megvii-research/AnchorDETR

Pretrained model

chenhbo opened this issue · 3 comments

Hi,
nice work!

Is there possible to put the pre-trained model up here in the future?
We would like to use the AnchorDETR to train our custom data (no more than 300 images).
Maybe fine-tuning the pre-trained model is better for us.
Thank you!

Hi, we have provided the trained models. You can refer to the readme.

I'm closing this issue as I think I have answered your question, but let us know if you have further questions.

Thank you very much for your reply~