Ets

polars_ts.streaming.ets

Online exponential smoothing (SES, Holt, Holt-Winters) with partial_fit.

StreamingETS

Streaming exponential smoothing with incremental updates.

Parameters

method One of "ses", "holt", "holt_winters". alpha Level smoothing parameter. beta Trend smoothing parameter (Holt / Holt-Winters only). gamma Seasonal smoothing parameter (Holt-Winters only). season_length Observations per season (Holt-Winters only). seasonal "additive" or "multiplicative" (Holt-Winters only). id_col, time_col, target_col Column name overrides.

state_ property

Per-series state dictionaries.

fit(df)

Initialize state from a batch of historical data.

partial_fit(df)

Incrementally update state with new observations.

predict(h)

Generate h-step-ahead forecasts from current state.

_fit_series(values, last_time)

Compute initial state for one series.

_update_series(sid, values, last_time)

Incrementally update state for one series.

_forecast_series(state, h)

Generate forecasts from current state.