This repository contains slide decks, programming exercises, and links to recorded lectures videos for the course "Natural Language Processing with Deep Learning" (Ruhr University Bochum, winter term 2024/2025).
This course is lectured by Prof. Dr. Ivan Habernal.
The slides are available as PDF as well as LaTeX source code (we've used Beamer because typesetting mathematics in PowerPoint or similar tools is painful). See the instructions below if you want to compile the slides yourselves.
The content is licensed under Creative Commons CC BY-SA 4.0 which means that you can re-use, adapt, modify, or publish it further, provided you keep the license and give proper credits.
Note: The following content is continuously updated as the winter term progresses.
To see the previous full 2023/24 course content, checkout the latest 2024 commit.
Subscribe the YouTube playlist to get updates on new lectures: (TO BE DECIDED/UPDATED)
2024-MM-DD
2024-MM-DD
- See the PDF (including LaTeX source) and Python code under exercises/ex01
- What are some essential pre-requisites?
- Math: Derivatives and partial derivatives. We cover them in Lecture 2. If you need more, I would recommend these sources:
- Jeremy Kun: A Programmer's Introduction to Mathematics. Absolutely amazing book. Pay-what-you-want for the PDF book. https://pimbook.org/
- Deisenroth, A. Aldo Faisal, and Cheng Soon Ong: Mathematics for Machine Learning. Excellent resource, freely available. Might be a bit dense. https://mml-book.github.io/
- Math: Derivatives and partial derivatives. We cover them in Lecture 2. If you need more, I would recommend these sources:
- Where do I find the code for plotting the functions?
- Most of the plots are generated in Python/Jupyter (in Colab). The links are included as comments in the respective LaTeX sources for the slides.
If you run a linux distribution (e.g., Ubuntu 20.04 and newer), all packages are provided as part of texlive
. Install the following packages
$ sudo apt-get install texlive-latex-recommended texlive-pictures texlive-latex-extra \
texlive-fonts-extra texlive-bibtex-extra texlive-humanities texlive-science \
texlive-luatex biber wget -y
To install MS Segoe UI fonts required by the beamer template locally, run the following script (works also similarly on Mac OS): https://gist.github.com/habernal/ad1085ce5dc5e8cb3fbead354d8f4190
Run the script compile-pdf.sh
in each lecture's folder to produce both handouts as well as unfolding PDFs used in the lecture.
If you don't run a linux system or don't want to mess up your latex packages, I've tested compiling the slides in a Docker.
Install Docker ( https://docs.docker.com/engine/install/ )
Create a folder to which you clone this repository (for example, $ mkdir -p /tmp/slides
)
Run Docker with Ubuntu 20.04 interactively; mount your slides directory under /mnt
in this Docker container
$ docker run -it --rm --mount type=bind,source=/tmp/slides,target=/mnt \
ubuntu:20.04 /bin/bash
Once the container is running, update, install packages and fonts as above
# apt-get update && apt-get dist-upgrade -y && apt-get install texlive-latex-recommended \
texlive-pictures texlive-latex-extra texlive-fonts-extra texlive-bibtex-extra \
texlive-humanities texlive-science texlive-luatex biber wget -y
Install fonts as above: https://gist.github.com/habernal/ad1085ce5dc5e8cb3fbead354d8f4190
Run the script compile-pdf.sh
in each lecture's folder to produce both handouts as well as unfolding PDFs used in the lecture, which generates the PDFs in your local folder (e.g, /tmp/slides
).