search

ifzhang / FairMOT

[IJCV-2021] FairMOT: On the Fairness of Detection and Re-Identification in Multi-Object Tracking
MIT License
4.03k stars 934 forks source link

Error of running demo.py #535

Open zjutwt opened 1 year ago

zjutwt commented 1 year ago

After I configured the environment according to the author's instructions, I ran the program, but the following error occurred.

(test) C:\FairMOT-master-test\FairMOT-master\src>python demo.py mot --load_model ../models/fairmot_dla34.pth --conf_thres 0.4 --arch dla_34 Fix size testing. training chunk_sizes: [6, 6] The output will be saved to C:\FairMOT-master-test\FairMOT-master\src\lib....\exp\mot\default heads {'hm': 1, 'wh': 4, 'id': 128, 'reg': 2} 2023-03-12 22:27:52 [INFO]: Starting tracking... Lenth of the video: 1500 frames Creating model... loaded ../models/fairmot_dla34.pth, epoch 30 Drop parameter base.fc.weight.If you see this, your model does not fully load the pre-trained weight. Please make sure you have correctly specified --arch xxx or set the correct --num_classes for your own dataset. Drop parameter base.fc.bias.If you see this, your model does not fully load the pre-trained weight. Please make sure you have correctly specified --arch xxx or set the correct --num_classes for your own dataset. 2023-03-12 22:27:53 [INFO]: Processing frame 0 (100000.00 fps) 2023-03-12 22:27:56 [INFO]: Processing frame 20 (8.92 fps) 2023-03-12 22:27:58 [INFO]: Processing frame 40 (12.80 fps) 2023-03-12 22:28:00 [INFO]: Processing frame 60 (15.03 fps) 2023-03-12 22:28:01 [INFO]: Processing frame 80 (16.48 fps) 2023-03-12 22:28:03 [INFO]: Processing frame 100 (17.50 fps) 2023-03-12 22:28:05 [INFO]: Processing frame 120 (18.24 fps) 2023-03-12 22:28:07 [INFO]: Processing frame 140 (18.79 fps) 2023-03-12 22:28:08 [INFO]: Processing frame 160 (19.23 fps) 2023-03-12 22:28:10 [INFO]: Processing frame 180 (19.55 fps) 2023-03-12 22:28:12 [INFO]: Processing frame 200 (19.73 fps) 2023-03-12 22:28:13 [INFO]: Processing frame 220 (19.95 fps) 2023-03-12 22:28:15 [INFO]: Processing frame 240 (20.15 fps) 2023-03-12 22:28:17 [INFO]: Processing frame 260 (20.37 fps) 2023-03-12 22:28:19 [INFO]: Processing frame 280 (20.56 fps) 2023-03-12 22:28:20 [INFO]: Processing frame 300 (20.72 fps) 2023-03-12 22:28:22 [INFO]: Processing frame 320 (20.85 fps)

I wonder why it happends and i wonder if dropping “base.fc.weight" affect the subsequent training process. Thanks in advance.

1351332484 commented 1 year ago

When you run demo.py again, you have encountered missing files such as ' no model opts ' or ' no model datasets '.What is the solution ?