tf.data.experimental.AutotuneOptions

Represents options for autotuning dataset performance.

tf.data.experimental.AutotuneOptions()

options = tf.data.Options()
options.autotune.enabled = False
dataset = dataset.with_options(options)

autotune_algorithm When autotuning is enabled (through autotune), determines the algorithm to use.
cpu_budget When autotuning is enabled (through autotune), determines the CPU budget to use. Values greater than the number of schedulable CPU cores are allowed but may result in CPU contention. If None, defaults to the number of schedulable CPU cores.
enabled Whether to automatically tune performance knobs. If None, defaults to True.
initial_parallelism The initial parallelism to use for parallel transformations before autotune has a chance to run. A higher value can help with quick startup, but may cause the ram_budget to temporarily be exceeded. Memory-sensitive datasets should consider setting this to 1 to avoid running out of memory. Defaults to 16.
min_parallelism When true, .map(num_parallel_calls=AUTOTUNE) and .batch(num_parallel_calls=AUTOTUNE) will be at least parallelized by min_parallelism threads.
ram_budget When autotuning is enabled (through autotune), determines the RAM budget to use. Values greater than the available RAM in bytes may result in OOM. If None, defaults to half of the available RAM in bytes.

Methods

__eq__

View source

__eq__(
    other
)

Return self==value.

__ne__

View source

__ne__(
    other
)

Return self!=value.