From 62809ce7538f0c0ec69fb0a3be623653709f08ea Mon Sep 17 00:00:00 2001
From: Tue Herlau <tuhe@dtu.dk>
Date: Sun, 9 Feb 2025 21:15:19 +0100
Subject: [PATCH] Solutions to exercises
---
solutions/ex00/fruit_homework_TODO_1.py | 1 +
solutions/ex00/fruit_homework_TODO_2.py | 1 +
solutions/ex00/fruit_homework_TODO_3.py | 1 +
solutions/ex00/fruit_homework_TODO_4.py | 1 +
solutions/ex01/bobs_friend_TODO_1.py | 3 +++
solutions/ex01/bobs_friend_TODO_2.py | 9 +++++++++
solutions/ex01/bobs_friend_TODO_3.py | 1 +
solutions/ex01/bobs_friend_TODO_4.py | 1 +
solutions/ex01/chess_TODO_1.py | 1 +
solutions/ex01/chess_TODO_2.py | 7 +++++++
solutions/ex01/chess_TODO_3.py | 1 +
solutions/ex01/chess_TODO_4.py | 1 +
solutions/ex01/chess_TODO_5.py | 1 +
solutions/ex01/inventory_environment_TODO_1.py | 5 +++++
solutions/ex01/inventory_environment_TODO_2.py | 1 +
solutions/ex01/inventory_environment_TODO_3.py | 7 +++++++
solutions/ex01/pacman_hardcoded_TODO_1.py | 7 +++++++
17 files changed, 49 insertions(+)
create mode 100644 solutions/ex00/fruit_homework_TODO_1.py
create mode 100644 solutions/ex00/fruit_homework_TODO_2.py
create mode 100644 solutions/ex00/fruit_homework_TODO_3.py
create mode 100644 solutions/ex00/fruit_homework_TODO_4.py
create mode 100644 solutions/ex01/bobs_friend_TODO_1.py
create mode 100644 solutions/ex01/bobs_friend_TODO_2.py
create mode 100644 solutions/ex01/bobs_friend_TODO_3.py
create mode 100644 solutions/ex01/bobs_friend_TODO_4.py
create mode 100644 solutions/ex01/chess_TODO_1.py
create mode 100644 solutions/ex01/chess_TODO_2.py
create mode 100644 solutions/ex01/chess_TODO_3.py
create mode 100644 solutions/ex01/chess_TODO_4.py
create mode 100644 solutions/ex01/chess_TODO_5.py
create mode 100644 solutions/ex01/inventory_environment_TODO_1.py
create mode 100644 solutions/ex01/inventory_environment_TODO_2.py
create mode 100644 solutions/ex01/inventory_environment_TODO_3.py
create mode 100644 solutions/ex01/pacman_hardcoded_TODO_1.py
diff --git a/solutions/ex00/fruit_homework_TODO_1.py b/solutions/ex00/fruit_homework_TODO_1.py
new file mode 100644
index 0000000..b498ceb
--- /dev/null
+++ b/solutions/ex00/fruit_homework_TODO_1.py
@@ -0,0 +1 @@
+ return a+b
\ No newline at end of file
diff --git a/solutions/ex00/fruit_homework_TODO_2.py b/solutions/ex00/fruit_homework_TODO_2.py
new file mode 100644
index 0000000..f546843
--- /dev/null
+++ b/solutions/ex00/fruit_homework_TODO_2.py
@@ -0,0 +1 @@
+ return ["mr " + a for a in animals]
\ No newline at end of file
diff --git a/solutions/ex00/fruit_homework_TODO_3.py b/solutions/ex00/fruit_homework_TODO_3.py
new file mode 100644
index 0000000..5be72c6
--- /dev/null
+++ b/solutions/ex00/fruit_homework_TODO_3.py
@@ -0,0 +1 @@
+ return sum([x * p for x, p in p_dict.items()])
\ No newline at end of file
diff --git a/solutions/ex00/fruit_homework_TODO_4.py b/solutions/ex00/fruit_homework_TODO_4.py
new file mode 100644
index 0000000..84c3b39
--- /dev/null
+++ b/solutions/ex00/fruit_homework_TODO_4.py
@@ -0,0 +1 @@
+ return self.prices[fruit]
\ No newline at end of file
diff --git a/solutions/ex01/bobs_friend_TODO_1.py b/solutions/ex01/bobs_friend_TODO_1.py
new file mode 100644
index 0000000..2d03d7c
--- /dev/null
+++ b/solutions/ex01/bobs_friend_TODO_1.py
@@ -0,0 +1,3 @@
+
+ self.s = self.x0
+
\ No newline at end of file
diff --git a/solutions/ex01/bobs_friend_TODO_2.py b/solutions/ex01/bobs_friend_TODO_2.py
new file mode 100644
index 0000000..9caf28a
--- /dev/null
+++ b/solutions/ex01/bobs_friend_TODO_2.py
@@ -0,0 +1,9 @@
+ terminated = True
+ if a == 0:
+ s_next = self.s * 1.1
+ else:
+ if np.random.rand() < 1/4:
+ s_next = 0
+ else:
+ s_next = self.s + 12
+ reward = s_next - self.s
\ No newline at end of file
diff --git a/solutions/ex01/bobs_friend_TODO_3.py b/solutions/ex01/bobs_friend_TODO_3.py
new file mode 100644
index 0000000..8399f7f
--- /dev/null
+++ b/solutions/ex01/bobs_friend_TODO_3.py
@@ -0,0 +1 @@
+ return 0
\ No newline at end of file
diff --git a/solutions/ex01/bobs_friend_TODO_4.py b/solutions/ex01/bobs_friend_TODO_4.py
new file mode 100644
index 0000000..36a268f
--- /dev/null
+++ b/solutions/ex01/bobs_friend_TODO_4.py
@@ -0,0 +1 @@
+ return 1
\ No newline at end of file
diff --git a/solutions/ex01/chess_TODO_1.py b/solutions/ex01/chess_TODO_1.py
new file mode 100644
index 0000000..f8752f9
--- /dev/null
+++ b/solutions/ex01/chess_TODO_1.py
@@ -0,0 +1 @@
+ self.s = []
\ No newline at end of file
diff --git a/solutions/ex01/chess_TODO_2.py b/solutions/ex01/chess_TODO_2.py
new file mode 100644
index 0000000..9b82990
--- /dev/null
+++ b/solutions/ex01/chess_TODO_2.py
@@ -0,0 +1,7 @@
+ if np.random.rand() < self.p_draw:
+ game_outcome = 0
+ else:
+ if np.random.rand() < self.p_win:
+ game_outcome = 1
+ else:
+ game_outcome = -1
\ No newline at end of file
diff --git a/solutions/ex01/chess_TODO_3.py b/solutions/ex01/chess_TODO_3.py
new file mode 100644
index 0000000..29e1443
--- /dev/null
+++ b/solutions/ex01/chess_TODO_3.py
@@ -0,0 +1 @@
+ done = len(self.s) >= 2 and self.s[-1] == self.s[-2] and self.s[-1] != 0
\ No newline at end of file
diff --git a/solutions/ex01/chess_TODO_4.py b/solutions/ex01/chess_TODO_4.py
new file mode 100644
index 0000000..d45e38a
--- /dev/null
+++ b/solutions/ex01/chess_TODO_4.py
@@ -0,0 +1 @@
+ r = self.s[-1] == 1 if done else 0
\ No newline at end of file
diff --git a/solutions/ex01/chess_TODO_5.py b/solutions/ex01/chess_TODO_5.py
new file mode 100644
index 0000000..c270359
--- /dev/null
+++ b/solutions/ex01/chess_TODO_5.py
@@ -0,0 +1 @@
+ stats, _ = train(env, Agent(env), num_episodes=T)
\ No newline at end of file
diff --git a/solutions/ex01/inventory_environment_TODO_1.py b/solutions/ex01/inventory_environment_TODO_1.py
new file mode 100644
index 0000000..5f5a775
--- /dev/null
+++ b/solutions/ex01/inventory_environment_TODO_1.py
@@ -0,0 +1,5 @@
+ s_next = max(0, min(2, self.s-w+a)) # next state; x_{k+1} = f_k(x_k, u_k, w_k)
+ reward = -(a + (self.s + a - w)**2) # reward = -cost = -g_k(x_k, u_k, w_k)
+ terminated = self.k == self.N-1 # Have we terminated? (i.e. is k==N-1)
+ self.s = s_next # update environment state
+ self.k += 1 # update current time step
\ No newline at end of file
diff --git a/solutions/ex01/inventory_environment_TODO_2.py b/solutions/ex01/inventory_environment_TODO_2.py
new file mode 100644
index 0000000..bebe04b
--- /dev/null
+++ b/solutions/ex01/inventory_environment_TODO_2.py
@@ -0,0 +1 @@
+ return np.random.choice(3) # Return a random action
\ No newline at end of file
diff --git a/solutions/ex01/inventory_environment_TODO_3.py b/solutions/ex01/inventory_environment_TODO_3.py
new file mode 100644
index 0000000..0855951
--- /dev/null
+++ b/solutions/ex01/inventory_environment_TODO_3.py
@@ -0,0 +1,7 @@
+ a = agent.pi(s, k)
+ sp, r, terminated, truncated, metadata = env.step(a)
+ agent.train(s, a, sp, r, terminated)
+ s = sp
+ J += r
+ if terminated or truncated:
+ break
\ No newline at end of file
diff --git a/solutions/ex01/pacman_hardcoded_TODO_1.py b/solutions/ex01/pacman_hardcoded_TODO_1.py
new file mode 100644
index 0000000..5c532d7
--- /dev/null
+++ b/solutions/ex01/pacman_hardcoded_TODO_1.py
@@ -0,0 +1,7 @@
+ if k < 7:
+ return 'South'
+ elif k < 14:
+ return 'East'
+ elif k < 21:
+ return 'North'
+ elif k < 28:
\ No newline at end of file
--
GitLab