search

JoaquinAmatRodrigo / skforecast

Time series forecasting with machine learning models
https://skforecast.org
BSD 3-Clause "New" or "Revised" License
1.08k stars 122 forks source link

Typos in "Recursive multi-step forecasting with exogenous variables" documentation #199

Closed kaionwong closed 2 years ago

kaionwong commented 2 years ago

I have found a bug in the python code on this page "Recursive multi-step forecasting with exogenous variables".

If exog = data_train[['exog_1', 'exog_2']].values is run directly, it will give this error Exception: "exog" must be "pd.Series" or "pd.DataFrame".

Correction, this should be exog = data_train[['exog_1', 'exog_2']] without the .values since a pandas pd.DataFrame is expected. Please make the correction in the appropriate documentation page and code examples, thanks.

JavierEscobarOrtiz commented 2 years ago

Hello @kaionwong,

The page you refer to is the documentation for an older version of skforecast, specifically 0.3.0 where the code was different. If you are using this version, I recommend updating the library to take advantage of the latest improvements.

Here is the same page for version 0.4.3: Recursive multi-step forecasting with exogenous variables.

Please note that you can change the version referenced in the documentation with this selector:

imagen

Hope it helps!