Confusion Matrix Generator developed by Damiano Perri, Phd, Research Fellow at the University of Perugia

This page allows you to create confusion matrices and obtain a usable image in papers.
The resulting images are of high resolution and high quality.
The page also calculates many statistics of crucial importance during data analysis.
All calculations are done by your browser, locally, and no data is sent to the server.
No data you enter will be stored, retained or saved.

Draw confusion matrix for classes.

1) ↑↑ Set the number of classes and press "apply". ↑↑
2) Click with the mouse on the left cells to set the class names
3) Set absolute values
4) The matrix will automatically calculate the remaining values and you can save it as a PNG picture by pressing the "Download picture" button
Numerical precision:


Statistical measures:
Numerical precision:

List of the formulas used
Precision TP

(TP + FP)
also called Sensitivity

(TP + FN)
F1 Score 2*Precision*Recall

Precision + Recall
Macro-F1 average of F1 Scores
Weighted-F1 weighted-averaged of F1 Scores
Misclassification Rate # incorrect predictions

# total predictions

The code is available on GitHub