boss/tigers-and-sheep.git/.gitignore, branch v0.1.0Unnamed repository; edit this file 'description' to name the repository.https://git.tehbox.org/boss/tigers-and-sheep.git/atom/.gitignore?h=v0.1.02026-11-05T00:46:28Zfeat: Neural network opponent trained by self-play2026-11-05T00:46:28ZDylanboss@tehbox.org2026-11-05T00:46:28Zurn:sha1:94f3df190a91898eb344b9214a57eb31792bcb1c
- NeuralPlayer: a MinimaxPlayer that scores positions with a small MLP
(network.rs, hand-written backprop and Adam, no new dependencies)
- train subcommand: self-play with TD(lambda) targets, cut at random
exploration moves so sheep blunders don't bias the net towards tigers
- neural subcommand: play against a trained network
- evaluate now gets the side to move
- Board symmetries; the search cache keys by the canonical board, which
halves the depth-9 bench time with identical moves and scores
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
feat: Initial commit2026-10-18T11:38:55ZDylanboss@tehbox.org2026-10-18T11:37:17Zurn:sha1:15bdaa5d550115d97be3f89126c924b49c10a70f
What works:
- Minimax search
- With a simple player claude designed
- Manual player
- Move generation
- Board display
What needs improvement:
- Manual player
- Game coordinator
- Draw checking
- Nix setup