Neo, XGBoost, and ONNX

[shadimari]

I have an XGBoost model thats trained over SageMaker and I am trying to understand how Neo can help me to do model inferencing from a .NET application.

Is there a way i can compile the XGBoost model into ONNX through Neo?
Or
Is there a way i can invoke Neo DLR API through C# to invoke the compiled XGBoost model?
Or
Is there a way i can export XGBoost model as ONNX, thus bypassing Neo?