A comprehensive MLOps catalog system for managing machine learning experiments, models, and artifacts.
- Experiment tracking and management
- Model registry with versioning
- Data source management
- Local and cloud runners
- CLI interface
# Clone the repository
git clone https://github.com/yourusername/mlops_catalog.git
cd mlops_catalog
# Create and activate virtual environment
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install development dependencies
pip install -e ".[dev]"- Register a model:
mlops model register config.yaml- Run an experiment:
mlops exp run experiment.yaml train.py- Register a data source:
mlops data register training-data data/train.csvFull documentation is available in the docs directory.
This project is licensed under the MIT License.