Mac M1 conda+python+tensorflow安装

安装conda

安装arm64的miniconda
https://github.com/conda-forge/miniforge
选择arm64 (Apple Silicon)下载
执行
bash Miniforge3-Linux-x86_64.sh
验证 conda --version
添加源

vim .condarc
添加下面到文件中

#auto_activate_base: false
channels: [https://mirrors.ustc.edu.cn/anaconda/cloud/menpo/, https://mirrors.ustc.edu.cn/anaconda/cloud/bioconda/,
  https://mirrors.ustc.edu.cn/anaconda/cloud/conda-forge/,
  https://mirrors.ustc.edu.cn/anaconda/pkgs/free/, https://mirrors.ustc.edu.cn/anaconda/pkgs/main/,
  https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/, https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge/,
  https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/msys2/, conda-forge]
show_channel_urls: true  

conda config --show # 查看配置信息

channel_priority: flexible
channels:
  - https://mirrors.ustc.edu.cn/anaconda/cloud/menpo/
  - https://mirrors.ustc.edu.cn/anaconda/cloud/bioconda/
  - https://mirrors.ustc.edu.cn/anaconda/cloud/conda-forge/
  - https://mirrors.ustc.edu.cn/anaconda/pkgs/free/
  - https://mirrors.ustc.edu.cn/anaconda/pkgs/main/
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/conda-forge/
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/msys2/
  - conda-forge
client_ssl_cert: None
client_ssl_cert_key: None

如果channels下面有 - default 需要删除

新建python3.8环境

conda create -n tensorflow-cpu python=3.8
conda activate tensorflow-cpu
conda install pip

tensorflow包下载

下载为m1优化的tensorflow及其插件
https://github.com/apple/tensorflow_macos/releases
下载后解压，进入tensorflow_macos/arm64目录的终端
执行下面语句

我下载的版本是tensorflow_macos-0.1alpha3.tar.gz

pip install --upgrade --no-dependencies --force numpy-1.18.5-cp38-cp38-macosx_11_0_arm64.whl grpcio-1.33.2-cp38-cp38-macosx_11_0_arm64.whl tensorflow_addons_macos-0.1a3-cp38-cp38-macosx_11_0_arm64.whl

pip install absl-py astunparse flatbuffers gast google_pasta keras_preprocessing opt_einsum protobuf tensorflow_estimator termcolor typing_extensions wrapt wheel tensorboard typeguard

pip install --upgrade --no-dependencies tensorflow_macos-0.1a3-cp38-cp38-macosx_11_0_arm64.whl

如果最终提示Successfully installed tensorflow-macos-…则安装成功。

# 打开python进行验证
import tensorflow as tf

tf.__version__