macos安装metal 加速版 pytorch

categories: [Python]

tags: Python MacOS

写在前面

试试 m3 的 metal 加速效果如何

Mac computers with Apple silicon or AMD GPUs

macOS 12.3 or later

Python 3.7 or later

Xcode command-line tools: xcode-select --install

安装 Python: conda-forge

bash 复制代码

brew install miniforge

镜像

yaml 复制代码

channels:
  - defaults
show_channel_urls: true
auto_activate_base: false
ssl-verify: false
default_channels:
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/r
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/msys2
custom_channels:
  conda-forge: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
  msys2: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
  bioconda: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
  menpo: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
  pytorch: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
  pytorch-lts: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
  simpleitk: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
  deepmodeling: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/

安装

开一个新的虚拟环境, 这个是重点, 如果不开的话, 原有的环境会污染 C 库的链接, 所以这一步是必须的

On Mac OS X, import numpy complaining about "Library not loaded: @rpath/libgfortran.3.dylib" · Issue #12970 · numpy/numpy;

这个方案不彻底, 直接卸载 numpy 然后重装不能解决问题...

bash 复制代码

conda create -n py3xi python=3.11
conda activate py3xi
# conda update --all -c conda-forge # optional
# 重点: 
conda install pytorch torchvision torchaudio -c pytorch-nightly

然后测试

Accelerated PyTorch training on Mac - Metal - Apple Developer;

python 复制代码

import torch
if torch.backends.mps.is_available():
    mps_device = torch.device("mps")
    x = torch.ones(1, device=mps_device)
    print (x)
else:
    print ("MPS device not found.")
'''
tensor([1.], device='mps:0')
'''

可以在 MacOS 上跑深度学习了.