gluonts.time_feature package#

class gluonts.time_feature.Constant(*, value: float = 0.0)[source]#

Bases: pydantic.v1.main.BaseModel

Constant time feature using a predefined value.

value: float#

class gluonts.time_feature.SpecialDateFeatureSet(feature_names: typing.List[str], kernel_function: typing.Callable[[int], int] = <function indicator>)[source]#

Bases: object

Implements calculation of holiday features. The SpecialDateFeatureSet is applied on a pandas Series with Datetimeindex and returns a 2D array of the shape (len(dates), num_features), where num_features are the number of holidays.

Note that for lower than daily granularity the distance to the holiday is still computed on a per-day basis.

Example use:

>>> from gluonts.time_feature.holiday import (
...    squared_exponential_kernel,
...    SpecialDateFeatureSet,
...    CHRISTMAS_DAY,
...    CHRISTMAS_EVE
... )
>>> import pandas as pd
>>> sfs = SpecialDateFeatureSet([CHRISTMAS_EVE, CHRISTMAS_DAY])
>>> date_indices = pd.date_range(
...     start="2016-12-24",
...     end="2016-12-31",
...     freq='D'
... )
>>> sfs(date_indices)
array([[1., 0., 0., 0., 0., 0., 0., 0.],
       [0., 1., 0., 0., 0., 0., 0., 0.]])

Example use for using a squared exponential kernel:

>>> kernel = squared_exponential_kernel(alpha=1.0)
>>> sfs = SpecialDateFeatureSet([CHRISTMAS_EVE, CHRISTMAS_DAY], kernel)
>>> sfs(date_indices)
array([[1.00000000e+00, 3.67879441e-01, 1.83156389e-02, 1.23409804e-04,
        1.12535175e-07, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00],
       [3.67879441e-01, 1.00000000e+00, 3.67879441e-01, 1.83156389e-02,
        1.23409804e-04, 1.12535175e-07, 0.00000000e+00, 0.00000000e+00]])

gluonts.time_feature.day_of_month(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Day of month encoded as value between [-0.5, 0.5]

gluonts.time_feature.day_of_month_index(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Day of month encoded as zero-based index, between 0 and 11.

gluonts.time_feature.day_of_week(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Day of week encoded as value between [-0.5, 0.5]

gluonts.time_feature.day_of_week_index(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Day of week encoded as zero-based index, between 0 and 6.

gluonts.time_feature.day_of_year(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Day of year encoded as value between [-0.5, 0.5]

gluonts.time_feature.day_of_year_index(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Day of year encoded as zero-based index, between 0 and 365.

gluonts.time_feature.get_lags_for_frequency(freq_str: str, lag_ub: int = 1200, num_lags: Optional[int] = None, num_default_lags: int = 7) → List[int][source]#

Generates a list of lags that that are appropriate for the given frequency string.

By default all frequencies have the following lags: [1, 2, 3, 4, 5, 6, 7]. Remaining lags correspond to the same season (+/- delta) in previous k cycles. Here delta and k are chosen according to the existing code.

Parameters

freq_str – Frequency string of the form [multiple][granularity] such as “12H”, “5min”, “1D” etc.
lag_ub – The maximum value for a lag.
num_lags – Maximum number of lags; by default all generated lags are returned.
num_default_lags – The number of default lags; by default it is 7.

gluonts.time_feature.get_seasonality(freq: str, seasonalities={'B': 5, 'D': 1, 'H': 24, 'M': 12, 'ME': 12, 'Q': 4, 'QE': 4, 'S': 3600, 'T': 1440, 'W': 1, 'h': 24, 'min': 1440, 's': 3600}) → int[source]#

Return the seasonality of a given frequency:

>>> get_seasonality("2h")
12

gluonts.time_feature.hour_of_day(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Hour of day encoded as value between [-0.5, 0.5]

gluonts.time_feature.hour_of_day_index(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Hour of day encoded as zero-based index, between 0 and 23.

gluonts.time_feature.minute_of_hour(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Minute of hour encoded as value between [-0.5, 0.5]

gluonts.time_feature.minute_of_hour_index(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Minute of hour encoded as zero-based index, between 0 and 59.

gluonts.time_feature.month_of_year(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Month of year encoded as value between [-0.5, 0.5]

gluonts.time_feature.month_of_year_index(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Month of year encoded as zero-based index, between 0 and 11.

gluonts.time_feature.norm_freq_str(freq_str: str) → str[source]#

gluonts.time_feature.second_of_minute(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Second of minute encoded as value between [-0.5, 0.5]

gluonts.time_feature.second_of_minute_index(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Second of minute encoded as zero-based index, between 0 and 59.

gluonts.time_feature.time_features_from_frequency_str(freq_str: str) → List[Callable[[pandas.core.indexes.period.PeriodIndex], numpy.ndarray]][source]#

Returns a list of time features that will be appropriate for the given frequency string.

Parameters: freq_str – Frequency string of the form [multiple][granularity] such as “12H”, “5min”, “1D” etc.

gluonts.time_feature.week_of_year(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Week of year encoded as value between [-0.5, 0.5]

gluonts.time_feature.week_of_year_index(index: pandas.core.indexes.period.PeriodIndex) → numpy.ndarray[source]#: Week of year encoded as zero-based index, between 0 and 52.

gluonts.time_feature package#

Submodules#