Converting a Table to Matrix in MATLAB

In this detailed article, we will explore how to convert a table to matrix using MATLAB. We’ll discuss its importance in data science, machine learning, and computer vision applications while also del …

Updated October 18, 2023

Today’s sponsor is Depot. Depot is a remote container build service that’s half the cost of GitHub Actions runners and up to 40x faster than other build services. Try it for free.

In this detailed article, we will explore how to convert a table to matrix using MATLAB. We’ll discuss its importance in data science, machine learning, and computer vision applications while also delving into Python’s potential for equivalent operations. By the end of this tutorial, you will have gained hands-on experience with transforming tabular data in MATLAB and a better understanding of its relevance to other programming languages like Python.


MATLAB is an advanced programming language and interactive development environment that provides extensive capabilities for performing mathematical operations, visualizing data, and exploring algorithms. In this article, we will focus on the process of converting a table (also known as a tabular structure) to matrix form in MATLAB. This transformation plays a vital role in various applications such as data science, machine learning, and computer vision, where data is often organized into structured tables containing numeric or categorical values.

Motivation for Converting Tables to Matrices

Converting tables into matrices can significantly simplify subsequent computations due to the following reasons:

  1. Linear algebra functions in MATLAB operate on matrices. Matrix operations such as solving linear systems, performing matrix decompositions, and calculating eigenvalues become more efficient when dealing with numerical data organized in a matrix form.

  2. Machine learning algorithms often require data in structured formats for optimal performance, which are typically represented by matrices or tensors in MATLAB. By representing the tabular data as a matrix, we can feed it directly into these algorithms without additional transformations.

  3. In computer vision applications, image manipulation and analysis operations require matrices to efficiently process large collections of images. Converting image data into matrices allows us to work with powerful image processing libraries in MATLAB such as Image Processing Toolbox and Computer Vision System Toolbox.

Converting a Table to Matrix in Python

Before diving into the MATLAB implementation, we will briefly discuss how to convert tables to matrices using Python for comparison purposes. The primary steps involve reading tabular data from CSV or other formats into Pandas DataFrame and then converting it to a NumPy array which corresponds to a matrix in MATLAB:

  1. Import necessary libraries: pandas for handling tabular data, numpy for working with arrays, and matplotlib for plotting.

    import pandas as pd import numpy as np import matplotlib.pyplot as plt

  2. Read the CSV file into a Pandas DataFrame using pandas.read_csv():

    df = pd.read_csv(‘data/table.csv’)

  3. Convert the DataFrame to a NumPy array by specifying the column names as the first argument and setting the index as False to avoid converting it into a row vector:

    arr = np.array(df) arr = arr[…, np.newaxis] # For NxMx1 arrays (3D)

  4. Check if we successfully converted the table to a matrix using Python:

    print(arr.shape) # Should now be a 2-dimensional array or (N, M) for an NxM table

  5. Convert back to the equivalent MATLAB data structure by reshaping the NumPy array and converting it to a MATLAB matrix using scipy.misc.toarray():

    arr_mat = scipy.misc.toarray(arr)

Converting Tables to Matrices in MATLAB

Now that we have reviewed the process for Python, let’s dive into MATLAB with an example where we convert a table to a matrix:

  1. Create a sample dataset: In this case, we will use a simple table containing numerical values. You can generate your own tabular data or download a dataset from a reputable repository such as Kaggle, UCI Machine Learning Repository, or any other relevant source.

    table_data = {…} # Populate the table with numeric and categorical columns

  2. Load the MATLAB environment and import the necessary libraries for manipulating data structures and visualization:

    % Open MATLAB environment % Import required libraries

    load(‘table_data’) clear all addpath(genpath(‘C:\Users\YourUser\Desktop’)) formatCompact library_path = [‘’,‘stats’,‘signal’,‘image’] path(library_path) import table_data

  3. Create a MATLAB structure variable to store the data in tabular form using struct:

    myTableData = table_data

  4. Verify that the input is indeed a table by printing its properties and values using who:

    % Check the structure of the data

    who(myTableData) % Show the tabular data as a formatted string for easy reading disp([str2func(‘myTableData.'){:},’ - ‘, myTableData])

  5. Convert the table to a matrix using the MATLAB function mat2cell which returns a matrix as an N-by-1 cell array:

    mat_table = mat2cell(myTableData) % Display the new matrix structure disp([str2func(‘mat_table{:}.'){:},’ - ‘, mat_table])

Now that we have successfully converted our table into a matrix, let’s proceed with visualizing and analyzing the data using MATLAB functions.


In this detailed article, we explored the process of transforming tabular data from tables to matrices in both Python and MATLAB. We discussed its relevance in various fields such as data science, machine learning, and computer vision. With the ability to work with structured datasets using both languages, you are now equipped with a powerful skill set that can help streamline your data analysis and modeling efforts.