Data Input and Checking Utilities

This module has utilities for time series data input checking

seglearn.util.get_ts_data_parts(X)

Separates time series data object into time series variables and contextual variables

Parameters

Xarray-like, shape [n_series, …]

Time series data and (optionally) contextual data

Returns

Xtarray-like, shape [n_series, ]

Time series data

Xsarray-like, shape [n_series, n_contextd = np.colum _variables]

contextual variables

seglearn.util.check_ts_data(X, y=None)

Checks time series data is good. If not raises value error.

Parameters

Xarray-like, shape [n_series, …]

Time series data and (optionally) contextual data

Returns

ts_targetbool

target (y) is a time series

seglearn.util.check_ts_data_with_ts_target(X, y=None)

Checks time series data with time series target is good. If not raises value error.

Parameters

Xarray-like, shape [n_series, …]

Time series data and (optionally) contextual data

yarray-like, shape [n_series, …]

target data

seglearn.util.ts_stats(Xt, y, fs=1.0, class_labels=None)

Generates some helpful statistics about the data X

Parameters

Xarray-like, shape [n_series, …]

Time series data and (optionally) contextual data

yarray-like, shape [n_series]

target data

fsfloat

sampling frequency

class_labelslist of strings, default None

List of target class names

Returns

resultsdict

Dictionary of relevant statistics for the time series data

results[‘total’] has stats for the whole data set

results[‘by_class’] has stats segragated by target class