From 595bee95c4951f4bd582b39da6c4de08894b9b5b Mon Sep 17 00:00:00 2001
From: LeoQuentin <leo.quentin.thorbjornsen.bakholt@nmbu.no>
Date: Sat, 10 Feb 2024 18:22:28 +0100
Subject: [PATCH] deleted old files

---
 reinforcement_learning/.DS_Store              | Bin 6148 -> 0 bytes
 .../Q_Learning/Frozen_Lake_Q.py               |  41 ------------------
 2 files changed, 41 deletions(-)
 delete mode 100644 reinforcement_learning/.DS_Store
 delete mode 100644 reinforcement_learning/Q_Learning/Frozen_Lake_Q.py

diff --git a/reinforcement_learning/.DS_Store b/reinforcement_learning/.DS_Store
deleted file mode 100644
index 329ab26df6fa89bd6d8a98cac00a7b0c5b445363..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 6148
zcmeHKO-sW-5S?wSO%<UB1-&hJE!tYGh?h|7LGYpxJ*d>g77fO1>4!a(LhkxsJo`ud
zJ36zw6{Ys5Seb#@H`$$+Y+jb#3;>8~uTcWX1Av20n95=EjnO#yn5Aq_8d2zb_|S$J
zLiC%lZ0`7r3edNkgCnTJAvEFB_dAHxbr7il$7yjqqp!Vb!)O90=yzI{^Afb8u+uFT
zUqvdNnVHQxS!dq4^lnw(>-ybZr|P%Q>FPwOC>X_la2huI_59+Vio1RoH`|&JHkz1n
zaTdl6)vu~v+~{boX9k>{ldI=f27^*%yIhdv($=sb2Rj?Zg4`@uhQpk*w7R}~T)&I%
zWA&(~4+0mgwgrnLc*5XYK_0wz9I2QV_J}*;FtP%y04p$M3fOJTnV+(G^K@AOR^aCq
zp#4Fj6S@`?gL>=0MwbAH8SK`EG5szQM_P0(CI)c?O&BVop$c1K2t&tyY4co*i9tgL
zVT%u8pDb*JBJ|Vo{!*uda1C<H3a|oS6`<M=El=nF{^a^!Cvn3Humb;60g<VBwJN@w
zJzK}#lFnL*euYj(d5OUf3I_Tp##}mzm(jIhzoY=swU`*h9yI<D&@ynt3jC@9?^k47
AXaE2J

diff --git a/reinforcement_learning/Q_Learning/Frozen_Lake_Q.py b/reinforcement_learning/Q_Learning/Frozen_Lake_Q.py
deleted file mode 100644
index fa40b2f..0000000
--- a/reinforcement_learning/Q_Learning/Frozen_Lake_Q.py
+++ /dev/null
@@ -1,41 +0,0 @@
-import gymnasium as gym
-import pandas as pd
-import numpy as np
-import matplotlib.pyplot as plt
-
-
-class QLearningAgent:
-    def __init__(self, state_space, action_space, lr,
-                 gamma, epsilon, epsilon_decay, epsilon_minimum):
-        self.state_space = state_space
-        self.action_space = action_space
-        self.lr = lr
-        self.gamma = gamma
-        self.epsilon = epsilon
-        self.epsilon_decay = epsilon_decay
-        self.epsilon_minimum = epsilon_minimum
-        self.Q_table = pd.DataFrame(data=np.zeros(shape=(action_space, state_space)),
-                                    columns=[f"State {i}" for i in range(state_space)],
-                                    index=[f"Action {i}" for i in range(action_space)])
-
-    def greedy_epsilon_action_choice(self, state):
-        if np.random.random() > self.epsilon:
-            action = self.greedy_action_choice(state)
-        else:
-            action = np.random.choice(range(self.action_space))
-        self.decrement_epsilon()
-        return action
-
-    def decrement_epsilon(self):
-        self.epsilon = max(self.epsilon_minimum, self.epsilon - self.epsilon_decay)
-
-    def greedy_action_choice(self, state):
-        action = self.Q_table.iloc[:, state].idxmax()
-        return int(action.split(" ")[-1])
-
-    def learn(self, old_state, action, reward, new_state):
-        Q_observed = reward + self.gamma * self.Q_table.iloc[:, new_state].max()
-        Q_expected = self.Q_table.at[f"Action {action}", f"State {old_state}"]
-        self.Q_table.at[f"Action {action}", f"State {old_state}"] = Q_expected + self.lr * (Q_observed - Q_expected)
-
-