faster_rcnn_pytorch

faster_rcnn_pytorch copied to clipboard

why multiple cpus are used in .cuda() mode

8

Hello, I migrate the model to gpu. however it still uses cpu. any idea why that is the case? thanks!

volkancirik

There maybe a potential bug in [__init__.py](https://github.com/longcw/faster_rcnn_pytorch/blob/d8f842dfa51e067105e6949999277a08daa3d743/faster_rcnn/datasets/__init__.py#L37C4-L37C4) due to the usage of `path.strip('"')`. The code in the `which()` function is similar to the code in [version 6 of this Stack...

Alfusainey