CODEWITHSUNDEEP
pythontensorflowtensorflow-datasets
Mr_and_Mrs_D
Mr_and_Mrs_D

Reputation: 34026

How can I return the same batch twice from a tensorflow dataset iterator?

I am converting some legacy code to use the Dataset API - this code uses feed_dict to feed one batch to the train operation (actually three times) and then recalculates the losses for display using the same batch. So I need to have an iterator that returns the exact same batch two (or several) times. Unfortunately, I can't seem to find a way of doing it with tensorflow datasets - is it possible?

Upvotes: 4

Views: 1508

Answers (1)

mrry
mrry

Reputation: 126154

You can repeat individual elements of a Dataset using Dataset.flat_map(), Dataset.from_tensors() and Dataset.repeat() together. For example, to repeat elements twice:

NUM_REPEATS = 2
dataset = tf.data.Dataset.range(10)  # ...or the output of `.batch()`, etc.

# Repeat each element of `dataset` NUM_REPEATS times.
dataset = dataset.flat_map(
    lambda x: tf.data.Dataset.from_tensors(x).repeat(NUM_REPEATS))

Upvotes: 8

Related Questions