datadings.commands.bench module

usage: bench.py [-h] [-r] [-s] [-b BUFFERING] [–separator SEPARATOR]: [–root-dir ROOT_DIR] [–include INCLUDE [INCLUDE …]] [–exclude EXCLUDE [EXCLUDE …]] infile

Run a read benchmark on a given dataset.

Support reading from msgpack files or directory trees.

Note

For directory trees, please refer to the DirectoryReader documentation for details on how to specify the dataset structure.

Positional arguments

infile Input file.

options:

-h, --help: show this help message and exit
-r, --raw: Do not decode samples. This is representative of performance in a multi-process environment, where workers take over decoding.
-s, --shuffle: Shuffle the reader.
-b BUFFERING, --buffering BUFFERING: MsgpackReader only: Buffer size of reader.
--separator SEPARATOR: DirectoryReader only: separator for file input mode.
--root-dir ROOT_DIR: DirectoryReader only: root directory for file input mode.