Plotly.express.scatter

Ternary plots is a triangular coordinate system that represents the ratio of three variables as positions on an equilateral triangle.

In this article, we will explore how we can create ternary plots using the Plotly express module.

Plotly.express.scatter_ternary()

To create a ternary plot in Plotly, we use the scatter_ternary() function. The function syntax is as shown below:

plotly.express.scatter_ternary(data_frame=None, a=None, b=None, c=None, color=None, symbol=None, size=None, text=None, hover_name=None, hover_data=None, custom_data=None, animation_frame=None, animation_group=None, category_orders=None, labels=None, color_discrete_sequence=None, color_discrete_map=None, color_continuous_scale=None, range_color=None, color_continuous_midpoint=None, symbol_sequence=None, symbol_map=None, opacity=None, size_max=None, title=None, template=None, width=None, height=None)

The function parameters are expressed as shown below:

Data_frame – specifies the data frame, dictionary or array_like object to be used as the column names.
a – specifies the values used to position the marks along the a axis in the ternary coordinate. You can pass the values as a column within a data frame, a Pandas series or an array_like object.
b – similar to a but the values are used to place the marks along the b axis in the ternary coordinate.
c – similar to a and b but specifies the values used to position the marks along the c axis.
color – specifies the values used to assign a unique color to the marks.
symbols – specifies the values used to assign unique symbols to the marks.
Title – specifies the title of the figure.
width/height – specifies the width and height dimensions of the figure in pixels.

Example 1: Simple Ternary Plot

Take the example code below that shows how we can create a simple ternary plot using the scatter_ternary() function.

import plotly.express as px

import pandas as pd

candidate_1 = [10000, 20000, 1000, 50000, 40000]

candidate_2 = [50000, 5000, 3000, 22000, 10000]

candidate_3 = [40000, 5500, 55000, 33000, 23333]

df = pd.DataFrame({'candidate_1': candidate_1,

"candidate_2": candidate_2, "candidate_3": candidate_3})

fig = px.scatter_ternary(df, a="candidate_1", b="candidate_2", c="candidate_3")

fig.show()

In the example above, we start by importing the plotly express module as px. We also import pandas to create a data frame from the data.

Once we have the DataFrame, we use the scatter_ternary() function to create a ternary plot of the election candidates.

The resulting figure is as shown:

Example 2

Suppose we want to label the marks based on the area of the election. We can update our DataFrame to include the area column as shown:

candidate_1 = [10000, 20000, 1000, 50000, 40000]

candidate_2 = [50000, 5000, 3000, 22000, 10000]

candidate_3 = [40000, 5500, 55000, 33000, 23333]

areas = ['area1', 'area2', 'area3', 'area4', 'area5']

df = pd.DataFrame({'candidate_1': candidate_1,

"candidate_2": candidate_2, "candidate_3": candidate_3, "areas": areas})

fig = px.scatter_ternary(df, a="candidate_1", b="candidate_2", c="candidate_3", color='areas')

fig.show()

Once we run the code above, we should see a plot as shown:

In this example, we specify the color parameter based on the area of votes.

Example 3

We can also assign the marks a distinct symbol by setting the symbol parameter.

fig = px.scatter_ternary(df, a="candidate_1", b="candidate_2", c="candidate_3", color='areas', symbol='areas')

Running the code above should assign a unique mark based on the area.

The resulting figure is as shown:

Conclusion

This article covers the fundamentals of creating and working with ternary plots in Plotly express.

Plotly.express.scatter_ternary