Plotly: geom_bar(stat='identity') has problems with negative values

Created on 24 Apr 2016  路  10Comments  路  Source: ropensci/plotly

http://community.plot.ly/t/inversion-negative-values-in-ggplotly/875

d <- data.frame(
    x = LETTERS[1:10], 
    y = -5:4, 
    g = rep(c('A','B'), each = 5)
)

p <- ggplot(data = d, aes(x = x, y = y, fill = g)) +
    geom_bar(stat='identity')
ggplotly(p)
bug ggplotly

Most helpful comment

Looks like I'm not the only one who has issue with this . . . Any update on the fix? Thanks!

All 10 comments

Any clues here? Happy to work on this. I tracked to here, and everything looks fine.

This is the source of the problem (that subtraction is essentially assuming positive ymin/ymax values)

We'll probably need to switch from barmode='stacked' to 'relative' to fix this issue -- https://github.com/plotly/plotly.js/pull/517

Just in case somebody needs this before we fix, we can accomplish with plot_ly.

plot_ly(
  data = d,
  x = x,
  y = y,
  color = g,
  type = "bar"
) %>%
  layout(
    xaxis = list(type="category", categoryorder="category ascending")
  )

image

+1, already an idea when this one will be fixed?

Looks like I'm not the only one who has issue with this . . . Any update on the fix? Thanks!

Unfortunately @timelyportfolio 's workaround does not work for orientation = 'h'

d <- data.frame(
x = LETTERS[1:10],
y = -5:4,
g = rep(c('A','B'), each = 5)
)

plot_ly(
data = d,
x = ~y,
y = ~x,
color = ~g,
type = "bar",
orientation ='h'
) %>%
layout(
xaxis = list(type="category", categoryorder="category ascending")
)

Any updates regarding a fix for the negative bars turning positive? thank you very much

hello! i am having the same issue. curious about whether a fix may be in the works. I use this with my class!

Hello, I'm have the same problem... UP this issue

Was this page helpful?
0 / 5 - 0 ratings