EWC Automation of ECMWF AI stacks

This repository contains Ansible playbooks for customising EWC instances with specific AI stacks, including:

• ai-models: featuring the most popular data driven forecast models such as Pangu Weather, Fourcastnet or graphcast.
• aifs-single-mse: featuring the ECMWF AIFS Single data-driven model
• aifs-ens-crps: featuring the ECMWF AIFS ENS CRPS data-driven model
• Anemoi framework: leverage the Anemoi framework to develop and run your AI-models or manage datasets.
• ML Basic: featuring an environment with the main machine learning packages like scikit-learn or torch.

Getting started

• Install ansible and other dependencies. You may want to do it in its own virtual environment (pip install -r requirements.txt)

• Fetch the external requirements

  ansible-galaxy role install -r requirements.yml roles/

• Define your inventory in inventory

• Run the apropriate playbook

  ansible-playbook -i inventory playbookname.yml
  

License

Apache 2.0.

Author Information

ECMWF for the European Weather Cloud