Exploratory Analysis of a Public Dataset

You will use

• PD-03Reading and Writing CSV
• PD-04Select and Filter Rows
• PD-06Finding and Handling Missing Values
• PD-08Group-by Aggregation
• PD-10Merging Tables

Milestones

Build the smallest working version first. Ticks are saved in this browser only.

1. 01Choose a dataset and write down three questions before looking at the numbers.
2. 02Load it, inspect dtypes and missing values, and fix types and obvious errors.
3. 03Answer each question with groupby, filtering or a merge with a second table.
4. 04Make one clear chart per question with pandas plotting or matplotlib.
5. 05Write a one-page summary with the answers, the charts and the caveats.

Stretch goals

• Add a time-series view with resampling.
• Turn the analysis into a script that regenerates the report from the raw file.
• Publish the report as an HTML page.

Data

Run it yourself

pip install pandas matplotlib
python eda.py