uct/policy/generic.h

   1 #ifndef ZZGO_UCT_POLICY_GENERIC_H
   2 #define ZZGO_UCT_POLICY_GENERIC_H
   3
   4 /* Some default policy routines and templates. */
   5
   6 #include "stone.h"
   7 #include "uct/internal.h"
   8
   9 struct board;
  10 struct tree_node;
  11
  12 struct tree_node *uctp_generic_choose(struct uct_policy *p, struct tree_node *node, struct board *b, enum stone color);
  13 struct tree_node *uctp_generic_winner(struct uct_policy *p, struct tree *tree, struct tree_node *node);
  14
  15
  16 /* Some generic stitching for tree descent. */
  17
  18 #define uctd_try_node_children(node, allow_pass, ni, urgency) \
  19         /* XXX: Stack overflow danger on big boards? */ \
  20         struct tree_node *nbest[512] = { node->children }; int nbests = 1; \
  21         float best_urgency = -9999; \
  22         \
  23         for (struct tree_node *ni = node->children; ni; ni = ni->sibling) { \
  24                 float urgency; \
  25                 /* Do not consider passing early. */ \
  26                 if (unlikely((!allow_pass && is_pass(ni->coord)) || (ni->hints & TREE_HINT_INVALID))) \
  27                         continue;
  28
  29                 /* ...your urgency computation code goes here... */
  30
  31 #define uctd_set_best_child(ni, urgency) \
  32                 if (urgency - best_urgency > __FLT_EPSILON__) { /* urgency > best_urgency */ \
  33                         best_urgency = urgency; nbests = 0; \
  34                 } \
  35                 if (urgency - best_urgency > -__FLT_EPSILON__) { /* urgency >= best_urgency */ \
  36                         /* We want to always choose something else than a pass \
  37                          * in case of a tie. pass causes degenerative behaviour. */ \
  38                         if (nbests == 1 && is_pass(nbest[0]->coord)) { \
  39                                 nbests--; \
  40                         } \
  41                         nbest[nbests++] = ni; \
  42                 } \
  43         }
  44
  45 #define uctd_get_best_child() nbest[fast_random(nbests)]
  46
  47
  48 #endif