tony/tensorneat-mend
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
67 Commits 2 Branches 0 Tags
7265e33c43cff81607aa297925fd843c9b12f278
Go to file
Clone
Open with VS Code Open with VSCodium Open with Intellij IDEA
Download ZIP Download TAR.GZ Download BUNDLE

README.md

NEATAX: Tensorized NEAT Implementation for Parallel Hardware Accelaration

NEATAX is a powerful tool that utilizes JAX to implement the NEAT (NeuroEvolution of Augmenting Topologies) algorithm. It provides support for parallel execution of tasks such as forward network computation, mutation, and crossover at the population level.

Performance

One of the standout features of NEATAX is its speed. Compared to traditional CPU implementations, NEATAX significantly improves the efficiency of the NEAT algorithm. It has been observed to boost the running speed of the NEAT algorithm by more than 10 times, offering considerable advantage in larger-scale and time-sensitive applications.

Installization

by git clone need JAX environment

Reference in New Issue View Git Blame Copy Permalink
Description
MEND: Modular Evolutionary Neuroduplication — TensorNEAT fork with module duplication operator
Readme 36 MiB
Languages
Python 54.3%
Jupyter Notebook 45.7%