Looks cool, but: "This installation is compatible with Linux/Mac OS X and Python 2.7."
Sad face.
Sad face.
Is Python 2.7 requirement really a big deal for mostly using other people's code (adding a bit of your own application layer code)?
I use Anaconda and just have different environments for 2.7 and 3.x, TensorFlow or spaCy, or NLTK, etc.
Really easy to switch to the environment I need.
The 3.x is catching up, but still not fast enough as the community wants it to. But 2.7 is here for a reason, and shaming people that used 2.7 to push 3.x adoption, in my opinion is a bad practice.
If either their library (or a dependency) depends on something which binds to C, it could take some effort to make it python 3 compatible.
(FWIW, I use 2 or 3 depending on the OS or libraries that I am dealing with)