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, coord_t exclude);
  13 void uctp_generic_winner(struct uct_policy *p, struct tree *tree, struct uct_descent *descent);
  14
  15
  16 /* Some generic stitching for tree descent. */
  17
  18 #define uctd_try_node_children(tree, descent, allow_pass, parity, tenuki_d, di, urgency) \
  19         /* Information abound best children. */ \
  20         /* XXX: Stack overflow danger on big boards? */ \
  21         struct uct_descent dbest[512] = { { .node = descent->node->children, .lnode = NULL } }; int dbests = 1; \
  22         float best_urgency = -9999; \
  23         /* Descent children iterator. */ \
  24         struct uct_descent dci = { .node = descent->node->children, .lnode = descent->lnode ? descent->lnode->children : NULL }; \
  25         \
  26         for (; dci.node; dci.node = dci.node->sibling) { \
  27                 float urgency; \
  28                 /* Do not consider passing early. */ \
  29                 if (unlikely((!allow_pass && is_pass(dci.node->coord)) || (dci.node->hints & TREE_HINT_INVALID))) \
  30                         continue; \
  31                 /* Position dci.lnode to point at or right after the local
  32                  * node corresponding to dci.node. */ \
  33                 while (dci.lnode && dci.lnode->coord < dci.node->coord) \
  34                         dci.lnode = dci.lnode->sibling; \
  35                 /* Set up descent-further iterator. This is the public-accessible
  36                  * one, and usually is similar to dci. However, in case of local
  37                  * trees, we may keep next-candidate pointer in dci while storing
  38                  * actual-specimen in di. */ \
  39                 struct uct_descent di = dci; \
  40                 if (dci.lnode) { \
  41                         /* Set lnode to local tree node corresponding
  42                          * to node (dci.lnode, pass-lnode or NULL). */ \
  43                         di.lnode = tree_lnode_for_node(tree, dci.node, dci.lnode, tenuki_d); \
  44                 }
  45
  46                 /* ...your urgency computation code goes here... */
  47
  48 #define uctd_set_best_child(di, urgency) \
  49                 if (urgency - best_urgency > __FLT_EPSILON__) { /* urgency > best_urgency */ \
  50                         best_urgency = urgency; dbests = 0; \
  51                 } \
  52                 if (urgency - best_urgency > -__FLT_EPSILON__) { /* urgency >= best_urgency */ \
  53                         /* We want to always choose something else than a pass \
  54                          * in case of a tie. pass causes degenerative behaviour. */ \
  55                         if (dbests == 1 && is_pass(dbest[0].node->coord)) { \
  56                                 dbests--; \
  57                         } \
  58                         struct uct_descent db = di; \
  59                         /* Make sure lnode information is meaningful. */ \
  60                         if (db.lnode && is_pass(db.lnode->coord)) \
  61                                 db.lnode = NULL; \
  62                         dbest[dbests++] = db; \
  63                 } \
  64         }
  65
  66 #define uctd_get_best_child(descent) *(descent) = dbest[fast_random(dbests)];
  67
  68
  69 #endif