snap-research / Panda-70M

[CVPR 2024] Panda-70M: Captioning 70M Videos with Multiple Cross-Modality Teachers
https://snap-research.github.io/Panda-70M/
505 stars 19 forks source link

generate Vicuna error and inference error #57

Open wen020 opened 3 months ago

wen020 commented 3 months ago

I used the following command to generate Vicuna, but the following error occurred. Which step did I do wrong?

python3 -m fastchat.model.apply_delta --base-model-path /path/to/llama-7b --target-model-path /path/to/output/vicuna-7b --delta-path lmsys/vicuna-7b-delta-v0

base-model:https://huggingface.co/huggyllama/llama-7b/tree/main delta-model:https://huggingface.co/lmsys/vicuna-7b-delta-v0 image

Then I change delta-model to delta-model:https://huggingface.co/lmsys/vicuna-7b-delta-v1.1 run python3 -m fastchat.model.apply_delta \ --base-model-path /app/wen/Panda-70M/captioning/llama-7b \ --target-model-path /app/wen/Panda-70M/captioning/vicuna_weights/vicuna-7b-v0 \ --delta-path /app/wen/Panda-70M/captioning/vicuna-7b-delta-v1.1

Vicuna was able to successfully generate image

But when I ran the following command: python inference.py --video-list inputs/video_list.txt --prompt-list inputs/prompt_list.txt

the following error occurred: image