05.02-classificação-sentimentos-liguagem-natural.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "# importação das bibliotecas necessarias\n",
    "\n",
    "from sklearn.metrics import accuracy_score        # acuracia do modelo (avaliação)\n",
    "from sklearn.metrics import confusion_matrix      # matriz confusão (avaliação)\n",
    "from spacy.lang.pt import STOP_WORDS              # stop_word\n",
    "import matplotlib.pyplot as plt                   # visualização   \n",
    "import pandas as pd                               # manipulação dados\n",
    "import numpy as np                                # calculos numericos\n",
    "import seaborn as sns                             # visualização\n",
    "import string                                     # manipulação strings\n",
    "import spacy                                      # liguagem natural\n",
    "import random                                     # valores aleatórios\n",
    "\n",
    "# ignorar os warnings de depreciação de metodos\n",
    "import warnings\n",
    "warnings.simplefilter(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "# carregamento da base de dados\n",
    "\n",
    "treino = pd.read_csv('./data/base_treinamento.txt')\n",
    "bd_teste = pd.read_csv('./data/base_teste.txt')"
   ]
  },
  {
   "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>texto</th>\n",
       "      <th>emocao</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>este trabalho é agradável</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>gosto de ficar no seu aconchego</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>fiz a adesão ao curso hoje</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>eu sou admirada por muitos</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>adoro como você</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             texto   emocao\n",
       "0        este trabalho é agradável  alegria\n",
       "1  gosto de ficar no seu aconchego  alegria\n",
       "2       fiz a adesão ao curso hoje  alegria\n",
       "3       eu sou admirada por muitos  alegria\n",
       "4                  adoro como você  alegria"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# primeiros registros\n",
    "\n",
    "treino.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>texto</th>\n",
       "      <th>emocao</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>não precisei pagar o ingresso</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>se eu ajeitar tudo fica bem</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>minha fortuna ultrapassará a sua</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>sou muito afortunado</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>é benéfico para todos esta nova medida</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                    texto   emocao\n",
       "0           não precisei pagar o ingresso  alegria\n",
       "1             se eu ajeitar tudo fica bem  alegria\n",
       "2        minha fortuna ultrapassará a sua  alegria\n",
       "3                    sou muito afortunado  alegria\n",
       "4  é benéfico para todos esta nova medida  alegria"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# primeiros registros teste\n",
    "\n",
    "bd_teste.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualização dos grupos ['emocao']\n",
    "\n",
    "sns.countplot(treino['emocao'], label='Contagem');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# pre processamento de texto"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'!\"#$%&\\'()*+,-./:;<=>?@[\\\\]^_`{|}~'"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# pontuações \n",
    "\n",
    "pontuacoes = string.punctuation\n",
    "pontuacoes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# stop_words\n",
    "\n",
    "stop_words = STOP_WORDS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<spacy.lang.pt.Portuguese at 0x224f3305ac0>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# modelo em portugues\n",
    "\n",
    "pln = spacy.load('pt')\n",
    "pln"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def processamento(text):\n",
    "    \n",
    "    text = text.lower()                  # todos em minuscula para evitar problemas com a lematização\n",
    "    documento = pln(text)\n",
    "    lista = []\n",
    "\n",
    "    # lematização\n",
    "    for token in documento:\n",
    "        #lista.append(token.text)\n",
    "        lista.append(token.lemma_)\n",
    "\n",
    "    #remoção de stop_words\n",
    "    lista = [palavra for palavra in lista if palavra not in stop_words and palavra not in pontuacoes]\n",
    "    lista = ' '.join([str(elemento) for elemento in lista if not elemento.isdigit()])\n",
    "\n",
    "    return lista\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'aprender liguagem natural python brasilia'"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "teste = processamento('Estou aprendendo liguagem natural com python, em brasilia')\n",
    "teste"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# pre processamento da base dados"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>texto</th>\n",
       "      <th>emocao</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>este trabalho é agradável</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>gosto de ficar no seu aconchego</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>fiz a adesão ao curso hoje</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>eu sou admirada por muitos</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>adoro como você</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             texto   emocao\n",
       "0        este trabalho é agradável  alegria\n",
       "1  gosto de ficar no seu aconchego  alegria\n",
       "2       fiz a adesão ao curso hoje  alegria\n",
       "3       eu sou admirada por muitos  alegria\n",
       "4                  adoro como você  alegria"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# visualização da base de dados\n",
    "\n",
    "treino.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# provessamento na base de dados\n",
    "\n",
    "treino['texto'] = treino['texto'].apply(processamento)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>texto</th>\n",
       "      <th>emocao</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>trabalhar agradável</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>gostar ficar o aconchegar</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>o adesão o o cursar hoje</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>admirar</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>adorar comer</td>\n",
       "      <td>alegria</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       texto   emocao\n",
       "0        trabalhar agradável  alegria\n",
       "1  gostar ficar o aconchegar  alegria\n",
       "2   o adesão o o cursar hoje  alegria\n",
       "3                    admirar  alegria\n",
       "4               adorar comer  alegria"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# visualização após o tratamento\n",
    "\n",
    "treino.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# tratamento das classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "exemplo_base_dados =[['este trabalho é agradavel', {'ALEGRIA':True, 'MEDO':False}],\n",
    "                     'este lugar continua assustador', {'ALEGRIA':False, 'MEDO':True}]            "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['este trabalho é agradavel', {'ALEGRIA': True, 'MEDO': False}]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "exemplo_base_dados[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'este trabalho é agradavel'"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "exemplo_base_dados[0][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'ALEGRIA': True, 'MEDO': False}"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "exemplo_base_dados[0][1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(exemplo_base_dados[0][1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# deixar a base de dados como no exemplo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0      alegria\n",
       "1      alegria\n",
       "2      alegria\n",
       "3      alegria\n",
       "4      alegria\n",
       "        ...   \n",
       "191       medo\n",
       "192       medo\n",
       "193       medo\n",
       "194       medo\n",
       "195       medo\n",
       "Name: emocao, Length: 196, dtype: object"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "treino['emocao']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# processamento da base de dados\n",
    "\n",
    "base_dados_final = []\n",
    "\n",
    "for texto, emocao in zip(treino['texto'], treino['emocao']):\n",
    "    #print(texto, emocao)\n",
    "    if emocao == 'alegria':\n",
    "        dic = ({'ALEGRIA':True, 'MEDO':False})\n",
    "    elif emocao == 'medo':\n",
    "        dic = ({'ALEGRIA':False, 'MEDO':True})\n",
    "    \n",
    "    base_dados_final.append([texto, dic.copy()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['trabalhar agradável', {'ALEGRIA': True, 'MEDO': False}]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# verificação após tratamento\n",
    "\n",
    "base_dados_final[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'ALEGRIA': True, 'MEDO': False}"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "base_dados_final[0][1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(base_dados_final[0][1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# criação do classificador"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# configurações do modelo\n",
    "\n",
    "model = spacy.blank('pt')\n",
    "\n",
    "categorias = model.create_pipe('textcat')\n",
    "categorias.add_label('ALEGRIA')\n",
    "categorias.add_label('MEDO')\n",
    "\n",
    "model.add_pipe(categorias)\n",
    "\n",
    "historico = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'textcat': 0.006111857364885509}\n",
      "{'textcat': 1.1213549733213313e-07}\n",
      "{'textcat': 3.419905469925766e-08}\n"
     ]
    }
   ],
   "source": [
    "model.begin_training()\n",
    "\n",
    "for epoch in range(300):  # teste com 1_000 mostrou que significancia até 300\n",
    "    random.shuffle(base_dados_final)\n",
    "    losses = {}\n",
    "    \n",
    "    for batch in spacy.util.minibatch(base_dados_final, 30):  # quantos registros a atualização dos pesos\n",
    "        \n",
    "        textos = [model(texto) for texto, entities in batch]\n",
    "        \n",
    "        annotations = [{'cats': entities} for texto, entities in batch]\n",
    "        \n",
    "        # atualização dos pesos\n",
    "        model.update(textos, annotations, losses=losses)\n",
    "    \n",
    "    if epoch % 100 == 0:\n",
    "        print(losses)\n",
    "        historico.append(losses)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# criação de lista com erros para visualização\n",
    "\n",
    "historico_loss = []\n",
    "for i in historico:\n",
    "    historico_loss.append(i.get('textcat'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([6.11185736e-03, 1.12135497e-07, 3.41990547e-08])"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# transformação em array\n",
    "\n",
    "historico_loss = np.array(historico_loss)\n",
    "historico_loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Erro')"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualização do erro\n",
    "\n",
    "plt.plot(historico_loss)\n",
    "plt.title('Progressão erro')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Erro')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# salvando modelo\n",
    "\n",
    "model.to_disk('./data/modelo_spacy_LN')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "# carregando modelo\n",
    "\n",
    "modelo_treinado = spacy.load('./data/modelo_spacy_LN')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<spacy.lang.pt.Portuguese at 0x224806582e0>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# verificação \n",
    "\n",
    "modelo_treinado"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "# teste modelo\n",
    "\n",
    "texto_positivo = 'eu adoro a cor dos seus olhos'\n",
    "texto_negativo = 'estou com medo dele'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "# preprocessing\n",
    "\n",
    "texto_positivo = processamento(texto_positivo)\n",
    "texto_negativo = processamento(texto_negativo)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "adorar o cor olho\n",
      "medo d\n"
     ]
    }
   ],
   "source": [
    "print(texto_positivo)\n",
    "print(texto_negativo)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "adorar o cor olho"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# previsão positivo\n",
    "\n",
    "prev = modelo_treinado(texto_positivo)\n",
    "prev"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'ALEGRIA': 0.9976531863212585, 'MEDO': 0.000844139838591218}"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# visualização positivo\n",
    "\n",
    "prev.cats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "medo d"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# previsão negativo\n",
    "\n",
    "prev = modelo_treinado(texto_negativo)\n",
    "prev"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'ALEGRIA': 0.0006764791905879974, 'MEDO': 0.9998795986175537}"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# visualizção negativo\n",
    "\n",
    "prev.cats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "# avaliação do modelo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "# lista com previsões na base de treino\n",
    "\n",
    "previsoes = []\n",
    "\n",
    "for texto in treino['texto']:\n",
    "    #print(texto)\n",
    "    prev_ = modelo_treinado(texto)\n",
    "    previsoes.append(prev_.cats)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'ALEGRIA': 0.9994485974311829, 'MEDO': 0.00011876882490469143},\n",
       " {'ALEGRIA': 0.9970086216926575, 'MEDO': 0.0018794306088238955},\n",
       " {'ALEGRIA': 0.999771773815155, 'MEDO': 0.0003599323390517384},\n",
       " {'ALEGRIA': 0.9994340538978577, 'MEDO': 0.00028366746846586466},\n",
       " {'ALEGRIA': 0.9999088048934937, 'MEDO': 4.539787187241018e-05},\n",
       " {'ALEGRIA': 0.9992040991783142, 'MEDO': 5.979924389976077e-05},\n",
       " {'ALEGRIA': 0.9976531863212585, 'MEDO': 0.000844139838591218},\n",
       " {'ALEGRIA': 0.9992800354957581, 'MEDO': 0.0006117734010331333},\n",
       " {'ALEGRIA': 0.9991470575332642, 'MEDO': 0.0006060979794710875},\n",
       " {'ALEGRIA': 0.9991080164909363, 'MEDO': 0.00017572428623680025},\n",
       " {'ALEGRIA': 0.9993316531181335, 'MEDO': 0.00046976981684565544},\n",
       " {'ALEGRIA': 0.9995914101600647, 'MEDO': 0.00010104260582011193},\n",
       " {'ALEGRIA': 0.9997665286064148, 'MEDO': 0.00013284303713589907},\n",
       " {'ALEGRIA': 0.99964439868927, 'MEDO': 0.00019938245532102883},\n",
       " {'ALEGRIA': 0.9993776679039001, 'MEDO': 0.0001642816059757024},\n",
       " {'ALEGRIA': 0.9992395639419556, 'MEDO': 0.0001757174322847277},\n",
       " {'ALEGRIA': 0.9986667633056641, 'MEDO': 0.0010213104542344809},\n",
       " {'ALEGRIA': 0.9999545812606812, 'MEDO': 4.539787187241018e-05},\n",
       " {'ALEGRIA': 0.9993002414703369, 'MEDO': 0.0004219029506202787},\n",
       " {'ALEGRIA': 0.9988573789596558, 'MEDO': 0.00035863008815795183},\n",
       " {'ALEGRIA': 0.9996234178543091, 'MEDO': 0.0002086824388243258},\n",
       " {'ALEGRIA': 0.9989025592803955, 'MEDO': 0.0004428619868122041},\n",
       " {'ALEGRIA': 0.9992660880088806, 'MEDO': 0.0006729028536938131},\n",
       " {'ALEGRIA': 0.9999374151229858, 'MEDO': 4.539787187241018e-05},\n",
       " {'ALEGRIA': 0.9995991587638855, 'MEDO': 6.48430795990862e-05},\n",
       " {'ALEGRIA': 0.9989801049232483, 'MEDO': 0.0004266900068614632},\n",
       " {'ALEGRIA': 0.9997050166130066, 'MEDO': 7.175179780460894e-05},\n",
       " {'ALEGRIA': 0.9991114735603333, 'MEDO': 0.00025555549655109644},\n",
       " {'ALEGRIA': 0.999548614025116, 'MEDO': 7.178725354606286e-05},\n",
       " {'ALEGRIA': 0.9996753931045532, 'MEDO': 0.0001633224601391703},\n",
       " {'ALEGRIA': 0.9972955584526062, 'MEDO': 0.0018644389929249883},\n",
       " {'ALEGRIA': 0.9994515776634216, 'MEDO': 0.0001422195928171277},\n",
       " {'ALEGRIA': 0.9987050294876099, 'MEDO': 0.0002446387952659279},\n",
       " {'ALEGRIA': 0.9987682700157166, 'MEDO': 0.0005321019561961293},\n",
       " {'ALEGRIA': 0.9994398951530457, 'MEDO': 0.00048797758063301444},\n",
       " {'ALEGRIA': 0.9991284012794495, 'MEDO': 0.0007960856310091913},\n",
       " {'ALEGRIA': 0.9984198808670044, 'MEDO': 0.0007441764464601874},\n",
       " {'ALEGRIA': 0.9988590478897095, 'MEDO': 0.0005032029584981501},\n",
       " {'ALEGRIA': 0.9996121525764465, 'MEDO': 0.00024690802092663944},\n",
       " {'ALEGRIA': 0.9992490410804749, 'MEDO': 0.0004792716645170003},\n",
       " {'ALEGRIA': 0.996753990650177, 'MEDO': 0.001952925231307745},\n",
       " {'ALEGRIA': 0.9989155530929565, 'MEDO': 0.0004100767255295068},\n",
       " {'ALEGRIA': 0.999860405921936, 'MEDO': 4.580070890369825e-05},\n",
       " {'ALEGRIA': 0.9991751313209534, 'MEDO': 0.00020167205366306007},\n",
       " {'ALEGRIA': 0.9992108345031738, 'MEDO': 0.0003113407292403281},\n",
       " {'ALEGRIA': 0.9992682337760925, 'MEDO': 0.0006105731590650976},\n",
       " {'ALEGRIA': 0.9983116388320923, 'MEDO': 0.0005198820726945996},\n",
       " {'ALEGRIA': 0.9991126656532288, 'MEDO': 0.0003958570014219731},\n",
       " {'ALEGRIA': 0.9993662238121033, 'MEDO': 0.0005364312673918903},\n",
       " {'ALEGRIA': 0.999711811542511, 'MEDO': 0.0005089255282655358},\n",
       " {'ALEGRIA': 0.9987867474555969, 'MEDO': 0.0007218241225928068},\n",
       " {'ALEGRIA': 0.9999086856842041, 'MEDO': 0.00011701269977493212},\n",
       " {'ALEGRIA': 0.9993627667427063, 'MEDO': 0.0003434466489125043},\n",
       " {'ALEGRIA': 0.9982567429542542, 'MEDO': 0.0008207857608795166},\n",
       " {'ALEGRIA': 0.9983012080192566, 'MEDO': 0.0005391971208155155},\n",
       " {'ALEGRIA': 0.9987080097198486, 'MEDO': 0.0010237181559205055},\n",
       " {'ALEGRIA': 0.9989457726478577, 'MEDO': 0.001050214166752994},\n",
       " {'ALEGRIA': 0.9993853569030762, 'MEDO': 0.0006308844313025475},\n",
       " {'ALEGRIA': 0.9988325238227844, 'MEDO': 0.0004997851210646331},\n",
       " {'ALEGRIA': 0.9993960857391357, 'MEDO': 0.0009995454456657171},\n",
       " {'ALEGRIA': 0.9990608096122742, 'MEDO': 0.00048540651914663613},\n",
       " {'ALEGRIA': 0.9990783929824829, 'MEDO': 0.0006317333900369704},\n",
       " {'ALEGRIA': 0.9999163150787354, 'MEDO': 4.539787187241018e-05},\n",
       " {'ALEGRIA': 0.999097466468811, 'MEDO': 0.0004939970676787198},\n",
       " {'ALEGRIA': 0.9978564381599426, 'MEDO': 0.001501200022175908},\n",
       " {'ALEGRIA': 0.9993128776550293, 'MEDO': 0.0008337357430718839},\n",
       " {'ALEGRIA': 0.9985525012016296, 'MEDO': 0.0004731407098006457},\n",
       " {'ALEGRIA': 0.9987030029296875, 'MEDO': 0.000503421004395932},\n",
       " {'ALEGRIA': 0.9991180300712585, 'MEDO': 0.0008954048389568925},\n",
       " {'ALEGRIA': 0.9998169541358948, 'MEDO': 0.00018151446420233697},\n",
       " {'ALEGRIA': 0.9994266033172607, 'MEDO': 0.00016600680828560144},\n",
       " {'ALEGRIA': 0.9993738532066345, 'MEDO': 0.00027844851138070226},\n",
       " {'ALEGRIA': 0.9989295601844788, 'MEDO': 0.00015951262321323156},\n",
       " {'ALEGRIA': 0.9985718727111816, 'MEDO': 0.0003607825783547014},\n",
       " {'ALEGRIA': 0.9985713958740234, 'MEDO': 0.0008206985075958073},\n",
       " {'ALEGRIA': 0.9983587861061096, 'MEDO': 0.0007189537864178419},\n",
       " {'ALEGRIA': 0.9989439845085144, 'MEDO': 0.0004521863302215934},\n",
       " {'ALEGRIA': 0.9980903267860413, 'MEDO': 0.0009024324244819582},\n",
       " {'ALEGRIA': 0.9997490048408508, 'MEDO': 6.253593892324716e-05},\n",
       " {'ALEGRIA': 0.9984022974967957, 'MEDO': 0.0009767062729224563},\n",
       " {'ALEGRIA': 0.9989511966705322, 'MEDO': 0.0007346607744693756},\n",
       " {'ALEGRIA': 0.9993773102760315, 'MEDO': 0.00017991458298638463},\n",
       " {'ALEGRIA': 0.9994267225265503, 'MEDO': 0.00041857894393615425},\n",
       " {'ALEGRIA': 0.9996436834335327, 'MEDO': 0.00038313562981784344},\n",
       " {'ALEGRIA': 0.9991542100906372, 'MEDO': 0.0008358553168363869},\n",
       " {'ALEGRIA': 0.9995699524879456, 'MEDO': 0.0006138155004009604},\n",
       " {'ALEGRIA': 0.9993422627449036, 'MEDO': 0.0001561554236104712},\n",
       " {'ALEGRIA': 0.9988096952438354, 'MEDO': 0.0006216321489773691},\n",
       " {'ALEGRIA': 0.9989653825759888, 'MEDO': 0.0002482506970409304},\n",
       " {'ALEGRIA': 0.9980587363243103, 'MEDO': 0.00027714966563507915},\n",
       " {'ALEGRIA': 0.9994650483131409, 'MEDO': 5.9335212426958606e-05},\n",
       " {'ALEGRIA': 0.9989149570465088, 'MEDO': 0.000426309066824615},\n",
       " {'ALEGRIA': 0.9998764991760254, 'MEDO': 4.539787187241018e-05},\n",
       " {'ALEGRIA': 0.9985185265541077, 'MEDO': 0.0005141426809132099},\n",
       " {'ALEGRIA': 0.9998339414596558, 'MEDO': 7.394221756840125e-05},\n",
       " {'ALEGRIA': 0.9986820816993713, 'MEDO': 0.000530171615537256},\n",
       " {'ALEGRIA': 0.9996165037155151, 'MEDO': 0.0004728409112431109},\n",
       " {'ALEGRIA': 0.9985381364822388, 'MEDO': 0.0002980573335662484},\n",
       " {'ALEGRIA': 0.9998169541358948, 'MEDO': 7.619056123076007e-05},\n",
       " {'ALEGRIA': 0.9992586970329285, 'MEDO': 0.00015183845243882388},\n",
       " {'ALEGRIA': 0.9985852241516113, 'MEDO': 0.00019729748601093888},\n",
       " {'ALEGRIA': 0.9990234375, 'MEDO': 0.0006298780790530145},\n",
       " {'ALEGRIA': 0.999298095703125, 'MEDO': 7.805332279531285e-05},\n",
       " {'ALEGRIA': 0.9988325238227844, 'MEDO': 0.0006940133753232658},\n",
       " {'ALEGRIA': 0.9990121126174927, 'MEDO': 0.0007243548170663416},\n",
       " {'ALEGRIA': 0.9977387189865112, 'MEDO': 0.0008743127691559494},\n",
       " {'ALEGRIA': 0.9997944235801697, 'MEDO': 0.00024598740856163204},\n",
       " {'ALEGRIA': 0.9991926550865173, 'MEDO': 0.0007162504480220377},\n",
       " {'ALEGRIA': 0.9992340803146362, 'MEDO': 0.0003310494066681713},\n",
       " {'ALEGRIA': 0.9995172023773193, 'MEDO': 4.539787187241018e-05},\n",
       " {'ALEGRIA': 0.9985734224319458, 'MEDO': 0.001330422586761415},\n",
       " {'ALEGRIA': 0.999680757522583, 'MEDO': 0.00037182099185884},\n",
       " {'ALEGRIA': 0.0002369007415836677, 'MEDO': 0.9999545812606812},\n",
       " {'ALEGRIA': 0.00031901043257676065, 'MEDO': 0.9993233680725098},\n",
       " {'ALEGRIA': 0.00031349179334938526, 'MEDO': 0.9998599290847778},\n",
       " {'ALEGRIA': 0.0007468370022252202, 'MEDO': 0.9993353486061096},\n",
       " {'ALEGRIA': 0.0004371047834865749, 'MEDO': 0.9997782111167908},\n",
       " {'ALEGRIA': 0.0014208785723894835, 'MEDO': 0.9992204904556274},\n",
       " {'ALEGRIA': 0.000574019446503371, 'MEDO': 0.9997684359550476},\n",
       " {'ALEGRIA': 0.001238198601640761, 'MEDO': 0.9999545812606812},\n",
       " {'ALEGRIA': 0.0015342319384217262, 'MEDO': 0.9990034699440002},\n",
       " {'ALEGRIA': 0.0007029862608760595, 'MEDO': 0.9996086955070496},\n",
       " {'ALEGRIA': 0.000492093909997493, 'MEDO': 0.9998123049736023},\n",
       " {'ALEGRIA': 0.0009971951367333531, 'MEDO': 0.9999473094940186},\n",
       " {'ALEGRIA': 0.00042001809924840927, 'MEDO': 0.9993988275527954},\n",
       " {'ALEGRIA': 0.0008079211693257093, 'MEDO': 0.9991469383239746},\n",
       " {'ALEGRIA': 0.000545125687494874, 'MEDO': 0.999671459197998},\n",
       " {'ALEGRIA': 0.0012067804345861077, 'MEDO': 0.9994389414787292},\n",
       " {'ALEGRIA': 0.0011563291773200035, 'MEDO': 0.9997934699058533},\n",
       " {'ALEGRIA': 0.00067197933094576, 'MEDO': 0.999729335308075},\n",
       " {'ALEGRIA': 0.0003449097275733948, 'MEDO': 0.9998069405555725},\n",
       " {'ALEGRIA': 0.0009940246818587184, 'MEDO': 0.9995566010475159},\n",
       " {'ALEGRIA': 0.0005843997350893915, 'MEDO': 0.9996500015258789},\n",
       " {'ALEGRIA': 0.0017063396517187357, 'MEDO': 0.9995204210281372},\n",
       " {'ALEGRIA': 0.0006345764268189669, 'MEDO': 0.9994502663612366},\n",
       " {'ALEGRIA': 0.00026467302814126015, 'MEDO': 0.9995903372764587},\n",
       " {'ALEGRIA': 7.673413347220048e-05, 'MEDO': 0.9999277591705322},\n",
       " {'ALEGRIA': 0.0007045338861644268, 'MEDO': 0.9998652935028076},\n",
       " {'ALEGRIA': 0.00031538729672320187, 'MEDO': 0.9998631477355957},\n",
       " {'ALEGRIA': 0.00118662987370044, 'MEDO': 0.9992836117744446},\n",
       " {'ALEGRIA': 0.0003135677252430469, 'MEDO': 0.9998477697372437},\n",
       " {'ALEGRIA': 0.0009047695784829557, 'MEDO': 0.9993036985397339},\n",
       " {'ALEGRIA': 0.00010683344589779153, 'MEDO': 0.9998650550842285},\n",
       " {'ALEGRIA': 0.0015922709135338664, 'MEDO': 0.9989104270935059},\n",
       " {'ALEGRIA': 0.0007048350526019931, 'MEDO': 0.99970942735672},\n",
       " {'ALEGRIA': 0.00038342372863553464, 'MEDO': 0.9998682737350464},\n",
       " {'ALEGRIA': 0.0009205284877680242, 'MEDO': 0.9995143413543701},\n",
       " {'ALEGRIA': 0.0001478528429288417, 'MEDO': 0.9997156262397766},\n",
       " {'ALEGRIA': 0.00038161338306963444, 'MEDO': 0.9998629093170166},\n",
       " {'ALEGRIA': 0.00018104311311617494, 'MEDO': 0.9999545812606812},\n",
       " {'ALEGRIA': 0.0007221864070743322, 'MEDO': 0.9999126195907593},\n",
       " {'ALEGRIA': 0.00040148888365365565, 'MEDO': 0.9998500347137451},\n",
       " {'ALEGRIA': 0.0005993213271722198, 'MEDO': 0.9995286464691162},\n",
       " {'ALEGRIA': 0.0011386354453861713, 'MEDO': 0.9998170733451843},\n",
       " {'ALEGRIA': 0.003820498241111636, 'MEDO': 0.9977924823760986},\n",
       " {'ALEGRIA': 0.0008963664877228439, 'MEDO': 0.9990767240524292},\n",
       " {'ALEGRIA': 0.0005552554503083229, 'MEDO': 0.9998964071273804},\n",
       " {'ALEGRIA': 0.0008132685907185078, 'MEDO': 0.9994934797286987},\n",
       " {'ALEGRIA': 0.0010961415246129036, 'MEDO': 0.999817430973053},\n",
       " {'ALEGRIA': 0.003717559389770031, 'MEDO': 0.9980440139770508},\n",
       " {'ALEGRIA': 0.0011514982907101512, 'MEDO': 0.9991534948348999},\n",
       " {'ALEGRIA': 0.0004198083479423076, 'MEDO': 0.9999223947525024},\n",
       " {'ALEGRIA': 0.001091932994313538, 'MEDO': 0.999351441860199},\n",
       " {'ALEGRIA': 0.0005851516616530716, 'MEDO': 0.9996355772018433},\n",
       " {'ALEGRIA': 0.0008614048711024225, 'MEDO': 0.9994696974754333},\n",
       " {'ALEGRIA': 0.0007355935522355139, 'MEDO': 0.9993964433670044},\n",
       " {'ALEGRIA': 0.0004484833625610918, 'MEDO': 0.9997705817222595},\n",
       " {'ALEGRIA': 0.00035942948306910694, 'MEDO': 0.9996321201324463},\n",
       " {'ALEGRIA': 0.0011705305660143495, 'MEDO': 0.9998645782470703},\n",
       " {'ALEGRIA': 0.0010640397667884827, 'MEDO': 0.9996335506439209},\n",
       " {'ALEGRIA': 0.0014753437135368586, 'MEDO': 0.9990721940994263},\n",
       " {'ALEGRIA': 0.00027845328440889716, 'MEDO': 0.9994168281555176},\n",
       " {'ALEGRIA': 0.0012984878849238157, 'MEDO': 0.9989631175994873},\n",
       " {'ALEGRIA': 0.0009042632300406694, 'MEDO': 0.9996826648712158},\n",
       " {'ALEGRIA': 6.691506132483482e-05, 'MEDO': 0.9999494552612305},\n",
       " {'ALEGRIA': 0.0014072994235903025, 'MEDO': 0.9995595812797546},\n",
       " {'ALEGRIA': 0.001489030197262764, 'MEDO': 0.9994602799415588},\n",
       " {'ALEGRIA': 0.0010780652519315481, 'MEDO': 0.9987909197807312},\n",
       " {'ALEGRIA': 0.001520373858511448, 'MEDO': 0.999714195728302},\n",
       " {'ALEGRIA': 0.003163154236972332, 'MEDO': 0.9985766410827637},\n",
       " {'ALEGRIA': 0.0011522526619955897, 'MEDO': 0.9990431666374207},\n",
       " {'ALEGRIA': 0.0005057667149230838, 'MEDO': 0.9999080896377563},\n",
       " {'ALEGRIA': 0.0023842244409024715, 'MEDO': 0.9989591836929321},\n",
       " {'ALEGRIA': 0.0005896877264603972, 'MEDO': 0.9999315738677979},\n",
       " {'ALEGRIA': 0.0002487100719008595, 'MEDO': 0.9999475479125977},\n",
       " {'ALEGRIA': 0.0010257296962663531, 'MEDO': 0.99989914894104},\n",
       " {'ALEGRIA': 0.0010259427363052964, 'MEDO': 0.9996597766876221},\n",
       " {'ALEGRIA': 0.0011662396136671305, 'MEDO': 0.9994069337844849},\n",
       " {'ALEGRIA': 0.001420450396835804, 'MEDO': 0.9993513226509094},\n",
       " {'ALEGRIA': 0.0013213155325502157, 'MEDO': 0.9997100234031677},\n",
       " {'ALEGRIA': 0.0008676526485942304, 'MEDO': 0.9984740614891052},\n",
       " {'ALEGRIA': 0.0016949542332440615, 'MEDO': 0.9989197254180908},\n",
       " {'ALEGRIA': 0.0004897686303593218, 'MEDO': 0.999729335308075},\n",
       " {'ALEGRIA': 0.0022974342573434114, 'MEDO': 0.9993436932563782},\n",
       " {'ALEGRIA': 0.0022374687250703573, 'MEDO': 0.9991162419319153},\n",
       " {'ALEGRIA': 0.0014254628913477063, 'MEDO': 0.999760091304779}]"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# visualização previsoes\n",
    "\n",
    "previsoes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "# transformação das previsoes para avaliação\n",
    "\n",
    "previsoes_final = []\n",
    "\n",
    "for previsao in previsoes:\n",
    "\n",
    "    if previsao['ALEGRIA'] > previsao['MEDO']:\n",
    "        previsoes_final.append('alegria')\n",
    "    \n",
    "    elif previsao['MEDO'] > previsao['ALEGRIA']:\n",
    "        previsoes_final.append('medo')\n",
    "\n",
    "previsoes_final = np.array(previsoes_final)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo'], dtype='<U7')"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# viz após a transformação\n",
    "\n",
    "previsoes_final"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'alegria', 'alegria',\n",
       "       'alegria', 'alegria', 'alegria', 'alegria', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo', 'medo',\n",
       "       'medo'], dtype=object)"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# transformação para comparação\n",
    "\n",
    "respostas_reais = treino['emocao'].values\n",
    "respostas_reais"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# acuracia do modelo\n",
    "\n",
    "accuracy_score(previsoes_final, respostas_reais)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# viz matriz de confusão\n",
    "\n",
    "cnn = confusion_matrix(previsoes_final, respostas_reais)\n",
    "sns.heatmap(cnn, annot=True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "# previsões com teste"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "# processamento na base de teste\n",
    "\n",
    "bd_teste['texto'] = bd_teste['texto'].apply(processamento)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "# respostas na base de teste\n",
    "\n",
    "previsoes = [ ]\n",
    "\n",
    "for texto in bd_teste['texto']:\n",
    "\n",
    "    previsao = modelo_treinado(texto)\n",
    "    previsoes.append(previsao.cats)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "# tratamento das previsões\n",
    "\n",
    "previsoes_final_test = []\n",
    "\n",
    "for previsao in previsoes:\n",
    "\n",
    "    if previsao['ALEGRIA'] > previsao['MEDO']:\n",
    "        previsoes_final_test.append('alegria')\n",
    "    \n",
    "    elif previsao['MEDO'] > previsao['ALEGRIA']:\n",
    "        previsoes_final_test.append('medo')\n",
    "\n",
    "previsoes_final_test = np.array(previsoes_final_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "# transformação valores teste em array \n",
    "\n",
    "respostas_reais_test = bd_teste['emocao'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "# avaliação"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.5238095238095238"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# acuracia do modelo na base de teste\n",
    "\n",
    "accuracy_score(respostas_reais_test, previsoes_final_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# matriz de confusão\n",
    "\n",
    "cnn = confusion_matrix(respostas_reais_test, previsoes_final_test)\n",
    "sns.heatmap(cnn, annot=True);"
   ]
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "ad2bdc8ecc057115af97d19610ffacc2b4e99fae6737bb82f5d7fb13d2f2c186"
  },
  "kernelspec": {
   "display_name": "Python 3.8.8 ('base')",
   "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"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}