Q.What would happen if we use df.plot(kind=’scatter’) instead of plt.scatter() in Program 4-13?
[Context: Program 4-13 is:
import numpy as np
import matplotlib.pyplot as plt
discount= np.array([10,20,30,40,50])
saleInRs=np.array([40000,45000,48000,50000,100000])
size=discount10
plt.scatter(x=discount,y=saleInRs,s=size,color='red',linewidth=3,marker='',edgecolor='blue')
plt.title('Sales Vs Discount')
plt.xlabel('Discount offered')
plt.ylabel('Sales in Rs')
plt.show()]

Shape of Question

This is a plain theory question asking about the difference between two plotting methods in Python.

Explanation

The two methods—plt.scatter() from Matplotlib and df.plot(kind='scatter') from pandas—are fundamentally different in how they accept data and what they expect as input.

plt.scatter() works directly with array-like objects (NumPy arrays, lists, etc.). You pass the x and y data explicitly as separate arguments, along with optional parameters like s (size), color, marker, edgecolor, and linewidth. This is what Program 4-13 does: it takes two NumPy arrays (discount and saleInRs) and plots them with customized marker properties.

df.plot(kind='scatter'), on the other hand, is a pandas DataFrame method. It requires:

• A DataFrame object to call it on
• Column names (as strings) for the x and y parameters, not raw arrays
• The data must already be organized in a DataFrame structure

In Program 4-13, discount and saleInRs are NumPy arrays, not a DataFrame. If you try to write df.plot(kind='scatter'), Python will immediately raise a NameError because no DataFrame named df exists in the code. Even if you had created a DataFrame, say:

import pandas as pd
df = pd.DataFrame({'discount': discount, 'saleInRs': saleInRs})

you would then need to call:

df.plot(kind='scatter', x='discount', y='saleInRs', s=size, color='red', ...)

Notice that x and y now take column names (strings), not the arrays themselves. The s, color, marker, edgecolor, and linewidth parameters are passed through to the underlying Matplotlib scatter function, so they work similarly—but the data structure requirement is different. …

Why the two approaches are fundamentally different

The question asks what happens when we swap a Matplotlib function for a pandas plotting method. The answer hinges on understanding that these tools operate on different data structures and have different interfaces.

Program 4-13 works directly with NumPy arrays (discount and saleInRs). The plt.scatter() function from Matplotlib is designed to accept raw array-like data — lists, NumPy arrays, or sequences — and plot them directly. You pass x and y coordinates explicitly, along with optional parameters like marker size (s), color, marker shape, edge color, and line width.

The pandas method df.plot(kind='scatter'), on the other hand, is a DataFrame method. It expects to be called on a pandas DataFrame object, and you specify column names (as strings) for the x and y axes. The syntax would look like:

df.plot(kind='scatter', x='column_name_for_x', y='column_name_for_y')

What would actually happen

If you tried to write:

df.plot(kind='scatter')

in Program 4-13, Python would raise an error because there is no DataFrame object named df in the code. The variables discount and saleInRs are NumPy arrays, not DataFrames, and NumPy arrays do not have a .plot() method.

Even if you first converted the arrays into a DataFrame:

import pandas as pd
df = pd.DataFrame({'discount': discount, 'saleInRs': saleInRs})
df.plot(kind='scatter', x='discount', y='saleInRs')

you would encounter functional limitations:

The pandas wrapper is convenient when your data is already in a DataFrame and you want quick exploratory plots, but it abstracts away some of Matplotlib's fine-grained control. Notice the parameter name differences: pandas uses edgecolors and linewidths (plural), while the raw Matplotlib function uses edgecolor and linewidth (singular) — though both accept the same values. …