search

dbolya / yolact

A simple, fully convolutional model for real-time instance segmentation.
MIT License
5.01k stars 1.32k forks source link

eval.py raising error when trying to use the custom-trained weights #501

Open BishwaBS opened 4 years ago

BishwaBS commented 4 years ago

I got this error. Any idea what this is about?

I used yolact, not yolact ++. It's weird that the weight file saved with name "yolact_plus", instead of just "yolact".

!python eval.py --trained_model=./weights/yolact_plus_resnet50_cottweeds_8_287_interrupt.pth --config=yolact_resnet50_cottweeds_config --score_threshold=0.3 --top_k=15 --images=test_images:output_images

Loading model...Traceback (most recent call last): File "eval.py", line 1098, in net.load_weights(args.trained_model) File "/home/ubuntu/anaconda3/yolact/yolact.py", line 479, in load_weights state_dict = torch.load(path) File "/home/ubuntu/anaconda3/envs/tensorflow2_latest_p37/lib/python3.7/site-packages/torch/serialization.py", line 529, in load return _legacy_load(opened_file, map_location, pickle_module, **pickle_load_args) File "/home/ubuntu/anaconda3/envs/tensorflow2_latest_p37/lib/python3.7/site-packages/torch/serialization.py", line 709, in _legacy_load deserialized_objects[key]._set_from_file(f, offset, f_should_read_directly) RuntimeError: unexpected EOF, expected 434506 more bytes. The file might be corrupted. Segmentation fault (core dumped)

ws312 commented 3 years ago

I met the same situation, have you solved it?