Davide Coccomini

Did you find the weights?

Hello, I am afraid that this issue is related to the used OS and I have no experience in running it on MAC. Could you emulate a Linux environment?

Hello, you can edit the patch size but you need to do it properly. The EfficientNet returns images with different shapes depending on the block: 1 torch.Size([64, 24, 56, 56])...

> Thanks for clearing that up. I have tried every single popular software on github to perform this task and none has worked thus far (some work for individual files,...

What do you mean by "moving face images"?

Hi, which version of the model are you using?

The EfficientViT obtain an AUC of 0.919 which is not too far from yours. If I understand well, your training set is not complete so it is normal to obtain...

This part has been hugely inspired by Selim Seferbekov work: https://github.com/selimsef/dfdc_deepfake_challenge/blob/master/preprocessing/face_detector.py We maintained the frame resize to be totally comparable, anyway, we think that it would not affect the result....

Can you provide the full stack trace?

Can you check inside the folder cross-efficient-vit/efficient_net/efficientnet_pytorch/model.py if you have the delete_blocks method?