An int, the axis that should be normalized (typically the features
axis). For instance, after a Convolution2D layer with
data_format="channels_first", set axis=1 in BatchNormalization.
Momentum for the moving average.
Small float added to variance to avoid dividing by zero.
If True, add offset of beta to normalized tensor. If False, beta
If True, multiply by gamma. If False, gamma is not used. When the
next layer is linear (also e.g. nn.relu), this can be disabled since the
scaling can be done by the next layer.
Initializer for the beta weight.
Initializer for the gamma weight.
Initializer for the moving mean.
Initializer for the moving variance.
Optional regularizer for the beta weight.
Optional regularizer for the gamma weight.
An optional projection function to be applied to the beta
weight after being updated by an Optimizer (e.g. used to implement norm
constraints or value constraints for layer weights). The function must
take as input the unprojected variable and must return the projected
variable (which must have the same shape). Constraints are not safe to use
when doing asynchronous distributed training.
An optional projection function to be applied to the
gamma weight after being updated by an Optimizer.
Either a Python boolean, or a TensorFlow boolean scalar tensor
(e.g. a placeholder). Whether to return the output in training mode
(normalized with statistics of the current batch) or in inference mode
(normalized with moving statistics). NOTE: make sure to set this
parameter correctly, or else your training/inference will not work
Boolean, if True also add variables to the graph collection
GraphKeys.TRAINABLE_VARIABLES (see tf.Variable).
String, the name of the layer.
Boolean, whether to reuse the weights of a previous layer by the same
Whether to use Batch Renormalization (Ioffe, 2017). This adds extra
variables during training. The inference is the same for either value of
A dictionary that may map keys 'rmax', 'rmin', 'dmax' to
scalar Tensors used to clip the renorm correction. The correction (r,
d) is used as corrected_value = normalized_value * r + d, with r
clipped to [rmin, rmax], and d to [-dmax, dmax]. Missing rmax, rmin,
dmax are set to inf, 0, inf, respectively.
Momentum used to update the moving means and standard
deviations with renorm. Unlike momentum, this affects training and
should be neither too small (which would add noise) nor too large (which
would give stale estimates). Note that momentum is still applied to get
the means and variances for inference.
if None or True, use a faster, fused implementation if possible.
If False, use the system recommended implementation.
An int. By default, virtual_batch_size is None,
which means batch normalization is performed across the whole batch. When
virtual_batch_size is not None, instead perform "Ghost Batch
Normalization", which creates virtual sub-batches which are each
normalized separately (with shared gamma, beta, and moving statistics).
Must divide the actual batch size during execution.
A function taking the Tensor containing the (dynamic) shape of
the input tensor and returning a pair (scale, bias) to apply to the
normalized values (before gamma and beta), only during training. For
example, if axis==-1,
adjustment = lambda shape: (
tf.random.uniform(shape[-1:], 0.93, 1.07),
tf.random.uniform(shape[-1:], -0.1, 0.1)) will scale the normalized
value by up to 7% up or down, then shift the result by up to 0.1
(with independent scaling and bias for each feature but shared
across all examples), and finally apply gamma and/or beta. If
None, no adjustment is applied. Cannot be specified if
virtual_batch_size is specified.
if eager execution is enabled.
Batch Normalization - Accelerating Deep Network Training by Reducing
Internal Covariate Shift:
Ioffe et al., 2015
Batch Renormalization - Towards Reducing Minibatch Dependence in
"label":"Missing the information I need"
"label":"Too complicated / too many steps"
"label":"Out of date"
"label":"Samples / code issue"
"label":"Easy to understand"
"label":"Solved my problem"