rbbt-dm 1.1.57 → 1.1.58

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 072d57536b223931947dfd71d5e359961c6bfe44da0870cd2dbab440aa91ba6c
-  data.tar.gz: 7c5d2a3fc627992231b621b6efe4d2160aa093b11c89f31f728ede5121d2edc9
+  metadata.gz: 286385d90b276d30cd5e1b21ae38c5e6a203e2ce3ac10673c434c19a2f45cfb1
+  data.tar.gz: 7879d74a364886ea8cb507be51c4979cfb598bdb273f948c3c3930a5dce199e6
 SHA512:
-  metadata.gz: 5216b3179bc4a809829b79bc4c447159e88444dc7943da3d0c3643b728e3eb5a6da7c08a8538fb88db1e463f95e55e104976ac81f010aeaf729a6cb1c4ca1374
-  data.tar.gz: 12baae07f04ca3770dfef81c3166385badb7944d25b7072364631e5f93b419e04c53d5b7157934b3d1522ef83e84acf0c4e2c5d39ad3604b7ef8f5f460d2f750
+  metadata.gz: b82c77bd736c8422e49c4dc83b63d6a91da6e76857af4b5cf5aff0a9a58b4147bc50b49b1b1534e8b07ca1bce5f6a5a673c5d688fb8cd7856623370d19fd1bda
+  data.tar.gz: 1b267a85ab600b878e99f414f725255cf086165a27f8cdec42ed83349b4f36bdb9e29615e0aaada9b30f098df8382e4778cebaf2b8649e17b8985e79d9b8bd23

data/LICENSE

@@ -1,4 +1,4 @@
-Copyright (c) 2010-2013 Miguel Vázquez García
+Copyright (c) 2010-2022 Miguel Vázquez García
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the

data/lib/rbbt/vector/model/random_forest.rb

@@ -16,11 +16,21 @@ model = randomForest(as.factor(label) ~ ., data = features);
 rbbt.require("randomForest");
 pred = names(model$forest$xlevels)
 for (p in pred) { 
-  if (class(features[[p]]) == "factor") { 
+  if (is.factor(features[[p]])) { 
       features[[p]] = factor(features[[p]], levels=model$forest$xlevels[[p]])
     } 
 }
 label = predict(model, features);
     EOF
   end
+
+  def importance
+    TmpFile.with_file do |tmp|
+      tsv = R.run <<-EOF
+load(file="#{model_file}");
+rbbt.tsv.write('#{tmp}', model$importance)
+      EOF
+      TSV.open(tmp)
+    end
+  end
 end

data/lib/rbbt/vector/model/spaCy.rb

@@ -30,6 +30,7 @@ class SpaCyModel < VectorModel
     @train_model = Proc.new do |file, features, labels|
       texts = features
       docs = []
+      unique_labels = labels.uniq
       tmpconfig = File.join(file, 'config')
       tmptrain = File.join(file, 'train.spacy')
       SpaCy.config(@config, tmpconfig)
@@ -37,14 +38,11 @@ class SpaCyModel < VectorModel
         nlp = SpaCy.nlp(lang)
         docs = []
         RbbtPython.iterate nlp.pipe(texts.zip(labels), as_tuples: true), :bar => "Training documents into spacy format" do |doc,label|
-          doc.cats[label] = 1
-          #if %w(1 true pos).include?(label.to_s.downcase) 
-          #  doc.cats["positive"] = 1
-          #  doc.cats["negative"] = 0
-          #else
-          #  doc.cats["positive"] = 0
-          #  doc.cats["negative"] = 1
-          #end
+          unique_labels.each do |other_label|
+            next if other_label == label
+            doc.cats[other_label] = false
+          end
+          doc.cats[label] = true
           docs << doc
         end
 
@@ -56,8 +54,9 @@ class SpaCyModel < VectorModel
       CMD.cmd_log(:spacy, "train #{tmpconfig} --output #{file} --paths.train #{tmptrain} --paths.dev #{tmptrain}",  "--gpu-id" => gpu)
     end
  
-    @eval_model = Proc.new do |file, features|
+    @eval_model = Proc.new do |file, features,list|
       texts = features
+      texts = [texts] unless list
 
       docs = []
       SpaCyModel.spacy do
@@ -68,7 +67,6 @@ class SpaCyModel < VectorModel
             cats = nlp.(text).cats
             bar.tick
             cats.sort_by{|l,v| v.to_f }.last.first
-            #cats['positive'] > cats['negative']  ? 1 : 0
           end
         end
       end

data/lib/rbbt/vector/model/svm.rb

@@ -3,16 +3,16 @@ class SVMModel < VectorModel
   def initialize(dir)
     super(dir)
 
-    @extract_features = Proc.new{|element|
+    @extract_features ||= Proc.new{|element|
       element
     }
 
-    @train_model =<<-EOF
+    @train_model ||=<<-EOF
 rbbt.require('e1071');
 model = svm(as.factor(label) ~ ., data = features);
     EOF
  
-    @eval_model =<<-EOF
+    @eval_model ||=<<-EOF
 rbbt.require('e1071');
 label = predict(model, features);
     EOF

data/lib/rbbt/vector/model.rb

@@ -53,6 +53,13 @@ features = cbind(features, label = labels);
     "features[['#{name}']] = factor(features[['#{name}']], levels=#{R.ruby2R levels})"
   end * "\n" if factor_levels }
 #{code}
+# Save used factor levels
+factor_levels = c()
+for (c in names(features)){
+  if (is.factor(features[[c]]))
+    factor_levels[c] = paste(levels(features[[c]]), collapse="\t")
+}
+rbbt.tsv.write("#{model_file}.factor_levels", factor_levels, names=c('Levels'), type='flat')
 save(model, file='#{model_file}')
       EOF
     end
@@ -150,6 +157,9 @@ cat(paste(label, sep="\\n", collapse="\\n"));
       if File.exists?(@levels_file)
         @factor_levels = YAML.load(Open.read(@levels_file))
       end
+      if File.exists?(@model_file + '.factor_levels')
+        @factor_levels = TSV.open(@model_file + '.factor_levels')
+      end
     else
       @factor_levels = factor_levels 
     end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: rbbt-dm
 version: !ruby/object:Gem::Version
-  version: 1.1.57
+  version: 1.1.58
 platform: ruby
 authors:
 - Miguel Vazquez
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2022-05-11 00:00:00.000000000 Z
+date: 2022-07-19 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rbbt-util