What file formats does this skill support?

The skill primarily handles common structured data formats like CSV files but can be used for any dataset readable via standard Python libraries like Pandas.

Can I specify custom split ratios?

Yes, you can define specific percentages for training, validation, and testing subsets (e.g., 80% training and 20% testing) to fit your specific project needs.

Can I use this for imbalanced datasets?

The skill follows best practices and can implement stratification to maintain consistent class distributions across all your partitioned subsets.

Does this skill execute code locally?

Yes, the skill generates and executes Python code within your environment to perform the data manipulation on your local files.

Does it handle data randomization?

Yes, the skill ensures the splitting process is randomized to avoid selection bias and ensure that your subsets are representative of the overall data.

Machine Learning Dataset Splitter

Name: Machine Learning Dataset Splitter
Author: jeremylongshore

byjeremylongshore

•

883

데이터 과학 및 ML

Partitions raw datasets into training, validation, and testing sets to streamline machine learning model development.

소개

This skill empowers Claude to automate the critical data preparation phase of machine learning projects by programmatically dividing datasets into optimized subsets. It handles CSV and other structured data formats, generating and executing Python code to create precise train-test-validation splits based on user-defined ratios. By automating the partitioning process and ensuring proper randomization and stratification, it helps developers maintain rigorous evaluation standards and accelerate the model development lifecycle directly within the Claude Code environment.

주요 기능

Automated train-validation-test partitioning
Python-based execution using industry-standard libraries
Support for CSV and large-scale data file formats
Randomized and stratified splitting to prevent evaluation bias
883 GitHub stars
Customizable subset ratios (e.g., 70/15/15 or 80/20)

사용 사례

Creating stratified splits for imbalanced classification datasets
Preparing raw CSV data for supervised model training and evaluation
Automating data preprocessing pipelines within a terminal-based AI workflow

소개

주요 기능

Automated train-validation-test partitioning
Python-based execution using industry-standard libraries
Support for CSV and large-scale data file formats
Randomized and stratified splitting to prevent evaluation bias
883 GitHub stars
Customizable subset ratios (e.g., 70/15/15 or 80/20)

사용 사례

Creating stratified splits for imbalanced classification datasets
Preparing raw CSV data for supervised model training and evaluation
Automating data preprocessing pipelines within a terminal-based AI workflow