2025-plovdiv-comp-arch/exercise.ipynb
2025-09-23 15:05:34 +03:00

244 lines
6 KiB
Text

{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import numpy as np"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"n_series = 32\n",
"len_one_series = 5*2**20\n",
"time_series = np.random.rand(n_series, len_one_series)\n",
"gap = 16*2**10"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Size of one time series: 40 M\n",
"Size of collection: 1280 M\n",
"Gap size: 128 K\n",
"Gapped series size: 2 K\n"
]
}
],
"source": [
"print(f'Size of one time series: {int(time_series[0].nbytes/2**20)} M')\n",
"print(f'Size of collection: {int(time_series.nbytes/2**20)} M')\n",
"print(f'Gap size: {int(gap*8/2**10)} K')\n",
"print(f'Gapped series size: {int(time_series[0, ::gap].nbytes/2**10)} K')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The following function implements an approximation of a power series of every `gap` value in our time series.\n",
"\n",
"If we define one time series of length `N` to be:\n",
"\n",
"$[x_0, x_1, x_2, ..., x_N]$,\n",
"\n",
"then the \"gapped\" series with `gap=g` is:\n",
"\n",
"$[x_0, x_g, x_{2g}, ..., x_{N/g}]$,\n",
"\n",
"where $N/g$ is the number of gaps.\n",
"\n",
"The approximation of the power series up to power `30` for our \"gapped\" series is defined as:\n",
"\n",
"$$\\mathbf{P} = \\sum_{p=0}^{30} \\sum_i x_i^{p} = \\sum_i x_i^0 + \\sum_i x_i^1 + \\sum_i x_i^2 + ... + \\sum_i x_i^{30} $$\n",
"\n",
"where $i \\in [0, g, 2g, ..., N/g]$"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"# compute an approximation of a power series for a collection of gapped timeseries\n",
"def power(time_series, P, gap):\n",
" for row in range(time_series.shape[0]):\n",
" for pwr in range(30):\n",
" P[row] += (time_series[row, ::gap]**pwr).sum()\n",
" return P\n",
" "
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
"P = np.zeros_like(time_series)"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"5.53 s ± 1.3 s per loop (mean ± std. dev. of 7 runs, 1 loop each)\n"
]
}
],
"source": [
"%timeit power(time_series, P, gap)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Challenge\n",
"- Can you improve on the above implementation of the `power` function?\n",
"- Change the following `power_improved` function and see what you can do\n",
"- **Remember**: you can't change any other cell in this notebook!"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"execution": {
"iopub.execute_input": "2024-03-04T10:06:08.461Z",
"iopub.status.busy": "2024-03-04T10:06:08.459Z",
"iopub.status.idle": "2024-03-04T10:06:08.466Z",
"shell.execute_reply": "2024-03-04T10:06:08.468Z"
}
},
"outputs": [],
"source": [
"def power_improved(time_series, P, gap):\n",
" for row in range(time_series.shape[0]):\n",
" for pwr in range(30):\n",
" P[row] += (time_series[row, ::gap]**pwr).sum()\n",
" return P"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
"def power_improved(time_series, P, gap):\n",
" for pwr in range(30):\n",
" for row_idx, row_values in enumerate(time_series):\n",
" P[row_idx] += (row_values[::gap]**pwr).sum()\n",
" return P"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [],
"source": [
"# verify that they yield the same results\n",
"P = np.zeros(n_series, dtype='float64')\n",
"out1 = power(time_series, P, gap)\n",
"P = np.zeros(n_series, dtype='float64')\n",
"out2 = power_improved(time_series, P, gap)\n",
"np.testing.assert_allclose(out1, out2)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"execution": {
"iopub.execute_input": "2024-03-04T10:06:14.959Z",
"iopub.status.busy": "2024-03-04T10:06:14.956Z",
"iopub.status.idle": "2024-03-04T10:06:17.437Z",
"shell.execute_reply": "2024-03-04T10:06:17.443Z"
}
},
"outputs": [],
"source": [
"P = np.zeros(n_series, dtype='float64')\n",
"%timeit power(time_series, P, gap)"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"17.8 ms ± 1.56 ms per loop (mean ± std. dev. of 7 runs, 100 loops each)\n"
]
}
],
"source": [
"P = np.zeros(n_series, dtype='float64')\n",
"%timeit power_improved(time_series, P, gap)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Learnings:\n",
"- just changing the order of the loops: 5.5 s -> 24.2 ms\n",
"- also enumerating and accessing the rows: 17.8 ms"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"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": "3.13.6"
},
"nteract": {
"version": "0.28.0"
}
},
"nbformat": 4,
"nbformat_minor": 4
}