From the course: Machine Learning with Python: Logistic Regression
Join today to access over 24,900 courses taught by industry experts.
Making predictions with logistic regression - Python Tutorial
From the course: Machine Learning with Python: Logistic Regression
Making predictions with logistic regression
- [Narrator] Logistic regression is most often used to solve classification problems because that allows us to predict the probability of a categorical value Y for any given X. For example. Suppose that we have a logistic regression function like this that models the probability of a customer renting an e-bike as a function of their age. By doing the math, we can predict that the probability of a 35 year old customer renting an e-bike is 14.8%. If we use a 50% cutoff threshold then we interpret the prediction of 0.148 as no. According to this model, a 35 year old customer is unlikely to rent an e-bike. Statistically speaking, modeling the value of a variable with a restricted range, such probability can be challenging. It is also rather difficult to properly interpret the non-linear relationship that exists between predictors and the response as represented by the logistic function. So instead of using the logistic…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.
