Selections in Python

How to use selections in Python. Examples of adding and styling selections.

Plotly Studio: Transform any dataset into an interactive data application in minutes with AI. Try Plotly Studio now.

Adding Selections to Cartesian Subplots¶

New in 5.10

You can add persistent selections to a rendered figure using the Box Select and Lasso Select tools in the mode bar. To add multiple selections, select Shift when making new selections. To clear a selection, double-click it. On a subplot you can clear all selections by double-clicking any unselected area of the subplot.

You can also add selections to a figure that displays when it renders using fig.add_selection. Here, we add a rectangular selection with a region between 3.0 and 6.5 on the x axis and between 3.5 and 5.5 on the y axis.

In [1]:

import plotly.express as px

df = px.data.iris()

fig = px.scatter(df, x="sepal_width", y="sepal_length")
fig.add_selection(x0=3.0, y0=6.5, x1=3.5, y1=5.5)

fig.show()

Selections Using a Custom SVG¶

In the above example, we added a rectangular selection. You can also render a custom SVG for a selection by defining a path that can include single or multiple polygons. Here, we create a selection with a single polygon path "M2,6.5L4,7.5L4,6Z".

Please note that multiple polygons e.g. "M0,0L0,10L10,10,L10,0Z M2,2L2,8L8,8,L8,2Z" could be used to subtract certain regions from the selection.

In [2]:

import plotly.express as px

df = px.data.iris()

fig = px.scatter(df, x="sepal_width", y="sepal_length")
fig.add_selection(path="M2,6.5L4,7.5L4,6Z")

fig.show()

Styling Selections¶

In the above example, we added a selection to the figure that is displayed when the figure renders. fig.add_selection accepts additional properties that you can use to style the selection. Here, we add a color, width, and specify the dash type for the selection.

In [3]:

import plotly.express as px

df = px.data.iris()

fig = px.scatter(df, x="sepal_width", y="sepal_length")
fig.add_selection(
    x0=2.5, y0=6.5, x1=3.5, y1=5.5,
    line=dict(
        color="Crimson",
        width=2,
        dash="dash",
    ))

fig.show()

Styling New Selections¶

You can style new selections made on the figure by setting properties on newselection. Try making a new selection on the figure to try it out.

In [4]:

import plotly.express as px

df = px.data.iris()

fig = px.scatter(df, x="sepal_width", y="sepal_length")

fig.update_layout(dragmode