.\" Text automatically generated by txt2man
.TH mlpack_nmf 1 "11 January 2024" "mlpack-4.3.0" "User Commands"
.SH NAME
\fBmlpack_nmf \fP- non-negative matrix factorization
.SH SYNOPSIS
.nf
.fam C
 \fBmlpack_nmf\fP \fB-i\fP \fIunknown\fP \fB-r\fP \fIint\fP [\fB-q\fP \fIunknown\fP] [\fB-p\fP \fIunknown\fP] [\fB-m\fP \fIint\fP] [\fB-e\fP \fIdouble\fP] [\fB-s\fP \fIint\fP] [\fB-u\fP \fIstring\fP] [\fB-V\fP \fIbool\fP] [\fB-H\fP \fIunknown\fP] [\fB-W\fP \fIunknown\fP] [\fB-h\fP \fB-v\fP] 
.fam T
.fi
.fam T
.fi
.SH DESCRIPTION


This program performs non-negative matrix factorization on the given dataset,
storing the resulting decomposed matrices in the specified files. For an
input dataset V, NMF decomposes V into two matrices W and H such that 
.PP
V = W * H
.PP
where all elements in W and H are non-negative. If V is of size (n x m), then
W will be of size (n x r) and H will be of size (r x m), where r is the rank
of the factorization (specified by the '\fB--rank\fP (\fB-r\fP)' parameter).
.PP
Optionally, the desired update rules for each NMF iteration can be chosen from
the following list:
.RS
.IP \(bu 3
multdist: multiplicative distance-based update rules (Lee and Seung 1999)
.IP \(bu 3
multdiv: multiplicative divergence-based update rules (Lee and Seung 1999)
.IP \(bu 3
als: alternating least squares update rules (Paatero and Tapper 1994)
.RE
.PP
The maximum number of iterations is specified with '\fB--max_iterations\fP (\fB-m\fP)',
and the minimum residue required for algorithm termination is specified with
the '\fB--min_residue\fP (\fB-e\fP)' parameter.
.PP
For example, to run NMF on the input matrix 'V.csv' using the 'multdist'
update rules with a rank-10 decomposition and storing the decomposed matrices
into 'W.csv' and 'H.csv', the following command could be used: 
.PP
$ \fBmlpack_nmf\fP \fB--input_file\fP V.csv \fB--w_file\fP W.csv \fB--h_file\fP H.csv \fB--rank\fP 10
\fB--update_rules\fP multdist
.RE
.PP

.SH REQUIRED INPUT OPTIONS 

.TP
.B
\fB--input_file\fP (\fB-i\fP) [\fIunknown\fP]
Input dataset to perform NMF on. 
.TP
.B
\fB--rank\fP (\fB-r\fP) [\fIint\fP]
Rank of the factorization.  
.SH OPTIONAL INPUT OPTIONS 

.TP
.B
\fB--help\fP (\fB-h\fP) [\fIbool\fP]
Default help info. 
.TP
.B
\fB--info\fP [\fIstring\fP]
Print help on a specific option. Default value ''. 
.TP
.B
\fB--initial_h_file\fP (\fB-q\fP) [\fIunknown\fP]
Initial H matrix. 
.TP
.B
\fB--initial_w_file\fP (\fB-p\fP) [\fIunknown\fP]
Initial W matrix. 
.TP
.B
\fB--max_iterations\fP (\fB-m\fP) [\fIint\fP]
Number of iterations before NMF terminates (0 runs until convergence. Default value 10000. 
.TP
.B
\fB--min_residue\fP (\fB-e\fP) [\fIdouble\fP]
The minimum root mean square residue allowed for each iteration, below which the program terminates. Default value 1e-05. 
.TP
.B
\fB--seed\fP (\fB-s\fP) [\fIint\fP]
Random seed. If 0, 'std::time(NULL)' is used.  Default value 0. 
\fB--update_rules\fP (\fB-u\fP) [\fIstring\fP] Update rules for each iteration; ( multdist | multdiv | als ). Default value 'multdist'. 
.TP
.B
\fB--verbose\fP (\fB-v\fP) [\fIbool\fP]
Display informational messages and the full list of parameters and timers at the end of execution. 
.TP
.B
\fB--version\fP (\fB-V\fP) [\fIbool\fP]
Display the version of mlpack.  
.SH OPTIONAL OUTPUT OPTIONS 

.TP
.B
\fB--h_file\fP (\fB-H\fP) [\fIunknown\fP]
Matrix to save the calculated H to. 
.TP
.B
\fB--w_file\fP (\fB-W\fP) [\fIunknown\fP]
Matrix to save the calculated W to.
.SH ADDITIONAL INFORMATION

For further information, including relevant papers, citations, and theory,
consult the documentation found at http://www.mlpack.org or included with your
distribution of mlpack.