mldebug 0.0.0

mldebug is a lightweight Python toolkit for debugging machine learning systems in production.

mldebug

mldebug is a lightweight Python toolkit for debugging machine learning systems in production.

Quick Start

git clone https://github.com/anpenta/mldebug
cd mldebug
direnv allow
poe test

Status

Active development (v0.x, not yet stable).

Features

• Debug ML systems in production environments
• Lightweight diagnostic utilities
• Designed for integration into modern ML pipelines
• Focus on observability and failure analysis

Installation

pip install mldebug

Example Usage

See tests/test_public_api.py for real usage examples.

Documentation

See github pages.

Development Setup

Requirements

• Ubuntu 24.04.4 (recommended) or WSL
• Git
• direnv

Environment Setup

If not already done via Quick Start:

direnv allow

Development Workflow

Tasks are managed via poe (available in the project environment via direnv).

Run tests

poe test

Run linting

poe lint

Run full CI parity checks

poe test-all
poe lint-ci-all

CI/CD

CI runs multi-Python version testing and linting. All pull requests must pass the checks. Merging is blocked unless all checks pass.

See .github/workflows/ci.yml for details.

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Ensure all CI checks pass
5. Open a pull request

Dependency Management

Dependencies are managed using uv and defined in pyproject.toml.

License

See LICENSE.