Applies Group Normalization over a mini-batch of inputs as described in the paper Group Normalization.
Arguments
- num_groups
(int): number of groups to separate the channels into
- num_channels
(int): number of channels expected in input
- eps
a value added to the denominator for numerical stability. Default: 1e-5
- affine
a boolean value that when set to
TRUE, this module has learnable per-channel affine parameters initialized to ones (for weights) and zeros (for biases). Default:TRUE.
Details
$$ y = \frac{x - \mathrm{E}[x]}{ \sqrt{\mathrm{Var}[x] + \epsilon}} * \gamma + \beta $$
The input channels are separated into num_groups groups, each containing
num_channels / num_groups channels. The mean and standard-deviation are calculated
separately over the each group. \(\gamma\) and \(\beta\) are learnable
per-channel affine transform parameter vectors of size num_channels if
affine is TRUE.
The standard-deviation is calculated via the biased estimator, equivalent to
torch_var(input, unbiased=FALSE).
Shape
Input: \((N, C, *)\) where \(C=\mbox{num\_channels}\)
Output: \((N, C, *)\)` (same shape as input)
Examples
if (torch_is_installed()) {
input <- torch_randn(20, 6, 10, 10)
# Separate 6 channels into 3 groups
m <- nn_group_norm(3, 6)
# Separate 6 channels into 6 groups (equivalent with [nn_instance_morm])
m <- nn_group_norm(6, 6)
# Put all 6 channels into a single group (equivalent with [nn_layer_norm])
m <- nn_group_norm(1, 6)
# Activating the module
output <- m(input)
}