support Animated Grouped Bubble Chart?

[XiangyunHuang]

I want to create an Animated Grouped Bubble Chart, but I don't know how to do that. This is my attempt

1. simulate some fake data

# some fake data
library(echarts4r)
set.seed(2022)
dat <- data.frame(
  x1 = rep(2020:2023, each = 4),
  x2 = rep(c("A", "A", "B", "B"), 4),
  x3 = runif(16),
  x4 = runif(16),
  x5 = abs(runif(16))
)

dat

     x1 x2       x3     x4      x5
1  2020  A 0.815978 0.7732 0.25652
2  2020  A 0.647259 0.5145 0.40536
3  2020  B 0.120329 0.5954 0.17088
4  2020  B 0.543800 0.8509 0.06798
5  2021  A 0.184730 0.8428 0.84603
6  2021  A 0.635791 0.5703 0.54988
7  2021  B 0.074299 0.4266 0.80484
8  2021  B 0.041976 0.1372 0.76506
9  2022  A 0.370317 0.1631 0.64224
10 2022  A 0.757253 0.9351 0.44170
11 2022  B 0.001862 0.5370 0.64938
12 2022  B 0.159799 0.9977 0.52454
13 2023  A 0.144742 0.4790 0.86723
14 2023  A 0.519397 0.7736 0.10065
15 2023  B 0.609476 0.4680 0.88715
16 2023  B 0.122511 0.6376 0.62795

2. try to create Animated Grouped Bubble Chart

I'm able to create Grouped Bubble Chart and Animated Bubble Chart.

# **Grouped Bubble Chart**
dat |>
  group_by(x1) |>
  e_charts(x = x3) |>
  e_scatter(serie = x5, size = x4, symbol_size = 5) |>
  e_tooltip()

# **Animated Bubble Chart**
dat |>
  group_by(x1) |>
  e_charts(x = x3, timeline = TRUE) |>
  e_scatter(serie = x5, size = x4, symbol_size = 5) |>
  e_tooltip()

But, I failed to create Animated Grouped Bubble Chart. when I add x2 into group_by(), Is there any other way to do that?

dat |>
  group_by(x1, x2) |>
  e_charts(x = x3, timeline = TRUE) |>
  e_scatter(serie = x5, size = x4, symbol_size = 5) |>
  e_tooltip()

Session info

xfun::session_info(packages = "echarts4r", dependencies = F)
#> R version 4.2.0 (2022-04-22)
#> Platform: x86_64-apple-darwin17.0 (64-bit)
#> Running under: macOS Big Sur/Monterey 10.16
#> 
#> Locale: en_US.UTF-8 / en_US.UTF-8 / en_US.UTF-8 / C / en_US.UTF-8 / en_US.UTF-8
#> 
#> Package version:
#>   echarts4r_0.4.4

^{Created on 2022-06-08 by the reprex package (v2.0.1)}

[rdatasculptor]

@XiangyunHuang (@JohnCoene) I think you can do this in echarts4r as well. No new features are needed I think.

Given your examnple I was not entirely sure which output you intended, to be honest. So I made two possible animated, grouped bubbel charts. Pick the one you want :)

library(echarts4r)
library(tidyverse)
set.seed(2022)
dat <- data.frame(
  x1 = rep(2020:2023, each = 4),
  x2 = rep(c("A", "A", "B", "B"), 4),
  x3 = runif(16),
  x4 = runif(16),
  x5 = abs(runif(16))
)

## Bubble chart 1
datlist <- list()
years <- 2020:2023
for (i in 1:length(years)){
  df <- dat |> filter(x1 == years[i]) |> select(-x1)
  names(df)[length(df)] <- years[i]
  datlist[[i]] <- df 
}
dat2 <- bind_rows(datlist)
dat2 |>
  group_by(x2) |>
  e_charts(x = x3, timeline = TRUE) |>
  e_scatter(`2020`, size = x4, symbol_size = 5) |>
  e_scatter(`2021`, size = x4, symbol_size = 5) |>
  e_scatter(`2022`, size = x4, symbol_size = 5) |>
  e_scatter(`2023`, size = x4, symbol_size = 5) |>
  e_tooltip()

# Bubble chart 2
datlist <- list()
ab <- c("A", "B")
for (i in 1:length(ab)){
  df <- dat |> filter(x2 == ab[i]) |> select(-x2)
  names(df)[length(df)] <- ab[i]
  datlist[[i]] <- df 
}
dat2 <- bind_rows(datlist)
dat2 |>
  group_by(x1) |>
  e_charts(x = x3, timeline = TRUE) |>
  e_scatter(A, size = x4, symbol_size = 5) |>
  e_scatter(B, size = x4, symbol_size = 5) |>
  e_tooltip()

Did this help you? Good luck!

If this was the disired outcome, then you can close the issue.

[XiangyunHuang]

@rdatasculptor Below is what I want. It is created by helgasoft using echarty.

[rdatasculptor]

Is'nt that my second bubble chart?

Also you can add a changing title, like this:

dat2 |>
  group_by(x1) |>
  e_charts(x = x3, timeline = TRUE) |>
  e_scatter(A, size = x4, symbol_size = 5) |>
  e_scatter(B, size = x4, symbol_size = 5) |>
  e_tooltip() |>
  e_timeline_serie(
    title = list(
      list(text = "2020"),
      list(text = "2021"),
      list(text = "2022"),
      list(text = "2023")
    )
  ) 

@XiangyunHuang if it is still not what wanted, can you point me at the differences?

[XiangyunHuang]

@rdatasculptor Thanks, your solution is good. Is there any other simpler method to reshape dat to dat2?

[rdatasculptor]

@XiangyunHuang Another way is using e.g. pivot_wider.
I sometimes use loops only for me to understand what's happening. But I guess that is not helpful for everyone :)

by using pivot_wider:

dat2 <- dat |> pivot_wider(names_from = "x2", values_from = "x5")
dat2 |>
  group_by(x1) |>
  e_charts(x = x3, timeline = TRUE) |>
  e_scatter(A, size = x4, symbol_size = 5) |>
  e_scatter(B, size = x4, symbol_size = 5) |>
  e_tooltip() |>
  e_timeline_serie(
    title = list(
      list(text = "2020"),
      list(text = "2021"),
      list(text = "2022"),
      list(text = "2023")
    )
  ) 

Also you can play with symbol size, because that's where the main difference is with @helgasoft approach is I think.

[rdatasculptor]

@XiangyunHuang playing with symbol size using scaling I mean:

dat2 <- dat |> pivot_wider(names_from = "x2", values_from = "x5")

my_scale <- function(x){
  scales::rescale(x, to = c(20, 50))
}

dat2 |>
  group_by(x1) |>
  e_charts(x = x3, timeline = TRUE) |>
  e_scatter(A, size = x4, scale = my_scale) |>
  e_scatter(B, size = x4, scale = my_scale) |>
  e_tooltip() |>
  e_timeline_serie(
    title = list(
      list(text = "2020"),
      list(text = "2021"),
      list(text = "2022"),
      list(text = "2023")
    )
  ) 

[XiangyunHuang]

Thanks! Just for myself, we can use Base R's reshape() function instead of tidyr's pivot_wider() function to convert dat to dat2.

library(echarts4r)

set.seed(2022)
dat <- data.frame(
  x1 = rep(2020:2023, each = 4),
  x2 = rep(c("A", "A", "B", "B"), 4),
  x3 = runif(16),
  x4 = runif(16),
  x5 = abs(runif(16))
)

dat2 <- reshape(
  data = dat,
  varying = c("A", "B"),
  timevar = c("x2"),
  idvar = c("x1", "x3", "x4"),
  direction = "wide"
)

dat2 |>
  group_by(x1) |>
  e_charts(x = x3, timeline = TRUE) |>
  e_scatter(A, size = x4, symbol_size = 5) |>
  e_scatter(B, size = x4, symbol_size = 5) |>
  e_tooltip() |>
  e_timeline_serie(
    title = list(
      list(text = "2020"),
      list(text = "2021"),
      list(text = "2022"),
      list(text = "2023")
    )
  )