https://github.com/mattbui/sent-comp

Sentence Compression with deletion, accepted at ICCCI
https://github.com/mattbui/sent-comp

Last synced: about 1 month ago
JSON representation

Sentence Compression with deletion, accepted at ICCCI

• Host: GitHub
• URL: https://github.com/mattbui/sent-comp
• Owner: mattbui
• Created: 2021-01-04T12:48:40.000Z (about 4 years ago)
• Default Branch: master
• Last Pushed: 2021-01-04T13:49:36.000Z (about 4 years ago)
• Last Synced: 2024-10-30T02:54:31.816Z (2 months ago)
• Language: Python
• Homepage:
• Size: 9.77 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# ICCCI - Sentence Compression with deletion

## Dataset

The dataset is available at: [https://github.com/google-research-datasets/sentence-compression](https://github.com/google-research-datasets/sentence-compression). Download and store the `*.gz` files in `data/` directory.

## Requirements

This project requires python3.6+ and pytorch1.1+. It used the models and embeddings from [FLAIR framework](https://github.com/flairNLP/flairhttps://github.com/flairNLP/flair):

```bash

pip install flair

```

## Preprocess data

In order to train a sequence tagging model, the original data need to be align into sequence tagging format. To align the downloaded data:

```bash

export PRJ_HOME=

bash $PRJ_HOME/runs/preprocess.sh

```

## Training

Different training configs for each settings are available in `runs/`. To start training:

```bash

export PRJ_HOME=

bash $PRJ_HOME/runs/train_.sh

```