[Avida-SVN] r1901 - development/documentation
matt at myxo.css.msu.edu
matt at myxo.css.msu.edu
Tue Aug 7 13:53:43 PDT 2007
Author: matt
Date: 2007-08-07 16:53:42 -0400 (Tue, 07 Aug 2007)
New Revision: 1901
Modified:
development/documentation/analyze.html
Log:
Added support in cAnalyzeGenotype for basic phenotypic plasticity information. Updated documentation for various new features.
Modified: development/documentation/analyze.html
===================================================================
--- development/documentation/analyze.html 2007-08-07 20:04:40 UTC (rev 1900)
+++ development/documentation/analyze.html 2007-08-07 20:53:42 UTC (rev 1901)
@@ -195,7 +195,7 @@
</p>
<dl>
-<dt><strong>RECALCULATE [<span class="cmdargopt">use_resources=0</span>] [<span class="cmdargopt">update=-1</span>] [<span class="cmdargopt">use_random_inputs=0</span>]</strong></dt>
+<dt><strong>RECALCULATE [<span class="cmdargopt">use_resources=0</span>] [<span class="cmdargopt">update=-1</span>] [<span class="cmdargopt">use_random_inputs=0</span>] [<span class="cmdargopt">env_input.1 env_input.2 env_input.3</span>]</strong></dt>
<dd>
Run all of the genotypes in the current batch through a test CPU
and record the measurements taken (fitness, gestation time, etc.).
@@ -208,7 +208,31 @@
is set, then organisms will be provided with new, random input strings for
each trace as they would experience during an actual Avida run. By default,
the same inputs are provided every time to organisms in analysis mode.
+ If additional arguments are specified after use_random_input's value, these
+ integers will be used as environmental inputs for the genotype's test cpu
+ recalculation. Manually specified test cpu inputs must conform to the pseudo-random
+ formatting described in cEnvironment::SetupInputs. Phenotypic plasticity information is not available from the
+ RECALCULATE command; use RECALC with <kbd>num_trials X</kbd>, where X is greater than 1,
+ for phenotypic plasticity statistics.
</dd>
+<dt><strong>RECALC [<span class="cmdargopt">use_resources</span>] [<span class="cmdargopt">use_random_inputs</span>] [<span class="cmdargopt">update N (N=-1)</span>] [<span class="cmdargopt">use_manual_inputs input.1 input.2 input.3</span>] [<span class="cmdargopt">num_trials T (T=1)</span>]</strong></dt>
+<dd>
+ This command will perform the same operations as RECALCULATE but has a few additional features.
+ Instead of having a specified ordering for inputs to this command, argument order does not matter.
+ To use resources, for instance, use the flag <kbd>use_resources</kbd> following RECALC.
+ Arguments with parameters must have the values specified immediately after them. For instance
+ <br>
+ <kbd> RECALC update 10 use_resources use_manual_inputs 256948023 870730840 1441302276</kbd>
+ <br>
+ will set the update to 10, request the use of resources, and test the genotypes using the inputs 256948023, 870730840, and 1441302276. Manually specified environment inputs must conform to the pseudo-random numbers as
+ described in cEnvironment::SetupInputs.
+ Typically, <kbd>use_manual_inputs</kbd> will override <kbd>use_random_inputs</kbd>, however if <kbd>num_trials</kbd> is set to greater than one, manual input specficiation will be overriden
+ and random inputs will be used to gather phenotypic plasticity information that will be available
+ for <a href="#GENOTYPE_STATS">genotype statistics</a> at the bottom of this document. <i>Please note that phenotypic plasticity
+ analysis perfoemd by using RECALC will reset genotype statistics to the values for the most <b>likely</b>
+ phenotype.</i> Implicit phenotypic plasticity analysis (e.g. by not calling RECALC or calling RECALC with <kbd>num_trials 1</kbd>)
+ will not re-evaluate the genotype statistics in this manner and instead rely on the initial values or
+ those values from a single recalculation.
<dt><strong>FILTER [<span class="cmdarg">stat</span>] [<span class="cmdarg">relation</span>] [<span class="cmdarg">test_value</span>] [<span class="cmdargopt">batch=current</span>]</strong></dt>
<dd>
Perform the given test on all genotypes and Remove all those that do not pass. Stat indicates which
@@ -308,7 +332,7 @@
the files will be named by the genotype name, with a <kbd>.gen</kbd> appended to
them. Specifying the filename is useful when printing a single genotype.
</dd>
-<dt><strong>TRACE [<span class="cmdargopt">dir='archive/'</span>] [ <span class="cmdargopt">use_resources=0</span>] [<span class="cmdargopt">update=-1</span>] [ <span class="cmdargopt">use_random_inputs=0</span>]</strong></dt>
+<dt><strong>TRACE [<span class="cmdargopt">dir='archive/'</span>] [ <span class="cmdargopt">use_resources=0</span>] [<span class="cmdargopt">update=-1</span>] [ <span class="cmdargopt">use_random_inputs=0</span>] [<span class="cmdargopt">env_input.1 env_input.2 env_input.3</span>]</strong></dt>
<dd>
Trace all of the genotypes and print a listing of their execution.
This will show step-by-step the status of all of the CPU components
@@ -320,13 +344,16 @@
For more information on resources, see the <a href="#USING_RESOURCES">summary</a> below.
If the use_random_inputs flag is set, then organisms will be provided with new,
random input strings for each trace as they would experience during an actual Avida run.
- By default, the same inputs are provided every time to organisms in analysis mode.
+ By default, the same inputs are provided every time to organisms in analysis mode. You can
+ manually specify environmental inputs by setting use_random_inputs to 0 setting env_input.X
+ values. Manually specified environment inputs must conform to the pseudo-random numbers as
+ described in cEnvironment::SetupInputs.
</dd>
<dt><strong>PRINT_TASKS [<span class="cmdargopt">file='tasks.dat'</span>]</strong></dt>
<dd>
This will print out the tasks doable by each genotype, one per line
in the output file specified. Note that this information must either
- have been loaded in, or a RECALCULATE must have been run to collect
+ have been loaded in, or a RECALCULATE (or RECALC) must have been run to collect
it.
</dd>
<dt><strong>DETAIL [<span class="cmdargopt">file='detail.dat'</span>] [<span class="cmdargopt">format ...</span>]</strong></dt>
@@ -402,7 +429,11 @@
the list of stats that you want to include as columns in the array
(see <a href="#GENOTYPE_STATS">Genotype Statistics</a> for more info). Additionally you can have
special format flags; the possible flags are 'html' to print output in HTML format, and
- 'link_maps' to create html links between consecutive genotypes in a lineage.
+ 'link_maps' to create html links between consecutive genotypes in a lineage. The flag
+ 'use_manual_inputs input.1 input.2 input.3' where input.X are integers
+ allow for designated environmental inputs to be used when evaluating the genotype-phenotype mapping.
+ Manually specified environment inputs must conform to the pseudo-random numbers as
+ described in cEnvironment::SetupInputs.
</dd>
<dt><strong>
MAP_MUTATIONS [<span class="cmdargopt">dir="mutations/"</span>]
@@ -508,7 +539,7 @@
<p>
In analyze, a new data structure was included which contains a time ordered
list of resource concentrations. This list can be used to set up resources
-from different time points. By using the update parameter in the RECALCULATE function,
+from different time points. By using the update parameter in the RECALCULATE (or RECALC) function,
you can use the resource concentrations from a specified time
point. If the LOAD_RESOURCES command is not called, the list defaults to
a single entry which is the the initial concentrations of the resources
@@ -572,7 +603,7 @@
</div>
<p>
-After a RECALCULATE, these additional formats become available:
+After a RECALCULATE (or RECALC), these additional formats become available:
</p>
<div align="center">
@@ -606,7 +637,7 @@
</div>
<p>
-If a FIND_LINEAGE was done before the RECALCULATE, the parent
+If a FIND_LINEAGE was done before the RECALCULATE (or RECALC), the parent
genotype for each regular genotype will be available, enabling the
additional formats:
</p>
@@ -644,7 +675,7 @@
</div>
<p>
-If a RECALCULATE was done before the ALIGN, the following format is available:
+If a RECALCULATE (or RECALC) was done before the ALIGN, the following format is available:
</p>
<div align="center">
@@ -669,7 +700,7 @@
</div>
<p>
-Finally, there are a handful of commands that will automatically perform
+There are a handful of commands that will automatically perform
landscapping. The landscape will only be run once per organism even when
multiple output variables are used. For enhanced performance on
multi-processor/multi-core systems, see the
@@ -693,6 +724,40 @@
</table>
</div>
+<p>
+Phenotypic plasticity information is available through a number
+of different commands. This information will be gathered in one
+of two manners. If <kbd>RECALC num_trials X</kbd> is called, where X is greater
+than 1, phenotypic plasticity information for each genotype in the
+batch will be collected. If RECALC is not called or is called with
+with just one trial (the default for RECALC), then using these commands will request 1000
+trials for each genotype to gather plasticity information. Requesting an analysis
+of phenotypic plasticity in this manner will not re-evaluate other genotype statistics.
+<p>
+<div align="center">
+<table width="95%">
+<tr>
+ <td width="50%"><strong>num_phen</strong> (Number of Phenotypes Identified)</td>
+ <td width="50%"><strong>phen_avg_fitness</strong> (Weighted Average Fitness)</td>
+</tr>
+<tr>
+ <td><strong>num_trials</strong> (Number of Phenotype Tests)</td>
+ <td><strong>phen_entropy</strong> (Phenotypic Entropy [bits])</td>
+</tr>
+<tr>
+ <td><strong>phen_max_fit_freq</strong> (Maximum Fitness Phenotype Frequency)</td>
+ <td><strong>phen_max_fitness</strong> (Maximum Phenotype Fitness)</td>
+</tr>
+<tr>
+ <td><strong>phen_min_fit_freq</strong> (Minimum Fitness Phenotype Frequency)</td>
+ <td><strong>phen_min_fitness</strong> (Minimum Phenotype Fitness)</td>
+</tr>
+<tr>
+ <td><strong>phen_likely_freq</strong> (Most Likely Phenotype Frequency)</td>
+ <td><strong>phen_likely_fitness</strong> (Fitness of the Most Likely Phenotype)</td>
+</tr>
+</table>
+</div>
<h3>Variables</h3>
More information about the Avida-cvs
mailing list