Yiping Wang

problems about poverty worst r

1

Thanks for your implementation of fish, but when I test poverty with algorithm erm, the worst r is just about 0.2, while the performance noted in wilds is about 0.5...

Fail to load Optix Liabrary

5

I want to use your OptixRender program and quote your paper in the research process, but we met some problems thus we ask you for help. I used the lab...

About update metadata with the corresponding image sample in shards

2

Hi all, thanks for your great work!! I want to ask some relevant technical questions: I have used dataset2metadata to generate the metadata set from the 'shards' and store some...

Thanks for your cool work! Your idea of replacing classical CLIPScore with better metrics obtained by MLM is interesting. I'm currently working on selecting more informative data for training CLIP...

Problems about resuming from checkpoint for finetune_with_lora

1

Hi, thanks for your great job. I found some errors when I wanted to continue my LoRA finetuning. ``` 09/04/2024 15:54:23 - INFO - accelerate.accelerator - Loading states from output/tulu_v2_dolly_openorca_1M_v2_64_7B_lora/step_15600...

Hi, thank you for your impressive work, I note that in the data directory, we cannot find a file named base and thus we cannot load Txt2ImgIterableBaseDataset class. Without this...