Input shape specification in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to specify the input shape for a Keras model.

TensorFlow

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

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

A(28, 28)

B[28, 28]

C28, 28

D28

Attempts:

3 left

💡 Hint

Common Mistakes

Using a list instead of a tuple for input_shape.

Including the batch size in the input shape.

The input_shape argument expects a tuple specifying the shape of each input sample, excluding the batch size. So (28, 28) is correct.

2fill in blank

medium

Complete the code to define an input layer with a 3D input shape.

TensorFlow

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

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

A(64, 64)

B(64, 64, 3)

C[64, 64, 3]

D64, 64, 3

Attempts:

3 left

💡 Hint

Common Mistakes

Using a list instead of a tuple.

Omitting the color channels dimension.

The input shape for a color image with height and width 64 pixels and 3 color channels is (64, 64, 3). It must be a tuple.

3fill in blank

hard

Fix the error in the input shape specification for a grayscale image.

TensorFlow

model = tf.keras.Sequential([
    tf.keras.layers.InputLayer(input_shape=[1]),
    tf.keras.layers.Flatten(),
    tf.keras.layers.Dense(10, activation='softmax')
])

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

A(28, 28, 1)

B28, 28, 1

C[28, 28, 1]

D(28, 28)

Attempts:

3 left

💡 Hint

Common Mistakes

Omitting the channel dimension for grayscale images.

Using a list instead of a tuple.

For grayscale images, the input shape should include the single channel dimension as (28, 28, 1). It must be a tuple.

4fill in blank

hard

Fill both blanks to create a model input layer and a dense layer with correct input shape.

TensorFlow

inputs = tf.keras.Input(shape=[1])
x = tf.keras.layers.Dense(32, activation=[2])(inputs)

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

A(100,)

B'relu'

C'sigmoid'

D(100, 1)

Attempts:

3 left

💡 Hint

Common Mistakes

Using a 2D shape for 1D input data.

Using 'sigmoid' activation in hidden layers instead of 'relu'.

The input shape for a 1D vector of length 100 is (100,). The dense layer activation is commonly set to 'relu' for hidden layers.

5fill in blank

hard

Fill all three blanks to define a model with input shape, flatten layer, and output layer activation.

TensorFlow

model = tf.keras.Sequential([
    tf.keras.layers.InputLayer(input_shape=[1]),
    tf.keras.layers.Flatten(),
    tf.keras.layers.Dense(10, activation=[2])
])

model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=[[3]])

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

A(28, 28)

B'softmax'

C'accuracy'

D(28, 28, 3)

Attempts:

3 left

💡 Hint

Common Mistakes

Using color image shape for grayscale input.

Using 'relu' instead of 'softmax' in output layer.

Omitting metrics or using incorrect metric names.

The input shape for grayscale images is (28, 28). The output layer uses 'softmax' activation for multi-class classification. The metric 'accuracy' measures how often predictions are correct.