Mercurial > hg > smallbox
comparison util/SMALL_learn.m @ 8:33850553b702
(none)
author | idamnjanovic |
---|---|
date | Mon, 22 Mar 2010 10:56:54 +0000 (2010-03-22) |
parents | |
children | fc395272d53e |
comparison
equal
deleted
inserted
replaced
7:0151f1ea080d | 8:33850553b702 |
---|---|
1 function DL = SMALL_learn(Problem,DL) | |
2 %%% SMALL Dictionary Learning | |
3 % Ivan Damnjanovic 2009 | |
4 % Function gets as input Problem and Dictionary Learning (DL) structures | |
5 % In Problem structure field b with the training set needs to be defined | |
6 % In DL fields with name of the toolbox and solver, and parameters file | |
7 % for particular dictionary learning technique needs to be present. | |
8 % | |
9 % Outputs are Learned dictionary and time spent as a part of DL structure | |
10 %% | |
11 | |
12 fprintf('\nStarting Dictionary Learning %s... \n', DL.name); | |
13 start=cputime; | |
14 | |
15 if strcmpi(DL.toolbox,'KSVD') | |
16 param=DL.param; | |
17 param.data=Problem.b; | |
18 | |
19 D = eval([DL.name,'(param, ''t'', 5);']); | |
20 elseif strcmpi(DL.toolbox,'KSVDS') | |
21 param=DL.param; | |
22 param.data=Problem.b; | |
23 | |
24 D = eval([DL.name,'(param, ''t'', 5);']); | |
25 elseif strcmpi(DL.toolbox,'SPAMS') | |
26 | |
27 X = Problem.b; | |
28 param=DL.param; | |
29 | |
30 D = eval([DL.name,'(X, param);']); | |
31 % As some versions of SPAMS does not produce unit norm column | |
32 % dictionaries, we need to make sure that columns are normalised to | |
33 % unit lenght. | |
34 | |
35 for i = 1: size(D,2) | |
36 D(:,i)=D(:,i)/norm(D(:,i)); | |
37 end | |
38 | |
39 % To introduce new dictionary learning technique put the files in | |
40 % your Matlab path. Next, unique name <TolboxID> for your toolbox needs | |
41 % to be defined and also prefferd API for toolbox functions <Preffered_API> | |
42 % | |
43 % elseif strcmpi(DL.toolbox,'<ToolboxID>') | |
44 % % This is an example of API that can be used: | |
45 % % - get training set from Problem part of structure | |
46 % % - assign parameters defined in the main program | |
47 % | |
48 % X = Problem.b; | |
49 % param=DL.param; | |
50 % | |
51 % % - Evaluate the function (DL.name - defined in the main) with | |
52 % % parameters given above | |
53 % | |
54 % D = eval([DL.name,'(<Preffered_API>);']); | |
55 | |
56 else | |
57 printf('\nToolbox has not been registered. Please change SMALL_learn file.\n'); | |
58 return | |
59 end | |
60 | |
61 %% | |
62 % Dictionary Learning time | |
63 | |
64 DL.time = cputime - start; | |
65 fprintf('\n%s finished task in %2f seconds. \n', DL.name, DL.time); | |
66 | |
67 % If dictionary is given as a sparse matrix change it to full | |
68 | |
69 DL.D = full(D); | |
70 | |
71 end | |
72 |