line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
package Statistics::Descriptive::Discrete; |
2
|
|
|
|
|
|
|
|
3
|
|
|
|
|
|
|
### This module draws heavily from Statistics::Descriptive |
4
|
|
|
|
|
|
|
|
5
|
3
|
|
|
3
|
|
211232
|
use strict; |
|
3
|
|
|
|
|
30
|
|
|
3
|
|
|
|
|
92
|
|
6
|
3
|
|
|
3
|
|
16
|
use warnings; |
|
3
|
|
|
|
|
7
|
|
|
3
|
|
|
|
|
111
|
|
7
|
3
|
|
|
3
|
|
16
|
use Carp; |
|
3
|
|
|
|
|
7
|
|
|
3
|
|
|
|
|
183
|
|
8
|
3
|
|
|
3
|
|
1541
|
use AutoLoader; |
|
3
|
|
|
|
|
4301
|
|
|
3
|
|
|
|
|
19
|
|
9
|
3
|
|
|
3
|
|
126
|
use vars qw($VERSION $AUTOLOAD $DEBUG $Tolerance %autosubs); |
|
3
|
|
|
|
|
7
|
|
|
3
|
|
|
|
|
6397
|
|
10
|
|
|
|
|
|
|
|
11
|
|
|
|
|
|
|
$VERSION = '0.10'; |
12
|
|
|
|
|
|
|
$DEBUG = 0; |
13
|
|
|
|
|
|
|
|
14
|
|
|
|
|
|
|
$Tolerance = 0.0; |
15
|
|
|
|
|
|
|
|
16
|
|
|
|
|
|
|
#what subs can be autoloaded? |
17
|
|
|
|
|
|
|
%autosubs = ( |
18
|
|
|
|
|
|
|
count => undef, |
19
|
|
|
|
|
|
|
mean => undef, |
20
|
|
|
|
|
|
|
geometric_mean=> undef, |
21
|
|
|
|
|
|
|
harmonic_mean=>undef, |
22
|
|
|
|
|
|
|
sum => undef, |
23
|
|
|
|
|
|
|
mode => undef, |
24
|
|
|
|
|
|
|
median => undef, |
25
|
|
|
|
|
|
|
min => undef, |
26
|
|
|
|
|
|
|
max => undef, |
27
|
|
|
|
|
|
|
mindex => undef, |
28
|
|
|
|
|
|
|
maxdex => undef, |
29
|
|
|
|
|
|
|
standard_deviation => undef, |
30
|
|
|
|
|
|
|
sample_range => undef, |
31
|
|
|
|
|
|
|
variance => undef, |
32
|
|
|
|
|
|
|
text => undef, |
33
|
|
|
|
|
|
|
); |
34
|
|
|
|
|
|
|
|
35
|
|
|
|
|
|
|
|
36
|
|
|
|
|
|
|
sub new |
37
|
|
|
|
|
|
|
{ |
38
|
17
|
|
|
17
|
1
|
271
|
my $proto = shift; |
39
|
17
|
|
33
|
|
|
78
|
my $class = ref($proto) || $proto; |
40
|
17
|
|
|
|
|
34
|
my $self = {}; |
41
|
17
|
|
|
|
|
52
|
$self->{_permitted} = \%autosubs; |
42
|
17
|
|
|
|
|
32
|
$self->{data} = (); |
43
|
17
|
|
|
|
|
31
|
$self->{_dataindex} = (); #index of where each value first seen when adding data |
44
|
17
|
|
|
|
|
31
|
$self->{dirty} = 1; #is the data dirty? |
45
|
17
|
|
|
|
|
30
|
$self->{_index} = 0; #current index of number of data items added |
46
|
|
|
|
|
|
|
|
47
|
17
|
|
|
|
|
35
|
bless ($self,$class); |
48
|
17
|
50
|
|
|
|
39
|
print __PACKAGE__,"->new(",join(',',@_),")\n" if $DEBUG; |
49
|
17
|
|
|
|
|
50
|
return $self; |
50
|
|
|
|
|
|
|
} |
51
|
|
|
|
|
|
|
|
52
|
|
|
|
|
|
|
# Clear the stat object & erase all data |
53
|
|
|
|
|
|
|
# Object will be ready to use as if new was called |
54
|
|
|
|
|
|
|
# Not sure this is more efficient than just creating a new object but |
55
|
|
|
|
|
|
|
# maintained for compatability with Statistics::Descriptive |
56
|
|
|
|
|
|
|
sub clear |
57
|
|
|
|
|
|
|
{ |
58
|
8
|
|
|
8
|
1
|
3842
|
my $self = shift; |
59
|
8
|
|
|
|
|
16
|
my %keys = %{ $self }; |
|
8
|
|
|
|
|
71
|
|
60
|
|
|
|
|
|
|
|
61
|
|
|
|
|
|
|
#remove _permitted from the deletion list |
62
|
8
|
|
|
|
|
24
|
delete $keys{"_permitted"}; |
63
|
|
|
|
|
|
|
|
64
|
8
|
|
|
|
|
34
|
foreach my $key (keys %keys) |
65
|
|
|
|
|
|
|
{ # Check each key in the object |
66
|
132
|
50
|
|
|
|
224
|
print __PACKAGE__,"->clear, deleting $key\n" if $DEBUG; |
67
|
132
|
|
|
|
|
210
|
delete $self->{$key}; # Delete any out of date cached key |
68
|
|
|
|
|
|
|
} |
69
|
8
|
|
|
|
|
19
|
$self->{data} = (); |
70
|
8
|
|
|
|
|
16
|
$self->{_dataindex} = (); |
71
|
8
|
|
|
|
|
12
|
$self->{dirty} = 1; |
72
|
8
|
|
|
|
|
40
|
$self->{_index} = 0; |
73
|
|
|
|
|
|
|
} |
74
|
|
|
|
|
|
|
|
75
|
|
|
|
|
|
|
sub add_data |
76
|
|
|
|
|
|
|
{ |
77
|
|
|
|
|
|
|
#add data but don't compute ANY statistics yet |
78
|
21
|
|
|
21
|
1
|
671
|
my $self = shift; |
79
|
21
|
50
|
|
|
|
47
|
print __PACKAGE__,"->add_data(",join(',',@_),")\n" if $DEBUG; |
80
|
|
|
|
|
|
|
|
81
|
|
|
|
|
|
|
#get each element and add 0 to force it be a number |
82
|
|
|
|
|
|
|
#that way, 0.000 and 0 are treated the same |
83
|
21
|
|
|
|
|
36
|
my $val = shift; |
84
|
21
|
|
|
|
|
62
|
while (defined $val) |
85
|
|
|
|
|
|
|
{ |
86
|
136
|
|
|
|
|
199
|
$val += 0; |
87
|
136
|
|
|
|
|
354
|
$self->{data}{$val}++; |
88
|
136
|
100
|
|
|
|
353
|
if (not exists $self->{_dataindex}{$val}) { |
89
|
74
|
|
|
|
|
158
|
$self->{_dataindex}{$val} = $self->{_index}; |
90
|
|
|
|
|
|
|
} |
91
|
136
|
|
|
|
|
185
|
$self->{_index}++; |
92
|
|
|
|
|
|
|
#set dirty flag so we know cached stats are invalid |
93
|
136
|
|
|
|
|
199
|
$self->{dirty}++; |
94
|
136
|
|
|
|
|
428
|
$val = shift; #get next element |
95
|
|
|
|
|
|
|
} |
96
|
|
|
|
|
|
|
} |
97
|
|
|
|
|
|
|
|
98
|
|
|
|
|
|
|
sub add_data_tuple |
99
|
|
|
|
|
|
|
{ |
100
|
|
|
|
|
|
|
#add data but don't compute ANY statistics yet |
101
|
|
|
|
|
|
|
#the data are pairs of values and occurrences |
102
|
|
|
|
|
|
|
#e.g. 4,2 means 2 occurrences of the value 4 |
103
|
|
|
|
|
|
|
#thanks to Bill Dueber for suggesting this |
104
|
|
|
|
|
|
|
|
105
|
5
|
|
|
5
|
1
|
26
|
my $self = shift; |
106
|
5
|
50
|
|
|
|
13
|
print __PACKAGE__,"->add_data_tuple(",join(',',@_),")\n" if $DEBUG; |
107
|
|
|
|
|
|
|
|
108
|
|
|
|
|
|
|
#we want an even number of arguments (tuples in the form (value, count)) |
109
|
5
|
50
|
|
|
|
15
|
carp "argument list must have even number of elements" if @_ % 2; |
110
|
|
|
|
|
|
|
|
111
|
|
|
|
|
|
|
#get each element and add 0 to force it be a number |
112
|
|
|
|
|
|
|
#that way, 0.000 and 0 are treated the same |
113
|
|
|
|
|
|
|
#if $count is 0, then this will set the dirty flag but have no effect on |
114
|
|
|
|
|
|
|
#the statistics |
115
|
5
|
|
|
|
|
7
|
my $val = shift; |
116
|
5
|
|
|
|
|
9
|
my $count = shift; |
117
|
5
|
|
|
|
|
13
|
while (defined $count) |
118
|
|
|
|
|
|
|
{ |
119
|
18
|
|
|
|
|
26
|
$val += 0; |
120
|
18
|
|
|
|
|
37
|
$self->{data}{$val} += $count; |
121
|
18
|
50
|
|
|
|
37
|
if (not exists $self->{_dataindex}{$val}) { |
122
|
18
|
|
|
|
|
32
|
$self->{_dataindex}{$val} = $self->{_index}; |
123
|
|
|
|
|
|
|
} |
124
|
18
|
|
|
|
|
27
|
$self->{_index} += $count; |
125
|
|
|
|
|
|
|
#set dirty flag so we know cached stats are invalid |
126
|
18
|
|
|
|
|
24
|
$self->{dirty}++; |
127
|
18
|
|
|
|
|
28
|
$val = shift; #get next element |
128
|
18
|
|
|
|
|
40
|
$count = shift; |
129
|
|
|
|
|
|
|
} |
130
|
|
|
|
|
|
|
} |
131
|
|
|
|
|
|
|
|
132
|
|
|
|
|
|
|
sub _test_for_too_small_val |
133
|
|
|
|
|
|
|
{ |
134
|
112
|
|
|
112
|
|
176
|
my $self = shift; |
135
|
112
|
|
|
|
|
159
|
my $val = shift; |
136
|
|
|
|
|
|
|
|
137
|
112
|
|
|
|
|
285
|
return (abs($val) <= $Statistics::Descriptive::Discrete::Tolerance); |
138
|
|
|
|
|
|
|
} |
139
|
|
|
|
|
|
|
|
140
|
|
|
|
|
|
|
sub _calc_harmonic_mean |
141
|
|
|
|
|
|
|
{ |
142
|
24
|
|
|
24
|
|
38
|
my $self = shift; |
143
|
24
|
|
|
|
|
48
|
my $count = shift; |
144
|
24
|
|
|
|
|
39
|
my $datakeys = shift; #array ref |
145
|
|
|
|
|
|
|
|
146
|
24
|
|
|
|
|
36
|
my $hs = 0; |
147
|
|
|
|
|
|
|
|
148
|
24
|
|
|
|
|
35
|
foreach my $val ( @{$datakeys} ) |
|
24
|
|
|
|
|
50
|
|
149
|
|
|
|
|
|
|
{ |
150
|
|
|
|
|
|
|
##Guarantee that there are no divide by zeros |
151
|
91
|
100
|
|
|
|
160
|
if ($self->_test_for_too_small_val($val)) |
152
|
|
|
|
|
|
|
{ |
153
|
3
|
|
|
|
|
10
|
return; |
154
|
|
|
|
|
|
|
} |
155
|
|
|
|
|
|
|
|
156
|
88
|
|
|
|
|
180
|
foreach (1..$self->{data}{$val}) |
157
|
|
|
|
|
|
|
{ |
158
|
201
|
|
|
|
|
355
|
$hs += 1/$val; |
159
|
|
|
|
|
|
|
} |
160
|
|
|
|
|
|
|
} |
161
|
|
|
|
|
|
|
|
162
|
21
|
100
|
|
|
|
40
|
if ($self->_test_for_too_small_val($hs)) |
163
|
|
|
|
|
|
|
{ |
164
|
2
|
|
|
|
|
5
|
return; |
165
|
|
|
|
|
|
|
} |
166
|
|
|
|
|
|
|
|
167
|
19
|
|
|
|
|
42
|
return $count/$hs; |
168
|
|
|
|
|
|
|
} |
169
|
|
|
|
|
|
|
|
170
|
|
|
|
|
|
|
sub _all_stats |
171
|
|
|
|
|
|
|
{ |
172
|
|
|
|
|
|
|
#compute all the stats in one sub to save overhead of sub calls |
173
|
|
|
|
|
|
|
#a little wasteful to do this if all we want is count or sum for example but |
174
|
|
|
|
|
|
|
#I want to keep add_data as lean as possible since it gets called a lot |
175
|
28
|
|
|
28
|
|
49
|
my $self = shift; |
176
|
28
|
50
|
|
|
|
59
|
print __PACKAGE__,"->_all_stats(",join(',',@_),")\n" if $DEBUG; |
177
|
|
|
|
|
|
|
|
178
|
|
|
|
|
|
|
#if data is empty, set all stats to undef and return |
179
|
28
|
100
|
|
|
|
70
|
if (!$self->{data}) |
180
|
|
|
|
|
|
|
{ |
181
|
4
|
|
|
|
|
6
|
foreach my $key (keys %{$self->{_permitted}}) |
|
4
|
|
|
|
|
21
|
|
182
|
|
|
|
|
|
|
{ |
183
|
60
|
|
|
|
|
95
|
$self->{$key} = undef; |
184
|
|
|
|
|
|
|
} |
185
|
4
|
|
|
|
|
13
|
$self->{count} = 0; |
186
|
4
|
|
|
|
|
7
|
return; |
187
|
|
|
|
|
|
|
} |
188
|
|
|
|
|
|
|
|
189
|
|
|
|
|
|
|
#count = total number of data values we have |
190
|
24
|
|
|
|
|
38
|
my $count = 0; |
191
|
24
|
|
|
|
|
38
|
$count += $_ foreach (values %{$self->{data}}); |
|
24
|
|
|
|
|
110
|
|
192
|
|
|
|
|
|
|
|
193
|
24
|
|
|
|
|
41
|
my @datakeys = keys %{$self->{data}}; |
|
24
|
|
|
|
|
68
|
|
194
|
|
|
|
|
|
|
|
195
|
|
|
|
|
|
|
#initialize min, max, mode to an arbitrary value that's in the hash |
196
|
24
|
|
|
|
|
49
|
my $default = $datakeys[0]; |
197
|
24
|
|
|
|
|
40
|
my $max = $default; |
198
|
24
|
|
|
|
|
34
|
my $min = $default; |
199
|
24
|
|
|
|
|
39
|
my $mode = $default; |
200
|
24
|
|
|
|
|
39
|
my $moden = 0; |
201
|
24
|
|
|
|
|
37
|
my $sum = 0; |
202
|
|
|
|
|
|
|
|
203
|
|
|
|
|
|
|
#find min, max, sum, and mode |
204
|
24
|
|
|
|
|
44
|
foreach (@datakeys) |
205
|
|
|
|
|
|
|
{ |
206
|
95
|
|
|
|
|
149
|
my $n = $self->{data}{$_}; |
207
|
95
|
|
|
|
|
225
|
$sum += $_ * $n; |
208
|
95
|
100
|
|
|
|
198
|
$min = $_ if $_ < $min; |
209
|
95
|
100
|
|
|
|
184
|
$max = $_ if $_ > $max; |
210
|
|
|
|
|
|
|
|
211
|
|
|
|
|
|
|
#only finds one mode but there could be more than one |
212
|
|
|
|
|
|
|
#also, there might not be any mode (all the same frequency) |
213
|
|
|
|
|
|
|
#todo: need to make this more robust |
214
|
95
|
100
|
|
|
|
192
|
if ($n > $moden) |
215
|
|
|
|
|
|
|
{ |
216
|
28
|
|
|
|
|
43
|
$mode = $_; |
217
|
28
|
|
|
|
|
52
|
$moden = $n; |
218
|
|
|
|
|
|
|
} |
219
|
|
|
|
|
|
|
} |
220
|
24
|
|
|
|
|
45
|
my $mindex = $self->{_dataindex}{$min}; |
221
|
24
|
|
|
|
|
41
|
my $maxdex = $self->{_dataindex}{$max}; |
222
|
|
|
|
|
|
|
|
223
|
24
|
|
|
|
|
49
|
my $mean = $sum/$count; |
224
|
|
|
|
|
|
|
|
225
|
24
|
|
|
|
|
35
|
my $stddev = 0; |
226
|
24
|
|
|
|
|
39
|
my $variance = 0; |
227
|
|
|
|
|
|
|
|
228
|
24
|
100
|
|
|
|
52
|
if ($count > 1) |
229
|
|
|
|
|
|
|
{ |
230
|
|
|
|
|
|
|
# Thanks to Peter Dienes for finding and fixing a round-off error |
231
|
|
|
|
|
|
|
# in the following variance calculation |
232
|
|
|
|
|
|
|
|
233
|
23
|
|
|
|
|
47
|
foreach my $val (@datakeys) |
234
|
|
|
|
|
|
|
{ |
235
|
94
|
|
|
|
|
213
|
$stddev += $self->{data}{$val} * (($val - $mean) ** 2); |
236
|
|
|
|
|
|
|
} |
237
|
23
|
|
|
|
|
41
|
$variance = $stddev / ($count - 1); |
238
|
23
|
|
|
|
|
41
|
$stddev = sqrt($variance); |
239
|
|
|
|
|
|
|
} |
240
|
1
|
|
|
|
|
3
|
else {$stddev = undef} |
241
|
|
|
|
|
|
|
|
242
|
|
|
|
|
|
|
#find median, and do it without creating a list of the all the data points |
243
|
|
|
|
|
|
|
#if n=count is odd and n=2k+1 then median = data(k+1) |
244
|
|
|
|
|
|
|
#if n=count is even and n=2k, then median = (data(k) + data(k+1))/2 |
245
|
24
|
|
|
|
|
46
|
my $odd = $count % 2; #odd or even number of points? |
246
|
24
|
|
|
|
|
53
|
my $even = !$odd; |
247
|
24
|
100
|
|
|
|
55
|
my $k = $odd ? ($count-1)/2 : $count/2; |
248
|
24
|
|
|
|
|
38
|
my $median = undef; |
249
|
24
|
|
|
|
|
39
|
my $temp = 0; |
250
|
24
|
|
|
|
|
82
|
MEDIAN: foreach my $val (sort {$a <=> $b} (@datakeys)) |
|
133
|
|
|
|
|
226
|
|
251
|
|
|
|
|
|
|
{ |
252
|
67
|
|
|
|
|
139
|
foreach (1..$self->{data}{$val}) |
253
|
|
|
|
|
|
|
{ |
254
|
124
|
|
|
|
|
173
|
$temp++; |
255
|
124
|
100
|
100
|
|
|
375
|
if (($temp == $k) && $even) |
|
|
100
|
|
|
|
|
|
256
|
|
|
|
|
|
|
{ |
257
|
16
|
|
|
|
|
33
|
$median += $val; |
258
|
|
|
|
|
|
|
} |
259
|
|
|
|
|
|
|
elsif ($temp == $k+1) |
260
|
|
|
|
|
|
|
{ |
261
|
24
|
|
|
|
|
39
|
$median += $val; |
262
|
24
|
100
|
|
|
|
48
|
$median /= 2 if $even; |
263
|
24
|
|
|
|
|
52
|
last MEDIAN; |
264
|
|
|
|
|
|
|
} |
265
|
|
|
|
|
|
|
} |
266
|
|
|
|
|
|
|
} |
267
|
|
|
|
|
|
|
|
268
|
|
|
|
|
|
|
#compute geometric mean |
269
|
24
|
|
|
|
|
45
|
my $gm = 1; |
270
|
24
|
|
|
|
|
41
|
my $exponent = 1/$count; |
271
|
24
|
|
|
|
|
44
|
foreach my $val (@datakeys) |
272
|
|
|
|
|
|
|
{ |
273
|
90
|
100
|
|
|
|
166
|
if ($val < 0) |
274
|
|
|
|
|
|
|
{ |
275
|
3
|
|
|
|
|
8
|
$gm = undef; |
276
|
3
|
|
|
|
|
5
|
last; |
277
|
|
|
|
|
|
|
} |
278
|
87
|
|
|
|
|
144
|
foreach (1..$self->{data}{$val}) |
279
|
|
|
|
|
|
|
{ |
280
|
200
|
|
|
|
|
393
|
$gm *= $val**$exponent; |
281
|
|
|
|
|
|
|
} |
282
|
|
|
|
|
|
|
} |
283
|
|
|
|
|
|
|
|
284
|
|
|
|
|
|
|
#compute harmonic mean |
285
|
24
|
|
|
|
|
65
|
my $harmonic_mean = scalar $self->_calc_harmonic_mean($count, \@datakeys); |
286
|
|
|
|
|
|
|
|
287
|
24
|
50
|
|
|
|
52
|
print __PACKAGE__,"count: $count, _index ",$self->{_index},"\n" if $DEBUG; |
288
|
|
|
|
|
|
|
|
289
|
24
|
|
|
|
|
55
|
$self->{count} = $count; |
290
|
24
|
|
|
|
|
51
|
$self->{sum} = $sum; |
291
|
24
|
|
|
|
|
60
|
$self->{standard_deviation} = $stddev; |
292
|
24
|
|
|
|
|
40
|
$self->{variance} = $variance; |
293
|
24
|
|
|
|
|
44
|
$self->{min} = $min; |
294
|
24
|
|
|
|
|
44
|
$self->{max} = $max; |
295
|
24
|
|
|
|
|
40
|
$self->{mindex} = $mindex; |
296
|
24
|
|
|
|
|
37
|
$self->{maxdex} = $maxdex; |
297
|
24
|
|
|
|
|
49
|
$self->{sample_range} = $max - $min; #todo: does this require any bounds checking |
298
|
24
|
|
|
|
|
40
|
$self->{mean} = $mean; |
299
|
24
|
|
|
|
|
58
|
$self->{geometric_mean} = $gm; |
300
|
24
|
|
|
|
|
39
|
$self->{harmonic_mean} = $harmonic_mean; |
301
|
24
|
|
|
|
|
40
|
$self->{median} = $median; |
302
|
24
|
|
|
|
|
39
|
$self->{mode} = $mode; |
303
|
|
|
|
|
|
|
|
304
|
|
|
|
|
|
|
#clear dirty flag so we don't needlessly recompute the statistics |
305
|
24
|
|
|
|
|
66
|
$self->{dirty} = 0; |
306
|
|
|
|
|
|
|
} |
307
|
|
|
|
|
|
|
|
308
|
|
|
|
|
|
|
sub set_text |
309
|
|
|
|
|
|
|
{ |
310
|
0
|
|
|
0
|
0
|
0
|
my $self = shift; |
311
|
0
|
|
|
|
|
0
|
$self->{text} = shift; |
312
|
|
|
|
|
|
|
} |
313
|
|
|
|
|
|
|
|
314
|
|
|
|
|
|
|
sub get_data |
315
|
|
|
|
|
|
|
{ |
316
|
|
|
|
|
|
|
#returns a list of the data in sorted order |
317
|
|
|
|
|
|
|
#the list could be very big an this defeat the purpose of using this module |
318
|
|
|
|
|
|
|
#use this only if you really need it |
319
|
1
|
|
|
1
|
1
|
2
|
my $self = shift; |
320
|
1
|
50
|
|
|
|
7
|
print __PACKAGE__,"->get_data(",join(',',@_),")\n" if $DEBUG; |
321
|
|
|
|
|
|
|
|
322
|
1
|
|
|
|
|
2
|
my @data; |
323
|
1
|
|
|
|
|
2
|
foreach my $val (sort {$a <=> $b} (keys %{$self->{data}})) |
|
9
|
|
|
|
|
17
|
|
|
1
|
|
|
|
|
6
|
|
324
|
|
|
|
|
|
|
{ |
325
|
5
|
|
|
|
|
18
|
push @data, $val foreach (1..$self->{data}{$val}); |
326
|
|
|
|
|
|
|
} |
327
|
1
|
|
|
|
|
5
|
return @data; |
328
|
|
|
|
|
|
|
} |
329
|
|
|
|
|
|
|
|
330
|
|
|
|
|
|
|
# this is the previous frequency_distribution code |
331
|
|
|
|
|
|
|
# redid this completely based on current implementation in |
332
|
|
|
|
|
|
|
# Statistics::Descriptive |
333
|
|
|
|
|
|
|
# sub frequency_distribution |
334
|
|
|
|
|
|
|
# { |
335
|
|
|
|
|
|
|
# #Compute frequency distribution (histogram), borrowed heavily from Statistics::Descriptive |
336
|
|
|
|
|
|
|
# #Behavior is slightly different than Statistics::Descriptive |
337
|
|
|
|
|
|
|
# #e.g. if partition is not specified, we use to set the number of partitions |
338
|
|
|
|
|
|
|
# # if partition = 0, then we return the data hash WITHOUT binning it into equal bins |
339
|
|
|
|
|
|
|
# # I often want to just see how many of each value I saw |
340
|
|
|
|
|
|
|
# #Also, you can manually pass in the bin info (min bin, bin size, and number of partitions) |
341
|
|
|
|
|
|
|
# #I don't cache the frequency data like Statistics::Descriptive does since it's not as expensive to compute |
342
|
|
|
|
|
|
|
# #but I might add that later |
343
|
|
|
|
|
|
|
# #todo: the minbin/binsize stuff is funky and not intuitive -- fix it |
344
|
|
|
|
|
|
|
# my $self = shift; |
345
|
|
|
|
|
|
|
# print __PACKAGE__,"->frequency_distribution(",join(',',@_),")\n" if $DEBUG; |
346
|
|
|
|
|
|
|
|
347
|
|
|
|
|
|
|
# my $partitions = shift; #how many partitions (bins)? |
348
|
|
|
|
|
|
|
# my $minbin = shift; #upper bound of first bin |
349
|
|
|
|
|
|
|
# my $binsize = shift; #how wide is each bin? |
350
|
|
|
|
|
|
|
|
351
|
|
|
|
|
|
|
# #if partition == 0, then return the data hash |
352
|
|
|
|
|
|
|
# if (not defined $partitions || ($partitions == 0)) |
353
|
|
|
|
|
|
|
# { |
354
|
|
|
|
|
|
|
# $self->{frequency_partitions} = 0; |
355
|
|
|
|
|
|
|
# %{$self->{frequency}} = %{$self->{data}}; |
356
|
|
|
|
|
|
|
# return %{$self->{frequency}}; |
357
|
|
|
|
|
|
|
# } |
358
|
|
|
|
|
|
|
|
359
|
|
|
|
|
|
|
# #otherwise, partition better be >= 1 |
360
|
|
|
|
|
|
|
# return undef unless $partitions >= 1; |
361
|
|
|
|
|
|
|
|
362
|
|
|
|
|
|
|
# $self->_all_stats() if $self->{dirty}; #recompute stats if dirty, (so we have count) |
363
|
|
|
|
|
|
|
# return undef if $self->{count} < 2; #must have at least 2 values |
364
|
|
|
|
|
|
|
|
365
|
|
|
|
|
|
|
# #set up the bins |
366
|
|
|
|
|
|
|
# my ($interval, $iter, $max); |
367
|
|
|
|
|
|
|
# if (defined $minbin && defined $binsize) |
368
|
|
|
|
|
|
|
# { |
369
|
|
|
|
|
|
|
# $iter = $minbin; |
370
|
|
|
|
|
|
|
# $max = $minbin+$partitions*$binsize - $binsize; |
371
|
|
|
|
|
|
|
# $interval = $binsize; |
372
|
|
|
|
|
|
|
# $iter -= $interval; #so that loop that sets up bins works correctly |
373
|
|
|
|
|
|
|
# } |
374
|
|
|
|
|
|
|
# else |
375
|
|
|
|
|
|
|
# { |
376
|
|
|
|
|
|
|
# $iter = $self->{min}; |
377
|
|
|
|
|
|
|
# $max = $self->{max}; |
378
|
|
|
|
|
|
|
# $interval = $self->{sample_range}/$partitions; |
379
|
|
|
|
|
|
|
# } |
380
|
|
|
|
|
|
|
# my @k; |
381
|
|
|
|
|
|
|
# my %bins; |
382
|
|
|
|
|
|
|
# while (($iter += $interval) < $max) |
383
|
|
|
|
|
|
|
# { |
384
|
|
|
|
|
|
|
# $bins{$iter} = 0; |
385
|
|
|
|
|
|
|
# push @k, $iter; |
386
|
|
|
|
|
|
|
# } |
387
|
|
|
|
|
|
|
# $bins{$max} = 0; |
388
|
|
|
|
|
|
|
# push @k, $max; |
389
|
|
|
|
|
|
|
|
390
|
|
|
|
|
|
|
# VALUE: foreach my $val (keys %{$self->{data}}) |
391
|
|
|
|
|
|
|
# { |
392
|
|
|
|
|
|
|
# foreach my $k (@k) |
393
|
|
|
|
|
|
|
# { |
394
|
|
|
|
|
|
|
# if ($val <= $k) |
395
|
|
|
|
|
|
|
# { |
396
|
|
|
|
|
|
|
# $bins{$k} += $self->{data}{$val}; #how many of this value do we have? |
397
|
|
|
|
|
|
|
# next VALUE; |
398
|
|
|
|
|
|
|
# } |
399
|
|
|
|
|
|
|
# } |
400
|
|
|
|
|
|
|
# # } |
401
|
|
|
|
|
|
|
|
402
|
|
|
|
|
|
|
# %{$self->{frequency}} = %bins; #save it for later in case I add caching |
403
|
|
|
|
|
|
|
# $self->{frequency_partitions} = $partitions; #in case I add caching in the future |
404
|
|
|
|
|
|
|
# return %{$self->{frequency}}; |
405
|
|
|
|
|
|
|
# } |
406
|
|
|
|
|
|
|
|
407
|
|
|
|
|
|
|
sub frequency_distribution_ref |
408
|
|
|
|
|
|
|
{ |
409
|
12
|
|
|
12
|
1
|
1756
|
my $self = shift; |
410
|
12
|
|
|
|
|
23
|
my @k = (); |
411
|
|
|
|
|
|
|
|
412
|
|
|
|
|
|
|
# If called with no parameters, return the cached hashref |
413
|
|
|
|
|
|
|
# if we have one and data is not dirty |
414
|
|
|
|
|
|
|
# This is implemented this way because that's how Statistics::Descriptive |
415
|
|
|
|
|
|
|
# implements this. I don't like it. |
416
|
12
|
100
|
100
|
|
|
68
|
if ((!@_) && (! $self->{dirty}) && (defined $self->{_frequency})) |
|
|
|
66
|
|
|
|
|
417
|
|
|
|
|
|
|
{ |
418
|
3
|
|
|
|
|
29
|
return $self->{_frequency}; |
419
|
|
|
|
|
|
|
} |
420
|
|
|
|
|
|
|
|
421
|
9
|
100
|
|
|
|
35
|
$self->_all_stats() if $self->{dirty}; #recompute stats if dirty, (so we have count) |
422
|
|
|
|
|
|
|
|
423
|
|
|
|
|
|
|
# Must have at least two elements |
424
|
9
|
100
|
|
|
|
45
|
if ($self->count() < 2) |
425
|
|
|
|
|
|
|
{ |
426
|
1
|
|
|
|
|
5
|
return undef; |
427
|
|
|
|
|
|
|
} |
428
|
|
|
|
|
|
|
|
429
|
8
|
|
|
|
|
27
|
my %bins; |
430
|
8
|
|
|
|
|
14
|
my $partitions = shift; |
431
|
|
|
|
|
|
|
|
432
|
8
|
100
|
|
|
|
24
|
if (ref($partitions) eq 'ARRAY') |
433
|
|
|
|
|
|
|
{ |
434
|
4
|
|
|
|
|
8
|
@k = @{ $partitions }; |
|
4
|
|
|
|
|
13
|
|
435
|
4
|
50
|
|
|
|
14
|
return undef unless @k; ##Empty array |
436
|
4
|
50
|
|
|
|
12
|
if (@k > 1) { |
437
|
|
|
|
|
|
|
##Check for monotonicity |
438
|
4
|
|
|
|
|
10
|
my $element = $k[0]; |
439
|
4
|
|
|
|
|
24
|
for my $next_elem (@k[1..$#k]) { |
440
|
18
|
50
|
|
|
|
37
|
if ($element > $next_elem) { |
441
|
0
|
|
|
|
|
0
|
carp "Non monotonic array cannot be used as frequency bins!\n"; |
442
|
0
|
|
|
|
|
0
|
return undef; |
443
|
|
|
|
|
|
|
} |
444
|
18
|
|
|
|
|
34
|
$element = $next_elem; |
445
|
|
|
|
|
|
|
} |
446
|
|
|
|
|
|
|
} |
447
|
4
|
|
|
|
|
10
|
%bins = map { $_ => 0 } @k; |
|
22
|
|
|
|
|
60
|
|
448
|
|
|
|
|
|
|
} |
449
|
|
|
|
|
|
|
else |
450
|
|
|
|
|
|
|
{ |
451
|
4
|
100
|
66
|
|
|
30
|
return undef unless (defined $partitions) && ($partitions >= 1); |
452
|
3
|
|
|
|
|
17
|
my $interval = $self->sample_range() / $partitions; |
453
|
3
|
|
|
|
|
11
|
foreach my $idx (1 .. ($partitions-1)) |
454
|
|
|
|
|
|
|
{ |
455
|
2
|
|
|
|
|
11
|
push @k, ($self->min() + $idx * $interval); |
456
|
|
|
|
|
|
|
} |
457
|
|
|
|
|
|
|
|
458
|
3
|
|
|
|
|
15
|
$bins{$self->max()} = 0; |
459
|
|
|
|
|
|
|
|
460
|
3
|
|
|
|
|
14
|
push @k, $self->max(); |
461
|
|
|
|
|
|
|
} |
462
|
|
|
|
|
|
|
|
463
|
|
|
|
|
|
|
ELEMENT: |
464
|
7
|
|
|
|
|
16
|
foreach my $element (keys %{$self->{data}}) |
|
7
|
|
|
|
|
26
|
|
465
|
|
|
|
|
|
|
{ |
466
|
45
|
|
|
|
|
70
|
foreach my $limit (@k) |
467
|
|
|
|
|
|
|
{ |
468
|
109
|
100
|
|
|
|
222
|
if ($element <= $limit) |
469
|
|
|
|
|
|
|
{ |
470
|
44
|
|
|
|
|
107
|
$bins{$limit} += $self->{data}{$element}; |
471
|
44
|
|
|
|
|
80
|
next ELEMENT; |
472
|
|
|
|
|
|
|
} |
473
|
|
|
|
|
|
|
} |
474
|
|
|
|
|
|
|
} |
475
|
|
|
|
|
|
|
|
476
|
7
|
|
|
|
|
39
|
$self->{_frequency} = \%bins; |
477
|
7
|
|
|
|
|
30
|
return $self->{_frequency}; |
478
|
|
|
|
|
|
|
} |
479
|
|
|
|
|
|
|
|
480
|
|
|
|
|
|
|
sub frequency_distribution { |
481
|
3
|
|
|
3
|
1
|
403
|
my $self = shift; |
482
|
|
|
|
|
|
|
|
483
|
3
|
|
|
|
|
12
|
my $ret = $self->frequency_distribution_ref(@_); |
484
|
|
|
|
|
|
|
|
485
|
3
|
50
|
|
|
|
31
|
if (!defined($ret)) |
486
|
|
|
|
|
|
|
{ |
487
|
0
|
|
|
|
|
0
|
return undef; |
488
|
|
|
|
|
|
|
} |
489
|
|
|
|
|
|
|
else |
490
|
|
|
|
|
|
|
{ |
491
|
3
|
|
|
|
|
19
|
return %$ret; |
492
|
|
|
|
|
|
|
} |
493
|
|
|
|
|
|
|
} |
494
|
|
|
|
|
|
|
|
495
|
|
|
|
|
|
|
# return count of unique values in data if called in scalar context |
496
|
|
|
|
|
|
|
# returns sorted array of unique data values if called in array context |
497
|
|
|
|
|
|
|
# returns undef if no data |
498
|
|
|
|
|
|
|
sub uniq |
499
|
|
|
|
|
|
|
{ |
500
|
8
|
|
|
8
|
1
|
38
|
my $self = shift; |
501
|
|
|
|
|
|
|
|
502
|
8
|
100
|
|
|
|
22
|
if (!$self->{data}) |
503
|
|
|
|
|
|
|
{ |
504
|
2
|
|
|
|
|
8
|
return undef; |
505
|
|
|
|
|
|
|
} |
506
|
|
|
|
|
|
|
|
507
|
6
|
|
|
|
|
11
|
my @datakeys = sort {$a <=> $b} keys %{$self->{data}}; |
|
39
|
|
|
|
|
74
|
|
|
6
|
|
|
|
|
30
|
|
508
|
|
|
|
|
|
|
|
509
|
6
|
100
|
|
|
|
20
|
if (wantarray) |
510
|
|
|
|
|
|
|
{ |
511
|
2
|
|
|
|
|
9
|
return @datakeys; |
512
|
|
|
|
|
|
|
} |
513
|
|
|
|
|
|
|
else |
514
|
|
|
|
|
|
|
{ |
515
|
4
|
|
|
|
|
8
|
my $uniq = @datakeys; |
516
|
4
|
|
|
|
|
18
|
return $uniq; |
517
|
|
|
|
|
|
|
} |
518
|
|
|
|
|
|
|
} |
519
|
|
|
|
|
|
|
|
520
|
|
|
|
|
|
|
sub AUTOLOAD { |
521
|
82
|
|
|
82
|
|
8604
|
my $self = shift; |
522
|
82
|
50
|
|
|
|
219
|
my $type = ref($self) |
523
|
|
|
|
|
|
|
or croak "$self is not an object"; |
524
|
82
|
|
|
|
|
153
|
my $name = $AUTOLOAD; |
525
|
82
|
|
|
|
|
434
|
$name =~ s/.*://; ##Strip fully qualified-package portion |
526
|
82
|
100
|
|
|
|
545
|
return if $name eq "DESTROY"; |
527
|
65
|
50
|
|
|
|
162
|
unless (exists $self->{_permitted}{$name} ) { |
528
|
0
|
|
|
|
|
0
|
croak "Can't access `$name' field in class $type"; |
529
|
|
|
|
|
|
|
} |
530
|
|
|
|
|
|
|
|
531
|
65
|
50
|
|
|
|
130
|
print __PACKAGE__,"->AUTOLOAD $name\n" if $DEBUG; |
532
|
|
|
|
|
|
|
|
533
|
|
|
|
|
|
|
#compute stats if necessary |
534
|
65
|
100
|
|
|
|
195
|
$self->_all_stats() if $self->{dirty}; |
535
|
65
|
|
|
|
|
280
|
return $self->{$name}; |
536
|
|
|
|
|
|
|
} |
537
|
|
|
|
|
|
|
|
538
|
|
|
|
|
|
|
1; |
539
|
|
|
|
|
|
|
|
540
|
|
|
|
|
|
|
__END__ |