blob: 7cebc9449295baf1f15ec9c8ffa21d4eb47f1c8b (
plain)
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
|
creating cubic features for triples: abc
Num weight bits = 18
learning rate = 0.5
initial_t = 0
power_t = 0.5
decay_learning_rate = 1
final_regressor = models/xxor.model
creating cache_file = train-sets/xxor.dat.cache
Reading datafile = train-sets/xxor.dat
num sources = 1
average since example example current current current
loss last counter weight label predict features
1.000000 1.000000 1 1.0 1.0000 0.0000 5
0.581525 0.163049 2 2.0 0.0000 0.4038 5
0.422704 0.105064 3 3.0 0.0000 0.3241 5
0.567028 1.000000 4 4.0 1.0000 0.0000 5
0.527970 0.449852 6 6.0 1.0000 0.1235 5
0.493858 0.391523 8 8.0 0.0000 0.6019 5
0.407022 0.175458 11 11.0 0.0000 0.4429 5
0.361109 0.234848 15 15.0 1.0000 0.5645 5
0.304274 0.133770 20 20.0 1.0000 0.5956 5
0.249203 0.091857 27 27.0 0.0000 0.2644 5
0.201723 0.059283 36 36.0 1.0000 0.7710 5
0.158986 0.030775 48 48.0 0.0000 0.1478 5
0.122699 0.013839 64 64.0 0.0000 0.0883 5
0.092576 0.004943 86 86.0 1.0000 0.9496 5
0.069532 0.001197 115 115.0 0.0000 0.0223 5
0.051971 0.000188 154 154.0 0.0000 0.0084 5
0.038856 0.000016 206 206.0 1.0000 0.9988 5
0.029107 0.000001 275 275.0 0.0000 0.0002 5
0.021811 0.000000 367 367.0 1.0000 1.0000 5
0.016336 0.000000 490 490.0 0.0000 0.0000 5
0.012239 0.000000 654 654.0 1.0000 1.0000 5
finished run
number of examples per pass = 8
passes used = 100
weighted example sum = 800
weighted label sum = 400
average loss = 0.0100056
best constant = 0.5
best constant's loss = 0.25
total feature number = 4000
|
