You have chosen to generate 3-grams
You have chosen to generate 1-skip-3-grams
final_regressor = models/0001.model
Num weight bits = 18
learning rate = 2.56e+06
initial_t = 128000
power_t = 1
decay_learning_rate = 1
creating cache_file = train-sets/0001.dat.cache
Reading from train-sets/0001.dat
num sources = 1
average    since         example     example  current  current  current
loss       last          counter      weight    label  predict features
0.667135   0.667135            3         3.0   0.0000   0.0375      326
0.500234   0.333333            6         6.0   1.0000   0.0000      170
0.441085   0.370105           11        11.0   0.0000   0.9034      212
0.430364   0.419643           22        22.0   1.0000   0.2339      512
0.470963   0.511563           44        44.0   0.0000   1.0000      380
0.418930   0.365687           87        87.0   1.0000   0.0000      350
0.381112   0.343295          174       174.0   1.0000   0.3448      140
0.253458   0.125803          348       348.0   1.0000   1.0000      152
0.138076   0.022694          696       696.0   0.0000   0.0000      296
0.069038   0.000000         1392      1392.0   0.0000   0.0000       62

finished run
number of examples = 1600
weighted example sum = 1600
weighted label sum = 728
average loss = 0.06006
best constant = 1.007
total feature number = 717536