One type label policy in trainers

src/Microsoft.ML.Core/Data/AnnotationUtils.cs

@@ -308,7 +308,7 @@ public static void GetSlotNames(RoleMappedSchema schema, RoleMappedSchema.Column
                 schema.Schema[list[0].Index].Annotations.GetValue(Kinds.SlotNames, ref slotNames);
         }
 
-        public static bool HasKeyValues(this SchemaShape.Column col)
+        public static bool NeedsSlotNames(this SchemaShape.Column col)
         {
             return col.Annotations.TryFindColumn(Kinds.KeyValues, out var metaCol)
                 && metaCol.Kind == SchemaShape.Column.VectorKind.Vector
@@ -442,7 +442,7 @@ public static bool TryGetCategoricalFeatureIndices(DataViewSchema schema, int co
         public static IEnumerable<SchemaShape.Column> AnnotationsForMulticlassScoreColumn(SchemaShape.Column? labelColumn = null)
         {
             var cols = new List<SchemaShape.Column>();
-            if (labelColumn != null && labelColumn.Value.IsKey && HasKeyValues(labelColumn.Value))
+            if (labelColumn != null && labelColumn.Value.IsKey && NeedsSlotNames(labelColumn.Value))
                 cols.Add(new SchemaShape.Column(Kinds.SlotNames, SchemaShape.Column.VectorKind.Vector, TextDataViewType.Instance, false));
             cols.AddRange(GetTrainerOutputAnnotation());
             return cols;

src/Microsoft.ML.Data/Data/Conversion.cs

@@ -111,6 +111,7 @@ private Conversions()
             AddStd<I1, R4>(Convert);
             AddStd<I1, R8>(Convert);
             AddAux<I1, SB>(Convert);
+            AddStd<I1, BL>(Convert);
 
             AddStd<I2, I1>(Convert);
             AddStd<I2, I2>(Convert);
@@ -119,6 +120,7 @@ private Conversions()
             AddStd<I2, R4>(Convert);
             AddStd<I2, R8>(Convert);
             AddAux<I2, SB>(Convert);
+            AddStd<I2, BL>(Convert);
 
             AddStd<I4, I1>(Convert);
             AddStd<I4, I2>(Convert);
@@ -127,6 +129,7 @@ private Conversions()
             AddStd<I4, R4>(Convert);
             AddStd<I4, R8>(Convert);
             AddAux<I4, SB>(Convert);
+            AddStd<I4, BL>(Convert);
 
             AddStd<I8, I1>(Convert);
             AddStd<I8, I2>(Convert);
@@ -135,6 +138,7 @@ private Conversions()
             AddStd<I8, R4>(Convert);
             AddStd<I8, R8>(Convert);
             AddAux<I8, SB>(Convert);
+            AddStd<I8, BL>(Convert);
 
             AddStd<U1, U1>(Convert);
             AddStd<U1, U2>(Convert);
@@ -144,6 +148,7 @@ private Conversions()
             AddStd<U1, R4>(Convert);
             AddStd<U1, R8>(Convert);
             AddAux<U1, SB>(Convert);
+            AddStd<U1, BL>(Convert);
 
             AddStd<U2, U1>(Convert);
             AddStd<U2, U2>(Convert);
@@ -153,6 +158,7 @@ private Conversions()
             AddStd<U2, R4>(Convert);
             AddStd<U2, R8>(Convert);
             AddAux<U2, SB>(Convert);
+            AddStd<U2, BL>(Convert);
 
             AddStd<U4, U1>(Convert);
             AddStd<U4, U2>(Convert);
@@ -162,6 +168,7 @@ private Conversions()
             AddStd<U4, R4>(Convert);
             AddStd<U4, R8>(Convert);
             AddAux<U4, SB>(Convert);
+            AddStd<U4, BL>(Convert);
 
             AddStd<U8, U1>(Convert);
             AddStd<U8, U2>(Convert);
@@ -171,6 +178,7 @@ private Conversions()
             AddStd<U8, R4>(Convert);
             AddStd<U8, R8>(Convert);
             AddAux<U8, SB>(Convert);
+            AddStd<U8, BL>(Convert);
 
             AddStd<UG, U1>(Convert);
             AddStd<UG, U2>(Convert);
@@ -180,11 +188,13 @@ private Conversions()
             AddAux<UG, SB>(Convert);
 
             AddStd<R4, R4>(Convert);
+            AddStd<R4, BL>(Convert);
             AddStd<R4, R8>(Convert);
             AddAux<R4, SB>(Convert);
 
             AddStd<R8, R4>(Convert);
             AddStd<R8, R8>(Convert);
+            AddStd<R8, BL>(Convert);
             AddAux<R8, SB>(Convert);
 
             AddStd<TX, I1>(Convert);
@@ -901,6 +911,19 @@ public void Convert(in BL src, ref SB dst)
         public void Convert(in DZ src, ref SB dst) { ClearDst(ref dst); dst.AppendFormat("{0:o}", src); }
         #endregion ToStringBuilder
 
+        #region ToBL
+        public void Convert(in R8 src, ref BL dst) => dst = System.Convert.ToBoolean(src);
+        public void Convert(in R4 src, ref BL dst) => dst = System.Convert.ToBoolean(src);
+        public void Convert(in I1 src, ref BL dst) => dst = System.Convert.ToBoolean(src);
+        public void Convert(in I2 src, ref BL dst) => dst = System.Convert.ToBoolean(src);
+        public void Convert(in I4 src, ref BL dst) => dst = System.Convert.ToBoolean(src);
+        public void Convert(in I8 src, ref BL dst) => dst = System.Convert.ToBoolean(src);
+        public void Convert(in U1 src, ref BL dst) => dst = System.Convert.ToBoolean(src);
+        public void Convert(in U2 src, ref BL dst) => dst = System.Convert.ToBoolean(src);
+        public void Convert(in U4 src, ref BL dst) => dst = System.Convert.ToBoolean(src);
+        public void Convert(in U8 src, ref BL dst) => dst = System.Convert.ToBoolean(src);
+        #endregion
+
         #region FromR4
         public void Convert(in R4 src, ref R4 dst) => dst = src;
         public void Convert(in R4 src, ref R8 dst) => dst = src;
@@ -1139,7 +1162,7 @@ private bool TryParseCore(ReadOnlySpan<char> span, out ulong dst)
             dst = res;
             return true;
 
-            LFail:
+        LFail:
             dst = 0;
             return false;
         }
@@ -1246,7 +1269,7 @@ private bool TryParseNonNegative(ReadOnlySpan<char> span, out long result)
             result = res;
             return true;
 
-            LFail:
+        LFail:
             result = 0;
             return false;
         }

src/Microsoft.ML.Data/Scorers/MultiClassClassifierScorer.cs

@@ -429,7 +429,7 @@ internal static bool CanWrap(ISchemaBoundMapper mapper, DataViewType labelNameTy
             var scoreType = outSchema[scoreIdx].Type;
 
             // Check that the type is vector, and is of compatible size with the score output.
-            return labelNameType is VectorType vectorType && vectorType.Size == scoreType.GetVectorSize();
+            return labelNameType is VectorType vectorType && vectorType.Size == scoreType.GetVectorSize() && vectorType.ItemType == TextDataViewType.Instance;
         }
 
         internal static ISchemaBoundMapper WrapCore<T>(IHostEnvironment env, ISchemaBoundMapper mapper, RoleMappedSchema trainSchema)

src/Microsoft.ML.Data/Training/TrainerEstimatorBase.cs

@@ -145,8 +145,15 @@ private protected virtual void CheckLabelCompatible(SchemaShape.Column labelCol)
         private protected TTransformer TrainTransformer(IDataView trainSet,
             IDataView validationSet = null, IPredictor initPredictor = null)
         {
+            CheckInputSchema(SchemaShape.Create(trainSet.Schema));
             var trainRoleMapped = MakeRoles(trainSet);
-            var validRoleMapped = validationSet == null ? null : MakeRoles(validationSet);
+            RoleMappedData validRoleMapped = null;
+
+            if (validationSet != null)
+            {
+                CheckInputSchema(SchemaShape.Create(validationSet.Schema));
+                validRoleMapped = MakeRoles(validationSet);
+            }
 
             var pred = TrainModelCore(new TrainContext(trainRoleMapped, validRoleMapped, null, initPredictor));
             return MakeTransformer(pred, trainSet.Schema);

src/Microsoft.ML.FastTree/FastTreeRanking.cs

@@ -105,6 +105,7 @@ private protected override void CheckLabelCompatible(SchemaShape.Column labelCol
             if (!labelCol.IsKey && labelCol.ItemType != NumberDataViewType.Single)
                 error();
         }
+
         private protected override float GetMaxLabel()
         {
             return GetLabelGains().Length - 1;

src/Microsoft.ML.StandardTrainers/Standard/MultiClass/MetaMulticlassTrainer.cs

@@ -8,10 +8,9 @@
 using Microsoft.ML.Calibrators;
 using Microsoft.ML.CommandLine;
 using Microsoft.ML.Data;
-using Microsoft.ML.Data.Conversion;
 using Microsoft.ML.Internal.Internallearn;
 using Microsoft.ML.Runtime;
-
+using Microsoft.ML.Transforms;
 namespace Microsoft.ML.Trainers
 {
     using TScalarTrainer = ITrainerEstimator<ISingleFeaturePredictionTransformer<IPredictorProducing<float>>, IPredictorProducing<float>>;
@@ -32,7 +31,7 @@ internal abstract class OptionsBase
             [Argument(ArgumentType.LastOccurenceWins, HelpText = "Number of instances to train the calibrator", SortOrder = 150, ShortName = "numcali")]
             internal int MaxCalibrationExamples = 1000000000;
 
-            [Argument(ArgumentType.Multiple, HelpText = "Whether to treat missing labels as having negative labels, instead of keeping them missing", SortOrder = 150, ShortName = "missNeg")]
+            [Argument(ArgumentType.Multiple, HelpText = "Whether to treat missing labels as having negative labels, or exclude their rows from dataview.", SortOrder = 150, ShortName = "missNeg")]
             public bool ImputeMissingLabelsAsNegative;
         }
 
@@ -98,20 +97,15 @@ private protected IDataView MapLabelsCore<T>(DataViewType type, InPredicate<T> e
             Host.AssertValue(data);
             Host.Assert(data.Schema.Label.HasValue);
 
-            var lab = data.Schema.Label.Value;
+            var label = data.Schema.Label.Value;
+            IDataView dataView = data.Data;
+            if (!Args.ImputeMissingLabelsAsNegative)
+                dataView = new NAFilter(Host, data.Data, false, label.Name);
 
-            InPredicate<T> isMissing;
-            if (!Args.ImputeMissingLabelsAsNegative && Conversions.Instance.TryGetIsNAPredicate(type, out isMissing))
-            {
-                return LambdaColumnMapper.Create(Host, "Label mapper", data.Data,
-                    lab.Name, lab.Name, type, NumberDataViewType.Single,
-                    (in T src, ref float dst) =>
-                        dst = equalsTarget(in src) ? 1 : (isMissing(in src) ? float.NaN : default(float)));
-            }
             return LambdaColumnMapper.Create(Host, "Label mapper", data.Data,
-                lab.Name, lab.Name, type, NumberDataViewType.Single,
-                (in T src, ref float dst) =>
-                    dst = equalsTarget(in src) ? 1 : default(float));
+                label.Name, label.Name, type, BooleanDataViewType.Instance,
+                (in T src, ref bool dst) =>
+                    dst = equalsTarget(in src) ? true : false);
         }
 
         private protected abstract TModel TrainCore(IChannel ch, RoleMappedData data, int count);

src/Microsoft.ML.StandardTrainers/Standard/MultiClass/OneVersusAllTrainer.cs

@@ -145,28 +145,18 @@ private ISingleFeaturePredictionTransformer<TScalarPredictor> TrainOne(IChannel
 
         private IDataView MapLabels(RoleMappedData data, int cls)
         {
-            var lab = data.Schema.Label.Value;
-            Host.Assert(!lab.IsHidden);
-            Host.Assert(lab.Type.GetKeyCount() > 0 || lab.Type == NumberDataViewType.Single || lab.Type == NumberDataViewType.Double);
+            var label = data.Schema.Label.Value;
+            Host.Assert(!label.IsHidden);
+            Host.Assert(label.Type.GetKeyCount() > 0 || label.Type == NumberDataViewType.Single || label.Type == NumberDataViewType.Double);
 
-            if (lab.Type.GetKeyCount() > 0)
+            if (label.Type.GetKeyCount() > 0)
             {
                 // Key values are 1-based.
                 uint key = (uint)(cls + 1);
                 return MapLabelsCore(NumberDataViewType.UInt32, (in uint val) => key == val, data);
             }
-            if (lab.Type == NumberDataViewType.Single)
-            {
-                float key = cls;
-                return MapLabelsCore(NumberDataViewType.Single, (in float val) => key == val, data);
-            }
-            if (lab.Type == NumberDataViewType.Double)
-            {
-                double key = cls;
-                return MapLabelsCore(NumberDataViewType.Double, (in double val) => key == val, data);
-            }
 
-            throw Host.ExceptNotSupp($"Label column type is not supported by OneVersusAllTrainer: {lab.Type.RawType}");
+            throw Host.ExceptNotSupp($"Label column type is not supported by OneVersusAllTrainer: {label.Type.RawType}");
         }
 
         /// <summary> Trains a <see cref="MulticlassPredictionTransformer{OneVersusAllModelParameters}"/> model.</summary>

src/Microsoft.ML.StandardTrainers/Standard/MultiClass/PairwiseCouplingTrainer.cs

@@ -142,31 +142,19 @@ private ISingleFeaturePredictionTransformer<TDistPredictor> TrainOne(IChannel ch
 
         private IDataView MapLabels(RoleMappedData data, int cls1, int cls2)
         {
-            var lab = data.Schema.Label.Value;
-            Host.Assert(!lab.IsHidden);
-            Host.Assert(lab.Type.GetKeyCount() > 0 || lab.Type == NumberDataViewType.Single || lab.Type == NumberDataViewType.Double);
+            var label = data.Schema.Label.Value;
+            Host.Assert(!label.IsHidden);
+            Host.Assert(label.Type.GetKeyCount() > 0 || label.Type == NumberDataViewType.Single || label.Type == NumberDataViewType.Double);
 
-            if (lab.Type.GetKeyCount() > 0)
+            if (label.Type.GetKeyCount() > 0)
             {
                 // Key values are 1-based.
                 uint key1 = (uint)(cls1 + 1);
                 uint key2 = (uint)(cls2 + 1);
                 return MapLabelsCore(NumberDataViewType.UInt32, (in uint val) => val == key1 || val == key2, data);
             }
-            if (lab.Type == NumberDataViewType.Single)
-            {
-                float key1 = cls1;
-                float key2 = cls2;
-                return MapLabelsCore(NumberDataViewType.Single, (in float val) => val == key1 || val == key2, data);
-            }
-            if (lab.Type == NumberDataViewType.Double)
-            {
-                double key1 = cls1;
-                double key2 = cls2;
-                return MapLabelsCore(NumberDataViewType.Double, (in double val) => val == key1 || val == key2, data);
-            }
 
-            throw Host.ExceptNotSupp($"Label column type is not supported by nameof(PairwiseCouplingTrainer): {lab.Type.RawType}");
+            throw Host.ExceptNotSupp($"Label column type is not supported by nameof(PairwiseCouplingTrainer): {label.Type.RawType}");
         }
 
         /// <summary>

src/Microsoft.ML.StandardTrainers/Standard/Online/AveragedPerceptron.cs

@@ -174,20 +174,6 @@ private protected override void CheckLabels(RoleMappedData data)
             data.CheckBinaryLabel();
         }
 
-        private protected override void CheckLabelCompatible(SchemaShape.Column labelCol)
-        {
-            Contracts.Assert(labelCol.IsValid);
-
-            Action error =
-                () => throw Host.ExceptSchemaMismatch(nameof(labelCol), "label", labelCol.Name, "float, double, bool or KeyType", labelCol.GetTypeString());
-
-            if (labelCol.Kind != SchemaShape.Column.VectorKind.Scalar)
-                error();
-
-            if (!labelCol.IsKey && labelCol.ItemType != NumberDataViewType.Single && labelCol.ItemType != NumberDataViewType.Double && !(labelCol.ItemType is BooleanDataViewType))
-                error();
-        }
-
         private protected override TrainStateBase MakeState(IChannel ch, int numFeatures, LinearModelParameters predictor)
         {
             return new TrainState(ch, numFeatures, predictor, this);

src/Microsoft.ML.StandardTrainers/Standard/SdcaBinary.cs

@@ -1512,20 +1512,6 @@ private protected SdcaBinaryTrainerBase(IHostEnvironment env, BinaryOptionsBase
 
         private protected abstract SchemaShape.Column[] ComputeSdcaBinaryClassifierSchemaShape();
 
-        private protected override void CheckLabelCompatible(SchemaShape.Column labelCol)
-        {
-            Contracts.Assert(labelCol.IsValid);
-
-            Action error =
-                () => throw Host.ExceptSchemaMismatch(nameof(labelCol), "label", labelCol.Name, "float, double, bool or KeyType", labelCol.GetTypeString());
-
-            if (labelCol.Kind != SchemaShape.Column.VectorKind.Scalar)
-                error();
-
-            if (!labelCol.IsKey && labelCol.ItemType != NumberDataViewType.Single && labelCol.ItemType != NumberDataViewType.Double && !(labelCol.ItemType is BooleanDataViewType))
-                error();
-        }
-
         private protected LinearBinaryModelParameters CreateLinearBinaryModelParameters(VBuffer<float>[] weights, float[] bias)
         {
             Host.CheckParam(Utils.Size(weights) == 1, nameof(weights));

src/Microsoft.ML.StandardTrainers/Standard/SdcaMultiClass.cs

@@ -121,19 +121,6 @@ private protected override SchemaShape.Column[] GetOutputColumnsCore(SchemaShape
             };
         }
 
-        private protected override void CheckLabelCompatible(SchemaShape.Column labelCol)
-        {
-            Contracts.Assert(labelCol.IsValid);
-
-            Action error =
-                () => throw Host.ExceptSchemaMismatch(nameof(labelCol), "label", labelCol.Name, "float, double or KeyType", labelCol.GetTypeString());
-
-            if (labelCol.Kind != SchemaShape.Column.VectorKind.Scalar)
-                error();
-            if (!labelCol.IsKey && labelCol.ItemType != NumberDataViewType.Single && labelCol.ItemType != NumberDataViewType.Double)
-                error();
-        }
-
         /// <inheritdoc/>
         private protected override void TrainWithoutLock(IProgressChannelProvider progress, FloatLabelCursor.Factory cursorFactory, Random rand,
             IdToIdxLookup idToIdx, int numThreads, DualsTableBase duals, float[] biasReg, float[] invariants, float lambdaNInv,

src/Microsoft.ML.Transforms/Text/NgramTransform.cs

@@ -887,7 +887,7 @@ public SchemaShape GetOutputSchema(SchemaShape inputSchema)
                 if (!IsSchemaColumnValid(col))
                     throw _host.ExceptSchemaMismatch(nameof(inputSchema), "input", colInfo.InputColumnName, ExpectedColumnType, col.GetTypeString());
                 var metadata = new List<SchemaShape.Column>();
-                if (col.HasKeyValues())
+                if (col.NeedsSlotNames())
                     metadata.Add(new SchemaShape.Column(AnnotationUtils.Kinds.SlotNames, SchemaShape.Column.VectorKind.Vector, TextDataViewType.Instance, false));
                 result[colInfo.Name] = new SchemaShape.Column(colInfo.Name, SchemaShape.Column.VectorKind.Vector, NumberDataViewType.Single, false, new SchemaShape(metadata));
             }

test/BaselineOutput/Common/FeatureContribution/AveragePerceptronBinary.tsv

@@ -4,14 +4,15 @@
 #@   col=X2VBuffer:R4:1-4
 #@   col=X3Important:R4:5
 #@   col=Label:R4:6
-#@   col=Features:R4:7-12
-#@   col=Features:R4:13-18
-#@   col=FeatureContributions:R4:19-24
-#@   col=FeatureContributions:R4:25-30
-#@   col=FeatureContributions:R4:31-36
-#@   col=FeatureContributions:R4:37-42
+#@   col=Label:BL:7
+#@   col=Features:R4:8-13
+#@   col=Features:R4:14-19
+#@   col=FeatureContributions:R4:20-25
+#@   col=FeatureContributions:R4:26-31
+#@   col=FeatureContributions:R4:32-37
+#@   col=FeatureContributions:R4:38-43
 #@ }
-950	757	692	720	297	7515	1	950	757	692	720	297	7515	0.956696868	0.760804	0.7872582	0.754716933	0.297893673	0.7578661	0	0.1527809	0	0	0	1	-0.6583012	0	-1	-0.517060339	0	0	12	2:-0.13028869	5:1	8:-0.370813	11:2.84608746
-459	961	0	659	274	2147	0	459	961	0	659	274	2147	0.462235659	0.965829134	0	0.690775633	0.27482447	0.21651876	0	0.6788808	0	0	0	0.99999994	-0.6720779	0	0	-1	-0.870772958	0	12	3:-0.215823054	5:0.99999994	9:-0.175488681	11:0.8131137
-672	275	0	65	195	9818	1	672	275	0	65	195	9818	0.6767372	0.2763819	0	0.06813417	0.195586756	0.990116954	0	0.04248268	0	0	0	1	-1	0	0	-0.100242466	-0.6298147	0	12	0:-0.04643902	5:1	6:-0.172673345	11:3.71828127
-186	301	0	681	526	1456	0	186	301	0	681	526	1456	0.187311172	0.302512556	0	0.713836432	0.527582765	0.1468334	0	0.313550383	0	0	0	1	-0.162922	0	0	-0.6181894	-1	0	12	4:-0.5319963	5:1	10:-0.293352127	11:0.5514176
+950	757	692	720	297	7515	1	1	950	757	692	720	297	7515	0.956696868	0.760804	0.7872582	0.754716933	0.297893673	0.7578661	0	0.1527809	0	0	0	1	-0.6583012	0	-1	-0.517060339	0	0	12	2:-0.13028869	5:1	8:-0.370813	11:2.84608746
+459	961	0	659	274	2147	0	0	459	961	0	659	274	2147	0.462235659	0.965829134	0	0.690775633	0.27482447	0.21651876	0	0.6788808	0	0	0	0.99999994	-0.6720779	0	0	-1	-0.870772958	0	12	3:-0.215823054	5:0.99999994	9:-0.175488681	11:0.8131137
+672	275	0	65	195	9818	1	1	672	275	0	65	195	9818	0.6767372	0.2763819	0	0.06813417	0.195586756	0.990116954	0	0.04248268	0	0	0	1	-1	0	0	-0.100242466	-0.6298147	0	12	0:-0.04643902	5:1	6:-0.172673345	11:3.71828127
+186	301	0	681	526	1456	0	0	186	301	0	681	526	1456	0.187311172	0.302512556	0	0.713836432	0.527582765	0.1468334	0	0.313550383	0	0	0	1	-0.162922	0	0	-0.6181894	-1	0	12	4:-0.5319963	5:1	10:-0.293352127	11:0.5514176