rbbt-text 1.3.0 → 1.3.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: c2a24d8e7faf30d53e41a00a27f6145e8e9f18f0c10af57cdddaea0ee18c35d6
-  data.tar.gz: 3475006965110391e35151cd1b5368028dacf467aa276f8eb68fce3320be1122
+  metadata.gz: a6965ecde1b38d5bc93d4836ee6d757e2add39a51d64c2f06142bbbd303e22d7
+  data.tar.gz: a5c32ea03ea8214dd8c94ef6e884b59e459e3a7a8e3d26065a0a046b5b9b4778
 SHA512:
-  metadata.gz: da40a039a4792eb5e7fa00270870279221c74dcbf51df1b5278b247496fefbfa888a87b7ab19f05676644c51a01177eb49e229cb0156fe7f0190dd4933d41e24
-  data.tar.gz: a32fca5f21a987dcbb6b5541015cc33879330e6f1ef7c4a28e75debe5bdd1dc8bf7b98bfc91d828e605f29868aa972b55cd59bb4f86e66d2fb0cfea31fac2ae0
+  metadata.gz: 756d240a796e5ac88b4b55368e0e4e3af14b3dd2d8b8b55e49839c3cdc3fa45ee807d648cf86b45b62e7f2f4d9e7fc15567ab21d3356e37a5c3c4316cbcaa841
+  data.tar.gz: 6caa03ec51185cac00cc436bac999b063fccfcc1dbf0e2c09359dad7171c0eea37f80436cc860038a2c1ad17eb9b67a03e88d1ae8ef406ce1c5c874d375d1abd

data/lib/rbbt/bow/bow.rb

@@ -69,6 +69,11 @@ module BagOfWords
     count = bigrams ? count(bigrams(text)) : count(words(text))
     count.values_at(*terms)
   end
+
+  def self.weighted_features(text, weights)
+    features = features(text, weights.keys)
+    features.zip(weights.values).collect{|f,w| f * w } 
+  end
 end
 
 class String

data/lib/rbbt/bow/dictionary.rb

@@ -74,28 +74,32 @@ class Dictionary::TF_IDF
   end
 
   def best(options = {})
-    high, low, limit = {
-      :low   => 0,
-      :high    => 1,
-    }.merge(options).
-    values_at(:high, :low, :limit)
-
-    num_docs = @num_docs.to_f
-    best = df.select{|term, value|
-      value >= low && value <= high
-    }.collect{|p| 
-      term     = p.first
-      df_value = p.last
-      [term,
-       @terms[term].to_f / num_docs * Math::log(1.0/df_value)
-      ]
-    }
-
-    if limit
-      Hash[*best.sort{|a,b| b[1] <=>  a[1]}.slice(0, limit).flatten]
-    else
-      Hash[*best.flatten]
-    end
+    key = Misc.obj2digest(options)
+    @best ||= {}
+    @best[key] ||= begin
+                     high, low, limit = {
+                       :low   => 0,
+                       :high    => 1,
+                     }.merge(options).
+                     values_at(:high, :low, :limit)
+
+                     num_docs = @num_docs.to_f
+                     best = df.select{|term, value|
+                       value >= low && value <= high
+                     }.collect{|p| 
+                       term     = p.first
+                       df_value = p.last
+                       [term,
+                        @terms[term].to_f / num_docs * Math::log(1.0/df_value)
+                       ]
+                     }
+
+                     if limit
+                       Hash[*best.sort{|a,b| b[1] <=>  a[1]}.slice(0, limit).flatten]
+                     else
+                       Hash[*best.flatten]
+                     end
+                   end
   end
 
   def weights(options = {})

data/lib/rbbt/segment/transformed.rb

@@ -68,6 +68,10 @@ module Transformed
 
     segments = [segments] unless Array === segments 
     orig_length = self.length
+
+    offset = self.respond_to?(:offset) ? self.offset : 0
+    segments = segments.select{|s| s.offset >= offset && s.offset <= offset + self.length - 1 }
+
     Segment.clean_sort(segments).each do |segment|
       next if segment.offset.nil?
 
@@ -86,7 +90,7 @@ module Transformed
 
       updated_text = self[updated_begin..updated_end]
       if updated_text.nil?
-        Log.warn "Range outside of segment: #{self.length} #{segment.locus} (#{updated_range})"
+        Log.warn "Range outside of segment: #{self.length} #{segment.range} (#{updated_range})"
         next
       end
 

data/test/rbbt/segment/test_transformed.rb

@@ -101,6 +101,35 @@ More recently, PPAR activators were shown to inhibit the activation of inflammat
     assert_equal "CDK5R1 protein", exp2
   end
 
+  def test_with_transform_sentences
+    a = "This first sentence mentions Bread. This sentence mentions the TP53 gene and the CDK5R1 protein"
+    original = a.dup
+
+    gene1 = "TP53"
+    gene1.extend NamedEntity
+    gene1.offset = a.index gene1
+
+    gene2 = "CDK5R1"
+    gene2.extend NamedEntity
+    gene2.offset = a.index gene2
+
+    bread = "Bread"
+    bread.extend NamedEntity
+    bread.offset = a.index bread
+
+    sentences = Segment.align(a, a.split(". "))
+
+    Transformed.with_transform(sentences[1], [gene1, gene2, bread], "GN") do
+      assert sentences[1].include?("GN gene and the GN protein")
+    end
+
+    Transformed.with_transform(sentences[0], [gene1, gene2, bread], "BR") do
+      assert sentences[0].include?("first sentence mentions BR")
+    end
+
+
+  end
+
   def test_html
     a = "This sentence mentions the TP53 gene and the CDK5R1 protein"
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: rbbt-text
 version: !ruby/object:Gem::Version
-  version: 1.3.0
+  version: 1.3.1
 platform: ruby
 authors:
 - Miguel Vazquez
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2020-04-18 00:00:00.000000000 Z
+date: 2020-04-19 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rbbt-util