AI Blocks › Build and ship software
Model decision explainer
Give it a trained model. Get which inputs matter, why it made each call, and where it goes wrong.
Model decision explainer 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
A clear answer to why your trained model says what it says. You see which inputs matter most, why it gave one case a certain score, and where it keeps getting things wrong. The skill works in MATLAB on models trained on rows and columns of data. It matches your question to a method: ranking the inputs, showing how one input moves the result, explaining a single prediction, studying the errors, or comparing groups of people. It works with built-in models and with custom ones you wrap in a prediction function. It is not for images or text models, and it does not train or tune anything.
You give it
- A trained model on table data
- The data it was trained or tested on
- The question, such as why a case or which inputs
- Groups to compare (optional)
- MATLAB with the Statistics and Machine Learning Toolbox
- Spreadsheet (optional)
You get
- A ranked list of the inputs that matter
- A plain explanation of single predictions
- Charts of how one input changes the answer
- A review of the cases it got wrong
- A comparison across groups
Use it when
- Your manager asks which factors drive the model.
- One customer got a bad score and you must explain why.
- The model does worse on one group and you need to see how much.
- You want to know what small change would flip a prediction.
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.