geosci-course

Resources for running a course using geosci-labs. To manage the course, we will use nblibrarian

why

geosci-labs is a library of notebook-apps for applied geophysics. We hope that they are useful for your courses! There are a large number of notebooks in geosci-labs and you may only want to use a subset of them to teach your course. This repository is intended to help you select the build and deploy a subset of these notebooks for use in your course.

scope

There is very little code in this repository; the repository is meant to serve as a template for deploying a course that uses the notebooks in geosci-labs, though we may one day expand to improving functionality for notebooks from other repositories as well.

The two main components of this repository are:

• Makefile: command line tools for setting up the repository and keeping it up to date with advancements in geosci-labs
• notebooks.yml: structured file for describing which notebooks you want included in your course and where they should be located.

usage

Step 1: create your own fork of this repository

• First, you will need to setup a github account if you do not already have one.
• Then, fork this repository so that you have your own copy.

Step 2: clone and install software requirements

• Clone the repository. You can copy the below line and replace {your-username} with your github username

  git clone https://github.com/{your-username}/geosci-labs
  

• change directory into the cloned repository

  cd geosci-labs
  

• install nblibrarian

  pip install nblibrarian
  

Step 3: specify the notebooks that you want downloaded

• edit the file .jupyter-include to specify which notebooks you want downloaded (see the instructions on nblibrarian

• run nblibrarian to setup the repository, this downloads the notebooks you specified in .jupyter-include

  nblibrarian 
  

  This also downloads an associated environment.yml file which explicitly specifies the exact versions of the direct software dependencies. This ensures that if you distribute your notebooks using binder.org or use conda environments to manage the software environment that is distributed to students, you can have more confidence that upstream changes will not break during the time you are teaching the course.

• to then run the notebooks, you can either create the conda environment

  conda env create
  

  and activate it

  conda activate ENVIRONMENT_NAME
  

  (you will need to look inside of the environment.yml file to see the environment name). Alternatively, you can pip install the requirements in the requirements file

  pip install -r requirements.txt
  

• Launch Jupyter

  jupyter notebook
  

Step 4: tailoring it to your course

• a template notebook, index.ipynb, is provided for you and this can serve as a landing-page for your course.
• once this repository has been set up, you may also want to edit the README so that it is specific to your course

Step 5: push your changes

• Before committing your notebooks, we recommend running kernel + Restart & clear output in the notebook in order to clear all output from the notebooks so that they are always archived and deployed in a state that does not include output.
• Then follow standard git procedure to commit and push your notebooks

  git add .
  git commit -m "commit message"
  git push
  

Step 5.5: updating the code and notebooks

The workflow is opinionated and assumes that you will use the notebooks as-is from geosci-labs. You are welcome to open issues and create pull requestsin geosci-labs to improve them. You are free forge your own path and tailor the notebooks to your course, in which case, you can ignore this section.

Before updating, make sure that you have committed and pushed any changes. You can always check if you have any uncomitted changes by running

git status

If everything is up to date, you can then update the various components of the course.

• If you have updated your .jupyter-include to add a new notebook from geosci-labs, you can run

  nblibrarian
  

  to fetch it. This will no overwrite the environment.yml, requirements.txt, or any of the exsisting notebooks.

• To fetch updated notebooks from geosci-labs, you can run:

  nblibrarian --overwrite=True
  

  Note that this will overwrite any changes you have made to the notebooks listed in .jupyter-include. It will not over-write and new notebooks that you have added that are commented-out or simply not specified in the .jupyter-include

For more details, see the nblibrarian repository.

Step 6: deploy your course

There are several options for deploying your course. Among them, here are a few of the more popular options:

• binder.org
• azure notebooks
• If you have a university-supported JupyterHub, then you can request support from the administrator.