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 "pattern.h"
  10 #include "patternsp.h"
  11 #include "patternprob.h"
  12 #include "playout.h"
  13 #include "stats.h"
  14
  15 struct tree;
  16 struct tree_node;
  17 struct uct_policy;
  18 struct uct_prior;
  19 struct uct_dynkomi;
  20 struct uct_pluginset;
  21 struct uct_pattern;
  22 struct joseki_dict;
  23
  24 /* How big proportion of ownermap counts must be of one color to consider
  25  * the point sure. */
  26 #define GJ_THRES        0.8
  27 /* How many games to consider at minimum before judging groups. */
  28 #define GJ_MINGAMES     500
  29
  30 /* TODO: Generalize this for other engines too. */
  31 struct uct_pattern {
  32         struct pattern_config pc;
  33         pattern_spec ps;
  34         struct pattern_pdict *pd;
  35 };
  36
  37 /* Internal engine state. */
  38 struct uct {
  39         int debug_level;
  40         int games, gamelen;
  41         floating_t resign_threshold, sure_win_threshold;
  42         double best2_ratio, bestr_ratio;
  43         floating_t max_maintime_ratio;
  44         bool pass_all_alive; /* Current value */
  45         bool territory_scoring;
  46         int expand_p;
  47         bool playout_amaf;
  48         bool amaf_prior;
  49         int playout_amaf_cutoff;
  50         int dumpthres;
  51         int force_seed;
  52         bool no_tbook;
  53         bool fast_alloc;
  54         unsigned long max_tree_size;
  55         unsigned long max_pruned_size;
  56         unsigned long pruning_threshold;
  57         int mercymin;
  58         int significant_threshold;
  59
  60         int threads;
  61         enum uct_thread_model {
  62                 TM_TREE, /* Tree parallelization w/o virtual loss. */
  63                 TM_TREEVL, /* Tree parallelization with virtual loss. */
  64         } thread_model;
  65         int virtual_loss;
  66         bool pondering_opt; /* User wants pondering */
  67         bool pondering; /* Actually pondering now */
  68         bool slave; /* Act as slave in distributed engine. */
  69         int max_slaves; /* Optional, -1 if not set */
  70         int slave_index; /* 0..max_slaves-1, or -1 if not set */
  71         enum stone my_color;
  72
  73         int fuseki_end;
  74         int yose_start;
  75
  76         int dynkomi_mask;
  77         int dynkomi_interval;
  78         struct uct_dynkomi *dynkomi;
  79
  80         floating_t val_scale;
  81         int val_points;
  82         bool val_extra;
  83
  84         int random_policy_chance;
  85         bool local_tree;
  86         int tenuki_d;
  87         floating_t local_tree_aging;
  88 #define LTREE_PLAYOUTS_MULTIPLIER 100
  89         floating_t local_tree_depth_decay;
  90         bool local_tree_allseq;
  91         bool local_tree_neival;
  92         enum {
  93                 LTE_ROOT,
  94                 LTE_EACH,
  95                 LTE_TOTAL,
  96         } local_tree_eval;
  97         bool local_tree_rootchoose;
  98
  99         char *banner;
 100
 101         struct uct_policy *policy;
 102         struct uct_policy *random_policy;
 103         struct playout_policy *playout;
 104         struct uct_prior *prior;
 105         struct uct_pluginset *plugins;
 106         struct joseki_dict *jdict;
 107         struct uct_pattern pat;
 108
 109         /* Used within frame of single genmove. */
 110         struct board_ownermap ownermap;
 111         /* Used for coordination among slaves of the distributed engine. */
 112         int stats_hbits;
 113         int shared_nodes;
 114         int shared_levels;
 115         double stats_delay; /* stored in seconds */
 116         int played_own;
 117         int played_all; /* games played by all slaves */
 118
 119         /* Game state - maintained by setup_state(), reset_state(). */
 120         struct tree *t;
 121 };
 122
 123 #define UDEBUGL(n) DEBUGL_(u->debug_level, n)
 124
 125 bool uct_pass_is_safe(struct uct *u, struct board *b, enum stone color, bool pass_all_alive);
 126
 127 void uct_prepare_move(struct uct *u, struct board *b, enum stone color);
 128 void uct_genmove_setup(struct uct *u, struct board *b, enum stone color);
 129 void uct_pondering_stop(struct uct *u);
 130
 131
 132 /* This is the state used for descending the tree; we use this wrapper
 133  * structure in order to be able to easily descend in multiple trees
 134  * in parallel (e.g. main tree and local tree) or compute cummulative
 135  * "path value" throughout the tree descent. */
 136 struct uct_descent {
 137         /* Active tree nodes: */
 138         struct tree_node *node; /* Main tree. */
 139         struct tree_node *lnode; /* Local tree. */
 140         /* Value of main tree node (with all value factors, but unbiased
 141          * - without exploration factor), from black's perspective. */
 142         struct move_stats value;
 143 };
 144
 145
 146 typedef struct tree_node *(*uctp_choose)(struct uct_policy *p, struct tree_node *node, struct board *b, enum stone color, coord_t exclude);
 147 typedef floating_t (*uctp_evaluate)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity);
 148 typedef void (*uctp_descend)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity, bool allow_pass);
 149 typedef void (*uctp_winner)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent);
 150 typedef void (*uctp_prior)(struct uct_policy *p, struct tree *tree, struct tree_node *node, struct board *b, enum stone color, int parity);
 151 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);
 152
 153 struct uct_policy {
 154         struct uct *uct;
 155         uctp_choose choose;
 156         uctp_winner winner;
 157         uctp_evaluate evaluate;
 158         uctp_descend descend;
 159         uctp_update update;
 160         uctp_prior prior;
 161         bool wants_amaf;
 162         void *data;
 163 };
 164
 165 #endif