未验证 提交 2539b850 编写于 作者: A acosta123 提交者: GitHub

Update compile_Windows_en.md

上级 aa3cbec3
...@@ -8,7 +8,7 @@ This instruction will show you how to compile PaddlePaddle on a *64-bit desktop ...@@ -8,7 +8,7 @@ This instruction will show you how to compile PaddlePaddle on a *64-bit desktop
## Determine which version to compile ## Determine which version to compile
* 1.3 PaddlePaddle that supports GPU.In order to make PaddlePaddle programs run faster,we usually use GPU to accelerate them.However,for the installation of GPU version of PaddlePaddle,NVIDIA GPU which satisfies with the following condictions is needed(please refer to the offical document of NVIDIA GPU for more installation and configuration details:[For CUDA](https://docs.nvidia.com/cuda/cuda-installation-guide-linux/)[For cuDNN](https://docs.nvidia.com/deeplearning/sdk/cudnn-install/) * 1.3 PaddlePaddle that supports GPU. In order to make PaddlePaddle programs run faster, we usually use GPU to accelerate them. However, for the installation of GPU version of PaddlePaddle, NVIDIA GPU which satisfies with the following condictions is needed(please refer to the offical document of NVIDIA GPU for more installation and configuration details: [For CUDA](https://docs.nvidia.com/cuda/cuda-installation-guide-linux/)[For cuDNN](https://docs.nvidia.com/deeplearning/sdk/cudnn-install/)
* Cuda tool package8.0 with cuDNN v7 * Cuda tool package8.0 with cuDNN v7
* Hardware device with GPU calculation performance beyond 1.0 * Hardware device with GPU calculation performance beyond 1.0
......
Markdown is supported
0% .
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