applies formatter

Author: OmidS

source/PSID/IPSID.py

@@ -1,4 +1,4 @@
-""" 
+"""
 Copyright (c) 2020 University of Southern California
 See full notice in LICENSE.md
 Parsa Vahidi, Omid G. Sani and Maryam M. Shanechi

source/PSID/PSIDModel.py

@@ -1,4 +1,4 @@
-""" 
+"""
 Copyright (c) 2020 University of Southern California
 See full notice in LICENSE.md
 Omid G. Sani and Maryam M. Shanechi

source/PSID/PrepModel.py

@@ -9,6 +9,7 @@
 
 Optimal linear 'bottlenecking' or 'multitask learning'.
 """
+
 import numpy as np
 from scipy import sparse
 
@@ -29,6 +30,7 @@ class ReducedRankRegressor(object):
     - rrank is a rank constraint.
     - reg is a regularization parameter (optional).
     """
+
     def __init__(self, X, Y, rank, reg=None):
         if np.size(np.shape(X)) == 1:
             X = np.reshape(X, (-1, 1))
@@ -45,7 +47,7 @@ def __init__(self, X, Y, rank, reg=None):
         self.A = np.dot(np.linalg.pinv(CXX), np.dot(CXY, self.W)).T
 
     def __str__(self):
-        return 'Reduced Rank Regressor (rank = {})'.format(self.rank)
+        return "Reduced Rank Regressor (rank = {})".format(self.rank)
 
     def predict(self, X):
         """Predict Y from X."""

source/PSID/ReducedRankRegressor.py

@@ -1,4 +1,4 @@
-""" 
+"""
 Copyright (c) 2020 University of Southern California
 See full notice in LICENSE.md
 Omid G. Sani and Maryam M. Shanechi
@@ -14,48 +14,51 @@
 
 logger = logging.getLogger(__name__)
 
+
 def addConjs(vals):
     """Adds complex conjugate values for each complex value
 
     Args:
         vals (list of number): list of numbers (e.g. eigenvalues)
 
     Returns:
-        output (list of numbers): new list of numbers where for each complex value in the original list, 
+        output (list of numbers): new list of numbers where for each complex value in the original list,
             the conjugate is also added to the new list. For real values an np.nan is added.
-    """    
+    """
     vals = np.atleast_2d(vals).T
     valsConj = vals.conj()
     valsConj[np.abs(vals - valsConj) < np.spacing(1)] = np.nan
     return np.concatenate((vals, valsConj), axis=1)
 
 
-def drawRandomPoles(N, poleDist = {}):
+def drawRandomPoles(N, poleDist={}):
     """Draws random eigenvalues from the unit dist
 
     Args:
         N (int): number of eigenvalues to draw
-        poleDist (dict, optional): information about the distribution. 
+        poleDist (dict, optional): information about the distribution.
             For options see the top of the code. Defaults to dict.
 
     Returns:
         valsA (list of numbers): drawn random values
-    """    
-    nCplx = int(np.floor(N/2))
-    
-    if 'magDist' not in poleDist: poleDist['magDist'] = 'beta'
-    if 'magDistParams' not in poleDist and poleDist['magDist'] == 'beta': 
-        poleDist['magDistParams'] = {'a': 2, 'b': 1}
-    if 'angleDist' not in poleDist: poleDist['angleDist'] = 'uniform'
-    
+    """
+    nCplx = int(np.floor(N / 2))
+
+    if "magDist" not in poleDist:
+        poleDist["magDist"] = "beta"
+    if "magDistParams" not in poleDist and poleDist["magDist"] == "beta":
+        poleDist["magDistParams"] = {"a": 2, "b": 1}
+    if "angleDist" not in poleDist:
+        poleDist["angleDist"] = "uniform"
+
     # mag = np.random.rand(nCplx) # Uniform dist
-    if poleDist['magDist'] == 'beta':
+    if poleDist["magDist"] == "beta":
         a, b = 2, 1  # Use a, b = 2, 1 for uniform prob over unit circle
-        if 'a' in poleDist['magDistParams']: 
-            a = poleDist['magDistParams']['a']
-        if 'b' in poleDist['magDistParams']: 
-            b = poleDist['magDistParams']['b']
-        
+        if "a" in poleDist["magDistParams"]:
+            a = poleDist["magDistParams"]["a"]
+        if "b" in poleDist["magDistParams"]:
+            b = poleDist["magDistParams"]["b"]
+
         """
         import matplotlib.pyplot as plt    
         fig, ax = plt.subplots(1, 1)
@@ -66,34 +69,35 @@ def drawRandomPoles(N, poleDist = {}):
         plt.show()
         """
 
-        mag = stats.beta.rvs(a=a, b=b, size=nCplx) # Beta dist
+        mag = stats.beta.rvs(a=a, b=b, size=nCplx)  # Beta dist
     else:
-        raise Exception('Only beta distribution is supported for the magnitude')
+        raise Exception("Only beta distribution is supported for the magnitude")
 
-    if poleDist['angleDist'] == 'uniform':
+    if poleDist["angleDist"] == "uniform":
         theta = np.random.rand(nCplx) * np.pi
     else:
-        raise Exception('Only uniform distribution is supported for the angle')
-    
-    vals = mag * np.exp( 1j * theta ) 
+        raise Exception("Only uniform distribution is supported for the angle")
+
+    vals = mag * np.exp(1j * theta)
 
     valsA = addConjs(vals)
     valsA = valsA.reshape(valsA.size)
-    valsA = valsA[ np.logical_not(np.isnan(valsA)) ]
+    valsA = valsA[np.logical_not(np.isnan(valsA))]
 
     # Add real mode(s) if needed
-    nReal = N-2*nCplx
+    nReal = N - 2 * nCplx
     if nReal > 0:
         # rVals = np.random.rand(nReal)
-        rVals = stats.beta.rvs(a=a, b=b, size=nReal) # Beta dist
-        rSign = 2*(((np.random.rand(nReal) > 0.5).astype(float))-0.5)
+        rVals = stats.beta.rvs(a=a, b=b, size=nReal)  # Beta dist
+        rSign = 2 * (((np.random.rand(nReal) > 0.5).astype(float)) - 0.5)
 
         valsA = np.concatenate((valsA, rVals * rSign))
 
-    return valsA 
+    return valsA
+
 
 def generate_random_eigenvalues(count):
     """Generates complex conjugate pairs of eigen values with a uniform distribution in the unit circle"""
     # eigvals = 0.95 * np.exp(1j * np.pi/8 * np.array([-1, +1]))
     eigvals = drawRandomPoles(count)
-    return eigvals
+    return eigvals