🎰🎰🎰 hackfest-ppl 🎰🎰🎰

Install dependencies

This step should work for everyone:

git clone https://github.com/csiro-mlai/fno_inversion_ml4ps2021.git
cd fno_inversion_ml4ps2021

Now, install the requirements. Local desktop, bash shell:

python3 -m venv --prompt fno_inversion_ml4ps2021 ./venv
source ./venv/bin/activate
pip install -r requirements.txt
jupyter lab operator_inversion.ipynb

This should work for Linux, macos, or Windows Susbystem for Linux. For windows native, you are on your own, good luck.

Choosing regularisation

To experiment with selecting the regularisation parameter you will need a large validation data set. Here is one.

wget https://cloudstor.aarnet.edu.au/plus/s/FblQ6LxQtCosPkq/download -O ./data/grf_forcing_mini_1.h5

Graphical models

If you wish to additionally visualize graphical models, you need graphviz. Depending on your platform this will be something like

brew install graphviz       # MacOS with homebrew
conda install graphviz      # anaconda
sudo apt install graphviz   # Debian/ubuntu/WSL default
# etc

Graphviz on Windows is complicated so once again, use WSL.

Developer setup

If you want to contribute back to this repository, please do. To keep the storage small(er) we strip out all the notebooks using nbstripout:

nbstripout --install --attributes .gitattributes

Authors

• Dan MacKinlay

With input from

• Alasdair Tran