[TOC]
A TFRecords file represents a sequence of (binary) strings. The format is not random access, so it is suitable for streaming large amounts of data but not suitable if fast sharding or other non-sequential access is desired.
A class to write records to a TFRecords file.
This class implements __enter__ and __exit__, and can be used
in with blocks like a normal file.
Opens file path and creates a TFRecordWriter writing to it.
path: The path to the TFRecords file.options: (optional) A TFRecordOptions object.
IOError: Ifpathcannot be opened for writing.
Write a string record to the file.
record: str
Close the file.
Enter a with block.
tf.python_io.TFRecordWriter.__exit__(unused_type, unused_value, unused_traceback) {#TFRecordWriter.exit}
Exit a with block, closing the file.
An iterator that read the records from a TFRecords file.
path: The path to the TFRecords file.options: (optional) A TFRecordOptions object.
Strings.
IOError: Ifpathcannot be opened for reading.
The type of compression for the record.
Options used for manipulating TFRecord files.
tf.python_io.TFRecordOptions.get_compression_type_string(cls, options) {#TFRecordOptions.get_compression_type_string}
A TFRecords file contains a sequence of strings with CRC hashes. Each record has the format
uint64 length
uint32 masked_crc32_of_length
byte data[length]
uint32 masked_crc32_of_data
and the records are concatenated together to produce the file. The CRC32s are described here, and the mask of a CRC is
masked_crc = ((crc >> 15) | (crc << 17)) + 0xa282ead8ul