no need to add combiner if you don't have transforms. (#172)

* no need to add combiner if you don't have transforms.

* fix NextSigned

Author: Ivanidzo4ka

src/Microsoft.ML.Core/Utilities/Random.cs

@@ -151,7 +151,7 @@ public int NextSigned()
         {
             // Note that, according to the documentation for System.Random,
             // this won't ever achieve int.MaxValue, but oh well.
-            return _rnd.Next(int.MinValue, int.MinValue);
+            return _rnd.Next(int.MinValue, int.MaxValue);
         }
     }
 

src/Microsoft.ML/LearningPipeline.cs

@@ -154,23 +154,29 @@ public PredictionModel<TInput, TOutput> Train<TInput, TOutput>()
                     step = currentItem.ApplyStep(step, experiment);
                     if (step is ILearningPipelineDataStep dataStep && dataStep.Model != null)
                         transformModels.Add(dataStep.Model);
-                     
+
                     else if (step is ILearningPipelinePredictorStep predictorDataStep)
                     {
                         if (lastTransformModel != null)
                             transformModels.Insert(0, lastTransformModel);
 
-                        var localModelInput = new Transforms.ManyHeterogeneousModelCombiner
+                        Var<IPredictorModel> predictorModel;
+                        if (transformModels.Count != 0)
                         {
-                            PredictorModel = predictorDataStep.Model,
-                            TransformModels = new ArrayVar<ITransformModel>(transformModels.ToArray())
-                        };
-
-                        var localModelOutput = experiment.Add(localModelInput);
+                            var localModelInput = new Transforms.ManyHeterogeneousModelCombiner
+                            {
+                                PredictorModel = predictorDataStep.Model,
+                                TransformModels = new ArrayVar<ITransformModel>(transformModels.ToArray())
+                            };
+                            var localModelOutput = experiment.Add(localModelInput);
+                            predictorModel = localModelOutput.PredictorModel;
+                        }
+                        else
+                            predictorModel = predictorDataStep.Model;
 
                         var scorer = new Transforms.Scorer
                         {
-                            PredictorModel = localModelOutput.PredictorModel
+                            PredictorModel = predictorModel
                         };
 
                         var scorerOutput = experiment.Add(scorer);

test/Microsoft.ML.Tests/LearningPipelineTests.cs

@@ -3,8 +3,10 @@
 // See the LICENSE file in the project root for more information.
 
 using Microsoft.ML;
+using Microsoft.ML.Data;
 using Microsoft.ML.Runtime.Api;
 using Microsoft.ML.TestFramework;
+using Microsoft.ML.Trainers;
 using Microsoft.ML.Transforms;
 using System.Linq;
 using Xunit;
@@ -79,5 +81,34 @@ public void TransformOnlyPipeline()
                 else
                     Assert.Equal(0, predictionModel.TransformedF1[index]);
         }
+
+        public class Data
+        {
+            [ColumnName("Features")]
+            [VectorType(2)]
+            public float[] Features;
+
+            [ColumnName("Label")]
+            public float Label;
+        }
+
+        public class Prediction
+        {
+            [ColumnName("PredictedLabel")]
+            public bool PredictedLabel;
+        }
+
+        [Fact]
+        public void NoTransformPipeline()
+        {
+            var data = new Data[1];
+            data[0] = new Data();
+            data[0].Features = new float[] { 0.0f, 1.0f };
+            data[0].Label = 0f;
+            var pipeline = new LearningPipeline();
+            pipeline.Add(CollectionDataSource.Create(data));
+            pipeline.Add(new FastForestBinaryClassifier());
+            var model = pipeline.Train<Data, Prediction>();
+        }
     }
 }