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, 8 insertions, 2 deletions

diff --git a/test/classification-model.rb b/test/classification-model.rb
index 79ccb98..c41b211 100644
--- a/test/classification-model.rb
+++ b/test/classification-model.rb
@@ -84,13 +84,19 @@ class ClassificationModelTest < MiniTest::Test
     assert_kind_of Dataset, result
     assert_equal 7, result.features.size
     assert_equal 85, result.compounds.size
-    prediction_feature = result.prediction_features.first
+    prediction_feature = result.prediction_feature
     assert_equal ["carcinogenic"], result.values(result.compounds[1], prediction_feature)
     assert_equal ["non-carcinogenic"], result.values(result.compounds[5], prediction_feature)
     assert_nil result.predictions[result.compounds.first][:value]
     assert_equal "carcinogenic", result.predictions[result.compounds[1]][:value]
     assert_equal 0.27, result.predictions[result.compounds[1]][:probabilities]["non-carcinogenic"].round(2)
-    assert_match /High/i, result.predictions[result.compounds[1]][:confidence]
+    assert_match /Similar/i, result.predictions[result.compounds[1]][:confidence]
+    csv = result.to_prediction_csv
+    rows = csv.split("\n")
+    assert_equal "ID,Original SMILES,Canonical SMILES,Prediction,Confidence,Probability: carcinogenic,Probability: non-carcinogenic,Measurements", rows[0]
+    items = rows[2].split(",")
+    assert_equal "carcinogenic", items[3]
+    assert_equal 0.27, items[6].to_f.round(2) # probabilities
   end
 
   def test_carcinogenicity_rf_classification