I have a data frame with the following columns dtype

{Int64Dtype(), UInt8Dtype(), dtype('float64'), dtype('int64')}

when I'm trying to fit xgb.XGBClassifier() I'm getting following error

ValueError: DataFrame.dtypes for data must be int, float, bool or category.  When
categorical type is supplied, DMatrix parameter `enable_categorical` must
be set to `True`. Invalid columns: NAME OF COLS THAT ARE UINT TYPE

CodePudding user response：

Here's the code which triggers the warning:

def _invalid_dataframe_dtype(data: DataType) -> None:
    # pandas series has `dtypes` but it's just a single object
    # cudf series doesn't have `dtypes`.
    if hasattr(data, "dtypes") and hasattr(data.dtypes, "__iter__"):
        bad_fields = [
            str(data.columns[i])
            for i, dtype in enumerate(data.dtypes)
            if dtype.name not in _pandas_dtype_mapper
        ]
        err = " Invalid columns:"   ", ".join(bad_fields)
    else:
        err = ""

    type_err = "DataFrame.dtypes for data must be int, float, bool or category."
    msg = f"""{type_err} {_ENABLE_CAT_ERR} {err}"""
    raise ValueError(msg)

(Source.)

It references another variable, _pandas_dtype_mapper, which is used to decide how to match each datatype. Here's how that is defined:

_pandas_dtype_mapper = {
    'int8': 'int',
    'int16': 'int',
    'int32': 'int',
    'int64': 'int',
    'uint8': 'int',
    'uint16': 'int',
    'uint32': 'int',
    'uint64': 'int',
    'float16': 'float',
    'float32': 'float',
    'float64': 'float',
    'bool': 'i',
    # nullable types
    "Int16": "int",
    "Int32": "int",
    "Int64": "int",
    "boolean": "i",
}

(Source.)

So, here we find the problem. It supports a uint datatype. It supports a nullable datatype. But it doesn't seem to support a nullable uint datatype.

This suggests two possible workarounds:

1. Use int instead of uint.
2. Fill in your null values in that column, and convert that column to a non-nullable datatype.