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_opt; /* User option */
  34         bool pass_all_alive; /* Current value */
  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_tbook;
  43         bool fast_alloc;
  44         unsigned long max_tree_size;
  45         unsigned long max_pruned_size;
  46         unsigned long pruning_threshold;
  47         int mercymin;
  48         int significant_threshold;
  49
  50         int threads;
  51         enum uct_thread_model {
  52                 TM_TREE, /* Tree parallelization w/o virtual loss. */
  53                 TM_TREEVL, /* Tree parallelization with virtual loss. */
  54         } thread_model;
  55         int 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         int max_slaves; /* Optional, -1 if not set */
  60         int slave_index; /* 0..max_slaves-1, or -1 if not set */
  61         enum stone my_color;
  62
  63         int fuseki_end;
  64         int yose_start;
  65
  66         int dynkomi_mask;
  67         int dynkomi_interval;
  68         struct uct_dynkomi *dynkomi;
  69
  70         floating_t val_scale;
  71         int val_points;
  72         bool val_extra;
  73
  74         int random_policy_chance;
  75         bool local_tree;
  76         int tenuki_d;
  77         floating_t local_tree_aging;
  78 #define LTREE_PLAYOUTS_MULTIPLIER 100
  79         floating_t local_tree_depth_decay;
  80         bool local_tree_allseq;
  81         bool local_tree_neival;
  82         bool local_tree_rootgoal;
  83         bool local_tree_rootchoose;
  84
  85         char *banner;
  86
  87         struct uct_policy *policy;
  88         struct uct_policy *random_policy;
  89         struct playout_policy *playout;
  90         struct uct_prior *prior;
  91         struct uct_pluginset *plugins;
  92         struct joseki_dict *jdict;
  93
  94         /* Used within frame of single genmove. */
  95         struct board_ownermap ownermap;
  96         /* Used for coordination among slaves of the distributed engine. */
  97         int stats_hbits;
  98         int shared_nodes;
  99         int shared_levels;
 100         double stats_delay; /* stored in seconds */
 101         int played_own;
 102         int played_all; /* games played by all slaves */
 103
 104         /* Game state - maintained by setup_state(), reset_state(). */
 105         struct tree *t;
 106 };
 107
 108 #define UDEBUGL(n) DEBUGL_(u->debug_level, n)
 109
 110 bool uct_pass_is_safe(struct uct *u, struct board *b, enum stone color, bool pass_all_alive);
 111
 112 void uct_prepare_move(struct uct *u, struct board *b, enum stone color);
 113 void uct_genmove_setup(struct uct *u, struct board *b, enum stone color);
 114 void uct_pondering_stop(struct uct *u);
 115
 116
 117 /* This is the state used for descending the tree; we use this wrapper
 118  * structure in order to be able to easily descend in multiple trees
 119  * in parallel (e.g. main tree and local tree) or compute cummulative
 120  * "path value" throughout the tree descent. */
 121 struct uct_descent {
 122         /* Active tree nodes: */
 123         struct tree_node *node; /* Main tree. */
 124         struct tree_node *lnode; /* Local tree. */
 125         /* Value of main tree node (with all value factors, but unbiased
 126          * - without exploration factor), from black's perspective. */
 127         struct move_stats value;
 128 };
 129
 130
 131 typedef struct tree_node *(*uctp_choose)(struct uct_policy *p, struct tree_node *node, struct board *b, enum stone color, coord_t exclude);
 132 typedef floating_t (*uctp_evaluate)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity);
 133 typedef void (*uctp_descend)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity, bool allow_pass);
 134 typedef void (*uctp_winner)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent);
 135 typedef void (*uctp_prior)(struct uct_policy *p, struct tree *tree, struct tree_node *node, struct board *b, enum stone color, int parity);
 136 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);
 137
 138 struct uct_policy {
 139         struct uct *uct;
 140         uctp_choose choose;
 141         uctp_winner winner;
 142         uctp_evaluate evaluate;
 143         uctp_descend descend;
 144         uctp_update update;
 145         uctp_prior prior;
 146         bool wants_amaf;
 147         void *data;
 148 };
 149
 150 #endif