{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"id": "88b6cfbb",
"metadata": {},
"outputs": [],
"source": [
"import numpy as np\n",
"import pandas as pd\n",
"import random\n",
"from lets_plot import *\n",
"from lets_plot.mapping import as_discrete"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "a7fd8774",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
" \n",
" \n",
" "
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"def dump_plot(plot, display=None):\n",
" import json\n",
"\n",
" try:\n",
" import clipboard\n",
" except:\n",
" clipboard = None\n",
" \n",
" from lets_plot._type_utils import standardize_dict\n",
" \n",
" plot_dict = standardize_dict(plot.as_dict())\n",
" plot_json = json.dumps(plot_dict, indent=2)\n",
" \n",
" if clipboard:\n",
" clipboard.copy('')\n",
" clipboard.copy(str(plot_json))\n",
" else:\n",
" if display is None:\n",
" display = True\n",
" \n",
" if display:\n",
" print(plot_json)\n",
"\n",
" return plot\n",
"\n",
"LetsPlot.setup_html()"
]
},
{
"cell_type": "markdown",
"id": "f51ee2cf",
"metadata": {},
"source": [
"## Regression Testing\n",
"Regression testing of geometries that may be affected by current changes."
]
},
{
"cell_type": "markdown",
"id": "eeb35e19",
"metadata": {},
"source": [
"### Scatter Plots"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "2297d2f8",
"metadata": {},
"outputs": [],
"source": [
"np.random.seed(42)\n",
"random.seed(42)\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": "code",
"execution_count": 4,
"id": "bb8a7928",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gggrid([\n",
" ggplot(data, aes(x='xvar', y='yvar')) + geom_point(shape=1),\n",
" ggplot(data, aes(x='yvar', y='xvar')) + geom_point(shape=1),\n",
" ggplot(data, aes(x='xvar', y='yvar')) + geom_point(shape=1) + geom_smooth(),\n",
" ggplot(data, aes(x='yvar', y='xvar')) + geom_point(shape=1) + geom_smooth()\n",
"])"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "1d892d2d",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gggrid([\n",
" ggplot(data, aes(x='xvar', y='yvar', color='cond')) + geom_point(shape=1) + geom_smooth(se=False),\n",
" ggplot(data, aes(x='yvar', y='xvar', color='cond')) + geom_point(shape=1) + geom_smooth(se=False)\n",
"])"
]
},
{
"cell_type": "markdown",
"id": "7531ddf4",
"metadata": {},
"source": [
"### Marginal Plots"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "3247f51a",
"metadata": {},
"outputs": [],
"source": [
"np.random.seed(42)\n",
"cov0=[[1, -.8], [-.8, 1]] \n",
"cov1=[[ 10, .1], [.1, .1]]\n",
"\n",
"x0, y0 = np.random.multivariate_normal(mean=[-2,0], cov=cov0, size=200).T\n",
"x1, y1 = np.random.multivariate_normal(mean=[0,1], cov=cov1, size=200).T\n",
"\n",
"data = dict(\n",
" x = np.concatenate((x0,x1)),\n",
" y = np.concatenate((y0,y1)),\n",
" c = [\"A\"]*200 + [\"B\"]*200\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "4cbc63a4",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gggrid([\n",
" ggplot(data, aes(\"x\", \"y\")) + geom_density2df(aes(fill=\"..level..\")) + coord_cartesian() \\\n",
" + ggmarginal(\"tr\", layer=geom_area(stat=\"density\")),\n",
" ggplot(data, aes(\"y\", \"x\")) + geom_density2df(aes(fill=\"..level..\")) + coord_cartesian() \\\n",
" + ggmarginal(\"tr\", layer=geom_area(stat=\"density\"))\n",
"])"
]
},
{
"cell_type": "markdown",
"id": "15bd6c1d",
"metadata": {},
"source": [
"### `coord_flip()`"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "c52de749",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"np.random.seed(42)\n",
"n = 10\n",
"x = np.arange(n)\n",
"y = 1 + np.random.randint(5, size=10)\n",
"gggrid([\n",
" ggplot() + geom_bar(aes(x='x', y='y'), data={'x': x, 'y': y}, stat='identity'),\n",
" ggplot() + geom_bar(aes(x='x', y='y'), data={'x': x, 'y': y}, stat='identity') + coord_flip()\n",
"])"
]
},
{
"cell_type": "markdown",
"id": "55bb8824",
"metadata": {},
"source": [
"### `geom_bar()`"
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "959a5ec0",
"metadata": {},
"outputs": [],
"source": [
"data = {\n",
" 'code': ['a','b','c','d','e'],\n",
" 'value': [2, 5, 3, 8, -1],\n",
"}"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "1a4077a0",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gggrid([\n",
" ggplot(data, aes('code', 'value')) + geom_bar(stat='identity'),\n",
" ggplot(data, aes('value', 'code')) + geom_bar(orientation='y', stat='identity'),\n",
" ggplot(data, aes('value', 'code')) + geom_bar(stat='identity'),\n",
" ggplot(data) + geom_bar(aes('value', 'code'), stat='identity'),\n",
" ggplot(data, aes('value', as_discrete('code'))) + geom_bar(stat='identity')\n",
"])"
]
},
{
"cell_type": "markdown",
"id": "e737e168",
"metadata": {},
"source": [
"### `geom_lollipop()`"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "27967418",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gggrid([\n",
" ggplot(data, aes('code', 'value')) + geom_lollipop(stat='identity'),\n",
" ggplot(data, aes('value', 'code')) + geom_lollipop(orientation='y', stat='identity'),\n",
" ggplot(data, aes('value', 'code')) + geom_lollipop(stat='identity'),\n",
" ggplot(data) + geom_lollipop(aes('value', 'code'), stat='identity'),\n",
" ggplot(data, aes('value', as_discrete('code'))) + geom_lollipop(stat='identity') \n",
"])"
]
},
{
"cell_type": "markdown",
"id": "e3b7b6b7",
"metadata": {},
"source": [
"### `geom_boxplot()`"
]
},
{
"cell_type": "code",
"execution_count": 12,
"id": "8e934911",
"metadata": {},
"outputs": [],
"source": [
"n = 100\n",
"np.random.seed(42)\n",
"data = {\n",
" 'code': np.random.choice(list('abcde'), size=100),\n",
" 'value': np.random.normal(size=100),\n",
" 'value_str': [str(i) for i in np.random.normal(size=100)],\n",
"}"
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "df453671",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gggrid([\n",
" ggplot(data, aes('code', 'value')) + geom_boxplot(),\n",
" ggplot(data, aes('value', 'code')) + geom_boxplot(orientation='y'),\n",
" ggplot(data, aes('value', 'code')) + geom_boxplot(),\n",
" ggplot(data) + geom_boxplot(aes('value', 'code')),\n",
" ggplot(data, aes('value', as_discrete('code'))) + geom_boxplot() \n",
"])"
]
},
{
"cell_type": "markdown",
"id": "0dacca7a",
"metadata": {},
"source": [
"### `geom_violin()`"
]
},
{
"cell_type": "code",
"execution_count": 14,
"id": "7db1d033",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gggrid([\n",
" ggplot(data, aes('code', 'value')) + geom_violin(),\n",
" ggplot(data, aes('value', 'code')) + geom_violin(orientation='y'),\n",
" ggplot(data, aes('value', 'code')) + geom_violin(),\n",
" ggplot(data) + geom_violin(aes('value', 'code')),\n",
" ggplot(data, aes('value', as_discrete('code'))) + geom_violin() \n",
"])"
]
},
{
"cell_type": "markdown",
"id": "e1d07a13",
"metadata": {},
"source": [
"### `geom_ydotplot()`"
]
},
{
"cell_type": "code",
"execution_count": 15,
"id": "5a05e47f",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"p = gggrid([\n",
" ggplot(data, aes('code', 'value')) + geom_ydotplot(),\n",
" ggplot(data, aes('value', 'code')) + geom_ydotplot(orientation='y'),\n",
" ggplot(data, aes('value', 'code')) + geom_ydotplot(),\n",
" ggplot(data) + geom_ydotplot(aes('value', 'code')),\n",
" ggplot(data, aes('value', as_discrete('code'))) + geom_ydotplot() \n",
"])\n",
"dump_plot(p)"
]
},
{
"cell_type": "markdown",
"id": "632d6de8",
"metadata": {},
"source": [
"## Some Special Cases"
]
},
{
"cell_type": "markdown",
"id": "92f95198",
"metadata": {},
"source": [
"### `geom_smooth()`\n",
"`geom_smooth()` needs `orientation=”y”` when we want to flip axis."
]
},
{
"cell_type": "code",
"execution_count": 16,
"id": "bb724b19",
"metadata": {},
"outputs": [],
"source": [
"np.random.seed(42)\n",
"n = 100\n",
"x = np.linspace(-2, 2, n)\n",
"y = x ** 2 + np.random.normal(size=n)"
]
},
{
"cell_type": "code",
"execution_count": 17,
"id": "d5bd40cb",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 17,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gggrid([\n",
" ggplot({'x': x, 'y': y}, aes(x='x', y='y')) + geom_point() + geom_smooth(deg=2, se=False),\n",
" ggplot({'x': x, 'y': y}, aes(x='y', y='x')) + geom_point() + geom_smooth(deg=2, se=False),\n",
" ggplot({'x': x, 'y': y}, aes(x='y', y='x')) + geom_point() + geom_smooth(deg=2, se=False, orientation=\"y\"),\n",
"])"
]
},
{
"cell_type": "markdown",
"id": "aa2e0c86",
"metadata": {},
"source": [
"### Few Datasets"
]
},
{
"cell_type": "markdown",
"id": "bbc0359b",
"metadata": {},
"source": [
"If we have some dataset in the `ggplot()` context, then we can use data from it in `geom_boxplot()`. And, in this case, we can use variable names from the `ggplot()` context even if we set another context in `geom_boxplot()`."
]
},
{
"cell_type": "code",
"execution_count": 18,
"id": "baf88e20",
"metadata": {},
"outputs": [],
"source": [
"np.random.seed(42)\n",
"data = {\n",
" 'code': np.random.choice(list('abcde'), size=100),\n",
" 'value': np.random.normal(size=100),\n",
" 'value_str': [str(i) for i in np.random.normal(size=100)],\n",
"}\n",
"data_num = {\n",
" 'number': np.random.choice(list('qwxyz'), size=100),\n",
" 'volume': np.random.normal(size=100),\n",
"}"
]
},
{
"cell_type": "code",
"execution_count": 19,
"id": "956932a3",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 19,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gggrid([\n",
" ggplot(data, aes('value', 'code')) + geom_boxplot(aes('number', 'volume'), data = data_num),\n",
" ggplot(data, aes('value', 'code')) + geom_boxplot(aes('value', 'code'), data = data_num)\n",
"])"
]
},
{
"cell_type": "markdown",
"id": "b9a2bded",
"metadata": {},
"source": [
"### `stat_summary(geom='crossbar')`"
]
},
{
"cell_type": "code",
"execution_count": 20,
"id": "f0976fe8",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 20,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gggrid([\n",
" ggplot(data, aes('code', 'value')) + stat_summary(geom='crossbar'),\n",
" ggplot(data, aes('value', 'code')) + stat_summary(geom='crossbar') \n",
"])"
]
},
{
"cell_type": "markdown",
"id": "f949d6de",
"metadata": {},
"source": [
"### Discrete Both Axes"
]
},
{
"cell_type": "code",
"execution_count": 21,
"id": "0aba6fbb",
"metadata": {},
"outputs": [],
"source": [
"x = ['a', 'a', 'b', 'b']\n",
"y = ['x','x', 'x', 'y']"
]
},
{
"cell_type": "markdown",
"id": "fa99080f",
"metadata": {},
"source": [
"For `stat_sum()` and `geom_point()`, rotation does not work for both discrete axes."
]
},
{
"cell_type": "code",
"execution_count": 22,
"id": "14e36676",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 22,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gggrid([\n",
" ggplot({'x': x, 'y': y}, aes(x='x', y='y')) + stat_sum(),\n",
" ggplot({'x': x, 'y': y}, aes(x='x', y='y')) + geom_point(orientation=\"y\")\n",
"])"
]
},
{
"cell_type": "markdown",
"id": "5a761ca2",
"metadata": {},
"source": [
"The lollipop can be drawing when both axes are discrete. But this case is not usual for a lollipop."
]
},
{
"cell_type": "code",
"execution_count": 23,
"id": "f5ccf129",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 23,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"x = ['a', 'a', 'a', 'b', 'b']\n",
"y = ['x','x', 'x', 'y', 'y']\n",
"gggrid([\n",
" ggplot({'x': x, 'y': y}, aes(x='x', y='y')) + geom_lollipop(stat='count'),\n",
" ggplot({'x': x, 'y': y}, aes(x='x')) + geom_lollipop(stat='count'),\n",
" ggplot({'x': x, 'y': y}, aes(y='y')) + geom_lollipop(stat='count'), \n",
" ggplot({'x': x, 'y': y}, aes(y='y')) + geom_lollipop(stat='count', orientation='x'), \n",
" ggplot({'x': x, 'y': y}, aes(x='x', y='y')) + geom_lollipop(stat='count', orientation='y') \n",
"])"
]
},
{
"cell_type": "markdown",
"id": "c5b003f3",
"metadata": {},
"source": [
"### `stat='boxplot'`"
]
},
{
"cell_type": "code",
"execution_count": 24,
"id": "70805e17",
"metadata": {},
"outputs": [],
"source": [
"n = 50\n",
"np.random.seed(42)\n",
"x = np.random.uniform(size=n)\n",
"c = np.random.choice(['a', 'b', 'c'], size=n)"
]
},
{
"cell_type": "code",
"execution_count": 25,
"id": "e27e1ed0",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"p1 = ggplot({'x': x, 'c': c}) + \\\n",
" geom_crossbar(aes(x='c', y='x', color='c'), \\\n",
" stat='boxplot') + \\\n",
" geom_point(aes(x='c', y='x', color='c'), \\\n",
" size=4, shape=21, fill='white',\n",
" position=position_jitterdodge(seed=42))\n",
"p2 = ggplot({'x': x, 'c': c}) + \\\n",
" geom_crossbar(aes(x='x', y='c', color='c'), \\\n",
" stat='boxplot') + \\\n",
" geom_point(aes(x='x', y='c', color='c'), \\\n",
" size=4, shape=21, fill='white',\n",
" position=position_jitterdodge(seed=42))\n",
"gggrid([p1, p2])"
]
},
{
"cell_type": "code",
"execution_count": 26,
"id": "0986ea7d",
"metadata": {},
"outputs": [],
"source": [
"n = 100\n",
"np.random.seed(42)\n",
"x = np.random.choice(['a', 'b', 'c'], size=n)\n",
"y = np.random.normal(size=n)"
]
},
{
"cell_type": "code",
"execution_count": 27,
"id": "e687f5c9",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 27,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gggrid([\n",
" ggplot({'x': x, 'y': y}, aes(x='x', y='y')) + geom_pointrange(stat='boxplot'),\n",
" ggplot({'x': x, 'y': y}, aes(x='y', y='x')) + geom_pointrange(stat='boxplot')\n",
"])"
]
},
{
"cell_type": "markdown",
"id": "6604561b",
"metadata": {},
"source": [
"### `stat='boxplot_outlier'`"
]
},
{
"cell_type": "code",
"execution_count": 28,
"id": "c7b7a899",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 28,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gggrid([\n",
" ggplot({'x': x, 'y': y}, aes(x='x', y='y')) + geom_pointrange(stat='boxplot_outlier'),\n",
" ggplot({'x': x, 'y': y}, aes(x='y', y='x')) + geom_pointrange(stat='boxplot_outlier')\n",
"])"
]
},
{
"cell_type": "markdown",
"id": "57f10f16",
"metadata": {},
"source": [
"### `stat_summary()`"
]
},
{
"cell_type": "code",
"execution_count": 29,
"id": "6bdc4826",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 29,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"p1 = ggplot({'x': x, 'y': y}, aes(x='x', y='y')) + \\\n",
" stat_summary(position=position_nudge(x=-.1), color=\"red\") + \\\n",
" stat_summary(fun='mq', fun_min='lq', fun_max='uq', quantiles=[.1, .5, .9], \\\n",
" position=position_nudge(x=.1), color=\"blue\")\n",
"p2 = p=ggplot({'x': x, 'y': y}, aes(x='y', y='x')) + \\\n",
" stat_summary(position=position_nudge(x=-.1), color=\"red\") + \\\n",
" stat_summary(fun='mq', fun_min='lq', fun_max='uq', quantiles=[.1, .5, .9], \\\n",
" position=position_nudge(x=.1), color=\"blue\")\n",
"gggrid([p1, p2])"
]
},
{
"cell_type": "markdown",
"id": "a44e0244",
"metadata": {},
"source": [
"### `stat='ydensity'`"
]
},
{
"cell_type": "code",
"execution_count": 30,
"id": "3248a6b1",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 30,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gggrid([ \n",
" ggplot({'x': x, 'y': y}, aes('x', 'y')) + geom_crossbar(aes(ymin='..y..', ymax='..y..', width='..violinwidth..'), stat='ydensity'),\n",
" ggplot({'x': x, 'y': y}, aes('y', 'x')) + geom_crossbar(aes(ymin='..y..', ymax='..y..', width='..violinwidth..'), stat='ydensity', orientation='y'),\n",
" ggplot({'x': x, 'y': y}, aes('y', 'x')) + geom_crossbar(aes(ymin='..y..', ymax='..y..', width='..violinwidth..'), stat='ydensity')\n",
"])"
]
},
{
"cell_type": "markdown",
"id": "5296f46a",
"metadata": {},
"source": [
"### `stat='ydotplot'`"
]
},
{
"cell_type": "code",
"execution_count": 31,
"id": "9d820d5e",
"metadata": {},
"outputs": [],
"source": [
"n = 100\n",
"np.random.seed(42)\n",
"x = np.random.choice(['a'], size=n)\n",
"y = np.random.normal(size=n)"
]
},
{
"cell_type": "code",
"execution_count": 32,
"id": "08d6ea01",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
],
"text/plain": [
""
]
},
"execution_count": 32,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"gggrid([ \n",
" ggplot({'x': x, 'y': y}, aes(x='x', y='y')) + geom_crossbar(aes(ymin='..y..', ymax='..y..', width='..count..'), stat='ydotplot'),\n",
" ggplot({'x': x, 'y': y}, aes(x='y', y='x')) + geom_crossbar(aes(ymin='..y..', ymax='..y..', width='..count..'), stat='ydotplot', orientation='y'),\n",
" ggplot({'x': x, 'y': y}, aes(x='y', y='x')) + geom_crossbar(aes(ymin='..y..', ymax='..y..', width='..count..'), stat='ydotplot')\n",
"])"
]
}
],
"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": "3.10.11"
}
},
"nbformat": 4,
"nbformat_minor": 5
}