Leeds Sharp – Machine Learning with ML.NET and Azure

As an IT graduate, I am definitely not an experienced ‘techie’ however as an avid learner when it comes to the technology world I have decided to write a monthly blog after every Leeds Sharp event about what I have learnt. I’m hoping my blog will help understand what the Leeds Sharp Meet Up is all about or for those new Developers out there that are just starting out.

Leeds Sharp is a great event for anyone with a passion for the Microsoft technology stack. This meet up gives me the chance to keep on learning and to meet new people that are passionate about what they do. This is for everyone and anyone from a novice to an expert when it comes to software development.

So here goes….

ML.NET is an open source, machine learning framework built in .NET and runs on Windows, Linux and macOS. It allows developers to integrate custom machine learning into their applications without any prior expertise in developing or tuning machine learning models. Enhance your .NET apps with sentiment analysis, price prediction, fraud detection and more using custom models built with ML.NET.

Machine learning for prediction talk quality is where a computer program is said to learn from experience E with respect to some class of tasks T and performance measure P if its performance at tasks in T, as measured by P, improves with experience E.

E = historical quantities of beer consumed and corresponding audience ratings

T = predicting talk quality

P = accuracy of the talk quality prediction

There are two most commonly used machine learning methods: –

  1. Supervised – the computer is presented with example inputs and their desired outputs, given by a “teacher”
  2. Unsupervised – no desired outputs (“labels”) are given to the computer, leaving it on its own to find structure in the input.

Machine Learning Algorithms

  1. Linear Regression
  2. Logistic Regression
  3. Decision Tree
  4. Naïve Bayes
  5. kNN
  6. K-Means
  7. Dimensionality Reduction Algorithms
  8. Gradient Boosting Algorithms
    1. GBM
    2. XGBoost

Machine learning should be approached in a methodical manner, in this way we are more likely to achieve accurate, reliable and generalisable models. Best practive mostly revolves around how the data is used:

  • Data Preparation
  • Cross Validation
  • Validation and testing data sets

This relates to Big Data as there is distributed computing which breaks a larger problem into smaller tasks and distributes tasks around multiple computation hosts and aggregate results into an overall result.

Various places for ML include:

  • Azure Databricks
  • Azure HDInsight
  • Azure Data Lake Analytics
  • Azure ML
  • R/Python in many hosts
  • C# and dotnet hosted in many places
  • Typical Azure DevOps pipelines more mature for .NET

I hope this helps you understand Machine Learning with ML.Net and Azure. Everyone is welcome to Leeds Sharp and you don’t have to be developer, you just have to have an interest in tech and the .NET Microsoft stack. Go check out the Leeds Sharp page to find out more about the next meetup on the 29th November 2018

This is the last one of the year and it’s the Christmas Panel Show!

See you all there!

Fran 🙂