Vedansh
Use the Anaconda (base) prompt to create the new environment with a specific version of python:
Use this cheat sheet below:
At your prompt type (base) $ conda create -n py37env python=3.7
conda create - you are instructing conda to create
-n py37env - (-n is a short cut for --name; i chose py37env as the name)
python=3.7 - specifies the version of python you want the environment to have
at the base prompt you can search what versions plus build of Python is available on conda repositories with
(base) $ conda search python
Your output will show (partial shown for 3.7)
python 3.7.0 h6e4f718_3 pkgs/main
python 3.7.0 hc3d631a_0 pkgs/main
python 3.7.1 h0371630_3 pkgs/main
python 3.7.1 h0371630_7 pkgs/main
python 3.7.2 h0371630_0 pkgs/main
python 3.7.3 h0371630_0 pkgs/main
python 3.7.4 h265db76_0 pkgs/main
python 3.7.4 h265db76_1 pkgs/main
python 3.7.5 h0371630_0 pkgs/main
python 3.7.6 h0371630_2 pkgs/main
python 3.7.7 h191fe78_0_cpython pkgs/main
python 3.7.7 hcf32534_0_cpython pkgs/main
python 3.7.7 hcff3b4d_4 pkgs/main
python 3.7.7 hcff3b4d_5 pkgs/main
python 3.7.9 h7579374_0 pkgs/main
python 3.8.0 h0371630_0 pkgs/main
Now you can choose the correct version of python for tensorflow. Let us say, for example, it was 3.7.2 then
At your prompt type (base) $ conda create -n py37env python=3.7.2
If you are going to use Anaconda on Windows (or Linux), instead of the Navigator, I would recommend using the command line / terminal, asin thelong run that will be lot more helpful for data science work.