How to use the rdt.transformers.DatetimeTransformer function in rdt

"""Create the transformer instances needed to process the given dtypes.

        Args:
            dtypes (dict):
                mapping of field names and dtypes.

        Returns:
            dict:
                mapping of field names and transformer instances.
        """
        transformer_templates = {
            'i': rdt.transformers.NumericalTransformer(dtype=int),
            'f': rdt.transformers.NumericalTransformer(dtype=float),
            'O': rdt.transformers.CategoricalTransformer,
            'b': rdt.transformers.BooleanTransformer,
            'M': rdt.transformers.DatetimeTransformer,
        }
        transformer_templates.update(self._transformer_templates)

        transformers = dict()
        for name, dtype in dtypes.items():
            transformer_template = transformer_templates[np.dtype(dtype).kind]
            if isinstance(transformer_template, type):
                transformer = transformer_template()
            else:
                transformer = copy.deepcopy(transformer_template)

            LOGGER.info('Loading transformer %s for field %s',
                        transformer.__class__.__name__, name)
            transformers[name] = transformer

        return transformers

mapping of field names and transformer instances.
        """
        transformers_dict = dict()
        for name, dtype in dtypes.items():
            dtype = np.dtype(dtype)
            if dtype.kind == 'i':
                transformer = transformers.NumericalTransformer(dtype=int)
            elif dtype.kind == 'f':
                transformer = transformers.NumericalTransformer(dtype=float)
            elif dtype.kind == 'O':
                anonymize = pii_fields.get(name)
                transformer = transformers.CategoricalTransformer(anonymize=anonymize)
            elif dtype.kind == 'b':
                transformer = transformers.BooleanTransformer()
            elif dtype.kind == 'M':
                transformer = transformers.DatetimeTransformer()
            else:
                raise ValueError('Unsupported dtype: {}'.format(dtype))

            LOGGER.info('Loading transformer %s for field %s',
                        transformer.__class__.__name__, name)
            transformers_dict[name] = transformer

        return transformers_dict