zwhus

[Fix] Fix mmdet flops error

1

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand...

Thank you for your excellent work. It is mentioned in the paper that it can be added to petrv2 or bevformer. Is there any plan to release the corresponding code?

Load weight error

6

Hi, Thanks for your excellent work. Now I ran into an issue when I tried to load GPT4ROI weights to perform stage2 training and there was an error ”Error(s) in...

权重载入

1

发现一个问题，因为微调为保存得到的是pt, 然后使用pth_to_hf后，保存为safetensor的形式，再多卡测试时载入，发现显存占用超过80g，但是将其保存为bin形式，每张卡显存均保持在30g，这是为什么呢，能否转换为hf的时候直接变成bin呢

Thanks for your work! I have a question about gradient accumulate on contrastive learning. you use the code as follows: ``` all_teacher_image_features = torch.cat( [teacher_image_features] + gathered_teacher_image_features[:rank] + gathered_teacher_image_features[rank +...

Bge-M3 训练

2

Hi, 我想问一下Bge-M3 在训练中，处理像NQ这样的问答数据，有没有加一下instruction进行训练呢？因为我直接在训练好的权重上继续训练nq, 发现loss很高

VISTA Train

3

Can you provide some details about the training of stage 2? When I trained using the same approach like learning rate and hard samples, I found that the IT-I loss...

Visualized BGE on COCO/Flickr

3

Thank you for your excellent work. I used Visualized BGE for testing on COCO/Flickr, but the retrieval performance was quite poor. However, I used the stage2 weights. I would like...

Visualized BGE Evaluation

12

Hi, i want to reproduce the result of Visualized BGE, but zero-shot benchmark not clear, such as WebQA. Can you provide evaluation dataset and codes for zero-shot benchmark. Thanks!

Visualized_BGE Train

1

您好，我想问一下在训练过程呢各种难过stage2阶段，难样本的选择是怎么样的呢，论文中我看说是it2i是3，那么是随机选机选择3吗，t2it需要使用难样本吗, 另外我想问一下stage训练的时候是一个batch是it2i或者t2it吗，是否有其他额外的操作呢