Open C1nDeRainBo0M opened 1 year ago
I use python=3.6, torch=1.9.0, torchvision = 0.10.0, torchmeta=1.8.
I changed if os.path.isfile(fpath) and check_integrity(fpath, md5):
in torchmeta\datasets\utils.py toif os.path.isfile(fpath):
, then the program runs normally and generates subsequent files. The current output tensor is the correct shape. Is this the correct way to handle it?
I don't find def check_integrity():
in utils.py, is there something wrong with my version? Because I downloaded mini-imagenet.tar.gz
manually, is it because this affects the correct operation of the program?
I use google drive to download and finish.
dataset = miniimagenet(..., meta_train=True, download=True)
and i got,
NameError: name 'check_integrity' is not defined
how should i fix this error. thx.