Skip to content

komm.convolve

Convolves two signals. This is like numpy.convolve, but one of the inputs may be multidimensional (the other input must still be $1$-dimensional). The convolution is applied independently to each array along a specified axis.

Note

This is a simple wrapper around numpy.convolve.

Parameters:

  • input1 (ArrayLike) –

    The first input array.

  • input2 (ArrayLike) –

    The second input array.

  • axis (int) –

    The axis along which the convolution is applied in the multidimensional input. Default is the last axis.

  • mode (Literal['full', 'valid', 'same']) –

    Convolution mode (same as numpy.convolve).

Examples:

>>> komm.convolve([1, 1], [1, 1])
array([1, 2, 1])

>>> komm.convolve([1, 1], [[1, 1, 0], [0, 1, 1]])
array([[1, 2, 1, 0],
       [0, 1, 2, 1]])

>>> komm.convolve([[1, 1, 0], [0, 1, 1]], [1, 1])
array([[1, 2, 1, 0],
       [0, 1, 2, 1]])

>>> komm.convolve([1, 1], [[1, 1, 0], [0, 1, 1]], axis=0)
array([[1, 1, 0],
       [1, 2, 1],
       [0, 1, 1]])

>>> komm.convolve([[1, 1]], [[1, 1]])
Traceback (most recent call last):
...
ValueError: at least one of the inputs must be 1-dimensional