Skip to content

Question Answering on bAbI using a single RNN

Notifications You must be signed in to change notification settings

andri27-ts/bAbI

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 

Repository files navigation

bAbI

This repository contains a simple Question Answering algorithm developed in Tensorflow. We'll use the bAbI tasks 1-20, created by facebook, for testing the capacity of our algorithm to understand and reason the text.

It uses only a single GRU layer but perform better than the Facebook LSTM Baseline and Keras babi_rnn

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.

Prerequisites

In order to run the notebook you need to install tensorflow and tqdm (to take progress).

$ pip install tensorflow --upgrade
$ pip install tqdm

Installing

Next, install git if you don't have already it and run the following commands on your terminal

$ cd $HOME  # or any other development directory you prefer
$ git clone 
$ cd bAbI

Tensorboard

To keep track of the progress and to visualize the histograms of the NN layers, use Tensorboard, running on your terminal

$ cd $HOME  # or any other development directory you prefer
$ cd bAbI
$ tensorboard --logdir tf_logs

Contributing

Please let me know if you need more information or if you want to contribute in any way