merge with opentox/development

1 files changed, 44 insertions, 30 deletions

diff --git a/lazar.rb b/lazar.rb
index 637b623..a45886c 100644
--- a/lazar.rb
+++ b/lazar.rb
@@ -35,17 +35,17 @@ post '/lazar/?' do
   halt 404, "Dataset #{dataset_uri} not found." unless training_activities = OpenTox::Dataset.new(dataset_uri)
   training_activities.load_all(@subjectid)
 
-  prediction_feature = params[:prediction_feature]
-  unless prediction_feature # try to read prediction_feature from dataset
+  prediction_feature = OpenTox::Feature.find(params[:prediction_feature],@subjectid)
+  unless params[:prediction_feature] # try to read prediction_feature from dataset
     halt 404, "#{training_activities.features.size} features in dataset #{dataset_uri}. Please provide a  prediction_feature parameter." unless training_activities.features.size == 1
-    prediction_feature = training_activities.features.keys.first
-    params[:prediction_feature] = prediction_feature
+    prediction_feature = OpenTox::Feature.find(training_activities.features.keys.first,@subjectid)
+    params[:prediction_feature] = prediction_feature.uri # pass to feature mining service
   end
 
   feature_generation_uri = @@feature_generation_default unless feature_generation_uri = params[:feature_generation_uri]
 
-	halt 404, "No feature #{prediction_feature} in dataset #{params[:dataset_uri]}. (features: "+
-    training_activities.features.inspect+")" unless training_activities.features and training_activities.features.include?(prediction_feature)
+	halt 404, "No feature #{prediction_feature.uri} in dataset #{params[:dataset_uri]}. (features: "+
+    training_activities.features.inspect+")" unless training_activities.features and training_activities.features.include?(prediction_feature.uri)
 
   task = OpenTox::Task.create("Create lazar model",url_for('/lazar',:full)) do |task|
 
@@ -77,18 +77,21 @@ post '/lazar/?' do
 		halt 404, "Dataset #{feature_dataset_uri} not found." if training_features.nil?
 
     # sorted features for index lookups
-    lazar.features = training_features.features.sort if training_features.feature_type(@subjectid) == "regression"
+
+    lazar.features = training_features.features.sort if prediction_feature.feature_type(@subjectid) == "regression" and lazar.feature_calculation_algorithm != "Substructure.match"
 
     training_features.data_entries.each do |compound,entry|
       lazar.fingerprints[compound] = [] unless lazar.fingerprints[compound]
       entry.keys.each do |feature|
-        if feature_generation_uri.match(/fminer/)
-          smarts = training_features.features[feature][OT.smarts]
-          lazar.fingerprints[compound] << smarts
-          unless lazar.features.include? smarts
-            lazar.features << smarts
-            lazar.p_values[smarts] = training_features.features[feature][OT.pValue]
-            lazar.effects[smarts] = training_features.features[feature][OT.effect]
+        if lazar.feature_calculation_algorithm == "Substructure.match"
+          if training_features.features[feature]
+            smarts = training_features.features[feature][OT.smarts]
+            lazar.fingerprints[compound] << smarts
+            unless lazar.features.include? smarts
+              lazar.features << smarts
+              lazar.p_values[smarts] = training_features.features[feature][OT.pValue]
+              lazar.effects[smarts] = training_features.features[feature][OT.effect]
+            end
           end
         else
           case training_features.feature_type(@subjectid)
@@ -112,38 +115,49 @@ post '/lazar/?' do
       end
     end
       
+    @training_classes = training_activities.feature_classes(prediction_feature.uri) if prediction_feature.feature_type == "classification"
+    lazar.prediction_algorithm = "Neighbors.local_svm_regression" if  prediction_feature.feature_type == "regression"
+
     training_activities.data_entries.each do |compound,entry| 
 			lazar.activities[compound] = [] unless lazar.activities[compound]
-      unless entry[params[:prediction_feature]].empty?
-        entry[params[:prediction_feature]].each do |value|
-          case value.to_s
-          when "true"
-            lazar.activities[compound] << true
-          when "false"
-            lazar.activities[compound] << false
-          else 
+      unless entry[prediction_feature.uri].empty?
+        entry[prediction_feature.uri].each do |value|
+          if prediction_feature.feature_type == "classification"
+            case value.to_s
+            when "true"
+              lazar.activities[compound] << true
+            when "false"
+              lazar.activities[compound] << false
+            when /#{@training_classes.last}/
+              lazar.activities[compound] << true
+            when /#{@training_classes.first}/
+              lazar.activities[compound] << false
+            else
+              LOGGER.warn "Unknown class \"#{value.to_s}\"."
+            end
+          elsif prediction_feature.feature_type == "regression"
             halt 404, "0 values not allowed in training dataset. log10 is calculated internally." if value.to_f == 0
             lazar.activities[compound] << value.to_f
-            lazar.prediction_algorithm = "Neighbors.local_svm_regression"
           end
         end
       end
     end
 
-    lazar.metadata[DC.title] = "lazar model for #{URI.decode(File.basename(prediction_feature))}"
+    lazar.metadata[DC.title] = "lazar model for #{URI.decode(File.basename(prediction_feature.uri))}"
     # TODO: fix dependentVariable
-    lazar.metadata[OT.dependentVariables] = params[:prediction_feature]
+    lazar.metadata[OT.dependentVariables] = prediction_feature.uri
     lazar.metadata[OT.trainingDataset] = dataset_uri
 		lazar.metadata[OT.featureDataset] = feature_dataset_uri
-    if training_activities.feature_type(@subjectid).to_s == "classification"
-      lazar.metadata[OT.isA] = OTA.ClassificationLazySingleTarget
-    elsif training_activities.feature_type(@subjectid).to_s == "regression"
-      lazar.metadata[OT.isA] = OTA.RegressionLazySingleTarget
+
+    if prediction_feature.feature_type(@subjectid) == "classification"
+      lazar.metadata[RDF.type] = [OTA.ClassificationLazySingleTarget]
+    elsif prediction_feature.feature_type(@subjectid) == "regression"
+      lazar.metadata[RDF.type] = [OTA.RegressionLazySingleTarget]
     end
 
     lazar.metadata[OT.parameters] = [
       {DC.title => "dataset_uri", OT.paramValue => dataset_uri},
-      {DC.title => "prediction_feature", OT.paramValue => prediction_feature},
+      {DC.title => "prediction_feature", OT.paramValue => prediction_feature.uri},
       {DC.title => "feature_generation_uri", OT.paramValue => feature_generation_uri}
     ]