boss/tigers-and-sheep.git/src, branch master

refactor: Drop the neural command in favour of play's -S/-T weights files

2026-11-09T10:00:58Z

-S and -T already accept a weights file, so the neural subcommand was only a shorthand for one of them. Update the README and CLAUDE.md to show the play form instead. Co-Authored-By: Claude Opus 5.5

feat: Colour the game result and show scores to 5 significant figures

2026-11-09T03:56:33Z

The game-over line is bold green for a win, naming the winning player, and bold yellow for a draw. The last-turn panel shows a minimax player's score to 5 significant figures. Co-Authored-By: Claude Opus 5.5

feat: Multi-threaded training; play matches every 10 generations

2026-11-09T03:38:13Z

Gradients and losses are computed on rayon's pool in fixed-size chunks summed in order, so the weights are the same on any number of threads. Self-play uses the same pool, and the output shows the training time. The matches against the earlier network and ClaudePlayer now run only every --match-every generations (default 10), against the generation that last played them, read back from its saved file. Co-Authored-By: Claude Opus 5.5

feat: Draw by threefold repetition

2026-11-09T03:17:04Z

A position coming up for the third time with the same side to move ends the game as a draw, in the TUI and in training self-play. Players now get the game's earlier boards in pick_move, and the search scores the first repetition on its path as a draw, so a side that is ahead avoids one and a side that is behind seeks one. Co-Authored-By: Claude Opus 5.5

feat: Parse the command line with clap

2026-11-09T03:02:43Z

Players are now chosen with --sheep/-S and --tiger/-T (human, claude or a weights file) and --time/-t, instead of by editing main.rs. Every training setting is a train option, and --help covers all commands. Co-Authored-By: Claude Opus 5.5

feat: Resume training from any weights; balance training by outcome

2026-11-08T04:32:09Z

- train --from starts from any saved network; without it, training still carries on from the out dir's newest generation - Each epoch takes as many positions from sheep-win games as from tiger-win games (Config::balance_outcomes, on by default) - Training output shows the error on tiger-win and sheep-win positions separately, before and after training - Fix a broken sentence in CLAUDE.md's cache description Co-Authored-By: Claude Opus 5.5

feat: Neural network opponent trained by self-play

2026-11-05T00:46:28Z

- 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

feat: Optimisation

2026-10-23T06:41:07Z

Added benchmarking capabilities

feat: More improvements to UI

2026-10-23T00:17:08Z

feat: Improvements to UI

2026-10-22T04:17:02Z

Now uses ratatui