[Avida-cvs] [Avida2-svn] r369 - in branches/brysonda: Avida2.xcodeproj source source/analyze source/main
brysonda@myxo.css.msu.edu
brysonda at myxo.css.msu.edu
Thu Nov 3 20:19:25 PST 2005
Author: brysonda
Date: 2005-11-03 22:50:46 -0500 (Thu, 03 Nov 2005)
New Revision: 369
Added:
branches/brysonda/source/analyze/
branches/brysonda/source/analyze/cAnalyze.cc
branches/brysonda/source/analyze/cAnalyze.h
branches/brysonda/source/analyze/cAnalyzeCommand.h
branches/brysonda/source/analyze/cAnalyzeCommandDef.h
branches/brysonda/source/analyze/cAnalyzeCommandDefBase.h
branches/brysonda/source/analyze/cAnalyzeFlowCommand.h
branches/brysonda/source/analyze/cAnalyzeFlowCommandDef.h
branches/brysonda/source/analyze/cAnalyzeFunction.h
branches/brysonda/source/analyze/cAnalyzeGenotype.cc
branches/brysonda/source/analyze/cAnalyzeGenotype.h
branches/brysonda/source/analyze/cAnalyzeUtil.cc
branches/brysonda/source/analyze/cAnalyzeUtil.h
Removed:
branches/brysonda/source/main/cAnalyze.cc
branches/brysonda/source/main/cAnalyze.h
branches/brysonda/source/main/cAnalyzeCommand.h
branches/brysonda/source/main/cAnalyzeCommandDef.h
branches/brysonda/source/main/cAnalyzeCommandDefBase.h
branches/brysonda/source/main/cAnalyzeFlowCommand.h
branches/brysonda/source/main/cAnalyzeFlowCommandDef.h
branches/brysonda/source/main/cAnalyzeFunction.h
branches/brysonda/source/main/cAnalyzeGenotype.cc
branches/brysonda/source/main/cAnalyzeGenotype.h
branches/brysonda/source/main/cAnalyzeUtil.cc
branches/brysonda/source/main/cAnalyzeUtil.h
Modified:
branches/brysonda/Avida2.xcodeproj/project.pbxproj
Log:
Move cAnalyze* into source/analyze/
Modified: branches/brysonda/Avida2.xcodeproj/project.pbxproj
===================================================================
--- branches/brysonda/Avida2.xcodeproj/project.pbxproj 2005-11-04 00:39:30 UTC (rev 368)
+++ branches/brysonda/Avida2.xcodeproj/project.pbxproj 2005-11-04 03:50:46 UTC (rev 369)
@@ -42,18 +42,18 @@
702D4F0608DA5341007BA469 /* cEnvironment.cc in Sources */ = {isa = PBXBuildFile; fileRef = 702D4EFC08DA5341007BA469 /* cEnvironment.cc */; };
702D4F0708DA5341007BA469 /* cPopulationInterface.cc in Sources */ = {isa = PBXBuildFile; fileRef = 702D4EFD08DA5341007BA469 /* cPopulationInterface.cc */; };
702D4F4508DA61FE007BA469 /* cBirthChamber.cc in Sources */ = {isa = PBXBuildFile; fileRef = 702D4F3F08DA61FE007BA469 /* cBirthChamber.cc */; };
- 702D4F4808DA61FE007BA469 /* cAnalyze.cc in Sources */ = {isa = PBXBuildFile; fileRef = 702D4F3C08DA61FE007BA469 /* cAnalyze.cc */; };
- 702D4F4908DA61FE007BA469 /* cAnalyzeGenotype.cc in Sources */ = {isa = PBXBuildFile; fileRef = 702D4F3D08DA61FE007BA469 /* cAnalyzeGenotype.cc */; };
- 702D4F4A08DA61FE007BA469 /* cAnalyzeUtil.cc in Sources */ = {isa = PBXBuildFile; fileRef = 702D4F3E08DA61FE007BA469 /* cAnalyzeUtil.cc */; };
702D4F4B08DA61FE007BA469 /* cBirthChamber.cc in Sources */ = {isa = PBXBuildFile; fileRef = 702D4F3F08DA61FE007BA469 /* cBirthChamber.cc */; };
7040CF1C0906A52E00AA820F /* cEventManager.cc in Sources */ = {isa = PBXBuildFile; fileRef = 7040CF1A0906A52E00AA820F /* cEventManager.cc */; };
7040CF1E0906A52E00AA820F /* cEventManager.cc in Sources */ = {isa = PBXBuildFile; fileRef = 7040CF1A0906A52E00AA820F /* cEventManager.cc */; };
7040CF1F0906A52E00AA820F /* cEventManager.h in CopyFiles */ = {isa = PBXBuildFile; fileRef = 7040CF1B0906A52E00AA820F /* cEventManager.h */; };
- 7040D22509071EE000AA820F /* cAnalyzeUtil.cc in Sources */ = {isa = PBXBuildFile; fileRef = 702D4F3E08DA61FE007BA469 /* cAnalyzeUtil.cc */; };
- 7040D36C09095E4E00AA820F /* cAnalyze.cc in Sources */ = {isa = PBXBuildFile; fileRef = 702D4F3C08DA61FE007BA469 /* cAnalyze.cc */; };
- 7040D36D09095E5200AA820F /* cAnalyzeGenotype.cc in Sources */ = {isa = PBXBuildFile; fileRef = 702D4F3D08DA61FE007BA469 /* cAnalyzeGenotype.cc */; };
7040D36E09095E5900AA820F /* cAvidaDriver_Analyze.cc in Sources */ = {isa = PBXBuildFile; fileRef = 702D4EF908DA5341007BA469 /* cAvidaDriver_Analyze.cc */; };
7040D3A6090964D100AA820F /* cMxCodeArray.cc in Sources */ = {isa = PBXBuildFile; fileRef = 70B0865808F4974300FC65FE /* cMxCodeArray.cc */; };
+ 70422A28091B141000A5E67F /* cAnalyze.cc in Sources */ = {isa = PBXBuildFile; fileRef = 70422A1C091B141000A5E67F /* cAnalyze.cc */; };
+ 70422A30091B141000A5E67F /* cAnalyzeGenotype.cc in Sources */ = {isa = PBXBuildFile; fileRef = 70422A24091B141000A5E67F /* cAnalyzeGenotype.cc */; };
+ 70422A32091B141000A5E67F /* cAnalyzeUtil.cc in Sources */ = {isa = PBXBuildFile; fileRef = 70422A26091B141000A5E67F /* cAnalyzeUtil.cc */; };
+ 70422A34091B141000A5E67F /* cAnalyze.cc in Sources */ = {isa = PBXBuildFile; fileRef = 70422A1C091B141000A5E67F /* cAnalyze.cc */; };
+ 70422A3C091B141000A5E67F /* cAnalyzeGenotype.cc in Sources */ = {isa = PBXBuildFile; fileRef = 70422A24091B141000A5E67F /* cAnalyzeGenotype.cc */; };
+ 70422A3E091B141000A5E67F /* cAnalyzeUtil.cc in Sources */ = {isa = PBXBuildFile; fileRef = 70422A26091B141000A5E67F /* cAnalyzeUtil.cc */; };
704866D3090B51310048600A /* cAvidaDriver_TextPopViewer.cc in Sources */ = {isa = PBXBuildFile; fileRef = 704866B3090B51310048600A /* cAvidaDriver_TextPopViewer.cc */; };
704866D5090B51310048600A /* cBarScreen.cc in Sources */ = {isa = PBXBuildFile; fileRef = 704866B5090B51310048600A /* cBarScreen.cc */; };
704866D8090B51310048600A /* cEnvironmentScreen.cc in Sources */ = {isa = PBXBuildFile; fileRef = 704866B8090B51310048600A /* cEnvironmentScreen.cc */; };
@@ -390,24 +390,24 @@
702D4EFB08DA5341007BA469 /* cAvidaDriver_Population.cc */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.cpp.cpp; path = cAvidaDriver_Population.cc; sourceTree = "<group>"; };
702D4EFC08DA5341007BA469 /* cEnvironment.cc */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.cpp.cpp; path = cEnvironment.cc; sourceTree = "<group>"; };
702D4EFD08DA5341007BA469 /* cPopulationInterface.cc */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.cpp.cpp; path = cPopulationInterface.cc; sourceTree = "<group>"; };
- 702D4F2F08DA61E2007BA469 /* cAnalyze.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyze.h; sourceTree = "<group>"; };
- 702D4F3008DA61E2007BA469 /* cAnalyzeCommand.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeCommand.h; sourceTree = "<group>"; };
- 702D4F3108DA61E2007BA469 /* cAnalyzeCommandDef.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeCommandDef.h; sourceTree = "<group>"; };
- 702D4F3208DA61E2007BA469 /* cAnalyzeCommandDefBase.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeCommandDefBase.h; sourceTree = "<group>"; };
- 702D4F3308DA61E2007BA469 /* cAnalyzeFlowCommand.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeFlowCommand.h; sourceTree = "<group>"; };
- 702D4F3408DA61E2007BA469 /* cAnalyzeFlowCommandDef.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeFlowCommandDef.h; sourceTree = "<group>"; };
- 702D4F3508DA61E2007BA469 /* cAnalyzeFunction.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeFunction.h; sourceTree = "<group>"; };
- 702D4F3608DA61E2007BA469 /* cAnalyzeGenotype.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeGenotype.h; sourceTree = "<group>"; };
- 702D4F3708DA61E2007BA469 /* cAnalyzeUtil.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeUtil.h; sourceTree = "<group>"; };
702D4F3808DA61E2007BA469 /* cAvidaTriggers.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAvidaTriggers.h; sourceTree = "<group>"; };
702D4F3908DA61E2007BA469 /* cBirthChamber.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cBirthChamber.h; sourceTree = "<group>"; };
- 702D4F3C08DA61FE007BA469 /* cAnalyze.cc */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.cpp.cpp; path = cAnalyze.cc; sourceTree = "<group>"; };
- 702D4F3D08DA61FE007BA469 /* cAnalyzeGenotype.cc */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.cpp.cpp; path = cAnalyzeGenotype.cc; sourceTree = "<group>"; };
- 702D4F3E08DA61FE007BA469 /* cAnalyzeUtil.cc */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.cpp.cpp; path = cAnalyzeUtil.cc; sourceTree = "<group>"; };
702D4F3F08DA61FE007BA469 /* cBirthChamber.cc */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.cpp.cpp; path = cBirthChamber.cc; sourceTree = "<group>"; };
703F684207B437B800C1CA76 /* status.xml */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = text.xml; path = status.xml; sourceTree = "<group>"; };
7040CF1A0906A52E00AA820F /* cEventManager.cc */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.cpp.cpp; path = cEventManager.cc; sourceTree = "<group>"; };
7040CF1B0906A52E00AA820F /* cEventManager.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cEventManager.h; sourceTree = "<group>"; };
+ 70422A1C091B141000A5E67F /* cAnalyze.cc */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.cpp.cpp; path = cAnalyze.cc; sourceTree = "<group>"; };
+ 70422A1D091B141000A5E67F /* cAnalyze.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyze.h; sourceTree = "<group>"; };
+ 70422A1E091B141000A5E67F /* cAnalyzeCommand.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeCommand.h; sourceTree = "<group>"; };
+ 70422A1F091B141000A5E67F /* cAnalyzeCommandDef.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeCommandDef.h; sourceTree = "<group>"; };
+ 70422A20091B141000A5E67F /* cAnalyzeCommandDefBase.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeCommandDefBase.h; sourceTree = "<group>"; };
+ 70422A21091B141000A5E67F /* cAnalyzeFlowCommand.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeFlowCommand.h; sourceTree = "<group>"; };
+ 70422A22091B141000A5E67F /* cAnalyzeFlowCommandDef.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeFlowCommandDef.h; sourceTree = "<group>"; };
+ 70422A23091B141000A5E67F /* cAnalyzeFunction.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeFunction.h; sourceTree = "<group>"; };
+ 70422A24091B141000A5E67F /* cAnalyzeGenotype.cc */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.cpp.cpp; path = cAnalyzeGenotype.cc; sourceTree = "<group>"; };
+ 70422A25091B141000A5E67F /* cAnalyzeGenotype.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeGenotype.h; sourceTree = "<group>"; };
+ 70422A26091B141000A5E67F /* cAnalyzeUtil.cc */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.cpp.cpp; path = cAnalyzeUtil.cc; sourceTree = "<group>"; };
+ 70422A27091B141000A5E67F /* cAnalyzeUtil.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAnalyzeUtil.h; sourceTree = "<group>"; };
704866B3090B51310048600A /* cAvidaDriver_TextPopViewer.cc */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.cpp.cpp; path = cAvidaDriver_TextPopViewer.cc; sourceTree = "<group>"; };
704866B4090B51310048600A /* cAvidaDriver_TextPopViewer.h */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.c.h; path = cAvidaDriver_TextPopViewer.h; sourceTree = "<group>"; };
704866B5090B51310048600A /* cBarScreen.cc */ = {isa = PBXFileReference; fileEncoding = 30; lastKnownFileType = sourcecode.cpp.cpp; path = cBarScreen.cc; sourceTree = "<group>"; };
@@ -1035,6 +1035,25 @@
/* End PBXFrameworksBuildPhase section */
/* Begin PBXGroup section */
+ 70422A1B091B141000A5E67F /* analyze */ = {
+ isa = PBXGroup;
+ children = (
+ 70422A1C091B141000A5E67F /* cAnalyze.cc */,
+ 70422A1D091B141000A5E67F /* cAnalyze.h */,
+ 70422A1E091B141000A5E67F /* cAnalyzeCommand.h */,
+ 70422A1F091B141000A5E67F /* cAnalyzeCommandDef.h */,
+ 70422A20091B141000A5E67F /* cAnalyzeCommandDefBase.h */,
+ 70422A21091B141000A5E67F /* cAnalyzeFlowCommand.h */,
+ 70422A22091B141000A5E67F /* cAnalyzeFlowCommandDef.h */,
+ 70422A23091B141000A5E67F /* cAnalyzeFunction.h */,
+ 70422A24091B141000A5E67F /* cAnalyzeGenotype.cc */,
+ 70422A25091B141000A5E67F /* cAnalyzeGenotype.h */,
+ 70422A26091B141000A5E67F /* cAnalyzeUtil.cc */,
+ 70422A27091B141000A5E67F /* cAnalyzeUtil.h */,
+ );
+ path = analyze;
+ sourceTree = "<group>";
+ };
704866B2090B51310048600A /* viewer */ = {
isa = PBXGroup;
children = (
@@ -1088,6 +1107,7 @@
DCC30C670762539A008F7A48 /* Source */ = {
isa = PBXGroup;
children = (
+ 70422A1B091B141000A5E67F /* analyze */,
DCC30F7C0762539D008F7A48 /* cpu */,
DCC30FD00762539D008F7A48 /* event */,
DCC310040762539D008F7A48 /* main */,
@@ -1184,18 +1204,6 @@
children = (
DCC3109C0762539E008F7A48 /* avida.cc */,
70B086BE08F5D86100FC65FE /* avida.h */,
- 702D4F3C08DA61FE007BA469 /* cAnalyze.cc */,
- 702D4F2F08DA61E2007BA469 /* cAnalyze.h */,
- 702D4F3008DA61E2007BA469 /* cAnalyzeCommand.h */,
- 702D4F3108DA61E2007BA469 /* cAnalyzeCommandDef.h */,
- 702D4F3208DA61E2007BA469 /* cAnalyzeCommandDefBase.h */,
- 702D4F3308DA61E2007BA469 /* cAnalyzeFlowCommand.h */,
- 702D4F3408DA61E2007BA469 /* cAnalyzeFlowCommandDef.h */,
- 702D4F3508DA61E2007BA469 /* cAnalyzeFunction.h */,
- 702D4F3D08DA61FE007BA469 /* cAnalyzeGenotype.cc */,
- 702D4F3608DA61E2007BA469 /* cAnalyzeGenotype.h */,
- 702D4F3E08DA61FE007BA469 /* cAnalyzeUtil.cc */,
- 702D4F3708DA61E2007BA469 /* cAnalyzeUtil.h */,
7013846009028B3E0087ED2E /* cAvidaConfig.cc */,
7013845F09028B3E0087ED2E /* cAvidaConfig.h */,
702D4EF908DA5341007BA469 /* cAvidaDriver_Analyze.cc */,
@@ -2111,9 +2119,6 @@
702D4F0508DA5341007BA469 /* cAvidaDriver_Population.cc in Sources */,
702D4F0608DA5341007BA469 /* cEnvironment.cc in Sources */,
702D4F0708DA5341007BA469 /* cPopulationInterface.cc in Sources */,
- 702D4F4808DA61FE007BA469 /* cAnalyze.cc in Sources */,
- 702D4F4908DA61FE007BA469 /* cAnalyzeGenotype.cc in Sources */,
- 702D4F4A08DA61FE007BA469 /* cAnalyzeUtil.cc in Sources */,
702D4F4B08DA61FE007BA469 /* cBirthChamber.cc in Sources */,
70CA6ED208DB7F8200068AC2 /* cFitnessMatrix.cc in Sources */,
70CA6ED308DB7F8200068AC2 /* cGenebank.cc in Sources */,
@@ -2209,6 +2214,9 @@
704866ED090B51310048600A /* cViewInfo.cc in Sources */,
704866EF090B51310048600A /* cZoomScreen.cc in Sources */,
704866F2090B51310048600A /* viewer.cc in Sources */,
+ 70422A28091B141000A5E67F /* cAnalyze.cc in Sources */,
+ 70422A30091B141000A5E67F /* cAnalyzeGenotype.cc in Sources */,
+ 70422A32091B141000A5E67F /* cAnalyzeUtil.cc in Sources */,
);
runOnlyForDeploymentPostprocessing = 0;
};
@@ -2322,11 +2330,11 @@
70C5BC6509059A970028A785 /* cWorld.cc in Sources */,
70C5BD6B0905CE5F0028A785 /* cHardwareManager.cc in Sources */,
7040CF1C0906A52E00AA820F /* cEventManager.cc in Sources */,
- 7040D22509071EE000AA820F /* cAnalyzeUtil.cc in Sources */,
- 7040D36C09095E4E00AA820F /* cAnalyze.cc in Sources */,
- 7040D36D09095E5200AA820F /* cAnalyzeGenotype.cc in Sources */,
7040D36E09095E5900AA820F /* cAvidaDriver_Analyze.cc in Sources */,
7040D3A6090964D100AA820F /* cMxCodeArray.cc in Sources */,
+ 70422A34091B141000A5E67F /* cAnalyze.cc in Sources */,
+ 70422A3C091B141000A5E67F /* cAnalyzeGenotype.cc in Sources */,
+ 70422A3E091B141000A5E67F /* cAnalyzeUtil.cc in Sources */,
);
runOnlyForDeploymentPostprocessing = 0;
};
Copied: branches/brysonda/source/analyze/cAnalyze.cc (from rev 368, branches/brysonda/source/main/cAnalyze.cc)
Copied: branches/brysonda/source/analyze/cAnalyze.h (from rev 368, branches/brysonda/source/main/cAnalyze.h)
Copied: branches/brysonda/source/analyze/cAnalyzeCommand.h (from rev 367, branches/brysonda/source/main/cAnalyzeCommand.h)
Copied: branches/brysonda/source/analyze/cAnalyzeCommandDef.h (from rev 367, branches/brysonda/source/main/cAnalyzeCommandDef.h)
Copied: branches/brysonda/source/analyze/cAnalyzeCommandDefBase.h (from rev 367, branches/brysonda/source/main/cAnalyzeCommandDefBase.h)
Copied: branches/brysonda/source/analyze/cAnalyzeFlowCommand.h (from rev 367, branches/brysonda/source/main/cAnalyzeFlowCommand.h)
Copied: branches/brysonda/source/analyze/cAnalyzeFlowCommandDef.h (from rev 367, branches/brysonda/source/main/cAnalyzeFlowCommandDef.h)
Copied: branches/brysonda/source/analyze/cAnalyzeFunction.h (from rev 367, branches/brysonda/source/main/cAnalyzeFunction.h)
Copied: branches/brysonda/source/analyze/cAnalyzeGenotype.cc (from rev 368, branches/brysonda/source/main/cAnalyzeGenotype.cc)
Copied: branches/brysonda/source/analyze/cAnalyzeGenotype.h (from rev 368, branches/brysonda/source/main/cAnalyzeGenotype.h)
Copied: branches/brysonda/source/analyze/cAnalyzeUtil.cc (from rev 367, branches/brysonda/source/main/cAnalyzeUtil.cc)
Copied: branches/brysonda/source/analyze/cAnalyzeUtil.h (from rev 367, branches/brysonda/source/main/cAnalyzeUtil.h)
Deleted: branches/brysonda/source/main/cAnalyze.cc
===================================================================
--- branches/brysonda/source/main/cAnalyze.cc 2005-11-04 00:39:30 UTC (rev 368)
+++ branches/brysonda/source/main/cAnalyze.cc 2005-11-04 03:50:46 UTC (rev 369)
@@ -1,7380 +0,0 @@
-//////////////////////////////////////////////////////////////////////////////
-// Copyright (C) 1993 - 2003 California Institute of Technology //
-// //
-// Read the COPYING and README files, or contact 'avida at alife.org', //
-// before continuing. SOME RESTRICTIONS MAY APPLY TO USE OF THIS FILE. //
-//////////////////////////////////////////////////////////////////////////////
-
-#include <fstream>
-#include <sstream>
-#include <string>
-#include <queue>
-#include <stack>
-
-#include "cAnalyze.h"
-
-#include "cAnalyzeCommand.h"
-#include "cAnalyzeCommandDef.h"
-#include "cAnalyzeCommandDefBase.h"
-#include "cAnalyzeFlowCommand.h"
-#include "cAnalyzeFlowCommandDef.h"
-#include "cAnalyzeFunction.h"
-#include "cAnalyzeGenotype.h"
-#include "cDataFile.h"
-#include "cEnvironment.h"
-#include "cFitnessMatrix.h"
-#include "cGenomeUtil.h"
-#include "cHardwareBase.h"
-#include "cHardwareManager.h"
-#include "cHardwareStatusPrinter.h"
-#include "cHelpManager.h"
-#include "cInitFile.h"
-#include "cInstSet.h"
-#include "cInstUtil.h"
-#include "cLandscape.h"
-#include "cPhenotype.h"
-#include "cSpecies.h"
-#include "tArgDataEntry.h"
-#include "cTaskEntry.h"
-#include "tDataEntry.h"
-#include "tDataEntryCommand.h"
-#include "tMatrix.h"
-#include "cTestCPU.h"
-#include "cCPUTestInfo.h"
-#include "cTestUtil.h"
-#include "cResource.h"
-#include "tHashTable.h"
-#include "cWorld.h"
-#ifdef WIN32
-# include "win32_mkdir_hack.hh"
-#endif
-
-extern "C" {
-#include <errno.h>
-#include <sys/stat.h>
-}
-
-using namespace std;
-
-//////////////
-// cAnalyze
-//////////////
-
-cAnalyze::cAnalyze(cWorld* world)
-: cur_batch(0)
-, m_world(world)
-, inst_set(world->GetHardwareManager().GetInstSet())
-, verbose(false)
-, interactive_depth(0)
-{
-
- random.ResetSeed(m_world->GetConfig().RANDOM_SEED.Get());
-
- for (int i = 0; i < MAX_BATCHES; i++) {
- batch[i].Name().Set("Batch%d", i);
- }
-
- // Initialize the time oriented resource list to be just the initial
- // concentrations of the resources in the environment. This will only
- // be changed if LOAD_RESOURCES analyze command is called. If there is
- // no resources in the environment or there is no environment, the list
- // is empty and will cause an assert to be thrown in FindResource function.
- const cResourceLib &resource_lib = m_world->GetEnvironment().GetResourceLib();
- if(resource_lib.GetSize() > 0) {
- vector<double> r;
- for(int i=0; i<resource_lib.GetSize(); i++) {
- cResource *resource = resource_lib.GetResource(i);
- assert(resource);
- r.push_back(resource->GetInitial());
- }
- resources.push_back(make_pair(0, r));
- }
- FillResources(0);
-
- cInitFile analyze_file(m_world->GetConfig().ANALYZE_FILE.Get());
- analyze_file.Load();
- analyze_file.Compress();
- analyze_file.Close();
-
- LoadCommandList(analyze_file, command_list);
- ProcessCommands(command_list);
-
- return;
-}
-
-cAnalyze::~cAnalyze()
-{
- data_file_manager.FlushAll();
- while (genotype_data_list.GetSize()) delete genotype_data_list.Pop();
- while (command_list.GetSize()) delete command_list.Pop();
- while (function_list.GetSize()) delete function_list.Pop();
-}
-
-
-//////////////// Loading methods...
-
-void cAnalyze::LoadOrganism(cString cur_string)
-{
- // LOAD_ORGANISM command...
-
- cString filename = cur_string.PopWord();
-
- cout << "Loading: " << filename << endl;
-
- // Setup the genome...
- cGenome genome( cInstUtil::LoadGenome(filename, inst_set) );
-
- // Construct the new genotype..
- cAnalyzeGenotype * genotype = new cAnalyzeGenotype(m_world, genome, inst_set);
-
- // Determine the organism's original name -- strip off directory...
- while (filename.Find('/') != -1) filename.Pop('/');
- while (filename.Find('\\') != -1) filename.Pop('\\');
- filename.Replace(".gen", ""); // Remove the .gen from the filename.
- genotype->SetName(filename);
-
- // And save it in the current batch.
- batch[cur_batch].List().PushRear(genotype);
-
- // Adjust the flags on this batch
- batch[cur_batch].SetLineage(false);
- batch[cur_batch].SetAligned(false);
-}
-
-void cAnalyze::LoadBasicDump(cString cur_string)
-{
- // LOAD_BASE_DUMP
-
- cString filename = cur_string.PopWord();
-
- cout << "Loading: " << filename << endl;
-
- cInitFile input_file(filename);
- if (!input_file.IsOpen()) {
- cerr << "Error: Cannot load file: \"" << filename << "\"." << endl;
- exit(1);
- }
- input_file.Load();
- input_file.Compress();
- input_file.Close();
-
- // Setup the genome...
-
- for (int line_id = 0; line_id < input_file.GetNumLines(); line_id++) {
- cString cur_line = input_file.GetLine(line_id);
-
- // Setup the genotype for this line...
- cAnalyzeGenotype * genotype = new cAnalyzeGenotype(m_world, cur_line.PopWord(), inst_set);
- int num_cpus = cur_line.PopWord().AsInt();
- int id_num = cur_line.PopWord().AsInt();
- cString name = cStringUtil::Stringf("org-%d", id_num);
-
- genotype->SetNumCPUs(num_cpus);
- genotype->SetID(id_num);
- genotype->SetName(name);
-
- // Add this genotype to the proper batch.
- batch[cur_batch].List().PushRear(genotype);
- }
-
- // Adjust the flags on this batch
- batch[cur_batch].SetLineage(false);
- batch[cur_batch].SetAligned(false);
-}
-
-void cAnalyze::LoadDetailDump(cString cur_string)
-{
- cout << "Warning: Use of LoadDetailDump() is deprecated. Use \"LOAD\" instead." << endl;
- // LOAD_DETAIL_DUMP
-
- cString filename = cur_string.PopWord();
-
- cout << "Loading: " << filename << endl;
-
- cInitFile input_file(filename);
- if (!input_file.IsOpen()) {
- cerr << "Error: Cannot load file: \"" << filename << "\"." << endl;
- exit(1);
- }
- input_file.Load();
- input_file.Compress();
- input_file.Close();
-
- // Setup the genome...
-
- for (int line_id = 0; line_id < input_file.GetNumLines(); line_id++) {
- cString cur_line = input_file.GetLine(line_id);
-
- // Setup the genotype for this line...
-
- int id_num = cur_line.PopWord().AsInt();
- int parent_id = cur_line.PopWord().AsInt();
- int parent_dist = cur_line.PopWord().AsInt();
- int num_cpus = cur_line.PopWord().AsInt();
- int total_cpus = cur_line.PopWord().AsInt();
- int length = cur_line.PopWord().AsInt();
- double merit = cur_line.PopWord().AsDouble();
- int gest_time = cur_line.PopWord().AsInt();
- double fitness = cur_line.PopWord().AsDouble();
- int update_born = cur_line.PopWord().AsInt();
- int update_dead = cur_line.PopWord().AsInt();
- int depth = cur_line.PopWord().AsInt();
- cString name = cStringUtil::Stringf("org-%d", id_num);
-
- cAnalyzeGenotype * genotype = new cAnalyzeGenotype(m_world, cur_line.PopWord(), inst_set);
-
- genotype->SetID(id_num);
- genotype->SetParentID(parent_id);
- genotype->SetParentDist(parent_dist);
- genotype->SetNumCPUs(num_cpus);
- genotype->SetTotalCPUs(total_cpus);
- genotype->SetLength(length);
- genotype->SetMerit(merit);
- genotype->SetGestTime(gest_time);
- genotype->SetFitness(fitness);
- genotype->SetUpdateBorn(update_born);
- genotype->SetUpdateDead(update_dead);
- genotype->SetDepth(depth);
- genotype->SetName(name);
-
- // Add this genotype to the proper batch.
- batch[cur_batch].List().PushRear(genotype);
- }
-
- // Adjust the flags on this batch
- batch[cur_batch].SetLineage(false);
- batch[cur_batch].SetAligned(false);
-}
-
-void cAnalyze::LoadMultiDetail(cString cur_string)
-{
- // LOAD_MULTI_DETAIL
-
- int start_UD = cur_string.PopWord().AsInt();
- int step_UD = cur_string.PopWord().AsInt();
- int stop_UD = cur_string.PopWord().AsInt();
- cString data_directory = PopDirectory(cur_string, "./");
- cur_batch = cur_string.PopWord().AsInt();
-
- if (step_UD == 0) {
- cerr << "Error: LOAD_MULTI_DETAIL must have a non-zero value for step."
- << endl;
- return;
- }
-
- int num_steps = (stop_UD - start_UD) / step_UD + 1;
- if (num_steps > 1000) {
- cerr << "Warning: Loading over 1000 files in LOAD_MULTI_DETAIL!"
- << endl;
- }
-
- if (verbose == true) {
- cout << "Loading in " << num_steps
- << " detail files from update " << start_UD
- << " to update " << stop_UD
- << endl;
- } else {
- cout << "Running LOAD_MULTI_DETAIL" << endl;
- }
-
- for (int cur_UD = start_UD; cur_UD <= stop_UD; cur_UD += step_UD) {
- cString filename = cStringUtil::Stringf("detail_pop.%d", cur_UD);
-
- cout << "Loading '" << filename
- << "' into batch " << cur_batch
- << endl;
-
- cInitFile input_file(filename);
- if (!input_file.IsOpen()) {
- cerr << "Error: Cannot load file: \"" << filename << "\"." << endl;
- exit(1);
- }
- input_file.Load();
- input_file.Compress();
- input_file.Close();
-
- // Setup the genome...
-
- for (int line_id = 0; line_id < input_file.GetNumLines(); line_id++) {
- cString cur_line = input_file.GetLine(line_id);
-
- // Setup the genotype for this line...
-
- int id_num = cur_line.PopWord().AsInt();
- int parent_id = cur_line.PopWord().AsInt();
- int parent_dist = cur_line.PopWord().AsInt();
- int num_cpus = cur_line.PopWord().AsInt();
- int total_cpus = cur_line.PopWord().AsInt();
- int length = cur_line.PopWord().AsInt();
- double merit = cur_line.PopWord().AsDouble();
- int gest_time = cur_line.PopWord().AsInt();
- double fitness = cur_line.PopWord().AsDouble();
- int update_born = cur_line.PopWord().AsInt();
- int update_dead = cur_line.PopWord().AsInt();
- int depth = cur_line.PopWord().AsInt();
- cString name = cStringUtil::Stringf("org-%d", id_num);
-
- cAnalyzeGenotype * genotype = new cAnalyzeGenotype(m_world, cur_line.PopWord(), inst_set);
-
- genotype->SetID(id_num);
- genotype->SetParentID(parent_id);
- genotype->SetParentDist(parent_dist);
- genotype->SetNumCPUs(num_cpus);
- genotype->SetTotalCPUs(total_cpus);
- genotype->SetLength(length);
- genotype->SetMerit(merit);
- genotype->SetGestTime(gest_time);
- genotype->SetFitness(fitness);
- genotype->SetUpdateBorn(update_born);
- genotype->SetUpdateDead(update_dead);
- genotype->SetDepth(depth);
- genotype->SetName(name);
-
- // Add this genotype to the proper batch.
- batch[cur_batch].List().PushRear(genotype);
- }
-
- // Adjust the flags on this batch
- batch[cur_batch].SetLineage(false);
- batch[cur_batch].SetAligned(false);
-
- cur_batch++;
- }
-}
-
-void cAnalyze::LoadSequence(cString cur_string)
-{
- // LOAD_SEQUENCE
-
- static int sequence_count = 1;
- cString sequence = cur_string.PopWord();
- cString seq_name = cur_string.PopWord();
-
- cout << "Loading: " << sequence << endl;
-
- // Setup the genotype...
- cAnalyzeGenotype * genotype = new cAnalyzeGenotype(m_world, sequence, inst_set);
-
- genotype->SetNumCPUs(1); // Initialize to a single organism.
- if (seq_name == "") {
- seq_name = cStringUtil::Stringf("org-Seq%d", sequence_count);
- }
- genotype->SetName(seq_name);
- sequence_count++;
-
- // Add this genotype to the proper batch.
- batch[cur_batch].List().PushRear(genotype);
-
- // Adjust the flags on this batch
- batch[cur_batch].SetLineage(false);
- batch[cur_batch].SetAligned(false);
-}
-
-void cAnalyze::LoadDominant(cString cur_string)
-{
- (void) cur_string;
- cerr << "Warning: \"LOAD_DOMINANT\" not implmented yet!"<<endl;
-}
-
-// Clears the current time oriented list of resources and loads in a new one
-// from a file specified by the user, or resource.dat by default.
-void cAnalyze::LoadResources(cString cur_string)
-{
- resources.clear();
-
- int words = cur_string.CountNumWords();
-
- cString filename = "resource.dat";
- if(words >= 1) {
- filename = cur_string.PopWord();
- }
-
- cout << "Loading Resources from: " << filename << endl;
-
- // Process the resource.dat, currently assuming this is the only possible
- // input file
- ifstream resourceFile(filename, ios::in);
- assert(resourceFile);
-
- // Read in each line of the resource file and process it
- char line[2048];
- while(!resourceFile.eof()) {
- resourceFile.getline(line, 2047);
- if(line[0] == '\0') { continue; }
-
- // Get rid of the whitespace at the beginning of the stream, then
- // see if the line begins with a #, if so move on to the next line.
- stringstream ss;
- ss << line; ss >> ws; if(ss.peek() == '#') { continue; }
-
- // Read the update number from the stream
- int update;
- ss >> update; assert(ss.good());
-
- // Read in the concentration values for the current update and put them
- // into a temporary vector.
- double x;
- vector<double> tempValues;
- // Loop until the stream is no longer valid, which means either all the
- // data has been processed or a non-numeric was encountered. Currently
- // assuming a non-numeric is a comment denoting the rest of the line as
- // not informational.
- while(true) {
- ss >> ws; ss >> x;
- tempValues.push_back(x);
- if(!ss.good()) { ss.clear(); break; }
- }
- // Can't have no resources, so assert
- if(tempValues.empty()) { assert(0); }
- // add the update to the list of updates. Also assuming the values
- // in the file are already sorted. If this turns out not to be the
- // case you will need to sort on resources[i].first
- resources.push_back(make_pair(update, tempValues));
- }
- resourceFile.close();
-
- return;
-}
-
-
-// Looks up the resource concentrations that are the closest to the
-// given update and then fill in those concentrations into the environment.
-void cAnalyze::FillResources(int update)
-{
- // There must be some resources for at least one update
- //assert(!resources.empty());
- if(resources.empty()) { return; }
-
- int which = -1;
- // Assuming resource vector is sorted by update, front to back
- if(update <= resources[0].first) {
- which = 0;
- } else if(update >= resources.back().first) {
- which = resources.size() - 1;
- } else {
- // Find the update that is closest to the born update
- for(unsigned int i=0; i<resources.size()-1; i++) {
- if(update > resources[i+1].first) { continue; }
- if(update - resources[i].first <=
- resources[i+1].first - update) {
- which = i;
- } else {
- which = i + 1;
- }
- break;
- }
- }
- if(which < 0) { assert(0); }
-
-
-tArray<double> temp(resources[which].second.size());
-for(unsigned int i=0; i<resources[which].second.size(); i++) {
- temp[i] = resources[which].second[i];
-}
-m_world->GetTestCPU().SetupResourceArray(temp);
-
-return;
-}
-
-double cAnalyze::AnalyzeEntropy(cAnalyzeGenotype * genotype, double mu)
-{
- double entropy = 0.0;
-
- // If the fitness is 0, the entropy is the length of genotype ...
- genotype->Recalculate();
- if (genotype->GetFitness() == 0) {
- return genotype->GetLength();
- }
-
- // Calculate the stats for the genotype we're working with ...
- const int num_insts = inst_set.GetSize();
- const int num_lines = genotype->GetLength();
- const cGenome & base_genome = genotype->GetGenome();
- cGenome mod_genome(base_genome);
- double base_fitness = genotype->GetFitness();
-
- // Loop through all the lines of code, testing all mutations...
- tArray<double> test_fitness(num_insts);
- tArray<double> prob(num_insts);
- for (int line_no = 0; line_no < num_lines; line_no ++) {
- int cur_inst = base_genome[line_no].GetOp();
-
- // Test fitness of each mutant.
- for (int mod_inst = 0; mod_inst < num_insts; mod_inst++) {
- mod_genome[line_no].SetOp(mod_inst);
- cAnalyzeGenotype test_genotype(m_world, mod_genome, inst_set);
- test_genotype.Recalculate();
- // Ajust fitness ...
- if (test_genotype.GetFitness() <= base_fitness) {
- test_fitness[mod_inst] = test_genotype.GetFitness();
- } else {
- test_fitness[mod_inst] = base_fitness;
- }
- }
-
- // Calculate probabilities at mut-sel balance
- double w_bar = 1;
-
- // Normalize fitness values
- double maxFitness = 0.0;
- for(int i=0; i<num_insts; i++) {
- if(test_fitness[i] > maxFitness) {
- maxFitness = test_fitness[i];
- }
- }
-
-
- for(int i=0; i<num_insts; i++) {
- test_fitness[i] /= maxFitness;
- }
-
- while(1) {
- double sum = 0.0;
- for (int mod_inst = 0; mod_inst < num_insts; mod_inst ++) {
- prob[mod_inst] = (mu * w_bar) /
- ((double)num_insts *
- (w_bar + test_fitness[mod_inst] * mu - test_fitness[mod_inst]));
- sum = sum + prob[mod_inst];
- }
- if ((sum-1.0)*(sum-1.0) <= 0.0001)
- break;
- else
- w_bar = w_bar - 0.000001;
- }
-
- // Calculate entropy ...
- double this_entropy = 0.0;
- for (int i = 0; i < num_insts; i ++) {
- this_entropy += prob[i] * log((double) 1.0/prob[i]) / log ((double) num_insts);
- }
- entropy += this_entropy;
-
- // Reset the mod_genome back to the original sequence.
- mod_genome[line_no].SetOp(cur_inst);
- }
- return entropy;
-}
-
-double cAnalyze::AnalyzeEntropyGivenParent(cAnalyzeGenotype * genotype,
- cAnalyzeGenotype * parent, double mut_rate)
-{
- double entropy = 0.0;
-
- // Calculate the stats for the genotype we're working with ...
- genotype->Recalculate();
- const int num_insts = inst_set.GetSize();
- const int num_lines = genotype->GetLength();
- const cGenome & base_genome = genotype->GetGenome();
- const cGenome & parent_genome = parent->GetGenome();
- cGenome mod_genome(base_genome);
-
- // Loop through all the lines of code, testing all mutations ...
- tArray<double> test_fitness(num_insts);
- tArray<double> prob(num_insts);
- for (int line_no = 0; line_no < num_lines; line_no ++) {
- int cur_inst = base_genome[line_no].GetOp();
- int parent_inst = parent_genome[line_no].GetOp();
-
- // Test fitness of each mutant.
- for (int mod_inst = 0; mod_inst < num_insts; mod_inst++) {
- mod_genome[line_no].SetOp(mod_inst);
- cAnalyzeGenotype test_genotype(m_world, mod_genome, inst_set);
- test_genotype.Recalculate();
- test_fitness[mod_inst] = test_genotype.GetFitness();
- }
-
-
- // Calculate probabilities at mut-sel balance
- double w_bar = 1;
-
- // Normalize fitness values, assert if they are all zero
- double maxFitness = 0.0;
- for(int i=0; i<num_insts; i++) {
- if ( i == parent_inst) { continue; }
- if (test_fitness[i] > maxFitness) {
- maxFitness = test_fitness[i];
- }
- }
-
- if(maxFitness > 0) {
- for(int i = 0; i < num_insts; i ++) {
- if (i == parent_inst) { continue; }
- test_fitness[i] /= maxFitness;
- }
- } else {
- // every other inst is equally likely to be mutated to
- for (int i = 0; i < num_insts; i ++) {
- if (i == parent_inst) { continue; }
- test_fitness[i] = 1;
- }
- }
-
- double double_num_insts = num_insts * 1.0;
- while(1) {
- double sum = 0.0;
- for (int mod_inst = 0; mod_inst < num_insts; mod_inst ++) {
- if (mod_inst == parent_inst) { continue; }
- prob[mod_inst] = (mut_rate * w_bar) /
- (double_num_insts-2) /
- (w_bar + test_fitness[mod_inst] * mut_rate * (double_num_insts-1) / (double_num_insts - 2)
- - test_fitness[mod_inst]);
- sum = sum + prob[mod_inst];
- }
- if ((sum-1.0)*(sum-1.0) <= 0.0001)
- break;
- else
- w_bar = w_bar - 0.000001;
- }
-
- // Calculate entropy ...
- double this_entropy = 0.0;
- this_entropy -= (1.0 - mut_rate) * log(1.0 - mut_rate) / log(static_cast<double>(num_insts));
- for (int i = 0; i < num_insts; i ++) {
- if (i == parent_inst) { continue; }
- prob[i] = prob[i] * mut_rate;
- this_entropy += prob[i] * log(static_cast<double>(1.0/prob[i])) / log (static_cast<double>(num_insts));
- }
- entropy += this_entropy;
-
- // Reset the mod_genome back to the base_genome.
- mod_genome[line_no].SetOp(cur_inst);
- }
- return entropy;
-}
-
-double cAnalyze::IncreasedInfo(cAnalyzeGenotype * genotype1,
- cAnalyzeGenotype * genotype2,
- double mu)
-{
- double increased_info = 0.0;
-
- // Calculate the stats for the genotypes we're working with ...
- if ( genotype1->GetLength() != genotype2->GetLength() ) {
- cerr << "Error: Two genotypes don't have same length.(cAnalyze::IncreasedInfo)" << endl;
- exit(1);
- }
-
- genotype1->Recalculate();
- if (genotype1->GetFitness() == 0) {
- return 0.0;
- }
-
- const int num_insts = inst_set.GetSize();
- const int num_lines = genotype1->GetLength();
- const cGenome & genotype1_base_genome = genotype1->GetGenome();
- cGenome genotype1_mod_genome(genotype1_base_genome);
- double genotype1_base_fitness = genotype1->GetFitness();
- vector<double> genotype1_info(num_lines, 0.0);
-
- // Loop through all the lines of code, calculate genotype1 information
- tArray<double> test_fitness(num_insts);
- tArray<double> prob(num_insts);
- for (int line_no = 0; line_no < num_lines; line_no ++) {
- int cur_inst = genotype1_base_genome[line_no].GetOp();
-
- // Test fitness of each mutant.
- for (int mod_inst = 0; mod_inst < num_insts; mod_inst++) {
- genotype1_mod_genome[line_no].SetOp(mod_inst);
- cAnalyzeGenotype test_genotype(m_world, genotype1_mod_genome, inst_set);
- test_genotype.Recalculate();
- // Ajust fitness ...
- if (test_genotype.GetFitness() <= genotype1_base_fitness) {
- test_fitness[mod_inst] = test_genotype.GetFitness();
- } else {
- test_fitness[mod_inst] = genotype1_base_fitness;
- }
- }
-
- // Calculate probabilities at mut-sel balance
- double w_bar = 1;
-
- // Normalize fitness values
- double maxFitness = 0.0;
- for(int i=0; i<num_insts; i++) {
- if(test_fitness[i] > maxFitness) {
- maxFitness = test_fitness[i];
- }
- }
-
- for(int i=0; i<num_insts; i++) {
- test_fitness[i] /= maxFitness;
- }
-
- while(1) {
- double sum = 0.0;
- for (int mod_inst = 0; mod_inst < num_insts; mod_inst ++) {
- prob[mod_inst] = (mu * w_bar) /
- ((double)num_insts *
- (w_bar + test_fitness[mod_inst] * mu - test_fitness[mod_inst]));
- sum = sum + prob[mod_inst];
- }
- if ((sum-1.0)*(sum-1.0) <= 0.0001)
- break;
- else
- w_bar = w_bar - 0.000001;
- }
-
- // Calculate entropy ...
- double this_entropy = 0.0;
- for (int i = 0; i < num_insts; i ++) {
- this_entropy += prob[i] * log((double) 1.0/prob[i]) / log ((double) num_insts);
- }
- genotype1_info[line_no] = 1 - this_entropy;
-
- // Reset the mod_genome back to the original sequence.
- genotype1_mod_genome[line_no].SetOp(cur_inst);
- }
-
- genotype2->Recalculate();
- if (genotype2->GetFitness() == 0) {
- for (int line_no = 0; line_no < num_lines; ++ line_no) {
- increased_info += genotype1_info[line_no];
- }
- return increased_info;
- }
-
- const cGenome & genotype2_base_genome = genotype2->GetGenome();
- cGenome genotype2_mod_genome(genotype2_base_genome);
- double genotype2_base_fitness = genotype2->GetFitness();
-
- // Loop through all the lines of code, calculate increased information
- for (int line_no = 0; line_no < num_lines; line_no ++) {
- int cur_inst = genotype2_base_genome[line_no].GetOp();
-
- // Test fitness of each mutant.
- for (int mod_inst = 0; mod_inst < num_insts; mod_inst++) {
- genotype2_mod_genome[line_no].SetOp(mod_inst);
- cAnalyzeGenotype test_genotype(m_world, genotype2_mod_genome, inst_set);
- test_genotype.Recalculate();
- // Ajust fitness ...
- if (test_genotype.GetFitness() <= genotype2_base_fitness) {
- test_fitness[mod_inst] = test_genotype.GetFitness();
- } else {
- test_fitness[mod_inst] = genotype2_base_fitness;
- }
- }
-
- // Calculate probabilities at mut-sel balance
- double w_bar = 1;
-
- // Normalize fitness values, assert if they are all zero
- double maxFitness = 0.0;
- for(int i=0; i<num_insts; i++) {
- if(test_fitness[i] > maxFitness) {
- maxFitness = test_fitness[i];
- }
- }
-
- for(int i=0; i<num_insts; i++) {
- test_fitness[i] /= maxFitness;
- }
-
- while(1) {
- double sum = 0.0;
- for (int mod_inst = 0; mod_inst < num_insts; mod_inst ++) {
- prob[mod_inst] = (mu * w_bar) /
- ((double)num_insts *
- (w_bar + test_fitness[mod_inst] * mu - test_fitness[mod_inst]));
- sum = sum + prob[mod_inst];
- }
- if ((sum-1.0)*(sum-1.0) <= 0.0001)
- break;
- else
- w_bar = w_bar - 0.000001;
- }
-
- // Calculate entropy ...
- double this_entropy = 0.0;
- for (int i = 0; i < num_insts; i ++) {
- this_entropy += prob[i] * log((double) 1.0/prob[i]) / log ((double) num_insts);
- }
-
- // Compare information
- if (genotype1_info[line_no] > 1 - this_entropy) {
- increased_info += genotype1_info[line_no] - (1 - this_entropy);
- } // else increasing is 0, do nothing
-
- // Reset the mod_genome back to the original sequence.
- genotype2_mod_genome[line_no].SetOp(cur_inst);
- }
-
-
- return increased_info;
-}
-
-void cAnalyze::LoadFile(cString cur_string)
-{
- // LOAD
-
- cString filename = cur_string.PopWord();
-
- cout << "Loading: " << filename << endl;
-
- cInitFile input_file(filename);
- if (!input_file.IsOpen()) {
- cerr << "Error: Cannot load file: \"" << filename << "\"." << endl;
- exit(1);
- }
- input_file.Load();
- input_file.ReadHeader();
- input_file.Compress();
- input_file.Close();
-
- const cString filetype = input_file.GetFiletype();
- if (filetype != "population_data" && // Depricated
- filetype != "genotype_data") {
- cerr << "Error: Cannot load files of type \"" << filetype << "\"." << endl;
- exit(1);
- }
-
- if (verbose == true) {
- cout << "Loading file of type: " << filetype << endl;
- }
-
-
- // Construct a linked list of data types that can be loaded...
- tList< tDataEntryCommand<cAnalyzeGenotype> > output_list;
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > output_it(output_list);
- LoadGenotypeDataList(input_file.GetFormat(), output_list);
- bool id_inc = input_file.GetFormat().HasString("id");
-
- // Setup the genome...
- cGenome default_genome(1);
- int load_count = 0;
-
- for (int line_id = 0; line_id < input_file.GetNumLines(); line_id++) {
- cString cur_line = input_file.GetLine(line_id);
-
- cAnalyzeGenotype * genotype = new cAnalyzeGenotype(m_world, default_genome, inst_set);
-
- output_it.Reset();
- tDataEntryCommand<cAnalyzeGenotype> * data_command = NULL;
- while ((data_command = output_it.Next()) != NULL) {
- data_command->SetTarget(genotype);
- // genotype->SetSpecialArgs(data_command->GetArgs());
- data_command->SetValue(cur_line.PopWord());
- }
-
- // Give this genotype a name. Base it on the ID if possible.
- if (id_inc == false) {
- cString name = cStringUtil::Stringf("org-%d", load_count++);
- genotype->SetName(name);
- }
- else {
- cString name = cStringUtil::Stringf("org-%d", genotype->GetID());
- genotype->SetName(name);
- }
-
- // Add this genotype to the proper batch.
- batch[cur_batch].List().PushRear(genotype);
- }
-
- // Adjust the flags on this batch
- batch[cur_batch].SetLineage(false);
- batch[cur_batch].SetAligned(false);
-}
-
-
-//////////////// Reduction....
-
-void cAnalyze::FindGenotype(cString cur_string)
-{
- // If no arguments are passed in, just find max num_cpus.
- if (cur_string.GetSize() == 0) cur_string = "num_cpus";
-
- if (verbose == true) {
- cout << "Reducing batch " << cur_batch << " to genotypes: ";
- }
-
- tListPlus<cAnalyzeGenotype> & gen_list = batch[cur_batch].List();
- tListPlus<cAnalyzeGenotype> found_list;
- while (cur_string.CountNumWords() > 0) {
- cString gen_desc(cur_string.PopWord());
- if (verbose) cout << gen_desc << " ";
-
- // Determine by lin_type which genotype were are tracking...
- cAnalyzeGenotype * found_gen = PopGenotype(gen_desc, cur_batch);
-
- if (found_gen == NULL) {
- cerr << " Warning: genotype not found!" << endl;
- continue;
- }
-
- // Save this genotype...
- found_list.Push(found_gen);
- }
- cout << endl;
-
- // Delete all genotypes other than the ones found!
- while (gen_list.GetSize() > 0) delete gen_list.Pop();
-
- // And fill it back in with the good stuff.
- while (found_list.GetSize() > 0) gen_list.Push(found_list.Pop());
-
- // Adjust the flags on this batch
- batch[cur_batch].SetLineage(false);
- batch[cur_batch].SetAligned(false);
-}
-
-void cAnalyze::FindOrganism(cString cur_string)
-{
- // At least one argument is rquired.
- if (cur_string.GetSize() == 0) {
- cerr << "Error: At least one argument is required in FIND_ORGANISM." << endl;
- return;
- }
-
- if (verbose == true) {
- cout << "Reducing batch " << cur_batch << " to organisms: " << endl;
- }
-
- tListPlus<cAnalyzeGenotype> & gen_list = batch[cur_batch].List();
- tListPlus<cAnalyzeGenotype> found_list;
-
- tArray<int> new_counts(gen_list.GetSize());
- new_counts.SetAll(0);
-
- while (cur_string.CountNumWords() > 0) {
- cString org_desc(cur_string.PopWord());
- if (verbose) cout << org_desc << " ";
-
- // Determine by org_desc which genotype were are tracking...
- if (org_desc == "random") {
- bool found = false;
- int num_orgs = gen_list.Count(&cAnalyzeGenotype::GetNumCPUs);
- while (found != true) {
- int org_chosen = random.GetUInt(num_orgs);
- cAnalyzeGenotype * found_genotype =
- gen_list.FindSummedValue(org_chosen, &cAnalyzeGenotype::GetNumCPUs);
- if ( found_genotype->GetNumCPUs() != 0 && found_genotype->GetViable()) {
- found_genotype->SetNumCPUs(found_genotype->GetNumCPUs()-1);
- new_counts[gen_list.FindPosPtr(found_genotype)] +=1;
- cout << "Found genotype " << gen_list.FindPosPtr(found_genotype) << endl;
- found = true;
- }
- }
- }
-
- // pick a random organisms, with replacement!
- if (org_desc == "randomWR") {
- bool found = false;
- int num_orgs = gen_list.Count(&cAnalyzeGenotype::GetNumCPUs);
- while (found != true) {
- int org_chosen = random.GetUInt(num_orgs);
- cAnalyzeGenotype * found_genotype =
- gen_list.FindSummedValue(org_chosen, &cAnalyzeGenotype::GetNumCPUs);
- if ( found_genotype->GetNumCPUs() != 0 && found_genotype->GetViable()) {
- new_counts[gen_list.FindPosPtr(found_genotype)] +=1;
- cout << "Found genotype " << gen_list.FindPosPtr(found_genotype) << endl;
- found = true;
- }
- }
- }
- }
-
- int pos_count = 0;
- cAnalyzeGenotype * genotype = NULL;
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
-
- while ((genotype = batch_it.Next()) != NULL) {
- genotype->SetNumCPUs(new_counts[pos_count]);
- if (genotype->GetNumCPUs() == 0) batch_it.Remove();
- else cout << "Genotype " << pos_count << " has " << new_counts[pos_count] << " organisms." << endl;
- pos_count++;
- }
-
- // Adjust the flags on this batch
- batch[cur_batch].SetLineage(false);
- batch[cur_batch].SetAligned(false);
-}
-
-void cAnalyze::FindLineage(cString cur_string)
-{
- cString lin_type = "num_cpus";
- if (cur_string.CountNumWords() > 0) lin_type = cur_string.PopWord();
-
- if (verbose == true) {
- cout << "Reducing batch " << cur_batch
- << " to " << lin_type << " lineage " << endl;
- } else cout << "Performing lineage scan..." << endl;
-
-
- // Determine by lin_type which genotype we are tracking...
- cAnalyzeGenotype * found_gen = PopGenotype(lin_type, cur_batch);
-
- if (found_gen == NULL) {
- cerr << " Warning: Genotype " << lin_type
- << " not found. Lineage scan aborted." << endl;
- return;
- }
-
- // Otherwise, trace back through the id numbers to mark all of those
- // in the ancestral lineage...
-
- // Construct a list of genotypes found...
-
- tListPlus<cAnalyzeGenotype> found_list;
- found_list.Push(found_gen);
- int next_id = found_gen->GetParentID();
- bool found = true;
- while (found == true) {
- found = false;
-
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- while ((found_gen = batch_it.Next()) != NULL) {
- if (found_gen->GetID() == next_id) {
- batch_it.Remove();
- found_list.Push(found_gen);
- next_id = found_gen->GetParentID();
- found = true;
- break;
- }
- }
- }
-
- // We now have all of the genotypes in this lineage, delete everything
- // else.
-
- const int total_removed = batch[cur_batch].List().GetSize();
- while (batch[cur_batch].List().GetSize() > 0) {
- delete batch[cur_batch].List().Pop();
- }
-
- // And fill it back in with the good stuff.
- int total_kept = found_list.GetSize();
- while (found_list.GetSize() > 0) {
- batch[cur_batch].List().PushRear(found_list.Pop());
- }
-
- if (verbose == true) {
- cout << " Lineage has " << total_kept << " genotypes; "
- << total_removed << " were removed." << endl;
- }
-
- // Adjust the flags on this batch
- batch[cur_batch].SetLineage(true);
- batch[cur_batch].SetAligned(false);
-}
-
-void cAnalyze::FindSexLineage(cString cur_string)
-{
-
- // detemine the method for construicting a lineage
- // by defauly, find the lineage of the final dominant
- cString lin_type = "num_cpus";
- if (cur_string.CountNumWords() > 0) lin_type = cur_string.PopWord();
-
- // parent_method determins which of the two parental lineages to use
- // "rec_region_size" :
- // "mother" (dominant parent) is the parent contributing
- // more to the offspring genome (default)
- // "genome_size" :
- // "mother" (dominant parent) is the longer parent
- cString parent_method = "rec_region_size";
- if (cur_string.CountNumWords() > 0) parent_method = cur_string.PopWord();
-
- if (verbose == true) {
- cout << "Reducing batch " << cur_batch
- << " to " << lin_type << " sexual lineage "
- << " using " << parent_method << " criteria." << endl;
- } else cout << "Performing sexual lineage scan..." << endl;
-
-
- // Determine by lin_type which genotype we are tracking...
- cAnalyzeGenotype * found_gen = PopGenotype(lin_type, cur_batch);
-
- cAnalyzeGenotype * found_dad;
- cAnalyzeGenotype * found_mom;
- cAnalyzeGenotype * found_temp;
-
- if (found_gen == NULL) {
- cerr << " Warning: Genotype " << lin_type
- << " not found. Sexual lineage scan aborted." << endl;
- return;
- }
-
- // Otherwise, trace back through the id numbers to mark all of those
- // in the ancestral lineage...
-
- // Construct a list of genotypes found...
-
- tListPlus<cAnalyzeGenotype> found_list;
- found_list.Push(found_gen);
- int next_id1 = found_gen->GetParentID();
- int next_id2 = found_gen->GetParent2ID();
-
- bool found_m = true;
- bool found_d = true;
-
- while (found_m == true & found_d == true) {
-
- //cout << "Searching for mom=" << next_id1
- // << " and dad=" << next_id2 << endl;
- found_m = false;
- found_d = false;
-
- // Look for the father first....
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- batch_it.Reset();
- while ((found_dad = batch_it.Next()) != NULL) {
- // Check if the father is found...
- if (found_dad->GetID() == next_id2) {
- //cout << "Found dad!" << endl;
- batch_it.Remove();
- found_list.Push(found_dad);
- found_d = true;
- break;
- }
- }
-
- // dad may have already been found, check the find list!
- if (found_d == false) {
- tListIterator<cAnalyzeGenotype> found_it(found_list);
- while ((found_dad = found_it.Next()) != NULL) {
- if (found_dad->GetID() == next_id2) {
- //cout << "Found dad in found list!" << endl;
- found_d = true;
- break;
- }
- }
- }
-
- // Next, look for the mother...
- batch_it.Reset();
- while ((found_mom = batch_it.Next()) != NULL) {
- if (found_mom->GetID() == next_id1) {
- //cout << "Found mom!" << endl;
- batch_it.Remove();
- found_list.Push(found_mom);
- // if finding lineages by parental length, may have to swap
- if (parent_method == "genome_size" &
- found_mom->GetLength() < found_dad->GetLength()) {
- //cout << "Swapping parents!" << endl;
- found_temp = found_mom;
- found_mom = found_dad;
- found_dad = found_temp;
- }
- next_id1 = found_mom->GetParentID();
- next_id2 = found_mom->GetParent2ID();
- found_m = true;
- break;
- }
- }
-
- // If the mother was not found, it may already have been placed in the
- // found list as a father...
- if (found_m == false) {
- tListIterator<cAnalyzeGenotype> found_it(found_list);
- while ((found_mom = found_it.Next()) != NULL) {
- if (found_mom->GetID() == next_id1) {
- //cout << "Found mom as dad!" << endl;
- // Don't move to found list, since its already there, but update
- // to the next ids.
- // if finding lineages by parental length, may have to swap
- if (parent_method == "genome_size" &
- found_mom->GetLength() < found_dad->GetLength()) {
- //cout << "Swapping parents!" << endl;
- found_temp = found_mom;
- found_mom = found_dad;
- found_dad = found_temp;
- }
- next_id1 = found_mom->GetParentID();
- next_id2 = found_mom->GetParent2ID();
- found_m = true;
- break;
- }
- }
- }
- }
-
- // We now have all of the genotypes in this lineage, delete everything
- // else.
-
- const int total_removed = batch[cur_batch].List().GetSize();
- while (batch[cur_batch].List().GetSize() > 0) {
- delete batch[cur_batch].List().Pop();
- }
-
- // And fill it back in with the good stuff.
- int total_kept = found_list.GetSize();
- while (found_list.GetSize() > 0) {
- batch[cur_batch].List().PushRear(found_list.Pop());
- }
-
- if (verbose == true) {
- cout << " Sexual lineage has " << total_kept << " genotypes; "
- << total_removed << " were removed." << endl;
- }
-
- // Adjust the flags on this batch
- batch[cur_batch].SetLineage(false);
- batch[cur_batch].SetAligned(false);
-}
-
-void cAnalyze::FindClade(cString cur_string)
-{
- if (cur_string.GetSize() == 0) {
- cerr << " Warning: No clade specified for FIND_CLADE. Aborting." << endl;
- return;
- }
-
- cString clade_type( cur_string.PopWord() );
-
- if (verbose == true) {
- cout << "Reducing batch " << cur_batch
- << " to clade " << clade_type << "." << endl;
- } else cout << "Performing clade scan..." << endl;
-
-
- // Determine by clade_type which genotype we are tracking...
- cAnalyzeGenotype * found_gen = PopGenotype(clade_type, cur_batch);
-
- if (found_gen == NULL) {
- cerr << " Warning: Ancestral genotype " << clade_type
- << " not found. Clade scan aborted." << endl;
- return;
- }
-
- // Do this the brute force way... scan for one step at a time.
-
- // Construct a list of genotypes found...
-
- tListPlus<cAnalyzeGenotype> found_list; // Found and finished.
- tListPlus<cAnalyzeGenotype> scan_list; // Found, but need to scan for children.
- scan_list.Push(found_gen);
-
- // Keep going as long as there is something in the scan list...
- while (scan_list.GetSize() > 0) {
- // Move the next genotype from the scan list to the found_list.
- found_gen = scan_list.Pop();
- int parent_id = found_gen->GetID();
- found_list.Push(found_gen);
-
- // Seach for all of the children of this genotype...
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- while ((found_gen = batch_it.Next()) != NULL) {
- // If we found a child, place it into the scan list.
- if (found_gen->GetParentID() == parent_id) {
- batch_it.Remove();
- scan_list.Push(found_gen);
- }
- }
- }
-
- // We now have all of the genotypes in this clade, delete everything else.
-
- const int total_removed = batch[cur_batch].List().GetSize();
- while (batch[cur_batch].List().GetSize() > 0) {
- delete batch[cur_batch].List().Pop();
- }
-
- // And fill it back in with the good stuff.
- int total_kept = found_list.GetSize();
- while (found_list.GetSize() > 0) {
- batch[cur_batch].List().PushRear(found_list.Pop());
- }
-
- if (verbose == true) {
- cout << " Clade has " << total_kept << " genotypes; "
- << total_removed << " were removed." << endl;
- }
-
- // Adjust the flags on this batch
- batch[cur_batch].SetLineage(false);
- batch[cur_batch].SetAligned(false);
-}
-
-void cAnalyze::SampleOrganisms(cString cur_string)
-{
- double fraction = cur_string.PopWord().AsDouble();
- int init_genotypes = batch[cur_batch].List().GetSize();
-
- double test_viable = 0;
- if (cur_string.GetSize() > 0) {
- test_viable = cur_string.PopWord().AsDouble();
- }
-
- if (verbose == true) {
- cout << "Sampling " << fraction << " organisms from batch "
- << cur_batch << "." << endl;
- }
- else cout << "Sampling Organisms..." << endl;
-
- cAnalyzeGenotype * genotype = NULL;
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
-
- // Loop through all genotypes to perform a census
- int org_count = 0;
- while ((genotype = batch_it.Next()) != NULL) {
- // If we require viables, reduce all non-viables to zero organisms.
- if (test_viable == 1 && genotype->GetViable() == 0) {
- genotype->SetNumCPUs(0);
- }
-
- // Count the number of organisms in this genotype.
- org_count += genotype->GetNumCPUs();
- }
-
- // Create an array to store pointers to the genotypes and fill it in
- // while temporarily resetting all of the organism counts to zero.
- tArray<cAnalyzeGenotype *> org_array(org_count);
- int cur_org = 0;
- batch_it.Reset();
- while ((genotype = batch_it.Next()) != NULL) {
- for (int i = 0; i < genotype->GetNumCPUs(); i++) {
- org_array[cur_org] = genotype;
- cur_org++;
- }
- genotype->SetNumCPUs(0);
- }
-
- assert(cur_org == org_count);
-
- // Determine how many organisms we want to keep.
- int new_org_count = (int) fraction;
- if (fraction < 1.0) new_org_count = (int) (fraction * (double) org_count);
- if (new_org_count > org_count) {
- cerr << "Warning: Trying to sample " << new_org_count
- << "organisms from a population of " << org_count
- << endl;
- new_org_count = org_count;
- }
-
- // Now pick those that we are keeping.
- tArray<int> keep_ids(new_org_count);
- random.Choose(org_count, keep_ids);
-
- // And increment the org counts for the associated genotypes.
- for (int i = 0; i < new_org_count; i++) {
- genotype = org_array[ keep_ids[i] ];
- genotype->SetNumCPUs(genotype->GetNumCPUs() + 1);
- }
-
-
- // Delete all genotypes with no remaining organisms...
- batch_it.Reset();
- while ((genotype = batch_it.Next()) != NULL) {
- if (genotype->GetNumCPUs() == 0) {
- batch_it.Remove();
- delete genotype;
- }
- }
-
- int num_genotypes = batch[cur_batch].List().GetSize();
- if (verbose) {
- cout << " Removed " << org_count - new_org_count
- << " organisms (" << init_genotypes - num_genotypes
- << " genotypes); " << new_org_count
- << " orgs (" << num_genotypes << " gens) remaining."
- << endl;
- }
-
- // Adjust the flags on this batch
- batch[cur_batch].SetLineage(false);
- batch[cur_batch].SetAligned(false);
-}
-
-
-void cAnalyze::SampleGenotypes(cString cur_string)
-{
- double fraction = cur_string.PopWord().AsDouble();
- int init_genotypes = batch[cur_batch].List().GetSize();
-
- double test_viable = 0;
- if (cur_string.GetSize() > 0) {
- test_viable = cur_string.PopWord().AsDouble();
- }
-
- if (verbose == true) {
- cout << "Sampling " << fraction << " genotypes from batch "
- << cur_batch << "." << endl;
- }
- else cout << "Sampling Genotypes..." << endl;
-
- double frac_remove = 1.0 - fraction;
-
- cAnalyzeGenotype * genotype = NULL;
-
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- while ((genotype = batch_it.Next()) != NULL) {
- if (random.P(frac_remove) || ((genotype->GetViable())==0 && test_viable==1) ) {
- batch_it.Remove();
- delete genotype;
- }
- }
-
- int num_genotypes = batch[cur_batch].List().GetSize();
- if (verbose == true) {
- cout << " Removed " << init_genotypes - num_genotypes
- << " genotypes; " << num_genotypes << " remaining."
- << endl;
- }
-
- // Adjust the flags on this batch
- batch[cur_batch].SetLineage(false);
- batch[cur_batch].SetAligned(false);
-}
-
-void cAnalyze::KeepTopGenotypes(cString cur_string)
-{
- const int num_kept = cur_string.PopWord().AsInt();
- const int num_genotypes = batch[cur_batch].List().GetSize();
- const int num_removed = num_genotypes - num_kept;
-
- for (int i = 0; i < num_removed; i++) {
- delete batch[cur_batch].List().PopRear();
- }
-
- // Adjust the flags on this batch
- // batch[cur_batch].SetLineage(false); // Should not destroy a lineage...
- batch[cur_batch].SetAligned(false);
-}
-
-
-
-//////////////// Output Commands...
-
-void cAnalyze::CommandPrint(cString cur_string)
-{
- if (verbose == true) cout << "Printing batch " << cur_batch << endl;
- else cout << "Printing organisms..." << endl;
-
- cString directory = PopDirectory(cur_string, "genebank/");
-
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- cString filename(directory);
-
- if (cur_string.GetSize() > 0) {
- filename += cur_string.PopWord();
- }
- else {
- filename += genotype->GetName();
- filename += ".gen";
- }
-
- cTestUtil::PrintGenome(m_world, genotype->GetGenome(), filename);
- if (verbose) cout << "Printing: " << filename << endl;
- }
-}
-
-void cAnalyze::CommandTrace(cString cur_string)
-{
- if (verbose == true) cout << "Tracing batch " << cur_batch << endl;
- else cout << "Tracing organisms..." << endl;
-
- int words = cur_string.CountNumWords();
-
- cString dir = cur_string.PopWord();
- cString defaultDirectory = "genebank/";
- cString directory = PopDirectory(dir, defaultDirectory);
-
- int useResources = 0;
- if(words >= 2) {
- useResources = cur_string.PopWord().AsInt();
- // All non-zero values are considered false
- if(useResources != 0 && useResources != 1) {
- useResources = 0;
- }
- }
-
- bool backupUsage;
- tArray<double> backupResources;
-
- if(useResources) {
- // Backup test cpu data
- backupUsage = m_world->GetTestCPU().UseResources();
- backupResources = m_world->GetTestCPU().GetResources();
-
- m_world->GetTestCPU().UseResources() = true;
- }
-
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- cString filename(directory);
- filename += genotype->GetName();
- filename += ".trace";
-
- if (genotype->GetGenome().GetSize() == 0) {
- break;
- }
-
- // Build the hardware status printer for tracing.
- ofstream trace_fp;
- trace_fp.open(filename);
- assert (trace_fp.good() == true); // Unable to open trace file.
- cHardwareStatusPrinter trace_printer(trace_fp);
-
- // Build the test info for printing.
- cCPUTestInfo test_info;
- test_info.TestThreads();
- test_info.SetTraceExecution(&trace_printer);
-
- m_world->GetTestCPU().TestGenome(test_info, genotype->GetGenome());
-
- if (verbose) cout << " Tracing: " << filename << endl;
- trace_fp.close();
- }
-
- if(useResources) {
- // Set the test cpu back to the state it was in before we messed with it
- m_world->GetTestCPU().UseResources() = backupUsage;
- m_world->GetTestCPU().SetupResourceArray(backupResources);
- }
-
- return;
-}
-
-void cAnalyze::CommandPrintTasks(cString cur_string)
-{
- if (verbose == true) cout << "Printing tasks in batch " << cur_batch << endl;
- else cout << "Printing tasks..." << endl;
-
- // Load in the variables...
- cString filename("tasks.dat");
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
-
- ofstream & fp = data_file_manager.GetOFStream(filename);
-
- // Loop through all of the genotypes in this batch...
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- fp << genotype->GetID() << " ";
- genotype->PrintTasks(fp);
- fp << endl;
- }
-}
-
-void cAnalyze::CommandDetail(cString cur_string)
-{
- if (verbose == true) cout << "Detailing batch " << cur_batch << endl;
- else cout << "Detailing..." << endl;
-
- // Load in the variables...
- cString filename("detail.dat");
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
-
- // Construct a linked list of details needed...
- tList< tDataEntryCommand<cAnalyzeGenotype> > output_list;
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > output_it(output_list);
- LoadGenotypeDataList(cur_string, output_list);
-
- // Determine the file type...
- int file_type = FILE_TYPE_TEXT;
- cString file_extension(filename);
- while (file_extension.Find('.') != -1) file_extension.Pop('.');
- if (file_extension == "html") file_type = FILE_TYPE_HTML;
-
- // Setup the file...
- if (filename == "cout") {
- CommandDetail_Header(cout, file_type, output_it);
- CommandDetail_Body(cout, file_type, output_it);
- } else {
- ofstream & fp = data_file_manager.GetOFStream(filename);
- CommandDetail_Header(fp, file_type, output_it);
- CommandDetail_Body(fp, file_type, output_it);
- }
-
- // And clean up...
- while (output_list.GetSize() != 0) delete output_list.Pop();
-}
-
-
-void cAnalyze::CommandDetailTimeline(cString cur_string)
-{
- if (verbose == true) cout << "Detailing batch "
- << cur_batch << " based on time" << endl;
- else cout << "Detailing..." << endl;
-
- // Load in the variables...
- cString filename("detail_timeline.dat");
- int time_step = 100;
- int max_time = 100000;
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
- if (cur_string.GetSize() != 0) time_step = cur_string.PopWord().AsInt();
- if (cur_string.GetSize() != 0) max_time = cur_string.PopWord().AsInt();
-
- if (verbose == true) {
- cout << " Time step = " << time_step << endl
- << " Max time = " << max_time << endl;
- }
-
- // Construct a linked list of details needed...
- tList< tDataEntryCommand<cAnalyzeGenotype> > output_list;
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > output_it(output_list);
- LoadGenotypeDataList(cur_string, output_list);
-
- // Determine the file type...
- int file_type = FILE_TYPE_TEXT;
- cString file_extension(filename);
- while (file_extension.Find('.') != -1) file_extension.Pop('.');
- if (file_extension == "html") file_type = FILE_TYPE_HTML;
-
- // Setup the file...
- if (filename == "cout") {
- CommandDetail_Header(cout, file_type, output_it, time_step);
- CommandDetail_Body(cout, file_type, output_it, time_step, max_time);
- } else {
- ofstream & fp = data_file_manager.GetOFStream(filename);
- CommandDetail_Header(fp, file_type, output_it, time_step);
- CommandDetail_Body(fp, file_type, output_it, time_step, max_time);
- }
-
- // And clean up...
- while (output_list.GetSize() != 0) delete output_list.Pop();
-}
-
-
-void cAnalyze::CommandDetail_Header(ostream & fp, int format_type,
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > & output_it,
- int time_step)
-{
- // Write out the header on the file
- if (format_type == FILE_TYPE_TEXT) {
- fp << "#filetype genotype_data" << endl;
- fp << "#format ";
- if (time_step > 0) fp << "update ";
- while (output_it.Next() != NULL) {
- const cString & entry_name = output_it.Get()->GetName();
- fp << entry_name << " ";
- }
- fp << endl << endl;
-
- // Give the more human-readable legend.
- fp << "# Legend:" << endl;
- int count = 0;
- if (time_step > 0) fp << "# " << ++count << ": Update" << endl;
- while (output_it.Next() != NULL) {
- const cString & entry_desc = output_it.Get()->GetDesc();
- fp << "# " << ++count << ": " << entry_desc << endl;
- }
- fp << endl;
- } else { // if (format_type == FILE_TYPE_HTML) {
- fp << "<html>" << endl
- << "<body bgcolor=\"#FFFFFF\"" << endl
- << " text=\"#000000\"" << endl
- << " link=\"#0000AA\"" << endl
- << " alink=\"#0000FF\"" << endl
- << " vlink=\"#000044\">" << endl
- << endl
- << "<h1 align=center>Run " << batch[cur_batch].Name() << endl
- << endl
- << "<center>" << endl
- << "<table border=1 cellpadding=2><tr>" << endl;
-
- if (time_step > 0) fp << "<th bgcolor=\"#AAAAFF\">Update ";
- while (output_it.Next() != NULL) {
- const cString & entry_desc = output_it.Get()->GetDesc();
- fp << "<th bgcolor=\"#AAAAFF\">" << entry_desc << " ";
- }
- fp << "</tr>" << endl;
-
- }
-
- }
-
-
-void cAnalyze::CommandDetail_Body(ostream & fp, int format_type,
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > & output_it,
- int time_step, int max_time)
-{
- // Loop through all of the genotypes in this batch...
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * cur_genotype = batch_it.Next();
- cAnalyzeGenotype * next_genotype = batch_it.Next();
- cAnalyzeGenotype * prev_genotype = NULL;
-
- int cur_time = 0;
- while (cur_genotype != NULL && cur_time <= max_time) {
- output_it.Reset();
- tDataEntryCommand<cAnalyzeGenotype> * data_command = NULL;
- if (format_type == FILE_TYPE_HTML) {
- fp << "<tr>";
- if (time_step > 0) fp << "<td>" << cur_time << " ";
- }
- else if (time_step > 0) { // TEXT file, printing times...
- fp << cur_time << " ";
- }
-
- while ((data_command = output_it.Next()) != NULL) {
- data_command->SetTarget(cur_genotype);
- cur_genotype->SetSpecialArgs(data_command->GetArgs());
- if (format_type == FILE_TYPE_HTML) {
- int compare = 0;
- if (prev_genotype) {
- prev_genotype->SetSpecialArgs(data_command->GetArgs());
- compare = data_command->Compare(prev_genotype);
- }
- data_command->HTMLPrint(fp, compare);
- }
- else { // if (format_type == FILE_TYPE_TEXT) {
- fp << data_command->GetEntry() << " ";
- }
- }
- if (format_type == FILE_TYPE_HTML) fp << "</tr>";
- fp << endl;
-
- cur_time += time_step;
- if (time_step > 0) {
- while (next_genotype && next_genotype->GetUpdateBorn() < cur_time) {
- prev_genotype = cur_genotype;
- cur_genotype = next_genotype;
- next_genotype = batch_it.Next();
- }
- }
- else {
- // Always moveon if we're not basing this on time, or if we've run out
- // of genotypes.
- prev_genotype = cur_genotype;
- cur_genotype = next_genotype;
- next_genotype = batch_it.Next();
- }
-
- }
-
- // If in HTML mode, we need to end the file...
- if (format_type == FILE_TYPE_HTML) {
- fp << "</table>" << endl
- << "</center>" << endl;
- }
-}
-
-void cAnalyze::CommandDetailAverage_Body(ostream & fp, int num_outputs,
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > & output_it)
-{
- // Loop through all of the genotypes in this batch...
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * cur_genotype = batch_it.Next();
- cAnalyzeGenotype * next_genotype = batch_it.Next();
- cAnalyzeGenotype * prev_genotype = NULL;
-
- tArray<cDoubleSum> output_counts(num_outputs);
- for (int i = 0; i < num_outputs; i++) { output_counts[i].Clear();}
- int count;
- while (cur_genotype != NULL) {
- count = 0;
- output_it.Reset();
- tDataEntryCommand<cAnalyzeGenotype> * data_command = NULL;
- while ((data_command = output_it.Next()) != NULL) {
- data_command->SetTarget(cur_genotype);
- cur_genotype->SetSpecialArgs(data_command->GetArgs());
- for (int j = 0; j < cur_genotype->GetNumCPUs(); j++) {
- output_counts[count].Add((double) data_command->GetEntry().AsString().AsDouble());
- }
- count++;
- }
-
- prev_genotype = cur_genotype;
- cur_genotype = next_genotype;
- next_genotype = batch_it.Next();
- }
- fp << batch[cur_batch].Name() << " ";
- for (int i = 0; i < num_outputs; i++) {
- fp << output_counts[i].Average() << " ";
- }
- fp << endl;
-}
-
-void cAnalyze::CommandDetailAverage(cString cur_string)
-{
- if (verbose == true) cout << "Average detailing batch " << cur_batch << endl;
- else cout << "Detailing..." << endl;
-
- // Load in the variables...
- cString filename("detail.dat");
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
-
- // Construct a linked list of details needed...
- tList< tDataEntryCommand<cAnalyzeGenotype> > output_list;
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > output_it(output_list);
- LoadGenotypeDataList(cur_string, output_list);
-
- // check if file is already in use.
- bool file_active = data_file_manager.IsOpen(filename);
-
- ofstream & fp = data_file_manager.GetOFStream(filename);
-
- // if it's a new file print out the header
- if (file_active == false) {
- CommandDetail_Header(fp, FILE_TYPE_TEXT, output_it);
- }
- CommandDetailAverage_Body(fp, cur_string.CountNumWords(), output_it);
-
- while (output_list.GetSize() != 0) delete output_list.Pop();
-
-}
-
-void cAnalyze::CommandDetailBatches(cString cur_string)
-{
- // Load in the variables...
- cString keyword("num_cpus");
- cString filename("detail_batch.dat");
- if (cur_string.GetSize() != 0) keyword = cur_string.PopWord();
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
-
- if (verbose == true) cout << "Detailing batches for " << keyword << endl;
- else cout << "Detailing Batches..." << endl;
-
- // Scan the functions list for the keyword we need...
- SetupGenotypeDataList();
- tListIterator< tDataEntryBase<cAnalyzeGenotype> >
- output_it(genotype_data_list);
-
- // Divide up the keyword into its acrual entry and its arguments...
- cString cur_args = keyword;
- cString cur_entry = cur_args.Pop(':');
-
- // Find its associated command...
- tDataEntryCommand<cAnalyzeGenotype> * cur_command = NULL;
- bool found = false;
- while (output_it.Next() != NULL) {
- if (output_it.Get()->GetName() == cur_entry) {
- cur_command = new tDataEntryCommand<cAnalyzeGenotype>
- (output_it.Get(), cur_args);
- found = true;
- break;
- }
- }
- if (found == false) {
- cout << "Error: Unknown data type: " << cur_entry << endl;
- return;
- }
-
-
- // Determine the file type...
- int file_type = FILE_TYPE_TEXT;
- cString file_extension(filename);
- while (file_extension.Find('.') != -1) file_extension.Pop('.');
- if (file_extension == "html") file_type = FILE_TYPE_HTML;
-
- ofstream & fp = data_file_manager.GetOFStream(filename);
-
- // Write out the header on the file
- if (file_type == FILE_TYPE_TEXT) {
- fp << "#filetype batch_data" << endl
- << "#format batch_id " << keyword << endl
- << endl;
-
- // Give the more human-readable legend.
- fp << "# Legend:" << endl
- << "# Column 1 = Batch ID" << endl
- << "# Remaining entries: " << cur_command->GetDesc() << endl
- << endl;
-
- } else { // if (file_type == FILE_TYPE_HTML) {
- fp << "<html>" << endl
- << "<body bgcolor=\"#FFFFFF\"" << endl
- << " text=\"#000000\"" << endl
- << " link=\"#0000AA\"" << endl
- << " alink=\"#0000FF\"" << endl
- << " vlink=\"#000044\">" << endl
- << endl
- << "<h1 align=center> Distribution of " << cur_command->GetDesc()
- << endl << endl
- << "<center>" << endl
- << "<table border=1 cellpadding=2>" << endl
- << "<tr><th bgcolor=\"#AAAAFF\">" << cur_command->GetDesc() << "</tr>"
- << endl;
- }
-
-
- // Loop through all of the batches...
- for (int i = 0; i < MAX_BATCHES; i++) {
- if (batch[i].List().GetSize() == 0) continue;
-
- if (file_type == FILE_TYPE_HTML) fp << "<tr><td>";
- fp << i << " ";
-
- tListIterator<cAnalyzeGenotype> batch_it(batch[i].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- output_it.Reset();
- if (file_type == FILE_TYPE_HTML) fp << "<td>";
-
- cur_command->SetTarget(genotype);
- genotype->SetSpecialArgs(cur_command->GetArgs());
- if (file_type == FILE_TYPE_HTML) {
- cur_command->HTMLPrint(fp, 0);
- }
- else { // if (file_type == FILE_TYPE_TEXT) {
- fp << cur_command->GetEntry() << " ";
- }
- }
- if (file_type == FILE_TYPE_HTML) fp << "</tr>";
- fp << endl;
- }
-
- // If in HTML mode, we need to end the file...
- if (file_type == FILE_TYPE_HTML) {
- fp << "</table>" << endl
- << "</center>" << endl;
- }
-
- delete cur_command;
- }
-
-
-
-void cAnalyze::CommandDetailIndex(cString cur_string)
-{
- cout << "Creating a Detail Index..." << endl;
-
- // A filename and min and max batches must be included.
- if (cur_string.CountNumWords() < 3) {
- cerr << "Error: must include filename, and min and max batch numbers."
- << endl;
- exit(1);
- }
-
- // Load in the variables...
- cString filename(cur_string.PopWord());
- int min_batch = cur_string.PopWord().AsInt();
- int max_batch = cur_string.PopWord().AsInt();
-
- if (max_batch < min_batch) {
- cerr << "Error: min_batch=" << min_batch
- << ", max_batch=" << max_batch << " (incorrect order?)" << endl;
- exit(1);
- }
-
- // Construct a linked list of details needed...
- tList< tDataEntryBase<cAnalyzeGenotype> > output_list;
- tListIterator< tDataEntryBase<cAnalyzeGenotype> > output_it(output_list);
-
- // For the moment, just put everything into the output list.
- SetupGenotypeDataList();
-
- // If no args were given, load all of the stats.
- if (cur_string.CountNumWords() == 0) {
- tListIterator< tDataEntryBase<cAnalyzeGenotype> >
- genotype_data_it(genotype_data_list);
- while (genotype_data_it.Next() != NULL) {
- output_list.PushRear(genotype_data_it.Get());
- }
- }
- // Otherwise, load only those listed.
- else {
- while (cur_string.GetSize() != 0) {
- // Setup the next entry
- cString cur_entry = cur_string.PopWord();
- bool found_entry = false;
-
- // Scan the genotype data list for the current entry
- tListIterator< tDataEntryBase<cAnalyzeGenotype> >
- genotype_data_it(genotype_data_list);
-
- while (genotype_data_it.Next() != NULL) {
- if (genotype_data_it.Get()->GetName() == cur_entry) {
- output_list.PushRear(genotype_data_it.Get());
- found_entry = true;
- break;
- }
- }
-
- // If the entry was not found, give a warning.
- if (found_entry == false) {
- int best_match = 1000;
- cString best_entry;
-
- genotype_data_it.Reset();
- while (genotype_data_it.Next() != NULL) {
- const cString & test_str = genotype_data_it.Get()->GetName();
- const int test_dist = cStringUtil::EditDistance(test_str, cur_entry);
- if (test_dist < best_match) {
- best_match = test_dist;
- best_entry = test_str;
- }
- }
-
- cerr << "Warning: Format entry \"" << cur_entry
- << "\" not found. Best match is \""
- << best_entry << "\"." << endl;
- }
-
- }
- }
-
- // Setup the file...
- ofstream fp;
- fp.open(filename);
-
- // Determine the file type...
- int file_type = FILE_TYPE_TEXT;
- while (filename.Find('.') != -1) filename.Pop('.'); // Grab only extension
- if (filename == "html") file_type = FILE_TYPE_HTML;
-
- // Write out the header on the file
- if (file_type == FILE_TYPE_TEXT) {
- fp << "#filetype genotype_data" << endl;
- fp << "#format ";
- while (output_it.Next() != NULL) {
- const cString & entry_name = output_it.Get()->GetName();
- fp << entry_name << " ";
- }
- fp << endl << endl;
-
- // Give the more human-readable legend.
- fp << "# Legend:" << endl;
- fp << "# 1: Batch Name" << endl;
- int count = 1;
- while (output_it.Next() != NULL) {
- const cString & entry_desc = output_it.Get()->GetDesc();
- fp << "# " << ++count << ": " << entry_desc << endl;
- }
- fp << endl;
- } else { // if (file_type == FILE_TYPE_HTML) {
- fp << "<html>" << endl
- << "<body bgcolor=\"#FFFFFF\"" << endl
- << " text=\"#000000\"" << endl
- << " link=\"#0000AA\"" << endl
- << " alink=\"#0000FF\"" << endl
- << " vlink=\"#000044\">" << endl
- << endl
- << "<h1 align=center>Batch Index" << endl
- << endl
- << "<center>" << endl
- << "<table border=1 cellpadding=2><tr>" << endl;
-
- fp << "<th bgcolor=\"#AAAAFF\">Batch ";
- while (output_it.Next() != NULL) {
- const cString & entry_desc = output_it.Get()->GetDesc();
- fp << "<th bgcolor=\"#AAAAFF\">" << entry_desc << " ";
- }
- fp << "</tr>" << endl;
-
- }
-
- // Loop through all of the batchs...
- for (int batch_id = min_batch; batch_id <= max_batch; batch_id++) {
- cAnalyzeGenotype * genotype = batch[batch_id].List().GetFirst();
- if (genotype == NULL) continue;
- output_it.Reset();
- tDataEntryBase<cAnalyzeGenotype> * data_entry = NULL;
- const cString & batch_name = batch[batch_id].Name();
- if (file_type == FILE_TYPE_HTML) {
- fp << "<tr><th><a href=lineage." << batch_name << ".html>"
- << batch_name << "</a> ";
- }
- else {
- fp << batch_name << " ";
- }
-
- while ((data_entry = output_it.Next()) != NULL) {
- data_entry->SetTarget(genotype);
- if (file_type == FILE_TYPE_HTML) {
- fp << "<td align=center><a href=\""
- << data_entry->GetName() << "." << batch_name << ".png\">"
- << *data_entry << "</a> ";
- }
- else { // if (file_type == FILE_TYPE_TEXT) {
- fp << *data_entry << " ";
- }
- }
- if (file_type == FILE_TYPE_HTML) fp << "</tr>";
- fp << endl;
- }
-
- // If in HTML mode, we need to end the file...
- if (file_type == FILE_TYPE_HTML) {
- fp << "</table>" << endl
- << "</center>" << endl;
- }
-}
-
-void cAnalyze::CommandHistogram(cString cur_string)
-{
- if (verbose == true) cout << "Histogram batch " << cur_batch << endl;
- else cout << "Histograming..." << endl;
-
- // Load in the variables...
- cString filename("histogram.dat");
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
-
- // Construct a linked list of details needed...
- tList< tDataEntryCommand<cAnalyzeGenotype> > output_list;
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > output_it(output_list);
- LoadGenotypeDataList(cur_string, output_list);
-
- // Determine the file type...
- int file_type = FILE_TYPE_TEXT;
- cString file_extension(filename);
- while (file_extension.Find('.') != -1) file_extension.Pop('.');
- if (file_extension == "html") file_type = FILE_TYPE_HTML;
-
- // Setup the file...
- if (filename == "cout") {
- CommandHistogram_Header(cout, file_type, output_it);
- CommandHistogram_Body(cout, file_type, output_it);
- } else {
- ofstream & fp = data_file_manager.GetOFStream(filename);
- CommandHistogram_Header(fp, file_type, output_it);
- CommandHistogram_Body(fp, file_type, output_it);
- }
-
- // And clean up...
- while (output_list.GetSize() != 0) delete output_list.Pop();
-}
-
-void cAnalyze::CommandHistogram_Header(ostream & fp, int format_type,
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > & output_it)
-{
- // Write out the header on the file
- if (format_type == FILE_TYPE_TEXT) {
- fp << "#filetype histogram_data" << endl;
- fp << "#format ";
- while (output_it.Next() != NULL) {
- const cString & entry_name = output_it.Get()->GetName();
- fp << entry_name << " ";
- }
- fp << endl << endl;
-
- // Give the more human-readable legend.
- fp << "# Histograms:" << endl;
- int count = 0;
- while (output_it.Next() != NULL) {
- const cString & entry_desc = output_it.Get()->GetDesc();
- fp << "# " << ++count << ": " << entry_desc << endl;
- }
- fp << endl;
- } else { // if (format_type == FILE_TYPE_HTML) {
- fp << "<html>" << endl
- << "<body bgcolor=\"#FFFFFF\"" << endl
- << " text=\"#000000\"" << endl
- << " link=\"#0000AA\"" << endl
- << " alink=\"#0000FF\"" << endl
- << " vlink=\"#000044\">" << endl
- << endl
- << "<h1 align=center>Histograms for " << batch[cur_batch].Name()
- << "</h1>" << endl
- << endl
- << "<center>" << endl
- << "<table border=1 cellpadding=2><tr>" << endl;
-
- while (output_it.Next() != NULL) {
- const cString & entry_desc = output_it.Get()->GetDesc();
- const cString & entry_name = output_it.Get()->GetName();
- fp << "<tr><th bgcolor=\"#AAAAFF\"><a href=\"#"
- << entry_name << "\">"
- << entry_desc << "</a></tr>";
- }
- fp << "</tr></table>" << endl;
- }
-}
-
-
-void cAnalyze::CommandHistogram_Body(ostream & fp, int format_type,
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > & output_it)
-{
- output_it.Reset();
- tDataEntryCommand<cAnalyzeGenotype> * data_command = NULL;
-
- while ((data_command = output_it.Next()) != NULL) {
- if (format_type == FILE_TYPE_TEXT) {
- fp << "# --- " << data_command->GetDesc() << " ---" << endl;
- } else {
- fp << "<table cellpadding=3>" << endl
- << "<tr><th colspan=3><a name=\"" << data_command->GetName() << "\">"
- << data_command->GetDesc() << "</th></tr>" << endl;
- }
-
- tDictionary<int> count_dict;
-
- // Loop through all genotypes in this batch to collect the info we need.
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * cur_genotype;
- while ((cur_genotype = batch_it.Next()) != NULL) {
- data_command->SetTarget(cur_genotype);
- const cString cur_name(data_command->GetEntry().AsString());
- int count = 0;
- count_dict.Find(cur_name, count);
- count += cur_genotype->GetNumCPUs();
- count_dict.SetValue(cur_name, count);
- }
-
- tList<cString> name_list;
- tList<int> count_list;
- count_dict.AsLists(name_list, count_list);
-
- // Figure out the maximum count and the maximum widths...
- int max_count = 0;
- int max_name_width = 0;
- int max_count_width = 0;
- tListIterator<int> count_it(count_list);
- tListIterator<cString> name_it(name_list);
- while (count_it.Next() != NULL) {
- const cString cur_name( *(name_it.Next()) );
- const int cur_count = *(count_it.Get());
- const int name_width = cur_name.GetSize();
- const int count_width = cStringUtil::Stringf("%d", cur_count).GetSize();
- if (cur_count > max_count) max_count = cur_count;
- if (name_width > max_name_width) max_name_width = name_width;
- if (count_width > max_count_width) max_count_width = count_width;
- }
-
- // Do some final calculations now that we know the maximums...
- const int max_stars = 75 - max_name_width - max_count_width;
-
- // Now print everything out...
- count_it.Reset();
- name_it.Reset();
- while (count_it.Next() != NULL) {
- const cString cur_name( *(name_it.Next()) );
- const int cur_count = *(count_it.Get());
- if (cur_count == 0) continue;
- int num_stars = (cur_count * max_stars) / max_count;
-
- if (format_type == FILE_TYPE_TEXT) {
- fp << setw(max_name_width) << cur_name << " "
- << setw(max_count_width) << cur_count << " ";
- for (int i = 0; i < num_stars; i++) { fp << '#'; }
- fp << endl;
- } else { // FILE_TYPE_HTML
- fp << "<tr><td>" << cur_name
- << "<td>" << cur_count
- << "<td>";
- for (int i = 0; i < num_stars; i++) { fp << '#'; }
- fp << "</tr>" << endl;
- }
- }
-
- if (format_type == FILE_TYPE_TEXT) {
- // Skip a line between histograms...
- fp << endl;
- } else {
- fp << "</table><br><br>" << endl << endl;;
- }
- }
-
- // If in HTML mode, we need to end the file...
- if (format_type == FILE_TYPE_HTML) {
- fp << "</table>" << endl
- << "</center>" << endl;
- }
-}
-
-
-///// Population Analysis Commands ////
-
-void cAnalyze::CommandPrintPhenotypes(cString cur_string)
-{
- if (verbose == true) cout << "Printing phenotypes in batch "
- << cur_batch << endl;
- else cout << "Printing phenotypes..." << endl;
-
- // Load in the variables...
- cString filename("phenotype.dat");
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
-
- // Make sure we have at least one genotype...
- if (batch[cur_batch].List().GetSize() == 0) return;
-
- // Setup the phenotype categories...
- const int num_tasks = batch[cur_batch].List().GetFirst()->GetNumTasks();
- const int num_phenotypes = 1 << (num_tasks + 1);
- tArray<int> phenotype_counts(num_phenotypes);
- tArray<int> genotype_counts(num_phenotypes);
- tArray<double> total_length(num_phenotypes);
- tArray<double> total_gest(num_phenotypes);
-
- phenotype_counts.SetAll(0);
- genotype_counts.SetAll(0);
- total_length.SetAll(0.0);
- total_gest.SetAll(0.0);
-
- // Loop through all of the genotypes in this batch...
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- int phen_id = 0;
- if (genotype->GetViable() == true) phen_id++;
- for (int i = 0; i < num_tasks; i++) {
- if (genotype->GetTaskCount(i) > 0) phen_id += 1 << (i+1);
- }
- phenotype_counts[phen_id] += genotype->GetNumCPUs();
- genotype_counts[phen_id]++;
- total_length[phen_id] += genotype->GetNumCPUs() * genotype->GetLength();
- total_gest[phen_id] += genotype->GetNumCPUs() * genotype->GetGestTime();
- }
-
- // Print out the results...
-
- ofstream & fp = data_file_manager.GetOFStream(filename);
-
- fp << "# 1: Number of organisms of this phenotype" << endl
- << "# 2: Number of genotypes of this phenotye" << endl
- << "# 3: Average Genome Length" << endl
- << "# 4: Average Gestation Time" << endl
- << "# 5: Viability of Phenotype" << endl
- << "# 6+: Tasks performed in this phenotype" << endl
- << endl;
-
- // @CAO Lets do this inefficiently for the moment, but print out the
- // phenotypes in order.
-
- while (true) {
- // Find the next phenotype to print...
- int max_count = phenotype_counts[0];
- int max_position = 0;
- for (int i = 0; i < num_phenotypes; i++) {
- if (phenotype_counts[i] > max_count) {
- max_count = phenotype_counts[i];
- max_position = i;
- }
- }
-
- if (max_count == 0) break; // we're done!
-
- fp << phenotype_counts[max_position] << " "
- << genotype_counts[max_position] << " "
- << total_length[max_position] / phenotype_counts[max_position] << " "
- << total_gest[max_position] / phenotype_counts[max_position] << " "
- << (max_position & 1) << " ";
- for (int i = 1; i <= num_tasks; i++) {
- if ((max_position >> i) & 1 > 0) fp << "1 ";
- else fp << "0 ";
- }
- fp << endl;
- phenotype_counts[max_position] = 0;
- }
-
- fp.close();
-}
-
-
-// Print various diversity metrics from the current batch of genotypes...
-void cAnalyze::CommandPrintDiversity(cString cur_string)
-{
- if (verbose == true) cout << "Printing diversity data for batch "
- << cur_batch << endl;
- else cout << "Printing diversity data..." << endl;
-
- // Load in the variables...
- cString filename("diversity.dat");
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
-
- // Make sure we have at least one genotype...
- if (batch[cur_batch].List().GetSize() == 0) return;
-
- // Setup the task categories...
- const int num_tasks = batch[cur_batch].List().GetFirst()->GetNumTasks();
- tArray<int> task_count(num_tasks);
- tArray<int> task_gen_count(num_tasks);
- tArray<double> task_gen_dist(num_tasks);
- tArray<double> task_site_entropy(num_tasks);
- task_count.SetAll(0);
- task_gen_count.SetAll(0);
-
- // We must determine the average hamming distance between genotypes in
- // this batch that perform each task. Levenstein distance would be ideal,
- // but takes a while, so we'll do it this way first. For the calculations,
- // we need to know home many times each instruction appears at each
- // position for each genotype collection that performs a particular task.
- const int num_insts = inst_set.GetSize();
- const int max_length = BatchUtil_GetMaxLength();
- tMatrix<int> inst_freq(max_length, num_insts+1);
-
- for (int task_id = 0; task_id < num_tasks; task_id++) {
- inst_freq.SetAll(0);
-
- // Loop through all genotypes, singling out those that do current task...
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- if (genotype->GetTaskCount(task_id) == 0) continue;
-
- const cGenome & genome = genotype->GetGenome();
- const int num_cpus = genotype->GetNumCPUs();
- task_count[task_id] += num_cpus;
- task_gen_count[task_id]++;
- for (int i = 0; i < genotype->GetLength(); i++) {
- inst_freq( i, genome[i].GetOp() ) += num_cpus;
- }
- for (int i = genotype->GetLength(); i < max_length; i++) {
- inst_freq(i, num_insts) += num_cpus; // Entry for "past genome end"
- }
- }
-
- // Analyze the data for this entry...
- const int cur_count = task_count[task_id];
- const int total_pairs = cur_count * (cur_count - 1) / 2;
- int total_dist = 0;
- double total_ent = 0;
- for (int pos = 0; pos < max_length; pos++) {
- // Calculate distance component...
- for (int inst1 = 0; inst1 < num_insts; inst1++) {
- if (inst_freq(pos, inst1) == 0) continue;
- for (int inst2 = inst1+1; inst2 <= num_insts; inst2++) {
- total_dist += inst_freq(pos, inst1) * inst_freq(pos, inst2);
- }
- }
-
- // Calculate entropy component...
- for (int i = 0; i <= num_insts; i++) {
- const int cur_freq = inst_freq(pos, i);
- if (cur_freq == 0) continue;
- const double p = ((double) cur_freq) / (double) cur_count;
- total_ent -= p * log(p);
- }
- }
-
- task_gen_dist[task_id] = ((double) total_dist) / (double) total_pairs;
- task_site_entropy[task_id] = total_ent;
- }
-
- // Print out the results...
- cDataFile & df = data_file_manager.Get(filename);
-
- for (int i = 0; i < num_tasks; i++) {
- df.Write(i, "# 1: Task ID");
- df.Write(task_count[i], "# 2: Number of organisms performing task");
- df.Write(task_gen_count[i], "# 3: Number of genotypes performing task");
- df.Write(task_gen_dist[i], "# 4: Average distance between genotypes performing task");
- df.Write(task_site_entropy[i], "# 5: Total per-site entropy of genotypes performing task");
- df.Endl();
- }
-}
-
-
-void cAnalyze::PhyloCommunityComplexity(cString cur_string)
-{
-
- /////////////////////////////////////////////////////////////////////////
- // Calculate the mutual information between all genotypes and environment
- /////////////////////////////////////////////////////////////////////////
-
- cout << "Analyze biocomplexity of current population about environment ...\n";
-
- // Get the number of genotypes that are gonna be analyzed.
- int max_genotypes = cur_string.PopWord().AsInt();
-
- // Get update
- int update = cur_string.PopWord().AsInt();
-
- // Get the directory
- cString dir = cur_string.PopWord();
- cString defaultDir = "community_cpx/";
- cString directory = PopDirectory(dir, defaultDir);
-
- // Get the file name that saves the result
- cString filename = cur_string.PopWord();
- if (filename.IsEmpty()) {
- filename = "community.complexity.dat";
- }
-
- filename.Set("%s%s", directory(), filename.GetData());
- ofstream cpx_fp(filename());
-
- cpx_fp << "# Legend:" << endl;
- cpx_fp << "# 1: Genotype ID" << endl;
- cpx_fp << "# 2: Entropy given Known Genotypes" << endl;
- cpx_fp << "# 3: Entropy given Both Known Genotypes and Env" << endl;
- cpx_fp << "# 4: New Information about Environment" << endl;
- cpx_fp << "# 5: Total Complexity" << endl;
- cpx_fp << endl;
-
- /////////////////////////////////////////////////////////////////////////////////
- // Loop through all of the genotypes in all batches and build id vs. genotype map
-
- map<int, cAnalyzeGenotype *> genotype_database;
- for (int i = 0; i < MAX_BATCHES; ++ i) {
- tListIterator<cAnalyzeGenotype> batch_it(batch[i].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- genotype_database.insert(make_pair(genotype->GetID(), genotype));
- }
- }
-
- ////////////////////////////////////////////////
- // Check if all the genotypes having same length
-
- int length_genome;
- if (genotype_database.size() > 0) {
- length_genome = genotype_database.begin()->second->GetLength();
- }
- map<int, cAnalyzeGenotype *>::iterator gen_iterator = genotype_database.begin();
- for (; gen_iterator != genotype_database.end(); ++ gen_iterator) {
- if (gen_iterator->second->GetLength() != length_genome) {
- cerr << "Genotype " << gen_iterator->first << " has different genome length." << endl;
- exit(1);
- }
- }
-
- ///////////////////////
- // Backup test CPU data
- bool backupUsage = m_world->GetTestCPU().UseResources();
- tArray<double> backupResources(m_world->GetTestCPU().GetResources());
- m_world->GetTestCPU().UseResources() = true;
- FillResources(update);
-
- ///////////////////////////////////////////////////////////////////////
- // Choose the first n most abundant genotypes and put them in community
-
- vector<cAnalyzeGenotype *> community;
- cAnalyzeGenotype * genotype = NULL;
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
-
- while (((genotype = batch_it.Next()) != NULL) && (community.size() < max_genotypes)) {
- community.push_back(genotype);
- }
-
- ///////////////////////////
- // Measure hamming distance
-
- int size_community = community.size();
- if (size_community == 0) {
- cerr << "There is no genotype in this community." << endl;
- cpx_fp.close();
- exit(1);
- }
- typedef pair<int,int> gen_pair;
- map<gen_pair, int> hamming_dist;
-
- for (int i = 0; i< size_community; ++ i) {
- for (int j = i+1; j < size_community; ++ j) {
- int dist = cGenomeUtil::FindHammingDistance(community[i]->GetGenome(),
- community[j]->GetGenome());
- int id1 = community[i]->GetID();
- int id2 = community[j]->GetID();
-
- hamming_dist.insert(make_pair(gen_pair(id1, id2), dist));
- hamming_dist.insert(make_pair(gen_pair(id2, id1), dist));
- }
- }
-
- //////////////////////////////////
- // Get Most Recent Common Ancestor
-
- map<gen_pair, cAnalyzeGenotype *> mrca;
- map<gen_pair, int> raw_dist;
- for (int i = 0; i< size_community; ++ i) {
- for (int j = i+1; j < size_community; ++ j) {
-
- cAnalyzeGenotype * lineage1_genotype = community[i];
- cAnalyzeGenotype * lineage2_genotype = community[j];
- int total_dist = 0;
-
- while (lineage1_genotype->GetID() != lineage2_genotype->GetID()) {
- if (lineage1_genotype->GetID() > lineage2_genotype->GetID()) {
- int parent_id = lineage1_genotype->GetParentID();
- cAnalyzeGenotype * parent = genotype_database.find(parent_id)->second;
-
- total_dist += cGenomeUtil::FindHammingDistance(lineage1_genotype->GetGenome(),
- parent->GetGenome());
- lineage1_genotype = parent;
- } else {
- int parent_id = lineage2_genotype->GetParentID();
- cAnalyzeGenotype * parent = genotype_database.find(parent_id)->second;
- total_dist += cGenomeUtil::FindHammingDistance(lineage2_genotype->GetGenome(),
- parent->GetGenome());
-
- lineage2_genotype = parent;
- }
- }
-
- int id1 = community[i]->GetID();
- int id2 = community[j]->GetID();
- mrca.insert(make_pair(gen_pair(id1, id2), lineage1_genotype));
- mrca.insert(make_pair(gen_pair(id2, id1), lineage1_genotype));
- raw_dist.insert(make_pair(gen_pair(id1, id2), total_dist));
- raw_dist.insert(make_pair(gen_pair(id2, id1), total_dist));
- }
- }
-
- /*
- ////////////////////////////////////////////////////////////////////////////////////////////
- // Sort the genotype that is next genotype is the most closest one to all previous genotypes
-
- vector<cAnalyzeGenotype *> sorted_community;
- vector<cAnalyzeGenotype *> left_genotypes = community;
-
- // Put the first genotype in left to sorted.
- sorted_community.push_back(*left_genotypes.begin());
- left_genotypes.erase(left_genotypes.begin());
-
- while (left_genotypes.size() > 0) {
- int min_total_hamming = size_community * length_genome;
- int index_next;
-
- for (int next = 0; next < left_genotypes.size(); ++ next) {
- int total_hamming = 0;
- int id1 = left_genotypes[next]->GetID();
-
- for (int given = 0; given < sorted_community.size(); ++ given) {
- int id2 = sorted_community[given]->GetID();
- total_hamming += hamming_dist.find(gen_pair(id1, id2))->second;
- }
-
- if (total_hamming < min_total_hamming) {
- min_total_hamming = total_hamming;
- index_next = next;
- }
- }
-
- sorted_community.push_back(left_genotypes[index_next]);
- left_genotypes.erase(left_genotypes.begin() + index_next);
- }
-
- */
-
- vector<cAnalyzeGenotype *> sorted_community = community;
- /////////////////////////////////////////////
- // Loop through genotypes in sorted community
-
- double complexity = 0.0;
- vector<cAnalyzeGenotype *> given_genotypes;
- int num_insts = inst_set.GetSize();
-
- for (int i = 0; i < size_community; ++ i) {
- genotype = sorted_community[i];
-
- // Skip the dead organisms
- genotype->Recalculate();
- if (genotype->GetFitness() == 0) {
- continue;
- }
-
- vector<double> one_line_prob(num_insts, 0.0);
- vector< vector<double> > prob(length_genome, one_line_prob);
-
- cout << endl << genotype->GetID() << endl;
-
- /*if (given_genotypes.size() >= 2) {
-
- ///////////////////////////////////////////////////////////////////
- // Look for two given genotypes that has minimun depth dist with it
-
- cAnalyzeGenotype * min_depth_gen = given_genotypes[0];
- cAnalyzeGenotype * tmrca = mrca.find(gen_pair(genotype->GetID(),
- given_genotypes[0]->GetID()))->second;
- int min_depth_dist = genotype->GetDepth() + given_genotypes[0]->GetDepth() - 2 * tmrca->GetDepth();
-
- cAnalyzeGenotype * second_min_gen = given_genotypes[1];
- tmrca = mrca.find(gen_pair(genotype->GetID(), given_genotypes[1]->GetID()))->second;
- int second_min_depth = genotype->GetDepth() + given_genotypes[1]->GetDepth() - 2 * tmrca->GetDepth();
-
- for (int i = 2; i < given_genotypes.size(); ++ i) {
- cAnalyzeGenotype * given_genotype = given_genotypes[i];
- cAnalyzeGenotype * tmrca = mrca.find(gen_pair(genotype->GetID(),
- given_genotype->GetID()))->second;
- int dist = genotype->GetDepth() + given_genotype->GetDepth() - 2 * tmrca->GetDepth();
-
- if (dist < min_depth_dist) {
- second_min_depth = min_depth_dist;
- second_min_gen = min_depth_gen;
- min_depth_dist = dist;
- min_depth_gen = given_genotype;
- } else if (dist >= min_depth_dist && dist < second_min_depth) {
- second_min_depth = dist;
- second_min_gen = given_genotype;
- }
- }
-
- const cGenome & given_genome1 = min_depth_gen->GetGenome();
- const cGenome & given_genome2 = second_min_gen->GetGenome();
- for (int line = 0; line < length_genome; ++ line) {
- int given_inst = given_genome1[line].GetOp();
- prob[line][given_inst] += pow(1 - 1.0/length_genome, min_depth_dist);
- given_inst = given_genome2[line].GetOp();
- prob[line][given_inst] += pow(1 - 1.0/length_genome, min_depth_dist);
- }
-
- cpx_fp << genotype->GetID() << " " << min_depth_dist << " " << second_min_depth
- << " " << raw_dist.find(gen_pair(genotype->GetID(), min_depth_gen->GetID()))->second
- << " " << raw_dist.find(gen_pair(genotype->GetID(), second_min_gen->GetID()))->second
- << " ";
-
-
- } else if (given_genotypes.size() == 1) {
- //////////////////////////////////////////////////////
- // Calculate the probability of each inst at each line
- cAnalyzeGenotype * tmrca = mrca.find(gen_pair(genotype->GetID(),
- given_genotypes[0]->GetID()))->second;
- int dist = genotype->GetDepth() + given_genotypes[0]->GetDepth() - 2 * tmrca->GetDepth();
- const cGenome & given_genome = given_genotypes[0]->GetGenome();
-
- for (int line = 0; line < length_genome; ++ line) {
- int given_inst = given_genome[line].GetOp();
- prob[line][given_inst] += pow(1 - 1.0/length_genome, dist);
- }
-
- cpx_fp << genotype->GetID() << " " << dist << " "
- << raw_dist.find(gen_pair(genotype->GetID(), given_genotypes[0]->GetID()))->second << " ";
- } else {
- cpx_fp << genotype->GetID() << " ";
- }*/
-
- if (given_genotypes.size() >= 1) {
- //////////////////////////////////////////////////
- // Look for a genotype that is closest to this one
-
- cAnalyzeGenotype * min_depth_gen = given_genotypes[0];
- cAnalyzeGenotype * tmrca = mrca.find(gen_pair(genotype->GetID(),
- given_genotypes[0]->GetID()))->second;
- int min_depth_dist = genotype->GetDepth() + given_genotypes[0]->GetDepth() - 2 * tmrca->GetDepth();
-
- for (int i = 1; i < given_genotypes.size() ; ++ i) {
- cAnalyzeGenotype * given_genotype = given_genotypes[i];
- cAnalyzeGenotype * tmrca = mrca.find(gen_pair(genotype->GetID(),
- given_genotype->GetID()))->second;
- int dist = genotype->GetDepth() + given_genotype->GetDepth() - 2 * tmrca->GetDepth();
-
- if (dist < min_depth_dist) {
- min_depth_dist = dist;
- min_depth_gen = given_genotype;
- }
- }
-
- const cGenome & given_genome = min_depth_gen->GetGenome();
- const cGenome & base_genome = genotype->GetGenome();
- cGenome mod_genome(base_genome);
- for (int line = 0; line < length_genome; ++ line) {
- int given_inst = given_genome[line].GetOp();
- mod_genome = base_genome;
- mod_genome[line].SetOp(given_inst);
- cAnalyzeGenotype test_genotype(m_world, mod_genome, inst_set);
- test_genotype.Recalculate();
-
- // Only when given inst make the genotype alive
- if (test_genotype.GetFitness() > 0) {
- prob[line][given_inst] += pow(1 - 1.0/length_genome, min_depth_dist);
- }
- }
-
- cpx_fp << genotype->GetID() << " " << min_depth_dist << " "
- << raw_dist.find(gen_pair(genotype->GetID(), min_depth_gen->GetID()))->second << " "
- << hamming_dist.find(gen_pair(genotype->GetID(), min_depth_gen->GetID()))->second << " ";
- } else {
- cpx_fp << genotype->GetID() << " ";
- }
-
- ///////////////////////////////////////////////////////////////////
- // Point mutation at all lines of code to look for neutral mutation
- // and the mutations that can make organism alive
- cout << "Test point mutation." << endl;
- vector<bool> one_line_neutral(num_insts, false);
- vector< vector<bool> > neutral_mut(length_genome, one_line_neutral);
- vector< vector<bool> > alive_mut(length_genome, one_line_neutral);
-// if (verbose == true) {
-// PrintTestCPUResources("");
-// }
-
- genotype->Recalculate();
- double base_fitness = genotype->GetFitness();
- cout << base_fitness << endl;
- const cGenome & base_genome = genotype->GetGenome();
- cGenome mod_genome(base_genome);
-
- for (int line = 0; line < length_genome; ++ line) {
- int cur_inst = base_genome[line].GetOp();
-
- for (int mod_inst = 0; mod_inst < num_insts; ++ mod_inst) {
- mod_genome[line].SetOp(mod_inst);
- cAnalyzeGenotype test_genotype(m_world, mod_genome, inst_set);
- test_genotype.Recalculate();
- if (test_genotype.GetFitness() >= base_fitness) {
- neutral_mut[line][mod_inst] = true;
- }
- if (test_genotype.GetFitness() > 0) {
- alive_mut[line][mod_inst] = true;
- }
- }
-
- mod_genome[line].SetOp(cur_inst);
- }
-
- /////////////////////////////////////////
- // Normalize the probability at each line
- vector< vector<double> > prob_before_env(length_genome, one_line_prob);
-
- for (int line = 0; line < length_genome; ++ line) {
- double cur_total_prob = 0.0;
- int num_alive = 0;
- for (int inst = 0; inst < num_insts; ++ inst) {
- if (alive_mut[line][inst] == true) {
- cur_total_prob += prob[line][inst];
- num_alive ++;
- }
- }
- if (cur_total_prob > 1) {
- cout << "Total probability at " << line << " is greater than 0." << endl;
- exit(1);
- }
- double left_prob = 1 - cur_total_prob;
-
- for (int inst = 0; inst < num_insts; ++ inst) {
- if (alive_mut[line][inst] == true) {
- prob_before_env[line][inst] = prob[line][inst] + left_prob / num_alive;
- } else {
- prob_before_env[line][inst] = 0;
- }
- }
-
- }
-
- /////////////////////////////////
- // Calculate entropy of each line
- vector<double> entropy(length_genome, 0.0);
- for (int line = 0; line < length_genome; ++ line) {
- double sum = 0;
- for (int inst = 0; inst < num_insts; ++ inst) {
- sum += prob_before_env[line][inst];
- if (prob_before_env[line][inst] > 0) {
- entropy[line] -= prob_before_env[line][inst] * log(prob_before_env[line][inst]) / log(num_insts*1.0);
- }
- }
- if (sum > 1.001 || sum < 0.999) {
- cout << "Sum of probability is not 1 at line " << line << endl;
- exit(1);
- }
- }
-
-
- /////////////////////////////////////////////////////
- // Redistribute the probability of insts at each line
- vector< vector<double> > prob_given_env(length_genome, one_line_prob);
-
- for (int line = 0; line < length_genome; ++ line) {
- double total_prob = 0.0;
- int num_neutral = 0;
- for (int inst = 0; inst < num_insts; ++ inst) {
- if (neutral_mut[line][inst] == true) {
- num_neutral ++;
- total_prob += prob[line][inst];
- }
- }
-
- double left = 1 - total_prob;
-
- for (int inst = 0; inst < num_insts; ++ inst) {
- if (neutral_mut[line][inst] == true) {
- prob_given_env[line][inst] = prob[line][inst] + left / num_neutral;
- } else {
- prob_given_env[line][inst] = 0.0;
- }
- }
-
- }
-
- ////////////////////////////////////////////////
- // Calculate the entropy given environment
-
- vector<double> entropy_given_env(length_genome, 0.0);
- for (int line = 0; line < length_genome; ++ line) {
- double sum = 0;
- for (int inst = 0; inst < num_insts; ++ inst) {
- sum += prob_given_env[line][inst];
- if (prob_given_env[line][inst] > 0) {
- entropy_given_env[line] -= prob_given_env[line][inst] * log(prob_given_env[line][inst]) /
- log(num_insts*1.0);
- }
- }
- if (sum > 1.001 || sum < 0.999) {
- cout << "Sum of probability is not 1 at line " << line << " " << sum << endl;
- exit(1);
- }
- }
-
-
- ///////////////////////////////////////////////////////////////////////////
- // Calculate the information between genotype and env given other genotypes
- double information = 0.0;
- double entropy_before = 0.0;
- double entropy_after = 0.0;
- for (int line = 0; line < length_genome; ++ line) {
- entropy_before += entropy[line];
- entropy_after += entropy_given_env[line];
-
- if (entropy[line] >= entropy_given_env[line]) {
- information += entropy[line] - entropy_given_env[line];
- } else { // Negative information is because given condition is not related with this genotype ...
-
- // Count the number of insts that can make genotype alive
- int num_inst_alive = 0;
- for (int inst = 0; inst < num_insts; ++ inst) {
- if (alive_mut[line][inst] == true) {
- num_inst_alive ++;
- }
- }
-
- double entropy_before = - log(1.0/num_inst_alive) / log(num_insts*1.0);
- information += entropy_before - entropy_given_env[line];
- if (information < 0) {
- cout << "Negative information at site " << line << endl;
- exit(1);
- }
- }
-
- }
- complexity += information;
-
- cpx_fp << entropy_before << " " << entropy_after << " "
- << information << " " << complexity << " ";
-
- genotype->PrintTasks(cpx_fp, 0, -1);
- cpx_fp << endl;
-
- /////////////////////////////////////////////////////////////
- // This genotype becomes the given condition of next genotype
-
- given_genotypes.push_back(genotype);
-
- }
-
- // Set the test CPU back to the state it was
- m_world->GetTestCPU().UseResources() = backupUsage;
- m_world->GetTestCPU().SetupResourceArray(backupResources);
-
- cpx_fp.close();
- return;
-
-}
-
-void cAnalyze::AnalyzeCommunityComplexity(cString cur_string)
-{
- /////////////////////////////////////////////////////////////////////
- // Calculate the mutual information between community and environment
- /////////////////////////////////////////////////////////////////////
-
- cout << "Analyze community complexity of current population about environment with Charles method ...\n";
-
- // Get the number of genotypes that are gonna be analyzed.
- int max_genotypes = cur_string.PopWord().AsInt(); // If it is 0, we sample
- //two genotypes for each task.
-
- // Get update
- int update = cur_string.PopWord().AsInt();
-
- // Get the directory
- cString dir = cur_string.PopWord();
- cString defaultDir = "community_cpx/";
- cString directory = PopDirectory(dir, defaultDir);
-
- // Get the file name that saves the result
- cString filename = cur_string.PopWord();
- if (filename.IsEmpty()) {
- filename = "community.complexity.dat";
- }
-
- filename.Set("%s%s", directory(), filename.GetData());
- ofstream cpx_fp(filename());
-
- cpx_fp << "# Legend:" << endl;
- cpx_fp << "# 1: Genotype ID" << endl;
- cpx_fp << "# 2: Entropy given Known Genotypes" << endl;
- cpx_fp << "# 3: Entropy given Both Known Genotypes and Env" << endl;
- cpx_fp << "# 4: New Information about Environment" << endl;
- cpx_fp << "# 5: Total Complexity" << endl;
- cpx_fp << "# 6: Hamming Distance to Closest Given Genotype" << endl;
- cpx_fp << "# 7: Total Hamming Distance to Closest Neighbor" << endl;
- cpx_fp << "# 8: Number of Organisms" << endl;
- cpx_fp << "# 9: Total Number of Organisms" << endl;
- cpx_fp << "# 10 - : Tasks Implemented" << endl;
- cpx_fp << endl;
-
- ///////////////////////
- // Backup test CPU data
- bool backupUsage = m_world->GetTestCPU().UseResources();
- tArray<double> backupResources(m_world->GetTestCPU().GetResources());
- m_world->GetTestCPU().UseResources() = true;
- FillResources(update);
-
-
- vector<cAnalyzeGenotype *> community;
- cAnalyzeGenotype * genotype = NULL;
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
-
-
- if (max_genotypes > 0) {
-
- ///////////////////////////////////////////////////////////////////////
- // Choose the first n most abundant genotypes and put them in community
-
- while (((genotype = batch_it.Next()) != NULL) && (community.size() < max_genotypes)) {
- community.push_back(genotype);
- }
- } else if (max_genotypes == 0) {
-
- /////////////////////////////////////
- // Choose two genotypes for each task
-
- genotype = batch_it.Next();
- if (genotype == NULL) {
- cpx_fp.close();
- return;
- }
- genotype->Recalculate();
- int num_tasks = genotype->GetNumTasks();
- vector< vector<cAnalyzeGenotype *> > genotype_class(num_tasks);
- do {
- for (int task_id = 0; task_id < num_tasks; ++ task_id) {
- int count = genotype->GetTaskCount(task_id);
- if (count > 0) {
- genotype_class[task_id].push_back(genotype);
- }
- }
- } while ((genotype = batch_it.Next()) != NULL);
-
- cRandom random;
- for (int task_id = 0; task_id < num_tasks; ++ task_id) {
- int num_genotype = genotype_class[task_id].size();
- if (num_genotype > 0) {
- int index = random.GetUInt(num_genotype);
- community.push_back(genotype_class[task_id][index]);
- index = random.GetUInt(num_genotype);
- community.push_back(genotype_class[task_id][index]);
- } else {
- // Pick up a class that is not empty
- int class_id = random.GetUInt(num_tasks);
- while (genotype_class[class_id].size() == 0) {
- class_id ++;
- if (class_id >= num_tasks) {
- class_id = 0;
- }
- }
- int num_genotype = genotype_class[class_id].size();
- int index = random.GetUInt(num_genotype);
- community.push_back(genotype_class[task_id][index]);
- index = random.GetUInt(num_genotype);
- community.push_back(genotype_class[task_id][index]);
- }
- }
-
- }
-
- ////////////////////////////////////////////////////
- // Test point mutation of each genotype in community
-
- int num_insts = inst_set.GetSize();
- map<int, tMatrix<double> > point_mut;
- int size_community = community.size();
- int length_genome;
- if (size_community > 1) {
- length_genome = community[0]->GetLength();
- }
-
- for (int i = 0; i < size_community; ++ i) {
- genotype = community[i];
-
- ///////////////////////////////////////////////////////////////////
- // Point mutation at all lines of code to look for neutral mutation
- cout << "Test point mutation for genotype " << genotype->GetID() << endl;
- tMatrix<double> prob(length_genome, num_insts);
-// if (verbose == true) {
-// PrintTestCPUResources("");
-// }
-
- genotype->Recalculate();
- double base_fitness = genotype->GetFitness();
- const cGenome & base_genome = genotype->GetGenome();
- cGenome mod_genome(base_genome);
-
- for (int line = 0; line < length_genome; ++ line) {
- int cur_inst = base_genome[line].GetOp();
- int num_neutral = 0;
-
- for (int mod_inst = 0; mod_inst < num_insts; ++ mod_inst) {
- mod_genome[line].SetOp(mod_inst);
- cAnalyzeGenotype test_genotype(m_world, mod_genome, inst_set);
- test_genotype.Recalculate();
- if (test_genotype.GetFitness() >= base_fitness) {
- prob[line][mod_inst] = 1.0;
- num_neutral ++;
- } else {
- prob[line][mod_inst] = 0.0;
- }
- }
-
- for (int mod_inst = 0; mod_inst < num_insts; ++ mod_inst) {
- prob[line][mod_inst] /= num_neutral;
- }
-
-
- mod_genome[line].SetOp(cur_inst);
- }
-
- point_mut.insert(make_pair(genotype->GetID(), prob));
- }
-
- //////////////////////////////////////
- // Loop through genotypes in community
-
- double complexity = 0.0;
- int total_dist = 0;
- int total_cpus = 0;
- vector<cAnalyzeGenotype *> given_genotypes;
-
- ////////////////////////////////////////
- // New information in the first gentoype
- genotype = community[0];
- double oo_initial_entropy = length_genome;
- double oo_conditional_entropy = 0.0;
- tMatrix<double> this_prob = point_mut.find(genotype->GetID())->second;
-
- for (int line = 0; line < length_genome; ++ line) {
- double oneline_entropy = 0.0;
- for (int inst = 0; inst < num_insts; ++ inst) {
- if (this_prob[line][inst] > 0) {
- oneline_entropy -= this_prob[line][inst] * (log(this_prob[line][inst]) /
- log(1.0*num_insts));
- }
- }
- oo_conditional_entropy += oneline_entropy;
- }
-
- double new_info = oo_initial_entropy - oo_conditional_entropy;
- complexity += new_info;
- given_genotypes.push_back(genotype);
-
- cpx_fp << genotype->GetID() << " "
- << oo_initial_entropy << " "
- << oo_conditional_entropy << " "
- << new_info << " "
- << complexity << " "
- << "0 0" << " ";
- int num_cpus = genotype->GetNumCPUs();
- total_cpus += num_cpus;
- cpx_fp << num_cpus << " " << total_cpus << " ";
- genotype->Recalculate();
- genotype->PrintTasks(cpx_fp, 0, -1);
- cpx_fp << endl;
-
-
- //////////////////////////////////////////////////////
- // New information in other genotypes in community ...
- for (int i = 1; i < size_community; ++ i) {
- genotype = community[i];
- if (genotype->GetLength() != length_genome) {
- cerr << "Genotypes in the community do not same genome length.\n";
- cpx_fp.close();
- exit(1);
- }
-
- // Skip the dead organisms
- genotype->Recalculate();
- cout << genotype->GetID() << " " << genotype->GetFitness() << endl;
- if (genotype->GetFitness() == 0) {
- continue;
- }
-
- double min_new_info = length_genome;
- double oo_initial_entropy, oo_conditional_entropy;
- cAnalyzeGenotype * used_genotype;
- tMatrix<double> this_prob = point_mut.find(genotype->GetID())->second;
-
- // For any given genotype, calculate the new information in genotype
- for (int j = 0; j < given_genotypes.size(); ++ j) {
-
- tMatrix<double> given_prob = point_mut.find(given_genotypes[j]->GetID())->second;
- double new_info = 0.0;
- double total_initial_entropy = 0.0;
- double total_conditional_entropy = 0.0;
-
- for (int line = 0; line < length_genome; ++ line) {
-
- // H(genotype|known_genotype)
- double prob_overlap = 0;
- for (int inst = 0; inst < num_insts; ++ inst) {
- if (this_prob[line][inst] < given_prob[line][inst]) {
- prob_overlap += this_prob[line][inst];
- } else {
- prob_overlap += given_prob[line][inst];
- }
- }
-
- double given_site_entropy = 0.0;
- for (int inst = 0; inst < num_insts; ++ inst) {
- if (given_prob[line][inst] > 0) {
- given_site_entropy -= given_prob[line][inst] * (log(given_prob[line][inst]) /
- log(1.0*num_insts));
- }
- }
-
-
- double entropy_overlap = 0.0;
- if (prob_overlap > 0 && (1 - prob_overlap > 0)) {
- entropy_overlap = (- prob_overlap * log(prob_overlap)
- - (1-prob_overlap) * log(1 - prob_overlap)) / log(1.0*num_insts);
- } else {
- entropy_overlap = 0;
- }
-
- double initial_entropy = prob_overlap * given_site_entropy
- + (1 - prob_overlap) * 1 + entropy_overlap;
- total_initial_entropy += initial_entropy;
-
- // H(genotype|E, known_genotype) = H(genotype|Env)
- double conditional_entropy = 0.0;
- for (int inst = 0; inst < num_insts; ++ inst) {
- if (this_prob[line][inst] > 0) {
- conditional_entropy -= this_prob[line][inst] * (log(this_prob[line][inst]) /
- log(1.0*num_insts));
- }
- }
- total_conditional_entropy += conditional_entropy;
-
- if (conditional_entropy > initial_entropy + 0.00001) {
- cerr << "Negative Information.\n";
- cout << line << endl;
- for (int inst = 0; inst < num_insts; ++ inst) {
- cout << this_prob[line][inst] << " ";
- }
- cout << endl;
- for (int inst = 0; inst < num_insts; ++ inst) {
- cout << given_prob[line][inst] << " ";
- }
- cout << endl;
-
- exit(1);
- }
-
- new_info += initial_entropy - conditional_entropy;
- }
-
- if (new_info < min_new_info) {
- min_new_info = new_info;
- oo_initial_entropy = total_initial_entropy;
- oo_conditional_entropy = total_conditional_entropy;
- used_genotype = given_genotypes[j];
- cout << " " << "New closest genotype " << used_genotype->GetID()
- << " " << new_info << endl;;
- }
-
- }
- complexity += min_new_info;
- cpx_fp << genotype->GetID() << " "
- << oo_initial_entropy << " "
- << oo_conditional_entropy << " "
- << min_new_info << " " << complexity << " ";
-
- int hamm_dist = cGenomeUtil::FindHammingDistance(genotype->GetGenome(),
- used_genotype->GetGenome());
- total_dist += hamm_dist;
- cpx_fp << hamm_dist << " " << total_dist << " ";
-
- int num_cpus = genotype->GetNumCPUs();
- total_cpus += num_cpus;
- cpx_fp << num_cpus << " " << total_cpus << " ";
-
-
- genotype->PrintTasks(cpx_fp, 0, -1);
- cpx_fp << endl;
- given_genotypes.push_back(genotype);
- }
-
-
- // Set the test CPU back to the state it was
- m_world->GetTestCPU().UseResources() = backupUsage;
- m_world->GetTestCPU().SetupResourceArray(backupResources);
-
- cpx_fp.close();
- return;
-
-
-
-}
-
-void cAnalyze::CommandLandscape(cString cur_string)
-{
- if (verbose == true) cout << "Landscaping batch " << cur_batch << endl;
- else cout << "Landscapping..." << endl;
-
- // Load in the variables...
- cString filename;
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
- int dist = 1;
- if (cur_string.GetSize() != 0) dist = cur_string.PopWord().AsInt();
- int num_test = 0;
- if (cur_string.GetSize() != 0) num_test=cur_string.PopWord().AsInt();
-
- // If we're given a file, write to it.
- ofstream & fp = data_file_manager.GetOFStream(filename);
-
- // Loop through all of the genotypes in this batch...
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- cLandscape landscape(m_world, genotype->GetGenome(), inst_set);
- if (num_test == 0) landscape.Process(dist);
- else landscape.RandomProcess(num_test,dist,num_test,num_test*2);
- landscape.PrintStats(fp);
- }
-}
-
-void cAnalyze::AnalyzeEpistasis(cString cur_string)
-{
- if (verbose == true) cout << "Epistasis on " << cur_batch << endl;
- else cout << "Calculating epistasis values..." << endl;
-
- // Load in the variables...
- cString filename;
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
- int test_num = 1;
- if (cur_string.GetSize() != 0) test_num = cur_string.PopWord().AsInt();
-
- // If we're given a file, write to it.
- ofstream & fp = data_file_manager.GetOFStream(filename);
-
- // Loop through all of the genotypes in this batch...
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- cLandscape landscape(m_world, genotype->GetGenome(), inst_set);
- if (test_num == 1) landscape.TestAllPairs(fp);
- else landscape.TestPairs(test_num,fp);
- landscape.PrintStats(fp);
- }
-}
-
-
-// This command will take the current batch and analyze how well organisms
-// cross-over with each other, both across the population and between mates.
-
-void cAnalyze::AnalyzeMateSelection(cString cur_string)
-{
- int sample_size = 10000;
- if (cur_string.GetSize() != 0) sample_size = cur_string.PopWord().AsInt();
- cString filename("none");
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
- double min_swap_frac = 0.0;
- if (cur_string.GetSize() != 0) min_swap_frac=cur_string.PopWord().AsDouble();
- double max_swap_frac = 1.0 - min_swap_frac;
-
- cout << "Analyzing Mate Selection... " << endl;
-
- // Do some quick tests before moving on...
- if (min_swap_frac < 0.0 || min_swap_frac >= 0.5) {
- cerr << "ERROR: Minimum swap fraction out of range [0.0, 0.5)." << endl;
- }
-
- // Next, we create an array that contains pointers to all of the organisms
- // in this batch. Note that we want to select genotypes based on their
- // abundance, so they will have one entry in the array per organism. Note
- // that we only consider viable genotypes.
-
- // Start by counting the total number of organisms (and do other such
- // data collection...
- tHashTable<int, int> mate_id_counts;
-
- int org_count = 0;
- int gen_count = 0;
- cAnalyzeGenotype * genotype = NULL;
- tListIterator<cAnalyzeGenotype> list_it(batch[cur_batch].List());
- while ((genotype = list_it.Next()) != NULL) {
- if (genotype->GetViable() == false || genotype->GetNumCPUs() == 0) {
- continue;
- }
- gen_count++;
- org_count += genotype->GetNumCPUs();
-
- // Keep track of how many organisms have each mate id...
- int mate_id = genotype->GetMateID();
- int count = 0;
- mate_id_counts.Find(mate_id, count);
- count += genotype->GetNumCPUs();
- mate_id_counts.SetValue(mate_id, count);
- }
-
- // Create an array of the correct size.
- tArray<cAnalyzeGenotype *> genotype_array(org_count);
-
- // And insert all of the organisms into the array.
- int cur_pos = 0;
- while ((genotype = list_it.Next()) != NULL) {
- if (genotype->GetViable() == false) continue;
- int cur_count = genotype->GetNumCPUs();
- for (int i = 0; i < cur_count; i++) {
- genotype_array[cur_pos++] = genotype;
- }
- }
-
-
- // Setup some variables to collect statistics.
- int total_matches_tested = 0;
- int fail_count = 0;
- int match_fail_count = 0;
-
- // Loop through all of the tests, picking random organisms each time and
- // performing a random cross test.
- cAnalyzeGenotype * genotype2 = NULL;
- for (int test_id = 0; test_id < sample_size; test_id++) {
- genotype = genotype_array[ m_world->GetRandom().GetUInt(org_count) ];
- genotype2 = genotype_array[ m_world->GetRandom().GetUInt(org_count) ];
-
- // Stop immediately if we're comparing a genotype to itself.
- if (genotype == genotype2) {
- total_matches_tested++;
- continue;
- }
-
- // Setup the random parameters for this test.
- cCPUMemory test_genome0 = genotype->GetGenome();
- cCPUMemory test_genome1 = genotype2->GetGenome();
-
- double start_frac = -1.0;
- double end_frac = -1.0;
- double swap_frac = -1.0;
- while (swap_frac < min_swap_frac || swap_frac > max_swap_frac) {
- start_frac = m_world->GetRandom().GetDouble();
- end_frac = m_world->GetRandom().GetDouble();
- if (start_frac > end_frac) nFunctions::Swap(start_frac, end_frac);
- swap_frac = end_frac - start_frac;
- }
-
- int start0 = (int) (start_frac * (double) test_genome0.GetSize());
- int end0 = (int) (end_frac * (double) test_genome0.GetSize());
- int size0 = end0 - start0;
-
- int start1 = (int) (start_frac * (double) test_genome1.GetSize());
- int end1 = (int) (end_frac * (double) test_genome1.GetSize());
- int size1 = end1 - start1;
-
- int new_size0 = test_genome0.GetSize() - size0 + size1;
- int new_size1 = test_genome1.GetSize() - size1 + size0;
-
- // Setup some statistics for this particular test.
- bool same_mate_id = ( genotype->GetMateID() == genotype2->GetMateID() );
- if (same_mate_id == true) total_matches_tested++;
-
- // Don't Crossover if offspring will be illegal!!!
- if (new_size0 < MIN_CREATURE_SIZE || new_size0 > MAX_CREATURE_SIZE ||
- new_size1 < MIN_CREATURE_SIZE || new_size1 > MAX_CREATURE_SIZE) {
- fail_count++;
- if (same_mate_id == true) match_fail_count++;
- continue;
- }
-
- // Do the replacement... We're only going to test genome0, so we only
- // need to modify that one.
- cGenome cross1 = cGenomeUtil::Crop(test_genome1, start1, end1);
- test_genome0.Replace(start0, size0, cross1);
-
- // Do the test.
- cCPUTestInfo test_info;
-
- // Run each side, and determine viability...
- m_world->GetTestCPU().TestGenome(test_info, test_genome0);
- if( test_info.IsViable() == false ) {
- fail_count++;
- if (same_mate_id == true) match_fail_count++;
- }
- }
-
- // Do some calculations on the sizes of the mate groups...
- const int num_mate_groups = mate_id_counts.GetSize();
-
- // Collect lists on all of the mate groups for the calculations...
- tList<int> key_list;
- tList<int> count_list;
- mate_id_counts.AsLists(key_list, count_list);
- tListIterator<int> count_it(count_list);
-
- int max_group_size = 0;
- double mate_id_entropy = 0.0;
- while (count_it.Next() != NULL) {
- int cur_count = *(count_it.Get());
- double cur_frac = ((double) cur_count) / ((double) org_count);
- if (cur_count > max_group_size) max_group_size = cur_count;
- mate_id_entropy -= cur_frac * log(cur_frac);
- }
-
- // Calculate the final answer
- double fail_frac = (double) fail_count / (double) sample_size;
- double match_fail_frac =
- (double) match_fail_count / (double) total_matches_tested;
- cout << " ave fraction failed = " << fail_frac << endl
- << " ave matches failed = " << match_fail_frac << endl
- << " total mate matches = " << total_matches_tested
- << " / " << sample_size<< endl;
-
- if (filename == "none") return;
-
- cDataFile & df = data_file_manager.Get(filename);
- df.WriteComment( "Mate selection information" );
- df.WriteTimeStamp();
-
- df.Write(fail_frac, "Average fraction failed");
- df.Write(match_fail_frac, "Average fraction of mate matches failed");
- df.Write(sample_size, "Total number of crossovers tested");
- df.Write(total_matches_tested, "Number of crossovers with matching mate IDs");
- df.Write(gen_count, "Number of genotypes in test batch");
- df.Write(org_count, "Number of organisms in test batch");
- df.Write(num_mate_groups, "Number of distinct mate IDs");
- df.Write(max_group_size, "Size of the largest distinct mate ID group");
- df.Write(mate_id_entropy, "Diversity of mate IDs (entropy)");
- df.Endl();
-}
-
-
-void cAnalyze::AnalyzeComplexityDelta(cString cur_string)
-{
- // This command will examine the current population, and sample mutations
- // to see what the distribution of complexity changes is. Only genotypes
- // with a certain abundance (default=3) will be tested to make sure that
- // the organism didn't already have hidden complexity due to a downward
- // step.
- cout << "Testing complexity delta." << endl;
-
- cString filename = "complexity_delta.dat";
- int num_tests = 10;
- double copy_mut_prob = m_world->GetConfig().COPY_MUT_PROB.Get();
- double ins_mut_prob = m_world->GetConfig().DIVIDE_INS_PROB.Get();
- double del_mut_prob = m_world->GetConfig().DIVIDE_DEL_PROB.Get();
- int count_threshold = 3;
-
- if (cur_string.GetSize() > 0) filename = cur_string.PopWord();
- if (cur_string.GetSize() > 0) num_tests = cur_string.PopWord().AsInt();
- if (cur_string.GetSize() > 0) copy_mut_prob = cur_string.PopWord().AsDouble();
- if (cur_string.GetSize() > 0) ins_mut_prob = cur_string.PopWord().AsDouble();
- if (cur_string.GetSize() > 0) del_mut_prob = cur_string.PopWord().AsDouble();
- if (cur_string.GetSize() > 0) count_threshold = cur_string.PopWord().AsInt();
-
- if (verbose == true) {
- cout << "...using:"
- << " filename='" << filename << "'"
- << " num_tests=" << num_tests
- << " copy_mut_prob=" << copy_mut_prob
- << " ins_mut_prob=" << ins_mut_prob
- << " del_mut_prob=" << del_mut_prob
- << " count_threshold=" << count_threshold
- << endl;
- }
-
- // Create an array of all of the genotypes above threshold.
- cAnalyzeGenotype * genotype = NULL;
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
-
- // Loop through all genotypes to perform a census
- int org_count = 0;
- while ((genotype = batch_it.Next()) != NULL) {
- // Only count genotypes above threshold
- if (genotype->GetNumCPUs() >= count_threshold) {
- org_count += genotype->GetNumCPUs();
- }
- }
-
- // Create an array to store pointers to the genotypes and fill it in.
- tArray<cAnalyzeGenotype *> org_array(org_count);
- int cur_org = 0;
- batch_it.Reset();
- while ((genotype = batch_it.Next()) != NULL) {
- // Ignore genotypes below threshold.
- if (genotype->GetNumCPUs() < count_threshold) continue;
-
- // Insert the remaining genotypes into the array.
- for (int i = 0; i < genotype->GetNumCPUs(); i++) {
- org_array[cur_org] = genotype;
- cur_org++;
- }
- }
-
- // Open up the file and prepare it for output.
- cDataFile & df = data_file_manager.Get(filename);
- df.WriteComment( "An analyze of expected complexity changes between parent and offspring" );
- df.WriteTimeStamp();
-
- // Next check the appropriate number of organisms, perform mutations, and
- // store the results.
- for (int cur_test = 0; cur_test < num_tests; cur_test++) {
- // Pick the genotype to test.
- int test_org_id = m_world->GetRandom().GetInt(org_count);
- genotype = org_array[test_org_id];
-
- // Create a copy of the genome.
- cCPUMemory mod_genome = genotype->GetGenome();
-
- if (copy_mut_prob == 0.0 &&
- ins_mut_prob == 0.0 &&
- del_mut_prob == 0.0) {
- cerr << "ERROR: All mutation rates are zero! No complexity delta analysis possible." << endl;
- return;
- }
-
- // Perform the per-site mutations -- we are going to keep looping until
- // we trigger at least one mutation.
- int num_mutations = 0;
- while (num_mutations == 0) {
- if (copy_mut_prob > 0.0) {
- for (int i = 0; i < mod_genome.GetSize(); i++) {
- if (m_world->GetRandom().P(copy_mut_prob)) {
- mod_genome[i] = inst_set.GetRandomInst();
- num_mutations++;
- }
- }
- }
-
- // Perform an Insertion if it has one.
- if (m_world->GetRandom().P(ins_mut_prob)) {
- int ins_line = m_world->GetRandom().GetInt(mod_genome.GetSize() + 1);
- mod_genome.Insert(ins_line, inst_set.GetRandomInst());
- num_mutations++;
- }
-
- // Perform a Deletion if it has one.
- if (m_world->GetRandom().P(del_mut_prob)) {
- int del_line = m_world->GetRandom().GetInt(mod_genome.GetSize());
- mod_genome.Remove(del_line);
- num_mutations++;
- }
- }
-
- // Calculate the complexities....
- genotype->Recalculate();
- double start_complexity = genotype->GetComplexity();
- double start_fitness = genotype->GetFitness();
- int start_length = genotype->GetLength();
- int start_gest = genotype->GetGestTime();
-
- cAnalyzeGenotype new_genotype(m_world, mod_genome, inst_set);
- new_genotype.Recalculate();
- double end_complexity = new_genotype.GetComplexity();
- double complexity_change = end_complexity - start_complexity;
- double end_fitness = new_genotype.GetFitness();
- int end_length = new_genotype.GetLength();
- int end_gest = new_genotype.GetGestTime();
-
- df.Write(num_mutations, "Number of mutational differences between original organism and mutant.");
- df.Write(complexity_change, "Complexity difference between original organism and mutant.");
- df.Write(start_complexity, "Complexity of initial organism.");
- df.Write(end_complexity, "Complexity of mutant.");
- df.Write(start_fitness, "Fitness of initial organism.");
- df.Write(end_fitness, "Fitness of mutant.");
- df.Write(start_length, "Length of initial organism.");
- df.Write(end_length, "Length of mutant.");
- df.Write(start_gest, "Gestation Time of initial organism.");
- df.Write(end_gest, "Gestation Time of mutant.");
- df.Write(genotype->GetID(), "ID of initial genotype.");
- df.Endl();
- }
-}
-
-void cAnalyze::AnalyzeKnockouts(cString cur_string)
-{
- cout << "Analyzing the effects of knockouts..." << endl;
-
- cString filename = "knockouts.dat";
- if (cur_string.GetSize() > 0) filename = cur_string.PopWord();
-
- int max_knockouts = 1;
- if (cur_string.GetSize() > 0) max_knockouts = cur_string.PopWord().AsInt();
-
- // Open up the data file...
- cDataFile & df = data_file_manager.Get(filename);
- df.WriteComment( "Analysis of knockouts in genomes" );
- df.WriteTimeStamp();
-
-
- // Setup a NULL instruction in a special inst set.
- cInstSet ko_inst_set(inst_set);
- // Locate the instruction corresponding to "NULL" in the instruction library.
- {
- cInstruction lib_null_inst = ko_inst_set.GetInstLib()->GetInst("NULL");
- if (lib_null_inst == ko_inst_set.GetInstLib()->GetInstError()) {
- cout << "<cAnalyze::AnalyzeKnockouts> got error:" << endl
- << " instruction 'NULL' not in current hardware type" << endl;
- exit(1);
- }
- // Add mapping to located instruction.
- ko_inst_set.Add2(lib_null_inst.GetOp());
- }
- const cInstruction null_inst = ko_inst_set.GetInst("NULL");
-
- // Loop through all of the genotypes in this batch...
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- if (verbose == true) cout << " Knockout: " << genotype->GetName() << endl;
-
- // Calculate the stats for the genotype we're working with...
- genotype->Recalculate();
- const double base_fitness = genotype->GetFitness();
-
- const int max_line = genotype->GetLength();
- const cGenome & base_genome = genotype->GetGenome();
- cGenome mod_genome(base_genome);
-
- // Loop through all the lines of code, testing the removal of each.
- int dead_count = 0;
- int neg_count = 0;
- int neut_count = 0;
- int pos_count = 0;
- for (int line_num = 0; line_num < max_line; line_num++) {
- // Save a copy of the current instruction and replace it with "NULL"
- int cur_inst = base_genome[line_num].GetOp();
- mod_genome[line_num] = null_inst;
- cAnalyzeGenotype ko_genotype(m_world, mod_genome, ko_inst_set);
- ko_genotype.Recalculate();
-
- double ko_fitness = ko_genotype.GetFitness();
- if (ko_fitness == 0.0) dead_count++;
- else if (ko_fitness < base_fitness) neg_count++;
- else if (ko_fitness == base_fitness) neut_count++;
- else if (ko_fitness > base_fitness) pos_count++;
- else cerr << "ERROR: illegal state in AnalyzeKnockouts()" << endl;
-
- // Reset the mod_genome back to the original sequence.
- mod_genome[line_num].SetOp(cur_inst);
- }
-
- df.Write(genotype->GetID(), "Genotype ID");
- df.Write(dead_count, "Count of lethal knockouts");
- df.Write(neg_count, "Count of detrimental knockouts");
- df.Write(neut_count, "Count of neutral knockouts");
- df.Write(pos_count, "Count of beneficial knockouts");
- df.Endl();
- }
-}
-
-
-void cAnalyze::CommandFitnessMatrix(cString cur_string)
-{
- if (verbose == true) cout << "Calculating fitness matrix for batch " << cur_batch << endl;
- else cout << "Calculating fitness matrix..." << endl;
-
- cout << "Warning: considering only first genotype of the batch!" << endl;
-
- // Load in the variables...
- int depth_limit = 4;
- if (cur_string.GetSize() != 0) depth_limit = cur_string.PopWord().AsInt();
-
- double fitness_threshold_ratio = .9;
- if (cur_string.GetSize() != 0) fitness_threshold_ratio = cur_string.PopWord().AsDouble();
-
- int ham_thresh = 1;
- if (cur_string.GetSize() != 0) ham_thresh = cur_string.PopWord().AsInt();
-
- double error_rate_min = 0.005;
- if (cur_string.GetSize() != 0) error_rate_min = cur_string.PopWord().AsDouble();
-
- double error_rate_max = 0.05;
- if (cur_string.GetSize() != 0) error_rate_max = cur_string.PopWord().AsDouble();
-
- double error_rate_step = 0.005;
- if (cur_string.GetSize() != 0) error_rate_step = cur_string.PopWord().AsDouble();
-
- double vect_fmax = 1.1;
- if (cur_string.GetSize() != 0) vect_fmax = cur_string.PopWord().AsDouble();
-
- double vect_fstep = .1;
- if (cur_string.GetSize() != 0) vect_fstep = cur_string.PopWord().AsDouble();
-
- int diag_iters = 200;
- if (cur_string.GetSize() != 0) diag_iters = cur_string.PopWord().AsInt();
-
- int write_ham_vector = 0;
- if (cur_string.GetSize() != 0) write_ham_vector = cur_string.PopWord().AsInt();
-
- int write_full_vector = 0;
- if (cur_string.GetSize() != 0) write_full_vector = cur_string.PopWord().AsInt();
-
- // Consider only the first genotypes in this batch...
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = batch_it.Next();
-
- cFitnessMatrix matrix(m_world, genotype->GetGenome(), &inst_set);
-
- matrix.CalcFitnessMatrix( depth_limit, fitness_threshold_ratio, ham_thresh, error_rate_min, error_rate_max, error_rate_step, vect_fmax, vect_fstep, diag_iters, write_ham_vector, write_full_vector );
-}
-
-
-void cAnalyze::CommandMapTasks(cString cur_string)
-{
- cout << "Constructing genotype-phenotype maps..." << endl;
-
- // Load in the variables...
- cString directory = PopDirectory(cur_string, "phenotype/");
- int print_mode = 0; // 0=Normal, 1=Boolean results
- int file_type = FILE_TYPE_TEXT;
-
- // HTML special flags...
- bool link_maps = false; // Should muliple maps be linked together?
- bool link_insts = false; // Should links be made to instruction descs?
-
- // Collect any other format information needed...
- tList< tDataEntryCommand<cAnalyzeGenotype> > output_list;
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > output_it(output_list);
-
- cStringList arg_list(cur_string);
-
- cout << "Found " << arg_list.GetSize() << " args." << endl;
-
- // Check for some command specific variables.
- if (arg_list.PopString("0") != "") print_mode = 0;
- if (arg_list.PopString("1") != "") print_mode = 1;
- if (arg_list.PopString("text") != "") file_type = FILE_TYPE_TEXT;
- if (arg_list.PopString("html") != "") file_type = FILE_TYPE_HTML;
- if (arg_list.PopString("link_maps") != "") link_maps = true;
- if (arg_list.PopString("link_insts") != "") link_insts = true;
-
- cout << "There are " << arg_list.GetSize() << " column args." << endl;
-
- LoadGenotypeDataList(arg_list, output_list);
-
- cout << "Args are loaded." << endl;
-
- const int num_cols = output_list.GetSize();
-
- // Give some information in verbose mode.
- if (verbose == true) {
- cout << " outputing as ";
- if (print_mode == 1) cout << "boolean ";
- if (file_type == FILE_TYPE_TEXT) {
- cout << "text files." << endl;
- } else { // if (file_type == FILE_TYPE_HTML) {
- cout << "HTML files";
- if (link_maps == true) cout << "; linking files together";
- if (link_maps == true) cout << "; linking inst names to descs";
- cout << "." << endl;
- }
- cout << " Format: ";
-
- output_it.Reset();
- while (output_it.Next() != NULL) {
- cout << output_it.Get()->GetName() << " ";
- }
- cout << endl;
- }
-
-
- ///////////////////////////////////////////////////////
- // Loop through all of the genotypes in this batch...
-
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- if (verbose == true) cout << " Mapping " << genotype->GetName() << endl;
-
- // Construct this filename...
- cString filename;
- if (file_type == FILE_TYPE_TEXT) {
- filename.Set("%stasksites.%s.dat", directory(), genotype->GetName()());
- } else { // if (file_type == FILE_TYPE_HTML) {
- filename.Set("%stasksites.%s.html", directory(), genotype->GetName()());
- }
- ofstream fp(filename());
-
- // Construct linked filenames...
- cString next_file("");
- cString prev_file("");
- if (link_maps == true) {
- // Check the next genotype on the list...
- if (batch_it.Next() != NULL) {
- next_file.Set("tasksites.%s.html", batch_it.Get()->GetName()());
- }
- batch_it.Prev(); // Put the list back where it was...
-
- // Check the previous genotype on the list...
- if (batch_it.Prev() != NULL) {
- prev_file.Set("tasksites.%s.html", batch_it.Get()->GetName()());
- }
- batch_it.Next(); // Put the list back where it was...
- }
-
- // Calculate the stats for the genotype we're working with...
- genotype->Recalculate();
-
- // Headers...
- if (file_type == FILE_TYPE_TEXT) {
- fp << "-1 " << batch[cur_batch].Name() << " "
- << genotype->GetID() << " ";
-
- tDataEntryCommand<cAnalyzeGenotype> * data_command = NULL;
- while ((data_command = output_it.Next()) != NULL) {
- data_command->SetTarget(genotype);
- fp << data_command->GetEntry() << " ";
- }
- fp << endl;
-
- } else { // if (file_type == FILE_TYPE_HTML) {
- // Mark file as html
- fp << "<html>" << endl;
-
- // Setup any javascript macros needed...
- fp << "<head>" << endl;
- if (link_insts == true) {
- fp << "<script language=\"javascript\">" << endl
- << "function Inst(inst_name)" << endl
- << "{" << endl
- << "var filename = \"help.\" + inst_name + \".html\";" << endl
- << "newwin = window.open(filename, 'Instruction', "
- << "'toolbar=0,status=0,location=0,directories=0,menubar=0,"
- << "scrollbars=1,height=150,width=300');" << endl
- << "newwin.focus();" << endl
- << "}" << endl
- << "</script>" << endl;
- }
- fp << "</head>" << endl;
-
- // Setup the body...
- fp << "<body bgcolor=\"#FFFFFF\"" << endl
- << " text=\"#000000\"" << endl
- << " link=\"#0000AA\"" << endl
- << " alink=\"#0000FF\"" << endl
- << " vlink=\"#000044\">" << endl
- << endl
- << "<h1 align=center>Run " << batch[cur_batch].Name()
- << ", ID " << genotype->GetID() << "</h1>" << endl
- << "<center>" << endl
- << endl;
-
- // Links?
- fp << "<table width=90%><tr><td align=left>";
- if (prev_file != "") fp << "<a href=\"" << prev_file << "\">Prev</a>";
- else fp << " ";
- fp << "<td align=right>";
- if (next_file != "") fp << "<a href=\"" << next_file << "\">Next</a>";
- else fp << " ";
- fp << "</tr></table>" << endl;
-
- // The table
- fp << "<table border=1 cellpadding=2>" << endl;
-
- // The headings...///
- fp << "<tr><td colspan=3> ";
- output_it.Reset();
- while (output_it.Next() != NULL) {
- fp << "<th>" << output_it.Get()->GetDesc() << " ";
- }
- fp << "</tr>" << endl;
-
- // The base creature...
- fp << "<tr><th colspan=3>Base Creature";
- tDataEntryCommand<cAnalyzeGenotype> * data_command = NULL;
- cAnalyzeGenotype null_genotype(m_world, "a", inst_set);
- while ((data_command = output_it.Next()) != NULL) {
- data_command->SetTarget(genotype);
- genotype->SetSpecialArgs(data_command->GetArgs());
- if (data_command->Compare(&null_genotype) > 0) {
- fp << "<th bgcolor=\"#00FF00\">";
- }
- else fp << "<th bgcolor=\"#FF0000\">";
-
- if (data_command->HasArg("blank") == true) fp << " " << " ";
- else fp << data_command->GetEntry() << " ";
- }
- fp << "</tr>" << endl;
- }
-
-
- const int max_line = genotype->GetLength();
- const cGenome & base_genome = genotype->GetGenome();
- cGenome mod_genome(base_genome);
-
- // Keep track of the number of failues/successes for attributes...
- int * col_pass_count = new int[num_cols];
- int * col_fail_count = new int[num_cols];
- for (int i = 0; i < num_cols; i++) {
- col_pass_count[i] = 0;
- col_fail_count[i] = 0;
- }
-
- cInstSet map_inst_set(inst_set);
- // Locate instruction corresponding to "NULL" in the instruction library.
- {
- const cInstruction inst_lib_null_inst = map_inst_set.GetInstLib()->GetInst("NULL");
- if(inst_lib_null_inst == map_inst_set.GetInstLib()->GetInstError()){
- cout << "<cAnalyze::CommandMapTasks> got error:" << endl;
- cout << " --- instruction \"NULL\" isn't in the instruction library for the" << endl;
- cout << " --- current hardware type." << endl;
- exit(1);
- }
- // Add mapping to located instruction.
- map_inst_set.Add2(inst_lib_null_inst.GetOp());
- }
- const cInstruction null_inst = map_inst_set.GetInst("NULL");
-
- // Loop through all the lines of code, testing the removal of each.
- for (int line_num = 0; line_num < max_line; line_num++) {
- int cur_inst = base_genome[line_num].GetOp();
- char cur_symbol = base_genome[line_num].GetSymbol();
-
- mod_genome[line_num] = null_inst;
- cAnalyzeGenotype test_genotype(m_world, mod_genome, map_inst_set);
- test_genotype.Recalculate();
-
- if (file_type == FILE_TYPE_HTML) fp << "<tr><td align=right>";
- fp << (line_num + 1) << " ";
- if (file_type == FILE_TYPE_HTML) fp << "<td align=center>";
- fp << cur_symbol << " ";
- if (file_type == FILE_TYPE_HTML) fp << "<td align=center>";
- if (link_insts == true) {
- fp << "<a href=\"javascript:Inst('"
- << map_inst_set.GetName(cur_inst)
- << "')\">";
- }
- fp << map_inst_set.GetName(cur_inst) << " ";
- if (link_insts == true) fp << "</a>";
-
-
- // Print the individual columns...
- output_it.Reset();
- tDataEntryCommand<cAnalyzeGenotype> * data_command = NULL;
- int cur_col = 0;
- while ((data_command = output_it.Next()) != NULL) {
- data_command->SetTarget(&test_genotype);
- test_genotype.SetSpecialArgs(data_command->GetArgs());
- int compare = data_command->Compare(genotype);
- if (file_type == FILE_TYPE_HTML) {
- data_command->HTMLPrint(fp, compare,
- !(data_command->HasArg("blank")));
- }
- else fp << data_command->GetEntry() << " ";
-
- if (compare == -2) col_fail_count[cur_col]++;
- else if (compare == 2) col_pass_count[cur_col]++;
- cur_col++;
- }
- if (file_type == FILE_TYPE_HTML) fp << "</tr>";
- fp << endl;
-
- // Reset the mod_genome back to the original sequence.
- mod_genome[line_num].SetOp(cur_inst);
- }
-
-
- // Construct the final line of the table with all totals...
- if (file_type == FILE_TYPE_HTML) {
- fp << "<tr><th colspan=3>Totals";
-
- for (int i = 0; i < num_cols; i++) {
- if (col_pass_count[i] > 0) {
- fp << "<th bgcolor=\"#00FF00\">" << col_pass_count[i];
- }
- else if (col_fail_count[i] > 0) {
- fp << "<th bgcolor=\"#FF0000\">" << col_fail_count[i];
- }
- else fp << "<th>0";
- }
- fp << "</tr>" << endl;
-
- // And close everything up...
- fp << "</table>" << endl
- << "</center>" << endl;
- }
-
- delete [] col_pass_count;
- delete [] col_fail_count;
- }
-}
-
-void cAnalyze::CommandAverageModularity(cString cur_string)
-{
- cout << "Average Modularity calculations" << endl;
-
- // Load in the variables...
- cString filename = cur_string.PopWord();
- // cString filename = "average.dat";
-
- int print_mode = 0; // 0=Normal, 1=Boolean results
-
- // Collect any other format information needed...
- tList< tDataEntryCommand<cAnalyzeGenotype> > output_list;
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > output_it(output_list);
-
- cStringList arg_list(cur_string);
-
- cout << "Found " << arg_list.GetSize() << " args." << endl;
-
- // Check for some command specific variables.
- if (arg_list.PopString("0") != "") print_mode = 0;
- if (arg_list.PopString("1") != "") print_mode = 1;
-
- cout << "There are " << arg_list.GetSize() << " column args." << endl;
-
- LoadGenotypeDataList(arg_list, output_list);
-
- cout << "Args are loaded." << endl;
-
- const int num_cols = output_list.GetSize();
-
- // Give some information in verbose mode.
- if (verbose == true) {
- cout << " outputing as ";
- if (print_mode == 1) cout << "boolean ";
- cout << "text files." << endl;
- cout << " Format: ";
-
- output_it.Reset();
- while (output_it.Next() != NULL) {
- cout << output_it.Get()->GetName() << " ";
- }
- cout << endl;
- }
-
- ofstream & fp = data_file_manager.GetOFStream(filename);
-
- // printing the headers
- // not done by default since many dumps may be analyzed at the same time
- // and results would be put in the same file
- if (arg_list.GetSize()==0) {
- // Headers
- fp << "# Avida analyze modularity data" << endl;
- fp << "# 1: organism length" << endl;
- fp << "# 2: number of tasks done" << endl;
- fp << "# 3: number of sites used in tasks" << endl;
- fp << "# 4: proportion of sites used in tasks" << endl;
- fp << "# 5: average number of tasks done per site" << endl;
- fp << "# 6: average number sites per task done" << endl;
- fp << "# 7: average number tasks per site per task" << endl;
- fp << "# 8: average proportion of the non-overlaping region of a task" << endl;
- fp << "# 9-17: average StDev in positions used for task 1-9" << endl;
- fp << "# 18-26: average number of sites necessary for each of the tasks" << endl;
- fp << "# 27-36: number of sites involved in 0-9 tasks" << endl;
- fp << "# 37-45: average task length (distance from first to last inst used)" << endl;
- fp << endl;
- return;
- }
-
- // initialize various variables used in calculations
-
- int num_orgs = 0; // number of organisms in the dump
-
- double av_length = 0; // average organism length
- double av_task = 0; // average # of tasks done
- double av_inst = 0; // average # instructions used in tasks
- double av_inst_len = 0; // proportion of sites used for tasks
- double av_site_task = 0; // average number of sites per task
- double av_task_site = 0; // average number of tasks per site
- double av_t_s_norm = 0; // average number of tasks per site per task
- double av_task_overlap = 0; // average overlap between tasks
-
- // average StDev in positions used for a task
- tArray<double> std_task_position(num_cols);
- std_task_position.SetAll(0.0);
-
- // # of organisms actually doing a task
- tArray<double> org_task(num_cols);
- org_task.SetAll(0.0);
-
- // av. # of sites necessary for each of the tasks
- tArray<double> av_num_inst(num_cols);
- av_num_inst.SetAll(0.0);
-
- // number of sites involved in 0-9 tasks
- tArray<double> av_inst_task(num_cols+1);
- av_inst_task.SetAll(0.0);
-
- // av. # task length (distance from first to last site used)
- tArray<double> av_task_length(num_cols);
- av_task_length.SetAll(0.0);
-
-
- ///////////////////////////////////////////////////////
- // Loop through all of the genotypes in this batch...
- ///////////////////////////////////////////////////////
-
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
-
- // would like to test oly the viable ones, but they can be non-viable
- // and still reproduce and do tasks
- // while ((genotype = batch_it.Next()) != NULL && genotype->GetViable()) {
- while ((genotype = batch_it.Next()) != NULL) {
-
- int num_cpus = genotype->GetNumCPUs();
-
- if (verbose == true) cout << " Mapping " << genotype->GetName() << endl;
-
- // Calculate the stats for the genotype we're working with...
- genotype->Recalculate();
-
- // Check if the organism does any tasks.
- int does_tasks = 0;
- for (int i = 0; i < num_cols; i++) {
- if (genotype->GetTaskCount(i) > 0) does_tasks = 1;
- }
-
- // Don't calculate the modularity if the organism doesn't reproduce
- // i.e. if the fitness is 0
- if (genotype->GetFitness() != 0 && does_tasks != 0) {
- num_orgs = num_orgs + num_cpus;
-
- const int max_line = genotype->GetLength();
- const cGenome & base_genome = genotype->GetGenome();
- cGenome mod_genome(base_genome);
-
- // Create and initialize the modularity matrix
- tMatrix<int> mod_matrix(num_cols, max_line);
- mod_matrix.SetAll(0);
-
- // Create and initialize the task overalp matrix
- tMatrix<int> task_overlap(num_cols, num_cols);
- task_overlap.SetAll(0);
-
- // Create an initialize the counters for modularity
- tArray<int> num_task(max_line); // number of tasks instruction is used in
- tArray<int> num_inst(num_cols); // number of instructions involved in a task
- tArray<int> sum(num_cols); // helps with StDev calculations
- tArray<int> sumsq(num_cols); // helps with StDev calculations
- tArray<int> inst_task(num_cols+1); // # of inst's involved in 0,1,2,3... tasks
- tArray<int> task_length(num_cols); // ditance between first and last inst involved in a task
-
- num_task.SetAll(0);
- num_inst.SetAll(0);
- sum.SetAll(0);
- sumsq.SetAll(0);
- inst_task.SetAll(0);
- task_length.SetAll(0);
-
- int total_task = 0; // total number of tasks done
- int total_inst = 0; // total number of instructions involved in tasks
- int total_all = 0; // sum of mod_matrix
- double sum_task_overlap = 0;// task overlap for for this geneome
-
- cInstSet map_inst_set(inst_set);
-
- // Locate instruction corresponding to "NULL" in the instruction library.
- {
- const cInstruction inst_lib_null_inst = map_inst_set.GetInstLib()->GetInst("NULL");
- if(inst_lib_null_inst == map_inst_set.GetInstLib()->GetInstError()){
- cout << "<cAnalyze::CommandMapTasks> got error:" << endl;
- cout << " --- instruction \"NULL\" isn't in the instruction library;" << endl;
- cout << " --- get somebody to map a function to \"NULL\" in the library." << endl;
- cout << " --- (probably to class method \"cHardware-of-some-type::initInstLib\"" << endl;
- cout << " --- in file named \"cpu/hardware-of-some-type.cc\".)" << endl;
- cout << " --- bailing-out." << endl;
- exit(1);
- }
- // Add mapping to located instruction.
- map_inst_set.Add2(inst_lib_null_inst.GetOp());
- }
- const cInstruction null_inst = map_inst_set.GetInst("NULL");
-
- // Loop through all the lines of code, testing the removal of each.
- for (int line_num = 0; line_num < max_line; line_num++) {
- int cur_inst = base_genome[line_num].GetOp();
-
- mod_genome[line_num] = null_inst;
- cAnalyzeGenotype test_genotype(m_world, mod_genome, map_inst_set);
- cAnalyzeGenotype old_genotype(m_world, base_genome, map_inst_set);
- test_genotype.Recalculate();
-
- // Print the individual columns...
- output_it.Reset();
- tDataEntryCommand<cAnalyzeGenotype> * data_command = NULL;
- int cur_col = 0;
- while ((data_command = output_it.Next()) != NULL) {
- data_command->SetTarget(&test_genotype);
- test_genotype.SetSpecialArgs(data_command->GetArgs());
- // This is done so that under 'binary' option it marks
- // the task as being influenced by the mutation iff
- // it is completely knocked out, not just decreased
- genotype->SetSpecialArgs(data_command->GetArgs());
-
- int compare = data_command->Compare(genotype);
-
- // If knocking out an instruction stops the expression of a
- // particular task, mark that in the modularity matrix
- // and add it to two counts
- // Only do the checking if the test_genotype replicate, i.e.
- // if it's fitness is not zeros
-
- if (compare < 0 && test_genotype.GetFitness() != 0) {
- mod_matrix(cur_col,line_num) = 1;
- num_inst[cur_col]++;
- num_task[line_num]++;
- }
- cur_col++;
- }
-
- // Reset the mod_genome back to the original sequence.
- mod_genome[line_num].SetOp(cur_inst);
- } // end of genotype-phenotype mapping for a single organism
-
- for (int i = 0; i < num_cols; i++) {if (num_inst[i] != 0) total_task++;}
- for (int i = 0; i < max_line; i++) {if (num_task[i] != 0) total_inst++;}
- for (int i = 0; i < num_cols; i++) {total_all = total_all + num_inst[i];}
-
- // Add the values to the av_ variables, used for calculating the average
- // in order to weigh them by abundance, multiply everything by num_cpus
-
- av_length = av_length + max_line*num_cpus;
- av_task = av_task + total_task*num_cpus;
- av_inst = av_inst + total_inst*num_cpus;
- av_inst_len = av_inst_len + (double) total_inst*num_cpus/max_line;
-
- if (total_task !=0) av_site_task = av_site_task + num_cpus * (double) total_all/total_task;
- if (total_inst !=0) av_task_site = av_task_site + num_cpus * (double) total_all/total_inst;
- if (total_inst !=0 && total_task !=0) {
- av_t_s_norm = av_t_s_norm + num_cpus * (double) total_all/(total_inst*total_task);
- }
-
- for (int i = 0; i < num_cols; i++) {
- if (num_inst[i] > 0) {
- av_num_inst[i] = av_num_inst[i] + num_inst[i] * num_cpus;
- org_task[i] = org_task[i] + num_cpus; // count how many are actually doing the task
- }
- }
-
- // calculate average task overlap
- // first construct num_task x num_task matrix with number of sites overlapping
- for (int i = 0; i < max_line; i++) {
- for (int j = 0; j < num_cols; j++) {
- for (int k = j; k < num_cols; k++) {
- if (mod_matrix(j,i)>0 && mod_matrix(k,i)>0) {
- task_overlap(j,k)++;
- if (j!=k) task_overlap(k,j)++;
- }
- }
- }
- }
-
- // go though the task_overlap matrix, add and average everything up.
- if (total_task > 1) {
- for (int i = 0; i < num_cols; i++) {
- double overlap_per_task = 0;
- for (int j = 0; j < num_cols; j++) {
- if (i!=j) {overlap_per_task = overlap_per_task + task_overlap(i,j);}
- }
- if (task_overlap(i,i) !=0){
- sum_task_overlap = sum_task_overlap + overlap_per_task / (task_overlap(i,i) * (total_task-1));
- }
- }
- }
-
- // now, divide that by number of tasks done and add to the grand sum, weigthed by num_cpus
- if (total_task!=0) {
- av_task_overlap = av_task_overlap + num_cpus * (double) sum_task_overlap/total_task ;
- }
- // calculate the first/last postion of a task, the task "spread"
- // starting from the top look for the fist command that matters for a task
-
- for (int i = 0; i < num_cols; i++) {
- int j = 0;
- while (j < max_line) {
- if (mod_matrix(i,j) > 0 && task_length[i] == 0 ) {
- task_length[i] = j;
- break;
- }
- j++;
- }
- }
-
- // starting frm the bottom look for the last command that matters for a task
- // and substract it from the first to get the task length
- // add one in order to account for both the beginning and the end instruction
- for (int i = 0; i < num_cols; i++) {
- int j = max_line - 1;
- while (j > -1) {
- if (mod_matrix(i,j) > 0) {
- task_length[i] = j - task_length[i] + 1;
- break;
- }
- j--;
- }
- }
- // add the task lengths to the average for the batch
- // weigthed by the number of cpus for that genotype
- for (int i = 0; i < num_cols; i++) {
- av_task_length[i] = av_task_length[i] + num_cpus * task_length[i];
- }
-
- // calculate the Standard Deviation in the mean position of the task
- for (int i = 0; i < num_cols; i++) {
- for (int j = 0; j < max_line; j++) {
- if (mod_matrix(i,j)>0) sum[i] = sum[i] + j;
- }
- }
-
- double temp = 0;
- for (int i = 0; i < num_cols; i++) {
- if (num_inst[i]>1) {
- double av_sum = sum[i]/num_inst[i];
- for (int j = 0; j < max_line; j++) {
- if (mod_matrix(i,j)>0) temp = (av_sum - j)*(av_sum - j);
- }
- std_task_position[i] = std_task_position[i] + sqrt(temp/(num_inst[i]-1))*num_cpus;
- }
- }
-
- for (int i = 0; i < max_line; i++) { inst_task[num_task[i]]++ ;}
- for (int i = 0; i < num_cols+1; i++) { av_inst_task[i] = av_inst_task[i] + inst_task[i] * num_cpus;}
-
- }
- } // this is the end of the loop going though all the organisms
-
- // make sure there are some organisms doing task in this batch
- // if not, return all zeros
-
- if (num_orgs != 0) {
- fp << (double) av_length/num_orgs << " "; // 1: average length
- fp << (double) av_task/num_orgs << " "; // 2: av. number of tasks done
- fp << (double) av_inst/num_orgs << " "; // 3: av. number of sites used for tasks
- fp << (double) av_inst_len/num_orgs << " "; // 4: proportion of sites used for tasks
- fp << (double) av_task_site/num_orgs << " "; // 5: av. number of tasks per site
- fp << (double) av_site_task/num_orgs << " "; // 6: av. number of sites per task
- fp << (double) av_t_s_norm/num_orgs << " "; // 7: av. number of tasks per site per task
- fp << (double) 1 - av_task_overlap/num_orgs << " "; // 8: av. proportion of a task that DOESN'T overlap
- for (int i = 0; i < num_cols; i++) {
- if (org_task[i] > 0) fp << std_task_position[i]/org_task[i] << " ";
- else fp << 0 << " ";
- }
- for (int i = 0; i < num_cols; i++) {
- if (org_task[i] > 0) fp << (double) av_num_inst[i]/org_task[i] << " ";
- else fp << 0 << " ";
- }
- for (int i = 0; i < num_cols+1; i++) { fp << (double) av_inst_task[i]/num_orgs << " ";}
- for (int i = 0; i < num_cols; i++) { fp << (double) av_task_length[i]/num_orgs << " ";}
- fp << endl;
- }
-
- else {
- for (int i = 0; i < 8+4*num_cols+1; i++) {fp << "0 ";}
- fp << endl;
- }
- }
-
-
-void cAnalyze::CommandMapMutations(cString cur_string)
-{
- cout << "Constructing genome mutations maps..." << endl;
-
- // Load in the variables...
- cString directory = PopDirectory(cur_string, "mutations/");
- int file_type = FILE_TYPE_TEXT;
-
- cStringList arg_list(cur_string);
-
- // Check for some command specific variables.
- if (arg_list.PopString("text") != "") file_type = FILE_TYPE_TEXT;
- if (arg_list.PopString("html") != "") file_type = FILE_TYPE_HTML;
-
- // Give some information in verbose mode.
- if (verbose == true) {
- cout << " outputing as ";
- if (file_type == FILE_TYPE_TEXT) cout << "text files." << endl;
- else cout << "HTML files." << endl;
- }
-
-
- ///////////////////////////////////////////////////////
- // Loop through all of the genotypes in this batch...
-
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- if (verbose == true) {
- cout << " Creating mutation map for " << genotype->GetName() << endl;
- }
-
- // Construct this filename...
- cString filename;
- if (file_type == FILE_TYPE_TEXT) {
- filename.Set("%smut_map.%s.dat", directory(), genotype->GetName()());
- } else { // if (file_type == FILE_TYPE_HTML) {
- filename.Set("%smut_map.%s.html", directory(), genotype->GetName()());
- }
- ofstream fp(filename());
-
- // Calculate the stats for the genotype we're working with...
- genotype->Recalculate();
- const double base_fitness = genotype->GetFitness();
- const int num_insts = inst_set.GetSize();
-
- // Headers...
- if (file_type == FILE_TYPE_TEXT) {
- fp << "# 1: Genome instruction ID (pre-mutation)" << endl;
- for (int i = 0; i < num_insts; i++) {
- fp << "# " << i+1 <<": Fit if mutated to '"
- << inst_set.GetName(i) << "'" << endl;
- }
- fp << "# " << num_insts + 2 << ": Knockout" << endl;
- fp << "# " << num_insts + 3 << ": Fraction Lethal" << endl;
- fp << "# " << num_insts + 4 << ": Fraction Detremental" << endl;
- fp << "# " << num_insts + 5 << ": Fraction Neutral" << endl;
- fp << "# " << num_insts + 6 << ": Fraction Beneficial" << endl;
- fp << "# " << num_insts + 7 << ": Average Fitness" << endl;
- fp << "# " << num_insts + 8 << ": Expected Entropy" << endl;
- fp << "# " << num_insts + 9 << ": Original Instruction Name" << endl;
- fp << endl;
-
- } else { // if (file_type == FILE_TYPE_HTML) {
- // Mark file as html
- fp << "<html>" << endl;
-
- // Setup the body...
- fp << "<body bgcolor=\"#FFFFFF\"" << endl
- << " text=\"#000000\"" << endl
- << " link=\"#0000AA\"" << endl
- << " alink=\"#0000FF\"" << endl
- << " vlink=\"#000044\">" << endl
- << endl
- << "<h1 align=center>Mutation Map for Run " << batch[cur_batch].Name()
- << ", ID " << genotype->GetID() << "</h1>" << endl
- << "<center>" << endl
- << endl;
-
- // The main chart...
- fp << "<table border=1 cellpadding=2>" << endl;
-
- // The headings...///
- fp << "<tr><th>Genome ";
- for (int i = 0; i < num_insts; i++) {
- fp << "<th>" << inst_set.GetName(i) << " ";
- }
- fp << "<th>Knockout ";
- fp << "<th>Frac. Lethal ";
- fp << "<th>Frac. Detremental ";
- fp << "<th>Frac. Neutral ";
- fp << "<th>Frac. Beneficial ";
- fp << "<th>Ave. Fitness ";
- fp << "<th>Expected Entropy ";
- fp << "</tr>" << endl << endl;
- }
-
- const int max_line = genotype->GetLength();
- const cGenome & base_genome = genotype->GetGenome();
- cGenome mod_genome(base_genome);
-
- // Keep track of the number of mutations in each category...
- int total_dead = 0, total_neg = 0, total_neut = 0, total_pos = 0;
- double total_fitness = 0.0;
- tArray<double> col_fitness(num_insts + 1);
- col_fitness.SetAll(0.0);
-
- // Build an empty instruction into the an instruction library.
- cInstSet map_inst_set(inst_set);
- // Locate instruction corresponding to "NULL" in the instruction library.
- {
- const cInstruction inst_lib_null_inst = map_inst_set.GetInstLib()->GetInst("NULL");
- if (inst_lib_null_inst == map_inst_set.GetInstLib()->GetInstError()){
- cout << "<cAnalyze::CommandMapMutations> got error:" << endl;
- cout << " --- instruction \"NULL\" isn't in the instruction library;" << endl;
- cout << " --- get somebody to map a function to \"NULL\" in the library." << endl;
- cout << " --- (probably to class method \"cHardware-of-some-type::initInstLib\"" << endl;
- cout << " --- in file named \"cpu/hardware-of-some-type.cc\".)" << endl;
- cout << " --- bailing-out." << endl;
- exit(1);
- }
- // Add mapping to located instruction.
- map_inst_set.Add2(inst_lib_null_inst.GetOp());
- }
- const cInstruction null_inst = map_inst_set.GetInst("NULL");
-
- cString color_string; // For coloring cells...
-
- // Loop through all the lines of code, testing all mutations...
- for (int line_num = 0; line_num < max_line; line_num++) {
- int cur_inst = base_genome[line_num].GetOp();
- char cur_symbol = base_genome[line_num].GetSymbol();
- int row_dead = 0, row_neg = 0, row_neut = 0, row_pos = 0;
- double row_fitness = 0.0;
-
- // Column 1... the original instruction in the geneome.
- if (file_type == FILE_TYPE_HTML) {
- fp << "<tr><td align=right>" << inst_set.GetName(cur_inst)
- << " (" << cur_symbol << ") ";
- } else {
- fp << cur_inst << " ";
- }
-
- // Columns 2 to D+1 (the possible mutations)
- for (int mod_inst = 0; mod_inst < num_insts; mod_inst++)
- {
- if (mod_inst == cur_inst) {
- if (file_type == FILE_TYPE_HTML) {
- color_string = "#FFFFFF";
- fp << "<th bgcolor=\"" << color_string << "\">";
- }
- }
- else {
- mod_genome[line_num].SetOp(mod_inst);
- cAnalyzeGenotype test_genotype(m_world, mod_genome, inst_set);
- test_genotype.Recalculate();
- const double test_fitness = test_genotype.GetFitness() / base_fitness;
- row_fitness += test_fitness;
- total_fitness += test_fitness;
- col_fitness[mod_inst] += test_fitness;
-
- // Categorize this mutation...
- if (test_fitness == 1.0) { // Neutral Mutation...
- row_neut++;
- total_neut++;
- if (file_type == FILE_TYPE_HTML) color_string = "#FFFFFF";
- } else if (test_fitness == 0.0) { // Lethal Mutation...
- row_dead++;
- total_dead++;
- if (file_type == FILE_TYPE_HTML) color_string = "#FF0000";
- } else if (test_fitness < 1.0) { // Detrimental Mutation...
- row_neg++;
- total_neg++;
- if (file_type == FILE_TYPE_HTML) color_string = "#FFFF00";
- } else { // Beneficial Mutation...
- row_pos++;
- total_pos++;
- if (file_type == FILE_TYPE_HTML) color_string = "#00FF00";
- }
-
- // Write out this cell...
- if (file_type == FILE_TYPE_HTML) {
- fp << "<th bgcolor=\"" << color_string << "\">";
- }
- fp << test_fitness << " ";
- }
- }
-
- // Column: Knockout
- mod_genome[line_num] = null_inst;
- cAnalyzeGenotype test_genotype(m_world, mod_genome, map_inst_set);
- test_genotype.Recalculate();
- const double test_fitness = test_genotype.GetFitness() / base_fitness;
- col_fitness[num_insts] += test_fitness;
-
- // Categorize this mutation if its in HTML mode (color only)...
- if (file_type == FILE_TYPE_HTML) {
- if (test_fitness == 1.0) color_string = "#FFFFFF";
- else if (test_fitness == 0.0) color_string = "#FF0000";
- else if (test_fitness < 1.0) color_string = "#FFFF00";
- else color_string = "#00FF00";
-
- fp << "<th bgcolor=\"" << color_string << "\">";
- }
-
- fp << test_fitness << " ";
-
- // Fraction Columns...
- if (file_type == FILE_TYPE_HTML) fp << "<th bgcolor=\"#FF0000\">";
- fp << (double) row_dead / (double) (num_insts-1) << " ";
-
- if (file_type == FILE_TYPE_HTML) fp << "<th bgcolor=\"#FFFF00\">";
- fp << (double) row_neg / (double) (num_insts-1) << " ";
-
- if (file_type == FILE_TYPE_HTML) fp << "<th bgcolor=\"#FFFFFF\">";
- fp << (double) row_neut / (double) (num_insts-1) << " ";
-
- if (file_type == FILE_TYPE_HTML) fp << "<th bgcolor=\"#00FF00\">";
- fp << (double) row_pos / (double) (num_insts-1) << " ";
-
-
- // Column: Average Fitness
- if (file_type == FILE_TYPE_HTML) fp << "<th>";
- fp << row_fitness / (double) (num_insts-1) << " ";
-
- // Column: Expected Entropy @CAO Implement!
- if (file_type == FILE_TYPE_HTML) fp << "<th>";
- fp << 0.0 << " ";
-
- // End this row...
- if (file_type == FILE_TYPE_HTML) fp << "</tr>";
- fp << endl;
-
- // Reset the mod_genome back to the original sequence.
- mod_genome[line_num].SetOp(cur_inst);
- }
-
-
- // Construct the final line of the table with all totals...
- if (file_type == FILE_TYPE_HTML) {
- fp << "<tr><th>Totals";
-
- // Instructions + Knockout
- for (int i = 0; i <= num_insts; i++) {
- fp << "<th>" << col_fitness[i] / max_line << " ";
- }
-
- int total_tests = max_line * (num_insts-1);
- fp << "<th>" << (double) total_dead / (double) total_tests << " ";
- fp << "<th>" << (double) total_neg / (double) total_tests << " ";
- fp << "<th>" << (double) total_neut / (double) total_tests << " ";
- fp << "<th>" << (double) total_pos / (double) total_tests << " ";
- fp << "<th>" << total_fitness / (double) total_tests << " ";
- fp << "<th>" << 0.0 << " ";
-
-
- // And close everything up...
- fp << "</table>" << endl
- << "</center>" << endl;
- }
-
- }
- }
-
-
-void cAnalyze::CommandMapDepth(cString cur_string)
-{
- cout << "Constructing depth map..." << endl;
-
- cString filename("depth_map.dat");
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
-
- int min_batch = 0;
- int max_batch = cur_batch - 1;
-
- if (cur_string.GetSize() != 0) min_batch = cur_string.PopWord().AsInt();
- if (cur_string.GetSize() != 0) max_batch = cur_string.PopWord().AsInt();
-
- // First, scan all of the batches to find the maximum depth.
- int max_depth = -1;
- cAnalyzeGenotype * genotype;
- for (int i = min_batch; i <= max_batch; i++) {
- tListIterator<cAnalyzeGenotype> list_it(batch[i].List());
- while ((genotype = list_it.Next()) != NULL) {
- if (genotype->GetDepth() > max_depth) max_depth = genotype->GetDepth();
- }
- }
-
- cout << "max_depth = " << max_depth << endl;
-
- ofstream & fp = data_file_manager.GetOFStream(filename);
-
- cout << "Output to " << filename << endl;
- tArray<int> depth_array(max_depth+1);
- for (cur_batch = min_batch; cur_batch <= max_batch; cur_batch++) {
- depth_array.SetAll(0);
- tListIterator<cAnalyzeGenotype> list_it(batch[cur_batch].List());
- while ((genotype = list_it.Next()) != NULL) {
- const int cur_depth = genotype->GetDepth();
- const int cur_count = genotype->GetNumCPUs();
- depth_array[cur_depth] += cur_count;
- }
-
- for (int i = 0; i <= max_depth; i++) {
- fp << depth_array[i] << " ";
- }
- fp << endl;
- }
-}
-
-void cAnalyze::CommandHamming(cString cur_string)
-{
- cString filename("hamming.dat");
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
-
- int batch1 = PopBatch(cur_string.PopWord());
- int batch2 = PopBatch(cur_string.PopWord());
-
- // We want batch2 to be the larger one for efficiency...
- if (batch[batch1].List().GetSize() > batch[batch2].List().GetSize()) {
- int tmp = batch1; batch1 = batch2; batch2 = tmp;
- }
-
- if (verbose == false) {
- cout << "Calculating Hamming Distance... ";
- cout.flush();
- } else {
- cout << "Calculating Hamming Distance between batches "
- << batch1 << " and " << batch2 << endl;
- cout.flush();
- }
-
- // Setup some variables;
- cAnalyzeGenotype * genotype1 = NULL;
- cAnalyzeGenotype * genotype2 = NULL;
- int total_dist = 0;
- int total_count = 0;
-
- tListIterator<cAnalyzeGenotype> list1_it(batch[batch1].List());
- tListIterator<cAnalyzeGenotype> list2_it(batch[batch2].List());
-
- while ((genotype1 = list1_it.Next()) != NULL) {
- list2_it.Reset();
- while ((genotype2 = list2_it.Next()) != NULL) {
- // Determine the counts...
- const int count1 = genotype1->GetNumCPUs();
- const int count2 = genotype2->GetNumCPUs();
- const int num_pairs = (genotype1 == genotype2) ?
- ((count1 - 1) * (count2 - 1)) : (count1 * count2);
- if (num_pairs == 0) continue;
-
- // And do the tests...
- const int dist =
- cGenomeUtil::FindHammingDistance(genotype1->GetGenome(),
- genotype2->GetGenome());
- total_dist += dist * num_pairs;
- total_count += num_pairs;
- }
- }
-
-
- // Calculate the final answer
- double ave_dist = (double) total_dist / (double) total_count;
- cout << " ave distance = " << ave_dist << endl;
-
- cDataFile & df = data_file_manager.Get(filename);
-
- df.WriteComment( "Hamming distance information" );
- df.WriteTimeStamp();
-
- df.Write(batch[batch1].Name(), "Name of First Batch");
- df.Write(batch[batch2].Name(), "Name of Second Batch");
- df.Write(ave_dist, "Average Hamming Distance");
- df.Write(total_count, "Total Pairs Test");
- df.Endl();
-}
-
-void cAnalyze::CommandLevenstein(cString cur_string)
-{
- cString filename("lev.dat");
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
-
- int batch1 = PopBatch(cur_string.PopWord());
- int batch2 = PopBatch(cur_string.PopWord());
-
- // We want batch2 to be the larger one for efficiency...
- if (batch[batch1].List().GetSize() > batch[batch2].List().GetSize()) {
- int tmp = batch1; batch1 = batch2; batch2 = tmp;
- }
-
- if (verbose == false) {
- cout << "Calculating Levenstein Distance... ";
- cout.flush();
- } else {
- cout << "Calculating Levenstein Distance between batch "
- << batch1 << " and " << batch2 << endl;
- cout.flush();
- }
-
- // Setup some variables;
- cAnalyzeGenotype * genotype1 = NULL;
- cAnalyzeGenotype * genotype2 = NULL;
- int total_dist = 0;
- int total_count = 0;
-
- tListIterator<cAnalyzeGenotype> list1_it(batch[batch1].List());
- tListIterator<cAnalyzeGenotype> list2_it(batch[batch2].List());
-
- // Loop through all of the genotypes in each batch...
- while ((genotype1 = list1_it.Next()) != NULL) {
- list2_it.Reset();
- while ((genotype2 = list2_it.Next()) != NULL) {
- // Determine the counts...
- const int count1 = genotype1->GetNumCPUs();
- const int count2 = genotype2->GetNumCPUs();
- const int num_pairs = (genotype1 == genotype2) ?
- ((count1 - 1) * (count2 - 1)) : (count1 * count2);
- if (num_pairs == 0) continue;
-
- // And do the tests...
- const int dist = cGenomeUtil::FindEditDistance(genotype1->GetGenome(),
- genotype2->GetGenome());
- total_dist += dist * num_pairs;
- total_count += num_pairs;
- }
- }
-
- // Calculate the final answer
- double ave_dist = (double) total_dist / (double) total_count;
- cout << " ave distance = " << ave_dist << endl;
-
- cDataFile & df = data_file_manager.Get(filename);
-
- df.WriteComment( "Levenstein distance information" );
- df.WriteTimeStamp();
-
- df.Write(batch[batch1].Name(), "Name of First Batch");
- df.Write(batch[batch2].Name(), "Name of Second Batch");
- df.Write(ave_dist, "Average Levenstein Distance");
- df.Write(total_count, "Total Pairs Test");
- df.Endl();
-}
-
-void cAnalyze::CommandSpecies(cString cur_string)
-{
- cString filename("species.dat");
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
-
- int batch1 = PopBatch(cur_string.PopWord());
- int batch2 = PopBatch(cur_string.PopWord());
- int num_compare = PopBatch(cur_string.PopWord());
-
- // We want batch2 to be the larger one for efficiency...
- if (batch[batch1].List().GetSize() > batch[batch2].List().GetSize()) {
- int tmp = batch1; batch1 = batch2; batch2 = tmp;
- }
-
- if (verbose == false) cout << "Calculating Species Distance... " << endl;
- else cout << "Calculating Species Distance between batch "
- << batch1 << " and " << batch2 << endl;
-
- // Setup some variables;
- cAnalyzeGenotype * genotype1 = NULL;
- cAnalyzeGenotype * genotype2 = NULL;
- int total_fail = 0;
- int total_count = 0;
-
- tListIterator<cAnalyzeGenotype> list1_it(batch[batch1].List());
- tListIterator<cAnalyzeGenotype> list2_it(batch[batch2].List());
-
- // Loop through all of the genotypes in each batch...
- while ((genotype1 = list1_it.Next()) != NULL) {
- list2_it.Reset();
- while ((genotype2 = list2_it.Next()) != NULL) {
- // Determine the counts...
- const int count1 = genotype1->GetNumCPUs();
- const int count2 = genotype2->GetNumCPUs();
- int num_pairs = count1 * count2;
- int fail_count = 0;
- bool cross1_viable = true;
- bool cross2_viable = true;
-
-
- if (genotype1 == genotype2) {
- total_count += num_pairs * 2 * num_compare;
- }
- else {
- assert(num_compare!=0);
- // And do the tests...
- for (int iter=1; iter < num_compare; iter++) {
- cCPUMemory test_genome0 = genotype1->GetGenome();
- cCPUMemory test_genome1 = genotype2->GetGenome();
-
- double start_frac = m_world->GetRandom().GetDouble();
- double end_frac = m_world->GetRandom().GetDouble();
- if (start_frac > end_frac) nFunctions::Swap(start_frac, end_frac);
-
- int start0 = (int) (start_frac * (double) test_genome0.GetSize());
- int end0 = (int) (end_frac * (double) test_genome0.GetSize());
- int start1 = (int) (start_frac * (double) test_genome1.GetSize());
- int end1 = (int) (end_frac * (double) test_genome1.GetSize());
- assert( start0 >= 0 && start0 < test_genome0.GetSize() );
- assert( end0 >= 0 && end0 < test_genome0.GetSize() );
- assert( start1 >= 0 && start1 < test_genome1.GetSize() );
- assert( end1 >= 0 && end1 < test_genome1.GetSize() );
-
- // Calculate size of sections crossing over...
- int size0 = end0 - start0;
- int size1 = end1 - start1;
-
- int new_size0 = test_genome0.GetSize() - size0 + size1;
- int new_size1 = test_genome1.GetSize() - size1 + size0;
-
- // Don't Crossover if offspring will be illegal!!!
- if (new_size0 < MIN_CREATURE_SIZE || new_size0 > MAX_CREATURE_SIZE ||
- new_size1 < MIN_CREATURE_SIZE || new_size1 > MAX_CREATURE_SIZE) {
- fail_count +=2;
- break;
- }
-
- // Swap the components
- cGenome cross0 = cGenomeUtil::Crop(test_genome0, start0, end0);
- cGenome cross1 = cGenomeUtil::Crop(test_genome1, start1, end1);
- test_genome0.Replace(start0, size0, cross1);
- test_genome1.Replace(start1, size1, cross0);
-
- // Run each side, and determine viability...
- cCPUTestInfo test_info;
- m_world->GetTestCPU().TestGenome(test_info, test_genome0);
- cross1_viable = test_info.IsViable();
-
- m_world->GetTestCPU().TestGenome(test_info, test_genome1);
- cross2_viable = test_info.IsViable();
-
- if (cross1_viable == false) fail_count++;
- if (cross2_viable == false) fail_count++;
- }
-
- total_fail += fail_count * num_pairs;
- total_count += num_pairs * 2 * num_compare;
- }
- }
- }
- // Calculate the final answer
- double ave_dist = (double) total_fail / (double) total_count;
- cout << " ave distance = " << ave_dist
- << " in " << total_count
- << " tests." << endl;
-
- cDataFile & df = data_file_manager.Get(filename);
-
- df.WriteComment( "Species information" );
- df.WriteTimeStamp();
-
- df.Write(batch[batch1].Name(), "Name of First Batch");
- df.Write(batch[batch2].Name(), "Name of Second Batch");
- df.Write(ave_dist, "Average Species Distance");
- df.Write(total_count, "Total Recombinants tested");
- df.Endl();
-}
-
-void cAnalyze::CommandRecombine(cString cur_string)
-{
- int batch1 = PopBatch(cur_string.PopWord());
- int batch2 = PopBatch(cur_string.PopWord());
- int batch3 = PopBatch(cur_string.PopWord());
- int num_compare = PopBatch(cur_string.PopWord());
-
- // We want batch2 to be the larger one for efficiency...
- if (batch[batch1].List().GetSize() > batch[batch2].List().GetSize()) {
- int tmp = batch1; batch1 = batch2; batch2 = tmp;
- }
-
- if (verbose == false) cout << "Creating recombinants... " << endl;
- else cout << "Creating recombinants between batch "
- << batch1 << " and " << batch2 << endl;
-
- // Setup some variables;
- cAnalyzeGenotype * genotype1 = NULL;
- cAnalyzeGenotype * genotype2 = NULL;
-
- tListIterator<cAnalyzeGenotype> list1_it(batch[batch1].List());
- tListIterator<cAnalyzeGenotype> list2_it(batch[batch2].List());
-
- // Loop through all of the genotypes in each batch...
- while ((genotype1 = list1_it.Next()) != NULL) {
- list2_it.Reset();
- while ((genotype2 = list2_it.Next()) != NULL) {
- // Determine the counts...
- int fail_count = 0;
-
-
- assert(num_compare!=0);
- // And do the tests...
- for (int iter=1; iter < num_compare; iter++) {
- cCPUMemory test_genome0 = genotype1->GetGenome();
- cCPUMemory test_genome1 = genotype2->GetGenome();
-
- double start_frac = m_world->GetRandom().GetDouble();
- double end_frac = m_world->GetRandom().GetDouble();
- if (start_frac > end_frac) nFunctions::Swap(start_frac, end_frac);
-
- int start0 = (int) (start_frac * (double) test_genome0.GetSize());
- int end0 = (int) (end_frac * (double) test_genome0.GetSize());
- int start1 = (int) (start_frac * (double) test_genome1.GetSize());
- int end1 = (int) (end_frac * (double) test_genome1.GetSize());
- assert( start0 >= 0 && start0 < test_genome0.GetSize() );
- assert( end0 >= 0 && end0 < test_genome0.GetSize() );
- assert( start1 >= 0 && start1 < test_genome1.GetSize() );
- assert( end1 >= 0 && end1 < test_genome1.GetSize() );
-
- // Calculate size of sections crossing over...
- int size0 = end0 - start0;
- int size1 = end1 - start1;
-
- int new_size0 = test_genome0.GetSize() - size0 + size1;
- int new_size1 = test_genome1.GetSize() - size1 + size0;
-
- // Don't Crossover if offspring will be illegal!!!
- if (new_size0 < MIN_CREATURE_SIZE || new_size0 > MAX_CREATURE_SIZE ||
- new_size1 < MIN_CREATURE_SIZE || new_size1 > MAX_CREATURE_SIZE) {
- fail_count +=2;
- break;
- }
-
- if (size0 > 0 && size1 > 0) {
- cGenome cross0 = cGenomeUtil::Crop(test_genome0, start0, end0);
- cGenome cross1 = cGenomeUtil::Crop(test_genome1, start1, end1);
- test_genome0.Replace(start0, size0, cross1);
- test_genome1.Replace(start1, size1, cross0);
- }
- else if (size0 > 0) {
- cGenome cross0 = cGenomeUtil::Crop(test_genome0, start0, end0);
- test_genome1.Replace(start1, size1, cross0);
- }
- else if (size1 > 0) {
- cGenome cross1 = cGenomeUtil::Crop(test_genome1, start1, end1);
- test_genome0.Replace(start0, size0, cross1);
- }
-
- cAnalyzeGenotype * new_genotype0 = new cAnalyzeGenotype(m_world, test_genome0, inst_set);
- cAnalyzeGenotype * new_genotype1 = new cAnalyzeGenotype(m_world, test_genome1, inst_set);
- new_genotype0->SetNumCPUs(1);
- new_genotype1->SetNumCPUs(1);
- new_genotype0->SetID(0);
- new_genotype1->SetID(0);
- new_genotype0->SetName("noname");
- new_genotype1->SetName("noname");
-
- batch[batch3].List().PushRear(new_genotype0);
- batch[batch3].List().PushRear(new_genotype1);
-
- //batch[batch3].List().PushRear(new cAnalyzeGenotype(test_genome0, inst_set));
- //batch[batch3].List().PushRear(new cAnalyzeGenotype(test_genome1, inst_set));
-
- }
- }
- }
-}
-
-void cAnalyze::CommandAlign(cString cur_string)
-{
- // Align does not need any args yet.
- (void) cur_string;
-
- cout << "Aligning sequences..." << endl;
-
- if (batch[cur_batch].IsLineage() == false && verbose == true) {
- cerr << " Warning: sequences may not be a consecutive lineage."
- << endl;
- }
-
- // Create an array of all the sequences we need to align.
- tListPlus<cAnalyzeGenotype> & glist = batch[cur_batch].List();
- tListIterator<cAnalyzeGenotype> batch_it(glist);
- const int num_sequences = glist.GetSize();
- cString * sequences = new cString[num_sequences];
-
- // Move through each sequence an update it.
- batch_it.Reset();
- cString diff_info;
- for (int i = 0; i < num_sequences; i++) {
- sequences[i] = batch_it.Next()->GetGenome().AsString();
- if (i == 0) continue;
- // Track of the number of insertions and deletions to shift properly.
- int num_ins = 0;
- int num_del = 0;
-
- // Compare each string to the previous.
- cStringUtil::EditDistance(sequences[i], sequences[i-1], diff_info, '_');
- while (diff_info.GetSize() != 0) {
- cString cur_mut = diff_info.Pop(',');
- const char mut_type = cur_mut[0];
- cur_mut.ClipFront(1); cur_mut.ClipEnd(1);
- int position = cur_mut.AsInt();
-
- // Nothing to do with Mutations
- if (mut_type == 'M') continue;
-
- // Handle insertions...
- if (mut_type == 'I') {
- // Loop back and insert an '_' into all previous sequences.
- for (int j = 0; j < i; j++) {
- sequences[j].Insert('_', position + num_del);
- }
- num_ins++;
- }
-
- // Handle Deletions...
- else if (mut_type == 'D') {
- // Insert '_' into the current sequence at the point of deletions.
- sequences[i].Insert("_", position + num_ins);
- num_del++;
- }
-
- }
- }
-
- batch_it.Reset();
- for (int i = 0; i < num_sequences; i++) {
- batch_it.Next()->SetAlignedSequence(sequences[i]);
- }
-
- // Cleanup
- delete [] sequences;
-
- // Adjust the flags on this batch
- // batch[cur_batch].SetLineage(false);
- batch[cur_batch].SetAligned(true);
-}
-
-// Now this command do not consider changing environment
-// and only work for lineage and fixed-length runs.
-void cAnalyze::AnalyzeNewInfo(cString cur_string)
-{
- cout << "Analyze new information in child about environment ..." << endl;
-
- // Load in the variables
- int words = cur_string.CountNumWords();
- if (words < 1) {
- cout << "This command requires mutation rate, skipping." << endl;
- return;
- }
-
- // Get the mutation rate ...
- double mu = cur_string.PopWord().AsDouble();
-
- // Create the directory using the string given as the second argument
- cString dir = cur_string.PopWord();
- cString defaultDir = "newinfo/";
- cString directory = PopDirectory(dir, defaultDir);
-
- ///////////////////////////////////////////////////////
- // Loop through all of the genotypes in this batch...
-
- cString newinfo_fn;
- ofstream newinfo_fp;
- if (batch[cur_batch].IsLineage() == true) {
- newinfo_fn.Set("%s%s.newinfo.dat", directory(), "lineage");
- newinfo_fp.open(newinfo_fn);
- } else {
- cout << "This command requires the lineage in the batch, skipping.\n";
- return;
- }
- newinfo_fp << "# Legend:" << endl;
- newinfo_fp << "# 1:Child Genotype ID" << endl;
- newinfo_fp << "# 2:Parent Genotype ID" << endl;
- newinfo_fp << "# 3:Information of Child about Environment I(C:E)" << endl;
- newinfo_fp << "# 4:Information of Parent about Environment I(P:E)" << endl;
- newinfo_fp << "# 5:I(C:E)-I(P:E)" << endl;
- newinfo_fp << "# 6:Information Gained in Child" << endl;
- newinfo_fp << "# 7:Information Decreased in Child" << endl;
- newinfo_fp << "# 8:Net Increasing of Information in Child" << endl;
- newinfo_fp << endl;
-
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * parent_genotype = batch_it.Next();
- if (parent_genotype == NULL) {
- newinfo_fp.close();
- return;
- }
- cAnalyzeGenotype * child_genotype = NULL;
- double I_P_E; // Information of parent about environment
- double H_P_E = AnalyzeEntropy(parent_genotype, mu);
- I_P_E = parent_genotype->GetLength() - H_P_E;
-
- while ((child_genotype = batch_it.Next()) != NULL) {
-
- if (verbose == true) {
- cout << "Analyze new information for " << child_genotype->GetName() << endl;
- }
-
- // Information of parent about its environment should not be zero.
- if (I_P_E == 0) {
- cerr << "Error: Information between parent and its enviroment is zero."
- << "(cAnalyze::AnalyzeNewInfo)" << endl;
- exit(1);
- }
-
- double H_C_E = AnalyzeEntropy(child_genotype, mu);
- double I_C_E = child_genotype->GetLength() - H_C_E;
- double net_gain = I_C_E - I_P_E;
-
- // Increased information in child compared to parent
- double child_increased_info = IncreasedInfo(child_genotype, parent_genotype, mu);
-
- // Lost information in child compared to parent
- double child_lost_info = IncreasedInfo(parent_genotype, child_genotype, mu);
-
- // Write information to file ...
- newinfo_fp << child_genotype->GetID() << " ";
- newinfo_fp << parent_genotype->GetID() << " ";
- newinfo_fp << I_C_E << " ";
- newinfo_fp << I_P_E << " ";
- newinfo_fp << net_gain << " ";
- newinfo_fp << child_increased_info << " ";
- newinfo_fp << child_lost_info << " ";
- newinfo_fp << child_increased_info - child_lost_info << endl;
-
- parent_genotype = child_genotype;
- I_P_E = I_C_E;
- }
-
- newinfo_fp.close();
- return;
-}
-
-
-
-void cAnalyze::WriteClone(cString cur_string)
-{
- // Load in the variables...
- cString filename("clone.dat");
- int num_cells = -1;
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
- if (cur_string.GetSize() != 0) num_cells = cur_string.PopWord().AsInt();
-
-
- ofstream & fp = data_file_manager.GetOFStream(filename);
-
- // Start up again at update zero...
- fp << "0 ";
-
- // Setup the genebank sizes of lists to all be zero.
- fp << MAX_CREATURE_SIZE << " ";
- for (int i = 0; i < MAX_CREATURE_SIZE; i++) {
- fp << "0 ";
- }
-
- // Save the individual genotypes
- fp << batch[cur_batch].List().GetSize() << " ";
-
- int org_count = 0;
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- org_count += genotype->GetNumCPUs();
- const int length = genotype->GetLength();
- const cGenome & genome = genotype->GetGenome();
-
- fp << genotype->GetID() << " "
- << length << " ";
-
- for (int i = 0; i < length; i++) {
- fp << genome[i].GetOp() << " ";
- }
- }
-
- // Write out the current state of the grid.
-
- if (num_cells == 0) num_cells = org_count;
- fp << num_cells << " ";
-
- batch_it.Reset();
- while ((genotype = batch_it.Next()) != NULL) {
- for (int i = 0; i < genotype->GetNumCPUs(); i++) {
- fp << genotype->GetID() << " ";
- }
- }
-
- // Fill out the remainder of the grid with -1
- for (int i = org_count; i < num_cells; i++) {
- fp << "-1 ";
- }
-}
-
-
-void cAnalyze::WriteInjectEvents(cString cur_string)
-{
- // Load in the variables...
- cString filename("events_inj.cfg");
- int start_cell = 0;
- int lineage = 0;
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
- if (cur_string.GetSize() != 0) start_cell = cur_string.PopWord().AsInt();
- if (cur_string.GetSize() != 0) lineage = cur_string.PopWord().AsInt();
-
- ofstream & fp = data_file_manager.GetOFStream(filename);
-
- int org_count = 0;
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- const int cur_count = genotype->GetNumCPUs();
- org_count += cur_count;
- const cGenome & genome = genotype->GetGenome();
-
- fp << "u 0 inject_sequence "
- << genome.AsString() << " "
- << start_cell << " "
- << start_cell + cur_count << " "
- << genotype->GetMerit() << " "
- << lineage << " "
- << endl;
- start_cell += cur_count;
- }
-}
-
-
-void cAnalyze::WriteCompetition(cString cur_string)
-{
- cout << "Writing Competition events..." << endl;
-
- // Load in the variables...
- int join_UD = 0;
- double start_merit = 50000;
- cString filename("events_comp.cfg");
- int batch_A = cur_batch - 1;
- int batch_B = cur_batch;
- int grid_side = -1;
- int lineage = 0;
-
- // Make sure we have reasonable default batches.
- if (cur_batch == 0) { batch_A = 0; batch_B = 1; }
-
- if (cur_string.GetSize() != 0) join_UD = cur_string.PopWord().AsInt();
- if (cur_string.GetSize() != 0) start_merit = cur_string.PopWord().AsDouble();
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
- if (cur_string.GetSize() != 0) batch_A = cur_string.PopWord().AsInt();
- if (cur_string.GetSize() != 0) batch_B = cur_string.PopWord().AsInt();
- if (cur_string.GetSize() != 0) grid_side = cur_string.PopWord().AsInt();
- if (cur_string.GetSize() != 0) lineage = cur_string.PopWord().AsInt();
-
- // Check inputs...
- if (join_UD < 0) join_UD = 0;
- if (batch_A < 0 || batch_B < 0) {
- cerr << "Error: Batch IDs must be positive!" << endl;
- return;
- }
-
- ofstream & fp = data_file_manager.GetOFStream(filename);
-
- // Count the number of organisms in each batch...
- cAnalyzeGenotype * genotype = NULL;
-
- int org_count_A = 0;
- tListIterator<cAnalyzeGenotype> batchA_it(batch[batch_A].List());
- while ((genotype = batchA_it.Next()) != NULL) {
- org_count_A += genotype->GetNumCPUs();
- }
-
- int org_count_B = 0;
- tListIterator<cAnalyzeGenotype> batchB_it(batch[batch_B].List());
- while ((genotype = batchB_it.Next()) != NULL) {
- org_count_B += genotype->GetNumCPUs();
- }
-
- int max_count = Max(org_count_A, org_count_B);
- if (max_count > 10000) {
- cout << "Warning: more than 10,000 organisms in sub-population!" << endl;
- }
-
- if (grid_side <= 0) {
- for (grid_side = 5; grid_side < 100; grid_side += 5) {
- if (grid_side * grid_side >= max_count) break;
- }
- if (verbose == true) {
- cout << "...assuming population size "
- << grid_side << "x" << grid_side << "." << endl;
- }
- }
-
-
- int pop_size = grid_side * grid_side;
-
- int inject_pos = 0;
- while ((genotype = batchA_it.Next()) != NULL) {
- const int cur_count = genotype->GetNumCPUs();
- const cGenome & genome = genotype->GetGenome();
- double cur_merit = start_merit;
- if (cur_merit < 0) cur_merit = genotype->GetMerit();
- fp << "u 0 inject_sequence "
- << genome.AsString() << " "
- << inject_pos << " "
- << inject_pos + cur_count << " "
- << cur_merit << " "
- << lineage << " "
- << endl;
- inject_pos += cur_count;
- }
-
- inject_pos = pop_size;
- while ((genotype = batchB_it.Next()) != NULL) {
- const int cur_count = genotype->GetNumCPUs();
- const cGenome & genome = genotype->GetGenome();
- double cur_merit = start_merit;
- if (cur_merit < 0) cur_merit = genotype->GetMerit();
- fp << "u 0 inject_sequence "
- << genome.AsString() << " "
- << inject_pos << " "
- << inject_pos + cur_count << " "
- << cur_merit << " "
- << lineage+1 << " "
- << endl;
- inject_pos += cur_count;
- }
-
- fp << "u 0 sever_grid_row" << grid_side << endl;
- fp << "u " << join_UD << " join_grid_row " << grid_side << endl;
-}
-
-
-void cAnalyze::AnalyzeMuts(cString cur_string)
-{
- cout << "Analyzing Mutations" << endl;
-
- // Make sure we have everything we need.
- if (batch[cur_batch].IsAligned() == false) {
- cout << " Error: sequences not aligned." << endl;
- return;
- }
-
- // Setup variables...
- cString filename("analyze_muts.dat");
- bool all_combos = false;
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
- if (cur_string.GetSize() != 0) all_combos = cur_string.PopWord().AsInt();
-
- tListPlus<cAnalyzeGenotype> & gen_list = batch[cur_batch].List();
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
-
- const int num_sequences = gen_list.GetSize();
- const int sequence_length =
- gen_list.GetFirst()->GetAlignedSequence().GetSize();
- cString * sequences = new cString[num_sequences];
- int * mut_count = new int[sequence_length];
- for (int i = 0; i < sequence_length; i++) mut_count[i] = 0;
-
- // Load in the sequences
- batch_it.Reset();
- int count = 0;
- while (batch_it.Next() != NULL) {
- sequences[count] = batch_it.Get()->GetAlignedSequence();
- count++;
- }
-
- // Count the number of changes at each site...
- for (int i = 1; i < num_sequences; i++) { // For each pair...
- cString & seq1 = sequences[i-1];
- cString & seq2 = sequences[i];
- for (int j = 0; j < sequence_length; j++) { // For each site...
- if (seq1[j] != seq2[j]) mut_count[j]++;
- }
- }
-
- // Grab the two strings we're actively going to be working with.
- cString & first_seq = sequences[0];
- cString & last_seq = sequences[num_sequences - 1];
-
- // Print out the header...
- ofstream & fp = data_file_manager.GetOFStream(filename);
- fp << "# " << sequences[0] << endl;
- fp << "# " << sequences[num_sequences - 1] << endl;
- fp << "# ";
- for (int i = 0; i < sequence_length; i++) {
- if (mut_count[i] == 0) fp << " ";
- else if (mut_count[i] > 9) fp << "+";
- else fp << mut_count[i];
- }
- fp << endl;
- fp << "# ";
- for (int i = 0; i < sequence_length; i++) {
- if (first_seq[i] == last_seq[i]) fp << " ";
- else fp << "^";
- }
- fp << endl << endl;
-
- // Count the number of diffs between the two strings we're interested in.
- const int total_diffs = cStringUtil::Distance(first_seq, last_seq);
- if (verbose) cout << " " << total_diffs << " mutations being tested."
- << endl;
-
- // Locate each difference.
- int * mut_positions = new int[total_diffs];
- int cur_mut = 0;
- for (int i = 0; i < first_seq.GetSize(); i++) {
- if (first_seq[i] != last_seq[i]) {
- mut_positions[cur_mut] = i;
- cur_mut++;
- }
- }
-
- // The number of mutations we need to deal with will tell us how much
- // we can attempt to do. (@CAO should be able to overide defaults)
- bool scan_combos = true; // Scan all possible combos of mutations?
- bool detail_muts = true; // Collect detailed info on all mutations?
- bool print_all = true; // Print everything we collect without digestion?
- if (total_diffs > 30) scan_combos = false;
- if (total_diffs > 20) detail_muts = false;
- if (total_diffs > 10) print_all = false;
-
- // Start moving through the difference combinations...
- if (scan_combos) {
- const int total_combos = 1 << total_diffs;
- cout << " Scanning through " << total_combos << " combos." << endl;
-
- double * total_fitness = new double[total_diffs + 1];
- double * total_sqr_fitness = new double[total_diffs + 1];
- double * max_fitness = new double[total_diffs + 1];
- cString * max_sequence = new cString[total_diffs + 1];
- int * test_count = new int[total_diffs + 1];
- for (int i = 0; i <= total_diffs; i++) {
- total_fitness[i] = 0.0;
- total_sqr_fitness[i] = 0.0;
- max_fitness[i] = 0.0;
- test_count[i] = 0;
- }
-
- // Loop through all of the combos...
- const int combo_step = total_combos / 79;
- for (int combo_id = 0; combo_id < total_combos; combo_id++) {
- if (combo_id % combo_step == 0) {
- cout << '.';
- cout.flush();
- }
- // Start at the first sequence and add needed changes...
- cString test_sequence = first_seq;
- int diff_count = 0;
- for (int mut_id = 0; mut_id < total_diffs; mut_id++) {
- if ((combo_id >> mut_id) & 1) {
- const int cur_pos = mut_positions[mut_id];
- test_sequence[cur_pos] = last_seq.GetData()[cur_pos];
- diff_count++;
- }
- }
-
- // Determine the fitness of the current sequence...
- cGenome test_genome(test_sequence);
- cCPUTestInfo test_info;
- test_info.TestThreads();
- m_world->GetTestCPU().TestGenome(test_info, test_genome);
- const double fitness = test_info.GetGenotypeFitness();
-
- //cAnalyzeGenotype test_genotype(test_sequence);
- //test_genotype.Recalculate();
- //const double fitness = test_genotype.GetFitness();
-
- total_fitness[diff_count] += fitness;
- total_sqr_fitness[diff_count] += fitness * fitness;
- if (fitness > max_fitness[diff_count]) {
- max_fitness[diff_count] = fitness;
- max_sequence[diff_count] = test_sequence;
- // cout << endl
- // << max_sequence[diff_count] << " "
- // << test_info.GetGenotypeMerit() << " "
- // << fitness << " "
- // << combo_id << endl;
- }
- test_count[diff_count]++;
- }
-
- // Output the results...
-
- for (int i = 0; i <= total_diffs; i++) {
- cAnalyzeGenotype max_genotype(m_world, max_sequence[i], inst_set);
- max_genotype.Recalculate();
- fp << i << " " // 1
- << test_count[i] << " " // 2
- << total_fitness[i] / (double) test_count[i] << " " // 3
- << max_fitness[i] << " " // 4
- << max_genotype.GetMerit() << " " // 5
- << max_genotype.GetGestTime() << " " // 6
- << max_genotype.GetLength() << " " // 7
- << max_genotype.GetCopyLength() << " " // 8
- << max_genotype.GetExeLength() << " "; // 9
- max_genotype.PrintTasks(fp, 3,12);
- fp << max_sequence[i] << endl;
- }
-
- // Cleanup
- delete [] total_fitness;
- delete [] total_sqr_fitness;
- delete [] max_fitness;
- delete [] max_sequence;
- delete [] test_count;
- }
- // If we can't scan through all combos, give wanring.
- else {
- cerr << " Warning: too many mutations (" << total_diffs
- << ") to scan through combos." << endl;
- }
-
-
- // Cleanup...
- delete [] sequences;
- delete [] mut_count;
- delete [] mut_positions;
-}
-
-void cAnalyze::AnalyzeInstructions(cString cur_string)
-{
- if (verbose == true) {
- cout << "Analyzing Instructions in batch " << cur_batch << endl;
- }
- else cout << "Analyzeing Instructions..." << endl;
-
- // Load in the variables...
- cString filename("inst_analyze.dat");
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
- const int num_insts = inst_set.GetSize();
-
- // Setup the file...
- ofstream fp(filename);
-
- // Determine the file type...
- int file_type = FILE_TYPE_TEXT;
- while (filename.Find('.') != -1) filename.Pop('.');
- if (filename == "html") file_type = FILE_TYPE_HTML;
-
- // If we're in HTML mode, setup the header...
- if (file_type == FILE_TYPE_HTML) {
- // Document header...
- fp << "<html>" << endl
- << "<body bgcolor=\"#FFFFFF\"" << endl
- << " text=\"#000000\"" << endl
- << " link=\"#0000AA\"" << endl
- << " alink=\"#0000FF\"" << endl
- << " vlink=\"#000044\">" << endl
- << endl
- << "<h1 align=center>Instruction Chart: "
- << batch[cur_batch].Name() << endl
- << "<br><br>" << endl
- << endl;
-
- // Instruction key...
- const int num_cols = 6;
- const int num_rows = ((num_insts - 1) / num_cols) + 1;
- fp << "<table border=2 cellpadding=3>" << endl
- << "<tr bgcolor=\"#AAAAFF\"><th colspan=6>Instruction Set Legend</tr>"
- << endl;
- for (int i = 0; i < num_rows; i++) {
- fp << "<tr>";
- for (int j = 0; j < num_cols; j++) {
- const int inst_id = i + j * num_rows;
- if (inst_id < num_insts) {
- cInstruction cur_inst(inst_id);
- fp << "<td><b>" << cur_inst.GetSymbol() << "</b> : "
- << inst_set.GetName(inst_id) << " ";
- }
- else {
- fp << "<td> ";
- }
- }
- fp << "</tr>" << endl;
- }
- fp << "</table>" << endl
- << "<br><br><br>" << endl;
-
- // Main table header...
- fp << "<center>" << endl
- << "<table border=1 cellpadding=2>" << endl
- << "<tr><th bgcolor=\"#AAAAFF\">Run # <th bgcolor=\"#AAAAFF\">Length"
- << endl;
- for (int i = 0; i < num_insts; i++) {
- cInstruction cur_inst(i);
- fp << "<th bgcolor=\"#AAAAFF\">" << cur_inst.GetSymbol() << " ";
- }
- fp << "</tr>" << endl;
- }
- else { // if (file_type == FILE_TYPE_TEXT) {
- fp << "#RUN_NAME LENGTH ";
- for (int i = 0; i < num_insts; i++) {
- cInstruction cur_inst(i);
- fp << cur_inst.GetSymbol() << ":" << inst_set.GetName(i) << " ";
- }
- fp << endl;
- }
-
- // Figure out how often we expect each instruction to appear...
- const double exp_freq = 1.0 / (double) num_insts;
- const double min_freq = exp_freq * 0.5;
- const double max_freq = exp_freq * 2.0;
-
- double total_length = 0.0;
- tArray<double> total_freq(num_insts);
- for (int i = 0; i < num_insts; i++) total_freq[i] = 0.0;
-
- // Loop through all of the genotypes in this batch...
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- // Setup for counting...
- tArray<int> inst_bin(num_insts);
- for (int i = 0; i < num_insts; i++) inst_bin[i] = 0;
-
- // Count it up!
- const int genome_size = genotype->GetLength();
- for (int i = 0; i < genome_size; i++) {
- const int inst_id = genotype->GetGenome()[i].GetOp();
- inst_bin[inst_id]++;
- }
-
- // Print it out...
- if (file_type == FILE_TYPE_HTML) fp << "<tr><th>";
- fp << genotype->GetName() << " ";
- if (file_type == FILE_TYPE_HTML) fp << "<td align=center>";
- total_length += genome_size;
- fp << genome_size << " ";
- for (int i = 0; i < num_insts; i++) {
- const double inst_freq = ((double) inst_bin[i]) / (double) genome_size;
- total_freq[i] += inst_freq;
- if (file_type == FILE_TYPE_HTML) {
- if (inst_freq == 0.0) fp << "<td bgcolor=\"FFAAAA\">";
- else if (inst_freq < min_freq) fp << "<td bgcolor=\"FFFFAA\">";
- else if (inst_freq < max_freq) fp << "<td bgcolor=\"AAAAFF\">";
- else fp << "<td bgcolor=\"AAFFAA\">";
- }
- fp << cStringUtil::Stringf("%04.3f", inst_freq) << " ";
- }
- if (file_type == FILE_TYPE_HTML) fp << "</tr>";
- fp << endl;
- }
-
- if (file_type == FILE_TYPE_HTML) {
- int num_genomes = batch[cur_batch].List().GetSize();
- fp << "<tr><th>Average <th>" << total_length / num_genomes << " ";
- for (int i = 0; i < num_insts; i++) {
- double inst_freq = total_freq[i] / num_genomes;
- if (inst_freq == 0.0) fp << "<td bgcolor=\"FF0000\">";
- else if (inst_freq < min_freq) fp << "<td bgcolor=\"FFFF00\">";
- else if (inst_freq < max_freq) fp << "<td bgcolor=\"0000FF\">";
- else fp << "<td bgcolor=\"00FF00\">";
- fp << cStringUtil::Stringf("%04.3f", inst_freq) << " ";
- }
- fp << "</tr>" << endl
- << "</table></center>" << endl;
- }
- }
-
-void cAnalyze::AnalyzeInstPop(cString cur_string)
-{
- if (verbose == true) {
- cout << "Analyzing Instructions in batch " << cur_batch << endl;
- }
- else cout << "Analyzeing Instructions..." << endl;
-
- // Load in the variables...
- cString filename("inst_analyze.dat");
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
- const int num_insts = inst_set.GetSize();
-
- // Setup the file...
- ofstream fp(filename);
-
- for (int i = 0; i < num_insts; i++) {
- cInstruction cur_inst(i);
- fp << cur_inst.GetSymbol() << ":" << inst_set.GetName(i) << " ";
- }
- fp << endl;
-
- double total_length = 0.0;
- tArray<double> total_freq(num_insts);
- for (int i = 0; i < num_insts; i++) total_freq[i] = 0.0;
- int num_orgs = 0;
-
- // Loop through all of the genotypes in this batch...
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
-
- num_orgs++;
-
- // Setup for counting...
- tArray<int> inst_bin(num_insts);
- for (int i = 0; i < num_insts; i++) inst_bin[i] = 0;
-
- // Count it up!
- const int genome_size = genotype->GetLength();
- for (int i = 0; i < genome_size; i++) {
- const int inst_id = genotype->GetGenome()[i].GetOp();
- inst_bin[inst_id]++;
- }
- total_length += genome_size;
- for (int i = 0; i < num_insts; i++) {
- const double inst_freq = ((double) inst_bin[i]) / (double) genome_size;
- total_freq[i] += inst_freq;
- }
- }
- // Print it out...
- // fp << total_length/num_orgs << " ";
- for (int i = 0; i < num_insts; i++) {
- fp << cStringUtil::Stringf("%04.3f", total_freq[i]/num_orgs) << " ";
- }
- fp << endl;
-
-}
-
-void cAnalyze::AnalyzeBranching(cString cur_string)
-{
- if (verbose == true) {
- cout << "Analyzing branching patterns in batch " << cur_batch << endl;
- }
- else cout << "Analyzeing Branches..." << endl;
-
- // Load in the variables...
- cString filename("branch_analyze.dat");
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
-
- // Setup the file...
- ofstream fp(filename);
-
- // UNFINISHED!
- // const int num_insts = inst_set.GetSize();
-}
-
-void cAnalyze::AnalyzeMutationTraceback(cString cur_string)
-{
- if (verbose == true) {
- cout << "Analyzing mutation traceback in batch " << cur_batch << endl;
- }
- else cout << "Analyzing mutation traceback..." << endl;
-
- // This works best on lineages, so warn if we don't have one.
- if (batch[cur_batch].IsLineage() == false && verbose == true) {
- cerr << " Warning: trying to traceback mutations outside of lineage."
- << endl;
- }
-
- if (batch[cur_batch].List().GetSize() == 0) {
- cerr << "Error: Trying to traceback mutations with no genotypes in batch."
- << endl;
- return;
- }
-
- // Make sure all genotypes are the same length.
- int size = -1;
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- if (size == -1) size = genotype->GetLength();
- if (size != genotype->GetLength()) {
- cerr << " Error: Trying to traceback mutations in genotypes of differing lengths." << endl;
- cerr << " Aborting." << endl;
- return;
- }
- }
-
- // Setup variables...
- cString filename("analyze_traceback.dat");
- if (cur_string.GetSize() != 0) filename = cur_string.PopWord();
-
- // Setup a genome to store the previous values before mutations.
- tArray<int> prev_inst(size);
- prev_inst.SetAll(-1); // -1 indicates never changed.
-
- // Open the output file...
- ofstream fp(filename);
-
- // Loop through all of the genotypes again, testing mutation reversions.
- cAnalyzeGenotype * prev_genotype = batch_it.Next();
- while ((genotype = batch_it.Next()) != NULL) {
- continue;
- // Check to see if any sites have changed...
- for (int i = 0; i < size; i++) {
- if (genotype->GetGenome()[i] != prev_genotype->GetGenome()[i]) {
- prev_inst[i] = prev_genotype->GetGenome()[i].GetOp();
- }
- }
-
- // Next, determine the fraction of mutations that are currently adaptive.
- int num_beneficial = 0;
- int num_neutral = 0;
- int num_detrimental = 0;
- int num_static = 0; // Sites that were never mutated.
-
- cGenome test_genome = genotype->GetGenome();
- cCPUTestInfo test_info;
- m_world->GetTestCPU().TestGenome(test_info, test_genome);
- const double base_fitness = test_info.GetGenotypeFitness();
-
- for (int i = 0; i < size; i++) {
- if (prev_inst[i] == -1) num_static++;
- else {
- test_genome[i].SetOp(prev_inst[i]);
- m_world->GetTestCPU().TestGenome(test_info, test_genome);
- const double cur_fitness = test_info.GetGenotypeFitness();
- if (cur_fitness > base_fitness) num_detrimental++;
- else if (cur_fitness < base_fitness) num_beneficial++;
- else num_neutral++;
- test_genome[i] = genotype->GetGenome()[i];
- }
- }
-
- fp << genotype->GetDepth() << " "
- << num_beneficial << " "
- << num_neutral << " "
- << num_detrimental << " "
- << num_static << " "
- << endl;
-
- prev_genotype = genotype;
- }
-}
-
-void cAnalyze::AnalyzeComplexity(cString cur_string)
-{
- cout << "Analyzing genome complexity..." << endl;
-
- // Load in the variables...
- // This command requires at least on arguement
- int words = cur_string.CountNumWords();
- if(words < 1) {
- cout << "Error: AnalyzeComplexity has no parameters, skipping." << endl;
- return;
- }
-
- // Get the mutation rate arguement
- double mut_rate = cur_string.PopWord().AsDouble();
-
- // Create the directory using the string given as the second arguement
- cString dir = cur_string.PopWord();
- cString defaultDirectory = "complexity/";
- cString directory = PopDirectory(dir, defaultDirectory);
-
- // Default for usage of resources is false
- int useResources = 0;
- // resource usage flag is an optional arguement, but is always the 3rd arg
- if(words >= 3) {
- useResources = cur_string.PopWord().AsInt();
- // All non-zero values are considered false
- if(useResources != 0 && useResources != 1) {
- useResources = 0;
- }
- }
-
- // Batch frequency begins with the first organism, but then skips that
- // amount ahead in the batch. It defaults to 1, so that default analyzes
- // all the organisms in the batch. It is always the 4th arg.
- int batchFrequency = 1;
- if(words == 4) {
- batchFrequency = cur_string.PopWord().AsInt();
- if(batchFrequency <= 0) {
- batchFrequency = 1;
- }
- }
-
- // These are used to backup the state of the environment and resource_count
- // for the test cpu, and will be reset to their original values at the end
- bool backupUsage;
- tArray<double> backupResources;
-
- if(useResources) {
- // Backup test cpu data
- backupUsage = m_world->GetTestCPU().UseResources();
- backupResources = m_world->GetTestCPU().GetResources();
-
- m_world->GetTestCPU().UseResources() = true;
- }
-
- ///////////////////////////////////////////////////////
- // Loop through all of the genotypes in this batch...
-
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
-
-
- bool islineage = false;
- cString lineage_filename;
- ofstream lineage_fp;
- ofstream non_lineage_fp;
- if (batch[cur_batch].IsLineage() == true) {
- lineage_filename.Set("%s%s.complexity.dat", directory(), "lineage");
- lineage_fp.open(lineage_filename);
- islineage = true;
- } else {
- cString non_lineage_file;
- non_lineage_file.Set("%s%s.complexity.dat", directory(), "nonlineage");
- non_lineage_fp.open(non_lineage_file);
- islineage = false;
- }
-
- while ((genotype = batch_it.Next()) != NULL) {
- if (verbose == true) {
- cout << " Analyzing complexity for " << genotype->GetName() << endl;
- }
-
- // Construct this filename...
- cString filename;
- filename.Set("%s%s.complexity.dat", directory(), genotype->GetName()());
- ofstream fp(filename());
- if (islineage) {
- lineage_fp << genotype->GetID() << " ";
- } else {
- non_lineage_fp << genotype->GetID() << " ";
- }
-
- int updateBorn = -1;
- if(useResources) {
- updateBorn = genotype->GetUpdateBorn();
- FillResources(updateBorn);
- }
-
- // Calculate the stats for the genotype we're working with ...
- genotype->Recalculate();
- cout << genotype->GetFitness() << endl;
- const int num_insts = inst_set.GetSize();
- const int max_line = genotype->GetLength();
- const cGenome & base_genome = genotype->GetGenome();
- cGenome mod_genome(base_genome);
-
- // Loop through all the lines of code, testing all mutations...
- tArray<double> test_fitness(num_insts);
- tArray<double> prob(num_insts);
- for (int line_num = 0; line_num < max_line; line_num++) {
- int cur_inst = base_genome[line_num].GetOp();
- //char cur_symbol = base_genome[line_num].GetSymbol();
-
- // Column 1 ... the original instruction in the genome.
- fp << cur_inst << " ";
-
- // Test fitness of each mutant.
- for (int mod_inst = 0; mod_inst < num_insts; mod_inst++) {
- mod_genome[line_num].SetOp(mod_inst);
- cAnalyzeGenotype test_genotype(m_world, mod_genome, inst_set);
- test_genotype.Recalculate();
- test_fitness[mod_inst] = test_genotype.GetFitness();
- }
-
- // Ajust fitness
- double cur_inst_fitness = test_fitness[cur_inst];
- for (int mod_inst = 0; mod_inst < num_insts; mod_inst++) {
- if (test_fitness[mod_inst] > cur_inst_fitness)
- test_fitness[mod_inst] = cur_inst_fitness;
- test_fitness[mod_inst] = test_fitness[mod_inst] / cur_inst_fitness;
- }
-
- // Calculate probabilities at mut-sel balance
- double w_bar = 1;
-
- // Normalize fitness values, assert if they are all zero
- double maxFitness = 0.0;
- for(int i=0; i<num_insts; i++) {
- if(test_fitness[i] > maxFitness) {
- maxFitness = test_fitness[i];
- }
- }
-
- if(maxFitness > 0) {
- for(int i=0; i<num_insts; i++) {
- test_fitness[i] /= maxFitness;
- }
- } else {
- fp << "All zero fitness, ERROR." << endl;
- continue;
- }
-
- while(1) {
- double sum = 0.0;
- for (int mod_inst = 0; mod_inst < num_insts; mod_inst ++) {
- prob[mod_inst] = (mut_rate * w_bar) /
- ((double)num_insts * (w_bar + test_fitness[mod_inst] * mut_rate - test_fitness[mod_inst]));
- sum = sum + prob[mod_inst];
- }
- if ((sum-1.0)*(sum-1.0) <= 0.0001)
- break;
- else
- w_bar = w_bar - 0.000001;
- }
- // Write probability
- for (int mod_inst = 0; mod_inst < num_insts; mod_inst ++) {
- fp << prob[mod_inst] << " ";
- }
-
- // Calculate complexity
- double entropy = 0;
- for (int i = 0; i < num_insts; i ++) {
- entropy += prob[i] * log((double) 1.0/prob[i]) / log ((double) num_insts);
- }
- double complexity = 1 - entropy;
- fp << complexity << endl;
-
- if (islineage) {
- lineage_fp << complexity << " ";
- } else {
- non_lineage_fp << complexity << " ";
- }
-
- // Reset the mod_genome back to the original sequence.
- mod_genome[line_num].SetOp(cur_inst);
- }
- fp.close();
- if (islineage) {
- lineage_fp << endl;
- } else {
- non_lineage_fp << endl;
- }
-
- // Always grabs the first one
- // Skip i-1 times, so that the beginning of the loop will grab the ith one
- // where i is the batchFrequency
- for(int count=0; genotype != NULL && count < batchFrequency - 1; count++) {
- genotype = batch_it.Next();
- if(genotype != NULL && verbose == true) {
- cout << "Skipping: " << genotype->GetName() << endl;
- }
- }
- if(genotype == NULL) { break; }
- }
- if (islineage) {
- lineage_fp.close();
- } else {
- non_lineage_fp.close();
- }
-
- if(useResources) {
- // Set the test cpu back to the state it was in before we messed with it
- m_world->GetTestCPU().UseResources() = backupUsage;
- m_world->GetTestCPU().SetupResourceArray(backupResources);
- }
-
- return;
-}
-
-void cAnalyze::AnalyzePopComplexity(cString cur_string)
-{
- cout << "Analyzing population complexity ..." << endl;
-
- // Load in the variables...
- //int batch_size = cur_string.PopWord().AsInt();
- //int sample_size = cur_string.PopWord().AsInt();
- cString directory = PopDirectory(cur_string, "pop_complexity/");
- cString file = cur_string;
-
- // Get sample index...
- /*if (sample_size > batch_size) {
- cout << " Sample size is greater than batch size." << endl;
- return;
- }
-int *index_array;
-int *flag_array;
-index_array = (int*) malloc (batch_size*sizeof(int));
-flag_array = (int*) malloc (batch_size*sizeof(int));
-for (int i=0; i<batch_size; i++) {
- index_array[i] = i;
- flag_array[i] = 0;
-}
-int max_num = batch_size;
-cRandom random;
-for (int i=0; i<sample_size; i++) {
- int index = random.GetUInt(max_num);
- flag_array[index_array[index]] = 1;
- index_array[index] = index_array[max_num-1];
- max_num--;
-}*/
-
- // Construct filename...
- cString filename;
- filename.Set("%spop%s.complexity.dat", directory(), file());
- ofstream fp(filename());
-
- //////////////////////////////////////////////////////////
- // Loop through all of the genotypes in this batch ...
-
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
-
-
- genotype = batch_it.Next();
-
-
- if (genotype == NULL) return;
- int seq_length = genotype->GetLength();
- const int num_insts = inst_set.GetSize();
- tMatrix<int> inst_stat(seq_length, num_insts);
-
- // Initializing inst_stat ...
- for (int line_num = 0; line_num < seq_length; line_num ++)
- for (int inst_num = 0; inst_num < num_insts; inst_num ++)
- inst_stat(line_num, inst_num) = 0;
-
- int num_cpus = 0;
- int actural_samples = 0;
- while (genotype != NULL) {
- num_cpus = genotype->GetNumCPUs();
- const cGenome & base_genome = genotype->GetGenome();
- for (int i=0; i<num_cpus; i++) { // Stat on every organism with same genotype.
- //if (flag_array[organism_index] == 0) {
- //organism_index++;
- //continue;
- //}
- for (int line_num = 0; line_num < seq_length; line_num ++) {
- int cur_inst = base_genome[line_num].GetOp();
- inst_stat(line_num, cur_inst) ++;
- }
- //organism_index++;
- actural_samples++;
- }
- genotype = batch_it.Next();
- }
-
-
-
-// Calculate complexity
-for (int line_num = 0; line_num < seq_length; line_num ++) {
- double entropy = 0.0;
- for (int inst_num = 0; inst_num < num_insts; inst_num ++) {
- if (inst_stat(line_num, inst_num) == 0) continue;
- float prob = (float) (inst_stat(line_num, inst_num)) / (float) (actural_samples);
- entropy += prob * log((double) 1.0/prob) / log((double) num_insts);
- }
- double complexity = 1 - entropy;
- fp << complexity << " ";
-};
-fp << endl;
-fp.close();
-//free(index_array);
-//free(flag_array);
-return;
-}
-
-
-void cAnalyze::EnvironmentSetup(cString cur_string)
-{
- cout << "Running environment command: " << endl
- << " " << cur_string << endl;
-
- m_world->GetEnvironment().LoadLine(cur_string);
-}
-
-
-void cAnalyze::CommandHelpfile(cString cur_string)
-{
- cout << "Printing helpfiles in: " << cur_string << endl;
-
- cHelpManager help_control;
- if (verbose == true) help_control.SetVerbose();
- while (cur_string.GetSize() > 0) {
- help_control.LoadFile(cur_string.PopWord());
- }
-
- help_control.PrintHTML();
-}
-
-void cAnalyze::CommandDocfile(cString cur_string)
-{
- cout << "Printing documentation files in: " << cur_string << endl;
-
- cHelpManager help_control;
- if (verbose == true) help_control.SetVerbose();
- while (cur_string.GetSize() > 0) {
- help_control.LoadFile(cur_string.PopWord());
- }
-
- help_control.PrintHTML();
-}
-
-
-
-//////////////// Control...
-
-void cAnalyze::VarSet(cString cur_string)
-{
- cString var = cur_string.PopWord();
-
- if (cur_string.GetSize() == 0) {
- cerr << "Error: No variable provided in SET command" << endl;
- return;
- }
-
- cString & cur_variable = GetVariable(var);
- cur_variable = cur_string.PopWord();
-
- if (verbose == true) {
- cout << "Setting " << var << " to " << cur_variable << endl;
- }
-}
-
-void cAnalyze::BatchSet(cString cur_string)
-{
- int next_batch = 0;
- if (cur_string.CountNumWords() > 0) {
- next_batch = cur_string.PopWord().AsInt();
- }
- if (verbose) cout << "Setting current batch to " << next_batch << endl;
- if (next_batch >= MAX_BATCHES) {
- cerr << " Error: max batches is " << MAX_BATCHES << endl;
- exit(1);
- } else {
- cur_batch = next_batch;
- }
-}
-
-void cAnalyze::BatchName(cString cur_string)
-{
- if (cur_string.CountNumWords() == 0) {
- if (verbose) cout << " Warning: No name given in NAME_BATCH!" << endl;
- return;
- }
-
- batch[cur_batch].Name() = cur_string.PopWord();
-}
-
-void cAnalyze::BatchTag(cString cur_string)
-{
- if (cur_string.CountNumWords() == 0) {
- if (verbose) cout << " Warning: No tag given in TAG_BATCH!" << endl;
- return;
- }
-
- if (verbose == true) {
- cout << "Tagging batch " << cur_batch
- << " with tag '" << cur_string << "'" << endl;
- }
-
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- genotype->SetTag(cur_string);
- }
-
-}
-
-void cAnalyze::BatchPurge(cString cur_string)
-{
- int batch_id = cur_batch;
- if (cur_string.CountNumWords() > 0) batch_id = cur_string.PopWord().AsInt();
-
- if (verbose) cout << "Purging batch " << batch_id << endl;
-
- while (batch[batch_id].List().GetSize() > 0) {
- delete batch[batch_id].List().Pop();
- }
-
- batch[batch_id].SetLineage(false);
- batch[batch_id].SetAligned(false);
-}
-
-void cAnalyze::BatchDuplicate(cString cur_string)
-{
- if (cur_string.GetSize() == 0) {
- cerr << "Duplicate Error: Must include from ID!" << endl;
- exit(1);
- }
- int batch_from = cur_string.PopWord().AsInt();
-
- int batch_to = cur_batch;
- if (cur_string.GetSize() > 0) batch_to = cur_string.PopWord().AsInt();
-
- if (verbose == true) {
- cout << "Duplicating from batch " << batch_from
- << " to batch " << batch_to << "." << endl;
- }
-
- tListIterator<cAnalyzeGenotype> batch_from_it(batch[batch_from].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_from_it.Next()) != NULL) {
- cAnalyzeGenotype * new_genotype = new cAnalyzeGenotype(*genotype);
- batch[batch_to].List().PushRear(new_genotype);
- }
-
- batch[batch_to].SetLineage(false);
- batch[batch_to].SetAligned(false);
-}
-
-void cAnalyze::BatchRecalculate(cString cur_string)
-{
- int words = cur_string.CountNumWords();
- int useResources = 0;
- int update = -1;
- if(words >= 1) {
- useResources = cur_string.PopWord().AsInt();
- // All non-zero values are considered false
- if(useResources != 0 && useResources != 1) {
- useResources = 0;
- }
- }
- if (words >= 2) {
- update = cur_string.PopWord().AsInt();
- }
-
- bool backupUsage;
- tArray<double> backupResources;
-
- if(useResources) {
- // Backup test cpu data
- backupUsage = m_world->GetTestCPU().UseResources();
- backupResources = m_world->GetTestCPU().GetResources();
-
- m_world->GetTestCPU().UseResources() = true;
- }
-
- if (verbose == true) {
- cout << "Running batch " << cur_batch << " through test CPUs..." << endl;
- } else cout << "Running through test CPUs..." << endl;
-
- if (verbose == true && batch[cur_batch].IsLineage() == false) {
- cerr << " Warning: batch may not be a linege; "
- << "parent and ancestor distances may not be correct" << endl;
- }
-
- if (useResources && update > -1) {
- FillResources(update);
- }
-
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- cAnalyzeGenotype * last_genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
-
- // If use resources, load proper resource according to update_born
- if(useResources && update == -1) {
- int updateBorn = -1;
- updateBorn = genotype->GetUpdateBorn();
- FillResources(updateBorn);
- }
-
- // If the previous genotype was the parent of this one, pass in a pointer
- // to it for improved recalculate (such as distance to parent, etc.)
-// if (verbose == true) {
-// PrintTestCPUResources("");
-// }
- if (last_genotype != NULL &&
- genotype->GetParentID() == last_genotype->GetID()) {
- genotype->Recalculate(last_genotype);
- }
- else genotype->Recalculate();
- last_genotype = genotype;
- }
-
- if(useResources) {
- // Set the test cpu back to the state it was in before we messed with it
- m_world->GetTestCPU().UseResources() = backupUsage;
- m_world->GetTestCPU().SetupResourceArray(backupResources);
- }
-
- return;
-}
-
-void cAnalyze::BatchRename(cString cur_string)
-{
- if (verbose == false) cout << "Renaming organisms..." << endl;
- else cout << "Renaming organisms in batch " << cur_batch << endl;
-
- // If a number is given with rename, start at that number...
-
- int id_num = cur_string.PopWord().AsInt();
- tListIterator<cAnalyzeGenotype> batch_it(batch[cur_batch].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- cString name = cStringUtil::Stringf("org-%d", id_num);
- genotype->SetID(id_num);
- genotype->SetName(name);
- id_num++;
- }
-}
-
-void cAnalyze::PrintStatus(cString cur_string)
-{
- // No Args needed...
- (void) cur_string;
-
- cout << "Status Report:" << endl;
- for (int i = 0; i < MAX_BATCHES; i++) {
- if (i == cur_batch || batch[i].List().GetSize() > 0) {
- cout << " Batch " << i << " -- "
- << batch[i].List().GetSize() << " genotypes.";
- if (i == cur_batch) cout << " <current>";
- if (batch[i].IsLineage() == true) cout << " <lineage>";
- if (batch[i].IsAligned() == true) cout << " <aligned>";
-
- cout << endl;
- }
- }
-}
-
-void cAnalyze::PrintDebug(cString cur_string)
-{
- cerr << "Debug Args: " << cur_string << endl;
-}
-
-void cAnalyze::ToggleVerbose(cString cur_string)
-{
- // No Args needed...
- (void) cur_string;
-
- if (verbose == false) {
- cout << "Using verbose log messages..." << endl;
- verbose = true;
- } else {
- cout << "Using non-verbose log messages..." << endl;
- verbose = false;
- }
-}
-
-void cAnalyze::IncludeFile(cString cur_string)
-{
- while (cur_string.GetSize() > 0) {
- cString filename = cur_string.PopWord();
-
- cInitFile include_file(filename);
- include_file.Load();
- include_file.Compress();
- include_file.Close();
-
- tList<cAnalyzeCommand> include_list;
- LoadCommandList(include_file, include_list);
- ProcessCommands(include_list);
- }
-}
-
-void cAnalyze::CommandSystem(cString cur_string)
-{
- cout << "Running System Command: " << cur_string << endl;
-
- system(cur_string());
-}
-
-void cAnalyze::CommandInteractive(cString cur_string)
-{
- // No Args needed...
- (void) cur_string;
-
- RunInteractive();
-}
-
-void cAnalyze::PrintTestCPUResources(cString cur_string)
-{
- cout << "TestCPU is using resources: ";
- cout << m_world->GetTestCPU().UseResources() << endl;
- cout << "Resources currently in TestCPU: ";
- const tArray<double> &quantity = m_world->GetTestCPU().GetResources();
- for(int i=0; i<quantity.GetSize(); i++) {
- cout << quantity.ElementAt(i) << " ";
- }
- cout << endl;
-
- return;
-}
-
-void cAnalyze::FunctionCreate(cString cur_string,
- tList<cAnalyzeCommand> & clist)
-{
- int num_args = cur_string.CountNumWords();
- if (num_args < 1) {
- cerr << "Error: Must provide function name when creating function.";
- exit(1);
- }
-
- cString fun_name = cur_string.PopWord();
- fun_name.ToUpper();
-
- if (FindAnalyzeCommandDef(fun_name) != NULL) {
- cerr << "Error: Cannot create function '" << fun_name
- << "'; already exists." << endl;
- exit(1);
- }
-
- if (verbose) cout << "Creating function: " << fun_name << endl;
-
- // Create the new function...
- cAnalyzeFunction * new_function = new cAnalyzeFunction(fun_name);
- while (clist.GetSize() > 0) {
- new_function->GetCommandList()->PushRear(clist.Pop());
- }
-
- // Save the function on the new list...
- function_list.PushRear(new_function);
-}
-
-bool cAnalyze::FunctionRun(const cString & fun_name, cString args)
-{
- if (verbose) {
- cout << "Running function: " << fun_name << endl;
- // << " with args: " << args << endl;
- }
-
- // Find the function we're about to run...
- cAnalyzeFunction * found_function = NULL;
- tListIterator<cAnalyzeFunction> function_it(function_list);
- while (function_it.Next() != NULL) {
- if (function_it.Get()->GetName() == fun_name) {
- found_function = function_it.Get();
- break;
- }
- }
-
- // If we were unable to find the command we're looking for, return false.
- if (found_function == NULL) return false;
-
- // Back up the local variables
- cString backup_arg_vars[10];
- cString backup_local_vars[26];
- for (int i = 0; i < 10; i++) backup_arg_vars[i] = arg_variables[i];
- for (int i = 0; i < 26; i++) backup_local_vars[i] = local_variables[i];
-
- // Set the arg variables to the passed-in args...
- arg_variables[0] = fun_name;
- for (int i = 1; i < 10; i++) arg_variables[i] = args.PopWord();
- for (int i = 0; i < 26; i++) local_variables[i] = "";
-
- ProcessCommands(*(found_function->GetCommandList()));
-
- // Restore the local variables
- for (int i = 0; i < 10; i++) arg_variables[i] = backup_arg_vars[i];
- for (int i = 0; i < 26; i++) local_variables[i] = backup_local_vars[i];
-
- return true;
-}
-
-
-int cAnalyze::BatchUtil_GetMaxLength(int batch_id)
-{
- if (batch_id < 0) batch_id = cur_batch;
-
- int max_length = 0;
-
- tListIterator<cAnalyzeGenotype> batch_it(batch[batch_id].List());
- cAnalyzeGenotype * genotype = NULL;
- while ((genotype = batch_it.Next()) != NULL) {
- if (genotype->GetLength() > max_length) max_length = genotype->GetLength();
- }
-
- return max_length;
-}
-
-
-void cAnalyze::CommandForeach(cString cur_string,
- tList<cAnalyzeCommand> & clist)
-{
- if (verbose) cout << "Initiating Foreach loop..." << endl;
-
- cString var = cur_string.PopWord();
- int num_args = cur_string.CountNumWords();
-
- cString & cur_variable = GetVariable(var);
-
- for (int i = 0; i < num_args; i++) {
- cur_variable = cur_string.PopWord();
-
- if (verbose == true) {
- cout << "Foreach: setting " << var << " to " << cur_variable << endl;
- }
- ProcessCommands(clist);
- }
-
- if (verbose == true) {
- cout << "Ending Foreach on " << var << endl;
- }
-}
-
-
-void cAnalyze::CommandForRange(cString cur_string,
- tList<cAnalyzeCommand> & clist)
-{
- if (verbose) cout << "Initiating FORRANGE loop..." << endl;
-
- int num_args = cur_string.CountNumWords();
- if (num_args < 3) {
- cerr << " Error: Must give variable, min and max with FORRANGE!"
- << endl;
- exit(1);
- }
-
- cString var = cur_string.PopWord();
- double min_val = cur_string.PopWord().AsDouble();
- double max_val = cur_string.PopWord().AsDouble();
- double step_val = 1.0;
- if (num_args >=4 ) step_val = cur_string.PopWord().AsDouble();
-
- cString & cur_variable = GetVariable(var);
-
- // Seperate out all ints from not all ints...
- if (min_val == (double) ((int) min_val) &&
- max_val == (double) ((int) max_val) &&
- step_val == (double) ((int) step_val)) {
- for (int i = (int) min_val; i <= (int) max_val; i += (int) step_val) {
- cur_variable.Set("%d", i);
-
- if (verbose == true) {
- cout << "FORRANGE: setting " << var << " to " << cur_variable << endl;
- }
- ProcessCommands(clist);
- }
- } else {
- for (double i = min_val; i <= max_val; i += step_val) {
- cur_variable.Set("%f", i);
-
- if (verbose == true) {
- cout << "FORRANGE: setting " << var << " to " << cur_variable << endl;
- }
- ProcessCommands(clist);
- }
- }
-
- if (verbose == true) {
- cout << "Ending FORRANGE on " << var << endl;
- }
-}
-
-
-/////////////////// Private Methods ///////////////////////////
-
-cString cAnalyze::PopDirectory(cString & in_string,
- const cString & default_dir)
-{
- // Determing the directory name
- cString directory(default_dir);
- if (in_string.GetSize() != 0) directory = in_string.PopWord();
-
- // Make sure the directory ends in a slash. If not, add one.
- int last_pos = directory.GetSize() - 1;
- if (directory[last_pos] != '/' && directory[last_pos] != '\\') {
- directory += '/';
- }
-
- // Make sure the directory exists.
- FILE *fp = fopen ( directory(), "r" );
- if ( fp == 0 ){
- if ( errno == ENOENT ){
- cerr << "Directory '" << directory
- << "' does not exist. Creating..." << endl;
- if ( mkdir( directory(), (S_IRWXU|S_IRWXG|S_IRWXO) ) )
- cerr << " Error creating '" << directory << "'" << endl;
- }
- else cerr << " Error opening '" << directory << "'" << endl;
- }
-
- return directory;
-}
-
-int cAnalyze::PopBatch(const cString & in_string)
-{
- int batch = cur_batch;
- if (in_string.GetSize() != 0 && in_string != "current") {
- batch = in_string.AsInt();
- }
-
- return batch;
-}
-
-cAnalyzeGenotype * cAnalyze::PopGenotype(cString gen_desc, int batch_id)
-{
- if (batch_id == -1) batch_id = cur_batch;
- tListPlus<cAnalyzeGenotype> & gen_list = batch[batch_id].List();
- gen_desc.ToLower();
-
- cAnalyzeGenotype * found_gen = NULL;
- if (gen_desc == "num_cpus")
- found_gen = gen_list.PopIntMax(&cAnalyzeGenotype::GetNumCPUs);
- else if (gen_desc == "total_cpus")
- found_gen = gen_list.PopIntMax(&cAnalyzeGenotype::GetTotalCPUs);
- else if (gen_desc == "merit")
- found_gen = gen_list.PopDoubleMax(&cAnalyzeGenotype::GetMerit);
- else if (gen_desc == "fitness")
- found_gen = gen_list.PopDoubleMax(&cAnalyzeGenotype::GetFitness);
- else if (gen_desc.IsNumeric(0))
- found_gen = gen_list.PopIntValue(&cAnalyzeGenotype::GetID,
- gen_desc.AsInt());
- else if (gen_desc == "random") {
- int gen_pos = random.GetUInt(gen_list.GetSize());
- found_gen = gen_list.PopPos(gen_pos);
- }
- else {
- cout << " Error: unknown type " << gen_desc << endl;
- exit(1);
- }
-
- return found_gen;
-}
-
-
-cString & cAnalyze::GetVariable(const cString & var)
-{
- if (var.GetSize() != 1 ||
- (var.IsLetter(0) == false && var.IsNumeric(0) == false)) {
- cerr << "Error: Illegal variable " << var << " being used." << endl;
- exit(1);
- }
-
- if (var.IsLowerLetter(0) == true) {
- int var_id = (int) (var[0] - 'a');
- return variables[var_id];
- }
- else if (var.IsUpperLetter(0) == true) {
- int var_id = (int) (var[0] - 'A');
- return local_variables[var_id];
- }
- // Otherwise it must be a number...
- int var_id = (int) (var[0] - '0');
- return arg_variables[var_id];
-}
-
-
-void cAnalyze::LoadCommandList(cInitFile & init_file,
- tList<cAnalyzeCommand> & clist)
-{
- while (init_file.GetLineNum() < init_file.GetNumLines()) {
- cString cur_string = init_file.GetNextLine();
- cString command = cur_string.PopWord();
- command.ToUpper();
-
- cAnalyzeCommand * cur_command;
- cAnalyzeCommandDefBase * command_def = FindAnalyzeCommandDef(command);
- if (command == "END") {
- // We are done with this section of code; break out...
- break;
- }
- else if (command_def != NULL && command_def->IsFlowCommand() == true) {
- // This code has a body to it... fill it out!
- cur_command = new cAnalyzeFlowCommand(command, cur_string);
- LoadCommandList( init_file, *(cur_command->GetCommandList()) );
- }
- else {
- // This is a normal command...
- cur_command = new cAnalyzeCommand(command, cur_string);
- }
-
- clist.PushRear(cur_command);
- }
-}
-
-void cAnalyze::InteractiveLoadCommandList(tList<cAnalyzeCommand> & clist)
-{
- interactive_depth++;
- char text_input[2048];
- while (true) {
- for (int i = 0; i <= interactive_depth; i++) {
- cout << ">>";
- }
- cout << " ";
- cout.flush();
- cin.getline(text_input, 2048);
- cString cur_input(text_input);
- cString command = cur_input.PopWord();
- command.ToUpper();
-
- cAnalyzeCommand * cur_command;
- cAnalyzeCommandDefBase * command_def = FindAnalyzeCommandDef(command);
- if (command == "END") {
- // We are done with this section of code; break out...
- break;
- }
- else if (command_def != NULL && command_def->IsFlowCommand() == true) {
- // This code has a body to it... fill it out!
- cur_command = new cAnalyzeFlowCommand(command, cur_input);
- InteractiveLoadCommandList(*(cur_command->GetCommandList()));
- }
- else {
- // This is a normal command...
- cur_command = new cAnalyzeCommand(command, cur_input);
- }
-
- clist.PushRear(cur_command);
- }
- interactive_depth--;
-}
-
-void cAnalyze::PreProcessArgs(cString & args)
-{
- int pos = 0;
- int search_start = 0;
- while ((pos = args.Find('$', search_start)) != -1) {
- // Setup the variable name that was found...
- char varlet = args[pos+1];
- cString varname("$");
- varname += varlet;
-
- // Determine the variable and act on it.
- int varsize = 0;
- if (varlet == '$') {
- args.Clip(pos+1, 1);
- varsize = 1;
- }
- else if (varlet >= 'a' && varlet <= 'z') {
- int var_id = (int) (varlet - 'a');
- args.Replace(varname, variables[var_id], pos);
- varsize = variables[var_id].GetSize();
- }
- else if (varlet >= 'A' && varlet <= 'Z') {
- int var_id = (int) (varlet - 'A');
- args.Replace(varname, local_variables[var_id], pos);
- varsize = local_variables[var_id].GetSize();
- }
- else if (varlet >= '0' && varlet <= '9') {
- int var_id = (int) (varlet - '0');
- args.Replace(varname, arg_variables[var_id], pos);
- varsize = arg_variables[var_id].GetSize();
- }
- search_start = pos + varsize;
- }
-}
-
-void cAnalyze::ProcessCommands(tList<cAnalyzeCommand> & clist)
-{
- // Process the command list...
- tListIterator<cAnalyzeCommand> command_it(clist);
- command_it.Reset();
- cAnalyzeCommand * cur_command = NULL;
- while ((cur_command = command_it.Next()) != NULL) {
- cString command = cur_command->GetCommand();
- cString args = cur_command->GetArgs();
- PreProcessArgs(args);
-
- cAnalyzeCommandDefBase * command_fun = FindAnalyzeCommandDef(command);
-
- if (command_fun != NULL) command_fun->Run(this, args, *cur_command);
- else if (FunctionRun(command, args) == true) {
- // Found a defined function by this name.
- }
- else {
- cerr << "Error: Unknown analysis keyword '" << command << "'." << endl;
- exit(1);
- }
-
- }
-}
-
-void cAnalyze::SetupGenotypeDataList()
-{
- if (genotype_data_list.GetSize() != 0) return; // List already setup.
-
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, bool>
- ("viable", "Is Viable (0/1)", &cAnalyzeGenotype::GetViable,
- &cAnalyzeGenotype::SetViable));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("id", "Genotype ID", &cAnalyzeGenotype::GetID,
- &cAnalyzeGenotype::SetID));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, const cString &>
- ("tag", "Genotype Tag", &cAnalyzeGenotype::GetTag,
- &cAnalyzeGenotype::SetTag,
- &cAnalyzeGenotype::CompareNULL, "(none)", ""));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("parent_id", "Parent ID", &cAnalyzeGenotype::GetParentID,
- &cAnalyzeGenotype::SetParentID));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("parent2_id", "Second Parent ID (for sexual genotypes)",
- &cAnalyzeGenotype::GetParent2ID, &cAnalyzeGenotype::SetParent2ID));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("parent_dist", "Parent Distance", &cAnalyzeGenotype::GetParentDist,
- &cAnalyzeGenotype::SetParentDist));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("ancestor_dist","Ancestor Distance",&cAnalyzeGenotype::GetAncestorDist,
- &cAnalyzeGenotype::SetAncestorDist));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("lineage", "Unique Lineage Label",&cAnalyzeGenotype::GetLineageLabel,
- &cAnalyzeGenotype::SetLineageLabel));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("num_cpus", "Number of CPUs", &cAnalyzeGenotype::GetNumCPUs,
- &cAnalyzeGenotype::SetNumCPUs));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("total_cpus", "Total CPUs Ever", &cAnalyzeGenotype::GetTotalCPUs,
- &cAnalyzeGenotype::SetTotalCPUs));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("length", "Genome Length", &cAnalyzeGenotype::GetLength,
- &cAnalyzeGenotype::SetLength, &cAnalyzeGenotype::CompareLength));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("copy_length", "Copied Length", &cAnalyzeGenotype::GetCopyLength,
- &cAnalyzeGenotype::SetCopyLength));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("exe_length", "Executed Length", &cAnalyzeGenotype::GetExeLength,
- &cAnalyzeGenotype::SetExeLength));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, double>
- ("merit", "Merit", &cAnalyzeGenotype::GetMerit,
- &cAnalyzeGenotype::SetMerit, &cAnalyzeGenotype::CompareMerit));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, double>
- ("comp_merit", "Computational Merit", &cAnalyzeGenotype::GetCompMerit,
- (void (cAnalyzeGenotype::*)(double)) NULL, &cAnalyzeGenotype::CompareCompMerit));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, double>
- ("comp_merit_ratio", "Computational Merit Ratio",
- &cAnalyzeGenotype::GetCompMeritRatio,
- (void (cAnalyzeGenotype::*)(double)) NULL,
- &cAnalyzeGenotype::CompareCompMerit));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("gest_time", "Gestation Time", &cAnalyzeGenotype::GetGestTime,
- &cAnalyzeGenotype::SetGestTime,
- &cAnalyzeGenotype::CompareGestTime, "Inf."));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, double>
- ("efficiency", "Rep. Efficiency", &cAnalyzeGenotype::GetEfficiency,
- (void (cAnalyzeGenotype::*)(double)) NULL,
- &cAnalyzeGenotype::CompareEfficiency));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, double>
- ("efficiency_ratio", "Rep. Efficiency Ratio",
- &cAnalyzeGenotype::GetEfficiencyRatio,
- (void (cAnalyzeGenotype::*)(double)) NULL,
- &cAnalyzeGenotype::CompareEfficiency));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, double>
- ("fitness", "Fitness", &cAnalyzeGenotype::GetFitness,
- &cAnalyzeGenotype::SetFitness, &cAnalyzeGenotype::CompareFitness));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, double>
- ("div_type", "Divide Type", &cAnalyzeGenotype::GetDivType,
- &cAnalyzeGenotype::SetDivType));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("mate_id", "Mate Selection ID Number", &cAnalyzeGenotype::GetMateID,
- &cAnalyzeGenotype::SetMateID));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, double>
- ("fitness_ratio", "Fitness Ratio", &cAnalyzeGenotype::GetFitnessRatio,
- (void (cAnalyzeGenotype::*)(double)) NULL,
- &cAnalyzeGenotype::CompareFitness));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("update_born", "Update Born", &cAnalyzeGenotype::GetUpdateBorn,
- &cAnalyzeGenotype::SetUpdateBorn));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("update_dead", "Update Dead", &cAnalyzeGenotype::GetUpdateDead,
- &cAnalyzeGenotype::SetUpdateDead));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("depth", "Tree Depth", &cAnalyzeGenotype::GetDepth,
- &cAnalyzeGenotype::SetDepth));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, double>
- ("frac_dead", "Fraction Mutations Lethal",
- &cAnalyzeGenotype::GetFracDead,
- (void (cAnalyzeGenotype::*)(double)) NULL));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, double>
- ("frac_neg", "Fraction Mutations Detrimental",
- &cAnalyzeGenotype::GetFracNeg,
- (void (cAnalyzeGenotype::*)(double)) NULL));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, double>
- ("frac_neut", "Fraction Mutations Neutral",
- &cAnalyzeGenotype::GetFracNeut,
- (void (cAnalyzeGenotype::*)(double)) NULL));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, double>
- ("frac_pos", "Fraction Mutations Beneficial",
- &cAnalyzeGenotype::GetFracPos,
- (void (cAnalyzeGenotype::*)(double)) NULL));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, double>
- ("complexity", "Basic Complexity (all neutral/beneficial muts are equal)",
- &cAnalyzeGenotype::GetComplexity,
- (void (cAnalyzeGenotype::*)(double)) NULL));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, const cString &>
- ("parent_muts", "Mutations from Parent",
- &cAnalyzeGenotype::GetParentMuts, &cAnalyzeGenotype::SetParentMuts,
- &cAnalyzeGenotype::CompareNULL, "(none)", ""));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, const cString &>
- ("task_order", "Task Performance Order",
- &cAnalyzeGenotype::GetTaskOrder, &cAnalyzeGenotype::SetTaskOrder,
- &cAnalyzeGenotype::CompareNULL, "(none)", ""));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, cString>
- ("sequence", "Genome Sequence",
- &cAnalyzeGenotype::GetSequence, &cAnalyzeGenotype::SetSequence,
- &cAnalyzeGenotype::CompareNULL, "(N/A)", ""));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, const cString &>
- ("alignment", "Aligned Sequence",
- &cAnalyzeGenotype::GetAlignedSequence,
- &cAnalyzeGenotype::SetAlignedSequence,
- &cAnalyzeGenotype::CompareNULL, "(N/A)", ""));
-
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, cString>
- ("task_list", "List of all tasks performed",
- &cAnalyzeGenotype::GetTaskList,
- (void (cAnalyzeGenotype::*)(cString)) NULL,
- &cAnalyzeGenotype::CompareNULL, "(N/A)", ""));
-
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, cString>
- ("link.tasksites", "Phenotype Map", &cAnalyzeGenotype::GetMapLink,
- (void (cAnalyzeGenotype::*)(cString)) NULL));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, cString>
- ("html.sequence", "Genome Sequence",
- &cAnalyzeGenotype::GetHTMLSequence,
- (void (cAnalyzeGenotype::*)(cString)) NULL,
- &cAnalyzeGenotype::CompareNULL, "(N/A)", ""));
-
- const cTaskLib & task_lib = m_world->GetEnvironment().GetTaskLib();
- for (int i = 0; i < task_lib.GetSize(); i++) {
- cString t_name, t_desc;
- t_name.Set("task.%d", i);
- t_desc = task_lib.GetTask(i).GetDesc();
- genotype_data_list.PushRear(new tArgDataEntry<cAnalyzeGenotype, int, int>
- (t_name, t_desc, &cAnalyzeGenotype::GetTaskCount, i,
- &cAnalyzeGenotype::CompareTaskCount));
- }
-
- // The remaining values should actually go in a seperate list called
- // "population_data_list", but for the moment we're going to put them
- // here so that we only need to worry about a single system to load and
- // save genotype information.
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("update", "Update Output",
- &cAnalyzeGenotype::GetUpdateDead, &cAnalyzeGenotype::SetUpdateDead));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("dom_num_cpus", "Number of Dominant Organisms",
- &cAnalyzeGenotype::GetNumCPUs, &cAnalyzeGenotype::SetNumCPUs));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("dom_depth", "Tree Depth of Dominant Genotype",
- &cAnalyzeGenotype::GetDepth, &cAnalyzeGenotype::SetDepth));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, int>
- ("dom_id", "Dominant Genotype ID",
- &cAnalyzeGenotype::GetID, &cAnalyzeGenotype::SetID));
- genotype_data_list.PushRear(new tDataEntry<cAnalyzeGenotype, cString>
- ("dom_sequence", "Dominant Genotype Sequence",
- &cAnalyzeGenotype::GetSequence, &cAnalyzeGenotype::SetSequence,
- &cAnalyzeGenotype::CompareNULL, "(N/A)", ""));
-}
-
-
-// Pass in the arguments for a command and fill out the entries in list
-// format....
-
-void cAnalyze::LoadGenotypeDataList(cStringList arg_list,
- tList< tDataEntryCommand<cAnalyzeGenotype> > & output_list)
-{
- // For the moment, just put everything into the output list.
- SetupGenotypeDataList();
-
- // If no args were given, load all of the stats.
- if (arg_list.GetSize() == 0) {
- tListIterator< tDataEntryBase<cAnalyzeGenotype> >
- genotype_data_it(genotype_data_list);
- while (genotype_data_it.Next() != (void *) NULL) {
- tDataEntryCommand<cAnalyzeGenotype> * entry_command =
- new tDataEntryCommand<cAnalyzeGenotype>(genotype_data_it.Get());
- output_list.PushRear(entry_command);
- }
- }
- // Otherwise, load only those listed.
- else {
- while (arg_list.GetSize() != 0) {
- // Setup the next entry
- cString cur_args = arg_list.Pop();
- cString cur_entry = cur_args.Pop(':');
- bool found_entry = false;
-
- // Scan the genotype data list for the current entry
- tListIterator< tDataEntryBase<cAnalyzeGenotype> >
- genotype_data_it(genotype_data_list);
-
- while (genotype_data_it.Next() != (void *) NULL) {
- if (genotype_data_it.Get()->GetName() == cur_entry) {
- tDataEntryCommand<cAnalyzeGenotype> * entry_command =
- new tDataEntryCommand<cAnalyzeGenotype>
- (genotype_data_it.Get(), cur_args);
- output_list.PushRear(entry_command);
- found_entry = true;
- break;
- }
- }
-
- // If the entry was not found, give a warning.
- if (found_entry == false) {
- int best_match = 1000;
- cString best_entry;
-
- genotype_data_it.Reset();
- while (genotype_data_it.Next() != (void *) NULL) {
- const cString & test_str = genotype_data_it.Get()->GetName();
- const int test_dist = cStringUtil::EditDistance(test_str, cur_entry);
- if (test_dist < best_match) {
- best_match = test_dist;
- best_entry = test_str;
- }
- }
-
- cerr << "Warning: Format entry \"" << cur_entry
- << "\" not found. Best match is \""
- << best_entry << "\"." << endl;
- }
-
- }
- }
-}
-
-
-void cAnalyze::AddLibraryDef(const cString & name,
- void (cAnalyze::*_fun)(cString))
-{
- command_lib.PushRear(new cAnalyzeCommandDef(name, _fun));
-}
-
-void cAnalyze::AddLibraryDef(const cString & name,
- void (cAnalyze::*_fun)(cString, tList<cAnalyzeCommand> &))
-{
- command_lib.PushRear(new cAnalyzeFlowCommandDef(name, _fun));
-}
-
-void cAnalyze::SetupCommandDefLibrary()
-{
- if (command_lib.GetSize() != 0) return; // Library already setup.
-
- AddLibraryDef("LOAD_ORGANISM", &cAnalyze::LoadOrganism);
- AddLibraryDef("LOAD_BASE_DUMP", &cAnalyze::LoadBasicDump);
- AddLibraryDef("LOAD_DETAIL_DUMP", &cAnalyze::LoadDetailDump);
- AddLibraryDef("LOAD_MULTI_DETAIL", &cAnalyze::LoadMultiDetail);
- AddLibraryDef("LOAD_SEQUENCE", &cAnalyze::LoadSequence);
- AddLibraryDef("LOAD_DOMINANT", &cAnalyze::LoadDominant);
- AddLibraryDef("LOAD_RESOURCES", &cAnalyze::LoadResources);
- AddLibraryDef("LOAD", &cAnalyze::LoadFile);
-
- // Reduction commands...
- AddLibraryDef("FIND_GENOTYPE", &cAnalyze::FindGenotype);
- AddLibraryDef("FIND_ORGANISM", &cAnalyze::FindOrganism);
- AddLibraryDef("FIND_LINEAGE", &cAnalyze::FindLineage);
- AddLibraryDef("FIND_SEX_LINEAGE", &cAnalyze::FindSexLineage);
- AddLibraryDef("FIND_CLADE", &cAnalyze::FindClade);
- AddLibraryDef("SAMPLE_ORGANISMS", &cAnalyze::SampleOrganisms);
- AddLibraryDef("SAMPLE_GENOTYPES", &cAnalyze::SampleGenotypes);
- AddLibraryDef("KEEP_TOP", &cAnalyze::KeepTopGenotypes);
-
- // Direct output commands...
- AddLibraryDef("PRINT", &cAnalyze::CommandPrint);
- AddLibraryDef("TRACE", &cAnalyze::CommandTrace);
- AddLibraryDef("PRINT_TASKS", &cAnalyze::CommandPrintTasks);
- AddLibraryDef("DETAIL", &cAnalyze::CommandDetail);
- AddLibraryDef("DETAIL_TIMELINE", &cAnalyze::CommandDetailTimeline);
- AddLibraryDef("DETAIL_BATCHES", &cAnalyze::CommandDetailBatches);
- AddLibraryDef("DETAIL_AVERAGE", &cAnalyze::CommandDetailAverage);
- AddLibraryDef("DETAIL_INDEX", &cAnalyze::CommandDetailIndex);
- AddLibraryDef("HISTOGRAM", &cAnalyze::CommandHistogram);
-
- // Population analysis commands...
- AddLibraryDef("PRINT_PHENOTYPES", &cAnalyze::CommandPrintPhenotypes);
- AddLibraryDef("PRINT_DIVERSITY", &cAnalyze::CommandPrintDiversity);
- AddLibraryDef("COMMUNITY_COMPLEXITY", &cAnalyze::AnalyzeCommunityComplexity);
-
- // Individual organism analysis...
- AddLibraryDef("LANDSCAPE", &cAnalyze::CommandLandscape);
- AddLibraryDef("FITNESS_MATRIX", &cAnalyze::CommandFitnessMatrix);
- AddLibraryDef("MAP", &cAnalyze::CommandMapTasks); // Deprecated...
- AddLibraryDef("MAP_TASKS", &cAnalyze::CommandMapTasks);
- AddLibraryDef("AVERAGE_MODULARITY", &cAnalyze::CommandAverageModularity);
- AddLibraryDef("MAP_MUTATIONS", &cAnalyze::CommandMapMutations);
- AddLibraryDef("ANALYZE_COMPLEXITY", &cAnalyze::AnalyzeComplexity);
- AddLibraryDef("ANALYZE_KNOCKOUTS", &cAnalyze::AnalyzeKnockouts);
- AddLibraryDef("ANALYZE_POP_COMPLEXITY", &cAnalyze::AnalyzePopComplexity);
- AddLibraryDef("MAP_DEPTH", &cAnalyze::CommandMapDepth);
-
- // Population comparison commands...
- AddLibraryDef("HAMMING", &cAnalyze::CommandHamming);
- AddLibraryDef("LEVENSTEIN", &cAnalyze::CommandLevenstein);
- AddLibraryDef("SPECIES", &cAnalyze::CommandSpecies);
- AddLibraryDef("RECOMBINE", &cAnalyze::CommandRecombine);
-
- // Lineage analysis commands...
- AddLibraryDef("ALIGN", &cAnalyze::CommandAlign);
- AddLibraryDef("ANALYZE_NEWINFO", &cAnalyze::AnalyzeNewInfo);
-
- // Build input files for avida...
- AddLibraryDef("WRITE_CLONE", &cAnalyze::WriteClone);
- AddLibraryDef("WRITE_INJECT_EVENTS", &cAnalyze::WriteInjectEvents);
- AddLibraryDef("WRITE_COMPETITION", &cAnalyze::WriteCompetition);
-
- // Automated analysis
- AddLibraryDef("ANALYZE_MUTS", &cAnalyze::AnalyzeMuts);
- AddLibraryDef("ANALYZE_INSTRUCTIONS", &cAnalyze::AnalyzeInstructions);
- AddLibraryDef("ANALYZE_INST_POP", &cAnalyze::AnalyzeInstPop);
- AddLibraryDef("ANALYZE_BRANCHING", &cAnalyze::AnalyzeBranching);
- AddLibraryDef("ANALYZE_MUTATION_TRACEBACK",
- &cAnalyze::AnalyzeMutationTraceback);
- AddLibraryDef("ANALYZE_EPISTASIS", &cAnalyze::AnalyzeEpistasis);
- AddLibraryDef("ANALYZE_MATE_SELECTION", &cAnalyze::AnalyzeMateSelection);
- AddLibraryDef("ANALYZE_COMPLEXITY_DELTA", &cAnalyze::AnalyzeComplexityDelta);
-
- // Environment manipulation
- AddLibraryDef("ENVIRONMENT", &cAnalyze::EnvironmentSetup);
-
- // Documantation...
- AddLibraryDef("HELPFILE", &cAnalyze::CommandHelpfile);
-
- // Control commands...
- AddLibraryDef("SET", &cAnalyze::VarSet);
- AddLibraryDef("SET_BATCH", &cAnalyze::BatchSet);
- AddLibraryDef("NAME_BATCH", &cAnalyze::BatchName);
- AddLibraryDef("TAG_BATCH", &cAnalyze::BatchTag);
- AddLibraryDef("PURGE_BATCH", &cAnalyze::BatchPurge);
- AddLibraryDef("DUPLICATE", &cAnalyze::BatchDuplicate);
- AddLibraryDef("RECALCULATE", &cAnalyze::BatchRecalculate);
- AddLibraryDef("RENAME", &cAnalyze::BatchRename);
- AddLibraryDef("STATUS", &cAnalyze::PrintStatus);
- AddLibraryDef("DEBUG", &cAnalyze::PrintDebug);
- AddLibraryDef("ECHO", &cAnalyze::PrintDebug);
- AddLibraryDef("VERBOSE", &cAnalyze::ToggleVerbose);
- AddLibraryDef("INCLUDE", &cAnalyze::IncludeFile);
- AddLibraryDef("SYSTEM", &cAnalyze::CommandSystem);
- AddLibraryDef("INTERACTIVE", &cAnalyze::CommandInteractive);
- AddLibraryDef("PRINT_TEST_CPU_RESOURCES", &cAnalyze::PrintTestCPUResources);
-
- // Functions...
- AddLibraryDef("FUNCTION", &cAnalyze::FunctionCreate);
-
- // Flow commands...
- AddLibraryDef("FOREACH", &cAnalyze::CommandForeach);
- AddLibraryDef("FORRANGE", &cAnalyze::CommandForRange);
-}
-
-cAnalyzeCommandDefBase * cAnalyze::FindAnalyzeCommandDef(const cString & name)
-{
- SetupCommandDefLibrary();
-
- tListIterator<cAnalyzeCommandDefBase> lib_it(command_lib);
- while (lib_it.Next() != (void *) NULL) {
- if (lib_it.Get()->GetName() == name) break;
- }
-
- return lib_it.Get();
-}
-
-void cAnalyze::RunInteractive()
-{
- cout << "Entering interactive mode..." << endl;
-
- char text_input[2048];
- while (true) {
- cout << ">> ";
- cout.flush();
- cin.getline(text_input, 2048);
- cString cur_input(text_input);
- cString command = cur_input.PopWord();
- command.ToUpper();
-
- cAnalyzeCommand * cur_command;
- cAnalyzeCommandDefBase * command_def = FindAnalyzeCommandDef(command);
- if (command == "") {
- // Don't worry about blank lines...
- continue;
- }
- else if (command == "END" || command == "QUIT" || command == "EXIT") {
- // We are done with interactive mode...
- break;
- }
- else if (command_def != NULL && command_def->IsFlowCommand() == true) {
- // This code has a body to it... fill it out!
- cur_command = new cAnalyzeFlowCommand(command, cur_input);
- InteractiveLoadCommandList(*(cur_command->GetCommandList()));
- }
- else {
- // This is a normal command...
- cur_command = new cAnalyzeCommand(command, cur_input);
- }
-
- cString args = cur_command->GetArgs();
- PreProcessArgs(args);
-
- cAnalyzeCommandDefBase * command_fun = FindAnalyzeCommandDef(command);
-
- // First check for built-in functions...
- if (command_fun != NULL) command_fun->Run(this, args, *cur_command);
-
- // Then for user defined functions
- else if (FunctionRun(command, args) == true) { }
-
- // Otherwise, give an error.
- else cerr << "Error: Unknown command '" << command << "'." << endl;
- }
-}
Deleted: branches/brysonda/source/main/cAnalyze.h
===================================================================
--- branches/brysonda/source/main/cAnalyze.h 2005-11-04 00:39:30 UTC (rev 368)
+++ branches/brysonda/source/main/cAnalyze.h 2005-11-04 03:50:46 UTC (rev 369)
@@ -1,255 +0,0 @@
-//////////////////////////////////////////////////////////////////////////////
-// Copyright (C) 1993 - 2003 California Institute of Technology //
-// //
-// Read the COPYING and README files, or contact 'avida at alife.org', //
-// before continuing. SOME RESTRICTIONS MAY APPLY TO USE OF THIS FILE. //
-//////////////////////////////////////////////////////////////////////////////
-
-#ifndef ANALYZE_HH
-#define ANALYZE_HH
-
-#include <vector>
-
-#ifndef DATA_FILE_MANAGER_HH
-#include "cDataFileManager.h"
-#endif
-#ifndef GENOTYPE_BATCH_HH
-#include "cGenotypeBatch.h"
-#endif
-#ifndef cRandom_h
-#include "cRandom.h"
-#endif
-#ifndef STRING_HH
-#include "cString.h"
-#endif
-#ifndef STRING_LIST_HH
-#include "cStringList.h"
-#endif
-#ifndef TLIST_HH
-#include "tList.h"
-#endif
-
-#define MAX_BATCHES 2000
-
-// cAnalyze : The master analyze object.
-
-class cGenotypeBatch; // array
-template <class T> class tList; // aggregate
-class cAnalyzeCommand;
-class cAnalyzeFunction;
-class cAnalyzeCommandDefBase;
-class cString; // aggregate
-class cDataFileManager; // aggregate
-template <class T> class tDataEntryBase;
-class cInstSet;
-class cRandom; // aggregate
-class cString; // aggregate
-class cAnalyzeGenotype;
-class cInitFile;
-class cStringList; // aggregate
-template <class T> class tDataEntryCommand;
-template <class T> class tListIterator;
-class cEnvironment;
-class cWorld;
-
-class cAnalyze {
-private:
- int cur_batch;
- cGenotypeBatch batch[MAX_BATCHES];
- tList<cAnalyzeCommand> command_list;
- tList<cAnalyzeFunction> function_list;
- tList<cAnalyzeCommandDefBase> command_lib;
- cString variables[26];
- cString local_variables[26];
- cString arg_variables[10];
-
- cWorld* m_world;
- cInstSet& inst_set;
-
- // This is the storage for the resource information from resource.dat. It
- // is a pair of the update and a vector of the resource concentrations
- std::vector<std::pair<int, std::vector<double> > > resources;
-
- bool verbose; // Should details be output to command line?
- int interactive_depth; // How nested are we if in interactive mode?
-
- cDataFileManager data_file_manager;
- tList< tDataEntryBase<cAnalyzeGenotype> > genotype_data_list;
-
- cRandom random;
-
-private:
- // Pop specific types of arguments from an arg list.
- cString PopDirectory(cString & in_string, const cString & default_dir);
- int PopBatch(const cString & in_string);
- cAnalyzeGenotype * PopGenotype(cString gen_desc, int batch_id=-1);
- cString & GetVariable(const cString & varname);
-
- // Other arg-list methods
- void LoadCommandList(cInitFile & init_file, tList<cAnalyzeCommand> & clist);
- void InteractiveLoadCommandList(tList<cAnalyzeCommand> & clist);
- void PreProcessArgs(cString & args);
- void ProcessCommands(tList<cAnalyzeCommand> & clist);
-
- void SetupGenotypeDataList();
- void LoadGenotypeDataList(cStringList arg_list,
- tList< tDataEntryCommand<cAnalyzeGenotype> > & output_list);
-
- void AddLibraryDef(const cString & name, void (cAnalyze::*_fun)(cString));
- void AddLibraryDef(const cString & name,
- void (cAnalyze::*_fun)(cString, tList<cAnalyzeCommand> &));
- cAnalyzeCommandDefBase * FindAnalyzeCommandDef(const cString & name);
- void SetupCommandDefLibrary();
- bool FunctionRun(const cString & fun_name, cString args);
-
- // Batch management...
- int BatchUtil_GetMaxLength(int batch_id=-1);
-
- // Comamnd helpers...
- void CommandDetail_Header(std::ostream & fp, int format_type,
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > & output_it,
- int time_step=-1);
- void CommandDetail_Body(std::ostream & fp, int format_type,
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > & output_it,
- int time_step=-1, int max_time=1);
- void CommandDetailAverage_Body(std::ostream & fp, int num_arguments,
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > & output_it);
- void CommandHistogram_Header(std::ostream & fp, int format_type,
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > & output_it);
- void CommandHistogram_Body(std::ostream & fp, int format_type,
- tListIterator< tDataEntryCommand<cAnalyzeGenotype> > & output_it);
-
-private:
- // Loading methods...
- void LoadOrganism(cString cur_string);
- void LoadBasicDump(cString cur_string);
- void LoadDetailDump(cString cur_string);
- void LoadMultiDetail(cString cur_string);
- void LoadSequence(cString cur_string);
- void LoadDominant(cString cur_string);
- // Clears the current time oriented list of resources and loads in a new one
- // from a file specified by the user, or resource.dat by default.
- void LoadResources(cString cur_string);
- void LoadFile(cString cur_string);
-
- // Reduction
- void FindGenotype(cString cur_string);
- void FindOrganism(cString cur_string);
- void FindLineage(cString cur_string);
- void FindSexLineage(cString cur_string);
- void FindClade(cString cur_string);
- void SampleOrganisms(cString cur_string);
- void SampleGenotypes(cString cur_string);
- void KeepTopGenotypes(cString cur_string);
-
- // Direct Output Commands...
- void CommandPrint(cString cur_string);
- void CommandTrace(cString cur_string);
- void CommandTraceWithResources(cString cur_string);
- void CommandPrintTasks(cString cur_string);
- void CommandDetail(cString cur_string);
- void CommandDetailTimeline(cString cur_string);
- void CommandDetailBatches(cString cur_string);
- void CommandDetailAverage(cString cur_string);
- void CommandDetailIndex(cString cur_string);
- void CommandHistogram(cString cur_string);
-
- // Population Analysis Commands...
- void CommandPrintPhenotypes(cString cur_string);
- void CommandPrintDiversity(cString cur_string);
- void PhyloCommunityComplexity(cString cur_string);
- void AnalyzeCommunityComplexity(cString cur_string);
-
- // Individual Organism Analysis...
- void CommandLandscape(cString cur_string);
- void CommandFitnessMatrix(cString cur_string);
- void CommandMapTasks(cString cur_string);
- void CommandAverageModularity(cString cur_string);
- void CommandMapMutations(cString cur_string);
- void CommandMapDepth(cString cur_string);
-
- // Population Comparison Commands...
- void CommandHamming(cString cur_string);
- void CommandLevenstein(cString cur_string);
- void CommandSpecies(cString cur_string);
- void CommandRecombine(cString cur_string);
-
- // Lineage Analysis Commands...
- void CommandAlign(cString cur_string);
- void AnalyzeNewInfo(cString cur_string);
-
- // Build Input Files for Avida
- void WriteClone(cString cur_string);
- void WriteInjectEvents(cString cur_string);
- void WriteCompetition(cString cur_string);
-
- // Automated analysis...
- void AnalyzeMuts(cString cur_string);
- void AnalyzeInstructions(cString cur_string);
- void AnalyzeInstPop(cString cur_string);
- void AnalyzeBranching(cString cur_string);
- void AnalyzeMutationTraceback(cString cur_string);
- void AnalyzeComplexity(cString cur_string);
- void AnalyzeKnockouts(cString cur_string);
- void AnalyzePopComplexity(cString cur_string);
- void AnalyzeEpistasis(cString cur_string);
- void AnalyzeMateSelection(cString cur_string);
- void AnalyzeComplexityDelta(cString cur_string);
-
- // Environment Manipulation
- void EnvironmentSetup(cString cur_string);
-
- // Documentation...
- void CommandHelpfile(cString cur_string);
- void CommandDocfile(cString cur_string);
-
- // Control...
- void VarSet(cString cur_string);
- void BatchSet(cString cur_string);
- void BatchName(cString cur_string);
- void BatchTag(cString cur_string);
- void BatchPurge(cString cur_string);
- void BatchDuplicate(cString cur_string);
- void BatchRecalculate(cString cur_string);
- void BatchRename(cString cur_string);
- void PrintStatus(cString cur_string);
- void PrintDebug(cString cur_string);
- void ToggleVerbose(cString cur_string);
- void IncludeFile(cString cur_string);
- void CommandSystem(cString cur_string);
- void CommandInteractive(cString cur_string);
- void PrintTestCPUResources(cString cur_string);
-
- // Functions...
- void FunctionCreate(cString cur_string, tList<cAnalyzeCommand> & clist);
- // Looks up the resource concentrations that are the closest to the
- // given update and then fill in those concentrations into the environment.
- void FillResources(int update);
- // Analyze the entropy of genotype under default environment
- double AnalyzeEntropy(cAnalyzeGenotype * genotype, double mut_rate);
- // Analyze the entropy of child given parent and default environment
- double AnalyzeEntropyGivenParent(cAnalyzeGenotype * genotype,
- cAnalyzeGenotype * parent,
- double mut_rate);
- // Calculate the increased information in genotype1 comparing to genotype2
- // line by line. If information in genotype1 is less than that in genotype2
- // in a line, increasing is 0. Usually this is used for child-parent comparison.
- double IncreasedInfo(cAnalyzeGenotype * genotype1,
- cAnalyzeGenotype * genotype2,
- double mut_rate);
-
- // Flow Control...
- void CommandForeach(cString cur_string, tList<cAnalyzeCommand> & clist);
- void CommandForRange(cString cur_string, tList<cAnalyzeCommand> & clist);
-
-private:
- // disabled copy constructor.
- cAnalyze(const cAnalyze &);
-public:
- cAnalyze(cWorld* world);
- ~cAnalyze();
-
- void RunInteractive();
-};
-
-#endif
Deleted: branches/brysonda/source/main/cAnalyzeCommand.h
===================================================================
--- branches/brysonda/source/main/cAnalyzeCommand.h 2005-11-04 00:39:30 UTC (rev 368)
+++ branches/brysonda/source/main/cAnalyzeCommand.h 2005-11-04 03:50:46 UTC (rev 369)
@@ -1,44 +0,0 @@
-//////////////////////////////////////////////////////////////////////////////
-// Copyright (C) 1993 - 2003 California Institute of Technology //
-// //
-// Read the COPYING and README files, or contact 'avida at alife.org', //
-// before continuing. SOME RESTRICTIONS MAY APPLY TO USE OF THIS FILE. //
-//////////////////////////////////////////////////////////////////////////////
-
-#ifndef ANALYZE_COMMAND_HH
-#define ANALYZE_COMMAND_HH
-
-#ifndef STRING_HH
-#include "cString.h"
-#endif
-
-// cAnalyzeCommand : A command in a loaded program
-
-class cString; // aggregate
-template <class T> class tList;
-
-class cAnalyzeCommand {
-protected:
- cString command;
- cString args;
-private:
- // disabled copy constructor.
- cAnalyzeCommand(const cAnalyzeCommand &);
-public:
- cAnalyzeCommand(const cString & _command, const cString & _args)
- : command(_command), args(_args) { command.ToUpper(); }
- virtual ~cAnalyzeCommand() { ; }
-
- const cString & GetCommand() { return command; }
- const cString & GetArgs() const { return args; }
- cString GetArgs() { return args; }
- virtual tList<cAnalyzeCommand> * GetCommandList() { return NULL; }
-
- /*
- added to satisfy Boost.Python; the semantics are fairly useless --
- equality of two references means that they refer to the same object.
- */
- bool operator==(const cAnalyzeCommand &in) const { return &in == this; }
-};
-
-#endif
Deleted: branches/brysonda/source/main/cAnalyzeCommandDef.h
===================================================================
--- branches/brysonda/source/main/cAnalyzeCommandDef.h 2005-11-04 00:39:30 UTC (rev 368)
+++ branches/brysonda/source/main/cAnalyzeCommandDef.h 2005-11-04 03:50:46 UTC (rev 369)
@@ -1,37 +0,0 @@
-//////////////////////////////////////////////////////////////////////////////
-// Copyright (C) 1993 - 2003 California Institute of Technology //
-// //
-// Read the COPYING and README files, or contact 'avida at alife.org', //
-// before continuing. SOME RESTRICTIONS MAY APPLY TO USE OF THIS FILE. //
-//////////////////////////////////////////////////////////////////////////////
-
-#ifndef ANALYZE_COMMAND_DEF_HH
-#define ANALYZE_COMMAND_DEF_HH
-
-#ifndef ANALYZE_COMMAND_DEF_BASE_HH
-#include "cAnalyzeCommandDefBase.h"
-#endif
-#ifndef STRING_HH
-#include "cString.h"
-#endif
-
-class cAnalyze;
-class cString; // aggregate
-class cAnalyzeCommand;
-
-class cAnalyzeCommandDef : public cAnalyzeCommandDefBase {
-private:
- void (cAnalyze::*CommandFunction)(cString);
-public:
- cAnalyzeCommandDef(const cString & _name, void (cAnalyze::*_cf)(cString))
- : cAnalyzeCommandDefBase(_name), CommandFunction(_cf) { ; }
- virtual ~cAnalyzeCommandDef() { ; }
-
- virtual void Run(cAnalyze * analyze, const cString & args,
- cAnalyzeCommand & command) const {
- (void) command; // used in other types of command defininitions.
- (analyze->*CommandFunction)(args);
- }
-};
-
-#endif
Deleted: branches/brysonda/source/main/cAnalyzeCommandDefBase.h
===================================================================
--- branches/brysonda/source/main/cAnalyzeCommandDefBase.h 2005-11-04 00:39:30 UTC (rev 368)
+++ branches/brysonda/source/main/cAnalyzeCommandDefBase.h 2005-11-04 03:50:46 UTC (rev 369)
@@ -1,33 +0,0 @@
-//////////////////////////////////////////////////////////////////////////////
-// Copyright (C) 1993 - 2003 California Institute of Technology //
-// //
-// Read the COPYING and README files, or contact 'avida at alife.org', //
-// before continuing. SOME RESTRICTIONS MAY APPLY TO USE OF THIS FILE. //
-//////////////////////////////////////////////////////////////////////////////
-
-#ifndef ANALYZE_COMMAND_DEF_BASE_HH
-#define ANALYZE_COMMAND_DEF_BASE_HH
-
-#ifndef STRING_HH
-#include "cString.h"
-#endif
-
-class cString; // aggregate
-class cAnalyze;
-class cAnalyzeCommand;
-
-class cAnalyzeCommandDefBase {
-protected:
- cString name;
-public:
- cAnalyzeCommandDefBase(const cString & _name) : name(_name) { ; }
- virtual ~cAnalyzeCommandDefBase() { ; }
-
- virtual void Run(cAnalyze * analyze, const cString & args,
- cAnalyzeCommand & command) const = 0;
- virtual bool IsFlowCommand() { return false; }
-
- const cString & GetName() const { return name; }
-};
-
-#endif
Deleted: branches/brysonda/source/main/cAnalyzeFlowCommand.h
===================================================================
--- branches/brysonda/source/main/cAnalyzeFlowCommand.h 2005-11-04 00:39:30 UTC (rev 368)
+++ branches/brysonda/source/main/cAnalyzeFlowCommand.h 2005-11-04 03:50:46 UTC (rev 369)
@@ -1,36 +0,0 @@
-//////////////////////////////////////////////////////////////////////////////
-// Copyright (C) 1993 - 2003 California Institute of Technology //
-// //
-// Read the COPYING and README files, or contact 'avida at alife.org', //
-// before continuing. SOME RESTRICTIONS MAY APPLY TO USE OF THIS FILE. //
-//////////////////////////////////////////////////////////////////////////////
-
-#ifndef ANALYZE_FLOW_COMMAND_HH
-#define ANALYZE_FLOW_COMMAND_HH
-
-#ifndef ANALYZE_COMMAND_HH
-#include "cAnalyzeCommand.h"
-#endif
-#ifndef TLIST_HH
-#include "tList.h"
-#endif
-
-// cAnalyzeFlowCommand : A cAnalyzeCommand containing other commands
-
-template <class T> class tList; // aggregate
-class cString;
-
-class cAnalyzeFlowCommand : public cAnalyzeCommand {
-protected:
- tList<cAnalyzeCommand> command_list;
-public:
- cAnalyzeFlowCommand(const cString & _command, const cString & _args)
- : cAnalyzeCommand(_command, _args) { ; }
- virtual ~cAnalyzeFlowCommand() {
- while ( command_list.GetSize() > 0 ) delete command_list.Pop();
- }
-
- tList<cAnalyzeCommand> * GetCommandList() { return &command_list; }
-};
-
-#endif
Deleted: branches/brysonda/source/main/cAnalyzeFlowCommandDef.h
===================================================================
--- branches/brysonda/source/main/cAnalyzeFlowCommandDef.h 2005-11-04 00:39:30 UTC (rev 368)
+++ branches/brysonda/source/main/cAnalyzeFlowCommandDef.h 2005-11-04 03:50:46 UTC (rev 369)
@@ -1,46 +0,0 @@
-//////////////////////////////////////////////////////////////////////////////
-// Copyright (C) 1993 - 2003 California Institute of Technology //
-// //
-// Read the COPYING and README files, or contact 'avida at alife.org', //
-// before continuing. SOME RESTRICTIONS MAY APPLY TO USE OF THIS FILE. //
-//////////////////////////////////////////////////////////////////////////////
-
-#ifndef ANALYZE_FLOW_COMMAND_DEF_HH
-#define ANALYZE_FLOW_COMMAND_DEF_HH
-
-#ifndef ANALYZE_COMMAND_HH
-#include "cAnalyzeCommand.h"
-#endif
-#ifndef ANALYZE_COMMAND_DEF_BASE_HH
-#include "cAnalyzeCommandDefBase.h"
-#endif
-#ifndef STRING_HH
-#include "cString.h"
-#endif
-#ifndef TLIST_HH
-#include "tList.h"
-#endif
-
-class cAnalyze;
-template <class T> class tList;
-class cAnalyzeCommand; // access
-class cString; // aggregate
-
-class cAnalyzeFlowCommandDef : public cAnalyzeCommandDefBase {
-private:
- void (cAnalyze::*CommandFunction)(cString, tList<cAnalyzeCommand> &);
-public:
- cAnalyzeFlowCommandDef(const cString &_name,
- void (cAnalyze::*_cf)(cString, tList<cAnalyzeCommand> &))
- : cAnalyzeCommandDefBase(_name), CommandFunction(_cf) { ; }
- virtual ~cAnalyzeFlowCommandDef() { ; }
-
- virtual void Run(cAnalyze * analyze, const cString & args,
- cAnalyzeCommand & command) const {
- (analyze->*CommandFunction)(args, *(command.GetCommandList()) );
- }
-
- virtual bool IsFlowCommand() { return true; }
-};
-
-#endif
Deleted: branches/brysonda/source/main/cAnalyzeFunction.h
===================================================================
--- branches/brysonda/source/main/cAnalyzeFunction.h 2005-11-04 00:39:30 UTC (rev 368)
+++ branches/brysonda/source/main/cAnalyzeFunction.h 2005-11-04 03:50:46 UTC (rev 369)
@@ -1,44 +0,0 @@
-//////////////////////////////////////////////////////////////////////////////
-// Copyright (C) 1993 - 2003 California Institute of Technology //
-// //
-// Read the COPYING and README files, or contact 'avida at alife.org', //
-// before continuing. SOME RESTRICTIONS MAY APPLY TO USE OF THIS FILE. //
-//////////////////////////////////////////////////////////////////////////////
-
-#ifndef ANALYZE_FUNCTION_HH
-#define ANALYZE_FUNCTION_HH
-
-#ifndef ANALYZE_COMMAND_HH
-#include "cAnalyzeCommand.h"
-#endif
-#ifndef STRING_HH
-#include "cString.h"
-#endif
-#ifndef TLIST_HH
-#include "tList.h"
-#endif
-
-// cAnalyzeFunction : User-defined function
-
-class cString; // aggregate
-template <class T> class tList; // aggregate
-class cAnalyzeCommand;
-
-class cAnalyzeFunction {
-private:
- cString name;
- tList<cAnalyzeCommand> command_list;
-private:
- // disabled copy constructor.
- cAnalyzeFunction(const cAnalyzeFunction &);
-public:
- cAnalyzeFunction(const cString & _name) : name(_name) { ; }
- ~cAnalyzeFunction() {
- while ( command_list.GetSize() > 0 ) delete command_list.Pop();
- }
-
- const cString & GetName() { return name; }
- tList<cAnalyzeCommand> * GetCommandList() { return &command_list; }
-};
-
-#endif
Deleted: branches/brysonda/source/main/cAnalyzeGenotype.cc
===================================================================
--- branches/brysonda/source/main/cAnalyzeGenotype.cc 2005-11-04 00:39:30 UTC (rev 368)
+++ branches/brysonda/source/main/cAnalyzeGenotype.cc 2005-11-04 03:50:46 UTC (rev 369)
@@ -1,347 +0,0 @@
-//////////////////////////////////////////////////////////////////////////////
-// Copyright (C) 1993 - 2003 California Institute of Technology //
-// //
-// Read the COPYING and README files, or contact 'avida at alife.org', //
-// before continuing. SOME RESTRICTIONS MAY APPLY TO USE OF THIS FILE. //
-//////////////////////////////////////////////////////////////////////////////
-
-#include "cAnalyzeGenotype.h"
-
-#include "cCPUTestInfo.h"
-#include "cInstSet.h"
-#include "cLandscape.h"
-#include "cOrganism.h"
-#include "cPhenotype.h"
-#include "cTestCPU.h"
-#include "cEnvironment.h"
-#include "cHardwareManager.h"
-#include "cWorld.h"
-
-using namespace std;
-
-//////////////////////
-// cAnalyzeGenotype
-//////////////////////
-
-cAnalyzeGenotype::cAnalyzeGenotype(cWorld* world, cString symbol_string, cInstSet& in_inst_set)
- : m_world(world)
- , genome(symbol_string)
- , inst_set(in_inst_set)
- , name("")
- , aligned_sequence("")
- , tag("")
- , viable(false)
- , id_num(-1)
- , parent_id(-1)
- , parent2_id(-1)
- , num_cpus(0)
- , total_cpus(0)
- , update_born(0)
- , update_dead(0)
- , depth(0)
- , length(0)
- , copy_length(0)
- , exe_length(0)
- , merit(0.0)
- , gest_time(INT_MAX)
- , fitness(0.0)
- , errors(0)
- , task_counts(0)
- , fitness_ratio(0.0)
- , efficiency_ratio(0.0)
- , comp_merit_ratio(0.0)
- , parent_dist(0)
- , ancestor_dist(0)
- , parent_muts("")
- , landscape_stats(NULL)
-{
- // Make sure that the sequences jive with the inst_set
- for (int i = 0; i < genome.GetSize(); i++) {
- if (genome[i].GetOp() >= inst_set.GetSize()) {
- cerr << "Error: Trying to load instruction " << genome[i].GetOp()
- << ". Max in set is" << inst_set.GetSize() - 1
- << endl;
- }
- }
-}
-
-cAnalyzeGenotype::cAnalyzeGenotype(cWorld* world, const cGenome& _genome, cInstSet& in_inst_set)
- : m_world(world)
- , genome(_genome)
- , inst_set(in_inst_set)
- , name("")
- , aligned_sequence("")
- , tag("")
- , viable(false)
- , id_num(-1)
- , parent_id(-1)
- , parent2_id(-1)
- , num_cpus(0)
- , total_cpus(0)
- , update_born(0)
- , update_dead(0)
- , depth(0)
- , length(0)
- , copy_length(0)
- , exe_length(0)
- , merit(0.0)
- , gest_time(INT_MAX)
- , fitness(0.0)
- , errors(0)
- , task_counts(0)
- , fitness_ratio(0.0)
- , efficiency_ratio(0.0)
- , comp_merit_ratio(0.0)
- , parent_dist(0)
- , ancestor_dist(0)
- , parent_muts("")
- , landscape_stats(NULL)
-{
-}
-
-cAnalyzeGenotype::cAnalyzeGenotype(const cAnalyzeGenotype & _gen)
- : m_world(_gen.m_world)
- , genome(_gen.genome)
- , inst_set(_gen.inst_set)
- , name(_gen.name)
- , aligned_sequence(_gen.aligned_sequence)
- , tag(_gen.tag)
- , viable(_gen.viable)
- , id_num(_gen.id_num)
- , parent_id(_gen.parent_id)
- , parent2_id(_gen.parent2_id)
- , num_cpus(_gen.num_cpus)
- , total_cpus(_gen.total_cpus)
- , update_born(_gen.update_born)
- , update_dead(_gen.update_dead)
- , depth(_gen.depth)
- , length(_gen.length)
- , copy_length(_gen.copy_length)
- , exe_length(_gen.exe_length)
- , merit(_gen.merit)
- , gest_time(_gen.gest_time)
- , fitness(_gen.fitness)
- , errors(_gen.errors)
- , task_counts(_gen.task_counts)
- , fitness_ratio(_gen.fitness_ratio)
- , efficiency_ratio(_gen.efficiency_ratio)
- , comp_merit_ratio(_gen.comp_merit_ratio)
- , parent_dist(_gen.parent_dist)
- , ancestor_dist(_gen.ancestor_dist)
- , parent_muts(_gen.parent_muts)
- , landscape_stats(NULL)
-{
- if (_gen.landscape_stats != NULL) {
- landscape_stats = new cAnalyzeLandscape;
- *landscape_stats = *(_gen.landscape_stats);
- }
-}
-
-cAnalyzeGenotype::~cAnalyzeGenotype()
-{
- if (landscape_stats != NULL) delete landscape_stats;
-}
-
-
-int cAnalyzeGenotype::CalcMaxGestation() const
-{
- return m_world->GetConfig().TEST_CPU_TIME_MOD.Get() * genome.GetSize();
-}
-
-void cAnalyzeGenotype::CalcLandscape() const
-{
- if (landscape_stats != NULL) return;
-
- landscape_stats = new cAnalyzeLandscape;
- cLandscape landscape(m_world, genome, inst_set);
- landscape.Process(1);
- landscape_stats->frac_dead = landscape.GetProbDead();
- landscape_stats->frac_neg = landscape.GetProbNeg();
- landscape_stats->frac_neut = landscape.GetProbNeut();
- landscape_stats->frac_pos = landscape.GetProbPos();
- landscape_stats->complexity = landscape.GetComplexity();
-}
-
-void cAnalyzeGenotype::Recalculate(cAnalyzeGenotype * parent_genotype)
-{
- // Build the test info for printing.
- cCPUTestInfo test_info;
- test_info.TestThreads();
- // test_info.TraceTaskOrder();
-
- // DDD - This does some 'interesting' things with the instruction set
-
- // Use the inst lib for this genotype... and syncrhonize environment
-
- // Backup old instruction set, update with new
- cInstSet env_inst_set_backup = m_world->GetHardwareManager().GetInstSet();
- m_world->GetHardwareManager().GetInstSet() = inst_set;
-
- m_world->GetTestCPU().TestGenome(test_info, genome);
-
- // Restore the instruction set
- m_world->GetHardwareManager().GetInstSet() = env_inst_set_backup;
-
- viable = test_info.IsViable();
-
- cOrganism * test_organism = test_info.GetTestOrganism();
- cPhenotype & test_phenotype = test_organism->GetPhenotype();
-
- length = test_organism->GetGenome().GetSize();
- copy_length = test_phenotype.GetCopiedSize();
- exe_length = test_phenotype.GetExecutedSize();
- merit = test_phenotype.GetMerit().GetDouble();
- gest_time = test_phenotype.GetGestationTime();
- fitness = test_phenotype.GetFitness();
- errors = test_phenotype.GetLastNumErrors();
- div_type = test_phenotype.GetDivType();
- mate_id = test_phenotype.MateSelectID();
- task_counts = test_phenotype.GetLastTaskCount();
-
- // Setup a new parent stats if we have a parent to work with.
- if (parent_genotype != NULL) {
- fitness_ratio = GetFitness() / parent_genotype->GetFitness();
- efficiency_ratio = GetEfficiency() / parent_genotype->GetEfficiency();
- comp_merit_ratio = GetCompMerit() / parent_genotype->GetCompMerit();
- parent_dist = cStringUtil::EditDistance(genome.AsString(),
- parent_genotype->GetGenome().AsString(), parent_muts);
- ancestor_dist = parent_genotype->GetAncestorDist() + parent_dist;
- }
-}
-
-
-void cAnalyzeGenotype::PrintTasks(ofstream & fp, int min_task, int max_task)
-{
- if (max_task == -1) max_task = task_counts.GetSize();
-
- for (int i = min_task; i < max_task; i++) {
- fp << task_counts[i] << " ";
- }
-}
-
-void cAnalyzeGenotype::SetSequence(cString _sequence)
-{
- cGenome new_genome(_sequence);
- genome = new_genome;
-}
-
-
-double cAnalyzeGenotype::GetFracDead() const
-{
- CalcLandscape(); // Make sure the landscape is calculated...
- return landscape_stats->frac_dead;
-}
-
-double cAnalyzeGenotype::GetFracNeg() const
-{
- CalcLandscape(); // Make sure the landscape is calculated...
- return landscape_stats->frac_neg;
-}
-
-double cAnalyzeGenotype::GetFracNeut() const
-{
- CalcLandscape(); // Make sure the landscape is calculated...
- return landscape_stats->frac_neut;
-}
-
-double cAnalyzeGenotype::GetFracPos() const
-{
- CalcLandscape(); // Make sure the landscape is calculated...
- return landscape_stats->frac_pos;
-}
-
-double cAnalyzeGenotype::GetComplexity() const
-{
- CalcLandscape(); // Make sure the landscape is calculated...
- return landscape_stats->complexity;
-}
-
-cString cAnalyzeGenotype::GetTaskList() const
-{
- const int num_tasks = task_counts.GetSize();
- cString out_string(num_tasks);
-
- for (int i = 0; i < num_tasks; i++) {
- const int cur_count = task_counts[i];
- if (cur_count < 10) {
- out_string[i] = '0' + cur_count;
- }
- else if (cur_count < 30) {
- out_string[i] = 'X';
- }
- else if (cur_count < 80) {
- out_string[i] = 'L';
- }
- else if (cur_count < 300) {
- out_string[i] = 'C';
- }
- else if (cur_count < 800) {
- out_string[i] = 'D';
- }
- else if (cur_count < 3000) {
- out_string[i] = 'M';
- }
- else {
- out_string[i] = '+';
- }
- }
-
- return out_string;
-}
-
-
-cString cAnalyzeGenotype::GetHTMLSequence() const
-{
- cString text_genome = genome.AsString();
- cString html_code("<tt>");
-
- cString diff_info = parent_muts;
- char mut_type = 'N';
- int mut_pos = -1;
-
- cString cur_mut = diff_info.Pop(',');
- if (cur_mut != "") {
- mut_type = cur_mut[0];
- cur_mut.ClipFront(1); cur_mut.ClipEnd(1);
- mut_pos = cur_mut.AsInt();
- }
-
- int ins_count = 0;
- for (int i = 0; i < genome.GetSize(); i++) {
- char symbol = text_genome[i];
- if (i != mut_pos) html_code += symbol;
- else {
- // Figure out the information for the type of mutation we had...
- cString color;
- if (mut_type == 'M') {
- color = "#FF0000";
- } else if (mut_type == 'I') {
- color = "#00FF00";
- ins_count++;
- } else { // if (mut_type == 'D') {
- color = "#0000FF";
- symbol = '*';
- i--; // Rewind - we didn't read the handle character yet!
- }
-
- // Move on to the next mutation...
- cur_mut = diff_info.Pop(',');
- if (cur_mut != "") {
- mut_type = cur_mut[0];
- cur_mut.ClipFront(1); cur_mut.ClipEnd(1);
- mut_pos = cur_mut.AsInt();
- if (mut_type == 'D') mut_pos += ins_count;
- } else mut_pos = -1;
-
- // Tack on the current symbol...
- cString symbol_string;
- symbol_string.Set("<b><font color=\"%s\">%c</font></b>",
- color(), symbol);
- html_code += symbol_string;
- }
- }
-
- html_code += "</tt>";
-
- return html_code;
-}
Deleted: branches/brysonda/source/main/cAnalyzeGenotype.h
===================================================================
--- branches/brysonda/source/main/cAnalyzeGenotype.h 2005-11-04 00:39:30 UTC (rev 368)
+++ branches/brysonda/source/main/cAnalyzeGenotype.h 2005-11-04 03:50:46 UTC (rev 369)
@@ -1,268 +0,0 @@
-//////////////////////////////////////////////////////////////////////////////
-// Copyright (C) 1993 - 2003 California Institute of Technology //
-// //
-// Read the COPYING and README files, or contact 'avida at alife.org', //
-// before continuing. SOME RESTRICTIONS MAY APPLY TO USE OF THIS FILE. //
-//////////////////////////////////////////////////////////////////////////////
-
-#ifndef ANALYZE_GENOTYPE_HH
-#define ANALYZE_GENOTYPE_HH
-
-#include <fstream>
-
-#ifndef FUNCTIONS_HH
-#include "functions.h"
-#endif
-#ifndef GENOME_HH
-#include "cGenome.h"
-#endif
-#ifndef STRING_HH
-#include "cString.h"
-#endif
-#ifndef STRING_LIST_HH
-#include "cStringList.h"
-#endif
-#ifndef STRING_UTIL_HH
-#include "cStringUtil.h"
-#endif
-#ifndef TARRAY_HH
-#include "tArray.h"
-#endif
-
-// cAnalyzeGenotype : Collection of information about loaded genotypes
-
-class cGenome; // aggregate
-class cInstSet;
-class cString; // aggregate
-template <class T> class tArray; // aggregate
-class cStringList; // aggregate
-class cStringUtil; // access
-class cWorld;
-
-class cAnalyzeGenotype {
-private:
- cWorld* m_world;
- cGenome genome; // Full Genome
- cInstSet& inst_set; // Instruction set used in this genome
- cString name; // Name, if one was provided in loading
- cString aligned_sequence; // Sequence (in ASCII) after alignment
- cString tag; // All genotypes in a batch can be tagged
-
- bool viable;
-
- // Group 1 : Load-in Stats (Only obtained if available for input file)
- int id_num;
- int parent_id;
- int parent2_id;
- int num_cpus;
- int total_cpus;
- int update_born;
- int update_dead;
- int depth;
-
- // Group 2 : Basic Execution Stats (Obtained from test CPUs)
- int length;
- int copy_length;
- int exe_length;
- double merit;
- int gest_time;
- double fitness;
- int errors;
- double div_type;
- int mate_id;
-
- tArray<int> task_counts;
-
- // Group 3 : Stats requiring parental genotype (Also from test CPUs)
- double fitness_ratio;
- double efficiency_ratio;
- double comp_merit_ratio;
- int parent_dist;
- int ancestor_dist;
- int lineage_label;
- cString parent_muts;
-
- // Group 4 : Landscape stats (obtained from testing all possible mutations)
- class cAnalyzeLandscape {
- public:
- double frac_dead;
- double frac_neg;
- double frac_neut;
- double frac_pos;
- double complexity;
- cAnalyzeLandscape() : frac_dead(0.0), frac_neg(0.0),
- frac_neut(0.0), frac_pos(0.0), complexity(0.0) { ; }
- };
- mutable cAnalyzeLandscape * landscape_stats;
-
- // Group 5 : More complex stats (obtained indvidually, through tests)
- cString task_order;
-
- cStringList special_args; // These are args placed after a ':' in details...
-
- int NumCompare(double new_val, double old_val) const {
- if (new_val == old_val) return 0;
- else if (new_val == 0) return -2;
- else if (old_val == 0) return +2;
- else if (new_val < old_val) return -1;
- // else if (new_val > old_val)
- return +1;
- }
-
- int CalcMaxGestation() const;
- void CalcLandscape() const;
-public:
- cAnalyzeGenotype(cWorld* world, cString symbol_string, cInstSet & in_inst_set);
- cAnalyzeGenotype(cWorld* world, const cGenome & _genome, cInstSet & in_inst_set);
- cAnalyzeGenotype(const cAnalyzeGenotype & _gen);
- ~cAnalyzeGenotype();
-
- const cStringList & GetSpecialArgs() { return special_args; }
- void SetSpecialArgs(const cStringList & _args) { special_args = _args; }
-
- void Recalculate(cAnalyzeGenotype * parent_genotype=NULL);
- void PrintTasks(std::ofstream & fp, int min_task=0, int max_task=-1);
-
- // Set...
- void SetSequence(cString _sequence);
-
- void SetName(const cString & _name) { name = _name; }
- void SetAlignedSequence(const cString & _seq) { aligned_sequence = _seq; }
- void SetTag(const cString & _tag) { tag = _tag; }
-
- void SetViable(bool _viable) { viable = _viable; }
-
- void SetID(int _id) { id_num = _id; }
- void SetParentID(int _id) { parent_id = _id; }
- void SetParent2ID(int _id) { parent2_id = _id; }
- void SetNumCPUs(int _cpus) { num_cpus = _cpus; }
- void SetTotalCPUs(int _cpus) { total_cpus = _cpus; }
- void SetUpdateBorn(int _born) { update_born = _born; }
- void SetUpdateDead(int _dead) { update_dead = _dead; }
- void SetDepth(int _depth) { depth = _depth; }
-
- void SetLength(int _length) { length = _length; }
- void SetCopyLength(int _length) { copy_length = _length; }
- void SetExeLength(int _length) { exe_length = _length; }
- void SetMerit(double _merit) { merit = _merit; }
- void SetGestTime(int _gest) { gest_time = _gest; }
- void SetFitness(double _fitness) { fitness = _fitness; }
- void SetDivType(double _div_type) { div_type = _div_type; }
- void SetMateID(int _mate_id) { mate_id = _mate_id; }
- void SetParentDist(int _dist) { parent_dist = _dist; }
- void SetAncestorDist(int _dist) { ancestor_dist = _dist; }
- void SetLineageLabel(int _label) { lineage_label = _label; }
-
- void SetParentMuts(const cString & in_muts) { parent_muts = in_muts; }
- void SetTaskOrder(const cString & in_order) { task_order = in_order; }
-
-// void SetFracDead(double in_frac);
-// void SetFracNeg(double in_frac);
-// void SetFracNeut(double in_frac);
-// void SetFracPos(double in_frac);
-
- // Accessors...
- const cGenome & GetGenome() const { return genome; }
- const cString & GetName() const { return name; }
- const cString & GetAlignedSequence() const { return aligned_sequence; }
- const cString & GetTag() const { return tag; }
-
- bool GetViable() const { return viable; }
-
- int GetID() const { return id_num; }
- int GetParentID() const { return parent_id; }
- int GetParent2ID() const { return parent2_id; }
- int GetParentDist() const { return parent_dist; }
- int GetAncestorDist() const { return ancestor_dist; }
- int GetLineageLabel() const { return lineage_label; }
- int GetNumCPUs() const { return num_cpus; }
- int GetTotalCPUs() const { return total_cpus; }
- int GetLength() const { return length; }
- int GetCopyLength() const { return copy_length; }
- int GetExeLength() const { return exe_length; }
- int GetMinLength() const { return Min(exe_length, copy_length); }
- double GetMerit() const { return merit; }
- double GetCompMerit() const { return merit / (double) GetMinLength(); }
- int GetGestTime() const { return gest_time; }
- double GetEfficiency() const
- { return ((double) GetMinLength()) / (double) gest_time; }
- double GetFitness() const { return fitness; }
- double GetDivType() const { return div_type; }
- int GetMateID() const { return mate_id; }
- int GetUpdateBorn() const { return update_born; }
- int GetUpdateDead() const { return update_dead; }
- int GetDepth() const { return depth; }
-
- const cString & GetParentMuts() const { return parent_muts; }
-
- // Landscape accessors
- double GetFracDead() const;
- double GetFracNeg() const;
- double GetFracNeut() const;
- double GetFracPos() const;
- double GetComplexity() const;
-
- double GetFitnessRatio() const { return fitness_ratio; }
- double GetEfficiencyRatio() const { return efficiency_ratio; }
- double GetCompMeritRatio() const { return comp_merit_ratio; }
-
- const cString & GetTaskOrder() const { return task_order; }
- cString GetTaskList() const;
-
- cString GetSequence() const { return genome.AsString(); }
- cString GetHTMLSequence() const;
-
- cString GetMapLink() const {
- return cStringUtil::Stringf("<a href=\"tasksites.%s.html\">Phenotype Map</a>", GetName()());
- }
-
- int GetNumTasks() const { return task_counts.GetSize(); }
- int GetTaskCount(int task_id) const {
- if (task_id >= task_counts.GetSize()) return 0;
- if (special_args.HasString("binary")) return (task_counts[task_id] > 0);
- return task_counts[task_id];
- }
-
- // Comparisons... Compares a genotype to the "previous" one, which is
- // passed in, in one specified phenotype.
- // Return values are:
- // -2 : Toggle; no longer has phenotype it used to...
- // -1 : Reduction in strength of phenotype
- // 0 : Identical in phenotype
- // +1 : Improvement in phenotype
- // +2 : Toggle; phenotype now present that wasn't.
- int CompareNULL(cAnalyzeGenotype * prev) const { (void) prev; return 0; }
- int CompareArgNULL(cAnalyzeGenotype * prev, int i) const
- { (void) prev; (void) i; return 0; }
- int CompareLength(cAnalyzeGenotype * prev) const {
- if (GetLength() < MIN_CREATURE_SIZE &&
- prev->GetLength() > MIN_CREATURE_SIZE) return -2;
- if (GetLength() > MIN_CREATURE_SIZE &&
- prev->GetLength() < MIN_CREATURE_SIZE) return 2;
- return 0;
- }
- int CompareMerit(cAnalyzeGenotype * prev) const
- { return NumCompare(GetMerit(), prev->GetMerit()); }
- int CompareCompMerit(cAnalyzeGenotype * prev) const
- { return NumCompare(GetCompMerit(), prev->GetCompMerit()); }
- int CompareGestTime(cAnalyzeGenotype * prev) const {
- const int max_time = CalcMaxGestation();
- const int cur_time = max_time - GetGestTime();
- const int prev_time = max_time - prev->GetGestTime();
- return NumCompare(cur_time, prev_time);
- }
- int CompareEfficiency(cAnalyzeGenotype * prev) const
- { return NumCompare(GetEfficiency(), prev->GetEfficiency()); }
- int CompareFitness(cAnalyzeGenotype * prev) const
- { return NumCompare(GetFitness(), prev->GetFitness()); }
- int CompareTaskCount(cAnalyzeGenotype * prev, int task_id) const
- { return NumCompare(GetTaskCount(task_id), prev->GetTaskCount(task_id)); }
-
- /*
- added to satisfy Boost.Python; the semantics are fairly useless --
- equality of two references means that they refer to the same object.
- */
- bool operator==(const cAnalyzeGenotype &in) const { return &in == this; }
-};
-
-#endif
Deleted: branches/brysonda/source/main/cAnalyzeUtil.cc
===================================================================
--- branches/brysonda/source/main/cAnalyzeUtil.cc 2005-11-04 00:39:30 UTC (rev 368)
+++ branches/brysonda/source/main/cAnalyzeUtil.cc 2005-11-04 03:50:46 UTC (rev 369)
@@ -1,961 +0,0 @@
-//////////////////////////////////////////////////////////////////////////////
-// Copyright (C) 1993 - 2001 California Institute of Technology //
-// //
-// Read the COPYING and README files, or contact 'avida at alife.org', //
-// before continuing. SOME RESTRICTIONS MAY APPLY TO USE OF THIS FILE. //
-//////////////////////////////////////////////////////////////////////////////
-
-#include "cAnalyzeUtil.h"
-
-#include "defs.h"
-#include "cEnvironment.h"
-#include "cGenebank.h"
-#include "cGenome.h"
-#include "cGenomeUtil.h"
-#include "cGenotype.h"
-#include "cHardwareBase.h"
-#include "cHardwareManager.h"
-#include "cHistogram.h"
-#include "cInstSet.h"
-#include "cInstUtil.h"
-#include "cLandscape.h"
-#include "cOrganism.h"
-#include "cPhenotype.h"
-#include "cPopulation.h"
-#include "cPopulationCell.h"
-#include "cSpecies.h"
-#include "cStats.h"
-#include "cTestCPU.h"
-#include "cTestUtil.h"
-#include "cTools.h"
-#include "cWorld.h"
-
-#include <vector>
-
-using namespace std;
-
-void cAnalyzeUtil::TestGenome(cWorld* world, const cGenome & genome, cInstSet & inst_set,
- ofstream & fp, int update)
-{
- cCPUTestInfo test_info;
- world->GetTestCPU().TestGenome(test_info, genome);
-
- cPhenotype &colony_phenotype = test_info.GetColonyOrganism()->GetPhenotype();
- fp << update << " " // 1
- << colony_phenotype.GetMerit().GetDouble() << " " // 2
- << colony_phenotype.GetGestationTime() << " " // 3
- << colony_phenotype.GetFitness() << " " // 4
- << 1.0 / (0.1 + colony_phenotype.GetGestationTime()) << " " // 5
- << genome.GetSize() << " " // 6
- << colony_phenotype.GetCopiedSize() << " " // 7
- << colony_phenotype.GetExecutedSize() << endl; // 8
-}
-
-
-
-void cAnalyzeUtil::TestInsSizeChangeRobustness(cWorld* world, ofstream & fp,
- const cInstSet & inst_set, const cGenome & in_genome,
- int num_trials, int update)
-{
- cCPUTestInfo test_info;
- const cInstruction inst_none = inst_set.GetInst("instruction_none");
-
- // Stats
- int num_viable = 0;
- int num_new_size = 0;
- int num_parent_size = 0;
-
- for (int i = 0; i < num_trials; i++) {
- cCPUMemory genome(in_genome);
- // Should check to only insert infront of an instruction (not a Nop)
- int ins_pos = -1;
- while (ins_pos < 0) {
- ins_pos = world->GetRandom().GetUInt(genome.GetSize());
- if( inst_set.IsNop(genome[ins_pos]) ) ins_pos = -1;
- }
-
- // Insert some "instruction_none" into the genome
- const int num_nops = world->GetRandom().GetUInt(5) + 5;
- for (int j = 0; j < num_nops; j++) genome.Insert(ins_pos, inst_none);
-
- // Test the genome and output stats
- if (world->GetTestCPU().TestGenome(test_info, genome)){ // Daughter viable...
- num_viable++;
- const double child_size =
- test_info.GetColonyOrganism()->GetGenome().GetSize();
-
- if (child_size == (double) in_genome.GetSize()) num_parent_size++;
- else if (child_size == (double) genome.GetSize()) num_new_size++;
- }
- } // for num_trials
-
- fp << update << " "
- << (double) num_viable / num_trials << " "
- << (double) num_new_size / num_trials << " "
- << (double) num_parent_size / num_trials << " "
- << endl;
-}
-
-
-
-// Returns the genome of maximal fitness.
-cGenome cAnalyzeUtil::CalcLandscape(cWorld* world, int dist, const cGenome & genome,
- cInstSet & inst_set)
-{
- cLandscape landscape(world, genome, inst_set);
- landscape.Process(dist);
- double peak_fitness = landscape.GetPeakFitness();
- cGenome peak_genome = landscape.GetPeakGenome();
-
- // Print the results.
-
- static ofstream fp("landscape.dat");
- static ofstream fp_entropy("land-entropy.dat");
- static ofstream fp_count("land-sitecount.dat");
-
- landscape.PrintStats(fp);
- landscape.PrintEntropy(fp_entropy);
- landscape.PrintSiteCount(fp_count);
-
- // Repeat for Insertions...
- landscape.Reset(genome);
- landscape.ProcessInsert();
- static ofstream fp_ins("landscape-ins.dat");
- static ofstream fp_ins_count("land-ins-sitecount.dat");
- landscape.PrintStats(fp_ins);
- landscape.PrintSiteCount(fp_ins_count);
- if (landscape.GetPeakFitness() > peak_fitness) {
- peak_fitness = landscape.GetPeakFitness();
- peak_genome = landscape.GetPeakGenome();
- }
-
- // And Deletions...
- landscape.Reset(genome);
- landscape.ProcessDelete();
- static ofstream fp_del("landscape-del.dat");
- static ofstream fp_del_count("land-del-sitecount.dat");
- landscape.PrintStats(fp_del);
- landscape.PrintSiteCount(fp_del_count);
- if (landscape.GetPeakFitness() > peak_fitness) {
- peak_fitness = landscape.GetPeakFitness();
- peak_genome = landscape.GetPeakGenome();
- }
-
- return peak_genome;
-}
-
-
-void cAnalyzeUtil::AnalyzeLandscape(cWorld* world, const cGenome & genome, cInstSet &inst_set,
- int sample_size, int min_found, int max_sample_size, int update)
-{
- cLandscape landscape(world, genome, inst_set);
-
- static ofstream fp("land_analyze.dat");
-
- int num_found = 0;
- for (int dist = 1; dist <= 10; dist++) {
- landscape.Reset(genome);
- if (dist == 1) {
- landscape.Process(dist);
- num_found = genome.GetSize() * (inst_set.GetSize() - 1);
- } else {
- num_found =
- landscape.RandomProcess(sample_size, dist, min_found, max_sample_size);
- }
-
- fp << update << " " // 1
- << dist << " " // 2
- << landscape.GetProbDead() << " " // 3
- << landscape.GetProbNeg() << " " // 4
- << landscape.GetProbNeut() << " " // 5
- << landscape.GetProbPos() << " " // 6
- << landscape.GetNumTrials() << " " // 7
- << num_found << " " // 8
- << landscape.GetAveFitness() << " " // 9
- << landscape.GetAveSqrFitness() << " " // 10
- << endl;
-
- if ((dist > 1) && (num_found < min_found)) break;
- }
-}
-
-
-void cAnalyzeUtil::PairTestLandscape(cWorld* world, const cGenome &genome, cInstSet &inst_set,
- int sample_size, int update)
-{
- cLandscape landscape(world, genome, inst_set);
-
- cString filename;
- filename.Set("pairtest.%d.dat", update);
- ofstream fp(filename());
-
- if (sample_size != 0) landscape.TestPairs(sample_size, fp);
- else landscape.TestAllPairs(fp);
-}
-
-
-void cAnalyzeUtil::CalcConsensus(cWorld* world, int lines_saved)
-{
- cPopulation* population = &world->GetPopulation();
- const int num_inst = world->GetHardwareManager().GetInstSet().GetSize();
- const int update = population->GetStats().GetUpdate();
- cGenebank & genebank = population->GetGenebank();
-
- // Setup the histogtams...
- cHistogram * inst_hist = new cHistogram[MAX_CREATURE_SIZE];
- for (int i = 0; i < MAX_CREATURE_SIZE; i++) inst_hist[i].Resize(num_inst,-1);
-
- // Loop through all of the genotypes adding them to the histograms.
- cGenotype * cur_genotype = genebank.GetBestGenotype();
- for (int i = 0; i < genebank.GetSize(); i++) {
- const int num_organisms = cur_genotype->GetNumOrganisms();
- const int length = cur_genotype->GetLength();
- const cGenome & genome = cur_genotype->GetGenome();
-
- // Place this genotype into the histograms.
- for (int j = 0; j < length; j++) {
- assert(genome[j].GetOp() < num_inst);
- inst_hist[j].Insert(genome[j].GetOp(), num_organisms);
- }
-
- // Mark all instructions beyond the length as -1 in histogram...
- for (int j = length; j < MAX_CREATURE_SIZE; j++) {
- inst_hist[j].Insert(-1, num_organisms);
- }
-
- // ...and advance to the next genotype...
- cur_genotype = cur_genotype->GetNext();
- }
-
- // Now, lets print something!
- static ofstream fp("consensus.dat");
- static ofstream fp_abundance("con-abundance.dat");
- static ofstream fp_var("con-var.dat");
- static ofstream fp_entropy("con-entropy.dat");
-
- // Determine the length of the concensus genome
- int con_length;
- for (con_length = 0; con_length < MAX_CREATURE_SIZE; con_length++) {
- if (inst_hist[con_length].GetMode() == -1) break;
- }
-
- // Build the concensus genotype...
- cGenome con_genome(con_length);
- double total_entropy = 0.0;
- for (int i = 0; i < MAX_CREATURE_SIZE; i++) {
- const int mode = inst_hist[i].GetMode();
- const int count = inst_hist[i].GetCount(mode);
- const int total = inst_hist[i].GetCount();
- const double entropy = inst_hist[i].GetNormEntropy();
- if (i < con_length) total_entropy += entropy;
-
- // Break out if ALL creatures have a -1 in this area, and we've
- // finished printing all of the files.
- if (mode == -1 && count == total) break;
-
- if ( i < con_length )
- con_genome[i].SetOp(mode);
-
- // Print all needed files.
- if (i < lines_saved) {
- fp_abundance << count << " ";
- fp_var << inst_hist[i].GetCountVariance() << " ";
- fp_entropy << entropy << " ";
- }
- }
-
- // Put end-of-lines on the files.
- if (lines_saved > 0) {
- fp_abundance << endl;
- fp_var << endl;
- fp_entropy << endl;
- }
-
- // --- Study the consensus genome ---
-
- // Loop through genotypes again, and determine the average genetic distance.
- cur_genotype = genebank.GetBestGenotype();
- cDoubleSum distance_sum;
- for (int i = 0; i < genebank.GetSize(); i++) {
- const int num_organisms = cur_genotype->GetNumOrganisms();
- const int cur_dist =
- cGenomeUtil::FindEditDistance(con_genome, cur_genotype->GetGenome());
- distance_sum.Add(cur_dist, num_organisms);
-
- // ...and advance to the next genotype...
- cur_genotype = cur_genotype->GetNext();
- }
-
- // Finally, gather last bits of data and print the results.
- cGenotype * con_genotype = genebank.FindGenotype(con_genome, -1);
- const int best_dist = cGenomeUtil::FindEditDistance(con_genome,
- genebank.GetBestGenotype()->GetGenome());
-
- const double ave_dist = distance_sum.Average();
- const double var_dist = distance_sum.Variance();
- const double complexity_base = (double) con_genome.GetSize() - total_entropy;
-
- cString con_name;
- con_name.Set("genebank/%03d-consensus-u%i.gen", con_genome.GetSize(),update);
- cTestUtil::PrintGenome(world, con_genome, con_name());
-
-
- if (con_genotype) {
- fp << update << " " // 1
- << con_genotype->GetMerit() << " " // 2
- << con_genotype->GetGestationTime() << " " // 3
- << con_genotype->GetFitness() << " " // 4
- << con_genotype->GetReproRate() << " " // 5
- << con_genotype->GetLength() << " " // 6
- << con_genotype->GetCopiedSize() << " " // 7
- << con_genotype->GetExecutedSize() << " " // 8
- << con_genotype->GetBirths() << " " // 9
- << con_genotype->GetBreedTrue() << " " // 10
- << con_genotype->GetBreedIn() << " " // 11
- << con_genotype->GetNumOrganisms() << " " // 12
- << con_genotype->GetDepth() << " " // 13
- << con_genotype->GetID() << " " // 14
- << update - con_genotype->GetUpdateBorn() << " " // 15
- << best_dist << " " // 16
- << ave_dist << " " // 17
- << var_dist << " " // 18
- << total_entropy << " " // 19
- << complexity_base << " " // 20
- << endl;
- }
- else {
- cCPUTestInfo test_info;
- world->GetTestCPU().TestGenome(test_info, con_genome);
- cPhenotype & colony_phenotype =
- test_info.GetColonyOrganism()->GetPhenotype();
- fp << update << " " // 1
- << colony_phenotype.GetMerit() << " " // 2
- << colony_phenotype.GetGestationTime() << " " // 3
- << colony_phenotype.GetFitness() << " " // 4
- << 1.0 / (0.1 + colony_phenotype.GetGestationTime()) << " " // 5
- << con_genome.GetSize() << " " // 6
- << colony_phenotype.GetCopiedSize() << " " // 7
- << colony_phenotype.GetExecutedSize() << " " // 8
- << 0 << " " // Births // 9
- << 0 << " " // Breed True // 10
- << 0 << " " // Breed In // 11
- << 0 << " " // Num CPUs // 12
- << -1 << " " // Depth // 13
- << -1 << " " // ID // 14
- << 0 << " " // Age // 15
- << best_dist << " " // 16
- << ave_dist << " " // 17
- << var_dist << " " // 18
- << total_entropy << " " // 19
- << complexity_base << " " // 20
- << endl;
- }
-
- // Flush the file...
- fp.flush();
-
- delete [] inst_hist;
-}
-
-
-
-/**
- * This function goes through all creatures in the soup, and saves the
- * basic landscape data (neutrality, fitness, and so on) into a stream.
- *
- * @param fp The stream into which the data should be saved.
- *
- * @param sample_prob The probability with which a particular creature should
- * be analyzed (a value of 1 analyzes all creatures, a value of 0.1 analyzes
- * 10%, and so on).
- *
- * @param landscape A bool that indicates whether the creatures should be
- * landscaped (calc. neutrality and so on) or not.
- *
- * @param save_genotype A bool that indicates whether the creatures should
- * be saved or not.
- **/
-
-void cAnalyzeUtil::AnalyzePopulation(cWorld* world, ofstream & fp,
- double sample_prob, bool landscape, bool save_genotype)
-{
- cPopulation* pop = &world->GetPopulation();
- fp << "# (1) cell number (2) genotype name (3) length (4) fitness [test-cpu] (5) fitness (actual) (6) merit (7) no of breed trues occurred (8) lineage label (9) neutral metric (10) -... landscape data" << endl;
-
- const double skip_prob = 1.0 - sample_prob;
- for (int i = 0; i < pop->GetSize(); i++) {
- if (pop->GetCell(i).IsOccupied() == false) continue; // No organism...
- if (world->GetRandom().P(skip_prob)) continue; // Not sampled...
-
- cOrganism * organism = pop->GetCell(i).GetOrganism();
- cGenotype * genotype = organism->GetGenotype();
- const cGenome & genome = organism->GetGenome();
-
- cString creature_name;
- if ( genotype->GetThreshold() ) creature_name = genotype->GetName();
- else creature_name.Set("%03d-no_name-u%i-c%i", genotype->GetLength(),
- pop->GetStats().GetUpdate(), i );
-
- fp << i << " " // 1 cell ID
- << creature_name() << " " // 2 name
- << genotype->GetLength() << " " // 3 length
- << genotype->GetTestFitness() << " " // 4 fitness (test-cpu)
- << organism->GetPhenotype().GetFitness() << " " // 5 fitness (actual)
- << organism->GetPhenotype().GetMerit() << " " // 6 merit
- << genotype->GetBreedTrue() << " " // 7 breed true?
- << organism->GetLineageLabel() << " " // 8 lineage label
- << organism->GetPhenotype().GetNeutralMetric() << " "; // 9 neut metric
-
- // create landscape object for this creature
- if (landscape && genotype->GetTestFitness() > 0) {
- cLandscape landscape(world, genome, world->GetHardwareManager().GetInstSet());
- landscape.Process(1);
- landscape.PrintStats(fp);
- }
- else fp << endl;
- if ( save_genotype ){
- char filename[40];
- sprintf( filename, "genebank/%s", creature_name() );
- cTestUtil::PrintGenome(world, genome, filename);
- }
- }
-}
-
-
-/**
- * This function prints out fitness data. The main point is that it
- * calculates the average fitness from info from the testCPU + the actual
- * merit of the organisms, and assigns zero fitness to those organisms
- * that will never reproduce.
- *
- * The function also determines the maximum fitness genotype, and can
- * produce fitness histograms.
- *
- * @param datafp A stream into which the fitness data should be written.
- * @param histofp A stream into which the fitness histogram should be
- * written.
- * @param histo_testCPU_fp A stream into which the fitness histogram as
- * determined exclusively from the test-CPU should be written.
- * @param save_max_f_genotype A bool that determines whether the genotype
- * with the maximum fitness should be saved into the genebank.
- * @param print_fitness_histo A bool that determines whether fitness
- * histograms should be written.
- * @param hist_fmax The maximum fitness value to be taken into account
- * for the fitness histograms.
- * @param hist_fstep The width of the individual bins in the fitness
- * histograms.
- **/
-
-void cAnalyzeUtil::PrintDetailedFitnessData(cWorld* world, ofstream &datafp,
- ofstream & hist_fp, ofstream & histo_testCPU_fp, bool save_max_f_genotype,
- bool print_fitness_histo, double hist_fmax, double hist_fstep)
-{
- cPopulation* pop = &world->GetPopulation();
- const int update = pop->GetStats().GetUpdate();
- const double generation = pop->GetStats().SumGeneration().Average();
-
- // the histogram variables
- vector<int> histo;
- vector<int> histo_testCPU;
- int bins;
-
- if ( print_fitness_histo ){
- bins = (int) (hist_fmax / hist_fstep) + 1;
- histo.resize( bins, 0 ); // resize and clear
- histo_testCPU.resize( bins, 0 );
- }
-
- int n = 0;
- int nhist_tot = 0;
- int nhist_tot_testCPU = 0;
- double fave = 0;
- double fave_testCPU = 0;
- double max_fitness = -1; // we set this to -1, so that even 0 is larger...
- cGenotype * max_f_genotype = NULL;
-
- for (int i = 0; i < pop->GetSize(); i++) {
- if (pop->GetCell(i).IsOccupied() == false) continue; // One use organisms.
-
- cOrganism * organism = pop->GetCell(i).GetOrganism();
- cGenotype * genotype = organism->GetGenotype();
-
- cCPUTestInfo test_info;
- world->GetTestCPU().TestGenome( test_info, genotype->GetGenome() );
- // We calculate the fitness based on the current merit,
- // but with the true gestation time. Also, we set the fitness
- // to zero if the creature is not viable.
- const double f = ( test_info.IsViable() ) ? organism->GetPhenotype().GetMerit().CalcFitness(test_info.GetTestOrganism()->GetPhenotype().GetGestationTime()) : 0;
- const double f_testCPU = test_info.GetColonyFitness();
-
- // Get the maximum fitness in the population
- // Here, we want to count only organisms that can truly replicate,
- // to avoid complications
- if ( f_testCPU > max_fitness &&
- test_info.GetTestOrganism()->GetPhenotype().CopyTrue() ){
- max_fitness = f_testCPU;
- max_f_genotype = genotype;
- }
-
- fave += f;
- fave_testCPU += f_testCPU;
- n += 1;
-
-
- // histogram
- if ( print_fitness_histo && f < hist_fmax ){
- histo[(int) (f / hist_fstep)] += 1;
- nhist_tot += 1;
- }
-
- if ( print_fitness_histo && f_testCPU < hist_fmax ){
- histo_testCPU[(int) (f_testCPU / hist_fstep)] += 1;
- nhist_tot_testCPU += 1;
- }
- }
-
- // determine the name of the maximum fitness genotype
- cString max_f_name;
- if ( max_f_genotype->GetThreshold() )
- max_f_name = max_f_genotype->GetName();
- else // we put the current update into the name, so that it becomes unique.
- max_f_name.Set("%03d-no_name-u%i", max_f_genotype->GetLength(),
- update );
-
- datafp << update << " " // 1 update
- << generation << " " // 2 generation
- << fave/ (double) n << " " // 3 average fitness
- << fave_testCPU/ (double) n << " " // 4 average test fitness
- << n << " " // 5 organism total
- << max_fitness << " " // 6 maximum fitness
- << max_f_name() << " " // 7 maxfit genotype name
- << endl;
-
- if (save_max_f_genotype) {
- char filename[40];
- sprintf( filename, "genebank/%s", max_f_name() );
- cTestUtil::PrintGenome(world, max_f_genotype->GetGenome(), filename);
- }
-
- if (print_fitness_histo) {
- hist_fp << update << " " // 1 update
- << generation << " " // 2 generation
- << fave/ (double) n << " "; // 3 average fitness
-
- // now output the fitness histo
- vector<int>::const_iterator it = histo.begin();
- for ( ; it != histo.end(); it++ )
- hist_fp << (double) (*it) / (double) nhist_tot << " ";
- hist_fp << endl;
-
- histo_testCPU_fp << update << " " // 1 update
- << generation << " " // 2 generation
- << fave_testCPU/ (double) n << " "; // 3 average fitness
-
- // now output the fitness histo
- it = histo_testCPU.begin();
- for ( ; it != histo_testCPU.end(); it++ )
- histo_testCPU_fp << (double) (*it) / (double) nhist_tot_testCPU << " ";
- histo_testCPU_fp << endl;
- }
-}
-
-
-/**
- * This function goes through all genotypes currently present in the soup,
- * and writes into an output file the average Hamming distance between the
- * creatures in the population and a given reference genome.
- *
- * @param fp The stream into which the data should be saved.
- * @param reference_genome The reference genome.
- * @param save_creatures A bool that indicates whether creatures should be
- * saved into the genebank or not.
- **/
-
-void cAnalyzeUtil::PrintGeneticDistanceData(cWorld* world, ofstream & fp,
- const char * creature_name)
-{
- cPopulation* pop = &world->GetPopulation();
- double hamming_m1 = 0;
- double hamming_m2 = 0;
- int count = 0;
- int dom_dist = 0;
-
- // load the reference genome
- cGenome reference_genome(cInstUtil::LoadGenome(creature_name, world->GetHardwareManager().GetInstSet()));
-
- // get the info for the dominant genotype
- cGenotype * cur_genotype = pop->GetGenebank().GetBestGenotype();
- cGenome genome = cur_genotype->GetGenome();
- dom_dist = cGenomeUtil::FindHammingDistance( reference_genome, genome );
- hamming_m1 += dom_dist;
- hamming_m2 += dom_dist*dom_dist;
- count += cur_genotype->GetNumOrganisms();
- // now cycle over the remaining genotypes
- for (int i = 1; i < pop->GetGenebank().GetSize(); i++) {
- cur_genotype = cur_genotype->GetNext();
- cGenome genome = cur_genotype->GetGenome();
-
- int dist = cGenomeUtil::FindHammingDistance( reference_genome, genome );
- hamming_m1 += dist;
- hamming_m2 += dist*dist;
- count += cur_genotype->GetNumOrganisms();
- }
-
- hamming_m1 /= (double) count;
- hamming_m2 /= (double) count;
-
- fp << pop->GetStats().GetUpdate() << " " // 1 update
- << hamming_m1 << " " // ave. Hamming dist
- << sqrt( ( hamming_m2 - hamming_m1*hamming_m1 ) / (double) count )
- << " " // std. error
- << cGenomeUtil::FindHammingDistance( reference_genome,
- pop->GetGenebank().GetBestGenotype()->GetGenome() ) << " "
- << endl;
-}
-
-
-/**
- * This function goes through all genotypes currently present in the soup,
- * and writes into an output file the names of the genotypes, the fitness
- * as determined in the test cpu, and the genetic distance to a reference
- * genome.
- *
- * @param fp The stream into which the data should be saved.
- * @param reference_genome The reference genome.
- * @param save_creatures A bool that indicates whether creatures should be
- * saved into the genebank or not.
- **/
-
-void cAnalyzeUtil::GeneticDistancePopDump(cWorld* world, ofstream & fp,
- const char * creature_name, bool save_creatures)
-{
- cPopulation* pop = &world->GetPopulation();
- double sum_fitness = 0;
- int sum_num_organisms = 0;
-
- // load the reference genome
- cGenome reference_genome( cInstUtil::LoadGenome(creature_name, world->GetHardwareManager().GetInstSet()) );
-
- // first, print out some documentation...
- fp << "# (1) genotype name (2) fitness [test-cpu] (3) abundance (4) Hamming distance to reference (5) Levenstein distance to reference" << endl;
- fp << "# reference genome is the START_CREATURE" << endl;
-
- // cycle over all genotypes
- cGenotype * cur_genotype = pop->GetGenebank().GetBestGenotype();
- for (int i = 0; i < pop->GetGenebank().GetSize(); i++) {
- const cGenome & genome = cur_genotype->GetGenome();
- const int num_orgs = cur_genotype->GetNumOrganisms();
-
- // now output
-
- sum_fitness += cur_genotype->GetTestFitness() * num_orgs;
- sum_num_organisms += num_orgs;
-
- fp << cur_genotype->GetName()() << " " // 1 name
- << cur_genotype->GetTestFitness() << " " // 2 fitness
- << num_orgs << " " // 3 abundance
- << cGenomeUtil::FindHammingDistance(reference_genome, genome) << " "
- << cGenomeUtil::FindEditDistance(reference_genome, genome) << " " // 5
- << genome.AsString()() << " " // 6 genome
- << endl;
-
- // save into genebank
- if (save_creatures) {
- char filename[40];
- sprintf( filename, "genebank/%s", cur_genotype->GetName()() );
- cTestUtil::PrintGenome(world, genome, filename);
- }
-
- // ...and advance to the next genotype...
- cur_genotype = cur_genotype->GetNext();
- }
- fp << "# ave fitness from Test CPU's: "
- << sum_fitness/sum_num_organisms << endl;
-}
-
-
-/**
- * This function goes through all creatures in the soup, and writes out
- * how many tasks the different creatures have done up to now. It counts
- * every task only once, i.e., if a creature does 'put' three times, that
- * will increase its count only by one.
- *
- * @param fp The file into which the result should be written.
- **/
-
-void cAnalyzeUtil::TaskSnapshot(cWorld* world, ofstream & fp)
-{
- cPopulation* pop = &world->GetPopulation();
- fp << "# (1) cell number\n# (2) number of rewarded tasks done so far\n# (3) total number of tasks done so far\n# (4) same as 2, but right before divide\n# (5) same as 3, but right before divide\n# (6) same as 2, but for parent\n# (7) same as 3, but for parent\n# (8) genotype fitness\n# (9) genotype name" << endl;
-
- for (int i = 0; i < pop->GetSize(); i++) {
- if (pop->GetCell(i).IsOccupied() == false) continue;
- cOrganism * organism = pop->GetCell(i).GetOrganism();
-
- // create a test-cpu for the current creature
- cCPUTestInfo test_info;
- world->GetTestCPU().TestGenome( test_info, organism->GetGenome() );
- cPhenotype & test_phenotype = test_info.GetTestOrganism()->GetPhenotype();
- cPhenotype & phenotype = organism->GetPhenotype();
-
- int num_tasks = world->GetEnvironment().GetTaskLib().GetSize();
- int sum_tasks_all = 0;
- int sum_tasks_rewarded = 0;
- int divide_sum_tasks_all = 0;
- int divide_sum_tasks_rewarded = 0;
- int parent_sum_tasks_all = 0;
- int parent_sum_tasks_rewarded = 0;
-
- for (int j = 0; j < num_tasks; j++) {
- // get the number of bonuses for this task
- int bonuses = 1; //phenotype.GetTaskLib().GetTaskNumBonus(j);
- int task_count = ( phenotype.GetCurTaskCount()[j] == 0 ) ? 0 : 1;
- int divide_tasks_count = (test_phenotype.GetLastTaskCount()[j] == 0)?0:1;
- int parent_task_count = (phenotype.GetLastTaskCount()[j] == 0) ? 0 : 1;
-
- // If only one bonus, this task is not rewarded, as last bonus is + 0.
- if (bonuses > 1) {
- sum_tasks_rewarded += task_count;
- divide_sum_tasks_rewarded += divide_tasks_count;
- parent_sum_tasks_rewarded += parent_task_count;
- }
- sum_tasks_all += task_count;
- divide_sum_tasks_all += divide_tasks_count;
- parent_sum_tasks_all += parent_task_count;
- }
-
- fp << i << " " // 1 cell number
- << sum_tasks_rewarded << " " // 2 number of tasks rewarded
- << sum_tasks_all << " " // 3 total number of tasks done
- << divide_sum_tasks_rewarded << " " // 4 num rewarded tasks on divide
- << divide_sum_tasks_all << " " // 5 num total tasks on divide
- << parent_sum_tasks_rewarded << " " // 6 parent number of tasks rewared
- << parent_sum_tasks_all << " " // 7 parent total num tasks done
- << test_info.GetColonyFitness() << " " // 8 genotype fitness
- << organism->GetGenotype()->GetName()() << " " // 9 genotype name
- << endl;
- }
-}
-
-void cAnalyzeUtil::TaskGrid(cWorld* world, ofstream & fp)
-{
- cPopulation* pop = &world->GetPopulation();
- for (int i = 0; i < pop->GetWorldX(); i++) {
- for (int j = 0; j < pop->GetWorldY(); j++) {
- int task_sum = 0;
- int cell_num = i*pop->GetWorldX()+j;
- if (pop->GetCell(cell_num).IsOccupied() == true) {
- cOrganism * organism = pop->GetCell(cell_num).GetOrganism();
- cCPUTestInfo test_info;
- world->GetTestCPU().TestGenome( test_info, organism->GetGenome() );
- cPhenotype & test_phenotype = test_info.GetTestOrganism()->GetPhenotype();
- int num_tasks = world->GetEnvironment().GetTaskLib().GetSize();
- for (int k = 0; k < num_tasks; k++) {
- if (test_phenotype.GetLastTaskCount()[k]>0) {
- task_sum = task_sum + (int) pow(2.0,k);
- }
- }
- }
- fp << task_sum << " ";
- }
- fp << endl;
- }
-}
-
-/**
- * This function prints all the tasks that viable creatures have performed
- * so far (compare with the event 'print_task_data', which prints all tasks.
- **/
-
-void cAnalyzeUtil::PrintViableTasksData(cWorld* world, ofstream & fp)
-{
- const int num_tasks = world->GetNumTasks();
- cPopulation* pop = &world->GetPopulation();
-
- static vector<int> tasks(num_tasks);
- vector<int>::iterator it;
-
- // clear task vector
- for (it = tasks.begin(); it != tasks.end(); it++) (*it) = 0;
-
- for (int i = 0; i < pop->GetSize(); i++) {
- if (pop->GetCell(i).IsOccupied() == false) continue;
- if (pop->GetCell(i).GetOrganism()->GetGenotype()->GetTestFitness() > 0.0) {
- cPhenotype & phenotype = pop->GetCell(i).GetOrganism()->GetPhenotype();
- for (int j = 0; j < num_tasks; j++) {
- if (phenotype.GetCurTaskCount()[j] > 0) tasks[j] += 1;
- }
- }
- }
-
- fp << pop->GetStats().GetUpdate();
- for (it = tasks.begin(); it != tasks.end(); it++) fp << " " << (*it);
- fp<<endl;
-}
-
-
-void cAnalyzeUtil::PrintTreeDepths(cPopulation * pop, ofstream & fp)
-{
- // cycle over all genotypes
- cGenotype * genotype = pop->GetGenebank().GetBestGenotype();
- for (int i = 0; i < pop->GetGenebank().GetSize(); i++) {
- fp << genotype->GetID() << " " // 1
- << genotype->GetTestFitness() << " " // 2
- << genotype->GetNumOrganisms() << " " // 3
- << genotype->GetDepth() << " " // 4
- << endl;
-
- // ...and advance to the next genotype...
- genotype = genotype->GetNext();
- }
-}
-
-
-void cAnalyzeUtil::PrintDepthHistogram(ofstream & fp, cPopulation * pop)
-{
- // Output format: update min max histogram_values...
- int min = INT_MAX;
- int max = 0;
- assert(fp.good());
-
- // Two pass method
-
- // Loop through all genotypes getting min and max values
- cGenebank & genebank = pop->GetGenebank();
- cGenotype * cur_genotype = genebank.GetBestGenotype();
- for (int i = 0; i < genebank.GetSize(); i++) {
- if (cur_genotype->GetDepth() < min) min = cur_genotype->GetDepth();
- if (cur_genotype->GetDepth() > max) max = cur_genotype->GetDepth();
- cur_genotype = cur_genotype->GetNext();
- }
- assert(max >= min);
-
- // Allocate the array for the bins (& zero)
- int * n = new int[max - min + 1];
- for (int i = 0; i < max - min + 1; i++) n[i] = 0;
-
- // Loop through all genotypes binning the values
- cur_genotype = genebank.GetBestGenotype();
- for (int i = 0; i < genebank.GetSize(); i++) {
- n[cur_genotype->GetDepth() - min] += cur_genotype->GetNumOrganisms();
- cur_genotype = cur_genotype->GetNext();
- }
-
- // Actual output
- fp << pop->GetStats().GetUpdate() << " "
- << min << " "
- << max;
-
- for (int i = 0; i < max - min + 1; i++) fp << " " << n[i];
- fp<<endl;
-}
-
-
-void cAnalyzeUtil::PrintGenotypeAbundanceHistogram(ofstream & fp,
- cPopulation * pop)
-{
- assert(fp.good());
-
- // Allocate array for the histogram & zero it
- tArray <int> hist(pop->GetGenebank().GetBestGenotype()->GetNumOrganisms());
- for (int i = 0; i < hist.GetSize(); i++) hist[i] = 0;
-
- // Loop through all genotypes binning the values
- cGenotype * cur_genotype = pop->GetGenebank().GetBestGenotype();
- for (int i = 0; i < pop->GetGenebank().GetSize(); i++) {
- assert( cur_genotype->GetNumOrganisms() - 1 >= 0 );
- assert( cur_genotype->GetNumOrganisms() - 1 < hist.GetSize() );
- hist[cur_genotype->GetNumOrganisms() - 1]++;
- cur_genotype = cur_genotype->GetNext();
- }
-
- // Actual output
- fp << pop->GetStats().GetUpdate() << " ";
- for (int i = 0; i < hist.GetSize(); i++) fp<<hist[i]<<" ";
- fp << endl;
-}
-
-
-void cAnalyzeUtil::PrintSpeciesAbundanceHistogram(ofstream & fp,
- cPopulation * pop)
-{
- int max = 0;
- assert(fp.good());
-
- // Find max species abundance...
- cGenebank & genebank = pop->GetGenebank();
- cSpecies * cur_species = genebank.GetFirstSpecies();
- for (int i = 0; i < genebank.GetNumSpecies(); i++) {
- if (max < cur_species->GetNumOrganisms()) {
- max = cur_species->GetNumOrganisms();
- }
- cur_species = cur_species->GetNext();
- }
-
- // Allocate array for the histogram & zero it
- tArray <int> hist(max);
- for (int i = 0; i < hist.GetSize(); i++) hist[i] = 0;
-
- // Loop through all species binning the values
- cur_species = genebank.GetFirstSpecies();
- for (int i = 0; i < genebank.GetNumSpecies(); i++) {
- assert( cur_species->GetNumOrganisms() - 1 >= 0 );
- assert( cur_species->GetNumOrganisms() - 1 < hist.GetSize() );
- hist[cur_species->GetNumOrganisms() -1]++;
- cur_species = cur_species->GetNext();
- }
-
- // Actual output
- fp << pop->GetStats().GetUpdate() << " ";
- for (int i = 0; i < hist.GetSize(); i++) fp<<hist[i]<<" ";
- fp<<endl;
-}
-
-
-// this adds support for evan dorn's InstructionHistogramEvent. -- kgn
-/**
- * Count the number of each instruction present in the population. Output
- * this info to a log file
- **/
-void cAnalyzeUtil::PrintInstructionAbundanceHistogram(cWorld* world, ofstream & fp)
-{
- cPopulation* pop = &world->GetPopulation();
- int i,x,y;
- int num_inst=0;
- int mem_size=0;
- int * inst_counts;
- cCPUMemory cpu_mem; // cCPUMemory is a subclass of cGenome
- assert(fp.good());
-
- // ----- number of instructions available?
- num_inst = world->GetNumInstructions();
- inst_counts= new int[num_inst];
-
- // ----- create and initialize counting array
- inst_counts = new int[num_inst];
- for (i=0;i<num_inst;i++)
- { inst_counts[i]=0;
- }
-
- int num_cells = pop->GetSize();
- //looping through all CPUs counting up instructions
- for( x=0; x<num_cells; x++ )
- { cPopulationCell & cell=pop->GetCell(x);
- if (cell.IsOccupied())
- {
- // access this CPU's code block
- cpu_mem=cell.GetOrganism()->GetHardware().GetMemory();
- mem_size=cpu_mem.GetSize();
- for (y=0; y<mem_size ; y++)
- { inst_counts[(cpu_mem[y]).GetOp()]++;
- }
- }
- }
-
- // ----- output instruction counts
- for(i=0; i<num_inst; i++)
- { fp<<inst_counts[i]<<" ";
- }
- fp<<endl;
-}
-
Deleted: branches/brysonda/source/main/cAnalyzeUtil.h
===================================================================
--- branches/brysonda/source/main/cAnalyzeUtil.h 2005-11-04 00:39:30 UTC (rev 368)
+++ branches/brysonda/source/main/cAnalyzeUtil.h 2005-11-04 03:50:46 UTC (rev 369)
@@ -1,72 +0,0 @@
-//////////////////////////////////////////////////////////////////////////////
-// Copyright (C) 1993 - 2001 California Institute of Technology //
-// //
-// Read the COPYING and README files, or contact 'avida at alife.org', //
-// before continuing. SOME RESTRICTIONS MAY APPLY TO USE OF THIS FILE. //
-//////////////////////////////////////////////////////////////////////////////
-
-#ifndef ANALYZE_UTIL_HH
-#define ANALYZE_UTIL_HH
-
-#include <fstream>
-
-class cGenome;
-class cInstSet;
-class cPopulation;
-class cWorld;
-
-// This is a static class used to do various forms of complex analysis
-// on genomes.
-
-class cAnalyzeUtil {
-private:
-public:
- // Generic test-CPU analysis
- static void TestGenome(cWorld* world, const cGenome & genome, cInstSet & inst_set,
- std::ofstream & fp, int update);
-
- static void TestInsSizeChangeRobustness(cWorld* world, std::ofstream & fp,
- const cInstSet & inst_set, const cGenome & in_genome,
- int num_trials, int update);
-
-
- // Landscape-based analysis
- static cGenome CalcLandscape(cWorld* world, int dist, const cGenome & genome,
- cInstSet & inst_set);
- static void AnalyzeLandscape(cWorld* world, const cGenome & genome, cInstSet & inst_set,
- int sample_size=1000, int min_found=0,
- int max_sample_size=0, int update=-1);
- static void PairTestLandscape(cWorld* world, const cGenome & genome, cInstSet & inst_set,
- int sample_size=0, int update=-1);
-
-
- // Population-wide analysis
- static void CalcConsensus(cWorld* world, int lines_saved);
-
- static void AnalyzePopulation(cWorld* world, std::ofstream & fp,
- double sample_prob=1, bool landscape=false,
- bool save_genotype=false);
-
- static void PrintDetailedFitnessData(cWorld* world, std::ofstream & datafp,
- std::ofstream & histofp, std::ofstream & histo_testCPU_fp, bool save_max_f_genotype,
- bool print_fitness_histo, double hist_fmax, double hist_fstep);
-
- static void PrintGeneticDistanceData(cWorld* world, std::ofstream & fp,
- const char *creature_name );
- static void GeneticDistancePopDump(cWorld* world, std::ofstream & fp,
- const char * creature_name, bool save_creatures=false);
-
- static void TaskSnapshot(cWorld* world, std::ofstream & fp);
- static void TaskGrid(cWorld* world, std::ofstream & fp);
- static void PrintViableTasksData(cWorld* world, std::ofstream &fp);
- static void PrintTreeDepths(cPopulation * pop, std::ofstream & fp);
-
- static void PrintDepthHistogram(std::ofstream &fp, cPopulation * pop);
- static void PrintGenotypeAbundanceHistogram(std::ofstream &fp, cPopulation * pop);
- static void PrintSpeciesAbundanceHistogram(std::ofstream &fp, cPopulation * pop);
-
- // this adds support for evan dorn's InstructionHistogramEvent. -- kgn
- static void PrintInstructionAbundanceHistogram(cWorld* world, std::ofstream &fp);
- // -- kgn
-};
-#endif
More information about the Avida-cvs
mailing list