# Chinese sentence sentiment classification ## Get model files and sample data ``` sh get_data.sh ``` ## Start http service ``` python senta_web_service.py senta_bilstm_model/ workdir 9292 ``` ## Client prediction ```
curl -H "Content-Type:application/json" -X POST -d '{"words": "天气不错 | 0", "fetch":["sentence_feature"]}' http://127.0.0.1:9292/senta/prediction
```