I'm trying to reproduce the results of the paper and I've downloaded the datasets using the download_datasets.sh script and preprocessed them usig the process_datasets.sh script. However, I encountered two erros during the preprocessing phase.
Below, the error output that happens for both, CLOCS.py and Mixing-up.py python scripts, called from process_datasets.sh:
Traceback (most recent call last):
File "/workspaces/hiaac-m4/TFC-pretraining/data_processing/Mixing-up.py", line 8, in <module>
train_dict = torch.load(os.path.join('datasets', dataset_name, 'train.pt'))
File "/home/vscode/.local/lib/python3.10/site-packages/torch/serialization.py", line 988, in load
with _open_file_like(f, 'rb') as opened_file:
File "/home/vscode/.local/lib/python3.10/site-packages/torch/serialization.py", line 437, in _open_file_like
return _open_file(name_or_buffer, mode)
File "/home/vscode/.local/lib/python3.10/site-packages/torch/serialization.py", line 418, in __init__
super().__init__(open(name, mode))
FileNotFoundError: [Errno 2] No such file or directory: 'datasets/ecg/train.pt'
Below, the error output that happens when exeuting the SimCLR.py from process_datasets.sh script:
Traceback (most recent call last):
File "/workspaces/hiaac-m4/TFC-pretraining/data_processing/SimCLR.py", line 79, in <module>
scatter_numpy(train_y, 1, np.expand_dims(train_dict['labels'].numpy().astype(int), axis=1), 1)
File "/workspaces/hiaac-m4/TFC-pretraining/data_processing/SimCLR.py", line 41, in scatter_numpy
idx = [[*np.indices(idx_xsection_shape).reshape(index.ndim - 1, -1),
File "/workspaces/hiaac-m4/TFC-pretraining/data_processing/SimCLR.py", line 42, in <listcomp>
index[make_slice(index, dim, i)].reshape(1, -1)[0]] for i in range(index.shape[dim])]
IndexError: only integers, slices (`:`), ellipsis (`...`), numpy.newaxis (`None`) and integer or boolean arrays are valid indices
NOTE: The first error happens in case-sensitive file-systems, as the datasets are downloaded with upper-case letters and the scripts process them using lower-case letters.
Hello.
DESCRIBE THE BUG
I'm trying to reproduce the results of the paper and I've downloaded the datasets using the
download_datasets.sh
script and preprocessed them usig theprocess_datasets.sh
script. However, I encountered two erros during the preprocessing phase.CLOCS.py
andMixing-up.py
python scripts, called fromprocess_datasets.sh
:SimCLR.py
fromprocess_datasets.sh
script:NOTE: The first error happens in case-sensitive file-systems, as the datasets are downloaded with upper-case letters and the scripts process them using lower-case letters.
SYSTEM SPECIFICATION