yfletberliac / mango

Question-Answering NLP model with character-level RNN (TensorFlow).

Question Answering NLP with character-level RNN

The different models will first be challenged with the bAbI dataset from FAIR and the SQuAD dataset from Stanford.

Content:

Name	Description
/babi	babi dataset and utilities
/squad	squad dataset and utilities
/qrn	contains the QRN cell
char2word.py	Char2Word-only module (on bAbI dataset)
qrn.py	implementation of the QRN model (on bAbI dataset)
char2word_qrn.py	implementation of the QRN model w/ Char2Word module (on bAbI dataset)
mango_squad.py	implementation of the QRN model w/ Char2Word module (on SQuAD dataset)

Usage (with the bAbI dataset):

Query-Reduction Network with Char2Word:

python char2word_qrn.py

Query-Reduction Network without Char2Word:

python qrn.py

Usage (with the SQuAD dataset):

Query-Reduction Network with Char2Word:

python mango_squad.py