# PLOME

**Repository Path**: weigy/PLOME

## Basic Information

- **Project Name**: PLOME
- **Description**: No description available
- **Primary Language**: Python
- **License**: Apache-2.0
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2021-12-24
- **Last Updated**: 2021-12-24

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# PLOME:Pre-training with Misspelled Knowledge for Chinese Spelling Correction (ACL2021)
This repository provides the code and data of the work in ACL2021: *PLOME: Pre-training with Misspelled Knowledge for Chinese Spelling Correction* https://aclanthology.org/2021.acl-long.233.pdf

**Requirements:**

- python3

- tensorflow1.14

- horovod

**Instructions:**

- Finetune: 

   train and evaluation file format: original sentence \t golden sentence 
   ```bash
   step1: cd finetune_src ; 
   step2: download the pretrained PLOME model and corpus from https://drive.google.com/file/d/1aip_siFdXynxMz6-2iopWvJqr5jtUu3F/view?usp=sharing ;
   step3: sh start.sh
   ```
   
 - Pre-train
   ```bash
   step1: cd pre_train_src ;
   step2: sh gen_train_tfrecords.sh ;
   step3: sh start.sh
   ```
   Our pre-trained model: https://drive.google.com/file/d/1aip_siFdXynxMz6-2iopWvJqr5jtUu3F/view?usp=sharing
   
   pre-trained cBERT model: https://drive.google.com/file/d/1cqSTpn7r9pnDcvMoM3BbX1X67JsPdZ8_/view?usp=sharing
   
   国内下载地址：https://share.weiyun.com/OREEY0H3