search

Johnathan-Xie / ZSD-YOLO

GNU General Public License v3.0
50 stars 2 forks source link

Hello, I downloaded your Kaggle data and encountered an error using test.py. #6

Closed Lz-Sadada closed 1 year ago

Lz-Sadada commented 1 year ago

Fusing layers... torch.Size([17, 512]) Model Summary: 226 layers, 8432590 parameters, 0 gradients, 20.7 GFLOPS <class 'utils.datasets.LoadImagesAndLabels'> val: Scanning 'data\coco\labels2014_zsd_self_test_l_48_17\val2014' images and labels... 0 found, 2729 missing, 0 empty, 0 corrupted: 100%|██████████| 2729/2729 [00:00<00:00, 8621.79it/s] val: WARNING: No labels found in data\coco\labels2014_zsd_self_test_l_48_17\val2014.cache. See https://github.com/ultralytics/yolov5/wiki/Train-Custom-Data val: New cache created: data\coco\labels2014_zsd_self_test_l_48_17\val2014.cache Class Images Labels P R mAP@.5 mAP@.5:.95: 100%|██████████| 86/86 [00:19<00:00, 4.40it/s] NO STATS Traceback (most recent call last): File "E:\pycharm-community-2022.3.2\PyCharm-Community-Edition-2022.3.2\code\ZSD-YOLO-main\test.py", line 494, in test(opt.data, File "E:\Anaconda3-2022.10\envs\ZSD-YOLO\lib\site-packages\torch\autograd\grad_mode.py", line 26, in decorate_context return func(*args, **kwargs) File "E:\pycharm-community-2022.3.2\PyCharm-Community-Edition-2022.3.2\code\ZSD-YOLO-main\test.py", line 382, in test recall_info = process_recall(zsd_recall_correct, zsd_recall_classes, nt) File "E:\pycharm-community-2022.3.2\PyCharm-Community-Edition-2022.3.2\code\ZSD-YOLO-main\test.py", line 51, in process_recall zsd_recall_correct, zsd_recall_classes = torch.stack(zsd_recall_correct), torch.Tensor(zsd_recall_classes).type(torch.LongTensor) RuntimeError: stack expects a non-empty TensorList

Johnathan-Xie commented 1 year ago

Could you share the information on your conda environment? It is possible the packages are not matching because this work was originally done over a year ago.