gabrielt-prathivys / atoms-ai Goto Github PK

Cut out existing complex codes, to create the simplest possible interaction of 3-5 kinds of atoms, in a interface that the A.I. can try to solve simple puzzles and get a score.

• Show to the A.I. that we want water, and a box with Hydrogen, Oxygen and some tools (yeah, let's simplify and get easy in the beginning)

EXABYTE - Data about chemical elements arranged in JSON format.
https://github.com/Exabyte-io/periodic-table.js

EXABYTE - JSON schemas and examples representing structural data, characteristic properties, modeling workflows and related data about materials
https://github.com/Exabyte-io/esse

EXABYTE - Design Materials from atoms up - https://exabyte.io/
https://github.com/exabyte-io/

PYSCF - Python Module do Quantum Chemistry
https://github.com/pyscf/pyscf

PYQUANTE2 - Another Python Module do Quantum Chemistry
https://github.com/rpmuller/pyquante2

@losnol suggestion:
Particles want to get the less energetic possible. One solution (simplifying a lot) it's to use particle swarm optimization to cauculate the local minimum for the potential energy of each particule and check all particle configurations. The problem with this solution is the complexity O(particules*number_of_possible_orbitals) that can be huge. Probably the AI will have to cut some branchs in this scope and get some common patterns. Trying to validade and make less "ifs" and checks, so it reduce the avarage complexity and the time of execution.