Overall Accuracy
{{ "%.2f"|format((report_data.best_model.metrics.test_accuracy|default(0)) * 100) }}%
Top 1 Accuracy
Precision
{{ "%.2f"|format((report_data.best_model.metrics.test_precision|default(0)) * 100) }}%
High precision
Recall
{{ "%.2f"|format((report_data.best_model.metrics.test_recall|default(0)) * 100) }}%
Good coverage
F1 Score
{{ "%.3f"|format(report_data.best_model.metrics.test_f1_score|default(0)) }}
Balanced

Confusion Matrix

Classification Report

Class Precision Recall F1-Score Support
Macro Avg 0.92 0.91 0.915 1000
Weighted Avg 0.93 0.92 0.925 -

ROC Curves

AUC-ROC: 0.98

Precision-Recall Curves

AUC-PR: 0.95

Per-Class Performance Metrics

Top Confusion Pairs

Error Distribution

Threshold Analysis

0.50

Calibration Plot

Confidence Distribution

Comprehensive Performance Summary

Metric Category Metric Name Teacher Model Student Model Difference Status
Classification Accuracy 95.2% 92.8% -2.4% Good
Precision 93.1% 91.5% -1.6% Good
Recall 94.0% 92.1% -1.9% Good
F1 Score 0.935 0.918 -0.017 Good
Probabilistic AUC-ROC 0.982 0.965 -0.017 Good
AUC-PR 0.956 0.938 -0.018 Good
Log Loss 0.142 0.198 +0.056 Marginal
Efficiency Inference Time 50ms 15ms -70% Excellent
Memory Usage 1024MB 256MB -75% Excellent
Model Size 400MB 100MB -75% Excellent