How to start the work ?

[chengjianhong]

Hi, I 'm interested in your works. But I 'm wondering how to run this training?

Best wish,
Jianhong Cheng

[MinaRe]

Hi, You need to install python 3.6 Keras 2.2+and Tensorflow 1.11+ then simply run python train.py. Let me know if you still have a problem.

[chengjianhong]

@MinaRe Yes, I run python train.py directly, it happens following error. What is more, I do not find the location where configures the BraTS 2018 training data.

Epoch 46
Epoch 47
Epoch 48
Epoch 49
Traceback (most recent call last):
File "train.py", line 88, in
tensorlog.on_train_end()
TypeError: on_train_end() missing 1 required positional argument: '_'

[chengjianhong]

I find the data_dir = 'data/train/**/*t1.nii.gz'. and I wonder that the other sequences(flair, t1ce, t2 ) don't participate in the training?

[MinaRe]

You have to update config.py with your data configuration, for example, replace your training data folder direction in "data_dir =" !
voxel-GAN takes multi-modalities, please take a look utils.py :)

[chengjianhong]

@MinaRe Thanks for your reply. I update the data_dir = '/root/userfolder/data/MICCAI_BraTS_2018_Data_Training/*/*/*t1.nii.gz'. And running the train.py ,it still happens the following errors. I also still notice the Brats18_TCIA05_444_1_seg0.nii.gz that not exists in my training data. and in utils.py shows that t1.nii.gz replace with seg0.nii.gz. But why is this happening? Am I doing something wrong in my configuration ?

Epoch 0
  1/256 [..............................] - ETA: 0sTraceback (most recent call last):
  File "/root/anaconda3/envs/py35/lib/python3.5/site-packages/nibabel/loadsave.py", line 40, in load
    stat_result = os.stat(filename)
FileNotFoundError: [Errno 2] No such file or directory: '/root/userfolder/data/MICCAI_BraTS_2018_Data_Training/HGG/Brats18_TCIA05_444_1/Brats18_TCIA05_444_1_seg0.nii.gz'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "train.py", line 57, in <module>
    x, y, idx = load(sequence, size)
  File "/root/userfolder/workspace/voxelGAN/utils.py", line 32, in load
    seg0 = nib.load(patient.replace("t1.nii.gz", "seg0.nii.gz")).get_data().astype(np.float)
  File "/root/anaconda3/envs/py35/lib/python3.5/site-packages/nibabel/loadsave.py", line 42, in load
    raise FileNotFoundError("No such file or no access: '%s'" % filename)
FileNotFoundError: No such file or no access: '/root/userfolder/data/MICCAI_BraTS_2018_Data_Training/HGG/Brats18_TCIA05_444_1/Brats18_TCIA05_444_1_seg0.nii.gz'

[chengjianhong]

I change the seg0.nii.gz, seg1.nii.gz, seg2.nii.gz , gz,seg3.nii.gz into seg.nii.gz in utils.py , and fix it.

[MinaRe]

the segXX.nii.gz is considered regarding internal bias correction which you can modify them with the corresponding segmentation label.

[chengjianhong]

How to make it about the four segmentation labels?