Visualizing & Communicating Results in Python with Jupyter

Description

Code and run your first Python program in minutes without installing anything!

This course is designed for learners with limited coding experience, providing a foundation for presenting data using visualization tools in Jupyter Notebook. This course helps learners describe and make inferences from data, and better communicate and present data.
The modules in this course will cover a wide range of visualizations which allow you to illustrate and compare the composition of the dataset, determine the distribution of the dataset, and visualize complex data such as geographically-based data. Completion of Data Analysis in Python with pandas & matplotlib in Spyder before taking this course is recommended.
To allow for a truly hands-on, self-paced learning experience, this course is video-free.
Assignments contain short explanations with images and runnable code examples with suggested edits to explore code examples further, building a deeper understanding by doing. You’ll benefit from instant feedback from a variety of assessment items along the way, gently progressing from quick understanding checks (multiple choice, fill in the blank, and un-scrambling code blocks) to small, approachable coding exercises that take minutes instead of hours. Finally, an accumulative lab at the end of the course will provide you an opportunity to apply all learned concepts within a real-world context.

What you will learn

Creating Comparison and Composition Charts
Creating Distribution Charts
Creating Specialized Visualizations
Communicating Data Using Jupyter notebook

What’s included