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 joseki_dict;
  22
  23 /* How big proportion of ownermap counts must be of one color to consider
  24  * the point sure. */
  25 #define GJ_THRES        0.8
  26 /* How many games to consider at minimum before judging groups. */
  27 #define GJ_MINGAMES     500
  28
  29 /* Internal engine state. */
  30 struct uct {
  31         int debug_level;
  32         enum uct_reporting {
  33                 UR_TEXT,
  34                 UR_JSON,
  35                 UR_JSON_BIG,
  36         } reporting;
  37         int reportfreq;
  38
  39         int games, gamelen;
  40         floating_t resign_threshold, sure_win_threshold;
  41         double best2_ratio, bestr_ratio;
  42         floating_t max_maintime_ratio;
  43         bool pass_all_alive; /* Current value */
  44         bool territory_scoring;
  45         int expand_p;
  46         bool playout_amaf;
  47         bool amaf_prior;
  48         int playout_amaf_cutoff;
  49         int dumpthres;
  50         int force_seed;
  51         bool no_tbook;
  52         bool fast_alloc;
  53         unsigned long max_tree_size;
  54         unsigned long max_pruned_size;
  55         unsigned long pruning_threshold;
  56         int mercymin;
  57         int significant_threshold;
  58
  59         int threads;
  60         enum uct_thread_model {
  61                 TM_TREE, /* Tree parallelization w/o virtual loss. */
  62                 TM_TREEVL, /* Tree parallelization with virtual loss. */
  63         } thread_model;
  64         int virtual_loss;
  65         bool pondering_opt; /* User wants pondering */
  66         bool pondering; /* Actually pondering now */
  67         bool slave; /* Act as slave in distributed engine. */
  68         int max_slaves; /* Optional, -1 if not set */
  69         int slave_index; /* 0..max_slaves-1, or -1 if not set */
  70         enum stone my_color;
  71
  72         int fuseki_end;
  73         int yose_start;
  74
  75         int dynkomi_mask;
  76         int dynkomi_interval;
  77         struct uct_dynkomi *dynkomi;
  78
  79         floating_t val_scale;
  80         int val_points;
  81         bool val_extra;
  82
  83         int random_policy_chance;
  84         bool local_tree;
  85         int tenuki_d;
  86         floating_t local_tree_aging;
  87 #define LTREE_PLAYOUTS_MULTIPLIER 100
  88         floating_t local_tree_depth_decay;
  89         bool local_tree_allseq;
  90         bool local_tree_neival;
  91         enum {
  92                 LTE_ROOT,
  93                 LTE_EACH,
  94                 LTE_TOTAL,
  95         } local_tree_eval;
  96         bool local_tree_rootchoose;
  97
  98         char *banner;
  99
 100         struct uct_policy *policy;
 101         struct uct_policy *random_policy;
 102         struct playout_policy *playout;
 103         struct uct_prior *prior;
 104         struct uct_pluginset *plugins;
 105         struct joseki_dict *jdict;
 106
 107         struct pattern_setup pat;
 108         /* Various modules (prior, policy, ...) set this if they want pattern
 109          * database to be loaded. */
 110         bool want_pat;
 111
 112         /* Used within frame of single genmove. */
 113         struct board_ownermap ownermap;
 114         /* Used for coordination among slaves of the distributed engine. */
 115         int stats_hbits;
 116         int shared_nodes;
 117         int shared_levels;
 118         double stats_delay; /* stored in seconds */
 119         int played_own;
 120         int played_all; /* games played by all slaves */
 121
 122         /* Game state - maintained by setup_state(), reset_state(). */
 123         struct tree *t;
 124 };
 125
 126 #define UDEBUGL(n) DEBUGL_(u->debug_level, n)
 127
 128 bool uct_pass_is_safe(struct uct *u, struct board *b, enum stone color, bool pass_all_alive);
 129
 130 void uct_prepare_move(struct uct *u, struct board *b, enum stone color);
 131 void uct_genmove_setup(struct uct *u, struct board *b, enum stone color);
 132 void uct_pondering_stop(struct uct *u);
 133
 134
 135 /* This is the state used for descending the tree; we use this wrapper
 136  * structure in order to be able to easily descend in multiple trees
 137  * in parallel (e.g. main tree and local tree) or compute cummulative
 138  * "path value" throughout the tree descent. */
 139 struct uct_descent {
 140         /* Active tree nodes: */
 141         struct tree_node *node; /* Main tree. */
 142         struct tree_node *lnode; /* Local tree. */
 143         /* Value of main tree node (with all value factors, but unbiased
 144          * - without exploration factor), from black's perspective. */
 145         struct move_stats value;
 146 };
 147
 148
 149 typedef struct tree_node *(*uctp_choose)(struct uct_policy *p, struct tree_node *node, struct board *b, enum stone color, coord_t exclude);
 150 typedef floating_t (*uctp_evaluate)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity);
 151 typedef void (*uctp_descend)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity, bool allow_pass);
 152 typedef void (*uctp_winner)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent);
 153 typedef void (*uctp_prior)(struct uct_policy *p, struct tree *tree, struct tree_node *node, struct board *b, enum stone color, int parity);
 154 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);
 155
 156 struct uct_policy {
 157         struct uct *uct;
 158         uctp_choose choose;
 159         uctp_winner winner;
 160         uctp_evaluate evaluate;
 161         uctp_descend descend;
 162         uctp_update update;
 163         uctp_prior prior;
 164         bool wants_amaf;
 165         void *data;
 166 };
 167
 168 #endif