*************************************************************************               gpc++ - The Genetic Programming Kernel                 **                         C++ Class Library                            *************************************************************************The source files================Uncompressing the archive file should have created a number ofdirectories (using "gzip -d <tarfile>.gz ; tar xf <tarfile>" in UNIX,for instance).  Have a look at file FILES for what files should be inthis release."src" contains the GP kernel.  The "lawn" and "symbreg" directorycontain examples of how to use the kernel.  "skeleton" contains thefile "skeleton.cc".  This file contains necessary definitions to starta new problem using the GP kernel.The "include" directory contains the include files of the kernel, andthe "lib" directory will hold the GP library after compilation.Technical documentation in postscript format can be found in the "doc"directory.  There is also a short introduction into geneticprogramming (by Adam Fraser).With most genetic programming systems the documentation is the lastthought (apart from John Koza's 'Genetic Programming' book, of course,which many would consider overkill on a documentation front).  Herethen is the compile and go painless method of getting gpc++ up andevolving something.Compiling under UNIX====================Have a look at the Makefiles, especially Makefile.ini.  It containssome important definitions, for example the name of the C++-Compileryou want to use, and general compiler and linker flags.Call "make" in the root GP directory.  The main makefile calls makefor each subdirectory, that is first for the basic kernel (directorysrc), then for the examples.  If any errors occur, modify themakefiles.If you receive any error on compilation of the source code check thatthe correct directories are being searched for the appropriatestandard header files in the compiler."make" can be called with the following arguments:- clean: Removes the object files as well as the files generated bythe executables, but not the libraries and executables.- superclean: Same as clean, but also removes all files that can beregenerated, e.g. Libraries and executables.- strip: Strips the debug information from the executablesIn the root GP directory, the following argument is allowed as well:- install: Use the program "install" to install the libraries andinclude files to the given directory (see the file "Makefile.ini").Dynamic Libraries-----------------If you want to use dynamic libraries, change the file names for thelibraries from .a to .so (in file Makefile.ini). Change the linkcommand in file src/Makefile so that "ld" will be used instead of"ar" to create the library.Installation------------The installation commands, when executing "make install", are nottested.  It's easy enough: Just copy the files in the lib and includedirectory into /usr/local/bin and /usr/local/include or whateverdirectory you prefer.Examples--------Once you compiled the code of the example, an executable named "sym"in directory symbreg should have been created, as well as "lawn" indirectory lawn and "ant" in directory ant.  When executing theprograms, they read configuration files named symbreg.ini, lawn.iniand ant.ini, respectively, before starting to evolve genetic programs.For more information on the examples either read the book or thesource code.The Symbolic Regression Problem tries to evolve a mathematicalfunction which produces a graph similar (nay identical) to X**4 + X**3+ X**2 + X described in Koza's book.The Lawn-mower Problem tries to find a genetic program that mows thelawn.The Artificial Ant Problem tries to follow the Santa Fe trail definedby Koza.Useless DOS memory allocation=============================gpc++ is memory hungry and is designed nowadays for machines which canaccess their memory.  The DOS operating system is a pain for this andwe did not want to reduce the portability of our code.  If you try anduse over 640k of memory (that includes the program and whatever othergarbage you have) the kernel might crash, if you do not set up a newhandler.  This is certainly done in all examples.  To get over thisyou can;a) Use UNIX (Linux for example).b) Use Windows (but this is nearly as bad as using DOS).c) Give up and use small populations.If you are not keen on any of these options you could buy the WatcomC/C++ compiler which produces the 32-bit code to counter DOS'sinadequacies.  If you have internet access you may wish to look atdjgpp, the DOS port of the GNU compiler.  It isn't the most documentedof systems and also doesn't have a nice front end but it does workwhich is always nice.  Just thought you would like to know.Thomas Weinbrenner (with parts by Adam Fraser)LICENCE=======This program is free software; you can redistribute it and/or modifyit under the terms of the GNU General Public License as published bythe Free Software Foundation; either version 1, or (at your option)any later version.This program is distributed in the hope that it will be useful, butWITHOUT ANY WARRANTY; without even the implied warranty ofMERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNUGeneral Public License for more details.You should have received a copy of the GNU General Public Licensealong with this program; if not, write to the Free SoftwareFoundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.Copyright 1993, 1994 Adam P. Fraser and 1996, 1997 Thomas WeinbrennerFor comments, improvements, additions (or even money) contact:Thomas WeinbrennerGrauensteinstr. 2635789 LaimbachGermanyE-mail: thomasw@emk.e-technik.th-darmstadt.deWWW:    http://www.emk.e-technik.th-darmstadt/~thomasw  or (Address may be out of date)Adam Fraser, Postgraduate Section, Dept of Elec & Elec Eng,Maxwell Building, University Of Salford, Salford, M5 4WT, United Kingdom.E-mail: a.fraser@eee.salford.ac.ukTel:    (UK) 061 745 5000 x3633Fax:    (UK) 061 745 5999