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

SysCV / qdtrack

Quasi-Dense Similarity Learning for Multiple Object Tracking, CVPR 2021 (Oral)

Apache License 2.0

384 stars 61 forks source link

The model and loaded state dict do not match exactly during inference #137

Open Faheem-Yousafzai opened 1 year ago

Faheem-Yousafzai commented 1 year ago

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 commented 1 year ago

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

reb1rth-l commented 6 months ago

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