吴恩达2022版机器学习pip环境

吴恩达机器学习pip环境,对应的讲义和可运行jupyter notebook为
https://github.com/kaieye/2022-Machine-Learning-Specialization

2024/02我本机测试通过的环境如下：
python 3.9.10
pip 24.0
ipympl 0.9.3
ipywidgets 8.1.1
keras 2.8.0
matplotlib 3.7.2
notebook 6.5.6
numpy 1.23.5
protobuf 3.20.0
pytorch-lightning 1.6.5
scikit-learn 1.2.1
tensorboard 2.8.0
tensorflow 2.8.0
torch 1.13.1

某些关键步骤：

1.卸载jupyter noteboook 7(如有)，安装notebook 6
pip uninstall notebook
pip install notebook==6.5.6

2.安装notebook扩展,使得matplotlib图表可以交互
pip install jupyter_contrib_nbextensions
jupyter contrib nbextension install --user
pip install jupyter_nbextensions_configurator
jupyter nbextensions_configurator enable --user
pip install ipympl

3.启动
jupyter notebook

4.第二部分神经网络需要安装tensorflow，安装如下老版本可以运行
tensorflow 2.8.0
keras 2.8.0
protobuf 3.20.0