|TensorFlow 1 version||View source on GitHub|
Fused implementation of
tf.data.experimental.map_and_batch( map_func, batch_size, num_parallel_batches=None, drop_remainder=False, num_parallel_calls=None )
batch_size consecutive elements of this dataset
and then combines them into a batch. Functionally, it is equivalent to
batch. However, by fusing the two transformations together, the
implementation can be more efficient. Surfacing this transformation in the API
is temporary. Once automatic input pipeline optimization is implemented,
the fusing of
batch will happen automatically and this API will be
map_func: A function mapping a nested structure of tensors to another nested structure of tensors.
tf.Tensor, representing the number of consecutive elements of this dataset to combine in a single batch.
num_parallel_batches: (Optional.) A
tf.Tensor, representing the number of batches to create in parallel. On one hand, higher values can help mitigate the effect of stragglers. On the other hand, higher values can increase contention if CPU is scarce.
drop_remainder: (Optional.) A
tf.Tensor, representing whether the last batch should be dropped in case its size is smaller than desired; the default behavior is not to drop the smaller batch.
num_parallel_calls: (Optional.) A
tf.Tensor, representing the number of elements to process in parallel. If not specified,
batch_size * num_parallel_batcheselements will be processed in parallel. If the value
tf.data.experimental.AUTOTUNEis used, then the number of parallel calls is set dynamically based on available CPU.
Dataset transformation function, which can be passed to
ValueError: If both