Generating 2-grams for all namespaces.
Generating 4-skips for all namespaces.
Num weight bits = 24
learning rate = 0.25
initial_t = 0
power_t = 0.5
using no cache
Reading datafile = train-sets/rcv1_raw_cb_small.vw
num sources = 1
average    since         example     example  current  current  current
loss       last          counter      weight    label  predict features
*estimate* *estimate*                                                avglossreg last pred  last correct
2.000000   2.000000          1      1.0    known        1      280   1.000000   0.000000   1.000000  
1.000000   0.000000          2      2.0    known        2      598   0.500501   0.031652   0.000000  
0.500000   0.000000          4      4.0    known        2      784   0.250859   0.026338   0.000000  
0.250000   0.000000          8      8.0    known        2      850   0.247080   0.075947   0.000000  
0.506238   0.762476         16     16.0    known        2      118   0.471313   0.072947   1.000000  
0.265628   0.025018         32     32.0    known        1      166   0.389623   0.043058   0.000000  
0.307697   0.349765         64     64.0    known        1      340   0.395963   0.222594   1.000000  
0.287462   0.267226        128    128.0    known        1      610   0.322881   0.222213   1.000000  
0.236404   0.185346        256    256.0    known        2      712   0.301560   0.518830   0.000000  
0.193768   0.151133        512    512.0    known        1      424   0.272979   0.302103   1.000000  
0.140346   0.086924       1024   1024.0    known        1      574   0.229385   0.760079   1.000000  
0.116006   0.091666       2048   2048.0    known        2      166   0.190320   0.209406   0.000000  
0.106417   0.096828       4096   4096.0    known        1      664   0.158365   0.348037   0.000000  
0.089500   0.072582       8192   8192.0    known        2      598   0.130558   0.812939   0.000000  

finished run
number of examples per pass = 10000
passes used = 1
weighted example sum = 10000
weighted label sum = 0
average loss = 0.0858454
best constant = 0
total feature number = 4476364