How to Run Machine Learning Experiments with Python Logging Module

Image

Running machine learning experiments involves a lot of tasks such as trying different algorithms to find the best one for a specific problem you want to solve (supervised, unsupervised or reinforcement), fining tuning your model to get better performance(grid search or Bayesian optimization), handling overfitting and underfitting on model performance and other related tasks that depend on the problem you are solving.

During the process of running a machine learning experiment, you can use the print statement to observe experiments output such as a list of algorithms and their performance (e.g accuracy score), best parameters that can improve your model performance, etc. When running ML experiments time to time print is not useful because when you run your next experiment, you lost all the outputs from the previous experiment unless you record them manually in the document. This becomes difficult to compare your ML experiment results because you need to have records of all experiments you have run so far. A python logging module can help you solve this problem by tracking events that happen when running machine learning experiments.

In this article, you will learn

  • what is Logging?
  • Why logging is important.
  • Applications of logging in different technology industries.
  • How to create your own python logger.
  • How to use python logger when running Machine learning experiments.

Read More on Medium