1 | .\"------------------------------------------------------------
|
---|
2 | .\" Id - set Rv,revision, and Dt, Date using rcs-Id tag.
|
---|
3 | .de Id
|
---|
4 | .ds Rv \\$3
|
---|
5 | .ds Dt \\$4
|
---|
6 | ..
|
---|
7 | .Id $Id: mg_weights_build.1 3745 2003-02-20 21:20:24Z mdewsnip $
|
---|
8 | .\"------------------------------------------------------------
|
---|
9 | .TH mg_weights_build 1 \*(Dt CITRI
|
---|
10 | .SH NAME
|
---|
11 | mg_weights_build \- builds the approximate weights file
|
---|
12 | .SH SYNOPSIS
|
---|
13 | .B mg_weights_build
|
---|
14 | [
|
---|
15 | .B \-h
|
---|
16 | ]
|
---|
17 | [
|
---|
18 | .B \-s
|
---|
19 | ]
|
---|
20 | [
|
---|
21 | .BI \-b " num"
|
---|
22 | ]
|
---|
23 | [
|
---|
24 | .BI \-d " directory"
|
---|
25 | ]
|
---|
26 | .if n .ti +9n
|
---|
27 | .BI \-f " name"
|
---|
28 | .SH DESCRIPTION
|
---|
29 | .B mg_weights_build
|
---|
30 | is the program that generates the approximate weights file and the
|
---|
31 | interleaved text index and exact weights file. To do this, it uses
|
---|
32 | the text index and either the exact weights file, or the inverted file
|
---|
33 | and stemmed dictionary.
|
---|
34 | .SH OPTIONS
|
---|
35 | Options may appear in any order.
|
---|
36 | .TP "\w'\fB\-d\fP \fIdirectory\fP'u+2n"
|
---|
37 | .B \-h
|
---|
38 | This displays a usage line on
|
---|
39 | .IR stderr .
|
---|
40 | .TP
|
---|
41 | .B \-s
|
---|
42 | Save the weights generated from the
|
---|
43 | .I *.invf
|
---|
44 | and
|
---|
45 | .I *.invf.dict
|
---|
46 | file in a
|
---|
47 | .I *.weight
|
---|
48 | file.
|
---|
49 | .TP
|
---|
50 | .BI \-b " num"
|
---|
51 | This specifies the number of bits of precision that should be used for
|
---|
52 | the approximate weights. If not specified, it defaults to 8.
|
---|
53 | .TP
|
---|
54 | .BI \-d " directory"
|
---|
55 | This specifies the directory where the document collection is.
|
---|
56 | .TP
|
---|
57 | .BI \-f " name"
|
---|
58 | This specifies the base name of the document collection.
|
---|
59 | .SH ENVIRONMENT
|
---|
60 | .TP "\w'\fBMGDATA\fP'u+2n"
|
---|
61 | .SB MGDATA
|
---|
62 | If this environment variable exists, then its value is used as the
|
---|
63 | default directory where the
|
---|
64 | .BR mg (1)
|
---|
65 | collection files are. If this variable does not exist, then the
|
---|
66 | directory \*(lq\fB.\fP\*(rq is used by default. The command line
|
---|
67 | option
|
---|
68 | .BI \-d " directory"
|
---|
69 | overrides the directory in
|
---|
70 | .BR MGDATA .
|
---|
71 | .SH FILES
|
---|
72 | .TP 20
|
---|
73 | .B *.invf
|
---|
74 | Inverted file.
|
---|
75 | .TP
|
---|
76 | .B *.invf.dict
|
---|
77 | Compressed stemmed dictionary.
|
---|
78 | .TP
|
---|
79 | .B *.weight
|
---|
80 | The exact weights file.
|
---|
81 | .TP
|
---|
82 | .B *.text.idx
|
---|
83 | Index into the compressed documents.
|
---|
84 | .TP
|
---|
85 | .B *.text.idx.wgt
|
---|
86 | Interleaved index into the compressed documents and document weights.
|
---|
87 | .TP
|
---|
88 | .B *.weight.approx
|
---|
89 | Approximate document weights.
|
---|
90 | .SH "SEE ALSO"
|
---|
91 | .na
|
---|
92 | .BR mg (1),
|
---|
93 | .BR mg_compression_dict (1),
|
---|
94 | .BR mg_fast_comp_dict (1),
|
---|
95 | .BR mg_get (1),
|
---|
96 | .BR mg_invf_dict (1),
|
---|
97 | .BR mg_invf_dump (1),
|
---|
98 | .BR mg_invf_rebuild (1),
|
---|
99 | .BR mg_passes (1),
|
---|
100 | .BR mg_perf_hash_build (1),
|
---|
101 | .BR mg_text_estimate (1),
|
---|
102 | .BR mgbilevel (1),
|
---|
103 | .BR mgbuild (1),
|
---|
104 | .BR mgdictlist (1),
|
---|
105 | .BR mgfelics (1),
|
---|
106 | .BR mgquery (1),
|
---|
107 | .BR mgstat (1),
|
---|
108 | .BR mgtic (1),
|
---|
109 | .BR mgticbuild (1),
|
---|
110 | .BR mgticdump (1),
|
---|
111 | .BR mgticprune (1),
|
---|
112 | .BR mgticstat (1).
|
---|