Iván Palomares Carrascosa writes about using the Scikit-LLM library alongside MLflow to build, track, compare, and register scikit-learn pipelines that incorporate large language models. The article provides a workflow for ensuring model versioning and reproducibility by logging different LLM backends as environment parameters and promoting successful pipeline versions into an MLflow Model Registry.
- Uses the `scikit-llm gpt4all » ` installation option to ensure compatibility with local execution.
- Demonstrates how to use `cloudpickle` for serialization when working with scikit-learn models in MLflow.
- Highlights a two-step workflow of logging experiments first and then registering only "winner" models to keep the registry clean.
- Uses the `scikit-llm gpt4all » ` installation option to ensure compatibility with local execution.
- Demonstrates how to use `cloudpickle` for serialization when working with scikit-learn models in MLflow.
- Highlights a two-step workflow of logging experiments first and then registering only "winner" models to keep the registry clean.
Ship measurable improvements in your GenAI systems with Opik, your open-source LLM observability and agent optimization platform. Trusted by over 150,000 developers and thousands of companies.
