I've been looking with envy and admiration at the various ggplot animations appearing on twitter since David Robinson released his gganimate package and thought I'd have a play myself. I am having an issue with gganimate when using geom_bar. Hopefully the following example demonstrates the problem.
First generate some data for a reproducible example:
df <- data.frame(x = c(1, 2, 1, 2),
y = c(1, 2, 3, 4),
z = c("A", "A", "B", "B"))
To demonstrate what I'm trying to do I thought it would be useful to plot an ordinary ggplot, facetted by z
. I'm trying to get gganimate to produce a gif that cycles between these 2 plots.
ggplot(df, aes(x = x, y = y)) +
geom_bar(stat = "Identity") +
facet_grid(~z)
But when I use gganimate the plot for B behaves oddly. In the second frame the bars start at the values that the first frame's bars finish at, rather than starting at the origin. As if it was a stacked bar chart.
p <- ggplot(df, aes(x = x, y = y, frame = z)) +
geom_bar(stat = "Identity")
gg_animate(p)
Incidentally when trying the same plot with geom_point
everything works as expected.
q <- ggplot(df, aes(x = x, y = y, frame = z)) +
geom_point()
gg_animate(q)
I tried to post some images, but apparently I don't have sufficient reputation, so I hope it makes sense without them. Is this a bug, or am I missing something?
Thanks in advance,
Thomas
The reason is that without faceting, the bars are stacked. Use position = "identity"
:
p <- ggplot(df, aes(x = x, y = y, frame = z)) +
geom_bar(stat = "Identity", position = "identity")
gg_animate(p)
In order to avoid confusion in situations like this, it is much more useful to replace frame
by fill
(or colour
, depending on the geom you are using`):
p <- ggplot(df, aes(x = x, y = y, fill = z)) +
geom_bar(stat = "Identity")
p
The two plots that are drawn, when you replace fill
by frame
correspond exactly to exclusively drawing one of the colours at a time.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With