From 4ebd80fee52c04bd36781f846eae60019918345d Mon Sep 17 00:00:00 2001
From: Christoph Helma <helma@in-silico.ch>
Date: Thu, 21 Apr 2016 14:29:23 +0200
Subject: initial classification probabilities

---
 lib/classification.rb           | 38 +++++++++++++++++++-------------------
 lib/crossvalidation.rb          |  2 +-
 lib/leave-one-out-validation.rb | 22 +++++++++++-----------
 test/setup.rb                   |  4 ++--
 4 files changed, 33 insertions(+), 33 deletions(-)

diff --git a/lib/classification.rb b/lib/classification.rb
index 0202940..b9b66f0 100644
--- a/lib/classification.rb
+++ b/lib/classification.rb
@@ -5,28 +5,28 @@ module OpenTox
 
       def self.weighted_majority_vote compound, params
         neighbors = params[:neighbors]
-        weighted_sum = {}
-        sim_sum = 0.0
-        confidence = 0.0
-        neighbors.each do |row|
-          sim = row["tanimoto"]
-          row["features"][params[:prediction_feature_id].to_s].each do |act|
-            weighted_sum[act] ||= 0
-            weighted_sum[act] += sim
+        feature_id = params[:prediction_feature_id].to_s
+        sims = {}
+        neighbors.each do |n|
+          sim = n["tanimoto"]
+          n["features"][feature_id].each do |act|
+            sims[act] ||= []
+            sims[act] << sim
+            #sims[act] << 0.5*sim+0.5 # scale to 1-0.5 
           end
         end
-        case weighted_sum.size
-        when 1
-          return {:value => weighted_sum.keys.first, :confidence => weighted_sum.values.first/neighbors.size.abs}
-        when 2
-          sim_sum = weighted_sum[weighted_sum.keys[0]]
-          sim_sum -= weighted_sum[weighted_sum.keys[1]]
-          sim_sum > 0 ? prediction = weighted_sum.keys[0] : prediction = weighted_sum.keys[1] 
-          confidence = (sim_sum/neighbors.size).abs 
-          return {:value => prediction,:confidence => confidence}
-        else
-          bad_request_error "Cannot predict more than 2 classes, multinomial classifications is not yet implemented. Received classes were: '#{weighted.sum.keys}'"
+        sim_all = sims.collect{|a,s| s}.flatten
+        sim_sum = sim_all.sum
+        sim_max = sim_all.max
+        probabilities = {}
+        sims.each do |a,s|
+          probabilities[a] = s.sum/sim_sum
         end
+        probabilities = probabilities.collect{|a,p| [a,sim_max*p]}.to_h
+        p_max = probabilities.collect{|a,p| p}.max
+        prediction = probabilities.key(p_max)
+        {:value => prediction,:probabilities => probabilities}
+
       end
     end
   end
diff --git a/lib/crossvalidation.rb b/lib/crossvalidation.rb
index 15dfb21..6ffeb25 100644
--- a/lib/crossvalidation.rb
+++ b/lib/crossvalidation.rb
@@ -52,7 +52,7 @@ module OpenTox
       cv.update_attributes(
         nr_instances: nr_instances,
         nr_unpredicted: nr_unpredicted,
-        predictions: predictions#.sort{|a,b| b[3] <=> a[3]} # sort according to confidence
+        predictions: predictions
       )
       $logger.debug "Nr unpredicted: #{nr_unpredicted}"
       cv.statistics
diff --git a/lib/leave-one-out-validation.rb b/lib/leave-one-out-validation.rb
index 2cd13db..0a131a4 100644
--- a/lib/leave-one-out-validation.rb
+++ b/lib/leave-one-out-validation.rb
@@ -51,18 +51,18 @@ module OpenTox
             if pred[:value] == db_act
               if pred[:value] == accept_values[0]
                 confusion_matrix[0][0] += 1
-                weighted_confusion_matrix[0][0] += pred[:confidence]
+                #weighted_confusion_matrix[0][0] += pred[:confidence]
               elsif pred[:value] == accept_values[1]
                 confusion_matrix[1][1] += 1
-                weighted_confusion_matrix[1][1] += pred[:confidence]
+                #weighted_confusion_matrix[1][1] += pred[:confidence]
               end
             else
               if pred[:value] == accept_values[0]
                 confusion_matrix[0][1] += 1
-                weighted_confusion_matrix[0][1] += pred[:confidence]
+                #weighted_confusion_matrix[0][1] += pred[:confidence]
               elsif pred[:value] == accept_values[1]
                 confusion_matrix[1][0] += 1
-                weighted_confusion_matrix[1][0] += pred[:confidence]
+                #weighted_confusion_matrix[1][0] += pred[:confidence]
               end
             end
           end
@@ -73,17 +73,17 @@ module OpenTox
         predictivity[v] = confusion_matrix[i][i]/confusion_matrix.collect{|n| n[i]}.reduce(:+).to_f
       end
       confidence_sum = 0
-      weighted_confusion_matrix.each do |r|
-        r.each do |c|
-          confidence_sum += c
-        end
-      end
+#      weighted_confusion_matrix.each do |r|
+#        r.each do |c|
+#          confidence_sum += c
+#        end
+#      end
       update_attributes(
         accept_values: accept_values,
         confusion_matrix: confusion_matrix,
-        weighted_confusion_matrix: weighted_confusion_matrix,
+#        weighted_confusion_matrix: weighted_confusion_matrix,
         accuracy: (confusion_matrix[0][0]+confusion_matrix[1][1])/(nr_instances-nr_unpredicted).to_f,
-        weighted_accuracy: (weighted_confusion_matrix[0][0]+weighted_confusion_matrix[1][1])/confidence_sum.to_f,
+#        weighted_accuracy: (weighted_confusion_matrix[0][0]+weighted_confusion_matrix[1][1])/confidence_sum.to_f,
         true_rate: true_rate,
         predictivity: predictivity,
         finished_at: Time.now
diff --git a/test/setup.rb b/test/setup.rb
index be3140a..e7c32b4 100644
--- a/test/setup.rb
+++ b/test/setup.rb
@@ -1,7 +1,7 @@
 ENV["LAZAR_ENV"] = "development"
 require 'minitest/autorun'
-#require_relative '../lib/lazar.rb'
-require 'lazar'
+require_relative '../lib/lazar.rb'
+#require 'lazar'
 include OpenTox
 TEST_DIR ||= File.expand_path(File.dirname(__FILE__))
 DATA_DIR ||= File.join(TEST_DIR,"data")
-- 
cgit v1.2.3