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
As a user of multiseries time series in EvalML, I would like the existing TimeSeriesImputer component to be able to handle my multiseries dataset, which has missing values, in a way that matches its current single series implementation.
Acceptance Criteria (AC)
When passed an unstacked multiseries dataset, the TimeSeriesImputer will impute the multiple target columns identically to how it imputes a single target column
One imputation method argument controls all the imputation methods for all target columns, i.e. all target columns are imputed with the same strategy
The text was updated successfully, but these errors were encountered:
Description
As a user of multiseries time series in EvalML, I would like the existing TimeSeriesImputer component to be able to handle my multiseries dataset, which has missing values, in a way that matches its current single series implementation.
Acceptance Criteria (AC)
When passed an unstacked multiseries dataset, the TimeSeriesImputer will impute the multiple target columns identically to how it imputes a single target column
One imputation method argument controls all the imputation methods for all target columns, i.e. all target columns are imputed with the same strategy
The text was updated successfully, but these errors were encountered: