AutoML Experiment example fails

[gewarren]

---

Issue moved from MicrosoftDocs/feedback#3979

• Please respond to @Elantonio.

---

From @Elantonio on Monday, January 22, 2024 1:15:45 PM

I'm following the example code on: https://learn.microsoft.com/en-us/dotnet/machine-learning/how-to-guides/how-to-use-the-automl-api

MyMLContext = new MLContext();
ColumnInferenceResults columnInference =
  MyMLContext.Auto().InferColumns(TrainFilePath, labelColumnName: LabelColumnName, groupColumns: false);

TextLoader loader = MyMLContext.Data.CreateTextLoader(columnInference.TextLoaderOptions);

// Load data into IDataView
IDataView data = loader.Load(TrainFilePath);

TrainTestData trainValidationData = MyMLContext.Data.TrainTestSplit(data, testFraction: 0.2);

SweepablePipeline pipeline =
  MyMLContext.Auto().Featurizer(data, columnInformation: columnInference.ColumnInformation)
      .Append(MyMLContext.Auto().Regression(labelColumnName: columnInference.ColumnInformation.LabelColumnName));

AutoMLExperiment experiment = MyMLContext.Auto().CreateExperiment();

experiment
  .SetPipeline(pipeline)
  .SetRegressionMetric(RegressionMetric.RSquared, labelColumn: columnInference.ColumnInformation.LabelColumnName)
  .SetTrainingTimeInSeconds((uint)Math.Max(seconds, 10))
  .SetDataset(trainValidationData);

TrialResult experimentResults = experiment.Run();

But it gives an error on experiment.Run(); Error = "One or more errors occurred. (Could not find feature column 'Features' (Parameter 'inputSchema')"

I can't seem to find what I'm missing here. Anyone an idea, and if so should the docu be updated?

To reproduce Steps to reproduce the behavior:

-in vs2022 create a project -Add nuget Microsoft.ML.AutoML 0.21.1 -create a class Predictor -add an instantiator with the above code; -Build -Instantiate the class

Expected behavior Run without error

Environment Desktop OS: Win11 . Visual Studio Community 2022 (64-bit) -17.6.4