From 241156da6f35cb888253e6b8814e7b61a76b4fc7 Mon Sep 17 00:00:00 2001 From: tslil Date: Sun, 30 Aug 2026 19:46:07 +0100 Subject: TD(lambda), and a big rework NOTE: 6x6 weights are as-yet untrained On the performance side: * switch from linked-list chaining hash table to direct mapping * cache check_win by hash * persist transposition tables across plies (this was a big deal and difficult to puzzle out) * add align directives to the network floats * prepare_nn_input gained a small LUT to avoid branching (i suppose you might call this 1D embedding) * apply mtune=native for the native builds On the correctness side: * Zobrist has now includes the ply parity * the network now predicts something in [-1, 1] using tanh (approximated by polynomials) * the training object is now MSE because we're learning a value estimate * use .clangd instead of compile_commands.json On the training side: * completely forgo human-derived data set * implement td(lambda), self-play, tournaments --- src/ctlm.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'src/ctlm.c') diff --git a/src/ctlm.c b/src/ctlm.c index dc4a17e..c9844c1 100644 --- a/src/ctlm.c +++ b/src/ctlm.c @@ -401,11 +401,11 @@ help, info, load , log, new, play (b|w), self-play, square\ } else if (!strcmp(line, "info")) { print_info(); } else if (!strcmp(line, "eval")) { - float eval = nn1986_evaluate_black_win(state) * 100; + float eval = (state->board_size == 5 ? five_nn_fn(state) : six_nn_fn(state))*100; if (state->ply & 1) { - printf("Black heuristic chance: %s%.2f%s\n", blk, eval, rst); + printf("Black heuristic: %s%.2f%s\n", blk, eval, rst); } else { - printf("White heruistic chance: %s%.2f%s\n", wht, -eval, rst); + printf("White heruistic: %s%.2f%s\n", wht, -eval, rst); } } else if (!strcmp(line, "log")) { puts(gamelog); -- cgit v1.3.1