49 lines
1.2 KiB
Markdown
49 lines
1.2 KiB
Markdown
查看当前你的电脑显卡支持的最高CUDA版本，后面的安装不能超过它：
|
||
nvidia-smi
|
||
|
||
安装Visual Studio
|
||
|
||
下载cuda toolkit
|
||
CUDA Toolkit Archive | NVIDIA Developer
|
||
Windows默认路径：
|
||
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA
|
||
|
||
查看CUDA是否安装成功
|
||
nvcc -V
|
||
|
||
下载cuDNN
|
||
https://developer.nvidia.com/cudnn-downloads
|
||
解压覆盖至cuda目录
|
||
|
||
通过NVIDIA提供的 deviceQuery.exe 和 bandwidthTest.exe 来查看GPU的状态，两者均在安装目录的 extras\demo_suite文件夹中
|
||
|
||
|
||
|
||
conda create -n yolov10 python=3.9
|
||
conda activate yolov10
|
||
|
||
|
||
|
||
|
||
conda create -n yolov10 python=3.12.4
|
||
|
||
安装pytorch
|
||
# CUDA 12.4
|
||
|
||
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu124
|
||
|
||
conda install pytorch torchvision torchaudio pytorch-cuda=12.4 -c pytorch -c nvidia
|
||
|
||
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu124
|
||
|
||
|
||
# conda install pytorch torchvision torchaudio pytorch-cuda=12.4 -c pytorch -c nvidia
|
||
# CPU Only
|
||
conda install pytorch==2.4.0 torchvision==0.19.0 torchaudio==2.4.0 cpuonly -c pytorch
|
||
|
||
在txt中删掉 pytorch、torchvision、torchaudio
|
||
|
||
pip install -r requirements.txt
|
||
pip install -e .
|
||
|