Summary: LEMGA: Learning Models and Generic
Algorithms
Introduction with Examples
(Updated to v0.1)
Ling Li
Learning Systems Group, Caltech
January 22, 2003
What is Lemga?
Lemga is a C++ library consisting of
· Learning models/frames: neural networks, decision stumps, bagging,
boosting, . . .
· Generic optimization algorithms: gradient descent, conjugate-gradient,
. . .
· Auxiliary parts: data set, . . .
Features
· Uniform interface for learning models: initialize(), set train data(),
train(), > >, < <, create(), . . .
· Flexible training methods: different optimization methods can be applied
to a same object
· Easily extendable

Source: Abu-Mostafa, Yaser S. - Department of Mechanical Engineering & Computer Science Department, California Institute of Technology

Collections: Computer Technologies and Information Sciences