{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import pandas as pd\n",
"import numpy as np"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"np.random.seed(101)\n",
"samples = ['A','B','C','D','E']\n",
"features = ['W','X','Y','Z']\n",
"df = pd.DataFrame(np.random.randn(5,4), samples, features)"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>W</th>\n",
" <th>X</th>\n",
" <th>Y</th>\n",
" <th>Z</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>A</th>\n",
" <td>2.706850</td>\n",
" <td>0.628133</td>\n",
" <td>0.907969</td>\n",
" <td>0.503826</td>\n",
" </tr>\n",
" <tr>\n",
" <th>B</th>\n",
" <td>0.651118</td>\n",
" <td>-0.319318</td>\n",
" <td>-0.848077</td>\n",
" <td>0.605965</td>\n",
" </tr>\n",
" <tr>\n",
" <th>C</th>\n",
" <td>-2.018168</td>\n",
" <td>0.740122</td>\n",
" <td>0.528813</td>\n",
" <td>-0.589001</td>\n",
" </tr>\n",
" <tr>\n",
" <th>D</th>\n",
" <td>0.188695</td>\n",
" <td>-0.758872</td>\n",
" <td>-0.933237</td>\n",
" <td>0.955057</td>\n",
" </tr>\n",
" <tr>\n",
" <th>E</th>\n",
" <td>0.190794</td>\n",
" <td>1.978757</td>\n",
" <td>2.605967</td>\n",
" <td>0.683509</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" W X Y Z\n",
"A 2.706850 0.628133 0.907969 0.503826\n",
"B 0.651118 -0.319318 -0.848077 0.605965\n",
"C -2.018168 0.740122 0.528813 -0.589001\n",
"D 0.188695 -0.758872 -0.933237 0.955057\n",
"E 0.190794 1.978757 2.605967 0.683509"
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"W 2.706850\n",
"X 0.628133\n",
"Y 0.907969\n",
"Z 0.503826\n",
"Name: A, dtype: float64"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.loc['A', :]"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"2.706849839399938"
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.loc['A', 'W']"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"A 2.706850\n",
"B 0.651118\n",
"C -2.018168\n",
"D 0.188695\n",
"E 0.190794\n",
"Name: W, dtype: float64"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.W"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
"df['X+Y'] = df.X + df.Y"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>W</th>\n",
" <th>X</th>\n",
" <th>Y</th>\n",
" <th>Z</th>\n",
" <th>X+Y</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>A</th>\n",
" <td>2.706850</td>\n",
" <td>0.628133</td>\n",
" <td>0.907969</td>\n",
" <td>0.503826</td>\n",
" <td>1.536102</td>\n",
" </tr>\n",
" <tr>\n",
" <th>B</th>\n",
" <td>0.651118</td>\n",
" <td>-0.319318</td>\n",
" <td>-0.848077</td>\n",
" <td>0.605965</td>\n",
" <td>-1.167395</td>\n",
" </tr>\n",
" <tr>\n",
" <th>C</th>\n",
" <td>-2.018168</td>\n",
" <td>0.740122</td>\n",
" <td>0.528813</td>\n",
" <td>-0.589001</td>\n",
" <td>1.268936</td>\n",
" </tr>\n",
" <tr>\n",
" <th>D</th>\n",
" <td>0.188695</td>\n",
" <td>-0.758872</td>\n",
" <td>-0.933237</td>\n",
" <td>0.955057</td>\n",
" <td>-1.692109</td>\n",
" </tr>\n",
" <tr>\n",
" <th>E</th>\n",
" <td>0.190794</td>\n",
" <td>1.978757</td>\n",
" <td>2.605967</td>\n",
" <td>0.683509</td>\n",
" <td>4.584725</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" W X Y Z X+Y\n",
"A 2.706850 0.628133 0.907969 0.503826 1.536102\n",
"B 0.651118 -0.319318 -0.848077 0.605965 -1.167395\n",
"C -2.018168 0.740122 0.528813 -0.589001 1.268936\n",
"D 0.188695 -0.758872 -0.933237 0.955057 -1.692109\n",
"E 0.190794 1.978757 2.605967 0.683509 4.584725"
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>W</th>\n",
" <th>X</th>\n",
" <th>Y</th>\n",
" <th>Z</th>\n",
" <th>X+Y</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>A</th>\n",
" <td>True</td>\n",
" <td>True</td>\n",
" <td>True</td>\n",
" <td>True</td>\n",
" <td>True</td>\n",
" </tr>\n",
" <tr>\n",
" <th>B</th>\n",
" <td>True</td>\n",
" <td>False</td>\n",
" <td>False</td>\n",
" <td>True</td>\n",
" <td>False</td>\n",
" </tr>\n",
" <tr>\n",
" <th>C</th>\n",
" <td>False</td>\n",
" <td>True</td>\n",
" <td>True</td>\n",
" <td>False</td>\n",
" <td>True</td>\n",
" </tr>\n",
" <tr>\n",
" <th>D</th>\n",
" <td>True</td>\n",
" <td>False</td>\n",
" <td>False</td>\n",
" <td>True</td>\n",
" <td>False</td>\n",
" </tr>\n",
" <tr>\n",
" <th>E</th>\n",
" <td>True</td>\n",
" <td>True</td>\n",
" <td>True</td>\n",
" <td>True</td>\n",
" <td>True</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" W X Y Z X+Y\n",
"A True True True True True\n",
"B True False False True False\n",
"C False True True False True\n",
"D True False False True False\n",
"E True True True True True"
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df > 0"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>W</th>\n",
" <th>X</th>\n",
" <th>Y</th>\n",
" <th>Z</th>\n",
" <th>X+Y</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>A</th>\n",
" <td>2.706850</td>\n",
" <td>0.628133</td>\n",
" <td>0.907969</td>\n",
" <td>0.503826</td>\n",
" <td>1.536102</td>\n",
" </tr>\n",
" <tr>\n",
" <th>B</th>\n",
" <td>0.651118</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>0.605965</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>C</th>\n",
" <td>NaN</td>\n",
" <td>0.740122</td>\n",
" <td>0.528813</td>\n",
" <td>NaN</td>\n",
" <td>1.268936</td>\n",
" </tr>\n",
" <tr>\n",
" <th>D</th>\n",
" <td>0.188695</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>0.955057</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>E</th>\n",
" <td>0.190794</td>\n",
" <td>1.978757</td>\n",
" <td>2.605967</td>\n",
" <td>0.683509</td>\n",
" <td>4.584725</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" W X Y Z X+Y\n",
"A 2.706850 0.628133 0.907969 0.503826 1.536102\n",
"B 0.651118 NaN NaN 0.605965 NaN\n",
"C NaN 0.740122 0.528813 NaN 1.268936\n",
"D 0.188695 NaN NaN 0.955057 NaN\n",
"E 0.190794 1.978757 2.605967 0.683509 4.584725"
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df[df > 0]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Groupby"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"data = {\n",
" 'company':'Google Google Microsoft Microsoft Facebook Facebook'.split(),\n",
" 'Person':'A B C D E F'.split(),\n",
" 'Sales':[200, 300, 100, 120, 400, 500]\n",
"}"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"{'Person': ['A', 'B', 'C', 'D', 'E', 'F'],\n",
" 'Sales': [200, 300, 100, 120, 400, 500],\n",
" 'company': ['Google',\n",
" 'Google',\n",
" 'Microsoft',\n",
" 'Microsoft',\n",
" 'Facebook',\n",
" 'Facebook']}"
]
},
"execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"data"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Person</th>\n",
" <th>Sales</th>\n",
" <th>company</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>A</td>\n",
" <td>200</td>\n",
" <td>Google</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>B</td>\n",
" <td>300</td>\n",
" <td>Google</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>C</td>\n",
" <td>100</td>\n",
" <td>Microsoft</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>D</td>\n",
" <td>120</td>\n",
" <td>Microsoft</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>E</td>\n",
" <td>400</td>\n",
" <td>Facebook</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>F</td>\n",
" <td>500</td>\n",
" <td>Facebook</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Person Sales company\n",
"0 A 200 Google\n",
"1 B 300 Google\n",
"2 C 100 Microsoft\n",
"3 D 120 Microsoft\n",
"4 E 400 Facebook\n",
"5 F 500 Facebook"
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df = pd.DataFrame(data)\n",
"df"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Person</th>\n",
" <th>Sales</th>\n",
" </tr>\n",
" <tr>\n",
" <th>company</th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>Facebook</th>\n",
" <td>F</td>\n",
" <td>500</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Google</th>\n",
" <td>B</td>\n",
" <td>300</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Microsoft</th>\n",
" <td>D</td>\n",
" <td>120</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Person Sales\n",
"company \n",
"Facebook F 500\n",
"Google B 300\n",
"Microsoft D 120"
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.groupby('company').max()"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Sales</th>\n",
" </tr>\n",
" <tr>\n",
" <th>company</th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>Facebook</th>\n",
" <td>450</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Google</th>\n",
" <td>250</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Microsoft</th>\n",
" <td>110</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Sales\n",
"company \n",
"Facebook 450\n",
"Google 250\n",
"Microsoft 110"
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.groupby('company').mean()"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Sales</th>\n",
" </tr>\n",
" <tr>\n",
" <th>company</th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>Facebook</th>\n",
" <td>70.710678</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Google</th>\n",
" <td>70.710678</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Microsoft</th>\n",
" <td>14.142136</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Sales\n",
"company \n",
"Facebook 70.710678\n",
"Google 70.710678\n",
"Microsoft 14.142136"
]
},
"execution_count": 16,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.groupby('company').std()"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead tr th {\n",
" text-align: left;\n",
" }\n",
"\n",
" .dataframe thead tr:last-of-type th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr>\n",
" <th></th>\n",
" <th colspan=\"8\" halign=\"left\">Sales</th>\n",
" </tr>\n",
" <tr>\n",
" <th></th>\n",
" <th>count</th>\n",
" <th>mean</th>\n",
" <th>std</th>\n",
" <th>min</th>\n",
" <th>25%</th>\n",
" <th>50%</th>\n",
" <th>75%</th>\n",
" <th>max</th>\n",
" </tr>\n",
" <tr>\n",
" <th>company</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>Facebook</th>\n",
" <td>2.0</td>\n",
" <td>450.0</td>\n",
" <td>70.710678</td>\n",
" <td>400.0</td>\n",
" <td>425.0</td>\n",
" <td>450.0</td>\n",
" <td>475.0</td>\n",
" <td>500.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Google</th>\n",
" <td>2.0</td>\n",
" <td>250.0</td>\n",
" <td>70.710678</td>\n",
" <td>200.0</td>\n",
" <td>225.0</td>\n",
" <td>250.0</td>\n",
" <td>275.0</td>\n",
" <td>300.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Microsoft</th>\n",
" <td>2.0</td>\n",
" <td>110.0</td>\n",
" <td>14.142136</td>\n",
" <td>100.0</td>\n",
" <td>105.0</td>\n",
" <td>110.0</td>\n",
" <td>115.0</td>\n",
" <td>120.0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Sales \n",
" count mean std min 25% 50% 75% max\n",
"company \n",
"Facebook 2.0 450.0 70.710678 400.0 425.0 450.0 475.0 500.0\n",
"Google 2.0 250.0 70.710678 200.0 225.0 250.0 275.0 300.0\n",
"Microsoft 2.0 110.0 14.142136 100.0 105.0 110.0 115.0 120.0"
]
},
"execution_count": 17,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.groupby('company').describe()"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Facebook 2\n",
"Google 2\n",
"Microsoft 2\n",
"Name: company, dtype: int64"
]
},
"execution_count": 18,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.company.value_counts()"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Person</th>\n",
" <th>Sales</th>\n",
" <th>company</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>A</td>\n",
" <td>200</td>\n",
" <td>Google</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>B</td>\n",
" <td>300</td>\n",
" <td>Google</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>C</td>\n",
" <td>100</td>\n",
" <td>Microsoft</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>D</td>\n",
" <td>120</td>\n",
" <td>Microsoft</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>E</td>\n",
" <td>400</td>\n",
" <td>Facebook</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>F</td>\n",
" <td>500</td>\n",
" <td>Facebook</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Person Sales company\n",
"0 A 200 Google\n",
"1 B 300 Google\n",
"2 C 100 Microsoft\n",
"3 D 120 Microsoft\n",
"4 E 400 Facebook\n",
"5 F 500 Facebook"
]
},
"execution_count": 19,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Person</th>\n",
" <th>Sales</th>\n",
" <th>company</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>F</td>\n",
" <td>500</td>\n",
" <td>Facebook</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>E</td>\n",
" <td>400</td>\n",
" <td>Facebook</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>B</td>\n",
" <td>300</td>\n",
" <td>Google</td>\n",
" </tr>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>A</td>\n",
" <td>200</td>\n",
" <td>Google</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>D</td>\n",
" <td>120</td>\n",
" <td>Microsoft</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>C</td>\n",
" <td>100</td>\n",
" <td>Microsoft</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" Person Sales company\n",
"5 F 500 Facebook\n",
"4 E 400 Facebook\n",
"1 B 300 Google\n",
"0 A 200 Google\n",
"3 D 120 Microsoft\n",
"2 C 100 Microsoft"
]
},
"execution_count": 20,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df.sort_values(by='Sales', ascending=False)"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Bank Name</th>\n",
" <th>City</th>\n",
" <th>ST</th>\n",
" <th>CERT</th>\n",
" <th>Acquiring Institution</th>\n",
" <th>Closing Date</th>\n",
" <th>Updated Date</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>Washington Federal Bank for Savings</td>\n",
" <td>Chicago</td>\n",
" <td>IL</td>\n",
" <td>30570</td>\n",
" <td>Royal Savings Bank</td>\n",
" <td>December 15, 2017</td>\n",
" <td>February 21, 2018</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>The Farmers and Merchants State Bank of Argonia</td>\n",
" <td>Argonia</td>\n",
" <td>KS</td>\n",
" <td>17719</td>\n",
" <td>Conway Bank</td>\n",
" <td>October 13, 2017</td>\n",
" <td>February 21, 2018</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>Fayette County Bank</td>\n",
" <td>Saint Elmo</td>\n",
" <td>IL</td>\n",
" <td>1802</td>\n",
" <td>United Fidelity Bank, fsb</td>\n",
" <td>May 26, 2017</td>\n",
" <td>July 26, 2017</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>Guaranty Bank, (d/b/a BestBank in Georgia & Mi...</td>\n",
" <td>Milwaukee</td>\n",
" <td>WI</td>\n",
" <td>30003</td>\n",
" <td>First-Citizens Bank & Trust Company</td>\n",
" <td>May 5, 2017</td>\n",
" <td>March 22, 2018</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>First NBC Bank</td>\n",
" <td>New Orleans</td>\n",
" <td>LA</td>\n",
" <td>58302</td>\n",
" <td>Whitney Bank</td>\n",
" <td>April 28, 2017</td>\n",
" <td>December 5, 2017</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>Proficio Bank</td>\n",
" <td>Cottonwood Heights</td>\n",
" <td>UT</td>\n",
" <td>35495</td>\n",
" <td>Cache Valley Bank</td>\n",
" <td>March 3, 2017</td>\n",
" <td>March 7, 2018</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td>Seaway Bank and Trust Company</td>\n",
" <td>Chicago</td>\n",
" <td>IL</td>\n",
" <td>19328</td>\n",
" <td>State Bank of Texas</td>\n",
" <td>January 27, 2017</td>\n",
" <td>May 18, 2017</td>\n",
" </tr>\n",
" <tr>\n",
" <th>7</th>\n",
" <td>Harvest Community Bank</td>\n",
" <td>Pennsville</td>\n",
" <td>NJ</td>\n",
" <td>34951</td>\n",
" <td>First-Citizens Bank & Trust Company</td>\n",
" <td>January 13, 2017</td>\n",
" <td>May 18, 2017</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8</th>\n",
" <td>Allied Bank</td>\n",
" <td>Mulberry</td>\n",
" <td>AR</td>\n",
" <td>91</td>\n",
" <td>Today's Bank</td>\n",
" <td>September 23, 2016</td>\n",
" <td>September 25, 2017</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9</th>\n",
" <td>The Woodbury Banking Company</td>\n",
" <td>Woodbury</td>\n",
" <td>GA</td>\n",
" <td>11297</td>\n",
" <td>United Bank</td>\n",
" <td>August 19, 2016</td>\n",
" <td>June 1, 2017</td>\n",
" </tr>\n",
" <tr>\n",
" <th>10</th>\n",
" <td>First CornerStone Bank</td>\n",
" <td>King of Prussia</td>\n",
" <td>PA</td>\n",
" <td>35312</td>\n",
" <td>First-Citizens Bank & Trust Company</td>\n",
" <td>May 6, 2016</td>\n",
" <td>September 6, 2016</td>\n",
" </tr>\n",
" <tr>\n",
" <th>11</th>\n",
" <td>Trust Company Bank</td>\n",
" <td>Memphis</td>\n",
" <td>TN</td>\n",
" <td>9956</td>\n",
" <td>The Bank of Fayette County</td>\n",
" <td>April 29, 2016</td>\n",
" <td>September 6, 2016</td>\n",
" </tr>\n",
" <tr>\n",
" <th>12</th>\n",
" <td>North Milwaukee State Bank</td>\n",
" <td>Milwaukee</td>\n",
" <td>WI</td>\n",
" <td>20364</td>\n",
" <td>First-Citizens Bank & Trust Company</td>\n",
" <td>March 11, 2016</td>\n",
" <td>March 13, 2017</td>\n",
" </tr>\n",
" <tr>\n",
" <th>13</th>\n",
" <td>Hometown National Bank</td>\n",
" <td>Longview</td>\n",
" <td>WA</td>\n",
" <td>35156</td>\n",
" <td>Twin City Bank</td>\n",
" <td>October 2, 2015</td>\n",
" <td>February 19, 2018</td>\n",
" </tr>\n",
" <tr>\n",
" <th>14</th>\n",
" <td>The Bank of Georgia</td>\n",
" <td>Peachtree City</td>\n",
" <td>GA</td>\n",
" <td>35259</td>\n",
" <td>Fidelity Bank</td>\n",
" <td>October 2, 2015</td>\n",
" <td>July 9, 2018</td>\n",
" </tr>\n",
" <tr>\n",
" <th>15</th>\n",
" <td>Premier Bank</td>\n",
" <td>Denver</td>\n",
" <td>CO</td>\n",
" <td>34112</td>\n",
" <td>United Fidelity Bank, fsb</td>\n",
" <td>July 10, 2015</td>\n",
" <td>February 20, 2018</td>\n",
" </tr>\n",
" <tr>\n",
" <th>16</th>\n",
" <td>Edgebrook Bank</td>\n",
" <td>Chicago</td>\n",
" <td>IL</td>\n",
" <td>57772</td>\n",
" <td>Republic Bank of Chicago</td>\n",
" <td>May 8, 2015</td>\n",
" <td>July 12, 2016</td>\n",
" </tr>\n",
" <tr>\n",
" <th>17</th>\n",
" <td>Doral Bank En Espanol</td>\n",
" <td>San Juan</td>\n",
" <td>PR</td>\n",
" <td>32102</td>\n",
" <td>Banco Popular de Puerto Rico</td>\n",
" <td>February 27, 2015</td>\n",
" <td>May 13, 2015</td>\n",
" </tr>\n",
" <tr>\n",
" <th>18</th>\n",
" <td>Capitol City Bank & Trust Company</td>\n",
" <td>Atlanta</td>\n",
" <td>GA</td>\n",
" <td>33938</td>\n",
" <td>First-Citizens Bank & Trust Company</td>\n",
" <td>February 13, 2015</td>\n",
" <td>April 21, 2015</td>\n",
" </tr>\n",
" <tr>\n",
" <th>19</th>\n",
" <td>Highland Community Bank</td>\n",
" <td>Chicago</td>\n",
" <td>IL</td>\n",
" <td>20290</td>\n",
" <td>United Fidelity Bank, fsb</td>\n",
" <td>January 23, 2015</td>\n",
" <td>November 15, 2017</td>\n",
" </tr>\n",
" <tr>\n",
" <th>20</th>\n",
" <td>First National Bank of Crestview</td>\n",
" <td>Crestview</td>\n",
" <td>FL</td>\n",
" <td>17557</td>\n",
" <td>First NBC Bank</td>\n",
" <td>January 16, 2015</td>\n",
" <td>November 15, 2017</td>\n",
" </tr>\n",
" <tr>\n",
" <th>21</th>\n",
" <td>Northern Star Bank</td>\n",
" <td>Mankato</td>\n",
" <td>MN</td>\n",
" <td>34983</td>\n",
" <td>BankVista</td>\n",
" <td>December 19, 2014</td>\n",
" <td>January 3, 2018</td>\n",
" </tr>\n",
" <tr>\n",
" <th>22</th>\n",
" <td>Frontier Bank, FSB D/B/A El Paseo Bank</td>\n",
" <td>Palm Desert</td>\n",
" <td>CA</td>\n",
" <td>34738</td>\n",
" <td>Bank of Southern California, N.A.</td>\n",
" <td>November 7, 2014</td>\n",
" <td>November 10, 2016</td>\n",
" </tr>\n",
" <tr>\n",
" <th>23</th>\n",
" <td>The National Republic Bank of Chicago</td>\n",
" <td>Chicago</td>\n",
" <td>IL</td>\n",
" <td>916</td>\n",
" <td>State Bank of Texas</td>\n",
" <td>October 24, 2014</td>\n",
" <td>January 6, 2016</td>\n",
" </tr>\n",
" <tr>\n",
" <th>24</th>\n",
" <td>NBRS Financial</td>\n",
" <td>Rising Sun</td>\n",
" <td>MD</td>\n",
" <td>4862</td>\n",
" <td>Howard Bank</td>\n",
" <td>October 17, 2014</td>\n",
" <td>February 19, 2018</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25</th>\n",
" <td>GreenChoice Bank, fsb</td>\n",
" <td>Chicago</td>\n",
" <td>IL</td>\n",
" <td>28462</td>\n",
" <td>Providence Bank, LLC</td>\n",
" <td>July 25, 2014</td>\n",
" <td>December 12, 2016</td>\n",
" </tr>\n",
" <tr>\n",
" <th>26</th>\n",
" <td>Eastside Commercial Bank</td>\n",
" <td>Conyers</td>\n",
" <td>GA</td>\n",
" <td>58125</td>\n",
" <td>Community & Southern Bank</td>\n",
" <td>July 18, 2014</td>\n",
" <td>October 6, 2017</td>\n",
" </tr>\n",
" <tr>\n",
" <th>27</th>\n",
" <td>The Freedom State Bank</td>\n",
" <td>Freedom</td>\n",
" <td>OK</td>\n",
" <td>12483</td>\n",
" <td>Alva State Bank & Trust Company</td>\n",
" <td>June 27, 2014</td>\n",
" <td>February 21, 2018</td>\n",
" </tr>\n",
" <tr>\n",
" <th>28</th>\n",
" <td>Valley Bank</td>\n",
" <td>Fort Lauderdale</td>\n",
" <td>FL</td>\n",
" <td>21793</td>\n",
" <td>Landmark Bank, National Association</td>\n",
" <td>June 20, 2014</td>\n",
" <td>February 14, 2018</td>\n",
" </tr>\n",
" <tr>\n",
" <th>29</th>\n",
" <td>Valley Bank</td>\n",
" <td>Moline</td>\n",
" <td>IL</td>\n",
" <td>10450</td>\n",
" <td>Great Southern Bank</td>\n",
" <td>June 20, 2014</td>\n",
" <td>June 26, 2015</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>525</th>\n",
" <td>ANB Financial, NA</td>\n",
" <td>Bentonville</td>\n",
" <td>AR</td>\n",
" <td>33901</td>\n",
" <td>Pulaski Bank and Trust Company</td>\n",
" <td>May 9, 2008</td>\n",
" <td>August 28, 2012</td>\n",
" </tr>\n",
" <tr>\n",
" <th>526</th>\n",
" <td>Hume Bank</td>\n",
" <td>Hume</td>\n",
" <td>MO</td>\n",
" <td>1971</td>\n",
" <td>Security Bank</td>\n",
" <td>March 7, 2008</td>\n",
" <td>August 28, 2012</td>\n",
" </tr>\n",
" <tr>\n",
" <th>527</th>\n",
" <td>Douglass National Bank</td>\n",
" <td>Kansas City</td>\n",
" <td>MO</td>\n",
" <td>24660</td>\n",
" <td>Liberty Bank and Trust Company</td>\n",
" <td>January 25, 2008</td>\n",
" <td>October 26, 2012</td>\n",
" </tr>\n",
" <tr>\n",
" <th>528</th>\n",
" <td>Miami Valley Bank</td>\n",
" <td>Lakeview</td>\n",
" <td>OH</td>\n",
" <td>16848</td>\n",
" <td>The Citizens Banking Company</td>\n",
" <td>October 4, 2007</td>\n",
" <td>September 12, 2016</td>\n",
" </tr>\n",
" <tr>\n",
" <th>529</th>\n",
" <td>NetBank</td>\n",
" <td>Alpharetta</td>\n",
" <td>GA</td>\n",
" <td>32575</td>\n",
" <td>ING DIRECT</td>\n",
" <td>September 28, 2007</td>\n",
" <td>August 28, 2012</td>\n",
" </tr>\n",
" <tr>\n",
" <th>530</th>\n",
" <td>Metropolitan Savings Bank</td>\n",
" <td>Pittsburgh</td>\n",
" <td>PA</td>\n",
" <td>35353</td>\n",
" <td>Allegheny Valley Bank of Pittsburgh</td>\n",
" <td>February 2, 2007</td>\n",
" <td>October 27, 2010</td>\n",
" </tr>\n",
" <tr>\n",
" <th>531</th>\n",
" <td>Bank of Ephraim</td>\n",
" <td>Ephraim</td>\n",
" <td>UT</td>\n",
" <td>1249</td>\n",
" <td>Far West Bank</td>\n",
" <td>June 25, 2004</td>\n",
" <td>April 9, 2008</td>\n",
" </tr>\n",
" <tr>\n",
" <th>532</th>\n",
" <td>Reliance Bank</td>\n",
" <td>White Plains</td>\n",
" <td>NY</td>\n",
" <td>26778</td>\n",
" <td>Union State Bank</td>\n",
" <td>March 19, 2004</td>\n",
" <td>April 9, 2008</td>\n",
" </tr>\n",
" <tr>\n",
" <th>533</th>\n",
" <td>Guaranty National Bank of Tallahassee</td>\n",
" <td>Tallahassee</td>\n",
" <td>FL</td>\n",
" <td>26838</td>\n",
" <td>Hancock Bank of Florida</td>\n",
" <td>March 12, 2004</td>\n",
" <td>April 17, 2018</td>\n",
" </tr>\n",
" <tr>\n",
" <th>534</th>\n",
" <td>Dollar Savings Bank</td>\n",
" <td>Newark</td>\n",
" <td>NJ</td>\n",
" <td>31330</td>\n",
" <td>No Acquirer</td>\n",
" <td>February 14, 2004</td>\n",
" <td>April 9, 2008</td>\n",
" </tr>\n",
" <tr>\n",
" <th>535</th>\n",
" <td>Pulaski Savings Bank</td>\n",
" <td>Philadelphia</td>\n",
" <td>PA</td>\n",
" <td>27203</td>\n",
" <td>Earthstar Bank</td>\n",
" <td>November 14, 2003</td>\n",
" <td>October 6, 2017</td>\n",
" </tr>\n",
" <tr>\n",
" <th>536</th>\n",
" <td>First National Bank of Blanchardville</td>\n",
" <td>Blanchardville</td>\n",
" <td>WI</td>\n",
" <td>11639</td>\n",
" <td>The Park Bank</td>\n",
" <td>May 9, 2003</td>\n",
" <td>June 5, 2012</td>\n",
" </tr>\n",
" <tr>\n",
" <th>537</th>\n",
" <td>Southern Pacific Bank</td>\n",
" <td>Torrance</td>\n",
" <td>CA</td>\n",
" <td>27094</td>\n",
" <td>Beal Bank</td>\n",
" <td>February 7, 2003</td>\n",
" <td>October 20, 2008</td>\n",
" </tr>\n",
" <tr>\n",
" <th>538</th>\n",
" <td>Farmers Bank of Cheneyville</td>\n",
" <td>Cheneyville</td>\n",
" <td>LA</td>\n",
" <td>16445</td>\n",
" <td>Sabine State Bank & Trust</td>\n",
" <td>December 17, 2002</td>\n",
" <td>October 20, 2004</td>\n",
" </tr>\n",
" <tr>\n",
" <th>539</th>\n",
" <td>Bank of Alamo</td>\n",
" <td>Alamo</td>\n",
" <td>TN</td>\n",
" <td>9961</td>\n",
" <td>No Acquirer</td>\n",
" <td>November 8, 2002</td>\n",
" <td>March 18, 2005</td>\n",
" </tr>\n",
" <tr>\n",
" <th>540</th>\n",
" <td>AmTrade International Bank En Espanol</td>\n",
" <td>Atlanta</td>\n",
" <td>GA</td>\n",
" <td>33784</td>\n",
" <td>No Acquirer</td>\n",
" <td>September 30, 2002</td>\n",
" <td>September 11, 2006</td>\n",
" </tr>\n",
" <tr>\n",
" <th>541</th>\n",
" <td>Universal Federal Savings Bank</td>\n",
" <td>Chicago</td>\n",
" <td>IL</td>\n",
" <td>29355</td>\n",
" <td>Chicago Community Bank</td>\n",
" <td>June 27, 2002</td>\n",
" <td>October 6, 2017</td>\n",
" </tr>\n",
" <tr>\n",
" <th>542</th>\n",
" <td>Connecticut Bank of Commerce</td>\n",
" <td>Stamford</td>\n",
" <td>CT</td>\n",
" <td>19183</td>\n",
" <td>Hudson United Bank</td>\n",
" <td>June 26, 2002</td>\n",
" <td>February 14, 2012</td>\n",
" </tr>\n",
" <tr>\n",
" <th>543</th>\n",
" <td>New Century Bank</td>\n",
" <td>Shelby Township</td>\n",
" <td>MI</td>\n",
" <td>34979</td>\n",
" <td>No Acquirer</td>\n",
" <td>March 28, 2002</td>\n",
" <td>March 18, 2005</td>\n",
" </tr>\n",
" <tr>\n",
" <th>544</th>\n",
" <td>Net 1st National Bank</td>\n",
" <td>Boca Raton</td>\n",
" <td>FL</td>\n",
" <td>26652</td>\n",
" <td>Bank Leumi USA</td>\n",
" <td>March 1, 2002</td>\n",
" <td>April 9, 2008</td>\n",
" </tr>\n",
" <tr>\n",
" <th>545</th>\n",
" <td>NextBank, NA</td>\n",
" <td>Phoenix</td>\n",
" <td>AZ</td>\n",
" <td>22314</td>\n",
" <td>No Acquirer</td>\n",
" <td>February 7, 2002</td>\n",
" <td>February 5, 2015</td>\n",
" </tr>\n",
" <tr>\n",
" <th>546</th>\n",
" <td>Oakwood Deposit Bank Co.</td>\n",
" <td>Oakwood</td>\n",
" <td>OH</td>\n",
" <td>8966</td>\n",
" <td>The State Bank & Trust Company</td>\n",
" <td>February 1, 2002</td>\n",
" <td>October 25, 2012</td>\n",
" </tr>\n",
" <tr>\n",
" <th>547</th>\n",
" <td>Bank of Sierra Blanca</td>\n",
" <td>Sierra Blanca</td>\n",
" <td>TX</td>\n",
" <td>22002</td>\n",
" <td>The Security State Bank of Pecos</td>\n",
" <td>January 18, 2002</td>\n",
" <td>November 6, 2003</td>\n",
" </tr>\n",
" <tr>\n",
" <th>548</th>\n",
" <td>Hamilton Bank, NA En Espanol</td>\n",
" <td>Miami</td>\n",
" <td>FL</td>\n",
" <td>24382</td>\n",
" <td>Israel Discount Bank of New York</td>\n",
" <td>January 11, 2002</td>\n",
" <td>September 21, 2015</td>\n",
" </tr>\n",
" <tr>\n",
" <th>549</th>\n",
" <td>Sinclair National Bank</td>\n",
" <td>Gravette</td>\n",
" <td>AR</td>\n",
" <td>34248</td>\n",
" <td>Delta Trust & Bank</td>\n",
" <td>September 7, 2001</td>\n",
" <td>October 6, 2017</td>\n",
" </tr>\n",
" <tr>\n",
" <th>550</th>\n",
" <td>Superior Bank, FSB</td>\n",
" <td>Hinsdale</td>\n",
" <td>IL</td>\n",
" <td>32646</td>\n",
" <td>Superior Federal, FSB</td>\n",
" <td>July 27, 2001</td>\n",
" <td>August 19, 2014</td>\n",
" </tr>\n",
" <tr>\n",
" <th>551</th>\n",
" <td>Malta National Bank</td>\n",
" <td>Malta</td>\n",
" <td>OH</td>\n",
" <td>6629</td>\n",
" <td>North Valley Bank</td>\n",
" <td>May 3, 2001</td>\n",
" <td>November 18, 2002</td>\n",
" </tr>\n",
" <tr>\n",
" <th>552</th>\n",
" <td>First Alliance Bank & Trust Co.</td>\n",
" <td>Manchester</td>\n",
" <td>NH</td>\n",
" <td>34264</td>\n",
" <td>Southern New Hampshire Bank & Trust</td>\n",
" <td>February 2, 2001</td>\n",
" <td>February 18, 2003</td>\n",
" </tr>\n",
" <tr>\n",
" <th>553</th>\n",
" <td>National State Bank of Metropolis</td>\n",
" <td>Metropolis</td>\n",
" <td>IL</td>\n",
" <td>3815</td>\n",
" <td>Banterra Bank of Marion</td>\n",
" <td>December 14, 2000</td>\n",
" <td>March 17, 2005</td>\n",
" </tr>\n",
" <tr>\n",
" <th>554</th>\n",
" <td>Bank of Honolulu</td>\n",
" <td>Honolulu</td>\n",
" <td>HI</td>\n",
" <td>21029</td>\n",
" <td>Bank of the Orient</td>\n",
" <td>October 13, 2000</td>\n",
" <td>March 17, 2005</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>555 rows × 7 columns</p>\n",
"</div>"
],
"text/plain": [
" Bank Name City \\\n",
"0 Washington Federal Bank for Savings Chicago \n",
"1 The Farmers and Merchants State Bank of Argonia Argonia \n",
"2 Fayette County Bank Saint Elmo \n",
"3 Guaranty Bank, (d/b/a BestBank in Georgia & Mi... Milwaukee \n",
"4 First NBC Bank New Orleans \n",
"5 Proficio Bank Cottonwood Heights \n",
"6 Seaway Bank and Trust Company Chicago \n",
"7 Harvest Community Bank Pennsville \n",
"8 Allied Bank Mulberry \n",
"9 The Woodbury Banking Company Woodbury \n",
"10 First CornerStone Bank King of Prussia \n",
"11 Trust Company Bank Memphis \n",
"12 North Milwaukee State Bank Milwaukee \n",
"13 Hometown National Bank Longview \n",
"14 The Bank of Georgia Peachtree City \n",
"15 Premier Bank Denver \n",
"16 Edgebrook Bank Chicago \n",
"17 Doral Bank En Espanol San Juan \n",
"18 Capitol City Bank & Trust Company Atlanta \n",
"19 Highland Community Bank Chicago \n",
"20 First National Bank of Crestview Crestview \n",
"21 Northern Star Bank Mankato \n",
"22 Frontier Bank, FSB D/B/A El Paseo Bank Palm Desert \n",
"23 The National Republic Bank of Chicago Chicago \n",
"24 NBRS Financial Rising Sun \n",
"25 GreenChoice Bank, fsb Chicago \n",
"26 Eastside Commercial Bank Conyers \n",
"27 The Freedom State Bank Freedom \n",
"28 Valley Bank Fort Lauderdale \n",
"29 Valley Bank Moline \n",
".. ... ... \n",
"525 ANB Financial, NA Bentonville \n",
"526 Hume Bank Hume \n",
"527 Douglass National Bank Kansas City \n",
"528 Miami Valley Bank Lakeview \n",
"529 NetBank Alpharetta \n",
"530 Metropolitan Savings Bank Pittsburgh \n",
"531 Bank of Ephraim Ephraim \n",
"532 Reliance Bank White Plains \n",
"533 Guaranty National Bank of Tallahassee Tallahassee \n",
"534 Dollar Savings Bank Newark \n",
"535 Pulaski Savings Bank Philadelphia \n",
"536 First National Bank of Blanchardville Blanchardville \n",
"537 Southern Pacific Bank Torrance \n",
"538 Farmers Bank of Cheneyville Cheneyville \n",
"539 Bank of Alamo Alamo \n",
"540 AmTrade International Bank En Espanol Atlanta \n",
"541 Universal Federal Savings Bank Chicago \n",
"542 Connecticut Bank of Commerce Stamford \n",
"543 New Century Bank Shelby Township \n",
"544 Net 1st National Bank Boca Raton \n",
"545 NextBank, NA Phoenix \n",
"546 Oakwood Deposit Bank Co. Oakwood \n",
"547 Bank of Sierra Blanca Sierra Blanca \n",
"548 Hamilton Bank, NA En Espanol Miami \n",
"549 Sinclair National Bank Gravette \n",
"550 Superior Bank, FSB Hinsdale \n",
"551 Malta National Bank Malta \n",
"552 First Alliance Bank & Trust Co. Manchester \n",
"553 National State Bank of Metropolis Metropolis \n",
"554 Bank of Honolulu Honolulu \n",
"\n",
" ST CERT Acquiring Institution Closing Date \\\n",
"0 IL 30570 Royal Savings Bank December 15, 2017 \n",
"1 KS 17719 Conway Bank October 13, 2017 \n",
"2 IL 1802 United Fidelity Bank, fsb May 26, 2017 \n",
"3 WI 30003 First-Citizens Bank & Trust Company May 5, 2017 \n",
"4 LA 58302 Whitney Bank April 28, 2017 \n",
"5 UT 35495 Cache Valley Bank March 3, 2017 \n",
"6 IL 19328 State Bank of Texas January 27, 2017 \n",
"7 NJ 34951 First-Citizens Bank & Trust Company January 13, 2017 \n",
"8 AR 91 Today's Bank September 23, 2016 \n",
"9 GA 11297 United Bank August 19, 2016 \n",
"10 PA 35312 First-Citizens Bank & Trust Company May 6, 2016 \n",
"11 TN 9956 The Bank of Fayette County April 29, 2016 \n",
"12 WI 20364 First-Citizens Bank & Trust Company March 11, 2016 \n",
"13 WA 35156 Twin City Bank October 2, 2015 \n",
"14 GA 35259 Fidelity Bank October 2, 2015 \n",
"15 CO 34112 United Fidelity Bank, fsb July 10, 2015 \n",
"16 IL 57772 Republic Bank of Chicago May 8, 2015 \n",
"17 PR 32102 Banco Popular de Puerto Rico February 27, 2015 \n",
"18 GA 33938 First-Citizens Bank & Trust Company February 13, 2015 \n",
"19 IL 20290 United Fidelity Bank, fsb January 23, 2015 \n",
"20 FL 17557 First NBC Bank January 16, 2015 \n",
"21 MN 34983 BankVista December 19, 2014 \n",
"22 CA 34738 Bank of Southern California, N.A. November 7, 2014 \n",
"23 IL 916 State Bank of Texas October 24, 2014 \n",
"24 MD 4862 Howard Bank October 17, 2014 \n",
"25 IL 28462 Providence Bank, LLC July 25, 2014 \n",
"26 GA 58125 Community & Southern Bank July 18, 2014 \n",
"27 OK 12483 Alva State Bank & Trust Company June 27, 2014 \n",
"28 FL 21793 Landmark Bank, National Association June 20, 2014 \n",
"29 IL 10450 Great Southern Bank June 20, 2014 \n",
".. .. ... ... ... \n",
"525 AR 33901 Pulaski Bank and Trust Company May 9, 2008 \n",
"526 MO 1971 Security Bank March 7, 2008 \n",
"527 MO 24660 Liberty Bank and Trust Company January 25, 2008 \n",
"528 OH 16848 The Citizens Banking Company October 4, 2007 \n",
"529 GA 32575 ING DIRECT September 28, 2007 \n",
"530 PA 35353 Allegheny Valley Bank of Pittsburgh February 2, 2007 \n",
"531 UT 1249 Far West Bank June 25, 2004 \n",
"532 NY 26778 Union State Bank March 19, 2004 \n",
"533 FL 26838 Hancock Bank of Florida March 12, 2004 \n",
"534 NJ 31330 No Acquirer February 14, 2004 \n",
"535 PA 27203 Earthstar Bank November 14, 2003 \n",
"536 WI 11639 The Park Bank May 9, 2003 \n",
"537 CA 27094 Beal Bank February 7, 2003 \n",
"538 LA 16445 Sabine State Bank & Trust December 17, 2002 \n",
"539 TN 9961 No Acquirer November 8, 2002 \n",
"540 GA 33784 No Acquirer September 30, 2002 \n",
"541 IL 29355 Chicago Community Bank June 27, 2002 \n",
"542 CT 19183 Hudson United Bank June 26, 2002 \n",
"543 MI 34979 No Acquirer March 28, 2002 \n",
"544 FL 26652 Bank Leumi USA March 1, 2002 \n",
"545 AZ 22314 No Acquirer February 7, 2002 \n",
"546 OH 8966 The State Bank & Trust Company February 1, 2002 \n",
"547 TX 22002 The Security State Bank of Pecos January 18, 2002 \n",
"548 FL 24382 Israel Discount Bank of New York January 11, 2002 \n",
"549 AR 34248 Delta Trust & Bank September 7, 2001 \n",
"550 IL 32646 Superior Federal, FSB July 27, 2001 \n",
"551 OH 6629 North Valley Bank May 3, 2001 \n",
"552 NH 34264 Southern New Hampshire Bank & Trust February 2, 2001 \n",
"553 IL 3815 Banterra Bank of Marion December 14, 2000 \n",
"554 HI 21029 Bank of the Orient October 13, 2000 \n",
"\n",
" Updated Date \n",
"0 February 21, 2018 \n",
"1 February 21, 2018 \n",
"2 July 26, 2017 \n",
"3 March 22, 2018 \n",
"4 December 5, 2017 \n",
"5 March 7, 2018 \n",
"6 May 18, 2017 \n",
"7 May 18, 2017 \n",
"8 September 25, 2017 \n",
"9 June 1, 2017 \n",
"10 September 6, 2016 \n",
"11 September 6, 2016 \n",
"12 March 13, 2017 \n",
"13 February 19, 2018 \n",
"14 July 9, 2018 \n",
"15 February 20, 2018 \n",
"16 July 12, 2016 \n",
"17 May 13, 2015 \n",
"18 April 21, 2015 \n",
"19 November 15, 2017 \n",
"20 November 15, 2017 \n",
"21 January 3, 2018 \n",
"22 November 10, 2016 \n",
"23 January 6, 2016 \n",
"24 February 19, 2018 \n",
"25 December 12, 2016 \n",
"26 October 6, 2017 \n",
"27 February 21, 2018 \n",
"28 February 14, 2018 \n",
"29 June 26, 2015 \n",
".. ... \n",
"525 August 28, 2012 \n",
"526 August 28, 2012 \n",
"527 October 26, 2012 \n",
"528 September 12, 2016 \n",
"529 August 28, 2012 \n",
"530 October 27, 2010 \n",
"531 April 9, 2008 \n",
"532 April 9, 2008 \n",
"533 April 17, 2018 \n",
"534 April 9, 2008 \n",
"535 October 6, 2017 \n",
"536 June 5, 2012 \n",
"537 October 20, 2008 \n",
"538 October 20, 2004 \n",
"539 March 18, 2005 \n",
"540 September 11, 2006 \n",
"541 October 6, 2017 \n",
"542 February 14, 2012 \n",
"543 March 18, 2005 \n",
"544 April 9, 2008 \n",
"545 February 5, 2015 \n",
"546 October 25, 2012 \n",
"547 November 6, 2003 \n",
"548 September 21, 2015 \n",
"549 October 6, 2017 \n",
"550 August 19, 2014 \n",
"551 November 18, 2002 \n",
"552 February 18, 2003 \n",
"553 March 17, 2005 \n",
"554 March 17, 2005 \n",
"\n",
"[555 rows x 7 columns]"
]
},
"execution_count": 21,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df = pd.read_html('http://www.fdic.gov/bank/individual/failed/banklist.html')\n",
"df[0]"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [],
"source": [
"#!pip install plotly"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [],
"source": [
"#!pip install --upgrade pip"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {},
"outputs": [],
"source": [
"#!pip install cufflinks"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {},
"outputs": [],
"source": [
"#!pip install pandas-datareader"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>score</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>1</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>2</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>2</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>2</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td>3</td>\n",
" </tr>\n",
" <tr>\n",
" <th>7</th>\n",
" <td>3</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8</th>\n",
" <td>3</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" score\n",
"0 1\n",
"1 1\n",
"2 1\n",
"3 2\n",
"4 2\n",
"5 2\n",
"6 3\n",
"7 3\n",
"8 3"
]
},
"execution_count": 26,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Create data\n",
"data = {'score': [1,1,1,2,2,2,3,3,3]}\n",
"\n",
"# Create dataframe\n",
"df = pd.DataFrame(data)\n",
"\n",
"# View dataframe\n",
"df"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>score</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>1.5</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>2.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>2.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td>2.5</td>\n",
" </tr>\n",
" <tr>\n",
" <th>7</th>\n",
" <td>3.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8</th>\n",
" <td>3.0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" score\n",
"0 NaN\n",
"1 1.0\n",
"2 1.0\n",
"3 1.5\n",
"4 2.0\n",
"5 2.0\n",
"6 2.5\n",
"7 3.0\n",
"8 3.0"
]
},
"execution_count": 27,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Calculate the moving average. That is, take\n",
"# the first two values, average them, \n",
"# then drop the first and add the third, etc.\n",
"df.rolling(window=2).mean()"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[<matplotlib.lines.Line2D at 0x11c6f04e0>]"
]
},
"execution_count": 28,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import seaborn as sns\n",
"import matplotlib.pyplot as plt\n",
"sns.set()\n",
"plt.plot(df.index, df.score)\n",
"plt.plot(df.index, df.rolling(window=2).mean())"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>x</th>\n",
" <th>y</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>1.000000</td>\n",
" <td>10.820262</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>1.191919</td>\n",
" <td>4.384624</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>1.383838</td>\n",
" <td>7.661367</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>1.575758</td>\n",
" <td>14.355981</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>1.767677</td>\n",
" <td>12.873143</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" x y\n",
"0 1.000000 10.820262\n",
"1 1.191919 4.384624\n",
"2 1.383838 7.661367\n",
"3 1.575758 14.355981\n",
"4 1.767677 12.873143"
]
},
"execution_count": 29,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"np.random.seed(0)\n",
"x = np.linspace(1,20,100)\n",
"y = 2 * x + 5 * np.random.randn(100)\n",
"df = pd.DataFrame({'x':x, 'y':y})\n",
"df.head()"
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[<matplotlib.lines.Line2D at 0x1a1ed57320>]"
]
},
"execution_count": 30,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x1a1ed57518>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.plot(df.x, df.y)\n",
"plt.plot(df.x, df.y.rolling(window=5).mean())"
]
},
{
"cell_type": "code",
"execution_count": 31,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[<matplotlib.lines.Line2D at 0x11c6f0208>]"
]
},
"execution_count": 31,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x11c6f0518>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.plot(df.x, df.y)\n",
"plt.plot(df.x, df.y.rolling(window=10).mean())"
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[<matplotlib.lines.Line2D at 0x11c66c668>]"
]
},
"execution_count": 32,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x1a1ed62ef0>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.plot(df.x, df.y - df.y.rolling(window=10).mean())"
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"0 NaN\n",
"1 NaN\n",
"2 NaN\n",
"3 NaN\n",
"4 NaN\n",
"5 NaN\n",
"6 NaN\n",
"7 NaN\n",
"8 NaN\n",
"9 0.090149\n",
"10 -0.432621\n",
"11 5.591471\n",
"12 2.234142\n",
"13 0.096937\n",
"14 2.419726\n",
"15 1.213305\n",
"16 6.743334\n",
"17 -1.725953\n",
"18 0.657034\n",
"19 -4.546436\n",
"20 -11.692390\n",
"21 4.740980\n",
"22 5.743369\n",
"23 -1.857718\n",
"24 12.288935\n",
"25 -5.437646\n",
"26 2.787135\n",
"27 1.613436\n",
"28 9.603395\n",
"29 8.124566\n",
" ... \n",
"70 7.581233\n",
"71 4.336427\n",
"72 8.412242\n",
"73 -3.704619\n",
"74 4.368761\n",
"75 -0.925483\n",
"76 -2.235119\n",
"77 -0.254566\n",
"78 0.784047\n",
"79 2.620526\n",
"80 -2.538929\n",
"81 7.405031\n",
"82 5.566080\n",
"83 -4.292742\n",
"84 10.286782\n",
"85 11.034617\n",
"86 6.424281\n",
"87 -0.568704\n",
"88 -4.643242\n",
"89 5.483636\n",
"90 -2.185494\n",
"91 5.781807\n",
"92 0.839650\n",
"93 3.425029\n",
"94 0.889609\n",
"95 3.235301\n",
"96 0.339075\n",
"97 8.233030\n",
"98 -0.660595\n",
"99 1.041023\n",
"Name: y, Length: 100, dtype: float64"
]
},
"execution_count": 33,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"w = df.y - df.y.rolling(window=10).mean()\n",
"w"
]
},
{
"cell_type": "code",
"execution_count": 34,
"metadata": {},
"outputs": [],
"source": [
"w = w[9:]"
]
},
{
"cell_type": "code",
"execution_count": 35,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x11c6633c8>"
]
},
"execution_count": 35,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x11c6babe0>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"sns.distplot(w)"
]
},
{
"cell_type": "code",
"execution_count": 36,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.axes._subplots.AxesSubplot at 0x11c6557f0>"
]
},
"execution_count": 36,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x11c52d7b8>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"sns.boxplot(w)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Outlier removal in Python using IQR rule\n",
" - http://stamfordresearch.com/outlier-removal-in-python-using-iqr-rule/"
]
},
{
"cell_type": "code",
"execution_count": 37,
"metadata": {},
"outputs": [],
"source": [
"q75, q25 = np.percentile(w, [75 ,25])\n",
"iqr = q75 - q25\n",
" \n",
"mini = q25 - (iqr*1.5)\n",
"maxi = q75 + (iqr*1.5)"
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(5.157466166849959, -0.7073976359559406)"
]
},
"execution_count": 38,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"q75, q25"
]
},
{
"cell_type": "code",
"execution_count": 39,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"5.8648638028059"
]
},
"execution_count": 39,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"iqr"
]
},
{
"cell_type": "code",
"execution_count": 40,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"-9.50469334016479"
]
},
"execution_count": 40,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"mini"
]
},
{
"cell_type": "code",
"execution_count": 41,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"13.954761871058809"
]
},
"execution_count": 41,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"maxi"
]
},
{
"cell_type": "code",
"execution_count": 42,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.lines.Line2D at 0x11c5e48d0>"
]
},
"execution_count": 42,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x11c607b38>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"sns.boxplot(w, whis=1.5)\n",
"plt.axvline(x=mini, c='r')\n",
"plt.axvline(x=maxi, c='r')\n",
"\n",
"plt.axvline(x=q75, c='r')\n",
"plt.axvline(x=q25, c='r')\n",
"\n",
"plt.axvline(x=w.median(), c='r')"
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {},
"outputs": [],
"source": [
"wr = w[(w>mini) & (w<maxi)]"
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {},
"outputs": [],
"source": [
"q75, q25 = np.percentile(wr, [75 ,25])\n",
"iqr = q75 - q25\n",
" \n",
"mini = q25 - (iqr*1.5)\n",
"maxi = q75 + (iqr*1.5)"
]
},
{
"cell_type": "code",
"execution_count": 45,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.lines.Line2D at 0x11c41a240>"
]
},
"execution_count": 45,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x11c5e4e10>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"sns.boxplot(w, whis=1.5)\n",
"plt.axvline(x=mini, c='r')\n",
"plt.axvline(x=maxi, c='r')\n",
"\n",
"plt.axvline(x=q75, c='r')\n",
"plt.axvline(x=q25, c='r')\n",
"\n",
"plt.axvline(x=wr.median(), c='r')"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"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": "3.6.4"
}
},
"nbformat": 4,
"nbformat_minor": 2
}