Skip to content

A playground to explore optimal Mario Kart 8 combos via Pareto Optimization and fancy plots.

License

Notifications You must be signed in to change notification settings

tejashah88/mk8d-combo-playground

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

14 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Mario Kart 8 Deluxe - Combo Playground

Live Website: https://app.mk8d-combo-playground.com

Screenshot of Kart Combo Playground - 3D View

A playground to explore optimal Mario Kart 8 build via Pareto Optimization and fancy plots. This repository contains 2 notebooks, one for scraping the in-game stats from http://japan-mk.blog.jp/ (as of August 16th, 2023) and one for setting up the combo playground visualizer.

NOTICE: The plots on the live website will take a few seconds to regenerate when changing the parameters or controls. PLEASE. BE. PATIENT.

Features

Let's say you want to find the best kart build based on 2 or 3 in-game stats. With this playground you can...

  • Visualize all possible builds and see how they stack against each other based on selected stats. Screenshot of Kart Combo Playground - All Combos View
  • View all pareto-optimal combos and find the best one based on what stats matter to you (via the importance weights). Screenshot of Kart Combo Playground - Pareto Combos View
  • Click on any scatterpoint to see a table of all the possible builds matching the selected stats. Screenshot of Kart Combo Playground - Selected Point Lookup Table
  • Choose your preferred build and see how it stacks up against the rest. Screenshot of Kart Combo Playground - Your Combo View
  • Compare in-game statistics for each part category (characters, frames, tires, and gliders) via the comparision heatmap. Screenshot of in-game statistics for each part category

Notes

  • The coloring for all combos is defined such that the middle color (yellow) is defined at the 70th percentile point to exaggerate the differences for near-meta builds.

Setup

Option 1: Commandline (recommended)

  1. (Setup & activate virtual python 3.x environment)
  2. pip install -U pip
  3. pip install -r requirements.txt
  4. jupyter nbconvert --execute --debug --clear-output notebooks/01_scrape_data.ipynb
  5. panel serve notebooks/02_analyze_combos.ipynb --warm --reuse-sessions --global-loading-spinner

Option 2: Jupyter Lab

  1. (Setup & activate virtual python 3.x environment)
  2. pip install -U pip
  3. pip install -r requirements.txt
  4. jupyter lab
  5. (Run all cells in notebooks/01_scrape_data.ipynb)
  6. (Run all cells in notebooks/02_analyze_combos.ipynb)

More resources / Credits

Releases

No releases published

Packages

No packages published