The primary purpose of the ACCESS-NRI Model Live Diagnostics package is to provide a simple, easy to use and accessible framework for the ACCESS modelling community to check, monitor, visualise and evaluate model behaviour and progress on currently running or ‘live’ ACCESS models on the Australian NCI supercomputer Gadi. In addition to monitoring a live model, the package provides the functionality to load, visualise and compare reference ACCESS model data with the selected live user model.
This first official Model Live Diagnostics release utilises the ACCESS-NRI Intake Catalog and provides support for monitoring the following ACCESS model configurations on Gadi:
- Earth System Model 1.5: ACCESS-ESM1.5
- Model for Ocean/Sea Ice: ACCESS-OM2
- Coupled Model 2: ACCESS-CM2
As Model Live Diagnostics runs within an ARE JupyterLab environment on Gadi, so before getting started started you will need an active NCI User Account and membership to a NCI project with available compute allowance.
The med-diagnostics package is pre-installed in the Gadi ACCESS-NRI MED conda environment. To invoke this environment for your ARE JupyterLab session simply set the following parameters in the ARE JupyterLab ‘Advanced options’:
Module directories | /g/data/xp65/public/modules
Modules | conda/access-med
conda | conda install -c accessnri med-diagnostics
PyPI | pip install med-diagnostics
To use the med-diagnostics package in your Python scripts, simply use:
An example Jupyter notebook describing usage options can be found in the Github repo docs/notebooks directory.
The full documentation is available from readthedocs.