Download Latest Version phr.zip (87.1 MB)
Email in envelope

Get an email when there's a new version of udop

Home / udop_en_c / get_rules
Name Modified Size InfoDownloads / Week
Parent folder
grammar1.txt 2011-02-27 14 Bytes
Val.java 2011-02-27 194 Bytes
Symbol.java 2011-02-27 236 Bytes
RuleSet.java 2011-02-27 1.3 kB
Rule.java 2011-02-27 226 Bytes
ProbRule.java 2011-02-27 3.6 kB
PCFGTerminalSymbol.java 2011-02-27 329 Bytes
PCFGSymbol.java 2011-02-27 34 Bytes
PCFGNonTerminalSymbol.java 2011-02-27 345 Bytes
Main.java 2011-02-27 4.5 kB
Key.java 2011-02-27 127 Bytes
Cyk.java 2011-02-27 9.5 kB
Totals: 12 Items   20.3 kB 0
1 
The files in the get_rules folder are to generate the PCFG rules "phr.mod" for the parser.
Change the path in the main.java to the folder containing PTB wsj data.
You can also unzip phr.zip to get "phr.mod".


2 compile
make


3 run
.\udop phr.mod wsj10

"wsj10" is the file that contains the sentences no greater than 10 english words in WSJ.
Source: readme.txt, updated 2011-02-27