tf.raw_ops.TFRecordDatasetV2

Creates a dataset that emits the records from one or more TFRecord files.

tf.raw_ops.TFRecordDatasetV2(
    filenames,
    compression_type,
    buffer_size,
    byte_offsets,
    metadata='',
    name=None
)

filenames A Tensor of type string. A scalar or vector containing the name(s) of the file(s) to be read.
compression_type A Tensor of type string. A scalar containing either (i) the empty string (no compression), (ii) "ZLIB", or (iii) "GZIP".
buffer_size A Tensor of type int64. A scalar representing the number of bytes to buffer. A value of 0 means no buffering will be performed.
byte_offsets A Tensor of type int64. A scalar or vector containing the number of bytes for each file that will be skipped prior to reading.
metadata An optional string. Defaults to "".
name A name for the operation (optional).

A Tensor of type variant.