Introduce SimpleColumnInfo class (#2641)

yaeldMS · web-flow · commit b0a170153ca2 · 2019-02-20T14:33:28.000-08:00
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/Normalizer.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/Normalizer.cs
@@ -66,7 +66,7 @@ public static void Example()
 
             // Composing a different pipeline if we wanted to normalize more than one column at a time. 
             // Using log scale as the normalization mode. 
-            var multiColPipeline = ml.Transforms.Normalize(NormalizingEstimator.NormalizerMode.LogMeanVariance, new[] { ("LogInduced", "Induced"), ("LogSpontaneous", "Spontaneous") });
+            var multiColPipeline = ml.Transforms.Normalize(NormalizingEstimator.NormalizerMode.LogMeanVariance, new SimpleColumnInfo[] { ("LogInduced", "Induced"), ("LogSpontaneous", "Spontaneous") });
             // The transformed data.
             var multiColtransformer = multiColPipeline.Fit(trainData);
             var multiColtransformedData = multiColtransformer.Transform(trainData);
diff --git a/docs/samples/Microsoft.ML.Samples/Dynamic/TensorFlow/TextClassification.cs b/docs/samples/Microsoft.ML.Samples/Dynamic/TensorFlow/TextClassification.cs
@@ -70,7 +70,7 @@ public static void Example()
             };
 
             var engine = mlContext.Transforms.Text.TokenizeWords("TokenizedWords", "Sentiment_Text")
-                .Append(mlContext.Transforms.Conversion.ValueMap(lookupMap, "Words", "Ids", new[] { ("VariableLenghtFeatures", "TokenizedWords") }))
+                .Append(mlContext.Transforms.Conversion.ValueMap(lookupMap, "Words", "Ids", new SimpleColumnInfo[] { ("VariableLenghtFeatures", "TokenizedWords") }))
                 .Append(mlContext.Transforms.CustomMapping(ResizeFeaturesAction, "Resize"))
                 .Append(mlContext.Transforms.ScoreTensorFlowModel(modelInfo, new[] { "Prediction/Softmax" }, new[] { "Features" }))
                 .Append(mlContext.Transforms.CopyColumns(("Prediction", "Prediction/Softmax")))
diff --git a/src/Microsoft.ML.Data/Transforms/ConversionsExtensionsCatalog.cs b/src/Microsoft.ML.Data/Transforms/ConversionsExtensionsCatalog.cs
@@ -86,8 +86,8 @@ public static KeyToValueMappingEstimator MapKeyToValue(this TransformsCatalog.Co
         ///  [!code-csharp[KeyToValueMappingEstimator](~/../docs/samples/docs/samples/Microsoft.ML.Samples/Dynamic/ValueMappingStringToKeyType.cs)]
         /// ]]></format>
         /// </example>
-        public static KeyToValueMappingEstimator MapKeyToValue(this TransformsCatalog.ConversionTransforms catalog, params (string outputColumnName, string inputColumnName)[] columns)
-             => new KeyToValueMappingEstimator(CatalogUtils.GetEnvironment(catalog), columns);
+        public static KeyToValueMappingEstimator MapKeyToValue(this TransformsCatalog.ConversionTransforms catalog, params SimpleColumnInfo[] columns)
+             => new KeyToValueMappingEstimator(CatalogUtils.GetEnvironment(catalog), SimpleColumnInfo.ConvertToValueTuples(columns));
 
         /// <summary>
         /// Convert the key types back to their original vectors.
@@ -175,8 +175,8 @@ public static ValueMappingEstimator<TInputType, TOutputType> ValueMap<TInputType
             this TransformsCatalog.ConversionTransforms catalog,
             IEnumerable<TInputType> keys,
             IEnumerable<TOutputType> values,
-            params (string outputColumnName, string inputColumnName)[] columns)
-            => new ValueMappingEstimator<TInputType, TOutputType>(CatalogUtils.GetEnvironment(catalog), keys, values, columns);
+            params SimpleColumnInfo[] columns)
+            => new ValueMappingEstimator<TInputType, TOutputType>(CatalogUtils.GetEnvironment(catalog), keys, values, SimpleColumnInfo.ConvertToValueTuples(columns));
 
         /// <summary>
         /// <see cref="ValueMappingEstimator"/>
@@ -201,8 +201,9 @@ public static ValueMappingEstimator<TInputType, TOutputType> ValueMap<TInputType
             IEnumerable<TInputType> keys,
             IEnumerable<TOutputType> values,
             bool treatValuesAsKeyType,
-            params (string outputColumnName, string inputColumnName)[] columns)
-            => new ValueMappingEstimator<TInputType, TOutputType>(CatalogUtils.GetEnvironment(catalog), keys, values, treatValuesAsKeyType, columns);
+            params SimpleColumnInfo[] columns)
+            => new ValueMappingEstimator<TInputType, TOutputType>(CatalogUtils.GetEnvironment(catalog), keys, values, treatValuesAsKeyType,
+                SimpleColumnInfo.ConvertToValueTuples(columns));
 
         /// <summary>
         /// <see cref="ValueMappingEstimator"/>
@@ -228,8 +229,9 @@ public static ValueMappingEstimator<TInputType, TOutputType> ValueMap<TInputType
             this TransformsCatalog.ConversionTransforms catalog,
             IEnumerable<TInputType> keys,
             IEnumerable<TOutputType[]> values,
-            params (string outputColumnName, string inputColumnName)[] columns)
-            => new ValueMappingEstimator<TInputType, TOutputType>(CatalogUtils.GetEnvironment(catalog), keys, values, columns);
+            params SimpleColumnInfo[] columns)
+            => new ValueMappingEstimator<TInputType, TOutputType>(CatalogUtils.GetEnvironment(catalog), keys, values,
+                SimpleColumnInfo.ConvertToValueTuples(columns));
 
         /// <summary>
         /// <see cref="ValueMappingEstimator"/>
@@ -251,7 +253,8 @@ public static ValueMappingEstimator<TInputType, TOutputType> ValueMap<TInputType
         /// </example>
         public static ValueMappingEstimator ValueMap(
             this TransformsCatalog.ConversionTransforms catalog,
-            IDataView lookupMap, string keyColumn, string valueColumn, params (string outputColumnName, string inputColumnName)[] columns)
-            => new ValueMappingEstimator(CatalogUtils.GetEnvironment(catalog), lookupMap, keyColumn, valueColumn, columns);
+            IDataView lookupMap, string keyColumn, string valueColumn, params SimpleColumnInfo[] columns)
+            => new ValueMappingEstimator(CatalogUtils.GetEnvironment(catalog), lookupMap, keyColumn, valueColumn,
+                SimpleColumnInfo.ConvertToValueTuples(columns));
     }
 }
diff --git a/src/Microsoft.ML.Data/Transforms/ExtensionsCatalog.cs b/src/Microsoft.ML.Data/Transforms/ExtensionsCatalog.cs
@@ -2,12 +2,36 @@
 // The .NET Foundation licenses this file to you under the MIT license.
 // See the LICENSE file in the project root for more information.
 
+using System.Linq;
 using Microsoft.Data.DataView;
 using Microsoft.ML.Data;
 using Microsoft.ML.Transforms;
 
 namespace Microsoft.ML
 {
+    public sealed class SimpleColumnInfo
+    {
+        private readonly string _outputColumnName;
+        private readonly string _inputColumnName;
+
+        public SimpleColumnInfo(string outputColumnName, string inputColumnName)
+        {
+            _outputColumnName = outputColumnName;
+            _inputColumnName = inputColumnName;
+        }
+
+        public static implicit operator SimpleColumnInfo((string outputColumnName, string inputColumnName) value)
+        {
+            return new SimpleColumnInfo(value.outputColumnName, value.inputColumnName);
+        }
+
+        [BestFriend]
+        internal static (string outputColumnName, string inputColumnName)[] ConvertToValueTuples(SimpleColumnInfo[] infos)
+        {
+            return infos.Select(info => (info._outputColumnName, info._inputColumnName)).ToArray();
+        }
+    }
+
     /// <summary>
     /// Extension methods for the <see cref="TransformsCatalog"/>.
     /// </summary>
@@ -42,8 +66,8 @@ public static ColumnCopyingEstimator CopyColumns(this TransformsCatalog catalog,
         /// ]]>
         /// </format>
         /// </example>
-        public static ColumnCopyingEstimator CopyColumns(this TransformsCatalog catalog, params (string outputColumnName, string inputColumnName)[] columns)
-            => new ColumnCopyingEstimator(CatalogUtils.GetEnvironment(catalog), columns);
+        public static ColumnCopyingEstimator CopyColumns(this TransformsCatalog catalog, params SimpleColumnInfo[] columns)
+            => new ColumnCopyingEstimator(CatalogUtils.GetEnvironment(catalog), SimpleColumnInfo.ConvertToValueTuples(columns));
 
         /// <summary>
         /// Concatenates columns together.
diff --git a/src/Microsoft.ML.Data/Transforms/NormalizerCatalog.cs b/src/Microsoft.ML.Data/Transforms/NormalizerCatalog.cs
@@ -46,8 +46,8 @@ public static NormalizingEstimator Normalize(this TransformsCatalog catalog,
         /// </example>
         public static NormalizingEstimator Normalize(this TransformsCatalog catalog,
             NormalizingEstimator.NormalizerMode mode,
-            params (string outputColumnName, string inputColumnName)[] columns)
-            => new NormalizingEstimator(CatalogUtils.GetEnvironment(catalog), mode, columns);
+            params SimpleColumnInfo[] columns)
+            => new NormalizingEstimator(CatalogUtils.GetEnvironment(catalog), mode, SimpleColumnInfo.ConvertToValueTuples(columns));
 
         /// <summary>
         /// Normalize (rescale) columns according to specified custom parameters.
diff --git a/src/Microsoft.ML.ImageAnalytics/ExtensionsCatalog.cs b/src/Microsoft.ML.ImageAnalytics/ExtensionsCatalog.cs
@@ -18,8 +18,8 @@ public static class ImageEstimatorsCatalog
         ///  [!code-csharp[ConvertToGrayscale](~/../docs/samples/docs/samples/Microsoft.ML.Samples/Dynamic/ImageAnalytics/ConvertToGrayscale.cs)]
         /// ]]></format>
         /// </example>
-        public static ImageGrayscalingEstimator ConvertToGrayscale(this TransformsCatalog catalog, params (string outputColumnName, string inputColumnName)[] columnPairs)
-            => new ImageGrayscalingEstimator(CatalogUtils.GetEnvironment(catalog), columnPairs);
+        public static ImageGrayscalingEstimator ConvertToGrayscale(this TransformsCatalog catalog, params SimpleColumnInfo[] columnPairs)
+            => new ImageGrayscalingEstimator(CatalogUtils.GetEnvironment(catalog), SimpleColumnInfo.ConvertToValueTuples(columnPairs));
 
         /// <summary>
         /// Loads the images from the <see cref="ImageLoadingTransformer.ImageFolder" /> into memory.
@@ -41,8 +41,8 @@ public static ImageGrayscalingEstimator ConvertToGrayscale(this TransformsCatalo
         ///  [!code-csharp[LoadImages](~/../docs/samples/docs/samples/Microsoft.ML.Samples/Dynamic/ImageAnalytics/LoadImages.cs)]
         /// ]]></format>
         /// </example>
-        public static ImageLoadingEstimator LoadImages(this TransformsCatalog catalog, string imageFolder, params (string outputColumnName, string inputColumnName)[] columnPairs)
-           => new ImageLoadingEstimator(CatalogUtils.GetEnvironment(catalog), imageFolder, columnPairs);
+        public static ImageLoadingEstimator LoadImages(this TransformsCatalog catalog, string imageFolder, params SimpleColumnInfo[] columnPairs)
+           => new ImageLoadingEstimator(CatalogUtils.GetEnvironment(catalog), imageFolder, SimpleColumnInfo.ConvertToValueTuples(columnPairs));
 
         /// <include file='doc.xml' path='doc/members/member[@name="ImagePixelExtractingEstimator"]/*' />
         /// <param name="catalog"> The transform's catalog.</param>
diff --git a/src/Microsoft.ML.Transforms/ConversionsCatalog.cs b/src/Microsoft.ML.Transforms/ConversionsCatalog.cs
@@ -19,8 +19,8 @@ public static class ConversionsCatalog
         /// <param name="catalog">The categorical transform's catalog.</param>
         /// <param name="columns">The input column.</param>
         public static KeyToBinaryVectorMappingEstimator MapKeyToBinaryVector(this TransformsCatalog.ConversionTransforms catalog,
-            params (string outputColumnName, string inputColumnName)[] columns)
-            => new KeyToBinaryVectorMappingEstimator(CatalogUtils.GetEnvironment(catalog), columns);
+            params SimpleColumnInfo[] columns)
+            => new KeyToBinaryVectorMappingEstimator(CatalogUtils.GetEnvironment(catalog), SimpleColumnInfo.ConvertToValueTuples(columns));
 
         /// <summary>
         ///  Convert the key types back to binary vector.
diff --git a/src/Microsoft.ML.Transforms/ExtensionsCatalog.cs b/src/Microsoft.ML.Transforms/ExtensionsCatalog.cs
@@ -16,8 +16,8 @@ public static class ExtensionsCatalog
         /// <param name="catalog">The transform extensions' catalog.</param>
         /// <param name="columns">The names of the input columns of the transformation and the corresponding names for the output columns.</param>
         public static MissingValueIndicatorEstimator IndicateMissingValues(this TransformsCatalog catalog,
-            params (string outputColumnName, string inputColumnName)[] columns)
-            => new MissingValueIndicatorEstimator(CatalogUtils.GetEnvironment(catalog), columns);
+            params SimpleColumnInfo[] columns)
+            => new MissingValueIndicatorEstimator(CatalogUtils.GetEnvironment(catalog), SimpleColumnInfo.ConvertToValueTuples(columns));
 
         /// <summary>
         /// Creates a new output column, or replaces the source with a new column
diff --git a/src/Microsoft.ML.Transforms/FeatureSelectionCatalog.cs b/src/Microsoft.ML.Transforms/FeatureSelectionCatalog.cs
@@ -29,8 +29,9 @@ public static MutualInformationFeatureSelectingEstimator SelectFeaturesBasedOnMu
             string labelColumn = MutualInfoSelectDefaults.LabelColumn,
             int slotsInOutput = MutualInfoSelectDefaults.SlotsInOutput,
             int numBins = MutualInfoSelectDefaults.NumBins,
-            params (string outputColumnName, string inputColumnName)[] columns)
-            => new MutualInformationFeatureSelectingEstimator(CatalogUtils.GetEnvironment(catalog), labelColumn, slotsInOutput, numBins, columns);
+            params SimpleColumnInfo[] columns)
+            => new MutualInformationFeatureSelectingEstimator(CatalogUtils.GetEnvironment(catalog), labelColumn, slotsInOutput, numBins,
+                SimpleColumnInfo.ConvertToValueTuples(columns));
 
         /// <include file='doc.xml' path='doc/members/member[@name="MutualInformationFeatureSelection"]/*' />
         /// <param name="catalog">The transform's catalog.</param>
diff --git a/src/Microsoft.ML.Transforms/Text/TextCatalog.cs b/src/Microsoft.ML.Transforms/Text/TextCatalog.cs
@@ -73,8 +73,8 @@ public static TokenizingByCharactersEstimator TokenizeCharacters(this Transforms
 
         public static TokenizingByCharactersEstimator TokenizeCharacters(this TransformsCatalog.TextTransforms catalog,
             bool useMarkerCharacters = CharTokenizingDefaults.UseMarkerCharacters,
-            params (string outputColumnName, string inputColumnName)[] columns)
-            => new TokenizingByCharactersEstimator(Contracts.CheckRef(catalog, nameof(catalog)).GetEnvironment(), useMarkerCharacters, columns);
+            params SimpleColumnInfo[] columns)
+            => new TokenizingByCharactersEstimator(Contracts.CheckRef(catalog, nameof(catalog)).GetEnvironment(), useMarkerCharacters, SimpleColumnInfo.ConvertToValueTuples(columns));
 
         /// <summary>
         /// Normalizes incoming text in <paramref name="inputColumnName"/> by changing case, removing diacritical marks, punctuation marks and/or numbers
diff --git a/test/Microsoft.ML.Tests/ScenariosWithDirectInstantiation/TensorflowTests.cs b/test/Microsoft.ML.Tests/ScenariosWithDirectInstantiation/TensorflowTests.cs
@@ -1003,7 +1003,7 @@ public void TensorFlowSentimentClassificationTest()
             // Then this integer vector is retrieved from the pipeline and resized to fixed length.
             // The second pipeline 'tfEnginePipe' takes the resized integer vector and passes it to TensoFlow and gets the classification scores.
             var estimator = mlContext.Transforms.Text.TokenizeWords("TokenizedWords", "Sentiment_Text")
-                .Append(mlContext.Transforms.Conversion.ValueMap(lookupMap, "Words", "Ids", new[] { ("Features", "TokenizedWords") }));
+                .Append(mlContext.Transforms.Conversion.ValueMap(lookupMap, "Words", "Ids", new SimpleColumnInfo[] { ("Features", "TokenizedWords") }));
             var dataPipe = estimator.Fit(dataView)
                 .CreatePredictionEngine<TensorFlowSentiment, TensorFlowSentiment>(mlContext);
 
diff --git a/test/Microsoft.ML.Tests/Transformers/FeatureSelectionTests.cs b/test/Microsoft.ML.Tests/Transformers/FeatureSelectionTests.cs
@@ -184,9 +184,9 @@ public void MutualInformationSelectionWorkout()
 
             var est = ML.Transforms.FeatureSelection.SelectFeaturesBasedOnMutualInformation("FeatureSelect", "VectorFloat", slotsInOutput: 1, labelColumn: "Label")
                 .Append(ML.Transforms.FeatureSelection.SelectFeaturesBasedOnMutualInformation(labelColumn: "Label", slotsInOutput: 2, numBins: 100,
-                    columns: new[] {
-                        (name: "out1", source: "VectorFloat"),
-                        (name: "out2", source: "VectorDouble")
+                    columns: new SimpleColumnInfo[] {
+                        ("out1", "VectorFloat"),
+                        ("out2", "VectorDouble")
                     }));
             TestEstimatorCore(est, data);
 
diff --git a/test/Microsoft.ML.Tests/Transformers/NAIndicatorTests.cs b/test/Microsoft.ML.Tests/Transformers/NAIndicatorTests.cs
@@ -46,7 +46,7 @@ public void NAIndicatorWorkout()
             };
 
             var dataView = ML.Data.ReadFromEnumerable(data);
-            var pipe = ML.Transforms.IndicateMissingValues(new (string outputColumnName, string inputColumnName)[] { ("NAA", "A"), ("NAB", "B"), ("NAC", "C"), ("NAD", "D") });
+            var pipe = ML.Transforms.IndicateMissingValues(new SimpleColumnInfo[] { ("NAA", "A"), ("NAB", "B"), ("NAC", "C"), ("NAD", "D") });
             TestEstimatorCore(pipe, dataView);
             Done();
         }
@@ -69,7 +69,7 @@ public void TestOldSavingAndLoading()
             };
 
             var dataView = ML.Data.ReadFromEnumerable(data);
-            var pipe = ML.Transforms.IndicateMissingValues(new (string outputColumnName, string inputColumnName)[] { ("NAA", "A"), ("NAB", "B"), ("NAC", "C"), ("NAD", "D") });
+            var pipe = ML.Transforms.IndicateMissingValues(new SimpleColumnInfo[] { ("NAA", "A"), ("NAB", "B"), ("NAC", "C"), ("NAD", "D") });
             var result = pipe.Fit(dataView).Transform(dataView);
             var resultRoles = new RoleMappedData(result);
             using (var ms = new MemoryStream())
@@ -94,7 +94,7 @@ public void NAIndicatorFileOutput()
             var data = reader.Read(new MultiFileSource(dataPath)).AsDynamic;
             var wrongCollection = new[] { new TestClass() { A = 1, B = 3, C = new float[2] { 1, 2 }, D = new double[2] { 3, 4 } } };
             var invalidData = ML.Data.ReadFromEnumerable(wrongCollection);
-            var est = ML.Transforms.IndicateMissingValues(new (string outputColumnName, string inputColumnName)[] 
+            var est = ML.Transforms.IndicateMissingValues(new SimpleColumnInfo[] 
             {
                 ("A", "ScalarFloat"), ("B", "ScalarDouble"),
                 ("C", "VectorFloat"), ("D", "VectorDoulbe")
@@ -127,7 +127,7 @@ public void NAIndicatorMetadataTest()
 
             var dataView = ML.Data.ReadFromEnumerable(data);
             var pipe = ML.Transforms.Categorical.OneHotEncoding("CatA", "A");
-            var newpipe = pipe.Append(ML.Transforms.IndicateMissingValues(new (string name, string source)[] { ("NAA", "CatA") }));
+            var newpipe = pipe.Append(ML.Transforms.IndicateMissingValues(("NAA", "CatA")));
             var result = newpipe.Fit(dataView).Transform(dataView);
             Assert.True(result.Schema.TryGetColumnIndex("NAA", out var col));
             // Check that the column is normalized.
diff --git a/test/Microsoft.ML.Tests/Transformers/ValueMappingTests.cs b/test/Microsoft.ML.Tests/Transformers/ValueMappingTests.cs
@@ -512,7 +512,7 @@ public void ValueMappingWorkout()
             var values = new List<int>() { 1, 2, 3, 4 };
 
             // Workout on value mapping
-            var est = ML.Transforms.Conversion.ValueMap(keys, values, new[] { ("D", "A"), ("E", "B"), ("F", "C") });
+            var est = ML.Transforms.Conversion.ValueMap(keys, values, new SimpleColumnInfo[] { ("D", "A"), ("E", "B"), ("F", "C") });
             TestEstimatorCore(est, validFitInput: dataView, invalidInput: badDataView);
         }
 
@@ -531,7 +531,7 @@ public void ValueMappingValueTypeIsVectorWorkout()
                 new int[] {400, 500, 600, 700 }};
 
             // Workout on value mapping
-            var est = ML.Transforms.Conversion.ValueMap(keys, values, new[] { ("D", "A"), ("E", "B"), ("F", "C") });
+            var est = ML.Transforms.Conversion.ValueMap(keys, values, new SimpleColumnInfo[] { ("D", "A"), ("E", "B"), ("F", "C") });
             TestEstimatorCore(est, validFitInput: dataView, invalidInput: badDataView);
         }
 
@@ -548,7 +548,7 @@ public void ValueMappingInputIsVectorWorkout()
             var values = new List<int>() { 1, 2, 3, 4 };
 
             var est = ML.Transforms.Text.TokenizeWords("TokenizeB", "B")
-                .Append(ML.Transforms.Conversion.ValueMap(keys, values, new[] { ("VecB", "TokenizeB") }));
+                .Append(ML.Transforms.Conversion.ValueMap(keys, values, new SimpleColumnInfo[] { ("VecB", "TokenizeB") }));
             TestEstimatorCore(est, validFitInput: dataView, invalidInput: badDataView);
         }
 
