New answers tagged

0

The key assumption you take when you are building a model is that the training features and the test features belong to the same distribution. Many algorithms are sensitive to subtle changes. For example, when Gradient Decent tries to find a local minima, it will be able to do so more easily if the data it is fed is normalized - with the mean and standard ...


0

This problem requires you to distinguish customer behavior . One customer’s purchase pattern may be different from another. I am assuming that you are collecting this data daily . Let’s assume that the data for each day is a vector In multi- dimensional space . you can calculate Mahalanobis distance between the vector for current day which will be of ...


Top 50 recent answers are included