MaJ

README.md

@@ -25,7 +25,7 @@ To use the scripts with Raven annotation software ([Raven Lite](https://www.rave
 
 * go into the folder that contains the scripts 
 * Run get_train_annot.py
-* Launch training
+* Launch training following instructions
 
 <br>
 
@@ -179,7 +179,7 @@ This script splits data into training, validation, and optionally test sets base
 To run the script, use the following command:
 
 ```bash
-python get_train_val.py.py <path_to_data> <directory> [options]
+python get_train_val.py <path_to_data> <directory> [options]
 ```
 # Arguments
 ## Positional Arguments
@@ -220,6 +220,7 @@ python get_train_annot.py <filename_path> <path_to_data> <directory> [options]
 --hop (float): Ratio of hop in window (e.g., 50% = 0.5). Default is 0.5.
 --cpu (int): Number of CPUs to speed up the process. Default is 1.
 --test (flag): If provided, splits the data into train/test/validation sets with the ratio 1 - Ratio / 2 for test and the same for validation. If not provided, only train and validation splits are created.
+--minimum : If True, vmin will be stft.mean(), else stft.min().
 ```
 </details>
 <br />
@@ -278,7 +279,7 @@ python yolov5/train.py --imgsz <IMG_SIZE> --batch <BATCH_SIZE> --epochs <NB_EPOC
 - Detect on audio files
 
 ```bash
-python detect.py --weights yolov5/runs/train/<EXP_NB>/weights/best.pt --imgsz <imgsz> --conf <conf> --source <PATH_TO_FOLDER_THAT_CONTAIN_WAV> --save-txt --sound --rf <RF> --sampleDur <SampleDur> --window <window> --hop <hop> --low <low> --high <high> --cmap <cmap> --save-conf
+python detect.py --weights yolov5/runs/train/<EXP_NB>/weights/best.pt --imgsz <imgsz> --conf <conf> --source <PATH_TO_FOLDER_THAT_CONTAIN_WAV> --save-txt --sound --rf <RF> --sampleDur <SampleDur> --minimum <True/False --window <window> --hop <hop> --low <low> --high <high> --cmap <cmap> --save-conf
 ```
 
 <br />
@@ -286,7 +287,7 @@ python detect.py --weights yolov5/runs/train/<EXP_NB>/weights/best.pt --imgsz <i
 - Detect on audio files without saving the detection images
 
 ```bash
-python detect.py --weights yolov5/runs/train/<EXP_NB>/weights/best.pt --imgsz <imgsz> --conf <conf> --source <PATH_TO_FOLDER_THAT_CONTAIN_WAV> --save-txt --sound --rf <RF> --sampleDur <SampleDur> --window <window> --hop <hop> --low <low> --high <high> --cmap <cmap> --save-conf --nosave
+python detect.py --weights yolov5/runs/train/<EXP_NB>/weights/best.pt --imgsz <imgsz> --conf <conf> --source <PATH_TO_FOLDER_THAT_CONTAIN_WAV> --save-txt --sound --rf <RF> --sampleDur <SampleDur> --minimum <True/False> --window <window> --hop <hop> --low <low> --high <high> --cmap <cmap> --save-conf --nosave
 ```
 
 <br />
@@ -300,9 +301,10 @@ python detect.py --weights yolov5/runs/train/<EXP_NB>/weights/best.pt --imgsz <i
 --window: Window size for each spectrogram for detection. Default is 1024.
 --hop: Hop length for each spectrogram for detection. Default is 50% of window = 512.
 --sound: Enable sound. Default is False. Action 'store_true'.
---low: Low pass filter value
---high: High pass filter
---cmap: Colormap for the Spectrograms
+--low: Low pass filter value.
+--high: High pass filter.
+--cmap: Colormap for the Spectrograms.
+--minimum : If True, vmin will be stft.mean(), else stft.min().
 ```
 <br />
 

get_train_annot.py

@@ -183,7 +183,7 @@ if __name__ == '__main__':
     parser.add_argument('--test', action='store_const', const=1,
                         help='Split into train/test/val. 1 - Ratio / 2 for test and'
                         ' same for validation', default=None)
-    parser.add_argument('--minimum', type=int,
+    parser.add_argument('--minimum', type=str,
                         help='If True, vmin will be stft.mean(), else stft.min()', default=True)
     args = parser.parse_args()
 

requirements.txt

@@ -28,3 +28,5 @@ soundfile>=0.11.0    # Sound library
 p_tqdm==1.4.0        # Parallel tqdm (progress bar) for Python
 albumentations>=1.0.3 # Image augmentation library
 scikit-learn>=1.2.2  # Machine learning library
+xarray>=2024.3.0     # NetCDF interpreter
+h5netcdf>=1.3.0      # NetCDF interpreter