Table 3 Comparisons of perplexities on test set of five million training corpus with different sizes of class layer
From: RNN language model with word clustering and class-based output layer
Class | RNNLM-Freq / +KN5 | RNNLM-Brown / +KN5 |
|---|---|---|
50 | 218.13 / 178.10 | 206.19 / 172.08 |
100 | 220.47 / 178.50 | 208.37 / 172.06 |
200 | 219.60 / 178.21 | 206.54 / 171.21 |
400 | 219.73 / 178.09 | 205.02 / 170.56 |