[R] Need help in ggplot

[Md. Moyazzem Hossain]

Dear Rui Barradas

Thank you very much for your prompt reply.

It works. I am grateful for your help.

Take care and best wishes.

Hossain


On Tue, Jun 28, 2022 at 5:45 PM Rui Barradas <ruipbarradas using sapo.pt> wrote:

> Hello,
>
> scale_*_manual expects a discrete vector as values, you are passing it a
> continuous one. The natural way of having a continuous variable set the
> color and fill values is with a gradient scale.
>
> I have changed geom_bar to geom_col.
> Map prob to the fill aesthetic. Then, in scale_fill_continuous set low
> and high values to
>
>   - the highest possible (all zeros is black).
>   - and to the lowest in the scale of probabilities (all binary ones or
> hexa FF's is bright white)
>
> Note that this reverses the real line order relation, you want to map
> low probabilities to high binary/hexa values and vice-versa.
>
> Don't mind the theme() setting a yellow panel background fill, it's just
> meant to see better that the bars are in shades of gray.
>
>
>
> library(ggplot2)
>
> predictors <- c("x1", "x2", "x3", "x4", "x5")
> values <- c(1.0, 0.67, 0.26, 0.18, 0.17)
> prob <- c(1.0, 0.001, 0.957, 0.924, 0.253)
>
> df <- data.frame(predictors, values, prob)
>
> ggplot(df, aes(x = predictors, y = values) ) +
>    geom_col(aes(fill = prob)) +
>    coord_flip() +
>    scale_fill_gradient(low = "#FFFFFF", high = "#000000") +
>    theme(panel.background = element_rect(fill = "yellow"))
>
>
>
> If you want to use scale_fill_manual, create a vector of values first, then
>
>   - Map variable predictors to the fill aesthetic since each precitor
> value has an associated probability prob;
>   - make values equal to gray_vals.
>
>
> gray_vals <- setNames(gray(df$prob), df$predictors)
>
> ggplot(df, aes(x = predictors, y = values) ) +
>    geom_col(aes(fill = predictors)) +
>    coord_flip() +
>    scale_fill_manual(values = gray_vals) +
>    theme(panel.background = element_rect(fill = "yellow"))
>
>
> Hope this helps,
>
> Rui Barradas
>
> Às 15:07 de 28/06/2022, Md. Moyazzem Hossain escreveu:
> > Dear R- Experts,
> >
> > I hope that you are doing well.
> >
> > I am facing a problem with adding color to a barplot. My target is to
> add a
> > bar color conditioning on the following
> >
> > Values close to 1 represent "White" and values tend to 0 represent
> "Black".
> >
> > However, I failed. The code is given below. I will be happy and thankful
> if
> > anyone helps me in this regard.
> >
> > Thanks in advance.
> >
> > library(ggplot2)
> > predictors=c("x1", "x2", "x3", "x4", "x5")
> > values=c(1.0, 0.67, 0.26, 0.18, 0.17)
> > prob=c(1.0, 0.001, 0.957, 0.924, 0.253)
> >
> > df <- data.frame(predictors, values, prob)
> > ggplot(df, aes(x=predictors, y=values) ) +
> >    geom_bar(stat="identity") +
> >    coord_flip()+
> >    scale_fill_manual(values=c("gray(prob)"))
> >
> > Best Regards,
> >
> > *Hossain*
> >
> >       [[alternative HTML version deleted]]
> >
> > ______________________________________________
> > R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> > and provide commented, minimal, self-contained, reproducible code.
>


-- 
Best Regards,
Md. Moyazzem Hossain
Associate Professor
Department of Statistics
Jahangirnagar University
Savar, Dhaka-1342, Bangladesh
Website: http://www.juniv.edu/teachers/hossainmm
Research: *[image: Google Scholar]
<https://scholar.google.com/citations?hl=en&user=-U03XCgAAAAJ>* | *ResearchGate
<https://www.researchgate.net/profile/Md_Hossain107>* | *ORCID iD
<https://orcid.org/0000-0003-3593-6936>*

	[[alternative HTML version deleted]]