Commit Graph

19 Commits

Author SHA1 Message Date
meelstorm
caf0376323 Add Plan 3: ASP.NET backend plan 2026-07-17 18:23:16 +00:00
meelstorm
bbd2c36091 feat(training): export SB3 PPO policy to ONNX with parity test 2026-07-17 18:23:16 +00:00
meelstorm
7b713ba361 feat(training): PPO training script with vectorized LanderCliEnv 2026-07-17 18:23:16 +00:00
meelstorm
edf815de4f feat(training): implement LanderCliEnv wrapping the GameCli subprocess 2026-07-17 18:23:16 +00:00
meelstorm
ba24c752c5 feat(training): scaffold Python training project 2026-07-17 18:23:16 +00:00
meelstorm
30ac7a8114 Add Plan 2: Python PPO training pipeline plan 2026-07-17 18:23:16 +00:00
meelstorm
d1290442ed fix(gamecli): wrap angle in reward to prevent unbounded spin penalty 2026-07-17 18:23:16 +00:00
meelstorm
238ff65db4 chore: gitignore publish/ directory 2026-07-17 18:23:16 +00:00
meelstorm
0b22d67316 test(gamecli): end-to-end smoke tests exercising the built binary 2026-07-17 18:23:16 +00:00
meelstorm
6340218f88 feat(gamecli): wire stdio read-line loop to physics + reward + observation 2026-07-17 18:23:16 +00:00
meelstorm
f5ee2049da feat(gamecli): add JSON protocol DTOs with source-gen context 2026-07-17 18:23:16 +00:00
meelstorm
5b35d65bd2 feat(gamecli): implement Observation.Build with unit tests 2026-07-17 18:23:16 +00:00
meelstorm
3bca85bcc4 feat(gamecli): implement pure Reward.Compute with unit tests 2026-07-17 18:23:16 +00:00
meelstorm
34fb901fe6 feat(gamecli): implement pure Physics.Step with unit tests 2026-07-17 18:23:16 +00:00
meelstorm
f31f01d321 feat(gamecli): add ShipState record struct 2026-07-17 18:23:16 +00:00
meelstorm
83c6b0f41f feat(gamecli): scaffold .NET solution and test project 2026-07-17 18:23:16 +00:00
meelstorm
33b9b8e77c Add Plan 1: Game CLI implementation plan 2026-07-17 18:23:16 +00:00
meelstorm
7d15323cb9 Add design spec for cursor-following lunar lander 2026-07-17 18:23:16 +00:00
1e1ca8bf88 Initial commit 2026-07-17 15:34:39 +00:00