evaluation — Turi Create API 6.4.1 documentation (original) (raw)

The evaluation module includes performance metrics to evaluate machine learning models. The metrics can be broadly categorized as: - Classification metrics - Regression metrics

The evaluation module supports the following classification metrics: - accuracy - auc - confusion_matrix - f1_score - fbeta_score - log_loss - precision - recall - roc_curve

The evaluation module supports the following regression metrics: - rmse - max_error

classifier metrics¶

accuracy	Compute the accuracy score; which measures the fraction of predictions made by the classifier that are exactly correct.
auc	Compute the area under the ROC curve for the given targets and predictions.
confusion_matrix	Compute the confusion matrix for classifier predictions.
f1_score	Compute the F1 score (sometimes known as the balanced F-score or F-measure).
fbeta_score	Compute the F-beta score.
log_loss	Compute the logloss for the given targets and the given predicted probabilities.
precision	Compute the precision score for classification tasks.
recall	Compute the recall score for classification tasks.
roc_curve	Compute an ROC curve for the given targets and predictions.

regression metrics¶

max_error	Compute the maximum absolute deviation between two SArrays.
rmse	Compute the root mean squared error between two SArrays.