Update docs

RapidAI · Apr 24, 2024 · ed92207 · ed92207
1 parent 3ce4f1a
commit ed92207
Showing 1 changed file with 5 additions and 4 deletions.
diff --git a/content/docs/inference_engine/onnxruntime/onnxruntime-gpu.md b/content/docs/inference_engine/onnxruntime/onnxruntime-gpu.md
@@ -25,12 +25,13 @@ onnxruntime-gpu版在动态输入情况下，推理速度要比CPU慢很多。
 ### 相关对比表格
 |设备|onnxruntime-gpu|CPU总耗时(s)|CPU平均耗时(s/img)|GPU总耗时(s)|GPU平均耗时(s/img)||
 |:---:|:---:|:---:|:---:|:---:|:---:|:---:|
-|设备1[^1]|1.11.0|296.8841|1.18282|646.14667|2.57429|
-|设备2[^2]|1.12.1|149.35427|0.50504|250.81760|0.99927|
+|设备1|1.11.0|296.8841|1.18282|646.14667|2.57429|
+|设备2|1.12.1|149.35427|0.50504|250.81760|0.99927|
 
-[^1]: 宏碁(Acer) 暗影骑士·威N50-N93游戏台式机 | Windows | 十代i5-10400F 16G 512G SSD | NVIDIA GeForce GTX 1660Super 6G
+设备1: 宏碁(Acer) 暗影骑士·威N50-N93游戏台式机 | Windows | 十代i5-10400F 16G 512G SSD | NVIDIA GeForce GTX 1660Super 6G
+
+设备2: Linux | AMD R9 5950X | NVIDIA GeForce RTX 3090
 
-[^2]: Linux | AMD R9 5950X | NVIDIA GeForce RTX 3090
 
 <script src="https://giscus.app/client.js"
         data-repo="RapidAI/RapidOCRDocs"