Scatter Plots in Julia

How to make scatter plots in Julia with Plotly.

Scatter Trace Type

The scatter trace type can be used to represent scatter charts (one point or marker per observation), line charts (a line drawn between each point), or bubble charts (points with size proportional to a dimension of the observation).

To draw a scatter chart, use the scatter trace type and set the mode parameter to markers.

using PlotlyJS
# x and y given as arrays
plot(scatter(x=1:10, y=rand(10), mode="markers"))

# x and y given as DataFrame columns
using PlotlyJS, CSV, DataFrames
df = dataset(DataFrame, "iris")
plot(scatter(df, x=:sepal_width, y=:sepal_length, mode="markers"))

Set size and color with column names

Note that you can set marker_size via column name and generate multiple traces using group.

using PlotlyJS, CSV, DataFrames
df = dataset(DataFrame, "iris")
plot(
    df, x=:sepal_width, y=:sepal_length, color=:species,
    marker=attr(size=:petal_length, sizeref=maximum(df.petal_length) / (20^2), sizemode="area"),
    mode="markers"
)

Line plots

By setting mode to lines, you can draw a line chart.

using PlotlyJS
t = 0:0.01:2π
plot(
    scatter(x=t, y=cos.(t), mode="lines"),
    Layout(yaxis_title="cos(t)", xaxis_title="t")
)

You can also plot functions directly:

using PlotlyJS
plot(cos, 0, 2π, mode="lines", Layout(title="cos(t)"))

As well as DataFrames:

using PlotlyJS, CSV, DataFrames
df = dataset(DataFrame, "gapminder")
df_ocean = df[df.continent .== "Oceania", :]
plot(
    df_ocean, x=:year, y=:lifeExp, color=:country, mode="lines"
)

Line and Scatter Plots

Use mode argument to choose between markers, lines, or a combination of both.

using PlotlyJS, Random

Random.seed!(42)

N = 100
random_x = range(0, stop=1, length=N)
random_y0 = randn(N) .+ 5
random_y1 = randn(N)
random_y2 = randn(N) .- 5

plot([
    scatter(x=random_x, y=random_y0, mode="markers", name="markers"),
    scatter(x=random_x, y=random_y1, mode="lines", name="lines"),
    scatter(x=random_x, y=random_y2, mode="markers+lines", name="markers+lines")
])

Bubble Scatter Plots

In bubble charts, a third dimension of the data is shown through the size of markers. For more examples, see the bubble chart docs

using PlotlyJS

plot(scatter(
    x=1:4, y=10:13, mode="markers", marker=attr(size=40:20:100, color=0:3)
))

Style Scatter Plots

There are many properties of the scatter trace type that control differetn aspects of the appearance of the trace. Here are a few examples

using PlotlyJS

p = plot(
    [sin, cos], 0, 10, mode="markers", marker=attr(size=10, line_width=2),
    Layout(title="Styled Scatter", yaxis_zeroline=false, xaxis_zeroline=false)
)
restyle!(p, 1, marker_color="rgba(152, 0, 0, 0.8)")
restyle!(p, 2, marker_color="rgba(255, 182, 193, 0.9)")
p

Data Labels on Hover

using PlotlyJS, HTTP, CSV, DataFrames

read_remote_csv(url) = DataFrame(CSV.File(HTTP.get(url).body))

df = read_remote_csv("https://raw.githubusercontent.com/plotly/datasets/master/2014_usa_states.csv")

plot(
    df, x=:Postal, y=:Population, mode="markers", text=:State, marker_color=:Population,
    Layout(title="Populations of USA States")
)

Scatter with a Color Dimension

using PlotlyJS

plot(scatter(
    y=randn(500), mode="markers",
    marker=attr(size=16, color=rand(500), colorscale="Viridis", showscale=true)
))

Large Data Sets

Now in Plotly you can implement WebGL with scattergl() in place of scatter() <br> for increased speed, improved interactivity, and the ability to plot even more data!

using PlotlyJS

N = 100000
plot(scattergl(
    x=randn(N), y=randn(N), mode="markers",
    marker=attr(color=randn(N), colorscale="Viridis", line_width=1)
))

Reference

See https://plotly.com/julia/reference/scatter/ or https://plotly.com/julia/reference/scattergl/ for more information and chart attribute options!