fastai/fastbook

This repository contains the Jupyter Notebook source files for the book "Deep Learning for Coders with Fastai and PyTorch" by Jeremy Howard and Sylvain Gugger. Jupyter Notebooks are interactive documents that mix explanatory text, code, and output, you can read the explanation and then run the code right in the same document. The book introduces deep learning (a branch of machine learning that uses layered neural networks) using the fastai library, which sits on top of PyTorch and makes common deep learning tasks much simpler to code. The notebooks are the same material used for the fast.ai MOOC (a free online course) at course.fast.ai. They cover topics from basic image recognition, natural language processing (teaching computers to understand text), tabular data analysis, and recommendation systems, all the way to building neural network architectures from scratch. The approach is deliberately top-down, you build working models early and understand the theory later, which is the opposite of traditional academic courses. You'd use this if you want to learn deep learning practically without a math-heavy academic background. The recommended way to start is opening the notebooks in Google Colab, a free browser-based environment where you can run the code without installing anything on your own computer. The physical book is available for purchase separately on Amazon; the notebooks are free to read and run but have copyright restrictions on redistribution. Written in Python using fastai and PyTorch.