Principles of Beautiful Data Visualization (2022)

Discover Python data visualization libraries such as Muttalib, Seabourne, and Gotlib, compare low- and high-level options, and learn to create interactive and animated charts with Plotline.

Explore loading and cleaning a penguins dataset in Python using Jupyter notebooks, handle missing values with a simple imputer, and compare visualizations with Matplotlib and Seaborn.

Explore common plots for data visualization, including histograms, density, bar charts, pie charts, tree maps, and scatterplots. Learn how these visualizations reveal amounts, distributions, proportions, and associations.

Explore linear, logarithmic, and curved scales for plotting data, including axis transformations, polar coordinates, and how figure size emphasizes changes and handles skewed distributions.

Explore qualitative and sequential color scales for data visualization, using distinct colors without order to distinguish categories, and ordered hues to convey magnitude.

Explore divergent color scales that balance a light midpoint with darker endpoints. Learn how accent color scales pair subdued and saturated colors to highlight select data.

Apply the less is more principle to data visualization, keeping displays simple and using fewer colors to achieve beauty.

Visualize penguin species counts with a seaborn bar plot in Python, customizing colors, order, and axis labels for a clear, well-labeled data visualization.

Learn to build bar plots with seaborn, adjust color palettes and axis labels, rotate labels or switch to horizontal bars, and refine figure size and borders for cleaner visuals.

Explore dot plots as an alternative to bar charts, noting how axis ranges affect interpretation, and learn to create vertical or horizontal dot plots that emphasize counts and order.

Improve bar plots by removing x axis labels and annotating counts next to bars, placing labels at bar centers, and fixing coding errors to produce a clearer figure.

Visualize data as a heat map by month and year to show passenger counts with Seaborn. Pivot the data, adjust color maps such as blues, and format values as integers.

Explore how histograms visualize distributions in the penguins dataset by adjusting bins and coloring by species. Incorporate kernel density estimates and consider log scale to compare multiple numerical distributions clearly.

Explore density plots as an alternative to histograms using seaborn to visualize distributions, compare bandwidth settings, and display multiple distributions side by side with density estimates.

Learn how box plots visualize distributions, alongside histograms and density plots, using the penguins dataset, highlighting median, quartiles, whiskers, and outliers.

Explore violin plots to visualize distributions and density estimates, using the penguins dataset and comparing them to box plots, with options like horizontal orientation and strip overlays for clearer visualization.

Explore how to visualize proportions with pie charts, noting there is no perfect chart and when tree maps may help for many slices.

Continue refining pie charts by converting to donut charts, adjusting inner and outer radii, removing explosions, and displaying percentages outside the donut for clearer data with improved title placement.

Explore proportion visualization of penguin species using side-by-side and stacked bars with the plotline library, transforming counts into percentages, customizing colors, and cleaning the figure for clear, interactive data storytelling.

visualize proportions with sankey diagrams by converting a penguin species–island dataset into a matrix-like frame, filling missing values with zeros, and mapping species to islands with labeled nodes and links.

Visualize proportions with three tree maps using the Titanic data set and stratify, labeling class and passenger counts with a color palette.

Learn to visualize associations with scatterplots in Seaborn using the penguins dataset, customize markers, colors, and sizes, place legends outside the plot, and overlay kernel density estimates for richer insights.

Learn how to use scales and redundancy to improve data visualizations, encoding penguin species with color and shape in a scatter plot and in a bubble chart.

Explore 3d scatterplots to visualize relationships among three features using the penguins dataset, color-coding by species and enabling hover to read x, y, z values.

Highlight data in scatterplots by changing the color palette to emphasize a specific group while greying others, and apply this approach to barcodes to tell a clearer data story.

Use a pair plot to visualize relationships among numerical features with scatter plots, diagonal histograms, and diagonal density plots, using species to reveal separations and overlaps.

Visualize time series data with line graphs of NYSE prices for Google, Amazon, and Microsoft. Filter to three companies and plot separate lines using Matplotlib and Seaborn.

Apply trendlines to emphasize overall stock price trends using a regression plot in seaborn, plotting the Google price and adjusting the x-axis to the index for clarity.

Visualize trendline uncertainty by adjusting the confidence interval and polynomial order, using a compact 300-point data subset; observe how higher orders cause overfitting while lower orders often generalize better.

Animate data visualizations by importing celluloid, creating a figure and camera, and progressively plotting date versus closing price to build a gif from snapshot frames.

Celebrate completing this course and apply principles like simplicity, redundancy, and highlighting figures to craft beautiful data visualizations.