sierra.plugins.compare.graphs.preprocess#

Preprocess inter-experiment outputs for stage 5.

Basically, gather statistics generated from controllers for graph generation in previous stages into the correct files(s) for comparison.

Classes#

IntraExpPreparer

Collate generated stats from previous stages into files(s) for comparison.

Module Contents#

class sierra.plugins.compare.graphs.preprocess.IntraExpPreparer(ipath_stem: pathlib.Path, ipath_leaf: str, opath_stem: pathlib.Path, criteria: sierra.core.variables.batch_criteria.XVarBatchCriteria)[source]#

Collate generated stats from previous stages into files(s) for comparison.

for_cc(controller: str, opath_leaf: str, index: int, inc_exps: str | None) None[source]#

Take batch-level dataframes and creates a new dataframe.

Has:

  • Experiment names for the index.

  • Controller names as column names (guaranteed to be unique, since that's what we are comparing).

  • df[controller] columns as timeslices across columns (i.e., across experiments in the batch) in the source dataframe.

for_sc(scenario: str, opath_leaf: str, index: int, inc_exps: str | None) None[source]#

Take batch-level dataframes and creates a new dataframe.

Has:

  • Experiment names for the index.

  • Scenario names as column names (guaranteed to be unique, since that's what we are comparing).

  • df[scenario] columns as timeslices across columns (i.e., across experiments in the batch) in the source dataframe.