X-AI workshops by Daniel Huynh
These lessons come in notebooks to introduce the basic ideas behind machine learning, from the pure maths to the code. Practice is put forward, and we will see how to code simple and complex algorithms with various libraries such as Sklearn, Pytorch, and how to share the results with plots and even a simple REST API with Flask.
We will introduce the most basics concept with the two most simple models : linear and logistic regressions, and see how we can code them from scratch using numpy. Then we will see how it is done in Sklearn and deploy a Flask app on Azure.
We will see the fundamentals of Neural Networks, and see how Pytorch handles most of the tedious stuff. We will code a simple Feed Forward NN, then a Convolutionnal NN on MNIST data. Then we will cover VAEs on MNIST, finally we will use Fastai framework to do transfer learning on a real world dataset.
We will see quickly how Pytorch works, and have a quick dive into NLP with Word2Vec.