Metadata-Version: 2.4
Name: ayase
Version: 0.1.20
Summary: Modular media quality metrics toolkit
Project-URL: Homepage, https://github.com/seruva19/ayase
Project-URL: Repository, https://github.com/seruva19/ayase
Project-URL: Issues, https://github.com/seruva19/ayase/issues
Project-URL: Changelog, https://github.com/seruva19/ayase/blob/main/CHANGELOG.md
Author: Ayase Contributors
License-Expression: MIT
License-File: LICENSE
Keywords: ai,computer-vision,dataset,image,metrics,ml,quality-assessment,video,vqa
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Multimedia :: Video
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.9
Requires-Dist: accelerate>=0.25.0
Requires-Dist: colour-science>=0.4.0
Requires-Dist: decord>=0.6.0
Requires-Dist: diffusers>=0.25.0
Requires-Dist: einops>=0.7.0
Requires-Dist: huggingface-hub>=0.20.0
Requires-Dist: imageio-ffmpeg<1.0,>=0.4.9
Requires-Dist: imageio<3.0,>=2.33.0
Requires-Dist: insightface>=0.7.0
Requires-Dist: invisible-watermark>=0.2.0
Requires-Dist: joblib>=1.3.0
Requires-Dist: librosa>=0.10.0
Requires-Dist: lpips>=0.1.4
Requires-Dist: matplotlib>=3.7.0
Requires-Dist: mediapipe<1.0,>=0.10.0
Requires-Dist: nltk>=3.8.0
Requires-Dist: numba>=0.58.0
Requires-Dist: numpy<3.0,>=1.24.0
Requires-Dist: onnxruntime>=1.14.0
Requires-Dist: open-clip-torch>=2.24.0
Requires-Dist: opencv-python<5.0,>=4.9.0
Requires-Dist: paddleocr<3.0,>=2.7.0
Requires-Dist: paddlepaddle<3.0,>=2.5.0
Requires-Dist: pandas<3.0,>=2.1.0
Requires-Dist: pillow<13.0,>=10.2.0
Requires-Dist: pyarrow<23.0,>=15.0.0
Requires-Dist: pydantic-settings<3.0,>=2.1.0
Requires-Dist: pydantic<3.0,>=2.5.0
Requires-Dist: pyiqa<1.0,>=0.1.13
Requires-Dist: pyrtools>=1.0.9
Requires-Dist: python-levenshtein>=0.21.0
Requires-Dist: pywavelets>=1.3.0
Requires-Dist: pyyaml>=6.0
Requires-Dist: requests>=2.31.0
Requires-Dist: rich<15.0,>=13.7.0
Requires-Dist: safetensors>=0.4.0
Requires-Dist: scikit-image>=0.19.0
Requires-Dist: scikit-learn>=1.3.0
Requires-Dist: scikit-video>=1.1.11
Requires-Dist: scipy>=1.11.0
Requires-Dist: soundfile>=0.12.0
Requires-Dist: t2v-metrics>=1.0.0
Requires-Dist: textual>=0.52.0
Requires-Dist: timm>=0.9.12
Requires-Dist: tomli-w<2.0,>=1.0.0
Requires-Dist: tomli<3.0,>=2.0.1; python_version < '3.11'
Requires-Dist: torch<3.0,>=2.1.0
Requires-Dist: torchvision<1.0,>=0.16.0
Requires-Dist: tqdm<5.0,>=4.66.0
Requires-Dist: transformers<6.0,>=4.45.0
Requires-Dist: transnetv2-pytorch>=1.0.0
Requires-Dist: typer[all]<1.0,>=0.12.0
Requires-Dist: typing-extensions>=4.0.0; python_version < '3.11'
Requires-Dist: ultralytics<9.0,>=8.0.0
Requires-Dist: umap-learn>=0.5.0
Provides-Extra: dev
Requires-Dist: black>=24.0.0; extra == 'dev'
Requires-Dist: mypy>=1.8.0; extra == 'dev'
Requires-Dist: pytest-cov>=4.1.0; extra == 'dev'
Requires-Dist: pytest-sugar>=1.0.0; extra == 'dev'
Requires-Dist: pytest>=7.4.0; extra == 'dev'
Requires-Dist: ruff>=0.1.0; extra == 'dev'
Requires-Dist: seaborn>=0.13.0; extra == 'dev'
Description-Content-Type: text/markdown

# Ayase

Modular media quality metrics toolkit for video, image, and audio datasets.

## Overview

- **324 modules**, **354 quality metrics** across visual, temporal, audio, perceptual, and safety categories
- Modular pipeline - modules compute raw values, downstream apps decide what to do with them
- CLI and Python API with profile-based configuration
- See [METRICS.md](METRICS.md) for the full reference, [MODELS.md](MODELS.md) for all pretrained weights

## Installation

```bash
pip install ayase
```

Ayase ships with the shared runtime dependencies used by the bundled metrics.
Some model implementations include source files under `src/ayase/third_party/`; model binaries are
downloaded and cached automatically on first use when they are not packaged in the repository.

## Quick Start

```bash
ayase scan ./my_dataset
ayase scan ./my_dataset --modules metadata,basic_quality,motion
ayase modules list
ayase modules check
ayase filter ./my_dataset --min-score 70 --output ./filtered
```

```python
from ayase import AyasePipeline

ayase = AyasePipeline(modules=["basic", "aesthetic", "motion"])
results = ayase.run("./my_dataset")

for path, sample in results.items():
    if sample.quality_metrics:
        print(f"{sample.path.name}: score={sample.quality_metrics.technical_score}")

ayase.export("report.json")
```

## Configuration

Create `ayase.toml` in your project root:

```toml
[general]
parallel_jobs = 8

[pipeline]
modules = ["metadata", "basic_quality", "motion"]

[output]
default_format = "json"
artifacts_dir = "reports"
```

## Writing Plugins

```python
from ayase.models import Sample, ValidationIssue, ValidationSeverity
from ayase.pipeline import PipelineModule

class MyCheck(PipelineModule):
    name = "my_check"
    description = "Custom quality check"
    default_config = {"threshold": 0.5}

    def process(self, sample: Sample) -> Sample:
        # Your logic here
        return sample
```

```bash
ayase scan ./data --modules metadata,my_check
```

## Development

```bash
git clone <repo-url> && cd ayase
pip install -e ".[dev]"
pytest
```

## License

MIT - see [LICENSE](LICENSE).

**Model licenses:** ML models downloaded at runtime have their own licenses (Apache 2.0, BSD, CC-BY-NC, etc.). Users are responsible for compliance. See [MODELS.md](MODELS.md) for the full catalog with license info. No model weights are bundled with this package.
