Recently, a friend of mine, Maxx and myself decided to embark on a quest to start learning ML.NET. It came to our attention that ML.NET is released on Preview in 2 April 2019 and subsequently on 3 May 2019 on Stable Release.

Being in a language that I am comfortable with, i decided to give it a try and see what capabilities are available and how we are able to build something.

Every the course of next few weeks, we are going to try

1. Coding it ourselves
2. Using Model Builder Tool
3. Using AutoML to perform model training and picking the best algorithm.

More information can be found on their site on
https://dotnet.microsoft.com/apps/machinelearning-ai/ml-dotnet

We are also going to follow the ML.NET tutorials released –
https://docs.microsoft.com/en-gb/dotnet/machine-learning/tutorials/ then attempt to run the sample –
https://github.com/dotnet/machinelearning-samples/blob/master/README.md

We have ran through the Microsoft Docs detailing ML.NET and will be extracting the important points out in the next various posts i made.

ML.NET will give developers the power to add in machine learning capabilities to .NET applications. With this ability, developers are now able to make predictions using the data presented.

The general steps required
1. Load the data
2. Prepare the data
3. Train and test the model
4. Get predictions / Using the model

ML.NET presents various functions and methods to do them and in the course of next few weeks and posts, we will present them in our blogs.

As of date of posting, ML.NET has the following Tasks.

1. Binary Classification
2. Multiclass Classification
3. Regression
4. Clustering
5. Anomaly Detection
6. Ranking
7. Recommendation

Posts-
ML.NET Introduction – Introduction
ML.NET Part 2 – Machine Learning – Environment setup