Attend the Women in ML Symposium on December 7 Register now

Module: tfp.sts

Stay organized with collections Save and categorize content based on your preferences.

Framework for Bayesian structural time series models.

See the blog post for an introductory example.

Classes

class AdditiveStateSpaceModel: A state space model representing a sum of component state space models.

class Autoregressive: Formal representation of an autoregressive model.

class AutoregressiveIntegratedMovingAverage: Represents an autoregressive integrated moving-average (ARIMA) model.

class AutoregressiveMovingAverageStateSpaceModel: State space model for an autoregressive moving average process.

class AutoregressiveStateSpaceModel: State space model for an autoregressive process.

class ConstrainedSeasonalStateSpaceModel: Seasonal state space model with effects constrained to sum to zero.

class DynamicLinearRegression: Formal representation of a dynamic linear regresson model.

class DynamicLinearRegressionStateSpaceModel: State space model for a dynamic linear regression from provided covariates.

class IntegratedStateSpaceModel: Integrates (/cumsums) a noise-free state space model.

class LinearRegression: Formal representation of a linear regression from provided covariates.

class LocalLevel: Formal representation of a local level model.

class LocalLevelStateSpaceModel: State space model for a local level.

class LocalLinearTrend: Formal representation of a local linear trend model.

class LocalLinearTrendStateSpaceModel: State space model for a local linear trend.

class MaskedTimeSeries: Named tuple encoding a time series Tensor and optional missingness mask.

class MissingValuesTolerance: MissingValuesTolerance(overall_fraction, fraction_low_missing_number, fraction_high_missing_number, low_missing_number, high_missing_number)

class Seasonal: Formal representation of a seasonal effect model.

class SeasonalStateSpaceModel: State space model for a seasonal effect.

class SemiLocalLinearTrend: Formal representation of a semi-local linear trend model.

class SemiLocalLinearTrendStateSpaceModel: State space model for a semi-local linear trend.

class SmoothSeasonal: Formal representation of a smooth seasonal effect model.

class SmoothSeasonalStateSpaceModel: State space model for a smooth seasonal effect.

class SparseLinearRegression: Formal representation of a sparse linear regression.

class StructuralTimeSeries: Base class for structural time series models.

class Sum: Sum of structural time series components.

Functions

build_factored_surrogate_posterior(...): Build a variational posterior that factors over model parameters.

build_factored_surrogate_posterior_stateless(...): Returns stateless functions for building a variational posterior.

decompose_by_component(...): Decompose an observed time series into contributions from each component.

decompose_forecast_by_component(...): Decompose a forecast distribution into contributions from each component.

fit_with_hmc(...): Draw posterior samples using Hamiltonian Monte Carlo (HMC).

forecast(...): Construct predictive distribution over future observations.

impute_missing_values(...): Runs posterior inference to impute the missing values in a time series. (deprecated argument values)

moments_of_masked_time_series(...): Compute mean and variance, accounting for a mask.

one_step_predictive(...): Compute one-step-ahead predictive distributions for all timesteps. (deprecated argument values)

regularize_series(...): Infers frequency and makes an irregular time series regular.

sample_uniform_initial_state(...): Initialize from a uniform [-2, 2] distribution in unconstrained space. (deprecated)