tfp.bijectors.BatchNormalization

View source on GitHub

Class BatchNormalization

Compute Y = g(X) s.t. X = g^-1(Y) = (Y - mean(Y)) / std(Y).

Inherits From: Bijector

Applies Batch Normalization [(Ioffe and Szegedy, 2015)][1] to samples from a data distribution. This can be used to stabilize training of normalizing flows ([Papamakarios et al., 2016][3]; [Dinh et al., 2017][2])

When training Deep Neural Networks (DNNs), it is common practice to normalize or whiten features by shifting them to have zero mean and scaling them to have unit variance.

The inverse() method of the BatchNormalization bijector, which is used in the log-likelihood computation of data samples, implements the normalization procedure (shift-and-scale) using the mean and standard deviation of the current minibatch.

Conversely, the forward() method of the bijector de-normalizes samples (e.g. X*std(Y) + mean(Y) with the running-average mean and standard deviation computed at training-time. De-normalization is useful for sampling.


dist = tfd.TransformedDistribution(
    distribution=tfd.Normal()),
    bijector=tfb.BatchNormalization())

y = tfd.MultivariateNormalDiag(loc=1., scale=2.).sample(100)  # ~ N(1, 2)
x = dist.bijector.inverse(y)  # ~ N(0, 1)
y = dist.sample()  # ~ N(1, 2)

During training time, BatchNormalization.inverse and BatchNormalization.forward are not guaranteed to be inverses of each other because inverse(y) uses statistics of the current minibatch, while forward(x) uses running-average statistics accumulated from training. In other words, BatchNormalization.inverse(BatchNormalization.forward(...)) and BatchNormalization.forward(BatchNormalization.inverse(...)) will be identical when training=False but may be different when training=True.

References

[1]: Sergey Ioffe and Christian Szegedy. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. In International Conference on Machine Learning, 2015. https://arxiv.org/abs/1502.03167

[2]: Laurent Dinh, Jascha Sohl-Dickstein, and Samy Bengio. Density Estimation using Real NVP. In International Conference on Learning Representations, 2017. https://arxiv.org/abs/1605.08803

[3]: George Papamakarios, Theo Pavlakou, and Iain Murray. Masked Autoregressive Flow for Density Estimation. In Neural Information Processing Systems, 2017. https://arxiv.org/abs/1705.07057

__init__

View source

__init__(
    batchnorm_layer=None,
    training=True,
    validate_args=False,
    name='batch_normalization'
)

Instantiates the BatchNormalization bijector.

Args:

  • batchnorm_layer: tf.layers.BatchNormalization layer object. If None, defaults to a tf.keras.layers.BatchNormalization with gamma_constraint=tf.nn.relu(x) + 1e-6). This ensures positivity of the scale variable.

  • training: If True, updates running-average statistics during call to inverse().

  • validate_args: Python bool indicating whether arguments should be checked for correctness.

  • name: Python str name given to ops managed by this object.

Raises:

  • ValueError: If bn_layer is not an instance of tf.layers.BatchNormalization, or if it is specified with renorm=True or a virtual batch size.

Properties

dtype

dtype of Tensors transformable by this distribution.

forward_min_event_ndims

Returns the minimal number of dimensions bijector.forward operates on.

graph_parents

Returns this Bijector's graph_parents as a Python list.

inverse_min_event_ndims

Returns the minimal number of dimensions bijector.inverse operates on.

is_constant_jacobian

Returns true iff the Jacobian matrix is not a function of x.

Returns:

  • is_constant_jacobian: Python bool.

name

Returns the string name of this Bijector.

name_scope

Returns a tf.name_scope instance for this class.

submodules

Sequence of all sub-modules.

Submodules are modules which are properties of this module, or found as properties of modules which are properties of this module (and so on).

a = tf.Module()
b = tf.Module()
c = tf.Module()
a.b = b
b.c = c
assert list(a.submodules) == [b, c]
assert list(b.submodules) == [c]
assert list(c.submodules) == []

Returns:

A sequence of all submodules.

trainable_variables

Sequence of variables owned by this module and it's submodules.

Returns:

A sequence of variables for the current module (sorted by attribute name) followed by variables from all submodules recursively (breadth first).

validate_args

Returns True if Tensor arguments will be validated.

variables

Sequence of variables owned by this module and it's submodules.

Returns:

A sequence of variables for the current module (sorted by attribute name) followed by variables from all submodules recursively (breadth first).

Methods

__call__

View source

__call__(
    value,
    name=None,
    **kwargs
)

Applies or composes the Bijector, depending on input type.

This is a convenience function which applies the Bijector instance in three different ways, depending on the input:

  1. If the input is a tfd.Distribution instance, return tfd.TransformedDistribution(distribution=input, bijector=self).
  2. If the input is a tfb.Bijector instance, return tfb.Chain([self, input]).
  3. Otherwise, return self.forward(input)

Args:

  • value: A tfd.Distribution, tfb.Bijector, or a Tensor.
  • name: Python str name given to ops created by this function.
  • **kwargs: Additional keyword arguments passed into the created tfd.TransformedDistribution, tfb.Bijector, or self.forward.

Returns:

  • composition: A tfd.TransformedDistribution if the input was a tfd.Distribution, a tfb.Chain if the input was a tfb.Bijector, or a Tensor computed by self.forward.

Examples

sigmoid = tfb.Reciprocal()(
    tfb.AffineScalar(shift=1.)(
      tfb.Exp()(
        tfb.AffineScalar(scale=-1.))))
# ==> `tfb.Chain([
#         tfb.Reciprocal(),
#         tfb.AffineScalar(shift=1.),
#         tfb.Exp(),
#         tfb.AffineScalar(scale=-1.),
#      ])`  # ie, `tfb.Sigmoid()`

log_normal = tfb.Exp()(tfd.Normal(0, 1))
# ==> `tfd.TransformedDistribution(tfd.Normal(0, 1), tfb.Exp())`

tfb.Exp()([-1., 0., 1.])
# ==> tf.exp([-1., 0., 1.])

forward

View source

forward(
    x,
    name='forward',
    **kwargs
)

Returns the forward Bijector evaluation, i.e., X = g(Y).

Args:

  • x: Tensor. The input to the 'forward' evaluation.
  • name: The name to give this op.
  • **kwargs: Named arguments forwarded to subclass implementation.

Returns:

Tensor.

Raises:

  • TypeError: if self.dtype is specified and x.dtype is not self.dtype.
  • NotImplementedError: if _forward is not implemented.

forward_event_shape

View source

forward_event_shape(input_shape)

Shape of a single sample from a single batch as a TensorShape.

Same meaning as forward_event_shape_tensor. May be only partially defined.

Args:

  • input_shape: TensorShape indicating event-portion shape passed into forward function.

Returns:

  • forward_event_shape_tensor: TensorShape indicating event-portion shape after applying forward. Possibly unknown.

forward_event_shape_tensor

View source

forward_event_shape_tensor(
    input_shape,
    name='forward_event_shape_tensor'
)

Shape of a single sample from a single batch as an int32 1D Tensor.

Args:

  • input_shape: Tensor, int32 vector indicating event-portion shape passed into forward function.
  • name: name to give to the op

Returns:

  • forward_event_shape_tensor: Tensor, int32 vector indicating event-portion shape after applying forward.

forward_log_det_jacobian

View source

forward_log_det_jacobian(
    x,
    event_ndims,
    name='forward_log_det_jacobian',
    **kwargs
)

Returns both the forward_log_det_jacobian.

Args:

  • x: Tensor. The input to the 'forward' Jacobian determinant evaluation.
  • event_ndims: Number of dimensions in the probabilistic events being transformed. Must be greater than or equal to self.forward_min_event_ndims. The result is summed over the final dimensions to produce a scalar Jacobian determinant for each event, i.e. it has shape rank(x) - event_ndims dimensions.
  • name: The name to give this op.
  • **kwargs: Named arguments forwarded to subclass implementation.

Returns:

Tensor, if this bijector is injective. If not injective this is not implemented.

Raises:

  • TypeError: if self.dtype is specified and y.dtype is not self.dtype.
  • NotImplementedError: if neither _forward_log_det_jacobian nor {_inverse, _inverse_log_det_jacobian} are implemented, or this is a non-injective bijector.

inverse

View source

inverse(
    y,
    name='inverse',
    **kwargs
)

Returns the inverse Bijector evaluation, i.e., X = g^{-1}(Y).

Args:

  • y: Tensor. The input to the 'inverse' evaluation.
  • name: The name to give this op.
  • **kwargs: Named arguments forwarded to subclass implementation.

Returns:

Tensor, if this bijector is injective. If not injective, returns the k-tuple containing the unique k points (x1, ..., xk) such that g(xi) = y.

Raises:

  • TypeError: if self.dtype is specified and y.dtype is not self.dtype.
  • NotImplementedError: if _inverse is not implemented.

inverse_event_shape

View source

inverse_event_shape(output_shape)

Shape of a single sample from a single batch as a TensorShape.

Same meaning as inverse_event_shape_tensor. May be only partially defined.

Args:

  • output_shape: TensorShape indicating event-portion shape passed into inverse function.

Returns:

  • inverse_event_shape_tensor: TensorShape indicating event-portion shape after applying inverse. Possibly unknown.

inverse_event_shape_tensor

View source

inverse_event_shape_tensor(
    output_shape,
    name='inverse_event_shape_tensor'
)

Shape of a single sample from a single batch as an int32 1D Tensor.

Args:

  • output_shape: Tensor, int32 vector indicating event-portion shape passed into inverse function.
  • name: name to give to the op

Returns:

  • inverse_event_shape_tensor: Tensor, int32 vector indicating event-portion shape after applying inverse.

inverse_log_det_jacobian

View source

inverse_log_det_jacobian(
    y,
    event_ndims,
    name='inverse_log_det_jacobian',
    **kwargs
)

Returns the (log o det o Jacobian o inverse)(y).

Mathematically, returns: log(det(dX/dY))(Y). (Recall that: X=g^{-1}(Y).)

Note that forward_log_det_jacobian is the negative of this function, evaluated at g^{-1}(y).

Args:

  • y: Tensor. The input to the 'inverse' Jacobian determinant evaluation.
  • event_ndims: Number of dimensions in the probabilistic events being transformed. Must be greater than or equal to self.inverse_min_event_ndims. The result is summed over the final dimensions to produce a scalar Jacobian determinant for each event, i.e. it has shape rank(y) - event_ndims dimensions.
  • name: The name to give this op.
  • **kwargs: Named arguments forwarded to subclass implementation.

Returns:

  • ildj: Tensor, if this bijector is injective. If not injective, returns the tuple of local log det Jacobians, log(det(Dg_i^{-1}(y))), where g_i is the restriction of g to the ith partition Di.

Raises:

  • TypeError: if self.dtype is specified and y.dtype is not self.dtype.
  • NotImplementedError: if _inverse_log_det_jacobian is not implemented.

with_name_scope

with_name_scope(
    cls,
    method
)

Decorator to automatically enter the module name scope.

class MyModule(tf.Module):
  @tf.Module.with_name_scope
  def __call__(self, x):
    if not hasattr(self, 'w'):
      self.w = tf.Variable(tf.random.normal([x.shape[1], 64]))
    return tf.matmul(x, self.w)

Using the above module would produce tf.Variables and tf.Tensors whose names included the module name:

mod = MyModule()
mod(tf.ones([8, 32]))
# ==> <tf.Tensor: ...>
mod.w
# ==> <tf.Variable ...'my_module/w:0'>

Args:

  • method: The method to wrap.

Returns:

The original method wrapped such that it enters the module's name scope.