UDF declaration throwing exception when unittesting

[nkhlss]

Hi, I am trying to write unittest for dotnet spark code. However, UDF declaration is throwing exception.

  Message: 
Test method ConsolidatedTests.NetCore.Test.TestExecute threw exception: 
System.InvalidCastException: Unable to cast object of type 'Microsoft.Spark.Interop.Ipc.JvmObjectReference' to type 'System.String'.

  Stack Trace: 
SparkEnvironment.GetSparkVersion()
<.cctor>b__20_0()
Lazy`1.ViaFactory(LazyThreadSafetyMode mode)
Lazy`1.ExecutionAndPublication(LazyHelper executionAndPublication, Boolean useDefaultConstructor)
Lazy`1.CreateValue()
Lazy`1.get_Value()
SparkEnvironment.get_SparkVersion()
UdfUtils.CreateEnvVarsForPythonFunction(IJvmBridge jvm)
UdfUtils.CreatePythonFunction(IJvmBridge jvm, Byte[] command)
UserDefinedFunction.Create(IJvmBridge jvm, String name, Byte[] command, PythonEvalType evalType, String returnType)
UserDefinedFunction.Create(String name, Byte[] command, PythonEvalType evalType, String returnType)
Functions.CreateUdf(String name, Delegate execute, PythonEvalType evalType, String returnType)
Functions.CreateUdf[TResult](String name, Delegate execute, PythonEvalType evalType)
Functions.CreateUdf[TResult](String name, Delegate execute)
Functions.Udf[T,TResult](Func`2 udf)
Test.TestExecute() line 69

its a simple Unittest (doing absolutely nothing)

[TestMethod]
public void TestExecute()
{
    Func<Column, Column> udf = Udf<int, int>(str => str);
}

versions of MSTest and Microsoft.Spark are,

<PackageReference Update="MSTest.TestFramework"              Version="[2.1.0]" />
<PackageReference Update="Microsoft.Spark"                   Version="[2.1.1]" />

What am I missing?

[AFFogarty]

What Spark version are you using?

[nkhlss]

The remote spark pool has version 3.1.
However, the above exception is thrown while performing unittest on the local machine which does not have Spark installed but the code references nuget package Microsoft.Spark (Version=[2.1.1])