diff --git a/neural-pde-surrogates-for-3d-electrostatics/NeuralPDEInferenceApp.mlapp b/neural-pde-surrogates-for-3d-electrostatics/NeuralPDEInferenceApp.mlapp new file mode 100644 index 0000000..f86fa74 Binary files /dev/null and b/neural-pde-surrogates-for-3d-electrostatics/NeuralPDEInferenceApp.mlapp differ diff --git a/neural-pde-surrogates-for-3d-electrostatics/README.md b/neural-pde-surrogates-for-3d-electrostatics/README.md new file mode 100644 index 0000000..1f8779b --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/README.md @@ -0,0 +1,74 @@ +# Rapid Design Exploration with Neural PDE Surrogates + +This demo shows how neural partial differential equation (PDE) surrogates can accelerate design exploration for a 3D electrostatics problem. +The goal is to predict the electric potential field throughout a transformer bushing insulator, +a cylindrical component with fins that provides electrical insulation between a high-voltage conductor and a grounded enclosure, +for varying bushing geometries. Two deep learning architectures are trained on finite element analysis (FEA) simulations to learn this mapping directly from geometry, enabling rapid evaluation of new designs without running a full simulation. + +## Architectures + +Two neural PDE surrogate architectures are demonstrated: + +- **Transolver** [\[1\]](README.md#references) — A transformer-based architecture for learning PDE solutions on unstructured meshes. Rather than applying attention across all mesh nodes, Transolver learns to group nodes into a small number of physics slices and applies attention across these slices, reducing the quadratic cost of global attention. Inputs are the 3D coordinates and material properties (relative permittivity) at each mesh node. + +- **MeshGraphNet** [\[2\]](README.md#references) — A graph neural network that represents the FEA mesh as a graph (nodes = mesh nodes, edges = element connectivity). Through message-passing layers, the network learns how information propagates through the geometry. Inputs are the 3D coordinates of the mesh nodes within the bushing insulator. + +Both architectures output the electric potential at every input node. + +## Training Data +The problem setup follows the documentation example [Electrostatic Analysis of Transformer Bushing Insulator](https://www.mathworks.com/help/pde/ug/electrostatic-analysis-of-transformer-bushing-insulator.html). +Boundary conditions are held fixed; only the transformer bushing insulator geometry varies between samples. + +The training dataset consists of 75 electrostatic simulations, each on a different procedurally generated transformer bushing geometry. For each simulation, we extract: +- FEA mesh nodes and element connectivity +- Material properties (relative permittivity at each node) +- Electric potential at each node +- Electric field at each node (for optional gradient regularization) + +## Results + +Below are example predictions from two trained models on two geometries. + +**Transolver** predictions: +![Transolver prediction on bushing geometry 39: side-by-side comparison of FEA solution, AI prediction with 2.8% relative L2 error, and relative error map](README_media/transolver_test_39.png) + +![Transolver prediction on bushing geometry 57: side-by-side comparison of FEA solution, AI prediction with 2.4% relative L2 error, and relative error map](README_media/transolver_test_57.png) + +**MeshGraphNet** predictions: +![MeshGraphNet prediction on bushing geometry 39: side-by-side comparison of FEA solution, AI prediction with 2.1% relative L2 error, and relative error map](README_media/meshgraphnet_test_39.png) + +![MeshGraphNet prediction on bushing geometry 57: side-by-side comparison of FEA solution, AI prediction with 2.2% relative L2 error, and relative error map](README_media/meshgraphnet_test_57.png) + + +# Getting Started + +**Setup:** +- Run [startup.m](startup.m) to set the MATLAB path. This also creates the `data/`, `STL/`, and `results/` subdirectories. + +**Data generation:** +- Run [generate\_data.m](generate_data.m) to generate the training dataset (~5 minutes with 8 parallel workers). This saves simulation results to `data/` and STL geometry files to `STL/`. + +**Training:** +- **Transolver:** Run [transolver\_script.m](transolver_script.m) (includes optional gradient regularization fine-tuning). +- **MeshGraphNet:** Run [meshgraphnet\_script.m](meshgraphnet_script.m). +- Set `doTrain = true` to train from scratch (GPU recommended). The trained model is saved to `results/`. +- Set `doTrain = false` and specify a filename in `loadFile` to load a pretrained model. +- Set `doFinetune = true` to finetune a trained model with gradient regularization. This adds a loss term that penalizes discrepancies between the predicted electric field $-\nabla V$ and the reference FEA electric field, where the spatial gradients are computed via automatic differentiation. This encourages spatially smooth predictions and improves derived electric field accuracy (Transolver only). + +**Inference app:** +- Type `NeuralPDEInferenceApp` in the Command Window. +- Click "Load AI Model" and select a model from the `results/` subdirectory. +- Click "Load Geometry" and choose an STL file from the `STL/` folder. +- Click "Predict!" to display the predicted electric potential. +- Toggle the switch to view electric field magnitude, which is numerically derived from the predicted potential. + +# Required Products +- MATLAB® (tested on R2026a) +- PDE Toolbox™ +- Deep Learning Toolbox™ +- Parallel Computing Toolbox™ +- Statistics and Machine Learning Toolbox™ + +# References +1. Wu, Haixu, Huakun Luo, Haowen Wang, Jianmin Wang, and Mingsheng Long. "Transolver: A Fast Transformer Solver for PDEs on General Geometries." arXiv, June 1, 2024. https://arxiv.org/abs/2402.02366. +2. Pfaff, Tobias, Meire Fortunato, Alvaro Sanchez-Gonzalez, and Peter W. Battaglia. "Learning Mesh-Based Simulation with Graph Networks." arXiv, June 18, 2021. https://arxiv.org/abs/2010.03409. diff --git a/neural-pde-surrogates-for-3d-electrostatics/README_media/meshgraphnet_test_39.png b/neural-pde-surrogates-for-3d-electrostatics/README_media/meshgraphnet_test_39.png new file mode 100644 index 0000000..721ba46 Binary files /dev/null and b/neural-pde-surrogates-for-3d-electrostatics/README_media/meshgraphnet_test_39.png differ diff --git a/neural-pde-surrogates-for-3d-electrostatics/README_media/meshgraphnet_test_57.png b/neural-pde-surrogates-for-3d-electrostatics/README_media/meshgraphnet_test_57.png new file mode 100644 index 0000000..3095a7d Binary files /dev/null and b/neural-pde-surrogates-for-3d-electrostatics/README_media/meshgraphnet_test_57.png differ diff --git a/neural-pde-surrogates-for-3d-electrostatics/README_media/transolver_test_39.png b/neural-pde-surrogates-for-3d-electrostatics/README_media/transolver_test_39.png new file mode 100644 index 0000000..26d5d3f Binary files /dev/null and b/neural-pde-surrogates-for-3d-electrostatics/README_media/transolver_test_39.png differ diff --git a/neural-pde-surrogates-for-3d-electrostatics/README_media/transolver_test_57.png b/neural-pde-surrogates-for-3d-electrostatics/README_media/transolver_test_57.png new file mode 100644 index 0000000..2da9161 Binary files /dev/null and b/neural-pde-surrogates-for-3d-electrostatics/README_media/transolver_test_57.png differ diff --git a/neural-pde-surrogates-for-3d-electrostatics/createBushing.m b/neural-pde-surrogates-for-3d-electrostatics/createBushing.m new file mode 100644 index 0000000..d1881ff --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/createBushing.m @@ -0,0 +1,209 @@ +function [gm, faceIDs] = createBushing(nFins, finRadiusBase, finRadiusTop, finWidth, tubeRadiusBase, tubeRadiusTop, boreRadius, totalLength) +%createBushing Create a parametric transformer bushing geometry. +% [gm, faceIDs] = createBushing(nFins, finRadiusBase, finRadiusTop, +% finWidth, tubeRadiusBase, tubeRadiusTop, boreRadius, totalLength) +% creates an axisymmetric transformer bushing with nFins cooling fin +% rings. The fins have a raised-cosine profile and are evenly spaced. +% Fin radii are linearly interpolated from finRadiusBase (first fin) to +% finRadiusTop (last fin). The tube tapers linearly from tubeRadiusBase +% at Z=0 to tubeRadiusTop at Z=totalLength. +% +% The geometry axis is along Z. The last (topmost) fin has a flat annular +% face at its peak for applying a boundary condition. +% +% Inputs: +% nFins - Number of cooling fin rings +% finRadiusBase - Outer radius of the first (bottom) fin +% finRadiusTop - Outer radius of the last (top) fin +% finWidth - Axial width of each fin (full width of cosine bump) +% tubeRadiusBase - Tube outer radius at Z=0 (bottom) +% tubeRadiusTop - Tube outer radius at Z=totalLength (top) +% boreRadius - Inner bore radius (central hole) +% totalLength - Total axial length of the bushing +% +% Outputs: +% gm - fegeometry object +% faceIDs - struct with fields: +% .bore - face ID for the inner bore surface +% .flatAnnular - face ID for the flat annular ring at +% the top of the last fin + + %% Compute fin positions and radii + % Asymmetric margins: larger at the bottom, small stub at the top. + % endStub is the distance from the last fin CENTER to the top of the + % bushing — this is the visible tube length above the split face. + endStub = 0.06 * totalLength; % short tube stub above the last fin peak + startMargin = 0.12 * totalLength; % longer tube section at the bottom + finPositions = linspace(startMargin + finWidth/2, ... + totalLength - endStub, nFins); + finRadii = linspace(finRadiusBase, finRadiusTop, nFins); + + % Clamp finWidth so adjacent fins don't overlap (leave a small tube gap) + if nFins > 1 + spacing = finPositions(2) - finPositions(1); + maxFinWidth = 0.9 * spacing; + if finWidth > maxFinWidth + finWidth = maxFinWidth; + end + end + + %% Helper: tube radius at any axial position (linear taper) + tubeRadiusAt = @(z) tubeRadiusBase + (tubeRadiusTop - tubeRadiusBase) * z / totalLength; + + %% Build the outer radius profile + nArc = 8; % points per fin bump (raised cosine) + + z_all = []; + r_all = []; + + % Tube section before first fin + z_start = linspace(0, finPositions(1) - finWidth/2, 3); + z_all = z_start; + r_all = tubeRadiusAt(z_start); + + for i = 1:nFins + hw = finWidth / 2; + center = finPositions(i); + + % Raised cosine bump for the fin, sitting on the local tube radius + zFin = linspace(center - hw, center + hw, nArc); + localTubeR = tubeRadiusAt(zFin); + bumpHeight = finRadii(i) - localTubeR; + rFin = localTubeR + 0.5 .* bumpHeight .* ... + (1 + cos(pi * (zFin - center) / hw)); + + z_all = [z_all, zFin]; %#ok + r_all = [r_all, rFin]; %#ok + + % Tube section after this fin + if i < nFins + zTube = linspace(center + hw + 0.002, ... + finPositions(i+1) - hw - 0.002, 3); + else + zTube = linspace(center + hw + 0.002, totalLength, 3); + end + z_all = [z_all, zTube]; %#ok + r_all = [r_all, tubeRadiusAt(zTube)]; %#ok + end + + % Remove duplicates and sort + [z_all, ia] = unique(z_all); + r_all = r_all(ia); + + %% Split at peak of last fin to create the flat annular face + lastFinCenter = finPositions(end); + [~, splitIdx] = min(abs(z_all - lastFinCenter)); + + %% Revolve lower body (all fins up to peak of last fin) + xv_low = [0, r_all(1:splitIdx), 0]; + yv_low = [z_all(1), z_all(1:splitIdx), z_all(splitIdx)]; + gm_lower = fegeometry(revolveProfile(xv_low, yv_low)); + + %% Revolve upper body (tube stub above last fin) + % Clamp radii to the local tube radius (remove any residual fin contribution) + localTubeAtSplit = tubeRadiusAt(z_all(splitIdx)); + r_upper_raw = r_all(splitIdx+1:end); + r_upper_clamped = min(r_upper_raw, tubeRadiusAt(z_all(splitIdx+1:end))); + r_upper = [localTubeAtSplit, r_upper_clamped]; + ax_upper = [z_all(splitIdx), z_all(splitIdx+1:end)]; + xv_up = [0, r_upper, 0]; + yv_up = [ax_upper(1), ax_upper, ax_upper(end)]; + gm_upper = fegeometry(revolveProfile(xv_up, yv_up)); + + %% Union the two halves and subtract the bore + gm = union(gm_lower, gm_upper); + bore = fegeometry(multicylinder(boreRadius, totalLength)); + gm = subtract(gm, bore); + + %% Find face IDs programmatically + z_split = z_all(splitIdx); + mid_r = (localTubeAtSplit + r_all(splitIdx)) / 2; + faceIDs.flatAnnular = nearestFace(gm, [mid_r, 0, z_split]); + faceIDs.bore = nearestFace(gm, [boreRadius, 0, totalLength/2]); + +end + + +function tri = revolveProfile(rProfile, zProfile, nAngles) +%revolveProfile Create a triangulated solid of revolution. +% tri = revolveProfile(rProfile, zProfile) revolves the closed 2D +% polygon defined by (rProfile, zProfile) around the Z axis to produce +% a closed triangulated surface mesh. rProfile contains radial +% coordinates (>= 0) and zProfile contains axial coordinates. +% Vertices with r ≈ 0 are treated as on-axis pole points. +% +% tri = revolveProfile(rProfile, zProfile, nAngles) specifies the +% number of angular divisions (default: 72, i.e. 5-degree steps). + + if nargin < 3 + nAngles = 72; + end + + theta = linspace(0, 2*pi, nAngles + 1); + theta(end) = []; % remove duplicate at 2*pi + + nPts = numel(rProfile); + onAxis = rProfile < 1e-10; + nOff = sum(~onAxis); + nOn = sum(onAxis); + + % Preallocate vertices + nVerts = nOff * nAngles + nOn; + verts = zeros(nVerts, 3); + vertMap = zeros(nPts, nAngles); + + k = 0; + for i = 1:nPts + if onAxis(i) + k = k + 1; + verts(k,:) = [0, 0, zProfile(i)]; + vertMap(i,:) = k; % all angles map to same pole vertex + else + for j = 1:nAngles + k = k + 1; + verts(k,:) = [rProfile(i)*cos(theta(j)), ... + rProfile(i)*sin(theta(j)), ... + zProfile(i)]; + vertMap(i,j) = k; + end + end + end + + % Preallocate faces (upper bound: 2 triangles per quad) + faces = zeros(2 * nPts * nAngles, 3); + f = 0; + + for i = 1:nPts + inext = mod(i, nPts) + 1; + for j = 1:nAngles + jnext = mod(j, nAngles) + 1; + + v1 = vertMap(i, j); + v2 = vertMap(i, jnext); + v3 = vertMap(inext, j); + v4 = vertMap(inext, jnext); + + if v1 == v2 && v3 == v4 + % Both on axis — degenerate, skip + continue; + elseif v1 == v2 + % Current point on axis — fan triangle + f = f + 1; + faces(f,:) = [v1, v4, v3]; + elseif v3 == v4 + % Next point on axis — fan triangle + f = f + 1; + faces(f,:) = [v1, v3, v2]; + else + % Quad — split into two triangles + f = f + 1; + faces(f,:) = [v1, v4, v3]; + f = f + 1; + faces(f,:) = [v1, v2, v4]; + end + end + end + + faces = faces(1:f,:); + tri = triangulation(faces, verts); +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/findProjectRoot.m b/neural-pde-surrogates-for-3d-electrostatics/findProjectRoot.m new file mode 100644 index 0000000..e5b0d4c --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/findProjectRoot.m @@ -0,0 +1,30 @@ +function rootDir = findProjectRoot(marker) +%findProjectRoot Locate the root directory of the project. +% rootDir = findProjectRoot(marker) searches upward from the current +% directory to find a directory containing the file or folder specified +% by marker (e.g., 'startup.m', '.git'). +% +% Inputs: +% marker - Name of a file or folder that identifies the project root +% +% Outputs: +% rootDir - Absolute path to the project root directory + +if nargin < 1 + marker = 'startup.m'; % Default marker file +end + +currentDir = pwd; +while true + if exist(fullfile(currentDir, marker), 'file') || ... + exist(fullfile(currentDir, marker), 'dir') + rootDir = currentDir; + return; + end + [parentDir, ~, ~] = fileparts(currentDir); + if strcmp(currentDir, parentDir) + error('Project root not found. Marker "%s" not found in any parent directory.', marker); + end + currentDir = parentDir; +end +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/generate_data.m b/neural-pde-surrogates-for-3d-electrostatics/generate_data.m new file mode 100644 index 0000000..9c80432 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/generate_data.m @@ -0,0 +1,186 @@ +%[text] # Generate Training Data (Geometry + 3D Simulation Dataset) +%[text] This script generates a paired dataset of **3D geometries** and **corresponding electrostatic simulation results** on those geometries for training and evaluating a deep learning surrogate model. +%[text] Each sample consists of: +%[text] - A parameterized transformer bushing insulator-like geometry (***Note***\*: the geometry parameters are only used to procedurally generate CAD geometry files and ensure reproducibility of the dataset. They are not provided as inputs to the surrogate model during training or inference. This mimics a common scenario where you may have access to the CAD geometry files, but no underlying parametric description of how those shapes were created.\*) +%[text] - A finite element solution of the electrostatic potential field and the derived electric field +%[text] - Per-node material properties +%[text] - Mesh information \ +%[text] The resulting dataset is intended for full-field surrogate modeling, where the model learns to predict spatially varying physical fields given geometry and material inputs. +%[text] The data generation takes several minutes to complete (~5 minutes using Parallel pool with 8 workers). +N = 75; % number of geometries to generate + +% Folders where we will store the generated data +projectRoot = findProjectRoot("startup.m"); +stldir = fullfile(projectRoot,"STL"); +datadir = fullfile(projectRoot,"data"); +%% +%[text] ## Generate CAD Data +%[text] This step creates a collection of parameterized axisymmetric 3D geometries representing variations of transformer bushing insulators. Each geometry consists of a central hollow tapered tube and a stack of annular fins, linearly increasing in radius from top to bottom. The geometry parameters are: +%[text] - number of fins +%[text] - fin radius at the base (smallest fin is at the bottom) +%[text] - fin radius at the top (largest fin is at the top) +%[text] - fin width +%[text] - tube radius at the base (narrower at the bottom) +%[text] - tube radius at the top (wider at the top) +%[text] - bore radius (fixed) +%[text] - total length (fixed) \ +%[text] The varying geometric parameters are sampled within prespecified bounds using Latin Hypercube Sampling (LHS) to efficiently cover the parameter space. +%[text] Define the parameter ranges. +% Varying parameters +nFinsRange = [6, 12]; % integer +finRadBaseRange = [0.06, 0.10]; % meters +finRadTopRange = [0.10, 0.16]; % meters +finWidthRange = [0.012, 0.05]; % meters +tubeRadBaseRange = [0.025, 0.045]; % meters (tube bottom) +tubeRadTopRange = [0.045, 0.065]; % meters (tube top) + +% Fixed parameters +boreRadius = 0.015; +totalLength = 0.6; +%[text] Generate the bushing parameters using (LHS). Uses [lhsdesign](https://www.mathworks.com/help/releases/R2026a/stats/lhsdesign.html?searchPort=58037) from Statistics and Machine Learning Toolbox™. +rng(42) % reproducibility +lhs = lhsdesign(N, 6); + +nFinsAll = round(nFinsRange(1) + lhs(:,1) * diff(nFinsRange)); +finRadBaseAll = finRadBaseRange(1) + lhs(:,2) * diff(finRadBaseRange); +finRadTopAll = finRadTopRange(1) + lhs(:,3) * diff(finRadTopRange); +finWidthAll = finWidthRange(1) + lhs(:,4) * diff(finWidthRange); +tubeRadBaseAll = tubeRadBaseRange(1) + lhs(:,5) * diff(tubeRadBaseRange); +tubeRadTopAll = tubeRadTopRange(1) + lhs(:,6) * diff(tubeRadTopRange); + +% Ensure finRadiusTop >= finRadiusBase for each sample +finRadTopAll = max(finRadTopAll, finRadBaseAll + 0.01); + +% Ensure tube top >= tube base +tubeRadTopAll = max(tubeRadTopAll, tubeRadBaseAll); +%% +%[text] Construct and export the geometries. Construct each geometry using primitive operations and Booleans. See `createBushing.m` for the geometry construction. +%[text] After the geometry is created, it is meshed and its surface triangulation is extracted. These triangulations are saved as STL files. +geometries = cell(N, 1); +faceIDsAll = cell(N, 1); +if isempty(gcp("nocreate")) + parpool; % use default cluster profile +end +parfor i = 1:N %[output:group:7298d8dd] %[output:4e79d9bc] + fprintf('Variant %d: nFins=%d, finRad=[%.3f,%.3f], width=%.3f, tubeRad=[%.3f,%.3f]\n', ... + i, nFinsAll(i), finRadBaseAll(i), finRadTopAll(i), finWidthAll(i), ... + tubeRadBaseAll(i), tubeRadTopAll(i)); + + [gm, fIDs] = createBushing(nFinsAll(i), finRadBaseAll(i), finRadTopAll(i), ... + finWidthAll(i), tubeRadBaseAll(i), tubeRadTopAll(i), ... + boreRadius, totalLength); + + geometries{i} = gm; + faceIDsAll{i} = fIDs; + + % Export to STL and save + filename = sprintf('%s/transformer_bushing_%03d.stl', stldir, i); + gm = generateMesh(gm,GeometricOrder="linear",Hmax=0.01); + + nodes = gm.Mesh.Nodes; + elements = gm.Mesh.Elements; + + [f, v] = freeBoundary(triangulation(elements', nodes')); + TR = triangulation(f, v); + stlwrite(TR,filename); +end %[output:group:7298d8dd] +%% +%[text] Save the parameters to a table. +params = table(nFinsAll, finRadBaseAll, finRadTopAll, finWidthAll, ... + tubeRadBaseAll, tubeRadTopAll, ... + repmat(boreRadius, N, 1), repmat(totalLength, N, 1), ... + VariableNames=["NFins", "FinRadiusBase", "FinRadiusTop", "FinWidth", ... + "TubeRadiusBase", "TubeRadiusTop", "BoreRadius", "TotalLength"]); +save("bushingParams.mat", "params") +fprintf('\nSaved %d parameter sets to bushingParams.mat\n', N); %[output:2096060e] +%% +%[text] Create one extra "inference-only" geometry with no simulation data, to demonstrate how the AI model can be used to make inference on new designs for which no high-fidelity simulation data exists. +rng(2); % for reproducibility +nFinsTest = randi(nFinsRange); +finRadBaseTest = (finRadBaseRange(2)-finRadBaseRange(1))*rand(1) + finRadBaseRange(1); +finRadTopTest = (finRadTopRange(2)-finRadTopRange(1))*rand(1) + finRadTopRange(1); +finWidthTest = (finWidthRange(2)-finWidthRange(1))*rand(1) + finWidthRange(1); +tubeRadBaseTest = (tubeRadBaseRange(2)-tubeRadBaseRange(1))*rand(1) + tubeRadBaseRange(1); +tubeRadTopTest = (tubeRadTopRange(2)-tubeRadTopRange(1))*rand(1) + tubeRadTopRange(1); + +gmTest = createBushing(nFinsTest, finRadBaseTest, finRadTopTest, ... + finWidthTest, tubeRadBaseTest, tubeRadTopTest, ... + boreRadius, totalLength); + +% Save parameters to a table +paramsTest = table(nFinsTest, finRadBaseTest, finRadTopTest, ... + finWidthTest, tubeRadBaseTest, tubeRadTopTest, ... + boreRadius, totalLength, ... + VariableNames=["NFins", "FinRadiusBase", "FinRadiusTop", "FinWidth", ... + "TubeRadiusBase", "TubeRadiusTop", "BoreRadius", "TotalLength"]); +save("testParams.mat", "paramsTest") +fprintf('\nSaved test parameter set to testParams.mat\n'); %[output:9651a736] + +% Export to STL and save +filename = sprintf('%s/TEST_bushing.stl', stldir); +gmTest = generateMesh(gmTest,GeometricOrder="linear",Hmax=0.01); + +nodes = gmTest.Mesh.Nodes; +elements = gmTest.Mesh.Elements; + +[f, v] = freeBoundary(triangulation(elements', nodes')); +TR = triangulation(f, v); +stlwrite(TR,filename); +%% +%[text] ## Generate CAE data (electrostatic simulation) +%[text] For each geometry, a 3D electrostatic finite element simulation is performed with PDE Toolbox. The experimental setup follows the documentation example [Electrostatic Analysis of Transformer Bushing Insulator](https://www.mathworks.com/help/pde/ug/electrostatic-analysis-of-transformer-bushing-insulator.html). +%[text] In the simulation setup, we embed the insulator geometry in a surrounding air volume. The materials used are: +%[text] - Air (relative permittivity = 1) +%[text] - Insulator (relative permittivity = 5) \ +%[text] The boundary conditions used are: +%[text] - Fixed high voltage applied to the inner bore surface, +%[text] - Ground applied to the flat annular ring. \ +%[text] To automate application of the boundary conditions, we identify the faces using [`nearestFace`](https://www.mathworks.com/help/pde/ug/discretegeometry.nearestface.html). For the full electrostatic simulation, see `solveBushingElectrostatic.m`. +%[text] For each geometry, the following simulation data is saved to a `.mat` file: +%[text] - Nodal coordinates (3xN array) +%[text] - Per-node relative permittivity values (1xN array) +%[text] - Per-node electric potential values (1xN array) +%[text] - Per-node electric field component values, derived from the FEA solution (3xN array) +%[text] - FEA mesh ([`FEMesh`](https://www.mathworks.com/help/pde/ug/pde.femesh.html) object) \ +%[text] This data is designed to support the creation of Transolver and MeshGraphNet models. +parfor i=1:N + gmBushing = fegeometry(sprintf('%s/transformer_bushing_%03d.stl', stldir, i)); + p = params(i,:); + [R,model] = solveBushingElectrostatic(p.NFins, p.FinRadiusBase, p.FinRadiusTop, ... + p.FinWidth, p.TubeRadiusBase, p.TubeRadiusTop, p.BoreRadius, p.TotalLength, ... + BushingGeometry=gmBushing); + % Save results in temporary struct + Di = struct(); + % Get the mesh nodes + Di.Coord = R.Mesh.Nodes; % 3 x numNodes array + % Get the material properties vector + epsilon = zeros(size(R.Mesh.Nodes,2),1); + idxAir = findNodes(R.Mesh,"Region",Cell=1); + epsilon(idxAir) = model.MaterialProperties(1).RelativePermittivity; + idxBushing = findNodes(R.Mesh,"Region",Cell=2); + epsilon(idxBushing) = model.MaterialProperties(2).RelativePermittivity; + Di.Epsilon = epsilon; + % Get the targets + Di.Potential = R.ElectricPotential; % numNodes x 1 + Di.ElectricField = [R.ElectricField.Ex'; R.ElectricField.Ey'; R.ElectricField.Ez']; % 3 x numNodes + % Get the mesh data + Di.Mesh = R.Mesh; + % Save the results to a .mat file for further analysis + Ssave = struct("D",Di); + save(sprintf('%s/EM_results_%03d.mat',datadir,i), "-fromstruct", Ssave); +end + +%[appendix]{"version":"1.0"} +%--- +%[metadata:view] +% data: {"layout":"onright"} +%--- +%[output:4e79d9bc] +% data: {"dataType":"text","outputData":{"text":"Variant 4: nFins=10, finRad=[0.078,0.143], width=0.024, tubeRad=[0.032,0.049]\nVariant 3: nFins=7, finRad=[0.063,0.152], width=0.025, tubeRad=[0.025,0.064]\nVariant 8: nFins=8, finRad=[0.066,0.123], width=0.015, tubeRad=[0.034,0.057]\nVariant 1: nFins=7, finRad=[0.071,0.147], width=0.029, tubeRad=[0.040,0.049]\nVariant 2: nFins=6, finRad=[0.081,0.129], width=0.016, tubeRad=[0.026,0.051]\nVariant 5: nFins=9, finRad=[0.082,0.160], width=0.020, tubeRad=[0.036,0.055]\nVariant 6: nFins=11, finRad=[0.074,0.120], width=0.038, tubeRad=[0.040,0.057]\nVariant 7: nFins=11, finRad=[0.089,0.140], width=0.026, tubeRad=[0.039,0.059]\nVariant 12: nFins=8, finRad=[0.084,0.156], width=0.044, tubeRad=[0.034,0.048]\nVariant 40: nFins=10, finRad=[0.087,0.114], width=0.019, tubeRad=[0.029,0.054]\nVariant 16: nFins=9, finRad=[0.090,0.138], width=0.042, tubeRad=[0.030,0.054]\nVariant 24: nFins=11, finRad=[0.098,0.108], width=0.044, tubeRad=[0.044,0.057]\nVariant 28: nFins=9, finRad=[0.094,0.137], width=0.036, tubeRad=[0.031,0.047]\nVariant 20: nFins=7, finRad=[0.075,0.117], width=0.021, tubeRad=[0.030,0.061]\nVariant 32: nFins=11, finRad=[0.085,0.103], width=0.019, tubeRad=[0.030,0.060]\nVariant 36: nFins=8, finRad=[0.068,0.127], width=0.042, tubeRad=[0.032,0.048]\nVariant 11: nFins=10, finRad=[0.078,0.146], width=0.015, tubeRad=[0.033,0.055]\nVariant 19: nFins=11, finRad=[0.081,0.121], width=0.029, tubeRad=[0.038,0.059]\nVariant 15: nFins=9, finRad=[0.091,0.125], width=0.030, tubeRad=[0.039,0.047]\nVariant 39: nFins=7, finRad=[0.062,0.158], width=0.038, tubeRad=[0.044,0.062]\nVariant 27: nFins=8, finRad=[0.079,0.121], width=0.043, tubeRad=[0.033,0.064]\nVariant 35: nFins=7, finRad=[0.077,0.153], width=0.013, tubeRad=[0.038,0.062]\nVariant 31: nFins=7, finRad=[0.084,0.142], width=0.039, tubeRad=[0.043,0.063]\nVariant 23: nFins=11, finRad=[0.066,0.101], width=0.034, tubeRad=[0.034,0.048]\nVariant 10: nFins=8, finRad=[0.088,0.134], width=0.035, tubeRad=[0.045,0.052]\nVariant 38: nFins=12, finRad=[0.098,0.108], width=0.022, tubeRad=[0.035,0.056]\nVariant 14: nFins=9, finRad=[0.070,0.125], width=0.025, tubeRad=[0.041,0.046]\nVariant 18: nFins=12, finRad=[0.071,0.104], width=0.034, tubeRad=[0.038,0.057]\nVariant 34: nFins=9, finRad=[0.071,0.149], width=0.046, tubeRad=[0.026,0.053]\nVariant 26: nFins=6, finRad=[0.068,0.126], width=0.037, tubeRad=[0.038,0.046]\nVariant 30: nFins=8, finRad=[0.091,0.154], width=0.031, tubeRad=[0.033,0.059]\nVariant 22: nFins=9, finRad=[0.096,0.114], width=0.041, tubeRad=[0.028,0.050]\nVariant 9: nFins=8, finRad=[0.064,0.128], width=0.019, tubeRad=[0.031,0.045]\nVariant 13: nFins=10, finRad=[0.076,0.140], width=0.026, tubeRad=[0.028,0.053]\nVariant 25: nFins=9, finRad=[0.076,0.157], width=0.044, tubeRad=[0.033,0.059]\nVariant 37: nFins=6, finRad=[0.100,0.133], width=0.017, tubeRad=[0.043,0.063]\nVariant 29: nFins=10, finRad=[0.088,0.110], width=0.035, tubeRad=[0.045,0.056]\nVariant 17: nFins=8, finRad=[0.086,0.115], width=0.024, tubeRad=[0.027,0.049]\nVariant 33: nFins=12, finRad=[0.093,0.123], width=0.047, tubeRad=[0.029,0.062]\nVariant 21: nFins=11, finRad=[0.065,0.118], width=0.046, tubeRad=[0.044,0.050]\nVariant 42: nFins=10, finRad=[0.067,0.135], width=0.042, tubeRad=[0.041,0.055]\nVariant 44: nFins=11, finRad=[0.095,0.129], width=0.012, tubeRad=[0.042,0.048]\nVariant 46: nFins=6, finRad=[0.086,0.108], width=0.031, tubeRad=[0.038,0.052]\nVariant 50: nFins=10, finRad=[0.098,0.119], width=0.033, tubeRad=[0.037,0.064]\nVariant 52: nFins=11, finRad=[0.093,0.156], width=0.040, tubeRad=[0.036,0.050]\nVariant 54: nFins=12, finRad=[0.073,0.150], width=0.040, tubeRad=[0.028,0.058]\nVariant 48: nFins=12, finRad=[0.096,0.109], width=0.032, tubeRad=[0.026,0.063]\nVariant 56: nFins=9, finRad=[0.062,0.108], width=0.048, tubeRad=[0.034,0.056]\nVariant 41: nFins=8, finRad=[0.097,0.152], width=0.041, tubeRad=[0.043,0.058]\nVariant 45: nFins=7, finRad=[0.061,0.113], width=0.021, tubeRad=[0.040,0.047]\nVariant 43: nFins=10, finRad=[0.080,0.132], width=0.027, tubeRad=[0.041,0.061]\nVariant 49: nFins=11, finRad=[0.099,0.142], width=0.020, tubeRad=[0.043,0.051]\nVariant 51: nFins=10, finRad=[0.077,0.102], width=0.027, tubeRad=[0.042,0.052]\nVariant 53: nFins=7, finRad=[0.092,0.149], width=0.017, tubeRad=[0.028,0.061]\nVariant 47: nFins=7, finRad=[0.072,0.111], width=0.050, tubeRad=[0.028,0.049]\nVariant 55: nFins=9, finRad=[0.092,0.159], width=0.028, tubeRad=[0.039,0.055]\nVariant 59: nFins=7, finRad=[0.066,0.137], width=0.047, tubeRad=[0.027,0.065]\nVariant 57: nFins=9, finRad=[0.082,0.122], width=0.023, tubeRad=[0.029,0.060]\nVariant 58: nFins=6, finRad=[0.069,0.139], width=0.032, tubeRad=[0.036,0.051]\nVariant 60: nFins=11, finRad=[0.065,0.106], width=0.016, tubeRad=[0.037,0.064]\nVariant 62: nFins=10, finRad=[0.075,0.113], width=0.036, tubeRad=[0.042,0.050]\nVariant 63: nFins=7, finRad=[0.087,0.111], width=0.014, tubeRad=[0.031,0.058]\nVariant 61: nFins=7, finRad=[0.086,0.145], width=0.048, tubeRad=[0.030,0.058]\nVariant 64: nFins=9, finRad=[0.064,0.130], width=0.022, tubeRad=[0.035,0.047]\nVariant 65: nFins=8, finRad=[0.073,0.131], width=0.036, tubeRad=[0.025,0.045]\nVariant 66: nFins=8, finRad=[0.079,0.131], width=0.014, tubeRad=[0.027,0.052]\nVariant 67: nFins=11, finRad=[0.090,0.103], width=0.029, tubeRad=[0.032,0.054]\nVariant 71: nFins=10, finRad=[0.083,0.150], width=0.045, tubeRad=[0.040,0.054]\nVariant 68: nFins=8, finRad=[0.073,0.148], width=0.023, tubeRad=[0.035,0.062]\nVariant 69: nFins=10, finRad=[0.083,0.116], width=0.031, tubeRad=[0.037,0.060]\nVariant 70: nFins=9, finRad=[0.095,0.105], width=0.013, tubeRad=[0.031,0.061]\nVariant 72: nFins=12, finRad=[0.069,0.155], width=0.049, tubeRad=[0.043,0.053]\nVariant 73: nFins=11, finRad=[0.061,0.145], width=0.049, tubeRad=[0.041,0.054]\nVariant 74: nFins=7, finRad=[0.094,0.143], width=0.038, tubeRad=[0.027,0.065]\nVariant 75: nFins=6, finRad=[0.060,0.136], width=0.018, tubeRad=[0.035,0.046]\n","truncated":false}} +%--- +%[output:2096060e] +% data: {"dataType":"text","outputData":{"text":"\nSaved 75 parameter sets to bushingParams.mat\n","truncated":false}} +%--- +%[output:9651a736] +% data: {"dataType":"text","outputData":{"text":"\nSaved test parameter set to testParams.mat\n","truncated":false}} +%--- diff --git a/neural-pde-surrogates-for-3d-electrostatics/meshgraphnet_script.m b/neural-pde-surrogates-for-3d-electrostatics/meshgraphnet_script.m new file mode 100644 index 0000000..1ea601a --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/meshgraphnet_script.m @@ -0,0 +1,2235 @@ +%[text] # Rapid Design Exploration for a 3D Electrostatics Problem Using MeshGraphNet +%[text] Accurately solving three-dimensional electrostatic problems is central to the design and analysis of many electrical and electromechanical systems, including transformer bushings, high-voltage insulation, and power electronics. While finite element analysis (FEA) provides high-fidelity solutions, repeatedly solving these problems across varying geometries can be computationally expensive, limiting its practicality for rapid design exploration, optimization, and real-time workflows. +%[text] This demo demonstrates how deep learning can be used to enable fast, full-field solutions of 3D electrostatic partial differential equations (PDEs). Using simulation data generated with MATLAB® PDE Toolbox™, we train a neural network to predict the electric potential field directly from mesh geometry. The geometries vary across samples and are discretized with different node counts, making the problem particularly challenging for conventional neural network architectures that assume fixed grids or inputs. +%[text] To address this challenge, we use **MeshGraphNet** \[[1](internal:M_5694)\], a graph neural network architecture designed for learning PDE solutions on unstructured meshes. MeshGraphNet represents the finite element mesh as a graph, where each mesh node becomes a graph node and edges are defined by the mesh connectivity. Each node is represented by its 3D coordinates. Through a sequence of message-passing layers, the network learns how information propagates locally through the geometry, enabling it to approximate the solution of the underlying PDE. The network outputs the electric potential at every node. Unlike Transolver, which models the full simulation domain (bushing and surrounding air) and uses material properties as additional input features, this MeshGraphNet example operates only on the bushing insulator mesh. +%[text] In this demo, we apply MeshGraphNet to a 3D electrostatics problem involving a transformer bushing insulator embedded in air. The trained model is evaluated on unseen geometries, and its predictions are compared against FEA solutions. +%[text] ![](text:image:372c) +%[text:tableOfContents]{"heading":"Table of Contents"} +%[text] +%% +%[text] %[text:anchor:TMP_4ac9] **Experiment Settings** +%[text] The function [`canUseGPU`](https://www.mathworks.com/help/releases/R2026a/matlab/ref/canusegpu.html) is used here to detect and enable GPU acceleration for training and inference. Set `doTrain` to `true` to train a new model; otherwise, specify the name of a pretrained model to load. +useGpu = canUseGPU; % use GPU if one is available +doTrain = true; % set to 1 if you would like to train the model from scratch, or 0 if you want to load in a pre-trained network %[control:checkbox:9e9f]{"position":[11,15]} +if doTrain + ts = string(datetime("now",Format="uuuuMMdd'T'HHmmss")); % specify timestamp to uniquely name results file +else + loadFile = "MeshGraphNet_XXXXXXXXXXXXXXX"; % name of results file (if loading a pretrained model) +end +projectRoot = findProjectRoot("startup.m"); +outdir = fullfile(projectRoot,"results"); +%% +%[text] ## Load Data +%[text] Load simulation data generated using MATLAB PDE Toolbox. The simulation data was generated by varying a parameterized geometry inspired by the one in the documentation example [Electrostatic Analysis of Transformer Bushing Insulator](https://www.mathworks.com/help/pde/ug/electrostatic-analysis-of-transformer-bushing-insulator.html). Data files are expected to be located in the `data/` subdirectory within the parent directory of this script. +%[text] Each file contains a struct with FEA data for that particular observation. The struct includes the following fields: +%[text] - `Coord`: 3 x $N$ array of node coordinates in the FEA mesh. Used as model input. +%[text] - `Potential`: $N$ x 1 array of electric potential values at each node. Used as training targets. +%[text] - `Mesh`: [FEMesh](https://www.mathworks.com/help/pde/ug/pde.femesh.html) object representing the FEA mesh. Used for graph construction and visualization. \ +%[text] Note that $N$(number of nodes) varies between samples, as each simulation uses a different geometry and therefore mesh. +if doTrain + datadir = fullfile("data"); + data = io.loadAllData(datadir); +else + % Skip data loading when loading a pretrained model +end +%% +%[text] ## Preprocess Data +%[text] Preprocess the finite element data into the graph-based representation required by MeshGraphNet. +%[text] For each training sample, construct: +%[text] - Node features (`Xs`): per-node input features, based on mesh node coordinates within the transformer bushing insulator +%[text] - Edge features (`Xes`): per-edge features derived from mesh connectivity +%[text] - Node targets (`Ys`): ground-truth field values (e.g. electric potential) at each mesh node +%[text] - Adjacency matrices (`As`): node-to-node connectivity defining the graph +%[text] - Node-edge adjacency matrices (`Bs`): mappings between nodes and edges \ +%[text] This representation allows MGN to operate directly on the finite element mesh and learn local physical interactions through message passing. +if doTrain + % Number of training samples + numObs = numel(data); + % Preallocate cell arrays for graph data + Xs = cell(numObs,1); + Xes = cell(numObs,1); + Ys = cell(numObs,1); + As = cell(numObs,1); + Bs = cell(numObs,1); + + % Iterate over samples and construct graph representations + for i = 1:numObs + bushing = findNodes(data(i).Mesh,"region",Cell=2); + [Xs{i},Xes{i},Ys{i},As{i},Bs{i}] = meshgraphnet.preprocessData(data(i),bushing); + end + clear data % free memory once raw data is no longer needed +%[text] Split the dataset into 80% training and 20% validation subsets. + [idxTrain,idxVal] = utils.trainTestSplit(numObs,0.8); + cfg.IdxVal = idxVal; % Save validation indices for later evaluation + XTrain = Xs(idxTrain); + XVal = Xs(idxVal); + XeTrain = Xes(idxTrain); + XeVal = Xes(idxVal); + YTrain = Ys(idxTrain); + YVal = Ys(idxVal); + ATrain = As(idxTrain); + AVal = As(idxVal); + BTrain = Bs(idxTrain); + BVal = Bs(idxVal); +%[text] Normalize Node Features, Edge Features, and Targets. Normalize inputs and targets using statistics computed from the training data only. Different normalization ranges are used for inputs and outputs: Node features (`X`) and edge features (`Xe`) are scaled to \[-1, 1\], while target values are scaled to \[0,1\]. Scaling is applied feature-wise using linear min–max normalization. + [inputOffset,inputScale] = utils.computeScaleStats(XTrain,'[-1,1]'); + [edgeOffset,edgeScale] = utils.computeScaleStats(XeTrain,'[-1,1]'); + [outputOffset,outputScale] = utils.computeScaleStats(YTrain,'[0,1]'); + normalizedXTrain = utils.applyScale(XTrain,inputOffset,inputScale); + normalizedXVal = utils.applyScale(XVal,inputOffset,inputScale); + normalizedXeTrain = utils.applyScale(XeTrain,edgeOffset,edgeScale); + normalizedXeVal = utils.applyScale(XeVal,edgeOffset,edgeScale); + normalizedYTrain = utils.applyScale(YTrain,outputOffset,outputScale); + normalizedYVal = utils.applyScale(YVal,outputOffset,outputScale); +%[text] Save the normalization parameters (`inputOffset`, `inputScale`, `edgeOffset`, `edgeScale`, `outputOffset`, `outputScale`) for use during inference. + cfg.InputOffset = inputOffset; cfg.InputScale = inputScale; + cfg.EdgeOffset = edgeOffset; cfg.EdgeScale = edgeScale; + cfg.OutputOffset = outputOffset; cfg.OutputScale = outputScale; +else + % Skip preprocessing when loading a pretrained model +end +%% +%[text] ## Define Network Architecture +%[text] Define the MeshGraphNet architecture and initialize the learnable parameters. +%[text] MGN is a graph neural network designed to learn solutions on unstructured meshes by performing message-passing over the mesh connectivity. +%[text] Conceptually, the architecture foolowns an Encoder $\\to$ Message-Passing Processors $\\to$ Decoder pipeline. The encoder maps raw node and edge features into a latent feature space. The processor blocks iteratively update node/edge representations using graph connectivity (adjacency / node-edge adjacency matrices). The decoder maps the final latent node features to the target field values (one scalar per node in this example). +%[text] The key hyperparameters of MGN are: +%[text] - `HiddenSize`: latent feature dimension, +%[text] - `NumLayersEncoder`: depth of the encoder MLP(s) +%[text] - `NumProcessors`: number of message-passing blocks +%[text] - `NumLayersProcessor`: depth of processor MLP(s) +%[text] - `NumLayersDecoder`: depth of the decoder MLP(s) +%[text] - `MaxFreq`: (optional) frequency/embedding parameter used in this implementation \ +if doTrain + % Demonstration call of model + cfg.ModelName = "MeshGraphNet"; + cfg.NodeInputSize = size(normalizedXTrain{1},1); + cfg.EdgeInputSize = size(normalizedXeTrain{1},1); + cfg.OutputSize = 1; + cfg.HiddenSize = 128; + cfg.NumLayersEncoder = 2; + cfg.NumProcessors = 3; + cfg.NumLayersProcessor = 3; + cfg.NumLayersDecoder = 3; + cfg.MaxFreq = 2; + cfg.Notes = "Predict 3D electric potential from electrostatic FEA data. Geometry varies across samples; other simulation settings (e.g. boundary conditions) are held fixed."; + % Initialize learnable parameters + p = meshgraphnet.initialize(cfg); + % Demonstration model forward pass + % z = meshgraphnet.meshgraphnet(p,dlarray(normalizedXTrain{1}),dlarray(normalizedXeTrain{1}),BTrain{1},cfg); +else + % Skip network construction when loading pretrained model +end +%% +%[text] ## Define the Model Loss Function +%[text] Define the loss function used to train the MeshGraphNet model. +%[text] This function computes a mean squared error (MSE) loss between the predicted and true field values at each mesh node. The loss is averaged over all nodes in the graph. +%[text] During training, the loss is differentiated with respect to the model parameters using automatic differentiation. +function [loss,grad] = modelLoss(p,x,xe,B,y,cfg) + % A mean squared error loss. + z = meshgraphnet.meshgraphnet(p,x,xe,B,cfg); + loss = mean((z - y).^2,"all"); + grad = dlgradient(loss,p); +end +%% +%[text] ## Train the Network +%[text] Train the MeshGraphNet model using a custom training loop using the ADAM optimizer. Training is performed on either a GPU or CPU, depending on availability, and the model parameters that achieve the best validation performance are retained. +%[text] MeshGraphNet training differs slightly from standard deep learning workflows because each training sample corresponds to a **graph with a variable number of nodes and edges**. As a result, we construct custom minibatches and explicitly manage the forward pass, loss computation, and parameter updates. +if doTrain %[output:group:8481ed1c] +%[text] Before training, intitialize the model parameters and configure optimization and bookkeeping settings. + % Initialize model parameters + if useGpu + gpuDevice(1).reset(); + p = meshgraphnet.initialize(cfg); + p = dlupdate(@gpuArray,p); + end + % Training hyperparameters + epochs = 300; + mbSize = 2; + lr = 1e-3; + % ADAM optimizer state + avgG = []; + avgSqG = []; + % Training set size + nobsTrain = numel(normalizedXTrain); + mbPerEpoch = ceil(nobsTrain/mbSize); + totalIters = epochs*mbPerEpoch; + iter = 0; + % Capture loss function used + myFun = @modelLoss; cfg.LossName = "MSE"; + % Preallocate arrays to track training and validation loss + trainIter = zeros(totalIters,1); + trainEpoch = zeros(totalIters,1); + trainLR = zeros(totalIters,1); + trainLoss = zeros(totalIters,1); + valIter = zeros(epochs,1); + valLoss = zeros(epochs,1); +%[text] Create a single batched representation of the validation set to efficiently evaluate validation loss at the end of each epoch. + [Xvb,Xevb,Bvb,Yvb] = meshgraphnet.minibatch(normalizedXVal,... + normalizedXeVal,BVal,normalizedYVal,useGpu); +%[text] Track the best-performing model parameters based on validation loss. + bestValLoss = inf; + pBest = p; + % Training progress monitor + monitor = trainingProgressMonitor( ... %[output:8efe7c47] + Metrics=["TrainingLoss","ValidationLoss"], ... %[output:8efe7c47] + Info=["Epoch","Solver","LearnRateSchedule","LearnRate", ... %[output:8efe7c47] + "ValidationFrequency","ValidationPatience", ... %[output:8efe7c47] + "ObjectiveMetric","OutputNetwork","HardwareResource"], ... %[output:8efe7c47] + XLabel="Iteration"); %[output:8efe7c47] + groupSubPlot(monitor, "Loss", ["TrainingLoss","ValidationLoss"]); + yscale(monitor, "Loss", "log"); + if useGpu, hw = "GPU"; else, hw = "CPU"; end + updateInfo(monitor, Solver="Adam", LearnRateSchedule="Cosine", ... %[output:8efe7c47] + ValidationFrequency=string(mbPerEpoch) + " iterations", ... %[output:8efe7c47] + ValidationPatience="Inf", ObjectiveMetric="mse", ... %[output:8efe7c47] + OutputNetwork="best-validation-loss", HardwareResource=hw); %[output:8efe7c47] + % Record initial validation loss before training begins + valPred0 = meshgraphnet.meshgraphnet(p,Xvb,Xevb,Bvb,cfg); + recordMetrics(monitor, 0, ValidationLoss=double(extractdata(gather(mean((valPred0-Yvb).^2,"all"))))); %[output:8efe7c47] +%[text] Train the network. The network is trained over multiple epochs. In each epoch, the training data is shuffled and processed in minibatches. For each minibatch, the loss and gradients are computed, and the model parameters are updated using the ADAM optimizer. + for epoch = 1:epochs + % Shuffle training samples + idx = randperm(nobsTrain); + for i = 1:mbPerEpoch + iter = iter+1; + % Minibatch indices + mbidx = (i-1)*(mbSize) + (1:mbSize); + mbidx(mbidx>nobsTrain) = []; + % Apply random augmentation + augX = normalizedXTrain(mbidx); + augXe = normalizedXeTrain(mbidx); + for s = 1:numel(augX) + % Random rotation around Z-axis (valid for axisymmetric geometry) + theta = 360*rand(); + ct = cosd(theta); st = sind(theta); + R = [ct -st 0; st ct 0; 0 0 1]; + augX{s} = R * augX{s}; + augXe{s}(1:3,:) = R * augXe{s}(1:3,:); % rotate displacement vectors; length (row 4) unchanged + % Random x-reflection (valid for axisymmetric geometry) + if rand() > 0.5 + augX{s}(1,:) = -augX{s}(1,:); + augXe{s}(1,:) = -augXe{s}(1,:); + end + end + % Contruct minibatch + [Xb,Xeb,Bb,Yb] = meshgraphnet.minibatch(augX,... + augXe, ... + BTrain(mbidx),... + normalizedYTrain(mbidx),... + useGpu); + % Evaluate loss and gradients + [loss,grad] = dlfeval(myFun,p,Xb,Xeb,Bb,Yb,cfg); + % Cosine learning rate schedule + currentLR = lr * 0.5 * (1 + cos(pi * iter / totalIters)); + % Update parameters using ADAM + [p,avgG,avgSqG] = adamupdate(p,grad,avgG,avgSqG,iter,currentLR); + % Log training history + trainIter(iter) = iter; + trainEpoch(iter) = epoch; + trainLR(iter) = currentLR; + trainLoss(iter) = gather(extractdata(loss)); + recordMetrics(monitor, iter, TrainingLoss=trainLoss(iter)); %[output:8efe7c47] + monitor.Progress = 100 * iter / totalIters; + % Console report + fprintf("Iter: %d, Epoch: %d, Loss: %.4f\n",... %[output:2388b472] %[output:9e95aef6] %[output:453605ac] %[output:187ab2a1] %[output:29eb44cd] %[output:9665f29d] %[output:2dd92824] %[output:9e79b3a9] %[output:145f82f0] %[output:5161a939] %[output:5fcb025f] %[output:5efc4a41] %[output:2e1e5c6b] %[output:20cf326e] %[output:0c0cfeda] %[output:36e06638] %[output:1b282164] %[output:3599dc82] %[output:8b7126e7] %[output:30e10416] %[output:519e33c4] %[output:46b75fc8] %[output:12e0435c] %[output:48d67fce] %[output:2ad740aa] %[output:48e6ee01] %[output:6e129df8] %[output:94c9a19d] %[output:8ef8e50f] %[output:6da734f3] %[output:6a80ef77] %[output:56e69f64] %[output:00d506a3] %[output:154ca774] %[output:7567b303] %[output:905b4938] %[output:60bb2365] %[output:2a067f77] %[output:192ec571] %[output:1f9c02ba] %[output:3470d1dc] %[output:880f85d0] %[output:76a6da47] %[output:48076b9d] %[output:9208fbb0] %[output:2c554e52] %[output:548afa29] %[output:2945cbb2] %[output:694de540] %[output:124473a9] %[output:55ee496b] %[output:1586ad98] %[output:7ccbbb25] %[output:802a9cda] %[output:2e9f6233] %[output:683d307a] %[output:32e82a2d] %[output:62df5769] %[output:3db1fd45] %[output:74c7b215] %[output:654071b7] %[output:1a0f64d4] %[output:69c511c7] %[output:80c50658] %[output:521ce776] %[output:6abfb90a] %[output:49258648] %[output:467e4aca] %[output:6e1a86eb] %[output:5eea11aa] %[output:747bc9b1] %[output:0a0fe026] %[output:3dff3184] %[output:3b6a102e] %[output:90476dad] %[output:4c128056] %[output:4d184a5c] %[output:189227c0] %[output:39ba7841] %[output:7c592cf0] %[output:8c86dca4] %[output:41b2baf4] %[output:84fd4b08] %[output:2ce57674] %[output:9803f580] %[output:4b4afd09] %[output:4b3de16c] %[output:7d213b50] %[output:00d641d0] %[output:6ce5ffd8] %[output:9ee1f7e4] %[output:56361963] %[output:35f19933] %[output:7f01428e] %[output:0fa9183e] %[output:0bf73890] %[output:58d76c03] %[output:63fe1cbf] %[output:8bae86ce] %[output:25455714] %[output:51438166] %[output:09c8c0a0] %[output:76141baf] %[output:4ecabd20] %[output:37ba81ed] %[output:296220cb] %[output:85737333] %[output:88775070] %[output:5dea9ab5] %[output:5c7b048f] %[output:7053153a] %[output:084c6bae] %[output:6b80aade] %[output:920fbb93] %[output:30756905] %[output:5f1a920e] %[output:9b5cb80a] %[output:09bba6c0] %[output:9284807a] %[output:17999174] %[output:1e371406] %[output:5444af52] %[output:4e07268e] %[output:3e8b13ca] %[output:605ff620] %[output:31369086] %[output:572767b5] %[output:971f3ed3] %[output:81e55fff] %[output:95a00284] %[output:51e9f898] %[output:448c9078] %[output:8e5fea78] %[output:7307a7df] %[output:366b2194] %[output:9cb564cd] %[output:9da4f824] %[output:02dcff7f] %[output:11080f9f] %[output:6087ca7c] %[output:9171ca81] %[output:6eda7cb5] %[output:58f8fccf] %[output:9a066e9f] %[output:15d4e80f] %[output:392b24cd] %[output:99a0520f] %[output:72bb2e0a] %[output:51372b91] %[output:0e0a21e9] %[output:1b40466f] %[output:9665fe6f] %[output:79f52345] %[output:4c6e0a04] %[output:4a771709] %[output:6f63bb9e] %[output:5beff449] %[output:7dbbd20d] %[output:2685e72a] %[output:707a11a3] %[output:8ec25ece] %[output:75ab77b0] %[output:8a991d2b] %[output:081f865d] %[output:75bfbbed] %[output:501736cc] %[output:699cbdea] %[output:3dd4bc93] %[output:6cde6acf] %[output:9a1bc941] %[output:3afe39b3] %[output:1684ec2d] %[output:5f1a718a] %[output:49b33c6e] %[output:8b6c8d64] %[output:4d6b2bc8] %[output:618c1793] %[output:3f4f78f2] %[output:8a6f0cf8] %[output:6f32ee91] %[output:38bab324] %[output:34c87860] %[output:9432b1fe] %[output:9823e54b] %[output:3ba13a05] %[output:314da832] %[output:88d51177] %[output:9c5ad102] %[output:97889ee2] %[output:11d9dde7] %[output:59da6fed] %[output:99376ae3] %[output:9b3f564f] %[output:5111034d] %[output:73677cba] %[output:84a1c31e] %[output:139d7b57] %[output:48d8a20e] %[output:782bb660] %[output:7ad79bda] %[output:07705166] %[output:51aacec6] %[output:4d0d3fa4] %[output:9069e957] %[output:8510bcb5] %[output:52fe6de5] %[output:7f5b687f] %[output:17689972] %[output:94236b53] %[output:71de17e9] %[output:7a589bc7] %[output:1eeb0a4c] %[output:509ab337] %[output:8e64767f] %[output:2c37d46c] %[output:4c553f76] %[output:4d80af44] %[output:53be6a8e] %[output:4e8cf01d] %[output:25607ab4] %[output:8aede320] %[output:77d2493d] %[output:390d1015] %[output:9c55235f] %[output:88f6cb63] %[output:24dce2b7] %[output:5a4a9c5f] %[output:886aea0b] %[output:4e5c6c95] %[output:2bc02e69] %[output:69baabec] %[output:0bfc7764] %[output:7e79f7b7] %[output:2b46a3f3] %[output:95f476b4] %[output:62248664] %[output:7665abf6] %[output:77fa26fc] %[output:1704c9b1] %[output:59928606] %[output:9554411f] %[output:64e41fcb] %[output:1668efdb] %[output:338f8083] %[output:8b4d0d9a] %[output:4b6cc66d] %[output:068739bf] %[output:7fd9b34c] %[output:05ca75b7] %[output:84846fe8] %[output:33b2fa90] %[output:4be59cf3] %[output:2ad7b632] %[output:7b93a34d] %[output:3ff98b25] %[output:720bbf08] %[output:4c94c0df] %[output:74c63ce7] %[output:0ea1556d] %[output:6e4ae90a] %[output:4ab8bed5] %[output:03ef4d44] %[output:31437bc9] %[output:10b1afd6] %[output:6b647426] %[output:18f55326] %[output:1c008433] %[output:831451bb] %[output:995c9a36] %[output:3b7e0b1c] %[output:44f95a0a] %[output:2ca643cb] %[output:302c0cd0] %[output:27c00cfa] %[output:4bbd852d] %[output:4aad7518] %[output:69f09dd7] %[output:2d31eba6] %[output:0be852e5] %[output:5b3381ba] %[output:7568c2ca] %[output:62de3957] %[output:86cfe797] %[output:18ee7b6e] %[output:5c3bffc5] %[output:339338b8] %[output:9f25a6ad] %[output:22d78126] %[output:07d6ecf4] %[output:55ec4c8a] %[output:2ba25360] %[output:02df5bbb] %[output:4e6fc9b0] %[output:0f215664] %[output:8628a909] %[output:10eb2dc6] %[output:4553bef2] %[output:1031648f] %[output:3b096af5] %[output:1451c82d] + i,epoch,extractdata(gather(loss))); %[output:2388b472] %[output:9e95aef6] %[output:453605ac] %[output:187ab2a1] %[output:29eb44cd] %[output:9665f29d] %[output:2dd92824] %[output:9e79b3a9] %[output:145f82f0] %[output:5161a939] %[output:5fcb025f] %[output:5efc4a41] %[output:2e1e5c6b] %[output:20cf326e] %[output:0c0cfeda] %[output:36e06638] %[output:1b282164] %[output:3599dc82] %[output:8b7126e7] %[output:30e10416] %[output:519e33c4] %[output:46b75fc8] %[output:12e0435c] %[output:48d67fce] %[output:2ad740aa] %[output:48e6ee01] %[output:6e129df8] %[output:94c9a19d] %[output:8ef8e50f] %[output:6da734f3] %[output:6a80ef77] %[output:56e69f64] %[output:00d506a3] %[output:154ca774] %[output:7567b303] %[output:905b4938] %[output:60bb2365] %[output:2a067f77] %[output:192ec571] %[output:1f9c02ba] %[output:3470d1dc] %[output:880f85d0] %[output:76a6da47] %[output:48076b9d] %[output:9208fbb0] %[output:2c554e52] %[output:548afa29] %[output:2945cbb2] %[output:694de540] %[output:124473a9] %[output:55ee496b] %[output:1586ad98] %[output:7ccbbb25] %[output:802a9cda] %[output:2e9f6233] %[output:683d307a] %[output:32e82a2d] %[output:62df5769] %[output:3db1fd45] %[output:74c7b215] %[output:654071b7] %[output:1a0f64d4] %[output:69c511c7] %[output:80c50658] %[output:521ce776] %[output:6abfb90a] %[output:49258648] %[output:467e4aca] %[output:6e1a86eb] %[output:5eea11aa] %[output:747bc9b1] %[output:0a0fe026] %[output:3dff3184] %[output:3b6a102e] %[output:90476dad] %[output:4c128056] %[output:4d184a5c] %[output:189227c0] %[output:39ba7841] %[output:7c592cf0] %[output:8c86dca4] %[output:41b2baf4] %[output:84fd4b08] %[output:2ce57674] %[output:9803f580] %[output:4b4afd09] %[output:4b3de16c] %[output:7d213b50] %[output:00d641d0] %[output:6ce5ffd8] %[output:9ee1f7e4] %[output:56361963] %[output:35f19933] %[output:7f01428e] %[output:0fa9183e] %[output:0bf73890] %[output:58d76c03] %[output:63fe1cbf] %[output:8bae86ce] %[output:25455714] %[output:51438166] %[output:09c8c0a0] %[output:76141baf] %[output:4ecabd20] %[output:37ba81ed] %[output:296220cb] %[output:85737333] %[output:88775070] %[output:5dea9ab5] %[output:5c7b048f] %[output:7053153a] %[output:084c6bae] %[output:6b80aade] %[output:920fbb93] %[output:30756905] %[output:5f1a920e] %[output:9b5cb80a] %[output:09bba6c0] %[output:9284807a] %[output:17999174] %[output:1e371406] %[output:5444af52] %[output:4e07268e] %[output:3e8b13ca] %[output:605ff620] %[output:31369086] %[output:572767b5] %[output:971f3ed3] %[output:81e55fff] %[output:95a00284] %[output:51e9f898] %[output:448c9078] %[output:8e5fea78] %[output:7307a7df] %[output:366b2194] %[output:9cb564cd] %[output:9da4f824] %[output:02dcff7f] %[output:11080f9f] %[output:6087ca7c] %[output:9171ca81] %[output:6eda7cb5] %[output:58f8fccf] %[output:9a066e9f] %[output:15d4e80f] %[output:392b24cd] %[output:99a0520f] %[output:72bb2e0a] %[output:51372b91] %[output:0e0a21e9] %[output:1b40466f] %[output:9665fe6f] %[output:79f52345] %[output:4c6e0a04] %[output:4a771709] %[output:6f63bb9e] %[output:5beff449] %[output:7dbbd20d] %[output:2685e72a] %[output:707a11a3] %[output:8ec25ece] %[output:75ab77b0] %[output:8a991d2b] %[output:081f865d] %[output:75bfbbed] %[output:501736cc] %[output:699cbdea] %[output:3dd4bc93] %[output:6cde6acf] %[output:9a1bc941] %[output:3afe39b3] %[output:1684ec2d] %[output:5f1a718a] %[output:49b33c6e] %[output:8b6c8d64] %[output:4d6b2bc8] %[output:618c1793] %[output:3f4f78f2] %[output:8a6f0cf8] %[output:6f32ee91] %[output:38bab324] %[output:34c87860] %[output:9432b1fe] %[output:9823e54b] %[output:3ba13a05] %[output:314da832] %[output:88d51177] %[output:9c5ad102] %[output:97889ee2] %[output:11d9dde7] %[output:59da6fed] %[output:99376ae3] %[output:9b3f564f] %[output:5111034d] %[output:73677cba] %[output:84a1c31e] %[output:139d7b57] %[output:48d8a20e] %[output:782bb660] %[output:7ad79bda] %[output:07705166] %[output:51aacec6] %[output:4d0d3fa4] %[output:9069e957] %[output:8510bcb5] %[output:52fe6de5] %[output:7f5b687f] %[output:17689972] %[output:94236b53] %[output:71de17e9] %[output:7a589bc7] %[output:1eeb0a4c] %[output:509ab337] %[output:8e64767f] %[output:2c37d46c] %[output:4c553f76] %[output:4d80af44] %[output:53be6a8e] %[output:4e8cf01d] %[output:25607ab4] %[output:8aede320] %[output:77d2493d] %[output:390d1015] %[output:9c55235f] %[output:88f6cb63] %[output:24dce2b7] %[output:5a4a9c5f] %[output:886aea0b] %[output:4e5c6c95] %[output:2bc02e69] %[output:69baabec] %[output:0bfc7764] %[output:7e79f7b7] %[output:2b46a3f3] %[output:95f476b4] %[output:62248664] %[output:7665abf6] %[output:77fa26fc] %[output:1704c9b1] %[output:59928606] %[output:9554411f] %[output:64e41fcb] %[output:1668efdb] %[output:338f8083] %[output:8b4d0d9a] %[output:4b6cc66d] %[output:068739bf] %[output:7fd9b34c] %[output:05ca75b7] %[output:84846fe8] %[output:33b2fa90] %[output:4be59cf3] %[output:2ad7b632] %[output:7b93a34d] %[output:3ff98b25] %[output:720bbf08] %[output:4c94c0df] %[output:74c63ce7] %[output:0ea1556d] %[output:6e4ae90a] %[output:4ab8bed5] %[output:03ef4d44] %[output:31437bc9] %[output:10b1afd6] %[output:6b647426] %[output:18f55326] %[output:1c008433] %[output:831451bb] %[output:995c9a36] %[output:3b7e0b1c] %[output:44f95a0a] %[output:2ca643cb] %[output:302c0cd0] %[output:27c00cfa] %[output:4bbd852d] %[output:4aad7518] %[output:69f09dd7] %[output:2d31eba6] %[output:0be852e5] %[output:5b3381ba] %[output:7568c2ca] %[output:62de3957] %[output:86cfe797] %[output:18ee7b6e] %[output:5c3bffc5] %[output:339338b8] %[output:9f25a6ad] %[output:22d78126] %[output:07d6ecf4] %[output:55ec4c8a] %[output:2ba25360] %[output:02df5bbb] %[output:4e6fc9b0] %[output:0f215664] %[output:8628a909] %[output:10eb2dc6] %[output:4553bef2] %[output:1031648f] %[output:3b096af5] %[output:1451c82d] + end + % Compute and log validation loss + valPred = meshgraphnet.meshgraphnet(p,Xvb,Xevb,Bvb,cfg); + valLoss(epoch) = double(extractdata(gather(mean((valPred-Yvb).^2,"all")))); + valIter(epoch) = iter; + recordMetrics(monitor, iter, ValidationLoss=valLoss(epoch)); %[output:8efe7c47] + fprintf(">> Epoch %d validation loss: %.4f\n", epoch, valLoss(epoch)); %[output:360ecf12] %[output:23bd0954] %[output:907bd166] %[output:1c762b07] %[output:0106249c] %[output:40928f55] %[output:7c6d92a1] %[output:11e4fd48] %[output:1a354d81] %[output:6dc8045d] %[output:2235f927] %[output:677a6e6b] %[output:7dd568a6] %[output:3a486642] %[output:1698649c] %[output:2f072a60] %[output:056c2c20] %[output:2137142f] %[output:6fd19edf] %[output:19ac1f21] %[output:74ccec9e] %[output:2c6526b2] %[output:8bf6db19] %[output:4cf183f5] %[output:20a62e59] %[output:1e3e292a] %[output:3f340bd3] %[output:34961fc8] %[output:858fae5a] %[output:286a721a] %[output:17157ff2] %[output:0697fd3c] %[output:964dbeb2] %[output:21875818] %[output:2c9fee55] %[output:24033b38] %[output:91359628] %[output:747f7326] %[output:0ed9eee7] %[output:65ba71ea] %[output:5092ec9d] %[output:093122bd] %[output:91664549] %[output:4e7d3724] %[output:054e6030] %[output:11b53b2a] %[output:65ec5c79] %[output:01f6d859] %[output:0e7dd202] %[output:585e4155] %[output:6e325963] %[output:128e68b5] %[output:0b1648fa] %[output:1b00417e] %[output:3dd8f80c] %[output:56684393] %[output:41097202] %[output:612b4908] %[output:9ab0a38a] %[output:3ffc5c33] %[output:82e2f9b4] %[output:2a095bb8] %[output:7d4b16d2] %[output:9d1a4a94] %[output:206aae00] %[output:9a189315] %[output:13f50c39] %[output:9c8f0d6a] %[output:2b3e43c3] %[output:49b55408] %[output:3596c36c] %[output:33a03ef6] %[output:8670b51c] %[output:13fd599d] %[output:046fc08a] %[output:7d8fa1d9] %[output:149b1853] %[output:64743cc5] %[output:0e698a3c] %[output:7d856e8d] %[output:843b70e3] %[output:67bae44b] %[output:1daec627] %[output:4efd54a7] %[output:09a37f06] %[output:2aa50591] %[output:94037ed4] %[output:02293857] %[output:9358c128] %[output:25820879] %[output:61ea05f6] %[output:045b74af] %[output:015a1954] %[output:73be09fa] %[output:69da0b2e] %[output:38181c2e] %[output:390c9784] %[output:1ce2cef7] %[output:683036bc] %[output:421760ee] %[output:7ad504df] %[output:3d73f615] %[output:262fd194] %[output:4a763068] %[output:22fad00a] %[output:428b823b] %[output:8f616aa8] %[output:6efa2340] %[output:6683307d] %[output:3a37c612] %[output:37c3c5b3] %[output:123a8c2a] %[output:0c96e117] %[output:77384522] %[output:7a845ec5] %[output:89e1eac0] %[output:27e1e622] %[output:3d5deeff] %[output:233f0a9e] %[output:31e769b7] %[output:8f9fec08] %[output:068740da] %[output:5548fd11] %[output:123a2d18] %[output:676f7048] %[output:393ba012] %[output:10beede1] %[output:39b35cec] %[output:4b4ba426] %[output:07a485e3] %[output:17923bfa] %[output:07b75789] %[output:68aad416] %[output:9ea02d93] %[output:0e329285] %[output:62d63ed5] %[output:67007505] %[output:5dbf1b45] %[output:81ee8730] %[output:976c9ab3] %[output:1cd9a3bd] %[output:60d0216c] %[output:71625090] %[output:1c036f02] %[output:75b2c1f3] %[output:801dabeb] %[output:852359ab] %[output:23c9ef94] %[output:89075115] %[output:6dd7b792] %[output:93408908] %[output:237859ab] %[output:25084d74] %[output:299c9459] %[output:7b68406a] %[output:0a38329a] %[output:8dc0adc2] %[output:4e3bedf4] %[output:7eb29abe] %[output:45e3cc9c] %[output:7521c589] %[output:156a892d] %[output:67da58b4] %[output:0591c023] %[output:44f26859] %[output:92a3ee44] %[output:6523e727] %[output:51305e72] %[output:2e8f6998] %[output:7cb89f58] %[output:93217151] %[output:92e234ed] %[output:519c44e7] %[output:03356122] %[output:01de9fb3] %[output:3062520d] %[output:9fdc266c] %[output:4511a275] %[output:85427654] %[output:53320511] %[output:1bdff14a] %[output:18d99f7f] %[output:6bce0e7d] %[output:14e87826] %[output:562cbc1d] %[output:1f0cdcc9] %[output:60aa40f7] %[output:8cf7f8a7] %[output:56002f24] %[output:92a3d8a2] %[output:762e98bd] %[output:95bf697f] %[output:46ffaf31] %[output:86d6200c] %[output:66f4739a] %[output:690f8572] %[output:09fa35d6] %[output:1921d716] %[output:560c46a2] %[output:158a2958] %[output:5be6a734] %[output:669452f8] %[output:6532506f] %[output:168fd3da] %[output:264b7315] %[output:2cfcd015] %[output:55f67715] %[output:924dfbe9] %[output:5568e291] %[output:7418a86d] %[output:9add4627] %[output:874eaa08] %[output:80a43eea] %[output:89d7f4d6] %[output:2212ccd6] %[output:4aca68fb] %[output:40ab6cf7] %[output:8aca6872] %[output:19b876f7] %[output:86d054d6] %[output:51bf7e05] %[output:9b07118b] %[output:26c4e02c] %[output:90c35afe] %[output:5d1ba28c] %[output:52a4ee15] %[output:25dea215] %[output:175ffb7e] %[output:7673353c] %[output:8511c6e7] %[output:8adf1ef2] %[output:8cf10a47] %[output:473e54f9] %[output:635b23f8] %[output:4b904e95] %[output:671adcfd] %[output:1d7006cf] %[output:4f069e1b] %[output:8925769b] %[output:4b12787c] %[output:74bdb1f6] %[output:1db4dd50] %[output:1c410eed] %[output:6b5c915e] %[output:74ca347c] %[output:42201a8c] %[output:290dbb85] %[output:60e57fc0] %[output:1f5501cc] %[output:11c26538] %[output:0583ed22] %[output:191c8a62] %[output:607270f2] %[output:5622e938] %[output:3be41725] %[output:4a9bc037] %[output:00066e3b] %[output:926ecf1c] %[output:0db85c76] %[output:07058dce] %[output:4207ba2b] %[output:39199c17] %[output:853642df] %[output:362e52a4] %[output:12f62da3] %[output:9b0b913c] %[output:975173df] %[output:1b65c002] %[output:53b87dba] %[output:556f3d01] %[output:0c2e92f6] %[output:54b0d143] %[output:17584135] %[output:1934d832] %[output:7ca352b0] %[output:9efdf189] %[output:64dc73e9] %[output:2497201a] %[output:78dd5006] %[output:86826ee9] %[output:084f3f73] %[output:4f5552f7] %[output:503c0319] %[output:579ebf63] %[output:0c7235ac] %[output:2f3c8b96] %[output:0bb4a76a] %[output:473e8680] %[output:4a6fe78b] %[output:45423fc0] %[output:437d8fd7] %[output:0adf8e1c] %[output:8b7c85e1] %[output:7e7c53a5] %[output:71e592c2] %[output:9379e722] %[output:59195fd4] %[output:821a7639] %[output:1a227dc6] %[output:057ff863] + % Track best params by validation loss + if valLoss(epoch) < bestValLoss + bestValLoss = valLoss(epoch); + pBest = p; + end + updateInfo(monitor, Epoch=epoch + " of " + epochs, LearnRate=currentLR); %[output:8efe7c47] + if monitor.Stop, break; end + end +%[text] After training, store the training and validation histories and save the model parameters corresponding to the best validation performance. +% Build tables to store the training and validation info + TrainingHistory = table(trainIter,trainEpoch,trainLR,trainLoss,... + VariableNames={'Iteration','Epoch','LearnRate','Loss'}); + ValidationHistory = table(valIter,valLoss,... + VariableNames={'Iteration','Loss'}); + info = struct(); + info.TrainingHistory = TrainingHistory; + info.ValidationHistory = ValidationHistory; + cfg.TrainingInfo = info; + % Save best parameters by validation loss + p = pBest; + % Save the network and informational struct cfg + mName = cfg.ModelName; + baseName = strjoin([mName, ts], "_"); + outpath = fullfile(outdir,baseName + ".mat"); + save(outpath,"p","cfg"); +else + % Skip training when loading pretrained model +end %[output:group:8481ed1c] +%% +%[text] ## Test the Network +%[text] Evaluate the trained model on randomly selected samples from the validation set and compare predictions against the original FEA solutions. +if doTrain + % Skip loading when the model was just trained +else + % Load in pretrained network and configuration + fName = fullfile(outdir, loadFile + ".mat"); + warning('off'); + S = load(fName,"p","cfg"); + warning('on'); + % Use gather on gpuArrays if running on CPU + if useGpu == false + S.p = utils.convertGpuToCpu(S.p); + end + p = S.p; cfg = S.cfg; +end +%[text] Choose a small number of random samples from the validation set to view the results and compare against FEA. +numTests = 3; +idxTest = randi([1,numel(cfg.IdxVal)],numTests,1); +for i=1:numTests %[output:group:0bb8ebe1] + % Load the corresponding FEA results for the selected sample + fileIdx = cfg.IdxVal(idxTest(i)); + testFileName = sprintf("EM_results_%03d.mat",fileIdx); + load(testFileName); + + % Prepare network inputs + bushing = findNodes(D.Mesh,"region",Cell=2); + [Xs,Xes,Ys,As,Bs] = meshgraphnet.preprocessData(D,bushing); + + % Normalize inputs using training-set statistics + normXs = utils.applyScale(Xs,cfg.InputOffset,cfg.InputScale); + normXes = utils.applyScale(Xes,cfg.EdgeOffset,cfg.EdgeScale); + Bs = blkdiag(Bs); + Bs = Bs ./ sum(Bs,2); + + % Predict electric potential + Ypred = meshgraphnet.meshgraphnet(p,normXs,normXes,Bs,cfg); + Ypred = extractdata(Ypred); + if useGpu + Ypred = gather(Ypred); + end + + % Rescale predictions back to physical units + Ypred = utils.applyInverseScale(Ypred,cfg.OutputOffset,cfg.OutputScale); + + % Visualization and error analysis + % Plot results on the transformer bushing only (exclude surrounding air) + elemsBushing = findElements(D.Mesh,"region",Cell=2); + nodesBushing = findNodes(D.Mesh,"region",Cell=2); + clims = [min(D.Potential), max(D.Potential)]; + figure(); %[output:49191cf1] %[output:66a349c3] %[output:005a8275] + tiledlayout(1,3); %[output:49191cf1] %[output:66a349c3] %[output:005a8275] + % FEA solution + nexttile; + pdeplot3D(D.Mesh.Nodes, ... + D.Mesh.Elements(:,elemsBushing), ... + ColorMapData=D.Potential); + title('FEA Solution'); clim(clims); + % AI prediction + nexttile; + % Our MGN makes predictions on the bushing nodes only, pad the rest with zeros + numNodes = size(D.Mesh.Nodes,2); + paddedYpred = zeros(numNodes,1); + paddedYpred(nodesBushing) = Ypred; + pdeplot3D(D.Mesh.Nodes, ... + D.Mesh.Elements(:,elemsBushing), ... + ColorMapData=paddedYpred); + title('AI Prediction'); clim(clims); + % Compute the relative L2 error on the bushing nodes + relErrorL2 = norm(Ypred'-D.Potential(nodesBushing))/norm(D.Potential(nodesBushing)); + title(sprintf('AI Prediction: Relative L2 error = %0.1e', relErrorL2)); + % Plot the relative error + nexttile; + absError = abs(paddedYpred-D.Potential); + relError = absError./max(D.Potential(nodesBushing)); + pdeplot3D(D.Mesh.Nodes, ... + D.Mesh.Elements(:,elemsBushing), ... + ColorMapData=relError); + title('Relative Error') + clim([min(relError(nodesBushing)) max(relError(nodesBushing))]) +end %[output:group:0bb8ebe1] +%% +%[text] ## Make Inference on New Design +%[text] This section demonstrates how to use the trained MeshGraphNet model to make predictions on a new design for which no high-fidelity simulation data is available. Given a new geometry defined by an STL file, the model predicts the electric potential field directly from geometry information. This enables rapid evaluation of new designs without running a full electrostatic simulation. +%[text] Load the STL file representing the transformer bushing insulator geometry. The STL file is assumed to be located in the `STL/` subdirectory. +gmBushing = fegeometry("STL/TEST_bushing.stl"); +%[text] Visualize the bushing insulator geometry. +figure(); %[output:6e6a25b7] +pdegplot(gmBushing); %[output:6e6a25b7] +title("New Bushing Insulator Geometry"); %[output:6e6a25b7] +%% +%[text] Create an electrostatic FEA model. +model = femodel(AnalysisType="electrostatic",Geometry=gmBushing); +%[text] Define points in the domain at which predictions are made. Do this by generating a mesh. +model = generateMesh(model,Hmax=0.025); +%% +%[text] Prepare the inputs. +%[text] Extract the mesh and preprocess to build graph. +data.Mesh = model.Mesh; +keep = []; % Keep all nodes since only passing the mesh on the bushing +[Xs,Xes,~,~,Bs] = meshgraphnet.preprocessData(data,keep); +%[text] Normalize inputs using training-set statistics. +normalXs = utils.applyScale(Xs,cfg.InputOffset,cfg.InputScale); +normalXes = utils.applyScale(Xes,cfg.EdgeOffset,cfg.EdgeScale); +Bs = blkdiag(Bs); +Bs = Bs ./ sum(Bs,2); +%[text] Predict electric potential. +PredPotential = meshgraphnet.meshgraphnet(p,normalXs,normalXes,Bs,cfg); +%[text] Rescale predictions back to physical units. +PredPotential = utils.applyInverseScale(PredPotential,cfg.OutputOffset,cfg.OutputScale); +PredPotential = extractdata(PredPotential)'; +if useGpu + PredPotential = gather(PredPotential); +end +%[text] Visualize predicted electric potential on the bushing insulator. +figure(); %[output:6f426dbf] +pdeplot3D(model.Mesh, ... %[output:6f426dbf] + ColorMapData=PredPotential); %[output:6f426dbf] +title('AI Predicted Electric Potential') %[output:6f426dbf] +%% +%[text] ## Derive the Electric Field from the Predicted Potential +%[text] In the FEA workflow, the electric field is derived from the potential: $\\mathbf{E} = -\\nabla V$. We follow the same approach here, computing the gradient of the AI-predicted potential using FEA shape function derivatives on the tetrahedral mesh (see the helper function `electricFieldFromTets.m` in `src/+utils`). +out = utils.electricFieldFromTets(model.Mesh.Nodes,model.Mesh.Elements,PredPotential); +Emag = out.Emag_node; +pdeplot3D(model.Mesh, ... %[output:677e5848] + ColorMapData=Emag); %[output:677e5848] +title('Electric field magnitude derived from AI predictions') %[output:677e5848] +%% +%[text] %[text:anchor:TMP_36ef] **References** +%[text] %[text:anchor:M_5694] 1. Pfaff, Tobias, Meire Fortunato, Alvaro Sanchez-Gonzalez, and Peter W. Battaglia. "Learning Mesh-Based Simulation with Graph Networks." arXiv, June 18, 2021. [https://arxiv.org/abs/2010.03409](https://arxiv.org/abs/2010.03409). \ + +%[appendix]{"version":"1.0"} +%--- +%[metadata:view] +% data: {"layout":"inline","rightPanelPercent":21} +%--- +%[text:image:372c] +% data: {"align":"baseline","height":325,"src":"data:image\/png;base64,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","width":614} +%--- +%[control:checkbox:9e9f] +% data: {"defaultValue":false,"label":"doTrain","run":"Section"} +%--- +%[output:2388b472] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 1, Loss: 0.7690\nIter: 2, Epoch: 1, Loss: 3.3390\nIter: 3, Epoch: 1, Loss: 0.3615\nIter: 4, Epoch: 1, Loss: 0.6648\nIter: 5, Epoch: 1, Loss: 0.7819\nIter: 6, Epoch: 1, Loss: 0.6871\nIter: 7, Epoch: 1, Loss: 0.4630\nIter: 8, Epoch: 1, Loss: 0.2442\nIter: 9, Epoch: 1, Loss: 0.3812\nIter: 10, Epoch: 1, Loss: 0.3051\nIter: 11, Epoch: 1, Loss: 0.1136\nIter: 12, Epoch: 1, Loss: 0.1054\nIter: 13, Epoch: 1, Loss: 0.1158\nIter: 14, Epoch: 1, Loss: 0.1308\nIter: 15, Epoch: 1, Loss: 0.1340\nIter: 16, Epoch: 1, Loss: 0.1182\nIter: 17, Epoch: 1, Loss: 0.0935\nIter: 18, Epoch: 1, Loss: 0.0733\nIter: 19, Epoch: 1, Loss: 0.0586\nIter: 20, Epoch: 1, Loss: 0.0457\nIter: 21, Epoch: 1, Loss: 0.0431\nIter: 22, Epoch: 1, Loss: 0.0490\nIter: 23, Epoch: 1, Loss: 0.0497\nIter: 24, Epoch: 1, Loss: 0.0491\nIter: 25, Epoch: 1, Loss: 0.0378\nIter: 26, Epoch: 1, Loss: 0.0358\nIter: 27, Epoch: 1, Loss: 0.0384\nIter: 28, Epoch: 1, Loss: 0.0338\nIter: 29, Epoch: 1, Loss: 0.0326\nIter: 30, Epoch: 1, Loss: 0.0231\n","truncated":false}} +%--- +%[output:360ecf12] +% data: {"dataType":"text","outputData":{"text":">> Epoch 1 validation loss: 0.0214\n","truncated":false}} +%--- +%[output:9e95aef6] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 2, Loss: 0.0226\nIter: 2, Epoch: 2, Loss: 0.0175\nIter: 3, Epoch: 2, Loss: 0.0170\nIter: 4, Epoch: 2, Loss: 0.0225\nIter: 5, Epoch: 2, Loss: 0.0220\nIter: 6, Epoch: 2, Loss: 0.0228\nIter: 7, Epoch: 2, Loss: 0.0192\nIter: 8, Epoch: 2, Loss: 0.0221\nIter: 9, Epoch: 2, Loss: 0.0140\nIter: 10, Epoch: 2, Loss: 0.0172\nIter: 11, Epoch: 2, Loss: 0.0128\nIter: 12, Epoch: 2, Loss: 0.0135\nIter: 13, Epoch: 2, Loss: 0.0113\nIter: 14, Epoch: 2, Loss: 0.0149\nIter: 15, Epoch: 2, Loss: 0.0137\nIter: 16, Epoch: 2, Loss: 0.0139\nIter: 17, Epoch: 2, Loss: 0.0098\nIter: 18, Epoch: 2, Loss: 0.0122\nIter: 19, Epoch: 2, Loss: 0.0129\nIter: 20, Epoch: 2, Loss: 0.0107\nIter: 21, Epoch: 2, Loss: 0.0092\nIter: 22, Epoch: 2, Loss: 0.0095\nIter: 23, Epoch: 2, Loss: 0.0111\nIter: 24, Epoch: 2, Loss: 0.0097\nIter: 25, Epoch: 2, Loss: 0.0078\nIter: 26, Epoch: 2, Loss: 0.0085\nIter: 27, Epoch: 2, Loss: 0.0088\nIter: 28, Epoch: 2, Loss: 0.0088\nIter: 29, Epoch: 2, Loss: 0.0077\nIter: 30, Epoch: 2, Loss: 0.0071\n","truncated":false}} +%--- +%[output:23bd0954] +% data: {"dataType":"text","outputData":{"text":">> Epoch 2 validation loss: 0.0085\n","truncated":false}} +%--- +%[output:453605ac] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 3, Loss: 0.0097\nIter: 2, Epoch: 3, Loss: 0.0070\nIter: 3, Epoch: 3, Loss: 0.0063\nIter: 4, Epoch: 3, Loss: 0.0090\nIter: 5, Epoch: 3, Loss: 0.0067\nIter: 6, Epoch: 3, Loss: 0.0074\nIter: 7, Epoch: 3, Loss: 0.0061\nIter: 8, Epoch: 3, Loss: 0.0086\nIter: 9, Epoch: 3, Loss: 0.0057\nIter: 10, Epoch: 3, Loss: 0.0070\nIter: 11, Epoch: 3, Loss: 0.0062\nIter: 12, Epoch: 3, Loss: 0.0075\nIter: 13, Epoch: 3, Loss: 0.0065\nIter: 14, Epoch: 3, Loss: 0.0082\nIter: 15, Epoch: 3, Loss: 0.0066\nIter: 16, Epoch: 3, Loss: 0.0082\nIter: 17, Epoch: 3, Loss: 0.0058\nIter: 18, Epoch: 3, Loss: 0.0067\nIter: 19, Epoch: 3, Loss: 0.0073\nIter: 20, Epoch: 3, Loss: 0.0061\nIter: 21, Epoch: 3, Loss: 0.0052\nIter: 22, Epoch: 3, Loss: 0.0055\nIter: 23, Epoch: 3, Loss: 0.0065\nIter: 24, Epoch: 3, Loss: 0.0065\nIter: 25, Epoch: 3, Loss: 0.0054\nIter: 26, Epoch: 3, Loss: 0.0059\nIter: 27, Epoch: 3, Loss: 0.0057\nIter: 28, Epoch: 3, Loss: 0.0067\nIter: 29, Epoch: 3, Loss: 0.0065\nIter: 30, Epoch: 3, Loss: 0.0046\n","truncated":false}} +%--- +%[output:907bd166] +% data: {"dataType":"text","outputData":{"text":">> Epoch 3 validation loss: 0.0065\n","truncated":false}} +%--- +%[output:187ab2a1] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 4, Loss: 0.0082\nIter: 2, Epoch: 4, Loss: 0.0054\nIter: 3, Epoch: 4, Loss: 0.0044\nIter: 4, Epoch: 4, Loss: 0.0066\nIter: 5, Epoch: 4, Loss: 0.0052\nIter: 6, Epoch: 4, Loss: 0.0053\nIter: 7, Epoch: 4, Loss: 0.0043\nIter: 8, Epoch: 4, Loss: 0.0072\nIter: 9, Epoch: 4, Loss: 0.0040\nIter: 10, Epoch: 4, Loss: 0.0056\nIter: 11, Epoch: 4, Loss: 0.0045\nIter: 12, Epoch: 4, Loss: 0.0059\nIter: 13, Epoch: 4, Loss: 0.0059\nIter: 14, Epoch: 4, Loss: 0.0062\nIter: 15, Epoch: 4, Loss: 0.0052\nIter: 16, Epoch: 4, Loss: 0.0069\nIter: 17, Epoch: 4, Loss: 0.0053\nIter: 18, Epoch: 4, Loss: 0.0052\nIter: 19, Epoch: 4, Loss: 0.0068\nIter: 20, Epoch: 4, Loss: 0.0054\nIter: 21, Epoch: 4, Loss: 0.0041\nIter: 22, Epoch: 4, Loss: 0.0041\nIter: 23, Epoch: 4, Loss: 0.0059\nIter: 24, Epoch: 4, Loss: 0.0057\nIter: 25, Epoch: 4, Loss: 0.0041\nIter: 26, Epoch: 4, Loss: 0.0046\nIter: 27, Epoch: 4, Loss: 0.0049\nIter: 28, Epoch: 4, Loss: 0.0059\nIter: 29, Epoch: 4, Loss: 0.0048\nIter: 30, Epoch: 4, Loss: 0.0038\n","truncated":false}} +%--- +%[output:1c762b07] +% data: {"dataType":"text","outputData":{"text":">> Epoch 4 validation loss: 0.0067\n","truncated":false}} +%--- +%[output:29eb44cd] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 5, Loss: 0.0090\nIter: 2, Epoch: 5, Loss: 0.0045\nIter: 3, Epoch: 5, Loss: 0.0034\nIter: 4, Epoch: 5, Loss: 0.0068\nIter: 5, Epoch: 5, Loss: 0.0049\nIter: 6, Epoch: 5, Loss: 0.0041\nIter: 7, Epoch: 5, Loss: 0.0035\nIter: 8, Epoch: 5, Loss: 0.0079\nIter: 9, Epoch: 5, Loss: 0.0034\nIter: 10, Epoch: 5, Loss: 0.0042\nIter: 11, Epoch: 5, Loss: 0.0045\nIter: 12, Epoch: 5, Loss: 0.0054\nIter: 13, Epoch: 5, Loss: 0.0048\nIter: 14, Epoch: 5, Loss: 0.0050\nIter: 15, Epoch: 5, Loss: 0.0052\nIter: 16, Epoch: 5, Loss: 0.0063\nIter: 17, Epoch: 5, Loss: 0.0041\nIter: 18, Epoch: 5, Loss: 0.0047\nIter: 19, Epoch: 5, Loss: 0.0072\nIter: 20, Epoch: 5, Loss: 0.0045\nIter: 21, Epoch: 5, Loss: 0.0035\nIter: 22, Epoch: 5, Loss: 0.0036\nIter: 23, Epoch: 5, Loss: 0.0057\nIter: 24, Epoch: 5, Loss: 0.0047\nIter: 25, Epoch: 5, Loss: 0.0031\nIter: 26, Epoch: 5, Loss: 0.0042\nIter: 27, Epoch: 5, Loss: 0.0043\nIter: 28, Epoch: 5, Loss: 0.0044\nIter: 29, Epoch: 5, Loss: 0.0035\nIter: 30, Epoch: 5, Loss: 0.0036\n","truncated":false}} +%--- +%[output:0106249c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 5 validation loss: 0.0058\n","truncated":false}} +%--- +%[output:9665f29d] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 6, Loss: 0.0080\nIter: 2, Epoch: 6, Loss: 0.0034\nIter: 3, Epoch: 6, Loss: 0.0033\nIter: 4, Epoch: 6, Loss: 0.0065\nIter: 5, Epoch: 6, Loss: 0.0038\nIter: 6, Epoch: 6, Loss: 0.0035\nIter: 7, Epoch: 6, Loss: 0.0033\nIter: 8, Epoch: 6, Loss: 0.0073\nIter: 9, Epoch: 6, Loss: 0.0030\nIter: 10, Epoch: 6, Loss: 0.0036\nIter: 11, Epoch: 6, Loss: 0.0042\nIter: 12, Epoch: 6, Loss: 0.0043\nIter: 13, Epoch: 6, Loss: 0.0037\nIter: 14, Epoch: 6, Loss: 0.0045\nIter: 15, Epoch: 6, Loss: 0.0047\nIter: 16, Epoch: 6, Loss: 0.0049\nIter: 17, Epoch: 6, Loss: 0.0031\nIter: 18, Epoch: 6, Loss: 0.0046\nIter: 19, Epoch: 6, Loss: 0.0058\nIter: 20, Epoch: 6, Loss: 0.0035\nIter: 21, Epoch: 6, Loss: 0.0032\nIter: 22, Epoch: 6, Loss: 0.0033\nIter: 23, Epoch: 6, Loss: 0.0047\nIter: 24, Epoch: 6, Loss: 0.0038\nIter: 25, Epoch: 6, Loss: 0.0026\nIter: 26, Epoch: 6, Loss: 0.0036\nIter: 27, Epoch: 6, Loss: 0.0034\nIter: 28, Epoch: 6, Loss: 0.0035\nIter: 29, Epoch: 6, Loss: 0.0030\nIter: 30, Epoch: 6, Loss: 0.0029\n","truncated":false}} +%--- +%[output:40928f55] +% data: {"dataType":"text","outputData":{"text":">> Epoch 6 validation loss: 0.0046\n","truncated":false}} +%--- +%[output:2dd92824] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 7, Loss: 0.0061\nIter: 2, Epoch: 7, Loss: 0.0028\nIter: 3, Epoch: 7, Loss: 0.0029\nIter: 4, Epoch: 7, Loss: 0.0053\nIter: 5, Epoch: 7, Loss: 0.0029\nIter: 6, Epoch: 7, Loss: 0.0032\nIter: 7, Epoch: 7, Loss: 0.0029\nIter: 8, Epoch: 7, Loss: 0.0061\nIter: 9, Epoch: 7, Loss: 0.0027\nIter: 10, Epoch: 7, Loss: 0.0030\nIter: 11, Epoch: 7, Loss: 0.0034\nIter: 12, Epoch: 7, Loss: 0.0035\nIter: 13, Epoch: 7, Loss: 0.0033\nIter: 14, Epoch: 7, Loss: 0.0040\nIter: 15, Epoch: 7, Loss: 0.0037\nIter: 16, Epoch: 7, Loss: 0.0038\nIter: 17, Epoch: 7, Loss: 0.0029\nIter: 18, Epoch: 7, Loss: 0.0041\nIter: 19, Epoch: 7, Loss: 0.0045\nIter: 20, Epoch: 7, Loss: 0.0029\nIter: 21, Epoch: 7, Loss: 0.0029\nIter: 22, Epoch: 7, Loss: 0.0032\nIter: 23, Epoch: 7, Loss: 0.0040\nIter: 24, Epoch: 7, Loss: 0.0031\nIter: 25, Epoch: 7, Loss: 0.0023\nIter: 26, Epoch: 7, Loss: 0.0032\nIter: 27, Epoch: 7, Loss: 0.0030\nIter: 28, Epoch: 7, Loss: 0.0029\nIter: 29, Epoch: 7, Loss: 0.0027\nIter: 30, Epoch: 7, Loss: 0.0025\n","truncated":false}} +%--- +%[output:7c6d92a1] +% data: {"dataType":"text","outputData":{"text":">> Epoch 7 validation loss: 0.0038\n","truncated":false}} +%--- +%[output:9e79b3a9] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 8, Loss: 0.0048\nIter: 2, Epoch: 8, Loss: 0.0025\nIter: 3, Epoch: 8, Loss: 0.0025\nIter: 4, Epoch: 8, Loss: 0.0041\nIter: 5, Epoch: 8, Loss: 0.0023\nIter: 6, Epoch: 8, Loss: 0.0029\nIter: 7, Epoch: 8, Loss: 0.0025\nIter: 8, Epoch: 8, Loss: 0.0050\nIter: 9, Epoch: 8, Loss: 0.0025\nIter: 10, Epoch: 8, Loss: 0.0024\nIter: 11, Epoch: 8, Loss: 0.0027\nIter: 12, Epoch: 8, Loss: 0.0030\nIter: 13, Epoch: 8, Loss: 0.0028\nIter: 14, Epoch: 8, Loss: 0.0033\nIter: 15, Epoch: 8, Loss: 0.0029\nIter: 16, Epoch: 8, Loss: 0.0033\nIter: 17, Epoch: 8, Loss: 0.0028\nIter: 18, Epoch: 8, Loss: 0.0033\nIter: 19, Epoch: 8, Loss: 0.0037\nIter: 20, Epoch: 8, Loss: 0.0026\nIter: 21, Epoch: 8, Loss: 0.0027\nIter: 22, Epoch: 8, Loss: 0.0030\nIter: 23, Epoch: 8, Loss: 0.0032\nIter: 24, Epoch: 8, Loss: 0.0025\nIter: 25, Epoch: 8, Loss: 0.0022\nIter: 26, Epoch: 8, Loss: 0.0031\nIter: 27, Epoch: 8, Loss: 0.0027\nIter: 28, Epoch: 8, Loss: 0.0025\nIter: 29, Epoch: 8, Loss: 0.0027\nIter: 30, Epoch: 8, Loss: 0.0024\n","truncated":false}} +%--- +%[output:11e4fd48] +% data: {"dataType":"text","outputData":{"text":">> Epoch 8 validation loss: 0.0033\n","truncated":false}} +%--- +%[output:145f82f0] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 9, Loss: 0.0041\nIter: 2, Epoch: 9, Loss: 0.0021\nIter: 3, Epoch: 9, Loss: 0.0021\nIter: 4, Epoch: 9, Loss: 0.0033\nIter: 5, Epoch: 9, Loss: 0.0020\nIter: 6, Epoch: 9, Loss: 0.0025\nIter: 7, Epoch: 9, Loss: 0.0020\nIter: 8, Epoch: 9, Loss: 0.0042\nIter: 9, Epoch: 9, Loss: 0.0025\nIter: 10, Epoch: 9, Loss: 0.0020\nIter: 11, Epoch: 9, Loss: 0.0023\nIter: 12, Epoch: 9, Loss: 0.0024\nIter: 13, Epoch: 9, Loss: 0.0022\nIter: 14, Epoch: 9, Loss: 0.0027\nIter: 15, Epoch: 9, Loss: 0.0025\nIter: 16, Epoch: 9, Loss: 0.0028\nIter: 17, Epoch: 9, Loss: 0.0024\nIter: 18, Epoch: 9, Loss: 0.0027\nIter: 19, Epoch: 9, Loss: 0.0034\nIter: 20, Epoch: 9, Loss: 0.0024\nIter: 21, Epoch: 9, Loss: 0.0024\nIter: 22, Epoch: 9, Loss: 0.0026\nIter: 23, Epoch: 9, Loss: 0.0026\nIter: 24, Epoch: 9, Loss: 0.0021\nIter: 25, Epoch: 9, Loss: 0.0021\nIter: 26, Epoch: 9, Loss: 0.0027\nIter: 27, Epoch: 9, Loss: 0.0022\nIter: 28, Epoch: 9, Loss: 0.0022\nIter: 29, Epoch: 9, Loss: 0.0028\nIter: 30, Epoch: 9, Loss: 0.0022\n","truncated":false}} +%--- +%[output:1a354d81] +% data: {"dataType":"text","outputData":{"text":">> Epoch 9 validation loss: 0.0030\n","truncated":false}} +%--- +%[output:5161a939] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 10, Loss: 0.0039\nIter: 2, Epoch: 10, Loss: 0.0019\nIter: 3, Epoch: 10, Loss: 0.0019\nIter: 4, Epoch: 10, Loss: 0.0029\nIter: 5, Epoch: 10, Loss: 0.0018\nIter: 6, Epoch: 10, Loss: 0.0022\nIter: 7, Epoch: 10, Loss: 0.0018\nIter: 8, Epoch: 10, Loss: 0.0037\nIter: 9, Epoch: 10, Loss: 0.0025\nIter: 10, Epoch: 10, Loss: 0.0018\nIter: 11, Epoch: 10, Loss: 0.0019\nIter: 12, Epoch: 10, Loss: 0.0020\nIter: 13, Epoch: 10, Loss: 0.0019\nIter: 14, Epoch: 10, Loss: 0.0024\nIter: 15, Epoch: 10, Loss: 0.0021\nIter: 16, Epoch: 10, Loss: 0.0024\nIter: 17, Epoch: 10, Loss: 0.0021\nIter: 18, Epoch: 10, Loss: 0.0023\nIter: 19, Epoch: 10, Loss: 0.0030\nIter: 20, Epoch: 10, Loss: 0.0022\nIter: 21, Epoch: 10, Loss: 0.0022\nIter: 22, Epoch: 10, Loss: 0.0024\nIter: 23, Epoch: 10, Loss: 0.0023\nIter: 24, Epoch: 10, Loss: 0.0019\nIter: 25, Epoch: 10, Loss: 0.0020\nIter: 26, Epoch: 10, Loss: 0.0024\nIter: 27, Epoch: 10, Loss: 0.0020\nIter: 28, Epoch: 10, Loss: 0.0021\nIter: 29, Epoch: 10, Loss: 0.0028\nIter: 30, Epoch: 10, Loss: 0.0020\n","truncated":false}} +%--- +%[output:6dc8045d] +% data: {"dataType":"text","outputData":{"text":">> Epoch 10 validation loss: 0.0029\n","truncated":false}} +%--- +%[output:5fcb025f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 11, Loss: 0.0039\nIter: 2, Epoch: 11, Loss: 0.0017\nIter: 3, Epoch: 11, Loss: 0.0018\nIter: 4, Epoch: 11, Loss: 0.0027\nIter: 5, Epoch: 11, Loss: 0.0016\nIter: 6, Epoch: 11, Loss: 0.0020\nIter: 7, Epoch: 11, Loss: 0.0016\nIter: 8, Epoch: 11, Loss: 0.0034\nIter: 9, Epoch: 11, Loss: 0.0024\nIter: 10, Epoch: 11, Loss: 0.0017\nIter: 11, Epoch: 11, Loss: 0.0017\nIter: 12, Epoch: 11, Loss: 0.0018\nIter: 13, Epoch: 11, Loss: 0.0017\nIter: 14, Epoch: 11, Loss: 0.0022\nIter: 15, Epoch: 11, Loss: 0.0018\nIter: 16, Epoch: 11, Loss: 0.0021\nIter: 17, Epoch: 11, Loss: 0.0019\nIter: 18, Epoch: 11, Loss: 0.0020\nIter: 19, Epoch: 11, Loss: 0.0027\nIter: 20, Epoch: 11, Loss: 0.0020\nIter: 21, Epoch: 11, Loss: 0.0021\nIter: 22, Epoch: 11, Loss: 0.0022\nIter: 23, Epoch: 11, Loss: 0.0021\nIter: 24, Epoch: 11, Loss: 0.0017\nIter: 25, Epoch: 11, Loss: 0.0019\nIter: 26, Epoch: 11, Loss: 0.0023\nIter: 27, Epoch: 11, Loss: 0.0018\nIter: 28, Epoch: 11, Loss: 0.0020\nIter: 29, Epoch: 11, Loss: 0.0028\nIter: 30, Epoch: 11, Loss: 0.0019\n","truncated":false}} +%--- +%[output:2235f927] +% data: {"dataType":"text","outputData":{"text":">> Epoch 11 validation loss: 0.0028\n","truncated":false}} +%--- +%[output:5efc4a41] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 12, Loss: 0.0040\nIter: 2, Epoch: 12, Loss: 0.0016\nIter: 3, Epoch: 12, Loss: 0.0018\nIter: 4, Epoch: 12, Loss: 0.0026\nIter: 5, Epoch: 12, Loss: 0.0015\nIter: 6, Epoch: 12, Loss: 0.0019\nIter: 7, Epoch: 12, Loss: 0.0015\nIter: 8, Epoch: 12, Loss: 0.0031\nIter: 9, Epoch: 12, Loss: 0.0024\nIter: 10, Epoch: 12, Loss: 0.0017\nIter: 11, Epoch: 12, Loss: 0.0015\nIter: 12, Epoch: 12, Loss: 0.0017\nIter: 13, Epoch: 12, Loss: 0.0016\nIter: 14, Epoch: 12, Loss: 0.0020\nIter: 15, Epoch: 12, Loss: 0.0017\nIter: 16, Epoch: 12, Loss: 0.0019\nIter: 17, Epoch: 12, Loss: 0.0019\nIter: 18, Epoch: 12, Loss: 0.0019\nIter: 19, Epoch: 12, Loss: 0.0025\nIter: 20, Epoch: 12, Loss: 0.0019\nIter: 21, Epoch: 12, Loss: 0.0020\nIter: 22, Epoch: 12, Loss: 0.0022\nIter: 23, Epoch: 12, Loss: 0.0020\nIter: 24, Epoch: 12, Loss: 0.0016\nIter: 25, Epoch: 12, Loss: 0.0019\nIter: 26, Epoch: 12, Loss: 0.0021\nIter: 27, Epoch: 12, Loss: 0.0016\nIter: 28, Epoch: 12, Loss: 0.0020\nIter: 29, Epoch: 12, Loss: 0.0027\nIter: 30, Epoch: 12, Loss: 0.0018\n","truncated":false}} +%--- +%[output:677a6e6b] +% data: {"dataType":"text","outputData":{"text":">> Epoch 12 validation loss: 0.0029\n","truncated":false}} +%--- +%[output:2e1e5c6b] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 13, Loss: 0.0042\nIter: 2, Epoch: 13, Loss: 0.0016\nIter: 3, Epoch: 13, Loss: 0.0018\nIter: 4, Epoch: 13, Loss: 0.0025\nIter: 5, Epoch: 13, Loss: 0.0014\nIter: 6, Epoch: 13, Loss: 0.0019\nIter: 7, Epoch: 13, Loss: 0.0015\nIter: 8, Epoch: 13, Loss: 0.0029\nIter: 9, Epoch: 13, Loss: 0.0025\nIter: 10, Epoch: 13, Loss: 0.0017\nIter: 11, Epoch: 13, Loss: 0.0014\nIter: 12, Epoch: 13, Loss: 0.0016\nIter: 13, Epoch: 13, Loss: 0.0016\nIter: 14, Epoch: 13, Loss: 0.0020\nIter: 15, Epoch: 13, Loss: 0.0015\nIter: 16, Epoch: 13, Loss: 0.0018\nIter: 17, Epoch: 13, Loss: 0.0019\nIter: 18, Epoch: 13, Loss: 0.0019\nIter: 19, Epoch: 13, Loss: 0.0023\nIter: 20, Epoch: 13, Loss: 0.0018\nIter: 21, Epoch: 13, Loss: 0.0019\nIter: 22, Epoch: 13, Loss: 0.0021\nIter: 23, Epoch: 13, Loss: 0.0019\nIter: 24, Epoch: 13, Loss: 0.0015\nIter: 25, Epoch: 13, Loss: 0.0019\nIter: 26, Epoch: 13, Loss: 0.0021\nIter: 27, Epoch: 13, Loss: 0.0015\nIter: 28, Epoch: 13, Loss: 0.0020\nIter: 29, Epoch: 13, Loss: 0.0027\nIter: 30, Epoch: 13, Loss: 0.0017\n","truncated":false}} +%--- +%[output:7dd568a6] +% data: {"dataType":"text","outputData":{"text":">> Epoch 13 validation loss: 0.0028\n","truncated":false}} +%--- +%[output:20cf326e] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 14, Loss: 0.0042\nIter: 2, Epoch: 14, Loss: 0.0016\nIter: 3, Epoch: 14, Loss: 0.0020\nIter: 4, Epoch: 14, Loss: 0.0023\nIter: 5, Epoch: 14, Loss: 0.0014\nIter: 6, Epoch: 14, Loss: 0.0020\nIter: 7, Epoch: 14, Loss: 0.0015\nIter: 8, Epoch: 14, Loss: 0.0027\nIter: 9, Epoch: 14, Loss: 0.0024\nIter: 10, Epoch: 14, Loss: 0.0016\nIter: 11, Epoch: 14, Loss: 0.0013\nIter: 12, Epoch: 14, Loss: 0.0016\nIter: 13, Epoch: 14, Loss: 0.0016\nIter: 14, Epoch: 14, Loss: 0.0019\nIter: 15, Epoch: 14, Loss: 0.0014\nIter: 16, Epoch: 14, Loss: 0.0017\nIter: 17, Epoch: 14, Loss: 0.0020\nIter: 18, Epoch: 14, Loss: 0.0018\nIter: 19, Epoch: 14, Loss: 0.0023\nIter: 20, Epoch: 14, Loss: 0.0017\nIter: 21, Epoch: 14, Loss: 0.0019\nIter: 22, Epoch: 14, Loss: 0.0020\nIter: 23, Epoch: 14, Loss: 0.0018\nIter: 24, Epoch: 14, Loss: 0.0015\nIter: 25, Epoch: 14, Loss: 0.0019\nIter: 26, Epoch: 14, Loss: 0.0019\nIter: 27, Epoch: 14, Loss: 0.0014\nIter: 28, Epoch: 14, Loss: 0.0019\nIter: 29, Epoch: 14, Loss: 0.0025\nIter: 30, Epoch: 14, Loss: 0.0017\n","truncated":false}} +%--- +%[output:3a486642] +% data: {"dataType":"text","outputData":{"text":">> Epoch 14 validation loss: 0.0027\n","truncated":false}} +%--- +%[output:0c0cfeda] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 15, Loss: 0.0041\nIter: 2, Epoch: 15, Loss: 0.0016\nIter: 3, Epoch: 15, Loss: 0.0019\nIter: 4, Epoch: 15, Loss: 0.0021\nIter: 5, Epoch: 15, Loss: 0.0013\nIter: 6, Epoch: 15, Loss: 0.0020\nIter: 7, Epoch: 15, Loss: 0.0014\nIter: 8, Epoch: 15, Loss: 0.0025\nIter: 9, Epoch: 15, Loss: 0.0024\nIter: 10, Epoch: 15, Loss: 0.0016\nIter: 11, Epoch: 15, Loss: 0.0013\nIter: 12, Epoch: 15, Loss: 0.0015\nIter: 13, Epoch: 15, Loss: 0.0016\nIter: 14, Epoch: 15, Loss: 0.0018\nIter: 15, Epoch: 15, Loss: 0.0014\nIter: 16, Epoch: 15, Loss: 0.0017\nIter: 17, Epoch: 15, Loss: 0.0020\nIter: 18, Epoch: 15, Loss: 0.0018\nIter: 19, Epoch: 15, Loss: 0.0024\nIter: 20, Epoch: 15, Loss: 0.0016\nIter: 21, Epoch: 15, Loss: 0.0020\nIter: 22, Epoch: 15, Loss: 0.0019\nIter: 23, Epoch: 15, Loss: 0.0018\nIter: 24, Epoch: 15, Loss: 0.0014\nIter: 25, Epoch: 15, Loss: 0.0019\nIter: 26, Epoch: 15, Loss: 0.0019\nIter: 27, Epoch: 15, Loss: 0.0013\nIter: 28, Epoch: 15, Loss: 0.0019\nIter: 29, Epoch: 15, Loss: 0.0025\nIter: 30, Epoch: 15, Loss: 0.0016\n","truncated":false}} +%--- +%[output:1698649c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 15 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:36e06638] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 16, Loss: 0.0040\nIter: 2, Epoch: 16, Loss: 0.0016\nIter: 3, Epoch: 16, Loss: 0.0019\nIter: 4, Epoch: 16, Loss: 0.0020\nIter: 5, Epoch: 16, Loss: 0.0013\nIter: 6, Epoch: 16, Loss: 0.0019\nIter: 7, Epoch: 16, Loss: 0.0013\nIter: 8, Epoch: 16, Loss: 0.0024\nIter: 9, Epoch: 16, Loss: 0.0023\nIter: 10, Epoch: 16, Loss: 0.0015\nIter: 11, Epoch: 16, Loss: 0.0013\nIter: 12, Epoch: 16, Loss: 0.0014\nIter: 13, Epoch: 16, Loss: 0.0015\nIter: 14, Epoch: 16, Loss: 0.0018\nIter: 15, Epoch: 16, Loss: 0.0013\nIter: 16, Epoch: 16, Loss: 0.0016\nIter: 17, Epoch: 16, Loss: 0.0020\nIter: 18, Epoch: 16, Loss: 0.0018\nIter: 19, Epoch: 16, Loss: 0.0024\nIter: 20, Epoch: 16, Loss: 0.0016\nIter: 21, Epoch: 16, Loss: 0.0021\nIter: 22, Epoch: 16, Loss: 0.0018\nIter: 23, Epoch: 16, Loss: 0.0017\nIter: 24, Epoch: 16, Loss: 0.0014\nIter: 25, Epoch: 16, Loss: 0.0018\nIter: 26, Epoch: 16, Loss: 0.0018\nIter: 27, Epoch: 16, Loss: 0.0012\nIter: 28, Epoch: 16, Loss: 0.0019\nIter: 29, Epoch: 16, Loss: 0.0025\nIter: 30, Epoch: 16, Loss: 0.0015\n","truncated":false}} +%--- +%[output:2f072a60] +% data: {"dataType":"text","outputData":{"text":">> Epoch 16 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:1b282164] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 17, Loss: 0.0041\nIter: 2, Epoch: 17, Loss: 0.0015\nIter: 3, Epoch: 17, Loss: 0.0018\nIter: 4, Epoch: 17, Loss: 0.0019\nIter: 5, Epoch: 17, Loss: 0.0013\nIter: 6, Epoch: 17, Loss: 0.0019\nIter: 7, Epoch: 17, Loss: 0.0012\nIter: 8, Epoch: 17, Loss: 0.0023\nIter: 9, Epoch: 17, Loss: 0.0023\nIter: 10, Epoch: 17, Loss: 0.0015\nIter: 11, Epoch: 17, Loss: 0.0012\nIter: 12, Epoch: 17, Loss: 0.0014\nIter: 13, Epoch: 17, Loss: 0.0015\nIter: 14, Epoch: 17, Loss: 0.0018\nIter: 15, Epoch: 17, Loss: 0.0013\nIter: 16, Epoch: 17, Loss: 0.0016\nIter: 17, Epoch: 17, Loss: 0.0021\nIter: 18, Epoch: 17, Loss: 0.0017\nIter: 19, Epoch: 17, Loss: 0.0024\nIter: 20, Epoch: 17, Loss: 0.0016\nIter: 21, Epoch: 17, Loss: 0.0022\nIter: 22, Epoch: 17, Loss: 0.0017\nIter: 23, Epoch: 17, Loss: 0.0017\nIter: 24, Epoch: 17, Loss: 0.0015\nIter: 25, Epoch: 17, Loss: 0.0018\nIter: 26, Epoch: 17, Loss: 0.0017\nIter: 27, Epoch: 17, Loss: 0.0012\nIter: 28, Epoch: 17, Loss: 0.0019\nIter: 29, Epoch: 17, Loss: 0.0025\nIter: 30, Epoch: 17, Loss: 0.0014\n","truncated":false}} +%--- +%[output:056c2c20] +% data: {"dataType":"text","outputData":{"text":">> Epoch 17 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:3599dc82] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 18, Loss: 0.0042\nIter: 2, Epoch: 18, Loss: 0.0015\nIter: 3, Epoch: 18, Loss: 0.0017\nIter: 4, Epoch: 18, Loss: 0.0019\nIter: 5, Epoch: 18, Loss: 0.0013\nIter: 6, Epoch: 18, Loss: 0.0019\nIter: 7, Epoch: 18, Loss: 0.0012\nIter: 8, Epoch: 18, Loss: 0.0022\nIter: 9, Epoch: 18, Loss: 0.0023\nIter: 10, Epoch: 18, Loss: 0.0015\nIter: 11, Epoch: 18, Loss: 0.0012\nIter: 12, Epoch: 18, Loss: 0.0013\nIter: 13, Epoch: 18, Loss: 0.0015\nIter: 14, Epoch: 18, Loss: 0.0018\nIter: 15, Epoch: 18, Loss: 0.0013\nIter: 16, Epoch: 18, Loss: 0.0016\nIter: 17, Epoch: 18, Loss: 0.0021\nIter: 18, Epoch: 18, Loss: 0.0017\nIter: 19, Epoch: 18, Loss: 0.0024\nIter: 20, Epoch: 18, Loss: 0.0016\nIter: 21, Epoch: 18, Loss: 0.0022\nIter: 22, Epoch: 18, Loss: 0.0017\nIter: 23, Epoch: 18, Loss: 0.0017\nIter: 24, Epoch: 18, Loss: 0.0015\nIter: 25, Epoch: 18, Loss: 0.0018\nIter: 26, Epoch: 18, Loss: 0.0016\nIter: 27, Epoch: 18, Loss: 0.0012\nIter: 28, Epoch: 18, Loss: 0.0019\nIter: 29, Epoch: 18, Loss: 0.0025\nIter: 30, Epoch: 18, Loss: 0.0014\n","truncated":false}} +%--- +%[output:2137142f] +% data: {"dataType":"text","outputData":{"text":">> Epoch 18 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:8b7126e7] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 19, Loss: 0.0041\nIter: 2, Epoch: 19, Loss: 0.0014\nIter: 3, Epoch: 19, Loss: 0.0017\nIter: 4, Epoch: 19, Loss: 0.0018\nIter: 5, Epoch: 19, Loss: 0.0013\nIter: 6, Epoch: 19, Loss: 0.0018\nIter: 7, Epoch: 19, Loss: 0.0012\nIter: 8, Epoch: 19, Loss: 0.0021\nIter: 9, Epoch: 19, Loss: 0.0023\nIter: 10, Epoch: 19, Loss: 0.0014\nIter: 11, Epoch: 19, Loss: 0.0012\nIter: 12, Epoch: 19, Loss: 0.0013\nIter: 13, Epoch: 19, Loss: 0.0015\nIter: 14, Epoch: 19, Loss: 0.0018\nIter: 15, Epoch: 19, Loss: 0.0012\nIter: 16, Epoch: 19, Loss: 0.0016\nIter: 17, Epoch: 19, Loss: 0.0020\nIter: 18, Epoch: 19, Loss: 0.0017\nIter: 19, Epoch: 19, Loss: 0.0023\nIter: 20, Epoch: 19, Loss: 0.0016\nIter: 21, Epoch: 19, Loss: 0.0022\nIter: 22, Epoch: 19, Loss: 0.0017\nIter: 23, Epoch: 19, Loss: 0.0017\nIter: 24, Epoch: 19, Loss: 0.0015\nIter: 25, Epoch: 19, Loss: 0.0017\nIter: 26, Epoch: 19, Loss: 0.0016\nIter: 27, Epoch: 19, Loss: 0.0011\nIter: 28, Epoch: 19, Loss: 0.0019\nIter: 29, Epoch: 19, Loss: 0.0025\nIter: 30, Epoch: 19, Loss: 0.0014\n","truncated":false}} +%--- +%[output:6fd19edf] +% data: {"dataType":"text","outputData":{"text":">> Epoch 19 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:30e10416] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 20, Loss: 0.0042\nIter: 2, Epoch: 20, Loss: 0.0014\nIter: 3, Epoch: 20, Loss: 0.0017\nIter: 4, Epoch: 20, Loss: 0.0018\nIter: 5, Epoch: 20, Loss: 0.0013\nIter: 6, Epoch: 20, Loss: 0.0018\nIter: 7, Epoch: 20, Loss: 0.0012\nIter: 8, Epoch: 20, Loss: 0.0021\nIter: 9, Epoch: 20, Loss: 0.0023\nIter: 10, Epoch: 20, Loss: 0.0014\nIter: 11, Epoch: 20, Loss: 0.0011\nIter: 12, Epoch: 20, Loss: 0.0013\nIter: 13, Epoch: 20, Loss: 0.0015\nIter: 14, Epoch: 20, Loss: 0.0017\nIter: 15, Epoch: 20, Loss: 0.0012\nIter: 16, Epoch: 20, Loss: 0.0016\nIter: 17, Epoch: 20, Loss: 0.0020\nIter: 18, Epoch: 20, Loss: 0.0016\nIter: 19, Epoch: 20, Loss: 0.0023\nIter: 20, Epoch: 20, Loss: 0.0016\nIter: 21, Epoch: 20, Loss: 0.0021\nIter: 22, Epoch: 20, Loss: 0.0016\nIter: 23, Epoch: 20, Loss: 0.0017\nIter: 24, Epoch: 20, Loss: 0.0015\nIter: 25, Epoch: 20, Loss: 0.0016\nIter: 26, Epoch: 20, Loss: 0.0016\nIter: 27, Epoch: 20, Loss: 0.0011\nIter: 28, Epoch: 20, Loss: 0.0018\nIter: 29, Epoch: 20, Loss: 0.0025\nIter: 30, Epoch: 20, Loss: 0.0013\n","truncated":false}} +%--- +%[output:19ac1f21] +% data: {"dataType":"text","outputData":{"text":">> Epoch 20 validation loss: 0.0027\n","truncated":false}} +%--- +%[output:519e33c4] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 21, Loss: 0.0043\nIter: 2, Epoch: 21, Loss: 0.0014\nIter: 3, Epoch: 21, Loss: 0.0017\nIter: 4, Epoch: 21, Loss: 0.0018\nIter: 5, Epoch: 21, Loss: 0.0013\nIter: 6, Epoch: 21, Loss: 0.0018\nIter: 7, Epoch: 21, Loss: 0.0011\nIter: 8, Epoch: 21, Loss: 0.0020\nIter: 9, Epoch: 21, Loss: 0.0023\nIter: 10, Epoch: 21, Loss: 0.0014\nIter: 11, Epoch: 21, Loss: 0.0011\nIter: 12, Epoch: 21, Loss: 0.0012\nIter: 13, Epoch: 21, Loss: 0.0015\nIter: 14, Epoch: 21, Loss: 0.0017\nIter: 15, Epoch: 21, Loss: 0.0011\nIter: 16, Epoch: 21, Loss: 0.0015\nIter: 17, Epoch: 21, Loss: 0.0019\nIter: 18, Epoch: 21, Loss: 0.0016\nIter: 19, Epoch: 21, Loss: 0.0022\nIter: 20, Epoch: 21, Loss: 0.0015\nIter: 21, Epoch: 21, Loss: 0.0021\nIter: 22, Epoch: 21, Loss: 0.0016\nIter: 23, Epoch: 21, Loss: 0.0017\nIter: 24, Epoch: 21, Loss: 0.0014\nIter: 25, Epoch: 21, Loss: 0.0016\nIter: 26, Epoch: 21, Loss: 0.0016\nIter: 27, Epoch: 21, Loss: 0.0011\nIter: 28, Epoch: 21, Loss: 0.0018\nIter: 29, Epoch: 21, Loss: 0.0025\nIter: 30, Epoch: 21, Loss: 0.0013\n","truncated":false}} +%--- +%[output:74ccec9e] +% data: {"dataType":"text","outputData":{"text":">> Epoch 21 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:46b75fc8] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 22, Loss: 0.0042\nIter: 2, Epoch: 22, Loss: 0.0014\nIter: 3, Epoch: 22, Loss: 0.0017\nIter: 4, Epoch: 22, Loss: 0.0017\nIter: 5, Epoch: 22, Loss: 0.0013\nIter: 6, Epoch: 22, Loss: 0.0018\nIter: 7, Epoch: 22, Loss: 0.0012\nIter: 8, Epoch: 22, Loss: 0.0020\nIter: 9, Epoch: 22, Loss: 0.0022\nIter: 10, Epoch: 22, Loss: 0.0014\nIter: 11, Epoch: 22, Loss: 0.0011\nIter: 12, Epoch: 22, Loss: 0.0012\nIter: 13, Epoch: 22, Loss: 0.0015\nIter: 14, Epoch: 22, Loss: 0.0017\nIter: 15, Epoch: 22, Loss: 0.0011\nIter: 16, Epoch: 22, Loss: 0.0015\nIter: 17, Epoch: 22, Loss: 0.0019\nIter: 18, Epoch: 22, Loss: 0.0016\nIter: 19, Epoch: 22, Loss: 0.0022\nIter: 20, Epoch: 22, Loss: 0.0015\nIter: 21, Epoch: 22, Loss: 0.0020\nIter: 22, Epoch: 22, Loss: 0.0016\nIter: 23, Epoch: 22, Loss: 0.0018\nIter: 24, Epoch: 22, Loss: 0.0014\nIter: 25, Epoch: 22, Loss: 0.0016\nIter: 26, Epoch: 22, Loss: 0.0016\nIter: 27, Epoch: 22, Loss: 0.0011\nIter: 28, Epoch: 22, Loss: 0.0018\nIter: 29, Epoch: 22, Loss: 0.0025\nIter: 30, Epoch: 22, Loss: 0.0013\n","truncated":false}} +%--- +%[output:2c6526b2] +% data: {"dataType":"text","outputData":{"text":">> Epoch 22 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:12e0435c] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 23, Loss: 0.0042\nIter: 2, Epoch: 23, Loss: 0.0014\nIter: 3, Epoch: 23, Loss: 0.0016\nIter: 4, Epoch: 23, Loss: 0.0017\nIter: 5, Epoch: 23, Loss: 0.0013\nIter: 6, Epoch: 23, Loss: 0.0018\nIter: 7, Epoch: 23, Loss: 0.0012\nIter: 8, Epoch: 23, Loss: 0.0019\nIter: 9, Epoch: 23, Loss: 0.0022\nIter: 10, Epoch: 23, Loss: 0.0014\nIter: 11, Epoch: 23, Loss: 0.0011\nIter: 12, Epoch: 23, Loss: 0.0012\nIter: 13, Epoch: 23, Loss: 0.0015\nIter: 14, Epoch: 23, Loss: 0.0016\nIter: 15, Epoch: 23, Loss: 0.0011\nIter: 16, Epoch: 23, Loss: 0.0015\nIter: 17, Epoch: 23, Loss: 0.0018\nIter: 18, Epoch: 23, Loss: 0.0016\nIter: 19, Epoch: 23, Loss: 0.0021\nIter: 20, Epoch: 23, Loss: 0.0015\nIter: 21, Epoch: 23, Loss: 0.0020\nIter: 22, Epoch: 23, Loss: 0.0016\nIter: 23, Epoch: 23, Loss: 0.0017\nIter: 24, Epoch: 23, Loss: 0.0014\nIter: 25, Epoch: 23, Loss: 0.0015\nIter: 26, Epoch: 23, Loss: 0.0016\nIter: 27, Epoch: 23, Loss: 0.0010\nIter: 28, Epoch: 23, Loss: 0.0017\nIter: 29, Epoch: 23, Loss: 0.0024\nIter: 30, Epoch: 23, Loss: 0.0012\n","truncated":false}} +%--- +%[output:8bf6db19] +% data: {"dataType":"text","outputData":{"text":">> Epoch 23 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:48d67fce] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 24, Loss: 0.0042\nIter: 2, Epoch: 24, Loss: 0.0014\nIter: 3, Epoch: 24, Loss: 0.0016\nIter: 4, Epoch: 24, Loss: 0.0017\nIter: 5, Epoch: 24, Loss: 0.0012\nIter: 6, Epoch: 24, Loss: 0.0018\nIter: 7, Epoch: 24, Loss: 0.0012\nIter: 8, Epoch: 24, Loss: 0.0019\nIter: 9, Epoch: 24, Loss: 0.0022\nIter: 10, Epoch: 24, Loss: 0.0014\nIter: 11, Epoch: 24, Loss: 0.0011\nIter: 12, Epoch: 24, Loss: 0.0011\nIter: 13, Epoch: 24, Loss: 0.0015\nIter: 14, Epoch: 24, Loss: 0.0016\nIter: 15, Epoch: 24, Loss: 0.0011\nIter: 16, Epoch: 24, Loss: 0.0014\nIter: 17, Epoch: 24, Loss: 0.0018\nIter: 18, Epoch: 24, Loss: 0.0016\nIter: 19, Epoch: 24, Loss: 0.0021\nIter: 20, Epoch: 24, Loss: 0.0015\nIter: 21, Epoch: 24, Loss: 0.0019\nIter: 22, Epoch: 24, Loss: 0.0016\nIter: 23, Epoch: 24, Loss: 0.0017\nIter: 24, Epoch: 24, Loss: 0.0013\nIter: 25, Epoch: 24, Loss: 0.0015\nIter: 26, Epoch: 24, Loss: 0.0016\nIter: 27, Epoch: 24, Loss: 0.0010\nIter: 28, Epoch: 24, Loss: 0.0017\nIter: 29, Epoch: 24, Loss: 0.0024\nIter: 30, Epoch: 24, Loss: 0.0012\n","truncated":false}} +%--- +%[output:4cf183f5] +% data: {"dataType":"text","outputData":{"text":">> Epoch 24 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:2ad740aa] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 25, Loss: 0.0042\nIter: 2, Epoch: 25, Loss: 0.0014\nIter: 3, Epoch: 25, Loss: 0.0017\nIter: 4, Epoch: 25, Loss: 0.0016\nIter: 5, Epoch: 25, Loss: 0.0012\nIter: 6, Epoch: 25, Loss: 0.0017\nIter: 7, Epoch: 25, Loss: 0.0011\nIter: 8, Epoch: 25, Loss: 0.0019\nIter: 9, Epoch: 25, Loss: 0.0021\nIter: 10, Epoch: 25, Loss: 0.0013\nIter: 11, Epoch: 25, Loss: 0.0011\nIter: 12, Epoch: 25, Loss: 0.0011\nIter: 13, Epoch: 25, Loss: 0.0015\nIter: 14, Epoch: 25, Loss: 0.0016\nIter: 15, Epoch: 25, Loss: 0.0011\nIter: 16, Epoch: 25, Loss: 0.0014\nIter: 17, Epoch: 25, Loss: 0.0018\nIter: 18, Epoch: 25, Loss: 0.0015\nIter: 19, Epoch: 25, Loss: 0.0021\nIter: 20, Epoch: 25, Loss: 0.0014\nIter: 21, Epoch: 25, Loss: 0.0019\nIter: 22, Epoch: 25, Loss: 0.0015\nIter: 23, Epoch: 25, Loss: 0.0018\nIter: 24, Epoch: 25, Loss: 0.0013\nIter: 25, Epoch: 25, Loss: 0.0014\nIter: 26, Epoch: 25, Loss: 0.0015\nIter: 27, Epoch: 25, Loss: 0.0010\nIter: 28, Epoch: 25, Loss: 0.0017\nIter: 29, Epoch: 25, Loss: 0.0024\nIter: 30, Epoch: 25, Loss: 0.0012\n","truncated":false}} +%--- +%[output:20a62e59] +% data: {"dataType":"text","outputData":{"text":">> Epoch 25 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:48e6ee01] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 26, Loss: 0.0042\nIter: 2, Epoch: 26, Loss: 0.0013\nIter: 3, Epoch: 26, Loss: 0.0016\nIter: 4, Epoch: 26, Loss: 0.0016\nIter: 5, Epoch: 26, Loss: 0.0012\nIter: 6, Epoch: 26, Loss: 0.0017\nIter: 7, Epoch: 26, Loss: 0.0011\nIter: 8, Epoch: 26, Loss: 0.0019\nIter: 9, Epoch: 26, Loss: 0.0021\nIter: 10, Epoch: 26, Loss: 0.0013\nIter: 11, Epoch: 26, Loss: 0.0011\nIter: 12, Epoch: 26, Loss: 0.0011\nIter: 13, Epoch: 26, Loss: 0.0014\nIter: 14, Epoch: 26, Loss: 0.0016\nIter: 15, Epoch: 26, Loss: 0.0010\nIter: 16, Epoch: 26, Loss: 0.0014\nIter: 17, Epoch: 26, Loss: 0.0017\nIter: 18, Epoch: 26, Loss: 0.0015\nIter: 19, Epoch: 26, Loss: 0.0020\nIter: 20, Epoch: 26, Loss: 0.0014\nIter: 21, Epoch: 26, Loss: 0.0018\nIter: 22, Epoch: 26, Loss: 0.0015\nIter: 23, Epoch: 26, Loss: 0.0018\nIter: 24, Epoch: 26, Loss: 0.0013\nIter: 25, Epoch: 26, Loss: 0.0014\nIter: 26, Epoch: 26, Loss: 0.0016\nIter: 27, Epoch: 26, Loss: 0.0009\nIter: 28, Epoch: 26, Loss: 0.0017\nIter: 29, Epoch: 26, Loss: 0.0023\nIter: 30, Epoch: 26, Loss: 0.0012\n","truncated":false}} +%--- +%[output:1e3e292a] +% data: {"dataType":"text","outputData":{"text":">> Epoch 26 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:6e129df8] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 27, Loss: 0.0042\nIter: 2, Epoch: 27, Loss: 0.0013\nIter: 3, Epoch: 27, Loss: 0.0016\nIter: 4, Epoch: 27, Loss: 0.0016\nIter: 5, Epoch: 27, Loss: 0.0012\nIter: 6, Epoch: 27, Loss: 0.0017\nIter: 7, Epoch: 27, Loss: 0.0011\nIter: 8, Epoch: 27, Loss: 0.0018\nIter: 9, Epoch: 27, Loss: 0.0021\nIter: 10, Epoch: 27, Loss: 0.0013\nIter: 11, Epoch: 27, Loss: 0.0011\nIter: 12, Epoch: 27, Loss: 0.0011\nIter: 13, Epoch: 27, Loss: 0.0014\nIter: 14, Epoch: 27, Loss: 0.0016\nIter: 15, Epoch: 27, Loss: 0.0010\nIter: 16, Epoch: 27, Loss: 0.0014\nIter: 17, Epoch: 27, Loss: 0.0017\nIter: 18, Epoch: 27, Loss: 0.0015\nIter: 19, Epoch: 27, Loss: 0.0020\nIter: 20, Epoch: 27, Loss: 0.0014\nIter: 21, Epoch: 27, Loss: 0.0018\nIter: 22, Epoch: 27, Loss: 0.0015\nIter: 23, Epoch: 27, Loss: 0.0018\nIter: 24, Epoch: 27, Loss: 0.0012\nIter: 25, Epoch: 27, Loss: 0.0013\nIter: 26, Epoch: 27, Loss: 0.0015\nIter: 27, Epoch: 27, Loss: 0.0009\nIter: 28, Epoch: 27, Loss: 0.0016\nIter: 29, Epoch: 27, Loss: 0.0023\nIter: 30, Epoch: 27, Loss: 0.0011\n","truncated":false}} +%--- +%[output:3f340bd3] +% data: {"dataType":"text","outputData":{"text":">> Epoch 27 validation loss: 0.0025\n","truncated":false}} +%--- +%[output:94c9a19d] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 28, Loss: 0.0041\nIter: 2, Epoch: 28, Loss: 0.0013\nIter: 3, Epoch: 28, Loss: 0.0016\nIter: 4, Epoch: 28, Loss: 0.0016\nIter: 5, Epoch: 28, Loss: 0.0012\nIter: 6, Epoch: 28, Loss: 0.0017\nIter: 7, Epoch: 28, Loss: 0.0011\nIter: 8, Epoch: 28, Loss: 0.0018\nIter: 9, Epoch: 28, Loss: 0.0021\nIter: 10, Epoch: 28, Loss: 0.0013\nIter: 11, Epoch: 28, Loss: 0.0011\nIter: 12, Epoch: 28, Loss: 0.0011\nIter: 13, Epoch: 28, Loss: 0.0014\nIter: 14, Epoch: 28, Loss: 0.0015\nIter: 15, Epoch: 28, Loss: 0.0010\nIter: 16, Epoch: 28, Loss: 0.0013\nIter: 17, Epoch: 28, Loss: 0.0017\nIter: 18, Epoch: 28, Loss: 0.0015\nIter: 19, Epoch: 28, Loss: 0.0020\nIter: 20, Epoch: 28, Loss: 0.0014\nIter: 21, Epoch: 28, Loss: 0.0017\nIter: 22, Epoch: 28, Loss: 0.0014\nIter: 23, Epoch: 28, Loss: 0.0018\nIter: 24, Epoch: 28, Loss: 0.0012\nIter: 25, Epoch: 28, Loss: 0.0013\nIter: 26, Epoch: 28, Loss: 0.0015\nIter: 27, Epoch: 28, Loss: 0.0009\nIter: 28, Epoch: 28, Loss: 0.0016\nIter: 29, Epoch: 28, Loss: 0.0022\nIter: 30, Epoch: 28, Loss: 0.0011\n","truncated":false}} +%--- +%[output:34961fc8] +% data: {"dataType":"text","outputData":{"text":">> Epoch 28 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:8ef8e50f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 29, Loss: 0.0042\nIter: 2, Epoch: 29, Loss: 0.0013\nIter: 3, Epoch: 29, Loss: 0.0016\nIter: 4, Epoch: 29, Loss: 0.0016\nIter: 5, Epoch: 29, Loss: 0.0012\nIter: 6, Epoch: 29, Loss: 0.0017\nIter: 7, Epoch: 29, Loss: 0.0011\nIter: 8, Epoch: 29, Loss: 0.0018\nIter: 9, Epoch: 29, Loss: 0.0021\nIter: 10, Epoch: 29, Loss: 0.0013\nIter: 11, Epoch: 29, Loss: 0.0011\nIter: 12, Epoch: 29, Loss: 0.0011\nIter: 13, Epoch: 29, Loss: 0.0014\nIter: 14, Epoch: 29, Loss: 0.0015\nIter: 15, Epoch: 29, Loss: 0.0010\nIter: 16, Epoch: 29, Loss: 0.0013\nIter: 17, Epoch: 29, Loss: 0.0016\nIter: 18, Epoch: 29, Loss: 0.0015\nIter: 19, Epoch: 29, Loss: 0.0020\nIter: 20, Epoch: 29, Loss: 0.0013\nIter: 21, Epoch: 29, Loss: 0.0017\nIter: 22, Epoch: 29, Loss: 0.0015\nIter: 23, Epoch: 29, Loss: 0.0018\nIter: 24, Epoch: 29, Loss: 0.0012\nIter: 25, Epoch: 29, Loss: 0.0013\nIter: 26, Epoch: 29, Loss: 0.0015\nIter: 27, Epoch: 29, Loss: 0.0008\nIter: 28, Epoch: 29, Loss: 0.0016\nIter: 29, Epoch: 29, Loss: 0.0022\nIter: 30, Epoch: 29, Loss: 0.0011\n","truncated":false}} +%--- +%[output:858fae5a] +% data: {"dataType":"text","outputData":{"text":">> Epoch 29 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:6da734f3] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 30, Loss: 0.0042\nIter: 2, Epoch: 30, Loss: 0.0013\nIter: 3, Epoch: 30, Loss: 0.0016\nIter: 4, Epoch: 30, Loss: 0.0016\nIter: 5, Epoch: 30, Loss: 0.0012\nIter: 6, Epoch: 30, Loss: 0.0017\nIter: 7, Epoch: 30, Loss: 0.0011\nIter: 8, Epoch: 30, Loss: 0.0018\nIter: 9, Epoch: 30, Loss: 0.0020\nIter: 10, Epoch: 30, Loss: 0.0012\nIter: 11, Epoch: 30, Loss: 0.0011\nIter: 12, Epoch: 30, Loss: 0.0010\nIter: 13, Epoch: 30, Loss: 0.0014\nIter: 14, Epoch: 30, Loss: 0.0015\nIter: 15, Epoch: 30, Loss: 0.0010\nIter: 16, Epoch: 30, Loss: 0.0013\nIter: 17, Epoch: 30, Loss: 0.0016\nIter: 18, Epoch: 30, Loss: 0.0015\nIter: 19, Epoch: 30, Loss: 0.0020\nIter: 20, Epoch: 30, Loss: 0.0013\nIter: 21, Epoch: 30, Loss: 0.0016\nIter: 22, Epoch: 30, Loss: 0.0015\nIter: 23, Epoch: 30, Loss: 0.0018\nIter: 24, Epoch: 30, Loss: 0.0012\nIter: 25, Epoch: 30, Loss: 0.0012\nIter: 26, Epoch: 30, Loss: 0.0015\nIter: 27, Epoch: 30, Loss: 0.0008\nIter: 28, Epoch: 30, Loss: 0.0016\nIter: 29, Epoch: 30, Loss: 0.0022\nIter: 30, Epoch: 30, Loss: 0.0011\n","truncated":false}} +%--- +%[output:286a721a] +% data: {"dataType":"text","outputData":{"text":">> Epoch 30 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:6a80ef77] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 31, Loss: 0.0041\nIter: 2, Epoch: 31, Loss: 0.0013\nIter: 3, Epoch: 31, Loss: 0.0016\nIter: 4, Epoch: 31, Loss: 0.0016\nIter: 5, Epoch: 31, Loss: 0.0011\nIter: 6, Epoch: 31, Loss: 0.0017\nIter: 7, Epoch: 31, Loss: 0.0011\nIter: 8, Epoch: 31, Loss: 0.0017\nIter: 9, Epoch: 31, Loss: 0.0020\nIter: 10, Epoch: 31, Loss: 0.0012\nIter: 11, Epoch: 31, Loss: 0.0011\nIter: 12, Epoch: 31, Loss: 0.0010\nIter: 13, Epoch: 31, Loss: 0.0014\nIter: 14, Epoch: 31, Loss: 0.0015\nIter: 15, Epoch: 31, Loss: 0.0010\nIter: 16, Epoch: 31, Loss: 0.0013\nIter: 17, Epoch: 31, Loss: 0.0016\nIter: 18, Epoch: 31, Loss: 0.0014\nIter: 19, Epoch: 31, Loss: 0.0020\nIter: 20, Epoch: 31, Loss: 0.0013\nIter: 21, Epoch: 31, Loss: 0.0016\nIter: 22, Epoch: 31, Loss: 0.0014\nIter: 23, Epoch: 31, Loss: 0.0018\nIter: 24, Epoch: 31, Loss: 0.0011\nIter: 25, Epoch: 31, Loss: 0.0012\nIter: 26, Epoch: 31, Loss: 0.0015\nIter: 27, Epoch: 31, Loss: 0.0008\nIter: 28, Epoch: 31, Loss: 0.0015\nIter: 29, Epoch: 31, Loss: 0.0021\nIter: 30, Epoch: 31, Loss: 0.0010\n","truncated":false}} +%--- +%[output:17157ff2] +% data: {"dataType":"text","outputData":{"text":">> Epoch 31 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:56e69f64] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 32, Loss: 0.0041\nIter: 2, Epoch: 32, Loss: 0.0012\nIter: 3, Epoch: 32, Loss: 0.0016\nIter: 4, Epoch: 32, Loss: 0.0016\nIter: 5, Epoch: 32, Loss: 0.0011\nIter: 6, Epoch: 32, Loss: 0.0017\nIter: 7, Epoch: 32, Loss: 0.0011\nIter: 8, Epoch: 32, Loss: 0.0017\nIter: 9, Epoch: 32, Loss: 0.0020\nIter: 10, Epoch: 32, Loss: 0.0012\nIter: 11, Epoch: 32, Loss: 0.0011\nIter: 12, Epoch: 32, Loss: 0.0010\nIter: 13, Epoch: 32, Loss: 0.0014\nIter: 14, Epoch: 32, Loss: 0.0015\nIter: 15, Epoch: 32, Loss: 0.0010\nIter: 16, Epoch: 32, Loss: 0.0013\nIter: 17, Epoch: 32, Loss: 0.0015\nIter: 18, Epoch: 32, Loss: 0.0014\nIter: 19, Epoch: 32, Loss: 0.0020\nIter: 20, Epoch: 32, Loss: 0.0013\nIter: 21, Epoch: 32, Loss: 0.0016\nIter: 22, Epoch: 32, Loss: 0.0014\nIter: 23, Epoch: 32, Loss: 0.0018\nIter: 24, Epoch: 32, Loss: 0.0012\nIter: 25, Epoch: 32, Loss: 0.0012\nIter: 26, Epoch: 32, Loss: 0.0015\nIter: 27, Epoch: 32, Loss: 0.0008\nIter: 28, Epoch: 32, Loss: 0.0015\nIter: 29, Epoch: 32, Loss: 0.0021\nIter: 30, Epoch: 32, Loss: 0.0010\n","truncated":false}} +%--- +%[output:0697fd3c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 32 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:00d506a3] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 33, Loss: 0.0042\nIter: 2, Epoch: 33, Loss: 0.0012\nIter: 3, Epoch: 33, Loss: 0.0015\nIter: 4, Epoch: 33, Loss: 0.0016\nIter: 5, Epoch: 33, Loss: 0.0011\nIter: 6, Epoch: 33, Loss: 0.0017\nIter: 7, Epoch: 33, Loss: 0.0011\nIter: 8, Epoch: 33, Loss: 0.0018\nIter: 9, Epoch: 33, Loss: 0.0020\nIter: 10, Epoch: 33, Loss: 0.0012\nIter: 11, Epoch: 33, Loss: 0.0011\nIter: 12, Epoch: 33, Loss: 0.0010\nIter: 13, Epoch: 33, Loss: 0.0013\nIter: 14, Epoch: 33, Loss: 0.0014\nIter: 15, Epoch: 33, Loss: 0.0010\nIter: 16, Epoch: 33, Loss: 0.0013\nIter: 17, Epoch: 33, Loss: 0.0015\nIter: 18, Epoch: 33, Loss: 0.0014\nIter: 19, Epoch: 33, Loss: 0.0020\nIter: 20, Epoch: 33, Loss: 0.0013\nIter: 21, Epoch: 33, Loss: 0.0015\nIter: 22, Epoch: 33, Loss: 0.0014\nIter: 23, Epoch: 33, Loss: 0.0018\nIter: 24, Epoch: 33, Loss: 0.0011\nIter: 25, Epoch: 33, Loss: 0.0012\nIter: 26, Epoch: 33, Loss: 0.0015\nIter: 27, Epoch: 33, Loss: 0.0008\nIter: 28, Epoch: 33, Loss: 0.0015\nIter: 29, Epoch: 33, Loss: 0.0021\nIter: 30, Epoch: 33, Loss: 0.0010\n","truncated":false}} +%--- +%[output:964dbeb2] +% data: {"dataType":"text","outputData":{"text":">> Epoch 33 validation loss: 0.0026\n","truncated":false}} +%--- +%[output:154ca774] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 34, Loss: 0.0042\nIter: 2, Epoch: 34, Loss: 0.0012\nIter: 3, Epoch: 34, Loss: 0.0016\nIter: 4, Epoch: 34, Loss: 0.0016\nIter: 5, Epoch: 34, Loss: 0.0011\nIter: 6, Epoch: 34, Loss: 0.0017\nIter: 7, Epoch: 34, Loss: 0.0011\nIter: 8, Epoch: 34, Loss: 0.0018\nIter: 9, Epoch: 34, Loss: 0.0020\nIter: 10, Epoch: 34, Loss: 0.0012\nIter: 11, Epoch: 34, Loss: 0.0011\nIter: 12, Epoch: 34, Loss: 0.0010\nIter: 13, Epoch: 34, Loss: 0.0013\nIter: 14, Epoch: 34, Loss: 0.0014\nIter: 15, Epoch: 34, Loss: 0.0010\nIter: 16, Epoch: 34, Loss: 0.0013\nIter: 17, Epoch: 34, Loss: 0.0015\nIter: 18, Epoch: 34, Loss: 0.0014\nIter: 19, Epoch: 34, Loss: 0.0020\nIter: 20, Epoch: 34, Loss: 0.0013\nIter: 21, Epoch: 34, Loss: 0.0015\nIter: 22, Epoch: 34, Loss: 0.0013\nIter: 23, Epoch: 34, Loss: 0.0019\nIter: 24, Epoch: 34, Loss: 0.0012\nIter: 25, Epoch: 34, Loss: 0.0011\nIter: 26, Epoch: 34, Loss: 0.0015\nIter: 27, Epoch: 34, Loss: 0.0008\nIter: 28, Epoch: 34, Loss: 0.0015\nIter: 29, Epoch: 34, Loss: 0.0020\nIter: 30, Epoch: 34, Loss: 0.0010\n","truncated":false}} +%--- +%[output:21875818] +% data: {"dataType":"text","outputData":{"text":">> Epoch 34 validation loss: 0.0027\n","truncated":false}} +%--- +%[output:7567b303] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 35, Loss: 0.0043\nIter: 2, Epoch: 35, Loss: 0.0011\nIter: 3, Epoch: 35, Loss: 0.0015\nIter: 4, Epoch: 35, Loss: 0.0017\nIter: 5, Epoch: 35, Loss: 0.0012\nIter: 6, Epoch: 35, Loss: 0.0016\nIter: 7, Epoch: 35, Loss: 0.0011\nIter: 8, Epoch: 35, Loss: 0.0018\nIter: 9, Epoch: 35, Loss: 0.0019\nIter: 10, Epoch: 35, Loss: 0.0011\nIter: 11, Epoch: 35, Loss: 0.0011\nIter: 12, Epoch: 35, Loss: 0.0011\nIter: 13, Epoch: 35, Loss: 0.0013\nIter: 14, Epoch: 35, Loss: 0.0014\nIter: 15, Epoch: 35, Loss: 0.0010\nIter: 16, Epoch: 35, Loss: 0.0013\nIter: 17, Epoch: 35, Loss: 0.0015\nIter: 18, Epoch: 35, Loss: 0.0014\nIter: 19, Epoch: 35, Loss: 0.0020\nIter: 20, Epoch: 35, Loss: 0.0013\nIter: 21, Epoch: 35, Loss: 0.0015\nIter: 22, Epoch: 35, Loss: 0.0013\nIter: 23, Epoch: 35, Loss: 0.0019\nIter: 24, Epoch: 35, Loss: 0.0012\nIter: 25, Epoch: 35, Loss: 0.0011\nIter: 26, Epoch: 35, Loss: 0.0014\nIter: 27, Epoch: 35, Loss: 0.0008\nIter: 28, Epoch: 35, Loss: 0.0016\nIter: 29, Epoch: 35, Loss: 0.0020\nIter: 30, Epoch: 35, Loss: 0.0010\n","truncated":false}} +%--- +%[output:2c9fee55] +% data: {"dataType":"text","outputData":{"text":">> Epoch 35 validation loss: 0.0027\n","truncated":false}} +%--- +%[output:905b4938] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 36, Loss: 0.0043\nIter: 2, Epoch: 36, Loss: 0.0011\nIter: 3, Epoch: 36, Loss: 0.0015\nIter: 4, Epoch: 36, Loss: 0.0016\nIter: 5, Epoch: 36, Loss: 0.0012\nIter: 6, Epoch: 36, Loss: 0.0016\nIter: 7, Epoch: 36, Loss: 0.0010\nIter: 8, Epoch: 36, Loss: 0.0018\nIter: 9, Epoch: 36, Loss: 0.0019\nIter: 10, Epoch: 36, Loss: 0.0011\nIter: 11, Epoch: 36, Loss: 0.0011\nIter: 12, Epoch: 36, Loss: 0.0010\nIter: 13, Epoch: 36, Loss: 0.0014\nIter: 14, Epoch: 36, Loss: 0.0014\nIter: 15, Epoch: 36, Loss: 0.0010\nIter: 16, Epoch: 36, Loss: 0.0014\nIter: 17, Epoch: 36, Loss: 0.0016\nIter: 18, Epoch: 36, Loss: 0.0014\nIter: 19, Epoch: 36, Loss: 0.0020\nIter: 20, Epoch: 36, Loss: 0.0013\nIter: 21, Epoch: 36, Loss: 0.0016\nIter: 22, Epoch: 36, Loss: 0.0013\nIter: 23, Epoch: 36, Loss: 0.0019\nIter: 24, Epoch: 36, Loss: 0.0012\nIter: 25, Epoch: 36, Loss: 0.0011\nIter: 26, Epoch: 36, Loss: 0.0014\nIter: 27, Epoch: 36, Loss: 0.0008\nIter: 28, Epoch: 36, Loss: 0.0016\nIter: 29, Epoch: 36, Loss: 0.0019\nIter: 30, Epoch: 36, Loss: 0.0010\n","truncated":false}} +%--- +%[output:24033b38] +% data: {"dataType":"text","outputData":{"text":">> Epoch 36 validation loss: 0.0028\n","truncated":false}} +%--- +%[output:60bb2365] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 37, Loss: 0.0044\nIter: 2, Epoch: 37, Loss: 0.0011\nIter: 3, Epoch: 37, Loss: 0.0015\nIter: 4, Epoch: 37, Loss: 0.0016\nIter: 5, Epoch: 37, Loss: 0.0012\nIter: 6, Epoch: 37, Loss: 0.0016\nIter: 7, Epoch: 37, Loss: 0.0010\nIter: 8, Epoch: 37, Loss: 0.0018\nIter: 9, Epoch: 37, Loss: 0.0019\nIter: 10, Epoch: 37, Loss: 0.0011\nIter: 11, Epoch: 37, Loss: 0.0011\nIter: 12, Epoch: 37, Loss: 0.0011\nIter: 13, Epoch: 37, Loss: 0.0014\nIter: 14, Epoch: 37, Loss: 0.0014\nIter: 15, Epoch: 37, Loss: 0.0009\nIter: 16, Epoch: 37, Loss: 0.0014\nIter: 17, Epoch: 37, Loss: 0.0017\nIter: 18, Epoch: 37, Loss: 0.0014\nIter: 19, Epoch: 37, Loss: 0.0020\nIter: 20, Epoch: 37, Loss: 0.0013\nIter: 21, Epoch: 37, Loss: 0.0017\nIter: 22, Epoch: 37, Loss: 0.0013\nIter: 23, Epoch: 37, Loss: 0.0019\nIter: 24, Epoch: 37, Loss: 0.0012\nIter: 25, Epoch: 37, Loss: 0.0012\nIter: 26, Epoch: 37, Loss: 0.0014\nIter: 27, Epoch: 37, Loss: 0.0008\nIter: 28, Epoch: 37, Loss: 0.0017\nIter: 29, Epoch: 37, Loss: 0.0020\nIter: 30, Epoch: 37, Loss: 0.0009\n","truncated":false}} +%--- +%[output:91359628] +% data: {"dataType":"text","outputData":{"text":">> Epoch 37 validation loss: 0.0027\n","truncated":false}} +%--- +%[output:2a067f77] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 38, Loss: 0.0043\nIter: 2, Epoch: 38, Loss: 0.0011\nIter: 3, Epoch: 38, Loss: 0.0015\nIter: 4, Epoch: 38, Loss: 0.0015\nIter: 5, Epoch: 38, Loss: 0.0012\nIter: 6, Epoch: 38, Loss: 0.0016\nIter: 7, Epoch: 38, Loss: 0.0010\nIter: 8, Epoch: 38, Loss: 0.0018\nIter: 9, Epoch: 38, Loss: 0.0020\nIter: 10, Epoch: 38, Loss: 0.0012\nIter: 11, Epoch: 38, Loss: 0.0010\nIter: 12, Epoch: 38, Loss: 0.0011\nIter: 13, Epoch: 38, Loss: 0.0014\nIter: 14, Epoch: 38, Loss: 0.0014\nIter: 15, Epoch: 38, Loss: 0.0009\nIter: 16, Epoch: 38, Loss: 0.0013\nIter: 17, Epoch: 38, Loss: 0.0017\nIter: 18, Epoch: 38, Loss: 0.0013\nIter: 19, Epoch: 38, Loss: 0.0020\nIter: 20, Epoch: 38, Loss: 0.0013\nIter: 21, Epoch: 38, Loss: 0.0017\nIter: 22, Epoch: 38, Loss: 0.0014\nIter: 23, Epoch: 38, Loss: 0.0018\nIter: 24, Epoch: 38, Loss: 0.0013\nIter: 25, Epoch: 38, Loss: 0.0012\nIter: 26, Epoch: 38, Loss: 0.0014\nIter: 27, Epoch: 38, Loss: 0.0008\nIter: 28, Epoch: 38, Loss: 0.0017\nIter: 29, Epoch: 38, Loss: 0.0020\nIter: 30, Epoch: 38, Loss: 0.0009\n","truncated":false}} +%--- +%[output:747f7326] +% data: {"dataType":"text","outputData":{"text":">> Epoch 38 validation loss: 0.0027\n","truncated":false}} +%--- +%[output:192ec571] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 39, Loss: 0.0042\nIter: 2, Epoch: 39, Loss: 0.0010\nIter: 3, Epoch: 39, Loss: 0.0015\nIter: 4, Epoch: 39, Loss: 0.0015\nIter: 5, Epoch: 39, Loss: 0.0012\nIter: 6, Epoch: 39, Loss: 0.0016\nIter: 7, Epoch: 39, Loss: 0.0010\nIter: 8, Epoch: 39, Loss: 0.0017\nIter: 9, Epoch: 39, Loss: 0.0020\nIter: 10, Epoch: 39, Loss: 0.0012\nIter: 11, Epoch: 39, Loss: 0.0010\nIter: 12, Epoch: 39, Loss: 0.0010\nIter: 13, Epoch: 39, Loss: 0.0014\nIter: 14, Epoch: 39, Loss: 0.0015\nIter: 15, Epoch: 39, Loss: 0.0008\nIter: 16, Epoch: 39, Loss: 0.0013\nIter: 17, Epoch: 39, Loss: 0.0018\nIter: 18, Epoch: 39, Loss: 0.0014\nIter: 19, Epoch: 39, Loss: 0.0019\nIter: 20, Epoch: 39, Loss: 0.0013\nIter: 21, Epoch: 39, Loss: 0.0018\nIter: 22, Epoch: 39, Loss: 0.0014\nIter: 23, Epoch: 39, Loss: 0.0018\nIter: 24, Epoch: 39, Loss: 0.0013\nIter: 25, Epoch: 39, Loss: 0.0013\nIter: 26, Epoch: 39, Loss: 0.0014\nIter: 27, Epoch: 39, Loss: 0.0008\nIter: 28, Epoch: 39, Loss: 0.0017\nIter: 29, Epoch: 39, Loss: 0.0021\nIter: 30, Epoch: 39, Loss: 0.0010\n","truncated":false}} +%--- +%[output:0ed9eee7] +% data: {"dataType":"text","outputData":{"text":">> Epoch 39 validation loss: 0.0025\n","truncated":false}} +%--- +%[output:1f9c02ba] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 40, Loss: 0.0040\nIter: 2, Epoch: 40, Loss: 0.0010\nIter: 3, Epoch: 40, Loss: 0.0016\nIter: 4, Epoch: 40, Loss: 0.0014\nIter: 5, Epoch: 40, Loss: 0.0011\nIter: 6, Epoch: 40, Loss: 0.0016\nIter: 7, Epoch: 40, Loss: 0.0011\nIter: 8, Epoch: 40, Loss: 0.0016\nIter: 9, Epoch: 40, Loss: 0.0019\nIter: 10, Epoch: 40, Loss: 0.0013\nIter: 11, Epoch: 40, Loss: 0.0010\nIter: 12, Epoch: 40, Loss: 0.0010\nIter: 13, Epoch: 40, Loss: 0.0013\nIter: 14, Epoch: 40, Loss: 0.0015\nIter: 15, Epoch: 40, Loss: 0.0008\nIter: 16, Epoch: 40, Loss: 0.0012\nIter: 17, Epoch: 40, Loss: 0.0018\nIter: 18, Epoch: 40, Loss: 0.0014\nIter: 19, Epoch: 40, Loss: 0.0017\nIter: 20, Epoch: 40, Loss: 0.0012\nIter: 21, Epoch: 40, Loss: 0.0019\nIter: 22, Epoch: 40, Loss: 0.0014\nIter: 23, Epoch: 40, Loss: 0.0016\nIter: 24, Epoch: 40, Loss: 0.0012\nIter: 25, Epoch: 40, Loss: 0.0013\nIter: 26, Epoch: 40, Loss: 0.0014\nIter: 27, Epoch: 40, Loss: 0.0007\nIter: 28, Epoch: 40, Loss: 0.0017\nIter: 29, Epoch: 40, Loss: 0.0022\nIter: 30, Epoch: 40, Loss: 0.0010\n","truncated":false}} +%--- +%[output:65ba71ea] +% data: {"dataType":"text","outputData":{"text":">> Epoch 40 validation loss: 0.0024\n","truncated":false}} +%--- +%[output:3470d1dc] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 41, Loss: 0.0037\nIter: 2, Epoch: 41, Loss: 0.0010\nIter: 3, Epoch: 41, Loss: 0.0016\nIter: 4, Epoch: 41, Loss: 0.0013\nIter: 5, Epoch: 41, Loss: 0.0010\nIter: 6, Epoch: 41, Loss: 0.0016\nIter: 7, Epoch: 41, Loss: 0.0011\nIter: 8, Epoch: 41, Loss: 0.0016\nIter: 9, Epoch: 41, Loss: 0.0019\nIter: 10, Epoch: 41, Loss: 0.0013\nIter: 11, Epoch: 41, Loss: 0.0010\nIter: 12, Epoch: 41, Loss: 0.0009\nIter: 13, Epoch: 41, Loss: 0.0012\nIter: 14, Epoch: 41, Loss: 0.0016\nIter: 15, Epoch: 41, Loss: 0.0008\nIter: 16, Epoch: 41, Loss: 0.0011\nIter: 17, Epoch: 41, Loss: 0.0017\nIter: 18, Epoch: 41, Loss: 0.0015\nIter: 19, Epoch: 41, Loss: 0.0016\nIter: 20, Epoch: 41, Loss: 0.0012\nIter: 21, Epoch: 41, Loss: 0.0019\nIter: 22, Epoch: 41, Loss: 0.0015\nIter: 23, Epoch: 41, Loss: 0.0015\nIter: 24, Epoch: 41, Loss: 0.0011\nIter: 25, Epoch: 41, Loss: 0.0014\nIter: 26, Epoch: 41, Loss: 0.0014\nIter: 27, Epoch: 41, Loss: 0.0007\nIter: 28, Epoch: 41, Loss: 0.0016\nIter: 29, Epoch: 41, Loss: 0.0022\nIter: 30, Epoch: 41, Loss: 0.0010\n","truncated":false}} +%--- +%[output:5092ec9d] +% data: {"dataType":"text","outputData":{"text":">> Epoch 41 validation loss: 0.0022\n","truncated":false}} +%--- +%[output:880f85d0] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 42, Loss: 0.0034\nIter: 2, Epoch: 42, Loss: 0.0009\nIter: 3, Epoch: 42, Loss: 0.0017\nIter: 4, Epoch: 42, Loss: 0.0014\nIter: 5, Epoch: 42, Loss: 0.0009\nIter: 6, Epoch: 42, Loss: 0.0016\nIter: 7, Epoch: 42, Loss: 0.0012\nIter: 8, Epoch: 42, Loss: 0.0015\nIter: 9, Epoch: 42, Loss: 0.0019\nIter: 10, Epoch: 42, Loss: 0.0013\nIter: 11, Epoch: 42, Loss: 0.0010\nIter: 12, Epoch: 42, Loss: 0.0009\nIter: 13, Epoch: 42, Loss: 0.0011\nIter: 14, Epoch: 42, Loss: 0.0016\nIter: 15, Epoch: 42, Loss: 0.0008\nIter: 16, Epoch: 42, Loss: 0.0010\nIter: 17, Epoch: 42, Loss: 0.0016\nIter: 18, Epoch: 42, Loss: 0.0015\nIter: 19, Epoch: 42, Loss: 0.0016\nIter: 20, Epoch: 42, Loss: 0.0011\nIter: 21, Epoch: 42, Loss: 0.0019\nIter: 22, Epoch: 42, Loss: 0.0016\nIter: 23, Epoch: 42, Loss: 0.0014\nIter: 24, Epoch: 42, Loss: 0.0011\nIter: 25, Epoch: 42, Loss: 0.0013\nIter: 26, Epoch: 42, Loss: 0.0014\nIter: 27, Epoch: 42, Loss: 0.0008\nIter: 28, Epoch: 42, Loss: 0.0015\nIter: 29, Epoch: 42, Loss: 0.0021\nIter: 30, Epoch: 42, Loss: 0.0011\n","truncated":false}} +%--- +%[output:093122bd] +% data: {"dataType":"text","outputData":{"text":">> Epoch 42 validation loss: 0.0021\n","truncated":false}} +%--- +%[output:76a6da47] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 43, Loss: 0.0032\nIter: 2, Epoch: 43, Loss: 0.0009\nIter: 3, Epoch: 43, Loss: 0.0017\nIter: 4, Epoch: 43, Loss: 0.0014\nIter: 5, Epoch: 43, Loss: 0.0009\nIter: 6, Epoch: 43, Loss: 0.0015\nIter: 7, Epoch: 43, Loss: 0.0012\nIter: 8, Epoch: 43, Loss: 0.0016\nIter: 9, Epoch: 43, Loss: 0.0018\nIter: 10, Epoch: 43, Loss: 0.0013\nIter: 11, Epoch: 43, Loss: 0.0010\nIter: 12, Epoch: 43, Loss: 0.0009\nIter: 13, Epoch: 43, Loss: 0.0010\nIter: 14, Epoch: 43, Loss: 0.0015\nIter: 15, Epoch: 43, Loss: 0.0008\nIter: 16, Epoch: 43, Loss: 0.0010\nIter: 17, Epoch: 43, Loss: 0.0015\nIter: 18, Epoch: 43, Loss: 0.0015\nIter: 19, Epoch: 43, Loss: 0.0016\nIter: 20, Epoch: 43, Loss: 0.0011\nIter: 21, Epoch: 43, Loss: 0.0018\nIter: 22, Epoch: 43, Loss: 0.0015\nIter: 23, Epoch: 43, Loss: 0.0015\nIter: 24, Epoch: 43, Loss: 0.0010\nIter: 25, Epoch: 43, Loss: 0.0013\nIter: 26, Epoch: 43, Loss: 0.0014\nIter: 27, Epoch: 43, Loss: 0.0008\nIter: 28, Epoch: 43, Loss: 0.0013\nIter: 29, Epoch: 43, Loss: 0.0020\nIter: 30, Epoch: 43, Loss: 0.0011\n","truncated":false}} +%--- +%[output:91664549] +% data: {"dataType":"text","outputData":{"text":">> Epoch 43 validation loss: 0.0021\n","truncated":false}} +%--- +%[output:48076b9d] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 44, Loss: 0.0030\nIter: 2, Epoch: 44, Loss: 0.0008\nIter: 3, Epoch: 44, Loss: 0.0016\nIter: 4, Epoch: 44, Loss: 0.0014\nIter: 5, Epoch: 44, Loss: 0.0008\nIter: 6, Epoch: 44, Loss: 0.0015\nIter: 7, Epoch: 44, Loss: 0.0012\nIter: 8, Epoch: 44, Loss: 0.0016\nIter: 9, Epoch: 44, Loss: 0.0017\nIter: 10, Epoch: 44, Loss: 0.0012\nIter: 11, Epoch: 44, Loss: 0.0010\nIter: 12, Epoch: 44, Loss: 0.0009\nIter: 13, Epoch: 44, Loss: 0.0010\nIter: 14, Epoch: 44, Loss: 0.0015\nIter: 15, Epoch: 44, Loss: 0.0009\nIter: 16, Epoch: 44, Loss: 0.0010\nIter: 17, Epoch: 44, Loss: 0.0014\nIter: 18, Epoch: 44, Loss: 0.0015\nIter: 19, Epoch: 44, Loss: 0.0017\nIter: 20, Epoch: 44, Loss: 0.0010\nIter: 21, Epoch: 44, Loss: 0.0018\nIter: 22, Epoch: 44, Loss: 0.0015\nIter: 23, Epoch: 44, Loss: 0.0014\nIter: 24, Epoch: 44, Loss: 0.0009\nIter: 25, Epoch: 44, Loss: 0.0013\nIter: 26, Epoch: 44, Loss: 0.0014\nIter: 27, Epoch: 44, Loss: 0.0009\nIter: 28, Epoch: 44, Loss: 0.0013\nIter: 29, Epoch: 44, Loss: 0.0020\nIter: 30, Epoch: 44, Loss: 0.0011\n","truncated":false}} +%--- +%[output:4e7d3724] +% data: {"dataType":"text","outputData":{"text":">> Epoch 44 validation loss: 0.0021\n","truncated":false}} +%--- +%[output:9208fbb0] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 45, Loss: 0.0030\nIter: 2, Epoch: 45, Loss: 0.0008\nIter: 3, Epoch: 45, Loss: 0.0016\nIter: 4, Epoch: 45, Loss: 0.0015\nIter: 5, Epoch: 45, Loss: 0.0008\nIter: 6, Epoch: 45, Loss: 0.0014\nIter: 7, Epoch: 45, Loss: 0.0012\nIter: 8, Epoch: 45, Loss: 0.0017\nIter: 9, Epoch: 45, Loss: 0.0017\nIter: 10, Epoch: 45, Loss: 0.0012\nIter: 11, Epoch: 45, Loss: 0.0010\nIter: 12, Epoch: 45, Loss: 0.0009\nIter: 13, Epoch: 45, Loss: 0.0009\nIter: 14, Epoch: 45, Loss: 0.0014\nIter: 15, Epoch: 45, Loss: 0.0009\nIter: 16, Epoch: 45, Loss: 0.0009\nIter: 17, Epoch: 45, Loss: 0.0013\nIter: 18, Epoch: 45, Loss: 0.0014\nIter: 19, Epoch: 45, Loss: 0.0017\nIter: 20, Epoch: 45, Loss: 0.0010\nIter: 21, Epoch: 45, Loss: 0.0017\nIter: 22, Epoch: 45, Loss: 0.0015\nIter: 23, Epoch: 45, Loss: 0.0015\nIter: 24, Epoch: 45, Loss: 0.0009\nIter: 25, Epoch: 45, Loss: 0.0012\nIter: 26, Epoch: 45, Loss: 0.0014\nIter: 27, Epoch: 45, Loss: 0.0009\nIter: 28, Epoch: 45, Loss: 0.0012\nIter: 29, Epoch: 45, Loss: 0.0018\nIter: 30, Epoch: 45, Loss: 0.0011\n","truncated":false}} +%--- +%[output:054e6030] +% data: {"dataType":"text","outputData":{"text":">> Epoch 45 validation loss: 0.0021\n","truncated":false}} +%--- +%[output:2c554e52] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 46, Loss: 0.0030\nIter: 2, Epoch: 46, Loss: 0.0008\nIter: 3, Epoch: 46, Loss: 0.0015\nIter: 4, Epoch: 46, Loss: 0.0016\nIter: 5, Epoch: 46, Loss: 0.0008\nIter: 6, Epoch: 46, Loss: 0.0013\nIter: 7, Epoch: 46, Loss: 0.0012\nIter: 8, Epoch: 46, Loss: 0.0017\nIter: 9, Epoch: 46, Loss: 0.0016\nIter: 10, Epoch: 46, Loss: 0.0011\nIter: 11, Epoch: 46, Loss: 0.0010\nIter: 12, Epoch: 46, Loss: 0.0009\nIter: 13, Epoch: 46, Loss: 0.0009\nIter: 14, Epoch: 46, Loss: 0.0013\nIter: 15, Epoch: 46, Loss: 0.0009\nIter: 16, Epoch: 46, Loss: 0.0010\nIter: 17, Epoch: 46, Loss: 0.0013\nIter: 18, Epoch: 46, Loss: 0.0014\nIter: 19, Epoch: 46, Loss: 0.0019\nIter: 20, Epoch: 46, Loss: 0.0010\nIter: 21, Epoch: 46, Loss: 0.0017\nIter: 22, Epoch: 46, Loss: 0.0015\nIter: 23, Epoch: 46, Loss: 0.0015\nIter: 24, Epoch: 46, Loss: 0.0009\nIter: 25, Epoch: 46, Loss: 0.0012\nIter: 26, Epoch: 46, Loss: 0.0014\nIter: 27, Epoch: 46, Loss: 0.0009\nIter: 28, Epoch: 46, Loss: 0.0012\nIter: 29, Epoch: 46, Loss: 0.0018\nIter: 30, Epoch: 46, Loss: 0.0011\n","truncated":false}} +%--- +%[output:11b53b2a] +% data: {"dataType":"text","outputData":{"text":">> Epoch 46 validation loss: 0.0021\n","truncated":false}} +%--- +%[output:548afa29] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 47, Loss: 0.0030\nIter: 2, Epoch: 47, Loss: 0.0007\nIter: 3, Epoch: 47, Loss: 0.0014\nIter: 4, Epoch: 47, Loss: 0.0016\nIter: 5, Epoch: 47, Loss: 0.0008\nIter: 6, Epoch: 47, Loss: 0.0013\nIter: 7, Epoch: 47, Loss: 0.0012\nIter: 8, Epoch: 47, Loss: 0.0018\nIter: 9, Epoch: 47, Loss: 0.0016\nIter: 10, Epoch: 47, Loss: 0.0011\nIter: 11, Epoch: 47, Loss: 0.0010\nIter: 12, Epoch: 47, Loss: 0.0009\nIter: 13, Epoch: 47, Loss: 0.0009\nIter: 14, Epoch: 47, Loss: 0.0013\nIter: 15, Epoch: 47, Loss: 0.0010\nIter: 16, Epoch: 47, Loss: 0.0010\nIter: 17, Epoch: 47, Loss: 0.0013\nIter: 18, Epoch: 47, Loss: 0.0014\nIter: 19, Epoch: 47, Loss: 0.0020\nIter: 20, Epoch: 47, Loss: 0.0010\nIter: 21, Epoch: 47, Loss: 0.0016\nIter: 22, Epoch: 47, Loss: 0.0014\nIter: 23, Epoch: 47, Loss: 0.0016\nIter: 24, Epoch: 47, Loss: 0.0010\nIter: 25, Epoch: 47, Loss: 0.0011\nIter: 26, Epoch: 47, Loss: 0.0014\nIter: 27, Epoch: 47, Loss: 0.0010\nIter: 28, Epoch: 47, Loss: 0.0012\nIter: 29, Epoch: 47, Loss: 0.0017\nIter: 30, Epoch: 47, Loss: 0.0010\n","truncated":false}} +%--- +%[output:65ec5c79] +% data: {"dataType":"text","outputData":{"text":">> Epoch 47 validation loss: 0.0021\n","truncated":false}} +%--- +%[output:2945cbb2] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 48, Loss: 0.0031\nIter: 2, Epoch: 48, Loss: 0.0008\nIter: 3, Epoch: 48, Loss: 0.0014\nIter: 4, Epoch: 48, Loss: 0.0016\nIter: 5, Epoch: 48, Loss: 0.0008\nIter: 6, Epoch: 48, Loss: 0.0012\nIter: 7, Epoch: 48, Loss: 0.0011\nIter: 8, Epoch: 48, Loss: 0.0019\nIter: 9, Epoch: 48, Loss: 0.0015\nIter: 10, Epoch: 48, Loss: 0.0010\nIter: 11, Epoch: 48, Loss: 0.0010\nIter: 12, Epoch: 48, Loss: 0.0009\nIter: 13, Epoch: 48, Loss: 0.0010\nIter: 14, Epoch: 48, Loss: 0.0012\nIter: 15, Epoch: 48, Loss: 0.0009\nIter: 16, Epoch: 48, Loss: 0.0010\nIter: 17, Epoch: 48, Loss: 0.0013\nIter: 18, Epoch: 48, Loss: 0.0013\nIter: 19, Epoch: 48, Loss: 0.0021\nIter: 20, Epoch: 48, Loss: 0.0010\nIter: 21, Epoch: 48, Loss: 0.0016\nIter: 22, Epoch: 48, Loss: 0.0014\nIter: 23, Epoch: 48, Loss: 0.0016\nIter: 24, Epoch: 48, Loss: 0.0010\nIter: 25, Epoch: 48, Loss: 0.0011\nIter: 26, Epoch: 48, Loss: 0.0013\nIter: 27, Epoch: 48, Loss: 0.0010\nIter: 28, Epoch: 48, Loss: 0.0012\nIter: 29, Epoch: 48, Loss: 0.0016\nIter: 30, Epoch: 48, Loss: 0.0010\n","truncated":false}} +%--- +%[output:01f6d859] +% data: {"dataType":"text","outputData":{"text":">> Epoch 48 validation loss: 0.0022\n","truncated":false}} +%--- +%[output:694de540] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 49, Loss: 0.0032\nIter: 2, Epoch: 49, Loss: 0.0008\nIter: 3, Epoch: 49, Loss: 0.0013\nIter: 4, Epoch: 49, Loss: 0.0016\nIter: 5, Epoch: 49, Loss: 0.0009\nIter: 6, Epoch: 49, Loss: 0.0012\nIter: 7, Epoch: 49, Loss: 0.0011\nIter: 8, Epoch: 49, Loss: 0.0019\nIter: 9, Epoch: 49, Loss: 0.0015\nIter: 10, Epoch: 49, Loss: 0.0010\nIter: 11, Epoch: 49, Loss: 0.0009\nIter: 12, Epoch: 49, Loss: 0.0009\nIter: 13, Epoch: 49, Loss: 0.0010\nIter: 14, Epoch: 49, Loss: 0.0011\nIter: 15, Epoch: 49, Loss: 0.0009\nIter: 16, Epoch: 49, Loss: 0.0011\nIter: 17, Epoch: 49, Loss: 0.0014\nIter: 18, Epoch: 49, Loss: 0.0013\nIter: 19, Epoch: 49, Loss: 0.0021\nIter: 20, Epoch: 49, Loss: 0.0010\nIter: 21, Epoch: 49, Loss: 0.0015\nIter: 22, Epoch: 49, Loss: 0.0014\nIter: 23, Epoch: 49, Loss: 0.0017\nIter: 24, Epoch: 49, Loss: 0.0011\nIter: 25, Epoch: 49, Loss: 0.0010\nIter: 26, Epoch: 49, Loss: 0.0013\nIter: 27, Epoch: 49, Loss: 0.0010\nIter: 28, Epoch: 49, Loss: 0.0012\nIter: 29, Epoch: 49, Loss: 0.0016\nIter: 30, Epoch: 49, Loss: 0.0010\n","truncated":false}} +%--- +%[output:0e7dd202] +% data: {"dataType":"text","outputData":{"text":">> Epoch 49 validation loss: 0.0023\n","truncated":false}} +%--- +%[output:124473a9] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 50, Loss: 0.0033\nIter: 2, Epoch: 50, Loss: 0.0009\nIter: 3, Epoch: 50, Loss: 0.0013\nIter: 4, Epoch: 50, Loss: 0.0015\nIter: 5, Epoch: 50, Loss: 0.0010\nIter: 6, Epoch: 50, Loss: 0.0012\nIter: 7, Epoch: 50, Loss: 0.0010\nIter: 8, Epoch: 50, Loss: 0.0019\nIter: 9, Epoch: 50, Loss: 0.0015\nIter: 10, Epoch: 50, Loss: 0.0011\nIter: 11, Epoch: 50, Loss: 0.0009\nIter: 12, Epoch: 50, Loss: 0.0009\nIter: 13, Epoch: 50, Loss: 0.0012\nIter: 14, Epoch: 50, Loss: 0.0011\nIter: 15, Epoch: 50, Loss: 0.0009\nIter: 16, Epoch: 50, Loss: 0.0011\nIter: 17, Epoch: 50, Loss: 0.0015\nIter: 18, Epoch: 50, Loss: 0.0013\nIter: 19, Epoch: 50, Loss: 0.0020\nIter: 20, Epoch: 50, Loss: 0.0010\nIter: 21, Epoch: 50, Loss: 0.0016\nIter: 22, Epoch: 50, Loss: 0.0013\nIter: 23, Epoch: 50, Loss: 0.0016\nIter: 24, Epoch: 50, Loss: 0.0012\nIter: 25, Epoch: 50, Loss: 0.0011\nIter: 26, Epoch: 50, Loss: 0.0012\nIter: 27, Epoch: 50, Loss: 0.0009\nIter: 28, Epoch: 50, Loss: 0.0014\nIter: 29, Epoch: 50, Loss: 0.0017\nIter: 30, Epoch: 50, Loss: 0.0009\n","truncated":false}} +%--- +%[output:585e4155] +% data: {"dataType":"text","outputData":{"text":">> Epoch 50 validation loss: 0.0024\n","truncated":false}} +%--- +%[output:55ee496b] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 51, Loss: 0.0035\nIter: 2, Epoch: 51, Loss: 0.0010\nIter: 3, Epoch: 51, Loss: 0.0013\nIter: 4, Epoch: 51, Loss: 0.0014\nIter: 5, Epoch: 51, Loss: 0.0010\nIter: 6, Epoch: 51, Loss: 0.0012\nIter: 7, Epoch: 51, Loss: 0.0011\nIter: 8, Epoch: 51, Loss: 0.0018\nIter: 9, Epoch: 51, Loss: 0.0016\nIter: 10, Epoch: 51, Loss: 0.0012\nIter: 11, Epoch: 51, Loss: 0.0009\nIter: 12, Epoch: 51, Loss: 0.0009\nIter: 13, Epoch: 51, Loss: 0.0013\nIter: 14, Epoch: 51, Loss: 0.0012\nIter: 15, Epoch: 51, Loss: 0.0008\nIter: 16, Epoch: 51, Loss: 0.0011\nIter: 17, Epoch: 51, Loss: 0.0016\nIter: 18, Epoch: 51, Loss: 0.0013\nIter: 19, Epoch: 51, Loss: 0.0018\nIter: 20, Epoch: 51, Loss: 0.0010\nIter: 21, Epoch: 51, Loss: 0.0016\nIter: 22, Epoch: 51, Loss: 0.0013\nIter: 23, Epoch: 51, Loss: 0.0015\nIter: 24, Epoch: 51, Loss: 0.0012\nIter: 25, Epoch: 51, Loss: 0.0012\nIter: 26, Epoch: 51, Loss: 0.0012\nIter: 27, Epoch: 51, Loss: 0.0008\nIter: 28, Epoch: 51, Loss: 0.0014\nIter: 29, Epoch: 51, Loss: 0.0018\nIter: 30, Epoch: 51, Loss: 0.0009\n","truncated":false}} +%--- +%[output:6e325963] +% data: {"dataType":"text","outputData":{"text":">> Epoch 51 validation loss: 0.0023\n","truncated":false}} +%--- +%[output:1586ad98] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 52, Loss: 0.0034\nIter: 2, Epoch: 52, Loss: 0.0009\nIter: 3, Epoch: 52, Loss: 0.0014\nIter: 4, Epoch: 52, Loss: 0.0013\nIter: 5, Epoch: 52, Loss: 0.0010\nIter: 6, Epoch: 52, Loss: 0.0013\nIter: 7, Epoch: 52, Loss: 0.0011\nIter: 8, Epoch: 52, Loss: 0.0017\nIter: 9, Epoch: 52, Loss: 0.0016\nIter: 10, Epoch: 52, Loss: 0.0012\nIter: 11, Epoch: 52, Loss: 0.0009\nIter: 12, Epoch: 52, Loss: 0.0008\nIter: 13, Epoch: 52, Loss: 0.0012\nIter: 14, Epoch: 52, Loss: 0.0013\nIter: 15, Epoch: 52, Loss: 0.0008\nIter: 16, Epoch: 52, Loss: 0.0010\nIter: 17, Epoch: 52, Loss: 0.0016\nIter: 18, Epoch: 52, Loss: 0.0014\nIter: 19, Epoch: 52, Loss: 0.0017\nIter: 20, Epoch: 52, Loss: 0.0010\nIter: 21, Epoch: 52, Loss: 0.0016\nIter: 22, Epoch: 52, Loss: 0.0014\nIter: 23, Epoch: 52, Loss: 0.0015\nIter: 24, Epoch: 52, Loss: 0.0011\nIter: 25, Epoch: 52, Loss: 0.0012\nIter: 26, Epoch: 52, Loss: 0.0012\nIter: 27, Epoch: 52, Loss: 0.0008\nIter: 28, Epoch: 52, Loss: 0.0014\nIter: 29, Epoch: 52, Loss: 0.0018\nIter: 30, Epoch: 52, Loss: 0.0009\n","truncated":false}} +%--- +%[output:128e68b5] +% data: {"dataType":"text","outputData":{"text":">> Epoch 52 validation loss: 0.0021\n","truncated":false}} +%--- +%[output:7ccbbb25] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 53, Loss: 0.0031\nIter: 2, Epoch: 53, Loss: 0.0009\nIter: 3, Epoch: 53, Loss: 0.0014\nIter: 4, Epoch: 53, Loss: 0.0013\nIter: 5, Epoch: 53, Loss: 0.0009\nIter: 6, Epoch: 53, Loss: 0.0012\nIter: 7, Epoch: 53, Loss: 0.0011\nIter: 8, Epoch: 53, Loss: 0.0017\nIter: 9, Epoch: 53, Loss: 0.0015\nIter: 10, Epoch: 53, Loss: 0.0012\nIter: 11, Epoch: 53, Loss: 0.0009\nIter: 12, Epoch: 53, Loss: 0.0008\nIter: 13, Epoch: 53, Loss: 0.0011\nIter: 14, Epoch: 53, Loss: 0.0013\nIter: 15, Epoch: 53, Loss: 0.0008\nIter: 16, Epoch: 53, Loss: 0.0010\nIter: 17, Epoch: 53, Loss: 0.0015\nIter: 18, Epoch: 53, Loss: 0.0014\nIter: 19, Epoch: 53, Loss: 0.0016\nIter: 20, Epoch: 53, Loss: 0.0009\nIter: 21, Epoch: 53, Loss: 0.0016\nIter: 22, Epoch: 53, Loss: 0.0014\nIter: 23, Epoch: 53, Loss: 0.0014\nIter: 24, Epoch: 53, Loss: 0.0010\nIter: 25, Epoch: 53, Loss: 0.0012\nIter: 26, Epoch: 53, Loss: 0.0012\nIter: 27, Epoch: 53, Loss: 0.0008\nIter: 28, Epoch: 53, Loss: 0.0013\nIter: 29, Epoch: 53, Loss: 0.0018\nIter: 30, Epoch: 53, Loss: 0.0009\n","truncated":false}} +%--- +%[output:0b1648fa] +% data: {"dataType":"text","outputData":{"text":">> Epoch 53 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:802a9cda] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 54, Loss: 0.0030\nIter: 2, Epoch: 54, Loss: 0.0009\nIter: 3, Epoch: 54, Loss: 0.0014\nIter: 4, Epoch: 54, Loss: 0.0013\nIter: 5, Epoch: 54, Loss: 0.0008\nIter: 6, Epoch: 54, Loss: 0.0013\nIter: 7, Epoch: 54, Loss: 0.0011\nIter: 8, Epoch: 54, Loss: 0.0016\nIter: 9, Epoch: 54, Loss: 0.0015\nIter: 10, Epoch: 54, Loss: 0.0012\nIter: 11, Epoch: 54, Loss: 0.0009\nIter: 12, Epoch: 54, Loss: 0.0008\nIter: 13, Epoch: 54, Loss: 0.0011\nIter: 14, Epoch: 54, Loss: 0.0013\nIter: 15, Epoch: 54, Loss: 0.0008\nIter: 16, Epoch: 54, Loss: 0.0009\nIter: 17, Epoch: 54, Loss: 0.0014\nIter: 18, Epoch: 54, Loss: 0.0014\nIter: 19, Epoch: 54, Loss: 0.0016\nIter: 20, Epoch: 54, Loss: 0.0009\nIter: 21, Epoch: 54, Loss: 0.0016\nIter: 22, Epoch: 54, Loss: 0.0014\nIter: 23, Epoch: 54, Loss: 0.0014\nIter: 24, Epoch: 54, Loss: 0.0010\nIter: 25, Epoch: 54, Loss: 0.0012\nIter: 26, Epoch: 54, Loss: 0.0012\nIter: 27, Epoch: 54, Loss: 0.0008\nIter: 28, Epoch: 54, Loss: 0.0012\nIter: 29, Epoch: 54, Loss: 0.0018\nIter: 30, Epoch: 54, Loss: 0.0009\n","truncated":false}} +%--- +%[output:1b00417e] +% data: {"dataType":"text","outputData":{"text":">> Epoch 54 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:2e9f6233] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 55, Loss: 0.0030\nIter: 2, Epoch: 55, Loss: 0.0009\nIter: 3, Epoch: 55, Loss: 0.0014\nIter: 4, Epoch: 55, Loss: 0.0013\nIter: 5, Epoch: 55, Loss: 0.0008\nIter: 6, Epoch: 55, Loss: 0.0012\nIter: 7, Epoch: 55, Loss: 0.0011\nIter: 8, Epoch: 55, Loss: 0.0016\nIter: 9, Epoch: 55, Loss: 0.0015\nIter: 10, Epoch: 55, Loss: 0.0011\nIter: 11, Epoch: 55, Loss: 0.0009\nIter: 12, Epoch: 55, Loss: 0.0008\nIter: 13, Epoch: 55, Loss: 0.0010\nIter: 14, Epoch: 55, Loss: 0.0012\nIter: 15, Epoch: 55, Loss: 0.0008\nIter: 16, Epoch: 55, Loss: 0.0009\nIter: 17, Epoch: 55, Loss: 0.0014\nIter: 18, Epoch: 55, Loss: 0.0013\nIter: 19, Epoch: 55, Loss: 0.0016\nIter: 20, Epoch: 55, Loss: 0.0009\nIter: 21, Epoch: 55, Loss: 0.0015\nIter: 22, Epoch: 55, Loss: 0.0014\nIter: 23, Epoch: 55, Loss: 0.0015\nIter: 24, Epoch: 55, Loss: 0.0010\nIter: 25, Epoch: 55, Loss: 0.0011\nIter: 26, Epoch: 55, Loss: 0.0012\nIter: 27, Epoch: 55, Loss: 0.0008\nIter: 28, Epoch: 55, Loss: 0.0012\nIter: 29, Epoch: 55, Loss: 0.0017\nIter: 30, Epoch: 55, Loss: 0.0009\n","truncated":false}} +%--- +%[output:3dd8f80c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 55 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:683d307a] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 56, Loss: 0.0030\nIter: 2, Epoch: 56, Loss: 0.0008\nIter: 3, Epoch: 56, Loss: 0.0013\nIter: 4, Epoch: 56, Loss: 0.0012\nIter: 5, Epoch: 56, Loss: 0.0008\nIter: 6, Epoch: 56, Loss: 0.0012\nIter: 7, Epoch: 56, Loss: 0.0011\nIter: 8, Epoch: 56, Loss: 0.0016\nIter: 9, Epoch: 56, Loss: 0.0015\nIter: 10, Epoch: 56, Loss: 0.0011\nIter: 11, Epoch: 56, Loss: 0.0008\nIter: 12, Epoch: 56, Loss: 0.0007\nIter: 13, Epoch: 56, Loss: 0.0010\nIter: 14, Epoch: 56, Loss: 0.0012\nIter: 15, Epoch: 56, Loss: 0.0008\nIter: 16, Epoch: 56, Loss: 0.0009\nIter: 17, Epoch: 56, Loss: 0.0014\nIter: 18, Epoch: 56, Loss: 0.0013\nIter: 19, Epoch: 56, Loss: 0.0017\nIter: 20, Epoch: 56, Loss: 0.0009\nIter: 21, Epoch: 56, Loss: 0.0016\nIter: 22, Epoch: 56, Loss: 0.0014\nIter: 23, Epoch: 56, Loss: 0.0014\nIter: 24, Epoch: 56, Loss: 0.0010\nIter: 25, Epoch: 56, Loss: 0.0011\nIter: 26, Epoch: 56, Loss: 0.0012\nIter: 27, Epoch: 56, Loss: 0.0008\nIter: 28, Epoch: 56, Loss: 0.0012\nIter: 29, Epoch: 56, Loss: 0.0017\nIter: 30, Epoch: 56, Loss: 0.0009\n","truncated":false}} +%--- +%[output:56684393] +% data: {"dataType":"text","outputData":{"text":">> Epoch 56 validation loss: 0.0021\n","truncated":false}} +%--- +%[output:32e82a2d] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 57, Loss: 0.0030\nIter: 2, Epoch: 57, Loss: 0.0008\nIter: 3, Epoch: 57, Loss: 0.0013\nIter: 4, Epoch: 57, Loss: 0.0012\nIter: 5, Epoch: 57, Loss: 0.0008\nIter: 6, Epoch: 57, Loss: 0.0012\nIter: 7, Epoch: 57, Loss: 0.0011\nIter: 8, Epoch: 57, Loss: 0.0016\nIter: 9, Epoch: 57, Loss: 0.0015\nIter: 10, Epoch: 57, Loss: 0.0011\nIter: 11, Epoch: 57, Loss: 0.0009\nIter: 12, Epoch: 57, Loss: 0.0007\nIter: 13, Epoch: 57, Loss: 0.0010\nIter: 14, Epoch: 57, Loss: 0.0012\nIter: 15, Epoch: 57, Loss: 0.0008\nIter: 16, Epoch: 57, Loss: 0.0009\nIter: 17, Epoch: 57, Loss: 0.0014\nIter: 18, Epoch: 57, Loss: 0.0013\nIter: 19, Epoch: 57, Loss: 0.0016\nIter: 20, Epoch: 57, Loss: 0.0008\nIter: 21, Epoch: 57, Loss: 0.0015\nIter: 22, Epoch: 57, Loss: 0.0014\nIter: 23, Epoch: 57, Loss: 0.0015\nIter: 24, Epoch: 57, Loss: 0.0010\nIter: 25, Epoch: 57, Loss: 0.0011\nIter: 26, Epoch: 57, Loss: 0.0012\nIter: 27, Epoch: 57, Loss: 0.0008\nIter: 28, Epoch: 57, Loss: 0.0012\nIter: 29, Epoch: 57, Loss: 0.0017\nIter: 30, Epoch: 57, Loss: 0.0009\n","truncated":false}} +%--- +%[output:41097202] +% data: {"dataType":"text","outputData":{"text":">> Epoch 57 validation loss: 0.0021\n","truncated":false}} +%--- +%[output:62df5769] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 58, Loss: 0.0030\nIter: 2, Epoch: 58, Loss: 0.0008\nIter: 3, Epoch: 58, Loss: 0.0014\nIter: 4, Epoch: 58, Loss: 0.0013\nIter: 5, Epoch: 58, Loss: 0.0008\nIter: 6, Epoch: 58, Loss: 0.0012\nIter: 7, Epoch: 58, Loss: 0.0011\nIter: 8, Epoch: 58, Loss: 0.0016\nIter: 9, Epoch: 58, Loss: 0.0015\nIter: 10, Epoch: 58, Loss: 0.0011\nIter: 11, Epoch: 58, Loss: 0.0008\nIter: 12, Epoch: 58, Loss: 0.0007\nIter: 13, Epoch: 58, Loss: 0.0009\nIter: 14, Epoch: 58, Loss: 0.0011\nIter: 15, Epoch: 58, Loss: 0.0008\nIter: 16, Epoch: 58, Loss: 0.0009\nIter: 17, Epoch: 58, Loss: 0.0013\nIter: 18, Epoch: 58, Loss: 0.0013\nIter: 19, Epoch: 58, Loss: 0.0017\nIter: 20, Epoch: 58, Loss: 0.0008\nIter: 21, Epoch: 58, Loss: 0.0015\nIter: 22, Epoch: 58, Loss: 0.0014\nIter: 23, Epoch: 58, Loss: 0.0015\nIter: 24, Epoch: 58, Loss: 0.0009\nIter: 25, Epoch: 58, Loss: 0.0010\nIter: 26, Epoch: 58, Loss: 0.0012\nIter: 27, Epoch: 58, Loss: 0.0008\nIter: 28, Epoch: 58, Loss: 0.0011\nIter: 29, Epoch: 58, Loss: 0.0016\nIter: 30, Epoch: 58, Loss: 0.0009\n","truncated":false}} +%--- +%[output:612b4908] +% data: {"dataType":"text","outputData":{"text":">> Epoch 58 validation loss: 0.0021\n","truncated":false}} +%--- +%[output:3db1fd45] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 59, Loss: 0.0030\nIter: 2, Epoch: 59, Loss: 0.0008\nIter: 3, Epoch: 59, Loss: 0.0013\nIter: 4, Epoch: 59, Loss: 0.0012\nIter: 5, Epoch: 59, Loss: 0.0008\nIter: 6, Epoch: 59, Loss: 0.0012\nIter: 7, Epoch: 59, Loss: 0.0011\nIter: 8, Epoch: 59, Loss: 0.0016\nIter: 9, Epoch: 59, Loss: 0.0015\nIter: 10, Epoch: 59, Loss: 0.0011\nIter: 11, Epoch: 59, Loss: 0.0008\nIter: 12, Epoch: 59, Loss: 0.0007\nIter: 13, Epoch: 59, Loss: 0.0009\nIter: 14, Epoch: 59, Loss: 0.0011\nIter: 15, Epoch: 59, Loss: 0.0008\nIter: 16, Epoch: 59, Loss: 0.0009\nIter: 17, Epoch: 59, Loss: 0.0013\nIter: 18, Epoch: 59, Loss: 0.0012\nIter: 19, Epoch: 59, Loss: 0.0017\nIter: 20, Epoch: 59, Loss: 0.0008\nIter: 21, Epoch: 59, Loss: 0.0015\nIter: 22, Epoch: 59, Loss: 0.0013\nIter: 23, Epoch: 59, Loss: 0.0015\nIter: 24, Epoch: 59, Loss: 0.0010\nIter: 25, Epoch: 59, Loss: 0.0010\nIter: 26, Epoch: 59, Loss: 0.0011\nIter: 27, Epoch: 59, Loss: 0.0008\nIter: 28, Epoch: 59, Loss: 0.0012\nIter: 29, Epoch: 59, Loss: 0.0016\nIter: 30, Epoch: 59, Loss: 0.0009\n","truncated":false}} +%--- +%[output:9ab0a38a] +% data: {"dataType":"text","outputData":{"text":">> Epoch 59 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:74c7b215] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 60, Loss: 0.0030\nIter: 2, Epoch: 60, Loss: 0.0008\nIter: 3, Epoch: 60, Loss: 0.0013\nIter: 4, Epoch: 60, Loss: 0.0012\nIter: 5, Epoch: 60, Loss: 0.0008\nIter: 6, Epoch: 60, Loss: 0.0011\nIter: 7, Epoch: 60, Loss: 0.0010\nIter: 8, Epoch: 60, Loss: 0.0016\nIter: 9, Epoch: 60, Loss: 0.0014\nIter: 10, Epoch: 60, Loss: 0.0011\nIter: 11, Epoch: 60, Loss: 0.0008\nIter: 12, Epoch: 60, Loss: 0.0007\nIter: 13, Epoch: 60, Loss: 0.0009\nIter: 14, Epoch: 60, Loss: 0.0011\nIter: 15, Epoch: 60, Loss: 0.0008\nIter: 16, Epoch: 60, Loss: 0.0009\nIter: 17, Epoch: 60, Loss: 0.0013\nIter: 18, Epoch: 60, Loss: 0.0012\nIter: 19, Epoch: 60, Loss: 0.0017\nIter: 20, Epoch: 60, Loss: 0.0008\nIter: 21, Epoch: 60, Loss: 0.0015\nIter: 22, Epoch: 60, Loss: 0.0013\nIter: 23, Epoch: 60, Loss: 0.0015\nIter: 24, Epoch: 60, Loss: 0.0009\nIter: 25, Epoch: 60, Loss: 0.0010\nIter: 26, Epoch: 60, Loss: 0.0011\nIter: 27, Epoch: 60, Loss: 0.0008\nIter: 28, Epoch: 60, Loss: 0.0011\nIter: 29, Epoch: 60, Loss: 0.0016\nIter: 30, Epoch: 60, Loss: 0.0009\n","truncated":false}} +%--- +%[output:3ffc5c33] +% data: {"dataType":"text","outputData":{"text":">> Epoch 60 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:654071b7] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 61, Loss: 0.0030\nIter: 2, Epoch: 61, Loss: 0.0008\nIter: 3, Epoch: 61, Loss: 0.0012\nIter: 4, Epoch: 61, Loss: 0.0012\nIter: 5, Epoch: 61, Loss: 0.0008\nIter: 6, Epoch: 61, Loss: 0.0011\nIter: 7, Epoch: 61, Loss: 0.0010\nIter: 8, Epoch: 61, Loss: 0.0016\nIter: 9, Epoch: 61, Loss: 0.0014\nIter: 10, Epoch: 61, Loss: 0.0010\nIter: 11, Epoch: 61, Loss: 0.0008\nIter: 12, Epoch: 61, Loss: 0.0007\nIter: 13, Epoch: 61, Loss: 0.0009\nIter: 14, Epoch: 61, Loss: 0.0011\nIter: 15, Epoch: 61, Loss: 0.0008\nIter: 16, Epoch: 61, Loss: 0.0009\nIter: 17, Epoch: 61, Loss: 0.0012\nIter: 18, Epoch: 61, Loss: 0.0012\nIter: 19, Epoch: 61, Loss: 0.0017\nIter: 20, Epoch: 61, Loss: 0.0008\nIter: 21, Epoch: 61, Loss: 0.0014\nIter: 22, Epoch: 61, Loss: 0.0013\nIter: 23, Epoch: 61, Loss: 0.0015\nIter: 24, Epoch: 61, Loss: 0.0009\nIter: 25, Epoch: 61, Loss: 0.0010\nIter: 26, Epoch: 61, Loss: 0.0012\nIter: 27, Epoch: 61, Loss: 0.0008\nIter: 28, Epoch: 61, Loss: 0.0011\nIter: 29, Epoch: 61, Loss: 0.0015\nIter: 30, Epoch: 61, Loss: 0.0009\n","truncated":false}} +%--- +%[output:82e2f9b4] +% data: {"dataType":"text","outputData":{"text":">> Epoch 61 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:1a0f64d4] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 62, Loss: 0.0029\nIter: 2, Epoch: 62, Loss: 0.0008\nIter: 3, Epoch: 62, Loss: 0.0012\nIter: 4, Epoch: 62, Loss: 0.0011\nIter: 5, Epoch: 62, Loss: 0.0008\nIter: 6, Epoch: 62, Loss: 0.0011\nIter: 7, Epoch: 62, Loss: 0.0010\nIter: 8, Epoch: 62, Loss: 0.0016\nIter: 9, Epoch: 62, Loss: 0.0013\nIter: 10, Epoch: 62, Loss: 0.0010\nIter: 11, Epoch: 62, Loss: 0.0008\nIter: 12, Epoch: 62, Loss: 0.0007\nIter: 13, Epoch: 62, Loss: 0.0009\nIter: 14, Epoch: 62, Loss: 0.0010\nIter: 15, Epoch: 62, Loss: 0.0008\nIter: 16, Epoch: 62, Loss: 0.0009\nIter: 17, Epoch: 62, Loss: 0.0012\nIter: 18, Epoch: 62, Loss: 0.0012\nIter: 19, Epoch: 62, Loss: 0.0017\nIter: 20, Epoch: 62, Loss: 0.0008\nIter: 21, Epoch: 62, Loss: 0.0014\nIter: 22, Epoch: 62, Loss: 0.0013\nIter: 23, Epoch: 62, Loss: 0.0015\nIter: 24, Epoch: 62, Loss: 0.0009\nIter: 25, Epoch: 62, Loss: 0.0009\nIter: 26, Epoch: 62, Loss: 0.0011\nIter: 27, Epoch: 62, Loss: 0.0008\nIter: 28, Epoch: 62, Loss: 0.0011\nIter: 29, Epoch: 62, Loss: 0.0015\nIter: 30, Epoch: 62, Loss: 0.0008\n","truncated":false}} +%--- +%[output:2a095bb8] +% data: {"dataType":"text","outputData":{"text":">> Epoch 62 validation loss: 0.0021\n","truncated":false}} +%--- +%[output:69c511c7] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 63, Loss: 0.0030\nIter: 2, Epoch: 63, Loss: 0.0008\nIter: 3, Epoch: 63, Loss: 0.0012\nIter: 4, Epoch: 63, Loss: 0.0012\nIter: 5, Epoch: 63, Loss: 0.0008\nIter: 6, Epoch: 63, Loss: 0.0011\nIter: 7, Epoch: 63, Loss: 0.0010\nIter: 8, Epoch: 63, Loss: 0.0016\nIter: 9, Epoch: 63, Loss: 0.0014\nIter: 10, Epoch: 63, Loss: 0.0010\nIter: 11, Epoch: 63, Loss: 0.0008\nIter: 12, Epoch: 63, Loss: 0.0007\nIter: 13, Epoch: 63, Loss: 0.0009\nIter: 14, Epoch: 63, Loss: 0.0011\nIter: 15, Epoch: 63, Loss: 0.0008\nIter: 16, Epoch: 63, Loss: 0.0009\nIter: 17, Epoch: 63, Loss: 0.0012\nIter: 18, Epoch: 63, Loss: 0.0012\nIter: 19, Epoch: 63, Loss: 0.0016\nIter: 20, Epoch: 63, Loss: 0.0008\nIter: 21, Epoch: 63, Loss: 0.0014\nIter: 22, Epoch: 63, Loss: 0.0013\nIter: 23, Epoch: 63, Loss: 0.0015\nIter: 24, Epoch: 63, Loss: 0.0009\nIter: 25, Epoch: 63, Loss: 0.0009\nIter: 26, Epoch: 63, Loss: 0.0011\nIter: 27, Epoch: 63, Loss: 0.0008\nIter: 28, Epoch: 63, Loss: 0.0011\nIter: 29, Epoch: 63, Loss: 0.0015\nIter: 30, Epoch: 63, Loss: 0.0008\n","truncated":false}} +%--- +%[output:7d4b16d2] +% data: {"dataType":"text","outputData":{"text":">> Epoch 63 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:80c50658] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 64, Loss: 0.0030\nIter: 2, Epoch: 64, Loss: 0.0008\nIter: 3, Epoch: 64, Loss: 0.0012\nIter: 4, Epoch: 64, Loss: 0.0011\nIter: 5, Epoch: 64, Loss: 0.0008\nIter: 6, Epoch: 64, Loss: 0.0011\nIter: 7, Epoch: 64, Loss: 0.0010\nIter: 8, Epoch: 64, Loss: 0.0016\nIter: 9, Epoch: 64, Loss: 0.0013\nIter: 10, Epoch: 64, Loss: 0.0010\nIter: 11, Epoch: 64, Loss: 0.0008\nIter: 12, Epoch: 64, Loss: 0.0007\nIter: 13, Epoch: 64, Loss: 0.0010\nIter: 14, Epoch: 64, Loss: 0.0011\nIter: 15, Epoch: 64, Loss: 0.0008\nIter: 16, Epoch: 64, Loss: 0.0008\nIter: 17, Epoch: 64, Loss: 0.0012\nIter: 18, Epoch: 64, Loss: 0.0012\nIter: 19, Epoch: 64, Loss: 0.0016\nIter: 20, Epoch: 64, Loss: 0.0008\nIter: 21, Epoch: 64, Loss: 0.0014\nIter: 22, Epoch: 64, Loss: 0.0013\nIter: 23, Epoch: 64, Loss: 0.0014\nIter: 24, Epoch: 64, Loss: 0.0009\nIter: 25, Epoch: 64, Loss: 0.0009\nIter: 26, Epoch: 64, Loss: 0.0011\nIter: 27, Epoch: 64, Loss: 0.0008\nIter: 28, Epoch: 64, Loss: 0.0011\nIter: 29, Epoch: 64, Loss: 0.0015\nIter: 30, Epoch: 64, Loss: 0.0008\n","truncated":false}} +%--- +%[output:9d1a4a94] +% data: {"dataType":"text","outputData":{"text":">> Epoch 64 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:521ce776] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 65, Loss: 0.0029\nIter: 2, Epoch: 65, Loss: 0.0008\nIter: 3, Epoch: 65, Loss: 0.0012\nIter: 4, Epoch: 65, Loss: 0.0011\nIter: 5, Epoch: 65, Loss: 0.0008\nIter: 6, Epoch: 65, Loss: 0.0011\nIter: 7, Epoch: 65, Loss: 0.0010\nIter: 8, Epoch: 65, Loss: 0.0016\nIter: 9, Epoch: 65, Loss: 0.0013\nIter: 10, Epoch: 65, Loss: 0.0010\nIter: 11, Epoch: 65, Loss: 0.0008\nIter: 12, Epoch: 65, Loss: 0.0007\nIter: 13, Epoch: 65, Loss: 0.0009\nIter: 14, Epoch: 65, Loss: 0.0010\nIter: 15, Epoch: 65, Loss: 0.0008\nIter: 16, Epoch: 65, Loss: 0.0009\nIter: 17, Epoch: 65, Loss: 0.0012\nIter: 18, Epoch: 65, Loss: 0.0011\nIter: 19, Epoch: 65, Loss: 0.0016\nIter: 20, Epoch: 65, Loss: 0.0008\nIter: 21, Epoch: 65, Loss: 0.0014\nIter: 22, Epoch: 65, Loss: 0.0013\nIter: 23, Epoch: 65, Loss: 0.0015\nIter: 24, Epoch: 65, Loss: 0.0009\nIter: 25, Epoch: 65, Loss: 0.0009\nIter: 26, Epoch: 65, Loss: 0.0011\nIter: 27, Epoch: 65, Loss: 0.0008\nIter: 28, Epoch: 65, Loss: 0.0011\nIter: 29, Epoch: 65, Loss: 0.0015\nIter: 30, Epoch: 65, Loss: 0.0008\n","truncated":false}} +%--- +%[output:206aae00] +% data: {"dataType":"text","outputData":{"text":">> Epoch 65 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:6abfb90a] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 66, Loss: 0.0030\nIter: 2, Epoch: 66, Loss: 0.0008\nIter: 3, Epoch: 66, Loss: 0.0012\nIter: 4, Epoch: 66, Loss: 0.0011\nIter: 5, Epoch: 66, Loss: 0.0007\nIter: 6, Epoch: 66, Loss: 0.0011\nIter: 7, Epoch: 66, Loss: 0.0010\nIter: 8, Epoch: 66, Loss: 0.0016\nIter: 9, Epoch: 66, Loss: 0.0013\nIter: 10, Epoch: 66, Loss: 0.0010\nIter: 11, Epoch: 66, Loss: 0.0008\nIter: 12, Epoch: 66, Loss: 0.0006\nIter: 13, Epoch: 66, Loss: 0.0009\nIter: 14, Epoch: 66, Loss: 0.0010\nIter: 15, Epoch: 66, Loss: 0.0008\nIter: 16, Epoch: 66, Loss: 0.0008\nIter: 17, Epoch: 66, Loss: 0.0012\nIter: 18, Epoch: 66, Loss: 0.0011\nIter: 19, Epoch: 66, Loss: 0.0015\nIter: 20, Epoch: 66, Loss: 0.0008\nIter: 21, Epoch: 66, Loss: 0.0013\nIter: 22, Epoch: 66, Loss: 0.0012\nIter: 23, Epoch: 66, Loss: 0.0014\nIter: 24, Epoch: 66, Loss: 0.0009\nIter: 25, Epoch: 66, Loss: 0.0009\nIter: 26, Epoch: 66, Loss: 0.0011\nIter: 27, Epoch: 66, Loss: 0.0008\nIter: 28, Epoch: 66, Loss: 0.0011\nIter: 29, Epoch: 66, Loss: 0.0015\nIter: 30, Epoch: 66, Loss: 0.0008\n","truncated":false}} +%--- +%[output:9a189315] +% data: {"dataType":"text","outputData":{"text":">> Epoch 66 validation loss: 0.0019\n","truncated":false}} +%--- +%[output:49258648] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 67, Loss: 0.0028\nIter: 2, Epoch: 67, Loss: 0.0008\nIter: 3, Epoch: 67, Loss: 0.0011\nIter: 4, Epoch: 67, Loss: 0.0010\nIter: 5, Epoch: 67, Loss: 0.0007\nIter: 6, Epoch: 67, Loss: 0.0010\nIter: 7, Epoch: 67, Loss: 0.0010\nIter: 8, Epoch: 67, Loss: 0.0015\nIter: 9, Epoch: 67, Loss: 0.0013\nIter: 10, Epoch: 67, Loss: 0.0010\nIter: 11, Epoch: 67, Loss: 0.0007\nIter: 12, Epoch: 67, Loss: 0.0006\nIter: 13, Epoch: 67, Loss: 0.0009\nIter: 14, Epoch: 67, Loss: 0.0010\nIter: 15, Epoch: 67, Loss: 0.0008\nIter: 16, Epoch: 67, Loss: 0.0008\nIter: 17, Epoch: 67, Loss: 0.0011\nIter: 18, Epoch: 67, Loss: 0.0011\nIter: 19, Epoch: 67, Loss: 0.0016\nIter: 20, Epoch: 67, Loss: 0.0008\nIter: 21, Epoch: 67, Loss: 0.0013\nIter: 22, Epoch: 67, Loss: 0.0013\nIter: 23, Epoch: 67, Loss: 0.0015\nIter: 24, Epoch: 67, Loss: 0.0008\nIter: 25, Epoch: 67, Loss: 0.0008\nIter: 26, Epoch: 67, Loss: 0.0011\nIter: 27, Epoch: 67, Loss: 0.0008\nIter: 28, Epoch: 67, Loss: 0.0010\nIter: 29, Epoch: 67, Loss: 0.0015\nIter: 30, Epoch: 67, Loss: 0.0008\n","truncated":false}} +%--- +%[output:13f50c39] +% data: {"dataType":"text","outputData":{"text":">> Epoch 67 validation loss: 0.0019\n","truncated":false}} +%--- +%[output:467e4aca] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 68, Loss: 0.0029\nIter: 2, Epoch: 68, Loss: 0.0008\nIter: 3, Epoch: 68, Loss: 0.0011\nIter: 4, Epoch: 68, Loss: 0.0011\nIter: 5, Epoch: 68, Loss: 0.0007\nIter: 6, Epoch: 68, Loss: 0.0010\nIter: 7, Epoch: 68, Loss: 0.0010\nIter: 8, Epoch: 68, Loss: 0.0015\nIter: 9, Epoch: 68, Loss: 0.0013\nIter: 10, Epoch: 68, Loss: 0.0010\nIter: 11, Epoch: 68, Loss: 0.0007\nIter: 12, Epoch: 68, Loss: 0.0006\nIter: 13, Epoch: 68, Loss: 0.0009\nIter: 14, Epoch: 68, Loss: 0.0010\nIter: 15, Epoch: 68, Loss: 0.0008\nIter: 16, Epoch: 68, Loss: 0.0008\nIter: 17, Epoch: 68, Loss: 0.0011\nIter: 18, Epoch: 68, Loss: 0.0011\nIter: 19, Epoch: 68, Loss: 0.0016\nIter: 20, Epoch: 68, Loss: 0.0008\nIter: 21, Epoch: 68, Loss: 0.0013\nIter: 22, Epoch: 68, Loss: 0.0012\nIter: 23, Epoch: 68, Loss: 0.0014\nIter: 24, Epoch: 68, Loss: 0.0008\nIter: 25, Epoch: 68, Loss: 0.0008\nIter: 26, Epoch: 68, Loss: 0.0011\nIter: 27, Epoch: 68, Loss: 0.0008\nIter: 28, Epoch: 68, Loss: 0.0010\nIter: 29, Epoch: 68, Loss: 0.0015\nIter: 30, Epoch: 68, Loss: 0.0008\n","truncated":false}} +%--- +%[output:9c8f0d6a] +% data: {"dataType":"text","outputData":{"text":">> Epoch 68 validation loss: 0.0019\n","truncated":false}} +%--- +%[output:6e1a86eb] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 69, Loss: 0.0028\nIter: 2, Epoch: 69, Loss: 0.0008\nIter: 3, Epoch: 69, Loss: 0.0011\nIter: 4, Epoch: 69, Loss: 0.0011\nIter: 5, Epoch: 69, Loss: 0.0007\nIter: 6, Epoch: 69, Loss: 0.0010\nIter: 7, Epoch: 69, Loss: 0.0009\nIter: 8, Epoch: 69, Loss: 0.0015\nIter: 9, Epoch: 69, Loss: 0.0013\nIter: 10, Epoch: 69, Loss: 0.0010\nIter: 11, Epoch: 69, Loss: 0.0007\nIter: 12, Epoch: 69, Loss: 0.0006\nIter: 13, Epoch: 69, Loss: 0.0009\nIter: 14, Epoch: 69, Loss: 0.0010\nIter: 15, Epoch: 69, Loss: 0.0007\nIter: 16, Epoch: 69, Loss: 0.0008\nIter: 17, Epoch: 69, Loss: 0.0011\nIter: 18, Epoch: 69, Loss: 0.0011\nIter: 19, Epoch: 69, Loss: 0.0016\nIter: 20, Epoch: 69, Loss: 0.0008\nIter: 21, Epoch: 69, Loss: 0.0013\nIter: 22, Epoch: 69, Loss: 0.0012\nIter: 23, Epoch: 69, Loss: 0.0015\nIter: 24, Epoch: 69, Loss: 0.0008\nIter: 25, Epoch: 69, Loss: 0.0008\nIter: 26, Epoch: 69, Loss: 0.0011\nIter: 27, Epoch: 69, Loss: 0.0008\nIter: 28, Epoch: 69, Loss: 0.0010\nIter: 29, Epoch: 69, Loss: 0.0015\nIter: 30, Epoch: 69, Loss: 0.0008\n","truncated":false}} +%--- +%[output:2b3e43c3] +% data: {"dataType":"text","outputData":{"text":">> Epoch 69 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:5eea11aa] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 70, Loss: 0.0030\nIter: 2, Epoch: 70, Loss: 0.0008\nIter: 3, Epoch: 70, Loss: 0.0011\nIter: 4, Epoch: 70, Loss: 0.0011\nIter: 5, Epoch: 70, Loss: 0.0007\nIter: 6, Epoch: 70, Loss: 0.0010\nIter: 7, Epoch: 70, Loss: 0.0010\nIter: 8, Epoch: 70, Loss: 0.0016\nIter: 9, Epoch: 70, Loss: 0.0013\nIter: 10, Epoch: 70, Loss: 0.0010\nIter: 11, Epoch: 70, Loss: 0.0008\nIter: 12, Epoch: 70, Loss: 0.0006\nIter: 13, Epoch: 70, Loss: 0.0009\nIter: 14, Epoch: 70, Loss: 0.0010\nIter: 15, Epoch: 70, Loss: 0.0008\nIter: 16, Epoch: 70, Loss: 0.0008\nIter: 17, Epoch: 70, Loss: 0.0011\nIter: 18, Epoch: 70, Loss: 0.0010\nIter: 19, Epoch: 70, Loss: 0.0016\nIter: 20, Epoch: 70, Loss: 0.0007\nIter: 21, Epoch: 70, Loss: 0.0012\nIter: 22, Epoch: 70, Loss: 0.0012\nIter: 23, Epoch: 70, Loss: 0.0015\nIter: 24, Epoch: 70, Loss: 0.0008\nIter: 25, Epoch: 70, Loss: 0.0008\nIter: 26, Epoch: 70, Loss: 0.0011\nIter: 27, Epoch: 70, Loss: 0.0008\nIter: 28, Epoch: 70, Loss: 0.0010\nIter: 29, Epoch: 70, Loss: 0.0014\nIter: 30, Epoch: 70, Loss: 0.0008\n","truncated":false}} +%--- +%[output:49b55408] +% data: {"dataType":"text","outputData":{"text":">> Epoch 70 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:747bc9b1] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 71, Loss: 0.0029\nIter: 2, Epoch: 71, Loss: 0.0008\nIter: 3, Epoch: 71, Loss: 0.0011\nIter: 4, Epoch: 71, Loss: 0.0011\nIter: 5, Epoch: 71, Loss: 0.0007\nIter: 6, Epoch: 71, Loss: 0.0010\nIter: 7, Epoch: 71, Loss: 0.0009\nIter: 8, Epoch: 71, Loss: 0.0015\nIter: 9, Epoch: 71, Loss: 0.0013\nIter: 10, Epoch: 71, Loss: 0.0010\nIter: 11, Epoch: 71, Loss: 0.0007\nIter: 12, Epoch: 71, Loss: 0.0006\nIter: 13, Epoch: 71, Loss: 0.0009\nIter: 14, Epoch: 71, Loss: 0.0009\nIter: 15, Epoch: 71, Loss: 0.0008\nIter: 16, Epoch: 71, Loss: 0.0008\nIter: 17, Epoch: 71, Loss: 0.0011\nIter: 18, Epoch: 71, Loss: 0.0010\nIter: 19, Epoch: 71, Loss: 0.0016\nIter: 20, Epoch: 71, Loss: 0.0008\nIter: 21, Epoch: 71, Loss: 0.0013\nIter: 22, Epoch: 71, Loss: 0.0012\nIter: 23, Epoch: 71, Loss: 0.0015\nIter: 24, Epoch: 71, Loss: 0.0008\nIter: 25, Epoch: 71, Loss: 0.0008\nIter: 26, Epoch: 71, Loss: 0.0011\nIter: 27, Epoch: 71, Loss: 0.0007\nIter: 28, Epoch: 71, Loss: 0.0010\nIter: 29, Epoch: 71, Loss: 0.0015\nIter: 30, Epoch: 71, Loss: 0.0008\n","truncated":false}} +%--- +%[output:3596c36c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 71 validation loss: 0.0019\n","truncated":false}} +%--- +%[output:0a0fe026] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 72, Loss: 0.0029\nIter: 2, Epoch: 72, Loss: 0.0008\nIter: 3, Epoch: 72, Loss: 0.0011\nIter: 4, Epoch: 72, Loss: 0.0011\nIter: 5, Epoch: 72, Loss: 0.0007\nIter: 6, Epoch: 72, Loss: 0.0010\nIter: 7, Epoch: 72, Loss: 0.0009\nIter: 8, Epoch: 72, Loss: 0.0015\nIter: 9, Epoch: 72, Loss: 0.0012\nIter: 10, Epoch: 72, Loss: 0.0009\nIter: 11, Epoch: 72, Loss: 0.0007\nIter: 12, Epoch: 72, Loss: 0.0006\nIter: 13, Epoch: 72, Loss: 0.0009\nIter: 14, Epoch: 72, Loss: 0.0009\nIter: 15, Epoch: 72, Loss: 0.0008\nIter: 16, Epoch: 72, Loss: 0.0009\nIter: 17, Epoch: 72, Loss: 0.0011\nIter: 18, Epoch: 72, Loss: 0.0010\nIter: 19, Epoch: 72, Loss: 0.0016\nIter: 20, Epoch: 72, Loss: 0.0008\nIter: 21, Epoch: 72, Loss: 0.0013\nIter: 22, Epoch: 72, Loss: 0.0011\nIter: 23, Epoch: 72, Loss: 0.0015\nIter: 24, Epoch: 72, Loss: 0.0008\nIter: 25, Epoch: 72, Loss: 0.0007\nIter: 26, Epoch: 72, Loss: 0.0011\nIter: 27, Epoch: 72, Loss: 0.0007\nIter: 28, Epoch: 72, Loss: 0.0010\nIter: 29, Epoch: 72, Loss: 0.0014\nIter: 30, Epoch: 72, Loss: 0.0008\n","truncated":false}} +%--- +%[output:33a03ef6] +% data: {"dataType":"text","outputData":{"text":">> Epoch 72 validation loss: 0.0019\n","truncated":false}} +%--- +%[output:3dff3184] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 73, Loss: 0.0029\nIter: 2, Epoch: 73, Loss: 0.0008\nIter: 3, Epoch: 73, Loss: 0.0010\nIter: 4, Epoch: 73, Loss: 0.0011\nIter: 5, Epoch: 73, Loss: 0.0007\nIter: 6, Epoch: 73, Loss: 0.0010\nIter: 7, Epoch: 73, Loss: 0.0009\nIter: 8, Epoch: 73, Loss: 0.0016\nIter: 9, Epoch: 73, Loss: 0.0012\nIter: 10, Epoch: 73, Loss: 0.0010\nIter: 11, Epoch: 73, Loss: 0.0007\nIter: 12, Epoch: 73, Loss: 0.0006\nIter: 13, Epoch: 73, Loss: 0.0009\nIter: 14, Epoch: 73, Loss: 0.0009\nIter: 15, Epoch: 73, Loss: 0.0008\nIter: 16, Epoch: 73, Loss: 0.0008\nIter: 17, Epoch: 73, Loss: 0.0011\nIter: 18, Epoch: 73, Loss: 0.0010\nIter: 19, Epoch: 73, Loss: 0.0016\nIter: 20, Epoch: 73, Loss: 0.0008\nIter: 21, Epoch: 73, Loss: 0.0012\nIter: 22, Epoch: 73, Loss: 0.0012\nIter: 23, Epoch: 73, Loss: 0.0016\nIter: 24, Epoch: 73, Loss: 0.0008\nIter: 25, Epoch: 73, Loss: 0.0007\nIter: 26, Epoch: 73, Loss: 0.0011\nIter: 27, Epoch: 73, Loss: 0.0007\nIter: 28, Epoch: 73, Loss: 0.0010\nIter: 29, Epoch: 73, Loss: 0.0014\nIter: 30, Epoch: 73, Loss: 0.0008\n","truncated":false}} +%--- +%[output:8670b51c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 73 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:3b6a102e] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 74, Loss: 0.0030\nIter: 2, Epoch: 74, Loss: 0.0008\nIter: 3, Epoch: 74, Loss: 0.0010\nIter: 4, Epoch: 74, Loss: 0.0011\nIter: 5, Epoch: 74, Loss: 0.0007\nIter: 6, Epoch: 74, Loss: 0.0010\nIter: 7, Epoch: 74, Loss: 0.0009\nIter: 8, Epoch: 74, Loss: 0.0015\nIter: 9, Epoch: 74, Loss: 0.0012\nIter: 10, Epoch: 74, Loss: 0.0009\nIter: 11, Epoch: 74, Loss: 0.0007\nIter: 12, Epoch: 74, Loss: 0.0007\nIter: 13, Epoch: 74, Loss: 0.0009\nIter: 14, Epoch: 74, Loss: 0.0009\nIter: 15, Epoch: 74, Loss: 0.0008\nIter: 16, Epoch: 74, Loss: 0.0009\nIter: 17, Epoch: 74, Loss: 0.0010\nIter: 18, Epoch: 74, Loss: 0.0010\nIter: 19, Epoch: 74, Loss: 0.0016\nIter: 20, Epoch: 74, Loss: 0.0008\nIter: 21, Epoch: 74, Loss: 0.0012\nIter: 22, Epoch: 74, Loss: 0.0011\nIter: 23, Epoch: 74, Loss: 0.0016\nIter: 24, Epoch: 74, Loss: 0.0008\nIter: 25, Epoch: 74, Loss: 0.0007\nIter: 26, Epoch: 74, Loss: 0.0011\nIter: 27, Epoch: 74, Loss: 0.0007\nIter: 28, Epoch: 74, Loss: 0.0010\nIter: 29, Epoch: 74, Loss: 0.0014\nIter: 30, Epoch: 74, Loss: 0.0008\n","truncated":false}} +%--- +%[output:13fd599d] +% data: {"dataType":"text","outputData":{"text":">> Epoch 74 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:90476dad] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 75, Loss: 0.0030\nIter: 2, Epoch: 75, Loss: 0.0008\nIter: 3, Epoch: 75, Loss: 0.0010\nIter: 4, Epoch: 75, Loss: 0.0011\nIter: 5, Epoch: 75, Loss: 0.0007\nIter: 6, Epoch: 75, Loss: 0.0010\nIter: 7, Epoch: 75, Loss: 0.0009\nIter: 8, Epoch: 75, Loss: 0.0015\nIter: 9, Epoch: 75, Loss: 0.0012\nIter: 10, Epoch: 75, Loss: 0.0009\nIter: 11, Epoch: 75, Loss: 0.0007\nIter: 12, Epoch: 75, Loss: 0.0006\nIter: 13, Epoch: 75, Loss: 0.0009\nIter: 14, Epoch: 75, Loss: 0.0009\nIter: 15, Epoch: 75, Loss: 0.0008\nIter: 16, Epoch: 75, Loss: 0.0009\nIter: 17, Epoch: 75, Loss: 0.0010\nIter: 18, Epoch: 75, Loss: 0.0010\nIter: 19, Epoch: 75, Loss: 0.0016\nIter: 20, Epoch: 75, Loss: 0.0008\nIter: 21, Epoch: 75, Loss: 0.0012\nIter: 22, Epoch: 75, Loss: 0.0012\nIter: 23, Epoch: 75, Loss: 0.0015\nIter: 24, Epoch: 75, Loss: 0.0008\nIter: 25, Epoch: 75, Loss: 0.0007\nIter: 26, Epoch: 75, Loss: 0.0010\nIter: 27, Epoch: 75, Loss: 0.0007\nIter: 28, Epoch: 75, Loss: 0.0010\nIter: 29, Epoch: 75, Loss: 0.0014\nIter: 30, Epoch: 75, Loss: 0.0008\n","truncated":false}} +%--- +%[output:046fc08a] +% data: {"dataType":"text","outputData":{"text":">> Epoch 75 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:4c128056] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 76, Loss: 0.0030\nIter: 2, Epoch: 76, Loss: 0.0008\nIter: 3, Epoch: 76, Loss: 0.0010\nIter: 4, Epoch: 76, Loss: 0.0011\nIter: 5, Epoch: 76, Loss: 0.0007\nIter: 6, Epoch: 76, Loss: 0.0010\nIter: 7, Epoch: 76, Loss: 0.0009\nIter: 8, Epoch: 76, Loss: 0.0015\nIter: 9, Epoch: 76, Loss: 0.0012\nIter: 10, Epoch: 76, Loss: 0.0009\nIter: 11, Epoch: 76, Loss: 0.0007\nIter: 12, Epoch: 76, Loss: 0.0007\nIter: 13, Epoch: 76, Loss: 0.0009\nIter: 14, Epoch: 76, Loss: 0.0009\nIter: 15, Epoch: 76, Loss: 0.0008\nIter: 16, Epoch: 76, Loss: 0.0009\nIter: 17, Epoch: 76, Loss: 0.0010\nIter: 18, Epoch: 76, Loss: 0.0009\nIter: 19, Epoch: 76, Loss: 0.0016\nIter: 20, Epoch: 76, Loss: 0.0008\nIter: 21, Epoch: 76, Loss: 0.0012\nIter: 22, Epoch: 76, Loss: 0.0011\nIter: 23, Epoch: 76, Loss: 0.0015\nIter: 24, Epoch: 76, Loss: 0.0008\nIter: 25, Epoch: 76, Loss: 0.0007\nIter: 26, Epoch: 76, Loss: 0.0010\nIter: 27, Epoch: 76, Loss: 0.0007\nIter: 28, Epoch: 76, Loss: 0.0010\nIter: 29, Epoch: 76, Loss: 0.0013\nIter: 30, Epoch: 76, Loss: 0.0008\n","truncated":false}} +%--- +%[output:7d8fa1d9] +% data: {"dataType":"text","outputData":{"text":">> Epoch 76 validation loss: 0.0021\n","truncated":false}} +%--- +%[output:4d184a5c] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 77, Loss: 0.0031\nIter: 2, Epoch: 77, Loss: 0.0008\nIter: 3, Epoch: 77, Loss: 0.0010\nIter: 4, Epoch: 77, Loss: 0.0011\nIter: 5, Epoch: 77, Loss: 0.0007\nIter: 6, Epoch: 77, Loss: 0.0009\nIter: 7, Epoch: 77, Loss: 0.0009\nIter: 8, Epoch: 77, Loss: 0.0016\nIter: 9, Epoch: 77, Loss: 0.0012\nIter: 10, Epoch: 77, Loss: 0.0009\nIter: 11, Epoch: 77, Loss: 0.0007\nIter: 12, Epoch: 77, Loss: 0.0007\nIter: 13, Epoch: 77, Loss: 0.0009\nIter: 14, Epoch: 77, Loss: 0.0009\nIter: 15, Epoch: 77, Loss: 0.0009\nIter: 16, Epoch: 77, Loss: 0.0009\nIter: 17, Epoch: 77, Loss: 0.0010\nIter: 18, Epoch: 77, Loss: 0.0009\nIter: 19, Epoch: 77, Loss: 0.0017\nIter: 20, Epoch: 77, Loss: 0.0008\nIter: 21, Epoch: 77, Loss: 0.0012\nIter: 22, Epoch: 77, Loss: 0.0011\nIter: 23, Epoch: 77, Loss: 0.0016\nIter: 24, Epoch: 77, Loss: 0.0008\nIter: 25, Epoch: 77, Loss: 0.0007\nIter: 26, Epoch: 77, Loss: 0.0010\nIter: 27, Epoch: 77, Loss: 0.0007\nIter: 28, Epoch: 77, Loss: 0.0010\nIter: 29, Epoch: 77, Loss: 0.0014\nIter: 30, Epoch: 77, Loss: 0.0007\n","truncated":false}} +%--- +%[output:149b1853] +% data: {"dataType":"text","outputData":{"text":">> Epoch 77 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:189227c0] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 78, Loss: 0.0030\nIter: 2, Epoch: 78, Loss: 0.0008\nIter: 3, Epoch: 78, Loss: 0.0010\nIter: 4, Epoch: 78, Loss: 0.0011\nIter: 5, Epoch: 78, Loss: 0.0008\nIter: 6, Epoch: 78, Loss: 0.0010\nIter: 7, Epoch: 78, Loss: 0.0009\nIter: 8, Epoch: 78, Loss: 0.0016\nIter: 9, Epoch: 78, Loss: 0.0012\nIter: 10, Epoch: 78, Loss: 0.0009\nIter: 11, Epoch: 78, Loss: 0.0007\nIter: 12, Epoch: 78, Loss: 0.0007\nIter: 13, Epoch: 78, Loss: 0.0009\nIter: 14, Epoch: 78, Loss: 0.0009\nIter: 15, Epoch: 78, Loss: 0.0008\nIter: 16, Epoch: 78, Loss: 0.0010\nIter: 17, Epoch: 78, Loss: 0.0011\nIter: 18, Epoch: 78, Loss: 0.0009\nIter: 19, Epoch: 78, Loss: 0.0016\nIter: 20, Epoch: 78, Loss: 0.0009\nIter: 21, Epoch: 78, Loss: 0.0011\nIter: 22, Epoch: 78, Loss: 0.0011\nIter: 23, Epoch: 78, Loss: 0.0016\nIter: 24, Epoch: 78, Loss: 0.0009\nIter: 25, Epoch: 78, Loss: 0.0006\nIter: 26, Epoch: 78, Loss: 0.0010\nIter: 27, Epoch: 78, Loss: 0.0007\nIter: 28, Epoch: 78, Loss: 0.0011\nIter: 29, Epoch: 78, Loss: 0.0013\nIter: 30, Epoch: 78, Loss: 0.0007\n","truncated":false}} +%--- +%[output:64743cc5] +% data: {"dataType":"text","outputData":{"text":">> Epoch 78 validation loss: 0.0021\n","truncated":false}} +%--- +%[output:39ba7841] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 79, Loss: 0.0032\nIter: 2, Epoch: 79, Loss: 0.0008\nIter: 3, Epoch: 79, Loss: 0.0009\nIter: 4, Epoch: 79, Loss: 0.0011\nIter: 5, Epoch: 79, Loss: 0.0008\nIter: 6, Epoch: 79, Loss: 0.0009\nIter: 7, Epoch: 79, Loss: 0.0009\nIter: 8, Epoch: 79, Loss: 0.0016\nIter: 9, Epoch: 79, Loss: 0.0013\nIter: 10, Epoch: 79, Loss: 0.0009\nIter: 11, Epoch: 79, Loss: 0.0007\nIter: 12, Epoch: 79, Loss: 0.0007\nIter: 13, Epoch: 79, Loss: 0.0010\nIter: 14, Epoch: 79, Loss: 0.0009\nIter: 15, Epoch: 79, Loss: 0.0008\nIter: 16, Epoch: 79, Loss: 0.0010\nIter: 17, Epoch: 79, Loss: 0.0011\nIter: 18, Epoch: 79, Loss: 0.0008\nIter: 19, Epoch: 79, Loss: 0.0016\nIter: 20, Epoch: 79, Loss: 0.0009\nIter: 21, Epoch: 79, Loss: 0.0013\nIter: 22, Epoch: 79, Loss: 0.0011\nIter: 23, Epoch: 79, Loss: 0.0016\nIter: 24, Epoch: 79, Loss: 0.0009\nIter: 25, Epoch: 79, Loss: 0.0007\nIter: 26, Epoch: 79, Loss: 0.0010\nIter: 27, Epoch: 79, Loss: 0.0007\nIter: 28, Epoch: 79, Loss: 0.0012\nIter: 29, Epoch: 79, Loss: 0.0014\nIter: 30, Epoch: 79, Loss: 0.0007\n","truncated":false}} +%--- +%[output:0e698a3c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 79 validation loss: 0.0022\n","truncated":false}} +%--- +%[output:7c592cf0] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 80, Loss: 0.0032\nIter: 2, Epoch: 80, Loss: 0.0008\nIter: 3, Epoch: 80, Loss: 0.0009\nIter: 4, Epoch: 80, Loss: 0.0011\nIter: 5, Epoch: 80, Loss: 0.0009\nIter: 6, Epoch: 80, Loss: 0.0010\nIter: 7, Epoch: 80, Loss: 0.0009\nIter: 8, Epoch: 80, Loss: 0.0016\nIter: 9, Epoch: 80, Loss: 0.0014\nIter: 10, Epoch: 80, Loss: 0.0010\nIter: 11, Epoch: 80, Loss: 0.0006\nIter: 12, Epoch: 80, Loss: 0.0008\nIter: 13, Epoch: 80, Loss: 0.0010\nIter: 14, Epoch: 80, Loss: 0.0009\nIter: 15, Epoch: 80, Loss: 0.0008\nIter: 16, Epoch: 80, Loss: 0.0010\nIter: 17, Epoch: 80, Loss: 0.0012\nIter: 18, Epoch: 80, Loss: 0.0008\nIter: 19, Epoch: 80, Loss: 0.0015\nIter: 20, Epoch: 80, Loss: 0.0009\nIter: 21, Epoch: 80, Loss: 0.0012\nIter: 22, Epoch: 80, Loss: 0.0011\nIter: 23, Epoch: 80, Loss: 0.0016\nIter: 24, Epoch: 80, Loss: 0.0009\nIter: 25, Epoch: 80, Loss: 0.0007\nIter: 26, Epoch: 80, Loss: 0.0010\nIter: 27, Epoch: 80, Loss: 0.0008\nIter: 28, Epoch: 80, Loss: 0.0012\nIter: 29, Epoch: 80, Loss: 0.0014\nIter: 30, Epoch: 80, Loss: 0.0007\n","truncated":false}} +%--- +%[output:7d856e8d] +% data: {"dataType":"text","outputData":{"text":">> Epoch 80 validation loss: 0.0021\n","truncated":false}} +%--- +%[output:8c86dca4] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 81, Loss: 0.0031\nIter: 2, Epoch: 81, Loss: 0.0008\nIter: 3, Epoch: 81, Loss: 0.0009\nIter: 4, Epoch: 81, Loss: 0.0011\nIter: 5, Epoch: 81, Loss: 0.0009\nIter: 6, Epoch: 81, Loss: 0.0010\nIter: 7, Epoch: 81, Loss: 0.0009\nIter: 8, Epoch: 81, Loss: 0.0015\nIter: 9, Epoch: 81, Loss: 0.0013\nIter: 10, Epoch: 81, Loss: 0.0010\nIter: 11, Epoch: 81, Loss: 0.0006\nIter: 12, Epoch: 81, Loss: 0.0007\nIter: 13, Epoch: 81, Loss: 0.0011\nIter: 14, Epoch: 81, Loss: 0.0009\nIter: 15, Epoch: 81, Loss: 0.0007\nIter: 16, Epoch: 81, Loss: 0.0010\nIter: 17, Epoch: 81, Loss: 0.0013\nIter: 18, Epoch: 81, Loss: 0.0008\nIter: 19, Epoch: 81, Loss: 0.0014\nIter: 20, Epoch: 81, Loss: 0.0010\nIter: 21, Epoch: 81, Loss: 0.0014\nIter: 22, Epoch: 81, Loss: 0.0011\nIter: 23, Epoch: 81, Loss: 0.0014\nIter: 24, Epoch: 81, Loss: 0.0010\nIter: 25, Epoch: 81, Loss: 0.0008\nIter: 26, Epoch: 81, Loss: 0.0010\nIter: 27, Epoch: 81, Loss: 0.0008\nIter: 28, Epoch: 81, Loss: 0.0013\nIter: 29, Epoch: 81, Loss: 0.0015\nIter: 30, Epoch: 81, Loss: 0.0007\n","truncated":false}} +%--- +%[output:843b70e3] +% data: {"dataType":"text","outputData":{"text":">> Epoch 81 validation loss: 0.0020\n","truncated":false}} +%--- +%[output:41b2baf4] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 82, Loss: 0.0030\nIter: 2, Epoch: 82, Loss: 0.0008\nIter: 3, Epoch: 82, Loss: 0.0010\nIter: 4, Epoch: 82, Loss: 0.0010\nIter: 5, Epoch: 82, Loss: 0.0008\nIter: 6, Epoch: 82, Loss: 0.0011\nIter: 7, Epoch: 82, Loss: 0.0009\nIter: 8, Epoch: 82, Loss: 0.0014\nIter: 9, Epoch: 82, Loss: 0.0014\nIter: 10, Epoch: 82, Loss: 0.0011\nIter: 11, Epoch: 82, Loss: 0.0006\nIter: 12, Epoch: 82, Loss: 0.0007\nIter: 13, Epoch: 82, Loss: 0.0011\nIter: 14, Epoch: 82, Loss: 0.0010\nIter: 15, Epoch: 82, Loss: 0.0007\nIter: 16, Epoch: 82, Loss: 0.0010\nIter: 17, Epoch: 82, Loss: 0.0013\nIter: 18, Epoch: 82, Loss: 0.0009\nIter: 19, Epoch: 82, Loss: 0.0013\nIter: 20, Epoch: 82, Loss: 0.0010\nIter: 21, Epoch: 82, Loss: 0.0014\nIter: 22, Epoch: 82, Loss: 0.0011\nIter: 23, Epoch: 82, Loss: 0.0014\nIter: 24, Epoch: 82, Loss: 0.0010\nIter: 25, Epoch: 82, Loss: 0.0008\nIter: 26, Epoch: 82, Loss: 0.0009\nIter: 27, Epoch: 82, Loss: 0.0008\nIter: 28, Epoch: 82, Loss: 0.0013\nIter: 29, Epoch: 82, Loss: 0.0015\nIter: 30, Epoch: 82, Loss: 0.0007\n","truncated":false}} +%--- +%[output:67bae44b] +% data: {"dataType":"text","outputData":{"text":">> Epoch 82 validation loss: 0.0019\n","truncated":false}} +%--- +%[output:84fd4b08] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 83, Loss: 0.0029\nIter: 2, Epoch: 83, Loss: 0.0008\nIter: 3, Epoch: 83, Loss: 0.0010\nIter: 4, Epoch: 83, Loss: 0.0009\nIter: 5, Epoch: 83, Loss: 0.0008\nIter: 6, Epoch: 83, Loss: 0.0011\nIter: 7, Epoch: 83, Loss: 0.0008\nIter: 8, Epoch: 83, Loss: 0.0014\nIter: 9, Epoch: 83, Loss: 0.0014\nIter: 10, Epoch: 83, Loss: 0.0011\nIter: 11, Epoch: 83, Loss: 0.0006\nIter: 12, Epoch: 83, Loss: 0.0006\nIter: 13, Epoch: 83, Loss: 0.0011\nIter: 14, Epoch: 83, Loss: 0.0011\nIter: 15, Epoch: 83, Loss: 0.0006\nIter: 16, Epoch: 83, Loss: 0.0009\nIter: 17, Epoch: 83, Loss: 0.0013\nIter: 18, Epoch: 83, Loss: 0.0009\nIter: 19, Epoch: 83, Loss: 0.0012\nIter: 20, Epoch: 83, Loss: 0.0010\nIter: 21, Epoch: 83, Loss: 0.0015\nIter: 22, Epoch: 83, Loss: 0.0012\nIter: 23, Epoch: 83, Loss: 0.0013\nIter: 24, Epoch: 83, Loss: 0.0009\nIter: 25, Epoch: 83, Loss: 0.0009\nIter: 26, Epoch: 83, Loss: 0.0010\nIter: 27, Epoch: 83, Loss: 0.0008\nIter: 28, Epoch: 83, Loss: 0.0012\nIter: 29, Epoch: 83, Loss: 0.0015\nIter: 30, Epoch: 83, Loss: 0.0007\n","truncated":false}} +%--- +%[output:1daec627] +% data: {"dataType":"text","outputData":{"text":">> Epoch 83 validation loss: 0.0018\n","truncated":false}} +%--- +%[output:2ce57674] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 84, Loss: 0.0026\nIter: 2, Epoch: 84, Loss: 0.0008\nIter: 3, Epoch: 84, Loss: 0.0011\nIter: 4, Epoch: 84, Loss: 0.0010\nIter: 5, Epoch: 84, Loss: 0.0007\nIter: 6, Epoch: 84, Loss: 0.0011\nIter: 7, Epoch: 84, Loss: 0.0009\nIter: 8, Epoch: 84, Loss: 0.0013\nIter: 9, Epoch: 84, Loss: 0.0014\nIter: 10, Epoch: 84, Loss: 0.0012\nIter: 11, Epoch: 84, Loss: 0.0006\nIter: 12, Epoch: 84, Loss: 0.0006\nIter: 13, Epoch: 84, Loss: 0.0009\nIter: 14, Epoch: 84, Loss: 0.0011\nIter: 15, Epoch: 84, Loss: 0.0006\nIter: 16, Epoch: 84, Loss: 0.0008\nIter: 17, Epoch: 84, Loss: 0.0012\nIter: 18, Epoch: 84, Loss: 0.0010\nIter: 19, Epoch: 84, Loss: 0.0011\nIter: 20, Epoch: 84, Loss: 0.0009\nIter: 21, Epoch: 84, Loss: 0.0016\nIter: 22, Epoch: 84, Loss: 0.0012\nIter: 23, Epoch: 84, Loss: 0.0012\nIter: 24, Epoch: 84, Loss: 0.0009\nIter: 25, Epoch: 84, Loss: 0.0009\nIter: 26, Epoch: 84, Loss: 0.0010\nIter: 27, Epoch: 84, Loss: 0.0008\nIter: 28, Epoch: 84, Loss: 0.0012\nIter: 29, Epoch: 84, Loss: 0.0015\nIter: 30, Epoch: 84, Loss: 0.0007\n","truncated":false}} +%--- +%[output:4efd54a7] +% data: {"dataType":"text","outputData":{"text":">> Epoch 84 validation loss: 0.0017\n","truncated":false}} +%--- +%[output:9803f580] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 85, Loss: 0.0025\nIter: 2, Epoch: 85, Loss: 0.0008\nIter: 3, Epoch: 85, Loss: 0.0011\nIter: 4, Epoch: 85, Loss: 0.0010\nIter: 5, Epoch: 85, Loss: 0.0007\nIter: 6, Epoch: 85, Loss: 0.0011\nIter: 7, Epoch: 85, Loss: 0.0009\nIter: 8, Epoch: 85, Loss: 0.0014\nIter: 9, Epoch: 85, Loss: 0.0013\nIter: 10, Epoch: 85, Loss: 0.0011\nIter: 11, Epoch: 85, Loss: 0.0006\nIter: 12, Epoch: 85, Loss: 0.0006\nIter: 13, Epoch: 85, Loss: 0.0008\nIter: 14, Epoch: 85, Loss: 0.0011\nIter: 15, Epoch: 85, Loss: 0.0007\nIter: 16, Epoch: 85, Loss: 0.0007\nIter: 17, Epoch: 85, Loss: 0.0012\nIter: 18, Epoch: 85, Loss: 0.0011\nIter: 19, Epoch: 85, Loss: 0.0011\nIter: 20, Epoch: 85, Loss: 0.0008\nIter: 21, Epoch: 85, Loss: 0.0016\nIter: 22, Epoch: 85, Loss: 0.0012\nIter: 23, Epoch: 85, Loss: 0.0012\nIter: 24, Epoch: 85, Loss: 0.0008\nIter: 25, Epoch: 85, Loss: 0.0009\nIter: 26, Epoch: 85, Loss: 0.0011\nIter: 27, Epoch: 85, Loss: 0.0009\nIter: 28, Epoch: 85, Loss: 0.0010\nIter: 29, Epoch: 85, Loss: 0.0015\nIter: 30, Epoch: 85, Loss: 0.0008\n","truncated":false}} +%--- +%[output:09a37f06] +% data: {"dataType":"text","outputData":{"text":">> Epoch 85 validation loss: 0.0017\n","truncated":false}} +%--- +%[output:4b4afd09] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 86, Loss: 0.0023\nIter: 2, Epoch: 86, Loss: 0.0007\nIter: 3, Epoch: 86, Loss: 0.0011\nIter: 4, Epoch: 86, Loss: 0.0010\nIter: 5, Epoch: 86, Loss: 0.0006\nIter: 6, Epoch: 86, Loss: 0.0010\nIter: 7, Epoch: 86, Loss: 0.0009\nIter: 8, Epoch: 86, Loss: 0.0014\nIter: 9, Epoch: 86, Loss: 0.0013\nIter: 10, Epoch: 86, Loss: 0.0010\nIter: 11, Epoch: 86, Loss: 0.0007\nIter: 12, Epoch: 86, Loss: 0.0006\nIter: 13, Epoch: 86, Loss: 0.0007\nIter: 14, Epoch: 86, Loss: 0.0011\nIter: 15, Epoch: 86, Loss: 0.0008\nIter: 16, Epoch: 86, Loss: 0.0007\nIter: 17, Epoch: 86, Loss: 0.0011\nIter: 18, Epoch: 86, Loss: 0.0011\nIter: 19, Epoch: 86, Loss: 0.0012\nIter: 20, Epoch: 86, Loss: 0.0007\nIter: 21, Epoch: 86, Loss: 0.0015\nIter: 22, Epoch: 86, Loss: 0.0013\nIter: 23, Epoch: 86, Loss: 0.0013\nIter: 24, Epoch: 86, Loss: 0.0007\nIter: 25, Epoch: 86, Loss: 0.0008\nIter: 26, Epoch: 86, Loss: 0.0011\nIter: 27, Epoch: 86, Loss: 0.0009\nIter: 28, Epoch: 86, Loss: 0.0009\nIter: 29, Epoch: 86, Loss: 0.0013\nIter: 30, Epoch: 86, Loss: 0.0008\n","truncated":false}} +%--- +%[output:2aa50591] +% data: {"dataType":"text","outputData":{"text":">> Epoch 86 validation loss: 0.0016\n","truncated":false}} +%--- +%[output:4b3de16c] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 87, Loss: 0.0022\nIter: 2, Epoch: 87, Loss: 0.0007\nIter: 3, Epoch: 87, Loss: 0.0011\nIter: 4, Epoch: 87, Loss: 0.0010\nIter: 5, Epoch: 87, Loss: 0.0006\nIter: 6, Epoch: 87, Loss: 0.0010\nIter: 7, Epoch: 87, Loss: 0.0009\nIter: 8, Epoch: 87, Loss: 0.0015\nIter: 9, Epoch: 87, Loss: 0.0012\nIter: 10, Epoch: 87, Loss: 0.0010\nIter: 11, Epoch: 87, Loss: 0.0007\nIter: 12, Epoch: 87, Loss: 0.0007\nIter: 13, Epoch: 87, Loss: 0.0007\nIter: 14, Epoch: 87, Loss: 0.0010\nIter: 15, Epoch: 87, Loss: 0.0008\nIter: 16, Epoch: 87, Loss: 0.0007\nIter: 17, Epoch: 87, Loss: 0.0010\nIter: 18, Epoch: 87, Loss: 0.0011\nIter: 19, Epoch: 87, Loss: 0.0014\nIter: 20, Epoch: 87, Loss: 0.0007\nIter: 21, Epoch: 87, Loss: 0.0015\nIter: 22, Epoch: 87, Loss: 0.0012\nIter: 23, Epoch: 87, Loss: 0.0013\nIter: 24, Epoch: 87, Loss: 0.0007\nIter: 25, Epoch: 87, Loss: 0.0008\nIter: 26, Epoch: 87, Loss: 0.0011\nIter: 27, Epoch: 87, Loss: 0.0010\nIter: 28, Epoch: 87, Loss: 0.0009\nIter: 29, Epoch: 87, Loss: 0.0013\nIter: 30, Epoch: 87, Loss: 0.0008\n","truncated":false}} +%--- +%[output:94037ed4] +% data: {"dataType":"text","outputData":{"text":">> Epoch 87 validation loss: 0.0016\n","truncated":false}} +%--- +%[output:7d213b50] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 88, Loss: 0.0022\nIter: 2, Epoch: 88, Loss: 0.0007\nIter: 3, Epoch: 88, Loss: 0.0010\nIter: 4, Epoch: 88, Loss: 0.0011\nIter: 5, Epoch: 88, Loss: 0.0005\nIter: 6, Epoch: 88, Loss: 0.0009\nIter: 7, Epoch: 88, Loss: 0.0009\nIter: 8, Epoch: 88, Loss: 0.0016\nIter: 9, Epoch: 88, Loss: 0.0011\nIter: 10, Epoch: 88, Loss: 0.0009\nIter: 11, Epoch: 88, Loss: 0.0007\nIter: 12, Epoch: 88, Loss: 0.0007\nIter: 13, Epoch: 88, Loss: 0.0007\nIter: 14, Epoch: 88, Loss: 0.0010\nIter: 15, Epoch: 88, Loss: 0.0008\nIter: 16, Epoch: 88, Loss: 0.0007\nIter: 17, Epoch: 88, Loss: 0.0009\nIter: 18, Epoch: 88, Loss: 0.0011\nIter: 19, Epoch: 88, Loss: 0.0015\nIter: 20, Epoch: 88, Loss: 0.0006\nIter: 21, Epoch: 88, Loss: 0.0014\nIter: 22, Epoch: 88, Loss: 0.0012\nIter: 23, Epoch: 88, Loss: 0.0014\nIter: 24, Epoch: 88, Loss: 0.0007\nIter: 25, Epoch: 88, Loss: 0.0007\nIter: 26, Epoch: 88, Loss: 0.0011\nIter: 27, Epoch: 88, Loss: 0.0010\nIter: 28, Epoch: 88, Loss: 0.0008\nIter: 29, Epoch: 88, Loss: 0.0012\nIter: 30, Epoch: 88, Loss: 0.0008\n","truncated":false}} +%--- +%[output:02293857] +% data: {"dataType":"text","outputData":{"text":">> Epoch 88 validation loss: 0.0017\n","truncated":false}} +%--- +%[output:00d641d0] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 89, Loss: 0.0023\nIter: 2, Epoch: 89, Loss: 0.0007\nIter: 3, Epoch: 89, Loss: 0.0010\nIter: 4, Epoch: 89, Loss: 0.0011\nIter: 5, Epoch: 89, Loss: 0.0006\nIter: 6, Epoch: 89, Loss: 0.0008\nIter: 7, Epoch: 89, Loss: 0.0008\nIter: 8, Epoch: 89, Loss: 0.0016\nIter: 9, Epoch: 89, Loss: 0.0011\nIter: 10, Epoch: 89, Loss: 0.0008\nIter: 11, Epoch: 89, Loss: 0.0007\nIter: 12, Epoch: 89, Loss: 0.0007\nIter: 13, Epoch: 89, Loss: 0.0008\nIter: 14, Epoch: 89, Loss: 0.0009\nIter: 15, Epoch: 89, Loss: 0.0008\nIter: 16, Epoch: 89, Loss: 0.0008\nIter: 17, Epoch: 89, Loss: 0.0009\nIter: 18, Epoch: 89, Loss: 0.0010\nIter: 19, Epoch: 89, Loss: 0.0016\nIter: 20, Epoch: 89, Loss: 0.0006\nIter: 21, Epoch: 89, Loss: 0.0013\nIter: 22, Epoch: 89, Loss: 0.0011\nIter: 23, Epoch: 89, Loss: 0.0015\nIter: 24, Epoch: 89, Loss: 0.0007\nIter: 25, Epoch: 89, Loss: 0.0007\nIter: 26, Epoch: 89, Loss: 0.0011\nIter: 27, Epoch: 89, Loss: 0.0011\nIter: 28, Epoch: 89, Loss: 0.0009\nIter: 29, Epoch: 89, Loss: 0.0012\nIter: 30, Epoch: 89, Loss: 0.0007\n","truncated":false}} +%--- +%[output:9358c128] +% data: {"dataType":"text","outputData":{"text":">> Epoch 89 validation loss: 0.0018\n","truncated":false}} +%--- +%[output:6ce5ffd8] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 90, Loss: 0.0025\nIter: 2, Epoch: 90, Loss: 0.0007\nIter: 3, Epoch: 90, Loss: 0.0010\nIter: 4, Epoch: 90, Loss: 0.0011\nIter: 5, Epoch: 90, Loss: 0.0007\nIter: 6, Epoch: 90, Loss: 0.0008\nIter: 7, Epoch: 90, Loss: 0.0008\nIter: 8, Epoch: 90, Loss: 0.0017\nIter: 9, Epoch: 90, Loss: 0.0011\nIter: 10, Epoch: 90, Loss: 0.0009\nIter: 11, Epoch: 90, Loss: 0.0006\nIter: 12, Epoch: 90, Loss: 0.0007\nIter: 13, Epoch: 90, Loss: 0.0009\nIter: 14, Epoch: 90, Loss: 0.0008\nIter: 15, Epoch: 90, Loss: 0.0008\nIter: 16, Epoch: 90, Loss: 0.0008\nIter: 17, Epoch: 90, Loss: 0.0010\nIter: 18, Epoch: 90, Loss: 0.0009\nIter: 19, Epoch: 90, Loss: 0.0016\nIter: 20, Epoch: 90, Loss: 0.0007\nIter: 21, Epoch: 90, Loss: 0.0012\nIter: 22, Epoch: 90, Loss: 0.0011\nIter: 23, Epoch: 90, Loss: 0.0015\nIter: 24, Epoch: 90, Loss: 0.0008\nIter: 25, Epoch: 90, Loss: 0.0007\nIter: 26, Epoch: 90, Loss: 0.0010\nIter: 27, Epoch: 90, Loss: 0.0010\nIter: 28, Epoch: 90, Loss: 0.0010\nIter: 29, Epoch: 90, Loss: 0.0012\nIter: 30, Epoch: 90, Loss: 0.0007\n","truncated":false}} +%--- +%[output:25820879] +% data: {"dataType":"text","outputData":{"text":">> Epoch 90 validation loss: 0.0019\n","truncated":false}} +%--- +%[output:9ee1f7e4] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 91, Loss: 0.0026\nIter: 2, Epoch: 91, Loss: 0.0008\nIter: 3, Epoch: 91, Loss: 0.0009\nIter: 4, Epoch: 91, Loss: 0.0010\nIter: 5, Epoch: 91, Loss: 0.0007\nIter: 6, Epoch: 91, Loss: 0.0009\nIter: 7, Epoch: 91, Loss: 0.0008\nIter: 8, Epoch: 91, Loss: 0.0016\nIter: 9, Epoch: 91, Loss: 0.0011\nIter: 10, Epoch: 91, Loss: 0.0010\nIter: 11, Epoch: 91, Loss: 0.0006\nIter: 12, Epoch: 91, Loss: 0.0007\nIter: 13, Epoch: 91, Loss: 0.0010\nIter: 14, Epoch: 91, Loss: 0.0008\nIter: 15, Epoch: 91, Loss: 0.0007\nIter: 16, Epoch: 91, Loss: 0.0008\nIter: 17, Epoch: 91, Loss: 0.0010\nIter: 18, Epoch: 91, Loss: 0.0009\nIter: 19, Epoch: 91, Loss: 0.0015\nIter: 20, Epoch: 91, Loss: 0.0008\nIter: 21, Epoch: 91, Loss: 0.0013\nIter: 22, Epoch: 91, Loss: 0.0011\nIter: 23, Epoch: 91, Loss: 0.0015\nIter: 24, Epoch: 91, Loss: 0.0009\nIter: 25, Epoch: 91, Loss: 0.0007\nIter: 26, Epoch: 91, Loss: 0.0010\nIter: 27, Epoch: 91, Loss: 0.0010\nIter: 28, Epoch: 91, Loss: 0.0011\nIter: 29, Epoch: 91, Loss: 0.0014\nIter: 30, Epoch: 91, Loss: 0.0007\n","truncated":false}} +%--- +%[output:61ea05f6] +% data: {"dataType":"text","outputData":{"text":">> Epoch 91 validation loss: 0.0019\n","truncated":false}} +%--- +%[output:56361963] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 92, Loss: 0.0026\nIter: 2, Epoch: 92, Loss: 0.0009\nIter: 3, Epoch: 92, Loss: 0.0010\nIter: 4, Epoch: 92, Loss: 0.0009\nIter: 5, Epoch: 92, Loss: 0.0007\nIter: 6, Epoch: 92, Loss: 0.0010\nIter: 7, Epoch: 92, Loss: 0.0009\nIter: 8, Epoch: 92, Loss: 0.0014\nIter: 9, Epoch: 92, Loss: 0.0011\nIter: 10, Epoch: 92, Loss: 0.0011\nIter: 11, Epoch: 92, Loss: 0.0006\nIter: 12, Epoch: 92, Loss: 0.0006\nIter: 13, Epoch: 92, Loss: 0.0010\nIter: 14, Epoch: 92, Loss: 0.0009\nIter: 15, Epoch: 92, Loss: 0.0007\nIter: 16, Epoch: 92, Loss: 0.0008\nIter: 17, Epoch: 92, Loss: 0.0011\nIter: 18, Epoch: 92, Loss: 0.0010\nIter: 19, Epoch: 92, Loss: 0.0013\nIter: 20, Epoch: 92, Loss: 0.0008\nIter: 21, Epoch: 92, Loss: 0.0013\nIter: 22, Epoch: 92, Loss: 0.0011\nIter: 23, Epoch: 92, Loss: 0.0014\nIter: 24, Epoch: 92, Loss: 0.0008\nIter: 25, Epoch: 92, Loss: 0.0007\nIter: 26, Epoch: 92, Loss: 0.0010\nIter: 27, Epoch: 92, Loss: 0.0009\nIter: 28, Epoch: 92, Loss: 0.0011\nIter: 29, Epoch: 92, Loss: 0.0015\nIter: 30, Epoch: 92, Loss: 0.0007\n","truncated":false}} +%--- +%[output:045b74af] +% data: {"dataType":"text","outputData":{"text":">> Epoch 92 validation loss: 0.0018\n","truncated":false}} +%--- +%[output:35f19933] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 93, Loss: 0.0025\nIter: 2, Epoch: 93, Loss: 0.0008\nIter: 3, Epoch: 93, Loss: 0.0011\nIter: 4, Epoch: 93, Loss: 0.0009\nIter: 5, Epoch: 93, Loss: 0.0007\nIter: 6, Epoch: 93, Loss: 0.0010\nIter: 7, Epoch: 93, Loss: 0.0010\nIter: 8, Epoch: 93, Loss: 0.0014\nIter: 9, Epoch: 93, Loss: 0.0012\nIter: 10, Epoch: 93, Loss: 0.0012\nIter: 11, Epoch: 93, Loss: 0.0006\nIter: 12, Epoch: 93, Loss: 0.0006\nIter: 13, Epoch: 93, Loss: 0.0009\nIter: 14, Epoch: 93, Loss: 0.0009\nIter: 15, Epoch: 93, Loss: 0.0007\nIter: 16, Epoch: 93, Loss: 0.0007\nIter: 17, Epoch: 93, Loss: 0.0010\nIter: 18, Epoch: 93, Loss: 0.0010\nIter: 19, Epoch: 93, Loss: 0.0012\nIter: 20, Epoch: 93, Loss: 0.0007\nIter: 21, Epoch: 93, Loss: 0.0014\nIter: 22, Epoch: 93, Loss: 0.0011\nIter: 23, Epoch: 93, Loss: 0.0013\nIter: 24, Epoch: 93, Loss: 0.0008\nIter: 25, Epoch: 93, Loss: 0.0007\nIter: 26, Epoch: 93, Loss: 0.0010\nIter: 27, Epoch: 93, Loss: 0.0009\nIter: 28, Epoch: 93, Loss: 0.0010\nIter: 29, Epoch: 93, Loss: 0.0014\nIter: 30, Epoch: 93, Loss: 0.0007\n","truncated":false}} +%--- +%[output:015a1954] +% data: {"dataType":"text","outputData":{"text":">> Epoch 93 validation loss: 0.0017\n","truncated":false}} +%--- +%[output:7f01428e] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 94, Loss: 0.0023\nIter: 2, Epoch: 94, Loss: 0.0008\nIter: 3, Epoch: 94, Loss: 0.0010\nIter: 4, Epoch: 94, Loss: 0.0009\nIter: 5, Epoch: 94, Loss: 0.0006\nIter: 6, Epoch: 94, Loss: 0.0009\nIter: 7, Epoch: 94, Loss: 0.0009\nIter: 8, Epoch: 94, Loss: 0.0014\nIter: 9, Epoch: 94, Loss: 0.0011\nIter: 10, Epoch: 94, Loss: 0.0011\nIter: 11, Epoch: 94, Loss: 0.0006\nIter: 12, Epoch: 94, Loss: 0.0006\nIter: 13, Epoch: 94, Loss: 0.0008\nIter: 14, Epoch: 94, Loss: 0.0010\nIter: 15, Epoch: 94, Loss: 0.0007\nIter: 16, Epoch: 94, Loss: 0.0007\nIter: 17, Epoch: 94, Loss: 0.0010\nIter: 18, Epoch: 94, Loss: 0.0010\nIter: 19, Epoch: 94, Loss: 0.0013\nIter: 20, Epoch: 94, Loss: 0.0007\nIter: 21, Epoch: 94, Loss: 0.0014\nIter: 22, Epoch: 94, Loss: 0.0012\nIter: 23, Epoch: 94, Loss: 0.0012\nIter: 24, Epoch: 94, Loss: 0.0008\nIter: 25, Epoch: 94, Loss: 0.0007\nIter: 26, Epoch: 94, Loss: 0.0009\nIter: 27, Epoch: 94, Loss: 0.0009\nIter: 28, Epoch: 94, Loss: 0.0010\nIter: 29, Epoch: 94, Loss: 0.0014\nIter: 30, Epoch: 94, Loss: 0.0008\n","truncated":false}} +%--- +%[output:73be09fa] +% data: {"dataType":"text","outputData":{"text":">> Epoch 94 validation loss: 0.0016\n","truncated":false}} +%--- +%[output:0fa9183e] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 95, Loss: 0.0022\nIter: 2, Epoch: 95, Loss: 0.0007\nIter: 3, Epoch: 95, Loss: 0.0011\nIter: 4, Epoch: 95, Loss: 0.0009\nIter: 5, Epoch: 95, Loss: 0.0006\nIter: 6, Epoch: 95, Loss: 0.0009\nIter: 7, Epoch: 95, Loss: 0.0009\nIter: 8, Epoch: 95, Loss: 0.0014\nIter: 9, Epoch: 95, Loss: 0.0011\nIter: 10, Epoch: 95, Loss: 0.0011\nIter: 11, Epoch: 95, Loss: 0.0007\nIter: 12, Epoch: 95, Loss: 0.0006\nIter: 13, Epoch: 95, Loss: 0.0008\nIter: 14, Epoch: 95, Loss: 0.0010\nIter: 15, Epoch: 95, Loss: 0.0007\nIter: 16, Epoch: 95, Loss: 0.0006\nIter: 17, Epoch: 95, Loss: 0.0009\nIter: 18, Epoch: 95, Loss: 0.0010\nIter: 19, Epoch: 95, Loss: 0.0013\nIter: 20, Epoch: 95, Loss: 0.0006\nIter: 21, Epoch: 95, Loss: 0.0014\nIter: 22, Epoch: 95, Loss: 0.0012\nIter: 23, Epoch: 95, Loss: 0.0011\nIter: 24, Epoch: 95, Loss: 0.0007\nIter: 25, Epoch: 95, Loss: 0.0007\nIter: 26, Epoch: 95, Loss: 0.0009\nIter: 27, Epoch: 95, Loss: 0.0009\nIter: 28, Epoch: 95, Loss: 0.0009\nIter: 29, Epoch: 95, Loss: 0.0013\nIter: 30, Epoch: 95, Loss: 0.0007\n","truncated":false}} +%--- +%[output:69da0b2e] +% data: {"dataType":"text","outputData":{"text":">> Epoch 95 validation loss: 0.0015\n","truncated":false}} +%--- +%[output:0bf73890] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 96, Loss: 0.0021\nIter: 2, Epoch: 96, Loss: 0.0007\nIter: 3, Epoch: 96, Loss: 0.0010\nIter: 4, Epoch: 96, Loss: 0.0009\nIter: 5, Epoch: 96, Loss: 0.0006\nIter: 6, Epoch: 96, Loss: 0.0008\nIter: 7, Epoch: 96, Loss: 0.0009\nIter: 8, Epoch: 96, Loss: 0.0014\nIter: 9, Epoch: 96, Loss: 0.0010\nIter: 10, Epoch: 96, Loss: 0.0010\nIter: 11, Epoch: 96, Loss: 0.0006\nIter: 12, Epoch: 96, Loss: 0.0006\nIter: 13, Epoch: 96, Loss: 0.0007\nIter: 14, Epoch: 96, Loss: 0.0009\nIter: 15, Epoch: 96, Loss: 0.0008\nIter: 16, Epoch: 96, Loss: 0.0007\nIter: 17, Epoch: 96, Loss: 0.0009\nIter: 18, Epoch: 96, Loss: 0.0010\nIter: 19, Epoch: 96, Loss: 0.0015\nIter: 20, Epoch: 96, Loss: 0.0006\nIter: 21, Epoch: 96, Loss: 0.0014\nIter: 22, Epoch: 96, Loss: 0.0012\nIter: 23, Epoch: 96, Loss: 0.0012\nIter: 24, Epoch: 96, Loss: 0.0007\nIter: 25, Epoch: 96, Loss: 0.0007\nIter: 26, Epoch: 96, Loss: 0.0009\nIter: 27, Epoch: 96, Loss: 0.0009\nIter: 28, Epoch: 96, Loss: 0.0009\nIter: 29, Epoch: 96, Loss: 0.0013\nIter: 30, Epoch: 96, Loss: 0.0008\n","truncated":false}} +%--- +%[output:38181c2e] +% data: {"dataType":"text","outputData":{"text":">> Epoch 96 validation loss: 0.0015\n","truncated":false}} +%--- +%[output:58d76c03] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 97, Loss: 0.0022\nIter: 2, Epoch: 97, Loss: 0.0007\nIter: 3, Epoch: 97, Loss: 0.0010\nIter: 4, Epoch: 97, Loss: 0.0009\nIter: 5, Epoch: 97, Loss: 0.0006\nIter: 6, Epoch: 97, Loss: 0.0008\nIter: 7, Epoch: 97, Loss: 0.0009\nIter: 8, Epoch: 97, Loss: 0.0014\nIter: 9, Epoch: 97, Loss: 0.0010\nIter: 10, Epoch: 97, Loss: 0.0010\nIter: 11, Epoch: 97, Loss: 0.0006\nIter: 12, Epoch: 97, Loss: 0.0006\nIter: 13, Epoch: 97, Loss: 0.0007\nIter: 14, Epoch: 97, Loss: 0.0009\nIter: 15, Epoch: 97, Loss: 0.0008\nIter: 16, Epoch: 97, Loss: 0.0007\nIter: 17, Epoch: 97, Loss: 0.0009\nIter: 18, Epoch: 97, Loss: 0.0010\nIter: 19, Epoch: 97, Loss: 0.0015\nIter: 20, Epoch: 97, Loss: 0.0006\nIter: 21, Epoch: 97, Loss: 0.0013\nIter: 22, Epoch: 97, Loss: 0.0012\nIter: 23, Epoch: 97, Loss: 0.0012\nIter: 24, Epoch: 97, Loss: 0.0007\nIter: 25, Epoch: 97, Loss: 0.0007\nIter: 26, Epoch: 97, Loss: 0.0009\nIter: 27, Epoch: 97, Loss: 0.0009\nIter: 28, Epoch: 97, Loss: 0.0008\nIter: 29, Epoch: 97, Loss: 0.0012\nIter: 30, Epoch: 97, Loss: 0.0008\n","truncated":false}} +%--- +%[output:390c9784] +% data: {"dataType":"text","outputData":{"text":">> Epoch 97 validation loss: 0.0015\n","truncated":false}} +%--- +%[output:63fe1cbf] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 98, Loss: 0.0021\nIter: 2, Epoch: 98, Loss: 0.0007\nIter: 3, Epoch: 98, Loss: 0.0009\nIter: 4, Epoch: 98, Loss: 0.0009\nIter: 5, Epoch: 98, Loss: 0.0006\nIter: 6, Epoch: 98, Loss: 0.0008\nIter: 7, Epoch: 98, Loss: 0.0009\nIter: 8, Epoch: 98, Loss: 0.0015\nIter: 9, Epoch: 98, Loss: 0.0010\nIter: 10, Epoch: 98, Loss: 0.0009\nIter: 11, Epoch: 98, Loss: 0.0006\nIter: 12, Epoch: 98, Loss: 0.0006\nIter: 13, Epoch: 98, Loss: 0.0007\nIter: 14, Epoch: 98, Loss: 0.0009\nIter: 15, Epoch: 98, Loss: 0.0008\nIter: 16, Epoch: 98, Loss: 0.0007\nIter: 17, Epoch: 98, Loss: 0.0008\nIter: 18, Epoch: 98, Loss: 0.0010\nIter: 19, Epoch: 98, Loss: 0.0015\nIter: 20, Epoch: 98, Loss: 0.0006\nIter: 21, Epoch: 98, Loss: 0.0013\nIter: 22, Epoch: 98, Loss: 0.0012\nIter: 23, Epoch: 98, Loss: 0.0011\nIter: 24, Epoch: 98, Loss: 0.0006\nIter: 25, Epoch: 98, Loss: 0.0007\nIter: 26, Epoch: 98, Loss: 0.0009\nIter: 27, Epoch: 98, Loss: 0.0009\nIter: 28, Epoch: 98, Loss: 0.0008\nIter: 29, Epoch: 98, Loss: 0.0012\nIter: 30, Epoch: 98, Loss: 0.0007\n","truncated":false}} +%--- +%[output:1ce2cef7] +% data: {"dataType":"text","outputData":{"text":">> Epoch 98 validation loss: 0.0015\n","truncated":false}} +%--- +%[output:8bae86ce] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 99, Loss: 0.0021\nIter: 2, Epoch: 99, Loss: 0.0007\nIter: 3, Epoch: 99, Loss: 0.0009\nIter: 4, Epoch: 99, Loss: 0.0009\nIter: 5, Epoch: 99, Loss: 0.0006\nIter: 6, Epoch: 99, Loss: 0.0008\nIter: 7, Epoch: 99, Loss: 0.0009\nIter: 8, Epoch: 99, Loss: 0.0015\nIter: 9, Epoch: 99, Loss: 0.0010\nIter: 10, Epoch: 99, Loss: 0.0009\nIter: 11, Epoch: 99, Loss: 0.0006\nIter: 12, Epoch: 99, Loss: 0.0006\nIter: 13, Epoch: 99, Loss: 0.0007\nIter: 14, Epoch: 99, Loss: 0.0009\nIter: 15, Epoch: 99, Loss: 0.0008\nIter: 16, Epoch: 99, Loss: 0.0007\nIter: 17, Epoch: 99, Loss: 0.0008\nIter: 18, Epoch: 99, Loss: 0.0010\nIter: 19, Epoch: 99, Loss: 0.0015\nIter: 20, Epoch: 99, Loss: 0.0006\nIter: 21, Epoch: 99, Loss: 0.0013\nIter: 22, Epoch: 99, Loss: 0.0011\nIter: 23, Epoch: 99, Loss: 0.0011\nIter: 24, Epoch: 99, Loss: 0.0007\nIter: 25, Epoch: 99, Loss: 0.0007\nIter: 26, Epoch: 99, Loss: 0.0009\nIter: 27, Epoch: 99, Loss: 0.0009\nIter: 28, Epoch: 99, Loss: 0.0008\nIter: 29, Epoch: 99, Loss: 0.0012\nIter: 30, Epoch: 99, Loss: 0.0007\n","truncated":false}} +%--- +%[output:683036bc] +% data: {"dataType":"text","outputData":{"text":">> Epoch 99 validation loss: 0.0016\n","truncated":false}} +%--- +%[output:25455714] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 100, Loss: 0.0022\nIter: 2, Epoch: 100, Loss: 0.0007\nIter: 3, Epoch: 100, Loss: 0.0009\nIter: 4, Epoch: 100, Loss: 0.0009\nIter: 5, Epoch: 100, Loss: 0.0006\nIter: 6, Epoch: 100, Loss: 0.0008\nIter: 7, Epoch: 100, Loss: 0.0008\nIter: 8, Epoch: 100, Loss: 0.0015\nIter: 9, Epoch: 100, Loss: 0.0010\nIter: 10, Epoch: 100, Loss: 0.0009\nIter: 11, Epoch: 100, Loss: 0.0006\nIter: 12, Epoch: 100, Loss: 0.0006\nIter: 13, Epoch: 100, Loss: 0.0007\nIter: 14, Epoch: 100, Loss: 0.0008\nIter: 15, Epoch: 100, Loss: 0.0008\nIter: 16, Epoch: 100, Loss: 0.0007\nIter: 17, Epoch: 100, Loss: 0.0008\nIter: 18, Epoch: 100, Loss: 0.0009\nIter: 19, Epoch: 100, Loss: 0.0015\nIter: 20, Epoch: 100, Loss: 0.0006\nIter: 21, Epoch: 100, Loss: 0.0012\nIter: 22, Epoch: 100, Loss: 0.0011\nIter: 23, Epoch: 100, Loss: 0.0012\nIter: 24, Epoch: 100, Loss: 0.0006\nIter: 25, Epoch: 100, Loss: 0.0006\nIter: 26, Epoch: 100, Loss: 0.0009\nIter: 27, Epoch: 100, Loss: 0.0009\nIter: 28, Epoch: 100, Loss: 0.0008\nIter: 29, Epoch: 100, Loss: 0.0012\nIter: 30, Epoch: 100, Loss: 0.0007\n","truncated":false}} +%--- +%[output:421760ee] +% data: {"dataType":"text","outputData":{"text":">> Epoch 100 validation loss: 0.0016\n","truncated":false}} +%--- +%[output:51438166] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 101, Loss: 0.0022\nIter: 2, Epoch: 101, Loss: 0.0007\nIter: 3, Epoch: 101, Loss: 0.0009\nIter: 4, Epoch: 101, Loss: 0.0009\nIter: 5, Epoch: 101, Loss: 0.0006\nIter: 6, Epoch: 101, Loss: 0.0008\nIter: 7, Epoch: 101, Loss: 0.0009\nIter: 8, Epoch: 101, Loss: 0.0015\nIter: 9, Epoch: 101, Loss: 0.0010\nIter: 10, Epoch: 101, Loss: 0.0009\nIter: 11, Epoch: 101, Loss: 0.0006\nIter: 12, Epoch: 101, Loss: 0.0006\nIter: 13, Epoch: 101, Loss: 0.0008\nIter: 14, Epoch: 101, Loss: 0.0008\nIter: 15, Epoch: 101, Loss: 0.0008\nIter: 16, Epoch: 101, Loss: 0.0007\nIter: 17, Epoch: 101, Loss: 0.0008\nIter: 18, Epoch: 101, Loss: 0.0009\nIter: 19, Epoch: 101, Loss: 0.0015\nIter: 20, Epoch: 101, Loss: 0.0006\nIter: 21, Epoch: 101, Loss: 0.0012\nIter: 22, Epoch: 101, Loss: 0.0011\nIter: 23, Epoch: 101, Loss: 0.0011\nIter: 24, Epoch: 101, Loss: 0.0006\nIter: 25, Epoch: 101, Loss: 0.0006\nIter: 26, Epoch: 101, Loss: 0.0008\nIter: 27, Epoch: 101, Loss: 0.0009\nIter: 28, Epoch: 101, Loss: 0.0008\nIter: 29, Epoch: 101, Loss: 0.0012\nIter: 30, Epoch: 101, Loss: 0.0007\n","truncated":false}} +%--- +%[output:7ad504df] +% data: {"dataType":"text","outputData":{"text":">> Epoch 101 validation loss: 0.0016\n","truncated":false}} +%--- +%[output:09c8c0a0] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 102, Loss: 0.0022\nIter: 2, Epoch: 102, Loss: 0.0007\nIter: 3, Epoch: 102, Loss: 0.0009\nIter: 4, Epoch: 102, Loss: 0.0009\nIter: 5, Epoch: 102, Loss: 0.0006\nIter: 6, Epoch: 102, Loss: 0.0008\nIter: 7, Epoch: 102, Loss: 0.0009\nIter: 8, Epoch: 102, Loss: 0.0015\nIter: 9, Epoch: 102, Loss: 0.0010\nIter: 10, Epoch: 102, Loss: 0.0009\nIter: 11, Epoch: 102, Loss: 0.0006\nIter: 12, Epoch: 102, Loss: 0.0006\nIter: 13, Epoch: 102, Loss: 0.0007\nIter: 14, Epoch: 102, Loss: 0.0008\nIter: 15, Epoch: 102, Loss: 0.0008\nIter: 16, Epoch: 102, Loss: 0.0007\nIter: 17, Epoch: 102, Loss: 0.0008\nIter: 18, Epoch: 102, Loss: 0.0009\nIter: 19, Epoch: 102, Loss: 0.0015\nIter: 20, Epoch: 102, Loss: 0.0006\nIter: 21, Epoch: 102, Loss: 0.0012\nIter: 22, Epoch: 102, Loss: 0.0011\nIter: 23, Epoch: 102, Loss: 0.0011\nIter: 24, Epoch: 102, Loss: 0.0006\nIter: 25, Epoch: 102, Loss: 0.0006\nIter: 26, Epoch: 102, Loss: 0.0008\nIter: 27, Epoch: 102, Loss: 0.0008\nIter: 28, Epoch: 102, Loss: 0.0008\nIter: 29, Epoch: 102, Loss: 0.0012\nIter: 30, Epoch: 102, Loss: 0.0007\n","truncated":false}} +%--- +%[output:3d73f615] +% data: {"dataType":"text","outputData":{"text":">> Epoch 102 validation loss: 0.0015\n","truncated":false}} +%--- +%[output:76141baf] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 103, Loss: 0.0022\nIter: 2, Epoch: 103, Loss: 0.0007\nIter: 3, Epoch: 103, Loss: 0.0009\nIter: 4, Epoch: 103, Loss: 0.0009\nIter: 5, Epoch: 103, Loss: 0.0006\nIter: 6, Epoch: 103, Loss: 0.0008\nIter: 7, Epoch: 103, Loss: 0.0009\nIter: 8, Epoch: 103, Loss: 0.0015\nIter: 9, Epoch: 103, Loss: 0.0010\nIter: 10, Epoch: 103, Loss: 0.0009\nIter: 11, Epoch: 103, Loss: 0.0006\nIter: 12, Epoch: 103, Loss: 0.0006\nIter: 13, Epoch: 103, Loss: 0.0007\nIter: 14, Epoch: 103, Loss: 0.0008\nIter: 15, Epoch: 103, Loss: 0.0009\nIter: 16, Epoch: 103, Loss: 0.0007\nIter: 17, Epoch: 103, Loss: 0.0008\nIter: 18, Epoch: 103, Loss: 0.0009\nIter: 19, Epoch: 103, Loss: 0.0016\nIter: 20, Epoch: 103, Loss: 0.0006\nIter: 21, Epoch: 103, Loss: 0.0012\nIter: 22, Epoch: 103, Loss: 0.0011\nIter: 23, Epoch: 103, Loss: 0.0011\nIter: 24, Epoch: 103, Loss: 0.0006\nIter: 25, Epoch: 103, Loss: 0.0006\nIter: 26, Epoch: 103, Loss: 0.0008\nIter: 27, Epoch: 103, Loss: 0.0008\nIter: 28, Epoch: 103, Loss: 0.0008\nIter: 29, Epoch: 103, Loss: 0.0012\nIter: 30, Epoch: 103, Loss: 0.0007\n","truncated":false}} +%--- +%[output:262fd194] +% data: {"dataType":"text","outputData":{"text":">> Epoch 103 validation loss: 0.0015\n","truncated":false}} +%--- +%[output:4ecabd20] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 104, Loss: 0.0022\nIter: 2, Epoch: 104, Loss: 0.0007\nIter: 3, Epoch: 104, Loss: 0.0009\nIter: 4, Epoch: 104, Loss: 0.0009\nIter: 5, Epoch: 104, Loss: 0.0006\nIter: 6, Epoch: 104, Loss: 0.0008\nIter: 7, Epoch: 104, Loss: 0.0009\nIter: 8, Epoch: 104, Loss: 0.0015\nIter: 9, Epoch: 104, Loss: 0.0010\nIter: 10, Epoch: 104, Loss: 0.0009\nIter: 11, Epoch: 104, Loss: 0.0006\nIter: 12, Epoch: 104, Loss: 0.0006\nIter: 13, Epoch: 104, Loss: 0.0008\nIter: 14, Epoch: 104, Loss: 0.0008\nIter: 15, Epoch: 104, Loss: 0.0009\nIter: 16, Epoch: 104, Loss: 0.0008\nIter: 17, Epoch: 104, Loss: 0.0008\nIter: 18, Epoch: 104, Loss: 0.0009\nIter: 19, Epoch: 104, Loss: 0.0017\nIter: 20, Epoch: 104, Loss: 0.0006\nIter: 21, Epoch: 104, Loss: 0.0012\nIter: 22, Epoch: 104, Loss: 0.0011\nIter: 23, Epoch: 104, Loss: 0.0011\nIter: 24, Epoch: 104, Loss: 0.0006\nIter: 25, Epoch: 104, Loss: 0.0006\nIter: 26, Epoch: 104, Loss: 0.0008\nIter: 27, Epoch: 104, Loss: 0.0008\nIter: 28, Epoch: 104, Loss: 0.0008\nIter: 29, Epoch: 104, Loss: 0.0012\nIter: 30, Epoch: 104, Loss: 0.0007\n","truncated":false}} +%--- +%[output:4a763068] +% data: {"dataType":"text","outputData":{"text":">> Epoch 104 validation loss: 0.0016\n","truncated":false}} +%--- +%[output:37ba81ed] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 105, Loss: 0.0023\nIter: 2, Epoch: 105, Loss: 0.0007\nIter: 3, Epoch: 105, Loss: 0.0009\nIter: 4, Epoch: 105, Loss: 0.0010\nIter: 5, Epoch: 105, Loss: 0.0006\nIter: 6, Epoch: 105, Loss: 0.0007\nIter: 7, Epoch: 105, Loss: 0.0009\nIter: 8, Epoch: 105, Loss: 0.0015\nIter: 9, Epoch: 105, Loss: 0.0009\nIter: 10, Epoch: 105, Loss: 0.0009\nIter: 11, Epoch: 105, Loss: 0.0006\nIter: 12, Epoch: 105, Loss: 0.0007\nIter: 13, Epoch: 105, Loss: 0.0008\nIter: 14, Epoch: 105, Loss: 0.0008\nIter: 15, Epoch: 105, Loss: 0.0009\nIter: 16, Epoch: 105, Loss: 0.0008\nIter: 17, Epoch: 105, Loss: 0.0008\nIter: 18, Epoch: 105, Loss: 0.0009\nIter: 19, Epoch: 105, Loss: 0.0017\nIter: 20, Epoch: 105, Loss: 0.0006\nIter: 21, Epoch: 105, Loss: 0.0011\nIter: 22, Epoch: 105, Loss: 0.0011\nIter: 23, Epoch: 105, Loss: 0.0011\nIter: 24, Epoch: 105, Loss: 0.0006\nIter: 25, Epoch: 105, Loss: 0.0006\nIter: 26, Epoch: 105, Loss: 0.0008\nIter: 27, Epoch: 105, Loss: 0.0009\nIter: 28, Epoch: 105, Loss: 0.0008\nIter: 29, Epoch: 105, Loss: 0.0012\nIter: 30, Epoch: 105, Loss: 0.0007\n","truncated":false}} +%--- +%[output:22fad00a] +% data: {"dataType":"text","outputData":{"text":">> Epoch 105 validation loss: 0.0016\n","truncated":false}} +%--- +%[output:296220cb] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 106, Loss: 0.0023\nIter: 2, Epoch: 106, Loss: 0.0007\nIter: 3, Epoch: 106, Loss: 0.0009\nIter: 4, Epoch: 106, Loss: 0.0010\nIter: 5, Epoch: 106, Loss: 0.0006\nIter: 6, Epoch: 106, Loss: 0.0007\nIter: 7, Epoch: 106, Loss: 0.0009\nIter: 8, Epoch: 106, Loss: 0.0015\nIter: 9, Epoch: 106, Loss: 0.0009\nIter: 10, Epoch: 106, Loss: 0.0009\nIter: 11, Epoch: 106, Loss: 0.0006\nIter: 12, Epoch: 106, Loss: 0.0007\nIter: 13, Epoch: 106, Loss: 0.0008\nIter: 14, Epoch: 106, Loss: 0.0008\nIter: 15, Epoch: 106, Loss: 0.0009\nIter: 16, Epoch: 106, Loss: 0.0008\nIter: 17, Epoch: 106, Loss: 0.0008\nIter: 18, Epoch: 106, Loss: 0.0009\nIter: 19, Epoch: 106, Loss: 0.0017\nIter: 20, Epoch: 106, Loss: 0.0006\nIter: 21, Epoch: 106, Loss: 0.0011\nIter: 22, Epoch: 106, Loss: 0.0011\nIter: 23, Epoch: 106, Loss: 0.0011\nIter: 24, Epoch: 106, Loss: 0.0006\nIter: 25, Epoch: 106, Loss: 0.0006\nIter: 26, Epoch: 106, Loss: 0.0008\nIter: 27, Epoch: 106, Loss: 0.0008\nIter: 28, Epoch: 106, Loss: 0.0009\nIter: 29, Epoch: 106, Loss: 0.0012\nIter: 30, Epoch: 106, Loss: 0.0007\n","truncated":false}} +%--- +%[output:428b823b] +% data: {"dataType":"text","outputData":{"text":">> Epoch 106 validation loss: 0.0017\n","truncated":false}} +%--- +%[output:85737333] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 107, Loss: 0.0024\nIter: 2, Epoch: 107, Loss: 0.0007\nIter: 3, Epoch: 107, Loss: 0.0009\nIter: 4, Epoch: 107, Loss: 0.0010\nIter: 5, Epoch: 107, Loss: 0.0006\nIter: 6, Epoch: 107, Loss: 0.0007\nIter: 7, Epoch: 107, Loss: 0.0009\nIter: 8, Epoch: 107, Loss: 0.0015\nIter: 9, Epoch: 107, Loss: 0.0010\nIter: 10, Epoch: 107, Loss: 0.0009\nIter: 11, Epoch: 107, Loss: 0.0006\nIter: 12, Epoch: 107, Loss: 0.0007\nIter: 13, Epoch: 107, Loss: 0.0008\nIter: 14, Epoch: 107, Loss: 0.0008\nIter: 15, Epoch: 107, Loss: 0.0009\nIter: 16, Epoch: 107, Loss: 0.0008\nIter: 17, Epoch: 107, Loss: 0.0008\nIter: 18, Epoch: 107, Loss: 0.0009\nIter: 19, Epoch: 107, Loss: 0.0017\nIter: 20, Epoch: 107, Loss: 0.0007\nIter: 21, Epoch: 107, Loss: 0.0011\nIter: 22, Epoch: 107, Loss: 0.0011\nIter: 23, Epoch: 107, Loss: 0.0011\nIter: 24, Epoch: 107, Loss: 0.0007\nIter: 25, Epoch: 107, Loss: 0.0006\nIter: 26, Epoch: 107, Loss: 0.0008\nIter: 27, Epoch: 107, Loss: 0.0008\nIter: 28, Epoch: 107, Loss: 0.0009\nIter: 29, Epoch: 107, Loss: 0.0012\nIter: 30, Epoch: 107, Loss: 0.0007\n","truncated":false}} +%--- +%[output:8f616aa8] +% data: {"dataType":"text","outputData":{"text":">> Epoch 107 validation loss: 0.0016\n","truncated":false}} +%--- +%[output:88775070] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 108, Loss: 0.0024\nIter: 2, Epoch: 108, Loss: 0.0007\nIter: 3, Epoch: 108, Loss: 0.0009\nIter: 4, Epoch: 108, Loss: 0.0010\nIter: 5, Epoch: 108, Loss: 0.0006\nIter: 6, Epoch: 108, Loss: 0.0008\nIter: 7, Epoch: 108, Loss: 0.0009\nIter: 8, Epoch: 108, Loss: 0.0015\nIter: 9, Epoch: 108, Loss: 0.0010\nIter: 10, Epoch: 108, Loss: 0.0009\nIter: 11, Epoch: 108, Loss: 0.0006\nIter: 12, Epoch: 108, Loss: 0.0007\nIter: 13, Epoch: 108, Loss: 0.0008\nIter: 14, Epoch: 108, Loss: 0.0008\nIter: 15, Epoch: 108, Loss: 0.0009\nIter: 16, Epoch: 108, Loss: 0.0008\nIter: 17, Epoch: 108, Loss: 0.0008\nIter: 18, Epoch: 108, Loss: 0.0009\nIter: 19, Epoch: 108, Loss: 0.0017\nIter: 20, Epoch: 108, Loss: 0.0007\nIter: 21, Epoch: 108, Loss: 0.0011\nIter: 22, Epoch: 108, Loss: 0.0011\nIter: 23, Epoch: 108, Loss: 0.0011\nIter: 24, Epoch: 108, Loss: 0.0007\nIter: 25, Epoch: 108, Loss: 0.0006\nIter: 26, Epoch: 108, Loss: 0.0008\nIter: 27, Epoch: 108, Loss: 0.0008\nIter: 28, Epoch: 108, Loss: 0.0009\nIter: 29, Epoch: 108, Loss: 0.0012\nIter: 30, Epoch: 108, Loss: 0.0007\n","truncated":false}} +%--- +%[output:6efa2340] +% data: {"dataType":"text","outputData":{"text":">> Epoch 108 validation loss: 0.0017\n","truncated":false}} +%--- +%[output:5dea9ab5] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 109, Loss: 0.0025\nIter: 2, Epoch: 109, Loss: 0.0007\nIter: 3, Epoch: 109, Loss: 0.0009\nIter: 4, Epoch: 109, Loss: 0.0010\nIter: 5, Epoch: 109, Loss: 0.0006\nIter: 6, Epoch: 109, Loss: 0.0007\nIter: 7, Epoch: 109, Loss: 0.0009\nIter: 8, Epoch: 109, Loss: 0.0015\nIter: 9, Epoch: 109, Loss: 0.0010\nIter: 10, Epoch: 109, Loss: 0.0009\nIter: 11, Epoch: 109, Loss: 0.0006\nIter: 12, Epoch: 109, Loss: 0.0007\nIter: 13, Epoch: 109, Loss: 0.0009\nIter: 14, Epoch: 109, Loss: 0.0008\nIter: 15, Epoch: 109, Loss: 0.0009\nIter: 16, Epoch: 109, Loss: 0.0008\nIter: 17, Epoch: 109, Loss: 0.0009\nIter: 18, Epoch: 109, Loss: 0.0008\nIter: 19, Epoch: 109, Loss: 0.0017\nIter: 20, Epoch: 109, Loss: 0.0007\nIter: 21, Epoch: 109, Loss: 0.0010\nIter: 22, Epoch: 109, Loss: 0.0011\nIter: 23, Epoch: 109, Loss: 0.0011\nIter: 24, Epoch: 109, Loss: 0.0007\nIter: 25, Epoch: 109, Loss: 0.0006\nIter: 26, Epoch: 109, Loss: 0.0008\nIter: 27, Epoch: 109, Loss: 0.0009\nIter: 28, Epoch: 109, Loss: 0.0010\nIter: 29, Epoch: 109, Loss: 0.0012\nIter: 30, Epoch: 109, Loss: 0.0007\n","truncated":false}} +%--- +%[output:6683307d] +% data: {"dataType":"text","outputData":{"text":">> Epoch 109 validation loss: 0.0017\n","truncated":false}} +%--- +%[output:5c7b048f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 110, Loss: 0.0025\nIter: 2, Epoch: 110, Loss: 0.0008\nIter: 3, Epoch: 110, Loss: 0.0009\nIter: 4, Epoch: 110, Loss: 0.0010\nIter: 5, Epoch: 110, Loss: 0.0007\nIter: 6, Epoch: 110, Loss: 0.0007\nIter: 7, Epoch: 110, Loss: 0.0009\nIter: 8, Epoch: 110, Loss: 0.0016\nIter: 9, Epoch: 110, Loss: 0.0010\nIter: 10, Epoch: 110, Loss: 0.0009\nIter: 11, Epoch: 110, Loss: 0.0006\nIter: 12, Epoch: 110, Loss: 0.0008\nIter: 13, Epoch: 110, Loss: 0.0009\nIter: 14, Epoch: 110, Loss: 0.0008\nIter: 15, Epoch: 110, Loss: 0.0009\nIter: 16, Epoch: 110, Loss: 0.0009\nIter: 17, Epoch: 110, Loss: 0.0009\nIter: 18, Epoch: 110, Loss: 0.0008\nIter: 19, Epoch: 110, Loss: 0.0017\nIter: 20, Epoch: 110, Loss: 0.0007\nIter: 21, Epoch: 110, Loss: 0.0010\nIter: 22, Epoch: 110, Loss: 0.0011\nIter: 23, Epoch: 110, Loss: 0.0011\nIter: 24, Epoch: 110, Loss: 0.0008\nIter: 25, Epoch: 110, Loss: 0.0006\nIter: 26, Epoch: 110, Loss: 0.0008\nIter: 27, Epoch: 110, Loss: 0.0009\nIter: 28, Epoch: 110, Loss: 0.0011\nIter: 29, Epoch: 110, Loss: 0.0012\nIter: 30, Epoch: 110, Loss: 0.0007\n","truncated":false}} +%--- +%[output:3a37c612] +% data: {"dataType":"text","outputData":{"text":">> Epoch 110 validation loss: 0.0017\n","truncated":false}} +%--- +%[output:7053153a] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 111, Loss: 0.0025\nIter: 2, Epoch: 111, Loss: 0.0008\nIter: 3, Epoch: 111, Loss: 0.0008\nIter: 4, Epoch: 111, Loss: 0.0011\nIter: 5, Epoch: 111, Loss: 0.0007\nIter: 6, Epoch: 111, Loss: 0.0008\nIter: 7, Epoch: 111, Loss: 0.0008\nIter: 8, Epoch: 111, Loss: 0.0016\nIter: 9, Epoch: 111, Loss: 0.0011\nIter: 10, Epoch: 111, Loss: 0.0010\nIter: 11, Epoch: 111, Loss: 0.0006\nIter: 12, Epoch: 111, Loss: 0.0008\nIter: 13, Epoch: 111, Loss: 0.0010\nIter: 14, Epoch: 111, Loss: 0.0008\nIter: 15, Epoch: 111, Loss: 0.0009\nIter: 16, Epoch: 111, Loss: 0.0009\nIter: 17, Epoch: 111, Loss: 0.0010\nIter: 18, Epoch: 111, Loss: 0.0008\nIter: 19, Epoch: 111, Loss: 0.0016\nIter: 20, Epoch: 111, Loss: 0.0008\nIter: 21, Epoch: 111, Loss: 0.0010\nIter: 22, Epoch: 111, Loss: 0.0011\nIter: 23, Epoch: 111, Loss: 0.0011\nIter: 24, Epoch: 111, Loss: 0.0008\nIter: 25, Epoch: 111, Loss: 0.0006\nIter: 26, Epoch: 111, Loss: 0.0008\nIter: 27, Epoch: 111, Loss: 0.0010\nIter: 28, Epoch: 111, Loss: 0.0011\nIter: 29, Epoch: 111, Loss: 0.0013\nIter: 30, Epoch: 111, Loss: 0.0007\n","truncated":false}} +%--- +%[output:37c3c5b3] +% data: {"dataType":"text","outputData":{"text":">> Epoch 111 validation loss: 0.0017\n","truncated":false}} +%--- +%[output:084c6bae] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 112, Loss: 0.0025\nIter: 2, Epoch: 112, Loss: 0.0008\nIter: 3, Epoch: 112, Loss: 0.0008\nIter: 4, Epoch: 112, Loss: 0.0010\nIter: 5, Epoch: 112, Loss: 0.0008\nIter: 6, Epoch: 112, Loss: 0.0008\nIter: 7, Epoch: 112, Loss: 0.0008\nIter: 8, Epoch: 112, Loss: 0.0015\nIter: 9, Epoch: 112, Loss: 0.0011\nIter: 10, Epoch: 112, Loss: 0.0010\nIter: 11, Epoch: 112, Loss: 0.0006\nIter: 12, Epoch: 112, Loss: 0.0007\nIter: 13, Epoch: 112, Loss: 0.0011\nIter: 14, Epoch: 112, Loss: 0.0008\nIter: 15, Epoch: 112, Loss: 0.0008\nIter: 16, Epoch: 112, Loss: 0.0009\nIter: 17, Epoch: 112, Loss: 0.0010\nIter: 18, Epoch: 112, Loss: 0.0008\nIter: 19, Epoch: 112, Loss: 0.0014\nIter: 20, Epoch: 112, Loss: 0.0008\nIter: 21, Epoch: 112, Loss: 0.0010\nIter: 22, Epoch: 112, Loss: 0.0011\nIter: 23, Epoch: 112, Loss: 0.0011\nIter: 24, Epoch: 112, Loss: 0.0009\nIter: 25, Epoch: 112, Loss: 0.0007\nIter: 26, Epoch: 112, Loss: 0.0008\nIter: 27, Epoch: 112, Loss: 0.0010\nIter: 28, Epoch: 112, Loss: 0.0012\nIter: 29, Epoch: 112, Loss: 0.0013\nIter: 30, Epoch: 112, Loss: 0.0007\n","truncated":false}} +%--- +%[output:123a8c2a] +% data: {"dataType":"text","outputData":{"text":">> Epoch 112 validation loss: 0.0017\n","truncated":false}} +%--- +%[output:6b80aade] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 113, Loss: 0.0025\nIter: 2, Epoch: 113, Loss: 0.0009\nIter: 3, Epoch: 113, Loss: 0.0008\nIter: 4, Epoch: 113, Loss: 0.0010\nIter: 5, Epoch: 113, Loss: 0.0008\nIter: 6, Epoch: 113, Loss: 0.0008\nIter: 7, Epoch: 113, Loss: 0.0008\nIter: 8, Epoch: 113, Loss: 0.0014\nIter: 9, Epoch: 113, Loss: 0.0012\nIter: 10, Epoch: 113, Loss: 0.0011\nIter: 11, Epoch: 113, Loss: 0.0005\nIter: 12, Epoch: 113, Loss: 0.0007\nIter: 13, Epoch: 113, Loss: 0.0011\nIter: 14, Epoch: 113, Loss: 0.0009\nIter: 15, Epoch: 113, Loss: 0.0008\nIter: 16, Epoch: 113, Loss: 0.0008\nIter: 17, Epoch: 113, Loss: 0.0010\nIter: 18, Epoch: 113, Loss: 0.0009\nIter: 19, Epoch: 113, Loss: 0.0013\nIter: 20, Epoch: 113, Loss: 0.0007\nIter: 21, Epoch: 113, Loss: 0.0010\nIter: 22, Epoch: 113, Loss: 0.0011\nIter: 23, Epoch: 113, Loss: 0.0011\nIter: 24, Epoch: 113, Loss: 0.0008\nIter: 25, Epoch: 113, Loss: 0.0008\nIter: 26, Epoch: 113, Loss: 0.0008\nIter: 27, Epoch: 113, Loss: 0.0010\nIter: 28, Epoch: 113, Loss: 0.0012\nIter: 29, Epoch: 113, Loss: 0.0014\nIter: 30, Epoch: 113, Loss: 0.0007\n","truncated":false}} +%--- +%[output:0c96e117] +% data: {"dataType":"text","outputData":{"text":">> Epoch 113 validation loss: 0.0015\n","truncated":false}} +%--- +%[output:920fbb93] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 114, Loss: 0.0023\nIter: 2, Epoch: 114, Loss: 0.0009\nIter: 3, Epoch: 114, Loss: 0.0008\nIter: 4, Epoch: 114, Loss: 0.0009\nIter: 5, Epoch: 114, Loss: 0.0007\nIter: 6, Epoch: 114, Loss: 0.0008\nIter: 7, Epoch: 114, Loss: 0.0008\nIter: 8, Epoch: 114, Loss: 0.0014\nIter: 9, Epoch: 114, Loss: 0.0011\nIter: 10, Epoch: 114, Loss: 0.0011\nIter: 11, Epoch: 114, Loss: 0.0005\nIter: 12, Epoch: 114, Loss: 0.0007\nIter: 13, Epoch: 114, Loss: 0.0011\nIter: 14, Epoch: 114, Loss: 0.0009\nIter: 15, Epoch: 114, Loss: 0.0007\nIter: 16, Epoch: 114, Loss: 0.0008\nIter: 17, Epoch: 114, Loss: 0.0010\nIter: 18, Epoch: 114, Loss: 0.0009\nIter: 19, Epoch: 114, Loss: 0.0012\nIter: 20, Epoch: 114, Loss: 0.0007\nIter: 21, Epoch: 114, Loss: 0.0010\nIter: 22, Epoch: 114, Loss: 0.0011\nIter: 23, Epoch: 114, Loss: 0.0011\nIter: 24, Epoch: 114, Loss: 0.0008\nIter: 25, Epoch: 114, Loss: 0.0008\nIter: 26, Epoch: 114, Loss: 0.0008\nIter: 27, Epoch: 114, Loss: 0.0010\nIter: 28, Epoch: 114, Loss: 0.0011\nIter: 29, Epoch: 114, Loss: 0.0014\nIter: 30, Epoch: 114, Loss: 0.0007\n","truncated":false}} +%--- +%[output:77384522] +% data: {"dataType":"text","outputData":{"text":">> Epoch 114 validation loss: 0.0015\n","truncated":false}} +%--- +%[output:30756905] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 115, Loss: 0.0022\nIter: 2, Epoch: 115, Loss: 0.0009\nIter: 3, Epoch: 115, Loss: 0.0008\nIter: 4, Epoch: 115, Loss: 0.0009\nIter: 5, Epoch: 115, Loss: 0.0007\nIter: 6, Epoch: 115, Loss: 0.0008\nIter: 7, Epoch: 115, Loss: 0.0008\nIter: 8, Epoch: 115, Loss: 0.0013\nIter: 9, Epoch: 115, Loss: 0.0011\nIter: 10, Epoch: 115, Loss: 0.0011\nIter: 11, Epoch: 115, Loss: 0.0005\nIter: 12, Epoch: 115, Loss: 0.0006\nIter: 13, Epoch: 115, Loss: 0.0011\nIter: 14, Epoch: 115, Loss: 0.0009\nIter: 15, Epoch: 115, Loss: 0.0007\nIter: 16, Epoch: 115, Loss: 0.0008\nIter: 17, Epoch: 115, Loss: 0.0010\nIter: 18, Epoch: 115, Loss: 0.0009\nIter: 19, Epoch: 115, Loss: 0.0011\nIter: 20, Epoch: 115, Loss: 0.0007\nIter: 21, Epoch: 115, Loss: 0.0010\nIter: 22, Epoch: 115, Loss: 0.0011\nIter: 23, Epoch: 115, Loss: 0.0011\nIter: 24, Epoch: 115, Loss: 0.0008\nIter: 25, Epoch: 115, Loss: 0.0008\nIter: 26, Epoch: 115, Loss: 0.0008\nIter: 27, Epoch: 115, Loss: 0.0009\nIter: 28, Epoch: 115, Loss: 0.0011\nIter: 29, Epoch: 115, Loss: 0.0014\nIter: 30, Epoch: 115, Loss: 0.0006\n","truncated":false}} +%--- +%[output:7a845ec5] +% data: {"dataType":"text","outputData":{"text":">> Epoch 115 validation loss: 0.0014\n","truncated":false}} +%--- +%[output:5f1a920e] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 116, Loss: 0.0021\nIter: 2, Epoch: 116, Loss: 0.0009\nIter: 3, Epoch: 116, Loss: 0.0008\nIter: 4, Epoch: 116, Loss: 0.0008\nIter: 5, Epoch: 116, Loss: 0.0006\nIter: 6, Epoch: 116, Loss: 0.0009\nIter: 7, Epoch: 116, Loss: 0.0008\nIter: 8, Epoch: 116, Loss: 0.0013\nIter: 9, Epoch: 116, Loss: 0.0011\nIter: 10, Epoch: 116, Loss: 0.0011\nIter: 11, Epoch: 116, Loss: 0.0005\nIter: 12, Epoch: 116, Loss: 0.0006\nIter: 13, Epoch: 116, Loss: 0.0010\nIter: 14, Epoch: 116, Loss: 0.0010\nIter: 15, Epoch: 116, Loss: 0.0006\nIter: 16, Epoch: 116, Loss: 0.0007\nIter: 17, Epoch: 116, Loss: 0.0010\nIter: 18, Epoch: 116, Loss: 0.0009\nIter: 19, Epoch: 116, Loss: 0.0010\nIter: 20, Epoch: 116, Loss: 0.0006\nIter: 21, Epoch: 116, Loss: 0.0010\nIter: 22, Epoch: 116, Loss: 0.0011\nIter: 23, Epoch: 116, Loss: 0.0011\nIter: 24, Epoch: 116, Loss: 0.0007\nIter: 25, Epoch: 116, Loss: 0.0008\nIter: 26, Epoch: 116, Loss: 0.0008\nIter: 27, Epoch: 116, Loss: 0.0009\nIter: 28, Epoch: 116, Loss: 0.0010\nIter: 29, Epoch: 116, Loss: 0.0013\nIter: 30, Epoch: 116, Loss: 0.0006\n","truncated":false}} +%--- +%[output:89e1eac0] +% data: {"dataType":"text","outputData":{"text":">> Epoch 116 validation loss: 0.0013\n","truncated":false}} +%--- +%[output:9b5cb80a] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 117, Loss: 0.0020\nIter: 2, Epoch: 117, Loss: 0.0008\nIter: 3, Epoch: 117, Loss: 0.0008\nIter: 4, Epoch: 117, Loss: 0.0008\nIter: 5, Epoch: 117, Loss: 0.0006\nIter: 6, Epoch: 117, Loss: 0.0008\nIter: 7, Epoch: 117, Loss: 0.0008\nIter: 8, Epoch: 117, Loss: 0.0013\nIter: 9, Epoch: 117, Loss: 0.0010\nIter: 10, Epoch: 117, Loss: 0.0010\nIter: 11, Epoch: 117, Loss: 0.0005\nIter: 12, Epoch: 117, Loss: 0.0006\nIter: 13, Epoch: 117, Loss: 0.0009\nIter: 14, Epoch: 117, Loss: 0.0009\nIter: 15, Epoch: 117, Loss: 0.0006\nIter: 16, Epoch: 117, Loss: 0.0007\nIter: 17, Epoch: 117, Loss: 0.0009\nIter: 18, Epoch: 117, Loss: 0.0009\nIter: 19, Epoch: 117, Loss: 0.0010\nIter: 20, Epoch: 117, Loss: 0.0006\nIter: 21, Epoch: 117, Loss: 0.0009\nIter: 22, Epoch: 117, Loss: 0.0011\nIter: 23, Epoch: 117, Loss: 0.0011\nIter: 24, Epoch: 117, Loss: 0.0007\nIter: 25, Epoch: 117, Loss: 0.0008\nIter: 26, Epoch: 117, Loss: 0.0008\nIter: 27, Epoch: 117, Loss: 0.0008\nIter: 28, Epoch: 117, Loss: 0.0010\nIter: 29, Epoch: 117, Loss: 0.0014\nIter: 30, Epoch: 117, Loss: 0.0006\n","truncated":false}} +%--- +%[output:27e1e622] +% data: {"dataType":"text","outputData":{"text":">> Epoch 117 validation loss: 0.0013\n","truncated":false}} +%--- +%[output:09bba6c0] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 118, Loss: 0.0020\nIter: 2, Epoch: 118, Loss: 0.0008\nIter: 3, Epoch: 118, Loss: 0.0008\nIter: 4, Epoch: 118, Loss: 0.0008\nIter: 5, Epoch: 118, Loss: 0.0006\nIter: 6, Epoch: 118, Loss: 0.0008\nIter: 7, Epoch: 118, Loss: 0.0008\nIter: 8, Epoch: 118, Loss: 0.0013\nIter: 9, Epoch: 118, Loss: 0.0010\nIter: 10, Epoch: 118, Loss: 0.0010\nIter: 11, Epoch: 118, Loss: 0.0005\nIter: 12, Epoch: 118, Loss: 0.0006\nIter: 13, Epoch: 118, Loss: 0.0009\nIter: 14, Epoch: 118, Loss: 0.0009\nIter: 15, Epoch: 118, Loss: 0.0006\nIter: 16, Epoch: 118, Loss: 0.0007\nIter: 17, Epoch: 118, Loss: 0.0009\nIter: 18, Epoch: 118, Loss: 0.0009\nIter: 19, Epoch: 118, Loss: 0.0010\nIter: 20, Epoch: 118, Loss: 0.0006\nIter: 21, Epoch: 118, Loss: 0.0009\nIter: 22, Epoch: 118, Loss: 0.0011\nIter: 23, Epoch: 118, Loss: 0.0011\nIter: 24, Epoch: 118, Loss: 0.0006\nIter: 25, Epoch: 118, Loss: 0.0007\nIter: 26, Epoch: 118, Loss: 0.0008\nIter: 27, Epoch: 118, Loss: 0.0008\nIter: 28, Epoch: 118, Loss: 0.0009\nIter: 29, Epoch: 118, Loss: 0.0013\nIter: 30, Epoch: 118, Loss: 0.0006\n","truncated":false}} +%--- +%[output:3d5deeff] +% data: {"dataType":"text","outputData":{"text":">> Epoch 118 validation loss: 0.0013\n","truncated":false}} +%--- +%[output:9284807a] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 119, Loss: 0.0020\nIter: 2, Epoch: 119, Loss: 0.0007\nIter: 3, Epoch: 119, Loss: 0.0008\nIter: 4, Epoch: 119, Loss: 0.0008\nIter: 5, Epoch: 119, Loss: 0.0005\nIter: 6, Epoch: 119, Loss: 0.0008\nIter: 7, Epoch: 119, Loss: 0.0008\nIter: 8, Epoch: 119, Loss: 0.0013\nIter: 9, Epoch: 119, Loss: 0.0009\nIter: 10, Epoch: 119, Loss: 0.0010\nIter: 11, Epoch: 119, Loss: 0.0005\nIter: 12, Epoch: 119, Loss: 0.0006\nIter: 13, Epoch: 119, Loss: 0.0008\nIter: 14, Epoch: 119, Loss: 0.0009\nIter: 15, Epoch: 119, Loss: 0.0006\nIter: 16, Epoch: 119, Loss: 0.0007\nIter: 17, Epoch: 119, Loss: 0.0008\nIter: 18, Epoch: 119, Loss: 0.0008\nIter: 19, Epoch: 119, Loss: 0.0011\nIter: 20, Epoch: 119, Loss: 0.0006\nIter: 21, Epoch: 119, Loss: 0.0009\nIter: 22, Epoch: 119, Loss: 0.0011\nIter: 23, Epoch: 119, Loss: 0.0012\nIter: 24, Epoch: 119, Loss: 0.0006\nIter: 25, Epoch: 119, Loss: 0.0007\nIter: 26, Epoch: 119, Loss: 0.0008\nIter: 27, Epoch: 119, Loss: 0.0008\nIter: 28, Epoch: 119, Loss: 0.0009\nIter: 29, Epoch: 119, Loss: 0.0012\nIter: 30, Epoch: 119, Loss: 0.0006\n","truncated":false}} +%--- +%[output:233f0a9e] +% data: {"dataType":"text","outputData":{"text":">> Epoch 119 validation loss: 0.0013\n","truncated":false}} +%--- +%[output:17999174] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 120, Loss: 0.0020\nIter: 2, Epoch: 120, Loss: 0.0007\nIter: 3, Epoch: 120, Loss: 0.0008\nIter: 4, Epoch: 120, Loss: 0.0008\nIter: 5, Epoch: 120, Loss: 0.0005\nIter: 6, Epoch: 120, Loss: 0.0008\nIter: 7, Epoch: 120, Loss: 0.0008\nIter: 8, Epoch: 120, Loss: 0.0013\nIter: 9, Epoch: 120, Loss: 0.0010\nIter: 10, Epoch: 120, Loss: 0.0010\nIter: 11, Epoch: 120, Loss: 0.0005\nIter: 12, Epoch: 120, Loss: 0.0006\nIter: 13, Epoch: 120, Loss: 0.0008\nIter: 14, Epoch: 120, Loss: 0.0009\nIter: 15, Epoch: 120, Loss: 0.0006\nIter: 16, Epoch: 120, Loss: 0.0007\nIter: 17, Epoch: 120, Loss: 0.0008\nIter: 18, Epoch: 120, Loss: 0.0008\nIter: 19, Epoch: 120, Loss: 0.0010\nIter: 20, Epoch: 120, Loss: 0.0006\nIter: 21, Epoch: 120, Loss: 0.0008\nIter: 22, Epoch: 120, Loss: 0.0010\nIter: 23, Epoch: 120, Loss: 0.0012\nIter: 24, Epoch: 120, Loss: 0.0006\nIter: 25, Epoch: 120, Loss: 0.0007\nIter: 26, Epoch: 120, Loss: 0.0008\nIter: 27, Epoch: 120, Loss: 0.0008\nIter: 28, Epoch: 120, Loss: 0.0008\nIter: 29, Epoch: 120, Loss: 0.0012\nIter: 30, Epoch: 120, Loss: 0.0006\n","truncated":false}} +%--- +%[output:31e769b7] +% data: {"dataType":"text","outputData":{"text":">> Epoch 120 validation loss: 0.0013\n","truncated":false}} +%--- +%[output:1e371406] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 121, Loss: 0.0020\nIter: 2, Epoch: 121, Loss: 0.0007\nIter: 3, Epoch: 121, Loss: 0.0008\nIter: 4, Epoch: 121, Loss: 0.0008\nIter: 5, Epoch: 121, Loss: 0.0005\nIter: 6, Epoch: 121, Loss: 0.0007\nIter: 7, Epoch: 121, Loss: 0.0008\nIter: 8, Epoch: 121, Loss: 0.0013\nIter: 9, Epoch: 121, Loss: 0.0009\nIter: 10, Epoch: 121, Loss: 0.0010\nIter: 11, Epoch: 121, Loss: 0.0006\nIter: 12, Epoch: 121, Loss: 0.0006\nIter: 13, Epoch: 121, Loss: 0.0008\nIter: 14, Epoch: 121, Loss: 0.0008\nIter: 15, Epoch: 121, Loss: 0.0006\nIter: 16, Epoch: 121, Loss: 0.0007\nIter: 17, Epoch: 121, Loss: 0.0008\nIter: 18, Epoch: 121, Loss: 0.0008\nIter: 19, Epoch: 121, Loss: 0.0011\nIter: 20, Epoch: 121, Loss: 0.0006\nIter: 21, Epoch: 121, Loss: 0.0008\nIter: 22, Epoch: 121, Loss: 0.0010\nIter: 23, Epoch: 121, Loss: 0.0012\nIter: 24, Epoch: 121, Loss: 0.0006\nIter: 25, Epoch: 121, Loss: 0.0007\nIter: 26, Epoch: 121, Loss: 0.0008\nIter: 27, Epoch: 121, Loss: 0.0008\nIter: 28, Epoch: 121, Loss: 0.0008\nIter: 29, Epoch: 121, Loss: 0.0012\nIter: 30, Epoch: 121, Loss: 0.0006\n","truncated":false}} +%--- +%[output:8f9fec08] +% data: {"dataType":"text","outputData":{"text":">> Epoch 121 validation loss: 0.0013\n","truncated":false}} +%--- +%[output:5444af52] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 122, Loss: 0.0020\nIter: 2, Epoch: 122, Loss: 0.0007\nIter: 3, Epoch: 122, Loss: 0.0008\nIter: 4, Epoch: 122, Loss: 0.0008\nIter: 5, Epoch: 122, Loss: 0.0005\nIter: 6, Epoch: 122, Loss: 0.0007\nIter: 7, Epoch: 122, Loss: 0.0008\nIter: 8, Epoch: 122, Loss: 0.0013\nIter: 9, Epoch: 122, Loss: 0.0009\nIter: 10, Epoch: 122, Loss: 0.0009\nIter: 11, Epoch: 122, Loss: 0.0005\nIter: 12, Epoch: 122, Loss: 0.0006\nIter: 13, Epoch: 122, Loss: 0.0008\nIter: 14, Epoch: 122, Loss: 0.0008\nIter: 15, Epoch: 122, Loss: 0.0006\nIter: 16, Epoch: 122, Loss: 0.0007\nIter: 17, Epoch: 122, Loss: 0.0008\nIter: 18, Epoch: 122, Loss: 0.0008\nIter: 19, Epoch: 122, Loss: 0.0010\nIter: 20, Epoch: 122, Loss: 0.0006\nIter: 21, Epoch: 122, Loss: 0.0008\nIter: 22, Epoch: 122, Loss: 0.0010\nIter: 23, Epoch: 122, Loss: 0.0013\nIter: 24, Epoch: 122, Loss: 0.0007\nIter: 25, Epoch: 122, Loss: 0.0007\nIter: 26, Epoch: 122, Loss: 0.0008\nIter: 27, Epoch: 122, Loss: 0.0008\nIter: 28, Epoch: 122, Loss: 0.0008\nIter: 29, Epoch: 122, Loss: 0.0013\nIter: 30, Epoch: 122, Loss: 0.0006\n","truncated":false}} +%--- +%[output:068740da] +% data: {"dataType":"text","outputData":{"text":">> Epoch 122 validation loss: 0.0013\n","truncated":false}} +%--- +%[output:4e07268e] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 123, Loss: 0.0020\nIter: 2, Epoch: 123, Loss: 0.0007\nIter: 3, Epoch: 123, Loss: 0.0008\nIter: 4, Epoch: 123, Loss: 0.0008\nIter: 5, Epoch: 123, Loss: 0.0006\nIter: 6, Epoch: 123, Loss: 0.0008\nIter: 7, Epoch: 123, Loss: 0.0008\nIter: 8, Epoch: 123, Loss: 0.0013\nIter: 9, Epoch: 123, Loss: 0.0009\nIter: 10, Epoch: 123, Loss: 0.0010\nIter: 11, Epoch: 123, Loss: 0.0005\nIter: 12, Epoch: 123, Loss: 0.0005\nIter: 13, Epoch: 123, Loss: 0.0009\nIter: 14, Epoch: 123, Loss: 0.0008\nIter: 15, Epoch: 123, Loss: 0.0006\nIter: 16, Epoch: 123, Loss: 0.0007\nIter: 17, Epoch: 123, Loss: 0.0008\nIter: 18, Epoch: 123, Loss: 0.0008\nIter: 19, Epoch: 123, Loss: 0.0010\nIter: 20, Epoch: 123, Loss: 0.0006\nIter: 21, Epoch: 123, Loss: 0.0008\nIter: 22, Epoch: 123, Loss: 0.0011\nIter: 23, Epoch: 123, Loss: 0.0013\nIter: 24, Epoch: 123, Loss: 0.0007\nIter: 25, Epoch: 123, Loss: 0.0007\nIter: 26, Epoch: 123, Loss: 0.0009\nIter: 27, Epoch: 123, Loss: 0.0008\nIter: 28, Epoch: 123, Loss: 0.0008\nIter: 29, Epoch: 123, Loss: 0.0013\nIter: 30, Epoch: 123, Loss: 0.0006\n","truncated":false}} +%--- +%[output:5548fd11] +% data: {"dataType":"text","outputData":{"text":">> Epoch 123 validation loss: 0.0013\n","truncated":false}} +%--- +%[output:3e8b13ca] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 124, Loss: 0.0020\nIter: 2, Epoch: 124, Loss: 0.0007\nIter: 3, Epoch: 124, Loss: 0.0008\nIter: 4, Epoch: 124, Loss: 0.0008\nIter: 5, Epoch: 124, Loss: 0.0006\nIter: 6, Epoch: 124, Loss: 0.0008\nIter: 7, Epoch: 124, Loss: 0.0008\nIter: 8, Epoch: 124, Loss: 0.0013\nIter: 9, Epoch: 124, Loss: 0.0009\nIter: 10, Epoch: 124, Loss: 0.0010\nIter: 11, Epoch: 124, Loss: 0.0005\nIter: 12, Epoch: 124, Loss: 0.0006\nIter: 13, Epoch: 124, Loss: 0.0008\nIter: 14, Epoch: 124, Loss: 0.0008\nIter: 15, Epoch: 124, Loss: 0.0005\nIter: 16, Epoch: 124, Loss: 0.0006\nIter: 17, Epoch: 124, Loss: 0.0008\nIter: 18, Epoch: 124, Loss: 0.0008\nIter: 19, Epoch: 124, Loss: 0.0009\nIter: 20, Epoch: 124, Loss: 0.0006\nIter: 21, Epoch: 124, Loss: 0.0008\nIter: 22, Epoch: 124, Loss: 0.0011\nIter: 23, Epoch: 124, Loss: 0.0013\nIter: 24, Epoch: 124, Loss: 0.0007\nIter: 25, Epoch: 124, Loss: 0.0007\nIter: 26, Epoch: 124, Loss: 0.0008\nIter: 27, Epoch: 124, Loss: 0.0008\nIter: 28, Epoch: 124, Loss: 0.0008\nIter: 29, Epoch: 124, Loss: 0.0013\nIter: 30, Epoch: 124, Loss: 0.0006\n","truncated":false}} +%--- +%[output:123a2d18] +% data: {"dataType":"text","outputData":{"text":">> Epoch 124 validation loss: 0.0013\n","truncated":false}} +%--- +%[output:605ff620] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 125, Loss: 0.0019\nIter: 2, Epoch: 125, Loss: 0.0007\nIter: 3, Epoch: 125, Loss: 0.0008\nIter: 4, Epoch: 125, Loss: 0.0007\nIter: 5, Epoch: 125, Loss: 0.0005\nIter: 6, Epoch: 125, Loss: 0.0008\nIter: 7, Epoch: 125, Loss: 0.0009\nIter: 8, Epoch: 125, Loss: 0.0012\nIter: 9, Epoch: 125, Loss: 0.0009\nIter: 10, Epoch: 125, Loss: 0.0010\nIter: 11, Epoch: 125, Loss: 0.0005\nIter: 12, Epoch: 125, Loss: 0.0005\nIter: 13, Epoch: 125, Loss: 0.0008\nIter: 14, Epoch: 125, Loss: 0.0008\nIter: 15, Epoch: 125, Loss: 0.0005\nIter: 16, Epoch: 125, Loss: 0.0006\nIter: 17, Epoch: 125, Loss: 0.0007\nIter: 18, Epoch: 125, Loss: 0.0008\nIter: 19, Epoch: 125, Loss: 0.0009\nIter: 20, Epoch: 125, Loss: 0.0006\nIter: 21, Epoch: 125, Loss: 0.0008\nIter: 22, Epoch: 125, Loss: 0.0011\nIter: 23, Epoch: 125, Loss: 0.0012\nIter: 24, Epoch: 125, Loss: 0.0006\nIter: 25, Epoch: 125, Loss: 0.0006\nIter: 26, Epoch: 125, Loss: 0.0008\nIter: 27, Epoch: 125, Loss: 0.0008\nIter: 28, Epoch: 125, Loss: 0.0008\nIter: 29, Epoch: 125, Loss: 0.0013\nIter: 30, Epoch: 125, Loss: 0.0007\n","truncated":false}} +%--- +%[output:676f7048] +% data: {"dataType":"text","outputData":{"text":">> Epoch 125 validation loss: 0.0012\n","truncated":false}} +%--- +%[output:31369086] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 126, Loss: 0.0018\nIter: 2, Epoch: 126, Loss: 0.0006\nIter: 3, Epoch: 126, Loss: 0.0008\nIter: 4, Epoch: 126, Loss: 0.0007\nIter: 5, Epoch: 126, Loss: 0.0005\nIter: 6, Epoch: 126, Loss: 0.0008\nIter: 7, Epoch: 126, Loss: 0.0009\nIter: 8, Epoch: 126, Loss: 0.0011\nIter: 9, Epoch: 126, Loss: 0.0009\nIter: 10, Epoch: 126, Loss: 0.0009\nIter: 11, Epoch: 126, Loss: 0.0005\nIter: 12, Epoch: 126, Loss: 0.0005\nIter: 13, Epoch: 126, Loss: 0.0007\nIter: 14, Epoch: 126, Loss: 0.0008\nIter: 15, Epoch: 126, Loss: 0.0005\nIter: 16, Epoch: 126, Loss: 0.0006\nIter: 17, Epoch: 126, Loss: 0.0007\nIter: 18, Epoch: 126, Loss: 0.0008\nIter: 19, Epoch: 126, Loss: 0.0009\nIter: 20, Epoch: 126, Loss: 0.0006\nIter: 21, Epoch: 126, Loss: 0.0008\nIter: 22, Epoch: 126, Loss: 0.0011\nIter: 23, Epoch: 126, Loss: 0.0011\nIter: 24, Epoch: 126, Loss: 0.0006\nIter: 25, Epoch: 126, Loss: 0.0006\nIter: 26, Epoch: 126, Loss: 0.0008\nIter: 27, Epoch: 126, Loss: 0.0008\nIter: 28, Epoch: 126, Loss: 0.0007\nIter: 29, Epoch: 126, Loss: 0.0012\nIter: 30, Epoch: 126, Loss: 0.0006\n","truncated":false}} +%--- +%[output:393ba012] +% data: {"dataType":"text","outputData":{"text":">> Epoch 126 validation loss: 0.0012\n","truncated":false}} +%--- +%[output:572767b5] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 127, Loss: 0.0018\nIter: 2, Epoch: 127, Loss: 0.0006\nIter: 3, Epoch: 127, Loss: 0.0008\nIter: 4, Epoch: 127, Loss: 0.0007\nIter: 5, Epoch: 127, Loss: 0.0005\nIter: 6, Epoch: 127, Loss: 0.0008\nIter: 7, Epoch: 127, Loss: 0.0009\nIter: 8, Epoch: 127, Loss: 0.0011\nIter: 9, Epoch: 127, Loss: 0.0009\nIter: 10, Epoch: 127, Loss: 0.0009\nIter: 11, Epoch: 127, Loss: 0.0005\nIter: 12, Epoch: 127, Loss: 0.0005\nIter: 13, Epoch: 127, Loss: 0.0007\nIter: 14, Epoch: 127, Loss: 0.0008\nIter: 15, Epoch: 127, Loss: 0.0005\nIter: 16, Epoch: 127, Loss: 0.0006\nIter: 17, Epoch: 127, Loss: 0.0007\nIter: 18, Epoch: 127, Loss: 0.0007\nIter: 19, Epoch: 127, Loss: 0.0009\nIter: 20, Epoch: 127, Loss: 0.0005\nIter: 21, Epoch: 127, Loss: 0.0008\nIter: 22, Epoch: 127, Loss: 0.0011\nIter: 23, Epoch: 127, Loss: 0.0011\nIter: 24, Epoch: 127, Loss: 0.0006\nIter: 25, Epoch: 127, Loss: 0.0006\nIter: 26, Epoch: 127, Loss: 0.0008\nIter: 27, Epoch: 127, Loss: 0.0007\nIter: 28, Epoch: 127, Loss: 0.0007\nIter: 29, Epoch: 127, Loss: 0.0012\nIter: 30, Epoch: 127, Loss: 0.0006\n","truncated":false}} +%--- +%[output:10beede1] +% data: {"dataType":"text","outputData":{"text":">> Epoch 127 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:971f3ed3] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 128, Loss: 0.0017\nIter: 2, Epoch: 128, Loss: 0.0006\nIter: 3, Epoch: 128, Loss: 0.0007\nIter: 4, Epoch: 128, Loss: 0.0007\nIter: 5, Epoch: 128, Loss: 0.0005\nIter: 6, Epoch: 128, Loss: 0.0007\nIter: 7, Epoch: 128, Loss: 0.0008\nIter: 8, Epoch: 128, Loss: 0.0012\nIter: 9, Epoch: 128, Loss: 0.0009\nIter: 10, Epoch: 128, Loss: 0.0009\nIter: 11, Epoch: 128, Loss: 0.0005\nIter: 12, Epoch: 128, Loss: 0.0005\nIter: 13, Epoch: 128, Loss: 0.0007\nIter: 14, Epoch: 128, Loss: 0.0007\nIter: 15, Epoch: 128, Loss: 0.0005\nIter: 16, Epoch: 128, Loss: 0.0006\nIter: 17, Epoch: 128, Loss: 0.0007\nIter: 18, Epoch: 128, Loss: 0.0007\nIter: 19, Epoch: 128, Loss: 0.0009\nIter: 20, Epoch: 128, Loss: 0.0005\nIter: 21, Epoch: 128, Loss: 0.0008\nIter: 22, Epoch: 128, Loss: 0.0010\nIter: 23, Epoch: 128, Loss: 0.0011\nIter: 24, Epoch: 128, Loss: 0.0006\nIter: 25, Epoch: 128, Loss: 0.0006\nIter: 26, Epoch: 128, Loss: 0.0008\nIter: 27, Epoch: 128, Loss: 0.0007\nIter: 28, Epoch: 128, Loss: 0.0007\nIter: 29, Epoch: 128, Loss: 0.0012\nIter: 30, Epoch: 128, Loss: 0.0006\n","truncated":false}} +%--- +%[output:39b35cec] +% data: {"dataType":"text","outputData":{"text":">> Epoch 128 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:81e55fff] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 129, Loss: 0.0017\nIter: 2, Epoch: 129, Loss: 0.0006\nIter: 3, Epoch: 129, Loss: 0.0007\nIter: 4, Epoch: 129, Loss: 0.0007\nIter: 5, Epoch: 129, Loss: 0.0005\nIter: 6, Epoch: 129, Loss: 0.0007\nIter: 7, Epoch: 129, Loss: 0.0008\nIter: 8, Epoch: 129, Loss: 0.0011\nIter: 9, Epoch: 129, Loss: 0.0009\nIter: 10, Epoch: 129, Loss: 0.0009\nIter: 11, Epoch: 129, Loss: 0.0005\nIter: 12, Epoch: 129, Loss: 0.0005\nIter: 13, Epoch: 129, Loss: 0.0006\nIter: 14, Epoch: 129, Loss: 0.0007\nIter: 15, Epoch: 129, Loss: 0.0005\nIter: 16, Epoch: 129, Loss: 0.0006\nIter: 17, Epoch: 129, Loss: 0.0007\nIter: 18, Epoch: 129, Loss: 0.0007\nIter: 19, Epoch: 129, Loss: 0.0009\nIter: 20, Epoch: 129, Loss: 0.0005\nIter: 21, Epoch: 129, Loss: 0.0008\nIter: 22, Epoch: 129, Loss: 0.0010\nIter: 23, Epoch: 129, Loss: 0.0011\nIter: 24, Epoch: 129, Loss: 0.0006\nIter: 25, Epoch: 129, Loss: 0.0006\nIter: 26, Epoch: 129, Loss: 0.0007\nIter: 27, Epoch: 129, Loss: 0.0007\nIter: 28, Epoch: 129, Loss: 0.0007\nIter: 29, Epoch: 129, Loss: 0.0011\nIter: 30, Epoch: 129, Loss: 0.0006\n","truncated":false}} +%--- +%[output:4b4ba426] +% data: {"dataType":"text","outputData":{"text":">> Epoch 129 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:95a00284] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 130, Loss: 0.0017\nIter: 2, Epoch: 130, Loss: 0.0006\nIter: 3, Epoch: 130, Loss: 0.0007\nIter: 4, Epoch: 130, Loss: 0.0007\nIter: 5, Epoch: 130, Loss: 0.0005\nIter: 6, Epoch: 130, Loss: 0.0007\nIter: 7, Epoch: 130, Loss: 0.0008\nIter: 8, Epoch: 130, Loss: 0.0012\nIter: 9, Epoch: 130, Loss: 0.0009\nIter: 10, Epoch: 130, Loss: 0.0009\nIter: 11, Epoch: 130, Loss: 0.0005\nIter: 12, Epoch: 130, Loss: 0.0005\nIter: 13, Epoch: 130, Loss: 0.0006\nIter: 14, Epoch: 130, Loss: 0.0007\nIter: 15, Epoch: 130, Loss: 0.0005\nIter: 16, Epoch: 130, Loss: 0.0006\nIter: 17, Epoch: 130, Loss: 0.0007\nIter: 18, Epoch: 130, Loss: 0.0007\nIter: 19, Epoch: 130, Loss: 0.0009\nIter: 20, Epoch: 130, Loss: 0.0005\nIter: 21, Epoch: 130, Loss: 0.0008\nIter: 22, Epoch: 130, Loss: 0.0010\nIter: 23, Epoch: 130, Loss: 0.0011\nIter: 24, Epoch: 130, Loss: 0.0006\nIter: 25, Epoch: 130, Loss: 0.0006\nIter: 26, Epoch: 130, Loss: 0.0007\nIter: 27, Epoch: 130, Loss: 0.0007\nIter: 28, Epoch: 130, Loss: 0.0007\nIter: 29, Epoch: 130, Loss: 0.0011\nIter: 30, Epoch: 130, Loss: 0.0006\n","truncated":false}} +%--- +%[output:07a485e3] +% data: {"dataType":"text","outputData":{"text":">> Epoch 130 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:51e9f898] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 131, Loss: 0.0017\nIter: 2, Epoch: 131, Loss: 0.0006\nIter: 3, Epoch: 131, Loss: 0.0007\nIter: 4, Epoch: 131, Loss: 0.0007\nIter: 5, Epoch: 131, Loss: 0.0004\nIter: 6, Epoch: 131, Loss: 0.0007\nIter: 7, Epoch: 131, Loss: 0.0008\nIter: 8, Epoch: 131, Loss: 0.0012\nIter: 9, Epoch: 131, Loss: 0.0009\nIter: 10, Epoch: 131, Loss: 0.0009\nIter: 11, Epoch: 131, Loss: 0.0005\nIter: 12, Epoch: 131, Loss: 0.0005\nIter: 13, Epoch: 131, Loss: 0.0006\nIter: 14, Epoch: 131, Loss: 0.0007\nIter: 15, Epoch: 131, Loss: 0.0005\nIter: 16, Epoch: 131, Loss: 0.0006\nIter: 17, Epoch: 131, Loss: 0.0007\nIter: 18, Epoch: 131, Loss: 0.0007\nIter: 19, Epoch: 131, Loss: 0.0009\nIter: 20, Epoch: 131, Loss: 0.0005\nIter: 21, Epoch: 131, Loss: 0.0008\nIter: 22, Epoch: 131, Loss: 0.0010\nIter: 23, Epoch: 131, Loss: 0.0011\nIter: 24, Epoch: 131, Loss: 0.0005\nIter: 25, Epoch: 131, Loss: 0.0006\nIter: 26, Epoch: 131, Loss: 0.0007\nIter: 27, Epoch: 131, Loss: 0.0007\nIter: 28, Epoch: 131, Loss: 0.0007\nIter: 29, Epoch: 131, Loss: 0.0011\nIter: 30, Epoch: 131, Loss: 0.0006\n","truncated":false}} +%--- +%[output:17923bfa] +% data: {"dataType":"text","outputData":{"text":">> Epoch 131 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:448c9078] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 132, Loss: 0.0017\nIter: 2, Epoch: 132, Loss: 0.0006\nIter: 3, Epoch: 132, Loss: 0.0007\nIter: 4, Epoch: 132, Loss: 0.0007\nIter: 5, Epoch: 132, Loss: 0.0004\nIter: 6, Epoch: 132, Loss: 0.0007\nIter: 7, Epoch: 132, Loss: 0.0008\nIter: 8, Epoch: 132, Loss: 0.0012\nIter: 9, Epoch: 132, Loss: 0.0009\nIter: 10, Epoch: 132, Loss: 0.0009\nIter: 11, Epoch: 132, Loss: 0.0005\nIter: 12, Epoch: 132, Loss: 0.0005\nIter: 13, Epoch: 132, Loss: 0.0006\nIter: 14, Epoch: 132, Loss: 0.0007\nIter: 15, Epoch: 132, Loss: 0.0005\nIter: 16, Epoch: 132, Loss: 0.0006\nIter: 17, Epoch: 132, Loss: 0.0007\nIter: 18, Epoch: 132, Loss: 0.0007\nIter: 19, Epoch: 132, Loss: 0.0009\nIter: 20, Epoch: 132, Loss: 0.0005\nIter: 21, Epoch: 132, Loss: 0.0008\nIter: 22, Epoch: 132, Loss: 0.0010\nIter: 23, Epoch: 132, Loss: 0.0010\nIter: 24, Epoch: 132, Loss: 0.0005\nIter: 25, Epoch: 132, Loss: 0.0006\nIter: 26, Epoch: 132, Loss: 0.0007\nIter: 27, Epoch: 132, Loss: 0.0006\nIter: 28, Epoch: 132, Loss: 0.0007\nIter: 29, Epoch: 132, Loss: 0.0011\nIter: 30, Epoch: 132, Loss: 0.0006\n","truncated":false}} +%--- +%[output:07b75789] +% data: {"dataType":"text","outputData":{"text":">> Epoch 132 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:8e5fea78] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 133, Loss: 0.0017\nIter: 2, Epoch: 133, Loss: 0.0006\nIter: 3, Epoch: 133, Loss: 0.0007\nIter: 4, Epoch: 133, Loss: 0.0007\nIter: 5, Epoch: 133, Loss: 0.0004\nIter: 6, Epoch: 133, Loss: 0.0007\nIter: 7, Epoch: 133, Loss: 0.0008\nIter: 8, Epoch: 133, Loss: 0.0012\nIter: 9, Epoch: 133, Loss: 0.0009\nIter: 10, Epoch: 133, Loss: 0.0009\nIter: 11, Epoch: 133, Loss: 0.0005\nIter: 12, Epoch: 133, Loss: 0.0005\nIter: 13, Epoch: 133, Loss: 0.0006\nIter: 14, Epoch: 133, Loss: 0.0007\nIter: 15, Epoch: 133, Loss: 0.0005\nIter: 16, Epoch: 133, Loss: 0.0006\nIter: 17, Epoch: 133, Loss: 0.0007\nIter: 18, Epoch: 133, Loss: 0.0007\nIter: 19, Epoch: 133, Loss: 0.0009\nIter: 20, Epoch: 133, Loss: 0.0005\nIter: 21, Epoch: 133, Loss: 0.0008\nIter: 22, Epoch: 133, Loss: 0.0010\nIter: 23, Epoch: 133, Loss: 0.0010\nIter: 24, Epoch: 133, Loss: 0.0005\nIter: 25, Epoch: 133, Loss: 0.0006\nIter: 26, Epoch: 133, Loss: 0.0007\nIter: 27, Epoch: 133, Loss: 0.0006\nIter: 28, Epoch: 133, Loss: 0.0007\nIter: 29, Epoch: 133, Loss: 0.0011\nIter: 30, Epoch: 133, Loss: 0.0006\n","truncated":false}} +%--- +%[output:68aad416] +% data: {"dataType":"text","outputData":{"text":">> Epoch 133 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:7307a7df] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 134, Loss: 0.0017\nIter: 2, Epoch: 134, Loss: 0.0006\nIter: 3, Epoch: 134, Loss: 0.0007\nIter: 4, Epoch: 134, Loss: 0.0007\nIter: 5, Epoch: 134, Loss: 0.0004\nIter: 6, Epoch: 134, Loss: 0.0007\nIter: 7, Epoch: 134, Loss: 0.0008\nIter: 8, Epoch: 134, Loss: 0.0012\nIter: 9, Epoch: 134, Loss: 0.0009\nIter: 10, Epoch: 134, Loss: 0.0008\nIter: 11, Epoch: 134, Loss: 0.0005\nIter: 12, Epoch: 134, Loss: 0.0005\nIter: 13, Epoch: 134, Loss: 0.0006\nIter: 14, Epoch: 134, Loss: 0.0007\nIter: 15, Epoch: 134, Loss: 0.0005\nIter: 16, Epoch: 134, Loss: 0.0006\nIter: 17, Epoch: 134, Loss: 0.0007\nIter: 18, Epoch: 134, Loss: 0.0007\nIter: 19, Epoch: 134, Loss: 0.0009\nIter: 20, Epoch: 134, Loss: 0.0005\nIter: 21, Epoch: 134, Loss: 0.0008\nIter: 22, Epoch: 134, Loss: 0.0010\nIter: 23, Epoch: 134, Loss: 0.0010\nIter: 24, Epoch: 134, Loss: 0.0005\nIter: 25, Epoch: 134, Loss: 0.0006\nIter: 26, Epoch: 134, Loss: 0.0007\nIter: 27, Epoch: 134, Loss: 0.0006\nIter: 28, Epoch: 134, Loss: 0.0007\nIter: 29, Epoch: 134, Loss: 0.0011\nIter: 30, Epoch: 134, Loss: 0.0006\n","truncated":false}} +%--- +%[output:9ea02d93] +% data: {"dataType":"text","outputData":{"text":">> Epoch 134 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:366b2194] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 135, Loss: 0.0018\nIter: 2, Epoch: 135, Loss: 0.0006\nIter: 3, Epoch: 135, Loss: 0.0007\nIter: 4, Epoch: 135, Loss: 0.0007\nIter: 5, Epoch: 135, Loss: 0.0004\nIter: 6, Epoch: 135, Loss: 0.0007\nIter: 7, Epoch: 135, Loss: 0.0008\nIter: 8, Epoch: 135, Loss: 0.0012\nIter: 9, Epoch: 135, Loss: 0.0009\nIter: 10, Epoch: 135, Loss: 0.0008\nIter: 11, Epoch: 135, Loss: 0.0005\nIter: 12, Epoch: 135, Loss: 0.0005\nIter: 13, Epoch: 135, Loss: 0.0006\nIter: 14, Epoch: 135, Loss: 0.0007\nIter: 15, Epoch: 135, Loss: 0.0005\nIter: 16, Epoch: 135, Loss: 0.0006\nIter: 17, Epoch: 135, Loss: 0.0007\nIter: 18, Epoch: 135, Loss: 0.0007\nIter: 19, Epoch: 135, Loss: 0.0009\nIter: 20, Epoch: 135, Loss: 0.0005\nIter: 21, Epoch: 135, Loss: 0.0008\nIter: 22, Epoch: 135, Loss: 0.0010\nIter: 23, Epoch: 135, Loss: 0.0011\nIter: 24, Epoch: 135, Loss: 0.0005\nIter: 25, Epoch: 135, Loss: 0.0006\nIter: 26, Epoch: 135, Loss: 0.0007\nIter: 27, Epoch: 135, Loss: 0.0006\nIter: 28, Epoch: 135, Loss: 0.0007\nIter: 29, Epoch: 135, Loss: 0.0011\nIter: 30, Epoch: 135, Loss: 0.0006\n","truncated":false}} +%--- +%[output:0e329285] +% data: {"dataType":"text","outputData":{"text":">> Epoch 135 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:9cb564cd] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 136, Loss: 0.0018\nIter: 2, Epoch: 136, Loss: 0.0006\nIter: 3, Epoch: 136, Loss: 0.0008\nIter: 4, Epoch: 136, Loss: 0.0007\nIter: 5, Epoch: 136, Loss: 0.0004\nIter: 6, Epoch: 136, Loss: 0.0007\nIter: 7, Epoch: 136, Loss: 0.0008\nIter: 8, Epoch: 136, Loss: 0.0012\nIter: 9, Epoch: 136, Loss: 0.0009\nIter: 10, Epoch: 136, Loss: 0.0008\nIter: 11, Epoch: 136, Loss: 0.0005\nIter: 12, Epoch: 136, Loss: 0.0005\nIter: 13, Epoch: 136, Loss: 0.0006\nIter: 14, Epoch: 136, Loss: 0.0006\nIter: 15, Epoch: 136, Loss: 0.0005\nIter: 16, Epoch: 136, Loss: 0.0006\nIter: 17, Epoch: 136, Loss: 0.0007\nIter: 18, Epoch: 136, Loss: 0.0007\nIter: 19, Epoch: 136, Loss: 0.0009\nIter: 20, Epoch: 136, Loss: 0.0005\nIter: 21, Epoch: 136, Loss: 0.0008\nIter: 22, Epoch: 136, Loss: 0.0010\nIter: 23, Epoch: 136, Loss: 0.0011\nIter: 24, Epoch: 136, Loss: 0.0006\nIter: 25, Epoch: 136, Loss: 0.0006\nIter: 26, Epoch: 136, Loss: 0.0007\nIter: 27, Epoch: 136, Loss: 0.0006\nIter: 28, Epoch: 136, Loss: 0.0007\nIter: 29, Epoch: 136, Loss: 0.0011\nIter: 30, Epoch: 136, Loss: 0.0006\n","truncated":false}} +%--- +%[output:62d63ed5] +% data: {"dataType":"text","outputData":{"text":">> Epoch 136 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:9da4f824] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 137, Loss: 0.0018\nIter: 2, Epoch: 137, Loss: 0.0006\nIter: 3, Epoch: 137, Loss: 0.0008\nIter: 4, Epoch: 137, Loss: 0.0007\nIter: 5, Epoch: 137, Loss: 0.0004\nIter: 6, Epoch: 137, Loss: 0.0007\nIter: 7, Epoch: 137, Loss: 0.0008\nIter: 8, Epoch: 137, Loss: 0.0013\nIter: 9, Epoch: 137, Loss: 0.0009\nIter: 10, Epoch: 137, Loss: 0.0008\nIter: 11, Epoch: 137, Loss: 0.0005\nIter: 12, Epoch: 137, Loss: 0.0005\nIter: 13, Epoch: 137, Loss: 0.0006\nIter: 14, Epoch: 137, Loss: 0.0006\nIter: 15, Epoch: 137, Loss: 0.0005\nIter: 16, Epoch: 137, Loss: 0.0006\nIter: 17, Epoch: 137, Loss: 0.0007\nIter: 18, Epoch: 137, Loss: 0.0007\nIter: 19, Epoch: 137, Loss: 0.0009\nIter: 20, Epoch: 137, Loss: 0.0005\nIter: 21, Epoch: 137, Loss: 0.0008\nIter: 22, Epoch: 137, Loss: 0.0010\nIter: 23, Epoch: 137, Loss: 0.0011\nIter: 24, Epoch: 137, Loss: 0.0006\nIter: 25, Epoch: 137, Loss: 0.0006\nIter: 26, Epoch: 137, Loss: 0.0007\nIter: 27, Epoch: 137, Loss: 0.0006\nIter: 28, Epoch: 137, Loss: 0.0007\nIter: 29, Epoch: 137, Loss: 0.0011\nIter: 30, Epoch: 137, Loss: 0.0006\n","truncated":false}} +%--- +%[output:67007505] +% data: {"dataType":"text","outputData":{"text":">> Epoch 137 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:02dcff7f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 138, Loss: 0.0018\nIter: 2, Epoch: 138, Loss: 0.0006\nIter: 3, Epoch: 138, Loss: 0.0008\nIter: 4, Epoch: 138, Loss: 0.0007\nIter: 5, Epoch: 138, Loss: 0.0004\nIter: 6, Epoch: 138, Loss: 0.0007\nIter: 7, Epoch: 138, Loss: 0.0008\nIter: 8, Epoch: 138, Loss: 0.0013\nIter: 9, Epoch: 138, Loss: 0.0009\nIter: 10, Epoch: 138, Loss: 0.0008\nIter: 11, Epoch: 138, Loss: 0.0005\nIter: 12, Epoch: 138, Loss: 0.0005\nIter: 13, Epoch: 138, Loss: 0.0006\nIter: 14, Epoch: 138, Loss: 0.0006\nIter: 15, Epoch: 138, Loss: 0.0005\nIter: 16, Epoch: 138, Loss: 0.0006\nIter: 17, Epoch: 138, Loss: 0.0007\nIter: 18, Epoch: 138, Loss: 0.0007\nIter: 19, Epoch: 138, Loss: 0.0009\nIter: 20, Epoch: 138, Loss: 0.0005\nIter: 21, Epoch: 138, Loss: 0.0008\nIter: 22, Epoch: 138, Loss: 0.0010\nIter: 23, Epoch: 138, Loss: 0.0011\nIter: 24, Epoch: 138, Loss: 0.0006\nIter: 25, Epoch: 138, Loss: 0.0006\nIter: 26, Epoch: 138, Loss: 0.0007\nIter: 27, Epoch: 138, Loss: 0.0006\nIter: 28, Epoch: 138, Loss: 0.0007\nIter: 29, Epoch: 138, Loss: 0.0011\nIter: 30, Epoch: 138, Loss: 0.0006\n","truncated":false}} +%--- +%[output:5dbf1b45] +% data: {"dataType":"text","outputData":{"text":">> Epoch 138 validation loss: 0.0012\n","truncated":false}} +%--- +%[output:11080f9f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 139, Loss: 0.0018\nIter: 2, Epoch: 139, Loss: 0.0006\nIter: 3, Epoch: 139, Loss: 0.0008\nIter: 4, Epoch: 139, Loss: 0.0007\nIter: 5, Epoch: 139, Loss: 0.0004\nIter: 6, Epoch: 139, Loss: 0.0007\nIter: 7, Epoch: 139, Loss: 0.0009\nIter: 8, Epoch: 139, Loss: 0.0013\nIter: 9, Epoch: 139, Loss: 0.0009\nIter: 10, Epoch: 139, Loss: 0.0008\nIter: 11, Epoch: 139, Loss: 0.0005\nIter: 12, Epoch: 139, Loss: 0.0005\nIter: 13, Epoch: 139, Loss: 0.0006\nIter: 14, Epoch: 139, Loss: 0.0006\nIter: 15, Epoch: 139, Loss: 0.0005\nIter: 16, Epoch: 139, Loss: 0.0006\nIter: 17, Epoch: 139, Loss: 0.0007\nIter: 18, Epoch: 139, Loss: 0.0008\nIter: 19, Epoch: 139, Loss: 0.0009\nIter: 20, Epoch: 139, Loss: 0.0005\nIter: 21, Epoch: 139, Loss: 0.0008\nIter: 22, Epoch: 139, Loss: 0.0010\nIter: 23, Epoch: 139, Loss: 0.0011\nIter: 24, Epoch: 139, Loss: 0.0006\nIter: 25, Epoch: 139, Loss: 0.0006\nIter: 26, Epoch: 139, Loss: 0.0007\nIter: 27, Epoch: 139, Loss: 0.0005\nIter: 28, Epoch: 139, Loss: 0.0007\nIter: 29, Epoch: 139, Loss: 0.0011\nIter: 30, Epoch: 139, Loss: 0.0006\n","truncated":false}} +%--- +%[output:81ee8730] +% data: {"dataType":"text","outputData":{"text":">> Epoch 139 validation loss: 0.0012\n","truncated":false}} +%--- +%[output:6087ca7c] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 140, Loss: 0.0019\nIter: 2, Epoch: 140, Loss: 0.0006\nIter: 3, Epoch: 140, Loss: 0.0008\nIter: 4, Epoch: 140, Loss: 0.0007\nIter: 5, Epoch: 140, Loss: 0.0004\nIter: 6, Epoch: 140, Loss: 0.0007\nIter: 7, Epoch: 140, Loss: 0.0009\nIter: 8, Epoch: 140, Loss: 0.0013\nIter: 9, Epoch: 140, Loss: 0.0009\nIter: 10, Epoch: 140, Loss: 0.0009\nIter: 11, Epoch: 140, Loss: 0.0005\nIter: 12, Epoch: 140, Loss: 0.0004\nIter: 13, Epoch: 140, Loss: 0.0006\nIter: 14, Epoch: 140, Loss: 0.0006\nIter: 15, Epoch: 140, Loss: 0.0005\nIter: 16, Epoch: 140, Loss: 0.0006\nIter: 17, Epoch: 140, Loss: 0.0008\nIter: 18, Epoch: 140, Loss: 0.0008\nIter: 19, Epoch: 140, Loss: 0.0010\nIter: 20, Epoch: 140, Loss: 0.0005\nIter: 21, Epoch: 140, Loss: 0.0008\nIter: 22, Epoch: 140, Loss: 0.0010\nIter: 23, Epoch: 140, Loss: 0.0012\nIter: 24, Epoch: 140, Loss: 0.0007\nIter: 25, Epoch: 140, Loss: 0.0007\nIter: 26, Epoch: 140, Loss: 0.0008\nIter: 27, Epoch: 140, Loss: 0.0005\nIter: 28, Epoch: 140, Loss: 0.0007\nIter: 29, Epoch: 140, Loss: 0.0011\nIter: 30, Epoch: 140, Loss: 0.0006\n","truncated":false}} +%--- +%[output:976c9ab3] +% data: {"dataType":"text","outputData":{"text":">> Epoch 140 validation loss: 0.0012\n","truncated":false}} +%--- +%[output:9171ca81] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 141, Loss: 0.0019\nIter: 2, Epoch: 141, Loss: 0.0006\nIter: 3, Epoch: 141, Loss: 0.0009\nIter: 4, Epoch: 141, Loss: 0.0008\nIter: 5, Epoch: 141, Loss: 0.0004\nIter: 6, Epoch: 141, Loss: 0.0007\nIter: 7, Epoch: 141, Loss: 0.0009\nIter: 8, Epoch: 141, Loss: 0.0014\nIter: 9, Epoch: 141, Loss: 0.0010\nIter: 10, Epoch: 141, Loss: 0.0009\nIter: 11, Epoch: 141, Loss: 0.0005\nIter: 12, Epoch: 141, Loss: 0.0005\nIter: 13, Epoch: 141, Loss: 0.0006\nIter: 14, Epoch: 141, Loss: 0.0007\nIter: 15, Epoch: 141, Loss: 0.0005\nIter: 16, Epoch: 141, Loss: 0.0006\nIter: 17, Epoch: 141, Loss: 0.0009\nIter: 18, Epoch: 141, Loss: 0.0009\nIter: 19, Epoch: 141, Loss: 0.0010\nIter: 20, Epoch: 141, Loss: 0.0006\nIter: 21, Epoch: 141, Loss: 0.0008\nIter: 22, Epoch: 141, Loss: 0.0011\nIter: 23, Epoch: 141, Loss: 0.0012\nIter: 24, Epoch: 141, Loss: 0.0007\nIter: 25, Epoch: 141, Loss: 0.0007\nIter: 26, Epoch: 141, Loss: 0.0009\nIter: 27, Epoch: 141, Loss: 0.0005\nIter: 28, Epoch: 141, Loss: 0.0008\nIter: 29, Epoch: 141, Loss: 0.0011\nIter: 30, Epoch: 141, Loss: 0.0006\n","truncated":false}} +%--- +%[output:1cd9a3bd] +% data: {"dataType":"text","outputData":{"text":">> Epoch 141 validation loss: 0.0012\n","truncated":false}} +%--- +%[output:6eda7cb5] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 142, Loss: 0.0020\nIter: 2, Epoch: 142, Loss: 0.0006\nIter: 3, Epoch: 142, Loss: 0.0010\nIter: 4, Epoch: 142, Loss: 0.0009\nIter: 5, Epoch: 142, Loss: 0.0005\nIter: 6, Epoch: 142, Loss: 0.0007\nIter: 7, Epoch: 142, Loss: 0.0009\nIter: 8, Epoch: 142, Loss: 0.0014\nIter: 9, Epoch: 142, Loss: 0.0010\nIter: 10, Epoch: 142, Loss: 0.0010\nIter: 11, Epoch: 142, Loss: 0.0006\nIter: 12, Epoch: 142, Loss: 0.0005\nIter: 13, Epoch: 142, Loss: 0.0007\nIter: 14, Epoch: 142, Loss: 0.0007\nIter: 15, Epoch: 142, Loss: 0.0005\nIter: 16, Epoch: 142, Loss: 0.0006\nIter: 17, Epoch: 142, Loss: 0.0009\nIter: 18, Epoch: 142, Loss: 0.0011\nIter: 19, Epoch: 142, Loss: 0.0011\nIter: 20, Epoch: 142, Loss: 0.0006\nIter: 21, Epoch: 142, Loss: 0.0008\nIter: 22, Epoch: 142, Loss: 0.0010\nIter: 23, Epoch: 142, Loss: 0.0012\nIter: 24, Epoch: 142, Loss: 0.0007\nIter: 25, Epoch: 142, Loss: 0.0007\nIter: 26, Epoch: 142, Loss: 0.0009\nIter: 27, Epoch: 142, Loss: 0.0005\nIter: 28, Epoch: 142, Loss: 0.0009\nIter: 29, Epoch: 142, Loss: 0.0011\nIter: 30, Epoch: 142, Loss: 0.0006\n","truncated":false}} +%--- +%[output:60d0216c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 142 validation loss: 0.0012\n","truncated":false}} +%--- +%[output:58f8fccf] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 143, Loss: 0.0019\nIter: 2, Epoch: 143, Loss: 0.0006\nIter: 3, Epoch: 143, Loss: 0.0011\nIter: 4, Epoch: 143, Loss: 0.0010\nIter: 5, Epoch: 143, Loss: 0.0005\nIter: 6, Epoch: 143, Loss: 0.0006\nIter: 7, Epoch: 143, Loss: 0.0008\nIter: 8, Epoch: 143, Loss: 0.0013\nIter: 9, Epoch: 143, Loss: 0.0010\nIter: 10, Epoch: 143, Loss: 0.0009\nIter: 11, Epoch: 143, Loss: 0.0006\nIter: 12, Epoch: 143, Loss: 0.0005\nIter: 13, Epoch: 143, Loss: 0.0007\nIter: 14, Epoch: 143, Loss: 0.0008\nIter: 15, Epoch: 143, Loss: 0.0005\nIter: 16, Epoch: 143, Loss: 0.0006\nIter: 17, Epoch: 143, Loss: 0.0009\nIter: 18, Epoch: 143, Loss: 0.0011\nIter: 19, Epoch: 143, Loss: 0.0012\nIter: 20, Epoch: 143, Loss: 0.0008\nIter: 21, Epoch: 143, Loss: 0.0007\nIter: 22, Epoch: 143, Loss: 0.0009\nIter: 23, Epoch: 143, Loss: 0.0011\nIter: 24, Epoch: 143, Loss: 0.0007\nIter: 25, Epoch: 143, Loss: 0.0007\nIter: 26, Epoch: 143, Loss: 0.0009\nIter: 27, Epoch: 143, Loss: 0.0005\nIter: 28, Epoch: 143, Loss: 0.0009\nIter: 29, Epoch: 143, Loss: 0.0011\nIter: 30, Epoch: 143, Loss: 0.0006\n","truncated":false}} +%--- +%[output:71625090] +% data: {"dataType":"text","outputData":{"text":">> Epoch 143 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:9a066e9f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 144, Loss: 0.0018\nIter: 2, Epoch: 144, Loss: 0.0006\nIter: 3, Epoch: 144, Loss: 0.0010\nIter: 4, Epoch: 144, Loss: 0.0010\nIter: 5, Epoch: 144, Loss: 0.0005\nIter: 6, Epoch: 144, Loss: 0.0006\nIter: 7, Epoch: 144, Loss: 0.0008\nIter: 8, Epoch: 144, Loss: 0.0013\nIter: 9, Epoch: 144, Loss: 0.0010\nIter: 10, Epoch: 144, Loss: 0.0009\nIter: 11, Epoch: 144, Loss: 0.0006\nIter: 12, Epoch: 144, Loss: 0.0005\nIter: 13, Epoch: 144, Loss: 0.0007\nIter: 14, Epoch: 144, Loss: 0.0008\nIter: 15, Epoch: 144, Loss: 0.0006\nIter: 16, Epoch: 144, Loss: 0.0006\nIter: 17, Epoch: 144, Loss: 0.0008\nIter: 18, Epoch: 144, Loss: 0.0010\nIter: 19, Epoch: 144, Loss: 0.0012\nIter: 20, Epoch: 144, Loss: 0.0008\nIter: 21, Epoch: 144, Loss: 0.0007\nIter: 22, Epoch: 144, Loss: 0.0009\nIter: 23, Epoch: 144, Loss: 0.0011\nIter: 24, Epoch: 144, Loss: 0.0006\nIter: 25, Epoch: 144, Loss: 0.0006\nIter: 26, Epoch: 144, Loss: 0.0009\nIter: 27, Epoch: 144, Loss: 0.0004\nIter: 28, Epoch: 144, Loss: 0.0009\nIter: 29, Epoch: 144, Loss: 0.0011\nIter: 30, Epoch: 144, Loss: 0.0006\n","truncated":false}} +%--- +%[output:1c036f02] +% data: {"dataType":"text","outputData":{"text":">> Epoch 144 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:15d4e80f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 145, Loss: 0.0017\nIter: 2, Epoch: 145, Loss: 0.0006\nIter: 3, Epoch: 145, Loss: 0.0009\nIter: 4, Epoch: 145, Loss: 0.0010\nIter: 5, Epoch: 145, Loss: 0.0005\nIter: 6, Epoch: 145, Loss: 0.0006\nIter: 7, Epoch: 145, Loss: 0.0008\nIter: 8, Epoch: 145, Loss: 0.0013\nIter: 9, Epoch: 145, Loss: 0.0009\nIter: 10, Epoch: 145, Loss: 0.0008\nIter: 11, Epoch: 145, Loss: 0.0006\nIter: 12, Epoch: 145, Loss: 0.0005\nIter: 13, Epoch: 145, Loss: 0.0008\nIter: 14, Epoch: 145, Loss: 0.0008\nIter: 15, Epoch: 145, Loss: 0.0006\nIter: 16, Epoch: 145, Loss: 0.0006\nIter: 17, Epoch: 145, Loss: 0.0008\nIter: 18, Epoch: 145, Loss: 0.0010\nIter: 19, Epoch: 145, Loss: 0.0012\nIter: 20, Epoch: 145, Loss: 0.0008\nIter: 21, Epoch: 145, Loss: 0.0006\nIter: 22, Epoch: 145, Loss: 0.0009\nIter: 23, Epoch: 145, Loss: 0.0011\nIter: 24, Epoch: 145, Loss: 0.0006\nIter: 25, Epoch: 145, Loss: 0.0006\nIter: 26, Epoch: 145, Loss: 0.0008\nIter: 27, Epoch: 145, Loss: 0.0004\nIter: 28, Epoch: 145, Loss: 0.0009\nIter: 29, Epoch: 145, Loss: 0.0011\nIter: 30, Epoch: 145, Loss: 0.0006\n","truncated":false}} +%--- +%[output:75b2c1f3] +% data: {"dataType":"text","outputData":{"text":">> Epoch 145 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:392b24cd] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 146, Loss: 0.0018\nIter: 2, Epoch: 146, Loss: 0.0006\nIter: 3, Epoch: 146, Loss: 0.0009\nIter: 4, Epoch: 146, Loss: 0.0010\nIter: 5, Epoch: 146, Loss: 0.0005\nIter: 6, Epoch: 146, Loss: 0.0006\nIter: 7, Epoch: 146, Loss: 0.0008\nIter: 8, Epoch: 146, Loss: 0.0013\nIter: 9, Epoch: 146, Loss: 0.0009\nIter: 10, Epoch: 146, Loss: 0.0008\nIter: 11, Epoch: 146, Loss: 0.0006\nIter: 12, Epoch: 146, Loss: 0.0005\nIter: 13, Epoch: 146, Loss: 0.0008\nIter: 14, Epoch: 146, Loss: 0.0008\nIter: 15, Epoch: 146, Loss: 0.0006\nIter: 16, Epoch: 146, Loss: 0.0007\nIter: 17, Epoch: 146, Loss: 0.0008\nIter: 18, Epoch: 146, Loss: 0.0009\nIter: 19, Epoch: 146, Loss: 0.0012\nIter: 20, Epoch: 146, Loss: 0.0008\nIter: 21, Epoch: 146, Loss: 0.0006\nIter: 22, Epoch: 146, Loss: 0.0009\nIter: 23, Epoch: 146, Loss: 0.0012\nIter: 24, Epoch: 146, Loss: 0.0006\nIter: 25, Epoch: 146, Loss: 0.0006\nIter: 26, Epoch: 146, Loss: 0.0008\nIter: 27, Epoch: 146, Loss: 0.0005\nIter: 28, Epoch: 146, Loss: 0.0009\nIter: 29, Epoch: 146, Loss: 0.0011\nIter: 30, Epoch: 146, Loss: 0.0006\n","truncated":false}} +%--- +%[output:801dabeb] +% data: {"dataType":"text","outputData":{"text":">> Epoch 146 validation loss: 0.0012\n","truncated":false}} +%--- +%[output:99a0520f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 147, Loss: 0.0018\nIter: 2, Epoch: 147, Loss: 0.0006\nIter: 3, Epoch: 147, Loss: 0.0009\nIter: 4, Epoch: 147, Loss: 0.0009\nIter: 5, Epoch: 147, Loss: 0.0005\nIter: 6, Epoch: 147, Loss: 0.0006\nIter: 7, Epoch: 147, Loss: 0.0007\nIter: 8, Epoch: 147, Loss: 0.0013\nIter: 9, Epoch: 147, Loss: 0.0009\nIter: 10, Epoch: 147, Loss: 0.0008\nIter: 11, Epoch: 147, Loss: 0.0005\nIter: 12, Epoch: 147, Loss: 0.0005\nIter: 13, Epoch: 147, Loss: 0.0008\nIter: 14, Epoch: 147, Loss: 0.0008\nIter: 15, Epoch: 147, Loss: 0.0005\nIter: 16, Epoch: 147, Loss: 0.0007\nIter: 17, Epoch: 147, Loss: 0.0008\nIter: 18, Epoch: 147, Loss: 0.0009\nIter: 19, Epoch: 147, Loss: 0.0012\nIter: 20, Epoch: 147, Loss: 0.0008\nIter: 21, Epoch: 147, Loss: 0.0006\nIter: 22, Epoch: 147, Loss: 0.0009\nIter: 23, Epoch: 147, Loss: 0.0012\nIter: 24, Epoch: 147, Loss: 0.0006\nIter: 25, Epoch: 147, Loss: 0.0005\nIter: 26, Epoch: 147, Loss: 0.0008\nIter: 27, Epoch: 147, Loss: 0.0005\nIter: 28, Epoch: 147, Loss: 0.0009\nIter: 29, Epoch: 147, Loss: 0.0011\nIter: 30, Epoch: 147, Loss: 0.0006\n","truncated":false}} +%--- +%[output:852359ab] +% data: {"dataType":"text","outputData":{"text":">> Epoch 147 validation loss: 0.0012\n","truncated":false}} +%--- +%[output:72bb2e0a] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 148, Loss: 0.0018\nIter: 2, Epoch: 148, Loss: 0.0006\nIter: 3, Epoch: 148, Loss: 0.0008\nIter: 4, Epoch: 148, Loss: 0.0009\nIter: 5, Epoch: 148, Loss: 0.0006\nIter: 6, Epoch: 148, Loss: 0.0006\nIter: 7, Epoch: 148, Loss: 0.0007\nIter: 8, Epoch: 148, Loss: 0.0012\nIter: 9, Epoch: 148, Loss: 0.0009\nIter: 10, Epoch: 148, Loss: 0.0008\nIter: 11, Epoch: 148, Loss: 0.0005\nIter: 12, Epoch: 148, Loss: 0.0005\nIter: 13, Epoch: 148, Loss: 0.0008\nIter: 14, Epoch: 148, Loss: 0.0008\nIter: 15, Epoch: 148, Loss: 0.0005\nIter: 16, Epoch: 148, Loss: 0.0007\nIter: 17, Epoch: 148, Loss: 0.0008\nIter: 18, Epoch: 148, Loss: 0.0009\nIter: 19, Epoch: 148, Loss: 0.0011\nIter: 20, Epoch: 148, Loss: 0.0008\nIter: 21, Epoch: 148, Loss: 0.0006\nIter: 22, Epoch: 148, Loss: 0.0009\nIter: 23, Epoch: 148, Loss: 0.0011\nIter: 24, Epoch: 148, Loss: 0.0006\nIter: 25, Epoch: 148, Loss: 0.0005\nIter: 26, Epoch: 148, Loss: 0.0008\nIter: 27, Epoch: 148, Loss: 0.0005\nIter: 28, Epoch: 148, Loss: 0.0009\nIter: 29, Epoch: 148, Loss: 0.0011\nIter: 30, Epoch: 148, Loss: 0.0006\n","truncated":false}} +%--- +%[output:23c9ef94] +% data: {"dataType":"text","outputData":{"text":">> Epoch 148 validation loss: 0.0012\n","truncated":false}} +%--- +%[output:51372b91] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 149, Loss: 0.0017\nIter: 2, Epoch: 149, Loss: 0.0006\nIter: 3, Epoch: 149, Loss: 0.0008\nIter: 4, Epoch: 149, Loss: 0.0009\nIter: 5, Epoch: 149, Loss: 0.0005\nIter: 6, Epoch: 149, Loss: 0.0006\nIter: 7, Epoch: 149, Loss: 0.0008\nIter: 8, Epoch: 149, Loss: 0.0012\nIter: 9, Epoch: 149, Loss: 0.0009\nIter: 10, Epoch: 149, Loss: 0.0008\nIter: 11, Epoch: 149, Loss: 0.0005\nIter: 12, Epoch: 149, Loss: 0.0005\nIter: 13, Epoch: 149, Loss: 0.0008\nIter: 14, Epoch: 149, Loss: 0.0008\nIter: 15, Epoch: 149, Loss: 0.0005\nIter: 16, Epoch: 149, Loss: 0.0006\nIter: 17, Epoch: 149, Loss: 0.0007\nIter: 18, Epoch: 149, Loss: 0.0009\nIter: 19, Epoch: 149, Loss: 0.0010\nIter: 20, Epoch: 149, Loss: 0.0008\nIter: 21, Epoch: 149, Loss: 0.0006\nIter: 22, Epoch: 149, Loss: 0.0009\nIter: 23, Epoch: 149, Loss: 0.0011\nIter: 24, Epoch: 149, Loss: 0.0006\nIter: 25, Epoch: 149, Loss: 0.0005\nIter: 26, Epoch: 149, Loss: 0.0008\nIter: 27, Epoch: 149, Loss: 0.0004\nIter: 28, Epoch: 149, Loss: 0.0008\nIter: 29, Epoch: 149, Loss: 0.0011\nIter: 30, Epoch: 149, Loss: 0.0006\n","truncated":false}} +%--- +%[output:89075115] +% data: {"dataType":"text","outputData":{"text":">> Epoch 149 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:0e0a21e9] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 150, Loss: 0.0016\nIter: 2, Epoch: 150, Loss: 0.0006\nIter: 3, Epoch: 150, Loss: 0.0009\nIter: 4, Epoch: 150, Loss: 0.0009\nIter: 5, Epoch: 150, Loss: 0.0005\nIter: 6, Epoch: 150, Loss: 0.0006\nIter: 7, Epoch: 150, Loss: 0.0008\nIter: 8, Epoch: 150, Loss: 0.0011\nIter: 9, Epoch: 150, Loss: 0.0009\nIter: 10, Epoch: 150, Loss: 0.0008\nIter: 11, Epoch: 150, Loss: 0.0005\nIter: 12, Epoch: 150, Loss: 0.0004\nIter: 13, Epoch: 150, Loss: 0.0007\nIter: 14, Epoch: 150, Loss: 0.0008\nIter: 15, Epoch: 150, Loss: 0.0005\nIter: 16, Epoch: 150, Loss: 0.0006\nIter: 17, Epoch: 150, Loss: 0.0007\nIter: 18, Epoch: 150, Loss: 0.0009\nIter: 19, Epoch: 150, Loss: 0.0010\nIter: 20, Epoch: 150, Loss: 0.0008\nIter: 21, Epoch: 150, Loss: 0.0006\nIter: 22, Epoch: 150, Loss: 0.0009\nIter: 23, Epoch: 150, Loss: 0.0010\nIter: 24, Epoch: 150, Loss: 0.0006\nIter: 25, Epoch: 150, Loss: 0.0005\nIter: 26, Epoch: 150, Loss: 0.0008\nIter: 27, Epoch: 150, Loss: 0.0004\nIter: 28, Epoch: 150, Loss: 0.0008\nIter: 29, Epoch: 150, Loss: 0.0011\nIter: 30, Epoch: 150, Loss: 0.0006\n","truncated":false}} +%--- +%[output:6dd7b792] +% data: {"dataType":"text","outputData":{"text":">> Epoch 150 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:1b40466f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 151, Loss: 0.0016\nIter: 2, Epoch: 151, Loss: 0.0006\nIter: 3, Epoch: 151, Loss: 0.0008\nIter: 4, Epoch: 151, Loss: 0.0009\nIter: 5, Epoch: 151, Loss: 0.0005\nIter: 6, Epoch: 151, Loss: 0.0006\nIter: 7, Epoch: 151, Loss: 0.0008\nIter: 8, Epoch: 151, Loss: 0.0011\nIter: 9, Epoch: 151, Loss: 0.0009\nIter: 10, Epoch: 151, Loss: 0.0008\nIter: 11, Epoch: 151, Loss: 0.0005\nIter: 12, Epoch: 151, Loss: 0.0004\nIter: 13, Epoch: 151, Loss: 0.0007\nIter: 14, Epoch: 151, Loss: 0.0008\nIter: 15, Epoch: 151, Loss: 0.0005\nIter: 16, Epoch: 151, Loss: 0.0006\nIter: 17, Epoch: 151, Loss: 0.0007\nIter: 18, Epoch: 151, Loss: 0.0009\nIter: 19, Epoch: 151, Loss: 0.0010\nIter: 20, Epoch: 151, Loss: 0.0008\nIter: 21, Epoch: 151, Loss: 0.0006\nIter: 22, Epoch: 151, Loss: 0.0009\nIter: 23, Epoch: 151, Loss: 0.0010\nIter: 24, Epoch: 151, Loss: 0.0005\nIter: 25, Epoch: 151, Loss: 0.0005\nIter: 26, Epoch: 151, Loss: 0.0007\nIter: 27, Epoch: 151, Loss: 0.0004\nIter: 28, Epoch: 151, Loss: 0.0008\nIter: 29, Epoch: 151, Loss: 0.0011\nIter: 30, Epoch: 151, Loss: 0.0006\n","truncated":false}} +%--- +%[output:93408908] +% data: {"dataType":"text","outputData":{"text":">> Epoch 151 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:9665fe6f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 152, Loss: 0.0016\nIter: 2, Epoch: 152, Loss: 0.0005\nIter: 3, Epoch: 152, Loss: 0.0008\nIter: 4, Epoch: 152, Loss: 0.0008\nIter: 5, Epoch: 152, Loss: 0.0005\nIter: 6, Epoch: 152, Loss: 0.0006\nIter: 7, Epoch: 152, Loss: 0.0008\nIter: 8, Epoch: 152, Loss: 0.0011\nIter: 9, Epoch: 152, Loss: 0.0009\nIter: 10, Epoch: 152, Loss: 0.0008\nIter: 11, Epoch: 152, Loss: 0.0005\nIter: 12, Epoch: 152, Loss: 0.0004\nIter: 13, Epoch: 152, Loss: 0.0007\nIter: 14, Epoch: 152, Loss: 0.0008\nIter: 15, Epoch: 152, Loss: 0.0005\nIter: 16, Epoch: 152, Loss: 0.0006\nIter: 17, Epoch: 152, Loss: 0.0007\nIter: 18, Epoch: 152, Loss: 0.0008\nIter: 19, Epoch: 152, Loss: 0.0010\nIter: 20, Epoch: 152, Loss: 0.0007\nIter: 21, Epoch: 152, Loss: 0.0006\nIter: 22, Epoch: 152, Loss: 0.0009\nIter: 23, Epoch: 152, Loss: 0.0009\nIter: 24, Epoch: 152, Loss: 0.0005\nIter: 25, Epoch: 152, Loss: 0.0005\nIter: 26, Epoch: 152, Loss: 0.0007\nIter: 27, Epoch: 152, Loss: 0.0004\nIter: 28, Epoch: 152, Loss: 0.0008\nIter: 29, Epoch: 152, Loss: 0.0011\nIter: 30, Epoch: 152, Loss: 0.0006\n","truncated":false}} +%--- +%[output:237859ab] +% data: {"dataType":"text","outputData":{"text":">> Epoch 152 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:79f52345] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 153, Loss: 0.0016\nIter: 2, Epoch: 153, Loss: 0.0005\nIter: 3, Epoch: 153, Loss: 0.0008\nIter: 4, Epoch: 153, Loss: 0.0008\nIter: 5, Epoch: 153, Loss: 0.0005\nIter: 6, Epoch: 153, Loss: 0.0006\nIter: 7, Epoch: 153, Loss: 0.0007\nIter: 8, Epoch: 153, Loss: 0.0011\nIter: 9, Epoch: 153, Loss: 0.0009\nIter: 10, Epoch: 153, Loss: 0.0008\nIter: 11, Epoch: 153, Loss: 0.0005\nIter: 12, Epoch: 153, Loss: 0.0004\nIter: 13, Epoch: 153, Loss: 0.0006\nIter: 14, Epoch: 153, Loss: 0.0008\nIter: 15, Epoch: 153, Loss: 0.0005\nIter: 16, Epoch: 153, Loss: 0.0006\nIter: 17, Epoch: 153, Loss: 0.0007\nIter: 18, Epoch: 153, Loss: 0.0008\nIter: 19, Epoch: 153, Loss: 0.0010\nIter: 20, Epoch: 153, Loss: 0.0007\nIter: 21, Epoch: 153, Loss: 0.0006\nIter: 22, Epoch: 153, Loss: 0.0009\nIter: 23, Epoch: 153, Loss: 0.0009\nIter: 24, Epoch: 153, Loss: 0.0005\nIter: 25, Epoch: 153, Loss: 0.0005\nIter: 26, Epoch: 153, Loss: 0.0007\nIter: 27, Epoch: 153, Loss: 0.0003\nIter: 28, Epoch: 153, Loss: 0.0008\nIter: 29, Epoch: 153, Loss: 0.0011\nIter: 30, Epoch: 153, Loss: 0.0006\n","truncated":false}} +%--- +%[output:25084d74] +% data: {"dataType":"text","outputData":{"text":">> Epoch 153 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:4c6e0a04] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 154, Loss: 0.0015\nIter: 2, Epoch: 154, Loss: 0.0005\nIter: 3, Epoch: 154, Loss: 0.0008\nIter: 4, Epoch: 154, Loss: 0.0008\nIter: 5, Epoch: 154, Loss: 0.0005\nIter: 6, Epoch: 154, Loss: 0.0006\nIter: 7, Epoch: 154, Loss: 0.0007\nIter: 8, Epoch: 154, Loss: 0.0011\nIter: 9, Epoch: 154, Loss: 0.0009\nIter: 10, Epoch: 154, Loss: 0.0007\nIter: 11, Epoch: 154, Loss: 0.0005\nIter: 12, Epoch: 154, Loss: 0.0004\nIter: 13, Epoch: 154, Loss: 0.0006\nIter: 14, Epoch: 154, Loss: 0.0008\nIter: 15, Epoch: 154, Loss: 0.0005\nIter: 16, Epoch: 154, Loss: 0.0006\nIter: 17, Epoch: 154, Loss: 0.0006\nIter: 18, Epoch: 154, Loss: 0.0008\nIter: 19, Epoch: 154, Loss: 0.0010\nIter: 20, Epoch: 154, Loss: 0.0007\nIter: 21, Epoch: 154, Loss: 0.0006\nIter: 22, Epoch: 154, Loss: 0.0009\nIter: 23, Epoch: 154, Loss: 0.0009\nIter: 24, Epoch: 154, Loss: 0.0005\nIter: 25, Epoch: 154, Loss: 0.0005\nIter: 26, Epoch: 154, Loss: 0.0007\nIter: 27, Epoch: 154, Loss: 0.0003\nIter: 28, Epoch: 154, Loss: 0.0008\nIter: 29, Epoch: 154, Loss: 0.0011\nIter: 30, Epoch: 154, Loss: 0.0006\n","truncated":false}} +%--- +%[output:299c9459] +% data: {"dataType":"text","outputData":{"text":">> Epoch 154 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:4a771709] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 155, Loss: 0.0015\nIter: 2, Epoch: 155, Loss: 0.0005\nIter: 3, Epoch: 155, Loss: 0.0008\nIter: 4, Epoch: 155, Loss: 0.0008\nIter: 5, Epoch: 155, Loss: 0.0005\nIter: 6, Epoch: 155, Loss: 0.0006\nIter: 7, Epoch: 155, Loss: 0.0007\nIter: 8, Epoch: 155, Loss: 0.0011\nIter: 9, Epoch: 155, Loss: 0.0009\nIter: 10, Epoch: 155, Loss: 0.0007\nIter: 11, Epoch: 155, Loss: 0.0005\nIter: 12, Epoch: 155, Loss: 0.0004\nIter: 13, Epoch: 155, Loss: 0.0006\nIter: 14, Epoch: 155, Loss: 0.0008\nIter: 15, Epoch: 155, Loss: 0.0005\nIter: 16, Epoch: 155, Loss: 0.0006\nIter: 17, Epoch: 155, Loss: 0.0006\nIter: 18, Epoch: 155, Loss: 0.0008\nIter: 19, Epoch: 155, Loss: 0.0010\nIter: 20, Epoch: 155, Loss: 0.0007\nIter: 21, Epoch: 155, Loss: 0.0006\nIter: 22, Epoch: 155, Loss: 0.0009\nIter: 23, Epoch: 155, Loss: 0.0009\nIter: 24, Epoch: 155, Loss: 0.0005\nIter: 25, Epoch: 155, Loss: 0.0005\nIter: 26, Epoch: 155, Loss: 0.0007\nIter: 27, Epoch: 155, Loss: 0.0003\nIter: 28, Epoch: 155, Loss: 0.0008\nIter: 29, Epoch: 155, Loss: 0.0011\nIter: 30, Epoch: 155, Loss: 0.0007\n","truncated":false}} +%--- +%[output:7b68406a] +% data: {"dataType":"text","outputData":{"text":">> Epoch 155 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:6f63bb9e] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 156, Loss: 0.0015\nIter: 2, Epoch: 156, Loss: 0.0005\nIter: 3, Epoch: 156, Loss: 0.0008\nIter: 4, Epoch: 156, Loss: 0.0008\nIter: 5, Epoch: 156, Loss: 0.0005\nIter: 6, Epoch: 156, Loss: 0.0006\nIter: 7, Epoch: 156, Loss: 0.0007\nIter: 8, Epoch: 156, Loss: 0.0011\nIter: 9, Epoch: 156, Loss: 0.0009\nIter: 10, Epoch: 156, Loss: 0.0007\nIter: 11, Epoch: 156, Loss: 0.0005\nIter: 12, Epoch: 156, Loss: 0.0004\nIter: 13, Epoch: 156, Loss: 0.0006\nIter: 14, Epoch: 156, Loss: 0.0008\nIter: 15, Epoch: 156, Loss: 0.0005\nIter: 16, Epoch: 156, Loss: 0.0006\nIter: 17, Epoch: 156, Loss: 0.0006\nIter: 18, Epoch: 156, Loss: 0.0008\nIter: 19, Epoch: 156, Loss: 0.0010\nIter: 20, Epoch: 156, Loss: 0.0007\nIter: 21, Epoch: 156, Loss: 0.0006\nIter: 22, Epoch: 156, Loss: 0.0008\nIter: 23, Epoch: 156, Loss: 0.0009\nIter: 24, Epoch: 156, Loss: 0.0004\nIter: 25, Epoch: 156, Loss: 0.0005\nIter: 26, Epoch: 156, Loss: 0.0007\nIter: 27, Epoch: 156, Loss: 0.0003\nIter: 28, Epoch: 156, Loss: 0.0008\nIter: 29, Epoch: 156, Loss: 0.0010\nIter: 30, Epoch: 156, Loss: 0.0007\n","truncated":false}} +%--- +%[output:0a38329a] +% data: {"dataType":"text","outputData":{"text":">> Epoch 156 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:5beff449] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 157, Loss: 0.0015\nIter: 2, Epoch: 157, Loss: 0.0005\nIter: 3, Epoch: 157, Loss: 0.0008\nIter: 4, Epoch: 157, Loss: 0.0008\nIter: 5, Epoch: 157, Loss: 0.0004\nIter: 6, Epoch: 157, Loss: 0.0006\nIter: 7, Epoch: 157, Loss: 0.0007\nIter: 8, Epoch: 157, Loss: 0.0011\nIter: 9, Epoch: 157, Loss: 0.0009\nIter: 10, Epoch: 157, Loss: 0.0007\nIter: 11, Epoch: 157, Loss: 0.0005\nIter: 12, Epoch: 157, Loss: 0.0004\nIter: 13, Epoch: 157, Loss: 0.0006\nIter: 14, Epoch: 157, Loss: 0.0008\nIter: 15, Epoch: 157, Loss: 0.0005\nIter: 16, Epoch: 157, Loss: 0.0006\nIter: 17, Epoch: 157, Loss: 0.0006\nIter: 18, Epoch: 157, Loss: 0.0008\nIter: 19, Epoch: 157, Loss: 0.0010\nIter: 20, Epoch: 157, Loss: 0.0007\nIter: 21, Epoch: 157, Loss: 0.0005\nIter: 22, Epoch: 157, Loss: 0.0008\nIter: 23, Epoch: 157, Loss: 0.0009\nIter: 24, Epoch: 157, Loss: 0.0004\nIter: 25, Epoch: 157, Loss: 0.0005\nIter: 26, Epoch: 157, Loss: 0.0007\nIter: 27, Epoch: 157, Loss: 0.0003\nIter: 28, Epoch: 157, Loss: 0.0008\nIter: 29, Epoch: 157, Loss: 0.0010\nIter: 30, Epoch: 157, Loss: 0.0007\n","truncated":false}} +%--- +%[output:8dc0adc2] +% data: {"dataType":"text","outputData":{"text":">> Epoch 157 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:7dbbd20d] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 158, Loss: 0.0015\nIter: 2, Epoch: 158, Loss: 0.0005\nIter: 3, Epoch: 158, Loss: 0.0007\nIter: 4, Epoch: 158, Loss: 0.0008\nIter: 5, Epoch: 158, Loss: 0.0004\nIter: 6, Epoch: 158, Loss: 0.0005\nIter: 7, Epoch: 158, Loss: 0.0007\nIter: 8, Epoch: 158, Loss: 0.0011\nIter: 9, Epoch: 158, Loss: 0.0009\nIter: 10, Epoch: 158, Loss: 0.0007\nIter: 11, Epoch: 158, Loss: 0.0005\nIter: 12, Epoch: 158, Loss: 0.0004\nIter: 13, Epoch: 158, Loss: 0.0006\nIter: 14, Epoch: 158, Loss: 0.0008\nIter: 15, Epoch: 158, Loss: 0.0005\nIter: 16, Epoch: 158, Loss: 0.0006\nIter: 17, Epoch: 158, Loss: 0.0006\nIter: 18, Epoch: 158, Loss: 0.0007\nIter: 19, Epoch: 158, Loss: 0.0010\nIter: 20, Epoch: 158, Loss: 0.0007\nIter: 21, Epoch: 158, Loss: 0.0005\nIter: 22, Epoch: 158, Loss: 0.0008\nIter: 23, Epoch: 158, Loss: 0.0009\nIter: 24, Epoch: 158, Loss: 0.0004\nIter: 25, Epoch: 158, Loss: 0.0005\nIter: 26, Epoch: 158, Loss: 0.0007\nIter: 27, Epoch: 158, Loss: 0.0003\nIter: 28, Epoch: 158, Loss: 0.0007\nIter: 29, Epoch: 158, Loss: 0.0010\nIter: 30, Epoch: 158, Loss: 0.0007\n","truncated":false}} +%--- +%[output:4e3bedf4] +% data: {"dataType":"text","outputData":{"text":">> Epoch 158 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:2685e72a] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 159, Loss: 0.0015\nIter: 2, Epoch: 159, Loss: 0.0005\nIter: 3, Epoch: 159, Loss: 0.0007\nIter: 4, Epoch: 159, Loss: 0.0008\nIter: 5, Epoch: 159, Loss: 0.0004\nIter: 6, Epoch: 159, Loss: 0.0005\nIter: 7, Epoch: 159, Loss: 0.0007\nIter: 8, Epoch: 159, Loss: 0.0011\nIter: 9, Epoch: 159, Loss: 0.0009\nIter: 10, Epoch: 159, Loss: 0.0007\nIter: 11, Epoch: 159, Loss: 0.0005\nIter: 12, Epoch: 159, Loss: 0.0004\nIter: 13, Epoch: 159, Loss: 0.0006\nIter: 14, Epoch: 159, Loss: 0.0007\nIter: 15, Epoch: 159, Loss: 0.0006\nIter: 16, Epoch: 159, Loss: 0.0006\nIter: 17, Epoch: 159, Loss: 0.0006\nIter: 18, Epoch: 159, Loss: 0.0007\nIter: 19, Epoch: 159, Loss: 0.0010\nIter: 20, Epoch: 159, Loss: 0.0007\nIter: 21, Epoch: 159, Loss: 0.0005\nIter: 22, Epoch: 159, Loss: 0.0008\nIter: 23, Epoch: 159, Loss: 0.0009\nIter: 24, Epoch: 159, Loss: 0.0004\nIter: 25, Epoch: 159, Loss: 0.0005\nIter: 26, Epoch: 159, Loss: 0.0007\nIter: 27, Epoch: 159, Loss: 0.0003\nIter: 28, Epoch: 159, Loss: 0.0007\nIter: 29, Epoch: 159, Loss: 0.0010\nIter: 30, Epoch: 159, Loss: 0.0007\n","truncated":false}} +%--- +%[output:7eb29abe] +% data: {"dataType":"text","outputData":{"text":">> Epoch 159 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:707a11a3] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 160, Loss: 0.0015\nIter: 2, Epoch: 160, Loss: 0.0005\nIter: 3, Epoch: 160, Loss: 0.0007\nIter: 4, Epoch: 160, Loss: 0.0008\nIter: 5, Epoch: 160, Loss: 0.0004\nIter: 6, Epoch: 160, Loss: 0.0005\nIter: 7, Epoch: 160, Loss: 0.0007\nIter: 8, Epoch: 160, Loss: 0.0011\nIter: 9, Epoch: 160, Loss: 0.0009\nIter: 10, Epoch: 160, Loss: 0.0007\nIter: 11, Epoch: 160, Loss: 0.0005\nIter: 12, Epoch: 160, Loss: 0.0004\nIter: 13, Epoch: 160, Loss: 0.0006\nIter: 14, Epoch: 160, Loss: 0.0007\nIter: 15, Epoch: 160, Loss: 0.0006\nIter: 16, Epoch: 160, Loss: 0.0006\nIter: 17, Epoch: 160, Loss: 0.0006\nIter: 18, Epoch: 160, Loss: 0.0007\nIter: 19, Epoch: 160, Loss: 0.0010\nIter: 20, Epoch: 160, Loss: 0.0007\nIter: 21, Epoch: 160, Loss: 0.0005\nIter: 22, Epoch: 160, Loss: 0.0008\nIter: 23, Epoch: 160, Loss: 0.0008\nIter: 24, Epoch: 160, Loss: 0.0004\nIter: 25, Epoch: 160, Loss: 0.0005\nIter: 26, Epoch: 160, Loss: 0.0007\nIter: 27, Epoch: 160, Loss: 0.0003\nIter: 28, Epoch: 160, Loss: 0.0008\nIter: 29, Epoch: 160, Loss: 0.0010\nIter: 30, Epoch: 160, Loss: 0.0007\n","truncated":false}} +%--- +%[output:45e3cc9c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 160 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:8ec25ece] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 161, Loss: 0.0014\nIter: 2, Epoch: 161, Loss: 0.0005\nIter: 3, Epoch: 161, Loss: 0.0007\nIter: 4, Epoch: 161, Loss: 0.0008\nIter: 5, Epoch: 161, Loss: 0.0004\nIter: 6, Epoch: 161, Loss: 0.0005\nIter: 7, Epoch: 161, Loss: 0.0007\nIter: 8, Epoch: 161, Loss: 0.0011\nIter: 9, Epoch: 161, Loss: 0.0008\nIter: 10, Epoch: 161, Loss: 0.0007\nIter: 11, Epoch: 161, Loss: 0.0005\nIter: 12, Epoch: 161, Loss: 0.0004\nIter: 13, Epoch: 161, Loss: 0.0006\nIter: 14, Epoch: 161, Loss: 0.0007\nIter: 15, Epoch: 161, Loss: 0.0006\nIter: 16, Epoch: 161, Loss: 0.0006\nIter: 17, Epoch: 161, Loss: 0.0006\nIter: 18, Epoch: 161, Loss: 0.0007\nIter: 19, Epoch: 161, Loss: 0.0010\nIter: 20, Epoch: 161, Loss: 0.0007\nIter: 21, Epoch: 161, Loss: 0.0005\nIter: 22, Epoch: 161, Loss: 0.0008\nIter: 23, Epoch: 161, Loss: 0.0008\nIter: 24, Epoch: 161, Loss: 0.0004\nIter: 25, Epoch: 161, Loss: 0.0004\nIter: 26, Epoch: 161, Loss: 0.0007\nIter: 27, Epoch: 161, Loss: 0.0003\nIter: 28, Epoch: 161, Loss: 0.0007\nIter: 29, Epoch: 161, Loss: 0.0010\nIter: 30, Epoch: 161, Loss: 0.0007\n","truncated":false}} +%--- +%[output:7521c589] +% data: {"dataType":"text","outputData":{"text":">> Epoch 161 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:75ab77b0] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 162, Loss: 0.0014\nIter: 2, Epoch: 162, Loss: 0.0005\nIter: 3, Epoch: 162, Loss: 0.0007\nIter: 4, Epoch: 162, Loss: 0.0008\nIter: 5, Epoch: 162, Loss: 0.0005\nIter: 6, Epoch: 162, Loss: 0.0005\nIter: 7, Epoch: 162, Loss: 0.0007\nIter: 8, Epoch: 162, Loss: 0.0011\nIter: 9, Epoch: 162, Loss: 0.0008\nIter: 10, Epoch: 162, Loss: 0.0007\nIter: 11, Epoch: 162, Loss: 0.0005\nIter: 12, Epoch: 162, Loss: 0.0004\nIter: 13, Epoch: 162, Loss: 0.0006\nIter: 14, Epoch: 162, Loss: 0.0007\nIter: 15, Epoch: 162, Loss: 0.0006\nIter: 16, Epoch: 162, Loss: 0.0006\nIter: 17, Epoch: 162, Loss: 0.0006\nIter: 18, Epoch: 162, Loss: 0.0007\nIter: 19, Epoch: 162, Loss: 0.0010\nIter: 20, Epoch: 162, Loss: 0.0007\nIter: 21, Epoch: 162, Loss: 0.0005\nIter: 22, Epoch: 162, Loss: 0.0008\nIter: 23, Epoch: 162, Loss: 0.0009\nIter: 24, Epoch: 162, Loss: 0.0004\nIter: 25, Epoch: 162, Loss: 0.0004\nIter: 26, Epoch: 162, Loss: 0.0007\nIter: 27, Epoch: 162, Loss: 0.0003\nIter: 28, Epoch: 162, Loss: 0.0008\nIter: 29, Epoch: 162, Loss: 0.0010\nIter: 30, Epoch: 162, Loss: 0.0007\n","truncated":false}} +%--- +%[output:156a892d] +% data: {"dataType":"text","outputData":{"text":">> Epoch 162 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:8a991d2b] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 163, Loss: 0.0014\nIter: 2, Epoch: 163, Loss: 0.0005\nIter: 3, Epoch: 163, Loss: 0.0007\nIter: 4, Epoch: 163, Loss: 0.0008\nIter: 5, Epoch: 163, Loss: 0.0005\nIter: 6, Epoch: 163, Loss: 0.0005\nIter: 7, Epoch: 163, Loss: 0.0007\nIter: 8, Epoch: 163, Loss: 0.0011\nIter: 9, Epoch: 163, Loss: 0.0008\nIter: 10, Epoch: 163, Loss: 0.0007\nIter: 11, Epoch: 163, Loss: 0.0005\nIter: 12, Epoch: 163, Loss: 0.0004\nIter: 13, Epoch: 163, Loss: 0.0006\nIter: 14, Epoch: 163, Loss: 0.0007\nIter: 15, Epoch: 163, Loss: 0.0006\nIter: 16, Epoch: 163, Loss: 0.0006\nIter: 17, Epoch: 163, Loss: 0.0006\nIter: 18, Epoch: 163, Loss: 0.0007\nIter: 19, Epoch: 163, Loss: 0.0010\nIter: 20, Epoch: 163, Loss: 0.0007\nIter: 21, Epoch: 163, Loss: 0.0004\nIter: 22, Epoch: 163, Loss: 0.0008\nIter: 23, Epoch: 163, Loss: 0.0009\nIter: 24, Epoch: 163, Loss: 0.0004\nIter: 25, Epoch: 163, Loss: 0.0004\nIter: 26, Epoch: 163, Loss: 0.0006\nIter: 27, Epoch: 163, Loss: 0.0003\nIter: 28, Epoch: 163, Loss: 0.0008\nIter: 29, Epoch: 163, Loss: 0.0010\nIter: 30, Epoch: 163, Loss: 0.0007\n","truncated":false}} +%--- +%[output:67da58b4] +% data: {"dataType":"text","outputData":{"text":">> Epoch 163 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:081f865d] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 164, Loss: 0.0014\nIter: 2, Epoch: 164, Loss: 0.0005\nIter: 3, Epoch: 164, Loss: 0.0007\nIter: 4, Epoch: 164, Loss: 0.0008\nIter: 5, Epoch: 164, Loss: 0.0005\nIter: 6, Epoch: 164, Loss: 0.0005\nIter: 7, Epoch: 164, Loss: 0.0006\nIter: 8, Epoch: 164, Loss: 0.0011\nIter: 9, Epoch: 164, Loss: 0.0008\nIter: 10, Epoch: 164, Loss: 0.0006\nIter: 11, Epoch: 164, Loss: 0.0005\nIter: 12, Epoch: 164, Loss: 0.0004\nIter: 13, Epoch: 164, Loss: 0.0006\nIter: 14, Epoch: 164, Loss: 0.0007\nIter: 15, Epoch: 164, Loss: 0.0006\nIter: 16, Epoch: 164, Loss: 0.0006\nIter: 17, Epoch: 164, Loss: 0.0006\nIter: 18, Epoch: 164, Loss: 0.0007\nIter: 19, Epoch: 164, Loss: 0.0010\nIter: 20, Epoch: 164, Loss: 0.0007\nIter: 21, Epoch: 164, Loss: 0.0004\nIter: 22, Epoch: 164, Loss: 0.0008\nIter: 23, Epoch: 164, Loss: 0.0009\nIter: 24, Epoch: 164, Loss: 0.0004\nIter: 25, Epoch: 164, Loss: 0.0004\nIter: 26, Epoch: 164, Loss: 0.0006\nIter: 27, Epoch: 164, Loss: 0.0003\nIter: 28, Epoch: 164, Loss: 0.0008\nIter: 29, Epoch: 164, Loss: 0.0010\nIter: 30, Epoch: 164, Loss: 0.0007\n","truncated":false}} +%--- +%[output:0591c023] +% data: {"dataType":"text","outputData":{"text":">> Epoch 164 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:75bfbbed] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 165, Loss: 0.0014\nIter: 2, Epoch: 165, Loss: 0.0006\nIter: 3, Epoch: 165, Loss: 0.0007\nIter: 4, Epoch: 165, Loss: 0.0008\nIter: 5, Epoch: 165, Loss: 0.0005\nIter: 6, Epoch: 165, Loss: 0.0005\nIter: 7, Epoch: 165, Loss: 0.0006\nIter: 8, Epoch: 165, Loss: 0.0011\nIter: 9, Epoch: 165, Loss: 0.0008\nIter: 10, Epoch: 165, Loss: 0.0006\nIter: 11, Epoch: 165, Loss: 0.0004\nIter: 12, Epoch: 165, Loss: 0.0004\nIter: 13, Epoch: 165, Loss: 0.0007\nIter: 14, Epoch: 165, Loss: 0.0007\nIter: 15, Epoch: 165, Loss: 0.0005\nIter: 16, Epoch: 165, Loss: 0.0006\nIter: 17, Epoch: 165, Loss: 0.0006\nIter: 18, Epoch: 165, Loss: 0.0007\nIter: 19, Epoch: 165, Loss: 0.0010\nIter: 20, Epoch: 165, Loss: 0.0007\nIter: 21, Epoch: 165, Loss: 0.0004\nIter: 22, Epoch: 165, Loss: 0.0008\nIter: 23, Epoch: 165, Loss: 0.0009\nIter: 24, Epoch: 165, Loss: 0.0004\nIter: 25, Epoch: 165, Loss: 0.0004\nIter: 26, Epoch: 165, Loss: 0.0007\nIter: 27, Epoch: 165, Loss: 0.0003\nIter: 28, Epoch: 165, Loss: 0.0008\nIter: 29, Epoch: 165, Loss: 0.0010\nIter: 30, Epoch: 165, Loss: 0.0007\n","truncated":false}} +%--- +%[output:44f26859] +% data: {"dataType":"text","outputData":{"text":">> Epoch 165 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:501736cc] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 166, Loss: 0.0014\nIter: 2, Epoch: 166, Loss: 0.0005\nIter: 3, Epoch: 166, Loss: 0.0007\nIter: 4, Epoch: 166, Loss: 0.0008\nIter: 5, Epoch: 166, Loss: 0.0005\nIter: 6, Epoch: 166, Loss: 0.0005\nIter: 7, Epoch: 166, Loss: 0.0006\nIter: 8, Epoch: 166, Loss: 0.0011\nIter: 9, Epoch: 166, Loss: 0.0009\nIter: 10, Epoch: 166, Loss: 0.0006\nIter: 11, Epoch: 166, Loss: 0.0004\nIter: 12, Epoch: 166, Loss: 0.0004\nIter: 13, Epoch: 166, Loss: 0.0007\nIter: 14, Epoch: 166, Loss: 0.0007\nIter: 15, Epoch: 166, Loss: 0.0005\nIter: 16, Epoch: 166, Loss: 0.0006\nIter: 17, Epoch: 166, Loss: 0.0006\nIter: 18, Epoch: 166, Loss: 0.0007\nIter: 19, Epoch: 166, Loss: 0.0009\nIter: 20, Epoch: 166, Loss: 0.0007\nIter: 21, Epoch: 166, Loss: 0.0004\nIter: 22, Epoch: 166, Loss: 0.0008\nIter: 23, Epoch: 166, Loss: 0.0009\nIter: 24, Epoch: 166, Loss: 0.0004\nIter: 25, Epoch: 166, Loss: 0.0004\nIter: 26, Epoch: 166, Loss: 0.0007\nIter: 27, Epoch: 166, Loss: 0.0003\nIter: 28, Epoch: 166, Loss: 0.0008\nIter: 29, Epoch: 166, Loss: 0.0010\nIter: 30, Epoch: 166, Loss: 0.0007\n","truncated":false}} +%--- +%[output:92a3ee44] +% data: {"dataType":"text","outputData":{"text":">> Epoch 166 validation loss: 0.0009\n","truncated":false}} +%--- +%[output:699cbdea] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 167, Loss: 0.0014\nIter: 2, Epoch: 167, Loss: 0.0005\nIter: 3, Epoch: 167, Loss: 0.0007\nIter: 4, Epoch: 167, Loss: 0.0007\nIter: 5, Epoch: 167, Loss: 0.0005\nIter: 6, Epoch: 167, Loss: 0.0005\nIter: 7, Epoch: 167, Loss: 0.0006\nIter: 8, Epoch: 167, Loss: 0.0011\nIter: 9, Epoch: 167, Loss: 0.0008\nIter: 10, Epoch: 167, Loss: 0.0006\nIter: 11, Epoch: 167, Loss: 0.0004\nIter: 12, Epoch: 167, Loss: 0.0004\nIter: 13, Epoch: 167, Loss: 0.0007\nIter: 14, Epoch: 167, Loss: 0.0008\nIter: 15, Epoch: 167, Loss: 0.0005\nIter: 16, Epoch: 167, Loss: 0.0006\nIter: 17, Epoch: 167, Loss: 0.0006\nIter: 18, Epoch: 167, Loss: 0.0006\nIter: 19, Epoch: 167, Loss: 0.0009\nIter: 20, Epoch: 167, Loss: 0.0007\nIter: 21, Epoch: 167, Loss: 0.0004\nIter: 22, Epoch: 167, Loss: 0.0008\nIter: 23, Epoch: 167, Loss: 0.0009\nIter: 24, Epoch: 167, Loss: 0.0004\nIter: 25, Epoch: 167, Loss: 0.0004\nIter: 26, Epoch: 167, Loss: 0.0007\nIter: 27, Epoch: 167, Loss: 0.0003\nIter: 28, Epoch: 167, Loss: 0.0008\nIter: 29, Epoch: 167, Loss: 0.0010\nIter: 30, Epoch: 167, Loss: 0.0007\n","truncated":false}} +%--- +%[output:6523e727] +% data: {"dataType":"text","outputData":{"text":">> Epoch 167 validation loss: 0.0009\n","truncated":false}} +%--- +%[output:3dd4bc93] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 168, Loss: 0.0014\nIter: 2, Epoch: 168, Loss: 0.0006\nIter: 3, Epoch: 168, Loss: 0.0006\nIter: 4, Epoch: 168, Loss: 0.0007\nIter: 5, Epoch: 168, Loss: 0.0005\nIter: 6, Epoch: 168, Loss: 0.0005\nIter: 7, Epoch: 168, Loss: 0.0006\nIter: 8, Epoch: 168, Loss: 0.0011\nIter: 9, Epoch: 168, Loss: 0.0008\nIter: 10, Epoch: 168, Loss: 0.0006\nIter: 11, Epoch: 168, Loss: 0.0004\nIter: 12, Epoch: 168, Loss: 0.0004\nIter: 13, Epoch: 168, Loss: 0.0006\nIter: 14, Epoch: 168, Loss: 0.0007\nIter: 15, Epoch: 168, Loss: 0.0005\nIter: 16, Epoch: 168, Loss: 0.0006\nIter: 17, Epoch: 168, Loss: 0.0006\nIter: 18, Epoch: 168, Loss: 0.0006\nIter: 19, Epoch: 168, Loss: 0.0009\nIter: 20, Epoch: 168, Loss: 0.0007\nIter: 21, Epoch: 168, Loss: 0.0004\nIter: 22, Epoch: 168, Loss: 0.0008\nIter: 23, Epoch: 168, Loss: 0.0009\nIter: 24, Epoch: 168, Loss: 0.0004\nIter: 25, Epoch: 168, Loss: 0.0004\nIter: 26, Epoch: 168, Loss: 0.0006\nIter: 27, Epoch: 168, Loss: 0.0003\nIter: 28, Epoch: 168, Loss: 0.0007\nIter: 29, Epoch: 168, Loss: 0.0010\nIter: 30, Epoch: 168, Loss: 0.0007\n","truncated":false}} +%--- +%[output:51305e72] +% data: {"dataType":"text","outputData":{"text":">> Epoch 168 validation loss: 0.0009\n","truncated":false}} +%--- +%[output:6cde6acf] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 169, Loss: 0.0014\nIter: 2, Epoch: 169, Loss: 0.0006\nIter: 3, Epoch: 169, Loss: 0.0006\nIter: 4, Epoch: 169, Loss: 0.0007\nIter: 5, Epoch: 169, Loss: 0.0004\nIter: 6, Epoch: 169, Loss: 0.0005\nIter: 7, Epoch: 169, Loss: 0.0006\nIter: 8, Epoch: 169, Loss: 0.0011\nIter: 9, Epoch: 169, Loss: 0.0008\nIter: 10, Epoch: 169, Loss: 0.0006\nIter: 11, Epoch: 169, Loss: 0.0004\nIter: 12, Epoch: 169, Loss: 0.0004\nIter: 13, Epoch: 169, Loss: 0.0006\nIter: 14, Epoch: 169, Loss: 0.0008\nIter: 15, Epoch: 169, Loss: 0.0005\nIter: 16, Epoch: 169, Loss: 0.0006\nIter: 17, Epoch: 169, Loss: 0.0006\nIter: 18, Epoch: 169, Loss: 0.0006\nIter: 19, Epoch: 169, Loss: 0.0009\nIter: 20, Epoch: 169, Loss: 0.0007\nIter: 21, Epoch: 169, Loss: 0.0004\nIter: 22, Epoch: 169, Loss: 0.0008\nIter: 23, Epoch: 169, Loss: 0.0009\nIter: 24, Epoch: 169, Loss: 0.0004\nIter: 25, Epoch: 169, Loss: 0.0004\nIter: 26, Epoch: 169, Loss: 0.0006\nIter: 27, Epoch: 169, Loss: 0.0003\nIter: 28, Epoch: 169, Loss: 0.0007\nIter: 29, Epoch: 169, Loss: 0.0010\nIter: 30, Epoch: 169, Loss: 0.0007\n","truncated":false}} +%--- +%[output:2e8f6998] +% data: {"dataType":"text","outputData":{"text":">> Epoch 169 validation loss: 0.0009\n","truncated":false}} +%--- +%[output:9a1bc941] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 170, Loss: 0.0014\nIter: 2, Epoch: 170, Loss: 0.0005\nIter: 3, Epoch: 170, Loss: 0.0006\nIter: 4, Epoch: 170, Loss: 0.0007\nIter: 5, Epoch: 170, Loss: 0.0004\nIter: 6, Epoch: 170, Loss: 0.0005\nIter: 7, Epoch: 170, Loss: 0.0006\nIter: 8, Epoch: 170, Loss: 0.0011\nIter: 9, Epoch: 170, Loss: 0.0008\nIter: 10, Epoch: 170, Loss: 0.0006\nIter: 11, Epoch: 170, Loss: 0.0004\nIter: 12, Epoch: 170, Loss: 0.0004\nIter: 13, Epoch: 170, Loss: 0.0006\nIter: 14, Epoch: 170, Loss: 0.0008\nIter: 15, Epoch: 170, Loss: 0.0005\nIter: 16, Epoch: 170, Loss: 0.0006\nIter: 17, Epoch: 170, Loss: 0.0006\nIter: 18, Epoch: 170, Loss: 0.0006\nIter: 19, Epoch: 170, Loss: 0.0008\nIter: 20, Epoch: 170, Loss: 0.0007\nIter: 21, Epoch: 170, Loss: 0.0004\nIter: 22, Epoch: 170, Loss: 0.0008\nIter: 23, Epoch: 170, Loss: 0.0009\nIter: 24, Epoch: 170, Loss: 0.0004\nIter: 25, Epoch: 170, Loss: 0.0004\nIter: 26, Epoch: 170, Loss: 0.0006\nIter: 27, Epoch: 170, Loss: 0.0003\nIter: 28, Epoch: 170, Loss: 0.0007\nIter: 29, Epoch: 170, Loss: 0.0010\nIter: 30, Epoch: 170, Loss: 0.0007\n","truncated":false}} +%--- +%[output:7cb89f58] +% data: {"dataType":"text","outputData":{"text":">> Epoch 170 validation loss: 0.0009\n","truncated":false}} +%--- +%[output:3afe39b3] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 171, Loss: 0.0014\nIter: 2, Epoch: 171, Loss: 0.0005\nIter: 3, Epoch: 171, Loss: 0.0006\nIter: 4, Epoch: 171, Loss: 0.0007\nIter: 5, Epoch: 171, Loss: 0.0004\nIter: 6, Epoch: 171, Loss: 0.0005\nIter: 7, Epoch: 171, Loss: 0.0006\nIter: 8, Epoch: 171, Loss: 0.0011\nIter: 9, Epoch: 171, Loss: 0.0008\nIter: 10, Epoch: 171, Loss: 0.0006\nIter: 11, Epoch: 171, Loss: 0.0004\nIter: 12, Epoch: 171, Loss: 0.0004\nIter: 13, Epoch: 171, Loss: 0.0006\nIter: 14, Epoch: 171, Loss: 0.0008\nIter: 15, Epoch: 171, Loss: 0.0005\nIter: 16, Epoch: 171, Loss: 0.0006\nIter: 17, Epoch: 171, Loss: 0.0005\nIter: 18, Epoch: 171, Loss: 0.0006\nIter: 19, Epoch: 171, Loss: 0.0008\nIter: 20, Epoch: 171, Loss: 0.0007\nIter: 21, Epoch: 171, Loss: 0.0004\nIter: 22, Epoch: 171, Loss: 0.0007\nIter: 23, Epoch: 171, Loss: 0.0008\nIter: 24, Epoch: 171, Loss: 0.0004\nIter: 25, Epoch: 171, Loss: 0.0004\nIter: 26, Epoch: 171, Loss: 0.0006\nIter: 27, Epoch: 171, Loss: 0.0003\nIter: 28, Epoch: 171, Loss: 0.0007\nIter: 29, Epoch: 171, Loss: 0.0010\nIter: 30, Epoch: 171, Loss: 0.0007\n","truncated":false}} +%--- +%[output:93217151] +% data: {"dataType":"text","outputData":{"text":">> Epoch 171 validation loss: 0.0009\n","truncated":false}} +%--- +%[output:1684ec2d] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 172, Loss: 0.0013\nIter: 2, Epoch: 172, Loss: 0.0005\nIter: 3, Epoch: 172, Loss: 0.0006\nIter: 4, Epoch: 172, Loss: 0.0007\nIter: 5, Epoch: 172, Loss: 0.0004\nIter: 6, Epoch: 172, Loss: 0.0005\nIter: 7, Epoch: 172, Loss: 0.0006\nIter: 8, Epoch: 172, Loss: 0.0011\nIter: 9, Epoch: 172, Loss: 0.0008\nIter: 10, Epoch: 172, Loss: 0.0006\nIter: 11, Epoch: 172, Loss: 0.0004\nIter: 12, Epoch: 172, Loss: 0.0003\nIter: 13, Epoch: 172, Loss: 0.0006\nIter: 14, Epoch: 172, Loss: 0.0008\nIter: 15, Epoch: 172, Loss: 0.0005\nIter: 16, Epoch: 172, Loss: 0.0006\nIter: 17, Epoch: 172, Loss: 0.0005\nIter: 18, Epoch: 172, Loss: 0.0006\nIter: 19, Epoch: 172, Loss: 0.0008\nIter: 20, Epoch: 172, Loss: 0.0007\nIter: 21, Epoch: 172, Loss: 0.0004\nIter: 22, Epoch: 172, Loss: 0.0008\nIter: 23, Epoch: 172, Loss: 0.0008\nIter: 24, Epoch: 172, Loss: 0.0004\nIter: 25, Epoch: 172, Loss: 0.0004\nIter: 26, Epoch: 172, Loss: 0.0006\nIter: 27, Epoch: 172, Loss: 0.0003\nIter: 28, Epoch: 172, Loss: 0.0007\nIter: 29, Epoch: 172, Loss: 0.0010\nIter: 30, Epoch: 172, Loss: 0.0007\n","truncated":false}} +%--- +%[output:92e234ed] +% data: {"dataType":"text","outputData":{"text":">> Epoch 172 validation loss: 0.0009\n","truncated":false}} +%--- +%[output:5f1a718a] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 173, Loss: 0.0013\nIter: 2, Epoch: 173, Loss: 0.0005\nIter: 3, Epoch: 173, Loss: 0.0006\nIter: 4, Epoch: 173, Loss: 0.0007\nIter: 5, Epoch: 173, Loss: 0.0004\nIter: 6, Epoch: 173, Loss: 0.0005\nIter: 7, Epoch: 173, Loss: 0.0005\nIter: 8, Epoch: 173, Loss: 0.0011\nIter: 9, Epoch: 173, Loss: 0.0008\nIter: 10, Epoch: 173, Loss: 0.0006\nIter: 11, Epoch: 173, Loss: 0.0004\nIter: 12, Epoch: 173, Loss: 0.0003\nIter: 13, Epoch: 173, Loss: 0.0006\nIter: 14, Epoch: 173, Loss: 0.0008\nIter: 15, Epoch: 173, Loss: 0.0005\nIter: 16, Epoch: 173, Loss: 0.0006\nIter: 17, Epoch: 173, Loss: 0.0005\nIter: 18, Epoch: 173, Loss: 0.0006\nIter: 19, Epoch: 173, Loss: 0.0008\nIter: 20, Epoch: 173, Loss: 0.0007\nIter: 21, Epoch: 173, Loss: 0.0004\nIter: 22, Epoch: 173, Loss: 0.0007\nIter: 23, Epoch: 173, Loss: 0.0008\nIter: 24, Epoch: 173, Loss: 0.0004\nIter: 25, Epoch: 173, Loss: 0.0004\nIter: 26, Epoch: 173, Loss: 0.0006\nIter: 27, Epoch: 173, Loss: 0.0003\nIter: 28, Epoch: 173, Loss: 0.0007\nIter: 29, Epoch: 173, Loss: 0.0010\nIter: 30, Epoch: 173, Loss: 0.0007\n","truncated":false}} +%--- +%[output:519c44e7] +% data: {"dataType":"text","outputData":{"text":">> Epoch 173 validation loss: 0.0009\n","truncated":false}} +%--- +%[output:49b33c6e] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 174, Loss: 0.0013\nIter: 2, Epoch: 174, Loss: 0.0005\nIter: 3, Epoch: 174, Loss: 0.0006\nIter: 4, Epoch: 174, Loss: 0.0007\nIter: 5, Epoch: 174, Loss: 0.0004\nIter: 6, Epoch: 174, Loss: 0.0005\nIter: 7, Epoch: 174, Loss: 0.0005\nIter: 8, Epoch: 174, Loss: 0.0011\nIter: 9, Epoch: 174, Loss: 0.0008\nIter: 10, Epoch: 174, Loss: 0.0006\nIter: 11, Epoch: 174, Loss: 0.0004\nIter: 12, Epoch: 174, Loss: 0.0003\nIter: 13, Epoch: 174, Loss: 0.0006\nIter: 14, Epoch: 174, Loss: 0.0008\nIter: 15, Epoch: 174, Loss: 0.0005\nIter: 16, Epoch: 174, Loss: 0.0006\nIter: 17, Epoch: 174, Loss: 0.0005\nIter: 18, Epoch: 174, Loss: 0.0006\nIter: 19, Epoch: 174, Loss: 0.0008\nIter: 20, Epoch: 174, Loss: 0.0007\nIter: 21, Epoch: 174, Loss: 0.0004\nIter: 22, Epoch: 174, Loss: 0.0007\nIter: 23, Epoch: 174, Loss: 0.0008\nIter: 24, Epoch: 174, Loss: 0.0004\nIter: 25, Epoch: 174, Loss: 0.0004\nIter: 26, Epoch: 174, Loss: 0.0006\nIter: 27, Epoch: 174, Loss: 0.0003\nIter: 28, Epoch: 174, Loss: 0.0007\nIter: 29, Epoch: 174, Loss: 0.0010\nIter: 30, Epoch: 174, Loss: 0.0007\n","truncated":false}} +%--- +%[output:03356122] +% data: {"dataType":"text","outputData":{"text":">> Epoch 174 validation loss: 0.0009\n","truncated":false}} +%--- +%[output:8b6c8d64] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 175, Loss: 0.0013\nIter: 2, Epoch: 175, Loss: 0.0005\nIter: 3, Epoch: 175, Loss: 0.0006\nIter: 4, Epoch: 175, Loss: 0.0007\nIter: 5, Epoch: 175, Loss: 0.0004\nIter: 6, Epoch: 175, Loss: 0.0005\nIter: 7, Epoch: 175, Loss: 0.0005\nIter: 8, Epoch: 175, Loss: 0.0012\nIter: 9, Epoch: 175, Loss: 0.0008\nIter: 10, Epoch: 175, Loss: 0.0006\nIter: 11, Epoch: 175, Loss: 0.0004\nIter: 12, Epoch: 175, Loss: 0.0003\nIter: 13, Epoch: 175, Loss: 0.0006\nIter: 14, Epoch: 175, Loss: 0.0008\nIter: 15, Epoch: 175, Loss: 0.0005\nIter: 16, Epoch: 175, Loss: 0.0006\nIter: 17, Epoch: 175, Loss: 0.0005\nIter: 18, Epoch: 175, Loss: 0.0006\nIter: 19, Epoch: 175, Loss: 0.0008\nIter: 20, Epoch: 175, Loss: 0.0007\nIter: 21, Epoch: 175, Loss: 0.0004\nIter: 22, Epoch: 175, Loss: 0.0007\nIter: 23, Epoch: 175, Loss: 0.0008\nIter: 24, Epoch: 175, Loss: 0.0003\nIter: 25, Epoch: 175, Loss: 0.0004\nIter: 26, Epoch: 175, Loss: 0.0006\nIter: 27, Epoch: 175, Loss: 0.0003\nIter: 28, Epoch: 175, Loss: 0.0007\nIter: 29, Epoch: 175, Loss: 0.0010\nIter: 30, Epoch: 175, Loss: 0.0007\n","truncated":false}} +%--- +%[output:01de9fb3] +% data: {"dataType":"text","outputData":{"text":">> Epoch 175 validation loss: 0.0009\n","truncated":false}} +%--- +%[output:4d6b2bc8] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 176, Loss: 0.0013\nIter: 2, Epoch: 176, Loss: 0.0005\nIter: 3, Epoch: 176, Loss: 0.0006\nIter: 4, Epoch: 176, Loss: 0.0007\nIter: 5, Epoch: 176, Loss: 0.0004\nIter: 6, Epoch: 176, Loss: 0.0005\nIter: 7, Epoch: 176, Loss: 0.0005\nIter: 8, Epoch: 176, Loss: 0.0012\nIter: 9, Epoch: 176, Loss: 0.0008\nIter: 10, Epoch: 176, Loss: 0.0006\nIter: 11, Epoch: 176, Loss: 0.0004\nIter: 12, Epoch: 176, Loss: 0.0003\nIter: 13, Epoch: 176, Loss: 0.0006\nIter: 14, Epoch: 176, Loss: 0.0008\nIter: 15, Epoch: 176, Loss: 0.0005\nIter: 16, Epoch: 176, Loss: 0.0006\nIter: 17, Epoch: 176, Loss: 0.0005\nIter: 18, Epoch: 176, Loss: 0.0006\nIter: 19, Epoch: 176, Loss: 0.0008\nIter: 20, Epoch: 176, Loss: 0.0007\nIter: 21, Epoch: 176, Loss: 0.0004\nIter: 22, Epoch: 176, Loss: 0.0007\nIter: 23, Epoch: 176, Loss: 0.0008\nIter: 24, Epoch: 176, Loss: 0.0003\nIter: 25, Epoch: 176, Loss: 0.0004\nIter: 26, Epoch: 176, Loss: 0.0007\nIter: 27, Epoch: 176, Loss: 0.0003\nIter: 28, Epoch: 176, Loss: 0.0007\nIter: 29, Epoch: 176, Loss: 0.0010\nIter: 30, Epoch: 176, Loss: 0.0007\n","truncated":false}} +%--- +%[output:3062520d] +% data: {"dataType":"text","outputData":{"text":">> Epoch 176 validation loss: 0.0009\n","truncated":false}} +%--- +%[output:618c1793] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 177, Loss: 0.0012\nIter: 2, Epoch: 177, Loss: 0.0005\nIter: 3, Epoch: 177, Loss: 0.0006\nIter: 4, Epoch: 177, Loss: 0.0007\nIter: 5, Epoch: 177, Loss: 0.0004\nIter: 6, Epoch: 177, Loss: 0.0005\nIter: 7, Epoch: 177, Loss: 0.0005\nIter: 8, Epoch: 177, Loss: 0.0012\nIter: 9, Epoch: 177, Loss: 0.0008\nIter: 10, Epoch: 177, Loss: 0.0006\nIter: 11, Epoch: 177, Loss: 0.0004\nIter: 12, Epoch: 177, Loss: 0.0003\nIter: 13, Epoch: 177, Loss: 0.0006\nIter: 14, Epoch: 177, Loss: 0.0008\nIter: 15, Epoch: 177, Loss: 0.0005\nIter: 16, Epoch: 177, Loss: 0.0006\nIter: 17, Epoch: 177, Loss: 0.0005\nIter: 18, Epoch: 177, Loss: 0.0006\nIter: 19, Epoch: 177, Loss: 0.0008\nIter: 20, Epoch: 177, Loss: 0.0007\nIter: 21, Epoch: 177, Loss: 0.0004\nIter: 22, Epoch: 177, Loss: 0.0007\nIter: 23, Epoch: 177, Loss: 0.0009\nIter: 24, Epoch: 177, Loss: 0.0003\nIter: 25, Epoch: 177, Loss: 0.0004\nIter: 26, Epoch: 177, Loss: 0.0007\nIter: 27, Epoch: 177, Loss: 0.0003\nIter: 28, Epoch: 177, Loss: 0.0007\nIter: 29, Epoch: 177, Loss: 0.0009\nIter: 30, Epoch: 177, Loss: 0.0007\n","truncated":false}} +%--- +%[output:9fdc266c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 177 validation loss: 0.0008\n","truncated":false}} +%--- +%[output:3f4f78f2] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 178, Loss: 0.0012\nIter: 2, Epoch: 178, Loss: 0.0005\nIter: 3, Epoch: 178, Loss: 0.0006\nIter: 4, Epoch: 178, Loss: 0.0007\nIter: 5, Epoch: 178, Loss: 0.0004\nIter: 6, Epoch: 178, Loss: 0.0005\nIter: 7, Epoch: 178, Loss: 0.0005\nIter: 8, Epoch: 178, Loss: 0.0012\nIter: 9, Epoch: 178, Loss: 0.0007\nIter: 10, Epoch: 178, Loss: 0.0006\nIter: 11, Epoch: 178, Loss: 0.0004\nIter: 12, Epoch: 178, Loss: 0.0003\nIter: 13, Epoch: 178, Loss: 0.0006\nIter: 14, Epoch: 178, Loss: 0.0008\nIter: 15, Epoch: 178, Loss: 0.0005\nIter: 16, Epoch: 178, Loss: 0.0006\nIter: 17, Epoch: 178, Loss: 0.0005\nIter: 18, Epoch: 178, Loss: 0.0006\nIter: 19, Epoch: 178, Loss: 0.0008\nIter: 20, Epoch: 178, Loss: 0.0007\nIter: 21, Epoch: 178, Loss: 0.0004\nIter: 22, Epoch: 178, Loss: 0.0007\nIter: 23, Epoch: 178, Loss: 0.0008\nIter: 24, Epoch: 178, Loss: 0.0003\nIter: 25, Epoch: 178, Loss: 0.0004\nIter: 26, Epoch: 178, Loss: 0.0007\nIter: 27, Epoch: 178, Loss: 0.0003\nIter: 28, Epoch: 178, Loss: 0.0006\nIter: 29, Epoch: 178, Loss: 0.0009\nIter: 30, Epoch: 178, Loss: 0.0007\n","truncated":false}} +%--- +%[output:4511a275] +% data: {"dataType":"text","outputData":{"text":">> Epoch 178 validation loss: 0.0008\n","truncated":false}} +%--- +%[output:8a6f0cf8] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 179, Loss: 0.0012\nIter: 2, Epoch: 179, Loss: 0.0005\nIter: 3, Epoch: 179, Loss: 0.0006\nIter: 4, Epoch: 179, Loss: 0.0007\nIter: 5, Epoch: 179, Loss: 0.0004\nIter: 6, Epoch: 179, Loss: 0.0005\nIter: 7, Epoch: 179, Loss: 0.0005\nIter: 8, Epoch: 179, Loss: 0.0012\nIter: 9, Epoch: 179, Loss: 0.0007\nIter: 10, Epoch: 179, Loss: 0.0006\nIter: 11, Epoch: 179, Loss: 0.0004\nIter: 12, Epoch: 179, Loss: 0.0003\nIter: 13, Epoch: 179, Loss: 0.0006\nIter: 14, Epoch: 179, Loss: 0.0008\nIter: 15, Epoch: 179, Loss: 0.0005\nIter: 16, Epoch: 179, Loss: 0.0006\nIter: 17, Epoch: 179, Loss: 0.0005\nIter: 18, Epoch: 179, Loss: 0.0006\nIter: 19, Epoch: 179, Loss: 0.0008\nIter: 20, Epoch: 179, Loss: 0.0007\nIter: 21, Epoch: 179, Loss: 0.0004\nIter: 22, Epoch: 179, Loss: 0.0007\nIter: 23, Epoch: 179, Loss: 0.0009\nIter: 24, Epoch: 179, Loss: 0.0003\nIter: 25, Epoch: 179, Loss: 0.0004\nIter: 26, Epoch: 179, Loss: 0.0007\nIter: 27, Epoch: 179, Loss: 0.0003\nIter: 28, Epoch: 179, Loss: 0.0006\nIter: 29, Epoch: 179, Loss: 0.0009\nIter: 30, Epoch: 179, Loss: 0.0007\n","truncated":false}} +%--- +%[output:85427654] +% data: {"dataType":"text","outputData":{"text":">> Epoch 179 validation loss: 0.0008\n","truncated":false}} +%--- +%[output:6f32ee91] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 180, Loss: 0.0012\nIter: 2, Epoch: 180, Loss: 0.0005\nIter: 3, Epoch: 180, Loss: 0.0006\nIter: 4, Epoch: 180, Loss: 0.0006\nIter: 5, Epoch: 180, Loss: 0.0004\nIter: 6, Epoch: 180, Loss: 0.0005\nIter: 7, Epoch: 180, Loss: 0.0005\nIter: 8, Epoch: 180, Loss: 0.0012\nIter: 9, Epoch: 180, Loss: 0.0007\nIter: 10, Epoch: 180, Loss: 0.0006\nIter: 11, Epoch: 180, Loss: 0.0004\nIter: 12, Epoch: 180, Loss: 0.0003\nIter: 13, Epoch: 180, Loss: 0.0006\nIter: 14, Epoch: 180, Loss: 0.0007\nIter: 15, Epoch: 180, Loss: 0.0005\nIter: 16, Epoch: 180, Loss: 0.0006\nIter: 17, Epoch: 180, Loss: 0.0005\nIter: 18, Epoch: 180, Loss: 0.0006\nIter: 19, Epoch: 180, Loss: 0.0008\nIter: 20, Epoch: 180, Loss: 0.0007\nIter: 21, Epoch: 180, Loss: 0.0004\nIter: 22, Epoch: 180, Loss: 0.0007\nIter: 23, Epoch: 180, Loss: 0.0009\nIter: 24, Epoch: 180, Loss: 0.0003\nIter: 25, Epoch: 180, Loss: 0.0004\nIter: 26, Epoch: 180, Loss: 0.0007\nIter: 27, Epoch: 180, Loss: 0.0003\nIter: 28, Epoch: 180, Loss: 0.0006\nIter: 29, Epoch: 180, Loss: 0.0009\nIter: 30, Epoch: 180, Loss: 0.0007\n","truncated":false}} +%--- +%[output:53320511] +% data: {"dataType":"text","outputData":{"text":">> Epoch 180 validation loss: 0.0008\n","truncated":false}} +%--- +%[output:38bab324] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 181, Loss: 0.0011\nIter: 2, Epoch: 181, Loss: 0.0005\nIter: 3, Epoch: 181, Loss: 0.0006\nIter: 4, Epoch: 181, Loss: 0.0006\nIter: 5, Epoch: 181, Loss: 0.0004\nIter: 6, Epoch: 181, Loss: 0.0005\nIter: 7, Epoch: 181, Loss: 0.0005\nIter: 8, Epoch: 181, Loss: 0.0012\nIter: 9, Epoch: 181, Loss: 0.0007\nIter: 10, Epoch: 181, Loss: 0.0006\nIter: 11, Epoch: 181, Loss: 0.0004\nIter: 12, Epoch: 181, Loss: 0.0003\nIter: 13, Epoch: 181, Loss: 0.0006\nIter: 14, Epoch: 181, Loss: 0.0007\nIter: 15, Epoch: 181, Loss: 0.0005\nIter: 16, Epoch: 181, Loss: 0.0006\nIter: 17, Epoch: 181, Loss: 0.0005\nIter: 18, Epoch: 181, Loss: 0.0007\nIter: 19, Epoch: 181, Loss: 0.0008\nIter: 20, Epoch: 181, Loss: 0.0006\nIter: 21, Epoch: 181, Loss: 0.0004\nIter: 22, Epoch: 181, Loss: 0.0007\nIter: 23, Epoch: 181, Loss: 0.0009\nIter: 24, Epoch: 181, Loss: 0.0003\nIter: 25, Epoch: 181, Loss: 0.0004\nIter: 26, Epoch: 181, Loss: 0.0007\nIter: 27, Epoch: 181, Loss: 0.0003\nIter: 28, Epoch: 181, Loss: 0.0006\nIter: 29, Epoch: 181, Loss: 0.0009\nIter: 30, Epoch: 181, Loss: 0.0007\n","truncated":false}} +%--- +%[output:1bdff14a] +% data: {"dataType":"text","outputData":{"text":">> Epoch 181 validation loss: 0.0008\n","truncated":false}} +%--- +%[output:34c87860] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 182, Loss: 0.0011\nIter: 2, Epoch: 182, Loss: 0.0005\nIter: 3, Epoch: 182, Loss: 0.0005\nIter: 4, Epoch: 182, Loss: 0.0006\nIter: 5, Epoch: 182, Loss: 0.0004\nIter: 6, Epoch: 182, Loss: 0.0005\nIter: 7, Epoch: 182, Loss: 0.0005\nIter: 8, Epoch: 182, Loss: 0.0012\nIter: 9, Epoch: 182, Loss: 0.0007\nIter: 10, Epoch: 182, Loss: 0.0006\nIter: 11, Epoch: 182, Loss: 0.0004\nIter: 12, Epoch: 182, Loss: 0.0003\nIter: 13, Epoch: 182, Loss: 0.0006\nIter: 14, Epoch: 182, Loss: 0.0007\nIter: 15, Epoch: 182, Loss: 0.0005\nIter: 16, Epoch: 182, Loss: 0.0006\nIter: 17, Epoch: 182, Loss: 0.0005\nIter: 18, Epoch: 182, Loss: 0.0007\nIter: 19, Epoch: 182, Loss: 0.0008\nIter: 20, Epoch: 182, Loss: 0.0007\nIter: 21, Epoch: 182, Loss: 0.0004\nIter: 22, Epoch: 182, Loss: 0.0007\nIter: 23, Epoch: 182, Loss: 0.0009\nIter: 24, Epoch: 182, Loss: 0.0003\nIter: 25, Epoch: 182, Loss: 0.0004\nIter: 26, Epoch: 182, Loss: 0.0007\nIter: 27, Epoch: 182, Loss: 0.0003\nIter: 28, Epoch: 182, Loss: 0.0006\nIter: 29, Epoch: 182, Loss: 0.0009\nIter: 30, Epoch: 182, Loss: 0.0007\n","truncated":false}} +%--- +%[output:18d99f7f] +% data: {"dataType":"text","outputData":{"text":">> Epoch 182 validation loss: 0.0008\n","truncated":false}} +%--- +%[output:9432b1fe] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 183, Loss: 0.0011\nIter: 2, Epoch: 183, Loss: 0.0005\nIter: 3, Epoch: 183, Loss: 0.0005\nIter: 4, Epoch: 183, Loss: 0.0006\nIter: 5, Epoch: 183, Loss: 0.0004\nIter: 6, Epoch: 183, Loss: 0.0005\nIter: 7, Epoch: 183, Loss: 0.0005\nIter: 8, Epoch: 183, Loss: 0.0012\nIter: 9, Epoch: 183, Loss: 0.0007\nIter: 10, Epoch: 183, Loss: 0.0005\nIter: 11, Epoch: 183, Loss: 0.0005\nIter: 12, Epoch: 183, Loss: 0.0003\nIter: 13, Epoch: 183, Loss: 0.0006\nIter: 14, Epoch: 183, Loss: 0.0007\nIter: 15, Epoch: 183, Loss: 0.0005\nIter: 16, Epoch: 183, Loss: 0.0006\nIter: 17, Epoch: 183, Loss: 0.0005\nIter: 18, Epoch: 183, Loss: 0.0007\nIter: 19, Epoch: 183, Loss: 0.0008\nIter: 20, Epoch: 183, Loss: 0.0006\nIter: 21, Epoch: 183, Loss: 0.0004\nIter: 22, Epoch: 183, Loss: 0.0007\nIter: 23, Epoch: 183, Loss: 0.0009\nIter: 24, Epoch: 183, Loss: 0.0003\nIter: 25, Epoch: 183, Loss: 0.0004\nIter: 26, Epoch: 183, Loss: 0.0007\nIter: 27, Epoch: 183, Loss: 0.0003\nIter: 28, Epoch: 183, Loss: 0.0006\nIter: 29, Epoch: 183, Loss: 0.0009\nIter: 30, Epoch: 183, Loss: 0.0007\n","truncated":false}} +%--- +%[output:6bce0e7d] +% data: {"dataType":"text","outputData":{"text":">> Epoch 183 validation loss: 0.0008\n","truncated":false}} +%--- +%[output:9823e54b] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 184, Loss: 0.0011\nIter: 2, Epoch: 184, Loss: 0.0005\nIter: 3, Epoch: 184, Loss: 0.0005\nIter: 4, Epoch: 184, Loss: 0.0006\nIter: 5, Epoch: 184, Loss: 0.0004\nIter: 6, Epoch: 184, Loss: 0.0005\nIter: 7, Epoch: 184, Loss: 0.0005\nIter: 8, Epoch: 184, Loss: 0.0012\nIter: 9, Epoch: 184, Loss: 0.0007\nIter: 10, Epoch: 184, Loss: 0.0005\nIter: 11, Epoch: 184, Loss: 0.0005\nIter: 12, Epoch: 184, Loss: 0.0003\nIter: 13, Epoch: 184, Loss: 0.0006\nIter: 14, Epoch: 184, Loss: 0.0007\nIter: 15, Epoch: 184, Loss: 0.0005\nIter: 16, Epoch: 184, Loss: 0.0006\nIter: 17, Epoch: 184, Loss: 0.0005\nIter: 18, Epoch: 184, Loss: 0.0007\nIter: 19, Epoch: 184, Loss: 0.0009\nIter: 20, Epoch: 184, Loss: 0.0006\nIter: 21, Epoch: 184, Loss: 0.0004\nIter: 22, Epoch: 184, Loss: 0.0007\nIter: 23, Epoch: 184, Loss: 0.0009\nIter: 24, Epoch: 184, Loss: 0.0003\nIter: 25, Epoch: 184, Loss: 0.0004\nIter: 26, Epoch: 184, Loss: 0.0007\nIter: 27, Epoch: 184, Loss: 0.0003\nIter: 28, Epoch: 184, Loss: 0.0006\nIter: 29, Epoch: 184, Loss: 0.0008\nIter: 30, Epoch: 184, Loss: 0.0007\n","truncated":false}} +%--- +%[output:14e87826] +% data: {"dataType":"text","outputData":{"text":">> Epoch 184 validation loss: 0.0008\n","truncated":false}} +%--- +%[output:3ba13a05] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 185, Loss: 0.0011\nIter: 2, Epoch: 185, Loss: 0.0005\nIter: 3, Epoch: 185, Loss: 0.0005\nIter: 4, Epoch: 185, Loss: 0.0006\nIter: 5, Epoch: 185, Loss: 0.0004\nIter: 6, Epoch: 185, Loss: 0.0005\nIter: 7, Epoch: 185, Loss: 0.0005\nIter: 8, Epoch: 185, Loss: 0.0012\nIter: 9, Epoch: 185, Loss: 0.0007\nIter: 10, Epoch: 185, Loss: 0.0005\nIter: 11, Epoch: 185, Loss: 0.0005\nIter: 12, Epoch: 185, Loss: 0.0003\nIter: 13, Epoch: 185, Loss: 0.0006\nIter: 14, Epoch: 185, Loss: 0.0007\nIter: 15, Epoch: 185, Loss: 0.0005\nIter: 16, Epoch: 185, Loss: 0.0006\nIter: 17, Epoch: 185, Loss: 0.0005\nIter: 18, Epoch: 185, Loss: 0.0007\nIter: 19, Epoch: 185, Loss: 0.0009\nIter: 20, Epoch: 185, Loss: 0.0006\nIter: 21, Epoch: 185, Loss: 0.0004\nIter: 22, Epoch: 185, Loss: 0.0007\nIter: 23, Epoch: 185, Loss: 0.0010\nIter: 24, Epoch: 185, Loss: 0.0003\nIter: 25, Epoch: 185, Loss: 0.0004\nIter: 26, Epoch: 185, Loss: 0.0007\nIter: 27, Epoch: 185, Loss: 0.0003\nIter: 28, Epoch: 185, Loss: 0.0006\nIter: 29, Epoch: 185, Loss: 0.0008\nIter: 30, Epoch: 185, Loss: 0.0007\n","truncated":false}} +%--- +%[output:562cbc1d] +% data: {"dataType":"text","outputData":{"text":">> Epoch 185 validation loss: 0.0008\n","truncated":false}} +%--- +%[output:314da832] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 186, Loss: 0.0011\nIter: 2, Epoch: 186, Loss: 0.0005\nIter: 3, Epoch: 186, Loss: 0.0005\nIter: 4, Epoch: 186, Loss: 0.0006\nIter: 5, Epoch: 186, Loss: 0.0004\nIter: 6, Epoch: 186, Loss: 0.0005\nIter: 7, Epoch: 186, Loss: 0.0005\nIter: 8, Epoch: 186, Loss: 0.0013\nIter: 9, Epoch: 186, Loss: 0.0006\nIter: 10, Epoch: 186, Loss: 0.0005\nIter: 11, Epoch: 186, Loss: 0.0005\nIter: 12, Epoch: 186, Loss: 0.0003\nIter: 13, Epoch: 186, Loss: 0.0006\nIter: 14, Epoch: 186, Loss: 0.0007\nIter: 15, Epoch: 186, Loss: 0.0006\nIter: 16, Epoch: 186, Loss: 0.0006\nIter: 17, Epoch: 186, Loss: 0.0005\nIter: 18, Epoch: 186, Loss: 0.0007\nIter: 19, Epoch: 186, Loss: 0.0009\nIter: 20, Epoch: 186, Loss: 0.0006\nIter: 21, Epoch: 186, Loss: 0.0004\nIter: 22, Epoch: 186, Loss: 0.0007\nIter: 23, Epoch: 186, Loss: 0.0010\nIter: 24, Epoch: 186, Loss: 0.0003\nIter: 25, Epoch: 186, Loss: 0.0004\nIter: 26, Epoch: 186, Loss: 0.0007\nIter: 27, Epoch: 186, Loss: 0.0003\nIter: 28, Epoch: 186, Loss: 0.0006\nIter: 29, Epoch: 186, Loss: 0.0008\nIter: 30, Epoch: 186, Loss: 0.0007\n","truncated":false}} +%--- +%[output:1f0cdcc9] +% data: {"dataType":"text","outputData":{"text":">> Epoch 186 validation loss: 0.0008\n","truncated":false}} +%--- +%[output:88d51177] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 187, Loss: 0.0011\nIter: 2, Epoch: 187, Loss: 0.0005\nIter: 3, Epoch: 187, Loss: 0.0005\nIter: 4, Epoch: 187, Loss: 0.0007\nIter: 5, Epoch: 187, Loss: 0.0004\nIter: 6, Epoch: 187, Loss: 0.0005\nIter: 7, Epoch: 187, Loss: 0.0005\nIter: 8, Epoch: 187, Loss: 0.0013\nIter: 9, Epoch: 187, Loss: 0.0006\nIter: 10, Epoch: 187, Loss: 0.0005\nIter: 11, Epoch: 187, Loss: 0.0005\nIter: 12, Epoch: 187, Loss: 0.0003\nIter: 13, Epoch: 187, Loss: 0.0006\nIter: 14, Epoch: 187, Loss: 0.0007\nIter: 15, Epoch: 187, Loss: 0.0006\nIter: 16, Epoch: 187, Loss: 0.0006\nIter: 17, Epoch: 187, Loss: 0.0005\nIter: 18, Epoch: 187, Loss: 0.0007\nIter: 19, Epoch: 187, Loss: 0.0009\nIter: 20, Epoch: 187, Loss: 0.0006\nIter: 21, Epoch: 187, Loss: 0.0004\nIter: 22, Epoch: 187, Loss: 0.0007\nIter: 23, Epoch: 187, Loss: 0.0010\nIter: 24, Epoch: 187, Loss: 0.0003\nIter: 25, Epoch: 187, Loss: 0.0004\nIter: 26, Epoch: 187, Loss: 0.0007\nIter: 27, Epoch: 187, Loss: 0.0003\nIter: 28, Epoch: 187, Loss: 0.0006\nIter: 29, Epoch: 187, Loss: 0.0008\nIter: 30, Epoch: 187, Loss: 0.0007\n","truncated":false}} +%--- +%[output:60aa40f7] +% data: {"dataType":"text","outputData":{"text":">> Epoch 187 validation loss: 0.0009\n","truncated":false}} +%--- +%[output:9c5ad102] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 188, Loss: 0.0012\nIter: 2, Epoch: 188, Loss: 0.0005\nIter: 3, Epoch: 188, Loss: 0.0005\nIter: 4, Epoch: 188, Loss: 0.0007\nIter: 5, Epoch: 188, Loss: 0.0004\nIter: 6, Epoch: 188, Loss: 0.0005\nIter: 7, Epoch: 188, Loss: 0.0005\nIter: 8, Epoch: 188, Loss: 0.0013\nIter: 9, Epoch: 188, Loss: 0.0006\nIter: 10, Epoch: 188, Loss: 0.0005\nIter: 11, Epoch: 188, Loss: 0.0005\nIter: 12, Epoch: 188, Loss: 0.0003\nIter: 13, Epoch: 188, Loss: 0.0006\nIter: 14, Epoch: 188, Loss: 0.0007\nIter: 15, Epoch: 188, Loss: 0.0006\nIter: 16, Epoch: 188, Loss: 0.0006\nIter: 17, Epoch: 188, Loss: 0.0005\nIter: 18, Epoch: 188, Loss: 0.0007\nIter: 19, Epoch: 188, Loss: 0.0010\nIter: 20, Epoch: 188, Loss: 0.0006\nIter: 21, Epoch: 188, Loss: 0.0004\nIter: 22, Epoch: 188, Loss: 0.0007\nIter: 23, Epoch: 188, Loss: 0.0011\nIter: 24, Epoch: 188, Loss: 0.0003\nIter: 25, Epoch: 188, Loss: 0.0004\nIter: 26, Epoch: 188, Loss: 0.0008\nIter: 27, Epoch: 188, Loss: 0.0003\nIter: 28, Epoch: 188, Loss: 0.0006\nIter: 29, Epoch: 188, Loss: 0.0008\nIter: 30, Epoch: 188, Loss: 0.0007\n","truncated":false}} +%--- +%[output:8cf7f8a7] +% data: {"dataType":"text","outputData":{"text":">> Epoch 188 validation loss: 0.0009\n","truncated":false}} +%--- +%[output:97889ee2] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 189, Loss: 0.0012\nIter: 2, Epoch: 189, Loss: 0.0005\nIter: 3, Epoch: 189, Loss: 0.0005\nIter: 4, Epoch: 189, Loss: 0.0007\nIter: 5, Epoch: 189, Loss: 0.0004\nIter: 6, Epoch: 189, Loss: 0.0004\nIter: 7, Epoch: 189, Loss: 0.0005\nIter: 8, Epoch: 189, Loss: 0.0014\nIter: 9, Epoch: 189, Loss: 0.0006\nIter: 10, Epoch: 189, Loss: 0.0005\nIter: 11, Epoch: 189, Loss: 0.0005\nIter: 12, Epoch: 189, Loss: 0.0003\nIter: 13, Epoch: 189, Loss: 0.0006\nIter: 14, Epoch: 189, Loss: 0.0007\nIter: 15, Epoch: 189, Loss: 0.0006\nIter: 16, Epoch: 189, Loss: 0.0006\nIter: 17, Epoch: 189, Loss: 0.0005\nIter: 18, Epoch: 189, Loss: 0.0008\nIter: 19, Epoch: 189, Loss: 0.0010\nIter: 20, Epoch: 189, Loss: 0.0006\nIter: 21, Epoch: 189, Loss: 0.0004\nIter: 22, Epoch: 189, Loss: 0.0007\nIter: 23, Epoch: 189, Loss: 0.0011\nIter: 24, Epoch: 189, Loss: 0.0003\nIter: 25, Epoch: 189, Loss: 0.0004\nIter: 26, Epoch: 189, Loss: 0.0008\nIter: 27, Epoch: 189, Loss: 0.0003\nIter: 28, Epoch: 189, Loss: 0.0005\nIter: 29, Epoch: 189, Loss: 0.0008\nIter: 30, Epoch: 189, Loss: 0.0007\n","truncated":false}} +%--- +%[output:56002f24] +% data: {"dataType":"text","outputData":{"text":">> Epoch 189 validation loss: 0.0009\n","truncated":false}} +%--- +%[output:11d9dde7] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 190, Loss: 0.0013\nIter: 2, Epoch: 190, Loss: 0.0005\nIter: 3, Epoch: 190, Loss: 0.0005\nIter: 4, Epoch: 190, Loss: 0.0007\nIter: 5, Epoch: 190, Loss: 0.0005\nIter: 6, Epoch: 190, Loss: 0.0004\nIter: 7, Epoch: 190, Loss: 0.0005\nIter: 8, Epoch: 190, Loss: 0.0015\nIter: 9, Epoch: 190, Loss: 0.0006\nIter: 10, Epoch: 190, Loss: 0.0005\nIter: 11, Epoch: 190, Loss: 0.0005\nIter: 12, Epoch: 190, Loss: 0.0004\nIter: 13, Epoch: 190, Loss: 0.0006\nIter: 14, Epoch: 190, Loss: 0.0007\nIter: 15, Epoch: 190, Loss: 0.0006\nIter: 16, Epoch: 190, Loss: 0.0007\nIter: 17, Epoch: 190, Loss: 0.0006\nIter: 18, Epoch: 190, Loss: 0.0008\nIter: 19, Epoch: 190, Loss: 0.0011\nIter: 20, Epoch: 190, Loss: 0.0006\nIter: 21, Epoch: 190, Loss: 0.0004\nIter: 22, Epoch: 190, Loss: 0.0007\nIter: 23, Epoch: 190, Loss: 0.0012\nIter: 24, Epoch: 190, Loss: 0.0003\nIter: 25, Epoch: 190, Loss: 0.0004\nIter: 26, Epoch: 190, Loss: 0.0008\nIter: 27, Epoch: 190, Loss: 0.0003\nIter: 28, Epoch: 190, Loss: 0.0005\nIter: 29, Epoch: 190, Loss: 0.0008\nIter: 30, Epoch: 190, Loss: 0.0007\n","truncated":false}} +%--- +%[output:92a3d8a2] +% data: {"dataType":"text","outputData":{"text":">> Epoch 190 validation loss: 0.0010\n","truncated":false}} +%--- +%[output:59da6fed] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 191, Loss: 0.0014\nIter: 2, Epoch: 191, Loss: 0.0004\nIter: 3, Epoch: 191, Loss: 0.0005\nIter: 4, Epoch: 191, Loss: 0.0008\nIter: 5, Epoch: 191, Loss: 0.0005\nIter: 6, Epoch: 191, Loss: 0.0004\nIter: 7, Epoch: 191, Loss: 0.0005\nIter: 8, Epoch: 191, Loss: 0.0015\nIter: 9, Epoch: 191, Loss: 0.0005\nIter: 10, Epoch: 191, Loss: 0.0005\nIter: 11, Epoch: 191, Loss: 0.0006\nIter: 12, Epoch: 191, Loss: 0.0004\nIter: 13, Epoch: 191, Loss: 0.0006\nIter: 14, Epoch: 191, Loss: 0.0007\nIter: 15, Epoch: 191, Loss: 0.0007\nIter: 16, Epoch: 191, Loss: 0.0007\nIter: 17, Epoch: 191, Loss: 0.0006\nIter: 18, Epoch: 191, Loss: 0.0008\nIter: 19, Epoch: 191, Loss: 0.0012\nIter: 20, Epoch: 191, Loss: 0.0006\nIter: 21, Epoch: 191, Loss: 0.0004\nIter: 22, Epoch: 191, Loss: 0.0007\nIter: 23, Epoch: 191, Loss: 0.0013\nIter: 24, Epoch: 191, Loss: 0.0004\nIter: 25, Epoch: 191, Loss: 0.0004\nIter: 26, Epoch: 191, Loss: 0.0008\nIter: 27, Epoch: 191, Loss: 0.0004\nIter: 28, Epoch: 191, Loss: 0.0005\nIter: 29, Epoch: 191, Loss: 0.0007\nIter: 30, Epoch: 191, Loss: 0.0007\n","truncated":false}} +%--- +%[output:762e98bd] +% data: {"dataType":"text","outputData":{"text":">> Epoch 191 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:99376ae3] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 192, Loss: 0.0015\nIter: 2, Epoch: 192, Loss: 0.0004\nIter: 3, Epoch: 192, Loss: 0.0005\nIter: 4, Epoch: 192, Loss: 0.0008\nIter: 5, Epoch: 192, Loss: 0.0005\nIter: 6, Epoch: 192, Loss: 0.0004\nIter: 7, Epoch: 192, Loss: 0.0005\nIter: 8, Epoch: 192, Loss: 0.0016\nIter: 9, Epoch: 192, Loss: 0.0005\nIter: 10, Epoch: 192, Loss: 0.0005\nIter: 11, Epoch: 192, Loss: 0.0006\nIter: 12, Epoch: 192, Loss: 0.0004\nIter: 13, Epoch: 192, Loss: 0.0007\nIter: 14, Epoch: 192, Loss: 0.0007\nIter: 15, Epoch: 192, Loss: 0.0007\nIter: 16, Epoch: 192, Loss: 0.0008\nIter: 17, Epoch: 192, Loss: 0.0006\nIter: 18, Epoch: 192, Loss: 0.0008\nIter: 19, Epoch: 192, Loss: 0.0013\nIter: 20, Epoch: 192, Loss: 0.0006\nIter: 21, Epoch: 192, Loss: 0.0004\nIter: 22, Epoch: 192, Loss: 0.0007\nIter: 23, Epoch: 192, Loss: 0.0014\nIter: 24, Epoch: 192, Loss: 0.0004\nIter: 25, Epoch: 192, Loss: 0.0004\nIter: 26, Epoch: 192, Loss: 0.0008\nIter: 27, Epoch: 192, Loss: 0.0004\nIter: 28, Epoch: 192, Loss: 0.0006\nIter: 29, Epoch: 192, Loss: 0.0007\nIter: 30, Epoch: 192, Loss: 0.0007\n","truncated":false}} +%--- +%[output:95bf697f] +% data: {"dataType":"text","outputData":{"text":">> Epoch 192 validation loss: 0.0012\n","truncated":false}} +%--- +%[output:9b3f564f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 193, Loss: 0.0016\nIter: 2, Epoch: 193, Loss: 0.0004\nIter: 3, Epoch: 193, Loss: 0.0005\nIter: 4, Epoch: 193, Loss: 0.0009\nIter: 5, Epoch: 193, Loss: 0.0006\nIter: 6, Epoch: 193, Loss: 0.0004\nIter: 7, Epoch: 193, Loss: 0.0005\nIter: 8, Epoch: 193, Loss: 0.0017\nIter: 9, Epoch: 193, Loss: 0.0005\nIter: 10, Epoch: 193, Loss: 0.0005\nIter: 11, Epoch: 193, Loss: 0.0006\nIter: 12, Epoch: 193, Loss: 0.0005\nIter: 13, Epoch: 193, Loss: 0.0008\nIter: 14, Epoch: 193, Loss: 0.0007\nIter: 15, Epoch: 193, Loss: 0.0007\nIter: 16, Epoch: 193, Loss: 0.0009\nIter: 17, Epoch: 193, Loss: 0.0006\nIter: 18, Epoch: 193, Loss: 0.0007\nIter: 19, Epoch: 193, Loss: 0.0014\nIter: 20, Epoch: 193, Loss: 0.0007\nIter: 21, Epoch: 193, Loss: 0.0003\nIter: 22, Epoch: 193, Loss: 0.0007\nIter: 23, Epoch: 193, Loss: 0.0015\nIter: 24, Epoch: 193, Loss: 0.0005\nIter: 25, Epoch: 193, Loss: 0.0003\nIter: 26, Epoch: 193, Loss: 0.0008\nIter: 27, Epoch: 193, Loss: 0.0005\nIter: 28, Epoch: 193, Loss: 0.0006\nIter: 29, Epoch: 193, Loss: 0.0007\nIter: 30, Epoch: 193, Loss: 0.0007\n","truncated":false}} +%--- +%[output:46ffaf31] +% data: {"dataType":"text","outputData":{"text":">> Epoch 193 validation loss: 0.0013\n","truncated":false}} +%--- +%[output:5111034d] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 194, Loss: 0.0018\nIter: 2, Epoch: 194, Loss: 0.0005\nIter: 3, Epoch: 194, Loss: 0.0005\nIter: 4, Epoch: 194, Loss: 0.0009\nIter: 5, Epoch: 194, Loss: 0.0007\nIter: 6, Epoch: 194, Loss: 0.0004\nIter: 7, Epoch: 194, Loss: 0.0005\nIter: 8, Epoch: 194, Loss: 0.0018\nIter: 9, Epoch: 194, Loss: 0.0005\nIter: 10, Epoch: 194, Loss: 0.0006\nIter: 11, Epoch: 194, Loss: 0.0006\nIter: 12, Epoch: 194, Loss: 0.0005\nIter: 13, Epoch: 194, Loss: 0.0010\nIter: 14, Epoch: 194, Loss: 0.0007\nIter: 15, Epoch: 194, Loss: 0.0008\nIter: 16, Epoch: 194, Loss: 0.0011\nIter: 17, Epoch: 194, Loss: 0.0007\nIter: 18, Epoch: 194, Loss: 0.0007\nIter: 19, Epoch: 194, Loss: 0.0014\nIter: 20, Epoch: 194, Loss: 0.0008\nIter: 21, Epoch: 194, Loss: 0.0003\nIter: 22, Epoch: 194, Loss: 0.0007\nIter: 23, Epoch: 194, Loss: 0.0016\nIter: 24, Epoch: 194, Loss: 0.0006\nIter: 25, Epoch: 194, Loss: 0.0003\nIter: 26, Epoch: 194, Loss: 0.0007\nIter: 27, Epoch: 194, Loss: 0.0006\nIter: 28, Epoch: 194, Loss: 0.0007\nIter: 29, Epoch: 194, Loss: 0.0008\nIter: 30, Epoch: 194, Loss: 0.0007\n","truncated":false}} +%--- +%[output:86d6200c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 194 validation loss: 0.0014\n","truncated":false}} +%--- +%[output:73677cba] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 195, Loss: 0.0019\nIter: 2, Epoch: 195, Loss: 0.0006\nIter: 3, Epoch: 195, Loss: 0.0005\nIter: 4, Epoch: 195, Loss: 0.0009\nIter: 5, Epoch: 195, Loss: 0.0009\nIter: 6, Epoch: 195, Loss: 0.0004\nIter: 7, Epoch: 195, Loss: 0.0005\nIter: 8, Epoch: 195, Loss: 0.0018\nIter: 9, Epoch: 195, Loss: 0.0005\nIter: 10, Epoch: 195, Loss: 0.0007\nIter: 11, Epoch: 195, Loss: 0.0005\nIter: 12, Epoch: 195, Loss: 0.0006\nIter: 13, Epoch: 195, Loss: 0.0012\nIter: 14, Epoch: 195, Loss: 0.0007\nIter: 15, Epoch: 195, Loss: 0.0007\nIter: 16, Epoch: 195, Loss: 0.0012\nIter: 17, Epoch: 195, Loss: 0.0008\nIter: 18, Epoch: 195, Loss: 0.0006\nIter: 19, Epoch: 195, Loss: 0.0013\nIter: 20, Epoch: 195, Loss: 0.0009\nIter: 21, Epoch: 195, Loss: 0.0004\nIter: 22, Epoch: 195, Loss: 0.0007\nIter: 23, Epoch: 195, Loss: 0.0016\nIter: 24, Epoch: 195, Loss: 0.0008\nIter: 25, Epoch: 195, Loss: 0.0004\nIter: 26, Epoch: 195, Loss: 0.0006\nIter: 27, Epoch: 195, Loss: 0.0006\nIter: 28, Epoch: 195, Loss: 0.0009\nIter: 29, Epoch: 195, Loss: 0.0010\nIter: 30, Epoch: 195, Loss: 0.0007\n","truncated":false}} +%--- +%[output:66f4739a] +% data: {"dataType":"text","outputData":{"text":">> Epoch 195 validation loss: 0.0015\n","truncated":false}} +%--- +%[output:84a1c31e] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 196, Loss: 0.0019\nIter: 2, Epoch: 196, Loss: 0.0007\nIter: 3, Epoch: 196, Loss: 0.0006\nIter: 4, Epoch: 196, Loss: 0.0009\nIter: 5, Epoch: 196, Loss: 0.0010\nIter: 6, Epoch: 196, Loss: 0.0005\nIter: 7, Epoch: 196, Loss: 0.0005\nIter: 8, Epoch: 196, Loss: 0.0016\nIter: 9, Epoch: 196, Loss: 0.0006\nIter: 10, Epoch: 196, Loss: 0.0008\nIter: 11, Epoch: 196, Loss: 0.0004\nIter: 12, Epoch: 196, Loss: 0.0005\nIter: 13, Epoch: 196, Loss: 0.0014\nIter: 14, Epoch: 196, Loss: 0.0007\nIter: 15, Epoch: 196, Loss: 0.0006\nIter: 16, Epoch: 196, Loss: 0.0012\nIter: 17, Epoch: 196, Loss: 0.0009\nIter: 18, Epoch: 196, Loss: 0.0006\nIter: 19, Epoch: 196, Loss: 0.0011\nIter: 20, Epoch: 196, Loss: 0.0010\nIter: 21, Epoch: 196, Loss: 0.0005\nIter: 22, Epoch: 196, Loss: 0.0007\nIter: 23, Epoch: 196, Loss: 0.0015\nIter: 24, Epoch: 196, Loss: 0.0009\nIter: 25, Epoch: 196, Loss: 0.0005\nIter: 26, Epoch: 196, Loss: 0.0006\nIter: 27, Epoch: 196, Loss: 0.0005\nIter: 28, Epoch: 196, Loss: 0.0010\nIter: 29, Epoch: 196, Loss: 0.0012\nIter: 30, Epoch: 196, Loss: 0.0006\n","truncated":false}} +%--- +%[output:690f8572] +% data: {"dataType":"text","outputData":{"text":">> Epoch 196 validation loss: 0.0013\n","truncated":false}} +%--- +%[output:139d7b57] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 197, Loss: 0.0018\nIter: 2, Epoch: 197, Loss: 0.0008\nIter: 3, Epoch: 197, Loss: 0.0007\nIter: 4, Epoch: 197, Loss: 0.0007\nIter: 5, Epoch: 197, Loss: 0.0009\nIter: 6, Epoch: 197, Loss: 0.0006\nIter: 7, Epoch: 197, Loss: 0.0006\nIter: 8, Epoch: 197, Loss: 0.0014\nIter: 9, Epoch: 197, Loss: 0.0006\nIter: 10, Epoch: 197, Loss: 0.0010\nIter: 11, Epoch: 197, Loss: 0.0004\nIter: 12, Epoch: 197, Loss: 0.0004\nIter: 13, Epoch: 197, Loss: 0.0014\nIter: 14, Epoch: 197, Loss: 0.0009\nIter: 15, Epoch: 197, Loss: 0.0005\nIter: 16, Epoch: 197, Loss: 0.0010\nIter: 17, Epoch: 197, Loss: 0.0010\nIter: 18, Epoch: 197, Loss: 0.0006\nIter: 19, Epoch: 197, Loss: 0.0008\nIter: 20, Epoch: 197, Loss: 0.0009\nIter: 21, Epoch: 197, Loss: 0.0006\nIter: 22, Epoch: 197, Loss: 0.0008\nIter: 23, Epoch: 197, Loss: 0.0012\nIter: 24, Epoch: 197, Loss: 0.0008\nIter: 25, Epoch: 197, Loss: 0.0006\nIter: 26, Epoch: 197, Loss: 0.0006\nIter: 27, Epoch: 197, Loss: 0.0004\nIter: 28, Epoch: 197, Loss: 0.0009\nIter: 29, Epoch: 197, Loss: 0.0014\nIter: 30, Epoch: 197, Loss: 0.0007\n","truncated":false}} +%--- +%[output:09fa35d6] +% data: {"dataType":"text","outputData":{"text":">> Epoch 197 validation loss: 0.0011\n","truncated":false}} +%--- +%[output:48d8a20e] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 198, Loss: 0.0015\nIter: 2, Epoch: 198, Loss: 0.0008\nIter: 3, Epoch: 198, Loss: 0.0008\nIter: 4, Epoch: 198, Loss: 0.0006\nIter: 5, Epoch: 198, Loss: 0.0008\nIter: 6, Epoch: 198, Loss: 0.0006\nIter: 7, Epoch: 198, Loss: 0.0007\nIter: 8, Epoch: 198, Loss: 0.0011\nIter: 9, Epoch: 198, Loss: 0.0006\nIter: 10, Epoch: 198, Loss: 0.0011\nIter: 11, Epoch: 198, Loss: 0.0004\nIter: 12, Epoch: 198, Loss: 0.0003\nIter: 13, Epoch: 198, Loss: 0.0013\nIter: 14, Epoch: 198, Loss: 0.0009\nIter: 15, Epoch: 198, Loss: 0.0005\nIter: 16, Epoch: 198, Loss: 0.0008\nIter: 17, Epoch: 198, Loss: 0.0009\nIter: 18, Epoch: 198, Loss: 0.0007\nIter: 19, Epoch: 198, Loss: 0.0007\nIter: 20, Epoch: 198, Loss: 0.0008\nIter: 21, Epoch: 198, Loss: 0.0006\nIter: 22, Epoch: 198, Loss: 0.0009\nIter: 23, Epoch: 198, Loss: 0.0011\nIter: 24, Epoch: 198, Loss: 0.0007\nIter: 25, Epoch: 198, Loss: 0.0006\nIter: 26, Epoch: 198, Loss: 0.0006\nIter: 27, Epoch: 198, Loss: 0.0003\nIter: 28, Epoch: 198, Loss: 0.0008\nIter: 29, Epoch: 198, Loss: 0.0014\nIter: 30, Epoch: 198, Loss: 0.0008\n","truncated":false}} +%--- +%[output:1921d716] +% data: {"dataType":"text","outputData":{"text":">> Epoch 198 validation loss: 0.0009\n","truncated":false}} +%--- +%[output:782bb660] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 199, Loss: 0.0012\nIter: 2, Epoch: 199, Loss: 0.0007\nIter: 3, Epoch: 199, Loss: 0.0009\nIter: 4, Epoch: 199, Loss: 0.0006\nIter: 5, Epoch: 199, Loss: 0.0006\nIter: 6, Epoch: 199, Loss: 0.0006\nIter: 7, Epoch: 199, Loss: 0.0007\nIter: 8, Epoch: 199, Loss: 0.0010\nIter: 9, Epoch: 199, Loss: 0.0006\nIter: 10, Epoch: 199, Loss: 0.0010\nIter: 11, Epoch: 199, Loss: 0.0004\nIter: 12, Epoch: 199, Loss: 0.0003\nIter: 13, Epoch: 199, Loss: 0.0011\nIter: 14, Epoch: 199, Loss: 0.0009\nIter: 15, Epoch: 199, Loss: 0.0005\nIter: 16, Epoch: 199, Loss: 0.0007\nIter: 17, Epoch: 199, Loss: 0.0008\nIter: 18, Epoch: 199, Loss: 0.0007\nIter: 19, Epoch: 199, Loss: 0.0006\nIter: 20, Epoch: 199, Loss: 0.0007\nIter: 21, Epoch: 199, Loss: 0.0006\nIter: 22, Epoch: 199, Loss: 0.0009\nIter: 23, Epoch: 199, Loss: 0.0010\nIter: 24, Epoch: 199, Loss: 0.0006\nIter: 25, Epoch: 199, Loss: 0.0006\nIter: 26, Epoch: 199, Loss: 0.0007\nIter: 27, Epoch: 199, Loss: 0.0003\nIter: 28, Epoch: 199, Loss: 0.0008\nIter: 29, Epoch: 199, Loss: 0.0013\nIter: 30, Epoch: 199, Loss: 0.0009\n","truncated":false}} +%--- +%[output:560c46a2] +% data: {"dataType":"text","outputData":{"text":">> Epoch 199 validation loss: 0.0008\n","truncated":false}} +%--- +%[output:7ad79bda] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 200, Loss: 0.0010\nIter: 2, Epoch: 200, Loss: 0.0006\nIter: 3, Epoch: 200, Loss: 0.0008\nIter: 4, Epoch: 200, Loss: 0.0006\nIter: 5, Epoch: 200, Loss: 0.0005\nIter: 6, Epoch: 200, Loss: 0.0005\nIter: 7, Epoch: 200, Loss: 0.0006\nIter: 8, Epoch: 200, Loss: 0.0010\nIter: 9, Epoch: 200, Loss: 0.0006\nIter: 10, Epoch: 200, Loss: 0.0009\nIter: 11, Epoch: 200, Loss: 0.0004\nIter: 12, Epoch: 200, Loss: 0.0003\nIter: 13, Epoch: 200, Loss: 0.0009\nIter: 14, Epoch: 200, Loss: 0.0009\nIter: 15, Epoch: 200, Loss: 0.0005\nIter: 16, Epoch: 200, Loss: 0.0006\nIter: 17, Epoch: 200, Loss: 0.0007\nIter: 18, Epoch: 200, Loss: 0.0007\nIter: 19, Epoch: 200, Loss: 0.0006\nIter: 20, Epoch: 200, Loss: 0.0007\nIter: 21, Epoch: 200, Loss: 0.0006\nIter: 22, Epoch: 200, Loss: 0.0009\nIter: 23, Epoch: 200, Loss: 0.0009\nIter: 24, Epoch: 200, Loss: 0.0006\nIter: 25, Epoch: 200, Loss: 0.0006\nIter: 26, Epoch: 200, Loss: 0.0007\nIter: 27, Epoch: 200, Loss: 0.0003\nIter: 28, Epoch: 200, Loss: 0.0007\nIter: 29, Epoch: 200, Loss: 0.0013\nIter: 30, Epoch: 200, Loss: 0.0009\n","truncated":false}} +%--- +%[output:158a2958] +% data: {"dataType":"text","outputData":{"text":">> Epoch 200 validation loss: 0.0007\n","truncated":false}} +%--- +%[output:07705166] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 201, Loss: 0.0010\nIter: 2, Epoch: 201, Loss: 0.0006\nIter: 3, Epoch: 201, Loss: 0.0008\nIter: 4, Epoch: 201, Loss: 0.0005\nIter: 5, Epoch: 201, Loss: 0.0005\nIter: 6, Epoch: 201, Loss: 0.0005\nIter: 7, Epoch: 201, Loss: 0.0006\nIter: 8, Epoch: 201, Loss: 0.0010\nIter: 9, Epoch: 201, Loss: 0.0005\nIter: 10, Epoch: 201, Loss: 0.0009\nIter: 11, Epoch: 201, Loss: 0.0004\nIter: 12, Epoch: 201, Loss: 0.0003\nIter: 13, Epoch: 201, Loss: 0.0009\nIter: 14, Epoch: 201, Loss: 0.0008\nIter: 15, Epoch: 201, Loss: 0.0005\nIter: 16, Epoch: 201, Loss: 0.0006\nIter: 17, Epoch: 201, Loss: 0.0007\nIter: 18, Epoch: 201, Loss: 0.0007\nIter: 19, Epoch: 201, Loss: 0.0005\nIter: 20, Epoch: 201, Loss: 0.0007\nIter: 21, Epoch: 201, Loss: 0.0005\nIter: 22, Epoch: 201, Loss: 0.0009\nIter: 23, Epoch: 201, Loss: 0.0009\nIter: 24, Epoch: 201, Loss: 0.0005\nIter: 25, Epoch: 201, Loss: 0.0005\nIter: 26, Epoch: 201, Loss: 0.0007\nIter: 27, Epoch: 201, Loss: 0.0003\nIter: 28, Epoch: 201, Loss: 0.0007\nIter: 29, Epoch: 201, Loss: 0.0012\nIter: 30, Epoch: 201, Loss: 0.0009\n","truncated":false}} +%--- +%[output:5be6a734] +% data: {"dataType":"text","outputData":{"text":">> Epoch 201 validation loss: 0.0007\n","truncated":false}} +%--- +%[output:51aacec6] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 202, Loss: 0.0009\nIter: 2, Epoch: 202, Loss: 0.0006\nIter: 3, Epoch: 202, Loss: 0.0008\nIter: 4, Epoch: 202, Loss: 0.0005\nIter: 5, Epoch: 202, Loss: 0.0005\nIter: 6, Epoch: 202, Loss: 0.0005\nIter: 7, Epoch: 202, Loss: 0.0006\nIter: 8, Epoch: 202, Loss: 0.0010\nIter: 9, Epoch: 202, Loss: 0.0005\nIter: 10, Epoch: 202, Loss: 0.0009\nIter: 11, Epoch: 202, Loss: 0.0004\nIter: 12, Epoch: 202, Loss: 0.0003\nIter: 13, Epoch: 202, Loss: 0.0008\nIter: 14, Epoch: 202, Loss: 0.0008\nIter: 15, Epoch: 202, Loss: 0.0005\nIter: 16, Epoch: 202, Loss: 0.0006\nIter: 17, Epoch: 202, Loss: 0.0007\nIter: 18, Epoch: 202, Loss: 0.0006\nIter: 19, Epoch: 202, Loss: 0.0005\nIter: 20, Epoch: 202, Loss: 0.0007\nIter: 21, Epoch: 202, Loss: 0.0005\nIter: 22, Epoch: 202, Loss: 0.0009\nIter: 23, Epoch: 202, Loss: 0.0008\nIter: 24, Epoch: 202, Loss: 0.0005\nIter: 25, Epoch: 202, Loss: 0.0005\nIter: 26, Epoch: 202, Loss: 0.0007\nIter: 27, Epoch: 202, Loss: 0.0003\nIter: 28, Epoch: 202, Loss: 0.0007\nIter: 29, Epoch: 202, Loss: 0.0012\nIter: 30, Epoch: 202, Loss: 0.0009\n","truncated":false}} +%--- +%[output:669452f8] +% data: {"dataType":"text","outputData":{"text":">> Epoch 202 validation loss: 0.0007\n","truncated":false}} +%--- +%[output:4d0d3fa4] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 203, Loss: 0.0009\nIter: 2, Epoch: 203, Loss: 0.0006\nIter: 3, Epoch: 203, Loss: 0.0007\nIter: 4, Epoch: 203, Loss: 0.0005\nIter: 5, Epoch: 203, Loss: 0.0004\nIter: 6, Epoch: 203, Loss: 0.0005\nIter: 7, Epoch: 203, Loss: 0.0005\nIter: 8, Epoch: 203, Loss: 0.0010\nIter: 9, Epoch: 203, Loss: 0.0005\nIter: 10, Epoch: 203, Loss: 0.0008\nIter: 11, Epoch: 203, Loss: 0.0004\nIter: 12, Epoch: 203, Loss: 0.0003\nIter: 13, Epoch: 203, Loss: 0.0008\nIter: 14, Epoch: 203, Loss: 0.0008\nIter: 15, Epoch: 203, Loss: 0.0005\nIter: 16, Epoch: 203, Loss: 0.0006\nIter: 17, Epoch: 203, Loss: 0.0007\nIter: 18, Epoch: 203, Loss: 0.0006\nIter: 19, Epoch: 203, Loss: 0.0005\nIter: 20, Epoch: 203, Loss: 0.0006\nIter: 21, Epoch: 203, Loss: 0.0005\nIter: 22, Epoch: 203, Loss: 0.0009\nIter: 23, Epoch: 203, Loss: 0.0008\nIter: 24, Epoch: 203, Loss: 0.0005\nIter: 25, Epoch: 203, Loss: 0.0005\nIter: 26, Epoch: 203, Loss: 0.0007\nIter: 27, Epoch: 203, Loss: 0.0003\nIter: 28, Epoch: 203, Loss: 0.0007\nIter: 29, Epoch: 203, Loss: 0.0011\nIter: 30, Epoch: 203, Loss: 0.0009\n","truncated":false}} +%--- +%[output:6532506f] +% data: {"dataType":"text","outputData":{"text":">> Epoch 203 validation loss: 0.0006\n","truncated":false}} +%--- +%[output:9069e957] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 204, Loss: 0.0008\nIter: 2, Epoch: 204, Loss: 0.0005\nIter: 3, Epoch: 204, Loss: 0.0007\nIter: 4, Epoch: 204, Loss: 0.0005\nIter: 5, Epoch: 204, Loss: 0.0004\nIter: 6, Epoch: 204, Loss: 0.0005\nIter: 7, Epoch: 204, Loss: 0.0005\nIter: 8, Epoch: 204, Loss: 0.0010\nIter: 9, Epoch: 204, Loss: 0.0005\nIter: 10, Epoch: 204, Loss: 0.0008\nIter: 11, Epoch: 204, Loss: 0.0004\nIter: 12, Epoch: 204, Loss: 0.0003\nIter: 13, Epoch: 204, Loss: 0.0008\nIter: 14, Epoch: 204, Loss: 0.0008\nIter: 15, Epoch: 204, Loss: 0.0005\nIter: 16, Epoch: 204, Loss: 0.0006\nIter: 17, Epoch: 204, Loss: 0.0006\nIter: 18, Epoch: 204, Loss: 0.0006\nIter: 19, Epoch: 204, Loss: 0.0005\nIter: 20, Epoch: 204, Loss: 0.0006\nIter: 21, Epoch: 204, Loss: 0.0005\nIter: 22, Epoch: 204, Loss: 0.0009\nIter: 23, Epoch: 204, Loss: 0.0008\nIter: 24, Epoch: 204, Loss: 0.0004\nIter: 25, Epoch: 204, Loss: 0.0005\nIter: 26, Epoch: 204, Loss: 0.0007\nIter: 27, Epoch: 204, Loss: 0.0003\nIter: 28, Epoch: 204, Loss: 0.0007\nIter: 29, Epoch: 204, Loss: 0.0011\nIter: 30, Epoch: 204, Loss: 0.0008\n","truncated":false}} +%--- +%[output:168fd3da] +% data: {"dataType":"text","outputData":{"text":">> Epoch 204 validation loss: 0.0006\n","truncated":false}} +%--- +%[output:8510bcb5] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 205, Loss: 0.0008\nIter: 2, Epoch: 205, Loss: 0.0005\nIter: 3, Epoch: 205, Loss: 0.0007\nIter: 4, Epoch: 205, Loss: 0.0005\nIter: 5, Epoch: 205, Loss: 0.0004\nIter: 6, Epoch: 205, Loss: 0.0004\nIter: 7, Epoch: 205, Loss: 0.0005\nIter: 8, Epoch: 205, Loss: 0.0010\nIter: 9, Epoch: 205, Loss: 0.0005\nIter: 10, Epoch: 205, Loss: 0.0008\nIter: 11, Epoch: 205, Loss: 0.0004\nIter: 12, Epoch: 205, Loss: 0.0003\nIter: 13, Epoch: 205, Loss: 0.0008\nIter: 14, Epoch: 205, Loss: 0.0007\nIter: 15, Epoch: 205, Loss: 0.0005\nIter: 16, Epoch: 205, Loss: 0.0006\nIter: 17, Epoch: 205, Loss: 0.0006\nIter: 18, Epoch: 205, Loss: 0.0006\nIter: 19, Epoch: 205, Loss: 0.0005\nIter: 20, Epoch: 205, Loss: 0.0006\nIter: 21, Epoch: 205, Loss: 0.0005\nIter: 22, Epoch: 205, Loss: 0.0009\nIter: 23, Epoch: 205, Loss: 0.0008\nIter: 24, Epoch: 205, Loss: 0.0004\nIter: 25, Epoch: 205, Loss: 0.0004\nIter: 26, Epoch: 205, Loss: 0.0006\nIter: 27, Epoch: 205, Loss: 0.0003\nIter: 28, Epoch: 205, Loss: 0.0006\nIter: 29, Epoch: 205, Loss: 0.0010\nIter: 30, Epoch: 205, Loss: 0.0008\n","truncated":false}} +%--- +%[output:264b7315] +% data: {"dataType":"text","outputData":{"text":">> Epoch 205 validation loss: 0.0006\n","truncated":false}} +%--- +%[output:52fe6de5] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 206, Loss: 0.0008\nIter: 2, Epoch: 206, Loss: 0.0005\nIter: 3, Epoch: 206, Loss: 0.0007\nIter: 4, Epoch: 206, Loss: 0.0005\nIter: 5, Epoch: 206, Loss: 0.0004\nIter: 6, Epoch: 206, Loss: 0.0004\nIter: 7, Epoch: 206, Loss: 0.0005\nIter: 8, Epoch: 206, Loss: 0.0010\nIter: 9, Epoch: 206, Loss: 0.0005\nIter: 10, Epoch: 206, Loss: 0.0008\nIter: 11, Epoch: 206, Loss: 0.0004\nIter: 12, Epoch: 206, Loss: 0.0003\nIter: 13, Epoch: 206, Loss: 0.0007\nIter: 14, Epoch: 206, Loss: 0.0007\nIter: 15, Epoch: 206, Loss: 0.0005\nIter: 16, Epoch: 206, Loss: 0.0006\nIter: 17, Epoch: 206, Loss: 0.0006\nIter: 18, Epoch: 206, Loss: 0.0006\nIter: 19, Epoch: 206, Loss: 0.0005\nIter: 20, Epoch: 206, Loss: 0.0006\nIter: 21, Epoch: 206, Loss: 0.0005\nIter: 22, Epoch: 206, Loss: 0.0009\nIter: 23, Epoch: 206, Loss: 0.0008\nIter: 24, Epoch: 206, Loss: 0.0004\nIter: 25, Epoch: 206, Loss: 0.0004\nIter: 26, Epoch: 206, Loss: 0.0006\nIter: 27, Epoch: 206, Loss: 0.0003\nIter: 28, Epoch: 206, Loss: 0.0006\nIter: 29, Epoch: 206, Loss: 0.0010\nIter: 30, Epoch: 206, Loss: 0.0008\n","truncated":false}} +%--- +%[output:2cfcd015] +% data: {"dataType":"text","outputData":{"text":">> Epoch 206 validation loss: 0.0006\n","truncated":false}} +%--- +%[output:7f5b687f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 207, Loss: 0.0008\nIter: 2, Epoch: 207, Loss: 0.0005\nIter: 3, Epoch: 207, Loss: 0.0006\nIter: 4, Epoch: 207, Loss: 0.0005\nIter: 5, Epoch: 207, Loss: 0.0004\nIter: 6, Epoch: 207, Loss: 0.0004\nIter: 7, Epoch: 207, Loss: 0.0005\nIter: 8, Epoch: 207, Loss: 0.0010\nIter: 9, Epoch: 207, Loss: 0.0005\nIter: 10, Epoch: 207, Loss: 0.0008\nIter: 11, Epoch: 207, Loss: 0.0004\nIter: 12, Epoch: 207, Loss: 0.0003\nIter: 13, Epoch: 207, Loss: 0.0007\nIter: 14, Epoch: 207, Loss: 0.0007\nIter: 15, Epoch: 207, Loss: 0.0005\nIter: 16, Epoch: 207, Loss: 0.0006\nIter: 17, Epoch: 207, Loss: 0.0006\nIter: 18, Epoch: 207, Loss: 0.0006\nIter: 19, Epoch: 207, Loss: 0.0004\nIter: 20, Epoch: 207, Loss: 0.0006\nIter: 21, Epoch: 207, Loss: 0.0005\nIter: 22, Epoch: 207, Loss: 0.0009\nIter: 23, Epoch: 207, Loss: 0.0008\nIter: 24, Epoch: 207, Loss: 0.0004\nIter: 25, Epoch: 207, Loss: 0.0004\nIter: 26, Epoch: 207, Loss: 0.0006\nIter: 27, Epoch: 207, Loss: 0.0003\nIter: 28, Epoch: 207, Loss: 0.0006\nIter: 29, Epoch: 207, Loss: 0.0010\nIter: 30, Epoch: 207, Loss: 0.0008\n","truncated":false}} +%--- +%[output:55f67715] +% data: {"dataType":"text","outputData":{"text":">> Epoch 207 validation loss: 0.0006\n","truncated":false}} +%--- +%[output:17689972] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 208, Loss: 0.0007\nIter: 2, Epoch: 208, Loss: 0.0005\nIter: 3, Epoch: 208, Loss: 0.0006\nIter: 4, Epoch: 208, Loss: 0.0005\nIter: 5, Epoch: 208, Loss: 0.0004\nIter: 6, Epoch: 208, Loss: 0.0004\nIter: 7, Epoch: 208, Loss: 0.0005\nIter: 8, Epoch: 208, Loss: 0.0010\nIter: 9, Epoch: 208, Loss: 0.0005\nIter: 10, Epoch: 208, Loss: 0.0008\nIter: 11, Epoch: 208, Loss: 0.0004\nIter: 12, Epoch: 208, Loss: 0.0003\nIter: 13, Epoch: 208, Loss: 0.0007\nIter: 14, Epoch: 208, Loss: 0.0007\nIter: 15, Epoch: 208, Loss: 0.0005\nIter: 16, Epoch: 208, Loss: 0.0006\nIter: 17, Epoch: 208, Loss: 0.0006\nIter: 18, Epoch: 208, Loss: 0.0006\nIter: 19, Epoch: 208, Loss: 0.0004\nIter: 20, Epoch: 208, Loss: 0.0006\nIter: 21, Epoch: 208, Loss: 0.0004\nIter: 22, Epoch: 208, Loss: 0.0009\nIter: 23, Epoch: 208, Loss: 0.0007\nIter: 24, Epoch: 208, Loss: 0.0004\nIter: 25, Epoch: 208, Loss: 0.0004\nIter: 26, Epoch: 208, Loss: 0.0006\nIter: 27, Epoch: 208, Loss: 0.0003\nIter: 28, Epoch: 208, Loss: 0.0006\nIter: 29, Epoch: 208, Loss: 0.0010\nIter: 30, Epoch: 208, Loss: 0.0008\n","truncated":false}} +%--- +%[output:924dfbe9] +% data: {"dataType":"text","outputData":{"text":">> Epoch 208 validation loss: 0.0006\n","truncated":false}} +%--- +%[output:94236b53] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 209, Loss: 0.0007\nIter: 2, Epoch: 209, Loss: 0.0005\nIter: 3, Epoch: 209, Loss: 0.0006\nIter: 4, Epoch: 209, Loss: 0.0005\nIter: 5, Epoch: 209, Loss: 0.0004\nIter: 6, Epoch: 209, Loss: 0.0004\nIter: 7, Epoch: 209, Loss: 0.0005\nIter: 8, Epoch: 209, Loss: 0.0010\nIter: 9, Epoch: 209, Loss: 0.0005\nIter: 10, Epoch: 209, Loss: 0.0007\nIter: 11, Epoch: 209, Loss: 0.0004\nIter: 12, Epoch: 209, Loss: 0.0003\nIter: 13, Epoch: 209, Loss: 0.0007\nIter: 14, Epoch: 209, Loss: 0.0007\nIter: 15, Epoch: 209, Loss: 0.0005\nIter: 16, Epoch: 209, Loss: 0.0005\nIter: 17, Epoch: 209, Loss: 0.0006\nIter: 18, Epoch: 209, Loss: 0.0006\nIter: 19, Epoch: 209, Loss: 0.0004\nIter: 20, Epoch: 209, Loss: 0.0006\nIter: 21, Epoch: 209, Loss: 0.0004\nIter: 22, Epoch: 209, Loss: 0.0009\nIter: 23, Epoch: 209, Loss: 0.0007\nIter: 24, Epoch: 209, Loss: 0.0004\nIter: 25, Epoch: 209, Loss: 0.0004\nIter: 26, Epoch: 209, Loss: 0.0006\nIter: 27, Epoch: 209, Loss: 0.0003\nIter: 28, Epoch: 209, Loss: 0.0006\nIter: 29, Epoch: 209, Loss: 0.0009\nIter: 30, Epoch: 209, Loss: 0.0008\n","truncated":false}} +%--- +%[output:5568e291] +% data: {"dataType":"text","outputData":{"text":">> Epoch 209 validation loss: 0.0006\n","truncated":false}} +%--- +%[output:71de17e9] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 210, Loss: 0.0007\nIter: 2, Epoch: 210, Loss: 0.0005\nIter: 3, Epoch: 210, Loss: 0.0006\nIter: 4, Epoch: 210, Loss: 0.0005\nIter: 5, Epoch: 210, Loss: 0.0004\nIter: 6, Epoch: 210, Loss: 0.0004\nIter: 7, Epoch: 210, Loss: 0.0005\nIter: 8, Epoch: 210, Loss: 0.0010\nIter: 9, Epoch: 210, Loss: 0.0005\nIter: 10, Epoch: 210, Loss: 0.0007\nIter: 11, Epoch: 210, Loss: 0.0003\nIter: 12, Epoch: 210, Loss: 0.0003\nIter: 13, Epoch: 210, Loss: 0.0007\nIter: 14, Epoch: 210, Loss: 0.0007\nIter: 15, Epoch: 210, Loss: 0.0004\nIter: 16, Epoch: 210, Loss: 0.0005\nIter: 17, Epoch: 210, Loss: 0.0006\nIter: 18, Epoch: 210, Loss: 0.0006\nIter: 19, Epoch: 210, Loss: 0.0004\nIter: 20, Epoch: 210, Loss: 0.0006\nIter: 21, Epoch: 210, Loss: 0.0004\nIter: 22, Epoch: 210, Loss: 0.0009\nIter: 23, Epoch: 210, Loss: 0.0007\nIter: 24, Epoch: 210, Loss: 0.0003\nIter: 25, Epoch: 210, Loss: 0.0004\nIter: 26, Epoch: 210, Loss: 0.0006\nIter: 27, Epoch: 210, Loss: 0.0003\nIter: 28, Epoch: 210, Loss: 0.0006\nIter: 29, Epoch: 210, Loss: 0.0009\nIter: 30, Epoch: 210, Loss: 0.0008\n","truncated":false}} +%--- +%[output:7418a86d] +% data: {"dataType":"text","outputData":{"text":">> Epoch 210 validation loss: 0.0006\n","truncated":false}} +%--- +%[output:7a589bc7] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 211, Loss: 0.0007\nIter: 2, Epoch: 211, Loss: 0.0005\nIter: 3, Epoch: 211, Loss: 0.0006\nIter: 4, Epoch: 211, Loss: 0.0005\nIter: 5, Epoch: 211, Loss: 0.0004\nIter: 6, Epoch: 211, Loss: 0.0004\nIter: 7, Epoch: 211, Loss: 0.0005\nIter: 8, Epoch: 211, Loss: 0.0010\nIter: 9, Epoch: 211, Loss: 0.0005\nIter: 10, Epoch: 211, Loss: 0.0007\nIter: 11, Epoch: 211, Loss: 0.0003\nIter: 12, Epoch: 211, Loss: 0.0003\nIter: 13, Epoch: 211, Loss: 0.0007\nIter: 14, Epoch: 211, Loss: 0.0007\nIter: 15, Epoch: 211, Loss: 0.0004\nIter: 16, Epoch: 211, Loss: 0.0005\nIter: 17, Epoch: 211, Loss: 0.0005\nIter: 18, Epoch: 211, Loss: 0.0006\nIter: 19, Epoch: 211, Loss: 0.0004\nIter: 20, Epoch: 211, Loss: 0.0006\nIter: 21, Epoch: 211, Loss: 0.0004\nIter: 22, Epoch: 211, Loss: 0.0009\nIter: 23, Epoch: 211, Loss: 0.0007\nIter: 24, Epoch: 211, Loss: 0.0003\nIter: 25, Epoch: 211, Loss: 0.0004\nIter: 26, Epoch: 211, Loss: 0.0006\nIter: 27, Epoch: 211, Loss: 0.0003\nIter: 28, Epoch: 211, Loss: 0.0006\nIter: 29, Epoch: 211, Loss: 0.0009\nIter: 30, Epoch: 211, Loss: 0.0008\n","truncated":false}} +%--- +%[output:9add4627] +% data: {"dataType":"text","outputData":{"text":">> Epoch 211 validation loss: 0.0006\n","truncated":false}} +%--- +%[output:1eeb0a4c] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 212, Loss: 0.0007\nIter: 2, Epoch: 212, Loss: 0.0005\nIter: 3, Epoch: 212, Loss: 0.0006\nIter: 4, Epoch: 212, Loss: 0.0004\nIter: 5, Epoch: 212, Loss: 0.0004\nIter: 6, Epoch: 212, Loss: 0.0004\nIter: 7, Epoch: 212, Loss: 0.0005\nIter: 8, Epoch: 212, Loss: 0.0010\nIter: 9, Epoch: 212, Loss: 0.0005\nIter: 10, Epoch: 212, Loss: 0.0007\nIter: 11, Epoch: 212, Loss: 0.0003\nIter: 12, Epoch: 212, Loss: 0.0003\nIter: 13, Epoch: 212, Loss: 0.0007\nIter: 14, Epoch: 212, Loss: 0.0007\nIter: 15, Epoch: 212, Loss: 0.0004\nIter: 16, Epoch: 212, Loss: 0.0005\nIter: 17, Epoch: 212, Loss: 0.0005\nIter: 18, Epoch: 212, Loss: 0.0006\nIter: 19, Epoch: 212, Loss: 0.0004\nIter: 20, Epoch: 212, Loss: 0.0006\nIter: 21, Epoch: 212, Loss: 0.0004\nIter: 22, Epoch: 212, Loss: 0.0009\nIter: 23, Epoch: 212, Loss: 0.0007\nIter: 24, Epoch: 212, Loss: 0.0003\nIter: 25, Epoch: 212, Loss: 0.0004\nIter: 26, Epoch: 212, Loss: 0.0006\nIter: 27, Epoch: 212, Loss: 0.0003\nIter: 28, Epoch: 212, Loss: 0.0006\nIter: 29, Epoch: 212, Loss: 0.0009\nIter: 30, Epoch: 212, Loss: 0.0008\n","truncated":false}} +%--- +%[output:874eaa08] +% data: {"dataType":"text","outputData":{"text":">> Epoch 212 validation loss: 0.0006\n","truncated":false}} +%--- +%[output:509ab337] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 213, Loss: 0.0007\nIter: 2, Epoch: 213, Loss: 0.0005\nIter: 3, Epoch: 213, Loss: 0.0006\nIter: 4, Epoch: 213, Loss: 0.0004\nIter: 5, Epoch: 213, Loss: 0.0004\nIter: 6, Epoch: 213, Loss: 0.0004\nIter: 7, Epoch: 213, Loss: 0.0005\nIter: 8, Epoch: 213, Loss: 0.0010\nIter: 9, Epoch: 213, Loss: 0.0005\nIter: 10, Epoch: 213, Loss: 0.0007\nIter: 11, Epoch: 213, Loss: 0.0003\nIter: 12, Epoch: 213, Loss: 0.0003\nIter: 13, Epoch: 213, Loss: 0.0007\nIter: 14, Epoch: 213, Loss: 0.0007\nIter: 15, Epoch: 213, Loss: 0.0004\nIter: 16, Epoch: 213, Loss: 0.0005\nIter: 17, Epoch: 213, Loss: 0.0005\nIter: 18, Epoch: 213, Loss: 0.0006\nIter: 19, Epoch: 213, Loss: 0.0004\nIter: 20, Epoch: 213, Loss: 0.0006\nIter: 21, Epoch: 213, Loss: 0.0004\nIter: 22, Epoch: 213, Loss: 0.0009\nIter: 23, Epoch: 213, Loss: 0.0007\nIter: 24, Epoch: 213, Loss: 0.0003\nIter: 25, Epoch: 213, Loss: 0.0004\nIter: 26, Epoch: 213, Loss: 0.0006\nIter: 27, Epoch: 213, Loss: 0.0003\nIter: 28, Epoch: 213, Loss: 0.0006\nIter: 29, Epoch: 213, Loss: 0.0009\nIter: 30, Epoch: 213, Loss: 0.0008\n","truncated":false}} +%--- +%[output:80a43eea] +% data: {"dataType":"text","outputData":{"text":">> Epoch 213 validation loss: 0.0006\n","truncated":false}} +%--- +%[output:8e64767f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 214, Loss: 0.0007\nIter: 2, Epoch: 214, Loss: 0.0005\nIter: 3, Epoch: 214, Loss: 0.0006\nIter: 4, Epoch: 214, Loss: 0.0004\nIter: 5, Epoch: 214, Loss: 0.0004\nIter: 6, Epoch: 214, Loss: 0.0004\nIter: 7, Epoch: 214, Loss: 0.0005\nIter: 8, Epoch: 214, Loss: 0.0010\nIter: 9, Epoch: 214, Loss: 0.0005\nIter: 10, Epoch: 214, Loss: 0.0007\nIter: 11, Epoch: 214, Loss: 0.0003\nIter: 12, Epoch: 214, Loss: 0.0003\nIter: 13, Epoch: 214, Loss: 0.0007\nIter: 14, Epoch: 214, Loss: 0.0007\nIter: 15, Epoch: 214, Loss: 0.0004\nIter: 16, Epoch: 214, Loss: 0.0005\nIter: 17, Epoch: 214, Loss: 0.0005\nIter: 18, Epoch: 214, Loss: 0.0006\nIter: 19, Epoch: 214, Loss: 0.0004\nIter: 20, Epoch: 214, Loss: 0.0006\nIter: 21, Epoch: 214, Loss: 0.0004\nIter: 22, Epoch: 214, Loss: 0.0009\nIter: 23, Epoch: 214, Loss: 0.0007\nIter: 24, Epoch: 214, Loss: 0.0003\nIter: 25, Epoch: 214, Loss: 0.0004\nIter: 26, Epoch: 214, Loss: 0.0006\nIter: 27, Epoch: 214, Loss: 0.0003\nIter: 28, Epoch: 214, Loss: 0.0006\nIter: 29, Epoch: 214, Loss: 0.0009\nIter: 30, Epoch: 214, Loss: 0.0008\n","truncated":false}} +%--- +%[output:89d7f4d6] +% data: {"dataType":"text","outputData":{"text":">> Epoch 214 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:2c37d46c] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 215, Loss: 0.0007\nIter: 2, Epoch: 215, Loss: 0.0005\nIter: 3, Epoch: 215, Loss: 0.0006\nIter: 4, Epoch: 215, Loss: 0.0004\nIter: 5, Epoch: 215, Loss: 0.0004\nIter: 6, Epoch: 215, Loss: 0.0004\nIter: 7, Epoch: 215, Loss: 0.0004\nIter: 8, Epoch: 215, Loss: 0.0010\nIter: 9, Epoch: 215, Loss: 0.0005\nIter: 10, Epoch: 215, Loss: 0.0007\nIter: 11, Epoch: 215, Loss: 0.0003\nIter: 12, Epoch: 215, Loss: 0.0003\nIter: 13, Epoch: 215, Loss: 0.0007\nIter: 14, Epoch: 215, Loss: 0.0007\nIter: 15, Epoch: 215, Loss: 0.0004\nIter: 16, Epoch: 215, Loss: 0.0005\nIter: 17, Epoch: 215, Loss: 0.0005\nIter: 18, Epoch: 215, Loss: 0.0005\nIter: 19, Epoch: 215, Loss: 0.0004\nIter: 20, Epoch: 215, Loss: 0.0006\nIter: 21, Epoch: 215, Loss: 0.0004\nIter: 22, Epoch: 215, Loss: 0.0009\nIter: 23, Epoch: 215, Loss: 0.0007\nIter: 24, Epoch: 215, Loss: 0.0003\nIter: 25, Epoch: 215, Loss: 0.0004\nIter: 26, Epoch: 215, Loss: 0.0006\nIter: 27, Epoch: 215, Loss: 0.0003\nIter: 28, Epoch: 215, Loss: 0.0006\nIter: 29, Epoch: 215, Loss: 0.0009\nIter: 30, Epoch: 215, Loss: 0.0008\n","truncated":false}} +%--- +%[output:2212ccd6] +% data: {"dataType":"text","outputData":{"text":">> Epoch 215 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:4c553f76] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 216, Loss: 0.0006\nIter: 2, Epoch: 216, Loss: 0.0005\nIter: 3, Epoch: 216, Loss: 0.0006\nIter: 4, Epoch: 216, Loss: 0.0004\nIter: 5, Epoch: 216, Loss: 0.0004\nIter: 6, Epoch: 216, Loss: 0.0004\nIter: 7, Epoch: 216, Loss: 0.0004\nIter: 8, Epoch: 216, Loss: 0.0010\nIter: 9, Epoch: 216, Loss: 0.0005\nIter: 10, Epoch: 216, Loss: 0.0007\nIter: 11, Epoch: 216, Loss: 0.0003\nIter: 12, Epoch: 216, Loss: 0.0003\nIter: 13, Epoch: 216, Loss: 0.0006\nIter: 14, Epoch: 216, Loss: 0.0007\nIter: 15, Epoch: 216, Loss: 0.0004\nIter: 16, Epoch: 216, Loss: 0.0005\nIter: 17, Epoch: 216, Loss: 0.0005\nIter: 18, Epoch: 216, Loss: 0.0005\nIter: 19, Epoch: 216, Loss: 0.0004\nIter: 20, Epoch: 216, Loss: 0.0006\nIter: 21, Epoch: 216, Loss: 0.0004\nIter: 22, Epoch: 216, Loss: 0.0009\nIter: 23, Epoch: 216, Loss: 0.0007\nIter: 24, Epoch: 216, Loss: 0.0003\nIter: 25, Epoch: 216, Loss: 0.0004\nIter: 26, Epoch: 216, Loss: 0.0006\nIter: 27, Epoch: 216, Loss: 0.0003\nIter: 28, Epoch: 216, Loss: 0.0006\nIter: 29, Epoch: 216, Loss: 0.0009\nIter: 30, Epoch: 216, Loss: 0.0008\n","truncated":false}} +%--- +%[output:4aca68fb] +% data: {"dataType":"text","outputData":{"text":">> Epoch 216 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:4d80af44] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 217, Loss: 0.0006\nIter: 2, Epoch: 217, Loss: 0.0005\nIter: 3, Epoch: 217, Loss: 0.0006\nIter: 4, Epoch: 217, Loss: 0.0004\nIter: 5, Epoch: 217, Loss: 0.0004\nIter: 6, Epoch: 217, Loss: 0.0004\nIter: 7, Epoch: 217, Loss: 0.0004\nIter: 8, Epoch: 217, Loss: 0.0010\nIter: 9, Epoch: 217, Loss: 0.0005\nIter: 10, Epoch: 217, Loss: 0.0007\nIter: 11, Epoch: 217, Loss: 0.0003\nIter: 12, Epoch: 217, Loss: 0.0003\nIter: 13, Epoch: 217, Loss: 0.0006\nIter: 14, Epoch: 217, Loss: 0.0006\nIter: 15, Epoch: 217, Loss: 0.0004\nIter: 16, Epoch: 217, Loss: 0.0005\nIter: 17, Epoch: 217, Loss: 0.0005\nIter: 18, Epoch: 217, Loss: 0.0005\nIter: 19, Epoch: 217, Loss: 0.0004\nIter: 20, Epoch: 217, Loss: 0.0005\nIter: 21, Epoch: 217, Loss: 0.0004\nIter: 22, Epoch: 217, Loss: 0.0009\nIter: 23, Epoch: 217, Loss: 0.0007\nIter: 24, Epoch: 217, Loss: 0.0003\nIter: 25, Epoch: 217, Loss: 0.0004\nIter: 26, Epoch: 217, Loss: 0.0006\nIter: 27, Epoch: 217, Loss: 0.0003\nIter: 28, Epoch: 217, Loss: 0.0006\nIter: 29, Epoch: 217, Loss: 0.0008\nIter: 30, Epoch: 217, Loss: 0.0008\n","truncated":false}} +%--- +%[output:40ab6cf7] +% data: {"dataType":"text","outputData":{"text":">> Epoch 217 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:53be6a8e] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 218, Loss: 0.0006\nIter: 2, Epoch: 218, Loss: 0.0005\nIter: 3, Epoch: 218, Loss: 0.0005\nIter: 4, Epoch: 218, Loss: 0.0004\nIter: 5, Epoch: 218, Loss: 0.0004\nIter: 6, Epoch: 218, Loss: 0.0004\nIter: 7, Epoch: 218, Loss: 0.0004\nIter: 8, Epoch: 218, Loss: 0.0010\nIter: 9, Epoch: 218, Loss: 0.0005\nIter: 10, Epoch: 218, Loss: 0.0007\nIter: 11, Epoch: 218, Loss: 0.0003\nIter: 12, Epoch: 218, Loss: 0.0003\nIter: 13, Epoch: 218, Loss: 0.0006\nIter: 14, Epoch: 218, Loss: 0.0006\nIter: 15, Epoch: 218, Loss: 0.0004\nIter: 16, Epoch: 218, Loss: 0.0005\nIter: 17, Epoch: 218, Loss: 0.0005\nIter: 18, Epoch: 218, Loss: 0.0005\nIter: 19, Epoch: 218, Loss: 0.0004\nIter: 20, Epoch: 218, Loss: 0.0005\nIter: 21, Epoch: 218, Loss: 0.0004\nIter: 22, Epoch: 218, Loss: 0.0009\nIter: 23, Epoch: 218, Loss: 0.0007\nIter: 24, Epoch: 218, Loss: 0.0003\nIter: 25, Epoch: 218, Loss: 0.0004\nIter: 26, Epoch: 218, Loss: 0.0006\nIter: 27, Epoch: 218, Loss: 0.0003\nIter: 28, Epoch: 218, Loss: 0.0006\nIter: 29, Epoch: 218, Loss: 0.0008\nIter: 30, Epoch: 218, Loss: 0.0008\n","truncated":false}} +%--- +%[output:8aca6872] +% data: {"dataType":"text","outputData":{"text":">> Epoch 218 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:4e8cf01d] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 219, Loss: 0.0006\nIter: 2, Epoch: 219, Loss: 0.0005\nIter: 3, Epoch: 219, Loss: 0.0005\nIter: 4, Epoch: 219, Loss: 0.0004\nIter: 5, Epoch: 219, Loss: 0.0004\nIter: 6, Epoch: 219, Loss: 0.0004\nIter: 7, Epoch: 219, Loss: 0.0004\nIter: 8, Epoch: 219, Loss: 0.0010\nIter: 9, Epoch: 219, Loss: 0.0005\nIter: 10, Epoch: 219, Loss: 0.0007\nIter: 11, Epoch: 219, Loss: 0.0003\nIter: 12, Epoch: 219, Loss: 0.0003\nIter: 13, Epoch: 219, Loss: 0.0006\nIter: 14, Epoch: 219, Loss: 0.0006\nIter: 15, Epoch: 219, Loss: 0.0004\nIter: 16, Epoch: 219, Loss: 0.0005\nIter: 17, Epoch: 219, Loss: 0.0005\nIter: 18, Epoch: 219, Loss: 0.0005\nIter: 19, Epoch: 219, Loss: 0.0004\nIter: 20, Epoch: 219, Loss: 0.0005\nIter: 21, Epoch: 219, Loss: 0.0004\nIter: 22, Epoch: 219, Loss: 0.0009\nIter: 23, Epoch: 219, Loss: 0.0007\nIter: 24, Epoch: 219, Loss: 0.0003\nIter: 25, Epoch: 219, Loss: 0.0004\nIter: 26, Epoch: 219, Loss: 0.0006\nIter: 27, Epoch: 219, Loss: 0.0003\nIter: 28, Epoch: 219, Loss: 0.0006\nIter: 29, Epoch: 219, Loss: 0.0008\nIter: 30, Epoch: 219, Loss: 0.0008\n","truncated":false}} +%--- +%[output:19b876f7] +% data: {"dataType":"text","outputData":{"text":">> Epoch 219 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:25607ab4] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 220, Loss: 0.0006\nIter: 2, Epoch: 220, Loss: 0.0005\nIter: 3, Epoch: 220, Loss: 0.0005\nIter: 4, Epoch: 220, Loss: 0.0004\nIter: 5, Epoch: 220, Loss: 0.0004\nIter: 6, Epoch: 220, Loss: 0.0004\nIter: 7, Epoch: 220, Loss: 0.0004\nIter: 8, Epoch: 220, Loss: 0.0010\nIter: 9, Epoch: 220, Loss: 0.0005\nIter: 10, Epoch: 220, Loss: 0.0007\nIter: 11, Epoch: 220, Loss: 0.0003\nIter: 12, Epoch: 220, Loss: 0.0003\nIter: 13, Epoch: 220, Loss: 0.0006\nIter: 14, Epoch: 220, Loss: 0.0006\nIter: 15, Epoch: 220, Loss: 0.0004\nIter: 16, Epoch: 220, Loss: 0.0005\nIter: 17, Epoch: 220, Loss: 0.0005\nIter: 18, Epoch: 220, Loss: 0.0005\nIter: 19, Epoch: 220, Loss: 0.0004\nIter: 20, Epoch: 220, Loss: 0.0005\nIter: 21, Epoch: 220, Loss: 0.0004\nIter: 22, Epoch: 220, Loss: 0.0009\nIter: 23, Epoch: 220, Loss: 0.0007\nIter: 24, Epoch: 220, Loss: 0.0003\nIter: 25, Epoch: 220, Loss: 0.0004\nIter: 26, Epoch: 220, Loss: 0.0006\nIter: 27, Epoch: 220, Loss: 0.0003\nIter: 28, Epoch: 220, Loss: 0.0006\nIter: 29, Epoch: 220, Loss: 0.0008\nIter: 30, Epoch: 220, Loss: 0.0008\n","truncated":false}} +%--- +%[output:86d054d6] +% data: {"dataType":"text","outputData":{"text":">> Epoch 220 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:8aede320] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 221, Loss: 0.0006\nIter: 2, Epoch: 221, Loss: 0.0005\nIter: 3, Epoch: 221, Loss: 0.0005\nIter: 4, Epoch: 221, Loss: 0.0004\nIter: 5, Epoch: 221, Loss: 0.0004\nIter: 6, Epoch: 221, Loss: 0.0004\nIter: 7, Epoch: 221, Loss: 0.0004\nIter: 8, Epoch: 221, Loss: 0.0010\nIter: 9, Epoch: 221, Loss: 0.0005\nIter: 10, Epoch: 221, Loss: 0.0007\nIter: 11, Epoch: 221, Loss: 0.0003\nIter: 12, Epoch: 221, Loss: 0.0003\nIter: 13, Epoch: 221, Loss: 0.0006\nIter: 14, Epoch: 221, Loss: 0.0007\nIter: 15, Epoch: 221, Loss: 0.0004\nIter: 16, Epoch: 221, Loss: 0.0005\nIter: 17, Epoch: 221, Loss: 0.0005\nIter: 18, Epoch: 221, Loss: 0.0005\nIter: 19, Epoch: 221, Loss: 0.0004\nIter: 20, Epoch: 221, Loss: 0.0005\nIter: 21, Epoch: 221, Loss: 0.0004\nIter: 22, Epoch: 221, Loss: 0.0009\nIter: 23, Epoch: 221, Loss: 0.0007\nIter: 24, Epoch: 221, Loss: 0.0003\nIter: 25, Epoch: 221, Loss: 0.0003\nIter: 26, Epoch: 221, Loss: 0.0006\nIter: 27, Epoch: 221, Loss: 0.0003\nIter: 28, Epoch: 221, Loss: 0.0006\nIter: 29, Epoch: 221, Loss: 0.0008\nIter: 30, Epoch: 221, Loss: 0.0008\n","truncated":false}} +%--- +%[output:51bf7e05] +% data: {"dataType":"text","outputData":{"text":">> Epoch 221 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:77d2493d] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 222, Loss: 0.0006\nIter: 2, Epoch: 222, Loss: 0.0005\nIter: 3, Epoch: 222, Loss: 0.0005\nIter: 4, Epoch: 222, Loss: 0.0004\nIter: 5, Epoch: 222, Loss: 0.0004\nIter: 6, Epoch: 222, Loss: 0.0004\nIter: 7, Epoch: 222, Loss: 0.0004\nIter: 8, Epoch: 222, Loss: 0.0010\nIter: 9, Epoch: 222, Loss: 0.0005\nIter: 10, Epoch: 222, Loss: 0.0007\nIter: 11, Epoch: 222, Loss: 0.0003\nIter: 12, Epoch: 222, Loss: 0.0003\nIter: 13, Epoch: 222, Loss: 0.0006\nIter: 14, Epoch: 222, Loss: 0.0007\nIter: 15, Epoch: 222, Loss: 0.0004\nIter: 16, Epoch: 222, Loss: 0.0005\nIter: 17, Epoch: 222, Loss: 0.0005\nIter: 18, Epoch: 222, Loss: 0.0005\nIter: 19, Epoch: 222, Loss: 0.0004\nIter: 20, Epoch: 222, Loss: 0.0005\nIter: 21, Epoch: 222, Loss: 0.0004\nIter: 22, Epoch: 222, Loss: 0.0009\nIter: 23, Epoch: 222, Loss: 0.0007\nIter: 24, Epoch: 222, Loss: 0.0003\nIter: 25, Epoch: 222, Loss: 0.0003\nIter: 26, Epoch: 222, Loss: 0.0006\nIter: 27, Epoch: 222, Loss: 0.0003\nIter: 28, Epoch: 222, Loss: 0.0006\nIter: 29, Epoch: 222, Loss: 0.0008\nIter: 30, Epoch: 222, Loss: 0.0008\n","truncated":false}} +%--- +%[output:9b07118b] +% data: {"dataType":"text","outputData":{"text":">> Epoch 222 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:390d1015] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 223, Loss: 0.0006\nIter: 2, Epoch: 223, Loss: 0.0005\nIter: 3, Epoch: 223, Loss: 0.0005\nIter: 4, Epoch: 223, Loss: 0.0004\nIter: 5, Epoch: 223, Loss: 0.0004\nIter: 6, Epoch: 223, Loss: 0.0004\nIter: 7, Epoch: 223, Loss: 0.0004\nIter: 8, Epoch: 223, Loss: 0.0010\nIter: 9, Epoch: 223, Loss: 0.0005\nIter: 10, Epoch: 223, Loss: 0.0007\nIter: 11, Epoch: 223, Loss: 0.0003\nIter: 12, Epoch: 223, Loss: 0.0003\nIter: 13, Epoch: 223, Loss: 0.0006\nIter: 14, Epoch: 223, Loss: 0.0007\nIter: 15, Epoch: 223, Loss: 0.0004\nIter: 16, Epoch: 223, Loss: 0.0005\nIter: 17, Epoch: 223, Loss: 0.0005\nIter: 18, Epoch: 223, Loss: 0.0005\nIter: 19, Epoch: 223, Loss: 0.0004\nIter: 20, Epoch: 223, Loss: 0.0005\nIter: 21, Epoch: 223, Loss: 0.0004\nIter: 22, Epoch: 223, Loss: 0.0009\nIter: 23, Epoch: 223, Loss: 0.0007\nIter: 24, Epoch: 223, Loss: 0.0003\nIter: 25, Epoch: 223, Loss: 0.0003\nIter: 26, Epoch: 223, Loss: 0.0006\nIter: 27, Epoch: 223, Loss: 0.0003\nIter: 28, Epoch: 223, Loss: 0.0006\nIter: 29, Epoch: 223, Loss: 0.0008\nIter: 30, Epoch: 223, Loss: 0.0008\n","truncated":false}} +%--- +%[output:26c4e02c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 223 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:9c55235f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 224, Loss: 0.0006\nIter: 2, Epoch: 224, Loss: 0.0005\nIter: 3, Epoch: 224, Loss: 0.0005\nIter: 4, Epoch: 224, Loss: 0.0004\nIter: 5, Epoch: 224, Loss: 0.0004\nIter: 6, Epoch: 224, Loss: 0.0004\nIter: 7, Epoch: 224, Loss: 0.0004\nIter: 8, Epoch: 224, Loss: 0.0010\nIter: 9, Epoch: 224, Loss: 0.0005\nIter: 10, Epoch: 224, Loss: 0.0007\nIter: 11, Epoch: 224, Loss: 0.0003\nIter: 12, Epoch: 224, Loss: 0.0003\nIter: 13, Epoch: 224, Loss: 0.0006\nIter: 14, Epoch: 224, Loss: 0.0007\nIter: 15, Epoch: 224, Loss: 0.0004\nIter: 16, Epoch: 224, Loss: 0.0005\nIter: 17, Epoch: 224, Loss: 0.0005\nIter: 18, Epoch: 224, Loss: 0.0005\nIter: 19, Epoch: 224, Loss: 0.0004\nIter: 20, Epoch: 224, Loss: 0.0005\nIter: 21, Epoch: 224, Loss: 0.0004\nIter: 22, Epoch: 224, Loss: 0.0009\nIter: 23, Epoch: 224, Loss: 0.0007\nIter: 24, Epoch: 224, Loss: 0.0003\nIter: 25, Epoch: 224, Loss: 0.0003\nIter: 26, Epoch: 224, Loss: 0.0006\nIter: 27, Epoch: 224, Loss: 0.0003\nIter: 28, Epoch: 224, Loss: 0.0006\nIter: 29, Epoch: 224, Loss: 0.0008\nIter: 30, Epoch: 224, Loss: 0.0008\n","truncated":false}} +%--- +%[output:90c35afe] +% data: {"dataType":"text","outputData":{"text":">> Epoch 224 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:88f6cb63] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 225, Loss: 0.0006\nIter: 2, Epoch: 225, Loss: 0.0005\nIter: 3, Epoch: 225, Loss: 0.0005\nIter: 4, Epoch: 225, Loss: 0.0004\nIter: 5, Epoch: 225, Loss: 0.0004\nIter: 6, Epoch: 225, Loss: 0.0004\nIter: 7, Epoch: 225, Loss: 0.0004\nIter: 8, Epoch: 225, Loss: 0.0010\nIter: 9, Epoch: 225, Loss: 0.0005\nIter: 10, Epoch: 225, Loss: 0.0007\nIter: 11, Epoch: 225, Loss: 0.0003\nIter: 12, Epoch: 225, Loss: 0.0003\nIter: 13, Epoch: 225, Loss: 0.0006\nIter: 14, Epoch: 225, Loss: 0.0007\nIter: 15, Epoch: 225, Loss: 0.0004\nIter: 16, Epoch: 225, Loss: 0.0005\nIter: 17, Epoch: 225, Loss: 0.0005\nIter: 18, Epoch: 225, Loss: 0.0005\nIter: 19, Epoch: 225, Loss: 0.0004\nIter: 20, Epoch: 225, Loss: 0.0005\nIter: 21, Epoch: 225, Loss: 0.0004\nIter: 22, Epoch: 225, Loss: 0.0009\nIter: 23, Epoch: 225, Loss: 0.0007\nIter: 24, Epoch: 225, Loss: 0.0003\nIter: 25, Epoch: 225, Loss: 0.0004\nIter: 26, Epoch: 225, Loss: 0.0006\nIter: 27, Epoch: 225, Loss: 0.0003\nIter: 28, Epoch: 225, Loss: 0.0005\nIter: 29, Epoch: 225, Loss: 0.0008\nIter: 30, Epoch: 225, Loss: 0.0008\n","truncated":false}} +%--- +%[output:5d1ba28c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 225 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:24dce2b7] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 226, Loss: 0.0005\nIter: 2, Epoch: 226, Loss: 0.0005\nIter: 3, Epoch: 226, Loss: 0.0005\nIter: 4, Epoch: 226, Loss: 0.0004\nIter: 5, Epoch: 226, Loss: 0.0004\nIter: 6, Epoch: 226, Loss: 0.0004\nIter: 7, Epoch: 226, Loss: 0.0004\nIter: 8, Epoch: 226, Loss: 0.0010\nIter: 9, Epoch: 226, Loss: 0.0005\nIter: 10, Epoch: 226, Loss: 0.0007\nIter: 11, Epoch: 226, Loss: 0.0003\nIter: 12, Epoch: 226, Loss: 0.0003\nIter: 13, Epoch: 226, Loss: 0.0006\nIter: 14, Epoch: 226, Loss: 0.0007\nIter: 15, Epoch: 226, Loss: 0.0004\nIter: 16, Epoch: 226, Loss: 0.0005\nIter: 17, Epoch: 226, Loss: 0.0005\nIter: 18, Epoch: 226, Loss: 0.0005\nIter: 19, Epoch: 226, Loss: 0.0004\nIter: 20, Epoch: 226, Loss: 0.0005\nIter: 21, Epoch: 226, Loss: 0.0004\nIter: 22, Epoch: 226, Loss: 0.0009\nIter: 23, Epoch: 226, Loss: 0.0007\nIter: 24, Epoch: 226, Loss: 0.0003\nIter: 25, Epoch: 226, Loss: 0.0003\nIter: 26, Epoch: 226, Loss: 0.0006\nIter: 27, Epoch: 226, Loss: 0.0003\nIter: 28, Epoch: 226, Loss: 0.0005\nIter: 29, Epoch: 226, Loss: 0.0008\nIter: 30, Epoch: 226, Loss: 0.0008\n","truncated":false}} +%--- +%[output:52a4ee15] +% data: {"dataType":"text","outputData":{"text":">> Epoch 226 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:5a4a9c5f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 227, Loss: 0.0005\nIter: 2, Epoch: 227, Loss: 0.0005\nIter: 3, Epoch: 227, Loss: 0.0005\nIter: 4, Epoch: 227, Loss: 0.0004\nIter: 5, Epoch: 227, Loss: 0.0004\nIter: 6, Epoch: 227, Loss: 0.0004\nIter: 7, Epoch: 227, Loss: 0.0004\nIter: 8, Epoch: 227, Loss: 0.0010\nIter: 9, Epoch: 227, Loss: 0.0005\nIter: 10, Epoch: 227, Loss: 0.0007\nIter: 11, Epoch: 227, Loss: 0.0003\nIter: 12, Epoch: 227, Loss: 0.0003\nIter: 13, Epoch: 227, Loss: 0.0006\nIter: 14, Epoch: 227, Loss: 0.0007\nIter: 15, Epoch: 227, Loss: 0.0004\nIter: 16, Epoch: 227, Loss: 0.0005\nIter: 17, Epoch: 227, Loss: 0.0005\nIter: 18, Epoch: 227, Loss: 0.0005\nIter: 19, Epoch: 227, Loss: 0.0004\nIter: 20, Epoch: 227, Loss: 0.0005\nIter: 21, Epoch: 227, Loss: 0.0004\nIter: 22, Epoch: 227, Loss: 0.0009\nIter: 23, Epoch: 227, Loss: 0.0007\nIter: 24, Epoch: 227, Loss: 0.0003\nIter: 25, Epoch: 227, Loss: 0.0003\nIter: 26, Epoch: 227, Loss: 0.0006\nIter: 27, Epoch: 227, Loss: 0.0003\nIter: 28, Epoch: 227, Loss: 0.0005\nIter: 29, Epoch: 227, Loss: 0.0008\nIter: 30, Epoch: 227, Loss: 0.0008\n","truncated":false}} +%--- +%[output:25dea215] +% data: {"dataType":"text","outputData":{"text":">> Epoch 227 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:886aea0b] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 228, Loss: 0.0005\nIter: 2, Epoch: 228, Loss: 0.0005\nIter: 3, Epoch: 228, Loss: 0.0005\nIter: 4, Epoch: 228, Loss: 0.0004\nIter: 5, Epoch: 228, Loss: 0.0004\nIter: 6, Epoch: 228, Loss: 0.0004\nIter: 7, Epoch: 228, Loss: 0.0004\nIter: 8, Epoch: 228, Loss: 0.0011\nIter: 9, Epoch: 228, Loss: 0.0005\nIter: 10, Epoch: 228, Loss: 0.0007\nIter: 11, Epoch: 228, Loss: 0.0003\nIter: 12, Epoch: 228, Loss: 0.0003\nIter: 13, Epoch: 228, Loss: 0.0006\nIter: 14, Epoch: 228, Loss: 0.0007\nIter: 15, Epoch: 228, Loss: 0.0004\nIter: 16, Epoch: 228, Loss: 0.0005\nIter: 17, Epoch: 228, Loss: 0.0005\nIter: 18, Epoch: 228, Loss: 0.0005\nIter: 19, Epoch: 228, Loss: 0.0004\nIter: 20, Epoch: 228, Loss: 0.0005\nIter: 21, Epoch: 228, Loss: 0.0004\nIter: 22, Epoch: 228, Loss: 0.0009\nIter: 23, Epoch: 228, Loss: 0.0007\nIter: 24, Epoch: 228, Loss: 0.0003\nIter: 25, Epoch: 228, Loss: 0.0003\nIter: 26, Epoch: 228, Loss: 0.0006\nIter: 27, Epoch: 228, Loss: 0.0004\nIter: 28, Epoch: 228, Loss: 0.0005\nIter: 29, Epoch: 228, Loss: 0.0007\nIter: 30, Epoch: 228, Loss: 0.0008\n","truncated":false}} +%--- +%[output:175ffb7e] +% data: {"dataType":"text","outputData":{"text":">> Epoch 228 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:4e5c6c95] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 229, Loss: 0.0005\nIter: 2, Epoch: 229, Loss: 0.0005\nIter: 3, Epoch: 229, Loss: 0.0005\nIter: 4, Epoch: 229, Loss: 0.0004\nIter: 5, Epoch: 229, Loss: 0.0004\nIter: 6, Epoch: 229, Loss: 0.0004\nIter: 7, Epoch: 229, Loss: 0.0004\nIter: 8, Epoch: 229, Loss: 0.0011\nIter: 9, Epoch: 229, Loss: 0.0005\nIter: 10, Epoch: 229, Loss: 0.0007\nIter: 11, Epoch: 229, Loss: 0.0003\nIter: 12, Epoch: 229, Loss: 0.0002\nIter: 13, Epoch: 229, Loss: 0.0006\nIter: 14, Epoch: 229, Loss: 0.0007\nIter: 15, Epoch: 229, Loss: 0.0004\nIter: 16, Epoch: 229, Loss: 0.0005\nIter: 17, Epoch: 229, Loss: 0.0005\nIter: 18, Epoch: 229, Loss: 0.0005\nIter: 19, Epoch: 229, Loss: 0.0004\nIter: 20, Epoch: 229, Loss: 0.0005\nIter: 21, Epoch: 229, Loss: 0.0004\nIter: 22, Epoch: 229, Loss: 0.0009\nIter: 23, Epoch: 229, Loss: 0.0007\nIter: 24, Epoch: 229, Loss: 0.0003\nIter: 25, Epoch: 229, Loss: 0.0003\nIter: 26, Epoch: 229, Loss: 0.0006\nIter: 27, Epoch: 229, Loss: 0.0004\nIter: 28, Epoch: 229, Loss: 0.0005\nIter: 29, Epoch: 229, Loss: 0.0007\nIter: 30, Epoch: 229, Loss: 0.0008\n","truncated":false}} +%--- +%[output:7673353c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 229 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:2bc02e69] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 230, Loss: 0.0005\nIter: 2, Epoch: 230, Loss: 0.0005\nIter: 3, Epoch: 230, Loss: 0.0004\nIter: 4, Epoch: 230, Loss: 0.0004\nIter: 5, Epoch: 230, Loss: 0.0004\nIter: 6, Epoch: 230, Loss: 0.0004\nIter: 7, Epoch: 230, Loss: 0.0004\nIter: 8, Epoch: 230, Loss: 0.0011\nIter: 9, Epoch: 230, Loss: 0.0005\nIter: 10, Epoch: 230, Loss: 0.0007\nIter: 11, Epoch: 230, Loss: 0.0003\nIter: 12, Epoch: 230, Loss: 0.0003\nIter: 13, Epoch: 230, Loss: 0.0006\nIter: 14, Epoch: 230, Loss: 0.0007\nIter: 15, Epoch: 230, Loss: 0.0004\nIter: 16, Epoch: 230, Loss: 0.0005\nIter: 17, Epoch: 230, Loss: 0.0005\nIter: 18, Epoch: 230, Loss: 0.0005\nIter: 19, Epoch: 230, Loss: 0.0004\nIter: 20, Epoch: 230, Loss: 0.0005\nIter: 21, Epoch: 230, Loss: 0.0004\nIter: 22, Epoch: 230, Loss: 0.0009\nIter: 23, Epoch: 230, Loss: 0.0007\nIter: 24, Epoch: 230, Loss: 0.0003\nIter: 25, Epoch: 230, Loss: 0.0003\nIter: 26, Epoch: 230, Loss: 0.0006\nIter: 27, Epoch: 230, Loss: 0.0004\nIter: 28, Epoch: 230, Loss: 0.0005\nIter: 29, Epoch: 230, Loss: 0.0007\nIter: 30, Epoch: 230, Loss: 0.0008\n","truncated":false}} +%--- +%[output:8511c6e7] +% data: {"dataType":"text","outputData":{"text":">> Epoch 230 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:69baabec] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 231, Loss: 0.0005\nIter: 2, Epoch: 231, Loss: 0.0005\nIter: 3, Epoch: 231, Loss: 0.0004\nIter: 4, Epoch: 231, Loss: 0.0004\nIter: 5, Epoch: 231, Loss: 0.0004\nIter: 6, Epoch: 231, Loss: 0.0004\nIter: 7, Epoch: 231, Loss: 0.0004\nIter: 8, Epoch: 231, Loss: 0.0011\nIter: 9, Epoch: 231, Loss: 0.0005\nIter: 10, Epoch: 231, Loss: 0.0007\nIter: 11, Epoch: 231, Loss: 0.0003\nIter: 12, Epoch: 231, Loss: 0.0003\nIter: 13, Epoch: 231, Loss: 0.0006\nIter: 14, Epoch: 231, Loss: 0.0006\nIter: 15, Epoch: 231, Loss: 0.0004\nIter: 16, Epoch: 231, Loss: 0.0005\nIter: 17, Epoch: 231, Loss: 0.0005\nIter: 18, Epoch: 231, Loss: 0.0005\nIter: 19, Epoch: 231, Loss: 0.0004\nIter: 20, Epoch: 231, Loss: 0.0005\nIter: 21, Epoch: 231, Loss: 0.0004\nIter: 22, Epoch: 231, Loss: 0.0009\nIter: 23, Epoch: 231, Loss: 0.0007\nIter: 24, Epoch: 231, Loss: 0.0003\nIter: 25, Epoch: 231, Loss: 0.0003\nIter: 26, Epoch: 231, Loss: 0.0006\nIter: 27, Epoch: 231, Loss: 0.0004\nIter: 28, Epoch: 231, Loss: 0.0005\nIter: 29, Epoch: 231, Loss: 0.0007\nIter: 30, Epoch: 231, Loss: 0.0008\n","truncated":false}} +%--- +%[output:8adf1ef2] +% data: {"dataType":"text","outputData":{"text":">> Epoch 231 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:0bfc7764] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 232, Loss: 0.0005\nIter: 2, Epoch: 232, Loss: 0.0005\nIter: 3, Epoch: 232, Loss: 0.0004\nIter: 4, Epoch: 232, Loss: 0.0004\nIter: 5, Epoch: 232, Loss: 0.0003\nIter: 6, Epoch: 232, Loss: 0.0004\nIter: 7, Epoch: 232, Loss: 0.0004\nIter: 8, Epoch: 232, Loss: 0.0011\nIter: 9, Epoch: 232, Loss: 0.0004\nIter: 10, Epoch: 232, Loss: 0.0007\nIter: 11, Epoch: 232, Loss: 0.0003\nIter: 12, Epoch: 232, Loss: 0.0003\nIter: 13, Epoch: 232, Loss: 0.0006\nIter: 14, Epoch: 232, Loss: 0.0006\nIter: 15, Epoch: 232, Loss: 0.0004\nIter: 16, Epoch: 232, Loss: 0.0005\nIter: 17, Epoch: 232, Loss: 0.0005\nIter: 18, Epoch: 232, Loss: 0.0005\nIter: 19, Epoch: 232, Loss: 0.0004\nIter: 20, Epoch: 232, Loss: 0.0005\nIter: 21, Epoch: 232, Loss: 0.0004\nIter: 22, Epoch: 232, Loss: 0.0009\nIter: 23, Epoch: 232, Loss: 0.0007\nIter: 24, Epoch: 232, Loss: 0.0003\nIter: 25, Epoch: 232, Loss: 0.0003\nIter: 26, Epoch: 232, Loss: 0.0006\nIter: 27, Epoch: 232, Loss: 0.0004\nIter: 28, Epoch: 232, Loss: 0.0005\nIter: 29, Epoch: 232, Loss: 0.0007\nIter: 30, Epoch: 232, Loss: 0.0007\n","truncated":false}} +%--- +%[output:8cf10a47] +% data: {"dataType":"text","outputData":{"text":">> Epoch 232 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:7e79f7b7] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 233, Loss: 0.0005\nIter: 2, Epoch: 233, Loss: 0.0005\nIter: 3, Epoch: 233, Loss: 0.0004\nIter: 4, Epoch: 233, Loss: 0.0004\nIter: 5, Epoch: 233, Loss: 0.0003\nIter: 6, Epoch: 233, Loss: 0.0004\nIter: 7, Epoch: 233, Loss: 0.0004\nIter: 8, Epoch: 233, Loss: 0.0011\nIter: 9, Epoch: 233, Loss: 0.0004\nIter: 10, Epoch: 233, Loss: 0.0008\nIter: 11, Epoch: 233, Loss: 0.0003\nIter: 12, Epoch: 233, Loss: 0.0003\nIter: 13, Epoch: 233, Loss: 0.0006\nIter: 14, Epoch: 233, Loss: 0.0006\nIter: 15, Epoch: 233, Loss: 0.0004\nIter: 16, Epoch: 233, Loss: 0.0005\nIter: 17, Epoch: 233, Loss: 0.0005\nIter: 18, Epoch: 233, Loss: 0.0005\nIter: 19, Epoch: 233, Loss: 0.0004\nIter: 20, Epoch: 233, Loss: 0.0005\nIter: 21, Epoch: 233, Loss: 0.0004\nIter: 22, Epoch: 233, Loss: 0.0009\nIter: 23, Epoch: 233, Loss: 0.0007\nIter: 24, Epoch: 233, Loss: 0.0003\nIter: 25, Epoch: 233, Loss: 0.0003\nIter: 26, Epoch: 233, Loss: 0.0006\nIter: 27, Epoch: 233, Loss: 0.0004\nIter: 28, Epoch: 233, Loss: 0.0005\nIter: 29, Epoch: 233, Loss: 0.0006\nIter: 30, Epoch: 233, Loss: 0.0007\n","truncated":false}} +%--- +%[output:473e54f9] +% data: {"dataType":"text","outputData":{"text":">> Epoch 233 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:2b46a3f3] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 234, Loss: 0.0005\nIter: 2, Epoch: 234, Loss: 0.0005\nIter: 3, Epoch: 234, Loss: 0.0004\nIter: 4, Epoch: 234, Loss: 0.0004\nIter: 5, Epoch: 234, Loss: 0.0003\nIter: 6, Epoch: 234, Loss: 0.0004\nIter: 7, Epoch: 234, Loss: 0.0004\nIter: 8, Epoch: 234, Loss: 0.0011\nIter: 9, Epoch: 234, Loss: 0.0004\nIter: 10, Epoch: 234, Loss: 0.0008\nIter: 11, Epoch: 234, Loss: 0.0003\nIter: 12, Epoch: 234, Loss: 0.0003\nIter: 13, Epoch: 234, Loss: 0.0006\nIter: 14, Epoch: 234, Loss: 0.0006\nIter: 15, Epoch: 234, Loss: 0.0004\nIter: 16, Epoch: 234, Loss: 0.0005\nIter: 17, Epoch: 234, Loss: 0.0005\nIter: 18, Epoch: 234, Loss: 0.0005\nIter: 19, Epoch: 234, Loss: 0.0004\nIter: 20, Epoch: 234, Loss: 0.0005\nIter: 21, Epoch: 234, Loss: 0.0004\nIter: 22, Epoch: 234, Loss: 0.0009\nIter: 23, Epoch: 234, Loss: 0.0007\nIter: 24, Epoch: 234, Loss: 0.0003\nIter: 25, Epoch: 234, Loss: 0.0003\nIter: 26, Epoch: 234, Loss: 0.0006\nIter: 27, Epoch: 234, Loss: 0.0004\nIter: 28, Epoch: 234, Loss: 0.0005\nIter: 29, Epoch: 234, Loss: 0.0006\nIter: 30, Epoch: 234, Loss: 0.0007\n","truncated":false}} +%--- +%[output:635b23f8] +% data: {"dataType":"text","outputData":{"text":">> Epoch 234 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:95f476b4] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 235, Loss: 0.0005\nIter: 2, Epoch: 235, Loss: 0.0005\nIter: 3, Epoch: 235, Loss: 0.0004\nIter: 4, Epoch: 235, Loss: 0.0004\nIter: 5, Epoch: 235, Loss: 0.0003\nIter: 6, Epoch: 235, Loss: 0.0004\nIter: 7, Epoch: 235, Loss: 0.0004\nIter: 8, Epoch: 235, Loss: 0.0011\nIter: 9, Epoch: 235, Loss: 0.0004\nIter: 10, Epoch: 235, Loss: 0.0008\nIter: 11, Epoch: 235, Loss: 0.0003\nIter: 12, Epoch: 235, Loss: 0.0003\nIter: 13, Epoch: 235, Loss: 0.0006\nIter: 14, Epoch: 235, Loss: 0.0006\nIter: 15, Epoch: 235, Loss: 0.0004\nIter: 16, Epoch: 235, Loss: 0.0005\nIter: 17, Epoch: 235, Loss: 0.0005\nIter: 18, Epoch: 235, Loss: 0.0005\nIter: 19, Epoch: 235, Loss: 0.0004\nIter: 20, Epoch: 235, Loss: 0.0005\nIter: 21, Epoch: 235, Loss: 0.0004\nIter: 22, Epoch: 235, Loss: 0.0009\nIter: 23, Epoch: 235, Loss: 0.0007\nIter: 24, Epoch: 235, Loss: 0.0003\nIter: 25, Epoch: 235, Loss: 0.0003\nIter: 26, Epoch: 235, Loss: 0.0006\nIter: 27, Epoch: 235, Loss: 0.0004\nIter: 28, Epoch: 235, Loss: 0.0005\nIter: 29, Epoch: 235, Loss: 0.0006\nIter: 30, Epoch: 235, Loss: 0.0007\n","truncated":false}} +%--- +%[output:4b904e95] +% data: {"dataType":"text","outputData":{"text":">> Epoch 235 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:62248664] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 236, Loss: 0.0005\nIter: 2, Epoch: 236, Loss: 0.0005\nIter: 3, Epoch: 236, Loss: 0.0004\nIter: 4, Epoch: 236, Loss: 0.0004\nIter: 5, Epoch: 236, Loss: 0.0003\nIter: 6, Epoch: 236, Loss: 0.0004\nIter: 7, Epoch: 236, Loss: 0.0004\nIter: 8, Epoch: 236, Loss: 0.0011\nIter: 9, Epoch: 236, Loss: 0.0004\nIter: 10, Epoch: 236, Loss: 0.0008\nIter: 11, Epoch: 236, Loss: 0.0004\nIter: 12, Epoch: 236, Loss: 0.0003\nIter: 13, Epoch: 236, Loss: 0.0005\nIter: 14, Epoch: 236, Loss: 0.0006\nIter: 15, Epoch: 236, Loss: 0.0004\nIter: 16, Epoch: 236, Loss: 0.0005\nIter: 17, Epoch: 236, Loss: 0.0005\nIter: 18, Epoch: 236, Loss: 0.0005\nIter: 19, Epoch: 236, Loss: 0.0004\nIter: 20, Epoch: 236, Loss: 0.0005\nIter: 21, Epoch: 236, Loss: 0.0004\nIter: 22, Epoch: 236, Loss: 0.0009\nIter: 23, Epoch: 236, Loss: 0.0007\nIter: 24, Epoch: 236, Loss: 0.0003\nIter: 25, Epoch: 236, Loss: 0.0003\nIter: 26, Epoch: 236, Loss: 0.0006\nIter: 27, Epoch: 236, Loss: 0.0004\nIter: 28, Epoch: 236, Loss: 0.0005\nIter: 29, Epoch: 236, Loss: 0.0006\nIter: 30, Epoch: 236, Loss: 0.0007\n","truncated":false}} +%--- +%[output:671adcfd] +% data: {"dataType":"text","outputData":{"text":">> Epoch 236 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:7665abf6] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 237, Loss: 0.0005\nIter: 2, Epoch: 237, Loss: 0.0005\nIter: 3, Epoch: 237, Loss: 0.0004\nIter: 4, Epoch: 237, Loss: 0.0004\nIter: 5, Epoch: 237, Loss: 0.0003\nIter: 6, Epoch: 237, Loss: 0.0003\nIter: 7, Epoch: 237, Loss: 0.0004\nIter: 8, Epoch: 237, Loss: 0.0011\nIter: 9, Epoch: 237, Loss: 0.0004\nIter: 10, Epoch: 237, Loss: 0.0008\nIter: 11, Epoch: 237, Loss: 0.0004\nIter: 12, Epoch: 237, Loss: 0.0003\nIter: 13, Epoch: 237, Loss: 0.0005\nIter: 14, Epoch: 237, Loss: 0.0006\nIter: 15, Epoch: 237, Loss: 0.0004\nIter: 16, Epoch: 237, Loss: 0.0005\nIter: 17, Epoch: 237, Loss: 0.0005\nIter: 18, Epoch: 237, Loss: 0.0005\nIter: 19, Epoch: 237, Loss: 0.0004\nIter: 20, Epoch: 237, Loss: 0.0005\nIter: 21, Epoch: 237, Loss: 0.0004\nIter: 22, Epoch: 237, Loss: 0.0009\nIter: 23, Epoch: 237, Loss: 0.0007\nIter: 24, Epoch: 237, Loss: 0.0003\nIter: 25, Epoch: 237, Loss: 0.0003\nIter: 26, Epoch: 237, Loss: 0.0006\nIter: 27, Epoch: 237, Loss: 0.0004\nIter: 28, Epoch: 237, Loss: 0.0005\nIter: 29, Epoch: 237, Loss: 0.0005\nIter: 30, Epoch: 237, Loss: 0.0006\n","truncated":false}} +%--- +%[output:1d7006cf] +% data: {"dataType":"text","outputData":{"text":">> Epoch 237 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:77fa26fc] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 238, Loss: 0.0005\nIter: 2, Epoch: 238, Loss: 0.0005\nIter: 3, Epoch: 238, Loss: 0.0004\nIter: 4, Epoch: 238, Loss: 0.0004\nIter: 5, Epoch: 238, Loss: 0.0003\nIter: 6, Epoch: 238, Loss: 0.0003\nIter: 7, Epoch: 238, Loss: 0.0004\nIter: 8, Epoch: 238, Loss: 0.0011\nIter: 9, Epoch: 238, Loss: 0.0004\nIter: 10, Epoch: 238, Loss: 0.0008\nIter: 11, Epoch: 238, Loss: 0.0004\nIter: 12, Epoch: 238, Loss: 0.0003\nIter: 13, Epoch: 238, Loss: 0.0005\nIter: 14, Epoch: 238, Loss: 0.0006\nIter: 15, Epoch: 238, Loss: 0.0004\nIter: 16, Epoch: 238, Loss: 0.0005\nIter: 17, Epoch: 238, Loss: 0.0005\nIter: 18, Epoch: 238, Loss: 0.0005\nIter: 19, Epoch: 238, Loss: 0.0004\nIter: 20, Epoch: 238, Loss: 0.0005\nIter: 21, Epoch: 238, Loss: 0.0003\nIter: 22, Epoch: 238, Loss: 0.0009\nIter: 23, Epoch: 238, Loss: 0.0007\nIter: 24, Epoch: 238, Loss: 0.0003\nIter: 25, Epoch: 238, Loss: 0.0003\nIter: 26, Epoch: 238, Loss: 0.0006\nIter: 27, Epoch: 238, Loss: 0.0004\nIter: 28, Epoch: 238, Loss: 0.0005\nIter: 29, Epoch: 238, Loss: 0.0005\nIter: 30, Epoch: 238, Loss: 0.0006\n","truncated":false}} +%--- +%[output:4f069e1b] +% data: {"dataType":"text","outputData":{"text":">> Epoch 238 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:1704c9b1] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 239, Loss: 0.0005\nIter: 2, Epoch: 239, Loss: 0.0005\nIter: 3, Epoch: 239, Loss: 0.0004\nIter: 4, Epoch: 239, Loss: 0.0004\nIter: 5, Epoch: 239, Loss: 0.0004\nIter: 6, Epoch: 239, Loss: 0.0003\nIter: 7, Epoch: 239, Loss: 0.0004\nIter: 8, Epoch: 239, Loss: 0.0010\nIter: 9, Epoch: 239, Loss: 0.0004\nIter: 10, Epoch: 239, Loss: 0.0008\nIter: 11, Epoch: 239, Loss: 0.0004\nIter: 12, Epoch: 239, Loss: 0.0003\nIter: 13, Epoch: 239, Loss: 0.0005\nIter: 14, Epoch: 239, Loss: 0.0006\nIter: 15, Epoch: 239, Loss: 0.0004\nIter: 16, Epoch: 239, Loss: 0.0005\nIter: 17, Epoch: 239, Loss: 0.0004\nIter: 18, Epoch: 239, Loss: 0.0005\nIter: 19, Epoch: 239, Loss: 0.0004\nIter: 20, Epoch: 239, Loss: 0.0005\nIter: 21, Epoch: 239, Loss: 0.0003\nIter: 22, Epoch: 239, Loss: 0.0009\nIter: 23, Epoch: 239, Loss: 0.0007\nIter: 24, Epoch: 239, Loss: 0.0003\nIter: 25, Epoch: 239, Loss: 0.0003\nIter: 26, Epoch: 239, Loss: 0.0006\nIter: 27, Epoch: 239, Loss: 0.0004\nIter: 28, Epoch: 239, Loss: 0.0005\nIter: 29, Epoch: 239, Loss: 0.0005\nIter: 30, Epoch: 239, Loss: 0.0006\n","truncated":false}} +%--- +%[output:8925769b] +% data: {"dataType":"text","outputData":{"text":">> Epoch 239 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:59928606] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 240, Loss: 0.0006\nIter: 2, Epoch: 240, Loss: 0.0005\nIter: 3, Epoch: 240, Loss: 0.0004\nIter: 4, Epoch: 240, Loss: 0.0004\nIter: 5, Epoch: 240, Loss: 0.0004\nIter: 6, Epoch: 240, Loss: 0.0003\nIter: 7, Epoch: 240, Loss: 0.0004\nIter: 8, Epoch: 240, Loss: 0.0010\nIter: 9, Epoch: 240, Loss: 0.0004\nIter: 10, Epoch: 240, Loss: 0.0008\nIter: 11, Epoch: 240, Loss: 0.0004\nIter: 12, Epoch: 240, Loss: 0.0003\nIter: 13, Epoch: 240, Loss: 0.0005\nIter: 14, Epoch: 240, Loss: 0.0006\nIter: 15, Epoch: 240, Loss: 0.0004\nIter: 16, Epoch: 240, Loss: 0.0005\nIter: 17, Epoch: 240, Loss: 0.0004\nIter: 18, Epoch: 240, Loss: 0.0005\nIter: 19, Epoch: 240, Loss: 0.0004\nIter: 20, Epoch: 240, Loss: 0.0005\nIter: 21, Epoch: 240, Loss: 0.0003\nIter: 22, Epoch: 240, Loss: 0.0008\nIter: 23, Epoch: 240, Loss: 0.0007\nIter: 24, Epoch: 240, Loss: 0.0003\nIter: 25, Epoch: 240, Loss: 0.0003\nIter: 26, Epoch: 240, Loss: 0.0006\nIter: 27, Epoch: 240, Loss: 0.0004\nIter: 28, Epoch: 240, Loss: 0.0005\nIter: 29, Epoch: 240, Loss: 0.0005\nIter: 30, Epoch: 240, Loss: 0.0006\n","truncated":false}} +%--- +%[output:4b12787c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 240 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:9554411f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 241, Loss: 0.0006\nIter: 2, Epoch: 241, Loss: 0.0005\nIter: 3, Epoch: 241, Loss: 0.0004\nIter: 4, Epoch: 241, Loss: 0.0004\nIter: 5, Epoch: 241, Loss: 0.0004\nIter: 6, Epoch: 241, Loss: 0.0003\nIter: 7, Epoch: 241, Loss: 0.0004\nIter: 8, Epoch: 241, Loss: 0.0010\nIter: 9, Epoch: 241, Loss: 0.0004\nIter: 10, Epoch: 241, Loss: 0.0008\nIter: 11, Epoch: 241, Loss: 0.0003\nIter: 12, Epoch: 241, Loss: 0.0003\nIter: 13, Epoch: 241, Loss: 0.0005\nIter: 14, Epoch: 241, Loss: 0.0006\nIter: 15, Epoch: 241, Loss: 0.0004\nIter: 16, Epoch: 241, Loss: 0.0005\nIter: 17, Epoch: 241, Loss: 0.0004\nIter: 18, Epoch: 241, Loss: 0.0005\nIter: 19, Epoch: 241, Loss: 0.0004\nIter: 20, Epoch: 241, Loss: 0.0005\nIter: 21, Epoch: 241, Loss: 0.0003\nIter: 22, Epoch: 241, Loss: 0.0008\nIter: 23, Epoch: 241, Loss: 0.0007\nIter: 24, Epoch: 241, Loss: 0.0003\nIter: 25, Epoch: 241, Loss: 0.0003\nIter: 26, Epoch: 241, Loss: 0.0006\nIter: 27, Epoch: 241, Loss: 0.0004\nIter: 28, Epoch: 241, Loss: 0.0005\nIter: 29, Epoch: 241, Loss: 0.0005\nIter: 30, Epoch: 241, Loss: 0.0006\n","truncated":false}} +%--- +%[output:74bdb1f6] +% data: {"dataType":"text","outputData":{"text":">> Epoch 241 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:64e41fcb] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 242, Loss: 0.0006\nIter: 2, Epoch: 242, Loss: 0.0005\nIter: 3, Epoch: 242, Loss: 0.0004\nIter: 4, Epoch: 242, Loss: 0.0004\nIter: 5, Epoch: 242, Loss: 0.0004\nIter: 6, Epoch: 242, Loss: 0.0003\nIter: 7, Epoch: 242, Loss: 0.0004\nIter: 8, Epoch: 242, Loss: 0.0010\nIter: 9, Epoch: 242, Loss: 0.0004\nIter: 10, Epoch: 242, Loss: 0.0008\nIter: 11, Epoch: 242, Loss: 0.0003\nIter: 12, Epoch: 242, Loss: 0.0003\nIter: 13, Epoch: 242, Loss: 0.0005\nIter: 14, Epoch: 242, Loss: 0.0006\nIter: 15, Epoch: 242, Loss: 0.0004\nIter: 16, Epoch: 242, Loss: 0.0005\nIter: 17, Epoch: 242, Loss: 0.0004\nIter: 18, Epoch: 242, Loss: 0.0005\nIter: 19, Epoch: 242, Loss: 0.0004\nIter: 20, Epoch: 242, Loss: 0.0005\nIter: 21, Epoch: 242, Loss: 0.0003\nIter: 22, Epoch: 242, Loss: 0.0008\nIter: 23, Epoch: 242, Loss: 0.0007\nIter: 24, Epoch: 242, Loss: 0.0002\nIter: 25, Epoch: 242, Loss: 0.0003\nIter: 26, Epoch: 242, Loss: 0.0006\nIter: 27, Epoch: 242, Loss: 0.0004\nIter: 28, Epoch: 242, Loss: 0.0005\nIter: 29, Epoch: 242, Loss: 0.0005\nIter: 30, Epoch: 242, Loss: 0.0006\n","truncated":false}} +%--- +%[output:1db4dd50] +% data: {"dataType":"text","outputData":{"text":">> Epoch 242 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:1668efdb] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 243, Loss: 0.0006\nIter: 2, Epoch: 243, Loss: 0.0005\nIter: 3, Epoch: 243, Loss: 0.0004\nIter: 4, Epoch: 243, Loss: 0.0004\nIter: 5, Epoch: 243, Loss: 0.0004\nIter: 6, Epoch: 243, Loss: 0.0003\nIter: 7, Epoch: 243, Loss: 0.0004\nIter: 8, Epoch: 243, Loss: 0.0009\nIter: 9, Epoch: 243, Loss: 0.0004\nIter: 10, Epoch: 243, Loss: 0.0007\nIter: 11, Epoch: 243, Loss: 0.0003\nIter: 12, Epoch: 243, Loss: 0.0003\nIter: 13, Epoch: 243, Loss: 0.0005\nIter: 14, Epoch: 243, Loss: 0.0006\nIter: 15, Epoch: 243, Loss: 0.0004\nIter: 16, Epoch: 243, Loss: 0.0005\nIter: 17, Epoch: 243, Loss: 0.0004\nIter: 18, Epoch: 243, Loss: 0.0005\nIter: 19, Epoch: 243, Loss: 0.0004\nIter: 20, Epoch: 243, Loss: 0.0005\nIter: 21, Epoch: 243, Loss: 0.0003\nIter: 22, Epoch: 243, Loss: 0.0008\nIter: 23, Epoch: 243, Loss: 0.0007\nIter: 24, Epoch: 243, Loss: 0.0002\nIter: 25, Epoch: 243, Loss: 0.0003\nIter: 26, Epoch: 243, Loss: 0.0006\nIter: 27, Epoch: 243, Loss: 0.0004\nIter: 28, Epoch: 243, Loss: 0.0005\nIter: 29, Epoch: 243, Loss: 0.0005\nIter: 30, Epoch: 243, Loss: 0.0006\n","truncated":false}} +%--- +%[output:1c410eed] +% data: {"dataType":"text","outputData":{"text":">> Epoch 243 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:338f8083] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 244, Loss: 0.0006\nIter: 2, Epoch: 244, Loss: 0.0005\nIter: 3, Epoch: 244, Loss: 0.0004\nIter: 4, Epoch: 244, Loss: 0.0004\nIter: 5, Epoch: 244, Loss: 0.0004\nIter: 6, Epoch: 244, Loss: 0.0003\nIter: 7, Epoch: 244, Loss: 0.0004\nIter: 8, Epoch: 244, Loss: 0.0009\nIter: 9, Epoch: 244, Loss: 0.0004\nIter: 10, Epoch: 244, Loss: 0.0007\nIter: 11, Epoch: 244, Loss: 0.0003\nIter: 12, Epoch: 244, Loss: 0.0003\nIter: 13, Epoch: 244, Loss: 0.0005\nIter: 14, Epoch: 244, Loss: 0.0006\nIter: 15, Epoch: 244, Loss: 0.0004\nIter: 16, Epoch: 244, Loss: 0.0005\nIter: 17, Epoch: 244, Loss: 0.0004\nIter: 18, Epoch: 244, Loss: 0.0005\nIter: 19, Epoch: 244, Loss: 0.0004\nIter: 20, Epoch: 244, Loss: 0.0005\nIter: 21, Epoch: 244, Loss: 0.0003\nIter: 22, Epoch: 244, Loss: 0.0008\nIter: 23, Epoch: 244, Loss: 0.0007\nIter: 24, Epoch: 244, Loss: 0.0002\nIter: 25, Epoch: 244, Loss: 0.0003\nIter: 26, Epoch: 244, Loss: 0.0005\nIter: 27, Epoch: 244, Loss: 0.0004\nIter: 28, Epoch: 244, Loss: 0.0005\nIter: 29, Epoch: 244, Loss: 0.0005\nIter: 30, Epoch: 244, Loss: 0.0006\n","truncated":false}} +%--- +%[output:6b5c915e] +% data: {"dataType":"text","outputData":{"text":">> Epoch 244 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:8b4d0d9a] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 245, Loss: 0.0006\nIter: 2, Epoch: 245, Loss: 0.0005\nIter: 3, Epoch: 245, Loss: 0.0004\nIter: 4, Epoch: 245, Loss: 0.0004\nIter: 5, Epoch: 245, Loss: 0.0004\nIter: 6, Epoch: 245, Loss: 0.0003\nIter: 7, Epoch: 245, Loss: 0.0004\nIter: 8, Epoch: 245, Loss: 0.0009\nIter: 9, Epoch: 245, Loss: 0.0004\nIter: 10, Epoch: 245, Loss: 0.0007\nIter: 11, Epoch: 245, Loss: 0.0003\nIter: 12, Epoch: 245, Loss: 0.0003\nIter: 13, Epoch: 245, Loss: 0.0005\nIter: 14, Epoch: 245, Loss: 0.0006\nIter: 15, Epoch: 245, Loss: 0.0004\nIter: 16, Epoch: 245, Loss: 0.0005\nIter: 17, Epoch: 245, Loss: 0.0004\nIter: 18, Epoch: 245, Loss: 0.0005\nIter: 19, Epoch: 245, Loss: 0.0004\nIter: 20, Epoch: 245, Loss: 0.0005\nIter: 21, Epoch: 245, Loss: 0.0003\nIter: 22, Epoch: 245, Loss: 0.0008\nIter: 23, Epoch: 245, Loss: 0.0007\nIter: 24, Epoch: 245, Loss: 0.0002\nIter: 25, Epoch: 245, Loss: 0.0003\nIter: 26, Epoch: 245, Loss: 0.0005\nIter: 27, Epoch: 245, Loss: 0.0004\nIter: 28, Epoch: 245, Loss: 0.0005\nIter: 29, Epoch: 245, Loss: 0.0005\nIter: 30, Epoch: 245, Loss: 0.0006\n","truncated":false}} +%--- +%[output:74ca347c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 245 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:4b6cc66d] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 246, Loss: 0.0006\nIter: 2, Epoch: 246, Loss: 0.0005\nIter: 3, Epoch: 246, Loss: 0.0004\nIter: 4, Epoch: 246, Loss: 0.0004\nIter: 5, Epoch: 246, Loss: 0.0004\nIter: 6, Epoch: 246, Loss: 0.0003\nIter: 7, Epoch: 246, Loss: 0.0004\nIter: 8, Epoch: 246, Loss: 0.0009\nIter: 9, Epoch: 246, Loss: 0.0004\nIter: 10, Epoch: 246, Loss: 0.0007\nIter: 11, Epoch: 246, Loss: 0.0003\nIter: 12, Epoch: 246, Loss: 0.0003\nIter: 13, Epoch: 246, Loss: 0.0005\nIter: 14, Epoch: 246, Loss: 0.0006\nIter: 15, Epoch: 246, Loss: 0.0004\nIter: 16, Epoch: 246, Loss: 0.0005\nIter: 17, Epoch: 246, Loss: 0.0004\nIter: 18, Epoch: 246, Loss: 0.0005\nIter: 19, Epoch: 246, Loss: 0.0004\nIter: 20, Epoch: 246, Loss: 0.0005\nIter: 21, Epoch: 246, Loss: 0.0003\nIter: 22, Epoch: 246, Loss: 0.0008\nIter: 23, Epoch: 246, Loss: 0.0007\nIter: 24, Epoch: 246, Loss: 0.0002\nIter: 25, Epoch: 246, Loss: 0.0003\nIter: 26, Epoch: 246, Loss: 0.0005\nIter: 27, Epoch: 246, Loss: 0.0004\nIter: 28, Epoch: 246, Loss: 0.0005\nIter: 29, Epoch: 246, Loss: 0.0005\nIter: 30, Epoch: 246, Loss: 0.0005\n","truncated":false}} +%--- +%[output:42201a8c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 246 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:068739bf] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 247, Loss: 0.0006\nIter: 2, Epoch: 247, Loss: 0.0005\nIter: 3, Epoch: 247, Loss: 0.0004\nIter: 4, Epoch: 247, Loss: 0.0004\nIter: 5, Epoch: 247, Loss: 0.0004\nIter: 6, Epoch: 247, Loss: 0.0003\nIter: 7, Epoch: 247, Loss: 0.0004\nIter: 8, Epoch: 247, Loss: 0.0009\nIter: 9, Epoch: 247, Loss: 0.0004\nIter: 10, Epoch: 247, Loss: 0.0007\nIter: 11, Epoch: 247, Loss: 0.0003\nIter: 12, Epoch: 247, Loss: 0.0003\nIter: 13, Epoch: 247, Loss: 0.0005\nIter: 14, Epoch: 247, Loss: 0.0006\nIter: 15, Epoch: 247, Loss: 0.0004\nIter: 16, Epoch: 247, Loss: 0.0005\nIter: 17, Epoch: 247, Loss: 0.0004\nIter: 18, Epoch: 247, Loss: 0.0005\nIter: 19, Epoch: 247, Loss: 0.0004\nIter: 20, Epoch: 247, Loss: 0.0005\nIter: 21, Epoch: 247, Loss: 0.0003\nIter: 22, Epoch: 247, Loss: 0.0008\nIter: 23, Epoch: 247, Loss: 0.0007\nIter: 24, Epoch: 247, Loss: 0.0002\nIter: 25, Epoch: 247, Loss: 0.0003\nIter: 26, Epoch: 247, Loss: 0.0005\nIter: 27, Epoch: 247, Loss: 0.0004\nIter: 28, Epoch: 247, Loss: 0.0005\nIter: 29, Epoch: 247, Loss: 0.0005\nIter: 30, Epoch: 247, Loss: 0.0005\n","truncated":false}} +%--- +%[output:290dbb85] +% data: {"dataType":"text","outputData":{"text":">> Epoch 247 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:7fd9b34c] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 248, Loss: 0.0006\nIter: 2, Epoch: 248, Loss: 0.0004\nIter: 3, Epoch: 248, Loss: 0.0004\nIter: 4, Epoch: 248, Loss: 0.0004\nIter: 5, Epoch: 248, Loss: 0.0004\nIter: 6, Epoch: 248, Loss: 0.0003\nIter: 7, Epoch: 248, Loss: 0.0004\nIter: 8, Epoch: 248, Loss: 0.0009\nIter: 9, Epoch: 248, Loss: 0.0004\nIter: 10, Epoch: 248, Loss: 0.0007\nIter: 11, Epoch: 248, Loss: 0.0003\nIter: 12, Epoch: 248, Loss: 0.0003\nIter: 13, Epoch: 248, Loss: 0.0005\nIter: 14, Epoch: 248, Loss: 0.0006\nIter: 15, Epoch: 248, Loss: 0.0004\nIter: 16, Epoch: 248, Loss: 0.0005\nIter: 17, Epoch: 248, Loss: 0.0004\nIter: 18, Epoch: 248, Loss: 0.0005\nIter: 19, Epoch: 248, Loss: 0.0004\nIter: 20, Epoch: 248, Loss: 0.0005\nIter: 21, Epoch: 248, Loss: 0.0003\nIter: 22, Epoch: 248, Loss: 0.0008\nIter: 23, Epoch: 248, Loss: 0.0007\nIter: 24, Epoch: 248, Loss: 0.0002\nIter: 25, Epoch: 248, Loss: 0.0003\nIter: 26, Epoch: 248, Loss: 0.0005\nIter: 27, Epoch: 248, Loss: 0.0004\nIter: 28, Epoch: 248, Loss: 0.0005\nIter: 29, Epoch: 248, Loss: 0.0005\nIter: 30, Epoch: 248, Loss: 0.0005\n","truncated":false}} +%--- +%[output:60e57fc0] +% data: {"dataType":"text","outputData":{"text":">> Epoch 248 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:05ca75b7] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 249, Loss: 0.0006\nIter: 2, Epoch: 249, Loss: 0.0004\nIter: 3, Epoch: 249, Loss: 0.0004\nIter: 4, Epoch: 249, Loss: 0.0004\nIter: 5, Epoch: 249, Loss: 0.0004\nIter: 6, Epoch: 249, Loss: 0.0003\nIter: 7, Epoch: 249, Loss: 0.0004\nIter: 8, Epoch: 249, Loss: 0.0009\nIter: 9, Epoch: 249, Loss: 0.0005\nIter: 10, Epoch: 249, Loss: 0.0007\nIter: 11, Epoch: 249, Loss: 0.0003\nIter: 12, Epoch: 249, Loss: 0.0003\nIter: 13, Epoch: 249, Loss: 0.0005\nIter: 14, Epoch: 249, Loss: 0.0006\nIter: 15, Epoch: 249, Loss: 0.0004\nIter: 16, Epoch: 249, Loss: 0.0005\nIter: 17, Epoch: 249, Loss: 0.0004\nIter: 18, Epoch: 249, Loss: 0.0005\nIter: 19, Epoch: 249, Loss: 0.0004\nIter: 20, Epoch: 249, Loss: 0.0005\nIter: 21, Epoch: 249, Loss: 0.0003\nIter: 22, Epoch: 249, Loss: 0.0008\nIter: 23, Epoch: 249, Loss: 0.0007\nIter: 24, Epoch: 249, Loss: 0.0002\nIter: 25, Epoch: 249, Loss: 0.0003\nIter: 26, Epoch: 249, Loss: 0.0005\nIter: 27, Epoch: 249, Loss: 0.0004\nIter: 28, Epoch: 249, Loss: 0.0005\nIter: 29, Epoch: 249, Loss: 0.0005\nIter: 30, Epoch: 249, Loss: 0.0005\n","truncated":false}} +%--- +%[output:1f5501cc] +% data: {"dataType":"text","outputData":{"text":">> Epoch 249 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:84846fe8] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 250, Loss: 0.0006\nIter: 2, Epoch: 250, Loss: 0.0004\nIter: 3, Epoch: 250, Loss: 0.0004\nIter: 4, Epoch: 250, Loss: 0.0004\nIter: 5, Epoch: 250, Loss: 0.0004\nIter: 6, Epoch: 250, Loss: 0.0003\nIter: 7, Epoch: 250, Loss: 0.0004\nIter: 8, Epoch: 250, Loss: 0.0009\nIter: 9, Epoch: 250, Loss: 0.0005\nIter: 10, Epoch: 250, Loss: 0.0007\nIter: 11, Epoch: 250, Loss: 0.0003\nIter: 12, Epoch: 250, Loss: 0.0003\nIter: 13, Epoch: 250, Loss: 0.0004\nIter: 14, Epoch: 250, Loss: 0.0006\nIter: 15, Epoch: 250, Loss: 0.0004\nIter: 16, Epoch: 250, Loss: 0.0005\nIter: 17, Epoch: 250, Loss: 0.0004\nIter: 18, Epoch: 250, Loss: 0.0004\nIter: 19, Epoch: 250, Loss: 0.0004\nIter: 20, Epoch: 250, Loss: 0.0005\nIter: 21, Epoch: 250, Loss: 0.0003\nIter: 22, Epoch: 250, Loss: 0.0008\nIter: 23, Epoch: 250, Loss: 0.0007\nIter: 24, Epoch: 250, Loss: 0.0002\nIter: 25, Epoch: 250, Loss: 0.0003\nIter: 26, Epoch: 250, Loss: 0.0005\nIter: 27, Epoch: 250, Loss: 0.0004\nIter: 28, Epoch: 250, Loss: 0.0005\nIter: 29, Epoch: 250, Loss: 0.0005\nIter: 30, Epoch: 250, Loss: 0.0005\n","truncated":false}} +%--- +%[output:11c26538] +% data: {"dataType":"text","outputData":{"text":">> Epoch 250 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:33b2fa90] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 251, Loss: 0.0006\nIter: 2, Epoch: 251, Loss: 0.0004\nIter: 3, Epoch: 251, Loss: 0.0004\nIter: 4, Epoch: 251, Loss: 0.0004\nIter: 5, Epoch: 251, Loss: 0.0004\nIter: 6, Epoch: 251, Loss: 0.0003\nIter: 7, Epoch: 251, Loss: 0.0004\nIter: 8, Epoch: 251, Loss: 0.0009\nIter: 9, Epoch: 251, Loss: 0.0005\nIter: 10, Epoch: 251, Loss: 0.0007\nIter: 11, Epoch: 251, Loss: 0.0003\nIter: 12, Epoch: 251, Loss: 0.0003\nIter: 13, Epoch: 251, Loss: 0.0004\nIter: 14, Epoch: 251, Loss: 0.0006\nIter: 15, Epoch: 251, Loss: 0.0004\nIter: 16, Epoch: 251, Loss: 0.0005\nIter: 17, Epoch: 251, Loss: 0.0004\nIter: 18, Epoch: 251, Loss: 0.0004\nIter: 19, Epoch: 251, Loss: 0.0004\nIter: 20, Epoch: 251, Loss: 0.0004\nIter: 21, Epoch: 251, Loss: 0.0003\nIter: 22, Epoch: 251, Loss: 0.0008\nIter: 23, Epoch: 251, Loss: 0.0006\nIter: 24, Epoch: 251, Loss: 0.0002\nIter: 25, Epoch: 251, Loss: 0.0003\nIter: 26, Epoch: 251, Loss: 0.0005\nIter: 27, Epoch: 251, Loss: 0.0004\nIter: 28, Epoch: 251, Loss: 0.0005\nIter: 29, Epoch: 251, Loss: 0.0005\nIter: 30, Epoch: 251, Loss: 0.0005\n","truncated":false}} +%--- +%[output:0583ed22] +% data: {"dataType":"text","outputData":{"text":">> Epoch 251 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:4be59cf3] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 252, Loss: 0.0006\nIter: 2, Epoch: 252, Loss: 0.0004\nIter: 3, Epoch: 252, Loss: 0.0004\nIter: 4, Epoch: 252, Loss: 0.0004\nIter: 5, Epoch: 252, Loss: 0.0004\nIter: 6, Epoch: 252, Loss: 0.0003\nIter: 7, Epoch: 252, Loss: 0.0004\nIter: 8, Epoch: 252, Loss: 0.0009\nIter: 9, Epoch: 252, Loss: 0.0005\nIter: 10, Epoch: 252, Loss: 0.0007\nIter: 11, Epoch: 252, Loss: 0.0003\nIter: 12, Epoch: 252, Loss: 0.0003\nIter: 13, Epoch: 252, Loss: 0.0004\nIter: 14, Epoch: 252, Loss: 0.0006\nIter: 15, Epoch: 252, Loss: 0.0004\nIter: 16, Epoch: 252, Loss: 0.0005\nIter: 17, Epoch: 252, Loss: 0.0004\nIter: 18, Epoch: 252, Loss: 0.0004\nIter: 19, Epoch: 252, Loss: 0.0004\nIter: 20, Epoch: 252, Loss: 0.0004\nIter: 21, Epoch: 252, Loss: 0.0003\nIter: 22, Epoch: 252, Loss: 0.0008\nIter: 23, Epoch: 252, Loss: 0.0006\nIter: 24, Epoch: 252, Loss: 0.0002\nIter: 25, Epoch: 252, Loss: 0.0003\nIter: 26, Epoch: 252, Loss: 0.0005\nIter: 27, Epoch: 252, Loss: 0.0003\nIter: 28, Epoch: 252, Loss: 0.0005\nIter: 29, Epoch: 252, Loss: 0.0005\nIter: 30, Epoch: 252, Loss: 0.0005\n","truncated":false}} +%--- +%[output:191c8a62] +% data: {"dataType":"text","outputData":{"text":">> Epoch 252 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:2ad7b632] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 253, Loss: 0.0006\nIter: 2, Epoch: 253, Loss: 0.0004\nIter: 3, Epoch: 253, Loss: 0.0004\nIter: 4, Epoch: 253, Loss: 0.0004\nIter: 5, Epoch: 253, Loss: 0.0004\nIter: 6, Epoch: 253, Loss: 0.0003\nIter: 7, Epoch: 253, Loss: 0.0004\nIter: 8, Epoch: 253, Loss: 0.0009\nIter: 9, Epoch: 253, Loss: 0.0005\nIter: 10, Epoch: 253, Loss: 0.0007\nIter: 11, Epoch: 253, Loss: 0.0003\nIter: 12, Epoch: 253, Loss: 0.0003\nIter: 13, Epoch: 253, Loss: 0.0004\nIter: 14, Epoch: 253, Loss: 0.0006\nIter: 15, Epoch: 253, Loss: 0.0004\nIter: 16, Epoch: 253, Loss: 0.0005\nIter: 17, Epoch: 253, Loss: 0.0004\nIter: 18, Epoch: 253, Loss: 0.0004\nIter: 19, Epoch: 253, Loss: 0.0004\nIter: 20, Epoch: 253, Loss: 0.0004\nIter: 21, Epoch: 253, Loss: 0.0003\nIter: 22, Epoch: 253, Loss: 0.0008\nIter: 23, Epoch: 253, Loss: 0.0006\nIter: 24, Epoch: 253, Loss: 0.0002\nIter: 25, Epoch: 253, Loss: 0.0003\nIter: 26, Epoch: 253, Loss: 0.0005\nIter: 27, Epoch: 253, Loss: 0.0003\nIter: 28, Epoch: 253, Loss: 0.0005\nIter: 29, Epoch: 253, Loss: 0.0005\nIter: 30, Epoch: 253, Loss: 0.0005\n","truncated":false}} +%--- +%[output:607270f2] +% data: {"dataType":"text","outputData":{"text":">> Epoch 253 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:7b93a34d] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 254, Loss: 0.0007\nIter: 2, Epoch: 254, Loss: 0.0004\nIter: 3, Epoch: 254, Loss: 0.0004\nIter: 4, Epoch: 254, Loss: 0.0004\nIter: 5, Epoch: 254, Loss: 0.0004\nIter: 6, Epoch: 254, Loss: 0.0003\nIter: 7, Epoch: 254, Loss: 0.0004\nIter: 8, Epoch: 254, Loss: 0.0009\nIter: 9, Epoch: 254, Loss: 0.0005\nIter: 10, Epoch: 254, Loss: 0.0007\nIter: 11, Epoch: 254, Loss: 0.0003\nIter: 12, Epoch: 254, Loss: 0.0003\nIter: 13, Epoch: 254, Loss: 0.0004\nIter: 14, Epoch: 254, Loss: 0.0006\nIter: 15, Epoch: 254, Loss: 0.0004\nIter: 16, Epoch: 254, Loss: 0.0005\nIter: 17, Epoch: 254, Loss: 0.0004\nIter: 18, Epoch: 254, Loss: 0.0004\nIter: 19, Epoch: 254, Loss: 0.0004\nIter: 20, Epoch: 254, Loss: 0.0004\nIter: 21, Epoch: 254, Loss: 0.0003\nIter: 22, Epoch: 254, Loss: 0.0008\nIter: 23, Epoch: 254, Loss: 0.0006\nIter: 24, Epoch: 254, Loss: 0.0002\nIter: 25, Epoch: 254, Loss: 0.0003\nIter: 26, Epoch: 254, Loss: 0.0005\nIter: 27, Epoch: 254, Loss: 0.0003\nIter: 28, Epoch: 254, Loss: 0.0005\nIter: 29, Epoch: 254, Loss: 0.0005\nIter: 30, Epoch: 254, Loss: 0.0005\n","truncated":false}} +%--- +%[output:5622e938] +% data: {"dataType":"text","outputData":{"text":">> Epoch 254 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:3ff98b25] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 255, Loss: 0.0007\nIter: 2, Epoch: 255, Loss: 0.0004\nIter: 3, Epoch: 255, Loss: 0.0004\nIter: 4, Epoch: 255, Loss: 0.0004\nIter: 5, Epoch: 255, Loss: 0.0004\nIter: 6, Epoch: 255, Loss: 0.0003\nIter: 7, Epoch: 255, Loss: 0.0004\nIter: 8, Epoch: 255, Loss: 0.0009\nIter: 9, Epoch: 255, Loss: 0.0005\nIter: 10, Epoch: 255, Loss: 0.0007\nIter: 11, Epoch: 255, Loss: 0.0003\nIter: 12, Epoch: 255, Loss: 0.0003\nIter: 13, Epoch: 255, Loss: 0.0004\nIter: 14, Epoch: 255, Loss: 0.0006\nIter: 15, Epoch: 255, Loss: 0.0004\nIter: 16, Epoch: 255, Loss: 0.0005\nIter: 17, Epoch: 255, Loss: 0.0004\nIter: 18, Epoch: 255, Loss: 0.0004\nIter: 19, Epoch: 255, Loss: 0.0004\nIter: 20, Epoch: 255, Loss: 0.0004\nIter: 21, Epoch: 255, Loss: 0.0003\nIter: 22, Epoch: 255, Loss: 0.0008\nIter: 23, Epoch: 255, Loss: 0.0006\nIter: 24, Epoch: 255, Loss: 0.0002\nIter: 25, Epoch: 255, Loss: 0.0003\nIter: 26, Epoch: 255, Loss: 0.0005\nIter: 27, Epoch: 255, Loss: 0.0003\nIter: 28, Epoch: 255, Loss: 0.0005\nIter: 29, Epoch: 255, Loss: 0.0005\nIter: 30, Epoch: 255, Loss: 0.0005\n","truncated":false}} +%--- +%[output:3be41725] +% data: {"dataType":"text","outputData":{"text":">> Epoch 255 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:720bbf08] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 256, Loss: 0.0007\nIter: 2, Epoch: 256, Loss: 0.0004\nIter: 3, Epoch: 256, Loss: 0.0004\nIter: 4, Epoch: 256, Loss: 0.0004\nIter: 5, Epoch: 256, Loss: 0.0004\nIter: 6, Epoch: 256, Loss: 0.0003\nIter: 7, Epoch: 256, Loss: 0.0004\nIter: 8, Epoch: 256, Loss: 0.0009\nIter: 9, Epoch: 256, Loss: 0.0005\nIter: 10, Epoch: 256, Loss: 0.0007\nIter: 11, Epoch: 256, Loss: 0.0003\nIter: 12, Epoch: 256, Loss: 0.0003\nIter: 13, Epoch: 256, Loss: 0.0004\nIter: 14, Epoch: 256, Loss: 0.0006\nIter: 15, Epoch: 256, Loss: 0.0004\nIter: 16, Epoch: 256, Loss: 0.0005\nIter: 17, Epoch: 256, Loss: 0.0004\nIter: 18, Epoch: 256, Loss: 0.0004\nIter: 19, Epoch: 256, Loss: 0.0004\nIter: 20, Epoch: 256, Loss: 0.0004\nIter: 21, Epoch: 256, Loss: 0.0003\nIter: 22, Epoch: 256, Loss: 0.0008\nIter: 23, Epoch: 256, Loss: 0.0006\nIter: 24, Epoch: 256, Loss: 0.0002\nIter: 25, Epoch: 256, Loss: 0.0003\nIter: 26, Epoch: 256, Loss: 0.0005\nIter: 27, Epoch: 256, Loss: 0.0003\nIter: 28, Epoch: 256, Loss: 0.0004\nIter: 29, Epoch: 256, Loss: 0.0005\nIter: 30, Epoch: 256, Loss: 0.0005\n","truncated":false}} +%--- +%[output:4a9bc037] +% data: {"dataType":"text","outputData":{"text":">> Epoch 256 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:4c94c0df] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 257, Loss: 0.0007\nIter: 2, Epoch: 257, Loss: 0.0004\nIter: 3, Epoch: 257, Loss: 0.0004\nIter: 4, Epoch: 257, Loss: 0.0004\nIter: 5, Epoch: 257, Loss: 0.0004\nIter: 6, Epoch: 257, Loss: 0.0003\nIter: 7, Epoch: 257, Loss: 0.0004\nIter: 8, Epoch: 257, Loss: 0.0009\nIter: 9, Epoch: 257, Loss: 0.0005\nIter: 10, Epoch: 257, Loss: 0.0007\nIter: 11, Epoch: 257, Loss: 0.0003\nIter: 12, Epoch: 257, Loss: 0.0003\nIter: 13, Epoch: 257, Loss: 0.0004\nIter: 14, Epoch: 257, Loss: 0.0006\nIter: 15, Epoch: 257, Loss: 0.0004\nIter: 16, Epoch: 257, Loss: 0.0005\nIter: 17, Epoch: 257, Loss: 0.0004\nIter: 18, Epoch: 257, Loss: 0.0004\nIter: 19, Epoch: 257, Loss: 0.0004\nIter: 20, Epoch: 257, Loss: 0.0004\nIter: 21, Epoch: 257, Loss: 0.0003\nIter: 22, Epoch: 257, Loss: 0.0008\nIter: 23, Epoch: 257, Loss: 0.0006\nIter: 24, Epoch: 257, Loss: 0.0002\nIter: 25, Epoch: 257, Loss: 0.0003\nIter: 26, Epoch: 257, Loss: 0.0005\nIter: 27, Epoch: 257, Loss: 0.0003\nIter: 28, Epoch: 257, Loss: 0.0004\nIter: 29, Epoch: 257, Loss: 0.0005\nIter: 30, Epoch: 257, Loss: 0.0005\n","truncated":false}} +%--- +%[output:00066e3b] +% data: {"dataType":"text","outputData":{"text":">> Epoch 257 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:74c63ce7] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 258, Loss: 0.0007\nIter: 2, Epoch: 258, Loss: 0.0004\nIter: 3, Epoch: 258, Loss: 0.0004\nIter: 4, Epoch: 258, Loss: 0.0004\nIter: 5, Epoch: 258, Loss: 0.0004\nIter: 6, Epoch: 258, Loss: 0.0003\nIter: 7, Epoch: 258, Loss: 0.0004\nIter: 8, Epoch: 258, Loss: 0.0009\nIter: 9, Epoch: 258, Loss: 0.0005\nIter: 10, Epoch: 258, Loss: 0.0007\nIter: 11, Epoch: 258, Loss: 0.0003\nIter: 12, Epoch: 258, Loss: 0.0003\nIter: 13, Epoch: 258, Loss: 0.0004\nIter: 14, Epoch: 258, Loss: 0.0006\nIter: 15, Epoch: 258, Loss: 0.0004\nIter: 16, Epoch: 258, Loss: 0.0005\nIter: 17, Epoch: 258, Loss: 0.0004\nIter: 18, Epoch: 258, Loss: 0.0004\nIter: 19, Epoch: 258, Loss: 0.0004\nIter: 20, Epoch: 258, Loss: 0.0004\nIter: 21, Epoch: 258, Loss: 0.0003\nIter: 22, Epoch: 258, Loss: 0.0008\nIter: 23, Epoch: 258, Loss: 0.0006\nIter: 24, Epoch: 258, Loss: 0.0002\nIter: 25, Epoch: 258, Loss: 0.0003\nIter: 26, Epoch: 258, Loss: 0.0005\nIter: 27, Epoch: 258, Loss: 0.0003\nIter: 28, Epoch: 258, Loss: 0.0004\nIter: 29, Epoch: 258, Loss: 0.0005\nIter: 30, Epoch: 258, Loss: 0.0005\n","truncated":false}} +%--- +%[output:926ecf1c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 258 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:0ea1556d] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 259, Loss: 0.0007\nIter: 2, Epoch: 259, Loss: 0.0004\nIter: 3, Epoch: 259, Loss: 0.0004\nIter: 4, Epoch: 259, Loss: 0.0004\nIter: 5, Epoch: 259, Loss: 0.0004\nIter: 6, Epoch: 259, Loss: 0.0003\nIter: 7, Epoch: 259, Loss: 0.0004\nIter: 8, Epoch: 259, Loss: 0.0009\nIter: 9, Epoch: 259, Loss: 0.0005\nIter: 10, Epoch: 259, Loss: 0.0007\nIter: 11, Epoch: 259, Loss: 0.0004\nIter: 12, Epoch: 259, Loss: 0.0003\nIter: 13, Epoch: 259, Loss: 0.0004\nIter: 14, Epoch: 259, Loss: 0.0006\nIter: 15, Epoch: 259, Loss: 0.0004\nIter: 16, Epoch: 259, Loss: 0.0005\nIter: 17, Epoch: 259, Loss: 0.0004\nIter: 18, Epoch: 259, Loss: 0.0004\nIter: 19, Epoch: 259, Loss: 0.0005\nIter: 20, Epoch: 259, Loss: 0.0004\nIter: 21, Epoch: 259, Loss: 0.0003\nIter: 22, Epoch: 259, Loss: 0.0007\nIter: 23, Epoch: 259, Loss: 0.0006\nIter: 24, Epoch: 259, Loss: 0.0002\nIter: 25, Epoch: 259, Loss: 0.0003\nIter: 26, Epoch: 259, Loss: 0.0005\nIter: 27, Epoch: 259, Loss: 0.0003\nIter: 28, Epoch: 259, Loss: 0.0004\nIter: 29, Epoch: 259, Loss: 0.0005\nIter: 30, Epoch: 259, Loss: 0.0005\n","truncated":false}} +%--- +%[output:0db85c76] +% data: {"dataType":"text","outputData":{"text":">> Epoch 259 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:6e4ae90a] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 260, Loss: 0.0006\nIter: 2, Epoch: 260, Loss: 0.0004\nIter: 3, Epoch: 260, Loss: 0.0004\nIter: 4, Epoch: 260, Loss: 0.0004\nIter: 5, Epoch: 260, Loss: 0.0004\nIter: 6, Epoch: 260, Loss: 0.0003\nIter: 7, Epoch: 260, Loss: 0.0004\nIter: 8, Epoch: 260, Loss: 0.0009\nIter: 9, Epoch: 260, Loss: 0.0005\nIter: 10, Epoch: 260, Loss: 0.0007\nIter: 11, Epoch: 260, Loss: 0.0004\nIter: 12, Epoch: 260, Loss: 0.0003\nIter: 13, Epoch: 260, Loss: 0.0004\nIter: 14, Epoch: 260, Loss: 0.0005\nIter: 15, Epoch: 260, Loss: 0.0004\nIter: 16, Epoch: 260, Loss: 0.0005\nIter: 17, Epoch: 260, Loss: 0.0004\nIter: 18, Epoch: 260, Loss: 0.0004\nIter: 19, Epoch: 260, Loss: 0.0005\nIter: 20, Epoch: 260, Loss: 0.0004\nIter: 21, Epoch: 260, Loss: 0.0003\nIter: 22, Epoch: 260, Loss: 0.0007\nIter: 23, Epoch: 260, Loss: 0.0006\nIter: 24, Epoch: 260, Loss: 0.0002\nIter: 25, Epoch: 260, Loss: 0.0003\nIter: 26, Epoch: 260, Loss: 0.0005\nIter: 27, Epoch: 260, Loss: 0.0003\nIter: 28, Epoch: 260, Loss: 0.0004\nIter: 29, Epoch: 260, Loss: 0.0005\nIter: 30, Epoch: 260, Loss: 0.0005\n","truncated":false}} +%--- +%[output:07058dce] +% data: {"dataType":"text","outputData":{"text":">> Epoch 260 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:4ab8bed5] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 261, Loss: 0.0006\nIter: 2, Epoch: 261, Loss: 0.0004\nIter: 3, Epoch: 261, Loss: 0.0004\nIter: 4, Epoch: 261, Loss: 0.0004\nIter: 5, Epoch: 261, Loss: 0.0004\nIter: 6, Epoch: 261, Loss: 0.0003\nIter: 7, Epoch: 261, Loss: 0.0004\nIter: 8, Epoch: 261, Loss: 0.0009\nIter: 9, Epoch: 261, Loss: 0.0005\nIter: 10, Epoch: 261, Loss: 0.0007\nIter: 11, Epoch: 261, Loss: 0.0004\nIter: 12, Epoch: 261, Loss: 0.0003\nIter: 13, Epoch: 261, Loss: 0.0004\nIter: 14, Epoch: 261, Loss: 0.0005\nIter: 15, Epoch: 261, Loss: 0.0004\nIter: 16, Epoch: 261, Loss: 0.0005\nIter: 17, Epoch: 261, Loss: 0.0004\nIter: 18, Epoch: 261, Loss: 0.0004\nIter: 19, Epoch: 261, Loss: 0.0005\nIter: 20, Epoch: 261, Loss: 0.0004\nIter: 21, Epoch: 261, Loss: 0.0003\nIter: 22, Epoch: 261, Loss: 0.0007\nIter: 23, Epoch: 261, Loss: 0.0006\nIter: 24, Epoch: 261, Loss: 0.0002\nIter: 25, Epoch: 261, Loss: 0.0003\nIter: 26, Epoch: 261, Loss: 0.0005\nIter: 27, Epoch: 261, Loss: 0.0003\nIter: 28, Epoch: 261, Loss: 0.0004\nIter: 29, Epoch: 261, Loss: 0.0005\nIter: 30, Epoch: 261, Loss: 0.0005\n","truncated":false}} +%--- +%[output:4207ba2b] +% data: {"dataType":"text","outputData":{"text":">> Epoch 261 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:03ef4d44] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 262, Loss: 0.0006\nIter: 2, Epoch: 262, Loss: 0.0004\nIter: 3, Epoch: 262, Loss: 0.0004\nIter: 4, Epoch: 262, Loss: 0.0004\nIter: 5, Epoch: 262, Loss: 0.0004\nIter: 6, Epoch: 262, Loss: 0.0003\nIter: 7, Epoch: 262, Loss: 0.0004\nIter: 8, Epoch: 262, Loss: 0.0009\nIter: 9, Epoch: 262, Loss: 0.0004\nIter: 10, Epoch: 262, Loss: 0.0007\nIter: 11, Epoch: 262, Loss: 0.0003\nIter: 12, Epoch: 262, Loss: 0.0003\nIter: 13, Epoch: 262, Loss: 0.0004\nIter: 14, Epoch: 262, Loss: 0.0005\nIter: 15, Epoch: 262, Loss: 0.0004\nIter: 16, Epoch: 262, Loss: 0.0005\nIter: 17, Epoch: 262, Loss: 0.0004\nIter: 18, Epoch: 262, Loss: 0.0004\nIter: 19, Epoch: 262, Loss: 0.0005\nIter: 20, Epoch: 262, Loss: 0.0004\nIter: 21, Epoch: 262, Loss: 0.0003\nIter: 22, Epoch: 262, Loss: 0.0007\nIter: 23, Epoch: 262, Loss: 0.0006\nIter: 24, Epoch: 262, Loss: 0.0002\nIter: 25, Epoch: 262, Loss: 0.0003\nIter: 26, Epoch: 262, Loss: 0.0005\nIter: 27, Epoch: 262, Loss: 0.0003\nIter: 28, Epoch: 262, Loss: 0.0004\nIter: 29, Epoch: 262, Loss: 0.0005\nIter: 30, Epoch: 262, Loss: 0.0005\n","truncated":false}} +%--- +%[output:39199c17] +% data: {"dataType":"text","outputData":{"text":">> Epoch 262 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:31437bc9] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 263, Loss: 0.0006\nIter: 2, Epoch: 263, Loss: 0.0004\nIter: 3, Epoch: 263, Loss: 0.0004\nIter: 4, Epoch: 263, Loss: 0.0004\nIter: 5, Epoch: 263, Loss: 0.0004\nIter: 6, Epoch: 263, Loss: 0.0003\nIter: 7, Epoch: 263, Loss: 0.0004\nIter: 8, Epoch: 263, Loss: 0.0009\nIter: 9, Epoch: 263, Loss: 0.0004\nIter: 10, Epoch: 263, Loss: 0.0007\nIter: 11, Epoch: 263, Loss: 0.0003\nIter: 12, Epoch: 263, Loss: 0.0003\nIter: 13, Epoch: 263, Loss: 0.0004\nIter: 14, Epoch: 263, Loss: 0.0005\nIter: 15, Epoch: 263, Loss: 0.0004\nIter: 16, Epoch: 263, Loss: 0.0005\nIter: 17, Epoch: 263, Loss: 0.0004\nIter: 18, Epoch: 263, Loss: 0.0004\nIter: 19, Epoch: 263, Loss: 0.0004\nIter: 20, Epoch: 263, Loss: 0.0004\nIter: 21, Epoch: 263, Loss: 0.0003\nIter: 22, Epoch: 263, Loss: 0.0007\nIter: 23, Epoch: 263, Loss: 0.0006\nIter: 24, Epoch: 263, Loss: 0.0002\nIter: 25, Epoch: 263, Loss: 0.0003\nIter: 26, Epoch: 263, Loss: 0.0005\nIter: 27, Epoch: 263, Loss: 0.0003\nIter: 28, Epoch: 263, Loss: 0.0004\nIter: 29, Epoch: 263, Loss: 0.0005\nIter: 30, Epoch: 263, Loss: 0.0005\n","truncated":false}} +%--- +%[output:853642df] +% data: {"dataType":"text","outputData":{"text":">> Epoch 263 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:10b1afd6] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 264, Loss: 0.0006\nIter: 2, Epoch: 264, Loss: 0.0004\nIter: 3, Epoch: 264, Loss: 0.0004\nIter: 4, Epoch: 264, Loss: 0.0004\nIter: 5, Epoch: 264, Loss: 0.0004\nIter: 6, Epoch: 264, Loss: 0.0003\nIter: 7, Epoch: 264, Loss: 0.0004\nIter: 8, Epoch: 264, Loss: 0.0009\nIter: 9, Epoch: 264, Loss: 0.0004\nIter: 10, Epoch: 264, Loss: 0.0007\nIter: 11, Epoch: 264, Loss: 0.0003\nIter: 12, Epoch: 264, Loss: 0.0003\nIter: 13, Epoch: 264, Loss: 0.0004\nIter: 14, Epoch: 264, Loss: 0.0005\nIter: 15, Epoch: 264, Loss: 0.0004\nIter: 16, Epoch: 264, Loss: 0.0005\nIter: 17, Epoch: 264, Loss: 0.0004\nIter: 18, Epoch: 264, Loss: 0.0004\nIter: 19, Epoch: 264, Loss: 0.0004\nIter: 20, Epoch: 264, Loss: 0.0004\nIter: 21, Epoch: 264, Loss: 0.0003\nIter: 22, Epoch: 264, Loss: 0.0007\nIter: 23, Epoch: 264, Loss: 0.0006\nIter: 24, Epoch: 264, Loss: 0.0002\nIter: 25, Epoch: 264, Loss: 0.0003\nIter: 26, Epoch: 264, Loss: 0.0005\nIter: 27, Epoch: 264, Loss: 0.0003\nIter: 28, Epoch: 264, Loss: 0.0004\nIter: 29, Epoch: 264, Loss: 0.0005\nIter: 30, Epoch: 264, Loss: 0.0005\n","truncated":false}} +%--- +%[output:362e52a4] +% data: {"dataType":"text","outputData":{"text":">> Epoch 264 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:6b647426] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 265, Loss: 0.0006\nIter: 2, Epoch: 265, Loss: 0.0004\nIter: 3, Epoch: 265, Loss: 0.0004\nIter: 4, Epoch: 265, Loss: 0.0004\nIter: 5, Epoch: 265, Loss: 0.0004\nIter: 6, Epoch: 265, Loss: 0.0003\nIter: 7, Epoch: 265, Loss: 0.0004\nIter: 8, Epoch: 265, Loss: 0.0009\nIter: 9, Epoch: 265, Loss: 0.0004\nIter: 10, Epoch: 265, Loss: 0.0007\nIter: 11, Epoch: 265, Loss: 0.0003\nIter: 12, Epoch: 265, Loss: 0.0003\nIter: 13, Epoch: 265, Loss: 0.0004\nIter: 14, Epoch: 265, Loss: 0.0005\nIter: 15, Epoch: 265, Loss: 0.0004\nIter: 16, Epoch: 265, Loss: 0.0005\nIter: 17, Epoch: 265, Loss: 0.0004\nIter: 18, Epoch: 265, Loss: 0.0004\nIter: 19, Epoch: 265, Loss: 0.0004\nIter: 20, Epoch: 265, Loss: 0.0004\nIter: 21, Epoch: 265, Loss: 0.0002\nIter: 22, Epoch: 265, Loss: 0.0007\nIter: 23, Epoch: 265, Loss: 0.0006\nIter: 24, Epoch: 265, Loss: 0.0002\nIter: 25, Epoch: 265, Loss: 0.0003\nIter: 26, Epoch: 265, Loss: 0.0005\nIter: 27, Epoch: 265, Loss: 0.0003\nIter: 28, Epoch: 265, Loss: 0.0004\nIter: 29, Epoch: 265, Loss: 0.0005\nIter: 30, Epoch: 265, Loss: 0.0005\n","truncated":false}} +%--- +%[output:12f62da3] +% data: {"dataType":"text","outputData":{"text":">> Epoch 265 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:18f55326] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 266, Loss: 0.0006\nIter: 2, Epoch: 266, Loss: 0.0004\nIter: 3, Epoch: 266, Loss: 0.0004\nIter: 4, Epoch: 266, Loss: 0.0004\nIter: 5, Epoch: 266, Loss: 0.0004\nIter: 6, Epoch: 266, Loss: 0.0003\nIter: 7, Epoch: 266, Loss: 0.0004\nIter: 8, Epoch: 266, Loss: 0.0009\nIter: 9, Epoch: 266, Loss: 0.0004\nIter: 10, Epoch: 266, Loss: 0.0007\nIter: 11, Epoch: 266, Loss: 0.0003\nIter: 12, Epoch: 266, Loss: 0.0003\nIter: 13, Epoch: 266, Loss: 0.0003\nIter: 14, Epoch: 266, Loss: 0.0005\nIter: 15, Epoch: 266, Loss: 0.0004\nIter: 16, Epoch: 266, Loss: 0.0005\nIter: 17, Epoch: 266, Loss: 0.0004\nIter: 18, Epoch: 266, Loss: 0.0004\nIter: 19, Epoch: 266, Loss: 0.0004\nIter: 20, Epoch: 266, Loss: 0.0004\nIter: 21, Epoch: 266, Loss: 0.0002\nIter: 22, Epoch: 266, Loss: 0.0007\nIter: 23, Epoch: 266, Loss: 0.0007\nIter: 24, Epoch: 266, Loss: 0.0002\nIter: 25, Epoch: 266, Loss: 0.0003\nIter: 26, Epoch: 266, Loss: 0.0005\nIter: 27, Epoch: 266, Loss: 0.0003\nIter: 28, Epoch: 266, Loss: 0.0004\nIter: 29, Epoch: 266, Loss: 0.0005\nIter: 30, Epoch: 266, Loss: 0.0006\n","truncated":false}} +%--- +%[output:9b0b913c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 266 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:1c008433] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 267, Loss: 0.0006\nIter: 2, Epoch: 267, Loss: 0.0004\nIter: 3, Epoch: 267, Loss: 0.0004\nIter: 4, Epoch: 267, Loss: 0.0004\nIter: 5, Epoch: 267, Loss: 0.0004\nIter: 6, Epoch: 267, Loss: 0.0003\nIter: 7, Epoch: 267, Loss: 0.0004\nIter: 8, Epoch: 267, Loss: 0.0009\nIter: 9, Epoch: 267, Loss: 0.0004\nIter: 10, Epoch: 267, Loss: 0.0007\nIter: 11, Epoch: 267, Loss: 0.0003\nIter: 12, Epoch: 267, Loss: 0.0003\nIter: 13, Epoch: 267, Loss: 0.0003\nIter: 14, Epoch: 267, Loss: 0.0005\nIter: 15, Epoch: 267, Loss: 0.0004\nIter: 16, Epoch: 267, Loss: 0.0005\nIter: 17, Epoch: 267, Loss: 0.0004\nIter: 18, Epoch: 267, Loss: 0.0004\nIter: 19, Epoch: 267, Loss: 0.0004\nIter: 20, Epoch: 267, Loss: 0.0004\nIter: 21, Epoch: 267, Loss: 0.0002\nIter: 22, Epoch: 267, Loss: 0.0007\nIter: 23, Epoch: 267, Loss: 0.0007\nIter: 24, Epoch: 267, Loss: 0.0002\nIter: 25, Epoch: 267, Loss: 0.0003\nIter: 26, Epoch: 267, Loss: 0.0005\nIter: 27, Epoch: 267, Loss: 0.0003\nIter: 28, Epoch: 267, Loss: 0.0004\nIter: 29, Epoch: 267, Loss: 0.0005\nIter: 30, Epoch: 267, Loss: 0.0006\n","truncated":false}} +%--- +%[output:975173df] +% data: {"dataType":"text","outputData":{"text":">> Epoch 267 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:831451bb] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 268, Loss: 0.0006\nIter: 2, Epoch: 268, Loss: 0.0004\nIter: 3, Epoch: 268, Loss: 0.0004\nIter: 4, Epoch: 268, Loss: 0.0004\nIter: 5, Epoch: 268, Loss: 0.0004\nIter: 6, Epoch: 268, Loss: 0.0003\nIter: 7, Epoch: 268, Loss: 0.0004\nIter: 8, Epoch: 268, Loss: 0.0009\nIter: 9, Epoch: 268, Loss: 0.0004\nIter: 10, Epoch: 268, Loss: 0.0007\nIter: 11, Epoch: 268, Loss: 0.0003\nIter: 12, Epoch: 268, Loss: 0.0003\nIter: 13, Epoch: 268, Loss: 0.0003\nIter: 14, Epoch: 268, Loss: 0.0005\nIter: 15, Epoch: 268, Loss: 0.0004\nIter: 16, Epoch: 268, Loss: 0.0005\nIter: 17, Epoch: 268, Loss: 0.0004\nIter: 18, Epoch: 268, Loss: 0.0004\nIter: 19, Epoch: 268, Loss: 0.0004\nIter: 20, Epoch: 268, Loss: 0.0004\nIter: 21, Epoch: 268, Loss: 0.0002\nIter: 22, Epoch: 268, Loss: 0.0007\nIter: 23, Epoch: 268, Loss: 0.0007\nIter: 24, Epoch: 268, Loss: 0.0002\nIter: 25, Epoch: 268, Loss: 0.0003\nIter: 26, Epoch: 268, Loss: 0.0005\nIter: 27, Epoch: 268, Loss: 0.0003\nIter: 28, Epoch: 268, Loss: 0.0004\nIter: 29, Epoch: 268, Loss: 0.0005\nIter: 30, Epoch: 268, Loss: 0.0006\n","truncated":false}} +%--- +%[output:1b65c002] +% data: {"dataType":"text","outputData":{"text":">> Epoch 268 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:995c9a36] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 269, Loss: 0.0006\nIter: 2, Epoch: 269, Loss: 0.0004\nIter: 3, Epoch: 269, Loss: 0.0004\nIter: 4, Epoch: 269, Loss: 0.0004\nIter: 5, Epoch: 269, Loss: 0.0003\nIter: 6, Epoch: 269, Loss: 0.0003\nIter: 7, Epoch: 269, Loss: 0.0004\nIter: 8, Epoch: 269, Loss: 0.0009\nIter: 9, Epoch: 269, Loss: 0.0004\nIter: 10, Epoch: 269, Loss: 0.0007\nIter: 11, Epoch: 269, Loss: 0.0003\nIter: 12, Epoch: 269, Loss: 0.0003\nIter: 13, Epoch: 269, Loss: 0.0003\nIter: 14, Epoch: 269, Loss: 0.0005\nIter: 15, Epoch: 269, Loss: 0.0004\nIter: 16, Epoch: 269, Loss: 0.0005\nIter: 17, Epoch: 269, Loss: 0.0004\nIter: 18, Epoch: 269, Loss: 0.0004\nIter: 19, Epoch: 269, Loss: 0.0004\nIter: 20, Epoch: 269, Loss: 0.0004\nIter: 21, Epoch: 269, Loss: 0.0002\nIter: 22, Epoch: 269, Loss: 0.0007\nIter: 23, Epoch: 269, Loss: 0.0007\nIter: 24, Epoch: 269, Loss: 0.0002\nIter: 25, Epoch: 269, Loss: 0.0003\nIter: 26, Epoch: 269, Loss: 0.0005\nIter: 27, Epoch: 269, Loss: 0.0003\nIter: 28, Epoch: 269, Loss: 0.0004\nIter: 29, Epoch: 269, Loss: 0.0005\nIter: 30, Epoch: 269, Loss: 0.0006\n","truncated":false}} +%--- +%[output:53b87dba] +% data: {"dataType":"text","outputData":{"text":">> Epoch 269 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:3b7e0b1c] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 270, Loss: 0.0006\nIter: 2, Epoch: 270, Loss: 0.0004\nIter: 3, Epoch: 270, Loss: 0.0004\nIter: 4, Epoch: 270, Loss: 0.0004\nIter: 5, Epoch: 270, Loss: 0.0003\nIter: 6, Epoch: 270, Loss: 0.0003\nIter: 7, Epoch: 270, Loss: 0.0004\nIter: 8, Epoch: 270, Loss: 0.0009\nIter: 9, Epoch: 270, Loss: 0.0004\nIter: 10, Epoch: 270, Loss: 0.0007\nIter: 11, Epoch: 270, Loss: 0.0003\nIter: 12, Epoch: 270, Loss: 0.0003\nIter: 13, Epoch: 270, Loss: 0.0003\nIter: 14, Epoch: 270, Loss: 0.0005\nIter: 15, Epoch: 270, Loss: 0.0004\nIter: 16, Epoch: 270, Loss: 0.0005\nIter: 17, Epoch: 270, Loss: 0.0004\nIter: 18, Epoch: 270, Loss: 0.0004\nIter: 19, Epoch: 270, Loss: 0.0004\nIter: 20, Epoch: 270, Loss: 0.0004\nIter: 21, Epoch: 270, Loss: 0.0002\nIter: 22, Epoch: 270, Loss: 0.0007\nIter: 23, Epoch: 270, Loss: 0.0007\nIter: 24, Epoch: 270, Loss: 0.0002\nIter: 25, Epoch: 270, Loss: 0.0003\nIter: 26, Epoch: 270, Loss: 0.0005\nIter: 27, Epoch: 270, Loss: 0.0003\nIter: 28, Epoch: 270, Loss: 0.0004\nIter: 29, Epoch: 270, Loss: 0.0005\nIter: 30, Epoch: 270, Loss: 0.0006\n","truncated":false}} +%--- +%[output:556f3d01] +% data: {"dataType":"text","outputData":{"text":">> Epoch 270 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:44f95a0a] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 271, Loss: 0.0006\nIter: 2, Epoch: 271, Loss: 0.0004\nIter: 3, Epoch: 271, Loss: 0.0004\nIter: 4, Epoch: 271, Loss: 0.0004\nIter: 5, Epoch: 271, Loss: 0.0003\nIter: 6, Epoch: 271, Loss: 0.0003\nIter: 7, Epoch: 271, Loss: 0.0004\nIter: 8, Epoch: 271, Loss: 0.0009\nIter: 9, Epoch: 271, Loss: 0.0004\nIter: 10, Epoch: 271, Loss: 0.0007\nIter: 11, Epoch: 271, Loss: 0.0003\nIter: 12, Epoch: 271, Loss: 0.0003\nIter: 13, Epoch: 271, Loss: 0.0003\nIter: 14, Epoch: 271, Loss: 0.0005\nIter: 15, Epoch: 271, Loss: 0.0004\nIter: 16, Epoch: 271, Loss: 0.0005\nIter: 17, Epoch: 271, Loss: 0.0004\nIter: 18, Epoch: 271, Loss: 0.0004\nIter: 19, Epoch: 271, Loss: 0.0004\nIter: 20, Epoch: 271, Loss: 0.0004\nIter: 21, Epoch: 271, Loss: 0.0002\nIter: 22, Epoch: 271, Loss: 0.0007\nIter: 23, Epoch: 271, Loss: 0.0007\nIter: 24, Epoch: 271, Loss: 0.0002\nIter: 25, Epoch: 271, Loss: 0.0003\nIter: 26, Epoch: 271, Loss: 0.0005\nIter: 27, Epoch: 271, Loss: 0.0003\nIter: 28, Epoch: 271, Loss: 0.0004\nIter: 29, Epoch: 271, Loss: 0.0005\nIter: 30, Epoch: 271, Loss: 0.0006\n","truncated":false}} +%--- +%[output:0c2e92f6] +% data: {"dataType":"text","outputData":{"text":">> Epoch 271 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:2ca643cb] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 272, Loss: 0.0005\nIter: 2, Epoch: 272, Loss: 0.0004\nIter: 3, Epoch: 272, Loss: 0.0004\nIter: 4, Epoch: 272, Loss: 0.0004\nIter: 5, Epoch: 272, Loss: 0.0003\nIter: 6, Epoch: 272, Loss: 0.0003\nIter: 7, Epoch: 272, Loss: 0.0004\nIter: 8, Epoch: 272, Loss: 0.0009\nIter: 9, Epoch: 272, Loss: 0.0004\nIter: 10, Epoch: 272, Loss: 0.0007\nIter: 11, Epoch: 272, Loss: 0.0003\nIter: 12, Epoch: 272, Loss: 0.0003\nIter: 13, Epoch: 272, Loss: 0.0003\nIter: 14, Epoch: 272, Loss: 0.0005\nIter: 15, Epoch: 272, Loss: 0.0004\nIter: 16, Epoch: 272, Loss: 0.0005\nIter: 17, Epoch: 272, Loss: 0.0004\nIter: 18, Epoch: 272, Loss: 0.0004\nIter: 19, Epoch: 272, Loss: 0.0004\nIter: 20, Epoch: 272, Loss: 0.0004\nIter: 21, Epoch: 272, Loss: 0.0002\nIter: 22, Epoch: 272, Loss: 0.0007\nIter: 23, Epoch: 272, Loss: 0.0007\nIter: 24, Epoch: 272, Loss: 0.0002\nIter: 25, Epoch: 272, Loss: 0.0003\nIter: 26, Epoch: 272, Loss: 0.0004\nIter: 27, Epoch: 272, Loss: 0.0003\nIter: 28, Epoch: 272, Loss: 0.0004\nIter: 29, Epoch: 272, Loss: 0.0005\nIter: 30, Epoch: 272, Loss: 0.0006\n","truncated":false}} +%--- +%[output:54b0d143] +% data: {"dataType":"text","outputData":{"text":">> Epoch 272 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:302c0cd0] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 273, Loss: 0.0005\nIter: 2, Epoch: 273, Loss: 0.0004\nIter: 3, Epoch: 273, Loss: 0.0004\nIter: 4, Epoch: 273, Loss: 0.0004\nIter: 5, Epoch: 273, Loss: 0.0003\nIter: 6, Epoch: 273, Loss: 0.0003\nIter: 7, Epoch: 273, Loss: 0.0004\nIter: 8, Epoch: 273, Loss: 0.0009\nIter: 9, Epoch: 273, Loss: 0.0004\nIter: 10, Epoch: 273, Loss: 0.0007\nIter: 11, Epoch: 273, Loss: 0.0003\nIter: 12, Epoch: 273, Loss: 0.0003\nIter: 13, Epoch: 273, Loss: 0.0003\nIter: 14, Epoch: 273, Loss: 0.0005\nIter: 15, Epoch: 273, Loss: 0.0004\nIter: 16, Epoch: 273, Loss: 0.0005\nIter: 17, Epoch: 273, Loss: 0.0004\nIter: 18, Epoch: 273, Loss: 0.0004\nIter: 19, Epoch: 273, Loss: 0.0004\nIter: 20, Epoch: 273, Loss: 0.0004\nIter: 21, Epoch: 273, Loss: 0.0002\nIter: 22, Epoch: 273, Loss: 0.0007\nIter: 23, Epoch: 273, Loss: 0.0007\nIter: 24, Epoch: 273, Loss: 0.0002\nIter: 25, Epoch: 273, Loss: 0.0003\nIter: 26, Epoch: 273, Loss: 0.0004\nIter: 27, Epoch: 273, Loss: 0.0002\nIter: 28, Epoch: 273, Loss: 0.0004\nIter: 29, Epoch: 273, Loss: 0.0005\nIter: 30, Epoch: 273, Loss: 0.0006\n","truncated":false}} +%--- +%[output:17584135] +% data: {"dataType":"text","outputData":{"text":">> Epoch 273 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:27c00cfa] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 274, Loss: 0.0005\nIter: 2, Epoch: 274, Loss: 0.0004\nIter: 3, Epoch: 274, Loss: 0.0004\nIter: 4, Epoch: 274, Loss: 0.0004\nIter: 5, Epoch: 274, Loss: 0.0003\nIter: 6, Epoch: 274, Loss: 0.0003\nIter: 7, Epoch: 274, Loss: 0.0004\nIter: 8, Epoch: 274, Loss: 0.0009\nIter: 9, Epoch: 274, Loss: 0.0004\nIter: 10, Epoch: 274, Loss: 0.0007\nIter: 11, Epoch: 274, Loss: 0.0003\nIter: 12, Epoch: 274, Loss: 0.0003\nIter: 13, Epoch: 274, Loss: 0.0003\nIter: 14, Epoch: 274, Loss: 0.0005\nIter: 15, Epoch: 274, Loss: 0.0004\nIter: 16, Epoch: 274, Loss: 0.0005\nIter: 17, Epoch: 274, Loss: 0.0004\nIter: 18, Epoch: 274, Loss: 0.0004\nIter: 19, Epoch: 274, Loss: 0.0004\nIter: 20, Epoch: 274, Loss: 0.0004\nIter: 21, Epoch: 274, Loss: 0.0002\nIter: 22, Epoch: 274, Loss: 0.0007\nIter: 23, Epoch: 274, Loss: 0.0007\nIter: 24, Epoch: 274, Loss: 0.0002\nIter: 25, Epoch: 274, Loss: 0.0003\nIter: 26, Epoch: 274, Loss: 0.0004\nIter: 27, Epoch: 274, Loss: 0.0002\nIter: 28, Epoch: 274, Loss: 0.0004\nIter: 29, Epoch: 274, Loss: 0.0005\nIter: 30, Epoch: 274, Loss: 0.0006\n","truncated":false}} +%--- +%[output:1934d832] +% data: {"dataType":"text","outputData":{"text":">> Epoch 274 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:4bbd852d] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 275, Loss: 0.0005\nIter: 2, Epoch: 275, Loss: 0.0004\nIter: 3, Epoch: 275, Loss: 0.0004\nIter: 4, Epoch: 275, Loss: 0.0004\nIter: 5, Epoch: 275, Loss: 0.0003\nIter: 6, Epoch: 275, Loss: 0.0003\nIter: 7, Epoch: 275, Loss: 0.0004\nIter: 8, Epoch: 275, Loss: 0.0009\nIter: 9, Epoch: 275, Loss: 0.0004\nIter: 10, Epoch: 275, Loss: 0.0007\nIter: 11, Epoch: 275, Loss: 0.0003\nIter: 12, Epoch: 275, Loss: 0.0003\nIter: 13, Epoch: 275, Loss: 0.0003\nIter: 14, Epoch: 275, Loss: 0.0005\nIter: 15, Epoch: 275, Loss: 0.0004\nIter: 16, Epoch: 275, Loss: 0.0005\nIter: 17, Epoch: 275, Loss: 0.0004\nIter: 18, Epoch: 275, Loss: 0.0004\nIter: 19, Epoch: 275, Loss: 0.0004\nIter: 20, Epoch: 275, Loss: 0.0004\nIter: 21, Epoch: 275, Loss: 0.0002\nIter: 22, Epoch: 275, Loss: 0.0007\nIter: 23, Epoch: 275, Loss: 0.0007\nIter: 24, Epoch: 275, Loss: 0.0002\nIter: 25, Epoch: 275, Loss: 0.0003\nIter: 26, Epoch: 275, Loss: 0.0004\nIter: 27, Epoch: 275, Loss: 0.0002\nIter: 28, Epoch: 275, Loss: 0.0004\nIter: 29, Epoch: 275, Loss: 0.0005\nIter: 30, Epoch: 275, Loss: 0.0006\n","truncated":false}} +%--- +%[output:7ca352b0] +% data: {"dataType":"text","outputData":{"text":">> Epoch 275 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:4aad7518] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 276, Loss: 0.0005\nIter: 2, Epoch: 276, Loss: 0.0005\nIter: 3, Epoch: 276, Loss: 0.0004\nIter: 4, Epoch: 276, Loss: 0.0004\nIter: 5, Epoch: 276, Loss: 0.0003\nIter: 6, Epoch: 276, Loss: 0.0003\nIter: 7, Epoch: 276, Loss: 0.0004\nIter: 8, Epoch: 276, Loss: 0.0009\nIter: 9, Epoch: 276, Loss: 0.0004\nIter: 10, Epoch: 276, Loss: 0.0007\nIter: 11, Epoch: 276, Loss: 0.0003\nIter: 12, Epoch: 276, Loss: 0.0003\nIter: 13, Epoch: 276, Loss: 0.0003\nIter: 14, Epoch: 276, Loss: 0.0005\nIter: 15, Epoch: 276, Loss: 0.0004\nIter: 16, Epoch: 276, Loss: 0.0005\nIter: 17, Epoch: 276, Loss: 0.0004\nIter: 18, Epoch: 276, Loss: 0.0004\nIter: 19, Epoch: 276, Loss: 0.0004\nIter: 20, Epoch: 276, Loss: 0.0004\nIter: 21, Epoch: 276, Loss: 0.0002\nIter: 22, Epoch: 276, Loss: 0.0007\nIter: 23, Epoch: 276, Loss: 0.0007\nIter: 24, Epoch: 276, Loss: 0.0002\nIter: 25, Epoch: 276, Loss: 0.0003\nIter: 26, Epoch: 276, Loss: 0.0004\nIter: 27, Epoch: 276, Loss: 0.0002\nIter: 28, Epoch: 276, Loss: 0.0004\nIter: 29, Epoch: 276, Loss: 0.0006\nIter: 30, Epoch: 276, Loss: 0.0006\n","truncated":false}} +%--- +%[output:9efdf189] +% data: {"dataType":"text","outputData":{"text":">> Epoch 276 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:69f09dd7] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 277, Loss: 0.0005\nIter: 2, Epoch: 277, Loss: 0.0005\nIter: 3, Epoch: 277, Loss: 0.0004\nIter: 4, Epoch: 277, Loss: 0.0004\nIter: 5, Epoch: 277, Loss: 0.0003\nIter: 6, Epoch: 277, Loss: 0.0003\nIter: 7, Epoch: 277, Loss: 0.0004\nIter: 8, Epoch: 277, Loss: 0.0009\nIter: 9, Epoch: 277, Loss: 0.0004\nIter: 10, Epoch: 277, Loss: 0.0007\nIter: 11, Epoch: 277, Loss: 0.0003\nIter: 12, Epoch: 277, Loss: 0.0003\nIter: 13, Epoch: 277, Loss: 0.0003\nIter: 14, Epoch: 277, Loss: 0.0005\nIter: 15, Epoch: 277, Loss: 0.0004\nIter: 16, Epoch: 277, Loss: 0.0005\nIter: 17, Epoch: 277, Loss: 0.0004\nIter: 18, Epoch: 277, Loss: 0.0004\nIter: 19, Epoch: 277, Loss: 0.0003\nIter: 20, Epoch: 277, Loss: 0.0004\nIter: 21, Epoch: 277, Loss: 0.0002\nIter: 22, Epoch: 277, Loss: 0.0007\nIter: 23, Epoch: 277, Loss: 0.0007\nIter: 24, Epoch: 277, Loss: 0.0002\nIter: 25, Epoch: 277, Loss: 0.0003\nIter: 26, Epoch: 277, Loss: 0.0004\nIter: 27, Epoch: 277, Loss: 0.0002\nIter: 28, Epoch: 277, Loss: 0.0004\nIter: 29, Epoch: 277, Loss: 0.0006\nIter: 30, Epoch: 277, Loss: 0.0006\n","truncated":false}} +%--- +%[output:64dc73e9] +% data: {"dataType":"text","outputData":{"text":">> Epoch 277 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:2d31eba6] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 278, Loss: 0.0005\nIter: 2, Epoch: 278, Loss: 0.0005\nIter: 3, Epoch: 278, Loss: 0.0004\nIter: 4, Epoch: 278, Loss: 0.0004\nIter: 5, Epoch: 278, Loss: 0.0003\nIter: 6, Epoch: 278, Loss: 0.0003\nIter: 7, Epoch: 278, Loss: 0.0004\nIter: 8, Epoch: 278, Loss: 0.0009\nIter: 9, Epoch: 278, Loss: 0.0004\nIter: 10, Epoch: 278, Loss: 0.0007\nIter: 11, Epoch: 278, Loss: 0.0003\nIter: 12, Epoch: 278, Loss: 0.0003\nIter: 13, Epoch: 278, Loss: 0.0003\nIter: 14, Epoch: 278, Loss: 0.0005\nIter: 15, Epoch: 278, Loss: 0.0004\nIter: 16, Epoch: 278, Loss: 0.0005\nIter: 17, Epoch: 278, Loss: 0.0004\nIter: 18, Epoch: 278, Loss: 0.0004\nIter: 19, Epoch: 278, Loss: 0.0003\nIter: 20, Epoch: 278, Loss: 0.0004\nIter: 21, Epoch: 278, Loss: 0.0002\nIter: 22, Epoch: 278, Loss: 0.0007\nIter: 23, Epoch: 278, Loss: 0.0006\nIter: 24, Epoch: 278, Loss: 0.0002\nIter: 25, Epoch: 278, Loss: 0.0003\nIter: 26, Epoch: 278, Loss: 0.0004\nIter: 27, Epoch: 278, Loss: 0.0002\nIter: 28, Epoch: 278, Loss: 0.0004\nIter: 29, Epoch: 278, Loss: 0.0006\nIter: 30, Epoch: 278, Loss: 0.0006\n","truncated":false}} +%--- +%[output:2497201a] +% data: {"dataType":"text","outputData":{"text":">> Epoch 278 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:0be852e5] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 279, Loss: 0.0005\nIter: 2, Epoch: 279, Loss: 0.0005\nIter: 3, Epoch: 279, Loss: 0.0004\nIter: 4, Epoch: 279, Loss: 0.0004\nIter: 5, Epoch: 279, Loss: 0.0003\nIter: 6, Epoch: 279, Loss: 0.0003\nIter: 7, Epoch: 279, Loss: 0.0004\nIter: 8, Epoch: 279, Loss: 0.0009\nIter: 9, Epoch: 279, Loss: 0.0004\nIter: 10, Epoch: 279, Loss: 0.0007\nIter: 11, Epoch: 279, Loss: 0.0003\nIter: 12, Epoch: 279, Loss: 0.0003\nIter: 13, Epoch: 279, Loss: 0.0003\nIter: 14, Epoch: 279, Loss: 0.0005\nIter: 15, Epoch: 279, Loss: 0.0004\nIter: 16, Epoch: 279, Loss: 0.0005\nIter: 17, Epoch: 279, Loss: 0.0004\nIter: 18, Epoch: 279, Loss: 0.0004\nIter: 19, Epoch: 279, Loss: 0.0003\nIter: 20, Epoch: 279, Loss: 0.0004\nIter: 21, Epoch: 279, Loss: 0.0002\nIter: 22, Epoch: 279, Loss: 0.0007\nIter: 23, Epoch: 279, Loss: 0.0006\nIter: 24, Epoch: 279, Loss: 0.0002\nIter: 25, Epoch: 279, Loss: 0.0003\nIter: 26, Epoch: 279, Loss: 0.0004\nIter: 27, Epoch: 279, Loss: 0.0002\nIter: 28, Epoch: 279, Loss: 0.0004\nIter: 29, Epoch: 279, Loss: 0.0006\nIter: 30, Epoch: 279, Loss: 0.0006\n","truncated":false}} +%--- +%[output:78dd5006] +% data: {"dataType":"text","outputData":{"text":">> Epoch 279 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:5b3381ba] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 280, Loss: 0.0005\nIter: 2, Epoch: 280, Loss: 0.0005\nIter: 3, Epoch: 280, Loss: 0.0004\nIter: 4, Epoch: 280, Loss: 0.0004\nIter: 5, Epoch: 280, Loss: 0.0003\nIter: 6, Epoch: 280, Loss: 0.0003\nIter: 7, Epoch: 280, Loss: 0.0004\nIter: 8, Epoch: 280, Loss: 0.0009\nIter: 9, Epoch: 280, Loss: 0.0004\nIter: 10, Epoch: 280, Loss: 0.0007\nIter: 11, Epoch: 280, Loss: 0.0003\nIter: 12, Epoch: 280, Loss: 0.0003\nIter: 13, Epoch: 280, Loss: 0.0003\nIter: 14, Epoch: 280, Loss: 0.0005\nIter: 15, Epoch: 280, Loss: 0.0004\nIter: 16, Epoch: 280, Loss: 0.0005\nIter: 17, Epoch: 280, Loss: 0.0004\nIter: 18, Epoch: 280, Loss: 0.0004\nIter: 19, Epoch: 280, Loss: 0.0003\nIter: 20, Epoch: 280, Loss: 0.0004\nIter: 21, Epoch: 280, Loss: 0.0002\nIter: 22, Epoch: 280, Loss: 0.0007\nIter: 23, Epoch: 280, Loss: 0.0006\nIter: 24, Epoch: 280, Loss: 0.0002\nIter: 25, Epoch: 280, Loss: 0.0003\nIter: 26, Epoch: 280, Loss: 0.0004\nIter: 27, Epoch: 280, Loss: 0.0002\nIter: 28, Epoch: 280, Loss: 0.0004\nIter: 29, Epoch: 280, Loss: 0.0006\nIter: 30, Epoch: 280, Loss: 0.0006\n","truncated":false}} +%--- +%[output:86826ee9] +% data: {"dataType":"text","outputData":{"text":">> Epoch 280 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:7568c2ca] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 281, Loss: 0.0005\nIter: 2, Epoch: 281, Loss: 0.0005\nIter: 3, Epoch: 281, Loss: 0.0004\nIter: 4, Epoch: 281, Loss: 0.0004\nIter: 5, Epoch: 281, Loss: 0.0003\nIter: 6, Epoch: 281, Loss: 0.0003\nIter: 7, Epoch: 281, Loss: 0.0004\nIter: 8, Epoch: 281, Loss: 0.0009\nIter: 9, Epoch: 281, Loss: 0.0004\nIter: 10, Epoch: 281, Loss: 0.0007\nIter: 11, Epoch: 281, Loss: 0.0003\nIter: 12, Epoch: 281, Loss: 0.0003\nIter: 13, Epoch: 281, Loss: 0.0003\nIter: 14, Epoch: 281, Loss: 0.0005\nIter: 15, Epoch: 281, Loss: 0.0004\nIter: 16, Epoch: 281, Loss: 0.0005\nIter: 17, Epoch: 281, Loss: 0.0004\nIter: 18, Epoch: 281, Loss: 0.0004\nIter: 19, Epoch: 281, Loss: 0.0003\nIter: 20, Epoch: 281, Loss: 0.0004\nIter: 21, Epoch: 281, Loss: 0.0002\nIter: 22, Epoch: 281, Loss: 0.0006\nIter: 23, Epoch: 281, Loss: 0.0006\nIter: 24, Epoch: 281, Loss: 0.0002\nIter: 25, Epoch: 281, Loss: 0.0003\nIter: 26, Epoch: 281, Loss: 0.0004\nIter: 27, Epoch: 281, Loss: 0.0002\nIter: 28, Epoch: 281, Loss: 0.0004\nIter: 29, Epoch: 281, Loss: 0.0006\nIter: 30, Epoch: 281, Loss: 0.0006\n","truncated":false}} +%--- +%[output:084f3f73] +% data: {"dataType":"text","outputData":{"text":">> Epoch 281 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:62de3957] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 282, Loss: 0.0005\nIter: 2, Epoch: 282, Loss: 0.0005\nIter: 3, Epoch: 282, Loss: 0.0004\nIter: 4, Epoch: 282, Loss: 0.0004\nIter: 5, Epoch: 282, Loss: 0.0003\nIter: 6, Epoch: 282, Loss: 0.0003\nIter: 7, Epoch: 282, Loss: 0.0004\nIter: 8, Epoch: 282, Loss: 0.0009\nIter: 9, Epoch: 282, Loss: 0.0004\nIter: 10, Epoch: 282, Loss: 0.0007\nIter: 11, Epoch: 282, Loss: 0.0003\nIter: 12, Epoch: 282, Loss: 0.0003\nIter: 13, Epoch: 282, Loss: 0.0003\nIter: 14, Epoch: 282, Loss: 0.0005\nIter: 15, Epoch: 282, Loss: 0.0004\nIter: 16, Epoch: 282, Loss: 0.0005\nIter: 17, Epoch: 282, Loss: 0.0004\nIter: 18, Epoch: 282, Loss: 0.0004\nIter: 19, Epoch: 282, Loss: 0.0003\nIter: 20, Epoch: 282, Loss: 0.0004\nIter: 21, Epoch: 282, Loss: 0.0003\nIter: 22, Epoch: 282, Loss: 0.0006\nIter: 23, Epoch: 282, Loss: 0.0006\nIter: 24, Epoch: 282, Loss: 0.0002\nIter: 25, Epoch: 282, Loss: 0.0003\nIter: 26, Epoch: 282, Loss: 0.0004\nIter: 27, Epoch: 282, Loss: 0.0002\nIter: 28, Epoch: 282, Loss: 0.0004\nIter: 29, Epoch: 282, Loss: 0.0006\nIter: 30, Epoch: 282, Loss: 0.0006\n","truncated":false}} +%--- +%[output:4f5552f7] +% data: {"dataType":"text","outputData":{"text":">> Epoch 282 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:86cfe797] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 283, Loss: 0.0005\nIter: 2, Epoch: 283, Loss: 0.0005\nIter: 3, Epoch: 283, Loss: 0.0004\nIter: 4, Epoch: 283, Loss: 0.0004\nIter: 5, Epoch: 283, Loss: 0.0003\nIter: 6, Epoch: 283, Loss: 0.0003\nIter: 7, Epoch: 283, Loss: 0.0004\nIter: 8, Epoch: 283, Loss: 0.0009\nIter: 9, Epoch: 283, Loss: 0.0004\nIter: 10, Epoch: 283, Loss: 0.0007\nIter: 11, Epoch: 283, Loss: 0.0003\nIter: 12, Epoch: 283, Loss: 0.0003\nIter: 13, Epoch: 283, Loss: 0.0003\nIter: 14, Epoch: 283, Loss: 0.0005\nIter: 15, Epoch: 283, Loss: 0.0004\nIter: 16, Epoch: 283, Loss: 0.0005\nIter: 17, Epoch: 283, Loss: 0.0004\nIter: 18, Epoch: 283, Loss: 0.0004\nIter: 19, Epoch: 283, Loss: 0.0003\nIter: 20, Epoch: 283, Loss: 0.0004\nIter: 21, Epoch: 283, Loss: 0.0003\nIter: 22, Epoch: 283, Loss: 0.0006\nIter: 23, Epoch: 283, Loss: 0.0006\nIter: 24, Epoch: 283, Loss: 0.0002\nIter: 25, Epoch: 283, Loss: 0.0003\nIter: 26, Epoch: 283, Loss: 0.0004\nIter: 27, Epoch: 283, Loss: 0.0002\nIter: 28, Epoch: 283, Loss: 0.0004\nIter: 29, Epoch: 283, Loss: 0.0006\nIter: 30, Epoch: 283, Loss: 0.0006\n","truncated":false}} +%--- +%[output:503c0319] +% data: {"dataType":"text","outputData":{"text":">> Epoch 283 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:18ee7b6e] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 284, Loss: 0.0005\nIter: 2, Epoch: 284, Loss: 0.0005\nIter: 3, Epoch: 284, Loss: 0.0004\nIter: 4, Epoch: 284, Loss: 0.0004\nIter: 5, Epoch: 284, Loss: 0.0003\nIter: 6, Epoch: 284, Loss: 0.0003\nIter: 7, Epoch: 284, Loss: 0.0004\nIter: 8, Epoch: 284, Loss: 0.0008\nIter: 9, Epoch: 284, Loss: 0.0004\nIter: 10, Epoch: 284, Loss: 0.0007\nIter: 11, Epoch: 284, Loss: 0.0003\nIter: 12, Epoch: 284, Loss: 0.0003\nIter: 13, Epoch: 284, Loss: 0.0003\nIter: 14, Epoch: 284, Loss: 0.0005\nIter: 15, Epoch: 284, Loss: 0.0004\nIter: 16, Epoch: 284, Loss: 0.0005\nIter: 17, Epoch: 284, Loss: 0.0004\nIter: 18, Epoch: 284, Loss: 0.0004\nIter: 19, Epoch: 284, Loss: 0.0003\nIter: 20, Epoch: 284, Loss: 0.0004\nIter: 21, Epoch: 284, Loss: 0.0003\nIter: 22, Epoch: 284, Loss: 0.0006\nIter: 23, Epoch: 284, Loss: 0.0006\nIter: 24, Epoch: 284, Loss: 0.0002\nIter: 25, Epoch: 284, Loss: 0.0003\nIter: 26, Epoch: 284, Loss: 0.0004\nIter: 27, Epoch: 284, Loss: 0.0002\nIter: 28, Epoch: 284, Loss: 0.0004\nIter: 29, Epoch: 284, Loss: 0.0006\nIter: 30, Epoch: 284, Loss: 0.0006\n","truncated":false}} +%--- +%[output:579ebf63] +% data: {"dataType":"text","outputData":{"text":">> Epoch 284 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:5c3bffc5] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 285, Loss: 0.0005\nIter: 2, Epoch: 285, Loss: 0.0005\nIter: 3, Epoch: 285, Loss: 0.0004\nIter: 4, Epoch: 285, Loss: 0.0004\nIter: 5, Epoch: 285, Loss: 0.0003\nIter: 6, Epoch: 285, Loss: 0.0003\nIter: 7, Epoch: 285, Loss: 0.0004\nIter: 8, Epoch: 285, Loss: 0.0008\nIter: 9, Epoch: 285, Loss: 0.0004\nIter: 10, Epoch: 285, Loss: 0.0007\nIter: 11, Epoch: 285, Loss: 0.0003\nIter: 12, Epoch: 285, Loss: 0.0003\nIter: 13, Epoch: 285, Loss: 0.0003\nIter: 14, Epoch: 285, Loss: 0.0005\nIter: 15, Epoch: 285, Loss: 0.0004\nIter: 16, Epoch: 285, Loss: 0.0005\nIter: 17, Epoch: 285, Loss: 0.0004\nIter: 18, Epoch: 285, Loss: 0.0004\nIter: 19, Epoch: 285, Loss: 0.0003\nIter: 20, Epoch: 285, Loss: 0.0004\nIter: 21, Epoch: 285, Loss: 0.0003\nIter: 22, Epoch: 285, Loss: 0.0006\nIter: 23, Epoch: 285, Loss: 0.0006\nIter: 24, Epoch: 285, Loss: 0.0002\nIter: 25, Epoch: 285, Loss: 0.0003\nIter: 26, Epoch: 285, Loss: 0.0004\nIter: 27, Epoch: 285, Loss: 0.0002\nIter: 28, Epoch: 285, Loss: 0.0004\nIter: 29, Epoch: 285, Loss: 0.0006\nIter: 30, Epoch: 285, Loss: 0.0006\n","truncated":false}} +%--- +%[output:0c7235ac] +% data: {"dataType":"text","outputData":{"text":">> Epoch 285 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:339338b8] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 286, Loss: 0.0005\nIter: 2, Epoch: 286, Loss: 0.0005\nIter: 3, Epoch: 286, Loss: 0.0004\nIter: 4, Epoch: 286, Loss: 0.0004\nIter: 5, Epoch: 286, Loss: 0.0003\nIter: 6, Epoch: 286, Loss: 0.0003\nIter: 7, Epoch: 286, Loss: 0.0004\nIter: 8, Epoch: 286, Loss: 0.0008\nIter: 9, Epoch: 286, Loss: 0.0005\nIter: 10, Epoch: 286, Loss: 0.0007\nIter: 11, Epoch: 286, Loss: 0.0003\nIter: 12, Epoch: 286, Loss: 0.0003\nIter: 13, Epoch: 286, Loss: 0.0003\nIter: 14, Epoch: 286, Loss: 0.0005\nIter: 15, Epoch: 286, Loss: 0.0004\nIter: 16, Epoch: 286, Loss: 0.0005\nIter: 17, Epoch: 286, Loss: 0.0004\nIter: 18, Epoch: 286, Loss: 0.0005\nIter: 19, Epoch: 286, Loss: 0.0003\nIter: 20, Epoch: 286, Loss: 0.0004\nIter: 21, Epoch: 286, Loss: 0.0003\nIter: 22, Epoch: 286, Loss: 0.0006\nIter: 23, Epoch: 286, Loss: 0.0006\nIter: 24, Epoch: 286, Loss: 0.0002\nIter: 25, Epoch: 286, Loss: 0.0003\nIter: 26, Epoch: 286, Loss: 0.0004\nIter: 27, Epoch: 286, Loss: 0.0002\nIter: 28, Epoch: 286, Loss: 0.0004\nIter: 29, Epoch: 286, Loss: 0.0005\nIter: 30, Epoch: 286, Loss: 0.0006\n","truncated":false}} +%--- +%[output:2f3c8b96] +% data: {"dataType":"text","outputData":{"text":">> Epoch 286 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:9f25a6ad] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 287, Loss: 0.0005\nIter: 2, Epoch: 287, Loss: 0.0005\nIter: 3, Epoch: 287, Loss: 0.0004\nIter: 4, Epoch: 287, Loss: 0.0004\nIter: 5, Epoch: 287, Loss: 0.0003\nIter: 6, Epoch: 287, Loss: 0.0003\nIter: 7, Epoch: 287, Loss: 0.0004\nIter: 8, Epoch: 287, Loss: 0.0008\nIter: 9, Epoch: 287, Loss: 0.0005\nIter: 10, Epoch: 287, Loss: 0.0007\nIter: 11, Epoch: 287, Loss: 0.0003\nIter: 12, Epoch: 287, Loss: 0.0003\nIter: 13, Epoch: 287, Loss: 0.0003\nIter: 14, Epoch: 287, Loss: 0.0005\nIter: 15, Epoch: 287, Loss: 0.0004\nIter: 16, Epoch: 287, Loss: 0.0005\nIter: 17, Epoch: 287, Loss: 0.0004\nIter: 18, Epoch: 287, Loss: 0.0005\nIter: 19, Epoch: 287, Loss: 0.0003\nIter: 20, Epoch: 287, Loss: 0.0004\nIter: 21, Epoch: 287, Loss: 0.0003\nIter: 22, Epoch: 287, Loss: 0.0006\nIter: 23, Epoch: 287, Loss: 0.0006\nIter: 24, Epoch: 287, Loss: 0.0002\nIter: 25, Epoch: 287, Loss: 0.0003\nIter: 26, Epoch: 287, Loss: 0.0005\nIter: 27, Epoch: 287, Loss: 0.0002\nIter: 28, Epoch: 287, Loss: 0.0004\nIter: 29, Epoch: 287, Loss: 0.0005\nIter: 30, Epoch: 287, Loss: 0.0006\n","truncated":false}} +%--- +%[output:0bb4a76a] +% data: {"dataType":"text","outputData":{"text":">> Epoch 287 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:22d78126] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 288, Loss: 0.0005\nIter: 2, Epoch: 288, Loss: 0.0005\nIter: 3, Epoch: 288, Loss: 0.0004\nIter: 4, Epoch: 288, Loss: 0.0004\nIter: 5, Epoch: 288, Loss: 0.0003\nIter: 6, Epoch: 288, Loss: 0.0003\nIter: 7, Epoch: 288, Loss: 0.0004\nIter: 8, Epoch: 288, Loss: 0.0008\nIter: 9, Epoch: 288, Loss: 0.0005\nIter: 10, Epoch: 288, Loss: 0.0006\nIter: 11, Epoch: 288, Loss: 0.0003\nIter: 12, Epoch: 288, Loss: 0.0003\nIter: 13, Epoch: 288, Loss: 0.0003\nIter: 14, Epoch: 288, Loss: 0.0005\nIter: 15, Epoch: 288, Loss: 0.0004\nIter: 16, Epoch: 288, Loss: 0.0005\nIter: 17, Epoch: 288, Loss: 0.0004\nIter: 18, Epoch: 288, Loss: 0.0005\nIter: 19, Epoch: 288, Loss: 0.0003\nIter: 20, Epoch: 288, Loss: 0.0004\nIter: 21, Epoch: 288, Loss: 0.0003\nIter: 22, Epoch: 288, Loss: 0.0006\nIter: 23, Epoch: 288, Loss: 0.0006\nIter: 24, Epoch: 288, Loss: 0.0002\nIter: 25, Epoch: 288, Loss: 0.0003\nIter: 26, Epoch: 288, Loss: 0.0005\nIter: 27, Epoch: 288, Loss: 0.0002\nIter: 28, Epoch: 288, Loss: 0.0004\nIter: 29, Epoch: 288, Loss: 0.0005\nIter: 30, Epoch: 288, Loss: 0.0006\n","truncated":false}} +%--- +%[output:473e8680] +% data: {"dataType":"text","outputData":{"text":">> Epoch 288 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:07d6ecf4] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 289, Loss: 0.0005\nIter: 2, Epoch: 289, Loss: 0.0005\nIter: 3, Epoch: 289, Loss: 0.0004\nIter: 4, Epoch: 289, Loss: 0.0004\nIter: 5, Epoch: 289, Loss: 0.0003\nIter: 6, Epoch: 289, Loss: 0.0003\nIter: 7, Epoch: 289, Loss: 0.0004\nIter: 8, Epoch: 289, Loss: 0.0008\nIter: 9, Epoch: 289, Loss: 0.0005\nIter: 10, Epoch: 289, Loss: 0.0006\nIter: 11, Epoch: 289, Loss: 0.0003\nIter: 12, Epoch: 289, Loss: 0.0003\nIter: 13, Epoch: 289, Loss: 0.0003\nIter: 14, Epoch: 289, Loss: 0.0005\nIter: 15, Epoch: 289, Loss: 0.0004\nIter: 16, Epoch: 289, Loss: 0.0005\nIter: 17, Epoch: 289, Loss: 0.0004\nIter: 18, Epoch: 289, Loss: 0.0004\nIter: 19, Epoch: 289, Loss: 0.0003\nIter: 20, Epoch: 289, Loss: 0.0004\nIter: 21, Epoch: 289, Loss: 0.0003\nIter: 22, Epoch: 289, Loss: 0.0006\nIter: 23, Epoch: 289, Loss: 0.0006\nIter: 24, Epoch: 289, Loss: 0.0002\nIter: 25, Epoch: 289, Loss: 0.0003\nIter: 26, Epoch: 289, Loss: 0.0004\nIter: 27, Epoch: 289, Loss: 0.0002\nIter: 28, Epoch: 289, Loss: 0.0004\nIter: 29, Epoch: 289, Loss: 0.0005\nIter: 30, Epoch: 289, Loss: 0.0006\n","truncated":false}} +%--- +%[output:4a6fe78b] +% data: {"dataType":"text","outputData":{"text":">> Epoch 289 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:55ec4c8a] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 290, Loss: 0.0005\nIter: 2, Epoch: 290, Loss: 0.0005\nIter: 3, Epoch: 290, Loss: 0.0004\nIter: 4, Epoch: 290, Loss: 0.0004\nIter: 5, Epoch: 290, Loss: 0.0003\nIter: 6, Epoch: 290, Loss: 0.0003\nIter: 7, Epoch: 290, Loss: 0.0004\nIter: 8, Epoch: 290, Loss: 0.0008\nIter: 9, Epoch: 290, Loss: 0.0005\nIter: 10, Epoch: 290, Loss: 0.0006\nIter: 11, Epoch: 290, Loss: 0.0003\nIter: 12, Epoch: 290, Loss: 0.0003\nIter: 13, Epoch: 290, Loss: 0.0003\nIter: 14, Epoch: 290, Loss: 0.0005\nIter: 15, Epoch: 290, Loss: 0.0004\nIter: 16, Epoch: 290, Loss: 0.0005\nIter: 17, Epoch: 290, Loss: 0.0004\nIter: 18, Epoch: 290, Loss: 0.0004\nIter: 19, Epoch: 290, Loss: 0.0003\nIter: 20, Epoch: 290, Loss: 0.0004\nIter: 21, Epoch: 290, Loss: 0.0003\nIter: 22, Epoch: 290, Loss: 0.0006\nIter: 23, Epoch: 290, Loss: 0.0006\nIter: 24, Epoch: 290, Loss: 0.0002\nIter: 25, Epoch: 290, Loss: 0.0003\nIter: 26, Epoch: 290, Loss: 0.0005\nIter: 27, Epoch: 290, Loss: 0.0002\nIter: 28, Epoch: 290, Loss: 0.0004\nIter: 29, Epoch: 290, Loss: 0.0005\nIter: 30, Epoch: 290, Loss: 0.0006\n","truncated":false}} +%--- +%[output:45423fc0] +% data: {"dataType":"text","outputData":{"text":">> Epoch 290 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:2ba25360] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 291, Loss: 0.0005\nIter: 2, Epoch: 291, Loss: 0.0005\nIter: 3, Epoch: 291, Loss: 0.0004\nIter: 4, Epoch: 291, Loss: 0.0004\nIter: 5, Epoch: 291, Loss: 0.0003\nIter: 6, Epoch: 291, Loss: 0.0003\nIter: 7, Epoch: 291, Loss: 0.0005\nIter: 8, Epoch: 291, Loss: 0.0008\nIter: 9, Epoch: 291, Loss: 0.0005\nIter: 10, Epoch: 291, Loss: 0.0006\nIter: 11, Epoch: 291, Loss: 0.0003\nIter: 12, Epoch: 291, Loss: 0.0003\nIter: 13, Epoch: 291, Loss: 0.0003\nIter: 14, Epoch: 291, Loss: 0.0005\nIter: 15, Epoch: 291, Loss: 0.0004\nIter: 16, Epoch: 291, Loss: 0.0004\nIter: 17, Epoch: 291, Loss: 0.0004\nIter: 18, Epoch: 291, Loss: 0.0004\nIter: 19, Epoch: 291, Loss: 0.0003\nIter: 20, Epoch: 291, Loss: 0.0004\nIter: 21, Epoch: 291, Loss: 0.0003\nIter: 22, Epoch: 291, Loss: 0.0006\nIter: 23, Epoch: 291, Loss: 0.0006\nIter: 24, Epoch: 291, Loss: 0.0002\nIter: 25, Epoch: 291, Loss: 0.0003\nIter: 26, Epoch: 291, Loss: 0.0005\nIter: 27, Epoch: 291, Loss: 0.0002\nIter: 28, Epoch: 291, Loss: 0.0004\nIter: 29, Epoch: 291, Loss: 0.0005\nIter: 30, Epoch: 291, Loss: 0.0006\n","truncated":false}} +%--- +%[output:437d8fd7] +% data: {"dataType":"text","outputData":{"text":">> Epoch 291 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:02df5bbb] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 292, Loss: 0.0005\nIter: 2, Epoch: 292, Loss: 0.0005\nIter: 3, Epoch: 292, Loss: 0.0004\nIter: 4, Epoch: 292, Loss: 0.0004\nIter: 5, Epoch: 292, Loss: 0.0003\nIter: 6, Epoch: 292, Loss: 0.0003\nIter: 7, Epoch: 292, Loss: 0.0005\nIter: 8, Epoch: 292, Loss: 0.0008\nIter: 9, Epoch: 292, Loss: 0.0005\nIter: 10, Epoch: 292, Loss: 0.0006\nIter: 11, Epoch: 292, Loss: 0.0003\nIter: 12, Epoch: 292, Loss: 0.0003\nIter: 13, Epoch: 292, Loss: 0.0003\nIter: 14, Epoch: 292, Loss: 0.0005\nIter: 15, Epoch: 292, Loss: 0.0004\nIter: 16, Epoch: 292, Loss: 0.0004\nIter: 17, Epoch: 292, Loss: 0.0004\nIter: 18, Epoch: 292, Loss: 0.0004\nIter: 19, Epoch: 292, Loss: 0.0003\nIter: 20, Epoch: 292, Loss: 0.0004\nIter: 21, Epoch: 292, Loss: 0.0003\nIter: 22, Epoch: 292, Loss: 0.0006\nIter: 23, Epoch: 292, Loss: 0.0006\nIter: 24, Epoch: 292, Loss: 0.0002\nIter: 25, Epoch: 292, Loss: 0.0003\nIter: 26, Epoch: 292, Loss: 0.0005\nIter: 27, Epoch: 292, Loss: 0.0002\nIter: 28, Epoch: 292, Loss: 0.0004\nIter: 29, Epoch: 292, Loss: 0.0005\nIter: 30, Epoch: 292, Loss: 0.0006\n","truncated":false}} +%--- +%[output:0adf8e1c] +% data: {"dataType":"text","outputData":{"text":">> Epoch 292 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:4e6fc9b0] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 293, Loss: 0.0005\nIter: 2, Epoch: 293, Loss: 0.0005\nIter: 3, Epoch: 293, Loss: 0.0004\nIter: 4, Epoch: 293, Loss: 0.0004\nIter: 5, Epoch: 293, Loss: 0.0003\nIter: 6, Epoch: 293, Loss: 0.0003\nIter: 7, Epoch: 293, Loss: 0.0005\nIter: 8, Epoch: 293, Loss: 0.0008\nIter: 9, Epoch: 293, Loss: 0.0005\nIter: 10, Epoch: 293, Loss: 0.0006\nIter: 11, Epoch: 293, Loss: 0.0003\nIter: 12, Epoch: 293, Loss: 0.0003\nIter: 13, Epoch: 293, Loss: 0.0003\nIter: 14, Epoch: 293, Loss: 0.0005\nIter: 15, Epoch: 293, Loss: 0.0004\nIter: 16, Epoch: 293, Loss: 0.0004\nIter: 17, Epoch: 293, Loss: 0.0004\nIter: 18, Epoch: 293, Loss: 0.0004\nIter: 19, Epoch: 293, Loss: 0.0003\nIter: 20, Epoch: 293, Loss: 0.0004\nIter: 21, Epoch: 293, Loss: 0.0003\nIter: 22, Epoch: 293, Loss: 0.0006\nIter: 23, Epoch: 293, Loss: 0.0006\nIter: 24, Epoch: 293, Loss: 0.0002\nIter: 25, Epoch: 293, Loss: 0.0003\nIter: 26, Epoch: 293, Loss: 0.0005\nIter: 27, Epoch: 293, Loss: 0.0002\nIter: 28, Epoch: 293, Loss: 0.0004\nIter: 29, Epoch: 293, Loss: 0.0005\nIter: 30, Epoch: 293, Loss: 0.0006\n","truncated":false}} +%--- +%[output:8b7c85e1] +% data: {"dataType":"text","outputData":{"text":">> Epoch 293 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:0f215664] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 294, Loss: 0.0005\nIter: 2, Epoch: 294, Loss: 0.0005\nIter: 3, Epoch: 294, Loss: 0.0004\nIter: 4, Epoch: 294, Loss: 0.0004\nIter: 5, Epoch: 294, Loss: 0.0003\nIter: 6, Epoch: 294, Loss: 0.0003\nIter: 7, Epoch: 294, Loss: 0.0005\nIter: 8, Epoch: 294, Loss: 0.0008\nIter: 9, Epoch: 294, Loss: 0.0005\nIter: 10, Epoch: 294, Loss: 0.0006\nIter: 11, Epoch: 294, Loss: 0.0003\nIter: 12, Epoch: 294, Loss: 0.0003\nIter: 13, Epoch: 294, Loss: 0.0003\nIter: 14, Epoch: 294, Loss: 0.0005\nIter: 15, Epoch: 294, Loss: 0.0004\nIter: 16, Epoch: 294, Loss: 0.0004\nIter: 17, Epoch: 294, Loss: 0.0004\nIter: 18, Epoch: 294, Loss: 0.0004\nIter: 19, Epoch: 294, Loss: 0.0003\nIter: 20, Epoch: 294, Loss: 0.0004\nIter: 21, Epoch: 294, Loss: 0.0003\nIter: 22, Epoch: 294, Loss: 0.0006\nIter: 23, Epoch: 294, Loss: 0.0006\nIter: 24, Epoch: 294, Loss: 0.0002\nIter: 25, Epoch: 294, Loss: 0.0003\nIter: 26, Epoch: 294, Loss: 0.0005\nIter: 27, Epoch: 294, Loss: 0.0002\nIter: 28, Epoch: 294, Loss: 0.0004\nIter: 29, Epoch: 294, Loss: 0.0005\nIter: 30, Epoch: 294, Loss: 0.0006\n","truncated":false}} +%--- +%[output:7e7c53a5] +% data: {"dataType":"text","outputData":{"text":">> Epoch 294 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:8628a909] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 295, Loss: 0.0005\nIter: 2, Epoch: 295, Loss: 0.0005\nIter: 3, Epoch: 295, Loss: 0.0004\nIter: 4, Epoch: 295, Loss: 0.0004\nIter: 5, Epoch: 295, Loss: 0.0003\nIter: 6, Epoch: 295, Loss: 0.0003\nIter: 7, Epoch: 295, Loss: 0.0005\nIter: 8, Epoch: 295, Loss: 0.0008\nIter: 9, Epoch: 295, Loss: 0.0005\nIter: 10, Epoch: 295, Loss: 0.0006\nIter: 11, Epoch: 295, Loss: 0.0003\nIter: 12, Epoch: 295, Loss: 0.0003\nIter: 13, Epoch: 295, Loss: 0.0003\nIter: 14, Epoch: 295, Loss: 0.0005\nIter: 15, Epoch: 295, Loss: 0.0004\nIter: 16, Epoch: 295, Loss: 0.0004\nIter: 17, Epoch: 295, Loss: 0.0004\nIter: 18, Epoch: 295, Loss: 0.0004\nIter: 19, Epoch: 295, Loss: 0.0003\nIter: 20, Epoch: 295, Loss: 0.0004\nIter: 21, Epoch: 295, Loss: 0.0003\nIter: 22, Epoch: 295, Loss: 0.0006\nIter: 23, Epoch: 295, Loss: 0.0006\nIter: 24, Epoch: 295, Loss: 0.0002\nIter: 25, Epoch: 295, Loss: 0.0003\nIter: 26, Epoch: 295, Loss: 0.0005\nIter: 27, Epoch: 295, Loss: 0.0002\nIter: 28, Epoch: 295, Loss: 0.0004\nIter: 29, Epoch: 295, Loss: 0.0005\nIter: 30, Epoch: 295, Loss: 0.0006\n","truncated":false}} +%--- +%[output:71e592c2] +% data: {"dataType":"text","outputData":{"text":">> Epoch 295 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:10eb2dc6] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 296, Loss: 0.0005\nIter: 2, Epoch: 296, Loss: 0.0005\nIter: 3, Epoch: 296, Loss: 0.0004\nIter: 4, Epoch: 296, Loss: 0.0004\nIter: 5, Epoch: 296, Loss: 0.0003\nIter: 6, Epoch: 296, Loss: 0.0003\nIter: 7, Epoch: 296, Loss: 0.0005\nIter: 8, Epoch: 296, Loss: 0.0008\nIter: 9, Epoch: 296, Loss: 0.0005\nIter: 10, Epoch: 296, Loss: 0.0006\nIter: 11, Epoch: 296, Loss: 0.0003\nIter: 12, Epoch: 296, Loss: 0.0003\nIter: 13, Epoch: 296, Loss: 0.0003\nIter: 14, Epoch: 296, Loss: 0.0005\nIter: 15, Epoch: 296, Loss: 0.0004\nIter: 16, Epoch: 296, Loss: 0.0004\nIter: 17, Epoch: 296, Loss: 0.0004\nIter: 18, Epoch: 296, Loss: 0.0004\nIter: 19, Epoch: 296, Loss: 0.0003\nIter: 20, Epoch: 296, Loss: 0.0004\nIter: 21, Epoch: 296, Loss: 0.0003\nIter: 22, Epoch: 296, Loss: 0.0006\nIter: 23, Epoch: 296, Loss: 0.0006\nIter: 24, Epoch: 296, Loss: 0.0002\nIter: 25, Epoch: 296, Loss: 0.0003\nIter: 26, Epoch: 296, Loss: 0.0005\nIter: 27, Epoch: 296, Loss: 0.0002\nIter: 28, Epoch: 296, Loss: 0.0004\nIter: 29, Epoch: 296, Loss: 0.0005\nIter: 30, Epoch: 296, Loss: 0.0006\n","truncated":false}} +%--- +%[output:9379e722] +% data: {"dataType":"text","outputData":{"text":">> Epoch 296 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:4553bef2] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 297, Loss: 0.0005\nIter: 2, Epoch: 297, Loss: 0.0005\nIter: 3, Epoch: 297, Loss: 0.0004\nIter: 4, Epoch: 297, Loss: 0.0004\nIter: 5, Epoch: 297, Loss: 0.0003\nIter: 6, Epoch: 297, Loss: 0.0003\nIter: 7, Epoch: 297, Loss: 0.0005\nIter: 8, Epoch: 297, Loss: 0.0008\nIter: 9, Epoch: 297, Loss: 0.0005\nIter: 10, Epoch: 297, Loss: 0.0006\nIter: 11, Epoch: 297, Loss: 0.0003\nIter: 12, Epoch: 297, Loss: 0.0003\nIter: 13, Epoch: 297, Loss: 0.0003\nIter: 14, Epoch: 297, Loss: 0.0005\nIter: 15, Epoch: 297, Loss: 0.0004\nIter: 16, Epoch: 297, Loss: 0.0004\nIter: 17, Epoch: 297, Loss: 0.0004\nIter: 18, Epoch: 297, Loss: 0.0004\nIter: 19, Epoch: 297, Loss: 0.0003\nIter: 20, Epoch: 297, Loss: 0.0004\nIter: 21, Epoch: 297, Loss: 0.0003\nIter: 22, Epoch: 297, Loss: 0.0006\nIter: 23, Epoch: 297, Loss: 0.0006\nIter: 24, Epoch: 297, Loss: 0.0002\nIter: 25, Epoch: 297, Loss: 0.0003\nIter: 26, Epoch: 297, Loss: 0.0005\nIter: 27, Epoch: 297, Loss: 0.0002\nIter: 28, Epoch: 297, Loss: 0.0004\nIter: 29, Epoch: 297, Loss: 0.0005\nIter: 30, Epoch: 297, Loss: 0.0006\n","truncated":false}} +%--- +%[output:59195fd4] +% data: {"dataType":"text","outputData":{"text":">> Epoch 297 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:1031648f] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 298, Loss: 0.0005\nIter: 2, Epoch: 298, Loss: 0.0005\nIter: 3, Epoch: 298, Loss: 0.0004\nIter: 4, Epoch: 298, Loss: 0.0004\nIter: 5, Epoch: 298, Loss: 0.0003\nIter: 6, Epoch: 298, Loss: 0.0003\nIter: 7, Epoch: 298, Loss: 0.0005\nIter: 8, Epoch: 298, Loss: 0.0008\nIter: 9, Epoch: 298, Loss: 0.0005\nIter: 10, Epoch: 298, Loss: 0.0006\nIter: 11, Epoch: 298, Loss: 0.0003\nIter: 12, Epoch: 298, Loss: 0.0003\nIter: 13, Epoch: 298, Loss: 0.0003\nIter: 14, Epoch: 298, Loss: 0.0005\nIter: 15, Epoch: 298, Loss: 0.0004\nIter: 16, Epoch: 298, Loss: 0.0004\nIter: 17, Epoch: 298, Loss: 0.0004\nIter: 18, Epoch: 298, Loss: 0.0004\nIter: 19, Epoch: 298, Loss: 0.0003\nIter: 20, Epoch: 298, Loss: 0.0004\nIter: 21, Epoch: 298, Loss: 0.0003\nIter: 22, Epoch: 298, Loss: 0.0006\nIter: 23, Epoch: 298, Loss: 0.0006\nIter: 24, Epoch: 298, Loss: 0.0002\nIter: 25, Epoch: 298, Loss: 0.0003\nIter: 26, Epoch: 298, Loss: 0.0005\nIter: 27, Epoch: 298, Loss: 0.0002\nIter: 28, Epoch: 298, Loss: 0.0004\nIter: 29, Epoch: 298, Loss: 0.0005\nIter: 30, Epoch: 298, Loss: 0.0006\n","truncated":false}} +%--- +%[output:821a7639] +% data: {"dataType":"text","outputData":{"text":">> Epoch 298 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:3b096af5] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 299, Loss: 0.0005\nIter: 2, Epoch: 299, Loss: 0.0005\nIter: 3, Epoch: 299, Loss: 0.0004\nIter: 4, Epoch: 299, Loss: 0.0004\nIter: 5, Epoch: 299, Loss: 0.0003\nIter: 6, Epoch: 299, Loss: 0.0003\nIter: 7, Epoch: 299, Loss: 0.0005\nIter: 8, Epoch: 299, Loss: 0.0008\nIter: 9, Epoch: 299, Loss: 0.0005\nIter: 10, Epoch: 299, Loss: 0.0006\nIter: 11, Epoch: 299, Loss: 0.0003\nIter: 12, Epoch: 299, Loss: 0.0003\nIter: 13, Epoch: 299, Loss: 0.0003\nIter: 14, Epoch: 299, Loss: 0.0005\nIter: 15, Epoch: 299, Loss: 0.0004\nIter: 16, Epoch: 299, Loss: 0.0004\nIter: 17, Epoch: 299, Loss: 0.0004\nIter: 18, Epoch: 299, Loss: 0.0004\nIter: 19, Epoch: 299, Loss: 0.0003\nIter: 20, Epoch: 299, Loss: 0.0004\nIter: 21, Epoch: 299, Loss: 0.0003\nIter: 22, Epoch: 299, Loss: 0.0006\nIter: 23, Epoch: 299, Loss: 0.0006\nIter: 24, Epoch: 299, Loss: 0.0002\nIter: 25, Epoch: 299, Loss: 0.0003\nIter: 26, Epoch: 299, Loss: 0.0005\nIter: 27, Epoch: 299, Loss: 0.0002\nIter: 28, Epoch: 299, Loss: 0.0004\nIter: 29, Epoch: 299, Loss: 0.0005\nIter: 30, Epoch: 299, Loss: 0.0006\n","truncated":false}} +%--- +%[output:1a227dc6] +% data: {"dataType":"text","outputData":{"text":">> Epoch 299 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:1451c82d] +% data: {"dataType":"text","outputData":{"text":"Iter: 1, Epoch: 300, Loss: 0.0005\nIter: 2, Epoch: 300, Loss: 0.0005\nIter: 3, Epoch: 300, Loss: 0.0004\nIter: 4, Epoch: 300, Loss: 0.0004\nIter: 5, Epoch: 300, Loss: 0.0003\nIter: 6, Epoch: 300, Loss: 0.0003\nIter: 7, Epoch: 300, Loss: 0.0005\nIter: 8, Epoch: 300, Loss: 0.0008\nIter: 9, Epoch: 300, Loss: 0.0005\nIter: 10, Epoch: 300, Loss: 0.0006\nIter: 11, Epoch: 300, Loss: 0.0003\nIter: 12, Epoch: 300, Loss: 0.0003\nIter: 13, Epoch: 300, Loss: 0.0003\nIter: 14, Epoch: 300, Loss: 0.0005\nIter: 15, Epoch: 300, Loss: 0.0004\nIter: 16, Epoch: 300, Loss: 0.0004\nIter: 17, Epoch: 300, Loss: 0.0004\nIter: 18, Epoch: 300, Loss: 0.0004\nIter: 19, Epoch: 300, Loss: 0.0003\nIter: 20, Epoch: 300, Loss: 0.0004\nIter: 21, Epoch: 300, Loss: 0.0003\nIter: 22, Epoch: 300, Loss: 0.0006\nIter: 23, Epoch: 300, Loss: 0.0006\nIter: 24, Epoch: 300, Loss: 0.0002\nIter: 25, Epoch: 300, Loss: 0.0003\nIter: 26, Epoch: 300, Loss: 0.0005\nIter: 27, Epoch: 300, Loss: 0.0002\nIter: 28, Epoch: 300, Loss: 0.0004\nIter: 29, Epoch: 300, Loss: 0.0005\nIter: 30, Epoch: 300, Loss: 0.0006\n","truncated":false}} +%--- +%[output:057ff863] +% data: {"dataType":"text","outputData":{"text":">> Epoch 300 validation loss: 0.0005\n","truncated":false}} +%--- +%[output:8efe7c47] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,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","height":640,"width":1024}} +%--- +%[output:49191cf1] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,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","height":337,"width":560}} +%--- +%[output:66a349c3] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,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","height":337,"width":560}} +%--- +%[output:005a8275] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,iVBORw0KGgoAAAANSUhEUgAAAjAAAAFRCAYAAABqsZcNAAAQAElEQVR4Aey9CbxV1Z3n+2eQ8aI4gQyac1U0lShIRVMklHBNNLGeSZn4YnVLrOJKuizz2ryknwndVpUBmuryU5bWJ0mR59CJQMc2lTJNHJIUcQgXEiIOCXFKq6j3iAqCqCjDBWTo\/3dz983mcM65Z1hn\/sH+3733Gv\/ru\/577\/9aezgDD+ifCIiACIiACIiACDQYgYGmfyIgAiIgAiIgAkUSUPJaE5ADU+seUP0iIAIiIAIiIAJFE5ADUzQyZRABERCB2hOQBiLQ6gTkwLS6Baj9IiACIiACItCABOTANGCnSWURqD0BaSACIiACtSUgB6a2\/FW7CIiACIiACIhACQTkwJQATVlqT0AaiIAIiIAItDYBOTCt3f9qvQiIgAiIgAg0JAE5MCV1mzKJgAiIgAiIgAjUkoAcmFrSV90iIAIiIAIi0EoEArZVDkxAmCpKBERABERABESgOgTkwFSHs2oRAREQARGoPQFp0EQE5MA0UWeqKSIgAiIgAiLQKgTkwLRKT6udIiACtScgDURABIIRkAMTDKUKEgEREAEREAERqBYBOTDVIq16RKD2BKSBCIiACDQNATkwTdOVaogIiIAIiIAItA4BOTCt09e1b6k0EAEREAEREIFABOTABAKpYkRABERABERABKpHoJUcmOpRVU0iIAIiIAIiIAIVJSAHpqJ4VbgIiIAIiIAINDqB+tRfDkx99ou0EgEREAEREAERyENADkweOIoSAREQARGoPQFpIALZCMiByUZFYSIgAiIgAiIgAnVNQA5MXXePlBMBEag9AWkgAiJQjwTkwNRjr0gnERABERABERCBvATkwOTFo0gRqD0BaSACIiACInA4ATkwhzNRiAiIgAiIgAiIQJ0TkANT5x1Ue\/WkgQiIgAiIgAjUHwE5MPXXJ9JIBERABERABESgHwJ178D0o7+iRUAEREAEREAEWpCAHJgW7HQ1WQREQAREoOkJNH0D5cA0fRergSIgAiIgAiLQfATkwDRfn6pFIiACIlB7AtJABCpMQA5MhQGreBEQAREQAREQgfAE5MCEZ6oSRUAEak9AGoiACDQ5ATkwTd7Bap4IiIAIiIAINCMBOTDN2KtqU+0JSAMREAEREIGKEpADU1G8KlwEREAEREAERKASBOTAVIJq7cuUBiIgAiIgAiLQ1ATkwDR196pxIiACIiACItCcBCrjwDQnK7VKBERABERABESgTgjIgamTjpAaIiACIiACIiAChROQA1M4K6UUAREQAREQARGoEwJyYOqkI6SGCIiACNSegDQQgcYhIAemcfpKmoqACIiACIiACPQSkAPTC0IrERCB2hOQBiIgAiJQKAE5MIWSUjoREAEREAEREIG6ISAHpm66QorUnoA0EAEREAERaBQCcmAapaekpwiIgAiIgAiIQB8BOTB9KGq\/IQ1EQAREQAREQAQKIyAHpjBOSiUCIiACIiACIlBHBBIOTB1pJVVEQAREQAREQAREIA8BOTB54ChKBERABERABPoloAQ1ISAHpibYVakIiIAIiIAIiEA5BOTAlENPeUVABESg9gSkgQi0JAE5MC3Z7Wq0CIiACIiACDQ2ATkwjd1\/0l4Eak9AGoiACIhADQjIgakBdFUpAiIgAiIgAiJQHgE5MOXxU+7aE5AGIiACIiACLUhADkwLdrqaLAIiIAIiIAKNTkAOTLk9qPwiIAIiIAIiIAJVJyAHpurIVaEIiIAIiIAIiEC5BOTAlEtQ+UVABERABERABKpOQA5M1ZGrQhEQAREQgdoTkAaNTkAOTKP3oPQXAREQAREQgRYkIAemBTtdTRYBEag9AWkgAiJQHgE5MOXxU24REAEREAEREIEaEJADUwPoqlIEak9AGoiACIhAYxOQA9PY\/SftRUAEREAERKAlCciBaclur32jpYEIiIAIiIAIlENADkw59JRXBERABERABESgJgRa1IGpCWtVKgIiIAIiIAIiEIiAHJhAIFWMCIiACIiACDQ9gTpqoByYOuoMqSICIiACIiACIlAYATkwhXFSKhEQAREQgdoTkAYi0EdADkwfCm2IgAiIgAiIgAg0CgE5MI3SU9JTBESg9gSkgQiIQN0QkANTN10hRURABERABERABAolIAemUFJKJwK1JyANREAEREAEegnIgekFoZUIiIAIiIAIiEDjEJAD0zh9VXtNpYEIiIAIiIAI1AkBOTB10hFSQwREQAREQAREoHACjeTAFN4qpRQBERABERABEWhqAnJgmrp71TgREAEREAERaE4CcmCas1\/VKhEQAREQARFoagJyYJq6e9U4ERABEag9AWkgApUgIAemElRVpgiIgAiIgAiIQEUJyIGpKF4VLgIiUHsC0kAERKAZCciBacZeVZtEQAREQAREoMkJyIFp8g5W82pPQBqIgAiIgAiEJyAHJjxTlSgCIiACIiACIlBhAnJgKgy49sVLAxEQAREQARFoPgJyYJqvT9UiERABERABEWh6AhV3YJqeoBooAiIgAiIgAiJQdQJyYKqOXBWKgAiIgAiIQL8ElKAfAnJg+gGkaBEQAREQAREQgfojIAem\/vpEGomACIhA7QlIAxGocwJyYOq8g6SeCIiACIiACIjA4QTkwPQy2bx5s82cOdNOPvnkQ+SrX\/1qlOKWW245JDxOF8dHifzPU089ZVOmTImEbQ\/Ku2TWm1lerszbt2+3WbNmRcJ2rnRxOGkom\/oIY5v2xvuE1VLQB6ask3qgd752kp58SSE9+ZLllLJNGZSFsE1\/0rfYQiHlzZs3z8hDWvSsBm\/qQ8ckD7bjNqBLPnnwwQcjO2edLx1xMKFd2BDb1IGwTXyZYpQLM+rIVRZxtA8hLXlypW2U8Mw+zGdvodpPHTBEKDPJin3CkRCMk+VRZiyF2FzMBn2TOubarvQxiB6x\/sk1bcylk8LDEZADk8HykksusZdeeqlPbrzxxr4Uo0aNsnvuuacvjnTJeBKuXr3atm3bFgnbhOWTG264wSZMmGBPPvmkzZ0715YtW2aFHMj5yswWN3\/+fHvsscf6otB75cqVNmbMmL6wWm1w0Yl1Y81+MbqceOKJtmbNmqhfbrvttmj7jjvuKKaIgtKeeeaZ9sQTT9hVV13Vb3pObHfffXdfumrzxpawTwSbfeaZZyw0k6RNtbW12Z133hkJ230Nr+AGxwnHC33O8cNxxPFUwSorXjTO3\/XXX28f\/OAHo3MCbbv55pv7HOGkAqHaj1NAHdSFrTzwwAOG\/VJXqDooKynJYxYb5by7cOHCyGlNpitnmzZU4xgs5LpQTjuUNzcBOTC52RQdw8ln1apV0cmHExDbhOUqiLgNGzYYF5fu7u7owsjBfP755\/dl4SCMPft8ox9ONKRjTWbyMRLn5MQ2J\/pXXnnFLrjgguhkyAghWR5OA\/uUgZCHcpA4LaMZ4uJyiQshXHzQjZMYa\/ZLLXfatGmGxOxj3ZkZQHf4wD3eJyzZ1j4OPhN32WWXRY5orAssaXsyPeVTBkKZlE0dXEhxZC+++OLIISUdfCmf8lizTz4ks0zicvEmLXqgD2X1J2PHjrXRo0fbCy+80JeUMqgXifXui0xsUAd1kQ4hH9Gskzb16KOPRrOBlLVp06a+bXiQPrP95Kc8hDxxOtIWKhwn8fGC0zR+\/PjISYdtZhn0CXUhsI3TxO37q7\/6q2jWFF1+8pOfRLNQ6Ex61pQXpyUMoUzC0Z18n\/70pw2BF2mJK1ZoR9IRpO8oI9tgqL\/2F8o4Lpu62tvbo\/NXfPz0Vwe6hZBTTz3Vtm7datgO5dE\/9BOckZg1cUnJTBf3FenzHYOko3zyUx6skv1GfupFkulIW4zE5VAfZbFGKBObIYw0lIkO7CPEx7phS+iWtFFsjjytLnJgAloATgjOyEUXXWQI24T1V0XyQpdMi0EzMmJUxEWdEeZXvvIVK9Z4mTHAOWDUw+iKmYRkPRwol156qZ1zzjl9sxgc\/PGBRVoci3PPPTea3eBiuHTpUoKDyPLlyw3dOjs7ozX7QQruLQTd58yZE7WNEzIzB6+99lrUFkadybayzYkU5n\/\/939v69ev7y3l8BX9w4wRsz+kp7+Z5aAOZkDikRn7ydzV5s1FgTZxkUAP+pV20nZ0hwVMiEsKdsZswE033RSxw4awR06omTb1gQ98oC\/ryJEjbcaMGUa5O3fujEbVcMK+mPErtP6+AgvYQCdsm2NkxIgRh+Qg7pprrolmOHMdRxyDXMhxHoYOHdqXHweJ2TPKuPzyy+2LX\/xixIL+pUzC48TYCjbDLF3mMcZFiwtTpsAizp9tzXGGbnHfZUtDGHok20+5hfQxeXFsOaZxYHCgcATjviM+lsw64vAQ66QO2N1X\/DxHX9Jf2VjHddJGBhr0E+lwqmk7xxz7uY7BCy+8sM9hoj4cNgadOHC0k74lP\/WjB\/qQLq63lDU6YkvkzTwncS7h2OI8kqtO7CC2UfqJclpd5MBkWAAHQHySwevFmOMkGBAj6jieNQdLHI9xsT19+nRD2I7D2M4UjPDaa681DjLirrzyymjkun379shJSR5UpOUizAUHIX0o4YDhgOKgpkxmMJDbb7890oMwdOQEx8WBA5qZo3IPaMrlYh5f3M4444zIiWKfcOKLFdggXEBhRn6co8mTJ7N52MWUcOI5gVIndTNLxQWIkxkntShjxh\/aTv\/EF2XSc+Hiwp6R9LDdcnlTB3VR52GF9wZwYsc+EWyWCy\/5iMZBpM20HYeCNmTrT\/hxQaffORY4NshfiGD\/OE20FUnaV6H1F1IPaei3q6++OpotS\/Y7cUh8DKITbSJN5gWaMOJIH0t8PLCfLIN9nDku+jgY7CPYCjbDdqZw4eIClilcaDPTxvtc1GAe91UcnrnO1v5qM87Uqb997IFzDPaJcNzddddd0S1tBn0MBuI+od8oD0ecdVLgSl8wY4HNJ+PybWP79B\/9mlkfYeSlXmwCPTLthfhY+rsukC5pS+wn+zQ+l8T2Q53ZzvXoQRz5JQcJyIE5yKHvLwdDfJLJvEhwEcdDjuNZxyegTCPkRIZBcpEjrq+CjA0uQhwwHMxEcfFlFM\/IlYOG0VCljZaLN23jQoUO2YSDPV98tjyFhGVe3DjQObkRXkh+0pAefpwIcQLpw\/hiTXxSOAlyYeXCQHrykR8GyXT9bcf901+6bPHUVWne8eiR9qFDPILHFnFWaDNxMIAFtkabSBsLF0YuDMw8YJOUGcf1t47tnwspEp+wi6m\/vzqIpzxGx3F70JXwpMA7eZHhQocNYAvJdPm2KaOcY6DYGRgGRuiJTtddd110YWc7U7K1n7BC+zizvGz7lNcf42z58oVhD5zrmAUkHYMiBkds0y\/0F+3HPnHA2acPiI8FvbgNgx3DKC4rjs+3jh13zs\/3339\/lBSHhQ3qoT7qpX70yGcvHMu5rguU159w3HH8VeNc358ujRaf3YFptFbUgb4cjLHg3SPxPut8KuKgkeOhxgAAEABJREFUMNLlIOBg4ADiYOag5kTEgZovf7lxXNw4YDlxlFtWMflpF7M85MHx4GTBmn3CiWe7P4lPhjiUCKOyXHlwwrgQ4eSQNpZknkKYl9M\/1eKNXcUzfPEDkoRxosxktjLLA904kTgG3ELC0c7FNFs49TBi5LYGz8cwy8NFg\/BC689WbmYYt744vmjPN77xDaP8zDTw5rji+Ir7O3Nwkpknc58y8l3EMtNn7mNfcd3JdTwASqZn1pdbGIThNGZLQxySrf0wyMUYu+Wiz7GG4Cgl28Yxh\/1z7iFtrjoIDyG0jWORPsRJpkyOUfqLtidZZQ5K4pkT0lEOeYsRBkvM9GAXDDZxuskPD+onPK6\/WHuhnEIFzvCGO\/wLzad0ZnJgAlkBo8xMo+cAIIy4bNVwomJqntFZMp4DiJMQFwAOMA5UDJuLOqMNJJme7figx\/khLSMLwgsRnC0uALGenEwQpjHRo5AySklDu2gfJ7D4RMGafcKJL6XcfHm4iHIxZcqaGQZO4PGJPI6L66Z+trOVB5dk\/1AWsxWZfZktb7m8ubWA3WA\/2cpPhuF4cEsMR4SZFuI4cbOPg5JP76RNke773\/8+2QuWeET76quvGnXGGdkupP44fa41fUebOMYWLVqUc5Yi1oOZTo4NLuAI27nKzgxPlkEc9eLQzJ49m91ggk48d8SAguMgedEmDr0RtvO1Pxdj7JbBEscZwoU\/bhsDmHg2ANsmbb46QjUahvQh9oWd4UjgUHAOo5257B37ZDASn\/M4PxajE8ch+bHPuL3kj3mUai+UUYzAmfo513DOoc20ZVrvCwnFlNVqaeXAFNHjnFTiaUUueggnk3Q6Hb39wEHHwRcXyTZh8cUyDo\/XXFwY3XIypCzK5mITn7RY8+wC4Rxs5GOaFINnOymURV6mOzkAOSEk4+MTGgcFJ6VkHBdu7j+jJ3owC0I9nNyS6UJvc4KAKboly2afcOKT4aG258+fH726DgvamhzBEUefwfyv\/\/qv7aSTTspZLf0Dc9JSFqMo8pOBPmBNHCdgtmOpNm\/ah4PKQ4I4PfQrYbQ9U+9YR9ZJmyIdNkW\/cKEjnn7CESGOGRzCkhLbP3XH9kt8ofWTFomPD2wT4ZjjJB873OgEZ+JwIrkIki8W2sFxxrGBHkzX55qtifNkrimDGQIYUg9r9gnPTFvOPhcwLmSUkWx3pg0Rn6\/9xTCmDfCJ7QEHH9vurw7iQwj1c57Dlugjzm\/0D\/1EfxGGfqRL1sdxxAO8cCIdjgg2ikNDunzHIPHkp63kidMSTj3UR72Uix7og17EZ0rS\/rANJLbRzLS59uENA+yYOknHOThXncRLNAPTZwMYMydhpnr7AhMbGBgjlkxhNJNKpYy8bCcNjm3CiKP8RHF9m5xokmVm1p+sl7Iok8ys2UfYJoy8lMV056233hp9s4SDkbhkPWyTNqkX+rFPfoQ05EOSaamLOhG2iS9V4rYl66Is9tGBeOqgLoRt4pOS1C0ZHm9ni6ccyqMOhHri9Mm4++67zxDSEg5L2CbTUz5lIHE6yorTEk560sEXzsSzZp94hDYTjiTTUi\/lImwTT3noQR3sJ4Uw4kgTh8d1EU484cRTL5IsGz0IY006dGEf+fGPfxy9gRPHsSYc4a07ykFiPVmzTzvRgfJiyVV\/HM+aPOSl\/KRQJmUndYvjSU8+8iclqWsyDTzggj5x+jgt6ziMdZyWusjDPuHogk4I24SVKpRJ2dSRFPSjbOpA2M7e\/t9\/24k8cRlxnlx60dY4LeXG6diOw+N1kl+crpg1ZWaWEetKHGXRh6SJ60Q\/wmM+pGefdZyGGRvYEUZcnJZ4wiibMimbeIQw8pCW\/Vioj3xIZp44DWvKJU2mxLzjcliTHqHObGUmy4rzkx7d0JF49iW\/J6AZmN+z0FYDEGAGge8hMAJHXdaMdhj1JEffzDIRhiRHr9ziISx5CyZXGZQvEYFQBGS7oUiqHBE4SEAOzEEO+tsABHBKmGJlyjZWl2l87h8zAuJNBKZ9uYXAPWSe4+E5j7Vr10Yf7yM\/+UhLPp6bwHlhO7MM0kmKJ6Ac2Qlge7Ld7GwUKgKlEpADUyo55asqARwNnr2IH4yOK9+4cWPfN3fie8c8kzRp0qTooU6m2qdOnRp94ZN74zy3Qd729vbo+zs8tJitDOojnUQEyiWALcl2y6Wo\/CJwOAE5MIczUUiZBHiYrlTJVTWOyOc\/\/\/lc0b3hZryK2NPT07cfb+C8xNvxmpkcLizxfrymDBybeF\/r1iFQqt2SLxcl2W4uMgoPRQD7K1VC6VCLcuTA1IJ6E9fJQfRnH\/5w9Cl5bssUKx8dP94oo1REfP9i+PDhh2Xn1fTMQN4+4FXMzHDK4NsMmeHab24C2J1st7n7uBlbV2u7rSVTOTAB6asoi5yP14cNs4+98YZd7Ld3ipFz3n7byFsMx3Hjxln8ujXPu5CXt8LWrVsX\/WwA0\/c8A4OjghMTv3rK66rMwOCoZCuDUTNlSVqHABcC7E+22zp93gwtrbbd1hMzOTD11BtNpMukXbvs9CLlFE9fLAI+Hc\/rk7xZxNdm586dGz37wkf4+D4Jz8XwDAyvIsavMpKWfPx+Do4K25llFKuH0jcPAdlu8\/RlK7WkWnZbIaYlFSsHpiRsytQfgaGeYFiRMsTT97fgiPCNGxwP0rLmmwm8WZT8tgLOCmFI8vsJfIOBML6rQFn5yiBO0noEZLut1+fN0OJK2G22z05kY0U63rSL4\/h0BQNFhLg4PPRaDkxooiovIjDS\/7YVKYc\/ueIFaBGBKhOQ7VYZeD1U1wQ6hLbb2CFhwMenJuLPTiRRcYue73DxNeQ4nO8dcdueW\/rkJTwui+2QIgcmJE2V1UegEqOBvsK1IQIVJCDbrSBcFV0xAqHtljc3s312ItmAp59+2vjBWH63Kw5nZjs5S86zh3Fc6LUcmNBEVV5EYIT\/1QyMQ9DScARqYLsNx0gK1x+BUux2\/TBu9BfWFl56yPzsBM8Z4rDkKoFbSThC3NLPlaaccDkw5dBT3pwEQo8GclakCBEITEC2GxioiqsKgVLsdtzevQXrluuzE7kKiJ994bnDXGnKDZcDUy5B5c9KIPT92KyVKDAMAZVyCAHZ7iE4tNMgBEqx2xPzODDc+sn22Yn+cMTPxXD7KfkCRX\/5SomXA1MKNeXpl0Apo4FC3kLqt2IlEIEyCch2ywSo7DUhENpu49s+vEnEpybiz07wQG48u5KtofwGHXLllVcaeRHyZEtbbpgcmHIJKn9WAkXcj7X4WRm9hZQVpQKrTEC2W2Xgqi4IgUrYLbd\/eJMo+dkJHBvCk0qzTzhhrMmTFMKICy1yYEITrbPy0umt9r2ul+y\/2CabYa\/YZHvKzrNu+5jLt+wt6+p6uSIahx4NVERJFVq3BLDbO7petGvcRj9kb9qp9qJ1WNo67GW70bZXzG7N\/8l2HYKWkgik0zttyZLf2RVXrLH29gdtwOBHfX2vy1KbvwC7faWkcgvJ1Ip2KwemEMtosDSc\/OfPX+UHzT\/bGQt+Zss71tjF9hm797+fZPeeMdl+9Ocn23d7Pm6v2632pdQrNmDAfzPSk88C\/SvlfuzwQHWrmMYkkE6\/43b4K7fb79gHF6ywVR0\/t0vsT+2hbx1nPz\/hVPvXT7Xbd3d93DbbzXZlarPb7d95+tVGvpAtlu2GpNn8ZaXTu9wO17vdPuZytzsvm2zJ\/3jG0rsmmp35ftv00dPcRlO2YP59dt75z7vdcr592cN2BYXTinYrByaoCdW+sLTPuJx33vdswQJO7Nvt+Nkn2On2nH0kvcbWXmm29Bmzb95h1v6v3fYRe9iOOslsQGq0p\/+ltbd\/25YsecJC\/Bs02GxwkTJoUIiaVUYjEjhot3e4HXb5if11O3r2WDva3rbpr6y2R79stnyT2c0\/MTvley\/ah+zXdvz492xQqs3Tr3S7vTWY3cJOtgsFSSEEDtrtv7gdPuR2+7LZkPFmxxwwe98MG3T6Pht81nrr+ZU7KgPfNBt2itm+DWYDzvf0D7jd3ud2+5SF+lcLuw2le6nlDCw1o\/LVJ4ErrrjPD6R3Xbl9kbyVHmj\/an9mX059w7Ye+Kyds3i8nb3ig\/YPs\/+z\/ZP9f\/bc+hPsQPptM9vvYj56uM+6QtxWqsQN2UhD\/WlGAldc8WO323e8aQft8J2XhtjddrF9+cRv2g632xMWt9ukFR+1eX+5wG6xq+x\/bzjV9qW320G7Hex2e38Yu\/USTbYLBUkBBK64YrXb7fFmA9w5aZvpjstZZm+\/Z9a92vatfNv2LhlkNsrjRwwz2\/WYpY7xc+2BV8zsWJdT3W5fd7t91bcDLC1ot3JgAthNvRTBaOCg83GgT6V3Hzlgb9hx9r8OfM6utkU2p\/M3dmXH\/fZN+7K9ZO321qiT7ff\/DuY7WMbvQ0vaasUbsiWBUqZ0+l0\/iW91EEe7HFy2L2+3QbbffuTOyxz7rnX+xeOR3X7rwP9rr9pE278NAxt1MLH1ROuurvXRuuw\/FO3XGytGhpRdqwoomkBtM6TT+6zrqY+ZnXSR2YHNZtt\/bj\/49hw74o8HuGLvumAU7qw8t9fj3DEfN87GnOQ2e8JksxMu9HifDndvuatrp28HWFrQbuXABLCbeikilRp9uCq3rLLNXWPttc0TbMueY23LvuPsrfeOsU27xtrWJ462\/R956rA8qdRRh4UVHTDSc7QVKcM9vZaWI5BKHelt3uGyzaX3lHTXz+3575xmrzxxou0+MMy27hhtA\/YdsB27R9gLPzjV3jnLLwrGDIxnsQP+Z4+lQtitl2SyXShI+iGQSvnsypurzNav9JQHbXH5CZ+y\/SN9Nua8L3oYNtputucxsyGbzY7\/uP3R3zxttsUdlhPxjon\/ndttr817jrKWFrTbQOTKwq7MAQncveJTh5d23t+ZnfAd23NM2vbN3Ga7PrTL9h\/1oL171rfM1t13SPqOjvdZZ+eUQ8JK2qnAaIBvD\/BNgSlTphg\/GIZe8UeTCJ85c6Zt3uwnCo\/guwOEIXzO2oOiJVsZUYT+1JTAshV\/4vUPdmHU6it7zvb95TfMpj5rPZOH24GTB9r2U9rsvSP3mP37H5jtv9kT4bj4ygbaQbv1kS275UqBtnvIDE2sdo66s9mdbDcHrAYKXrz4D80Gn+Qa+9pG2+LJk23Uql\/Yopv+o834ny\/Y\/\/X4YzbzN0fY397znH3x5P\/f\/vmy\/2A28yyzxx4ys\/fcblPW2XmqbwdYKmC3AbSqaBEDK1q6Cq86gZEdT9vXV3Tb+JR7+YfU\/rrZiPvNVi8y2\/A9s+OeOyR2Qmq7zZr3in1+hY8oDokpcSfwwYRDgiZ8WyD5y6hsz5gxwwi\/7rrr7IYbboicmNtvv93WrFlj\/CIqv4yKw5OrDMqV1JbA0I7n7D+ueMOOT+1xRQa5sHBr6Idm3f9stsWd8M2+HrHEI9a5sAz09PvtknlbrHPFA+iRjOQAABAASURBVASEEdluGI4tUMrKX79l9uHxZseO8NZyzn3Stp54mV39rRts\/b9MsZ+e91F7+TPb7I4vfNhuXnWl2ZDfmj20zOx9nnbINOvqetLzBVoC220grSpajByYiuKtfuHbbYud2fGO3df9M\/td98027z\/93KZ39D4k9gYPSbpOb\/p6wzt2YmqbdX5mrS2+bZmt7P6h\/cX8tA2yffaKveiJylw4ntu8jGJkuKfPsYwdO9ZGjmSO9GCCE044wdra2mzjxo02ffr0KHCyj37YSKfTNmnSJBszZkyUZurUqcaPkPGjYnzemjTt7e3Gb3vs3OknEgIaTppL4Vdsr33A7fbW7sfs3u7lNnfeb2xKxw5v5BFmO3b52g2pZ5\/ZO3ttbMrsss4XbMHix+073Y\/b9PkD7Hk73dL+30L8k+2GoNgSZaQnn262e4jZn\/ssyg8\/YYO\/8hkbNMqd6SWvWfo+P48e2GLp9WdZesN0s+Fv24n\/frKd+p1RZvf8qQ14\/Ayz265xJ+bg7aeygQW2W\/T56le\/Gn1Nd0pi1pvwTCFdPEDMjKvkvhyYStKtQdkjbYzttBH2lh0TvWb6t\/\/0C\/vl\/7jVDjx7nXX\/6ibrfsRlzU3R\/vqf3WCL\/2WZffIvN9rrNta22ZF2hI23E83v4VqZ\/4Z6\/mHFySN7yeB5siz84um5554bHUzXX3+9fe1rX8uSymzDhg3W09NzWBzOS2Zgtl9XzUyj\/eoQONkO2HYb5XY4zvamjrHPzN9o9\/7sTttz4G\/tke7bbc1Lt9qadbfa2\/sX2PPP\/6P9jTsv0zq3+yT8EBtuPXaqO0Apc8\/GAvyT7QaA2BpFpJ7fYHbKOzbqtHdt4IThduxNe+zkVafbiN+m7ENPj7cTlrfZ6T\/aaWe\/\/aoNeexE2\/aNCbZ+5gyz\/2R24JK9lvrmauvoaLMg\/0qwW3PfK1fdsUPy0ksvGTPdixYtMm57JtOzP2vWLFu2zGeVkhFV2h5YpXpUTZUIXGCzbMS7Y+xVm2jP2Wn2qJ1jL5x4ir15+jE24SM7LPXhrTbxj7bb9tNH2qunTbDfDJ1qT9lkW28n2e6tQ+0P7U8L0rTfREyWcFwWIRNH8VBb9pJ5jmX58uXRraJrr73WLrvssuhWUWbq8ePH2\/Dhh0\/l8MNkmWmZgWFmJzNc+9UngN0e926bDbD9hiOzyU6wJ4ZMtidtig1NDbW29kE28tTBtm7AqfbYEedY2t5n2PgWO9YGvrPfzrRLLNg\/2W4wlM1e0OJ\/HG+pN9+ybV\/6ne2\/YK9tOvMEW\/fIJNv5o5H2Yvf77d19J9rmPZPs8WfOtj1rh9rWz4+2Pf\/NPY1fbDJ77WX7+2\/7zE0oSCXYrR1+quzThkFffzPWTz\/9tHE+vuSSgMdfnwb9b8iB6Z9RY6V4rcs+94832Qfu\/p29aKfac3a6\/drOtl\/ZdHvIPm732wW+\/ph12Xm2xj5iz9gZ9saG4+2Yrrfs4kX32Af+7WYL8s+PUWNCpQiZ0JbbgUGn2AnB6cD5IGzcuHG2evVqNqPnXdhIpVK2bt26yMFhhLB27VojD\/lxgkjT3d1tzMCMGDGCXUmtCbzeZXP+4et29o9\/ba\/aRHvZTrI3bIw9YZPtt3aWPW5n2y9tuj3sNvu8TbL17sDs2TLUxj+80WbfvMTOfvAmC\/ZPthsMZbMX1NW1wdIP\/dRsf4\/Lw2YvvGZ2\/vNm3zbbunC07bxhpL394tFmv3ASX9tt9pNnzVZ62tF+rtv7kN2\/dLNHBFpKsNtHejhBF1Y\/50tuxSdTT5s2zZgdT4ZVc7tBHZhqImqwut5NRwp\/+N5H7Ssf\/4Z99ry77aR\/WG9Dlu227Y+32YaNE3w9ynY9MsxG\/cu7NvUra+3zn\/2f9slv329tb243S3dF+cv+M8JLaCtS8owGLr\/8clu1alV0C4mDZs6cOdEzLsnwhQsX2ty5c6Nw4knHczE8A8NBdv7557tCFpVBvquvvjp6RiYK1J\/aEngrbeYn4LMfeNyu\/XfX2+f\/3Z128i0v2TEPvW37nh5kWzePtpHP7LT2p9P2B\/c9a9Pnr7arvnSLfXrZfTZsh18YNgWyWyjIdqEgKYAA3zAym2B24ANmx043+6RvX3S02TS3yaF7zH7xG7Pv\/c7s\/gfNBjxudoZfcr823Ow\/eLrhH7auh4dYsH8l2O3Eo9yRKlABBo0MBAtMXpVkTrMq9aiSahGY2GGGIR\/lFZ5gNtq22ke+\/7B98iv32+c++UPrPGOJ\/d8X\/i\/73Kd+aJ+cf7+d8cjTZuPN7FgXHI5TPL9vlr34xajYGRjLcyzzwO6dd94Z3ULinmzsjCTDV65cGTkv6E486ZCrrrqKoEj41VTCkr+uGkXoT20JYLejXAVscKLZccdssY5\/67LPzL\/b5vz57fblj3\/T\/vyy79mn\/+I+6\/hWl01+6kkz0g\/2PNja+A7fCLRQ3jAvqxgZ4ulzLEkbxfawTZImw2W7EGk86ejwk6wNd8XdHk\/31ZUuZxxvNsWNaIobxQV\/aHaWOzfv98HTTHdwjp9k9g13Gv7pPbOBv7aOab7tWYIsXmVB59yEXU\/I48A0woy1HJggllNHhYxKmZ0+z2yc63SSCwfVH\/iaT7v4seQz8n5A+f7ZLme4vN\/Fs9iJvvaZTfvEYt8IsAS+HxtAIxVRzwSOdCNsd7sd60q2uUx08fO9\/ZGvp7lgrx\/yNXKKr7HtY3ztA1nza4GdF8huvUiT7UJBUgCBVGq0zZvnxjjQnZgHfZZludlR171jhs1+1As42QV7ftnXD7j8cqvZ83fb4PH\/Zra\/3RYv\/qAHBloC223saPMtreSMNQ\/38tZRIK3LKkYOTFn46jTzlxaY3WS2ywcFdprr+H4XnJWzfD21V9h2x2b38Wa7fRCw\/14P5xqwjO9s+Ha5iw8+bKgXUowc4em1tC4B7PY6s70+627MIGIPrHnO8X2OBTtu9\/V4s\/d8ln4fDvf3ff82l3sD2a0XZbJdKFRKmq7cBQv8RLvfnRZzT2Xpr+2d8dvNPv2M2Xe9qf\/cbfZrd1p+4WnW\/cANd78HHm97148we3evLVnypu8HWipgt9lmrHFsCE9qzT7hybBqbMuBqQblatdxwCv04+m9n5p1X2P25Dd9vcJs4+9cnnXxYyvdZfb0nR7+j34cLTIb8KiZke\/VtG8EWPz45HguSnwQE6BmFdGoBAYdVPzAj83e+JLZS3PNNv\/M7B0f2G593O10jdlbfg1Y7xeGDfPMer7l6R9zYQbmxUB268WZbBcKkoIJMErjvox73D1jzN52z7rNnZnfbDIb4B74znfNUn7r6Bi\/jbT\/FTNz73v\/Hl+\/aun0C74OtLSg3cqBCWQ7dVXMzA7DcRjlF4TjXDF+AHXvI2ZvuEOz0WcuX3fZ5ReDtpfMiPfDyQb48Wanp3zq0\/N6nrKX+Jge5iUVKkM8rZbWJfDHbns+DX7EKLPj\/Vow1mcGB652x8Wn5d\/6idmWu812+kB21HqPd0ptbt8HPJ293+32XM\/rYf0uhSSQ7RZCSWl6CXR0uJNifBDTT6JHTTQb6PY40KcMD\/j90B6Pe9ZPgE+5U\/MWsy9u1LbRc261VOpY6+jw20++F2RpQbsdGAScCqkrAvuuX2y7\/Jgxvz00yo+piX6i98PJuB3r4wBjRj7lGhN2nKcb7F7Me6PN9nx2ttmHA10I\/EKEE1WUaAbGe6V1l31\/t9h6cEjcFnmofOSRZscNNps4wCLbPcnt+ATHc5SnGc7J2u12rwfs+rTb7dmB7NbL1zMwQJAUSmDx4g970t0uz\/v9+F+b7X\/RvW6f5t6a9jC\/RTTgNV\/zg4\/cG\/2lb7\/ucpTNnt1uHR08rOi7IZYWPOfKgQlhOHVWxo4lS6zbj5nf+LGzjlkN91SGjDdr8xP+qF4Z5vs7\/NjZcqzZs++ZPfOq2Yb\/vCBcS7jAuHNkxciQcNXXaUlSKw+BHrfbTT678qyf319kht3t03xAO8AHsjbBDEeb383rSZm96Xb84nazF3wWceN\/CWi36CfbhYKkQAJLlvi9+QEMC302ZYdPGZpPdxvOC78r547MAd8f1ftzGObGPNCN1\/bZggU+DW4B\/7Wg3Q4MiE9F1QmBvS+\/bANcl73umLz8ltmP\/CR\/72az+7aa\/dxP+gj7P\/WZzMfc0dnyjpkPbqNHYPal054zwDLCy\/DZHytGNAPj0Fp32e12u8Obv8vt9k23258+Z\/bgJrOH3HYfdmf8YY98wO31gXVmv\/X1m77PhLxnsffSaVZhRLYbhmOLlPLyy+5xH+jy1voJdeBlvj7SxT1u400Jny4ccY7ZNndczA3Z\/DbSfhfrNrO3LJ0+4OtASwvarRyYQLZTT8W0zZ4dTXxgz8zGH+3KjfAz\/UCX7T4Q2OkyxLf9DpMRj4+B73B0Z6cNSqUsyL8WHA0E4dbChRzpdhvZqjPAfHyS0Eb2mA3x8z12u8Od78H7zBi\/xjbLGrs9IpXyXIEWKi9m5pC0mj0MBL\/xipk9+2xXmpvzO\/32EW9DfML3cWL8\/qa58e58w\/fdKzeMxJ0de9n3R1hn5wWWSg3w7UBLC9qtHJhAtlNPxQzp6LCB7oxwS3S0KzbGhZM+YwAuChxWyPEefozLCBc\/9GygX0B8M8xC5VxdihG8qDC1q5QGJHCE2+1et9vBrjunf\/wC7Bfb5W4Sgt0e5fE4OpyvfRLGDoS0Wy\/bZLtQkBRIoKNjvDsjfhYd5t62veu5Frs86eLThIaVYsk4MD51aD4jY+6J216bPZuPb3myUEsL2m01HZhQ3aRy+iHwTleXPb9kiT3h6RCfhTf8fE78XBhYY+uc\/Nd7Gg61Db7esHSp\/w20xMctx26hwgAlUPUqpvEIvO12+4zbLfaIYLc+fjW\/NPRdBjhhbfOmMYZ9ytfY76aQdutl9lVWqN2STrYLuZaUrq63bMkSvyW0a4u3\/20XnyY0rBeHZa3vv+jC69I4LDgwGMt2W7r0MQ8PuLTgOZfzQUCCKqoeCOxKpyM1Bvlf\/H5O9g\/79oMuD7nEay4AHGaMeD3Y4nxsly1cdYqZfSGtZmDKxt7IBexOp6NJds7DOC4bvTGPuGCzK3x9vwtrLgtcJrBvniDo8XweFW6R7YZj2QIlpdNYI7eJmC9kXhtHhW2GiVgyjg1rZmgIw8KHWDrNk+qFAiogXQXsli\/u8iXeKVOm2FNPccU4VA\/CiCMNaeNYvtZLGHLLLbfEwcHXcmCCI619gWN9Gn6MT8fjD+AXIBxOo1y1WJiFIYw4BpCjUylrnzfPgv3jGKXgYoSBSTAFVFCjETjB7fY4t1tMhnMxtnqsN4LbnNgr6\/gWEpcBHO82t9tUSLv1+jQDAwRJoQQ6O0+xjg6cFtxprBcnpt2zn+TyPpfTXPhtAWZmuFk\/1FKpM2zevI95eMAl8DkXJwTt+P2uO+64wxYtWmTbt3P7i1CLtgkjjjSEkmfz5s12++2325o1a+zJJ5+0tWvn+l6gAAAQAElEQVTXZnV+SF+uyIEpl2Cd5ueBSPx9LgKjXcdM4RkC4jjcOOzeS6XsKL94eNIwC1cYvKNiBI8rT+14+Hj0CAcKSTmgZs2aFf3C9MyZM42Dh3DiSYckRwBxGYwaGD2QVlI\/BEbNnm3cImK8ysmJczIS3\/rEnHBuiMNuze326JB2CwrZLhQOEe3kJzB79sc9gd8eGo6lcgOU77383MO4yfmGr\/0Wk3Gjnpv5Q9xsd1pHx2gPD7gEttsXXnjBLrzwwkjB9vZ2GzVqlO3ciQMWBUXbhBFHCGnJM2LECBs\/nqctCbUo39ixYw\/uBP7LeSBwkSquHgisueIKe8sVedYlnnLnmZe9vo\/g3PBSH4cVt5Fe6+qyl5cs8dhAy1AvB++oGBnieXIsOCT8OiqePp798uXLoxEA3v+MGTOM8Ouuu85uuOGGyInJNgKgDIonLfkYPeAAESapDwKPuN1udVXSLjw1gA3zWOQe39\/lwjaPRj7v27zP8ZLbbfeSJb4XcJHtBoTZGkVdccXd3lC\/VdSDk8J84Rjfx5nhJj4uOWdcvhXDhf2AdXUNsyVLsHJPFmopwW4f+S0n6MIU2LZtm23axNUid3ocmLa2Nps7d65deumlNn36dJvtg5IxY+CRO1+pMXJgSiVX5\/mOTKVskOuI4PP7oRX5\/y95GBcGnot5x7d9zOB\/Dy4jPc\/BrfL\/7h401HYNHlaU7BmU24P5xS9+YTt27IhmWqZNm2Z4+xwoGzdujA4SNJ48eTIrv7ectkmTJhkHDWmmTp0aHXgcXOQjUXuWEQXhktoSGNFrg1gCt4jedHU2mhmPQXK6R7gUcIkgDfY9vDePJw2yyHaDYGypQlIpLtA4AwwJEYaH3G5B8Cz4Ui9hPC\/DmXezpU6ZEJRRKXY79iSuDoWpwWxLfzMpDDKZ2f7CF75gd911l61evdquv\/56iwePhdVUeCo5MIWzaqiU5yxefMitfO7OcMuI6XdO\/kzFIxxyxE2eN894\/iBUI3d77bvMHZgiZI9xScquAc4LzgqzJ8zAMMMS3y5K5tiwYYP19HCJS4aa4bwcGmJWyIgiM4\/2K0vgj91u93oVONYItomdMjvONmNabJYTF\/Fnud2e0NHhOcItst1wLFulpMWL\/8hs4pneXJ7U4nYJVoyjgrXu9nBmYbBYzk1H2bx551rHuex7VKClFLs9bgJDgOwK4Iww001sd3d3dL7k9hD7CNucQ4ljn7TkYZtbSMQzgGSbsEoI54FKlKsya0yAUSmHDIcRF4BYcGJiYYTLIcQhtSewvjttpG23tqLk4Udwr7IrMnLkyGgqklgODEYD2aYzOViGD+dSR8rfS3xg\/T7ECr43m8yj7coSOMJnU3jUEbvFNrFRhB7Fhllz0uIZGdKEtltaJ9uFgqQYAtEMzKvMFx7p2c5ymdIrODRDfJu1OzYDZ\/g2Vvv7Z0k8IMhSit32GEdU9urPP\/\/8KILnCC+\/\/HK7+uqrDYeE2RSeJWSbMOJIQ2LynHnmmcasNzPiyXDiQwvngtBlqrw6IPDCkiXGYcIEJtPur7pO3L1kTMBzBbyiyt1a1jxb8PiCsL8ns7uEGZhjJ3LJckWzLOeee64t7f3eBx4\/My1MZ44bNy6apiQLT7yzTvlFcN26ddGzMDzjwlPwpMWJYZRAGspg9IAzxL6kPghgtzgnPK\/Fo4\/c+nzXVWOf9Tbfxo6xYWw7tN168SbbhYKkGAJLlvBDjdycf8Wz\/dLl1y44KZxdGTJi1X4m3v8rD3\/PFizg+zC+GXApxW735Jn1RrUbb7wxer7wiSeeMBwTwnBSCGebMOKYGY\/DCL\/qqquifJnhxIUUOTAhaVakrNIKbfOLOCNXJghZU8oe\/8NFgBM\/o1fGBXHcKE\/v0cGWHTaiqNmX7T5bc+SEwkYDF198sX35y1+OnnHB+1+1alX0bMzChQujh8d49mXOnDk2bdo0YxTAaIADjQOPBjIqIB+jB0YRhEnqgwB2iE3iymK7aLXb\/+C44Hxju0zKM1fHyYuZRo8Oush2g+JsicJSKd4oGuptPeCCBfNsCa9P4IIzM8MaJ2arx++2VIrbTb4ZcNlRwjm3x3KfcwOqVrGiOAdUrHAVXDsCp3Z22sSOjujD1Zzw4+l31pz8CYvlmFTKLlrBJ8LC6bu7hBmYPQWOBvDqY2cEB+TOO++MvP2VK1dGTg2tIJ50CKMBwhBGCYQxasCpIUxSPwQmud2OdbvlVM\/lABvlFIvNYrvsH+HqEo+T\/tnAdutFm2wXCpJiCHR2nmYdHWd7ltNceGILK8WJwWJ5EwlrZv9Ed17OthWrzvB0YZdK2G1YDcOXNrC\/IhXfmAS2pdO2x1Xn+RZO9nQ0FwIOJ4TDizDSvOFp33Xx5MGWnSU8A9NjaBhMBRXUgASwQybeOeVjt5zymYnBVrFb9mkWcW+7zWLn7IeUnbLdkDhboqx02q32yNHeVt4y2u3rcWYDp5oNwlEZ5fsnuRzr0mPp9O8s\/SJvJPluwKUV7ZbzQkCEKqpeCHBiX9\/VZTwQyfMDHC4878K3MxCeL+A5AhwcJj2fW7o0qOq7KzADE1RBFVaXBLDb191ucVCw262uJXaKYLdcHrZ52B4X0vwusN16sbZbtgsGSREE0ul3reveLs\/hjowhbq3715rt49kYbtjz9SLOwjx52GNLl\/JUoicPuLSi3cqBCWhA9VTUeJ+GZ+TK3VhmWxi9sh1LvE8a9B4\/cyarYLLDRhT9DEyPaQYmWAc0aEET3G5jGx3ibWDiHftFCE\/aLQ7MxMB261WabBcKkmIIdHTwIxfbPAtuN0NCnBYsGMtlmMhQkgd6B3iaU2zmTNL7ZsClFe1WDkxAA6q3oi5escKOTqUsPvFz8s+U4z1+2rx5dnpnZ1D1W3E0EBRgCxeG3R7jdjnQGRzRK1wKcGawZZzuIz0eu\/2Dzk5PEXZpGtsNi0Wl9UNgxYq\/sFTqaE91nAtPbPnKBvgfLBerPdbjp9m8edOts3O8h4ddWtFuOUeEpajS6oYAszCf7+622S7jfGQ7yDWLhfHBJxYvtj91J+fs+fM9JuyyU88RhAXaQqUxC4Pdzum1W2ZaaD6XAtYXut1+zu32I\/PnsxtcZLvBkbZEgR0d46y7+yqXS62jY5K3+XQXvtDLI+jH2eLFn7AVKz5i8+fzTRiPCry0ot3KgQlsRPVYHK+mHj9zpvEaaiyDUik7rbPTiKuEzq04GqgExwYus2zVsc2xbre8Os3E\/DYvEbt9v9stMzC+W5FFtlsRrC1TaCo10m8RHent5alDZIvPvPT4rMvJvh7q4ZVZWtFu5cBUxpZavtRWPJhavtMrAICnCap9kpLtVqAjW65IZl22eqt5A4ln+3gGxncruOSz2x3WZtlkVz\/PHfLFXb6bNWXKFOM3jjLVJ4w40pA2judrvYQhyfA4PtS62ueGUHqrnCIJpDo6LPlvyuzZyd3g2zv9wNjht5GKkV3GEzqBVFExTUEg1dFhnKS49cktpKkVtlvzf7Jdh6ClLAIdH\/+A5+eJLZ7e2m2zZ5\/j+5Vd8tntrujNuqGWud5jPGWWXS+cEGL4btYdd9xhixYtMr5sThjCNmHEkYYw8vAbdXzxnC+jE87PwODoEB9aODeELlPl1SGBk\/xC8LcHDtiXuruN9Yz58yuqZb7RAI5KNtljHOwVVUuFNxiBlNvt37jd\/j+9dvvH8+dbpf\/JditNuPnL7zh3sB04cI11d\/+Zr6+1+fMr78CEtlt+APfCCy+MOqu9vT367bidO3lFPAoytvlNOuIIIS15cFxwWi666KLoC+n8EG+lPhoqBwby4aVuSzwqlaqKbjt99mW7T1sWIz0+a1MV5VRJwxGolt0CZqdsFwySAARSKZ6FCVBQAUWUYrcbHuGLYAUU7kn47bhNmzb5Vu4FB4bYrq4uu+uuu6IvpLPPzAzr0CIHJjRRlRcRCD0aiArVHxGoAgHZbhUgq4rgBEqx28ETjytYD2Zb+FHcfBn4wVziL7vssr6fdSEsdmyI618KTyEHpnBWSlkEgR36kF0RtJS0ngjIduupN6RLoQRKsdv9E8blLB7Hg2dZSNDtt3CZgRkxgoeTCTFjmzDiCCEtefgBXX5gl2dheE6GbcJJE1rkwIQmqvIiAqWMBvQMTIROf2pMQLZb4w6ocfWNWn1ouz3\/\/PMjFLxJdPnll9vVV19tbW1txu0g3iximzDiSENi8owZM8bmzJlj06ZNM5yZ8ePHG+HEhxY5MKGJqryIQCn3Y3sKeAYGj37WrFnRQURF8T4H0MyZMw2vn3AOMsKQW265haBIOPAI49W\/Sj0ZH1WkPw1LYGeFnoGJbRXbBE68jz3KdiEiKYdAJez2xhtvjJ5jeeKJJyx+EBdnhHB0JYw43jaKwwgnDWFIMpy4kCIHJiRNldVHIPRoIC54\/vz5tmbNmnjXeIVvxowZ0UF23XXX2Q033BA5MbfffnuUjifi165dG33DIL5wcFCRj1cAuYj0FaYNEXACu6NXTodZtjflcoUVMntYmO1uNtmud4KWoglUym6LVqSKGeTAVBF2K1VVyv3Ynn5mYGIH5JJLLulDuXHjRps+fXq0z3QlG+l02iZNmhQ9RMY059SpU42n53mQjFf9SNOe5bVAwiUiINuVDTQigUrYbb1zkANT7z3UoPrtLmUU+8j\/ztlabvfwWh6j2JyJPGLDhg3W08Ovv\/pOYsF5SexGmzyAhmMT7ehPREB\/zGS7soJGJFCS3Tb4t7fkwDSipTaAzjtLeI5g98T35WzZ6tWr7YEHHogeClu2bJldc8010W2hzAw8MDZ8OJ\/uPjQm21PwhbwWeGgp2msFArLdVujl5mtjKXbbY4efKxuJjByYRuqtBtK1lNHArgm5HZirrroqes6F51e4hXTTTTdFD5WNGzfOcG5Aw\/MurFOplK1bty56FoZnXHgGhu8X4MTwqh9pePWPGRheBWRfIgIxAdluTELrRiJQit0W8uxWPTOQA1PPvdPAulXrfiyv8PGdAd7kWLhwoc2dOzd69iX5Gh\/PwPC0PE\/Gg5S05OMVQJ6RIUwiAjEB2W5MQutGIlAtu60nJnJgEr2hzYAEdntZu4qUPZ6+gIXX8mJnBAfkzjvvjGZnVq5cGTkvFEE8szUIszeEIeQljFf\/cGoIk4jAIQRku4fg0E6DEKig3dYrATkw9dozja4Xv\/m13RtRjBz+7K0XoEUEqkxAtltl4KouCIHa2m2QJhRbiByYYokpfWEEWnA0UBgYpap7ArLduu8iKZiFQAXstr8Pf\/J2KB8F5bY8aTO14iOi2cIz05W6LwemVHLKl5\/ADo8uZvaFtJqBcWhaak5AtlvzLihaAWUwC2y38Xe3uOWe7cOfvCDBx0CJIw1dEOeJt\/mwKNuVEjkwlSLb6uVWYDTQ6kjV\/ioRkO1WCbSqCUogsN3y7ax8H\/7cuXOn8SkKPgpKO0hLHrb5SRe+KM2LFexXSuTAmuMIfQAAEABJREFU5CHL1BdTY0khLE8WRcUEWvB+bNz0Wq+x0aTNsk1YrfVqmPqLt10zzR6W3b3YKLaaFMLKLrhVCijBboetfaRgOnx2or8Pf+LAMDPDz7pce+21xucrCq6ghIRyYPJAi99YYXrsnnvusYkTJ9rs2bPz5FBUH4HAo4G+crXRLwHZbb+I8ieQ7ebnU6FY2W2ZYEuw271HTSy4UmZb+nNI+NYW39ji9+ouvvhiu\/LKK40Pj\/IsTMEVFZFQDkwBsJgO45shX\/\/616OPpxWQRUkC348V0DwEckTJbnOA6S9YttsfoYrGy25LxFuC3e4dMSFnZTgj+T78yUdAmZXBYaEQ0pKHz1PwmQoG\/rfddpvx4dHkpyxIG0rkwPRDkumwr8yebZddeKHxbZF+kis6JlDCaMAK\/A5MXIXWuQnIbnOz6TdGttsvokolkN2WQTaw3cbXO27pJT\/8yYO63NrjG1wM7IkjDZrHediuhsiB6YcyPx44\/vHH7apx4\/pJ2bLR2Rtewv1YPUeQHWUpobLbUqj15pHt9oKo\/kp2WwbzCthtfFuPGRVmVtAOJ4VwtgkjjtmWOIzwWJJp47CQazkweWhy327Dhg02f8uWPKkUlZVA4NFA1joUmJWA7DYrlsIDZbuFswqYUnZbJswWtNvGd2DK7PNc2bkP+\/3vf994GGlye7ud\/F\/\/qzFNNmvWLGOaM1c+hfcSKOF+rGZgetmVsZLdlgEvzirbjUlUbS27DYC6Be1WDkwOuxkzZozx2zpMjb3U3W0vff3r0e\/t8Ls73PvLkU3BMQGeZ2FEUIy8F2fWulQCsttSySXyyXYTMKqzKbsNwLkEu7UGP+fKgQlgNyoiC4HA92OZ9WL2i1kwhAfJqDUZPnPmTGMkRzjxpEOYmiYM4eEzwvj8NZ\/BJkwiAocQkO0egkM7DUIgsN02QqvlwDRCLzWijsy87HLFixFGEJ4l28LnqmfMmBHNgnFbj6884qwkw\/l4Ep+uJpx40j355JO2du1aw1nBqaFsZtXIx2ewcYAIk4hAHwHZbh+Kym6o9KAEAtttUN0qVJgcmABguQgyOxBfICmSUT+jfbZbUgLfj+U7Aggs+f7A+PHj2bSNGzfa9OnTo+3JkydH63Q6bZMmTTKmpbndN3XqVOMLknwlks9dk6i9vT36DDafw2a\/FUV2m6PXZbs5wNRHsOw2Rz8EttsctdRVsByYAN3BRZLZAT7kQ3EcYKtWrbL4YklYy0mFRgOw5euOsMVByeTKW2M9PYd\/1x3nJTMtH2HCsckMb5X9VrfbnP0s282Jph4iZLc5eqFCdpujtroIlgMTqBs+tWWLzV6yJHoGgy8TcnGMZwQCVdFYxZRwP3bYlvy\/y8Gtocsuu8z4jQ2+L5ANCDMzw4cPPyyKL0RmBhbyaezMPM22L7vN0qOy3SxQ6itIdpulP0qw20Z\/81MOTBY7KDpo\/nybeN11NmjQIOOZC0b1p59+enQLo+iymiVDCaOBvQNy\/y4Hz7B84QtfsO9+97uH\/JzDuHHjbPXq1RE12LORSqVs3bp1kTPJjA3PwPAbHjgx8SxZ7GRyO4o81Zc6qFF2m70TZLvZudRLqOw2e0+UYLf9ff2cxyDyvfTAeZkXIkhD2lgxHqEgDEmGx\/Gh1nJgyiXpB5MtWGA2b56t8jUXSIRbHOUW3dD5S7gfu\/e93L\/LsXTpUnvmmWds2rRp0fd44jeO+Iw1t+s4UBYuXGhz586NHMc5c+ZEaZkF4xkYvhgZz9qQlnx8Bpvp6IbmXKrystvc5GS7udnUOkZ2m7sHSrDbfDMw8TOduV56YHDIixC8EEEaFCMPTg2DRgaUyXDiQ4scmFKJptNmHEx+YcV5YZuHSR944AFDGPGXWnQ18lW8jsCjAT5TzcEQy8qVKyNHBQeEb\/MQHofRNpwVwpD44V\/C43L4\/DVODWEtJem0mew2f5fLdvPzqUVsOm0mu81PPrDd8txgPBDP9tIDL0BwG544FCMteTiv3nrrrca5mXBmvllXQuTAlEI1nTa74gqzpUvNZs8248DycujID37wg4aw7UGtuwQ+mFoXZMCWp9Oy20JwynYLoVS9NOm07LYQ2iXY7bBdjxRScpSG5zp5PCLayfEHByYZxa0kwhhQJsNDbdfIgQmlfg3KSafNzjvPLJ02W7HCrNd5QRM8Th4i5Y0ktglrWWnBB8rquq\/TadltoR0k2y2UVOXTpdOy20Ipl2C3e\/fkfu4ws1pmW\/q7s5CcbYmffWHWO7OsUPtyYIohmU4fPJjIg\/OSSrHVJ7wl89xzz\/V9l6QvohU3ShgN9PdAWStiDNLmdNoip5vCZLdQyC+y3fx8qhWbTstus7HOFVaC3e7dnfu5Q5wRnuekumwvPfACBLMyxJGGtOTh2Ri+i3bhhRda8vY9aUKLHJhCiXK7qL39YOrubrNU6uB2718eXuIB04suuuiQt2R6o1tvFfiBstYDGKjFstviQcp2i2cWOofstniige02vu2T+dID1zpmV7jLwIsQvBBBGhQmD19AR\/heF+EIeYgPLXJgCiXa1RW9aWQ4L1ny0HGZD4xmSdY6QSWMBjQDUwHz6OqS3RaLVbabj1h14rq6ZLfFkq6A3XL7h+ta8qUHrnWEox4P7BJHmjiMePaTQhjpQ4scmEKIdnYeTMWogGdeuroO7utvHgIl3JDN905fnpoUlYNAZ+fBCNntQQ4F\/5XtFoyqEgk7Ow+WKrs9yKHgv61nt3JgCjGOxYstmnmZPfvgm0c8xDtggBm3lHBo0ulCSmmxNBUYDrQYwbKb24x2WzaUQgqQ7RZCqWJpZLclom09u5UDU6ippFJmOCvcQkLmzTNLpSz6iB2ODEJ8V1ehJTZ5usA3ZJucVsWal0qZYZfYLCK7LQC1bLcASJVNkkqZyW6LZNx6disHpkgTiZKnUmYcXLzRceBA9tkZZmlIk05HWVrvT1OOBhq7G1MpM2xSdttPP8p2+wFU3ehUykx2WwDz1rNbOTAFmEW\/SVIpMw4wRrgIo1wyLVhg3Gba\/r732az3v99m\/cmfGK+YEcUHfvgNCT67zH7zSevdj224PkylzGS3WbpNtpsFSv0EpVJmstss\/dF6disHJosZlBWUSplxcMWjXF+3XXGF3fbcc1Gx\/G4ETgufv2ebp7ijiEL+NFSa1hsNNFT3ZCqbSpnJbnupyHZ7QdT\/KpUyk9329lPr2a0cmN6ur9iqo8PMD7A2v9V07Q032M0332wXX3yx8aGf5nVeoNl692NpddNIR4dZS9otPSjbhUJDSkeHmezWu257gdLj6Rp3KcaBadxW1onmOCwXXHCB8cE7Pv5TJ2pVSI3WGw1UCGTNi20tuwW3bBcKjS6y213ehf3JHk\/TuIscmCr2HV8jXLZsmfGVQm4fVbHqGlQV\/n4sX3\/kq47N\/exQDbqqnypby26BIduFQqNLfdltNWiGt9tqaF1OHXJgyqFXRF5+J2nhwoV222232T333GM8A8OzMEUU0WBJw45iORkBgK874vwtWrSo74FowiWVIdB6dgtH2S4UGllkt\/3NvMTxmoFpZDuviu4cTJdeeqmdc845xieVmdrkGRh+R4K4qihR5UoGD95ogwdvKEosz5d4X3jhBePHwWhGe3u78cuoO3cy4iBEUgkC2Gar2S0cBzeh7dKuVhHZbeHn3Xzn3EawF83AVKGXxowZYytXrrT4tyKokl\/pJIw49ptFJk6caLt2jbNx45bZiScuKUrIQ17K6I8Hv4K6adOm\/pIpvgwC2CY22gp2CybsDvvDDmW7EGlMkd0Wft7F1rF5bL8Re1sOTCP2Wh3rzIHw6KM\/iG6RcZusWCFvIc1jBmbs2LGFJFWamhNoDAVku43RT9LyUALVsttDa62PPTkw9dEPTaUFB9S0adOit62KXZM3G4xTTz3Vli9fHkV1d3cbMzAjRoyI9vVHBEIRwP6Ktdk4PXmz6SHbzUZFYSEJYHuxHRa7Jm9IXapZlhyYatJWXSUT4NkhMvMWEq+g8\/xQW1sbQf2KEohALQnIdmtJX3U3MwE5MM3cu03WNp7F4C2kJ554wngQusmap+Y0MQHZbhN3rppWMwJyYCqOXhWIgAiIgAiIgAiEJiAHJjRRlScCIiACIiACIlA+gX5KkAPTDyBFi4AIiIAIiIAI1B8BOTD11yfSSAREQAREoPYEpEGdE5ADU+cdJPVEQAREQAREQAQOJyAH5nAmChEBERCB2hOQBiIgAnkJyIHJi0eRIiACIiACIiAC9UhADkw99op0EoHaE5AGIiACIlDXBOTA1HX3SDkREAEREAEREIFsBOTAZKOisNoTkAYiIAIiIAIikIeAHJg8cBQlAiIgAiIgAiJQnwTkwGTvF4WKgAiIgAiIgAjUMQE5MHXcOVJNBERABERABBqLQPW0lQNTPdaqSQREQAREQAREIBABOTCBQKoYERABERCB2hOQBq1DQA5M6\/S1WioCIiACIiACTUNADkzTdKUaIgIiUHsC0kAERKBaBOTAVIu06hEBERABERABEQhGQA5MMJQqSARqT0AaiIAIiECrEJAD0yo9rXaKgAiIgAiIQBMRkAPTRJ1Z+6ZIAxEQAREQARGoDgE5MNXhrFpEQAREQAREQAQCEmgqByYgFxUlAiIgAiIgAiJQxwTkwNRx50g1ERABERABEagCgYasQg5MQ3ablBYBERABERCB1iYgB6a1+1+tFwEREIHaE5AGIlACATkwJUBTFhEQAREQAREQgdoSkANTW\/6qXQREoPYEpIEIiEADEpAD04CdJpVFQAREQAREoNUJyIFpdQtQ+2tPQBqIgAiIgAgUTUAOTNHIlEEEREAEREAERKDWBOTA1LoHal+\/NBABERABERCBhiMgB6bhukwKi4AIiIAIiIAI1N6BUR+IgAiIgAiIgAiIQJEE5MAUCUzJRUAEREAERKAeCLS6DnJgWt0C1H4REAEREAERaEACcmAasNOksgiIgAjUnoA0EIHaEpADU1v+ql0EREAEREAERKAEAnJgSoCmLCIgArUnIA1EQARam4AcmNbuf7VeBERABERABBqSgByYhuw2KV17AtJABERABESglgTkwNSSvuoWAREQAREQAREoiYAcmJKw1T6TNBABERABERCBViYgB6aVe19tFwEREAEREIEGJVCiA9OgrZXaIiACIiACIiACTUFADkxTdKMaIQIiIAIi0BAEpGQwAnJggqFUQSIgAiIgAiIgAtUiIAemWqRVjwiIgAjUnoA0EIGmISAHpmm6Ug0RAREQAREQgdYhIAemdfpaLRWB2hOQBiIgAiIQiIAcmEAgVYwIiIAIiIAIiED1CMiBqR5r1VR7AtJABERABESgSQjIgWmSjlQzREAEREAERKCVCMiBqWZvqy4REAEREAEREIEgBOTABMGoQkRABERABERABCpFIFu5cmCyUVGYCIiACIiACIhAXROQA1PX3SPlREAEREAEatfp0nsAAARNSURBVE9AGtQjATkw9dgr0kkEREAEREAERCAvATkwefEoUgREQARqT0AaiIAIHE5ADszhTBQiAiIgAiIgAiJQ5wTkwNR5B0k9Eag9AWkgAiIgAvVHQA5M\/fWJNBIBERABERABEeiHgByYfgApuvYEpIEIiIAIiIAIZBKQA5NJRPsiIAIiIAIiIAJ1T0AOTL9dpAQiIAIiIAIiIAL1RkAOTL31iPQRAREQAREQgWYgUOE2yIGpMGAVLwIiIAIiIAIiEJ6AHJjwTFWiCIiACIhA7QlIgyYnIAemyTtYzRMBERABERCBZiQgB6YZe1VtEgERqD0BaSACIlBRAnJgKopXhYuACIiACIiACFSCgByYSlBVmSJQewLSQAREQASamoAcmKbuXjVOBERABERABJqTgByY5uzX2rdKGoiACIiACIhABQnIgakgXBUtAiIgAiIgAiJQGQLN6sBUhpZKFQEREAEREAERqAsCcmDqohukhAiIgAiIgAjUA4HG0UEOTOP0lTQVAREQAREQARHoJSAHpheEViIgAiIgArUnIA1EoFACcmAKJaV0IiACIiACIiACdUNADkzddIUUEQERqD0BaSACItAoBOTANEpPSU8REAEREAEREIE+AnJg+lBoQwRqT0AaiIAIiIAIFEZADkxhnJRKBERABERABESgjgjIgamjzqi9KtJABERABERABBqDgByYxugnaSkCIiACIiACIpAgUFcOTEIvbYqACIiACIiACIhATgJyYHKiUYQIiIAIiIAINASBllRSDkxLdrsaLQIiIAIiIAKNTUAOTGP3n7QXAREQgdoTkAYiUAMCcmBqAF1VioAIiIAIiIAIlEdADkx5\/JRbBESg9gSkgQiIQAsSkAPTgp2uJouACIiACIhAoxOQA9PoPSj9a09AGoiACIiACFSdgByYqiNXhSIgAiIgAiIgAuUSkANTLsHa55cGIiACIiACItByBOTAtFyXq8EiIAIiIAIi0PgEyndgGp+BWiACIiACIiACItBgBOTANFiHSV0REAEREIHmIKBWlEdADkx5\/JRbBERABERABESgBgTkwNQAuqoUAREQgdoTkAYi0NgE5MA0dv9JexEQAREQARFoSQJyYFqy29VoEag9AWkgAiIgAuUQkANTDj3lFQEREAEREAERqAkBOTA1wa5Ka09AGoiACIiACDQyATkwjdx70l0EREAEREAEWpSAHJgadbyqFQEREAEREAERKJ2AHJjS2SmnCIiACIiACIhAdQn01SYHpg+FNkRABERABERABBqFgByYRukp6SkCIiACIlB7AtKgbgjIgambrpAiIiACIiACIiAChRKQA1MoKaUTAREQgdoTkAYiIAK9BOTA9ILQSgREQAREQAREoHEIyIFpnL6SpiJQewLSQAREQATqhIAcmDrpCKkhAiIgAiIgAiJQOAE5MIWzUsraE5AGIiACIiACIhAR+D8AAAD\/\/+VMXicAAAAGSURBVAMATVQbDWMu2sUAAAAASUVORK5CYII=","height":337,"width":560}} +%--- +%[output:6e6a25b7] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,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","height":337,"width":560}} +%--- +%[output:6f426dbf] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,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","height":337,"width":560}} +%--- +%[output:677e5848] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,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","height":337,"width":560}} +%--- diff --git a/neural-pde-surrogates-for-3d-electrostatics/solveBushingElectrostatic.m b/neural-pde-surrogates-for-3d-electrostatics/solveBushingElectrostatic.m new file mode 100644 index 0000000..fa534c7 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/solveBushingElectrostatic.m @@ -0,0 +1,124 @@ +function [R,model] = solveBushingElectrostatic(nFins, finRadiusBase, finRadiusTop, finWidth, tubeRadiusBase, tubeRadiusTop, boreRadius, totalLength, options) +%solveBushingElectrostatic Electrostatic analysis of a parametric transformer bushing. +% [R, model] = solveBushingElectrostatic(...) creates a parametric +% bushing geometry, embeds it in an air domain, applies electrostatic +% boundary conditions, and solves. Replicates the workflow from the +% MathWorks documentation example "Electrostatic Analysis of Transformer +% Bushing Insulator". +% +% Inputs: +% nFins - Number of fin-like rings +% finRadiusBase - Outer radius of the first (bottom) fin +% finRadiusTop - Outer radius of the last (top) fin +% finWidth - Axial width of each fin +% tubeRadiusBase - Tube outer radius at the bottom +% tubeRadiusTop - Tube outer radius at the top +% boreRadius - Inner bore radius +% totalLength - Total axial length of the bushing +% +% Name-Value Arguments: +% DoPlot - Plot results. Default: false +% BushingGeometry - fegeometry object to use instead of creating the +% bushing from parameters. Default: [] +% +% Outputs: +% R - ElectrostaticResults object from solve() +% model - femodel object with geometry, mesh, and material properties +% +% Boundary conditions: +% - Inner bore surface: Voltage = 10 kV (conductor) +% - Flat annular ring: Voltage = 0 V (oil tank ground) +% +% Material properties: +% - Air domain (Cell 1): Relative permittivity = 1 +% - Bushing insulator (Cell 2): Relative permittivity = 5 + + arguments + nFins + finRadiusBase + finRadiusTop + finWidth + tubeRadiusBase + tubeRadiusTop + boreRadius + totalLength + options.DoPlot = false + options.BushingGeometry = [] + end + + %% Step 1: Create the parametric bushing geometry (or use provided one) + if isempty(options.BushingGeometry) + [gmBushing, ~] = createBushing(nFins, finRadiusBase, finRadiusTop, ... + finWidth, tubeRadiusBase, tubeRadiusTop, ... + boreRadius, totalLength); + else + gmBushing = options.BushingGeometry; + end + + %% Step 2: Create the air domain surrounding the bushing + % Fixed-size air cuboid matching the documentation example dimensions + % (1, 0.4, 0.4) but with the long axis along Z to match our bushing. + % Centered in XY, translated so the bushing (Z=0 to totalLength) is + % fully enclosed. + gmAir = fegeometry(multicuboid(0.4, 0.4, 1)); + gmAir = translate(gmAir, [0, 0, totalLength/2 - 0.5]); + + %% Step 3: Combine air + bushing + % Cell 1 = air, Cell 2 = bushing insulator + gmModel = addCell(gmAir, gmBushing); + + %% Step 4: Find boundary condition faces in the combined geometry + % Face IDs change after addCell, so re-detect using probe points. + + % Bore face: probe at the bore radius, mid-length + boreFace = nearestFace(gmModel, [boreRadius, 0, totalLength/2]); + + % Flat annular face: at the peak of the last fin + % The last fin center is at totalLength - endStub, where endStub = 0.06*totalLength + lastFinZ = totalLength - 0.06 * totalLength; + % Probe at mid-radius between tube and fin + tubeRAtSplit = tubeRadiusBase + (tubeRadiusTop - tubeRadiusBase) * lastFinZ / totalLength; + midR = (tubeRAtSplit + finRadiusTop) / 2; + annularFace = nearestFace(gmModel, [midR, 0, lastFinZ]); + + %% Step 5: Set up the electrostatic model + model = femodel(AnalysisType="electrostatic", Geometry=gmModel); + model.VacuumPermittivity = 8.8541878128E-12; + + % Material properties + model.MaterialProperties(1) = materialProperties(RelativePermittivity=1); % air + model.MaterialProperties(2) = materialProperties(RelativePermittivity=5); % insulator + + % Boundary conditions + model.FaceBC(boreFace) = faceBC(Voltage=10E3); % 10 kV on conductor + model.FaceBC(annularFace) = faceBC(Voltage=0); % ground at oil tank + + %% Step 6: Mesh and solve + model = generateMesh(model, Hmax=0.025); + R = solve(model); + + %% Step 7 (optional): Visualize results on the bushing + if options.DoPlot + elemsBushing = findElements(R.Mesh, "Region", Cell=2); + + figure('Position', [100 100 1400 500]) + tiledlayout(1, 2) + + % Electric potential + nexttile + pdeplot3D(R.Mesh.Nodes, R.Mesh.Elements(:, elemsBushing), ... + ColorMapData=R.ElectricPotential); + title('Electric Potential (V)') + colorbar + + % Electric field magnitude + nexttile + Emag = sqrt(R.ElectricField.Ex.^2 + R.ElectricField.Ey.^2 + ... + R.ElectricField.Ez.^2); + pdeplot3D(R.Mesh.Nodes, R.Mesh.Elements(:, elemsBushing), ... + ColorMapData=Emag); + title('Electric Field Magnitude (V/m)') + colorbar + end + +end diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+io/loadAllData.m b/neural-pde-surrogates-for-3d-electrostatics/src/+io/loadAllData.m new file mode 100644 index 0000000..504ffec --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+io/loadAllData.m @@ -0,0 +1,11 @@ +function data = loadAllData(dataRootDir) + metadata = dir(fullfile(dataRootDir,"*.mat")); + fname = fullfile(dataRootDir,{metadata.name}); + + function data = loadData(fname) + data = load(fname); + data = data.D; + end + + data = arrayfun(@loadData, fname, UniformOutput = true); +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/edge_to_node_accumulation.m b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/edge_to_node_accumulation.m new file mode 100644 index 0000000..8ad63d4 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/edge_to_node_accumulation.m @@ -0,0 +1,9 @@ +function x = edge_to_node_accumulation(x,xe,B) + % B is the node-edge adjacency, x the node features and xe the edge + % features. x is H x NumNodes and xe is H x NumEdges. + % Requires hidden size to be same for nodes and edges so that we can add x + % and xe. + % xe * B' accumulates edge features onto nodes. + xe = xe*B'; + x = x+xe; +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/fourierPositionalEmbedding.m b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/fourierPositionalEmbedding.m new file mode 100644 index 0000000..39432dc --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/fourierPositionalEmbedding.m @@ -0,0 +1,15 @@ +function x = fourierPositionalEmbedding(x,maxfreq) + % This function extends an input x with "Fourier features", + % cos(f*x) and sin(f*x) + % for frequencies f = 1:maxfreq. + % + % There's various references to this sort of thing in literature. + % + % Assume x is CxN and in [-1,1] + f = x*pi; + F = permute(1:maxfreq,[1,3,2]); + f = f.*F; + f = permute(f,[1,3,2]); + f = reshape(f,[],size(x,2)); + x = cat(1,x,cos(f),sin(f)); +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/glorot.m b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/glorot.m new file mode 100644 index 0000000..c0591f1 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/glorot.m @@ -0,0 +1,6 @@ +function W = glorot(nin,nout) + % Glorot initialization + scale = sqrt(6/(nin+nout)); + W = 2*rand(nout,nin) - 1; + W = scale*W; +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/initialize.m b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/initialize.m new file mode 100644 index 0000000..8fd4d43 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/initialize.m @@ -0,0 +1,9 @@ +function p = initialize(cfg) + % Initialization function for the mgn. + % + % cfg is a struct of configuration data. + p.node_encoder = meshgraphnet.initialize_mlp(cfg.NodeInputSize*(2*cfg.MaxFreq+1),cfg.HiddenSize,cfg.NumLayersEncoder); + p.edge_encoder = meshgraphnet.initialize_mlp(cfg.EdgeInputSize,cfg.HiddenSize,cfg.NumLayersEncoder); + p.processors = meshgraphnet.initialize_processor(cfg); + p.node_decoder = meshgraphnet.initialize_mlp(cfg.HiddenSize,cfg.OutputSize,cfg.NumLayersDecoder); +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/initialize_mlp.m b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/initialize_mlp.m new file mode 100644 index 0000000..8e99544 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/initialize_mlp.m @@ -0,0 +1,10 @@ +function p = initialize_mlp(inputSize,hiddenSize,numLayers) + % Initializer for the weights used in mlp. + p = struct(); + sz = [inputSize, repelem(hiddenSize,numLayers)]; + for i = 1:numLayers + p.("fc_"+i).W = meshgraphnet.glorot(sz(i),sz(i+1)); + p.("fc_"+i).b = zeros(sz(i+1),1); + end + p = dlupdate(@dlarray,p); +end diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/initialize_processor.m b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/initialize_processor.m new file mode 100644 index 0000000..b638921 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/initialize_processor.m @@ -0,0 +1,11 @@ +function p = initialize_processor(cfg) + % Initialization function for the message passing processor layers. + % + % cfg is a struct of configuration data. + p = struct(); + for i = 1:cfg.NumProcessors + p.("node_processor_"+i) = meshgraphnet.initialize_mlp(cfg.HiddenSize,cfg.HiddenSize,cfg.NumLayersProcessor); + p.("edge_processor_"+i) = meshgraphnet.initialize_mlp(cfg.HiddenSize,cfg.HiddenSize,cfg.NumLayersProcessor); + end + p = dlupdate(@dlarray,p); +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/mesh2adjacency.m b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/mesh2adjacency.m new file mode 100644 index 0000000..fb49547 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/mesh2adjacency.m @@ -0,0 +1,32 @@ +function A = mesh2adjacency(mesh) + % This function assumes mesh is an FEMesh for 3d data and of quadratic + % order. + + numNodes = size(mesh.Nodes,2); + + % https://uk.mathworks.com/help/pde/ug/mesh-data.html + % Specify the edges from the above webpage. + r = [1,5,2,6,3,10,4,8,1,7,2,9]; + c = [5,2,6,3,10,4,8,1,7,3,9,4]; + + % Get the node indices of those edges from the elements. + R = mesh.Elements(r,:); + C = mesh.Elements(c,:); + + % Construct a sparse adjacency matrix. + V = ones(size(R)); + R = reshape(R,[],1); + C = reshape(C,[],1); + V = reshape(V,[],1); + A = sparse(R,C,V,numNodes,numNodes); + + % Add self loops - we'll typically use those in a GNN as a node can feed + % features back to itself. In principle you could also just add residual + % connections around any graph convolution or node-node message pass. + A = A + speye(numNodes); + + % Symmetrise the adjacency (we didn't include the other direction of edges + % in the r,c above) and make sure all the entries are 0 or 1 (in case we + % double counted anywhere). + A = double((A>0) | (A'>0)); +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/meshgraphnet.m b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/meshgraphnet.m new file mode 100644 index 0000000..249bcf4 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/meshgraphnet.m @@ -0,0 +1,33 @@ +function x = meshgraphnet(p,x,xe,B,cfg) + % A mesh graph network. + + % Add Fourier position embeddings. + x = meshgraphnet.fourierPositionalEmbedding(x,cfg.MaxFreq); + + % Encode node and edge features with MLPs. + x = meshgraphnet.mlp(p.node_encoder,x,cfg.NumLayersEncoder); + xe = meshgraphnet.mlp(p.edge_encoder,xe,cfg.NumLayersEncoder); + + % Loop over "processors" - message passing layers between edges and nodes. + z = x; + for i = 1:cfg.NumProcessors + + % Accumulate node features onto edges + xe = meshgraphnet.node_to_edge_accumulation(x,xe,B); + + % Apply MLP to edge representations + xe = meshgraphnet.mlp(p.processors.("edge_processor_"+i),xe,cfg.NumLayersProcessor); + + % Accumulate edge features onto nodes. + x = meshgraphnet.edge_to_node_accumulation(x,xe,B); + + % Apply MLP to node representations. + x = meshgraphnet.mlp(p.processors.("node_processor_"+i),x,cfg.NumLayersProcessor); + end + + % Residual connection around the processors. + x = x+z; + + % Node decoder. + x = meshgraphnet.mlp(p.node_decoder,x,cfg.NumLayersDecoder); +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/minibatch.m b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/minibatch.m new file mode 100644 index 0000000..7af1acc --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/minibatch.m @@ -0,0 +1,35 @@ +function [X,Xe,B,Y] = minibatch(X,Xe,B,Y,useGpu) + % Batch together a cell of node features, edge features, node-edge + % adjacencies, and node targets. + % + % The main interesting thing is that you compute the blkdiag of adjacencies + % to batch them and stack together the node/edge and batch dimensions for the + % other data. + + if useGpu + X = gpuArray(dlarray(cat(2,X{:}))); + else + X = dlarray(cat(2,X{:})); + end + B = blkdiag(B{:}); + + + % Normalize by num edges connected to a node. This helps prevent the model + % "blowing up" initially (suppose B==1 for all edges connected to a node, + % then for a node with many edges connected to it, the edge-to-node + % accumulation will sum up a bunch of randomly initialized vectors, and + % this tends to output very large values -- or else very large variance. So + % normalizing by the number of edges connected to a node helps mitigate + % this. This is standard, though different normalizations are used by + % different GNN methods, e.g. see the GCN examples which take a slightly + % different approach). + B = B ./ sum(B,2); + if useGpu + Xe = gpuArray(dlarray(cat(2,Xe{:}))); + Y = gpuArray(dlarray(cat(2,Y{:}))); + B = gpuArray(B); + else + Xe = dlarray(cat(2,Xe{:})); + Y = dlarray(cat(2,Y{:})); + end +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/mlp.m b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/mlp.m new file mode 100644 index 0000000..ae1766a --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/mlp.m @@ -0,0 +1,19 @@ +function x = mlp(params,x,numLayers) + % A simple multi-layer perceptron. + for i = 1:(numLayers-1) + p = params.("fc_"+i); + z = pagemtimes(p.W,x) + p.b; + z = gelu(z); + if i > 1 + % Add a residual connection. This is possible when x and z have the + % same size, which is always the case after the first layer in the + % below setup. + x = x + z; + else + x = z; + end + end + p = params.("fc_"+numLayers); + z = pagemtimes(p.W,x) + p.b; + x = x+z; +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/node_to_edge_accumulation.m b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/node_to_edge_accumulation.m new file mode 100644 index 0000000..853e0af --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/node_to_edge_accumulation.m @@ -0,0 +1,14 @@ +function xe = node_to_edge_accumulation(x,xe,B) + % B is the node-edge adjacency, x the node features and xe the edge + % features. x is H x NumNodes and xe is H x NumEdges. + % Requires hidden size to be same for nodes and edges so that we can add x + % and xe. + % x*B accumulates node features onto the edges (i.e. if edge i is defined + % by nodes j and k, then + % (x*B)(:,i) == x(:,j) + x(:,k) + % more or less (B may be normalized, e.g. by the number of edges connected + % to a node, so in practice there are coefficients in front of the terms + % on the RHS). + x = x*B; + xe = x + xe; +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/preprocessData.m b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/preprocessData.m new file mode 100644 index 0000000..f6d8f7d --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+meshgraphnet/preprocessData.m @@ -0,0 +1,60 @@ +function [X,Xe,Y,A,B] = preprocessData(data,keep) + + % Set coordinates as node features. + % Note: the mesh graph network paper suggests to not use coordinates as + % node features, as many physics problems will be translation invariant, + % and coordinate node features would cause overfitting. Instead they + % propose to include only relative positional information such as the + % displacement vectors between nodes as edge features (added below). + % However we have no other node features (epsilon is constant and equal to + % 5 on the transformer bushing), so I include the coordinate features so we + % have some node feature. + % We could add a node feature that is 1 if the node is on the boundary of + % the transformer bushing. I've been interested in adding geometric + % features too, such as curvatures of the 2d surface of the object at the + % node. + + % Only consider the nodes designated by "keep" + if isempty(keep) + keep = 1:size(data.Mesh.Nodes,2); + end + X = data.Mesh.Nodes(:,keep); + if isfield(data,'Potential') + Y = data.Potential(keep)'; + else + Y = []; % don't return targets + end + + % Get the adjacency matrix. + A = meshgraphnet.mesh2adjacency(data.Mesh); + + % Only retain the nodes we designated to keep. + A = A(keep,:); + A = A(:,keep); + + % Now construct the "node-edge adjacency", a matrix B of size + % NumNodes x NumEdges + % where B(i,j) = 1 if node i is one of the two nodes attached to edge j. + % + % This will be used in mesh graph net type networks that accumulate edge + % features onto nodes and node features onto edges. + % + % We recompute R,C here to include the self loops and the edges from + % symmetrisation above. + [r,c] = find(A); + rc = [r;c]; + + % Number the edges 1 -> numEdges following R and C + edgeLabels = [(1:numel(r))';(1:numel(r))']; + + % Construct the sparse node-edge adjacency. + B = sparse(rc,edgeLabels,ones(size(rc)),size(A,1),numel(r)); + B = double(B>0); + + % Construct edge features. Here I just use displacements and lengths. + % Angles could also be included. + % When dealing with multiple materials, you could + % include a feature for crossing a boundary. + Xe = X(:,r) - X(:,c); + Xe = [Xe; sqrt(sum(Xe(1:3,:).^2,1))]; +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+transolver/+datastore/RandomAugmentationDatastore.m b/neural-pde-surrogates-for-3d-electrostatics/src/+transolver/+datastore/RandomAugmentationDatastore.m new file mode 100644 index 0000000..a3e7913 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+transolver/+datastore/RandomAugmentationDatastore.m @@ -0,0 +1,42 @@ +classdef RandomAugmentationDatastore < matlab.io.Datastore & ... + matlab.io.datastore.Subsettable + properties + InputFeatures + Index + end + methods + function this = RandomAugmentationDatastore(inputFeatures) + this.InputFeatures = inputFeatures; + this.Index = 1; + end + function tf = hasdata(this) + tf = this.Index<=size(this.InputFeatures,3); + end + function [data,info] = read(this) + data = this.InputFeatures(:,:,this.Index); + % Random rotation around Z-axis (valid for axisymmetric geometry) + theta = 360*rand(); + ct = cosd(theta); st = sind(theta); + R = [ct -st 0; st ct 0; 0 0 1]; + data(1:3,:,:) = R * data(1:3,:,:); + % Random reflection across x (valid for axisymmetric geometry) + if rand() > 0.5 + data(1,:,:) = -data(1,:,:); + end + this.Index = this.Index + 1; + info = []; + end + function reset(this) + this.Index = 1; + end + end + methods(Access=protected) + function n = maxpartitions(this) + n = size(this.InputFeatures,3); + end + function subds = subsetByReadIndices(this,idx) + subds = copy(this); + subds.InputFeatures = subds.InputFeatures(:,:,idx); + end + end +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+transolver/padAndCreateDatastores.m b/neural-pde-surrogates-for-3d-electrostatics/src/+transolver/padAndCreateDatastores.m new file mode 100644 index 0000000..bf34f66 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+transolver/padAndCreateDatastores.m @@ -0,0 +1,18 @@ +function [cds,cdsVal] = padAndCreateDatastores(... + normalizedInputsTrain,normalizedInputsVal,... + normalizedTargetsTrain,normalizedTargetsVal) + [paddedInputsTrain,maskTrain] = padsequences(normalizedInputsTrain,2); + [paddedInputsVal,maskVal] = padsequences(normalizedInputsVal,2); + paddedTargetsTrain = padsequences(normalizedTargetsTrain,2); + paddedTargetsVal = padsequences(normalizedTargetsVal,2); + + inputDsTrain = transolver.datastore.RandomAugmentationDatastore(paddedInputsTrain); + maskDsTrain = arrayDatastore(maskTrain(1,:,:),IterationDimension=3); + targetDsTrain = arrayDatastore(paddedTargetsTrain,IterationDimension = 3); + inputDsVal = arrayDatastore(paddedInputsVal,IterationDimension=3); + maskDsVal = arrayDatastore(maskVal(1,:,:),IterationDimension=3); + targetDsVal = arrayDatastore(paddedTargetsVal,IterationDimension=3); + + cds = combine(inputDsTrain,maskDsTrain,targetDsTrain,maskDsTrain); + cdsVal = combine(inputDsVal,maskDsVal,targetDsVal,maskDsVal); +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+transolver/preprocessData.m b/neural-pde-surrogates-for-3d-electrostatics/src/+transolver/preprocessData.m new file mode 100644 index 0000000..1a8058e --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+transolver/preprocessData.m @@ -0,0 +1,35 @@ +function [inputs,targets,efieldTargets] = preprocessData(data,stride) + inputs = cell(numel(data),1); + targets = cell(numel(data),1); + efieldTargets = cell(numel(data),1); + + function [input,target,efield] = extractInputAndTarget(data,stride) + + input = cat(1, ... + data.Coord,... + data.Epsilon'); + + % target is the potential + target = data.Potential'; + + % electric field target (3 x N) + efield = data.ElectricField; + + % Find any nodes with a NaN input, target, or E-field. Remove them. + nanInput = isnan(input); + nanTarget = isnan(target); + nanEfield = isnan(efield); + nanNode = any(nanInput,1) | any(nanTarget,1) | any(nanEfield,1); + input(:,nanNode) = []; + target(:,nanNode) = []; + efield(:,nanNode) = []; + input = input(:,1:stride:end); + target = target(:,1:stride:end); + efield = efield(:,1:stride:end); + end + + % Loop over data + for i = 1:numel(data) + [inputs{i},targets{i},efieldTargets{i}] = extractInputAndTarget(data(i),stride); + end +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+transolver/transolver.m b/neural-pde-surrogates-for-3d-electrostatics/src/+transolver/transolver.m new file mode 100644 index 0000000..100d1b4 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+transolver/transolver.m @@ -0,0 +1,60 @@ +function net = transolver(args) +arguments + args.NumLayers (1,1) double {mustBePositive,mustBeInteger} = 2 + args.NumHeads (1,1) double {mustBePositive,mustBeInteger} = 1 + args.HiddenSize (1,1) double {mustBePositive,mustBeInteger} = 64 + args.NumSlices (1,1) double {mustBePositive,mustBeInteger} = 32 + args.OutputSize (1,1) double {mustBePositive,mustBeInteger} = 1 + args.IncludeTemperature (1,1) logical = false + args.Dropout (1,1) double = 0 + args.Rezero (1,1) logical = false + args.NormalizationLayer = @layerNormalizationLayer +end + +transolvers = repmat(transolverBlock(args.NumHeads,args.HiddenSize,args.NumSlices,args.IncludeTemperature,args.Dropout,args.Rezero,args.NormalizationLayer), args.NumLayers, 1); +for i = 1:args.NumLayers + transolvers(i).Name = "transolver_"+i; +end + +layers = [ + convolution1dLayer(1,args.HiddenSize) + transolvers + args.NormalizationLayer() + convolution1dLayer(1,args.OutputSize)]; +net = dlnetwork(layers,Initialize = false); +net = addLayers(net,identityLayer(Name="mask")); +for i = 1:args.NumLayers + net = connectLayers(net,"mask", "transolver_"+i+"/mask"); +end +end + +function block = transolverBlock(numHeads,hiddenSize,numSlices,temp,dropout,rezero,norm) +block = [ + identityLayer(Name="X") + norm() + transolver.transolverLayer(hiddenSize,numSlices,numHeads,Name="transolver",IncludeTemperature=temp) + dropoutLayer(dropout) + rezeroLayer(hiddenSize) + additionLayer(2,Name="add1") + norm() + convolution1dLayer(1,hiddenSize*4) + geluLayer + convolution1dLayer(1,hiddenSize) + dropoutLayer(dropout)]; + +if rezero + block = [block; rezeroLayer(hiddenSize)]; +end +block = [block;additionLayer(2,Name="add2")]; +block = dlnetwork(block,Initialize=false); +block = addLayers(block,identityLayer(Name="mask")); +block = connectLayers(block,"mask","transolver/mask"); +block = connectLayers(block,"X","add1/in2"); +block = connectLayers(block,"add1","add2/in2"); +block = networkLayer(block); +end + +function layer = rezeroLayer(hiddenSize) + +layer = convolution1dLayer(1,hiddenSize,WeightsInitializer="zeros",BiasLearnRateFactor=0); +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+transolver/transolverLayer.m b/neural-pde-surrogates-for-3d-electrostatics/src/+transolver/transolverLayer.m new file mode 100644 index 0000000..b1ddcb9 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+transolver/transolverLayer.m @@ -0,0 +1,73 @@ +function layer = transolverLayer(hiddenSize,numSlices,numHeads,args) +arguments + hiddenSize (1,1) {mustBePositive,mustBeInteger} + numSlices (1,1) {mustBePositive,mustBeInteger} + numHeads (1,1) {mustBePositive,mustBeInteger} + args.Name (1,1) string = "" + args.IncludeTemperature (1,1) logical = false +end + + +layers = [ + identityLayer(Name="X") + convolution1dLayer(1,numSlices)]; +if args.IncludeTemperature + layers = [layers + functionLayer(@(x,temp) x./max(temp,0.01), InputNames=["x","temp"],Acceleratable=true,Name="temp")]; +end +layers = [layers + softmaxLayer(Name="sliceWeights") + sliceRepresentationLayer(Name="sliceRepresentation") + selfAttentionLayer(numHeads,hiddenSize) + nodeRepresentationLayer(Name="nodeRepresentation") + convolution1dLayer(1,hiddenSize)]; + +net = dlnetwork(layers,Initialize=false); +net = addLayers(net,identityLayer(Name="mask")); +net = connectLayers(net,"mask","sliceRepresentation/mask"); +net = connectLayers(net,"X","sliceRepresentation/X"); +net = connectLayers(net,"sliceWeights","nodeRepresentation/W"); + +if args.IncludeTemperature + % see proj_temperature here: https://github.com/thuml/Transolver_plus/blob/main/models/Transolver_plus.py + net = addLayers(net,[convolution1dLayer(1,numSlices,Name="tempProjectionIn"); geluLayer; convolution1dLayer(1,1); geluLayer(); convolution1dLayer(1,1,WeightLearnRateFactor=0,Weights=1,Bias=0.5,Name="tempProjection")]); + net = connectLayers(net,"X","tempProjectionIn"); + net = connectLayers(net,"tempProjection","temp/temp"); +end +layer = networkLayer(net,Name=args.Name); +end + +function layer = sliceRepresentationLayer(args) +arguments + args.Name (1,1) string = "" +end +layer = functionLayer(@sliceLinear,... + InputNames=["W","X","mask"],... + Formattable=true,... + Acceleratable=true,... + Name=args.Name); +end + +function Z = sliceLinear(W,X,mask) +W = W.*mask; +W = stripdims(W,"CSB"); +Z = pagemtimes(W,stripdims(X)); +Z = Z ./ (sum(W,2)+1e-6); +Z = dlarray(Z,"SCB"); +end + +function layer = nodeRepresentationLayer(args) +arguments + args.Name (1,1) string = "" +end +layer = functionLayer(@nodeLinear,... + Formattable=true,... + Acceleratable=true,... + InputNames=["Z","W"],... + Name=args.Name); +end + +function Z = nodeLinear(Z,W) +Z = pagemtimes(stripdims(W), stripdims(Z)); +Z = dlarray(Z,"SCB"); +end diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+utils/applyInverseScale.m b/neural-pde-surrogates-for-3d-electrostatics/src/+utils/applyInverseScale.m new file mode 100644 index 0000000..6c8ec6e --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+utils/applyInverseScale.m @@ -0,0 +1,11 @@ +function out = applyInverseScale(data,offset,scale) + +rowMap = @(X) X.*scale + offset; + +if iscell(data) + out = cellfun(@(X) rowMap(X), data, 'UniformOutput', false); +else + out = rowMap(data); +end + +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+utils/applyScale.m b/neural-pde-surrogates-for-3d-electrostatics/src/+utils/applyScale.m new file mode 100644 index 0000000..f894e18 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+utils/applyScale.m @@ -0,0 +1,11 @@ +function out = applyScale(data,offset,scale) + +rowMap = @(X) (X-offset)./scale; + +if iscell(data) + out = cellfun(@(X) rowMap(X), data, 'UniformOutput', false); +else + out = rowMap(data); +end + +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+utils/computeScaleStats.m b/neural-pde-surrogates-for-3d-electrostatics/src/+utils/computeScaleStats.m new file mode 100644 index 0000000..40ebe7a --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+utils/computeScaleStats.m @@ -0,0 +1,59 @@ +function [offset,scale] = computeScaleStats(data, rangeSpec, varargin) +%computeScaleStats Compute row-wise scaling statistics for [0,1] or [-1,1]. +% [offset, scale] = computeScaleStats(data, rangeSpec) computes per-row +% offset and scale so that scaled = (data - offset) ./ scale maps each +% row to the specified range. +% +% Inputs: +% data - 1xK cell array (each DxNi) or matrix (DxN) +% rangeSpec - Target range: '[0,1]' or '[-1,1]' +% +% Name-Value Arguments: +% CoordIdx - Indices of coordinate rows (e.g., 1:3). Default: [] +% CoordPolicy - 'none' (default) or 'zero-center-uniform' +% +% Outputs: +% offset - Dx1 offset vector +% scale - Dx1 scale vector +% +% Note: 'zero-center-uniform' sets offset to 0 and uses a uniform scale +% factor across coordinate rows to preserve aspect ratio. + +p = inputParser; +addParameter(p, 'CoordIdx', [], @(x) isnumeric(x)); +addParameter(p, 'CoordPolicy', 'none', @(s) ismember(s, {'none','zero-center-uniform'})); +parse(p, varargin{:}); + +coordIdx = p.Results.CoordIdx; +coordPol = p.Results.CoordPolicy; + +% Concatenate training samples along columns +if iscell(data) + allMat = cat(2, data{:}); % [D x sum(Ni)] +else + allMat = data; % [D x N] +end + +% Concatenate training samples along columns +rowMin = min(allMat, [], 2); +rowMax = max(allMat, [], 2); + +switch rangeSpec +case '[0,1]' + offset = rowMin; + scale = rowMax - rowMin; +case '[-1,1]' + offset = 0.5 * (rowMin + rowMax); % midrange + scale = 0.5 * (rowMax - rowMin); % half-range +otherwise + error('rangeSpec must be "[0,1]" or "[-1,1]".'); +end + +% Optional coordinate policy +if ~isempty(coordIdx) && strcmp(coordPol, 'zero-center-uniform') + offset(coordIdx) = 0; % zero-center coordinates + s = max(scale(coordIdx)); % uniform scaling across x,y,z + scale(coordIdx) = s; +end + +end diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+utils/convertGpuToCpu.m b/neural-pde-surrogates-for-3d-electrostatics/src/+utils/convertGpuToCpu.m new file mode 100644 index 0000000..481951c --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+utils/convertGpuToCpu.m @@ -0,0 +1,18 @@ +function out = convertGpuToCpu(in) + if isa(in, 'dlarray') || isa(in, 'gpuArray') + out = gather(in); + elseif isstruct(in) + % Apply recursively to each field + out = struct(); + fields = fieldnames(in); + for i = 1:numel(fields) + out.(fields{i}) = utils.convertGpuToCpu(in.(fields{i})); + end + elseif iscell(in) + % Apply recursively to each cell element + out = cellfun(@utils.convertGpuToCpu, in, 'UniformOutput', false); + else + % Leave other data types unchanged + out = in; + end +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+utils/electricFieldFromTets.m b/neural-pde-surrogates-for-3d-electrostatics/src/+utils/electricFieldFromTets.m new file mode 100644 index 0000000..2120a23 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+utils/electricFieldFromTets.m @@ -0,0 +1,212 @@ +function out = electricFieldFromTets(nodes, elements, V) +%electricFieldFromTets Compute electric field from nodal potentials on tetrahedra. +% out = electricFieldFromTets(nodes, elements, V) computes E = -grad(V) +% on Tet4 or Tet10 meshes from MATLAB PDE Toolbox. +% For Tet4: grad(V) is constant per element. +% For Tet10: grad(V) is linear per element, evaluated at quadrature +% points (for element values) and at each node's barycentric position +% (for nodal values). +% +% Inputs: +% nodes - Node coordinates (3xN or Nx3) +% elements - Tetrahedral connectivity (4xNe/10xNe or Nex4/Nex10) +% V - Nodal electric potential (Nx1 or 1xN) +% +% Outputs: +% out.E_node - Nx3 nodal electric field [Ex Ey Ez] +% out.E_elem - Nex3 element electric field (quadrature average +% for Tet10, constant for Tet4) +% out.Emag_node - Nx1 nodal electric field magnitude +% out.Emag_elem - Nex1 element electric field magnitude + + % Normalize inputs + if size(nodes,1)==3 && size(nodes,2)~=3, X = nodes.'; else, X = nodes; end + V = V(:); % ensure column vector + + % Normalize elements to Ne x nen + [r,c] = size(elements); + if r==4 || r==10, T = elements.'; + elseif c==4 || c==10, T = elements; + else, error('elements must be 4- or 10-node tetrahedral connectivity.'); + end + [Ne, nen] = size(T); + if nen~=4 && nen~=10, error('Only Tet4 and Tet10 supported.'); end + + N = size(X,1); + + E_node_sum = zeros(N, 3); + E_node_cnt = zeros(N, 1); + + if nen == 4 + % Tet4 version + ids = T'; % 4 x Ne + + % Build [1 x y z] matrices for all elements: 4 x 4 x Ne + Xe_all = reshape(X(ids(:),:), 4, Ne, 3); + Xe_all = permute(Xe_all, [1,3,2]); % 4 x 3 x Ne + A_all = cat(2, ones(4,1,Ne), Xe_all); % 4 x 4 x Ne + + % Batch solve A \ I for all elements + I4 = repmat(eye(4), 1, 1, Ne); + C_all = pagemldivide(A_all, I4); % 4 x 4 x Ne + dN_all = C_all(2:4,:,:); % 3 x 4 x Ne + + % Potentials at element nodes: 4 x 1 x Ne + Ve_all = reshape(V(ids(:)), 4, Ne); + Ve_all = reshape(Ve_all, 4, 1, Ne); + + % Gradient of V for all elements: 3 x 1 x Ne + gradV_all = pagemtimes(dN_all, Ve_all); + E_elem = -squeeze(gradV_all)'; % Ne x 3 + + % Accumulate element E-field to nodes + for d = 1:3 + vals = repmat(E_elem(:,d)', 4, 1); % 4 x Ne + E_node_sum(:,d) = accumarray(ids(:), vals(:), [N,1]); + end + E_node_cnt = accumarray(ids(:), 1, [N,1]); + + else + % Tet10 version + % Nodal barycentric coordinates in SF canonical order + nodeRST = [ + 1.0 0.0 0.0; % N1 (corner) + 0.0 1.0 0.0; % N2 (corner) + 0.0 0.0 1.0; % N3 (corner) + 0.0 0.0 0.0; % N4 (corner, L4=1) + 0.5 0.5 0.0; % N5: edge 1-2 + 0.0 0.5 0.5; % N6: edge 2-3 + 0.5 0.0 0.5; % N7: edge 3-1 + 0.5 0.0 0.0; % N8: edge 1-4 + 0.0 0.5 0.0; % N9: edge 2-4 + 0.0 0.0 0.5]; % N10: edge 3-4 + + % 4-point quadrature for element-average + a = 0.5854101966249685; b = 0.1381966011250105; + quadPts = [a b b; b a b; b b a; b b b]; + + % PDE Toolbox convention: nodes 1-4 are corners, 5-10 are mid-edge. + % Reorder mid-edge nodes to match canonical SF edge ordering. + cornerIdx = T(:,1:4); % Ne x 4 + midIdx = T(:,5:10); % Ne x 6 + + % Canonical edges connecting corners + edgePairs = [1 2; 2 3; 3 1; 1 4; 2 4; 3 4]; + + % Compute expected midpoints of canonical edges for all elements + cornerCoords = reshape(X(cornerIdx(:),:), Ne, 4, 3); % Ne x 4 x 3 + targets = zeros(Ne, 6, 3); + for k = 1:6 + targets(:,k,:) = 0.5*(cornerCoords(:,edgePairs(k,1),:) + ... + cornerCoords(:,edgePairs(k,2),:)); + end + + % Actual mid-edge node coordinates + midCoords = reshape(X(midIdx(:),:), Ne, 6, 3); % Ne x 6 x 3 + + % Pairwise squared distances between mid-nodes and target midpoints + dist2 = sum((reshape(midCoords,Ne,6,1,3) - ... + reshape(targets,Ne,1,6,3)).^2, 4); % Ne x 6 x 6 + + % Match each canonical target to the closest mid-edge node + [~, assignment] = min(dist2, [], 2); % Ne x 1 x 6 + assignment = squeeze(assignment); % Ne x 6 + if Ne == 1, assignment = assignment'; end % handle single-element case + + % Reorder mid-edge nodes to canonical SF order + rowIdx = repmat((1:Ne)', 1, 6); + reorderedMid = midIdx(sub2ind([Ne,6], rowIdx, assignment)); + ids_sf_all = [cornerIdx, reorderedMid]; % Ne x 10 + + % Gather coordinates and potentials for all elements + Xe_all = reshape(X(ids_sf_all(:),:), Ne, 10, 3); + Xe_all = permute(Xe_all, [2,3,1]); % 10 x 3 x Ne + + Ve_all = reshape(V(ids_sf_all(:)), Ne, 10); + Ve_all = reshape(Ve_all', 10, 1, Ne); % 10 x 1 x Ne + + % Precompute shape function derivatives + dN_quad = zeros(10, 3, 4); + for q = 1:4 + [dNr,dNs,dNt] = tet10_shape_derivs( ... + quadPts(q,1), quadPts(q,2), quadPts(q,3)); + dN_quad(:,:,q) = [dNr dNs dNt]; + end + + dN_nodal = zeros(10, 3, 10); + for n = 1:10 + [dNr,dNs,dNt] = tet10_shape_derivs( ... + nodeRST(n,1), nodeRST(n,2), nodeRST(n,3)); + dN_nodal(:,:,n) = [dNr dNs dNt]; + end + + % Element-average E-field (4-point quadrature) + E_elem_acc = zeros(3, 1, Ne); + for q = 1:4 + dN = dN_quad(:,:,q); % 10 x 3 (fixed) + % Jacobian: J = Xe' * dN -> 3 x 3 x Ne + J = pagemtimes(permute(Xe_all,[2,1,3]), dN); + % Physical gradients: J' \ dN' -> 3 x 10 x Ne + dN_xyz = pagemldivide(permute(J,[2,1,3]), repmat(dN',1,1,Ne)); + % grad(V) = dN_xyz * Ve -> 3 x 1 x Ne + E_elem_acc = E_elem_acc - pagemtimes(dN_xyz, Ve_all); + end + E_elem = squeeze(E_elem_acc / 4)'; % Ne x 3 + if Ne == 1, E_elem = E_elem(:)'; end + + % Nodal E-field (accumulate from element contributions) + for n = 1:10 + dN = dN_nodal(:,:,n); % 10 x 3 (fixed) + J = pagemtimes(permute(Xe_all,[2,1,3]), dN); + dN_xyz = pagemldivide(permute(J,[2,1,3]), repmat(dN',1,1,Ne)); + gradV = pagemtimes(dN_xyz, Ve_all); % 3 x 1 x Ne + Enode = -squeeze(gradV)'; % Ne x 3 + if Ne == 1, Enode = Enode(:)'; end + + nids = ids_sf_all(:, n); + for d = 1:3 + E_node_sum(:,d) = E_node_sum(:,d) + ... + accumarray(nids, Enode(:,d), [N,1]); + end + E_node_cnt = E_node_cnt + accumarray(nids, 1, [N,1]); + end + end + + E_node = E_node_sum ./ max(E_node_cnt, 1); + Emag_node = sqrt(sum(E_node.^2, 2)); + Emag_elem = sqrt(sum(E_elem.^2, 2)); + + out.E_node = E_node; + out.E_elem = E_elem; + out.Emag_node = Emag_node; + out.Emag_elem = Emag_elem; + end + + % Tet10 shape function derivatives w.r.t. (r,s,t) + function [dNr, dNs, dNt] = tet10_shape_derivs(r, s, t) + L1=r; L2=s; L3=t; L4=1-r-s-t; + dL = [1 0 0 -1; % dLi/dr + 0 1 0 -1; % dLi/ds + 0 0 1 -1]; % dLi/dt + L = [L1; L2; L3; L4]; + + dNr = zeros(10,1); dNs = zeros(10,1); dNt = zeros(10,1); + + % Corner nodes: Ni = Li(2Li - 1), dNi/dLi = 4Li - 1 + for i = 1:4 + c = 4*L(i) - 1; + dNr(i) = c*dL(1,i); + dNs(i) = c*dL(2,i); + dNt(i) = c*dL(3,i); + end + + % Mid-edge nodes in canonical SF order: + % N5=4L1L2, N6=4L2L3, N7=4L3L1, N8=4L1L4, N9=4L2L4, N10=4L3L4 + edges = [1 2; 2 3; 3 1; 1 4; 2 4; 3 4]; + for k = 1:6 + i = edges(k,1); j = edges(k,2); + dNr(4+k) = 4*(dL(1,i)*L(j) + L(i)*dL(1,j)); + dNs(4+k) = 4*(dL(2,i)*L(j) + L(i)*dL(2,j)); + dNt(4+k) = 4*(dL(3,i)*L(j) + L(i)*dL(3,j)); + end + end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/src/+utils/trainTestSplit.m b/neural-pde-surrogates-for-3d-electrostatics/src/+utils/trainTestSplit.m new file mode 100644 index 0000000..a28acc7 --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/src/+utils/trainTestSplit.m @@ -0,0 +1,6 @@ +function [trainIdx,valIdx] = trainTestSplit(numObs,trainRatio) + idx = randperm(numObs); + numTrain = floor(trainRatio*numObs); + trainIdx = idx(1:numTrain); + valIdx = idx(numTrain+1:end); +end \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/startup.m b/neural-pde-surrogates-for-3d-electrostatics/startup.m new file mode 100644 index 0000000..eeec02b --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/startup.m @@ -0,0 +1,18 @@ +% Get the project root folder +projectRoot = fileparts(mfilename('fullpath')); + +% Add the project root to the path +addpath(projectRoot); + +% Add all subfolders to the path, create them if they don't exist +dirs = {'data','STL','results','src'}; +for k=1:numel(dirs) + d = fullfile(projectRoot, dirs{k}); + if ~exist(d, 'dir') + mkdir(d); + end + addpath(d); +end + +% Display a message +disp(['Project path set. Root: ', projectRoot]); \ No newline at end of file diff --git a/neural-pde-surrogates-for-3d-electrostatics/transolver_script.m b/neural-pde-surrogates-for-3d-electrostatics/transolver_script.m new file mode 100644 index 0000000..4cb22bd --- /dev/null +++ b/neural-pde-surrogates-for-3d-electrostatics/transolver_script.m @@ -0,0 +1,562 @@ +%[text] # Rapid Design Exploration for a 3D Electrostatics Problem Using Transolver +%[text] Accurately solving three-dimensional electrostatic problems is central to the design and analysis of many electrical and electromechanical systems, including transformer bushings, high-voltage insulation, and power electronics. While finite element analysis (FEA) provides high-fidelity solutions, repeatedly solving these problems across varying geometries can be computationally expensive, limiting its practicality for rapid design exploration, optimization, and real-time workflows. +%[text] This demo demonstrates how deep learning can be used to enable fast, full-field solutions of 3D electrostatic partial differential equations (PDEs). Using simulation data generated with MATLAB® PDE Toolbox™, we train a neural network to predict the electric potential field directly from geometry and material properties. The geometries vary across samples and are discretized with different node counts, making the problem particularly challenging for conventional neural network architectures that assume fixed grids or inputs. +%[text] To address this challenge, we use **Transolver** \[[1](internal:M_07b5)\], a transformer-based architecture designed for learning PDE solutions on unstructured meshes. Each mesh node is represented by its 3D coordinates and material properties (relative permittivity). A key innovation in Transolver is its "physics-attention" mechanism: rather than applying standard global attention across all nodes (which scales quadratically), Transolver softly assigns nodes to a fixed number of learned "physics slices" and applies attention within each slice. This enables scalable learning on large, irregular meshes commonly encountered in 3D FEA. The network learns a direct mapping from geometry and material information to the electric potential at every node, effectively acting as a learned PDE solver. +%[text] In this demo, we apply Transolver to a 3D electrostatics problem involving a transformer bushing insulator embedded in air. The trained model is evaluated on unseen geometries, and its predictions are compared against FEA solutions. We also show how to use the trained model to make inferences when no high-fidelity simulation data is present. +%[text] ![](text:image:68f9) +%[text:tableOfContents]{"heading":"Table of Contents"} +%[text] +%% +%[text] **Experiment Settings** +%[text] The function [`canUseGPU`](https://www.mathworks.com/help/releases/R2026a/matlab/ref/canusegpu.html) is used here to detect and enable GPU acceleration for training and inference. Set `doTrain` to `true` to train a new model; otherwise, specify the name of a pretrained model to load. Set `doFinetune` to `true` to fine-tune the trained model with gradient regularization. +useGpu = canUseGPU; % use GPU if one is available +doTrain = true; % set to true to train the model, else load in a pre-trained network %[control:checkbox:999a]{"position":[11,15]} +doFinetune = true; % fine-tune with autodiff gradient regularizer %[control:checkbox:6714]{"position":[14,18]} +if doTrain + ts = string(datetime("now",Format="uuuuMMdd'T'HHmmss")); % specify timestamp to uniquely name results file +else + loadFile = "TransolverFinetune_XXXXXXXXXXXXXXX"; % name of results file (if loading a pretrained model) +end +projectRoot = findProjectRoot("startup.m"); +outdir = fullfile(projectRoot,"results"); +%% +%[text] ## Load Data +%[text] Load simulation data generated using MATLAB PDE Toolbox. The simulation data was generated by varying a parameterized geometry inspired by the one in the documentation example [Electrostatic Analysis of Transformer Bushing Insulator](https://www.mathworks.com/help/pde/ug/electrostatic-analysis-of-transformer-bushing-insulator.html). Data files are expected to be located in the `data/` subdirectory within the parent directory of this script. +%[text] Each file contains a struct with FEA data for that particular observation. The struct includes the following fields: +%[text] - `Coord`: 3 x $N$ array of node coordinates in the FEA mesh. Used as model input. +%[text] - `Epsilon`: $N$ x 1 array of relative permittivity values at each node (1 for air, 5 for bushing insulator). Used as model input. +%[text] - `Potential`: $N$ x 1 array of electric potential values at each node. Used as training targets. +%[text] - `ElectricField`: 3 x $N$ array of electric field components ($E\_x$, $E\_y$, $E\_z$) at each node. Used as training targets during the optional fine-tuning step. +%[text] - `Mesh`: [`FEMesh`](https://www.mathworks.com/help/pde/ug/pde.femesh.html) object representing the FEA mesh. Used for visualization. \ +%[text] Note that $N$ (number of nodes) varies between samples, as each simulation uses a different geometry and therefore mesh. +if doTrain + datadir = fullfile("data"); + data = io.loadAllData(datadir); +else + % Skip data loading when loading a pretrained model +end +%% +%[text] ## Preprocess Data +%[text] Extract input features and targets from the raw simulation data. The helper function `preprocessData` subsamples every 10th node in each mesh and extracts the node coordinates and relative permittivity, resulting in 4 input features per node. This subsampling reduces computational cost while preserving the geometric and material structure of the problem. +if doTrain + [inputs,targets,efieldTargets] = transolver.preprocessData(data,10); % efieldTargets only used for fine-tuning step + clear data % free memory once raw data is no longer needed +%[text] After preprocessing, `inputs` is a $S$ x 1 cell array, where the $i$th cell contains a 4 x $N\_i$ matrix of input features for sample $i$, and `targets` is a $S$ x 1 cell array, where the $i$th cell contains a 1 x $N\_i$ vector of electric potential values. Here $S$ is the number of samples, and $N\_i$ is the (subsampled) number of nodes for sample $i$. +%[text] Split the dataset into 80% training and 20% validation subsets. + numObs = numel(inputs); + [idxTrain,idxVal] = utils.trainTestSplit(numObs,0.8); + inputsTrain = inputs(idxTrain); inputsVal = inputs(idxVal); + targetsTrain = targets(idxTrain); targetsVal = targets(idxVal); + cfg.IdxVal = idxVal; % Save validation indices for later evaluation +%[text] Normalize inputs and targets. Apply feature-wise min-max scaling to map values to \[-1,1\]. +%[text] For the coordinate features $(x,y,z)$, the coordinates are first zero-centered and then scaled using a single uniform scale factor. This preserves the aspect ratio of the geometry and avoids mesh distortion. +%[text] Relative permittivity and target potential values are scaled independently. + [inputOffset,inputScale] = utils.computeScaleStats(inputsTrain, '[-1,1]', CoordIdx=1:3, CoordPolicy='zero-center-uniform'); + [outputOffset,outputScale] = utils.computeScaleStats(targetsTrain, '[-1,1]'); + normalizedInputsTrain = utils.applyScale(inputsTrain,inputOffset,inputScale); + normalizedInputsVal = utils.applyScale(inputsVal,inputOffset,inputScale); + normalizedTargetsTrain = utils.applyScale(targetsTrain,outputOffset,outputScale); + normalizedTargetsVal = utils.applyScale(targetsVal,outputOffset,outputScale); +%[text] Save the normalization parameters (`inputOffset`, `inputScale`, `outputOffset`, `outputScale`) for use during inference. + cfg.InputOffset = inputOffset; cfg.InputScale = inputScale; + cfg.OutputOffset = outputOffset; cfg.OutputScale = outputScale; +%[text] Pad variable-size samples and create datastores. To enable batch processing, all samples are padded to the same number of nodes. Dummy nodes are added where necessary. A corresponding mask is created so that padded nodes are ignored during training and loss computation. The training datastore also applies random data augmentation (rotation around the Z-axis and reflection across X) to exploit the axial symmetry of the bushing geometry and increase the effective training set size. + [cds,cdsV] = transolver.padAndCreateDatastores(... + normalizedInputsTrain,normalizedInputsVal,... + normalizedTargetsTrain,normalizedTargetsVal); +else + % Skip preprocessing when loading a pretrained model +end +%% +%[text] ## Define Network Architecture +%[text] Construct a Transolver model (implemented in `transolver.m` and `transolverLayer.m`). +%[text] In other transformer-based PDE surrogate architectures, each mesh node is assigned its own token, and standard self-attention is applied across all tokens. This scales quadratically: if $N$ is the number of mesh nodes, global attention has $\\mathcal{O}(N^2)$ complexity, which becomes prohibitive for large 3D meshes. +%[text] Transolver reduces this cost by introducing "slices": +%[text] 1. Project $N$ node representations into $S$ slices ($S\\ll N$), +%[text] 2. Apply attention over the slice representations, +%[text] 3. Map the updated slice information back to the original $N$ nodes. \ +%[text] This reduces the dominant attention cost from $\\mathcal{O}(N^2)$ to approximately $\\mathcal{O}(NS+S^2)$, enabling scalability to large, irregular meshes. +%[text] The key hyperparameters of Transolver are: +%[text] - `NumLayers`: number of transformer blocks, +%[text] - `HiddenSize`: feature dimension/width of the model, +%[text] - `NumSlices`: number of slices, +%[text] - `NumHeads`: number of attention heads. \ +if doTrain + cfg.ModelName = "Transolver"; % save relevant information in cfg struct + numLayers = 4; cfg.NumLayers = numLayers; + inputSize = size(normalizedInputsTrain{1},1); cfg.InputSize = inputSize; + outputSize = 1; cfg.OutputSize = outputSize; + hiddenSize = 128; cfg.HiddenSize = hiddenSize; + numSlices = 8; cfg.NumSlices = numSlices; + numHeads = 4; cfg.NumHeads = numHeads; + cfg.Notes = "Predict 3D electric potential from electrostatic FEA data. Geometry varies across samples; other simulation settings (e.g. boundary conditions) are held fixed."; + + net = transolver.transolver(NumLayers=numLayers,... + NumHeads=numHeads,... + HiddenSize=hiddenSize,... + NumSlices=numSlices,... + OutputSize=outputSize,... + IncludeTemperature=true,... + Dropout=0.1,... + Rezero=false,... + NormalizationLayer=@layerNormalizationLayer); +%[text] Initialize the network. + sz = size(normalizedInputsTrain{1}); + net = initialize(net, ... + networkDataLayout(sz, "CS"), networkDataLayout([1, sz(2)],"CS")); +else + % Skip network construction when loading a pretrained model +end +%% +%[text] ## Define the Model Loss Function +%[text] Define custom loss functions for training the model. Different loss formulations can be used depending on the desired tradeoff between point-wise accuracy and global field characteristics. +%[text] Mean squared error (MSE) loss: The function `modelLoss` computes the MSE between the predicted and true electric potential values. Padded nodes are ignored using a binary mask. +function loss = modelLoss(Ypred,Y,mask) + totalNonPadding = sum(mask,"all"); + loss = sum(((Ypred - Y).*mask).^2,"all") / totalNonPadding; +end +%[text] Weighted loss with variance penalty: In addition to minimizing point-wise prediction error, the variance term penalizes differences between the predicted and true variance of the electric potential field. +function loss = weightedLoss(Ypred,Y,mask) + alpha = 0.05; % variance penalty weight + totalNonPadding = sum(mask,"all"); + mseLoss = sum(((Ypred - Y).*mask).^2,"all") / totalNonPadding; + Ypred_mean = sum(Ypred.*mask,"all") / totalNonPadding; + Y_mean = sum(Y.*mask,"all") / totalNonPadding; + Ypred_var = sum(((Ypred - Ypred_mean).*mask).^2,"all") / totalNonPadding; + Y_var = sum(((Y - Y_mean).*mask).^2,"all") / totalNonPadding; + variance_loss = (Ypred_var - Y_var)^2; + loss = (1-alpha)*mseLoss + alpha*variance_loss; +end +%% +%[text] ## Set the Training Options and Train the Network +%[text] Train the Transolver model using the Adam optimizer and selected loss function. +%[text] Configure training options using `trainingOptions`: +%[text] - `Optimizer`: `"adam"` for adaptive learning rate optimization +%[text] - `MaxEpochs`: `3000` training epochs (number of passes through the training data) +%[text] - `MiniBatchSize`: `6` samples per mini-batch +%[text] - `ValidationData`: `cdsV` (validation datastore). +%[text] - `Plots`: `"training-progress"` to visualize training and validation metrics +%[text] - `InputDataFormats: ["CSB","CSB"]` (channel-spatial-batch) +%[text] - `TargetDataFormats`: `["CSB","CSB"]` (channel-spatial-batch) +%[text] - `ExecutionEnvironment`: `"gpu"` or `"cpu"`, based on `useGpu` +%[text] - `OutputNetwork`: `"best-validation"` to save the model with the lowest validation loss +%[text] - `LearnRateSchedule`: `"cosine"` learning rate decay \ +if doTrain %[output:group:3c8d5695] + % Select execution environment + if useGpu + exEnv = "gpu"; + else + exEnv = "cpu"; + end + opts = trainingOptions("adam",... + MaxEpochs=3000,... + Plots="training-progress",... + ValidationData=cdsV,... + InputDataFormats=["CSB","CSB"],... + TargetDataFormats=["CSB","CSB"],... + MiniBatchSize=6,... + ExecutionEnvironment=exEnv,... + OutputNetwork="best-validation",... + LearnRateSchedule="cosine"); + myFun = @weightedLoss; cfg.LossName = "WeightedMSEVariance"; + [net,info] = trainnet(cds,net,myFun,opts); %[output:3610ba2c] %[output:6122d9b1] + cfg.TrainingInfo = info; + % Save trained model and configuration + mName = cfg.ModelName; + baseName = strjoin([mName, ts], "_"); + outpath = fullfile(outdir,baseName + ".mat"); + save(outpath,"net","cfg"); +else + % Skip training when loading a pretrained model +end %[output:group:3c8d5695] +%% +%[text] ## Fine-Tune with Gradient Regularization +%[text] Fine-tune the trained (or loaded) model using a custom training loop with a gradient regularizer. The regularizer penalizes discrepancies between $-\\nabla V$ (computed via [`dlgradient`](https://www.mathworks.com/help/deeplearning/ref/dlarray.dlgradient.html)) and the reference electric field, encouraging the model to produce spatially smooth potential predictions whose derived electric field matches the FEA results. +%[text] Use a low learning rate and small weight $\\beta$ to gently regularize without degrading the pretrained potential accuracy. +if ~doTrain + fName = fullfile(outdir, loadFile + ".mat"); + S = load(fName,"net","cfg"); + net = S.net; cfg = S.cfg; + fprintf("Loaded pretrained model: %s (OutputSize=%d)\n",loadFile,cfg.OutputSize); +end +if doFinetune %[output:group:53bf9cbe] + beta = 0.001; % gradient regularizer weight + finetuneEpochs = 500; % fewer epochs needed for fine-tuning + finetuneLearnRate = 1e-4; % lower than training from scratch + gradClipThreshold = 1; % element-wise gradient clipping + miniBatchSize = 6; +%[text] Load and preprocess data for fine-tuning. If the model was just trained, the electric field targets are already in the workspace. Otherwise, reload the data and extract the electric field targets. + if ~doTrain + datadir = fullfile("data"); + data = io.loadAllData(datadir); + [inputs,targets,efieldTargets] = transolver.preprocessData(data,10); + clear data + + numObs = numel(inputs); + idxVal = cfg.IdxVal; + idxTrain = setdiff(1:numObs,idxVal); + + inputOffset = cfg.InputOffset; inputScale = cfg.InputScale; + outputOffset = cfg.OutputOffset; outputScale = cfg.OutputScale; + normalizedInputsTrain = utils.applyScale(inputs(idxTrain),inputOffset,inputScale); + normalizedInputsVal = utils.applyScale(inputs(idxVal),inputOffset,inputScale); + normalizedTargetsTrain = utils.applyScale(targets(idxTrain),outputOffset,outputScale); + normalizedTargetsVal = utils.applyScale(targets(idxVal),outputOffset,outputScale); + end +%[text] Normalize E-field targets to the same normalized coordinate/potential space. Pad to 4 channels with zeros in the 4th channel to match the input shape. + coordScale = inputScale(1:3); + normalizedEfieldTrain = cellfun(@(E) [E .* coordScale ./ outputScale; zeros(1,size(E,2))], efieldTargets(idxTrain), UniformOutput=false); + normalizedEfieldVal = cellfun(@(E) [E .* coordScale ./ outputScale; zeros(1,size(E,2))], efieldTargets(idxVal), UniformOutput=false); +%[text] Pad variable-size samples for batch processing. + [paddedInputsTrain,maskTrain] = padsequences(normalizedInputsTrain,2); + [paddedInputsVal,maskVal] = padsequences(normalizedInputsVal,2); + paddedTargetsTrain = padsequences(normalizedTargetsTrain,2); + paddedTargetsVal = padsequences(normalizedTargetsVal,2); + paddedEfieldTrain = padsequences(normalizedEfieldTrain,2); + paddedEfieldVal = padsequences(normalizedEfieldVal,2); + maskTrain = maskTrain(1,:,:); + maskVal = maskVal(1,:,:); +%[text] Fine-tuning uses a [custom training loop](https://www.mathworks.com/help/deeplearning/custom-training-loops.html) because the gradient regularization loss requires calling `dlgradient` to differentiate the predicted potential with respect to the network inputs (*x,y,z*). This autodiff operation is not supported by `trainnet`, which only differentiates with respect to the learnable parameters. The loss function computes: $\\mathcal{L} = (1-\\beta)\\mathcal{L}\_V + \\beta\\mathcal{L}\_E$, where $\\mathcal{L}\_V$ is the MSE on predicted potential and $\\mathcal{L}\_E$ is the MSE between the autodiff-computed spatial gradient $-\\nabla V$ and the reference electric field. +%[text] The custom loss function `gradRegularizedLoss` used in the fine-tuning step is defined at the end of the script. + if useGpu + net = dlupdate(@gpuArray,net); + end + + numTrain = size(paddedInputsTrain,3); + numVal = size(paddedInputsVal,3); + numIterationsPerEpoch = ceil(numTrain / miniBatchSize); + numIterations = finetuneEpochs * numIterationsPerEpoch; + + % Prepare validation data + XVal = dlarray(single(paddedInputsVal),"CSB"); + mVal = dlarray(single(maskVal),"CSB"); + VVal = dlarray(single(paddedTargetsVal),"CSB"); + EVal = dlarray(single(paddedEfieldVal),"CSB"); + if useGpu + XVal = gpuArray(XVal); mVal = gpuArray(mVal); + VVal = gpuArray(VVal); EVal = gpuArray(EVal); + end + validationFrequency = 25; + + % Initialize Adam optimizer state + avgGrad = []; + avgSqGrad = []; + + % Training progress + lossHistory = zeros(numIterations,1); + valLossHistory = nan(numIterations,1); + monitor = trainingProgressMonitor(Metrics=["TrainingLoss","ValidationLoss"],Info="Epoch"); %[output:5b7527e4] + groupSubPlot(monitor,"Loss",["TrainingLoss","ValidationLoss"]); + + % Track best validation network + bestValLoss = Inf; + bestNet = []; + + iteration = 0; + for epoch = 1:finetuneEpochs + shuffleIdx = randperm(numTrain); + + for batch = 1:numIterationsPerEpoch + iteration = iteration + 1; + + s = (batch-1)*miniBatchSize + 1; + e = min(batch*miniBatchSize, numTrain); + bIdx = shuffleIdx(s:e); + + XBatch = dlarray(single(paddedInputsTrain(:,:,bIdx)),"CSB"); + mBatch = dlarray(single(maskTrain(:,:,bIdx)),"CSB"); + VBatch = dlarray(single(paddedTargetsTrain(:,:,bIdx)),"CSB"); + EBatch = dlarray(single(paddedEfieldTrain(:,:,bIdx)),"CSB"); + + if useGpu + XBatch = gpuArray(XBatch); + mBatch = gpuArray(mBatch); + VBatch = gpuArray(VBatch); + EBatch = gpuArray(EBatch); + end + + [loss,gradients] = dlfeval(@gradRegularizedLoss,net,XBatch,mBatch,VBatch,EBatch,beta); + + % Clip gradients + gradients = dlupdate(@(g) min(max(g,-gradClipThreshold),gradClipThreshold),gradients); + + % Cosine learning rate schedule + lr = finetuneLearnRate * 0.5 * (1 + cos(pi * iteration / numIterations)); + + [net,avgGrad,avgSqGrad] = adamupdate(net,gradients,avgGrad,avgSqGrad,iteration,lr); + + lossHistory(iteration) = double(extractdata(loss)); + recordMetrics(monitor,iteration,TrainingLoss=lossHistory(iteration)); %[output:5b7527e4] + end + + % Validation loss (potential-only MSE) + if mod(epoch,validationFrequency) == 0 + valLoss = 0; + numValBatches = ceil(numVal / miniBatchSize); + for vb = 1:numValBatches + vs = (vb-1)*miniBatchSize + 1; + ve = min(vb*miniBatchSize, numVal); + vIdx = vs:ve; + Vpred = predict(net,XVal(:,:,vIdx),mVal(:,:,vIdx),InputDataFormats=["CSB","CSB"]); + batchMask = mVal(:,:,vIdx); + batchTarget = VVal(:,:,vIdx); + nNonPad = sum(batchMask,"all"); + valLoss = valLoss + double(extractdata(sum(((Vpred - batchTarget).*batchMask).^2,"all") / nNonPad)) * numel(vIdx); + end + valLoss = valLoss / numVal; + valLossHistory(iteration) = valLoss; + recordMetrics(monitor,iteration,ValidationLoss=valLoss); + + if valLoss < bestValLoss + bestValLoss = valLoss; + bestNet = net.Learnables; + end + end + + updateInfo(monitor,Epoch=epoch + " of " + finetuneEpochs); + + if monitor.Stop + break; + end + end + + % Restore best validation network + if ~isempty(bestNet) + net.Learnables = bestNet; + end + + % Save fine-tuned model + if ~doTrain + ts = string(datetime("now",Format="uuuuMMdd'T'HHmmss")); + end + cfg.Beta = beta; + cfg.LossName = "GradRegularizedMSE"; + cfg.Notes = cfg.Notes + " Fine-tuned with autodiff gradient regularizer (beta=" + beta + ")."; + % Replace the original trainnet TrainingInfo with the fine-tuning + % loss history, mimicking the trainnet convention. + iters = (1:iteration)'; + epochs = ceil(iters / numIterationsPerEpoch); + lrs = finetuneLearnRate * 0.5 .* (1 + cos(pi * iters / numIterations)); + trainHist = table(iters,epochs,lrs,lossHistory(1:iteration), ... + VariableNames=["Iteration","Epoch","LearnRate","Loss"]); + valIdx = find(~isnan(valLossHistory(1:iteration))); + valHist = table(valIdx,valLossHistory(valIdx), ... + VariableNames=["Iteration","Loss"]); + cfg.TrainingInfo = struct("TrainingHistory",trainHist,"ValidationHistory",valHist); + baseName = strjoin(["TransolverFinetune",ts],"_"); + outpath = fullfile(outdir,baseName + ".mat"); + if useGpu + net = dlupdate(@gather,net); + end + save(outpath,"net","cfg"); + fprintf("Saved fine-tuned model: %s\n",outpath); %[output:8369fad8] +end %[output:group:53bf9cbe] +%% +%[text] ## Test the Network +%[text] Evaluate the trained model on randomly selected samples from the validation set and compare predictions against the original FEA solutions. +%[text] Tip: To view the training progress plot on a logarithmic scale, open the Training Progress Monitor, hover over the plot and click the "Log scale y-axis" widget, like so: +%[text] ![](text:image:7aa2) +%[text] Choose a small number of random samples from the validation set to view the results and compare against FEA. +numTests = 3; % Must be less than numel(idxVal) +idxTest = randi([1,numel(cfg.IdxVal)],numTests,1); +for i=1:numTests %[output:group:92e2baba] + % Load the corresponding FEA results for the selected sample + fileIdx = cfg.IdxVal(idxTest(i)); + testFileName = sprintf("EM_results_%03d.mat",fileIdx); + load(testFileName) + % Prepare network inputs + U = [D.Coord; D.Epsilon']; + % Normalize inputs using training-set statistics + normU = utils.applyScale(U,cfg.InputOffset,cfg.InputScale); + % Predict electric potential + if useGpu + V = predict(net,dlarray(gpuArray(normU)),gpuArray(ones(1,size(normU,2))),InputDataFormats=["CS","CS"]); + V = extractdata(V); + V = gather(V); + else + V = predict(net,dlarray(normU),ones(1,size(normU,2)),InputDataFormats=["CS","CS"]); + V = extractdata(V); + end + % Rescale predictions back to physical units + V = utils.applyInverseScale(V,cfg.OutputOffset,cfg.OutputScale); + + % Visualization and error analysis + % Plot results on the transformer bushing only (exclude surrounding air) + elemsBushing = findElements(D.Mesh,"Region",Cell=2); + nodesBushing = findNodes(D.Mesh,"Region",Cell=2); + clims = [min(D.Potential), max(D.Potential)]; + figure(); %[output:8ee38f52] %[output:1ef43198] %[output:0bd82ac8] + tiledlayout(1,3); %[output:8ee38f52] %[output:1ef43198] %[output:0bd82ac8] + % FEA solution + nexttile; + pdeplot3D(D.Mesh.Nodes, ... + D.Mesh.Elements(:,elemsBushing), ... + ColorMapData=D.Potential); + title('FEA Solution'); + clim(clims); + % AI prediction + nexttile; + pdeplot3D(D.Mesh.Nodes, ... + D.Mesh.Elements(:,elemsBushing), ... + ColorMapData=V) + % Compute the relative L2 error on the bushing nodes + relErrorL2 = norm(V(nodesBushing)-D.Potential(nodesBushing))/norm(D.Potential(nodesBushing)); + title(sprintf('AI Prediction: Relative L2 error = %0.1e', relErrorL2)); + clim(clims); + % Plot the relative error + nexttile; + absError = abs(V-D.Potential); + relError = absError./max(D.Potential(nodesBushing)); + pdeplot3D(D.Mesh.Nodes, ... + D.Mesh.Elements(:,elemsBushing), ... + ColorMapData=relError); + title('Relative Error') + clim([min(relError(nodesBushing)) max(relError(nodesBushing))]) +end %[output:group:92e2baba] +%% +%[text] ## Make Inference on New Design +%[text] This section demonstrates how to use the trained Transolver model to make predictions on a new design for which no high-fidelity simulation data is available. Given a new geometry defined in an STL file, the model predicts the electric potential field directly from geometry and material properties. This enables rapid evaluation of new designs without running a full electrostatic simulation. +%[text] Load the STL file representing the transformer bushing insulator geometry. The STL file is assumed to be located in the `STL/` subdirectory. +gmBushing = fegeometry("STL/TEST_bushing.stl"); +%[text] Visualize the bushing insulator geometry. +figure(); %[output:6f7fe7c2] +pdegplot(gmBushing); %[output:6f7fe7c2] +title("New Bushing Insulator Geometry"); %[output:6f7fe7c2] +%[text] Define the simulation domain. In this example, the surrounding air is modeled as a cuboid domain. +gmAir = multicuboid(0.4,0.4,1); +gmAir = translate(gmAir,[0 0 -0.2]); +gmAir = fegeometry(gmAir); +% Combine air and bushing into a single geometry +gmModel = addCell(gmAir,gmBushing); +%[text] Visualize the combined geometry and inspect the cell labels. +figure(); %[output:4155c762] +pdegplot(gmModel,FaceAlpha=0.3,CellLabels="on"); %[output:4155c762] +title("Bushing Embedded in Air Domain") %[output:4155c762] +%[text] Create an electrostatic FEA model and assign relative permittivity values based on the cell labels. +model = femodel(AnalysisType="electrostatic",Geometry=gmModel); +%[text] From the cell labels: +%[text] - Cell 1: air +%[text] - Cell 2: bushing insulator \ +cellAir = 1; cellBushing = 2; +model.MaterialProperties(cellAir) = materialProperties(RelativePermittivity=1); +model.MaterialProperties(cellBushing) = materialProperties(RelativePermittivity=5); +model.VacuumPermittivity = 8.8541878128E-12; +%[text] Define points in the domain at which predictions are made. Do this by generating a mesh. +model = generateMesh(model,Hmax=0.025); +%[text] Prepare the inputs. +%[text] Extract node coordinates and assign relative permittivity values to each node based on its region. +elemsBushing = findElements(model.Mesh,"Region",Cell=cellBushing); +nodesBushing = findNodes(model.Mesh,"Region",Cell=cellBushing); +nodesAir = findNodes(model.Mesh,"Region",Cell=cellAir); +coord = model.Mesh.Nodes; +relPermVec = zeros(size(model.Mesh.Nodes,2),1); +relPermVec(nodesAir) = model.MaterialProperties(cellAir).RelativePermittivity; +relPermVec(nodesBushing) = model.MaterialProperties(cellBushing).RelativePermittivity; +%[text] Concatenate coordinates and material properties. +U = [coord; relPermVec']; +%[text] Normalize inputs using training-set statistics. +normU = utils.applyScale(U,cfg.InputOffset,cfg.InputScale); +%[text] Predict electric potential. +if useGpu + V = predict(net,dlarray(gpuArray(normU)),gpuArray(ones(1,size(normU,2))),InputDataFormats=["CS","CS"]); + V = extractdata(V); + V = gather(V); +else + V = predict(net,dlarray(normU),ones(1,size(normU,2)),InputDataFormats=["CS","CS"]); + V = extractdata(V); +end +%[text] Rescale predictions back to physical units. +V = utils.applyInverseScale(V,cfg.OutputOffset,cfg.OutputScale); +%[text] Visualize predicted electric potential on the bushing insulator. +figure(); %[output:2cf257b8] +pdeplot3D(model.Mesh.Nodes, ... %[output:2cf257b8] + model.Mesh.Elements(:,elemsBushing), ... %[output:2cf257b8] + ColorMapData=V); %[output:2cf257b8] + clim([min(V(nodesBushing)) max(V(nodesBushing))]); %[output:2cf257b8] + title('AI Predicted Electric Potential') %[output:2cf257b8] +%% +%[text] ## Derive the Electric Field from the Predicted Potential +%[text] In the FEA workflow, the electric field is derived from the potential: $\\mathbf{E}=-\\nabla V$. We follow the same approach here, computing the gradient of the AI-predicted potential using FEA shape function derivatives on the tetrahedral mesh (see the helper function `electricFieldFromTets.m` in `src/+utils`). The gradient regularization applied during fine-tuning encourages the model to produce spatially smooth potential fields, which improves the quality of the derived electric field. +out = utils.electricFieldFromTets(model.Mesh.Nodes,model.Mesh.Elements,V); +%[text] Visualize the derived electric field magnitude on the bushing insulator. +figure(); %[output:751ae428] +pdeplot3D(model.Mesh.Nodes, ... %[output:751ae428] + model.Mesh.Elements(:,elemsBushing), ... %[output:751ae428] + ColorMapData=out.Emag_node); %[output:751ae428] +clim([min(out.Emag_node(nodesBushing)) max(out.Emag_node(nodesBushing))]); %[output:751ae428] +title('Electric field magnitude derived from AI-predicted potential') %[output:751ae428] +%% +%[text] ## Helper Functions +function [loss,gradients] = gradRegularizedLoss(net,X,mask,Vtarget,Etarget,beta) +%gradRegularizedLoss Combined potential + E-field gradient loss. +% Computes MSE on predicted potential (weighted by 1-beta) plus MSE on +% the autodiff-derived spatial gradient vs. reference E-field (weighted +% by beta). + + V = forward(net,X,mask); + totalNonPadding = sum(mask,"all"); + + potLoss = sum(((V - Vtarget) .* mask).^2,"all") / totalNonPadding; + + dVdX = dlgradient(sum(V .* mask,"all"),X); + coordW = dlarray(single([1;1;1;0]),"C"); + gradLoss = sum(((-dVdX - Etarget) .* coordW .* mask).^2,"all") / (3 * totalNonPadding); + + loss = (1-beta) * potLoss + beta * gradLoss; + gradients = dlgradient(loss,net.Learnables); +end +%% +%[text] %[text:anchor:M_07b5] **References** +%[text] %[text:anchor:M_99d2] 1. Wu, Haixu, Huakun Luo, Haowen Wang, Jianmin Wang, and Mingsheng Long. "Transolver: A Fast Transformer Solver for PDEs on General Geometries." arXiv, June 1, 2024. [https://arxiv.org/abs/2402.02366](https://arxiv.org/abs/2402.02366). \ +%[text] + +%[appendix]{"version":"1.0"} +%--- +%[metadata:view] +% data: {"layout":"inline","rightPanelPercent":37.9} +%--- +%[text:image:68f9] +% data: {"align":"baseline","height":325,"src":"data:image\/png;base64,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","width":614} +%--- +%[control:checkbox:999a] +% data: {"defaultValue":false,"label":"doTrain","run":"Section"} +%--- +%[control:checkbox:6714] +% data: {"defaultValue":false,"label":"doFinetune","run":"Section"} +%--- +%[text:image:7aa2] +% data: {"align":"baseline","height":60,"src":"data:image\/png;base64,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","width":306} +%--- +%[output:3610ba2c] +% data: {"dataType":"text","outputData":{"text":" Iteration Epoch TimeElapsed LearnRate TrainingLoss ValidationLoss\n _________ _____ ___________ _________ ____________ ______________\n 0 0 00:00:04 0.001 0.8641\n 1 1 00:00:04 0.001 1.0057 \n 50 5 00:00:23 0.001 0.18358 0.17868\n 100 10 00:00:38 0.00099998 0.046259 0.032392\n 150 15 00:00:54 0.00099995 0.039501 0.028247\n 200 20 00:01:09 0.0009999 0.034234 0.025484\n 250 25 00:01:25 0.00099984 0.031304 0.024029\n 300 30 00:01:40 0.00099977 0.029805 0.023745\n 350 35 00:01:56 0.00099968 0.029051 0.023927\n 400 40 00:02:12 0.00099958 0.028234 0.023845\n 450 45 00:02:28 0.00099947 0.027786 0.023544\n 500 50 00:02:43 0.00099934 0.027524 0.02337\n 550 55 00:02:59 0.0009992 0.027175 0.023223\n 600 60 00:03:15 0.00099905 0.026688 0.02306\n 650 65 00:03:30 0.00099888 0.026607 0.022838\n 700 70 00:03:46 0.00099869 0.026256 0.022718\n 750 75 00:04:01 0.0009985 0.025787 0.022637\n 800 80 00:04:17 0.00099829 0.026074 0.022532\n 850 85 00:04:33 0.00099807 0.025555 0.022523\n 900 90 00:04:48 0.00099783 0.02557 0.022466\n 950 95 00:05:04 0.00099758 0.025465 0.022452\n 1000 100 00:05:20 0.00099731 0.025523 0.022514\n 1050 105 00:05:35 0.00099704 0.025474 0.022449\n 1100 110 00:05:51 0.00099674 0.025126 0.022467\n 1150 115 00:06:07 0.00099644 0.02525 0.022425\n 1200 120 00:06:22 0.00099612 0.025018 0.022545\n 1250 125 00:06:38 0.00099579 0.025051 0.022347\n 1300 130 00:06:54 0.00099544 0.024846 0.02235\n 1350 135 00:07:09 0.00099508 0.024801 0.022228\n 1400 140 00:07:25 0.00099471 0.024677 0.022391\n 1450 145 00:07:40 0.00099432 0.024685 0.022267\n 1500 150 00:07:56 0.00099392 0.024664 0.021987\n 1550 155 00:08:12 0.00099351 0.024409 0.021519\n 1600 160 00:08:27 0.00099308 0.023848 0.021653\n 1650 165 00:08:43 0.00099264 0.023978 0.021357\n 1700 170 00:08:58 0.00099219 0.023724 0.020857\n 1750 175 00:09:14 0.00099172 0.023756 0.02083\n 1800 180 00:09:30 0.00099124 0.023382 0.019927\n 1850 185 00:09:45 0.00099074 0.023754 0.019284\n 1900 190 00:10:01 0.00099023 0.023658 0.017162\n 1950 195 00:10:17 0.00098971 0.021858 0.014447\n 2000 200 00:10:32 0.00098918 0.023062 0.014413\n 2050 205 00:10:48 0.00098863 0.016667 0.019562\n 2100 210 00:11:03 0.00098806 0.017817 0.013884\n 2150 215 00:11:19 0.00098749 0.013461 0.012159\n 2200 220 00:11:34 0.0009869 0.012407 0.011151\n 2250 225 00:11:50 0.0009863 0.011914 0.010897\n 2300 230 00:12:05 0.00098568 0.012094 0.011447\n 2350 235 00:12:21 0.00098505 0.012403 0.012983\n 2400 240 00:12:37 0.00098441 0.011863 0.010779\n 2450 245 00:12:52 0.00098376 0.011736 0.010943\n 2500 250 00:13:08 0.00098309 0.012341 0.013512\n 2550 255 00:13:23 0.00098241 0.011818 0.014085\n 2600 260 00:13:39 0.00098171 0.011599 0.010518\n 2650 265 00:13:55 0.000981 0.013797 0.01211\n 2700 270 00:14:10 0.00098028 0.011736 0.010943\n 2750 275 00:14:26 0.00097954 0.011773 0.01074\n 2800 280 00:14:41 0.0009788 0.011806 0.011529\n 2850 285 00:15:01 0.00097804 0.013498 0.017845\n 2900 290 00:15:16 0.00097726 0.015635 0.011739\n 2950 295 00:15:32 0.00097647 0.015469 0.012332\n 3000 300 00:15:47 0.00097567 0.013075 0.01188\n 3050 305 00:16:03 0.00097486 0.012547 0.01048\n 3100 310 00:16:18 0.00097403 0.012402 0.010937\n 3150 315 00:16:34 0.00097319 0.011256 0.010809\n 3200 320 00:16:49 0.00097234 0.011095 0.010203\n 3250 325 00:17:05 0.00097148 0.014276 0.015333\n 3300 330 00:17:20 0.0009706 0.011081 0.011284\n 3350 335 00:17:36 0.00096971 0.010818 0.010908\n 3400 340 00:17:51 0.0009688 0.014645 0.014554\n 3450 345 00:18:07 0.00096789 0.012358 0.016823\n 3500 350 00:18:22 0.00096696 0.011328 0.011683\n 3550 355 00:18:38 0.00096601 0.010835 0.01056\n 3600 360 00:18:53 0.00096506 0.013 0.014451\n 3650 365 00:19:09 0.00096409 0.013144 0.010792\n 3700 370 00:19:24 0.00096311 0.013493 0.013569\n 3750 375 00:19:40 0.00096212 0.012964 0.012741\n 3800 380 00:19:55 0.00096111 0.010593 0.010798\n 3850 385 00:20:10 0.00096009 0.010195 0.010825\n 3900 390 00:20:26 0.00095906 0.010957 0.012233\n 3950 395 00:20:41 0.00095801 0.011773 0.013495\n 4000 400 00:20:57 0.00095696 0.0095422 0.011027\n 4050 405 00:21:12 0.00095589 0.010316 0.011294\n 4100 410 00:21:28 0.00095481 0.0096185 0.008988\n 4150 415 00:21:43 0.00095371 0.012751 0.012763\n 4200 420 00:21:59 0.00095261 0.0091142 0.0095776\n 4250 425 00:22:14 0.00095149 0.009287 0.0086816\n 4300 430 00:22:30 0.00095035 0.011615 0.009973\n 4350 435 00:22:45 0.00094921 0.0098641 0.010227\n 4400 440 00:23:01 0.00094805 0.012169 0.011243\n 4450 445 00:23:16 0.00094689 0.0082828 0.0082376\n 4500 450 00:23:32 0.00094571 0.0089179 0.0091716\n 4550 455 00:23:47 0.00094451 0.0089333 0.0084023\n 4600 460 00:24:03 0.00094331 0.007815 0.0087017\n 4650 465 00:24:18 0.00094209 0.011177 0.010054\n 4700 470 00:24:34 0.00094086 0.0090667 0.010204\n 4750 475 00:24:49 0.00093962 0.0070031 0.0082151\n 4800 480 00:25:05 0.00093837 0.006895 0.0092023\n 4850 485 00:25:20 0.0009371 0.0087184 0.0090327\n 4900 490 00:25:36 0.00093582 0.0071912 0.0065055\n 4950 495 00:25:51 0.00093453 0.0093862 0.018092\n 5000 500 00:26:07 0.00093323 0.0084042 0.009523\n 5050 505 00:26:22 0.00093192 0.0065816 0.0072109\n 5100 510 00:26:38 0.00093059 0.0085271 0.0077448\n 5150 515 00:26:53 0.00092926 0.006702 0.0072406\n 5200 520 00:27:09 0.00092791 0.0077569 0.0098281\n 5250 525 00:27:24 0.00092655 0.0065537 0.0061519\n 5300 530 00:27:40 0.00092517 0.0063353 0.0073275\n 5350 535 00:27:55 0.00092379 0.015452 0.012401\n 5400 540 00:28:11 0.00092239 0.0086239 0.0055275\n 5450 545 00:28:27 0.00092099 0.011424 0.0055557\n 5500 550 00:28:42 0.00091957 0.0062264 0.0059035\n 5550 555 00:28:58 0.00091814 0.0081577 0.0081625\n 5600 560 00:29:13 0.0009167 0.0087789 0.0061737\n 5650 565 00:29:29 0.00091524 0.0068302 0.0069394\n 5700 570 00:29:44 0.00091378 0.0081886 0.0062035\n 5750 575 00:30:00 0.0009123 0.0082499 0.0062724\n 5800 580 00:30:16 0.00091082 0.0066968 0.0062787\n 5850 585 00:30:32 0.00090932 0.0068112 0.0064778\n 5900 590 00:30:48 0.00090781 0.0095535 0.0085045\n 5950 595 00:31:04 0.00090629 0.0059705 0.0059136\n 6000 600 00:31:20 0.00090476 0.0063758 0.0078358\n 6050 605 00:31:36 0.00090321 0.005601 0.0053154\n 6100 610 00:31:52 0.00090166 0.0061005 0.0069498\n 6150 615 00:32:08 0.00090009 0.0070097 0.0069178\n 6200 620 00:32:24 0.00089852 0.0097288 0.0097095\n 6250 625 00:32:40 0.00089693 0.0054692 0.0053901\n 6300 630 00:32:55 0.00089533 0.0084084 0.0086272\n 6350 635 00:33:11 0.00089372 0.0059256 0.0058983\n 6400 640 00:33:27 0.0008921 0.0077387 0.0087663\n 6450 645 00:33:43 0.00089047 0.0060383 0.0054145\n 6500 650 00:33:59 0.00088883 0.0059485 0.0076511\n 6550 655 00:34:15 0.00088718 0.0067045 0.0075349\n 6600 660 00:34:31 0.00088552 0.0058125 0.0055066\n 6650 665 00:34:47 0.00088385 0.0067766 0.006873\n 6700 670 00:35:03 0.00088216 0.0068555 0.0085167\n 6750 675 00:35:19 0.00088047 0.0058325 0.0056718\n 6800 680 00:35:35 0.00087876 0.0066163 0.0057625\n 6850 685 00:35:50 0.00087705 0.0054057 0.0063899\n 6900 690 00:36:06 0.00087532 0.0060242 0.0077561\n 6950 695 00:36:22 0.00087359 0.0078927 0.0082319\n 7000 700 00:36:38 0.00087184 0.0060943 0.0053026\n 7050 705 00:36:54 0.00087009 0.0064059 0.0056959\n 7100 710 00:37:10 0.00086832 0.0061713 0.0074147\n 7150 715 00:37:26 0.00086654 0.0062149 0.006763\n 7200 720 00:37:42 0.00086476 0.0051645 0.0049459\n 7250 725 00:37:58 0.00086296 0.0085941 0.0063174\n 7300 730 00:38:14 0.00086116 0.0050023 0.0053915\n 7350 735 00:38:30 0.00085934 0.0049581 0.006383\n 7400 740 00:38:46 0.00085751 0.0052173 0.004874\n 7450 745 00:39:01 0.00085568 0.0051945 0.0043461\n 7500 750 00:39:17 0.00085383 0.0053168 0.0072798\n 7550 755 00:39:33 0.00085198 0.0052003 0.0047035\n 7600 760 00:39:49 0.00085011 0.0047887 0.0054121\n 7650 765 00:40:05 0.00084824 0.0064361 0.0050864\n 7700 770 00:40:21 0.00084635 0.0048274 0.0052752\n 7750 775 00:40:37 0.00084446 0.0056569 0.0063247\n 7800 780 00:40:53 0.00084256 0.0046624 0.0045618\n 7850 785 00:41:09 0.00084065 0.0048522 0.0050385\n 7900 790 00:41:25 0.00083872 0.0050696 0.0059915\n 7950 795 00:41:41 0.00083679 0.0061549 0.0054478\n 8000 800 00:41:57 0.00083485 0.0062461 0.0046517\n 8050 805 00:42:13 0.0008329 0.0062869 0.0044078\n 8100 810 00:42:29 0.00083094 0.0047935 0.0042977\n 8150 815 00:42:45 0.00082898 0.0056394 0.0045981\n 8200 820 00:43:01 0.000827 0.0068088 0.0053592\n 8250 825 00:43:17 0.00082501 0.0047475 0.0053531\n 8300 830 00:43:33 0.00082302 0.0063541 0.0052936\n 8350 835 00:43:49 0.00082102 0.0044934 0.0044086\n 8400 840 00:44:05 0.000819 0.0042048 0.0044732\n 8450 845 00:44:21 0.00081698 0.0040283 0.0046519\n 8500 850 00:44:37 0.00081495 0.0040599 0.0064072\n 8550 855 00:44:53 0.00081292 0.0043192 0.0040858\n 8600 860 00:45:09 0.00081087 0.0041443 0.0042541\n 8650 865 00:45:24 0.00080881 0.0046126 0.006429\n 8700 870 00:45:40 0.00080675 0.0052411 0.0040136\n 8750 875 00:45:56 0.00080468 0.0042144 0.0055315\n 8800 880 00:46:12 0.0008026 0.0040535 0.0038102\n 8850 885 00:46:28 0.00080051 0.0038047 0.0049905\n 8900 890 00:46:44 0.00079841 0.0052055 0.0062948\n 8950 895 00:47:00 0.0007963 0.0039927 0.0044481\n 9000 900 00:47:16 0.00079419 0.0039317 0.0046134\n 9050 905 00:47:32 0.00079207 0.0036886 0.0036194\n 9100 910 00:47:48 0.00078994 0.0045894 0.0045133\n 9150 915 00:48:04 0.0007878 0.0039448 0.0037831\n 9200 920 00:48:20 0.00078566 0.0036084 0.0034195\n 9250 925 00:48:36 0.0007835 0.0048239 0.005734\n 9300 930 00:48:52 0.00078134 0.0038397 0.0036154\n 9350 935 00:49:08 0.00077917 0.0039968 0.0043042\n 9400 940 00:49:24 0.000777 0.0036057 0.0040988\n 9450 945 00:49:40 0.00077481 0.0055798 0.0040157\n 9500 950 00:49:56 0.00077262 0.005316 0.0034046\n 9550 955 00:50:12 0.00077042 0.0048613 0.0031826\n 9600 960 00:50:28 0.00076822 0.0046209 0.0046535\n 9650 965 00:50:44 0.000766 0.0042702 0.0070814\n 9700 970 00:51:00 0.00076378 0.0028866 0.0031442\n 9750 975 00:51:16 0.00076155 0.003109 0.0033589\n 9800 980 00:51:32 0.00075932 0.0032501 0.0032099\n 9850 985 00:51:48 0.00075707 0.0030503 0.003223\n 9900 990 00:52:04 0.00075483 0.002811 0.0046857\n 9950 995 00:52:20 0.00075257 0.0030388 0.0037791\n 10000 1000 00:52:36 0.0007503 0.0037056 0.0031482\n 10050 1005 00:52:52 0.00074803 0.0026506 0.0031978\n 10100 1010 00:53:08 0.00074576 0.0024784 0.0028794\n 10150 1015 00:53:24 0.00074347 0.0034958 0.0034462\n 10200 1020 00:53:40 0.00074118 0.0025269 0.0030562\n 10250 1025 00:53:56 0.00073888 0.0027473 0.0036803\n 10300 1030 00:54:12 0.00073658 0.0028621 0.0030814\n 10350 1035 00:54:28 0.00073427 0.0031095 0.0021902\n 10400 1040 00:54:44 0.00073195 0.0029946 0.0035428\n 10450 1045 00:55:00 0.00072963 0.0026707 0.0031266\n 10500 1050 00:55:16 0.0007273 0.0029126 0.0023953\n 10550 1055 00:55:32 0.00072497 0.0021614 0.0027946\n 10600 1060 00:55:48 0.00072262 0.0032055 0.0034411\n 10650 1065 00:56:04 0.00072028 0.0025963 0.0025546\n 10700 1070 00:56:20 0.00071792 0.0029571 0.0034346\n 10750 1075 00:56:36 0.00071556 0.0026377 0.0031612\n 10800 1080 00:56:52 0.0007132 0.0023568 0.0028712\n 10850 1085 00:57:08 0.00071082 0.0022323 0.0030361\n 10900 1090 00:57:24 0.00070845 0.002433 0.0029169\n 10950 1095 00:57:40 0.00070606 0.0026361 0.0030457\n 11000 1100 00:57:56 0.00070367 0.0025743 0.0027106\n 11050 1105 00:58:12 0.00070128 0.0054939 0.0027221\n 11100 1110 00:58:28 0.00069888 0.0023053 0.0028323\n 11150 1115 00:58:44 0.00069647 0.0030638 0.0039619\n 11200 1120 00:59:00 0.00069406 0.0021272 0.0034332\n 11250 1125 00:59:16 0.00069165 0.0032257 0.0028605\n 11300 1130 00:59:32 0.00068923 0.0026998 0.002786\n 11350 1135 00:59:48 0.0006868 0.002349 0.0033932\n 11400 1140 01:00:04 0.00068437 0.0024726 0.0038569\n 11450 1145 01:00:20 0.00068193 0.0021615 0.0031552\n 11500 1150 01:00:36 0.00067949 0.0020876 0.0026508\n 11550 1155 01:00:52 0.00067704 0.0024662 0.0023897\n 11600 1160 01:01:08 0.00067459 0.0020017 0.0024203\n 11650 1165 01:01:24 0.00067213 0.0022404 0.0025954\n 11700 1170 01:01:40 0.00066967 0.0029461 0.0020102\n 11750 1175 01:01:56 0.00066721 0.0021257 0.0041509\n 11800 1180 01:02:12 0.00066474 0.00229 0.0026472\n 11850 1185 01:02:28 0.00066226 0.0025374 0.0026332\n 11900 1190 01:02:44 0.00065978 0.0021453 0.0027612\n 11950 1195 01:03:00 0.0006573 0.0022691 0.0027813\n 12000 1200 01:03:16 0.00065481 0.0023873 0.0026034\n 12050 1205 01:03:32 0.00065232 0.0024548 0.0030672\n 12100 1210 01:03:48 0.00064982 0.0028361 0.003714\n 12150 1215 01:04:04 0.00064732 0.0022495 0.0026284\n 12200 1220 01:04:20 0.00064482 0.0021187 0.0038155\n 12250 1225 01:04:36 0.00064231 0.0021783 0.0024608\n 12300 1230 01:04:52 0.0006398 0.0022455 0.0027035\n 12350 1235 01:05:08 0.00063728 0.002342 0.0021228\n 12400 1240 01:05:24 0.00063476 0.0021471 0.0024213\n 12450 1245 01:05:40 0.00063224 0.0020683 0.0029241\n 12500 1250 01:05:56 0.00062971 0.0020834 0.0020658\n 12550 1255 01:06:12 0.00062718 0.0031044 0.0037576\n 12600 1260 01:06:28 0.00062464 0.0026255 0.0032252\n 12650 1265 01:06:44 0.00062211 0.002041 0.0021648\n 12700 1270 01:07:00 0.00061956 0.0021084 0.003686\n 12750 1275 01:07:16 0.00061702 0.0021969 0.0029639\n 12800 1280 01:07:32 0.00061447 0.002249 0.0031241\n 12850 1285 01:07:48 0.00061192 0.00244 0.0023013\n 12900 1290 01:08:04 0.00060937 0.0023925 0.0022925\n 12950 1295 01:08:20 0.00060681 0.002856 0.0034456\n 13000 1300 01:08:36 0.00060425 0.0028025 0.0025975\n 13050 1305 01:08:52 0.00060169 0.0021048 0.0018367\n 13100 1310 01:09:08 0.00059912 0.0019493 0.0021189\n 13150 1315 01:09:24 0.00059655 0.0020659 0.0022836\n 13200 1320 01:09:40 0.00059398 0.0019947 0.0027278\n 13250 1325 01:09:56 0.00059141 0.0021062 0.0024234\n 13300 1330 01:10:12 0.00058883 0.0017516 0.0025194\n 13350 1335 01:10:28 0.00058626 0.0021857 0.0025718\n 13400 1340 01:10:44 0.00058367 0.0020874 0.0029131\n 13450 1345 01:11:00 0.00058109 0.0022741 0.002449\n 13500 1350 01:11:16 0.00057851 0.0020277 0.003294\n 13550 1355 01:11:32 0.00057592 0.0027412 0.0026126\n 13600 1360 01:11:48 0.00057333 0.0016066 0.0023317\n 13650 1365 01:12:05 0.00057074 0.0019897 0.0032557\n 13700 1370 01:12:20 0.00056814 0.0021644 0.0027634\n 13750 1375 01:12:37 0.00056555 0.0020115 0.0020817\n 13800 1380 01:12:53 0.00056295 0.0020026 0.0022431\n 13850 1385 01:13:09 0.00056035 0.0019667 0.0026013\n 13900 1390 01:13:25 0.00055775 0.0016694 0.0024656\n 13950 1395 01:13:41 0.00055515 0.001729 0.0031378\n 14000 1400 01:13:57 0.00055255 0.0021263 0.0029858\n 14050 1405 01:14:13 0.00054994 0.0023363 0.0025838\n 14100 1410 01:14:29 0.00054734 0.0018443 0.0021834\n 14150 1415 01:14:45 0.00054473 0.0017418 0.0023037\n 14200 1420 01:15:01 0.00054212 0.0027345 0.0026583\n 14250 1425 01:15:17 0.00053951 0.001837 0.0026537\n 14300 1430 01:15:33 0.0005369 0.0018602 0.0024828\n 14350 1435 01:15:49 0.00053428 0.0020003 0.0023746\n 14400 1440 01:16:05 0.00053167 0.0021004 0.0030399\n 14450 1445 01:16:21 0.00052906 0.001697 0.0028326\n 14500 1450 01:16:37 0.00052644 0.001782 0.0027563\n 14550 1455 01:16:53 0.00052383 0.0018179 0.0021725\n 14600 1460 01:17:10 0.00052121 0.0017951 0.0023296\n 14650 1465 01:17:26 0.00051859 0.0016025 0.0023709\n 14700 1470 01:17:42 0.00051598 0.0027345 0.0031629\n 14750 1475 01:17:58 0.00051336 0.0015954 0.0024857\n 14800 1480 01:18:14 0.00051074 0.0017852 0.0029012\n 14850 1485 01:18:30 0.00050812 0.0017514 0.0026738\n 14900 1490 01:18:46 0.0005055 0.0015201 0.0023805\n 14950 1495 01:19:02 0.00050289 0.0018113 0.0024279\n 15000 1500 01:19:18 0.00050027 0.0016366 0.0026208\n 15050 1505 01:19:34 0.00049765 0.0022222 0.0030568\n 15100 1510 01:19:50 0.00049503 0.0021572 0.002356\n 15150 1515 01:20:06 0.00049241 0.0022763 0.0036881\n 15200 1520 01:20:22 0.00048979 0.0021401 0.0025897\n 15250 1525 01:20:38 0.00048717 0.0022776 0.0022092\n 15300 1530 01:20:54 0.00048456 0.0018786 0.0021764\n 15350 1535 01:21:10 0.00048194 0.0018066 0.0016468\n 15400 1540 01:21:26 0.00047932 0.0018577 0.0016957\n 15450 1545 01:21:43 0.00047671 0.0016235 0.0019651\n 15500 1550 01:21:59 0.00047409 0.0020395 0.0023918\n 15550 1555 01:22:15 0.00047148 0.0017087 0.002023\n 15600 1560 01:22:31 0.00046886 0.001628 0.0022732\n 15650 1565 01:22:47 0.00046625 0.0016506 0.0021371\n 15700 1570 01:23:03 0.00046364 0.0017275 0.0021996\n 15750 1575 01:23:19 0.00046102 0.0026451 0.0028609\n 15800 1580 01:23:35 0.00045841 0.002788 0.0020499\n 15850 1585 01:23:51 0.0004558 0.0040824 0.003473\n 15900 1590 01:24:06 0.0004532 0.0016423 0.0025589\n 15950 1595 01:24:22 0.00045059 0.0017866 0.002646\n 16000 1600 01:24:38 0.00044798 0.0021176 0.0036517\n 16050 1605 01:24:54 0.00044538 0.0032128 0.0039766\n 16100 1610 01:25:11 0.00044278 0.0024041 0.002711\n 16150 1615 01:25:27 0.00044018 0.0021574 0.0027829\n 16200 1620 01:25:43 0.00043758 0.0019144 0.0021685\n 16250 1625 01:25:59 0.00043498 0.0019858 0.0018922\n 16300 1630 01:26:15 0.00043238 0.0019833 0.0022488\n 16350 1635 01:26:31 0.00042979 0.0020166 0.0022879\n 16400 1640 01:26:47 0.0004272 0.0018761 0.0021364\n 16450 1645 01:27:03 0.00042461 0.0018617 0.0023143\n 16500 1650 01:27:19 0.00042202 0.0018081 0.0022441\n 16550 1655 01:27:35 0.00041944 0.0019587 0.0026376\n 16600 1660 01:27:51 0.00041685 0.0017312 0.0024834\n 16650 1665 01:28:07 0.00041427 0.0018109 0.0020409\n 16700 1670 01:28:22 0.00041169 0.001615 0.0022005\n 16750 1675 01:28:39 0.00040912 0.0018642 0.0019236\n 16800 1680 01:28:55 0.00040654 0.0018204 0.0021713\n 16850 1685 01:29:11 0.00040397 0.0018005 0.0021067\n 16900 1690 01:29:27 0.0004014 0.0016739 0.0022078\n 16950 1695 01:29:43 0.00039884 0.0015365 0.0017443\n 17000 1700 01:29:59 0.00039627 0.0016695 0.0024143\n 17050 1705 01:30:15 0.00039371 0.0019337 0.0023614\n 17100 1710 01:30:31 0.00039115 0.0016085 0.0018719\n 17150 1715 01:30:47 0.0003886 0.0029839 0.0020613\n 17200 1720 01:31:03 0.00038605 0.0021822 0.001965\n 17250 1725 01:31:19 0.0003835 0.002355 0.0029472\n 17300 1730 01:31:35 0.00038095 0.0015627 0.0017846\n 17350 1735 01:31:51 0.00037841 0.0013718 0.0016907\n 17400 1740 01:32:07 0.00037587 0.0014573 0.0015151\n 17450 1745 01:32:23 0.00037334 0.0017053 0.0021884\n 17500 1750 01:32:39 0.00037081 0.0016724 0.0020601\n 17550 1755 01:32:55 0.00036828 0.0014671 0.0018353\n 17600 1760 01:33:11 0.00036575 0.0016462 0.0016916\n 17650 1765 01:33:27 0.00036323 0.0015472 0.0016747\n 17700 1770 01:33:43 0.00036072 0.0018633 0.001649\n 17750 1775 01:33:59 0.0003582 0.0025974 0.0020588\n 17800 1780 01:34:15 0.00035569 0.002356 0.0023926\n 17850 1785 01:34:31 0.00035319 0.0018696 0.0022794\n 17900 1790 01:34:47 0.00035069 0.0014419 0.0016805\n 17950 1795 01:35:03 0.00034819 0.0013786 0.0015304\n 18000 1800 01:35:19 0.0003457 0.0014951 0.001383\n 18050 1805 01:35:35 0.00034321 0.0014986 0.0018211\n 18100 1810 01:35:51 0.00034072 0.0012721 0.0017414\n 18150 1815 01:36:07 0.00033824 0.0020776 0.0018284\n 18200 1820 01:36:23 0.00033577 0.0013199 0.0013384\n 18250 1825 01:36:39 0.0003333 0.0013481 0.0015085\n 18300 1830 01:36:55 0.00033083 0.0013924 0.0015471\n 18350 1835 01:37:11 0.00032837 0.0015559 0.0015356\n 18400 1840 01:37:27 0.00032591 0.0015095 0.0015076\n 18450 1845 01:37:44 0.00032346 0.0016296 0.0014499\n 18500 1850 01:38:00 0.00032101 0.0013478 0.001694\n 18550 1855 01:38:16 0.00031857 0.0012833 0.0017312\n 18600 1860 01:38:32 0.00031613 0.0013092 0.0016675\n 18650 1865 01:38:48 0.0003137 0.0015068 0.0017694\n 18700 1870 01:39:04 0.00031127 0.0013057 0.0018385\n 18750 1875 01:39:20 0.00030885 0.001497 0.0019182\n 18800 1880 01:39:36 0.00030643 0.0012058 0.0014288\n 18850 1885 01:39:52 0.00030402 0.001238 0.0016571\n 18900 1890 01:40:08 0.00030161 0.0013303 0.0017283\n 18950 1895 01:40:24 0.00029921 0.0012873 0.0020364\n 19000 1900 01:40:40 0.00029681 0.0026647 0.0020157\n 19050 1905 01:40:56 0.00029442 0.0018135 0.0022697\n 19100 1910 01:41:12 0.00029204 0.0012852 0.0015388\n 19150 1915 01:41:28 0.00028966 0.0018375 0.0014238\n 19200 1920 01:41:44 0.00028729 0.0013132 0.0016411\n 19250 1925 01:42:00 0.00028492 0.0017705 0.0016317\n 19300 1930 01:42:16 0.00028256 0.0016894 0.0021083\n 19350 1935 01:42:32 0.0002802 0.0016904 0.0015283\n 19400 1940 01:42:48 0.00027786 0.0015774 0.0014632\n 19450 1945 01:43:04 0.00027551 0.0012766 0.0015901\n 19500 1950 01:43:20 0.00027318 0.0010416 0.001245\n 19550 1955 01:43:36 0.00027084 0.0010556 0.0011244\n 19600 1960 01:43:52 0.00026852 0.0010909 0.00094426\n 19650 1965 01:44:08 0.0002662 0.00097145 0.00109\n 19700 1970 01:44:24 0.00026389 0.00095441 0.0011452\n 19750 1975 01:44:40 0.00026159 0.0010913 0.0011513\n 19800 1980 01:44:56 0.00025929 0.0009123 0.0010749\n 19850 1985 01:45:12 0.00025699 0.00107 0.0011298\n 19900 1990 01:45:28 0.00025471 0.0010332 0.0013017\n 19950 1995 01:45:44 0.00025243 0.0012376 0.0011547\n 20000 2000 01:46:00 0.00025016 0.00096726 0.00098116\n 20050 2005 01:46:16 0.00024789 0.0010202 0.0010052\n 20100 2010 01:46:32 0.00024564 0.00094221 0.0010229\n 20150 2015 01:46:49 0.00024338 0.00085431 0.0010943\n 20200 2020 01:47:05 0.00024114 0.0011372 0.0010772\n 20250 2025 01:47:21 0.0002389 0.0010034 0.00098106\n 20300 2030 01:47:37 0.00023667 0.00095551 0.00097051\n 20350 2035 01:47:53 0.00023445 0.00098049 0.0016128\n 20400 2040 01:48:09 0.00023224 0.0011153 0.0011304\n 20450 2045 01:48:25 0.00023003 0.0021347 0.001687\n 20500 2050 01:48:41 0.00022783 0.0015434 0.0016615\n 20550 2055 01:48:57 0.00022563 0.0015185 0.0016536\n 20600 2060 01:49:13 0.00022345 0.00091351 0.0011983\n 20650 2065 01:49:29 0.00022127 0.0010078 0.0012009\n 20700 2070 01:49:45 0.0002191 0.00092273 0.0009209\n 20750 2075 01:50:01 0.00021694 0.0010393 0.00094644\n 20800 2080 01:50:17 0.00021478 0.0009314 0.00099959\n 20850 2085 01:50:33 0.00021264 0.0011772 0.0012051\n 20900 2090 01:50:49 0.0002105 0.0010429 0.0010475\n 20950 2095 01:51:05 0.00020837 0.0012619 0.0011346\n 21000 2100 01:51:21 0.00020624 0.0014028 0.0015166\n 21050 2105 01:51:37 0.00020413 0.002182 0.0017179\n 21100 2110 01:51:53 0.00020202 0.0010472 0.0010081\n 21150 2115 01:52:09 0.00019992 0.0010738 0.0010961\n 21200 2120 01:52:25 0.00019783 0.0010998 0.0010807\n 21250 2125 01:52:41 0.00019575 0.0010799 0.0011845\n 21300 2130 01:52:58 0.00019367 0.0011378 0.0010161\n 21350 2135 01:53:14 0.00019161 0.0010385 0.00094433\n 21400 2140 01:53:30 0.00018955 0.0010601 0.0010589\n 21450 2145 01:53:46 0.0001875 0.00098625 0.0011407\n 21500 2150 01:54:02 0.00018546 0.0018647 0.0018763\n 21550 2155 01:54:18 0.00018343 0.00098855 0.0011061\n 21600 2160 01:54:34 0.00018141 0.00091989 0.0010024\n 21650 2165 01:54:50 0.0001794 0.00091874 0.00099932\n 21700 2170 01:55:06 0.00017739 0.00092038 0.00091465\n 21750 2175 01:55:22 0.00017539 0.00088197 0.0010527\n 21800 2180 01:55:38 0.00017341 0.0010255 0.0010631\n 21850 2185 01:55:54 0.00017143 0.00085444 0.00090728\n 21900 2190 01:56:10 0.00016946 0.00084727 0.0010766\n 21950 2195 01:56:26 0.0001675 0.00088647 0.0010296\n 22000 2200 01:56:42 0.00016555 0.00084739 0.0010236\n 22050 2205 01:56:58 0.0001636 0.00090044 0.001109\n 22100 2210 01:57:14 0.00016167 0.00092583 0.001217\n 22150 2215 01:57:30 0.00015975 0.00087582 0.00094238\n 22200 2220 01:57:46 0.00015783 0.000863 0.0011632\n 22250 2225 01:58:02 0.00015593 0.0010102 0.0010763\n 22300 2230 01:58:18 0.00015403 0.00099515 0.0010094\n 22350 2235 01:58:34 0.00015215 0.00093626 0.0010619\n 22400 2240 01:58:50 0.00015027 0.00091844 0.0009811\n 22450 2245 01:59:06 0.0001484 0.0011521 0.00096817\n 22500 2250 01:59:22 0.00014655 0.00088231 0.0010092\n 22550 2255 01:59:38 0.0001447 0.0010246 0.0010228\n 22600 2260 01:59:54 0.00014286 0.0010971 0.0011736\n 22650 2265 02:00:10 0.00014103 0.0042712 0.0051264\n 22700 2270 02:00:26 0.00013922 0.0014129 0.0013977\n 22750 2275 02:00:42 0.00013741 0.0011759 0.0012689\n 22800 2280 02:00:58 0.00013561 0.0012122 0.0012862\n 22850 2285 02:01:15 0.00013382 0.00099445 0.0011123\n 22900 2290 02:01:31 0.00013204 0.00093522 0.00097945\n 22950 2295 02:01:47 0.00013028 0.00085753 0.0009903\n 23000 2300 02:02:02 0.00012852 0.00087358 0.00094513\n 23050 2305 02:02:18 0.00012677 0.00099145 0.00094497\n 23100 2310 02:02:35 0.00012503 0.00088152 0.0009134\n 23150 2315 02:02:51 0.00012331 0.00092086 0.00099135\n 23200 2320 02:03:07 0.00012159 0.00092854 0.00091853\n 23250 2325 02:03:23 0.00011988 0.00092683 0.0010603\n 23300 2330 02:03:39 0.00011819 0.0012571 0.0010679\n 23350 2335 02:03:55 0.0001165 0.00088566 0.0010385\n 23400 2340 02:04:11 0.00011483 0.001005 0.00092071\n 23450 2345 02:04:27 0.00011316 0.00098361 0.00092014\n 23500 2350 02:04:43 0.00011151 0.0012323 0.0013903\n 23550 2355 02:04:59 0.00010986 0.0012213 0.0011035\n 23600 2360 02:05:15 0.00010823 0.0013001 0.0012856\n 23650 2365 02:05:31 0.00010661 0.0014167 0.0014498\n 23700 2370 02:05:47 0.000105 0.0024125 0.0038592\n 23750 2375 02:06:03 0.0001034 0.0017876 0.0018078\n 23800 2380 02:06:19 0.00010181 0.0015193 0.001909\n 23850 2385 02:06:35 0.00010023 0.0010862 0.0011527\n 23900 2390 02:06:51 9.8664e-05 0.0013709 0.00094224\n 23950 2395 02:07:07 9.7107e-05 0.0013856 0.00088534\n 24000 2400 02:07:23 9.5562e-05 0.0011348 0.00088116\n 24050 2405 02:07:39 9.4028e-05 0.0010557 0.00086249\n 24100 2410 02:07:55 9.2505e-05 0.0012366 0.00090103\n 24150 2415 02:08:11 9.0993e-05 0.0010839 0.00077987\n 24200 2420 02:08:27 8.9492e-05 0.001122 0.00079427\n 24250 2425 02:08:44 8.8003e-05 0.0010429 0.00077344\n 24300 2430 02:09:00 8.6525e-05 0.0010395 0.00090613\n 24350 2435 02:09:16 8.5058e-05 0.00095784 0.00077051\n 24400 2440 02:09:32 8.3603e-05 0.0010223 0.000771\n 24450 2445 02:09:48 8.2159e-05 0.0010814 0.00074239\n 24500 2450 02:10:04 8.0726e-05 0.00097704 0.00074942\n 24550 2455 02:10:20 7.9305e-05 0.00099932 0.00073714\n 24600 2460 02:10:36 7.7896e-05 0.0010589 0.00076202\n 24650 2465 02:10:52 7.6498e-05 0.0010034 0.00077246\n 24700 2470 02:11:08 7.5112e-05 0.00097715 0.00074769\n 24750 2475 02:11:24 7.3737e-05 0.0012611 0.0008424\n 24800 2480 02:11:40 7.2374e-05 0.00093554 0.00072949\n 24850 2485 02:11:56 7.1023e-05 0.0010296 0.00071859\n 24900 2490 02:12:12 6.9684e-05 0.00092649 0.00076156\n 24950 2495 02:12:28 6.8356e-05 0.0010038 0.00072601\n 25000 2500 02:12:44 6.704e-05 0.00096368 0.00072476\n 25050 2505 02:13:00 6.5736e-05 0.00098903 0.00074766\n 25100 2510 02:13:16 6.4444e-05 0.0010097 0.00072006\n 25150 2515 02:13:32 6.3164e-05 0.00094253 0.00072841\n 25200 2520 02:13:48 6.1896e-05 0.0010246 0.00074081\n 25250 2525 02:14:04 6.064e-05 0.00089725 0.00073981\n 25300 2530 02:14:20 5.9396e-05 0.00097605 0.00077159\n 25350 2535 02:14:36 5.8165e-05 0.0010684 0.00074877\n 25400 2540 02:14:52 5.6945e-05 0.0010021 0.00083028\n 25450 2545 02:15:08 5.5737e-05 0.0010762 0.00079952\n 25500 2550 02:15:24 5.4542e-05 0.00095972 0.00087132\n 25550 2555 02:15:40 5.3359e-05 0.0012367 0.0008153\n 25600 2560 02:15:57 5.2188e-05 0.0010316 0.00081752\n 25650 2565 02:16:13 5.1029e-05 0.0011539 0.00082998\n 25700 2570 02:16:29 4.9883e-05 0.0010823 0.0010713\n 25750 2575 02:16:45 4.8749e-05 0.001038 0.0008421\n 25800 2580 02:17:01 4.7627e-05 0.0011896 0.00095383\n 25850 2585 02:17:17 4.6518e-05 0.0010405 0.00092353\n 25900 2590 02:17:33 4.5421e-05 0.0010727 0.00090705\n 25950 2595 02:17:49 4.4337e-05 0.0011211 0.00099545\n 26000 2600 02:18:05 4.3265e-05 0.0012394 0.0012794\n 26050 2605 02:18:21 4.2206e-05 0.0015267 0.0014068\n 26100 2610 02:18:37 4.1159e-05 0.0014255 0.0010372\n 26150 2615 02:18:53 4.0125e-05 0.0015816 0.0011535\n 26200 2620 02:19:09 3.9104e-05 0.0013712 0.0011619\n 26250 2625 02:19:25 3.8095e-05 0.0014794 0.001187\n 26300 2630 02:19:41 3.7099e-05 0.0013275 0.0011328\n 26350 2635 02:19:57 3.6115e-05 0.0014279 0.0012803\n 26400 2640 02:20:13 3.5145e-05 0.0012899 0.0010859\n 26450 2645 02:20:29 3.4187e-05 0.00149 0.0012144\n 26500 2650 02:20:45 3.3241e-05 0.0014024 0.0013736\n 26550 2655 02:21:01 3.2309e-05 0.0014844 0.0012479\n 26600 2660 02:21:18 3.1389e-05 0.0015835 0.0016185\n 26650 2665 02:21:34 3.0483e-05 0.0058991 0.0060732\n 26700 2670 02:21:50 2.9589e-05 0.0041194 0.0036974\n 26750 2675 02:22:06 2.8708e-05 0.0019904 0.0017468\n 26800 2680 02:22:22 2.784e-05 0.0013358 0.0012805\n 26850 2685 02:22:38 2.6985e-05 0.0011744 0.0011167\n 26900 2690 02:22:54 2.6143e-05 0.0010823 0.00099406\n 26950 2695 02:23:10 2.5314e-05 0.0010501 0.00090623\n 27000 2700 02:23:26 2.4498e-05 0.0010592 0.00090321\n 27050 2705 02:23:42 2.3695e-05 0.0011382 0.00087822\n 27100 2710 02:23:58 2.2905e-05 0.00094988 0.00085277\n 27150 2715 02:24:14 2.2128e-05 0.000997 0.00086328\n 27200 2720 02:24:30 2.1364e-05 0.0010763 0.00085394\n 27250 2725 02:24:46 2.0614e-05 0.00096375 0.0008247\n 27300 2730 02:25:02 1.9876e-05 0.00096438 0.00083616\n 27350 2735 02:25:19 1.9152e-05 0.0009698 0.00085386\n 27400 2740 02:25:35 1.8441e-05 0.00097542 0.00082172\n 27450 2745 02:25:51 1.7743e-05 0.00093997 0.00084637\n 27500 2750 02:26:07 1.7058e-05 0.0009962 0.00086072\n 27550 2755 02:26:23 1.6387e-05 0.00092882 0.00084878\n 27600 2760 02:26:39 1.5729e-05 0.00093525 0.00085172\n 27650 2765 02:26:55 1.5084e-05 0.0011558 0.00086144\n 27700 2770 02:27:11 1.4452e-05 0.00096671 0.00087266\n 27750 2775 02:27:27 1.3834e-05 0.0010529 0.00085203\n 27800 2780 02:27:44 1.3229e-05 0.00093961 0.00089148\n 27850 2785 02:28:00 1.2638e-05 0.0010011 0.00084165\n 27900 2790 02:28:16 1.2059e-05 0.00095072 0.00083931\n 27950 2795 02:28:32 1.1495e-05 0.0011275 0.00085928\n 28000 2800 02:28:48 1.0943e-05 0.0010559 0.00083574\n 28050 2805 02:29:04 1.0405e-05 0.00095486 0.00086766\n 28100 2810 02:29:20 9.8809e-06 0.001031 0.00090354\n 28150 2815 02:29:36 9.3698e-06 0.00098583 0.00088427\n 28200 2820 02:29:52 8.8722e-06 0.00097426 0.00087277\n 28250 2825 02:30:08 8.388e-06 0.00098936 0.00086164\n 28300 2830 02:30:24 7.9174e-06 0.0010206 0.000891\n 28350 2835 02:30:40 7.4602e-06 0.0010582 0.00088516\n 28400 2840 02:30:56 7.0166e-06 0.0010947 0.0009366\n 28450 2845 02:31:12 6.5865e-06 0.0010767 0.00089407\n 28500 2850 02:31:28 6.1699e-06 0.0011903 0.00090544\n 28550 2855 02:31:44 5.7668e-06 0.0010079 0.00090236\n 28600 2860 02:32:00 5.3774e-06 0.0010096 0.00088854\n 28650 2865 02:32:16 5.0014e-06 0.00099521 0.00090043\n 28700 2870 02:32:32 4.6391e-06 0.0011222 0.00088661\n 28750 2875 02:32:48 4.2904e-06 0.0011128 0.00089644\n 28800 2880 02:33:04 3.9552e-06 0.0011693 0.00092828\n 28850 2885 02:33:20 3.6337e-06 0.001071 0.00091526\n 28900 2890 02:33:36 3.3258e-06 0.0010639 0.00094324\n 28950 2895 02:33:52 3.0315e-06 0.0011189 0.00092664\n 29000 2900 02:34:08 2.7509e-06 0.0010719 0.00094841\n 29050 2905 02:34:25 2.4838e-06 0.0011975 0.00095103\n 29100 2910 02:34:41 2.2305e-06 0.0011187 0.00094303\n 29150 2915 02:34:57 1.9908e-06 0.0011262 0.00093658\n 29200 2920 02:35:13 1.7647e-06 0.0011177 0.0009395\n 29250 2925 02:35:29 1.5523e-06 0.0011116 0.00094363\n 29300 2930 02:35:45 1.3536e-06 0.0010867 0.00096602\n 29350 2935 02:36:01 1.1686e-06 0.0011155 0.00094136\n 29400 2940 02:36:17 9.9728e-07 0.0010685 0.00095303\n 29450 2945 02:36:33 8.3964e-07 0.0011687 0.00095275\n 29500 2950 02:36:49 6.9568e-07 0.0010902 0.00095546\n 29550 2955 02:37:05 5.6543e-07 0.0011249 0.0009563\n 29600 2960 02:37:21 4.4887e-07 0.0010921 0.00095011\n 29650 2965 02:37:37 3.4602e-07 0.0010889 0.00095198\n 29700 2970 02:37:53 2.5688e-07 0.0011085 0.00095074\n 29750 2975 02:38:10 1.8145e-07 0.0010358 0.00095092\n 29800 2980 02:38:26 1.1973e-07 0.0010591 0.00095221\n 29850 2985 02:38:42 7.1724e-08 0.0011161 0.00095279\n 29900 2990 02:38:57 3.7433e-08 0.00108 0.00095213\n 29950 2995 02:39:13 1.6858e-08 0.0010591 0.00095227\n 30000 3000 02:39:28 1e-08 0.0011262 0.00095234\nTraining stopped: Max epochs completed\n","truncated":false}} +%--- +%[output:6122d9b1] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,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","height":629,"width":1024}} +%--- +%[output:5b7527e4] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,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","height":629,"width":1024}} +%--- +%[output:8369fad8] +% data: {"dataType":"text","outputData":{"text":"Saved fine-tuned model: \/home\/mmarkows\/Documents\/geometric-deep-learning-to-publish-realistic-geoms-and-finetuning-5\/results\/TransolverFinetune_20260427T150324.mat\n","truncated":false}} +%--- +%[output:8ee38f52] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,iVBORw0KGgoAAAANSUhEUgAAAjAAAAFRCAYAAABqsZcNAAAQAElEQVR4AeydB5wU5fnHf3tSpZ0URYouKKLYUGNF4UBjQETRxKCicERjVFRiLLHEP5DYSxRLEqMRjAqW2BEbeocK0YgKFoKo3NIEQaqHHPX+z29gLsuxu7c7O9t\/+9lnZ\/bt832feed5y8wUVesjAiIgAiIgAiIgAjlGoAj6iIAIiIAIiIAIJEhAwTNNQAZMpmtA+YuACIiACIiACCRMQAZMwsgUQQREQAQyT0AlEIFCJyADptA1QMcvAiIgAiIgAjlIQAZMDlaaiiwCmSegEoiACIhAZgnIgMksf+UuAiIgAiIgAiLggYAMGA\/QFCXzBFQCERABERCBwiYgA6aw619HLwIiIAIiIAI5SUAGjKdqUyQREAEREAEREIFMEpABk0n6ylsEREAEREAEComAj8cqA8ZHmEpKBERABERABEQgPQRkwKSHs3IRAREQARHIPAGVII8IyIDJo8rUoYiACIiACIhAoRCQAVMoNa3jFAERyDwBlUAERMA3AjJgfEOphERABERABERABNJFQAZMukgrHxHIPAGVQAREQATyhoAMmLypSh2ICIiACIiACBQOARkwhVPXmT9SlUAEREAEREAEfCIgA8YnkEpGBERABERABEQgfQQKyYBJH1XlJAIiIAIiIAIikFICMmBSileJi4AIiIAIiECuE8jO8suAyc56UalEQAREQAREQARiEJABEwOOvERABERABDJPQCUQgUgEZMBEoiI3ERABERABERCBrCYgAyarq0eFEwERyDwBlUAERCAbCciAycZaUZlEQAREQAREQARiEpABExOPPEUg8wRUAhEQAREQgR0JyIDZkYlcREAEREAEREAEspyADJgsr6DMF08lEAEREAEREIHsIyADJvvqRCUSAREQAREQARGog0DWGzB1lF\/eIiACIiACIiACBUhABkwBVroOWQREQAREIO8J5P0ByoDJ+yrWAYqACIiACIhA\/hGQAZN\/daojEgEREIHME1AJRCDFBGTApBiwkhcBERABERABEfCfgAwY\/5kqRREQgcwTUAlEQATynIAMmDyvYB2eCIiACIiACOQjARkw+VirOqbME1AJREAEREAEUkpABkxK8SpxERABERABERCBVBCQAZMKqplPUyUQAREQAREQgbwmIAMmr6tXBycCIiACIiAC+UkgNQZMfrLSUYmACIiACIiACGQJARkwWVIRKoYIiIAIiIAIiED8BGTAxM9KIUVABERABERABLKEgAyYLKkIFUMEREAEMk9AJRCB3CEgAyZ36kolFQEREAEREAER2EZABsw2ENqIgAhknoBKIAIiIALxEpABEy8phRMBERABERABEcgaAjJgsqYqVJDME1AJREAEREAEcoWADJhcqSmVUwREQAREQAREoIaADJgaFJnfUQlEQAREQAREQATiIyADJj5OCiUCIiACIiACIpBFBMIMmCwqlYoiAiIgAiIgAiIgAjEIyICJAUdeIiACIiACIlAnAQXICAEZMBnBrkxFQAREQAREQASSISADJhl6iisCIiACmSegEohAQRKQAVOQ1a6DFgEREAEREIHcJiADJrfrT6UXgcwTUAlEQAREIAMEZMBkALqyFAEREAEREAERSI6ADJjk+Cl25gmoBCIgAiIgAgVIQAZMAVa6DlkEREAEREAEcp2ADJhka1DxRUAEREAEREAE0k5ABkzakStDERABERABERCBZAnIgEmWoOKLgAiIgAiIgAiknYAMmLQjV4YiIAIiIAKZJ6AS5DoBGTC5XoMqvwiIgAiIgAgUIAEZMAVY6TpkERCBzBNQCURABJIjIAMmOX6KLQIiIAIiIAIikAECMmAyAF1ZikDmCagEIiACIpDbBGTA5Hb9qfQiIAIiIAIiUJAEZMAUZLVn\/qBVAhEQAREQARFIhoAMmGToKa4IiIAIiIAIiEBGCBSoAZMR1spUBERABERABETAJwIyYHwCqWREQAREQAREIO8JZNEByoDJospQUURABERABERABOIjIAMmPk4KJQIiIAIikHkCKoEI1BCQAVODQjsiIAIiIAIiIAK5QkAGTK7UlMopAiKQeQIqgQiIQNYQkAGTNVWhgoiACIiACIiACMRLQAZMvKQUTgQyT0AlEAEREAER2EZABsw2ENqIgAiIgAiIgAjkDgEZMLlTV5kvqUogAiIgAiIgAllCQAZMllSEiiECIiACIiACIhA\/gVwyYOI\/KoUUAREQAREQARHIawIyYPK6enVwIiACIiACIpCfBGTA5Ge96qhEQAREQAREIK8JyIDJ6+rVwYmACIhA5gmoBCKQCgIyYFJBVWmKgAiIgAiIgAiklIAMmJTiVeIiIAKZJ6ASiIAI5CMBGTD5WKs6JhEQAREQARHIcwIyYPK8gnV4mSegEoiACIiACPhPQAaM\/0yVogiIgAiIgAiIQIoJyIBJMeDMJ68SiIAIiIAIiED+EZABk391qiMSAREQAREQgbwnkHIDJu8J6gBFQAREQAREQATSTkAGTNqRK0MREAEREAERqJOAAtRBQAZMHYDkLQIiIAIiIAIikH0EZMBkX52oRCIgAiKQeQIqgQhkOQEZMFleQSqeCIiACIiACIjAjgRkwIQx+e9\/\/4vOnTtHlK+++goXX3xxRD\/G+f7772tSWr9+PQ444AAn7F\/+8pca90g7a9aswU033YTjjz8eXbt2xYABAzB69GgwjUjha7sde+yx6Nu3b23niP8XLVqEv\/71rzV+V1xxhVPG8LLXeGZg54MPPnDKQ54fffTRdiUYPny447dw4cLt3PnnkEMOcfwYj3LEEUfg8ssvx7fffkvvpIV1yHRZvgULFjh5\/d\/\/\/V+d6VZXV2PMmDFYtmyZEzZdvB999FGnjCyzKz169MDVV19dUxbE+DzzzDNO\/EmTJsUItdWrtk5RH0888cStnsn9OrHdc5LsHIcIPw899BBOOOEEHHTQQc45yjJFCJZTTl9++SWGDRuGbt264aSTTsLzzz8fsfzUsfvvv99pP3j8l112GZYsWRIxbF2Ojz\/+OE477TQceOCBuOSSS7Bhw4aaKH4zrn3Osu37+c9\/jldeeaUmz2g7bAOo1yNHjowWpMadfFJ9DsZ7XagplHZ8I1DkW0p5lNC+++6LSy+9dDvZZZddao5w6NCh2\/kxbJMmTWr833rrLfz4449gYz5x4sQa90g7bJgffvhhDBo0CA888AAOP\/xw8AJ09913RwqelNuTTz6Jv\/\/97zVpXHjhhWCj1aJFixq3TO7wgtmhQwdQ4mnIwsvarFkzp05o6LRs2RIvvvgirrzyyvAgvuzvuuuuDrNhdnGpK8F\/\/\/vfjgHjGojp5s2LOnWTTPbZZx\/861\/\/AvfrKnci\/rV1iheL22+\/PZEkkgrL\/G+55Ra0bt0aJ598Mt58803n4ptUohmOXFlZiTPPPBOff\/45zjjjDMeQYDvx4Ycf7lCyBx98EH\/+85\/RsWNHpyPz8ssve9J7pk2jfNOmTejfvz94Lt57771Ofqli7J6z1NHBgweDhif3p02b5uTrx086z8G6rgt+HI\/S2J6ADJjteTj\/9ttvP6cHz168K2wgHU\/7+dWvfrWDf+PGjc1n6\/ell14CjaAhQ4Zg9uzZ4OjNVp8df6dOneqEveCCC5xeJBuR66+\/Hn369HECswfBHlavXr3Acv3yl7\/EjBkzHL\/aP\/fdd5\/Tc+ZIAf04isGeCt2vu+46x0BavXq1E4Y927\/97W8455xzQDeGf\/vtt50RIPaG+vXrh6eeeorOjrBnx4sgDYvjjjsOhx12mGNoOZ4+\/GzevBk09nr37u00xGS4ZcuWuFNmY8i6YkP\/6quvgkbZ+++\/7zT+48ePd46ZvUiOdNFoY3633norjjrqKHTv3h3kzsabGfICcqEZd+RNw3LVqlV0dmTp0qUOs7Fjxzr\/mc5dd90FjnAwHfYK161bhxdeeMEJx0C8IDxqoyK1eYdCIbDR42gdjd2bb74ZGzduZBS4ZWYdcFSOYa655poaf47asW5nzpzphI\/0w2N1mbC87dq1w\/Tp0+FyJWMaOezl0yBzR4pqp0V9oz97+BwNeOONN5wgkXRqxIgRzkgPGVJfeCF2AtsPdYdlZj3YX8SbP8NGE3YceNF74oknQH6HHnooyITnTe04NAhOP\/105zyiscPzg2EqKioc\/WDdsT6YnsufZSVH6gzDxjpHjjzySPCc\/8Mf\/uDoA8OHy1577eXkQwbh8u6774YHc85HpkN94Wgsy8UAn3zyCTfbCdslGurjxo0DDUf3+N1A8TLmcTEOjRaeF+xIuaM+iTBmGvGKe85SR3n+8TgYl0YHtytXrnQMbuod2xzWMd1ry4oVK5z2mCOvPA9pRDNMXecgR5vYTruj3UynU6dOOP\/88xndMSAj6YvjGeEn2nUhki657SmNQx4frxHU2WhtfSQdjVCEgnOSAeNzlbPhZmPARq9nz55o0KCB01BHy4bKy+Hi3\/zmN044XkTOO+888GRkHDZi7GHx4siLFi+mvCiwt0L\/eIUnJS8oTZs2dUYQ9thjj+2isjFnmHr16oE9Wo6CXHvttTVDunTnBZ4nHE++Ro0aYdSoUZg\/f\/526Xj9884774AN1s9+9jNnOJyNideeWCAQcKYTWBbWx0477cRdx+C66KKLHOOLDTVHo2hc0GB8+umnwYsVA\/JCwIs0jcVTTjnFMUboHkmYDkfOfvrTn4LpsNFkY8yG9NRTT3WikCf9nT\/bfmg0Mn3WPTnTMOBI3J\/+9CcnhFtm1j+NzP333x8sI9NnAPa4qSOsT\/6PJTSy2KMm04MPPhhFRUUg29\/+9rdo37497rjjDixevDhiz53GDi\/ovPhT\/zi6xekFDuNTX6LpFMtFo5t6xTpg+SZPnsyN08OPN38nQowfGne8AJLX3Llz8emnn4LnSiAQ2C4WR8HYy1++fLmj3zxunmc0NhmXgcmWF24ara4bDU9XZ3gsPGaeC6zT2udI\/fr1HeOJekzjjmmGCzlGEtZleDiWjcfETgINZo6cMc+jjz46PJiz\/4tf\/MKZNuPFj3x5ISR3eibCmOyYRzAYZFTsvffejk5wijtexk5Ejz9szzh6xuicXuKWI5ZsS2nYsAw33HADIhlx1EuOuHJ0kQY5DZjnnnvOMSJjnYM85zhNxk4k8+M5HwgEnJG8WPrCsIlIJF0ia7anLCfLzxF+nuvR2no3jXAdTaQM+RpWBkyEmmXPI7yHxH0qmxuUDQTdXGFD6Pqx98+wPDlovHAkhb0g17\/2lgo7cOBAZ+ibFxT24s444wx89tlnTlBa74FAwOldsTfAho0nHcvoBIjzh2Vlj4cnwjHHHAP2qsKjTpgwwfl71VVXgfnwIk4HjgBwGwhsvSCwMeGwNo2oQCCAOXPm0DtpYe+8efPmYCPNCzNHUDgik2jCvOCyIeLoC4+ZF1w3Dfa42OBzZOXZZ58FRyTYKPJ4yJ3D7wzLRpM9W\/Z+edHrdUc6\/AAAEABJREFUG2ONEdNhPdNoYTrkyLn8Nm3aOMYB06ORyry47wobazaSvFgyD17w2BukkeKG4fass85yphd\/97vf8W\/NaB576DQm2at3PCL80DAigy5dujjTKhxZYw+PQXlR5JaNJ404GmocCeDFl+6u0NihO6cYeBGhXlO\/OcrBtGPpFEfxaDxNmTLFGQnjBZYGAlnEm79bjrq2vHCTJet\/9OjROwTntO4PP\/zgcOD5xhFPXpzLy8sRCGzVbfKnccbRMDeBcJ1h3dI92jlCv6qqKudcJVP+Dxee3zx\/a4trNISH5T65c4SQbcof\/\/hHZ20K3SMJjWEeEzsmrhEcizFHq1yhYcspb15UA4GtLBo2bOhkQ4PP2bGfuhhbkIS+XKNGHaJwhIV1wVFGtpnsGFHnyJG6TuY0EHkBr50JR0BpgHH0rLS01PGmIVvXOUg9YGB2nrh97bXXwDyot7H0hWEjSazrAsOH61IgsJUzO4M8r9jexGrrA4Gt4SPpKNMuVJEBE6HmeTJxyDhceOF3g9LoCPfjRd31cy+6PPF4YvKk4MJPNvhumPAtG3OmxxOQF042Ql988QXY62M49kzYy3UblLZt29LZ6R05Oz79sKfHpHhx55Y9Ap7MbGT43xWWl\/tcC8KtO\/zKfa\/CaRNy4gWFF2QKRyjIkn7xpMtykjd7juy5BQIBsNEPj8sLufuf4SmMQ+HFgqMQ9GeDHm74sCGkeyRhGrwA8UJPf07H8KLD\/Vji5sW6dcORKY1TjsK5brvvvruz63LnBdJxiOOHukQ95QWTwdkws3fPfTcPjhLy+O+88046I9ICUPqxZ8twnB5hwHjqhYt5eVF87733wGnNtWvXOlOUjJ9I\/gwfS9grp9FIY4oXOC5CrR3eze\/3v\/+9M41DY5NhXL3nPnWH23AJ1xk3bKxzhKMytTsH4eklss9pQ44SnnnmmaCBy2OLFp\/TTDRIaIicffbZjsHoHnOkOuZonys819hhoGHqpu+e12RKt3gYM1wiwos29ZNCvWQHhmVnGt999x034AgF9Y5tAnXOPSbHc9sPudA44LnEkW868zziNpbwvOZoDw0Y6iY7PdRZtrVuPrH0pXbabMd5LK6EXxcYNlyX+J\/CTgW3lHja+kg6yriFKpENmEKlse24eWJxlCJcAoGtFjCDcGg33M9t0NijZmPNk8lVYg5BMw4vxtyGC+c12bN253Z5IWQPhNMFvMCxQeGJHd4L4pQD06A7t+HCaR3+Zzm4nTdvHjdxCS+eDOjGZdnYYNTOJxD4HweG90PYM2fDyykYlxtHD+jG3ns8ebDO3Ljcch6ddRQeN9wI5fFy6J5hXXHv0KLxwuNnb57x3caU+7WFhkU4Z\/Z6uQDbjVs7vPvfNUzCRzxouNJoZMPqhnMNI\/d\/IlvO55MBh9a5z7VQbsPMY2RavEC6x89t7fqmYce7sGjU08geZdOGjBePcBqJa5rYmy0rKwP507Bn3HjzZ9hYwkafvW4y42iYey7WjkP9oBuNOR6nK+4Fj34sH7fhEu5GnaFfrHMkPDzDhgsvwrwY1xYyDg\/H0VfqEPWKhiPLTH+eJ9yGC+uPUw80dmjo0FhhG8H1VbEYsy5doW5Qj9neUO+Z\/tdffw3WX6tWrZzFtfEwZrxEhCOX1E8Kj5HtHKdkmYZbXxz9dOuKW46Q0d8VGh40MmiwcF0fO0KuXzxb6iNZPfbYY47Rx9EQxnPzZ7mYryvh+sJw4RLtuuCGiaQbgcD\/2lOee2Tghmc9cp\/u3FIipUH3QhUZMB5q\/pFHHgEbmHDh8CrXGXAumkOTPCkpbPg5J885WvqFZ8fRFFr9vCiwZ8tpGy6W5Fw747L3yh4JG0xeXNkbYYO18847w53bDU\/Ptc4Zjo2iaxi5YWjgsLfFk5xb151bTlVwywaEDSXLxP+c3uA2lcLpI56YXOvB46bwuOlGbvHkHd4YMj5HkGLF41QSR0\/IlsYG63TWrFlOFA4F03jj9BJZcErK8Yjww54\/e6ucQuJQP40wGiI0PMibURifo3Dcd4VrYthIcvEl64q82XjFy\/uf\/\/ync6fKN9984yYZdcuycHEzG\/vbbrvNCeeOQLgjg9QJ9mRrjx7w2BiBPLimgg09\/\/MYqUM8Rm4Zn1v6hQtHfciYhh0vrm69xJu\/mxYvTuHnG\/dpvHAtCqeGOPJFlnSnuIaaG58XHl6QuSCZblzTwzt4qDf8H48ke47wfHQvhOFbGh\/h+bOcnOrjCBqng9jJoT8X6HLLaTgahtznRc5tN1i3PPdpuOy5557OHUwMU1cdMwwv5NxypId5cvqGbVEgEHDWDMXDmPG9CqetORrBOqQRRaOKBgHridOQNPTZNnGUNjwP+tF4oR5ylI+jhaxTTkHxnKM7w0c6B+lO\/eSWRjq5u1y96AvbEOpeuPC6wPTjEbZ5PFfiaevjSa8QwsiA8VDLHFVhbzZc2GPhKAtPHq7hCE+WDTcVkydYuDvvXKJBwqFFzn\/+4x\/\/AIfwOQXChaEMy33O\/3L+nRdH9gK5TiLcKmc4Ck8+NjpsfNjocU6Y7jzJuaUfe1qci+cJTjdX2FgwfzYe7LHzAsDFcBxSdcOkYssLJNeD\/OQnPwGHbt08eCFlmdjwMIzr7teWDNir5BoQjnpx6N+9QPF5KawzXgzYKDEc861tgNKNc9hcgE1Dh9w4fcQ5efrxThcumiVHTg\/SzRUO2XN9EUdiwuPzAuKGibVl4\/yf\/\/wH7gLZWGHpx7UrvEBwlIIXBa7zYGNPo5s6xlEmloNGI8O7Qp2iPvPiywsbh8k5ckVdYQ8+lk4xDS7KpiHOcroXSbrHmz\/DUng3X\/j5xn2eKywD\/WkE080Vnm90d4UjNDwGTg9yIW55eTnOPfdc8Hxyw9S1pT7yuL2eI9QNGte1xTXq3Px58eb0J5lR\/1jXbAPI2g3jbqm71DGOVPL4uJaLW55LiTBmGmxfmNfrr78OjoBy9IH5xMuYYZMRGtkcBeJUGNOhUcERNZ5XvKhzbQwZ0s8VTjtxVIZTXGTE9op6RmOb7XGsc5Bp8Pxj20PWHO3hCCjdvehLtOsC04tHeB7yGOJp6+NJrxDCyIAJq2WeLDxZ3QtQmJezyxOK\/pGEljwNCza0bDycCNt+eLFknNonH73Z++JFhP7sTbEXzwsoT0z6BwIBZz0M52nZC+XCTfY26UfhlBV7v9yncNSAC9hoxHBhJtNl\/vRjA8DwLCPXCfA46c+RAPrzYsWTnuVgmdzhVPrVDssFk4zLxoL+XoWsmB9P2tpp8FjpxzA06HicXNDHenDD8i4SXhDY4LAHQ8OLfjQoeNFm+dgoumWlMcFGjSMw9KdRSQOK8Rifwh4de4LkTcOSD6pifNYfL97c5wWG+fCCzyFsNph0ZyNG44R+nBLkYmu605iszZAXKo5ocOSH9cIGnBd7xq3Nt3a+vNgwXV54GD5cmBf9mEa4O3WLF1\/qHN3Z6+X0DvWFoxGc4qA7R0cYn3x4fKwHxmN86g0vlvTn+oHaOsXjoNHJdCg00rnQm+GZH91c4f9I+bv+3LrnJOPXFhoU1NPa7vzPeIwfLiw764O8Wa+cnqF\/bbZ0IzumQ\/3hf1dinSO1j92Nwy0ZkFVt3eXFmVPO1DvqIMNSd2mEUD9uvPFG8D\/XxHEEgrrLzgsNSYblqAE7JdQD1uvNN9\/sPBCTfpR4GDMchTcj8Bgov\/71r+nkSCKMnQhx\/NDgYD7hQVkfZO7eDchzmnpJ\/WS7xBEphmdng+Hcxdo8b\/ifdxPxvHbPMzKr6xxkemy3GZ8jWfzvSjR9cf3dLeuU8SMJrwuRdMkto9v2Mq1AIHpbH0lHGafQRQZMoWtAjhw\/R5HYK+QFJLzIvBBx8R4vTJx+ueeee5xnpXBqgcP0dKdxxDVFHHngc1B4VxgNIqbHtCKlQXeJCPhBQLrrB0WlIQI7EpABsyMTuWQhAa7joKHB3mZ48Tga4I4UcfiYIyFc58BRKq4F4JQejR72zriQlMPETIu9WI6UcB1FpDTC89B+\/AQUckcC1Dfp7o5c5CICyRKQAZMsQcVPC4FAILDdGhk3U651cNcQcP6ac9l0490TbhgO0\/IWWLoXFxe7zmAYrqOge+00agJpRwSSJBAISHeTRKjoGSLgTrnzxhRO+9UuBm+A4Fo6rq+r7cf\/vJWcU3zcT4UUpSJRpSkCmSGgXEVABERABPwgEG3KPTxtrlFi54+j2eHu3OfdtFxvxv1UiQyYVJEt4HSPadfOeVhY7eddxPOfcRNBx9EV3mLJOLzVlw\/eohtvSaYbhSMsvNuE7nwVA90oDMNb2eleOw36SwqPAPUvHj2NFIZxEyEWSe\/oRr1005HuuiS0jUWAuhdJJ+NxY9xIaUebcg8Pyzs0ufg73I37XPfFRdG8wSEQ+N+zbujnp8iA8ZGmktpKYEmjRri4osKTMO7WVOL75TNbeAstQ3PL25\/53B1a\/rxllncU8U4XhuNaGN4iymFP3i7K6SauhaEf44anwX1J4RGg\/kl3C6\/ec\/2IU6G3NJ4jTbmHs+JdhuH\/3X3e\/s872XiXXKTHT7jhkt3KgEmWoOJHJNDUXL2IRYv45W2oZ555JvgAN95VxH0+y4YPv+Nt4bwVlQ9i49uQuRaGtxnzlmvO3fK2Yt6SydtM+UwT3l7O57fwwV\/MLFIadJcUJgEvess40WhJd6ORkbufBKiDicrM4mI\/i+CkxTs+2YHkbfGOQ3w\/nkLJgPGETZHqItDIAngRixbxy1ul+TwS3jHEJ9dyny+uc58Rw9ul+ewWzscyAT5zg8+O4DMs+BwGulH40Cu6MzwNGrpFS4N+ksIj4EVvGScaKeluNDJy95MAdTBROX7VqqhF4HRmpCn3qBG2efABnXyCOqev+EwptrWpWsgrA2YbdG38JZBoT8AN728plJoIJE7A1cVEt4nnpBhZQyAPCpKovrrhox16tCl3Ti1xVDFaPD441X2o3zPPPAOOjvPBfdHCJ+MuAyYZeooblUCiPQE3fNQE5SECaSLg6mKi2zQVT9mIQEQCieqrGz5iYubIEepIU+58yjtf2WFBwNc+cDqfIzXcpspQYV6RRAZMJCpyS5qAa90nuk06YyUgAkkSSFRn3fBJZKuoIpA0AVcPE93GyjjSlDtfj+AaMHx3G6fzOSLDbe2pIr7uI5VGjQyYWLUnP88EXOs+0a3nDBVRBHwikKjOuuF9yl7JiIAnAq4eJrr1lFmWRJIBkyUVkW\/FSLQX4IbPNw45cTwq5HYEXF1MdLtdIvojAmkmkKi+uuHTXExfs5MB4ytOJeYSSLQX4IZ342srApki4OpiottMlVf5igAJJKqvbnjGzVWRAZOrNeeh3KFQ9FvmPCQXM4pr3cexRXiYmInKsyAJpFNvCThcHxPZZ1yJCLgE0tjcOlkmoqvhYZ3IOfojAyZHKy5Wsdngjxs3E8OGvYTevR9DIHCjI5063e9s+b9Tp\/scv1Gj3okuErIAABAASURBVEF5+bxYyXnyc637RLeeMlOkvCCQDXpLkInqrBuecSWFRyA0Hxj3IjDsz0Dv24HAtSZ3AJ2ese0TJv+y\/dfM721g1MdA+SKk5OPqYaLblBQmTYnKgEkT6FRn4zb+NFhoqAwb9jLGjfsU5V\/\/gBYju6N1WT\/sVnYi2po0HXkYFqAaNFxGj37HMWRo0IwyY8avcoZb+Ins+5W\/0skNAtH0dlpoKQ4b2QT9y1ri9LImOKtsJxw3MoCVWJ1SvSW1RPQ1PCzjSgqDQCi0wdrXTeh9qhknJsPuBsa9Dnw2Zzl6H1aGs38xHlefeDseO+5cjNx5FFAWQvm7wOiZZsi8ZXEeA0aV+8sqXBcT2Y9VijFjxuDkk08Gn2j+ySef7BCUr2W588470bVr1+38+PA6Pv\/lpJNOAt+JtJ2nj39kwPgIM1NJ8SJAg4VCowTY+vKsQLA5TlnQAY+OehhfzfgJXjv\/SHzx5JF4Y+Sf0LeiM4pKOsD9hEKrMdqMmU42MrM1DdfH27ZePcCLeMtNsXKRQMjG2KmzlHCdaxKsh79WzMK\/rr0XL5QPx92\/\/h3Gj70OT17\/F9xasQi7lrCPufWIQyF\/9ZapetFbxmFcSf4TCIWqbHS7EsMuqYfyaVXAXBtSmfIqdptzLx5\/+BzMOrYbxvfsi9uPPQnX\/\/SXGPPmJbjytjvRrfwF4HfGZyoQWgmM\/rcZMv8HlH9qbj58qYNeJFrW8byN+sEHH8Quu+yCFi1a1CSzevVq5\/kwDzzwAJ5++mm8\/vrr+Oqrr2r8\/dwp8jMxpZV+AiG7CHDUpXy7aaBqpyANSnbHnpiHk1e8jH\/bifPiN8ATDwKHzf0Ih2AGmvdqidqfkF0QhtnUU8jSre2X0P+mFtqLWDR9858A9WtHveVxN0LbkhbYDUsQnBPCpD8CL38NPPk40G7Wt+iCOejUawsDbichv\/SWqXrRW8ZhXEleE6Dx0rv3apS\/1xhYWwF8PxlY86QdcwN0KNkP\/8bR+G54a6Dl5+j+8SY0uHBnrPrLdDyP07DvSRuBVe8Dj6wDbAPbhL63qadbzaBZYEkk+6UOepEo+ZaVlaFv374oKioCX3pLI2XRIjPWwsKXlpbivPPOC3MBmjdvjueeew7t2rUDX+bYpUsXLF++fLswfv2RAeMXyQylQ8OFjXek7KurA3biDMQvWj6Lj7dcj45jD8WW6hEYutejGI+zsXHKt5GigemNttGYiJ7xOjaygF7Eoumb\/wQi6209O\/AA1qA5JuEkPHHgYOy65SjsN7YlOm7ugbEHDzP3\/lg4ZUcDxiL6o7dMyIveMg7jStJIIP1ZlZfXR2jJbsCmnSzzH01oeext25b46PvD0R4LMe3ZYzHo48VY88kW\/NhjP1y0vgJXVt+JKd\/1AXodZXFnbDVgGHUDEFoMG\/22JJL9Uge9SJR8+cqA4uLiGt9WrVphyZIlNf+507hxY262k0AggI4dOzpus2bNwvz588En+joOPv\/IgPEZaLqTe\/TR6OOPGyatQyWa4SMchr9tvBB\/GvIs7tx0JT7E4ahCQ6yt1y1qcXmBieoZj4eXngDjREl7\/fr14FMfzzzzTJx77rmOhc+g69atw\/Dhw533bZx99tlYtmwZnTF58mRw\/rV\/\/\/544oknHDf+1DWnyzCS1BOIrrcbsOzFzpiF\/fEOemIcSvFk6S34Gy50\/i\/G7li0bgCifZLWWyZMPfQijBtBpLsRoOSo06PvmeFykBW+YwP7aWtS32SjyVdof+LX+M3nf8fr6IsBRS\/j1BOn4Pgj3nL0tvSbv2G\/zu+ZpWJBOx0JzN8EzLX9ZSargPK31thOkl8vOss4SWYbLfqMGTNw6aWXglNJHMWJFi4ZdxkwydDLgrjBYIvopVg2D6sOWY2Fn3bAt6vaYdGP7bBwTQd8Zz2Iby9oAkx+J2rcYPB\/lnfUQLE8vPQEGCdKmpPNIAkEAuDjqv\/617\/iD3\/4A6qrqzF+\/Hjwbb9cNDZ48GDcc8892LhxI2655RY8\/vjjoPuECROwePFixDOnGyV7OftMILLeWqOOzcDKMpQfeAz+PuMCPLp5CJ6t\/jmeXDsI4z4sxYSfngV88GrU0iStt0y5kf3EIagdxqJF+kp3I1HJTbfgoVZuitkg2L+V\/fmZyXyTCiy6bCZK\/jYOoxucjnOOuB93\/\/ICPNb8KHzRpgHWP74Kn79XD0U72ejhgXbZrb8TsMyifWfyDRAM2jCM7Sb1ra2PcfwfU1QcNcvWrVtj1apVNf6cBmrbtm3N\/1g7CxcuxGWXXQa+cqBTp06xgiblV5RUbEXOOIGRI3vGLsOMx1B98P3Ysuu92NjsPqDVLajc\/R7gofujxguaUVRnulFjb\/OI4+TZ4QLAONui197wxPnuu++wadMmrFixAnvuuScCgQCmTJmCAQMGOMFPPPFETJs2DTPM8u\/WrRtatmyJBg0agG9VnTp1KuKZ03US0k\/KCcTWL2vkP78dOOTvqKr3H6woWoRNTV8AjrjRjG4TVEUsny96y5Sph16EcSOIdDcClBx1GjnICr6Hic0Eob9t+3Gq5ELb2dekK8of6Ir9+6zBzVcPx323X4Q2L21Ax\/MDWLW+GPWPOQJbflcNvGHSOgCYDYN5QLDVPIwc2dTiJ\/n1oLMjmv\/PQKmdO9tNLsDlnUYLFixAZWWlsxaGU0t891Ht8O5\/huc7kR599FGwHXbdU7GVAZMKqmlMM2DGxi4Vg+vIcTXqB+2iEGyARiWtY4atH9wZS4d2QacSnqXw\/uH56EWi5MiXgjVp0gRnnHGGY7CMGDHCCcmTiavg+ad+\/frOSUY3ztfSjcKexNKlS0H34uJiOjnCMLXndB2PnPjJ7ULuFGyGDhUD6ziIpUArmyItngJ0jmy0uAnUCzbB90P3QucSXlBcV49bL3rLOFGyk+5GAZOLzj\/YKOHtNuKyixX+cJNfmIzcGeh+qu0cYNIJX5QdhjG\/GY27T78STU76Bgtu3Q24vRjL\/mBt74oiYP8A0MaCLjJZ8glCodcRDJqf\/U3qSx30IlEyPeCAA5y1K6eccoozmnLbbbc5IT\/88ENnZIV\/br75ZnBanyM13PLFjTNnzsQXX3yBa6+91vGj+xtvvMHgvovR9D1NJZhGAjYOAV4MghUnY8+Re0bNeWPoRyC0GlXl30UM0zS4E3YvbYM9ynph51GHWcfAegkRQ8bp6KE3MKayOGriHIanZf\/MM8\/gzTffBE+UtWvXRg0vj+wm0NG6n\/XM6OhW0QedRnaIXtjlq4FVJnOtqxohVPNgAB1Ld0Hnsh5oMeogLOAUFJL8SHeTBJi\/0YPBesCqjcAVs4Bxn8B5jDjtlqvtmG+z9uuytsDPG+G7PntjbqeDEDrjFKC0GTCkBdAvAHQxMXXGJ2aQr5xqkaaZFCMUMsPI9pL6etBbME6MTL28jfqQQw5xpus53e8KR8djZOPZSwaMZ3TZE7E96gPBYrQetR+Or9gPPx3bCP1H\/oDdg+sQ69PW\/I8prUTfsfVxgsXrMPZQbA62Qjs0RE\/EVo1Y6Tp+HnoCI4LRhzM\/+ugjHHrooahXrx523XVX7Lbbbs7qdo6urFy50smSa1+aNWsGunG+1nG0H468MA7d2VMwJ+fLMBzed\/7oJ+0EOmInFAWbo\/2ovR297T22KY4fuQGtgptjlmWXYDW6l1bjlLHAuRXF2H9sZzQINgbT68FzIWbsODylu3FAKtwgwX2bmBFjhswr1hk835Twjx8B\/1gDx5jpaFy6mxxk0sWEuvSjbb8yscFEfLQe+NoMn0MbAc17mGMJgsEDUVJiadq\/pL7My4sklWlmIxdlNnvl7geBMViBSjt7lqM1VgQ7oaj0ADNmuuFXFTvj\/urFeLDic4yvmIYHyt7H3WUf46GKmXiwej4uqGiI\/cZ2sfAHYgE6YJnF36V6JWzgM\/li2fnpWPeJbqPk3KVLF7z33nvOGhgaLLw1j7fq9erVC6+88ooTi9uePXuie\/fu4O17XCuzYcMGvPXWW2C4aHO6TmT9pJ3A7VvWgXq7CsWoCu6GJqU2BTSqIy6pAO6oXo1RFcvxx4pluKxsJYaXrcH1FWtwa3UlLq\/YhBPMSC8u7Ygv0RVLsDtabF6D36E5fPkkqrNu+CiZS3ejgMlR55EjW1vJdzbZC\/jBLJbPZgBvPQUMfwi4yvZfMQN8ZxulOcaCHGnS3jqSHy8DZr0OfP8Z0LI7MNMMmTXfmmcjjPzTPrb14evqYaJbH7LOVBI5asBkCld25tsT8zF9zU\/QZMNaLLbxk2+wFz7HAZiKHngTJ+Lt4Gl4M\/gLfFwyEJ+VnIzy4KmYhP54Dz0wAweD4RfbRaDZhkr8ct1TGIz\/3XYMr5+mFtGLWLRIXz7OmqMnPXr0wK9+9SuMGjUKfEjSOeecg9mzZzu3Ub\/wwgvg2hiuhbn++usxZMgQDBw4EEOHDkWbNm0QbU43Un5ySz2BozYvxjsLe6LR+ip8j9ZYhA74Cl0wEwdhFrrh+2Bn0CBHyZ7YUhLEguAB+ABHYAp6mvTCdBzu6G7D9esxuPIJnLHxGX8K7UVvGSdK7tLdKGBy1DnYyUZf0MpK\/4NJV5NeJj8F6p0DNOoOvLsIuGoxcGo5MPRJ4O5\/AOtspAYfA9VvAyvM0Nn8psWxEZx9G8F6Zbbvw5c66EV8yDpTSRRlKmPl6x+BaszDnlvm4fO5B2DEN\/dg74XfIIQ9sfVicDBmoDvex1HYKkdaw\/8TfIYDMAddEUIQ9ddsxCVfP4BP5xyEa368DUAISX\/svIQXiZIx7yZ69tln8cEHH+D5559Hnz59nJANGzZ0njNAt3HjxjmPtaYH\/SdOnIhJkyZh0KBBdHIk0pyu46Gf9BPYHELHqgX48IvDcdnc+9DpuwpwJPArdIGrs+\/hWLxr8h6Ow4c4HJ\/iYNDgXrClIxqtrcKls+\/F+\/8+Chct+ytg6cGPjxe9ZZwoeUt3o4DJUefQEiv4TqzwNcDe84CddjaHXYBNtqn60YyUBrbDQCtsS7ERGOxk+8Ume5h0AhoHgaaHALNbIxSinzkn+2WR4pHaYZLNN4PxizKYt7L2iUARhgANLbGmwDUrbsOr0\/vhk3GHYsLfzsKv\/\/V3lLxWjuD0edhj+gLsOX0+er9ajp+WT8atE67By\/86BZ9NPhDXfn+LnVBMw04slCDpj5UFXiTpjJVAzhBoVAo0sNKaXLzkL3j2Pz\/HtH\/1wCPPn4dzp\/wTPT6Yho6zFiA4ax72\/u\/XOOE\/kzFg5su4\/e2r8VpZP3z0xmE2nXQPHEO5geltPR\/01orjSW+p64wryXsCpafaIe4TsJ8Dga8XApsfAzqsAE5vAPx6J+BndlndqwWwb2fgoKNRdMpA4NRzgbOsnT7sTIs3B9i4L9ANCP6sIUr61jM3H77UQS\/iQ9aZSsJIZypr5esngep142Cj8EB7S3UvoOMeC3Bcq3dxnRkmf5t5ISZNPAmTXjR5+SSSKTAUAAAQAElEQVT8deZF+NusCzE48ASOLv43YOcZGC9gnYcfe6MIdTxbxrKo80uDyovUmbAC5BWBIhtaZ6Pb3I6qNbBr46U4Yt1\/cPnce3DPzN\/iuXdOx3NTTN47HffOugx3f3Y5zvjuGRy8ZiYcQ8PimMICm0qAeiaWTNJfL3rLOElnXBAJ5MVBjv2LHUZXTiOZcYJ+wMJZwHOPAA9NAd7eCOzSFdjDGuL2B2PL1x2AiTYiM+FB4KP3gOYXA\/vVA6qAEhtILjnK0vLj29AS8SIWLdq3rieX887QO++8E1272vFGSySF7kUpTFtJp5FA4PFx2HJ1EaqXWqZBE64LO8i2R5oca8K2vbdtKcfZ9giT7ibWEdjS2eLVC2DLIwEEJlabow9fXpS8iA9ZK4kcIvDgo8B1Vt7FJm1M2pt0NNl9m7QN27azfVe39wO2dAtsNWIsCWxdx20BfPh60VvG8SFrJZEbBB6dYOX81trKrmaFdLaGdrf+QNtfA8UnAhtbAtO\/Bt6oAF6dCswqAzavswhmrcCMmTULgM\/WACvNyaLbrz\/fppaMF7Fokb7xPLk80tuoI6WVKjcZMKkim+50WwZR9P0WBK6yhv2XRdgyPoDN3wawZX+rYhvpBI2VQwHYdkv3ADY1AzY3CmDzu0UoumwLAudWo+hLOyGX+7D+BfZp5FEsmr4FRGBXs0hs9B1\/tGO+xISGyPe2NQMFXWxrBjb2t639rz4A2Gg9zGrrvMIGDot+b\/p6vvlZ5xeL49RbC17nV7pbJ6JCDxAMGoFNZkB\/adNB340HWq4HOAixt7n\/pDHQyXaammGDY8yBQyxmuKAt0GQnoM0n5va+KTMQWm27fn191tt4nlxeWlqK8847z68jSDgdu7olHEcRspHA0UPt5LCCtQCKqrag6LFq7DS4GkXFW7DezqM1ZristrnbH0w27l+NegcDOx1djZ1u2gLMtXhm0MDOO\/SwdOxv0l8vPQHGSTpjJZBTBPqYvtW3EtMoYWP+pO1fbXK0iXVm1x9vI+2Dzdg+Gwj0AOpbRzfwCwD3mnxj0mCbnGDp2K4vX+qhF\/ElcyWSCwSGnmalpIENa0jX2tDgf8cCUz6wkZc1Np30JfCDWdi7fWaGzCKbF7UppSJrZ2HTSGtfB5atssg2BL4HMHSg7fr19aKzjBMlfz4\/q7i4uMY30pPLI72NuiZCGnaK0pCHskgHgb1sjmiPUqCVZdbWhA\/l3cu21oNt2BhoboZ\/i\/lAs4UA\/8N6tLBOAjhcb+cfGlrYLdatOMLSsN2kvz73BpIuT3YkoFLUJrB\/CTZ3NZ3jGhg2pmaAg23mbgDMraH9b1RpxjaH2um+h7m3NmHYnYEt1Wbc1DO97WNpmLMvX+muLxjzOZESa1dLh9gRHtoQaGltL86yP2ao4BlgySTge2tsv\/kOqLDpoqUfmqKaG76wMBa2xVDgoKYI2jR\/6XHm5Ne3APVWBoxfypMN6dw+DptsbvZHjqZ3sAJZu45tRgz2t\/82BI9utrWTD3bywPw32kWhyobwN\/H8es4iLjSBDx9eYLyID1kridwisO6ucVjzBvDjQit3KxPTSVB2AcD\/FDNkQLdtsqGBjcxYZ7fKOr0bJ4VQHQrBt48XvWUc3wqghHKBwLiHrJQhM1B2t+mgHqag3Y8Fgjad0vxyoOiX5mn\/YfP2RX2Bxr8BdrOhxH2td7n7TGD+mwhZWx1aZsH8+lIHE5QxU4qj5p4LTy6XARO1+nLQo3MQ9X6wHqmdT\/MeBGa9bHOsdlGYsxiYtw5YZo3+\/A3AgrXA7M+Br94GFtmQfdWbcOJhD7NoOpj4cegF2BvwA1shphEIBrHFRlh+nG2d15eAuTb6HjKdnWWN+zc\/AtaPhfMqL9Ptj23aaOZ\/gG9Md3+YbtcJi1fP4jMN39hJd31Dmc8JBdkpXLEr8MVcYOorWxV316lAZzOmDzXl7WENbffvgYNt6mi3ldbIfmsNrzXOs20kZlU1gh3XIsiF635B8qC3I37O6azIBYj25HJOLc2ZMydypDS7FqU5P2WXSgJ\/HQuYBc7lLC0tn11WA5vsfGEvddlkoOI5YKn1dFeYwbLTLGBnO59of1sU+2MRhg61H5++TNSL+JS9kskdAo3HjkU9Ky6FbXBTa\/c3fwmsN1n2MTDnPWCxbVd8ZGpq14bWNmJo\/ViqOuoFgHp+6q2Vw0lYuksSkhgExt5onvvZFBJOsB0zUtZYo\/ofs65nvApMf9eMGlPcGVOAT8y4CT1hvcWnLJwpNXa27V4Yel4T2\/r4bWppeRGLFul7gMe3UUdKK1VuRalKOEK6ckoxgU3tgpi\/3jLZBWhm55X1DdDB\/u5hEjTpbNLJhMtjuPRlN9svrg9sbm5TtrZfdYTNz9rWly+vRF7El8yVSC4R2BgMosIK3MCERgybd3ZMqadcJ8mOLpdscZ96yxklGulmu+C7auCHkhKL6ePXi94yjo9FUFLZTyBo7SxW2AjGYWxpuQaLc\/StreCbTPj2RhorNpQIGzqEDYGDzwb4ifkdDuzcECW0e+yfb1\/qoBeJUYBITy7v168f7rvvPvBz3XXX4cknnwRHZLi94oor6Jw2kQGTNtSpz2hTKIQ1G4H3VwJzrWYXmjXeyE6yYmvtW5tiU1rZlv832ZTtSpMv7IrxyRpgiQ3VV5WX+1dIyxtexL8SKKUcIVBleltpZZ1pYoOCvFcDjW2f\/VMKDRpTW9DAsRkjLDc\/G4yBDS7C+r1YW15uLj5+vegt4\/hYBCWV\/QQ4zYnvzDj56C5gnw+BDtbYtjwXKLrYCn+hyfkml5ixciXQzv7vczKwO++csLnPH99H+WvWUFuI+L5xhGpqYbyIRcvVr13mcrXoKndtAvWsJ2v2CCiLzOD\/r10V\/mXnyPM2JP+vDcDTJtz+086512x66UOTZRaO4XeyxBqVlNivT19ecbxIjOz5XqOTTjrJeUnjfzhUa2HXrVuH4cOHOy9zPPvss7Fs2TJzBSZPngyG7d+\/P554woZvHVegridLbgumTRoJNDS93Wz5BUy+N\/na5EmT501eCpPxtj\/JhOYK18XYLmwABs1KSrjrn3jRW8aJUQLpbgw4OeoV5HBgE\/7sbvOcNnW00LRzhWlnI5s+avwY0Ni0uMnjwI82WvGtmdzfmIYv3mJH285kV5SU0CS3Xb++1EEv4lf+GUhHBkwGoKcqSxow7ceOdQY+mlkmXN\/S2rY72znT2MTdcsCT7vSnwc7e7m4jR8JXA4YJexErb6TvmjVrMHr0aDz++OPO8OU999zjBBs\/fjz22Wcf8GWOgwcPBt03btyIW265xQlL9wkTJmDx4sWI58mSTqL6SSuBRsEgupjeMlM26TYwCLskwGZBYWoLDsjTj+7UWz6slzpM\/d3D9LZJSQm9\/RMvess4UUqQD7ob5dAK2jloyjj2OSI41X7YitowNuYAP5oWr7PRlnX9gLU9za+XiWnx5rdsO9VkGUaO7IqSEo4v2l+\/vtRBL+JX\/hlIRwZMBqCnMksbXMF8y8AGW8AG3gY1sRvgSNuwLd250Jen3QpzXzxvnv36+PXSE2CcKEWYMmUKTjjhBLRs2RItWrQADRcGpfuAAQO4ixNPPBHTpk3DjBkz0K1bNydsgwYNwNX0U6dORTxPlnQS0k\/aCXBqiNNBNigIqgH1k+u3glaSvUw6m3BNDPuurWy\/vgnfmvGt33pr6ToFYCESFcaNINRR6W4EMPng9K0NYTeyURccb0dzmAknPLn25U3bf9GE44gTbfuxyXITdi3rY948trr2189vI0vMi1i0XP3KgMnVmotS7pV2oafxMtf8+cDqkG053M6+gc0kwU438CJB9\/+a32cmPK3WhkK259+3ql4jeJFoJVixYgXq1auH66+\/Hpdffjnef58rIADe0rfLLrzcAfXr10dlZaXjxqdGumnxeQZLly513IuLi11nMMySJbxs1jhpJ0MElpreMmsaJZ\/aDrWRNbPG9lcDzroX6jF1dpa5cb0M9bYyFLJ\/\/n696C3jRCuFdDcamdx3nzJlPVC1wA7kVRO2vE1syy4jb53oavtHmNhIzE5H29ZGZsBu424IhSyeufj5pQ56ET\/LkO60itKdofJLLYFdevWCs57FsrEZV6y0LUdkvrAtjRUK18fTnb1e9mTrmV+TYNB+\/ftWWTfWi0QrwYYNGzB79mz89re\/xcUXX4yrr77aMVaihZd7bhFobXrLErMDyUaJxgkNGOrqbPOwgXnwMkEjnMY4dZuyczBovv5+vegt40QrhXQ3Gpncd+\/1M3aIaLCssoPhqMom236+TXhvHc3tycBmtrwMy8nPJggGubVgPn6rfG5zfSxaypJiW5GyxJVw+gm0Ky1FBxNeCJpa9hQOWnKMwqZsHfufbhQOdrK\/sKeF33\/bGgSL4su3Ek2RqNw3pnnUvDnKcsghh6BNmzbo0qULOnXqhAULFoCjKytX0hwDuPalWbNmjtvy5bwEbk2OozS77rqr475q1aqtjvbLMG3btrW92F\/5pp5AJ9PB3U1oTFNfKdxnn5V6yqlO7lO4T03pYuEP9VlveaSJ6i3DS3dJrvCk9EygtJT3Q3PEZY0BaGhiIy44ybbdTXqanGhyqMkuJnNQWlqNsWMZ3v76+K300OYyTqwi1HXTw\/z583HGGWc4N1Fcc8012LyZy\/GBRx55BOeeey7OOuss3HbbbdiyhavZYuXkzU8GjDduWR1rSSiEpVZC9lZZwTxteEGg0cKFkMXmRwNnk22\/N\/m6vNx+\/f1WeegNnDWiQdRCHH300fjggw9QXV0NGiwhO8aOHTuil\/XcX3nlFScetz179kT37t0xa9YscOievd+33nrLCce1MK+\/\/rpzMtH44XRT+\/btnbj6yTyB761O+dSMH60oHEXkWpc2tk\/Dm\/1VCvWYflzrNae83Hz9\/0p3\/WeazymGQhxp4dQQW9c1dqjfmHAdTMi2002mmXAEhuPgc1BezpbZnHz+etFbxolWjHhuerjxxhtx1VVXOTdRcL0h22B2GLlG8Z\/\/\/Cd4AwWn+z\/66KNo2STlXpRUbEVOA4HEs\/iuvBy0gzkuwekjnjo8pbguhkPy\/M8tbzjmNBLXv1ASzyl6jEqfewM0NGjRDxs2DKWlpbjhhhvQtGlTnHPOOc7U0mmnnYYXXngBI0aMcNbCcK3MkCFDMHDgQAwdOtQZuYn2ZMnoRyGfdBJYbHrLfhqbdz4PhlNHvARQqK9cG8Ptd1aoTSY\/hEJIxRoY6a7B1TduAuXlNGDWWPhKE2outZQ3RbCFpTnujiVy0rMSodAaE66XseA+fit9bnPLysrQt29fFBUVge0vb55YtGjRdiWmkXPEEVznA+cminfeeQfNmzd3Opo0ZPiYC46+tGvXbrt4fv2RAeMXySxKp\/vIkXDXtnCkhcLnZdCo4QWisZWVwtOK0sUMgibBoLn6963yMALDOLFKcPrpp2PcuHF48cUXnTuSGLZhv1v+DgAAEABJREFUw4Z44IEHnB4A\/TjVRPc+ffpg4sSJ4PM3Bg0aRCdHIj1Z0vHQT8YJ\/MT0lmNw1FNObXLqiIXif466UGfpRn3mpaCr6W3TYJBBfBXqoReJVQjpbiw6ue03cuRBdgBclMsxbsom+88WmNrKFpat7mpzW2WyG0pL90IwSE23vz5+q3xuc2mAFBcX15Sw9k0PHAmnUeMGoD+fw8WRmAsuuMB5Dhc7ljRwaAC54fzcFtWVmPxzj8ABo0Zh52AQbPDZ2POU4gWBw+\/cp9Cdp1BzC3dklqwjYA8i92irxH4R6L5Nb7mKgE0+t9RVrnfhlkKdpR\/1u1cK9JbHQj30IowrKTwCo0Z1QtB5qh01dG8D0M6EBg2NFt73aX\/R2X5+YuH2xNix9Le\/Pn+96OyEMf6PBPHOTk4fvfHGG05n88svv8Qnn\/CeWJ8P2JKTAWMQ8u3LofWlNrzOAUza\/FV2gJtM3C\/\/89TimnmG49C96+fXtsrn3oBf5VI62UtgjensChPqLNe4UE\/Z\/G+0IlO4v8b2Kcst3Lc25WR\/ff9Kd31HmtcJhkIbEArNsGNcaMKJeo4PtrX9PU12M6EpzlWJiyzcNygv5zSTOfv89aK3Px0R3ZjiDRKrVq2qKWXtmx442r16Na8kW4PQnzdL\/Pe\/\/wXXJ9KfI+SHH364DJitiPQbD4HKUMgJxmH3TbZHFaTQoOGiXe7zgpBK67XS5\/lYOwx985wADW8eIgfdueVaGK4qYBPJe8q4pd7Sj3rNbSpEupsKqvmbZsh5pgunjNbYQXLkhdrKlVrz7T8NF5riHDukIcOx8NS0vJU+t7m9e\/dGpJseOLXElzfawYHrCnlzBfc5ZV9SUuI8Gf3jjz8Gn0C9adMmcAHvvvvuyyC+S2pI+l5MJZgIgd1NiTqa8JTiVJEr7pSS+5+nVLFNITE8fP5UaQTGZ6L5n1x701nqoqu3NGS4T0OcTX+43rY2vW1n4VNBJW90NxVwlOYOBEpKmqGk5DBzZ4vKKRmufeG9n7yPjncncfKel9qNNoVUz8JydMaC+\/z1W29pnHD9yimnnILLLrvMuR2aRf7www+d17lwnzdT3HXXXc5t1I0aNXLWvXC9C2+ecOMeeeSROOaYYxjcdyFV3xNVgpkn0L+sDKea7GGNfCTDpZO59xk7FmdWcAW9\/+X1uzfgfwmVYjYSOM10doBJB9NPrijgpYDCfiv1uLO59zW9PSdFeksm0l1SkCRCoKysB8rKfmXGSSeLxrHDNbbl+CHHu5eZe0OMHXsIKipOM\/fUfCt9HoFhKSPd9NCvX78aA4ZTRU8\/\/bRzE8WoUaMQCATAz3XXXQc+eJQ3UVx00UV0SokUpSRVJZoVBNib7WmN\/TIrzcow4dqXA0eOxD6lpeaamq\/fvYHUlFKpppCA56Q5ElNieutOd66xlHgZoA4fnGK9tawg3SUFSaIESkpam5HCh9rx1mkaMRSu3FqHkSOPQGlpaqZR3HIWot7KgHFrP0+3xcEgaBNvBpxnw3Bruyn\/FuLJlHKoBZRBcTAIfnjHkSv8nw6R7qaDcn7mEQwW24Fx0pMPrqDmcsv\/bIXNK4XfWHq71kZnokkKi5TypGXApBxx5jPg6VPPiuFKq2AQLUzMKWXftWgCL+JbgZRQ3hGgzlJSfWBe9JZxUl0upZ8rBDjh2cAKyxVc9cH3HgWDXAtjTin8UgejSRUa2shiZElhkVKetAyYlCPOfAYXVVTgWBt6P7C0FJSzy8pSbsDE6g3E8ss8LZUgWwhcYnrb0\/T2INNbyrmmt8UpNrxhn1j6GcvPouorAqioON+mjE60KaOjTI5GWdkvzYjhSq7Uwomlm7H8Uluq1KYuAyY1fLMqVfZajx01Cv3HjnWE\/1NdwEobsvQiqS6X0s8dAsVmrPQ0vT3F9JbC\/0jDx4veMk4aiqYscoBAMLgzRo06CGPHHmtyTFqMF2KhDnoRxs1VkQGTqzWX5eWOZfHH8svyw1LxCoBALP2M5VcAaHSIWUwglm7G8ot1SF7fRv3VV1+Br8\/g26gff\/zxWFlE8IvfSQZM\/KwUMgEClRqBSYCWgmYTAeluNtWGyhIvAb\/1li9qnD59Ol566SXnfXPXXnvtDkW58cYbcdVVVzm3UfMdSK+88ooT5uqrrwZlwoQJ4CsFwp\/Y6wTw6UcGjE8glcz2BGJZ\/LH8tk9lx38bN27E8ccfj6eeesrx5NtOhw8f7jxI6eyzzwZfJkaPyZMnOw9V6t+\/P5544gk6OVJXj8IJpJ+CJhBLP2P51QVNulsXoezwz9VSxNLNWH7RjresrAx9+\/ZFUVER+HC6Fi1aYNGiRdsFp5HDB9bR8cQTT8Q777zjtMF8Cu9RRx1FZ\/C9SIzr\/PH5RwaMz0CV3FYCfvcGtqYKPPjgg+D7Ndz\/48ePdx5d\/fzzz2Pw4MG45557wAvFLbfcAg5d0p29gMWLF4MnW109CjddbQuXgHS3cOs+l4\/cb73lKwOKi4trkLRq1Qp8UaPrsHLlSoQbJvRnB3LFihVo164d+ITeCy64AE8++aQbxfetDBjfkSpBEohl8cfyY9xoQuuf79UYMGCA0ytguClTpoD\/uc8ewLRp0zBjxgx069YNLVu2BIc1+U6PqVOnIp4eBdORFDaBWPoZyy8Wtfh1V7obi6P8ohOIpZux\/KKn6M2nqqoKn376qTMCfuutt+K1114D22VvqcWOJQMmNh\/5eiTgd2+AxeCoyk033YTq6mps2cKHRAHsJfCtp\/SvX78+KisrHTf2BuhG4VtVly5d6rjH6lEwrEQEpLvSgVwk4EVvV415Meqhst308jZqxuvcuTP2228\/pxPJKaYvv\/wyaj7JeMiASYae4kYlEMvij+o3Zuu6lkiJciEYX8vOoclI\/nLzh4BSAaLqJxohqp90V6qTYQJRdRPR9XbTiPOilpoj16+\/\/rrTWVywYIHTOeRaGHYa58yZ48TjCx8\/+OADZ3\/ixIkoKSlxpo\/Wr1\/vrIXZtGkTOGretWtXJ4zfPzJg\/Caq9BwCXnoDlSOGO3Ej\/TzyyCMYPXo0aNnfeeed4Ip4LuSltc+5WMbh2pdmzZqBbsuXL6eTIzzhdt11V8c9Vo\/CCayfgicg3S14FchJAJ70FnxlauTDpXHC0ZNE30YdCASc9S\/XX389GPeggw7S26gjI5ZrthLw0huoQqOoh8OFYHPnzgXlyiuvBKeTBg0ahF69esG9dY\/bnj17onv37pg1axa4mGzDhg146623nHDRehRRM5VHQRKQ7hZktef8QfuttwRyySWXgCMrvBmCBg3d+vXrh\/vuuw\/8dOzYEU8\/\/bRzG\/WoUaMQCGx95xOnj\/7+979j0qRJGDFiBIOmRDQCkxKsStTv3kA0oueccw5mz57t3Eb9wgsvOCcL18LQ+h8yZAgGDhyIoUOHok2bNuAJGKlHES1tuRcmAeluYdZ7rh91uvQ2mzjJgAmrDe36SKDK0vIiFq2u78UXXwyOvjAcb6l+4IEHnB7AuHHj4C7o7dOnj9NzYA\/ADcvwkXoUdJeIQA0BL3rLODUJRN+R7kZnI58kCVAHvUiS2WYyugyYTNLP57wr7eC8iEXTVwQySsCL3jJORgutzAueAHXQi\/gDLiOpyIDJCPYCyNRLT4BxCgCNDjHLCVAPvUiWH5aKl+cEvOgs4+QwFhkwOVx5WV10Lz0Bxsnqg1LhCoIA9dCLFAScLD1IFQvworOMk8PsZMDkcOVlddFp2XuRrD4oFa4gCHjRW8YpCDg6yKwlQB30IjEOqK53x82fPx9nnHGGcxPFNddcg82bN2+XGu8YveKKK7Zz8\/OPDJgoNHmbGJ85Ei58u2aU4HKuTYCWvRepnY7+J0RAepsQrsiBE9fbrb3fyKnJNQ4C0ts4INUVxGe9jefdcTfeeCOuuuoq5yYKvraFj7Jwi8kH2PFxFu7\/VGxlwESheumllzrPHOFzR\/guh0aNGtXc+RIlipzDCXjpCTBOeBraT5iA9DZhZDtGoB56kR1TkkucBKS3cYKKFcyLzjJOlDTjeXccjRw+moJJ8F10fBs19zkSc\/vtt+P3v\/99zbNh6O63FPmdYL6lt27dOvzmN79xrMzDDjss3w4vdcfjc28gdQXNg5QjHIL0NgKUeJ2ku\/GS8j2c9DYJpD7rLZ9gXlxcXFMgvl9uyZIlNf\/5BPQWLVrU\/Kf\/smXLnP98cvqvf\/1rNG3a1Hl3neOYgh8ZMHVAverii3GgVcKwYcPqCCnv7QjQsvci2yWiP14JSG+9krN4XvSWcSyqvskRkN4mwY866EWSyDJS1MWLFztPQj\/hhBMiefvqJgMmBk4+vv4bszjvmDgxRqiC94oMwOfeQORM5BqJgPQ2EpUE3KS7CcDyL6j0NkmWHvS2+NUxUTPlO+VivTuODw1dvXp1TXy+f47vnOM76l588UXnvXVc4MvXEKRqIa8MmBr82+\/wbZt\/\/vOf8fDDD6Ph9l76Fw8BLz0BxoknbYWJSkB6GxVN\/B7UQy8Sfw4KWYuA9LYWEC9\/PejsqsOjv6eod+\/e8PI26t\/+9rc160efeeYZ5w6lu+66y8sR1Rkn9w2YOg\/RWwAC5xzgcccdh86dOjnW5JlnnuktsUKM5aE34DzHoBBZ+XjM0lsfYEp3fYCYWBLS28R4RQzts95Ge3fchx9+WPMyxxtuuAGsu9NOOw280eWkk06KWLRUOcqAiUL2wQcfrLEi51ZUOPsc4owSXM61Caw3By9i0aJ9b7rpJvAEOfXUU\/Hyyy87wbjob\/jw4Y6Vf\/bZZ8NdRDZ58mQnbP\/+\/fHEE084YflT13MNGCaXRXrrQ+150VvGiZG1dDcGHPOS3hqEZL\/UwQQFDB8j30jvjuvXr1+NAdOxY0c8\/fTTzm3Uo0aN2uGOI974QgMHKfrIgEkR2IJP1ufewNtvv42ZM2c6J8odd9yBP\/3pTw7i8ePHY5999nHcBw8ejHvuuQcbN27ELbfcgscff9xxnzBhAriwjLf8TZ8+HS+99BL4Ashrr73WSUM\/IrAdAenudjj0J0cI+Ky3uXDUMmCSrKU1a9aga9eu4LCam1RlZSX23Xdf54LruhXctsqO2ItYtEjfY489Fvfeey\/49mkaLNXV1di0aROmTJmCAQMGOFH4HIJp06ZhxowZ6NatG1q2bAk+XIlzuVOnTkU8zzVwEiqAH+ltjEr2oreMEyVJ6W4UMEjcXXobgxl10IvESDLbvWTAJFlDzZs3d6YquPLaTeqNN95A+\/btcfDBB7tOhbf1uTdAQ6Rt27YOR04JHX\/88ahXrx64Tomr4elRv3590HikG59JQDcKV9MvXbrUCVtcXEwnRxgm\/LkGjmOB\/EhvY1S0dDcGnMx6SW9j8PdZb2PklDVeMmB8qAreKvbJ889j7dq1Tmp8nDLdnD+F+uOlJ8A4dfDilBGngEaOHFlHSHnXRYA6Wsh6G5UP9dCLRE1wq4d0dyuHZH+lt1EIetFZxi\/A34sAABAASURBVImSXC44y4DxoZaOadcOdy5fjiV9+4L3xXO64uc\/\/7kPKedwEh56A8Vzx8Q8YI5scV3LQw89hMaNGzthObqycuVKZ59rX5o1awa68ZkEjqP9cESGzyege6znGljQgvpKb6NUt3Q3CpjscJbeRqkHD3qb63d+yoCJogtxO4dCQO\/e2HjMMbiydWu8+uqr6NGjB9q0aYOC\/tCyT1BWNY\/+TAIuwqXhwucKcBjZZdurVy9wxIv\/ue3Zsye6d+\/uPAlyxYoV2LBhA9566y0wHNfCRHquAeOmXzKcYygkvY1WBQnqLSy8dDcaTJ\/dQyHpbTSkpofUxYQlWno54C4DJplKCoWckwlDh6LDww87i0d5kR04cGAyqeZH3Eo7DC9i0SJ9X3jhBSxatAjnnXce+DweCv+fc845mD17tnMbNcOMGDECXAtz\/fXXY8iQIWBdDLX6oUEZ7bkGkfLLa7dQSHobq4K96C3jREmTekldle5GARSvcygkvY3FijroRWKkWddjJ+bPnw9O6fE5MNdccw34Ekcmx6fv0o2PveBLHemWCpEB45VqKFRzMmHUKLSzaST28vkckr42leQ12XTFS3k+PvcGLrroIvAOIz6LxxUulOZdSbwlmifMuHHj4C7o7dOnDyZOnIhJkyZh0KBBNYcb6bkGNZ6FsBMKSW\/rqmfpbl2E0u8fCklv66Lus97G89iJG2+8EVdddZXzuAreaMFRcC6juPnmm51HVfAZMRz1\/uqrr+oqvSd\/GTBesJnBwmkjcCEp97elseeee+L00093RgC2ORXuxueTqXBB+njk1FWb7pTe1sFUulsHoDR7S2\/jA+6z3sbz2AkaOUcccYRTPj7G4p133gGn+J977jmnU8+3UXfp0gXhaxKdwD79ZMiA8an0mUiGJ9OjjwJlZUBpqVOCLVu2OKMDtDZLS0sdt4L\/8TKUyTgFDy5FAKS38YOlHnqR+HNQyHgJSG\/jJQVnQa6PesubH4qLi2vyr\/3YCd480aJFi+38OQMRCATAJ\/TSY9asWeA0k2vk0M1PkQETL81QCAg\/mYLBmpgnn3wyzj\/\/fOvYjkQwGKxxL+gdn3sDBc0ymYMPhQDpbWIEpbuJ8UpF6FAIkN5GJhvN1YPeFm+JfedntKzicecDRS+99FJnKqmoKDWmRmpSjefoci3M6NHAvHlARQUQDG5Xeq6zoKXJhaXbeRTyHy89AcYpZGapOHbpbeJUqYdeJPGcFCMaAeltNDLR3T3o7KoVI6KmV9djJ7jekOtd3AQ4TcTHVfD\/woULcdlllznvTOrUqROdUiIyYOLFWl4OjB0bb2iF89AbcG7\/Ezl\/CZSXS28TJSrdjUUsPX7l5dLbREn7rLfRHjvBqaU5c+Y4peOdnR988IGzz5smSkpKwCUVV1xxBR599FHnlS6OZ4p+ZMDEA5aGy9ChgM3tgdYkhzbLy6FPLAIeugPOJG6sNOWXEAHpbUK4\/hdYuvs\/FhnYk956hO6v3tI44dqVU045xRlNue2225xy8b1\/9913n7N\/ww03gG+b5i3TjRo1cl6rw5fufvHFF+DLcjkrQeFDSJ0IPv\/IgIkHaGkpQKOluhrg4l3G4RAnjRne1UE\/uknCCPjcHQhLWbtxEigtBaib+aS3cR56csGku8nxSzJ2aSkgvfUA0X+9jfTYiX79+jlTQ7APF+vy5hU+xmKU1VkgEMAhhxwC3p3kPu6CW96hZMF9\/8qASRRpMAhYRTmGDI2ZXr22pmAVp9GZrSi2\/vrbG9iapn49EwgGAeltnPiku3GCSn2wYBCQ3sbJufD0VgZMnKoRMVgwCPDkooT3cocNQ40xQ79QCHxC7O9+97uaZDjsNmDAAGe+sMYxr3b87w1kAZ78KEIwCFAvKdLbCHUq3Y0AJfNOwSBAnaVIbyPUR+HprQyYCGrg2SkYBHhy8U4ljs4wId65ZFNNf\/r+e7z33nt4\/\/33MXfuXIy1ed57770Xqbq9jFlnVgqvN5BZ3knkHgwC0tswgNLdMBjZuxsMAtLbsPopPL2VARNW\/b7uBoMATy4zVHjrdfObbsIdd9yBq6++Gr\/\/\/e+dbSczbBLKM6cCF15vIKeqJ1phg0GgoPWWYKS7pJBTEgwC0lurMi+6a9Fy9CsDJh0VFwwCwSD4rqS2bduaPVOBc889Nx05ZzCPwusNZBB2arI2nS08vSVK6S4p5KxIb63qEtFhC56j30QMmBw9xOwpNm8l4z30e+21Fx5++OHsKVhKSuKlJ8A40QszZswY8KnHAwcOxCeffBI9oHx8JVBYekt01EMvwriRRbobmUsqXaW38epwKmshtWnLgEkt35rUly5d6kwd3X333fjzn\/+M+++\/37nVrCZA3u0k0gMIDxsZBG\/Lmz59Ol566SXn0dR8xkDkkHL1k0Dh6S3phetjIvuMu6NId3dkkmqX7NPbVB8x009EV8PDMm5uigyYNNUb3wkxZMgQHHzwwWjfvj2uu+46XHTRRVi7dm2aSpDubKosQy9i0SJ8y8rK0LdvX2fRM\/nxJWKLFi2KEFJOfhIoPL0lPS96yziMu6NId3dkkmoX6S31MV5JdW2kLn0ZMKlju13KTz31FC6\/\/PIat0GDBuHdd99FkyZNatzyayfcwk9kPzIFTr3FejNq5FhyTZZA4ektiSWir+FhGXdHyQbd3bFU+e0ivQ3Xy7r2c1cXZMDkbt1lbcmrqnZHp073ehLGzdoDU8HyngD1T7qb99WcdwdYqHorAybvVDnzB\/Ttt1OdZ93weTeJCuNGOoK63owaKY7csoVA7pSD+peozrrhGTfSkUp3I1GRm58EqHuuHia6ZVw\/y5LOtGTApJO28vJMoHfv3nj99dedJxcvWLAAlZWVzloizwkqogikiYB0N02glU3BEZABU3BVnpsHHO3NqPEcjcKIQCYJSHczSV955zMBGTD5XLt5dmyR3oyaZ4eow8lTAtLdPK1YHVZGCciASTl+ZSACIiACIiACIuA3ARkwfhNVeiIgAiIgAiIgAskTqCMFGTB1AJK3CIiACIiACIhA9hGQAZN9daISiYAIiIAIZJ6ASpDlBGTAZHkFqXgiIAIiIAIiIAI7EpABsyMTuYiACIhA5gmoBCIgAjEJyICJiUeeIiACIiACIiAC2UhABkw21orKJAKZJ6ASiIAIiEBWE5ABk9XVo8KJgAiIgAiIgAhEIiADJhIVuWWegEogAiIgAiIgAjEIyICJAUdeIiACIiACIiAC2UlABkzkepGrCIiACIiACIhAFhOQAZPFlaOiiYAIiIAIiEBuEUhfaWXApI+1chIBERABERABEfCJgAwYn0AqGREQAREQgcwTUAkKh4AMmMKpax2pCIiACIiACOQNARkweVOVOhAREIHME1AJREAE0kVABky6SCsfERABERABERAB3wjIgPENpRISgcwTUAlEQAREoFAIyIAplJrWcYqACIiACIhAHhGQAZNHlZn5Q1EJREAEREAERCA9BGTApIezchEBERABERABEfCRQF4ZMD5yUVIiIAIiIAIiIAJZTEAGTBZXjoomAiIgAiIgAmkgkJNZyIDJyWpToUVABERABESgsAnIgCns+tfRi4AIiEDmCagEIuCBgAwYD9AURQREQAREQAREILMEZMBklr9yFwERyDwBlUAERCAHCciAycFKU5FFQAREQAREoNAJyIApdA3Q8WeegEogAiIgAiKQMAEZMAkjUwQREAEREAEREIFME5ABk+kayHz+KoEIiIAIiIAI5BwBGTA5V2UqsAiIgAiIgAiIQOYNGNWBCIiACIiACIiACCRIQAZMgsAUXAREQAREQASygUChl0EGTKFrgI5fBERABERABHKQgAyYHKw0FVkEREAEMk9AJRCBzBKQAZNZ\/spdBERABERABETAAwEZMB6gKYoIiEDmCagEIiAChU1ABkxh17+OXgREQAREQARykoAMmJysNhU68wRUAhEQAREQgUwSkAGTSfrKWwREQAREQAREwBMBGTCesGU+kkogAiIgAiIgAoVMQAZMIde+jl0EREAEREAEcpSARwMmR49WxRYBERABERABEcgLAjJg8qIadRAiIAIiIAI5QUCF9I2ADBjfUCohERABERABERCBdBGQAZMu0spHBERABDJPQCUQgbwhIAMmb6pSByICIiACIiAChUNABkzh1LWOVAQyT0AlEAEREAGfCMiA8QmkkhEBERABERABEUgfARkw6WOtnDJPQCUQAREQARHIEwIyYPKkInUYIiACIiACIlBIBGTApLO2lZcIiIAIiIAIiIAvBGTA+IJRiYiACIiACIiACKSKQKR0ZcBEoiI3ERABERABERCBrCYgAyarq0eFEwEREAERyDwBlSAbCciAycZaUZlEQAREQAREQARiEpABExOPPEVABEQg8wRUAhEQgR0JyIDZkYlcREAEREAEREAEspyADJgsryAVTwQyT0AlEAEREIHsIyADJvvqRCUSAREQAREQARGog4AMmDoAyTvzBFQCERABERABEahNQAZMbSL6LwIiIAIiIAIikPUEZMDUWUUKIAIiIAIiIAIikG0EZMBkW42oPCIgAiIgAiKQDwRSfAwyYFIMWMmLgAiIgAiIgAj4T0AGjP9MlaIIiIAIiEDmCagEeU5ABkyeV7AOTwREQAREQATykYAMmHysVR2TCIhA5gmoBCIgAiklIAMmpXiVuAiIgAiIgAiIQCoIyIBJBVWlKQKZJ6ASiIAIiEBeE5ABk9fVq4MTAREQAREQgfwkIAMmP+s180elEoiACIiACIhACgnIgEkhXCUtAiIgAiIgAiKQGgL5asCkhpZSFQEREAEREAERyAoCMmCyohpUCBEQAREQARHIBgK5UwYZMLlTVyqpCIiACIiACIjANgIyYLaB0EYEREAERCDzBFQCEYiXgAyYeEkpnAiIgAiIgAiIQNYQkAGTNVWhgoiACGSegEogAiKQKwRkwORKTamcIiACIiACIiACNQRkwNSg0I4IZJ6ASiACIiACIhAfARkw8XFSKBEQAREQAREQgSwiIAMmiyoj80VRCURABERABEQgNwjIgMmNelIpRUAEREAEREAEwghklQETVi7tioAIiIAIiIAIiEBUAjJgoqKRhwiIgAiIgAjkBIGCLKQMmIKsdh20CIiACIiACOQ2ARkwuV1\/Kr0IiIAIZJ6ASiACGSAgAyYD0JWlCIiACIiACIhAcgRkwCTHT7FFQAQyT0AlEAERKEACMmAKsNJ1yCIgAiIgAiKQ6wRkwOR6Dar8mSegEoiACIiACKSdgAyYtCNXhiIgAiIgAiIgAskSkAGTLMHMx1cJREAEREAERKDgCMiAKbgq1wGLgAiIgAiIQO4TSN6AyX0GOgIREAEREAEREIEcIyADJscqTMUVAREQARHIDwI6iuQIyIBJjp9ii4AIiIAIiIAIZICADJgMQFeWIiACIpB5AiqBCOQ2ARkwuV3agFMNAAABGElEQVR\/Kr0IiIAIiIAIFCQBGTAFWe06aBHIPAGVQAREQASSISADJhl6iisCIiACIiACIpARAjJgMoJdmWaegEogAiIgAiKQywRkwORy7ansIiACIiACIlCgBGTAZKjila0IiIAIiIAIiIB3AjJgvLNTTBEQAREQAREQgfQSqMlNBkwNCu2IgAiIgAiIgAjkCgEZMLlSUyqnCIiACIhA5gmoBFlDQAZM1lSFCiICIiACIiACIhAvARkw8ZJSOBEQARHIPAGVQAREYBsBGTDbQGgjAiIgAiIgAiKQOwRkwOROXamkIpB5AiqBCIiACGQJARkwWVIRKoYIiIAIiIAIiED8BGTAxM9KITNPQCUQAREQAREQAYfA\/wMAAP\/\/IlkXdQAAAAZJREFUAwBXN1M6lgZ4RQAAAABJRU5ErkJggg==","height":337,"width":560}} +%--- +%[output:1ef43198] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,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","height":337,"width":560}} +%--- +%[output:0bd82ac8] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,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","height":337,"width":560}} +%--- +%[output:6f7fe7c2] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,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","height":337,"width":560}} +%--- +%[output:4155c762] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,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","height":337,"width":560}} +%--- +%[output:2cf257b8] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,iVBORw0KGgoAAAANSUhEUgAAAjAAAAFRCAYAAABqsZcNAAAQAElEQVR4AeydC3yU1Z3+nySA3EISC4jIJSCKWBAoyLrUrdgCCpWutR+FBcqlS6W7Fai0aAtYxn8tyJa6bXGRtrZgKbvQliJykdoiWC1QhKLWgtySeEFQCEkgBIIJ+Z\/nhDNOQu55Z+admYfP\/N5z3nP9ne8ZZp6c97zvJJfpnwiIgAiIgAiIgAjEGIFk6J8IiIAIiIAIiEA9Cah4tAlIwER7BtS\/CIiACIiACIhAvQlIwNQbmSqIgAiIQPQJyAMRSHQCEjCJ\/g7Q+EVABERABEQgBglIwMTgpMllEYg+AXkgAiIgAtElIAETXf7qXQREQAREQAREoAEEJGAaAE1Vok9AHoiACIiACCQ2AQmYxJ7\/hBz9yZMn0b17d\/Tv3z84\/muvvdamFRcXB9MiERk+fLjtd\/\/+\/Y3qjuOhNaqRBlaOFrsGuutZtfqO+8tf\/rKd661bt3rmgxoSgUQmIAHToNlXpVACq1atsh\/M\/AJ96qmnQrNs\/L777rP5zz33nD2vfFi0aJHNZ31nffv2xZgxY1BdncptNPZ82rRpoKWkpNSpqXPnzqFnz5749a9\/XafyDS107733XsbGMWJ46NChhjZdaz32TSa1FWQZWl3ZVW6P\/XAsoTZ48GBMnToVe\/fuRX3+vfzyy5aX11yqapdjpjV03PUZl8qKgAhcTkAC5nImSqkngWeffdbWaNWqFX7\/+9\/beEMObdu2xaBBg3DzzTfj\/Pnz2LVrF77xjW9g06ZNDWmuXnXYz4MPPogmTZrUqd6LL76Ijz76qE5lvSjEL3eyqWwtW7b0ovnL2jh16hR27959WXpVCfVlV1UbTHNjpHg9duwY\/vjHP2LcuHF49913mV0nY506Faxnoara9Wrc9XRFxUUgtgl46L0EjIcwE7EpfrlQaNx6660YMmQIjhw5gjfffLNBKIYOHQqu5qxevRp\/+9vfwDbZ0PPPP88Ao0ePtn9dUzB98YtfxPjx4216Xl4evvWtb4F\/td9444223Ouvv27zeOBf46x7ww03gJdsduzYweQKVvlyQGlpKbia9LnPfQ6s96\/\/+q\/43e9+Z+t897vftas1PGH8tttuYxSs86Mf\/Qisw9WZkSNHBuuwwOnTpzFjxgzcdNNN+PSnP23br6tgeuCBBywb8gm1a665hk1fZrX5wvzqxrdx40YMHDgQSUlJYJzC4ujRo5Yr45X514fdZY6GJLgxrl27Fn\/5y19w5ZVXWiH75z\/\/2ZYqKyvDypUrcdddd9nVL843V0AodliAc+tWxO644w489NBDTEZt7w++Nzguvme+973vWQHNufvTn\/5k61fXbuVxc7Xoa1\/7mn0fUoT953\/+J7Kzs20bOoiACHhPQALGe6YJ1aJbcRk1apQVBxz8mjVrGDTKuJrDLyE2kpubywBuqf6HP\/whOnXqhFtuuQX8d\/\/999uVHwqg2bNn2y8N\/uXOvS4XL160lyJeffVV8AvnU5\/6FJ544glWq9EoRH7wgx\/Yv\/4plgoLC+0X4rJly\/CZz3wGn\/zkJ23922+\/HRMnTrRx1vnJT34CiooFCxZYf\/kl+tJLL9n8xx57DOvXrwdXTShyfvOb3+D999+3eV4favOF+dWN77rrrsOdd95pXerRo4cVa6mpqXY8TKzMn2mhVlPboeVqinfs2BH\/\/M\/\/bItwHhn5+c9\/jkceeQT79u2z77XOnTtbgfWlL30JvKRHQZuWlsaiVtzy\/cCTmt4fzHfvq5kzZ+LDDz8E+37rrbfAFTnOe3Xtsq4zCsKvfvWreOGFF\/Dv\/\/7vVuxt3rw5+N5w5RRGnYAciCMCEjBxNJnRGApXJZKSkuyqA78wmjZtar+k+YHeGH8oOPjXNtvgigXD5OTytyu\/2BYvXgz+xc6\/mvfs2QP+xfv\/\/t\/\/s19c06dPR1FRkd0\/w7ycnBy0b9\/e+vX444+D5dhedca\/9H\/2s5\/Z7CVLloBi5Mknn7SXt7iaw3H26dPH5lPAfOUrX7HxX\/ziFzb86U9\/invuuScolP73f\/8XFFJcuWABntMHihleKmNabcYvV64ShJoTTlXVrcmX2sZ3\/fXXW4HAdrmSxC\/yNm3aoCr+LBNqtbUdWra6+IULF+xlw23bttkijvX\/\/M\/\/2PP\/+q\/\/AuefK3UUk8ePH8e6deswYcIEu2rDQtwwy5WT2t4fLOvGxVUXtssVIAros2fP4sCBA1W2y3qhxtU1iuZvf\/vbVsBQSLdr1w7vvfde2ERqaP+Ki0AiEij\/RkjEkWvMjSZAkcFLC9yzwuX+Fi1agJeRuH+iIXda8NKI+4Lmsj7\/CuZqBjdzhjrrLtkwjZesGPKLytXlZR2mvf3223YFhXHmuS8qXhJiWnXGL0S3v4WXeliuV69e9hLO\/PnzeXqZ8TKGEyNcnWF\/bgWJl9k++OADlJSUgF\/wzGMDrVu3tqs1jNdmrFN5\/wvFRVX1avOlIeML7SeUf2g6441p24k0zg\/FKQUExeJnP\/tZsN0zZ85YfsOGDWNX1nr37m1DCksbqXSo7f0RWpzzxnO+T\/r168eovfxkI7UcMjIy7MoNhRRXsDhfJ06csLU4DhvhQSYCIuAZAQkYz1AmXkPuUhH3wPADm+Y2O7rVhvpQcZt4+UXNv4YDgYC9RMAvh9B2KJTcOb9sGOcqDfc\/hBr\/Ik5KSmI2+BeyjZgDhYQJPH0lJZX3w8sRoT4wzn0VFC7sMCkpyV7uYJzm0hmvyfiFToEXavwrv6o6SUk1+1JVnfqkhfKvT73ayvL9w7mn\/du\/\/ZtdZVm6dGmFatwzxNUgl8j9LYy79wHjoebSq3t\/hJZt1qxZ8DQ0HkysIcJLR9\/5znfsnp1nnnnG3p3G93MNVZQlAiLQSAISMI0EmKjVuULhNtdy4yI3Uzrjl\/gf\/vAHFBQU1AsP\/9p2X9C8ZMNLAqFfVlU11q1bN5vMv3Z5aYkbO7t27WrTuMJx9dVX2zgvBTjh8tprr9m06g4dOnSA+wJzG5K5GXPMmDF2P0hoPXep7KqrrrJ1eM4+6Qe\/NFmWfjCfXHjO1SKG\/PLlKhHjXhr7ov9V+AL6Up\/x8dJXfXyrT9uV2w0Vad\/\/\/vfx+c9\/PnjZiu1yHw7H5PixPjfOMuReHYbOnN+1vT9c+bqGrt3K5bnpnGlcnXJz7\/Zu1VWksr5MBESg7gQkYOrOSiVDCHCDIpf0ebcKN6pyn4QzLvHzi2bDhg0hNcIT5d6XAQMGgJdNeIcP75qh8OFmT35Z0z9uyqQ\/XNXhM2fc3prqPEpKSrIbf5n\/9a9\/HVxBoTjjShP3hzCdbTPkPgzum0hKSgI3cTLtG9\/4hl05Ig9u6KTYo3hh\/8xnW9wDQ26VV5eYX5VR0FFAVbaqnpOTlFSzL0lJtY\/PCUc+\/4Sbdsm3Kr8qpyUl1d525Tp1PedcsCxXOjiP3DjOy4zczHv33XczC85v5nODeW3vD1upDofK7Vau4oTS9u3bQXv44YeRmZlpi23ZsqXeYt5W1EEERKBGAhIwNeJRZnUE3CUi3l5cuQy\/WJjmyjBuLUwHbprlFxgFBv9yp4DipRt+6fCSAy9DcCMo91G888474CZQulLTU3e5EZi3ZvOv\/l\/96lfgSgm\/QLlKwLoUErzUwSfoctMn0yigaBR2vLzDulxRcvs0eEmMq0y8Y4Z7Nrhy1aVLF1YFN67aSDWHrKws+1wcjjHUOKaqqtAPWnW+1DY+7v3h7cr0ixuCeTdOVf1UlVZb21XVqUsaBSL3IFEM8o4krmBx3n\/729\/iiiuusE1wjigmKBrcc2xqen\/YSnU4VNVuaLUvfOEL9vb4gwcP2lv6+X7j3W4UvNwIzn1QoeUVFwERaDwBCZjGM0zIFvilxi9V7jOpDGDEiBFgHr9YmMfbhXnOD3meVzYKBebzy6lyXug5RQnLcVNnaDo3EPPLYufOnfavX94Jwss4rgwFBDdXvvHGG+BtzlyRYTu8HdeV4WZPprkvQn5J8jke\/CKk2Pj73\/+Ob37zm\/bZKKzD\/RoUJ7y0xIfaMY1iiaKB57zMQZ+4AsA8GldbeHcTL0utWLECXDniKgH7DS3Hss7IkPnVGW8RZlmXzzitNl9qGx85kBVXOMiJG1Or419fdvQv1NwYKUZC0yvHk5KS7NOZubLHS4KvvPKKvdOLd5i5shSInDPycO+n2t4fVY3Lvb\/ZHttmWLnd0HFzXxAFO\/vlCgzFKeeUK5V87\/A9yDlnPu9cY5syERCBxhFIJAHTOFKqLQIiIAIiIAIi4BsCEjC+mQo5IgIiIAIiIAJ+JOBPnyRg\/Dkv8koEREAEREAERKAGAhIwNcBRlgiIgAiIQPQJyAMRqIqABExVVJQmAiIgAiIgAiLgawISML6eHjknAiIQfQLyQARqJzC4Y0fw7kSvjO3V3mtil5CASez51+hFQAREQAQ8IHC8eXN8JTvbM2N7HrgV101IwMT19Gpw8UBAYxABEYgNAk2Mm16ZaUqvWghIwNQCSNkiIAIiIAIiUBcCXokXtlOX\/hK9jARMor8Dah2\/CoiACIiACNSFAIWHV1aX\/hK9jARMor8DNH4REAEREAFPCHglXtiOJw7FeSO+FzBxzl\/DEwEREAERiBMCFB5eWZwgCeswJGDCileNi4AIiIAIiEBUCMR9pxIwcT\/FGqAIiIAIiEAkCHi1+sJ2IuFvrPchARPrMyj\/RUAERMCPBBLQJwoPrywB8dV7yBIw9UamCiIgAiIgAiJwOQGvxAvbubx1pVQmIAFTmYjORUAE4oGAxiACESdA4eGVRdz5GOxQAiYGJ00ui4AIiIAI+I+AV+KF7dQ0utLSUjz++OPo1q0bTp48WaHoypUrMXjwYPTt2xfz588P5hUUFGDMmDHo1asXRo4ciSNHjgTz9u7di2HDhtm8KVOmoLi4OJhXXXvBAlGMSMBEEb66jmMCGpoIiEDCEaDw8MpqgjdjxgxcffXVSE6u+BV+8OBBLF26FEuWLMHGjRuxZ88ebN682TY1d+5c9O\/fH9u3b8fEiRMxbdo0m04xxPj06dNtXocOHbBw4UKbV1N7tkCUDxVHH2Vn1L0IiIAIiIAIxCoBr8QL26mJwcyZM60IqVxm\/fr1GDt2LPr164dOnTph6tSp2LBhA86dO4edO3eC9TIyMjB69GgrfihQduzYgR49emDUqFFg3qxZs8B2ysrKbFhVe5X7jdZ5crQ6Vr9hJaDGRUAEREAEIkyAwsMrq8n17t27V5lNQULh4jI7d+6M\/fv3IycnB+np6WjatKnLQpcuXbBv3z5UrpOWloaSkhKcOHHisjzXXrCRKEckYKI8AepeBERABEQgPgh4JV7YTkOI5Ofno0WLFsGqrVu3Rm5uLvLy8tCqVatgOiMsx3Ray5YtmRQ0lmW96toLFoxyJDwCJsqDUvciIAIiIAIiEGkCFB6NtY1mpeT+bt0a5HqbNm1QVFQUrFtYWAiuqKSmpuLMmTPBdEZC886ePcukoLEs26JV1V6wYJQjeAY3zwAAEABJREFUEjBRngB1LwIiIAIiEB8EGiteWP9LZhXlV9nZDQLCy0eHDh0K1j1w4AC6du1q98McPXrU7oVxmYcPH7Z5vCzEuEs\/fvy4vQupffv2tl5V7bmy0Q4lYKI9A+pfBERABEQgLghQgHhlDQHC26Q3bdpkhQo34a5evRrjx4+3m3OHDx+OtWvX2ma5oZeXkAYMGIChQ4fi2LFjdr8LM1etWoVx48bZ\/TLVtcdyfjAJGD\/MgnwQAREQAV8QkBONIeCVeGE71fnB575wEy+NImXQoEFg\/MMPP0TPnj2xYMECK1ooTCZMmAAKF7a1aNEiu5l34MCBWLduHVasWMFkK1QoWhYvXgy2lZKSgjlz5ti8mtqzBaJ8kICJ8gSoexEQAREQgfggQOHhlVVHpG3btsjKyrrMeMmHdShC1qxZgy1btmDEiBFMstasWTPMnj0bu3fvtiKHt0zbDHPo2LEjKGB27dplnw8T+nyZ6toz1aL+koCJ+hTIAREQAUdAoQiIgAjUlYAETF1JqZwIiIAIiIAI1EDAq9UXtlNDN8q6REAC5hIIBSIAiIEIiIAINJwAhYdX1nAvEqemBEzizLVGKgIiIAIiEEYCXokXthNGN+OmaQkYH02lXBEBERABEYhdAhQeXlnsUoic5xIwkWOtnkRABERABOKYgFfihe3EMSbPhhYiYDxrUw2JgAiIgAiIQMIRoPDwyhIOXgMGLAHTAGiqIgIiEF4COTnn8LF9\/Nsu4e1VrYtAAwlcquaVeGE7l5pUUAMBCZga4ChLBEQgfAQoUJYvP4rJk\/+O22\/fZa1bt5eQlLQB3bptCbEXbRrTacy7\/fYdpt5rCAQOWsvJkcgJ30yp5boSoPDwyuraZyKXk4BJ5NnX2EUgggRyzKpKIHDYCpWkpM1GoLxkRMjfsXz5O9i27SRey3kf6UNycPeyHIzfehhzyl7HN7P34tvZrxrbha9ufRNfWnYE103cj23bTpl6R\/Hoowetdev2omlvi2l7ByhqcnISStBEcBbVVU0EvBIvbKemfpRXTkACppyDjiIgAmEgkBMiWri68uijh434yLvUU5IJk9EmswkmGMHyoBErX1m2B2MmvYLJQ57DF\/AcxmX+HvdlrseXMp\/HxCEbMHXSRkwI7MTiss14KHsPug05b9rgx30ycnKKTdunKggaihlTQC8RiAgBvhO9sog4HOOdSMDE+ATKfRGIOoEqHCgXLllmVeQlIygoWvJNKQoWfrw3M\/HmSM1Mwu3z3sPC7Gdx15CXcTtexCisxx34Az770VYMemcX+v\/jNWN7MfDobgw+9xcMwG7cjFfRD3txS+ZefGvrVtyz7AgAtpliQmcUNOdM3wfBy04SMgaNXmEnwHe3VxZ2Z+OgAwmYOJhEDUEE\/EKgonA5aNziRww\/0pua+BXGXLwJBkw8jrGBHeiMd3A9DqE33kTmOzlIX2bEzn8CH90GFPcz1h8oGwYkzQSuXJOHnqcP4jocxjU4io54H8MmvYm7lx0CQPHi2nf98jw5RMgcNuX0EoHwEOC7zSsLj4fx1Sr\/l8fXiDSaRCOg8fqEAMULN+NyX8rHYoIrI\/xITzJeMiy3NpmluCvwBprjPNJwGu3wITLeMcJlrSn2Q2D708CfcoCtJcCLHwG79gNvLzV5vzb2InBVwQe4ytRJRz5aoxCfnpSFiv\/K+ylP48dc+TkvYfFS1rZtp8qzdBQBLwmUv80AL0Iv\/YrTtpLjdFwalgiIQAQJBALll4tycopNr\/z0doLFnFb4NOdKTBOkZZagDElmzaTEWrI5wxkA54wZbfGeCbKNHTX2rjGu5bzeLhMweSg1CcZYvwk+QgpKrKVlFpmMi8ZCX\/SF9nGaE1rbtuV9nKiYCMQQgeeeew4jR45E7969cf\/99+P06dNB71euXInBgwejb9++mD9\/fjC9oKAAY8aMQa9evWzdI0eOBPP27t2LYcOG2bwpU6aguJj\/j4PZvo1IwDR2alRfBBKcwPLl79tLNDAyAnAfKU40MAy1FEur1JQ9i9ag5eITKEAblHzS5I0w2fcDo24CepjotcauM\/ZJ08QX\/sUsyXzZnPQC8q\/k2ku5FaEVLqCZ0TUlJvOCMYYmqPAyDVQ4B26\/fQ8kYipB0WnjCPBt5pVV48k777yDhx9+GNOnT8cLL7yA1q1b4\/HHH7elDx48iKVLl2LJkiXYuHEj9uzZg82bN9u8uXPnon\/\/\/ti+fTsmTpyIadOm2fTS0lIbZ3vM69ChAxYuXGjz\/H5wnzZ+91P+iYAI+JAAVzMmT37DeGbEB5JMyBc\/wUPDynGA6y\/njeg4ibbgxaBDuB5vog9O9m2Liw8noYW5lPTp\/wNuMn9A3mQ+mz\/1kmnjSeD8lCtw9MZrcBjX4j10Qi7a4hSuNALmClxEsSlEo4ChmVP7CvWLftLKfZw8+R\/gGGwxHUSgsQT4tvLKqvFl165dGDRoEO6880507NgRDzzwADZt2mRLr1+\/HmPHjkW\/fv3QqVMnTJ06FRs2bMC5c+ewc+dOzJw5ExkZGRg9ejSSk5NBwbNjxw706NEDo0aNsnmzZs0C2ykrK7NthvPQ2LaTG9uA6ouACCQugcmT\/17N4PkpziyGTkDwnJd4PkJeTga4cnISn8B76AxzAQpvojd24BbsbPHPeL17Xxwb0wNF3+mEwoe7Yt\/gXnj16oHYjsF4FTebsn1wCD3wNrriDFKtFeW0NB0UGTtvjAKGZqKgYGFIoy80xoGcnPNm9ejt8hMdRaCxBPh298qq8YUrJqFZqamp9hISL\/tQkFC4uPzOnTtj\/\/795n2eg\/T0dDRt2tRloUuXLti3b58VMaF10tLSUFJSghMnTgTL+jUiAePXmZFfIuBzAly52BbcDFtai7cUEzSWa46PProSH6I9zqMFCtEaJ9HOCJLrQBFDgUJ7BbdiCz6Hrbgd2\/FpI24GYzduxuvoi4O4HqeRho\/QFMW4AmwLoIBpbvzgRppQEVNi0vj6WLjwrNxSsHz58fKojglGIAzD9Uq8sJ1q3OPqy1\/\/+lccOHAAFy5cwI9+9CMkJSWhqKgI+fn5aNGiRbAmLy\/l5uYiLy8PrVq1CqYzwnJMp7Vsyf87TC03lmW98jP\/HiVg\/Ds38kwEfE1g+fKjdfSv8lK0WRE5ehX+jt7gCgoFTCFaocgIkCIT5iMDR9HJrMp0xwFcj0O4Dix3Au2Qh3SwbAlSkI80FBh715R98b3bAXDXTFMTlhq7YOy8MYoXmomiKgFTnrZ8+QcsIBOBxhGg8Gik\/bh5OrqndqvWj27duiEQCICbbe+++27cfPPNdmWFKyxt2rSxQsZVLiwsBFdUuEpz5swZl2zD0LyzZ8\/aNHdgWbblzv0aSsD4dWbklwjENAEnGioPguKCaVfgH9t74yCuw1u4Ae+iC86iJT4WMy1wDrSWKDIhRQsFSyFSUYzmyEc6uGpzBNeCKzLvvtQFQIax1saaG+OlKvpAM6egiKIx7ozfNC4e+VA9xiEBvqUaaTOQj6yPeA8eqv3Hu4lefvllu\/elZ8+eoCUlJdl9L4cOHQrW4ypN165dbfrRo0ftXhiXefjwYTCPl5kYd+nHjx+3dyG1b9\/eJfk2lIDx7dTIMRHwOwEnEpyfTiy4kOmMs1ypOWGcdt7EC4DlwMFdPfEGbsJbuAEHcT0Oowdy0BXvoAveQyfk4kqcwifAFZmTaIv3cA2y0A2H0QNvoA9exr9g3+ufBOweRi6d8yONfbBPhs7Yf2UBg0v\/+I1zKapABBpDgG8lr6waP7g35ZZbbgFvfWaRxYsX47777mPU3ibNDb3ctMtNuKtXr8b48ePBjbvDhw\/H2rVrbTlu6OUlpAEDBmDo0KE4duyY3QvDzFWrVmHcuHF2VYfnfjb+b\/ezf\/JNBEQgLAQa32hmJgUD26FIcOKAcabRXJxiwuUz5OWd94FlJv+XwKnfXYk3c3qDe17+jj5WnBzCdTiEHtiLT+E19MPb6IJ9+CTewE34M27D1gufxRuH+uLM\/6YCPwGMkjGHY8bYNvsL\/WhjnzSTXc3r47FUU0DJIlAXAl6JF7ZTTX\/t2rXDgw8+iNmzZ9tnvVx11VVWcLA4V2IWLFhgRQuFyYQJE0DhwrxFixYhJycHAwcOxLp167BixQomW6FC0UIhxP01KSkpmDNnjs3z+yH0f7nffZV\/IiACPiIwaVJnfPzFzxUO51yJidBMABcyTksyh4+MnQBK3rarMDAihlb8q+Z4++WueO3N\/tjz3gDsOTEAb57pjTdO34Rdx\/8Jbxy5CVm7rsWZTanAUtPEj4ytMvZHY+\/ycXf8mQCzsgP2QRHDb4EUk8kX0xg6c+nJGDIkzVgbl6FQBBpOgG85r6wGL3gb9PPPP4\/XX38d8+bNq1CSImTNmjXYsmULRowYEcxr1qyZFT27d+8GRQ5XZVwmb8emgOEt2nw+DG+xdnl+DiVg\/Dw7ceybhhYfBJYt6xcyEIoVGpMYVmXMo9gpMpH3gOK9wPPmkhL\/GPw\/k\/S\/xlYbM5eX8FMjOri68t\/m\/MfGnjDGkGkULs+Z8\/W5wLu7TMS0AxPHBRPnakvoTxiYdkxqxZdLSzZfAF0rZulMBBpKwCvxwnYa6kMC1ZOASaDJ1lBFwGsCQ4Z8ApMmdarULIULkyqHFBYUL8yjgKCI+cCc7AFyXgM2HwVXYvArk0QRw8v1a0z8d5fsNyb8rbE\/mDo7TJ33uPSy1SRw5eWUCYuNcWWFm3iduR+QDP1GYBlTFE2MeMnU6gv0zzMCfJt5ZZ45Fb8NJaiAid8J1chEINIEuArzySEUDKE9h4oXxmkULxQxLMeQIoZpeSbhfWNvABf+BLxjRMmbfwH+9jKw9xXgTXN+eDNw8lngo9+bctuMHTDGFRdejmI7vPuIz7ng8yxo9IffJLytmqEpDgoXGss3wZ2T0hEIVBZfLCcTgQYS4FvNK2ugC4lUTQImkWZbYxWBMBD4B06g\/9bj6Dev0LROoUIzUbv\/pXKcgsWlsQzFBMUM07iCwjYoaLgh16y0gOGHpuAZYxQr3NtCEcJLRFxd4UbiVJPHkAKGwoXmvkVMln3xnH0loVUmcMM8c9lq2Xs4DhPafB1EwAMCTUwbXplpypcvHzklAeOjyZArIhCLBH6Pv1u3BwYK8cWtJ9Amk0KDgoTChFmM0xh3aYzT+BHkzH3yU5y4uBMmDGkUJ1xtobmVFq6yUMy4Omw31JjO8ybGt1LctfVt3BQ4ZS4glWAFjjBDJgLeEOBbzSvzxqO4boWfHHE9QA1OBEQgvARO4OOneF495CN8NfsdfGYeV01KTcdOuJhocEUmNI3pKebAT30XMk6hQmO8OnPlGZomKrwq1snIvIDb52XhP7NfRUZmEZrhghUwJ0N8r1BdJ34lIL9EIEhAAiaIQhEREIHGELiIJFxEMkqRgsFmheN7ZXtwx7wcfCKTl38qt1xiEiobL\/GY5Dq\/yqooyY80ihdmNTJ74RwAABAASURBVMGVmcW4c95bmJP9AoYF9lvR0hzFNmxiBFWK8ZYlZSLgCQG+9bwyTxyK70b4vz2+R6jRiYAIhJUAJQs7uGiES6mxkqA8aIohgQ8QyP4LHs\/+I+6etw83DvnAFC0zVtWrxCQ6M9FqXyzDzFDBw3gTtMs8i+uH5GKEES3fzX4ej2RvxvDAPrDHMiOuWItG8UJjvF6mwiJQEwGvxAvbqakf5VkCEjAWgw4iIAINJfBJtDNrGeWfuGVmFabkkoApNhdqitEc59ECaZkX8PnAIcza+gp+lL0eP8l+FvfOew19hryPqzP58LnKvZeZBJoJKrz4kVXeF5M7ZBZi2KT9eGDZVjy89QV8P\/s5TNu6DSMD\/7CXikqMoLpohMvFSyHrhNqnkBp6qrgINI4A35peWeM8SYja\/DRIiIFqkCIQBwR8OYTeaGvkQYkVMU68fIQmRrg0hwvPo3kwPy3zIyMuinGPERnzjOj4afZqbCpbgpXZv8Rvs5di6dZfYd6ytXhs2W+Mrcajy9Zg7rLnMHvZRjyx9f\/w+NbfYm3Z\/+D3xv7b1P3ysl24adIH6DokD+fxcT8OVpkRVaVGxJQan5x\/zEsxHtF3xmUi4AkBr8QL2\/HEofhuRAImvudXoxOBsBP4Z\/TETfiEkQO8KFNuxZeERDGuAK0QrXEeH4sLCgmen0MLnEUray0yy1CW2QJth5RiwKTjRpScRJ9Jp4zl2nh\/k3b1kPO4ZkgR8pGOfGTgjFlBKUA62F+pESi8kIRL\/0qQYmJJJu8KXDB9nze+sM8SU64ILYzPV+IGdDBl9BIBjwhQeHhlHrkUz81IwMTz7Ho9NrUnAtUQ+HfcauRCKSgOnFEs0CgWio14KERrOGM6yzGkMd0JkXNoiQKkIR\/pNmScIue0ESuFpo1zcCs7V4AiJRkX4f6VWi9STHoTY01xHryERbsCLs42OprLW1\/BZ1w1hSLgDQGvxAvb8cajuG5FAiaup1eDE4HIEPiEEReLcRcuIgklaIILaGpDioYLl8QLRchZlK\/EMJ1CgiGNdWjnQLFRbsUmTuHDdNpFI044mlLTPu2ivSyUYvpMRqnJO2\/KnzMrK7Szph\/2R\/HDfvKRDoYFaGNiaZiN4WhrykD\/RMBLAhQeXpmXfsVpW7EkYOJ0CjQsEYgPAhmn1mLpG\/+BmwrewHkjJM6jXIhQSLh44aXLRRQTLo2hO6do4TmFi7vkw3May9AYZ5vn0BKFSAXFST4yUGBWbQqMQCkw4UlciZNoa\/PyUJ6Xi0+gr\/Ht\/+2biIwi\/vgj9E8EvCXQxDTnlZmm9KqZgARMzXyUKwIiUFcCp54xwiAP3\/n7AkzO+iUuIgkXzEpMuRhpjkK0BsUHwwK0AUUIV0p4Xtm4t6XQiJOzaGnLFZq6FD8F4GWldOQjHRQkzk4acULBchLtcBIULhlgGVoe0pFadAZfOfwLPLx3ITLO5gHGV+ifCHhNwCvxwnZq8O2VV17BF7\/4RYwcORIPPPAATpw4ESy9cuVKDB48GH379sX8+fMvpQMFBQUYM2YMevXqZesdOXIkmLd3714MGzbM5k2ZMgXFxcXBPD9HJGD8PDvyTQRihcCFHKCQP7JY7vBdWRuxdMt\/4MtvrTAJZThrVl6ceCkXI27lJA0FZsWE+QwrWz4yrCApFyptTZwrK5\/ACSNSaBQrzvJM2Xw4cXMl2FZ6UT4mv7UcP31xKu46vBEoAVBs7ORyc9BLBDwmQOHhlVXjGsUKRcsTTzyBTZs2oVu3bpg9e7YtffDgQSxduhRLlizBxo0bsWfPHmzevNnmzZ07F\/3798f27dsxceJETJs2zaaXlpba+PTp021ehw4dsHDhQpvn94MEjN9nSP6JQIwSyCjOw11vbcST66dh0hvL0Of433EBzVCEligXMeWberna4oRH5ZDihGkMTyED+cZOoq0J01GAdOQjHa6tArOqU2DEUNrpAgw6\/ipm\/3U+fr7pq7j70LO47B+FzIdGdF2WoYRwEEiYNr0SL2ynGmjZ2dmgyOhmhAuL3HHHHXj99dcZxfr16zF27Fj069cPnTp1wtSpU7FhwwacO3cOO3fuxMyZM5GRkYHRo0cjOTkZFDw7duxAjx49MGrUKJs3a9Ys205ZWZlt08+HZD87J99EQARihID7rHMfvAwvfbpQyIzc9zwe2vxfeHLVA\/jyq7\/C6O2rkIyLRsy0MEKkfBWGYuQMWoPGOO00UsGQaaeNODljzk+jjSmTCuadR3OzoHIFrjr1Af5173N4ZOP3sPxXkxB4IYBbs18B6EfovdUw\/0qMfWQszZheIuAlAb7fvLJq\/OrTpw\/y8vLw7rvv2hJcURk0aJCNU5BQuNgTc+jcuTP279+PnJwcpKeno2lT\/vCpyTCvLl26YN++fVbEhNZJS0tDSUlJhctSprgvX8m+9EpOiYAIxBaBKzKB0ksu8wOcooGhM2aZeMb5PNz5+h9w196N+MWif8eTi6fhsZ\/PxTd+\/SOMf\/bX+NxLL6LPP95E731v4pP\/+AfafXgSV334AToY67fvNQz4xx6Memk9\/v3ZX+CRXz2GH\/3PN7DpOyPx9OIp+NqWpfiX7JdhNBDQzHTYxFiKMRhj3AT2xU89+kuzCTqIgEcE+D5rpP34cDq6b+xWrUMtWrQAV0k+85nPoHv37nj66afx0EMP2fL5+flgvj0xh9atWyM3N9cKnlatWpmUj18sRyFEa9my5ccZJsayrGeivn7xv7KvHZRzIiACMUKg9RCAnyhJxl8Kh5o+yK8wZVoDGSV5uDb3CAYd2YXP79qIyc8uw8NLF2LO4u9j7uLH8OScB8pt3gOY99SjeOTp7+FrG5binj1r7QpLz7wDwCdMW7R0E5o2zaIM7MoL+6cvDE1WMC3ZnLQ2vppALxHwG4EZN+Qj61+zq3Ur21xCeuqpp+zeljfeeMPuf\/nKV75iV03atGmDoqKiYN3CwkJwRSU1NRVnzpwJpjMSmnf27FkmBY1l2VYwwacR\/lf2qWtySwTig0DCjKLjPICXkkLFC4UKBURzQ4GhM5fewqS3NsbLOW1MSCFylQk7GGtvrNMlu9qETHfGcrS2Jj3DGOuzHRr7orm+TLYVLwxpzcxqUZd5jMlEwFsC7j3nRViNZ1u3bsUNN9yAnj17gissd999N7KysvDBBx\/YfS+HDh0K1jxw4AC6du1q048ePWr3wrjMw4cP2zxeZmLcpR8\/ftzehdS+Pf8DulR\/hhIw\/pwXeSUCsUcg1axqdFkG8DK7+wAPjTtRwZDpDCk4nGWYIdMoZLiaws9PihTalSaPIfOZF2qsTyHE9tgu+zbFg6KF505UtTbipbMRL81NyDIyEfCSAN9rXlk1ft14443gbdS8G4lFGKfYuPrqq+1t0rwziZt2uQl39erVGD9+vN2cO3z4cKxdu5ZV7IZeXkIaMGAAhg4dimPHjtm9MMxctWoVxo0bV2G\/DNP9aMl+dEo+eUlAbYlABAm0mwR0rCRimpv+3Yd6aDw0jem8RO\/EC1dUUk291sacWGGeS2dZ1mEbbjWHIoVmqljxwjjzaRQ2aUa0fGIicLXxkWVkIuA1Ab7XvLJqfLvlllvsHUMMufrCPTDPPPOMvauIqzILFiywooXCZMKECaBwYVOLFi2ym3kHDhyIdevWYcWKFUy2QoWiZfHixeBm4JSUFMyZM8fm+f0gAeP3GZJ\/IhBrBN4yDvOzkftPKBz4gU6xUZ2F5rN8M1OfZSlM3MqKO2dZChOWY7wqY1mmswzjbKPQtLkkBzhvVolMVC8RCAsBvu+8shocfOyxx3DkyBE8++yzWL58ub2c5IpThKxZswZbtmzBiBEjXDKaNWtm98vs3r0bFDm8ndplduzYERQwu3btss+E4S3WLs\/PYdgFjJ8HL99EQAQ8JnAyB\/jto8Dbpt2fGNtizAkShqEf7hQXPHehi\/PcGUUM053xnHk8Z0hzaRQqbmWmtemXKzjnTLjV2E+N5Rv7nfHNBHqJQFgIUFx7ZWFxML4alYCJr\/nUaEQgugROGAFDoxenzeFlY\/9ljA8D5acN79ak6GiIUZxQpLAuBYoLGXfG9ilcuOLyvOl3ubG\/GnOvfdsAiix3rlAEvCRAYe2VAV56Fpdt8SMlLgemQYmACESBwH4jECp3W2ASXjL2HWO8tLTBhO7yDgUHhQjFCcOqzIkTihcXdyHFCvfG8DzPtMvunzLhI8bY5ykTVn5V5WPlMjoXgYYQ8Eq8sJ2G9J9gdSRgEmzCNVwRiCgBLqfzw9jZa6Z3ioyvmfAbxn5t7FfGmE5RQ5FCoyCpyphHyzV1\/mbsWWPfNTbN2FeM\/dwYV1y4\/4a3dJtT+2L\/NmIOoenmVK9qCCi5\/gT4PvPK6t97wtWQgEm4KdeARSCMBNpmwt4B5D7EKSRCu2M6hQpXWvjsrD+azE3GAsaGG7sVKPuiCe829vVLNsWERvCUTTXh540NNPYFY9809iNjfzF22BjbZNvsg8KJfTM0WfbFdIoX+mgTdBABjwnwPeaVeexaPDaXHI+D0phEQASiROAzk4CLIX1TQFT1gR668ZYrKrwMREsHkrjxtgDAmwD4G3X\/MKGxpH0mpOgxZcCyXKFhyMtIjNNCL0WxXydoTFX7amcE1o26E8my0MF7AnzPeWXeexd3LUrAxN2UakAiEGUC98wD3Ic4V0Hojjt3IVdLKDi4B4YhjWKEYZqpQJGSYcK2l4znocaH2jGPD7hz6axLMcM22T7NVLcv9svVly8Z32yCDiIQBgJ8n3llYXAv3ppMjrcBaTwi4AkBNdJwAl8y14OuN6scoR\/kbC303K2McPWFwoNGAUMxQhHCkAKGxjiNgoXGeGg66zqjaGGboX0xzv6HmNUhGuMyEQgHAb7XvLJw+BdnbUrAxNmEajgi4AsCj2wFKGKqu4TkPuT5bBiKDq6aUIQ4EeNWYaoSK0yjyGF5ihWGbIPGdp04YpzGVaDbjHj52jJfoJETcUyA7zevLI4xeTU0CRivSHrbjloTgdgncMNE4IgZBj9l+KFuosFLSzynOQHjQidknEChOOG+FoYUKC50e2gYsp2qjP1dMIdsY32NLybQSwTCSqCq92FD08LqaHw0zo+W+BiJRiECIuAfAu\/lADMnAybAnwG8a6ymD3KKk9ryayvj6puuUGwOvDOJt1q\/Z+L0xQR6iYAIxA+BqgVM\/IxPIxEBEYgGgd8sB0ovdXzehP8w9idjvLPIBHYlpvLqCQUKV2KcEGHINBrjoRZal5eMLpW5WARcZB\/8CQP+JhOfyFtiOnzPKKnfGZ9MVC8RCBuB0PdoY+NhczJ+GpaAiZ+51EhEwD8EdvIxuCHu8JOGQuYgUPI74OxzQNHfgeKTl8q4D3sKGO5roSChufTQkOkULSat1AiWCx8Aha8B+X8ATpt2SylgTLptmeLFRszhr5V8Mkl6iYCnBJoAVpyb92ajQ08di8\/G+LESnyPTqERABKJHgLe3d0t4AAAQAElEQVQscwMvPeCHOT9pLoVNTHiFERgX9hnRsQ14zwiad4zwOG5WaPLfMCLkdWMmPGsuAVnLAs5ml1uhEScFrwCnjFg58XvghKl36kWA6SUnYL8z3MIPnHihL\/RDJgLhJmDe2\/ZN6EUYbl\/joP3kOBiDhiACIuA3Ap0zL\/co5EOdUS6k8EoQw2QjaD4yqzH5RtScumR5uwFru0z413LLNwLmnLkadMGIFa6+uE7YHo2aiaFNdxHehcSEa7ryGOem4UWVAN9zXlk1A1m1ahW6d+9+mZ08af4DmTorV67E4MGD0bdvX8yfP9+klL8KCgowZswY9OrVCyNHjsSRI0fKM8xx7969GDZsmM2bMmUKiou5icxk+PwlAePzCZJ7IhCTBP7ptnK3qSgY44c6w0uXfmBUi3kxAEUMrxoxi3EWZbXKCydMozGdxuZ47tphyDa4wmP\/CmYBNsaQFf5pCGMyEQgfAb7fvLJqvKQIycrKgrM1a9bgxhtvRNu2bXHw4EEsXboUS5YswcaNG7Fnzx5s3rzZtjR37lz0798f27dvx8SJEzFtGn9ADCgtLbXx6dOn27wOHTpg4cKFto7fDxIwfp8h+ScCMUQg6OroSUDHTICrH\/xAh\/nHkIqDIc0oDvMCxQvvlGbcGYWMFSOXqrE4z52xTmuTx9DVYRgUL2zAJphC7PMa48s\/S8AYGnqFkwDfqF5ZHfwsKyvDnDlzMHv2bFt6\/fr1GDt2LPr164dOnTph6tSp2LBhA86dO4edO3di5syZyMjIwOjRo5GcnGwFz44dO9CjRw+MGjXK5s2aNQtsh23bRn18SPaxb3JNBEQglgn8d8iD4\/ih7sQM405gGJFB0WECOEHC0D0OhvHKxmfdhaY1N+01cQ0wg3EqHZNuV2J4OWvN1lgmKd9jhYB7z3kR1mHM69atw9VXX20vGbE4V2AoXBinde7cGfv370dOTg7S09PRtCn\/YzAH6NKlC\/bt22dFTGidtLQ0lJSU4MQJc522vKhvjxIwvp0aOVZ\/AqrhKwKDzYrH6hDhwJWQJsbDUGtuzmlGsVCEUIw0N+WoQyqbzTN1m5nyLGsVT2ghk26vSbnQlAXFy2+MD53MCozpSi8RCCsBvucaaT9el47uE7rV6iZXSHip6MEHHwyWzc\/PR4sWXJcsT2rdujVyc3ORl5eHVq24zlmeziPLMZ3WsqX5D8jES8ayrHfp1LeBBIxvp0aOiUAcEDAipux\/t6KUvyLthsMP+CRzwtCtxLiQgoSfvwxDLdWUDz1nnELFhWyL5wyNXTTFS04DZfPNKhBFjDnXSwTCTsC89+yqXyPCGaPzkfWb7Fpd5aUfipjevXsHy7Zp0wZFRUXB88LCQnBFJTU1FWfOnAmmMxKad\/Zs6H9Q2LJsi+X8bBIwHs6OmhIBEahIoMwsXRd\/5nacywMKjgHn3GcrV7L5Ie9CJ2CYRiESahQp\/OPRpbGOi7O8Ozdx3jl91vRxxqx+nzcC5sLkyRUd0pkIhJOAeQ82VsAE69fi55YtW3DHHXdUKMVLQYcOHQqmHThwAF27drX7YY4ePQruhXGZhw8ftnm8zMS4Sz9+\/Li9C6l9+\/YuybehBIxvp0aOiUDsE\/goREB8VGpEzCngPSNkcguAInenJj\/0KUIYOmESGjLdGctR7LhzE35UBpw\/D5wybWe\/B5ww4TnTNsXMR0ZAhfoQ+0Q1Al8TMO\/HoABpbLyWgXJvS8+ePSuU4h1KmzZtskKFqzOrV6\/G+PHj7ebc4cOHY+3atbY8N\/TyEtKAAQMwdOhQHDt2zO6FYSZv0x43blyF\/TJMD7M1qHkJmAZhUyUREIG6ELi4bVuwWMqlWIkRMqfM6kiWWSU5YsTM0Q+N+DCC5rRZxabwsMV4iYmfTpW+BChWzhmxUlAIHDP13jH13zaixfzRCLZptIytTvHiLNQHm6mDCISLQKX3a6PETC0+cu8K7ygKLUZBs2DBAitaKEwmTJgACheWWbRokd3MO3DgQHDz74oVK5hshQpFy+LFizFo0CCkpKTYO5tsps8P\/IjwuYtyTwREIB4IUJO4z3cX56rMuQtGwBhB86G5zHTUiJrDRpAcfhc4\/A6QY0La4Rxzboxi5Z3jwPsngVxzqYiCh205Poxz\/wvPS3gw5kIT1UsEqibgVap7g3sR1uLT888\/j1tvvfWyUhQha9asAS8xjRgxIpjfrFkze7v17t27QZETKn46duwICphdu3bZZ8LwFutgRR9HJGB8PDlyTQRinYBbdXGf5zx3cYZcMfnIDLLEWOjLnXO1xprJdOVYx5zaG454NYltsjzTKWD0oUY6MhGIfwL6vx7\/c6wRikD0CAwZYlfR6QAFCwUGQ25xYcgtLUyj+KAIYTkXMo3mzlmWdWguzjwKG9ZjOxQzjLMMjfGkzEwGfjb5Fi8E+KbzyuKFSRjHIQETRrhqWgQSnUDKvHkWAT\/TGXHCg+dOxDBO8REqVlwa0xmnMc42nFG4UMC4dJZh+1yVcWVSMjNxxSUfXJpCEQgbAb4JvbKwORk\/DUvAxM9caiQi0DACYayVbFZgki4JCH6uc4WEoTOKGGcUH8ynOxQmlc0JFu5xodihcGFdtsWQxjjbYZxh04kTkWJ8YJsyEQg7Ab4BvbKwOxv7HUjAxP4cagQi4GsCKYEAkpctsz6GfrZzpYRCI9QoOkLPqyrDZ4byWXfu0TB8TAzrhJZtemnlpbnp23asgwhEgkDoG7yx8Uj4G+N9SMDE+ATGgfsaQgIQSJ40CWVbt+K8Gav7XKdYcXEKEJoTI4zTKFQYOnP5DCvnsT2WO2f6aGH6kngxIPSKLAH3hvYijKznMdmbBExMTpucFoHYIlCSk4OjkyfjPeP2YWOnjfHDh5\/zFB2VQ6ZVNoqWymk8p3BhWxRHx0y7ucbeuf12c9RLBCJMgG9kryzCrsdid\/x\/H4t+e+ezWhIBEQg7gZNGvFDElJqeuJfllAnfMcaQgoNp\/NynIAk1rrI4o1BxeSzLvTB5po33jfGXYxgvNnGmnzeC6QPTpznVSwQiR4BvTK8scl7HbE8SMDE7dXJcBGKDAIXL+W3bwI23oR5TtFDAnDCJXJl5y4RcnXnbhFxJOWnCD4wxnyKFlmPO+Usvr5nwH8aOGiswxrZp3ATMzb2Mn16+HOzbZOslApEh4JV4YTuR8bhRvUS7sgRMtGdA\/YtAnBMozcmxI6SwoMDghw5XYpjIz2maO+edR9zDQlFCAfOhKcSQ51ypyTfnZ41RoJgArMt2GWca4zSes02mMS4TgYgQ4BvSK4uIw7HdCT9LYnsE8l4ERMDXBKpaBeFlHgoMOu4+7xl3aYxTzLjQpTtBQpHCS0rMpyhiGwx5TmM+w6r6ZrosHgj4bwwlRlJ7Zf4bnf88koDx35zIIxGIKwKtJk0CxQVFBo1xCoxQMcJ0Gm+F5uCZx3KM01ie5xQtLMe9MKEh4ywXaryVuoWeAROKRPEwE\/BKvLCdMLsaF81LwMTFNGoQIuBvAs2NkKDICDWKEq6yUKww7sRJMzMUxmkUKjSXxjjNtWOKWnHEkMb2GNJamj4ZhsvUrghUJkDh4ZVVblvnlxOQgLmciVJEQAQ8JtBu61ZcYQSFEx4uZDcUHdzQ64QMP5QoXmgs50KKGJZ3RtHD\/IsmgW3Q2IY5BcXL1ZcensdzmQhEgkBJhC4hnTp1CmPGjEGfPn3w1a9+Fe+9x23w5SNcuXIlBg8ejL59+2L+\/PnlieZYUFBg6\/Tq1QsjR47EkSNHTGr5a+\/evRg2bBiYN2XKFBQX836+8jw\/H\/lZ4Wf\/5JsIiECVBGIvsToRw0tDHA0FCIUM406IhMYvmpPK6SzPNJrJBkVN23nz0MUIJp7LRCCSBEoiJGAeeugh3HLLLfjzn\/+M3r17Y+nSpXaYBw8etPElS5Zg48aN2LNnDzZv3mzz5s6di\/79+2P79u2YOHEipk2bZtNLS0ttfPr06TavQ4cOWLhwoc3z+0ECxu8zJP9EII4IUMS0MSsjF82YuHpCcyssjFPMUMjQWIbChGaK29uwKVAYd8ZzfoixDaZ1McKlrX4+gChkcUrg2LFjeOuttzBjxgxkZGTY8LHHHrOjXb9+PcaOHYt+\/fqhU6dOmDp1KjZs2IBz585h586dmDlzpq0zevRoJCcng4Jnx44d6NGjB0aNGmXzZs2aBbZTVub+59mmfXlI9qVXcsr3BOSgCDSEwIWcHJx85hlwgbrINHDBGF8ULzQKEYY0ihNnoWWY58wJHrbHMicffZSBTASiQqAkAiswhw8fRseOHTFp0iS7+nLPPffg9ddft+OlIKFwsSfm0LlzZ+zfvx855v9deno6mjbl\/zCTYV5dunTBvn37rIgJrZOWloaSkhKcOMEnMJmCPn5JwPh4cuSaCMQbgfcmT8aFbdvA26IpXihics0g+XwX\/hQAz\/l3H40fTqHGVRmWoZ0xdQqMsS7r8FIS84tM2\/oZAQNGr6gQiISAOXnyJF577TW7Z2WrWXH8p3\/6J3zrW9+y483Pz0eLFnx2tT1F69atkZubi7y8PLRqxZ8\/LU\/nkeWYTmvZkj+RytRyY1nWKz\/z75GfD\/71rlrPlCECIhBrBAqNuKDAoHgJ9d2tohSaRIoRChMan9IbakxjGRpXXCiATBX7onhxxj5oNkMHEYggAS8EzC9+3ASf6f7xSkll9yk8rr\/+eowfPx7t2rWzl4WysrKsSGnTpg2Kivi\/qLxWYWEhuKKSmpqKM2fOlCdeOobmnT3Lx0NeyjABy7ItE\/X1SwLG19Mj50QgfghQwHA0XF3hpSHGeSmIYUPMtVFVXQmYqqgoLdwEvBAw42Y0wQtZlPVVe9vFXPqpLDhSUlJA46WgQ4f4YxvldQ8cOICuXbva\/TBHjx61e2FQngVeimIeLzMxfikZx48ft3chtW\/f3iX5Nkz2rWdyTAREIK4IXHibv3JUPiR+PIeKl9B4eYmqjxQ\/oTmungtD8xQXgUgT8ELAuDaq8\/3GG2+0G3BfeOEFW2TFihW47rrrwBWTMWPGYNOmTVaocBPu6tWr7UoNN\/sOHz4ca9eutXW4oZcrOQMGDMDQoUPBjcHcP8PMVatWYdy4cRX2yzDdjyYB48dZkU8iEIcEWt92W3BUFCIUHc64muLiwUJVRFjOJbM84+5DzJ2z7ZZDhjBL5j8Cce2REx9ehDWBevLJJ\/HUU0+BYuaPf\/wjFi1aZIv37NkTCxYssKKFwmTChAmgcGEmy3Az78CBA7Fu3TpQ+DCdG3spWhYvXoxBgwbZlZw5c+Ywy\/fm\/u\/73lE5KAIiENsErpw0KTiA0ksxChIKjxRzztAEYFgXc2X5Icby7rxZZqZ9kB3PZSIQSQJeCBfXRk1+84FzXE3hXUT\/93\/\/Zx9A58pThKxZswZbtmzBiBEjXDKaNWuG2bNnY\/fu3VbkcFXGZfKuJgqYXbt22WfC8BZrl+fnkP\/3\/eyffBMBEYgjAt22So5rMwAAEABJREFUbq0wGidcKECqM5bhBxXD6sqwUebx94\/4LBieV2lKFIEwEnDiw4swjG7GTdP8XIibwWggIiAC\/ibQasgQXJ+dbVdISkJcpfiozng\/Bn9GgGFomeamfui5Ey8MTZZeIhBxAl4IF9dGxJ2PwQ4lYGJw0uRyzBKQ44YAL\/F0Nysx7efNQ5K53GOSgq9QQVJbnJV4CYqChb97dK0RRowzXSYC0SDgxIcXYTT8j7U+JWBibcbkrwjECYGLZhxFOTngEyjOmXjoiow5rfbF\/TN8cB2fdnHGlGJ41rRjonqJQFQJlKAJvLKoDiRGOpeAiZGJ8sRNNSICPiFwPBDAe48+an9SgGKED6bLN77RTpuQD7Djg+sq20mTl2eM6RQurEvx8r5pK3vyZJOjlwhEj4BX4oXtRG8UsdOzBEzszJU8FYG4IHBq+XIrXqoaTKlJ5BN2L5qQ4qSymeQKL5ZnAldvck27Z7Zt46lMBKJCoEQrMBHlHkkBE9GBqTMREAF\/Ejj1zDOXOUYBwkQ+w4VhQ40\/FNnQuqonAiIQWwQkYGJrvuStCMQ8Af4ideVBcMMub5OunF7bOeuFlqmq7dB8xUUgnATCtwITTq9jt20JmNidO3kuAjFLwAkPF3IgvKMo9Jxp1RnL0arLV7oIRIOABExkqUvARJa3ehOBhCfQYd48y8AJEIY0m2gOjNdmpliFF8szoeOlthmXeUdALdWNgARM3Th5VUoCxiuSakcERKBOBPiTApVFDCs6EcJ4XYyXnFiHxvIUL6lDhjAqE4GoEJCAiSx2CZjI8lZvIiAChkCHQABdli0zMcAJEJ4wfrmVl6mczktOrMMH4\/XcuhUdTZs8l4lAtAhIwESWfHJku1NvIiACIlBOgCsx\/crKwNUYihEnUMpzaz7ybiU+dTfTiKCbsrOhlZeaeSk3MgQkYCLD2fUiAeNIKBSBaggoObwEuBpzMTPTPpGXD6fjLdU0PuPloumaxjiNz4Xhc2L44LtkU6ftpEmmhF4i4A8CJWY90Svzx4j87YUEjL\/nR96JQMIQ4KrKRTNaiheaEysULIzTKGJYxhTTSwR8R6AkAgKmuLgY3bt3r2D3339\/kMXKlSsxePBg9O3bF\/Pnzw+mFxQUYMyYMejVqxdGjhyJI0eOBPP27t2LYcOG2bwpU6aAfQQzfRyRgPHx5JS7pqMIiEBVBK4wKzBVpStNBKJFIBIC5vTp00hLS0NWVlbQfvazn9khHzx4EEuXLsWSJUuwceNG7NmzB5s3b7Z5c+fORf\/+\/bF9+3ZMnDgR06ZNs+mlpaU2Pn36dJvXoUMHLFy40Ob5\/SAB4\/cZkn8iIAIiIAIxQSASAqawsBBt2rSpksf69esxduxY9OvXD506dcLUqVOxYcMGnDt3Djt37sTMmTORkZGB0aNHIzk5GRQ8O3bsQI8ePTBq1CibN2vWLLCdsjKuiVbZjW8Sk2vzRPkiIAIiEG4CzRuwmtK8a9dwu6X2RaBeBCIlYChIJkyYgJtuugn\/9m\/\/hv3791s\/KUgoXOyJOXTu3Nnm5eTkID09HU2bNjWp5a8uXbpg3759VsSE1uHqTklJCU6cOFFe0MfHZB\/7JtdEQAREQAREIGYIVBIwaMx5dYNOSkpCz549MWnSJGzZssXudZkxY4Ytnp+fjxYtWtg4D61bt0Zubi7y8vLQqlUrJgWN5ZhOa9myZTCdEZZlPcb9bBIwfp4d+SYCCULgigaswCQIGg0zhgg0RrC4uq\/+eB9+1n11taPu3bs3VqxYgc9+9rNo164dHn74YWRnZ+PDDz+0l5aKingvX3l1Xm7iikpqairOnDlTnnjpGJp39uzZS6nlActWd5mqvIQ\/jhIw\/pgHeSECIiAC3hJQaxEnUIomaKzdNONTmJg1sVrfjx49WuEOIrdXJSkpye57OXToULDugQMH0NVcauUlItbjpSeXefjwYZvHy0yMu\/Tjx4\/bu5Dat2\/vknwbSsD4dmrkmAiIgAiIgAhUJMDVlvvuuw9vvvkmKF6efvppXHfddXY1hrdJb9q0yW7aZd7q1asxfvx4uzl3+PDhWLt2rW2MG3p5CWnAgAEYOnQojh07ZvfCMHPVqlUYN25chf0yTPejScD4cVbkkwjEPoF6jSDtttvqVZ6FddmJFGR+IlCCFHhl1Y3r1ltvxb333mvvGurTpw9effVVPPnkk7Y498YsWLDAihYKE270pXBh5qJFi8DNvAMHDsS6devsZSimc2MvRcvixYsxaNAgpKSkYM6cOczyvUnA+H6K5KAIxD+BqyZNqtcgKV7qW6deHaiwCDSAgNvH4kVYU\/ff\/va3wZUYrsL8\/Oc\/tw+1c+UpQtasWWM3+I4YMcIlo1mzZpg9ezZ2794NihzeTu0yO3bsCAqYXbt22WfC8BZrl+fnUALGz7Mj3xpOQDVjjsDA7Gx0mTevRr+vyMy0ZW42ZWssqEwRiAIBL4SLayMK7sdclxIwMTdlclgE4pNAc4qTQAAUMn22bsV1y5ZZsUJRwzjTKFy6mDLxSUCjinUCTnx4EcY6i0j4LwETHspqVQREoIEEKGTShgwBLxFRrNAYZ1oDm1Q1EYgIAS+Ei2sjIg7HeCcSMDE+gXJfBERABETAHwSc+PAi9MeIouFF3fuUgKk7K5UUAREQAREQgWoJeCFcXBvVdqKMIAEJmCAKRURABERABBKdQGPG78SHF2Fj\/EiUuhIwiTLTGqcIiIAIiEBYCXghXFwbYXU0ThqXgImTidQwREAE4oGAxhDLBJz48CKMZQ6R8l0CJlKk1Y8IiIAIiEBcE\/BCuLg24hqUR4OTgPEIpJoRgXggoDGIgAg0nIATH16EDfcicWpKwCTOXGukIiACIiACYSTghXBxbYTRzbhpWgImbqYyHgaiMYiACIiACIhA3QhIwNSNk0qJgAiIgAiIQI0E3OqJF2GNHSnTEpCAsRjKDzqKgAiIgAiIQEMJeCFcXBsN9SGR6knAJNJsa6wiIAIiIAJhI+DEhxdhXZzcuXMnunfvjkOHDgWLr1y5EoMHD0bfvn0xf\/78YHpBQQHGjBmDXr16YeTIkThy5Egwb+\/evRg2bJjNmzJlCoqLi4N5dYxEpZgETFSwq1MREAEREIF4I+CFcHFt1MaGImPOnDm48sorg0UPHjyIpUuXYsmSJdi4cSP27NmDzZs32\/y5c+eif\/\/+2L59OyZOnIhp06bZ9NLSUhufPn26zevQoQMWLlxo8\/x+kIDx+wzJPxEQAREQgZoJ+CTXiQ8vwtqG9OSTT+KLX\/xiBQGzfv16jB07Fv369UOnTp0wdepUbNiwAefOnQNXa2bOnImMjAyMHj0aycnJoODZsWMHevTogVGjRtm8WbNmge2UlZXV5kLU85Oj7oEcEAEREAEREIE4IOCFcHFt1IQjKysLmzZtsgIltBwFCYWLS+vcuTP279+PnJwcpKeno2nTpi4LXbp0wb59+6yICa2TlpaGkpISnDhxIljWrxEJGL\/OjPwSARGIFQLyUwQsASc+GhMW\/ngNcrvfZ9ur7vDNb34T3\/ve9yoIEpbNz89HixYtGLXWunVr5ObmIi8vD61atbJp7sByTKe1bNnSJduQZVnPnvj4IAHj48mRayIgAiIgArFDoDHCxdVtMmMcWmU9V+2guUmXKybcqFu5UJs2bVBUVBRMLiwsBFdUUlNTcebMmWA6I6F5Z8+eZVLQWJZtBRN8GpGA8enEyC0RqDMBFRQBEfAFASdCvAirG9ALL7xgN+jy7iPa4cOHcccdd9g0CpvQO5IOHDiArl272v0wR48etXthXLusxzxeZmLcpR8\/ftzehdS+fXuX5NtQAsa3UyPHREAEREAEYomAF8LFtVHduJ955hlwD4wzbsD9wx\/+gM9\/\/vP2NmnujeGmXW7CXb16NcaPH2835w4fPhxr1661zXJDLy8hDRgwAEOHDsWxY8fsXhhmrlq1CuPGjbvs8hTz\/GYSMH6bkdjzRx6LgAiIgAgYAk58eBGa5ur96tmzJxYsWGBFC4XJhAkTQOHChhYtWmQ38w4cOBDr1q3DihUrmGyFCkXL4sWLMWjQIKSkpIC3Z9tMnx8kYHw+QXJPBERABEQgNgh4IVxcG3UdMS8pXXfddcHiFCFr1qzBli1bMGLEiGB6s2bNMHv2bOzevduKHN5O7TI7duwICphdu3bZZ8LwFmuX5+cw9gWMn+nKNxEQAREQgYQh4MSHF2HCQGvEQCVgGgFPVUVABERABEQgVgnEut8SMLE+g\/JfBERABETAFwS8WHlxbfhiQD53QgLG5xMk90RABEQgPgnE36ic+PAijD863o9IAsZ7pmpRBERABEQgAQl4IVxcGwmIr95DloCpNzJVEAERiAcCGoMIeE3AiQ8vQq99i8f2JGDicVY1JhEQAREQgYgT8EK4uDYi7nwMdigBE4OTJpfjgYDGIAIiEG8EnPjwIow3NuEYjwRMOKiqTREQAREQgcQjUGKG7JWZpvSqmYAETM184jZXAxMBERABEfCYgFfihe147Fo8NicBE4+zqjGJgAiIgAhEngCFh1cWee9jrscoCZiY4ySHRUAEREAERKBmAl6JF7ZTc0\/KNQQkYAwEvURABERABESg0QQoPLyy6pxRepCABEwQhSIiIAIiIAIi0AgCXokXttMINxKlqgRMosy0xikCIiACjSegFmoiQOHhldXQzzPPPIO77roLN910Ex544AGcOHEiWHrlypUYPHgw+vbti\/nz5wfTCwoKMGbMGPTq1QsjR47EkSNHgnl79+7FsGHDbN6UKVNQXFwczPNzRALGz7Mj30RABERABGKHgFfihe1UM+p9+\/bhJz\/5CR5\/\/HG88sor6NatG+bMmWNLHzx4EEuXLsWSJUuwceNG7NmzB5s3b7Z5c+fORf\/+\/bF9+3ZMnDgR06ZNs+mlpaU2Pn36dJvXoUMHLFy40Ob5\/SAB4\/cZkn8iIAIfE1BMBBKcQHp6Oh577DH07t0bbdq0wec+9zm89dZblsr69esxduxY9OvXD506dcLUqVOxYcMGnDt3Djt37sTMmTORkZGB0aNHIzk5GRQ8O3bsQI8ePTBq1CibN2vWLLCdsrIy26afD8l+dk6+iYAIiIAIiEDMEODKiVdWzaA7duyIESNG2FwKkKefftpe\/mECzylcGKd17twZ+\/fvR05ODtKN8GnatCmTrXXp0gVczalcJy0tDSUlJRUuS9kKPjxIwPhwUuSSbwnIMREQARGonoAH4iX9mR+j2\/Du1fdxKefRRx\/FnXfeiaNHj2LGjBk2NT8\/Hy1atLBxHlq3bo3c3Fzk5eWhVatWTAoayzGd1rJly2A6IyzLeoz72SRg\/Dw78k0EREAERCB2CHggYPJHz0D2c1mo7d+8efPwxz\/+ET179sSkSZPASz68pFRUVBSsWlhYCK6opKam4syZM8F0RkLzzp49y6SgsSzbCib4NCIB49OJqdItJYqACIiACPiXgO+U41IAABAASURBVAcCBq6NakbJu4deeOEFm3vttdfi+9\/\/PngX0cmTJ+2+l0OHDtk8Hg4cOICuXbvadK7UcC8M02mHDx+2ebzMxDjTaMePH7d3IbVv356nvjYJGF9Pj5wTAREQARGIGQJOfHgRVjNo7ln51re+Zfe2XLx4EVu3bsU111yDtm3b2tukN23aZDftckVm9erVGD9+vN2cO3z4cKxdu9a2yg29vIQ0YMAADB06FMeOHbMbepm5atUqjBs3DqH7ZZjuR6uPgPGj\/\/JJBERABERABPxBwAvh4tqoZkTcwHvvvffi85\/\/PG688Ub8+te\/xi9\/+UskJSXZy0kLFiywooXCZMKECaBwYVOLFi0CN\/MOHDgQ69atw4oVK5hshQpFy+LFizFo0CCkpKQEb8u2BXx8kIDx8eTINREQAREQgRgiUGJ8rdIakG6qVPd65JFHkJWVBd4+vXz5clx\/\/fXBohQha9aswZYtW4J3KzGzWbNmmD17Nnbv3g2KHN5OzXQa72yigNm1a5d9JgxvsWa6300Cxu8zJP9EQAREQARig0CJcdMrM03pVTMBCZia+Sg3TASo9rt3745Qe+ihh8LUm5oVgcQmoNFHiIBX4oXtRMjlWO5GAiaWZy+GfedjrLkESuOjrps3b26fDhnDQ5LrIiACiU6AwsMrS3SWdRi\/BEwdIKlI+Ajwtj4+7pqPr+aO+PD1pJajR0A9i0CCEPBKvLCdBEHWmGFKwDSGnuo2mgCFS58+fTB58uRGt6UGREAERCCqBCg8vLKoDiQ2OpeAiY15iksveeseH8r0gx\/8IKzjU+MiIAIiEBECXokXthMRh2O7EwmY2J6\/mPWeD2N64oknwB8iu+KKK2J2HHJcBERABEQgOgQkYMLOXR1UReCHP\/wh+Ojrf\/mXfwneiTRmzJiqiipNBERABGKDAFdOvLLYGHFUvZSAiSr+xO38pz\/9qX0QU9Yvf4mssjIb5yWlxCWikYuACMQ8Aa\/EC9uJeRgeDKCWJiRgagGk7AgQyMmJQCfqQgREQATCTIDCwysLs6vx0LwETDzMYjyMIScnHkahMYiACMQPgfqPxCvxwnbq33vC1ZCASbgp99mAMzOBzEyfOSV3REAERKABBCg8vLIGdJ9oVSRgEm3G\/TrezEy\/eia\/RCA6BNRr7BHwSrywndgbfcQ9loCJOHJ1KAIiIAIiEJcEKDy8shoAvfrqq7jnnnvQu3dv3H\/\/\/Th27Fiw9MqVKzF48GD07dsX8+fPD6YXFBSAd3r26tULI0eOBJ\/B5TL37t2LYcOGgXlTpkxBcXGxy\/J1KAHj6+mRcyIQNQLqWAREoL4EvBIvbKeavk+cOIGvf\/3r+M53voOdO3fipptuwne\/+11bms\/XWrp0KZYsWYKNGzdiz5494G\/NMXPu3Lno378\/tm\/fjokTJ4K\/R8f00tJSG58+fbrN69ChAxYuXMgs35sEjO+nSA6KgAiIgAjEBAEKD6+shgE\/8sgjuPnmm9G6dWt87nOfwxtvvGFLr1+\/HmPHjkW\/fv3QqVMn8HfmNmzYAP7mHMXOzJkzkZGRYX84Nzk5GRQ8O3bsQI8ePTBq1Cibx593YTtlZWW2TT8fkv3snHxLYAIaugiIgAjEGgGvxAvbqWbs7dq1s2LDZf\/tb3+zl4t4TkFC4cI4rXPnzti\/fz9ycnKQnp6Opk2bMtlaly5dsG\/fPitiQuukpaWhpKQEXOmxBX18kIDx8eTINREQAREQgRgiUGp89cpMU7W9uKry1FNPgSsyLJufn48WLVowao0rNLm5ucjLy0OrVq1smjuwHNNpLVu2dMk2ZFnWsyc+PkjAVD05ShUBERABERCB+hHgykkjLf1vP0a3Z7rX2i\/3tjz00EOggOFKCyu0adMGRUVFjForLCwEV1RSU1Nx5swZm+YOoXlnz551yTZkWbZlT3x8kIDx8eTINREQAREQgRgi0EjxAlM\/\/8YZyB6dhZr+cSPu8uXLsW7dOvTp0ydYlJeCDh06FDw\/cOAAunbtavfDHD161O6FcZmHDx+2eRQ\/jLv048eP27uQ2rdv75LqGUauuARM5FirJxEQAREQgXgmYAQIRYgnVg0n7k1xKy\/ckBtajLdJb9q0yQoVbsJdvXo1xo8fbzfnDh8+HGvXrrXFeemJl5AGDBiAoUOH2tuwuX+GmfxNunHjxlXYL8N0P5oEjB9nRT6JgAiIgAg0iEC8V3rxxRfx\/vvvg+Kje\/fucHby5En07NkTCxYssKKFwmTChAmgcCGTRYsW2c28AwcOtCs3K1asYLIVKhQtixcvxqBBg5CSkoI5c+bYPL8fJGD8PkPyTwREQAREIDYIlBg3vTLTVFWv0aNHIysr6zJr27atLU4RsmbNGmzZsgUjRoywaTw0a9YMs2fPxu7du63ICV296dixIyhgdu3aZZ8Jw1usWcfvJgHj9xmSfyIgAjFEQK4mNIESM3qvzDSlV80EJGBq5qNcERABERABEagbAa\/EC9upW48JXUoCJqGnX4OPNwIajwiIQBQJUHh4ZVEcRqx0LQETKzMlP0VABERABPxNwCvxwnb8PVJfeCcB44tpiBcn\/D2O06dP2136r776atBRPszphhtuwOuvvx5MU0QEREAEGkSAwsMra5ADiVVJAiax5juhR8snS44cORJ8NoID8cILL+Caa64J\/paIS1coAiIgAvUm4JV4YTv17jzxKsSVgEm86dOI60vg3nvvxYYNG+Aenb1x40Ywrb7tqLwIiIAIXEaAwsMru6xxJVQmIAFTmYjO45dATg4GDx4MPiKbz0koKCjAX\/7yF3zpS1+K3zFrZCIgApEj4JV4YTuR85o9xaRJwMTktMnpehEwwgWBANCtG7BtG+677z77SO3nn38en\/70p9GuXbt6NafCIiACIlAlAQoPr6zKDpQYSkACJpSG4vFJwIgWPPMMkJkJDBmCe+65B6+99hp+\/vOf4+67747PMWtUIhBLBOLFV6\/EC9uJFyZhHIcETBjhqmkfEFi+HHj0UWDiRCA72zrEx2bfdttt4I+i3XnnnTZNBxEQARFoNAEKD6+s0c7EfwMSMPE\/x4k7wkCgXLxs3QoEAhU4dO3a1a7ENG3atEK6ThKSgAYtAt4Q8Eq8sB1vPIrrViRg4np6E3Rwbs8LLxtRvGRmBkFcvHgR27dvx29+8xtMmjQpmK6ICIiACIhAbBGQgImt+ZK3tRGgeFm+HHbPCy8ZZWZWqHHXXXdhypQpmDdvHjIzMyvkRe1EHYuACMQHAa6ceGXxQSSso5CACSteNR5RAhQvt99e3iXFS3mswnHTpk3Yt28fxowZUyFdJyIgAiLQaAJeiRe2U4MzpaWlePzxx9GtWzecPHmyQsmVK1fax0X07dsX8+fPD+bxsRH83OvVqxf4QM8jR44E8\/bu3Ythw4aBefwDr7i4OJjn54gEjJ9nJzK+xUcvTrxws24gEB9j0ihEQARiiwCFh1dWw8hnzJiBq6++GsnJFb\/CDx48iKVLl2LJkiXgQzr37NmDzZs325bmzp2L\/v3720voE83n5LRp02w6xRDj06dPt3kdOnTAwoULbZ7fDxVH73dv5Z8IVEVA4qUqKkoTARGINAGvxAvbqcH3mTNngiKkcpH169dj7Nix6NevHzp16oSpU6faJ4+fO3cOO3fuBOtlZGRg9OjRVvxQ8OzYsQM9evTAqFGjwLxZs2aB7ZSVlVVu3nfnyVH3SA6IQGMIBAIALxstWwYEAo1pSXVFQAREoHEEKDy8sho86d69e5W5FCQULi6zc+fO2L9\/P3LMH3np6ekIveuyS5cu9nJ65TppaWkoKSmxj5lw7fg1lIDx68zIr9oJBALlm3V5p9GQIbWXVwkREAERCCcBD8RL+vkfo9vZqgVKZdcrn+fn56NFixbB5NatWyM3Nxd5eXlo1apVMJ0RlmM6rWXLlkwKGsuyXjDBpxEJGJ9OjNyqgYD5awKBwMfiJTOzhsLKEgEREIEIEfBAwORjBrJTstCQf23atEFRUVGwamFhIbiikpqaijNnzgTTGQnNcz9uy3Qay7Itxv1sEjB+nh35djkBipfJk4G33wZ4p1Fm5uVllCICIhABAuriMgIeCBi4Ni5rvPYEXj46dOhQsOCBAwfAh3Yy\/ejRo+BeGJd5+PBhm8fLTIy79OPHj4N3IbVv394l+TaUgPHt1MixywhQvHC\/y223AdzzclkBJYiACIhAFAk48eFF2IBh8DZpPiqCQoWbcFevXo3x48fbzbnDhw+3P2LLZrmhl5eQBgwYgKFDh+LYsWPgXhjmrVq1CuPGjauwX4bpfjQJGD\/Oiny6nADFC1deJk4EAoHL85WScAQ0YBHwHQEvhItro5rB8bkv3MRLo0gZNGgQGP\/www\/Rs2dPLFiwwIoWCpMJEyaAwoVNLVq0CNzMO3DgQKxbtw4rVqxgshUqFC2LFy8G20pJScGcOXNsnt8PEjB+nyH5ByxfDnunEVdeAgEREQEREAF\/EigxbnllpqmqXm3btkVWVtZl5i75UISsWbMGW7ZswYgRI4JNNGvWDLNnz8bu3butyOEt0y6TP3BLAbNr1y7wmTCVny\/jyvktlIDx24zIn4oEAgHYX5PmnUaBQMW8qJ6pcxEQARGoRKDEnHtlpim9aiYgAVMzH+VGi0BODhAIwP6mEcVLZma0PFG\/IiACIlA3Al6JF7ZTtx4TupQETIxOf1y7nZMDLF8OvPSS7jSK64nW4EQgzghQeHhlcYYmHMORgAkHVbXZcAI5ObD7XdgCV14YykRABERABESgEoEGCphKrehUBLwgkJNTLl50p5EXNNWGCIhApAl4tfrCdiLtewz2JwETg5MWly7n5Ei8xOXEalAikEAEKDxqs7rmJxC2hg5VAqah5FTPOwI5OeXihQ+nCwS8a1ctiYAIiEBECdRVndSlXEQdj8nOJGBictriyOnMTCAzs3xA27YB27aVx3UUAREIBwG1GVYCdREmdS0TVkfjonEJmLiYxhgfBH\/TyG3Y5dN2u3UDAgEgEIjxgcl9ERCBxCJQV3FSl3KJRa4ho5WAaQg11fGeQGYmEAiU3zbtxAxvo05KApygycnxvl+1GFkC6k0E4ppAXYRJXcvENShPBicB4wlGNeIpgcxMIBAAKGTKyj7+4Ub+kKMTM9u2edqlGhMBERCBxhOoqzipS7nGexPvLUjAxPsMx8P4hgwBAoGKqzOPPgrUf3UG+icCIiAC4SNQF2FS1zLh8zJeWpaAiZeZTJRxZGYCgUD56gz3zvDOJY6dKzO0QADYto0pMhEQARGIMIG6ipO6lIuw6zHYnQRMJCdNfXlLIDMTGDIECAQAXmriJSf24FZneMkpEABycpgqEwEREIEwE6iLMKlrmTC7GgfNS8DEwSRqCJcIZGYCgcDHqzPz5pVncGWGFggA27aVp+koAiIgAp4TqKs4qUu56p3bu3cvhg0bhl69emHKlCkoLi6uvnCc5FRPf2dtAAAHH0lEQVQ1DAmYqqgoLfYJZGYCQ4YAgQBQeXWGYsatzsT+SDUCERAB3xCoizCpa5mqB1VaWopp06Zh+vTp2L59Ozp06ICFCxdWXTjOUyVg4nyCNbxLBDIzgUCgfHWGl5puu608I3Qj8LZt5Wk6ioAIiEAFAnU9qas4qUu5qvvcsWMHevTogVGjRiEjIwOzZs3C+vXrzd9pZVVXiOPU5Dgem4YmAlUTyMwEAgEgENDqTNWElCoCIuBTAgcPHkSnTp2C3qWlpaGkpAQnTpwIpiVKRAImUWZa46yeQGYmEAhodaZ6QsqJMgF1HysE6rKyUnOZ9PSV6NbtC9UOOC8vDy1btqyQ36pVK+Tm5lZIS4QTCZhEmGWNse4EMjOBQAAIBCquzoT+xIH2z9Sdp0qKQEIRqFmcALXn5+d\/CdnZv6mWWmpqKs6ePVsh\/8yZM2jTpk2FtEQ4kYBJhFnWGBtOIDMTCAQ+fogef95g2zaAt2pTyFxqecaMGZg5c+alM+CHP\/yhvUZ98eLFYFrsRuS5CIhA3QjULlDqImLKy1TdY+fOnXH48OFg5vHjx+1dSO3btw+mJUpEAiZRZlrjbDyBzMzyy0y8q4kP0Zs3L9jm9773PbzyyivYuXMnsrKysGzZMvzkJz9BcrL+iwUhKSICcU8g\/AJm6NChOHbsGLgXhjhXrVqFcePGoWnTpjxNKNOna0JNd2wO1pdeZ2YCQ4YEXePy7Q9+8AM89NBDePjhh23YjbdrB0soIgIiEP8Ewi9gKFQoWhYvXoxBgwYhJSUFc+bMiX+0VYxQAqYKKEoSgYYQuO222+wzGbLN6syXv\/zlhjShOiIgAjFNIPwChng6duwICphdu3bZZ8Ik6kqvBAzfDTWaMkWgbgReeOEFnDx5Etdeey2efvrpulVSKREQgTgiUGLG4pWZpvSqkYAETI14lCkCdSPw4Ycf2ktH\/\/3f\/40nnngCTz75JN588826VVYpERCBOCHglXhhO3GAJMxDkIAJM2A1nxgE+GjvCRMmoG\/fvrjmmmswe\/Zs\/Md\/\/MdltzsmBg2NUgQSlQCFh1eWqAzrPm4JmLqzUkkRqJbA6tWr8eCDDwbzR48ejZdffhl8wFQwUREREIFIEohCX16JF7YTBfdjrEsJmBibMLkrAiIgAiLgVwIUHl6ZX8foH78kYPwzF\/JEBEQgnghoLAlIwCvxwnYSEF89hywBU09gKi4CIiACIiACVROg8PDKqu5BqR8TkID5mIViIhBPBDQWERCBiBPwSrywnYg7H3MdSsDE3JTJYREQAREQAREQAQkYvQfCQ0CtioAIiEDCEeDKiVeWcPDqPWAJmHojUwUREAEREAERqIqAV+KF7VTVvtJCCcSrgAkdo+IiIAIiIAIiEAECFB5eWQTcjfEuJGBifALlvgiIgAiIgF8IeCVe2E60xhQ7\/UrAxM5cyVMREAEREAFfE6Dw8Mp8PVBfOCcB44tpkBMiIAIiIAIkEKt2\/nwHdOv2M8+M7cUqi0j5LQETKdLqRwREQAREIG4JvP\/+dmRlZXlmbC9uYXk0MAkYj0CqGREQgXggoDGIgAjECgEJmFiZKfkpAiIgAiIgAiIQJCABE0ShiAhEn4A8EAEREAERqBsBCZi6cVIpERABERABERABHxGQgPHRZETfFXkgAiIgAiIgArFBQAImNuZJXoqACIiACIiACIQQ8JWACfFLUREQAREQAREQARGoloAETLVolCECIiACIiACMUEgIZ2UgEnIadegRUAEREAERCC2CUjAxPb8yXsREAERiD4BeSACUSAgARMF6OpSBERABERABESgcQQkYBrHT7VFQASiT0AeiIAIJCABCZgEnHQNWQREQAREQARinYAETKzPoPyPPgF5IAIiIAIiEHECEjARR64ORUAEREAEREAEGktAAqaxBKNfXx6IgAiIgAiIQMIRkIBJuCnXgEVABERABEQg9gk0XsDEPgONQAREQAREQAREIMYISMDE2ITJXREQAREQgfggoFE0joAETOP4qbYIiIAIiIAIiEAUCEjARAG6uhQBERCB6BOQByIQ2wQkYGJ7\/uS9CIiACIiACCQkAQmYhJx2DVoEok9AHoiACIhAYwhIwDSGnuqKgAiIgAiIgAhEhYAETFSwq9PoE5AHIiACIiACsUxAAiaWZ0++i4AIiIAIiECCEpCAidLEq1sREAEREAEREIGGE5CAaTg71RQBERABERABEYgsgWBvEjBBFIqIgAiIgAiIgAjECgEJmFiZKfkpAiIgAiIQfQLywDcEJGB8MxVyRAREQAREQAREoK4EJGDqSkrlREAERCD6BOSBCIjAJQISMJdAKBABERABERABEYgdAhIwsTNX8lQEok9AHoiACIiATwhIwPhkIuSGCIiACIiACIhA3QlIwNSdlUpGn4A8EAEREAEREAFL4P8DAAD\/\/7r6oUcAAAAGSURBVAMAPOJZwrjqbTgAAAAASUVORK5CYII=","height":337,"width":560}} +%--- +%[output:751ae428] +% data: {"dataType":"image","outputData":{"dataUri":"data:image\/png;base64,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","height":337,"width":560}} +%---