2

Here is an example of a seaborn bivariate KDE plot from the Iris dataset. Code:

from sklearn import datasets
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
  
# Setting up the Data Frame
iris = datasets.load_iris()
  
iris_df = pd.DataFrame(iris.data, columns=['Sepal_Length',
                      'Sepal_Width', 'Patal_Length', 'Petal_Width'])
  
# Adding the Target column to the df which indicates the species
iris_df['Target'] = iris.target

# Iris Setosa corresponds to 0 value in Target column
iris_setosa = iris_df.query("Target==0")

# Plotting the KDE Plot
sns.kdeplot(iris_setosa['Sepal_Length'], iris_setosa['Sepal_Width'], shade=True, Label='Iris_Setosa',cbar=True)

plt.show()

Click here to view the KDE plot

I wish to extract the data from the plot and store it. I am able to extract the polygons for each contour line. However, I can't seem to find a way to extract the z / threshold values for each contour line, i.e. the values along the right-hand side of the plot which correspond to a certain colour for the contour lines.

Can anyone help me with this?

JohanC
  • 71,591
  • 8
  • 33
  • 66
Lucy
  • 29
  • 1
  • 3
    Unless you're trying to manipulate the figure itself, you should use a stats library for this, not a visualization library. – mwaskom Aug 12 '22 at 13:25

0 Answers0