Function _transform_cudf_df

python-package/xgboost/data.py:1074–1148

(
    data: DataType,
    feature_names: Optional[FeatureNames],
    feature_types: Optional[Union[FeatureTypes, Categories]],
    enable_categorical: bool,
)

def _transform_cudf_df(
    data: DataType,
    feature_names: Optional[FeatureNames],
    feature_types: Optional[Union[FeatureTypes, Categories]],
    enable_categorical: bool,
) -> Tuple[
    CudfTransformed,
    Optional[FeatureNames],
    Optional[FeatureTypes],
]:
    is_bool_dtype = _lazy_load_cudf_is_bool()

    is_categorical_dtype = _lazy_load_cudf_is_cat()
    # Work around https://github.com/dmlc/xgboost/issues/10181
    if _is_cudf_ser(data):
        if is_bool_dtype(data.dtype):
            data = data.astype(np.uint8)
        dtypes = [data.dtype]
    else:
        data = data.astype(
            {col: np.uint8 for col in data.select_dtypes(include="bool")}
        )
        dtypes = data.dtypes

    if not all(
        dtype.name in _pandas_dtype_mapper
        or (is_categorical_dtype(dtype) and enable_categorical)
        for dtype in dtypes
    ):
        _invalid_dataframe_dtype(data)

    # handle feature names
    if feature_names is None:
        if _is_cudf_ser(data):
            feature_names = [data.name]
        elif lazy_isinstance(data.columns, "cudf.core.multiindex", "MultiIndex"):
            feature_names = [" ".join([str(x) for x in i]) for i in data.columns]
        else:
            feature_names = list(data.columns.map(str))

    # handle feature types
    feature_types, ref_categories = get_ref_categories(feature_types)
    if feature_types is None:
        feature_types = []
        for dtype in dtypes:
            if is_categorical_dtype(dtype) and enable_categorical:
                feature_types.append(CAT_T)
            else:
                feature_types.append(_pandas_dtype_mapper[dtype.name])

    # handle categorical data
    result = []
    if _is_cudf_ser(data):
        # unlike pandas, cuDF uses NA for missing data.
        if is_categorical_dtype(data.dtype) and enable_categorical:
            result.append(data.cat)
        elif is_categorical_dtype(data.dtype) and not enable_categorical:
            raise ValueError(_ENABLE_CAT_ERR)