💻 Week 02 - Class Roadmap (90 min)

2022/23 Autumn Term

Welcome to the second seminar/lab class of DS101A.

This week’s content is more hands-on. We will work with Python to explore, pre-process and tidy data. You will see how data scientists use programming to make their work more efficient.

Step 01 - Nuvolos Cloud Account Setup (10 min)

You must all have received an invite to create an account on the Nuvolos cloud. Please follow the instructions below to set up the account:

1. Click on the ‘Review Invitation’ button.

2. It will redirect you to the Nuvolos Cloud website, where it will ask you to join. After clicking the ‘Access through your institution’ button, please use your LSE email address.

3. On the next page, ‘Find your institution’, please enter ‘London School of Economics’ and select the ‘London School of Economics and Political Science’ option from the dropdown.

4. And sign in using your LSE email address.

5. In a few seconds, it will redirect you to the Nuvolos Dashboard page, as shown below:

6. Now, click on the ‘JupyterLab 4’ section.

7. This will open a JupyterLab session.

8. On the top-left corner, click on the ‘+’ icon.

9. It will open a window on the right-hand. Please select the ‘Python 3’ option from the Notebook section.

10. It will open an ‘Untitled.ipynb’ notebook. [Note: .ipynb is the extension for jupyter python notebooks].

11. To rename your file, right-click on the ‘Untitled.ipynb’ and it will give you an option for it, as below:

12. Voila! Your first Jupyter Notebook is ready to go!

Step 02 - Reflection on the data tidying task (20min)

• Share your CSVs with others on your class group Slack channel (i.e #class-group1 or #class-group2 or #class-group3).
• In groups, discuss how you collected data from Wikipedia and how your tidy data representations diverge.
• Share your observations with the wider class. What did you learn? Any particular difficulties/challenges encountered when tidying data?

Step 03 - Tidying data programmatically (45-50 min)

• Your class teacher will demonstrate how one could use Python (and packages called BeautifulSoup and pandas) to extract data from Wikipedia and tidy that data. The script that was shown in class can be found here
• You can try and follow along the Python program demonstration on Nuvolos.