mlx.core.conv1d#

conv1d(input: array, weight: array, /, stride: int = 1, padding: int = 0, dilation: int = 1, groups: int = 1, *, stream: None | Stream | Device = None) → array#

1D convolution over an input with several channels

Note: Only the default groups=1 is currently supported.

Parameters:

  • input (array) – input array of shape (N, H, C_in)

  • weight (array) – weight array of shape (C_out, H, C_in)

  • stride (int, optional) – kernel stride. Default: 1.

  • padding (int, optional) – input padding. Default: 0.

  • dilation (int, optional) – kernel dilation. Default: 1.

  • groups (int, optional) – input feature groups. Default: 1.

Returns:

The convolved array.

Return type:

array