Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
xiongying 370c018046 | 2 years ago | |
---|---|---|
script | 2 years ago | |
source | 2 years ago | |
README.md | 2 years ago |
3实体
python run.py train -d 1 -- GPU编号
-c ./configs/config.ini -- 配置文件
--entity3 True -- 识别的实体类型
--model ./save/3entity.bin -- 保存模型
5实体
python run.py train -d 1 -- GPU编号
-c ./configs/config.ini -- 配置文件
--entity5 True -- 识别的实体类型
--model ./save/5entity.bin -- 保存模型
3实体
python run.py eval -d 1 -- GPU编号
-c ./configs/config.ini -- 配置文件
--entity3 True -- 识别的实体类型
--save_entity entity3.out -- NER预测文件
--save_relation relation3.out -- RE预测文件
--model ./save/3entity.bin -- 保存模型
5实体
python run.py eval -d 1 -- GPU编号
-c ./configs/config.ini -- 配置文件
--entity5 True -- 识别的实体类型
--save_entity entity5.out -- NER预测文件
--save_relation relation5.out -- RE预测文件
--model ./save/5entity.bin -- 保存模型
corpus:原始的数据
script:保存评测用到的预处理及后处理脚本
FH2019:祖传代码,后处理时使用,在上面改动了一些
src:祖传代码
parse_xml.py: 从xml文件中提取数据,预处理时使用
dnnfeature_extract.sh: 提取特征,
输入:目录,目录包含如下所示的分句后的文件,分句的脚本在祖传代码里,记不得是哪个了
输出:目录,输出的内容如下图所示
dealoutput_3entity_test.sh: 3实体后处理脚本,在python里调用
dealoutput_5entity_test.sh: 5实体后处理脚本,在python里调用
source:模型对应的代码
data: 数据部分
code: 模型代码部分
utils: 数据预处理
modules: 模型各部分
cmds: train或者eval
metric.py: 用于模型评估,使用F1值作为评价指标
model.py: 包括model.train()和model.eval()以及model.predict()三个函数
configs: 模型的配置部分
save: 模型的保存部分
config.py: 解析配置文件
run.py: 运行脚本
Zhan K, Peng W, Xiong Y, Fu H, Chen Q, Wang X, Tang B
Novel Graph-Based Model With Biaffine Attention for Family History Extraction From Clinical Text: Modeling Study
JMIR Med Inform 2021;9(4):e23587
URL: https://medinform.jmir.org/2021/4/e23587
DOI: 10.2196/23587
The model for 2019 N2C2 family history extraction track. Including named entity recognition and relation extraction.
Python Shell other
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》