stateinterpreter.ml.prepare_training_dataset

class stateinterpreter.ml.prepare_training_dataset(descriptors, states_labels, n_configs, regex_filter=None, states_subset=None)

Bases:

Sample points from trajectory

Args:

n_configs (int): number of points to sample for each metastable state regex_filter (str, optional): regex to filter the features. Defaults to ‘.*’. states_subset (list, optional): list of integers corresponding to the metastable states to sample. Defaults to None take all states. states_names (list, optional): list of strings corresponding to the name of the states. Defaults to None.

Returns:

(configurations, labels), features_names, states_names

__init__(**kwargs)