How to Read CSV File with NumPy (Step-by-Step)

by Tutor Aspire January 17, 2023

You can use the following basic syntax to read a CSV file into a record array in NumPy:

from numpy import genfromtxt

my_data = genfromtxt('data.csv', delimiter=',', dtype=None)

The following step-by-step example shows how to use this syntax in practice.

Step 1: View the CSV File

Suppose we have the following CSV file called data.csv that we’d like to read into NumPy:

The following code shows how to read in this CSV file into a Numpy array:

from numpy import genfromtxt

#import CSV file
my_data = genfromtxt('data.csv', delimiter=',', dtype=None)

Note the following:

delimiter: This specifies the delimiter that separates the data values in the CSV file.
dtype: This specifies the data type for the NumPy array. By using None, we allow multiple data types to be imported at once within the array.

Once we’ve imported the CSV file, we can view it:

#view imported CSV file
my_data

array([[1, 2, 2, 2, 3, 4],
       [5, 5, 6, 8, 9, 9]])

We can see that the data in the NumPy array matches the data shown in the CSV file.

Note: You can find the complete online documentation for the genfromtxt() function here.

The following tutorials explain how to perform other common functions with CSV files in pandas: