D3partitionR
D3partitionR copied to clipboard
AntoineGuillot2
Limiting colors to a subset of steps/levels
Is there a way to limit the number of colors to only a specified subset of the steps/levels in the data? The lowest level in my dataset has many unique values, and assigning colors to them all makes the chart look very messy.
When I assign a categorical variable to the color parameter to add_data, everything turns into one color.
Is there a way to limit the number of colors to only a specified subset of the steps/levels in the data?
Currently, there is not. I hope to be able to add it for the next version.
When I assign a categorical variable to the color parameter to add_data, everything turns into one color.
Did you specify this extra variable in add_nodes_data ? If you want to use a specific color you can also use set_discrete_color_scale and use a named list with the name of your leaves.
Did you specify this extra variable in add_nodes_data ? If you want to use a specific color you can also use set_discrete_color_scale and use a named list with the name of your leaves.
The variable is one of the steps in add_data. For example, using the Titanic dataset, I would like to assign colors based on the variable "Sex"- blue for males and red for females
D3partitionR() %>%
add_data(data_plot, count = 'N', steps=c('Sex','Embarked','Pclass','Survived'), tooltip=c('name','N'), label='name', color='Sex') %>%
set_chart_type('treemap') %>%
set_discrete_color_scale(list("Sex male"="blue","Sex female"="green")) %>%
plot()
Sorry if this is a dumb question.
Actually, you can only use a numeric variable as color when you use a variable from the dataframe, since we are doing some aggregations, we cannot deal with categorical variable (not sure if it is clear). To have categorical coloring you have two options:
- Adding a color variable with add_nodes_data, but this is currently broken ...
- Adding colors with set_discrete_color_scale, for instance:
set_discrete_color_scale(c("#ff0000","#ff0000")%>%
names<-(c('Sex male','Sex female'))) You need to download it from github as the option is broken in the current CRAN version.
option 2 is giving me an error (I downloaded the github version of D3partitionR)
This should work (a quote disappeared in GitHub):
## Loading packages
library("data.table")
library("D3partitionR")
library(titanic)
require(magrittr)
require(RColorBrewer)
## Reading data
titanic_data = data.table(titanic::titanic_train)
##Agregating data to have unique sequence for the 4 variables
var_names=c('Sex','Embarked','Pclass','Survived')
data_plot=titanic_data[,.N,by=var_names]
data_plot[,(var_names):=lapply(var_names,function(x){data_plot[[x]]=paste0(x,' ',data_plot[[x]])
})]
color_list=c("#ffff00","#0000ff","#ff0000")
names(color_list)<-c('Sex male','Sex female','Survived 0')
D3partitionR() %>%
D3partitionR::add_data(data_plot,count = 'N',tooltip=c('name'),steps=c('Sex','Embarked','Pclass','Survived')) %>%
set_chart_type('treemap') %>%
set_legend_parameters(zoom_subset=T)%>%
set_discrete_color_scale(color_list)%>%
plot()
Really appreciate your help, but I think I must have confused you by splitting up my initial question. I was wondering if there's a way to assign colors only based on Sex, i.e. blue for male (and all the children nodes) and green for female (and all the children nodes). I'm guessing that might not be possible based on your first response.
It's possible but tedious, you have to manually add the color for each of the child nodes using the function set_discrete_color_scale.
Actually, you can only use a numeric variable as color when you use a variable from the dataframe, since we are doing some aggregations, we cannot deal with categorical variable (not sure if it is clear). To have categorical coloring you have two options:
- Adding a color variable with add_nodes_data, but this is currently broken ...
- Adding colors with set_discrete_color_scale, for instance: set_discrete_color_scale(c("#ff0000","#ff0000")%>%
names<-(c('Sex male','Sex female'))) You need to download it from github as the option is broken in the current CRAN version.
Has the CRAN version been amended to account for this? I have been trying to set discrete colors with no success.
Hi,
I have the same issues with the Github version. set_continous_color_scale doesn't work and set_discrete_color_scale doesn't change anything.
My problem is that I have too many categories set in black in my circle_treemap and I don't know how to change that.
