Num weight bits = 18
learning rate = 0.5
initial_t = 0
power_t = 0.5
using no cache
Reading datafile = train-sets/0001.dat
num sources = 1
average    since         example     example  current  current  current
loss       last          counter      weight    label  predict features
0.251938   0.251938           10        10.0   1.0000   0.2253       34
0.235272   0.218607           20        20.0   0.0000   0.2122      104
0.240930   0.252246           30        30.0   0.0000   0.3390       82
0.229258   0.194240           40        40.0   1.0000   0.5489       42
0.225418   0.210062           50        50.0   0.0000   0.2070       60
0.232323   0.266844           60        60.0   0.0000   0.3381      147
0.229975   0.215893           70        70.0   1.0000   0.4733      134
0.226344   0.200927           80        80.0   0.0000   0.2216      136
0.217409   0.145923           90        90.0   0.0000   0.2602      139
0.216703   0.210351          100       100.0   1.0000   0.3171       56
0.218356   0.234887          110       110.0   1.0000   0.5796       97
0.226824   0.319970          120       120.0   0.0000   0.4137      120
0.223091   0.178299          130       130.0   1.0000   0.4105       54
0.218306   0.156103          140       140.0   0.0000   0.3486       82
0.212760   0.135116          150       150.0   1.0000   0.4022      148
0.211695   0.195720          160       160.0   0.0000   0.6071       63
0.206603   0.125119          170       170.0   1.0000   0.7091       69
0.202214   0.127603          180       180.0   1.0000   0.7582       42
0.198753   0.136460          190       190.0   1.0000   0.7647       34
0.195760   0.138892          200       200.0   1.0000   0.5244       56

finished run
number of examples per pass = 200
passes used = 1
weighted example sum = 200
weighted label sum = 91
average loss = 0.19576
best constant = 0.455
best constant's loss = 0.247975
total feature number = 15482