TensorFlow Advanced TensorFlow Techniques 2

Question 1

What is the purpose of tf.keras.callbacks.LambdaCallback?

Accepted Answer

Allows arbitrary functions to be called at training events without writing a full callback class

Answer

Schedules lambda layer updates

Answer

Applies gradient transformations

Answer

Logs lambda function outputs

Question 2

What does tf.config.list_physical_devices('GPU') return?

Accepted Answer

A list of available GPU devices on the system

Answer

GPU configuration settings

Answer

The number of GPU cores

Answer

GPU memory capacity

Question 3

Which strategy distributes training across multiple GPUs on a single machine?

Accepted Answer

tf.distribute.MirroredStrategy

Answer

tf.distribute.MultiWorkerMirroredStrategy

Answer

tf.distribute.CentralStorageStrategy

Answer

tf.distribute.OneDeviceStrategy

Question 4

What is the purpose of tf.debugging.assert_shapes()?

Accepted Answer

Verifies tensor shapes match expectations and raises an error if not

Answer

Asserts that shapes are undefined

Answer

Enforces shapes during graph compilation only

Answer

Checks shape compatibility between layers

Question 5

What does tf.keras.backend.clear_session() do?

Accepted Answer

Resets Keras state including the global graph and frees GPU memory

Answer

Clears the training data cache

Answer

Resets model weights only

Answer

Closes all open files

Question 6

Which TensorFlow API generates text, image, or other creative content using generative models?

Accepted Answer

TF-GAN library

Answer

TensorFlow Generative (tf.keras with custom generators/discriminators)

Answer

tf.generate()

Answer

tf.creative()