반응형
# Problem
When I execute NVDS repo for video mono depth estimation
$ CUDA_VISIBLE_DEVICES=1 python infer_NVDS_midas_bi.py --base_dir ./demo_outputs/dpt_init/ --vnum con2 --infer_w 1920 --infer_h 1080
I got a below Error:
urllib.error.HTTPError: HTTP Error 403: rate limit exceeded
# Solution
This is because we are using GitHub API to check the existence of the repo. I have submitted a PR to ignore HTTP error 403. To work around this error temporarily, you can delete line 163 on torch/hub.py
_validate_not_a_forked_repo(repo_owner, repo_name, branch)
# Reference
[Github Issue] HTTP Error 403: rate limit exceeded when loading model : https://github.com/pytorch/vision/issues/4156
반응형