In this tutorial you will launch a new virtual environment by providing only a URL to a GitHub repo, relying on a Jupyter technology called Binder to install all dependencies and launch Jupyter Lab.
Enter Binder URL
In ContainDS, click the ‘+ NEW’ button to access the new container screen. Click on the ‘Binder’ tab.
In ‘GitHub repository name or URL’ enter the URL
Make sure ‘GitHub’ is selected in the dropdown to the right. Leave the other two text fields empty, then click ‘LAUNCH’.
You will start to see some logs towards the bottom of the screen, showing that the repo is being downloaded and processed. A lot of the processing is carried out within various Docker containers which may be downloaded and run in the background.
You can see these running under the ‘Other’ section in the container list on the left-hand side of ContainDS. They can be safely ignored.
Once the final image is ready, it will appear under ‘Pending’ in the container list so you can confirm the name of the container and where to locate the workspace folder:
The folder should normally be a new folder that doesn’t exist yet, so just leave the default provided. Click CREATE.
ContainDS will create the folder and extract the files from the image into that workspace folder, before making the final container based on the original Binder-ready repo:
Click the ‘WEB’ button to launch Jupyter Lab in a browser, then open the ‘index.ipynb’ notebook. You can shift-enter through the sample notebook.
Find out more
You can read more about the options for starting a container through different Binder sources including local folders on your hard drive, as well as other ways to create new containers in New Container Options.
Managing Containers will explain how to manipulate your container now that it is running.