bartczernicki/baseballmachinelearningworkbench repository overview

Baseball Machine Learning Workbench A web application that showcases performing advanced analysis (decision thresholding, what-if analysis) using in-memory Machine Learning models.

GitHub Repository: https://github.com/bartczernicki/MachineLearning-BaseballPrediction-BlazorApp⁠
Live Demo (Hosted on Azure App Service): https://aka.ms/BaseballMLWorkbench⁠
Live Demo (Docker container hosted on Azure Container Instances): http://baseballmachinelearningworkbench.eastus2.azurecontainer.io⁠

Howto use the Docker Container:
Use this command to launch the Docker container: docker run -it --rm -p 8080:80 --name baseballmlworkbench bartczernicki/baseballmachinelearningworkbench:v1
Using your browser, navigate to http://localhost:8080⁠

The application has the following features:

Three different decision analysis mechanisms performing what-if analysis
A simple rules engine to predict baseball hall of fame induction contrasted with Machine Intelligence
Single and multiple machine learning models working together to predict baseball hall of fame ballot and induction
Machine Learning models are surfaced via ML.NET in-memory for very quick inference (predictions)
Surfaced via the Server-Side Blazor .NET Core application framework using SignalR to deliver the predictions from the server to the web client at scale

Architecture - Cloud Deployment Diagram:

Project Structure:
Visual Studio 2019 v4.0, .NET Core 3.1, Server-Side Blazor, ML.NET v1.5, Azure SignalR (optional)

More Information:

ML.NET: https://dotnet.microsoft.com/apps/machinelearning-ai/ml-dotnet⁠
Blazor: https://dotnet.microsoft.com/apps/aspnet/web-apps/blazor⁠
Historical Baseball Statistics Database: http://www.seanlahman.com/baseball-archive/statistics/⁠
Decision Management Systems (Amazon book): https://www.amazon.com/Decision-Management-Systems-Practical-Predictive/dp/0132884380⁠