← Home GitHub

ML Surprise Test - 3

Data Visualizations

A compact gallery of model and dataset outputs generated from the language detection pipeline.

Model performance summary

Accuracy Summary

Overall model accuracy, language count, and dataset size.

Open Image
Dataset language distribution

Language Distribution

Number of text samples available for each language.

Open Image
Normalized confusion matrix

Confusion Matrix

Actual versus predicted language classes, normalized for easier comparison.

Open Image
Per-language F1 score

Language Performance

Per-language F1 score for comparing classification quality across classes.

Open Image