MetricTask

class lsst.verify.tasks.MetricTask(**kwargs)

Bases: PipelineTask

A base class for tasks that compute one metric from input datasets.

Parameters

*args **kwargs

Constructor parameters are the same as for lsst.pipe.base.PipelineTask.

Notes

In general, both the MetricTask’s metric and its input data are configurable. Metrics may be associated with a data ID at any level of granularity, including repository-wide.

Like lsst.pipe.base.PipelineTask, this class should be customized by overriding run and by providing a lsst.pipe.base.connectionTypes.Input for each parameter of run. For requirements that are specific to MetricTask, see run.

Methods Summary

run(**kwargs)	Run the MetricTask on in-memory data.
runQuantum(butlerQC, inputRefs, outputRefs)	Do Butler I/O to provide in-memory objects for run.

Methods Documentation

abstract run(**kwargs)

Run the MetricTask on in-memory data.

Parameters

**kwargs

Keyword arguments matching the inputs given in the class config; see lsst.pipe.base.PipelineTask.run for more details.

Returns

structlsst.pipe.base.Struct

A Struct containing at least the following component:

• measurement: the value of the metric (lsst.verify.Measurement or None). This method is not responsible for adding mandatory metadata (e.g., the data ID); this is handled by the caller. None may be used to indicate that a metric is undefined or irrelevant instead of raising NoWorkFound.

Raises

lsst.verify.tasks.MetricComputationError

Raised if an algorithmic or system error prevents calculation of the metric. Examples include corrupted input data or unavoidable exceptions raised by analysis code. The MetricComputationError should be chained to a more specific exception describing the root cause.

Not having enough data for a metric to be applicable is not an error, and should raise NoWorkFound (see below) instead of this exception.

lsst.pipe.base.NoWorkFound

Raised if the metric is ill-defined or otherwise inapplicable to the data. Typically this means that the pipeline step or option being measured was not run.

runQuantum(butlerQC, inputRefs, outputRefs)

Do Butler I/O to provide in-memory objects for run.

This specialization of runQuantum performs error-handling specific to MetricTasks.