BatchNorm1d
Applies Batch Normalization over a 2D or 3D input as described in the paper Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift .
$$y = \frac{x - \mathrm{E}[x]}{\sqrt{\mathrm{Var}[x] + \epsilon}} * \gamma + \beta$$
The mean and standard-deviation are calculated per-dimension over the mini-batches and $\gamma$ and $\beta$ are learnable parameter vectors of size [C] (where [C] is the number of features or channels of the input). By default, the elements of $\gamma$ are set to 1 and the elements of $\beta$ are set to 0. The standard-deviation is calculated via the biased estimator, equivalent to [torch.var(input, unbiased=False)].
Also by default, during training this layer keeps running estimates of its computed mean and variance, which are then used for normalization during evaluation. The running estimates are kept with a default momentum of 0.1.
If trackRunningStats is set to false, this layer then does not keep running estimates, and batch statistics are instead used during evaluation time as well.
Example:
import torch.nn
// With Learnable Parameters
var m = nn.BatchNorm1d(numFeatures = 100)
// Without Learnable Parameters
m = nn.BatchNorm1d(100, affine = false)
val input = torch.randn(Seq(20, 100))
val output = m(input)
Value parameters
- affine:
-
a boolean value that when set to
true, this module has learnable affine parameters. Default:True - eps:
-
a value added to the denominator for numerical stability. Default: 1e-5
- momentum
-
the value used for the runningVean and runningVar computation. Can be set to
Nonefor cumulative moving average (i.e. simple average). Default: 0.1 - numFeatures
-
number of features or channels $C$ of the input
- trackRunningStats:
-
a boolean value that when set to
true, this module tracks the running mean and variance, and when set tofalse, this module does not track such statistics, and initializes statistics buffersrunningMeanandrunningVarasNone. When these buffers areNone, this module always uses batch statistics. in both training and eval modes. Default:trueShape:- Input: $(N, C)$ or $(N, C, L)$, where $N$ is the batch size, $C$ is the number of features or channels, and $L$ is the sequence length
- Output: $(N, C)$ or $(N, C, L)$ (same shape as input)
Attributes
- Note
-
This
momentumargument is different from one used in optimizer classes and the conventional notion of momentum. Mathematically, the update rule for running statistics here is $\hat{x}_\text{new} = (1 - \text{momentum}) \times \hat{x} + \text{momentum} \times x_t$, where $\hat{x}$ is the estimated statistic and $x_t$ is the new observed value. Because the Batch Normalization is done over the [C] dimension, computing statistics on [(N, L)] slices, it's common terminology to call this Temporal Batch Normalization. Args: - Source
- BatchNorm1d.scala
- Graph
-
- Supertypes