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 /* Stats for each child of the root node. */
  21 struct node_stats {
  22         struct move_stats last_sent_own;
  23         struct move_stats added_from_others;
  24         struct tree_node *node;
  25 };
  26
  27 /* Internal engine state. */
  28 struct uct {
  29         int debug_level;
  30         int games, gamelen;
  31         float resign_ratio;
  32         float loss_threshold;
  33         double best2_ratio, bestr_ratio;
  34         bool pass_all_alive;
  35         bool territory_scoring;
  36         int expand_p;
  37         bool playout_amaf, playout_amaf_nakade;
  38         bool amaf_prior;
  39         int playout_amaf_cutoff;
  40         int dumpthres;
  41         int force_seed;
  42         bool no_book;
  43         bool fast_alloc;
  44         unsigned long max_tree_size;
  45         int mercymin;
  46
  47         int threads;
  48         enum uct_thread_model {
  49                 TM_ROOT, /* Root parallelization. */
  50                 TM_TREE, /* Tree parallelization w/o virtual loss. */
  51                 TM_TREEVL, /* Tree parallelization with virtual loss. */
  52         } thread_model;
  53         bool parallel_tree;
  54         bool virtual_loss;
  55         bool pondering_opt; /* User wants pondering */
  56         bool pondering; /* Actually pondering now */
  57         bool slave; /* Act as slave in distributed engine. */
  58         enum stone my_color;
  59
  60         int fuseki_end;
  61         int yose_start;
  62
  63         int dynkomi_mask;
  64         int dynkomi_interval;
  65         struct uct_dynkomi *dynkomi;
  66
  67         float val_scale;
  68         int val_points;
  69         bool val_extra;
  70
  71         int random_policy_chance;
  72         int local_tree;
  73         int tenuki_d;
  74         float local_tree_aging;
  75         bool local_tree_allseq;
  76         /* Playout-localtree integration. */
  77         bool local_tree_playout; // can be true only if ELO playout
  78         bool local_tree_pseqroot;
  79
  80         char *banner;
  81
  82         struct uct_policy *policy;
  83         struct uct_policy *random_policy;
  84         struct playout_policy *playout;
  85         struct uct_prior *prior;
  86
  87         /* Used within frame of single genmove. */
  88         struct board_ownermap ownermap;
  89         struct node_stats *stats;
  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 extern volatile sig_atomic_t uct_halt;
 100 extern __thread int thread_id;
 101
 102 bool uct_pass_is_safe(struct uct *u, struct board *b, enum stone color, bool pass_all_alive);
 103
 104
 105 /* This is the state used for descending the tree; we use this wrapper
 106  * structure in order to be able to easily descend in multiple trees
 107  * in parallel (e.g. main tree and local tree) or compute cummulative
 108  * "path value" throughout the tree descent. */
 109 struct uct_descent {
 110         /* Active tree nodes: */
 111         struct tree_node *node; /* Main tree. */
 112         struct tree_node *lnode; /* Local tree. */
 113         /* Value of main tree node (with all value factors, but unbiased
 114          * - without exploration factor), from black's perspective. */
 115         struct move_stats value;
 116 };
 117
 118
 119 typedef struct tree_node *(*uctp_choose)(struct uct_policy *p, struct tree_node *node, struct board *b, enum stone color, coord_t exclude);
 120 typedef float (*uctp_evaluate)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity);
 121 typedef void (*uctp_descend)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity, bool allow_pass);
 122 typedef void (*uctp_winner)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent);
 123 typedef void (*uctp_prior)(struct uct_policy *p, struct tree *tree, struct tree_node *node, struct board *b, enum stone color, int parity);
 124 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);
 125
 126 struct uct_policy {
 127         struct uct *uct;
 128         uctp_choose choose;
 129         uctp_winner winner;
 130         uctp_evaluate evaluate;
 131         uctp_descend descend;
 132         uctp_update update;
 133         uctp_prior prior;
 134         bool wants_amaf;
 135         void *data;
 136 };
 137
 138 #endif