Recently Microsoft announced ML.NET, a machine learning framework for .NET. This is exciting news. So my mind immediately goes to: how does this look with F#? The current post will take a look at using ML.NET’s regression module to predict concrete compressive strength based on its composite ingredients.
F# and Word Stems
Read Time: 6 minutesToday’s post discusses performing word stemming with F#. This will be an expansion on a previous post, Comparing Quora question intent. As a result, it will also address some feature engineering.
F# and DotEnv
Read Time: 3 minutesOften the question arises, what is the best way to manage environment variables for a project. Many languages have libaries precisely for this issue. A common approach is the use of a .env file, .NET and F# are no different. In today’s post I’ll take a brief look at leveraging DotEnv in an F# application.
Coding with Color
Read Time: 3 minutesCode comes in many shapes and sizes. Often this means differing paradigms with text or gui workflow tools, but there are other ways. Enter Evo, a robot that leverages coding in the form of color. I recently had the opportunity to play with one of the little bots. It is intended as an educational toy for kids, and it fits the bill perfectly. More than that, it is an interesting case study in seeing the world differently. Take code beyond text, and into a realm that can easily appeal to kids (young and old).
Tackling Kaggle with F# and XGBoost
Read Time: 16 minutesToday’s topic will be to demonstrate tackling a Kaggle problem with XGBoost and F#. Comparing Quora question intent offers a perfect opportunity to work with XGBoost, a common tool used in Kaggle competitions. Luckily there is a .NET wrapper around the XGBoost library, XGBoost.Net.