How OnnxRuntime C# API can address the pain points of ~900 ML.NET Apr2021 survey responses?

[GeorgeS2019]

Is your feature request related to a problem? Please describe.

The pain points of the Apr 2021 ML.NET survey and the result discussions

Describe the solution you'd like

It is clear that NLP is high on priority This means more deep learning NLP use cases e.g. using ML.NET or just ONNX C# API to load pretrained Hugging Face transformer models using OnnxRuntime

With recent renaming effort to make TorchSharp more PyTorch like, it is NOW more feasible to port the PyTorch NLP Transformer codes [The ONNX API inference session] to .NET (c# or f#)

Here are some of the PyTorch NLP Transformer code candidates (porting to c#) to address the pain points of ML.NET!!

• Inference_GPT2_with_OnnxRuntime_on_CPU.ipynb
• Inference_GPT2-OneStepSearch_OnnxRuntime_CPU.ipynb

[tianleiwu]

Some partners are using OnnxRuntime nuget package for NLP model (like BERT and GPT-2). It might worth to add an example of C#, although we do have no timeline for that yet.

[GeorgeS2019]

FYI Journey to optimize large scale transformer model inference with ONNX Runtime -June 30, 2021

The article that cite the notebook: Inference_GPT2-OneStepSearch_OnnxRuntime_CPU.ipynb