pysteps.utils.transformation.boxcox_transform¶

pysteps.utils.transformation.boxcox_transform(R, metadata=None, Lambda=None, threshold=None, zerovalue=None, inverse=False)¶

The one-parameter Box-Cox transformation. Default parameters will produce a log transform (i.e. Lambda=0).

Parameters:

R : array-like: Array of any shape to be transformed.
metadata : dict, optional: Metadata dictionary containing the transform, zerovalue and threshold attributes as described in the documentation of pysteps.io.importers.
Lambda : float, optional: Parameter lambda of the Box-Cox transformation. It is 0 by default, which produces the log transformation.
threshold : float, optional: The value that is used for thresholding with the same units as R. If None, the threshold contained in metadata is used. If no threshold is found in the metadata, a value of 0.1 is used as default.
zerovalue : float, optional: The value to be assigned to no rain pixels as defined by the threshold. It is equal to the threshold - 1 by default.
inverse : bool, optional: If set to True, it performs the inverse transform. False by default.

Returns: