Describe the bug
Prophet allows logistic grown capacity and/ or floor to be either scalar or array. While fixing #1079 noticed that when set to array fit does the job but predict fails, as it attempts to reuse the same cap and floor params. predict expects cap and floor to be same length as fh and thus will need to passed into predict.
To Reproduce
import numpy as np
import pandas as pd
from sktime.forecasting.fbprophet import Prophet
from sktime.forecasting.model_selection import temporal_train_test_split
from sktime.datasets import load_airline
y = load_airline()
y = y.to_timestamp(freq="M")
y_train, y_test = temporal_train_test_split(y, test_size=36)
y_train_df = pd.DataFrame(y_train)
y_train_df.columns = ["y"]
y_train_df["cap"] = 1000
cap = np.full_like(y_train_df["y"], 1000)
forecaster = Prophet(growth="logistic", growth_cap=cap)
forecaster.fit(y_train_df)
y_pred = forecaster.predict([1, 2, 3])
Expected behavior
y_pred with valid values.
Additional context
only fails if cap is array.
I'll continue looking into this.
Describe the bug
Prophet allows logistic grown capacity and/ or floor to be either scalar or array. While fixing #1079 noticed that when set to array fit does the job but predict fails, as it attempts to reuse the same
capandfloorparams.predictexpectscapandfloorto be same length asfhand thus will need to passed intopredict.To Reproduce
Expected behavior
y_pred with valid values.
Additional context
only fails if cap is array.
I'll continue looking into this.