Why neural networks excel at classification in TensorFlow - Test Your Understanding

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a simple neural network layer using TensorFlow.

TensorFlow

import tensorflow as tf

model = tf.keras.Sequential([
    tf.keras.layers.Dense(10, activation=[1])
])

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

A"relu"

B"linear"

C"sigmoid"

D"softmax"

Attempts:

3 left

💡 Hint

Common Mistakes

Using linear activation which does not help learn complex patterns.

Choosing softmax in hidden layers instead of output layer.

The ReLU activation function helps neural networks learn complex patterns by introducing non-linearity.

2fill in blank

medium

Complete the code to compile the model with an appropriate loss function for classification.

TensorFlow

model.compile(optimizer='adam', loss=[1], metrics=['accuracy'])

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

A"mean_squared_error"

B"mean_absolute_error"

C"categorical_crossentropy"

D"hinge"

Attempts:

3 left

💡 Hint

Common Mistakes

Using mean squared error which is better for regression.

Choosing hinge loss which is for SVMs.

Categorical crossentropy is the right loss function for multi-class classification problems.

3fill in blank

hard

Fix the error in the code to correctly predict classes from model output probabilities.

TensorFlow

predictions = model.predict(test_data)
predicted_classes = tf.argmax(predictions, axis=[1])

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

A2

B1

C-1

D0

Attempts:

3 left

💡 Hint

Common Mistakes

Using axis 0 which selects the batch dimension.

Using axis -1 which may work but is less clear here.

Axis 1 selects the class dimension to find the index of the highest probability.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that maps words to their lengths only if length is greater than 3.

TensorFlow

word_lengths = {word: [1] for word in words if [2]

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

Alen(word)

Blen(word) > 3

Cword.startswith('a')

Dword.isalpha()

Attempts:

3 left

💡 Hint

Common Mistakes

Using the condition incorrectly or missing it.

Mapping to the word itself instead of its length.

The dictionary comprehension maps each word to its length only if the length is greater than 3.

5fill in blank

hard

Fill all three blanks to create a dictionary comprehension that maps uppercase words to their counts only if count is greater than 1.

TensorFlow

word_counts = [1]: [2] for word, count in counts.items() if [3]

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

Aword.upper()

Bcount

Ccount > 1

Dword.lower()

Attempts:

3 left

💡 Hint

Common Mistakes

Using lowercase instead of uppercase for keys.

Not filtering counts properly.

This comprehension creates a dictionary with uppercase words as keys and their counts as values, filtering counts greater than 1.