/src/multi-app/ReadMe.txt
http://github.com/beechung/Latent-Factor-Models · Plain Text · 24 lines · 16 code · 8 blank · 0 comment · 0 complexity · 89cec9f2773425fe37865bf3ffb31b0f MD5 · raw file
- This directory contains code for fitting the localized factor model described
- in the following paper:
- Deepak Agarwal, Bee-Chung Chen, Bo Long. Localized factor models for
- multi-context recommendation. KDD 2011.
- This model jointly factorizes multiple matrices to provide transfer learning
- across multiple contexts or applications.
- (0) See Notation.txt for the model first.
- (1) To compile, just type make in this directory.
- see Makefile for details.
-
- (2) Some C and R functions used are in
- ../R/utils.{c,h,hpp,R}
- (3) Please DO NOT check in .o and .so files into the repository
- Directories:
- C/ C and C++ code
- R/example Examples of how to run different functions
- R/model Code for modeling