This repository contains useful resources for the Stanford CS25 course, including homework solutions and reading notes.
- Read through The Illustrated Transformer — a visual guide to the Transformer architecture
- Implement an exercise from The Annotated Transformer. Note that code contains some bugs, so you may need to fix them.
torchtext
had a bug with expired hash for the dataset, so you may need to install the latest version from source (which will require also compile torch, which isn't trivial) or monkeypatch it.- Walkthrough of the model in the comments of resuling
transformer
library.