Broadcasting rules in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to add two tensors using broadcasting.

TensorFlow

import tensorflow as tf

a = tf.constant([1, 2, 3])
b = tf.constant([[10], [20], [30]])
result = a [1] b
print(result.numpy())

Drag options to blanks, or click blank then click option'

A+

B-

C*

D/

Attempts:

3 left

💡 Hint

Common Mistakes

Using subtraction (-) instead of addition (+).

Trying to multiply (*) when the task is to add.

Using the plus sign (+) adds the two tensors element-wise with broadcasting.

2fill in blank

medium

Complete the code to reshape a tensor for broadcasting.

TensorFlow

import tensorflow as tf

x = tf.constant([1, 2, 3])
x_reshaped = tf.reshape(x, [1])
print(x_reshaped.shape)

Drag options to blanks, or click blank then click option'

A(1, 3)

B(3, 1)

C(3,)

D(1,)

Attempts:

3 left

💡 Hint

Common Mistakes

Using (1, 3) which changes the shape incorrectly.

Using (3,) which is the original shape and may not broadcast as intended.

Reshaping to (3, 1) prepares the tensor for broadcasting with a (3, 4) tensor along the second dimension.

3fill in blank

hard

Fix the error in the code to correctly broadcast and multiply tensors.

TensorFlow

import tensorflow as tf

x = tf.constant([[1, 2, 3], [4, 5, 6]])
y = tf.constant([10, 20])
result = x [1] y
print(result.numpy())

Drag options to blanks, or click blank then click option'

A-

B+

C*

D/

Attempts:

3 left

💡 Hint

Common Mistakes

Using addition (+) instead of multiplication (*).

Trying to divide (/) which is not the intended operation.

The multiplication operator (*) correctly broadcasts y to match x's shape for element-wise multiplication.

4fill in blank

hard

Fill in the blank to create a tensor and broadcast it for addition.

TensorFlow

import tensorflow as tf

x = tf.constant([[1, 2, 3], [4, 5, 6]])
y = tf.constant([1])
result = x + y
print(result.numpy())

Drag options to blanks, or click blank then click option'

A[10, 20, 30]

B[10, 20]

C[[10, 20, 30]]

D[[10], [20]]

Attempts:

3 left

💡 Hint

Common Mistakes

Using a 1D list like [10, 20] which has shape (2,) and may not broadcast as intended.

Using a list with shape (1, 3) which does not match x's shape for broadcasting.

Using [[10], [20]] allows broadcasting along the columns to add to x.

5fill in blank

hard

Fill in the blank to create a tensor and broadcast it for multiplication.

TensorFlow

import tensorflow as tf

x = tf.constant([[1, 2, 3], [4, 5, 6]])
y = tf.constant([1])
result = x * y
print(result.numpy())

Drag options to blanks, or click blank then click option'

A[[10], [20]]

B[10, 20]

C[[1, 2, 3]]

D[[10, 20, 30]]

Attempts:

3 left

💡 Hint

Common Mistakes

Using a 1D list like [10, 20] which won't broadcast correctly.

Using a shape that does not match for broadcasting.

y is [[10], [20]] to broadcast along columns; x shape is (2,3); multiplication works element-wise.