January 13, 2022
Presenter: Alfred Essa
In this presentation we will study some Python code from both a computational and aesthetic lens. We will write code together for an artificial neuron, using simple functions and then building to a Neuron class. Our canvas will be the Numpy library. Neurons are computationally simple. A single neuron is a simple mathematical function. But it can yield complex models, such as multiple regression and logistic regression in machine learning. Artificial neurons are also the atoms of deep learning. The aim of the presentation is to demonstrate how simple computational patterns can yield complex representations.
Prerequisites: Basic Python
The presentation will be in the form of Jupyter Notebooks. I will prepare MyBinder so you can interact with the code during the presentation. The notebooks will also be on GitHub if you want to download them.
Link to materials: https://alfredessa.github.io/aes/intro.html
Meetup link: https://www.meetup.com/bostonpython/events/283085024/