Metadata-Version: 2.4
Name: docling-metrics-core
Version: 0.1.0
Summary: Metric types and evaluation utilities for Docling document processing
Project-URL: homepage, https://github.com/docling-project/docling-metrics
Project-URL: repository, https://github.com/docling-project/docling-metrics
Project-URL: issues, https://github.com/docling-project/docling-metrics/issues
Author-email: Christoph Auer <cau@zurich.ibm.com>, Panos Vagenas <pva@zurich.ibm.com>, Ahmed Nassar <ahn@zurich.ibm.com>, Nikos Livathinos <nli@zurich.ibm.com>, Maxim Lysak <mly@zurich.ibm.com>, Michele Dolfi <dol@zurich.ibm.com>, Peter Staar <taa@zurich.ibm.com>
License-Expression: MIT
Keywords: docling,evaluation,metrics,types
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: Operating System :: MacOS :: MacOS X
Classifier: Operating System :: POSIX :: Linux
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: <4.0,>=3.10
Requires-Dist: pydantic<3.0.0,>=2.0.0
Description-Content-Type: text/markdown

# docling-metrics-core

A minimal interface for computing and aggregating metrics on paired data samples.

## Overview

This package provides base types for building metrics that:
- Evaluate pairs of input samples (e.g., ground-truth vs. prediction)
- Produce per-sample results traceable by sample ID
- Aggregate results across multiple samples

## Core Types

- **`BaseInputSample`** — Input data with a unique `id` shared between sample pairs
- **`BaseSampleResult`** — Output of evaluating a single sample pair
- **`BaseAggregateResult`** — Output of aggregating multiple sample results
- **`BaseMetric`** — Abstract interface defining `evaluate_sample`, `aggregate`, and `evaluate_dataset`

## Installation

```bash
pip install docling-metrics-core
```

## Requirements

- Python >=3.10,<4.0

## Usage

*Coming soon*

## Development

```bash
# Install in development mode
uv sync

# Run tests
uv run pytest
```

## License

MIT
