This document will detail how to use an environment to access AWS files

In order to use the code examples along with this archive, it is necessary to have full root write access to your package installation path.

The process for setting this up will vary depending upon your system. In some cirumstances, you may just need to reach out to your Sys Admin to configure this for you. Otherwise, the recommended process is to build a miniconda environment. You can use this command to get a new miniconda environment:

wget 'https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh'

Once you have miniconda, navigate to this path to activate it:

/miniconda3/bin/activate.sh

Run the script to activate your miniconda. Now, you will need to create an env specifically for the packages you need for this code base. You can do that with this command:

conda create --name myenv # Replace my env with the name

Once you have the envrionment created, you will also need to source the env so that Python uses it for code execution. That is done by running this command:

source activate /absolute/path/to/env/myenv

If run successfully (either via the Anaconda shell or via a linux bash terminal) the result should be a set of () with the env name in front of your id like so:

(myenv) zach@utah.edu bin

Before doing any package installations, I would strongly recommend adding conda-forge as a source. If you aren;t familiar with conda-forge, it is a robust library of packages for python that is far more extensive than the default source environments. You can add it by running this command:

conda config --add channels conda-forge

Your final step to be able to run the examples would be to install all of the reqired packages to your environment via conda install. Here is a recommended list to start:

-numpy -xarray -h5netcdf -pandas -numcodecs -netcdf4 -zarr -matplotlib -cartopy -pygrib -s3fs -Blosc -metpy -python-rclone -cfgrib