# Conflicts:
#	fastNLP/modules/encoder/embedding.py
#	reproduction/seqence_labelling/ner/train_ontonote.py
#	reproduction/text_classification/model/lstm.py

2. sequence labeling中增加BiLSTMCRF
3. embedding中修复bug
4. 将之前版本的reproduction移动到legacy文件夹
5. sequence labeling中增加中文NER
6. text classification中loader增强，修改lstm模型，增加train_bert.py

-use spacy tokenizer for yelp data
-add set_rng_seed

[add] HAN train_HAN

LSTM、AWDLSTM和LSTM+self attention三个模型

[add] model: char_cnn dpcnn
[test] train_char_cnn train_dpcnn,dataloader的均在自己的main方法内测试

-use spacy tokenizer for yelp data
-add set_rng_seed