LSTM layer in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to add an LSTM layer with 50 units to the model.

TensorFlow

model.add(tf.keras.layers.LSTM([1]))

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

Aactivation='relu'

BDense

C50

Dinput_shape=(10, 1)

Attempts:

3 left

💡 Hint

Common Mistakes

Using layer types like 'Dense' instead of a number.

Passing activation functions inside LSTM constructor incorrectly.

The LSTM layer requires the number of units as an integer. Here, 50 units are specified.

2fill in blank

medium

Complete the code to define the input shape for the LSTM layer with 100 time steps and 20 features.

TensorFlow

model.add(tf.keras.layers.LSTM(64, input_shape=[1]))

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

A(100, 20)

B[100, 20]

C(20, 100)

D100, 20

Attempts:

3 left

💡 Hint

Common Mistakes

Swapping time steps and features order.

Using list instead of tuple for input_shape.

The input_shape for LSTM is a tuple (timesteps, features). Here, 100 time steps and 20 features are correct.

3fill in blank

hard

Fix the error in the code by completing the LSTM layer to return sequences for stacking.

TensorFlow

model.add(tf.keras.layers.LSTM(32, return_sequences=[1]))

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

ATrue

BNone

C0

DFalse

Attempts:

3 left

💡 Hint

Common Mistakes

Leaving return_sequences as False causes shape errors when stacking.

Using None or 0 instead of True.

Setting return_sequences=True makes the LSTM output a sequence, which is needed when stacking LSTM layers.

4fill in blank

hard

Fill both blanks to create a stacked LSTM model with 64 units in the first layer returning sequences, and 32 units in the second layer.

TensorFlow

model = tf.keras.Sequential()
model.add(tf.keras.layers.LSTM([1], return_sequences=[2], input_shape=(50, 10)))
model.add(tf.keras.layers.LSTM(32))

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

A64

BTrue

CFalse

D32

Attempts:

3 left

💡 Hint

Common Mistakes

Setting return_sequences to False in the first layer.

Using wrong number of units in the first layer.

The first LSTM layer has 64 units and must return sequences (True) to feed the next LSTM layer.

5fill in blank

hard

Fill all three blanks to build a model with an LSTM layer of 100 units, dropout rate 0.2, and a Dense output layer with 1 unit.

TensorFlow

model = tf.keras.Sequential([
    tf.keras.layers.LSTM([1], dropout=[2], input_shape=(30, 5)),
    tf.keras.layers.Dense([3])
])

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

A100

B0.2

C1

D0.5

Attempts:

3 left

💡 Hint

Common Mistakes

Using dropout values outside 0 to 1 range.

Setting wrong number of units in Dense layer.

The LSTM layer has 100 units with dropout 0.2, and the Dense layer outputs 1 unit for prediction.