This repository contains the implementation of paper "Hierarchical Attentional Hybrid Neural Networks for Document Classification"
Branch: master
Clone or download
Latest commit 60b0ab4 Feb 6, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
input code of paper Jan 28, 2019
model Update hierarchical_attention_network.py Jan 29, 2019
saved_models
README.md Update README.md Feb 6, 2019
dataset.py code of paper Jan 28, 2019
train.py code updates Jan 28, 2019

README.md

Hierarchical Attentional Hybrid Neural Networks for Document Classification

J. Abreu , L. Fred, D. Macêdo, C. Zanchettin, "Hierarchical Attentional Hybrid Neural Networks for Document Classification", Submitted to IJCNN on 15 Jan, 2019.

Datasets:

Dataset Classes Documents source
Yelp Review Polarity 5 1569264 link
IMDb Movie Review 2 50000 link

To download datasets, install the kaggle tool:

pip install kaggle

then run follow commands:

kaggle datasets download -d luisfredgs/in1164-deep-learning

kaggle datasets download -d luisfredgs/wiki-news-300d-1m-subword

put all data on input/ folder

Requirements

  • tensorflow 1.10
  • Keras 2.x
  • gensim
  • tqdm
  • matplotlib

A GPU with CUDA support is required to run this code

Execution

run python train.py