gluonts.mx.model.deepstate.issm module#

class gluonts.mx.model.deepstate.issm.CompositeISSM(seasonal_issms: List[gluonts.mx.model.deepstate.issm.SeasonalityISSM], add_trend: bool = True)[source]#

Bases: gluonts.mx.model.deepstate.issm.ISSM

DEFAULT_ADD_TREND: bool = True#
classmethod get_from_freq(freq: str, add_trend: bool = True)[source]#
get_issm_coeff(features: Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol]) → Tuple[Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol], Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol], Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol]][source]#
latent_dim() → int[source]#
output_dim() → int[source]#
time_features() → List[Callable[[pandas.core.indexes.period.PeriodIndex], numpy.ndarray]][source]#
gluonts.mx.model.deepstate.issm.DayOfWeekSeasonalISSM()[source]#
gluonts.mx.model.deepstate.issm.HourOfDaySeasonalISSM()[source]#
class gluonts.mx.model.deepstate.issm.ISSM[source]#

Bases: object

An abstract class for providing the basic structure of Innovation State Space Model (ISSM).

The structure of ISSM is given by

  • dimension of the latent state

  • transition and innovation coefficients of the transition model

  • emission coefficient of the observation model

emission_coeff(features: Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol]) → Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol][source]#
get_issm_coeff(features: Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol]) → Tuple[Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol], Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol], Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol]][source]#
innovation_coeff(features: Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol]) → Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol][source]#
latent_dim() → int[source]#
output_dim() → int[source]#
time_features() → List[Callable[[pandas.core.indexes.period.PeriodIndex], numpy.ndarray]][source]#
transition_coeff(features: Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol]) → Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol][source]#
class gluonts.mx.model.deepstate.issm.LevelISSM[source]#

Bases: gluonts.mx.model.deepstate.issm.ISSM

emission_coeff(feature: Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol]) → Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol][source]#
innovation_coeff(feature: Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol]) → Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol][source]#
latent_dim() → int[source]#
output_dim() → int[source]#
time_features() → List[Callable[[pandas.core.indexes.period.PeriodIndex], numpy.ndarray]][source]#
transition_coeff(feature: Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol]) → Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol][source]#
class gluonts.mx.model.deepstate.issm.LevelTrendISSM[source]#

Bases: gluonts.mx.model.deepstate.issm.LevelISSM

latent_dim() → int[source]#
output_dim() → int[source]#
time_features() → List[Callable[[pandas.core.indexes.period.PeriodIndex], numpy.ndarray]][source]#
transition_coeff(feature: Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol]) → Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol][source]#
gluonts.mx.model.deepstate.issm.MinuteOfHourSeasonalISSM()[source]#
gluonts.mx.model.deepstate.issm.MonthOfYearSeasonalISSM()[source]#
class gluonts.mx.model.deepstate.issm.SeasonalityISSM(num_seasons: int, time_feature: Callable[[pandas.core.indexes.period.PeriodIndex], numpy.ndarray])[source]#

Bases: gluonts.mx.model.deepstate.issm.LevelISSM

Implements periodic seasonality which is entirely determined by the period num_seasons.

emission_coeff(feature: Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol]) → Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol][source]#
innovation_coeff(feature: Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol]) → Union[mxnet.ndarray.ndarray.NDArray, mxnet.symbol.symbol.Symbol][source]#
latent_dim() → int[source]#
output_dim() → int[source]#
time_features() → List[Callable[[pandas.core.indexes.period.PeriodIndex], numpy.ndarray]][source]#
gluonts.mx.model.deepstate.issm.WeekOfYearSeasonalISSM()[source]#