-
Notifications
You must be signed in to change notification settings - Fork 0
/
config.yaml
81 lines (60 loc) · 1.66 KB
/
config.yaml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
############ options ####################
#generate_null_rule : yes
#comma_as_dead_span : yes
#use_mmap : yes
#remove_alignments: yes
# separate lm for lexical/normal fragment
#double_lm : yes
############ Rule Extraction ############
# source side with parsed dependency tree
# support format : stanford parser ( tag,base dependency )
file_source_tree : testdata/data.en.tree
file_source_dep : testdata/data.en.dep
# taget side with plain text
file_target : testdata/data.ja
# aligment file
file_alignment : testdata/aligned.grow-diag-final-and
# lexical probability table
file_lex_e2f : testdata/lex.e2f
file_lex_f2e : testdata/lex.f2e
# output file of rules extracted
rule_table_path : testdata/ruletables
# language model setting
file_lm : testdata/small.ja.lm
max_gram : 5
x_as_tag : no
max_merge_levels : 4
max_tokens : 6
min_deep_extract_terminals : 3
############ Probability Estimate ############
# input file of rules_extracted
dispersion_tables : 5
max_rules_for_each_source: 1000
############ Decoder ############
debug : no
#input file of rule table is same as file_rules_final
#input translation data
file_translation_input_tree : testdata/data.en.tree
file_translation_input_dep : testdata/data.en.dep
# output translation data
file_translation_output : testdata/out.ja
# the number of nbest in the mode of nbest output , make sure it bigger than 0;
nbest : 50
size_cube_pruning : 100
size_beam : 100
# max non-terminals in tree reconstruction
reconstruction_max_nt : 2
# weight settings
weights :
#Statistical Features (in ruletable):
#Pf2e Pe2f Frequency frag_penalty strength
- 1.0
- 1.0
- 1.0
- 1.0
- 1.0
#CONTEXTMATCHED
- 1.0
# languagemodel
- 1.0
###########