Loading the LM will be faster if you build a binary file.
Reading ../standard/nc-wmt11.en.srilm.gz
----5---10---15---20---25---30---35---40---45---50---55---60---65---70---75---80---85---90---95--100
****************************************************************************************************
dtrain
Parameters:
                       k 100
                       N 4
                       T 1
           learning rate 0.0001
            error margin 1
                  l1 reg 0
            decoder conf 'cdec.ini'
                   input 'work/shard.2.0.in'
                  output 'work/weights.2.2'
              weights in 'work/weights.1'
(a dot per input)
Iteration #1 of 1.
 .... 3
WEIGHTS
              Glue = -0.90863
       WordPenalty = +0.10819
     LanguageModel = +0.5239
 LanguageModel_OOV = -0.41623
     PhraseModel_0 = -0.86868
     PhraseModel_1 = +0.40784
     PhraseModel_2 = +1.1793
     PhraseModel_3 = -0.24698
     PhraseModel_4 = -1.2353
     PhraseModel_5 = +0.03375
     PhraseModel_6 = -0.17883
       PassThrough = -0.44638
        ---
       1best avg score: 0.12788 (+0.12788)
 1best avg model score: 41.302
           avg # pairs: 2246.3
   non-0 feature count: 12
           avg list sz: 100
           avg f count: 10.98
(time 0.35 min, 7 s/S)

---
Best iteration: 1 [GOLD = 0.12788].
This took 0.35 min.