tim/bridzik
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
74 Commits 4 Branches 0 Tags
8f2449a408c985132e4dff786f37df559c146a12
Commit Graph
3 Commits
Author SHA1 Message Date
timandClaude Fable 5 8f2449a408 RL: self-play PPO trening siete 
BridzikNet (trup + guess/play/value hlavy s maskovanim), self-play
generator so zdielanou sietou na 4 sedadlach a opponent mixingom
(random/heuristicke sedadla pre robustnost), vlastny clipped-PPO
so skalovanim odmien a lr/entropy annealom. Torch je len trenovacia
zavislost na hoste (requirements-rl.txt); checkpointy a logy su
gitignorovane. Spustenie: py -m rl.train.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
 2026-07-07 18:49:55 +02:00
timandClaude Fable 5 e1733f4943 RL: baseline boti a evaluacny harness 
RandomPlayer, HeuristicPlayer (MC tipper nad rozdaniami neznamych kariet
+ tipom riadena hracia heuristika) a McPlayer (MC ohodnotenie kazdeho
kandidatskeho tahu nad rozdaniami konzistentnymi s dedukovanymi voidmi).
Evaluacia s rotaciou sedadiel: py -m rl.evaluate.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
 2026-07-07 18:49:55 +02:00
timandClaude Fable 5 3710a68e37 RL: encoding observacii, akcne masky a Round prostredie 
Zaklad RL vrstvy (rl/DESIGN.md): egocentricky rotovana observacia
(ruka, videne karty, tipy, kopka, dedukovane voidy -- 233 dim), masky
legalnych tipov/kariet zrkadliace pravidla enginu a RoundEnv (jedno kolo
= jedna self-play epizoda). Fuzz-testy vynucuju zhodu masiek s enginom.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
 2026-07-07 18:49:55 +02:00