System.OverflowException: Arithmetic operation resulted in an overflow

[vanderryan]

Describe the bug
Expanded dataset. 14 Training data inputs,14 Validation data inputs. Each data point has 29 features:
[
65389,
80,
420,
3,
etc, etc up to 28 items
],
Running with data like this throws:
"Unhandled exception. System.OverflowException: Arithmetic operation resulted in an overflow. at System.Convert.ToInt64(Double value)"
Error bubbles up from

let (optimizedModel, nMisses) = TrainSequentialClassifier(
Mapped(
SequentialModel(ClassifierStructure(), _, 0.0),
initialParameters
),

To Reproduce
Steps to reproduce the behavior:
Follow instructions on Github to run via 'dotnet run'. Please note the only version I could get installed was: Microsoft.Quantum.MachineLearning --version 0.23.195983

Expected behavior
Program to run despit increase number of features.

Screenshots
If applicable, add screenshots to help explain your problem.

System information

• OS: Windows 10.0.19043
• .NET 5.0.202

Additional context
Add any other context about the problem here.

[cgranade]

Thanks for your report, @vanderryan! It looks like this may be an issue with the QML library, so transferring to the quantumlibraries repo where that can be properly triaged. Thank you!