{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This page is available as an executable or viewable Jupyter Notebook:\n",
"
\n",
"\n",
" ![](\"https://mybinder.org/badge_logo.svg\")
\n",
"\n",
"\n",
" ![](\"https://raw.githubusercontent.com/jupyter/design/master/logos/Badges/nbviewer_badge.png\"\n",)
\n",
"\n",
"
\n",
"
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Stratified sampling\n",
"\n",
"In large dataset a relatively small group of points might be overplotted by the dominant group. In this case __stratified sampling__ can help."
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"%useLatestDescriptors\n",
"%use lets-plot"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"import java.util.Random\n",
"\n",
"val N = 5000 \n",
"val smallGroup = 3\n",
"val largeGroup = N - smallGroup\n",
"\n",
"val rand = Random(123)\n",
"val data = mapOf (\n",
" \"x\" to List(N) { rand.nextGaussian() },\n",
" \"y\" to List(N) { rand.nextGaussian() },\n",
" \"cond\" to List(smallGroup) { 'A' } + List(largeGroup) { 'B' }\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
]
},
"execution_count": 3,
"metadata": {
"new_classpath": []
},
"output_type": "execute_result"
}
],
"source": [
"// Data points in group 'A' (small group) are overplotted by the dominant group 'B'.\n",
"val p = ggplot(data) { x=\"x\"; y=\"y\"; color=\"cond\" } + \n",
" scale_color_manual(values=listOf(\"red\", \"#1C9E77\"), breaks=listOf('A', 'B'))\n",
"p + geom_point(size=5, alpha=.2)"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
]
},
"execution_count": 4,
"metadata": {
"new_classpath": []
},
"output_type": "execute_result"
}
],
"source": [
"// The 'random' sampling loses the group 'A' altogether.\n",
"p + geom_point(size=5, sampling=sampling_random(50, seed=2))"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
]
},
"execution_count": 5,
"metadata": {
"new_classpath": []
},
"output_type": "execute_result"
}
],
"source": [
"// Stratified sampling ensures that group 'A' is represented.\n",
"p + geom_point(size=5, sampling=sampling_random_stratified(50, seed=2))"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Kotlin",
"language": "kotlin",
"name": "kotlin"
},
"language_info": {
"codemirror_mode": "text/x-kotlin",
"file_extension": ".kt",
"mimetype": "text/x-kotlin",
"name": "kotlin",
"pygments_lexer": "kotlin",
"version": "1.4.30-dev-2223"
}
},
"nbformat": 4,
"nbformat_minor": 4
}