Installing specific tensorflow branch with pip

Question

I am working on this project which was tested with tensorflow v0.10.0rc0.

I don't want to take any dependency risks, and I'm a bit confused about this syntax version (instead of less cryptic 1.0, 1.1, 1.2, 1.3and 1.4 versions)

so how do I safely install this v0.10.0rc0 version using a pip command?

Answer 1

You can see the history of official pip dependencies by doing the following:

pypi.python.org/simple/<nameofpackage>/

The one you're interested in is linked here. These are all the available packages and there are some v0.12 versions and alike there you can try.

To install a specific version of the package, do the following:

pip install -I tensorflow==<version>

NOTE: You can also install packages which are hosted externally by replacing tensorflow== with the HTTP link.

Answer 2

To install a specific version with pip:

pip install tensorflow==<VERSION>

However, the specific version you want is not available via pip and I'm afraid you're going to have to compile from source if you want to use it.

Available pip3 versions (nov 18, 2017):

0.12.1, 1.0.0, 1.0.1, 1.1.0rc0, 1.1.0rc1, 1.1.0rc2, 1.1.0, 1.2.0rc0, 1.2.0rc1, 1.2.0rc2, 1.2.0, 1.2.1, 1.3.0rc0, 1.3.0rc1, 1.3.0rc2, 1.3.0, 1.4.0rc0, 1.4.0rc1, 1.4.0

Available pip2 versions (nov 18, 2017):

0.12.0rc0, 0.12.0rc1, 0.12.0, 0.12.1, 1.0.0, 1.0.1, 1.1.0rc0, 1.1.0rc1, 1.1.0rc2, 1.1.0, 1.2.0rc0, 1.2.0rc1, 1.2.0rc2, 1.2.0, 1.2.1, 1.3.0rc0, 1.3.0rc1, 1.3.0rc2, 1.3.0, 1.4.0rc0, 1.4.0rc1, 1.4.0