uct/internal.h

   1 #ifndef ZZGO_UCT_INTERNAL_H
   2 #define ZZGO_UCT_INTERNAL_H
   3
   4 /* Internal UCT structures */
   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 /* How big proportion of ownermap counts must be of one color to consider
  19  * the point sure. */
  20 #define GJ_THRES        0.8
  21 /* How many games to consider at minimum before judging groups. */
  22 #define GJ_MINGAMES     500
  23
  24 /* Internal engine state. */
  25 struct uct {
  26         int debug_level;
  27         int games, gamelen;
  28         float resign_ratio;
  29         float loss_threshold;
  30         double best2_ratio, bestr_ratio;
  31         bool pass_all_alive;
  32         bool territory_scoring;
  33         int expand_p;
  34         bool playout_amaf, playout_amaf_nakade;
  35         bool amaf_prior;
  36         int playout_amaf_cutoff;
  37         int dumpthres;
  38         int force_seed;
  39         bool no_book;
  40         bool fast_alloc;
  41         unsigned long max_tree_size;
  42         int mercymin;
  43
  44         int threads;
  45         enum uct_thread_model {
  46                 TM_TREE, /* Tree parallelization w/o virtual loss. */
  47                 TM_TREEVL, /* Tree parallelization with virtual loss. */
  48         } thread_model;
  49         bool virtual_loss;
  50         bool pondering_opt; /* User wants pondering */
  51         bool pondering; /* Actually pondering now */
  52         bool slave; /* Act as slave in distributed engine. */
  53         enum stone my_color;
  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         /* Used for coordination among slaves of the distributed engine. */
  85         int stats_hbits;
  86         int shared_nodes;
  87         int shared_levels;
  88         int played_own;
  89         int played_all; /* games played by all slaves */
  90
  91         /* Game state - maintained by setup_state(), reset_state(). */
  92         struct tree *t;
  93 };
  94
  95 #define UDEBUGL(n) DEBUGL_(u->debug_level, n)
  96
  97 bool uct_pass_is_safe(struct uct *u, struct board *b, enum stone color, bool pass_all_alive);
  98
  99 void uct_prepare_move(struct uct *u, struct board *b, enum stone color);
 100 void uct_genmove_setup(struct uct *u, struct board *b, enum stone color);
 101 void uct_pondering_stop(struct uct *u);
 102
 103
 104 /* This is the state used for descending the tree; we use this wrapper
 105  * structure in order to be able to easily descend in multiple trees
 106  * in parallel (e.g. main tree and local tree) or compute cummulative
 107  * "path value" throughout the tree descent. */
 108 struct uct_descent {
 109         /* Active tree nodes: */
 110         struct tree_node *node; /* Main tree. */
 111         struct tree_node *lnode; /* Local tree. */
 112         /* Value of main tree node (with all value factors, but unbiased
 113          * - without exploration factor), from black's perspective. */
 114         struct move_stats value;
 115 };
 116
 117
 118 typedef struct tree_node *(*uctp_choose)(struct uct_policy *p, struct tree_node *node, struct board *b, enum stone color, coord_t exclude);
 119 typedef float (*uctp_evaluate)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity);
 120 typedef void (*uctp_descend)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity, bool allow_pass);
 121 typedef void (*uctp_winner)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent);
 122 typedef void (*uctp_prior)(struct uct_policy *p, struct tree *tree, struct tree_node *node, struct board *b, enum stone color, int parity);
 123 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);
 124
 125 struct uct_policy {
 126         struct uct *uct;
 127         uctp_choose choose;
 128         uctp_winner winner;
 129         uctp_evaluate evaluate;
 130         uctp_descend descend;
 131         uctp_update update;
 132         uctp_prior prior;
 133         bool wants_amaf;
 134         void *data;
 135 };
 136
 137 #endif