This tutorial aims to provide an introduction to machine learning and scikit-learn "from the ground up". We will start with basic concepts of machine learning and implementing these using scikit-learn. Going in detail through the characteristics of several methods, we will discuss how to pick an algorithm for your application, how to set its parameters, and how to evaluate performance.
Machine learning is the task of extracting knowledge from data, often with the goal to generalize to new, unseen data. Applications of machine learning now touch nearly every aspect of everyday life, from the face detection in our phones and the streams of social media we consume to picking restaurants, partners, and movies. I has also become indispensable to many empirical sciences, from physics, astronomy and biology to social sciences.
Scikit-learn has emerged of one of the most popular toolkits for machine learning, and is now widely used in industry and academia. The goal of the tutorial is to enable participants to use the wide variety of machine learning algorithms available in scikit-learn on their own data sets, for their own domains.
This tutorial will comprise an introductory morning session and an advanced afternoon session. The morning part of the tutorial will cover basic concepts of machine learning, data representation and preprocessing. We will explain different problem settings, and which algorithms to use in each. We will then go through some simple sample applications using algorithms implemented in Scikit-Learn, including SVMs, Random Forests, K-Means, PCA, T-SNE and others.
In the afternoon session, we will discuss setting hyper-parameters, and how to prevent overfitting. We will go in-depth into the trade-off of model complexity and dataset size. We will also discuss complexity of learning algorithms and how to cope with very large datasets. We will also go through the process of building machine learning pipelines, consisting of feature extraction, preprocessing and supervised learning.
I see a few small tweaks - let me try a PR. Overall it looks pretty good though!