SuanShu numerical library

{{Infobox software

| name = SuanShu

| latest release version = 20120606

| latest release date = 2012-06-06

| programming language = Java

| genre = Math

| license = Apache License 2.0

| website = {{URL|https://github.com/nmltd/SuanShu}}

}}

SuanShu is a Java math library. It is open-source under Apache License 2.0 available in [https://github.com/nmltd/SuanShu GitHub]. SuanShu is a large collection of Java classes for basic numerical analysis, statistics, and optimization.{{Cite web|title=Java Numerics: Main|url=https://math.nist.gov/javanumerics/|access-date=2021-03-23|website=math.nist.gov}} It implements a parallel version of the adaptive strassen's algorithm for fast matrix multiplication.{{Cite web|date=2015-08-07|title=Fastest Java Matrix Multiplication {{!}} NM DEV|url=https://nm.dev/2015/08/07/accelerating-matrix-multiplication/|access-date=2021-08-02|website=NM DEV {{!}} Mathematics at Your Fingertips}} SuanShu has been quoted and used in a number of academic works.{{Cite thesis|title=Automatic stability verification via Lyapunov functions: representations, transformations, and practical issues|url=http://oops.uni-oldenburg.de/4217/|publisher=Universität Oldenburg.|date=2018|degree=phd|first=Eike|last=Möhlmann}}{{Cite journal|last1=Christou|first1=Ioannis T.|last2=Vassilaras|first2=Spyridon|date=2013-10-01|title=A parallel hybrid greedy branch and bound scheme for the maximum distance-2 matching problem|url=https://www.sciencedirect.com/science/article/pii/S0305054813001159|journal=Computers & Operations Research|language=en|volume=40|issue=10|pages=2387–2397|doi=10.1016/j.cor.2013.04.009|issn=0305-0548|url-access=subscription}}{{Cite journal|last1=Łukawska|first1=Barbara|last2=Łukawski|first2=Grzegorz|last3=Sapiecha|first3=Krzysztof|date=2016-10-04|title=An implementation of artificial advisor for dynamic classification of objects|url=https://reshistorica.journals.umcs.pl/ai/article/view/3502|journal=Annales Universitatis Mariae Curie-Skłodowska, Sectio AI – Informatica|language=en|volume=16|issue=1|pages=40|doi=10.17951/ai.2016.16.1.40|issn=2083-3628|doi-access=free}}{{Cite book|last=Ansari|first=Mohd Samar|url=https://books.google.com/books?id=mi66BQAAQBAJ&q=%22suanshu%22+java&pg=PR10|title=Non-Linear Feedback Neural Networks: VLSI Implementations and Applications|date=2013-09-03|publisher=Springer|isbn=978-81-322-1563-9|language=en}}

Features

linear algebra
root finding
curve fitting and interpolation
unconstrained and constrained optimization
statistical analysis
linear regression
probability distributions and random number generation
ordinary and partial differential equation solvers

License terms

SuanShu is released under the terms of the Apache License 2.0

Examples of usage

The following code shows the object-oriented design of the library (in contrast to the traditional procedural design of many other FORTRAN and C numerical libraries) by a simple example of minimization.

LogGamma logGamma = new LogGamma(); // the log-gamma function

BracketSearchMinimizer solver = new BrentMinimizer(1e-8, 10); // precision, max number of iterations

UnivariateMinimizer.Solution soln = solver.solve(logGamma); // optimization

double x_min = soln.search(0, 5); // bracket = [0, 5]

System.out.println(String.format("f(%f) = %f", x_min, logGamma.evaluate(x_min)));