Skip to content

Machine learning in Slay the Spire to predict fight outcomes and evaluate cards. Available on the Steam Workshop!

Notifications You must be signed in to change notification settings

alexdriedger/SlayTheSpireFightPredictor

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

28 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Slay-I

For the mod that runs inference on the model in the game, see https://github.com/alexdriedger/SlayTheSpireFightPredictorMod

Slay the Spire + Neural Networks. Available on the Steam Workshop!

Slay-I uses a machine learning model trained on over 325, 000 fights!

Predict Damage Taken in a Fight

Ironclad Fight Prediction

Silent Fight Prediction

Evaluating Cards

Slay-I evaluates cards based on average damage saved in a fight if the card is added to your deck.

Silend Card Reward

Ironclad Card Reward

Upgrades and Removals

Slay-I also evaluates upgrading and removing cards

Silent Card Removal

Data Set

Data is from Spire Logs and Jorbs.

Accuracy

Slay-I is generally within +/- 7 HP of how much damage you will take in a fight. For adding, removing, and upgrading cards, the score is the average damage saved in a fight. The model isn't perfect, but it does a pretty good job at evaluating cards!

About

Machine learning in Slay the Spire to predict fight outcomes and evaluate cards. Available on the Steam Workshop!

Resources

Stars

Watchers

Forks