locan.analysis.pipeline#

Building an analysis pipeline.

Pipeline refers to sequential analysis steps that are applied to a single LocData object. An analysis pipeline here includes true piped analysis, where a preliminary result serves as input to the next analysis step, but also workflows that provide different results in parallel.

A batch process is a procedure for running a pipeline over multiple LocData objects while collecting and combing results.

This module provides a class Pipeline to combine the analysis procedure, parameters and results in a single pickleable object.

Classes

Pipeline(computation[, meta])

The base class for a specialized analysis pipeline to be used on LocData objects.

Functions

computation_test(self[, locdata, parameter])

A pipeline definition for testing.