gluonts.time_feature.holiday 模块

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

基类: object

实现节假日特征的计算。SpecialDateFeatureSet 应用于具有 Datetimeindex 的 pandas Series，并返回形状为 (len(dates), num_features) 的二维数组，其中 num_features 是节假日的数量。

请注意，对于低于每日粒度的数据，节假日的距离仍然按天计算。

使用示例

>>> 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.]])

使用平方指数核函数的使用示例

>>> 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.holiday.distance_to_holiday(holiday)

gluonts.time_feature.holiday.exponential_kernel(alpha=1.0, tol=1e-09)

gluonts.time_feature.holiday.indicator(distance)

gluonts.time_feature.holiday.squared_exponential_kernel(alpha=1.0, tol=1e-09)