uct/uct.c

   1 #include <assert.h>
   2 #include <pthread.h>
   3 #include <signal.h>
   4 #include <stdio.h>
   5 #include <stdlib.h>
   6 #include <string.h>
   7
   8 #define DEBUG
   9
  10 #include "debug.h"
  11 #include "board.h"
  12 #include "gtp.h"
  13 #include "move.h"
  14 #include "mq.h"
  15 #include "playout.h"
  16 #include "playout/elo.h"
  17 #include "playout/moggy.h"
  18 #include "playout/light.h"
  19 #include "random.h"
  20 #include "tactics.h"
  21 #include "uct/internal.h"
  22 #include "uct/prior.h"
  23 #include "uct/tree.h"
  24 #include "uct/uct.h"
  25 #include "uct/walk.h"
  26
  27 struct uct_policy *policy_ucb1_init(struct uct *u, char *arg);
  28 struct uct_policy *policy_ucb1amaf_init(struct uct *u, char *arg);
  29
  30
  31 #define MC_GAMES        80000
  32 #define MC_GAMELEN      MAX_GAMELEN
  33
  34 /* How big proportion of ownermap counts must be of one color to consider
  35  * the point sure. */
  36 #define GJ_THRES        0.8
  37 /* How many games to consider at minimum before judging groups. */
  38 #define GJ_MINGAMES     500
  39
  40
  41 static void
  42 setup_state(struct uct *u, struct board *b, enum stone color)
  43 {
  44         u->t = tree_init(b, color);
  45         if (u->force_seed)
  46                 fast_srandom(u->force_seed);
  47         if (UDEBUGL(0))
  48                 fprintf(stderr, "Fresh board with random seed %lu\n", fast_getseed());
  49         //board_print(b, stderr);
  50         if (!u->no_book && b->moves == 0) {
  51                 assert(color == S_BLACK);
  52                 tree_load(u->t, b);
  53         }
  54 }
  55
  56 static void
  57 reset_state(struct uct *u)
  58 {
  59         assert(u->t);
  60         tree_done(u->t); u->t = NULL;
  61 }
  62
  63 static void
  64 prepare_move(struct engine *e, struct board *b, enum stone color)
  65 {
  66         struct uct *u = e->data;
  67
  68         if (u->t) {
  69                 /* Verify that we have sane state. */
  70                 assert(b->es == u);
  71                 assert(u->t && b->moves);
  72                 if (color != stone_other(u->t->root_color)) {
  73                         fprintf(stderr, "Fatal: Non-alternating play detected %d %d\n",
  74                                 color, u->t->root_color);
  75                         exit(1);
  76                 }
  77
  78         } else {
  79                 /* We need fresh state. */
  80                 b->es = u;
  81                 setup_state(u, b, color);
  82         }
  83
  84         if (u->dynkomi && u->dynkomi > b->moves && (color & u->dynkomi_mask))
  85                 u->t->extra_komi = uct_get_extra_komi(u, b);
  86
  87         u->ownermap.playouts = 0;
  88         memset(u->ownermap.map, 0, board_size2(b) * sizeof(u->ownermap.map[0]));
  89 }
  90
  91 static void
  92 dead_group_list(struct uct *u, struct board *b, struct move_queue *mq)
  93 {
  94         struct group_judgement gj;
  95         gj.thres = GJ_THRES;
  96         gj.gs = alloca(board_size2(b) * sizeof(gj.gs[0]));
  97         board_ownermap_judge_group(b, &u->ownermap, &gj);
  98         groups_of_status(b, &gj, GS_DEAD, mq);
  99 }
 100
 101 bool
 102 uct_pass_is_safe(struct uct *u, struct board *b, enum stone color, bool pass_all_alive)
 103 {
 104         if (u->ownermap.playouts < GJ_MINGAMES)
 105                 return false;
 106
 107         struct move_queue mq = { .moves = 0 };
 108         if (!pass_all_alive)
 109                 dead_group_list(u, b, &mq);
 110         return pass_is_safe(b, color, &mq);
 111 }
 112
 113
 114 static void
 115 uct_printhook_ownermap(struct board *board, coord_t c, FILE *f)
 116 {
 117         struct uct *u = board->es;
 118         assert(u);
 119         const char chr[] = ":XO,"; // dame, black, white, unclear
 120         const char chm[] = ":xo,";
 121         char ch = chr[board_ownermap_judge_point(&u->ownermap, c, GJ_THRES)];
 122         if (ch == ',') { // less precise estimate then?
 123                 ch = chm[board_ownermap_judge_point(&u->ownermap, c, 0.67)];
 124         }
 125         fprintf(f, "%c ", ch);
 126 }
 127
 128 static char *
 129 uct_notify_play(struct engine *e, struct board *b, struct move *m)
 130 {
 131         struct uct *u = e->data;
 132         if (!u->t) {
 133                 /* No state, create one - this is probably game beginning
 134                  * and we need to load the opening book right now. */
 135                 prepare_move(e, b, m->color);
 136                 assert(u->t);
 137         }
 138
 139         if (is_resign(m->coord)) {
 140                 /* Reset state. */
 141                 reset_state(u);
 142                 return NULL;
 143         }
 144
 145         /* Promote node of the appropriate move to the tree root. */
 146         assert(u->t->root);
 147         if (!tree_promote_at(u->t, b, m->coord)) {
 148                 if (UDEBUGL(0))
 149                         fprintf(stderr, "Warning: Cannot promote move node! Several play commands in row?\n");
 150                 reset_state(u);
 151                 return NULL;
 152         }
 153
 154         return NULL;
 155 }
 156
 157 static char *
 158 uct_chat(struct engine *e, struct board *b, char *cmd)
 159 {
 160         struct uct *u = e->data;
 161         static char reply[1024];
 162
 163         cmd += strspn(cmd, " \n\t");
 164         if (!strncasecmp(cmd, "winrate", 7)) {
 165                 if (!u->t)
 166                         return "no game context (yet?)";
 167                 enum stone color = u->t->root_color;
 168                 struct tree_node *n = u->t->root;
 169                 snprintf(reply, 1024, "In %d*%d playouts, %s %s can win with %.2f%% probability",
 170                          n->u.playouts, u->threads, stone2str(color), coord2sstr(n->coord, b),
 171                          tree_node_get_value(u->t, -1, n->u.value) * 100);
 172                 if (abs(u->t->extra_komi) >= 0.5) {
 173                         sprintf(reply + strlen(reply), ", while self-imposing extra komi %.1f",
 174                                 u->t->extra_komi);
 175                 }
 176                 strcat(reply, ".");
 177                 return reply;
 178         }
 179         return NULL;
 180 }
 181
 182 static void
 183 uct_dead_group_list(struct engine *e, struct board *b, struct move_queue *mq)
 184 {
 185         struct uct *u = e->data;
 186         if (u->pass_all_alive)
 187                 return; // no dead groups
 188
 189         bool mock_state = false;
 190
 191         if (!u->t) {
 192                 /* No state, but we cannot just back out - we might
 193                  * have passed earlier, only assuming some stones are
 194                  * dead, and then re-connected, only to lose counting
 195                  * when all stones are assumed alive. */
 196                 /* Mock up some state and seed the ownermap by few
 197                  * simulations. */
 198                 prepare_move(e, b, S_BLACK); assert(u->t);
 199                 for (int i = 0; i < GJ_MINGAMES; i++)
 200                         uct_playout(u, b, S_BLACK, u->t);
 201                 mock_state = true;
 202         }
 203
 204         dead_group_list(u, b, mq);
 205
 206         if (mock_state) {
 207                 /* Clean up the mock state in case we will receive
 208                  * a genmove; we could get a non-alternating-move
 209                  * error from prepare_move() in that case otherwise. */
 210                 reset_state(u);
 211         }
 212 }
 213
 214 static void
 215 playout_policy_done(struct playout_policy *p)
 216 {
 217         if (p->done) p->done(p);
 218         if (p->data) free(p->data);
 219         free(p);
 220 }
 221
 222 static void
 223 uct_done(struct engine *e)
 224 {
 225         /* This is called on engine reset, especially when clear_board
 226          * is received and new game should begin. */
 227         struct uct *u = e->data;
 228         if (u->t) reset_state(u);
 229         free(u->ownermap.map);
 230
 231         free(u->policy);
 232         free(u->random_policy);
 233         playout_policy_done(u->playout);
 234         uct_prior_done(u->prior);
 235 }
 236
 237
 238 /* Pachi threading structure (if uct_playouts_parallel() is used):
 239  *
 240  * main thread
 241  *   |         main(), GTP communication, ...
 242  *   |
 243  * thread_manager
 244  *   |         spawns and manages worker threads
 245  *   |
 246  * worker0
 247  * worker1
 248  * ...
 249  * workerK
 250  *             uct_playouts() loop, doing descend-playout N=games times
 251  */
 252
 253 /* Set in thread manager in case the workers should stop. */
 254 volatile sig_atomic_t uct_halt = 0;
 255 /* ID of the running worker thread. */
 256 __thread int thread_id = -1;
 257 /* ID of the thread manager. */
 258 static pthread_t thread_manager;
 259 static bool thread_manager_running;
 260
 261 static pthread_mutex_t finish_mutex = PTHREAD_MUTEX_INITIALIZER;
 262 static pthread_cond_t finish_cond = PTHREAD_COND_INITIALIZER;
 263 static volatile int finish_thread;
 264 static pthread_mutex_t finish_serializer = PTHREAD_MUTEX_INITIALIZER;
 265
 266 struct spawn_ctx {
 267         int tid;
 268         struct uct *u;
 269         struct board *b;
 270         enum stone color;
 271         struct tree *t;
 272         unsigned long seed;
 273         int games;
 274 };
 275
 276 static void *
 277 spawn_worker(void *ctx_)
 278 {
 279         struct spawn_ctx *ctx = ctx_;
 280         /* Setup */
 281         fast_srandom(ctx->seed);
 282         thread_id = ctx->tid;
 283         /* Run */
 284         ctx->games = uct_playouts(ctx->u, ctx->b, ctx->color, ctx->t, ctx->games);
 285         /* Finish */
 286         pthread_mutex_lock(&finish_serializer);
 287         pthread_mutex_lock(&finish_mutex);
 288         finish_thread = ctx->tid;
 289         pthread_cond_signal(&finish_cond);
 290         pthread_mutex_unlock(&finish_mutex);
 291         return ctx;
 292 }
 293
 294 /* Thread manager, controlling worker threads. It must be called with
 295  * finish_mutex lock held, and the finish_cond can be signalled for it
 296  * to stop; in that case, the caller should set finish_thread = -1. */
 297 static void *
 298 spawn_thread_manager(void *ctx_)
 299 {
 300         /* In thread_manager, we use only some of the ctx fields. */
 301         struct spawn_ctx *mctx = ctx_;
 302         struct uct *u = mctx->u;
 303         bool shared_tree = u->parallel_tree;
 304         fast_srandom(mctx->seed);
 305
 306         int played_games = 0;
 307         pthread_t threads[u->threads];
 308         int joined = 0;
 309
 310         uct_halt = 0;
 311         /* Spawn threads... */
 312         for (int ti = 0; ti < u->threads; ti++) {
 313                 struct spawn_ctx *ctx = malloc(sizeof(*ctx));
 314                 ctx->u = u; ctx->b = mctx->b; ctx->color = mctx->color;
 315                 ctx->t = shared_tree ? mctx->t : tree_copy(mctx->t);
 316                 ctx->tid = ti; ctx->games = mctx->games;
 317                 ctx->seed = fast_random(65536) + ti;
 318                 pthread_create(&threads[ti], NULL, spawn_worker, ctx);
 319                 if (UDEBUGL(2))
 320                         fprintf(stderr, "Spawned worker %d\n", ti);
 321         }
 322
 323         /* ...and collect them back: */
 324         while (joined < u->threads) {
 325                 /* Wait for some thread to finish... */
 326                 pthread_cond_wait(&finish_cond, &finish_mutex);
 327                 if (finish_thread < 0) {
 328                         /* Stop-by-caller. Tell the workers to wrap up. */
 329                         uct_halt = 1;
 330                         continue;
 331                 }
 332                 /* ...and gather its remnants. */
 333                 struct spawn_ctx *ctx;
 334                 pthread_join(threads[finish_thread], (void **) &ctx);
 335                 played_games += ctx->games;
 336                 joined++;
 337                 if (!shared_tree) {
 338                         tree_merge(mctx->t, ctx->t);
 339                         tree_done(ctx->t);
 340                 }
 341                 free(ctx);
 342                 if (UDEBUGL(2))
 343                         fprintf(stderr, "Joined worker %d\n", finish_thread);
 344                 /* Do not get stalled by slow threads. */
 345                 if (joined >= u->threads / 2)
 346                         uct_halt = 1;
 347                 pthread_mutex_unlock(&finish_serializer);
 348         }
 349
 350         if (!shared_tree)
 351                 tree_normalize(mctx->t, u->threads);
 352
 353         mctx->games = played_games;
 354         return mctx;
 355 }
 356
 357 static void
 358 uct_pondering_start(struct uct *u, struct board *b0, enum stone color, struct tree *t, int games)
 359 {
 360         assert(u->threads > 0);
 361         assert(!thread_manager_running);
 362
 363         /* *b0 can change in the meantime. */
 364         struct board b; board_copy(&b, b0);
 365
 366         struct spawn_ctx ctx = { .u = u, .b = &b, .color = color, .t = t, .games = games, .seed = fast_random(65536) };
 367         static struct spawn_ctx mctx; mctx = ctx;
 368         pthread_mutex_lock(&finish_mutex);
 369         pthread_create(&thread_manager, NULL, spawn_thread_manager, &mctx);
 370         thread_manager_running = true;
 371 }
 372
 373 static int
 374 uct_pondering_stop(void)
 375 {
 376         assert(thread_manager_running);
 377
 378         struct spawn_ctx *pctx;
 379         thread_manager_running = false;
 380         pthread_join(thread_manager, (void **) &pctx);
 381         pthread_mutex_unlock(&finish_mutex);
 382         return pctx->games;
 383 }
 384
 385 static int
 386 uct_playouts_threaded(struct uct *u, struct board *b, enum stone color, struct tree *t, int games)
 387 {
 388         uct_pondering_start(u, b, color, t, games);
 389         /* We just wait until the thread manager finishes. */
 390         return uct_pondering_stop();
 391 }
 392
 393
 394 static coord_t *
 395 uct_genmove(struct engine *e, struct board *b, enum stone color, bool pass_all_alive)
 396 {
 397         struct uct *u = e->data;
 398
 399         if (b->superko_violation) {
 400                 fprintf(stderr, "!!! WARNING: SUPERKO VIOLATION OCCURED BEFORE THIS MOVE\n");
 401                 fprintf(stderr, "Maybe you play with situational instead of positional superko?\n");
 402                 fprintf(stderr, "I'm going to ignore the violation, but note that I may miss\n");
 403                 fprintf(stderr, "some moves valid under this ruleset because of this.\n");
 404                 b->superko_violation = false;
 405         }
 406
 407         /* Seed the tree. */
 408         prepare_move(e, b, color);
 409         assert(u->t);
 410
 411         /* Determine number of simulations. */
 412         int games = u->games;
 413         if (u->t->root->children) {
 414                 int delta = u->t->root->u.playouts * 2 / 3;
 415                 if (u->parallel_tree) delta /= u->threads;
 416                 games -= delta;
 417         }
 418         /* else this is highly read-out but dead-end branch of opening book;
 419          * we need to start from scratch; XXX: Maybe actually base the readout
 420          * count based on number of playouts of best node? */
 421         if (games < u->games && UDEBUGL(2))
 422                 fprintf(stderr, "<pre-simulated %d games skipped>\n", u->games - games);
 423
 424         /* Perform the Monte Carlo Tree Search! */
 425         int played_games = uct_playouts_threaded(u, b, color, u->t, games);
 426
 427         if (UDEBUGL(2))
 428                 tree_dump(u->t, u->dumpthres);
 429
 430         /* Choose the best move from the tree. */
 431         struct tree_node *best = u->policy->choose(u->policy, u->t->root, b, color);
 432         if (!best) {
 433                 reset_state(u);
 434                 return coord_copy(pass);
 435         }
 436         if (UDEBUGL(0)) {
 437                 uct_progress_status(u, u->t, color, played_games);
 438         }
 439         if (UDEBUGL(1))
 440                 fprintf(stderr, "*** WINNER is %s (%d,%d) with score %1.4f (%d/%d:%d games)\n",
 441                         coord2sstr(best->coord, b), coord_x(best->coord, b), coord_y(best->coord, b),
 442                         tree_node_get_value(u->t, 1, best->u.value),
 443                         best->u.playouts, u->t->root->u.playouts, played_games);
 444         if (tree_node_get_value(u->t, 1, best->u.value) < u->resign_ratio && !is_pass(best->coord)) {
 445                 reset_state(u);
 446                 return coord_copy(resign);
 447         }
 448
 449         /* If the opponent just passed and we win counting, always
 450          * pass as well. */
 451         if (b->moves > 1 && is_pass(b->last_move.coord)) {
 452                 /* Make sure enough playouts are simulated. */
 453                 while (u->ownermap.playouts < GJ_MINGAMES)
 454                         uct_playout(u, b, color, u->t);
 455                 if (uct_pass_is_safe(u, b, color, u->pass_all_alive || pass_all_alive)) {
 456                         if (UDEBUGL(0))
 457                                 fprintf(stderr, "<Will rather pass, looks safe enough.>\n");
 458                         best->coord = pass;
 459                 }
 460         }
 461
 462         tree_promote_node(u->t, best);
 463         return coord_copy(best->coord);
 464 }
 465
 466
 467 bool
 468 uct_genbook(struct engine *e, struct board *b, enum stone color)
 469 {
 470         struct uct *u = e->data;
 471         if (!u->t) prepare_move(e, b, color);
 472         assert(u->t);
 473
 474         uct_playouts_threaded(u, b, color, u->t, u->games);
 475
 476         tree_save(u->t, b, u->games / 100);
 477
 478         return true;
 479 }
 480
 481 void
 482 uct_dumpbook(struct engine *e, struct board *b, enum stone color)
 483 {
 484         struct tree *t = tree_init(b, color);
 485         tree_load(t, b);
 486         tree_dump(t, 0);
 487         tree_done(t);
 488 }
 489
 490
 491 struct uct *
 492 uct_state_init(char *arg, struct board *b)
 493 {
 494         struct uct *u = calloc(1, sizeof(struct uct));
 495
 496         u->debug_level = 1;
 497         u->games = MC_GAMES;
 498         u->gamelen = MC_GAMELEN;
 499         u->mercymin = 0;
 500         u->expand_p = 2;
 501         u->dumpthres = 1000;
 502         u->playout_amaf = true;
 503         u->playout_amaf_nakade = false;
 504         u->amaf_prior = false;
 505         u->max_tree_size = 3072ULL * 1048576;
 506
 507         if (board_size(b) - 2 >= 19)
 508                 u->dynkomi = 200;
 509         u->dynkomi_mask = S_BLACK;
 510
 511         u->threads = 1;
 512         u->thread_model = TM_TREEVL;
 513         u->parallel_tree = true;
 514         u->virtual_loss = true;
 515
 516         u->val_scale = 0.02; u->val_points = 20;
 517
 518         if (arg) {
 519                 char *optspec, *next = arg;
 520                 while (*next) {
 521                         optspec = next;
 522                         next += strcspn(next, ",");
 523                         if (*next) { *next++ = 0; } else { *next = 0; }
 524
 525                         char *optname = optspec;
 526                         char *optval = strchr(optspec, '=');
 527                         if (optval) *optval++ = 0;
 528
 529                         if (!strcasecmp(optname, "debug")) {
 530                                 if (optval)
 531                                         u->debug_level = atoi(optval);
 532                                 else
 533                                         u->debug_level++;
 534                         } else if (!strcasecmp(optname, "games") && optval) {
 535                                 u->games = atoi(optval);
 536                         } else if (!strcasecmp(optname, "mercy") && optval) {
 537                                 /* Minimal difference of black/white captures
 538                                  * to stop playout - "Mercy Rule". Speeds up
 539                                  * hopeless playouts at the expense of some
 540                                  * accuracy. */
 541                                 u->mercymin = atoi(optval);
 542                         } else if (!strcasecmp(optname, "gamelen") && optval) {
 543                                 u->gamelen = atoi(optval);
 544                         } else if (!strcasecmp(optname, "expand_p") && optval) {
 545                                 u->expand_p = atoi(optval);
 546                         } else if (!strcasecmp(optname, "dumpthres") && optval) {
 547                                 u->dumpthres = atoi(optval);
 548                         } else if (!strcasecmp(optname, "playout_amaf")) {
 549                                 /* Whether to include random playout moves in
 550                                  * AMAF as well. (Otherwise, only tree moves
 551                                  * are included in AMAF. Of course makes sense
 552                                  * only in connection with an AMAF policy.) */
 553                                 /* with-without: 55.5% (+-4.1) */
 554                                 if (optval && *optval == '0')
 555                                         u->playout_amaf = false;
 556                                 else
 557                                         u->playout_amaf = true;
 558                         } else if (!strcasecmp(optname, "playout_amaf_nakade")) {
 559                                 /* Whether to include nakade moves from playouts
 560                                  * in the AMAF statistics; this tends to nullify
 561                                  * the playout_amaf effect by adding too much
 562                                  * noise. */
 563                                 if (optval && *optval == '0')
 564                                         u->playout_amaf_nakade = false;
 565                                 else
 566                                         u->playout_amaf_nakade = true;
 567                         } else if (!strcasecmp(optname, "playout_amaf_cutoff") && optval) {
 568                                 /* Keep only first N% of playout stage AMAF
 569                                  * information. */
 570                                 u->playout_amaf_cutoff = atoi(optval);
 571                         } else if ((!strcasecmp(optname, "policy") || !strcasecmp(optname, "random_policy")) && optval) {
 572                                 char *policyarg = strchr(optval, ':');
 573                                 struct uct_policy **p = !strcasecmp(optname, "policy") ? &u->policy : &u->random_policy;
 574                                 if (policyarg)
 575                                         *policyarg++ = 0;
 576                                 if (!strcasecmp(optval, "ucb1")) {
 577                                         *p = policy_ucb1_init(u, policyarg);
 578                                 } else if (!strcasecmp(optval, "ucb1amaf")) {
 579                                         *p = policy_ucb1amaf_init(u, policyarg);
 580                                 } else {
 581                                         fprintf(stderr, "UCT: Invalid tree policy %s\n", optval);
 582                                         exit(1);
 583                                 }
 584                         } else if (!strcasecmp(optname, "playout") && optval) {
 585                                 char *playoutarg = strchr(optval, ':');
 586                                 if (playoutarg)
 587                                         *playoutarg++ = 0;
 588                                 if (!strcasecmp(optval, "moggy")) {
 589                                         u->playout = playout_moggy_init(playoutarg);
 590                                 } else if (!strcasecmp(optval, "light")) {
 591                                         u->playout = playout_light_init(playoutarg);
 592                                 } else if (!strcasecmp(optval, "elo")) {
 593                                         u->playout = playout_elo_init(playoutarg);
 594                                 } else {
 595                                         fprintf(stderr, "UCT: Invalid playout policy %s\n", optval);
 596                                         exit(1);
 597                                 }
 598                         } else if (!strcasecmp(optname, "prior") && optval) {
 599                                 u->prior = uct_prior_init(optval, b);
 600                         } else if (!strcasecmp(optname, "amaf_prior") && optval) {
 601                                 u->amaf_prior = atoi(optval);
 602                         } else if (!strcasecmp(optname, "threads") && optval) {
 603                                 /* By default, Pachi will run with only single
 604                                  * tree search thread! */
 605                                 u->threads = atoi(optval);
 606                         } else if (!strcasecmp(optname, "thread_model") && optval) {
 607                                 if (!strcasecmp(optval, "root")) {
 608                                         /* Root parallelization - each thread
 609                                          * does independent search, trees are
 610                                          * merged at the end. */
 611                                         u->thread_model = TM_ROOT;
 612                                         u->parallel_tree = false;
 613                                         u->virtual_loss = false;
 614                                 } else if (!strcasecmp(optval, "tree")) {
 615                                         /* Tree parallelization - all threads
 616                                          * grind on the same tree. */
 617                                         u->thread_model = TM_TREE;
 618                                         u->parallel_tree = true;
 619                                         u->virtual_loss = false;
 620                                 } else if (!strcasecmp(optval, "treevl")) {
 621                                         /* Tree parallelization, but also
 622                                          * with virtual losses - this discou-
 623                                          * rages most threads choosing the
 624                                          * same tree branches to read. */
 625                                         u->thread_model = TM_TREEVL;
 626                                         u->parallel_tree = true;
 627                                         u->virtual_loss = true;
 628                                 } else {
 629                                         fprintf(stderr, "UCT: Invalid thread model %s\n", optval);
 630                                         exit(1);
 631                                 }
 632                         } else if (!strcasecmp(optname, "force_seed") && optval) {
 633                                 u->force_seed = atoi(optval);
 634                         } else if (!strcasecmp(optname, "no_book")) {
 635                                 u->no_book = true;
 636                         } else if (!strcasecmp(optname, "dynkomi")) {
 637                                 /* Dynamic komi in handicap game; linearly
 638                                  * decreases to basic settings until move
 639                                  * #optval. */
 640                                 u->dynkomi = optval ? atoi(optval) : 150;
 641                         } else if (!strcasecmp(optname, "dynkomi_mask") && optval) {
 642                                 /* Bitmask of colors the player must be
 643                                  * for dynkomi be applied; you may want
 644                                  * to use dynkomi_mask=3 to allow dynkomi
 645                                  * even in games where Pachi is white. */
 646                                 u->dynkomi_mask = atoi(optval);
 647                         } else if (!strcasecmp(optname, "val_scale") && optval) {
 648                                 /* How much of the game result value should be
 649                                  * influenced by win size. Zero means it isn't. */
 650                                 u->val_scale = atof(optval);
 651                         } else if (!strcasecmp(optname, "val_points") && optval) {
 652                                 /* Maximum size of win to be scaled into game
 653                                  * result value. Zero means boardsize^2. */
 654                                 u->val_points = atoi(optval) * 2; // result values are doubled
 655                         } else if (!strcasecmp(optname, "val_extra")) {
 656                                 /* If false, the score coefficient will be simply
 657                                  * added to the value, instead of scaling the result
 658                                  * coefficient because of it. */
 659                                 u->val_extra = !optval || atoi(optval);
 660                         } else if (!strcasecmp(optname, "root_heuristic") && optval) {
 661                                 /* Whether to bias exploration by root node values
 662                                  * (must be supported by the used policy).
 663                                  * 0: Don't.
 664                                  * 1: Do, value = result.
 665                                  * Try to temper the result:
 666                                  * 2: Do, value = 0.5+(result-expected)/2.
 667                                  * 3: Do, value = 0.5+bzz((result-expected)^2). */
 668                                 u->root_heuristic = atoi(optval);
 669                         } else if (!strcasecmp(optname, "pass_all_alive")) {
 670                                 /* Whether to consider all stones alive at the game
 671                                  * end instead of marking dead groupd. */
 672                                 u->pass_all_alive = !optval || atoi(optval);
 673                         } else if (!strcasecmp(optname, "random_policy_chance") && optval) {
 674                                 /* If specified (N), with probability 1/N, random_policy policy
 675                                  * descend is used instead of main policy descend; useful
 676                                  * if specified policy (e.g. UCB1AMAF) can make unduly biased
 677                                  * choices sometimes, you can fall back to e.g.
 678                                  * random_policy=UCB1. */
 679                                 u->random_policy_chance = atoi(optval);
 680                         } else if (!strcasecmp(optname, "max_tree_size") && optval) {
 681                                 /* Maximum amount of memory [MiB] consumed by the move tree.
 682                                  * Default is 3072 (3 GiB). Note that if you use TM_ROOT,
 683                                  * this limits size of only one of the trees, not all of them
 684                                  * together. */
 685                                 u->max_tree_size = atol(optval) * 1048576;
 686                         } else if (!strcasecmp(optname, "banner") && optval) {
 687                                 /* Additional banner string. This must come as the
 688                                  * last engine parameter. */
 689                                 if (*next) *--next = ',';
 690                                 u->banner = strdup(optval);
 691                                 break;
 692                         } else {
 693                                 fprintf(stderr, "uct: Invalid engine argument %s or missing value\n", optname);
 694                                 exit(1);
 695                         }
 696                 }
 697         }
 698
 699         u->resign_ratio = 0.2; /* Resign when most games are lost. */
 700         u->loss_threshold = 0.85; /* Stop reading if after at least 5000 playouts this is best value. */
 701         if (!u->policy)
 702                 u->policy = policy_ucb1amaf_init(u, NULL);
 703
 704         if (!!u->random_policy_chance ^ !!u->random_policy) {
 705                 fprintf(stderr, "uct: Only one of random_policy and random_policy_chance is set\n");
 706                 exit(1);
 707         }
 708
 709         if (!u->prior)
 710                 u->prior = uct_prior_init(NULL, b);
 711
 712         if (!u->playout)
 713                 u->playout = playout_moggy_init(NULL);
 714         u->playout->debug_level = u->debug_level;
 715
 716         u->ownermap.map = malloc(board_size2(b) * sizeof(u->ownermap.map[0]));
 717
 718         /* Some things remain uninitialized for now - the opening book
 719          * is not loaded and the tree not set up. */
 720         /* This will be initialized in setup_state() at the first move
 721          * received/requested. This is because right now we are not aware
 722          * about any komi or handicap setup and such. */
 723
 724         return u;
 725 }
 726
 727 struct engine *
 728 engine_uct_init(char *arg, struct board *b)
 729 {
 730         struct uct *u = uct_state_init(arg, b);
 731         struct engine *e = calloc(1, sizeof(struct engine));
 732         e->name = "UCT Engine";
 733         e->printhook = uct_printhook_ownermap;
 734         e->notify_play = uct_notify_play;
 735         e->chat = uct_chat;
 736         e->genmove = uct_genmove;
 737         e->dead_group_list = uct_dead_group_list;
 738         e->done = uct_done;
 739         e->data = u;
 740
 741         const char banner[] = "I'm playing UCT. When I'm losing, I will resign, "
 742                 "if I think I win, I play until you pass. "
 743                 "Anyone can send me 'winrate' in private chat to get my assessment of the position.";
 744         if (!u->banner) u->banner = "";
 745         e->comment = malloc(sizeof(banner) + strlen(u->banner) + 1);
 746         sprintf(e->comment, "%s %s", banner, u->banner);
 747
 748         return e;
 749 }