Stan  2.10.0
probability, sampling & optimization
Classes | Functions
stan::variational Namespace Reference

Classes

class  advi
 Automatic Differentiation Variational Inference. More...
 
class  base_family
 
class  normal_fullrank
 Variational family approximation with full-rank multivariate normal distribution. More...
 
class  normal_meanfield
 Variational family approximation with mean-field (diagonal covariance) multivariate normal distribution. More...
 

Functions

base_family operator+ (base_family lhs, const base_family &rhs)
 
base_family operator/ (base_family lhs, const base_family &rhs)
 
base_family operator+ (double scalar, base_family rhs)
 
base_family operator* (double scalar, base_family rhs)
 
normal_fullrank operator+ (normal_fullrank lhs, const normal_fullrank &rhs)
 Return a new approximation resulting from adding the mean and covariance matrix Cholesky factor of the specified approximations. More...
 
normal_fullrank operator/ (normal_fullrank lhs, const normal_fullrank &rhs)
 Return a new approximation resulting from elementwise division of of the first specified approximation by the second. More...
 
normal_fullrank operator+ (double scalar, normal_fullrank rhs)
 Return a new approximation resulting from elementwise addition of the specified scalar to the mean and Cholesky factor of covariance entries for the specified approximation. More...
 
normal_fullrank operator* (double scalar, normal_fullrank rhs)
 Return a new approximation resulting from elementwise multiplication of the specified scalar to the mean and Cholesky factor of covariance entries for the specified approximation. More...
 
normal_meanfield operator+ (normal_meanfield lhs, const normal_meanfield &rhs)
 Return a new approximation resulting from adding the mean and log standard deviation of the specified approximations. More...
 
normal_meanfield operator/ (normal_meanfield lhs, const normal_meanfield &rhs)
 Return a new approximation resulting from elementwise division of of the first specified approximation by the second. More...
 
normal_meanfield operator+ (double scalar, normal_meanfield rhs)
 Return a new approximation resulting from elementwise addition of the specified scalar to the mean and log standard deviation entries of the specified approximation. More...
 
normal_meanfield operator* (double scalar, normal_meanfield rhs)
 Return a new approximation resulting from elementwise multiplication of the specified scalar to the mean and log standard deviation vectors of the specified approximation. More...
 

Function Documentation

base_family stan::variational::operator* ( double  scalar,
base_family  rhs 
)
normal_meanfield stan::variational::operator* ( double  scalar,
normal_meanfield  rhs 
)

Return a new approximation resulting from elementwise multiplication of the specified scalar to the mean and log standard deviation vectors of the specified approximation.

Parameters
[in]scalarScalar value
[in]rhsApproximation.
Returns
Multiplication of scalar by the specified approximation.

Definition at line 491 of file normal_meanfield.hpp.

normal_fullrank stan::variational::operator* ( double  scalar,
normal_fullrank  rhs 
)

Return a new approximation resulting from elementwise multiplication of the specified scalar to the mean and Cholesky factor of covariance entries for the specified approximation.

Parameters
[in]scalarScalar value
[in]rhsApproximation.
Returns
Multiplication of scalar by the specified approximation.

Definition at line 531 of file normal_fullrank.hpp.

base_family stan::variational::operator+ ( base_family  lhs,
const base_family rhs 
)
base_family stan::variational::operator+ ( double  scalar,
base_family  rhs 
)
normal_meanfield stan::variational::operator+ ( normal_meanfield  lhs,
const normal_meanfield rhs 
)

Return a new approximation resulting from adding the mean and log standard deviation of the specified approximations.

Parameters
[in]lhsFirst approximation.
[in]rhsSecond approximation.
Returns
Sum of the specified approximations.
Exceptions
std::domain_errorIf the dimensionalities do not match.

Definition at line 450 of file normal_meanfield.hpp.

normal_meanfield stan::variational::operator+ ( double  scalar,
normal_meanfield  rhs 
)

Return a new approximation resulting from elementwise addition of the specified scalar to the mean and log standard deviation entries of the specified approximation.

Parameters
[in]scalarScalar value
[in]rhsApproximation.
Returns
Addition of scalar to specified approximation.

Definition at line 478 of file normal_meanfield.hpp.

normal_fullrank stan::variational::operator+ ( normal_fullrank  lhs,
const normal_fullrank rhs 
)

Return a new approximation resulting from adding the mean and covariance matrix Cholesky factor of the specified approximations.

Parameters
[in]lhsFirst approximation.
[in]rhsSecond approximation.
Returns
Sum of the specified approximations.
Exceptions
std::domain_errorIf the dimensionalities do not match.

Definition at line 492 of file normal_fullrank.hpp.

normal_fullrank stan::variational::operator+ ( double  scalar,
normal_fullrank  rhs 
)

Return a new approximation resulting from elementwise addition of the specified scalar to the mean and Cholesky factor of covariance entries for the specified approximation.

Parameters
[in]scalarScalar value
[in]rhsApproximation.
Returns
Addition of scalar to specified approximation.

Definition at line 518 of file normal_fullrank.hpp.

base_family stan::variational::operator/ ( base_family  lhs,
const base_family rhs 
)
normal_meanfield stan::variational::operator/ ( normal_meanfield  lhs,
const normal_meanfield rhs 
)

Return a new approximation resulting from elementwise division of of the first specified approximation by the second.

Parameters
[in]lhsFirst approximation.
[in]rhsSecond approximation.
Returns
Elementwise division of the specified approximations.
Exceptions
std::domain_errorIf the dimensionalities do not match.

Definition at line 464 of file normal_meanfield.hpp.

normal_fullrank stan::variational::operator/ ( normal_fullrank  lhs,
const normal_fullrank rhs 
)

Return a new approximation resulting from elementwise division of of the first specified approximation by the second.

Parameters
[in]lhsFirst approximation.
[in]rhsSecond approximation.
Returns
Elementwise division of the specified approximations.
Exceptions
std::domain_errorIf the dimensionalities do not match.

Definition at line 505 of file normal_fullrank.hpp.


     [ Stan Home Page ] © 2011–2016, Stan Development Team.