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 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         enum uct_reporting {
  30                 UR_TEXT,
  31                 UR_JSON,
  32                 UR_JSON_BIG,
  33         } reporting;
  34         int reportfreq;
  35
  36         int games, gamelen;
  37         floating_t resign_threshold, sure_win_threshold;
  38         double best2_ratio, bestr_ratio;
  39         floating_t max_maintime_ratio;
  40         bool pass_all_alive; /* Current value */
  41         bool allow_losing_pass;
  42         bool territory_scoring;
  43         int expand_p;
  44         bool playout_amaf;
  45         bool amaf_prior;
  46         int playout_amaf_cutoff;
  47         double dumpthres;
  48         int force_seed;
  49         bool no_tbook;
  50         bool fast_alloc;
  51         unsigned long max_tree_size;
  52         unsigned long max_pruned_size;
  53         unsigned long pruning_threshold;
  54         int mercymin;
  55         int significant_threshold;
  56
  57         int threads;
  58         enum uct_thread_model {
  59                 TM_TREE, /* Tree parallelization w/o virtual loss. */
  60                 TM_TREEVL, /* Tree parallelization with virtual loss. */
  61         } thread_model;
  62         int virtual_loss;
  63         bool pondering_opt; /* User wants pondering */
  64         bool pondering; /* Actually pondering now */
  65         bool slave; /* Act as slave in distributed engine. */
  66         int max_slaves; /* Optional, -1 if not set */
  67         int slave_index; /* 0..max_slaves-1, or -1 if not set */
  68         enum stone my_color;
  69
  70         int fuseki_end;
  71         int yose_start;
  72
  73         int dynkomi_mask;
  74         int dynkomi_interval;
  75         struct uct_dynkomi *dynkomi;
  76         floating_t initial_extra_komi;
  77
  78         floating_t val_scale;
  79         int val_points;
  80         bool val_extra;
  81         bool val_byavg;
  82         bool val_bytemp;
  83         floating_t val_bytemp_min;
  84
  85         int random_policy_chance;
  86         bool local_tree;
  87         int tenuki_d;
  88         floating_t local_tree_aging;
  89 #define LTREE_PLAYOUTS_MULTIPLIER 100
  90         floating_t local_tree_depth_decay;
  91         bool local_tree_allseq;
  92         bool local_tree_neival;
  93         enum {
  94                 LTE_ROOT,
  95                 LTE_EACH,
  96                 LTE_TOTAL,
  97         } local_tree_eval;
  98         bool local_tree_rootchoose;
  99
 100         struct {
 101                 int level;
 102                 int playouts;
 103         } debug_after;
 104
 105         char *banner;
 106
 107         struct uct_policy *policy;
 108         struct uct_policy *random_policy;
 109         struct playout_policy *playout;
 110         struct uct_prior *prior;
 111         struct uct_pluginset *plugins;
 112         struct joseki_dict *jdict;
 113
 114         struct pattern_setup pat;
 115         /* Various modules (prior, policy, ...) set this if they want pattern
 116          * database to be loaded. */
 117         bool want_pat;
 118
 119         /* Used within frame of single genmove. */
 120         struct board_ownermap ownermap;
 121         /* Used for coordination among slaves of the distributed engine. */
 122         int stats_hbits;
 123         int shared_nodes;
 124         int shared_levels;
 125         double stats_delay; /* stored in seconds */
 126         int played_own;
 127         int played_all; /* games played by all slaves */
 128
 129         /* Game state - maintained by setup_state(), reset_state(). */
 130         struct tree *t;
 131 };
 132
 133 #define UDEBUGL(n) DEBUGL_(u->debug_level, n)
 134
 135 bool uct_pass_is_safe(struct uct *u, struct board *b, enum stone color, bool pass_all_alive);
 136
 137 void uct_prepare_move(struct uct *u, struct board *b, enum stone color);
 138 void uct_genmove_setup(struct uct *u, struct board *b, enum stone color);
 139 void uct_pondering_stop(struct uct *u);
 140
 141
 142 /* This is the state used for descending the tree; we use this wrapper
 143  * structure in order to be able to easily descend in multiple trees
 144  * in parallel (e.g. main tree and local tree) or compute cummulative
 145  * "path value" throughout the tree descent. */
 146 struct uct_descent {
 147         /* Active tree nodes: */
 148         struct tree_node *node; /* Main tree. */
 149         struct tree_node *lnode; /* Local tree. */
 150         /* Value of main tree node (with all value factors, but unbiased
 151          * - without exploration factor), from black's perspective. */
 152         struct move_stats value;
 153 };
 154
 155
 156 typedef struct tree_node *(*uctp_choose)(struct uct_policy *p, struct tree_node *node, struct board *b, enum stone color, coord_t exclude);
 157 typedef floating_t (*uctp_evaluate)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity);
 158 typedef void (*uctp_descend)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent, int parity, bool allow_pass);
 159 typedef void (*uctp_winner)(struct uct_policy *p, struct tree *tree, struct uct_descent *descent);
 160 typedef void (*uctp_prior)(struct uct_policy *p, struct tree *tree, struct tree_node *node, struct board *b, enum stone color, int parity);
 161 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);
 162 typedef void (*uctp_done)(struct uct_policy *p);
 163
 164 struct uct_policy {
 165         struct uct *uct;
 166         uctp_choose choose;
 167         uctp_winner winner;
 168         uctp_evaluate evaluate;
 169         uctp_descend descend;
 170         uctp_update update;
 171         uctp_prior prior;
 172         uctp_done done;
 173         bool wants_amaf;
 174         void *data;
 175 };
 176
 177 #endif