search

LCFractal / AIC21-MTMC

🏆The 1st place solution of track3 (City-Scale Multi-Camera Vehicle Tracking) in the NVIDIA AI City Challenge at CVPR 2021 Workshop.
MIT License
128 stars 35 forks source link

When run_all.sh, always cannot find 'c041_dets_feat.pkl' file. #13

Open wanglu362 opened 2 years ago

wanglu362 commented 2 years ago

Whenever I run the whole process with run_all.sh, I'm always caught by this error and the program stops:

Traceback (most recent call last): File "merge_reid_feat.py", line 59, in main() File "merge_reid_feat.py", line 55, in main merge_feat(cfg) File "merge_reid_feat.py", line 27, in merge_feat feat_mode_dic = pickle.load(open(feat_pkl_file, 'rb')) FileNotFoundError: [Errno 2] No such file or directory: '/opt/Program/PycharmProjects/AIC21-MTMC/datasets/detect_reid1/c041/c041_dets_feat.pkl'

Checking the detect_provided path, there is one such file. Is this caused by a failure to extract reid features? The output of extract_image_feat.py is:

51370 images for c041 51370 images in total init reid model abs path is /media/lu/71274469-3e57-4411-bd33-12b766d409a4/opt/Program/PycharmProjects/AIC21-MTMC/reid/reid_inference/reid_model.py 0%| | 0/803 [00:00<?, ?it/s]===========ResNet=========== using resnet101_ibn_a as a backbone Loading pretrained model from reid_model/resnet101_ibn_a_2.pth /opt/Softwares/anaconda3/envs/track/lib/python3.8/site-packages/torchvision/transforms/transforms.py:280: UserWarning: Argument interpolation should be of type InterpolationMode instead of int. Please, use InterpolationMode enum. warnings.warn( /opt/Softwares/anaconda3/envs/track/lib/python3.8/site-packages/torch/nn/functional.py:718: UserWarning: Named tensors and all their associated APIs are an experimental feature and subject to change. Please do not use them for anything important until they are released as stable. (Triggered internally at /opt/conda/conda-bld/pytorch_1623448278899/work/c10/core/TensorImpl.h:1156.) return torch.max_pool2d(input, kernel_size, stride, padding, dilation, ceil_mode) 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 803/803 [13:41<00:00, 1.02s/it] 821.8415 s

It seems successful. But how does it fail to save a dets_feat.pkl file?

LCFractal commented 2 years ago

Maybe you should check whether the feature dets_feat.pkl is output in the path under aic_reid1.yml: DATA_DIR

wanglu362 commented 2 years ago

Maybe you should check whether the feature dets_feat.pkl is output in the path under aic_reid1.yml: DATA_DIR

Thanks and it works!

cmtsai commented 2 years ago

Do you know how to obtain or produce 'test_cluster.pkl' ?