clothing_type_classifier_ruby/results at main · plartoo/clothing_type_classifier_ruby · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
fbt (x one out)
average: 0.798725217216277

nbt (x one out validation)
average: 0.784628803381529

fbt (five fold x validation)
average: 0.795256466167936


nbt (five fold x validation)
iter:	0 ==> 0.776678184440941
iter:	1 ==> 0.780468985876749
iter:	2 ==> 0.785098460196585
iter:	3 ==> 0.780905095776443
iter:	4 ==> 0.781542487168305
======
average: 0.780938642691804

***************
Nov 29
======

fbt
   -- use just one word per feature
iter:	0 ==> 0.70414201183432
iter:	1 ==> 0.715976331360947
iter:	2 ==> 0.727810650887574
iter:	3 ==> 0.775147928994083
iter:	4 ==> 0.680473372781065
======
average: 0.720710059171598

- (five fold with clothing_type_examples==36KB)
  -- include two words pairs as features
iter:	0 ==> 0.710059171597633
iter:	1 ==> 0.70414201183432
iter:	2 ==> 0.72189349112426
iter:	3 ==> 0.763313609467456
iter:	4 ==> 0.674556213017751
======
average: 0.714792899408284

   -- use size of class-diversity as denominator
iter:	0 ==> 0.698224852071006
iter:	1 ==> 0.751479289940828
iter:	2 ==> 0.710059171597633
iter:	3 ==> 0.763313609467456
iter:	4 ==> 0.650887573964497
======
average: 0.714792899408284

:::::::

x_one_out
   -- one word per feature
average: 0.799054373522459
   -- included both one and two words per feature
average: 0.822695035460993
   -- use size of class-diversity as denominator
average: 0.74468085106383

****************
phyothiha
screen -r data_collect