<br />

|---|
| [![](https://www.tensorflow.org/images/GitHub-Mark-32px.png) View source on GitHub](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/experimental/distributions/importance_resample.py#L33-L645) |

Models the distribution of finitely many importance-reweighted samples.

Inherits From: [`Distribution`](https://www.tensorflow.org/probability/api_docs/python/tfp/distributions/Distribution)

    tfp.experimental.distributions.ImportanceResample(
        proposal_distribution,
        target_log_prob_fn,
        importance_sample_size,
        sample_size=1,
        stochastic_approximation_seed=None,
        validate_args=False,
        name=None
    )

This wrapper adapts a proposal distribution towards a target density using
[importance sampling](https://en.wikipedia.org/wiki/Importance_sampling).
Given a proposal `q`, a target density `p` (which may be unnormalized), and
an integer `importance_sample_size`, it models the result of the following
sampling process:

1. Draw `importance_sample_size` samples `z[k] ~ q` from the proposal.
2. Compute an importance weight `w[k] = p(z[k]) / q(z[k])` for each sample.
3. Return a sample `z[k*]` selected with probability proportional to the importance weights, i.e., with `k* ~ Categorical(probs=w/sum(w))`.

In the limit where `importance_sample_size -> inf`, the result `z[k*]` of this
procedure would be distributed according to the target density `p`. On the
other hand, if `importance_sample_size == 1`, then the reweighting has no
effect and the result `z[k*]` is simply a sample from `q`. Finite values
of `importance_sample_size` describe distributions that are intermediate
between `p` and `q`.

This distribution may also be understood as an explicit representation of the
surrogate posterior that is implicitly assumed by importance-weighted
variational objectives. \[1, 2\]

#### Examples

This distribution can be used directly for posterior inference via importance
sampling:

    tfd = tfp.distributions
    tfed = tfp.experimental.distributions

    def target_log_prob_fn(x):
      prior = tfd.Normal(loc=0., scale=1.).log_prob(x)
      # Multimodal likelihood.
      likelihood = tf.reduce_logsumexp(
        tfd.Normal(loc=x, scale=0.1).log_prob([-1., 1.]))
      return prior + likelihood

    # Use importance sampling to infer an approximate posterior.
    approximate_posterior = tfed.ImportanceResample(
      proposal_distribution=tfd.Normal(loc=0., scale=2.),
      target_log_prob_fn=target_log_prob_fn,
      importance_sample_size=100)

We can estimate posterior expectations directly using an importance-weighted
sum of proposal samples:

    # Directly compute expectations under the posterior via importance weights.
    posterior_mean = approximate_posterior.self_normalized_expectation(
      lambda x: x, importance_sample_size=1000)
    posterior_variance = approximate_posterior.self_normalized_expectation(
      lambda x: (x - posterior_mean)**2, importance_sample_size=1000)

Alternately, the same expectations can be estimated from explicit (unweighted)
samples. Note that sampling may be expensive because it performs resampling
internally. For example, to produce `sample_size` samples requires first
proposing values of shape `[sample_size, importance_sample_size]`
(`[1000, 100]` in the code below) and then resampling down to `[sample_size]`,
throwing most of the proposals away. For this reason you should prefer calling
`self_normalized_expectation` over naive sampling to compute expectations.

    posterior_samples = approximate_posterior.sample(1000)
    posterior_mean_inefficient = tf.reduce_mean(posterior_samples)
    posterior_variance_inefficient = tf.math.reduce_variance(posterior_samples)

    # Calling `self_normalized_expectation` allows for a much lower `sample_size`
    # because it uses the full set of `importance_sample_size` proposal samples to
    # approximate the expectation at each of the `sample_size` Monte Carlo
    # evaluations. This is formalized in Eq. 9 of [3].
    posterior_mean_efficient = approximate_posterior.self_normalized_expectation(
      lambda x: x, sample_size=10)
    posterior_variance_efficient = (
      approximate_posterior.self_normalized_expectation(
        lambda x: (x - posterior_mean_efficient)**2, sample_size=10))

The posterior (log-)density cannot be computed directly, but may be
stochastically approximated. The `prob` and `log_prob` methods accept
arguments `seed` and `sample_size` to control the variance of the
approximation.

    # Plot the posterior density.
    from matplotlib import pylab as plt
    xs = tf.linspace(-3., 3., 101)
    probs = approximate_posterior.prob(xs, sample_size=10, seed=(42, 42))
    plt.plot(xs, probs)

#### Connections to importance-weighted variational inference

Optimizing an importance-weighted variational bound provides a natural
approach to choose a proposal distribution for importance sampling.
Importance-weighted bounds are available directly in TFP via the
`importance_sample_size` argument to [`tfp.vi.monte_carlo_variational_loss`](https://www.tensorflow.org/probability/api_docs/python/tfp/vi/monte_carlo_variational_loss)
and [`tfp.vi.fit_surrogate_posterior`](https://www.tensorflow.org/probability/api_docs/python/tfp/vi/fit_surrogate_posterior). For example, we might improve on the
example above by replacing the fixed proposal distribution with a learned
proposal:

    proposal_distribution = tfp.experimental.util.make_trainable(tfd.Normal)
    importance_sample_size = 100
    importance_weighted_losses = tfp.vi.fit_surrogate_posterior(
      target_log_prob_fn,
      surrogate_posterior=proposal_distribution,
      optimizer=tf_keras.optimizers.Adam(0.1),
      num_steps=200,
      importance_sample_size=importance_sample_size)
    approximate_posterior = tfed.ImportanceResample(
      proposal_distribution=proposal_distribution,
      target_log_prob_fn=target_log_prob_fn,
      importance_sample_size=importance_sample_size)

Note that although the importance-resampled `approximate_posterior` serves
ultimately as the surrogate posterior, only the bare proposal distribution
is passed as the `surrogate_posterior` argument to `fit_surrogate_posterior`.
This is because the `importance_sample_size` argument tells
`fit_surrogate_posterior` to compute an importance-weighted bound directly
from the proposal distribution. Mathematically, it would be equivalent to omit
the `importance_sample_size` argument and instead pass an `ImportanceResample`
distribution as the surrogate posterior:

    equivalent_but_less_efficient_losses = tfp.vi.fit_surrogate_posterior(
      target_log_prob_fn,
      surrogate_posterior=tfed.ImportanceResample(
        proposal_distribution=proposal_distribution,
        target_log_prob_fn=target_log_prob_fn,
        importance_sample_size=importance_sample_size),
      optimizer=tf_keras.optimizers.Adam(0.1),
      num_steps=200)

but this approach is not recommended, because it performs redundant
evaluations of the `target_log_prob_fn` compared to the direct bound shown
above.

#### References

\[1\] Yuri Burda, Roger Grosse, and Ruslan Salakhutdinov. Importance Weighted
Autoencoders. In *International Conference on Learning
Representations* , 2016. <https://arxiv.org/abs/1509.00519>
\[2\] Chris Cremer, Quaid Morris, and David Duvenaud. Reinterpreting
Importance-Weighted Autoencoders. In *International Conference on Learning
Representations* , Workshop track, 2017. <https://arxiv.org/abs/1704.02916>
\[3\] Justin Domke, Daniel Sheldon. Importance Weighting and Variational
Inference. In *Neural Information Processing Systems (NIPS)* , 2018.
<https://arxiv.org/abs/1808.09034>

<br />

<br />

<br />

| ## Args ||
|---|---|
| `proposal_distribution` | Instance of `tfd.Distribution` used to generate proposals. This may be a joint distribution. |
| `target_log_prob_fn` | Python `callable` representation of a (potentially unnormalized) target log-density. This should accept samples from the proposal, i.e., `lp = target_log_prob_fn(proposal_distribution.sample())`. |
| `importance_sample_size` | integer `Tensor` number of proposals used in the distribution of a single sample. Larger values better approximate the target distribution, at the cost of increased computation and memory usage. |
| `sample_size` | integer `Tensor` number of Monte Carlo samples used to reduce variance in stochastic methods such as `log_prob`, `prob`, and `self_normalized_expectation`. Note that increasing `importance_sample_size` leads to a more accurate approximation of the target distribution (reducing bias and variance), while increasing `sample_size` improves the precision of estimates under the intermediate distribution corresponding to a particular finite `importance_sample_size` (i.e., it reduces variance only and does not affect the sampling distribution). If unsure, it's generally safe to leave `sample_size` at its default value of `1` and focus on increasing `importance_sample_size` instead. Default value: `1`. |
| `stochastic_approximation_seed` | optional PRNG key used in stochastic approximations for methods such as `log_prob`, `prob`, and `self_normalized_expectation`. This seed does not affect sampling. Default value: `None`. |
| `validate_args` | Python `bool`. Whether to validate input with asserts. If `validate_args` is `False`, and the inputs are invalid, correct behavior is not guaranteed. Default value: `False`. |
| `name` | Python `str` name for this distribution. If `None`, defaults to 'importance_resample'. Default value: `None`. |

<br />

<br />

<br />

<br />

| ## Attributes ||
|---|---|
| `allow_nan_stats` | Python `bool` describing behavior when a stat is undefined. <br /> Stats return +/- infinity when it makes sense. E.g., the variance of a Cauchy distribution is infinity. However, sometimes the statistic is undefined, e.g., if a distribution's pdf does not achieve a maximum within the support of the distribution, the mode is undefined. If the mean is undefined, then by definition the variance is undefined. E.g. the mean for Student's T for df = 1 is undefined (no clear way to say it is either + or - infinity), so the variance = E\[(X - mean)\*\*2\] is also undefined. |
| `batch_shape` | Shape of a single sample from a single event index as a `TensorShape`. <br /> May be partially defined or unknown. The batch dimensions are indexes into independent, non-identical parameterizations of this distribution. |
| `dtype` | The `DType` of `Tensor`s handled by this `Distribution`. |
| `event_shape` | Shape of a single sample from a single batch as a `TensorShape`. <br /> May be partially defined or unknown. |
| `experimental_shard_axis_names` | The list or structure of lists of active shard axis names. |
| `importance_sample_size` | <br /> <br /> |
| `name` | Name prepended to all ops created by this `Distribution`. |
| `name_scope` | Returns a [`tf.name_scope`](https://www.tensorflow.org/api_docs/python/tf/name_scope) instance for this class. |
| `non_trainable_variables` | Sequence of non-trainable variables owned by this module and its submodules. <br /> > [!NOTE] > **Note:** this method uses reflection to find variables on the current instance and submodules. For performance reasons you may wish to cache the result of calling this method if you don't expect the return value to change. |
| `parameters` | Dictionary of parameters used to instantiate this `Distribution`. |
| `proposal_distribution` | <br /> |
| `reparameterization_type` | Describes how samples from the distribution are reparameterized. <br /> Currently this is one of the static instances `tfd.FULLY_REPARAMETERIZED` or `tfd.NOT_REPARAMETERIZED`. |
| `sample_size` | <br /> <br /> |
| `stochastic_approximation_seed` | <br /> <br /> |
| `submodules` | Sequence of all sub-modules. <br /> 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 <br /> |
| `target_log_prob_fn` | <br /> <br /> |
| `trainable_variables` | Sequence of trainable variables owned by this module and its submodules. <br /> > [!NOTE] > **Note:** this method uses reflection to find variables on the current instance and submodules. For performance reasons you may wish to cache the result of calling this method if you don't expect the return value to change. |
| `validate_args` | Python `bool` indicating possibly expensive checks are enabled. |
| `variables` | Sequence of variables owned by this module and its submodules. > [!NOTE] > **Note:** this method uses reflection to find variables on the current instance and submodules. For performance reasons you may wish to cache the result of calling this method if you don't expect the return value to change. |

## Methods

### `batch_shape_tensor`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L992-L1030)

    batch_shape_tensor(
        name='batch_shape_tensor'
    )

Shape of a single sample from a single event index as a 1-D `Tensor`.

The batch dimensions are indexes into independent, non-identical
parameterizations of this distribution.

<br />

<br />

<br />

| Args ||
|---|---|
| `name` | name to give to the op |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `batch_shape` | `Tensor`. |

<br />

### `cdf`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1411-L1429)

    cdf(
        value, name='cdf', **kwargs
    )

Cumulative distribution function.

Given random variable `X`, the cumulative distribution function `cdf` is:

    cdf(x) := P[X <= x]

<br />

<br />

<br />

| Args ||
|---|---|
| `value` | `float` or `double` `Tensor`. |
| `name` | Python `str` prepended to names of ops created by this function. |
| `**kwargs` | Named arguments forwarded to subclass implementation. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `cdf` | a `Tensor` of shape `sample_shape(x) + self.batch_shape` with values of type `self.dtype`. |

<br />

### `copy`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L896-L924)

    copy(
        **override_parameters_kwargs
    )

Creates a deep copy of the distribution.

> [!NOTE]
> **Note:** the copy distribution may continue to depend on the original initialization arguments.

<br />

<br />

<br />

| Args ||
|---|---|
| `**override_parameters_kwargs` | String/value dictionary of initialization arguments to override with new values. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `distribution` | A new instance of `type(self)` initialized from the union of self.parameters and override_parameters_kwargs, i.e., `dict(self.parameters, **override_parameters_kwargs)`. |

<br />

### `covariance`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1648-L1686)

    covariance(
        name='covariance', **kwargs
    )

Covariance.

Covariance is (possibly) defined only for non-scalar-event distributions.

For example, for a length-`k`, vector-valued distribution, it is calculated
as,

    Cov[i, j] = Covariance(X_i, X_j) = E[(X_i - E[X_i]) (X_j - E[X_j])]

where `Cov` is a (batch of) `k x k` matrix, `0 <= (i, j) < k`, and `E`
denotes expectation.

Alternatively, for non-vector, multivariate distributions (e.g.,
matrix-valued, Wishart), `Covariance` shall return a (batch of) matrices
under some vectorization of the events, i.e.,

    Cov[i, j] = Covariance(Vec(X)_i, Vec(X)_j) = [as above]

where `Cov` is a (batch of) `k' x k'` matrices,
`0 <= (i, j) < k' = reduce_prod(event_shape)`, and `Vec` is some function
mapping indices of this distribution's event dimensions to indices of a
length-`k'` vector.

<br />

<br />

<br />

| Args ||
|---|---|
| `name` | Python `str` prepended to names of ops created by this function. |
| `**kwargs` | Named arguments forwarded to subclass implementation. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `covariance` | Floating-point `Tensor` with shape `[B1, ..., Bn, k', k']` where the first `n` dimensions are batch coordinates and `k' = reduce_prod(self.event_shape)`. |

<br />

### `cross_entropy`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1701-L1724)

    cross_entropy(
        other, name='cross_entropy'
    )

Computes the (Shannon) cross entropy.

Denote this distribution (`self`) by `P` and the `other` distribution by
`Q`. Assuming `P, Q` are absolutely continuous with respect to
one another and permit densities `p(x) dr(x)` and `q(x) dr(x)`, (Shannon)
cross entropy is defined as:

    H[P, Q] = E_p[-log q(X)] = -int_F p(x) log q(x) dr(x)

where `F` denotes the support of the random variable `X ~ P`.

<br />

<br />

<br />

| Args ||
|---|---|
| `other` | [`tfp.distributions.Distribution`](https://www.tensorflow.org/probability/api_docs/python/tfp/distributions/Distribution) instance. |
| `name` | Python `str` prepended to names of ops created by this function. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `cross_entropy` | `self.dtype` `Tensor` with shape `[B1, ..., Bn]` representing `n` different calculations of (Shannon) cross entropy. |

<br />

### `entropy`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1524-L1527)

    entropy(
        name='entropy', **kwargs
    )

Shannon entropy in nats.

### `event_shape_tensor`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1099-L1125)

    event_shape_tensor(
        name='event_shape_tensor'
    )

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

<br />

<br />

<br />

| Args ||
|---|---|
| `name` | name to give to the op |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `event_shape` | `Tensor`. |

<br />

### `experimental_default_event_space_bijector`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1768-L1797)

    experimental_default_event_space_bijector(
        *args, **kwargs
    )

Bijector mapping the reals (R\*\*n) to the event space of the distribution.

Distributions with continuous support may implement
`_default_event_space_bijector` which returns a subclass of
[`tfp.bijectors.Bijector`](https://www.tensorflow.org/probability/api_docs/python/tfp/bijectors/Bijector) that maps R\*\*n to the distribution's event space.
For example, the default bijector for the `Beta` distribution
is [`tfp.bijectors.Sigmoid()`](https://www.tensorflow.org/probability/api_docs/python/tfp/bijectors/Sigmoid), which maps the real line to `[0, 1]`, the
support of the `Beta` distribution. The default bijector for the
`CholeskyLKJ` distribution is [`tfp.bijectors.CorrelationCholesky`](https://www.tensorflow.org/probability/api_docs/python/tfp/bijectors/CorrelationCholesky), which
maps R\^(k \* (k-1) // 2) to the submanifold of k x k lower triangular
matrices with ones along the diagonal.

The purpose of `experimental_default_event_space_bijector` is
to enable gradient descent in an unconstrained space for Variational
Inference and Hamiltonian Monte Carlo methods. Some effort has been made to
choose bijectors such that the tails of the distribution in the
unconstrained space are between Gaussian and Exponential.

For distributions with discrete event space, or for which TFP currently
lacks a suitable bijector, this function returns `None`.

<br />

<br />

<br />

| Args ||
|---|---|
| `*args` | Passed to implementation `_default_event_space_bijector`. |
| `**kwargs` | Passed to implementation `_default_event_space_bijector`. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `event_space_bijector` | `Bijector` instance or `None`. |

<br />

### `experimental_fit`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1799-L1847)

    @classmethod
    experimental_fit(
        value, sample_ndims=1, validate_args=False, **init_kwargs
    )

Instantiates a distribution that maximizes the likelihood of `x`.

<br />

<br />

<br />

| Args ||
|---|---|
| `value` | a `Tensor` valid sample from this distribution family. |
| `sample_ndims` | Positive `int` Tensor number of leftmost dimensions of `value` that index i.i.d. samples. Default value: `1`. |
| `validate_args` | Python `bool`, default `False`. When `True`, distribution parameters are checked for validity despite possibly degrading runtime performance. When `False`, invalid inputs may silently render incorrect outputs. Default value: `False`. |
| `**init_kwargs` | Additional keyword arguments passed through to `cls.__init__`. These take precedence in case of collision with the fitted parameters; for example, `tfd.Normal.experimental_fit([1., 1.], scale=20.)` returns a Normal distribution with `scale=20.` rather than the maximum likelihood parameter `scale=0.`. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `maximum_likelihood_instance` | instance of `cls` with parameters that maximize the likelihood of `value`. |

<br />

### `experimental_local_measure`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1872-L1914)

    experimental_local_measure(
        value, backward_compat=False, **kwargs
    )

Returns a log probability density together with a `TangentSpace`.

A `TangentSpace` allows us to calculate the correct push-forward
density when we apply a transformation to a `Distribution` on
a strict submanifold of R\^n (typically via a `Bijector` in the
`TransformedDistribution` subclass). The density correction uses
the basis of the tangent space.

<br />

<br />

<br />

| Args ||
|---|---|
| `value` | `float` or `double` `Tensor`. |
| `backward_compat` | `bool` specifying whether to fall back to returning `FullSpace` as the tangent space, and representing R\^n with the standard basis. |
| `**kwargs` | Named arguments forwarded to subclass implementation. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `log_prob` | a `Tensor` representing the log probability density, of shape `sample_shape(x) + self.batch_shape` with values of type `self.dtype`. |
| `tangent_space` | a `TangentSpace` object (by default `FullSpace`) representing the tangent space to the manifold at `value`. |

<br />

<br />

<br />

<br />

| Raises ||
|---|---|
| UnspecifiedTangentSpaceError if `backward_compat` is False and the `_experimental_tangent_space` attribute has not been defined. ||

<br />

### `experimental_sample_and_log_prob`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1229-L1259)

    experimental_sample_and_log_prob(
        sample_shape=(), seed=None, name='sample_and_log_prob', **kwargs
    )

Samples from this distribution and returns the log density of the sample.

The default implementation simply calls `sample` and `log_prob`:

    def _sample_and_log_prob(self, sample_shape, seed, **kwargs):
      x = self.sample(sample_shape=sample_shape, seed=seed, **kwargs)
      return x, self.log_prob(x, **kwargs)

However, some subclasses may provide more efficient and/or numerically
stable implementations.

Additional documentation from `ImportanceResample`:

Re-use proposal samples for lower variance in the log-prob estimate.

    Note: this method reuses the same proposal samples `z[k]` for both sampling
    and approximate `log_prob` evaluation. Thus, calling `sample_and_log_prob`
    is *not* equivalent to calling `sample` followed by `log_prob`, which would
    use two independent sets of proposal samples and in general return a
    different stochastic approximation to the log-density of the sampled points.

    In particular, `log_prob` returns a stochastic lower bound (which becomes
    tighter as `sample_size` increases) on the log-density of the
    importance-resampled distribution , while this method returns a
    single-sample stochastic *upper* bound. This guarantees that plugging an
    `ImportanceResample` surrogate posterior into a variational evidence lower
    bound (ELBO) preserves a valid lower bound---in fact, the IWAE bound [1]---
    which would otherwise not be the case for `log_prob` with finite values of
    `sample_size`. (This said, explicitly computing an IWAE bound via
    <a href="../../../tfp/vi/monte_carlo_variational_loss"><code>tfp.vi.monte_carlo_variational_loss</code></a> is more efficient and stable than
    this implicit construction using an `ImportanceResample` surrogate, and so
    should be the preferred approach in general.)

    #### Mathematical details

    The `log_prob` estimate computed in this method is given by

    ```
    surrogate_log_prob(x) = target_log_prob_fn(x) - log(mean(weights(z)))
    ```

    where
    `weights(z)[k] = exp(target_log_prob_fn(z[k]) - proposal.log_prob(z[k]))`
    are the importance weights of the proposal samples `z[k]` from which `x` was
    selected. Since we know that we selected `x` from among these
    proposal samples, we may conclude that these samples are more likely
    to lead to us selecting `x` than would be the case for 'typical' proposal
    samples in the absence of such knowledge. The implied estimate of
    `prob(x)` is therefore biased upwards.

    The motivation for this estimate is that plugging it into the ELBO recovers
    the IWAE objective:

    ```
      ELBO = target_log_prob(x) - surrogate_log_prob(x)
               (for x ~ surrogate)
           = target_log_prob(x) - (target_log_prob(x) - log(mean(weights(z))))
               (for z[k] ~ proposal)
           = log(mean(weights(z)))
           = IWAE
    ```

    Because the IWAE objective lower-bounds the *true* ELBO
    of the importance-resampled distribution (i.e., the ELBO that we would
    compute using
    `surrogate_log_prob(x) = ImportanceResample.log_prob(x, sample_size=inf)`;
    see section 5.3 of Cremer et al. [2]),
    it follows that the quantity `surrogate_log_prob(x)` estimated here is an
    upper bound on the *true* log_prob of the importance-resampled distribution.

##### `kwargs`:

- `importance_sample_size`: optional integer `Tensor` number of proposals used to define the distribution. If `None`, defaults to `self.importance_sample_size`.

<br />

<br />

<br />

| Args ||
|---|---|
| `sample_shape` | integer `Tensor` desired shape of samples to draw. Default value: `()`. |
| `seed` | PRNG seed; see [`tfp.random.sanitize_seed`](https://www.tensorflow.org/probability/api_docs/python/tfp/random/sanitize_seed) for details. Default value: `None`. |
| `name` | name to give to the op. Default value: `'sample_and_log_prob'`. |
| `**kwargs` | Named arguments forwarded to subclass implementation. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `samples` | a `Tensor`, or structure of `Tensor`s, with prepended dimensions `sample_shape`. |
| `log_prob` | a `Tensor` of shape `sample_shape(x) + self.batch_shape` with values of type `self.dtype`. |

<br />

### `is_scalar_batch`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1156-L1168)

    is_scalar_batch(
        name='is_scalar_batch'
    )

Indicates that `batch_shape == []`.

<br />

<br />

<br />

| Args ||
|---|---|
| `name` | Python `str` prepended to names of ops created by this function. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `is_scalar_batch` | `bool` scalar `Tensor`. |

<br />

### `is_scalar_event`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1142-L1154)

    is_scalar_event(
        name='is_scalar_event'
    )

Indicates that `event_shape == []`.

<br />

<br />

<br />

| Args ||
|---|---|
| `name` | Python `str` prepended to names of ops created by this function. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `is_scalar_event` | `bool` scalar `Tensor`. |

<br />

### `kl_divergence`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1730-L1761)

    kl_divergence(
        other, name='kl_divergence'
    )

Computes the Kullback--Leibler divergence.

Denote this distribution (`self`) by `p` and the `other` distribution by
`q`. Assuming `p, q` are absolutely continuous with respect to reference
measure `r`, the KL divergence is defined as:

    KL[p, q] = E_p[log(p(X)/q(X))]
             = -int_F p(x) log q(x) dr(x) + int_F p(x) log p(x) dr(x)
             = H[p, q] - H[p]

where `F` denotes the support of the random variable `X ~ p`, `H[., .]`
denotes (Shannon) cross entropy, and `H[.]` denotes (Shannon) entropy.

<br />

<br />

<br />

| Args ||
|---|---|
| `other` | [`tfp.distributions.Distribution`](https://www.tensorflow.org/probability/api_docs/python/tfp/distributions/Distribution) instance. |
| `name` | Python `str` prepended to names of ops created by this function. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `kl_divergence` | `self.dtype` `Tensor` with shape `[B1, ..., Bn]` representing `n` different calculations of the Kullback-Leibler divergence. |

<br />

### `log_cdf`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1373-L1395)

    log_cdf(
        value, name='log_cdf', **kwargs
    )

Log cumulative distribution function.

Given random variable `X`, the cumulative distribution function `cdf` is:

    log_cdf(x) := Log[ P[X <= x] ]

Often, a numerical approximation can be used for `log_cdf(x)` that yields
a more accurate answer than simply taking the logarithm of the `cdf` when
`x << -1`.

<br />

<br />

<br />

| Args ||
|---|---|
| `value` | `float` or `double` `Tensor`. |
| `name` | Python `str` prepended to names of ops created by this function. |
| `**kwargs` | Named arguments forwarded to subclass implementation. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `logcdf` | a `Tensor` of shape `sample_shape(x) + self.batch_shape` with values of type `self.dtype`. |

<br />

### `log_prob`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1275-L1287)

    log_prob(
        value, name='log_prob', **kwargs
    )

Log probability density/mass function.

Additional documentation from `ImportanceResample`:

The density of an importance-resampled distribution is
not generally available in closed form. This method follows algorithm (2) of
Cremer et al. \[2\] to compute an unbiased estimate of `prob(x)`, which
corresponds by [Jensen's inequality](https://en.wikipedia.org/wiki/Jensen%27s_inequality) to a stochastic lower
bound on `log_prob(x)`. The estimation variance decreases, and the corresponding
bound tightens, as `sample_size` increases; an infinitely large `sample_size`
would recover the true (log-)density.

##### `kwargs`:

- `importance_sample_size`: optional integer `Tensor` number of proposals used to define the distribution. If not specified, defaults to `self.importance_sample_size`.
- `sample_size`: int `Tensor` number of samples used to reduce variance in the estimated density for a given `importance_sample_size`. If `None`, defaults to `self.sample_size`.
- `seed`: PRNG seed; see [`tfp.random.sanitize_seed`](https://www.tensorflow.org/probability/api_docs/python/tfp/random/sanitize_seed) for details.

<br />

<br />

<br />

| Args ||
|---|---|
| `value` | `float` or `double` `Tensor`. |
| `name` | Python `str` prepended to names of ops created by this function. |
| `**kwargs` | Named arguments forwarded to subclass implementation. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `log_prob` | a `Tensor` of shape `sample_shape(x) + self.batch_shape` with values of type `self.dtype`. |

<br />

### `log_survival_function`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1452-L1476)

    log_survival_function(
        value, name='log_survival_function', **kwargs
    )

Log survival function.

Given random variable `X`, the survival function is defined:

    log_survival_function(x) = Log[ P[X > x] ]
                             = Log[ 1 - P[X <= x] ]
                             = Log[ 1 - cdf(x) ]

Typically, different numerical approximations can be used for the log
survival function, which are more accurate than `1 - cdf(x)` when `x >> 1`.

<br />

<br />

<br />

| Args ||
|---|---|
| `value` | `float` or `double` `Tensor`. |
| `name` | Python `str` prepended to names of ops created by this function. |
| `**kwargs` | Named arguments forwarded to subclass implementation. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `Tensor` of shape `sample_shape(x) + self.batch_shape` with values of type `self.dtype`. ||

<br />

### `mean`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1533-L1536)

    mean(
        name='mean', **kwargs
    )

Mean.

### `mode`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1692-L1695)

    mode(
        name='mode', **kwargs
    )

Mode.

### `param_shapes`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L701-L727)

    @classmethod
    param_shapes(
        sample_shape, name='DistributionParamShapes'
    )

Shapes of parameters given the desired shape of a call to `sample()`. (deprecated)

> [!WARNING]
> **Deprecated:** THIS FUNCTION IS DEPRECATED. It will be removed after 2021-03-01. Instructions for updating: The `param_shapes` method of `tfd.Distribution` is deprecated; use `parameter_properties` instead.

This is a class method that describes what key/value arguments are required
to instantiate the given `Distribution` so that a particular shape is
returned for that instance's call to `sample()`.

Subclasses should override class method `_param_shapes`.

<br />

<br />

<br />

| Args ||
|---|---|
| `sample_shape` | `Tensor` or python list/tuple. Desired shape of a call to `sample()`. |
| `name` | name to prepend ops with. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `dict` of parameter name to `Tensor` shapes. ||

<br />

### `param_static_shapes`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L729-L769)

    @classmethod
    param_static_shapes(
        sample_shape
    )

param_shapes with static (i.e. `TensorShape`) shapes. (deprecated)

> [!WARNING]
> **Deprecated:** THIS FUNCTION IS DEPRECATED. It will be removed after 2021-03-01. Instructions for updating: The `param_static_shapes` method of `tfd.Distribution` is deprecated; use `parameter_properties` instead.

This is a class method that describes what key/value arguments are required
to instantiate the given `Distribution` so that a particular shape is
returned for that instance's call to `sample()`. Assumes that the sample's
shape is known statically.

Subclasses should override class method `_param_shapes` to return
constant-valued tensors when constant values are fed.

<br />

<br />

<br />

| Args ||
|---|---|
| `sample_shape` | `TensorShape` or python list/tuple. Desired shape of a call to `sample()`. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `dict` of parameter name to `TensorShape`. ||

<br />

<br />

<br />

<br />

| Raises ||
|---|---|
| `ValueError` | if `sample_shape` is a `TensorShape` and is not fully defined. |

<br />

### `parameter_properties`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L655-L699)

    @classmethod
    parameter_properties(
        dtype=tf.float32, num_classes=None
    )

Returns a dict mapping constructor arg names to property annotations.

This dict should include an entry for each of the distribution's
`Tensor`-valued constructor arguments.

Distribution subclasses are not required to implement
`_parameter_properties`, so this method may raise `NotImplementedError`.
Providing a `_parameter_properties` implementation enables several advanced
features, including:

- Distribution batch slicing (`sliced_distribution = distribution[i:j]`).
- Automatic inference of `_batch_shape` and `_batch_shape_tensor`, which must otherwise be computed explicitly.
- Automatic instantiation of the distribution within TFP's internal property tests.
- Automatic construction of 'trainable' instances of the distribution using appropriate bijectors to avoid violating parameter constraints. This enables the distribution family to be used easily as a surrogate posterior in variational inference.

In the future, parameter property annotations may enable additional
functionality; for example, returning Distribution instances from
[`tf.vectorized_map`](https://www.tensorflow.org/api_docs/python/tf/vectorized_map).

<br />

<br />

<br />

| Args ||
|---|---|
| `dtype` | Optional float `dtype` to assume for continuous-valued parameters. Some constraining bijectors require advance knowledge of the dtype because certain constants (e.g., `tfb.Softplus.low`) must be instantiated with the same dtype as the values to be transformed. |
| `num_classes` | Optional `int` `Tensor` number of classes to assume when inferring the shape of parameters for categorical-like distributions. Otherwise ignored. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `parameter_properties` | A `str ->`tfp.python.internal.parameter_properties.ParameterProperties`dict mapping constructor argument names to`ParameterProperties\` instances. |

<br />

<br />

<br />

<br />

| Raises ||
|---|---|
| `NotImplementedError` | if the distribution class does not implement `_parameter_properties`. |

<br />

### `prob`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1303-L1315)

    prob(
        value, name='prob', **kwargs
    )

Probability density/mass function.

<br />

<br />

<br />

| Args ||
|---|---|
| `value` | `float` or `double` `Tensor`. |
| `name` | Python `str` prepended to names of ops created by this function. |
| `**kwargs` | Named arguments forwarded to subclass implementation. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `prob` | a `Tensor` of shape `sample_shape(x) + self.batch_shape` with values of type `self.dtype`. |

<br />

### `quantile`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1555-L1573)

    quantile(
        value, name='quantile', **kwargs
    )

Quantile function. Aka 'inverse cdf' or 'percent point function'.

Given random variable `X` and `p in [0, 1]`, the `quantile` is:

    quantile(p) := x such that P[X <= x] == p

<br />

<br />

<br />

| Args ||
|---|---|
| `value` | `float` or `double` `Tensor`. |
| `name` | Python `str` prepended to names of ops created by this function. |
| `**kwargs` | Named arguments forwarded to subclass implementation. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `quantile` | a `Tensor` of shape `sample_shape(x) + self.batch_shape` with values of type `self.dtype`. |

<br />

### `sample`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1189-L1205)

    sample(
        sample_shape=(), seed=None, name='sample', **kwargs
    )

Generate samples of the specified shape.

Note that a call to `sample()` without arguments will generate a single
sample.

Additional documentation from `ImportanceResample`:

##### `kwargs`:

- `importance_sample_size`: optional integer `Tensor` number of proposals used to define the distribution. If `None`, defaults to `self.importance_sample_size`.

<br />

<br />

<br />

| Args ||
|---|---|
| `sample_shape` | 0D or 1D `int32` `Tensor`. Shape of the generated samples. |
| `seed` | PRNG seed; see [`tfp.random.sanitize_seed`](https://www.tensorflow.org/probability/api_docs/python/tfp/random/sanitize_seed) for details. |
| `name` | name to give to the op. |
| `**kwargs` | Named arguments forwarded to subclass implementation. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `samples` | a `Tensor` with prepended dimensions `sample_shape`. |

<br />

### `self_normalized_expectation`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/experimental/distributions/importance_resample.py#L577-L645)

    self_normalized_expectation(
        fn,
        importance_sample_size=None,
        sample_size=None,
        seed=None,
        name='self_normalized_expectation'
    )

Approximates the expectation of fn(x).

This function applies self-normalized importance sampling with the given
proposal distribution to approximate expectations under the target
distribution. By using all of the `importance_sample_size` proposal
samples to approximate the expectation, this will in general give
lower-variance estimates than those obtained by explicit sampling
(`tf.reduce_sum(fn(self.sample(sample_size)), axis=0)`), since the latter
returns only one point from each set of `importance_sample_size` proposals.

Concretely, this function draws `importance_sample_size` samples
`x[1], x[2], ...` from
`self.proposal_distribution`, computes their importance weights
`w[k] = target_log_prob_fn(x[k]) / proposal_distribution.log_prob(x[k])`,
and returns the weighted sum
`sum(w[k]/sum(w) * fn(x[k]) for k in range(importance_sample_size))`. If
`sample_size > 1` is specified, the previous procedure is performed
multiple times and the results averaged to reduce variance.

> [!NOTE]
> **Note:** to approximate expectations under the target distribution you should prefer to increase `importance_sample_size` (which reduces both bias and variance) rather than `sample_size` (which reduces variance only). Values of `sample_size > 1` are needed only if you specifically want expectations under the intermediate distribution that arises from considering a particular finite number of importance samples.

<br />

<br />

<br />

| Args ||
|---|---|
| `fn` | Python `callable` that takes samples from `self.proposal_distribution` and returns a (structure of) `Tensor` value(s). This may represent a prediction derived from a posterior sample, or even a simple statistic; for example, the expectation of `fn = lambda x: x` is the posterior mean. |
| `importance_sample_size` | int `Tensor` number of samples used to define the distribution under which the expectation is taken. If `None`, defaults to `self.importance_sample_size`. Default value: `None`. |
| `sample_size` | int `Tensor` number of samples used to reduce variance in the expectation for a given `importance_sample_size`. If `None`, defaults to `self.sample_size`. Default value: `None`. |
| `seed` | PRNG seed; see [`tfp.random.sanitize_seed`](https://www.tensorflow.org/probability/api_docs/python/tfp/random/sanitize_seed) for details. If `None`, defaults to `self.stochastic_approximation_seed`. Default value: `None`. |
| `name` | Python string name for ops created by this function. Default value: `self_normalized_expectation`. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `expected_value` | (structure of) `Tensor` value(s) estimate of the expectation of `fn(x)` under the target distribution. |

<br />

### `stddev`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1613-L1642)

    stddev(
        name='stddev', **kwargs
    )

Standard deviation.

Standard deviation is defined as,

    stddev = E[(X - E[X])**2]**0.5

where `X` is the random variable associated with this distribution, `E`
denotes expectation, and `stddev.shape = batch_shape + event_shape`.

<br />

<br />

<br />

| Args ||
|---|---|
| `name` | Python `str` prepended to names of ops created by this function. |
| `**kwargs` | Named arguments forwarded to subclass implementation. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `stddev` | Floating-point `Tensor` with shape identical to `batch_shape + event_shape`, i.e., the same shape as `self.mean()`. |

<br />

### `survival_function`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1498-L1518)

    survival_function(
        value, name='survival_function', **kwargs
    )

Survival function.

Given random variable `X`, the survival function is defined:

    survival_function(x) = P[X > x]
                         = 1 - P[X <= x]
                         = 1 - cdf(x).

<br />

<br />

<br />

| Args ||
|---|---|
| `value` | `float` or `double` `Tensor`. |
| `name` | Python `str` prepended to names of ops created by this function. |
| `**kwargs` | Named arguments forwarded to subclass implementation. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `Tensor` of shape `sample_shape(x) + self.batch_shape` with values of type `self.dtype`. ||

<br />

### `unnormalized_log_prob`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1335-L1357)

    unnormalized_log_prob(
        value, name='unnormalized_log_prob', **kwargs
    )

Potentially unnormalized log probability density/mass function.

This function is similar to `log_prob`, but does not require that the
return value be normalized. (Normalization here refers to the total
integral of probability being one, as it should be by definition for any
probability distribution.) This is useful, for example, for distributions
where the normalization constant is difficult or expensive to compute. By
default, this simply calls `log_prob`.

<br />

<br />

<br />

| Args ||
|---|---|
| `value` | `float` or `double` `Tensor`. |
| `name` | Python `str` prepended to names of ops created by this function. |
| `**kwargs` | Named arguments forwarded to subclass implementation. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `unnormalized_log_prob` | a `Tensor` of shape `sample_shape(x) + self.batch_shape` with values of type `self.dtype`. |

<br />

### `variance`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L1579-L1607)

    variance(
        name='variance', **kwargs
    )

Variance.

Variance is defined as,

    Var = E[(X - E[X])**2]

where `X` is the random variable associated with this distribution, `E`
denotes expectation, and `Var.shape = batch_shape + event_shape`.

<br />

<br />

<br />

| Args ||
|---|---|
| `name` | Python `str` prepended to names of ops created by this function. |
| `**kwargs` | Named arguments forwarded to subclass implementation. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `variance` | Floating-point `Tensor` with shape identical to `batch_shape + event_shape`, i.e., the same shape as `self.mean()`. |

<br />

### `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.Variable`](https://www.tensorflow.org/api_docs/python/tf/Variable)s and [`tf.Tensor`](https://www.tensorflow.org/api_docs/python/tf/Tensor)s 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)>

<br />

<br />

<br />

| Args ||
|---|---|
| `method` | The method to wrap. |

<br />

<br />

<br />

<br />

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

<br />

### `__getitem__`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L825-L852)

    __getitem__(
        slices
    )

Slices the batch axes of this distribution, returning a new instance.

    b = tfd.Bernoulli(logits=tf.zeros([3, 5, 7, 9]))
    b.batch_shape  # => [3, 5, 7, 9]
    b2 = b[:, tf.newaxis, ..., -2:, 1::2]
    b2.batch_shape  # => [3, 1, 5, 2, 4]

    x = tf.random.normal([5, 3, 2, 2])
    cov = tf.matmul(x, x, transpose_b=True)
    chol = tf.linalg.cholesky(cov)
    loc = tf.random.normal([4, 1, 3, 1])
    mvn = tfd.MultivariateNormalTriL(loc, chol)
    mvn.batch_shape  # => [4, 5, 3]
    mvn.event_shape  # => [2]
    mvn2 = mvn[:, 3:, ..., ::-1, tf.newaxis]
    mvn2.batch_shape  # => [4, 2, 3, 1]
    mvn2.event_shape  # => [2]

<br />

<br />

<br />

| Args ||
|---|---|
| `slices` | slices from the \[\] operator |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `dist` | A new `tfd.Distribution` instance with sliced parameters. |

<br />

### `__iter__`

[View source](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/distributions/distribution.py#L854-L855)

    __iter__()