Machine Learning Model Performance Metrics: RMSE, MAE, F1-Score, and Confidence Intervals

A summary of performance metrics for a machine learning model. The 9.5 KB Excel file contains results for global RMSE, MAE, TPR, PPV, TNR, NPV, F1-score, and the 95% Wilson confidence interval for the TPR of each class. It was authored by Max Wiedemann and last updated on May 5, 2026.

Use Cases

• Benchmark model performance based on reported global RMSE and MAE values
• Evaluate classification model sensitivity and specificity based on TPR and TNR metrics
• Assess model precision and recall trade-offs using the reported PPV and F1-scores
• Analyze the statistical reliability of true positive rates using the provided 95% Wilson confidence intervals

Strengths

• Includes a standard suite of evaluation metrics (RMSE, MAE, TPR, PPV, TNR, NPV, F1-score) for comprehensive assessment.
• Provides 95% Wilson confidence intervals for TPR, offering a measure of statistical uncertainty.
• Released under the permissive CC-BY-4.0 license, allowing for broad reuse.

Limitations

• The dataset is very small at 9.5 KB, indicating a limited scope, likely a summary for a single model or experiment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for certain analyses.

Provenance

Source

figshare

Freshness

Last updated 2026-05-05 17:31:03