Tensor shapes and reshaping in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a tensor of shape (2, 3) filled with zeros.

TensorFlow

import tensorflow as tf

zeros_tensor = tf.zeros([1])
print(zeros_tensor.shape)

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

A3, 2

B[2, 3]

C2, 3

D(2, 3)

Attempts:

3 left

💡 Hint

Common Mistakes

Using a list instead of a tuple for the shape.

Passing dimensions without parentheses.

The tf.zeros function expects the shape as a tuple. So (2, 3) correctly creates a tensor with 2 rows and 3 columns.

2fill in blank

medium

Complete the code to reshape a tensor of shape (6,) into shape (2, 3).

TensorFlow

import tensorflow as tf

original = tf.constant([1, 2, 3, 4, 5, 6])
reshaped = tf.reshape(original, [1])
print(reshaped.shape)

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

A[2, 3]

B(2, 3)

C2, 3

D[3, 2]

Attempts:

3 left

💡 Hint

Common Mistakes

Using a list instead of a tuple for the shape.

Passing dimensions without parentheses.

The tf.reshape function expects the new shape as a tuple. So (2, 3) reshapes the tensor into 2 rows and 3 columns.

3fill in blank

hard

Fix the error in the code to get the shape of a tensor correctly.

TensorFlow

import tensorflow as tf

tensor = tf.constant([[1, 2], [3, 4]])
shape = tensor.[1]()
print(shape)

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

Ashape

Bget_shape()

Cshape()

Dget_shape

Attempts:

3 left

💡 Hint

Common Mistakes

Using shape() which is not a method.

Using get_shape() which returns a TensorShape object but is less common in eager execution.

In TensorFlow, shape is a property that returns the shape of the tensor. Using tensor.shape correctly gets the shape.

4fill in blank

hard

Fill both blanks to create a tensor of shape (3, 4) and then reshape it to (4, 3).

TensorFlow

import tensorflow as tf

initial_tensor = tf.constant([[1, 2, 3, 4], [5, 6, 7, 8], [9, 10, 11, 12]])
reshaped_tensor = tf.reshape(initial_tensor, [1])
print(reshaped_tensor.shape == [2])

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

A(4, 3)

B(3, 4)

C[4, 3]

D[3, 4]

Attempts:

3 left

💡 Hint

Common Mistakes

Using lists instead of tuples for shapes.

Mixing up the order of dimensions.

The initial tensor has shape (3, 4). We reshape it to (4, 3). The print statement checks if the reshaped tensor's shape equals the original shape (3, 4).

5fill in blank

hard

Fill all three blanks to create a tensor, reshape it, and check if the reshaped tensor has the expected shape.

TensorFlow

import tensorflow as tf

original = tf.constant([1, 2, 3, 4, 5, 6])
reshaped = tf.reshape(original, [1])
expected_shape = [2]
print(reshaped.shape == expected_shape and reshaped.shape[0] == [3])

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

A(2, 3)

B(3, 2)

C2

D3

Attempts:

3 left

💡 Hint

Common Mistakes

Mixing up the order of dimensions in the shape.

Using lists instead of tuples for shapes.

The tensor is reshaped to (2, 3). The expected shape is (3, 2). The print checks if the reshaped tensor's shape matches the expected shape and if the first dimension is 2.

Note: The print will output False because the shapes differ, showing the importance of matching shapes.