complearn 0.6.2

data/tests/sshagent-test.rb

@@ -0,0 +1,48 @@
+require 'test/unit'
+require 'CompLearnLib/RunEnv'
+include MRunEnv
+require 'CompLearnLib/CLConfig'
+require 'CompLearnLib/Task'
+include MTask
+
+class TaskMasterParTestTask < Task
+  attr_reader :misc
+  def initialize(i) @i = i end
+  def execute() reply(@i * @i) end
+end
+
+class TC_SSHAgentTest < Test::Unit::TestCase
+  def test_SSHAgent
+    assert(ENV.has_key?('SSH_AGENT_PID'), 'ssh-agent is not available')
+  end
+  def test_livingHosts
+    cfg = CLConfig.getDefaultCLConfig()
+    assert(cfg.hosts != nil)
+    assert(cfg.hosts.size >= 1)
+    if (cfg.hosts.size > 1)
+      h = cfg.hosts[1]
+      cmd = 'echo testing-party'
+      result = SSHAgent.runAt(cmd, h)
+      assert(result =~ /testing-party/, "Cannot run command #{cmd} on #{h}")
+    end
+  end
+
+# This test may work, but not until LAM/MPI is installed
+  def test_MPI
+    if CLConfig.getDefaultCLConfig().hosts.size > 1 && !CLConfig.getDefaultCLConfig().isSingleProcess?()
+      TaskMaster.init
+      sum = 0
+      todo = 4
+      4.times { |i|
+        TaskMaster.enqueue(TaskMasterParTestTask.new(i)) { |s, t, src|
+        sum += s
+        todo -= 1
+        }
+      }
+      while todo > 0
+        TaskMaster.waitForSlave()
+      end
+    assert(sum == 14, "TaskMaster parallel-threading is not working")
+    end
+  end
+end

data/tests/tests.rb

@@ -0,0 +1,275 @@
+##########################################################################
+#
+# Test suite for CompLearn system
+#
+# Invoke with
+# ruby tests/tests.rb # normal test suite
+# ruby tests/alltests.rb # run all tests, including ssh-agent (password req'd)
+#
+
+require 'test/unit'
+
+require 'CompLearn'
+require 'CompLearnLib/CLConfig'
+require 'CompLearnLib/Tree'
+require 'CompLearnLib/RunEnv'
+require 'CompLearnLib/DistMatrix'
+
+#require 'bz2'
+require 'CompLearnLib/Task'
+
+include MRunEnv
+include MTree
+include MTask
+
+class TaskMasterTestTask < Task
+  attr_reader :misc
+  def initialize(i) @i = i end
+  def execute() reply(@i * @i) end
+end
+
+class TC_TaskMasterTest < Test::Unit::TestCase
+  def test_TaskMasterSingle
+    CLConfig.setSingleUser
+    TaskMaster.init
+    sum = 0
+    4.times { |i|
+      TaskMaster.enqueue(TaskMasterTestTask.new(i)) { |s, t, src|
+        sum += s
+      }
+    }
+    assert(sum == 14, "TaskMaster single-threading is not working")
+  end
+end
+
+class TC_EnvironmentTest < Test::Unit::TestCase
+  def test_configFile
+    cfg = CLConfig.getDefaultCLConfig()
+    assert(cfg != nil)
+    assert(cfg.hosts.size >= 1)
+  end
+#  def test_BZ2
+#    a = 'the best coders are the lazy ones.'
+#    b = BZ2.compress(a)
+#    c = BZ2.decompress(b)
+#    assert(c == a)
+#  end
+end
+
+class TC_CLConfigFileParserTest < Test::Unit::TestCase
+  def test_CLConfigFileParser
+    fname = RunEnv.getTemporaryFilename()
+    assert(fname, "Cannot get temporary filename")
+    f = File.open(fname, "w")
+    assert(f, "Cannot open #{fname} for writing")
+    f.write <<EOF
+#
+# Test config file used by testconfig.rb
+#
+# $Id: tests.rb,v 1.6 2004/04/06 11:38:27 cilibrar Exp $
+
+# comment
+InputDir:     testInputDir
+OutputDir:    testOutputDir
+WorkDir:      testWorkDir
+
+Symmetric:    false
+
+Hosts:        
+  - host1
+  - host2
+  - host3
+
+EOF
+    f.close
+
+    cfg = CLConfig.new(fname)
+
+    assert("testInputDir" == cfg.inputDir, "Error with InputDir")
+    assert("testOutputDir"== cfg.outputDir, "Error with OutputDir")
+    assert("testWorkDir"  == cfg.workDir, "Error with WorkDir")
+    assert(!cfg.symmetric, "Error reading Symmetric")
+    assert(["host1","host2","host3"] == cfg.hosts, "Error with hosts")
+
+    File.unlink(fname)
+
+    cfg = CLConfig.getDefaultCLConfig
+
+  end
+
+end
+
+class TC_Tree < Test::Unit::TestCase
+  def test_TreeScore
+    dm = [ 
+          [ 0.1, 0.5, 0.1, 0.1],
+          [ 0.5, 0.1, 0.1, 0.1],
+          [ 0.1, 0.1, 0.1, 0.1],
+          [ 0.1, 0.1, 0.1, 0.1],
+         ]
+    t = TreeScore.makeFullList(dm)
+    assert(t.best == -0.2, "Best tree score is not -0.2 like it should be; got #{t.best}.")
+    assert(t.worst == -0.6, "Worst tree score is not -0.6 like it should be; got #{t.worst}.")
+  end
+  def test_TreeBasics
+    t = Tree.randomTree(8)
+    assert(t.verifyTree)
+    ot = t.clone()
+    assert((ot <=> t) == 0)
+    assert(t.getSpeciesCount() == 8)
+    assert(t.getKernelCount() == 6)
+    assert(t.getNodeCount() == 14)
+    k = t.randomKernelNodes(3)
+    k.each { |n| assert(t.isKernel?(n)) }
+    s = t.randomSpeciesNodes(5)
+    s.each { |n| assert(t.isSpecies?(n)) }
+    a = t.randomNode()
+    p = t.findPath(a,a)
+    assert(p.size == 1)
+    assert(p[0] == a)
+    20.times {
+      a,b = t.randomNodes(2)
+      p = t.findPath(a,b)
+      assert(p[0] == a)
+      assert(p[-1] == b)
+      (p.size-1).times { |n|
+        assert(p[n] != p[n+1])
+        assert(t.connected?(p[n],p[n+1]))
+      }
+    }
+    50.times { # should n3v3r be the sam3 aft3r a mut4t10n
+      t = ot.clone
+      assert((ot <=> t) == 0)
+      t.mutateSpecies
+      assert((ot <=> t) != 0)
+      assert(t.verifyTree())
+      t = ot.clone
+      t.mutateSubtreeInterchange
+      assert(t.verifyTree())
+      assert((ot <=> t) != 0)
+      t = ot.clone
+      t.mutateSubtreeTransfer
+      assert((ot <=> t) != 0)
+      assert(t.verifyTree())
+      t = ot.clone
+      t.mutateRandom
+      assert((ot <=> t) != 0)
+      assert(t.verifyTree())
+      t = ot.clone
+      t.mutateComplex
+      assert((ot <=> t) != 0)
+      assert(t.verifyTree())
+    }
+    t = ot.clone
+    t.mutateSpecies
+    begin
+      z = t.clone
+      z.mutateSpecies
+    end until ((ot <=> z) == 0) # should be able to mutate back
+    t = ot.clone
+    t.mutateSubtreeInterchange
+    begin
+      z = t.clone
+      z.mutateSubtreeInterchange
+    end until ((ot <=> z) == 0)
+    5.times {
+      t = ot.clone
+      t.mutateSubtreeTransfer
+      counter=0
+      begin
+        counter += 1
+        z = t.clone
+        z.mutateSubtreeTransfer
+      end until ((ot <=> z) == 0)
+    }
+    t = ot.clone
+    t.mutateRandom
+    begin
+      z = t.clone
+      z.mutateComplex
+    end until ((ot <=> z) == 0)
+end
+
+class TC_RunEnv < Test::Unit::TestCase
+  def test_tmpnames
+    a,b,c = RunEnv.getTemporaryFilenames(3)
+    assert( (a <= b) != 0 )
+    assert( (b <=> c) != 0 )
+    assert( (a <=> c) != 0 )
+    f = File.open(a, "wb")
+    f.write("hi!\n")
+    f.close()
+  end
+end
+
+class TC_DistMatrix < Test::Unit::TestCase
+  def test_distmatrix()
+    a,b,c = RunEnv.getTemporaryFilenames(3)
+    f = File.open(a, "wb")
+    f.write <<EOF
+   20
+BlueWhale   0.000000 0.956323 0.972438 0.933378 0.602634 0.976144 0.979629 0.952030 0.951910 0.944889 0.969844 0.976034 0.981995 0.977991 0.982811 0.974382 0.972313 0.976209 0.982206 0.945881 
+Cat         0.956323 0.000000 0.974313 0.944650 0.956363 0.975953 0.975125 0.926396 0.928524 0.932140 0.964397 0.977328 0.979132 0.977075 0.982982 0.974569 0.964113 0.979347 0.976386 0.937550 
+Chimpanzee  0.972438 0.974313 0.000000 0.970806 0.977246 0.873617 0.727796 0.972297 0.971715 0.968993 0.981926 0.657387 0.985329 0.851432 0.987256 0.428991 0.976425 0.850383 0.986032 0.969842 
+Cow         0.933378 0.944650 0.970806 0.000000 0.938797 0.972798 0.972197 0.941317 0.942067 0.940301 0.967403 0.973620 0.978456 0.973127 0.980380 0.970828 0.969176 0.977041 0.973974 0.933010 
+FinbackWhale  0.602634 0.956363 0.977246 0.938797 0.000000 0.981417 0.979665 0.957940 0.958546 0.947599 0.976451 0.980493 0.981616 0.978559 0.982824 0.979584 0.970869 0.980316 0.982494 0.948306 
+Gibbon      0.976144 0.975953 0.873617 0.972798 0.981417 0.000000 0.880709 0.973661 0.976154 0.969450 0.981696 0.880203 0.984314 0.891729 0.990322 0.875655 0.979186 0.891283 0.985513 0.968633 
+Gorilla     0.979629 0.975125 0.727796 0.972197 0.979665 0.880709 0.000000 0.975026 0.974955 0.973320 0.982952 0.732325 0.986312 0.857432 0.986569 0.720617 0.982154 0.854869 0.986469 0.969977 
+GreySeal    0.952030 0.926396 0.972297 0.941317 0.957940 0.973661 0.975026 0.000000 0.428044 0.938868 0.968053 0.975760 0.979947 0.978015 0.981756 0.975585 0.967721 0.976995 0.977296 0.933762 
+HarborSeal  0.951910 0.928524 0.971715 0.942067 0.958546 0.976154 0.974955 0.428044 0.000000 0.934823 0.965112 0.974737 0.980340 0.978275 0.981906 0.974849 0.966301 0.975989 0.977183 0.934832 
+Horse       0.944889 0.932140 0.968993 0.940301 0.947599 0.969450 0.973320 0.938868 0.934823 0.000000 0.972751 0.971558 0.977220 0.975414 0.984268 0.970180 0.964328 0.975455 0.978211 0.879802 
+HouseMouse  0.969844 0.964397 0.981926 0.967403 0.976451 0.981696 0.982952 0.968053 0.965112 0.972751 0.000000 0.980400 0.974406 0.983850 0.981355 0.980623 0.901139 0.983575 0.973949 0.968500 
+Human       0.976034 0.977328 0.657387 0.973620 0.980493 0.880203 0.732325 0.975760 0.974737 0.971558 0.980400 0.000000 0.986243 0.847139 0.988041 0.654234 0.981715 0.841775 0.985926 0.973694 
+Opossum     0.981995 0.979132 0.985329 0.978456 0.981616 0.984314 0.986312 0.979947 0.980340 0.977220 0.974406 0.986243 0.000000 0.987712 0.984083 0.985832 0.979062 0.986586 0.942194 0.979566 
+Orangutan   0.977991 0.977075 0.851432 0.973127 0.978559 0.891729 0.857432 0.978015 0.978275 0.975414 0.983850 0.847139 0.987712 0.000000 0.990018 0.849235 0.982004 0.569297 0.986478 0.978454 
+Platypus    0.982811 0.982982 0.987256 0.980380 0.982824 0.990322 0.986569 0.981756 0.981906 0.984268 0.981355 0.988041 0.984083 0.990018 0.000000 0.987128 0.983857 0.989407 0.981747 0.986069 
+PygmyChimpanzee 0.974382 0.974569 0.428991 0.970828 0.979584 0.875655 0.720617 0.975585 0.974849 0.970180 0.980623 0.654234 0.985832 0.849235 0.987128 0.000000 0.976925 0.849978 0.984617 0.971307 
+Rat         0.972313 0.964113 0.976425 0.969176 0.970869 0.979186 0.982154 0.967721 0.966301 0.964328 0.901139 0.981715 0.979062 0.982004 0.983857 0.976925 0.000000 0.979808 0.976307 0.964887 
+SumatranOrangutan 0.976209 0.979347 0.850383 0.977041 0.980316 0.891283 0.854869 0.976995 0.975989 0.975455 0.983575 0.841775 0.986586 0.569297 0.989407 0.849978 0.979808 0.000000 0.988176 0.975742 
+Wallaroo    0.982206 0.976386 0.986032 0.973974 0.982494 0.985513 0.986469 0.977296 0.977183 0.978211 0.973949 0.985926 0.942194 0.986478 0.981747 0.984617 0.976307 0.988176 0.000000 0.972083 
+WhiteRhino  0.945881 0.937550 0.969842 0.933010 0.948306 0.968633 0.969977 0.933762 0.934832 0.879802 0.968500 0.973694 0.979566 0.978454 0.986069 0.971307 0.964887 0.975742 0.972083 0.000000 
+EOF
+    f.close
+
+#    dm, names = DistMatrix.readFile(a)
+#    DistMatrix.writeFile(b, dm, names)
+#    dm2, names2 = DistMatrix.readFile(b)
+#    DistMatrix.writeFile(c, dm2, names2)
+#    dm3, names3 = DistMatrix.readFile(c)
+#    assert(names.size == 20)
+#    assert(names2.size == 20)
+#    assert(names3.size == 20)
+#    assert((dm2 <=> dm3) == 0)
+#    [a,b,c].each { |n| File.unlink(n) }
+#    ts = TreeScore.makeFullList(dm)
+#    randtree = Tree.randomTree(dm.size)
+#    assert(randtree.verifyTree)
+#    score = ts.score(randtree)
+#    assert(score >= 0.0 && score <= 1.0)
+    #DistMatrix.makeFromDir("/ufs/cilibrar/src/mpicquart/data/genres", "genresdist.txt")
+  end
+end
+
+end
+
+class TC_WholeSystem < Test::Unit::TestCase
+  def test_wholesystem
+    exampledir = "../examples/genes"
+    a,b,treename = RunEnv.getTemporaryFilenames(3)
+    `ncdmatrix #{exampledir} >#{a}`
+    assert( File.exist?(a) && File.stat(a).size > 100, "Distance Matrix file #{a} problem!" )
+    `maketree #{a} #{treename} >#{b}`
+    lastscore = nil
+    File.open(b, 'r') { |f|
+      while line = f.gets
+        lastscore = $1.to_f if line =~ /([0-9.]+)\s*$/
+      end
+    }
+    assert(lastscore > 0.7, "Final tree score too low: #{lastscore}")
+    assert(File.exist?(treename), "No #{treename} file created!")
+    File.unlink(a)
+    File.unlink(b)
+    File.unlink(treename)
+  end
+end
+

metadata

@@ -0,0 +1,164 @@
+--- !ruby/object:Gem::Specification 
+rubygems_version: 0.8.1
+specification_version: 1
+name: complearn
+version: !ruby/object:Gem::Version 
+  version: 0.6.2
+date: 2004-10-02
+summary: The CompLearn Toolkit enables you to do data mining using data compression. It also allows you to do Quartet Tree Reconstruction as well as Support Vector Machine experiments with these techniques.
+require_paths: 
+  - lib
+author: Rudi Cilibrasi
+email: cilibrar@ofb.net
+homepage: http://complearn.sourceforge.net/
+rubyforge_project: complearn
+description: "Using this package, you can do powerful pattern recognition easily from your
+  shell command line using a simple suite of tools.  Applications include music
+  cateogrization, text classification, handwritten digit recognition, and
+  evolutionary tree reconstruction.  The technique is based on a very general
+  theoretical notion of Kolmogorov Complexity, and so is undoubtedly applicable to
+  many as yet unknown disciplines.  See the webpage for research papers explaining
+  the math behind these techniques."
+autorequire: 
+default_executable: ncd
+bindir: bin
+has_rdoc: true
+required_ruby_version: !ruby/object:Gem::Version::Requirement 
+  requirements: 
+    - 
+      - ">"
+      - !ruby/object:Gem::Version 
+        version: 0.0.0
+  version: 
+platform: ruby
+files: 
+  - AUTHORS
+  - COPYING
+  - ChangeLog
+  - INSTALL
+  - Makefile.am
+  - Makefile.in
+  - NEWS
+  - README
+  - aclocal.m4
+  - bin
+  - configure
+  - configure.ac
+  - doc
+  - examples
+  - ext
+  - install-sh
+  - missing
+  - mkinstalldirs
+  - scripts
+  - tests
+  - complearn.gemspec
+  - complearn-0.6.2.gem
+  - config.log
+  - o
+  - config.status
+  - Makefile
+  - confstat5FpLBf
+  - bin/Makefile.am
+  - bin/Makefile.in
+  - bin/labeltree.in
+  - bin/makesvm.in
+  - bin/maketree.in
+  - bin/ncd.in
+  - bin/ncdmatrix.in
+  - bin/ncdvector.in
+  - bin/Makefile
+  - bin/ncd
+  - bin/ncdvector
+  - bin/ncdmatrix
+  - bin/labeltree
+  - bin/maketree
+  - bin/makesvm
+  - doc/FAQ.txt
+  - doc/Makefile.am
+  - doc/Makefile.in
+  - doc/devguide.txt
+  - doc/example.complearnrc
+  - doc/examples.txt
+  - doc/man
+  - doc/readme.txt
+  - doc/userguide.txt
+  - doc/Makefile
+  - doc/man/Makefile.am
+  - doc/man/Makefile.in
+  - doc/man/complearn.5
+  - doc/man/labeltree.1
+  - doc/man/makesvm.1
+  - doc/man/maketree.1
+  - doc/man/ncd.1
+  - doc/man/ncdmatrix.1
+  - doc/man/ncdvector.1
+  - doc/man/Makefile
+  - examples/genes
+  - examples/genes/blueWhale.txt
+  - examples/genes/cat.txt
+  - examples/genes/chimpanzee.txt
+  - examples/genes/finWhale.txt
+  - examples/genes/graySeal.txt
+  - examples/genes/harborSeal.txt
+  - examples/genes/horse.txt
+  - examples/genes/human.txt
+  - examples/genes/mouse.txt
+  - examples/genes/rat.txt
+  - ext/Quartet.c
+  - ext/Quartet.h
+  - ext/TreeScore.c
+  - ext/TreeScore.h
+  - ext/config.h.in
+  - ext/extconf.rb
+  - ext/lib
+  - ext/config.h
+  - ext/Makefile
+  - ext/lib/CompLearnLib
+  - ext/lib/CompLearnLib/CompressionObject.rb
+  - ext/lib/CompLearnLib/CompressionTask.rb
+  - ext/lib/CompLearnLib/FoundComp.rb
+  - ext/lib/CompLearnLib/DistMatrix.rb
+  - ext/lib/CompLearnLib/FoundComp.rb.in
+  - ext/lib/CompLearnLib/Ncd.rb
+  - ext/lib/CompLearnLib/RunEnv.rb
+  - ext/lib/CompLearnLib/Task.rb
+  - ext/lib/CompLearnLib/TaskMaster.rb
+  - ext/lib/CompLearnLib/TaskMasterMPI.rb
+  - ext/lib/CompLearnLib/TaskMasterSingle.rb
+  - ext/lib/CompLearnLib/Tree.rb
+  - ext/lib/CompLearnLib/CLConfig.rb
+  - scripts/CompLearn.iss.in
+  - scripts/debian
+  - scripts/makeSetup.sh.in
+  - scripts/makedeb.zsh.in
+  - scripts/CompLearn.iss
+  - scripts/makeSetup.sh
+  - scripts/makedeb.zsh
+  - scripts/debian/changelog
+  - scripts/debian/control
+  - tests/alltests.rb
+  - tests/bz2test.rb
+  - tests/sshagent-test.rb
+  - tests/tests.rb
+  - confstat5FpLBf/subs.sed
+  - confstat5FpLBf/subs.frag
+  - confstat5FpLBf/subs-1.sed
+  - confstat5FpLBf/subs-2.sed
+  - confstat5FpLBf/undefs.sed
+  - confstat5FpLBf/config.h
+test_files: []
+rdoc_options: []
+extra_rdoc_files: []
+executables: 
+  - ncd
+  - ncdvector
+  - ncdmatrix
+  - maketree
+  - makesvm
+  - labeltree
+extensions: 
+  - ext/extconf.rb
+requirements: 
+  - "Data compressors such as bzip, gzip, or ppmz"
+dependencies: []