toyot-li
toyot-li
@FDInSky could you please specify it? Or post a link for the relevant issue?
Hi @ZwwWayne, as my reply was initially posted in the `Reproduction` part: 1. The dataset is official COCO'17 detection. 2. I did not make any modifications to the code or...
@RangiLyu @ZwwWayne any fix?
@hyz-xmaster Thanks for your prompt reply. For `pos_decoded_target_preds`, I guess it may not be a must. It seems common in machine learning that the gradient does not flow back to...
@hyz-xmaster any update?
@hyz-xmaster Exactly, I see that QFL does not and should not propagate gradients to `pos_decoded_bbox_preds`. Then back to my above question, why did VFL choose to propagate gradients to `pos_decoded_bbox_preds`,...
@tangjiuqi097 Regarding the first question, one line is inserted after [this line](https://github.com/megvii-research/AnchorDETR/blob/main/models/transformer.py#L158) for refinement ```python reference_points = output_coord[..., :2].clone().detach() ``` But the performance is slightly lower (~1% AP) than the...
@tangjiuqi097 Thanks for your rapid reply. Although a simple line is changed, it is after my second thought. As far as I understand, the `output_coord[.., :2]` of the previous decoder...
@danthe3rd But I think it means `attn_mask` in https://pytorch.org/docs/stable/generated/torch.nn.MultiheadAttention.html#torch.nn.MultiheadAttention, not corresponding to `key_padding_mask`? correct me if I miss something.
same question here.