uct/internal.h

   1 #ifndef ZZGO_UCT_INTERNAL_H
   2 #define ZZGO_UCT_INTERNAL_H
   3
   4 #include <signal.h> // sig_atomic_t
   5
   6 #include "debug.h"
   7 #include "move.h"
   8 #include "ownermap.h"
   9 #include "playout.h"
  10 #include "stats.h"
  11
  12 struct tree;
  13 struct tree_node;
  14 struct uct_policy;
  15 struct uct_prior;
  16 struct uct_dynkomi;
  17
  18 /* Internal UCT structures */
  19
  20
  21 /* Internal engine state. */
  22 struct uct {
  23         int debug_level;
  24         int games, gamelen;
  25         float resign_ratio;
  26         float loss_threshold;
  27         double best2_ratio, bestr_ratio;
  28         bool pass_all_alive;
  29         bool territory_scoring;
  30         int expand_p;
  31         bool playout_amaf, playout_amaf_nakade;
  32         bool amaf_prior;
  33         int playout_amaf_cutoff;
  34         int dumpthres;
  35         int force_seed;
  36         bool no_book;
  37         bool fast_alloc;
  38         unsigned long max_tree_size;
  39         int mercymin;
  40
  41         int threads;
  42         enum uct_thread_model {
  43                 TM_ROOT, /* Root parallelization. */
  44                 TM_TREE, /* Tree parallelization w/o virtual loss. */
  45                 TM_TREEVL, /* Tree parallelization with virtual loss. */
  46         } thread_model;
  47         bool parallel_tree;
  48         bool virtual_loss;
  49         bool pondering_opt; /* User wants pondering */
  50         bool pondering; /* Actually pondering now */
  51
  52         int fuseki_end;
  53         int yose_start;
  54
  55         int dynkomi_mask;
  56         int dynkomi_interval;
  57         struct uct_dynkomi *dynkomi;
  58
  59         float val_scale;
  60         int val_points;
  61         bool val_extra;
  62
  63         int random_policy_chance;
  64         int local_tree;
  65         int tenuki_d;
  66         float local_tree_aging;
  67         bool local_tree_allseq;
  68         /* Playout-localtree integration. */
  69         bool local_tree_playout; // can be true only if ELO playout
  70         bool local_tree_pseqroot;
  71
  72         char *banner;
  73
  74         struct uct_policy *policy;
  75         struct uct_policy *random_policy;
  76         struct playout_policy *playout;
  77         struct uct_prior *prior;
  78
  79         /* Used within frame of single genmove. */
  80         struct board_ownermap ownermap;
  81
  82         /* Game state - maintained by setup_state(), reset_state(). */
  83         struct tree *t;
  84 };
  85
  86 #define UDEBUGL(n) DEBUGL_(u->debug_level, n)
  87
  88 extern volatile sig_atomic_t uct_halt;
  89 extern __thread int thread_id;
  90
  91 bool uct_pass_is_safe(struct uct *u, struct board *b, enum stone color, bool pass_all_alive);
  92
  93
  94 /* This is the state used for descending the tree; we use this wrapper
  95  * structure in order to be able to easily descend in multiple trees
  96  * in parallel (e.g. main tree and local tree) or compute cummulative
  97  * "path value" throughout the tree descent. */
  98 struct uct_descent {
  99         /* Active tree nodes: */
 100         struct tree_node *node; /* Main tree. */
 101         struct tree_node *lnode; /* Local tree. */
 102         /* Value of main tree node (with all value factors, but unbiased
 103          * - without exploration factor), from black's perspective. */
 104         struct move_stats value;
 105 };
 106
 107
 108 typedef struct tree_node *(*uctp_choose)(struct uct_policy *p, struct tree_node *node, struct board *b, enum stone color, coord_t exclude);
 109 typedef float (*uctp_evaluate)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity);
 110 typedef void (*uctp_descend)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity, bool allow_pass);
 111 typedef void (*uctp_winner)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent);
 112 typedef void (*uctp_prior)(struct uct_policy *p, struct tree *tree, struct tree_node *node, struct board *b, enum stone color, int parity);
 113 typedef void (*uctp_update)(struct uct_policy *p, struct tree *tree, struct tree_node *node, enum stone node_color, enum stone player_color, struct playout_amafmap *amaf, float result);
 114
 115 struct uct_policy {
 116         struct uct *uct;
 117         uctp_choose choose;
 118         uctp_winner winner;
 119         uctp_evaluate evaluate;
 120         uctp_descend descend;
 121         uctp_update update;
 122         uctp_prior prior;
 123         bool wants_amaf;
 124         void *data;
 125 };
 126
 127 #endif