为什么我训练的特别慢 总是卡在 partII 就不动了 怎么办

[autopost-get]

art I: train segmenter
words in dictionary: 200000
num features: 271
now do training
C: 20
epsilon: 0.01
num threads: 1
cache size: 5
max iterations: 2000
loss per missed segment: 3
C: 20 loss: 3 0.795918
C: 35 loss: 3 0.784257
C: 20 loss: 4.5 0.804665
C: 5 loss: 3 0.790087
C: 20 loss: 1.5 0.74344
C: 17.5 loss: 4.05 0.803207
C: 20 loss: 4.8 0.80758
C: 16.7825 loss: 4.99261 0.816327
C: 10.769 loss: 5.44081 0.816327
C: 18.1567 loss: 5.22501 0.819242
C: 20.1353 loss: 5.54356 0.822157
C: 25.3591 loss: 6.08173 0.816327
C: 20.3988 loss: 5.69292 0.822157
C: 17.6169 loss: 5.82141 0.819242
C: 21.7838 loss: 5.45424 0.814869
C: 19.1881 loss: 5.57563 0.8207
best C: 20.1353
best loss: 5.54356
num feats in chunker model: 4095
train: precision, recall, f1-score: 0.897574 0.970845 0.932773
Part I: elapsed time: 269 seconds.

Part II: train segment classifier
now do training
num training samples: 762

到这就不走了 好苦恼

[crownpku]

请参考 #13 后面的讨论。
暂时还没有好的解决方法。。

[libindavis]

please check this comment mit-nlp/MITIE#11 (comment)