For the machine learning notebook and pre-processing scripts, see https://github.com/alexdriedger/SlayTheSpireFightPredictor
Slay the Spire + Neural Networks. Available on the Steam Workshop!
Slay-I uses a machine learning model trained on over 325, 000 fights!
Slay-I evaluates cards based on average damage saved in a fight if the card is added to your deck.
Slay-I also evaluates upgrading and removing cards
Data is from Spire Logs and Jorbs.
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!