“RNN test”版本间的差异
来自cslt Wiki
（→wsj_data） |
（→wsj_data） |
||
| 第13行： | 第13行： | ||
bptt=2 # length of BPTT unfolding in RNNLM | bptt=2 # length of BPTT unfolding in RNNLM | ||
bptt_block=20 # length of BPTT unfolding in RNNLM | bptt_block=20 # length of BPTT unfolding in RNNLM | ||
| + | |||
| + | {| border="2px" | ||
| + | |+ Train Set Environment | ||
| + | |- | ||
| + | ! Parameters !! hidden !! class !! direct !! bbt !! bptt_block !! threads | ||
| + | |- | ||
| + | !set1 | ||
| + | | 320 || 300 || 2000 || 2 || 20 || 1 | ||
== daily work == | == daily work == | ||
[[ 140902 ]] | [[ 140902 ]] | ||
2014年9月5日 (五) 01:18的版本
140901
wsj_data
1.parameter
rand_seed=1
nwords=10000 # This is how many words we're putting in the vocab of the RNNLM.
hidden=320
class=300 # Num-classes... should be somewhat larger than sqrt of nwords.
direct=2000 # Number of weights that are used for "direct" connections, in millions.
rnnlm_ver=rnnlm-0.3e # version of RNNLM to use
threads=1 # for RNNLM-HS
bptt=2 # length of BPTT unfolding in RNNLM
bptt_block=20 # length of BPTT unfolding in RNNLM
| Parameters | hidden | class | direct | bbt | bptt_block | threads |
|---|---|---|---|---|---|---|
| set1 | 320 | 300 | 2000 | 2 | 20 | 1
daily work |
