You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: README.md
+32-13Lines changed: 32 additions & 13 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -21,7 +21,7 @@ To address these issues, we developed Flood Inundation Mapping Prediction Evalua
21
21
22
22
23
23
24
-
### **Repository structure**
24
+
### Repository structure
25
25
<hrstyle="border: 1pxsolidblack; margin: 0;">
26
26
27
27
The architecture of the ```fimeval``` integrates different modules to which helps the automation of flood evaluation. All those modules codes are in source (```src``` ) folder.
@@ -53,7 +53,7 @@ The graphical representation of fimeval pipeline can be summarized as follows in
53
53
</div>
54
54
Figure 1: Flowchart showing the entire framework pipeline.
55
55
56
-
### **Framework Installation and Usage**
56
+
### Framework Installation and Usage
57
57
<hrstyle="border: 1pxsolidblack; margin: 0;">
58
58
59
59
This framework is published as a python package in PyPI (https://pypi.org/project/fimeval/).For directly using the package, the user can install this package using python package installer 'pip' and can import on their workflows:
@@ -74,7 +74,9 @@ import fimeval as fp
74
74
```
75
75
**Note: The framework usage provided in detailed in [Here (docs/fimeval_usage.ipynb)](./docs/fimeval_usage.ipynb)**. It has detail documentation from installation, setup, running- until results.
76
76
77
-
#### **Main Directory Structure**
77
+
### Main Directory Structure
78
+
<hrstyle="border: 1pxsolidblack; margin: 0;">
79
+
78
80
The main directory contains the primary folder for storing the case studies. If there is one case study, user can directly pass the case study folder as the main directory. Each case study folder must include a Benchmark FIM (B-FIM) with a 'benchmark' word assigned within the B-FIM file and different Model Predicted FIM (M-FIM)
79
81
in tif format.
80
82
For mutilple case studies,the main directory could be structure in such a way that contain the seperate folders for individual case studies.For example, if a user has two case studies they should create two seperate folders as shown in the Figure below.
@@ -95,7 +97,7 @@ If user have more precise PWB, they can input their own PWB boundary as .shp and
95
97
```bash
96
98
PWD_dir = Path('./path/to/PWB/vector/file')
97
99
```
98
-
#### **Methods for Extracting Flood Extents**
100
+
#### Methods for Extracting Flood Extents
99
101
1.**```smallest_extent```**
100
102
The framework will first check all the raster extents (benchmark and FIMs). It will then determine the smallest among all the rasters. A shape file will then be created to mask all the rasters.
101
103
@@ -125,7 +127,7 @@ method_name = "AOI"
125
127
AOI = Path('./path/to/AOI/vectorfile')
126
128
```
127
129
128
-
#### **Executing the Evaluation framework**
130
+
#### Executing the Evaluation framework
129
131
The complete description of different modules, what they are meant for, arguments taken to run that module and what will be the end results from each is described in below **Table 1**. If user import `fimeval` framework as `fp` into workflows, they can call each module mentioned in **Table 1** as `fp.Module_Name(args)`. Here arguments in italic represents the optional field, depending upon the user requirement.
130
132
131
133
Table 1: Modules in `fimeval` are in order of execution.
@@ -143,8 +145,9 @@ Table 1: Modules in `fimeval` are in order of execution.
143
145
Figure 4: Combined raw output from framework for different two method. First row (subplot a and b) and second row (subplot c and d) is contingency maps and evaluation metrics of FIM derived using `PrintContingencyMaP` and `PlotEvaluationMetrics` module. Third row (subplot e and f) is the output after processing and calculating of evaluation with BF by unsing `EvaluateWithBuildingFoorprint` module.
144
146
145
147
## Installation Instructions
148
+
<hrstyle="border: 1pxsolidblack; margin: 0;">
146
149
147
-
### 1. Prerequisites
150
+
####1. Prerequisites
148
151
149
152
Before installing `fimeval`, ensure the following software are installed:
150
153
@@ -158,13 +161,13 @@ Before installing `fimeval`, ensure the following software are installed:
158
161
159
162
---
160
163
161
-
### 2. Install Anaconda
164
+
####2. Install Anaconda
162
165
163
166
If Anaconda is not installed, download and install it from the [official website](https://www.anaconda.com/products/distribution).
164
167
165
168
---
166
169
167
-
### 3. Set Up Virtual Environment
170
+
####3. Set Up Virtual Environment
168
171
169
172
#### For Mac Users
170
173
@@ -182,24 +185,40 @@ uv pip install fimeval
182
185
```
183
186
184
187
### Google Colab Version
188
+
<hrstyle="border: 1pxsolidblack; margin: 0;">
185
189
186
190
To use fimeval in Google Colab, follow the steps below:
187
191
188
-
## Upload Files
192
+
####Upload Files
189
193
Upload all necessary input files (e.g., raster, shapefiles, model outputs) to your Google Drive.
190
-
## Open Google Colab
194
+
####Open Google Colab
191
195
Go to Google Colab and sign in with a valid Google account.
192
-
## Mount Google Drive
196
+
####Mount Google Drive
193
197
In a new Colab notebook, mount the Google Drive
194
198
```bash
195
199
pip install fimeval
196
200
```
197
-
### **Acknowledgements**
201
+
### Citing our work
202
+
<hrstyle="border: 1pxsolidblack; margin: 0;">
203
+
204
+
- Devi, D., Dipsikha, Supath Dhital, Dinuke Munasinghe, Sagy Cohen, Anupal Baruah, Yixian Chen, Dan Tian, & Carson Pruitt (2025).
205
+
*A framework for the evaluation of flood inundation predictions over extensive benchmark databases.*
206
+
**Environmental Modelling & Software**, 106786.
207
+
https://doi.org/10.1016/j.envsoft.2025.106786
208
+
209
+
- Cohen, S., Baruah, A., Nikrou, P., Tian, D., & Liu, H. (2025).
210
+
*Toward robust evaluations of flood inundation predictions using remote sensing–derived benchmark maps.*
211
+
**Water Resources Research**, 61(8).
212
+
https://doi.org/10.1029/2024WR039574
213
+
214
+
### Acknowledgements
215
+
<hrstyle="border: 1pxsolidblack; margin: 0;">
216
+
198
217
|||
199
218
| --- | --- |
200
219
|  | Funding for this project was provided by the National Oceanic & Atmospheric Administration (NOAA), awarded to the Cooperative Institute for Research to Operations in Hydrology (CIROH) through the NOAA Cooperative Agreement with The University of Alabama.
0 commit comments