can't install older TensorFlow versions

[sloganking]

after running pip install --ignore-installed --upgrade tensorflow-gpu==1.15.0 I get:

ERROR: Could not find a version that satisfies the requirement tensorflow-gpu==1.15.0 (from versions: 2.2.0rc1, 2.2.0rc2, 2.2.0rc3, 2.2.0rc4, 2.2.0, 2.3.0rc0, 2.3.0rc1, 2.3.0rc2, 2.3.0)
ERROR: No matching distribution found for tensorflow-gpu==1.15.0

It seems pip no longer offers older TensorFlow versions by default. Is there a way to get around this?

[badjano]

@SleepingFox88 take a look at this https://stackoverflow.com/questions/41937915/how-to-pip-install-old-version-of-librarytensorflow Another approach could be using Docker, like this: https://hub.docker.com/layers/tensorflow/tensorflow/1.3.0-py3/images/sha256-d2775f748ed3caded7a7c2b7d86ec8292d16ece7836457bc9a26dd3336dd2ca6?context=explore But you´d have to use a dockerfile to use that image and pull this repo into it

[Sanaxen]

Please try the following methods. https://github.com/Sanaxen/tecoGAN_app/blob/master/requirements.txt

pip install -r required.txt In my case, this worked fine. (I tried it on 2020-09-28)

[mirh]

You won't find tf1.15 on pypi if you have python >3.7. https://stackoverflow.com/questions/61491893/i-cannot-install-tensorflow-version-1-15-through-pip/66158049

You'll either need an unofficial tf build, or update tecogan