OOFramework
From OpenOpt
 Free (license: BSD) crossplatform (Linux, Windows, Mac etc) software, written in modern highstream Python language, alternative to commercial frameworks with obsolete and/or banausic programming constructs like AMPL, GAMS, LINGO, AIMMS, TOMOPT's TOMLAB / TOMNET (see for example TOMLAB users and prices)
 Primary goals:
 providing connections to lots of solvers, e.g. IPOPT, cplex, algencan, glpk, lpsolve, CVXOPT, PSwarm, nlopt, BOBYQA, lbfgsb (some are C or Fortranwritten) with easy and unified syntax
 trying to enhance algorithms for numerical optimization invented by Ukrainian academicians Boris Pshenichniy and Naum Z. Shor
 Organized as an ordinary Python module, thus you can easily stack it with NumPy arrays, SciPy sparse matrices and other Python functions and code
 Graphical output, automatic derivatives check for usersupplied objFunc and constraints, userdefined callback functions and much more
 Capable of handling FuncDesigner code with Automatic differentiation (possibly largescale sparse) for gradientbased solvers and interval analysis for solver with guaranteed precision interalg (most of OpenOpt users after initial coding in pure Python move to FuncDesigner, that also allows connecting C/Fortran/General Python code to its functions)

Some native solvers (Pythonwritten, wellknown Python library NumPy is required), constrained and unconstrained, e.g.
 nonlinear/nonsmooth solvers ralg, gsubg, amsg2p
 global nonlinear solver de (differential evolution)
 interalg  solver with specifiable accuracy for global nonlinear and multiobjective problems

 Now interalg can find all solutions of nonlinear equation(s) and perform numerical integration with specifiable accuracy
Made by Dmitrey 
See also:
 DerApproximator  finitedifferences derivatives approximation
 SpaceFuncs  2D, 3D, Ndimensional geometric modeling with possibilities of parametrized calculations, numerical optimization and solving systems of geometrical equations
20072012