Catch22

Quickstart

from sktime.transformations.catch22 import Catch22

estimator = Catch22(features: int | str | list [int | str ]='all', catch24: bool=False, outlier_norm: bool=False, replace_nans: bool=False, col_names: str='range')

Parameters(6)

featuresint/str or List of int/str, optional, default=”all”

The Catch22 features to extract by feature index, feature name as a str or as a list of names or indices for multiple features. If “all”, all features are extracted. Valid features and their corresponding short feature names are as follows:

{ “DN_HistogramMode_5”: “mode_5”, “DN_HistogramMode_10”: “mode_10”, “SB_BinaryStats_diff_longstretch0”: “stretch_decreasing”, “DN_OutlierInclude_p_001_mdrmd”: “outlier_timing_pos”, “DN_OutlierInclude_n_001_mdrmd”: “outlier_timing_neg”, “CO_f1ecac”: “acf_timescale”, “CO_FirstMin_ac”: “acf_first_min”, “SP_Summaries_welch_rect_area_5_1”: “centroid_freq”, “SP_Summaries_welch_rect_centroid”: “low_freq_power”, “FC_LocalSimple_mean3_stderr”: “forecast_error”, “CO_trev_1_num”: “trev”, “CO_HistogramAMI_even_2_5”: “ami2”, “IN_AutoMutualInfoStats_40_gaussian_fmmi”: “ami_timescale”, “MD_hrv_classic_pnn40”: “high_fluctuation”, “SB_BinaryStats_mean_longstretch1”: “stretch_high”, “SB_MotifThree_quantile_hh”: “entropy_pairs”, “FC_LocalSimple_mean1_tauresrat”: “whiten_timescale”, “CO_Embed2_Dist_tau_d_expfit_meandiff”: “embedding_dist”, “SC_FluctAnal_2_dfa_50_1_2_logi_prop_r1”: “dfa”, “SC_FluctAnal_2_rsrangefit_50_1_logi_prop_r1”: “rs_range”, “SB_TransitionMatrix_3ac_sumdiagcov”: “transition_matrix”, “PD_PeriodicityWang_th0_01”: “periodicity”,

} Additionally, if catch24 is true, two additional features are available: {

“DN_Mean”: “mean”, “DN_Spread_Std”: “std”,

} The overview of these features is available at: https://time-series-features.gitbook.io/catch22-features/feature-overview-table

catch24bool, optional, default=False

Extract the mean and standard deviation as well as the 22 Catch22 features if true. If a List of specific features to extract is provided, “Mean” and/or “StandardDeviation” must be added to the List to extract these features.

outlier_normbool, optional, default=False

Normalise each series during the two outlier Catch22 features: _DN_OutlierInclude_p_001_mdrmd and _DN_OutlierInclude_n_001_mdrmd, which can take a while to process for large values.

replace_nansbool, optional, default=True

Replace NaN or inf values from the Catch22 transform with 0.

col_namesstr, one of {“range”, “int_feat”, “str_feat”, “short_str_feat”, “auto”},

optional, default=”range”

The type of column names to return. If “range”, column names will be a regular range of integers, as in a RangeIndex. If “int_feat”, column names will be the integer feature indices, as defined in pycatch22. If “str_feat”, column names will be the string feature names. If “short_str_feat”, column names will be the short string feature names as defined in pycatch22. If “auto”, column names will be the same as defined in features.

References