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         bool slave; /* Act as slave in distributed engine. */
  52         enum stone my_color;
  53         int gtp_id; /* id of the last gtp command */
  54
  55         int fuseki_end;
  56         int yose_start;
  57
  58         int dynkomi_mask;
  59         int dynkomi_interval;
  60         struct uct_dynkomi *dynkomi;
  61
  62         float val_scale;
  63         int val_points;
  64         bool val_extra;
  65
  66         int random_policy_chance;
  67         int local_tree;
  68         int tenuki_d;
  69         float local_tree_aging;
  70         bool local_tree_allseq;
  71         /* Playout-localtree integration. */
  72         bool local_tree_playout; // can be true only if ELO playout
  73         bool local_tree_pseqroot;
  74
  75         char *banner;
  76
  77         struct uct_policy *policy;
  78         struct uct_policy *random_policy;
  79         struct playout_policy *playout;
  80         struct uct_prior *prior;
  81
  82         /* Used within frame of single genmove. */
  83         struct board_ownermap ownermap;
  84
  85         /* Game state - maintained by setup_state(), reset_state(). */
  86         struct tree *t;
  87 };
  88
  89 #define UDEBUGL(n) DEBUGL_(u->debug_level, n)
  90
  91 extern volatile sig_atomic_t uct_halt;
  92 extern __thread int thread_id;
  93
  94 bool uct_pass_is_safe(struct uct *u, struct board *b, enum stone color, bool pass_all_alive);
  95
  96
  97 /* This is the state used for descending the tree; we use this wrapper
  98  * structure in order to be able to easily descend in multiple trees
  99  * in parallel (e.g. main tree and local tree) or compute cummulative
 100  * "path value" throughout the tree descent. */
 101 struct uct_descent {
 102         /* Active tree nodes: */
 103         struct tree_node *node; /* Main tree. */
 104         struct tree_node *lnode; /* Local tree. */
 105         /* Value of main tree node (with all value factors, but unbiased
 106          * - without exploration factor), from black's perspective. */
 107         struct move_stats value;
 108 };
 109
 110
 111 typedef struct tree_node *(*uctp_choose)(struct uct_policy *p, struct tree_node *node, struct board *b, enum stone color, coord_t exclude);
 112 typedef float (*uctp_evaluate)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity);
 113 typedef void (*uctp_descend)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity, bool allow_pass);
 114 typedef void (*uctp_winner)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent);
 115 typedef void (*uctp_prior)(struct uct_policy *p, struct tree *tree, struct tree_node *node, struct board *b, enum stone color, int parity);
 116 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);
 117
 118 struct uct_policy {
 119         struct uct *uct;
 120         uctp_choose choose;
 121         uctp_winner winner;
 122         uctp_evaluate evaluate;
 123         uctp_descend descend;
 124         uctp_update update;
 125         uctp_prior prior;
 126         bool wants_amaf;
 127         void *data;
 128 };
 129
 130 #endif