add support for regressors in make_future_dataframe · facebook/prophet#2068

(4 comments) (3 reactions) (0 assignees)Python (17,686 stars) (4,451 forks)batch import

enhancementgood first issue

説明

make_future_dataframe doesn't support regressors currently. So code like:

m = Prophet()
m.add_regressor('var')
m.fit(df)
forecasts = m.predict(m.make_future_dataframe(periods=7))

gives an error like:

ValueError: Regressor 'var' missing from dataframe when attempting to generate forecasts

I know prophet may not know what exact values to put for var in each of the rows added but it can create a dataframe with thes(e) column(s) with default values for this.

There is a post on stackoverflow a year ago looking for this: https://stackoverflow.com/questions/64822488/how-to-use-prophets-make-future-dataframe-with-multiple-regressors

コントリビューターガイド

技術スタック: python
領域: backenddata
Issue 種別: feature
難度: 3
推定時間: 1-3 hours
活動状況: fresh
明確さ: mostly clear
前提条件: Basic knowledge of ProphetPython programmingUnderstanding of time series forecastingFamiliarity with regressors in Prophet
初心者向け度: 60
調査方針: The issue requests that `make future dataframe` include columns for regressors added via `add regressor`. Investigate the Prophet source code, particularly the `Prophet` class and the `make future dataframe` method (likely in `prophet/forecaster.py`). Determine how regressors are stored (e.g., in `self.extra regressors`) and modify the method to add placeholder columns with default values (like 0 or mean) for future dates. Ensure that the new columns are compatible with the prediction method. Check existing comments for any proposed solutions or discussions.