Brasileirao.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Prediction of game results in Brazilian Football Championship"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from datetime import datetime\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n",
    "import tensorflow as tf\n",
    "from sklearn.preprocessing import StandardScaler, OneHotEncoder\n",
    "from sklearn.compose import make_column_transformer, make_column_selector\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Histórico de partidas - brasileirao 2003 a 2020\n",
    "hist_jogos = pd.read_csv(\"/home/rodrigogsvaz/projeto_predicao_brasileirao/Datasets_brasileirao/campeonato-brasileiro-pontos-corridos-2003-2020-jogos.csv\", sep=\";\")\n",
    "\n",
    "# Dados de idade e economico, por time participante, por ano\n",
    "hist_valores = pd.read_excel(\"/home/rodrigogsvaz/projeto_predicao_brasileirao/Datasets_brasileirao/Dados.xlsx\", \"Dados idade, valor\")\n",
    "\n",
    "# Histórico recente de conquistas\n",
    "hist_conquistas = pd.read_excel(\"/home/rodrigogsvaz/projeto_predicao_brasileirao/Datasets_brasileirao/Dados.xlsx\", \"Historico ano anterior\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Treat data and add new features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "hist_jogos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Cria possíveis targets\n",
    "\n",
    "def f_resultado(df):\n",
    "    if df[\"Mandante Placar\"] > df[\"Visitante Placar\"]:\n",
    "        return \"vitoria\"\n",
    "    elif df[\"Mandante Placar\"] == df[\"Visitante Placar\"]:\n",
    "        return \"empate\"\n",
    "    elif df[\"Mandante Placar\"] < df[\"Visitante Placar\"]:\n",
    "        return \"derrota\"\n",
    "    \n",
    "def f_vitoria(df):\n",
    "    if df[\"Mandante Placar\"] > df[\"Visitante Placar\"]:\n",
    "        return 1\n",
    "    else:\n",
    "        return 0\n",
    "    \n",
    "hist_jogos[\"resultado_mandante\"] = hist_jogos.apply(f_resultado, axis=1)\n",
    "hist_jogos[\"vitoria_mandante\"] = hist_jogos.apply(f_vitoria, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Indica se mesmo estado\n",
    "hist_jogos[\"mesmo_estado\"] = hist_jogos.apply(lambda x: 1 if x[\"Estado Mandante\"] == x[\"Estado Visitante\"] else 0, axis = 1)\n",
    "\n",
    "\n",
    "# Indica se mesma região\n",
    "sul = [\"PR\", \"RS\", \"SC\"]\n",
    "sudeste = [\"ES\", \"MG\", \"SP\", \"RJ\"]\n",
    "centro_oeste = [\"GO\", \"MT\", \"MS\", \"DF\"]\n",
    "nordeste = [\"AL\", \"BA\", \"CE\", \"MA\", \"PB\", \"PE\", \"PI\", \"RN\", \"SE\"]\n",
    "norte = [\"AC\", \"AP\", \"AM\", \"PA\", \"RO\", \"RR\", \"TO\"]\n",
    "\n",
    "def f_classifica_regiao(estado):\n",
    "    if estado in sul:\n",
    "        return \"sul\"\n",
    "    elif estado in sudeste:\n",
    "        return \"sudeste\"\n",
    "    elif estado in centro_oeste:\n",
    "        return \"centro_oeste\"\n",
    "    elif estado in nordeste:\n",
    "        return \"nordeste\"\n",
    "    elif estado in norte:\n",
    "        return \"norte\"\n",
    "    \n",
    "\n",
    "\n",
    "hist_jogos[\"regiao_mandante\"] = hist_jogos[\"Estado Mandante\"].apply(lambda x: f_classifica_regiao(x))\n",
    "hist_jogos[\"regiao_visitante\"] = hist_jogos[\"Estado Visitante\"].apply(lambda x: f_classifica_regiao(x))\n",
    "\n",
    "\n",
    "hist_jogos[\"mesma_regiao\"] = hist_jogos.apply(lambda x: 1 if x[\"regiao_mandante\"] == x[\"regiao_visitante\"] else 0, axis = 1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Deixa o ano explícito\n",
    "hist_jogos['data'] = pd.to_datetime(hist_jogos['Data'], format='%d/%m/%Y')\n",
    "hist_jogos['ano'] = hist_jogos['data'].apply(lambda x: x.year)\n",
    "\n",
    "# Pega só o mais importante\n",
    "columns = [\"ano\",\n",
    "           \"Rodada\", \n",
    "           \"Dia\",\n",
    "           \"Horário\",\n",
    "           \"Mandante\",\n",
    "           \"Visitante\",\n",
    "           \"mesmo_estado\",\n",
    "           \"mesma_regiao\",\n",
    "           \"resultado_mandante\",\n",
    "           \"vitoria_mandante\"]\n",
    "\n",
    "hist_jogos = hist_jogos.loc[:, columns]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "hist_valores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "hist_valores.valor_de_mercado_medio.replace({' €':''}, regex=True, inplace=True)\n",
    "hist_valores.valor_de_mercado_total.replace({' €':''}, regex=True, inplace=True)\n",
    "\n",
    "def f_multiplica_valor(x):\n",
    "    if \"mi.\" in x:\n",
    "        return 1000000\n",
    "    elif \"mil\" in x:\n",
    "        return 1000\n",
    "\n",
    "hist_valores[\"valor_de_mercado_medio_aux\"] = hist_valores[\"valor_de_mercado_medio\"].apply(f_multiplica_valor)\n",
    "hist_valores[\"valor_de_mercado_total_aux\"] = hist_valores[\"valor_de_mercado_total\"].apply(f_multiplica_valor)\n",
    "\n",
    "hist_valores.valor_de_mercado_medio.replace({' mi.':''}, regex=True, inplace=True)\n",
    "hist_valores.valor_de_mercado_total.replace({' mi.':''}, regex=True, inplace=True)\n",
    "\n",
    "hist_valores.valor_de_mercado_medio.replace({' mil':''}, regex=True, inplace=True)\n",
    "hist_valores.valor_de_mercado_total.replace({' mil':''}, regex=True, inplace=True)\n",
    "\n",
    "hist_valores.valor_de_mercado_medio.replace({',':'.'}, regex=True, inplace=True)\n",
    "hist_valores.valor_de_mercado_total.replace({',':'.'}, regex=True, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "hist_valores[\"valor_de_mercado_medio\"] = hist_valores[\"valor_de_mercado_medio\"].astype(\"float\")\n",
    "hist_valores[\"valor_de_mercado_total\"] = hist_valores[\"valor_de_mercado_total\"].astype(\"float\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "hist_valores[\"valor_de_mercado_medio\"] = hist_valores[\"valor_de_mercado_medio\"] * hist_valores[\"valor_de_mercado_medio_aux\"]\n",
    "hist_valores[\"valor_de_mercado_total\"] = hist_valores[\"valor_de_mercado_total\"] * hist_valores[\"valor_de_mercado_total_aux\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "hist_valores = hist_valores.rename(columns={\"valor_de_mercado_medio\":\"valor_de_mercado_medio_euros\",\n",
    "                                            \"valor_de_mercado_total\":\"valor_de_mercado_total_euros\"})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Pega só o mais importante\n",
    "columns = [\"ano\", \n",
    "           \"clube\", \n",
    "           \"plantel\", \n",
    "           \"idade_media\", \n",
    "           \"estrangeiros\", \n",
    "           \"valor_de_mercado_medio_euros\", \n",
    "           \"valor_de_mercado_total_euros\"]\n",
    "\n",
    "hist_valores = hist_valores.loc[:, columns]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "hist_conquistas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Indica se houve título no ano anterior\n",
    "def f_titulos(df):\n",
    "    if (df[\"campeao_brasileirao_ano_anterior\"] + \n",
    "        df[\"campeao_copa_do_brasil_ano_anterior\"] + \n",
    "        df[\"campeao_sul_americana_ano_anterior\"] + \n",
    "        df[\"campeao_libertadores_ano_anterior\"]) >= 1:\n",
    "        return 1\n",
    "    else:\n",
    "        return 0\n",
    "\n",
    "hist_conquistas[\"titulo_ano_anterior\"] = hist_conquistas.apply(f_titulos, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Indica se houve título ou G4 no ano anterior ---> Joga libertadores\n",
    "\n",
    "def f_deve_jogar_libertadores(df):\n",
    "    if (df[\"campeao_brasileirao_ano_anterior\"] + \n",
    "        df[\"campeao_copa_do_brasil_ano_anterior\"] + \n",
    "        df[\"campeao_sul_americana_ano_anterior\"] + \n",
    "        df[\"campeao_libertadores_ano_anterior\"]) >= 1:\n",
    "        return 1\n",
    "    elif df[\"posicao_brasileirao_ano_anterior\"] <= 4:\n",
    "        return 1\n",
    "    else:\n",
    "        return 0\n",
    "\n",
    "hist_conquistas[\"posicao_brasileirao_ano_anterior\"] = hist_conquistas[\"posicao_brasileirao_ano_anterior\"].astype(\"int\")\n",
    "hist_conquistas[\"jogando_libertadores\"] = hist_conquistas.apply(f_deve_jogar_libertadores, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Pega só o mais importante\n",
    "columns = [\"ano\",\n",
    "           \"posicao_brasileirao_ano_anterior\", \n",
    "           \"clube\", \n",
    "           \"saldo_de_gols_ano_anterior\", \n",
    "           \"pontos\", \n",
    "           \"campeao_brasileirao_ano_anterior\", \n",
    "           \"campeao_copa_do_brasil_ano_anterior\", \n",
    "           \"campeao_sul_americana_ano_anterior\",\n",
    "           \"campeao_libertadores_ano_anterior\",\n",
    "           \"titulo_ano_anterior\",\n",
    "           \"jogando_libertadores\"]\n",
    "\n",
    "hist_conquistas = hist_conquistas.loc[:, columns]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Padroniza as bases e junta tudo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Padroniza nomes dos times\n",
    "\n",
    "def f_padroniza_nome_times(df):\n",
    "  df = df.replace(['Athlético-PR', 'Athletico Paranaense', 'Clube Atlético Paranaense'], 'Athlético-PR')\n",
    "  df = df.replace(['Guarani', 'Guarani FC'], 'Guarani')\n",
    "  df = df.replace(['Fortaleza', 'Fortaleza EC'], 'Fortaleza')\n",
    "  df = df.replace(['Flamengo', 'FLAMENGO', 'CR Flamengo', 'CR Flamengo\\xa0'], 'Flamengo')\n",
    "  df = df.replace(['Vitória', 'EC Vitória'], 'Vitória')\n",
    "  df = df.replace(['Santos', 'SANTOS', 'Santos FC'], 'Santos')\n",
    "  df = df.replace(['Goiás', 'Goiás EC', ''], 'Goiás')\n",
    "  df = df.replace(['Internacional', 'SC Internacional'], 'Internacional')\n",
    "  df = df.replace(['Cruzeiro', 'CRUZEIRO'], 'Cruzeiro EC')\n",
    "  df = df.replace(['Juventude', 'EC Juventude'], 'Juventude')\n",
    "  df = df.replace(['Corinthians', 'CORINTHIANS', 'SC Corinthians'], 'Corinthians')\n",
    "  df = df.replace(['Criciúma', 'Criciúma EC'], 'Criciúma')\n",
    "  df = df.replace(['Fluminense', 'FLUMINENSE', 'Fluminense FC'], 'Fluminense')\n",
    "  df = df.replace(['Atlético-MG', 'Atlético Mineiro', ''], 'Atlético-MG')\n",
    "  df = df.replace(['Grêmio', 'Grêmio FBPA'], 'Grêmio')\n",
    "  df = df.replace(['Coritiba', 'Coritiba FC', ''], 'Coritiba')\n",
    "  df = df.replace(['Figueirense', 'Figueirense FC'], 'Figueirense')\n",
    "  df = df.replace(['São Paulo', 'São Paulo FC', 'São Paulo FC\\xa0'], 'São Paulo')\n",
    "  df = df.replace(['Bahia', 'EC Bahia', ''], 'Bahia')\n",
    "  df = df.replace(['Paraná', 'Paraná Clube'], 'Paraná')\n",
    "  df = df.replace(['Ponte Preta', 'AA Ponte Preta'], 'Ponte Preta')\n",
    "  df = df.replace(['São Caetano'], 'São Caetano')\n",
    "  df = df.replace(['Vasco', 'CR Vasco da Gama', 'VASCO'], 'Vasco')\n",
    "  df = df.replace(['Paysandu'], 'Paysandu')\n",
    "  df = df.replace(['Palmeiras', 'SE Palmeiras', 'SE Palmeiras\\xa0'], 'Palmeiras')\n",
    "  df = df.replace(['Botafogo-RJ', 'Botafogo-rj', 'Botafogo FR'], 'Botafogo-RJ')\n",
    "  df = df.replace(['Brasiliense'], 'Brasiliense')\n",
    "  df = df.replace(['Santa Cruz', 'Santa Cruz FC'], 'Santa Cruz')\n",
    "  df = df.replace(['Sport', 'Sport Recife'], 'Sport')\n",
    "  df = df.replace(['América-RN'], 'América-RN')\n",
    "  df = df.replace(['Náutico'], 'Náutico')\n",
    "  df = df.replace(['Portuguesa'], 'Portuguesa')\n",
    "  df = df.replace(['Ipatinga'], 'Ipatinga')\n",
    "  df = df.replace(['Avaí', 'Avaí FC'], 'Avaí')\n",
    "  df = df.replace(['Santo André', 'EC Santo André'], 'Santo André')\n",
    "  df = df.replace(['Barueri', 'Grêmio Barueri Futebol Ltda.'], 'Barueri')\n",
    "  df = df.replace(['Atlético-GO', 'Atlético Goianiense'], 'Atlético-GO')\n",
    "  df = df.replace(['Ceará', 'Ceará SC'], 'Ceará')\n",
    "  df = df.replace(['Grêmio Prudente'], 'Grêmio Prudente')\n",
    "  df = df.replace(['América-MG', 'América Mineiro'], 'América-MG')\n",
    "  df = df.replace(['Chapecoense'], 'Chapecoense')\n",
    "  df = df.replace(['Joinville', 'Joinville EC'], 'Joinville')\n",
    "  df = df.replace(['CSA'], 'CSA')\n",
    "  df = df.replace(['Bragantino', 'RB Bragantino'], 'Bragantino')\n",
    "  df = df.replace(['Cuiabá EC'], 'Cuiabá EC')\n",
    "\n",
    "  return df\n",
    "\n",
    "\n",
    "hist_jogos = f_padroniza_nome_times(hist_jogos)\n",
    "hist_valores = f_padroniza_nome_times(hist_valores)\n",
    "hist_conquistas = f_padroniza_nome_times(hist_conquistas)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = hist_jogos.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.merge(df, hist_valores, left_on=[\"ano\", \"Mandante\"], right_on=[\"ano\", \"clube\"], how=\"left\")\n",
    "\n",
    "df = df.rename(columns={\"clube\":\"clube_mandante\",\n",
    "                        \"plantel\":\"plantel_mandante\",\n",
    "                        \"idade_media\":\"idade_media_mandante\",\n",
    "                        \"estrangeiros\":\"estrangeiros_mandante\",\n",
    "                        \"valor_de_mercado_medio_euros\":\"valor_de_mercado_medio_euros_mandante\",\n",
    "                        \"valor_de_mercado_total_euros\":\"valor_de_mercado_total_euros_mandante\"})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.merge(df, hist_valores, left_on=[\"ano\", \"Visitante\"], right_on=[\"ano\", \"clube\"], how=\"left\")\n",
    "\n",
    "df = df.rename(columns={\"clube\":\"clube_visitante\",\n",
    "                        \"plantel\":\"plantel_visitante\",\n",
    "                        \"idade_media\":\"idade_media_visitante\",\n",
    "                        \"estrangeiros\":\"estrangeiros_visitante\",\n",
    "                        \"valor_de_mercado_medio_euros\":\"valor_de_mercado_medio_euros_visitante\",\n",
    "                        \"valor_de_mercado_total_euros\":\"valor_de_mercado_total_euros_visitante\"})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.merge(df, hist_conquistas, left_on=[\"ano\", \"Mandante\"], right_on=[\"ano\", \"clube\"], how=\"left\")\n",
    "\n",
    "df = df.rename(columns={\"clube\":\"clube_mandante\",\n",
    "                        \"posicao_brasileirao_ano_anterior\":\"posicao_brasileirao_ano_anterior_mandante\",\n",
    "                        \"saldo_de_gols_ano_anterior\":\"saldo_de_gols_ano_anterior_mandante\",\n",
    "                        \"pontos\":\"pontos_mandante\",\n",
    "                        \"campeao_brasileirao_ano_anterior\":\"campeao_brasileirao_ano_anterior_mandante\",\n",
    "                        \"campeao_copa_do_brasil_ano_anterior\":\"campeao_copa_do_brasil_ano_anterior_mandante\",\n",
    "                        \"campeao_sul_americana_ano_anterior\":\"campeao_sul_americana_ano_anterior_mandante\",\n",
    "                        \"campeao_libertadores_ano_anterior\":\"campeao_libertadores_ano_anterior_mandante\",\n",
    "                        \"titulo_ano_anterior\":\"titulo_ano_anterior_mandante\",\n",
    "                        \"jogando_libertadores\":\"jogando_libertadores_mandante\"\n",
    "                        })"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.merge(df, hist_conquistas, left_on=[\"ano\", \"Visitante\"], right_on=[\"ano\", \"clube\"], how=\"left\")\n",
    "\n",
    "df = df.rename(columns={\"clube\":\"clube_visitante\",\n",
    "                        \"posicao_brasileirao_ano_anterior\":\"posicao_brasileirao_ano_anterior_visitante\",\n",
    "                        \"saldo_de_gols_ano_anterior\":\"saldo_de_gols_ano_anterior_visitante\",\n",
    "                        \"pontos\":\"pontos_visitante\",\n",
    "                        \"campeao_brasileirao_ano_anterior\":\"campeao_brasileirao_ano_anterior_visitante\",\n",
    "                        \"campeao_copa_do_brasil_ano_anterior\":\"campeao_copa_do_brasil_ano_anterior_visitante\",\n",
    "                        \"campeao_sul_americana_ano_anterior\":\"campeao_sul_americana_ano_anterior_visitante\",\n",
    "                        \"campeao_libertadores_ano_anterior\":\"campeao_libertadores_ano_anterior_visitante\",\n",
    "                        \"titulo_ano_anterior\":\"titulo_ano_anterior_visitante\",\n",
    "                        \"jogando_libertadores\":\"jogando_libertadores_visitante\"\n",
    "                        })"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Filtra apenas os anos que tem info de tudo (2010 a 2020)\n",
    "df = df[(df.ano >= 2010) & (df.ano <= 2020)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Filtragem de colunas\n",
    "\n",
    "columns = [\"ano\",\n",
    "           'Rodada', \n",
    "           'Dia', \n",
    "           'Horário', \n",
    "           'mesmo_estado', \n",
    "           'mesma_regiao', \n",
    "           'resultado_mandante',\n",
    "           'vitoria_mandante', \n",
    "           'plantel_mandante',\n",
    "           'idade_media_mandante', \n",
    "           'estrangeiros_mandante',\n",
    "           'valor_de_mercado_medio_euros_mandante',\n",
    "           'valor_de_mercado_total_euros_mandante',\n",
    "           'plantel_visitante', \n",
    "           'idade_media_visitante', \n",
    "           'estrangeiros_visitante',\n",
    "           'valor_de_mercado_medio_euros_visitante',\n",
    "           'valor_de_mercado_total_euros_visitante',\n",
    "           'posicao_brasileirao_ano_anterior_mandante', \n",
    "           'saldo_de_gols_ano_anterior_mandante', \n",
    "           'pontos_mandante',\n",
    "           'campeao_brasileirao_ano_anterior_mandante',\n",
    "           'campeao_copa_do_brasil_ano_anterior_mandante',\n",
    "           'campeao_sul_americana_ano_anterior_mandante',\n",
    "           'campeao_libertadores_ano_anterior_mandante',\n",
    "           'titulo_ano_anterior_mandante', \n",
    "           'jogando_libertadores_mandante',\n",
    "           'posicao_brasileirao_ano_anterior_visitante', \n",
    "           'clube_visitante',\n",
    "           'saldo_de_gols_ano_anterior_visitante', \n",
    "           'pontos_visitante',\n",
    "           'campeao_brasileirao_ano_anterior_visitante',\n",
    "           'campeao_copa_do_brasil_ano_anterior_visitante',\n",
    "           'campeao_sul_americana_ano_anterior_visitante',\n",
    "           'campeao_libertadores_ano_anterior_visitante',\n",
    "           'titulo_ano_anterior_visitante', \n",
    "           'jogando_libertadores_visitante']\n",
    "\n",
    "df = df.loc[:, columns]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature Engineering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Preenche infos faltantes\n",
    "df[\"campeao_brasileirao_ano_anterior_mandante\"] = df[\"campeao_brasileirao_ano_anterior_mandante\"].fillna(0)\n",
    "df[\"campeao_copa_do_brasil_ano_anterior_mandante\"] = df[\"campeao_copa_do_brasil_ano_anterior_mandante\"].fillna(0)\n",
    "df[\"campeao_sul_americana_ano_anterior_mandante\"] = df[\"campeao_sul_americana_ano_anterior_mandante\"].fillna(0)\n",
    "df[\"campeao_libertadores_ano_anterior_mandante\"] = df[\"campeao_libertadores_ano_anterior_mandante\"].fillna(0)\n",
    "df[\"campeao_brasileirao_ano_anterior_visitante\"] = df[\"campeao_brasileirao_ano_anterior_visitante\"].fillna(0)\n",
    "df[\"campeao_copa_do_brasil_ano_anterior_visitante\"] = df[\"campeao_copa_do_brasil_ano_anterior_visitante\"].fillna(0)\n",
    "df[\"campeao_sul_americana_ano_anterior_visitante\"] = df[\"campeao_sul_americana_ano_anterior_visitante\"].fillna(0)\n",
    "df[\"campeao_libertadores_ano_anterior_visitante\"] = df[\"campeao_libertadores_ano_anterior_visitante\"].fillna(0)\n",
    "df[\"titulo_ano_anterior_mandante\"] = df[\"titulo_ano_anterior_mandante\"].fillna(0)\n",
    "df[\"titulo_ano_anterior_visitante\"] = df[\"titulo_ano_anterior_visitante\"].fillna(0)\n",
    "df[\"jogando_libertadores_mandante\"] = df[\"jogando_libertadores_mandante\"].fillna(0)\n",
    "df[\"jogando_libertadores_visitante\"] = df[\"jogando_libertadores_visitante\"].fillna(0)\n",
    "df[\"posicao_brasileirao_ano_anterior_mandante\"] = df[\"posicao_brasileirao_ano_anterior_mandante\"].fillna(21)\n",
    "df[\"posicao_brasileirao_ano_anterior_visitante\"] = df[\"posicao_brasileirao_ano_anterior_visitante\"].fillna(21)\n",
    "df['Horário'] = df['Horário'].fillna(df['Horário'].mode()[0])\n",
    "df['valor_de_mercado_medio_euros_mandante'] = df['valor_de_mercado_medio_euros_mandante'].fillna(df['valor_de_mercado_medio_euros_mandante'].median())\n",
    "df['valor_de_mercado_medio_euros_visitante'] = df['valor_de_mercado_medio_euros_visitante'].fillna(df['valor_de_mercado_medio_euros_visitante'].median())\n",
    "df['valor_de_mercado_total_euros_mandante'] = df['valor_de_mercado_total_euros_mandante'].fillna(df['valor_de_mercado_total_euros_mandante'].median())\n",
    "df['valor_de_mercado_total_euros_visitante'] = df['valor_de_mercado_total_euros_visitante'].fillna(df['valor_de_mercado_total_euros_visitante'].median())\n",
    "df['idade_media_mandante'] = df['idade_media_mandante'].fillna(df['idade_media_mandante'].mean())\n",
    "df['idade_media_visitante'] = df['idade_media_visitante'].fillna(df['idade_media_visitante'].mean())\n",
    "df['plantel_mandante'] = df['plantel_mandante'].fillna(df['plantel_mandante'].mean())\n",
    "df['plantel_visitante'] = df['plantel_visitante'].fillna(df['plantel_visitante'].mean())\n",
    "df['estrangeiros_mandante'] = df['estrangeiros_mandante'].fillna(df['estrangeiros_mandante'].mean())\n",
    "df['estrangeiros_visitante'] = df['estrangeiros_visitante'].fillna(df['estrangeiros_visitante'].mean())\n",
    "df['estrangeiros_visitante'] = df['estrangeiros_visitante'].fillna(df['estrangeiros_visitante'].mean())\n",
    "df['saldo_de_gols_ano_anterior_mandante'] = df['saldo_de_gols_ano_anterior_mandante'].fillna(df['saldo_de_gols_ano_anterior_mandante'].median())\n",
    "df['saldo_de_gols_ano_anterior_visitante'] = df['saldo_de_gols_ano_anterior_visitante'].fillna(df['saldo_de_gols_ano_anterior_visitante'].median())\n",
    "df['pontos_mandante'] = df['pontos_mandante'].fillna(df['pontos_mandante'].median())\n",
    "df['pontos_visitante'] = df['pontos_visitante'].fillna(df['pontos_visitante'].median())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "New features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"diferenca_valor_de_mercado_medio_euros\"] = df[\"valor_de_mercado_medio_euros_mandante\"] - df[\"valor_de_mercado_medio_euros_visitante\"]\n",
    "df[\"diferenca_valor_de_mercado_total_euros\"] = df[\"valor_de_mercado_total_euros_mandante\"] - df[\"valor_de_mercado_total_euros_visitante\"]\n",
    "df[\"diferenca_idade_media\"] = df[\"idade_media_mandante\"] - df[\"idade_media_visitante\"]\n",
    "df[\"diferenca_plantel\"] = df[\"plantel_mandante\"] - df[\"plantel_visitante\"]\n",
    "df[\"diferenca_estrangeiros\"] = df[\"estrangeiros_mandante\"] - df[\"estrangeiros_visitante\"]\n",
    "df[\"posicao_brasileirao_ano_anterior_diferenca\"] = df[\"posicao_brasileirao_ano_anterior_mandante\"] - df[\"posicao_brasileirao_ano_anterior_visitante\"]\n",
    "df[\"fim_de_semana\"] = df[\"Dia\"].apply(lambda x: 1 if x in ['Sábado', 'Domingo'] else 2)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Feature transformation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Agrupa feature de Horário\n",
    "dia = ['11h00', '11:00','15:00', '15h00', '16h00','16:00', '16:15', '16h20', '16:20', '16:30', '17h00', '17:00', '17:30']\n",
    "tardinha = ['18:00', '18h00', '18:15', '18h30','18:30', '18:45', '19:00', '19h00', '19:15', '19:30', '19h30', '19:45']\n",
    "noite = ['20:00', '20h00', '20:15', '20h30', '20:30', '22h30', '21h00', '21:00', '21h10', '21h30', '21:30', '21:45', '21:46', '21h50', '21:50', '22:00', '22h00', '22:30', '22:50']\n",
    "\n",
    "def f_agrupa_horario(x):\n",
    "    if x in dia:\n",
    "        return \"dia\"\n",
    "    elif x in tardinha:\n",
    "        return \"tardinha\"\n",
    "    elif x in noite:\n",
    "        return 'noite'\n",
    "\n",
    "df[\"horario\"] = df.Horário.apply(f_agrupa_horario)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "colunas_finais = [\"ano\",\n",
    "                  \"Rodada\",\n",
    "                  \"fim_de_semana\",\n",
    "                  \"horario\",\n",
    "                  \"mesmo_estado\",\n",
    "                  \"mesma_regiao\",\n",
    "                  \"titulo_ano_anterior_mandante\",\n",
    "                  \"titulo_ano_anterior_visitante\", \n",
    "                  \"jogando_libertadores_mandante\",\n",
    "                  \"jogando_libertadores_visitante\",\n",
    "                  \"diferenca_valor_de_mercado_medio_euros\",\n",
    "                  \"diferenca_valor_de_mercado_total_euros\",\n",
    "                  \"diferenca_idade_media\",\n",
    "                  \"diferenca_plantel\",\n",
    "                  \"diferenca_estrangeiros\",\n",
    "                  \"posicao_brasileirao_ano_anterior_diferenca\",\n",
    "                  \"vitoria_mandante\",\"resultado_mandante\"]\n",
    "\n",
    "df = df.loc[:, colunas_finais]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data normalization and split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Descarta o que não utilizaremos como target\n",
    "df = df.drop(\"resultado_mandante\", axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Normalization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "features_num = [\n",
    "    \"Rodada\",\n",
    "    \"diferenca_valor_de_mercado_medio_euros\",\n",
    "    \"diferenca_valor_de_mercado_total_euros\",\n",
    "    \"diferenca_idade_media\",\n",
    "    \"diferenca_plantel\",\n",
    "    \"diferenca_estrangeiros\",\n",
    "    \"posicao_brasileirao_ano_anterior_diferenca\",\n",
    " \n",
    "]\n",
    "features_cat = [\n",
    "    \"fim_de_semana\", \n",
    "    \"horario\",\n",
    "    \"mesmo_estado\",\n",
    "    \"mesma_regiao\",\n",
    "    \"titulo_ano_anterior_mandante\",\n",
    "    \"titulo_ano_anterior_visitante\",\n",
    "    \"jogando_libertadores_mandante\",\n",
    "    \"jogando_libertadores_visitante\",\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler, OneHotEncoder\n",
    "from sklearn.preprocessing import StandardScaler, OneHotEncoder\n",
    "from sklearn.pipeline import make_pipeline\n",
    "\n",
    "transformer_num = make_pipeline(\n",
    "    StandardScaler(),\n",
    ")\n",
    "transformer_cat = make_pipeline(\n",
    "    OneHotEncoder(handle_unknown='ignore'),\n",
    ")\n",
    "\n",
    "preprocessor = make_column_transformer(\n",
    "    (transformer_num, features_num),\n",
    "    (transformer_cat, features_cat),\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df.copy()\n",
    "y = X.pop('vitoria_mandante')\n",
    "\n",
    "# Split pelo ano\n",
    "X_train = X[(X.ano >= 2010) & (X.ano <= 2018)]\n",
    "X_valid = X[(X.ano >= 2019) & (X.ano <= 2019)]\n",
    "X_test = X[(X.ano >= 2020) & (X.ano <= 2020)]\n",
    "\n",
    "y_train = y[y.index.isin(X_train.index)]\n",
    "y_valid = y[y.index.isin(X_valid.index)]\n",
    "y_test = y[y.index.isin(X_test.index)]\n",
    "\n",
    "#Dropa o ano\n",
    "X_train = X_train.drop(\"ano\", axis=1)\n",
    "X_valid = X_valid.drop(\"ano\", axis=1)\n",
    "X_test = X_test.drop(\"ano\", axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = preprocessor.fit_transform(X_train)\n",
    "X_valid = preprocessor.transform(X_valid)\n",
    "X_test = preprocessor.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Deep Learning model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[24]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "input_shape = [X_train.shape[1]]\n",
    "input_shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-02-03 12:45:00.437017: E external/local_xla/xla/stream_executor/cuda/cuda_driver.cc:274] failed call to cuInit: UNKNOWN ERROR (100)\n"
     ]
    }
   ],
   "source": [
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n",
    "\n",
    "model = keras.Sequential([\n",
    "    layers.BatchNormalization(input_shape=input_shape),\n",
    "    layers.Dense(64, activation='relu'),\n",
    "    layers.BatchNormalization(),\n",
    "    layers.Dropout(0.3),\n",
    "    layers.Dense(64, activation='relu'),\n",
    "    layers.BatchNormalization(),\n",
    "    layers.Dropout(0.3),\n",
    "    layers.Dense(1, activation='sigmoid'),\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(\n",
    "    optimizer='adam',\n",
    "    loss='binary_crossentropy',\n",
    "    metrics=['binary_accuracy'],\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3420, 24)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "early_stopping = keras.callbacks.EarlyStopping(\n",
    "    patience=100,\n",
    "    min_delta=0.001,\n",
    "    restore_best_weights=True,\n",
    ")\n",
    "history = model.fit(\n",
    "    X_train, y_train,\n",
    "    validation_data=(X_valid, y_valid),\n",
    "    batch_size=512,\n",
    "    epochs=2000,\n",
    "    callbacks=[early_stopping],\n",
    "    verbose=0\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Validation Loss: 0.6479\n",
      "Best Validation Accuracy: 0.6395\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "history_df = pd.DataFrame(history.history)\n",
    "# Start the plot at epoch 5\n",
    "history_df.loc[5:, ['loss', 'val_loss']].plot()\n",
    "history_df.loc[5:, ['binary_accuracy', 'val_binary_accuracy']].plot()\n",
    "\n",
    "print((\"Best Validation Loss: {:0.4f}\" +\\\n",
    "      \"\\nBest Validation Accuracy: {:0.4f}\")\\\n",
    "      .format(history_df['val_loss'].min(), \n",
    "              history_df['val_binary_accuracy'].max()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Parameters Optimization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "def f_treina_rede_parametros(quantidade_camadas, tamanho_camada, tamanho_batch, drop_out):\n",
    " \n",
    "    if quantidade_camadas == 2:\n",
    "        model = keras.Sequential([\n",
    "            layers.BatchNormalization(input_shape=input_shape),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(1, activation='sigmoid'),\n",
    "        ])\n",
    "    elif quantidade_camadas == 3:\n",
    "        model = keras.Sequential([\n",
    "            layers.BatchNormalization(input_shape=input_shape),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(1, activation='sigmoid'),\n",
    "        ])\n",
    "    elif quantidade_camadas == 4:\n",
    "        model = keras.Sequential([\n",
    "            layers.BatchNormalization(input_shape=input_shape),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(1, activation='sigmoid'),\n",
    "        ])\n",
    "    elif quantidade_camadas == 5:\n",
    "        model = keras.Sequential([\n",
    "            layers.BatchNormalization(input_shape=input_shape),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(1, activation='sigmoid'),\n",
    "        ])\n",
    "    elif quantidade_camadas == 10:\n",
    "        model = keras.Sequential([\n",
    "            layers.BatchNormalization(input_shape=input_shape),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(tamanho_camada, activation='relu'),\n",
    "            layers.BatchNormalization(),\n",
    "            layers.Dropout(drop_out),\n",
    "            layers.Dense(1, activation='sigmoid'),\n",
    "        ])\n",
    "\n",
    "\n",
    "    early_stopping = keras.callbacks.EarlyStopping(\n",
    "        patience=100,\n",
    "        min_delta=0.001,\n",
    "        restore_best_weights=True,\n",
    "    )\n",
    "\n",
    "    model.compile(\n",
    "        optimizer='adam',\n",
    "        loss='binary_crossentropy',\n",
    "        metrics=['binary_accuracy'],\n",
    "    )\n",
    "\n",
    "    history = model.fit(\n",
    "        X_train, y_train,\n",
    "        validation_data=(X_valid, y_valid),\n",
    "        batch_size=tamanho_batch,\n",
    "        epochs=2000,\n",
    "        callbacks=[early_stopping],\n",
    "        verbose=0\n",
    "    )\n",
    "\n",
    "    history_df = pd.DataFrame(history.history)\n",
    "    # Start the plot at epoch 5\n",
    "    history_df.loc[5:, ['loss', 'val_loss']].plot()\n",
    "    history_df.loc[5:, ['binary_accuracy', 'val_binary_accuracy']].plot()\n",
    "\n",
    "    print((\"Best Validation Loss: {:0.4f}\" +\\\n",
    "          \"\\nBest Validation Accuracy: {:0.4f}\")\\\n",
    "          .format(history_df['val_loss'].min(), \n",
    "                  history_df['val_binary_accuracy'].max()))\n",
    "\n",
    "    return round(history_df['val_binary_accuracy'].max(),4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 64, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6289 # Melhor Acurácia = 0.6289\n",
      "2 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 64, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6737 # Melhor Acurácia = 0.6737\n",
      "3 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 64, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6684 # Melhor Acurácia = 0.6737\n",
      "4 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 64, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.6342 # Melhor Acurácia = 0.6737\n",
      "5 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 64, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6421 # Melhor Acurácia = 0.6737\n",
      "6 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 64, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6605 # Melhor Acurácia = 0.6737\n",
      "7 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 64, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6184 # Melhor Acurácia = 0.6737\n",
      "8 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 64, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6737 # Melhor Acurácia = 0.6737\n",
      "9 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 64, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.6842 # Melhor Acurácia = 0.6842\n",
      "10 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 64, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6263 # Melhor Acurácia = 0.6842\n",
      "11 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 64, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.6789 # Melhor Acurácia = 0.6842\n",
      "12 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 64, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.6737 # Melhor Acurácia = 0.6842\n",
      "13 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 128, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6421 # Melhor Acurácia = 0.6842\n",
      "14 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 128, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6579 # Melhor Acurácia = 0.6842\n",
      "15 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 128, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6632 # Melhor Acurácia = 0.6842\n",
      "16 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 128, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.6447 # Melhor Acurácia = 0.6842\n",
      "17 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 128, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6658 # Melhor Acurácia = 0.6842\n",
      "18 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 128, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6526 # Melhor Acurácia = 0.6842\n",
      "19 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 128, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6105 # Melhor Acurácia = 0.6842\n",
      "20 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 128, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6579 # Melhor Acurácia = 0.6842\n",
      "21 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 128, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.6553 # Melhor Acurácia = 0.6842\n",
      "22 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 128, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.5816 # Melhor Acurácia = 0.6842\n",
      "23 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 128, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.6395 # Melhor Acurácia = 0.6842\n",
      "24 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 128, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.6579 # Melhor Acurácia = 0.6842\n",
      "25 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 256, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6184 # Melhor Acurácia = 0.6842\n",
      "26 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 256, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6658 # Melhor Acurácia = 0.6842\n",
      "27 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 256, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6658 # Melhor Acurácia = 0.6842\n",
      "28 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 256, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.6132 # Melhor Acurácia = 0.6842\n",
      "29 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 256, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6289 # Melhor Acurácia = 0.6842\n",
      "30 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 256, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6553 # Melhor Acurácia = 0.6842\n",
      "31 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 256, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.5947 # Melhor Acurácia = 0.6842\n",
      "32 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 256, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6237 # Melhor Acurácia = 0.6842\n",
      "33 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 256, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.6553 # Melhor Acurácia = 0.6842\n",
      "34 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 256, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6 # Melhor Acurácia = 0.6842\n",
      "35 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 256, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.6421 # Melhor Acurácia = 0.6842\n",
      "36 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 256, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.6711 # Melhor Acurácia = 0.6842\n",
      "37 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 512, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6447 # Melhor Acurácia = 0.6842\n",
      "38 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 512, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6211 # Melhor Acurácia = 0.6842\n",
      "39 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 512, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6553 # Melhor Acurácia = 0.6842\n",
      "40 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 512, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.6132 # Melhor Acurácia = 0.6842\n",
      "41 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 512, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6263 # Melhor Acurácia = 0.6842\n",
      "42 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 512, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6579 # Melhor Acurácia = 0.6842\n",
      "43 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 512, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6316 # Melhor Acurácia = 0.6842\n",
      "44 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 512, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6316 # Melhor Acurácia = 0.6842\n",
      "45 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 512, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.6632 # Melhor Acurácia = 0.6842\n",
      "46 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 512, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6053 # Melhor Acurácia = 0.6842\n",
      "47 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 512, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.6368 # Melhor Acurácia = 0.6842\n",
      "48 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 512, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.6684 # Melhor Acurácia = 0.6842\n",
      "49 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 1024, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6263 # Melhor Acurácia = 0.6842\n",
      "50 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 1024, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6368 # Melhor Acurácia = 0.6842\n",
      "51 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 1024, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.65 # Melhor Acurácia = 0.6842\n",
      "52 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 1024, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.5868 # Melhor Acurácia = 0.6842\n",
      "53 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 1024, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6447 # Melhor Acurácia = 0.6842\n",
      "54 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 1024, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6684 # Melhor Acurácia = 0.6842\n",
      "55 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 1024, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6237 # Melhor Acurácia = 0.6842\n",
      "56 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 1024, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6526 # Melhor Acurácia = 0.6842\n",
      "57 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 1024, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.6421 # Melhor Acurácia = 0.6842\n",
      "58 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 1024, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6079 # Melhor Acurácia = 0.6842\n",
      "59 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 1024, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.65 # Melhor Acurácia = 0.6842\n",
      "60 - Set of parameters: quantidade_camadas: 3, tamanho_camada: 1024, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.6526 # Melhor Acurácia = 0.6842\n",
      "61 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 64, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6342 # Melhor Acurácia = 0.6842\n",
      "62 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 64, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6447 # Melhor Acurácia = 0.6842\n",
      "63 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 64, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6579 # Melhor Acurácia = 0.6842\n",
      "64 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 64, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.6184 # Melhor Acurácia = 0.6842\n",
      "65 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 64, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6605 # Melhor Acurácia = 0.6842\n",
      "66 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 64, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6526 # Melhor Acurácia = 0.6842\n",
      "67 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 64, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6342 # Melhor Acurácia = 0.6842\n",
      "68 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 64, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6579 # Melhor Acurácia = 0.6842\n",
      "69 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 64, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.65 # Melhor Acurácia = 0.6842\n",
      "70 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 64, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6 # Melhor Acurácia = 0.6842\n",
      "71 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 64, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.6553 # Melhor Acurácia = 0.6842\n",
      "72 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 64, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.6553 # Melhor Acurácia = 0.6842\n",
      "73 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 128, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6211 # Melhor Acurácia = 0.6842\n",
      "74 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 128, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6447 # Melhor Acurácia = 0.6842\n",
      "75 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 128, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6684 # Melhor Acurácia = 0.6842\n",
      "76 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 128, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.5947 # Melhor Acurácia = 0.6842\n",
      "77 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 128, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.65 # Melhor Acurácia = 0.6842\n",
      "78 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 128, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6553 # Melhor Acurácia = 0.6842\n",
      "79 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 128, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6263 # Melhor Acurácia = 0.6842\n",
      "80 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 128, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6684 # Melhor Acurácia = 0.6842\n",
      "81 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 128, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.6711 # Melhor Acurácia = 0.6842\n",
      "82 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 128, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6342 # Melhor Acurácia = 0.6842\n",
      "83 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 128, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.6526 # Melhor Acurácia = 0.6842\n",
      "84 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 128, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.6632 # Melhor Acurácia = 0.6842\n",
      "85 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 256, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6053 # Melhor Acurácia = 0.6842\n",
      "86 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 256, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6526 # Melhor Acurácia = 0.6842\n",
      "87 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 256, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6684 # Melhor Acurácia = 0.6842\n",
      "88 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 256, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.6237 # Melhor Acurácia = 0.6842\n",
      "89 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 256, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6684 # Melhor Acurácia = 0.6842\n",
      "90 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 256, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6632 # Melhor Acurácia = 0.6842\n",
      "91 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 256, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.65 # Melhor Acurácia = 0.6842\n",
      "92 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 256, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.65 # Melhor Acurácia = 0.6842\n",
      "93 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 256, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.6684 # Melhor Acurácia = 0.6842\n",
      "94 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 256, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6026 # Melhor Acurácia = 0.6842\n",
      "95 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 256, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.6553 # Melhor Acurácia = 0.6842\n",
      "96 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 256, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.6658 # Melhor Acurácia = 0.6842\n",
      "97 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 512, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6053 # Melhor Acurácia = 0.6842\n",
      "98 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 512, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6421 # Melhor Acurácia = 0.6842\n",
      "99 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 512, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6605 # Melhor Acurácia = 0.6842\n",
      "100 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 512, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.6421 # Melhor Acurácia = 0.6842\n",
      "101 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 512, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6421 # Melhor Acurácia = 0.6842\n",
      "102 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 512, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6605 # Melhor Acurácia = 0.6842\n",
      "103 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 512, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.5868 # Melhor Acurácia = 0.6842\n",
      "104 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 512, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6342 # Melhor Acurácia = 0.6842\n",
      "105 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 512, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.6684 # Melhor Acurácia = 0.6842\n",
      "106 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 512, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6342 # Melhor Acurácia = 0.6842\n",
      "107 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 512, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.65 # Melhor Acurácia = 0.6842\n",
      "108 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 512, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.6711 # Melhor Acurácia = 0.6842\n",
      "109 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 1024, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6026 # Melhor Acurácia = 0.6842\n",
      "110 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 1024, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6158 # Melhor Acurácia = 0.6842\n",
      "111 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 1024, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6658 # Melhor Acurácia = 0.6842\n",
      "112 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 1024, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.6026 # Melhor Acurácia = 0.6842\n",
      "113 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 1024, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6342 # Melhor Acurácia = 0.6842\n",
      "114 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 1024, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6658 # Melhor Acurácia = 0.6842\n",
      "115 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 1024, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6395 # Melhor Acurácia = 0.6842\n",
      "116 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 1024, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6158 # Melhor Acurácia = 0.6842\n",
      "117 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 1024, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.65 # Melhor Acurácia = 0.6842\n",
      "118 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 1024, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6026 # Melhor Acurácia = 0.6842\n",
      "119 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 1024, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.6211 # Melhor Acurácia = 0.6842\n",
      "120 - Set of parameters: quantidade_camadas: 4, tamanho_camada: 1024, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.6579 # Melhor Acurácia = 0.6842\n",
      "121 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 64, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6553 # Melhor Acurácia = 0.6842\n",
      "122 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 64, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6605 # Melhor Acurácia = 0.6842\n",
      "123 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 64, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6605 # Melhor Acurácia = 0.6842\n",
      "124 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 64, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.6263 # Melhor Acurácia = 0.6842\n",
      "125 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 64, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6579 # Melhor Acurácia = 0.6842\n",
      "126 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 64, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6605 # Melhor Acurácia = 0.6842\n",
      "127 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 64, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6079 # Melhor Acurácia = 0.6842\n",
      "128 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 64, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6763 # Melhor Acurácia = 0.6842\n",
      "129 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 64, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.6605 # Melhor Acurácia = 0.6842\n",
      "130 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 64, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6237 # Melhor Acurácia = 0.6842\n",
      "131 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 64, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.6632 # Melhor Acurácia = 0.6842\n",
      "132 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 64, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.6737 # Melhor Acurácia = 0.6842\n",
      "133 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 128, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6395 # Melhor Acurácia = 0.6842\n",
      "134 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 128, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6447 # Melhor Acurácia = 0.6842\n",
      "135 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 128, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6658 # Melhor Acurácia = 0.6842\n",
      "136 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 128, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.6132 # Melhor Acurácia = 0.6842\n",
      "137 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 128, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6632 # Melhor Acurácia = 0.6842\n",
      "138 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 128, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6579 # Melhor Acurácia = 0.6842\n",
      "139 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 128, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6 # Melhor Acurácia = 0.6842\n",
      "140 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 128, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.65 # Melhor Acurácia = 0.6842\n",
      "141 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 128, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.6658 # Melhor Acurácia = 0.6842\n",
      "142 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 128, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6211 # Melhor Acurácia = 0.6842\n",
      "143 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 128, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.6526 # Melhor Acurácia = 0.6842\n",
      "144 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 128, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.6447 # Melhor Acurácia = 0.6842\n",
      "145 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 256, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6447 # Melhor Acurácia = 0.6842\n",
      "146 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 256, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6579 # Melhor Acurácia = 0.6842\n",
      "147 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 256, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6579 # Melhor Acurácia = 0.6842\n",
      "148 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 256, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.6158 # Melhor Acurácia = 0.6842\n",
      "149 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 256, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6421 # Melhor Acurácia = 0.6842\n",
      "150 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 256, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6632 # Melhor Acurácia = 0.6842\n",
      "151 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 256, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6316 # Melhor Acurácia = 0.6842\n",
      "152 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 256, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6447 # Melhor Acurácia = 0.6842\n",
      "153 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 256, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.6711 # Melhor Acurácia = 0.6842\n",
      "154 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 256, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6053 # Melhor Acurácia = 0.6842\n",
      "155 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 256, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.6632 # Melhor Acurácia = 0.6842\n",
      "156 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 256, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.6474 # Melhor Acurácia = 0.6842\n",
      "157 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 512, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.5974 # Melhor Acurácia = 0.6842\n",
      "158 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 512, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6342 # Melhor Acurácia = 0.6842\n",
      "159 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 512, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.65 # Melhor Acurácia = 0.6842\n",
      "160 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 512, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.6079 # Melhor Acurácia = 0.6842\n",
      "161 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 512, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6158 # Melhor Acurácia = 0.6842\n",
      "162 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 512, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6605 # Melhor Acurácia = 0.6842\n",
      "163 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 512, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6237 # Melhor Acurácia = 0.6842\n",
      "164 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 512, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6447 # Melhor Acurácia = 0.6842\n",
      "165 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 512, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.6684 # Melhor Acurácia = 0.6842\n",
      "166 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 512, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6237 # Melhor Acurácia = 0.6842\n",
      "167 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 512, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.6237 # Melhor Acurácia = 0.6842\n",
      "168 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 512, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.6684 # Melhor Acurácia = 0.6842\n",
      "169 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 1024, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6289 # Melhor Acurácia = 0.6842\n",
      "170 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 1024, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6579 # Melhor Acurácia = 0.6842\n",
      "171 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 1024, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6526 # Melhor Acurácia = 0.6842\n",
      "172 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 1024, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.65 # Melhor Acurácia = 0.6842\n",
      "173 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 1024, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6263 # Melhor Acurácia = 0.6842\n",
      "174 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 1024, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6684 # Melhor Acurácia = 0.6842\n",
      "175 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 1024, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6079 # Melhor Acurácia = 0.6842\n",
      "176 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 1024, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6184 # Melhor Acurácia = 0.6842\n",
      "177 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 1024, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.6711 # Melhor Acurácia = 0.6842\n",
      "178 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 1024, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6237 # Melhor Acurácia = 0.6842\n",
      "179 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 1024, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.6368 # Melhor Acurácia = 0.6842\n",
      "180 - Set of parameters: quantidade_camadas: 5, tamanho_camada: 1024, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.6447 # Melhor Acurácia = 0.6842\n",
      "181 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 64, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6421 # Melhor Acurácia = 0.6842\n",
      "182 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 64, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6658 # Melhor Acurácia = 0.6842\n",
      "183 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 64, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.5395 # Melhor Acurácia = 0.6842\n",
      "184 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 64, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.6342 # Melhor Acurácia = 0.6842\n",
      "185 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 64, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6526 # Melhor Acurácia = 0.6842\n",
      "186 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 64, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.5684 # Melhor Acurácia = 0.6842\n",
      "187 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 64, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6263 # Melhor Acurácia = 0.6842\n",
      "188 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 64, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6605 # Melhor Acurácia = 0.6842\n",
      "189 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 64, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.5447 # Melhor Acurácia = 0.6842\n",
      "190 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 64, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6395 # Melhor Acurácia = 0.6842\n",
      "191 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 64, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.6447 # Melhor Acurácia = 0.6842\n",
      "192 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 64, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.55 # Melhor Acurácia = 0.6842\n",
      "193 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 128, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6421 # Melhor Acurácia = 0.6842\n",
      "194 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 128, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6579 # Melhor Acurácia = 0.6842\n",
      "195 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 128, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6737 # Melhor Acurácia = 0.6842\n",
      "196 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 128, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.65 # Melhor Acurácia = 0.6842\n",
      "197 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 128, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6658 # Melhor Acurácia = 0.6842\n",
      "198 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 128, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.5158 # Melhor Acurácia = 0.6842\n",
      "199 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 128, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6316 # Melhor Acurácia = 0.6842\n",
      "200 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 128, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6474 # Melhor Acurácia = 0.6842\n",
      "201 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 128, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.5263 # Melhor Acurácia = 0.6842\n",
      "202 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 128, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.6474 # Melhor Acurácia = 0.6842\n",
      "203 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 128, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.65 # Melhor Acurácia = 0.6842\n",
      "204 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 128, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.5395 # Melhor Acurácia = 0.6842\n",
      "205 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 256, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6211 # Melhor Acurácia = 0.6842\n",
      "206 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 256, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6658 # Melhor Acurácia = 0.6842\n",
      "207 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 256, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6632 # Melhor Acurácia = 0.6842\n",
      "208 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 256, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.6105 # Melhor Acurácia = 0.6842\n",
      "209 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 256, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6579 # Melhor Acurácia = 0.6842\n",
      "210 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 256, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.5342 # Melhor Acurácia = 0.6842\n",
      "211 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 256, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6026 # Melhor Acurácia = 0.6842\n",
      "212 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 256, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6395 # Melhor Acurácia = 0.6842\n",
      "213 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 256, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.5237 # Melhor Acurácia = 0.6842\n",
      "214 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 256, tamanho_batch: 1024, drop_out=0.1 ---> Acurácia = 0.5921 # Melhor Acurácia = 0.6842\n",
      "215 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 256, tamanho_batch: 1024, drop_out=0.3 ---> Acurácia = 0.65 # Melhor Acurácia = 0.6842\n",
      "216 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 256, tamanho_batch: 1024, drop_out=0.5 ---> Acurácia = 0.5632 # Melhor Acurácia = 0.6842\n",
      "217 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 512, tamanho_batch: 128, drop_out=0.1 ---> Acurácia = 0.6184 # Melhor Acurácia = 0.6842\n",
      "218 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 512, tamanho_batch: 128, drop_out=0.3 ---> Acurácia = 0.6395 # Melhor Acurácia = 0.6842\n",
      "219 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 512, tamanho_batch: 128, drop_out=0.5 ---> Acurácia = 0.6632 # Melhor Acurácia = 0.6842\n",
      "220 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 512, tamanho_batch: 256, drop_out=0.1 ---> Acurácia = 0.6263 # Melhor Acurácia = 0.6842\n",
      "221 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 512, tamanho_batch: 256, drop_out=0.3 ---> Acurácia = 0.6526 # Melhor Acurácia = 0.6842\n",
      "222 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 512, tamanho_batch: 256, drop_out=0.5 ---> Acurácia = 0.6395 # Melhor Acurácia = 0.6842\n",
      "223 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 512, tamanho_batch: 512, drop_out=0.1 ---> Acurácia = 0.6 # Melhor Acurácia = 0.6842\n",
      "224 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 512, tamanho_batch: 512, drop_out=0.3 ---> Acurácia = 0.6237 # Melhor Acurácia = 0.6842\n",
      "225 - Set of parameters: quantidade_camadas: 10, tamanho_camada: 512, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.6579 # Melhor Acurácia = 0.6842\n"
     ]
    },
    {
     "ename": "",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31mThe Kernel crashed while executing code in the the current cell or a previous cell. Please review the code in the cell(s) to identify a possible cause of the failure. Click <a href='https://aka.ms/vscodeJupyterKernelCrash'>here</a> for more info. View Jupyter <a href='command:jupyter.viewOutput'>log</a> for further details."
     ]
    }
   ],
   "source": [
    "quantidade_camadas = 2\n",
    "tamanho_camada = 64\n",
    "tamanho_batch = 512\n",
    "drop_out = 0.3\n",
    "\n",
    "exp_quantidade_camadas = [3, 4, 5, 10]\n",
    "exp_tamanho_camada = [64, 128, 256, 512, 1024]\n",
    "exp_tamanho_batch = [128, 256, 512, 1024]\n",
    "exp_drop_out = [0.1, 0.3, 0.5]\n",
    "\n",
    "lista_resultados = []\n",
    "iter = 0\n",
    "melhor_acuracia = 0\n",
    "for quantidade_camadas in exp_quantidade_camadas:\n",
    "    for tamanho_camada in exp_tamanho_camada:\n",
    "        for tamanho_batch in exp_tamanho_batch:\n",
    "            for drop_out in exp_drop_out:\n",
    "\n",
    "                acuracia = f_treina_rede_parametros(quantidade_camadas = quantidade_camadas, \n",
    "                                                    tamanho_camada = tamanho_camada, \n",
    "                                                    tamanho_batch = tamanho_batch, \n",
    "                                                    drop_out = drop_out)\n",
    "                \n",
    "                resumo = [quantidade_camadas, tamanho_camada, tamanho_batch, drop_out, acuracia]\n",
    "                lista_resultados.append(resumo)\n",
    "                iter = iter + 1\n",
    "                if acuracia  > melhor_acuracia:\n",
    "                    melhor_acuracia = acuracia\n",
    "                print(f\"{iter} - Set of parameters: quantidade_camadas: {quantidade_camadas}, tamanho_camada: {tamanho_camada}, tamanho_batch: {tamanho_batch}, drop_out={drop_out} ---> Acurácia = {acuracia} # Melhor Acurácia = {melhor_acuracia}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Best set of parameters:\n",
    "Set of parameters: quantidade_camadas: 3, tamanho_camada: 64, tamanho_batch: 512, drop_out=0.5 ---> Acurácia = 0.6842 # Melhor Acurácia = 0.6842"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Validation Loss: 0.6456\n",
      "Best Validation Accuracy: 0.6605\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.6605"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f_treina_rede_parametros(quantidade_camadas = 3, \n",
    "                         tamanho_camada = 64, \n",
    "                         tamanho_batch = 512, \n",
    "                         drop_out = 0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n",
    "\n",
    "model = keras.Sequential([\n",
    "    layers.BatchNormalization(input_shape=input_shape),\n",
    "    layers.Dense(64, activation='relu'),\n",
    "    layers.BatchNormalization(),\n",
    "    layers.Dropout(0.5),\n",
    "    layers.Dense(64, activation='relu'),\n",
    "    layers.BatchNormalization(),\n",
    "    layers.Dropout(0.5),\n",
    "    layers.Dense(64, activation='relu'),\n",
    "    layers.BatchNormalization(),\n",
    "    layers.Dropout(0.5),\n",
    "    layers.Dense(1, activation='sigmoid'),\n",
    "])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(\n",
    "    optimizer='adam',\n",
    "    loss='binary_crossentropy',\n",
    "    metrics=['binary_accuracy'],\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "early_stopping = keras.callbacks.EarlyStopping(\n",
    "    patience=100,\n",
    "    min_delta=0.001,\n",
    "    restore_best_weights=True,\n",
    ")\n",
    "history = model.fit(\n",
    "    X_train, y_train,\n",
    "    validation_data=(X_valid, y_valid),\n",
    "    batch_size=512,\n",
    "    epochs=2000,\n",
    "    callbacks=[early_stopping],\n",
    "    verbose=0\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Validation Loss: 0.6439\n",
      "Best Validation Accuracy: 0.6658\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "history_df = pd.DataFrame(history.history)\n",
    "# Start the plot at epoch 5\n",
    "history_df.loc[5:, ['loss', 'val_loss']].plot()\n",
    "history_df.loc[5:, ['binary_accuracy', 'val_binary_accuracy']].plot()\n",
    "\n",
    "print((\"Best Validation Loss: {:0.4f}\" +\\\n",
    "      \"\\nBest Validation Accuracy: {:0.4f}\")\\\n",
    "      .format(history_df['val_loss'].min(), \n",
    "              history_df['val_binary_accuracy'].max()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Using the model to predict for test dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predicted_labels = model.predict(X_test)\n",
    "predicted_labels = np.round(predicted_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final accuracy is in predicting for test dataset is 56.34%\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import mean_absolute_error, accuracy_score\n",
    "\n",
    "accuracy = accuracy_score(y_test, predicted_labels)\n",
    "\n",
    "print(f\"Final accuracy is in predicting for test dataset is {round(accuracy*100, 2)}%\")"
   ]
  }
 ],
 "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": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}