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 struct uct_pluginset;
  18
  19 /* How big proportion of ownermap counts must be of one color to consider
  20  * the point sure. */
  21 #define GJ_THRES        0.8
  22 /* How many games to consider at minimum before judging groups. */
  23 #define GJ_MINGAMES     500
  24
  25 /* Internal engine state. */
  26 struct uct {
  27         int debug_level;
  28         int games, gamelen;
  29         float resign_ratio;
  30         float loss_threshold;
  31         double best2_ratio, bestr_ratio;
  32         bool pass_all_alive;
  33         bool territory_scoring;
  34         int expand_p;
  35         bool playout_amaf, playout_amaf_nakade;
  36         bool amaf_prior;
  37         int playout_amaf_cutoff;
  38         int dumpthres;
  39         int force_seed;
  40         bool no_book;
  41         bool fast_alloc;
  42         unsigned long max_tree_size;
  43         int mercymin;
  44
  45         int threads;
  46         enum uct_thread_model {
  47                 TM_TREE, /* Tree parallelization w/o virtual loss. */
  48                 TM_TREEVL, /* Tree parallelization with virtual loss. */
  49         } thread_model;
  50         bool virtual_loss;
  51         bool pondering_opt; /* User wants pondering */
  52         bool pondering; /* Actually pondering now */
  53         bool slave; /* Act as slave in distributed engine. */
  54         enum stone my_color;
  55
  56         int fuseki_end;
  57         int yose_start;
  58
  59         int dynkomi_mask;
  60         int dynkomi_interval;
  61         struct uct_dynkomi *dynkomi;
  62
  63         float val_scale;
  64         int val_points;
  65         bool val_extra;
  66
  67         int random_policy_chance;
  68         int local_tree;
  69         int tenuki_d;
  70         float local_tree_aging;
  71         bool local_tree_allseq;
  72         /* Playout-localtree integration. */
  73         bool local_tree_playout; // can be true only if ELO playout
  74         bool local_tree_pseqroot;
  75
  76         char *banner;
  77
  78         struct uct_policy *policy;
  79         struct uct_policy *random_policy;
  80         struct playout_policy *playout;
  81         struct uct_prior *prior;
  82         struct uct_pluginset *plugins;
  83
  84         /* Used within frame of single genmove. */
  85         struct board_ownermap ownermap;
  86         /* Used for coordination among slaves of the distributed engine. */
  87         int stats_hbits;
  88         int shared_nodes;
  89         int shared_levels;
  90         int played_own;
  91         int played_all; /* games played by all slaves */
  92
  93         /* Game state - maintained by setup_state(), reset_state(). */
  94         struct tree *t;
  95 };
  96
  97 #define UDEBUGL(n) DEBUGL_(u->debug_level, n)
  98
  99 bool uct_pass_is_safe(struct uct *u, struct board *b, enum stone color, bool pass_all_alive);
 100
 101 void uct_prepare_move(struct uct *u, struct board *b, enum stone color);
 102 void uct_genmove_setup(struct uct *u, struct board *b, enum stone color);
 103 void uct_pondering_stop(struct uct *u);
 104
 105
 106 /* This is the state used for descending the tree; we use this wrapper
 107  * structure in order to be able to easily descend in multiple trees
 108  * in parallel (e.g. main tree and local tree) or compute cummulative
 109  * "path value" throughout the tree descent. */
 110 struct uct_descent {
 111         /* Active tree nodes: */
 112         struct tree_node *node; /* Main tree. */
 113         struct tree_node *lnode; /* Local tree. */
 114         /* Value of main tree node (with all value factors, but unbiased
 115          * - without exploration factor), from black's perspective. */
 116         struct move_stats value;
 117 };
 118
 119
 120 typedef struct tree_node *(*uctp_choose)(struct uct_policy *p, struct tree_node *node, struct board *b, enum stone color, coord_t exclude);
 121 typedef float (*uctp_evaluate)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity);
 122 typedef void (*uctp_descend)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity, bool allow_pass);
 123 typedef void (*uctp_winner)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent);
 124 typedef void (*uctp_prior)(struct uct_policy *p, struct tree *tree, struct tree_node *node, struct board *b, enum stone color, int parity);
 125 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);
 126
 127 struct uct_policy {
 128         struct uct *uct;
 129         uctp_choose choose;
 130         uctp_winner winner;
 131         uctp_evaluate evaluate;
 132         uctp_descend descend;
 133         uctp_update update;
 134         uctp_prior prior;
 135         bool wants_amaf;
 136         void *data;
 137 };
 138
 139 #endif