search

IDEA-Research / TAPTR

[ECCV 2024] Official implementation of the paper "TAPTR: Tracking Any Point with Transformers as Detection"
https://taptr.github.io
Other
196 stars 12 forks source link

tracking point demo error #5

Closed jiangziben closed 3 months ago

jiangziben commented 3 months ago

i test tracking point demo in 2080Ti,but get error info below, could you help me how to fix this problem?

torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 1.99 GiB (GPU 0; 10.75 GiB total capacity; 8.28 GiB already allocated; 1.59 GiB free; 8.89 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation.  See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
LHY-HongyangLi commented 3 months ago

Hy @jiangziben , thanks for your attention. This error is due to the memory of your GPU is smaller than the requirement. You can refer to this answer for help temporally: https://github.com/IDEA-Research/TAPTR/issues/4#issuecomment-2262200153

LHY-HongyangLi commented 3 months ago

Hi @jiangziben , This issue has been fixed, please refer to this https://github.com/IDEA-Research/TAPTR/issues/4#issuecomment-2264277331

LHY-HongyangLi commented 3 months ago

I will close this issue, if you have any questions, feel free to reopen this issue.