Get startedGet started for free

Wrap-up

1. Congratulations!

Congrats! You made it through the course!

2. What you've learned in this course

In this course, you learned what hyperparameters are, how they are different from model parameters and why it is useful to tune them in your machine learning models. You also learned HOW to tune hyperparameters with the R packages caret mlr and h2o.

3. Terms you can understand and apply

Specifically, you learned what the following terms mean and how to apply them: - Cartesian grid search for tuning all possible combinations of a hyperparameter set - Random Search for tuning with random sampling from a hyperparameter set - Adaptive Resampling for more efficient tuning of a hyperparameter search space - and Automatic Machine Learning for quick and easy model and hyperparameter optimization. - You know how to evaluate tuning results by examining and visualizing performance metrics - and how to use performance metrics in early stopping

4. How you can use this knowledge

With these techniques, you should now be able to find the best (or most optimal) hyperparameter combinations for your own models. You got to know three R packages for machine learning and hyperparameter tuning, so you will now have a feeling for how they work and how they differ. Maybe you even found a favorite, which you prefer to work with from now on. So, where do you start, if you want to go further? For additional information and tutorials, read the package manuals and vignettes. And of course: try it out! Maybe you already have data of your own. Or look for example datasets at the UC Irvine Machine Learning Repository or Kaggle. Follow the links to get there.

5. Thank you and have fun!

With this, I want to thank you for completing this course! I hope you benefit from what you've learned and that you will have a lot of fun trying things out! Bye!