rev InferColumns to accept ColumnInfo input param (dotnet#186)

Author: daholste

src/Microsoft.ML.Auto/API/AutoInferenceCatalog.cs

@@ -2,6 +2,8 @@
 // The .NET Foundation licenses this file to you under the MIT license.
 // See the LICENSE file in the project root for more information.
 
+using Microsoft.ML.Data;
+
 namespace Microsoft.ML.Auto
 {
     public class AutoInferenceCatalog
@@ -52,11 +54,18 @@ public MulticlassClassificationExperiment CreateMulticlassClassificationExperime
             return new MulticlassClassificationExperiment(_context, experimentSettings);
         }
 
-        public ColumnInferenceResults InferColumns(string path, string label,char? separatorChar = null, bool? allowQuotedStrings = null, 
+        public ColumnInferenceResults InferColumns(string path, string labelColumn = DefaultColumnNames.Label, char? separatorChar = null, bool? allowQuotedStrings = null, 
+            bool? supportSparse = null, bool trimWhitespace = false, bool groupColumns = true)
+        {
+            //UserInputValidationUtil.ValidateInferColumnsArgs(path, label);
+            return ColumnInferenceApi.InferColumns(_context, path, labelColumn, separatorChar, allowQuotedStrings, supportSparse, trimWhitespace, groupColumns);
+        }
+
+        public ColumnInferenceResults InferColumns(string path, ColumnInformation columnInformation, char? separatorChar = null, bool? allowQuotedStrings = null,
             bool? supportSparse = null, bool trimWhitespace = false, bool groupColumns = true)
         {
             //UserInputValidationUtil.ValidateInferColumnsArgs(path, label);
-            return ColumnInferenceApi.InferColumns(_context, path, label, separatorChar, allowQuotedStrings, supportSparse, trimWhitespace, groupColumns);
+            return ColumnInferenceApi.InferColumns(_context, path, columnInformation, separatorChar, allowQuotedStrings, supportSparse, trimWhitespace, groupColumns);
         }
 
         public ColumnInferenceResults InferColumns(string path, uint labelColumnIndex, bool hasHeader = false, char? separatorChar = null, 

src/Microsoft.ML.Auto/ColumnInference/ColumnInferenceApi.cs

@@ -24,20 +24,28 @@ public static ColumnInferenceResults InferColumns(MLContext context, string path
                 typeInference.Columns[labelColumnIndex].SuggestedName = DefaultColumnNames.Label;
             }
 
-            return InferColumns(context, path, typeInference.Columns[labelColumnIndex].SuggestedName,
-                hasHeader, splitInference, typeInference, trimWhitespace, groupColumns);
+            var columnInfo = new ColumnInformation() { LabelColumn = typeInference.Columns[labelColumnIndex].SuggestedName };
+
+            return InferColumns(context, path, columnInfo, hasHeader, splitInference, typeInference, trimWhitespace, groupColumns);
+        }
+
+        public static ColumnInferenceResults InferColumns(MLContext context, string path, string labelColumn,
+            char? separatorChar, bool? allowQuotedStrings, bool? supportSparse, bool trimWhitespace, bool groupColumns)
+        {
+            var columnInfo = new ColumnInformation() { LabelColumn = labelColumn };
+            return InferColumns(context, path, columnInfo, separatorChar, allowQuotedStrings, supportSparse, trimWhitespace, groupColumns);
         }
 
-        public static ColumnInferenceResults InferColumns(MLContext context, string path, string label,
+        public static ColumnInferenceResults InferColumns(MLContext context, string path, ColumnInformation columnInfo,
             char? separatorChar, bool? allowQuotedStrings, bool? supportSparse, bool trimWhitespace, bool groupColumns)
         {
             var sample = TextFileSample.CreateFromFullFile(path);
             var splitInference = InferSplit(context, sample, separatorChar, allowQuotedStrings, supportSparse);
-            var typeInference = InferColumnTypes(context, sample, splitInference, true, null, label);
-            return InferColumns(context, path, label, true, splitInference, typeInference, trimWhitespace, groupColumns);
+            var typeInference = InferColumnTypes(context, sample, splitInference, true, null, columnInfo.LabelColumn);
+            return InferColumns(context, path, columnInfo, true, splitInference, typeInference, trimWhitespace, groupColumns);
         }
 
-        public static ColumnInferenceResults InferColumns(MLContext context, string path, string label, bool hasHeader,
+        public static ColumnInferenceResults InferColumns(MLContext context, string path, ColumnInformation columnInfo, bool hasHeader,
             TextFileContents.ColumnSplitResult splitInference, ColumnTypeInference.InferenceResult typeInference,
             bool trimWhitespace, bool groupColumns)
         {
@@ -54,7 +62,6 @@ public static ColumnInferenceResults InferColumns(MLContext context, string path
             var textLoader = context.Data.CreateTextLoader(typedLoaderArgs);
             var dataView = textLoader.Read(path);
 
-            var columnInfo = new ColumnInformation() { LabelColumn = label };
             var purposeInferenceResult = PurposeInference.InferPurposes(context, dataView, columnInfo);
 
             // start building result objects

src/Test/ColumnInferenceTests.cs

@@ -112,5 +112,18 @@ public void DefaultColumnNamesInferredCorrectly()
             Assert.AreEqual(DefaultColumnNames.GroupId, result.ColumnInformation.GroupIdColumn);
             Assert.AreEqual(result.ColumnInformation.NumericColumns.Count(), 3);
         }
+
+        [TestMethod]
+        public void InferColumnsColumnInfoParam()
+        {
+            var columnInfo = new ColumnInformation() { LabelColumn = DatasetUtil.MlNetGeneratedRegressionLabel };
+            var result = new MLContext().AutoInference().InferColumns(DatasetUtil.DownloadMlNetGeneratedRegressionDataset(), 
+                columnInfo);
+            var labelCol = result.TextLoaderArgs.Column.First(c => c.Name == DatasetUtil.MlNetGeneratedRegressionLabel);
+            Assert.AreEqual(DataKind.R4, labelCol.Type);
+            Assert.AreEqual(DatasetUtil.MlNetGeneratedRegressionLabel, result.ColumnInformation.LabelColumn);
+            Assert.AreEqual(1, result.ColumnInformation.NumericColumns.Count());
+            Assert.AreEqual(DefaultColumnNames.Features, result.ColumnInformation.NumericColumns.First());
+        }
     }
 }