proto_tower is stacked by 32 convolutions according to your code (fcos/modeling/fcos/fcos.py Line 166). Did you try less number convolutions? I have also recently combined yolact and fcos, but my accuracy is only 10
@Epiphqny hi thanks for the source code , i am trying to understand how you have integrated the FCOS into the yolact can you provide pointers on where the changes have been made since u build detectron i am not able to figure out . I am more interested in lincomb_mask_loss function which takes anchors into the consideration. Thanks in advance