Model Audit - Verification, Validation, and Error Analysis
Deprecated
Automated tests for model residuals
Checks for autocorrelation in target variable or in residuals
Checks for outliers
Checks for trend in residuals
Cook's distances
Create model evaluation explanation
Create Halfnormal Explanation
Create Model Performance Explanation
Create Model Residuals Explanation
Model Diagnostic Plots
Autocorrelation Function Plot
Autocorrelation of Residuals Plot
Influence of Observations Plot
Correlation of Model's Residuals Plot
Half-Normal plot
LIFT Chart
Principal Component Analysis of models
Predicted response vs Observed or Variable Values
Model Ranking Plot
Regression Error Characteristic Curves (REC)
Plot Residuals vs Observed, Fitted or Variable Values
Plot Boxplots of Residuals
Residual Density Plot
Precision-Recall Curve (PRC)
Regression Receiver Operating Characteristic (RROC)
Scale location plot
Two-sided Cumulative Distribution Function
Model Diagnostic Plots in D3 with r2d3 package.
Plot Autocorrelation Function in D3 with r2d3 package.
Autocorrelation Plot in D3 with r2d3 package.
Influence of observations Plot in D3 with r2d3 package.
Plot Half-Normal in D3 with r2d3 package.
Plot LIFT in D3 with r2d3 package.
Plot Prediction vs Target, Observed or Variable Values in D3 with r2d3...
Regression Error Characteristic Curves (REC) in D3 with r2d3 package.
Plot Residuals vs Observed, Fitted or Variable Values in D3 with r2d3 ...
Receiver Operating Characteristic (ROC) in D3 with r2d3 package.
Regression Receiver Operating Characteristic (RROC) in D3 with r2d3 pa...
Scale Location Plot in D3 with r2d3 package.
Prints Model Cook's Distances Summary
Prints Model Evaluation Summary
Prints Model Halfnormal Summary
Prints Model Performance Summary
Prints Model Residual Summary
Prints of Models Scores
Model Scores computations
Accuracy
Area Under ROC Curve (AUC)
Area under precision-recall curve
Score based on Cooks Distance
Durbin-Watson Score
F1 Score
Gini Coefficient
Half-Normal Score
Mean Absolute Error
Mean Square Error
One minus accuracy
One minus Area Under ROC Curve (AUC)
One Minus area under precision-recall curve
One Minus F1 Score
One minus Gini Coefficient
One Minus Precision
One minus recall
One minus specificity
Peak Score
Precision
R-squared
Area Over the Curve for REC Curves
Recall
Root Mean Square Error
Area Over the Curve for RROC Curves
Runs Score
Specificity
Provides an easy to use unified interface for creating validation plots for any model. The 'auditor' helps to avoid repetitive work consisting of writing code needed to create residual plots. This visualizations allow to asses and compare the goodness of fit, performance, and similarity of models.
Useful links