something like

mlf = MLForecast(
    models = [LinearRegression(), lgb.LGBMRegressor(),AutoARIMA(),NBEATS()],
    lags=[1, 12],(
)
mlf.fit(df)
mlf.predict(12)

Sadly this is still not possible. However, here is a way to combine them: https://nixtla.github.io/statsforecast/docs/tutorials/statisticalneuralmethods.html

That is a super useful ! thanks for your work..I think I might juts switch from darts 😉 ..really like the prediction interval with CP.

Woop woop! Please tell us if we can help anymore