AI4Math-ShanZhang

mismatch owing to pretrain model_final.pkl.

2

After i download the model_final.pkl, runing the model shows the issue, like "expected 4-dimensional weight (64,3,7,7) but got input size of (4,10,3,320,320)".

Hi, thank you for this influential work! I have found in [this line](https://github.com/IDEACVR/DINO/blob/main/models/dino/deformable_transformer.py#:~:text=delta_unsig%20%3D%20self.bbox_embed%5Blayer_id%5D(output)) where you use the features without normalization when computing the delta of boxes. For the auxiliary one,...

The training log and the query embedding

3

Thanks for release the great job! Would you please provide the base training/ few-shot fine-tuning logs, eg., the 10-shot setting on split-1 (seed0) of voc dataset? And, i have noticed...