tensorflow-gpu 2.4.0 安装

z1315975700

已于 2023-07-19 15:38:08 修改

阅读量445

点赞数

文章标签： tensorflow 人工智能 python

于 2023-07-19 14:09:11 首次发布

本文链接：https://blog.csdn.net/z1315975700/article/details/131807495

版权

1. 安装cuda

（1）没有sudo命令

su -

yum install sudo

（2）没有wget命令

sudo yum install wget

(3)安装cuda11.0

wget http://developer.download.nvidia.com/compute/cuda/11.0.2/local_installers/cuda_11.0.2_450.51.05_linux.run

sudo sh cuda_11.0.2_450.51.05_linux.run

注意，这里按回车取消勾选driver：

然后选择Install。

但给出如下日志：

===========
= Summary =
===========

Driver: Not Selected
Toolkit: Installed in /usr/local/cuda-11.0/
Samples: Installed in /root/, but missing recommended libraries

Please make sure that
- PATH includes /usr/local/cuda-11.0/bin
- LD_LIBRARY_PATH includes /usr/local/cuda-11.0/lib64, or, add /usr/local/cuda-11.0/lib64 to /etc/ld.so.conf and run ldconfig as root

To uninstall the CUDA Toolkit, run cuda-uninstaller in /usr/local/cuda-11.0/bin

Please see CUDA_Installation_Guide_Linux.pdf in /usr/local/cuda-11.0/doc/pdf for detailed information on setting up CUDA.
***WARNING: Incomplete installation! This installation did not install the CUDA Driver. A driver of version at least .00 is required for CUDA 11.0 functionality to work.
To install the driver using this installer, run the following command, replacing <CudaInstaller> with the name of this run file:
sudo <CudaInstaller>.run --silent --driver

Logfile is /var/log/cuda-installer.log

此时修改环境变量：

vim ~/.bashrc

进入以后输入：

export PATH=/usr/local/cuda-11.0/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-11.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
这里一定要输入$后面的，不然会覆盖原有的PATH变量。

出来后输入：

source ~/.bashrc

此时输入：nvcc -V测试一下，应当显示安装成功：

2. 安装cudnn8.0

（1）官网（cuDNN Archive | NVIDIA Developer）下载：我下载的cuDNN v8.0.5 for CUDA11.0

(2)

tar -xzvf cudnn-11.0-linux-x64-v8.0.5.39.tgz -C /root # -C后跟要保存的目录

# 注意,/root/cuda是cudnn目录，/usr/local/cuda是cuda的目录
(venv-kdfast) [root@0f518f45dc79 cuda] sudo cp /root/cuda/include/cudnn*.h /usr/local/cuda/include
(venv-kdfast) [root@0f518f45dc79 cuda] sudo cp /root/cuda/lib64/libcudnn* /usr/local/cuda/lib64
(venv-kdfast) [root@0f518f45dc79 cuda] sudo chmod a+r /usr/local/cuda/include/cudnn*.h
(venv-kdfast) [root@0f518f45dc79 cuda] sudo chmod a+r /usr/local/cuda/lib64/libcudnn*

（3）测试一下：cat /usr/local/cuda/include/cudnn_version.h | grep CUDNN_MAJOR -A 2

如果成功，应当显示：

#define CUDNN_MAJOR 8
#define CUDNN_MINOR 0
#define CUDNN_PATCHLEVEL 5
--
#define CUDNN_VERSION (CUDNN_MAJOR * 1000 + CUDNN_MINOR * 100 + CUDNN_PATCHLEVEL)

#endif /* CUDNN_VERSION_H */

最后，测试一下tensorflow：

>>> import tensorflow as tf
2023-07-19 15:27:48.291353: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
>>> tf.test.is_gpu_available(
... )
WARNING:tensorflow:From <stdin>:1: is_gpu_available (from tensorflow.python.framework.test_util) is deprecated and will be removed in a future version.
Instructions for updating:
Use `tf.config.list_physical_devices('GPU')` instead.
2023-07-19 15:27:57.989208: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX512F
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-07-19 15:27:58.000793: I tensorflow/compiler/jit/xla_gpu_device.cc:99] Not creating XLA devices, tf_xla_enable_xla_devices not set
2023-07-19 15:27:58.001899: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcuda.so.1
2023-07-19 15:27:58.429787: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties:
pciBusID: 0000:3b:00.0 name: GeForce RTX 3090 computeCapability: 8.6
coreClock: 1.695GHz coreCount: 82 deviceMemorySize: 23.70GiB deviceMemoryBandwidth: 871.81GiB/s
2023-07-19 15:27:58.431068: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 1 with properties:
pciBusID: 0000:af:00.0 name: GeForce RTX 3090 computeCapability: 8.6
coreClock: 1.695GHz coreCount: 82 deviceMemorySize: 23.70GiB deviceMemoryBandwidth: 871.81GiB/s
2023-07-19 15:27:58.432032: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 2 with properties:
pciBusID: 0000:d8:00.0 name: GeForce RTX 3090 computeCapability: 8.6
coreClock: 1.695GHz coreCount: 82 deviceMemorySize: 23.70GiB deviceMemoryBandwidth: 871.81GiB/s
2023-07-19 15:27:58.432133: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
2023-07-19 15:27:58.435234: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11
2023-07-19 15:27:58.435324: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublasLt.so.11
2023-07-19 15:27:58.436859: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10
2023-07-19 15:27:58.437224: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10
2023-07-19 15:27:58.441241: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.10
2023-07-19 15:27:58.442235: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11
2023-07-19 15:27:58.442442: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8
2023-07-19 15:27:58.447471: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] Adding visible gpu devices: 0, 1, 2
2023-07-19 15:27:58.447514: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
2023-07-19 15:28:00.726430: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1261] Device interconnect StreamExecutor with strength 1 edge matrix:
2023-07-19 15:28:00.726501: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1267]      0 1 2
2023-07-19 15:28:00.726512: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1280] 0:   N N N
2023-07-19 15:28:00.726520: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1280] 1:   N N N
2023-07-19 15:28:00.726527: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1280] 2:   N N N
2023-07-19 15:28:00.730421: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1406] Created TensorFlow device (/device:GPU:0 with 14343 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3090, pci bus id: 0000:3b:00.0, compute capability: 8.6)
2023-07-19 15:28:00.732704: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1406] Created TensorFlow device (/device:GPU:1 with 22428 MB memory) -> physical GPU (device: 1, name: GeForce RTX 3090, pci bus id: 0000:af:00.0, compute capability: 8.6)
2023-07-19 15:28:00.734959: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1406] Created TensorFlow device (/device:GPU:2 with 18690 MB memory) -> physical GPU (device: 2, name: GeForce RTX 3090, pci bus id: 0000:d8:00.0, compute capability: 8.6)
True

成功！