PowerShell 7 环境已加载 (版本: 7.5.2)
PS C:\Users\Administrator\Desktop> cd E:\PyTorch_Build\pytorch
PS E:\PyTorch_Build\pytorch> python -m venv rtx5070_env
PS E:\PyTorch_Build\pytorch> .\rtx5070_env\Scripts\activate
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 查找最新 Windows SDK 库路径
(rtx5070_env) PS E:\PyTorch_Build\pytorch> $sdkLibPath = @(
>> "${env:ProgramFiles(x86)}\Windows Kits\10\Lib",
>> "${env:ProgramFiles}\Windows Kits\10\Lib"
>> ) | Where-Object { Test-Path $_ } | Select-Object -First 1
(rtx5070_env) PS E:\PyTorch_Build\pytorch>
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 获取最新版本路径
(rtx5070_env) PS E:\PyTorch_Build\pytorch> $latestSdkVersion = Get-ChildItem -Path $sdkLibPath |
>> Where-Object { $_.Name -match '^\d+\.\d+' } |
>> Sort-Object { [version]$_.Name } -Descending |
>> Select-Object -First 1
(rtx5070_env) PS E:\PyTorch_Build\pytorch>
(rtx5070_env) PS E:\PyTorch_Build\pytorch> if ($latestSdkVersion) {
>> $umPath = Join-Path $latestSdkVersion.FullName "um\x64"
>> $ucrtPath = Join-Path $latestSdkVersion.FullName "ucrt\x64"
>>
>> # 添加到环境变量
>> $env:LIB = "$umPath;$ucrtPath;$env:LIB"
>> $env:INCLUDE = "$(Join-Path $latestSdkVersion.FullName 'um');$(Join-Path $latestSdkVersion.FullName 'ucrt');$env:INCLUDE"
>>
>> Write-Host "已添加 LIB 路径: $umPath, $ucrtPath"
>> Write-Host "已添加 INCLUDE 路径: $(Join-Path $latestSdkVersion.FullName 'um')"
>> } else {
>> Write-Host "错误: 未找到 Windows SDK 库路径" -ForegroundColor Red
>> }
已添加 LIB 路径: C:\Program Files (x86)\Windows Kits\10\Lib\10.0.22621.0\um\x64, C:\Program Files (x86)\Windows Kits\10\Lib\10.0.22621.0\ucrt\x64
已添加 INCLUDE 路径: C:\Program Files (x86)\Windows Kits\10\Lib\10.0.22621.0\um
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 验证关键工具可用性
(rtx5070_env) PS E:\PyTorch_Build\pytorch> $tools = @("cl.exe", "link.exe", "rc.exe", "mt.exe", "nvcc.exe")
(rtx5070_env) PS E:\PyTorch_Build\pytorch> $missingTools = @()
(rtx5070_env) PS E:\PyTorch_Build\pytorch>
(rtx5070_env) PS E:\PyTorch_Build\pytorch> foreach ($tool in $tools) {
>> if (-not (Get-Command $tool -ErrorAction SilentlyContinue)) {
>> $missingTools += $tool
>> }
>> }
(rtx5070_env) PS E:\PyTorch_Build\pytorch>
(rtx5070_env) PS E:\PyTorch_Build\pytorch> if ($missingTools.Count -gt 0) {
>> Write-Host "缺失工具: $($missingTools -join ', ')" -ForegroundColor Red
>> Write-Host "当前 PATH: $env:PATH"
>> } else {
>> Write-Host "所有必要工具已配置" -ForegroundColor Green
>> }
缺失工具: cl.exe, link.exe, rc.exe, mt.exe
当前 PATH: E:\PyTorch_Build\pytorch\rtx5070_env\Scripts;C:\Program Files\PowerShell\7;C:\Miniconda3\condabin;C:\Miniconda3\Scripts;E:\PyTorch_Build\pytorch\pytorch_env\Scripts;C:\Program Files\PowerShell\7;E:\PyTorch_Build\pytorch\pytorch_env\Scripts;C:\Program Files\PowerShell\7;E:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.0\bin\x64;E:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.0\bin;E:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.0\bin\x64;;E:\PyTorch_Build\pytorch\build\lib.win-amd64-cpython-310\torch\lib;E:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.0\bin;E:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v13.0\bin;C:\Users\Administrator\AppData\Local\Microsoft\dotnet;C:\Users\Administrator\AppData\Local\Microsoft\dotnet;C:\Users\Administrator\AppData\Local\Microsoft\dotnet\;C:\Program Files\dotnet;C:\WINDOWS\system32;C:\WINDOWS;C:\WINDOWS\System32\Wbem;C:\WINDOWS\System32\WindowsPowerShell\v1.0\;C:\WINDOWS\System32\OpenSSH\;E:\Python310;C:\Program Files\dotnet\;C:\Users\Administrator\AppData\Local\Microsoft\WindowsApps;C:\Users\Administrator\.dotnet\tools;C:\Users\Administrator\AppData\Local\Microsoft\WindowsApps;C:\Users\Administrator\.dotnet\tools;E:\Python310\Scripts;E:\Python310\Scripts;C:\Program Files\PowerShell\7\;E:\Program Files\Microsoft VS Code\bin;E:\Program Files\Git\cmd;C:\Program Files\NVIDIA Corporation\Nsight Compute 2025.3.0\;E:\Program Files\CMake\bin;C:\Program Files\Microsoft SQL Server\150\Tools\Binn\;C:\Program Files\Microsoft SQL Server\Client SDK\ODBC\170\Tools\Binn\;C:\Program Files (x86)\Incredibuild;E:\PyTorch_Build\pytorch\build\lib.win-amd64-cpython-310\torch\lib;E:\Program Files (x86)\Windows Kits\10\Windows Performance Toolkit\;C:\ProgramData\chocolatey\bin;E:\Program Files\Rust\.cargo\bin;C:\Users\Administrator\AppData\Local\Microsoft\WindowsApps;C:\Users\Administrator\.dotnet\tools;C:\Users\Administrator\miniconda3\Scripts;E:\Program Files\Rust\.cargo\bin;C:\Users\Administrator\AppData\Local\Microsoft\WindowsApps;C:\Users\Administrator\.dotnet\tools;C:\Users\Administrator\miniconda3\Scripts;E:\Program Files\Rust\.cargo\bin;C:\Users\Administrator\AppData\Local\Microsoft\WindowsApps;C:\Users\Administrator\.dotnet\tools;C:\ProgramData\mingw64\mingw64\bin
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 获取 SDK 版本
(rtx5070_env) PS E:\PyTorch_Build\pytorch> $sdkVersion = Get-ChildItem -Path "${env:ProgramFiles(x86)}\Windows Kits\10\Include" |
>> Where-Object { $_.PSIsContainer -and $_.Name -match '^\d+\.\d+\.\d+' } |
>> Sort-Object { [version]$_.Name } -Descending |
>> Select-Object -First 1 -ExpandProperty Name
(rtx5070_env) PS E:\PyTorch_Build\pytorch>
(rtx5070_env) PS E:\PyTorch_Build\pytorch> $cmakeArgs = @(
>> "-G Ninja",
>> "-DCMAKE_BUILD_TYPE=Release",
>> "-DCMAKE_C_COMPILER=cl.exe",
>> "-DCMAKE_CXX_COMPILER=cl.exe",
>> "-DCMAKE_CUDA_COMPILER=E:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v13.0/bin/nvcc.exe",
>> "-DCMAKE_CUDA_HOST_COMPILER=cl.exe",
>> "-DCMAKE_SYSTEM_VERSION=$sdkVersion", # 关键: 指定 SDK 版本
>> "-DCUDA_NVCC_FLAGS=-Xcompiler /wd4819 -gencode arch=compute_89,code=sm_89",
>> "-DTORCH_CUDA_ARCH_LIST=8.9",
>> "-DUSE_CUDA=ON",
>> "-DUSE_NCCL=OFF",
>> "-DUSE_DISTRIBUTED=OFF",
>> "-DBUILD_TESTING=OFF",
>> "-DBLAS=OpenBLAS",
>> "-DCUDA_TOOLKIT_ROOT_DIR=E:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v13.0",
>> "-DCUDNN_ROOT_DIR=E:/Program Files/NVIDIA/CUDNN/v9.12",
>> "-DPYTHON_EXECUTABLE=$((Get-Command python).Source)",
>> # 显式指定库路径
>> "-DCMAKE_LIBRARY_PATH=C:/Program Files (x86)/Windows Kits/10/Lib/$sdkVersion/um/x64",
>> "-DCMAKE_INCLUDE_PATH=C:/Program Files (x86)/Windows Kits/10/Include/$sdkVersion/um"
>> )
(rtx5070_env) PS E:\PyTorch_Build\pytorch>
(rtx5070_env) PS E:\PyTorch_Build\pytorch> cmake .. @cmakeArgs
CMake Warning:
Ignoring extra path from command line:
".."
CMake Error: The source directory "E:/PyTorch_Build" does not appear to contain CMakeLists.txt.
Specify --help for usage, or press the help button on the CMake GUI.
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 加载完整 VC 环境
(rtx5070_env) PS E:\PyTorch_Build\pytorch> & "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\Build\vcvars64.bat" -vcvars_ver=14.4 -winsdk=10.0.22621.0
[ERROR:vcvarsall.bat] Invalid argument found : .4
[ERROR:vcvarsall.bat] Invalid argument found : -winsdk=10
[ERROR:vcvarsall.bat] Invalid argument found : .0.22621.0
[ERROR:vcvarsall.bat] Error in script usage. The correct usage is:
Syntax:
vcvarsall.bat [arch] [platform_type] [winsdk_version] [-vcvars_ver=vc_version] [-vcvars_spectre_libs=spectre_mode]
where :
[arch]: x86 | amd64 | x86_amd64 | x86_arm | x86_arm64 | amd64_x86 | amd64_arm | amd64_arm64
[platform_type]: {empty} | store | uwp
[winsdk_version] : full Windows 10 SDK number (e.g. 10.0.10240.0) or "8.1" to use the Windows 8.1 SDK.
[vc_version] : {none} for latest installed VC++ compiler toolset |
"14.0" for VC++ 2015 Compiler Toolset |
"14.xx" for the latest 14.xx.yyyyy toolset installed (e.g. "14.11") |
"14.xx.yyyyy" for a specific full version number (e.g. "14.11.25503")
[spectre_mode] : {none} for libraries without spectre mitigations |
"spectre" for libraries with spectre mitigations
The store parameter sets environment variables to support Universal Windows Platform application
development and is an alias for 'uwp'.
For example:
vcvarsall.bat x86_amd64
vcvarsall.bat x86_amd64 10.0.10240.0
vcvarsall.bat x86_arm uwp 10.0.10240.0
vcvarsall.bat x86_arm onecore 10.0.10240.0 -vcvars_ver=14.0
vcvarsall.bat x64 8.1
vcvarsall.bat x64 store 8.1
Please make sure either Visual Studio or C++ Build SKU is installed.
(rtx5070_env) PS E:\PyTorch_Build\pytorch>
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 验证环境
(rtx5070_env) PS E:\PyTorch_Build\pytorch> cl.exe
cl.exe: The term 'cl.exe' is not recognized as a name of a cmdlet, function, script file, or executable program.
Check the spelling of the name, or if a path was included, verify that the path is correct and try again.
(rtx5070_env) PS E:\PyTorch_Build\pytorch> link.exe
link.exe: The term 'link.exe' is not recognized as a name of a cmdlet, function, script file, or executable program.
Check the spelling of the name, or if a path was included, verify that the path is correct and try again.
(rtx5070_env) PS E:\PyTorch_Build\pytorch>
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 设置 CUDA 路径
(rtx5070_env) PS E:\PyTorch_Build\pytorch> $env:CUDA_PATH = "E:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v13.0"
(rtx5070_env) PS E:\PyTorch_Build\pytorch> $env:PATH = "$env:CUDA_PATH/bin;$env:PATH"
(rtx5070_env) PS E:\PyTorch_Build\pytorch>
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 直接运行 CMake
(rtx5070_env) PS E:\PyTorch_Build\pytorch> cmake .. -G Ninja -DCMAKE_BUILD_TYPE=Release -DUSE_CUDA=ON ...
CMake Warning:
Ignoring extra path from command line:
"E:/PyTorch_Build"
CMake Warning:
Ignoring extra path from command line:
".."
-- The CXX compiler identification is GNU 13.2.0
-- The C compiler identification is GNU 13.2.0
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Check for working CXX compiler: C:/ProgramData/mingw64/mingw64/bin/c++.exe - skipped
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Check for working C compiler: C:/ProgramData/mingw64/mingw64/bin/gcc.exe - skipped
-- Detecting C compile features
-- Detecting C compile features - done
-- Not forcing any particular BLAS to be found
CMake Warning at CMakeLists.txt:418 (message):
TensorPipe cannot be used on Windows. Set it to OFF
CMake Warning at CMakeLists.txt:430 (message):
Libuv is not installed in current conda env. Set USE_DISTRIBUTED to OFF.
Please run command 'conda install -c conda-forge libuv=1.39' to install
libuv.
-- Performing Test C_HAS_AVX_1
-- Performing Test C_HAS_AVX_1 - Failed
-- Performing Test C_HAS_AVX_2
-- Performing Test C_HAS_AVX_2 - Success
-- Performing Test C_HAS_AVX2_1
-- Performing Test C_HAS_AVX2_1 - Failed
-- Performing Test C_HAS_AVX2_2
-- Performing Test C_HAS_AVX2_2 - Success
-- Performing Test C_HAS_AVX512_1
-- Performing Test C_HAS_AVX512_1 - Failed
-- Performing Test C_HAS_AVX512_2
-- Performing Test C_HAS_AVX512_2 - Success
-- Performing Test CXX_HAS_AVX_1
-- Performing Test CXX_HAS_AVX_1 - Failed
-- Performing Test CXX_HAS_AVX_2
-- Performing Test CXX_HAS_AVX_2 - Success
-- Performing Test CXX_HAS_AVX2_1
-- Performing Test CXX_HAS_AVX2_1 - Failed
-- Performing Test CXX_HAS_AVX2_2
-- Performing Test CXX_HAS_AVX2_2 - Success
-- Performing Test CXX_HAS_AVX512_1
-- Performing Test CXX_HAS_AVX512_1 - Failed
-- Performing Test CXX_HAS_AVX512_2
-- Performing Test CXX_HAS_AVX512_2 - Success
-- Current compiler supports avx2 extension. Will build perfkernels.
-- Performing Test CAFFE2_COMPILER_SUPPORTS_AVX512_EXTENSIONS
-- Performing Test CAFFE2_COMPILER_SUPPORTS_AVX512_EXTENSIONS - Success
-- Current compiler supports avx512f extension. Will build fbgemm.
-- Performing Test COMPILER_SUPPORTS_HIDDEN_VISIBILITY
-- Performing Test COMPILER_SUPPORTS_HIDDEN_VISIBILITY - Success
-- Performing Test COMPILER_SUPPORTS_HIDDEN_INLINE_VISIBILITY
-- Performing Test COMPILER_SUPPORTS_HIDDEN_INLINE_VISIBILITY - Success
-- Performing Test COMPILER_SUPPORTS_RDYNAMIC
-- Performing Test COMPILER_SUPPORTS_RDYNAMIC - Failed
-- Found CUDA: E:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v13.0 (found version "13.0")
CMake Error at E:/PyTorch_Build/pytorch/rtx5070_env/Lib/site-packages/cmake/data/share/cmake-4.1/Modules/CMakeDetermineCompilerId.cmake:928 (message):
Compiling the CUDA compiler identification source file
"CMakeCUDACompilerId.cu" failed.
Compiler: E:/Program Files/NVIDIA GPU Computing
Toolkit/CUDA/v13.0/bin/nvcc.exe
Build flags:
Id flags: --keep;--keep-dir;tmp -v
The output was:
1
nvcc fatal : Cannot find compiler 'cl.exe' in PATH
Compiling the CUDA compiler identification source file
"CMakeCUDACompilerId.cu" failed.
Compiler: E:/Program Files/NVIDIA GPU Computing
Toolkit/CUDA/v13.0/bin/nvcc.exe
Build flags:
-DLIBCUDACXX_ENABLE_SIMPLIFIED_COMPLEX_OPERATIONS;-Xfatbin;-compress-all
Id flags: --keep;--keep-dir;tmp -v
The output was:
1
nvcc fatal : Cannot find compiler 'cl.exe' in PATH
Call Stack (most recent call first):
E:/PyTorch_Build/pytorch/rtx5070_env/Lib/site-packages/cmake/data/share/cmake-4.1/Modules/CMakeDetermineCompilerId.cmake:8 (CMAKE_DETERMINE_COMPILER_ID_BUILD)
E:/PyTorch_Build/pytorch/rtx5070_env/Lib/site-packages/cmake/data/share/cmake-4.1/Modules/CMakeDetermineCompilerId.cmake:53 (__determine_compiler_id_test)
E:/PyTorch_Build/pytorch/rtx5070_env/Lib/site-packages/cmake/data/share/cmake-4.1/Modules/CMakeDetermineCUDACompiler.cmake:162 (CMAKE_DETERMINE_COMPILER_ID)
cmake/public/cuda.cmake:47 (enable_language)
cmake/Dependencies.cmake:43 (include)
CMakeLists.txt:853 (include)
-- Configuring incomplete, errors occurred!
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 创建测试文件
(rtx5070_env) PS E:\PyTorch_Build\pytorch> @'
>> #include <iostream>
>> int main() {
>> std::cout << "Hello from C++ & CUDA toolchain!" << std::endl;
>> return 0;
>> }
>> '@ | Set-Content test.cpp
(rtx5070_env) PS E:\PyTorch_Build\pytorch>
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 编译测试
(rtx5070_env) PS E:\PyTorch_Build\pytorch> cl.exe test.cpp /EHsc /link kernel32.lib user32.lib gdi32.lib
cl.exe: The term 'cl.exe' is not recognized as a name of a cmdlet, function, script file, or executable program.
Check the spelling of the name, or if a path was included, verify that the path is correct and try again.
(rtx5070_env) PS E:\PyTorch_Build\pytorch> if (Test-Path test.exe) {
>> .\test.exe
>> } else {
>> Write-Host "编译失败，错误信息:"
>> cl.exe test.cpp /EHsc /link kernel32.lib /v
>> }
编译失败，错误信息:
cl.exe:
Line |
5 | cl.exe test.cpp /EHsc /link kernel32.lib /v
| ~~~~~~
| The term 'cl.exe' is not recognized as a name of a cmdlet, function, script file, or executable program.
Check the spelling of the name, or if a path was included, verify that the path is correct and try again.
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 1. 设置基本环境变量
(rtx5070_env) PS E:\PyTorch_Build\pytorch> $env:CUDA_PATH = "E:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v13.0"
(rtx5070_env) PS E:\PyTorch_Build\pytorch> $env:CUDNN_PATH = "E:/Program Files/NVIDIA/CUDNN/v9.12"
(rtx5070_env) PS E:\PyTorch_Build\pytorch> $env:OpenBLAS_HOME = "E:/Libs/OpenBLAS_Prebuilt"
(rtx5070_env) PS E:\PyTorch_Build\pytorch> $env:PATH = "$env:CUDA_PATH/bin;$env:OpenBLAS_HOME/bin;$env:PATH"
(rtx5070_env) PS E:\PyTorch_Build\pytorch>
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 2. 加载 Visual Studio 环境
(rtx5070_env) PS E:\PyTorch_Build\pytorch> & "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\Build\vcvars64.bat"
**********************************************************************
** Visual Studio 2022 Developer Command Prompt v17.14.13
** Copyright (c) 2025 Microsoft Corporation
**********************************************************************
[vcvarsall.bat] Environment initialized for: 'x64'
(rtx5070_env) PS E:\PyTorch_Build\pytorch>
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 3. 添加编译器路径
(rtx5070_env) PS E:\PyTorch_Build\pytorch> $clPath = "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.44.35207\bin\Hostx64\x64"
(rtx5070_env) PS E:\PyTorch_Build\pytorch> $env:PATH = "$clPath;$env:PATH"
(rtx5070_env) PS E:\PyTorch_Build\pytorch>
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 4. 修复 Windows SDK 库路径
(rtx5070_env) PS E:\PyTorch_Build\pytorch> .\fix_kernel32_error.ps1
.\fix_kernel32_error.ps1: The term '.\fix_kernel32_error.ps1' is not recognized as a name of a cmdlet, function, script file, or executable program.
Check the spelling of the name, or if a path was included, verify that the path is correct and try again.
(rtx5070_env) PS E:\PyTorch_Build\pytorch>
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 5. 验证工具链
(rtx5070_env) PS E:\PyTorch_Build\pytorch> .\verify_toolchain.ps1
.\verify_toolchain.ps1: The term '.\verify_toolchain.ps1' is not recognized as a name of a cmdlet, function, script file, or executable program.
Check the spelling of the name, or if a path was included, verify that the path is correct and try again.
(rtx5070_env) PS E:\PyTorch_Build\pytorch>
(rtx5070_env) PS E:\PyTorch_Build\pytorch> # 6. 准备构建目录
(rtx5070_env) PS E:\PyTorch_Build\pytorch> Set-Location E:\PyTorch_Build\pytorch
(rtx5070_env) PS E:\PyTorch_Build\pytorch> Remove-Item build -Recurse -Force -ErrorAction SilentlyContinue
(rtx5070_env) PS E:\PyTorch_Build\pytorch> New-Item -Path build -ItemType Directory | Out-Null
(rtx5070_env) PS E:\PyTorch_Build\pytorch> Set-Location build
(rtx5070_env) PS E:\PyTorch_Build\pytorch\build>
(rtx5070_env) PS E:\PyTorch_Build\pytorch\build> # 7. 配置 CMake
(rtx5070_env) PS E:\PyTorch_Build\pytorch\build> .\cmake_configure_final.ps1
.\cmake_configure_final.ps1: The term '.\cmake_configure_final.ps1' is not recognized as a name of a cmdlet, function, script file, or executable program.
Check the spelling of the name, or if a path was included, verify that the path is correct and try again.
(rtx5070_env) PS E:\PyTorch_Build\pytorch\build>
(rtx5070_env) PS E:\PyTorch_Build\pytorch\build> # 8. 编译和安装
(rtx5070_env) PS E:\PyTorch_Build\pytorch\build> if ($LASTEXITCODE -eq 0) {
>> cmake --build . --config Release --parallel 8
>> if ($LASTEXITCODE -eq 0) {
>> cmake --install .
>> }
>> }
Error: not a CMake build directory (missing CMakeCache.txt)
(rtx5070_env) PS E:\PyTorch_Build\pytorch\build>
(rtx5070_env) PS E:\PyTorch_Build\pytorch\build> # 9. 验证安装
(rtx5070_env) PS E:\PyTorch_Build\pytorch\build> python -c "import torch; print(f'PyTorch版本: {torch.__version__}'); print(f'CUDA可用: {torch.cuda.is_available()}')"
PyTorch版本: 2.8.0+cpu
CUDA可用: False
(rtx5070_env) PS E:\PyTorch_Build\pytorch\build>
最新发布
本文分析了在使用Caffe进行网络训练时遇到的一个错误，该错误与特征图尺寸和池化层设置有关。通过调整输入图像大小或更改核大小可以解决此问题。
扫一扫
2118
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
