{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Random Sampling"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"execution": {
"iopub.execute_input": "2024-03-26T14:35:59.058093Z",
"iopub.status.busy": "2024-03-26T14:35:59.058003Z",
"iopub.status.idle": "2024-03-26T14:35:59.380841Z",
"shell.execute_reply": "2024-03-26T14:35:59.380389Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" \n",
" "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"import numpy as np\n",
"from lets_plot import *\n",
"\n",
"LetsPlot.setup_html()"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"execution": {
"iopub.execute_input": "2024-03-26T14:35:59.393131Z",
"iopub.status.busy": "2024-03-26T14:35:59.393010Z",
"iopub.status.idle": "2024-03-26T14:35:59.395079Z",
"shell.execute_reply": "2024-03-26T14:35:59.394890Z"
}
},
"outputs": [],
"source": [
"def data(n):\n",
" np.random.seed(123)\n",
" cov=[[1, -.8], \n",
" [-.8, 1]]\n",
" x, y = np.random.multivariate_normal(mean=[0,0], cov=cov, size=n).T\n",
" return dict(x=x, y=y)"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"execution": {
"iopub.execute_input": "2024-03-26T14:35:59.396197Z",
"iopub.status.busy": "2024-03-26T14:35:59.396033Z",
"iopub.status.idle": "2024-03-26T14:35:59.760205Z",
"shell.execute_reply": "2024-03-26T14:35:59.759693Z"
}
},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# N exceeds default point sampling threshold (50,000)\n",
"N=75000\n",
"ggplot(data(N), aes('x', 'y')) + geom_point()"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"execution": {
"iopub.execute_input": "2024-03-26T14:35:59.772080Z",
"iopub.status.busy": "2024-03-26T14:35:59.771946Z",
"iopub.status.idle": "2024-03-26T14:35:59.902374Z",
"shell.execute_reply": "2024-03-26T14:35:59.902164Z"
}
},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Set 'random' sample size 500\n",
"ggplot(data(N), aes('x', 'y')) + geom_point(sampling=sampling_random(500))"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"execution": {
"iopub.execute_input": "2024-03-26T14:35:59.903695Z",
"iopub.status.busy": "2024-03-26T14:35:59.903584Z",
"iopub.status.idle": "2024-03-26T14:36:00.238478Z",
"shell.execute_reply": "2024-03-26T14:36:00.238249Z"
}
},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Use wrong sampling ('group_random' will do nothing in this context)\n",
"# It will override the default 'random' sampling and effectively disable proper sampling in this plot.\n",
"# However, if N will grow above 'safety' threshold, 'safety' sampling will be applied (see next plot). \n",
"ggplot(data(N), aes('x', 'y')) + geom_point(sampling=sampling_group_random(1))"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"execution": {
"iopub.execute_input": "2024-03-26T14:36:00.248311Z",
"iopub.status.busy": "2024-03-26T14:36:00.248198Z",
"iopub.status.idle": "2024-03-26T14:36:01.189365Z",
"shell.execute_reply": "2024-03-26T14:36:01.189106Z"
}
},
"outputs": [
{
"data": {
"text/html": [
" \n",
"