Difference between data frames and matrices in Python Pandas?

In this article, we will explore the differences between DataFrames and matrices in Python Pandas. Both are 2-dimensional data structures, but they serve different purposes and have distinct characteristics.

Both DataFrames and matrices are 2-dimensional data structures. In general, DataFrames can include multiple types of data (numeric, character, factor, etc) while matrices can only store one type of data.

DataFrame in Python

In Python, a DataFrame is a two-dimensional, tabular, mutable data structure that may store tabular data containing objects of various data types. A DataFrame has axes that are labeled in the form of rows and columns. DataFrames are useful tools in data preprocessing because they provide valuable data handling methods. DataFrames can also be used to create pivot tables and plot data with Matplotlib.

Applications of DataFrame

• Data frames can perform a variety of tasks, such as fit statistical formulas.

• Data processing (Not possible with Matrix, first converting to Data Frame is mandatory)

• Transposing rows to columns and vice versa is feasible, which is useful in Data Science.

Creating a Sample DataFrame

Example

The following program creates a DataFrame using the DataFrame() function ?

# importing pandas, numpy modules with alias names
import pandas as pd
import numpy as np

# creating a dataframe
inputDataframe = pd.DataFrame({
    'Name': ['Virat', 'Rohit', 'Meera', 'Nick', 'Sana'], 
    'Jobrole': ['Developer', 'Analyst', 'Help Desk', 'Database Developer', 'Finance accountant'], 
    'Age': [25, 30, 28, 25, 40]
})

# displaying the dataframe
print(inputDataframe)

    Name             Jobrole  Age
0  Virat           Developer   25
1  Rohit            Analyst   30
2  Meera          Help Desk   28
3   Nick  Database Developer   25
4   Sana  Finance accountant   40

Matrix in Python

Matrix is a homogeneous collection of data sets organized in a two-dimensional rectangular grid. It's an m*n array with the same data type. There are a fixed number of rows and columns. Python supports numerous arithmetic operations such as addition, subtraction, multiplication, and division on Matrix.

Applications of Matrix

• It is very useful in Economics for calculating statistics such as GDP (Gross Domestic Product) or PI (Price per capita income).

• It's also useful for studying electrical and electronic circuits.

• Matrices are utilized in survey research, such as plotting graphs.

• This is useful in probability and statistics.

Matrix Multiplication by Converting a Matrix to DataFrame

Example

The following program demonstrates matrix multiplication using DataFrames ?

# importing pandas module
import pandas as pd

# input matrix 1
inputMatrix_1 = [[1, 2, 2],
                 [1, 2, 0],
                 [1, 0, 2]]

# input matrix 2
inputMatrix_2 = [[1, 0, 1],
                 [2, 1, 1],
                 [2, 1, 2]]

# creating a dataframe of first matrix
df_1 = pd.DataFrame(data=inputMatrix_1)

# creating a dataframe of second matrix
df_2 = pd.DataFrame(data=inputMatrix_2)

# printing the dataframe of input matrix 1
print("inputMatrix_1:")
print(df_1)
print("The dimensions(shape) of input matrix 1:")
print(df_1.shape)
print()

# printing the dataframe of input matrix 2
print("inputMatrix_2:")
print(df_2)
print("The dimensions(shape) of input matrix 2:")
print(df_2.shape)
print()

# multiplying both the matrices inputMatrix_1 and inputMatrix_2
result_mult = df_1.dot(df_2)

# Printing the resultant of matrix multiplication
print("Resultant Matrix after Matrix multiplication:")
print(result_mult)
print("The dimensions(shape) of Resultant Matrix:")
print(result_mult.shape)

inputMatrix_1:
   0  1  2
0  1  2  2
1  1  2  0
2  1  0  2

The dimensions(shape) of input matrix 1:
(3, 3)

inputMatrix_2:
   0  1  2
0  1  0  1
1  2  1  1
2  2  1  2

The dimensions(shape) of input matrix 2:
(3, 3)

Resultant Matrix after Matrix multiplication:
   0  1  2
0  9  4  7
1  5  2  3
2  5  2  5
The dimensions(shape) of Resultant Matrix:
(3, 3)

Comparison

Conclusion

DataFrames are ideal for data analysis with mixed data types and labeled axes, while matrices are better for mathematical computations with homogeneous numeric data. Choose based on your specific use case and data requirements.