Tensorflow和Keras安装流程,jupyter无法使用keras解决方案

1.Base: anaconda https://www.anaconda.com/download

2.安装python=3.8，Tensorflow=2.13.0，Keras=2.13.1

–创建conda环境-在Anaconda Prompt中输入命令， conda create -n tensorflow python=3.8

–激活环境 activate tensorflow

–安装Tensorflow pip install tensorflow==2.13.0 -i https://pypi.tuna.tsinghua.edu.cn/simple

–安装Keras pip install keras==2.13.1 -i https://pypi.tuna.tsinghua.edu.cn/simple

–在tensorflow环境中，进入python，输入import tensorflow 和 import keras测试

3.打开jupyter notebook，运行import keras会报错：‘No module name tensorflow’ ，jupyter notebook内核没有切换环境。

在Anaconda Prompt中操作：

–安装ipython conda install ipython

–安装jupyter conda install jupyter

–查看可用kernel：jupyter kernelspec list

–在tensorflow虚拟环境下，安装ipykernel： conda install ipykernel

python -m ipykernel install --name tensorflow

运行jupyter notebook：jupyter notebook

Success!

tensorflow、python版本对应关系 https://tensorflow.google.cn/install/source?hl=en#cpu