1mlpack_sparse_coding(1)          User Commands         mlpack_sparse_coding(1)
2
3
4

NAME

6       mlpack_sparse_coding - sparse coding
7

SYNOPSIS

9        mlpack_sparse_coding [-k int] [-i string] [-m unknown] [-l double] [-L double] [-n int] [-w double] [-N bool] [-o double] [-s int] [-T string] [-t string] [-V bool] [-c string] [-d string] [-M unknown] [-h -v]
10

DESCRIPTION

12       An  implementation  of  Sparse  Coding  with Dictionary Learning, which
13       achieves sparsity via an l1-norm regularizer on the codes (LASSO) or an
14       (l1+l2)-norm  regularizer on the codes (the Elastic Net). Given a dense
15       data matrix X with d dimensions and n points, sparse  coding  seeks  to
16       find  a  dense  dictionary matrix D with k atoms in d dimensions, and a
17       sparse coding matrix Z with n points in k dimensions.
18
19       The original data matrix X can then be reconstructed as Z *  D.  There‐
20       fore,  this  program  finds  a  representation  of each point in X as a
21       sparse linear combination of atoms in the dictionary D.
22
23       The sparse coding is found with an algorithm which alternates between a
24       dictionary  step,  which  updates the dictionary D, and a sparse coding
25       step, which updates the sparse coding matrix.
26
27       Once a dictionary D is found, the sparse coding model may  be  used  to
28       encode other matrices, and saved for future usage.
29
30       To  run this program, either an input matrix or an already-saved sparse
31       coding model must be specified. An input matrix may be  specified  with
32       the  ’--training_file  (-t)'  option, along with the number of atoms in
33       the dictionary (specified with the '--atoms  (-k)'  parameter).  It  is
34       also  possible  to  specify an initial dictionary for the optimization,
35       with the ’--initial_dictionary_file (-i)' parameter. An input model may
36       be specified with the '--input_model_file (-m)' parameter.
37
38       As an example, to build a sparse coding model on the dataset 'data.csv'
39       using 200 atoms and an l1-regularization parameter of 0.1,  saving  the
40       model into ’model.bin', use
41
42       $  sparse_coding  --training_file  data.csv  --atoms  200 --lambda1 0.1
43       --output_model_file model.bin
44
45       Then, this model could be used to encode a new matrix, 'otherdata.csv',
46       and save the output codes to 'codes.csv':
47
48       $  sparse_coding --input_model_file model.bin --test_file otherdata.csv
49       --codes_file codes.csv
50

OPTIONAL INPUT OPTIONS

52       --atoms (-k) [int]
53              Number of atoms in the dictionary. Default value 15.
54
55       --help (-h) [bool]
56              Default help info.
57
58       --info [string]
59              Get help on a specific module or option.  Default value ''.
60
61       --initial_dictionary_file (-i) [string]
62              Optional initial dictionary matrix. Default value ''.
63
64       --input_model_file (-m) [unknown]
65              File containing input sparse coding model.  Default value ''.
66
67       --lambda1 (-l) [double]
68              Sparse coding l1-norm regularization parameter.   Default  value
69              0.
70
71       --lambda2 (-L) [double]
72              Sparse  coding  l2-norm regularization parameter.  Default value
73              0.
74
75       --max_iterations (-n) [int]
76              Maximum number of iterations for sparse coding (0  indicates  no
77              limit). Default value 0.
78
79       --newton_tolerance (-w) [double]
80              Tolerance  for  convergence  of  Newton  method.   Default value
81              1e-06.
82
83       --normalize (-N) [bool]
84              If set, the input data matrix will be normalized before coding.
85
86       --objective_tolerance (-o) [double]
87              Tolerance for convergence of  the  objective  function.  Default
88              value 0.01.
89
90       --seed (-s) [int]
91              Random seed. If 0, 'std::time(NULL)' is used.  Default value 0.
92
93       --test_file (-T) [string]
94              Optional  matrix  to be encoded by trained model.  Default value
95              ''.
96
97       --training_file (-t) [string]
98              Matrix of training data (X). Default value ''.
99
100       --verbose (-v) [bool]
101              Display informational messages and the full list  of  parameters
102              and timers at the end of execution.
103
104       --version (-V) [bool]
105              Display the version of mlpack.
106

OPTIONAL OUTPUT OPTIONS

108       --codes_file (-c) [string]
109              Matrix  to  save  the  output  sparse  codes  of the test matrix
110              (--test_file) to. Default value ''.
111
112       --dictionary_file (-d) [string]
113              Matrix to save the output dictionary to.  Default value ''.
114
115       --output_model_file (-M) [unknown]
116              File to save trained sparse coding model to.  Default value ''.
117

ADDITIONAL INFORMATION

119       For further information, including relevant papers, citations, and the‐
120       ory,  consult  the  documentation  found  at  http://www.mlpack.org  or
121       included with your distribution of mlpack.
122
123
124
125mlpack-3.0.4                   21 February 2019        mlpack_sparse_coding(1)