Does this skill help with data preprocessing?

Yes, it provides comprehensive guidance for scaling, encoding categorical variables, and handling missing values using ColumnTransformers and Pipelines.

Does it support unsupervised learning tasks?

Yes, it covers dimensionality reduction (like PCA and t-SNE) and various clustering algorithms including K-Means, DBSCAN, and Gaussian Mixture Models.

Can I use this skill for hyperparameter tuning?

Absolutely. It includes implementation patterns for GridSearchCV, RandomizedSearchCV, and successive halving techniques to optimize model performance.

How do I install scikit-learn for use with this skill?

You can install scikit-learn and its common dependencies using uv with the command: uv pip install scikit-learn pandas numpy.

Why does the skill recommend using Pipelines?

Pipelines are recommended because they chain transformers and estimators together, ensuring reproducibility and preventing data leakage during cross-validation.

Scikit-learn Machine Learning

Name: Scikit-learn Machine Learning
Author: x-cmd

byx-cmd

•

データサイエンスとML

Streamlines the implementation, evaluation, and deployment of classical machine learning models using the scikit-learn library.

概要

This skill equips Claude with specialized knowledge for building robust machine learning workflows using Python's industry-standard scikit-learn library. It provides domain-specific patterns for the entire ML lifecycle, including data preprocessing with ColumnTransformers, implementing supervised and unsupervised algorithms, and performing rigorous model evaluation through cross-validation. By emphasizing the use of production-grade Pipelines, the skill helps developers prevent data leakage and ensure that preprocessing steps are consistently applied during both training and inference. It is an essential companion for tasks ranging from exploratory data analysis and dimensionality reduction to complex hyperparameter tuning and model selection.

主な機能

8 GitHub stars
Automated preprocessing, scaling, and categorical encoding
Comprehensive supervised and unsupervised algorithm implementations
Robust model evaluation using cross-validation and performance metrics
End-to-end ML pipeline construction to prevent data leakage
Advanced hyperparameter tuning via Grid and Random search

ユースケース

Building predictive regression models for financial or demand forecasting
Creating automated data cleaning and transformation workflows for tabular datasets
Developing customer segmentation models using K-Means or density-based clustering

概要

主な機能

8 GitHub stars
Automated preprocessing, scaling, and categorical encoding
Comprehensive supervised and unsupervised algorithm implementations
Robust model evaluation using cross-validation and performance metrics
End-to-end ML pipeline construction to prevent data leakage
Advanced hyperparameter tuning via Grid and Random search

ユースケース

Building predictive regression models for financial or demand forecasting
Creating automated data cleaning and transformation workflows for tabular datasets
Developing customer segmentation models using K-Means or density-based clustering