GIF Tutorial in R
Michael Siebel
December 2020
Eye-Catching Graphs:
Creating a GIF in R
Using ggplot and gifski
Introduction
This was a component of a tutorial I presented at the offices of Fors Marsh Group.
Below, we looked at a few GIF vinettes featuring three potential utilitarian purposes. Then we viewed some applied examples. Finally, we looked line-by-line at the code ourselves.
For pronouncation guide, see Conclusion.
Setup
The two packages we need our
library(pacman)
pacman::p_load(tidyverse, ggplot2, gifski, formatR, RColorBrewer)
knitr::opts_chunk$set(echo=TRUE, message=FALSE, comment=NA, warning=FALSE, tidy=TRUE, results="hold", cache=FALSE, dpi=120)Aesthetic Example
Log-Sigmoid Curve
The link below shows a representation of a Log-Sigmoid function like that used in a logistic regression.
\[\sigma(x) = \frac{1}{1+e^{-x}}\]
View the GIF version here:
Log-Sigmoid GIF
Three Uses
Next, I show three utilitarian uses for GIFs
The Gapminder example at the end will utilize all three(!)
Showing Multiple X, Y Charts
Inverse Predictor
# Create Function for Chart
inverse_rel_bars <- function(iter, bar1, bar2) {
for (i in 1:iter) {
# Create Dataset with Random data
scl <- abs(bar1 - bar2)/iter
chg <- scl * i
Variables <- c("Exp Var", "Target")
Values <- c(bar1 - chg, bar2 + (chg/3))
data <- data.frame(Variables, Values)
# Grouped Bar Char
plot <- ggplot(data, aes(fill = Variables, y = Values, x = Variables)) +
geom_bar(stat = "identity") + ggtitle("Inverse Relationship") + ylim(0,
bar1) + scale_fill_manual(values = c("#7570B3", "maroon")) + theme_minimal()
print(plot)
}
}
# Save Chart to Hard Drive
invisible(save_gif(inverse_rel_bars(8, 90, 10), "inverse_rel_bars.gif", progress = FALSE))
# Display with R (Need animation.hook='gifski' in chunk options)
inverse_rel_bars(8, 90, 10)
Inverse Predictor and Controls
# Create Function for Chart
inverse_rel_ctrl_bars <- function(iter, bar1, bar2) {
for (i in 0:(iter - 1)) {
# Create Dataset with Random data
scl <- abs(bar1 - bar2)/iter
chg <- scl * i
Variables <- c("Control 1", "Control 2", "Control 3", "Exp Var", "Target")
Values <- c(75, 65, 75, bar1 - chg, bar2 + (chg/3))
data <- data.frame(Variables, Values)
# Grouped Bar Char
plot <- ggplot(data, aes(fill = Variables, y = Values, x = Variables)) +
geom_bar(stat = "identity") + ggtitle("Inverse Relationship") + ylim(0,
bar1) + scale_fill_manual(values = c(rep("steelblue4", 3), "#7570B3",
"maroon")) + theme_minimal()
print(plot)
}
}
# Save Chart to Hard Drive
invisible(save_gif(inverse_rel_ctrl_bars(8, 90, 10), "inverse_rel_ctrl_bars.gif",
progress = FALSE))
Showing a Z Parameter
Random Seed
# Create Function for Chart
random_grouped_bars <- function(iter, bar1, bar2) {
for (i in 1:iter) {
# Create Dataset with Random data
Variables <- c(rep("Var 1", 2), rep("Var 2", 2), rep("Var 3", 2), rep("Var 4",
2))
Group <- rep(c("Group 1", "Group 2"), 4)
Values <- c(bar1, abs(rnorm(6, 0, 5)), bar2)
data <- data.frame(Variables, Group, Values)
seed <- abs(rnorm(1, 0, 100)) %>% round(1)
set.seed(seed)
# Grouped Bar Char
plot <- ggplot(data, aes(fill = Group, y = Values, x = Variables)) + geom_bar(position = "dodge",
stat = "identity") + ggtitle("Random Data", paste("Seed", seed)) + ylim(0,
15) + theme_minimal()
print(plot)
}
}
# Save Chart to Hard Drive
invisible(save_gif(random_grouped_bars(8, 5, 15), "random_grouped_bars.gif", progress = FALSE))
Over Time
# Create Function for Chart
opposing_grouped_bars_2 <- function(iter, startyear) {
for (i in 1:iter) {
# Create Dataset with Random Data
year <- startyear + i - 1
chg <- year - (startyear - 2)
chg1_func <- function(value, chg) {
x <- ifelse((value + chg) > 10, value + chg - 10, value + chg)
x <- x * x
return(x)
}
chg2_func <- function(value, chg) {
x <- ifelse((value - chg) < 2, value - chg + 10, value - chg)
x <- x * x
return(x)
}
Variables <- c(rep("Var 1", 2), rep("Var 2", 2), rep("Var 3", 2))
Group <- rep(c("Group 1", "Group 2"), 3)
Values <- c(chg1_func(3, chg), chg2_func(10, chg), chg1_func(9, chg), chg2_func(2,
chg), chg1_func(3, chg), chg2_func(9, chg))
data <- data.frame(Variables, Group, Values)
# Grouped Bar Char
plot <- ggplot(data, aes(fill = Group, y = Values, x = Variables)) + geom_bar(position = "dodge",
stat = "identity") + ggtitle("Opposing Groups Data", paste("Year", year)) +
ylim(0, 100) + theme_minimal()
print(plot)
}
}
# Save Chart to Hard Drive
invisible(save_gif(opposing_grouped_bars_2(6, 2001), delay = 0.9, "opposing_grouped_bars_2.gif",
progress = FALSE))
Waves
# Create Function for Chart
opposing_grouped_bars <- function(iter, startyear) {
for (i in 1:iter) {
# Create Dataset with Random Data
year <- startyear + i - 1
chg <- year - (startyear - 2)
chg1_func <- function(value, chg) {
x <- ifelse((value + chg) > 8, value + chg - 8, value + chg)
x <- x * x
return(x)
}
chg2_func <- function(value, chg) {
x <- ifelse((value - chg) < 2, value - chg + 8, value - chg)
x <- x * x
return(x)
}
Variables <- c(rep("Var 1", 2), rep("Var 2", 2), rep("Var 3", 2), rep("Var 4",
2), rep("Var 5", 2), rep("Var 6", 2), rep("Var 7", 2), rep("Var 8", 2))
Group <- rep(c("Group 1", "Group 2"), 8)
Values <- c(chg1_func(3, chg), chg2_func(10, chg), chg1_func(4, chg), chg2_func(9,
chg), chg1_func(5, chg), chg2_func(8, chg), chg1_func(6, chg), chg2_func(7,
chg), chg1_func(7, chg), chg2_func(6, chg), chg1_func(8, chg), chg2_func(5,
chg), chg1_func(9, chg), chg2_func(4, chg), chg1_func(10, chg), chg2_func(3,
chg))
data <- data.frame(Variables, Group, Values)
# Grouped Bar Char
plot <- ggplot(data, aes(fill = Group, y = Values, x = Variables)) + geom_bar(position = "dodge",
stat = "identity") + ggtitle("Opposing Groups Data", paste("Year", year)) +
ylim(0, 100) + theme_minimal()
print(plot)
}
}
# Save Chart to Hard Drive
invisible(save_gif(opposing_grouped_bars(12, 2001), delay = 0.3, "opposing_grouped_bars.gif",
progress = FALSE))
Drawing a Chart
Dramatic Reveal
# Create Dataset with Random Data
Year <- c(rep(2001, 2), rep(2002, 2), rep(2003, 2), rep(2004, 2), rep(2005, 2))
Group <- rep(c("Group 1", "Group 2"), 5)
Values <- c(1, 2, 3, 4, 5, 8, 7, 16, 9, 32)
data <- data.frame(Year, Group, Values)
# Create Function for Chart
grouped_lines <- function(iter, startyear, data) {
for (i in 1:iter) {
year <- startyear + i - 1
# Censor Years
plot_df <- data[seq(1, i * 2), ]
# Grouped Bar Char
plot <- ggplot(na.omit(plot_df), aes(y = Values, x = Year, group = Group,
color = Group)) + geom_line(size = 2) + ggtitle("Line Chart", paste("Year",
year)) + ylim(0, 35) + xlim(2001, 2005) + theme_minimal()
print(plot)
}
}
# Save Chart to Hard Drive
invisible(save_gif(grouped_lines(5, 2001, data), delay = 0.6, "grouped_lines.gif",
progress = FALSE))
Drawing Complicated Lines
# Create Dataset with Random Data
Year <- c(rep(2001, 4), rep(2002, 4), rep(2003, 4), rep(2004, 4), rep(2005, 4))
Group <- rep(c("Group 1", "Group 2", "Group 3", "Group 4"), 5)
Values <- c(1, 2, 8, 32, 3, 4, 14, 16, 5, 8, 20, 8, 7, 16, 26, 4, 9, 32, 32, 2)
data <- data.frame(Year, Group, Values)
# Create Function for Chart
draw_lines <- function(iter, startyear, data) {
for (i in 2:(iter + 1)) {
year <- startyear + i
# Censor Years
plot_df1 <- data[seq(1, i * 4 - 3, by = 4), ]
# Grouped Bar Char
plot <- ggplot(na.omit(plot_df1), aes(y = Values, x = Year, group = Group,
color = Group)) + geom_line(size = 2) + ggtitle("Line Chart", paste("Group 1, Year",
year)) + ylim(0, 35) + xlim(2001, 2005) + scale_color_manual(values = brewer.pal(n = 4,
name = "Dark2")[1]) + theme_minimal()
print(plot)
}
for (i in 1:iter) {
year <- startyear + i - 1
# Censor Years
plot_df2 <- data[seq(2, i * 4 - 2, by = 4), ]
plot_df <- rbind(plot_df1, plot_df2)
# Grouped Bar Char
plot <- ggplot(na.omit(plot_df), aes(y = Values, x = Year, group = Group,
color = Group)) + geom_line(size = 2) + ggtitle("Line Chart", paste("Group 2, Year",
year)) + ylim(0, 35) + xlim(2001, 2005) + scale_color_manual(values = brewer.pal(n = 4,
name = "Dark2")[1:2]) + theme_minimal()
print(plot)
}
for (i in 1:iter) {
year <- startyear + i - 1
# Censor Years
plot_df3 <- data[seq(3, i * 4 - 1, by = 4), ]
plot_df <- rbind(plot_df1, plot_df2, plot_df3)
# Grouped Bar Char
plot <- ggplot(na.omit(plot_df), aes(y = Values, x = Year, group = Group,
color = Group)) + geom_line(size = 2) + ggtitle("Line Chart", paste("Group 3, Year",
year)) + ylim(0, 35) + xlim(2001, 2005) + scale_color_manual(values = brewer.pal(n = 4,
name = "Dark2")[1:3]) + theme_minimal()
print(plot)
}
for (i in 1:iter) {
year <- startyear + i - 1
# Censor Years
plot_df4 <- data[seq(4, i * 4, by = 4), ]
plot_df <- rbind(plot_df1, plot_df2, plot_df3, plot_df4)
# Grouped Bar Char
plot <- ggplot(na.omit(plot_df), aes(y = Values, x = Year, group = Group,
color = Group)) + geom_line(size = 2) + ggtitle("Line Chart", paste("Group 4, Year",
year)) + ylim(0, 35) + xlim(2001, 2005) + scale_color_manual(values = brewer.pal(n = 4,
name = "Dark2")[1:4]) + theme_minimal()
print(plot)
}
}
# Save Chart to Hard Drive
invisible(save_gif(draw_lines(5, 2001, data), delay = 0.6, "draw_lines.gif", progress = FALSE))
Applied Examples
Example deliverable
Using Python’s Matplotlib, I created a HTML Report as a summary version of my Capstone Project
The link can be found here:
Bad Banking Behavior
Gapminder
Hans Rosling is one of the best known practicioners of this form of data visualization. In his most famous lecture, he plots 200 countries over 200 years using 120,000 numbers. This entails plotting life expectancy against income for every country since 1810. By having the plots scroll through each consective year, he shows the progress in which the world has made in becoming “less sick” and “more wealthy”.
An example of his lecture can be found here:
Youtube Link
You can find my recreation of his famous lecture using d3.js:
200 Countries
# Load Gapminder Data
pacman::p_load(gapminder)
# Create Function for Chart
makeplot <- function() {
datalist <- split(gapminder, gapminder$year)
lapply(datalist, function(data) {
p <- ggplot(data, aes(gdpPercap, lifeExp, size = pop, color = continent)) +
scale_size("population", limits = range(gapminder$pop)) + geom_point() +
ylim(20, 90) + scale_x_log10(limits = range(gapminder$gdpPercap)) + ggtitle(data$year) +
theme_classic()
print(p)
})
}
# Save Chart to Hard Drive
invisible(save_gif(makeplot(), "gapminder.gif", delay = 0.5, progress = FALSE))Here is an ugly GIF version that you can create simply by copying and pasting code from 
Conclusion
Now that you know how to make a GIF, please make sure you are pronouncing it correctly!
Youtube Link