I’m trying to write a custom PyTorch FasterRCNN that’s based on an existing PyTorch model, fasterrcnn_resnet50_fpn, but I’m getting stuck on how to correctly write the forward pass. Following this recommendation, I’m basing the forward pass off of the GeneralizedRCNN forward pass, because I’d like to eventually modify the loss function. However, I’m getting errors on the forward pass formulation without loss function modifications, and not sure how to proceed.