# RSN

**Repository Path**: thunlp/RSN

## Basic Information

- **Project Name**: RSN
- **Description**: No description available
- **Primary Language**: Python
- **License**: MIT
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2020-05-29
- **Last Updated**: 2020-12-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# OpenRE
The source code of Relational Siamese Network (EMNLP2019)<br>
Future up-to-date versions will be available in:
https://github.com/thunlp/RSN

# Environment
You only need TensorFlow(>=1.12.0) to run this code.<br>
`conda create -n rsn python=3.6`<br>
`pip install networkx python-louvain tensorflow-gpu==1.12`<br>

# Data
You can download data by:<br>
`bash download.sh`<br>
The data is already preprocessed from the original FewRel and Glove data.

# Running CNN
A standard CNN for supervised RE can be trained by<br>
`cd CNN`<br>
`python train_CNN.py`<br>

# Running RSN
The Relational Siamese Network for OpenRE can be trained by<br>
`cd RSN`<br>
`python train_RSN.py`<br>
By default it will be trained as a semi-supervised RSN.<br>

A supervised RSN can be trained by<br>
`python train_RSN.py --trainset_loss_type cross --testset_loss_type none`<br>

# Future Work
The FewRel-distant dataset and more models will be released in the future.<br>