Skip to content

Home

Jump to bottom
marysa edited this page Sep 4, 2022 · 8 revisions

Welcome to the ACDC2022 wiki!

There are folders here for both Python and Julia.

  • Python

To create a python environment with at least some of the packages that will be helpful to do some analysis, use the environment.yml file in the python folder of the github repo. If you're using a linux machine, try using the linux_environment.yml file instead.

If you don't have python running locally on your computer, try using anaconda to install and manage your python environments. You're welcome to use pip/anything else too, I'm just most familiar with anaconda so will be of limited help in sorting out environment issues with anything else!

If you are having problems with dependencies of the packages, remove:

  • esmf=8.2.0
  • esmpy=8.2.0

from the environment.yml file and try again. You don't really need those; they're good for regridding but you really don't need them otherwise.

  1. download anaconda: https://www.anaconda.com/products/distribution

  2. follow instructions from anaconda on how to clone and environment from a .yml file: https://conda.io/projects/conda/en/latest/user-guide/tasks/manage-environments.html#cloning-an-environment

  3. From the Anaconda Navigator, you'll have the option to download some additional software. I suggest downloading JupyterLab, JupyterNotebook, PyCharm, and Spyder before the summer school.

  4. Open the ACDC2022_ERA5.ipynb file in a JupterLab (or JupyterNotebook) session using the acdc2022 environment.

  5. Hopefully, the file will read in the .nc file also in this repo and make a plot. For the summer school, we'll have more data on a hard drive.

  • Julia

The julia/Project.toml file includes all the information needed to create a Julia environment with tools for reading files, statistics, linear algebra, and oceanography. Credit to Yoni Nazarathy for his JuliaCon 2021 presentation "Statistics from the Ground Up" which serves as the basis for this environment.

In order to configure the environment make sure that you have Julia executable installed from here. Julia 1.8.0 is the latest stable version.

One way to configurate the environment is to run as follows (you may also view the installation video):

  1. Clone this repository to a local folder on your computer (or download the zip file).
  2. Start Julia at the command line, or using Virtual Studio Code, or emacs, etc.
  3. If you don't have IJulia installed install it via (] puts you in package manager mode):
] 
add IJulia

Now hit backspace to exit the package manager.

  1. Start Jupyter Notebook with:
using IJulia
notebook()

The first time you run notebook you may be asked if to install Conda. Recommended to hit y.

  1. In the Jupyter environment that opens in your web browser, navigate to the correct folder which you downloaded, open the Sample.ipynb file, and run the first few cells up to the section Why Julia?. These cells will install the needed packages in the environment for the workshop. This will take considerable time but only needs to be run once. Note the line with using RCall is commented out. If you have R installed on your system you may uncomment it (delete #uncomment if using R: ).
Clone this wiki locally