Perform a inference with pretrained model but stuck in processing and loop semaphore_tracker warning after loading model, anyone has idea what's going on?

[yang074]

(pytorch_gpu) [stu3@gpu7 AlphaPose-master]$ python scripts/demo_inference.py --cfg configs/coco/resnet/256x192_res50_lr1e-3_1x.yaml --checkpoint pretrained_models/fast_res50_256x192.pth --indir data/input/03_0032 --outdir data/output --gpus 1 Loading YOLO model.. Loading pose model from pretrained_models/fast_res50_256x192.pth... Network successfully loaded 0%| | 0/409 [00:00<?, ?it/s] /public/home/stu3/anaconda3/envs/pytorch_gpu/lib/python3.6/multiprocessing/semaphore_tracker.py:143: UserWarning: semaphore_tracker: There appear to be 1 leaked semaphores to clean up at shutdown len(cache)) /public/home/stu/anaconda3/envs/pytorch_gpu/lib/python3.6/multiprocessing/semaphore_tracker.py:143: UserWarning: semaphore_tracker: There appear to be 1 leaked semaphores to clean up at shutdown len(cache))

[Fang-Haoshu]

Hi, can you see if this help: https://discuss.pytorch.org/t/issue-with-multiprocessing-semaphore-tracking/22943/6 ?

[yang074]

Thks, I had seen this help before, and I tried to:

1. wrapped the codes which in demo_inference.py inside if name == "main": but still had the problem;
2. cmd with -W ignore and this had no warning,but still stuck in processing

[Fang-Haoshu]

It's pretty weird. Does enable --sp flag help?

[wu-yue-yu]

I run into this on my jetson nano 4GB.

[atpug22]

I am too getting this error. @yang074 did you find any solution?