Skip to content

Commit 1abb5c0

Browse files
nrposnernrposner
committed
adding readme
updating readme updating title in README updating notebook filled out README noting time of original talk
1 parent 021fb8f commit 1abb5c0

File tree

3 files changed

+756
-0
lines changed

3 files changed

+756
-0
lines changed

‎.ipynb_checkpoints/benchmark-checkpoint.ipynb‎

Lines changed: 219 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,219 @@
1+
{
2+
"cells": [
3+
{
4+
"cell_type": "code",
5+
"execution_count": 1,
6+
"id": "f03c3997-70f0-4196-b909-d84162d8c2c7",
7+
"metadata": {},
8+
"outputs": [],
9+
"source": [
10+
"import coding_club as cc\n",
11+
"\n",
12+
"import numpy as np\n",
13+
"import time\n",
14+
"import matplotlib.pyplot as plt"
15+
]
16+
},
17+
{
18+
"cell_type": "code",
19+
"execution_count": 2,
20+
"id": "220a9b52-0091-4b07-b926-5d8311844e5d",
21+
"metadata": {},
22+
"outputs": [],
23+
"source": [
24+
"def create_random_matrix(n: int):\n",
25+
" \"\"\"\n",
26+
" Initializes an n x n matrix with random numbers from 0 to 1.\n",
27+
"\n",
28+
" Args:\n",
29+
" n: The dimension of the square matrix (e.g., for an n x n matrix).\n",
30+
"\n",
31+
" Returns:\n",
32+
" A numpy array representing the n x n matrix with random numbers.\n",
33+
" \"\"\"\n",
34+
" matrix = np.random.rand(n, n)\n",
35+
" return matrix\n",
36+
"\n",
37+
"def manual_matrix_multiply(A: list[list[float]], B: list[list[float]]) -> list[list[float]]:\n",
38+
" \"\"\"\n",
39+
" Multiplies two matrices A and B using pure Python.\n",
40+
" Assumes inputs are well-formed (rectangular) lists of lists.\n",
41+
" \"\"\"\n",
42+
" # Get dimensions\n",
43+
" rows_A = len(A)\n",
44+
" cols_A = len(A[0])\n",
45+
" rows_B = len(B)\n",
46+
" cols_B = len(B[0])\n",
47+
"\n",
48+
" # Check if multiplication is possible\n",
49+
" if cols_A != rows_B:\n",
50+
" raise ValueError(f\"Cannot multiply matrices of shapes ({rows_A}, {cols_A}) and ({rows_B}, {cols_B}). Inner dimensions must match.\")\n",
51+
"\n",
52+
" # Initialize the result matrix C with zeros.\n",
53+
" # Dimensions of C will be (rows_A, cols_B)\n",
54+
" C = [[0 for _ in range(cols_B)] for _ in range(rows_A)]\n",
55+
"\n",
56+
" # Perform the multiplication\n",
57+
" # Iterate through each row of A\n",
58+
" for i in range(rows_A):\n",
59+
" # Iterate through each column of B\n",
60+
" for j in range(cols_B):\n",
61+
" # Calculate the dot product of A's row i and B's column j\n",
62+
" dot_product = 0\n",
63+
" for k in range(cols_A): # or range(rows_B)\n",
64+
" dot_product += A[i][k] * B[k][j]\n",
65+
" C[i][j] = dot_product\n",
66+
" \n",
67+
" return C\n",
68+
"\n",
69+
"def matrix_power(matrix: np.ndarray[np.ndarray], exponent: int, nrows: int):\n",
70+
" \"\"\"\n",
71+
" Multiplies two matrices using the manual Python matrix multiplication function\n",
72+
" \"\"\"\n",
73+
"\n",
74+
" result = np.eye(nrows, dtype = \"f8\")\n",
75+
" base = matrix\n",
76+
"\n",
77+
" while exponent > 0:\n",
78+
" if exponent % 2 == 1:\n",
79+
" result = manual_matrix_multiply(result, base)\n",
80+
" base = manual_matrix_multiply(base, base)\n",
81+
" exponent //= 2\n",
82+
" return result\n",
83+
"\n",
84+
"def np_matrix_power(matrix: np.ndarray[np.ndarray], exponent: int, nrows: int):\n",
85+
" \"\"\"\n",
86+
" Multiplies two matrices using the numpy.dot matrix multiplication function\n",
87+
" \"\"\"\n",
88+
"\n",
89+
" result = np.eye(nrows, dtype = \"f8\")\n",
90+
" base = matrix\n",
91+
"\n",
92+
" while exponent > 0:\n",
93+
" if exponent % 2 == 1:\n",
94+
" result = np.dot(result, base)\n",
95+
" base = np.dot(base, base)\n",
96+
" exponent //= 2\n",
97+
" return result"
98+
]
99+
},
100+
{
101+
"cell_type": "code",
102+
"execution_count": null,
103+
"id": "5019de12-010f-4d59-9b69-479a485d6b3a",
104+
"metadata": {},
105+
"outputs": [],
106+
"source": [
107+
"%%time\n",
108+
"\n",
109+
"python_times = np.array([])\n",
110+
"\n",
111+
"numpy_times = np.array([])\n",
112+
"\n",
113+
"py_numpy_times = np.array([])\n",
114+
"\n",
115+
"rust_python_times = np.array([])\n",
116+
"\n",
117+
"size = [5, 7, 10, 20, 50, 70, 100, 200, 500, 700]\n",
118+
"\n",
119+
"iterations = 10\n",
120+
"\n",
121+
"for i in size:\n",
122+
" matrix = create_random_matrix(i)\n",
123+
"\n",
124+
" s = time.process_time(); py_result = matrix_power(matrix, iterations, i); e = time.process_time();\n",
125+
" python_times = np.append(python_times, e-s)\n",
126+
"\n",
127+
"\n",
128+
" s = time.process_time(); numpy_result = np.linalg.matrix_power(matrix, iterations); e = time.process_time();\n",
129+
" numpy_times = np.append(numpy_times, e-s)\n",
130+
"\n",
131+
"\n",
132+
" s = time.process_time(); rust_result = cc.matrix_power(matrix, iterations); e = time.process_time();\n",
133+
" rust_python_times = np.append(rust_python_times, e-s)\n",
134+
"\n",
135+
" if np.all(np.isclose(py_result, numpy_result)) & np.all(np.isclose(numpy_result, rust_result)) & np.all(np.isclose(py_result, rust_result)):\n",
136+
" continue\n",
137+
" else: \n",
138+
" print(\"Results are inconsistent\")\n",
139+
" break\n"
140+
]
141+
},
142+
{
143+
"cell_type": "code",
144+
"execution_count": null,
145+
"id": "2123601b-fed4-418c-8107-c8b98db516be",
146+
"metadata": {},
147+
"outputs": [],
148+
"source": [
149+
"plt.plot(size, python_times, label = \"Python\")\n",
150+
"plt.plot(size, numpy_times, label = \"Numpy\")\n",
151+
"plt.plot(size, rust_python_times, label = \"Rust-Python\")\n",
152+
"plt.legend()\n",
153+
"plt.title(\"Runtime Performance on Matrix Benchmark\")\n",
154+
"plt.ylabel(\"Runtime (seconds)\")\n",
155+
"plt.xlabel(\"Matrix size\")"
156+
]
157+
},
158+
{
159+
"cell_type": "code",
160+
"execution_count": null,
161+
"id": "314bd286-0c80-45e4-9954-6c91fc93d578",
162+
"metadata": {},
163+
"outputs": [],
164+
"source": [
165+
"plt.plot(size, python_times, label = \"Python\")\n",
166+
"plt.plot(size, numpy_times, label = \"Numpy\")\n",
167+
"plt.plot(size, rust_python_times, label = \"Rust-Python\")\n",
168+
"plt.xscale(\"log\")\n",
169+
"plt.yscale(\"log\")\n",
170+
"plt.legend()\n",
171+
"plt.title(\"Runtime Performance on Matrix Benchmark (Log-Log)\")\n",
172+
"plt.ylabel(\"Runtime (Log(seconds))\")\n",
173+
"plt.xlabel(\"Log(Matrix size)\")\n",
174+
"plt.savefig('benchmark.png')"
175+
]
176+
},
177+
{
178+
"cell_type": "code",
179+
"execution_count": null,
180+
"id": "68a5392b-6bd0-4aa6-94db-763579611feb",
181+
"metadata": {},
182+
"outputs": [],
183+
"source": [
184+
"rust_python_times/numpy_times"
185+
]
186+
},
187+
{
188+
"cell_type": "code",
189+
"execution_count": null,
190+
"id": "c3c9b488-ff19-4d78-8606-2d91a4ab3c98",
191+
"metadata": {},
192+
"outputs": [],
193+
"source": [
194+
"python_times/rust_python_times"
195+
]
196+
}
197+
],
198+
"metadata": {
199+
"kernelspec": {
200+
"display_name": "Python 3 (ipykernel)",
201+
"language": "python",
202+
"name": "python3"
203+
},
204+
"language_info": {
205+
"codemirror_mode": {
206+
"name": "ipython",
207+
"version": 3
208+
},
209+
"file_extension": ".py",
210+
"mimetype": "text/x-python",
211+
"name": "python",
212+
"nbconvert_exporter": "python",
213+
"pygments_lexer": "ipython3",
214+
"version": "3.13.4"
215+
}
216+
},
217+
"nbformat": 4,
218+
"nbformat_minor": 5
219+
}

0 commit comments

Comments
 (0)