interpreTS.core package

Subpackages

• interpreTS.core.features package
  • Submodules
  • interpreTS.core.features.feature_length module
    • calculate_length()
  • interpreTS.core.features.feature_mean module
    • calculate_mean()
  • interpreTS.core.features.feature_peak module
    • calculate_peak()
  • interpreTS.core.features.feature_spikeness module
    • calculate_spikeness()
  • interpreTS.core.features.feature_std_1st_der module
    • calculate_std_1st_der()
  • interpreTS.core.features.feature_trough module
    • calculate_trough()
  • interpreTS.core.features.feature_variance module
    • calculate_variance()
  • interpreTS.core.features.histogram_dominant module
    • calculate_dominant()
  • interpreTS.core.features.seasonality_strength module
    • calculate_seasonality_strength()
  • interpreTS.core.features.trend_strength module
    • calculate_trend_strength()
  • Module contents

Submodules

interpreTS.core.feature_extractor module

class interpreTS.core.feature_extractor.FeatureExtractor(features=None, feature_params=None, window_size=5, stride=1, id_column=None, sort_column=None)

Bases: object

static available_features()

Returns a list of all available features.

Returns

list

List of feature names.

extract_features(data)

Extract features from a time series dataset.

Parameters

datapd.DataFrame or pd.Series

The time series data for which features are to be extracted.

Returns

pd.DataFrame

A DataFrame containing calculated features for each window.

class interpreTS.core.feature_extractor.Features

Bases: object

ABSOLUTE_ENERGY = 'absolute_energy'

BINARIZE_MEAN = 'binarize_mean'

CALCULATE_SEASONALITY_STRENGTH = 'seasonality_strength'

CROSSING_POINTS = 'crossing_points'

ENTROPY = 'entropy'

FLAT_SPOTS = 'flat_spots'

LENGTH = 'length'

MEAN = 'mean'

MISSING_POINTS = 'missing_points'

PEAK = 'peak'

SPIKENESS = 'spikeness'

STABILITY = 'stability'

STD_1ST_DER = 'std_1st_der'

TROUGH = 'trough'

VARIANCE = 'variance'

interpreTS.core.time_series_data module

class interpreTS.core.time_series_data.TimeSeriesData(data)

Bases: object

A class to manage and process time series data.

resample(interval)

Resample the time series data to a specified interval.

Parameters

intervalstr

The interval to resample the data, e.g., ‘D’ for daily, ‘H’ for hourly.

Returns

TimeSeriesData

A new TimeSeriesData object with resampled data.

Examples

>>> data = pd.Series([1, 2, 3, 4, 5], index=pd.date_range("2023-01-01", periods=5, freq="D"))
>>> ts_data = TimeSeriesData(data)
>>> resampled_data = ts_data.resample("2D")

split(train_size=0.7)

Split the time series data into training and test sets.

Parameters

train_sizefloat, optional

The proportion of the data to use for training, by default 0.7.

Returns

tuple of TimeSeriesData

A tuple containing the training and test sets as TimeSeriesData objects.

Examples

>>> data = pd.Series([1, 2, 3, 4, 5])
>>> ts_data = TimeSeriesData(data)
>>> train, test = ts_data.split(0.6)

Module contents