ds-utils-metrics

To evaluate machine learning model performance, you can generate and annotate plots like confusion matrices, ROC curves, and precision-recall curves using sklearn-compatible APIs for rapid model diagnostics.

Plotting a confusion matrix allows you to visually interpret misclassifications by annotating prediction errors, helping you quickly diagnose where your classification model is failing on real-world datasets.

You can generate ROC and precision-recall curves to assess your classification model's discrimination capabilities, providing annotated visualizations that integrate with common ML workflows.

Yes, you can analyze probability calibration and prediction errors using dedicated charts that help you understand how well your model's predicted probabilities align with actual outcomes.

Yes, it seamlessly integrates with sklearn-compatible models to plot learning curves, allowing you to assess learning dynamics and quickly diagnose issues across various classification and regression tasks.

Plot learning curves during model evaluation to assess learning dynamics and identify issues like overfitting or underfitting, ensuring your model improves consistently as training data increases.