← Back to branch summary

~ubuntu-branches/ubuntu/vivid/vowpal-wabbit/vivid

« back to all changes in this revision

Viewing changes to test/test-sets/ref/0002c.stderr

  • Committer: Package Import Robot
  • Author(s): Yaroslav Halchenko
  • Date: 2013-08-27 20:52:23 UTC
  • mfrom: (1.2.1) (7.1.2 experimental)
  • Revision ID: package-import@ubuntu.com-20130827205223-q005ps71tqinh25v
Tags: 7.3-1
New upstream release

Show diffs side-by-side

added added

removed removed

Lines of Context:
test/test-sets/ref/0002c.stderr
1
 
using no cache
2
 
Reading from train-sets/0002.dat
3
 
num sources = 1
 
1
only testing
4
2
Num weight bits = 18
5
3
learning rate = 10
6
4
initial_t = 1
7
5
power_t = 0.5
8
6
predictions = 0002c.predict
9
 
only testing
10
 
average    since       example  example    current  current  current
11
 
loss       last        counter   weight      label  predict features
12
 
0.002276   0.002276          3      3.0     0.5498   0.5361      184
13
 
0.031357   0.060437          6      6.0     0.2681   0.6668      184
14
 
0.027939   0.023839         11     11.0     0.4315   0.6007      184
15
 
0.042069   0.056199         22     22.0     0.5519   0.4997      184
16
 
0.025585   0.009101         44     44.0     0.5514   0.5331      184
17
 
0.024013   0.022405         87     87.0     0.5140   0.5308      184
18
 
0.018521   0.013029        174    174.0     0.5596   0.4972      184
19
 
0.016111   0.013701        348    348.0     0.5475   0.4388      184
20
 
0.015105   0.014098        696    696.0     0.3421   0.7898      184
21
 
0.014822   0.014539       1392   1392.0     0.4996   0.5059      184
22
 
0.013016   0.011211       2784   2784.0     0.5090   0.3869      184
23
 
0.012228   0.011440       5568   5568.0     0.6413   0.7550      184
24
 
0.011123   0.010018      11135  11135.0     0.3869   0.4848      184
25
 
0.011121   0.011119      22269  22269.0     0.5063   0.4510      184
26
 
0.015806   0.020491      44537  44537.0     0.4905   0.5220      184
 
7
using no cache
 
8
Reading datafile = train-sets/0002.dat
 
9
num sources = 1
 
10
average    since         example     example  current  current  current
 
11
loss       last          counter      weight    label  predict features
 
12
0.003238   0.003238            3         3.0   0.5498   0.4859       15
 
13
0.023991   0.044745            6         6.0   0.2681   0.5733       15
 
14
0.025514   0.027340           11        11.0   0.4315   0.5098       15
 
15
0.039356   0.053198           22        22.0   0.5519   0.4656       15
 
16
0.029043   0.018730           44        44.0   0.5514   0.5175       15
 
17
0.026338   0.023570           87        87.0   0.5140   0.4811       15
 
18
0.023165   0.019991          174       174.0   0.5596   0.4556       15
 
19
0.018138   0.013112          348       348.0   0.5475   0.4087       15
 
20
0.014870   0.011601          696       696.0   0.3421   0.7525       15
 
21
0.013755   0.012641         1392      1392.0   0.4996   0.4691       15
 
22
0.011204   0.008653         2784      2784.0   0.5090   0.4431       15
 
23
0.009764   0.008324         5568      5568.0   0.6413   0.7044       15
 
24
0.008678   0.007592        11135     11135.0   0.3869   0.4372       15
 
25
0.008074   0.007470        22269     22269.0   0.5063   0.4477       15
 
26
0.011837   0.015601        44537     44537.0   0.4905   0.3856       15
27
27
 
28
28
finished run
29
29
number of examples = 74746
30
30
weighted example sum = 6.952e+04
31
31
weighted label sum = 3.511e+04
32
 
average loss = 0.01421
 
32
average loss = 0.01063
33
33
best constant = 0.5051
34
34
best constant's loss = 0.25
35
 
total feature number = 13753264
 
35
total feature number = 1119986