SMALLbox

h1. FAQ

h2. Q1: What is SMALLbox?

*SMALLbox* is an evaluation framework for processing signals using adaptive sparse structured representations. SMALLbox is built within FP7 EU FET project called "SMALL" that is exploring new provably good methods to obtain inherently data-driven sparse models, which are able to cope with large-scale and complicated data. The main focus of research in the area of *sparse representations* is in developing reliable algorithms with provable performance and bounded complexity. There exist many applications for which it was proven beneficial to sparsely represent the data in some transform domain (i.e. "dictionary"). Moreover, the success of sparse models heavily depends on the choice of a “dictionary” to reflect the natural structures of a class of data. *Dictionary learning for sparse representation* deals with inferring such a dictionary from training data and is a key to the extension of sparse models for new exotic types of data.

SMALLbox provides an easy way to evaluate these methods against state-of-the art alternatives in a variety of standard signal processing problems. This is achieved trough a unifying interface that enables a seamless connection between the three types of modules: problems, dictionary learning algorithms and sparse solvers. In addition, it provides interoperability between existing state-of-the-art toolboxes. 

As an open source MATLAB toolbox, the SMALLbox can be seen as not only as a evaluation and educational tool, but as a tool for reproducible research in the sparse representations research community.

h2. Q2: How to obtain SMALLbox?

The SMALLbox project is maintained by people at the "Centre for Digital Music at SEECS, Queen Mary University of London":http://www.elec.qmul.ac.uk/digitalmusic/. To access the SMALLbox project page follow the link bellow:

https://code.soundsoftware.ac.uk/projects/smallbox/

If you want to try the latest stable public release please go to *Downloads* section. If you want to check the latest development you can obtain it from the project's code repository. It is hosted at soundsoftware.ac.uk, and is using Mercurial distributed version control, so you will need mercurial installed on your system.  If you are new to mercurial the easiest way to start is to install EasyMercurial, which you can find at https://code.soundsoftware.ac.uk/projects/easyhg.

To check out SMALLbox repository please hg clone the following URL, or provide this URL to your preferred Mercurial client (e.g. EasyMercurial):

https://code.soundsoftware.ac.uk/hg/smallbox

If you want to to contribute to the project then please first register to soundsoftware.ac.uk following the link in the upper right corner of the page.

h2. Q3: How to install SMALLbox?

To install the toolbox run the script *SMALLboxSetup.m* from the MATLAB command prompt and follow the instructions. *SmallboxSetup.m* is in the root SMALLbox directory. The SMALLbox installation involves the automatic download of several existing toolboxes. These are described in Q5. Due to the automatic download of toolboxes you must have an active internet connection.

Please note that within the toolboxes are several MEX components that must be compiled. If you do not already have MEX setup, run "mex -setup" or type "help mex" in the MATLAB command prompt.

Once installed, there are two optional demo functions that can be run. Further information can be found in the README.txt in the main SMALLbox directory.

For more detailed instructions regarding SMALLbox installation, please refer to the following [[InstallationGuide|page]].

h2. Q4: What are the Problem, solver, and DL structures in SMALLbox?

There are three main structures in SMALLbox that describe common parts of problem solving using sparse representation and dictionary learning - *Problem*, *DL* and *solver* structures.

The *Problem* structure defines all necessary aspects of a problem to be solved. To be compatible with the SPARCO, it needs to have five fields defined prior to any sparse representation of the data: *A* – a matrix or operator representing dictionary in which signal is sparse, *b* – a vector or matrix representing signal or signals to be represented, *reconstruct* – a function handle to reconstruct the signal from coefficients, *signalSize* – the dimension of the signal, *sizeA* – if matrix A is given as an operator the size of the dictionary needs to be defined in advance. Other fields that further describe the problem, which are useful for either reconstruction of the signal or representation of the results, might be generated by the SPARCO generateProblem function or the SMALLbox problem functions. The new problems implemented in the SMALLbox version 1.0 are: Image De-noising, Automatic Music Transcription and Image Representation using another image as a dictionary. In the case of a dictionary learning problem, fields *A* and *reconstruct* are not defined while generating the problem, but after the dictionary is learned and prior to the sparse representation. In this case, field *b* needs to be given in matrix form to represent the training data and another field *p* defining the number of dictionary elements to be learned needs to be specified.

The structure for dictionary learning - *DL* is a structure that defines dictionary learning algorithm to be used. It is initialised with a utility function *SMALL_init_DL*, which will define five mandatory fields:	*toolbox* - a field used to discriminate the API, *name* - the name of dictionary learning function from the particular toolbox,	*param* - a field containing parameters for the particular DL technique and in the form given by the toolbox API, *D* - a field where the learned dictionary will be stored, *time* - a field to store learning time. After *toolbox*, *name* and *param* fields are set, the function *SMALL_learn* is called with *Problem* and *DL* structures as inputs. According to the DL.toolbox field, the function calls the DL.name algorithm with its API and outputs learned dictionary D and time spent. The DL.param field contains parameters such as dictionary size, the number of iterations, the error goal or similar depending on the particular algorithm used. 

Similar to dictionary learning every instance of the sparse representation needs to be initialised with the *SMALL_init_solver* function. It will define mandatory fields of the *solver* structure: *toolbox* - a field with toolbox name (e.g. sparselab), *name* - the name of solver from the particular toolbox (e.g. SolveOMP), *param* - the parameters in the form given by the toolbox API, *solution* - the output representation, *reconstructed* - the signal reconstructed from solution, *time* - the time spent for sparse representation. With the input parameters of the solver structure set, the *SMALL_solve* function is called with *Problem* and *solver* structure as inputs. The function calls *solver.name* algorithm with API specified by *solver.toolbox* and outputs solution, reconstructed and time fields.

h2. Q5: What is included in SMALLbox?

To enable easy comparison with the existing state-of-the-art algorithms, during the installation procedure SMALLbox checks the Matlab path for existence of the following freely available toolboxes and will automatically download and install them, as required:

-	SPARCO (v.1.2) - set of sparse representation problems

-	SparseLab (v.2.1) - set of sparse solvers

-	Sparsify (v.0.4) - set of greedy and hard thresholding algorithms

-	SPGL1 (v.1.7) - large-scale sparse reconstruction solver

-	GPSR (v.6.0) - Gradient projection for sparse reconstruction

-	KSVD-box (v.13) and OMP-box (v.10) - dictionary learning

-	KSVDS-box (v.11) and OMPS-box (v.1) - sparse dictionary learning

In addition there are also implementations of three solvers in the *solver* directory (MP, OMP and PCGP) and our implementation of recursive least square dictionary learning algorithm (RLS-DLA) in *DL* directory. 

h2. Q6: How do I contribute?

If you want to contribute to the project then please first register to soundsoftware.ac.uk following the link in the upper right corner of the page. 

The code repository hosted at soundsoftware.ac.uk is using Mercurial distributed version control, so you will need mercurial installed on your system.  If you are new to mercurial the easiest way to start is to install EasyMercurial, which you can find at https://code.soundsoftware.ac.uk/projects/easyhg.

To check out SMALLbox repository please hg clone the following URL, or provide this URL to your preferred Mercurial client (e.g. EasyMercurial):

https://code.soundsoftware.ac.uk/hg/smallbox

There are three ways how you can contribute your code to SMALLbox:

a) I have a toolbox that I maintain myself and it is available at my repository, but I want SMALLbox users to be able to use it within SMALLbox.

> For example you have toolbox called *my_dummy_toolbox* and it is available at *my_dummy_url*. You should add the following lines to *SMALLboxSetup.m* script and of course change the parts in bold letters:

> > % check if toolbox is already installed (assuming that your toolbox have setup file, but any other script that is unique to your toolbox will do.

> >

> > if ~exist(' *my_dummy_toolbox_setup.m*','file')

> >

> > > fprintf('\n ******************************************************************');

> > >

> > > fprintf('\n\n Initialising *My_dummy_toolbox* Setup');

> > > 

> > > try

> > > > % setting up the path where toolbox will be installed

> > > >

> > > > *my_dummy_toolbox_path* =[SMALL_path,FS,'toolboxes',FS,' *my_dummy_toolbox* '];

> > > >

> > > > %setting up the url of the file to be downloaded

> > > >

> > > > *my_dummy_toolbox_zip*='http:// *my_dummy_url*/ *my_dummy_toolbox.zip*';

> > > >

> > > > fprintf('\n\n Downloading toolbox, please be patient\n\n');

> > > > 

> > > > unzip( *my_dummy_toolbox_zip*, *my_dummy_toolbox_path*);

> > > > 

> > > > % generate the path for the toolbox and add it to the MATLAB search path

> > > >

> > > > *my_dummy_toolbox_p*=genpath( *my_dummy_toolbox_path*);

> > > > 

> > > > addpath( *my_dummy_toolbox_p*);

> > > > 

> > > > % go to the installation directory and run the setup script if needed (e.g. there are mex files that needs to be compiled)

> > > > 

> > > > cd([ *my_dummy_toolbox_path*]);

> > > >

> > > > try

> > > >> *my_dummy_toolbox_setup.m*;

> > > >>

> > > >> fprintf('\n *My_dummy_toolbox* Installation Successful!\n');

> > > > catch

> > > >> warning('*My_dummy_toolbox* setup failed');

> > > > end

> > > catch

> > > > fprintf('\n *My_dummy_toolbox* Installation Failed\n');

> > > end

> > >

> > > % return to the SMALL root directory

> > >

> > > cd(SMALL_path);

> > else

> >> fprintf('\n ******************************************************************');

> > >

> >> fprintf('\n\n *My_dummy_toolbox* is already installed');

> > end

> 

> Once you made this changes, you should also follow the steps in Q7, Q8 and Q9 to integrate your API with *SMALLbox* API.

>

b) I have a toolbox that I would like to incorporate into SMALLbox and to make it maintained and developed through the SMALLbox project.

> Make a folder in *toolboxes* directory, add your files and commit to the repository. If your toolbox needs setup (e.g. mex files need to be compiled), add the lines in the *SMALLboxSetup.m* script  and of course change the parts in bold letters:

> > *my_dummy_toolbox_path* =[SMALL_path,FS,'toolboxes',FS,' *my_dummy_toolbox* '];

> >

> > cd([*my_dummy_toolbox_path*]);

> >

> > try

> > >

> > > *my_dummy_toolbox_setup.m*;

> > >

> > > fprintf('\n *My_dummy_toolbox* Installation Successful!\n');

> > >

> > catch

> > > warning('*My_dummy_toolbox* setup failed');

> > end

> >

> > cd(SMALL_path);

> 

> If setup is not needed for your toolbox, no changes are required and the SMALLbox setup script will automatically generate the path for your toolbox.

> Once you made this changes, you should also follow the steps in Q7, Q8 and Q9 to integrate your API with *SMALLbox* API.

c) I want to develop solver/DL algorithm/problem through the SMALLbox project. 

> If you want to develop your algorithms within the project and use SMALLbox API then put your scripts in dedicated folders ( *DL*, *solvers* and *Problems* ) and make sure you are using following API:

>

> SMALL solver API: 

>

> >*y = my_dummy_solver(A, b, n, params, AT );*

> >

> >where *y* is the solution, *A* is the dictionary to be used, *b* is signal or matrix of signals to be represented, n is size of the signal, *params* are the parameters needed and *AT* is the dictionary transpose if it is given in implicit form.

> 

> SMALL dictionary learning API: 

> >

> >*D = my_dummy_DL(X, param);*

> >

> >where *D* is the learned dictionary, *X* is the matrix with training vectors as its columns and param is a structure with parameters that you algorithm is using. 

>

> SMALL problem API: 

> >

> >*Problem=my_dummy_problem(varargin);*

> >

> >where *Problem* is the structure with fields as explained in *Q4*.

h2. Q7: I want to add my solver API to SMALLbox. How?

To introduce a new sparse representation algorithm to the SMALLbox, the file containing the code for the algorithm needs to be put into the MATLAB path (follow the steps in *Q6*). For example, one has a function called *my_dummy_solver*, to be used in the SMALLbox, with the following API call:

> y= *my_dummy_solver* (size_y, dictionary, signal, error_goal, iter_num);

A name needs to be defined for your toolbox in order to differentiate your API from other toolboxes. Using the example name *my_dummy_toolbox* the following line needs to be inserted to the if statement in the *SMALL_solve.m* script:

> elseif strcmpi(solver.toolbox,' *my_dummy_toolbox* ')

>

>>y =eval([solver.name,'(n,A,b,',solver.param,');']);

To test the function, the *SMALL_solver_test.m* script from the *examples* directory can be modified as follows:

>

>SMALL.Problem = generateProblem(6, 'P', 6, 'm', 270,'n',1024, 'show');

>

>i=1;

>

>SMALL.solver(i)=SMALL_init_solver;

>

>SMALL.solver(i).toolbox=' *my_dummy_toolbox* ';

>

>SMALL.solver(i).name=' *my_dummy_solver* ';

>

> % In the following string all parameters except matrix, measurement vector and size of solution need to be specified. If you are not sure which parameters are needed for particular solver type "help <Solver name>" in MATLAB command line

>

> SMALL.solver(i).param='1e-14, 200';

>

> SMALL.solver(i)=SMALL_solve(SMALL.Problem, SMALL.solver(i));

h2. Q8: I want to add my dictionary learning algorithm API to SMALLbox. How?

To introduce a new dictionarylearning algorithm to the SMALLbox, the file containing the code for the algorithm needs to be put into the MATLAB path (follow the steps in *Q6*). For example, one has a function called *my_dummy_DL*, to be used in the SMALLbox, with the following API call:

> D= *my_dummy_DL* (TrainingMatrix, params);

>

>where *D* is learned dictionary, *TrainingMatrix' is matrix that has training signals as its columns and *params* is structure with parameters needed for learning (e.g. initial dictionary, stopping criteria etc.)

A name needs to be defined for your toolbox in order to differentiate your API from other toolboxes. Using the example name *my_dummy_toolbox* the following line needs to be inserted to the if statement in the *SMALL_learn.m* script:

> elseif strcmpi(solver.toolbox,' *my_dummy_toolbox* ')

>

>>D =eval([DL.name,'(X,param);']);

You can then try to modify any of Image denoising example scripts to test your algorithm against the ones provided.

h2. Q9: I want to add a new problem. How? 

Assuming that your script that generates the problem is in the path (follow the steps in *Q6*), you just need to make sure that it generates *Problem* sturctures with the fileds explained in *Q4*. If you want to reconstruct the signal from the solution then you should also provide the recontruction function (follow the examples provided in the SMALLbox documentation and also in the *Problems* directory).