Function libwalrus::hafnian_rpt_quad(std::vector<double>&, std::vector<int>&)

• Defined in File repeated_hafnian.hpp

Function Documentation

double libwalrus::hafnian_rpt_quad(std::vector<double> &mat, std::vector<int> &rpt)

Returns the hafnian of a matrix using the algorithm described in From moments of sum to moments of product, doi:10.1016/j.jmva.2007.01.013.

Note that this algorithm, while generally slower than others, can be significantly more efficient in the cases where the matrix has repeated rows and columns.

This is a wrapper around the templated function libwalrus::hafnian_rpt for Python integration. It accepts and returns double numeric types, and returns sensible values for empty and non-even matrices.

In addition, this wrapper function automatically casts all matrices to type long double, allowing for greater precision than supported by Python and NumPy.

Return

hafnian of the input matrix

Parameters

• mat: a flattened vector of size \(n^2\), representing an \(n\times n\) row-ordered symmetric matrix.

• rpt: a vector of integers, representing the number of times each row/column in mat is repeated. For example, mat = [1] and rpt = [6] represents a \(6\times 6\) matrix of all ones.

libwalrus_cpp_api/function_namespacelibwalrus_1a79aebbde595b7f9b86e74e5ba8e898fa

 

Download Python script

 

Download Notebook

 

View on GitHub