search

okankop / MFF-pytorch

Motion Fused Frames implementation in PyTorch, codes and pretrained models.
Other
131 stars 33 forks source link

input specifities #7

Closed Omenranr closed 5 years ago

Omenranr commented 5 years ago

Hello, I'm trying to feed the pretrained model to get prediction, but I have some problems with data preprocessing. I'm wondering what is the input specifities: size, how much frames, type ?

the test_videos.py file in TRN repository gives: torch.Size([1, 80, 3, 224, 224])

okankop commented 5 years ago

Can you please write your question more clearly? If you look at the code for validation, you would see how a video is preprocessed to get its prediction.

Omenranr commented 5 years ago

ok thank you