Calculated!{"precision": 0, "recall": 0.0, "hmean": 0, "AP": 0}

MhLiao / MaskTextSpotter

A PyTorch implementation of Mask TextSpotter

https://github.com/MhLiao/MaskTextSpotter

412 stars 94 forks source link

Calculated!{"precision": 0, "recall": 0.0, "hmean": 0, "AP": 0} #68

Open zhongdajian opened 4 years ago

zhongdajian commented 4 years ago

i have run the command "python -m torch.distributed.launch --nproc_per_node=1 test_net.py ". In the config file, i used the download “model_finetune.pth” and test it on icdar_test dataset. The output images seems to be have the right results, but the precision is 0. I don't know why.

MhLiao commented 4 years ago

@zhongdajian I think you should not use the distributed mode for testing. Generally, it is only used in training.

zhongdajian commented 4 years ago

@MhLiao Sorry to bother you again! When i run the test_net.py with the model_finetune.pth, there comes an error: MaskTextSpotter/maskrcnn_benchmark/engine/text_inference.py", line 232, in process_char_mask box = list(boxes[index]) IndexError: index 1 is out of bounds for axis 0 with size 1 Could you help me?