components model_performance_compute_metrics

Model Performance - Compute Metrics

model_performance_compute_metrics

Overview

Compute model performance metrics leveraged by the model performance monitor.

Version: 0.0.19

View in Studio: https://ml.azure.com/registries/azureml/components/model_performance_compute_metrics/version/0.0.19

Inputs

Name	Description	Type	Default	Optional	Enum
task	Task type	string	tabular-classification		['tabular-classification', 'tabular-regression']
baseline_data_target_column	Column name which contains ground truths in provided input for baseline_data. (Optional if we have one column name.)	string
production_data	The mltable generated by data joiner to join ground truth and prediction data in a table	mltable
production_data_target_column	Column name which contains predictions in provided input for production_data_target_column. (Optional if we have one column name.)	string
regression_rmse_threshold		number		True
regression_meanabserror_threshold		number		True
classification_precision_threshold		number		True
classification_accuracy_threshold		number		True
classification_recall_threshold		number		True

Outputs

Name	Description	Type
signal_metrics		mltable