Freezing and unfreezing layers in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to freeze all layers in a TensorFlow Keras model.

TensorFlow

for layer in model.layers:
    layer.[1] = False

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

Afrozen

Btrainable

Cactive

Dlocked

Attempts:

3 left

💡 Hint

Common Mistakes

Using a non-existent property like 'frozen' or 'locked'.

Confusing the value True/False for freezing.

Setting layer.trainable = True or False controls whether the layer's weights are updated during training. To freeze a layer, set trainable = False.

2fill in blank

medium

Complete the code to unfreeze only the last layer of the model.

TensorFlow

model.layers[-1].[1] = True

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

Aactive

Bfrozen

Ctrainable

Dlocked

Attempts:

3 left

💡 Hint

Common Mistakes

Trying to set a property that does not exist.

Forgetting to set trainable to True.

To allow training on the last layer, set its trainable property to True.

3fill in blank

hard

Fix the error in the code to freeze all layers before compiling the model.

TensorFlow

for layer in model.layers:
    layer.[1] = False
model.compile(optimizer='adam', loss='categorical_crossentropy')

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

Atrainable

Bfrozen

Cactive

Dlocked

Attempts:

3 left

💡 Hint

Common Mistakes

Using a non-existent property like 'frozen'.

Setting the property after compiling the model.

The correct property to freeze layers is trainable. Setting trainable = False freezes the layer.

4fill in blank

hard

Fill both blanks to freeze all layers except the last one.

TensorFlow

for layer in model.layers[:[1]]:
    layer.[2] = False

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

Atrainable

B-1

Cfrozen

Dlocked

Attempts:

3 left

💡 Hint

Common Mistakes

Using incorrect slice indices.

Using a wrong property name to freeze layers.

Using model.layers[:-1] selects all layers except the last. Setting trainable = False freezes those layers.

5fill in blank

hard

Fill all three blanks to freeze the first two layers and unfreeze the rest.

TensorFlow

for layer in model.layers[:[1]]:
    layer.[2] = False
for layer in model.layers[[3]:]:
    layer.trainable = True

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

Atrainable

B2

Dfrozen

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong slice indices.

Using incorrect property names.

Slice model.layers[:2] selects the first two layers to freeze by setting trainable = False. Slice model.layers[2:] selects the rest to unfreeze.