Peering into the Black Box
Before moving on in your machine learning journey, it’s important to know about one more type of optimization problem you’ll encounter in machine learning: hyperparameter search.
Hyperparameters are the details about an algorithm that aren’t directly learnable but affect the structure and outcome of the model training process. For example, the learning rate 1.0e-2 used in the previous section is a hyperparameter.
Hyperparameter search is concerned with finding hyperparameters that lead to the optimal performance of a model on an evaluation set. Hyperparameters can have a big impact on the evaluation of a model, so finding better hyperparameters can drastically improve model performance. Unfortunately, it’s difficult ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access