# TextING

**Repository Path**: magicat128/texting

## Basic Information

- **Project Name**: TextING
- **Description**: No description available
- **Primary Language**: Python
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2020-12-23
- **Last Updated**: 2020-12-24

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# TextING

The code for ACL2020 paper Every Document Owns Its Structure: Inductive Text Classification via Graph Neural Networks [(https://arxiv.org/abs/2004.13826)](https://arxiv.org/abs/2004.13826), implemented in Tensorflow on Ascend 910 environment.

# Usage 

Download the data `mr` from obs bucket https://dataset-mr.obs.dualstack.cn-north-4.myhuaweicloud.com

Start training and inference as:

```
python3 train.py \
	--dataset mr \
	--data_url PATH_TO_DATA \
	--train_url PATH_TO_DATA \
	--learning_rate 0.005 \
	--epochs 50 \
	--batch_size 1024 \
	--hidden 96
```

The reported result from the original paper is 79.8, and this implementation achieves around 79.4