simrun ❭ modular_reduced_model_inference ❭ reduced_model ❭ Rm

Rm¶

class simrun.modular_reduced_model_inference.reduced_model.Rm(name, db, tmin=None, tmax=None, width=None, selected_indices=None)¶

Reduced model class.

This is the top-level class for running reduced model inference. It provides an interface to add data extractors and strategies, and to run the optimization.

Parameters:¶

• name (str) – Name of the reduced model

• db (str) – Path to the database

• tmin (float) – Minimum time for the simulation

• tmax (float) – Maximum time for the simulation

• width (float) – Width of the temporal kernel. This will be used as the \(\tau\) domain in e.g. RaisedCosineBasis

• selected_indices (list) – List/nested list of integer indices for selected simulation trials

Attributes:¶

name¶

Name of the reduced model

Type:¶

str

db¶

Path to the database

Type:¶

str

tmin¶

Minimum time for the simulation

Type:¶

float

tmax¶

Maximum time for the simulation

Type:¶

float

width¶

Width of the temporal kernel. This will be used as the domain e.g. the \(\\tau\) domain in RaisedCosineBasis

Type:¶

float

n_trials¶

Amount of trials

Type:¶

int

data_extractors¶

Dictionary of _DataExtractor objects

Type:¶

dict

strategies¶

Dictionary of Strategy objects

Type:¶

dict

Data¶

Data view object

Type:¶

DataView

DataSplitEvaluation¶

Data split evaluation object

Type:¶

DataSplitEvaluation

selected_indices¶

List/nested list of integer indices for selected simulation trials

Type:¶

list

results_remote¶

Flag that keeps track whether results are stored locally or on a remote scheduler.

Type:¶

bool

Methods:¶

add_data_extractor(name, data_extractor, setup)	Add a data extractor to the reduced model.
add_strategy(strategy, setup, view)	Add a strategy to the reduced model.
get_n_trials()	Get the amount of trials.
extract(name)	Extract data using the data extractor.
run(client, n_workers, strategy_selection)	Run one or more strategies on the data.
_gather_results(client)	Fetch the solver results from the dask scheduler.
get_results(client)	Get the results of the optimization.