Addressing PR comments

Author: sfilipi

docs/samples/Microsoft.ML.Samples/Dynamic/KeyToValue_Term.cs

@@ -12,11 +12,11 @@ public static void KeyToValue_Term()
         {
             // 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 ml = new MLContext();
+            var mlContext = new MLContext();
 
             // Get a small dataset as an IEnumerable.
             IEnumerable<SamplesUtils.DatasetUtils.SampleTopicsData> data = SamplesUtils.DatasetUtils.GetTopicsData();
-            var trainData = ml.CreateStreamingDataView(data);
+            var trainData = mlContext.CreateStreamingDataView(data);
 
             // Preview of one of the columns of the the topics data. 
             // The Review column contains the keys associated with a particular body of text.  
@@ -31,16 +31,16 @@ public static void KeyToValue_Term()
             // making use of default settings.
             string defaultColumnName = "DefaultKeys";
             // REVIEW create through the catalog extension
-            var default_pipeline = new WordTokenizingEstimator(ml, "Review")
-                .Append(new ValueToKeyMappingEstimator(ml, "Review", defaultColumnName));
+            var default_pipeline = mlContext.Transforms.Text.TokenizeWords("Review")
+                                   .Append(mlContext.Transforms.Conversion.MapValueToKey(defaultColumnName, "Review"));
 
             // Another pipeline, that customizes the advanced settings of the TermEstimator.
             // We can change the maxNumTerm to limit how many keys will get generated out of the set of words, 
             // and condition the order in which they get evaluated by changing sort from the default Occurence (order in which they get encountered) 
             // to value/alphabetically.
             string customizedColumnName = "CustomizedKeys";
-            var customized_pipeline = new WordTokenizingEstimator(ml, "Review")
-                .Append(new ValueToKeyMappingEstimator(ml, "Review", customizedColumnName, maxNumTerms: 10, sort: ValueToKeyMappingTransformer.SortOrder.Value));
+            var customized_pipeline = mlContext.Transforms.Text.TokenizeWords("Review")
+                .Append(mlContext.Transforms.Conversion.MapValueToKey(customizedColumnName, "Review", maxNumTerms: 10, sort: ValueToKeyMappingTransformer.SortOrder.Value));
 
             // The transformed data.
             var transformedData_default = default_pipeline.Fit(trainData).Transform(trainData);
@@ -61,7 +61,7 @@ public static void KeyToValue_Term()
             };
 
             // Preview of the DefaultKeys column obtained after processing the input.
-            var defaultColumn = transformedData_default.GetColumn<VBuffer<uint>>(ml, defaultColumnName);
+            var defaultColumn = transformedData_default.GetColumn<VBuffer<uint>>(mlContext, defaultColumnName);
             printHelper(defaultColumnName, defaultColumn);
 
             // DefaultKeys column obtained post-transformation.
@@ -72,7 +72,7 @@ public static void KeyToValue_Term()
             // 9 10 11 12 13 6
 
             // Previewing the CustomizedKeys column obtained after processing the input.
-            var customizedColumn = transformedData_customized.GetColumn<VBuffer<uint>>(ml, customizedColumnName);
+            var customizedColumn = transformedData_customized.GetColumn<VBuffer<uint>>(mlContext, customizedColumnName);
             printHelper(customizedColumnName, customizedColumn);
 
             // CustomizedKeys column obtained post-transformation.
@@ -84,11 +84,11 @@ public static void KeyToValue_Term()
 
             // Retrieve the original values, by appending the KeyToValue etimator to the existing pipelines
             // to convert the keys back to the strings.
-            var pipeline = default_pipeline.Append(new KeyToValueMappingEstimator(ml, defaultColumnName));
+            var pipeline = default_pipeline.Append(new KeyToValueMappingEstimator(mlContext, defaultColumnName));
             transformedData_default = pipeline.Fit(trainData).Transform(trainData);
 
             // Preview of the DefaultColumnName column obtained.
-            var originalColumnBack = transformedData_default.GetColumn<VBuffer<ReadOnlyMemory<char>>>(ml, defaultColumnName);
+            var originalColumnBack = transformedData_default.GetColumn<VBuffer<ReadOnlyMemory<char>>>(mlContext, defaultColumnName);
 
             foreach (var row in originalColumnBack)
             {

src/Microsoft.ML.Data/Commands/CrossValidationCommand.cs

@@ -330,7 +330,7 @@ private string GetSplitColumn(IChannel ch, IDataView input, ref IDataView output
                     int inc = 0;
                     while (input.Schema.TryGetColumnIndex(stratificationColumn, out tmp))
                         stratificationColumn = string.Format("{0}_{1:000}", origStratCol, ++inc);
-                    output = new HashingEstimator(Host, origStratCol, stratificationColumn, 30).Fit(input).Transform(input);
+                    output = new HashingEstimator(Host, stratificationColumn, origStratCol, 30).Fit(input).Transform(input);
                 }
             }
 

src/Microsoft.ML.Data/TrainContext.cs

@@ -151,7 +151,7 @@ private void EnsureStratificationColumn(ref IDataView data, ref string stratific
                     // Generate a new column with the hashed stratification column.
                     while (data.Schema.TryGetColumnIndex(stratificationColumn, out tmp))
                         stratificationColumn = string.Format("{0}_{1:000}", origStratCol, ++inc);
-                    data = new HashingEstimator(Host, origStratCol, stratificationColumn, 30).Fit(data).Transform(data);
+                    data = new HashingEstimator(Host, stratificationColumn, origStratCol, 30).Fit(data).Transform(data);
                 }
             }
         }

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

@@ -20,16 +20,17 @@ public static class ConversionsExtensionsCatalog
         /// Hashes the values in the input column.
         /// </summary>
         /// <param name="catalog">The transform's catalog.</param>
-        /// <param name="inputColumn">Name of the input column.</param>
         /// <param name="outputColumn">Name of the column to be transformed.</param>
+        /// <param name="inputColumn">Name of the input column. If set to <see langword="null"/>, the value of the <paramref name="outputColumn"/>
+        /// will be used as input.</param>
         /// <param name="hashBits">Number of bits to hash into. Must be between 1 and 31, inclusive.</param>
         /// <param name="invertHash">During hashing we constuct mappings between original values and the produced hash values.
         /// Text representation of original values are stored in the slot names of the  metadata for the new column.Hashing, as such, can map many initial values to one.
         /// <paramref name="invertHash"/> specifies the upper bound of the number of distinct input values mapping to a hash that should be retained.
         /// <value>0</value> does not retain any input values. <value>-1</value> retains all input values mapping to each hash.</param>
-        public static HashingEstimator Hash(this TransformsCatalog.ConversionTransforms catalog, string outputColumn, string inputColumn,
+        public static HashingEstimator Hash(this TransformsCatalog.ConversionTransforms catalog, string outputColumn, string inputColumn = null,
             int hashBits = HashDefaults.HashBits, int invertHash = HashDefaults.InvertHash)
-            => new HashingEstimator(CatalogUtils.GetEnvironment(catalog), inputColumn, outputColumn, hashBits, invertHash);
+            => new HashingEstimator(CatalogUtils.GetEnvironment(catalog), outputColumn, inputColumn ?? outputColumn, hashBits, invertHash);
 
         /// <summary>
         /// Hashes the values in the input column.
@@ -43,10 +44,11 @@ public static HashingEstimator Hash(this TransformsCatalog.ConversionTransforms
         /// Changes column type of the input column.
         /// </summary>
         /// <param name="catalog">The transform's catalog.</param>
-        /// <param name="inputColumn">Name of the input column.</param>
         /// <param name="outputColumn">Name of the column to be transformed.</param>
+        /// <param name="inputColumn">Name of the input column. If set to <see langword="null"/>, the value of the <paramref name="outputColumn"/>
+        /// will be used as input.</param>
         /// <param name="outputKind">Number of bits to hash into. Must be between 1 and 31, inclusive.</param>
-        public static TypeConvertingEstimator ConvertType(this TransformsCatalog.ConversionTransforms catalog, string outputColumn, string inputColumn,
+        public static TypeConvertingEstimator ConvertType(this TransformsCatalog.ConversionTransforms catalog, string outputColumn, string inputColumn = null,
             DataKind outputKind = ConvertDefaults.DefaultOutputKind)
             => new TypeConvertingEstimator(CatalogUtils.GetEnvironment(catalog), outputColumn, inputColumn, outputKind);
 
@@ -88,44 +90,31 @@ public static KeyToVectorMappingEstimator MapKeyToVector(this TransformsCatalog.
         /// Convert the key types back to their original vectors.
         /// </summary>
         /// <param name="catalog">The categorical transform's catalog.</param>
-        /// <param name="inputColumn">The name of the input column.</param>
+        /// <param name="inputColumn">The name of the input column. If set to <see langword="null"/>, the value of the <paramref name="outputColumn"/>
+        /// will be used as input.</param>
         /// <param name="outputColumn">The name of the output column.</param>
         /// <param name="bag">Whether bagging is used for the conversion. </param>
         public static KeyToVectorMappingEstimator MapKeyToVector(this TransformsCatalog.ConversionTransforms catalog,
-            string outputColumn, string inputColumn, bool bag = KeyToVectorMappingEstimator.Defaults.Bag)
-            => new KeyToVectorMappingEstimator(CatalogUtils.GetEnvironment(catalog), outputColumn, inputColumn, bag);
+            string outputColumn, string inputColumn = null, bool bag = KeyToVectorMappingEstimator.Defaults.Bag)
+            => new KeyToVectorMappingEstimator(CatalogUtils.GetEnvironment(catalog), outputColumn, inputColumn ?? outputColumn, bag);
 
         /// <summary>
         /// Converts value types into <see cref="KeyType"/>.
         /// </summary>
         /// <param name="catalog">The categorical transform's catalog.</param>
-        /// <param name="inputColumn">Name of the column to be transformed.</param>
+        /// <param name="inputColumn">Name of the column to be transformed. If set to <see langword="null"/>, the value of the <paramref name="outputColumn"/>
+        /// will be used as input.</param>
         /// <param name="outputColumn">Name of the output column.</param>
         /// <param name="maxNumTerms">Maximum number of keys to keep per column when auto-training.</param>
         /// <param name="sort">How items should be ordered when vectorized. If <see cref="ValueToKeyMappingTransformer.SortOrder.Occurrence"/> choosen they will be in the order encountered.
         /// If <see cref="ValueToKeyMappingTransformer.SortOrder.Value"/>, items are sorted according to their default comparison, for example, text sorting will be case sensitive (for example, 'A' then 'Z' then 'a').</param>
         public static ValueToKeyMappingEstimator MapValueToKey(this TransformsCatalog.ConversionTransforms catalog,
             string outputColumn,
-            string inputColumn,
+            string inputColumn = null,
             int maxNumTerms = ValueToKeyMappingEstimator.Defaults.MaxNumTerms,
             ValueToKeyMappingTransformer.SortOrder sort = ValueToKeyMappingEstimator.Defaults.Sort)
            => new ValueToKeyMappingEstimator(CatalogUtils.GetEnvironment(catalog), outputColumn, inputColumn, maxNumTerms, sort);
 
-        /// <summary>
-        /// Converts value types into <see cref="KeyType"/> loading the keys to use from <paramref name="file"/>.
-        /// </summary>
-        /// <param name="catalog">The categorical transform's catalog.</param>
-        /// <param name="columns">The data columns to map to keys.</param>
-        /// <param name="file">The path of the file containing the terms.</param>
-        /// <param name="termsColumn"></param>
-        /// <param name="loaderFactory"></param>
-        public static ValueToKeyMappingEstimator MapValueToKey(this TransformsCatalog.ConversionTransforms catalog,
-            ValueToKeyMappingTransformer.ColumnInfo[] columns,
-            string file = null,
-            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>
@@ -141,7 +130,7 @@ public static ValueMappingEstimator<TInputType, TOutputType> ValueMap<TInputType
             this TransformsCatalog.ConversionTransforms catalog,
             IEnumerable<TInputType> keys,
             IEnumerable<TOutputType> values,
-            params (string source, string name)[] columns)
+            params (string inputColumn, string outputColumn)[] columns)
             => new ValueMappingEstimator<TInputType, TOutputType>(CatalogUtils.GetEnvironment(catalog), keys, values, columns);
     }
 }

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

@@ -135,8 +135,8 @@ public sealed class ColumnInfo
             /// Text representation of original values are stored in the slot names of the  metadata for the new column.Hashing, as such, can map many initial values to one.
             /// <paramref name="invertHash"/> specifies the upper bound of the number of distinct input values mapping to a hash that should be retained.
             /// <value>0</value> does not retain any input values. <value>-1</value> retains all input values mapping to each hash.</param>
-            public ColumnInfo(string input,
-                string output = null,
+            public ColumnInfo(string output,
+                string input = null,
                 int hashBits = HashingEstimator.Defaults.HashBits,
                 uint seed = HashingEstimator.Defaults.Seed,
                 bool ordered = HashingEstimator.Defaults.Ordered,
@@ -146,7 +146,7 @@ public ColumnInfo(string input,
                     throw Contracts.ExceptParam(nameof(invertHash), "Value too small, must be -1 or larger");
                 if (invertHash != 0 && hashBits >= 31)
                     throw Contracts.ExceptParam(nameof(hashBits), $"Cannot support invertHash for a {0} bit hash. 30 is the maximum possible.", hashBits);
-                Contracts.CheckNonWhiteSpace(input, nameof(input));
+                Contracts.CheckNonWhiteSpace(output, nameof(output));
                 Input = input;
                 Output = output ?? input;
                 HashBits = hashBits;
@@ -157,6 +157,8 @@ public ColumnInfo(string input,
 
             internal ColumnInfo(string input, string output, ModelLoadContext ctx)
             {
+                Contracts.CheckNonWhiteSpace(output, nameof(output));
+
                 Input = input;
                 Output = output;
                 // *** Binary format ***
@@ -386,8 +388,8 @@ private static IDataTransform Create(IHostEnvironment env, Arguments args, IData
             {
                 var item = args.Column[i];
                 var kind = item.InvertHash ?? args.InvertHash;
-                cols[i] = new ColumnInfo(item.Source ?? item.Name,
-                    item.Name,
+                cols[i] = new ColumnInfo(item.Name,
+                    item.Source ?? item.Name,
                     item.HashBits ?? args.HashBits,
                     item.Seed ?? args.Seed,
                     item.Ordered ?? args.Ordered,
@@ -1211,16 +1213,17 @@ internal static bool IsColumnTypeValid(ColumnType type)
         /// Initializes a new instance of <see cref="HashingEstimator"/>.
         /// </summary>
         /// <param name="env">Host Environment.</param>
-        /// <param name="inputColumn">Name of the column to be transformed.</param>
+        /// <param name="inputColumn">Name of the column to be transformed.
+        /// If set to <see langword="null"/> the value specified for the <paramref name="outputColumn"/> will be used.</param>
         /// <param name="outputColumn">Name of the output column. </param>
         /// <param name="hashBits">Number of bits to hash into. Must be between 1 and 31, inclusive.</param>
         /// <param name="invertHash">During hashing we constuct mappings between original values and the produced hash values.
         /// Text representation of original values are stored in the slot names of the  metadata for the new column.Hashing, as such, can map many initial values to one.
         /// <paramref name="invertHash"/> specifies the upper bound of the number of distinct input values mapping to a hash that should be retained.
         /// <value>0</value> does not retain any input values. <value>-1</value> retains all input values mapping to each hash.</param>
-        internal HashingEstimator(IHostEnvironment env, string inputColumn, string outputColumn,
+        internal HashingEstimator(IHostEnvironment env, string outputColumn, string inputColumn = null,
             int hashBits = Defaults.HashBits, int invertHash = Defaults.InvertHash)
-            : this(env, new HashingTransformer.ColumnInfo(inputColumn, outputColumn, hashBits: hashBits, invertHash: invertHash))
+            : this(env, new HashingTransformer.ColumnInfo(outputColumn, inputColumn ?? outputColumn, hashBits: hashBits, invertHash: invertHash))
         {
         }