Skip to content

Implementation of the paper, Artificial Neurogenesis, by Mixter et al, to grow neural networks to allow for dynamic scaling of neurons in a network rather than pruning. Request access if required

License

Notifications You must be signed in to change notification settings

rohanphil/Artificial-Neurogenesis

Repository files navigation

Artificial-Neurogenesis

Implementation of the paper, Artificial Neurogenesis, by Mixter et al.

Artificial Neurogenesis allows for the creation of scalable models that grow rather than prune pre-trained models. Request access if required.

Steps to create and use python bindings.

clone the pybind repository into this folder

mkdir build

cd build

cmake .. && make

cd out of build

(Alter the contents of the CMAKE file. It is currently set to compile both the pybind11test.cpp file and the ANG.cpp file)

Open a python session in this folder and run:

from build.example1 (or any other name) import *

About

Implementation of the paper, Artificial Neurogenesis, by Mixter et al, to grow neural networks to allow for dynamic scaling of neurons in a network rather than pruning. Request access if required

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published