Throw error on incorrect Label name in InferColumns API (#47)

* Added sequential grouping of columns * reverted the file * addded infer columns label name checking * added column detection error * removed unsed usings * added quotes * replace Where with Any clause * replace Where with Any clause
dotnet · Jan 30, 2019 · d254f4e · d254f4e
1 parent bd900ff
commit d254f4e
Show file tree

Hide file tree

Showing 2 changed files with 12 additions and 0 deletions.
diff --git a/src/AutoML/ColumnInference/ColumnInferenceApi.cs b/src/AutoML/ColumnInference/ColumnInferenceApi.cs
@@ -16,6 +16,10 @@ public static ColumnInferenceResult InferColumns(MLContext context, string path,
             var splitInference = InferSplit(sample, separatorChar, allowQuotedStrings, supportSparse);
             var typeInference = InferColumnTypes(context, sample, splitInference, hasHeader);
             var loaderColumns = ColumnTypeInference.GenerateLoaderColumns(typeInference.Columns);
+            if (!loaderColumns.Any(t => label.Equals(t.Name)))
+            {
+                throw new InferenceException(InferenceType.Label, $"Specified Label Column '{label}' was not found.");
+            }
             var typedLoaderArgs = new TextLoader.Arguments
             {
                 Column = loaderColumns,

diff --git a/src/Test/ColumnInferenceTests.cs b/src/Test/ColumnInferenceTests.cs
@@ -20,5 +20,13 @@ public void UnGroupColumnsTest()
             var columnInferenceWithGrouping = context.Data.InferColumns(dataPath, DatasetUtil.UciAdultLabel, true, groupColumns: true);
             Assert.IsTrue(columnInferenceWithGrouping.Columns.Count() < columnInferenceWithoutGrouping.Columns.Count());
         }
+
+        [TestMethod]
+        public void IncorrectLabelColumnTest()
+        {
+            var dataPath = DatasetUtil.DownloadUciAdultDataset();
+            var context = new MLContext();
+            Assert.ThrowsException<InferenceException>(new System.Action(() => context.Data.InferColumns(dataPath, "Junk", true, groupColumns: false)));
+        }
     }
 }