LCSL / GURLS

GURLS: a Least Squares Library for Supervised Learning
http://lcsl.mit.edu/#/downloads/gurls
63 stars 37 forks source link

Recursive Cholesky functions (Init+update) added. Utility function added... #17

Closed raffaello-camoriano closed 9 years ago

raffaello-camoriano commented 9 years ago

The following GURLS functions have been added:

These functions have been successfully tested in the experiments available at: https://github.com/Ommac/gurlsRRLSmatlab/tree/master/experiments

The following plot, produced by experiment 5 (https://github.com/Ommac/gurlsRRLSmatlab/blob/master/experiments/5_RRLS_bench_icubdyn_RF_nMSE/RRLS_bench_icubdyn_RF_nMSE.m), shows a comparison between the performance (nMSE) of a batch estimator and a recursive one using the proposed functions for performing 100 predictions and updates.

res_pullreq

An utility script (clearAllButBP.m) for cleaning the workspace, keeping breakpoints unaltered, has alse been added to gurls/utils.