[loxias]

I wish alphago was more "democratized" -- that is to say, I have many questions and experiments I'd love to run on it (a friend of mine and I have frequently pondered Go played in various different topological spaces, and I'd love to see an AI's result, for example).

[kadoban]

Look into Katago. It's an open source AI in the same general style as AlphaGo, with an empasis on training speed. On 9x9 you can get up to superhuman really quickly on just a decent home machine (I think hours/days, can't remember exactly and it's probably improved since I looked).

[elefantastisch]

You can also just download pre-trained models. Get those set up and then install Sabaki (https://sabaki.yichuanshen.de/) and connect it to your KataGo... instant (ok, a few hours probably if it's your first time setting it up) superhuman Go AI. There's even an npm package you can use to process SGF files and automatically score moves as good/questionable/bad + generate variations that were better choices: https://github.com/9beach/analyze-sgf/blob/master/README.en-...

(Edit: Misread what the other poster was trying to do, but I'm leaving this here as a reference for anyone else who just wants to use KataGo on their own machine on their own Go games.)

[loxias]

Awesome! Thanks! Checking it out now.

[franknstein]

Fun idea. Did you reach any interesting conclusions?