tf.raw_ops.SnapshotDatasetV2

Creates a dataset that will write to / read from a snapshot.

tf.raw_ops.SnapshotDatasetV2(
    input_dataset,
    path,
    reader_func_other_args,
    shard_func_other_args,
    output_types,
    output_shapes,
    reader_func,
    shard_func,
    compression='',
    reader_prefix='',
    writer_prefix='',
    hash_valid=False,
    hash=0,
    metadata='',
    name=None
)

This dataset attempts to determine whether a valid snapshot exists at the snapshot_path, and reads from the snapshot in lieu of using input_dataset. If not, it will run the preprocessing pipeline as usual, and write out a snapshot of the data processed for future use.

input_datasetA Tensor of type variant. A variant tensor representing the input dataset.
pathA Tensor of type string. The path we should write snapshots to / read snapshots from.
reader_func_other_argsA list of Tensor objects.
shard_func_other_argsA list of Tensor objects.
output_typesA list of tf.DTypes that has length >= 1.
output_shapesA list of shapes (each a tf.TensorShape or list of ints) that has length >= 1.
reader_funcA function decorated with @Defun. Optional. A function to control how to read data from snapshot shards.
shard_funcA function decorated with @Defun. Optional. A function to control how to shard data when writing a snapshot.
compressionAn optional string. Defaults to "". The type of compression to be applied to the saved snapshot files.
reader_prefixAn optional string. Defaults to "".
writer_prefixAn optional string. Defaults to "".
hash_validAn optional bool. Defaults to False.
hashAn optional int. Defaults to 0.
metadataAn optional string. Defaults to "".
nameA name for the operation (optional).

A Tensor of type variant.