UserWarning: indexing with dtype torch.uint8 is now deprecated, please use a dtype torch.bool instead.

[onefish51]

when I train yolo v3 on coco, it WARNING that "/pytorch/aten/src/ATen/native/IndexingUtils.h:20: UserWarning: indexing with dtype torch.uint8 is now deprecated, please use a dtype torch.bool instead."

It's so many! How to fix it?

[onefish51]

[HG1998-JM]

torch1.2.0--->torch1.1.0 torchvision=0.2.0

[runningJ]

I met the same problem. Did you fixed it? @onefish51

[paojianghu]

I met the same problem. Did you fixed it? @onefish51

https://github.com/eriklindernoren/PyTorch-YOLOv3/blob/master/models.py#L191 obj_mask=obj_mask.bool() # convert int8 to bool noobj_mask=noobj_mask.bool() #convert int8 to bool

[iosfixed]

Replace in utils:269

ByteTensor = torch.cuda.ByteTensor if pred_boxes.is_cuda else torch.ByteTensor

With: BoolTensor = torch.cuda.BoolTensor if pred_boxes.is_cuda else torch.BoolTensor

And it usage in lines 278, 279 obj_mask = BoolTensor(nB, nA, nG, nG).fill_(0) noobj_mask = BoolTensor(nB, nA, nG, nG).fill_(1)

[ujsyehao]

@paojianghu You should replace obj_mask=obj_mask.bool() # convert int8 to bool

noobj_mask=noobj_mask.bool() #convert int8 to bool

in utils/utils.py after line 279