readme

贝叶斯优化算法是一种新的演化算法

--------------------------------------------------
          A Simple Implementation of 
 the Bayesian Optimization Algorithm (BOA) in C++ 
--------------------------------------------------

author:   Martin Pelikan
version:  1.0
released: March 1999

1. INTRODUCTION
----------------

The instructions for compiling and using the simple implementation of
the BOA, version 1.0, can be found in an IlliGAL tech-report
99011. The report can be downloaded from the IlliGAL www-site: 
http://www-illigal.ge.uiuc.edu/

The short version of the instructions for compiling the code, using
the resulting executable and some more comments follow. However, we
encourage you to read the report in order to take advantage of the
features of the implementation and understand what is actually going
on when you see all the outputs. 


2. COMPILATION
---------------

In Makefile, change the following two lines:

1) Line 36 - In the statement CC = CC, the CC on the right-hand side should be
changed to the name of a preferred C++ compiler on your machine. With
a gcc, for instance, the line should be changed to CC = gcc. 

2) Line 42 - In the statement OPTIMIZE = -O3, the required optimization level 
should be set (for GNU gcc this is -O4, for SGI CC it is -O3). For no code 
optimization, use only OPTIMIZE = . For instance, for a maximal optimization 
with gcc, i.e. -O4, use OPTIMIZE = -O4. All modules are compiled at once since 
some compilers (as SGI CC) use intermodule optimization that does not allow
them to compile each source file separately and link them together afterwards. 

Run the following command line:

make all

After compiling the source codes, your directory should contain an
executable file boa.


3. COMMAND LINE PARAMETERS
---------------------------

There are three parameters that can be passed to the boa:

<input file>      -> input file name
-h                -> help on command line parameters
-paramDescription -> print out the description of input file parameters


4. EXAMPLE INPUT FILES
-----------------------

Example input files are located in a sub-directory examples. Files with
the names starting with input are input files and files starting with
output are output files produced with the parameters specified in the
corresponding input files.


4. COMMENTS
------------

This code is distributed for academic purposes with absolutely no
warranty of any kind, either expressed or implied, to the extent
permitted by applicable state law. We are not responsible for any
damage resulting from its proper or improper use.

If you have any comments or identify any bugs, please contact the
author at the following address (email is the preferred form of
communication):

Martin Pelikan
Illinois Genetic Algorithms Laboratory
Department of General Engeneering
University of Illinois at Urbana-Champaign
117 Transportation Building
104 South Mathews Avenue
Urbana, IL 61801
email: [e-mail]


Happy BOAing!