mia/nim
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Implementiere update_q_value

Browse source
This commit is contained in:
mia 2026-04-23 09:39:16 +02:00
parent a74be2fc31
commit 933d0c5d27
3 changed files with 14 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

BIN
__pycache__/nim.cpython-313.pyc
View file

Binary file not shown.

12
nim.py
View file

@ -31,10 +31,16 @@ class Nim():
class NimAI(): class NimAI():
def __init__(self, alpha=0.5, epsilon=0.1): def __init__(self, alpha=0.5, epsilon=1):
self.q = dict() # Q-value table self.q = dict() # Q-value table
self.q[(0, 0, 0, 2), (3, 2)] = -1 # Test Q-Value # self.q[(0, 0, 0, 2), (3, 2)] = -1 # Test Q-Value
self.q[(0, 0, 0, 2), (3, 1)] = 10 # Test Q-Value # self.q[(0, 0, 0, 2), (3, 1)] = 10 # Test Q-Value
self.q[((1,1,1,0), (0,1))] = 0.4
self.q[((1,1,1,0), (1,1))] = 0.9
self.q[((1,1,1,0), (2,1))] = 0.7
self.q[((2,1,1,0), (0,1))] = 0.2
self.alpha = alpha # Learning rate self.alpha = alpha # Learning rate
self.epsilon = epsilon # Exploration rate self.epsilon = epsilon # Exploration rate

5
test.py
View file

@ -10,6 +10,11 @@ def test_get_q_value(ai):
def test_update_q_value(ai): def test_update_q_value(ai):
print("\n--- Testing update_q_value ---") print("\n--- Testing update_q_value ---")
state = (2, 1, 1, 0)
action = (0, 1)
print(ai.q)
print(ai.update_q_value([2, 1, 1, 0], (0, 1), 0.2, 1, 0.8))
print(ai.q)
def test_best_future_reward(ai): def test_best_future_reward(ai):