Activation functions (ReLU, sigmoid, softmax) in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to apply the ReLU activation function to the tensor.

TensorFlow

import tensorflow as tf

x = tf.constant([-3.0, 0.0, 5.0])
relu_output = tf.nn.[1](x)
print(relu_output.numpy())

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

Arelu

Bsigmoid

Csoftmax

Dtanh

Attempts:

3 left

💡 Hint

Common Mistakes

Using sigmoid or softmax instead of relu for this task.

Trying to call activation as a method on the tensor.

The ReLU activation function is applied using tf.nn.relu. It outputs zero for negative inputs and the input itself if positive.

2fill in blank

medium

Complete the code to apply the sigmoid activation function to the tensor.

TensorFlow

import tensorflow as tf

x = tf.constant([-1.0, 0.0, 1.0])
sigmoid_output = tf.nn.[1](x)
print(sigmoid_output.numpy())

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

Asoftmax

Brelu

Csigmoid

Delu

Attempts:

3 left

💡 Hint

Common Mistakes

Using relu or softmax instead of sigmoid.

Confusing sigmoid with tanh.

The sigmoid activation function is applied using tf.nn.sigmoid. It outputs values between 0 and 1, useful for probabilities.

3fill in blank

hard

Fix the error in the code to correctly apply the softmax activation function along the last axis.

TensorFlow

import tensorflow as tf

logits = tf.constant([[1.0, 2.0, 3.0]])
softmax_output = tf.nn.softmax(logits, axis=[1])
print(softmax_output.numpy())

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

A0

B-1

C1

D2

Attempts:

3 left

💡 Hint

Common Mistakes

Using axis=0 or axis=1 which may not be the last axis.

Omitting the axis argument causing default behavior that may be incorrect.

The softmax function should be applied along the last axis of the logits tensor. Using axis=-1 ensures this regardless of tensor shape.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that maps each word to its sigmoid activation applied to its length.

TensorFlow

words = ['hi', 'hello', 'hey']
import tensorflow as tf
result = {word: tf.nn.[1](tf.constant(len(word), dtype=tf.float32)) for word in [2]
print(result)

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

Asigmoid

Brelu

Cwords

Drange(3)

Attempts:

3 left

💡 Hint

Common Mistakes

Using relu instead of sigmoid for activation.

Iterating over range instead of the list of words.

The sigmoid activation is applied to the length of each word. The dictionary comprehension iterates over the list words.

5fill in blank

hard

Fill all three blanks to create a dictionary comprehension that maps each word to its softmax probability over lengths greater than 2.

TensorFlow

words = ['hi', 'hello', 'hey', 'greetings']
import tensorflow as tf
lengths = [len(word) for word in words if len(word) [1] 2]
lengths_tensor = tf.constant(lengths, dtype=tf.float32)
softmax_vals = tf.nn.[2](lengths_tensor, axis=[3])
result = {word: softmax_vals[i].numpy() for i, word in enumerate(words) if len(word) > 2}
print(result)

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

A>

Bsigmoid

C-1

Dsoftmax

Attempts:

3 left

💡 Hint

Common Mistakes

Using sigmoid instead of softmax for activation.

Using wrong axis value for softmax.

Using '<' instead of '>' in the filter condition.

The dictionary comprehension maps words with length greater than 2 to their softmax probabilities. The softmax is applied along the last axis (-1).