sneh-debug

Decoder loss and dice coefficient is 0 even after i trained it for 90 epochs?

2

Epoch 90/90 1/2 [==============>...............] - ETA: 15s - loss: 0.0271 - Dec_GT_Output_loss: 0.0000e+00 - Dec_VAE_Output_loss: 0.0271 - Dec_GT_Output_d2/2 [==============================] - 30s 15s/step - loss: 0.0282 - Dec_GT_Output_loss: 0.0000e+00 - Dec_VAE_Output_loss:...

Also in custom_generator script. I used on_epoch_end(self) rather than self.on_epoch_end(). Please suggest what can be the reasons?

@fitushar please let me know if this can be applied for multiple modalities .nii.gz ?

for preprocessing , in preprocess.py script only changes in def convert_brats_data(brats_folder, out_folder, bias_correct=True, overwrite=True, no_bias_correction_modalities=("flair",)): need to be done?

1

def convert_brats_data(brats_folder, out_folder, bias_correct=True, overwrite=True, no_bias_correction_modalities=("flair",)): for subject_folder in tqdm(glob.glob(os.path.join("data", "*", "*"))): # continue if os.path.isdir(subject_folder): subject = os.path.basename(subject_folder) new_subject_folder = os.path.join(preprocessed, os.path.basename(os.path.dirname(subject_folder)), subject) if not os.path.exists(new_subject_folder) or overwrite: if...

demo_task1, running demo.py script

1

i created folders in **Data** folder which include original and preprocessed. And then i ran demo.py i got this output: wcclab@wcclab-HP-280-Pro-G4-SFF-Business-PC:~/brats_2019-master/demo_task1$ python3 demo.py /home/wcclab/.local/lib/python3.5/site-packages/dicom/__init__.py:53: UserWarning: This code is using an...

difference between dice coefficient and weighted dice coefficient

2

why in weighted dice coefficient axis=[-3, -2, -1] ?? ?