rbbt 1.2.5 → 2.0.0

data/lib/rbbt/util/tmpfile.rb

@@ -1,35 +0,0 @@
-require 'fileutils'
-require 'rbbt' 
-
-
-module TmpFile
-
-  # Creates a random file name, with the given suffix and a random number
-  # up to +max+
-  def self.random_name( s="",max=10000000)
-    n = rand(max)
-    s << n.to_s
-    s
-  end
-
-  # Creates a random filename in the temporary directory
-  def self.tmp_file(s = "",max=10000000)
-    File.join(Rbbt.tmpdir,random_name(s,max))
-  end
-
-  def self.with_file(content = nil)
-    tmpfile = tmp_file
-
-    File.open(tmpfile, 'w') do |f| f.write content end if content != nil
-
-    result = yield(tmpfile)
-
-    FileUtils.rm tmpfile if File.exists? tmpfile
-
-    result
-  end
-
-  class << self
-    alias :new :tmp_file
-  end
-end

data/tasks/install.rake

@@ -1,124 +0,0 @@
-require 'rbbt'
-
-$datadir = Rbbt.datadir
-$scriptdir = File.join(File.expand_path(Rbbt.rootdir), '/install_scripts')
-
-
-task 'abner' do
-  directory = "#{$datadir}/third_party/abner/"
-  if !File.exists?(File.join(directory, 'abner.jar')) || $force
-    FileUtils.mkdir_p directory
-    `cd #{directory};rm -Rf *; #{$scriptdir}/get_abner.sh;cd -`
-  end
-end
-
-task 'banner' do
-  directory = "#{$datadir}/third_party/banner/"
-  if !File.exists?(File.join(directory, 'banner.jar')) || $force 
-    FileUtils.mkdir_p directory
-    `cd #{directory};rm -Rf *; #{$scriptdir}/get_banner.sh;cd -`
-  end
-end
-
-task 'crf++' do
-  directory = "#{$datadir}/third_party/crf++/"
-  if !File.exists?(File.join(directory, 'ruby/CRFPP.so')) || $force
-    FileUtils.mkdir_p directory
-    `cd #{directory};rm -Rf *; #{$scriptdir}/get_crf++.sh;cd -`
-  end
-end
-
-
-
-task 'wordlists' do
-  FileUtils.cp_r File.join($scriptdir, 'wordlists/'), $datadir
-end
-
-task 'polysearch' do
-  directory = "#{$datadir}/dbs/polysearch/"
-  if !File.exists?(File.join(directory,'disease.txt')) || $force
-    FileUtils.mkdir_p directory
-    `cd #{directory}/; rm * -Rf; #{$scriptdir}/get_polysearch.sh;cd -`
-  end
-end
-
-
-task '3party' => %w(abner banner crf++)
-
-task 'entrez' do
-  directory = "#{$datadir}/dbs/entrez/"
-  if !File.exists?(File.join(directory,'gene_info')) || $force
-    FileUtils.mkdir_p directory
-    `cd #{directory}/; rm * -Rf; #{$scriptdir}/get_entrez.sh;cd -`
-  end
-end
-
-task 'go' do
-  directory = "#{$datadir}/dbs/go/"
-  if !File.exists?(File.join(directory,'gene_ontology.obo')) || $force
-    FileUtils.mkdir_p directory
-    `cd #{directory}/; rm * -Rf; #{$scriptdir}/get_go.sh;cd -`
-  end
-end
-
-task 'biocreative' do
-  directory = "#{$datadir}/biocreative/"
-  if !File.exists?(File.join(directory, 'BC2GN')) || $force
-    FileUtils.mkdir_p directory
-    `cd #{directory};rm -Rf *; #{$scriptdir}/get_biocreative.sh;cd -`
-  end
-end
-
-
-task 'datasets' => %w(entrez biocreative)
-
-task 'organisms' do
-  directory = "#{$datadir}/organisms"
-  FileUtils.mkdir_p directory
-  %w(Rakefile rake-include.rb).each{|f|
-    FileUtils.cp_r File.join($scriptdir, "organisms/#{ f }"), directory
-  }
-  Dir.glob(File.join($scriptdir, "organisms/*.Rakefile")).each{|f|
-    org = File.basename(f).sub(/.Rakefile/,'')
-    if !File.exists?(File.join(directory, org))
-      FileUtils.mkdir_p File.join(directory, org)
-    end
-    FileUtils.cp f , File.join(directory, "#{ org }/Rakefile")
-  }
-  `cd #{directory}; rake names`
-end
-
-task 'ner' do
-  directory = "#{$datadir}/ner"
-  FileUtils.mkdir_p directory
-  %w(Rakefile config).each{|f|
-    FileUtils.cp_r File.join($scriptdir, "ner/#{ f }"), directory 
-  }
-  
-  %w(data model results).each{|d|
-    FileUtils.mkdir_p File.join(directory, d)
-  }
-end
-
-task 'norm' do
-  directory = "#{$datadir}/norm"
-  FileUtils.mkdir_p directory
-  %w(Rakefile config functions.sh).each{|f|
-    FileUtils.cp_r File.join($scriptdir, "norm/#{ f }"), directory 
-  }
- %w(results models).each{|d|
-  FileUtils.mkdir_p File.join(directory, d)
-  }
-end
-
-task 'classifier' do
-  directory = "#{$datadir}/classifier"
-  FileUtils.mkdir_p directory
-  %w(Rakefile R).each{|f|
-    FileUtils.cp_r File.join($scriptdir, "classifier/#{ f }"), directory
-  }
-  %w(data model results).each{|d|
-    FileUtils.mkdir_p File.join(directory, d)
-  }
-end
-

data/test/rbbt/bow/test_bow.rb

@@ -1,33 +0,0 @@
-require File.dirname(__FILE__) + '/../../test_helper'
-require 'rbbt/bow/bow'
-require 'test/unit'
-
-class TestBow < Test::Unit::TestCase
-
-  def test_words
-    assert_equal(["hello", "world"], "Hello World".words)
-
-  end
-   
-  def test_terms
-    text = "Hello World"
-    assert_equal(["hello", "world"], BagOfWords.terms(text,false).keys.sort)
-    assert_equal(["hello", "hello world", "world"], BagOfWords.terms(text,true).keys.sort)
-  end
-
-  def test_features
-
-    text = "Hello world!"
-    text += "Hello World Again!"
-
-    assert_equal([2, 2], BagOfWords.features(text, "Hello World".words.uniq.sort))
-  end
-
-  def test_stem
-    assert_equal(["protein"], "Proteins".words)
-  end
-
-
-end
-
-

data/test/rbbt/bow/test_classifier.rb

@@ -1,72 +0,0 @@
-require File.dirname(__FILE__) + '/../../test_helper'
-require 'rbbt/bow/classifier'
-require 'rbbt/util/tmpfile'
-require 'rbbt/util/open'
-require 'test/unit'
-
-class TestClassifier < Test::Unit::TestCase
-
-  def test_build_model
-    features =<<-EOT
-Name	Class	hello	world
-row1	-	2	0
-row2	+	0	2
-    EOT
-
-    featuresfile = TmpFile.tmp_file("test_classifier")
-    modelfile = TmpFile.tmp_file("test_classifier")
-    Open.write(featuresfile, features)
-    Classifier.create_model(featuresfile, modelfile)
-
-    assert(File.exist? modelfile)
-
-    FileUtils.rm featuresfile
-    FileUtils.rm modelfile
-
-  end
-
-  def test_classifier
-    features =<<-EOT
-Name	Class	hello	world
-row1	-	2	0
-row2	+	0	2
-    EOT
-
-    featuresfile = TmpFile.tmp_file("test_classifier")
-    modelfile = TmpFile.tmp_file("test_classifier")
-    Open.write(featuresfile, features)
-    Classifier.create_model(featuresfile, modelfile)
-
-    FileUtils.rm featuresfile
-
-    classifier = Classifier.new(modelfile)
-
-    assert_equal(["hello", "world"], classifier.terms)
-
-    assert_equal(["-", "+"], classifier.classify_feature_array([[1,0],[0,1]]))
-
-
-    assert_equal({"negative"=>"-", "positive"=>"+"}, classifier.classify_feature_hash({:positive => [0,1], :negative => [1,0]}))
-    assert_equal({"negative"=>"-", "positive"=>"+"}, classifier.classify_feature_hash({:positive => [0,1], :negative => [1,0]}))
-
-    assert_equal(["-", "+"], classifier.classify_text_array(["Hello","World"]))
-
-    assert_equal({"negative"=>"-", "positive"=>"+"}, classifier.classify_text_hash({:negative => "Hello", :positive =>"World"}))
-
-    assert_equal('-', classifier.classify("Hello"))                  
-    assert_equal(["-", "+"],classifier.classify([[1,0],[0,1]]))
-    assert_equal({"negative"=>"-", "positive"=>"+"},classifier.classify({:positive => [0,1], :negative => [1,0]}))   
-    assert_equal(["-", "+"],classifier.classify(["Hello","World"]))        
-    #assert_equal({"negative"=>"-", "positive"=>"+"},classifier.classify({:negative => "Hello", :positive => "World"}))   
-
-
-    #assert_nothing_raised do classifier.classify("Unknown terms") end
-    #assert_nothing_raised do classifier.classify([]) end
-
-    FileUtils.rm modelfile
-
-
-  end
-
-end
-

data/test/rbbt/bow/test_dictionary.rb

@@ -1,91 +0,0 @@
-require File.dirname(__FILE__) + '/../../test_helper'
-require 'rbbt/bow/dictionary'
-require 'rbbt/bow/bow'
-require 'test/unit'
-
-class TestDictionary < Test::Unit::TestCase
-  
-  def test_standard
-    docs = []
-    docs << BagOfWords.terms("Hello World", false)
-    docs << BagOfWords.terms("Hello Yin Yin", false)
-
-    dict = Dictionary.new
-    docs.each{|doc| dict.add doc}
-    
-    assert_equal(2, dict.terms["hello"])
-    assert_equal(2, dict.terms["yin"])
-    assert_equal(0, dict.terms["bye"])
-    assert_equal(1, dict.terms["world"])
-  end
-
-  def test_tf_idf
-    docs = []
-    docs << BagOfWords.terms("Hello World", false)
-    docs << BagOfWords.terms("Hello Yin Yin", false)
-
-
-    dict = Dictionary::TF_IDF.new
-    docs.each{|doc| dict.add doc}
-
-    assert_equal(2, dict.terms["hello"])
-    assert_equal(2, dict.terms["yin"])
-    assert_equal(0, dict.terms["bye"])
-    assert_equal(1, dict.terms["world"])
- 
-
-    assert_equal(1,   dict.df["hello"])
-    assert_equal(0.5, dict.df["yin"])
-    assert_equal(0,   dict.df["bye"])
-    assert_equal(0.5, dict.df["world"])
-
-    assert_equal(2.0/5, dict.tf["hello"])
-    assert_equal(2.0/5, dict.tf["yin"])
-    assert_equal(0,     dict.tf["bye"])
-    assert_equal(1.0/5,   dict.tf["world"])
-
-    assert_equal(Math::log(1), dict.idf["hello"])
-    assert_equal(Math::log(2), dict.idf["yin"])
-    assert_equal(0,            dict.idf["bye"])
-    assert_equal(Math::log(2), dict.idf["world"])
-
-    assert_equal(2.0/5 * Math::log(1),   dict.tf_idf["hello"])
-    assert_equal(2.0/5 * Math::log(2), dict.tf_idf["yin"])
-    assert_equal(0,                      dict.tf_idf["bye"])
-    assert_equal(1.0/5 * Math::log(2), dict.tf_idf["world"])
-  end
-
-  def test_best
-    docs = []
-    docs << BagOfWords.terms("Hello World", false)
-    docs << BagOfWords.terms("Hello Yin Yin", false)
-
-
-    dict = Dictionary::TF_IDF.new
-    docs.each{|doc| dict.add doc}
-
-    assert_equal(1, dict.best(:limit => 1).length)
-    assert(dict.best(:limit => 1).include? "yin")
-  end
- 
-  def test_kl
-    docs = []
-    docs << [BagOfWords.terms("Hello World", false), :+]
-    docs << [BagOfWords.terms("Hello Cruel World", false), :+]
-    docs << [BagOfWords.terms("Hello Yan Yan", false), :-]
-    docs << [BagOfWords.terms("Hello Yin Yin", false), :-]
-
-
-    dict = Dictionary::KL.new
-    docs.each{|doc| dict.add *doc}
-
-    assert_equal(0, dict.kl["hello"])
-    assert_equal(dict.kl['yan'], dict.kl['yin'])
-    assert_in_delta(1 * Math::log(1 / 0.000001), dict.kl["world"],0.01)
-    assert_in_delta(0.5 * Math::log(0.5 / 0.000001), dict.kl["cruel"],0.01)
-  end
-
-
-end
-
- 

data/test/rbbt/ner/rnorm/test_cue_index.rb

@@ -1,57 +0,0 @@
-require File.dirname(__FILE__) + '/../../../test_helper'
-require 'rbbt/ner/rnorm/cue_index'
-require 'rbbt/util/misc'
-require 'rbbt/util/tmpfile'
-require 'rbbt/util/open'
-require 'test/unit'
-
-class TestCUE < Test::Unit::TestCase
-
-  def setup
-    @index = CueIndex.new do
-      equal    do |w| [w] end
-      standard do |w| [w.downcase.split(/\s+/).sort.join("")] end
-      special  do |w| s = w.split.select{|w| w.is_special?}.collect{|w| w.downcase.sub(/p$/,'')} end
-      words    do |w| 
-        w.scan(/[a-z]+/i).
-          select{|w| w.length > 2}.
-          sort{|a,b| b.length <=> a.length}. 
-          collect{|n| n.downcase}
-      end
-    end
-  end
-
-  def test_cue
-    assert_equal([["Hsp70 gene"], ["genehsp70"], ["hsp70"], ["gene", "hsp"]], @index.cues("Hsp70 gene"))
-  end
-
-  def test_load
-    tmp = TmpFile.tmp_file("test_cue")
-
-    lexicon =<<-EOT
-code1\tNAME1\tname 1
-code2\tNAME2\tname 2
-    EOT
-    Open.write(tmp,lexicon)
-
-    assert_raise(CueIndex::LexiconMissingError){@index.match("NAME2")}
-    @index.load(tmp)
-    assert_equal(["code2"], @index.match("NAME2"))
-
-    FileUtils.rm tmp
-  end
-
-  #def test_yeast
-  #  index  = CueIndex.new
-  #  index.load(File.join(Rbbt.datadir,'biocreative','BC1GN','yeast','synonyms.list'))
-  #  assert(index.match("Met - 31").include? 'S0005959')
-  #end
-
-  #def test_mouse
-  #  index  = CueIndex.new
-  #  index.load(File.join(Rbbt.datadir,'biocreative','BC1GN','mouse','synonyms.list'))
-  #  puts index.match("kreisler gene").length
-  #end
-
-
-end

data/test/rbbt/ner/rnorm/test_tokens.rb

@@ -1,70 +0,0 @@
-require File.dirname(__FILE__) + '/../../../test_helper'
-require 'rbbt/ner/rnorm/tokens'
-require 'rbbt/util/misc'
-require 'rbbt/util/tmpfile'
-require 'rbbt/util/open' 
-require 'test/unit' 
-
-class TestCompare < Test::Unit::TestCase 
- 
-  def setup
-    @index = Tokenizer.new  
-  end
-  
-  def test_type   
-    assert_equal(:gene, @index.type("gene"))
-    assert_equal(:dna, @index.type("dna"))
-    assert_equal(:number, @index.type("121"))
-  end
-
-  def test_token_types
-    assert_equal([["dna", :dna], ["12", :number]], @index.token_types("dna12"))
-    assert_equal([["REX", :special], ["12", :number]], @index.token_types("REX12"))
-    assert_equal([["SSH", :special], ["3", :number], ["BP", :special]], @index.token_types("SSH3BP"))
-    assert_equal([["HP", :special], ["1", :number], ["gamma", :greek]], @index.token_types("HP1gamma"))
-    assert_equal([["HP", :special], ["1", :number], ["GAMMA", :greek]], @index.token_types("HP1-GAMMA"))
-  end
-
-  def test_eval 
-    assert_equal(3, @index.evaluate_tokens(@index.token_types("1"), @index.token_types("1"))) 
-  end 
-
-  def test_transforms 
-    t = Tokenizer::Transform.new.unknown do |t| [t, if t.length < 4 then :special else :unknown end] end 
-    assert_equal(["BP", :special], t.transform(["BP",:unknown]))
-  end
-  def test_comparisons
-    assert_equal(0, Tokenizer::Operation.new(:same).number(3).eval(@index.token_types("SSH1"),@index.token_types("SSH2")))
-    assert_equal(3, Tokenizer::Operation.new(:same).number(3).eval(@index.token_types("SSH1"),@index.token_types("SSH1")))
-    assert_equal(0, Tokenizer::Operation.new(:same).special(1).eval([["SSH", :special],["1", :number]],[["SSH", :special],["3", :number],["BP",:special]]))
-    assert_equal(-1, Tokenizer::Operation.new(:diff).special(-1).eval([["SSH", :special],["1", :number]],[["SSH", :special],["3", :number],["BP",:special]]))
-    assert_equal(-1, Tokenizer::Operation.new(:extr).special(-1).eval([["SSH", :special],["1", :number]],[["SSH", :special],["3", :number],["BP",:special]]))
-    assert_equal(-1, Tokenizer::Operation.new(:miss).special(-1).eval([["SSH", :special],["3", :number],["BP",:special]],[["SSH", :special],["1", :number]]))
-  end
-  def test_ignore_case
-    assert_equal(-1, Tokenizer::Operation.new(:diff).ignore_case(false).special(-1).eval([["ssh", :special]],[["SSH", :special]]))
-    assert_equal(0, Tokenizer::Operation.new(:diff).ignore_case(true).special(-1).eval([["ssh", :special]],[["SSH", :special]]))
-  end
-
-  def test_compare
-     assert_equal(-10, @index.evaluate("DNA1", "GENE2"))
-     assert_equal(3, @index.evaluate("DNA1", "GENE1")) 
-     assert_equal(3, @index.evaluate("DNA1", "RNA1")) 
-     assert_equal(-1, @index.evaluate("SSH", "SSH1")) 
-     assert_equal(7, @index.evaluate("pol III", "POL3")) 
-  end 
-
-  def test_default
-    index = Tokenizer.new 
-    assert(index.evaluate("SSH", "SSH1") > index.evaluate("SSH", "SSH3BP")) 
-    assert(index.evaluate("HP1gamma", "HP1-GAMMA") > 1) 
-    assert(index.evaluate("HP1alpha", "HP1 alpha") > 1) 
-    assert(index.evaluate("IL-1beta", "IL-1 beta") > 1) 
-    assert(index.evaluate("IL-1RI", "IL-1R-1") > 1) 
-    assert(index.evaluate("MODI", "MOD 1") > 1) 
-    assert(index.evaluate("MOD 1", "MODI") > 1) 
-    assert(index.evaluate("Ubc3", "Ubc3b") > 1) 
-  end
-  
-
-end