blob: 92b91e6494dece31811b855ae24841be1551df2c (
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
|
only testing
Num weight bits = 16
learning rate = 10
initial_t = 1
power_t = 0.5
using no cache
Reading datafile = test-sets/ml100k_small_test
num sources = 1
average since example example current current current
loss last counter weight label predict features
2.655687 2.655687 1 1.0 5.0000 3.3704 23
2.862807 3.069926 2 2.0 5.0000 3.2479 23
2.392509 1.922212 4 4.0 4.0000 3.2218 23
1.468514 0.544518 8 8.0 4.0000 3.2538 23
1.515712 1.562910 16 16.0 5.0000 3.1581 23
finished run
number of examples per pass = 30
passes used = 1
weighted example sum = 30
weighted label sum = 110
average loss = 1.4718
best constant = 3.66667
total feature number = 690
|