| line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
|
1
|
|
|
|
|
|
|
package Algorithm::Classifier::NaiveBayes::App::Command::untrain; |
|
2
|
|
|
|
|
|
|
|
|
3
|
1
|
|
|
1
|
|
716
|
use 5.006; |
|
|
1
|
|
|
|
|
3
|
|
|
4
|
1
|
|
|
1
|
|
4
|
use strict; |
|
|
1
|
|
|
|
|
2
|
|
|
|
1
|
|
|
|
|
38
|
|
|
5
|
1
|
|
|
1
|
|
5
|
use warnings; |
|
|
1
|
|
|
|
|
1
|
|
|
|
1
|
|
|
|
|
36
|
|
|
6
|
1
|
|
|
1
|
|
4
|
use Algorithm::Classifier::NaiveBayes (); |
|
|
1
|
|
|
|
|
1
|
|
|
|
1
|
|
|
|
|
17
|
|
|
7
|
1
|
|
|
1
|
|
3
|
use Algorithm::Classifier::NaiveBayes::App -command; |
|
|
1
|
|
|
|
|
1
|
|
|
|
1
|
|
|
|
|
8
|
|
|
8
|
1
|
|
|
1
|
|
289
|
use File::Slurp qw(read_file); |
|
|
1
|
|
|
|
|
1
|
|
|
|
1
|
|
|
|
|
372
|
|
|
9
|
|
|
|
|
|
|
|
|
10
|
|
|
|
|
|
|
sub options { |
|
11
|
|
|
|
|
|
|
return ( |
|
12
|
3
|
|
|
3
|
0
|
31
|
[ 'm=s', 'Model JSON file path/name.', { 'default' => 'nb_model.json', 'completion' => 'files' } ], |
|
13
|
|
|
|
|
|
|
[ 'c=s', 'Class to untrain.' ], |
|
14
|
|
|
|
|
|
|
[ 'f=s', 'File to read the text from.', { 'completion' => 'files' } ], |
|
15
|
|
|
|
|
|
|
); |
|
16
|
|
|
|
|
|
|
} |
|
17
|
|
|
|
|
|
|
|
|
18
|
0
|
|
|
0
|
1
|
0
|
sub abstract { 'Untrain a class on the specified text' } |
|
19
|
|
|
|
|
|
|
|
|
20
|
|
|
|
|
|
|
sub description { |
|
21
|
0
|
|
|
0
|
1
|
0
|
return 'Untrain a class on the specified text, reversing a previous train. |
|
22
|
|
|
|
|
|
|
|
|
23
|
|
|
|
|
|
|
The text is taken from the file specified via -f, the remaining args |
|
24
|
|
|
|
|
|
|
joined by a space, or from stdin. |
|
25
|
|
|
|
|
|
|
|
|
26
|
|
|
|
|
|
|
nb_tool untrain -m model.json -c spam something that is not spam |
|
27
|
|
|
|
|
|
|
nb_tool untrain -m model.json -c spam -f some_ham.txt |
|
28
|
|
|
|
|
|
|
cat ham | nb_tool untrain -m model.json -c spam |
|
29
|
|
|
|
|
|
|
'; |
|
30
|
|
|
|
|
|
|
} ## end sub description |
|
31
|
|
|
|
|
|
|
|
|
32
|
|
|
|
|
|
|
sub validate { |
|
33
|
3
|
|
|
3
|
0
|
7
|
my ( $self, $opt, $args ) = @_; |
|
34
|
|
|
|
|
|
|
|
|
35
|
3
|
50
|
|
|
|
10
|
if ( !defined( $opt->{'c'} ) ) { |
|
36
|
0
|
|
|
|
|
0
|
$self->usage_error('-c has not been specified'); |
|
37
|
|
|
|
|
|
|
} |
|
38
|
|
|
|
|
|
|
|
|
39
|
3
|
50
|
|
|
|
78
|
if ( !-f $opt->{'m'} ) { |
|
40
|
0
|
|
|
|
|
0
|
$self->usage_error( '-m, "' . $opt->{'m'} . '", is not a file or does not exist' ); |
|
41
|
|
|
|
|
|
|
} |
|
42
|
|
|
|
|
|
|
|
|
43
|
3
|
100
|
|
|
|
11
|
if ( defined( $opt->{'f'} ) ) { |
|
44
|
2
|
100
|
|
|
|
31
|
if ( @{$args} ) { |
|
|
2
|
|
|
|
|
12
|
|
|
45
|
1
|
|
|
|
|
16
|
$self->usage_error('-f and text args may not be used together'); |
|
46
|
|
|
|
|
|
|
} |
|
47
|
1
|
50
|
|
|
|
27
|
if ( !-f $opt->{'f'} ) { |
|
|
|
50
|
|
|
|
|
|
|
48
|
0
|
|
|
|
|
0
|
$self->usage_error( '-f, "' . $opt->{'f'} . '", is not a file or does not exist' ); |
|
49
|
|
|
|
|
|
|
} elsif ( !-r $opt->{'f'} ) { |
|
50
|
0
|
|
|
|
|
0
|
$self->usage_error( '-f, "' . $opt->{'f'} . '", is not readable' ); |
|
51
|
|
|
|
|
|
|
} |
|
52
|
|
|
|
|
|
|
} ## end if ( defined( $opt->{'f'} ) ) |
|
53
|
|
|
|
|
|
|
|
|
54
|
2
|
|
|
|
|
6
|
return 1; |
|
55
|
|
|
|
|
|
|
} ## end sub validate |
|
56
|
|
|
|
|
|
|
|
|
57
|
|
|
|
|
|
|
sub execute { |
|
58
|
2
|
|
|
2
|
1
|
10
|
my ( $self, $opt, $args ) = @_; |
|
59
|
|
|
|
|
|
|
|
|
60
|
2
|
|
|
|
|
17
|
my $nb = Algorithm::Classifier::NaiveBayes->new; |
|
61
|
2
|
|
|
|
|
10
|
$nb->load( $opt->{'m'} ); |
|
62
|
|
|
|
|
|
|
|
|
63
|
2
|
100
|
|
|
|
17
|
my $text = defined( $opt->{'f'} ) ? read_file( $opt->{'f'} ) : $self->text_from($args); |
|
64
|
2
|
|
|
|
|
108
|
$nb->untrain( $opt->{'c'}, $text ); |
|
65
|
2
|
|
|
|
|
9
|
$nb->save( $opt->{'m'} ); |
|
66
|
|
|
|
|
|
|
|
|
67
|
2
|
|
|
|
|
17
|
print 'Untrained "' . $opt->{'c'} . '", ' . $nb->{'model'}{'total_docs'} . ' total documents in the model' . "\n"; |
|
68
|
|
|
|
|
|
|
} ## end sub execute |
|
69
|
|
|
|
|
|
|
|
|
70
|
|
|
|
|
|
|
1; |