Conversion catalog samples

docs/samples/Microsoft.ML.Samples/Dynamic/Transforms/Conversion/ConvertType.cs

@@ -5,18 +5,6 @@ namespace Microsoft.ML.Samples.Dynamic
 {
     public static class ConvertType
     {
-        private sealed class InputData
-        {
-            public bool Survived;
-        }
-
-        private sealed class TransformedData
-        {
-            public bool Survived { get; set; }
-
-            public Int32 SurvivedInt32 { get; set; }
-        }
-
         public static void Example()
         {
             var mlContext = new MLContext(seed: 1);
@@ -51,5 +39,13 @@ public static void Example()
             // A: False    Aconv:0
             // A: False    Aconv:0
         }
+        private class InputData
+        {
+            public bool Survived;
+        }
+        private sealed class TransformedData : InputData
+        {
+            public Int32 SurvivedInt32 { get; set; }
+        }
     }
 }

docs/samples/Microsoft.ML.Samples/Dynamic/Transforms/Conversion/ConvertTypeMultiColumn.cs

@@ -0,0 +1,78 @@
+using System;
+using Microsoft.ML.Data;
+
+namespace Microsoft.ML.Samples.Dynamic
+{
+    // This example illustrates how to convert multiple columns of different types to one type, in this case System.Single. 
+    // This is often a useful data transformation before concatenating the features together and passing them to a particular estimator.
+    public static class ConvertTypeMultiColumn
+    { 
+        public static void Example()
+        {
+            // Create a new ML context, for ML.NET operations. It can be used for exception tracking and logging, 
+            // as well as the source of randomness.
+            var mlContext = new MLContext(seed: 1);
+
+            var rawData = new[] {
+                new InputData() { Feature1 = true, Feature2 = "0.4", Feature3 = DateTime.Now, Feature4 = 0.145},
+                new InputData() { Feature1 = false, Feature2 = "0.5", Feature3 = DateTime.Today, Feature4 = 3.14},
+                new InputData() { Feature1 = false, Feature2 = "14", Feature3 = DateTime.Today, Feature4 = 0.2046},
+                new InputData() { Feature1 = false, Feature2 = "23", Feature3 = DateTime.Now, Feature4 = 0.1206},
+                new InputData() { Feature1 = true, Feature2 = "8904", Feature3 = DateTime.UtcNow, Feature4 = 8.09},
+            };
+
+            // Convert the data to an IDataView.
+            var data = mlContext.Data.LoadFromEnumerable(rawData);
+
+            // Construct the pipeline.
+            var pipeline = mlContext.Transforms.Conversion.ConvertType(new[]
+            {
+                    new InputOutputColumnPair("Converted1", "Feature1"),
+                    new InputOutputColumnPair("Converted2", "Feature2"),
+                    new InputOutputColumnPair("Converted3", "Feature3"),
+                    new InputOutputColumnPair("Converted4", "Feature4"),
+             }, 
+             DataKind.Single);
+
+            // Let's fit our pipeline to the data.
+            var transformer = pipeline.Fit(data);
+            // Transforming the same data. This will add the 4 columns defined in the pipeline, containing the converted
+            // values of the initial columns. 
+            var transformedData = transformer.Transform(data);
+
+            // Shape the transformed data as a strongly typed IEnumerable.
+            var convertedData = mlContext.Data.CreateEnumerable<TransformedData>(transformedData, true);
+
+            // Printing the results.
+            Console.WriteLine("Converted1\t Converted2\t Converted3\t Converted4");
+            foreach (var item in convertedData)
+                Console.WriteLine($"\t{item.Converted1}\t {item.Converted2}\t\t  {item.Converted3}\t {item.Converted4}");
+
+            // Transformed data.
+            //
+            // Converted1   Converted2    Converted3     Converted4
+            //      1        0.4        6.368921E+17        0.145
+            //      0        0.5        6.368916E+17        3.14
+            //      0        14         6.368916E+17        0.2046
+            //      0        23         6.368921E+17        0.1206
+            //      1       8904        6.368924E+17        8.09
+
+        }
+        // The initial data type
+        private class InputData
+        {
+            public bool Feature1;
+            public string Feature2;
+            public DateTime Feature3;
+            public double Feature4;
+        }
+        // The resulting data type after the transformation
+        private class TransformedData : InputData
+        {
+            public float Converted1 { get; set; }
+            public float Converted2 { get; set; }
+            public float Converted3 { get; set; }
+            public float Converted4 { get; set; }
+        }
+    }
+}

docs/samples/Microsoft.ML.Samples/Dynamic/Transforms/Conversion/MapValue.cs

@@ -0,0 +1,94 @@
+using System;
+using System.Collections.Generic;
+using Microsoft.ML.Data;
+
+
+namespace Microsoft.ML.Samples.Dynamic
+{
+    public static class MapValue
+    {
+        /// This example demonstrates the use of the ValueMappingEstimator by mapping strings to other string values, or floats to strings. 
+        /// This is useful to map types to a category. 
+        public static void Example()
+        {
+            // Create a new ML context, for ML.NET operations. It can be used for exception tracking and logging, 
+            // as well as the source of randomness.
+            var mlContext = new MLContext();
+
+            // Get a small dataset as an IEnumerable.
+            var rawData = new[] {
+                new DataPoint() { Timeframe = "0-4yrs" , Score = 1 },
+                new DataPoint() { Timeframe = "6-11yrs" , Score = 2 },
+                new DataPoint() { Timeframe = "12-25yrs" , Score = 3 },
+                new DataPoint() { Timeframe = "0-5yrs" , Score = 4 },
+                new DataPoint() { Timeframe = "12-25yrs" , Score = 5 },
+                new DataPoint() { Timeframe = "25+yrs" , Score = 5 },
+            };
+
+            var data = mlContext.Data.LoadFromEnumerable(rawData);
+
+            // Construct the mapping to other strings for the Timeframe column.  
+            var timeframeMap = new Dictionary<string, string>();
+            timeframeMap["0-4yrs"] = "Short";
+            timeframeMap["0-5yrs"] = "Short";
+            timeframeMap["6-11yrs"] = "Medium";
+            timeframeMap["12-25yrs"] = "Long";
+            timeframeMap["25+yrs"] = "Long";
+
+            // Construct the mapping of strings to keys(uints) for the Timeframe column. 
+            var timeframeKeyMap = new Dictionary<string, uint>();
+            timeframeKeyMap["0-4yrs"] = 1;
+            timeframeKeyMap["0-5yrs"] = 1;
+            timeframeKeyMap["6-11yrs"] = 2;
+            timeframeKeyMap["12-25yrs"] = 3;
+            timeframeKeyMap["25+yrs"] = 3;
+
+            // Construct the mapping of ints to strings for the Score column. 
+            var scoreMap = new Dictionary<int, string>();
+            scoreMap[1] = "Low";
+            scoreMap[2] = "Low";
+            scoreMap[3] = "Average";
+            scoreMap[4] = "High";
+            scoreMap[5] = "High";
+
+            // Constructs the ML.net pipeline
+            var pipeline = mlContext.Transforms.Conversion.MapValue("TimeframeCategory", timeframeMap, "Timeframe")
+                           .Append(mlContext.Transforms.Conversion.MapValue("ScoreCategory", scoreMap, "Score"))
+                           // on the MapValue below, the treatValuesAsKeyType is set to true. The type of the Label column will be a KeyDataViewType type, 
+                           // and it can be used as input for trainers performing multiclass classification.
+                           .Append(mlContext.Transforms.Conversion.MapValue("Label", timeframeKeyMap, "Timeframe", treatValuesAsKeyType: true));
+
+            // Fits the pipeline to the data.
+            IDataView transformedData = pipeline.Fit(data).Transform(data);
+
+            // Getting the resulting data as an IEnumerable.
+            // This will contain the newly created columns.
+            IEnumerable<TransformedData> features = mlContext.Data.CreateEnumerable<TransformedData>(transformedData, reuseRowObject: false);
+
+            Console.WriteLine($" Timeframe   TimeframeCategory   Label    Score   ScoreCategory");
+            foreach (var featureRow in features)
+                Console.WriteLine($"{featureRow.Timeframe}\t\t{featureRow.TimeframeCategory}\t\t\t{featureRow.Label}\t\t{featureRow.Score}\t{featureRow.ScoreCategory}");
+
+            // TransformedData obtained post-transformation.
+            //
+            //  Timeframe   TimeframeCategory   Label    Score   ScoreCategory
+            // 0-4yrs         Short              1       1       Low
+            // 6-11yrs        Medium             2       2       Low
+            // 12-25yrs       Long               3       3       Average
+            // 0-5yrs         Short              1       4       High
+            // 12-25yrs       Long               3       5       High
+            // 25+yrs         Long               3       5       High
+        }
+        private class DataPoint
+        {
+            public string Timeframe { get; set; }
+            public int Score { get; set; }
+        }
+        private class TransformedData : DataPoint
+        {
+            public string TimeframeCategory { get; set; }
+            public string ScoreCategory { get; set; }
+            public uint Label { get; set; }
+        }
+    }
+}

docs/samples/Microsoft.ML.Samples/Dynamic/Transforms/Conversion/MapValueIdvLookup.cs

@@ -0,0 +1,80 @@
+using System;
+using System.Collections.Generic;
+
+namespace Microsoft.ML.Samples.Dynamic
+{
+    public static class MapValueIdvLookup
+    {
+        /// This example demonstrates the use of MapValue by mapping floats to strings, looking up the mapping in an IDataView. 
+        /// This is useful to map types to a grouping. 
+        public static void Example()
+        {
+            // Create a new ML context, for ML.NET operations. It can be used for exception tracking and logging, 
+            // as well as the source of randomness.
+            var mlContext = new MLContext();
+
+            // Get a small dataset as an IEnumerable.
+            var rawData = new[] {
+                new DataPoint() { Price = 3.14f },
+                new DataPoint() { Price = 2000f },
+                new DataPoint() { Price = 1.19f },
+                new DataPoint() { Price = 2.17f },
+                new DataPoint() { Price = 33.784f },
+
+            };
+
+            // Convert to IDataView
+            var data = mlContext.Data.LoadFromEnumerable(rawData);
+
+            // Create the lookup map data IEnumerable.   
+            var lookupData = new[] {
+                new LookupMap { Value = 3.14f, Category = "Low" },
+                new LookupMap { Value = 1.19f , Category = "Low" },
+                new LookupMap { Value = 2.17f , Category = "Low" },
+                new LookupMap { Value = 33.784f, Category = "Medium" },
+                new LookupMap { Value = 2000f, Category = "High"}
+
+            };
+
+            // Convert to IDataView
+            var lookupIdvMap = mlContext.Data.LoadFromEnumerable(lookupData);
+
+            // Constructs the ValueMappingEstimator making the ML.NET pipeline
+            var pipeline = mlContext.Transforms.Conversion.MapValue("PriceCategory", lookupIdvMap, lookupIdvMap.Schema["Value"], lookupIdvMap.Schema["Category"], "Price");
+
+            // Fits the ValueMappingEstimator and transforms the data converting the Price to PriceCategory.
+            IDataView transformedData = pipeline.Fit(data).Transform(data);
+
+            // Getting the resulting data as an IEnumerable.
+            IEnumerable<TransformedData> features = mlContext.Data.CreateEnumerable<TransformedData>(transformedData, reuseRowObject: false);
+
+            Console.WriteLine($" Price   PriceCategory");
+            foreach (var featureRow in features)
+                Console.WriteLine($"{featureRow.Price}\t\t{featureRow.PriceCategory}");
+
+            // TransformedData obtained post-transformation.
+            //
+            // Price        PriceCategory
+            // 3.14            Low
+            // 2000            High
+            // 1.19            Low
+            // 2.17            Low
+            // 33.784          Medium
+        }
+
+        // Type for the IDataView that will be serving as the map
+        private class LookupMap
+        {
+            public float Value { get; set; }
+            public string Category { get; set; }
+        }
+        private class DataPoint
+        {
+            public float Price { get; set; }
+        }
+        private class TransformedData : DataPoint
+        {
+            public string PriceCategory { get; set; }
+        }
+    }
+}

docs/samples/Microsoft.ML.Samples/Dynamic/Transforms/Conversion/MapValueToArray.cs

@@ -0,0 +1,67 @@
+using System;
+using System.Collections.Generic;
+namespace Microsoft.ML.Samples.Dynamic
+{
+    public static class MapValueToArray
+    { 
+        /// This example demonstrates the use of MapValue by mapping strings to array values, which allows for mapping data to numeric arrays. 
+        /// This functionality is useful when the generated column will serve as the Features column for a trainer. Most of the trainers take a numeric vector, as the Features column. 
+        /// In this example, we are mapping the Timeframe data to arbitrary integer arrays.
+        public static void Example()
+        {
+            // Create a new ML context, for ML.NET operations. It can be used for exception tracking and logging, 
+            // as well as the source of randomness.
+            var mlContext = new MLContext();
+
+            // Get a small dataset as an IEnumerable.
+            var rawData = new[] {
+                new DataPoint() { Timeframe = "0-4yrs" },
+                new DataPoint() { Timeframe = "6-11yrs" },
+                new DataPoint() { Timeframe = "12-25yrs" },
+                new DataPoint() { Timeframe = "0-5yrs" },
+                new DataPoint() { Timeframe = "12-25yrs" },
+                new DataPoint() { Timeframe = "25+yrs" },
+            };
+
+            var data = mlContext.Data.LoadFromEnumerable(rawData);
+
+            // Creating a list of key-value pairs to indicate the mapping between the 
+            // DataPoint values, and the arrays they should map to. 
+            var timeframeMap = new Dictionary<string, int[]>();
+            timeframeMap["0-4yrs"] = new int[] { 0, 5, 300 };
+            timeframeMap["0-5yrs"] = new int[] { 0, 5, 300 };
+            timeframeMap["6-11yrs"] = new int[] { 6, 11, 300 };
+            timeframeMap["12-25yrs"] = new int[] { 12, 50, 300 };
+            timeframeMap["25+yrs"] = new int[] { 12, 50, 300 };
+
+            // Constructs the ValueMappingEstimator making the ML.NET pipeline.
+            var pipeline = mlContext.Transforms.Conversion.MapValue("Features", timeframeMap, "Timeframe");
+
+            // Fits the ValueMappingEstimator and transforms the data adding the Features column.
+            IDataView transformedData = pipeline.Fit(data).Transform(data);
+
+            // Getting the resulting data as an IEnumerable.
+            IEnumerable<TransformedData> featuresColumn = mlContext.Data.CreateEnumerable<TransformedData>(transformedData, reuseRowObject: false);
+
+            Console.WriteLine($"Timeframe     Features");
+            foreach (var featureRow in featuresColumn)
+                Console.WriteLine($"{featureRow.Timeframe}\t\t {string.Join(",", featureRow.Features)}");
+
+            // Timeframe      Features
+            // 0-4yrs       0, 5, 300
+            // 6-11yrs      6, 11, 300
+            // 12-25yrs     12, 50, 300
+            // 0-5yrs       0, 5, 300
+            // 12-25yrs     12, 50,300
+            // 25+yrs       12, 50, 300
+        }
+        public class DataPoint
+        {
+            public string Timeframe { get; set; }
+        }
+        public class TransformedData : DataPoint
+        {
+            public int[] Features { get; set; }
+        }
+    }
+}