diff --git a/.DS_Store b/.DS_Store
new file mode 100644
index 00000000..007b67b1
Binary files /dev/null and b/.DS_Store differ
diff --git a/.gitignore b/.gitignore
new file mode 100644
index 00000000..afed0735
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1 @@
+*.csv
diff --git a/Labs/.DS_Store b/Labs/.DS_Store
new file mode 100644
index 00000000..b6430a22
Binary files /dev/null and b/Labs/.DS_Store differ
diff --git a/Labs/Lab.2/.DS_Store b/Labs/Lab.2/.DS_Store
new file mode 100644
index 00000000..5008ddfc
Binary files /dev/null and b/Labs/Lab.2/.DS_Store differ
diff --git a/Labs/Lab.2/Lab.2.ipynb b/Labs/Lab.2/Lab.2.ipynb
index f0b38e38..fb55044d 100644
--- a/Labs/Lab.2/Lab.2.ipynb
+++ b/Labs/Lab.2/Lab.2.ipynb
@@ -84,20 +84,41 @@
},
{
"cell_type": "code",
- "execution_count": 1,
+ "execution_count": 83,
"metadata": {},
"outputs": [],
"source": [
- "# Write your solution here"
+ "# Write your solution here\n",
+ "def make_board(n):\n",
+ " empty = 0\n",
+ " X = 1\n",
+ " O = 2\n",
+ " board = list()\n",
+ " for i in range(n):\n",
+ " row = list()\n",
+ " for j in range(n):\n",
+ " row.append(1)\n",
+ " board.append(row)\n",
+ " return board"
]
},
{
"cell_type": "code",
- "execution_count": 2,
+ "execution_count": 6,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "[[1, 1, 1], [1, 1, 1], [1, 1, 1]]\n"
+ ]
+ }
+ ],
"source": [
- "# Test your solution here"
+ "# Test your solution here\n",
+ "board = make_board(3)\n",
+ "print(board)"
]
},
{
@@ -107,7 +128,96 @@
"outputs": [],
"source": [
"# (Optional) Ask an LLM for 3 different solutions here\n",
- "# Then compare them to your own."
+ "#Solution 1\n",
+ "def make_board_1(n):\n",
+ " board = []\n",
+ " for i in range(n):\n",
+ " row = []\n",
+ " for j in range(n):\n",
+ " row.append(0)\n",
+ " board.append(row)\n",
+ " return board\n",
+ "\n",
+ "#solution 2\n",
+ "def make_board_2(n):\n",
+ " return [[0 for _ in range(n)] for _ in range(n)]\n",
+ "\n",
+ "#solution 3\n",
+ "def make_board_3(n):\n",
+ " board = []\n",
+ " for _ in range(n):\n",
+ " board.append([0] * n)\n",
+ " return board"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "
For Solution 1
\n",
+ "\n",
+ "
Same:
\n",
+ "\n",
+ "Both use nested loops\n",
+ "\n",
+ "Both construct the board row by row\n",
+ "\n",
+ "Both use append() to add values\n",
+ "\n",
+ "Both return an n × n list of lists\n",
+ "\n",
+ "Both are easy to understand and beginner-friendly\n",
+ "\n",
+ "\n",
+ "
Different:
\n",
+ "\n",
+ "Your solution defines empty, X, and O, while Solution 1 directly uses 0\n",
+ "\n",
+ "Your solution documents the meaning of values more clearly\n",
+ "\n",
+ "Solution 1 is slightly shorter\n",
+ "\n",
+ "\n",
+ "
For Solution 2
\n",
+ "\n",
+ "
Same:
\n",
+ "\n",
+ "Both return an n × n Tic Tac Toe board\n",
+ "\n",
+ "Both use 0 to represent empty spaces\n",
+ "\n",
+ "Both solve the same problem correctly\n",
+ "\n",
+ "\n",
+ "
Different:
\n",
+ "\n",
+ "Your solution uses explicit loops, Solution 2 uses list comprehension\n",
+ "\n",
+ "Your solution is more readable for beginners\n",
+ "\n",
+ "Solution 2 is more compact but less transparent\n",
+ "\n",
+ "Your solution clearly shows how the matrix is built step by step\n",
+ "\n",
+ "\n",
+ "
For Solution 3
\n",
+ "\n",
+ "
Same:
\n",
+ "\n",
+ "Both create each row separately\n",
+ "\n",
+ "Both store rows in a board list\n",
+ "\n",
+ "Both correctly initialize all cells as empty (0)\n",
+ "\n",
+ "\n",
+ "
Different:
\n",
+ "\n",
+ "Your solution fills rows using an inner loop, Solution 3 uses [0] * n\n",
+ "\n",
+ "Your approach is more detailed and instructional\n",
+ "\n",
+ "Solution 3 is more concise and slightly more efficient\n"
]
},
{
@@ -117,6 +227,19 @@
"**Question:** Which solution most closely matches your solution? What are the main differences?"
]
},
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "
Solution 1 is the closest match because:
\n",
+ "\n",
+ "It follows the same structure\n",
+ "\n",
+ "Uses the same logic\n",
+ "\n",
+ "Builds the matrix step-by-step just like my implementation"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -138,20 +261,45 @@
},
{
"cell_type": "code",
- "execution_count": 4,
+ "execution_count": 84,
"metadata": {},
"outputs": [],
"source": [
- "# Write your solution here"
+ "# Write your solution \n",
+ "def draw_board(r,c):\n",
+ " for i in range(r):\n",
+ " row = \"\"\n",
+ " empty = \" \"\n",
+ " print(\" --- \" *c)\n",
+ " for j in range(c):\n",
+ " row += \"| \" + empty + \" |\"\n",
+ " \n",
+ " print(row)\n",
+ " print(\" --- \" *c)"
]
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": 86,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " --- --- --- \n",
+ "| || || |\n",
+ " --- --- --- \n",
+ "| || || |\n",
+ " --- --- --- \n",
+ "| || || |\n",
+ " --- --- --- \n"
+ ]
+ }
+ ],
"source": [
- "# Test your solution here"
+ "# Test your solution \n",
+ "draw_board(3,3)"
]
},
{
@@ -163,20 +311,48 @@
},
{
"cell_type": "code",
- "execution_count": 6,
+ "execution_count": 88,
"metadata": {},
"outputs": [],
"source": [
- "# Write your solution here"
+ "# Write your solution \n",
+ "def draw_board(board):\n",
+ " n = len(board)\n",
+ " m = len(board[0])\n",
+ " icons = {1:\"X\",2:\"O\",0:\" \"}\n",
+ " for i in range(n):\n",
+ " row = \"\"\n",
+ " print(\" --- \"*m)\n",
+ " for j in range(m):\n",
+ " row += (f\"| {icons[int(board[i][j])]} |\")\n",
+ " print(row)\n",
+ "\n",
+ " print(\" --- \"*m)"
]
},
{
"cell_type": "code",
- "execution_count": 7,
+ "execution_count": 89,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " --- --- --- \n",
+ "| X || X || X |\n",
+ " --- --- --- \n",
+ "| X || X || X |\n",
+ " --- --- --- \n",
+ "| X || X || X |\n",
+ " --- --- --- \n"
+ ]
+ }
+ ],
"source": [
- "# Test your solution here"
+ "# Test your solution \n",
+ "b= make_board(3)\n",
+ "draw_board(b)"
]
},
{
@@ -193,27 +369,51 @@
"Here are some example inputs you can use to test your code:\n"
]
},
- {
- "cell_type": "code",
- "execution_count": 8,
- "metadata": {},
- "outputs": [],
- "source": [
- "# Write your solution here"
- ]
- },
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
- "# Test your solution here"
+ "# Write your solution \n",
+ "def check_play(board):\n",
+ " r = len(board)\n",
+ " c = len(board[0])\n",
+ "\n",
+ " for i in range(r):\n",
+ " if all(board[i][j] == 1 for j in range(c)):\n",
+ " return 1\n",
+ " elif all(board[i][j] == 2 for j in range(c)):\n",
+ " return 2\n",
+ " \n",
+ " for j in range(c):\n",
+ " if all(board[i][j] == 1 for i in range(r)):\n",
+ " return 1\n",
+ " elif all(board[i][j] == 2 for i in range(r)):\n",
+ " return 2\n",
+ " \n",
+ " \n",
+ " if all(board[i][i] == 1 for i in range(r)):\n",
+ " return 1\n",
+ " elif all(board[i][i] == 2 for i in range(r)):\n",
+ " return 2\n",
+ "\n",
+ " if all(board[i][r-1-i] == 1 for i in range(r)):\n",
+ " return 1\n",
+ " elif all(board[i][r-1-i] == 2 for i in range(r)):\n",
+ " return 2\n",
+ "\n",
+ "\n",
+ " for i in range(r):\n",
+ " if any(board[i][j] == 0 for j in range(c)):\n",
+ " return -1\n",
+ " \n",
+ " return 0"
]
},
{
"cell_type": "code",
- "execution_count": 10,
+ "execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
@@ -235,7 +435,38 @@
"\n",
"also_no_winner = [[1, 2, 0],\n",
"\t[2, 1, 0],\n",
- "\t[2, 1, 0]]"
+ "\t[2, 1, 0]]\n",
+ "\n",
+ "draw = [[1,2,1],\n",
+ " [2,2,1],\n",
+ " [1,1,2]]"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 18,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "2\n",
+ "1\n",
+ "1\n",
+ "-1\n",
+ "-1\n",
+ "0\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(check_play(winner_is_2))\n",
+ "print(check_play(winner_is_1))\n",
+ "print(check_play(winner_is_also_1))\n",
+ "print(check_play(no_winner))\n",
+ "print(check_play(also_no_winner))\n",
+ "print(check_play(draw))"
]
},
{
@@ -252,20 +483,85 @@
},
{
"cell_type": "code",
- "execution_count": 11,
+ "execution_count": 52,
"metadata": {},
"outputs": [],
"source": [
- "# Write your solution here"
+ "# Write your solution here\n",
+ "def player_move(board, player, x, y):\n",
+ " col = ord(x.upper()) - ord('A')\n",
+ " row = int(y) - 1\n",
+ "\n",
+ " if board[row][col] != 0:\n",
+ " return \"Invalid move!\"\n",
+ "\n",
+ " board[row][col] = player\n",
+ " return True"
]
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": 59,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | X | | X |\n",
+ " --- --- --- \n",
+ "2 | | X | |\n",
+ " --- --- --- \n",
+ "3 | O | O | |\n",
+ " --- --- --- \n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Press 1 for X and 2 for O 1\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Enter co-ordinates to make the move\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the x co-ordinates: b\n",
+ "Enter the y co-ordinates: 1\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "True"
+ ]
+ },
+ "execution_count": 59,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
"source": [
- "# Test your solution here"
+ "# Test your solution here\n",
+ "b = [[1, 0, 1],\n",
+ "\t[0, 1, 0],\n",
+ "\t[2, 2, 0]]\n",
+ "draw_board(b)\n",
+ "a = input(\"Press 1 for X and 2 for O\")\n",
+ "print(\"Enter co-ordinates to make the move\")\n",
+ "x1 = input(\"Enter the x co-ordinates:\")\n",
+ "x2 = input(\"Enter the y co-ordinates:\")\n",
+ "player_move(b,a,x1,x2)"
]
},
{
@@ -277,20 +573,60 @@
},
{
"cell_type": "code",
- "execution_count": 13,
+ "execution_count": 92,
"metadata": {},
"outputs": [],
"source": [
- "# Write your solution here"
+ "def draw_board(board):\n",
+ " n = len(board)\n",
+ " m = len(board[0])\n",
+ "\n",
+ " icons = {1: \"X\", 2: \"O\", 0: \" \"}\n",
+ "\n",
+ " header = \" \"\n",
+ " for j in range(m):\n",
+ " header += f\" {chr(65 + j)} \"\n",
+ " print(header)\n",
+ "\n",
+ " for i in range(n):\n",
+ " print(\" \" + \"--- \" * m)\n",
+ " row = f\"{i + 1} \"\n",
+ " for j in range(m):\n",
+ " row += f\"| {icons[int(board[i][j])]} \"\n",
+ " row += \"|\"\n",
+ " print(row)\n",
+ "\n",
+ " print(\" \" + \"--- \" * m)"
]
},
{
"cell_type": "code",
- "execution_count": 14,
+ "execution_count": 24,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | X | X | X |\n",
+ " --- --- --- \n",
+ "2 | O | X | O |\n",
+ " --- --- --- \n",
+ "3 | O | X | X |\n",
+ " --- --- --- \n",
+ "4 | X | O | |\n",
+ " --- --- --- \n"
+ ]
+ }
+ ],
"source": [
- "# Test your solution here"
+ "# Test your solution \n",
+ "draw_board([[1, 1, 1],\n",
+ "\t[2, 1, 2],\n",
+ "\t[2, 1, 1],\n",
+ " [1,2,0]])"
]
},
{
@@ -302,7 +638,7 @@
},
{
"cell_type": "code",
- "execution_count": 15,
+ "execution_count": 70,
"metadata": {
"ExecuteTime": {
"end_time": "2026-01-26T21:09:58.171399Z",
@@ -311,16 +647,39 @@
},
"outputs": [],
"source": [
- "# Write your solution here"
+ "# Write your solution here\n",
+ "def move(board,player,a,b):\n",
+ " player_move(board,player,a,b)\n",
+ " draw_board(board)\n",
+ " "
]
},
{
"cell_type": "code",
- "execution_count": 16,
+ "execution_count": 71,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | X | X | X |\n",
+ " --- --- --- \n",
+ "2 | O | | O |\n",
+ " --- --- --- \n",
+ "3 | O | X | X |\n",
+ " --- --- --- \n"
+ ]
+ }
+ ],
"source": [
- "# Test your solution here"
+ "# Test your solution here\n",
+ "board1=[[1, 0, 1],\n",
+ "\t[2, 0, 2],\n",
+ "\t[2, 1, 1]]\n",
+ "move(board1,1,'b',1)"
]
},
{
@@ -334,20 +693,90 @@
},
{
"cell_type": "code",
- "execution_count": 17,
+ "execution_count": 72,
"metadata": {},
"outputs": [],
"source": [
- "# Write your solution here"
+ "# Write your solution here\n",
+ "def play_turn(board, player):\n",
+ " while True:\n",
+ " a = input(\"Enter the column letter: \")\n",
+ " b = input(\"Enter the row number: \")\n",
+ "\n",
+ " try:\n",
+ " if player_move(board, player, a, b) == True:\n",
+ " move(board, player, a, b)\n",
+ " break\n",
+ " else:\n",
+ " print(\"Invalid move! Try again.\")\n",
+ " except:\n",
+ " print(\"Invalid input! Try again.\")"
]
},
{
"cell_type": "code",
- "execution_count": 18,
+ "execution_count": 73,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter (e.g., A, B, C): s\n",
+ "Enter the row number (e.g., 1, 2, 3): 9\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Invalid input! Please enter a valid letter and row number.\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter (e.g., A, B, C): a\n",
+ "Enter the row number (e.g., 1, 2, 3): 1\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Invalid move! That spot is already taken.\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter (e.g., A, B, C): b\n",
+ "Enter the row number (e.g., 1, 2, 3): 1\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | X | X | X |\n",
+ " --- --- --- \n",
+ "2 | O | | O |\n",
+ " --- --- --- \n",
+ "3 | O | X | X |\n",
+ " --- --- --- \n"
+ ]
+ }
+ ],
"source": [
- "# Test your solution here"
+ "# Test your solution here\n",
+ "bord1=[[1, 0, 1],\n",
+ "\t[2, 0, 2],\n",
+ "\t[2, 1, 1]]\n",
+ "play_game(bord1,1)"
]
},
{
@@ -364,20 +793,258 @@
},
{
"cell_type": "code",
- "execution_count": 19,
+ "execution_count": 90,
"metadata": {},
"outputs": [],
"source": [
- "# Write yourrr solution here"
+ "# Write yourrr solution here\n",
+ "def make_board(n,m):\n",
+ " empty = 0\n",
+ " X = 1\n",
+ " O = 2\n",
+ " board = list()\n",
+ " for i in range(n):\n",
+ " row = list()\n",
+ " for j in range(m):\n",
+ " row.append(0)\n",
+ " board.append(row)\n",
+ " return board\n",
+ " \n",
+ "def play_game():\n",
+ " n = int(input(\"Enter the number of rows on board\"))\n",
+ " m = int(input(\"Enter the number of columns on board: \"))\n",
+ " board = make_board(n, m)\n",
+ " current_player = 1\n",
+ " \n",
+ " draw_board(board)\n",
+ " \n",
+ " while check_play(board) == -1:\n",
+ " print(f\"Player {current_player}'s turn\")\n",
+ " play_turn(board,current_player)\n",
+ " \n",
+ " status = check_play(board)\n",
+ " \n",
+ " if status == 1 or status == 2:\n",
+ " print(f\"Player {status} wins!\")\n",
+ " return\n",
+ " elif status == 0:\n",
+ " print(\"It's a draw!\")\n",
+ " return\n",
+ " \n",
+ " if current_player == 1:\n",
+ " current_player = 2\n",
+ " else:\n",
+ " current_player = 1"
]
},
{
"cell_type": "code",
- "execution_count": 20,
+ "execution_count": 93,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the number of rows on board 4\n",
+ "Enter the number of columns on board: 4\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D \n",
+ " --- --- --- --- \n",
+ "1 | | | | |\n",
+ " --- --- --- --- \n",
+ "2 | | | | |\n",
+ " --- --- --- --- \n",
+ "3 | | | | |\n",
+ " --- --- --- --- \n",
+ "4 | | | | |\n",
+ " --- --- --- --- \n",
+ "Player 1's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: a\n",
+ "Enter the row number: 1\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D \n",
+ " --- --- --- --- \n",
+ "1 | X | | | |\n",
+ " --- --- --- --- \n",
+ "2 | | | | |\n",
+ " --- --- --- --- \n",
+ "3 | | | | |\n",
+ " --- --- --- --- \n",
+ "4 | | | | |\n",
+ " --- --- --- --- \n",
+ "Player 2's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: b\n",
+ "Enter the row number: 4\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D \n",
+ " --- --- --- --- \n",
+ "1 | X | | | |\n",
+ " --- --- --- --- \n",
+ "2 | | | | |\n",
+ " --- --- --- --- \n",
+ "3 | | | | |\n",
+ " --- --- --- --- \n",
+ "4 | | O | | |\n",
+ " --- --- --- --- \n",
+ "Player 1's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: b\n",
+ "Enter the row number: 2\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D \n",
+ " --- --- --- --- \n",
+ "1 | X | | | |\n",
+ " --- --- --- --- \n",
+ "2 | | X | | |\n",
+ " --- --- --- --- \n",
+ "3 | | | | |\n",
+ " --- --- --- --- \n",
+ "4 | | O | | |\n",
+ " --- --- --- --- \n",
+ "Player 2's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: b\n",
+ "Enter the row number: 1\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D \n",
+ " --- --- --- --- \n",
+ "1 | X | O | | |\n",
+ " --- --- --- --- \n",
+ "2 | | X | | |\n",
+ " --- --- --- --- \n",
+ "3 | | | | |\n",
+ " --- --- --- --- \n",
+ "4 | | O | | |\n",
+ " --- --- --- --- \n",
+ "Player 1's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: c\n",
+ "Enter the row number: 3\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D \n",
+ " --- --- --- --- \n",
+ "1 | X | O | | |\n",
+ " --- --- --- --- \n",
+ "2 | | X | | |\n",
+ " --- --- --- --- \n",
+ "3 | | | X | |\n",
+ " --- --- --- --- \n",
+ "4 | | O | | |\n",
+ " --- --- --- --- \n",
+ "Player 2's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: d\n",
+ "Enter the row number: 3\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D \n",
+ " --- --- --- --- \n",
+ "1 | X | O | | |\n",
+ " --- --- --- --- \n",
+ "2 | | X | | |\n",
+ " --- --- --- --- \n",
+ "3 | | | X | O |\n",
+ " --- --- --- --- \n",
+ "4 | | O | | |\n",
+ " --- --- --- --- \n",
+ "Player 1's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: d\n",
+ "Enter the row number: 4\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D \n",
+ " --- --- --- --- \n",
+ "1 | X | O | | |\n",
+ " --- --- --- --- \n",
+ "2 | | X | | |\n",
+ " --- --- --- --- \n",
+ "3 | | | X | O |\n",
+ " --- --- --- --- \n",
+ "4 | | O | | X |\n",
+ " --- --- --- --- \n",
+ "Player 1 wins!\n"
+ ]
+ }
+ ],
"source": [
- "# Test your solution here"
+ "play_game()"
]
},
{
@@ -389,11 +1056,283 @@
},
{
"cell_type": "code",
- "execution_count": 21,
+ "execution_count": 94,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the number of rows on board 5\n",
+ "Enter the number of columns on board: 5\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D E \n",
+ " --- --- --- --- --- \n",
+ "1 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "2 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "3 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "4 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "5 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "Player 1's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: a\n",
+ "Enter the row number: 1\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D E \n",
+ " --- --- --- --- --- \n",
+ "1 | X | | | | |\n",
+ " --- --- --- --- --- \n",
+ "2 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "3 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "4 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "5 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "Player 2's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: e\n",
+ "Enter the row number: 3\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D E \n",
+ " --- --- --- --- --- \n",
+ "1 | X | | | | |\n",
+ " --- --- --- --- --- \n",
+ "2 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "3 | | | | | O |\n",
+ " --- --- --- --- --- \n",
+ "4 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "5 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "Player 1's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: b\n",
+ "Enter the row number: 2\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D E \n",
+ " --- --- --- --- --- \n",
+ "1 | X | | | | |\n",
+ " --- --- --- --- --- \n",
+ "2 | | X | | | |\n",
+ " --- --- --- --- --- \n",
+ "3 | | | | | O |\n",
+ " --- --- --- --- --- \n",
+ "4 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "5 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "Player 2's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: c\n",
+ "Enter the row number: 2\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D E \n",
+ " --- --- --- --- --- \n",
+ "1 | X | | | | |\n",
+ " --- --- --- --- --- \n",
+ "2 | | X | O | | |\n",
+ " --- --- --- --- --- \n",
+ "3 | | | | | O |\n",
+ " --- --- --- --- --- \n",
+ "4 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "5 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "Player 1's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: c\n",
+ "Enter the row number: 3\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D E \n",
+ " --- --- --- --- --- \n",
+ "1 | X | | | | |\n",
+ " --- --- --- --- --- \n",
+ "2 | | X | O | | |\n",
+ " --- --- --- --- --- \n",
+ "3 | | | X | | O |\n",
+ " --- --- --- --- --- \n",
+ "4 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "5 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "Player 2's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: d\n",
+ "Enter the row number: 3\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D E \n",
+ " --- --- --- --- --- \n",
+ "1 | X | | | | |\n",
+ " --- --- --- --- --- \n",
+ "2 | | X | O | | |\n",
+ " --- --- --- --- --- \n",
+ "3 | | | X | O | O |\n",
+ " --- --- --- --- --- \n",
+ "4 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "5 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "Player 1's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: d\n",
+ "Enter the row number: 4\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D E \n",
+ " --- --- --- --- --- \n",
+ "1 | X | | | | |\n",
+ " --- --- --- --- --- \n",
+ "2 | | X | O | | |\n",
+ " --- --- --- --- --- \n",
+ "3 | | | X | O | O |\n",
+ " --- --- --- --- --- \n",
+ "4 | | | | X | |\n",
+ " --- --- --- --- --- \n",
+ "5 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "Player 2's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: e\n",
+ "Enter the row number: 1\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D E \n",
+ " --- --- --- --- --- \n",
+ "1 | X | | | | O |\n",
+ " --- --- --- --- --- \n",
+ "2 | | X | O | | |\n",
+ " --- --- --- --- --- \n",
+ "3 | | | X | O | O |\n",
+ " --- --- --- --- --- \n",
+ "4 | | | | X | |\n",
+ " --- --- --- --- --- \n",
+ "5 | | | | | |\n",
+ " --- --- --- --- --- \n",
+ "Player 1's turn\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column letter: e\n",
+ "Enter the row number: 5\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C D E \n",
+ " --- --- --- --- --- \n",
+ "1 | X | | | | O |\n",
+ " --- --- --- --- --- \n",
+ "2 | | X | O | | |\n",
+ " --- --- --- --- --- \n",
+ "3 | | | X | O | O |\n",
+ " --- --- --- --- --- \n",
+ "4 | | | | X | |\n",
+ " --- --- --- --- --- \n",
+ "5 | | | | | X |\n",
+ " --- --- --- --- --- \n",
+ "Player 1 wins!\n"
+ ]
+ }
+ ],
"source": [
- "# Test your solution here"
+ "# Test your solution here\n",
+ "play_game()"
]
},
{
@@ -409,20 +1348,274 @@
},
{
"cell_type": "code",
- "execution_count": 22,
+ "execution_count": 95,
"metadata": {},
"outputs": [],
"source": [
- "# Write your solution here"
+ "# Write your solution here\n",
+ "import random\n",
+ "\n",
+ "def computer_move(board, computer, human):\n",
+ " r = len(board)\n",
+ " c = len(board[0])\n",
+ "\n",
+ " def try_move(player):\n",
+ " for i in range(r):\n",
+ " for j in range(c):\n",
+ " if board[i][j] == 0:\n",
+ " board[i][j] = player\n",
+ " if check_play(board) == player:\n",
+ " return (i, j)\n",
+ " board[i][j] = 0\n",
+ " return \n",
+ "\n",
+ " move = try_move(computer)\n",
+ " if move:\n",
+ " board[move[0]][move[1]] = computer\n",
+ " return\n",
+ " move = try_move(human)\n",
+ " if move:\n",
+ " board[move[0]][move[1]] = computer\n",
+ " return\n",
+ " empty = [(i, j) for i in range(r) for j in range(c) if board[i][j] == 0]\n",
+ " if empty:\n",
+ " i, j = random.choice(empty)\n",
+ " board[i][j] = computer"
]
},
{
"cell_type": "code",
- "execution_count": 23,
+ "execution_count": 96,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the number of rows on board: 3\n",
+ "Enter the number of columns on board: 3\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | | | |\n",
+ " --- --- --- \n",
+ "2 | | | |\n",
+ " --- --- --- \n",
+ "3 | | | |\n",
+ " --- --- --- \n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Press 1 for 2-player game, 2 to play vs computer: 2\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Enter coordinates to make the move\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column (A, B, C...): c\n",
+ "Enter the row (1, 2, 3...): 3\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | | | |\n",
+ " --- --- --- \n",
+ "2 | | | |\n",
+ " --- --- --- \n",
+ "3 | | | X |\n",
+ " --- --- --- \n",
+ "Computer is making a move...\n",
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | O | | |\n",
+ " --- --- --- \n",
+ "2 | | | |\n",
+ " --- --- --- \n",
+ "3 | | | X |\n",
+ " --- --- --- \n",
+ "Enter coordinates to make the move\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column (A, B, C...): c\n",
+ "Enter the row (1, 2, 3...): 1\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | O | | X |\n",
+ " --- --- --- \n",
+ "2 | | | |\n",
+ " --- --- --- \n",
+ "3 | | | X |\n",
+ " --- --- --- \n",
+ "Computer is making a move...\n",
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | O | | X |\n",
+ " --- --- --- \n",
+ "2 | | | O |\n",
+ " --- --- --- \n",
+ "3 | | | X |\n",
+ " --- --- --- \n",
+ "Enter coordinates to make the move\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column (A, B, C...): a\n",
+ "Enter the row (1, 2, 3...): 3\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | O | | X |\n",
+ " --- --- --- \n",
+ "2 | | | O |\n",
+ " --- --- --- \n",
+ "3 | X | | X |\n",
+ " --- --- --- \n",
+ "Computer is making a move...\n",
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | O | | X |\n",
+ " --- --- --- \n",
+ "2 | | O | O |\n",
+ " --- --- --- \n",
+ "3 | X | | X |\n",
+ " --- --- --- \n",
+ "Enter coordinates to make the move\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column (A, B, C...): a\n",
+ "Enter the row (1, 2, 3...): 2\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | O | | X |\n",
+ " --- --- --- \n",
+ "2 | X | O | O |\n",
+ " --- --- --- \n",
+ "3 | X | | X |\n",
+ " --- --- --- \n",
+ "Computer is making a move...\n",
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | O | | X |\n",
+ " --- --- --- \n",
+ "2 | X | O | O |\n",
+ " --- --- --- \n",
+ "3 | X | O | X |\n",
+ " --- --- --- \n",
+ "Enter coordinates to make the move\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column (A, B, C...): b\n",
+ "Enter the row (1, 2, 3...): 1\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | O | X | X |\n",
+ " --- --- --- \n",
+ "2 | X | O | O |\n",
+ " --- --- --- \n",
+ "3 | X | O | X |\n",
+ " --- --- --- \n",
+ "Its a draw.\n"
+ ]
+ }
+ ],
"source": [
- "# Test your solution here"
+ "# Test your solution here\n",
+ "x_size = int(input(\"Enter the number of rows on board: \"))\n",
+ "y_size = int(input(\"Enter the number of columns on board: \"))\n",
+ "\n",
+ "b = make_board(x_size, y_size)\n",
+ "draw_board(b)\n",
+ "\n",
+ "mode = int(input(\"Press 1 for 2-player game, 2 to play vs computer: \"))\n",
+ "player = 1\n",
+ "computer = 2 if mode == 2 else None\n",
+ "moves = -1\n",
+ "\n",
+ "while moves == -1:\n",
+ "\n",
+ " if mode == 2 and player == computer:\n",
+ " print(\"Computer is making a move...\")\n",
+ " computer_move(b, computer, 1)\n",
+ " else:\n",
+ " print(\"Enter coordinates to make the move\")\n",
+ " x1 = input(\"Enter the column (A, B, C...): \")\n",
+ " x2 = input(\"Enter the row (1, 2, 3...): \")\n",
+ "\n",
+ " result = player_move(b, player, x1, x2)\n",
+ " if result != True:\n",
+ " print(result)\n",
+ " continue\n",
+ "\n",
+ " draw_board(b)\n",
+ " moves = check_play(b)\n",
+ "\n",
+ " if moves == -1:\n",
+ " player = 2 if player == 1 else 1\n",
+ "\n",
+ "if moves == 1:\n",
+ " print(\"X won.\")\n",
+ "elif moves == 2:\n",
+ " print(\"O won.\")\n",
+ "else:\n",
+ " print(\"Its a draw.\")"
]
},
{
@@ -434,20 +1627,319 @@
},
{
"cell_type": "code",
- "execution_count": 24,
+ "execution_count": 97,
"metadata": {},
"outputs": [],
"source": [
- "# Write your solution here"
+ "# Write your solution here\n",
+ "import random\n",
+ "\n",
+ "def minimax(board, depth, max_depth, player, smart, dumb):\n",
+ " result = check_play(board)\n",
+ "\n",
+ " if result == smart:\n",
+ " return 100 - depth\n",
+ " if result == dumb:\n",
+ " return depth - 100\n",
+ " if result == 0:\n",
+ " return 0\n",
+ "\n",
+ " if depth == max_depth:\n",
+ " return 0\n",
+ "\n",
+ " r = len(board)\n",
+ " c = len(board[0])\n",
+ "\n",
+ " if player == smart:\n",
+ " best = -float(\"inf\")\n",
+ "\n",
+ " for i in range(r):\n",
+ " for j in range(c):\n",
+ " if board[i][j] == 0:\n",
+ " board[i][j] = smart\n",
+ " score = minimax(board, depth + 1, max_depth,\n",
+ " dumb, smart, dumb)\n",
+ " board[i][j] = 0\n",
+ " best = max(best, score)\n",
+ "\n",
+ " return best\n",
+ "\n",
+ " else:\n",
+ " best = float(\"inf\")\n",
+ "\n",
+ " for i in range(r):\n",
+ " for j in range(c):\n",
+ " if board[i][j] == 0:\n",
+ " board[i][j] = dumb\n",
+ " score = minimax(board, depth + 1, max_depth,\n",
+ " smart, smart, dumb)\n",
+ " board[i][j] = 0\n",
+ " best = min(best, score)\n",
+ "\n",
+ " return best\n",
+ "\n",
+ "\n",
+ "def computer_move_exhaustive(board, player, opponent, depth):\n",
+ " r = len(board)\n",
+ " c = len(board[0])\n",
+ "\n",
+ " best_score = -float(\"inf\")\n",
+ " best_move = None\n",
+ "\n",
+ " for i in range(r):\n",
+ " for j in range(c):\n",
+ " if board[i][j] == 0:\n",
+ " board[i][j] = player\n",
+ " score = minimax(board, 0, depth,\n",
+ " opponent, player, opponent)\n",
+ " board[i][j] = 0\n",
+ "\n",
+ " if score > best_score:\n",
+ " best_score = score\n",
+ " best_move = (i, j)\n",
+ "\n",
+ " if best_move:\n",
+ " board[best_move[0]][best_move[1]] = player\n"
]
},
{
"cell_type": "code",
- "execution_count": 25,
+ "execution_count": 99,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the number of rows on board: 3\n",
+ "Enter the number of columns on board: 3\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | | | |\n",
+ " --- --- --- \n",
+ "2 | | | |\n",
+ " --- --- --- \n",
+ "3 | | | |\n",
+ " --- --- --- \n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Press 1 for 2-player game, 2 to play vs computer: 2\n",
+ "Enter search depth (recommended 3-5): 6\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Enter coordinates to make the move\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column (A, B, C...): c\n",
+ "Enter the row (1, 2, 3...): 3\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | | | |\n",
+ " --- --- --- \n",
+ "2 | | | |\n",
+ " --- --- --- \n",
+ "3 | | | X |\n",
+ " --- --- --- \n",
+ "Computer is making a move...\n",
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | | | |\n",
+ " --- --- --- \n",
+ "2 | | O | |\n",
+ " --- --- --- \n",
+ "3 | | | X |\n",
+ " --- --- --- \n",
+ "Enter coordinates to make the move\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column (A, B, C...): a\n",
+ "Enter the row (1, 2, 3...): 1\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | X | | |\n",
+ " --- --- --- \n",
+ "2 | | O | |\n",
+ " --- --- --- \n",
+ "3 | | | X |\n",
+ " --- --- --- \n",
+ "Computer is making a move...\n",
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | X | O | |\n",
+ " --- --- --- \n",
+ "2 | | O | |\n",
+ " --- --- --- \n",
+ "3 | | | X |\n",
+ " --- --- --- \n",
+ "Enter coordinates to make the move\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column (A, B, C...): b\n",
+ "Enter the row (1, 2, 3...): 3\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | X | O | |\n",
+ " --- --- --- \n",
+ "2 | | O | |\n",
+ " --- --- --- \n",
+ "3 | | X | X |\n",
+ " --- --- --- \n",
+ "Computer is making a move...\n",
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | X | O | |\n",
+ " --- --- --- \n",
+ "2 | | O | |\n",
+ " --- --- --- \n",
+ "3 | O | X | X |\n",
+ " --- --- --- \n",
+ "Enter coordinates to make the move\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column (A, B, C...): c\n",
+ "Enter the row (1, 2, 3...): 1\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | X | O | X |\n",
+ " --- --- --- \n",
+ "2 | | O | |\n",
+ " --- --- --- \n",
+ "3 | O | X | X |\n",
+ " --- --- --- \n",
+ "Computer is making a move...\n",
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | X | O | X |\n",
+ " --- --- --- \n",
+ "2 | | O | O |\n",
+ " --- --- --- \n",
+ "3 | O | X | X |\n",
+ " --- --- --- \n",
+ "Enter coordinates to make the move\n"
+ ]
+ },
+ {
+ "name": "stdin",
+ "output_type": "stream",
+ "text": [
+ "Enter the column (A, B, C...): a\n",
+ "Enter the row (1, 2, 3...): 2\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " A B C \n",
+ " --- --- --- \n",
+ "1 | X | O | X |\n",
+ " --- --- --- \n",
+ "2 | X | O | O |\n",
+ " --- --- --- \n",
+ "3 | O | X | X |\n",
+ " --- --- --- \n",
+ "Its a draw.\n"
+ ]
+ }
+ ],
"source": [
- "# Test your solution here"
+ "# Test your solution \n",
+ "# Test your solution here\n",
+ "x_size = int(input(\"Enter the number of rows on board: \"))\n",
+ "y_size = int(input(\"Enter the number of columns on board: \"))\n",
+ "\n",
+ "b = make_board(x_size, y_size)\n",
+ "draw_board(b)\n",
+ "\n",
+ "mode = int(input(\"Press 1 for 2-player game, 2 to play vs computer: \"))\n",
+ "player = 1\n",
+ "computer = 2 if mode == 2 else None\n",
+ "if mode == 2:\n",
+ " max_depth = int(input(\"Enter search depth (recommended 3-5): \"))\n",
+ "\n",
+ "moves = -1\n",
+ "\n",
+ "while moves == -1:\n",
+ "\n",
+ " if mode == 2 and player == computer:\n",
+ " print(\"Computer is making a move...\")\n",
+ " computer_move_exhaustive(b, computer, 1, max_depth)\n",
+ " else:\n",
+ " print(\"Enter coordinates to make the move\")\n",
+ " x1 = input(\"Enter the column (A, B, C...): \")\n",
+ " x2 = input(\"Enter the row (1, 2, 3...): \")\n",
+ "\n",
+ " result = player_move(b, player, x1, x2)\n",
+ " if result != True:\n",
+ " print(result)\n",
+ " continue\n",
+ "\n",
+ " draw_board(b)\n",
+ " moves = check_play(b)\n",
+ "\n",
+ " if moves == -1:\n",
+ " player = 2 if player == 1 else 1\n",
+ "\n",
+ "if moves == 1:\n",
+ " print(\"X won.\")\n",
+ "elif moves == 2:\n",
+ " print(\"O won.\")\n",
+ "else:\n",
+ " print(\"Its a draw.\")"
]
},
{
@@ -459,20 +1951,67 @@
},
{
"cell_type": "code",
- "execution_count": 26,
+ "execution_count": 103,
"metadata": {},
"outputs": [],
"source": [
- "# Write your solution here"
+ "# Write your solution \n",
+ "def play_ai_vs_ai(size, depth_smart, depth_dumb, games=10):\n",
+ " smart_wins = 0\n",
+ " draw= 0\n",
+ " for g in range(games):\n",
+ " board = make_board(size, size)\n",
+ "\n",
+ " smart = 1 if g % 2 == 0 else 2\n",
+ " dumb = 2 if smart == 1 else 1\n",
+ "\n",
+ " player = 1\n",
+ " result = -1\n",
+ "\n",
+ " while result == -1:\n",
+ " if player == smart:\n",
+ " computer_move_exhaustive(board, smart, dumb, depth_smart)\n",
+ " else:\n",
+ " computer_move_exhaustive(board, dumb, smart, depth_dumb)\n",
+ "\n",
+ " result = check_play(board)\n",
+ " player = 2 if player == 1 else 1\n",
+ "\n",
+ " if result == smart:\n",
+ " smart_wins += 1\n",
+ " elif result == 0:\n",
+ " draw += 1\n",
+ "\n",
+ " win_rate = smart_wins / games * 100\n",
+ " print(f\"{size}x{size} grid → Smart AI win rate: {win_rate:.1f}% with {draw} draws.\")\n"
]
},
{
"cell_type": "code",
- "execution_count": 27,
+ "execution_count": 104,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "AI vs AI experiment:\n",
+ "\n",
+ "3x3 grid → Smart AI win rate: 50.0% with 5 draws.\n",
+ "4x4 grid → Smart AI win rate: 0.0% with 10 draws.\n",
+ "5x5 grid → Smart AI win rate: 0.0% with 10 draws.\n"
+ ]
+ }
+ ],
"source": [
- "# Test your solution here"
+ "# Test your solution here\n",
+ "# Test your solution here\n",
+ "print(\"\\nAI vs AI experiment:\\n\")\n",
+ "\n",
+ "play_ai_vs_ai(3, depth_smart=5, depth_dumb=2)\n",
+ "play_ai_vs_ai(4, depth_smart=4, depth_dumb=2)\n",
+ "play_ai_vs_ai(5, depth_smart=3, depth_dumb=1)"
]
},
{
@@ -510,7 +2049,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.13.7"
+ "version": "3.12.12"
}
},
"nbformat": 4,
diff --git a/Labs/Lab.3/.DS_Store b/Labs/Lab.3/.DS_Store
new file mode 100644
index 00000000..b64e8a66
Binary files /dev/null and b/Labs/Lab.3/.DS_Store differ
diff --git a/Labs/Lab.3/Lab.3.ipynb b/Labs/Lab.3/Lab.3.ipynb
index 3dc0438e..0154b718 100644
--- a/Labs/Lab.3/Lab.3.ipynb
+++ b/Labs/Lab.3/Lab.3.ipynb
@@ -207,9 +207,17 @@
},
{
"cell_type": "code",
- "execution_count": null,
- "metadata": {},
- "outputs": [],
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "The Value of x is 0.9036066841881488\n"
+ ]
+ }
+ ],
"source": [
"import random\n",
"x=random.random()\n",
@@ -227,7 +235,7 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
@@ -235,18 +243,30 @@
"def generate_uniform(N,x_min,x_max):\n",
" out = []\n",
" ### BEGIN SOLUTION\n",
- "\n",
- " # Fill in your solution here \n",
- " \n",
+ " for _ in range(N):\n",
+ " d = random.random()\n",
+ " out.append(int(x_min + d * (x_max - x_min)))\n",
" ### END SOLUTION\n",
" return out"
]
},
{
"cell_type": "code",
- "execution_count": null,
- "metadata": {},
- "outputs": [],
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Data Type: \n",
+ "Data Length: 1000\n",
+ "Type of Data Contents: \n",
+ "Data Minimum: -9\n",
+ "Data Maximum: 9\n"
+ ]
+ }
+ ],
"source": [
"# Test your solution here\n",
"data=generate_uniform(1000,-10,10)\n",
@@ -268,7 +288,7 @@
},
{
"cell_type": "code",
- "execution_count": 1,
+ "execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
@@ -276,10 +296,10 @@
"def mean(Data):\n",
" m=0.\n",
" \n",
- " ### BEGIN SOLUTION\n",
- "\n",
- " # Fill in your solution here \n",
- " \n",
+ " ### BEGIN SOLUTION \n",
+ " n = len(Data)\n",
+ " Summation = sum(Data)\n",
+ " m = Summation/n\n",
" ### END SOLUTION\n",
" \n",
" return m"
@@ -287,11 +307,20 @@
},
{
"cell_type": "code",
- "execution_count": null,
- "metadata": {},
- "outputs": [],
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Mean of Data: 9.0\n"
+ ]
+ }
+ ],
"source": [
"# Test your solution here\n",
+ "data = [8,10]\n",
"print (\"Mean of Data:\", mean(data))"
]
},
@@ -305,7 +334,7 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
@@ -314,9 +343,14 @@
" m=0.\n",
" \n",
" ### BEGIN SOLUTION\n",
- "\n",
- " # Fill in your solution here \n",
- " \n",
+ " n = len(Data)\n",
+ " x_bar = mean(Data)\n",
+ " summation = 0\n",
+ " for i in range(n):\n",
+ " difference_sq = (Data[i] - x_bar) * (Data[i] - x_bar)\n",
+ " summation += difference_sq\n",
+ " \n",
+ " m = (1/(n-1)) * summation\n",
" ### END SOLUTION\n",
" \n",
" return m"
@@ -324,11 +358,20 @@
},
{
"cell_type": "code",
- "execution_count": null,
- "metadata": {},
- "outputs": [],
+ "execution_count": 8,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Variance of Data: 542.1944444444445\n"
+ ]
+ }
+ ],
"source": [
"# Test your solution here\n",
+ "data = [12,19,45,68,90,42,47,50,48]\n",
"print (\"Variance of Data:\", variance(data))"
]
},
@@ -358,16 +401,24 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 9,
"metadata": {},
"outputs": [],
"source": [
"# Solution\n",
"def histogram(x,n_bins=10,x_min=None,x_max=None):\n",
" ### BEGIN SOLUTION\n",
- "\n",
- " # Fill in your solution here \n",
- " \n",
+ " x_min, x_max = min(x), max(x)\n",
+ " bin_size = (x_max - x_min) / n_bins\n",
+ " hist = [0] * n_bins\n",
+ " bin_edges = [x_min + i * bin_size for i in range(n_bins + 1)]\n",
+ " for value in x:\n",
+ " for i in range(n_bins):\n",
+ " lower_bound = bin_edges[i]\n",
+ " upper_bound = bin_edges[i+1]\n",
+ " if lower_bound <= value <= upper_bound:\n",
+ " hist[i] += 1\n",
+ " break\n",
" ### END SOLUTION\n",
"\n",
" return hist,bin_edges"
@@ -375,13 +426,24 @@
},
{
"cell_type": "code",
- "execution_count": null,
- "metadata": {},
- "outputs": [],
+ "execution_count": 10,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "[3, 2, 2]\n",
+ "[1.0, 4.666666666666666, 8.333333333333332, 12.0]\n"
+ ]
+ }
+ ],
"source": [
"# Test your solution here\n",
- "h,b=histogram(data,100)\n",
- "print(h)"
+ "data = [1, 2, 2, 5, 8, 10, 12]\n",
+ "h,b=histogram(data,3)\n",
+ "print(h)\n",
+ "print(b)"
]
},
{
@@ -407,29 +469,59 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
- "# Solution\n",
- "def draw_histogram(x,n_bins,x_min=None,x_max=None,character=\"#\",max_character_per_line=20):\n",
+ "def draw_histogram(x, n_bins, x_min=None, x_max=None, character=\"#\", max_character_per_line=20):\n",
" ### BEGIN SOLUTION\n",
- "\n",
- " # Fill in your solution here \n",
+ " hist, bin_edges = histogram(x, n_bins)\n",
+ " max_h = max(hist)\n",
" \n",
+ " for i in range(len(hist)):\n",
+ " if max_h > 0:\n",
+ " num_chars = int((hist[i] / max_h) * max_character_per_line)\n",
+ " else:\n",
+ " num_chars = 0\n",
+ " \n",
+ " bar = character * num_chars\n",
+ " lower = bin_edges[i]\n",
+ " upper = bin_edges[i+1]\n",
+ " print(f\"[{lower:>3.0f}, {upper:>3.0f}] : {bar}\")\n",
" ### END SOLUTION\n",
"\n",
- " return hist,bin_edges"
+ " return hist, bin_edges"
]
},
{
"cell_type": "code",
- "execution_count": null,
- "metadata": {},
- "outputs": [],
+ "execution_count": 12,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "[ 0, 1] : ######\n",
+ "[ 1, 2] : ######\n",
+ "[ 2, 3] : #############\n",
+ "[ 3, 4] : ####################\n",
+ "[ 4, 4] : #############\n",
+ "[ 4, 5] : ####################\n",
+ "[ 5, 6] : #############\n",
+ "[ 6, 7] : ####################\n",
+ "[ 7, 8] : #############\n",
+ "[ 8, 9] : ####################\n"
+ ]
+ }
+ ],
"source": [
"# Test your solution here\n",
- "h,b=histogram(data,20)"
+ "# Create simple data from 0 to 10 to match your image\n",
+ "data = [0, 1, 2, 2, 3, 3, 3, 4, 4, 5, 5, 5, 6, 6, 7, 7, 7, 8, 8, 9, 9, 9]\n",
+ "\n",
+ "# This line will now print the graph to the screen\n",
+ "h, b = draw_histogram(data, n_bins=10)"
]
},
{
@@ -443,29 +535,40 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 13,
"metadata": {},
"outputs": [],
"source": [
- "def where(mylist,myfunc):\n",
- " out= []\n",
- " \n",
- " ### BEGIN SOLUTION\n",
+ "def where(mylist, myfunc):\n",
+ " out = []\n",
"\n",
- " # Fill in your solution here \n",
- " \n",
+ " ### BEGIN SOLUTION\n",
+ " for i in range(len(mylist)):\n",
+ " if myfunc(mylist[i]):\n",
+ " out.append(i)\n",
" ### END SOLUTION\n",
- " \n",
+ "\n",
" return out"
]
},
{
"cell_type": "code",
- "execution_count": null,
- "metadata": {},
- "outputs": [],
+ "execution_count": 28,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "[1, 3]\n"
+ ]
+ }
+ ],
"source": [
- "# Test your solution here"
+ "# Test your solution here\n",
+ "data = [0.1, 0.6, 0.3, 0.8, 0.2]\n",
+ "indices = where(data, lambda x: x > 0.5)\n",
+ "print(indices)"
]
},
{
@@ -483,9 +586,20 @@
},
{
"cell_type": "code",
- "execution_count": null,
- "metadata": {},
- "outputs": [],
+ "execution_count": 15,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "True True False False False\n",
+ "False False True True False\n",
+ "Number of Entries passing F1: 5\n",
+ "Number of Entries passing F2: 0\n"
+ ]
+ }
+ ],
"source": [
"def in_range(mymin,mymax):\n",
" def testrange(x):\n",
@@ -493,8 +607,8 @@
" return testrange\n",
"\n",
"# Examples:\n",
- "F1=inrange(0,10)\n",
- "F2=inrange(10,20)\n",
+ "F1=in_range(0,10)\n",
+ "F2=in_range(10,20)\n",
"\n",
"# Test of in_range\n",
"print (F1(0), F1(1), F1(10), F1(15), F1(20))\n",
@@ -506,24 +620,63 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
"### BEGIN SOLUTION\n",
- "\n",
- " # Fill in your solution here \n",
- " \n",
+ "def is_even(x):\n",
+ " return x % 2 == 0\n",
+ "\n",
+ "def is_odd(x):\n",
+ " return x % 2 != 0\n",
+ "\n",
+ "def greater_than(val):\n",
+ " def tester(x):\n",
+ " return x > val\n",
+ " return tester\n",
+ "\n",
+ "def less_than(val):\n",
+ " def tester(x):\n",
+ " return x < val\n",
+ " return tester\n",
+ "\n",
+ "def equal(val):\n",
+ " def tester(x):\n",
+ " return x == val\n",
+ " return tester\n",
+ "\n",
+ "def divisible_by(val):\n",
+ " def tester(x):\n",
+ " return x % val == 0\n",
+ " return tester \n",
"### END SOLUTION"
]
},
{
"cell_type": "code",
- "execution_count": null,
- "metadata": {},
- "outputs": [],
+ "execution_count": 17,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Even indices: [1, 3, 5, 6]\n",
+ "Greater than 5 indices: [5, 6, 7]\n",
+ "Divisible by 3 indices: [2, 5, 7]\n"
+ ]
+ }
+ ],
"source": [
- "# Test your solution"
+ "# Test your solution\n",
+ "data = [1, 2, 3, 4, 5, 6, 10, 15]\n",
+ "\n",
+ "print(\"Even indices:\", where(data, is_even))\n",
+ "\n",
+ "print(\"Greater than 5 indices:\", where(data, greater_than(5)))\n",
+ "\n",
+ "print(\"Divisible by 3 indices:\", where(data, divisible_by(3)))"
]
},
{
@@ -535,14 +688,30 @@
},
{
"cell_type": "code",
- "execution_count": null,
- "metadata": {},
- "outputs": [],
+ "execution_count": 18,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Even: 4\n",
+ "Odd: 4\n",
+ "Greater than: 8\n",
+ "Less than: 0\n",
+ "Equal: 0\n",
+ "Divisible by: 3\n"
+ ]
+ }
+ ],
"source": [
"### BEGIN SOLUTION\n",
- "\n",
- " # Fill in your solution here \n",
- " \n",
+ "print(\"Even:\", sum(map(lambda x: x % 2 == 0, data)))\n",
+ "print(\"Odd:\", sum(map(lambda x: x % 2 != 0, data)))\n",
+ "print(\"Greater than:\", sum(map(lambda x: x > 0.5, data)))\n",
+ "print(\"Less than:\", sum(map(lambda x: x < 0.5, data)))\n",
+ "print(\"Equal:\", sum(map(lambda x: x == 0, data)))\n",
+ "print(\"Divisible by:\", sum(map(lambda x: x % 3 == 0, data)))\n",
"### END SOLUTION"
]
},
@@ -561,30 +730,73 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 19,
"metadata": {},
"outputs": [],
"source": [
"def generate_function(func,x_min,x_max,N=1000):\n",
" out = list()\n",
" ### BEGIN SOLUTION\n",
+ " import random\n",
"\n",
- " # Fill in your solution here \n",
+ " y_max = 0\n",
+ " steps = 1000\n",
+ " step_size = (x_max - x_min) / steps\n",
" \n",
+ " for i in range(steps):\n",
+ " val = func(x_min + i * step_size)\n",
+ " if val > y_max:\n",
+ " y_max = val\n",
+ " \n",
+ " while len(out) < N:\n",
+ " test_x = random.uniform(x_min, x_max)\n",
+ " p = random.uniform(0, y_max)\n",
+ " \n",
+ " if p <= func(test_x):\n",
+ " out.append(test_x)\n",
" ### END SOLUTION\n",
- " \n",
" return out"
]
},
{
"cell_type": "code",
- "execution_count": null,
- "metadata": {},
- "outputs": [],
+ "execution_count": 20,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "[ -5, -5] : ####################\n",
+ "[ -5, -4] : ##################\n",
+ "[ -4, -4] : ################\n",
+ "[ -4, -3] : ##############\n",
+ "[ -3, -3] : ##########\n",
+ "[ -3, -3] : #########\n",
+ "[ -3, -2] : #######\n",
+ "[ -2, -2] : ####\n",
+ "[ -2, -1] : ##\n",
+ "[ -1, -1] : #\n",
+ "[ -1, -1] : #\n",
+ "[ -1, -0] : ##\n",
+ "[ -0, 0] : #####\n",
+ "[ 0, 1] : #######\n",
+ "[ 1, 1] : #########\n",
+ "[ 1, 1] : ###########\n",
+ "[ 1, 2] : ##############\n",
+ "[ 2, 2] : ###############\n",
+ "[ 2, 3] : #################\n",
+ "[ 3, 3] : ##################\n",
+ "([950, 863, 792, 686, 521, 432, 354, 219, 142, 54, 54, 137, 262, 362, 442, 565, 695, 741, 835, 894], [-4.999913621869641, -4.599928144478, -4.199942667086358, -3.7999571896947164, -3.3999717123030746, -2.999986234911433, -2.6000007575197914, -2.2000152801281496, -1.8000298027365078, -1.400044325344866, -1.0000588479532242, -0.6000733705615824, -0.20008789316994147, 0.19989758422170034, 0.5998830616133422, 0.999868539004984, 1.3998540163966258, 1.7998394937882676, 2.1998249711799094, 2.599810448571551, 2.999795925963193])\n"
+ ]
+ }
+ ],
"source": [
"# A test function\n",
"def test_func(x,a=1,b=1):\n",
- " return abs(a*x+b)"
+ " return abs(a*x+b)\n",
+ "data = generate_function(test_func, x_min=-5, x_max=3, N=10000)\n",
+ "print(draw_histogram(data, n_bins=20))"
]
},
{
@@ -596,9 +808,38 @@
},
{
"cell_type": "code",
- "execution_count": null,
- "metadata": {},
- "outputs": [],
+ "execution_count": 25,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Mean: -0.0052738494072836215\n",
+ "Variance: 0.9638367958021714\n",
+ "[ -4, -3] : \n",
+ "[ -3, -3] : \n",
+ "[ -3, -3] : \n",
+ "[ -3, -2] : \n",
+ "[ -2, -2] : ##\n",
+ "[ -2, -1] : ###\n",
+ "[ -1, -1] : #########\n",
+ "[ -1, -1] : #############\n",
+ "[ -1, -0] : ################\n",
+ "[ -0, 0] : ####################\n",
+ "[ 0, 0] : ###############\n",
+ "[ 0, 1] : ###############\n",
+ "[ 1, 1] : ###########\n",
+ "[ 1, 2] : ######\n",
+ "[ 2, 2] : ###\n",
+ "[ 2, 2] : ##\n",
+ "[ 2, 3] : \n",
+ "[ 3, 3] : \n",
+ "[ 3, 3] : \n",
+ "[ 3, 4] : \n"
+ ]
+ }
+ ],
"source": [
"import math\n",
"\n",
@@ -607,6 +848,17 @@
" return math.exp(-((x-mean)**2)/(2*sigma**2))/math.sqrt(math.pi*sigma)\n",
" return f\n",
"\n",
+ "g1 = gaussian(0, 1)\n",
+ "data = generate_function(g1, -5, 5, N=1000)\n",
+ "\n",
+ "calc_mean = sum(data) / len(data)\n",
+ "calc_var = sum([(x - calc_mean)**2 for x in data]) / len(data)\n",
+ "\n",
+ "print(f\"Mean: {calc_mean}\")\n",
+ "print(f\"Variance: {calc_var}\")\n",
+ "\n",
+ "draw_histogram(data, n_bins=20)\n",
+ "\n",
"# Example Instantiation\n",
"g1=gaussian(0,1)\n",
"g2=gaussian(10,3)"
@@ -621,19 +873,42 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 26,
"metadata": {},
"outputs": [],
"source": [
"def integrate(func, x_min, x_max, n_points=1000):\n",
- " \n",
+ " ### BEGIN SOLUTION\n",
+ " all_data = generate_function(func, x_min - 5, x_max + 5, N=n_points)\n",
+ " checker = in_range(x_min, x_max)\n",
+ " valid_indices = where(all_data, checker)\n",
+ " integral = len(valid_indices) / n_points\n",
+ " ### END SOLUTION\n",
" return integral"
]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 27,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "0.6908\n"
+ ]
+ }
+ ],
+ "source": [
+ "g1 = gaussian(0, 1)\n",
+ "print(integrate(g1, -1, 1, n_points=5000))"
+ ]
}
],
"metadata": {
"kernelspec": {
- "display_name": "Python 3 (ipykernel)",
+ "display_name": "ds",
"language": "python",
"name": "python3"
},
@@ -647,7 +922,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.13.7"
+ "version": "3.11.14"
}
},
"nbformat": 4,
diff --git a/Labs/Lab.4/.ipynb_checkpoints/Lab.4-checkpoint.ipynb b/Labs/Lab.4/.ipynb_checkpoints/Lab.4-checkpoint.ipynb
new file mode 100644
index 00000000..2b630cf9
--- /dev/null
+++ b/Labs/Lab.4/.ipynb_checkpoints/Lab.4-checkpoint.ipynb
@@ -0,0 +1,397 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Lab 4- Object Oriented Programming\n",
+ "\n",
+ "For all of the exercises below, make sure you provide tests of your solutions.\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "1. Write a \"counter\" class that can be incremented up to a specified maximum value, will print an error if an attempt is made to increment beyond that value, and allows reseting the counter. "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Error: Attempted to increment beyond maximum value.\n",
+ "Error: Attempted to increment beyond maximum value.\n"
+ ]
+ }
+ ],
+ "source": [
+ "class SimpleCounter:\n",
+ " def __init__(self, max_value):\n",
+ " self.count = 0\n",
+ " self.max_value = max_value\n",
+ "\n",
+ " def increment(self):\n",
+ " if self.count >= self.max_value:\n",
+ " print(\"Error: Attempted to increment beyond maximum value.\")\n",
+ " else:\n",
+ " self.count += 1\n",
+ "\n",
+ " def reset(self):\n",
+ " self.count = 0\n",
+ "\n",
+ "# --- Test ---\n",
+ "c1 = SimpleCounter(2)\n",
+ "c1.increment()\n",
+ "c1.increment()\n",
+ "c1.increment() # This will print the error\n",
+ "c1.reset()\n",
+ "c1.increment() \n",
+ "c1.increment()\n",
+ "c1.increment() # This will print the error again"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "2. Copy and paste your solution to question 1 and modify it so that all the data held by the counter is private. Implement functions to check the value of the counter, check the maximum value, and check if the counter is at the maximum."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Value: 2, Is Max?: False\n"
+ ]
+ }
+ ],
+ "source": [
+ "class Counter:\n",
+ " def __init__(self, max_value):\n",
+ " self.__count = 0 \n",
+ " self.__max_value = max_value\n",
+ "\n",
+ " def increment(self):\n",
+ " if self.__count >= self.__max_value:\n",
+ " print(\"Error: Attempted to increment beyond maximum value.\")\n",
+ " else:\n",
+ " self.__count += 1\n",
+ "\n",
+ " def reset(self):\n",
+ " self.__count = 0\n",
+ "\n",
+ " def get_value(self):\n",
+ " return self.__count\n",
+ "\n",
+ " def get_max(self):\n",
+ " return self.__max_value\n",
+ "\n",
+ " def is_at_max(self):\n",
+ " return self.__count == self.__max_value\n",
+ "\n",
+ "c2 = Counter(3)\n",
+ "c2.increment()\n",
+ "c2.increment()\n",
+ "print(f\"Value: {c2.get_value()}, Is Max?: {c2.is_at_max()}\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "3. Implement a class to represent a rectangle, holding the length, width, and $x$ and $y$ coordinates of a corner of the object. Implement functions that compute the area and perimeter of the rectangle. Make all data members private and privide accessors to retrieve values of data members. "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Area: 8, Perimeter: 12.\n"
+ ]
+ }
+ ],
+ "source": [
+ "class Rectangle:\n",
+ " def __init__(self, length, width, x, y):\n",
+ " self.__length = length\n",
+ " self.__width = width\n",
+ " self.__x = x\n",
+ " self.__y = y\n",
+ "\n",
+ " def get_length(self): return self.__length\n",
+ " def get_width(self): return self.__width\n",
+ " def get_x(self): return self.__x\n",
+ " def get_y(self): return self.__y\n",
+ "\n",
+ " def area(self):\n",
+ " return self.__length * self.__width\n",
+ " \n",
+ " def perimeter(self):\n",
+ " return 2 * (self.__length + self.__width)\n",
+ " \n",
+ "# Test\n",
+ "rect = Rectangle(4, 2, 0, 0)\n",
+ "print(f\"Area: {rect.area()}, Perimeter: {rect.perimeter()}.\")\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "4. Implement a class to represent a circle, holding the radius and $x$ and $y$ coordinates of center of the object. Implement functions that compute the area and perimeter of the rectangle. Make all data members private and privide accessors to retrieve values of data members. "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "5. Implement a common base class for the classes implemented in 3 and 4 above which implements all common methods as not implemented functions (virtual). Re-implement your regtangle and circule classes to inherit from the base class and overload the functions accordingly. "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "6. Implement a triangle class analogous to the rectangle and circle in question 5."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "7. Add a function to the object classes, including the base, that returns a list of up to 16 pairs of $x$ and $y$ points on the parameter of the object. "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "8. Add a function to the object classes, including the base, that tests if a given set of $x$ and $y$ coordinates are inside of the object. You'll have to think through how to determine if a set of coordinates are inside an object for each object type."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "9. Add a function in the base class of the object classes that returns true/false testing that the object overlaps with another object."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "10. Copy the `Canvas` class from lecture to in a python file creating a `paint` module. Copy your classes from above into the module and implement paint functions. Implement a `CompoundShape` class. Create a simple drawing demonstrating that all of your classes are working."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "11. Create a `RasterDrawing` class. Demonstrate that you can create a drawing made of several shapes, paint the drawing, modify the drawing, and paint it again. "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": []
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "12. Implement the ability to load/save raster drawings and demonstate that your method works. One way to implement this ability:\n",
+ "\n",
+ " * Overload `__repr__` functions of all objects to return strings of the python code that would construct the object.\n",
+ " \n",
+ " * In the save method of raster drawing class, store the representations into the file.\n",
+ " * Write a loader function that reads the file and uses `eval` to instantiate the object.\n",
+ "\n",
+ "For example:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "class foo:\n",
+ " def __init__(self,a,b=None):\n",
+ " self.a=a\n",
+ " self.b=b\n",
+ " \n",
+ " def __repr__(self):\n",
+ " return \"foo(\"+repr(self.a)+\",\"+repr(self.b)+\")\"\n",
+ " \n",
+ " def save(self,filename):\n",
+ " f=open(filename,\"w\")\n",
+ " f.write(self.__repr__())\n",
+ " f.close()\n",
+ " \n",
+ " \n",
+ "def foo_loader(filename):\n",
+ " f=open(filename,\"r\")\n",
+ " tmp=eval(f.read())\n",
+ " f.close()\n",
+ " return tmp\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "foo(1,'hello')\n"
+ ]
+ }
+ ],
+ "source": [
+ "# Test\n",
+ "print(repr(foo(1,\"hello\")))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# Create an object and save it\n",
+ "ff=foo(1,\"hello\")\n",
+ "ff.save(\"Test.foo\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "foo(1,'hello')"
+ ]
+ }
+ ],
+ "source": [
+ "# Check contents of the saved file\n",
+ "!cat Test.foo"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "foo(1,'hello')"
+ ]
+ },
+ "execution_count": 5,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# Load the object\n",
+ "ff_reloaded=foo_loader(\"Test.foo\")\n",
+ "ff_reloaded"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "ds",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.11.14"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/Labs/Lab.4/Lab.4.ipynb b/Labs/Lab.4/Lab.4.ipynb
index 98e6e434..02b5083c 100644
--- a/Labs/Lab.4/Lab.4.ipynb
+++ b/Labs/Lab.4/Lab.4.ipynb
@@ -18,10 +18,43 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 4,
"metadata": {},
- "outputs": [],
- "source": []
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Error: Attempted to increment beyond maximum value.\n",
+ "Error: Attempted to increment beyond maximum value.\n"
+ ]
+ }
+ ],
+ "source": [
+ "class SimpleCounter:\n",
+ " def __init__(self, max_value):\n",
+ " self.count = 0\n",
+ " self.max_value = max_value\n",
+ "\n",
+ " def increment(self):\n",
+ " if self.count >= self.max_value:\n",
+ " print(\"Error: Attempted to increment beyond maximum value.\")\n",
+ " else:\n",
+ " self.count += 1\n",
+ "\n",
+ " def reset(self):\n",
+ " self.count = 0\n",
+ "\n",
+ "# --- Test ---\n",
+ "c1 = SimpleCounter(2)\n",
+ "c1.increment()\n",
+ "c1.increment()\n",
+ "c1.increment() # This will print the error\n",
+ "c1.reset()\n",
+ "c1.increment() \n",
+ "c1.increment()\n",
+ "c1.increment() # This will print the error again"
+ ]
},
{
"cell_type": "markdown",
@@ -32,10 +65,46 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 7,
"metadata": {},
- "outputs": [],
- "source": []
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Value: 2, Is Max?: False\n"
+ ]
+ }
+ ],
+ "source": [
+ "class Counter:\n",
+ " def __init__(self, max_value):\n",
+ " self.__count = 0 \n",
+ " self.__max_value = max_value\n",
+ "\n",
+ " def increment(self):\n",
+ " if self.__count >= self.__max_value:\n",
+ " print(\"Error: Attempted to increment beyond maximum value.\")\n",
+ " else:\n",
+ " self.__count += 1\n",
+ "\n",
+ " def reset(self):\n",
+ " self.__count = 0\n",
+ "\n",
+ " def get_value(self):\n",
+ " return self.__count\n",
+ "\n",
+ " def get_max(self):\n",
+ " return self.__max_value\n",
+ "\n",
+ " def is_at_max(self):\n",
+ " return self.__count == self.__max_value\n",
+ "\n",
+ "c2 = Counter(3)\n",
+ "c2.increment()\n",
+ "c2.increment()\n",
+ "print(f\"Value: {c2.get_value()}, Is Max?: {c2.is_at_max()}\")"
+ ]
},
{
"cell_type": "markdown",
@@ -46,10 +115,40 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 13,
"metadata": {},
- "outputs": [],
- "source": []
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Area: 8, Perimeter: 12.\n"
+ ]
+ }
+ ],
+ "source": [
+ "class Rectangle:\n",
+ " def __init__(self, length, width, x, y):\n",
+ " self.__length = length\n",
+ " self.__width = width\n",
+ " self.__x = x\n",
+ " self.__y = y\n",
+ "\n",
+ " def get_length(self): return self.__length\n",
+ " def get_width(self): return self.__width\n",
+ " def get_x(self): return self.__x\n",
+ " def get_y(self): return self.__y\n",
+ "\n",
+ " def area(self):\n",
+ " return self.__length * self.__width\n",
+ " \n",
+ " def perimeter(self):\n",
+ " return 2 * (self.__length + self.__width)\n",
+ " \n",
+ "# Test\n",
+ "rect = Rectangle(4, 2, 0, 0)\n",
+ "print(f\"Area: {rect.area()}, Perimeter: {rect.perimeter()}.\")\n"
+ ]
},
{
"cell_type": "markdown",
@@ -60,10 +159,44 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 2,
"metadata": {},
- "outputs": [],
- "source": []
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Circle Area: 50.27, Perimeter: 25.13\n"
+ ]
+ }
+ ],
+ "source": [
+ "import numpy as np\n",
+ "\n",
+ "class Circle:\n",
+ " def __init__(self, radius, x, y):\n",
+ " self.__radius = radius\n",
+ " self.__x = x\n",
+ " self.__y = y\n",
+ "\n",
+ " def get_radius(self):\n",
+ " return self.__radius\n",
+ " def get_x(self):\n",
+ " return self.__x\n",
+ " def get_y(self):\n",
+ " return self.__y\n",
+ "\n",
+ " def area(self):\n",
+ " return np.pi * (self.__radius ** 2)\n",
+ "\n",
+ " def perimeter(self):\n",
+ " return 2 * np.pi * self.__radius\n",
+ "\n",
+ "# Test\n",
+ "c = Circle(4, 5, 5)\n",
+ "print(f\"Circle Area: {c.area():.2f}, Perimeter: {c.perimeter():.2f}\")\n",
+ " "
+ ]
},
{
"cell_type": "markdown",
@@ -74,10 +207,50 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 4,
"metadata": {},
- "outputs": [],
- "source": []
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Rectangle Area: 12.00\n",
+ "Circle Area: 12.57\n"
+ ]
+ }
+ ],
+ "source": [
+ "class Shape:\n",
+ " def area(self):\n",
+ " raise NotImplementedError\n",
+ " \n",
+ " def perimeter(self):\n",
+ " raise NotImplementedError\n",
+ "\n",
+ "class Rectangle(Shape):\n",
+ " def __init__(self, length, width, x, y):\n",
+ " self.__length = length\n",
+ " self.__width = width\n",
+ " self.__x = x\n",
+ " self.__y = y\n",
+ " \n",
+ " def area(self): return self.__length * self.__width\n",
+ " def perimeter(self): return 2 * (self.__length + self.__width)\n",
+ "\n",
+ "class Circle(Shape):\n",
+ " def __init__(self, radius, cx, cy):\n",
+ " self.__radius = radius\n",
+ " self.__cx = cx\n",
+ " self.__cy = cy\n",
+ " \n",
+ " def area(self): return np.pi * (self.__radius ** 2)\n",
+ " def perimeter(self): return 2 * np.pi * self.__radius\n",
+ "\n",
+ "# --- Test ---\n",
+ "shapes = [Rectangle(4, 3, 0, 0), Circle(2, 0, 0)]\n",
+ "for s in shapes:\n",
+ " print(f\"{s.__class__.__name__} Area: {s.area():.2f}\")"
+ ]
},
{
"cell_type": "markdown",
@@ -88,10 +261,36 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 6,
"metadata": {},
- "outputs": [],
- "source": []
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Triangle Area: 6.0, Perimeter: 12.0\n"
+ ]
+ }
+ ],
+ "source": [
+ "class Triangle(Shape):\n",
+ " def __init__(self, base, height, x, y):\n",
+ " self.__base = base\n",
+ " self.__height = height\n",
+ " self.__x = x\n",
+ " self.__y = y\n",
+ " \n",
+ " def area(self): \n",
+ " return 0.5 * self.__base * self.__height\n",
+ " \n",
+ " def perimeter(self): \n",
+ " hypotenuse = np.sqrt(self.__base**2 + self.__height**2)\n",
+ " return self.__base + self.__height + hypotenuse\n",
+ "\n",
+ "# --- Test ---\n",
+ "t = Triangle(3, 4, 0, 0)\n",
+ "print(f\"Triangle Area: {t.area()}, Perimeter: {t.perimeter()}\")"
+ ]
},
{
"cell_type": "markdown",
@@ -102,10 +301,73 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 7,
"metadata": {},
"outputs": [],
- "source": []
+ "source": [
+ "import math\n",
+ "\n",
+ "class Shape:\n",
+ " def area(self): raise NotImplementedError\n",
+ " def perimeter(self): raise NotImplementedError\n",
+ " \n",
+ " def get_perimeter_points(self): raise NotImplementedError\n",
+ "\n",
+ "class Rectangle(Shape):\n",
+ " def __init__(self, length, width, x, y):\n",
+ " self.__length = length\n",
+ " self.__width = width\n",
+ " self.__x = x\n",
+ " self.__y = y\n",
+ " \n",
+ " def area(self): return self.__length * self.__width\n",
+ " def perimeter(self): return 2 * (self.__length + self.__width)\n",
+ " \n",
+ " def get_perimeter_points(self):\n",
+ " points = []\n",
+ " for i in range(4):\n",
+ " points.append((self.__x + (self.__length * i / 4), self.__y)) # Top edge\n",
+ " points.append((self.__x + (self.__length * i / 4), self.__y - self.__width)) # Bottom edge\n",
+ " points.append((self.__x, self.__y - (self.__width * i / 4))) # Left edge\n",
+ " points.append((self.__x + self.__length, self.__y - (self.__width * i / 4))) # Right edge\n",
+ " return points\n",
+ "\n",
+ "class Circle(Shape):\n",
+ " def __init__(self, radius, cx, cy):\n",
+ " self.__radius = radius\n",
+ " self.__cx = cx\n",
+ " self.__cy = cy\n",
+ " \n",
+ " def area(self): return math.pi * (self.__radius ** 2)\n",
+ " def perimeter(self): return 2 * math.pi * self.__radius\n",
+ " \n",
+ " def get_perimeter_points(self):\n",
+ " points = []\n",
+ " for i in range(16):\n",
+ " angle = 2 * math.pi * i / 16\n",
+ " px = self.__cx + self.__radius * math.cos(angle)\n",
+ " py = self.__cy + self.__radius * math.sin(angle)\n",
+ " points.append((px, py))\n",
+ " return points\n",
+ "\n",
+ "class Triangle(Shape):\n",
+ " def __init__(self, base, height, x, y):\n",
+ " self.__base = base\n",
+ " self.__height = height\n",
+ " self.__x = x\n",
+ " self.__y = y\n",
+ " \n",
+ " def area(self): return 0.5 * self.__base * self.__height\n",
+ " def perimeter(self): return self.__base + self.__height + math.sqrt(self.__base**2 + self.__height**2)\n",
+ " \n",
+ " def get_perimeter_points(self):\n",
+ " points = []\n",
+ " for i in range(5):\n",
+ " points.append((self.__x + (self.__base * i / 5), self.__y)) \n",
+ " points.append((self.__x, self.__y + (self.__height * i / 5))) \n",
+ " points.append((self.__x + (self.__base * i / 5), self.__y + self.__height - (self.__height * i / 5))) \n",
+ " points.append((self.__x + self.__base, self.__y)) "
+ ]
},
{
"cell_type": "markdown",
@@ -116,10 +378,90 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 8,
"metadata": {},
"outputs": [],
- "source": []
+ "source": [
+ "class Shape:\n",
+ " def area(self): raise NotImplementedError\n",
+ " def perimeter(self): raise NotImplementedError\n",
+ " def get_perimeter_points(self): raise NotImplementedError\n",
+ " def contains(self, x, y): raise NotImplementedError\n",
+ "\n",
+ "class Rectangle(Shape):\n",
+ " def __init__(self, length, width, x, y):\n",
+ " self.length = length\n",
+ " self.width = width\n",
+ " self.x = x\n",
+ " self.y = y\n",
+ " \n",
+ " def area(self): return self.length * self.width\n",
+ " def perimeter(self): return 2 * (self.length + self.width)\n",
+ " \n",
+ " def get_perimeter_points(self):\n",
+ " points = []\n",
+ " for i in range(4):\n",
+ " points.append((self.x + (self.length * i / 4), self.y))\n",
+ " points.append((self.x + (self.length * i / 4), self.y - self.width))\n",
+ " points.append((self.x, self.y - (self.width * i / 4)))\n",
+ " points.append((self.x + self.length, self.y - (self.width * i / 4)))\n",
+ " return points\n",
+ "\n",
+ " def contains(self, px, py):\n",
+ " return (self.x <= px <= self.x + self.length) and (self.y - self.width <= py <= self.y)\n",
+ "\n",
+ " def __repr__(self):\n",
+ " return f\"Rectangle({self.length}, {self.width}, {self.x}, {self.y})\"\n",
+ "\n",
+ "class Circle(Shape):\n",
+ " def __init__(self, radius, cx, cy):\n",
+ " self.radius = radius\n",
+ " self.cx = cx\n",
+ " self.cy = cy\n",
+ " \n",
+ " def area(self): return math.pi * (self.radius ** 2)\n",
+ " def perimeter(self): return 2 * math.pi * self.radius\n",
+ " \n",
+ " def get_perimeter_points(self):\n",
+ " points = []\n",
+ " for i in range(16):\n",
+ " angle = 2 * math.pi * i / 16\n",
+ " points.append((self.cx + self.radius * math.cos(angle), self.cy + self.radius * math.sin(angle)))\n",
+ " return points\n",
+ "\n",
+ " def contains(self, px, py):\n",
+ " return math.sqrt((px - self.cx)**2 + (py - self.cy)**2) <= self.radius\n",
+ "\n",
+ " def __repr__(self):\n",
+ " return f\"Circle({self.radius}, {self.cx}, {self.cy})\"\n",
+ "\n",
+ "class Triangle(Shape):\n",
+ " def __init__(self, base, height, x, y):\n",
+ " self.base = base\n",
+ " self.height = height\n",
+ " self.x = x\n",
+ " self.y = y\n",
+ " \n",
+ " def area(self): return 0.5 * self.base * self.height\n",
+ " def perimeter(self): return self.base + self.height + math.sqrt(self.base**2 + self.height**2)\n",
+ " \n",
+ " def get_perimeter_points(self):\n",
+ " points = []\n",
+ " for i in range(5):\n",
+ " points.append((self.x + (self.base * i / 5), self.y))\n",
+ " points.append((self.x, self.y + (self.height * i / 5)))\n",
+ " points.append((self.x + (self.base * i / 5), self.y + self.height - (self.height * i / 5)))\n",
+ " points.append((self.x + self.base, self.y))\n",
+ " return points\n",
+ "\n",
+ " def contains(self, px, py):\n",
+ " if not ((self.x <= px <= self.x + self.base) and (self.y <= py <= self.y + self.height)):\n",
+ " return False\n",
+ " return py <= self.y + self.height - (self.height / self.base) * (px - self.x)\n",
+ "\n",
+ " def __repr__(self):\n",
+ " return f\"Triangle({self.base}, {self.height}, {self.x}, {self.y})\""
+ ]
},
{
"cell_type": "markdown",
@@ -130,10 +472,87 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 10,
"metadata": {},
- "outputs": [],
- "source": []
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "True\n"
+ ]
+ }
+ ],
+ "source": [
+ "class Shape:\n",
+ " def area(self): raise NotImplementedError\n",
+ " def perimeter(self): raise NotImplementedError\n",
+ " def get_perimeter_points(self): raise NotImplementedError\n",
+ " def contains(self, x, y): raise NotImplementedError\n",
+ " \n",
+ " def overlaps(self, other):\n",
+ " for px, py in other.get_perimeter_points():\n",
+ " if self.contains(px, py):\n",
+ " return True\n",
+ " \n",
+ " for px, py in self.get_perimeter_points():\n",
+ " if other.contains(px, py):\n",
+ " return True\n",
+ " \n",
+ " return False\n",
+ "\n",
+ "class Rectangle(Shape):\n",
+ " def __init__(self, length, width, x, y):\n",
+ " self.length = length\n",
+ " self.width = width\n",
+ " self.x = x\n",
+ " self.y = y\n",
+ " \n",
+ " def area(self): return self.length * self.width\n",
+ " def perimeter(self): return 2 * (self.length + self.width)\n",
+ " \n",
+ " def get_perimeter_points(self):\n",
+ " points = []\n",
+ " for i in range(4):\n",
+ " points.append((self.x + (self.length * i / 4), self.y))\n",
+ " points.append((self.x + (self.length * i / 4), self.y - self.width))\n",
+ " points.append((self.x, self.y - (self.width * i / 4)))\n",
+ " points.append((self.x + self.length, self.y - (self.width * i / 4)))\n",
+ " return points\n",
+ "\n",
+ " def contains(self, px, py):\n",
+ " return (self.x <= px <= self.x + self.length) and (self.y - self.width <= py <= self.y)\n",
+ "\n",
+ " def __repr__(self):\n",
+ " return f\"Rectangle({self.length}, {self.width}, {self.x}, {self.y})\"\n",
+ "\n",
+ "class Circle(Shape):\n",
+ " def __init__(self, radius, cx, cy):\n",
+ " self.radius = radius\n",
+ " self.cx = cx\n",
+ " self.cy = cy\n",
+ " \n",
+ " def area(self): return math.pi * (self.radius ** 2)\n",
+ " def perimeter(self): return 2 * math.pi * self.radius\n",
+ " \n",
+ " def get_perimeter_points(self):\n",
+ " points = []\n",
+ " for i in range(16):\n",
+ " angle = 2 * math.pi * i / 16\n",
+ " points.append((self.cx + self.radius * math.cos(angle), self.cy + self.radius * math.sin(angle)))\n",
+ " return points\n",
+ "\n",
+ " def contains(self, px, py):\n",
+ " return math.sqrt((px - self.cx)**2 + (py - self.cy)**2) <= self.radius\n",
+ "\n",
+ " def __repr__(self):\n",
+ " return f\"Circle({self.radius}, {self.cx}, {self.cy})\"\n",
+ "\n",
+ "# --- Q9 Test ---\n",
+ "r1 = Rectangle(10, 10, 0, 10)\n",
+ "c1 = Circle(5, 5, 5)\n",
+ "print(r1.overlaps(c1))"
+ ]
},
{
"cell_type": "markdown",
@@ -144,10 +563,184 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 11,
"metadata": {},
- "outputs": [],
- "source": []
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " \n",
+ " \n",
+ " ########### \n",
+ " ########### \n",
+ " ########### O \n",
+ " ########### OOOOO \n",
+ " ########### OOOOOOO \n",
+ " ########### OOOOOOO \n",
+ " OOOOOOOOO \n",
+ " OOOOOOO \n",
+ " OOOOOOO \n",
+ " OOOOO \n",
+ " O \n",
+ " \n",
+ " \n",
+ " \n",
+ " \n",
+ " \n",
+ " \n",
+ " \n"
+ ]
+ }
+ ],
+ "source": [
+ "import math\n",
+ "\n",
+ "class Canvas:\n",
+ " def __init__(self, width, height):\n",
+ " self.width = width\n",
+ " self.height = height\n",
+ " self.data = [[' '] * width for i in range(height)]\n",
+ "\n",
+ " def set_pixel(self, row, col, char='*'):\n",
+ " if 0 <= row < self.height and 0 <= col < self.width:\n",
+ " self.data[row][col] = char\n",
+ "\n",
+ " def get_pixel(self, row, col):\n",
+ " return self.data[row][col]\n",
+ " \n",
+ " def clear_canvas(self):\n",
+ " self.data = [[' '] * self.width for i in range(self.height)]\n",
+ " \n",
+ " def display(self):\n",
+ " print(\"\\n\".join([\"\".join(row) for row in self.data]))\n",
+ "\n",
+ "class Shape:\n",
+ " def __init__(self, name=\"\", char=\"*\"):\n",
+ " self.name = name\n",
+ " self.char = char\n",
+ " \n",
+ " def area(self): raise NotImplementedError\n",
+ " def perimeter(self): raise NotImplementedError\n",
+ " def get_perimeter_points(self): raise NotImplementedError\n",
+ " def contains(self, x, y): raise NotImplementedError\n",
+ " \n",
+ " def overlaps(self, other):\n",
+ " for px, py in other.get_perimeter_points():\n",
+ " if self.contains(px, py): return True\n",
+ " for px, py in self.get_perimeter_points():\n",
+ " if other.contains(px, py): return True\n",
+ " return False\n",
+ " \n",
+ " def paint(self, canvas):\n",
+ " for row in range(canvas.height):\n",
+ " for col in range(canvas.width):\n",
+ " if self.contains(col, row):\n",
+ " canvas.set_pixel(row, col, self.char)\n",
+ "\n",
+ "class Rectangle(Shape):\n",
+ " def __init__(self, length, width, x, y, name=\"\", char=\"*\"):\n",
+ " super().__init__(name, char)\n",
+ " self.length = length\n",
+ " self.width = width\n",
+ " self.x = x\n",
+ " self.y = y\n",
+ " \n",
+ " def area(self): return self.length * self.width\n",
+ " def perimeter(self): return 2 * (self.length + self.width)\n",
+ " \n",
+ " def get_perimeter_points(self):\n",
+ " points = []\n",
+ " for i in range(4):\n",
+ " points.append((self.x + (self.length * i / 4), self.y))\n",
+ " points.append((self.x + (self.length * i / 4), self.y + self.width))\n",
+ " points.append((self.x, self.y + (self.width * i / 4)))\n",
+ " points.append((self.x + self.length, self.y + (self.width * i / 4)))\n",
+ " return points\n",
+ "\n",
+ " def contains(self, px, py):\n",
+ " return (self.x <= px <= self.x + self.length) and (self.y <= py <= self.y + self.width)\n",
+ "\n",
+ " def __repr__(self):\n",
+ " return f\"Rectangle({self.length}, {self.width}, {self.x}, {self.y}, '{self.name}', '{self.char}')\"\n",
+ "\n",
+ "class Circle(Shape):\n",
+ " def __init__(self, radius, cx, cy, name=\"\", char=\"*\"):\n",
+ " super().__init__(name, char)\n",
+ " self.radius = radius\n",
+ " self.cx = cx\n",
+ " self.cy = cy\n",
+ " \n",
+ " def area(self): return math.pi * (self.radius ** 2)\n",
+ " def perimeter(self): return 2 * math.pi * self.radius\n",
+ " \n",
+ " def get_perimeter_points(self):\n",
+ " points = []\n",
+ " for i in range(16):\n",
+ " angle = 2 * math.pi * i / 16\n",
+ " points.append((self.cx + self.radius * math.cos(angle), self.cy + self.radius * math.sin(angle)))\n",
+ " return points\n",
+ "\n",
+ " def contains(self, px, py):\n",
+ " return math.sqrt((px - self.cx)**2 + (py - self.cy)**2) <= self.radius\n",
+ "\n",
+ " def __repr__(self):\n",
+ " return f\"Circle({self.radius}, {self.cx}, {self.cy}, '{self.name}', '{self.char}')\"\n",
+ "\n",
+ "class Triangle(Shape):\n",
+ " def __init__(self, base, height, x, y, name=\"\", char=\"*\"):\n",
+ " super().__init__(name, char)\n",
+ " self.base = base\n",
+ " self.height = height\n",
+ " self.x = x\n",
+ " self.y = y\n",
+ " \n",
+ " def area(self): return 0.5 * self.base * self.height\n",
+ " def perimeter(self): return self.base + self.height + math.sqrt(self.base**2 + self.height**2)\n",
+ " \n",
+ " def get_perimeter_points(self):\n",
+ " points = []\n",
+ " for i in range(5):\n",
+ " points.append((self.x + (self.base * i / 5), self.y))\n",
+ " points.append((self.x, self.y + (self.height * i / 5)))\n",
+ " points.append((self.x + (self.base * i / 5), self.y + self.height - (self.height * i / 5)))\n",
+ " points.append((self.x + self.base, self.y))\n",
+ " return points\n",
+ "\n",
+ " def contains(self, px, py):\n",
+ " if not ((self.x <= px <= self.x + self.base) and (self.y <= py <= self.y + self.height)):\n",
+ " return False\n",
+ " return py <= self.y + self.height - (self.height / self.base) * (px - self.x)\n",
+ "\n",
+ " def __repr__(self):\n",
+ " return f\"Triangle({self.base}, {self.height}, {self.x}, {self.y}, '{self.name}', '{self.char}')\"\n",
+ "\n",
+ "class CompoundShape(Shape):\n",
+ " def __init__(self, shapes, name=\"\", char=\"*\"):\n",
+ " super().__init__(name, char)\n",
+ " self.shapes = shapes\n",
+ " \n",
+ " def get_perimeter_points(self):\n",
+ " points = []\n",
+ " for s in self.shapes: \n",
+ " points.extend(s.get_perimeter_points())\n",
+ " return points\n",
+ " \n",
+ " def contains(self, x, y):\n",
+ " return any(s.contains(x, y) for s in self.shapes)\n",
+ " \n",
+ " def paint(self, canvas):\n",
+ " for s in self.shapes:\n",
+ " s.paint(canvas)\n",
+ " \n",
+ " def __repr__(self):\n",
+ " return f\"CompoundShape({self.shapes}, '{self.name}', '{self.char}')\"\n",
+ "\n",
+ "my_canvas = Canvas(40, 20)\n",
+ "comp = CompoundShape([Rectangle(10, 5, 2, 2, char=\"#\"), Circle(4, 25, 8, char=\"O\")])\n",
+ "comp.paint(my_canvas)\n",
+ "my_canvas.display()"
+ ]
},
{
"cell_type": "markdown",
@@ -158,10 +751,100 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 12,
"metadata": {},
- "outputs": [],
- "source": []
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " \n",
+ " \n",
+ " \n",
+ " \n",
+ " \n",
+ " O TTTTTTTTT \n",
+ " OOOOOOO TTTTTTTT \n",
+ " OOOOOOOOO TTTTTTT \n",
+ " OOOOOOOOO TTTTTT \n",
+ " OOOOOOOOO TTTTT \n",
+ " OOOOOOOOOOO TTTT \n",
+ " OOOOOOOOO TTT \n",
+ " OOOOOOOOO TT \n",
+ " OOOOOOOOO T \n",
+ " OOOOOOO \n",
+ " O \n",
+ " \n",
+ " \n",
+ " \n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ " \n",
+ " O \n",
+ " OOOOOOO \n",
+ " OOOOOOOOOOO \n",
+ " OOOOOOOOOOOOO TTTTTTTTT \n",
+ " OOOOOOOOOOOOO TTTTTTTT \n",
+ " OOOOOOOOOOOOOOO TTTTTTT \n",
+ " OOOOOOOOOOOOOOO TTTTTT \n",
+ " OOOOOOOOOOOOOOO TTTTT \n",
+ " OOOOOOOOOOOOOOOOO TTTT \n",
+ " OOOOOOOOOOOOOOO TTT \n",
+ " OOOOOOOOOOOOOOO TT \n",
+ " OOOOOOOOOOOOOOO T \n",
+ " OOOOOOOOOOOOO \n",
+ " OOOOOOOOOOOOO \n",
+ " OOOOOOOOOOO \n",
+ " OOOOOOO \n",
+ " O \n",
+ " \n"
+ ]
+ }
+ ],
+ "source": [
+ "class RasterDrawing:\n",
+ " def __init__(self):\n",
+ " self.shapes = dict()\n",
+ " self.shape_names = list()\n",
+ " \n",
+ " def add_shape(self, shape):\n",
+ " if shape.name == \"\":\n",
+ " shape.name = self.assign_name()\n",
+ " self.shapes[shape.name] = shape\n",
+ " self.shape_names.append(shape.name)\n",
+ "\n",
+ " def update(self, canvas):\n",
+ " canvas.clear_canvas()\n",
+ " self.paint(canvas)\n",
+ " \n",
+ " def paint(self, canvas):\n",
+ " for shape_name in self.shape_names:\n",
+ " self.shapes[shape_name].paint(canvas)\n",
+ " \n",
+ " def assign_name(self):\n",
+ " name_base = \"shape\"\n",
+ " name = name_base + \"_0\"\n",
+ " i = 1\n",
+ " while name in self.shapes:\n",
+ " name = name_base + \"_\" + str(i)\n",
+ " i += 1\n",
+ " return name\n",
+ "\n",
+ "drawing_canvas = Canvas(40, 20)\n",
+ "rd = RasterDrawing()\n",
+ "rd.add_shape(Circle(5, 10, 10, char=\"O\"))\n",
+ "rd.add_shape(Triangle(8, 8, 20, 5, char=\"T\"))\n",
+ "rd.paint(drawing_canvas)\n",
+ "drawing_canvas.display()\n",
+ "\n",
+ "print(\"\\n\")\n",
+ "\n",
+ "rd.shapes[\"shape_0\"].radius = 8\n",
+ "rd.update(drawing_canvas)\n",
+ "drawing_canvas.display()"
+ ]
},
{
"cell_type": "markdown",
@@ -177,6 +860,65 @@
"For example:"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " \n",
+ " \n",
+ " \n",
+ " \n",
+ " \n",
+ " ########### \n",
+ " ########### \n",
+ " ########### O \n",
+ " ########### OOOOO \n",
+ " ########### OOOOO \n",
+ " ########### OOOOOOO \n",
+ " OOOOO \n",
+ " OOOOO \n",
+ " O \n",
+ " \n",
+ " \n",
+ " \n",
+ " \n",
+ " \n",
+ " \n"
+ ]
+ }
+ ],
+ "source": [
+ "class RasterDrawingSaveable(RasterDrawing):\n",
+ " def save(self, filename):\n",
+ " with open(filename, \"w\") as f:\n",
+ " f.write(repr(list(self.shapes.values())))\n",
+ " \n",
+ "def load_drawing(filename):\n",
+ " with open(filename, \"r\") as f:\n",
+ " shapes_list = eval(f.read())\n",
+ " \n",
+ " new_drawing = RasterDrawingSaveable()\n",
+ " for s in shapes_list:\n",
+ " new_drawing.add_shape(s)\n",
+ " return new_drawing\n",
+ "\n",
+ "drawing = RasterDrawingSaveable()\n",
+ "drawing.add_shape(Rectangle(10, 5, 5, 5, char=\"#\"))\n",
+ "drawing.add_shape(Circle(3, 25, 10, char=\"O\"))\n",
+ "\n",
+ "drawing.save(\"lab4_drawing.txt\")\n",
+ "\n",
+ "loaded = load_drawing(\"lab4_drawing.txt\")\n",
+ "test_canvas = Canvas(40, 20)\n",
+ "loaded.paint(test_canvas)\n",
+ "test_canvas.display()"
+ ]
+ },
{
"cell_type": "code",
"execution_count": 1,
@@ -276,7 +1018,7 @@
],
"metadata": {
"kernelspec": {
- "display_name": "Python 3 (ipykernel)",
+ "display_name": "ds",
"language": "python",
"name": "python3"
},
diff --git a/Labs/Lab.4/lab4_drawing.txt b/Labs/Lab.4/lab4_drawing.txt
new file mode 100644
index 00000000..0f104b0f
--- /dev/null
+++ b/Labs/Lab.4/lab4_drawing.txt
@@ -0,0 +1 @@
+[Rectangle(10, 5, 5, 5, 'shape_0', '#'), Circle(3, 25, 10, 'shape_1', 'O')]
\ No newline at end of file
diff --git a/Labs/Lab.5/Lab.5.ipynb b/Labs/Lab.5/Lab.5.ipynb
index cba02709..e0dfc58d 100644
--- a/Labs/Lab.5/Lab.5.ipynb
+++ b/Labs/Lab.5/Lab.5.ipynb
@@ -20,7 +20,91 @@
" * Matrix instances `M` can be indexed with `M[i][j]` and `M[i,j]`.\n",
" * Matrix assignment works in 2 ways:\n",
" 1. If `M_1` and `M_2` are `matrix` instances `M_1=M_2` sets the values of `M_1` to those of `M_2`, if they are the same size. Error otherwise.\n",
- " 2. In example above `M_2` can be a list of lists of correct size.\n"
+ " 2. In example above `M_2` can be a list of lists of correct size."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 19,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "--- Q1 Tests ---\n",
+ "Q1 Initialization, Indexing, and Assignment tests passed!\n"
+ ]
+ }
+ ],
+ "source": [
+ "class matrix:\n",
+ " def __init__(self, *args):\n",
+ " # I'm checking if the arguments are two integers (n, m) to create a zero matrix\n",
+ " if len(args) == 2 and isinstance(args[0], int) and isinstance(args[1], int):\n",
+ " self.n, self.m = args[0], args[1]\n",
+ " self.data = [[0 for _ in range(self.m)] for _ in range(self.n)]\n",
+ " \n",
+ " # Or if the argument is a list of lists, I need to copy it over\n",
+ " elif len(args) == 1 and isinstance(args[0], list):\n",
+ " input_list = args[0]\n",
+ " self.n = len(input_list)\n",
+ " self.m = len(input_list[0]) if self.n > 0 else 0\n",
+ " \n",
+ " # Making sure every row has the same number of columns to catch bad inputs\n",
+ " if not all(isinstance(row, list) and len(row) == self.m for row in input_list):\n",
+ " raise ValueError(\"All rows must have the same number of columns.\")\n",
+ " \n",
+ " self.data = [[val for val in row] for row in input_list]\n",
+ " else:\n",
+ " raise ValueError(\"Need either (n, m) or a list of lists.\")\n",
+ "\n",
+ " def __getitem__(self, index):\n",
+ " # Python passes M[i,j] as a tuple, so I unpack it\n",
+ " if isinstance(index, tuple):\n",
+ " i, j = index\n",
+ " return self.data[i][j]\n",
+ " # For M[i][j], the first brackets pass an int, returning the whole row\n",
+ " elif isinstance(index, int):\n",
+ " return self.data[index]\n",
+ " else:\n",
+ " raise TypeError(\"Index needs to be an int or tuple.\")\n",
+ "\n",
+ " def __setitem__(self, index, value):\n",
+ " # Allowing value assignment like M[0,1] = 5\n",
+ " if isinstance(index, tuple):\n",
+ " i, j = index\n",
+ " self.data[i][j] = value\n",
+ " elif isinstance(index, int):\n",
+ " self.data[index] = value\n",
+ "\n",
+ " def assign(self, other):\n",
+ " # The assignment operator '=' cannot be overloaded in Python. \n",
+ " # Writing M_1 = M_2 just creates a reference copy, it doesn't modify the object's values.\n",
+ " # So, I'm using an assign() method to handle the specific assignment logic requested.\n",
+ " if isinstance(other, matrix):\n",
+ " if self.n != other.n or self.m != other.m:\n",
+ " raise ValueError(\"Size mismatch.\")\n",
+ " self.data = [[val for val in row] for row in other.data]\n",
+ " elif isinstance(other, list):\n",
+ " if len(other) != self.n or not all(len(row) == self.m for row in other):\n",
+ " raise ValueError(\"Size mismatch.\")\n",
+ " self.data = [[val for val in row] for row in other]\n",
+ "\n",
+ " def __str__(self):\n",
+ " return '\\n'.join([str(row) for row in self.data])\n",
+ "\n",
+ "# --- EXPLICIT TESTS FOR Q1 ---\n",
+ "print(\"--- Q1 Tests ---\")\n",
+ "m1 = matrix(2, 3)\n",
+ "assert m1.data == [[0, 0, 0], [0, 0, 0]]\n",
+ "m2 = matrix([[1, 2], [3, 4]])\n",
+ "assert m2.data == [[1, 2], [3, 4]]\n",
+ "\n",
+ "m_target = matrix(2, 2)\n",
+ "m_target.assign(m2)\n",
+ "assert m_target.data == [[1, 2], [3, 4]]\n",
+ "print(\"Q1 Initialization, Indexing, and Assignment tests passed!\")"
]
},
{
@@ -37,6 +121,100 @@
" "
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 20,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "--- Q2 Tests ---\n",
+ "Q2 Properties and Slicing tests passed!\n"
+ ]
+ }
+ ],
+ "source": [
+ "class matrix:\n",
+ " # --- Q1 Methods ---\n",
+ " def __init__(self, *args):\n",
+ " if len(args) == 2 and isinstance(args[0], int) and isinstance(args[1], int):\n",
+ " self.n, self.m = args[0], args[1]\n",
+ " self.data = [[0 for _ in range(self.m)] for _ in range(self.n)]\n",
+ " elif len(args) == 1 and isinstance(args[0], list):\n",
+ " input_list = args[0]\n",
+ " self.n, self.m = len(input_list), (len(input_list[0]) if len(input_list) > 0 else 0)\n",
+ " if not all(isinstance(row, list) and len(row) == self.m for row in input_list):\n",
+ " raise ValueError(\"All rows must have the same number of columns.\")\n",
+ " self.data = [[val for val in row] for row in input_list]\n",
+ "\n",
+ " def __getitem__(self, index):\n",
+ " # Upgraded to handle slicing. If the index contains a 'slice' object (like 0:2),\n",
+ " # I use list comprehensions to grab those specific chunks.\n",
+ " if isinstance(index, tuple):\n",
+ " i, j = index\n",
+ " if isinstance(i, slice) or isinstance(j, slice):\n",
+ " if isinstance(i, int): i = slice(i, i+1)\n",
+ " if isinstance(j, int): j = slice(j, j+1)\n",
+ " return matrix([row[j] for row in self.data[i]])\n",
+ " return self.data[i][j]\n",
+ " elif isinstance(index, slice):\n",
+ " return matrix(self.data[index])\n",
+ " elif isinstance(index, int):\n",
+ " return self.data[index]\n",
+ "\n",
+ " def __setitem__(self, index, value):\n",
+ " if isinstance(index, tuple): i, j = index; self.data[i][j] = value\n",
+ " elif isinstance(index, int): self.data[index] = value\n",
+ "\n",
+ " def assign(self, other):\n",
+ " if isinstance(other, matrix):\n",
+ " if self.shape() != other.shape(): raise ValueError(\"Size mismatch.\")\n",
+ " self.data = [[val for val in row] for row in other.data]\n",
+ " elif isinstance(other, list):\n",
+ " self.data = [[val for val in row] for row in other]\n",
+ " \n",
+ " def __str__(self): return '\\n'.join([str(row) for row in self.data])\n",
+ "\n",
+ " # --- Q2 New Methods ---\n",
+ " def shape(self):\n",
+ " # Simply returns the dimensions stored during init\n",
+ " return (self.n, self.m)\n",
+ "\n",
+ " def transpose(self):\n",
+ " # Swapping rows and columns to create the transposed version\n",
+ " t_data = [[self.data[i][j] for i in range(self.n)] for j in range(self.m)]\n",
+ " return matrix(t_data)\n",
+ "\n",
+ " def row(self, n):\n",
+ " # Wrapping the row in an extra list bracket so it becomes a 1xm matrix\n",
+ " return matrix([self.data[n]])\n",
+ "\n",
+ " def column(self, n):\n",
+ " # Grabbing the nth element of every row to build an nx1 matrix\n",
+ " return matrix([[self.data[i][n]] for i in range(self.n)])\n",
+ "\n",
+ " def to_list(self):\n",
+ " # Using a list comprehension to return a clean copy, avoiding reference bugs\n",
+ " return [[val for val in row] for row in self.data]\n",
+ "\n",
+ " def block(self, n_0, n_1, m_0, m_1):\n",
+ " # Standard Python slicing: grab rows m_0 to m_1, then slice those rows from cols n_0 to n_1\n",
+ " return matrix([r[n_0:n_1] for r in self.data[m_0:m_1]])\n",
+ "\n",
+ "# --- EXPLICIT TESTS FOR Q2 ---\n",
+ "print(\"--- Q2 Tests ---\")\n",
+ "m_test = matrix([[1, 2, 3], [4, 5, 6], [7, 8, 9]])\n",
+ "assert m_test.shape() == (3, 3)\n",
+ "assert m_test.transpose().data == [[1, 4, 7], [2, 5, 8], [3, 6, 9]]\n",
+ "assert m_test.row(1).data == [[4, 5, 6]]\n",
+ "assert m_test.column(2).data == [[3], [6], [9]]\n",
+ "assert m_test.to_list() == [[1, 2, 3], [4, 5, 6], [7, 8, 9]]\n",
+ "assert m_test.block(1, 3, 0, 2).data == [[2, 3], [5, 6]]\n",
+ "print(\"Q2 Properties and Slicing tests passed!\")"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -47,6 +225,49 @@
" * `eye(n)`: returns the n by n identity matrix."
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 21,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "--- Q3 Tests ---\n",
+ "Q3 Special Matrices tests passed!\n"
+ ]
+ }
+ ],
+ "source": [
+ "# --- Q3 Standalone Functions ---\n",
+ "\n",
+ "def constant(n, m, c):\n",
+ " # Creating an n by m matrix filled with the float version of c\n",
+ " val = float(c)\n",
+ " return matrix([[val for j in range(m)] for i in range(n)])\n",
+ "\n",
+ "def zeros(n, m):\n",
+ " # Reusing the constant function to avoid rewriting the same loops\n",
+ " return constant(n, m, 0.0)\n",
+ "\n",
+ "def ones(n, m):\n",
+ " # Same trick, reusing constant for 1.0\n",
+ " return constant(n, m, 1.0)\n",
+ "\n",
+ "def eye(n):\n",
+ " # Building the identity matrix: 1.0 on the diagonal (where i == j), 0.0 elsewhere\n",
+ " return matrix([[1.0 if i == j else 0.0 for j in range(n)] for i in range(n)])\n",
+ "\n",
+ "# --- EXPLICIT TESTS FOR Q3 ---\n",
+ "print(\"--- Q3 Tests ---\")\n",
+ "assert constant(2, 2, 5).data == [[5.0, 5.0], [5.0, 5.0]]\n",
+ "assert zeros(2, 3).data == [[0.0, 0.0, 0.0], [0.0, 0.0, 0.0]]\n",
+ "assert ones(2, 2).data == [[1.0, 1.0], [1.0, 1.0]]\n",
+ "assert eye(3).data == [[1.0, 0.0, 0.0], [0.0, 1.0, 0.0], [0.0, 0.0, 1.0]]\n",
+ "print(\"Q3 Special Matrices tests passed!\")"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -74,6 +295,142 @@
" * M=N\n"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 23,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "--- Q4 & Q5 Tests ---\n",
+ "Q4 Math Methods and Q5 Overloaded Operators tests passed!\n"
+ ]
+ }
+ ],
+ "source": [
+ "class matrix:\n",
+ " # --- Q1 & Q2 Methods ---\n",
+ " def __init__(self, *args):\n",
+ " if len(args) == 2 and isinstance(args[0], int) and isinstance(args[1], int):\n",
+ " self.n, self.m = args[0], args[1]\n",
+ " self.data = [[0 for _ in range(self.m)] for _ in range(self.n)]\n",
+ " elif len(args) == 1 and isinstance(args[0], list):\n",
+ " input_list = args[0]\n",
+ " self.n, self.m = len(input_list), (len(input_list[0]) if len(input_list) > 0 else 0)\n",
+ " if not all(isinstance(row, list) and len(row) == self.m for row in input_list):\n",
+ " raise ValueError(\"All rows must have the same number of columns.\")\n",
+ " self.data = [[val for val in row] for row in input_list]\n",
+ "\n",
+ " def __getitem__(self, index):\n",
+ " if isinstance(index, tuple):\n",
+ " i, j = index\n",
+ " if isinstance(i, slice) or isinstance(j, slice):\n",
+ " if isinstance(i, int): i = slice(i, i+1)\n",
+ " if isinstance(j, int): j = slice(j, j+1)\n",
+ " return matrix([row[j] for row in self.data[i]])\n",
+ " return self.data[i][j]\n",
+ " elif isinstance(index, slice): return matrix(self.data[index])\n",
+ " elif isinstance(index, int): return self.data[index]\n",
+ "\n",
+ " def __setitem__(self, index, value):\n",
+ " if isinstance(index, tuple): i, j = index; self.data[i][j] = value\n",
+ " elif isinstance(index, int): self.data[index] = value\n",
+ "\n",
+ " def assign(self, other):\n",
+ " if isinstance(other, matrix):\n",
+ " if self.shape() != other.shape(): raise ValueError(\"Size mismatch.\")\n",
+ " self.data = [[val for val in row] for row in other.data]\n",
+ " elif isinstance(other, list):\n",
+ " self.data = [[val for val in row] for row in other]\n",
+ "\n",
+ " def shape(self): return (self.n, self.m)\n",
+ " def transpose(self): return matrix([[self.data[i][j] for i in range(self.n)] for j in range(self.m)])\n",
+ " def row(self, n): return matrix([self.data[n]])\n",
+ " def column(self, n): return matrix([[self.data[i][n]] for i in range(self.n)])\n",
+ " def to_list(self): return [[val for val in row] for row in self.data]\n",
+ " def block(self, n_0, n_1, m_0, m_1): return matrix([r[n_0:n_1] for r in self.data[m_0:m_1]])\n",
+ " def __str__(self): return '\\n'.join([str(row) for row in self.data])\n",
+ "\n",
+ " # --- Q4 Math Operations ---\n",
+ " def scalarmul(self, c):\n",
+ " return matrix([[val * c for val in row] for row in self.data])\n",
+ "\n",
+ " def add(self, N):\n",
+ " if self.shape() != N.shape(): raise ValueError(\"Addition error: Size mismatch.\")\n",
+ " return matrix([[self.data[i][j] + N.data[i][j] for j in range(self.m)] for i in range(self.n)])\n",
+ "\n",
+ " def sub(self, N):\n",
+ " if self.shape() != N.shape(): raise ValueError(\"Subtraction error: Size mismatch.\")\n",
+ " return matrix([[self.data[i][j] - N.data[i][j] for j in range(self.m)] for i in range(self.n)])\n",
+ "\n",
+ " def mat_mult(self, N):\n",
+ " if self.m != N.n: raise ValueError(\"Matrix multiplication error: M columns must equal N rows.\")\n",
+ " result_data = []\n",
+ " for i in range(self.n):\n",
+ " new_row = []\n",
+ " for j in range(N.m):\n",
+ " # taking the dot product of M's row and N's column\n",
+ " new_row.append(sum(self.data[i][k] * N.data[k][j] for k in range(self.m)))\n",
+ " result_data.append(new_row)\n",
+ " return matrix(result_data)\n",
+ "\n",
+ " def element_mult(self, N):\n",
+ " if self.shape() != N.shape(): raise ValueError(\"Element multiplication error: Size mismatch.\")\n",
+ " return matrix([[self.data[i][j] * N.data[i][j] for j in range(self.m)] for i in range(self.n)])\n",
+ "\n",
+ " def equals(self, N):\n",
+ " if not isinstance(N, matrix) or self.shape() != N.shape(): return False\n",
+ " return self.data == N.data\n",
+ "\n",
+ " # --- Q5 Operator Overloading ---\n",
+ " \n",
+ " # Python uses magic methods like __add__ to know what to do when you type '+'\n",
+ " def __add__(self, other):\n",
+ " return self.add(other)\n",
+ "\n",
+ " def __sub__(self, other):\n",
+ " return self.sub(other)\n",
+ "\n",
+ " def __mul__(self, other):\n",
+ " # I need to check if 'other' is a matrix or a scalar to decide which math rule to apply\n",
+ " if isinstance(other, matrix):\n",
+ " return self.mat_mult(other)\n",
+ " elif isinstance(other, (int, float)):\n",
+ " return self.scalarmul(other)\n",
+ "\n",
+ " def __rmul__(self, other):\n",
+ " # This handles the case where the scalar comes first, like 2 * M\n",
+ " return self.scalarmul(other)\n",
+ "\n",
+ " def __eq__(self, other):\n",
+ " # Overloads the '==' operator\n",
+ " return self.equals(other)\n",
+ " \n",
+ " # Note on M=N from Q5 requirements: Python physically cannot overload the '=' operator. \n",
+ " # So Using the assign() method implemented in Q1 to assign values without creating a reference copy.\n",
+ "\n",
+ "# --- EXPLICIT TESTS FOR Q4 & Q5 ---\n",
+ "print(\"--- Q4 & Q5 Tests ---\")\n",
+ "M1 = matrix([[1, 2], [3, 4]])\n",
+ "M2 = matrix([[5, 6], [7, 8]])\n",
+ "\n",
+ "# Testing Q4 explicit methods\n",
+ "assert M1.scalarmul(3).data == [[3, 6], [9, 12]]\n",
+ "assert M1.mat_mult(M2).data == [[19, 22], [43, 50]]\n",
+ "\n",
+ "# Testing Q5 overloaded operators\n",
+ "assert (M1 + M2).data == [[6, 8], [10, 12]]\n",
+ "assert (M1 - M2).data == [[-4, -4], [-4, -4]]\n",
+ "assert (M1 * M2).data == [[19, 22], [43, 50]]\n",
+ "assert (M1 * 2).data == [[2, 4], [6, 8]]\n",
+ "assert (2 * M1).data == [[2, 4], [6, 8]]\n",
+ "assert (M1 == matrix([[1, 2], [3, 4]])) == True\n",
+ "\n",
+ "print(\"Q4 Math Methods and Q5 Overloaded Operators tests passed!\")"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -94,6 +451,72 @@
"$$"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 25,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "--- Q6 Mathematical Demonstrations ---\n",
+ "\n",
+ "1. Demonstrating (AB)C = A(BC)\n",
+ "Is it true? True\n",
+ "\n",
+ "2. Demonstrating A(B+C) = AB + AC\n",
+ "Is it true? True\n",
+ "\n",
+ "3. Demonstrating AB != BA\n",
+ "Is it true? True\n",
+ "\n",
+ "4. Demonstrating AI = A\n",
+ "Is it true? True\n",
+ "\n",
+ "All Q6 linear algebra properties successfully demonstrated!\n"
+ ]
+ }
+ ],
+ "source": [
+ "# Create test matrices\n",
+ "A = matrix([[1, 2], [3, 4]])\n",
+ "B = matrix([[5, 6], [7, 8]])\n",
+ "C = matrix([[9, 10], [11, 12]])\n",
+ "I = eye(2) \n",
+ "\n",
+ "print(\"--- Q6 Mathematical Demonstrations ---\\n\")\n",
+ "\n",
+ "# 1. Associativity: (AB)C = A(BC)\n",
+ "left_side = (A * B) * C\n",
+ "right_side = A * (B * C)\n",
+ "print(\"1. Demonstrating (AB)C = A(BC)\")\n",
+ "print(f\"Is it true? {left_side == right_side}\\n\")\n",
+ "assert left_side == right_side\n",
+ "\n",
+ "# 2. Distributivity: A(B+C) = AB + AC\n",
+ "left_side = A * (B + C)\n",
+ "right_side = (A * B) + (A * C)\n",
+ "print(\"2. Demonstrating A(B+C) = AB + AC\")\n",
+ "print(f\"Is it true? {left_side == right_side}\\n\")\n",
+ "assert left_side == right_side\n",
+ "\n",
+ "# 3. Non-commutativity: AB != BA\n",
+ "AB = A * B\n",
+ "BA = B * A\n",
+ "print(\"3. Demonstrating AB != BA\")\n",
+ "print(f\"Is it true? {not (AB == BA)}\\n\")\n",
+ "assert not (AB == BA)\n",
+ "\n",
+ "# 4. Identity: AI = A\n",
+ "AI = A * I\n",
+ "print(\"4. Demonstrating AI = A\")\n",
+ "print(f\"Is it true? {AI == A}\\n\")\n",
+ "assert AI == A\n",
+ "\n",
+ "print(\"All Q6 linear algebra properties successfully demonstrated!\")"
+ ]
+ },
{
"cell_type": "code",
"execution_count": null,
@@ -104,7 +527,7 @@
],
"metadata": {
"kernelspec": {
- "display_name": "Python 3 (ipykernel)",
+ "display_name": "base",
"language": "python",
"name": "python3"
},
@@ -118,7 +541,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.13.12"
+ "version": "3.12.12"
}
},
"nbformat": 4,
diff --git a/Labs/Lab.6/.DS_Store b/Labs/Lab.6/.DS_Store
new file mode 100644
index 00000000..36125665
Binary files /dev/null and b/Labs/Lab.6/.DS_Store differ
diff --git a/Labs/Lab.6/Lab.6.ipynb b/Labs/Lab.6/Lab.6.ipynb
index 3bc4e512..3cc6d7a7 100755
--- a/Labs/Lab.6/Lab.6.ipynb
+++ b/Labs/Lab.6/Lab.6.ipynb
@@ -29,6 +29,66 @@
"1. Begin by creating a classes to represent cards and decks. The deck should support more than one 52-card set. The deck should allow you to shuffle and draw cards. Include a \"plastic\" card, placed randomly in the deck. Later, when the plastic card is dealt, shuffle the cards before the next deal."
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "class Player:\n",
+ " def __init__(self, name, starting_chips=1000):\n",
+ " self.name = name\n",
+ " self.chips = starting_chips\n",
+ " self.hand = []\n",
+ " \n",
+ " # Storing the bet on the player object so we can easily calculate \n",
+ " # payouts or losses at the end of the round.\n",
+ " self.current_bet = 0 \n",
+ " \n",
+ " def receive_card(self, card):\n",
+ " self.hand.append(card)\n",
+ "\n",
+ " def calculate_score(self):\n",
+ " \"\"\"Calculates the best possible hand value.\"\"\"\n",
+ " score = 0\n",
+ " aces = 0\n",
+ " \n",
+ " # Logic: It's easier to assume all Aces are 11s initially. \n",
+ " # We can just downgrade them to 1s later if the total goes over 21.\n",
+ " for card in self.hand:\n",
+ " if card.rank in ['J', 'Q', 'K']:\n",
+ " score += 10\n",
+ " elif card.rank == 'A':\n",
+ " aces += 1\n",
+ " score += 11\n",
+ " else:\n",
+ " score += int(card.rank)\n",
+ " \n",
+ " # If we bust, check if we have an Ace we can shrink from 11 to 1.\n",
+ " # We subtract 10 for each Ace we convert until we are safe.\n",
+ " while score > 21 and aces > 0:\n",
+ " score -= 10\n",
+ " aces -= 1\n",
+ " \n",
+ " return score\n",
+ " \n",
+ " def clear_hand(self):\n",
+ " # Resetting for the next simulation round\n",
+ " self.hand = []\n",
+ " self.current_bet = 0\n",
+ "\n",
+ "\n",
+ "class Dealer(Player):\n",
+ " # The dealer is just a player with forced actions, so inheritance makes sense here.\n",
+ " def __init__(self):\n",
+ " super().__init__(\"Dealer\", starting_chips=0) # Dealer chips don't matter\n",
+ "\n",
+ " def should_hit(self):\n",
+ " # Assignment requirement: Dealer hits on 16 and stands on 17.\n",
+ " # Returning True means \"Hit\", False means \"Stand\".\n",
+ " return self.calculate_score() < 17"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -43,6 +103,153 @@
"3. Begin with implementing the skeleton (ie define data members and methods/functions, but do not code the logic) of the classes in your UML diagram."
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "class Card:\n",
+ " \"\"\"Represents a single playing card.\"\"\"\n",
+ " def __init__(self, rank, suit):\n",
+ " # Storing basic card properties. \n",
+ " self.rank = rank\n",
+ " self.suit = suit\n",
+ "\n",
+ " def __str__(self):\n",
+ " pass\n",
+ "\n",
+ " def __repr__(self):\n",
+ " pass\n",
+ "\n",
+ "\n",
+ "class Deck:\n",
+ " \"\"\"Manages the shoe of multiple decks.\"\"\"\n",
+ " def __init__(self, num_decks=6):\n",
+ " self.num_decks = num_decks\n",
+ " # Using a list to hold Card objects so we maintain drawing order\n",
+ " self.cards = [] \n",
+ " # Flag to tell the Game loop when the plastic card was hit\n",
+ " self.needs_reshuffle = False \n",
+ "\n",
+ " def build_deck(self):\n",
+ " pass\n",
+ "\n",
+ " def shuffle(self):\n",
+ " pass\n",
+ "\n",
+ " def draw(self):\n",
+ " pass\n",
+ "\n",
+ "\n",
+ "class Player:\n",
+ " def __init__(self, name, strategy=\"Human\", starting_chips=1000):\n",
+ " self.name = name\n",
+ " self.chips = starting_chips\n",
+ " self.current_bet = 0\n",
+ " self.hand = []\n",
+ " self.strategy = strategy \n",
+ "\n",
+ " def receive_card(self, card):\n",
+ " self.hand.append(card)\n",
+ "\n",
+ " def calculate_score(self):\n",
+ " score = 0\n",
+ " aces = 0\n",
+ " \n",
+ " for card in self.hand:\n",
+ " if card.rank in ['J', 'Q', 'K']:\n",
+ " score += 10\n",
+ " elif card.rank == 'A':\n",
+ " aces += 1\n",
+ " score += 11\n",
+ " else:\n",
+ " score += int(card.rank)\n",
+ " \n",
+ " while score > 21 and aces > 0:\n",
+ " score -= 10\n",
+ " aces -= 1\n",
+ " \n",
+ " return score\n",
+ "\n",
+ " def clear_hand(self):\n",
+ " self.hand = []\n",
+ " self.current_bet = 0\n",
+ "\n",
+ " def place_bet(self):\n",
+ " \"\"\"Routes betting logic based on player strategy.\"\"\"\n",
+ " if self.strategy == \"Human\":\n",
+ " while True:\n",
+ " try:\n",
+ " bet = int(input(f\"{self.name}, you have {self.chips} chips. Enter bet: \"))\n",
+ " if 0 < bet <= self.chips:\n",
+ " self.current_bet = bet\n",
+ " self.chips -= bet\n",
+ " break\n",
+ " else:\n",
+ " print(\"Invalid amount. Must be greater than 0.\")\n",
+ " except ValueError:\n",
+ " print(\"Please enter a valid number.\")\n",
+ " else:\n",
+ " # Automated bot logic: Always bet 10 chips to keep the simulation simple\n",
+ " if self.chips > 0:\n",
+ " bet = 10 if self.chips >= 10 else self.chips\n",
+ " self.current_bet = bet\n",
+ " self.chips -= bet\n",
+ "\n",
+ " def decide_action(self, dealer_upcard, true_count):\n",
+ " \"\"\"Routes hitting/standing logic based on player strategy.\"\"\"\n",
+ " if self.strategy == \"Human\":\n",
+ " while True:\n",
+ " choice = input(f\"Your score is {self.calculate_score()}. Hit or Stand? (h/s): \").lower()\n",
+ " if choice in ['h', 's']:\n",
+ " return \"Hit\" if choice == 'h' else \"Stand\"\n",
+ " print(\"Invalid input. Please type 'h' or 's'.\")\n",
+ " \n",
+ " elif self.strategy == \"Dealer\":\n",
+ " # Dealer bots ignore the true_count and just hit if under 17\n",
+ " if self.calculate_score() < 17:\n",
+ " return \"Hit\"\n",
+ " return \"Stand\"\n",
+ "\n",
+ "\n",
+ "class Dealer(Player):\n",
+ " \"\"\"The dealer is a specific type of player with fixed rules.\"\"\"\n",
+ " def __init__(self):\n",
+ " # Calling the parent Player __init__, but hardcoding the name and strategy.\n",
+ " # Dealers don't need chips, so we set it to 0.\n",
+ " super().__init__(name=\"Dealer\", strategy=\"Fixed Rules\", starting_chips=0)\n",
+ "\n",
+ " def should_hit(self):\n",
+ " # This will override normal player logic since the dealer must hit on 16.\n",
+ " pass\n",
+ "\n",
+ "\n",
+ "class Game:\n",
+ " \"\"\"The central controller that runs the simulation.\"\"\"\n",
+ " def __init__(self, num_players=1, decks_in_shoe=6):\n",
+ " # Aggregating our other classes here to build the actual table environment\n",
+ " self.deck = Deck(num_decks=decks_in_shoe)\n",
+ " self.dealer = Dealer()\n",
+ " self.players = []\n",
+ " \n",
+ " # Tracking the global card counting state\n",
+ " self.running_count = 0\n",
+ " self.true_count = 0.0\n",
+ "\n",
+ " def update_count(self, card):\n",
+ " # This will adjust the running_count based on the drawn card's value\n",
+ " pass\n",
+ "\n",
+ " def play_round(self):\n",
+ " # This will handle the sequence: bets -> deal -> player turns -> dealer turn -> payouts\n",
+ " pass\n",
+ "\n",
+ " def run_simulation(self, num_games):\n",
+ " # The main loop to run play_round() thousands of times for our data analysis\n",
+ " pass"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -50,6 +257,225 @@
"4. Complete the implementation by coding the logic of all functions. For now, just implement the dealer player and human player."
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import random\n",
+ "\n",
+ "class Card:\n",
+ " def __init__(self, rank, suit):\n",
+ " self.rank = rank\n",
+ " self.suit = suit\n",
+ " def __str__(self): return f\"{self.rank} of {self.suit}\"\n",
+ " def __repr__(self): return self.__str__()\n",
+ "\n",
+ "class Deck:\n",
+ " suits = ['Hearts', 'Diamonds', 'Clubs', 'Spades']\n",
+ " ranks = ['2', '3', '4', '5', '6', '7', '8', '9', '10', 'J', 'Q', 'K', 'A']\n",
+ "\n",
+ " def __init__(self, num_decks=6):\n",
+ " self.num_decks = num_decks\n",
+ " self.cards = []\n",
+ " self.needs_reshuffle = False\n",
+ " self.build_deck()\n",
+ " self.shuffle()\n",
+ "\n",
+ " def build_deck(self):\n",
+ " self.cards = [Card(rank, suit) for _ in range(self.num_decks) for suit in self.suits for rank in self.ranks]\n",
+ "\n",
+ " def shuffle(self):\n",
+ " random.shuffle(self.cards)\n",
+ " self.needs_reshuffle = False\n",
+ " total_cards = len(self.cards)\n",
+ " insert_idx = random.randint(total_cards // 2, int(total_cards * 0.85))\n",
+ " self.cards.insert(insert_idx, \"PLASTIC_CARD\")\n",
+ "\n",
+ " def draw(self):\n",
+ " if not self.cards:\n",
+ " self.build_deck()\n",
+ " self.shuffle()\n",
+ " drawn_item = self.cards.pop(0)\n",
+ " if drawn_item == \"PLASTIC_CARD\":\n",
+ " self.needs_reshuffle = True\n",
+ " return self.draw()\n",
+ " return drawn_item\n",
+ "\n",
+ "class Player:\n",
+ " def __init__(self, name, strategy=\"Human\", starting_chips=1000):\n",
+ " self.name = name\n",
+ " self.chips = starting_chips\n",
+ " self.current_bet = 0\n",
+ " self.hand = []\n",
+ " self.strategy = strategy \n",
+ "\n",
+ " def receive_card(self, card):\n",
+ " self.hand.append(card)\n",
+ "\n",
+ " def calculate_score(self):\n",
+ " score = 0\n",
+ " aces = 0\n",
+ " for card in self.hand:\n",
+ " if card.rank in ['J', 'Q', 'K']: score += 10\n",
+ " elif card.rank == 'A':\n",
+ " aces += 1\n",
+ " score += 11\n",
+ " else: score += int(card.rank)\n",
+ " \n",
+ " while score > 21 and aces > 0:\n",
+ " score -= 10\n",
+ " aces -= 1\n",
+ " return score\n",
+ "\n",
+ " def clear_hand(self):\n",
+ " self.hand = []\n",
+ " self.current_bet = 0\n",
+ "\n",
+ " def place_bet(self, true_count=0.0):\n",
+ " if self.chips <= 0:\n",
+ " self.current_bet = 0\n",
+ " return \n",
+ "\n",
+ " if self.strategy == \"Human\":\n",
+ " while True:\n",
+ " try:\n",
+ " bet = int(input(f\"{self.name}, you have {self.chips} chips. Enter bet: \"))\n",
+ " if 0 < bet <= self.chips:\n",
+ " self.current_bet = bet\n",
+ " self.chips -= bet\n",
+ " break\n",
+ " else: print(\"Invalid amount.\")\n",
+ " except ValueError: print(\"Please enter a valid number.\")\n",
+ " elif self.strategy == \"Counter\":\n",
+ " if true_count >= 2.0: bet = 50 if self.chips >= 50 else self.chips\n",
+ " else: bet = 10 if self.chips >= 10 else self.chips\n",
+ " self.current_bet = bet\n",
+ " self.chips -= bet\n",
+ " else:\n",
+ " bet = 10 if self.chips >= 10 else self.chips\n",
+ " self.current_bet = bet\n",
+ " self.chips -= bet\n",
+ "\n",
+ " # CRITICAL FIX: The base player now officially accepts running_count\n",
+ " def decide_action(self, dealer_upcard, true_count, running_count=0):\n",
+ " if self.strategy == \"Human\":\n",
+ " while True:\n",
+ " choice = input(f\"Your score is {self.calculate_score()}. Hit or Stand? (h/s): \").lower()\n",
+ " if choice in ['h', 's']: return \"Hit\" if choice == 'h' else \"Stand\"\n",
+ " print(\"Invalid input.\")\n",
+ " elif self.strategy == \"Dealer\":\n",
+ " if self.calculate_score() < 17: return \"Hit\"\n",
+ " return \"Stand\"\n",
+ " elif self.strategy == \"Counter\":\n",
+ " dealer_val = 10 if dealer_upcard.rank in ['J', 'Q', 'K', 'A'] else int(dealer_upcard.rank)\n",
+ " score = self.calculate_score()\n",
+ " if score >= 17: return \"Stand\"\n",
+ " elif score >= 12 and dealer_val <= 6: return \"Stand\"\n",
+ " else: return \"Hit\"\n",
+ "\n",
+ "class Dealer(Player):\n",
+ " def __init__(self):\n",
+ " super().__init__(name=\"Dealer\", strategy=\"Fixed Rules\", starting_chips=0)\n",
+ " def should_hit(self):\n",
+ " return self.calculate_score() < 17\n",
+ "\n",
+ "class Lab6Player(Player):\n",
+ " def __init__(self, name, starting_chips=1000, hit_threshold=-1):\n",
+ " super().__init__(name=name, strategy=\"Count_Action\", starting_chips=starting_chips)\n",
+ " self.hit_threshold = hit_threshold\n",
+ "\n",
+ " def decide_action(self, dealer_upcard, true_count, running_count=0):\n",
+ " if self.calculate_score() >= 21: return \"Stand\"\n",
+ " if running_count <= self.hit_threshold: return \"Hit\"\n",
+ " else: return \"Stand\"\n",
+ "\n",
+ "class Game:\n",
+ " def __init__(self, players_list, decks_in_shoe=6, verbose=True):\n",
+ " self.deck = Deck(num_decks=decks_in_shoe)\n",
+ " self.dealer = Dealer()\n",
+ " self.players = players_list \n",
+ " self.running_count = 0\n",
+ " self.true_count = 0.0\n",
+ " self.verbose = verbose \n",
+ "\n",
+ " def update_count(self, card):\n",
+ " if card.rank in ['2', '3', '4', '5', '6']: self.running_count += 1\n",
+ " elif card.rank in ['10', 'J', 'Q', 'K', 'A']: self.running_count -= 1\n",
+ " decks_remaining = max(1, len(self.deck.cards) / 52)\n",
+ " self.true_count = self.running_count / decks_remaining\n",
+ "\n",
+ " def deal_and_count(self, person):\n",
+ " card = self.deck.draw()\n",
+ " self.update_count(card)\n",
+ " person.receive_card(card)\n",
+ "\n",
+ " def play_round(self):\n",
+ " if self.verbose: print(\"\\n--- NEW ROUND ---\")\n",
+ " if self.deck.needs_reshuffle:\n",
+ " if self.verbose: print(\"Dealer is shuffling the shoe...\")\n",
+ " self.deck.build_deck()\n",
+ " self.deck.shuffle()\n",
+ " self.running_count = 0 \n",
+ " self.true_count = 0.0\n",
+ "\n",
+ " for player in self.players: \n",
+ " if player.chips > 0: player.place_bet(self.true_count)\n",
+ "\n",
+ " for _ in range(2):\n",
+ " for player in self.players: \n",
+ " if player.current_bet > 0: self.deal_and_count(player)\n",
+ " self.deal_and_count(self.dealer)\n",
+ "\n",
+ " if self.verbose: print(f\"Dealer shows: {self.dealer.hand[0]} (Hidden Card)\")\n",
+ "\n",
+ " for player in self.players:\n",
+ " if player.current_bet == 0: continue\n",
+ " if self.verbose: print(f\"\\n{player.name}'s turn. Hand: {player.hand}\")\n",
+ " while player.calculate_score() < 21:\n",
+ " action = player.decide_action(self.dealer.hand[0], self.true_count, self.running_count)\n",
+ " if action == \"Hit\":\n",
+ " self.deal_and_count(player)\n",
+ " if self.verbose: print(f\"Drew a card. Hand: {player.hand}\")\n",
+ " else: break\n",
+ " if player.calculate_score() > 21 and self.verbose: print(f\"{player.name} BUSTS!\")\n",
+ "\n",
+ " if self.verbose: print(f\"\\nDealer's turn. Hand reveals: {self.dealer.hand}\")\n",
+ " while self.dealer.should_hit():\n",
+ " self.deal_and_count(self.dealer)\n",
+ " if self.verbose: print(f\"Dealer hits. Hand: {self.dealer.hand}\")\n",
+ "\n",
+ " dealer_score = self.dealer.calculate_score()\n",
+ "\n",
+ " for player in self.players:\n",
+ " if player.current_bet == 0: continue\n",
+ " player_score = player.calculate_score()\n",
+ " if player_score <= 21:\n",
+ " if dealer_score > 21 or player_score > dealer_score:\n",
+ " player.chips += (player.current_bet * 2)\n",
+ " if self.verbose: print(f\"{player.name} WINS!\")\n",
+ " elif player_score == dealer_score:\n",
+ " player.chips += player.current_bet\n",
+ " if self.verbose: print(f\"{player.name} PUSHES.\")\n",
+ " else: \n",
+ " if self.verbose: print(f\"{player.name} LOSES.\")\n",
+ " else:\n",
+ " if self.verbose: print(f\"{player.name} LOSES.\")\n",
+ " player.clear_hand()\n",
+ " \n",
+ " self.dealer.clear_hand()\n",
+ "\n",
+ " def run_simulation(self, num_games=1, target_player_name=None):\n",
+ " for i in range(1, num_games + 1):\n",
+ " if target_player_name:\n",
+ " target_player = next((p for p in self.players if p.name == target_player_name), None)\n",
+ " if target_player and target_player.chips <= 0:\n",
+ " print(f\"Simulation stopped early at round {i}: {target_player.name} is out of money!\")\n",
+ " break\n",
+ " self.play_round()"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -57,6 +483,128 @@
"5. Test. Demonstrate game play. For example, create a game of several dealer players and show that the game is functional through several rounds."
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def place_bet(self):\n",
+ " \"\"\"Handles betting for both humans and computer players.\"\"\"\n",
+ " if self.strategy == \"Human\":\n",
+ " # ... (Keep your existing human input loop here) ...\n",
+ " pass\n",
+ " else:\n",
+ " # Logic: Automated testing. Computer players bet a flat 10 chips \n",
+ " # so we can easily track their win/loss rate over time.\n",
+ " if self.chips > 0:\n",
+ " bet = 10 if self.chips >= 10 else self.chips\n",
+ " self.current_bet = bet\n",
+ " self.chips -= bet\n",
+ "\n",
+ "def decide_action(self, dealer_upcard, true_count, running_count=0):\n",
+ " \"\"\"Routes the decision based on the player's assigned strategy.\"\"\"\n",
+ " if self.strategy == \"Human\":\n",
+ " # ... (Keep your existing human input loop here) ...\n",
+ " pass\n",
+ " elif self.strategy == \"Dealer\":\n",
+ " # Logic: These players mimic the dealer. They ignore the true_count \n",
+ " # and dealer_upcard, and simply hit if they are under 17.\n",
+ " if self.calculate_score() < 17:\n",
+ " return \"Hit\"\n",
+ " return \"Stand\"\n",
+ " \n",
+ "def __init__(self, players_list, decks_in_shoe=6):\n",
+ " self.deck = Deck(num_decks=decks_in_shoe)\n",
+ " self.dealer = Dealer()\n",
+ " \n",
+ " # Logic: Instead of hardcoding one human, we pass a list of Player objects\n",
+ " # so we can test different table compositions (e.g., 3 bots, 1 human).\n",
+ " self.players = players_list \n",
+ " \n",
+ " self.running_count = 0\n",
+ " self.true_count = 0.0"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Starting automated test for 3 rounds...\n",
+ "\n",
+ "\n",
+ "--- NEW ROUND ---\n",
+ "Dealer shows: 4 of Diamonds (Hidden Card)\n",
+ "\n",
+ "Bot Alice's turn. Hand: [4 of Clubs, J of Spades]\n",
+ "\n",
+ "Bot Bob's turn. Hand: [4 of Diamonds, 2 of Spades]\n",
+ "\n",
+ "Bot Charlie's turn. Hand: [9 of Diamonds, 6 of Spades]\n",
+ "\n",
+ "Dealer's turn. Hand reveals: [4 of Diamonds, 4 of Clubs]\n",
+ "Dealer hits. Hand: [4 of Diamonds, 4 of Clubs, 2 of Diamonds]\n",
+ "Dealer hits. Hand: [4 of Diamonds, 4 of Clubs, 2 of Diamonds, 5 of Clubs]\n",
+ "Dealer hits. Hand: [4 of Diamonds, 4 of Clubs, 2 of Diamonds, 5 of Clubs, J of Hearts]\n",
+ "\n",
+ "--- NEW ROUND ---\n",
+ "Dealer shows: A of Hearts (Hidden Card)\n",
+ "\n",
+ "Bot Alice's turn. Hand: [6 of Clubs, 9 of Spades]\n",
+ "\n",
+ "Bot Bob's turn. Hand: [K of Spades, 6 of Hearts]\n",
+ "\n",
+ "Bot Charlie's turn. Hand: [2 of Diamonds, 6 of Diamonds]\n",
+ "\n",
+ "Dealer's turn. Hand reveals: [A of Hearts, A of Clubs]\n",
+ "Dealer hits. Hand: [A of Hearts, A of Clubs, Q of Hearts]\n",
+ "Dealer hits. Hand: [A of Hearts, A of Clubs, Q of Hearts, 7 of Spades]\n",
+ "\n",
+ "--- NEW ROUND ---\n",
+ "Dealer shows: 6 of Clubs (Hidden Card)\n",
+ "\n",
+ "Bot Alice's turn. Hand: [A of Spades, Q of Hearts]\n",
+ "\n",
+ "Bot Bob's turn. Hand: [9 of Clubs, 6 of Diamonds]\n",
+ "\n",
+ "Bot Charlie's turn. Hand: [9 of Spades, K of Clubs]\n",
+ "\n",
+ "Dealer's turn. Hand reveals: [6 of Clubs, 2 of Spades]\n",
+ "Dealer hits. Hand: [6 of Clubs, 2 of Spades, Q of Clubs]\n",
+ "\n",
+ "--- FINAL CHIP COUNTS ---\n",
+ "Bot Alice: 1800 chips\n",
+ "Bot Bob: 0 chips\n",
+ "Bot Charlie: 1800 chips\n"
+ ]
+ }
+ ],
+ "source": [
+ "# --- SIMULATION TEST ---\n",
+ "\n",
+ "# 1. Create our automated players\n",
+ "bot_1 = Player(name=\"Bot Alice\", strategy=\"Dealer\", starting_chips=1000)\n",
+ "bot_2 = Player(name=\"Bot Bob\", strategy=\"Dealer\", starting_chips=1000)\n",
+ "bot_3 = Player(name=\"Bot Charlie\", strategy=\"Dealer\", starting_chips=1000)\n",
+ "\n",
+ "# 2. Initialize the game with these players\n",
+ "test_table = Game(players_list=[bot_1, bot_2, bot_3], decks_in_shoe=6)\n",
+ "\n",
+ "# 3. Run a short simulation to prove it works\n",
+ "print(\"Starting automated test for 3 rounds...\\n\")\n",
+ "test_table.run_simulation(num_games=3)\n",
+ "\n",
+ "# 4. Print final results to verify chips updated correctly\n",
+ "print(\"\\n--- FINAL CHIP COUNTS ---\")\n",
+ "for p in test_table.players:\n",
+ " print(f\"{p.name}: {p.chips} chips\")"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -71,6 +619,66 @@
" * Hit if sum is very negative, stay if sum is very positive. Select a threshold for hit/stay, e.g. 0 or -2. "
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "class Lab6Player(Player):\n",
+ " def __init__(self, name, starting_chips=1000, hit_threshold=-1):\n",
+ " # Call the parent __init__ to set up the hand, chips, etc.\n",
+ " super().__init__(name=name, strategy=\"Count_Action\", starting_chips=starting_chips)\n",
+ " \n",
+ " # Storing the threshold so we can easily test different values later (like 0 or -2)\n",
+ " self.hit_threshold = hit_threshold\n",
+ "\n",
+ " def decide_action(self, dealer_upcard, true_count, running_count=0):\n",
+ " \"\"\"\n",
+ " LAB 6 STRATEGY: \n",
+ " Ignore the dealer's card and our own hand score (mostly).\n",
+ " Hit if the sum is very negative, stay if very positive.\n",
+ " \"\"\"\n",
+ " # Absolute safeguard: It is mathematically useless to hit on a 21 or higher.\n",
+ " if self.calculate_score() >= 21:\n",
+ " return \"Stand\"\n",
+ " \n",
+ " # The core Lab 6 logic: Compare the running count sum to our threshold\n",
+ " if running_count <= self.hit_threshold:\n",
+ " return \"Hit\"\n",
+ " else:\n",
+ " return \"Stand\""
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "--- NEW ROUND ---\n",
+ "Dealer shows: 10 of Diamonds (Hidden Card)\n",
+ "\n",
+ "Q6 Bot's turn. Hand: [J of Spades, 6 of Diamonds]\n",
+ "Drew a card. Hand: [J of Spades, 6 of Diamonds, 3 of Hearts]\n",
+ "\n",
+ "Dealer's turn. Hand reveals: [10 of Diamonds, Q of Clubs]\n",
+ "Q6 Bot LOSES.\n"
+ ]
+ }
+ ],
+ "source": [
+ "count_bot = Lab6Player(name=\"Q6 Bot\", hit_threshold=-2)\n",
+ "\n",
+ "# Load it into a game and play just ONE round to see the prints\n",
+ "q6_table = Game(players_list=[count_bot], decks_in_shoe=6)\n",
+ "q6_table.run_simulation(num_games=1)"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -78,6 +686,159 @@
"7. Create a test scenario where one player, using the above strategy, is playing with a dealer and 3 other players that follow the dealer's strategy. Each player starts with same number of chips. Play 50 rounds (or until the strategy player is out of money). Compute the strategy player's winnings. You may remove unnecessary printouts from your code (perhaps implement a verbose/quiet mode) to reduce the output."
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "class Game:\n",
+ " # 1. Added the verbose parameter here\n",
+ " def __init__(self, players_list, decks_in_shoe=6, verbose=True):\n",
+ " self.deck = Deck(num_decks=decks_in_shoe)\n",
+ " self.dealer = Dealer()\n",
+ " self.players = players_list \n",
+ " self.running_count = 0\n",
+ " self.true_count = 0.0\n",
+ " self.verbose = verbose \n",
+ "\n",
+ " def update_count(self, card):\n",
+ " if card.rank in ['2', '3', '4', '5', '6']: self.running_count += 1\n",
+ " elif card.rank in ['10', 'J', 'Q', 'K', 'A']: self.running_count -= 1\n",
+ " decks_remaining = max(1, len(self.deck.cards) / 52)\n",
+ " self.true_count = self.running_count / decks_remaining\n",
+ "\n",
+ " def deal_and_count(self, person):\n",
+ " card = self.deck.draw()\n",
+ " self.update_count(card)\n",
+ " person.receive_card(card)\n",
+ "\n",
+ " def play_round(self):\n",
+ " # 2. Wrapped print statements in \"if self.verbose:\" checks\n",
+ " if self.verbose: print(\"\\n--- NEW ROUND ---\")\n",
+ " \n",
+ " if self.deck.needs_reshuffle:\n",
+ " if self.verbose: print(\"Dealer is shuffling the shoe...\")\n",
+ " self.deck.build_deck()\n",
+ " self.deck.shuffle()\n",
+ " self.running_count = 0 \n",
+ " self.true_count = 0.0\n",
+ "\n",
+ " for player in self.players: \n",
+ " player.place_bet(self.true_count)\n",
+ "\n",
+ " for _ in range(2):\n",
+ " for player in self.players: self.deal_and_count(player)\n",
+ " self.deal_and_count(self.dealer)\n",
+ "\n",
+ " if self.verbose: print(f\"Dealer shows: {self.dealer.hand[0]} (Hidden Card)\")\n",
+ "\n",
+ " for player in self.players:\n",
+ " if self.verbose: print(f\"\\n{player.name}'s turn. Hand: {player.hand}\")\n",
+ " while player.calculate_score() < 21:\n",
+ " action = player.decide_action(self.dealer.hand[0], self.true_count, self.running_count)\n",
+ " if action == \"Hit\":\n",
+ " self.deal_and_count(player)\n",
+ " if self.verbose: print(f\"Drew a card. Hand: {player.hand}\")\n",
+ " else: break\n",
+ " if player.calculate_score() > 21:\n",
+ " if self.verbose: print(f\"{player.name} BUSTS!\")\n",
+ "\n",
+ " if self.verbose: print(f\"\\nDealer's turn. Hand reveals: {self.dealer.hand}\")\n",
+ " while self.dealer.should_hit():\n",
+ " self.deal_and_count(self.dealer)\n",
+ " if self.verbose: print(f\"Dealer hits. Hand: {self.dealer.hand}\")\n",
+ "\n",
+ " dealer_score = self.dealer.calculate_score()\n",
+ "\n",
+ " for player in self.players:\n",
+ " player_score = player.calculate_score()\n",
+ " if player_score <= 21:\n",
+ " if dealer_score > 21 or player_score > dealer_score:\n",
+ " player.chips += (player.current_bet * 2)\n",
+ " if self.verbose: print(f\"{player.name} WINS!\")\n",
+ " elif player_score == dealer_score:\n",
+ " player.chips += player.current_bet\n",
+ " if self.verbose: print(f\"{player.name} PUSHES.\")\n",
+ " else: \n",
+ " if self.verbose: print(f\"{player.name} LOSES.\")\n",
+ " else:\n",
+ " if self.verbose: print(f\"{player.name} LOSES.\")\n",
+ " player.clear_hand()\n",
+ " \n",
+ " self.dealer.clear_hand()\n",
+ "\n",
+ " # 3. Updated simulation loop to check for bankruptcy\n",
+ " def run_simulation(self, num_games=1, target_player_name=None):\n",
+ " for i in range(1, num_games + 1):\n",
+ " # If a target player is provided, check if they are broke before dealing\n",
+ " if target_player_name:\n",
+ " target_player = next((p for p in self.players if p.name == target_player_name), None)\n",
+ " if target_player and target_player.chips <= 0:\n",
+ " print(f\"Simulation stopped early at round {i}: {target_player.name} is out of money!\")\n",
+ " break\n",
+ " \n",
+ " self.play_round()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Starting 50-round simulation... (Quiet Mode Active)\n",
+ "\n",
+ "--- FINAL RESULTS AFTER 50 ROUNDS ---\n",
+ "Dealer Bot 1:\n",
+ " Ending Chips: 930\n",
+ " Net Winnings: -70 chips\n",
+ "\n",
+ "Dealer Bot 2:\n",
+ " Ending Chips: 830\n",
+ " Net Winnings: -170 chips\n",
+ "\n",
+ "Dealer Bot 3:\n",
+ " Ending Chips: 950\n",
+ " Net Winnings: -50 chips\n",
+ "\n",
+ "Strategy Player:\n",
+ " Ending Chips: 790\n",
+ " Net Winnings: -210 chips\n",
+ "\n"
+ ]
+ }
+ ],
+ "source": [
+ "# --- STEP 7: TEST SCENARIO ---\n",
+ "\n",
+ "# 1. Create the players (Each starts with exactly 1000 chips)\n",
+ "bot_1 = Player(name=\"Dealer Bot 1\", strategy=\"Dealer\", starting_chips=1000)\n",
+ "bot_2 = Player(name=\"Dealer Bot 2\", strategy=\"Dealer\", starting_chips=1000)\n",
+ "bot_3 = Player(name=\"Dealer Bot 3\", strategy=\"Dealer\", starting_chips=1000)\n",
+ "\n",
+ "# The Lab 6 Strategy Player. We will use a threshold of 0 for this test.\n",
+ "strat_player = Lab6Player(name=\"Strategy Player\", starting_chips=1000, hit_threshold=0)\n",
+ "\n",
+ "# 2. Initialize the game table. Note that we set verbose=False!\n",
+ "q7_table = Game(players_list=[bot_1, bot_2, bot_3, strat_player], decks_in_shoe=6, verbose=False)\n",
+ "\n",
+ "# 3. Play 50 rounds, targeting our Strategy Player for the bankruptcy check\n",
+ "print(\"Starting 50-round simulation... (Quiet Mode Active)\\n\")\n",
+ "q7_table.run_simulation(num_games=50, target_player_name=\"Strategy Player\")\n",
+ "\n",
+ "# 4. Compute and print the final winnings\n",
+ "print(\"--- FINAL RESULTS AFTER 50 ROUNDS ---\")\n",
+ "for p in q7_table.players:\n",
+ " winnings = p.chips - 1000\n",
+ " print(f\"{p.name}:\")\n",
+ " print(f\" Ending Chips: {p.chips}\")\n",
+ " print(f\" Net Winnings: {winnings} chips\\n\")"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -85,6 +846,122 @@
"8. Create a loop that runs 100 games of 50 rounds, as setup in previous question, and store the strategy player's chips at the end of the game (aka \"winnings\") in a list. Histogram the winnings. What is the average winnings per round? What is the standard deviation. What is the probabilty of net winning or lossing after 50 rounds?\n"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Running 100 games of 50 rounds each... Please wait.\n",
+ "\n",
+ "--- FINAL STATISTICAL REPORT ---\n",
+ "Average Winnings (Per 50-Round Game): -208.60 chips\n",
+ "Average Winnings (PER ROUND): -4.17 chips\n",
+ "Standard Deviation of Winnings: 67.25 chips\n",
+ "Probability of a Net Win: 0.0%\n",
+ "Probability of a Net Loss: 100.0%\n",
+ "Probability of Breaking Even: 0.0%\n",
+ "\n"
+ ]
+ },
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ "
"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "import matplotlib.pyplot as plt\n",
+ "import numpy as np\n",
+ "\n",
+ "# --- 1. THE MONTE CARLO LOOP ---\n",
+ "num_simulations = 100\n",
+ "rounds_per_game = 50\n",
+ "winnings_list = []\n",
+ "\n",
+ "print(f\"Running {num_simulations} games of {rounds_per_game} rounds each... Please wait.\\n\")\n",
+ "\n",
+ "for _ in range(num_simulations):\n",
+ " # CRITICAL LOGIC: We MUST re-instantiate the players inside the loop. \n",
+ " # If we don't, they will carry their chip counts over from the previous game!\n",
+ " # We need everyone starting fresh with exactly 1000 chips.\n",
+ " bot_1 = Player(name=\"Dealer Bot 1\", strategy=\"Dealer\", starting_chips=1000)\n",
+ " bot_2 = Player(name=\"Dealer Bot 2\", strategy=\"Dealer\", starting_chips=1000)\n",
+ " bot_3 = Player(name=\"Dealer Bot 3\", strategy=\"Dealer\", starting_chips=1000)\n",
+ " strat_player = Lab6Player(name=\"Strategy Player\", starting_chips=1000, hit_threshold=0)\n",
+ " \n",
+ " # Initialize the table in quiet mode\n",
+ " q8_table = Game(players_list=[bot_1, bot_2, bot_3, strat_player], decks_in_shoe=6, verbose=False)\n",
+ " \n",
+ " # Run the 50 rounds\n",
+ " q8_table.run_simulation(num_games=rounds_per_game, target_player_name=\"Strategy Player\")\n",
+ " \n",
+ " # Record the net winnings (ending chips minus starting chips)\n",
+ " net_winnings = strat_player.chips - 1000\n",
+ " winnings_list.append(net_winnings)\n",
+ "\n",
+ "# --- 2. STATISTICAL ANALYSIS ---\n",
+ "# Convert our list to a numpy array for easy mathematical calculations\n",
+ "winnings_array = np.array(winnings_list)\n",
+ "\n",
+ "# 1. Average winnings per game, and per round\n",
+ "avg_winnings_per_game = np.mean(winnings_array)\n",
+ "avg_winnings_per_round = avg_winnings_per_game / rounds_per_game \n",
+ "\n",
+ "# 2. Standard Deviation\n",
+ "std_dev = np.std(winnings_array)\n",
+ "\n",
+ "# 3. Probabilities (Counting outcomes and dividing by total games)\n",
+ "prob_winning = np.sum(winnings_array > 0) / num_simulations\n",
+ "prob_losing = np.sum(winnings_array < 0) / num_simulations\n",
+ "prob_breaking_even = np.sum(winnings_array == 0) / num_simulations\n",
+ "\n",
+ "# Print the final report\n",
+ "print(\"--- FINAL STATISTICAL REPORT ---\")\n",
+ "print(f\"Average Winnings (Per 50-Round Game): {avg_winnings_per_game:.2f} chips\")\n",
+ "print(f\"Average Winnings (PER ROUND): {avg_winnings_per_round:.2f} chips\")\n",
+ "print(f\"Standard Deviation of Winnings: {std_dev:.2f} chips\")\n",
+ "print(f\"Probability of a Net Win: {prob_winning * 100:.1f}%\")\n",
+ "print(f\"Probability of a Net Loss: {prob_losing * 100:.1f}%\")\n",
+ "print(f\"Probability of Breaking Even: {prob_breaking_even * 100:.1f}%\\n\")\n",
+ "\n",
+ "# --- 3. HISTOGRAM GENERATION ---\n",
+ "plt.figure(figsize=(10, 6))\n",
+ "# Using 15 bins to get a nice distribution curve\n",
+ "plt.hist(winnings_list, bins=15, color='cornflowerblue', edgecolor='black')\n",
+ "\n",
+ "# Adding lines to easily visualize the mean and the break-even point\n",
+ "plt.axvline(avg_winnings_per_game, color='red', linestyle='dashed', linewidth=2, label=f'Mean Winnings: {avg_winnings_per_game:.2f}')\n",
+ "plt.axvline(0, color='green', linestyle='dashed', linewidth=2, label='Break Even (0)')\n",
+ "\n",
+ "plt.title('Histogram of Lab 6 Strategy Player Winnings\\n(100 Games, 50 Rounds Each)')\n",
+ "plt.xlabel('Net Winnings (Chips)')\n",
+ "plt.ylabel('Frequency (Number of Games)')\n",
+ "plt.legend()\n",
+ "plt.grid(axis='y', alpha=0.75)\n",
+ "\n",
+ "# Display the chart\n",
+ "plt.show()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "
Conclusion & Analysis
\n",
+ "\n",
+ "\n",
+ "The Monte Carlo simulation of 100 games demonstrates that the custom card-counting strategy outlined in Step 6 yields a negative expected value, with the average net winnings consistently falling below the break-even point. While tracking the running count provides valuable information about deck composition, strictly dictating hit/stay actions based solely on a count threshold—while completely ignoring the player's current hand total and the dealer's upcard—leads to mathematically unfavorable decisions, such as standing on a low total or hitting on a 19. Ultimately, this data proves that while card counting is an effective tool for adjusting bet sizes, it cannot profitably replace standard Basic Strategy for actual gameplay decisions."
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -92,19 +969,281 @@
"9. Repeat previous questions scanning the value of the threshold. Try at least 5 different threshold values. Can you find an optimal value?"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Starting threshold scan. Testing 5 values...\n",
+ "Threshold -4: Average Net Winnings = -138.70 chips\n",
+ "Threshold -2: Average Net Winnings = -183.90 chips\n",
+ "Threshold 0: Average Net Winnings = -205.60 chips\n",
+ "Threshold 2: Average Net Winnings = -246.80 chips\n",
+ "Threshold 4: Average Net Winnings = -272.80 chips\n"
+ ]
+ },
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ "
"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "CONCLUSION: The 'optimal' threshold found is -4, yielding -138.70 chips on average.\n"
+ ]
+ }
+ ],
+ "source": [
+ "import numpy as np\n",
+ "import matplotlib.pyplot as plt\n",
+ "\n",
+ "\n",
+ "thresholds_to_test = [-4, -2, 0, 2, 4] \n",
+ "num_simulations = 100\n",
+ "rounds_per_game = 50\n",
+ "\n",
+ "# Dictionary to store the average winnings for each threshold\n",
+ "results = {}\n",
+ "\n",
+ "print(f\"Starting threshold scan. Testing {len(thresholds_to_test)} values...\")\n",
+ "for threshold in thresholds_to_test:\n",
+ " # print(f\"Testing threshold: {threshold}...\")\n",
+ " winnings_for_this_threshold = []\n",
+ " \n",
+ " for _ in range(num_simulations):\n",
+ " # We must re-initialize the table with fresh players for EVERY single game\n",
+ " bot_1 = Player(name=\"Dealer Bot 1\", strategy=\"Dealer\", starting_chips=1000)\n",
+ " bot_2 = Player(name=\"Dealer Bot 2\", strategy=\"Dealer\", starting_chips=1000)\n",
+ " bot_3 = Player(name=\"Dealer Bot 3\", strategy=\"Dealer\", starting_chips=1000)\n",
+ " \n",
+ " # NOTE: We pass the current 'threshold' variable from our outer loop into the bot!\n",
+ " strat_player = Lab6Player(name=\"Strategy Player\", starting_chips=1000, hit_threshold=threshold)\n",
+ " \n",
+ " # Set up and run the game quietly\n",
+ " q9_table = Game(players_list=[bot_1, bot_2, bot_3, strat_player], decks_in_shoe=6, verbose=False)\n",
+ " q9_table.run_simulation(num_games=rounds_per_game, target_player_name=\"Strategy Player\")\n",
+ " \n",
+ " # Record net winnings\n",
+ " winnings_for_this_threshold.append(strat_player.chips - 1000)\n",
+ " \n",
+ " # Calculate the average winnings for this specific threshold and save it to the dictionary\n",
+ " avg_win = np.mean(winnings_for_this_threshold)\n",
+ " results[threshold] = avg_win\n",
+ " print(f\"Threshold {threshold:>2}: Average Net Winnings = {avg_win:.2f} chips\")\n",
+ "\n",
+ "# --- PLOTTING THE RESULTS ---\n",
+ "# Extract data for our line graph\n",
+ "x_values = list(results.keys())\n",
+ "y_values = list(results.values())\n",
+ "\n",
+ "plt.figure(figsize=(10, 6))\n",
+ "plt.plot(x_values, y_values, marker='o', linestyle='-', color='purple', linewidth=2, markersize=8)\n",
+ "\n",
+ "# Programmatically find the \"optimal\" (maximum) value in our dictionary\n",
+ "optimal_threshold = max(results, key=results.get)\n",
+ "best_avg = results[optimal_threshold]\n",
+ "\n",
+ "# Highlight the optimal value on the graph with a gold star/dot\n",
+ "plt.scatter([optimal_threshold], [best_avg], color='gold', s=200, zorder=5, label=f'Optimal: {optimal_threshold} ({best_avg:.1f} chips)')\n",
+ "plt.axhline(0, color='red', linestyle='dashed', alpha=0.5, label='Break Even (0)')\n",
+ "\n",
+ "plt.title('Average Winnings vs. Hit Threshold\\n(100 Games of 50 Rounds per Threshold)')\n",
+ "plt.xlabel('Hit Threshold Value')\n",
+ "plt.ylabel('Average Net Winnings (Chips)')\n",
+ "plt.xticks(thresholds_to_test)\n",
+ "plt.grid(True, alpha=0.3)\n",
+ "plt.legend()\n",
+ "\n",
+ "# Display the final chart\n",
+ "plt.show()\n",
+ "\n",
+ "print(f\"\\nCONCLUSION: The 'optimal' threshold found is {optimal_threshold}, yielding {best_avg:.2f} chips on average.\")"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
"source": [
"10. Create a new strategy based on web searches or your own ideas. Demonstrate that the new strategy will result in increased or decreased winnings. "
]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "
The Strategy: Martingale
\n",
+ "\n",
+ "\n",
+ "Instead of changing how we play our cards, we are going to change how we bet.\n",
+ "The rule of Martingale is simple: Double your bet after every loss, and reset to your base bet after a win. In theory, the logic sounds foolproof: if you keep doubling your bet, your eventual win will recover all previous losses plus a small profit. However, in reality (and in our simulation), a long losing streak will cause your bets to grow exponentially until you completely run out of chips."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "class MartingalePlayer(Player):\n",
+ " def __init__(self, name, starting_chips=1000, base_bet=10):\n",
+ " # Initialize using the parent class\n",
+ " super().__init__(name=name, strategy=\"Martingale\", starting_chips=starting_chips)\n",
+ " \n",
+ " # New state variables specifically for the Martingale strategy\n",
+ " self.base_bet = base_bet\n",
+ " self.last_bet = base_bet\n",
+ " self.last_chips = starting_chips \n",
+ " self.first_bet_made = False\n",
+ "\n",
+ " def place_bet(self, true_count=0.0):\n",
+ " if self.chips <= 0:\n",
+ " self.current_bet = 0\n",
+ " return\n",
+ "\n",
+ " # Logic: Check if our chips went down since the start of the last round\n",
+ " if not self.first_bet_made:\n",
+ " bet = self.base_bet\n",
+ " self.first_bet_made = True\n",
+ " else:\n",
+ " if self.chips < self.last_chips:\n",
+ " # We lost the last hand! Double the bet.\n",
+ " bet = self.last_bet * 2\n",
+ " else:\n",
+ " # We won or pushed. Reset to base bet.\n",
+ " bet = self.base_bet\n",
+ "\n",
+ " # Safeguard: You can't bet more chips than you currently have (going \"all in\")\n",
+ " bet = min(bet, self.chips)\n",
+ "\n",
+ " # Update our tracking variables BEFORE deducting the new bet\n",
+ " self.last_chips = self.chips\n",
+ " self.last_bet = bet\n",
+ " \n",
+ " self.current_bet = bet\n",
+ " self.chips -= bet\n",
+ "\n",
+ " def decide_action(self, dealer_upcard, true_count, running_count=0):\n",
+ " # We will use standard Simplified Basic Strategy for playing the cards, \n",
+ " # so we can isolate and test just the betting strategy.\n",
+ " dealer_val = 10 if dealer_upcard.rank in ['J', 'Q', 'K', 'A'] else int(dealer_upcard.rank)\n",
+ " score = self.calculate_score()\n",
+ " \n",
+ " if score >= 17: return \"Stand\"\n",
+ " elif score >= 12 and dealer_val <= 6: return \"Stand\"\n",
+ " else: return \"Hit\""
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Simulating 100 games to compare Martingale vs. Flat Betting... \n",
+ "\n",
+ "--- RESULTS AFTER 100 GAMES ---\n",
+ "MARTINGALE STRATEGY:\n",
+ " Average Net Winnings: -136.10 chips\n",
+ " Bankruptcy Rate: 20.0%\n",
+ " Win Rate (Finished with >1000 chips): 62.0%\n",
+ "\n",
+ "CONTROL STRATEGY (Flat Betting):\n",
+ " Average Net Winnings: -35.40 chips\n",
+ " Bankruptcy Rate: 0.0%\n",
+ " Win Rate (Finished with >1000 chips): 37.0%\n"
+ ]
+ },
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ "
"
]
@@ -1238,6 +1091,60 @@
"Now use `matplotlib` to reproduce as closely as you can figures 5 and 6 from the paper. This exercise is intended to get you to familiarize yourself with making nicely formatted `matplotlib` figures with multiple plots. Note that the plots in the paper are actually wrong!"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 34,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ "
"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "# Exercise 4.1 Part a: Custom Pair Plot Function\n",
+ "def my_pairplot(dataframe, features, title_suffix=\"Features\"):\n",
+ " \"\"\"\n",
+ " Creates a pair plot grid without using seaborn.\n",
+ " Diagonals: 1D histograms showing overlapping class distributions.\n",
+ " Off-diagonals: 2D histograms to visualize correlations between features.\n",
+ " \"\"\"\n",
+ " n = len(features)\n",
+ " fig, axes = plt.subplots(n, n, figsize=(n*2, n*2))\n",
+ " fig.suptitle(f'Pair Plots of SUSY {title_suffix}', fontsize=16)\n",
+ "\n",
+ " # Pre-split data to avoid repeated filtering inside the loops\n",
+ " sig = dataframe[dataframe['signal'] == 1]\n",
+ " bkg = dataframe[dataframe['signal'] == 0]\n",
+ "\n",
+ " for i in range(n):\n",
+ " for j in range(n):\n",
+ " ax = axes[i, j]\n",
+ " \n",
+ " if i == j: # Diagonal: 1D Histograms\n",
+ " ax.hist(sig[features[i]], bins=30, histtype='step', color='red', density=True)\n",
+ " ax.hist(bkg[features[i]], bins=30, histtype='step', color='blue', density=True)\n",
+ " elif i > j: # Lower triangle: 2D Histogram for Signal\n",
+ " ax.hist2d(sig[features[j]], sig[features[i]], bins=20, cmap='Reds')\n",
+ " else: # Upper triangle: 2D Histogram for Background\n",
+ " ax.hist2d(bkg[features[j]], bkg[features[i]], bins=20, cmap='Blues')\n",
+ "\n",
+ " # Formatting to keep the grid clean\n",
+ " if i < n - 1: ax.set_xticklabels([])\n",
+ " if j > 0: ax.set_yticklabels([])\n",
+ " if i == n - 1: ax.set_xlabel(features[j], fontsize=8)\n",
+ " if j == 0: ax.set_ylabel(features[i], fontsize=8)\n",
+ "\n",
+ " plt.tight_layout(rect=[0, 0.03, 1, 0.95])\n",
+ " plt.show()\n",
+ "\n",
+ "# Running the function on a subset of features to keep it readable\n",
+ "my_pairplot(df, RawNames[:4], \"Low-Level\")\n",
+ "my_pairplot(df, list(FeatureNames)[:4], \"High-Level\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 26,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "\"\\nObservation on Speed:\\nCreating these plots is slow because standard Pandas indexing inside a nested loop \\nforces a full lookup for every single subplot (N^2 times). \\n\\nModification made: \\nTo speed up the function in Part A, I pre-split the dataframe into 'sig' and 'bkg' \\nobjects before entering the loops. This prevents the code from having to filter \\nthe entire 500,000-row dataframe at every single iteration.\\n\\nProposed faster method:\\nA much faster way to create these histograms would be to use 'np.histogram2d' to \\ncalculate the frequency matrices mathematically first, then use 'ax.imshow()' \\nto render them as images. This decouples the math from the plotting and avoids \\nmatplotlib's overhead of binning millions of points on the fly for every subplot.\\n\""
+ ]
+ },
+ "execution_count": 26,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# Exercise 4.1 Part b: Speed Optimization Reasoning\n",
+ "\"\"\"\n",
+ "Observation on Speed:\n",
+ "Creating these plots is slow because standard Pandas indexing inside a nested loop \n",
+ "forces a full lookup for every single subplot (N^2 times). \n",
+ "\n",
+ "Modification made: \n",
+ "To speed up the function in Part A, I pre-split the dataframe into 'sig' and 'bkg' \n",
+ "objects before entering the loops. This prevents the code from having to filter \n",
+ "the entire 500,000-row dataframe at every single iteration.\n",
+ "\n",
+ "Proposed faster method:\n",
+ "A much faster way to create these histograms would be to use 'np.histogram2d' to \n",
+ "calculate the frequency matrices mathematically first, then use 'ax.imshow()' \n",
+ "to render them as images. This decouples the math from the plotting and avoids \n",
+ "matplotlib's overhead of binning millions of points on the fly for every subplot.\n",
+ "\"\"\""
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 27,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "\"\\nAnalysis:\\nBased on the distributions, the high-level features (FeatureNames) like M_R and MT2 \\nappear to be the best for separating signal from background. \\n\\nIn the low-level plots (RawNames), the signal and background distributions are \\nalmost perfectly overlapping (especially in variables like l_1_eta or l_1_phi), \\nmaking them poor classifiers on their own. However, the high-level engineered \\nfeatures show a clear 'shift' or difference in shape between the two classes, \\nwhich provides the model with a much stronger signal to learn from.\\n\""
+ ]
+ },
+ "execution_count": 27,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# Exercise 4.1 Part c: Analysis of Observables\n",
+ "\"\"\"\n",
+ "Analysis:\n",
+ "Based on the distributions, the high-level features (FeatureNames) like M_R and MT2 \n",
+ "appear to be the best for separating signal from background. \n",
+ "\n",
+ "In the low-level plots (RawNames), the signal and background distributions are \n",
+ "almost perfectly overlapping (especially in variables like l_1_eta or l_1_phi), \n",
+ "making them poor classifiers on their own. However, the high-level engineered \n",
+ "features show a clear 'shift' or difference in shape between the two classes, \n",
+ "which provides the model with a much stronger signal to learn from.\n",
+ "\"\"\""
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -1275,6 +1317,208 @@
"Write a function that takes a dataset and appropriate arguments and performs steps b and c. "
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 28,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Collecting tabulate\n",
+ " Downloading tabulate-0.10.0-py3-none-any.whl.metadata (40 kB)\n",
+ "Downloading tabulate-0.10.0-py3-none-any.whl (39 kB)\n",
+ "Installing collected packages: tabulate\n",
+ "Successfully installed tabulate-0.10.0\n"
+ ]
+ }
+ ],
+ "source": [
+ "!pip install tabulate"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 31,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "--- Low-Level Features ---\n",
+ "Covariance Matrix:\n"
+ ]
+ },
+ {
+ "data": {
+ "text/html": [
+ "
"
+ ],
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "import numpy as np\n",
+ "import pandas as pd\n",
+ "from IPython.display import HTML, display\n",
+ "import tabulate\n",
+ "\n",
+ "# Exercise 4.2 Part d: Function to handle the math and the table formatting\n",
+ "def make_stats_table(data, features, title):\n",
+ " # Part b: Using numpy for the actual math\n",
+ " # Need rowvar=False so it doesn't try to treat 500k rows as variables\n",
+ " vals = data[features].values\n",
+ " cov_mat = np.cov(vals, rowvar=False)\n",
+ " corr_mat = np.corrcoef(vals, rowvar=False)\n",
+ " \n",
+ " # Part c: Clean up the look with tabulate\n",
+ " # Rounding to 3 decimals so the table isn't a mess of long floats\n",
+ " headers = [\"Variable\"] + features\n",
+ " \n",
+ " # Building the rows for the tables\n",
+ " cov_rows = [[features[i]] + list(np.round(cov_mat[i], 3)) for i in range(len(features))]\n",
+ " corr_rows = [[features[i]] + list(np.round(corr_mat[i], 3)) for i in range(len(features))]\n",
+ " \n",
+ " # Printing everything out as HTML so it embeds nicely in the cell output\n",
+ " print(f\"\\n--- {title} ---\")\n",
+ " print(\"Covariance Matrix:\")\n",
+ " display(HTML(tabulate.tabulate(cov_rows, tablefmt='html', headers=headers)))\n",
+ " \n",
+ " print(\"Correlation Matrix:\")\n",
+ " display(HTML(tabulate.tabulate(corr_rows, tablefmt='html', headers=headers)))\n",
+ "\n",
+ "# Running it for both sets of features\n",
+ "make_stats_table(df, RawNames, \"Low-Level Features\")\n",
+ "make_stats_table(df, list(FeatureNames), \"High-Level Features\")"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -1439,7 +1683,7 @@
],
"metadata": {
"kernelspec": {
- "display_name": "Python 3 (ipykernel)",
+ "display_name": "ds",
"language": "python",
"name": "python3"
},
@@ -1453,7 +1697,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.13.12"
+ "version": "3.11.14"
}
},
"nbformat": 4,
diff --git a/Lectures/.DS_Store b/Lectures/.DS_Store
new file mode 100644
index 00000000..6ffb5a0d
Binary files /dev/null and b/Lectures/.DS_Store differ
diff --git a/Lectures/Lecture.16/Lecture.16.ipynb b/Lectures/Lecture.16/Lecture.16.ipynb
deleted file mode 100644
index 989236c8..00000000
--- a/Lectures/Lecture.16/Lecture.16.ipynb
+++ /dev/null
@@ -1,2639 +0,0 @@
-{
- "cells": [
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "# Lecture 16- Data Analysis"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 1,
- "metadata": {},
- "outputs": [],
- "source": [
- "import pandas as pd\n",
- "import numpy as np\n",
- "import matplotlib.pyplot as plt\n",
- "%matplotlib inline"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "From the last lab, here are the column names for the SUSY Data:"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 2,
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "['S_R',\n",
- " 'MT2',\n",
- " 'R',\n",
- " 'M_TR_2',\n",
- " 'MET_rel',\n",
- " 'M_Delta_R',\n",
- " 'axial_MET',\n",
- " 'M_R',\n",
- " 'cos_theta_r1',\n",
- " 'dPhi_r_b']"
- ]
- },
- "execution_count": 2,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "VarNames=[\"signal\", \"l_1_pT\", \"l_1_eta\",\"l_1_phi\", \"l_2_pT\", \"l_2_eta\", \"l_2_phi\", \"MET\", \"MET_phi\", \"MET_rel\", \"axial_MET\", \"M_R\", \"M_TR_2\", \"R\", \"MT2\", \"S_R\", \"M_Delta_R\", \"dPhi_r_b\", \"cos_theta_r1\"]\n",
- "RawNames=[\"l_1_pT\", \"l_1_eta\",\"l_1_phi\", \"l_2_pT\", \"l_2_eta\", \"l_2_phi\", \"MET\", \"MET_phi\"]\n",
- "FeatureNames=list(set(VarNames[1:]).difference(RawNames))\n",
- "FeatureNames"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "Let's not download the data again... just use the copy we have in lab 7."
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 3,
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "total 5152472\n",
- "-rw-------@ 1 afarbin staff 401K Mar 20 19:37 Lab.7.ipynb\n",
- "-rw-r--r--@ 1 afarbin staff 8.0M Mar 20 12:18 Lab.7.pdf\n",
- "-rw-r--r--@ 1 afarbin staff 228M Mar 20 13:50 SUSY-small.csv\n",
- "-rw-r--r--@ 1 afarbin staff 2.2G Mar 20 11:36 SUSY.csv\n"
- ]
- }
- ],
- "source": [
- "!ls -lh ../../Labs/Lab.7"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "How much data? Let's count the lines in the CSV file"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 4,
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- " 5000000 5000000 2390277560 ../../Labs/Lab.7/SUSY.csv\n"
- ]
- }
- ],
- "source": [
- "!wc ../../Labs/Lab.7/SUSY.csv"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "Just to make things go faster, lets create a smaller file with 10% of the data."
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 5,
- "metadata": {},
- "outputs": [],
- "source": [
- "!head -500000 ../../Labs/Lab.7/SUSY.csv > SUSY-Small.csv"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "Now load it with Pandas:"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 6,
- "metadata": {},
- "outputs": [],
- "source": [
- "filename=\"SUSY-Small.csv\"\n",
- "df = pd.read_csv(filename, dtype='float64', names=VarNames)"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 7,
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/html": [
- "
\n",
- "\n",
- "
\n",
- " \n",
- "
\n",
- "
\n",
- "
signal
\n",
- "
l_1_pT
\n",
- "
l_1_eta
\n",
- "
l_1_phi
\n",
- "
l_2_pT
\n",
- "
l_2_eta
\n",
- "
l_2_phi
\n",
- "
MET
\n",
- "
MET_phi
\n",
- "
MET_rel
\n",
- "
axial_MET
\n",
- "
M_R
\n",
- "
M_TR_2
\n",
- "
R
\n",
- "
MT2
\n",
- "
S_R
\n",
- "
M_Delta_R
\n",
- "
dPhi_r_b
\n",
- "
cos_theta_r1
\n",
- "
\n",
- " \n",
- " \n",
- "
\n",
- "
0
\n",
- "
0.0
\n",
- "
0.972861
\n",
- "
0.653855
\n",
- "
1.176225
\n",
- "
1.157156
\n",
- "
-1.739873
\n",
- "
-0.874309
\n",
- "
0.567765
\n",
- "
-0.175000
\n",
- "
0.810061
\n",
- "
-0.252552
\n",
- "
1.921887
\n",
- "
0.889637
\n",
- "
0.410772
\n",
- "
1.145621
\n",
- "
1.932632
\n",
- "
0.994464
\n",
- "
1.367815
\n",
- "
0.040714
\n",
- "
\n",
- "
\n",
- "
1
\n",
- "
1.0
\n",
- "
1.667973
\n",
- "
0.064191
\n",
- "
-1.225171
\n",
- "
0.506102
\n",
- "
-0.338939
\n",
- "
1.672543
\n",
- "
3.475464
\n",
- "
-1.219136
\n",
- "
0.012955
\n",
- "
3.775174
\n",
- "
1.045977
\n",
- "
0.568051
\n",
- "
0.481928
\n",
- "
0.000000
\n",
- "
0.448410
\n",
- "
0.205356
\n",
- "
1.321893
\n",
- "
0.377584
\n",
- "
\n",
- "
\n",
- "
2
\n",
- "
1.0
\n",
- "
0.444840
\n",
- "
-0.134298
\n",
- "
-0.709972
\n",
- "
0.451719
\n",
- "
-1.613871
\n",
- "
-0.768661
\n",
- "
1.219918
\n",
- "
0.504026
\n",
- "
1.831248
\n",
- "
-0.431385
\n",
- "
0.526283
\n",
- "
0.941514
\n",
- "
1.587535
\n",
- "
2.024308
\n",
- "
0.603498
\n",
- "
1.562374
\n",
- "
1.135454
\n",
- "
0.180910
\n",
- "
\n",
- "
\n",
- "
3
\n",
- "
1.0
\n",
- "
0.381256
\n",
- "
-0.976145
\n",
- "
0.693152
\n",
- "
0.448959
\n",
- "
0.891753
\n",
- "
-0.677328
\n",
- "
2.033060
\n",
- "
1.533041
\n",
- "
3.046260
\n",
- "
-1.005285
\n",
- "
0.569386
\n",
- "
1.015211
\n",
- "
1.582217
\n",
- "
1.551914
\n",
- "
0.761215
\n",
- "
1.715464
\n",
- "
1.492257
\n",
- "
0.090719
\n",
- "
\n",
- "
\n",
- "
4
\n",
- "
1.0
\n",
- "
1.309996
\n",
- "
-0.690089
\n",
- "
-0.676259
\n",
- "
1.589283
\n",
- "
-0.693326
\n",
- "
0.622907
\n",
- "
1.087562
\n",
- "
-0.381742
\n",
- "
0.589204
\n",
- "
1.365479
\n",
- "
1.179295
\n",
- "
0.968218
\n",
- "
0.728563
\n",
- "
0.000000
\n",
- "
1.083158
\n",
- "
0.043429
\n",
- "
1.154854
\n",
- "
0.094859
\n",
- "
\n",
- "
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
...
\n",
- "
\n",
- "
\n",
- "
499995
\n",
- "
0.0
\n",
- "
0.719035
\n",
- "
1.091879
\n",
- "
0.291540
\n",
- "
1.205962
\n",
- "
-1.599117
\n",
- "
-1.139445
\n",
- "
0.424546
\n",
- "
1.154849
\n",
- "
0.637185
\n",
- "
-0.091178
\n",
- "
1.972156
\n",
- "
0.697028
\n",
- "
0.313636
\n",
- "
0.988602
\n",
- "
1.981573
\n",
- "
0.744828
\n",
- "
1.095080
\n",
- "
0.006546
\n",
- "
\n",
- "
\n",
- "
499996
\n",
- "
1.0
\n",
- "
0.910016
\n",
- "
-0.364544
\n",
- "
-0.777120
\n",
- "
0.543648
\n",
- "
-0.910632
\n",
- "
-1.723707
\n",
- "
2.864673
\n",
- "
1.458272
\n",
- "
2.176558
\n",
- "
-0.590911
\n",
- "
0.673695
\n",
- "
1.662140
\n",
- "
2.189362
\n",
- "
1.195041
\n",
- "
0.910815
\n",
- "
1.181893
\n",
- "
1.252362
\n",
- "
0.826035
\n",
- "
\n",
- "
\n",
- "
499997
\n",
- "
1.0
\n",
- "
0.842954
\n",
- "
0.332476
\n",
- "
-1.048564
\n",
- "
1.347989
\n",
- "
0.320496
\n",
- "
-0.666358
\n",
- "
0.450433
\n",
- "
-0.411872
\n",
- "
0.293407
\n",
- "
0.630491
\n",
- "
0.859920
\n",
- "
0.403371
\n",
- "
0.416258
\n",
- "
0.591989
\n",
- "
0.372003
\n",
- "
0.716788
\n",
- "
0.366991
\n",
- "
0.265798
\n",
- "
\n",
- "
\n",
- "
499998
\n",
- "
0.0
\n",
- "
1.370760
\n",
- "
-1.162912
\n",
- "
0.893499
\n",
- "
2.118091
\n",
- "
1.248496
\n",
- "
-0.887211
\n",
- "
0.164659
\n",
- "
0.316840
\n",
- "
0.215165
\n",
- "
0.280418
\n",
- "
3.087083
\n",
- "
0.526929
\n",
- "
0.151467
\n",
- "
0.308067
\n",
- "
3.098183
\n",
- "
0.233042
\n",
- "
0.876216
\n",
- "
0.000593
\n",
- "
\n",
- "
\n",
- "
499999
\n",
- "
0.0
\n",
- "
0.762400
\n",
- "
0.440924
\n",
- "
0.342885
\n",
- "
1.034283
\n",
- "
1.740353
\n",
- "
-1.083314
\n",
- "
0.872145
\n",
- "
-1.519894
\n",
- "
0.284328
\n",
- "
-0.360861
\n",
- "
0.956828
\n",
- "
0.965979
\n",
- "
0.895881
\n",
- "
1.020396
\n",
- "
0.996446
\n",
- "
0.943458
\n",
- "
1.299870
\n",
- "
0.197220
\n",
- "
\n",
- " \n",
- "
\n",
- "
500000 rows × 19 columns
\n",
- "
"
- ],
- "text/plain": [
- " signal l_1_pT l_1_eta l_1_phi l_2_pT l_2_eta l_2_phi \\\n",
- "0 0.0 0.972861 0.653855 1.176225 1.157156 -1.739873 -0.874309 \n",
- "1 1.0 1.667973 0.064191 -1.225171 0.506102 -0.338939 1.672543 \n",
- "2 1.0 0.444840 -0.134298 -0.709972 0.451719 -1.613871 -0.768661 \n",
- "3 1.0 0.381256 -0.976145 0.693152 0.448959 0.891753 -0.677328 \n",
- "4 1.0 1.309996 -0.690089 -0.676259 1.589283 -0.693326 0.622907 \n",
- "... ... ... ... ... ... ... ... \n",
- "499995 0.0 0.719035 1.091879 0.291540 1.205962 -1.599117 -1.139445 \n",
- "499996 1.0 0.910016 -0.364544 -0.777120 0.543648 -0.910632 -1.723707 \n",
- "499997 1.0 0.842954 0.332476 -1.048564 1.347989 0.320496 -0.666358 \n",
- "499998 0.0 1.370760 -1.162912 0.893499 2.118091 1.248496 -0.887211 \n",
- "499999 0.0 0.762400 0.440924 0.342885 1.034283 1.740353 -1.083314 \n",
- "\n",
- " MET MET_phi MET_rel axial_MET M_R M_TR_2 R \\\n",
- "0 0.567765 -0.175000 0.810061 -0.252552 1.921887 0.889637 0.410772 \n",
- "1 3.475464 -1.219136 0.012955 3.775174 1.045977 0.568051 0.481928 \n",
- "2 1.219918 0.504026 1.831248 -0.431385 0.526283 0.941514 1.587535 \n",
- "3 2.033060 1.533041 3.046260 -1.005285 0.569386 1.015211 1.582217 \n",
- "4 1.087562 -0.381742 0.589204 1.365479 1.179295 0.968218 0.728563 \n",
- "... ... ... ... ... ... ... ... \n",
- "499995 0.424546 1.154849 0.637185 -0.091178 1.972156 0.697028 0.313636 \n",
- "499996 2.864673 1.458272 2.176558 -0.590911 0.673695 1.662140 2.189362 \n",
- "499997 0.450433 -0.411872 0.293407 0.630491 0.859920 0.403371 0.416258 \n",
- "499998 0.164659 0.316840 0.215165 0.280418 3.087083 0.526929 0.151467 \n",
- "499999 0.872145 -1.519894 0.284328 -0.360861 0.956828 0.965979 0.895881 \n",
- "\n",
- " MT2 S_R M_Delta_R dPhi_r_b cos_theta_r1 \n",
- "0 1.145621 1.932632 0.994464 1.367815 0.040714 \n",
- "1 0.000000 0.448410 0.205356 1.321893 0.377584 \n",
- "2 2.024308 0.603498 1.562374 1.135454 0.180910 \n",
- "3 1.551914 0.761215 1.715464 1.492257 0.090719 \n",
- "4 0.000000 1.083158 0.043429 1.154854 0.094859 \n",
- "... ... ... ... ... ... \n",
- "499995 0.988602 1.981573 0.744828 1.095080 0.006546 \n",
- "499996 1.195041 0.910815 1.181893 1.252362 0.826035 \n",
- "499997 0.591989 0.372003 0.716788 0.366991 0.265798 \n",
- "499998 0.308067 3.098183 0.233042 0.876216 0.000593 \n",
- "499999 1.020396 0.996446 0.943458 1.299870 0.197220 \n",
- "\n",
- "[500000 rows x 19 columns]"
- ]
- },
- "execution_count": 7,
- "metadata": {},
- "output_type": "execute_result"
- }
- ],
- "source": [
- "df"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "And separate signal and background into different DataFrames for easy reference:"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 8,
- "metadata": {},
- "outputs": [],
- "source": [
- "df_sig=df[df.signal==1]\n",
- "df_bkg=df[df.signal==0]"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 9,
- "metadata": {},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "(229245, 19)\n",
- "(270755, 19)\n"
- ]
- },
- {
- "name": "stderr",
- "output_type": "stream",
- "text": [
- "ERROR:tornado.general:Uncaught exception in ZMQStream callback\n",
- "Traceback (most recent call last):\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/zmq/eventloop/zmqstream.py\", line 565, in _log_error\n",
- " f.result()\n",
- " ~~~~~~~~^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/ipykernel/kernelbase.py\", line 584, in shell_channel_thread_main\n",
- " _, msg2 = self.session.feed_identities(msg, copy=False)\n",
- " ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/jupyter_client/session.py\", line 998, in feed_identities\n",
- " raise ValueError(msg)\n",
- "ValueError: DELIM not in msg_list\n",
- "ERROR:tornado.general:Uncaught exception in ZMQStream callback\n",
- "Traceback (most recent call last):\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/zmq/eventloop/zmqstream.py\", line 565, in _log_error\n",
- " f.result()\n",
- " ~~~~~~~~^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/ipykernel/kernelbase.py\", line 584, in shell_channel_thread_main\n",
- " _, msg2 = self.session.feed_identities(msg, copy=False)\n",
- " ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/jupyter_client/session.py\", line 998, in feed_identities\n",
- " raise ValueError(msg)\n",
- "ValueError: DELIM not in msg_list\n",
- "ERROR:tornado.general:Uncaught exception in ZMQStream callback\n",
- "Traceback (most recent call last):\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/zmq/eventloop/zmqstream.py\", line 565, in _log_error\n",
- " f.result()\n",
- " ~~~~~~~~^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/ipykernel/kernelbase.py\", line 584, in shell_channel_thread_main\n",
- " _, msg2 = self.session.feed_identities(msg, copy=False)\n",
- " ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/jupyter_client/session.py\", line 998, in feed_identities\n",
- " raise ValueError(msg)\n",
- "ValueError: DELIM not in msg_list\n",
- "ERROR:tornado.general:Uncaught exception in ZMQStream callback\n",
- "Traceback (most recent call last):\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/zmq/eventloop/zmqstream.py\", line 565, in _log_error\n",
- " f.result()\n",
- " ~~~~~~~~^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/ipykernel/kernelbase.py\", line 584, in shell_channel_thread_main\n",
- " _, msg2 = self.session.feed_identities(msg, copy=False)\n",
- " ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/jupyter_client/session.py\", line 998, in feed_identities\n",
- " raise ValueError(msg)\n",
- "ValueError: DELIM not in msg_list\n",
- "ERROR:tornado.general:Uncaught exception in ZMQStream callback\n",
- "Traceback (most recent call last):\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/zmq/eventloop/zmqstream.py\", line 565, in _log_error\n",
- " f.result()\n",
- " ~~~~~~~~^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/ipykernel/kernelbase.py\", line 348, in dispatch_control\n",
- " await self.process_control(msg)\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/ipykernel/kernelbase.py\", line 354, in process_control\n",
- " idents, msg = self.session.feed_identities(msg, copy=False)\n",
- " ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/jupyter_client/session.py\", line 998, in feed_identities\n",
- " raise ValueError(msg)\n",
- "ValueError: DELIM not in msg_list\n",
- "ERROR:tornado.general:Uncaught exception in ZMQStream callback\n",
- "Traceback (most recent call last):\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/zmq/eventloop/zmqstream.py\", line 565, in _log_error\n",
- " f.result()\n",
- " ~~~~~~~~^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/ipykernel/kernelbase.py\", line 348, in dispatch_control\n",
- " await self.process_control(msg)\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/ipykernel/kernelbase.py\", line 354, in process_control\n",
- " idents, msg = self.session.feed_identities(msg, copy=False)\n",
- " ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/jupyter_client/session.py\", line 998, in feed_identities\n",
- " raise ValueError(msg)\n",
- "ValueError: DELIM not in msg_list\n",
- "ERROR:tornado.general:Uncaught exception in ZMQStream callback\n",
- "Traceback (most recent call last):\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/zmq/eventloop/zmqstream.py\", line 565, in _log_error\n",
- " f.result()\n",
- " ~~~~~~~~^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/ipykernel/kernelbase.py\", line 348, in dispatch_control\n",
- " await self.process_control(msg)\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/ipykernel/kernelbase.py\", line 354, in process_control\n",
- " idents, msg = self.session.feed_identities(msg, copy=False)\n",
- " ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/jupyter_client/session.py\", line 998, in feed_identities\n",
- " raise ValueError(msg)\n",
- "ValueError: DELIM not in msg_list\n",
- "ERROR:tornado.general:Uncaught exception in ZMQStream callback\n",
- "Traceback (most recent call last):\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/zmq/eventloop/zmqstream.py\", line 565, in _log_error\n",
- " f.result()\n",
- " ~~~~~~~~^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/ipykernel/kernelbase.py\", line 348, in dispatch_control\n",
- " await self.process_control(msg)\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/ipykernel/kernelbase.py\", line 354, in process_control\n",
- " idents, msg = self.session.feed_identities(msg, copy=False)\n",
- " ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/jupyter_client/session.py\", line 998, in feed_identities\n",
- " raise ValueError(msg)\n",
- "ValueError: DELIM not in msg_list\n",
- "ERROR:tornado.general:Uncaught exception in ZMQStream callback\n",
- "Traceback (most recent call last):\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/zmq/eventloop/zmqstream.py\", line 565, in _log_error\n",
- " f.result()\n",
- " ~~~~~~~~^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/ipykernel/kernelbase.py\", line 348, in dispatch_control\n",
- " await self.process_control(msg)\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/ipykernel/kernelbase.py\", line 354, in process_control\n",
- " idents, msg = self.session.feed_identities(msg, copy=False)\n",
- " ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^\n",
- " File \"/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/site-packages/jupyter_client/session.py\", line 998, in feed_identities\n",
- " raise ValueError(msg)\n",
- "ValueError: DELIM not in msg_list\n",
- "Bad pipe message: %s [b'\\xd8\\xc5\\xe0\\xcd\\xd9yAB\\x81\\xe7XIpUda\\xfar\\x00\\x01|\\x00\\x00\\x00\\x01\\x00\\x02\\x00\\x03\\x00\\x04\\x00\\x05\\x00\\x06\\x00\\x07\\x00\\x08\\x00\\t\\x00\\n\\x00\\x0b\\x00\\x0c\\x00\\r\\x00\\x0e\\x00\\x0f\\x00\\x10\\x00\\x11\\x00\\x12\\x00\\x13\\x00\\x14\\x00\\x15\\x00\\x16\\x00\\x17\\x00\\x18\\x00\\x19\\x00\\x1a\\x00\\x1b\\x00/\\x000\\x001\\x002\\x003\\x004\\x005\\x006\\x007\\x008\\x009\\x00:\\x00;\\x00<\\x00=\\x00>\\x00?\\x00@\\x00A\\x00B\\x00C\\x00D\\x00E\\x00F\\x00g\\x00h\\x00i\\x00j\\x00k\\x00l\\x00m\\x00\\x84\\x00\\x85\\x00\\x86\\x00\\x87\\x00\\x88\\x00\\x89\\x00\\x96\\x00\\x97\\x00\\x98\\x00\\x99\\x00\\x9a\\x00\\x9b\\x00\\x9c\\x00\\x9d\\x00\\x9e\\x00\\x9f\\x00\\xa0\\x00\\xa1\\x00\\xa2\\x00\\xa3\\x00\\xa4\\x00\\xa5\\x00\\xa6\\x00\\xa7\\x00\\xba\\x00\\xbb\\x00\\xbc\\x00\\xbd\\x00\\xbe\\x00\\xbf\\x00\\xc0\\x00\\xc1\\x00\\xc2\\x00\\xc3\\x00\\xc4\\x00\\xc5\\x13\\x01\\x13\\x02\\x13\\x03', b'\\x13\\x05\\xc0\\x01\\xc0\\x02\\xc0\\x03\\xc0\\x04\\xc0\\x05\\xc0\\x06\\xc0\\x07\\xc0\\x08']\n",
- "Bad pipe message: %s [b'Y\\xec.\\xb5k\\xb8\\xb4&u\\xd5Bd\\xa4\\xca\\xf9\\x1d]\\x83\\x00\\x01|\\x00\\x00\\x00\\x01\\x00\\x02\\x00\\x03\\x00\\x04\\x00\\x05\\x00\\x06\\x00\\x07\\x00\\x08\\x00\\t\\x00\\n\\x00\\x0b\\x00\\x0c\\x00\\r\\x00\\x0e\\x00\\x0f\\x00\\x10\\x00\\x11\\x00']\n",
- "Bad pipe message: %s [b'\\x13\\x00\\x14\\x00\\x15\\x00\\x16\\x00\\x17\\x00\\x18\\x00\\x19\\x00\\x1a\\x00\\x1b']\n",
- "Bad pipe message: %s [b'&\\x81 \\xdb\\xf6!N\\x16\\x7f2)\\x17\\xb5K\\x11\\x02\\xac\\xab\\x00\\x01|\\x00\\x00\\x00\\x01\\x00\\x02\\x00\\x03\\x00\\x04\\x00\\x05\\x00\\x06\\x00\\x07\\x00\\x08\\x00\\t\\x00\\n\\x00\\x0b\\x00\\x0c\\x00\\r\\x00\\x0e\\x00\\x0f\\x00\\x10\\x00\\x11\\x00\\x12\\x00\\x13\\x00\\x14\\x00\\x15\\x00\\x16\\x00\\x17\\x00\\x18\\x00\\x19\\x00\\x1a\\x00\\x1b\\x00/\\x000\\x001\\x002\\x003\\x004\\x005\\x006\\x007\\x008\\x009\\x00:\\x00;\\x00<\\x00=\\x00>\\x00?\\x00@\\x00A\\x00B\\x00C\\x00D\\x00E\\x00F\\x00g']\n"
- ]
- }
- ],
- "source": [
- "print(df_sig.shape)\n",
- "print(df_bkg.shape)"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Create Data Like Sample\n",
- "\n",
- "First, lets look back to Lab 7 and try to identify one feature that seems to exhibit good separation between signal and background events. \n",
- "\n",
- "`M_TR_2` seems to by such variable...\n",
- "\n",
- "Now lets try to recreate what the distribution of this variable in real data coming from the accelerator would look like. Remember that the data we are using now is simulated, not real. And it was created in a way that we have roughly the same number of signal and background events, and lots of them. \n",
- "\n",
- "In the real data, we would expect very different mixture of signal and background events. Let's imagine we expected 100 background events and 10 signal events. We can use the simulated data to estimate what the distribution of `M_TR_2` may look like in real data."
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 10,
- "metadata": {},
- "outputs": [
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "
"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
- "source": [
- "import math\n",
- "n_bkg_expected = 100\n",
- "n_sig_expected = 10\n",
- "\n",
- "# Mix signal and background events:\n",
- "data = df_sig[:n_sig_expected][\"M_TR_2\"].tolist()+df_bkg[:n_bkg_expected][\"M_TR_2\"].tolist()\n",
- "\n",
- "# Histogram the data\n",
- "c , bin_edges= np.histogram(data,bins=20)\n",
- "\n",
- "# Correctly compute the errors on each bin\n",
- "# Each bin obey Poisson statistics, which means \n",
- "# that the standard deviation can be approximated \n",
- "# as the sqrt of the mean value\n",
- "error = np.sqrt(c)\n",
- "\n",
- "# Now we need to plot everything together...\n",
- "# Unfortunately it's not so pretty in Matplotlib\n",
- "bin_width = bin_edges[1]-bin_edges[0]\n",
- "bin_centers = 0.5*(bin_edges[1:]+bin_edges[:-1])\n",
- "\n",
- "plt.errorbar(bin_centers, c, yerr=error, marker=\"o\")\n",
- "plt.show()"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Obtain Probability Density Functions from Simulation\n",
- "\n",
- "Now lets use the simulation to see what the probability distributions of `M_TR_2` looks like for the signal and background. Note that we are really wanting to see the Probability Distribution Functions (PDFs). `matplotlib` histogram function plots PDFs for us by specifying `density=1`. In order to understand what density means, lets use `numpy` and do the computation ourselves."
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 11,
- "metadata": {},
- "outputs": [
- {
- "data": {
- "text/plain": [
- "[]"
- ]
- },
- "execution_count": 11,
- "metadata": {},
- "output_type": "execute_result"
- },
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "
"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
- "source": [
- "plt.errorbar(bin_centers, c, yerr=error, marker=\"o\",label=\"data\")\n",
- "plt.yscale(\"log\")\n",
- "plt.step(sim_bin[1:],c_bkg_sim_normalized,label=\"background\")\n",
- "plt.step(sim_bin[1:],(c_bkg_sim_normalized+c_sig_sim_normalized),label=\"background + signal\")\n",
- "plt.legend()\n",
- "plt.show()"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Calculate TPR / FPR\n",
- "\n",
- "Next we want to think about how we could use this feature to alter counting of signal and backgrounds in a way that is most favorable to signal. As we see, signal events have larger values of our feature, so if we select events with high values of this feature, we'll be left with a sample with a more favorable signal to background ratio. Now the question is, what threshold value should we select?\n",
- "\n",
- "First, lets compute the probability of signal and background passing the \"threshold cut\", as function of where we place the \"cut\". These probabilties are know as the True Positive Rate (TPR) and False Positive Rate (FPR).\n",
- "\n",
- "To to this, we have to integrate (actually just count) from the threshold to postive infinity:"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 16,
- "metadata": {},
- "outputs": [
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "
"
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
- "source": [
- "# Plot the signal distribution\n",
- "_=plt.hist(df_sig[\"M_TR_2\"],bins=100,histtype=\"step\")"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "We can just count and sum above a selected threshold... this operation is referred to as a cumlative sum. \n",
- "\n",
- "`matplotlib` will actually do this in one step using the `cumulative` argument: "
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 17,
- "metadata": {},
- "outputs": [
- {
- "data": {
- "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiMAAAGdCAYAAADAAnMpAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAhv0lEQVR4nO3df1SWhf3/8Reg9w2CIuoAIRwWbWqYlEwOWqd1uhdrzo5np8XMJaPlTg2+U/msKZU6P5VUm4x+MJk21/Y9OW2dai2dHSO144miIHbqrDSXpYcG6DcQvb3jZjf3948d77oD1BuBN3A/H+fcf9wX18X95joKz3Pd131dEX6/3y8AAAAjkdYDAACA8EaMAAAAU8QIAAAwRYwAAABTxAgAADBFjAAAAFPECAAAMEWMAAAAU6OsBzgfXV1d+uSTTzR27FhFRERYjwMAAM6D3+/XyZMnlZKSosjI3o9/DIsY+eSTT5SWlmY9BgAA6IOjR4/qoosu6vXrwyJGxo4dK+m/P8y4ceOMpwEAAOejvb1daWlpgb/jvRkWMXLmrZlx48YRIwAADDPnOsWCE1gBAIApYgQAAJgiRgAAgCliBAAAmCJGAACAKWIEAACYIkYAAIApYgQAAJgiRgAAgCliBAAAmAo5Rl599VUtWLBAKSkpioiI0PPPP3/Obfbu3asrr7xSTqdTGRkZevLJJ/swKgAAGIlCjhG3261Zs2apsrLyvNY/fPiw5s+fr2uvvVYNDQ1avny5br/9dr300kshDwsAAEaekG+Ud8MNN+iGG2447/Wrqqo0depUbdiwQZI0ffp07d+/X7/5zW+Ul5cX6ssDAIARZsDv2ltTUyOXyxW0LC8vT8uXL+91m46ODnV0dASet7e3D9R4AC5AY5tHrW6v9RgALlBCrEOp42PMXn/AY6SpqUlJSUlBy5KSktTe3i6Px6OYmO4/fFlZmdatWzfQowGDZiT+0f5/bq/u+L918nT6rEcBcIFiRkfp5f+5xixIBjxG+qK0tFQlJSWB5+3t7UpLSzOcCOGqPyJiJP/RjhkdpT/eNkcTYx3WowDoo0Mtp7R8e4Na3d6RGyPJyclqbm4OWtbc3Kxx48b1eFREkpxOp5xO50CPhjB3rtDoz4gYqX+0rQ/tAhgZBjxGcnNztXPnzqBlu3fvVm5u7kC/NMJYf4VGf0UEf7QBoHchx8ipU6d06NChwPPDhw+roaFBEyZM0JQpU1RaWqrGxkb96U9/kiTdcccdevzxx/WLX/xCt912m1555RU9/fTT2rFjR//9FAhrXw6P/gwNIgIABl7IMfLWW2/p2muvDTw/c25HQUGBnnzySf373//WkSNHAl+fOnWqduzYoRUrVuiRRx7RRRddpCeeeIKP9aJPzjc8CA0AGD5CjpFvfvOb8vv9vX69p6urfvOb39Tbb78d6ksBQfERSngQGgAwfAzJT9MgPJ3PUQ/CAwBGHmIEQ0Jjm0euDfvOedSD8ACAkYcYgZkvHgk51HJKnk6fKvKzlJEYF1iH+ACAkY8YgYmejoTEjI7SN6ZOID4AIMwQIxgUXz4fpKcjIRwFAYDwRIxgwJ3tfBCOhAAAiBEMCM4HAQCcL2IE/Y7zQQAAoSBG0O9a3V7OBwEAnDdiBBesp5NTJSkjMU6ZqfFWYwEAhgliBBfkbCenJlzgnW4BAOGBGMEF6ektGYm3ZQAA548YQb/gLRkAQF8RIwjZlz+2CwDAhSBGEJLePrbL+SEAgL4iRhASPrYLAOhvxAj6hHNEAAD9JdJ6AAAAEN44MoKz6u2CZgAA9BdiBL3igmYAgMFAjKBXXNAMADAYiBGcEyerAgAGEiewAgAAU8QIAAAwRYwAAABTnDOCINx3BgAw2IgRBHDfGQCABWIEAdx3BgBggRhBN3yUFwAwmDiBFQAAmCJGAACAKWIEAACYIkYAAIApTmANY1+8pojEdUUAADaIkTDV0zVFJK4rAgAYfMRImOrpmiIS1xUBAAw+YiTMcU0RAIA1TmAFAACmiBEAAGCKGAEAAKaIEQAAYIoYAQAApogRAABgihgBAACmuM5IGPni5d+59DsAYKggRsJET5d/59LvAIChgBgJEz1d/p1LvwMAhgJiJMxw+XcAwFDDCawAAMAUMQIAAEwRIwAAwBQxAgAATBEjAADAFDECAABMESMAAMAUMQIAAEwRIwAAwBQxAgAATHE5+BHqi3folbhLLwBg6CJGRqCe7tArcZdeAMDQ1Ke3aSorK5Wenq7o6Gjl5OSotrb2rOtXVFTo61//umJiYpSWlqYVK1bos88+69PAOLcv3qH3xf9zVeDx8v9cw116AQBDTshHRrZv366SkhJVVVUpJydHFRUVysvL04EDB5SYmNht/a1bt2rVqlXasmWL5s6dq4MHD+pHP/qRIiIiVF5e3i8/BHrGHXoBAMNByEdGysvLtXTpUhUWFmrGjBmqqqrSmDFjtGXLlh7Xf+211zRv3jzdcsstSk9P1/XXX69Fixad82gKAAAIDyHFiNfrVV1dnVwu1+ffIDJSLpdLNTU1PW4zd+5c1dXVBeLjww8/1M6dO/Wd73yn19fp6OhQe3t70AMAAIxMIb1Nc/z4cfl8PiUlJQUtT0pK0vvvv9/jNrfccouOHz+uq666Sn6/X//5z390xx136O677+71dcrKyrRu3bpQRgMAAMPUgF9nZO/evVq/fr1++9vfqr6+Xs8++6x27Nih++67r9dtSktLdeLEicDj6NGjAz0mAAAwEtKRkUmTJikqKkrNzc1By5ubm5WcnNzjNqtXr9att96q22+/XZI0c+ZMud1u/eQnP9E999yjyMjuPeR0OuV0OkMZDQAADFMhHRlxOByaPXu2qqurA8u6urpUXV2t3NzcHrc5ffp0t+CIioqSJPn9/lDnBQAAI0zIH+0tKSlRQUGBsrOzNWfOHFVUVMjtdquwsFCStGTJEqWmpqqsrEyStGDBApWXl+uKK65QTk6ODh06pNWrV2vBggWBKAEAAOEr5BjJz8/XsWPHtGbNGjU1NSkrK0u7du0KnNR65MiRoCMh9957ryIiInTvvfeqsbFRX/nKV7RgwQI98MAD/fdTAACAYatPl4MvLi5WcXFxj1/bu3dv8AuMGqW1a9dq7dq1fXkpAAAwwnHXXgAAYIoYAQAApogRAABgihgBAACm+nQCK4aexjaPWt1eSdKhllPG0wAAcP6IkRGgsc0j14Z98nT6AstiRkcpIdZhOBUAAOeHGBkBWt1eeTp9qsjPUkZinCQpIdah1PExxpMBAHBuxMgIkpEYp8zUeOsxAAAICSewAgAAU8QIAAAwRYwAAABTxAgAADBFjAAAAFPECAAAMEWMAAAAU8QIAAAwRYwAAABTxAgAADBFjAAAAFPECAAAMEWMAAAAU8QIAAAwRYwAAABTxAgAADBFjAAAAFPECAAAMEWMAAAAU6OsB0DoGts8anV7A88PtZwynAYAgAtDjAwzjW0euTbsk6fTF7Q8ZnSUEmIdRlMBANB3xMgw0+r2ytPpU0V+ljIS4wLLE2IdSh0fYzgZAAB9Q4wMUxmJccpMjbceAwCAC8YJrAAAwBQxAgAATBEjAADAFDECAABMESMAAMAUMQIAAEwRIwAAwBQxAgAATBEjAADAFDECAABMESMAAMAUMQIAAEwRIwAAwBQxAgAATBEjAADAFDECAABMESMAAMAUMQIAAEwRIwAAwBQxAgAATBEjAADAFDECAABMjbIeAOfW2OZRq9srSTrUcsp4GgAA+hcxMsQ1tnnk2rBPnk5fYFnM6CglxDoMpwIAoP8QI0Ncq9srT6dPFflZykiMkyQlxDqUOj7GeDIAAPoHMTJMZCTGKTM13noMAAD6HSewAgAAU8QIAAAw1acYqaysVHp6uqKjo5WTk6Pa2tqzrt/W1qaioiJNnjxZTqdTX/va17Rz584+DQwAAEaWkM8Z2b59u0pKSlRVVaWcnBxVVFQoLy9PBw4cUGJiYrf1vV6vvvWtbykxMVHPPPOMUlNT9fHHH2v8+PH9MT8AABjmQo6R8vJyLV26VIWFhZKkqqoq7dixQ1u2bNGqVau6rb9lyxZ9+umneu211zR69GhJUnp6+oVNDQAARoyQ3qbxer2qq6uTy+X6/BtERsrlcqmmpqbHbV544QXl5uaqqKhISUlJyszM1Pr16+Xz+XpcX5I6OjrU3t4e9AAAACNTSDFy/Phx+Xw+JSUlBS1PSkpSU1NTj9t8+OGHeuaZZ+Tz+bRz506tXr1aGzZs0P3339/r65SVlSk+Pj7wSEtLC2VMAAAwjAz4p2m6urqUmJioTZs2afbs2crPz9c999yjqqqqXrcpLS3ViRMnAo+jR48O9JgAAMBISOeMTJo0SVFRUWpubg5a3tzcrOTk5B63mTx5skaPHq2oqKjAsunTp6upqUler1cOR/fLmjudTjmdzlBGAwAAw1RIR0YcDodmz56t6urqwLKuri5VV1crNze3x23mzZunQ4cOqaurK7Ds4MGDmjx5co8hAgAAwkvIb9OUlJRo8+bN+uMf/6j33ntPd955p9xud+DTNUuWLFFpaWlg/TvvvFOffvqpli1bpoMHD2rHjh1av369ioqK+u+nAAAAw1bIH+3Nz8/XsWPHtGbNGjU1NSkrK0u7du0KnNR65MgRRUZ+3jhpaWl66aWXtGLFCl1++eVKTU3VsmXLtHLlyv77KQAAwLDVpxvlFRcXq7i4uMev7d27t9uy3Nxcvf766315KQAAMMJxbxoAAGCKGAEAAKaIEQAAYIoYAQAApogRAABgihgBAACmiBEAAGCKGAEAAKaIEQAAYIoYAQAApogRAABgihgBAACmiBEAAGCKGAEAAKZGWQ+AYI1tHrW6vYHnh1pOGU4DAMDAI0aGkMY2j1wb9snT6QtaHjM6SgmxDqOpAAAYWMTIENLq9srT6VNFfpYyEuMCyxNiHUodH2M4GQAAA4cYGYIyEuOUmRpvPQYAAIOCE1gBAIApYgQAAJgiRgAAgCliBAAAmCJGAACAKWIEAACYIkYAAIApYgQAAJgiRgAAgCliBAAAmCJGAACAKWIEAACYIkYAAIApYgQAAJgiRgAAgCliBAAAmCJGAACAKWIEAACYIkYAAIApYgQAAJgiRgAAgCliBAAAmCJGAACAKWIEAACYIkYAAIApYgQAAJgiRgAAgCliBAAAmCJGAACAKWIEAACYIkYAAICpUdYDhLvGNo9a3V5J0qGWU8bTAAAw+IgRQ41tHrk27JOn0xdYFjM6SgmxDsOpAAAYXMSIoVa3V55Onyrys5SRGCdJSoh1KHV8jPFkAAAMHmJkCMhIjFNmarz1GAAAmOAEVgAAYIoYAQAApogRAABgihgBAACmiBEAAGCKGAEAAKb6FCOVlZVKT09XdHS0cnJyVFtbe17bbdu2TREREVq4cGFfXhYAAIxAIcfI9u3bVVJSorVr16q+vl6zZs1SXl6eWlpazrrdRx99pJ///Oe6+uqr+zwsAAAYeUKOkfLyci1dulSFhYWaMWOGqqqqNGbMGG3ZsqXXbXw+nxYvXqx169bp4osvvqCBAQDAyBJSjHi9XtXV1cnlcn3+DSIj5XK5VFNT0+t2//u//6vExET9+Mc/Pq/X6ejoUHt7e9ADAACMTCHFyPHjx+Xz+ZSUlBS0PCkpSU1NTT1us3//fv3+97/X5s2bz/t1ysrKFB8fH3ikpaWFMiYAABhGBvTTNCdPntStt96qzZs3a9KkSee9XWlpqU6cOBF4HD16dACnBAAAlkK6Ud6kSZMUFRWl5ubmoOXNzc1KTk7utv6//vUvffTRR1qwYEFgWVdX139feNQoHThwQJdcckm37ZxOp5xOZyijAQCAYSqkIyMOh0OzZ89WdXV1YFlXV5eqq6uVm5vbbf1p06bpnXfeUUNDQ+Bx44036tprr1VDQwNvvwAAgNCOjEhSSUmJCgoKlJ2drTlz5qiiokJut1uFhYWSpCVLlig1NVVlZWWKjo5WZmZm0Pbjx4+XpG7LAQBAeAo5RvLz83Xs2DGtWbNGTU1NysrK0q5duwIntR45ckSRkVzYFQAAnJ+QY0SSiouLVVxc3OPX9u7de9Ztn3zyyb68JAAAGKE4hAEAAEwRIwAAwBQxAgAATBEjAADAFDECAABMESMAAMAUMQIAAEwRIwAAwBQxAgAATBEjAADAFDECAABMESMAAMAUMQIAAEwRIwAAwNQo6wHCSWObR61ub+D5oZZThtMAADA0ECODpLHNI9eGffJ0+oKWx4yOUkKsw2gqAADsESODpNXtlafTp4r8LGUkxgWWJ8Q6lDo+xnAyAABsESODLCMxTpmp8dZjAAAwZHACKwAAMEWMAAAAU8QIAAAwRYwAAABTxAgAADBFjAAAAFPECAAAMEWMAAAAU8QIAAAwRYwAAABTxAgAADBFjAAAAFPECAAAMEWMAAAAU8QIAAAwRYwAAABTxAgAADBFjAAAAFPECAAAMEWMAAAAU8QIAAAwRYwAAABTxAgAADBFjAAAAFPECAAAMEWMAAAAU8QIAAAwRYwAAABTxAgAADBFjAAAAFOjrAcYyRrbPGp1eyVJh1pOGU8DAMDQRIwMkMY2j1wb9snT6QssixkdpYRYh+FUAAAMPcTIAGl1e+Xp9KkiP0sZiXGSpIRYh1LHxxhPBgDA0EKMDLCMxDhlpsZbjwEAwJDFCawAAMAUMQIAAEwRIwAAwBQxAgAATBEjAADAFDECAABMESMAAMAUMQIAAEz1KUYqKyuVnp6u6Oho5eTkqLa2ttd1N2/erKuvvloJCQlKSEiQy+U66/oAACC8hBwj27dvV0lJidauXav6+nrNmjVLeXl5amlp6XH9vXv3atGiRdqzZ49qamqUlpam66+/Xo2NjRc8PAAAGP5CjpHy8nItXbpUhYWFmjFjhqqqqjRmzBht2bKlx/Wfeuop/fSnP1VWVpamTZumJ554Ql1dXaqurr7g4QEAwPAXUox4vV7V1dXJ5XJ9/g0iI+VyuVRTU3Ne3+P06dPq7OzUhAkTel2no6ND7e3tQQ8AADAyhRQjx48fl8/nU1JSUtDypKQkNTU1ndf3WLlypVJSUoKC5svKysoUHx8feKSlpYUyJgAAGEYG9dM0Dz74oLZt26bnnntO0dHRva5XWlqqEydOBB5Hjx4dxCkBAMBgGhXKypMmTVJUVJSam5uDljc3Nys5Ofms2/7617/Wgw8+qJdfflmXX375Wdd1Op1yOp2hjAYAAIapkI6MOBwOzZ49O+jk0zMno+bm5va63cMPP6z77rtPu3btUnZ2dt+nBQAAI05IR0YkqaSkRAUFBcrOztacOXNUUVEht9utwsJCSdKSJUuUmpqqsrIySdJDDz2kNWvWaOvWrUpPTw+cWxIXF6e4uLh+/FEAAMBwFHKM5Ofn69ixY1qzZo2ampqUlZWlXbt2BU5qPXLkiCIjPz/gsnHjRnm9Xt10001B32ft2rX65S9/eWHTAwCAYS/kGJGk4uJiFRcX9/i1vXv3Bj3/6KOP+vISAAAgTHBvGgAAYIoYAQAApogRAABgihgBAACmiBEAAGCKGAEAAKaIEQAAYKpP1xlBd41tHrW6vYHnh1pOGU4DAMDwQYz0g8Y2j1wb9snT6QtaHjM6SgmxDqOpAAAYHoiRftDq9srT6VNFfpYyEj+/305CrEOp42MMJwMAYOgjRvpRRmKcMlPjrccAAGBY4QRWAABgihgBAACmiBEAAGCKGAEAAKaIEQAAYIoYAQAApogRAABgihgBAACmiBEAAGCKGAEAAKaIEQAAYIoYAQAApogRAABgihgBAACmiBEAAGCKGAEAAKaIEQAAYIoYAQAApogRAABgihgBAACmRlkPMFw1tnnU6vZKkg61nDKeBgCA4YsY6YPGNo9cG/bJ0+kLLIsZHaWEWIfhVAAADE/ESB+0ur3ydPpUkZ+ljMQ4SVJCrEOp42OMJwMAYPghRi5ARmKcMlPjrccAAGBY4wRWAABgihgBAACmiBEAAGCKGAEAAKaIEQAAYIoYAQAApogRAABgihgBAACmiBEAAGCKGAEAAKaIEQAAYIoYAQAApogRAABgirv2nofGNo9a3d7A80MtpwynAQBgZCFGzqGxzSPXhn3ydPqClseMjlJCrMNoKgAARg5i5Bxa3V55On2qyM9SRmJcYHlCrEOp42MMJwMAYGQgRs5TRmKcMlPjrccAAGDE4QRWAABgihgBAACmiBEAAGCKGAEAAKaIEQAAYIoYAQAApvhobw++eMVVrrYKAMDA6tORkcrKSqWnpys6Olo5OTmqra096/p/+ctfNG3aNEVHR2vmzJnauXNnn4YdDGeuuPrdx/bru4/t1/LtDVxtFQCAARTykZHt27erpKREVVVVysnJUUVFhfLy8nTgwAElJiZ2W/+1117TokWLVFZWpu9+97vaunWrFi5cqPr6emVmZvbLD9GferriKldbBQBg4IR8ZKS8vFxLly5VYWGhZsyYoaqqKo0ZM0Zbtmzpcf1HHnlE3/72t3XXXXdp+vTpuu+++3TllVfq8ccfv+Dh+0Njm0fvNp4IPM68LXPmiquZqfGECAAAAyikIyNer1d1dXUqLS0NLIuMjJTL5VJNTU2P29TU1KikpCRoWV5enp5//vleX6ejo0MdHR2B5ydOnJAktbe3hzLuOX3S5tGNj+/XZ51dQcujR0dqlO8ztbdH9OvrAQAw1Jw62a6ujtM6dbK93//unfm77ff7z7peSDFy/Phx+Xw+JSUlBS1PSkrS+++/3+M2TU1NPa7f1NTU6+uUlZVp3bp13ZanpaWFMu4Fmf6rQXspAADM5VYM3Pc+efKk4uN7v7/bkPw0TWlpadDRlK6uLn366aeaOHGiIiL6r9ra29uVlpamo0ePaty4cf32fYcr9kd37JNg7I9g7I/u2CfBwn1/+P1+nTx5UikpKWddL6QYmTRpkqKiotTc3By0vLm5WcnJyT1uk5ycHNL6kuR0OuV0OoOWjR8/PpRRQzJu3Liw/EfSG/ZHd+yTYOyPYOyP7tgnwcJ5f5ztiMgZIZ3A6nA4NHv2bFVXVweWdXV1qbq6Wrm5uT1uk5ubG7S+JO3evbvX9QEAQHgJ+W2akpISFRQUKDs7W3PmzFFFRYXcbrcKCwslSUuWLFFqaqrKysokScuWLdM111yjDRs2aP78+dq2bZveeustbdq0qX9/EgAAMCyFHCP5+fk6duyY1qxZo6amJmVlZWnXrl2Bk1SPHDmiyMjPD7jMnTtXW7du1b333qu7775bl156qZ5//vkhcY0Rp9OptWvXdntLKFyxP7pjnwRjfwRjf3THPgnG/jg/Ef5zfd4GAABgAHGjPAAAYIoYAQAApogRAABgihgBAACmwjZGKisrlZ6erujoaOXk5Ki2ttZ6JDNlZWX6xje+obFjxyoxMVELFy7UgQMHrMcaMh588EFFRERo+fLl1qOYamxs1A9/+ENNnDhRMTExmjlzpt566y3rsUz4fD6tXr1aU6dOVUxMjC655BLdd99957z/xkjx6quvasGCBUpJSVFERES3e435/X6tWbNGkydPVkxMjFwulz744AObYQfJ2fZJZ2enVq5cqZkzZyo2NlYpKSlasmSJPvnkE7uBh5iwjJHt27erpKREa9euVX19vWbNmqW8vDy1tLRYj2Zi3759Kioq0uuvv67du3ers7NT119/vdxut/Vo5t5880397ne/0+WXX249iqnW1lbNmzdPo0eP1t///nf985//1IYNG5SQkGA9momHHnpIGzdu1OOPP6733ntPDz30kB5++GE99thj1qMNCrfbrVmzZqmysrLHrz/88MN69NFHVVVVpTfeeEOxsbHKy8vTZ599NsiTDp6z7ZPTp0+rvr5eq1evVn19vZ599lkdOHBAN954o8GkQ5Q/DM2ZM8dfVFQUeO7z+fwpKSn+srIyw6mGjpaWFr8k/759+6xHMXXy5En/pZde6t+9e7f/mmuu8S9btsx6JDMrV670X3XVVdZjDBnz58/333bbbUHLvve97/kXL15sNJEdSf7nnnsu8Lyrq8ufnJzs/9WvfhVY1tbW5nc6nf4///nPBhMOvi/vk57U1tb6Jfk//vjjwRlqiAu7IyNer1d1dXVyuVyBZZGRkXK5XKqpqTGcbOg4ceKEJGnChAnGk9gqKirS/Pnzg/6thKsXXnhB2dnZ+v73v6/ExERdccUV2rx5s/VYZubOnavq6modPHhQkvSPf/xD+/fv1w033GA8mb3Dhw+rqakp6P9NfHy8cnJy+B37BSdOnFBERMSA3ndtOBmSd+0dSMePH5fP5wtcMfaMpKQkvf/++0ZTDR1dXV1avny55s2bNySukmtl27Ztqq+v15tvvmk9ypDw4YcfauPGjSopKdHdd9+tN998Uz/72c/kcDhUUFBgPd6gW7Vqldrb2zVt2jRFRUXJ5/PpgQce0OLFi61HM9fU1CRJPf6OPfO1cPfZZ59p5cqVWrRoUdjePO/Lwi5GcHZFRUV69913tX//futRzBw9elTLli3T7t27FR0dbT3OkNDV1aXs7GytX79eknTFFVfo3XffVVVVVVjGyNNPP62nnnpKW7du1WWXXaaGhgYtX75cKSkpYbk/cP46Ozt18803y+/3a+PGjdbjDBlh9zbNpEmTFBUVpebm5qDlzc3NSk5ONppqaCguLtaLL76oPXv26KKLLrIex0xdXZ1aWlp05ZVXatSoURo1apT27dunRx99VKNGjZLP57MecdBNnjxZM2bMCFo2ffp0HTlyxGgiW3fddZdWrVqlH/zgB5o5c6ZuvfVWrVixInCD0HB25vcov2O7OxMiH3/8sXbv3s1RkS8IuxhxOByaPXu2qqurA8u6urpUXV2t3Nxcw8ns+P1+FRcX67nnntMrr7yiqVOnWo9k6rrrrtM777yjhoaGwCM7O1uLFy9WQ0ODoqKirEccdPPmzev2ce+DBw/qq1/9qtFEtk6fPh10Q1BJioqKUldXl9FEQ8fUqVOVnJwc9Du2vb1db7zxRtj+jpU+D5EPPvhAL7/8siZOnGg90pASlm/TlJSUqKCgQNnZ2ZozZ44qKirkdrtVWFhoPZqJoqIibd26VX/96181duzYwPu68fHxiomJMZ5u8I0dO7bb+TKxsbGaOHFi2J5Hs2LFCs2dO1fr16/XzTffrNraWm3atEmbNm2yHs3EggUL9MADD2jKlCm67LLL9Pbbb6u8vFy33Xab9WiD4tSpUzp06FDg+eHDh9XQ0KAJEyZoypQpWr58ue6//35deumlmjp1qlavXq2UlBQtXLjQbugBdrZ9MnnyZN10002qr6/Xiy++KJ/PF/g9O2HCBDkcDquxhw7rj/NYeeyxx/xTpkzxOxwO/5w5c/yvv/669UhmJPX4+MMf/mA92pAR7h/t9fv9/r/97W/+zMxMv9Pp9E+bNs2/adMm65HMtLe3+5ctW+afMmWKPzo62n/xxRf777nnHn9HR4f1aINiz549Pf7OKCgo8Pv9//147+rVq/1JSUl+p9Ppv+666/wHDhywHXqAnW2fHD58uNffs3v27LEefUiI8PvD5JKBAABgSAq7c0YAAMDQQowAAABTxAgAADBFjAAAAFPECAAAMEWMAAAAU8QIAAAwRYwAAABTxAgAADBFjAAAAFPECAAAMEWMAAAAU/8fZ57Lw5XtyaQAAAAASUVORK5CYII=\n",
- "text/plain": [
- "
"
- ],
- "text/plain": [
- ""
- ]
- },
- "metadata": {},
- "output_type": "display_data"
- }
- ],
- "source": [
- "scenarios = { \"1\": (10,100), \n",
- " \"2\": (100,1000),\n",
- " \"2.1\": (200,2000),\n",
- " \"2.2\": (300,3000),\n",
- " \"2.3\": (400,4000),\n",
- " \"2.4\": (500,5000),\n",
- " \"3\": (1000,10000),\n",
- " \"4\": (10000,100000)\n",
- " }\n",
- "_=compare_significance(scenarios)"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "## Generalize\n",
- "\n",
- "This function above isn't general, in the sense that it relies on various objects we computed before it (namely TPR and FPR). Lets generalize to get a function that can make a similar plot for any observeable. \n",
- "\n",
- "One problem is that we are using `matplotlib` to compute the TPR/FPR, eventhough we never plot it. Let's instead compute it with `numpy`:\n"
- ]
- },
- {
- "cell_type": "code",
- "execution_count": 29,
- "metadata": {},
- "outputs": [
- {
- "data": {
- "image/png": "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\n",
- "text/plain": [
- "