You can set the color palette in Seaborn with the following code. The given example helps you to understand how to set the color palette of the Seaborn plot. I highly recommend you “Python Crash Course Book” to learn Python.
Example 1: Set your own colors
# Import the required libraries import matplotlib.pyplot as plt import seaborn as sns # Define color codes colors = ["#0000FF", "#FFA500","#00FF00","#FFFF00","#FFCC99"] # Data data = [10, 30, 20, 15, 25] # Labels labels = ['A', 'B', 'C', 'D', 'E'] # Set Color Palette sns.set_palette(sns.color_palette(colors)) # Create Pie Chart plt.pie(data, labels = labels) # Display Pie Chart plt.show()
Output:

Example 2: Set predefined color palette (rocket)
# Import the required libraries import matplotlib.pyplot as plt import seaborn as sns # Data data = [10, 30, 20, 15, 25] # Labels labels = ['A', 'B', 'C', 'D', 'E'] # Set Color Palette sns.set_palette(sns.color_palette("rocket")) # Create Pie Chart plt.pie(data, labels = labels) # Display Pie Chart plt.show()

Example 3: Set predefined color palette (Paired)
# Import the required libraries import matplotlib.pyplot as plt import seaborn as sns # Data data = [10, 30, 20, 15, 25] # Labels labels = ['A', 'B', 'C', 'D', 'E'] # Set Color Palette sns.set_palette(sns.color_palette("Paired")) # Create Pie Chart plt.pie(data, labels = labels) # Display Pie Chart plt.show()
