# Module: tf.contrib.distributions

<br />

|---|
| [![](https://www.tensorflow.org/images/GitHub-Mark-32px.png) View source on GitHub](https://github.com/tensorflow/tensorflow/blob/v1.15.0/tensorflow/contrib/distributions/__init__.py) |

Classes representing statistical distributions and ops for working with them.

Use [tfp.distributions](https://www.tensorflow.org/probability/api_docs/python/tfp/distributions) instead.

## Modules

[`bijectors`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/bijectors) module: Bijector Ops.

## Classes

[`class Autoregressive`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/Autoregressive): Autoregressive distributions.

[`class BatchReshape`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/BatchReshape): The Batch-Reshaping distribution.

[`class Bernoulli`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/Bernoulli): Bernoulli distribution.

[`class Beta`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/Beta): Beta distribution.

[`class BetaWithSoftplusConcentration`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/BetaWithSoftplusConcentration): Beta with softplus transform of `concentration1` and `concentration0`.

[`class Binomial`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/Binomial): Binomial distribution.

[`class Categorical`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/Categorical): Categorical distribution.

[`class Cauchy`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/Cauchy): The Cauchy distribution with location `loc` and scale `scale`.

[`class Chi2`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/Chi2): Chi2 distribution.

[`class Chi2WithAbsDf`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/Chi2WithAbsDf): Chi2 with parameter transform `df = floor(abs(df))`.

[`class ConditionalDistribution`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/ConditionalDistribution): Distribution that supports intrinsic parameters (local latents).

[`class ConditionalTransformedDistribution`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/ConditionalTransformedDistribution): A TransformedDistribution that allows intrinsic conditioning.

[`class Deterministic`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/Deterministic): Scalar `Deterministic` distribution on the real line.

[`class Dirichlet`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/Dirichlet): Dirichlet distribution.

[`class DirichletMultinomial`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/DirichletMultinomial): Dirichlet-Multinomial compound distribution.

[`class Distribution`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/Distribution): A generic probability distribution base class.

[`class ExpRelaxedOneHotCategorical`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/ExpRelaxedOneHotCategorical): ExpRelaxedOneHotCategorical distribution with temperature and logits.

[`class Exponential`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/Exponential): Exponential distribution.

[`class ExponentialWithSoftplusRate`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/ExponentialWithSoftplusRate): Exponential with softplus transform on `rate`.

[`class Gamma`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/Gamma): Gamma distribution.

[`class GammaWithSoftplusConcentrationRate`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/GammaWithSoftplusConcentrationRate): `Gamma` with softplus of `concentration` and `rate`.

[`class Geometric`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/Geometric): Geometric distribution.

[`class HalfNormal`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/HalfNormal): The Half Normal distribution with scale `scale`.

[`class Independent`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/Independent): Independent distribution from batch of distributions.

[`class InverseGamma`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/InverseGamma): InverseGamma distribution.

[`class InverseGammaWithSoftplusConcentrationRate`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/InverseGammaWithSoftplusConcentrationRate): `InverseGamma` with softplus of `concentration` and `rate`.

[`class Kumaraswamy`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/Kumaraswamy): Kumaraswamy distribution.

[`class Laplace`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/Laplace): The Laplace distribution with location `loc` and `scale` parameters.

[`class LaplaceWithSoftplusScale`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/LaplaceWithSoftplusScale): Laplace with softplus applied to `scale`.

[`class Logistic`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/Logistic): The Logistic distribution with location `loc` and `scale` parameters.

[`class Mixture`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/Mixture): Mixture distribution.

[`class MixtureSameFamily`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/MixtureSameFamily): Mixture (same-family) distribution.

[`class Multinomial`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/Multinomial): Multinomial distribution.

[`class MultivariateNormalDiag`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/MultivariateNormalDiag): The multivariate normal distribution on `R^k`.

[`class MultivariateNormalDiagPlusLowRank`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/MultivariateNormalDiagPlusLowRank): The multivariate normal distribution on `R^k`.

[`class MultivariateNormalDiagWithSoftplusScale`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/MultivariateNormalDiagWithSoftplusScale): MultivariateNormalDiag with `diag_stddev = softplus(diag_stddev)`.

[`class MultivariateNormalFullCovariance`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/MultivariateNormalFullCovariance): The multivariate normal distribution on `R^k`.

[`class MultivariateNormalTriL`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/MultivariateNormalTriL): The multivariate normal distribution on `R^k`.

[`class NegativeBinomial`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/NegativeBinomial): NegativeBinomial distribution.

[`class Normal`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/Normal): The Normal distribution with location `loc` and `scale` parameters.

[`class NormalWithSoftplusScale`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/NormalWithSoftplusScale): Normal with softplus applied to `scale`.

[`class OneHotCategorical`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/OneHotCategorical): OneHotCategorical distribution.

[`class Poisson`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/Poisson): Poisson distribution.

[`class PoissonLogNormalQuadratureCompound`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/PoissonLogNormalQuadratureCompound): `PoissonLogNormalQuadratureCompound` distribution.

[`class QuantizedDistribution`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/QuantizedDistribution): Distribution representing the quantization `Y = ceiling(X)`.

[`class RegisterKL`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/RegisterKL): Decorator to register a KL divergence implementation function.

[`class RelaxedBernoulli`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/RelaxedBernoulli): RelaxedBernoulli distribution with temperature and logits parameters.

[`class RelaxedOneHotCategorical`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/RelaxedOneHotCategorical): RelaxedOneHotCategorical distribution with temperature and logits.

[`class ReparameterizationType`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/ReparameterizationType): Instances of this class represent how sampling is reparameterized.

[`class SeedStream`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/SeedStream): Local PRNG for amplifying seed entropy into seeds for base operations.

[`class SinhArcsinh`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/SinhArcsinh): The SinhArcsinh transformation of a distribution on `(-inf, inf)`.

[`class StudentT`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/StudentT): Student's t-distribution.

[`class StudentTWithAbsDfSoftplusScale`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/StudentTWithAbsDfSoftplusScale): StudentT with `df = floor(abs(df))` and `scale = softplus(scale)`.

[`class TransformedDistribution`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/TransformedDistribution): A Transformed Distribution.

[`class Uniform`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/Uniform): Uniform distribution with `low` and `high` parameters.

[`class VectorDeterministic`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/VectorDeterministic): Vector `Deterministic` distribution on `R^k`.

[`class VectorDiffeomixture`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/VectorDiffeomixture): VectorDiffeomixture distribution.

[`class VectorExponentialDiag`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/VectorExponentialDiag): The vectorization of the Exponential distribution on `R^k`.

[`class VectorLaplaceDiag`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/VectorLaplaceDiag): The vectorization of the Laplace distribution on `R^k`.

[`class VectorSinhArcsinhDiag`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/VectorSinhArcsinhDiag): The (diagonal) SinhArcsinh transformation of a distribution on `R^k`.

[`class WishartCholesky`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/WishartCholesky): The matrix Wishart distribution on positive definite matrices.

[`class WishartFull`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/WishartFull): The matrix Wishart distribution on positive definite matrices.

## Functions

[`assign_log_moving_mean_exp(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/assign_log_moving_mean_exp): Compute the log of the exponentially weighted moving mean of the exp.

[`assign_moving_mean_variance(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/assign_moving_mean_variance): Compute exponentially weighted moving {mean,variance} of a streaming value.

[`auto_correlation(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/auto_correlation): Auto correlation along one axis.

[`estimator_head_distribution_regression(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/estimator_head_distribution_regression): Creates a `Head` for regression under a generic distribution. (deprecated)

[`fill_triangular(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/fill_triangular): Creates a (batch of) triangular matrix from a vector of inputs.

[`fill_triangular_inverse(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/fill_triangular_inverse): Creates a vector from a (batch of) triangular matrix.

[`kl_divergence(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/distributions/kl_divergence): Get the KL-divergence KL(distribution_a \|\| distribution_b). (deprecated)

[`matrix_diag_transform(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/matrix_diag_transform): Transform diagonal of \[batch-\]matrix, leave rest of matrix unchanged.

[`moving_mean_variance(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/moving_mean_variance): Compute exponentially weighted moving {mean,variance} of a streaming value.

[`normal_conjugates_known_scale_posterior(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/normal_conjugates_known_scale_posterior): Posterior Normal distribution with conjugate prior on the mean.

[`normal_conjugates_known_scale_predictive(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/normal_conjugates_known_scale_predictive): Posterior predictive Normal distribution w. conjugate prior on the mean.

[`percentile(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/percentile): Compute the `q`-th percentile of `x`.

[`quadrature_scheme_lognormal_gauss_hermite(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/quadrature_scheme_lognormal_gauss_hermite): Use Gauss-Hermite quadrature to form quadrature on positive-reals. (deprecated)

[`quadrature_scheme_lognormal_quantiles(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/quadrature_scheme_lognormal_quantiles): Use LogNormal quantiles to form quadrature on positive-reals. (deprecated)

[`quadrature_scheme_softmaxnormal_gauss_hermite(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/quadrature_scheme_softmaxnormal_gauss_hermite): Use Gauss-Hermite quadrature to form quadrature on `K - 1` simplex. (deprecated)

[`quadrature_scheme_softmaxnormal_quantiles(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/quadrature_scheme_softmaxnormal_quantiles): Use SoftmaxNormal quantiles to form quadrature on `K - 1` simplex. (deprecated)

[`reduce_weighted_logsumexp(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/reduce_weighted_logsumexp): Computes `log(abs(sum(weight * exp(elements across tensor dimensions))))`.

[`softplus_inverse(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/softplus_inverse): Computes the inverse softplus, i.e., x = softplus_inverse(softplus(x)).

[`tridiag(...)`](https://www.tensorflow.org/versions/r1.15/api_docs/python/tf/contrib/distributions/tridiag): Creates a matrix with values set above, below, and on the diagonal.

## Other Members

- `FULLY_REPARAMETERIZED`
- `NOT_REPARAMETERIZED`