Hi everyone!
One talented Godot developer and I, we are creating a game related to classical Asian board game named Go/Weiqi/Baduk (https://en.wikipedia.org/wiki/Go_(game)) that becomes very famous in non-Asian world after the triumph of AlphaGo and AlphaGo Zero. We are not intended to beat the AlphaGo ;), but rather add some interesting mechanics. The main is to be able to play on a custom board (named Goban). Rather than play just on boring squared 19x19 board, you can create your own or use pre-created one, like a map of Japan, China, or another country and try yourself on the role of strategist/general capturing the country (not the squared abstraction =) ).
For now, MVP of the game is almost ready, but we are facing the next problem. The game is fine for two-person, but it would be much more interesting to add an AI to be able to play a single-player mode. For this purpose, we are trying to use OpenAI Gym (https://github.com/lupoglaz/GodotGymAI) and Tensorflow 2.0 (https://github.com/godot-extended-libraries/godot-tensorflow-workspace) but without any success yet. So, colleagues, maybe you would like to help us if you have experience in the AI field with Godot and try to accomplish the task for basic AI implementation (not necessarily a "strong player", but not a "random bugger"). Of course, we may discuss a budget for task completion. So please let me know.
Best wishes,
Igor
