tfc.entropy_models.LocationScaleIndexedEntropyModel

Indexed entropy model for location-scale family of random variables.

Inherits From: ContinuousIndexedEntropyModel

This class is a common special case of ContinuousIndexedEntropyModel. The specified distribution is parameterized with num_scales values of scale parameters. An element-wise location parameter is handled by shifting the distributions to zero.

This method is illustrated in Figure 10 of:

"Nonlinear Transform Coding"
J. Ballé, P.A. Chou, D. Minnen, S. Singh, N. Johnston, E. Agustsson, S.J. Hwang, G. Toderici
https://doi.org/10.1109/JSTSP.2020.3034501

prior_fn A callable returning a tfp.distributions.Distribution object, typically a Distribution class or factory function. This is a density model fitting the marginal distribution of the bottleneck data with additive uniform noise, which is shared a priori between the sender and the receiver. For best results, the distributions should be flexible enough to have a unit-width uniform distribution as a special case, since this is the marginal distribution for bottleneck dimensions that are constant. The callable will receive keyword arguments as determined by parameter_fns.
num_scales Integer. Values in indexes must be in the range [0, num_scales).
scale_fn Callable. indexes is passed to the callable, and the return value is given as scale keyword argument to prior_fn.
coding_rank Integer. Number of innermost dimensions considered a coding unit. Each coding unit is compressed to its own bit string, and the bits in the __call__ method are summed over each coding unit.
compression Boolean. If set to True, the range coding tables used by compress() and decompress() will be built on instantiation. If set to False, these two methods will not be accessible.
stateless Boolean. If False, range coding tables are created as Variables. This allows the entropy model to be serialized using the SavedModel protocol, so that both the encoder and the decoder use identical tables when loading the stored model. If True, creates range coding tables as Tensors. This makes the entropy model stateless and allows it to be constructed within a tf.function body, for when the range coding tables are provided manually. If compression=False, then stateless=True is implied and the provided value is ignored.
expected_grads If True, will use analytical expected gradients during backpropagation w.r.t. additive uniform noise.
tail_mass Float. Approximate probability mass which is encoded using an Elias gamma code embedded into the range coder.
range_coder_precision Integer. Precision passed to the range coding op.
bottleneck_dtype tf.dtypes.DType. Data type of bottleneck tensor. Defaults to tf.keras.mixed_precision.global_policy().compute_dtype.
prior_dtype tf.dtypes.DType. Data type of prior and probability computations. Defaults to tf.float32.
laplace_tail_mass Float. If positive, will augment the prior with a laplace mixture for training stability. (experimental)

bottleneck_dtype Data type of the bottleneck tensor.
cdf The CDFs used by range coding.
cdf_offset The CDF offsets used by range coding.
channel_axis Position of channel axis in indexes tensor.
coding_rank Number of innermost dimensions considered a coding unit.
compression Whether this entropy model is prepared for compression.
expected_grads Whether to use analytical expected gradients during backpropagation.
index_ranges Upper bound(s) on values allowed in indexes tensor.
laplace_tail_mass Whether to augment the prior with a Laplace mixture.
name Returns the name of this module as passed or determined in the ctor.

name_scope Returns a tf.name_scope instance for this class.
non_trainable_variables Sequence of non-trainable variables owned by this module and its submodules.
parameter_fns Functions mapping indexes to each distribution parameter.
prior Prior distribution, used for deriving range coding tables.
prior_dtype Data type of prior.
prior_fn Class or factory function returning a Distribution object.
range_coder_precision Precision used in range coding op.
stateless Whether range coding tables are created as Tensors or Variables.
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
list(a.submodules) == [b, c]
True
list(b.submodules) == [c]
True
list(c.submodules) == []
True

tail_mass Approximate probability mass which is range encoded with overflow.
trainable_variables Sequence of trainable variables owned by this module and its submodules.

variables Sequence of variables owned by this module and its submodules.

Methods

compress

View source

Compresses a floating-point tensor.

Compresses the tensor to bit strings. bottleneck is first quantized as in quantize(), and then compressed using the probability tables derived from indexes. The quantized tensor can later be recovered by calling decompress().

The innermost self.coding_rank dimensions are treated as one coding unit, i.e. are compressed into one string each. Any additional dimensions to the left are treated as batch dimensions.

Args
bottleneck tf.Tensor containing the data to be compressed.
scale_indexes tf.Tensor indexing the scale parameter for each element in bottleneck. Must have the same shape as bottleneck.
loc None or tf.Tensor. If None, the location parameter for all elements is assumed to be zero. Otherwise, specifies the location parameter for each element in bottleneck. Must have the same shape as bottleneck.

Returns
A tf.Tensor having the same shape as bottleneck without the self.coding_rank innermost dimensions, containing a string for each coding unit.

decompress

View source

Decompresses a tensor.

Reconstructs the quantized tensor from bit strings produced by compress().

Args
strings tf.Tensor containing the compressed bit strings.
scale_indexes tf.Tensor indexing the scale parameter for each output element.
loc None or tf.Tensor. If None, the location parameter for all output elements is assumed to be zero. Otherwise, specifies the location parameter for each output element. Must have the same shape as scale_indexes.

Returns
A tf.Tensor of the same shape as scale_indexes.

from_config

View source

Instantiates an entropy model from a configuration dictionary.

get_config

View source

Returns the configuration of the entropy model.

get_weights

View source

quantize

View source

Quantizes a floating-point tensor.

To use this entropy model as an information bottleneck during training, pass a tensor through this function. The tensor is rounded to integer values modulo the location parameters of the prior distribution given in loc.

The gradient of this rounding operation is overridden with the identity (straight-through gradient estimator).

Args
bottleneck tf.Tensor containing the data to be quantized.
loc None or tf.Tensor. If None, the location parameter for all elements is assumed to be zero. Otherwise, specifies the location parameter for each element in bottleneck. Must have the same shape as bottleneck.

Returns
A tf.Tensor containing the quantized values.

set_weights

View source

with_name_scope

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], 3]))
    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([1, 2]))
<tf.Tensor: shape=(1, 3), dtype=float32, numpy=..., dtype=float32)>
mod.w
<tf.Variable 'my_module/Variable:0' shape=(2, 3) dtype=float32,
numpy=..., dtype=float32)>

Args
method The method to wrap.

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

__call__

View source

Perturbs a tensor with (quantization) noise and estimates rate.

Args
bottleneck tf.Tensor containing the data to be compressed.
scale_indexes tf.Tensor indexing the scale parameter for each element in bottleneck. Must have the same shape as bottleneck.
loc None or tf.Tensor. If None, the location parameter for all elements is assumed to be zero. Otherwise, specifies the location parameter for each element in bottleneck. Must have the same shape as bottleneck.
training Boolean. If False, computes the Shannon information of bottleneck under the distribution computed by self.prior_fn, which is a non-differentiable, tight lower bound on the number of bits needed to compress bottleneck using compress(). If True, returns a somewhat looser, but differentiable upper bound on this quantity.

Returns
A tuple (bottleneck_perturbed, bits) where bottleneck_perturbed is bottleneck perturbed with (quantization) noise and bits is the rate. bits has the same shape as bottleneck without the self.coding_rank innermost dimensions.