search

KumapowerLIU / Rethinking-Inpainting-MEDFE

Rethinking Image Inpainting via a Mutual Encoder Decoder with Feature Equalizations. ECCV 2020 Oral
Other
371 stars 50 forks source link

The expanded size of the tensor (313) must match the existing size (256) at non-singleton dimension 2. #32

Open Jamie-61 opened 2 years ago

Jamie-61 commented 2 years ago

I had some problems when remerging your codes: RuntimeError: The expanded size of the tensor (313) must match the existing size (256) at non-singleton dimension 2. Target sizes: [1, 1, 313, 256]. Tensor sizes: [1, 1, 256, 256]

Yaque commented 2 years ago

I change this code in the dataprocess. transforms.Resize(opt.fineSize) transforms.Resize((opt.fineSize, opt.fineSize)) it works. my env: torch-1.8.1+cu102 torchaudio-0.8.1 torchtext-0.9.1 torchvision-0.9.1+cu102 image

camellia120 commented 2 years ago

I am very glad to hear from you.Thank you very much!

Jamie-61 commented 2 years ago

Thank you very very very much!!! I have solved it!!!