What is the RadViz plot in pandas?

RadViz plot is a visualization technique used to display a dataset’s features in a circle, simply put, as a 2-dimension dataset.

RadViz plots display the data points in a circle by showing which data points tend to have certain features and therefore have similar characteristics.

Since the position of each data point is determined by the values of all the variables that make it, data points with extreme characteristics will appear on extreme sides of the circle while those that stand out less will mostly be at the center of the circle.

It’s helpful in displaying cluster attributes or different class characteristics in classification problems as they allow for as many features as will fit within a circle’s perimeter to be displayed and therefore well visualized.

Practical use of RadViz plots

Suppose we have a dataset containing diabetes patients. This dataset has 10 features that contain the patients’ names, ages, blood sugar levels, and other relevant characteristics useful in identifying whether a patient has diabetes or not.

In this case, a RadViz plot will be helpful in displaying which features are related to patients with diabetes and which ones are not.

This allows for distinguishable characteristic features belonging to a certain class, to be known.

Implementing RadViz plots in pandas

Syntax

pandas.plotting.radviz(frame, class_column, ax=None, color=None, colormap=None, **kwds)

Parameters

frame: Dataset/dataframe under consideration

class_column: Column containing the class attribute

ax(optional): A plot to which the information will be added

color(optional): Assigns a color to each category.

colormap(optional): Colormap to select colors from

**kwds: Options to pass to matplotlib scatter plotting methods

Coding example

import pandas as pd

from sklearn.datasets import load_iris

import matplotlib.pyplot as plt



data = load_iris(as_frame = True)

df = pd.DataFrame(data=data['data'],columns=data['feature_names'])

df['class'] = data.target



plt.figure(figsize = (10,6),dpi=600)

pd.plotting.radviz(frame = df,class_column = 'class')

Explanation

Lines 1 and 3: We import needed libraries.

Lines 5–7: We load the iris dataset into a DataFrame.

Line 10: We plot the RadViz plots of the dataset.

Conclusion

The RadViz plot shows three classes of the iris dataset with their different characteristic features along a 2-D space/circle.

  • If two variables have a strong correlation, their data points will be located close to each other on the RadViz plot. Conversely, if two variables are uncorrelated, their dots will be located far apart from each other. This RadViz plot, therefore, clearly shows that classes 1 and 2 are strongly correlated.

  • Further, if the dots for each species are well separated, it suggests that the species can be differentiated based on the features. So, it’s clear that class 0 can be differentiated from classes 1 and 2 on the basis of features.

  • Lastly, it is also evident that sepal width has more influence in the classification of class 0 from other classes.

Free Resources

Copyright ©2026 Educative, Inc. All rights reserved