Installing Tensorflow when Python 3.9 is installed on Path?

Question

2 Versions of Python 3.9 and 3.8.6 are installed on Win 10. I Also want Tensorflow installed. But Pip recognises only Py 3.9 and hence does not installs tf locally. Is there a way out for this.

Solution 1

Try this command

python -m pip install --upgrade https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-1.12.0-py3-none-any.whl

Solution 2

python -m pip install --upgrade https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-1.12.0-py3-none-any.whl

This work with my python 3.9

Solution 3

The following command should be work for python version 3.9: python -m pip install --upgrade https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-1.12.0-py3-none-any.whl

Solution 4

This worked for me:

python3.8 -m pip install tensorflow

I think pip uses the latest python version, so you have to tell pip to install with python3.8 or use the beta version of tensorflow:

pip install tf-nightly

Solution 5

For dealing with multiple Python Versions it is highly recommended to use virtual environments. It makes live much easier. For a detailed guide how to set them up read this: https://www.freecodecamp.org/news/manage-multiple-python-versions-and-virtual-environments-venv-pyenv-pyvenv-a29fb00c296f/

View more solutions

Comments

[Divyessh]

source the path to the bin of python

[Djib2011]

which python do you want to install tensorflow for?

[hoefling]

py -3.8 -m pip install tensorflow

[danihelovij]

@hoefling It works, thank you

[usman mohammed]

use this : stackoverflow.com/questions/64447565/…

[Suthiro]

I suppose it will use python3.9, isn't it?

[RobbieTheK]

This appears to be for Mac OS. I got this error on Fedora: ImportError: /usr/local/lib/python3.9/site-packages/tensorflow/python/_py‌​wrap_tensorflow_inte‌​rnal.so: invalid ELF header During handling of the above exception, another exception occurred:

[EpixyCoder101]

@RobbieTheK that's odd.. seems to work on my Windows system.

[brunocrt]

the last command worked for me because I was using Python 3.9 and will only work with the most recent version/builds of tensorflow (tf-nightly) thanks for the help!