search

ifzhang / ByteTrack

[ECCV 2022] ByteTrack: Multi-Object Tracking by Associating Every Detection Box
MIT License
4.64k stars 884 forks source link

Installation error on rtx 3090 and my solution #54

Open SnowPye opened 2 years ago

SnowPye commented 2 years ago

Dear author,

When I use RTX 3090, the PyTorch installation will come to an error with 'pip3 install -r requirements.txt'.

3090 is only adapted to the Cuda version above 11.0 while 'pip3 install -r requirements.txt' will bring a torch with Cuda version 10.2. This will cause problems in GPU calls.

Thus if u use RTX 3090, install torch and torchvision from the official website and then install other packages.

I suggest you add this hint to the readme.

Thanks for your excellent work.

ifzhang commented 2 years ago

Thank you very much!

OceanStarHit commented 2 years ago

Good suggestion that I have just seen. I have already tried those for cpu and rtx3080 before finding your suggestion, but your suggestion verifies my trials. In addition it's faster than 'pip3 install -r requirements.txt'. Thanks.