HRNet / HRNet-MaskRCNN-Benchmark

Object detection with multi-level representations generated from deep high-resolution representation learning (HRNetV2h).
MIT License
138 stars 37 forks source link

coco data load error #2

Open stigma0617 opened 5 years ago

stigma0617 commented 5 years ago

Hi, to test your HRNet, I change paths_catalog.py to original maskrcnn-benchmark one because it is hard-coded by your path.

And then using this command, this error occurs.

CUDA_VISIBLE_DEVICES=0, python tools/test_net.py --config-file "configs/hrnet/e2e_faster_rcnn_hrnet_w48_2x.yaml" MODEL.WEIGHT FasterR-CNN-HR48-2x.pth TEST.IMS_PER_BATCH 1

2019-08-07 10:55:33,655 maskrcnn_benchmark.inference INFO: Start evaluation on coco_2017_val dataset(5000 images). 0%| | 0/5000 [00:00<?, ?it/s$ zip file 'datasets/coco/val2017.zip' is not found zip file 'datasets/coco/val2017.zip' is not found zip file 'datasets/coco/val2017.zip' is not found zip file 'datasets/coco/val2017.zip' is not found zip file 'datasets/coco/val2017.zip' is not found zip file 'datasets/coco/val2017.zip' is not found Traceback (most recent call last): File "tools/test_net.py", line 258, in zip file 'datasets/coco/val2017.zip' is not found main() File "tools/test_net.py", line 217, in main output_folder=output_folder, File "/home/lsrock1/HRNet-MaskRCNN-Benchmark/maskrcnn_benchmark/engine/inference.py", line 76, in inference predictions = compute_on_dataset(model, data_loader, device) File "/home/lsrock1/HRNet-MaskRCNN-Benchmark/maskrcnn_benchmark/engine/inference.py", line 20, in compute_on_dataset for i, batch in enumerate(tqdm(data_loader)): File "/home/lsrock1/anaconda3/envs/pytorch1.1/lib/python3.7/site-packages/tqdm/_tqdm.py", line 1022, in iter zip file 'datasets/coco/val2017.zip' is not found for obj in iterable: File "/home/lsrock1/anaconda3/envs/pytorch1.1/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 582, in next zip file 'datasets/coco/val2017.zip' is not found return self._process_next_batch(batch) File "/home/lsrock1/anaconda3/envs/pytorch1.1/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 608, in _process_next_batch raise batch.exc_type(batch.exc_msg) AssertionError: Traceback (most recent call last): File "/home/lsrock1/anaconda3/envs/pytorch1.1/lib/python3.7/site-packages/torch/utils/data/_utils/worker.py", line 99, in _worker_loop samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/lsrock1/anaconda3/envs/pytorch1.1/lib/python3.7/site-packages/torch/utils/data/_utils/worker.py", line 99, in samples = collate_fn([dataset[i] for i in batch_indices]) File "/home/lsrock1/HRNet-MaskRCNN-Benchmark/maskrcnn_benchmark/data/datasets/coco.py", line 56, in getitem cv2.IMREAD_COLOR | cv2.IMREAD_IGNORE_ORIENTATION) File "/home/lsrock1/HRNet-MaskRCNN-Benchmark/maskrcnn_benchmark/utils/zipreader.py", line 32, in imread assert 0 AssertionError

Should I prepare .zip file?

In origin maskrcnn-benchmark, they only need images/annotations folders.

I can't find your coco dataset setting instructions.

wondervictor commented 4 years ago

Because zip file 'datasets/coco/val2017.zip' is not found no zip file. you can directly use unzipped format. see path_catalog.py and change datasets coco_val2017