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         floating_t max_maintime_ratio;
  33         bool pass_all_alive;
  34         bool territory_scoring;
  35         int expand_p;
  36         bool playout_amaf, playout_amaf_nakade;
  37         bool amaf_prior;
  38         int playout_amaf_cutoff;
  39         int dumpthres;
  40         int force_seed;
  41         bool no_tbook;
  42         bool fast_alloc;
  43         unsigned long max_tree_size;
  44         unsigned long max_pruned_size;
  45         unsigned long pruning_threshold;
  46         int mercymin;
  47         int significant_threshold;
  48
  49         int threads;
  50         enum uct_thread_model {
  51                 TM_TREE, /* Tree parallelization w/o virtual loss. */
  52                 TM_TREEVL, /* Tree parallelization with virtual loss. */
  53         } thread_model;
  54         int 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         floating_t val_scale;
  68         int val_points;
  69         bool val_extra;
  70
  71         int treepool_chance[2];
  72         int treepool_size;
  73         enum treepool_type {
  74                 UTT_RAVE_PLAYOUTS,
  75                 UTT_RAVE_VALUE,
  76                 UTT_UCT_PLAYOUTS,
  77                 UTT_UCT_VALUE,
  78                 UTT_EVALUATE,
  79         } treepool_type;
  80         int treepool_pickfactor;
  81
  82         int random_policy_chance;
  83         int local_tree;
  84         int tenuki_d;
  85         floating_t local_tree_aging;
  86 #define LTREE_PLAYOUTS_MULTIPLIER 100
  87         floating_t local_tree_depth_decay;
  88         bool local_tree_allseq;
  89         /* Playout-localtree integration. */
  90         bool local_tree_pseqroot;
  91         bool local_tree_rootseqval;
  92
  93         char *banner;
  94
  95         struct uct_policy *policy;
  96         struct uct_policy *random_policy;
  97         struct playout_policy *playout;
  98         struct uct_prior *prior;
  99         struct uct_pluginset *plugins;
 100         struct joseki_dict *jdict;
 101
 102         /* Used within frame of single genmove. */
 103         struct board_ownermap ownermap;
 104         /* Used for coordination among slaves of the distributed engine. */
 105         int stats_hbits;
 106         int shared_nodes;
 107         int shared_levels;
 108         int played_own;
 109         int played_all; /* games played by all slaves */
 110
 111         /* Game state - maintained by setup_state(), reset_state(). */
 112         struct tree *t;
 113 };
 114
 115 #define UDEBUGL(n) DEBUGL_(u->debug_level, n)
 116
 117 bool uct_pass_is_safe(struct uct *u, struct board *b, enum stone color, bool pass_all_alive);
 118
 119 void uct_prepare_move(struct uct *u, struct board *b, enum stone color);
 120 void uct_genmove_setup(struct uct *u, struct board *b, enum stone color);
 121 void uct_pondering_stop(struct uct *u);
 122
 123
 124 /* This is the state used for descending the tree; we use this wrapper
 125  * structure in order to be able to easily descend in multiple trees
 126  * in parallel (e.g. main tree and local tree) or compute cummulative
 127  * "path value" throughout the tree descent. */
 128 struct uct_descent {
 129         /* Active tree nodes: */
 130         struct tree_node *node; /* Main tree. */
 131         struct tree_node *lnode; /* Local tree. */
 132         /* Value of main tree node (with all value factors, but unbiased
 133          * - without exploration factor), from black's perspective. */
 134         struct move_stats value;
 135 };
 136
 137
 138 typedef struct tree_node *(*uctp_choose)(struct uct_policy *p, struct tree_node *node, struct board *b, enum stone color, coord_t exclude);
 139 typedef floating_t (*uctp_evaluate)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity);
 140 typedef void (*uctp_descend)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity, bool allow_pass);
 141 typedef void (*uctp_winner)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent);
 142 typedef void (*uctp_prior)(struct uct_policy *p, struct tree *tree, struct tree_node *node, struct board *b, enum stone color, int parity);
 143 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);
 144
 145 struct uct_policy {
 146         struct uct *uct;
 147         uctp_choose choose;
 148         uctp_winner winner;
 149         uctp_evaluate evaluate;
 150         uctp_descend descend;
 151         uctp_update update;
 152         uctp_prior prior;
 153         bool wants_amaf;
 154         void *data;
 155 };
 156
 157 #endif