mirror of https://github.com/percyliang/sempre
ea2beaee7e
|
||
|---|---|---|
| data | ||
| demo-www | ||
| freebase | ||
| interactive | ||
| overnight | ||
| resources | ||
| scripts | ||
| src/main/java/edu/stanford/nlp/sempre | ||
| tables | ||
| unittest-files | ||
| .gitattributes | ||
| .gitignore | ||
| DOCUMENTATION.md | ||
| LICENSE.txt | ||
| README.md | ||
| TUTORIAL.md | ||
| build.xml | ||
| config.properties | ||
| pom.xml | ||
| pull-dependencies | ||
| quick_run.sh | ||
| run | ||
| testng.xml | ||
README.md
ROBOY SEMANTIC PARSER
Semantic Parser based on SEMPRE.
Installation
Requirements
You must have the following already installed on your system.
- Java 8 (not 7)
- Maven
- Ruby 1.8.7 or 1.9
- wget
- make (for compiling fig and Virtuoso)
- zip (for unzip downloaded dependencies)
Other dependencies will be downloaded as you need them. SEMPRE has been tested on Ubuntu Linux 12.04 and MacOS X. Your mileage will vary depending on how similar your system is.
Build
In order to build application with all needed dependencies run:
./pull-dependencies roboy
mvn clean
mvn install
Word2Vec Model
Small model for parser has to be manually downloaded from this link and put it in data/word2vec. Other models: google_model
Run
Roboy Talk Grammar and Lexicon
To run SEMPRE with Roboy grammar in interactive mode (from command line):
mvn exec:java@interactive -Dexec.mainClass=edu.stanford.nlp.sempre.Main
To run SEMPRE with Roboy grammar in socket mode (port 5000):
mvn exec:java@demo -Dexec.mainClass=edu.stanford.nlp.sempre.Main
To run SEMPRE with Roboy grammar in socket mode (port 5000) with training:
mvn exec:java@test -Dexec.mainClass=edu.stanford.nlp.sempre.Main
To run SEMPRE with Roboy grammar in web server mode:
mvn exec:java@debug -Dexec.mainClass=edu.stanford.nlp.sempre.Main