From d9c9d78e49d886ea91386adbbd2b523347df226e Mon Sep 17 00:00:00 2001
From: "helma@in-silico.ch" <helma@in-silico.ch>
Date: Mon, 29 Oct 2018 20:34:39 +0100
Subject: dataset predictions fixed

---
 lib/dataset.rb               | 23 +++++++++++++++++++++++
 lib/feature.rb               | 12 ++++++++++++
 lib/model.rb                 |  3 ++-
 lib/validation-statistics.rb |  3 +--
 4 files changed, 38 insertions(+), 3 deletions(-)

(limited to 'lib')

diff --git a/lib/dataset.rb b/lib/dataset.rb
index 41d7b5c..78f5633 100644
--- a/lib/dataset.rb
+++ b/lib/dataset.rb
@@ -86,6 +86,10 @@ module OpenTox
       features.select{|f| f._type.match("SubstanceProperty")}
     end
 
+    def prediction_features
+      features.select{|f| f._type.match("Prediction")}
+    end
+
     # Writers
 
     # Add a value for a given substance and feature
@@ -352,6 +356,25 @@ module OpenTox
       sdf
     end
 
+    def predictions
+      predictions = {}
+      substances.each do |s| 
+        predictions[s] ||= {}
+        prediction_feature = prediction_features.first
+        predictions[s][:value] = values(s,prediction_feature).first
+        predictions[s][:warnings] = []
+        warnings_features.each { |w| predictions[s][:warnings] += values(s,w) }
+        if predictions[s][:value] and prediction_feature.is_a? NominalLazarPrediction
+          prediction_feature.accept_values.each do |v|
+            f = LazarPredictionProbability.find_by(:name => v, :model_id => prediction_feature.model_id, :training_feature_id => prediction_feature.training_feature_id)
+            predictions[s][:probabilities] ||= {}
+            predictions[s][:probabilities][v] = values(s,f).first
+          end
+        end
+      end
+      predictions
+    end
+
     # Dataset operations
 
     # Merge an array of datasets 
diff --git a/lib/feature.rb b/lib/feature.rb
index be07e7a..c18b0b8 100644
--- a/lib/feature.rb
+++ b/lib/feature.rb
@@ -54,18 +54,30 @@ module OpenTox
   class NominalLazarPrediction < NominalFeature
     field :model_id, type: BSON::ObjectId
     field :training_feature_id, type: BSON::ObjectId
+    def name
+      "#{self[:name]} Prediction"
+    end
   end
 
   class LazarPredictionProbability < NominalLazarPrediction
+    def name
+      "probability(#{self[:name]})"
+    end
   end
 
   # Numeric lazar prediction
   class NumericLazarPrediction < NumericFeature
     field :model_id, type: BSON::ObjectId
     field :training_feature_id, type: BSON::ObjectId
+    def name
+      "#{name} Prediction"
+    end
   end
 
   class LazarPredictionInterval < NumericLazarPrediction
+    def name
+      "prediction_interval_#{self[:name]}"
+    end
   end
 
   class NominalSubstanceProperty < NominalFeature
diff --git a/lib/model.rb b/lib/model.rb
index fc98e09..7eaa469 100644
--- a/lib/model.rb
+++ b/lib/model.rb
@@ -346,7 +346,8 @@ module OpenTox
 
           # add predictions to dataset
           predictions.each do |substance_id,p|
-            d.add substance_id,warning_feature,p[:warnings].join(" ") if p[:warnings]
+            substance_id = BSON::ObjectId.from_string(substance_id)
+            d.add substance_id,warning_feature,p[:warnings].join(" ") unless p[:warnings].empty?
             unless p[:value].nil?
               d.add substance_id,f,p[:value]
               p[:probabilities].each {|name,p| d.add substance_id,probability_features[name],p} if p[:probabilities]
diff --git a/lib/validation-statistics.rb b/lib/validation-statistics.rb
index e440731..7bae891 100644
--- a/lib/validation-statistics.rb
+++ b/lib/validation-statistics.rb
@@ -150,8 +150,7 @@ module OpenTox
         y = {:all => [],:without_warnings => []}
         self.nr_predictions = {:all =>0,:without_warnings => 0}
         predictions.each do |cid,pred|
-          p pred
-          if pred[:value] and pred[:measurements] 
+          !if pred[:value] and pred[:measurements] and !pred[:measurements].empty?
             self.nr_predictions[:all] +=1
             x[:all] << pred[:measurements].median
             y[:all] << pred[:value]
-- 
cgit v1.2.3