安装 tensorflow-2.10.0 支持 gpu

1. 起因， 目的:

• 重新安装 tensorflow gpu， 因为之前的项目都删掉了。遇到一点问题，记录一下。
• 来回折腾，本质是 我的 python3.11 版本太高了。换成 python3.9 就行了。
• 坚决不使用 AI 扩写。垃圾而恶心。

2. 先看效果

无。

3. 过程:

代码

import platform
import sys
import os
import subprocess
import shutildef check_os():os_type = platform.system()arch = platform.machine()print(f"操作系统: {os_type}")print(f"系统架构: {arch}")def check_python():print(f"Python 版本: {platform.python_version()}")print(f"Python 位数: {platform.architecture()[0]}")def check_cuda_toolkit():possible_paths = ["C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA","/usr/local/cuda",]found = Falsefor path in possible_paths:if os.path.exists(path):print(f"已检测到 CUDA Toolkit 安装路径: {path}")found = Truebreakif not found:print("未检测到 CUDA Toolkit 安装路径")def check_nvidia_smi():try:result = subprocess.run(["nvidia-smi"], capture_output=True, text=True)if result.returncode == 0:print("✅ 已检测到 NVIDIA GPU：")print(result.stdout.split("\n")[0])  # 打印首行信息else:print("⚠️ 未检测到 NVIDIA GPU 或未安装驱动")except FileNotFoundError:print("⚠️ 未安装 NVIDIA 驱动（nvidia-smi 命令不可用）")def main():print("=== 系统信息检测 ===")check_os()print()check_python()print()check_cuda_toolkit()print()check_nvidia_smi()print("====================")if __name__ == "__main__":main()"""
输出：=== 系统信息检测 ===
操作系统: Windows
系统架构: AMD64Python 版本: 3.11.1
Python 位数: 64bit已检测到 CUDA Toolkit 安装路径: C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA✅ 已检测到 NVIDIA GPU：
Fri May 23 05:02:23 2025       
====================而且：(venv) C:\Users\Administrator\Code\python_nlp>nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver                  
Copyright (c) 2005-2022 NVIDIA Corporation             
Built on Tue_Mar__8_18:36:24_Pacific_Standard_Time_2022
Cuda compilation tools, release 11.6, V11.6.124        
Build cuda_11.6.r11.6/compiler.31057947_0              (venv) C:\Users\Administrator\Code\python_nlp>         (venv) C:\Users\Administrator\Code\python_nlp>nvidia-smi
Fri May 23 05:07:25 2025                                                                   
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 572.70                 Driver Version: 572.70         CUDA Version: 12.8     |
|-----------------------------------------+------------------------+----------------------+
| GPU  Name                  Driver-Model | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|    0   N/A  N/A            2300    C+G   ...ram Files\Tencent\QQNT\QQ.exe      N/A      |
|    0   N/A  N/A            8288    C+G   ...h_cw5n1h2txyewy\SearchApp.exe      N/A      |
|    0   N/A  N/A            9224    C+G   ...xyewy\ShellExperienceHost.exe      N/A      |
|    0   N/A  N/A           10644    C+G   C:\Windows\explorer.exe               N/A      |
|    0   N/A  N/A           11168    C+G   ...1.1.1\jbr\bin\jcef_helper.exe      N/A      |
|    0   N/A  N/A           12428    C+G   ...5n1h2txyewy\TextInputHost.exe      N/A      |
|    0   N/A  N/A           13452    C+G   ...Chrome\Application\chrome.exe      N/A      |
|    0   N/A  N/A           14156    C+G   ...acted\runtime\WeChatAppEx.exe      N/A      |
|    0   N/A  N/A           14396    C+G   ...Chrome\Application\chrome.exe      N/A      |
|    0   N/A  N/A           16664    C+G   ....0.3240.76\msedgewebview2.exe      N/A      |
+-----------------------------------------------------------------------------------------+(venv) C:\Users\Administrator\Code\python_nlp>
(venv) C:\Users\Administrator\Code\python_nlp>请根据我的系统信息，帮我配置 tensorflow gpu , 如何安装，请给出具体步骤。# 我的 python3.11， 版本太高了， 不行。换成 python3.9 可以的！
pip install tensorflow==2.10.0 -i https://pypi.org/simple"""

4. 结论 + todo

• 繁琐是难免的。容易吗，不容易！

---

希望对大家有帮助。