Skip to content

Instantly share code, notes, and snippets.

@ClebsonDantasUchoa

ClebsonDantasUchoa/treineirosENEM.ipynb

Created September 8, 2018 15:06
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save ClebsonDantasUchoa/efbbcb3e9e742ba08a06175b17ec9dc6 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save ClebsonDantasUchoa/efbbcb3e9e742ba08a06175b17ec9dc6 to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
treineirosENEM.ipynb
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Predizer se um candidado do ENEM fez a prova como treineiro apartir das suas notas na prova"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Importação das bibliotecas"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"%matplotlib inline\n",
"import pandas as pd\n",
"import numpy as np\n",
"from sklearn import linear_model\n",
"from sklearn import metrics\n",
"import matplotlib.pyplot as plt\n",
"import seaborn as sns\n",
"from sklearn import tree"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## leitura dos dados"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"(13730, 167)\n"
]
},
{
"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>Unnamed: 0</th>\n",
" <th>NU_INSCRICAO</th>\n",
" <th>NU_ANO</th>\n",
" <th>CO_MUNICIPIO_RESIDENCIA</th>\n",
" <th>NO_MUNICIPIO_RESIDENCIA</th>\n",
" <th>CO_UF_RESIDENCIA</th>\n",
" <th>SG_UF_RESIDENCIA</th>\n",
" <th>NU_IDADE</th>\n",
" <th>TP_SEXO</th>\n",
" <th>TP_ESTADO_CIVIL</th>\n",
" <th>...</th>\n",
" <th>Q041</th>\n",
" <th>Q042</th>\n",
" <th>Q043</th>\n",
" <th>Q044</th>\n",
" <th>Q045</th>\n",
" <th>Q046</th>\n",
" <th>Q047</th>\n",
" <th>Q048</th>\n",
" <th>Q049</th>\n",
" <th>Q050</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>1</td>\n",
" <td>ed50e8aaa58e7a806c337585efee9ca41f1eb1ad</td>\n",
" <td>2016</td>\n",
" <td>4314902</td>\n",
" <td>Porto Alegre</td>\n",
" <td>43</td>\n",
" <td>RS</td>\n",
" <td>24</td>\n",
" <td>M</td>\n",
" <td>0.0</td>\n",
" <td>...</td>\n",
" <td>5.0</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>B</td>\n",
" <td>D</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>2</td>\n",
" <td>2c3acac4b33ec2b195d77e7c04a2d75727fad723</td>\n",
" <td>2016</td>\n",
" <td>2304707</td>\n",
" <td>Granja</td>\n",
" <td>23</td>\n",
" <td>CE</td>\n",
" <td>17</td>\n",
" <td>F</td>\n",
" <td>0.0</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>C</td>\n",
" <td>A</td>\n",
" <td>B</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>C</td>\n",
" <td>A</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>3</td>\n",
" <td>f4545f8ccb9ff5c8aad7d32951b3f251a26e6568</td>\n",
" <td>2016</td>\n",
" <td>2304400</td>\n",
" <td>Fortaleza</td>\n",
" <td>23</td>\n",
" <td>CE</td>\n",
" <td>21</td>\n",
" <td>F</td>\n",
" <td>0.0</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>C</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>B</td>\n",
" <td>A</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>4</td>\n",
" <td>3d6ec248fef899c414e77f82d5c6d2bffbeaf7fe</td>\n",
" <td>2016</td>\n",
" <td>3304557</td>\n",
" <td>Rio de Janeiro</td>\n",
" <td>33</td>\n",
" <td>RJ</td>\n",
" <td>25</td>\n",
" <td>F</td>\n",
" <td>0.0</td>\n",
" <td>...</td>\n",
" <td>5.0</td>\n",
" <td>C</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>D</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>5</td>\n",
" <td>bf896ac8d3ecadd6dba1dfbf50110afcbf5d3268</td>\n",
" <td>2016</td>\n",
" <td>1302603</td>\n",
" <td>Manaus</td>\n",
" <td>13</td>\n",
" <td>AM</td>\n",
" <td>28</td>\n",
" <td>M</td>\n",
" <td>0.0</td>\n",
" <td>...</td>\n",
" <td>NaN</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" <td>A</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>5 rows × 167 columns</p>\n",
"</div>"
],
"text/plain": [
" Unnamed: 0 NU_INSCRICAO NU_ANO \\\n",
"0 1 ed50e8aaa58e7a806c337585efee9ca41f1eb1ad 2016 \n",
"1 2 2c3acac4b33ec2b195d77e7c04a2d75727fad723 2016 \n",
"2 3 f4545f8ccb9ff5c8aad7d32951b3f251a26e6568 2016 \n",
"3 4 3d6ec248fef899c414e77f82d5c6d2bffbeaf7fe 2016 \n",
"4 5 bf896ac8d3ecadd6dba1dfbf50110afcbf5d3268 2016 \n",
"\n",
" CO_MUNICIPIO_RESIDENCIA NO_MUNICIPIO_RESIDENCIA CO_UF_RESIDENCIA \\\n",
"0 4314902 Porto Alegre 43 \n",
"1 2304707 Granja 23 \n",
"2 2304400 Fortaleza 23 \n",
"3 3304557 Rio de Janeiro 33 \n",
"4 1302603 Manaus 13 \n",
"\n",
" SG_UF_RESIDENCIA NU_IDADE TP_SEXO TP_ESTADO_CIVIL ... Q041 Q042 Q043 \\\n",
"0 RS 24 M 0.0 ... 5.0 A A \n",
"1 CE 17 F 0.0 ... NaN A A \n",
"2 CE 21 F 0.0 ... NaN A A \n",
"3 RJ 25 F 0.0 ... 5.0 C A \n",
"4 AM 28 M 0.0 ... NaN A A \n",
"\n",
" Q044 Q045 Q046 Q047 Q048 Q049 Q050 \n",
"0 A A A A A B D \n",
"1 C A B A A C A \n",
"2 A A C A A B A \n",
"3 A A A D A A A \n",
"4 A A A A A A A \n",
"\n",
"[5 rows x 167 columns]"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"data = pd.read_csv(\"enem.csv\")\n",
"print(data.shape)\n",
"data.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Quantidade de treineiros"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"0\n",
"-------------------\n",
"IN_TREINEIRO\n",
"0 11947\n",
"1 1783\n",
"dtype: int64\n"
]
}
],
"source": [
"print(data['IN_TREINEIRO'].isnull().sum())\n",
"print('-------------------')\n",
"print(data.groupby('IN_TREINEIRO').size())"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Filtrando dados importantes"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"(13730, 6)\n",
"---------------\n"
]
},
{
"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>NU_NOTA_CN</th>\n",
" <th>NU_NOTA_CH</th>\n",
" <th>NU_NOTA_LC</th>\n",
" <th>NU_NOTA_REDACAO</th>\n",
" <th>NU_NOTA_MT</th>\n",
" <th>IN_TREINEIRO</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>436.3</td>\n",
" <td>495.4</td>\n",
" <td>581.2</td>\n",
" <td>520.0</td>\n",
" <td>399.4</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>474.5</td>\n",
" <td>544.1</td>\n",
" <td>599.0</td>\n",
" <td>580.0</td>\n",
" <td>459.8</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" NU_NOTA_CN NU_NOTA_CH NU_NOTA_LC NU_NOTA_REDACAO NU_NOTA_MT \\\n",
"0 436.3 495.4 581.2 520.0 399.4 \n",
"1 474.5 544.1 599.0 580.0 459.8 \n",
"2 NaN NaN NaN NaN NaN \n",
"3 NaN NaN NaN NaN NaN \n",
"4 NaN NaN NaN NaN NaN \n",
"\n",
" IN_TREINEIRO \n",
"0 0 \n",
"1 0 \n",
"2 0 \n",
"3 0 \n",
"4 0 "
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dados = pd.DataFrame()\n",
"dados['NU_NOTA_CN'] = data['NU_NOTA_CN']\n",
"dados['NU_NOTA_CH'] = data['NU_NOTA_CH']\n",
"dados['NU_NOTA_LC'] = data['NU_NOTA_LC']\n",
"dados['NU_NOTA_REDACAO'] = data['NU_NOTA_REDACAO']\n",
"dados['NU_NOTA_MT'] = data['NU_NOTA_MT']\n",
"dados['IN_TREINEIRO'] = data['IN_TREINEIRO']\n",
"print(dados.shape)\n",
"print(\"---------------\")\n",
"dados.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Visualização dos dados"
]
},
{
"cell_type": "code",
"execution_count": 5,
"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>NU_NOTA_CN</th>\n",
" <th>NU_NOTA_CH</th>\n",
" <th>NU_NOTA_LC</th>\n",
" <th>NU_NOTA_REDACAO</th>\n",
" <th>NU_NOTA_MT</th>\n",
" <th>IN_TREINEIRO</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>count</th>\n",
" <td>10341.000000</td>\n",
" <td>10341.000000</td>\n",
" <td>10133.000000</td>\n",
" <td>10133.000000</td>\n",
" <td>10133.000000</td>\n",
" <td>13730.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>mean</th>\n",
" <td>473.495155</td>\n",
" <td>529.661087</td>\n",
" <td>516.472841</td>\n",
" <td>529.048258</td>\n",
" <td>482.497928</td>\n",
" <td>0.129862</td>\n",
" </tr>\n",
" <tr>\n",
" <th>std</th>\n",
" <td>71.093674</td>\n",
" <td>73.726344</td>\n",
" <td>68.688190</td>\n",
" <td>154.294758</td>\n",
" <td>99.826323</td>\n",
" <td>0.336163</td>\n",
" </tr>\n",
" <tr>\n",
" <th>min</th>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" <td>0.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25%</th>\n",
" <td>419.900000</td>\n",
" <td>480.400000</td>\n",
" <td>468.100000</td>\n",
" <td>440.000000</td>\n",
" <td>408.900000</td>\n",
" <td>0.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>50%</th>\n",
" <td>459.800000</td>\n",
" <td>532.000000</td>\n",
" <td>520.900000</td>\n",
" <td>540.000000</td>\n",
" <td>461.200000</td>\n",
" <td>0.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>75%</th>\n",
" <td>514.500000</td>\n",
" <td>581.200000</td>\n",
" <td>564.900000</td>\n",
" <td>600.000000</td>\n",
" <td>537.600000</td>\n",
" <td>0.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>max</th>\n",
" <td>806.400000</td>\n",
" <td>807.000000</td>\n",
" <td>763.600000</td>\n",
" <td>1000.000000</td>\n",
" <td>952.000000</td>\n",
" <td>1.000000</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" NU_NOTA_CN NU_NOTA_CH NU_NOTA_LC NU_NOTA_REDACAO \\\n",
"count 10341.000000 10341.000000 10133.000000 10133.000000 \n",
"mean 473.495155 529.661087 516.472841 529.048258 \n",
"std 71.093674 73.726344 68.688190 154.294758 \n",
"min 0.000000 0.000000 0.000000 0.000000 \n",
"25% 419.900000 480.400000 468.100000 440.000000 \n",
"50% 459.800000 532.000000 520.900000 540.000000 \n",
"75% 514.500000 581.200000 564.900000 600.000000 \n",
"max 806.400000 807.000000 763.600000 1000.000000 \n",
"\n",
" NU_NOTA_MT IN_TREINEIRO \n",
"count 10133.000000 13730.000000 \n",
"mean 482.497928 0.129862 \n",
"std 99.826323 0.336163 \n",
"min 0.000000 0.000000 \n",
"25% 408.900000 0.000000 \n",
"50% 461.200000 0.000000 \n",
"75% 537.600000 0.000000 \n",
"max 952.000000 1.000000 "
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dados.describe()"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 2 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"corr = dados.corr()\n",
"sns.heatmap(corr, xticklabels=dados.columns, yticklabels=dados.columns, cmap='RdBu');"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1080x1080 with 6 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"dados.hist(figsize=(15,15), bins=30);"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Removendo linhas com dados faltantes"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"(10097, 6)\n"
]
},
{
"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>NU_NOTA_CN</th>\n",
" <th>NU_NOTA_CH</th>\n",
" <th>NU_NOTA_LC</th>\n",
" <th>NU_NOTA_REDACAO</th>\n",
" <th>NU_NOTA_MT</th>\n",
" <th>IN_TREINEIRO</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>436.3</td>\n",
" <td>495.4</td>\n",
" <td>581.2</td>\n",
" <td>520.0</td>\n",
" <td>399.4</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>474.5</td>\n",
" <td>544.1</td>\n",
" <td>599.0</td>\n",
" <td>580.0</td>\n",
" <td>459.8</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>439.7</td>\n",
" <td>583.2</td>\n",
" <td>410.9</td>\n",
" <td>620.0</td>\n",
" <td>364.5</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td>420.1</td>\n",
" <td>604.2</td>\n",
" <td>484.5</td>\n",
" <td>560.0</td>\n",
" <td>529.2</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>7</th>\n",
" <td>619.6</td>\n",
" <td>625.8</td>\n",
" <td>611.2</td>\n",
" <td>620.0</td>\n",
" <td>566.7</td>\n",
" <td>0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" NU_NOTA_CN NU_NOTA_CH NU_NOTA_LC NU_NOTA_REDACAO NU_NOTA_MT \\\n",
"0 436.3 495.4 581.2 520.0 399.4 \n",
"1 474.5 544.1 599.0 580.0 459.8 \n",
"5 439.7 583.2 410.9 620.0 364.5 \n",
"6 420.1 604.2 484.5 560.0 529.2 \n",
"7 619.6 625.8 611.2 620.0 566.7 \n",
"\n",
" IN_TREINEIRO \n",
"0 0 \n",
"1 0 \n",
"5 0 \n",
"6 0 \n",
"7 0 "
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dados = dados.dropna()\n",
"print(dados.shape)\n",
"dados.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Separando em dados de treino e dados de teste\n",
"### Usaremos 75% dos valores para treino e 25% para teste"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"(7572, 5) (2525, 5) (7572,) (2525,)\n"
]
}
],
"source": [
"values = dados.values\n",
"np.random.seed(1)\n",
"np.random.shuffle(values)\n",
"x_train = values[ :int(10097*0.75), 0:5]\n",
"x_test = values[ int(10097*0.75):, 0:5]\n",
"y_train = values[ :int(10097*0.75), 5]\n",
"y_test = values[ int(10097*0.75):, 5]\n",
"print(x_train.shape, x_test.shape, y_train.shape, y_test.shape)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Criação do modelos de predição\n",
"### Criação do modelo de Regressão Logistica"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"0.8598019801980198"
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"lr = linear_model.LogisticRegression()\n",
"lr.fit(x_train, y_train)\n",
"lr_pred = lr.predict(x_test)\n",
"metrics.accuracy_score(y_test, lr_pred)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Criação do modelo de árvore de classificação"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"0.8598019801980198"
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"clf = tree.DecisionTreeClassifier()\n",
"clf.fit(x_train, y_train)\n",
"clf_pred = lr.predict(x_test)\n",
"metrics.accuracy_score(y_test, clf_pred)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Geração dos gráficos de comparação entre os modelos"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
"predicoes = pd.DataFrame()\n",
"predicoes['Regressao Linear'] = lr_pred\n",
"predicoes['Arvore de decisao'] = clf_pred"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" precision recall f1-score support\n",
"\n",
" 0.0 0.86 1.00 0.92 2172\n",
" 1.0 0.00 0.00 0.00 353\n",
"\n",
"avg / total 0.74 0.86 0.80 2525\n",
"\n"
]
}
],
"source": [
"print(metrics.classification_report(y_test, lr_pred))"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" precision recall f1-score support\n",
"\n",
" 0.0 0.86 1.00 0.92 2172\n",
" 1.0 0.00 0.00 0.00 353\n",
"\n",
"avg / total 0.74 0.86 0.80 2525\n",
"\n"
]
}
],
"source": [
"print(metrics.classification_report(y_test, clf_pred))"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.plot(lr_pred, clf_pred);"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAELCAYAAAA86fb3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3Xt4FOXd//H3l5OcGqEgigLRCqK2pYighMgpKOUkIHiAX9XaSkF9qIVWqdZWLU+fVmutXoB4qLXSXvXxBJQoqKgoInjisQoijZwqUFAIKiFyCEm+vz9mNlmWHDbJJoHx87quXMnO3HPPPZvZz87O3HuPuTsiIhJNDeq7ASIiUnsU8iIiEaaQFxGJMIW8iEiEKeRFRCJMIS8iEmEKeRGRCFPIi4hEmEJeRCTCGtXXitPS0rxdu3b1tXoRkaPShg0bct39uGTL11vIt2vXjrvvvru+Vi8iclQaPXr0x1Upr9M1IiIRppAXEYkwhbyISITV2zl5EUlegwYNaNu2LU2aNKnvpkgdKSgoIDc3l+Li4hrVo5AXOQq0bduWE088kbS0NMysvpsjtczdycvLA2DHjh01qqvS0zVm9oiZ7TCzD8qZb2Y2w8zWm9kqM+tRoxaJyGGaNGmigP8KMTPS0tJS8sktmXPyjwJDKpg/FOgS/kwE7q9xq0TkMAr4r5ZU/b8rDXl3fw34rIIio4C/euBNoJWZtU9J60REpEZScU7+JGBL3OOt4bTtiQXNbCLB0T7HHZf0F7ZEJEH3O5antL73bsqstEybNm0488wzKSwsJD09nQcffJBjjz02pe2oL9dddx3f/e53GTVqVMm07du3c9NNNzFnzpx6bFnNpaILZVmfKcq8O7i7P+TuPd29Z1paWgpWLSJ1pVmzZixbtow33niD1q1b86c//Skl9RYWFqaknlRr3759rQd8XWx7KkJ+K9Ax7nEHYFsK6hWRI1SvXr3Yvr30w/qMGTPIysoiMzOT3/3udyXT77rrLs455xwuuugirr76ambOnAnAiBEjmD59OsOHD+eBBx4gNzeXK6+8kqysLLKysnjzzTcBWL58OX379qVv377069ePPXv28MknnzBs2DD69u1LRkYGK1asAOCnP/0pAwcOJCMj45A2LF26lH79+tGnTx8mT57MgQMHktrGzZs3k5GRAcBjjz3GFVdcwcUXX8zZZ5/NrbfeWlJuyZIlDB48mP79+3PVVVeRn58PwO9//3uysrLIyMhgypQpuHuZ217bUnG6JhuYbGaPA+cCu939sFM1IhINRUVFvPbaa1x++eVAEHIbN27k5Zdfxt0ZP348y5cvp3nz5mRnZ7N06VIKCwsZMGAA3bt3L6ln9+7dLFy4EIAJEyZw7bXXkpGRwZYtW7j44ot56623mDlzJnfddRe9e/cmPz+fpk2bMmfOHLKysrjhhhsoKipi7969APzqV7+idevWFBUVMWrUKD744AM6d+7Mddddx4IFC+jcuTPXXHMNjzzyCNdee22Vt3v16tUsXbqUY445hl69ejFx4kSaNWvGH/7wB+bPn0+LFi249957mT17NtOmTeNHP/oR06ZNA2DSpEk8//zzDB069LBtr22VhryZ/S8wAGhrZluB24DGAO7+ALAIGAasB/YCP6itxopI/dm3bx99+/Zl8+bNdO/enYEDBwLwyiuvsGTJEvr16wfAl19+ycaNG8nPz2fYsGE0a9YMgCFDDu2kN2bMmJK/ly5dSk5OTsnjPXv2sGfPHs4991x++ctfcskllzBixAhatmzJWWedxY9//GMKCwsZPnw43/72twGYP38+c+bMobCwkE8//ZScnBzcnfT0dDp37gzA+PHjefjhh6sV8v379y+5BtG1a1e2bNnC7t27ycnJKdm2gwcP0qtXLwCWLVvGjBkz2Lt3L1988QWnn356ScjHb3ttqzTk3X18JfMd+K+UtUhEjkixc/K7d+9m3LhxPPzww0yaNAl3Z+rUqfzgB4ce382ePbvC+po3b17yd3FxMYsXLy55Q4iZOnUqgwcP5sUXX2Tw4MHMnz+fzMxMFi5cyOLFi5k0aRLXX389GRkZzJo1iyVLltCqVSuuu+66pE/LJCu+z3rDhg0pKirC3RkwYAB//vOfDym7f/9+brjhBpYsWUKHDh244447DmlP/LbXNo1dIyJVcuyxx3LnnXcyc+ZMDh48SFZWFn//+99LzkVv27aNnTt30rt3b55//nn2799Pfn4+ixcvLrfOgQMHHnIhd/Xq1QBs2rSJb37zm0yZMoXu3buzbt06Nm/ezHHHHcf3v/99rrjiCt5//33y8vJo3rw5aWlp7Nixg5deegmALl26sHnzZjZu3AjAE088QWZm5T2JktWrVy/eeuutkvr37t3L+vXrSwK9TZs25Ofns2DBgpSts6o0rIHIUSiZLo+1qVu3bnzrW99i7ty5jBs3jo8++ojBgwcD0LJlSx588EF69OjB0KFD6du3Lx07dqR79+6U16vuzjvv5MYbbyQzM5OioiIyMjK45557uP/++1m2bBkNGzaka9eunH/++cybN48ZM2bQuHFjWrRowQMPPEB6ejrdunUjIyOD9PR0zj33XACaNm3Kfffdx1VXXUVhYSE9evQ47BNHzNSpU/nFL34BwEknncTDDz9c6fPQtm1bZs+ezYQJE0qC/ZZbbqFz585ceeWVZGZm0qlTJ3r0qL+BACx2xbeude7c2XXTEJHkdOjQgVNPPbW+m1Fl+fn5tGzZkr179zJ8+HDuvfdevvOd79R3s44aGzZsYOvWrYdMGz169P+5e89k69CRvIjUmilTppCTk8OBAwcYN26cAr4eKORFpNYkc8pDapcuvIqIRJhCXkQkwhTyIiIRppAXEYkwXXgVOQq1ujc9pfV9MeXjpMo9++yzXHHFFbz11lucdtppKW1DTbz++uvMnDmTJ554olrLlzXUcDKOhuGIdSQvIkmbO3cuvXv3Zt68eWXOLyoqqlH9NV2+rtXFcMQ1pZAXkaTk5+eXjAwZH/Kvv/46F154IRMmTCAzM5PbbrvtkK6Td9xxB7NmzcLd+dWvfkVGRgZ9+vQpqSNxeQiGHxg0aBB9+/ZlypQpZYb/Sy+9xDnnnMOQIUN45plnSqZ/+eWXTJ48maysLPr168eiRYsOW9bdufHGG+nduzeXXnopO3fuLJn33nvvMXz4cAYMGMDYsWP55JNPANi4cSOjR4/mvPPOo3///mzatOmQ4YjXrl1b0ubMzEw2bNgAwPe+9z0GDBhARkYGjz76aMl6nn76afr06UNGRga33XZblf8fydLpGhFJysKFCxk0aBCdO3emVatWvP/++yVfbnr33XdZsWIF6enprFq1iptvvpkJEyYA8I9//IOnnnqKZ555hg8++IDXX3+dXbt2kZWVRZ8+fQ5bPicnh/nz5/P888/TuHFjfvazn/HUU08xbty4krbs37+fn/zkJ2RnZ/ONb3yDH/7whyXz7r77bvr27cusWbPYvXs3gwYNon///rRo0aKkzLPPPsv69etZvnw5O3bsoHfv3lx++eUcPHiQadOm8dhjj9G2bVvmzZvHb37zG2bNmsXEiROZMmUKI0aMYP/+/RQXF5Obm1tS51/+8hcmTZrEpZdeSkFBQckb06xZs2jdujX79u0jKyuLkSNHcuDAAW6//XZeffVVWrVqxZgxY1i4cCHDhw9P+f9NIS8iSZk7d27JEL1jx47l6aefLgn5Hj16kJ4eXCfo1q0bubm5bN++ndzcXI499lg6duzI/fffz9ixY2nYsCHt2rUjMzOTd999l7S0tEOWX7p0Ke+//z5ZWVlAEOiJtwtdt24d6enpJUM9XHLJJSWnTV555RWee+45Zs2aVbL81q1b6dq1a8nyK1asKGlL+/btS4ZJXrduHf/617+46KKLgOD00QknnMCePXvYvn07I0aMAIIxcRL16tWLP/7xj2zbto0LL7ywpG0PPvggzz77LAD/+c9/2LBhAzt37uS8886jbdu2Je1fsWKFQl5E6sdnn33GsmXLWLt2LWZGUVERZsb06dOBw4fOHTlyJNnZ2Xz66aeMHTsWgIrGyUpcfty4cZWewjAr686jwXr++te/0qVLlyov7+6cfvrph42YmZeXV2FdEAR1z549Wbx4MWPHjmXGjBk0aNCAV199lcWLF9O8eXNGjBjBgQMHKnwuUk3n5EWkUgsWLOCyyy5j9erVrFq1ijVr1pCens4bb7xRZvkxY8Ywd+5csrOzGTlyJAB9+vRh/vz5FBUVkZuby4oVKzj77LMPW7Zfv35kZ2eXnCf//PPP2bx58yFlunTpwscff8ymTZuA4FNGTFZWFg899FBJkK5ateqwdcSuCRQVFfHJJ5+wbNmyknpzc3N5++23geAmIGvXriUtLY0TTzyx5G5OBw4cKLkjVcy///1vTj75ZCZNmsTQoUNZs2YNeXl5tGrViubNm/PRRx+xcuVKAM4++2yWL1/Orl27KCoqYu7cuSkdAjmejuRFjkLJdnlMlblz5zJlypRDpo0cOZKnn366zLscnXHGGeTn59O+fXtOOOEEILi36dtvv815552HmfHrX/+a448/nnXr1h2y7Omnn84tt9zCmDFjKC4upnHjxtx111106tSppEzTpk259957ueyyy/j6179O7969Wbt2LQA33ngjN998M5mZmbg7nTp1Oqxr5YgRI3jttdfIzMzk1FNPLQnYJk2aMGfOHH7+85+Tl5dHUVER11xzDWeccQYPPPAAU6dO5be//S2NGzfm0UcfpUGD0uPkefPm8dRTT9GoUSOOP/54pk2bRvPmzXnkkUfIzMykS5cu9OwZDB55wgkncOutt3LhhRfi7lxwwQUMGzasuv+eCmmoYZGjwNE61LDUTCqGGtbpGhGRCFPIi4hEmEJe5ChRX6dWpX6k6v+tkBc5ChQUFJCXl6eg/4pwd/Ly8igoKKhxXepdI3IUiH2zMv4blhJtBQUFKfl/K+RFjgLFxcXs2LGjvpshRyGdrhERiTCFvIhIhCnkRUQiTCEvIhJhCnkRkQhTyIuIRJhCXkQkwhTyIiIRllTIm9kQM8sxs/VmdlMZ8zuZ2Stm9k8zW2VmtTMwsoiIVEmlIW9mDYH7gKHAmcB4MzszodgvgSfd/SxgHDA71Q0VEZGqS+ZI/hxgvbtvdPcC4HFgVEIZB9LCv48FtqWuiSIiUl3JjF1zErAl7vFW4NyEMrcDi83sx0AL4PyUtE5ERGokmSP5sm6Jnjje6XjgUXfvAAwD/mZmh9VtZhPNbKWZrUzm7uciIlIzyYT8VqBj3OMOHH465mrgSQB3fwNoCrRNrMjdH3L3nu7eMy0tLXG2iIikWDIh/w7QxcxOMbMmBBdWsxPKbAYGAZjZGQQhvzOVDRURkaqrNOTdvRCYDLwArCXoRbPGzKab2ciw2M+AH5nZ+8D/Ale5bmEjIlLvkrppiLsvAhYlTLs17u8PgczUNk1ERGpK33gVEYkwhbyISIQp5EVEIkwhLyISYQp5EZEIU8iLiESYQl5EJMIU8iIiEaaQFxGJMIW8iEiEKeRFRCJMIS8iEmEKeRGRCFPIi4hEmEJeRCTCFPIiIhGmkBcRiTCFvIhIhCnkRUQiTCEvIhJhCnkRkQhTyIuIRJhCXkQkwhTyIiIRppAXEYkwhbyISIQp5EVEIkwhLyISYQp5EZEIU8iLiESYQl5EJMIU8iIiEaaQFxGJsKRC3syGmFmOma03s5vKKXOpmX1oZmvM7LHUNlNERKqjUWUFzKwhcB9wAbAVeMfMst39w7gyXYCbgUx3/9zM2tVWg0VEJHnJHMmfA6x3943uXgA8DoxKKPMj4D53/xzA3XektpkiIlIdyYT8ScCWuMdbw2nxTgNOM7PlZvammQ0pqyIzm2hmK81sZV5eXvVaLCIiSUsm5K2MaZ7wuBHQBRgAjAceNrNWhy3k/pC793T3nmlpaVVtq4iIVFEyIb8V6Bj3uAOwrYwyC9z9oLtvAnIIQl9EROpRMiH/DtDFzE4xsybAOCA7ocw/gIEAZtaW4PTNxlQ2VEREqq7SkHf3QmAy8AKwFnjS3deY2XQzGxkWewHYZWYfAq8AN7r7rtpqtIiIJKfSLpQA7r4IWJQw7da4vx34afgjIiJHCH3jVUQkwhTyIiIRppAXEYkwhbyISIQp5EVEIkwhLyISYQp5EZEIU8iLiESYQl5EJMIU8iIiEaaQFxGJMIW8iEiEKeRFRCJMIS8iEmEKeRGRCFPIi4hEmEJeRCTCFPIiIhGmkBcRiTCFvIhIhCnkRUQiTCEvIhJhCnkRkQhTyIuIRJhCXkQkwhTyIiIRppAXEYkwhbyISIQp5EVEIkwhLyISYQp5EZEIU8iLiERYUiFvZkPMLMfM1pvZTRWUu9jM3Mx6pq6JIiJSXZWGvJk1BO4DhgJnAuPN7Mwyyn0NuB54K9WNFBGR6knmSP4cYL27b3T3AuBxYFQZ5f4b+D2wP4XtExGRGkgm5E8CtsQ93hpOK2FmZwEd3f3Ziioys4lmttLMVubl5VW5sSIiUjXJhLyVMc1LZpo1AO4BflZZRe7+kLv3dPeeaWlpybdSRESqJZmQ3wp0jHvcAdgW9/hrwLeAV83s30BvIFsXX0VE6l8yIf8O0MXMTjGzJsA4IDs20913u3tbdz/Z3U8G3gRGuvvKWmmxiIgkrdKQd/dCYDLwArAWeNLd15jZdDMbWdsNFBGR6muUTCF3XwQsSph2azllB9S8WSIikgr6xquISIQp5EVEIkwhLyISYQp5EZEIU8iLiESYQl5EJMIU8iIiEaaQFxGJMIW8iEiEKeRFRCJMIS8iEmEKeRGRCFPIi4hEmEJeRCTCFPIiIhGmkBcRiTCFvIhIhCnkRUQiTCEvIhJhCnkRkQhTyIuIRJhCXkQkwhTyIiIRppAXEYkwhbyISIQp5EVEIkwhLyISYQp5EZEIU8iLiESYQl5EJMIU8iIiEaaQFxGJMIW8iEiEJRXyZjbEzHLMbL2Z3VTG/J+a2YdmtsrMXjaz9NQ3VUREqqrSkDezhsB9wFDgTGC8mZ2ZUOyfQE937wY8Dfw+1Q0VEZGqS+ZI/hxgvbtvdPcC4HFgVHwBd3/F3feGD98EOqS2mSIiUh3JhPxJwJa4x1vDaeW5GniurBlmNtHMVprZyry8vORbKSIi1ZJMyFsZ07zMgmaXAz2Bu8qa7+4PuXtPd++ZlpaWfCtFRKRaGiVRZivQMe5xB2BbYiEzOx+4Bejv7gdS0zwREamJZI7k3wG6mNkpZtYEGAdkxxcws7OAB4GR7r4j9c0UEZHqqDTk3b0QmAy8AKwFnnT3NWY23cxGhsXuAloCT5nZe2aWXU51IiJSh5I5XYO7LwIWJUy7Ne7v81PcLhERSQF941VEJMIU8iIiEaaQFxGJMIW8iEiEKeRFRCJMIS8iEmEKeRGRCFPIi4hEmEJeRCTCFPIiIhGmkBcRiTCFvIhIhCnkRUQiTCEvIhJhCnkRkQhTyIuIRJhCXkQkwhTyIiIRppAXEYkwhbyISIQp5EVEIkwhLyISYQp5EZEIU8iLiESYQl5EJMIU8iIiEaaQFxGJMIW8iEiEKeRFRCJMIS8iEmEKeRGRCFPIi4hEWFIhb2ZDzCzHzNab2U1lzD/GzJ4I579lZienuqEiIlJ1lYa8mTUE7gOGAmcC483szIRiVwOfu3tn4B7gzlQ3VEREqi6ZI/lzgPXuvtHdC4DHgVEJZUYBc8K/nwYGmZmlrpkiIlIdjZIocxKwJe7xVuDc8sq4e6GZ7QbaALnxhcxsIjAxfFg8evRovRGIiFTNd6pSOJmQLyuIvRplcPeHgIcAzKyonOVERKR8Veowk0zhrUDHuMcdgG3llTGzRsCxwGdVaYiIiKReMiH/DtDFzE4xsybAOCA7oUw28P3w74uBJe5+2JG8iIjUrUpP14Tn2CcDLwANgUfcfY2ZTQdWuns28Gfgb2a2nuAIflwS634HOBsoCutN/E0F81KxTG2VVf2qX/Wr/tpsy3tUgemAW0QkuvSNVxGRCFPIi4hEmEJeRCTCkuknnxJmNh74IZAf/gwDVhF8Q/Yhdz9YV20REfmqqJMLr2a2AzgOKCb4kpRR+imiGFjk7hfGle8KfE5wJbmtu+eYWVd3z0liXa2A5sBAoC9BD54uBG9oDjQJ17k3nNY8bE8xsDN83BjYBbxP0P+/McE3ej8P53cL69oFfAG0BloC+8I27wZOSFimO3AQyAvntwe+BuwPyzYM6ywK13cg/E3428NpsW3w8Kc4nJ+4TQCF4TJNw23cH04/Jny8N1xfi4T1Nwh/FwM7gE/DthYCm8NtOiVst4VlOoT1x9bdhtL/8YFwnfE7W+wA44twuYbAU+4+BflKMrPbgZ5AL6AZwX5Z1hcmjdL93+KmFcWV2Q18SfDaaB1O2xiWywG+DvQOy20Ezoir9wugHcH++x+C/fcTgn34FIIsKyTYh+PbF1s+9ru8dpe5+WVsE3Hl9wN/Ba6v6gFxXYV8bCWJG1AdxQQbXEwQdkUE/4QGBAFTZ59OpNbE9pfKXjQQvHE2BAoI3kAbAGkE+8XycFkjGFyvLcEb1j7gZGAP5QfBZoIgOI3gDTzWnlj7YvXGtytx/67sRU3C/Fj5AoL9eH/YRgi+YFgEfESwn28jCKaqbFMs3LqH81dRdrjFQuQ/JBduFW1vstt0OqUHCFKxIuB5dx+RTOG6Cvlign/2ZwQ7mYiI1Mxqd+9WWaG6DnkREUmNk93948oK1VXIFxB8pFZvHhGRFHD3pA6c6yp05wL/VUfrEhGRUJ0Oa2BmuwiumB+TMKuY2nvDifV2yQe+6+7/CttyHcHFtX8BFwBPuvuucJ4BX3P3vPjHALFpcdtkBBfm8qu6jLvvoRxm9mvgCYKLUb3c/f7KNjRum94AfgD8JbZNFSwzAujk7rPN7Apgm7u/XMnzsCd+ejLbZGbtCN7o7wO+DQwGJhFcJNWpPKlIMXA7wes0dgF4PcH1vcYEF4PbUNpj7EuCC8YtCS4+7w1/FxJkQVOCXGhFaa+7/QS9xgoIevbsAdLDuhrF1Xs8pT3ydhO83l4My18A9AjbtIagB93BcD3NwmX2AduBBwj2/SyC3n9tE7bpYDjNwrp3ACvD9h0A7nf3Vck8eXU+do2ZFVI66E513efuk+PqjIXbB8ACgi6OqRJrb1WCqDrLVFcqeiwlux5qeV2FwMvAPwl27IEEdyaLvWg6hO1oSvBmU0Tw4isk6BlykOAgogXBC3FnWL45wUFEfFfRXQQvtjSCAMgP/4797/YT9NY5ieCgpEG4zDKCN9HTCW6e0zlcZlO4fBOCF2qTcPqOcLkXytimHIJuqEXhOmJBFL9NBWHZNpQOULWHIDQaEARZ4jZ9LWzLvrDO2OnSg5QG1TEE4fk5pd2AD4Tz24TPcWG4rn3hMrF1fknQY+dFguD5OfCNcL0flrFNB8M6CuOe10bhehpR2m04jdSqq9dGXXqE4HnbBPyPu+dWUr5eQj5VK4ziP1Bqpjr7xJG+H9XlNtXlAcOR/JwfLf7p7j0qK1SnF0JTGPCgnUQOV5194kjfj+pym+rquTjSn/OjxRF14bU8r9Xz+kVEjlZJHTTXR8gXE5yL/K6793d3S+gKFLsIUlDO8oU1WHfsa8NHquq0r66Wqa4j+fkWOVrEXkfFcdO6JLNgnQ4BUFG/zrLmmdm1wOyEybE2V+VCYFFCubK+ip4YRtV5A6zOx9DYWDHxKgrGxHVU53xteev4LKyvVUK9Nfl4rY/mkgrvEQwx0SxuWuJrOn7fzicYe6olQY8XODwz4q8NxIZLWUtwkTgzoVzicoXA/xAMC3EawcX9b3Bopsa37yDBRfj/I+h10zRsWxvCXnjlbFP837nAAILX5zPAH0nCUXtnKDM7QOktsWIhGRv4TMFy9FpN8GZzAhW/aOJf0IUEXeY+BL5F0Asnvkx5L+rd4foqCoLYtAJgOsGgVqdSOp5LbHC7+P0u1qYcYAVwFkHvFIATObR3WW1tUxrwnXLKxYsNbLcpXLaycHOCHh6NCAYSc6Aryf2fDhL05nkP+CbB4H9lraem1hKM55Pp7k0AzOw5YFA4fwWl3RZzCf43BwmCF4JeRj8Eng3riXW5TGW2VOfi8z5giLtX7TS3ux+VP5TucPrRz+ME3fkK4vaPnQTBXEDQJ3kvwYvkRYIXbGHc8tuA/wc8SvBid0pH/KyLn+Jypu8HXk+YX0jQn3oXwdFqWcsVlDGtvLKxn9XAwPC5ezdc58cE92Ku7e0vqsPnOio/7wHHJ5OVR/SRfIp744iIRMkCdx9dWaH67l0jRxa9qYocufYnPD45mYW+amOvH6R07PniuN8HCb7ZdzBh3jEEH31j5/7jb3oS+x0bX7sThz+fiWVjv+PnrQBGApcDw8P1NQ3XVUDwzcnYV63bEnzVukXcOuJvlBAv/g28mGBs8H8TnO+ODcMQ/xvgyXDdlxB8U7EpwY7UKKyvojeB8s73luertu+J1FTi6y+pg/Qj+nSNRJeZDQGeq+92SCRVdHABtX8G4yCld3WLFzufnrj+2EFkRZzgonVrSg+o/ubuV1bWGJ2ukXrh7s8T9E5Zk4Lqiiv5qW37wt/xR0xO2Z98Cin/01e88pYvb3ptKSS4k9S2OlxnTTVI+IndxSv2OFnVfZ7LCnji2pAombG8jGBMpNjNl7YDjyXTGB3JyxGnkgvu8ae8yrKP4EXTpIxpRXHzYuuwuPmNObQ7YmHCMrG64vtq14VY+xJPm8XujQvJbZOXsUys/mYcus3x2xh/yjJWX02fh1idsfsPS9UlNXaNQl7qhXpOidTYe+5+VmWFdLpGROTolNQnIB3JS70ws7KGcxBJVmWn7epTRTdBqmheVbep2N0rPZ+vbmxSL9y9oZnlEAwR0IDgQtJl7v56GW8AHxOM11HWXaTq6sVe0XoeB8aE8+MvujnQD3iFw19rR8I2VVdZXYEh6I67jWAsl4qWqc56ZgJjKb1hyT0EN2AZQjAWTGOC7tFrCa41dCEI1LTwd+wOcbGbrEDpXaG+DMu0DP+OdSk2gm9ONyG4/pBP6V2hCigdsuJtguEq2hN0u76fYNwsSy27AAAAVUlEQVScb4dtNoKhE74ArifoKn0KwXM1Omz3FuBO4GzgsrC+4rA9PyG4eU1fgm7UXQiuiUxM5knUkbyISITp47KISIQp5EVEIkwhLyISYQp5EZEI+/9UWmfHNn7PSAAAAABJRU5ErkJggg==\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"predicoes.plot(kind='bar', grid=True);"
]
}
],
"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": "3.6.5"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment