API Reference

Forecasters

High-level interfaces for fitting and forecasting.

forecaster.Forecaster: Fit a forecasting model with stochastic variational inference.
forecaster.HMCForecaster: Fit a forecasting model with MCMC (NUTS by default).
forecaster.PathfinderForecaster: Fit a forecasting model with BlackJAX Pathfinder variational inference.

Building forecasting models (object-oriented and functional).

forecaster.ForecastingModel: Abstract base class for forecasting models.
functional.models.forecasting_model(): Build a NumPyro model from a functional model body.

Pure functional primitives for the train/forecast split.

functional.models.Horizon: The train/forecast split for a single model call.
functional.models.time_series(): Sample a time-varying latent over the full horizon.
functional.models.markov_time_series(): Sample a Markov (state-space) latent over the full horizon.
functional.models.predict(): Register the observation/forecast sites for the model.
functional.models.predict_glm(): Register GLM-style observation/forecast sites from a latent predictor.

Optimizer/guide/kernel resolution and the SVI and MCMC fit entry points.

functional.svi.resolve_optimizer(): Normalize an optimizer specification into a NumPyro optimizer.
functional.svi.resolve_guide(): Normalize a guide specification against model.
functional.svi.fit_svi(): Fit a forecasting model with stochastic variational inference.
functional.svi.SVIFit: The result of fitting a forecasting model with SVI.
functional.mcmc.resolve_kernel(): Normalize a kernel specification.
functional.mcmc.fit_mcmc(): Fit a forecasting model with MCMC.
functional.mcmc.MCMCFit: The result of fitting a forecasting model with MCMC.

Drawing posterior samples and generating forecasts and in-sample predictions.

functional.posterior.draw_posterior(): Draw num_samples posterior samples of the latent sites from a fit.
functional.prediction.forecast(): Sample forecasts for the steps in [t, duration) from a posterior.
functional.prediction.predict_in_sample(): Sample the in-sample posterior predictive of the obs site.

Rolling-window backtesting and forecast metrics.

evaluate.backtest(): Backtest a forecasting model on a moving window of (train, test) data.
evaluate.backtest_vectorized(): Rolling-window backtest with all windows fitted in one vmapped SVI run.
evaluate.BacktestResult: Per-window result of a backtest() run.
evaluate.VectorizedBacktestResult: Result of a backtest_vectorized() run (all windows at once).
evaluate.evaluate_forecast(): Evaluate forecast samples against ground truth for several metrics at once.
evaluate.results_to_dataframe(): Flatten backtest results into a tidy one-row-per-window DataFrame.
evaluate.eval_crps(): Empirical CRPS averaged over all data elements.
evaluate.eval_mae(): Mean absolute error using the forecast sample median as point estimate.
evaluate.eval_rmse(): Root mean squared error using the forecast sample mean as point estimate.
evaluate.eval_coverage(): Empirical coverage of the central alpha prediction interval.
metrics.crps_empirical(): Compute the empirical Continuous Ranked Probability Score (CRPS).
metrics.eval_pinball(): Mean pinball (quantile) loss of the forecast quantile.
metrics.eval_interval_score(): Mean Winkler interval score for the central alpha prediction interval.
metrics.make_mase(): Build a Mean Absolute Scaled Error metric scaled by train_data.

Batched autocorrelation and partial autocorrelation diagnostics.

acf.acf(): Compute the empirical autocorrelation function up to max_lag.
acf.pacf(): Compute the empirical partial autocorrelation function up to max_lag.

Fourier design matrices and seasonal tiling.

features.fourier_features(): Build a Fourier seasonality design matrix.
features.periodic_repeat(): Tile a seasonal pattern to cover duration time steps.

Time-axis array shaping for the train/forecast split.

arrays.zero_data_like(): Return zeros shaped like data but extended to the covariate duration.
arrays.concat_future(): Concatenate in-sample and forecast-horizon arrays along the time axis.

Time-axis operations on observation distributions, extensible via singledispatch.

surgery.shift_loc(): Re-center a zero-centered noise distribution at loc.
surgery.slice_time(): Slice an elementwise distribution along the time axis -2.
surgery.prefix_condition(): Condition a (t+f)-length distribution on a t-length data prefix.
surgery.register_elementwise(): Declare a distribution family elementwise (usable as a decorator).

Lazy imports behind pyproject extras.

optional.require(): Import an optional dependency, or raise a targeted ImportError.

Package exception hierarchy raised at resolution and validation boundaries.

exceptions.NumpyroForecastError: Base class for all deliberate numpyro_forecast errors.
exceptions.BacktestWindowError: A backtest window configuration is invalid.
exceptions.VectorizedGuideError: The vectorized backtest requires an AutoGuide.
exceptions.VectorizedMetricError: A metric is not vmappable in the vectorized backtest.
exceptions.OptimizerResolutionError: An optimizer specification could not be resolved.
exceptions.GuideResolutionError: A guide specification could not be resolved.
exceptions.GuideSampleArgsError: Drawing from a hand-written guide needs the in-sample arguments.
exceptions.KernelResolutionError: A kernel specification could not be resolved.
exceptions.KernelConfigError: A kernel is combined with an invalid configuration.
exceptions.CovariateDimsError: Covariate dimension names are inconsistent or malformed.
exceptions.MVNLayoutError: A MultivariateNormal layout is unsupported for time-axis surgery.
exceptions.DeviceMemoryError: The accelerator ran out of memory during posterior or predictive sampling.

Convert fits into ArviZ-schema xarray DataTrees for diagnostics and plotting.

convert.to_datatree(): Convert a fit into an ArviZ-schema xarray.DataTree.
convert.add_forecast_groups(): Attach out-of-sample forecast groups to a copy of tree.
convert.predictions_to_datatree(): Pack prediction draws into a DataTree laid out for per-series plot_lm faceting.

Optional backends behind pyproject extras (never imported by default).

contrib.blackjax.BlackjaxNUTSKernel: BlackJAX NUTS with Stan-style window adaptation.
contrib.blackjax.BlackjaxMCLMCKernel: BlackJAX Microcanonical Langevin Monte Carlo (MCLMC).
contrib.blackjax.BlackjaxCustomKernel: Adapt an arbitrary BlackJAX sampler via a user-supplied build_fn.
contrib.blackjax.PathfinderFit: The result of fitting a forecasting model with BlackJAX Pathfinder.
contrib.blackjax.fit_pathfinder(): Fit a forecasting model with BlackJAX Pathfinder variational inference.

Example datasets used in the tutorials.

datasets.load_bart_weekly(): Load total weekly BART ridership (log scale) for the univariate example.
datasets.load_bart_hierarchical(): Load the windowed hierarchical BART panel for the hierarchical example.
datasets.load_victoria_electricity(): Load hourly Victoria (Australia) electricity demand and temperature.
datasets.bart_available(): Return whether the BART dataset can be loaded (download succeeds).