Common

Core APIs

Recipe

Recipe(self, *steps)

Attributes

Name	Description
output_format	The output format to use for transform

Methods

Name	Description
fit	Fit a recipe.
fit_transform	Fit and transform in one step.
get_params	Get parameters for this recipe.
is_fitted	Check if this recipe has already been fit.
set_output	Set output type returned by transform.
set_params	Set the parameters of this recipe.
to_dask_dataframe	Transform X and return a dask.dataframe.DataFrame.
to_dask_dmatrix	Transform X and return a xgboost.dask.DMatrix
to_dmatrix	Transform X and return a xgboost.DMatrix
to_ibis	Transform X and return an ibis table.
to_numpy	Transform X and return a numpy.ndarray.
to_pandas	Transform X and return a pandas.DataFrame.
to_polars	Transform X and return a polars.DataFrame.
to_pyarrow	Transform X and return a pyarrow.Table.
to_pyarrow_batches	Transform X and return a pyarrow.RecordBatchReader.
transform	Transform the data.

fit

fit(X, y=None)

Fit a recipe.

Parameters

Name	Type	Description	Default
X	table - like	Training data.	required
y	column - like	Training targets.	None

Returns

Type	Description
self	Returns the same instance.

fit_transform

fit_transform(X, y=None)

Fit and transform in one step.

Parameters

Name	Type	Description	Default
X	table - like	Training data.	required
y	column - like	Training targets.	None

Returns

Type	Description
Xt	Transformed training data.

get_params

get_params(deep=True)

Get parameters for this recipe.

Returns the parameters given in the constructor as well as the steps contained within the steps of the Recipe.

Parameters

Name	Type	Description	Default
deep	bool	If True, will return the parameters for this recipe and contained steps.	True

Returns

Type	Description
mapping of string to any	Parameter names mapped to their values.

Notes

Derived from [1]_.

References

.. [1] https://github.com/scikit-learn/scikit-learn/blob/ee5a1b6/sklearn/utils/metaestimators.py#L30-L50

is_fitted

is_fitted()

Check if this recipe has already been fit.

set_output

set_output(transform=None)

Set output type returned by transform.

Parameters

Name	Type	Description	Default
transform	(‘default’, ‘pandas’)	Configure output of transform and fit_transform. - "default": Default output format of a transformer - "pandas": Pandas dataframe - "polars": Polars dataframe - "pyarrow": PyArrow table - None: Transform configuration is unchanged	"default"

set_params

set_params(**params)

Set the parameters of this recipe.

Valid parameter keys can be listed with get_params(). Note that you can directly set the parameters of the steps contained in steps.

Parameters

Name	Type	Description	Default
**params	dict	Parameters of this recipe or parameters of steps contained in steps. Parameters of the steps may be set using its name and the parameter name separated by a ’__’.	{}

Returns

Type	Description
object	Recipe class instance.

Notes

Derived from [1]_ and [2]_.

References

.. [1] https://github.com/scikit-learn/scikit-learn/blob/ff1c6f3/sklearn/utils/metaestimators.py#L51-L70 .. [2] https://github.com/scikit-learn/scikit-learn/blob/74016ab/sklearn/base.py#L214-L256

to_dask_dataframe

to_dask_dataframe(X, categories=False)

Transform X and return a dask.dataframe.DataFrame.

Parameters

Name	Type	Description	Default
X	table - like	The input data to transform.	required
categories	bool	Whether to return any categorical columns as dask categorical series. If False (the default) these columns will be returned as numeric columns containing only their integral categorical codes.	False

to_dask_dmatrix

to_dask_dmatrix(X)

Transform X and return a xgboost.dask.DMatrix

Parameters

Name	Type	Description	Default
X	table - like	The input data to transform.	required

to_dmatrix

to_dmatrix(X)

Transform X and return a xgboost.DMatrix

Parameters

Name	Type	Description	Default
X	table - like	The input data to transform.	required

to_ibis

to_ibis(X)

Transform X and return an ibis table.

Parameters

Name	Type	Description	Default
X	table - like	The input data to transform.	required

to_numpy

to_numpy(X)

Transform X and return a numpy.ndarray.

Parameters

Name	Type	Description	Default
X	table - like	The input data to transform.	required

to_pandas

to_pandas(X, categories=False)

Transform X and return a pandas.DataFrame.

Parameters

Name	Type	Description	Default
X	table - like	The input data to transform.	required
categories	bool	Whether to return any categorical columns as pandas categorical series. If False (the default) these columns will be returned as numeric columns containing only their integral categorical codes.	False

to_polars

to_polars(X)

Transform X and return a polars.DataFrame.

Parameters

Name	Type	Description	Default
X	table - like	The input data to transform.	required

to_pyarrow

to_pyarrow(X, categories=False)

Transform X and return a pyarrow.Table.

Parameters

Name	Type	Description	Default
X	table - like	The input data to transform.	required
categories	bool	Whether to return any categorical columns as dictionary-encoded columns. If False (the default) these columns will be returned as numeric columns containing only their integral categorical codes.	False

to_pyarrow_batches

to_pyarrow_batches(X, categories=False)

Transform X and return a pyarrow.RecordBatchReader.

Parameters

Name	Type	Description	Default
X	table - like	The input data to transform.	required
categories	bool	Whether to return any categorical columns as dictionary-encoded columns. If False (the default) these columns will be returned as numeric columns containing only their integral categorical codes.	False

transform

transform(X)

Transform the data.

Parameters

Name	Type	Description	Default
X	table - like	Data to transform.	required

Returns

Type	Description
Xt	Transformed data.