如何在Mac M1和M2芯片上部署TensorFlow并启动GPU支持

默认你已经会安装anaconda和搭建虚拟环境。

在虚拟环境下，我们先使用如下命令安装tensorflow-macos版本：

pip install tensorflow-macos

pip install tensorflow-metal

结束后，我们还需要更新numpy之类的库来和tensorflow-macos兼容，命令如下：

pip install numpy==1.24.3
pip install pandas  --upgrade
pip install matplotlib  --upgrade
pip install scikit-learn  --upgrade
pip install scipy  --upgrade
pip install plotly  --upgrade

此时，大功告成。我们来测试下：

import sys
import tensorflow.keras
import pandas as pd
import sklearn as sk
import scipy as sp
import tensorflow as tf
import platform
print(f"Python Platform: {platform.platform()}")
print(f"Tensor Flow Version: {tf.__version__}")
# print(f"Keras Version: {tensorflow.keras.__version__}")
print()
print(f"Python {sys.version}")
print(f"Pandas {pd.__version__}")
print(f"Scikit-Learn {sk.__version__}")
print(f"SciPy {sp.__version__}")
gpu = len(tf.config.list_physical_devices('GPU'))>0
print("GPU is", "available" if gpu else "NOT AVAILABLE")

这时，运行会出现如下的画面：

总结：安装需要三条命令：

pip install tensorflow-macos
pip install tensorflow-metal
pip install numpy==1.24.3
pip install pandas  --upgrade
pip install matplotlib  --upgrade
pip install scikit-learn  --upgrade
pip install scipy  --upgrade
pip install plotly  --upgrade