tf.keras.layers.BatchNormalization

Layer that normalizes its inputs.

Inherits From: Layer, Module

Batch normalization applies a transformation that maintains the mean output close to 0 and the output standard deviation close to 1.

Importantly, batch normalization works differently during training and during inference.

During training (i.e. when using fit() or when calling the layer/model with the argument training=True), the layer normalizes its output using the mean and standard deviation of the current batch of inputs. That is to say, for each channel being normalized, the layer returns (batch - mean(batch)) / (var(batch) + epsilon) * gamma + beta, where:

  • epsilon is small constant (configurable as part of the constructor arguments)
  • gamma is a learned scaling factor (initialized as 1), which can be disabled by passing scale=False to the constructor.
  • beta is a learned offset factor (initialized as 0), which can be disabled by passing center=False to the constructor.

During inference (i.e. when using evaluate() or predict() or when calling the layer/model with the argument training=False (which is the default), the layer normalizes its output using a moving average of the mean and standard deviation of the batches it has seen during training. That is to say, it returns (batch - self.moving_mean) / (self.moving_var + epsilon) * gamma + beta.

self.moving_mean and self.moving_var are non-trainable variables that are updated each time the layer in called in training mode, as such:

  • moving_mean = moving_mean * momentum + mean(batch) * (1 - momentum)
  • moving_var = moving_var * momentum + var(batch) * (1 - momentum)

As such, the layer will only normalize its inputs during inference after having been trained on data that has similar statistics as the inference data.

axis Integer or a list of integers, the axis that should be normalized (typically the features axis). For instance, after a Conv2D layer with data_format="channels_first", set axis=1 in BatchNormalization.
momentum Momentum for the moving average.
epsilon Small float added to variance to avoid dividing by zero.
center If True, add offset of beta to normalized tensor. If False, beta is ignored.
scale 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 will be done by the next layer.
beta_initializer Initializer for the beta weight.
gamma_initializer Initializer for the gamma weight.
moving_mean_initializer Initializer for the moving mean.
moving_variance_initializer Initializer for the moving variance.
beta_regularizer Optional regularizer for the beta weight.
gamma_regularizer Optional regularizer for the gamma weight.
beta_constraint Optional constraint for the beta weight.
gamma_constraint Optional constraint for the gamma weight.
renorm Whether to use Batch Renormalization. This adds extra variables during training. The inference is the same for either value of this parameter.
renorm_clipping 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.
renorm_momentum 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.
fused if True, use a faster, fused implementation, or raise a ValueError if the fused implementation cannot be used. If None, use the faster implementation if possible. If False, do not used the fused implementation.
trainable Boolean, if True the variables will be marked as trainable.
virtual_batch_size 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.
adjustment 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.

Call arguments:

  • inputs: Input tensor (of any rank).
  • training: Python boolean indicating whether the layer should behave in training mode or in inference mode.
    • training=True: The layer will normalize its inputs using the mean and variance of the current batch of inputs.
    • training=False: The layer will normalize its inputs using the mean and variance of its moving statistics, learned during training.

Input shape: Arbitrary. Use the keyword argument input_shape (tuple of integers, does not include the samples axis) when using this layer as the first layer in a model.

Output shape: Same shape as input.
About setting layer.trainable = False on a BatchNormalization layer:

The meaning of setting layer.trainable = False is to freeze the layer, i.e. its internal state will not change during training: its trainable weights will not be updated during fit() or train_on_batch(), and its state updates will not be run.

Usually, this does not necessarily mean that the layer is run in inference mode (which is normally controlled by the training argument that can be passed when calling a layer). "Frozen state" and "inference mode" are two separate concepts.

However, in the case of the BatchNormalization layer, setting trainable = False on the layer means that the layer will be subsequently run in inference mode (meaning that it will use the moving mean and the moving variance to normalize the current batch, rather than using the mean and variance of the current batch).

This behavior has been introduced in TensorFlow 2.0, in order to enable layer.trainable = False to produce the most commonly expected behavior in the convnet fine-tuning use case.

Note that:

  • This behavior only occurs as of TensorFlow 2.0. In 1.*, setting layer.trainable = False would freeze the layer but would not switch it to inference mode.
  • Setting trainable on an model containing other layers will recursively set the trainable value of all inner layers.
  • If the value of the trainable attribute is changed after calling compile() on a model, the new value doesn't take effect for this model until compile() is called again.

Reference: