I changed the categories and set num_classes=9, on line 237 of the detectron2/modeling/roi_heads/fast_rcnn.py, the number of output layers is num_classes+1 when calculates cls_score, and a "/pytorch/aten/src/ATen/native/cuda/IndexKernel.cu:142: operator(): block: [0,0,0], thred: [75,0,0] Assertion index >= -size[i] && index < size[i] && "index out of bounds" failed." error occurs. What's the matter?
I changed the categories and set num_classes=9, on line 237 of the detectron2/modeling/roi_heads/fast_rcnn.py, the number of output layers is num_classes+1 when calculates cls_score, and a "/pytorch/aten/src/ATen/native/cuda/IndexKernel.cu:142: operator(): block: [0,0,0], thred: [75,0,0] Assertion
index >= -size[i] && index < size[i] && "index out of bounds"failed." error occurs. What's the matter?