Where is the implementation of torchvision roi_align?

I can’t find anywhere the implementation of this code:

may be this is what you are looking

@talhaanwarch so the method is basically calling itself? Line 61 is calling line 13?

The CPU kernel can be found here the CUDA kernel here.

Thank you @ptrblck ! That responds my question since I was interested in learning about how the ROI Align algorith was implemented since I was not able to understand just reading the Mask R CNN Paper.
If you don’t mind I’d like to ask another question now. Is there some documentation explaining how this cpp code is loaded into the python side? When I call torch.ops.torchvision.roi_align, how does that load the cpp kernel that you sent above?
Thank you!