Small Double > 0. Fuzz parameter. Default is 0.001.
Double. Momentum in the computation of the exponential average of the mean and standard deviation of the data, for feature-wise normalization. Default is 0.99.
Name of initialization function for shift parameter. Default is 'zero'.
Name of initialization function for scale parameter. Default is 'one'.
Format of input data. Either DataFormat.NCHW (dimOrdering='th') or DataFormat.NHWC (dimOrdering='tf'). Default is NCHW. For NCHW, axis along which to normalize is 1. For NHWC, axis is 3.
A Single Shape, does not include the batch dimension.
Name of initialization function for shift parameter.
Name of initialization function for shift parameter. Default is 'zero'.
Format of input data.
Format of input data. Either DataFormat.NCHW (dimOrdering='th') or DataFormat.NHWC (dimOrdering='tf'). Default is NCHW. For NCHW, axis along which to normalize is 1. For NHWC, axis is 3.
Small Double > 0.
Small Double > 0. Fuzz parameter. Default is 0.001.
Build graph: some other modules point to current module
Build graph: some other modules point to current module
upstream variables
Variable containing current module
Name of initialization function for scale parameter.
Name of initialization function for scale parameter. Default is 'one'.
A Single Shape, does not include the batch dimension.
A Single Shape, does not include the batch dimension.
Double.
Double. Momentum in the computation of the exponential average of the mean and standard deviation of the data, for feature-wise normalization. Default is 0.99.
(Since version 0.3.0) please use recommended saveModule(path, overWrite)
Batch normalization layer. Normalize the activations of the previous layer at each batch, i.e. applies a transformation that maintains the mean activation close to 0 and the activation standard deviation close to 1. It is a feature-wise normalization, each feature map in the input will be normalized separately. The input of this layer should be 4D.
When you use this layer as the first layer of a model, you need to provide the argument inputShape (a Single Shape, does not include the batch dimension).
Numeric type of parameter(e.g. weight, bias). Only support float/double now.