From the course: Python: Working with Predictive Analytics

Unlock the full course today

Join today to access over 22,500 courses taught by industry experts or purchase this course individually.

Hyperparameter optimization

Hyperparameter optimization - Python Tutorial

From the course: Python: Working with Predictive Analytics

Start my 1-month free trial

Hyperparameter optimization

“

- [Narrator] We came a long way from preparing our data, to working with the predictive models. In most cases this an iterative approach of evaluation and modeling until we reach a satisfactory result. We can find the best parameters by trial and error, but it will take a very long time. Imagine trying and evaluating a model on a very large data set by changing one parameter at a time. It would work, only if we had unlimited time. This is not the most time efficient option. Instead I want to introduce you to a new term, hyperparameter optimization. This is the technique of identifying an ideal set of parameters for a prediction algorithm, which provides the optimum performance. The algorithm learns which parameter provides us with better performance by iteratively working on a pre-defined set of parameters. For example, one method traditionally used is grid search. Let's say we have parameter A and parameter B. Sometimes we may have more or less depending on the learning model. We…

Contents