AI Blocks › Build and ship software
Data classifier comparer
Give it a data table. Get several classifiers trained and ranked by accuracy, with tests that show real differences.
Data classifier comparer is a free AI skill for the job "Build and ship software". Give the instructions to an AI assistant that supports the required files and tools, then review its output.
Free · No account · BSD 2 Clause licence · Made by MathWorks · Level: advanced · Thinking: high
Made by MathWorks. See the original on GitHub; we copied it as it is, with its licence.
What it does
Find out which classification models really are the best on your data, and which ones are only slightly different by chance. Rankings alone can mislead, so this adds statistical tests. You bring a table with predictors and a category column to predict. The skill cleans and prepares it, picks promising classifiers, and trains them in MATLAB. It compares accuracy with cross-validation, a holdout split or hyperparameter search. It then runs McNemar, 5 by 2 cv and Friedman tests to show which models sit in the top group with no real difference. It supports a uniform prior for imbalanced classes but nothing more custom. It does not handle images, sequences or numeric targets, and it needs MATLAB with statistics tools.
You give it
- A data table with a category column
- Evaluation method preference (optional)
- Whether classes are imbalanced (optional)
You get
- Trained classifiers
- Accuracy ranking
- Statistical test results
- Top tier of equal models
Use it when
- You need to choose a model for churn, fraud or diagnosis data.
- Two models score 91 and 92 percent and you wonder if that gap is real.
- You want a fair accuracy comparison across many classifiers.
- Your classes are unbalanced and you need to handle that.
How to use it
- Add it to your AI once: download it and add the folder to Claude, ChatGPT or Codex as a skill.
- Ask for the job in your own words. Your AI opens the skill when it fits.
- It works best when your AI can run MATLAB with the Statistics and Machine Learning Toolbox.