Trabalho Final.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Trabalho Final da disciplina de Machine Learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tipo de problema\n",
    "NER - Named Entity Recognition.\n",
    "\n",
    "Neste tipo de problema um modelo de predição para textos tentará predizer da melhor maneira possível quais são as entidades em um texto. Os problemas de NER surgem devido diversas maneiras de expressar uma determinada coisa e também onde uma coisa pode significar várias outras.\n",
    "A Língua Portuguesa por exemplo possui muitos tempos verbais e maneiras de conjugação de verbos.\n",
    "\n",
    "Ex: conjugar, conjugado, conjugaria, conjugarei, conjugaríamos, conjugaríeis, conjugaste, conjugou, conjuguemos...\n",
    "\n",
    "Tente explicar por exemplo para um extrangeiro a diferença entre \"bota a calça e calça a bota\"."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Problema\n",
    "O problema a ser resolvido com a resolução deste trabalho é utilizar um dataset com milhares de textos e palavras rotulados e a partir de um modelo de predições de texto, realizar o ranqueamento das melhores classificações de entidades.\n",
    "\n",
    "Nós nos focaremos em 4 tipos de named entities: persons, locations, organizations e nomes de demais entidades que não pertençam as anteriores."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Datasets\n",
    "Será utilizado um dataset amplamente difundido que foi utilizado primariamente na Conference on Computational Natural Language Learning (CoNLL-2003) acessível a partir do seguinte link: https://www.clips.uantwerpen.be/conll2003/ner/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Divisão do dataset - rows\n",
    "Os arquivos de dados de tarefas compartilhadas CoNLL-2003 contêm quatro colunas separadas por um único espaço.\n",
    "\n",
    "- O primeiro item de cada linha no dataset é uma palavra.  \n",
    "- O segundo é um Part-of-Speech (POS) tag. \n",
    "- O terceiro é uma tag de fragmento sintático. \n",
    "- A quarta é a tag de entidade nomeada.\n",
    "\n",
    "As tags de fragmento e os nomes de entidades tem o formato I-TYPE que significa que a palavra está dentro de uma frase do tipo TYPE.\n",
    "\n",
    " Uma palavra com a tag O não é parte de uma frase."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ferramenta\n",
    "Será utilizado o [Spacy](https://spacy.io/)\n",
    " que é uma ferramenta para o emprego de técnicas de Processamento de Linguagem Natural - PLN. Será utilizado voltado ao NER - Named Entity Recognition.\n",
    "\n",
    "O Spacy é uma ferramenta de código aberto muito poderosa que já possui diversos modelos prontos para uso para diversos idiomas.\n",
    "\n",
    "Será utilizado o modelo para classificação de textos em inglês.\n",
    "\n",
    "Algumas funções do spacy estão definidas mais abaixo no documento."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#U.N.         NNP  I-NP  I-ORG \n",
    "#official     NN   I-NP  O \n",
    "#Ekeus        NNP  I-NP  I-PER \n",
    "#heads        VBZ  I-VP  O \n",
    "#for          IN   I-PP  O \n",
    "#Baghdad      NNP  I-NP  I-LOC \n",
    "#.            .    O     O "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Divisão dos arquivos de dados\n",
    "A divisão dos arquivos do dataset se dão da seguinte forma:\n",
    "    - Consiste de 3 arquivos por linguagem\n",
    "    - Um arquivo de treino\n",
    "    - Dois arquivos de teste, testeA e testeB\n",
    "    - O primeiro arquivo de teste será usado em produção para encontrar os melhores parametros\n",
    "    - O segundo arquivo de teste será usado para a avaliação final\n",
    "    - Os dados estão disponíveis em dois datasets, um em Inglês e também em Alemão. Para o propósito deste trabalho será usado apenas a versão em inglês.\n",
    "   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Arquivos\n",
    "- eng.raw.tar   - 13,930 MB - Conjunto de dados em inglês\n",
    "- ner.tgz       - 3,374  MB - Contém o software para fazer o build dos dados\n",
    "- 000README.txt - 8      KB - Instruções para descompactação\n",
    "\n",
    "Os dados em inglês são uma coleção de artigos de notícias do Reuters Corpus."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Métricas de Avaliação\n",
    "A competição utiliza três métricas principais que são:\n",
    "- Precision\n",
    "- Recall\n",
    "- F-Score\n",
    "\n",
    "Precision é a porcentagem de named entities encontradas pelo sistema de aprendizado que estão corretas.\n",
    "\n",
    "Recall é a porcentagem de named entities presentes no Corpus que são encontradas pelo sistema.\n",
    "\n",
    "Uma named entity só está correta se for uma correspondência exata da entidade correspondente no\n",
    "arquivo de dados."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dados de treino: Exemplos e suas anotações.\n",
    "\n",
    "### Texto: O texto de entrada para o qual o modelo deve prever um rótulo.\n",
    "\n",
    "### Label: O label que o modelo deve predizer\n",
    "\n",
    "### Gradient: Gradient da função de perca calculando a diferença entre a entrada e a saída esperada"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import Image\n",
    "Image(filename=\"training model.png\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Importando bibliotecas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [],
   "source": [
    "import spacy\n",
    "from spacy import displacy\n",
    "from spacy.tokens import Span, Doc\n",
    "from spacy.vocab import Vocab\n",
    "from spacy.pipeline import EntityRuler\n",
    "from spacy.lang.en import English\n",
    "\n",
    "from sklearn.metrics import classification_report\n",
    "\n",
    "import numpy as np \n",
    "import random"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "nlp = spacy.load(\"en_core_web_sm\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div class=\"entities\" style=\"line-height: 2.5; direction: ltr\">\n",
       "<mark class=\"entity\" style=\"background: #7aecec; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em; box-decoration-break: clone; -webkit-box-decoration-break: clone\">\n",
       "    Apple\n",
       "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; text-transform: uppercase; vertical-align: middle; margin-left: 0.5rem\">ORG</span>\n",
       "</mark>\n",
       " is looking at buying \n",
       "<mark class=\"entity\" style=\"background: #feca74; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em; box-decoration-break: clone; -webkit-box-decoration-break: clone\">\n",
       "    U.K.\n",
       "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; text-transform: uppercase; vertical-align: middle; margin-left: 0.5rem\">GPE</span>\n",
       "</mark>\n",
       " startup for \n",
       "<mark class=\"entity\" style=\"background: #e4e7d2; padding: 0.45em 0.6em; margin: 0 0.25em; line-height: 1; border-radius: 0.35em; box-decoration-break: clone; -webkit-box-decoration-break: clone\">\n",
       "    $1 billion\n",
       "    <span style=\"font-size: 0.8em; font-weight: bold; line-height: 1; border-radius: 0.35em; text-transform: uppercase; vertical-align: middle; margin-left: 0.5rem\">MONEY</span>\n",
       "</mark>\n",
       "</div>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "doc = nlp(u\"Apple is looking at buying U.K. startup for $1 billion\")\n",
    "spacy.displacy.render(doc, style='ent', jupyter=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Jonas Lopes\\Anaconda3\\lib\\runpy.py:193: UserWarning: [W011] It looks like you're calling displacy.serve from within a Jupyter notebook or a similar environment. This likely means you're already running a local web server, so there's no need to make displaCy start another one. Instead, you should be able to replace displacy.serve with displacy.render to show the visualization.\n",
      "  \"__main__\", mod_spec)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<!DOCTYPE html>\n",
       "<html lang=\"en\">\n",
       "    <head>\n",
       "        <title>displaCy</title>\n",
       "    </head>\n",
       "\n",
       "    <body style=\"font-size: 16px; font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif, 'Apple Color Emoji', 'Segoe UI Emoji', 'Segoe UI Symbol'; padding: 4rem 2rem; direction: ltr\">\n",
       "<figure style=\"margin-bottom: 6rem\">\n",
       "<svg xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\" xml:lang=\"en\" id=\"577214f5f2b7446ba4faef2d08af4f77-0\" class=\"displacy\" width=\"750\" height=\"312.0\" direction=\"ltr\" style=\"max-width: none; height: 312.0px; color: #000000; background: #ffffff; font-family: Arial; direction: ltr\">\n",
       "<text class=\"displacy-token\" fill=\"currentColor\" text-anchor=\"middle\" y=\"222.0\">\n",
       "    <tspan class=\"displacy-word\" fill=\"currentColor\" x=\"50\">This</tspan>\n",
       "    <tspan class=\"displacy-tag\" dy=\"2em\" fill=\"currentColor\" x=\"50\">DET</tspan>\n",
       "</text>\n",
       "\n",
       "<text class=\"displacy-token\" fill=\"currentColor\" text-anchor=\"middle\" y=\"222.0\">\n",
       "    <tspan class=\"displacy-word\" fill=\"currentColor\" x=\"225\">is</tspan>\n",
       "    <tspan class=\"displacy-tag\" dy=\"2em\" fill=\"currentColor\" x=\"225\">VERB</tspan>\n",
       "</text>\n",
       "\n",
       "<text class=\"displacy-token\" fill=\"currentColor\" text-anchor=\"middle\" y=\"222.0\">\n",
       "    <tspan class=\"displacy-word\" fill=\"currentColor\" x=\"400\">a</tspan>\n",
       "    <tspan class=\"displacy-tag\" dy=\"2em\" fill=\"currentColor\" x=\"400\">DET</tspan>\n",
       "</text>\n",
       "\n",
       "<text class=\"displacy-token\" fill=\"currentColor\" text-anchor=\"middle\" y=\"222.0\">\n",
       "    <tspan class=\"displacy-word\" fill=\"currentColor\" x=\"575\">sentence.</tspan>\n",
       "    <tspan class=\"displacy-tag\" dy=\"2em\" fill=\"currentColor\" x=\"575\">NOUN</tspan>\n",
       "</text>\n",
       "\n",
       "<g class=\"displacy-arrow\">\n",
       "    <path class=\"displacy-arc\" id=\"arrow-577214f5f2b7446ba4faef2d08af4f77-0-0\" stroke-width=\"2px\" d=\"M70,177.0 C70,89.5 220.0,89.5 220.0,177.0\" fill=\"none\" stroke=\"currentColor\"/>\n",
       "    <text dy=\"1.25em\" style=\"font-size: 0.8em; letter-spacing: 1px\">\n",
       "        <textPath xlink:href=\"#arrow-577214f5f2b7446ba4faef2d08af4f77-0-0\" class=\"displacy-label\" startOffset=\"50%\" side=\"left\" fill=\"currentColor\" text-anchor=\"middle\">nsubj</textPath>\n",
       "    </text>\n",
       "    <path class=\"displacy-arrowhead\" d=\"M70,179.0 L62,167.0 78,167.0\" fill=\"currentColor\"/>\n",
       "</g>\n",
       "\n",
       "<g class=\"displacy-arrow\">\n",
       "    <path class=\"displacy-arc\" id=\"arrow-577214f5f2b7446ba4faef2d08af4f77-0-1\" stroke-width=\"2px\" d=\"M420,177.0 C420,89.5 570.0,89.5 570.0,177.0\" fill=\"none\" stroke=\"currentColor\"/>\n",
       "    <text dy=\"1.25em\" style=\"font-size: 0.8em; letter-spacing: 1px\">\n",
       "        <textPath xlink:href=\"#arrow-577214f5f2b7446ba4faef2d08af4f77-0-1\" class=\"displacy-label\" startOffset=\"50%\" side=\"left\" fill=\"currentColor\" text-anchor=\"middle\">det</textPath>\n",
       "    </text>\n",
       "    <path class=\"displacy-arrowhead\" d=\"M420,179.0 L412,167.0 428,167.0\" fill=\"currentColor\"/>\n",
       "</g>\n",
       "\n",
       "<g class=\"displacy-arrow\">\n",
       "    <path class=\"displacy-arc\" id=\"arrow-577214f5f2b7446ba4faef2d08af4f77-0-2\" stroke-width=\"2px\" d=\"M245,177.0 C245,2.0 575.0,2.0 575.0,177.0\" fill=\"none\" stroke=\"currentColor\"/>\n",
       "    <text dy=\"1.25em\" style=\"font-size: 0.8em; letter-spacing: 1px\">\n",
       "        <textPath xlink:href=\"#arrow-577214f5f2b7446ba4faef2d08af4f77-0-2\" class=\"displacy-label\" startOffset=\"50%\" side=\"left\" fill=\"currentColor\" text-anchor=\"middle\">attr</textPath>\n",
       "    </text>\n",
       "    <path class=\"displacy-arrowhead\" d=\"M575.0,179.0 L583.0,167.0 567.0,167.0\" fill=\"currentColor\"/>\n",
       "</g>\n",
       "</svg>\n",
       "</figure>\n",
       "</body>\n",
       "</html>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Using the 'dep' visualizer\n",
      "Serving on http://0.0.0.0:5000 ...\n",
      "\n",
      "Shutting down server on port 5000.\n"
     ]
    }
   ],
   "source": [
    "doc_dep = nlp(u\"This is a sentence.\")\n",
    "displacy.serve(doc_dep, style=\"dep\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Reconheça e atualize named entities"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "FB 0 2 ORG\n"
     ]
    }
   ],
   "source": [
    "doc = nlp(u\"FB is hiring a new VP of global policy\")\n",
    "doc.ents = [Span(doc, 0, 1, label=doc.vocab.strings[u\"ORG\"])] # onde 0 é a posição inícial e 1 é a posição final da string\n",
    "for ent in doc.ents:\n",
    "    print(ent.text, ent.start_char, ent.end_char, ent.label_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Treinando e atualizando modelos de rede neural"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nlp = spacy.load(\"en_core_web_sm\")\n",
    "train_data = [(u\"Uber blew through $1 million\", {\"entities\": [(0, 4, \"ORG\")]})]\n",
    "\n",
    "other_pipes = [pipe for pipe in nlp.pipe_names if pipe != \"ner\"]\n",
    "\n",
    "with nlp.disable_pipes(*other_pipes):\n",
    "    optimizer = nlp.begin_training()\n",
    "    for i in range(10):\n",
    "        random.shuffle(train_data)\n",
    "        for text, annotations in train_data:\n",
    "            nlp.update([text], [annotations], sgd=optimizer)\n",
    "            \n",
    "# salvando o modelo no disco         \n",
    "nlp.to_disk(\"models/modelo.bin\")\n",
    "# trazendo o modelo de volta ao disco\n",
    "nlp = spacy.load(\"models/modelo.bin\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# utilizando o novo modelo\n",
    "eng_testa = open(\"dados/eng.testb.txt\").read()\n",
    "doc = nlp(eng_testa)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Serialização simples e eficiente"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from spacy.tokens import Doc\n",
    "from spacy.vocab import Vocab\n",
    "\n",
    "nlp = spacy.load(\"en_core_web_sm\")\n",
    "saida = open(\"dados/eng.testb.txt\").read()\n",
    "doc = nlp(saida)\n",
    "\n",
    "doc.to_disk(\"dados/saida.bin\")\n",
    "\n",
    "new_doc = Doc(Vocab()).from_disk(\"dados/saida.bin\")\n",
    "\n",
    "new_doc.ents"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "spacy.explain(\"dobj\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Matcher Pattern"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "New York\n"
     ]
    }
   ],
   "source": [
    "# Matcher para palavras\n",
    "\n",
    "# Matcher is initialized with the shared vocab\n",
    "from spacy.matcher import Matcher\n",
    "\n",
    "# Each dict represents one token and its attributes\n",
    "matcher = Matcher(nlp.vocab)\n",
    "\n",
    "# Add with ID, optional callback and pattern(s)\n",
    "pattern = [{\"LOWER\": \"new\"}, {\"LOWER\": \"york\"}]\n",
    "matcher.add('CITIES', None, pattern)\n",
    "\n",
    "# Match by calling the matcher on a Doc object\n",
    "doc = nlp(\"I live in New York\")\n",
    "matches = matcher(doc)\n",
    "\n",
    "# Matches are (match_id, start, end) tuples\n",
    "for match_id, start, end in matches:\n",
    "     # Get the matched span by slicing the Doc\n",
    "    span = doc[start:end]\n",
    "    print(span.text)\n",
    "# 'New York'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Utilizando Match Pattern para criar um modelo de NER com Spacy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 385,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "818268"
      ]
     },
     "execution_count": 385,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train = open(\"dados/eng.train.txt\").read()\n",
    "train = train.split()\n",
    "len(train)\n",
    "# O spacy só aceita por vez 100000 caracteres"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 386,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "#Tentativa com arquivo de treino\n",
    "nlp = English();\n",
    "ruler = EntityRuler(nlp)\n",
    "patterns = []\n",
    "i = 0\n",
    "\n",
    "while(i <= len(train)-4):\n",
    "    if train[i] == '-DOCSTART-':\n",
    "        i += 4\n",
    "    #print(\"Pattern\",train[i] + \" Label\",train[i+3])   \n",
    "    patterns += [{\"label\": train[i+3], \"pattern\" : train[i]}]\n",
    "\n",
    "    i += 4\n",
    "\n",
    "ruler.add_patterns(patterns)\n",
    "\n",
    "#adicionando o pipe ruler dentro do nlp\n",
    "nlp.add_pipe(ruler)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 387,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Enviando o ruler para o disco e salvando-o.\n",
    "ruler.to_disk(\"models/patterns.json\")\n",
    "nlp.to_disk(\"models/modelo.bin\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Adicionando pipe com um modelo do zero\n",
    "nlp = English()\n",
    "\n",
    "ruler_disk = EntityRuler(nlp).from_disk(\"models/patterns.json\")\n",
    "\n",
    "nlp.add_pipe(ruler_disk)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# carregando o modelo criado a partir do disco\n",
    "nlp = spacy.load(\"models/modelo.bin\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[('blew', 'O'), ('through', 'O'), ('$', 'O'), ('1', 'I-MISC'), ('million', 'O')]\n"
     ]
    }
   ],
   "source": [
    "doc = nlp(u\"Uber blew through $1 million\")\n",
    "print([(ent.text, ent.label_) for ent in doc.ents])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[('Peter', 'I-PER'), ('is', 'I-MISC'), ('a', 'O'), ('good', 'O'), ('guy', 'O')]\n"
     ]
    }
   ],
   "source": [
    "doc = nlp(u\"Peter is a good guy\")\n",
    "print([(ent.text, ent.label_) for ent in doc.ents])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Avaliando com as métricas do CoNLL"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "O CoNNL trabalha com o EM - Exact Match utilizando a precision e recall\n",
    " - Onde precision como mostrado no início , é a porcentagem de NEs encontradas pelo sistema que estão corretas\n",
    " - e o recall que é a porcentagem de NEs na solução que são encontradas pelo sistema.\n",
    " - Uma NE só está correta se é uma combinação exata da entidade na solução."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- ### Antes de mais nada, um exemplo para melhor entendimento"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Digamos que tenhamos as seguintes anotações de treino corretas\n",
    "\n",
    "Unlike <ENAMEX TYPE=\"PERSON\">Robert</ENAMEX>, <ENAMEX TYPE=\"PERSON\">John\n",
    "Briggs Jr</ENAMEX> contacted <ENAMEX TYPE=\"ORGANIZATION\">Wonderful\n",
    "Stockbrockers Inc</ENAMEX> in <ENAMEX TYPE=\"LOCATION\">New York</ENAMEX> and\n",
    "instructed them to sell all his shares in <ENAMEX\n",
    "TYPE=\"ORGANIZATION\">Acme</ENAMEX>. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vamos supor que um sistema produziu as seguintes saídas\n",
    "\n",
    "<ENAMEX TYPE=\"LOCATION\">Unlike</ENAMEX> Robert, <ENAMEX\n",
    "TYPE=\"ORGANIZATION\">John Briggs Jr</ENAMEX> contacted Wonderful <ENAMEX\n",
    "TYPE=\"ORGANIZATION\">Stockbrockers</ENAMEX> Inc <TIMEX TYPE=\"DATE\">in New\n",
    "York</TIMEX> and instructed them to sell all his shares in <ENAMEX\n",
    "TYPE=\"ORGANIZATION\">Acme</ENAMEX>. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- ### O sistema gera os seguintes erros:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 16,
     "metadata": {
      "image/png": {
       "height": 500,
       "width": 700
      }
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(filename=\"saidas eplicar.png\", width=700, height=500)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Para o exemplo acima, existem 5 entidades verdadeiras, 5 palpites do sistema onde somente um desses palpites combina exatamente com a solução:  <Organization> Acme </Organization>\n",
    "A precision é então 20% (1/5)\n",
    "A recall é também 20% (1/5) "
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "F-score\n",
    "\n",
    "F1 = 2 * (precision * recall) / (precision + recall)\n",
    "F1 = 2 * (20        *     20) / (20        +     20)\n",
    "F1 = 20"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Utilizando as métricas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Primeiro vamos definir o texto em uma string normal Python para entao manipularmos com o nosso modelo criado"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 258,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "testa =     u\"\"\"Their PRP$ I-NP O\n",
    "                stay NN I-NP O\n",
    "                on IN I-PP O\n",
    "                top NN I-NP O\n",
    "                , , O O\n",
    "                though RB I-ADVP O\n",
    "                , , O O\n",
    "                may MD I-VP O\n",
    "                be VB I-VP O\n",
    "                short-lived JJ I-ADJP O\n",
    "                as IN I-PP O\n",
    "                title NN I-NP O\n",
    "                rivals NNS I-NP O\n",
    "                Essex NNP I-NP I-ORG\n",
    "                , , O O\n",
    "                Derbyshire NNP I-NP I-ORG\n",
    "                and CC I-NP O\n",
    "                Surrey NNP I-NP I-ORG\n",
    "                all DT O O\n",
    "                closed VBD I-VP O\n",
    "                in RP I-PRT O\n",
    "                on IN I-PP O\n",
    "                victory NN I-NP O\n",
    "                while IN I-SBAR O\n",
    "                Kent NNP I-NP I-ORG\n",
    "                made VBD I-VP O\n",
    "                up RP I-PRT O\n",
    "                for IN I-PP O\n",
    "                lost VBN I-NP O\n",
    "                time NN I-NP O\n",
    "                in IN I-PP O\n",
    "                their PRP$ I-NP O\n",
    "                rain-affected JJ I-NP O\n",
    "                match NN I-NP O\n",
    "                against IN I-PP O\n",
    "                Nottinghamshire NNP I-NP I-ORG\n",
    "                . . O O\"\"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "metadata": {},
   "outputs": [],
   "source": [
    "testa = open(\"dados/eng.testa.txt\").read()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Após isso utilizamos a função split para realizar a separação dos tokens, onde cada palavra será separada e adicionada a uma lista Python"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 276,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "206312"
      ]
     },
     "execution_count": 276,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "testa = testa.split()\n",
    "len(testa)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Feito isto poderemos utilizar um numpy array para facilitar a comparação a posteriori das entidades corretas do arquivo de teste, com as entidades preditas pelo nosso modelo\n",
    "- Para isso iremos ignorar as palavras marcadas como 'O', já que elas não sao entidades"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 277,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Entidades corretas do dado de teste\n",
    "correct_ent = [['                                       ','                                       ']]\n",
    "correct_ent = np.array(correct_ent)\n",
    "correct_ent = np.delete(correct_ent, np.s_[:1], axis=0)\n",
    "\n",
    "i = 0\n",
    "\n",
    "while(i <= len(testa)-4):\n",
    "    if testa[i] == '-DOCSTART-':\n",
    "        i += 4\n",
    "        \n",
    "    if testa[i+3] != 'O':   \n",
    "        correct_ent = np.insert(correct_ent, len(correct_ent), [testa[i], testa[i+3]], axis=0)\n",
    "\n",
    "    i += 4"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Temos agora um numpy array com as entidades corretas que serão utilizadas para comparação com as predições\n",
    "- Na célula abaixo iremos criar uma string pegando apenas a primeira posição de cada linha ou seja, a palavra. Colocando todas as palavras em uma string criaremos um \"documento\", assim chamado pelo spacy que será usado como entrada para o modelo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 283,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Colocando o texto em uma String para poder usar nlp do spacy\n",
    "testa_doc = ''\n",
    "\n",
    "i = 0\n",
    "\n",
    "while(i <= len(testa)-4):\n",
    "    if testa[i] == '-DOCSTART-':\n",
    "        i += 4\n",
    "        \n",
    "    testa_doc += testa[i]\n",
    "    testa_doc += ' '\n",
    "        \n",
    "    i += 4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 292,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'CRICKET - LEICESTERSHIRE TAKE OVER AT TOP AFTER INNINGS VICTORY . LONDON 1996-08-30 West Indian all-rounder Phil Simmons took four for 38 on Friday as Leicestershire beat Somerset by an innings and 39 runs in two days to take over at the head of the county championship . Their stay on top , though , may be short-lived as title rivals Essex , Derbyshire and Surrey all closed in on victory while Kent made up for lost time in their rain-affected match against Nottinghamshire . After bowling Somerse'"
      ]
     },
     "execution_count": 292,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 500 primeiros caracteres\n",
    "testa_doc[:500]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 291,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "275559"
      ]
     },
     "execution_count": 291,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# tamanho da String\n",
    "len(testa_doc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Chamamos então a função  nlp que é responsável por atribuir a um documento as palavras e suas entidades preditas pelo modelo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 293,
   "metadata": {},
   "outputs": [],
   "source": [
    "doc = nlp(testa_doc)\n",
    "predicted_ent = [(ent.text, ent.label_) for ent in doc.ents]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 294,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "('CRICKET', 'O')"
      ]
     },
     "execution_count": 294,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "doc.ents[0].text, doc.ents[0].label_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- O modelo foi treinando utilizando as anotações 'O', como dito acima não precisaremos delas já que não são entidades.\n",
    "- Utilizaremos um laço para remover as palavras anotadas com 'O' e adicionaremos as restantes em um numpy array do mesmo tipo das entidades corretas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui\n",
      "aqui"
     ]
    }
   ],
   "source": [
    "# Transformando as entidades preditas no mesmo formato das entidades corretas\n",
    "predicted_ent = [['                                       ','                                       ']]\n",
    "predicted_ent = np.array(predicted_ent)\n",
    "predicted_ent = np.delete(predicted_ent, np.s_[:1], axis=0)\n",
    "\n",
    "i = 0\n",
    "\n",
    "while(i <= len(doc.ents)-1):\n",
    "        \n",
    "    if doc.ents[i].label_ != 'O':   \n",
    "        predicted_ent = np.insert(predicted_ent, len(predicted_ent), [doc.ents[i].text, doc.ents[i].label_], axis=0)\n",
    "        print(\"aqui\")\n",
    "    i += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 267,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([['Essex', 'I-ORG'],\n",
       "       ['Derbyshire', 'I-ORG'],\n",
       "       ['Surrey', 'I-ORG'],\n",
       "       ['Kent', 'I-ORG'],\n",
       "       ['Nottinghamshire', 'I-ORG']], dtype='<U39')"
      ]
     },
     "execution_count": 267,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "correct_ent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 268,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([['on', 'I-MISC'],\n",
       "       [',', 'I-LOC'],\n",
       "       [',', 'I-LOC'],\n",
       "       ['Essex', 'I-LOC'],\n",
       "       [',', 'I-LOC'],\n",
       "       ['Derbyshire', 'I-ORG'],\n",
       "       ['and', 'I-LOC'],\n",
       "       ['Surrey', 'I-LOC'],\n",
       "       ['on', 'I-MISC'],\n",
       "       ['Kent', 'I-ORG'],\n",
       "       ['for', 'I-MISC'],\n",
       "       ['against', 'I-MISC'],\n",
       "       ['Nottinghamshire', 'I-ORG'],\n",
       "       ['.', 'I-LOC']], dtype='<U39')"
      ]
     },
     "execution_count": 268,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predicted_ent"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Agora iremos calcular a precision e a recall a partir das duas listas: correct_ent e predicted_ent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 118,
     "metadata": {
      "image/png": {
       "height": 500,
       "width": 500
      }
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(filename=\"precision.png\", width=500, height=500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 269,
   "metadata": {},
   "outputs": [],
   "source": [
    "TP = 0\n",
    "FP = 0\n",
    "FN = 0\n",
    "\n",
    "for i in range(0, len(correct_ent)):\n",
    "    for j in range(0, len(predicted_ent)):\n",
    "        compare = correct_ent[i] == predicted_ent[j]\n",
    "        if compare[0] and compare[1]:\n",
    "            TP += 1\n",
    "        elif compare[0] == True and compare[1] == False:\n",
    "            FP += 1\n",
    "            \n",
    "FN = len(predicted_ent) - TP - FP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 270,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([False,  True])"
      ]
     },
     "execution_count": 270,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compare = correct_ent[0] == predicted_ent[9]\n",
    "compare"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "True Positives: 3\n",
      "False Positives: 2\n",
      "False Negatives: 9\n"
     ]
    }
   ],
   "source": [
    "print(\"True Positives:\",TP)\n",
    "print(\"False Positives:\",FP)\n",
    "print(\"False Negatives:\",FN)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 272,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6"
      ]
     },
     "execution_count": 272,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "precision = 0\n",
    "precision = TP / (TP + FP)\n",
    "precision"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Cálculo do Recall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmQAAABxCAYAAAB7lb+9AAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAAAJcEhZcwAAEnQAABJ0Ad5mH3gAAFIOSURBVHhe7d0FVBRr48dxVETEQMXu7sDu7sDu7u5uMLC7u7tFERVFMVBSBUERVLq7+/ufXRZcQsX7vu/fe73P55znHGViZ2d3Z37zxIwKgiAIgiAIwm8lApkgCIIgCMJvJgKZIAiCIAjCbyYCmSAIgiAIwm8mApkgCIIgCMJvJgKZIAiCIAjCbyYCmSAIgiAIwm8mApkgCIIgCMJvJgKZIAiCIAjCbyYCmSAIgiAIwm8mApkgCIIgCMJvJgKZIAiCIAjCbyYCmSAIgiAIwm8mApkgCIIgCMJvJgKZIAiCIAjCbyYCmSAIgiAIwm8mApkgCIIgCMJvJgKZIAiCIAjCbyYCmSAIgiAIwm8mApkgCIIgCMJvJgKZIAiCIAjCbyYCmSAIgiAIwm8mApkgCIIgCMJvJgKZIAiCIAjCbyYCmSAIgiAIwm8mApkgCIIgCMJvJgKZIAiCIAjCbyYCmSAIwr9CHIkJ3jg+ecRjAzMs7DwJUkz5a+JISvTB8ekjTAxeYm7r8R+uTxD+3UQgEwQhS+IiAgjy88TTM4vF2w+fwEjiEpNIUqzjz5EohZEoQn288c3svSuKj18AoTHS3H+HHZAUSGzYNVbWLEtltS70W3gNa8WkvyQpiISom6yuW4Eqap3oPfcKlopJgiD8OhHIBEHIkrfHRjCtS0lKlsxiqdmBeuMuYB0cRZRiHX+OQKJC7rCpuTbNMnvvitKg82BWmULA32EHRHsR82EnY6qUoFiL2Uw/845YxaS/JMaH+E+7GV+jFMWbz2DyyTf/2foE4V9OBDJBEH5Cdpp9y5nRzWndqAPNp+3EwMBAXnZMbkibSioUrd6MAfoGXL4h+/sOVgxpQ/3SjSk14y5fQqJJSF7RHySWhDg3PjzazqKWNamqUoGK2hNZK+2Tq/J9c4Yd8/vTv3ZhKrcZxLRTb3nr+ZtTWUI0iWHO2Dx+yAMze+w8wxQTlPni5XCTvX37seKUJZZuP9jmxBj5+t6YSOt7+R5bj8zWJwhCVolAJgjCjyVGQNAV1nTsz+BRW9hp4SX7o1SCeLy2E32qaVK17US2vIGIONkCH3m0dTyjW7SlyWYb/MPlf/zzJElBNfgmW/o0oVb+5jQedgIr6c/R8onheJqf5ujkxpRR06DMkGNcsPQiUj7t7yva+yXPj4yikWouem54zmPnGMUUQRD+10QgEwThxxKkGOFjxLElxzhz2Qpn+R9ltWbvuTi5Je3K1KLRwM3c8pVO6LKcRiAf7h/k5KZV6D/0JTxG+mO0D652rzGR1R4Z3sfAxgMfWecqaT2RgS44PjPA6K6sZukJr9+74SNLNUkJEOmKvdkTjGXLGT/ngb0/0XHyF0mVGB+Dv8NDnj8y4bWdK97JiSgTsuUCcLF8wTN5LZas3JXKM958DSDoV9vbEqQXctjH3M61KVtNhx5rTPCW/Tl5qpRXrbA8M52mOdRRb7CS3Y8c8VdMSiPSD38nc8X2pJSnPH3pjJ80+bu1i6GufH73PN1ysvKQR89s+Sp9bLGytxzjh9dHS57Kp92RihX27kGk1mclJRLlboPl8wdc2r2Uxb3LkV01Jy2n7WXLyeR1PjB5jq20MdHx0vyy9TlaKa3PkvduSuuTifLAyeo5j2Tz3DPmzosveIfHIFuc+DDCvex5Lk27J5XH1p9w9s9s50cRFfKFd7LXl7+WrDzA+MkbvkjXCLKvlSD8SUQgEwTh1yVJZ/uYu2zp2oDa5XvRZYUhbtKfM4QH6WSfEBlIwIcrHJg3kOZFClFAqxDZBh7nho0b4WGu2N7bz/pOxalYWJ1cubXpNf80dz2iiQz4jLfNIZb1bUO9QvnIV7EBxabf5pNfFAmpneSTiAn1xHhBKZqVrUW3Wce446OYpCQpIY64cE98vK+xbWAn2hYuTsmSxSlRvAh5c2gzfLMhxu6xv9a0GhdBwqPFDGhZnjKdpzD/liyOKclCIIuLDCLY+iI3dHuSR6skxUrI+p4Vo0jBxjRsrcdFrxD846V9qJg/WZK0T4MIfLqHDRPbU7BYSp+1YhTKJ72WWhkqt5jLMRdpE2SVk35PMdg0mtYF86Cmokq2nENZdO41H5NXJt83HtdnMaFHDbQ086CeLQfZcmlSqGhxiiv6wtVto8OyJ+AVLi3gb4rh1rG0ka8vu7S+ISw4ZYaDYl3x4b742p5mw+jONCiYj7wFypKjyXZufPQlOC6SCE9zXpxfTI+SRSmpoUqVgXqsN/JM8x4TY8IJD5JC4sONjNMsSo3iJaTtKELhAuWpWH04ayzCcYtIlEdsQfhTiEAmCMKviw6C58vp26QcORuMpM9ei0zDTHxEAG7nRjNg/CoWHjbmifEJ9gzIS4FCE9G/d4HDW9exuu8y9j+34+mmzrTsOZvph55jbfeQG/OqULPPevbfs8To9EIWdCpC/rJD2fo6AJfUtr8oIoMsOTmgOHW1OjJkxQ1eZ7IhUW5W2OzuTv0aJWgwdis7b7/H09OOT3YHmVm6BNUbzGD+aUsyyXLflRgThvuZofSoX4Q6Axax9VW6F85CIHt/fhoLumtRs2l7Jlz05J2TbHTmW+4fmMekOlUoVWUlZ5390mxXUnw0HlenM7ZjJ7qN38wRa9kyHlKx4PzCrnRv1JIGE89hHy9lZllwTYwh2uMZTw9OoEH2nOSutZBNd+2V1pkc8IL8Tbi6bjRd85YhZ/P1nDF7j6NitKi3rz8hMVLgTl3fC54fmURDKbzlrjUf/dt28vXF+DnidKQPbYetZ8PlW1w/vpDJNYuSPfdMDrxxw+PDVc7s1KPtjKNYeDxn15CqNG09klE7TfGUb0uygOf72TmlCRXqNaHNmpc8f/dF2g4TbmyeiE7+Smj2OModR7+0tXKC8A8nApkgCL8sIdwXn3ODaV+7MFX6LWXNoxDFlLSSEmKJdDXHzNKB927BuL69xdkRuSio0Z6O/Weju+80t83scQmLJczFQj6fnWsQoaE+uL814v5LR1wCPXh9biGzGhQkX+FerHnhx+cIxQvEeBJid4gpNQpRssJwJu1+ymfFpFRelpifmEM77TLUGr2ZPUYfcJG3T4ZJYe4pezqUoFYJHUauM8Q2eYksiCE24iMGM+rStHQFWo/bxrnkttxU4Q63uLGiDYXV89Fu5RMefIpIHYUoC1W+jzezoF912vTox/QjZtj5SWuVZ7oYPE0Ps3doVdQ0ajDj+mdsA+WLSWKJj/nM3TkNadmsP0PW3sEutYk2Cr/7+qxauoAWek/k/dVSa5CC3/Du0mya5VBDo8cOjjxzSd2WVF9vcGpRN4oXrkauUdcxdwtJbmLMTMg73l+dS/PsOcjTbSsHn3yVr08WUsM/v8DktSOffKx4cmI+w/MXIXuOqey1ccMtzB2XTw48ffuZ8ERHzo2tRXOdKYw/Zq0YiSslPocrbJ/RhWYdutFt1VVefIlIbvbGHZvLq5heuSjZCs3nsLQ+WW9GQfhTiEAmCMIviw32wHpDPZpXKkPbKbs590kx4YcCcH56EN26auTL2Ziu43ZySgpcPz+pOvJ45yQGFy6OZtGx7HkfgFtKmgj+iL/RXJoUzUeZzqtYe8sxXa2JF59ubUavfUM0CnZg+q0PvA9ULBznJ4WHiyxpUpQKtccydd8LPJKnZEEIUSHGbG9Zihoajek1+zTGSi/sbXWFS2v6069jXfK0nsjh5z6kDmqMCyXWw4SjY5ugXbs53Rcc5X66Fw62vsC5mbVQVc9Dpy3mmH5NecNRxEXbcnpYRRpWq0vXeYcxcldMkvGx4ZX5C46/UP6jxNsMi8OjKZtTHe3F97hrn7FuKeLlbjaNrkve8vVpttsJJ/8fdOj3eY3N8bGUzaFKvYUG3H4fqpigJMIMo21jaatWkuzlV3LmvZe8T5xcfAhJ/ndZ16USXcevR9/YV/pjDElJnzBdMZCulRui3Xc1e22Vgn6EHc+OzKJfhdJka7iD6/be0qcgCH8OEcgEQfhFMUT42nBhsBZ1Cjek75ILGGflFu2x77G6vJg+OTRQLz2YVVff8jErbU7RL7i+YjBtcldBq9EW7nmHEqyYlOBlw5ej/SilpUH9GSc4bZVuhSGPubC0H83yVqNC673c80peNi7YDVeLy1zcPobmWtVoMHEfB15lPY4R60n4p72MLVeEYiratBywjO2pHc8N2DWjK0M71abRkNkMOfIGL1l7n0JCiDO+D+fSpXRBSjadxfzTb6SomlZKIMuRS3pfyx/yMHVHxZIQ8wWTtd3oUVWTGs36MG7bQ+4a2uESHqMY4ZlRlP1NDJa1RDVXScac/oBlJm2zrreXsqRnOUrW7sri51KUTamFzESUgwH3VrQiu2pRRp+wwzxd9zk5tztcXNGb8poVyNH9OA+dA1JHmSaGuxP2fDF9q2ozauUFDFykP8b5k+RznJn1ylOtZH9G6j2Q90uTifJ8i7XhVlaO7E6tMu1ous8SB98fbKAg/AOJQCYIwq9J8ibwy3FmF9GktLoUCDY/yFpTn/8Lnu4fR708hdEYewlDWRtdFiS4XuPQjE5Ur9CMCitf4JkabhIJczbl6Qptimnmobvebe6kazaMe7UTveH10KzSjI5bXvHW0VXeJ8recA+7J9RHs2gFSpSYzqa7thmbOn8gyc8e/wsjqFA8H6oammjKBwmkdK6XldYMWnQKw0yCSqSrJZbr61OucC5qjNvDgRep7ZGpAi3PcXpaNXkga7zqMcaOSuEjKR48rrJ9bDcaaWmSX70oarnHsfnpG96HRhGVoZ0xkS/GO9jUU5Oc6p1Z8eArH2Wd89OIxerQSCY2LkWFBtM5+BUCMrRppkjk6+M9bO2VnxyqHVh2zxmHDOuT5np7gn3TWpCtaDVyzzTmjZsiVCbGEeFmhbleI7SrT2PJaXP5AIOk4C9EXx9B/UqFKNd7MSuv2Mo/K1mxPDCKYR2qoVW8LS267uJxUhKZ1MkJwj+aCGSCIPyayM/4vFxL70J5KdR+JXq3PpKlO40538FArzt5NIuhrfuc558zOYtnwsdoJQt6VaRcw+7MMIwhKCpliGUwrq9PsqW1BgVz1WPsvqe8SNdr3vbYUMY1z0l2VTVyaRaleAnZaL2SFC9Sk8YdprP5lReOrkGERcd9v79UJqLcLHm7pQVlC6tTach6NtySDRJIDg/JxZeA4AiiMxlgEOj4hFsT1SilmY12i85z2T7jWEGXRzvZ2ldVClB5GXDQntceKe9ZISGSsEA/LC/psrJzdrJly0OBosXov/YaBhmaj/14e3U187SLScFtCvveeeKeZnWy1zfn3KQutFWrSfXm+tyVdkZoupf8xh/bm2tZUL8IOXJOYre1G26ZzOv/eDPrhlQhT9k6NNn+kY++ivq7sA843llJzwLqDN3wjIdOsfLvT5irNY9na1C1aHZUc+cnv9a3kFusYEMGrzjFZVtffP3CkEXy726eIPxDiUAmCMKv8bfF/cokqhTITcXR+zn4PGtjEz0fb2f7kCLkLlSSvkc/Y+OV2bjMjN4fH864FmWo1mYqO99LJ+6U9Bduh/WlxfTSykPuqtPYaPge13S1OtYH+jG6cWHK1O7LjKMGXL6Z0qxozJOX9vLRmvG/fGYPx/vddY711qJonir0WXeLey7frU7KIODjY66NzU5JzUoM3mSIsfLwQplgK4y2jaNbsULkytOXDWbufP5OW2Sk/2ecn5/i+t5xtC1RgIqt57H4vF3aJtDgVxhuGU27YuXIpb2J605+aR8CLrvfm991tg5qQY2yHWk87668r9d3A2rwa4y2j6N90TKo1tnAlY8+ZKzjA1/jDawZVBGtyg0ZcS2Cr8FS8At5y6uzKxjXsTVa3fW4Zu5FoOJhACEulhhOVqFy4fK0HrmS9We/NQHL7stm9cmbTG9XJgh/CBHI/ijxJMb78u76bo5s0eXARSPS9u2NwNveBKMDuqxZu44rlr64iXp/4ReFO5lguqoOhfKq0nzRZa7YZuWRQGG8vbCIeU0LoFmsG2vMAr6NlPwuWc1NME/0u9G/enW0u6znWrAUQuQVSmF4mB3jwPiGlMylSY6uuzn72k0xUu+b5EBWmhpt5nP4i7RUmpQRR3x8ADY2nwkMzFptXTJPnJ7tYUGpnBRQ7cysEy+xSb0Nx88lB7IcUiBrwOhdxjxLV6vnbboR/WHVqVisDi0n3MPUKzx5oEK0F16Orzh12gKXkKhvoyTjA4n1vMHqNmWorD2SsbtM0w5OcL7K8fldKFO8FpWn38fCM7mGKVWitFNs9zCvR10K1tKhxeZXignf8fm6fDRmmaLVqTDViFfuoWnXp/D5xkIWdy9DyZo6rHydhNcHU24fms8QnW7UbTWZOXeccFfqW/ctkNVlyPpbPJb1808jAl9fL2xtZXe8E4Q/zx8RyOQddN89UbqayryYWn/gS0pv4D9SNPExdpwZXpnGRVSo3WcOaY+tPry5soo5dVTIkTMXAw/aYfYL/ZgFQXZSdHl5kq1tVcinpkL3tfcwTNdvK3NfeLpnMiMqVKJwrTXc9A3LtFYlLVl6+orB4o50L1Obht22YBCZRFRSJD4OphhJJ/c5PatSOncBcow4w803rgS7vcfB8iXmrpHExCfhfHEy09pXpmKzCawxjyRYfmMuSVwwQa4WmNw4gt7qy7yw85DeWRZFvsP6ygI6ZFNHQ1Na76230rvLujDX1zxbW4vahWvQbelFrio6dCUlJsjvl3ZpZSd6tyxLgwHzOGwjRdKU2jG/11hcXEm9+jM4bemNV8rf44KIdb3Nqq5lqddpBrNPWqetIXt7gl1TWpC/YkPa73PmS0AIfo4WfHxvh7OsqkwWyKy2Mq1bLTSqdqfpahNpXySRGOXBJ8vnmJvb4qRcpWZ7in3TW5KvvDZt9shGYwbj98mSj3Zp57M9PYEZrQpQpHwXJh++zbn1I+nSqgmlWoxn4h5T+eha5abuSE9bLFeXo0aJqnRdfJbrzkrVYVGeOFsbcv7oKbbvNkaW1bLUTC4I/yB/RCALeHWMIxOroKKi8sMiCyhrjXzxDowkNjHpD+yDIAKZ8L8UR3SoFabH5tJP+j2pS6X9orNcehOmeIblDySYc3PlUDpq1aXYgJNYSL/Bn1cqyZo03Xio25PelcpRtcVUtr52w8nNmivLF3Jgrz47Tm9gYiFN8g7Yw/GHJjw6sohV4/sz6pwr/hGymp/T7JzWlnLardN06vf8+IgHh2bTs2xlenfbxuGXX7LWqT8ukvC3V7i4pCMqOdRRqTOX9VctcQ6NznpAiPYk0m4vUxpWolynucw4YirfJncXZyx2DaB3qyrU6j+f5XfTjghI/PyAxxu6o5pLA50tpty1UfRXs3/B5xPDqFuzCk2mH+LYq3RR1/4s+2e0oXDZurTQf8mr9xbcXDuc9StXsd9ami5rsvS4ztZhLahRuim1Rx6SjgvufH1zkvUjuzBqzAr2yR7SmcLhPIdnt0OrdC2arnvJSztzbumPQn/5MvZaKuaR2J+bzOy2OVKPv+qaRSjcdBrTDrxIcxPYVGHuJD1eROf6pak+YGmaTv2eb8+weXwPetXrzcR5N3ksHbxF5b7wp/kjAtnne/ro91JNE74yKzlz50WzUhtqjDyLmXSF/ucNmhaBTPhfesv9TZMYWCgfuaXfUzapqOfXov2MXZz+2TBLnzscn92dWmWbUnL+QzxCYrJ4QZRA9PNtrBhUXwoi0u+3aDlKlOjMzH0PMHUP5avVTU4MzIaWVgEKFWnOgLn7ufIhgMDIBKRrLilAReBmsp8dY2ul6dRfsv4Aei2+hLWXF95+0rEgNiFrnfrtzrB3Vke08kthTCUbKjnzoak1gInrDXmjmOWnpACUFBtEkPUeFvRoRvWCRZO3qURpihVozEj9K9z+FExIVNo+dqFvLnNpbj1UsmUjd4GiFE55ZFLxopQurkW/bc8w+pByE1UlcZbc3zaerjlySvugOMWKt2fSlus8+BJCeEolVEIk5geHM66pGjk1ClK0VBlKdFjJ5qsv+RSoNJ9MnDXGuybRXb6+YhSV1jdh4xWMPqedz/v+OnT7l5Mfe7Nlz0GP9Q+4+SaI4PCYTJ/qIA+G0cG8PTyM/q0rpunUX7LdUtZdeMZHb38Cg6NEp37hj/RHBDKnu2tZ210KXWp5UGk7l+W7zmRorjy9aiB9GxWRDqB5yVm6Gc3X3+eZvL7+TyICmfC/FIz3B2tepfttPXvzia8/u0NnysPFHz3noYM/MfHpQsOPhHzF0eaZ4vWSHwb+1iVQ/jDwmFAfPGwMuG8om5b5w8Uz7dJg/IqX9t7yWrpfOrGHuOD0Jv0DvV9h5eCVem+0LEmS3n+UGx/NTZMfwC0vsgd1m2Lz2Y+Ue9cqiwvxxMMu864Zd+4aYuUSTlCmnf9D8f9ii4V8XsVrOPtmuK1FqNsbbF8o1nlHms/UASfpwjXjpoQS8DXt+qydfDN0uI/xd8LR6pF8fXek9VlKATClA/+PRHm8wfrFg+TtSClP7fnkndm2CMKf488KZLkLoDL8NDfeZBz1Fe74iIt6/WhfRZpPVQpudVay97FThhsy/rOJQCYIgiAI/0T/mkAm42W6l70jiqKikksqvVlw3hy7P+qSSwQyQRAEQfgn+lcFMpwNub+uhxTGckqlKRMPPOXV91otE+NJjArGx9vrW8dSWfGS/h8USVRcpr0gMpA9bDc00CftOuTFD7+AiO/0hUgiITKIQD/vTJbzxDd9n45UIpAJgiAIwj/RvzSQyTrkDmLpFSs+fK8rS+Angg3m0Fy76reOpbJSvgolp13E4F2Gm+Rkys9kK/pjm6Rdh7x0pOPAQ5hK86QfbZYYF43HlWmM7VY7k+VK0mG8Lgdko6MyEIFMEARBEP6J/p1Nljk0UKm8gG0PPsjvhZPWZ6yu7GRN13Z0a1gGjVzpRm9ml/5fsh4NW49lxT5DzDLthBZGbORLzk4ezeimVahcPHfadchLfvKX6Ui7lYZY+YUrbmjpg6fddXb17EFv7VIU01TLZDkV8hevSA2dGQw8ZINbml68IpAJgiAIwj/RvyaQeVle4viijnSoooKqNF+daTe5bReY4c7eXla72TpamypSYFLL3Yw+cxaxQFcXXXlZwcplk+lTU5PiGlpUaDeTBZcd5I8h+daAGU2o1yue7BtMt6L5KKgizdegB4Pnp6wjucwZ1YVOjSuTp9MBHroFK+7E7cD7OyvoJYW+vLV0GDBpQZplZGVq77o0KC2917ylUW+xiotv/JRGN/3vApnP21vcPZ52W35a9PTQvWyOtau4Y5AgCIIg/MifFch+cNuLndOaS2FMnbyFKtJQZwQbjX2T71KdQnYPnCg3DPW606tKfjSLtKH98GPc9w5Veu5bDHFRn3i0qTd9qxWigFY9tEds55EfRKYkssgvfDDaxJiq2civpkmZ2gOYtv0uL9PVpPm8ucG1bZMZuNKQNyk1ZBHOfH56mAV9+zNg8wOeOGa8l/mXh6vRHVCOYioa5FSvz+ybn7FN3cD/XSCzPT2e6S2/1dJlqWTLhkr/vRx5Lh51IgiCIAg/8mcFssxCQWpRJZdGdZr01uWsa/pn2kkSokhwPMiizrUplbMRDXvt55n054y3zZEtaMbJke1oqaZCiXodmGGUgG9Ectf8ROd7GK/vJn89lZzNmbDPhPQ3zv7P2PF491T6qarI79g9+LADr1JD1f8ukDlcmc+yPhn7s/2wlCpFyYknOGee6X25MwgODs50EIMooogiiij/nOLj40Ni4i/ca1CQ+xcFskq0GbON0+9D5LVZ6Uc2JkaH4n5+BDoNipKjVn86bnrx3RGQEIv53gGMqq9CgWot6HHEFY+Q5AenhNhc4PzsuorHqixkq5ED3v/V76UnVheWMr36/28gi4sMJsQ/8x/fD0tgBJGxWRuRumTJksyDnSiiiCKKKP+Y0rhxY/z90z01X/ipP7zJ8irXLm5jdnMtyufLS5Hy7ek79xhGPkpNjApxkYGY62vTtqoaKvlLUqRWa3R0dL5b2tQtSZkCKmhWbU6Xg59xD04OZG4mu9k5KB8qOTVQ6bWXU2Zuf+Hu0rIE956H25azMMNrd6FNgypUyPf/G8j+P0yZMiWTIC2KKKKIIso/qZQtWxZf36zdiUD45g/v1B9HfLQ7NpdXMa9LJapo5qdw1V4M2mfFx5CYNEEpLiKAJ8sr0KpCxi/Xj0r6QPb5/mY29lIhe668lJh+jdu2v/YsgKiArzjcWs32jePoXbcCpTN5zZTypwUy2eNVMh0cIIoooogiyj+m7Nixg4iIP+9p0f9rf3ggk5HVNvlisWc0I+tL0/NUQLPZGq45BRGg1I9MOZCpFy5HxUad09VMZV6GTFnE6nveBEQkryw1kKnnpdK8Oxja/8IT7iL9+Gp6guWdVNDKo0Ke0rWo3aJrutf8PTVkaZ5zl9Uiex6e5Rdcs/IAO0EQBEH4F/sXBDIF2+NsHt8cdZX8aBToyDzDrziEfOvcFRcRyPNVVWldKTslOkxmxtW/NjIwJZCpqGmgMvAw58095ZEwKxI+GWK8rjMqKtnk76XB1AMcyzAiwAebS8uZXfv/N5CJUZaCIAiC8L/z7wlkca+5rTecDirZyZa7MPkmXuD2Oz/FRNkgyxCcj/ama92C5GgwlO57rBRTfk1qIFPNjUoTXfY8/kRWB1nK+58NUJeWlUrvrRw0cSYiNn2cE4FMEARBEP40/55ARgifDbej370MKtk1UCk1my337EnJI0mx4UQZzqZvM2l6/jpoD9ySaef/n4m0vcr1RY2lMJITFc1uzDr+gjeZ3Rc1whlni9usWGmIveI+ZF8ebmOzjvQ+ckphbvCR79wu4gOmB2YyRDRZCoIgCMIf418UyCDK2ZA767pTSkUVVZU69Nt4m/tuirCQEANfr7N+RGOqauZHq0xb+i0+i5FzGIGZDJP0trrCraMHOXT6FZ+l/yd36ZcEWWJxZhraKtnJpVKCet3msfaSBQ7yW/EreFvz8vJapgwbRNUaa7n52R9Z13/vF0c4OKqoFBjVUKk1Cb2rb3BJedBlYjy4mHD91HJGd2tINZU/r1O/IAiCIPxb/asCGeHveXN1KToFVMiTXYVSffXZYuQixRiZ5M7/lmfnMrNNGYqq5CFXnoYM0T/F4SsZa392zWhFp6r10O64A2NpyW8PCPfG3eYs29pUp3JuNdRUSlK3y1SWnlRafvdMxnephmqeipRqspNbXwPlzZoRzkYYbuhGGSnM5VSpSPvxa9h4TrHMrRsYbBxKuzolUpsERSATBEEQhD/DvyuQEYy71Tl2ttdAK3c2VCqNYNLep3xN0yxph8meuYwsVpjC+XKSI7s0X0qfKKWiqqGJZuGOdB58GFNpqTSNctGeRNrtY0rjGlTOp456jvTL5yJ33oJUaNGTIadc8FTcVBbc+Gy2j3klilA2dw5ySqHx2zLSduTMi6ZWUYpoaZJfXQQyQRAEQfhT/MsCWSIJ3m9wPTaQckXyoqJah26zjnPXWzFZLo6Y8GA8LAwxWqpNqYK5lELRt1Jp8Do23HLELyAi4x39kxJIjAslyMeSC/M60UX2MPA0y7eix9QjGPr4yW+XkZCYsnQC8bH++H02Zk+fstTVUlpG1oxZexarz5ty75IuK7uIQCYIgiAIf4o/IpBF+Dji+NoAA0MjDKzd8Q6VRaTviA0j1vMNjx7cw8DAhFe2Lngnt1mmkRAVTIjTc4zvG35ralQqJm++8DWzzmWpZCErCt+P5lg8Sr+8GZb2XkoPLVcmhbmEUNytH2N6X2mZO3cxMP2As284wf6fcbYw4M5dQ6xcwglK3f4EkhLDcLd5zLMHBjyxsEv7AHUpOoZ6feT9U9myd6Vlw5SWFYT/lOz38JY7evOZozOPhesNsZX+ktq/8pfJlnyH4bqFzNOZy/w1d6X//SfrE4R/A0fMTm5ipc4kRk86w4uIGJS7MAt/X39EIBP+HCEuVry+lPndn79bDtznqYMPf9Z9oWWxw1k6sB7gYGbvWVE2btvNLfsY/P8OA1kTpU8g+CrrOtWjWsnudFpwmy/Sn39xoPI3idKbCr6GfrcGVC/RlfZzb8rXp3Q/538wWZ9VF6yvnuRoJp9rxrKHU7deY5/ZiO2/KNrbHjujA99eY891br76yn/xJf5DIQR7vsZA2rYtKdu47xyHTNPdRifGl4CPDziup8cGxXxbj13hpm1IJs8i/hcIM+b0rJ40LqhN2T7HsQ6LUerjLPydiUAm/I2E8fHeNla3zUO2ck1p3bGb/OkEnVrWo3YJFbKpqFC8VmtadJI9saATndrUply2bKg1W8HGOw78wjMR/gFkVZdvuLl8JAMrlaCwigbqearQqHsPusqf2CAFlOa1qVehBF2WXeLsS/fUB9z/NglSgPK5z/Fl0xk7bwfrDZ0VE5TFEBftjv0DI16/c8FN8cixTCXGSOt7wKmVMxk3dxtr7jopJvwJZDHVjgdbpjOmZlmKqaiTQ7UC9Tp0obPSkzl69ehOi6p50cpTl86TD3P3157E9kMRX17wfP84ejYuS77cOcnTeBozj1vJR3z/Pfjh63iNfR1qUTt/bnLJum6UbkvdGbfxkZJW6jcn0gXPV4dY3LE2VaX5CpaoSrOxa9n51J/U3iD/aLLvih/OL59iYeHAR98ftADJBLzEcN8KpkxayqQzdkTE/OVLIuH/mQhkwt9H4jueHpxD/2J10Vr0mDfusor2RFxND7GzdzayZVNj4F5LnrnLZvbA7c0BZubOTqlBezNeNf8pEsy4PL8f7bJXplRNPW6FRituNBzI52eH2NAuL5o5s1F94jFOvpZOQPJpf19JCZ74fTrE9GIF6Dr1OCct/vkxOjEmjNDAwMz7k/5MojX3N4+je47SaBRbzGknP5R7wCbEhPNiTW3a1+7OwCU3sVH8/b8lKdCJ8OtjqFWuINVGbWOX6d8njsnF+ILTLsbXrkhZNTXU1KpTUXsl16OTCFLe0dHSfB+2M6ZORVoP28A2KYz9KWTdUOIjbrCudS169VvOmgffbmgu/FlEIBP+PgKeYLBhNQO0V3A2SAoeCbIjbgD2dzaypIEG2VQaMeeqPW/l5/BgfD/eYHuTHHRddoWrb/6s+rFU3jc5MKkjNQppU67fMSwjYhXND4nEB9rieGMObfPmolD9uay49AbptPS3FulixuutbSmbX41uq25z2/HvHiF/zvfRJtaNGUvnIUcyjrj+Gb8HXFrej3rFapNvogF23uFpmngT46L4fHwAvWatY/Z5W3kvvf+mGC87Pu5sRdUSeWg99wTn7f5mn0dqIOtK76aNaNq0EkWqtGD24yRcldtWw6QLsoczaFmtLf1XXeGexz//e5UiLsQDl1N9aVutEC3GbeaQ9Z/z3oS0RCAT/j5i/fH+5Iil6Sd5sEhukvjE69OLGFm0GNlqLOKYhTve8gnxxIR78+XFHcw/euEVKv0x0hVP8xMs7dObQbImnyUH2f3YRZpXtoAXz/bOZsloWVPQSMZM3cFNtwSCZGe4EFssr29jtrRMb9ly629yxzZjtHF7dpjjy0cyevI2brjGZXrD4GRuOD45wSZpXX1l65OXCcxbdwET6eJWnjOz6t0BVg1pRMkaneiyxRrfmIRvJ+xwR9wfraRLPnU0a01n8Tkr0gwYThHkjLfxFoYN6pfaFKajM5FRE8/wPCz6+32GPF5hdnaV0jIpZThDx+hz2TFc2h5pvrAPvL+3h3nStOT3q8uOS5YoN1gGvj7OodXj6N6hhXTSLICaanaKVG1Ckw7J6xw4bAzbnwbxNUSaOewj9vf3ytfXT76+1Ww7b06aBssgc4yP6jFNNr3vYHTmXeDee+/kzssxPtKuuYz+6KEMl6aPXX0wk2fCyoQSHfqM0xNGMlH+OrIylIFDdaVgEoRXFge8eBuuYJFOb7Q77Ux3T8IscDjL3ultKVahMQ232PI1MG1zVFJiAlHu1pjbO/PeM1zx13TCPYl5tZvp44Yq3oNUBk+i3+LrnN45inl629l52y7TpsjQr68xnlOICoUrM2DdTe5nNura15YPN9d/W3dqGUTvPks5aeGFa6Zv2gW7e4fQT7PMJBZvvY5pViuwwt3BdAEdqo1m0uhhjBnTGLUS1Sm32pJ3Ht96jcYFOON1rDva7Wey8PQrnL7bqieLy9ZcXTiVGanbNFAqizj6woUvP+qIGi99u7zusn/uREanLjtMKns4cmQly3YeRP+SDbL6qww/8eggsDnE0hmjFctJZcBIdBZc5/D2SSxft551F6zTLBv6/g63dkyhZ7dOdKhTmAJ5VClYtib12qS8tg5rr9liI6tSjfIg4O1ZVg3qzxD5tAUs3XwfO2lS6vEi1J53BruYK03vI5tn+gFOmDglDy5LkD5AHyOOLJ7KWGna8Nm6bDT2JTLDo/uSB+4Y6M5ljvx1ZKW/VOaz3/gj0iFB+A+IQCb8vfk+4sb6wdQoUB6VPmd45BQgbxZKL+LLS57dOsTaTdvZoqfHpI6lqNpwAAPXXMXO24o7a7ZxaNNaVk7sQqdmzajQYQ23feKxfXKUC0c3s37DFtbOn8bIxjko0nAU80+Y4Zzm4ByK5dHJTGlYhlLVZ3HcMxbfTHqXh76/y82TS5k9czxD+81l9SpZJ+OJDJRes0X9YYw//hbPuIQsd0wPfbiSGT0qU6pZP6bdCiYyTulQn4VA5u/wkEf7J7NoRGuaj1zBjMWy7VnMrDHD6VGjI8OXXeG+s1+GEb9hH4y4u3Mhs8eNoc9cXVasli23gPG9GlG/THmKNF7CRecw/GRZN8oNl5en2DGuHdVzq6FeZAhTdplIUfqbsI8PMTy7gqlD21Inhzo5S7ZHZ/wclig6Yetv3sb1d2H4yPZ5lDtur06za3x7aubJRe7Cg5iw/RGO0iRZjVGw9QVOH9+Hvr4uS8br0L5MHtTLzGbLfQfcAx359Pwyeluk78Ha+YxqVZGG7YYydLelvI9hyuklyuMNNne2sHLFSAb3ncm82cul7ZjN9JHdaaRRk87LDLj/KTBLtV3/SSALf76NtcNqU7hGK4ZdCsIzNPn0Ge1li+Or25y2CCIk+vt9gII/m/Hi+CxWjWhCu5ELmThftj+XMn/2RLq16Ey3Rnkp2WUqU45ZZhK8g/F8e4ktDXJQvMJA5p8ywz7dxkc4m/JEChsLx46gu3SSXrRStv6FTB3clmZliqJaYSr7XrnjrvyjTEqUrnFucvnoQmZOm8DwgfPkn7Gu7nj6NWhMq2bjmX7WTt70/rPfQXygM4Fn+6LdYw3rDu7g0LpB1NAohkrDJZyQXtdH8boRXu8xX1mOmsM2ss3IMdMBPjG+H/n4cBdr1oyUtmkaM6culbZpHvOn9aVR9rJ0nHGai1K6yXRZv098ur+DtXN70KPfZMbLl9Vl2fz50jGjMb06VqS8zjSG7HqJ7JpCOZCFub/F6txC1o1uQpeRsxkt/Z50dZexZNFUerTqTrfGBSnfcTgDd7xIs2ykqzlmN/RZOEuHpmqq5CvahFYDJrNQ8ZuRlfPPvySPpI8NINTpHmcX9aVNaS3yFehAp+mXpOiUHMjCHO5hcPEA6zaul45zY+hVKRdaZYYzfb8pTlKgD3p7g0279rBp3RKm9GxIi8ataLbsiRRQ41L76sUGfuWryT426Y9l9NBpTJu0RNqGhSyeO4DmGhXpMPEQx195IDLZXycCmfC3Fmt9jH1TmsivisuuskhzVaws5N0Nrh7ZxvSdT3FPTOLpxqb0rNmIZj0XsP3GAVb3XsSFF668fXWTa0e3MXXHE9wSErG9toh1azey4bIFTm7mXBuVi2qF6zBg9VUeKHfmibbltt5AepavSfnmW3gYHSc/eKaKjwY/O26v70evTk1oPGIV20z8iJVXh73l9vJB9ChZj/I9DmIWGZuFg5YsOkRid3gkwxqXplqHCWy0kDZD6dwc7/+WD5cn00gKLRW6rGOLoeJqVyHay45720cwtkNJ6vacxLZn4bjJz8phUuA5xe7uxcmr1pEFl615l3oilm1vJPbHxjG6fVsaD9rKbV/phCRPMoF8uLaaZUO7UGrURexCor+F4whnfJ/p0VMrLwUbzmHxeZuMt3WJsMHs9Cza5SpArvbbOfPa9fvD8SO+EGC2lt5FNSnUYCbzz1jLT+Cy/lo+D9ezctNJDj14ivnTvSytkpv8qiNYdfMNn7wssbxznOFrDLEPceDq0s70aNeT9ivvy/tmyXdfiAs2l/VYMEibEk36MO20Lbby6jA3Pj6WvkNaGuSvuoBtDz7gJZv/J/5aIJPt5yg+np/DjPZlKVO/N8tfSF8heQIM5euDvRydP5S+p7xxD8k8tsgChunRWUxqX4BiDbqja+CCo7wiMAJvu5vs76RCybwqaPVdi969TPpYxjnh+Ggdg7PnRLPlGnYZf0r7nZa279PlRczr2ZIq7ZdzWdoZyWMwgvnyaA9bhrcgZ8/jPPoc+O09x0WS5G3DtVXd6NK+GS3H67Pnmaw6TPZ+rbk8uyedSjWm1qDj8rDw48AbT7iHFS8W16DJgjOcNHvN26trmV2pKCoq1Rl9zIJX3rJ9E0/w15dcHZ6dBrMOceJlJp9amAcf7u1Cb3gdCtfuwJh9z3gprw7zxuv9CRYVVaVEpcksPW8pfQvSCfPE0WgPa0fUoUid9oze85QXiqq0SO8PPF9SgjqlVNFoP4tpFz7K/54iNuALFhdXMauDBkVrt2XRBVveyo8r0oWF2zNO9dGgakEVCnaZx5xrsjHE6UR/xsNsA/3yqlOo8SJWX7P9/uClOGnK5wPMalaFUlUH0nvDY3lLg+ynG2B2mH2796N3Vrqwcb/Llqb5qZSnJ+M2GmET5IyP6SGGr7zJ409vuX9wCuPaNKTsmEvYh8Yk/8YjvPlicoTNo2tTWrsdw7fdl+aVHcX8CXK7wuoqBShfcTQzDj7nq2x+4S8RgUz4W/M23sjaQRUpUrkp0+4n4vLTMfmyA38MT3Qb061qRUqW7sewGdKJUgpfP140BH+n2+xsoUb5vNoM0rvOI+VmFZcLbB3VikolmlFz/GU+xMQrHrklk0BiyBdCLo6nebWyVB+yls1PlDreJnzh0bZRDKlSjbL19TEMj/n+QTWVrGnAkctTm9K2eEUa6qzhrLv0V0UVT3x0KF4Wl7k8T5s8BQszZKcpT12Tp8lqKGRNEI5npjC6bXmqdhyNvpl0ClA6r4c7mfBkVQM0NVTpusYIQyfFiuXXw1+4Oa81nZt2pMOCy9hLR+R42W6V+Xyfexe30XjDSwKkq+dU4U74PFlJ50J5Kdp7A5vuZzwsJ368yY0VHcmWtzAaE69x7/0POpCHO+P/bDXdi+anmM561t/L5GSV6Mg7g9UMU8lNnpxj0Lv9VqmZVHqzCR+5vqAjPfuOpO8h2ek/+bsRYbqDlYMaSSfmLgw664pvuGLHJPrx5cVhVtVXp2CBsejdVF7f9/21QCbb35+5s7wHvcoUp0KDcWw088TBxRNPz4dc0h3F2HqdmPeY5FrDDGJxubGcBd0qULBKG3qfdMFVacRqkNMzbo5XoWzBnHRafI6raXOCXGKAFdZnp9Iwhxr5hx7h7GtPxRQZ2b5y5eH6/gxo2pS6Y6UAJX0PolK+Bx5mWN/Qp+zSJ9h7pVxeJEgXCQ74nx5OnXIlqDVmO3teKH3GCc7cWzuAvpXqUK31NkykP/34/ljB+Hy4xrZGKvTWu83t92GEOdzl7vIm5MqmQs0ZF7liE0hiUhA+9tfY1ESdfutuY2Cffq2xRFkcY+/EVuQrU58uh534kDJSUVrW98N1tjRTpazWYGYdeoZD8hQF2bInODC5NXmlC6pOBx15n1ItJ4kJ+MzHnS1oUKEgVQevYdNj5Q73sXjc38L6/lXIU6I2nfY7YKd008sIKcw9mJWH2iVUaTllFydlN+1LJ8njNQ6HBshHwaoP2MGuxyk/8kzESkeVT3uZ2rQyZVtOZtTJ94oJSpJkg2qOMLNgXkqq9mbClgfye/ulSvjK0z0TGdO1DTXXmBEo/43HEf3mAmdmtye3VhXa77TByl0RpZNCCXF/yO5O0vsv2odxmx/I7z0o/DUikAl/aw4XZzC3XXFK1RzJVvvETG/im5bs5PCE7W1rUFe1Io1663JMOpBHJiWlNldlTjoR2h1mdr6cFM/Wk8m7HvFGeQGL7czrW5di9XUYfNSJiPhE+SkrmTc+H0+xonRBSqt2Y8zme7xOrlJK5nVbOqB3pFblNlRbZExAdPxPtkVGdlp/zK7O9dDO0ZjWgw/xOEEeM+RkT6fQH6CFVonSZB96lNtvvEkd3S7r6+J6muVd6lKxUjc6L7oqD1XKfdeUA1mdGWc4Z5VSN6IUyMqpU7n9SNa9lE4eKef6+GiiI8Pwka6cE6V9msr3Le7nR1OqQG46rrzFzQ9K6U8h2Pw4RybXQqNQSQYcd8HG+wd7wc8W70tjKaeVhw7LrnHdIeP6CLbg1fGpNM+Zj1x1V7LvqZN0va4guwWH21lW99Cm95ClbHmZfGKBl5wZ3YnWGg3Q7ryJW5EJ0ndDvoS0Phssz8ymTd58qPc7xEULz582qcn8tUAmW/MLjo9oR4scOVBVy4Nm0ZIUL1GSkiWLoJW/ArWaTGG3I5n0VZR90K85N7Er7dRrU6PlWq5HxBOqdI+H5ECWXQpktRi22RAT5dpehVDbG1xf1JBcqrmovdCAu++V68dk61IEsspqlJDC4aIn0lc5NXvFEBslXRSExBCX+sXywM3qIAtloVy1F1P3m2Cd+juQ5vG8yrYRLalWvQsNVz2R17x82+LMuONqvZ/paipM2Pmapy5JUn56w/srs2icLRsFGi9l4x17/MMccDLUp5N6Y5ZeeodNYPrvlSU3Fg6kW56alKuzjIshsQSkbHP4B+kaYyVdNHKg2X0jOx9+ln9LvrHi1pIhdM9TgzI1l3A+MBp/pR9SuJc9j+bkl0JVDXSWnueOu/JrW3Nr6WC6a1ShVLVFnPWLTLPst0BWjd5LzmOonIcVIhyNMVndmDy5ctB53UMMPylVkacXKYVBkwV00S5Fo9H67DVP+07kIpzwNNWjp6YGBatOYf4pc+lTU+J9h2Mzu9GzdT/m3Y8hPFa2vW8wWjeWvvmqUKTcXI65BuOd8j4iv+DzQp/+xfOi1XEVerc+ptt/wq8QgUz4G3PAYPkAemlVooL2Mi6HJRH0g3O4XJx0Re51nDn1K1C25iAGbTbGQzo3fzsMfkfUJ74+1qOXmioFmi5nw22HNDVqAQbzmNCxHBXaDmfRgyhilY+LAW/5dGUa1dRzUrXXNvYbu8iXjQ/zlZ+s541qTe0yLWneYw17PwYQm5WbI0VL7+ONPkMalSe/SkEKlaxLCx0deik60rZv1Jyu3YYw9/RDbth44h/+7aydEO6D97XR9KhdlLJtpzHz3Ed505DyqyoHsprTTnLGMqWBUTZXJN7PtqHbT5uqhUpQufkguvfcwSVr1+824QV/eIDh3PIUylOdUftMeZ5JAHC5vwH9PsXJX7QtSx758+kHVZYhH425P78cRfJVZ8QuE0wzG63g+pAHm/pRQF2T7H0OccnSk5S9kBgdgu+tKQxq3JPB807xUFYlGR8BttuZ3L46lbSHMmaXlTzAyT7KIMszHF/Zn9baDSmhOYD5RvY4SGEjlaxJ2vYMG5dMVHRk/lY6N61CpWLS+ypci2bS/3soTx8ibcOqe7wNiEjbPCcFGux2M7NLLQqWb0Gd8du+PZVj90wmdOlDrdZruSmlu7B03/kkaVk/k6VMbluRSrX7MWSLubwz+LevZEqTZQ5K5u3BrKPPeZMh1Enx6flh9g8tJoXBeow/+w6rDJ0io/CzOMru8S2omrcg5Zv2ppMUWo4ZO0hRLRN+Frw5NYFy2XNQtf9+Tj53l9eAxQW74XFrIdOHNKdGyVa0GbSZI04ZGrQz8rPE9uwCyql0Zf3DT3yU7cB4HzysTrCkRjaKF+0gBbUHWHx4je2x2WhlH8YWU2ec0yeCD0dYObARFat1pdeal3hJvz/ZLLJ+blc3DKNDE21p2a5MOfMS84A0nxJ8PMrqIU3QKtWS6lNu4h6XoBQ4oghyecJJnVxULtiJsVuMsFD6ygRIQWVx75pUrNKRrque4yodNJSX/dZk2ZExGw0xV1o2ha/NVc6NL4KGWjXGHzPnlXIFXDoxgV+lQ0ZtGpevSte5x7guG8+Unt9bvl6cQIV86qi2X8Xa2x++dTuQBJqsZXEfHdrprOOKdAiS5+lPp9k8rhUVK7Si/VJTPknHGtnXSdbP1HDHKLq0qEfRnJ0Zd+ARpr6ZdykRskYEMuHvK/QeB8d2QLtwM2qNOIejdCD9aQVZpBQZzFehU6skZbotYsGNz4oJPxbj/pLXBwZQKWcO6s6+zKXU22jIT9eYbdKhd60y1Okxl9220klG6SQZ5mCI0ZJa5FbLTtGmAxg4JbnD76oli5jdszX9Js5hytJjnLxmjewY+bNMKZMQ6knQpWG0qlGYAtVa0XbEwtSOvMllJ4fOmSDbTKWLbrnoQBes1tWicQV1ag3RlU5SGXusBb435OaMEuTPnZ0Wi69xzS7diSjsA5Z3pQAyZQh966iRS7Ue7UdOZ9PNZ1hmSGVRuL0+y+6uucmfqzsLL1lhm251shpIs4MzGV+pBAVLjGffh2A8vnspHYW7xQX2dlOXwlY35p234F2G9Ulz2Uqf04Lm5MhTiJKz73DfQXGSjwsm8usDDg7Xps/g9ay/8lF6dWk/RQXx9WhPutQpQP6KjWg+dEHq/lwwvB8jh49g6ILtbNh0X9qvkWn7+cVJJ5onyxnYqsK358tmpWhWJV+XAzxyD06zvqS4SIJvTaNfs1IUazeBKVeV6imcL3NkxQq6DTqGpfTf9N/5xNhI3u1qR9faGlTqNo3lD9Il2+ivfDLewriyahSst4Sthg7ywJZWBB/ubmR1m0LkzDWIjS9c+ZxJIJD1DbR/cpkdsycwrEE2CuSuSbM+49G7YMjzdJ2tQt5e5drc6qhky07xFsMYNj35d7By4Txmd29J3ykLmL70JOfu2Gbsp5WJmE/3Mdk0DJUyKzj1zkNxS5cYQl1fcG9yWaoUKUOrmYfYe+Est5e1J3vZ5Zx4457h1i9uF8YwvGUR8pSuSb3+81it+MwXjZFC+eBB9J+3Wfr/bczcgtL1oZMOQfeXMqlzRfLU7UmnPbbExiv/ev3xdbyIbmU1ypQZzfzjZigfbeyP9Gdok3yUaTmYmQYhxKS2+0tiPfG2OsyMquoUrzuTZeesMxkh7c8Hox0sq5WXXDl6sdzADvvvHgBjCfOy4tLw/NQs0pKhejd4lvFnT6zLC2y2d6ZAnpwUGb6X/aaKajnZTZiDLLmysCejBsxk+m4reV8w2dHP4/oMpnYqTu5ilajRZy5LV6yW77/FE0YybmBf+szVZ8XqG5g6+SrukSj8VSKQCX9fjsfRG9iYcnV60W6HdZbuuh0f+Bnfkz1oUFGTmiM2suPZz3tryQS+k66WZ1ZBTbUQw/ab8zyl+SAxUjopmXFgSB0altKm9Zi9GEpnN+UQ5PHqDIcG5ZBCS0EqNGhLm24pNSQDpLKAw6Zf+PxLF47xRPk78GpVQxqULUbdERvY+TJr70Mm0s8Z47kF0C6Vj9ZTdnM6bacY6agciNOj/dLJWBWNnGUYue8ZzzK9XwbyUXyPtnRnSLe6VNTKS52eC1h9x0feLJe6C2JdeW+4kWlV86GhNY3NDzOpQYl+ztWlA2mtUY0iTbbxwC88w8kvVawbDkabpZNVHvIWnsKGe+/lQTa9wNcnODq5Bmr5i9BuizVmLtLZSgpjQZ+MMdo5jha1WzF//xNSukbFhvnxZGlZWlTIS5HydWncWakmS2cqq/YbkendMWT+qzVkCSTE+GK9pQudapSk0ZBVHFDuyBNshemd2+w59FJ+kk5fb5UQE4Hp6uq0r5yHhoOWs1/WPU5JpIsxxtt6UV1VnSJjT3PZKn1EkcR84sXxuYyoVBK1Eks45eCd5oa06cma18z39GJCv/pUL1GA6q1HMeuyp3xEYkpEcX26n519VcmWrRCVmnSgXervYBC9ey/hhLnnd26PkZlYvF+f4fistqh0PYrRJ//UpuC4EBe8DMbSrUphKneZxZDpc1g3tDyFBp7ngVNAhiZjM/kAHw0KlqxB/U7fapllt35ZsOkKsu6e3zu0fDo5jKHNClKqxWCmXPdTap6VPpcwBz5IgU0nV04KNl3C2pvv0/QNfbWllXQxo0HNLlPZLCVr5TtIRHu+wvLoIOqrq1JsyB72P5Xf7Tqt2He8PDGHXqpaqJZZyuFXX7//GcX7E/DhMqsb5aFssUFM3W2Sri9csvBPjzFR1Iw3mn+Zy2+l1BYfQbSPDa9PzqJ/246MW34KI6VkabGrC0PqSxdbRaqg3VH6fvdK2X/jmbXqNMbS10u5h4bw14lAJvxtxZqsYmKXKpRoMZSJ0sH/54EsSbp6tuLORFWqFlGn3ZwjXMjsqJRBIm4vjrO3bx5y5mjN/Gt2vJNXOiSRGOVNhJU+Y+uVpWjlgfTXN5b3uVDelORAllsKZM2YffEtNhmyUwIxMbFS+W6VUDqhBLkasbtNHirlq0e/5ZekAKOYlAXfAlktdBadxTBd2Ir3fITR1iHUy65JvoKz2GbimBx4khJIjI0gUAoP0XHSv+VzS+R90k6xtGMtylfoSeeld1KvnuX8X/B031jqqRdEo/VWTpm7Zhy04H6DQ9M7U7lcM0otNsE7NLPqGAX/lzw7MB7tXJpotNrEcbOvmQ6C8Dc7yqEJ1chdsDRjznvw1iOCsK8PMdgxjvqaZSk57QqGdt923LdAVo3O0w5zM0OfnUQSEuKIjIwhSbl/3E/8eh+ySGIjLDnWrxzahWrQcdIBbn4nEGfmWyCrQrtxu1GuXJPVdDjf38r6HqpkV8tNo5UPefAxk6oSr3tcWtGXGpqVyNf9KA++BH6rwZO+B0nSSTo4KIJI5f6OssEiHlfYMrQ51Uq3QnvSJfmgh5RPMjmQ5SW79BtaZPCJ9xmapFN+B1nqmce762uY364qGrMfY+2q3FFf9m8TtrWuRp1KTancqCmDu9Si7d7POCh1uE+RHMgq0mzQBs65SbsozUcr/cYT44mIiJE++4yp4tXmFujUUqFqp7HoPZN+CimzSPvZ3+Y6l2bUkt9TT33gbilUpa33Sw5kFWjUdy1npCuU1AoyaVm3FyfY2z+ndMxQod5cKRi9yaT93vcp97eNoLZmKXKMu86T5CG0mQv9gNuDpbTLp556k+jMDhlhH6Wwvqwu+XPnpvemx1KAjSbKyxKba/Npp1WUskN2sNsk7eVPciArS72uyzkh/fCjvrWNS5Kk30q8/DeT2f4Tfo0IZMLfluW2VujUkcJVlwnomUoniZ+eI8PwdzLkQPsclM/Xjcm7jbHOUgYK4IPRVpY3LIhq9jFseebEF/mxJZwgF2NO9C1C9YI5yN5sMiOP2nwLIgo/D2RevHplxZMn9or//4w3vp9OMF8rN6Wyd2fc5vv8ys25fxbInO/qsqaXBgWLlGPwoa9Ye8Unv6eIL/haHWXEsKPcs3FXCkHSTkz4wOWZrWjfNJNA5mTAbb3u5NEsjvZKU545h307iaewOYDeiEaoZiWQORtyd20PNPIVRnu5CU8+ZbI+yed761nXU4t8Wl1Y+dQXJ7PT7JzZkcJl61Gs01ZufvIjUKlt+eeBLBh3d3vOnn1GWNhPG8dT/XogCyQ69BZ6NUtROXs7+s2/jFn6L9UP/DCQ+T3hyeHZDNSujmouDfrtt8XMPZMfjt1p9k1vTYFy9Wmx1R5HfymEKibJbqwb53SUWVOOcvnxJ6UbykpzJDhhqNeXPk3/SiDz5NkzC168yGTIZwb2PD86hynV6jPyRgxfgpXfg+zfMTxZ1ZAu1dVQVVOjVDVtxt2OTTdfsh8HsnACAx05ftwEX9+MdbbfDWSBZjzaM5E2eXOSPZsKXfRNuO+c9kP8biALfIXl+cWMalxbugDMTo+NJjz4nMk33PE6F5d2JqcUyFR+Fsi8rXA+NYLCedSoPvH4dx+j5mN9ldMj85I3V32mnDDH3OIedzYMoHjJsuRqsIzdjz/ike7Zlz8OZJGEhztJv5mn0m9HaUSt8JeIQCb8DclOaVYc6lWThnlVqN5tEutfSIfhjMfadLzxtD3OoiI5KFF5CssvWGWpr4p0FsPecDNLtPNJgawzSw3ssY8Iwc3qPtd0Z7P70mqGVitDuRZTGX3iFWE+Vtxd3p910on7+ecIop0f8Xx9W3KrFaad7h3uflS0T8pqlrylgDhvFBP6LUR39xNkA9EzO1CmEWjLlxszqKmhhnrZ8cyXXjOzJrvvSYj0w+/RbEbWr07N7ouYd+tb+0Og2RE2T2hKgyYNaLP4Ai9dYghL6fDt/w7Xq9OpVFabqYde8ColyMVKZ9a3e5jSpRo1u0xm2ukP8k8o9eNwvMXNVV3IVaAEjde/xuyrOx/uH+TSrvWcspU+TVko/nKJPZPbU7qoNsVHXOBtUBQxga95eFgX3YXrOPlOMZ/MJwMMdLvKmyIbrXkp7WN3Pj48wuWdazghzZcy4tPp7hrWdMspnZCLUa1lVzq0qEO1eu2oPXgTp81d8YtRBE0FWUf/gMuj6NmgItX7LmH9E6UTXJAlJsdXMrPPCEaNPctjPymMKyb9zC8HsrAvhJsuoW2ZQmiUHMzoLcZZur1GCtnNcT8c7EGPeiWp0mMOqx5JJ8LEeHh/nm3b9zB3znSWDm9CTvX6zL7hjO0XG56fPc+WJdflz8KUN52+O8HuqS3IV6EB7fZ+4nPAG4x3rGTfukNct/hKzKN5tNRuxsgNtzBO+RElSXvT/gQrhzaiZqv+DNr3Lk2TZdSHOxitbEW27EXoJIWMx58VeyIumESPW+ycPoRx/Zaif/hFps1pabjc5vzyntSp3JDJ96T/ZtK+HWQwizHtyqCiUoAS1fuz1iYRz0x2ftCdOYxtX53y7SaywMjvW3eDEFssrm5gTm/pvQw8wm1H3wz9z4LuygbzlCVvPR0677MnVpaqAp5z6+QJdKdOY+nAGlKoqsGkk9aYK98mR+IkBaQRzUtRtvUY5hj6ESN74Y83OLp/D9Nmz0N/YlNy5azBxBOWvHZ+j83Ni6ydeo7X0XHJtwPxNObuxkFU1ChG9k57ufbOi4jQ97y5vZPlY6aw52UkHikVh14WOB0biqYUyGrJR0274vnWgOtrJnHgVQTyxwLLZrO6zImh2aVAVoDy9VvTulUjGjZoTOHWC9nxwB6n4CilgQfJgu8vZ07PWpRqMphpt30JS2mfDPuAneEO5vXWYfCAfVx5456l+/YJ3ycCmfA3E0Kwp6l0Qh6BTslCFFFRoXDlBvRaeEg6YCYR/IOKFWKc+GK6ngGq2SnYVo+t37ljd0aR+Fld5NyEetLBtYx0JT2NqUs3sGH9Ho4cvYiZnyNHRzegdcNG1Oszk3VrN7Nr6xqOPLDjvew+HGHueD87woLeldDuOjy1U7/umnXobtnFprXr2L/rIgZPnH7eqd/fHtsbG5ilU0s6WOdApUhjWg1Yy96rWR8QkNJB12jndIb27kfrXlOSt0cq84foMHzkGCZuPctlm+A0HY1jXc14s7MLhfKqUaX9SEbMTl5Gd8USdEc3p9uI6Uzfex/j9M9HCXjJE1kTo3peynWYyIQ5a9m0YQNHz17E+Iv0schSUdhHHu4cz4DaRclTrTcTFi9n+catbNiwiYNHL/BQmi86pSUr0IxnhyegnUuDcu3HM06xviNnzqeZz8NkFzuGFE7tQF+p7VCGrTjDyQcf5bV7GfZVgpQ8vxpzbIkOXbp2TNOpX1faFn39TezauJfTZy2xC4/51oT3E78UyIKc+fJwD8sGa6OVXx0VrXo06LGUzafM5KEs/ckwM0mJcYS8Pc66YfVprK1N40Hz0dXTQ3fbNnQP3uHixWNc3zKQcqoVaD9yDjP0tqC/8STHj7+Qv4Y8f/sYcXFVP+pKIbq8FOrmLN7EZj19Tlww5KnNB7xP9qZmmfyUb96XAdMV+0h3Nbrj29Nv5ATGbLqOQfoqsJCvuBjvYWa3stTuNjq1U7+u9P3X27qLzbLfwe6r3Hvx5QcXStL30U36/e+YjE7TcmgUKkHDEbocNvXMGMrc7rJnentqlalOpc5ruesvHR8yaw11e84V/RH07toqTad+3Q1bkD2hY4f+No4dM8PKNyxjv0b3F5xe3ovmDStStN04Vq6S9sGm7WzYf569G1exaWBV6ZjRm1WGDnxIV6ka+v4Cu6a2plntatTuN1daVnrNrVvRO3iDU+fP8fjwcKqol6fNgKlMWb2Z9RuPc2CPCR9i45NDc6Qrb66tYbJ2HtTKdqL/lPkskb6j66Xv6q5tu7llH4N/SsdE6ffl9mg5bfOpU6lxP/pPXiUdvzaxf/8uab5o/BVfykC7O9yYUYJ86tnlv5lS2p3QmXeQ7Zdt8JQuYDL9/nm84s6OiQzo2jRNp35d/c2sk7Zl+7pNHD1iirl7sOjU/x8SgUz4m/HH18mA/dJV14jUzrfSvyfPZZ9lEj8cVa38LEv922n6D/1UwAdc7m2if98+itecyHz9S/IOv/IL2+uLWTdD9vfMn2Upv8XF3RXMmzQ4dZt1BoySP6vO9EtAFoOhRDoBmJ5a8W0d8rKQZVsepH0uXVZIwcZw3wqmplnXdNaeMCH1LhdKZI/rsTs/l6ED+yrN/63oXbLGKtNLYE++vDrHNmmeAfJ5p6F39BEW6V7D0/wc51crrXPaPo48/JhJTZQXX80vsF2aZ6BifbqHH/I63dE+3NGYB/umpa5v9blXqR34fyTI4jTH1oz/th2yMnU3+40c/tIJ5ZcCmY8N766uTfvaOrOYtfwmb6TJWWsolYVoP6wvrEN/gmIdffqjs8kouVkr4rN0YXKYhdLf5Z/H5MyeZfmVdwb7WS9/fdl3fhPHH0sndWmK7BE5rjcXMmW07BmPivUrlWXHjDH9zv1JZbe4cL+1iOljlZYdNJ7eSwww9wz5ee2h7L05XGGv7rfPVVY2GDjxPsPP2Z9317eyeZH0ejuf4ikt+r1AK3+Sx1Yp5CmtU2fiRjZdybyvVXx0BH62L3EJCMPu8SGOLVNabuYRzpk84tHllYyvVgTVcss5ZuGaoXZN1hXC7vZOtk9WWnbddW7JbtUf5Y6vzRlWDOjHYNnfJ2R8lqVMoONTHmyWPse+iuWl+daet5K/lvJ80pyEeD3kyJCBjJG/1njmrDnHI2lG5RHh0Z7veHduLkMGJB/nFuy6mqYD//eEORhxd/fUb+9DVsavYdVpC/m2ZOlCUfgpEcgEQRD+A76PNrFuzFg6DzmCqfT/TO7QIfyDJMRG4utgxoUxzdhx3wargIxxI97DiJtr+1OnUHnyzjDC7PN3xwwLQpaJQCYIgvAfkD1fMzQwEL+ACHkH97Q1F8I/jZflJU5OKErRfLlotvQGN+0yjoD8dHMxy7rmpnCZmgy7FIJjJqFNEH6VCGSCIAiCoOD+8gQHBqqgml2Fzusfc98pbUNowPP9rB3dgIp16lFp/H4sPOK/PVpMEP4DIpAJgiAIgkKo2xteHJvG6CY5aNV/AmPmKgYBKMr8QY3oNWAQfVecZt8TF+KzcsdqQcgCEcgEQRAEQUlUgAvWB/oxfZRSJ3alsmi/Acaym/EJwn+RCGSCIAiCIAi/mQhkgiAIgiAIv5kIZIIgCIIgCL+ZCGSCIAiCIAi/mQhkgiAIgiAIv5kIZIIgCIIgCL+ZCGSCIAiCIAi/mQhkgiAIgiAIv5kIZIIgCIIgCL+ZCGSCIAiCIAi/mQhkgiAIgiAIv5kIZIIgCIIgCL+ZCGSCIAiCIAi/mQhkgiAIgiAIv5kIZIIgCIIgCL+ZCGSCIAiCIAi/mQhkgiAIgiAIv5kIZIIgCIIgCL+ZCGSCIAiCIAi/mQhkgiAIgiAIv5kIZIIgCIIgCL+ZCGSCIAiCIAi/mQhkgiAIgiAIv5kIZIIgCIIgCL+ZCGSCIAiCIAi/FfwfNhqAaG8W/UcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 53,
     "metadata": {
      "image/png": {
       "height": 500,
       "width": 500
      }
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(filename=\"recall.png\", width=500, height=500)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Somente pare relembrar, o recall para o CoNNL é a porcentagem de entidades nomeadas - NE na solução, que foram encontradas pelo sistema"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 273,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.25"
      ]
     },
     "execution_count": 273,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recall = 0\n",
    "recall = TP / (TP + FN)\n",
    "recall"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Por fim calcularemos o F-measure amenizando as duas métricas calculadas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 254,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 254,
     "metadata": {
      "image/png": {
       "height": 400,
       "width": 400
      }
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Image(filename=\"fscore.png\", width=400, height=400)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 274,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.35294117647058826"
      ]
     },
     "execution_count": 274,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "F1_score = 2 * (precision * recall) / (precision + recall)\n",
    "F1_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Training model with train method"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# encontrar uma maneira de treinar para cada entidade\n",
    "nlp = spacy.load(\"en_core_web_sm\")\n",
    "train_data = [(u\"Uber blew through $1 million\", {\"entities\": [(0, 4, \"ORG\")]})]\n",
    "\n",
    "other_pipes = [pipe for pipe in nlp.pipe_names if pipe != \"ner\"]\n",
    "\n",
    "with nlp.disable_pipes(*other_pipes):\n",
    "    optimizer = nlp.begin_training()\n",
    "    for i in range(10):\n",
    "        random.shuffle(train_data)\n",
    "        for text, annotations in train_data:\n",
    "            nlp.update([text], [annotations], sgd=optimizer)\n",
    "            \n",
    "# salvando o modelo no disco         \n",
    "nlp.to_disk(\"models/modelo.bin\")\n",
    "# trazendo o modelo de volta ao disco\n",
    "nlp = spacy.load(\"models/modelo.bin\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# usando um modelo já pronto do Spacy\n",
    "nlp = spacy.load(\"en_core_web_sm\")\n",
    "# sobrescrevendo as entidades atuais com as novas entidades dos dados de treino a partir do patterns.json\n",
    "ruler_disk = EntityRuler(nlp, overwrite_ents=True).from_disk(\"models/patterns.json\")\n",
    "\n",
    "nlp.add_pipe(ruler_disk)\n",
    "\n",
    "doc = nlp(u\"MyCorp Inc. is a company in the U.S., Microsoft too\")\n",
    "print([(ent.text, ent.label_) for ent in doc.ents])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nlp.to_disk(\"models/modelo.bin\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "-"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}