search

Ambrosys / glyph

a python 3 library based on deap providing abstraction layers for symbolic regression problems.
GNU Lesser General Public License v3.0
91 stars 16 forks source link

please add tensorflow 2.0 keras #65

Open bionicles opened 4 years ago

bionicles commented 4 years ago

hi,

Do you like AI? im sure many users and researchers and developers would appreciate the ability to apply genetic programming to neural networks / computation graphs... but i find support for this seemingly obvious idea to be really lacking in 2019. This looks like a cool project. I would like to use it, but if we can't hook it up to our existing tensorflow / keras stuff, it seems we can't use Glyph!

How hard do you think it would be to add tf.keras support to glyph so we could make computation graphs that run fast with tensorflow and have strongly typed / SHAPE CHECKING (to avoid infinity bugs and frustration) for tensors?

sohrabtowfighi commented 4 years ago

@bionicles , i think you are in the wrong repository - this glyph package is focused on symbolic regression. try: https://github.com/joeddav/devol

Ohjeah commented 4 years ago

We are thinking about dedicated tensorflow support by providing a suitable primitive set and examples.

Besides that, other concepts (algorithms, genetic operations) will work right out of the box.