anaconda really useful, eliminating a lot of trouble, and now I personally recommend the direct use anaconda.
anaconda features: a plurality of python may be present in the environment, to use a certain environment, then it needs to switch to the environment, installing, uninstalling the package are carried out in a particular environment. The so-called environment is actually different from python and install packages it in a different directory, and the environment is to switch to switch through the catalog.
Basic operation:
Create an environment python3.4: Conda the Create --name = 3.4 python34 Python activation: of an activate python34 # for Windows Source of an activate python34 for Linux & Mac # # If you want to return to the default setting, run deactivate python34 # for Windows Source deactivate python34 for # Mac & Linux # List all environmental Conda env List # delete an existing environment conda remove --name python34 --all
Connection vscode:
Similarly clicking the lower left corner of the gear, selection comand palette, choice python interpreter which will automatically connect the anaconda environment. (At the time of installation Note To check the recommended option)
Solution powershell script execution is prohibited: The administrator permission to open powershell, and then execute the command Set-ExecutionPolicy -ExecutionPolicy RemoteSigned
For details refer to: https://blog.csdn.net/qq_42739865/article/details/88855495
How comes jupyter switch Kernel :
conda create -n py3 python = 3 # Create a python3 environment, called py3 of an activate py3 # activate py3 environment conda install ipykernel # installation ipykernel module python -m ipykernel install --user --name py3 --display -name "py3 "# configure jupyter notebook # start jupyter notebook, and then in the" New "in the kernel will have a py3
Installation package
You can use conda or pip install, but not both used interchangeably, what is installed on what to uninstall.
Installation tensorflow:
Cpu version Installation: enter the corresponding ambient
pip install --upgrade --ignore-installed tensorflow
If errors are reported missing msgpack, pip install it
If an error occurs it: Could not install Packages Standard Package Due to AN EnvironmentError: [WinError 5] Access Denied: Just after install add --user
Run the program appears about warning: Your the CPU that the this TensorFlow the Supports Instructions Compiled binary WAS not to use: AVX2
Solution Reference: https://blog.csdn.net/hq86937375/article/details/79696023
supplement:
anaconda specific installation procedures and use refer to: https://blog.csdn.net/ITLearnHall/article/details/81708148