2 files changed, 52 insertions, 3 deletions
diff --git a/cv/data/utils/count_features.rb b/cv/data/utils/count_features.rb
new file mode 100644
index 0000000..1e272ac
--- /dev/null
+++ b/cv/data/utils/count_features.rb
@@ -0,0 +1,48 @@
+require 'rubygems'
+require 'opentox-ruby'
+require 'yaml'
+
+@subjectid = nil
+
+
+
+def count_features(ds_uri)
+  puts ds_uri
+
+  dataset = OpenTox::Dataset.find(ds_uri, @subjectid)
+
+  features = dataset.features.keys
+  puts "# all features: #{features.size}"
+  
+  delete_features = []
+  features.each{ |fn|
+    dataset.features[fn][RDF.type].each { |typestr|
+      if typestr.include? "MissingFeature"
+        delete_features << fn 
+        @missing_features << dataset.features[fn][DC.title]
+      end
+    }
+  }
+  puts "# Missingfeatures: #{delete_features.size}"
+  features = features - delete_features
+  puts "# numeric features: #{features.size}"
+  puts "-----"
+end
+
+
+@missing_features = []
+
+ds = YAML::load_file("../datasets.yaml")
+ds.keys.each { |dataset|
+  puts "----------"
+  puts dataset
+  ds[dataset].keys.each { |pc|
+    puts pc unless (pc == "dataset") || (pc == "test") || (pc == "training")
+    count_features(ds[dataset][pc]) unless (pc == "dataset") || (pc == "test") || (pc == "training") 
+  }
+  puts "----------"
+  puts
+}
+puts
+puts "Missing features over all datasets:"
+puts @missing_features.uniq!.to_yaml 
diff --git a/cv/data/utils/create_selected_feature_ds.rb b/cv/data/utils/create_selected_feature_ds.rb
index 0e5f063..c30a23e 100644
--- a/cv/data/utils/create_selected_feature_ds.rb
+++ b/cv/data/utils/create_selected_feature_ds.rb
@@ -35,10 +35,11 @@ ds = YAML::load_file("../datasets.yaml")
 ds.keys.each { |dataset|
   puts "----------------- next dataset -----------------"
   ds[dataset].keys.each { |pc|
-    puts pc unless pc == "dataset"
-    [false, true].each { |del_missing|
+    puts pc unless (pc == "dataset") || (pc == "test") || (pc == "training")
+    #[false, true].each { |del_missing| 
+    [false].each { |del_missing| #false is default 
       begin
-        create_f_ds(ds[dataset]["dataset"], ds[dataset][pc], del_missing) unless pc == "dataset"
+        create_f_ds(ds[dataset]["dataset"], ds[dataset][pc], del_missing) unless (pc == "dataset") || (pc == "test") || (pc == "training")
       rescue
       end
     }