I want to validate on my own dataset.

[pandascd]

Hello, I would like to perform validation on my own dataset. First, I arranged my dataset according to the DanceTrack dataset format, and the dataset name remains as "dancetrack". Then, I ran the following command: python main.py --mode eval --data-root /home2/cxz/python_project/MeMOTR-main/DATADIR --eval-mode specific --evdir ./outputs/memotr_dancetrack/ --eval-threads 1 An error occurred.：

I don’t know where the problem occurred, and how the file pedestrian_summary.txt is generated.

[HELLORPG]

According to the error, you are missing the output file ./outputs/memotr_dancetrack/val/memotr_dancetracker_tracker/cam_28.txt. Did you really have this file in your outputs dir?

[pandascd]

Thank you for your response. I checked and found that the file ./outputs/memotr_dancetrack/val/memotr_dancetracker_tracker/cam_28.txt does not exist, but the file ./outputs/memotr_dancetrack/val/memotr_dancetrack_tracker/tracker/cam_28.txt does exist. It seems that this file is automatically generated. I'm not sure if the issue lies with the path specified in the code. Where should I modify it?

[HELLORPG]

Have you specified the checkpoint name using --eval-model?

[HELLORPG]

BTW, what system did you use for this experiment? Windows or Linux?

[HELLORPG]

I have closed this issue due to a long period of inactivity.