Replaces ChooseColumnsTransform and DropColumnsTransform with SelectColumnsTransform (#1371)

* Removes ChooseColumnsTransform and DropColumnsTransform classes
replacing them with SelectColumnsTransform. These changes include:
* Updates to SelectColumnsTransform to respect ordering when keeping
columns. For example, if the input is ABC and CB is selected, the output
will be CB.
* Updates to code that used Choose or Drop columns, replacing with
SelectColumns.
* Updates to baseline output for tests to pass
* Re-enabled the SavePipeline tests

This fixes #1342
These changes are also related to #754

Author: singlis

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

@@ -130,11 +130,10 @@ private void RunCore(IChannel ch)
 
             if (!string.IsNullOrWhiteSpace(Args.Columns))
             {
-                var args = new ChooseColumnsTransform.Arguments();
-                args.Column = Args.Columns
-                    .Split(new char[] { ',' }, StringSplitOptions.RemoveEmptyEntries).Select(s => new ChooseColumnsTransform.Column() { Name = s }).ToArray();
-                if (Utils.Size(args.Column) > 0)
-                    data = new ChooseColumnsTransform(Host, args, data);
+                var keepColumns = Args.Columns
+                    .Split(new char[] { ',' }, StringSplitOptions.RemoveEmptyEntries).ToArray();
+                if (Utils.Size(keepColumns) > 0)
+                    data = SelectColumnsTransform.CreateKeep(Host, data, keepColumns);
             }
 
             IDataSaver saver;

src/Microsoft.ML.Data/EntryPoints/ScoreColumnSelector.cs

@@ -101,7 +101,7 @@ public static CommonOutputs.TransformOutput RenameBinaryPredictionScoreColumns(I
                     }
 
                     var copyColumn = new CopyColumnsTransform(env, copyCols.ToArray()).Transform(input.Data);
-                    var dropColumn = new DropColumnsTransform(env, new DropColumnsTransform.Arguments() { Column = copyCols.Select(c => c.Source).ToArray() }, copyColumn);
+                    var dropColumn = SelectColumnsTransform.CreateDrop(env, copyColumn, copyCols.Select(c => c.Source).ToArray());
                     return new CommonOutputs.TransformOutput { Model = new TransformModel(env, dropColumn, input.Data), OutputData = dropColumn };
                 }
             }

src/Microsoft.ML.Data/Evaluators/AnomalyDetectionEvaluator.cs

@@ -703,59 +703,42 @@ protected override void PrintFoldResultsCore(IChannel ch, Dictionary<string, IDa
                 }
             }
 
-            var args = new ChooseColumnsTransform.Arguments();
-            var cols = new List<ChooseColumnsTransform.Column>()
-                {
-                    new ChooseColumnsTransform.Column()
-                    {
-                        Name = string.Format(FoldDrAtKFormat, _k),
-                        Source = AnomalyDetectionEvaluator.OverallMetrics.DrAtK
-                    },
-                    new ChooseColumnsTransform.Column()
-                    {
-                        Name = string.Format(FoldDrAtPFormat, _p),
-                        Source = AnomalyDetectionEvaluator.OverallMetrics.DrAtPFpr
-                    },
-                    new ChooseColumnsTransform.Column()
-                    {
-                        Name = string.Format(FoldDrAtNumAnomaliesFormat, numAnomalies),
-                        Source=AnomalyDetectionEvaluator.OverallMetrics.DrAtNumPos
-                    },
-                    new ChooseColumnsTransform.Column()
-                    {
-                        Name=AnomalyDetectionEvaluator.OverallMetrics.ThreshAtK
-                    },
-                    new ChooseColumnsTransform.Column()
-                    {
-                        Name=AnomalyDetectionEvaluator.OverallMetrics.ThreshAtP
-                    },
-                    new ChooseColumnsTransform.Column()
-                    {
-                        Name=AnomalyDetectionEvaluator.OverallMetrics.ThreshAtNumPos
-                    },
-                    new ChooseColumnsTransform.Column()
-                    {
-                        Name = BinaryClassifierEvaluator.Auc
-                    }
-                };
+            var kFormatName = string.Format(FoldDrAtKFormat, _k);
+            var pFormatName = string.Format(FoldDrAtPFormat, _p);
+            var numAnomName = string.Format(FoldDrAtNumAnomaliesFormat, numAnomalies);
+
+            (string Source, string Name)[] cols =
+            {
+                (AnomalyDetectionEvaluator.OverallMetrics.DrAtK, kFormatName),
+                (AnomalyDetectionEvaluator.OverallMetrics.DrAtPFpr, pFormatName),
+                (AnomalyDetectionEvaluator.OverallMetrics.DrAtNumPos, numAnomName)
+            };
+
+            // List of columns to keep, note that the order specified determines the order of the output
+            var colsToKeep = new List<string>();
+            colsToKeep.Add(kFormatName);
+            colsToKeep.Add(pFormatName);
+            colsToKeep.Add(numAnomName);
+            colsToKeep.Add(AnomalyDetectionEvaluator.OverallMetrics.ThreshAtK);
+            colsToKeep.Add(AnomalyDetectionEvaluator.OverallMetrics.ThreshAtP);
+            colsToKeep.Add(AnomalyDetectionEvaluator.OverallMetrics.ThreshAtNumPos);
+            colsToKeep.Add(BinaryClassifierEvaluator.Auc);
+
+            overall = new CopyColumnsTransform(Host, cols).Transform(overall);
+            IDataView fold = SelectColumnsTransform.CreateKeep(Host, overall, colsToKeep.ToArray());
 
-            args.Column = cols.ToArray();
-            IDataView fold = new ChooseColumnsTransform(Host, args, overall);
             string weightedFold;
             ch.Info(MetricWriter.GetPerFoldResults(Host, fold, out weightedFold));
         }
 
         protected override IDataView GetOverallResultsCore(IDataView overall)
         {
-            var args = new DropColumnsTransform.Arguments();
-            args.Column = new[]
-            {
-                AnomalyDetectionEvaluator.OverallMetrics.NumAnomalies,
-                AnomalyDetectionEvaluator.OverallMetrics.ThreshAtK,
-                AnomalyDetectionEvaluator.OverallMetrics.ThreshAtP,
-                AnomalyDetectionEvaluator.OverallMetrics.ThreshAtNumPos
-            };
-            return new DropColumnsTransform(Host, args, overall);
+            return SelectColumnsTransform.CreateDrop(Host,
+                                                    overall,
+                                                    AnomalyDetectionEvaluator.OverallMetrics.NumAnomalies,
+                                                    AnomalyDetectionEvaluator.OverallMetrics.ThreshAtK,
+                                                    AnomalyDetectionEvaluator.OverallMetrics.ThreshAtP,
+                                                    AnomalyDetectionEvaluator.OverallMetrics.ThreshAtNumPos);
         }
 
         protected override IEnumerable<string> GetPerInstanceColumnsToSave(RoleMappedSchema schema)

src/Microsoft.ML.Data/Evaluators/BinaryClassifierEvaluator.cs

@@ -1333,43 +1333,33 @@ protected override void PrintFoldResultsCore(IChannel ch, Dictionary<string, IDa
             if (!metrics.TryGetValue(MetricKinds.ConfusionMatrix, out conf))
                 throw ch.Except("No overall metrics found");
 
-            var args = new ChooseColumnsTransform.Arguments();
-            var cols = new List<ChooseColumnsTransform.Column>()
-                {
-                    new ChooseColumnsTransform.Column()
-                    {
-                        Name = FoldAccuracy,
-                        Source = BinaryClassifierEvaluator.Accuracy
-                    },
-                    new ChooseColumnsTransform.Column()
-                    {
-                        Name = FoldLogLoss,
-                        Source = BinaryClassifierEvaluator.LogLoss
-                    },
-                    new ChooseColumnsTransform.Column()
-                    {
-                        Name = BinaryClassifierEvaluator.Entropy
-                    },
-                    new ChooseColumnsTransform.Column()
-                    {
-                        Name = FoldLogLosRed,
-                        Source = BinaryClassifierEvaluator.LogLossReduction
-                    },
-                    new ChooseColumnsTransform.Column()
-                    {
-                        Name = BinaryClassifierEvaluator.Auc
-                    }
-                };
+            (string Source, string Name)[] cols =
+            {
+                (BinaryClassifierEvaluator.Accuracy, FoldAccuracy),
+                (BinaryClassifierEvaluator.LogLoss, FoldLogLoss),
+                (BinaryClassifierEvaluator.LogLossReduction, FoldLogLosRed)
+            };
+
+            var colsToKeep = new List<string>();
+            colsToKeep.Add(FoldAccuracy);
+            colsToKeep.Add(FoldLogLoss);
+            colsToKeep.Add(BinaryClassifierEvaluator.Entropy);
+            colsToKeep.Add(FoldLogLosRed);
+            colsToKeep.Add(BinaryClassifierEvaluator.Auc);
+
             int index;
             if (fold.Schema.TryGetColumnIndex(MetricKinds.ColumnNames.IsWeighted, out index))
-                cols.Add(new ChooseColumnsTransform.Column() { Name = MetricKinds.ColumnNames.IsWeighted });
+                colsToKeep.Add(MetricKinds.ColumnNames.IsWeighted);
             if (fold.Schema.TryGetColumnIndex(MetricKinds.ColumnNames.StratCol, out index))
-                cols.Add(new ChooseColumnsTransform.Column() { Name = MetricKinds.ColumnNames.StratCol });
+                colsToKeep.Add(MetricKinds.ColumnNames.StratCol);
             if (fold.Schema.TryGetColumnIndex(MetricKinds.ColumnNames.StratVal, out index))
-                cols.Add(new ChooseColumnsTransform.Column() { Name = MetricKinds.ColumnNames.StratVal });
+                colsToKeep.Add(MetricKinds.ColumnNames.StratVal);
+
+            fold = new CopyColumnsTransform(Host, cols).Transform(fold);
+
+            // Select the columns that are specified in the Copy
+            fold = SelectColumnsTransform.CreateKeep(Host, fold, colsToKeep.ToArray());
 
-            args.Column = cols.ToArray();
-            fold = new ChooseColumnsTransform(Host, args, fold);
             string weightedConf;
             var unweightedConf = MetricWriter.GetConfusionTable(Host, conf, out weightedConf);
             string weightedFold;
@@ -1386,9 +1376,7 @@ protected override void PrintFoldResultsCore(IChannel ch, Dictionary<string, IDa
 
         protected override IDataView GetOverallResultsCore(IDataView overall)
         {
-            var args = new DropColumnsTransform.Arguments();
-            args.Column = new[] { BinaryClassifierEvaluator.Entropy };
-            return new DropColumnsTransform(Host, args, overall);
+            return SelectColumnsTransform.CreateDrop(Host, overall, BinaryClassifierEvaluator.Entropy);
         }
 
         protected override void PrintAdditionalMetricsCore(IChannel ch, Dictionary<string, IDataView>[] metrics)

src/Microsoft.ML.Data/Evaluators/EvaluatorUtils.cs

@@ -933,7 +933,7 @@ private static IDataView AppendPerInstanceDataViews(IHostEnvironment env, string
                                  variableSizeVectorColumnName, type);
 
                         // Drop the old column that does not have variable length.
-                        idv = new DropColumnsTransform(env, new DropColumnsTransform.Arguments() { Column = new[] { variableSizeVectorColumnName } }, idv);
+                        idv = SelectColumnsTransform.CreateDrop(env, idv, variableSizeVectorColumnName);
                     }
                     return idv;
                 };
@@ -1059,8 +1059,7 @@ internal static IDataView GetOverallMetricsData(IHostEnvironment env, IDataView
             {
                 if (Utils.Size(nonAveragedCols) > 0)
                 {
-                    var dropArgs = new DropColumnsTransform.Arguments() { Column = nonAveragedCols.ToArray() };
-                    data = new DropColumnsTransform(env, dropArgs, data);
+                    data = SelectColumnsTransform.CreateDrop(env, data, nonAveragedCols.ToArray());
                 }
                 idvList.Add(data);
             }
@@ -1734,9 +1733,7 @@ public static IDataView GetNonStratifiedMetrics(IHostEnvironment env, IDataView
             var found = data.Schema.TryGetColumnIndex(MetricKinds.ColumnNames.StratVal, out stratVal);
             env.Check(found, "If stratification column exist, data view must also contain a StratVal column");
 
-            var dropArgs = new DropColumnsTransform.Arguments();
-            dropArgs.Column = new[] { data.Schema.GetColumnName(stratCol), data.Schema.GetColumnName(stratVal) };
-            data = new DropColumnsTransform(env, dropArgs, data);
+            data = SelectColumnsTransform.CreateDrop(env, data, data.Schema.GetColumnName(stratCol), data.Schema.GetColumnName(stratVal));
             return data;
         }
     }

src/Microsoft.ML.Data/Evaluators/MamlEvaluator.cs

@@ -213,13 +213,14 @@ private IDataView WrapPerInstance(RoleMappedData perInst)
             var idv = perInst.Data;
 
             // Make a list of column names that Maml outputs as part of the per-instance data view, and then wrap
-            // the per-instance data computed by the evaluator in a ChooseColumnsTransform.
-            var cols = new List<ChooseColumnsTransform.Column>();
+            // the per-instance data computed by the evaluator in a SelectColumnsTransform.
+            var cols = new List<(string Source, string Name)>();
+            var colsToKeep = new List<string>();
 
             // If perInst is the result of cross-validation and contains a fold Id column, include it.
             int foldCol;
             if (perInst.Schema.Schema.TryGetColumnIndex(MetricKinds.ColumnNames.FoldIndex, out foldCol))
-                cols.Add(new ChooseColumnsTransform.Column() { Source = MetricKinds.ColumnNames.FoldIndex });
+                colsToKeep.Add(MetricKinds.ColumnNames.FoldIndex);
 
             // Maml always outputs a name column, if it doesn't exist add a GenerateNumberTransform.
             if (perInst.Schema.Name == null)
@@ -228,22 +229,24 @@ private IDataView WrapPerInstance(RoleMappedData perInst)
                 args.Column = new[] { new GenerateNumberTransform.Column() { Name = "Instance" } };
                 args.UseCounter = true;
                 idv = new GenerateNumberTransform(Host, args, idv);
-                cols.Add(new ChooseColumnsTransform.Column() { Name = "Instance" });
+                colsToKeep.Add("Instance");
             }
             else
-                cols.Add(new ChooseColumnsTransform.Column() { Source = perInst.Schema.Name.Name, Name = "Instance" });
+            {
+                cols.Add((perInst.Schema.Name.Name, "Instance"));
+                colsToKeep.Add("Instance");
+            }
 
             // Maml outputs the weight column if it exists.
             if (perInst.Schema.Weight != null)
-                cols.Add(new ChooseColumnsTransform.Column() { Name = perInst.Schema.Weight.Name });
+                colsToKeep.Add(perInst.Schema.Weight.Name);
 
             // Get the other columns from the evaluator.
             foreach (var col in GetPerInstanceColumnsToSave(perInst.Schema))
-                cols.Add(new ChooseColumnsTransform.Column() { Name = col });
+                colsToKeep.Add(col);
 
-            var chooseArgs = new ChooseColumnsTransform.Arguments();
-            chooseArgs.Column = cols.ToArray();
-            idv = new ChooseColumnsTransform(Host, chooseArgs, idv);
+            idv = new CopyColumnsTransform(Host, cols.ToArray()).Transform(idv);
+            idv = SelectColumnsTransform.CreateKeep(Host, idv, colsToKeep.ToArray());
             return GetPerInstanceMetricsCore(idv, perInst.Schema);
         }
 

src/Microsoft.ML.Data/Evaluators/MulticlassClassifierEvaluator.cs

@@ -1051,22 +1051,14 @@ protected override IDataView GetOverallResultsCore(IDataView overall)
         private IDataView ChangeTopKAccColumnName(IDataView input)
         {
             input = new CopyColumnsTransform(Host, (MultiClassClassifierEvaluator.TopKAccuracy, string.Format(TopKAccuracyFormat, _outputTopKAcc))).Transform(input);
-            var dropArgs = new DropColumnsTransform.Arguments
-            {
-                Column = new[] { MultiClassClassifierEvaluator.TopKAccuracy }
-            };
-            return new DropColumnsTransform(Host, dropArgs, input);
+            return SelectColumnsTransform.CreateDrop(Host, input, MultiClassClassifierEvaluator.TopKAccuracy );
         }
 
         private IDataView DropPerClassColumn(IDataView input)
         {
             if (input.Schema.TryGetColumnIndex(MultiClassClassifierEvaluator.PerClassLogLoss, out int perClassCol))
             {
-                var args = new DropColumnsTransform.Arguments
-                {
-                    Column = new[] { MultiClassClassifierEvaluator.PerClassLogLoss }
-                };
-                input = new DropColumnsTransform(Host, args, input);
+                input = SelectColumnsTransform.CreateDrop(Host, input, MultiClassClassifierEvaluator.PerClassLogLoss);
             }
             return input;
         }