ALGLIB - C++/C#/Java numerical analysis library (original) (raw)
About ALGLIB
ALGLIB is a cross-platform numerical analysis and data processing library. It supports five programming languages (C++, C#, Java, Python, Delphi) and several operating systems (Windows and POSIX, including Linux). ALGLIB features include:
- Data analysis (classification/regression, statistics)
- Optimization and nonlinear solvers
- Interpolation and linear/nonlinear least-squares fitting
- Linear algebra (direct algorithms, EVD/SVD), direct and iterative linear solvers
- Fast Fourier Transform and many other algorithms
ALGLIB Project offers you several editions of ALGLIB:
ALGLIB Free Edition (download):
+delivered for free under GPL or Personal/Academic license
+offers full set of numerical functionality
+extensive algorithmic optimizations
-single-threaded
-license agreement does not suit most commercial applications
ALGLIB Commercial Edition (more information):
+flexible commercial license without royalties or distribution fees
+extensive algorithmic optimizations
+high performance C++ version (SMP, commercial HPC kernels)
+two C# versions - managed and HPC one (native code, SMP/SIMD)
+commercial support and warranties
Why to choose ALGLIB? Because it is:
- portable. It can be compiled almost anywhere with almost any compiler.
- easy to use. Easy integration, comes with extensive documentation.
- efficient. Deep algorithmic and low-level optimizations inside.
- trusted by leading companies. From nuclear research to aerospace.
Announcements
ALGLIB NEWS (archive):
23.12.2024 ALGLIB 4.04 is released
01.12.2024 Announcing pricing changes for 2025
26.09.2024 ALGLIB 4.03 is released
30.05.2024 ALGLIB 4.02 is released
Resources and links
ALGLIB User Guide online
Dense and sparse linear solvers
- dense direct linear solvers
- sparse iterative/direct linear solvers
Matrix operations and decompositions
- dense BLAS
- LU, Cholesky, QR/LQ and SVD decompositions
- matrix inversion, norms and condition numbers
- generation of random matrices
- Sparse BLAS
- LU, Cholesky decompositions
- linear solvers
- eigensolvers
- dense symmetric/Hermitian EVD
- dense nonsymmetric EVD
- sparse symmetric EVD
- single-dimensional interpolation
- 1D, 2D and 3D splines
- fast scattered N-dimensional interpolation
- least squares curve fitting
Thin plate spline interpolation and fitting
- Thin plate splines
- Interpolation and fitting
- Large-scale algorithms
- IDW
- original, modified and modified-stabilized algorithms
- linear programming
- simplex method
- interior point method
Convex/non-convex QP and QCQP solver
- quadratic programming
- interior point method
Conic solver (SOCP and beyond)
- conic programming
- second-order cones
- power cones
- nonlinear programming
- augmented Lagrangian methods
- SQP
- SLP
Optimization (nonlinear and quadratic)
- unconstrained nonlinear optimization
- constrained nonlinear optimization (box, linear, nonlinear constraints)
- constrained quadratic programming
- nonsmooth optimization
Multi-objective optimization solver
- Multi-objective optimization
- Derivative-free solvers
- Global optimization solver
- Differential evolution
- FFT
- convolution
- correlation
Data analysis: classification, regression, other tasks
- LDA, PCA
- hierarchical and k-means clustering
- decision forests
- nonlinear classifiers
Decision forest (regression and classification)
- Randomized trees and decision forests
- Regression and classification
- Variable importance
- Filtering and smoothing
- Predicting
Statistics: general algorithms
- descriptive statistics
- parametric and non-parametric tests
- nearest neighbor search
Special functions
Numerical integration
Nonlinear and polynomial equations
Differential equations
Other articles
DOCUMENTATION LICENSE:
1. ALGLIB User Guide is licensed for personal use only. See ALGLIB Reference Manual for a free documentation under BSD-like license
2. You may read the Guide and make unlimited copies for personal use.
3. Any other kinds of using the Guide, specifically, sales or any other commercial use, distribution on any material media, through computer networks or any other ways, are prohibited.