课程: Python: Working with Predictive Analytics

免费学习该课程！

今天就开通帐号，24,700 门业界名师课程任您挑！

Evaluation of predictive models

Evaluation of predictive models - Python教程

课程: Python: Working with Predictive Analytics

开始一个月试用 购买团队方案

Evaluation of predictive models

“

- [Instructor] Now that you've seen how to build a few regression models, we are moving on to the evaluation section of our roadmap. I'm going to summarize the strengths and weaknesses of each model in this video. So far, we have used R-squared as a way of measuring the success scores of the regression models. Please keep in mind that this score by itself is not enough to make decisions. It's recommended to further visualize, combine it with domain knowledge, and do further tests before making a final judgment. Now, let's look at each model individually. Linear regression has an advantage when there is a linear relationship between the independent variables and dependent variable. However, we need to keep in mind that this may become a disadvantage when we do not have a linear relationship between the independent variables and the dependent variable. Polynomial regression can be a strong model when there is a non-linear relationship between the independent variables and dependent…

内容