fnnx-ai/scikit-llm

Seamlessly integrate LLMs into scikit-learn.

What it solves

It simplifies the integration of Large Language Models (LLMs) like ChatGPT into the familiar scikit-learn ecosystem for text analysis tasks, allowing users to leverage LLM capabilities within a standard machine learning pipeline.

How it works

The library provides scikit-learn compatible wrappers for LLMs. For example, it offers a ZeroShotGPTClassifier that allows users to perform zero-shot text classification by configuring an OpenAI API key and using the standard .fit() and .predict() methods common to scikit-learn models.

Who it’s for

Data scientists and machine learning engineers who are already familiar with scikit-learn and want to use LLMs for text analysis without having to rewrite their existing workflows.

Highlights

  • Seamless integration with scikit-learn's API
  • Support for zero-shot text classification using GPT models
  • Easy configuration via SKLLMConfig

Related

  • Project
    MLCIL/scikit-fingerprints

    A scikit-learn compatible library for chemoinformatics that enables efficient computation of molecular fingerprints, filters, and similarity measures for ML pipelines.

  • Project
    jpmml/sklearn2pmml

    A Python package that converts Scikit-Learn pipelines to PMML format to improve model portability and deployment across different environments.

  • Project
    feature-engine/feature_engine

    A Python library providing Scikit-learn compatible transformers for comprehensive feature engineering and selection to improve machine learning model performance.

  • Project
    skorch-dev/skorch

    A scikit-learn compatible neural network library that wraps PyTorch, enabling the use of scikit-learn's workflow tools with PyTorch models.

  • Project
    Tejas-TA/predikit

    A library that wraps scikit-learn and XGBoost models as LLM-callable tools with auto-generated JSON schemas and Pydantic validation.