AKPIK Workshop | Resource-Aware Deep Learning: Tracking Energy Consumption in Scientific AI Applications
This repository contains all files for the DPG Spring Meeting 2026 AKPIK Workshop Resource-Aware Deep Learning: Tracking Energy Consumption in Scientific AI Applications.
You can find the slides for the presentation inside the slides directory. Before building the slides
with the provided Makefile, make sure to install the pygments style located in slides/pygments_style
using pip. Then call make inside the slides directory.
To participate in the hands-on session, make sure to install the virtual environment we provide
in this repository, e.g. using uv or conda/mamba:
$ uv pip install -r requirements.txt
$ mamba env create --file=environment.yml
The repository contains two notebooks for the hands-on session. resource_awareness.ipynb contains a template
for the tasks during the session, resource_awareness_solution.ipynb contains the solution to the tasks.
The notebooks will teach you one way to track the emissions of your code/deep-learning models.
Additionally, we provide a small script visualise.py which will allow you to visualise your results
using STREP.