ml.net model works with console application but not web api - asp.net-web-api

Question

I have created an MLModel with ml.net model builder and it works with a 'consume' console application and hard coded values for input. I wanted to go to the next step and feed values with web api. I am using a modified version of their tutorial example on using a mlmodel in a web api. When I use postman with the webapi I get the following error;
System.InvalidOperationException: 'Could not apply a map over type 'Single' to column 'PG' since it has type 'String''
The mlmodel builds successfully from model builder and the default code is generated from the model builder. Running a consume console app with hard coded values on the model works.
Here is the relevant code for modelinput
//*****************************************************************************************
//* *
//* This is an auto-generated file by Microsoft ML.NET CLI (Command-Line Interface) tool. *
//* *
//*****************************************************************************************
using Microsoft.ML.Data;
namespace JobClassMLModelML.Model.DataModels
{
public class ModelInput
{
[ColumnName("EducationLevel"), LoadColumn(0)]
public float EducationLevel { get; set; }
[ColumnName("Experience"), LoadColumn(1)]
public float Experience { get; set; }
[ColumnName("OrgImpact"), LoadColumn(2)]
public float OrgImpact { get; set; }
[ColumnName("ProblemSolving"), LoadColumn(3)]
public float ProblemSolving { get; set; }
[ColumnName("Supervision"), LoadColumn(4)]
public float Supervision { get; set; }
[ColumnName("ContactLevel"), LoadColumn(5)]
public float ContactLevel { get; set; }
[ColumnName("FinancialBudget"), LoadColumn(6)]
public float FinancialBudget { get; set; }
[ColumnName("PG"), LoadColumn(7)]
public string PG { get; set; }
}
}
And here it is for modeloutput
//*****************************************************************************************
//* *
//* This is an auto-generated file by Microsoft ML.NET CLI (Command-Line Interface) tool. *
//* *
//*****************************************************************************************
using System;
using Microsoft.ML.Data;
namespace JobClassMLModelML.Model.DataModels
{
public class ModelOutput
{
// ColumnName attribute is used to change the column name from
// its default value, which is the name of the field.
[ColumnName("PredictedLabel")]
public String Prediction { get; set; }
public float[] Score { get; set; }
}
}
I cloned the modelinput and modeloutput classes in equivalent webapi classes of jobclassdata and jobclassprediction respectively
The console application that works:
using System;
using Microsoft.ML;
using JobClassMLModelML;
using JobClassMLModelML.Model.DataModels;
namespace ConsumeApp
{
class Program
{
static void Main(string[] args)
{ // Load the model
MLContext mlContext = new MLContext();
ITransformer mlModel = mlContext.Model.Load("MLModel.zip", out
var modelInputSchema);
var predEngine =
mlContext.Model.CreatePredictionEngine<ModelInput, ModelOutput>(mlModel);
// Use the code below to add input data
var input = new ModelInput();
// input.
input.ContactLevel = 5;
input.EducationLevel = 5;
input.Experience = 5;
input.FinancialBudget = 5;
input.OrgImpact = 5;
input.ProblemSolving = 5;
input.Supervision =5;
//input.PG = "PG01";
// Try model on sample data
ModelOutput result = predEngine.Predict(input);
Console.WriteLine($"Predicted value: {result.Prediction} ");
}
}
}
Here is the code for the predict controller
using System;
using System.Collections.Generic;
using System.Linq;
using System.Threading.Tasks;
using Microsoft.AspNetCore.Http;
using Microsoft.AspNetCore.Mvc;
using Microsoft.Extensions.ML;
using JobClassMLModelWebAPI.DataModels;
namespace JobClassMLModelWebAPI.Controllers
{
[Route("api/[controller]")]
[ApiController]
public class PredictController : ControllerBase
{
private readonly PredictionEnginePool<JobClassData,
JobClassPrediction> _predictionEnginePool;
public PredictController(PredictionEnginePool<JobClassData,
JobClassPrediction> predictionEnginePool)
{
_predictionEnginePool = predictionEnginePool;
}
[HttpGet]
public JobClassData Get(JobClassData jobClassData)
{
return new JobClassData
{
// insert test data here
};
}
[HttpPost]
public ActionResult<string> Post([FromBody] JobClassData input)
{
if (!ModelState.IsValid)
{
return BadRequest();
}
JobClassPrediction prediction =
_predictionEnginePool.Predict(input);
string PredictedJobGrade = prediction.Prediction;
return Ok(PredictedJobGrade);
}
}
}
Here is json body code posted
{
"educationLevel": 3,
"experience": 3,
"orgImpact": 3,
"problemSolving": 3,
"supervision": 3,
"contactLevel": 4,
"financialBudget": 4,
"pg" : "PG01"
}
The expected result would be predicted paygrade such as PG03. Instead the error is
System.InvalidOperationException: 'Could not apply a map over type 'Single' to column 'PG' since it has type 'String''
at the line
JobClassPrediction prediction =
_predictionEnginePool.Predict(input);
with no compile error and with console app working is the format of the json input?
Thanks