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 #include "distributed/distributed.h"
  12
  13 struct tree;
  14 struct tree_node;
  15 struct uct_policy;
  16 struct uct_prior;
  17 struct uct_dynkomi;
  18
  19 /* Internal UCT structures */
  20
  21 /* Stats for each child of the root node. */
  22 struct node_stats {
  23         struct move_stats2 last_sent_own;
  24         struct move_stats2 added_from_others;
  25         struct tree_node *node;
  26 };
  27
  28 /* Internal engine state. */
  29 struct uct {
  30         int debug_level;
  31         int games, gamelen;
  32         float resign_ratio;
  33         float loss_threshold;
  34         double best2_ratio, bestr_ratio;
  35         bool pass_all_alive;
  36         bool territory_scoring;
  37         int expand_p;
  38         bool playout_amaf, playout_amaf_nakade;
  39         bool amaf_prior;
  40         int playout_amaf_cutoff;
  41         int dumpthres;
  42         int force_seed;
  43         bool no_book;
  44         bool fast_alloc;
  45         unsigned long max_tree_size;
  46         int mercymin;
  47
  48         int threads;
  49         enum uct_thread_model {
  50                 TM_ROOT, /* Root parallelization. */
  51                 TM_TREE, /* Tree parallelization w/o virtual loss. */
  52                 TM_TREEVL, /* Tree parallelization with virtual loss. */
  53         } thread_model;
  54         bool parallel_tree;
  55         bool virtual_loss;
  56         bool pondering_opt; /* User wants pondering */
  57         bool pondering; /* Actually pondering now */
  58         bool slave; /* Act as slave in distributed engine. */
  59         enum stone my_color;
  60
  61         int fuseki_end;
  62         int yose_start;
  63
  64         int dynkomi_mask;
  65         int dynkomi_interval;
  66         struct uct_dynkomi *dynkomi;
  67
  68         float val_scale;
  69         int val_points;
  70         bool val_extra;
  71
  72         int random_policy_chance;
  73         int local_tree;
  74         int tenuki_d;
  75         float local_tree_aging;
  76         bool local_tree_allseq;
  77         /* Playout-localtree integration. */
  78         bool local_tree_playout; // can be true only if ELO playout
  79         bool local_tree_pseqroot;
  80
  81         char *banner;
  82
  83         struct uct_policy *policy;
  84         struct uct_policy *random_policy;
  85         struct playout_policy *playout;
  86         struct uct_prior *prior;
  87
  88         /* Used within frame of single genmove. */
  89         struct board_ownermap ownermap;
  90         struct node_stats *stats;
  91         int played_own;
  92         int played_all; /* games played by all slaves */
  93
  94         /* Game state - maintained by setup_state(), reset_state(). */
  95         struct tree *t;
  96 };
  97
  98 #define UDEBUGL(n) DEBUGL_(u->debug_level, n)
  99
 100 extern volatile sig_atomic_t uct_halt;
 101 extern __thread int thread_id;
 102
 103 bool uct_pass_is_safe(struct uct *u, struct board *b, enum stone color, bool pass_all_alive);
 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