PolynomialChaosExpansion

class pyapprox.surrogates.PolynomialChaosExpansion

Bases: object

A polynomial chaos expansion for independent random variables.

Methods Summary

__call__(samples[, return_grad])	Call self as a function.
basis_matrix(samples[, opts])
canonical_basis_matrix(canonical_samples[, opts])
configure(opts)	Parameters:
covariance()	Compute the covariance between each quantity of interest of the polynomial chaos expansion
get_coefficients()
get_indices()
jacobian(sample)
mean()	Compute the mean of the polynomial chaos expansion
num_terms()
num_vars()
set_coefficients(coefficients)
set_indices(indices)
update_recursion_coefficients(num_coefs_per_var)
value(samples)
variance()	Compute the variance of the polynomial chaos expansion

Methods Documentation

__call__(samples, return_grad=False)

Call self as a function.

basis_matrix(samples, opts={})

canonical_basis_matrix(canonical_samples, opts={})

configure(opts)

Parameters:

var_transpyapprox.variables.transforms.AffineTransform

Variable transformation mapping user samples into the canonical domain of the polynomial basis

optsdictionary

Options defining the configuration of the polynomial chaos expansion basis with the following attributes

poly_optsdictionary

Options to configure each unique univariate polynomial basis with attibutes

var_numsiterable

List of variables dimension which use the ith unique basis

The remaining options are specific to a given basis type. See

• pyapprox.surrogates.orthopoly.quadrature.get_recursion_coefficients_from_variable()

covariance()

Compute the covariance between each quantity of interest of the polynomial chaos expansion

Returns:

covarnp.ndarray (nqoi)

The covariance between each quantitity of interest

get_coefficients()

get_indices()

jacobian(sample)

mean()

Compute the mean of the polynomial chaos expansion

Returns:

meannp.ndarray (nqoi)

The mean of each quantitity of interest

num_terms()

num_vars()

set_coefficients(coefficients)

set_indices(indices)

update_recursion_coefficients(num_coefs_per_var)

value(samples)

variance()

Compute the variance of the polynomial chaos expansion

Returns:

varnp.ndarray (nqoi)

The variance of each quantitity of interest