Source code for the paper "CRASpell: A Contextual Typo Robust Approach to Improve Chinese Spelling Correction" in ACL2022 findings https://aclanthology.org/2022.findings-acl.237.pdf .
-python 3.7
-tensorflow 1.14
Step1: Download the pretrained cBERT from https://drive.google.com/file/d/1cqSTpn7r9pnDcvMoM3BbX1X67JsPdZ8_/view?usp=sharing (our previous work),
and save it in ./datas/init_bert/cbert
Step2: Run the training script: sh start_train.sh
The best model will be saved when it is finished.
Step3: Run the evaluation script to obtain the results on whole set and multi-typo set, respectively:
sh start_eval.sh sighan15_test.sh
sh start_eval.sh sighan15_multierror.txt
sh start_eval.sh sighan14_test.sh
sh start_eval.sh sighan14_multierror.txt