Text-Matching-master
所属分类：Windows编程
开发工具：Python
文件大小：6564KB
下载次数：3
上传日期：2020-06-11 13:01:31
上 传 者：
yyxyyx123
说明： 在LCQMC数据集上的=测试几种常见的语义匹配模型
(Text matching models on LCQMC datasets)
文件列表:
ABCNN (0, 2019-12-05)
ABCNN\__pycache__ (0, 2019-12-05)
ABCNN\__pycache__\__init__.cpython-36.pyc (103, 2019-12-05)
ABCNN\__pycache__\__init__.cpython-37.pyc (107, 2019-12-05)
ABCNN\__pycache__\args.cpython-36.pyc (384, 2019-12-05)
ABCNN\__pycache__\graph.cpython-36.pyc (3708, 2019-12-05)
ABCNN\__pycache__\graph.cpython-37.pyc (3662, 2019-12-05)
ABCNN\abcnn.py (5048, 2019-12-05)
ABCNN\args.py (240, 2019-12-05)
ABCNN\test.py (814, 2019-12-05)
ABCNN\train.py (2606, 2019-12-05)
DSSM (0, 2019-12-05)
DSSM\__pycache__ (0, 2019-12-05)
DSSM\__pycache__\__init__.cpython-36.pyc (102, 2019-12-05)
DSSM\__pycache__\args.cpython-36.pyc (287, 2019-12-05)
DSSM\__pycache__\graph.cpython-36.pyc (2410, 2019-12-05)
DSSM\args.py (152, 2019-12-05)
DSSM\dssm.py (2280, 2019-12-05)
DSSM\test.py (758, 2019-12-05)
DSSM\train.py (2565, 2019-12-05)
ESIM (0, 2019-12-05)
ESIM\__pycache__ (0, 2019-12-05)
ESIM\__pycache__\__init__.cpython-36.pyc (102, 2019-12-05)
ESIM\__pycache__\args.cpython-36.pyc (348, 2019-12-05)
ESIM\__pycache__\graph.cpython-36.pyc (3262, 2019-12-05)
ESIM\args.py (206, 2019-12-05)
ESIM\esim.py (3973, 2019-12-05)
ESIM\test.py (800, 2019-12-05)
ESIM\train.py (2564, 2019-12-05)
data (0, 2019-12-05)
data\LCQMC_dev.csv (699252, 2019-12-05)
data\LCQMC_test.csv (789072, 2019-12-05)
data\LCQMC_train.csv (16750782, 2019-12-05)
data\vocab.txt (31368, 2019-12-05)
utils (0, 2019-12-05)
utils\__pycache__ (0, 2019-12-05)
utils\__pycache__\__init__.cpython-36.pyc (103, 2019-12-05)
utils\__pycache__\data_utils.cpython-36.pyc (2506, 2019-12-05)
... ...
Text matching models on LCQMC datasets
## Requrement
- python 3.6
- tensorflow-gpu 1.12
- gensim 3.8.1
- jieba 0.39
- numpy 1.16
- pandas 0.23
## To Do List
### Single Model
- [x] DSSM
- [x] ABCNN
- [x] ESIM
- [ ] BIMPM
- [ ] DIIN
- [ ] DRCN
- [ ] RE2
### Classic Algorithm
- [ ] TFIDF
- [ ] BM25
- [ ] VSM
### LM Fintune
- [ ] ELMo
- [ ] BERT
- [ ] ALBERT
## Result
Model| accuracy | loss | word/char
:-: | :-: | :-: | :-: |
DSSM | 63.336% | 0.***119714 | char
ABCNN | 79.928% | 0.***21789 | char
ESIM | 81.8% | 0.48200694 | char
BIMPM | | |
DIIN | 84.472% | 0.34605518 | char + dynamic word
DRCN | | |
RE2 | | |
近期下载者：
相关文件：
收藏者：