Overview

If you are looking for the right way how to organize your data science projects, and the right tool and workflow according to the latest trends, you are on the right place.

In this course you will learn everything you need to be comfortable to work on your data science projects. Primary, you will get detailed knowledge of the concepts and possibilities of the most used tool currently, jupyter, its extension jupyter lab, how to set jupyter hub to work with multiple users. You will get the basics on how to work with docker containers. Main programming language will be python, but we cover the foundations on how to work with other programing languages in jupyter as well, like R and Julia. We will show you how to work with it on cyber security use cases, how to manipulate data which fits into RAM with pandas, but we show you the options on how to work with big data too using dask. You will learn how to effectively visualize data with static or interactive charts in jupyter, and ways on how to present your work with various export options.

Machine learning is one of the ultimate topics today, and it will not be left out. You will learn basic concepts of selected supervised and unsupervised ML algorithms, how to build optimal ML pipelines using sklearn, starting from data preparation, variables transformations and selections and the final model. You will be able to build by yourself “good enough” models with minimum effort and theory, but with maximum practical usability using state-of-art algorithms like xgboost and lightgbm, and you will get the foundations on how to automatically tweak the hyperparameters of the pipeline to find the best ones. All of it in jupyter, jupyter lab and python.