About download link authorization

[liushangke]

Dear author,

I'm doing OOD-related research and encountered a little problem when I try to download your datasets through scripts. Here is the error message. Could you please help me with the download link access authorization? Thank you.

Best,

Shangke Northwestern University

Flags: data_path: /Users/shangke/Desktop/ideas_lab/WOODS/path/to/data dataset: ['PCL'] mode: gdrive Access denied with the following error: Cannot retrieve the public link of the file. You may need to change the permission to 'Anyone with the link', or have had many accesses. You may still be able to access the file from the browser: https://drive.google.com/uc?id=118DNxHpzeJwVTM22wzZhSiOuDsno0nay Traceback (most recent call last): File "/opt/anaconda3/envs/d2l/lib/python3.8/runpy.py", line 194, in _run_module_as_main return _run_code(code, main_globals, None, File "/opt/anaconda3/envs/d2l/lib/python3.8/runpy.py", line 87, in _run_code exec(code, run_globals) File "/Users/shangke/Desktop/ideas_lab/WOODS/woods/scripts/download_datasets.py", line 32, in download_pcl(flags.data_path, flags.mode) File "/Users/shangke/Desktop/ideas_lab/WOODS/woods/download.py", line 90, in download_pcl download_gdrive(archive_url, path, 'pcl.zip') File "/Users/shangke/Desktop/ideas_lab/WOODS/woods/download.py", line 16, in download_gdrive with zipfile.ZipFile(r, 'r') as zip_ref: File "/opt/anaconda3/envs/d2l/lib/python3.8/zipfile.py", line 1269, in init self._RealGetContents() File "/opt/anaconda3/envs/d2l/lib/python3.8/zipfile.py", line 1332, in _RealGetContents endrec = _EndRecData(fp) File "/opt/anaconda3/envs/d2l/lib/python3.8/zipfile.py", line 264, in _EndRecData fpin.seek(0, 2) AttributeError: 'NoneType' object has no attribute 'seek'

[jc-audet]

hmm yeah looks like there is some issue with authorization. I will try to fix it, in the meantime you can try and use the academic torrent option of the download script!

[liushangke]

Thank you. And just to add one point, seem that the training script doesn't work with the manually downloaded dataset. Somehow it can't handle extracted .h5 file. Looking forward to your update!

Best,