PredictionDrift.prediction_drift#
- PredictionDrift.prediction_drift(train_prediction, test_prediction, model_classes, with_display, proba_drift, cat_plot) CheckResult[source]#
Calculate prediction drift.
- Args:
- train_predictionnp.ndarray
train prediction or probabilities
- test_predictionnp.ndarray
test prediction or probabilities
- model_classeslist
list of model classes
- with_displaybool
flag for displaying the prediction distribution graph
- proba_driftbool
flag for computing drift on the probabilities rather than the predicted labels
- cat_plotbool
flag for plotting the distribution of the predictions as a categorical plot
- CheckResult
value: drift score. display: prediction distribution graph, comparing the train and test distributions.