Skip to content

Commit b509bc0

Browse files
committed
docs: update docs
1 parent ebbcd11 commit b509bc0

File tree

1 file changed

+14
-2
lines changed

1 file changed

+14
-2
lines changed

docs/install_usage/rapidocr/usage.md

Lines changed: 14 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -742,7 +742,13 @@ RapidOCR输出包括4种类型:`Union[TextDetOutput, TextClsOutput, TextRecOut
742742
```python linenums="1" hl_lines="3"
743743
from rapidocr import RapidOCR
744744

745-
engine = RapidOCR(params={"Global.with_torch": True})
745+
engine = RapidOCR(
746+
params={
747+
"Det.engine_type": EngineType.TORCH,
748+
"Cls.engine_type": EngineType.TORCH,
749+
"Rec.engine_type": EngineType.TORCH,
750+
}
751+
)
746752

747753
img_url = "https://github.com/RapidAI/RapidOCR/blob/main/python/tests/test_files/ch_en_num.jpg?raw=true"
748754
result = engine(img_url)
@@ -761,6 +767,8 @@ RapidOCR输出包括4种类型:`Union[TextDetOutput, TextClsOutput, TextRecOut
761767
"Det.engine_type": EngineType.TORCH,
762768
"Cls.engine_type": EngineType.TORCH,
763769
"Rec.engine_type": EngineType.TORCH,
770+
"EngineConfig.torch.use_cuda": True, # 使用torch GPU版推理
771+
"EngineConfig.torch.gpu_id": 0, # 指定GPU id
764772
}
765773
)
766774

@@ -784,7 +792,11 @@ RapidOCR输出包括4种类型:`Union[TextDetOutput, TextClsOutput, TextRecOut
784792

785793
### 使用本地已下载模型
786794

787-
使用方法同样也是有两种:一是通过配置文件传入;二是通过初始化参数传入。下面以通过初始化参数传入为例:
795+
使用方法同样也是有两种:一是通过配置文件传入;二是通过初始化参数传入。
796+
797+
⚠️注意:Paddle格式模型需要传入`model_dir`来传入
798+
799+
下面以通过初始化参数传入为例:
788800

789801
```python linenums="1" hl_lines="4"
790802
from rapidocr import RapidOCR

0 commit comments

Comments
 (0)