requirements.txt

# numpy==1.24.3
# torch==2.6.0
# torchvision==0.21.0
# torchaudio==2.6.0
# torchdiffeq==0.2.3
# omegaconf>=2.3.0
# huggingface_hub>=0.26.1
# Pillow>=9.5
# opencv-python-headless>=4.8
# gradio>=4.19.2
# einops>=0.6
# open_clip_torch>=2.29.0
# av>=14.0.1
# transformers>=4.36.2
# ffmpeg-python>=0.2.0
# moviepy>=1.0.3
# python-multipart>=0.0.9
# colorlog==6.8.2
# pandas>=2.0.0
# tensordict>=0.2.0
# hydra-core>=1.3.2
# tqdm>=4.65.0
# librosa>=0.10.1
# timm>=0.9.12
# requests>=2.31.0