nojhan/paradiseo
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Added normal mutation for reals, and corrected error of segment-> arithmetic

Browse source 
for the crossover operators
This commit is contained in:
evomarc 2000-12-04 16:49:42 +00:00
commit e5ff1bf59c
3 changed files with 63 additions and 20 deletions
Download patch file Download diff file Expand all files Collapse all files

2
eo/tutorial/html/FirstRealGA.html
View file

@ -272,7 +272,7 @@ The actual code is in boldface and the comment in normal face.
<td>
<tt><font color="#993399">
<b> &nbsp;</b>// offspring(i) is a linear combination of parent(i)<br>
<b> &nbsp;eoArithmeticCrossover&lt;Indi> xover;</b><br>
<b> &nbsp;eoSegmentCrossover&lt;Indi> xover;</b><br>
</font></tt>
</td>
</tr>

4
eo/tutorial/html/eoLesson1.html
View file

@ -219,7 +219,7 @@ objects of class</font> <b><font face="Arial,Helvetica"><font size=+1><a href=".
(unary operator).&nbsp; These operators are applied in turn to all selected
parents, according to user-defined probabilities.&nbsp; These probabilities
are defined with all other <a href="#parametres">parameters</a>, and will
be passed to the <b><tt><font color="#FF6666"><font size=+1>eoSGA</font></font></tt></b><a href="#parametres">algorithm</a><font color="#000000">.</font></li>
be passed to the <b><tt><font color="#FF6666"><font size=+1>eoSGA </font></font></tt></b><a href="#parametres">algorithm</a><font color="#000000">.</font></li>
<ul>
<li>
@ -240,7 +240,7 @@ bit, which is specific of the bit-flip mutation.&nbsp; Hence, to run the
same algorithm as Goldberg's SGA, the mutation probability (at individual
level) is 1, and the probability of flipping each bit is <b><tt>P_MUT_PER_BIT.</tt></b>
<li>
<a href="FirstRealGA.html#operators">Real</a> The crossover <b><font face="Arial,Helvetica"><font size=+1><a href="../../doc/html/class_eoarithmeticcrossover.html">eoArithmeticCrossover</a></font></font></b>
<a href="FirstRealGA.html#operators">Real</a> The crossover <b><font face="Arial,Helvetica"><font size=+1><a href="../../doc/html/class_eosegmentcrossover.html">eoSegmentCrossover</a></font></font></b>
is the standard <font color="#CC33CC">arithmetic crossover</font> for real-valued
vectors, that chooses a point randomly on the segment between both parents
(also termed <font color="#CC33CC">BLX-0</font>). <b><font face="Arial,Helvetica"><font size=+1><a href="../../doc/html/class_eouniformmutation.html">eoUniformMutation</a></font></font></b>

77
eo/tutorial/html/eoLesson2.html
View file

@ -4,7 +4,7 @@
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<meta name="GENERATOR" content="Mozilla/4.75 [en] (X11; U; Linux 2.2.17-21mdksmp i686) [Netscape]">
<title>Tutorial: Lesson 2</title>
<!-- Changed by: Marc Schoenauer, 29-Nov-2000 -->
<!-- Changed by: Marc Schoenauer, 29-Nov-2000 -->
</head>
<body text="#000000" link="#0000EE" vlink="#551A8B" alink="#FF0000" background="beige009.jpg">
<a href="eoLesson1.html">Lesson 1</a> -
@ -67,7 +67,7 @@ argument is a <a href="binary_value.html">vector&lt;bool></a> or a <a href="real
and not an unknown type. This will allow to use the same file for any EO
object that is a sub-class of the corresponding STL vector class.</font></li>
<p><b><font color="#FF0000">Note:</font></b> <font color="#000000">Also,
<p><br><b><font color="#FF0000">Note:</font></b> <font color="#000000">Also,
a non-templatized fitness can be </font><b><font color="#FF6600">compiled
separately</font></b><font color="#000000"> (not done here) into an object
file once and for all (<a href="eoProgramming.html#templates">remember</a>
@ -82,8 +82,8 @@ have to declare 3 template arguments: the type of EO object it will be
applied to, the return type and the type of argument the function actually
requires.</font></li>
<p><b><font color="#FF0000">Note:</font></b> <font color="#000000">In the
previous files (<a href="FirstBitGA.html#eval">Bit</a> - <a href="FirstRealGA.html#eval">Real</a>)
<p><br><b><font color="#FF0000">Note:</font></b> <font color="#000000">In
the previous files (<a href="FirstBitGA.html#eval">Bit</a> - <a href="FirstRealGA.html#eval">Real</a>)
, the last 2 types were deduced from the first (2nd argument = fitness
type of EO object, third = first).</font>
<br>&nbsp;
@ -105,7 +105,7 @@ You can also use different initializers and call them in turn through the
call to <a href="../../doc/html/class_eopop.html#a2">pop.append()</a> function
(see <a href="#exercise2">Exercise 2</a>).</font></li>
<p><b><font color="#FF0000">Note</font><font color="#CC33CC">: </font></b><font color="#000000">Don't
<p><br><b><font color="#FF0000">Note</font><font color="#CC33CC">: </font></b><font color="#000000">Don't
forget to </font><b><font color="#CC0000">evaluate the population</font></b><font color="#000000">:
the eoPop has no idea of the eval function, so it has to be done from outside!!!</font>
<br>&nbsp;
@ -113,24 +113,67 @@ the eoPop has no idea of the eval function, so it has to be done from outside!!!
<a NAME="combined_operators"></a><font color="#000000">You can now use
</font><font color="#CC33CC"><b>different
</b>crossover
and mutation
<b>operators</b></font><font color="#000000">in the same algorithm
(<a href="FirstBitEA.html#operators">Bit</a> - <a href="FirstRealEA.html#operators">Real</a>),
choosing among them according to
</font><font color="#CC33CC"><b>relative
rates.</b> </font><font color="#000000">The class </font><font color="#CC33CC"><b>eoPropCombinedxxxOp</b>,
</font><font color="#000000">and</font><font color="#CC33CC"> mutation
<b>operators</b></font><font color="#000000">in
the same algorithm, choosing among them according to
</font><b><font color="#FF6600">relative
rates.</font></b><font color="#CC33CC"> </font><font color="#000000">The
class </font><font color="#CC33CC"><b>eoPropCombinedxxxOp</b>,
</font><font color="#000000">where
xxx is either Mon (for mutation, of class </font><font color="#CC33CC"><b><tt>eoMonOp</tt></b>)</font><font color="#000000">
xxx is either Mon (for mutations, of class </font><font color="#CC33CC"><b><tt>eoMonOp</tt></b>)</font><font color="#000000">
or Quad (for crossovers, of class </font><b><tt><font color="#CC33CC">eoQuadOp</font></tt></b><font color="#000000">),
is derived from the corresponding eoxxxOp class. When applying the eoPropCombinedxxxOp,
one of the eoxxxOp it contains is chosen by a <a href="../../doc/html/class_eorng.html#a12">roulette
wheel,</a> according to their respective rates, and is applied to the arguments.</font></li>
<p><b><font color="#FF0000">Note:</font></b> A third optional argument
in method <b><tt><font color="#660000">add</font></tt></b> is a boolean
(defaulted to false). When true, the actual rates for all operators are
displayed on the screen as percentages: you don't have to input rates that
sum up to 1, all rates are scaled anyway.
<ul>
<li>
<font color="#000000"><a href="FirstBitEA.html#operators">Bit</a>&nbsp;<br>
Three </font><b><font color="#FF6600">crossover operators</font></b><font color="#000000">
are available: the </font><font color="#FF6600">one-point</font><font color="#000000">
crossover is still there (class ), but now you also have the </font><font color="#FF6600">N-point</font><font color="#000000">
crossover </font><font color="#CC33CC">eoBinNxOver</font><font color="#000000">
(the&nbsp; number of points is 2 by default, but as always you can change
that in the constructor), and the </font><font color="#FF6600">Uniform</font><font color="#000000">
crossover </font><font color="#CC33CC">eoBinUxOver</font><font color="#000000">
(where you can eventually twidle the choice from one parent to the other
by providing a probability in the constructore - defaulted to 0.5, which
amounts to symmetrical choice).<br>
As for </font><b><font color="#FF6600">mutation operators</font></b><font color="#000000">,
apart from the </font><font color="#CC33CC">eoBinMutation</font><font color="#000000">
(standard bitstring mutation flipping one bit with a given probability)
you can also use the </font><font color="#CC33CC">eoDetBitFlip</font><font color="#000000">
that always filps the same number of bits (1 by default, but you can change
that in the constructor), randomly chosen in the bitstring. Even though
the average number of bits flipped is the same if the </font><font color="#CC33CC">eoBinMutation
</font><font color="#000000">is used with a rate of 1/N (N is the bitstring
length) </font><font color="#FF6600">the behavior of these mutation can
be very different</font><font color="#000000"> on many problems.</font></li>
<li>
<font color="#000000"><a href="FirstRealEA.html#operators">Real</a>&nbsp;<br>
Two </font><b><font color="#FF6600">crossover operators</font></b><font color="#000000">
are available: the </font><font color="#CC33CC">eoSegmentCrossover</font><font color="#000000">
chooses one point uniformly on the segment joining the parents, while the
</font><font color="#CC33CC">eoArithmeticCrossover</font><font color="#000000">
performs a segment crossover on each coordinate independently, which amount
to choosing the offspring uniformly in the hypercube whose diagonal is
the segment joining the parents.<br>
As for </font><b><font color="#FF6600">mutation operators</font></b><font color="#000000">,
apart from the </font><font color="#CC33CC">eoBinMutation</font><font color="#000000">
(standard bitstring mutation flipping one bit with a given probability)
you can also use the </font><font color="#CC33CC">eoDetBitFlip</font><font color="#000000">
that always filps the same number of bits (1 by default, but you can change
that in the constructor), randomly chosen in the bitstring. And last but
not least, the normal mutation eoNormMutation modifies all coordinates
with a Gaussian noise, with standard deviation passed in the constructor.</font><br>
<BR></li>
</ul>
<b><font color="#FF0000">Note:</font></b> A third optional argument in
method <b><tt><font color="#660000">add</font></tt></b> is a boolean (defaulted
to false). When true, the actual rates for all operators are displayed
on the screen as percentages: you don't have to input rates that sum up
to 1, all rates are scaled anyway.
<p><b><font color="#FF0000">Note:</font></b> The operators have to be encapsulated
into an <b><tt><font color="#CC33CC">eoTransform</font></tt></b> object
(<a href="FirstBitEA.html#transform">Bit</a> - <a href="FirstRealEA.html#transform">Real</a>)