Welcome to Smogon! Take a moment to read the Introduction to Smogon for a run-down on everything Smogon, and make sure you take some time to read the global rules.
Yeah, I wouldn't expect it to work out of the box with the same NN architectures, but I think the key insight of self-play combined with deep learning will be the only possible path to a world-champion Pokemon AI. Can't find the link now but pretty sure David Silver said in his Deep RL course...
I'm working on a project in which I'll need to simulate tons of games between bots and am going to need to write some tools to interact with the simulator CLI. Before I do that I wanted to see if anyone had any repos or previous projects that already did that though. Preferably in python but...
If you're looking into MCTS then a better route might be a self-play model using deep reinforcement learning (an architecture similar to AlphaGo Zero). These models do not require any historical play data and outperform historical data models over time. The showdown simulator conmmand-line...
I've been exploring the play.pokemonshowdown.com/data directory to do some data analysis. However, there's one piece I'm missing which is the mon list for each format. I'm most interested in VGC but others would be nice as well. Is there a data file somewhere that defines which mons are legal...
I run an R blog mainly for VGC and worked with a player to provide insights based on parsing replay files. You can read more about that here: https://pokemon-data-analysis.netlify.app/posts/2022-04-25-moneyball-vgc-featuring-chef/
I've been working on automating this type of analysis so any...
I built a parser from replay data in R and ran a team clustering algorithm on it. Most of the info is in my blog:
Clustering: https://pokemon-data-analysis.netlify.app/posts/2022-03-28-meta-cores/
Parsing (click "hide" buttons to reveal code)...