Learning Python for Data Analysis and Visualization is a great Udemy course that has a lot more than the title implies. Besides reviewing the key concepts and libraries to do data analysis, it also has a few great notebook projects that guide you through doing projects on different datasets.
The examples provided cover distinct areas such as:
- The famous Kaggle Titanic dataset
- 2012 Election Donor datasets analysis
- Basic Stock market analysis of technology stocks
The course has a few years but it’s content is still relevant (and I think it will continue to be for a very long time). It was built using Python 2 and with the most recent pandas, matplotlib and seaborn versions available at the time.
When I started to go through the course I set the goal of using Python 3 as well as the most recent versions of the libraries, and thus my challenge became not only to do the full course but also to close the gap, as much as possible, between the original notebooks and the ones that I built.
As you may expect, it was not possible to replicate exactly the same outputs as the course notebooks, and this was mainly due to the following:
- Both matplotlib and seaborn have updated their themes colours and styles.
- Some of the functions on seaborn have been deprecated, thus replacements were needed
- Some data sources extraction methods are no longer available, such as the Yahoo stock feed.
Nevertheless, the final result was pretty close and it was fun to find solutions to close the gaps between the original course material and the notebooks I was building. Yes, it increased a bit the exercises’ challenges but it was definitely worth the effort.
Hope you find it useful and have fun!
