data_describe.misc.preprocessing¶
|
Simple preprocessing pipeline for ML. |
-
data_describe.misc.preprocessing.
preprocess
(data, target, impute='simple', encode='label')¶ Simple preprocessing pipeline for ML.
- Parameters
data – A Pandas dataframe
target – Name of the target feature
impute – Method to use for imputing numeric variables. Only ‘simple’ is implemented.
encode – Method to use for encoding categorical variables. Only ‘label’ is implemented.
- Raises
NotImplementedError – Imputation or encoding method not implemented.
ValueError – No columns left to preprocess.
- Returns
(X, y) tuple of numpy arrays