Does this skill support deep learning algorithms?

This skill focuses on classical machine learning algorithms provided by Scikit-learn. For neural networks and deep learning, specialized skills for frameworks like PyTorch or TensorFlow are recommended.

How does this skill help with data leakage?

The skill provides patterns for implementing Scikit-learn Pipelines, which encapsulate preprocessing and modeling steps together to ensure transformations are only learned from the training data.

Can I use this for high-dimensional data visualization?

Yes, it includes specific patterns for dimensionality reduction using PCA, t-SNE, and UMAP to prepare complex data for 2D or 3D visualization.

What algorithms are recommended for very large datasets?

The skill suggests using out-of-core learners like SGDClassifier or high-performance implementations like HistGradientBoosting for datasets exceeding 100,000 samples.

Scikit-learn Machine Learning Toolkit

Name: Scikit-learn Machine Learning Toolkit
Author: lifangda

bylifangda

•

データサイエンスとML

Implements comprehensive machine learning workflows including classification, regression, and data preprocessing using the industry-standard Scikit-learn library.

概要

This skill equips Claude with specialized knowledge for building end-to-end machine learning pipelines using Python's Scikit-learn library. It provides domain-specific guidance on data preprocessing, algorithm selection, and robust model evaluation. From implementing baseline linear models to complex ensemble methods like Gradient Boosting, this skill ensures your ML workflows follow best practices, avoid common pitfalls like data leakage, and include systematic hyperparameter optimization for production-grade results.

主な機能

End-to-end machine learning pipeline construction to prevent data leakage
Automated data preprocessing, feature scaling, and categorical encoding
Guided algorithm selection for classification, regression, and clustering
Systematic hyperparameter tuning using GridSearch and RandomSearch
Comprehensive model evaluation using cross-validation and advanced metrics
16 GitHub stars

ユースケース

Building predictive models for business forecasting, price prediction, and trend analysis
Performing unsupervised learning for customer segmentation and anomaly detection
Developing classification systems for spam detection, sentiment analysis, or medical diagnosis