Difference between revisions of "Lexical selection module scripts reference"

From Apertium
Jump to navigation Jump to search
(cat)
 
(11 intermediate revisions by one other user not shown)
Line 1: Line 1:
[[ apertium-lex-evaluate.py ]]
apertium-lex-evaluate.py <br/>
input:
[[ biltrans-count-patterns-frac-maxent.py ]]
- src = output of translator up to lt-proc -b
[[ biltrans-count-patterns-frac.py ]]
- ref = reference corpus
[[ biltrans-count-patterns-me.py ]]
- tst = output of lexical selection module
[[ biltrans-count-patterns.py ]]
output:
[[ biltrans-extract-frac-freq.py ]]
-
[[ biltrans-extract-freq.py ]]

[[ biltrans-extract-poormans-freq.py ]]
[[ biltrans-line-only-pos-ambig.py ]]
biltrans-count-patterns-frac-maxent.py <br/>

[[ biltrans-no-retained.py ]]
[[ biltrans-only-retained.py ]]
biltrans-count-patterns-frac.py <br/>
input:
[[ biltrans-to-multitrans-line-recursive.py ]]
- frequency lexicon
[[ biltrans-to-multitrans.py ]]
- biltrans output
[[ biltrans-trim-uncovered.py ]]
output:
[[ eval-corp-from-alig.py ]]
-
[[ extract-alig-lrx.py ]]

[[ extract-biltrans-cand.py ]]
biltrans-count-patterns-me.py <br/>
[[ extract-default-ldx.py ]]
biltrans-count-patterns.py <br/>
[[ extract-default-lrx.py ]]
input:
[[ extract-freq-lexicon.py ]]
- frequency lexicon
[[ extract-sentences.py ]]
- biltrans output
[[ giza-to-moses.awk ]]
- disambiguated biltrans output
[[ lrx2cg.py ]]
- crispiness threshold
[[ merge-ngrams-lambdas.py ]]
output:
[[ ngram-count-patterns-maxent.py ]]
-
[[ ngram-count-patterns.py ]]

[[ ngrams-to-rules.py ]]
[[ process-biltrans-output.py ]]
biltrans-extract-frac-freq.py <br/>
input:
[[ process-tagger-output_2.py ]]
- biltrans output
[[ process-tagger-output.py ]]
- disambiguated biltrans output
[[ split-sentences.py ]]
output:
[[ trim-fertile-lines.py ]]
-

biltrans-extract-freq.py <br/>
input:
- biltrans output
- disambiguated biltrans output
output:
-

biltrans-extract-poormans-freq.py <br/>
biltrans-line-only-pos-ambig.py <br/>
biltrans-no-retained.py <br/>
biltrans-only-retained.py <br/>

biltrans-to-multitrans-line-recursive.py
biltrans-to-multitrans.py <br/>
input:
- ambiguous biltrans output

biltrans-trim-uncovered.py <br/>
eval-corp-from-alig.py <br/>

extract-alig-lrx.py <br/>
extract-biltrans-cand.py <br/>
extract-default-ldx.py <br/>
extract-default-lrx.py <br/>
extract-freq-lexicon.py <br/>
extract-sentences.py <br/>
giza-to-moses.awk <br/>
lrx2cg.py <br/>
merge-ngrams-lambdas.py <br/>
ngram-count-patterns-maxent.py <br/>
ngram-count-patterns.py <br/>
ngrams-to-rules.py <br/>

process-biltrans-output.py <br/>
process-tagger-output_2.py <br/>
input:
- tagger output
- bidix path
output:
- tagger output with trimmed tags

process-tagger-output.py <br/>
split-sentences.py <br/>
trim-fertile-lines.py <br/>

[[Category:Lexical selection]]

Latest revision as of 12:32, 29 April 2015

apertium-lex-evaluate.py 
input: - src = output of translator up to lt-proc -b - ref = reference corpus - tst = output of lexical selection module output: -
biltrans-count-patterns-frac-maxent.py 
biltrans-count-patterns-frac.py 
input: - frequency lexicon - biltrans output output: -
biltrans-count-patterns-me.py 
biltrans-count-patterns.py
input: - frequency lexicon - biltrans output - disambiguated biltrans output - crispiness threshold output: -
biltrans-extract-frac-freq.py 
input: - biltrans output - disambiguated biltrans output output: -
biltrans-extract-freq.py 
input: - biltrans output - disambiguated biltrans output output: -
biltrans-extract-poormans-freq.py 
biltrans-line-only-pos-ambig.py
biltrans-no-retained.py
biltrans-only-retained.py
biltrans-to-multitrans-line-recursive.py 
biltrans-to-multitrans.py 
input: - ambiguous biltrans output
biltrans-trim-uncovered.py 
eval-corp-from-alig.py
extract-alig-lrx.py 
extract-biltrans-cand.py
extract-default-ldx.py
extract-default-lrx.py
extract-freq-lexicon.py
extract-sentences.py
giza-to-moses.awk
lrx2cg.py
merge-ngrams-lambdas.py
ngram-count-patterns-maxent.py
ngram-count-patterns.py
ngrams-to-rules.py
process-biltrans-output.py 
process-tagger-output_2.py
input: - tagger output - bidix path output: - tagger output with trimmed tags
process-tagger-output.py 
split-sentences.py
trim-fertile-lines.py