zewditu
commented
1 year ago @SSlavaV it seems that you are running a multiclass classification for Image but you have only one label(class) which does not make sense for multi-classification, at least you should have two categories or types of datasets
version ML.Net 17.14.4.2312404
Only one class found in the Label column. To build a multiclass classification model, the number of classes needs to be 2 or greater Имя параметра: labelCount
в Microsoft.ML.Vision.ImageClassificationTrainer.InitializeTrainingGraph(IDataView input) в Microsoft.ML.Vision.ImageClassificationTrainer.TrainModelCore(TrainContext trainContext) в Microsoft.ML.Trainers.TrainerEstimatorBased24.MoveNext()
--- Конец трассировка стека из предыдущего расположения, где возникло исключение ---
в System.Runtime.ExceptionServices.ExceptionDispatchInfo.Throw()
в System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
в Microsoft.ML.ModelBuilder.AutoMLService.LocalImageClassificationExperiment.d 13.MoveNext() в /_/src/Microsoft.ML.ModelBuilder.AutoMLService/Experiments/LocalImageClassificationExperiment.cs:строка 116
--- Конец трассировка стека из предыдущего расположения, где возникло исключение ---
в System.Runtime.ExceptionServices.ExceptionDispatchInfo.Throw()
в System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
в Microsoft.ML.ModelBuilder.AutoMLEngine.d_21.MoveNext() в //src/Microsoft.ML.ModelBuilder.AutoMLService/AutoMLEngineService/AutoMLEngine.cs:строка 145
2.TrainTransformer(IDataView trainSet, IDataView validationSet, IPredictor initPredictor) в Microsoft.ML.Data.EstimatorChain1.Fit(IDataView input) в Microsoft.ML.AutoML.SweepablePipelineRunner.Run(TrialSettings settings) в Microsoft.ML.AutoML.SweepablePipelineRunner.RunAsync(TrialSettings settings, CancellationToken ct) в Microsoft.ML.AutoML.AutoMLExperiment.