Data Visualization and Modeling in Python

Overview

Put the keystone in your Python Data Science skills by becoming proficient with Data Visualization and Modeling. This course is suited for intermediate programmers, who have some experience with NumPy and Pandas, that want to expand their skills for any career in data science. Whether you come to data science through social sciences and Statistics, or from a programming background, this course will integrate the two perspectives and offer unique insights from each. You’ll begin by becoming adept with matplotlib, an essential plotting library in Python that will enable you to discover and communicate insights about data effectively. You’ll progress to classification algorithms by creating a K-Nearest Neighbors (KNN) classifier, a foundational algorithm used in data science and machine learning. Finally, you will write Python programs that leverage your newfound data science skills based on inferential statistics, and be able to describe relationships between variables in your data. By the end of the course, you’ll be able to quickly visualize a dataset, explore it for insights, determine relationships between data, and communicate it all with effective plots. In the last module of this course, you’ll produce a publication-quality figure based on data that you’ve prepared and cleaned yourself; the first artifact in your data science portfolio. Throughout this course you’ll get plenty of hands-on experience through interactive programming assignments, live coding demos from data scientists, and analyzing the data behind important real-world problems (like carbon emissions, real estate prices, and infant mortality). Guided activities throughout each module will reinforce your proficiency with data science techniques and analytical approach as a data scientist. Solidify your understanding of these critical data science concepts and begin your data science portfolio by mastering visualization and modeling. Start this integrative and transformative learning journey today!

Syllabus

Plotting

In this module, you will learn about plotting in Python—an important technique for exploring a dataset, and an indispensable tool for communicating insights. We’ll learn to make all the most common types of plots used in data science including the basics like line, bar, and scatter plots, as well as more advanced plot types including histograms and heatmaps. We’ll learn both how to make these plots and how they can be customized for your needs using a core plotting library for python, matplotlib, which serves as the backbone for many python plotting tools. You’ll learn how to create professional, accessible, and information-rich plots, which you will leverage to quickly identify trends in data that would be difficult to otherwise recognize. We've also included some optional additional readings if you want to further enhance your learning!

Prediction

This module, you will learn the basics of how to use code to make predictions based on data. After discussing what prediction is, you’ll learn to describe the concepts that underlie predictive algorithms within the context of the K-Nearest Neighbors (KNN) algorithm for both classification and regression. Additionally, you’ll learn to evaluate the accuracy of a predictive algorithm to assess its ability to generalize to new data. You will build your own KNN classification and regression algorithms from scratch and make predictions with each of them. At the end of this module, we’ll have a quiz to give you the opportunity to evaluate your understanding of predictive algorithms and reflect on your experience implementing your own.

Regression

This module, you will learn how to describe the differences between prediction and inference, two key Data Science concepts. You’ll learn how to implement linear regressions — one of the most useful tools that data scientists have for inference and prediction — and other statistical models in Python. You’ll apply this knowledge by examining a dataset and regressing multiple variables on each other, and describing the insights on their relationships.

Final Project

This module, you’ll bring together the concepts and skills you’ve developed throughout the course to create a final project for your data science portfolio. You’ll recreate a now-famous data visualization that illustrates the relationship between the income of countries and their greenhouse gas emissions on a global scale. To do this, you’ll explore and prepare 4 datasets and merge them into a composite dataset that you’ll plot. Creating this merged dataset is an important step, and you’ll validate your merged dataset with a short quiz on the insights within. The end result of this effort will be a publication-quality plot that makes a compelling point about the relationship between emissions and income—an impactful visualization that showcases your growing programming skills for data science applications.