Baptiste Standaert

I have exactly the same issue. It happens when I am trying to fine-tune a model (using `--checkpoint`) that has not been trained on my custom dataset. Surprisingly, I don't...

I managed to finetune a `checkoint` by decreasing the learning rate. I use now `--lr=0.0001` with `SGD` optimizer and everything seems to work.

I got the same problem here with M2. I managed to solve this issue very badly but things still work : I changed the package to `pycocotools` which is "a...

Hello, I have the same troubles. Did you find a solution @MintMAO ? I think it would be a great idea to work on the same detections for fair comparison....

We need to clean the printings and test the PR. Thank you for your contribution.

Hello, The device assignment is handled [here](https://github.com/TrackingLaboratory/tracklab/blob/095306aa4bd89c94ead5c579b15c41870d1356f2/tracklab/main.py#L82). You can modify the behavior directly in this function if you need a different setup. Hope this helps! Let me know if you...

Hello, sorry for the late reply. If it’s not too late: yes, you can. You can use the ExternalVideo module, which is configurable—you can specify your video path in config/dataset/video.yaml....