2

It says: Stable-Baselines supports Tensorflow versions from 1.8.0 to 1.15.0, and does not work on Tensorflow versions 2.0.0 and above.

So I tried to install "sudo pip3 install tensorflow==1.15.0"

But I get the message: ERROR: Could not find a version that satisfies the requirement tensorflow==1.15.0 (from versions: 2.2.0rc1, 2.2.0rc2, 2.2.0rc3, 2.2.0rc4, 2.2.0) ERROR: No matching distribution found for tensorflow==1.15.0

I'm using: Ubuntu 20.04 LTS Python 3.8.2 pip 20.1.1 from .../python3.8/site-packages/pip (python 3.8)

How can I install a tensorflow version that works with stable-baselines?

Caspar Bm
  • 258
  • 4
  • 12
  • 2
    Most likely because you are using Python 3.8 and TF 1.x don't support it. Edit: yepp, I see it from `pip --version` output. Use Python 3.7 and you should be fine. – hoefling May 27 '20 at 12:16
  • Thanks. Sadly this didn't fix my bug. I created a virtualenv with python 3.7.7 (and therefore also pip used python 3.7) but I still get the same message – Caspar Bm May 27 '20 at 21:19
  • 1
    First, double-check that `pip` refers to one from the 3.7 venv, then run `python -c "from pip._internal.utils.compatibility_tags import get_supported; print(get_supported()[0])"` and post the output. You may need to update `pip` for that command to work; in older versions, the import is `pip._internal.pep425tags`. – hoefling May 28 '20 at 16:45

1 Answers1

0

I solved the problem by using anaconda instead. Thanks for the help!

Caspar Bm
  • 258
  • 4
  • 12