{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"execution": {
"iopub.execute_input": "2024-04-17T07:33:19.985737Z",
"iopub.status.busy": "2024-04-17T07:33:19.985331Z",
"iopub.status.idle": "2024-04-17T07:33:20.300691Z",
"shell.execute_reply": "2024-04-17T07:33:20.300454Z"
},
"pycharm": {
"name": "#%%\n"
}
},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" \n",
" "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"import numpy as np\n",
"import random\n",
"from lets_plot import *\n",
"\n",
"LetsPlot.setup_html()"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"execution": {
"iopub.execute_input": "2024-04-17T07:33:20.313779Z",
"iopub.status.busy": "2024-04-17T07:33:20.313635Z",
"iopub.status.idle": "2024-04-17T07:33:20.315674Z",
"shell.execute_reply": "2024-04-17T07:33:20.315488Z"
}
},
"outputs": [],
"source": [
"# This example was found at: www.cookbook-r.com/Graphs/Scatterplots_(ggplot2)\n",
"random.seed(123)\n",
"data = dict(\n",
" cond=np.repeat(['A','B'], 10),\n",
" xvar=[i + random.normalvariate(0, 3) for i in range(0,20)],\n",
" yvar=[i + random.normalvariate(0, 3) for i in range(0,20)]\n",
")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Basic scatter-plot"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"execution": {
"iopub.execute_input": "2024-04-17T07:33:20.316776Z",
"iopub.status.busy": "2024-04-17T07:33:20.316604Z",
"iopub.status.idle": "2024-04-17T07:33:20.348245Z",
"shell.execute_reply": "2024-04-17T07:33:20.348044Z"
}
},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"p = ggplot(data, aes(x='xvar', y='yvar')) + ggsize(300, 250)\n",
"p + geom_point(shape=1) "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Add regression line"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"execution": {
"iopub.execute_input": "2024-04-17T07:33:20.349898Z",
"iopub.status.busy": "2024-04-17T07:33:20.349787Z",
"iopub.status.idle": "2024-04-17T07:33:20.352932Z",
"shell.execute_reply": "2024-04-17T07:33:20.352691Z"
}
},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"p + geom_point(shape=1) + geom_smooth() "
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"execution": {
"iopub.execute_input": "2024-04-17T07:33:20.354207Z",
"iopub.status.busy": "2024-04-17T07:33:20.354052Z",
"iopub.status.idle": "2024-04-17T07:33:20.356862Z",
"shell.execute_reply": "2024-04-17T07:33:20.356661Z"
}
},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Without standard error band.\n",
"p + geom_point(shape=1) + geom_smooth(se=False) "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Split dataset by the `cond` variable"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"execution": {
"iopub.execute_input": "2024-04-17T07:33:20.357930Z",
"iopub.status.busy": "2024-04-17T07:33:20.357772Z",
"iopub.status.idle": "2024-04-17T07:33:20.361344Z",
"shell.execute_reply": "2024-04-17T07:33:20.361126Z"
}
},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"p1 = ggplot(data, aes(x='xvar', y='yvar', color='cond')) + ggsize(500, 250)\n",
"p1 + geom_point(shape=1) + geom_smooth(se=False)"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"execution": {
"iopub.execute_input": "2024-04-17T07:33:20.362249Z",
"iopub.status.busy": "2024-04-17T07:33:20.362171Z",
"iopub.status.idle": "2024-04-17T07:33:20.364694Z",
"shell.execute_reply": "2024-04-17T07:33:20.364517Z"
}
},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Map `shape` to the `cond` variable.\n",
"p2 = p1 + geom_point(aes(shape='cond'), size=5)\n",
"p2"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"execution": {
"iopub.execute_input": "2024-04-17T07:33:20.365499Z",
"iopub.status.busy": "2024-04-17T07:33:20.365423Z",
"iopub.status.idle": "2024-04-17T07:33:20.367710Z",
"shell.execute_reply": "2024-04-17T07:33:20.367541Z"
},
"scrolled": true
},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Choose different shapes using `scale_shape_manual`:\n",
"# 1 - hollow circle \n",
"# 2 - hollow triangle\n",
"\n",
"p2 + scale_shape_manual(values=[1,2])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### Handling overplotting"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"execution": {
"iopub.execute_input": "2024-04-17T07:33:20.368535Z",
"iopub.status.busy": "2024-04-17T07:33:20.368463Z",
"iopub.status.idle": "2024-04-17T07:33:20.369987Z",
"shell.execute_reply": "2024-04-17T07:33:20.369810Z"
}
},
"outputs": [],
"source": [
"# Create data containing overlapping points.\n",
"data['xrnd'] = [round(v / 5) * 5 for v in data['xvar']]\n",
"data['yrnd'] = [round(v / 5) * 5 for v in data['yvar']]"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"execution": {
"iopub.execute_input": "2024-04-17T07:33:20.370761Z",
"iopub.status.busy": "2024-04-17T07:33:20.370685Z",
"iopub.status.idle": "2024-04-17T07:33:20.373156Z",
"shell.execute_reply": "2024-04-17T07:33:20.372982Z"
}
},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"p3 = ggplot(data, aes(x='xrnd', y='yrnd')) + ggsize(500, 250)\n",
"\n",
"# Use `alpha` to show overplotting.\n",
"p3 + geom_point(alpha=.3, size=5)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"execution": {
"iopub.execute_input": "2024-04-17T07:33:20.373957Z",
"iopub.status.busy": "2024-04-17T07:33:20.373885Z",
"iopub.status.idle": "2024-04-17T07:33:20.376198Z",
"shell.execute_reply": "2024-04-17T07:33:20.376028Z"
}
},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# `jitter` points to show overplotting in another way.\n",
"p3 + geom_point(shape=1, position=position_jitter(width=.2,height=.5))"
]
}
],
"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": "3.10.13"
}
},
"nbformat": 4,
"nbformat_minor": 4
}