The following UATs are designed to test some selected libraries contained within the NCI-geophys/23.04 environment.
To run these tests, you will first need to join the following NCI project codes via the mynci portal ( https://my.nci.org.au ):
my80: https://my.nci.org.au/mancini/project/my80
up99: https://my.nci.org.au/mancini/project/up99
qt52: https://my.nci.org.au/mancini/project/qt52
rt52: https://my.nci.org.au/mancini/project/rt52
dk92: https://my.nci.org.au/mancini/project/dk92
iv65: https://my.nci.org.au/mancini/project/iv65
Once you are approved as a member of these projects, you can next log on to Gadi and run the following commands:
$ module use /g/data/up99/modulefiles $ module load NCI-geophys/23.04 $ cp -r ${NCI_GEOPHYS_ROOT}/UAT ~/
The above commands will copy four Jupyter notebooks to your home directory. To check that these notebooks are available, run:
$ cd ~/UAT $ ls
and you should see the following Jupyter Notebooks:
Julia_examples.ipynb Python_examples.ipynb R_examples.ipynb Geophysics_data_indexing.ipynb
Now we can exit Gadi and launch a JupyterLab session on the Australian Research Environment (ARE - https://are.nci.org.au ). For more information on ARE, please refer to the ARE User Guide.
First log in to ARE using your NCI user name and password:
Next, click on the JupyterLab app:
To run these examples, let's request:
Walltime (hours): 2 |
Under the “Advanced options…” tab, add in the following Module:
Module directories: /g/data/up99/modulefiles |
Now click the “Launch” tab and your JupyterLab job will join the queue:
Once the JupyterLab is running, click the “Open JupyterLab” tab:
On the left filesystem pane of the JupyterLab session, you will need to navigate to the directory containing the four Jupyter Notebooks that were copied over. Click on the /home folder and search for the UAT subfolder:
This is where you should find the four test Jupyter Notebooks:
Open up Python_examples.ipynb and start running through this Jupyter notebook by clicking the “run selected cells” button in the top panel:
Note that after Cell 21, there is a step that requires you to run an Occam inversion either on Gadi or using JupyterLab. To do this within JupyterLab, open up a Launcher Tab and click on the “Terminal” app:
Then follow the steps in the notebook to run the example Occam inversion:
Once this Occam inversion is complete (it should only take a few minutes), please continue running the remaining cells in the Python_examples notebook.
Finally, you can also try running through the R_examples, Julia_examples and Geophysics_data_indexing notebooks: