Course website: https://uvadlc.github.io/
Course edition: DL1 - Fall 2024, DL2 - Spring 2023, Being kept up to date Documentation: https://uvadlc-notebooks.readthedocs.io/
Repository: https://github.com/phlippe/uvadlc_notebooks
Recordings: YouTube Playlist https://www.youtube.com/playlist?list=PLdlPlO1QhMiAkedeu0aJixfkknLRxk1nA
Author: Phillip Lippe
This directory contains a selection of notebooks from the 2025 PyTorch deep learning course of the University of Amsterdam, adapted to run on the ScienceData Kubernetes service and save/load data to/from ScienceData.
The notebooks download data files from various sources. These downloads can be quite time consuming. To save time, or if the data source is unavailable, you can grab all downloaded data from ScienceData by firing up a shell on your notebook server and running:
curl -LO https://sciencedata.dk/public/kubefiles_public/pytorch_data_uva/data.tar
tar -xvf data.tarYou can grab all saved models by running:
curl -LO https://sciencedata.dk/public/kubefiles_public/pytorch_data_uva/saved_models.tar
tar -xvf saved_models.tarIMPORTANT: The notebooks should be opened from a pod running the image jupyterlab_sciencedata_pytorch.
For the originals, see the links above.