A tool for sending logs via email at regular intervals during model training, to conveniently check the progress and status of the model training process.
The third-party libraries required for this tool are as follows:
pip install easydict
pip install apscheduler
pip install pyyamlPlace the kkreporter module in your project directory, and then import the KukaReporter class using the following import statement:
from kkreporter.reporter import KukaReporter# The config_path is the file path of the configuration file for the KukaReporter class.
# The configuration file is located at kkreporter/kuka_reporter_config.yaml within
# the project directory.
kkreporter = KukaReporter(config_path=[your config path], log_path=[your log path])- Send log reports on a scheduled basis
kkreporter.recurring_report()After executing this statement, the tool will send email reports on the training progress on a scheduled basis.
- Final Training Report
Execute after training completion
kkreporter.ending_report()