Metrics Module

UnaryMetric

class metrics.UnaryMetric

Abstract Base Class from which all unary metrics inherit.

run(target_dataset)

Run the metric for a given target dataset.

Parameters:target_dataset – The dataset on which the current metric will be run. Returns:The result of evaluating the metric on the target_dataset.

BinaryMetric

class metrics.BinaryMetric

Abstract Base Class from which all binary metrics inherit.

run(ref_dataset, target_dataset)

Run the metric for the given reference and target datasets.

Parameters:

• ref_dataset (Dataset) – The Dataset to use as the reference dataset when running the evaluation.
• target_dataset – The Dataset to use as the target dataset when running the evaluation.

Returns:

The result of evaluation the metric on the reference and target dataset.

Bias

class metrics.Bias

Calculate the bias between a reference and target dataset.

run(ref_dataset, target_dataset)

Calculate the bias between a reference and target dataset.

Note

Overrides BinaryMetric.run()

Parameters:

• ref_dataset (Dataset.) – The reference dataset to use in this metric run.
• target_dataset (Dataset.) – The target dataset to evaluate against the reference dataset in this metric run.

Returns:

The difference between the reference and target datasets.

Return type:

Numpy Array

TemporalStdDev

class metrics.TemporalStdDev

Calculate the standard deviation over the time.

run(target_dataset)

Calculate the temporal std. dev. for a datasets.

Note

Overrides UnaryMetric.run()

Parameters:target_dataset (Dataset) – The target_dataset on which to calculate the temporal standard deviation. Returns:The temporal standard deviation of the target dataset Return type:Numpy Array