Class DataStats

monai/transforms/utility/array.py:640–746

Utility transform to show the statistics of data for debug or analysis. It can be inserted into any place of a transform chain and check results of previous transforms. It support both `numpy.ndarray` and `torch.tensor` as input data, so it can be used in pre-processing and post-pro

class DataStats(Transform):
    """
    Utility transform to show the statistics of data for debug or analysis.
    It can be inserted into any place of a transform chain and check results of previous transforms.
    It support both `numpy.ndarray` and `torch.tensor` as input data,
    so it can be used in pre-processing and post-processing.

    It gets logger from `logging.getLogger(name)`, we can setup a logger outside first with the same `name`.
    If the log level of `logging.RootLogger` is higher than `INFO`, will add a separate `StreamHandler`
    log handler with `INFO` level and record to `stdout`.

    """

    backend = [TransformBackends.TORCH, TransformBackends.NUMPY]

    def __init__(
        self,
        prefix: str = "Data",
        data_type: bool = True,
        data_shape: bool = True,
        value_range: bool = True,
        data_value: bool = False,
        meta_info: bool = False,
        additional_info: Callable | None = None,
        name: str = "DataStats",
    ) -> None:
        """
        Args:
            prefix: will be printed in format: "{prefix} statistics".
            data_type: whether to show the type of input data.
            data_shape: whether to show the shape of input data.
            value_range: whether to show the value range of input data.
            data_value: whether to show the raw value of input data.
                a typical example is to print some properties of Nifti image: affine, pixdim, etc.
            meta_info: whether to show the data of MetaTensor.
            additional_info: user can define callable function to extract additional info from input data.
            name: identifier of `logging.logger` to use, defaulting to "DataStats".

        Raises:
            TypeError: When ``additional_info`` is not an ``Optional[Callable]``.

        """
        if not isinstance(prefix, str):
            raise ValueError(f"prefix must be a string, got {type(prefix)}.")
        self.prefix = prefix
        self.data_type = data_type
        self.data_shape = data_shape
        self.value_range = value_range
        self.data_value = data_value
        self.meta_info = meta_info
        if additional_info is not None and not callable(additional_info):
            raise TypeError(f"additional_info must be None or callable but is {type(additional_info).__name__}.")
        self.additional_info = additional_info
        self._logger_name = name
        _logger = logging.getLogger(self._logger_name)
        _logger.setLevel(logging.INFO)
        if logging.root.getEffectiveLevel() > logging.INFO:
            # Avoid duplicate stream handlers to be added when multiple DataStats are used in a chain.