东方耀AI技术分享

标题: 安装PaddlePaddle深度学习框架 [打印本页]

作者: 东方耀 时间: 2021-6-25 11:46
标题: 安装PaddlePaddle深度学习框架
安装PaddlePaddle深度学习框架

运行PaddleClas需要PaddlePaddle 2.0或更高版本

环境要求:
python 3.x
cuda >= 10.1 (如果使用paddlepaddle-gpu)
cudnn >= 7.6.4 (如果使用paddlepaddle-gpu)
nccl >= 2.1.2 (如果使用分布式训练/评估)
gcc >= 8.2 为什么gcc版本要求这么高？

在cuda10.1时，建议显卡驱动版本大于等于418.39；在使用cuda10.2时，建议显卡驱动版本大于440.33，

jiang@jiang-Ubuntu:~$ nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2019 NVIDIA Corporation
Built on Sun_Jul_28_19:07:16_PDT_2019
Cuda compilation tools, release 10.1, V10.1.243

jiang@jiang-Ubuntu:~$ nvidia-smi
Fri Jun 25 10:49:03 2021
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 455.45.01 Driver Version: 455.45.01 CUDA Version: 11.1    |
|-------------------------------+----------------------+----------------------+
| GPU  Name       Persistence-M| Bus-Id       Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|       Memory-Usage | GPU-Util  Compute M. |
|                            |                   |             MIG M. |
|===============================+======================+======================|
| 0  GeForce RTX 2070 Off  | 00000000:01:00.0  On |                N/A |
| 17% 37C P8 14W / 175W | 383MiB /  7979MiB |    3%    Default |

jiang@jiang-Ubuntu:~$ cat /usr/include/x86_64-linux-gnu/cudnn_v*.h | grep CUDNN_MAJOR -A 2
#define CUDNN_MAJOR 7
#define CUDNN_MINOR 6
#define CUDNN_PATCHLEVEL 5
--
#define CUDNN_VERSION (CUDNN_MAJOR * 1000 + CUDNN_MINOR * 100 + CUDNN_PATCHLEVEL)

#include "driver_types.h"

jiang@jiang-Ubuntu:~$ gcc --version
gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
Copyright (C) 2017 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

Ubuntu18.04新安装一个高版本gcc g++，再设置优先级来随意切换：
http://www.ai111.vip/thread-1052-1-1.html

sudo apt-get update
sudo apt-get install gcc-8 g++-8

首先查看当前系统装了几个GCC G++版本:
ls /usr/bin/gcc*
ls /usr/bin/g++*

给各版本设置优先级,最后设置的数字越小,优先级越高,
sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-7 60
sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-8 30

sudo update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-7 60
sudo update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-8 30

jiang@jiang-Ubuntu:~$ gcc --version
gcc (Ubuntu 8.4.0-1ubuntu1~18.04) 8.4.0
jiang@jiang-Ubuntu:~$ g++ --version
g++ (Ubuntu 8.4.0-1ubuntu1~18.04) 8.4.0

环境检查后，正式开始安装：
conda create -n py3_baidu_pp python=3.7.10

conda activate py3_baidu_pp
pip3 install paddlepaddle-gpu --upgrade -i https://mirror.baidu.com/pypi/simple

Looking in indexes: https://mirror.baidu.com/pypi/simple
Collecting paddlepaddle-gpu
  Downloading https://mirror.baidu.com/pypi/pa ... nylinux1_x86_64.whl (349.2 MB)
   |████████████████████████████████| 349.2 MB 66 kB/s

验证是否安装成功:
(py3_baidu_pp) jiang@jiang-Ubuntu:~/py3_baidu_pp_works$ python3
Python 3.7.10 (default, Jun  4 2021, 14:48:32)
[GCC 7.5.0] :: Anaconda, Inc. on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import paddle
>>> paddle.utils.run_check()
Running verify PaddlePaddle program ...
W0625 13:46:36.840833 16880 device_context.cc:404] Please NOTE: device: 0, GPU Compute Capability: 7.5, Driver API Version: 11.1, Runtime API Version: 10.2
W0625 13:46:36.890359 16880 device_context.cc:422] device: 0, cuDNN Version: 7.6.
PaddlePaddle works well on 1 GPU.
PaddlePaddle works well on 1 GPUs.
PaddlePaddle is installed successfully! Let's start deep learning with PaddlePaddle now.
>>> paddle.__version__
'2.1.0'
>>> exit()

欢迎光临东方耀AI技术分享 (http://www.ai111.vip/)