JackieWu

Hi @giginho83 , I have uploaded the example for the inference with auto inheritance in PR #223 . Before the code is merged into the main branch, you can try...

Hi @Zhong1015， thanks for your attention to our work! I wonder which vision task the model handles, and which evaluation metric is used.

@Zhong1015 What are the mAP of the baseline and the experiments equipped iRPE on k and qkv? The iRPE on qkv may have little improvement. You can conduct multiple experiments...

Hi @gudrb , thanks for your attention to our work! Does the class token exist in the fine-tuned model? If the class token exists `(use_cls_token=True)`, please set `skip=1` in https://github.com/microsoft/Cream/blob/main/MiniViT/Mini-DeiT/mini_deit_models.py#L16....

@gudrb > Is it possible to use a pretrained model that utilizes IRPE for a different sequence length, such as a varying number of patches? Yes. You need to pass...

@gudrb No. You don't. Relative position encoding can be adapted with the longer sequence.

@Gumpest I observed `--train-data synthetic` in the training command. Did you replace the dataloader with the one loading LAION-400M image-text pairs?

Sorry for that. Regarding to the data loader, you can refer to the repo OpenCLIP (https://github.com/mlfoundations/open_clip?tab=readme-ov-file#data).

@Gumpest Sorry for late reply. > @wkcn Sorry to bother you, (https://github.com/mlfoundations/open_clip?tab=readme-ov-file#data) tells me how to download the laion-400m dataset, and "replace the dataloader with the one loading LAION-400M image-text...

Thanks for your attention to our work! Let me read the paper and check whether RPE works with approximate attention mechanisms.