Skip to content

Enable SymbolicSgdLogisticRegression (SymSgdNative) on arm64 - #7671

Open
vladimir-aubrecht wants to merge 2 commits into
dotnet:mainfrom
vladimir-aubrecht:feature/symsgd-arm64
Open

Enable SymbolicSgdLogisticRegression (SymSgdNative) on arm64#7671
vladimir-aubrecht wants to merge 2 commits into
dotnet:mainfrom
vladimir-aubrecht:feature/symsgd-arm64

Address PR review: const-correct cblas_sdoti, null handle in DftiFree…

4de75e4
Select commit
Loading
Failed to load commit list.
Build Analysis / Build Analysis failed Aug 11, 2026 in 0s

.NET Result Analysis

Details

Tip

To unconditionally bypass the build analysis check (turn it green), you can use the escape mechanism feature. The completion time may vary, potentially taking several minutes, depending on the build analysis workload at the moment.

Build Failures

MachineLearning-CodeCoverage / Windows_x64 / Windows_x64 Build_Debug / Run All Tests.

[ 🚧 Create issue against infrastructure team] [ 📄 Create issue in this repo]
    ❌Cmd.exe exited with code '1'.

MachineLearning-CI / MacOS_cross_arm64 / MacOS_cross_arm64 Debug_Build / Build

[ 🚧 Create issue against infrastructure team] [ 📄 Create issue in this repo]
    ❌Bash exited with code '1'.

MachineLearning-CI / MacOS_cross_arm64 / MacOS_cross_arm64 Release_Build / Build

[ 🚧 Create issue against infrastructure team] [ 📄 Create issue in this repo]
    ❌Bash exited with code '1'.

2 other jobs are failing and were not displayed due to space limitations.
Builds with failures: [MachineLearning-CodeCoverage] [MachineLearning-CI]

Known test errors

Test Failures (73 tests failed)

🔹 [All failing tests from MachineLearning-CI]

    ⚠️ 67 failing tests not displayed for this pipeline due to space limitations

    ❌ Microsoft.ML.IntegrationTests.Validation.CrossValidation [Console] [Details] [Artifacts] [1.55% failure rate]
    [ 🚧 Create issue against infrastructure team] [ 📄 Create issue in this repo]

    Failing Configurations (2)
    Exception Message
    System.EntryPointNotFoundException : Unable to find an entry point named 'LAPACKE_dpptrf' in DLL 'MklImports'.
    CallStack
       at Microsoft.ML.Trainers.OlsTrainer.Mkl.PptrfInternal(Layout layout, UpLo uplo, Int32 n, Double[] ap)
   at Microsoft.ML.Trainers.OlsTrainer.Mkl.Pptrf(Layout layout, UpLo uplo, Int32 n, Double[] ap) in D:\a\_work\1\s\src\Microsoft.ML.Mkl.Components\OlsLinearRegression.cs:line 568
   at Microsoft.ML.Trainers.OlsTrainer.ComputeMklRegression(IChannel ch, Factory cursorFactory, Int32 m, Double[]& beta, Double[] xtx, Int64& n, Double& yMean) in D:\a\_work\1\s\src\Microsoft.ML.Mkl.Components\OlsLinearRegression.cs:line 392
   at Microsoft.ML.Trainers.OlsTrainer.TrainCore(IChannel ch, Factory cursorFactory, Int32 featureCount) in D:\a\_work\1\s\src\Microsoft.ML.Mkl.Components\OlsLinearRegression.cs:line 439
   at Microsoft.ML.Trainers.OlsTrainer.TrainModelCore(TrainContext context) in D:\a\_work\1\s\src\Microsoft.ML.Mkl.Components\OlsLinearRegression.cs:line 178
   at Microsoft.ML.Trainers.TrainerEstimatorBase`2.TrainTransformer(IDataView trainSet, IDataView validationSet, IPredictor initPredictor) in D:\a\_work\1\s\src\Microsoft.ML.Data\Training\TrainerEstimatorBase.cs:line 157
   at Microsoft.ML.Trainers.TrainerEstimatorBase`2.Fit(IDataView input) in D:\a\_work\1\s\src\Microsoft.ML.Data\Training\TrainerEstimatorBase.cs:line 77
   at Microsoft.ML.Data.EstimatorChain`1.Fit(IDataView input) in D:\a\_work\1\s\src\Microsoft.ML.Data\DataLoadSave\EstimatorChain.cs:line 68
   at Microsoft.ML.TrainCatalogBase.CrossValidateTrain(IDataView data, IEstimator`1 estimator, Int32 numFolds, String samplingKeyColumn, Nullable`1 seed) in D:\a\_work\1\s\src\Microsoft.ML.Data\TrainCatalog.cs:line 105
   at Microsoft.ML.RegressionCatalog.CrossValidate(IDataView data, IEstimator`1 estimator, Int32 numberOfFolds, String labelColumnName, String samplingKeyColumnName, Nullable`1 seed) in D:\a\_work\1\s\src\Microsoft.ML.Data\TrainCatalog.cs:line 628
   at Microsoft.ML.IntegrationTests.Validation.CrossValidation() in D:\a\_work\1\s\test\Microsoft.ML.IntegrationTests\Validation.cs:line 45
   at System.RuntimeMethodHandle.InvokeMethod(Object target, Void** arguments, Signature sig, Boolean isConstructor)
   at System.Reflection.MethodBaseInvoker.InvokeWithNoArgs(Object obj, BindingFlags invokeAttr)
    ❌ Microsoft.ML.RunTests.TestEntryPoints.EntryPointLinearPredictorSummary [Console] [Details] [Artifacts] [1.97% failure rate]
    [ 🚧 Create issue against infrastructure team] [ 📄 Create issue in this repo]

    Failing Configurations (2)
    Exception Message
    System.EntryPointNotFoundException : Unable to find an entry point named 'LAPACKE_dpptrf' in DLL 'MklImports'.
    CallStack
       at Microsoft.ML.Trainers.OlsTrainer.Mkl.PptrfInternal(Layout layout, UpLo uplo, Int32 n, Double[] ap)
   at Microsoft.ML.Trainers.OlsTrainer.Mkl.Pptrf(Layout layout, UpLo uplo, Int32 n, Double[] ap) in D:\a\_work\1\s\src\Microsoft.ML.Mkl.Components\OlsLinearRegression.cs:line 568
   at Microsoft.ML.Trainers.ComputeLRTrainingStdThroughMkl.ComputeStandardDeviation(Double[] hessian, Int32[] weightIndices, Int32 numSelectedParams, Int32 currentWeightsCount, IChannel ch, Single l2Weight) in D:\a\_work\1\s\src\Microsoft.ML.Mkl.Components\ComputeLRTrainingStdThroughHal.cs:line 40
   at Microsoft.ML.Trainers.LbfgsLogisticRegressionBinaryTrainer.ComputeTrainingStatistics(IChannel ch, Factory cursorFactory, Single loss, Int32 numParams) in D:\a\_work\1\s\src\Microsoft.ML.StandardTrainers\Standard\LogisticRegression\LogisticRegression.cs:line 413
   at Microsoft.ML.Trainers.LbfgsTrainerBase`3.TrainCore(IChannel ch, RoleMappedData data) in D:\a\_work\1\s\src\Microsoft.ML.StandardTrainers\Standard\LogisticRegression\LbfgsPredictorBase.cs:line 720
   at Microsoft.ML.Trainers.LbfgsTrainerBase`3.TrainModelCore(TrainContext context) in D:\a\_work\1\s\src\Microsoft.ML.StandardTrainers\Standard\LogisticRegression\LbfgsPredictorBase.cs:line 458
   at Microsoft.ML.Data.TrainUtils.TrainCore(IHostEnvironment env, IChannel ch, RoleMappedData data, ITrainer trainer, RoleMappedData validData, IComponentFactory`1 calibrator, Int32 maxCalibrationExamples, Nullable`1 cacheData, IPredictor inputPredictor, RoleMappedData testData) in D:\a\_work\1\s\src\Microsoft.ML.Data\Commands\TrainCommand.cs:line 280
   at Microsoft.ML.EntryPoints.TrainerEntryPointsUtils.Train[TArg,TOut](IHost host, TArg input, Func`1 createTrainer, Func`1 getLabel, Func`1 getWeight, Func`1 getGroup, Func`1 getName, Func`1 getCustom, ICalibratorTrainerFactory calibrator, Int32 maxCalibrationExamples) in D:\a\_work\1\s\src\Microsoft.ML.Data\EntryPoints\InputBase.cs:line 116
   at Microsoft.ML.Trainers.LbfgsLogisticRegressionBinaryTrainer.TrainBinary(IHostEnvironment env, Options input) in D:\a\_work\1\s\src\Microsoft.ML.StandardTrainers\Standard\LogisticRegression\LogisticRegression.cs:line 473
   at Microsoft.ML.RunTests.TestEntryPoints.EntryPointLinearPredictorSummary() in D:\a\_work\1\s\test\Microsoft.ML.Core.Tests\UnitTests\TestEntryPoints.cs:line 4257
   at System.RuntimeMethodHandle.InvokeMethod(Object target, Void** arguments, Signature sig, Boolean isConstructor)
   at System.Reflection.MethodBaseInvoker.InvokeWithNoArgs(Object obj, BindingFlags invokeAttr)
    ❌ Microsoft.ML.RunTests.TestEntryPoints.EntryPointPipelineEnsembleGetSummary [Console] [Details] [Artifacts] [1.97% failure rate]
    [ 🚧 Create issue against infrastructure team] [ 📄 Create issue in this repo]

    Failing Configurations (2)
    Exception Message
    System.EntryPointNotFoundException : Unable to find an entry point named 'LAPACKE_dpptrf' in DLL 'MklImports'.
    CallStack
       at Microsoft.ML.Trainers.OlsTrainer.Mkl.PptrfInternal(Layout layout, UpLo uplo, Int32 n, Double[] ap)
   at Microsoft.ML.Trainers.OlsTrainer.Mkl.Pptrf(Layout layout, UpLo uplo, Int32 n, Double[] ap) in D:\a\_work\1\s\src\Microsoft.ML.Mkl.Components\OlsLinearRegression.cs:line 568
   at Microsoft.ML.Trainers.ComputeLRTrainingStdThroughMkl.ComputeStandardDeviation(Double[] hessian, Int32[] weightIndices, Int32 numSelectedParams, Int32 currentWeightsCount, IChannel ch, Single l2Weight) in D:\a\_work\1\s\src\Microsoft.ML.Mkl.Components\ComputeLRTrainingStdThroughHal.cs:line 40
   at Microsoft.ML.Trainers.LbfgsLogisticRegressionBinaryTrainer.ComputeTrainingStatistics(IChannel ch, Factory cursorFactory, Single loss, Int32 numParams) in D:\a\_work\1\s\src\Microsoft.ML.StandardTrainers\Standard\LogisticRegression\LogisticRegression.cs:line 413
   at Microsoft.ML.Trainers.LbfgsTrainerBase`3.TrainCore(IChannel ch, RoleMappedData data) in D:\a\_work\1\s\src\Microsoft.ML.StandardTrainers\Standard\LogisticRegression\LbfgsPredictorBase.cs:line 720
   at Microsoft.ML.Trainers.LbfgsTrainerBase`3.TrainModelCore(TrainContext context) in D:\a\_work\1\s\src\Microsoft.ML.StandardTrainers\Standard\LogisticRegression\LbfgsPredictorBase.cs:line 458
   at Microsoft.ML.Data.TrainUtils.TrainCore(IHostEnvironment env, IChannel ch, RoleMappedData data, ITrainer trainer, RoleMappedData validData, IComponentFactory`1 calibrator, Int32 maxCalibrationExamples, Nullable`1 cacheData, IPredictor inputPredictor, RoleMappedData testData) in D:\a\_work\1\s\src\Microsoft.ML.Data\Commands\TrainCommand.cs:line 280
   at Microsoft.ML.EntryPoints.TrainerEntryPointsUtils.Train[TArg,TOut](IHost host, TArg input, Func`1 createTrainer, Func`1 getLabel, Func`1 getWeight, Func`1 getGroup, Func`1 getName, Func`1 getCustom, ICalibratorTrainerFactory calibrator, Int32 maxCalibrationExamples) in D:\a\_work\1\s\src\Microsoft.ML.Data\EntryPoints\InputBase.cs:line 116
   at Microsoft.ML.Trainers.LbfgsLogisticRegressionBinaryTrainer.TrainBinary(IHostEnvironment env, Options input) in D:\a\_work\1\s\src\Microsoft.ML.StandardTrainers\Standard\LogisticRegression\LogisticRegression.cs:line 473
   at Microsoft.ML.RunTests.TestEntryPoints.EntryPointPipelineEnsembleGetSummary() in D:\a\_work\1\s\test\Microsoft.ML.Core.Tests\UnitTests\TestEntryPoints.cs:line 2166
   at System.RuntimeMethodHandle.InvokeMethod(Object target, Void** arguments, Signature sig, Boolean isConstructor)
   at System.Reflection.MethodBaseInvoker.InvokeWithNoArgs(Object obj, BindingFlags invokeAttr)
    ❌ Microsoft.ML.RunTests.TestPredictors.BinaryClassifierSymSgdTest [Console] [Details] [Artifacts] [1.97% failure rate]
    [ 🚧 Create issue against infrastructure team] [ 📄 Create issue in this repo]

    Failing Configurations (2)
    Exception Message
    Assert.Equal() Failure: Values differ
Expected: 0
Actual:   10
    CallStack
       at Microsoft.ML.RunTests.BaseTestBaseline.Done() in D:\a\_work\1\s\test\Microsoft.ML.TestFramework\BaseTestBaseline.cs:line 221
   at Microsoft.ML.RunTests.TestPredictors.BinaryClassifierSymSgdTest() in D:\a\_work\1\s\test\Microsoft.ML.Predictor.Tests\TestPredictors.cs:line 289
   at System.RuntimeMethodHandle.InvokeMethod(Object target, Void** arguments, Signature sig, Boolean isConstructor)
   at System.Reflection.MethodBaseInvoker.InvokeWithNoArgs(Object obj, BindingFlags invokeAttr)
    ❌ Microsoft.ML.RunTests.TestPredictors.RegressorOlsTestOne [Console] [Details] [Artifacts] [1.97% failure rate]
    [ 🚧 Create issue against infrastructure team] [ 📄 Create issue in this repo]

    Failing Configurations (2)
    Exception Message
    System.AggregateException : One or more errors occurred. (Value cannot be null. (Parameter 'input')) (Assert.Equal() Failure: Values differ
Expected: 0
Actual:   1)
---- System.ArgumentNullException : Value cannot be null. (Parameter 'input')
---- Assert.Equal() Failure: Values differ
Expected: 0
Actual:   1
    CallStack
    ----- Inner Stack Trace #1 (System.ArgumentNullException) -----
   at System.Text.RegularExpressions.ThrowHelper.ThrowArgumentNullException(ExceptionArgument arg)
   at Microsoft.ML.RunTests.BaseTestBaseline.CheckEqualityFromPathsCore(String relPath, String basePath, String outPath, Int32 skip, Int32 digitsOfPrecision, NumberParseOption parseOption) in D:\a\_work\1\s\test\Microsoft.ML.TestFramework\BaseTestBaseline.cs:line 556
   at Microsoft.ML.RunTests.BaseTestBaseline.CheckEqualityNormalized(String dir, String name, String nameBase, Int32 digitsOfPrecision, NumberParseOption parseOption) in D:\a\_work\1\s\test\Microsoft.ML.TestFramework\BaseTestBaseline.cs:line 397
   at Microsoft.ML.RunTests.BaseTestPredictors.Run(RunContext ctx, Int32 digitsOfPrecision, NumberParseOption parseOption) in D:\a\_work\1\s\test\Microsoft.ML.TestFramework\BaseTestPredictorsMaml.cs:line 213
   at Microsoft.ML.RunTests.BaseTestPredictors.Run_TrainTest(PredictorAndArgs predictor, TestDataset dataset, String[] extraSettings, String extraTag, Boolean expectFailure, Boolean summary, Boolean saveAsIni, Int32 digitsOfPrecision, NumberParseOption parseOption) in D:\a\_work\1\s\test\Microsoft.ML.TestFramework\BaseTestPredictorsMaml.cs:line 396
   at Microsoft.ML.RunTests.TestPredictors.RegressorOlsTestOne() in D:\a\_work\1\s\test\Microsoft.ML.Predictor.Tests\TestPredictors.cs:line 1201
   at System.RuntimeMethodHandle.InvokeMethod(Object target, Void** arguments, Signature sig, Boolean isConstructor)
   at System.Reflection.MethodBaseInvoker.InvokeWithNoArgs(Object obj, BindingFlags invokeAttr)
----- Inner Stack Trace #2 (Xunit.Sdk.EqualException) -----
   at Microsoft.ML.RunTests.BaseTestBaseline.Cleanup() in D:\a\_work\1\s\test\Microsoft.ML.TestFramework\BaseTestBaseline.cs:line 200
   at Microsoft.ML.TestFramework.BaseTestClass.System.IDisposable.Dispose() in D:\a\_work\1\s\test\Microsoft.ML.TestFramework\BaseTestClass.cs:line 80
   at ReflectionAbstractionExtensions.DisposeTestClass(ITest test, Object testClass, IMessageBus messageBus, ExecutionTimer timer, CancellationTokenSource cancellationTokenSource) in /_/src/xunit.execution/Extensions/ReflectionAbstractionExtensions.cs:line 79

🔹 [All failing tests from MachineLearning-CodeCoverage]

    ⚠️ 1 failing tests not displayed for this pipeline due to space limitations

Was this helpful? Yes No

View more details on Build Analysis