====================================================================================================
模型：Bi_LSTM+CRF，test结果如下：
precision recall f1-score support
E-ORG 0.9372 0.9168 0.9269 553
B-NAME 1.0000 0.9292 0.9633 113
M-CONT 1.0000 1.0000 1.0000 53
B-TITLE 0.9374 0.9313 0.9344 772
M-TITLE 0.9509 0.8965 0.9229 1922
E-TITLE 0.9921 0.9793 0.9857 772
E-RACE 1.0000 1.0000 1.0000 14
M-ID 1.0000 1.0000 1.0000 16
B-LOC 0.8571 0.8571 0.8571 7
E-PRO 0.9118 0.9394 0.9254 33
B-CONT 1.0000 1.0000 1.0000 28
B-PRO 0.9375 0.9091 0.9231 33
M-EDU 0.9767 0.9385 0.9573 179
E-CONT 1.0000 1.0000 1.0000 28
B-RACE 1.0000 0.9286 0.9630 14
B-ORG 0.9587 0.9656 0.9622 553
M-NAME 0.9425 0.9880 0.9647 83
E-EDU 0.9732 0.9732 0.9732 112
E-LOC 1.0000 1.0000 1.0000 7
M-PRO 0.8125 0.9559 0.8784 68
B-EDU 0.9821 0.9821 0.9821 112
M-ORG 0.9553 0.9690 0.9621 4325
M-LOC 1.0000 0.8571 0.9231 28
E-ID 1.0000 1.0000 1.0000 1
O 0.9681 0.9822 0.9751 5222
E-NAME 1.0000 0.9912 0.9956 113
B-ID 1.0000 1.0000 1.0000 1
avg/total 0.9605 0.9604 0.9603 15162
基于Bilstm + CRF的信息抽取模型
197 浏览量
2024-05-20
13:06:31
上传
评论
收藏 21.93MB ZIP 举报
PytorchBilstmCRF-Information-Extraction-main.zip （36个子文件） 
PytorchBilstmCRF-Information-Extraction-main 
utils.py 2KB __init__.py 0B evaluate.py 1KB ckpts bilstm_crf.pkl 23.41MB main.py 1KB data crf_word2id.pkl 15KB train.char 1.05MB test.char 132KB dev.char 119KB crf_tag2id.pkl 378B lables.char 197B evaluating.py 7KB predict.py 2KB .idea ailabner.iml 438B misc.xml 302B inspectionProfiles profiles_settings.xml 174B modules.xml 268B .gitignore 47B encodings.xml 181B modelgraph __init__.py 0B BILSTM_CRF.py 4KB __pycache__ BILSTM.cpython-37.pyc 2KB BILSTM_CRF.cpython-37.pyc 4KB __init__.cpython-37.pyc 131B BILSTM.py 1KB data.py 1019B
requirements.txt 25B result.txt 2KB operate_bilstm.py 7KB __pycache__ data.cpython-37.pyc 937B operate_bilstm.cpython-37.pyc 5KB evaluate.cpython-37.pyc 1KB utils.cpython-37.pyc 3KB evaluating.cpython-37.pyc 6KB config.cpython-37.pyc 535B config.py 263B资源评论
