Modified scripts.

2 files changed, 50 insertions, 19 deletions

diff --git a/cv/data/utils/count_features.rb b/cv/data/utils/count_features.rb
index 1e272ac..c8a5893 100644
--- a/cv/data/utils/count_features.rb
+++ b/cv/data/utils/count_features.rb
@@ -4,7 +4,11 @@ require 'yaml'
 
 @subjectid = nil
 
-
+if ARGV.size != 1
+  puts "Args: path/to/dataset.yaml"
+  puts ARGV.size
+  exit
+end
 
 def count_features(ds_uri)
   puts ds_uri
@@ -13,9 +17,10 @@ def count_features(ds_uri)
 
   features = dataset.features.keys
   puts "# all features: #{features.size}"
-  
+  feature_names = []
   delete_features = []
   features.each{ |fn|
+    feature_names << fn.split("\/feature\/").last
     dataset.features[fn][RDF.type].each { |typestr|
       if typestr.include? "MissingFeature"
         delete_features << fn 
@@ -23,6 +28,9 @@ def count_features(ds_uri)
       end
     }
   }
+  @all_feature_names << feature_names.sort
+  @all_feature_names << "" 
+
   puts "# Missingfeatures: #{delete_features.size}"
   features = features - delete_features
   puts "# numeric features: #{features.size}"
@@ -31,18 +39,30 @@ end
 
 
 @missing_features = []
-
-ds = YAML::load_file("../datasets.yaml")
+@all_feature_names = []
+path = ARGV[0]
+puts path
+ds = YAML::load_file("#{path}")
+#ds = YAML::load_file("../datasets.yaml")
 ds.keys.each { |dataset|
   puts "----------"
   puts dataset
+  @all_feature_names << "" 
+  @all_feature_names << "------ new dataset ------" 
+  @all_feature_names << "-------- #{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") 
+    if !(pc == "dataset") || (pc == "test") || (pc == "training")
+      puts pc
+      @all_feature_names << "--- new feature: #{pc} ---" 
+      count_features(ds[dataset][pc])
+    end
   }
   puts "----------"
   puts
 }
 puts
 puts "Missing features over all datasets:"
-puts @missing_features.uniq!.to_yaml 
+puts @missing_features.uniq!.to_yaml
+puts 
+puts "All feature names:"
+puts @all_feature_names
diff --git a/cv/data/utils/create_selected_feature_ds.rb b/cv/data/utils/create_selected_feature_ds.rb
index c30a23e..c7c3445 100644
--- a/cv/data/utils/create_selected_feature_ds.rb
+++ b/cv/data/utils/create_selected_feature_ds.rb
@@ -4,14 +4,14 @@ require 'yaml'
 
 @subjectid = nil
 
-
+@dataset = []
 
 def create_f_ds(t_ds_uri, f_ds_uri, del)
 
   regression_training_dataset = OpenTox::Dataset.find(t_ds_uri, @subjectid)
   prediction_feature = regression_training_dataset.features.keys.first
   regression_feature_dataset = OpenTox::Dataset.find(f_ds_uri, @subjectid)
-  
+
   params = {} 
   params[:dataset_uri] = regression_training_dataset.uri
   params[:prediction_feature_uri] = prediction_feature
@@ -20,12 +20,12 @@ def create_f_ds(t_ds_uri, f_ds_uri, del)
   puts params.to_yaml
   feature_selection_algo_uri = File.join(CONFIG[:services]["opentox-algorithm"],"feature_selection/rfe")
   puts feature_selection_algo_uri
+  puts "--- Feature dataset is: ---"
 
   result = OpenTox::RestClientWrapper.post( feature_selection_algo_uri, params)
-  puts "--- Feature dataset is: ---"
   puts result
-  
   puts
+  result
 end
 
 
@@ -34,15 +34,26 @@ end
 ds = YAML::load_file("../datasets.yaml")
 ds.keys.each { |dataset|
   puts "----------------- next dataset -----------------"
+  @dataset << "\"#{dataset}\": {" 
   ds[dataset].keys.each { |pc|
-    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") || (pc == "test") || (pc == "training")
-      rescue
-      end
-    }
+    if !((pc == "dataset") || (pc == "test") || (pc == "training") || (pc == "hybrid"))
+      puts pc
+      #[false, true].each { |del_missing| 
+      [false].each { |del_missing| #false is default 
+        begin
+          result = create_f_ds(ds[dataset]["dataset"], ds[dataset][pc], del_missing) 
+          @dataset << "  \"#{pc}\": \"#{result}\"," 
+        rescue
+        end
+      }
+    else
+      @dataset << "  \"#{pc}\": \"#{ds[dataset][pc]}\"," 
+
+    end
     puts "-----------------" unless pc == "dataset"
   }
+  @dataset << "}," 
 }
+
+puts @dataset
+