ValueError: num_samples should be a positive integer value, but got num_samples=0

[Argha Dhar]

I am using CoPINET on my Colaboratory . while running
!python /content/CoPINet/src/main.py train --dataset "/content/drive/MyDrive/sample2/center_single"

the following error occurs
Traceback (most recent call last):
File "/content/CoPINet/src/main.py", line 262, in
main()
File "/content/CoPINet/src/main.py", line 253, in main
train(args, device)
File "/content/CoPINet/src/main.py", line 122, in train
train_loader = DataLoader(train_set, batch_size=args.batch_size, shuffle=True, num_workers=args.num_workers)
File "/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py", line 268, in init
sampler = RandomSampler(dataset, generator=generator)
File "/usr/local/lib/python3.7/dist-packages/torch/utils/data/sampler.py", line 103, in init
"value, but got num_samples={}".format(self.num_samples))
ValueError: num_samples should be a positive integer value, but got num_samples=0

**I have mounted google drive on my Colab and this folder /content/drive/MyDrive/sample2/center_single contains 10 sample training data (RAVEN_0_train.npz , RAVEN_0_train.xml , ........... , RAVEN_9_train.npz , RAVEN_9_train.xml)

[Chi Zhang]

Sorry that I have never tried running the code on colab. But it looks like a data loading problem. Any possibility of trying a local version? The code works fine on my local device.

[Argha Dhar]

trying it on local device ? What could be the solution?

Thanks in advance

[Chi Zhang]

Try Python2? Looks like there are some compatibility issues with Python3.

[Argha Dhar]

now using python2 but the same error occurs....any other solutions?

[Chi Zhang]

Should set the dataset arg to the parent folder, not a specific split