LostTech.TensorFlow : API Documentation

Type WishartCholesky

Namespace tensorflow.contrib.distributions

Parent _WishartLinearOperator

Interfaces IWishartCholesky

The matrix Wishart distribution on positive definite matrices.

This distribution is defined by a scalar degrees of freedom `df` and a lower, triangular Cholesky factor which characterizes the scale matrix.

Using WishartCholesky is a constant-time improvement over WishartFull. It saves an O(nbk^3) operation, i.e., a matrix-product operation for sampling and a Cholesky factorization in log_prob. For most use-cases it often saves another O(nbk^3) operation since most uses of Wishart will also use the Cholesky factorization.

#### Mathematical Details

The probability density function (pdf) is,

```none pdf(X; df, scale) = det(X)**(0.5 (df-k-1)) exp(-0.5 tr[inv(scale) X]) / Z Z = 2**(0.5 df k) |det(scale)|**(0.5 df) Gamma_k(0.5 df) ```

where: * `df >= k` denotes the degrees of freedom, * `scale` is a symmetric, positive definite, `k x k` matrix, * `Z` is the normalizing constant, and, * `Gamma_k` is the [multivariate Gamma function]( https://en.wikipedia.org/wiki/Multivariate_gamma_function).

#### Examples
Show Example 
import tensorflow_probability as tfp
            tfd = tfp.distributions 
 # Initialize a single 3x3 Wishart with Cholesky factored scale matrix and 5
# degrees-of-freedom.(*)
df = 5
chol_scale = tf.linalg.cholesky(...)  # Shape is [3, 3].
dist = tfd.WishartCholesky(df=df, scale=chol_scale) 
 # Evaluate this on an observation in R^3, returning a scalar.
x =...  # A 3x3 positive definite matrix.
dist.prob(x)  # Shape is [], a scalar. 
 # Evaluate this on a two observations, each in R^{3x3}, returning a length two
# Tensor.
x = [x0, x1]  # Shape is [2, 3, 3].
dist.prob(x)  # Shape is [2]. 
 # Initialize two 3x3 Wisharts with Cholesky factored scale matrices.
df = [5, 4]
chol_scale = tf.linalg.cholesky(...)  # Shape is [2, 3, 3].
dist = tfd.WishartCholesky(df=df, scale=chol_scale) 
 # Evaluate this on four observations.
x = [[x0, x1], [x2, x3]]  # Shape is [2, 2, 3, 3].
dist.prob(x)  # Shape is [2, 2]. 
 # (*) - To efficiently create a trainable covariance matrix, see the example
#   in tfp.distributions.matrix_diag_transform.

Properties

Public properties

object allow_nan_stats get;

object allow_nan_stats_dyn get;

TensorShape batch_shape get;

object batch_shape_dyn get;

bool cholesky_input_output_matrices get;

object cholesky_input_output_matrices_dyn get;

object df get;

object df_dyn get;

object dimension get;

object dimension_dyn get;

object dtype get;

object dtype_dyn get;

TensorShape event_shape get;

object event_shape_dyn get;

string name get;

object name_dyn get;

IDictionary<object, object> parameters get;

object parameters_dyn get;

object PythonObject get;

object reparameterization_type get;

object reparameterization_type_dyn get;

object scale_operator get;

object scale_operator_dyn get;

object validate_args get;

object validate_args_dyn get;