my_dureader
运行环境：python 3.5/3.6，pytorch 0.2.0，自行安装fasttext
参考百度给的tensorflow基线代码，改成了pytorch版本
- 该版本实现的是match-lstm，bidaf可以后期实现
- 和基线代码的不同点：
- match层的输入不同
- pointer network的初始化隐层向量不同
- 如果结果差太多，看看是否是不同框架之间参数初始化不同
- 词向量初始化采用fasttext
- 训练的时候采用单篇文章方式，测试的时候选range概率最大的答案
WSTC2007 / my_dureader_single
Licence: other
No description or website provided.
Stars: ✭ 14
Projects that are alternatives of or similar to my dureader single
DuReader QANet BiDAF
Using QANet and BiDAF on DuReader datasets
Stars: ✭ 34 (+142.86%)
Mutual labels: dureader
AiSpace
AiSpace: Better practices for deep learning model development and deployment For Tensorflow 2.0
Stars: ✭ 28 (+100%)
Mutual labels: dureader
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].