normalize

function of dascore.proc.basic source

normalize(
    self: Patch ,
    dim: str ,
    norm: Literal[‘l1’, ‘l2’, ‘max’, ‘bit’] = l2,
)-> ‘PatchType’

Normalize a patch along a specified dimension.

Parameters

Parameter Description
dim The dimension along which the normalization takes place.
norm Determines the value to divide each sample by along a given axis.
Options are:
l1 - divide each sample by the l1 of the axis.
l2 - divide each sample by the l2 of the axis.
max - divide each sample by the maximum of the absolute value of the axis.
bit - sample-by-sample normalization (-1/+1)

Examples

import dascore as dc
patch = dc.get_example_patch()

# L2 normalization along time dimension
l2_norm = patch.normalize(dim="time", norm="l2")

# Max normalization along distance dimension
max_norm = patch.normalize(dim="distance", norm="max")

# Bit normalization (sign only)
bit_norm = patch.normalize(dim="time", norm="bit")