Supporting code and data: Simulating multivariate hazards with generative deep learning

This repository contains all code to reproduce the figures in the paper Simulating mulitvariate hazards with generative deep learning. Note that due to stochasticity in the StyleGAN training, some results may vary slightly. The purpose of this repo is to demonstrate how the results were generated, but the model will be published as a snakemake workflow later. Installation git clone [e-mail]:alisonpeard/hazGAN.git cd environments mamba create -f apple-silicon.yaml # or yaml for other OS env mamba activate hazGANv0 The path to the data folder hazGAN-data is set in the .env file in the parent directory. Change it to point to the location of the extracted data zipfile. The StyleGAN model is included in the styleGAN-DA folder. This model requires an nVIDIA GPU to run. The environment definition in environments/linux-gpu.yaml creates an environment that works with the model on linux machines.