rfwtools.extractor.tsf.tsfresh_extractor

rfwtools.extractor.tsf.tsfresh_extractor(example, signals=None, query=None, impute_function=<function impute>, disable_progress_bar=True, n_jobs=0, default_fc_parameters=None, **kwargs)[source]

Use tsfresh to extract features specified.

This is a thin wrapper over tsfresh.feature_extraction.extraction.extract_features. See that method for more details.

Parameters:

  • example (Example) – The Example for which features are extracted

  • signals (Optional[List[str]]) – A list of signal names to extract features from. Default: combination of cavities 1-8 and waveforms = [‘GMES’, ‘GASK’, ‘CRFP’, ‘DETA2’]

  • query (Optional[str]) – Argument passed to the ex.event_df to filter data prior to feature extraction, e.g. “Time <= 0”.

  • impute_function (Optional[callable]) – The function used to impute missing values about the data

  • disable_progress_bar (bool) – Should the progress bar be displayed?

  • n_jobs (int) – The number of jobs should be run concurrently. Defaults to zero, which disables parallelization.

  • default_fc_parameters (Optional[dict]) – mapping of feature calculator names to parameters. If None, defaults to EfficientFCParameters(). See tsfresh.feature_extraction.extraction.extract_features for more details.

  • **kwargs – All other key word arguments are passed directly to tsfresh.extract_features

Return type:

DataFrame

Returns:

A DataFrame of the calculated features.