Accelerate

Logging with Accelerate

Accelerate has its own logging utility to handle logging while in a distributed system. To utilize this replace cases of logging with accelerate.logging:

- import logging
+ from accelerate.logging import get_logger
- logger = logging.getLogger(__name__)
+ logger = get_logger(__name__)

accelerate.logging.get_logger

< source >

( name: str )

Parameters

name (str) — The name for the logger, such as __file__

Returns a logging.Logger for name that can handle multiprocessing.

If a log should be called on all processes, pass main_process_only=False

E.g.

logger.info("My log", main_process_only=False)
logger.debug("My log", main_process_only=False)