Python code for training Paragram word embeddings. These achieve human-level performance on some word similiarty tasks including SimLex-999.This code was used to obtain results in the appendix of our 2015 TACL paper "From Paraphrase Database to Compositional Paraphrase Model and Back".
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
main
.gitignore
README.md
demo.sh
setup.sh
train.sh

README.md

paragram-word

Code to train Paragram word embeddings from the appendix of "From Paraphrase Database to Compositional Paraphrase Model and Back".

The code is written in python and requires numpy, scipy, theano and the lasagne library.

To get started, run setup.sh which will download the required files. Then run demo.sh to start training a model. Check main/train.py for command line options.

If you use our code for your work please cite:

@article{wieting2015ppdb, title={From Paraphrase Database to Compositional Paraphrase Model and Back}, author={John Wieting and Mohit Bansal and Kevin Gimpel and Karen Livescu and Dan Roth}, journal={Transactions of the ACL (TACL)}, year={2015}}