TensorFlow tf.data Pipelines

Question 1

Which tf.data method applies a function to each element of a dataset?

Accepted Answer

dataset.map()

Answer

dataset.apply()

Answer

dataset.transform()

Answer

dataset.process()

Question 2

What does dataset.batch(32) do in tf.data?

Accepted Answer

Groups consecutive elements into batches of 32

Answer

Shuffles 32 elements

Answer

Repeats dataset 32 times

Answer

Takes first 32 elements

Question 3

Which tf.data method shuffles dataset elements?

Accepted Answer

dataset.shuffle(buffer_size)

Answer

dataset.randomize()

Answer

dataset.mix()

Answer

dataset.permute()

Question 4

What does dataset.prefetch(tf.data.AUTOTUNE) do?

Accepted Answer

Prepares the next batch while the current one is being processed

Answer

Caches dataset to memory

Answer

Parallelizes map operations

Answer

Compresses dataset elements

Question 5

Which function creates a tf.data.Dataset from a list of NumPy arrays?

Accepted Answer

tf.data.Dataset.from_tensor_slices()

Answer

tf.data.Dataset.from_array()

Answer

tf.data.Dataset.load()

Answer

tf.data.Dataset.from_numpy()

Question 6

What does dataset.repeat(3) do in tf.data?

Accepted Answer

Iterates through the dataset 3 times

Answer

Repeats each element 3 times

Answer

Duplicates the dataset into 3 copies

Answer

Creates 3 parallel data streams