Utilities for Metropolis-Hastings random walk tasks (tomographer.mhrwtasks)

class tomographer.BinningAnalysis

A dummy class whose sole purpose is to expose the following constants to Python code.

CONVERGED

The error bar has converged over the different binning levels, the error bar can be considered reliable.

NOT_CONVERGED

The error bar has not converged over the different binning levels, and it should not be considered as reliable.

UNKNOWN_CONVERGENCE

The convergence of the error bar over the different binning levels is uknown, or couldn’t be determined. It may or may not be reliable.

Utilities for tasks running Metropolis-Hastings random walks. These classes shouldn’t be used directly; rather, corresponding instances are returned by, e.g., tomographer.tomorun.tomorun().

class tomographer.mhrwtasks.MHRandomWalkTaskResult

Bases: pybind11_builtins.pybind11_object

The result of an executed Metropolis-Hastings random walk task.

This class interfaces the corresponding C++ class Tomographer::MHRWTasks::MHRandomWalkTaskResult (the stats results type can be anything, represented in a Python object).

This class is pickleable.

stats_results

An object containing the results of the stats collected during the random walk. This can be any Python object.

Changed in version 5.0: Previously, this attribute was called stats_collector_result and necessarily had the type tomographer.ValueHistogramWithBinningMHRWStatsCollectorResult.

mhrw_params

The parameters of the executed random walk, as an MHRWParams instance.

acceptance_ratio

The average acceptance ratio of the random walk (excluding the thermalization sweeps).

tomographer.mhrwtasks.MHRandomWalkValueHistogramTaskResult

alias of tomographer.mhrwtasks.MHRandomWalkTaskResult