In thebioengineer/camcorder: Record Your Plot History

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  warning = FALSE,
  message = FALSE
)

camcorder

{camcorder} is an an R package to track and automatically save graphics generated with {ggplot2} that are created across one or multiple sessions with the eventual goal of creating a GIF showing all the plots saved sequentially during the design process.

After installation, the package enables you to:

save a ggplot automatically every time you run ggplot() in any format with given specifications
generate a GIF that showcases every step of the design process using those image files
inspect the ggplot output directly with your specifications in the RStudio IDE—you'll get what you see[^f1]

[^f1]: In case you are saving to PDF, the file will automatically open in your default PDF viewer.

Installation

Currently {camcorder} is only available on GitHub, and can be installed using the following command.

# install.packages("camcorder")
remotes::install_github("thebioengineer/camcorder")

Goal End Product

The idea of tracking your plots as part of your development process and generating a making-of movie was popularized by two contributors to this project: Georgios Karamanis and Cédric Scherer. They have wowed the R community weekly with their "how its made" videos. Below are a few examples of the goal end products.

TidyTuesday 2020/28
by Cédric Scherer

TidyTuesday 2020/15
by Georgios Karamanis

How To

To get started, load {camcorder} and initialize recording using the gg_record() function. This function has several options, such as where to save the recordings, device to use to save the recordings, and the height/width of the image to create. By default it will save to a temporary directory so recordings will go away once the R session is closed.

library(ggplot2)
library(camcorder)

gg_record(
  dir = file.path(tempdir(), "recording100"), # where to save the recording
  device = "png", # device to use to save images
  width = 4,      # width of saved image
  height = 6,     # height of saved image
  units = "in",   # units for width and height
  dpi = 300       # dpi to use when saving image
)

Once the recorder is initialized, any ggplot that is made and printed will be automatically (or automagically[^f2]) recorded.

[^f2]: A previous typo but actually it fits quite well.

ggplot(mtcars, aes(x = mpg, y = hp)) +
  geom_point()

ggplot(mtcars, aes(x = mpg, y = hp)) + 
  geom_point(aes(shape = as.factor(gear)))

ggplot(mtcars, aes(x = mpg, y = hp)) + 
  geom_point(aes(color = as.factor(gear)))

ggplot(mtcars, aes(x = mpg, y = hp)) + 
  geom_point(aes(color = as.factor(gear))) +
  geom_path()

ggplot(mtcars, aes(x = mpg, y = hp)) +
  geom_point(aes(color = disp)) +
  geom_smooth()

ggplot(mtcars, aes(x = mpg, y = hp)) +
  geom_smooth() +
  geom_point(aes(color = disp))

ggplot(mtcars, aes(x = mpg, y = hp)) + 
  geom_smooth() +
  geom_point(aes(color = disp)) +
  scale_color_viridis_c() +
  theme_light()

ggplot(mtcars, aes(x = mpg, y = hp)) + 
  geom_smooth() +
  geom_point(aes(color = disp)) +
  scale_color_viridis_c() +
  theme_light() +
  labs(
    title = "MPG vs Horse Power!",
    subtitle = "Power and economy, the classic compromise!"
  )

ggplot(mtcars, aes(x = mpg, y = hp)) + 
  geom_smooth() +
  geom_point(aes(color = disp)) +
  scale_color_viridis_c() +
  theme_light(base_family = "Roboto Mono") +
  labs(
    title = "MPG vs Horse Power!",
    subtitle = "Power and economy, the classic compromise!"
  )

ggplot(mtcars, aes(x = mpg, y = hp)) + 
  geom_smooth() +
  geom_point(aes(color = disp)) +
  scale_color_viridis_c() +
  theme_light(base_family = "Roboto Mono") +
  labs(
    title = "MPG vs Horse Power!",
    subtitle = "Power and economy, the classic compromise!", 
    x = "Efficiency (Miles/Gallon)",
    y = "Power (Horsepower)",
    color = "Displacement\n(Cubic Inch)"
  )

If at any point, that you want to save your plots in a different format than what the recorder was initialized with this can be done through the gg_resize_film() function. This will set the size and dpi of all plots going forward.

gg_resize_film(
  height = 4,
  width = 6,
  units = "in",
  dpi = 350
)

ggplot(mtcars, aes(x = mpg, y = hp)) + 
  geom_smooth() +
  geom_point(aes(color = disp)) +
  scale_color_viridis_c() +
  theme_light(base_family = "Roboto Mono") +
  labs(
    title = "MPG vs Horse Power!",
    subtitle = "Power and economy, the classic compromise!", 
    x = "Efficiency (Miles/Gallon)",
    y = "Power (Horsepower)",
    color = "Displacement\n(Cubic Inch)"
  ) +
  theme(
    plot.title.position = "plot",
    plot.title = element_text(face = "bold")
  )

ggplot(mtcars, aes(x = mpg, y = hp)) + 
  geom_smooth() +
  geom_point(aes(color = disp)) +
  scale_color_viridis_c() +
  theme_light(base_family = "Roboto Mono")  +
  labs(
    title = "MPG vs Horse Power!",
    subtitle = "Power and economy, the classic compromise!", 
    x = "Efficiency (Miles/Gallon)",
    y = "Power (Horsepower)",
    color = "Displacement\n(Cubic Inch)"
  ) +
  theme(
    plot.title.position = "plot",
    plot.title = element_text(face = "bold"),
    panel.background = element_rect(colour = "turquoise", fill = "turquoise")
  )

Finally, to generate the final GIF, use the gg_playback() function. The user can define: - where the final GIF gets saved by setting the name argument, - duration of the first and last images with first_image_duration or last_image_duration - delay between frames in seconds with frame_duration

gg_playback(
  name = file.path(tempdir(), "recording", "vignette_gif.gif"),
  first_image_duration = 5,
  last_image_duration = 15,
  frame_duration = .4,
  image_resize = 800
)