[FLINK-2013] Create generalized linear model framework

• Type: New Feature
• Status: Closed
• Priority: Major
• Resolution: Won't Do
• Component/s: Library / Machine Learning
• Labels: ML pull-request-available
• Assignee: Theodore Vasiloudis
• Reporter: Theodore Vasiloudis
• Created: 14/May/15 14:35
• Updated: 28/Feb/19 22:59
• Resolved: 28/Feb/19 22:57

Description

Generalized linear models (GLMs) provide an abstraction for many learning models that can be used for regression and classification tasks.

Some example GLMs are linear regression, logistic regression, LASSO and ridge regression.

The goal for this JIRA is to provide interfaces for the set of common properties and functions these models share.
In order to achieve that, a design pattern similar to the one that sklearn and MLlib employ will be used.

Issue Links

• incorporates: FLINK-1743 Add multinomial logistic regression to machine learning library
• incorporates: FLINK-2014 Add LASSO regression
• incorporates: FLINK-2015 Add ridge regression
• incorporates: FLINK-2016 Add elastic net regression
• incorporates: FLINK-1696 Add multiple linear regression to ML library
• incorporates: FLINK-1979 Implement Loss Functions
• is required by: FLINK-5525 Streaming Version of a Linear Regression model
• relates to: FLINK-1729 Assess performance of classification algorithms
• relates to: FLINK-1932 Add L-BFGS to the optimization framework
• links to: GitHub Pull Request #3756
• links to: GitHub Pull Request #3756