b2txt25/language_model/srilm-1.7.3/lm/test/reference/ngram-count-maxent-unk.stderr

Counting counts of order 2
Counting counts of order 3
Starting fitting...
Starting OWL-BFGS with c1=1.56621e-07, sigma2=1.91546e+07, max_iters=1000
  No of NaNs in logZs: 0, No infs: 0
  dual is 10.4076
  regularized dual is 10.4076
  norm of gradient =0.123201
  norm of regularized gradient =0.123201
  No of NaNs in logZs: 0, No infs: 0
  dual is 10.2844
  regularized dual is 10.2844
  norm of gradient =0.123166
  norm of regularized gradient =0.123166
Iteration 1
  No of NaNs in logZs: 0, No infs: 414877
  dual is inf
  regularized dual is inf
  norm of gradient =0.0570462
  norm of regularized gradient =0.0569473
  No of NaNs in logZs: 0, No infs: 414877
  dual is inf
  regularized dual is inf
  norm of gradient =0.0950417
  norm of regularized gradient =0.0949613
  No of NaNs in logZs: 0, No infs: 0
  dual is 367.058
  regularized dual is 367.08
  norm of gradient =0.932065
  norm of regularized gradient =0.932086
  No of NaNs in logZs: 0, No infs: 0
  dual is 183.783
  regularized dual is 183.788
  norm of gradient =0.932065
  norm of regularized gradient =0.932076
  No of NaNs in logZs: 0, No infs: 0
  dual is 92.145
  regularized dual is 92.1464
  norm of gradient =0.932065
  norm of regularized gradient =0.932071
  No of NaNs in logZs: 0, No infs: 0
  dual is 46.3263
  regularized dual is 46.3266
  norm of gradient =0.932065
  norm of regularized gradient =0.932068
  No of NaNs in logZs: 0, No infs: 0
  dual is 23.4169
  regularized dual is 23.417
  norm of gradient =0.932059
  norm of regularized gradient =0.932061
  No of NaNs in logZs: 0, No infs: 0
  dual is 11.9711
  regularized dual is 11.9711
  norm of gradient =0.923145
  norm of regularized gradient =0.923145
  No of NaNs in logZs: 0, No infs: 0
  dual is 8.64134
  regularized dual is 8.64135
  norm of gradient =0.0945428
  norm of regularized gradient =0.0945422
Iteration 2
  No of NaNs in logZs: 0, No infs: 0
  dual is 34.1053
  regularized dual is 34.1066
  norm of gradient =0.706205
  norm of regularized gradient =0.706207
  No of NaNs in logZs: 0, No infs: 0
  dual is 18.608
  regularized dual is 18.6084
  norm of gradient =0.624414
  norm of regularized gradient =0.624416
  No of NaNs in logZs: 0, No infs: 0
  dual is 10.9824
  regularized dual is 10.9825
  norm of gradient =0.53305
  norm of regularized gradient =0.533051
  No of NaNs in logZs: 0, No infs: 0
  dual is 7.82864
  regularized dual is 7.82868
  norm of gradient =0.249817
  norm of regularized gradient =0.249817
Iteration 3
  No of NaNs in logZs: 0, No infs: 0
  dual is 7.45707
  regularized dual is 7.45711
  norm of gradient =0.0813196
  norm of regularized gradient =0.0813195
Iteration 4
  No of NaNs in logZs: 0, No infs: 0
  dual is 7.39068
  regularized dual is 7.39072
  norm of gradient =0.127317
  norm of regularized gradient =0.127317
Iteration 5
  No of NaNs in logZs: 0, No infs: 0
  dual is 7.1284
  regularized dual is 7.12844
  norm of gradient =0.0591887
  norm of regularized gradient =0.0591879
Iteration 6
  No of NaNs in logZs: 0, No infs: 0
  dual is 6.89141
  regularized dual is 6.89147
  norm of gradient =0.0589822
  norm of regularized gradient =0.0589816
Iteration 7
  No of NaNs in logZs: 0, No infs: 0
  dual is 6.65125
  regularized dual is 6.65134
  norm of gradient =0.0819502
  norm of regularized gradient =0.0819505
Iteration 8
  No of NaNs in logZs: 0, No infs: 0
  dual is 6.43328
  regularized dual is 6.43337
  norm of gradient =0.045384
  norm of regularized gradient =0.0453839
Iteration 9
  No of NaNs in logZs: 0, No infs: 0
  dual is 6.30993
  regularized dual is 6.31004
  norm of gradient =0.046049
  norm of regularized gradient =0.0460488
Iteration 10
  No of NaNs in logZs: 0, No infs: 0
  dual is 6.15737
  regularized dual is 6.1575
  norm of gradient =0.0466352
  norm of regularized gradient =0.0466349
Iteration 11
  No of NaNs in logZs: 0, No infs: 0
  dual is 6.06057
  regularized dual is 6.06071
  norm of gradient =0.0313801
  norm of regularized gradient =0.0313797
Iteration 12
  No of NaNs in logZs: 0, No infs: 0
  dual is 5.90965
  regularized dual is 5.90982
  norm of gradient =0.0332189
  norm of regularized gradient =0.0332188
Iteration 13
  No of NaNs in logZs: 0, No infs: 0
  dual is 5.84861
  regularized dual is 5.8488
  norm of gradient =0.0534722
  norm of regularized gradient =0.0534723
Iteration 14
  No of NaNs in logZs: 0, No infs: 0
  dual is 5.78141
  regularized dual is 5.78161
  norm of gradient =0.0229882
  norm of regularized gradient =0.0229883
Iteration 15
  No of NaNs in logZs: 0, No infs: 0
  dual is 5.68465
  regularized dual is 5.6849
  norm of gradient =0.0227859
  norm of regularized gradient =0.0227859
Iteration 16
  No of NaNs in logZs: 0, No infs: 0
  dual is 5.59792
  regularized dual is 5.5982
  norm of gradient =0.0360479
  norm of regularized gradient =0.0360477
Iteration 17
  No of NaNs in logZs: 0, No infs: 0
  dual is 5.52285
  regularized dual is 5.52317
  norm of gradient =0.0454409
  norm of regularized gradient =0.045441
Iteration 18
  No of NaNs in logZs: 0, No infs: 0
  dual is 5.46329
  regularized dual is 5.46365
  norm of gradient =0.0163726
  norm of regularized gradient =0.0163723
Iteration 19
  No of NaNs in logZs: 0, No infs: 0
  dual is 5.38658
  regularized dual is 5.387
  norm of gradient =0.0162138
  norm of regularized gradient =0.016213
Iteration 20
  No of NaNs in logZs: 0, No infs: 0
  dual is 5.32452
  regularized dual is 5.325
  norm of gradient =0.0218967
  norm of regularized gradient =0.0218962
Iteration 21
  No of NaNs in logZs: 0, No infs: 0
  dual is 5.23473
  regularized dual is 5.23536
  norm of gradient =0.0258103
  norm of regularized gradient =0.0258101
Iteration 22
  No of NaNs in logZs: 0, No infs: 0
  dual is 5.20341
  regularized dual is 5.20412
  norm of gradient =0.031137