Welcome to the second part of the forester blog. In the previous part, we explained the main idea of the forester package, the motivations behind it, its advantages, and the innovations it brings to the ML world. You should definitely check it out! In this part, however, we will focus on showing the wide range of possibilities of the forester package and things you can achieve with it. We will present you the main functions of the package with their parameters and show how you can use them in your problems.

Yesterday Olga Tokarczuk (2018 Nobel Prize in Literature) said in an interview that when she thinks about literature, she no longer thinks about books!!! So, how should we effectively tell the most important story in predictive modelling i.e. We (MI2DataLab) are currently working on an exciting and interdisciplinary experiment combining a classic textbook with a comic book, combining a description of methods and software with a description of process, combining a description of a specific use-case about COVID-19 data analysis with universal best practices. These 52 page long teaching materials describe how to build a predictive model, compare the developed models, and use XAI to analyze them, plus a bonus -- how to deploy model with explanations in a similar form to https://crs19.pl/. The material is prepared as a starter for predictive modelling. The included code examples can be executed and experimented with on your own (the first version has examples in R, but there will be albo translation for Python).

Machine learning is an important part of working in R. Packages like mlr3 simplify the whole process. Its no need to manually split data into training and test set, no need to manually fit linear models. Even more, packages like mlr3pipelines let you crate complex pipelines and include feature engineering and transformation in your models. R is also used by statisticians, from statisticians we have advanced methods of imputing missing data like mice or Amelia. What happens when we want to connect machine learning with a statistical approach.