Jihao Liu

As shown in the picture.(click the link below you will get it) [loss curve link](http://116.62.134.81/images/loss.PNG) Is it a bug ?

Thanks for your great effort in making such a collection. We recently published a paper about multi-camera perception, GeoMIM in ICCV 2023. Can you add it to your collection? Thanks...

RuntimeError: Expected object of type torch.FloatTensor but found type torch.DoubleTensor for argument #3 'other'

2

The error occurs in functions/LBFGS.py, line 854. I think this error comes from t. It becomes a double precision number with the process of convergence. I tried ``` if(F_new >...

Difference between eva-clip model and MIM pretrained model

2

I find that the eva-clip model has an extra inner_attn_ln layer compared to the original pretrained model. [EVA-CLIP](https://github.com/baaivision/EVA/blob/master/EVA-CLIP/rei/eva_clip/eva_vit_model.py#L164) [EVA](https://github.com/baaivision/EVA/blob/master/EVA-02/asuka/modeling_finetune.py#L188) Any reason to use another layer-norm in the clip model?

Hi, I'm trying to reproduce the provided results on my cluster. I found the training time is really slow, and the depth loss seems unstable. I'm working on checking the...

生成车牌时，W字母会明显的被遮挡，请问这个问题的原因是什么呢？

Request for middle checkpoint

3

Thank you for the amazing release! Do you plan to release the checkpoints from different stages, e.g., checkpoint before SFT? These checkpoints would be valuable for further fine-tuning.