Tanvir Khan
Tanvir Khan

Reputation: 113

Single legend for Plotly subplot for line plots created from two data frames in R

I am trying to plot line charts from two data frames where first columns of both data frames are to be plotted in one frame and so forth. And finally all the plots are to be put under one subplot. But I am getting multiple legends with same symbols. The data frames are-

d1 <- data.frame(x = 1:5,
                 y = c(2, 3, 4, 1, 5),
                 z = c(2, 1, 4, 6, 8))

d2 <- data.frame(x = 1:5,
                 y = c(1, 5, 8, 9, 11),
                 z = c(3, 5, 8, 13, 11))

The code I am trying to generate the subplot is-

py <-
  plot_ly(
    x = d1$x,
    y = d1$y,
    type = "scatter",
    mode = "lines",
    name = names(d1)[2],
    line = list(color = "#56B4E9")
  ) %>% add_trace(y = d2$y,
                  name = names(d1)[3],
                  line = list(color = "#D55E00"))

pz <-
  plot_ly(
    x = d1$x,
    y = d1$z,
    type = "scatter",
    mode = "lines",
    name = names(d1)[2],
    line = list(color = "#56B4E9")
  ) %>% add_trace(y = d2$z,
                  name = names(d1)[3],
                  line = list(color = "#D55E00"))

subplot(py, pz)

The output is- enter image description here

Is there any way to get rid of the duplicate legends?

Thanks in advance.

Upvotes: 3

Views: 2608

Answers (1)

stefan
stefan

Reputation: 125572

This can be achieved by first bringing the data in the right shape which also simplifies the plotting. Simply row bind your dfs e.g. via dplyr::bindrows and you have the variable you need for setting up the legendgroup. Also, your colors don't reflect the variables y and z but the datasets. Try this:

library(dplyr)
library(plotly)

d1 <- data.frame(x = 1:5,
                 y = c(2, 3, 4, 1, 5),
                 z = c(2, 1, 4, 6, 8))

d2 <- data.frame(x = 1:5,
                 y = c(1, 5, 8, 9, 11),
                 z = c(3, 5, 8, 13, 11))

# Bind the dfs
d3 <- bind_rows(list(d1 = d1, d2 = d2), .id = "id")

py <- d3 %>% 
  plot_ly(x = ~x, y = ~y, color = ~id, legendgroup= ~id) %>% 
  add_lines(colors = c("#D55E00", "#56B4E9"))

pz <- d3 %>% 
  plot_ly(x = ~x, y = ~z, color = ~id, legendgroup= ~id) %>% 
  add_lines(colors = c("#D55E00", "#56B4E9"), showlegend = FALSE)

subplot(py, pz) %>% 
  layout(legend=list(title=list(text='<b> Dataset </b>')))

Created on 2020-04-10 by the reprex package (v0.3.0)

enter image description here

Upvotes: 5

Related Questions