Repository navigation
FRLG League sim: planned policy (plan JSON + override table), plan search, console regression cases - #111
Draft
dayfine wants to merge 1 commit into
Draft
FRLG League sim: planned policy (plan JSON + override table), plan search, console regression cases#111dayfine wants to merge 1 commit into
dayfine wants to merge 1 commit into
Conversation
…rch, regression cases - League plan JSON (per-foe move ranking, setup goals tied to the move they serve, PP/item reserves, between-fight targets) and the override table (planned.py). - Engine: our HM users on the bench, so Roar/Whirlwind succeed and reset stages. - Trainer items follow pret ShouldUseItem's 4-slot gate. - Lv54 stats from the console's summary read. - Regression cases for the 2026-11-16 console failures, kill order and the Max Elixir reserve; Champion setup analysis; coordinate plan search.
This was referenced Nov 17, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Depends on #110
The greedy per-turn League policy made the mistakes seen on console on 2026-11-16: it spent Blizzard PP before Dragonite, set up for moves with no PP left, used Strength finishers that ran into the trainer's Full Restore, chained Super Potions, and read the wrong threat rows. This adds a plan-as-data policy: a League plan JSON (per-foe move ranking, setup goals tied to the move they serve, PP and item reserves, between-fight targets) plus a small override table (sure KO, KO-risk heal, cures, trainer-heal-aware kill order, PP restore). It also adds a coordinate plan search with Monte Carlo over the whole chain and a Champion setup analysis. The engine now puts HM users on our bench, so Roar and Whirlwind succeed, and trainer items follow pret's 4-slot gate. Schema, rows and the C++ port list are in docs/battles/frlg-league-plan.md. Tested with the FRLG unit tests (new test_planned regression cases for each console failure, the old policy pinned red where it made the mistake) and the engine node tests.