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 struct joseki_dict;
  19
  20 /* How big proportion of ownermap counts must be of one color to consider
  21  * the point sure. */
  22 #define GJ_THRES        0.8
  23 /* How many games to consider at minimum before judging groups. */
  24 #define GJ_MINGAMES     500
  25
  26 /* Internal engine state. */
  27 struct uct {
  28         int debug_level;
  29         int games, gamelen;
  30         floating_t resign_threshold, sure_win_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_tbook;
  41         bool fast_alloc;
  42         unsigned long max_tree_size;
  43         unsigned long max_pruned_size;
  44         unsigned long pruning_threshold;
  45         int mercymin;
  46         int significant_threshold;
  47
  48         int threads;
  49         enum uct_thread_model {
  50                 TM_TREE, /* Tree parallelization w/o virtual loss. */
  51                 TM_TREEVL, /* Tree parallelization with virtual loss. */
  52         } thread_model;
  53         bool virtual_loss;
  54         bool pondering_opt; /* User wants pondering */
  55         bool pondering; /* Actually pondering now */
  56         bool slave; /* Act as slave in distributed engine. */
  57         enum stone my_color;
  58
  59         int fuseki_end;
  60         int yose_start;
  61
  62         int dynkomi_mask;
  63         int dynkomi_interval;
  64         struct uct_dynkomi *dynkomi;
  65
  66         floating_t val_scale;
  67         int val_points;
  68         bool val_extra;
  69
  70         int treepool_chance[2];
  71         int treepool_size;
  72         enum treepool_type {
  73                 UTT_RAVE_PLAYOUTS,
  74                 UTT_RAVE_VALUE,
  75                 UTT_UCT_PLAYOUTS,
  76                 UTT_UCT_VALUE,
  77                 UTT_EVALUATE,
  78         } treepool_type;
  79         int treepool_pickfactor;
  80
  81         int random_policy_chance;
  82         int local_tree;
  83         int tenuki_d;
  84         floating_t local_tree_aging;
  85         bool local_tree_allseq;
  86         /* Playout-localtree integration. */
  87         bool local_tree_playout; // can be true only if ELO playout
  88         bool local_tree_pseqroot;
  89
  90         char *banner;
  91
  92         struct uct_policy *policy;
  93         struct uct_policy *random_policy;
  94         struct playout_policy *playout;
  95         struct uct_prior *prior;
  96         struct uct_pluginset *plugins;
  97         struct joseki_dict *jdict;
  98
  99         /* Used within frame of single genmove. */
 100         struct board_ownermap ownermap;
 101         /* Used for coordination among slaves of the distributed engine. */
 102         int stats_hbits;
 103         int shared_nodes;
 104         int shared_levels;
 105         int played_own;
 106         int played_all; /* games played by all slaves */
 107
 108         /* Game state - maintained by setup_state(), reset_state(). */
 109         struct tree *t;
 110 };
 111
 112 #define UDEBUGL(n) DEBUGL_(u->debug_level, n)
 113
 114 bool uct_pass_is_safe(struct uct *u, struct board *b, enum stone color, bool pass_all_alive);
 115
 116 void uct_prepare_move(struct uct *u, struct board *b, enum stone color);
 117 void uct_genmove_setup(struct uct *u, struct board *b, enum stone color);
 118 void uct_pondering_stop(struct uct *u);
 119
 120
 121 /* This is the state used for descending the tree; we use this wrapper
 122  * structure in order to be able to easily descend in multiple trees
 123  * in parallel (e.g. main tree and local tree) or compute cummulative
 124  * "path value" throughout the tree descent. */
 125 struct uct_descent {
 126         /* Active tree nodes: */
 127         struct tree_node *node; /* Main tree. */
 128         struct tree_node *lnode; /* Local tree. */
 129         /* Value of main tree node (with all value factors, but unbiased
 130          * - without exploration factor), from black's perspective. */
 131         struct move_stats value;
 132 };
 133
 134
 135 typedef struct tree_node *(*uctp_choose)(struct uct_policy *p, struct tree_node *node, struct board *b, enum stone color, coord_t exclude);
 136 typedef floating_t (*uctp_evaluate)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity);
 137 typedef void (*uctp_descend)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity, bool allow_pass);
 138 typedef void (*uctp_winner)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent);
 139 typedef void (*uctp_prior)(struct uct_policy *p, struct tree *tree, struct tree_node *node, struct board *b, enum stone color, int parity);
 140 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, floating_t result);
 141
 142 struct uct_policy {
 143         struct uct *uct;
 144         uctp_choose choose;
 145         uctp_winner winner;
 146         uctp_evaluate evaluate;
 147         uctp_descend descend;
 148         uctp_update update;
 149         uctp_prior prior;
 150         bool wants_amaf;
 151         void *data;
 152 };
 153
 154 #endif