51-modelo-genero-nome-sklearn.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "NWJ6b5VDKBhs"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import math"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Carga dos dados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# definir arquivo Pickle de entrada\n",
    "#arquivo = 'dados/afastamentos-nomes-sexo-preparado.pkl.bz2'\n",
    "arquivo = 'dados/cadastro-nomes-sexo-preparado.pkl.bz2'\n",
    "#arquivo = 'extra/nome_sexo_pf-preparado.pkl.bz2'"
   ]
  },
  {
   "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>SEXO</th>\n",
       "      <th>QTDE</th>\n",
       "      <th>L0</th>\n",
       "      <th>L1</th>\n",
       "      <th>L2</th>\n",
       "      <th>L3</th>\n",
       "      <th>L4</th>\n",
       "      <th>L5</th>\n",
       "      <th>L6</th>\n",
       "      <th>L7</th>\n",
       "      <th>L8</th>\n",
       "      <th>L9</th>\n",
       "      <th>L10</th>\n",
       "      <th>L11</th>\n",
       "      <th>L12</th>\n",
       "      <th>L13</th>\n",
       "      <th>L14</th>\n",
       "      <th>L15</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NOME</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",
       "      <th></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",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MARIA</th>\n",
       "      <td>0</td>\n",
       "      <td>20389</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOSE</th>\n",
       "      <td>1</td>\n",
       "      <td>15925</td>\n",
       "      <td>5</td>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ANA</th>\n",
       "      <td>0</td>\n",
       "      <td>11497</td>\n",
       "      <td>1</td>\n",
       "      <td>14</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOAO</th>\n",
       "      <td>1</td>\n",
       "      <td>7515</td>\n",
       "      <td>15</td>\n",
       "      <td>1</td>\n",
       "      <td>15</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CARLOS</th>\n",
       "      <td>1</td>\n",
       "      <td>7370</td>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>12</td>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        SEXO   QTDE  L0  L1  L2  L3  L4  L5  L6  L7  L8  L9  L10  L11  L12  \\\n",
       "NOME                                                                         \n",
       "MARIA      0  20389   1   9  18   1  13   0   0   0   0   0    0    0    0   \n",
       "JOSE       1  15925   5  19  15  10   0   0   0   0   0   0    0    0    0   \n",
       "ANA        0  11497   1  14   1   0   0   0   0   0   0   0    0    0    0   \n",
       "JOAO       1   7515  15   1  15  10   0   0   0   0   0   0    0    0    0   \n",
       "CARLOS     1   7370  19  15  12  18   1   3   0   0   0   0    0    0    0   \n",
       "\n",
       "        L13  L14  L15  \n",
       "NOME                   \n",
       "MARIA     0    0    0  \n",
       "JOSE      0    0    0  \n",
       "ANA       0    0    0  \n",
       "JOAO      0    0    0  \n",
       "CARLOS    0    0    0  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_pickle(arquivo)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(41146, 18)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Expansão dos dados agregados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "41146 1 20389 143\n",
      "CPU times: user 498 ms, sys: 0 ns, total: 498 ms\n",
      "Wall time: 3.67 s\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>SEXO</th>\n",
       "      <th>QTDE</th>\n",
       "      <th>L0</th>\n",
       "      <th>L1</th>\n",
       "      <th>L2</th>\n",
       "      <th>L3</th>\n",
       "      <th>L4</th>\n",
       "      <th>L5</th>\n",
       "      <th>L6</th>\n",
       "      <th>L7</th>\n",
       "      <th>L8</th>\n",
       "      <th>L9</th>\n",
       "      <th>L10</th>\n",
       "      <th>L11</th>\n",
       "      <th>L12</th>\n",
       "      <th>L13</th>\n",
       "      <th>L14</th>\n",
       "      <th>L15</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NOME</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",
       "      <th></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",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MARIA</th>\n",
       "      <td>0</td>\n",
       "      <td>20389</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MARIA</th>\n",
       "      <td>0</td>\n",
       "      <td>20389</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MARIA</th>\n",
       "      <td>0</td>\n",
       "      <td>20389</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MARIA</th>\n",
       "      <td>0</td>\n",
       "      <td>20389</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MARIA</th>\n",
       "      <td>0</td>\n",
       "      <td>20389</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</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",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\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>MARCOS</th>\n",
       "      <td>1</td>\n",
       "      <td>4542</td>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MARCOS</th>\n",
       "      <td>1</td>\n",
       "      <td>4542</td>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MARCOS</th>\n",
       "      <td>1</td>\n",
       "      <td>4542</td>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MARCOS</th>\n",
       "      <td>1</td>\n",
       "      <td>4542</td>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MARCOS</th>\n",
       "      <td>1</td>\n",
       "      <td>4542</td>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 18 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        SEXO   QTDE  L0  L1  L2  L3  L4  L5  L6  L7  L8  L9  L10  L11  L12  \\\n",
       "NOME                                                                         \n",
       "MARIA      0  20389   1   9  18   1  13   0   0   0   0   0    0    0    0   \n",
       "MARIA      0  20389   1   9  18   1  13   0   0   0   0   0    0    0    0   \n",
       "MARIA      0  20389   1   9  18   1  13   0   0   0   0   0    0    0    0   \n",
       "MARIA      0  20389   1   9  18   1  13   0   0   0   0   0    0    0    0   \n",
       "MARIA      0  20389   1   9  18   1  13   0   0   0   0   0    0    0    0   \n",
       "...      ...    ...  ..  ..  ..  ..  ..  ..  ..  ..  ..  ..  ...  ...  ...   \n",
       "MARCOS     1   4542  19  15   3  18   1  13   0   0   0   0    0    0    0   \n",
       "MARCOS     1   4542  19  15   3  18   1  13   0   0   0   0    0    0    0   \n",
       "MARCOS     1   4542  19  15   3  18   1  13   0   0   0   0    0    0    0   \n",
       "MARCOS     1   4542  19  15   3  18   1  13   0   0   0   0    0    0    0   \n",
       "MARCOS     1   4542  19  15   3  18   1  13   0   0   0   0    0    0    0   \n",
       "\n",
       "        L13  L14  L15  \n",
       "NOME                   \n",
       "MARIA     0    0    0  \n",
       "MARIA     0    0    0  \n",
       "MARIA     0    0    0  \n",
       "MARIA     0    0    0  \n",
       "MARIA     0    0    0  \n",
       "...     ...  ...  ...  \n",
       "MARCOS    0    0    0  \n",
       "MARCOS    0    0    0  \n",
       "MARCOS    0    0    0  \n",
       "MARCOS    0    0    0  \n",
       "MARCOS    0    0    0  \n",
       "\n",
       "[1000 rows x 18 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "df2 = df.head(50000).copy()\n",
    "#df2 = df.copy()\n",
    "menor = df2['QTDE'].min()\n",
    "maior = df2['QTDE'].max()\n",
    "print(df2.shape[0], menor, maior, math.ceil(np.sqrt(maior / menor)))\n",
    "#df2['TAM'] = df2['QTDE'].apply(lambda x: math.ceil(np.sqrt(x / menor)))\n",
    "df2['TEMP'] = df2[['SEXO', 'QTDE']].apply(\n",
    "    lambda x: [x[0]] * math.ceil(np.sqrt(x[1] / menor)), axis=1)\n",
    "dfe = df2.explode('TEMP').drop(['TEMP'], axis=1) #.reset_index()\n",
    "dfe.head(1000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# usar os dados expandidos invés dos agregados\n",
    "dfo = df # guardar o dataframe original\n",
    "#df = dfe # apontar para o dataframe expandido\n",
    "df = dfe.sample(frac=1) # embaralhar as linhas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Decomposição com PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#from sklearn.decomposition import PCA\n",
    "#X_reduced = PCA(n_components=4).fit_transform(df.drop(['SEXO'], axis=1))\n",
    "#X_reduced[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "#X = pd.DataFrame(X_reduced, index=df.index)\n",
    "#y = df['SEXO'] # apenas a coluna alvo\n",
    "#print('Forma dos dados:', X.shape, y.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Uso dos dados originais"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "dFztelfbKBkQ",
    "outputId": "0eae88b2-28e8-4b92-bfa2-90aed1c2be79"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Forma dos dados: (82709, 16) (82709,)\n"
     ]
    }
   ],
   "source": [
    "X = df.drop(['SEXO', 'QTDE'], axis=1) # tudo, exceto a coluna alvo\n",
    "y = df['SEXO'] # apenas a coluna alvo\n",
    "print('Forma dos dados:', X.shape, y.shape)"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "id": "ADE2HILPiuxH"
   },
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.model_selection import RepeatedStratifiedKFold\n",
    "from sklearn.feature_selection import RFE\n",
    "from sklearn.feature_selection import RFECV\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.pipeline import Pipeline\n",
    "\n",
    "# create pipeline\n",
    "#rfe = RFE(estimator=DecisionTreeClassifier(), n_features_to_select=5)\n",
    "rfe = RFECV(estimator=DecisionTreeClassifier())\n",
    "model = DecisionTreeClassifier()\n",
    "pipeline = Pipeline(steps=[('s',rfe), ('m',model)])\n",
    "\n",
    "# evaluate model\n",
    "cv = RepeatedStratifiedKFold(n_splits=10, n_repeats=3, random_state=1)\n",
    "n_scores = cross_val_score(pipeline, X, y, scoring='accuracy', cv=cv, n_jobs=-1, error_score='raise')\n",
    "\n",
    "# report performance\n",
    "print('Accuracy: %.3f (%.3f)' % (np.mean(n_scores)*100, np.std(n_scores)*100))\n",
    "\n",
    "rfe.fit(X, y)\n",
    "\n",
    "# summarize all features\n",
    "for i in range(X.shape[1]):\n",
    "  print('Column: %d [%s], Rank: %.3f %s' % (\n",
    "      i, X.columns[i], rfe.ranking_[i],\n",
    "      'Selected ' if rfe.support_[i] else ''))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "id": "eLu3BAdlKBkT"
   },
   "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>L0</th>\n",
       "      <th>L1</th>\n",
       "      <th>L2</th>\n",
       "      <th>L3</th>\n",
       "      <th>L4</th>\n",
       "      <th>L5</th>\n",
       "      <th>L6</th>\n",
       "      <th>L7</th>\n",
       "      <th>L8</th>\n",
       "      <th>L9</th>\n",
       "      <th>L10</th>\n",
       "      <th>L11</th>\n",
       "      <th>L12</th>\n",
       "      <th>L13</th>\n",
       "      <th>L14</th>\n",
       "      <th>L15</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NOME</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",
       "      <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>ANNELISE</th>\n",
       "      <td>5</td>\n",
       "      <td>19</td>\n",
       "      <td>9</td>\n",
       "      <td>12</td>\n",
       "      <td>5</td>\n",
       "      <td>14</td>\n",
       "      <td>14</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ADAIR</th>\n",
       "      <td>18</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ROZANA</th>\n",
       "      <td>1</td>\n",
       "      <td>14</td>\n",
       "      <td>1</td>\n",
       "      <td>26</td>\n",
       "      <td>15</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CELIMAR</th>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>9</td>\n",
       "      <td>12</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>INES</th>\n",
       "      <td>19</td>\n",
       "      <td>5</td>\n",
       "      <td>14</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          L0  L1  L2  L3  L4  L5  L6  L7  L8  L9  L10  L11  L12  L13  L14  L15\n",
       "NOME                                                                          \n",
       "ANNELISE   5  19   9  12   5  14  14   1   0   0    0    0    0    0    0    0\n",
       "ADAIR     18   9   1   4   1   0   0   0   0   0    0    0    0    0    0    0\n",
       "ROZANA     1  14   1  26  15  18   0   0   0   0    0    0    0    0    0    0\n",
       "CELIMAR   18   1  13   9  12   5   3   0   0   0    0    0    0    0    0    0\n",
       "INES      19   5  14   9   0   0   0   0   0   0    0    0    0    0    0    0"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "nE7LUtIkKBkV"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NOME\n",
       "ANNELISE    0\n",
       "ADAIR       1\n",
       "ROZANA      0\n",
       "CELIMAR     1\n",
       "INES        0\n",
       "Name: SEXO, dtype: uint8"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "id": "ZwMDGS37iqNj"
   },
   "outputs": [],
   "source": [
    "NUMBER_KFOLD_SPLITS = 5 # number of splits in cross-validation\n",
    "NUMBER_GRID_ITERATIONS = 10 # number of grid iterations to parameters testing\n",
    "SCORING_METRIC = 'accuracy' # the scoring metric to be used"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "C08iaGbOKBkX"
   },
   "outputs": [],
   "source": [
    "from datetime import datetime\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.model_selection import KFold\n",
    "\n",
    "models = {}\n",
    "\n",
    "def evaluate_model(name, model, X=X, y=y):\n",
    "\n",
    "  start = datetime.now()\n",
    "  kfold = KFold(n_splits=NUMBER_KFOLD_SPLITS, shuffle=True, random_state=42)\n",
    "  results = cross_val_score(model, X, y, cv=kfold,\n",
    "                            scoring=SCORING_METRIC, verbose=1, n_jobs=-1)\n",
    "  end = datetime.now()\n",
    "\n",
    "  elapsed = int((end - start).total_seconds() * 1000)\n",
    "  score = results.mean() * 100\n",
    "  stddev = results.std() * 100\n",
    "\n",
    "  models[name] = (model, score, stddev, elapsed)\n",
    "  print(model, '\\nCross-Validation Score: %.2f (+/- %.2f) [%5s ms]' % \\\n",
    "        (score, stddev, elapsed))\n",
    "  return score, stddev, elapsed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "gt9-P6gDKBka"
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import KFold\n",
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "#from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "# faz o ajuste fino do modelo, calculando os melhores hiperparâmetros\n",
    "def fine_tune_model(model, params, X=X, y=y):\n",
    "\n",
    "  print('\\nFine Tuning Model:')\n",
    "  print(model, \"\\nparams:\", params)\n",
    "  \n",
    "  kfold = KFold(n_splits=NUMBER_KFOLD_SPLITS, shuffle=True, random_state=42)\n",
    "\n",
    "  search = RandomizedSearchCV(model, param_distributions=params, \n",
    "                              n_iter=NUMBER_GRID_ITERATIONS, \n",
    "                              scoring=SCORING_METRIC, cv=kfold, \n",
    "                              verbose=1, n_jobs=-1)\n",
    "  #search = GridSearchCV(estimator=model, param_grid=params, scoring='accuracy', cv=kfold, verbose=1)\n",
    "\n",
    "  search.fit(X, y)\n",
    "  print('\\nBest Score: %.2f %%' % (search.best_score_ * 100))\n",
    "  print('Best Params:', search.best_params_)\n",
    "  return search"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "AKjSn0fiKBkd"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LogisticRegression(C=0.1, max_iter=1000, random_state=42, solver='liblinear') \n",
      "Cross-Validation Score: 93.78 (+/- 0.21) [10147 ms]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   10.0s finished\n"
     ]
    }
   ],
   "source": [
    "# Logistic Regression\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "model = LogisticRegression(\n",
    "    random_state=42, multi_class='auto', \n",
    "    max_iter=1000, solver='liblinear', C=0.1) # 93.10\n",
    "evaluate_model('LR', model)\n",
    "\n",
    "params = dict(\n",
    "    solver=['liblinear','lbfgs'], \n",
    "    C=np.logspace(-3,3,7)\n",
    ")\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "id": "stLXLiv5KBkm"
   },
   "source": [
    "# Support Vector Machine (SVM)\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "model = SVC(random_state=42, C=10, gamma=0.001, kernel='rbf') # 89.49\n",
    "evaluate_model('SVM', model)\n",
    "\n",
    "#params = dict(C=np.logspace(-1,1,3)) # 87.90\n",
    "#params = dict(gamma=np.logspace(-3,0,4)) # 89.46\n",
    "#params = dict(kernel=['linear','rbf','poly','sigmoid']) # 89.46\n",
    "params = dict(C=np.logspace(-1,1,3),\n",
    "              gamma=np.logspace(-3,0,4),\n",
    "              kernel=['linear','rbf','poly','sigmoid'])\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "ESU3vXP1KBkx"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LinearSVC(C=0.1, random_state=42) \n",
      "Cross-Validation Score: 93.85 (+/- 0.24) [73954 ms]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  1.2min finished\n"
     ]
    }
   ],
   "source": [
    "# Linear SVM\n",
    "from sklearn.svm import LinearSVC\n",
    "\n",
    "model = LinearSVC(random_state=42, max_iter=1000, C=0.1) # 93.07\n",
    "evaluate_model('LSVM', model)\n",
    "\n",
    "#from scipy.stats import reciprocal, uniform\n",
    "#params = dict(C=uniform(1, 10))\n",
    "params = dict(\n",
    "    C=np.logspace(-4,3,8)\n",
    ")\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "id": "I1ZG1hT_KBkj"
   },
   "source": [
    "# K-Nearest Neighbours (KNN)\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "\n",
    "model = KNeighborsClassifier(n_neighbors=12, p=1, weights='distance') # 93.90\n",
    "evaluate_model('KNN', model)\n",
    "\n",
    "params = dict(\n",
    "    n_neighbors=[7,9,11],\n",
    "    p=[1,2],\n",
    "    weights=['uniform','distance']\n",
    ")\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "id": "iWdLq2ULKBk3"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LinearDiscriminantAnalysis() \n",
      "Cross-Validation Score: 93.76 (+/- 0.22) [ 5184 ms]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    5.1s finished\n"
     ]
    }
   ],
   "source": [
    "# Linear Discriminant Analysis (LDA)\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
    "\n",
    "model = LinearDiscriminantAnalysis(solver='svd') # 93.20\n",
    "evaluate_model('LDA', model)\n",
    "\n",
    "params = dict(\n",
    "    solver=['svd','lsqr','eigen']\n",
    ")\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "id": "n1gzX_b395Q4"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "QuadraticDiscriminantAnalysis() \n",
      "Cross-Validation Score: 52.62 (+/- 0.31) [ 1342 ms]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    1.2s finished\n"
     ]
    }
   ],
   "source": [
    "# Quadratic Discriminant Analysis (QDA)\n",
    "from sklearn.discriminant_analysis import QuadraticDiscriminantAnalysis\n",
    "\n",
    "model = QuadraticDiscriminantAnalysis() # 44.99\n",
    "evaluate_model('QDA', model)\n",
    "\n",
    "params = dict()\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "id": "sTI50bmiKBkv"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GaussianNB(var_smoothing=1.0) \n",
      "Cross-Validation Score: 93.53 (+/- 0.21) [ 1016 ms]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    0.9s finished\n"
     ]
    }
   ],
   "source": [
    "# Naïve Bayes - Gaussian\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "\n",
    "model = GaussianNB(priors=None, var_smoothing=1.0) # 92.95\n",
    "evaluate_model('GNB', model)\n",
    "\n",
    "params = dict(\n",
    "    var_smoothing=np.logspace(0,-9,num=10)\n",
    ")\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "id": "I-ML--bl-abs"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BernoulliNB() \n",
      "Cross-Validation Score: 53.06 (+/- 0.35) [ 1223 ms]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    1.1s finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(53.05831299883603, 0.3488284627106127, 1223)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Naïve Bayes - Bernoulli\n",
    "from sklearn.naive_bayes import BernoulliNB\n",
    "\n",
    "model = BernoulliNB() # 52.24\n",
    "evaluate_model('BNB', model)\n",
    "\n",
    "#params = dict(var_smoothing=np.logspace(0,-9,num=10))\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "id": "LFta0nqyKBkg"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DecisionTreeClassifier(max_depth=17, random_state=42) \n",
      "Cross-Validation Score: 99.76 (+/- 0.03) [ 2111 ms]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    2.0s finished\n"
     ]
    }
   ],
   "source": [
    "# Decision Tree\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "model = DecisionTreeClassifier(\n",
    "    random_state=42, criterion='gini', max_depth=17) # 97.23\n",
    "evaluate_model('DT', model)\n",
    "\n",
    "params = dict(\n",
    "    criterion=['gini','entropy'],\n",
    "    max_depth=[3,5,7,11,13,17,19]\n",
    ")\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "id": "zi5Wzk_iKBko"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RandomForestClassifier(random_state=42) \n",
      "Cross-Validation Score: 99.86 (+/- 0.03) [33243 ms]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:   33.1s finished\n"
     ]
    }
   ],
   "source": [
    "# Ensemble - Random Forest\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "model = RandomForestClassifier(\n",
    "    random_state=42, max_features='auto', n_estimators=100) # 99.80\n",
    "evaluate_model('RF', model)\n",
    "\n",
    "params = dict(\n",
    "    n_estimators=[10,50,100,500], \n",
    "    max_features=['auto','sqrt','log2']\n",
    ")\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "id": "U6AvlHQkKBky"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AdaBoostClassifier(base_estimator=DecisionTreeClassifier(criterion='entropy',\n",
      "                                                         max_depth=17,\n",
      "                                                         random_state=42),\n",
      "                   n_estimators=3) \n",
      "Cross-Validation Score: 99.85 (+/- 0.03) [ 3192 ms]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    3.1s finished\n"
     ]
    }
   ],
   "source": [
    "# Ensemble - Ada Boost\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "model = AdaBoostClassifier(\n",
    "    DecisionTreeClassifier(random_state=42, criterion='entropy', max_depth=17), \n",
    "    n_estimators=3) # 99.76\n",
    "evaluate_model('ABDT', model)\n",
    "\n",
    "params = dict(\n",
    "    n_estimators=[1,3,5,7,9,11,13]\n",
    ")\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "id": "MNsYeL07KBk0"
   },
   "source": [
    "# Ensemble - Gradient Boosting\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "\n",
    "model = GradientBoostingClassifier(random_state=42, max_depth=35, n_estimators=250, learning_rate=0.1) # 96.75\n",
    "evaluate_model('GB', model)\n",
    "\n",
    "params = dict(max_depth=[1,3,5,7,9], n_estimators=[5,50,250,500], learning_rate=[0.01,0.1,1,10,100])\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "id": "67Y9pbl-KBkq"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SGDClassifier(max_iter=100, random_state=42, tol=0.01) \n",
      "Cross-Validation Score: 88.28 (+/- 5.49) [ 1968 ms]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    1.9s finished\n"
     ]
    }
   ],
   "source": [
    "# Stochastic Gradient Descent (SGD)\n",
    "from sklearn.linear_model import SGDClassifier\n",
    "\n",
    "model = SGDClassifier(random_state=42, max_iter=100, tol=0.01) # 91.59\n",
    "evaluate_model('SGD', model)\n",
    "\n",
    "params = dict(\n",
    "    max_iter=[100, 200, 350, 500, 1000], \n",
    "    tol=[0.01, 0.1, 1.0]\n",
    ")\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "id": "xbBC6p2wKBkt"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Perceptron(max_iter=750, random_state=42, tol=0.1) \n",
      "Cross-Validation Score: 82.73 (+/- 7.63) [  983 ms]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    0.9s finished\n"
     ]
    }
   ],
   "source": [
    "# Perceptron\n",
    "from sklearn.linear_model import Perceptron\n",
    "\n",
    "model = Perceptron(random_state=42, max_iter=750, tol=0.1) # 90.87\n",
    "evaluate_model('PCT', model)\n",
    "\n",
    "params = dict(\n",
    "    max_iter=[100, 200, 350, 500, 750, 1000], \n",
    "    tol=[0.1, 0.01, 0.001]\n",
    ")\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "id": "d1VQL0brKBk1"
   },
   "source": [
    "# Multi-Layer Perceptron (MLP)\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "\n",
    "model = MLPClassifier(random_state=42, solver='lbfgs', alpha=1e-07, hidden_layer_sizes=(50, 100, 50)) # 88.67\n",
    "evaluate_model('MLP', model)\n",
    "\n",
    "params = dict(alpha=np.logspace(-9,-1,5), \n",
    "              solver=['lbfgs','sgd','adam'], \n",
    "              hidden_layer_sizes=[(50,50,50), (50,100,50), (100,)])\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "id": "LwPteD3pTyih"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "XGBClassifier(base_score=None, booster=None, colsample_bylevel=None,\n",
      "              colsample_bynode=None, colsample_bytree=0.6,\n",
      "              eval_metric='logloss', gamma=0.4, gpu_id=None,\n",
      "              importance_type='gain', interaction_constraints=None,\n",
      "              learning_rate=None, max_delta_step=None, max_depth=9,\n",
      "              min_child_weight=1, missing=nan, monotone_constraints=None,\n",
      "              n_estimators=100, n_jobs=None, num_parallel_tree=None,\n",
      "              random_state=None, reg_alpha=1e-05, reg_lambda=None,\n",
      "              scale_pos_weight=None, subsample=0.9, tree_method=None,\n",
      "              use_label_encoder=False, validate_parameters=None,\n",
      "              verbosity=None) \n",
      "Cross-Validation Score: 99.93 (+/- 0.02) [348552 ms]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  5.8min finished\n"
     ]
    }
   ],
   "source": [
    "# Extreme Gradient Boosting Machine (XGBM)\n",
    "from xgboost import XGBClassifier\n",
    "#!pip install xgboost\n",
    "\n",
    "model = XGBClassifier(\n",
    "    use_label_encoder=False, eval_metric='logloss',\n",
    "    max_depth=9, min_child_weight=1, gamma=0.4, \n",
    "    subsample=0.9, colsample_bytree=0.6, reg_alpha=1e-05) # 99.20\n",
    "evaluate_model('XGB', model)\n",
    "\n",
    "params = dict(max_depth=range(3,10,2), min_child_weight=range(1,6,2)) # 99.85\n",
    "#params = dict(gamma=[i/10.0 for i in range(0,5)]) # 99.85\n",
    "#params = dict(subsample=[i/10.0 for i in range(6,10)], colsample_bytree=[i/10.0 for i in range(6,10)]) # 99.85\n",
    "#params = dict(reg_alpha=[1e-5, 1e-2, 0.1, 1, 100]) # 99.85\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "id": "KBmR5q4c_X0A"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LGBMClassifier(feature_fraction=0.5, max_depth=11, num_leaves=97) \n",
      "Cross-Validation Score: 99.86 (+/- 0.03) [ 9375 ms]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:    9.3s finished\n"
     ]
    }
   ],
   "source": [
    "# LightGBM\n",
    "from lightgbm import LGBMClassifier\n",
    "#!pip install lightgbm\n",
    "\n",
    "model = LGBMClassifier(\n",
    "    num_leaves=97, max_depth=11, \n",
    "    learning_rate=0.1, feature_fraction=0.5) # 97.09\n",
    "evaluate_model('LGBM', model)\n",
    "\n",
    "#params = dict(num_leaves=range(10,100), max_depth=range(1,30)) # 99.85\n",
    "#params = dict(learning_rate=[0.01,0.05,0.1,0.5]) # 99.85\n",
    "params = dict(feature_fraction=[0.1,0.25,0.5,0.75,1.0]) # 99.87\n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "id": "7JVuRL_xIXLQ"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 12 concurrent workers.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<catboost.core.CatBoostClassifier object at 0x7f668a8f4160> \n",
      "Cross-Validation Score: 99.83 (+/- 0.02) [140043 ms]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   5 out of   5 | elapsed:  2.3min finished\n"
     ]
    }
   ],
   "source": [
    "# CatBoost\n",
    "from catboost import CatBoostClassifier\n",
    "#!pip install catboost\n",
    "\n",
    "model = CatBoostClassifier(\n",
    "    silent=True, iterations=500,\n",
    "    learning_rate=0.05, depth=9) # 99.61\n",
    "#learning_rate=0.1, iterations=1000, depth=5\n",
    "evaluate_model('CB', model)\n",
    "\n",
    "params = dict(\n",
    "    iterations=[10,50,100,250,500,1000],\n",
    "    learning_rate=[0.01,0.05,0.1,0.5],\n",
    "    depth=range(1,11,2)\n",
    ") \n",
    "#fine_tune_model(model, params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "id": "ikE7JK4TnsmA"
   },
   "outputs": [],
   "source": [
    "names = []\n",
    "estimators = []\n",
    "scores = []\n",
    "stdevs = []\n",
    "times = []\n",
    "\n",
    "for key, value in models.items():\n",
    "    (model, score, stdev, elapsed) = value\n",
    "    names.append(key)\n",
    "    estimators.append(model)\n",
    "    scores.append(score)\n",
    "    stdevs.append(stdev)\n",
    "    times.append(elapsed)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "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>Model</th>\n",
       "      <th>Score</th>\n",
       "      <th>Std Dev</th>\n",
       "      <th>Time (ms)</th>\n",
       "      <th>Estimator</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>XGB</td>\n",
       "      <td>99.928666</td>\n",
       "      <td>0.019644</td>\n",
       "      <td>348552</td>\n",
       "      <td>XGBClassifier(base_score=None, booster=None, c...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>RF</td>\n",
       "      <td>99.858540</td>\n",
       "      <td>0.029515</td>\n",
       "      <td>33243</td>\n",
       "      <td>RandomForestClassifier(random_state=42)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>LGBM</td>\n",
       "      <td>99.858540</td>\n",
       "      <td>0.028767</td>\n",
       "      <td>9375</td>\n",
       "      <td>LGBMClassifier(feature_fraction=0.5, max_depth...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>ABDT</td>\n",
       "      <td>99.848868</td>\n",
       "      <td>0.030825</td>\n",
       "      <td>3192</td>\n",
       "      <td>AdaBoostClassifier(base_estimator=DecisionTree...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>CB</td>\n",
       "      <td>99.833150</td>\n",
       "      <td>0.022813</td>\n",
       "      <td>140043</td>\n",
       "      <td>&lt;catboost.core.CatBoostClassifier object at 0x...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>DT</td>\n",
       "      <td>99.756979</td>\n",
       "      <td>0.028462</td>\n",
       "      <td>2111</td>\n",
       "      <td>DecisionTreeClassifier(max_depth=17, random_st...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>LSVM</td>\n",
       "      <td>93.845890</td>\n",
       "      <td>0.242873</td>\n",
       "      <td>73954</td>\n",
       "      <td>LinearSVC(C=0.1, random_state=42)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LR</td>\n",
       "      <td>93.776974</td>\n",
       "      <td>0.211234</td>\n",
       "      <td>10147</td>\n",
       "      <td>LogisticRegression(C=0.1, max_iter=1000, rando...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>LDA</td>\n",
       "      <td>93.757628</td>\n",
       "      <td>0.218362</td>\n",
       "      <td>5184</td>\n",
       "      <td>LinearDiscriminantAnalysis()</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>GNB</td>\n",
       "      <td>93.526699</td>\n",
       "      <td>0.212234</td>\n",
       "      <td>1016</td>\n",
       "      <td>GaussianNB(var_smoothing=1.0)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>SGD</td>\n",
       "      <td>88.284212</td>\n",
       "      <td>5.485385</td>\n",
       "      <td>1968</td>\n",
       "      <td>SGDClassifier(max_iter=100, random_state=42, t...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>PCT</td>\n",
       "      <td>82.730857</td>\n",
       "      <td>7.633005</td>\n",
       "      <td>983</td>\n",
       "      <td>Perceptron(max_iter=750, random_state=42, tol=...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>BNB</td>\n",
       "      <td>53.058313</td>\n",
       "      <td>0.348828</td>\n",
       "      <td>1223</td>\n",
       "      <td>BernoulliNB()</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>QDA</td>\n",
       "      <td>52.620631</td>\n",
       "      <td>0.308413</td>\n",
       "      <td>1342</td>\n",
       "      <td>QuadraticDiscriminantAnalysis()</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Model      Score   Std Dev  Time (ms)  \\\n",
       "11   XGB  99.928666  0.019644     348552   \n",
       "7     RF  99.858540  0.029515      33243   \n",
       "12  LGBM  99.858540  0.028767       9375   \n",
       "8   ABDT  99.848868  0.030825       3192   \n",
       "13    CB  99.833150  0.022813     140043   \n",
       "6     DT  99.756979  0.028462       2111   \n",
       "1   LSVM  93.845890  0.242873      73954   \n",
       "0     LR  93.776974  0.211234      10147   \n",
       "2    LDA  93.757628  0.218362       5184   \n",
       "4    GNB  93.526699  0.212234       1016   \n",
       "9    SGD  88.284212  5.485385       1968   \n",
       "10   PCT  82.730857  7.633005        983   \n",
       "5    BNB  53.058313  0.348828       1223   \n",
       "3    QDA  52.620631  0.308413       1342   \n",
       "\n",
       "                                            Estimator  \n",
       "11  XGBClassifier(base_score=None, booster=None, c...  \n",
       "7             RandomForestClassifier(random_state=42)  \n",
       "12  LGBMClassifier(feature_fraction=0.5, max_depth...  \n",
       "8   AdaBoostClassifier(base_estimator=DecisionTree...  \n",
       "13  <catboost.core.CatBoostClassifier object at 0x...  \n",
       "6   DecisionTreeClassifier(max_depth=17, random_st...  \n",
       "1                   LinearSVC(C=0.1, random_state=42)  \n",
       "0   LogisticRegression(C=0.1, max_iter=1000, rando...  \n",
       "2                        LinearDiscriminantAnalysis()  \n",
       "4                       GaussianNB(var_smoothing=1.0)  \n",
       "9   SGDClassifier(max_iter=100, random_state=42, t...  \n",
       "10  Perceptron(max_iter=750, random_state=42, tol=...  \n",
       "5                                       BernoulliNB()  \n",
       "3                     QuadraticDiscriminantAnalysis()  "
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results_df = pd.DataFrame({\n",
    "    'Model': names,\n",
    "    'Score': scores,\n",
    "    'Std Dev': stdevs,\n",
    "    'Time (ms)': times,\n",
    "    'Estimator': estimators})\n",
    "\n",
    "results_df.sort_values(by='Score', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "id": "dGp1H7jqY4Vu"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<catboost.core.CatBoostClassifier at 0x7f668a8f4160>"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "escolhido = 'CB'\n",
    "model = models[escolhido][0]\n",
    "model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "id": "_mAciwGo5iOE",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 52.9 s, sys: 5.83 s, total: 58.8 s\n",
      "Wall time: 6.19 s\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<catboost.core.CatBoostClassifier at 0x7f668a8f4160>"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "# treinar o melhor modelo com todos os dados\n",
    "model.fit(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['modelos/modelo-genero-nome-CB.joblib.bz2']"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# gravar em disco o modelo treinado\n",
    "from joblib import dump\n",
    "#dump(model, 'modelos/modelo-genero-nome-' + escolhido + '.joblib.bz2', compress=('bz2', 9))\n",
    "dump(model, 'modelos/modelo-genero-nome.joblib.bz2', compress=('bz2', 9))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Avaliação do melhor modelo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "AutoSklearnClassifier(include_preprocessors=['no_preprocessing'],\n",
       "                      memory_limit=2048, n_jobs=-1, per_run_time_limit=300,\n",
       "                      seed=42)"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# carregar um modelo previamente gravado em disco\n",
    "from joblib import load\n",
    "#model = load('modelos/modelo-genero-nome.joblib.bz2')\n",
    "model = load('modelos/modelo-genero-nome-autoskl2.joblib.bz2')\n",
    "model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "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>SEXO</th>\n",
       "      <th>QTDE</th>\n",
       "      <th>L0</th>\n",
       "      <th>L1</th>\n",
       "      <th>L2</th>\n",
       "      <th>L3</th>\n",
       "      <th>L4</th>\n",
       "      <th>L5</th>\n",
       "      <th>L6</th>\n",
       "      <th>L7</th>\n",
       "      <th>L8</th>\n",
       "      <th>L9</th>\n",
       "      <th>L10</th>\n",
       "      <th>L11</th>\n",
       "      <th>L12</th>\n",
       "      <th>L13</th>\n",
       "      <th>L14</th>\n",
       "      <th>L15</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NOME</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",
       "      <th></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",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MARIA</th>\n",
       "      <td>0</td>\n",
       "      <td>2937</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>18</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOSE</th>\n",
       "      <td>1</td>\n",
       "      <td>2766</td>\n",
       "      <td>5</td>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ANTONIO</th>\n",
       "      <td>1</td>\n",
       "      <td>1257</td>\n",
       "      <td>15</td>\n",
       "      <td>9</td>\n",
       "      <td>14</td>\n",
       "      <td>15</td>\n",
       "      <td>20</td>\n",
       "      <td>14</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FRANCISCO</th>\n",
       "      <td>1</td>\n",
       "      <td>1014</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>19</td>\n",
       "      <td>9</td>\n",
       "      <td>3</td>\n",
       "      <td>14</td>\n",
       "      <td>1</td>\n",
       "      <td>18</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOAO</th>\n",
       "      <td>1</td>\n",
       "      <td>993</td>\n",
       "      <td>15</td>\n",
       "      <td>1</td>\n",
       "      <td>15</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           SEXO  QTDE  L0  L1  L2  L3  L4  L5  L6  L7  L8  L9  L10  L11  L12  \\\n",
       "NOME                                                                           \n",
       "MARIA         0  2937   1   9  18   1  13   0   0   0   0   0    0    0    0   \n",
       "JOSE          1  2766   5  19  15  10   0   0   0   0   0   0    0    0    0   \n",
       "ANTONIO       1  1257  15   9  14  15  20  14   1   0   0   0    0    0    0   \n",
       "FRANCISCO     1  1014  15   3  19   9   3  14   1  18   6   0    0    0    0   \n",
       "JOAO          1   993  15   1  15  10   0   0   0   0   0   0    0    0    0   \n",
       "\n",
       "           L13  L14  L15  \n",
       "NOME                      \n",
       "MARIA        0    0    0  \n",
       "JOSE         0    0    0  \n",
       "ANTONIO      0    0    0  \n",
       "FRANCISCO    0    0    0  \n",
       "JOAO         0    0    0  "
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# definir arquivo Pickle de entrada\n",
    "arquivo = 'dados/afastamentos-nomes-sexo-preparado.pkl.bz2'\n",
    "#arquivo = 'dados/cadastro-nomes-sexo-preparado.pkl.bz2'\n",
    "#arquivo = 'extra/nome_sexo_pf-preparado.pkl.bz2'\n",
    "\n",
    "df2 = pd.read_pickle(arquivo)\n",
    "df2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8711, 18)"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test = df2.drop(['SEXO', 'QTDE'], axis=1) # tudo, exceto a coluna alvo\n",
    "y_test = df2['SEXO'] # apenas a coluna alvo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "id": "Pwcb-2OeY_Fv"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 1, 1, ..., 1, 1, 1], dtype=uint8)"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred = model.predict(X_test)\n",
    "y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score: 99.99 %\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "print(\"Accuracy score: %.2f %%\" % (accuracy_score(y_test, y_pred) * 100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "id": "F8w7q7WlaxYG"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix:\n",
      " [[4383    1]\n",
      " [   0 4327]]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "print(\"Confusion matrix:\\n\", confusion_matrix(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "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>REAL</th>\n",
       "      <th>PREV</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NOME</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>MARIA</th>\n",
       "      <td>F</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOSE</th>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ANTONIO</th>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>FRANCISCO</th>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JOAO</th>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CARLOS</th>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ANA</th>\n",
       "      <td>F</td>\n",
       "      <td>F</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAULO</th>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LUIZ</th>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MARCOS</th>\n",
       "      <td>M</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          REAL PREV\n",
       "NOME               \n",
       "MARIA        F    F\n",
       "JOSE         M    M\n",
       "ANTONIO      M    M\n",
       "FRANCISCO    M    M\n",
       "JOAO         M    M\n",
       "CARLOS       M    M\n",
       "ANA          F    F\n",
       "PAULO        M    M\n",
       "LUIZ         M    M\n",
       "MARCOS       M    M"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dados = pd.DataFrame({'REAL': y_test, 'PREV': y_pred}, index=X_test.index)\n",
    "for col in dados.columns:\n",
    "    dados[col] = dados[col].map({0: 'F', 1: 'M', 9: 'X'})\n",
    "dados.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "id": "POsNbYe9Zhou"
   },
   "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>REAL</th>\n",
       "      <th>PREV</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NOME</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>IGNEZ</th>\n",
       "      <td>F</td>\n",
       "      <td>M</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      REAL PREV\n",
       "NOME           \n",
       "IGNEZ    F    M"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dados[dados['REAL'] != dados['PREV']].head(50)"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "61-modelo-genero-nome.ipynb",
   "provenance": []
  },
  "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}