User Skills

Machine Learning with Tidymodels

These skills help you build and evaluate predictive models using Tidymodels for data analysis and machine learning tasks.

Installation

To use user skills, first add the Tidymodels skills marketplace, then install the user skills collection:

/plugin marketplace add tidymodels/skills
/plugin install tidymodels-users@tidymodels-skills

See the Getting Started Guide for Claude Code installation and detailed setup instructions.

Available Skills

Tabular Data Machine Learning

Build predictive models for tabular data with proper validation practices.

What you’ll learn:

  • Data Spending: Proper train/test splitting with strict test set protection
  • Empirical Validation: Cross-validation and resampling strategies
  • Performance Metrics: Classification (ROC-AUC, Brier score) and regression (RMSE, R²)
  • Feature Engineering: Preprocessing and transformation techniques
  • Model Tuning: Hyperparameter optimization with Tidymodels
  • Model Evaluation: Visualization and final test set evaluation

Key principle: All model development happens on training data using out-of-sample validation. Test set evaluation only occurs with explicit user permission.

Prerequisites

Before using user skills, ensure you have:

  • R 4.1.0 or higher
  • tidymodels packages: install.packages("tidymodels")

See the Getting Started Guide for detailed setup instructions.

Workflow

  1. Load the skill when starting a modeling project
  2. Describe your data and goal - be specific about prediction task
  3. Iterate with Claude Code through data splitting, feature engineering, and model development
  4. Review visualizations and metrics from cross-validation
  5. Finalize model and evaluate on test set (with permission)

Get Involved

Interested in contributing more user skills? Explore Developer Skills to create Tidymodels extensions.