Guided Projects and How to Save your Work

While you’re learning with Dataquest, you want to make sure your work is saved - and we do too! Here are some best practices for avoiding losing your work while you’re working on any missions with Jupyter Notebook.

Things you may encounter

1. Jupyter Notebook Is Not Saving The File

Jupyter notebook is an awesome open-source tool, and it has a great auto-save and manual saving feature and is under normal circumstances pretty reliable. Unfortunately, there can be some saving issues are due to an unstable connection. If the connection is unstable, there is a chance the auto save will not work.

How to avoid this issue

Always try to Save And Checkpoint before you close the Guided Project mission or before you leave the project idle to do some other work. This step is essential because it helps you to identify whether you are experiencing a connection issue or not. Here is an example:

As you can see, when we performed File → Save and Checkpoint, because there was a connection issue, it created a small error message. If you think your connection is stable again, then click on Kernel → Reconnect and try saving once again. If you received “Checkpoint created” message like this:

Then all good! If you are still getting an error. Then copy all of your notebook contents to make sure you don’t lose the code.

2. Lost Code Despite Saving Manually

Although this can be a frustrating thing to see - it does happen. We recommend you to take time to back up your projects every single time you work on it.

You can refer to this post for instructions:

3. Opening Multiple Projects At Once

Having multiple projects open at once may seem convenient - especially if you’re taking a look at them for your portfolio - but if you save any Guided Project while you have several open, each Guided Project you have open will be overwritten by the project you’ve saved. So when you accidentally save any of the Guided Projects, all of the opened Guided Projects are overwritten.

How to avoid this issue

Only have one Guided Project open at a time! This will prevent any overwriting and create peace of mind for you.

Other Alternatives

Although there are advantages to using Jupyter Notebook on our platform, there are also other alternatives you may choose to use.

1. Installing Jupyter Notebook Locally

Installing a Jupyter Notebook on your machine is a great way to be sure your work is saved. To download the datasets from our Guided Projects, please click the Download button at the top of the Jupyter Notebook interface in our platform.


It will download a .tar file that contains both your notebook file and the required datasets. Now extract the .tar file and paste the datasets in the same folder as your Jupyter notebook file. Now all you need to do is open your locally installed Jupyter notebook and our Guided Project mission in a split-screen mode like this:

If the space taken by the Jupyter notebook is making it difficult for you then you can try running this JavaScript Code in the browser console:

document.getElementsByClassName("Pane1")[0].style.width = "100%";

It will hide the embedded Jupyter notebook like this:

Please note that you have to do this every time you refresh the page.


  • Offline Support
  • No Connection Issues
  • Performance is dependent on your machine


  • Cannot easily continue your work from a different computer
  • Troubleshooting issues can be difficult
  • Performance is dependent on your machine

2. Using Google Colaboratory

Google Colaboratory is an excellent alternative if you don’t want to install the Jupyter notebook on your computer. You can use it in the same manner as you would use the Jupyter notebook installed locally.


  • Highly Stable
  • High Performance
  • Easy To Share


  • Requires Internet