Skip to content

nick-kann/Xatu-AI

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

32 Commits
 
 
 
 
 
 
 
 

Repository files navigation

What is Xatu AI?

Xatu AI is a project focused on developing a machine learning model to predict the first Pokémon a player is likely to send out in Pokémon Showdown. This gives users a signifcant edge by allowing them to instantly counter their opponent's first move. The name draws inspiration from the Pokémon Xatu, a psychic bird known for its ability to see into the future with its right eye.
Xatu

Source: Pokémon.com

Xatu AI Project Checklist 📋

  • Fetch Data 🌐
    Retrieve game replays from the Pokémon Showdown server to use for training

  • Prepare Dataset 🗂️
    Process the raw data into a workable dataset suitable for model training

  • Create/Train Model 🔧
    Find the best ML algorithm, hyperparameters, and feature transformations to create/train the model.

    • Current best-performing model: CatBoost with 46.7% accuracy.
  • HuggingFace Deployment 🤗
    Deploy the model to HuggingFace

  • HuggingFace Deployment 🤗
    Deploy my model to HuggingFace

  • Website Deployment 🌐
    Use Streamlit to create a UI to make it easy for anyone to use

Optional:

  • Chrome Extension Integration 🚀

About

WIP: training a model to predict which Pokemon a player is most likely to send out first

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors