echarts4r Cookbook

Notes on Apache ECharts for R
Author

Nathan States

Published

January 10, 2023

0. Intro Thoughts

There’s a ton of different charting libraries in R, with a few of them being interactive. plotly is one of the most popular, having over 2.3k stars on GitHub. highcharter is another such package. With ggiraph, it’s possible to turn ggplot2 objects into interactive charts, so there’s quite a selection of tools to choose from.

Personally, I find echarts4r to be the best out of all of them specifically for exploratory data analysis. The charts are easy to create, they look nice out of the box, and it’s quick and easy to gain insight into your data. The interactive element makes identifying outliers or other specific values very easy. Compared to other packages, plotly is way too ugly, highcharter requires a license for paid applications, and ggiraph is too time consuming for regular charts. echarts4r offers a nice solution to all these problems (though if you need a deeper statistical breakdown, I’d recommend ggstatsplot).

There is a noticeable limit on customization, though. I’d still recommend ggplot2 for high quality static charts, and D3 for more advanced interactive ones.

echarts4r is a wrapper for the JS library developed by Apache Software Foundation, meaning changes in the original code will cause changes in this package, regardless of if echarts4r itself is updated. Apache ECharts is listed as one of Apache’s top projects, so hopefully, we see even more improvements in the future.

1. Installation

1.A. Stable Install

Github Page.

install.packages("echarts4r")

1.B. Developmental Build

Requites the remotes package to install.

install.packages("remotes")
remotes::install_github("JohnCoene/echarts4r")

1.C. Complimentary Packages

Neither of these packages have been updated in years, but are still usable in their current formats. They allow for maps and image assets to be used while charting.

remotes::install_github('JohnCoene/echarts4r.assets')
remotes::install_github('JohnCoene/echarts4r.maps')

1.D. Loading Packages

Also loading dplyr for data manipulation and palmerpenguins for dummy data.

library(dplyr)

Attaching package: 'dplyr'
The following objects are masked from 'package:stats':

    filter, lag
The following objects are masked from 'package:base':

    intersect, setdiff, setequal, union
library(palmerpenguins)
library(echarts4r)
library(echarts4r.maps)

e_common(
  font_family = "Georgia",
  theme = "dark-mushroom"
)

2. Basic Charts

2.A. Column

penguins |>
  group_by(island) |>
  summarise(average_mass = mean(body_mass_g, na.rm = TRUE)) |>
  e_chart(island) |>
  e_bar(average_mass) |>
  e_tooltip(trigger = " axis")

2.B. Bar Chart

penguins |>
  group_by(island) |>
  summarise(average_mass = mean(body_mass_g, na.rm = TRUE)) |>
  e_chart(island) |>
  e_bar(average_mass) |>
  e_flip_coords() |> # THIS LINE OF CODE # 
  e_tooltip()

2.C. Scatter Chart

penguins |>
  group_by(species) |>
  e_chart(bill_length_mm) |>
  e_scatter(bill_depth_mm) |>
  e_tooltip()

2. . Heatmap

2. . Candlestick

2. . Treemap

2. . Calendars

dates <- seq.Date(as.Date("2017-01-01"), as.Date("2018-12-31"), by = "day")
values <- rnorm(length(dates), 20, 6)

year <- data.frame(date = dates, values = values)

year |> 
  e_charts(date) |> 
  e_calendar(range = "2018") |> 
  e_heatmap(values, coord_system = "calendar") |> 
  e_visual_map(max = 30) |> 
  e_title("Calendar", "Heatmap") |>
  e_tooltip()

When charting by multiple years, call e_calendar again and group by year.

year |> 
  dplyr::mutate(year = format(date, "%Y")) |>
  group_by(year) |> 
  e_charts(date) |> 
  e_calendar(range = "2017",top="40") |> 
  e_calendar(range = "2018",top="260") |> 
  e_heatmap(values, coord_system = "calendar") |> 
  e_visual_map(max = 30) |> 
  e_title("Calendar", "Heatmap")|>
  e_tooltip("item")

2.J. Pie / Donut Chart

No.

3. Theme Customizations

3.A. Set Common Theme

Set theme for all charts on the page and set font.

e_common(
  theme = "my-theme",
  font_family = "my-font"
)

3.B. Tooltip Theme

I always customize the tooltip to get rid of the ugly border and the white background.

penguins |>
  group_by(island) |>
  summarise(average_mass = mean(body_mass_g, na.rm = TRUE)) |>
  e_chart(island) |>
  e_bar(average_mass) |>
  e_legend(show = FALSE) |>
  e_axis_labels(
    x = "Island",
    y = "Averge Mass"
  ) |>
  e_title(
    text = "Some meaningful title."
  ) |>
  e_tooltip(
    trigger = " axis",
    backgroundColor = "rgba(40, 40, 40, 0.75)",
    borderColor = "rgba(0, 0, 0, 0)",
    textStyle = list(
      color = "#fcfcff"
    )
  )

3.C. Remove Legend

penguins |>
  group_by(island) |>
  summarise(average_mass = mean(body_mass_g, na.rm = TRUE)) |>
  e_chart(island) |>
  e_bar(average_mass) |>
  e_legend(show = FALSE) |>
  e_axis_labels(
    x = "Island",
    y = "Averge Mass"
  ) |>
  e_legend(show = FALSE) |> # THIS LINE OF CODE # 
  e_title(
    text = "Some meaningful title."
  ) |>
  e_tooltip(trigger = " axis")

4. Tooltips

3.C. Default Tooltip Formatter

One of the arguments in e_tooltip is formatter. There are three different types of default formats that make it easy to display tooltips nicer:

my_echart |>
  e_tooltip(
    formatter = e_tooltip_pointer_formatter(style = "percent"),
    # OR #
    formatter = e_tooltip_pointer_formatter(style = "percent"),
    # OR #
    formatter = e_tooltip_pointer_formatter(style = "percent"),
  )

5. Maps