Logistic regression (Part 1)
Logistic regression (https://en.wikipedia.org/wiki/Logisti...) is one of the most fundamental binary classification algorithms. It uses a linear transformation similar to the linear regression algorithm, but it adds the logistic function on top of it, so the output is always between 0 and 1. This allows us to interpret its output as confidence of predicting 1 or 0. In addition to it being a popular machine learning algorithm in its own right, logistic regression can also serve as a stepping stone to understanding deep neural networks.
⏱️TIMESTAMPS
0:00 - Intro
0:50 - Example of logistic regression
2:46 - Why sigmoid?
4:40 - Making predictions
Credits:
All animations were made using the manim library: https://docs.manim.community/en/stable/
Photo of Pierre Francois Verhulst source: https://commons.wikimedia.org/wiki/Fi...
Penguin image source: https://openclipart.org/detail/27563/...
"Little penguin" by "lemmling"
Puffin image source: https://openclipart.org/detail/23564/...
"A puffin bird" by "Aki G. Karlsson"
Student image source: "https://openclipart.org/detail/297144..."
"student" by "jabernal"
Book image source "https://openclipart.org/detail/169379..."
"open book" by "hatalar205"
The exam example is taken from the Wikipedia page about logistic regression: https://en.wikipedia.org/wiki/Logisti...