uct/internal.h

   1 #ifndef PACHI_UCT_INTERNAL_H
   2 #define PACHI_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; /* Current value */
  34         bool territory_scoring;
  35         int expand_p;
  36         bool playout_amaf;
  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         int max_slaves; /* Optional, -1 if not set */
  59         int slave_index; /* 0..max_slaves-1, or -1 if not set */
  60         enum stone my_color;
  61
  62         int fuseki_end;
  63         int yose_start;
  64
  65         int dynkomi_mask;
  66         int dynkomi_interval;
  67         struct uct_dynkomi *dynkomi;
  68
  69         floating_t val_scale;
  70         int val_points;
  71         bool val_extra;
  72
  73         int random_policy_chance;
  74         bool local_tree;
  75         int tenuki_d;
  76         floating_t local_tree_aging;
  77 #define LTREE_PLAYOUTS_MULTIPLIER 100
  78         floating_t local_tree_depth_decay;
  79         bool local_tree_allseq;
  80         bool local_tree_neival;
  81         enum {
  82                 LTE_ROOT,
  83                 LTE_EACH,
  84         } local_tree_eval;
  85         bool local_tree_rootchoose;
  86
  87         char *banner;
  88
  89         struct uct_policy *policy;
  90         struct uct_policy *random_policy;
  91         struct playout_policy *playout;
  92         struct uct_prior *prior;
  93         struct uct_pluginset *plugins;
  94         struct joseki_dict *jdict;
  95
  96         /* Used within frame of single genmove. */
  97         struct board_ownermap ownermap;
  98         /* Used for coordination among slaves of the distributed engine. */
  99         int stats_hbits;
 100         int shared_nodes;
 101         int shared_levels;
 102         double stats_delay; /* stored in seconds */
 103         int played_own;
 104         int played_all; /* games played by all slaves */
 105
 106         /* Game state - maintained by setup_state(), reset_state(). */
 107         struct tree *t;
 108 };
 109
 110 #define UDEBUGL(n) DEBUGL_(u->debug_level, n)
 111
 112 bool uct_pass_is_safe(struct uct *u, struct board *b, enum stone color, bool pass_all_alive);
 113
 114 void uct_prepare_move(struct uct *u, struct board *b, enum stone color);
 115 void uct_genmove_setup(struct uct *u, struct board *b, enum stone color);
 116 void uct_pondering_stop(struct uct *u);
 117
 118
 119 /* This is the state used for descending the tree; we use this wrapper
 120  * structure in order to be able to easily descend in multiple trees
 121  * in parallel (e.g. main tree and local tree) or compute cummulative
 122  * "path value" throughout the tree descent. */
 123 struct uct_descent {
 124         /* Active tree nodes: */
 125         struct tree_node *node; /* Main tree. */
 126         struct tree_node *lnode; /* Local tree. */
 127         /* Value of main tree node (with all value factors, but unbiased
 128          * - without exploration factor), from black's perspective. */
 129         struct move_stats value;
 130 };
 131
 132
 133 typedef struct tree_node *(*uctp_choose)(struct uct_policy *p, struct tree_node *node, struct board *b, enum stone color, coord_t exclude);
 134 typedef floating_t (*uctp_evaluate)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity);
 135 typedef void (*uctp_descend)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity, bool allow_pass);
 136 typedef void (*uctp_winner)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent);
 137 typedef void (*uctp_prior)(struct uct_policy *p, struct tree *tree, struct tree_node *node, struct board *b, enum stone color, int parity);
 138 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, struct board *final_board, floating_t result);
 139
 140 struct uct_policy {
 141         struct uct *uct;
 142         uctp_choose choose;
 143         uctp_winner winner;
 144         uctp_evaluate evaluate;
 145         uctp_descend descend;
 146         uctp_update update;
 147         uctp_prior prior;
 148         bool wants_amaf;
 149         void *data;
 150 };
 151
 152 #endif