Addition of the ValueMappingEstimator and ValueMappingTransform. (#1710)

* Addition of the ValueMappingEstimator and ValueMappingTransform.
This will be replacing the TermLookupTransform and provide a way to
specify the mapping betweeen two values (note this is specified and not
trained). A user can specify the mapping by providing a keys list and
values list that must be equal in size. The Estimator will then generate
a 1-1 mapping based on the two lists.

The PR references #754 which covers the conversion of Transformer to use
the new Estimator API.

Author: singlis

src/Microsoft.ML.Data/DataView/ArrayDataViewBuilder.cs

@@ -75,12 +75,18 @@ public void AddColumn<T>(string name, PrimitiveType type, params T[] values)
         /// Constructs a new key column from an array where values are copied to output simply
         /// by being assigned.
         /// </summary>
-        public void AddColumn(string name, ValueGetter<VBuffer<ReadOnlyMemory<char>>> getKeyValues, ulong keyMin, int keyCount, params uint[] values)
+        /// <param name="name">The name of the column.</param>
+        /// <param name="getKeyValues">The delegate that does a reverse lookup based upon the given key. This is for metadata creation</param>
+        /// <param name="keyMin">The <see cref="KeyType"/> minimum to use.</param>
+        /// <param name="keyCount">The count of unique keys specified in values</param>
+        /// <param name="values">The values to add to the column. Note that since this is creating a <see cref="KeyType"/> column, the values will be offset by 1.</param>
+        public void AddColumn<T1>(string name, ValueGetter<VBuffer<ReadOnlyMemory<char>>> getKeyValues, ulong keyMin, int keyCount, params T1[] values)
         {
             _host.CheckValue(getKeyValues, nameof(getKeyValues));
             _host.CheckParam(keyCount > 0, nameof(keyCount));
             CheckLength(name, values);
-            _columns.Add(new AssignmentColumn<uint>(new KeyType(DataKind.U4, keyMin, keyCount), values));
+            values.GetType().GetElementType().TryGetDataKind(out DataKind kind);
+            _columns.Add(new AssignmentColumn<T1>(new KeyType(kind, keyMin, keyCount), values));
             _getKeyValues.Add(name, getKeyValues);
             _names.Add(name);
         }

src/Microsoft.ML.Data/Transforms/ConversionsExtensionsCatalog.cs

@@ -4,7 +4,9 @@
 
 using Microsoft.ML.Runtime;
 using Microsoft.ML.Runtime.Data;
+using Microsoft.ML.Transforms;
 using Microsoft.ML.Transforms.Conversions;
+using System.Collections.Generic;
 
 namespace Microsoft.ML
 {
@@ -125,5 +127,23 @@ public static ValueToKeyMappingEstimator MapValueToKey(this TransformsCatalog.Co
             string termsColumn = null,
             IComponentFactory<IMultiStreamSource, IDataLoader> loaderFactory = null)
             => new ValueToKeyMappingEstimator(CatalogUtils.GetEnvironment(catalog), columns, file, termsColumn, loaderFactory);
+
+        /// <summary>
+        /// Maps specified keys to specified values
+        /// </summary>
+        /// <typeparam name="TInputType">The key type.</typeparam>
+        /// <typeparam name="TOutputType">The value type.</typeparam>
+        /// <param name="catalog">The categorical transform's catalog</param>
+        /// <param name="keys">The list of keys to use for the mapping. The mapping is 1-1 with values. This list must be the same length as values and
+        /// cannot contain duplicate keys.</param>
+        /// <param name="values">The list of values to pair with the keys for the mapping. This list must be equal to the same length as keys.</param>
+        /// <param name="columns">The columns to apply this transform on.</param>
+        /// <returns></returns>
+        public static ValueMappingEstimator<TInputType, TOutputType> ValueMap<TInputType, TOutputType>(
+            this TransformsCatalog.ConversionTransforms catalog,
+            IEnumerable<TInputType> keys,
+            IEnumerable<TOutputType> values,
+            params (string source, string name)[] columns)
+            => new ValueMappingEstimator<TInputType, TOutputType>(CatalogUtils.GetEnvironment(catalog), keys, values, columns);
     }
 }