Mercurial > hg > smallbox
comparison DL/two-step DL/SMALL_two_step_DL.m @ 195:d50f5bdbe14c luisf_dev
- Added SMALL_DL_test: simple DL showcase
- Added dico_decorr_symmetric: improved version of INK-SVD decorrelation step
- Debugged SMALL_learn, SMALLBoxInit and SMALL_two_step_DL
author | Daniele Barchiesi <daniele.barchiesi@eecs.qmul.ac.uk> |
---|---|
date | Wed, 14 Mar 2012 14:42:52 +0000 |
parents | 759313488e7b |
children | 83af80baf959 |
comparison
equal
deleted
inserted
replaced
190:759313488e7b | 195:d50f5bdbe14c |
---|---|
96 | 96 |
97 % This is a small patch that needs to be resolved in dictionary learning we | 97 % This is a small patch that needs to be resolved in dictionary learning we |
98 % want sparse representation of training set, and in Problem.b1 in this | 98 % want sparse representation of training set, and in Problem.b1 in this |
99 % version of software we store the signal that needs to be represented | 99 % version of software we store the signal that needs to be represented |
100 % (for example the whole image) | 100 % (for example the whole image) |
101 global SMALL_path | |
101 | 102 |
102 tmpTraining = Problem.b1; | 103 tmpTraining = Problem.b1; |
103 Problem.b1 = sig; | 104 Problem.b1 = sig; |
104 if isfield(Problem,'reconstruct') | 105 if isfield(Problem,'reconstruct') |
105 Problem = rmfield(Problem, 'reconstruct'); | 106 Problem = rmfield(Problem, 'reconstruct'); |