From 2328b9f4ceea5c96065b2439b775908aaf3bcf2e Mon Sep 17 00:00:00 2001
From: Petr Baudis <pasky@ucw.cz>
Date: Fri, 19 Mar 2010 15:05:37 +0100
Subject: [PATCH] UCT dynkomi: Also track value in resettable movestats

---
 uct/dynkomi.c | 15 +++++++++------
 uct/tree.h    | 12 ++++++------
 uct/walk.c    |  1 +
 3 files changed, 16 insertions(+), 12 deletions(-)

diff --git a/uct/dynkomi.c b/uct/dynkomi.c
index c7e5663..007dccb 100644
--- a/uct/dynkomi.c
+++ b/uct/dynkomi.c
@@ -230,9 +230,13 @@ komi_by_score(struct dynkomi_adaptive *a, struct board *b, struct tree *tree)
 static float
 komi_by_value(struct dynkomi_adaptive *a, struct board *b, struct tree *tree)
 {
-	if (tree->root->u.playouts < TRUSTWORTHY_KOMI_PLAYOUTS)
+	if (tree->value.playouts < TRUSTWORTHY_KOMI_PLAYOUTS)
 		return tree->extra_komi;
 
+	struct move_stats value = tree->value;
+	/* Almost-reset tree->value to gather fresh stats. */
+	tree->value.playouts = 1;
+
 	/* We have three "value zones":
 	 * red zone | yellow zone | green zone
 	 *        ~45%           ~60%
@@ -249,19 +253,18 @@ komi_by_value(struct dynkomi_adaptive *a, struct board *b, struct tree *tree)
 	 * to try to reduce extra komi we take.
 	 *
 	 * TODO: Make the latch expire after a while. */
-	float value = tree->root->u.value;
 	float extra_komi = tree->extra_komi;
 
-	if (value < a->zone_red) {
+	if (value.value < a->zone_red) {
 		/* Red zone. Take extra komi. */
 		if (DEBUGL(3))
 			fprintf(stderr, "[red] %f, komi latch %f -> %f\n",
-				value, a->komi_latch, extra_komi);
+				value.value, a->komi_latch, extra_komi);
 		if (extra_komi > 0) a->komi_latch = extra_komi;
 		extra_komi -= a->score_step; // XXX: we depend on being black
 		return extra_komi;
 
-	} else if (value < a->zone_green) {
+	} else if (value.value < a->zone_green) {
 		/* Yellow zone, do nothing. */
 		return extra_komi;
 
@@ -270,7 +273,7 @@ komi_by_value(struct dynkomi_adaptive *a, struct board *b, struct tree *tree)
 		extra_komi += a->score_step; // XXX: we depend on being black
 		if (DEBUGL(3))
 			fprintf(stderr, "[green] %f, += %d | komi latch %f\n",
-				value, a->score_step, a->komi_latch);
+				value.value, a->score_step, a->komi_latch);
 		return !a->use_komi_latch || extra_komi < a->komi_latch ? extra_komi : a->komi_latch - 1;
 	}
 }
diff --git a/uct/tree.h b/uct/tree.h
index 78acaaa..2ba9edf 100644
--- a/uct/tree.h
+++ b/uct/tree.h
@@ -95,13 +95,13 @@ struct tree {
 	 * based on leaf node depth. */
 	float extra_komi;
 
-	/* Information on average score at the game end (from black's
-	 * perspective). */
-	/* N.B.: Some users may need average score carried over from previous
-	 * move to initialize. Thus, this information is NOT zeroed on node
-	 * promotion, you are responsible for zeroing it for yourself if you
-	 * use it. */
+	/*** Only for dynkomi use: */
+	/* Information on average score at the simulation end (black's
+	 * perspective) since last dynkomi adjustment. */
 	struct move_stats score;
+	/* Information on average winrate of simulations since last
+	 * dynkomi adjustment. */
+	struct move_stats value;
 
 	/* We merge local (non-tenuki) sequences for both colors, occuring
 	 * anywhere in the tree; nodes are created on-demand, special 'pass'
diff --git a/uct/walk.c b/uct/walk.c
index a68bdf0..2e6a15f 100644
--- a/uct/walk.c
+++ b/uct/walk.c
@@ -438,6 +438,7 @@ uct_playout(struct uct *u, struct board *b, enum stone player_color, struct tree
 		stats_add_result(&t->score, result / 2, 1);
 
 		float rval = scale_value(u, b, result);
+		stats_add_result(&t->value, rval, 1);
 		u->policy->update(u->policy, t, n, node_color, player_color, amaf, rval);
 
 		if (u->local_tree && n->parent && !is_pass(n->coord) && dlen > 0) {
-- 
2.11.4.GIT