未验证 提交 0b07bd89 编写于 作者: C Cheerego 提交者: GitHub

fix_format (#669) (#670)

* fix_format

* Update Tables.md

* Update Tables_en.md
上级 b2544dd4
......@@ -259,7 +259,8 @@ PaddePaddle通过编译时指定路径来实现引用各种BLAS/CUDA/cuDNN库。
</p>
您可以在 [Release History](https://pypi.org/project/paddlepaddle-gpu/#history) 中找到PaddlePaddle-gpu的各个发行版本。
需要注意的是,在v1.3版本中,<code> paddlepaddle-gpu </code> 命令在windows环境下,会默认安装CUDA 8.0和cuDNN 5编译的PaddlePaddle
需要注意的是,在v1.3版本中,<code> paddlepaddle-gpu </code> 命令在windows环境下,会默认安装CUDA 8.0和cuDNN 7编译的PaddlePaddle
***
......@@ -361,7 +362,7 @@ PaddePaddle通过编译时指定路径来实现引用各种BLAS/CUDA/cuDNN库。
paddlepaddle_gpu-1.3.0-cp37-cp37m-linux_x86_64.whl</a></td>
</tr>
<tr>
<td> win_amd64 </td>
<td> win_cpu_noavx_openblas </td>
<td> - </td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-openblas/paddlepaddle-1.3.0-cp27-cp27m-win_amd64.whl">
paddlepaddle-1.3.0-cp27-cp27m-win_amd64.whl</a></td>
......@@ -373,15 +374,39 @@ PaddePaddle通过编译时指定路径来实现引用各种BLAS/CUDA/cuDNN库。
paddlepaddle-1.3.0-cp37-cp37m-win_amd64.whl</a></td>
</tr>
<tr>
<td> cuda8.0_cudnn5_win_amd64 </td>
<td> win_cpu_noavx_mkl </td>
<td> - </td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-mkl/paddlepaddle-1.3.0-cp27-cp27m-win_amd64.whl">
paddlepaddle-1.3.0-cp27-cp27m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-mkl/paddlepaddle-1.3.0-cp35-cp35m-win_amd64.whl">
paddlepaddle-1.3.0-cp35-cp35m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-mkl/paddlepaddle-1.3.0-cp36-cp36m-win_amd64.whl">
paddlepaddle-1.3.0-cp36-cp36m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-mkl/paddlepaddle-1.3.0-cp37-cp37m-win_amd64.whl">
paddlepaddle-1.3.0-cp37-cp37m-win_amd64.whl</a></td>
</tr>
<tr>
<td> win_cpu_avx_openblas </td>
<td> - </td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-openblas/paddlepaddle_gpu-1.3.0-cp27-cp27m-win_amd64.whl">
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle-1.3.0-cp27-cp27m-win_amd64.whl">
paddlepaddle-1.3.0-cp27-cp27m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle-1.3.0-cp35-cp35m-win_amd64.whl">
paddlepaddle-1.3.0-cp35-cp35m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle-1.3.0-cp36-cp36m-win_amd64.whl">
paddlepaddle-1.3.0-cp36-cp36m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle-1.3.0-cp37-cp37m-win_amd64.whl">
paddlepaddle-1.3.0-cp37-cp37m-win_amd64.whl</a></td>
</tr>
<tr>
<td> win_cuda8.0_cudnn7_cpu_avx_openblas </td>
<td> - </td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle_gpu-1.3.0-cp27-cp27m-win_amd64.whl">
paddlepaddle_gpu-1.3.0-cp27-cp27m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-openblas/paddlepaddle_gpu-1.3.0-cp35-cp35m-win_amd64.whl">
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle_gpu-1.3.0-cp35-cp35m-win_amd64.whl">
paddlepaddle_gpu-1.3.0-cp35-cp35m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-openblas/paddlepaddle_gpu-1.3.0-cp36-cp36m-win_amd64.whl">
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle_gpu-1.3.0-cp36-cp36m-win_amd64.whl">
paddlepaddle_gpu-1.3.0-cp36-cp36m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-openblas/paddlepaddle_gpu-1.3.0-cp37-cp37m-win_amd64.whl">
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle_gpu-1.3.0-cp37-cp37m-win_amd64.whl">
paddlepaddle_gpu-1.3.0-cp37-cp37m-win_amd64.whl</a></td>
</tr>
</tbody>
......@@ -406,9 +431,6 @@ cuda8.0_cudnn7_avx_mkl: 支持GPU训练和预测,使用avx指令集和Intel mk
cuda9.0_cudnn7_avx_mkl: 支持GPU训练和预测,使用avx指令集和Intel mkl数学库
win_amd64:只支持CPU训练和预测,使用AMD64指令集
cuda8.0_cudnn5_win_amd64:支持支持GPU训练和预测,使用AMD64指令集
- 横轴
......
......@@ -261,7 +261,8 @@ PaddePaddle implements references to various BLAS/CUDA/cuDNN libraries by specif
You can find various distributions of PaddlePaddle-gpu in [the Release History](https://pypi.org/project/paddlepaddle-gpu/#history).
Please note that: paddlepaddle-gpu==1.3.0 in windows, will download package compiled with CUDA 8.0 and cuDNN 5
Please note that: paddlepaddle-gpu==1.3.0 in windows, will download package compiled with CUDA 8.0 and cuDNN 7
***
<a name="dockers"></a>
......@@ -399,7 +400,7 @@ You can find the docker image for each release of PaddlePaddle in the [DockerHub
paddlepaddle_gpu-1.3.0-cp37-cp37m-linux_x86_64.whl</a></td>
</tr>
<tr>
<td> win_amd64 </td>
<td> win_cpu_noavx_openblas </td>
<td> - </td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-openblas/paddlepaddle-1.3.0-cp27-cp27m-win_amd64.whl">
paddlepaddle-1.3.0-cp27-cp27m-win_amd64.whl</a></td>
......@@ -411,15 +412,39 @@ You can find the docker image for each release of PaddlePaddle in the [DockerHub
paddlepaddle-1.3.0-cp37-cp37m-win_amd64.whl</a></td>
</tr>
<tr>
<td> cuda8.0_cudnn5_win_amd64 </td>
<td> win_cpu_noavx_mkl </td>
<td> - </td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-mkl/paddlepaddle-1.3.0-cp27-cp27m-win_amd64.whl">
paddlepaddle-1.3.0-cp27-cp27m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-mkl/paddlepaddle-1.3.0-cp35-cp35m-win_amd64.whl">
paddlepaddle-1.3.0-cp35-cp35m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-mkl/paddlepaddle-1.3.0-cp36-cp36m-win_amd64.whl">
paddlepaddle-1.3.0-cp36-cp36m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-mkl/paddlepaddle-1.3.0-cp37-cp37m-win_amd64.whl">
paddlepaddle-1.3.0-cp37-cp37m-win_amd64.whl</a></td>
</tr>
<tr>
<td> win_cpu_avx_openblas </td>
<td> - </td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle-1.3.0-cp27-cp27m-win_amd64.whl">
paddlepaddle-1.3.0-cp27-cp27m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle-1.3.0-cp35-cp35m-win_amd64.whl">
paddlepaddle-1.3.0-cp35-cp35m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle-1.3.0-cp36-cp36m-win_amd64.whl">
paddlepaddle-1.3.0-cp36-cp36m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle-1.3.0-cp37-cp37m-win_amd64.whl">
paddlepaddle-1.3.0-cp37-cp37m-win_amd64.whl</a></td>
</tr>
<tr>
<td> win_cuda8.0_cudnn7_cpu_avx_openblas </td>
<td> - </td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-openblas/paddlepaddle_gpu-1.3.0-cp27-cp27m-win_amd64.whl">
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle_gpu-1.3.0-cp27-cp27m-win_amd64.whl">
paddlepaddle_gpu-1.3.0-cp27-cp27m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-openblas/paddlepaddle_gpu-1.3.0-cp35-cp35m-win_amd64.whl">
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle_gpu-1.3.0-cp35-cp35m-win_amd64.whl">
paddlepaddle_gpu-1.3.0-cp35-cp35m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-openblas/paddlepaddle_gpu-1.3.0-cp36-cp36m-win_amd64.whl">
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle_gpu-1.3.0-cp36-cp36m-win_amd64.whl">
paddlepaddle_gpu-1.3.0-cp36-cp36m-win_amd64.whl</a></td>
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-noavx-openblas/paddlepaddle_gpu-1.3.0-cp37-cp37m-win_amd64.whl">
<td> <a href="https://paddle-wheel.bj.bcebos.com/1.3.0-win-avx-openblas/paddlepaddle_gpu-1.3.0-cp37-cp37m-win_amd64.whl">
paddlepaddle_gpu-1.3.0-cp37-cp37m-win_amd64.whl</a></td>
</tr>
</tbody>
......
......@@ -9,7 +9,7 @@
### 注意事项
* 默认提供的安装包需要计算机支持AVX指令集和MKL
* 默认提供的安装包需要计算机支持AVX指令集和MKL,如果您的环境不支持,请在[这里](./Tables.html/#ciwhls-release)下载`no_avx``openblas`等版本的安装包
* 当前版本暂不支持NCCL,分布式等相关功能
## 选择CPU/GPU
......
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