Functional API basics in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create an input layer with shape (28, 28).

TensorFlow

inputs = tf.keras.Input(shape=[1])

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

A(28, 28)

B28

C[28, 28]

D(784,)

Attempts:

3 left

💡 Hint

Common Mistakes

Using a list instead of a tuple for shape.

Specifying the total number of pixels instead of the shape tuple.

The Input layer requires the shape as a tuple. (28, 28) correctly specifies the input shape for images of 28 by 28 pixels.

2fill in blank

medium

Complete the code to add a Dense layer with 64 units and ReLU activation.

TensorFlow

x = tf.keras.layers.Dense([1], activation='relu')(inputs)

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

A128

B64

C32

D10

Attempts:

3 left

💡 Hint

Common Mistakes

Choosing the wrong number of units.

Confusing activation function with units.

The Dense layer should have 64 units as specified, with ReLU activation.

3fill in blank

hard

Fix the error in the code to create the model using the Functional API.

TensorFlow

model = tf.keras.Model(inputs=[1], outputs=x)

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

Aoutputs

Bx

Cmodel

Dinputs

Attempts:

3 left

💡 Hint

Common Mistakes

Passing the output tensor as inputs.

Passing the model variable itself.

The inputs argument must be the input tensor, which is 'inputs' here, not 'x' or other variables.

4fill in blank

hard

Fill both blanks to flatten the input and add a Dense layer with 10 units.

TensorFlow

x = tf.keras.layers.[1]()(inputs)
x = tf.keras.layers.Dense([2])(x)

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

AFlatten

B10

CDense

D64

Attempts:

3 left

💡 Hint

Common Mistakes

Using Dense instead of Flatten for the first layer.

Choosing wrong number of units for the Dense layer.

First, the Flatten layer converts the input to 1D. Then, a Dense layer with 10 units is added for output classes.

5fill in blank

hard

Fill all three blanks to compile the model with Adam optimizer, sparse categorical crossentropy loss, and accuracy metric.

TensorFlow

model.compile(optimizer='[1]', loss='[2]', metrics=['[3]'])

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

Aadam

Bsparse_categorical_crossentropy

Caccuracy

Dsgd

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong optimizer like 'sgd' when 'adam' is expected.

Using categorical_crossentropy instead of sparse_categorical_crossentropy.

Forgetting to include accuracy metric.

Adam optimizer is popular for training. Sparse categorical crossentropy is used for integer labels in classification. Accuracy is a common metric to track.