This repository provides a Docker-based environment to get started with the Lightning-Fast Modulation Classification with Hardware-Efficient Neural Networks problem statement of the ITU AI/ML in 5G Challenge. The sandbox environment includes PyTorch and Brevitas and serves a Jupyter notebook that guides you through definition, training, and evaluation of an exemplary quantized CNN model.
The sandbox was tested on Ubuntu, but the containerized setup should work on most platforms.
sudo
you should follow these instructions to add your user to the docker group
DATASET_DIR
: This directory will be mounted inside the container at "/workspace/dataset", download instructions can be found inside the Jupyter notebookDOCKER_GPUS
: Select GPUs which will be accessible from within the container, for example all
or device=0
JUPYTER_PORT
: Override default port (8888)NETRON_PORT
: Override default port (8081)JUPYTER_PASSWD_HASH
: Override default password ("radioml")LOCALHOST_URL
: Set the IP/URL of the machine if you don't access it via localhost
./run_docker.sh
inside sandbox/
to launch the Jupyter notebook server
./run_docker.sh bash
to launch an interactive shellhttp://HOSTNAME:JUPYTER_PORT
from a browser and login with password "radioml"Connect with the challenge organizers and other participants on GitHub discussion. For questions related to quantization-aware training with Brevitas, there is also a separate Gitter channel: