sample-loggers 0.0.5

This package provides sample loggers.

Deep Learning Tools for Pytorch

This package provides some common configurations to quickly create your own logger through the logging library.

Installation

pip install sample_loggers

Example for using

To ues the logger directly, you can use the logger object provided by the package.

import os
from sample_loggers import logger

# you can set the log dir and log file name through environment variables.
os.environ["LOG_DIR"] = "./output/log"
os.environ["LOG_FILE_NAME"] = "log.txt"

logger.info("hello world")

To define your own logger, you can use the create_logger method. This method allows you to specify the name of the logger and the log level.

from sample_loggers import create_logger, LEVEL

# The format can be defined as a string, 
# you can refer to the official website 
# https://docs.python.org/3/library/logging.html#logrecord-attributes.
define_format = "Define your own format here, you can use the default format or define your own format."

logger = create_logger(
    logger_name="logger",
    log_level=LEVEL.INFO,
    log_file_name="log.txt",
    output_dir="./output/log",
    format_type="default",
    define_format=define_format
)

Update

• 0.0.5 - improve the default logger.
• 0.0.4post1 - Fix the bug.
• 0.0.4 - Samplify thr method to create logger and provide the default logger that can be used directly.
• 0.0.3post1 - Fix the bug.
• 0.0.3 - Add the level definition.
• 0.0.2post2 - Fix the bug.
• 0.0.2post1 - Fix the bug.
• 0.0.2 - Add the summery and throughput method.
• 0.0.1 - We provide the common config for logging to create logger.

License

Sample Loggers is MIT licensed. See the LICENSE for details.