How to downgrade Python from 3.7 to 3.5 in Anaconda [closed]
You can create a new environment for your experiments:
conda create -n new_environment python=3.5
Or install anaconda with another python version (http://docs.anaconda.com/anaconda/user-guide/faq/#how-do-i-get-the-latest-anaconda-with-python-3-5)
conda install downgrade python version
If you want to set specific version, use it like this:
WARNING: This command will overwrite the default python version system-wise
conda install python=3.6
To create environment with a specific version, you can do:
conda create -n $PYTHON36_ENV_NAME python=3.6 anaconda # set custom env name
anacondaat the end allows the env to use all anaconda packages
For more information refere to Anaconda documentation
Anaconda cannot downgrade python to 3.6, and cannot install tensorflow
Anaconda wants to update itself and use new version of python. Usually I don't touch my base environment and just let anaconda update it. I don't install any extra packages into the base environment.
If I want to run a script that needs an old version of python, I create a new environment with that old version, and then use this new environment to run the script:
conda create --name py2 python=2.7
conda create --name py3 python=3.5
See How to run anaconda2 python on anaconda3 python
How to downgrade to Python 3.4 from 3.5
Uninstall Python 3.5 by going to control panel, uninstall a program, and then look for Python 3.5, click on it and press uninstall. After that download Python 3.4. I don't know any other ways but I think that should work.
How to change python version in Anaconda?
A better (recommended) alternative is to create a virtual environment of the desired Python version and then use that environment to run Tensorflow and other scripts.
To do that, you can follow the instructions given here.
BUT, if you don't want to create a separate environment, then
conda install python=<version> should do.
OR (not recommended) you can download the "latest" Anaconda installer with your required Python version bundled.