No Install Options
The following options provide alternatives for those who are unable to install the full programming environment on their main computer and would otherwise be unable to continue with the module. Each of these has pros and cons but they allow you to run code ‘in the cloud’ and so mean that any internet-connected device can be used to write and run code.
UCL JupyterHub
Please see the JupyterHub page for information about configuring a new environment.
To connect to JupyterHub:
- Start up the UCL VPN.
- Connect to JupyterHub
- Authenticate using your UCL credentials.
- If you see a URL that ends in
tree?
please replace this withlab?
to get the JupyterLab interface and not the original Jupyter Notebook interface (you can bookmark this page once it opens). - Create a new terminal: File > New > Terminal
Note that you need to replace {group_name}
with the appropriate group role (this value will be obvious once you’re logged in and can run ls /shared/groups/
in the Terminal):
conda config --append envs_dirs /shared/groups/{group_name}/casa
jupyter contrib nbextension install --user
Google Colab
Google’s Colaboratory is an easy way to get started without needing to install anything, while still being able to save your work. This approach will work well for the first few weeks, up to the point where we start loading larger data sets that require more memory than Google’s free tier provides. At that point you’ll either need to start sampling from the full data set, or you’re back to Docker!
Here’s how to get started with Colab:
- Go to Google’s Colab site and sign in using your Google account details (or create an account if you don’t already have one).
- If you are trying to clone a practical from Foundations then: choose
File
thenOpen
and select theGitHub
tab from the open options (it’s betweenGoogle Drive
andUpload
). - Enter the following URL: https://github.com/jreades/fsds.
- You should now see a list of notebooks to choose from – select the appropriate one from the sequence and click
Open in Colab
. - You will need to trust the notebook in order to run it.
When you are done for the day you can save the notebook to your Google Drive – to re-run that notebook you can just load it from Google Drive, rather than going back to GitHub, but remember that you need to re-run all the cells up to the point where you saved the notebook. Being able to save and re-run a notebook is the main difference from Binder.
Binder
While we support Binder for Code Camp, this is not a platform that we are supporting as part of our masters programme since it is impossible to save your work without downloading the notebook each time.