TensorFlow安裝CPU版本和GPU版本的實(shí)現(xiàn)步驟
更新時(shí)間：2025年03月12日 09:41:54 作者：席惜兮兮
本文主要介紹了TensorFlow安裝CPU版本和GPU版本的實(shí)現(xiàn)步驟,文中通過圖文示例介紹的非常詳細(xì),對(duì)大家的學(xué)習(xí)或者工作具有一定的參考學(xué)習(xí)價(jià)值,需要的朋友們下面隨著小編來(lái)一起學(xué)習(xí)學(xué)習(xí)吧
前言
下載的Anaconda是Anaconda3-2024.02-1-Windows-x86_64版本
一、TensorFlow安裝CPU版本
本例子，下載的Python版本為3.11.7和tensorflow版本為2.16.1
1.新建虛擬環(huán)境
打開Anaconda Prompt，輸入
conda create -n myenvname python=3.11.7
“myenvname”為自己的虛擬環(huán)境名字
2.激活虛擬環(huán)境
繼續(xù)輸入
activate myenvname
“myenvname”為自己的虛擬環(huán)境名字
3.下載tensorflow
直接安裝tensorflow會(huì)遇到以下報(bào)錯(cuò)，這是提示有一些依賴沒有安裝
所以我先安裝了依賴再下載tensorflow
pip install joblib==1.2.0 scipy==1.11.4 tabulate==0.9.0 tqdm==4.65.0 tensorflow==2.16.1 -i https://mirrors.aliyun.com/pypi/simple
4.驗(yàn)證是否下載成功
輸入ipython，進(jìn)入交互環(huán)境（要是報(bào)錯(cuò)，那可能是沒有ipython，可以pip list查看一下，沒有的話需要下載一個(gè)）
導(dǎo)入tensorflow
import tensorflow as tf
成功
二、TensorFlow安裝GPU版本
本例子，下載的CUDA版本是11.5.2，cuDNN的版本是8.3.2，Python環(huán)境是3.9，tensorflow-gpu的版本是2.7.0。注：CUDA、cuDNN、python的環(huán)境要對(duì)應(yīng)，不然會(huì)安裝失?。ê苤匾。。。?。
1.新建虛擬環(huán)境
打開Anaconda Prompt，輸入
conda create -n myenvname python=3.9
“myenvname”為自己的虛擬環(huán)境名字
2.激活虛擬環(huán)境
activate myenvname
“myenvname”為自己的虛擬環(huán)境名字
3.安裝tensorflow-gpu
pip install tensorflow-gpu==2.7.0 -i https://pypi.tuna.tsinghua.edu.cn/simple
4.驗(yàn)證是否下載成功
進(jìn)入python環(huán)境，導(dǎo)入tensorflow
import tensorflow as tf
要是遇到這個(gè)問題，提示protobuf版本過低
(tensorflow2) C:\Users\asus>python
Python 3.9.19 (main, May 6 2024, 20:12:36) [MSC v.1916 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow as tf
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\__init__.py", line 41, in <module>
from tensorflow.python.tools import module_util as _module_util
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\__init__.py", line 41, in <module>
from tensorflow.python.eager import context
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\eager\context.py", line 33, in <module>
from tensorflow.core.framework import function_pb2
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\core\framework\function_pb2.py", line 16, in <module>
from tensorflow.core.framework import attr_value_pb2 as tensorflow_dot_core_dot_framework_dot_attr__value__pb2
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\core\framework\attr_value_pb2.py", line 16, in <module>
from tensorflow.core.framework import tensor_pb2 as tensorflow_dot_core_dot_framework_dot_tensor__pb2
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\core\framework\tensor_pb2.py", line 16, in <module>
from tensorflow.core.framework import resource_handle_pb2 as tensorflow_dot_core_dot_framework_dot_resource__handle__pb2
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\core\framework\resource_handle_pb2.py", line 16, in <module>
from tensorflow.core.framework import tensor_shape_pb2 as tensorflow_dot_core_dot_framework_dot_tensor__shape__pb2
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\core\framework\tensor_shape_pb2.py", line 36, in <module>
_descriptor.FieldDescriptor(
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\google\protobuf\descriptor.py", line 553, in __new__
_message.Message._CheckCalledFromGeneratedFile()
TypeError: Descriptors cannot be created directly.
If this call came from a _pb2.py file, your generated code is out of date and must be regenerated with protoc >= 3.19.0.
If you cannot immediately regenerate your protos, some other possible workarounds are:
1. Downgrade the protobuf package to 3.20.x or lower.
2. Set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python (but this will use pure-Python parsing and will be much slower).
More information: https://developers.google.com/protocol-buffers/docs/news/2022-05-06#python-updates
>>>
輸入exit()退出python環(huán)境，回到虛擬環(huán)境
pip install protobuf==3.19.6 -i https://pypi.tuna.tsinghua.edu.cn/simple --trusted-host pypi.tuna.tsinghua.edu.cn
再次進(jìn)入python環(huán)境，輸入“import tensorflow as tf”，要是遇到如下問題，提示TensorFlow與NumPy的版本不兼容
(tensorflow2) C:\Users\asus>python
Python 3.9.19 (main, May 6 2024, 20:12:36) [MSC v.1916 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow as tf
A module that was compiled using NumPy 1.x cannot be run in
NumPy 2.0.0 as it may crash. To support both 1.x and 2.x
versions of NumPy, modules must be compiled with NumPy 2.0.
Some module may need to rebuild instead e.g. with 'pybind11>=2.12'.
If you are a user of the module, the easiest solution will be to
downgrade to 'numpy<2' or try to upgrade the affected module.
We expect that some modules will need time to support NumPy 2.
Traceback (most recent call last): File "<stdin>", line 1, in <module>
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\__init__.py", line 41, in <module>
from tensorflow.python.tools import module_util as _module_util
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\__init__.py", line 41, in <module>
from tensorflow.python.eager import context
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\eager\context.py", line 38, in <module>
from tensorflow.python.client import pywrap_tf_session
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\client\pywrap_tf_session.py", line 23, in <module>
from tensorflow.python.client._pywrap_tf_session import *
AttributeError: _ARRAY_API not found
A module that was compiled using NumPy 1.x cannot be run in
NumPy 2.0.0 as it may crash. To support both 1.x and 2.x
versions of NumPy, modules must be compiled with NumPy 2.0.
Some module may need to rebuild instead e.g. with 'pybind11>=2.12'.
If you are a user of the module, the easiest solution will be to
downgrade to 'numpy<2' or try to upgrade the affected module.
We expect that some modules will need time to support NumPy 2.
Traceback (most recent call last): File "<stdin>", line 1, in <module>
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\__init__.py", line 41, in <module>
from tensorflow.python.tools import module_util as _module_util
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\__init__.py", line 46, in <module>
from tensorflow.python import data
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\data\__init__.py", line 25, in <module>
from tensorflow.python.data import experimental
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\data\experimental\__init__.py", line 98, in <module>
from tensorflow.python.data.experimental import service
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\data\experimental\service\__init__.py", line 374, in <module>
from tensorflow.python.data.experimental.ops.data_service_ops import distribute
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\data\experimental\ops\data_service_ops.py", line 27, in <module>
from tensorflow.python.data.experimental.ops import compression_ops
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\data\experimental\ops\compression_ops.py", line 20, in <module>
from tensorflow.python.data.util import structure
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\data\util\structure.py", line 26, in <module>
from tensorflow.python.data.util import nest
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\data\util\nest.py", line 40, in <module>
from tensorflow.python.framework import sparse_tensor as _sparse_tensor
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\framework\sparse_tensor.py", line 28, in <module>
from tensorflow.python.framework import constant_op
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\framework\constant_op.py", line 29, in <module>
from tensorflow.python.eager import execute
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\eager\execute.py", line 27, in <module>
from tensorflow.python.framework import dtypes
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\framework\dtypes.py", line 30, in <module>
from tensorflow.python.lib.core import _pywrap_bfloat16
AttributeError: _ARRAY_API not found
ImportError: numpy.core._multiarray_umath failed to import
ImportError: numpy.core.umath failed to import
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\__init__.py", line 41, in <module>
from tensorflow.python.tools import module_util as _module_util
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\__init__.py", line 46, in <module>
from tensorflow.python import data
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\data\__init__.py", line 25, in <module>
from tensorflow.python.data import experimental
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\data\experimental\__init__.py", line 98, in <module>
from tensorflow.python.data.experimental import service
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\data\experimental\service\__init__.py", line 374, in <module>
from tensorflow.python.data.experimental.ops.data_service_ops import distribute
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\data\experimental\ops\data_service_ops.py", line 27, in <module>
from tensorflow.python.data.experimental.ops import compression_ops
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\data\experimental\ops\compression_ops.py", line 20, in <module>
from tensorflow.python.data.util import structure
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\data\util\structure.py", line 26, in <module>
from tensorflow.python.data.util import nest
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\data\util\nest.py", line 40, in <module>
from tensorflow.python.framework import sparse_tensor as _sparse_tensor
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\framework\sparse_tensor.py", line 28, in <module>
from tensorflow.python.framework import constant_op
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\framework\constant_op.py", line 29, in <module>
from tensorflow.python.eager import execute
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\eager\execute.py", line 27, in <module>
from tensorflow.python.framework import dtypes
File "C:\Users\asus\.conda\envs\tensorflow2\lib\site-packages\tensorflow\python\framework\dtypes.py", line 33, in <module>
_np_bfloat16 = _pywrap_bfloat16.TF_bfloat16_type()
TypeError: Unable to convert function return value to a Python type! The signature was
() -> handle
>>>
輸入exit()退出python環(huán)境，回到虛擬環(huán)境
pip install numpy==1.21.6 -i https://pypi.tuna.tsinghua.edu.cn/simple/
進(jìn)入python環(huán)境，輸入
import tensorflow as tf tf.__version__ tf.test.is_gpu_available()
查看版本2.7.0，版本正確。末尾顯示True，TensorFlow檢測(cè)到可用的GPU，安裝成功，exit()退出python環(huán)境
后續(xù)我想用ipython查看是否安裝成功，出現(xiàn)以下問題進(jìn)入ipython環(huán)境，輸入
import tensorflow as tf tf.__version__ tf.test.is_gpu_available()
創(chuàng)建虛擬環(huán)境的時(shí)候指定python版本為3.9，但是這里卻顯示3.11.7。
末尾顯示False，TensorFlow沒有檢測(cè)到可用的GPU。
猜測(cè)可能是這個(gè)虛擬環(huán)境沒有ipython，可能用了其他環(huán)境的ipython。
解決方案，可以在虛擬環(huán)境中用pip list查看虛擬環(huán)境中是否有ipython，要是沒有，需要安裝一個(gè)，然后就可以解決了
到此這篇關(guān)于TensorFlow安裝CPU版本和GPU版本的實(shí)現(xiàn)步驟的文章就介紹到這了,更多相關(guān)TensorFlow安裝CPU和GPU內(nèi)容請(qǐng)搜索腳本之家以前的文章或繼續(xù)瀏覽下面的相關(guān)文章希望大家以后多多支持腳本之家！
您可能感興趣的文章:
- 運(yùn)行tensorflow python程序,限制對(duì)GPU和CPU的占用操作
- 基于Tensorflow使用CPU而不用GPU問題的解決
- 在tensorflow中設(shè)置使用某一塊GPU、多GPU、CPU的操作
- tensorflow指定CPU與GPU運(yùn)算的方法實(shí)現(xiàn)
- 卸載tensorflow-cpu重裝tensorflow-gpu操作
- 使用Tensorflow-GPU禁用GPU設(shè)置(CPU與GPU速度對(duì)比)
- 詳解tf.device()指定tensorflow運(yùn)行的GPU或CPU設(shè)備實(shí)現(xiàn)
- 用gpu訓(xùn)練好的神經(jīng)網(wǎng)絡(luò),用tensorflow-cpu跑出錯(cuò)的原因及解決方案
- Tensorflow中使用cpu和gpu有什么區(qū)別
- tensorflow之如何使用GPU而不是CPU問題
相關(guān)文章
python序列化與數(shù)據(jù)持久化實(shí)例詳解
這篇文章主要介紹了python序列化與數(shù)據(jù)持久化,結(jié)合實(shí)例形式詳細(xì)分析了Python序列化與數(shù)據(jù)持久化相關(guān)原理、實(shí)現(xiàn)技巧與操作注意事項(xiàng),需要的朋友可以參考下2019-12-12
Python中五個(gè)你不知道的print函數(shù)用法詳解
今天我們來(lái)聊聊Python里最 平平無(wú)奇卻又無(wú)處不在的print()函數(shù),作為Python開發(fā)者,我們幾乎每天都在用它,但你真的了解它的所有玩法嗎,下面就跟隨小編來(lái)看看五個(gè)你不知道的print函數(shù)用法吧2026-01-01
在Python中,heapq模塊提供了實(shí)現(xiàn)最小堆算法的數(shù)據(jù)結(jié)構(gòu),能夠用作優(yōu)先隊(duì)列,本文將詳細(xì)介紹heapq模塊,包括堆的基本概念、heapq的功能和示例代碼,需要的可以參考下2023-12-12
基于Python自制一個(gè)圖片批量處理工具實(shí)現(xiàn)格式統(tǒng)一和大小壓縮
在日常工作和生活中,我們經(jīng)常需要處理大量圖片,今天就給大家分享一款自制的Python圖片批量處理工具,無(wú)需復(fù)雜操作,圖形化界面一鍵搞定所有需求2025-11-11
Python（Pandas、Dask、PySpark等庫(kù)）在大數(shù)據(jù)處理中的學(xué)習(xí)心得
作者分享了他對(duì)Python在大數(shù)據(jù)處理中的學(xué)習(xí)心得,介紹了大數(shù)據(jù)的概念、特點(diǎn)及處理挑戰(zhàn),討論了Python的優(yōu)勢(shì)及常用的大數(shù)據(jù)處理庫(kù),比較了Python與Rust在大數(shù)據(jù)處理中的優(yōu)劣,并提出了大數(shù)據(jù)處理的最佳實(shí)踐2026-04-04
這篇文章主要和大家講解一下常見的圖像銳化和邊緣檢測(cè)方法，即Roberts算子和Prewitt算子。文中的示例代碼講解詳細(xì)，感興趣的小伙伴可以了解一下2022-12-12
Python實(shí)現(xiàn)繁體轉(zhuǎn)簡(jiǎn)體功能的三種方案
在中文信息處理中,繁體字與簡(jiǎn)體字的轉(zhuǎn)換是一個(gè)常見需求,無(wú)論是處理港澳臺(tái)地區(qū)的文本數(shù)據(jù),還是開發(fā)面向不同中文用戶群體的應(yīng)用,繁簡(jiǎn)轉(zhuǎn)換都是不可或缺的功能,本文將詳細(xì)介紹如何在Python中實(shí)現(xiàn)高效準(zhǔn)確的繁體轉(zhuǎn)簡(jiǎn)體功能,需要的朋友可以參考下2025-11-11
python機(jī)器學(xué)習(xí)高數(shù)篇之泰勒公式
這篇文章主要介紹了python機(jī)器學(xué)習(xí)高數(shù)篇之函數(shù)極限和導(dǎo)數(shù)，本文給大家介紹的非常詳細(xì)，對(duì)大家的學(xué)習(xí)或工作具有一定的參考借鑒價(jià)值，需要的朋友可以參考下2021-08-08
Python?是一種代表簡(jiǎn)單思想的語(yǔ)言，其語(yǔ)法相對(duì)簡(jiǎn)單，很容易上手。本文精心篩選了Python中的八大核心語(yǔ)句，快來(lái)看看你都掌握了幾個(gè)呢2023-02-02
這篇文章主要介紹了Pycharm+Python+PyQt5使用詳解，文中通過示例代碼介紹的非常詳細(xì)，對(duì)大家的學(xué)習(xí)或者工作具有一定的參考學(xué)習(xí)價(jià)值，需要的朋友們下面隨著小編來(lái)一起學(xué)習(xí)學(xué)習(xí)吧2019-09-09