Related Articles

# Python – Iterate over Columns in NumPy

• Last Updated : 03 Jun, 2020

`Numpy` (abbreviation for ‘Numerical Python‘) is a library for performing large scale mathematical operations in fast and efficient manner. This article serves to educate you about methods one could use to iterate over columns in an 2D `NumPy` array. Since a single dimensional array only consists of linear elements, there doesn’t exists a distinguished definition of rows and columns in them. Therefore, in order to perform such operations we need a array whose `len(ary.shape) > 1 `.

To install `NumPy` on your python environment, type the following code in your OS’s Command Processor (CMD, Bash etc):

pip install numpy

We would be taking a look at several methods of iterating over a column of an Array/Matrix:-

METHOD 1:

CODE: Use of primitive 2D Slicing operation on an array to get the desired column/columns

 `import` `numpy as np`` ` `# Creating a sample numpy array (in 1D)``ary ``=` `np.arange(``1``, ``25``, ``1``)`` ` `# Converting the 1 Dimensional array to a 2D array ``# (to allow explicitly column and row operations)``ary ``=` `ary.reshape(``5``, ``5``)`` ` `# Displaying the Matrix (use print(ary) in IDE)``print``(ary)`` ` `# This for loop will iterate over all columns of the array one at a time``for` `col ``in` `range``(ary.shape[``1``]):``    ``print``(ary[:, col])`

Output:

```[[ 0,  1,  2,  3,  4],
[ 5,  6,  7,  8,  9],
[10, 11, 12, 13, 14],
[15, 16, 17, 18, 19],
[20, 21, 22, 23, 24]])

[ 0  5 10 15 20]
[ 1  6 11 16 21]
[ 2  7 12 17 22]
[ 3  8 13 18 23]
[ 4  9 14 19 24]

```

Explanation:

In the above code, we firstly create an linear array of 25 elements (0-24) using `np.arange(25)`. Then we reshape (transform 1D to 2D) using `np.reshape()` to create a 2D array out of a linear array. Then we output the transformed array. Now we used a for loop which would iterate x times (where x is the number of columns in the array) for which we used `range()` with the argument `ary.shape[1]` (where `shape[1]` = number of columns in a 2D symmetric array). In each iteration we output a column out of the array using `ary[:, col]` which means that give give all elements of the column number = `col`.

METHOD 2:
In this method we would transpose the array to treat each column element as a row element (which in turn is equivalent of column iteration).

Code:

 `# libraries``import` `numpy as np`` ` `# Creating an 2D array of 25 elements ``ary ``=` `np.array([[ ``0``,  ``1``,  ``2``,  ``3``,  ``4``],``                ``[ ``5``,  ``6``,  ``7``,  ``8``,  ``9``],``                ``[``10``, ``11``, ``12``, ``13``, ``14``],``                ``[``15``, ``16``, ``17``, ``18``, ``19``],``                ``[``20``, ``21``, ``22``, ``23``, ``24``]])`` ` ` ` `# This loop will iterate through each row of the transposed ``# array (equivalent of iterating through each column)``for` `col ``in` `ary.T:``    ``print``(col)`

Output:

```
[ 0  5 10 15 20]
[ 1  6 11 16 21]
[ 2  7 12 17 22]
[ 3  8 13 18 23]
[ 4  9 14 19 24]
```

Explanation:
Firstly, we created an 2D array (same as the previous example) using `np.array()` and initialized it with 25 values. Then we transposed the array, using `ary.T` which in turn switches the rows with the columns and columns with the rows. Then we iterated over each row of this transposed array and printed the row values.

Attention geek! Strengthen your foundations with the Python Programming Foundation Course and learn the basics.

To begin with, your interview preparations Enhance your Data Structures concepts with the Python DS Course. And to begin with your Machine Learning Journey, join the Machine Learning – Basic Level Course

My Personal Notes arrow_drop_up