8000 GitHub - pncakespoon1/SpeedrunStatsTools
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

pncakespoon1/SpeedrunStatsTools

Repository files navigation

SpeedrunStatsTools

Tools which I'm working on to track stats of runs beyond what the typical tracker does

Links

Rawalle

Grind-Simulator

Specnr's Reset Tracker

Talking Mime's Reset Tracker

Efficiency Score Calculator

Stats Analysis Website

Dependencies

pillow, tensorflow, numpy, cv2, pygsheets, twitch-dl, seaborn, scipy, pandas, matplotlib, keras-tuner, keras, pathlib, shutil

Requirements

ffmpeg.exe and credentials.json in the folder, cuda

Scripts

analyzeVod.py

This is the heart of the program, it downloads a bunch of vods and writes various stats to a google sheet. It requires models in the models folder to detect buried treasures.

makeImages.py

This script downloads a frame from every second of a vod specified to the unsorted_images folder. It requires a vod in the vods folder.

breakWalls.py

This script "uncollages" wall images

trainModel_bgss.py

This script trains a model for classifying specific traits of spawn images. It requires images to be sorted in the designated folders in the spawn_image_classification folder. It stores the models in the models folder.

trainModel_bt.py

This script trains a model for identifying buried treasure chest loot guis. It requires images to be sorted in the designated folders in the bt_identification folder. It stores the model in the models folder.

trainModel_wall.py

This script trains a model for identifying wall interfaces. It requires images to be sorted in the designated folders in the wall_identification folder. It stores the model in the models folder.

Credits

  • Emily
  • Ravalle
  • Specnr
  • Pokemon
  • TalkingMime
  • Lemon
  • Sharpie
  • Llama
  • Peej

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published
0