Dachuan Shi
Dachuan Shi
Hello, thanks for your nice work! I am now having trouble reproducing the reported score on the VQA task. I evaluated the checkpoint downloaded from https://storage.googleapis.com/sfr-vision-language-research/BLIP/models/model_base_vqa_capfilt_large.pth and followed the default...
Hello, thanks for your nice work! Are there scripts and configuration files that can be used to finetune CLIP on COCO and Flickr30K, like BLIP ([retrieval_coco_ft.yaml](https://github.com/salesforce/LAVIS/blob/main/lavis/projects/blip/train/retrieval_coco_ft.yaml) and [train_retrieval_coco](https://github.com/salesforce/LAVIS/blob/main/run_scripts/blip/train/train_retrieval_coco.sh))? Thanks again!
Hi, thanks for sharing this great open-source project! When using multiple GPUs for evaluation, I found partition tasks sometimes will fail due to occupied ports. ### Prerequisite - [X] I...
[Feature] Can we find the config version that is used to produce the score displayed on leaderboard?
### Describe the feature Hi, thanks for sharing this wonderful project! I am wondering is there somewhere we can find the config version that is used to produce the score...
Hi, @he-y, very thanks for your great list from which I learned a lot! I am also working on pruning and I would like to contribute the full list of...
Hi, @pliang279, very thanks for your great list from which I learned a lot! Recently, we have a new work about compressing multimodal models, i.e., making them more lightweight and...
Hi, thanks for building this wonderful open-source project! I am using GPTQ to first quantize a llama2-7b-chat-hf model: ```bash python quantize.py --checkpoint_path checkpoints/meta-llama/Llama-2-7b-chat-hf/model.pth --mode int4-gptq --calibration_tasks wikitext --calibration_seq_length 2048 ```...