Plotting plots in same colors

Question

I working with plotly in R. I am trying to plot charts with the same colors. Below you can see data and charts.

library(plotly)
library(reshape2)
library(dplyr)

df<-data.frame(city=c("NYC","Seattle","Boston","LA","Seattle"),
               value=c(100,200,300,400,500))
df <-melt(df)

Now I am plotting pie chart with colors shown below:

fig<-df %>% 
  plot_ly(labels = ~city, values = ~value)
fig <- fig %>% add_pie(hole = 0.6)
fig

Finally, I want to plot a bar chart with the same colors as the pie plot, shown above. In order to do this, I tried this command lines :

df <-melt(df)
fig <- plot_ly(df, x = ~city, y = ~value, type = 'bar')
fig

So can anybody help me with how to plot a barplot with the same colors as pie chart ?

score 1 · Accepted Answer · answered Jun 21 '22 at 13:29

Here's a somewhat hacky but effective solution:

fig <- ggplot(df, aes(city, value, fill = city)) + 
       geom_col() +
       scale_fill_manual(values = c("#2ca02c", "#ff7f0e", 
                                    "#d62728", "#1f77b4")) +
       theme_minimal() +
       theme(panel.grid.major.x = element_blank())

ggplotly(fig)

Archeologist · Answer 2 · 2022-06-21T13:52:03.627

You may use ggplot2 package for both charts and it shall give matching colours to the city in each chart:

#install.packages('ggplot2')
library(ggplot2)
library(dplyr)

#your data frame
df<-data.frame(city=c("NYC","Seattle","Boston","LA","Seattle"),
value=c(100,200,300,400,500))
df <-melt(df)


PieChart <- df %>% ggplot(aes(x="", y=value, fill=city)) + 
  geom_bar(stat = "identity", width=1) + 
  coord_polar("y",start=0) + 
  theme_void() 
PieChart

The resulted plot:

BarChart <- df %>% ggplot(aes(x=city, y=value, fill=city)) + 
  geom_bar(stat = "identity") +
  theme_void() + 
  xlab("City") + ylab("Value")
BarChart

The resulted plot:

I prefer Plotly and solution with Plotly, because this library is compatible with my work — silent_hunter, Jun 21 '22 at 13:49

Archeologist · Answer 3 · 2022-06-21T18:05:39.680

You may find this helpful.

In addition, here's what you can do if you want to completely use plotly.

#install.packages('RColorBrewer')
library(RColorBrewer)
library(dplyr)
library(plotly)

df<-data.frame(city=c("NYC","Seattle","Boston","LA","Seattle"),
               value=c(100,200,300,400,500))
df <-melt(df)
df
#define a vector with colour palette of four colours
cols <- brewer.pal(12, "Set3")[1:4]
cols

Bar Chart:

BarChart <- df %>% plot_ly(x = ~city, y = ~value, showlegend = TRUE, 
type = 'bar', color = ~city, colors = cols)

BarChart

The resulting plot:

Pie Chart:

PieChart <- df %>% plot_ly(labels = ~city, values = ~value, 
    marker = list(colors = cols), showlegend = TRUE) %>% 
    add_pie(hole = 0.6)
PieChart

The resulting plot:

Note that this would still give you different colours for the cities, so another suggestion that I have besides the linked answer above is to try creating a new data frame that has the unique four cities i.e no repetition, the summation of the values for the repeated cities, and bind another column that has a unique colour per city.

Plotting plots in same colors

3 Answers3