tfp.util.DeferredTensor

Variable tracking object which applies function upon convert_to_tensor.

tfp.util.DeferredTensor(
    pretransformed_input,
    transform_fn,
    dtype=None,
    shape=NONE_SPECIFIED,
    also_track=None,
    name=None
)

Example

import tensorflow.compat.v2 as tf
import tensorflow_probability as tfp
tfb = tfp.bijectors
tfd = tfp.distributions

# Note: it'd be better to use `tfp.util.TransformedVariable`;
#       this example is for illustration only.
trainable_normal = tfd.Normal(
    loc=tf.Variable(0.),
    scale=tfp.util.DeferredTensor(tf.Variable(0.), tf.math.exp))

trainable_normal.loc
# ==> <tf.Variable 'Variable:0' shape=() dtype=float32, numpy=0.0>

trainable_normal.scale
# ==> <DeferredTensor: dtype=float32, shape=[], fn=exp>

# Operators work with `DeferredTensor`.
trainable_normal.scale + 1.
# ==> 2.

with tf.GradientTape() as tape:
  negloglik = -trainable_normal.log_prob(0.5)
g = tape.gradient(negloglik, trainable_normal.trainable_variables)
# ==> (-0.5, 0.75)

Which we could then fit as:

opt = tf.optimizers.Adam(learning_rate=0.05)
loss = tf.function(lambda: -trainable_normal.log_prob(0.5), autograph=True)
for _ in range(int(1e3)):
  opt.minimize(loss, trainable_normal.trainable_variables)
trainable_normal.mean()
# ==> 0.5
trainable_normal.stddev()
# ==> (approximately) 0.0075

It is also possible to parameterize a DeferredTensor with a bijector, e.g.:

# Note: it'd be better to use `tfp.util.TransformedVariable`;
#       this example is for illustration only.
d = tfd.Normal(loc=0.,
               scale=tfp.util.DeferredTensor(tf.Variable([0.54, 1.85]),
                                             tfb.Softplus()))
d.stddev()
# ==> [1., 2.]
tf.convert_to_tensor(d.scale)
# ==> [1., 2.]

pretransformed_input object with shape, dtype properties (typically a tf.Variable) passed into transform_fn when this object is acted upon in a Tensor context, eg, tf.convert_to_tensor, +, tf.math.exp, etc.
transform_fn Python callable or tfp.bijectors.Bijector-like instance. When callable, should take pretransformed_input and return a Tensor (representing by this object).
dtype Equivalent to what would otherwise be transform_fn(pretransformed_input).dtype. Default value: None (i.e., getattr(transform_fn, 'dtype', None) or pretransformed_input.dtype).
shape Equivalent to what would otherwise be transform_fn(pretransformed_input).shape. Default value: 'None' (i.e., getattr(transform_fn, 'forward_event_shape', lambda x: x)( pretransformed_input.shape)).
also_track Optional instance or structure of instances of tf.Variable and/or tf.Module, containing any additional trainable variables that the transform_fn may access beyond the given pretransformed_input. This ensures that such variables will be correctly tracked in self.trainable_variables. Default value: None.
name Python str representing this object's name; used only in graph mode. Default value: None (i.e., (getattr(transform_fn, 'name', None) or transform_fn.__name__ + '_' + pretransformed_input.name)).

TypeError if transform_fn is not callable.
TypeError if pretransformed_input lacks dtype and/or shape properties (and dtype and/or shape arguments are unspecified).

also_track Additional variables tracked by tf.Module in self.trainable_variables.
dtype Represents the type of the elements in a Tensor.
name The string name of this object.
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.
pretransformed_input Input to transform_fn.
shape Represents the shape of a Tensor.
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

trainable_variables Sequence of trainable variables owned by this module and its submodules.
transform_fn Function which characterizes the Tensorization of this object.
variables Sequence of variables owned by this module and its submodules.

Methods

numpy

View source

numpy()

Returns (copy of) deferred values as a NumPy array or scalar.

set_shape

View source

set_shape(
    shape
)

Updates the shape of this pretransformed_input.

This method can be called multiple times, and will merge the given shape with the current shape of this object. It can be used to provide additional information about the shape of this object that cannot be inferred from the graph alone.

Args
shape A TensorShape representing the shape of this pretransformed_input, a TensorShapeProto, a list, a tuple, or None.

Raises
ValueError If shape is not compatible with the current shape of this pretransformed_input.

with_name_scope

@classmethod
with_name_scope(
    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], 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.

__abs__

View source

__abs__(
    *args, **kwargs
)

__add__

View source

__add__(
    *args, **kwargs
)

__and__

View source

__and__(
    *args, **kwargs
)

__array__

View source

__array__(
    dtype=None
)

__bool__

__bool__()

Dummy method to prevent a tensor from being used as a Python bool.

This overload raises a TypeError when the user inadvertently treats a Tensor as a boolean (most commonly in an if or while statement), in code that was not converted by AutoGraph. For example:

if tf.constant(True):  # Will raise.
  # ...

if tf.constant(5) < tf.constant(7):  # Will raise.
  # ...

Raises
TypeError.

__div__

View source

__div__(
    *args, **kwargs
)

__floordiv__

View source

__floordiv__(
    *args, **kwargs
)

__ge__

View source

__ge__(
    *args, **kwargs
)

__getitem__

View source

__getitem__(
    *args, **kwargs
)

__gt__

View source

__gt__(
    *args, **kwargs
)

__invert__

View source

__invert__(
    *args, **kwargs
)

__iter__

View source

__iter__(
    *args, **kwargs
)

__le__

View source

__le__(
    *args, **kwargs
)

__lt__

View source

__lt__(
    *args, **kwargs
)

__matmul__

View source

__matmul__(
    *args, **kwargs
)

__mod__

View source

__mod__(
    *args, **kwargs
)

__mul__

View source

__mul__(
    *args, **kwargs
)

__neg__

View source

__neg__(
    *args, **kwargs
)

__nonzero__

__nonzero__()

Dummy method to prevent a tensor from being used as a Python bool.

This is the Python 2.x counterpart to __bool__() above.

Raises
TypeError.

__or__

View source

__or__(
    *args, **kwargs
)

__pow__

View source

__pow__(
    *args, **kwargs
)

__radd__

View source

__radd__(
    *args, **kwargs
)

__rand__

View source

__rand__(
    *args, **kwargs
)

__rdiv__

View source

__rdiv__(
    *args, **kwargs
)

__rfloordiv__

View source

__rfloordiv__(
    *args, **kwargs
)

__rmatmul__

View source

__rmatmul__(
    *args, **kwargs
)

__rmod__

View source

__rmod__(
    *args, **kwargs
)

__rmul__

View source

__rmul__(
    *args, **kwargs
)

__ror__

View source

__ror__(
    *args, **kwargs
)

__rpow__

View source

__rpow__(
    *args, **kwargs
)

__rsub__

View source

__rsub__(
    *args, **kwargs
)

__rtruediv__

View source

__rtruediv__(
    *args, **kwargs
)

__rxor__

View source

__rxor__(
    *args, **kwargs
)

__sub__

View source

__sub__(
    *args, **kwargs
)

__truediv__

View source

__truediv__(
    *args, **kwargs
)

__xor__

View source

__xor__(
    *args, **kwargs
)