merging ML.ColumnOptions and ML.InputOutputColumnPair

artidoro · artidoro · commit 5272c0c5efce · 2019-03-22T15:54:41.000-07:00
diff --git a/docs/code/MlNetCookBook.md b/docs/code/MlNetCookBook.md
@@ -424,7 +424,7 @@ var pipeline =
     // Use the multi-class SDCA model to predict the label using features.
     .Append(mlContext.MulticlassClassification.Trainers.SdcaCalibrated())
     // Apply the inverse conversion from 'PredictedLabel' column back to string value.
-    .Append(mlContext.Transforms.Conversion.MapKeyToValue(("PredictedLabel", "Data")));
+    .Append(mlContext.Transforms.Conversion.MapKeyToValue("Data", "PredictedLabel"));
 
 // Train the model.
 var model = pipeline.Fit(trainData);
diff --git a/src/Microsoft.ML.Data/Transforms/ConversionsExtensionsCatalog.cs b/src/Microsoft.ML.Data/Transforms/ConversionsExtensionsCatalog.cs
@@ -111,22 +111,6 @@ public static KeyToValueMappingEstimator MapKeyToValue(this TransformsCatalog.Co
         public static KeyToValueMappingEstimator MapKeyToValue(this TransformsCatalog.ConversionTransforms catalog, InputOutputColumnPair[] columns)
             => new KeyToValueMappingEstimator(CatalogUtils.GetEnvironment(catalog), columns.Select(x => (x.OutputColumnName, x.InputColumnName)).ToArray());
 
-        /// <summary>
-        ///  Convert the key types (name of the column specified in the first item of the tuple) back to their original values
-        ///  (named as specified in the second item of the tuple).
-        /// </summary>
-        /// <param name="catalog">The conversion transform's catalog</param>
-        /// <param name="columns">The pairs of input and output columns.</param>
-        /// <example>
-        /// <format type="text/markdown">
-        /// <![CDATA[
-        ///  [!code-csharp[KeyToValueMappingEstimator](~/../docs/samples/docs/samples/Microsoft.ML.Samples/Dynamic/ValueMappingStringToKeyType.cs)]
-        /// ]]></format>
-        /// </example>
-        [BestFriend]
-        internal static KeyToValueMappingEstimator MapKeyToValue(this TransformsCatalog.ConversionTransforms catalog, params ColumnOptions[] columns)
-             => new KeyToValueMappingEstimator(CatalogUtils.GetEnvironment(catalog), ColumnOptions.ConvertToValueTuples(columns));
-
         /// <summary>
         /// Maps key types or key values into a floating point vector.
         /// </summary>
@@ -292,11 +276,11 @@ public static ValueMappingEstimator<TInputType, TOutputType> MapValue<TInputType
         internal static ValueMappingEstimator<TInputType, TOutputType> MapValue<TInputType, TOutputType>(
             this TransformsCatalog.ConversionTransforms catalog,
             IEnumerable<KeyValuePair<TInputType, TOutputType>> keyValuePairs,
-            params ColumnOptions[] columns)
+            params InputOutputColumnPair[] columns)
         {
             var keys = keyValuePairs.Select(pair => pair.Key);
             var values = keyValuePairs.Select(pair => pair.Value);
-            return new ValueMappingEstimator<TInputType, TOutputType>(CatalogUtils.GetEnvironment(catalog), keys, values, ColumnOptions.ConvertToValueTuples(columns));
+            return new ValueMappingEstimator<TInputType, TOutputType>(CatalogUtils.GetEnvironment(catalog), keys, values, InputOutputColumnPair.ConvertToValueTuples(columns));
         }
 
         /// <summary>
@@ -320,12 +304,12 @@ internal static ValueMappingEstimator<TInputType, TOutputType> MapValue<TInputTy
             this TransformsCatalog.ConversionTransforms catalog,
             IEnumerable<KeyValuePair<TInputType, TOutputType>> keyValuePairs,
             bool treatValuesAsKeyType,
-            params ColumnOptions[] columns)
+            params InputOutputColumnPair[] columns)
         {
             var keys = keyValuePairs.Select(pair => pair.Key);
             var values = keyValuePairs.Select(pair => pair.Value);
             return new ValueMappingEstimator<TInputType, TOutputType>(CatalogUtils.GetEnvironment(catalog), keys, values, treatValuesAsKeyType,
-                  ColumnOptions.ConvertToValueTuples(columns));
+                  InputOutputColumnPair.ConvertToValueTuples(columns));
         }
 
         /// <summary>
@@ -381,12 +365,12 @@ public static ValueMappingEstimator<TInputType, TOutputType> MapValue<TInputType
         internal static ValueMappingEstimator<TInputType, TOutputType> MapValue<TInputType, TOutputType>(
             this TransformsCatalog.ConversionTransforms catalog,
             IEnumerable<KeyValuePair<TInputType, TOutputType[]>> keyValuePairs,
-            params ColumnOptions[] columns)
+            params InputOutputColumnPair[] columns)
         {
             var keys = keyValuePairs.Select(pair => pair.Key);
             var values = keyValuePairs.Select(pair => pair.Value);
             return new ValueMappingEstimator<TInputType, TOutputType>(CatalogUtils.GetEnvironment(catalog), keys, values,
-                    ColumnOptions.ConvertToValueTuples(columns));
+                    InputOutputColumnPair.ConvertToValueTuples(columns));
         }
 
         /// <summary>
@@ -437,8 +421,8 @@ public static ValueMappingEstimator MapValue(
         [BestFriend]
         internal static ValueMappingEstimator MapValue(
             this TransformsCatalog.ConversionTransforms catalog,
-            IDataView lookupMap, DataViewSchema.Column keyColumn, DataViewSchema.Column valueColumn, params ColumnOptions[] columns)
+            IDataView lookupMap, DataViewSchema.Column keyColumn, DataViewSchema.Column valueColumn, params InputOutputColumnPair[] columns)
             => new ValueMappingEstimator(CatalogUtils.GetEnvironment(catalog), lookupMap, keyColumn.Name, valueColumn.Name,
-                ColumnOptions.ConvertToValueTuples(columns));
+                InputOutputColumnPair.ConvertToValueTuples(columns));
     }
 }
diff --git a/src/Microsoft.ML.Data/Transforms/ExtensionsCatalog.cs b/src/Microsoft.ML.Data/Transforms/ExtensionsCatalog.cs
@@ -32,40 +32,11 @@ public InputOutputColumnPair(string outputColumnName, string inputColumnName = n
             InputColumnName = inputColumnName ?? outputColumnName;
             OutputColumnName = outputColumnName;
         }
-    }
-
-    /// <summary>
-    /// Specifies input and output column names for a transformation.
-    /// </summary>
-    [BestFriend]
-    internal sealed class ColumnOptions
-    {
-        private readonly string _outputColumnName;
-        private readonly string _inputColumnName;
-
-        /// <summary>
-        /// Specifies input and output column names for a transformation.
-        /// </summary>
-        /// <param name="outputColumnName">Name of the column resulting from the transformation of <paramref name="inputColumnName"/>.</param>
-        /// <param name="inputColumnName">Name of the column to transform. If set to <see langword="null"/>, the value of the <paramref name="outputColumnName"/> will be used as source.</param>
-        public ColumnOptions(string outputColumnName, string inputColumnName = null)
-        {
-            _outputColumnName = outputColumnName;
-            _inputColumnName = inputColumnName ?? outputColumnName;
-        }
-
-        /// <summary>
-        /// Instantiates a <see cref="ColumnOptions"/> from a tuple of input and output column names.
-        /// </summary>
-        public static implicit operator ColumnOptions((string outputColumnName, string inputColumnName) value)
-        {
-            return new ColumnOptions(value.outputColumnName, value.inputColumnName);
-        }
 
         [BestFriend]
-        internal static (string outputColumnName, string inputColumnName)[] ConvertToValueTuples(ColumnOptions[] infos)
+        internal static (string outputColumnName, string inputColumnName)[] ConvertToValueTuples(InputOutputColumnPair[] infos)
         {
-            return infos.Select(info => (info._outputColumnName, info._inputColumnName)).ToArray();
+            return infos.Select(info => (info.OutputColumnName, info.InputColumnName)).ToArray();
         }
     }
 
@@ -104,8 +75,8 @@ public static ColumnCopyingEstimator CopyColumns(this TransformsCatalog catalog,
         /// </format>
         /// </example>
         [BestFriend]
-        internal static ColumnCopyingEstimator CopyColumns(this TransformsCatalog catalog, params ColumnOptions[] columns)
-            => new ColumnCopyingEstimator(CatalogUtils.GetEnvironment(catalog), ColumnOptions.ConvertToValueTuples(columns));
+        internal static ColumnCopyingEstimator CopyColumns(this TransformsCatalog catalog, params InputOutputColumnPair[] columns)
+            => new ColumnCopyingEstimator(CatalogUtils.GetEnvironment(catalog), InputOutputColumnPair.ConvertToValueTuples(columns));
 
         /// <summary>
         /// Concatenates columns together.
diff --git a/src/Microsoft.ML.ImageAnalytics/ExtensionsCatalog.cs b/src/Microsoft.ML.ImageAnalytics/ExtensionsCatalog.cs
@@ -32,8 +32,8 @@ public static ImageGrayscalingEstimator ConvertToGrayscale(this TransformsCatalo
         /// ]]></format>
         /// </example>
         [BestFriend]
-        internal static ImageGrayscalingEstimator ConvertToGrayscale(this TransformsCatalog catalog, params ColumnOptions[] columns)
-            => new ImageGrayscalingEstimator(CatalogUtils.GetEnvironment(catalog), ColumnOptions.ConvertToValueTuples(columns));
+        internal static ImageGrayscalingEstimator ConvertToGrayscale(this TransformsCatalog catalog, params InputOutputColumnPair[] columns)
+            => new ImageGrayscalingEstimator(CatalogUtils.GetEnvironment(catalog), InputOutputColumnPair.ConvertToValueTuples(columns));
 
         /// <summary>
         /// Loads the images from the <see cref="ImageLoadingTransformer.ImageFolder" /> into memory.
@@ -80,8 +80,8 @@ public static ImageLoadingEstimator LoadImages(this TransformsCatalog catalog, s
         /// ]]></format>
         /// </example>
         [BestFriend]
-        internal static ImageLoadingEstimator LoadImages(this TransformsCatalog catalog, string imageFolder, params ColumnOptions[] columns)
-           => new ImageLoadingEstimator(CatalogUtils.GetEnvironment(catalog), imageFolder, ColumnOptions.ConvertToValueTuples(columns));
+        internal static ImageLoadingEstimator LoadImages(this TransformsCatalog catalog, string imageFolder, params InputOutputColumnPair[] columns)
+           => new ImageLoadingEstimator(CatalogUtils.GetEnvironment(catalog), imageFolder, InputOutputColumnPair.ConvertToValueTuples(columns));
 
         /// <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
@@ -20,8 +20,8 @@ public static class ConversionsCatalog
         /// <param name="columns">Specifies the output and input columns on which the transformation should be applied.</param>
         [BestFriend]
         internal static KeyToBinaryVectorMappingEstimator MapKeyToBinaryVector(this TransformsCatalog.ConversionTransforms catalog,
-            params ColumnOptions[] columns)
-            => new KeyToBinaryVectorMappingEstimator(CatalogUtils.GetEnvironment(catalog), ColumnOptions.ConvertToValueTuples(columns));
+            params InputOutputColumnPair[] columns)
+            => new KeyToBinaryVectorMappingEstimator(CatalogUtils.GetEnvironment(catalog), InputOutputColumnPair.ConvertToValueTuples(columns));
 
         /// <summary>
         ///  Convert the key types back to binary vector.
diff --git a/src/Microsoft.ML.Transforms/NormalizerCatalog.cs b/src/Microsoft.ML.Transforms/NormalizerCatalog.cs
@@ -43,8 +43,8 @@ public static NormalizingEstimator Normalize(this TransformsCatalog catalog,
         [BestFriend]
         internal static NormalizingEstimator Normalize(this TransformsCatalog catalog,
             NormalizingEstimator.NormalizationMode mode,
-            params ColumnOptions[] columns)
-            => new NormalizingEstimator(CatalogUtils.GetEnvironment(catalog), mode, ColumnOptions.ConvertToValueTuples(columns));
+            params InputOutputColumnPair[] columns)
+            => new NormalizingEstimator(CatalogUtils.GetEnvironment(catalog), mode, InputOutputColumnPair.ConvertToValueTuples(columns));
 
         /// <summary>
         /// Normalize (rescale) columns according to specified custom parameters.
diff --git a/src/Microsoft.ML.Transforms/Text/TextCatalog.cs b/src/Microsoft.ML.Transforms/Text/TextCatalog.cs
@@ -75,8 +75,8 @@ public static TokenizingByCharactersEstimator TokenizeIntoCharactersAsKeys(this
         [BestFriend]
         internal static TokenizingByCharactersEstimator TokenizeIntoCharactersAsKeys(this TransformsCatalog.TextTransforms catalog,
             bool useMarkerCharacters = CharTokenizingDefaults.UseMarkerCharacters,
-            params ColumnOptions[] columns)
-            => new TokenizingByCharactersEstimator(Contracts.CheckRef(catalog, nameof(catalog)).GetEnvironment(), useMarkerCharacters, ColumnOptions.ConvertToValueTuples(columns));
+            params InputOutputColumnPair[] columns)
+            => new TokenizingByCharactersEstimator(Contracts.CheckRef(catalog, nameof(catalog)).GetEnvironment(), useMarkerCharacters, InputOutputColumnPair.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/Scenarios/Api/CookbookSamples/CookbookSamplesDynamicApi.cs b/test/Microsoft.ML.Tests/Scenarios/Api/CookbookSamples/CookbookSamplesDynamicApi.cs
@@ -187,7 +187,7 @@ private ITransformer TrainOnIris(string irisDataPath)
             // 		[2]	-9.709775	float
 
             // Apply the inverse conversion from 'PredictedLabel' column back to string value.
-            var finalPipeline = pipeline.Append(mlContext.Transforms.Conversion.MapKeyToValue(("Data", "PredictedLabel")));
+            var finalPipeline = pipeline.Append(mlContext.Transforms.Conversion.MapKeyToValue("Data", "PredictedLabel"));
             dataPreview = finalPipeline.Preview(trainData);
 
             return finalPipeline.Fit(trainData);
diff --git a/test/Microsoft.ML.Tests/Scenarios/IrisPlantClassificationWithStringLabelTests.cs b/test/Microsoft.ML.Tests/Scenarios/IrisPlantClassificationWithStringLabelTests.cs
@@ -39,7 +39,7 @@ public void TrainAndPredictIrisModelWithStringLabelTest()
                 .AppendCacheCheckpoint(mlContext)
                 .Append(mlContext.MulticlassClassification.Trainers.SdcaCalibrated(
                     new SdcaCalibratedMulticlassTrainer.Options { NumberOfThreads = 1 }))
-                .Append(mlContext.Transforms.Conversion.MapKeyToValue(("Plant", "PredictedLabel")));
+                .Append(mlContext.Transforms.Conversion.MapKeyToValue("Plant", "PredictedLabel"));
 
             // Train the pipeline
             var trainedModel = pipe.Fit(trainData);
diff --git a/test/Microsoft.ML.Tests/ScenariosWithDirectInstantiation/TensorflowTests.cs b/test/Microsoft.ML.Tests/ScenariosWithDirectInstantiation/TensorflowTests.cs
@@ -391,7 +391,7 @@ public void TensorFlowTransformInceptionTest()
                );
 
             var data = reader.Load(new MultiFileSource(dataFile));
-            var images = mlContext.Transforms.LoadImages(imageFolder, ("ImageReal", "ImagePath")).Fit(data).Transform(data);
+            var images = mlContext.Transforms.LoadImages("ImageReal", "ImagePath", imageFolder).Fit(data).Transform(data);
             var cropped = mlContext.Transforms.ResizeImages("ImageCropped", 224, 224, "ImageReal").Fit(images).Transform(images);
             var pixels = mlContext.Transforms.ExtractPixels(inputName, "ImageCropped", interleavePixelColors: true).Fit(cropped).Transform(cropped);
             var tf = mlContext.Model.LoadTensorFlowModel(modelLocation).ScoreTensorFlowModel(outputName, inputName, true).Fit(pixels).Transform(pixels);
@@ -507,7 +507,7 @@ public void TensorFlowTransformMNISTConvTest()
             var trainData = reader.Load(GetDataPath(TestDatasets.mnistTiny28.trainFilename));
             var testData = reader.Load(GetDataPath(TestDatasets.mnistOneClass.testFilename));
 
-            var pipe = mlContext.Transforms.CopyColumns(("reshape_input", "Placeholder"))
+            var pipe = mlContext.Transforms.CopyColumns("reshape_input", "Placeholder")
                 .Append(mlContext.Model.LoadTensorFlowModel("mnist_model/frozen_saved_model.pb").ScoreTensorFlowModel(new[] { "Softmax", "dense/Relu" }, new[] { "Placeholder", "reshape_input" }))
                 .Append(mlContext.Transforms.Concatenate("Features", "Softmax", "dense/Relu"))
                 .Append(mlContext.MulticlassClassification.Trainers.LightGbm("Label", "Features"));
@@ -662,7 +662,7 @@ private void ExecuteTFTransformMNISTConvTrainingTest(bool shuffle, int? shuffleS
                     preprocessedTestData = testData;
                 }
 
-                var pipe = mlContext.Transforms.CopyColumns(("Features", "Placeholder"))
+                var pipe = mlContext.Transforms.CopyColumns("Features", "Placeholder")
                     .Append(mlContext.Model.LoadTensorFlowModel(modelLocation).RetrainTensorFlowModel(
                         inputColumnNames: new[] { "Features" },
                         outputColumnNames: new[] { "Prediction" },
@@ -729,7 +729,7 @@ public void TensorFlowTransformMNISTConvSavedModelTest()
             var trainData = reader.Load(GetDataPath(TestDatasets.mnistTiny28.trainFilename));
             var testData = reader.Load(GetDataPath(TestDatasets.mnistOneClass.testFilename));
 
-            var pipe = mlContext.Transforms.CopyColumns(("reshape_input", "Placeholder"))
+            var pipe = mlContext.Transforms.CopyColumns("reshape_input", "Placeholder")
                 .Append(mlContext.Model.LoadTensorFlowModel("mnist_model").ScoreTensorFlowModel(new[] { "Softmax", "dense/Relu" }, new[] { "Placeholder", "reshape_input" }))
                 .Append(mlContext.Transforms.Concatenate("Features", new[] { "Softmax", "dense/Relu" }))
                 .Append(mlContext.MulticlassClassification.Trainers.LightGbm("Label", "Features"));
@@ -898,7 +898,7 @@ public void TensorFlowTransformCifarSavedModel()
                         new TextLoader.Column("Name", DataKind.String, 1),
                 }
             );
-            var images = mlContext.Transforms.LoadImages(imageFolder, ("ImageReal", "ImagePath")).Fit(data).Transform(data);
+            var images = mlContext.Transforms.LoadImages("ImageReal", imageFolder, "ImagePath").Fit(data).Transform(data);
             var cropped = mlContext.Transforms.ResizeImages("ImageCropped", imageWidth, imageHeight, "ImageReal").Fit(images).Transform(images);
             var pixels = mlContext.Transforms.ExtractPixels("Input", "ImageCropped", interleavePixelColors: true).Fit(cropped).Transform(cropped);
             IDataView trans = tensorFlowModel.ScoreTensorFlowModel("Output", "Input").Fit(pixels).Transform(pixels);
@@ -1000,15 +1000,15 @@ public void TensorFlowSentimentClassificationTest()
             // The second pipeline 'tfEnginePipe' takes the resized integer vector and passes it to TensoFlow and gets the classification scores.
             var estimator = mlContext.Transforms.Text.TokenizeIntoWords("TokenizedWords", "Sentiment_Text")
                 .Append(mlContext.Transforms.Conversion.MapValue(lookupMap, lookupMap.Schema["Words"], lookupMap.Schema["Ids"], 
-                    new ColumnOptions[] { ("Features", "TokenizedWords") }));
+                    new[] { new InputOutputColumnPair("Features", "TokenizedWords") }));
             var model = estimator.Fit(dataView);
             var dataPipe = mlContext.Model.CreatePredictionEngine<TensorFlowSentiment, TensorFlowSentiment>(model);
 
             // For explanation on how was the `sentiment_model` created 
             // c.f. https://github.com/dotnet/machinelearning-testdata/blob/master/Microsoft.ML.TensorFlow.TestModels/sentiment_model/README.md
             string modelLocation = @"sentiment_model";
             var pipelineModel = mlContext.Model.LoadTensorFlowModel(modelLocation).ScoreTensorFlowModel(new[] { "Prediction/Softmax" }, new[] { "Features" })
-                .Append(mlContext.Transforms.CopyColumns(("Prediction", "Prediction/Softmax")))
+                .Append(mlContext.Transforms.CopyColumns("Prediction", "Prediction/Softmax"))
                 .Fit(dataView);
             var tfEnginePipe = mlContext.Model.CreatePredictionEngine<TensorFlowSentiment, TensorFlowSentiment>(pipelineModel);
 
@@ -1052,7 +1052,7 @@ public void TensorFlowStringTest()
             var dataview = mlContext.Data.CreateTextLoader<TextInput>().Load(new MultiFileSource(null));
 
             var pipeline = tensorFlowModel.ScoreTensorFlowModel(new[] { "Original_A", "Joined_Splited_Text" }, new[] { "A", "B" })
-                .Append(mlContext.Transforms.CopyColumns(("AOut", "Original_A"), ("BOut", "Joined_Splited_Text")));
+                .Append(mlContext.Transforms.CopyColumns(new[] { new InputOutputColumnPair("AOut", "Original_A"), new InputOutputColumnPair("BOut", "Joined_Splited_Text") }));
             var transformer = mlContext.Model.CreatePredictionEngine<TextInput, TextOutput>(pipeline.Fit(dataview));
 
             var input = new TextInput
diff --git a/test/Microsoft.ML.Tests/Transformers/KeyToBinaryVectorEstimatorTest.cs b/test/Microsoft.ML.Tests/Transformers/KeyToBinaryVectorEstimatorTest.cs
diff --git a/test/Microsoft.ML.Tests/Transformers/NormalizerTests.cs b/test/Microsoft.ML.Tests/Transformers/NormalizerTests.cs
diff --git a/test/Microsoft.ML.Tests/Transformers/ValueMappingTests.cs b/test/Microsoft.ML.Tests/Transformers/ValueMappingTests.cs
