catch22: CAnonical Time-series CHaracteristics (for NodeJS)

A fork of catch22 to support Nodejs. Just npm i catch22 and you are ready.

Example

const catch22 = require('catch22');

const data = [1,2,3,4,5,6,7,8,9,10];

const features22 = catch22.catch22(data);
const features24 = catch22.catch24(data);

const mode5 = catch22.DN_HistogramMode_5(data);

console.log('features22: ', features22);
console.log('features24: ', features24);
console.log('DN_HistogramMode_5: ', mode5);

For more information, check the official 📗catch22 documentation

Functions to short names map:

Following the feature descriptions:

Function name	Short name
DN_HistogramMode_5	mode_5
DN_HistogramMode_10	mode_10
CO_f1ecac	acf_timescale
CO_FirstMin_ac	acf_first_min
CO_HistogramAMI_even_2_5	ami2
CO_trev_1_num	trev
MD_hrv_classic_pnn40	high_fluctuation
SB_BinaryStats_mean_longstretch1	stretch_high
SB_TransitionMatrix_3ac_sumdiagcov	transition_matrix
PD_PeriodicityWang_th0_01	periodicity
CO_Embed2_Dist_tau_d_expfit_meandiff	embedding_dist
IN_AutoMutualInfoStats_40_gaussian_fmmi	ami_timescale
FC_LocalSimple_mean1_tauresrat	whiten_timescale
DN_OutlierInclude_p_001_mdrmd	outlier_timing_pos
DN_OutlierInclude_n_001_mdrmd	outlier_timing_neg
SP_Summaries_welch_rect_area_5_1	centroid_freq
SB_BinaryStats_diff_longstretch0	stretch_decreasing
SB_MotifThree_quantile_hh	entropy_pairs
SC_FluctAnal_2_rsrangefit_50_1_logi_prop_r1	rs_range
SC_FluctAnal_2_dfa_50_1_2_logi_prop_r1	dfa
SP_Summaries_welch_rect_centroid	low_freq_power
FC_LocalSimple_mean3_stderr	forecast_error
DN_Mean	mean
DN_Spread_Std	SD