LPPool1d

Contents

LPPool1d#

class brainstate.nn.LPPool1d(norm_type, kernel_size, stride=None, padding='VALID', channel_axis=-1, name=None, in_size=None)#

Applies a 1D power-average pooling over an input signal composed of several input planes.

On each window, the function computed is:

\[f(X) = \sqrt[p]{\sum_{x \in X} |x|^{p}}\]

  • At \(p = \infty\), one gets max pooling

  • At \(p = 1\), one gets average pooling (with absolute values)

  • At \(p = 2\), one gets root mean square (RMS) pooling

Shape:

  • Input: \((N, L_{in}, C)\) or \((L_{in}, C)\).

  • Output: \((N, L_{out}, C)\) or \((L_{out}, C)\), where

    \[L_{out} = \left\lfloor \frac{L_{in} + 2 \times \text{padding} - \text{kernel\_size}}{\text{stride}} + 1\right\rfloor\]
Parameters:

  • norm_type (float) – Exponent for the pooling operation. Default: 2.0

  • kernel_size (int | Sequence[int] | integer | Sequence[integer]) – An integer, or a sequence of integers defining the window to reduce over.

  • stride (int | Sequence[int]) – An integer, or a sequence of integers, representing the inter-window stride. Default: kernel_size

  • padding (str | int | Tuple[int, ...] | Sequence[Tuple[int, int]]) – Either the string ‘SAME’, the string ‘VALID’, or a sequence of n (low, high) integer pairs that give the padding to apply before and after each spatial dimension. Default: ‘VALID’

  • channel_axis (int | None) – Axis of the spatial channels for which pooling is skipped. If None, there is no channel axis. Default: -1

  • name (str | None) – The object name.

  • in_size (int | Sequence[int] | integer | Sequence[integer] | None) – The shape of the input tensor.

Examples

>>> import brainstate
>>> # power-average pooling of window of size=3, stride=2 with norm_type=2.0
>>> m = LPPool1d(2, 3, stride=2)
>>> input = brainstate.random.randn(20, 50, 16)
>>> output = m(input)
>>> output.shape
(20, 24, 16)
Contents