2021-08-19 15:33:03 +08:00
|
|
|
|
## 快速上手
|
|
|
|
|
|
|
2021-09-12 15:35:43 +08:00
|
|
|
|
### 环境依赖
|
|
|
|
|
|
|
|
|
|
|
|
> python == 3.8
|
|
|
|
|
|
|
|
|
|
|
|
- torch == 1.5
|
|
|
|
|
|
- hydra-core == 1.0.6
|
|
|
|
|
|
- tensorboard == 2.4.1
|
|
|
|
|
|
- matplotlib == 3.4.1
|
|
|
|
|
|
- scikit-learn == 0.24.1
|
|
|
|
|
|
- transformers == 4.5.0
|
|
|
|
|
|
- jieba == 0.42.1
|
|
|
|
|
|
- deepke
|
|
|
|
|
|
|
2021-08-19 15:33:03 +08:00
|
|
|
|
### 克隆代码
|
|
|
|
|
|
```
|
|
|
|
|
|
git clone git@github.com:zjunlp/DeepKE.git
|
|
|
|
|
|
```
|
|
|
|
|
|
### 使用pip安装
|
|
|
|
|
|
|
2021-09-12 20:22:08 +08:00
|
|
|
|
首先创建python虚拟环境，再进入虚拟环境
|
2021-08-19 15:33:03 +08:00
|
|
|
|
|
2021-09-12 15:35:43 +08:00
|
|
|
|
- 安装依赖: ```pip install -r requirements.txt```
|
2021-08-19 15:33:03 +08:00
|
|
|
|
|
2021-09-12 15:40:05 +08:00
|
|
|
|
### 使用数据进行训练预测
|
|
|
|
|
|
|
|
|
|
|
|
- 存放数据：在 `data/origin` 文件夹下存放训练数据。训练文件主要有三个文件。
|
2021-09-12 15:35:43 +08:00
|
|
|
|
|
|
|
|
|
|
- `train.csv`：存放训练数据集
|
|
|
|
|
|
|
|
|
|
|
|
- `valid.csv`：存放验证数据集
|
|
|
|
|
|
|
|
|
|
|
|
- `test.csv`：存放测试数据集
|
|
|
|
|
|
|
|
|
|
|
|
- `relation.csv`：存放关系种类
|
|
|
|
|
|
|
2021-10-24 14:18:09 +08:00
|
|
|
|
- 开始训练：```python run.py``` (训练所用到参数都在conf文件夹中，修改即可使用lm时，可修改'lm_file'使用下载至本地的模型)
|
2021-09-12 15:35:43 +08:00
|
|
|
|
|
|
|
|
|
|
- 每次训练的日志保存在 `logs` 文件夹内，模型结果保存在 `checkpoints` 文件夹内。
|
|
|
|
|
|
|
|
|
|
|
|
- 进行预测 ```python predict.py```
|
2021-08-19 15:33:03 +08:00
|
|
|
|
|
|
|
|
|
|
|
2021-09-12 15:35:43 +08:00
|
|
|
|
## 模型内容
|
2021-08-19 15:33:03 +08:00
|
|
|
|
1、CNN
|
|
|
|
|
|
|
|
|
|
|
|
2、RNN
|
|
|
|
|
|
|
|
|
|
|
|
3、Capsule
|
|
|
|
|
|
|
|
|
|
|
|
4、GCN
|
|
|
|
|
|
|
|
|
|
|
|
5、Transformer
|
|
|
|
|
|
|
|
|
|
|
|
6、预训练模型
|
