sklearn-dummies¶
Scikit-learn label binarizer with support for missing values.
Usage example¶
import pandas as pd
import sklearn_dummies as skdm
df = pd.DataFrame(['A', 'B', None, 'A'], columns=['val'])
df_dummy = skdm.DataFrameDummies().fit_transform(df)
Result:
| idx | val_A | val_B |
|---|---|---|
| 0 | 1.0 | 0.0 |
| 1 | 0.0 | 1.0 |
| 2 | NaN | NaN |
| 3 | 1.0 | 0.0 |
Installing¶
Sklearn-dummies is available in PyPI. Install via pip:
pip install sklearn_dummies
sklearn_dummies API¶
Base module.
-
class
sklearn_dummies.base.DataFrameDummies¶ Bases:
sklearn.base.TransformerMixin-
cat_cols¶ List of categorical columns
Type: list
-
final_cols¶ List of all columns with dummy values
Type: list
-
fit(df, y=None)¶ Parameters: df (pd.DataFrame) – Provides the column names to be used. Returns: Itself. Return type: DataFrameDummies
-
transform(df, y=None)¶ Parameters: df (pd.DataFrame) – Data to be dummified. Returns: Transformed data Return type: pd.DataFrame
-
-
class
sklearn_dummies.base.NPArrayDummies¶ Bases:
sklearn.base.TransformerMixin-
labels¶ List of labels
Type: list
-
fit(X, y=None)¶ Parameters: X (np.ndarray) – Provides the labels. Returns: Itself. Return type: NPArrayDummies
-
transform(X, y=None)¶ Parameters: X (np.ndarray) – Data to be dummified. Returns: Xt – Transformed data Return type: np.ndarray
-