You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In many of the "Model Evaluation" sections within the statsforecast docs, it's mentioned that datasetsforecast is a required installation in order to get access to error metrics.
However, this is not the case, as one can simply import them from utilsforecast, which comes installed with statsforecast.
from utilsforecast.losses import *
My sense is that the docs would read cleaner without this additional unnecessary external install requirement. Let me know what you think.
Description
In many of the "Model Evaluation" sections within the
statsforecast
docs, it's mentioned thatdatasetsforecast
is a required installation in order to get access to error metrics.However, this is not the case, as one can simply import them from
utilsforecast
, which comes installed withstatsforecast
.My sense is that the docs would read cleaner without this additional unnecessary external install requirement. Let me know what you think.
Link
https://nixtlaverse.nixtla.io/statsforecast/docs/models/autoarima.html#model-evaluation
https://nixtlaverse.nixtla.io/statsforecast/docs/models/optimizedtheta.html#model-evaluation
And others.
The text was updated successfully, but these errors were encountered: