separate csv serialisations for batch predictions and training data, repeated measurements in mutagenicity dataset fixed, daphnia import fixed, CENTRAL_MONGO_IP removed

1 files changed, 15 insertions, 2 deletions

diff --git a/test/regression-model.rb b/test/regression-model.rb
index 7f667dc..3b41171 100644
--- a/test/regression-model.rb
+++ b/test/regression-model.rb
@@ -173,13 +173,26 @@ class LazarRegressionTest < MiniTest::Test
     model = Model::Lazar.create training_dataset: training_dataset
     result = model.predict training_dataset
     assert_kind_of Dataset, result
-    assert_equal 6, result.features.size
+    assert_equal 8, result.features.size
     assert_equal 88, result.compounds.size
     assert_equal [1.95], result.values(result.compounds.first, result.bioactivity_features[0]).collect{|v| v.round(2)}
     assert_equal [1.37], result.values(result.compounds[6], result.bioactivity_features[0]).collect{|v| v.round(2)}
-    assert_equal [1.79], result.values(result.compounds[6], result.prediction_features[0]).collect{|v| v.round(2)}
+    assert_equal [1.79], result.values(result.compounds[6], result.prediction_feature).collect{|v| v.round(2)}
     assert_equal [1.84,1.73], result.values(result.compounds[7], result.bioactivity_features[0]).collect{|v| v.round(2)}
     assert_match /Low/i, result.predictions[result.compounds[6]][:confidence]
+    csv = result.to_prediction_csv
+    rows = csv.split("\n")
+    assert_equal "ID,Original SMILES,Canonical SMILES,Prediction,Confidence,Lower prediction interval,Upper prediction interval,Measurements", rows[0]
+    items = rows[3].split(",")
+    # prediction and measurement within prediciton interval
+    prediction = items[3].to_f
+    pi_low = items[5].to_f
+    pi_hi = items[6].to_f
+    measurement = items[7].to_f
+    [prediction,measurement].each do |v|
+      assert(v > pi_low)
+      assert(v < pi_hi)
+    end
   end
 
 end