Mac M1开发环境配置---tensorflow

1. 安装conda环境， miniconda

这里采用的Miniconda（精简版Anaconda）下载地址：Miniconda — Conda documentation 

安装完成后，需要运行～/.bash_profile, 路径设置才能起作用

2. 建立python虚拟开发环境

可以python3.8或者3.9

我们这里用的3.9

conda create -n py39tf python=3.9 

3. 安装tensorflow包支持

conda install -c apple tensorflow-deps # Step 1: Environment setup Install the TensorFlow dependencies: 
pip install tensorflow-macos # Step 2: Install base TensorFlow 
pip install tensorflow-metal # Step 3: Install tensorflow-metal plugin

4. 环境测试验证

(py39) myself@192 tensorflow-dev % python
Python 3.9.12 (main, Jun  1 2022, 06:34:44) 
[Clang 12.0.0 ] :: Anaconda, Inc. on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow as tf
tf.te>>> tf.test.is_gpu_available()
WARNING:tensorflow:From :1: is_gpu_available (from tensorflow.python.framework.test_util) is deprecated and will be removed in a future version.
Instructions for updating:
Use `tf.config.list_physical_devices('GPU')` instead.
Metal device set to: Apple M1
 
systemMemory: 8.00 GB
maxCacheSize: 2.67 GB
 
2022-07-27 20:32:19.519444: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:305] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2022-07-27 20:32:19.520494: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:271] Created TensorFlow device (/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: )
True
>>> tf.config.list_physical_devices('GPU')
[PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]
>>> 

 搞定。