From 9245dbb13a6add795b66053bc4d2a0485dee496b Mon Sep 17 00:00:00 2001
From: davor <vorgrimmlerdavid@gmx.de>
Date: Wed, 7 Mar 2012 17:07:44 +0100
Subject: Added del_nominal switch to load_csv which removes all nominal
 features (del_nomnial is false by default).

---
 lib/parser.rb | 28 ++++++++++++++++++++++++++--
 lib/utils.rb  |  2 +-
 2 files changed, 27 insertions(+), 3 deletions(-)

diff --git a/lib/parser.rb b/lib/parser.rb
index 53765b1..f11ed48 100644
--- a/lib/parser.rb
+++ b/lib/parser.rb
@@ -351,12 +351,13 @@ module OpenTox
       # @param [String] csv CSV representation of the dataset
       # @param [Boolean] drop_missing Whether completely missing rows should be droppped
       # @param [Boolean] all_numeric Whether all features should be treated as numeric
+      # @param [Boolean] del_nominal All nominal features will be removed
       # @return [OpenTox::Dataset] Dataset object with CSV data
-      def load_csv(csv, drop_missing=false, all_numeric=false)
+      def load_csv(csv, drop_missing=false, all_numeric=false, del_nominal=false)
+        
         row = 0
         input = csv.split("\n")
         headers = split_row(input.shift)
-        add_features(headers)
         value_maps = Array.new
         regression_features=Array.new
 
@@ -371,6 +372,29 @@ module OpenTox
             end
           }
         }
+        
+        if del_nominal
+          # Collect all nominal features 
+          del_features = Array.new
+          regression_features.each_with_index {|value, i|
+            if value == false
+              del_features << i
+            end
+          }
+          del_features = del_features.reverse
+          # Delete nominal entries in each row
+          del_features.each{|f|
+            regression_features.delete_at(f)
+            headers.delete_at(f+1)
+            input.map!{|row|
+              new_row = split_row(row)
+              new_row.delete_at(f+1)
+              row = new_row.join(",")
+            }
+          }
+        end
+        
+        add_features(headers)
 
         input.each_with_index { |row, i| 
           drop=false
diff --git a/lib/utils.rb b/lib/utils.rb
index 40988db..c9dd32a 100644
--- a/lib/utils.rb
+++ b/lib/utils.rb
@@ -60,7 +60,7 @@ module OpenTox
         ds = OpenTox::Dataset.new
         ds.save
         parser.dataset = ds
-        ds = parser.load_csv(master.collect{|r| r.join(",")}.join("\n"),false,true)
+        ds = parser.load_csv(master.collect{|r| r.join(",")}.join("\n"),false,false,true)
         ds.save
 
       rescue Exception => e
-- 
cgit v1.2.3