Broadcasting

Introduction to NumPy

Izzy Weber

Core Curriculum Manager, DataCamp

Broadcasting introduced

A graphic showing two arrays with different shapes begin added together

Image showing a 2D array being added to the number 2

Image showing a 2D array being added to any array full of twos which has the same shape

A graphic applying the compatibility rules to arrays shapes of (10, 5 and (10, 1))

NumPy compares sets of array dimensions from right to left
Two dimensions are compatible when...
- One of them has a length of one or
- They are of equal lengths
All dimension sets must be compatible

A graphic applying the compatibility rules to arrays shapes of (10, 5 and (10, 1))

Broadcastable shapes:

Shapes which are not broadcastable:

An array with shape (2, 5) being added to an array with shape (1, 5)

array = np.arange(10).reshape((2, 5))
array + np.array([0, 1, 2, 3, 4])

array([[ 0,  2,  4,  6,  8],
       [ 5,  7,  9, 11, 13]])

An array with shape (2, 5) being added to an array with shape (2, 5)

A graphic showing that adding an array of shape (2, 5) to one of shape (2,) generates a ValueError.

array = np.arange(10).reshape((2, 5))
array + np.array([0, 1])

ValueError: operands could not be broadcast together with shapes (2,5) (2,)

A graphic showing an array of shape (2, 5) and one of shape (2, 1) being added together

array = np.arange(10).reshape((2, 5))
array + np.array([0, 1]).reshape((2, 1))

array([[ 0,  1,  2,  3,  4],
       [ 6,  7,  8,  9, 10]])

A graphic showing the single-column array stretched column-wise across an array with the same shape as the larger array

An array with shape (2, 2) being multiplied by an array with shape (2, 1)

An array with shape (2, ) being subtracted from an array with shape (2, 2)

Introduction to NumPy