Why are torch.mm and torch.bmm deleted in pytorch_ops.py

[wzk1015]

When I use calculate_flops to calculate flops of a local model (e.g. openai/clip-vit-large-patch14-336 downloaded locally), the result is smaller than the FLOPs calculated manually (use the flops calculation from mmcv and add the self-attention flops).

I found that the modeling_clip.py in the huggingface transformers package use torch.bmm to calculate attention weights, but the torch.mm and torch.bmm are commented in _patch_tensor_methods in pytorch_ops.py. When I uncomment these lines, the result is consistent with my manual calculation.

Are there any reasons why they are deleted? From commit 4bcf54c it seems to fix the bug for calculate_flops_hf, but I suspect this introduces bugs to calculate_flops.

Maybe related to #26

[MrYxJ]

谢谢，MrYx已经收到您的邮件!

[ChangeFWorld]

谢谢，MrYx已经收到您的邮件!

encounter the same bug. Any update?