Automates the partitioning of raw data into training, validation, and testing sets for machine learning workflows.
The ML Dataset Splitter skill streamlines data preparation by automatically dividing datasets into optimized subsets for model development. It handles the generation and execution of Python code to perform data partitioning, allowing developers to define custom ratios, implement stratification for imbalanced data, and ensure randomization. By automating these foundational steps, the skill ensures data integrity and facilitates more robust model evaluation within the Claude Code environment.
Características Principales
01Automatic Python code generation using standard ML libraries
02Automated train-validation-test partitioning
03Customizable split ratios (e.g., 70/15/15 or 80/20)
04Support for CSV and large-scale tabular datasets
05Built-in support for data stratification and randomization
060 GitHub stars
Casos de Uso
01Preparing structured data for supervised learning model training
02Creating hold-out test sets to evaluate final model performance
03Partitioning data subsets to ensure balanced class distributions
