Skip to content

P
PaddleOCR

PaddlePaddle / PaddleOCR
大约 1 年 前同步成功

通知 1528
Star 32962
Fork 6643
P
PaddleOCR
未验证 提交 d7b107d1 编写于 作者: Z zhoujun 提交者： GitHub
浏览文件
操作

Merge pull request #6030 from WenmuZhou/test1 

update config.md
上级 6eb2c4ab f7c439a0
隐藏空白更改
内联 并排
Showing with 13 addition and 13 deletion
doc/doc_ch/config.md
浏览文件 @ d7b107d1
...@@ -45,18 +45,18 @@ ...@@ -45,18 +45,18 @@
### Optimizer ([ppocr/optimizer](../../ppocr/optimizer)) ### Optimizer ([ppocr/optimizer](../../ppocr/optimizer))
| 字段 | 用途 | 默认值 | 备注 | | 字段 | 用途 | 默认值 | 备注 |
| :---------------------: | :---------------------: | :--------------: | :--------------------: | | :---------------------: |:-------------:|:-------------:| :--------------------: |
| name | 优化器类名 | Adam | 目前支持`Momentum`,`Adam`,`RMSProp`, 见[ppocr/optimizer/optimizer.py](../../ppocr/optimizer/optimizer.py) | | name | 优化器类名 | Adam | 目前支持`Momentum`,`Adam`,`RMSProp`, 见[ppocr/optimizer/optimizer.py](../../ppocr/optimizer/optimizer.py) |
| beta1 | 设置一阶矩估计的指数衰减率 | 0.9 | \ | | beta1 | 设置一阶矩估计的指数衰减率 | 0.9 | \ |
| beta2 | 设置二阶矩估计的指数衰减率 | 0.999 | \ | | beta2 | 设置二阶矩估计的指数衰减率 | 0.999 | \ |
| clip_norm | 所允许的二范数最大值 | | \ | | clip_norm | 所允许的二范数最大值 | | \ |
| **lr** | 设置学习率decay方式 | - | \ | | **lr** | 设置学习率decay方式 | - | \ |
| name | 学习率decay类名 | Cosine | 目前支持`Linear`,`Cosine`,`Step`,`Piecewise`, 见[ppocr/optimizer/learning_rate.py](../../ppocr/optimizer/learning_rate.py) | | name | 学习率decay类名 | Cosine | 目前支持`Linear`,`Cosine`,`Step`,`Piecewise`, 见[ppocr/optimizer/learning_rate.py](../../ppocr/optimizer/learning_rate.py) |
| learning_rate | 基础学习率 | 0.001 | \ | | learning_rate | 基础学习率 | 0.001 | \ |
| **regularizer** | 设置网络正则化方式 | - | \ | | **regularizer** | 设置网络正则化方式 | - | \ |
| name | 正则化类名 | L2 | 目前支持`L1`,`L2`, 见[ppocr/optimizer/regularizer.py](../../ppocr/optimizer/regularizer.py) | | name | 正则化类名 | L2 | 目前支持`L1`,`L2`, 见[ppocr/optimizer/regularizer.py](../../ppocr/optimizer/regularizer.py) |
| factor | 学习率衰减系数 | 0.00004 | \ | | factor | 正则化系数 | 0.00001 | \ |
### Architecture ([ppocr/modeling](../../ppocr/modeling)) ### Architecture ([ppocr/modeling](../../ppocr/modeling))
......
doc/doc_en/config_en.md
浏览文件 @ d7b107d1
...@@ -56,7 +56,7 @@ Take rec_chinese_lite_train_v2.0.yml as an example ...@@ -56,7 +56,7 @@ Take rec_chinese_lite_train_v2.0.yml as an example
| learning_rate | Set the base learning rate | 0.001 | \ | | learning_rate | Set the base learning rate | 0.001 | \ |
| **regularizer** | Set network regularization method | - | \ | | **regularizer** | Set network regularization method | - | \ |
| name | Regularizer class name | L2 | Currently support`L1`,`L2`, see[ppocr/optimizer/regularizer.py](../../ppocr/optimizer/regularizer.py) | | name | Regularizer class name | L2 | Currently support`L1`,`L2`, see[ppocr/optimizer/regularizer.py](../../ppocr/optimizer/regularizer.py) |
| factor | Learning rate decay coefficient | 0.00004 | \ | | factor | Regularizer coefficient | 0.00001 | \ |
### Architecture ([ppocr/modeling](../../ppocr/modeling)) ### Architecture ([ppocr/modeling](../../ppocr/modeling))
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或