mikel-brostrom/Yolov7_StrongSORT_OSNet

no support for cuda? can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.

akashAD98 opened this issue · 3 comments

akashAD98 commented

Search before asking

  • I have searched the Yolov7_StrongSORT_OSNet issues and discussions and found no similar questions.

Yolov7_StrongSORT_OSNet Component

No response

Bug

i saw there is same issue in github issue, but still this isssue is not solved,im running it on google collab, code working firne for cpu but not on gpu-cuda @mikel-brostrom

Environment

google collab

TaplierShiru commented

Hi!
Face same issue here, it seems that there is an error in yolo7 (dataset.py) and in the track.py scripts. I suggest pull request for each repo to fix it, after them repo work properly.
For anyone as temporal solution I have Colab where repo is cloned from my profile, and all seems work as intended.

mikel-brostrom commented

I have updated the repo with your PRs @TaplierShiru. Could you pleas double check that this is solved now?

TaplierShiru commented

I have updated the repo with your PRs @TaplierShiru. Could you pleas double check that this is solved now?

Checked current work of the repo - all seems good, so I think this issue is resolved.

  • 🚀1