complearn 0.6.2
Sign up to get free protection for your applications and to get access to all the features.
- data/AUTHORS +13 -0
- data/COPYING +340 -0
- data/ChangeLog +0 -0
- data/INSTALL +231 -0
- data/Makefile +352 -0
- data/Makefile.am +76 -0
- data/Makefile.in +352 -0
- data/NEWS +7 -0
- data/README +0 -0
- data/aclocal.m4 +104 -0
- data/bin/Makefile +209 -0
- data/bin/Makefile.am +8 -0
- data/bin/Makefile.in +209 -0
- data/bin/labeltree +68 -0
- data/bin/labeltree.in +68 -0
- data/bin/makesvm +70 -0
- data/bin/makesvm.in +70 -0
- data/bin/maketree +98 -0
- data/bin/maketree.in +98 -0
- data/bin/ncd +43 -0
- data/bin/ncd.in +43 -0
- data/bin/ncdmatrix +54 -0
- data/bin/ncdmatrix.in +54 -0
- data/bin/ncdvector +50 -0
- data/bin/ncdvector.in +50 -0
- data/complearn-0.6.2.gem +0 -0
- data/complearn.gemspec +57 -0
- data/config.log +597 -0
- data/config.status +1082 -0
- data/configure +4922 -0
- data/configure.ac +91 -0
- data/confstat5FpLBf/config.h +65 -0
- data/confstat5FpLBf/subs-1.sed +50 -0
- data/confstat5FpLBf/subs-2.sed +13 -0
- data/confstat5FpLBf/subs.frag +0 -0
- data/confstat5FpLBf/subs.sed +59 -0
- data/confstat5FpLBf/undefs.sed +24 -0
- data/doc/FAQ.txt +67 -0
- data/doc/Makefile +286 -0
- data/doc/Makefile.am +11 -0
- data/doc/Makefile.in +286 -0
- data/doc/devguide.txt +15 -0
- data/doc/example.complearnrc +14 -0
- data/doc/examples.txt +35 -0
- data/doc/man/Makefile +255 -0
- data/doc/man/Makefile.am +11 -0
- data/doc/man/Makefile.in +255 -0
- data/doc/man/complearn.5 +91 -0
- data/doc/man/labeltree.1 +35 -0
- data/doc/man/makesvm.1 +60 -0
- data/doc/man/maketree.1 +58 -0
- data/doc/man/ncd.1 +51 -0
- data/doc/man/ncdmatrix.1 +40 -0
- data/doc/man/ncdvector.1 +42 -0
- data/doc/readme.txt +101 -0
- data/doc/userguide.txt +46 -0
- data/examples/genes/blueWhale.txt +1 -0
- data/examples/genes/cat.txt +1 -0
- data/examples/genes/chimpanzee.txt +1 -0
- data/examples/genes/finWhale.txt +1 -0
- data/examples/genes/graySeal.txt +1 -0
- data/examples/genes/harborSeal.txt +1 -0
- data/examples/genes/horse.txt +1 -0
- data/examples/genes/human.txt +1 -0
- data/examples/genes/mouse.txt +1 -0
- data/examples/genes/rat.txt +1 -0
- data/ext/Makefile +167 -0
- data/ext/Quartet.c +399 -0
- data/ext/Quartet.h +62 -0
- data/ext/TreeScore.c +244 -0
- data/ext/TreeScore.h +3 -0
- data/ext/config.h +65 -0
- data/ext/config.h.in +64 -0
- data/ext/extconf.rb +3 -0
- data/ext/lib/CompLearnLib/CLConfig.rb +241 -0
- data/ext/lib/CompLearnLib/CompressionObject.rb +59 -0
- data/ext/lib/CompLearnLib/CompressionTask.rb +99 -0
- data/ext/lib/CompLearnLib/DistMatrix.rb +18 -0
- data/ext/lib/CompLearnLib/FoundComp.rb +10 -0
- data/ext/lib/CompLearnLib/FoundComp.rb.in +10 -0
- data/ext/lib/CompLearnLib/Ncd.rb +248 -0
- data/ext/lib/CompLearnLib/RunEnv.rb +150 -0
- data/ext/lib/CompLearnLib/Task.rb +39 -0
- data/ext/lib/CompLearnLib/TaskMaster.rb +13 -0
- data/ext/lib/CompLearnLib/TaskMasterMPI.rb +112 -0
- data/ext/lib/CompLearnLib/TaskMasterSingle.rb +39 -0
- data/ext/lib/CompLearnLib/Tree.rb +300 -0
- data/install-sh +294 -0
- data/missing +336 -0
- data/mkinstalldirs +111 -0
- data/o +24 -0
- data/scripts/CompLearn.iss +89 -0
- data/scripts/CompLearn.iss.in +89 -0
- data/scripts/debian/changelog +6 -0
- data/scripts/debian/control +14 -0
- data/scripts/makeSetup.sh +23 -0
- data/scripts/makeSetup.sh.in +23 -0
- data/scripts/makedeb.zsh +46 -0
- data/scripts/makedeb.zsh.in +46 -0
- data/tests/alltests.rb +2 -0
- data/tests/bz2test.rb +516 -0
- data/tests/sshagent-test.rb +48 -0
- data/tests/tests.rb +275 -0
- metadata +164 -0
data/doc/man/Makefile.am
ADDED
data/doc/man/Makefile.in
ADDED
@@ -0,0 +1,255 @@
|
|
1
|
+
# Makefile.in generated automatically by automake 1.4-p4 from Makefile.am
|
2
|
+
|
3
|
+
# Copyright (C) 1994, 1995-8, 1999 Free Software Foundation, Inc.
|
4
|
+
# This Makefile.in is free software; the Free Software Foundation
|
5
|
+
# gives unlimited permission to copy and/or distribute it,
|
6
|
+
# with or without modifications, as long as this notice is preserved.
|
7
|
+
|
8
|
+
# This program is distributed in the hope that it will be useful,
|
9
|
+
# but WITHOUT ANY WARRANTY, to the extent permitted by law; without
|
10
|
+
# even the implied warranty of MERCHANTABILITY or FITNESS FOR A
|
11
|
+
# PARTICULAR PURPOSE.
|
12
|
+
|
13
|
+
|
14
|
+
SHELL = @SHELL@
|
15
|
+
|
16
|
+
srcdir = @srcdir@
|
17
|
+
top_srcdir = @top_srcdir@
|
18
|
+
VPATH = @srcdir@
|
19
|
+
prefix = @prefix@
|
20
|
+
exec_prefix = @exec_prefix@
|
21
|
+
|
22
|
+
bindir = @bindir@
|
23
|
+
sbindir = @sbindir@
|
24
|
+
libexecdir = @libexecdir@
|
25
|
+
datadir = @datadir@
|
26
|
+
sysconfdir = @sysconfdir@
|
27
|
+
sharedstatedir = @sharedstatedir@
|
28
|
+
localstatedir = @localstatedir@
|
29
|
+
libdir = @libdir@
|
30
|
+
infodir = @infodir@
|
31
|
+
mandir = @mandir@
|
32
|
+
includedir = @includedir@
|
33
|
+
oldincludedir = /usr/include
|
34
|
+
|
35
|
+
DESTDIR =
|
36
|
+
|
37
|
+
pkgdatadir = $(datadir)/@PACKAGE@
|
38
|
+
pkglibdir = $(libdir)/@PACKAGE@
|
39
|
+
pkgincludedir = $(includedir)/@PACKAGE@
|
40
|
+
|
41
|
+
top_builddir = ../..
|
42
|
+
|
43
|
+
ACLOCAL = @ACLOCAL@
|
44
|
+
AUTOCONF = @AUTOCONF@
|
45
|
+
AUTOMAKE = @AUTOMAKE@
|
46
|
+
AUTOHEADER = @AUTOHEADER@
|
47
|
+
|
48
|
+
INSTALL = @INSTALL@
|
49
|
+
INSTALL_PROGRAM = @INSTALL_PROGRAM@ $(AM_INSTALL_PROGRAM_FLAGS)
|
50
|
+
INSTALL_DATA = @INSTALL_DATA@
|
51
|
+
INSTALL_SCRIPT = @INSTALL_SCRIPT@
|
52
|
+
transform = @program_transform_name@
|
53
|
+
|
54
|
+
NORMAL_INSTALL = :
|
55
|
+
PRE_INSTALL = :
|
56
|
+
POST_INSTALL = :
|
57
|
+
NORMAL_UNINSTALL = :
|
58
|
+
PRE_UNINSTALL = :
|
59
|
+
POST_UNINSTALL = :
|
60
|
+
CC = @CC@
|
61
|
+
FOUNDBZIP2 = @FOUNDBZIP2@
|
62
|
+
FOUNDGZIP = @FOUNDGZIP@
|
63
|
+
MAKEINFO = @MAKEINFO@
|
64
|
+
PACKAGE = @PACKAGE@
|
65
|
+
RANLIB = @RANLIB@
|
66
|
+
RUBYBIN = @RUBYBIN@
|
67
|
+
SET_MAKE = @SET_MAKE@
|
68
|
+
TOPDIR = @TOPDIR@
|
69
|
+
VERSION = @VERSION@
|
70
|
+
WINBUILDDIR = @WINBUILDDIR@
|
71
|
+
|
72
|
+
man_MANS = complearn.5 labeltree.1 makesvm.1 maketree.1 ncd.1 ncdmatrix.1 ncdvector.1
|
73
|
+
|
74
|
+
|
75
|
+
EXTRA_DIST = $(man_MANS)
|
76
|
+
mkinstalldirs = $(SHELL) $(top_srcdir)/mkinstalldirs
|
77
|
+
CONFIG_CLEAN_FILES =
|
78
|
+
man1dir = $(mandir)/man1
|
79
|
+
man5dir = $(mandir)/man5
|
80
|
+
MANS = $(man_MANS)
|
81
|
+
|
82
|
+
NROFF = nroff
|
83
|
+
DIST_COMMON = Makefile.am Makefile.in
|
84
|
+
|
85
|
+
|
86
|
+
DISTFILES = $(DIST_COMMON) $(SOURCES) $(HEADERS) $(TEXINFOS) $(EXTRA_DIST)
|
87
|
+
|
88
|
+
TAR = tar
|
89
|
+
GZIP_ENV = --best
|
90
|
+
all: all-redirect
|
91
|
+
.SUFFIXES:
|
92
|
+
$(srcdir)/Makefile.in: Makefile.am $(top_srcdir)/configure.ac $(ACLOCAL_M4)
|
93
|
+
cd $(top_srcdir) && $(AUTOMAKE) --gnu --include-deps doc/man/Makefile
|
94
|
+
|
95
|
+
Makefile: $(srcdir)/Makefile.in $(top_builddir)/config.status
|
96
|
+
cd $(top_builddir) \
|
97
|
+
&& CONFIG_FILES=$(subdir)/$@ CONFIG_HEADERS= $(SHELL) ./config.status
|
98
|
+
|
99
|
+
|
100
|
+
install-man1:
|
101
|
+
$(mkinstalldirs) $(DESTDIR)$(man1dir)
|
102
|
+
@list='$(man1_MANS)'; \
|
103
|
+
l2='$(man_MANS)'; for i in $$l2; do \
|
104
|
+
case "$$i" in \
|
105
|
+
*.1*) list="$$list $$i" ;; \
|
106
|
+
esac; \
|
107
|
+
done; \
|
108
|
+
for i in $$list; do \
|
109
|
+
if test -f $(srcdir)/$$i; then file=$(srcdir)/$$i; \
|
110
|
+
else file=$$i; fi; \
|
111
|
+
ext=`echo $$i | sed -e 's/^.*\\.//'`; \
|
112
|
+
inst=`echo $$i | sed -e 's/\\.[0-9a-z]*$$//'`; \
|
113
|
+
inst=`echo $$inst | sed '$(transform)'`.$$ext; \
|
114
|
+
echo " $(INSTALL_DATA) $$file $(DESTDIR)$(man1dir)/$$inst"; \
|
115
|
+
$(INSTALL_DATA) $$file $(DESTDIR)$(man1dir)/$$inst; \
|
116
|
+
done
|
117
|
+
|
118
|
+
uninstall-man1:
|
119
|
+
@list='$(man1_MANS)'; \
|
120
|
+
l2='$(man_MANS)'; for i in $$l2; do \
|
121
|
+
case "$$i" in \
|
122
|
+
*.1*) list="$$list $$i" ;; \
|
123
|
+
esac; \
|
124
|
+
done; \
|
125
|
+
for i in $$list; do \
|
126
|
+
ext=`echo $$i | sed -e 's/^.*\\.//'`; \
|
127
|
+
inst=`echo $$i | sed -e 's/\\.[0-9a-z]*$$//'`; \
|
128
|
+
inst=`echo $$inst | sed '$(transform)'`.$$ext; \
|
129
|
+
echo " rm -f $(DESTDIR)$(man1dir)/$$inst"; \
|
130
|
+
rm -f $(DESTDIR)$(man1dir)/$$inst; \
|
131
|
+
done
|
132
|
+
|
133
|
+
install-man5:
|
134
|
+
$(mkinstalldirs) $(DESTDIR)$(man5dir)
|
135
|
+
@list='$(man5_MANS)'; \
|
136
|
+
l2='$(man_MANS)'; for i in $$l2; do \
|
137
|
+
case "$$i" in \
|
138
|
+
*.5*) list="$$list $$i" ;; \
|
139
|
+
esac; \
|
140
|
+
done; \
|
141
|
+
for i in $$list; do \
|
142
|
+
if test -f $(srcdir)/$$i; then file=$(srcdir)/$$i; \
|
143
|
+
else file=$$i; fi; \
|
144
|
+
ext=`echo $$i | sed -e 's/^.*\\.//'`; \
|
145
|
+
inst=`echo $$i | sed -e 's/\\.[0-9a-z]*$$//'`; \
|
146
|
+
inst=`echo $$inst | sed '$(transform)'`.$$ext; \
|
147
|
+
echo " $(INSTALL_DATA) $$file $(DESTDIR)$(man5dir)/$$inst"; \
|
148
|
+
$(INSTALL_DATA) $$file $(DESTDIR)$(man5dir)/$$inst; \
|
149
|
+
done
|
150
|
+
|
151
|
+
uninstall-man5:
|
152
|
+
@list='$(man5_MANS)'; \
|
153
|
+
l2='$(man_MANS)'; for i in $$l2; do \
|
154
|
+
case "$$i" in \
|
155
|
+
*.5*) list="$$list $$i" ;; \
|
156
|
+
esac; \
|
157
|
+
done; \
|
158
|
+
for i in $$list; do \
|
159
|
+
ext=`echo $$i | sed -e 's/^.*\\.//'`; \
|
160
|
+
inst=`echo $$i | sed -e 's/\\.[0-9a-z]*$$//'`; \
|
161
|
+
inst=`echo $$inst | sed '$(transform)'`.$$ext; \
|
162
|
+
echo " rm -f $(DESTDIR)$(man5dir)/$$inst"; \
|
163
|
+
rm -f $(DESTDIR)$(man5dir)/$$inst; \
|
164
|
+
done
|
165
|
+
install-man: $(MANS)
|
166
|
+
@$(NORMAL_INSTALL)
|
167
|
+
$(MAKE) $(AM_MAKEFLAGS) install-man1 install-man5
|
168
|
+
uninstall-man:
|
169
|
+
@$(NORMAL_UNINSTALL)
|
170
|
+
$(MAKE) $(AM_MAKEFLAGS) uninstall-man1 uninstall-man5
|
171
|
+
tags: TAGS
|
172
|
+
TAGS:
|
173
|
+
|
174
|
+
|
175
|
+
distdir = $(top_builddir)/$(PACKAGE)-$(VERSION)/$(subdir)
|
176
|
+
|
177
|
+
subdir = doc/man
|
178
|
+
|
179
|
+
distdir: $(DISTFILES)
|
180
|
+
@for file in $(DISTFILES); do \
|
181
|
+
d=$(srcdir); \
|
182
|
+
if test -d $$d/$$file; then \
|
183
|
+
cp -pr $$d/$$file $(distdir)/$$file; \
|
184
|
+
else \
|
185
|
+
test -f $(distdir)/$$file \
|
186
|
+
|| ln $$d/$$file $(distdir)/$$file 2> /dev/null \
|
187
|
+
|| cp -p $$d/$$file $(distdir)/$$file || :; \
|
188
|
+
fi; \
|
189
|
+
done
|
190
|
+
info-am:
|
191
|
+
info: info-am
|
192
|
+
dvi-am:
|
193
|
+
dvi: dvi-am
|
194
|
+
check-am: all-am
|
195
|
+
check: check-am
|
196
|
+
installcheck-am:
|
197
|
+
installcheck: installcheck-am
|
198
|
+
install-exec-am:
|
199
|
+
install-exec: install-exec-am
|
200
|
+
|
201
|
+
install-data-am: install-man
|
202
|
+
install-data: install-data-am
|
203
|
+
|
204
|
+
install-am: all-am
|
205
|
+
@$(MAKE) $(AM_MAKEFLAGS) install-exec-am install-data-am
|
206
|
+
install: install-am
|
207
|
+
uninstall-am: uninstall-man
|
208
|
+
uninstall: uninstall-am
|
209
|
+
all-am: Makefile $(MANS)
|
210
|
+
all-redirect: all-am
|
211
|
+
install-strip:
|
212
|
+
$(MAKE) $(AM_MAKEFLAGS) AM_INSTALL_PROGRAM_FLAGS=-s install
|
213
|
+
installdirs:
|
214
|
+
$(mkinstalldirs) $(DESTDIR)$(mandir)/man1 $(DESTDIR)$(mandir)/man5
|
215
|
+
|
216
|
+
|
217
|
+
mostlyclean-generic:
|
218
|
+
|
219
|
+
clean-generic:
|
220
|
+
|
221
|
+
distclean-generic:
|
222
|
+
-rm -f Makefile $(CONFIG_CLEAN_FILES)
|
223
|
+
-rm -f config.cache config.log stamp-h stamp-h[0-9]*
|
224
|
+
|
225
|
+
maintainer-clean-generic:
|
226
|
+
mostlyclean-am: mostlyclean-generic
|
227
|
+
|
228
|
+
mostlyclean: mostlyclean-am
|
229
|
+
|
230
|
+
clean-am: clean-generic mostlyclean-am
|
231
|
+
|
232
|
+
clean: clean-am
|
233
|
+
|
234
|
+
distclean-am: distclean-generic clean-am
|
235
|
+
|
236
|
+
distclean: distclean-am
|
237
|
+
|
238
|
+
maintainer-clean-am: maintainer-clean-generic distclean-am
|
239
|
+
@echo "This command is intended for maintainers to use;"
|
240
|
+
@echo "it deletes files that may require special tools to rebuild."
|
241
|
+
|
242
|
+
maintainer-clean: maintainer-clean-am
|
243
|
+
|
244
|
+
.PHONY: install-man1 uninstall-man1 install-man5 uninstall-man5 \
|
245
|
+
install-man uninstall-man tags distdir info-am info dvi-am dvi check \
|
246
|
+
check-am installcheck-am installcheck install-exec-am install-exec \
|
247
|
+
install-data-am install-data install-am install uninstall-am uninstall \
|
248
|
+
all-redirect all-am all installdirs mostlyclean-generic \
|
249
|
+
distclean-generic clean-generic maintainer-clean-generic clean \
|
250
|
+
mostlyclean distclean maintainer-clean
|
251
|
+
|
252
|
+
|
253
|
+
# Tell versions [3.59,3.63) of GNU make to not export all variables.
|
254
|
+
# Otherwise a system limit (for SysV at least) may be exceeded.
|
255
|
+
.NOEXPORT:
|
data/doc/man/complearn.5
ADDED
@@ -0,0 +1,91 @@
|
|
1
|
+
.TH COMPLEARN 5
|
2
|
+
.SH NAME
|
3
|
+
complearn \- file format for the complearnrc initialization file
|
4
|
+
.SH SYNOPSIS
|
5
|
+
.B complearnrc
|
6
|
+
controls options for the CompLearn Toolkit
|
7
|
+
.SH DESCRIPTION
|
8
|
+
|
9
|
+
The CompLearn toolkit is a suite of utilities to analyze arbitrary data.
|
10
|
+
The commands
|
11
|
+
.BR ncd (1),
|
12
|
+
.BR ncdvector(1),
|
13
|
+
.BR ncdmatrix (1),
|
14
|
+
.BR makesvm (1),
|
15
|
+
.BR maketree (1),
|
16
|
+
and
|
17
|
+
.BR labeltree (1)
|
18
|
+
all use the \fBcomplearnrc\fR configuration file.
|
19
|
+
First, the user home directory (\fI$HOME\fR) is searched for a file called \fB.complearnrc\fR. If this is not found, then \fB/etc/complearnrc\fR is next searched. If neither of these are found, default values are used. All CompLearn utilities support the \fB-f\fR option, which overrides the standard search with a specific custom complearnrc configuration file.
|
20
|
+
|
21
|
+
For more information on the CompLearn project, please see
|
22
|
+
\fIAlgorithmic Clustering of Music\fR by \fBCilibrasi, Vitanyi,\fR and \fBde Wolf\fR at \fBhttp://www.arxiv.org/abs/cs.SD/0303025\fR
|
23
|
+
|
24
|
+
The format of this file is as follows:
|
25
|
+
|
26
|
+
.B<VariableName>: <value>
|
27
|
+
|
28
|
+
Blank lines are allowed. Comments are designated with a \fB#\fR sign.
|
29
|
+
Variables come in one of four types: boolean, integer, floating-point,
|
30
|
+
or string. Variables may be single or Array parameters, which are
|
31
|
+
separated by commas.
|
32
|
+
|
33
|
+
The following VariableNames are valid.
|
34
|
+
|
35
|
+
.BI InputDir String (general)
|
36
|
+
the directory where input files are to be read from (relative paths only!)
|
37
|
+
|
38
|
+
.BI Symmetric Bool (general)
|
39
|
+
whether or not it is safe to assume that compressed size is irrelevant of concatenation order when compressing. This can save time but may produce slightly different results with certain compressors.
|
40
|
+
|
41
|
+
.BI Hosts IntArray (general)
|
42
|
+
a list of hosts to use when running in multiple-host MPI parallel execution mode
|
43
|
+
|
44
|
+
.BI SingleProcess Bool (general)
|
45
|
+
if true, indicates that only a single host should be used, even if the
|
46
|
+
utilities are started from within the MPI execution environment.
|
47
|
+
|
48
|
+
.BI Compressor String (ncd, ncdvector, ncdmatrix)
|
49
|
+
the name of the compressor to use. Allowable values are bz2 and gzip.
|
50
|
+
|
51
|
+
.BI UnpairedPenalty Float \fB(maketree (1))\fR
|
52
|
+
a floating-point number representing the penalty associated with leaves that do not cluster adjacent to other leaves. A value of 0, the default, means no penalty. A value above 0 (typically 0.05-0.3) indicates an additional cost associated with each of these "unpaired" nodes. Some types of distance matrices, particularly inconsistent ones, can often lead to degenerate trees that are linear and heavily imbalanced without this penalty.
|
53
|
+
|
54
|
+
.BI UseBestThirdOnly Bool \fB(maketree (1))\fR
|
55
|
+
a boolean value indicating if the quartet tree evaluation function should
|
56
|
+
restrict its calculations to only include the single best quartet out of every triple of quartets associated with each group of four labels. This option is off by default. If activated, it may speed up searches but may also lead to different results.
|
57
|
+
|
58
|
+
.BI TreesPerTry Int \fB(maketree (1))\fR
|
59
|
+
an integer specifying the number of mutated trees to try per failure-counter increment. Default 1000.
|
60
|
+
|
61
|
+
.BR MaxFailedTries Int \fB(maketree (1))\fR
|
62
|
+
an integer specifying how many failed batches of trees must occur in succession before the program exits. Default 10.
|
63
|
+
|
64
|
+
.SH EXAMPLE
|
65
|
+
|
66
|
+
.nf
|
67
|
+
#
|
68
|
+
# comments are written like this
|
69
|
+
#
|
70
|
+
InputDir: /home/cilibrar/data/genres
|
71
|
+
Symmetric: yes
|
72
|
+
UnpairedPenalty: 0.05
|
73
|
+
MaxFailedTries: 10000
|
74
|
+
Hosts: localhost, remote1, remote2
|
75
|
+
SingleProcess: yes
|
76
|
+
# etc
|
77
|
+
|
78
|
+
.fi
|
79
|
+
|
80
|
+
.SH FILES
|
81
|
+
.I $HOME/.complearnrc
|
82
|
+
.RS per-user configuration file, overrides systemwide default
|
83
|
+
.I /etc/complearnrc
|
84
|
+
.RS the systemwide configuration file
|
85
|
+
.SH "SEE ALSO"
|
86
|
+
.BR ncd (1),
|
87
|
+
.BR ncdvector (1),
|
88
|
+
.BR ncdmatrix (1),
|
89
|
+
.BR labeltree (1),
|
90
|
+
.BR makesvm (1),
|
91
|
+
.BR maketree (1)
|
data/doc/man/labeltree.1
ADDED
@@ -0,0 +1,35 @@
|
|
1
|
+
.TH LABELTREE 1
|
2
|
+
.SH NAME
|
3
|
+
labeltree \- relabel the vertices of an unlabelled (or integer-labelled) tree
|
4
|
+
.SH SYNOPSIS
|
5
|
+
.B labeltree [-h] [-f
|
6
|
+
.I configfile
|
7
|
+
.B ]
|
8
|
+
.I tree.dot featurelist
|
9
|
+
.SH DESCRIPTION
|
10
|
+
.B labeltree
|
11
|
+
takes as input a tree \fBtree.dot\fR whose leaves are labelled with integers the
|
12
|
+
way \fBmaketree (1)\fR produces. It also must be given a file listing or directory name \fBfeaturelist\fR, and from this changes \fBtree.dot\fR to show the filenames associated with each leaf. Characters are stripped from the filenames so that only alphanumeric characters are shown in the resultant labelled \fBtree.dot\fR file. The original file is renamed to \fBtree.dot.orig\fR as a backup.
|
13
|
+
.SH OPTIONS
|
14
|
+
.IP -h
|
15
|
+
Print usage information and exit.
|
16
|
+
.IP "-f configfile"
|
17
|
+
Use a different configuration file than the default search locations
|
18
|
+
.SH FILES
|
19
|
+
.I $HOME/.complearnrc
|
20
|
+
.RS per-user configuration file, overrides systemwide default
|
21
|
+
.I /etc/complearnrc
|
22
|
+
.RS the systemwide configuration file
|
23
|
+
see
|
24
|
+
.BR complearn (5)
|
25
|
+
for further details.
|
26
|
+
.SH ENVIRONMENT
|
27
|
+
none
|
28
|
+
.SH DIAGNOSTICS
|
29
|
+
If any of the files cannot be read, an error will be printed and the program exits with a nonzero exit code.
|
30
|
+
.SH "SEE ALSO"
|
31
|
+
.BR maketree (1),
|
32
|
+
.BR ncd (1),
|
33
|
+
.BR makesvm (1),
|
34
|
+
.BR ncdmatrix (1),
|
35
|
+
.BR complearn (5)
|
data/doc/man/makesvm.1
ADDED
@@ -0,0 +1,60 @@
|
|
1
|
+
.TH MAKESVM 1
|
2
|
+
.SH NAME
|
3
|
+
makesvm \- Prepare a set of training and testing files for LIBSVM using the NCD and a classification specification file
|
4
|
+
.SH SYNOPSIS
|
5
|
+
.B makesvm [-h] [-f
|
6
|
+
.I configfile
|
7
|
+
.B ]
|
8
|
+
.I classfile.txt
|
9
|
+
.SH DESCRIPTION
|
10
|
+
.B makesvm
|
11
|
+
reads a classification specification file and produces two output files
|
12
|
+
called \fBtrainsvm.txt\fR and \fBtestsvm.txt\fR. These files are suitable
|
13
|
+
for use with the \fBLIBSVM\fR package, particularly \fBgrid.py (1)\fR,
|
14
|
+
\fBsvm-train (1)\fR, and \fBsvm-predict (1)\fR. Each line of
|
15
|
+
\fBclassfile.txt\fR must be either a blank line, a comment denoted with \fB#\fR, or a line with three whitespace-separated fields. These three
|
16
|
+
fields must be of the form
|
17
|
+
|
18
|
+
.B <intclass> <typecode> <filename>
|
19
|
+
|
20
|
+
\fB<intclass>\fR is a positive integer indicating to which class
|
21
|
+
this file belongs. \fB<typecode>\fR is one of the three letters
|
22
|
+
\fBF\fR, \fBG\fR, or \fBP\fR, standing for "feature", "given", or "predict".
|
23
|
+
Feature files are used to create NCD components of non-feature files.
|
24
|
+
Given files are used in \fBtrainsvm.txt\fR. Predict files are used in
|
25
|
+
\fBtestsvm.txt\fR. The number of feature files determines the dimensionality
|
26
|
+
of the resultant vectors used in training and testing. The third field,
|
27
|
+
\fB<filename>\fR, specifies the input file to be analyzed with compression,
|
28
|
+
and may be relative to the \fIInputDir\fR configuration variable. In
|
29
|
+
typical usage,
|
30
|
+
\fBtrainsvm.txt\fR will be used by \fBgrid.py\fR and \fBsvm-train\fR, whereas
|
31
|
+
\fBtestsvm.txt\fR will be used by \fBsvm-predict\fR.
|
32
|
+
|
33
|
+
Each line of the output files corresponds to a vector of the form
|
34
|
+
produced by \fBncdvector (1)\fR. The NCD is defined in
|
35
|
+
.BR ncd (1).
|
36
|
+
|
37
|
+
.SH OPTIONS
|
38
|
+
.IP -h
|
39
|
+
Print usage information and exit.
|
40
|
+
.IP "-f configfile"
|
41
|
+
Use a different configuration file than the default search locations
|
42
|
+
.SH FILES
|
43
|
+
.I $HOME/.complearnrc
|
44
|
+
.RS per-user configuration file, overrides systemwide default
|
45
|
+
.I /etc/complearnrc
|
46
|
+
.RS the systemwide configuration file
|
47
|
+
see
|
48
|
+
.BR complearn (5)
|
49
|
+
for further details.
|
50
|
+
.SH ENVIRONMENT
|
51
|
+
none
|
52
|
+
.SH DIAGNOSTICS
|
53
|
+
If any of the files cannot be read, an error will be printed and the program exits with a nonzero exit code.
|
54
|
+
.SH "SEE ALSO"
|
55
|
+
.BR ncd (1),
|
56
|
+
.BR maketree (1),
|
57
|
+
.BR ncdvector (1),
|
58
|
+
.BR svm-train (1),
|
59
|
+
.BR svm-predict (1),
|
60
|
+
.BR complearn (5)
|