shaoxiongduan
f6275f8005
Adds an in-process evaluation suite covering native (SSIM/PSNR/LPIPS/ optical_flow), audio, physics_iq, vbench (16 sub-metrics), and videoscore2. Public API: create_evaluator/evaluate, BaseMetric + @register, ensure_checkpoint, get_cache_dir. CLI: fastvideo eval list/run. VBench upstream pinned as a git submodule at fastvideo/third_party/eval/vbench (Vchitect/VBench@45e79ec); modern-dep compat is achieved via runtime shims in vbench/__init__.py rather than on-disk patches. CLIP/torch.hub caches are routed through ${FASTVIDEO_EVAL_CACHE}; HF cache stays at the system default. ensure_checkpoint delegates to fastvideo.utils.get_lock + huggingface_hub primitives. Evaluator supports release_cuda_memory(), unload(), reload() for training-time eval that frees GPU between calls. Verified parity against upstream vbench (5/8 metrics bit-exact, others within 1% drift driven by transformers/torch version skew) and against upstream VideoScore2 (regex anchored on the actual model's output format, soft-score formula matches upstream's argmax*max_prob/total). Includes docs/contributing/eval-metrics.md as the porting guide. Out of scope (deferred follow-ups): MIND, VBench-2.0, FVD as a registered metric, training-time EvalCallback.
233 lines
4.7 KiB
TOML
233 lines
4.7 KiB
TOML
[build-system]
|
|
requires = ["setuptools>=61.0"]
|
|
build-backend = "setuptools.build_meta"
|
|
|
|
[project]
|
|
name = "fastvideo"
|
|
version = "0.1.7"
|
|
description = "FastVideo"
|
|
readme = "README.md"
|
|
requires-python = ">=3.10"
|
|
classifiers = [
|
|
"Programming Language :: Python :: 3",
|
|
"License :: OSI Approved :: Apache Software License",
|
|
]
|
|
|
|
dependencies = [
|
|
# Core Libraries
|
|
"scipy==1.14.1",
|
|
"six==1.16.0",
|
|
"h5py==3.12.1",
|
|
"requests>=2.32.2",
|
|
|
|
# Machine Learning & Transformers
|
|
"transformers==4.57.3",
|
|
"tokenizers>=0.20.1",
|
|
"sentencepiece==0.2.0",
|
|
"timm==1.0.11",
|
|
"peft>=0.15.0",
|
|
"diffusers>=0.33.1",
|
|
"torch>=2.10.0",
|
|
"torchvision",
|
|
"torchaudio",
|
|
|
|
# Acceleration & Optimization
|
|
"accelerate==1.0.1",
|
|
|
|
# Computer Vision & Image Processing
|
|
"opencv-python==4.10.0.84",
|
|
"pillow>=10.3.0",
|
|
"imageio==2.36.0",
|
|
"imageio-ffmpeg==0.5.1",
|
|
"einops",
|
|
|
|
# Experiment Tracking & Logging
|
|
"wandb>=0.21.0",
|
|
"loguru",
|
|
"test-tube==0.7.5",
|
|
|
|
# Miscellaneous Utilities
|
|
"tqdm",
|
|
"pytest",
|
|
"PyYAML==6.0.1",
|
|
"protobuf>=5.28.3",
|
|
"gradio==5.32.0",
|
|
"moviepy>=2.0.0",
|
|
"flask",
|
|
"flask_restful",
|
|
"aiohttp",
|
|
"aiofiles",
|
|
"huggingface_hub",
|
|
"cloudpickle",
|
|
"omegaconf",
|
|
|
|
# System & Monitoring Tools
|
|
"gpustat",
|
|
"watch",
|
|
"remote-pdb",
|
|
|
|
# Kernel & Packaging
|
|
"fastvideo-kernel==0.2.6",
|
|
"wheel",
|
|
|
|
# Training Dependencies
|
|
"torchdata",
|
|
"pyarrow",
|
|
"datasets==4.0.0",
|
|
"av",
|
|
|
|
# Preprocessing Dependencies
|
|
"torchcodec",
|
|
"ray>=2.49.1",
|
|
"ftfy==6.3.1",
|
|
|
|
# Job Runner
|
|
"fastapi==0.129.0",
|
|
"uvicorn==0.41.0"
|
|
]
|
|
|
|
[tool.uv]
|
|
prerelease = "allow"
|
|
|
|
[tool.uv.sources]
|
|
torch = [
|
|
{ index = "pytorch-cpu", marker = "sys_platform != 'linux'" },
|
|
{ index = "pytorch-cu128", marker = "sys_platform == 'linux'" },
|
|
]
|
|
torchvision = [
|
|
{ index = "pytorch-cpu", marker = "sys_platform != 'linux'" },
|
|
{ index = "pytorch-cu128", marker = "sys_platform == 'linux'" },
|
|
]
|
|
|
|
[[tool.uv.index]]
|
|
name = "pytorch-cpu"
|
|
url = "https://download.pytorch.org/whl/cpu"
|
|
explicit = true
|
|
|
|
[[tool.uv.index]]
|
|
name = "pytorch-cu128"
|
|
url = "https://download.pytorch.org/whl/cu128"
|
|
explicit = true
|
|
|
|
[project.optional-dependencies]
|
|
|
|
# flash-attn: pip install flash-attn==2.8.1 --no-cache-dir --no-build-isolation
|
|
|
|
|
|
lint = [
|
|
"pre-commit==4.0.1",
|
|
"types-aiofiles",
|
|
"types-requests",
|
|
]
|
|
|
|
test = [
|
|
"av",
|
|
"pytorch-msssim==1.0.0",
|
|
"pytest",
|
|
]
|
|
|
|
# Extra deps for `fastvideo.eval`. The vbench upstream submodule ships
|
|
# its own ad-hoc requirements list; this group covers the imports that
|
|
# the in-tree metric adapters (and vbench's pure-Python helpers) need.
|
|
# Heavier per-metric deps (detectron2, mmcv, ...) are left for users to
|
|
# install on demand — the metric registry surfaces a clear ImportError.
|
|
eval = [
|
|
"decord",
|
|
"openai-clip",
|
|
"open_clip_torch",
|
|
"pyiqa",
|
|
"ptlflow",
|
|
"lpips",
|
|
"torchmetrics",
|
|
"easydict",
|
|
"qwen-vl-utils",
|
|
]
|
|
|
|
dev = [ "fastvideo[lint]", "fastvideo[test]", ]
|
|
|
|
rocm = [
|
|
"amdsmi",
|
|
]
|
|
|
|
[project.scripts]
|
|
fastvideo = "fastvideo.entrypoints.cli.main:main"
|
|
|
|
[tool.setuptools.packages.find]
|
|
exclude = ["assets*", "docker*", "docs", "scripts*"]
|
|
|
|
[tool.wheel]
|
|
exclude = ["assets*", "docker*", "docs", "scripts*"]
|
|
|
|
|
|
[tool.mypy]
|
|
warn_unused_configs = true
|
|
ignore_missing_imports = true
|
|
disallow_untyped_calls = true
|
|
check_untyped_defs = true
|
|
follow_imports = "silent"
|
|
|
|
[tool.codespell]
|
|
skip = "./data,./wandb,ui/package-lock.json"
|
|
|
|
[tool.ruff]
|
|
# Allow lines to be as long as 120.
|
|
line-length = 120
|
|
extend-exclude = [
|
|
# Upstream submodules under fastvideo/third_party/ are not our style
|
|
# territory; modern-dep compat lives as runtime shims in our metric
|
|
# __init__.py files.
|
|
"fastvideo/third_party/**",
|
|
]
|
|
|
|
[tool.ruff.lint]
|
|
select = [
|
|
# pycodestyle
|
|
"E",
|
|
# Pyflakes
|
|
"F",
|
|
# pyupgrade
|
|
"UP",
|
|
# flake8-bugbear
|
|
"B",
|
|
# flake8-simplify
|
|
"SIM",
|
|
# isort
|
|
# "I",
|
|
"G",
|
|
]
|
|
ignore = [
|
|
# star imports
|
|
"F405", "F403",
|
|
# lambda expression assignment
|
|
"E731",
|
|
# Loop control variable not used within loop body
|
|
"B007",
|
|
# f-string format
|
|
"UP032",
|
|
# line too long
|
|
"E501",
|
|
]
|
|
|
|
[tool.ruff.lint.per-file-ignores]
|
|
# Ignore all files that end in `_test.py`.
|
|
"fastvideo/models/hunyuan/diffusion/pipelines/pipeline_hunyuan_video.py" = ["E741"]
|
|
|
|
|
|
[tool.yapf]
|
|
column_limit = 120
|
|
|
|
[tool.isort]
|
|
line_length = 120
|
|
use_parentheses = true
|
|
skip_gitignore = true
|
|
|
|
|
|
[project.urls]
|
|
Repository = "https://github.com/hao-ai-lab/FastVideo"
|
|
# Used by Comfy Registry https://comfyregistry.org
|
|
|
|
[tool.comfy]
|
|
PublisherId = "fastvideo"
|
|
DisplayName = "FastVideo"
|
|
Icon = "https://raw.githubusercontent.com/hao-ai-lab/FastVideo/main/comfyui/assets/icon_simple.svg"
|