lqdev / FSAccordSuaveDemo

F# Naive Bayes Classification Model built with Accord.NET and exposed via a Suave API

F# Suave Accord Demo

This sample project shows how to expose a machine learning model built with Accord.NET via a Suave API

Dependencies

F# Tools

Windows
Linux
Mac

Install

Windows

.paket\paket.exe install

Mac/Linux

mono .paket\paket.exe install

Train Model

To train the model, run the App.fsx script

Start Server

To initialize the server, run the Server.fsx script

Get Prediction

With the server running, using a REST client like POSTMAN or Insomnia make an HTTP POST request to the enpoint http://localhost:3001/api/predict. In the request body include the data property which is a 2D array of type float.

Sample Request Body

{
	"data": [[3.3,1.6,0.2,5.1]] 
}

Sample Response

{
    "label": [
        "Iris-virginica"
    ]
}