Generate built-in SweepableEstimator classes for all available estimators

src/Microsoft.ML.AutoML/API/AutoCatalog.cs

@@ -3,6 +3,10 @@
 // See the LICENSE file in the project root for more information.
 
 using System;
+using System.Collections;
+using System.Collections.Generic;
+using System.Diagnostics.Contracts;
+using Microsoft.ML.AutoML.CodeGen;
 using Microsoft.ML.Data;
 using Microsoft.ML.SearchSpace;
 
@@ -289,5 +293,170 @@ internal SweepableEstimator CreateSweepableEstimator<T>(Func<MLContext, T, IEsti
         {
             return new SweepableEstimator((MLContext context, Parameter param) => factory(context, param.AsType<T>()), ss);
         }
+
+        internal SweepableEstimator[] BinaryClassification(string labelColumnName = DefaultColumnNames.Label, string featureColumnName = DefaultColumnNames.Features, string exampleWeightColumnName = null, bool useFastForest = true, bool useLgbm = true, bool useFastTree = true, bool useLbfgs = true, bool useSdca = true,
+            FastTreeOption fastTreeOption = null, LgbmOption lgbmOption = null, FastForestOption fastForestOption = null, LbfgsOption lbfgsOption = null, SdcaOption sdcaOption = null,
+            SearchSpace<FastTreeOption> fastTreeSearchSpace = null, SearchSpace<LgbmOption> lgbmSearchSpace = null, SearchSpace<FastForestOption> fastForestSearchSpace = null, SearchSpace<LbfgsOption> lbfgsSearchSpace = null, SearchSpace<SdcaOption> sdcaSearchSpace = null)
+        {
+            var res = new List<SweepableEstimator>();
+
+            if (useFastTree)
+            {
+                fastTreeOption = fastTreeOption ?? new FastTreeOption();
+                fastTreeOption.LabelColumnName = labelColumnName;
+                fastTreeOption.FeatureColumnName = featureColumnName;
+                fastTreeOption.ExampleWeightColumnName = exampleWeightColumnName;
+                res.Add(SweepableEstimatorFactory.CreateFastTreeBinary(fastTreeOption, fastTreeSearchSpace ?? new SearchSpace<FastTreeOption>()));
+            }
+
+            if (useFastForest)
+            {
+                fastForestOption = fastForestOption ?? new FastForestOption();
+                fastForestOption.LabelColumnName = labelColumnName;
+                fastForestOption.FeatureColumnName = featureColumnName;
+                fastForestOption.ExampleWeightColumnName = exampleWeightColumnName;
+                res.Add(SweepableEstimatorFactory.CreateFastForestBinary(fastForestOption, fastForestSearchSpace ?? new SearchSpace<FastForestOption>()));
+            }
+
+            if (useLgbm)
+            {
+                lgbmOption = lgbmOption ?? new LgbmOption();
+                lgbmOption.LabelColumnName = labelColumnName;
+                lgbmOption.FeatureColumnName = featureColumnName;
+                lgbmOption.ExampleWeightColumnName = exampleWeightColumnName;
+                res.Add(SweepableEstimatorFactory.CreateLightGbmBinary(lgbmOption, lgbmSearchSpace ?? new SearchSpace<LgbmOption>()));
+            }
+
+            if (useLbfgs)
+            {
+                lbfgsOption = lbfgsOption ?? new LbfgsOption();
+                lbfgsOption.LabelColumnName = labelColumnName;
+                lbfgsOption.FeatureColumnName = featureColumnName;
+                lbfgsOption.ExampleWeightColumnName = exampleWeightColumnName;
+                res.Add(SweepableEstimatorFactory.CreateLbfgsLogisticRegressionBinary(lbfgsOption, lbfgsSearchSpace ?? new SearchSpace<LbfgsOption>()));
+            }
+
+            if (useSdca)
+            {
+                sdcaOption = sdcaOption ?? new SdcaOption();
+                sdcaOption.LabelColumnName = labelColumnName;
+                sdcaOption.FeatureColumnName = featureColumnName;
+                sdcaOption.ExampleWeightColumnName = exampleWeightColumnName;
+                res.Add(SweepableEstimatorFactory.CreateSdcaLogisticRegressionBinary(sdcaOption, sdcaSearchSpace ?? new SearchSpace<SdcaOption>()));
+            }
+
+            return res.ToArray();
+        }
+
+        internal SweepableEstimator[] MultiClassification(string labelColumnName = DefaultColumnNames.Label, string featureColumnName = DefaultColumnNames.Features, string exampleWeightColumnName = null, bool useFastForest = true, bool useLgbm = true, bool useFastTree = true, bool useLbfgs = true, bool useSdca = true,
+            FastTreeOption fastTreeOption = null, LgbmOption lgbmOption = null, FastForestOption fastForestOption = null, LbfgsOption lbfgsOption = null, SdcaOption sdcaOption = null,
+            SearchSpace<FastTreeOption> fastTreeSearchSpace = null, SearchSpace<LgbmOption> lgbmSearchSpace = null, SearchSpace<FastForestOption> fastForestSearchSpace = null, SearchSpace<LbfgsOption> lbfgsSearchSpace = null, SearchSpace<SdcaOption> sdcaSearchSpace = null)
+        {
+            var res = new List<SweepableEstimator>();
+
+            if (useFastTree)
+            {
+                fastTreeOption = fastTreeOption ?? new FastTreeOption();
+                fastTreeOption.LabelColumnName = labelColumnName;
+                fastTreeOption.FeatureColumnName = featureColumnName;
+                fastTreeOption.ExampleWeightColumnName = exampleWeightColumnName;
+                res.Add(SweepableEstimatorFactory.CreateFastTreeOva(fastTreeOption, fastTreeSearchSpace ?? new SearchSpace<FastTreeOption>()));
+            }
+
+            if (useFastForest)
+            {
+                fastForestOption = fastForestOption ?? new FastForestOption();
+                fastForestOption.LabelColumnName = labelColumnName;
+                fastForestOption.FeatureColumnName = featureColumnName;
+                fastForestOption.ExampleWeightColumnName = exampleWeightColumnName;
+                res.Add(SweepableEstimatorFactory.CreateFastForestOva(fastForestOption, fastForestSearchSpace ?? new SearchSpace<FastForestOption>()));
+            }
+
+            if (useLgbm)
+            {
+                lgbmOption = lgbmOption ?? new LgbmOption();
+                lgbmOption.LabelColumnName = labelColumnName;
+                lgbmOption.FeatureColumnName = featureColumnName;
+                lgbmOption.ExampleWeightColumnName = exampleWeightColumnName;
+                res.Add(SweepableEstimatorFactory.CreateLightGbmMulti(lgbmOption, lgbmSearchSpace ?? new SearchSpace<LgbmOption>()));
+            }
+
+            if (useLbfgs)
+            {
+                lbfgsOption = lbfgsOption ?? new LbfgsOption();
+                lbfgsOption.LabelColumnName = labelColumnName;
+                lbfgsOption.FeatureColumnName = featureColumnName;
+                lbfgsOption.ExampleWeightColumnName = exampleWeightColumnName;
+                res.Add(SweepableEstimatorFactory.CreateLbfgsLogisticRegressionOva(lbfgsOption, lbfgsSearchSpace ?? new SearchSpace<LbfgsOption>()));
+                res.Add(SweepableEstimatorFactory.CreateLbfgsMaximumEntropyMulti(lbfgsOption, lbfgsSearchSpace ?? new SearchSpace<LbfgsOption>()));
+            }
+
+            if (useSdca)
+            {
+                sdcaOption = sdcaOption ?? new SdcaOption();
+                sdcaOption.LabelColumnName = labelColumnName;
+                sdcaOption.FeatureColumnName = featureColumnName;
+                sdcaOption.ExampleWeightColumnName = exampleWeightColumnName;
+                res.Add(SweepableEstimatorFactory.CreateSdcaMaximumEntropyMulti(sdcaOption, sdcaSearchSpace ?? new SearchSpace<SdcaOption>()));
+                res.Add(SweepableEstimatorFactory.CreateSdcaLogisticRegressionOva(sdcaOption, sdcaSearchSpace ?? new SearchSpace<SdcaOption>()));
+            }
+
+            return res.ToArray();
+        }
+
+        internal SweepableEstimator[] Regression(string labelColumnName = DefaultColumnNames.Label, string featureColumnName = DefaultColumnNames.Features, string exampleWeightColumnName = null, bool useFastForest = true, bool useLgbm = true, bool useFastTree = true, bool useLbfgs = true, bool useSdca = true,
+            FastTreeOption fastTreeOption = null, LgbmOption lgbmOption = null, FastForestOption fastForestOption = null, LbfgsOption lbfgsOption = null, SdcaOption sdcaOption = null,
+            SearchSpace<FastTreeOption> fastTreeSearchSpace = null, SearchSpace<LgbmOption> lgbmSearchSpace = null, SearchSpace<FastForestOption> fastForestSearchSpace = null, SearchSpace<LbfgsOption> lbfgsSearchSpace = null, SearchSpace<SdcaOption> sdcaSearchSpace = null)
+        {
+            var res = new List<SweepableEstimator>();
+
+            if (useFastTree)
+            {
+                fastTreeOption = fastTreeOption ?? new FastTreeOption();
+                fastTreeOption.LabelColumnName = labelColumnName;
+                fastTreeOption.FeatureColumnName = featureColumnName;
+                fastTreeOption.ExampleWeightColumnName = exampleWeightColumnName;
+                res.Add(SweepableEstimatorFactory.CreateFastTreeRegression(fastTreeOption, fastTreeSearchSpace ?? new SearchSpace<FastTreeOption>()));
+                res.Add(SweepableEstimatorFactory.CreateFastTreeTweedieRegression(fastTreeOption, fastTreeSearchSpace ?? new SearchSpace<FastTreeOption>()));
+            }
+
+            if (useFastForest)
+            {
+                fastForestOption = fastForestOption ?? new FastForestOption();
+                fastForestOption.LabelColumnName = labelColumnName;
+                fastForestOption.FeatureColumnName = featureColumnName;
+                fastForestOption.ExampleWeightColumnName = exampleWeightColumnName;
+                res.Add(SweepableEstimatorFactory.CreateFastForestRegression(fastForestOption, fastForestSearchSpace ?? new SearchSpace<FastForestOption>()));
+            }
+
+            if (useLgbm)
+            {
+                lgbmOption = lgbmOption ?? new LgbmOption();
+                lgbmOption.LabelColumnName = labelColumnName;
+                lgbmOption.FeatureColumnName = featureColumnName;
+                lgbmOption.ExampleWeightColumnName = exampleWeightColumnName;
+                res.Add(SweepableEstimatorFactory.CreateLightGbmRegression(lgbmOption, lgbmSearchSpace ?? new SearchSpace<LgbmOption>()));
+            }
+
+            if (useLbfgs)
+            {
+                lbfgsOption = lbfgsOption ?? new LbfgsOption();
+                lbfgsOption.LabelColumnName = labelColumnName;
+                lbfgsOption.FeatureColumnName = featureColumnName;
+                lbfgsOption.ExampleWeightColumnName = exampleWeightColumnName;
+                res.Add(SweepableEstimatorFactory.CreateLbfgsPoissonRegressionRegression(lbfgsOption, lbfgsSearchSpace ?? new SearchSpace<LbfgsOption>()));
+            }
+
+            if (useSdca)
+            {
+                sdcaOption = sdcaOption ?? new SdcaOption();
+                sdcaOption.LabelColumnName = labelColumnName;
+                sdcaOption.FeatureColumnName = featureColumnName;
+                sdcaOption.ExampleWeightColumnName = exampleWeightColumnName;
+                res.Add(SweepableEstimatorFactory.CreateSdcaRegression(sdcaOption, sdcaSearchSpace ?? new SearchSpace<SdcaOption>()));
+            }
+
+            return res.ToArray();
+        }
     }
 }

src/Microsoft.ML.AutoML/API/SweepableExtension.cs

@@ -4,6 +4,7 @@
 
 using System;
 using System.Collections.Generic;
+using System.Linq;
 using System.Text;
 
 namespace Microsoft.ML.AutoML
@@ -27,7 +28,34 @@ public static SweepableEstimatorPipeline Append(this SweepableEstimator estimato
 
         public static SweepableEstimatorPipeline Append(this SweepableEstimator estimator, IEstimator<ITransformer> estimator1)
         {
-            return estimator.Append(estimator1);
+            return new SweepableEstimatorPipeline().Append(estimator).Append(estimator1);
+        }
+
+        public static MultiModelPipeline Append(this IEstimator<ITransformer> estimator, params SweepableEstimator[] estimators)
+        {
+            var sweepableEstimator = new SweepableEstimator((context, parameter) => estimator, new SearchSpace.SearchSpace());
+            var multiModelPipeline = new MultiModelPipeline().Append(sweepableEstimator).Append(estimators);
+
+            return multiModelPipeline;
+        }
+
+        public static MultiModelPipeline Append(this SweepableEstimatorPipeline pipeline, params SweepableEstimator[] estimators)
+        {
+            var multiModelPipeline = new MultiModelPipeline();
+            foreach (var estimator in pipeline.Estimators)
+            {
+                multiModelPipeline = multiModelPipeline.Append(estimator);
+            }
+
+            return multiModelPipeline.Append(estimators);
+        }
+
+        public static MultiModelPipeline Append(this SweepableEstimator estimator, params SweepableEstimator[] estimators)
+        {
+            var multiModelPipeline = new MultiModelPipeline();
+            multiModelPipeline = multiModelPipeline.Append(estimator);
+
+            return multiModelPipeline.Append(estimators);
         }
     }
 }

src/Microsoft.ML.AutoML/AutoMlUtils.cs

@@ -3,12 +3,46 @@
 // See the LICENSE file in the project root for more information.
 
 using System;
+using System.Collections.Generic;
 using System.Threading;
 
 namespace Microsoft.ML.AutoML
 {
     internal static class AutoMlUtils
     {
+        private const string MLNetMaxThread = "MLNET_MAX_THREAD";
+
         public static readonly ThreadLocal<Random> Random = new ThreadLocal<Random>(() => new Random());
+
+        /// <summary>
+        /// Return number of thread if MLNET_MAX_THREAD is set, otherwise return null.
+        /// </summary>
+        public static int? GetNumberOfThreadFromEnvrionment()
+        {
+            var res = Environment.GetEnvironmentVariable(MLNetMaxThread);
+
+            if (int.TryParse(res, out var numberOfThread))
+            {
+                return numberOfThread;
+            }
+
+            return null;
+        }
+
+        public static InputOutputColumnPair[] CreateInputOutputColumnPairsFromStrings(string[] inputs, string[] outputs)
+        {
+            if (inputs.Length != outputs.Length)
+            {
+                throw new Exception("inputs and outputs count must match");
+            }
+
+            var res = new List<InputOutputColumnPair>();
+            for (int i = 0; i != inputs.Length; ++i)
+            {
+                res.Add(new InputOutputColumnPair(outputs[i], inputs[i]));
+            }
+
+            return res.ToArray();
+        }
     }
 }

src/Microsoft.ML.AutoML/CodeGen/code_gen_flag.json

@@ -1,7 +1,7 @@
 {
   "EstimatorFactoryGenerator": false,
-  "CodeGenCatalogGenerator": false,
+  "SweepableEstimatorFactory": true,
   "EstimatorTypeGenerator": true,
   "SearchSpaceGenerator": true,
-  "SweepableEstimatorGenerator": false
+  "SweepableEstimatorGenerator": true
 }

src/Microsoft.ML.AutoML/Microsoft.ML.AutoML.csproj

@@ -14,11 +14,13 @@
       <PrivateAssets>all</PrivateAssets>
     </ProjectReference>
     <ProjectReference Include="..\Microsoft.ML.CpuMath\Microsoft.ML.CpuMath.csproj" />
+    <ProjectReference Include="..\Microsoft.ML.OnnxTransformer\Microsoft.ML.OnnxTransformer.csproj" />
     <ProjectReference Include="..\Microsoft.ML.SearchSpace\Microsoft.ML.SearchSpace.csproj">
       <PrivateAssets>all</PrivateAssets>
       <IncludeInNuget>true</IncludeInNuget>
     </ProjectReference>
     <PackageReference Include="Microsoft.CodeAnalysis.CSharp" Version="$(MicrosoftCodeAnalysisCSharpVersion)" />
+    <ProjectReference Include="..\Microsoft.ML.TimeSeries\Microsoft.ML.TimeSeries.csproj" />
     <ProjectReference Include="..\Microsoft.ML.Vision\Microsoft.ML.Vision.csproj" />
     <ProjectReference Include="..\Microsoft.ML.ImageAnalytics\Microsoft.ML.ImageAnalytics.csproj" />
     <ProjectReference Include="..\Microsoft.ML.LightGbm\Microsoft.ML.LightGbm.csproj" />
@@ -43,13 +45,13 @@
   <Target DependsOnTargets="ResolveReferences" Name="CopyProjectReferencesToPackage">
     <ItemGroup>
       <!--Include DLLs of Project References-->
-      <BuildOutputInPackage Include="@(ReferenceCopyLocalPaths->WithMetadataValue('ReferenceSourceTarget', 'ProjectReference')->WithMetadataValue('IncludeInNuget','true'))"/>
+      <BuildOutputInPackage Include="@(ReferenceCopyLocalPaths-&gt;WithMetadataValue('ReferenceSourceTarget', 'ProjectReference')-&gt;WithMetadataValue('IncludeInNuget','true'))" />
       <!--Include PDBs of Project References-->
-      <BuildOutputInPackage Include="@(ReferenceCopyLocalPaths->WithMetadataValue('ReferenceSourceTarget', 'ProjectReference')->WithMetadataValue('IncludeInNuget','true')->Replace('.dll', '.pdb'))"/>
+      <BuildOutputInPackage Include="@(ReferenceCopyLocalPaths-&gt;WithMetadataValue('ReferenceSourceTarget', 'ProjectReference')-&gt;WithMetadataValue('IncludeInNuget','true')-&gt;Replace('.dll', '.pdb'))" />
       <!--Include PDBs for Native binaries-->
       <!--The path needed to be hardcoded for this to work on our publishing CI-->
-      <BuildOutputInPackage Condition="Exists('$(PackageAssetsPath)$(PackageIdFolderName)\runtimes\win-x86\native\LdaNative.pdb')" Include="$(PackageAssetsPath)$(PackageIdFolderName)\runtimes\win-x86\native\LdaNative.pdb" TargetPath="..\..\runtimes\win-x86\native"/>
-      <BuildOutputInPackage Condition="Exists('$(PackageAssetsPath)$(PackageIdFolderName)\runtimes\win-x64\native\LdaNative.pdb')" Include="$(PackageAssetsPath)$(PackageIdFolderName)\runtimes\win-x64\native\LdaNative.pdb" TargetPath="..\..\runtimes\win-x64\native"/>
+      <BuildOutputInPackage Condition="Exists('$(PackageAssetsPath)$(PackageIdFolderName)\runtimes\win-x86\native\LdaNative.pdb')" Include="$(PackageAssetsPath)$(PackageIdFolderName)\runtimes\win-x86\native\LdaNative.pdb" TargetPath="..\..\runtimes\win-x86\native" />
+      <BuildOutputInPackage Condition="Exists('$(PackageAssetsPath)$(PackageIdFolderName)\runtimes\win-x64\native\LdaNative.pdb')" Include="$(PackageAssetsPath)$(PackageIdFolderName)\runtimes\win-x64\native\LdaNative.pdb" TargetPath="..\..\runtimes\win-x64\native" />
     </ItemGroup>
   </Target>
 

src/Microsoft.ML.AutoML/SweepableEstimator/Converter/MultiModelPipelineConverter.cs

@@ -0,0 +1,34 @@
+// Licensed to the .NET Foundation under one or more agreements.
+// The .NET Foundation licenses this file to you under the MIT license.
+// See the LICENSE file in the project root for more information.
+
+using System;
+using System.Collections.Generic;
+using System.Text;
+using System.Text.Json;
+using System.Text.Json.Nodes;
+using System.Text.Json.Serialization;
+
+namespace Microsoft.ML.AutoML
+{
+    internal class MultiModelPipelineConverter : JsonConverter<MultiModelPipeline>
+    {
+        public override MultiModelPipeline Read(ref Utf8JsonReader reader, Type typeToConvert, JsonSerializerOptions options)
+        {
+            var jValue = JsonValue.Parse(ref reader);
+            var schema = jValue["schema"].GetValue<string>();
+            var estimators = jValue["estimator"].GetValue<Dictionary<string, SweepableEstimator>>();
+
+            return new MultiModelPipeline(estimators, Entity.FromExpression(schema));
+        }
+
+        public override void Write(Utf8JsonWriter writer, MultiModelPipeline value, JsonSerializerOptions options)
+        {
+            var jsonObject = JsonNode.Parse("{}");
+            jsonObject["schema"] = value.Schema.ToString();
+            jsonObject["estimators"] = JsonValue.Create(value.Estimators);
+
+            jsonObject.WriteTo(writer, options);
+        }
+    }
+}