np.asarray() Errno 21 IsADirectoryError

[ghost]

Hi,

I have downloaded the datasets and placed them in to the "data" folder, ensuring the names matched as per segmentation/header.py. However, in attempting to run segmentation/train.py for the JSRT dataset I get the following error:

[...]/segmentation$ python3 train.py

train.py


>> Generate network - v1.1

>> Load data
  >>> Total # of trainset : 7
  >>> Total # of valset : 1

>> Train Network - [1 epoch]
Traceback (most recent call last):
  File "train.py", line 208, in <module>
    main()
  File "train.py", line 111, in main
    for i, data in enumerate(trainloader, 0):
  File "/[...]/anaconda3/envs/Oh_et_al_2020/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 345, in __next__
    data = self._next_data()
  File "/[...]/anaconda3/envs/Oh_et_al_2020/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 385, in _next_data
    data = self._dataset_fetcher.fetch(index)  # may raise StopIteration
  File "/[...]/anaconda3/envs/Oh_et_al_2020/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in fetch
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "/[...]/anaconda3/envs/Oh_et_al_2020/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in <listcomp>
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "/[...]/Deep-Learning-COVID-19-on-CXR-using-Limited-Training-Data-Sets/segmentation/mydataset.py", line 60, in __getitem__
    images = np.asarray(np.reshape(np.fromfile(self.images[index], dtype='>u2', sep="", ), (header.orig_height, header.orig_width)))
IsADirectoryError: [Errno 21] Is a directory: '../data/JSRT/Clinical_Information'

Clearly, you cannot use np.asarray() on a directory

I found where this issue is and did a rough workaround: MyTrainDataset.__init__() assumes any items returned by glob(image_path + /*) are image files and appends them to the file image list. Will do a pull request for this one.