VideoRefer-VideoLLaMA3 / requirements.txt
lixin4ever's picture
Update requirements.txt
7265d87 verified
raw
history blame
642 Bytes
# basic dependencies
torch
torchvision
datasets
transformers==4.46.3
tokenizers
deepspeed
accelerate
peft
timm
numpy
# data processing
decord
imageio
imageio-ffmpeg
moviepy
scenedetect
opencv-python
pyarrow
pysubs2
ffmpeg-python
# misc
scikit-learn
huggingface_hub
sentencepiece
shortuuid
einops
einops-exts
bitsandbytes
pydantic>=2.0
markdown2[all]
gradio==5.34.0
gradio_client==1.10.3
httpx==0.24.1
requests
openai
uvicorn
fastapi
tensorboard
wandb
tabulate
Levenshtein
pycocotools
spaces
https://github.com/mjun0812/flash-attention-prebuild-wheels/releases/download/v0.0.8/flash_attn-2.7.4.post1+cu126torch2.7-cp310-cp310-linux_x86_64.whl