- Generalized Linear Models
- Generalized linear models unify various statistical models such as linear and logistic regression through the specification of a model family and link function. In R, such models can be fitted by passing an R model formula, family, and training dataset to the glm() function. Spark 1.5 extends glm() to operate over Spark DataFrames, which are distributed data collections managed by Spark. We also support elastic-net regularization for these models, the same as in R’s glmnet package.