erocoar / ggpol Goto Github PK

Hi Developers.

Is there any chance to support box with dots as full box plot, instead of just half box. An example of style as below, a plot from poltly.

boxplot with dots side by side

Thank you very much for this great work.

Best regards
Wallace

Hello,

I want to keep only the half boxplot and replace the jitter points with the points from the ggbeeswarm package. Is it possible? Could you add an option to turn on/off the plotting of all the points?

In the example below, I can disable the color of all the jitter points but it leaves an empty space between the boxplot and the beeswarm.

Thanks!

library(ggplot2)
library(ggbeeswarm)
library(ggpol)

ggplot(iris, aes(y = Petal.Width)) + 
  geom_boxjitter(jitter.shape = NA, 
                 jitter.color = NA,
                 jitter.height = 0, 
                 jitter.width = 0,
                 outlier.color = NA, 
                 errorbar.draw = TRUE) +
  geom_quasirandom(
    aes(x = 1, color = Species),
    width = 0.25,
    size = 2, 
    groupOnX = TRUE, alpha = 0.7) +
  scale_color_brewer(palette = 'Set2') +
  theme_classic()

^{Created on 2019-05-18 by the reprex package (v0.3.0)}

The following ggplot2 code works as expected even without attaching the package first:

set.seed(221)
df <- data.frame(score = rgamma(150, 4, 1),
                 gender = sample(c("M", "F"), 150, replace = TRUE),
                 genotype = factor(sample(1:3, 150, replace = TRUE)))

ggplot2::ggplot(df, ggplot2::aes(x = gender, y = score)) +
  ggplot2::geom_boxplot() ## works

However, the same does not work with geom_boxjitter:

ggplot2::ggplot(df, ggplot2::aes(x = gender, y = score)) +
  ggpol::geom_boxjitter() ##fails
## Error in alpha(data$fill, data$alpha) : could not find function "alpha"

We need to attach the package first, before it works:

library("ggpol")
ggplot2::ggplot(df, ggplot2::aes(x = gender, y = score)) +
  geom_boxjitter() ##works

Then it also works with :::

ggplot2::ggplot(df, ggplot2::aes(x = gender, y = score)) +
  ggpol::geom_boxjitter()  ## works now

Hi,

I encountered the following error (R 3.6.1):

* installing *source* package 'ggpol' ...
** using staged installation
** R
Error in .install_package_code_files(".", instdir) : 
files in '/AppData/Local/Temp/Rtmpsdn2EM/R.INSTALL66303af073c1/ggpol/R' missing from 'Collate' field:
  geom_half_boxplot.R
  geom_half_point.R
  stat-half-point.R
ERROR: unable to collate and parse R files for package 'ggpol'

Can you please check? Thanks!

Hey. I have installed the library, following the steps written on the repository, but I still get the error that the function geom_parliament() wasn't found. I have run the same code which is on ReadMe. It also says that the package 'ggpol' wasn't found, although I have installed it. Thank you for the attention.

Dear Frederik, thank you very much for your nice package. I am opening this issue because the example of facet_share from https://erocoar.github.io/ggpol/ is not working anymore. Specifically, I am getting this error:

df <- data.frame(sex = sample(c("M", "F"), 1000, replace = TRUE),
                 age = rnorm(1000, 45, 12))

df$age_bins <- cut(df$age, 15)
df$count <- 1
df <- aggregate(count ~ sex + age_bins, data = df, length)

df_h <- df
df_h$count <- ifelse(df_h$sex == "F", df_h$count * -1, df_h$count)

ggplot(df_h, aes(x = age_bins, y = count, fill = sex)) + 
  geom_bar(stat = "identity") +
  facet_share(~sex, dir = "h", scales = "free", reverse_num = TRUE) + 
  coord_flip()

Error in unit(c(as.numeric(axes$y$left[[1]]$children$axis$widths[[tick_idx]]), :
'list' object cannot be coerced to type 'double'

Do you know how to fix it? I was trying to update with new data a plot that I created with facet_share this time last year, but I cannot due to this facet_share problem.

Thank you in advance for any comment you might provide.

Kind regards,

Guillermo

Hello,

I decided to try out your pkg and made a rainbow colored ggpol.

when I save this image or use RStudio clipboard it shows white bg edges so i screen captured this

my initial try was to put ggdark::dark_mode() at the end for a dark plot but that did not work. here is the code i used to make my plot dark, maybe you will find it useful

# install.packages("ggpol")
library(ggpol)

rb = data.frame(
  flag = factor( c('red','orange','yellow',
                   'green','blue','purple')),
  seats = c(200, 46, 92, 80, 153, 69),
  colors = c('red','orange','yellow',
             'green','blue','purple')
)

ggplot(rb) + 
  geom_parliament(aes(seats = seats,
                      fill = flag), 
                  color = "black") + 
  scale_fill_manual(values = rb$colors, 
                    labels = rb$flag) +
  labs(title = "\nParliamentary diagram {ggpol} #PrideFlag")+
  coord_fixed() + 
  theme(
    text = element_text(family = "Rubik"),
    plot.title = element_text(size = 14, 
                              hjust = 0.5
                              ),
    panel.grid.major = element_line(linetype = "blank"),
    panel.grid.minor = element_line(linetype = "blank"),
    legend.text = element_text(colour = "white"),
    legend.title = element_markdown(size = 12,
                                    colour = "white"), 
    panel.background = element_rect(fill = "gray0"),
    plot.background = element_rect(fill = "gray0"),
    legend.background = element_rect(fill = "gray0"),
    legend.position = "bottom", legend.direction = "horizontal",
    axis.text.x = element_blank(),
    axis.text.y = element_blank(),
    axis.ticks = element_blank()
    )

Thanks for the package. : )

Would something like this possible? solid fills instead of points?

I would like to have a common limit for both facets when using facet_share(). In the example below the largest value on the left is 10 whereas it is 5 on the right. Because the limits are not the same, it looks as though the values are equal.

library(ggplot2)
sex <- rep(c("male", "female"), each = 5)
cat <- rep(LETTERS[1:5], 2)
val <- c(1:5, -(6:10))
df <- tibble::tibble(sex, cat, val)

p <- ggplot(df, aes(cat, val, fill = sex)) +
  geom_col() +
  ggpol::facet_share(vars(sex), scales = "free", reverse_num = TRUE) + 
  coord_flip()
p

A naive way to set the limits failed as expected.

p + ylim(0, 10)

It would be great to have an option for both facets to have the same limits.

facet_share.R has a bug which produces the error when dir="v" option is used.

Error in unit(c(axes$x$bottom[[1]]$children$axis$heights[[tick_idx]], :
'list' object cannot be coerced to type 'double'

The heights parameter needs to be defined as:

heights = unit(c(as.numeric(axes$x$bottom[[1]]$children$axis$heights[[tick_idx]]), 1,
as.numeric(axes$x$bottom[[1]]$children$axis$heights[[tick_idx]])),
c("pt", "grobwidth", "pt"),
list(NULL, axes$x$bottom[[1]]$children$axis$grobs[[lab_idx]], NULL))

I tried to install it a couple of times and both times it failed installation. I am using R version 3.4 on MacOS

I get the following error:

** R
Error in .install_package_code_files(".", instdir) :
files in '/private/var/folders/w0/zf6n6gbd2kd0jlvsm9hlb6kw0000gn/T/RtmpEZgP7p/devtoolsfc6e722fff86/erocoar-ggpol-5840b1a/R' missing from 'Collate' field:
geom_bartext.R
ERROR: unable to collate and parse R files for package ‘ggpol’

• removing ‘/Library/Frameworks/R.framework/Versions/3.4/Resources/library/ggpol’
  Installation failed: Command failed (1)

I'd like to remove outliers from a box jitter plot. I tried outlier.color =NA, outlier.alpla=0, and outlier.fill=NA but none of these alone or in comb did not work. Is there a way to remove outliners from the box jitter plot without manipulating the original df in question ?

Thanks.