How to plot Andrews curves using Pandas in Python?

Andrews curves are used for visualizing high-dimensional data by mapping each observation onto a function. It preserves means, distance, and variances. It is given by formula:

T(n) = x_1/sqrt(2) + x_2 sin(n) + x_3 cos(n) + x_4 sin(2n) + x_5 cos(2n) + …

Plotting Andrews curves on a graph can be done using the andrews_curves() method of the plotting module. This function generates a matplotlib plot of Andrews curves, for visualising clusters of multivariate data.

Syntax: andrews_curves(frame, class_column, ax=None, samples=200, color=None, colormap=None, **kwargs)

Parameters:

• frame: It is the data to be plotted.
• class_column: This is the name of the column containing class names.
• ax: This parameter is a matplotlib axes object. Its default value is None.
• samples: This parameter is the number of points to plot in each curve.
• color: This parameter is an optional parameter and it is the list or tuple of colors to use for the different classes.
• colormap: This parameter is the string/matplotlib colormap object. Its default value is None.

Returns: This function returns an object of class matplotlib.axis.Axes

Example 1: In the following example, A data frame is made from the CSV file and the data frame is used to plot andrews_curves.

Python3

# importing various package
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

# making data frame from csv file
df = pd.read_csv(
    'C:\\Users\\digital india\\Desktop\\pand.csv'
)

# Creating Andrews curves
x = pd.plotting.andrews_curves(df, 'animal')

# plotting the Curve
x.plot()

# Display
plt.show()

Output:

Example 2: 

Python3

# importing various package
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

# making data frame from csv file
df = pd.read_csv(
    'https://raw.github.com/pandas-dev/'
    'pandas/master/pandas/tests/io/data/csv/iris.csv'
)

# Creating Andrews curves
x = pd.plotting.andrews_curves(df, 'Name')

# plotting the Curve
x.plot()

# Display
plt.show()

Output: