The model and loaded state dict do not match exactly during inference

[Faheem-Yousafzai]

Hi,
Firstly thanks for the great work.

I am using the inference script to check the result on a video. However using given pre-trained model qdtrack-frcnn_r50_fpn_12e_bdd100k-13328aed.pth and the config file qdtrack/configs/bdd100k/qdtrack-frcnn_r50_fpn_12e_bdd100k.py, I got issue that checkpoint file and loaded configuration did not match.

Although an output viddeo is created, but without any tracking results.
Any help would be appreciated.

Regards,
Muhammad Faheem

[lixb3499]

Hi,
I had the same problem when i use the inference script. Did you solve it?

[reb1rth-l]

I'm facing the same problem, did you solve it?