Skip to content
master
Switch branches/tags
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
 
 
 
 
 
 
 
 
 
 

PyAdaGram

PyAdaGram is an online Adaptor Grammar model package, developed by the Cloud Computing Research Team in University of Maryland, College Park. You may find more details about this project on our papaer Online Adaptor Grammars with Hybrid Inference appeared in TACL 2014.

Please download the latest version from our GitHub repository.

Please send any bugs of problems to Ke Zhai (kzhai@umd.edu).

Install and Build

This package depends on many external python libraries, such as numpy, scipy and nltk.

Launch and Execute

Assume the PyAdaGram package is downloaded under directory $PROJECT_SPACE/src/, i.e.,

$PROJECT_SPACE/src/PyAdaGram

To prepare the example dataset,

tar zxvf brent-phone.tar.gz

To launch PyAdaGram, first redirect to the directory of PyAdaGram source code,

cd $PROJECT_SPACE/src/PyAdaGram

and run the following command on example dataset,

python -m launch_train \
--input_directory=./brent-phone/ \
--output_directory=./ \
--grammar_file=./brent-phone/grammar.unigram \
--number_of_documents=9790 \
--batch_size=10

The generic argument to run PyAdaGram is

python -m launch_train \
--input_directory=$INPUT_DIRECTORY/$CORPUS_NAME \
--output_directory=$OUTPUT_DIRECTORY \
--grammar_file=$GRAMMAR_FILE \
--number_of_documents=$NUMBER_OF_DOCUMENTS \
--batch_size=$BATCH_SIZE

You should be able to find the output at directory $OUTPUT_DIRECTORY/$CORPUS_NAME.

Under any circumstances, you may also get help information and usage hints by running the following command

python -m launch_train --help

To launch test script, run the following command

python -m launch_test \
--input_directory=$DATA_DIRECTORY \
--model_directory=$MODEL_DIRECTORY \
--non_terminal_symbol=$NON_TERMINAL_SYMBOL

About

An Adaptor Grammar model implementation in Python.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages