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

1 files changed, 3 insertions, 3 deletions

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 <file.ptn>, 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);