From 88addb1cf5c6ea3e1a2013836ac2a04439661d5e Mon Sep 17 00:00:00 2001
From: Josh Creek <8179928+jcreek@users.noreply.github.com>
Date: Thu, 12 Oct 2023 21:27:13 +0100
Subject: [PATCH] feat(*): Update at 1 billion timesteps

---
 README.md      | 10 ++++++++--
 rl-training.py |  4 ++--
 2 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/README.md b/README.md
index 5eabcd5..d007b2d 100644
--- a/README.md
+++ b/README.md
@@ -80,7 +80,7 @@ This is to try to get the bot to gain experience in other areas of the game now
 
 #### ~500 million steps
 
-After 528 million steps I changed things again, switching to 1v1 instead of 3v3, and changing the rewards as below: 
+After 528 million steps I changed things again, changing the rewards as below: 
 
 - VelocityBallToGoalReward -> 10
 - EventReward -> 10
@@ -90,7 +90,13 @@ After 528 million steps I changed things again, switching to 1v1 instead of 3v3,
 
 #### ~1 billion steps
 
-The bot should now have acceptable mechanics and a vague understanding of 1v1 strategies.
+The bot should now have acceptable mechanics and a vague understanding of 3v3 strategies.
+
+However, I noticed that there's a lot of dead time, where it's not necessarily driving towards the ball or preparing to defend or receive a pass, so I changed things again, changing the rewards as below: 
+
+- VelocityPlayerToBallReward -> 50
+- LiuDistanceBallToGoalReward -> 20
+- RewardIfClosestToBall -> 10
 
 ### Stage 2
 
diff --git a/rl-training.py b/rl-training.py
index 764986f..aad391b 100644
--- a/rl-training.py
+++ b/rl-training.py
@@ -61,7 +61,7 @@ if __name__ == "__main__":  # Required for multiprocessing
     def exit_save(model):
         model.save("models/exit_save")
 
-    team_size = 1
+    team_size = 3
 
     def get_match():  # Need to use a function so that each instance can call it and produce their own objects
         return Match(
@@ -87,7 +87,7 @@ if __name__ == "__main__":  # Required for multiprocessing
                     AlignBallGoal(),
                     KickoffReward(),
                 ),
-                (1.0, 10.0, 50.0, 10.0, 0.1, 1.0, 1.0, 0.2, 10.0, 20.0, 10.0),
+                (50.0, 10.0, 20.0, 10.0, 0.1, 10.0, 1.0, 0.2, 10.0, 20.0, 10.0),
             ),
             spawn_opponents=True,
             terminal_conditions=[