whoosh-mb-tf.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "c56e13e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "import csv\n",
    "import nltk\n",
    "from whoosh.fields import Schema, TEXT, ID"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a0c1a917",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to\n",
      "[nltk_data]     C:\\Users\\meriem\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import csv\n",
    "from whoosh.fields import Schema, TEXT, ID\n",
    "from whoosh import index\n",
    "from whoosh.analysis import RegexTokenizer, StopFilter, Filter\n",
    "from whoosh.qparser import QueryParser\n",
    "from nltk.stem.isri import ISRIStemmer\n",
    "import os.path\n",
    "from nltk.corpus import stopwords\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "nltk.download('stopwords')\n",
    "\n",
    "# Charger la liste des mots vides en arabe de NLTK\n",
    "arabic_stopwords = set(stopwords.words(\"arabic\"))\n",
    "\n",
    "class ArabicStemFilter(Filter):\n",
    "    def __init__(self):\n",
    "        self.stemmer = ISRIStemmer()\n",
    "\n",
    "    def __call__(self, tokens):\n",
    "        for t in tokens:\n",
    "            t.text = self.stemmer.stem(t.text)\n",
    "            yield t\n",
    "\n",
    "# Créer un analyseur personnalisé avec la liste des mots vides de NLTK\n",
    "analyzer = RegexTokenizer() | StopFilter(stoplist=arabic_stopwords) | ArabicStemFilter()\n",
    "\n",
    "# Créer le schéma en utilisant l'analyseur personnalisé\n",
    "schema = Schema(\n",
    "    title=TEXT(stored=True),\n",
    "    path=ID(stored=True),\n",
    "    content=TEXT(stored=True, analyzer=analyzer, vector=True)  # Ajouter \"vector=True\"\n",
    ")\n",
    "\n",
    "# Créer le répertoire d'index s'il n'existe pas\n",
    "#if not os.path.exists(\"Index\"):\n",
    " #   os.mkdir(\"Index\")\n",
    "\n",
    "# Lister les fichiers dans le répertoire \"Corpus\"\n",
    "corpus_dir = \"E:/M2/RI/Corpus/Corpus\"\n",
    "file_names = os.listdir(corpus_dir)\n",
    "\n",
    "# Créer l'index avec le modèle BM25F\n",
    "ix = index.create_in(\"Index-2\", schema)  \n",
    "\n",
    "# Indexer les documents avec prétraitement\n",
    "with ix.writer() as writer:\n",
    "    for file_name in file_names:\n",
    "        with open(os.path.join(corpus_dir, file_name), 'r', encoding='windows-1256') as file:\n",
    "            content = file.read()\n",
    "            writer.add_document(title=file_name, content=content, path='')\n",
    "\n",
    "\n",
    "def search_vectorial(query_str, relevant_documents, limit=20):\n",
    "    with ix.searcher() as searcher:\n",
    "        # Définir un analyseur et un parseur liés à l'index\n",
    "        query_parser = QueryParser(\"content\", ix.schema)\n",
    "        query = query_parser.parse(query_str)\n",
    "\n",
    "        # Recherche Vectorielle (utilisation par défaut de Whoosh)\n",
    "        results_vectorial = searcher.search(query, terms=True, limit=limit)\n",
    "\n",
    "        # Marquer les résultats en fonction de la liste des documents pertinents\n",
    "        marked_results = []\n",
    "        true_positives = 0  # Count of relevant documents retrieved\n",
    "        false_positives = 0  # Count of irrelevant documents retrieved\n",
    "        total_relevant_documents = len(relevant_documents)\n",
    "\n",
    "        if total_relevant_documents == 0:\n",
    "            recall = precision = 0\n",
    "        else:\n",
    "            for rank, r in enumerate(results_vectorial, start=1):\n",
    "                is_relevant = r['title'] in relevant_documents\n",
    "                mark = 'YES' if is_relevant else 'NO'\n",
    "\n",
    "                # Calculate recall and precision for each document\n",
    "                if is_relevant:\n",
    "                    true_positives += 1\n",
    "                else:\n",
    "                    false_positives += 1\n",
    "\n",
    "                recall = round(true_positives / total_relevant_documents, 3)\n",
    "                precision = round(true_positives / (true_positives + false_positives), 3)\n",
    "\n",
    "                # Append information to the marked result\n",
    "                marked_results.append((rank, r['title'], mark, recall, precision))\n",
    "\n",
    "        # Write detailed results to CSV\n",
    "        csv_file_path = f\"results_{query_str.replace(' ', '_')}.csv\"\n",
    "        with open(csv_file_path, mode='w', encoding='utf-8', newline='') as csv_file:\n",
    "            csv_writer = csv.writer(csv_file)\n",
    "            csv_writer.writerow(['Rang', 'document', 'Pertinent', 'Rappel', 'Precision'])\n",
    "            csv_writer.writerows(marked_results)\n",
    "\n",
    "        return marked_results\n",
    "\n",
    "def calculate_interpolated_precision(results, recall_levels):\n",
    "    recall_values = [result[3] for result in results]\n",
    "    precision_values = [result[4] for result in results]\n",
    "\n",
    "    interpolated_precisions = []\n",
    "    for recall_level in recall_levels:\n",
    "        max_precision = 0\n",
    "        for recall, precision in zip(recall_values, precision_values):\n",
    "            if recall >= recall_level:\n",
    "                max_precision = max(max_precision, precision)\n",
    "        interpolated_precisions.append(max_precision)\n",
    "\n",
    "    return interpolated_precisions\n",
    "\n",
    "# Define the queries and relevant documents for each query\n",
    "queries = [\n",
    "    (\"الكلاسيكية OR الحلول\", []),\n",
    "    (\"أوبك\", [\"D493 A.txt\", \"D330 A.txt\"]),\n",
    "    (\"التفاضلية OR المعادلات\", [\"D508 A.txt\", \"D081 A.txt\", \"D105 A.txt\", \"D162 A.txt\", \"D329 A.txt\", \"D552 A.txt\", \"D464 A.txt\"])\n",
    "]\n",
    "\n",
    "# Interpolated Precision Recall Curve\n",
    "recall_levels = np.arange(0, 1.1, 0.1)\n",
    "interpolated_precision_all_queries = []\n",
    "\n",
    "# Rechercher et marquer les documents pertinents pour chaque requête\n",
    "for query, relevant_documents in queries:\n",
    "    marked_results_query = search_vectorial(query, relevant_documents, limit=20)\n",
    "    interpolated_precision_query = calculate_interpolated_precision(marked_results_query, recall_levels)\n",
    "    interpolated_precision_all_queries.append((query, interpolated_precision_query))\n",
    "\n",
    "# Plot Interpolated Precision Recall Curve for all queries\n",
    "plt.figure(figsize=(10, 6))\n",
    "for query, interpolated_precision_query in interpolated_precision_all_queries:\n",
    "    plt.plot(recall_levels, interpolated_precision_query, label=query)\n",
    "\n",
    "plt.title('Interpolated Precision-Recall Curve for Queries')\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Interpolated Precision')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "2d371af7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Rang    document Pertinent  Rappel  Precision\n",
      "0      1  D162 A.txt       YES   0.143      1.000\n",
      "1      2  D185 A.txt        NO   0.143      0.500\n",
      "2      3  D334 A.txt        NO   0.143      0.333\n",
      "3      4  D471 A.txt        NO   0.143      0.250\n",
      "4      5  D468 A.txt        NO   0.143      0.200\n",
      "5      6  D250 A.txt        NO   0.143      0.167\n",
      "6      7  D560 A.txt        NO   0.143      0.143\n",
      "7      8  D359 A.txt        NO   0.143      0.125\n",
      "8      9  D329 A.txt       YES   0.286      0.222\n",
      "9     10  D139 A.txt        NO   0.286      0.200\n",
      "10    11  D264 A.txt        NO   0.286      0.182\n",
      "11    12  D325 A.txt        NO   0.286      0.167\n",
      "12    13  D062 A.txt        NO   0.286      0.154\n",
      "13    14  D225 A.txt        NO   0.286      0.143\n",
      "14    15  D551 A.txt        NO   0.286      0.133\n",
      "15    16  D241 A.txt        NO   0.286      0.125\n",
      "16    17  D568 A.txt        NO   0.286      0.118\n",
      "17    18  D066 A.txt        NO   0.286      0.111\n",
      "18    19  D289 A.txt        NO   0.286      0.105\n",
      "19    20  D500 A.txt        NO   0.286      0.100\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "# Provide the correct file path as a string\n",
    "file_path = r'C:\\Users\\meriem\\Documents\\results_التفاضلية_OR_المعادلات.csv'\n",
    "\n",
    "# Read the CSV file into a pandas DataFrame\n",
    "df = pd.read_csv(file_path)\n",
    "\n",
    "# Print the DataFrame\n",
    "print(df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "49006ebf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   Rang    document Pertinent  Rappel  Precision\n",
      "0     1  D030 A.txt        NO     0.0      0.000\n",
      "1     2  D345 A.txt        NO     0.0      0.000\n",
      "2     3  D493 A.txt       YES     0.5      0.333\n",
      "3     4  D455 A.txt        NO     0.5      0.250\n",
      "4     5  D330 A.txt       YES     1.0      0.400\n",
      "5     6  D443 A.txt        NO     1.0      0.333\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "# Provide the correct file path as a string\n",
    "file_path = r'C:\\Users\\meriem\\Documents\\results_أوبك.csv'\n",
    "\n",
    "# Read the CSV file into a pandas DataFrame\n",
    "df = pd.read_csv(file_path)\n",
    "\n",
    "# Print the DataFrame\n",
    "print(df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "30622e04",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Empty DataFrame\n",
      "Columns: [Rang, document, Pertinent, Rappel, Precision]\n",
      "Index: []\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "# Provide the correct file path as a string\n",
    "file_path = r'C:\\Users\\meriem\\Documents\\results_الكلاسيكية_OR_الحلول.csv'\n",
    "\n",
    "# Read the CSV file into a pandas DataFrame\n",
    "df = pd.read_csv(file_path)\n",
    "\n",
    "# Print the DataFrame\n",
    "print(df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "72f98abd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "# Function to plot the Precision-Recall Curve\n",
    "def plot_precision_recall_curve(results, query):\n",
    "    recalls = [result[3] for result in results]\n",
    "    precisions = [result[4] for result in results]\n",
    "\n",
    "    plt.plot(recalls, precisions, label=f'Query: {query}')\n",
    "\n",
    "# Define the queries and relevant documents for each query\n",
    "queries = [\n",
    "    (\"الكلاسيكية OR الحلول\", []),\n",
    "    (\"أوبك\", [\"D493 A.txt\", \"D330 A.txt\"]),\n",
    "    (\"التفاضلية OR المعادلات\", [\"D508 A.txt\", \"D081 A.txt\", \"D105 A.txt\", \"D162 A.txt\", \"D329 A.txt\", \"D552 A.txt\", \"D464 A.txt\"])\n",
    "]\n",
    "\n",
    "# Plot non-interpolated Precision-Recall Curve for each query\n",
    "plt.figure(figsize=(8, 5))\n",
    "for query, relevant_documents in queries:\n",
    "    marked_results_query = search_vectorial(query, relevant_documents, limit=20)\n",
    "    plot_precision_recall_curve(marked_results_query, query)\n",
    "\n",
    "plt.title('Precision-Recall Curve')\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Precision')\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "bb3d75c2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No results for query: 'الكلاسيكية OR الحلول'\n",
      "--------------------\n",
      "Average Precision: 0.276\n",
      "F-measure: 0.324\n",
      "Precision at specific points:\n",
      "P5: 0.300\n",
      "P10: 0.267\n",
      "P15: 0.233\n",
      "P20: 0.217\n"
     ]
    }
   ],
   "source": [
    "def calculate_average_precision(results):\n",
    "    precision_values = [result[4] for result in results if result[3] > 0]  # Exclude queries with zero recall\n",
    "    if not precision_values:\n",
    "        return 0\n",
    "    return sum(precision_values) / len(precision_values)\n",
    "\n",
    "def calculate_f_measure(precision, recall):\n",
    "    if precision + recall == 0:\n",
    "        return 0\n",
    "    return 2 * (precision * recall) / (precision + recall)\n",
    "\n",
    "\n",
    "# Calculate average precision, F-measure, and precision at specific points\n",
    "average_precisions = []\n",
    "f_measures = []\n",
    "precisions_at_points = {\"P5\": [], \"P10\": [], \"P15\": [], \"P20\": []}\n",
    "\n",
    "for query, relevant_documents in queries:\n",
    "    marked_results_query = search_vectorial(query, relevant_documents, limit=20)\n",
    "    \n",
    "    recall_values = [result[3] for result in marked_results_query]\n",
    "    precision_values = [result[4] for result in marked_results_query]\n",
    "    \n",
    "    if recall_values and precision_values:\n",
    "        average_precisions.append(calculate_average_precision(marked_results_query))\n",
    "        f_measures.append(calculate_f_measure(precision_values[-1], recall_values[-1]))\n",
    "\n",
    "        precisions_at_points[\"P5\"].append(precision_values[min(4, len(precision_values)-1)])\n",
    "        precisions_at_points[\"P10\"].append(precision_values[min(9, len(precision_values)-1)])\n",
    "        precisions_at_points[\"P15\"].append(precision_values[min(14, len(precision_values)-1)])\n",
    "        precisions_at_points[\"P20\"].append(precision_values[min(19, len(precision_values)-1)])\n",
    "    else:\n",
    "        print(f\"No results for query: '{query}'\")\n",
    "\n",
    "# ...\n",
    "\n",
    "\n",
    "# Calculate overall averages\n",
    "average_precision_overall = sum(average_precisions) / len(average_precisions)\n",
    "f_measure_overall = sum(f_measures) / len(f_measures)\n",
    "precision_at_points_overall = {point: sum(values) / len(values) for point, values in precisions_at_points.items()}\n",
    "\n",
    "# Print the results\n",
    "print('--------------------')\n",
    "print(f\"Average Precision: {average_precision_overall:.3f}\")\n",
    "print(f\"F-measure: {f_measure_overall:.3f}\")\n",
    "print(\"Precision at specific points:\")\n",
    "for point, value in precision_at_points_overall.items():\n",
    "    print(f\"{point}: {value:.3f}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e58c317a",
   "metadata": {},
   "source": [
    "-----------------------------------------------------reformulation automatique--------------------------------------------------"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "57ed8cc0",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to\n",
      "[nltk_data]     C:\\Users\\meriem\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import csv\n",
    "from whoosh.fields import Schema, TEXT, ID\n",
    "from whoosh import index\n",
    "from whoosh.analysis import RegexTokenizer, StopFilter, Filter\n",
    "from whoosh.qparser import QueryParser\n",
    "from nltk.stem.isri import ISRIStemmer\n",
    "import os.path\n",
    "from nltk.corpus import stopwords\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "nltk.download('stopwords')\n",
    "\n",
    "# Charger la liste des mots vides en arabe de NLTK\n",
    "arabic_stopwords = set(stopwords.words(\"arabic\"))\n",
    "\n",
    "class ArabicStemFilter(Filter):\n",
    "    def __init__(self):\n",
    "        self.stemmer = ISRIStemmer()\n",
    "\n",
    "    def __call__(self, tokens):\n",
    "        for t in tokens:\n",
    "            t.text = self.stemmer.stem(t.text)\n",
    "            yield t\n",
    "\n",
    "# Créer un analyseur personnalisé avec la liste des mots vides de NLTK\n",
    "analyzer = RegexTokenizer() | StopFilter(stoplist=arabic_stopwords) | ArabicStemFilter()\n",
    "\n",
    "# Créer le schéma en utilisant l'analyseur personnalisé\n",
    "schema = Schema(\n",
    "    title=TEXT(stored=True),\n",
    "    path=ID(stored=True),\n",
    "    content=TEXT(stored=True, analyzer=analyzer, vector=True)  # Ajouter \"vector=True\"\n",
    ")\n",
    "\n",
    "# Créer le répertoire d'index s'il n'existe pas\n",
    "#if not os.path.exists(\"Index\"):\n",
    " #   os.mkdir(\"Index\")\n",
    "\n",
    "# Lister les fichiers dans le répertoire \"Corpus\"\n",
    "corpus_dir = \"E:/M2/RI/Corpus/Corpus\"\n",
    "file_names = os.listdir(corpus_dir)\n",
    "\n",
    "# Créer l'index avec le modèle BM25F\n",
    "ix = index.create_in(\"Index-2\", schema)  # Ajouter le paramètre weighting ici\n",
    "\n",
    "# Indexer les documents avec prétraitement\n",
    "with ix.writer() as writer:\n",
    "    for file_name in file_names:\n",
    "        with open(os.path.join(corpus_dir, file_name), 'r', encoding='windows-1256') as file:\n",
    "            content = file.read()\n",
    "            writer.add_document(title=file_name, content=content, path='')\n",
    "\n",
    "# ...\n",
    "\n",
    "# ...\n",
    "\n",
    "# ...\n",
    "\n",
    "# ...\n",
    "\n",
    "def search_vectorial(query_str, relevant_documents, limit=20):\n",
    "    with ix.searcher() as searcher:\n",
    "        # Définir un analyseur et un parseur liés à l'index\n",
    "        query_parser = QueryParser(\"content\", ix.schema)\n",
    "        query = query_parser.parse(query_str)\n",
    "\n",
    "        # Recherche Vectorielle (utilisation par défaut de Whoosh)\n",
    "        results_vectorial = searcher.search(query, terms=True, limit=limit)\n",
    "\n",
    "        # Marquer les résultats en fonction de la liste des documents pertinents\n",
    "        marked_results = []\n",
    "        true_positives = 0  # Count of relevant documents retrieved\n",
    "        false_positives = 0  # Count of irrelevant documents retrieved\n",
    "        total_relevant_documents = len(relevant_documents)\n",
    "\n",
    "        if total_relevant_documents == 0:\n",
    "            recall = precision = 0\n",
    "        else:\n",
    "            for rank, r in enumerate(results_vectorial, start=1):\n",
    "                is_relevant = r['title'] in relevant_documents\n",
    "                mark = 'YES' if is_relevant else 'NO'\n",
    "\n",
    "                # Calculate recall and precision for each document\n",
    "                if is_relevant:\n",
    "                    true_positives += 1\n",
    "                else:\n",
    "                    false_positives += 1\n",
    "\n",
    "                recall = round(true_positives / total_relevant_documents, 3)\n",
    "                precision = round(true_positives / (true_positives + false_positives), 3)\n",
    "\n",
    "                # Append information to the marked result\n",
    "                marked_results.append((rank, r['title'], mark, recall, precision))\n",
    "\n",
    "        # Write detailed results to CSV\n",
    "        csv_file_path = f\"results_{query_str.replace(' ', '_')}.csv\"\n",
    "        with open(csv_file_path, mode='w', encoding='utf-8', newline='') as csv_file:\n",
    "            csv_writer = csv.writer(csv_file)\n",
    "            csv_writer.writerow(['Rang', 'document', 'Pertinent', 'Rappel', 'Precision'])\n",
    "            csv_writer.writerows(marked_results)\n",
    "\n",
    "        return marked_results\n",
    "\n",
    "# ...\n",
    "\n",
    "\n",
    "\n",
    "def calculate_interpolated_precision(results, recall_levels):\n",
    "    recall_values = [result[3] for result in results]\n",
    "    precision_values = [result[4] for result in results]\n",
    "\n",
    "    interpolated_precisions = []\n",
    "    for recall_level in recall_levels:\n",
    "        max_precision = 0\n",
    "        for recall, precision in zip(recall_values, precision_values):\n",
    "            if recall >= recall_level:\n",
    "                max_precision = max(max_precision, precision)\n",
    "        interpolated_precisions.append(max_precision)\n",
    "\n",
    "    return interpolated_precisions\n",
    "\n",
    "# Define the queries and relevant documents for each query\n",
    "queries = [\n",
    "    (\"الكلاسيكية OR الحلول OR حل OR فك ORجذر OR بقاء ORلبث OR مكوث  \", []),\n",
    "    (\"أوبك OR منظمة OR الدول OR المصدرة OR للنفط \", [\"D493 A.txt\", \"D330 A.txt\"]),\n",
    "    (\"التفاضلية OR المعادلات\", [\"D508 A.txt\", \"D081 A.txt\", \"D105 A.txt\", \"D162 A.txt\", \"D329 A.txt\", \"D552 A.txt\", \"D464 A.txt\"])\n",
    "]\n",
    "\n",
    "# Interpolated Precision Recall Curve\n",
    "recall_levels = np.arange(0, 1.1, 0.1)\n",
    "interpolated_precision_all_queries = []\n",
    "\n",
    "# Rechercher et marquer les documents pertinents pour chaque requête\n",
    "for query, relevant_documents in queries:\n",
    "    marked_results_query = search_vectorial(query, relevant_documents, limit=20)\n",
    "    interpolated_precision_query = calculate_interpolated_precision(marked_results_query, recall_levels)\n",
    "    interpolated_precision_all_queries.append((query, interpolated_precision_query))\n",
    "\n",
    "# Plot Interpolated Precision Recall Curve for all queries\n",
    "plt.figure(figsize=(10, 6))\n",
    "for query, interpolated_precision_query in interpolated_precision_all_queries:\n",
    "    plt.plot(recall_levels, interpolated_precision_query, label=query)\n",
    "\n",
    "plt.title('Interpolated Precision-Recall Curve for Queries')\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Interpolated Precision')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "adf6bac6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Rang    document Pertinent  Rappel  Precision\n",
      "0      1  D162 A.txt       YES   0.143      1.000\n",
      "1      2  D185 A.txt        NO   0.143      0.500\n",
      "2      3  D334 A.txt        NO   0.143      0.333\n",
      "3      4  D471 A.txt        NO   0.143      0.250\n",
      "4      5  D468 A.txt        NO   0.143      0.200\n",
      "5      6  D250 A.txt        NO   0.143      0.167\n",
      "6      7  D560 A.txt        NO   0.143      0.143\n",
      "7      8  D359 A.txt        NO   0.143      0.125\n",
      "8      9  D329 A.txt       YES   0.286      0.222\n",
      "9     10  D139 A.txt        NO   0.286      0.200\n",
      "10    11  D264 A.txt        NO   0.286      0.182\n",
      "11    12  D325 A.txt        NO   0.286      0.167\n",
      "12    13  D062 A.txt        NO   0.286      0.154\n",
      "13    14  D225 A.txt        NO   0.286      0.143\n",
      "14    15  D551 A.txt        NO   0.286      0.133\n",
      "15    16  D241 A.txt        NO   0.286      0.125\n",
      "16    17  D568 A.txt        NO   0.286      0.118\n",
      "17    18  D066 A.txt        NO   0.286      0.111\n",
      "18    19  D289 A.txt        NO   0.286      0.105\n",
      "19    20  D500 A.txt        NO   0.286      0.100\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Provide the correct file path as a string\n",
    "file_path = r'C:\\Users\\meriem\\Documents\\results_التفاضلية_OR_المعادلات.csv'\n",
    "\n",
    "# Read the CSV file into a pandas DataFrame\n",
    "df = pd.read_csv(file_path)\n",
    "\n",
    "# Print the DataFrame\n",
    "print(df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "df6c2d0d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Empty DataFrame\n",
      "Columns: [Rang, document, Pertinent, Rappel, Precision]\n",
      "Index: []\n"
     ]
    }
   ],
   "source": [
    "\n",
    "\n",
    "# Provide the correct file path as a string\n",
    "file_path = r'C:\\Users\\meriem\\Documents\\results_الكلاسيكية_OR_الحلول_OR_حل_OR_فك_ORجذر_OR_بقاء_ORلبث_OR_مكوث__.csv'\n",
    "\n",
    "# Read the CSV file into a pandas DataFrame\n",
    "df = pd.read_csv(file_path)\n",
    "\n",
    "# Print the DataFrame\n",
    "print(df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "b12c5d49",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Rang    document Pertinent  Rappel  Precision\n",
      "0      1  D011 A.txt        NO     0.0      0.000\n",
      "1      2  D035 A.txt        NO     0.0      0.000\n",
      "2      3  D112 A.txt        NO     0.0      0.000\n",
      "3      4  D330 A.txt       YES     0.5      0.250\n",
      "4      5  D037 A.txt        NO     0.5      0.200\n",
      "5      6  D298 A.txt        NO     0.5      0.167\n",
      "6      7  D028 A.txt        NO     0.5      0.143\n",
      "7      8  D506 A.txt        NO     0.5      0.125\n",
      "8      9  D267 A.txt        NO     0.5      0.111\n",
      "9     10  D483 A.txt        NO     0.5      0.100\n",
      "10    11  D272 A.txt        NO     0.5      0.091\n",
      "11    12  D027 A.txt        NO     0.5      0.083\n",
      "12    13  D321 A.txt        NO     0.5      0.077\n",
      "13    14  D493 A.txt       YES     1.0      0.143\n",
      "14    15  D536 A.txt        NO     1.0      0.133\n",
      "15    16  D301 A.txt        NO     1.0      0.125\n",
      "16    17  D030 A.txt        NO     1.0      0.118\n",
      "17    18  D227 A.txt        NO     1.0      0.111\n",
      "18    19  D257 A.txt        NO     1.0      0.105\n",
      "19    20  D194 A.txt        NO     1.0      0.100\n"
     ]
    }
   ],
   "source": [
    "\n",
    "\n",
    "# Provide the correct file path as a string\n",
    "file_path = r'C:\\Users\\meriem\\Documents\\results_أوبك_OR_منظمة_OR_الدول_OR_المصدرة_OR_للنفط_.csv'\n",
    "\n",
    "# Read the CSV file into a pandas DataFrame\n",
    "df = pd.read_csv(file_path)\n",
    "\n",
    "# Print the DataFrame\n",
    "print(df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "dcb5f29e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "# Function to plot the Precision-Recall Curve\n",
    "def plot_precision_recall_curve(results, query):\n",
    "    recalls = [result[3] for result in results]\n",
    "    precisions = [result[4] for result in results]\n",
    "\n",
    "    plt.plot(recalls, precisions, label=f'Query: {query}')\n",
    "\n",
    "# Define the queries and relevant documents for each query\n",
    "queries = [\n",
    "    (\"الكلاسيكية OR الحلول OR حل OR فك ORجذر OR بقاء ORلبث OR مكوث  \", []),\n",
    "    (\"أوبك OR منظمة OR الدول OR المصدرة OR للنفط \", [\"D493 A.txt\", \"D330 A.txt\"]),\n",
    "    (\"التفاضلية OR المعادلات\", [\"D508 A.txt\", \"D081 A.txt\", \"D105 A.txt\", \"D162 A.txt\", \"D329 A.txt\", \"D552 A.txt\", \"D464 A.txt\"])\n",
    "]\n",
    "# Plot non-interpolated Precision-Recall Curve for each query\n",
    "plt.figure(figsize=(8, 5))\n",
    "for query, relevant_documents in queries:\n",
    "    marked_results_query = search_vectorial(query, relevant_documents, limit=20)\n",
    "    plot_precision_recall_curve(marked_results_query, query)\n",
    "\n",
    "plt.title('Precision-Recall Curve')\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Precision')\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "27d768b4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No results for query: 'الكلاسيكية OR الحلول OR حل OR فك ORجذر OR بقاء ORلبث OR مكوث  '\n",
      "--------------------\n",
      "Average Precision: 0.176\n",
      "F-measure: 0.165\n",
      "Precision at specific points:\n",
      "P5: 0.200\n",
      "P10: 0.150\n",
      "P15: 0.133\n",
      "P20: 0.100\n"
     ]
    }
   ],
   "source": [
    "# ...\n",
    "\n",
    "# Calculate average precision, F-measure, and precision at specific points\n",
    "average_precisions = []\n",
    "f_measures = []\n",
    "precisions_at_points = {\"P5\": [], \"P10\": [], \"P15\": [], \"P20\": []}\n",
    "\n",
    "# ...\n",
    "\n",
    "# ...\n",
    "\n",
    "# Assuming you have a list of transformed queries and relevant documents\n",
    "queries = [\n",
    "    (\"الكلاسيكية OR الحلول OR حل OR فك ORجذر OR بقاء ORلبث OR مكوث  \", []),\n",
    "    (\"أوبك OR منظمة OR الدول OR المصدرة OR للنفط \", [\"D493 A.txt\", \"D330 A.txt\"]),\n",
    "    (\"التفاضلية OR المعادلات\", [\"D508 A.txt\", \"D081 A.txt\", \"D105 A.txt\", \"D162 A.txt\", \"D329 A.txt\", \"D552 A.txt\", \"D464 A.txt\"])\n",
    "]\n",
    "\n",
    "# ...\n",
    "\n",
    "# Calculate average precision, F-measure, and precision at specific points\n",
    "average_precisions = []\n",
    "f_measures = []\n",
    "precisions_at_points = {\"P5\": [], \"P10\": [], \"P15\": [], \"P20\": []}\n",
    "\n",
    "# ...\n",
    "\n",
    "for query, relevant_documents in queries:\n",
    "    marked_results_query = search_vectorial(query, relevant_documents, limit=20)\n",
    "    \n",
    "    recall_values = [result[3] for result in marked_results_query]\n",
    "    precision_values = [result[4] for result in marked_results_query]\n",
    "    \n",
    "    if recall_values and precision_values:\n",
    "        average_precisions.append(calculate_average_precision(marked_results_query))\n",
    "        f_measures.append(calculate_f_measure(precision_values[-1], recall_values[-1]))\n",
    "\n",
    "        precisions_at_points[\"P5\"].append(precision_values[min(4, len(precision_values)-1)])\n",
    "        precisions_at_points[\"P10\"].append(precision_values[min(9, len(precision_values)-1)])\n",
    "        precisions_at_points[\"P15\"].append(precision_values[min(14, len(precision_values)-1)])\n",
    "        precisions_at_points[\"P20\"].append(precision_values[min(19, len(precision_values)-1)])\n",
    "    else:\n",
    "        print(f\"No results for query: '{query}'\")\n",
    "\n",
    "# ...\n",
    "\n",
    "# Calculate overall averages\n",
    "average_precision_overall = sum(average_precisions) / len(average_precisions)\n",
    "f_measure_overall = sum(f_measures) / len(f_measures)\n",
    "precision_at_points_overall = {point: sum(values) / len(values) for point, values in precisions_at_points.items()}\n",
    "\n",
    "# Print the results\n",
    "print('--------------------')\n",
    "print(f\"Average Precision: {average_precision_overall:.3f}\")\n",
    "print(f\"F-measure: {f_measure_overall:.3f}\")\n",
    "print(\"Precision at specific points:\")\n",
    "for point, value in precision_at_points_overall.items():\n",
    "    print(f\"{point}: {value:.3f}\")\n"
   ]
  },
  {
   "cell_type": "raw",
   "id": "84a94581",
   "metadata": {},
   "source": [
    "--------------------------------------------------REFORMULATION+validation------------------------------------------------------"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "2814a063",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to\n",
      "[nltk_data]     C:\\Users\\meriem\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import csv\n",
    "from whoosh.fields import Schema, TEXT, ID\n",
    "from whoosh import index\n",
    "from whoosh.analysis import RegexTokenizer, StopFilter, Filter\n",
    "from whoosh.qparser import QueryParser\n",
    "from nltk.stem.isri import ISRIStemmer\n",
    "import os.path\n",
    "from nltk.corpus import stopwords\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "nltk.download('stopwords')\n",
    "\n",
    "# Charger la liste des mots vides en arabe de NLTK\n",
    "arabic_stopwords = set(stopwords.words(\"arabic\"))\n",
    "\n",
    "class ArabicStemFilter(Filter):\n",
    "    def __init__(self):\n",
    "        self.stemmer = ISRIStemmer()\n",
    "\n",
    "    def __call__(self, tokens):\n",
    "        for t in tokens:\n",
    "            t.text = self.stemmer.stem(t.text)\n",
    "            yield t\n",
    "\n",
    "# Créer un analyseur personnalisé avec la liste des mots vides de NLTK\n",
    "analyzer = RegexTokenizer() | StopFilter(stoplist=arabic_stopwords) | ArabicStemFilter()\n",
    "\n",
    "# Créer le schéma en utilisant l'analyseur personnalisé\n",
    "schema = Schema(\n",
    "    title=TEXT(stored=True),\n",
    "    path=ID(stored=True),\n",
    "    content=TEXT(stored=True, analyzer=analyzer, vector=True)  # Ajouter \"vector=True\"\n",
    ")\n",
    "\n",
    "# Créer le répertoire d'index s'il n'existe pas\n",
    "#if not os.path.exists(\"Index\"):\n",
    " #   os.mkdir(\"Index\")\n",
    "\n",
    "# Lister les fichiers dans le répertoire \"Corpus\"\n",
    "corpus_dir = \"E:/M2/RI/Corpus/Corpus\"\n",
    "file_names = os.listdir(corpus_dir)\n",
    "\n",
    "# Créer l'index avec le modèle BM25F\n",
    "ix = index.create_in(\"Index-2\", schema)  # Ajouter le paramètre weighting ici\n",
    "\n",
    "# Indexer les documents avec prétraitement\n",
    "with ix.writer() as writer:\n",
    "    for file_name in file_names:\n",
    "        with open(os.path.join(corpus_dir, file_name), 'r', encoding='windows-1256') as file:\n",
    "            content = file.read()\n",
    "            writer.add_document(title=file_name, content=content, path='')\n",
    "\n",
    "# ...\n",
    "\n",
    "# ...\n",
    "\n",
    "# ...\n",
    "\n",
    "# ...\n",
    "\n",
    "def search_vectorial(query_str, relevant_documents, limit=20):\n",
    "    with ix.searcher() as searcher:\n",
    "        # Définir un analyseur et un parseur liés à l'index\n",
    "        query_parser = QueryParser(\"content\", ix.schema)\n",
    "        query = query_parser.parse(query_str)\n",
    "\n",
    "        # Recherche Vectorielle (utilisation par défaut de Whoosh)\n",
    "        results_vectorial = searcher.search(query, terms=True, limit=limit)\n",
    "\n",
    "        # Marquer les résultats en fonction de la liste des documents pertinents\n",
    "        marked_results = []\n",
    "        true_positives = 0  # Count of relevant documents retrieved\n",
    "        false_positives = 0  # Count of irrelevant documents retrieved\n",
    "        total_relevant_documents = len(relevant_documents)\n",
    "\n",
    "        if total_relevant_documents == 0:\n",
    "            recall = precision = 0\n",
    "        else:\n",
    "            for rank, r in enumerate(results_vectorial, start=1):\n",
    "                is_relevant = r['title'] in relevant_documents\n",
    "                mark = 'YES' if is_relevant else 'NO'\n",
    "\n",
    "                # Calculate recall and precision for each document\n",
    "                if is_relevant:\n",
    "                    true_positives += 1\n",
    "                else:\n",
    "                    false_positives += 1\n",
    "\n",
    "                recall = round(true_positives / total_relevant_documents, 3)\n",
    "                precision = round(true_positives / (true_positives + false_positives), 3)\n",
    "\n",
    "                # Append information to the marked result\n",
    "                marked_results.append((rank, r['title'], mark, recall, precision))\n",
    "\n",
    "        # Write detailed results to CSV\n",
    "        csv_file_path = f\"results_{query_str.replace(' ', '_')}.csv\"\n",
    "        with open(csv_file_path, mode='w', encoding='utf-8', newline='') as csv_file:\n",
    "            csv_writer = csv.writer(csv_file)\n",
    "            csv_writer.writerow(['Rang', 'document', 'Pertinent', 'Rappel', 'Precision'])\n",
    "            csv_writer.writerows(marked_results)\n",
    "\n",
    "        return marked_results\n",
    "\n",
    "# ...\n",
    "\n",
    "\n",
    "\n",
    "def calculate_interpolated_precision(results, recall_levels):\n",
    "    recall_values = [result[3] for result in results]\n",
    "    precision_values = [result[4] for result in results]\n",
    "\n",
    "    interpolated_precisions = []\n",
    "    for recall_level in recall_levels:\n",
    "        max_precision = 0\n",
    "        for recall, precision in zip(recall_values, precision_values):\n",
    "            if recall >= recall_level:\n",
    "                max_precision = max(max_precision, precision)\n",
    "        interpolated_precisions.append(max_precision)\n",
    "\n",
    "    return interpolated_precisions\n",
    "\n",
    "# Define the queries and relevant documents for each query\n",
    "queries = [\n",
    "    (\"الكلاسيكية OR الحلول OR حل ORجذر  \", []),\n",
    "    (\"أوبك OR منظمة OR الدول OR المصدرة OR للنفط \", [\"D493 A.txt\", \"D330 A.txt\"]),\n",
    "    (\"التفاضلية OR المعادلات\", [\"D508 A.txt\", \"D081 A.txt\", \"D105 A.txt\", \"D162 A.txt\", \"D329 A.txt\", \"D552 A.txt\", \"D464 A.txt\"])\n",
    "]\n",
    "\n",
    "# Interpolated Precision Recall Curve\n",
    "recall_levels = np.arange(0, 1.1, 0.1)\n",
    "interpolated_precision_all_queries = []\n",
    "\n",
    "# Rechercher et marquer les documents pertinents pour chaque requête\n",
    "for query, relevant_documents in queries:\n",
    "    marked_results_query = search_vectorial(query, relevant_documents, limit=20)\n",
    "    interpolated_precision_query = calculate_interpolated_precision(marked_results_query, recall_levels)\n",
    "    interpolated_precision_all_queries.append((query, interpolated_precision_query))\n",
    "\n",
    "# Plot Interpolated Precision Recall Curve for all queries\n",
    "plt.figure(figsize=(10, 6))\n",
    "for query, interpolated_precision_query in interpolated_precision_all_queries:\n",
    "    plt.plot(recall_levels, interpolated_precision_query, label=query)\n",
    "\n",
    "plt.title('Interpolated Precision-Recall Curve for Queries')\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Interpolated Precision')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "1ae2d64b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Rang    document Pertinent  Rappel  Precision\n",
      "0      1  D162 A.txt       YES   0.143      1.000\n",
      "1      2  D185 A.txt        NO   0.143      0.500\n",
      "2      3  D334 A.txt        NO   0.143      0.333\n",
      "3      4  D471 A.txt        NO   0.143      0.250\n",
      "4      5  D468 A.txt        NO   0.143      0.200\n",
      "5      6  D250 A.txt        NO   0.143      0.167\n",
      "6      7  D560 A.txt        NO   0.143      0.143\n",
      "7      8  D359 A.txt        NO   0.143      0.125\n",
      "8      9  D329 A.txt       YES   0.286      0.222\n",
      "9     10  D139 A.txt        NO   0.286      0.200\n",
      "10    11  D264 A.txt        NO   0.286      0.182\n",
      "11    12  D325 A.txt        NO   0.286      0.167\n",
      "12    13  D062 A.txt        NO   0.286      0.154\n",
      "13    14  D225 A.txt        NO   0.286      0.143\n",
      "14    15  D551 A.txt        NO   0.286      0.133\n",
      "15    16  D241 A.txt        NO   0.286      0.125\n",
      "16    17  D568 A.txt        NO   0.286      0.118\n",
      "17    18  D066 A.txt        NO   0.286      0.111\n",
      "18    19  D289 A.txt        NO   0.286      0.105\n",
      "19    20  D500 A.txt        NO   0.286      0.100\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Provide the correct file path as a string\n",
    "file_path = r'C:\\Users\\meriem\\Documents\\results_التفاضلية_OR_المعادلات.csv'\n",
    "\n",
    "# Read the CSV file into a pandas DataFrame\n",
    "df = pd.read_csv(file_path)\n",
    "\n",
    "# Print the DataFrame\n",
    "print(df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "d11d3aa9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Empty DataFrame\n",
      "Columns: [Rang, document, Pertinent, Rappel, Precision]\n",
      "Index: []\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Provide the correct file path as a string\n",
    "file_path = r'C:\\Users\\meriem\\Documents\\results_الكلاسيكية_OR_الحلول_OR_حل_ORجذر__.csv'\n",
    "\n",
    "# Read the CSV file into a pandas DataFrame\n",
    "df = pd.read_csv(file_path)\n",
    "\n",
    "# Print the DataFrame\n",
    "print(df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "0dfc18f6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Rang    document Pertinent  Rappel  Precision\n",
      "0      1  D011 A.txt        NO     0.0      0.000\n",
      "1      2  D035 A.txt        NO     0.0      0.000\n",
      "2      3  D112 A.txt        NO     0.0      0.000\n",
      "3      4  D330 A.txt       YES     0.5      0.250\n",
      "4      5  D037 A.txt        NO     0.5      0.200\n",
      "5      6  D298 A.txt        NO     0.5      0.167\n",
      "6      7  D028 A.txt        NO     0.5      0.143\n",
      "7      8  D506 A.txt        NO     0.5      0.125\n",
      "8      9  D267 A.txt        NO     0.5      0.111\n",
      "9     10  D483 A.txt        NO     0.5      0.100\n",
      "10    11  D272 A.txt        NO     0.5      0.091\n",
      "11    12  D027 A.txt        NO     0.5      0.083\n",
      "12    13  D321 A.txt        NO     0.5      0.077\n",
      "13    14  D493 A.txt       YES     1.0      0.143\n",
      "14    15  D536 A.txt        NO     1.0      0.133\n",
      "15    16  D301 A.txt        NO     1.0      0.125\n",
      "16    17  D030 A.txt        NO     1.0      0.118\n",
      "17    18  D227 A.txt        NO     1.0      0.111\n",
      "18    19  D257 A.txt        NO     1.0      0.105\n",
      "19    20  D194 A.txt        NO     1.0      0.100\n"
     ]
    }
   ],
   "source": [
    "# Provide the correct file path as a string\n",
    "file_path = r'C:\\Users\\meriem\\Documents\\results_أوبك_OR_منظمة_OR_الدول_OR_المصدرة_OR_للنفط_.csv'\n",
    "\n",
    "# Read the CSV file into a pandas DataFrame\n",
    "df = pd.read_csv(file_path)\n",
    "\n",
    "# Print the DataFrame\n",
    "print(df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "4e7e20b2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No results for query: 'الكلاسيكية OR الحلول OR حل ORجذر  '\n",
      "--------------------\n",
      "Average Precision: 0.176\n",
      "F-measure: 0.165\n",
      "Precision at specific points:\n",
      "P5: 0.200\n",
      "P10: 0.150\n",
      "P15: 0.133\n",
      "P20: 0.100\n"
     ]
    }
   ],
   "source": [
    "# ...\n",
    "\n",
    "# Calculate average precision, F-measure, and precision at specific points\n",
    "average_precisions = []\n",
    "f_measures = []\n",
    "precisions_at_points = {\"P5\": [], \"P10\": [], \"P15\": [], \"P20\": []}\n",
    "\n",
    "# ...\n",
    "\n",
    "# ...\n",
    "\n",
    "# Assuming you have a list of transformed queries and relevant documents\n",
    "queries = [\n",
    "    (\"الكلاسيكية OR الحلول OR حل ORجذر  \", []),\n",
    "    (\"أوبك OR منظمة OR الدول OR المصدرة OR للنفط \", [\"D493 A.txt\", \"D330 A.txt\"]),\n",
    "    (\"التفاضلية OR المعادلات\", [\"D508 A.txt\", \"D081 A.txt\", \"D105 A.txt\", \"D162 A.txt\", \"D329 A.txt\", \"D552 A.txt\", \"D464 A.txt\"])\n",
    "]\n",
    "\n",
    "\n",
    "# ...\n",
    "\n",
    "# Calculate average precision, F-measure, and precision at specific points\n",
    "average_precisions = []\n",
    "f_measures = []\n",
    "precisions_at_points = {\"P5\": [], \"P10\": [], \"P15\": [], \"P20\": []}\n",
    "\n",
    "# ...\n",
    "\n",
    "for query, relevant_documents in queries:\n",
    "    marked_results_query = search_vectorial(query, relevant_documents, limit=20)\n",
    "    \n",
    "    recall_values = [result[3] for result in marked_results_query]\n",
    "    precision_values = [result[4] for result in marked_results_query]\n",
    "    \n",
    "    if recall_values and precision_values:\n",
    "        average_precisions.append(calculate_average_precision(marked_results_query))\n",
    "        f_measures.append(calculate_f_measure(precision_values[-1], recall_values[-1]))\n",
    "\n",
    "        precisions_at_points[\"P5\"].append(precision_values[min(4, len(precision_values)-1)])\n",
    "        precisions_at_points[\"P10\"].append(precision_values[min(9, len(precision_values)-1)])\n",
    "        precisions_at_points[\"P15\"].append(precision_values[min(14, len(precision_values)-1)])\n",
    "        precisions_at_points[\"P20\"].append(precision_values[min(19, len(precision_values)-1)])\n",
    "    else:\n",
    "        print(f\"No results for query: '{query}'\")\n",
    "\n",
    "# ...\n",
    "\n",
    "# Calculate overall averages\n",
    "average_precision_overall = sum(average_precisions) / len(average_precisions)\n",
    "f_measure_overall = sum(f_measures) / len(f_measures)\n",
    "precision_at_points_overall = {point: sum(values) / len(values) for point, values in precisions_at_points.items()}\n",
    "\n",
    "# Print the results\n",
    "print('--------------------')\n",
    "print(f\"Average Precision: {average_precision_overall:.3f}\")\n",
    "print(f\"F-measure: {f_measure_overall:.3f}\")\n",
    "print(\"Precision at specific points:\")\n",
    "for point, value in precision_at_points_overall.items():\n",
    "    print(f\"{point}: {value:.3f}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a2a6b9df",
   "metadata": {},
   "source": [
    "-------------------------with scoring.TF_IDF()\"original-querry\"------------------------------------- "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "640db25c",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to\n",
      "[nltk_data]     C:\\Users\\meriem\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA04AAAIhCAYAAAB5deq6AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB+FUlEQVR4nO3deVhUZf/H8c+wIwiICyoiIJpLaZqmiZlbuZWVWVqWS2ppm5lt2uaSZfW02KYtpmZZWZalPWbuqWnmllZapoIruKCCCirL/fuD38wjsQ4CZwbfr+ua62nOnOV7hgPPfLzv8x2bMcYIAAAAAJAvD6sLAAAAAABXR3ACAAAAgEIQnAAAAACgEAQnAAAAACgEwQkAAAAACkFwAgAAAIBCEJwAAAAAoBAEJwAAAAAoBMEJAAAAAApBcAIuUjNmzJDNZtOGDRuc3jY1NVVjx47VihUrSr6wUjJ27FjZbLZibbtgwQKNHTu2ZAv6f+3bt1f79u0LXS8qKko2m83xCAwMVKtWrTRz5sxSqasg8fHxstlsmjFjhlPbDRw4UFFRUaVSU2HsNdsfHh4eqlSpkjp16qRFixZZUtO/5fX+REVFaeDAgUXa/uzZs3rnnXd09dVXq1KlSvLx8VF4eLh69+6tn376qeQLLiNvv/226tatKx8fH9lsNp04caLUj/nLL7/otttuU40aNeTj46MaNWqod+/eWr9+fakfuzBF/ZsBoOQRnAA4LTU1VePGjXOr4HQhFixYoHHjxlldhtq0aaO1a9dq7dq1juA7YMAATZkypUzrqFGjhtauXavrr7/eqe2effZZzZ07t5SqKpqHHnpIa9eu1apVq/Tqq6/qn3/+Uffu3bVy5UpL67pQR48eVZs2bTRy5EhddtllmjFjhpYuXarXXntNnp6e6tSpk7Zs2WJ1mU777bffNHz4cHXo0EHLli3T2rVrVbFixVI95ttvv602bdpo//79euWVV7RkyRL95z//0b59+3TVVVfpgw8+KNXjF2by5MmaPHmypTUAFysvqwsAALv09HTZbDZ5efGnKS8hISG66qqrHM+vvfZaRUZG6vXXX9d9992X5zaZmZnKyMiQr69vidXh6+ubo46iiomJKbEaiqt27dqO2tu0aaN69eqpXbt2+uijj3TNNddYXF3x9e/fX1u2bNGPP/6ojh075njt9ttv18iRI1WpUqUSOVZaWpr8/f1LZF+F+fPPPyVJ99xzj1q2bFki+0xNTVWFChXyfO3nn3/WiBEj1L17d82dOzfH36Lbb79dPXv21P33369mzZrpyiuvLJF6isped6NGjcr0uAD+hxEnAA4DBw5UYGCgdu7cqe7duyswMFARERF69NFHdfbsWUnZU56qVq0qSRo3bpxj6tP504n++ecf9e3bV9WqVZOvr68aNmyod999N8exVqxYIZvNpk8++USPPvqowsPD5evrq507dzpGUxYvXqy7775boaGhCggIUI8ePbR79+5cdU+bNk2XX365/Pz8FBoaqp49e2r79u2Fnu/s2bPVuXNn1ahRQ/7+/mrYsKFGjRql06dP53hP7LWfP9UrPj5ekmSM0eTJk9W0aVP5+/urUqVKuvXWW3PVaYzRK6+8osjISPn5+emKK67QDz/8UPgPpQAhISGqX7++9uzZI+l/09FeeeUVTZgwQdHR0fL19dXy5cslSRs2bNCNN96o0NBQ+fn5qVmzZvryyy9z7ffAgQO69957FRERIR8fH9WsWVO33nqrDh06lOM450/VO3LkiGMbX19fVa1aVW3atNGSJUtyvJf/nop25swZjR49WtHR0Y6pZQ888ECu6VhRUVG64YYbtHDhQl1xxRXy9/dXgwYNNG3atAt6D1u0aCFJjnOzS0xM1NChQ1WrVi35+PgoOjpa48aNU0ZGRo71zp49q/Hjx6thw4by8/NT5cqV1aFDB61Zs8axzrvvvqtrrrlG1apVU0BAgBo3bqxXXnlF6enpF1S73caNG/XDDz9o8ODBuUKT3ZVXXqnatWtLyn/aqv33zn5tS/9737/55hs1a9ZMfn5+GjdunJo1a6a2bdvm2kdmZqbCw8N1yy23OJadO3dOEyZMUIMGDRzXxt13360jR44UeF7t27fXXXfdJUlq1apVrr8zRfm9t/9N+/3339W5c2dVrFhRnTp1yveYEydOlM1m05QpU3L9A46Xl5djpGfixIk5jpHXFNS83uei/r1o3769LrvsMq1cuVKxsbGqUKGCBg0a5Hjt31P1ivoeL1u2TO3bt1flypXl7++v2rVrq1evXkpNTc33PQHwP/yzLoAc0tPTdeONN2rw4MF69NFHtXLlSj3//PMKDg7Wc889pxo1amjhwoXq2rWrBg8erCFDhkiSI0xt27ZNsbGxql27tl577TVVr15dP/74o4YPH66jR49qzJgxOY43evRotW7dWu+99548PDxUrVo1x2uDBw/Wddddp88++0z79u3TM888o/bt22vr1q0KCQmRlP0B5qmnntIdd9yhiRMnKikpSWPHjlXr1q21fv161atXL99ztU/TGjFihAICAvTXX3/p5Zdf1q+//qply5ZJyp5edvr0ac2ZM0dr1651bFujRg1J0tChQzVjxgwNHz5cL7/8so4dO6bx48crNjZWW7ZsUVhYmKTskDlu3DgNHjxYt956q/bt26d77rlHmZmZql+/frF/Vnv27HG893ZvvfWWLrnkEr366qsKCgpSvXr1tHz5cnXt2lWtWrXSe++9p+DgYH3xxRfq06ePUlNTHR9IDxw4oCuvvFLp6el66qmn1KRJEyUlJenHH3/U8ePHHefzb/369dOmTZv0wgsv6JJLLtGJEye0adMmJSUl5Vu/MUY333yzli5dqtGjR6tt27baunWrxowZ45iSeP5I2ZYtW/Too49q1KhRCgsL09SpUzV48GDVrVu32KNFcXFxkqRLLrnEsSwxMVEtW7aUh4eHnnvuOcXExGjt2rWaMGGC4uPjNX36dElSRkaGunXrplWrVmnEiBHq2LGjMjIy9Msvv2jv3r2KjY2VJO3atUt9+/Z1hMMtW7bohRde0F9//XXBwU+S4x6tm2+++YL3lZdNmzZp+/bteuaZZxQdHa2AgADVrFlTDz/8sP75558cv2OLFi3SwYMHdffdd0uSsrKydNNNN2nVqlV64oknFBsbqz179mjMmDFq3769NmzYkO/o1eTJk/X5559rwoQJmj59uho0aOC41p35vT937pxuvPFGDR06VKNGjcoVfu0yMzO1fPlytWjRQrVq1cpznYiICDVv3lxLlixRVlaWPDyc+/fnov69kKSEhATdddddeuKJJ/Tiiy/me6yivsfx8fG6/vrr1bZtW02bNk0hISE6cOCAFi5cqHPnzuU7CgfgPAbARWn69OlGklm/fr1j2YABA4wk8+WXX+ZYt3v37qZ+/fqO50eOHDGSzJgxY3Ltt0uXLqZWrVomOTk5x/IHH3zQ+Pn5mWPHjhljjFm+fLmRZK655pp8a+vZs2eO5T///LORZCZMmGCMMeb48ePG39/fdO/ePcd6e/fuNb6+vqZv376OZWPGjDEF/cnLysoy6enp5qeffjKSzJYtWxyvPfDAA3luu3btWiPJvPbaazmW79u3z/j7+5snnnjCUaefn1++59OuXbt867KLjIw03bt3N+np6SY9Pd3ExcU5fl6PP/64McaYuLg4I8nExMSYc+fO5di+QYMGplmzZiY9PT3H8htuuMHUqFHDZGZmGmOMGTRokPH29jbbtm3Ltxb7caZPn+5YFhgYaEaMGFHgOQwYMMBERkY6ni9cuNBIMq+88kqO9WbPnm0kmQ8++CDH+fv5+Zk9e/Y4lqWlpZnQ0FAzdOjQAo97fs0vv/yySU9PN2fOnDG//fabad26talRo4aJi4tzrDt06FATGBiY41jGGPPqq68aSebPP/80xhgzc+ZMI8l8+OGHhR7fLjMz06Snp5uZM2caT09Px++DMbnfH/t5DxgwoMB9Dhs2zEgyf/31V5FqyO93wf57d/57ERkZaTw9Pc3ff/+dY92jR48aHx8f89RTT+VY3rt3bxMWFua4zj7//HMjyXz99dc51lu/fr2RZCZPnlxgrXn9nXLm997+OzJt2rQCj2OMMYmJiUaSuf322wtcr0+fPkaSOXLkiOMY//65GZP7fS7q3wtjjGnXrp2RZJYuXZprv+3atcvxN6Oo7/GcOXOMJPPbb78VeH4A8sdUPQA52Gw29ejRI8eyJk2aOKaDFeTMmTNaunSpevbsqQoVKigjI8Px6N69u86cOaNffvklxza9evXKd3933nlnjuexsbGKjIx0TD1bu3at0tLScnUdi4iIUMeOHbV06dIC6929e7f69u2r6tWry9PTU97e3mrXrp0kFWmq3/fffy+bzaa77rorx7lWr15dl19+uaN5xtq1a3XmzJl8z6eoFixYIG9vb3l7eys6OlpffvmlHnroIU2YMCHHejfeeKO8vb0dz3fu3Km//vrLcfx//1wSEhL0999/S5J++OEHdejQQQ0bNixyXZLUsmVLzZgxQxMmTNAvv/xSpGlo9lG9f//8brvtNgUEBOT6+TVt2tQx3UyS/Pz8dMkll+S4Ns8/t4yMDBljcuzjySeflLe3t/z8/NS0aVP98ccfmj9/fo6pVt9//706dOigmjVr5thXt27dJMnRoe6HH36Qn5+fYwpVfjZv3qwbb7xRlStXdlxn/fv3V2Zmpnbs2FHo+2S1Jk2a5BiRk6TKlSurR48e+vjjj5WVlSVJOn78uL777jv179/fMc3t+++/V0hIiHr06JHjvWzatKmqV69erAYzxfm9L+jvjLPs15SzXTqL+vfCrlKlSvlOvfz3fovyHjdt2lQ+Pj6699579fHHH+c57RlAwQhOAHKoUKGC/Pz8cizz9fXVmTNnCt02KSlJGRkZevvttx0f8O2P7t27S8ru/nU++5S3vFSvXj3PZfbpX/b/zWsfNWvWLHCa2KlTp9S2bVutW7dOEyZM0IoVK7R+/Xp98803krJvgC/MoUOHZIxRWFhYrvP95ZdfHOdqryO/8ymqq6++WuvXr9eGDRu0bds2nThxQm+99ZZ8fHxyrPfv98N+/85jjz2Wq877779f0v9+LkeOHMl3mlJBZs+erQEDBmjq1Klq3bq1QkND1b9/fyUmJua7TVJSkry8vHJNNbTZbDl+znaVK1fOtQ9fX98cP6t/n9/HH3+cY/2HH35Y69ev1+rVq/Xqq68qPT1dN910U45jHTp0SPPnz8+1r0svvVRSzveqZs2aBU7X2rt3r9q2basDBw7ozTff1KpVq7R+/XrHfXNFuc4KYw+T9mmHJS2/39FBgwbpwIEDWrx4sSTp888/19mzZ3MEmkOHDunEiRPy8fHJ9X4mJibm+ntQFM7+3leoUEFBQUGF7rdKlSqqUKFCoe9jfHy8/P3987weC1LUvxd2Bf1t/Pd+i/Iex8TEaMmSJapWrZoeeOABxcTEKCYmRm+++aZT5wFczLjHCUCJqVSpkjw9PdWvXz898MADea4THR2d43lB/2qb14fuxMRE1a1bV9L/PkgnJCTkWu/gwYOqUqVKvvtetmyZDh48qBUrVjhGmSQ59R0xVapUkc1m06pVq/LsWmdfZq8zv/Mp6ncbBQcHO5oZFOTf76n9fRg9enSOm/bPZ7/PqmrVqtq/f3+R6vn3MSZNmqRJkyZp7969mjdvnkaNGqXDhw9r4cKFeW5TuXJlZWRk6MiRIznCkzFGiYmJxepa9u/v2fn39VarVi3He9imTRtVr15dd911l8aMGaN33nnHcS5NmjTRCy+8kOcxatasKSn7vVq9enWB97p8++23On36tL755psco4u//fab0+eWny5duuipp57St99+q65duxa6vv0fRs6ePZvjus0vxOT3O9qlSxfVrFlT06dPV5cuXTR9+nS1atUqR9e3KlWqqHLlyvleA8VpLe7s731RR4Y8PT3VsWNH/fDDD9q/f3+e/4Cwf/9+bdy4Mcf77Ofn52iec75/v59F/XvhbN3OvMdt27ZV27ZtlZmZqQ0bNujtt9/WiBEjFBYWpttvv71IxwMuZow4AXCa/f/g//2v5RUqVFCHDh20efNmNWnSRC1atMj1cOZfaWfNmpXj+Zo1a7Rnzx5HR6nWrVvL399fn376aY719u/fr2XLlhXYPcv+oeTfH1bef//9XOvmd7433HCDjDE6cOBAnufauHFjSdJVV10lPz+/fM+ntNWvX1/16tXTli1b8qyzRYsWjg9X3bp10/Llyx1T94qjdu3aevDBB3Xddddp06ZN+a5n//n8++f39ddf6/Tp0wX+/PLj7PV25513qn379vrwww8dP4sbbrhBf/zxh2JiYvJ8r+zBqVu3bjpz5kyBXwSc13VmjNGHH37o9Lnl54orrlC3bt300UcfOaY//tuGDRu0d+9eSXIE9a1bt+ZYZ/78+U4d1/6PJN9++61WrVqlDRs25Jq2eMMNNygpKUmZmZl5vpfFaYxyIb/3hRk1apSMMbr//vuVmZmZ47XMzEzdd999yszM1MMPP+xYHhUVpcOHD+fozHju3Dn9+OOPObYv6t8LZxXnPfb09FSrVq0cI58F/Z4C+B9GnAA4rWLFioqMjNR3332nTp06KTQ0VFWqVFFUVJTefPNNXX311Wrbtq3uu+8+RUVF6eTJk9q5c6fmz5+f7we7vGzYsEFDhgzRbbfdpn379unpp59WeHi4Y3pZSEiInn32WT311FPq37+/7rjjDiUlJWncuHHy8/PL1cHvfLGxsapUqZKGDRumMWPGyNvbW7NmzcrzS0LtH2hefvlldevWTZ6enmrSpInatGmje++9V3fffbc2bNiga665RgEBAUpISNDq1avVuHFj3XfffapUqZIee+wxTZgwIcf5jB071qmpehfi/fffV7du3dSlSxcNHDhQ4eHhOnbsmLZv365Nmzbpq6++kiSNHz9eP/zwg6655ho99dRTaty4sU6cOKGFCxdq5MiRatCgQa59Jycnq0OHDurbt68aNGigihUrav369Vq4cGG+I1ySdN1116lLly568sknlZKSojZt2ji66jVr1kz9+vUrtffjfC+//LJatWql559/XlOnTtX48eO1ePFixcbGavjw4apfv77OnDmj+Ph4LViwQO+9955q1aqlO+64Q9OnT9ewYcP0999/q0OHDsrKytK6devUsGFD3X777bruuuvk4+OjO+64Q0888YTOnDmjKVOm6Pjx4yV6DjNnzlTXrl3VrVs3DRo0SN26dVOlSpWUkJCg+fPn6/PPP9fGjRtVu3Ztde/eXaGhoRo8eLDGjx8vLy8vzZgxQ/v27XP6uIMGDdLLL7+svn37yt/fX3369Mnx+u23365Zs2ape/fuevjhh9WyZUt5e3tr//79Wr58uW666Sb17NnTqWNeyO99Ydq0aaNJkybp4Ycf1tVXX60HH3xQtWvX1t69e/Xuu+9q7dq1Gjt2rK677jrHNn369NFzzz2n22+/XY8//rjOnDmjt956K1fwKurfC2cV9T1+7733tGzZMl1//fWqXbu2zpw54+jqeO211xb7PQMuKpa1pQBgqfy66gUEBORaN68uXEuWLDHNmjUzvr6+RlKOzl9xcXFm0KBBJjw83Hh7e5uqVaua2NhYRzc8Y/7XVe+rr77Kt7ZFixaZfv36mZCQEEcXrX/++SfX+lOnTjVNmjQxPj4+Jjg42Nx0002OzmcFncOaNWtM69atTYUKFUzVqlXNkCFDzKZNm3J1jDt79qwZMmSIqVq1qrHZbLk6j02bNs20atXKBAQEGH9/fxMTE2P69+9vNmzY4FgnKyvLTJw40URERBgfHx/TpEkTM3/+/FwdsvITGRlprr/++gLXsXeO+89//pPn61u2bDG9e/c21apVM97e3qZ69eqmY8eO5r333sux3r59+8ygQYNM9erVjbe3t6lZs6bp3bu3OXToUI7j2N+jM2fOmGHDhpkmTZqYoKAg4+/vb+rXr2/GjBljTp8+7dhvXt3H0tLSzJNPPmkiIyONt7e3qVGjhrnvvvvM8ePHi3T+RX3/CntvbrvtNuPl5WV27txpjMnuHDl8+HATHR1tvL29TWhoqGnevLl5+umnzalTp3LU/9xzz5l69eoZHx8fU7lyZdOxY0ezZs0axzrz5883l19+ufHz8zPh4eHm8ccfNz/88IORZJYvX17g+1OUrnrn1/LWW2+Z1q1bm6CgIOPl5WVq1qxpbrnlFvPf//43x7q//vqriY2NNQEBASY8PNyMGTPGTJ06Nc+ueoVdd7GxsUaSufPOO/N8PT093bz66quO9yAwMNA0aNDADB06NM/f5/Pl9XfKrii/9/n9TSvMmjVrTK9evUxYWJjx8PAwkoyfn1+u99FuwYIFpmnTpsbf39/UqVPHvPPOO/l2LyzK34t27dqZSy+9NM9j5XXNF+U9Xrt2renZs6eJjIw0vr6+pnLlyqZdu3Zm3rx5Tr8/wMXKZsy/Wg4BgMVmzJihu+++W+vXry/SPT0AUJpmzpypAQMG6IknntDLL79sdTkALMJUPQAAgAL0799fCQkJGjVqlAICAvTcc89ZXRIACxCcAAAACvHkk0/qySeftLoMABZiqh4AAAAAFIJ25AAAAABQCIITAAAAABSC4AQAAAAAhbjomkNkZWXp4MGDqlixouMb3QEAAABcfIwxOnnypGrWrCkPj4LHlC664HTw4EFFRERYXQYAAAAAF7Fv3z7VqlWrwHUuuuBUsWJFSdlvTlBQkMXVAAAAALBKSkqKIiIiHBmhIBddcLJPzwsKCiI4AQAAACjSLTw0hwAAAACAQhCcAAAAAKAQBCcAAAAAKMRFd48TAABASTDGKCMjQ5mZmVaXAqAA3t7e8vT0vOD9EJwAAACcdO7cOSUkJCg1NdXqUgAUwmazqVatWgoMDLyg/RCcAAAAnJCVlaW4uDh5enqqZs2a8vHxKVJHLgBlzxijI0eOaP/+/apXr94FjTwRnAAAAJxw7tw5ZWVlKSIiQhUqVLC6HACFqFq1quLj45Wenn5BwYnmEAAAAMXg4cHHKMAdlNSIML/xAAAAAFAIghMAAAAAFILgBAAAcJFo3769JkyYUOh6AwcO1JAhQ4q0z08//VRRUVGFrhcfHy+bzab9+/dLkmbMmKG6desW6Rhlrajn5C6ioqL06aefFrpeUa+Pohg7dqyuvfbaC9qHM9dhWSA4AQAAABb6888/1bt3b1WtWlW+vr6qV6+enn322Vzt7qOiomSz2WSz2eTv768GDRroP//5j4wxFlV+cSE4AQAAACUoMzNTxhhlZWUVuu4vv/yiVq1a6dy5c/rvf/+rHTt26MUXX9THH3+s6667TufOncux/vjx45WQkKDt27frscce01NPPaUPPvigtE4F5yE4AQAAXCBjjFLPZVjyKOvRhh07dqhhw4Z68cUXy/S4pakkzyk+Pl4xMTH6+OOPNWjQoALXNcZo8ODBatiwob755hu1bNlSkZGRuu222zR//nytXbtWb7zxRo5tKlasqOrVqysqKkpDhgxRkyZNtGjRoguuG4Wz9HucVq5cqf/85z/auHGjEhISNHfuXN18880FbvPTTz9p5MiR+vPPP1WzZk098cQTGjZsWNkUDAAAkIe09Ew1eu5HS469bXwXVfApu490R44c0R133KF+/fqV2TFLW0meU82aNTV//nzVrFlT7du3L3Dd3377Tdu2bdNnn32Wq7395ZdfrmuvvVaff/65nnzyyVzbGmP0008/afv27apXr94F143CWTridPr0aV1++eV65513irR+XFycunfvrrZt22rz5s166qmnNHz4cH399delXCkAAAAkqVmzZurevbt27NhhdSnaunWrJk2alO/rM2fO1GWXXSYfHx81bdpUW7duzXM9Z8/p119/1TXXXKOAgABFREToiy++cLzm4+Ojxo0bq3LlyoU2mLAfr2HDhnm+3rBhw1w1PfnkkwoMDJSvr686dOggY4yGDx9epLqdcezYMfXs2VN16tSRn5+fPD09ValSJV155ZX66aefSvx47sDSEadu3bqpW7duRV7/vffeU+3atR2/IA0bNtSGDRv06quvqlevXqVUZenZd3Kf/j72t9VluKzo4GjFhMRYXQYAAIXy9/bUtvFdLDt2WQoICJCnp6f++9//SpKysrJK9cuA3377bbVr105NmjSRJL355ptq166dIwhNmjRJI0aMyLXd0aNHdffdd+uZZ55Ru3btNHnyZD322GN5Tmtz9pzuueceXXLJJfr222+1atUqPfTQQ7rhhhsUGBiozZs36/rrr9epU6d088036+WXX1aNGjWKde7GmFxf3vr4449r4MCBOnLkiJ5++ml17NhRsbGxxdp/QdLT0xUeHq5bb71VX3/9tb777js1aNBAV111lUJCQkr8eO7A0uDkrLVr16pz5845lnXp0kUfffSR0tPT5e3tnWubs2fP6uzZs47nKSkppV5nUa05sEYT1pVMy8fyyNfTV4tuXaRQv1CrSwEAoEA2m61Mp8uVtvT0dPn6+ub52s6dOxUUFKSqVatKkpKSkhQcHFykfUqSp6dzQW/16tVKTEx0BKeJEyeqQoUKatq0qSTlChZ2AQEBCgkJ0a5du3T99ddrzpw5+R7D2XMKDw/X/v37lZaWpueee07jx493vPb000+rcePGGjlypN599121adNGv/76q6pUqZJrP5dccokkadu2bY7zOd9ff/2VaxpelSpVVLduXdWtW1dff/216tatq6uuuuqCW39L2T8j+88nLCxM77zzjgYMGKA///xTa9euVcuWLYu1L0nas2eP+vXrp3Xr1ql79+765JNPFBgYWOg+8rsOreBWv+GJiYkKCwvLsSwsLEwZGRk6evRonml+4sSJGjduXFmV6JQq/lXUrFozq8twSX8f+1upGan6K+kvxYaX/L+iAABwsTt16pRiY2P10Ucf6corr8zx2r59+9S6dWtJUmpqqgYNGqQ33nhDNWrUUExMTI59fP755zk+9L/xxhvatGmTPvnkk1z7lJRvINm8ebNWrlyphx9+OMfy1q1b6/PPP9cLL7yglJQUHT58WDt37pSUPZ0sNPR//8A6evRoderUSddee638/f31yy+/aMyYMerYsaOaNWum//73vwoKCnL6nGbOnKnk5GQ99NBDkqQvvvhCL7zwgoYOHSofHx999913jvV37typJ598Ul26dFHnzp3VsGFDff311xo6dGiuc27atKkaNGigN954Q7fffnuOUa4tW7ZoyZIlmjhxYp7vlyRVqlRJDz30kB577DFt3rw53xB5vj179qhz585atmyZwsPDc7y2b9++HKNJs2fP1oIFC/T777+revXqufZljFG7du30yCOPqGfPngXu68knn9TZs2c1b9483XvvvXr11Vc1duzYIl+HrsCtgpOU+18V7J1k8rtQRo8erZEjRzqep6SkKCIiovQKdEKnyE7qFNnJ6jJc0iPLH9GSvUu088ROghMAAKVg27Zt+v333x0jOXY7duzQmjVr9MADD0jKnrY2f/58nTlzRjfffLMiIiLk4eGhjRs36t1339Xp06c1Y8YMx/aLFy/Oc/Rk+vTpatCggSpUqJBnPePGjVOtWrVyLQ8ICNCvv/6qN954Q3FxcfL399cHH3ygq666Sp9++qk6duwoSTp+/Lheeukl9e3b17FtvXr19Nlnn+nEiROqVauWpk+frocfftjpc3rttddyfBFrUFCQXn75Zb3wwgvq1KmTxo0bp7lz50qSqlWrpi+++EIRERHaunWr4uPjFRkZmec522w2TZ06VZ07d1avXr00evRoVa9eXevWrdOjjz6q1q1b5zkN8XwPPPCAXn75ZX399de69dZbC1xXktatW6eTJ0/mCk1Lly7Vvn37dMUVVziWTZs2TUOGDMkzNEnZTTVWrVqlqVOn5lielJSk+fPna9SoUTmWp6am6syZM2rVqpU2b94sqejXoStwq3bk1atXV2JiYo5lhw8flpeXlypXrpznNr6+vgoKCsrxgOuz39u0K3mXxZUAAFA+RUZGysfHRxMnTtSxY8e0d+9eTZs2TZ06ddJVV13lGEEIDAzU8uXLZYzRE088oa5du+qGG27Q5MmT1blzZ23YsEGNGjVy7LdevXqaM2eONm7cqJSUFK1bt079+/fXF198UeDoybp169S2bdtcy//55x9FREToxRdf1GeffaavvvpKXbt21V133aWqVavqmWeekST5+fnJ19dXe/fuzbH9qVOn9PrrrystLU0tWrQo1jkFBwfn2m9GRoa+/vprbdy4Ua1atXIs/89//qOjR4/q5ptv1qxZszRlyhR17do13/Nu06aNfvnlF3l6eqp79+6qW7euRo8erQEDBmjx4sWFTlWrWrWq+vXrp7Fjxxbpe6Pq1aunxMREffDBB0pOTtauXbs0adIk3Xrrrbr11ltznMvBgwdVu3btfPcVGhqq0NBQvfrqq0pMTFRCQoK+/PJLtW3bVmFhYbrnnnsc67766quKjIxU//799dVXXzmm8RX1OnQJxkVIMnPnzi1wnSeeeMI0bNgwx7Jhw4aZq666qsjHSU5ONpJMcnJyccpEGVmwe4G5bMZlpu9/+1pdCgAAOaSlpZlt27aZtLQ0q0u5YJ988ompXLmykWQkmapVq5onn3zSnD59utj7TEpKMl27dnXsU5Lp0KGDWbZsWYHb+fr6mu+//z7HssTERFO1alUzZsyYIh373XffNf7+/qZ9+/amZ8+eplmzZsbHx8c0btzYzJ8/v7inZNauXWtCQ0PNFVdcYXr16mViY2NNcHCwqV69unnllVdMZmZmsfdthVdeecUEBAQ4fj61a9c2L774osnIyMix3i233GL69OlT4L4WLVpkatWq5dhXxYoVzbBhw8zhw4fzXP/o0aMmIiLCvPrqq45lpXEdnq+g31lnsoHNmDL+1rTznDp1yjFHtVmzZnr99dfVoUMHhYaGqnbt2ho9erQOHDigmTNnSspuR37ZZZdp6NChuueee7R27VoNGzZMn3/+eZG76qWkpCg4OFjJycmMPrmwHcd3qNe8Xgr0DtSaO9YUac4uAABl4cyZM4qLi1N0dLT8/PysLueCnTt3Trt375anp6fq1KnjdPOG/Bw+fFiJiYmKiIhQpUqVCl3/yiuvVEhIiN566y3Vrl1bW7du1QMPPKATJ05o/fr1+c4u+rc9e/Zo4cKFOnPmjGrUqKGWLVsW2ha8KE6ePKl58+bp0KFDqlKlipo0aaLLL7/cbT+jpKWlKS4uThUqVFBkZGSe57F27Vq1adNG9957r4YNG6Z69erp9OnT2rhxo7Zs2eKYipeZmandu3crMzNTMTExORq2ZWVlKS0tTYmJiVqzZo2efvppVapUST///HOO5hCldR1KBf/OOpMNLJ2qt2HDBjVr1kzNmmU3SBg5cqSaNWum5557TpKUkJCQY1g0OjpaCxYs0IoVK9S0aVM9//zzeuutt9yyFTkKFhUUJU+bp06ln9Kh1ENWlwMAQLnl4+OjBg0aqF69eiX6YbVatWpq0qRJkUKTlP21M/v27VOjRo0UGBio2NhY1apVS6tWrSpyaJKyp36tXbtWwcHBeuKJJ4pbfi4VK1ZUenq6tm7dqmXLlum3335z29AkSf7+/mrUqJGioqLyPY/WrVtr3rx5Wrt2rZo1a6bAwECFhYWpV69eyszMdKzn6empevXqqUGDBrm6XI8fP16BgYGqW7eunnjiCfXu3TtXaJJK7zosSZaOOFmBESf3ceO3NyouOU7vXfue2oS3sbocAAAklb8RJ1dz5MgRJSQkqFq1avk2JShMfHy8KlasqIMHD6p+/fry8fEpkdqSkpJ08uRJSdlByplA5+6SkpKUkJAgX19f1a5du8htwhMSEpSUlKRq1aqpWrVqpVxl3kpqxMntuurh4lE3pK7ikuO088ROghMAABeJqlWrOr5PqbjsU/NKOthUrlz5ogpL5yvuudeoUaPYXwDsatyqqx4uLo7OeiforAcAAABrEZzgsghOAAAAcBUEJ7isusF1JWV/l9NFdiseAAAAXAzBCS4rMihSXjYvnU4/rcTTiYVvAAAAAJQSghNclrentyKDIiVJO0/stLgaAAAAXMwITnBp3OcEAEDJOHDggKKiotSxY0erSwHcEsEJLq1uSPZ9Tow4AQBwYXbt2qUePXpo/PjxVpcCuCWCE1waI04AAJSMsLAwnTp1SmvXrrW6FMAtEZzg0uwjTruSdynLZFlcDQAA7qt+/fqKi4vT2bNny/zYQ4YM0cCBA8v8uEBJIjjBpUUERcjLw0tpGWlKOJ1gdTkAAAC4SBGc4NK8PbwVFRQliel6AAAAsA7BCS6PBhEAALiXOXPmqHr16vr555+tLgUoMV5WFwAUpk5IHUmMOAEAXJgxUnqqNcf2riDZbNYcOx979uzRW2+9pSZNmlhdClBiCE5weYw4AQBcXnqq9GJNa4791EHJJ8CaY+chMzNTQ4cO1ffff69Tp06pYsWKVpcElAim6sHl2VuSxyXH0VkPAIBStmPHDr300ksyxhS43ttvv63169fnWr5q1SpVrFhRDz30kDw8sj9qZmVlOf4bcFeMOMHl1a5YW94e3krLSNPBUwdVq2Itq0sCACAn7wrZIz9WHbsYli5dqmHDhunAgQN6/PHHNW7cOEnSRx99pG3btslWyPS/GTNmKDU1VVdeeWWO5VdffbW2bdumyMhIVaiQXVtSUpLq1KlTrDoBV0H0h8vz8vBSVHCUJO5zAgC4KJste7qcFY9i3t905513qkOHDvrggw80ceJErVixQpJ05syZQr/r6bvvvtOmTZt0xRVXSJI2bNig++67T5Lk5eWlhg0bOkLTzp07tXz5cjVt2rRYdQKughEnuIW6wXX1z/F/tPPETrWLaGd1OQAAlAtJSUkKCwtTZGSkNm/erPbt26tfv366+uqrdcMNN+jOO+9URESE0tLStG/fPv3xxx9auXKlNm/erFGjRum6666TJJ08eVIffvih/Pz8dNVVV6lGjRpKTU3VsmXL9N5776lBgwbq3bu3xWcLXBiCE9yC/T4nRpwAACgZX3zxhUaNGqUePXro7Nmz8vT0lCS1aNFCa9as0SuvvKLHHntMhw8flq+vr2rUqKEGDRropptu0qxZs1S/fn3Hvjp06KCPP/5YU6ZM0bRp03Tq1CkFBgYqMjJSI0eO1KhRo+Tn52fVqQIlguAEt0BnPQAALpx9Op4ktW/fXr/88ovWrFmjdu3aqU2bNo7XrrjiCn3xxReO55mZmY5glZ8777xTd955Z4nXDLgK7nGCW6CzHgAAJSMjI0PJycn6448/NHHiRF177bXq37+/mjdvnu82MTExio+PL7siARfEiBPcQkTFCPl4+OhM5hkdOHlAEUERVpcEAIBbuvbaa/XTTz9Jkho0aKDXXntNw4YNK3Cb+fPnq2ZNi76nCnARBCe4BU8PT0UHR+vv439r54mdBCcAAIrpww8/VGZmpsLDw4v85bSNGzcu5aoA18dUPbgNR4OIZBpEAABQXPXq1VODBg2KHJoAZCM4wW3QIAIAAABWITjBbdCSHAAAAFYhOMFt2Eec4pLjlJmVaXE1AAAAuJgQnOA2wgPD5evpq7OZZ7X/1H6rywEAAMBFhOAEt+Hp4ak6wXUkcZ8TAAAAyhbBCW6F+5wAACi+9u3ba8KECYWuN3DgQA0ZMqQMKsptxowZqlu3bpkca+zYsbr22mvL5FilLT4+XjabTfv3Fz4rx2azafXq1SV27Av9mf279rK8BpxBcIJbsQcnRpwAAHB/WVlZMsZYXUaZ+f7779W+fXtVrFhRFSpU0JVXXqkZM2bkWMceIuyP4OBgXXXVVZo/f741RcOB4AS3EhPMiBMAAOXFoEGD9PHHH1tdRrEYY5SVlVXk8Pf222/rpptuUmxsrNatW6etW7fq9ttv17Bhw/TYY4/lWn/JkiVKSEjQunXr1LJlS/Xq1Ut//PFHaZwKisjL6gIAZ9g768UnxysjK0NeHlzCAAC4q7Fjx17QF/Haw4unp2cJVlU0H3/8sVasWCEpewrkwIED81133759evTRRzVixAi9+OKLjuWPPvqofHx8NHz4cN12221q1aqV47XKlSurevXqql69ul544QW9/fbbWr58uS677LLSOiUUghEnuJXwiuHy8/TTuaxz2n+SznoAANdgjFFqeqolj7Ke6ubsSEt+PvroI11zzTVq0KCB5s6dW6x9/PTTT+rUqVOxa7gQPXr00NixYzV27Fj16NGjwHXnzJmj9PT0PEeWhg4dqsDAQH3++ed5bpuenq4PP/xQkuTt7X3hhRfTmjVr1LVrV1WqVElvvvmmZXVIUmZmpuM6LEv8cz3ciofNQ9HB0dp+bLt2ndilqOAoq0sCAEBpGWlq9VmrwlcsBev6rlMF7wpldjxnRlryc+DAAQ0dOlTvvPOOzp49q7vuukvx8fGqWrWqU/tp3ry53n//faePX1QzZ87UK6+8oh07dqhRo0aaOXOmmjRpIil7RKhy5cpF2s+OHTsUHBysGjVq5HrNx8dHderU0Y4dO3Isj42NlYeHh9LS0pSVlaWoqCj17t37wk+qGP7880916tRJAwYM0EMPPaTmzZsXus0PP/ygO+64Q1999ZWuu+66EqslPj5e7du319ixY7VixYpc94iVJkac4Hbs0/VoEAEAQNl4/vnn9eeff0rKf6Tl3XffVZs2bRzP7aNSeUlKSlJmZqaaNGmiLl26KDU1NVdwKIqKFSuqfv36OnfunMaNG6fIyEgFBwerffv22rZtm9P7O9/Ro0d19913q1evXlq4cKHq1q2bY8Ro+vTpCggIUHh4uMaMGaOzZ88W+1jGGNlsthzLZs+erc2bN2vevHmqW7eupk6dqtDQ0GIfozBHjhzRzTffLD8/P7Vo0cIRjiVpwYIFuvTSS/Xee+/p+uuvV2ZmphYtWlTg/rp166aXXnpJ/fr1U1pa2gXV9umnnzqaY9SsWVPz5893XId2W7duVWBgoFJSUi7oWAVhxAluh5bkAABX4+/lr3V911l27JKWnp4uX19fSdmjDc8995weeOABSfmPtBw/ftzxoTUzM1O1atXSO++8o169euVa97LLLlOXLl3Upk0b+fn5ydfXV0FBQU7VmJmZKQ8PD506dUpdunRRYmKixo4dq5iYGD3zzDMaOXKkFi5cWOh55nd/VEBAgEJCQrRr1y5df/31mjNnjuO1jIwMjRgxQg899JCuuOIKjRs3zhFy8nLJJZcoOTlZBw8eVM2aNXO8du7cOe3evVsdO3bMsTwiIkL16tVTvXr1FBgYqF69emnbtm2qVq1aUd6eAs9ZUq7zHj9+vDZu3Kh58+Zp8eLFuvnmm3Xo0CH5+vqqXbt2euGFF9SjRw/t2bNHv//+u+rUqaNduwr+LHbvvffq+eef13fffafbb7/d6ZrsHnnkEcfIoo+Pjxo3bixJOa7DlJQUnT592rGv0sCIE9yOY8QpmREnAIBrsNlsquBdwZLHv0cqiurUqVNq0qSJ1q9fn+u1ffv2KSQkRNL/PtQWNGqwdetWvfPOO+rWrZuk7BGlxMRENWrUKM/1PTw8tHDhQv3999965ZVX5O/vr3r16uW57u7du/XVV1/lWn711Vdr4sSJuummmxQQEKBNmzbp7rvv1jXXXKMWLVooMTFRUvZozjXXXJPnfVTnn6ckjR49WkuWLJEk+fv765dfflFWVpY6duyotm3bOoLh4cOHlZKSonvuuUe9e/fWkiVLNH/+fP399995nkOvXr3k5eWl1157Lddr7733nk6fPq077rgjz20lqV27drrsssv0wgsv5LvOv3311Vfq2LFjrvvQ9u3bJ0mO887KylKfPn00ZcoUNWrUSM2aNVOzZs2UnJyskydPSpLq1Kmjpk2b6rffflP//v11+PDhQkOTlP1zDg0N1dGjRyVJb7zxhvr165drPXtNwcHBee4nPT29wOvv6NGjeuqpp3TllVcWefpkcTDiBLdjH3Gisx4AAMW3bds2/f777457dux27NihNWvWOEaYmjRpopYtW6pNmzYaPny4mjZtKg8PDyUmJmrr1q36+eeftXr1avXo0UPjx4+XJFWtWlX169fX3Xffreeff15XXHGFgoKClJGRoZMnT2rv3r3auHGjZs+erZUrV2ry5Mny8/PLs85nnnlGfn5+uu2223Isj4qK0tNPP60aNWrok08+0caNGxUXF6fZs2dr6dKljoYKR44c0apVqzR16tQc2yclJWn+/PkaNWqUpOwRs5deekl9+/Z1rFOvXj199tlnOnHihGrVqqXp06fr4YcfVnBwsHx8fPTWW2/pxhtv1JdffilfX19Vr149z3OoXbu2XnnlFT322GPy8/NTv3795O3tre+++05PPfWUHn300Rwd9fLy6KOP6rbbbtMTTzyh8PDwAteVpOXLl6t69eq5gvX06dMVFBTk+ILZ7777Tl9++aXatm2rP/74Q9WqVZPNZtM999yjKlWqSJJuu+02eXt7a/v27QoMDCz02J9++qlq1Kihjz/+WNu2bVO7du0kSYsXL1bTpk1zrT99+nQ1aNBAFSrkfa/ekCFDNHToUP3+++9q166dAgMDlZSUpG3btunXX3/V4sWLFRUVle+IX4kxF5nk5GQjySQnJ1tdCoopMyvTXPnpleayGZeZXSd2WV0OAOAik5aWZrZt22bS0tKsLsVp7dq1M88//7wxxpjExETj4+NjxowZY5KSksyePXvMRx99ZGrVqmXatGljzp4969guJSXFPPPMM+bSSy81/v7+xsfHx4SHh5trr73WjBo1yqxduzbXsfbs2WN69eplvL29jaQcDy8vL9OoUSMzbNgws2nTphzbTZ8+3cTExDieh4eHm88++yzX/rt06WKioqJMQECAkWS8vb1NVFSU6du3b4560tPTTWhoqLnnnntMQkKCOXjwoJk9e7Zp2LChqV+/vklKSjLGGJOammp8fX3N999/n+M4J0+eNM8++6zx8PAwq1evdix///33TXR0tAkMDDTXXnttjtfy891335m2bduagIAA4+fnZ5o3b26mTZuWY524uDgjyWzevDnH8qysLFO/fn1z33335blv+3b79u0zxhjz+uuvm6pVq5ply5aZlJQUs3nzZjNixAjj4eFh3n77bcd2zz//vKlWrZoxxpiMjAzzzz//OPZhjDHHjx/Ps57z/ftn1qBBA+Pr62uaNGli5syZ41g+fPhwU69ePbNhwwaTnJxsfvnlF9OvXz/j5eVl5s6dm+/+srKyzIcffmhiY2NNUFCQ8fLyMlWqVDFXXXWVue+++8zcuXNNenp6vvUV9DvrTDYgOMEt9Znfx1w24zKzKH6R1aUAAC4y7hyc/u2TTz4xlStXdgSaqlWrmieffNKcPn26xI6RlpZmduzYYTZt2mS2bt1q4uPjc4Sywvj6+pr//ve/OZadPn3ahIWFmRdeeKFI+1i0aJGpVauW4zwrVqxohg0bZg4fPpxjvXfffdf4+/ub9u3bm549e5pmzZoZHx8f07hxYzN//vwi1+wK0tLSzF133WU8PDwc592iRYscQcYYY8aMGWMiIyPz3c+pU6eMpDzDsbOSkpJM165dc4ToDh06mGXLll3wvgtSUsHJZkwZN/+3WEpKioKDg5WcnOz0TYhwHU+vflrzds3T/U3v132X32d1OQCAi8iZM2cUFxen6OjofKeXuRN7cwJPT0/VqVPHki+TLUjz5s1VpUoVTZo0SeHh4dq+fbueeeYZrVmzRn/++aeioqKKtJ/MzEzt3r1bmZmZiomJyfc7kfbs2aOFCxfqzJkzqlGjhlq2bFnkY7iiY8eOaf/+/apevXqejSXGjh2rGTNmKD4+Pt993Hjjjfrjjz80fvx4xcbGKjAwUPv27dP8+fPVs2dPXX755U7VdPjwYSUmJioiIkKVKlVy9pScVtDvrDPZgJtD4JbsDSLorAcAwIXx8fFRgwYNrC4jX++//77uuuuuHI0mWrRooSVLljgVaDw9PfNtQHG+yMhIrV27Vu3bt9cTTzyRoy23OwoNDb3gNuazZs3S6NGj9fDDD+vYsWOO5bGxsRo0aJDT+6tWrdoFdwe0AiNOcEsr96/UA0sfUN2Qupp7U/G+bRwAgOIobyNO7uLo0aM6ePCgQkNDVatWrVI9Vnx8vCpWrKiDBw+qfv368vHxKdXjuZPExESlpqaqevXq+TZzcDWMOOGiZh9xik+JV3pWurw98h5uBwAA5UOVKlUcXd5Km30kqzRbW7ur/DoHXgz4Hie4pRoBNVTBq4IysjK0N2Wv1eUAAACgnCM4wS3ZbDbH9zntPMEX4QIAyl5WVpbVJQAogpK6M4mpenBbMSEx+v3o7zSIAACUKR8fH3l4eOjgwYOqWrWqfHx8cn3JKADXYIzRkSNHZLPZ8u2kWFQEJ7itmGBGnAAAZc/Dw0PR0dFKSEjQwYMHrS4HQCFsNptq1ap1wa32CU5wW/apeow4AQDKmo+Pj2rXrq2MjAxlZmZaXQ6AAnh7e5fI95MRnOC27J319qbsVXpmurw96awHACg79qk/Fzr9B4B7oDkE3Fb1gOoK8A5QhsnQnpQ9VpcDAACAcozgBLdls9n+d59TMvc5AQAAoPQQnODWuM8JAAAAZYHgBLdGcAIAAEBZIDjBrdkbRNCSHAAAAKWJ4AS3Zh9x2puyV+cyz1lcDQAAAMorghPcWliFMAV6ByrTZCo+Jd7qcgAAAFBOEZzg1mw2G/c5AQAAoNQRnOD2uM8JAAAApY3gBLfHiBMAAABKG8EJbo/gBAAAgNJGcILbs0/V23tyr85mnrW4GgAAAJRHBCe4var+VVXRp6KyTJbik+OtLgcAAADlEMEJbs9ms9EgAgAAAKWK4IRygfucAAAAUJoITigXYoKzgxMjTgAAACgNBCeUC4w4AQAAoDQRnFAu2O9x2n9qv85knLG4GgAAAJQ3BCeUC1X8qyjIJyi7s15KvNXlAAAAoJwhOKFcoLMeAAAAShPBCeUG9zkBAACgtFgenCZPnqzo6Gj5+fmpefPmWrVqVYHrz5o1S5dffrkqVKigGjVq6O6771ZSUlIZVQtXZg9OjDgBAACgpFkanGbPnq0RI0bo6aef1ubNm9W2bVt169ZNe/fuzXP91atXq3///ho8eLD+/PNPffXVV1q/fr2GDBlSxpXDFdmn6jHiBAAAgJJmaXB6/fXXNXjwYA0ZMkQNGzbUpEmTFBERoSlTpuS5/i+//KKoqCgNHz5c0dHRuvrqqzV06FBt2LChjCuHK7KPOO0/uV9pGWkWVwMAAIDyxLLgdO7cOW3cuFGdO3fOsbxz585as2ZNntvExsZq//79WrBggYwxOnTokObMmaPrr78+3+OcPXtWKSkpOR4onyr7VVaIb4iMjOKS46wuBwAAAOWIZcHp6NGjyszMVFhYWI7lYWFhSkxMzHOb2NhYzZo1S3369JGPj4+qV6+ukJAQvf322/keZ+LEiQoODnY8IiIiSvQ84DpsNhsNIgAAAFAqLG8OYbPZcjw3xuRaZrdt2zYNHz5czz33nDZu3KiFCxcqLi5Ow4YNy3f/o0ePVnJysuOxb9++Eq0froWW5AAAACgNXlYduEqVKvL09Mw1unT48OFco1B2EydOVJs2bfT4449Lkpo0aaKAgAC1bdtWEyZMUI0aNXJt4+vrK19f35I/AbgkRpwAAABQGiwbcfLx8VHz5s21ePHiHMsXL16s2NjYPLdJTU2Vh0fOkj09PSVlj1QBjDgBAACgNFg6VW/kyJGaOnWqpk2bpu3bt+uRRx7R3r17HVPvRo8erf79+zvW79Gjh7755htNmTJFu3fv1s8//6zhw4erZcuWqlmzplWnARdiH3E6cOqAUtNTLa4GAAAA5YVlU/UkqU+fPkpKStL48eOVkJCgyy67TAsWLFBkZKQkKSEhIcd3Og0cOFAnT57UO++8o0cffVQhISHq2LGjXn75ZatOAS4m1C9UoX6hOnbmmOKS43RplUutLgkAAADlgM1cZHPcUlJSFBwcrOTkZAUFBVldDkrBoB8HaX3iek1oM0E31b3J6nIAAADgopzJBpZ31QNKWkwwDSIAAABQsghOKHfs9znRIAIAAAAlheCEcoeW5AAAAChpBCeUO/aW5AdPH6SzHgAAAEoEwQnlTiW/Sgr1C5Uk7U7ebXE1AAAAKA8ITiiX+CJcAAAAlCSCE8ol7nMCAABASSI4oVxixAkAAAAlieCEcokRJwAAAJQkghPKJfuIU8LpBJ1OP21xNQAAAHB3BCeUS8G+wariX0USo04AAAC4cAQnlFtM1wMAAEBJITih3KJBBAAAAEoKwQnlFiNOAAAAKCkEJ5RbjDgBAACgpBCcUG7ZR5wOpR7SyXMnLa4GAAAA7ozghHIryCdI1fyrSWK6HgAAAC4MwQnlGvc5AQAAoCQQnFCu2YMT9zkBAADgQhCcUK4x4gQAAICSQHBCuWbvrEdwAgAAwIUgOKFcqxNSR5J0OO2wUs6lWFwNAAAA3BXBCeVakE+QqlXI7qy3+8Rui6sBAACAuyI4odzji3ABAABwoQhOKPdoEAEAAIALRXBCuceIEwAAAC4UwQnlHiNOAAAAuFAEJ5R7McHZwelI2hEln022uBoAAAC4I4ITyr1An0BVD6guiVEnAAAAFA/BCRcF+3Q97nMCAABAcRCccFGoG5zdIIIRJwAAABQHwQkXBRpEAAAA4EIQnHBRoCU5AAAALgTBCRcF+4hT0pkknThzwtpiAAAA4HYITrgoVPCuoJoBNSUx6gQAAADnEZxw0eA+JwAAABQXwQkXDVqSAwAAoLgITrhoOEackhlxAgAAgHMITrho2DvrMVUPAAAAziI44aJRJ7iOJOnYmWM6fua4xdUAAADAnRCccNGo4F1B4YHhkrjPCQAAAM4hOOGiQmc9AAAAFAfBCRcVOusBAACgOAhOuKjQIAIAAADFQXDCRYWpegAAACgOr+JstGPHDq1YsUKHDx9WVlZWjteee+65EikMKA11guvIJpuOnz2upLQkVfavbHVJAAAAcANOB6cPP/xQ9913n6pUqaLq1avLZrM5XrPZbAQnuDR/L3+FB4Zr/6n92nViF8EJAAAAReJ0cJowYYJeeOEFPfnkk6VRD1Dq6obU1f5T+7XzxE61rNHS6nIAAADgBpy+x+n48eO67bbbSqMWoExwnxMAAACc5XRwuu2227Ro0aLSqAUoE7QkBwAAgLOcnqpXt25dPfvss/rll1/UuHFjeXt753h9+PDhJVYcUBocLcmTd8kYk+M+PQAAACAvNmOMcWaD6Ojo/Hdms2n37t0XXFRpSklJUXBwsJKTkxUUFGR1ObDAmYwzavVZK2WZLC3vvVxV/KtYXRIAAAAs4Ew2cHrEKS4urtiFAa7Az8tPtQJrae/Jvdp5YifBCQAAAIW6oC/ANcbIyQErwCXQIAIAAADOKFZwmjlzpho3bix/f3/5+/urSZMm+uSTT0q6NqDU0CACAAAAznB6qt7rr7+uZ599Vg8++KDatGkjY4x+/vlnDRs2TEePHtUjjzxSGnUCJYoRJwAAADjD6eD09ttva8qUKerfv79j2U033aRLL71UY8eOJTjBLTg6652gsx4AAAAK5/RUvYSEBMXGxuZaHhsbq4SEhBIpCiht0cHR8rB5KOVcio6mHbW6HAAAALg4p4NT3bp19eWXX+ZaPnv2bNWrV69EigJKm6+nryIqRkjiPicAAAAUzumpeuPGjVOfPn20cuVKtWnTRjabTatXr9bSpUvzDFSAq4oJjtGelD3adWKXWtdsbXU5AAAAcGFOjzj16tVL69atU5UqVfTtt9/qm2++UZUqVfTrr7+qZ8+epVEjUCrorAcAAICicnrESZKaN2+uTz/9tKRrAcrU+Q0iAAAAgIIUKTilpKQoKCjI8d8Fsa8HuLrzW5LTWQ8AAAAFKVJwqlSpkhISElStWjWFhITk+QHT/sEzMzOzxIsESkN0cLQ8bZ46mX5Sh1MPKywgzOqSAAAA4KKKFJyWLVum0NBQSdLy5ctLtSCgrPh4+iiiYoTiU+K168QughMAAADyVaTg1K5duzz/G3B3dUPqKj4lXjtP7FRseO7vJwMAAACkYnTVW7hwoVavXu14/u6776pp06bq27evjh8/XqLFAaXNcZ9TMg0iAAAAkD+ng9Pjjz/uaBDx+++/a+TIkerevbt2796tkSNHlniBQGmyd9ajJTkAAAAK4nQ78ri4ODVq1EiS9PXXX6tHjx568cUXtWnTJnXv3r3ECwRKk33EafeJ3XTWAwAAQL6cHnHy8fFRamqqJGnJkiXq3LmzJCk0NLTQVuWAq4kKipKXzUun0k/pUOohq8sBAACAi3J6xOnqq6/WyJEj1aZNG/3666+aPXu2JGnHjh2qVatWiRcIlCZvT2/VDqqt3cm7tfPETlUPqG51SQAAAHBBTo84vfPOO/Ly8tKcOXM0ZcoUhYeHS5J++OEHde3atcQLBErb+V+ECwAAAOTF6RGn2rVr6/vvv8+1/I033iiRgoCyFhMSo8V7FtMgAgAAAPkq0ojT+fcupaSkFPhw1uTJkxUdHS0/Pz81b95cq1atKnD9s2fP6umnn1ZkZKR8fX0VExOjadOmOX1cwI4RJwAAABSmSCNOlSpVUkJCgqpVq6aQkJA8O4/ZO5JlZmYW+eCzZ8/WiBEjNHnyZLVp00bvv/++unXrpm3btql27dp5btO7d28dOnRIH330kerWravDhw8rIyOjyMcE/q1ucHZL8l0ndtFZDwAAAHkqUnBatmyZQkNDJUnLly8vsYO//vrrGjx4sIYMGSJJmjRpkn788UdNmTJFEydOzLX+woUL9dNPP2n37t2OeqKiokqsHlycIoMi5WXzUmpGqhJPJ6pGYA2rSwIAAICLKVJwateuXZ7/fSHOnTunjRs3atSoUTmWd+7cWWvWrMlzm3nz5qlFixZ65ZVX9MknnyggIEA33nijnn/+efn7++e5zdmzZ3X27FnHc1qm49+8Pb0VGRSpXcm7tPPEToITAAAAcnG6q9706dP11Vdf5Vr+1Vdf6eOPPy7yfo4eParMzEyFhYXlWB4WFqbExMQ8t9m9e7dWr16tP/74Q3PnztWkSZM0Z84cPfDAA/keZ+LEiQoODnY8IiIiilwjLh7c5wQAAICCOB2cXnrpJVWpUiXX8mrVqunFF190uoB/309S0D0mWVlZstlsmjVrllq2bKnu3bvr9ddf14wZM5SWlpbnNqNHj1ZycrLjsW/fPqdrRPlXNyT7Pic66wEAACAvTrcj37Nnj6Kjo3Mtj4yM1N69e4u8nypVqsjT0zPX6NLhw4dzjULZ1ahRQ+Hh4QoODnYsa9iwoYwx2r9/v+rVq5drG19fX/n6+ha5LlycGHECAABAQZwecapWrZq2bt2aa/mWLVtUuXLlIu/Hx8dHzZs31+LFi3MsX7x4sWJjY/Pcpk2bNjp48KBOnTrlWLZjxw55eHioVq1aRT428G/2EaddybuUZbIsrgYAAACuxungdPvtt2v48OFavny5MjMzlZmZqWXLlunhhx/W7bff7tS+Ro4cqalTp2ratGnavn27HnnkEe3du1fDhg2TlD3Nrn///o71+/btq8qVK+vuu+/Wtm3btHLlSj3++OMaNGhQvs0hgKKICIqQl4eX0jLSlHA6wepyAAAA4GKcnqo3YcIE7dmzR506dZKXV/bmWVlZ6t+/v9P3OPXp00dJSUkaP368EhISdNlll2nBggWKjIyUJCUkJOSY/hcYGKjFixfroYceUosWLVS5cmX17t1bEyZMcPY0gBy8PbwVFRSlnSd2ateJXQoPDLe6JAAAALgQmzHGFGfDHTt2aMuWLfL391fjxo0dYcfVpaSkKDg4WMnJyQoKCrK6HLiQx396XAvjF+qR5o9o0GWDrC4HAAAApcyZbOD0iJNdVFSUjDGKiYlxjDwB7owGEQAAAMiP0/c4paamavDgwapQoYIuvfRSx1S64cOH66WXXirxAoGyQktyAAAA5Mfp4DR69Ght2bJFK1askJ+fn2P5tddeq9mzZ5docUBZso84xSXH0VkPAAAAOTg9x+7bb7/V7NmzddVVV+X4otpGjRpp1y6mOMF9RVSMkLeHt9Iy0nTg1AFFVIywuiQAAAC4CKdHnI4cOaJq1arlWn769OkcQQpwN14eXooOzv5yZ+5zAgAAwPmcDk5XXnml/vvf/zqe28PShx9+qNatW5dcZYAF7NP1uM8JAAAA53N6qt7EiRPVtWtXbdu2TRkZGXrzzTf1559/au3atfrpp59Ko0agzMQE01kPAAAAuTk94hQbG6s1a9YoNTVVMTExWrRokcLCwrR27Vo1b968NGoEyoy9sx7BCQAAAOdzasQpPT1d9957r5599ll9/PHHpVUTYBn7VL3dybuVmZUpTw9PiysCAACAK3BqxMnb21tz584trVoAy0VUjJCPh4/OZp7VwVMHrS4HAAAALsLpqXo9e/bUt99+WwqlANbz9PB0dNajQQQAAADsnG4OUbduXT3//PNas2aNmjdvroCAgByvDx8+vMSKA6wQExKjv4//rV3Ju9RBHawuBwAAAC7A6eA0depUhYSEaOPGjdq4cWOO12w2G8EJbs/eIIIRJwAAANg5HZzi4uJKow7AZdgbRNBZDwAAAHZOBad169Zp3rx5ysjIUKdOndS5c+fSqguwjH3EKS45js56AAAAkOREc4i5c+eqTZs2evPNN/X++++rW7dumjRpUimWBlgjPDBcvp6+Opt5VvtP7be6HAAAALiAIgenF198UQMHDtSJEyd04sQJjRs3ThMmTCjN2gBLeHp4qk5wHUnc5wQAAIBsRQ5Of//9t5544gl5eWXP7nv88cd14sQJHT16tNSKA6zCfU4AAAA4X5GD06lTpxQSEuJ47uvrK39/f6WkpJRGXYCl7MGJEScAAABITjaH+PHHHxUcHOx4npWVpaVLl+qPP/5wLLvxxhtLrjrAIvYGEYw4AQAAQHIyOA0YMCDXsqFDhzr+22azKTMz88KrAixmH3GKS45TRlaGvDyc7twPAACAcqTIU/WysrIKfRCaUF6EB4bL38tf6Vnp2ndyn9XlAAAAwGJFDk7AxcTD5qHo4GhJTNcDAAAAwQnIV0wwDSIAAACQjeAE5IOW5AAAALAjOAH5sHfWY8QJAAAABCcgH/YRp/iUeGVkZVhcDQAAAKxEcALyUTOwpvy9/JWRlaG9J/daXQ4AAAAsVKQvp6lUqZJsNluRdnjs2LELKghwFR42D9UJrqM/k/7UrhO7VCe4jtUlAQAAwCJFCk6TJk1y/HdSUpImTJigLl26qHXr1pKktWvX6scff9Szzz5bKkUCVokJidGfSX9q54mdui7yOqvLAQAAgEWKFJwGDBjg+O9evXpp/PjxevDBBx3Lhg8frnfeeUdLlizRI488UvJVAhaxN4igsx4AAMDFzel7nH788Ud17do11/IuXbpoyZIlJVIU4CpoSQ4AAACpGMGpcuXKmjt3bq7l3377rSpXrlwiRQGuwj7iFJ8Sr/SsdIurAQAAgFWKNFXvfOPGjdPgwYO1YsUKxz1Ov/zyixYuXKipU6eWeIHl2rE4KfF3q6tAAWoYowoePkrNOqe9mz9WjH81q0sCUJIiWkoVq1tdBQDADTgdnAYOHKiGDRvqrbfe0jfffCNjjBo1aqSff/5ZrVq1Ko0ay69dS6X/Pmp1FSiATVJMjTD97uernUufUUxqmtUlAShJvkHSXd9IEVdaXQkAwMU5HZwkqVWrVpo1a1ZJ13LxCQyTIq6yugoUIsbjuH5XmnZVjZaygqwuB0BJOZUoHY+XPukp9ZtLeAIAFKhYwWnXrl2aPn26du/erUmTJqlatWpauHChIiIidOmll5Z0jeVXwx7ZD7i0un9+LG14VTujW0vtX7e6HAAl5dxpaVZvac/q/w9P32RP3QMAIA9ON4f46aef1LhxY61bt05ff/21Tp06JUnaunWrxowZU+IFAlajsx5QTvkESHd+KUW1lc6dlD65Rdq7zuqqAAAuyungNGrUKE2YMEGLFy+Wj4+PY3mHDh20du3aEi0OcAX2znp7U/YqPZPOekC54hMg9Z39v/D0KeEJAJA3p4PT77//rp49e+ZaXrVqVSUlJZVIUYArCasQpkDvQGWYDMWnxFtdDoCS5hMg9bWPPJ36//D0i9VVAQBcjNPBKSQkRAkJCbmWb968WeHh4SVSFOBKbDab6oTUkcR0PaDc8qmQHZ6ir/n/8NRL2sMsCgDA/zgdnPr27asnn3xSiYmJstlsysrK0s8//6zHHntM/fv3L40aAcvFBGff57TzxE6LKwFQanwqSHfMlqLbnRee1lhdFQDARTgdnF544QXVrl1b4eHhOnXqlBo1aqRrrrlGsbGxeuaZZ0qjRsByNIgALhI+FbLvearTXko/LX16qxT/s9VVAQBcgNPBydvbW7NmzdI///yjL7/8Up9++qn++usvffLJJ/L09CyNGgHL2RtE7EomOAHlnre/dMcXUp0O2eFp1m2EJwCA88Fp/PjxSk1NVZ06dXTrrbeqd+/eqlevntLS0jR+/PjSqBGwnH3EaW/KXp3LPGdxNQBKnbe/dMfnUkzH/w9Pt0rxq62uCgBgIaeD07hx4xzf3XS+1NRUjRs3rkSKAlyNvbNepsmksx5wsfD2l27/TIrpJKWnZo88xa2yuioAgEWcDk7GGNlstlzLt2zZotDQ0BIpCnA1NpuN+5yAi9G/w9NnvaW4lVZXBQCwQJGDU6VKlRQaGiqbzaZLLrlEoaGhjkdwcLCuu+469e7duzRrBSxlv8+JznrARcbbLzs81b32/0eeeku7f7K6KgBAGfMq6oqTJk2SMUaDBg3SuHHjFBwc7HjNx8dHUVFRat26dakUCbgCRpyAi5i3n9RnlvRlP+mfRdJnfaS+X2R33wMAXBSKHJwGDBggSYqOjlZsbKy8vb1LrSjAFRGcgIuct5/U51Np9l3nhafZhCcAuEg4fY9Tu3btHKEpLS1NKSkpOR5AeWWfqrf35F6dzTxrcTUALOHlmx2e6nWRMs5kh6ddy62uCgBQBpwOTqmpqXrwwQdVrVo1BQYGqlKlSjkeQHlV1b+qKvpUVJbJUnxyvNXlALCKl6/U5xPpkq7Z4enz26Vdy6yuCgBQypwOTo8//riWLVumyZMny9fXV1OnTtW4ceNUs2ZNzZw5szRqBFyCzWajQQSAbF6+Uu+Z0iXd/j883SHtXGp1VQCAUuR0cJo/f74mT56sW2+9VV5eXmrbtq2eeeYZvfjii5o1a1Zp1Ai4DO5zAuDg5Sv1/vhf4WmJ1VUBAEqJ08Hp2LFjio6OliQFBQXp2LFjkqSrr75aK1fy3RYo3xhxApCDfeSp/vVS5lnp877SP4QnACiPnA5OderUUXx8vCSpUaNG+vLLLyVlj0SFhISUZG2Ay2HECUAuXj7SbTOkBjdkh6cvCE8AUB45HZzuvvtubdmyRZI0evRox71OjzzyiB5//PESLxBwJfYRp30n9+lMxhmLqwHgMrx8pFunnxee7pD+WWx1VQCAEmQzxpgL2cHevXu1YcMGxcTE6PLLLy+pukpNSkqKgoODlZycrKCgIKvLgZsxxqjt7LZKPpusL2/4Ug0rN7S6JACuJDNd+mqg9Nf3kqdP9pfmXtLZ6qoAAPlwJhs4PeL0b7Vr19Ytt9ziFqEJuFA2m00xwdnT9bjPCUAunt7Z0/Ya9pAyz0mz75R2/Gh1VQCAEuBVlJXeeuutIu9w+PDhxS4GcAcxITHadHgT9zkByJund/a0vTmDpO3zpC/uzP7S3Ppdra4MAHABihSc3njjjSLtzGazEZxQ7tEgAkChPL2lW6dJXw+Wtn0nzb4r+0tz63ezujIAQDEVKTjFxcWVdh2A27A3iNiVTHACUABPb6nXR5Js0rZvpdn9sluXN+hudWUAgGK4oHucjDG6wN4SgNuxjzjtP7lfaRlpFlcDwKXZw9OlPaWsdOnL/tJf/7W6KgBAMRQrOM2cOVONGzeWv7+//P391aRJE33yySclXRvgkir7VVaIb4iMjOKSGY0FUAhPL+mWqdKlt/x/eBpAeAIAN+R0cHr99dd13333qXv37vryyy81e/Zsde3aVcOGDSvyvVCAO7PZbNznBMA5nl7SLR9Kl/X638jT9u+trgoA4IQi3eN0vrfffltTpkxR//79HctuuukmXXrppRo7dqweeeSREi0QcEV1Q+pq46GNtCQHUHSeXlLPDyTZpD/mSF8N+F/rcgCAy3N6xCkhIUGxsbG5lsfGxiohIaFEigJcHSNOAIrF00vq+b7U+DYpKyP7y3K3fWd1VQCAInA6ONWtW1dffvllruWzZ89WvXr1SqQowNXZO+sx4gTAaZ5e0s3vnRee7iY8AYAbcHqq3rhx49SnTx+tXLlSbdq0kc1m0+rVq7V06dI8AxVQHtlHnA6cOqDU9FRV8K5gcUUA3Ip95MnmIW2dnR2ebp0mXXqz1ZUBAPLh9IhTr169tG7dOlWpUkXffvutvvnmG1WpUkW//vqrevbsWRo1Ai4n1C9UoX6hkkRnPQDF4+Ep3TxFanK7ZDKlOYOkP+daXRUAIB9OjzhJUvPmzfXpp5+WdC2AW4kJidGxxGPaeWKnLq1yqdXlAHBHHp7SzZMlm03a8rk0Z7BkjHTZLVZXBgD4l2IFp8zMTM2dO1fbt2+XzWZTw4YNddNNN8nLq1i7A9xSTHCM1ieup0EEgAvj4Snd9K4km7TlM+nrIZJMdutyAIDLcDrp/PHHH7rpppuUmJio+vXrS5J27NihqlWrat68eWrcuHGJFwm4IhpEACgxHp7STe9kjzz9Nis7PBkjNb7V6soAAP/P6XuchgwZoksvvVT79+/Xpk2btGnTJu3bt09NmjTRvffe63QBkydPVnR0tPz8/NS8eXOtWrWqSNv9/PPP8vLyUtOmTZ0+JlASaEkOoER5eEo3vi01vUsyWdI390i/z7G6KgDA/3M6OG3ZskUTJ05UpUqVHMsqVaqkF154Qb/99ptT+5o9e7ZGjBihp59+Wps3b1bbtm3VrVs37d27t8DtkpOT1b9/f3Xq1MnZ8oESYx9xOnj6oFLTUy2uBkC5kFd42vqV1VUBAFSM4FS/fn0dOnQo1/LDhw+rbt26Tu3r9ddf1+DBgzVkyBA1bNhQkyZNUkREhKZMmVLgdkOHDlXfvn3VunVrp44HlKQQvxBV9qssiVEnACXIwyM7PDXrlx2e5t4rbeXrPgDAak4HpxdffFHDhw/XnDlztH//fu3fv19z5szRiBEj9PLLLyslJcXxKMi5c+e0ceNGde7cOcfyzp07a82aNfluN336dO3atUtjxowpUr1nz57NUVNhdQHO4D4nAKXCw0Pq8ZZ0Rf//D09DpS2zra4KAC5qTjeHuOGGGyRJvXv3ls1mkyQZYyRJPXr0cDy32WzKzMzMdz9Hjx5VZmamwsLCciwPCwtTYmJintv8888/GjVqlFatWlXkDn4TJ07UuHHjirQu4Kw6IXW0LnEdI04ASp6Hh3TDm5Js0qaPpW+HSTLS5bdbXRkAXJScDk7Lly8v0QLs4cvOHrr+LTMzU3379tW4ceN0ySWXFHn/o0eP1siRIx3PU1JSFBERUfyCgfM4RpySGXECUAo8PKQbJmV329s4Q5o7LLvbXtM7rK4MAC46TgWnjIwMrVixQoMGDbrg8FGlShV5enrmGl06fPhwrlEoSTp58qQ2bNigzZs368EHH5QkZWVlyRgjLy8vLVq0SB07dsy1na+vr3x9fS+oViA/9s56u0/strgSAOWWh4d0/RuSbNLG6dK390kyUtO+VlcGABcVp+5x8vLy0quvvlrgFLyi8vHxUfPmzbV48eIcyxcvXqzY2Nhc6wcFBen333/Xb7/95ngMGzZM9evX12+//aZWrVpdcE2As+wjTgmnE3Q6/bTF1QAotzw8pOtfl1oMkmSkb++XNs+yuioAuKg4PVWvU6dOWrFihQYOHHjBBx85cqT69eunFi1aqHXr1vrggw+0d+9eDRs2TFL2NLsDBw5o5syZ8vDw0GWXXZZj+2rVqsnPzy/XcqCsBPsGq4p/FR1NO6pdJ3apSdUmVpcEoLzy8JC6vybJJm34SPruAUlGanaX1ZUBwEXB6eDUrVs3jR49Wn/88YeaN2+ugICAHK/feOONRd5Xnz59lJSUpPHjxyshIUGXXXaZFixYoMjISElSQkJCod/pBFgtJiSG4ASgbHh4SNe/ln3P0/qp0ncPZt/zdEU/qysDgHLPZuwt8YrIwyP/2X2FddJzBSkpKQoODlZycrKCgoKsLgflwEu/vqRZ22epf6P+evzKx60uB8DFwBhpwePS+g8l2aQb/791OQDAKc5kA6dHnLKysopdGFAe2RtE0JIcQJmx2aTu/8n+318/kOY9lB2mmg+wujIAKLec/gLc8505c6ak6gDcFl+CC8ASNpvU7RWp5dDs5/OHZ7csBwCUCqeDU2Zmpp5//nmFh4crMDBQu3dnt2F+9tln9dFHH5V4gYCrs484HUo9pJPnTlpcDYCLis0mdXtZapXdVEnzH5Y2TLe2JgAop5wOTi+88IJmzJihV155RT4+Po7ljRs31tSpU0u0OMAdBPkEqZp/NUlM1wNgAZtN6vqS1Oq+7Offj5A2TLO0JAAoj5wOTjNnztQHH3ygO++8U56eno7lTZo00V9//VWixQHugvucAFjKZpO6TpSuuj/7+fePSOuZBQIAJcnp4HTgwAHVrVs31/KsrCylp6eXSFGAu7EHJ+5zAmAZm03q8qJ01QPZz/87MrtlOQCgRDgdnC699FKtWrUq1/KvvvpKzZo1K5GiAHdjbxDBiBMAS9lsUpcXpNYPZj//76PSrx9aWxMAlBNOtyMfM2aM+vXrpwMHDigrK0vffPON/v77b82cOVPff/99adQIuDym6gFwGTab1HlC9v+ueVta8Fj28pb3WFsXALg5p0ecevToodmzZ2vBggWy2Wx67rnntH37ds2fP1/XXXddadQIuDx7cDqcdlgp51IsrgbARc9mk657Xoodnv18wWPSug+srQkA3JzTI06S1KVLF3Xp0qWkawHcVkWfigqrEKZDqYe068QuNavGtFUAFrPZpOvGZ//vz29KPzwuyUithlpdGQC4JadHnOrUqaOkpKRcy0+cOKE6deqUSFGAO+KLcAG4HJtNunac1GZE9vMfnpB+ec/SkgDAXTkdnOLj45WZmZlr+dmzZ3XgwIESKQpwR3VCsv/hgPucALgUm026dqx09SPZzxc+Kf0yxdKSAMAdFXmq3rx58xz//eOPPyo4ONjxPDMzU0uXLlVUVFSJFge4E0acALgsm03qNEayeUirXpMWjpJMltT6AasrAwC3UeTgdPPNN0uSbDabBgwYkOM1b29vRUVF6bXXXivR4gB3Ym8QsfvEbosrAYA82GxSx2cl2aRVr0o/PiUZI8U+aHVlAOAWihycsrKyJEnR0dFav369qlSpUmpFAe4oJjg7OB1JO6Lks8kK9g0uZAsAKGM2m9Txmez/XfkfadHTkowU+5DVlQGAy3P6Hqe4uDhCE5CHQJ9AVQ+oLon7nAC4MJtN6vC0dM0T2c8XPSP9/Ja1NQGAGyhWO/KlS5dq6dKlOnz4sGMkym7atGklUhjgjmJCYpR4OlE7T+zUFWFXWF0OAOTNZpM6PJX9vz+9LC1+VpKR2jxsdWUA4LKcHnEaN26cOnfurKVLl+ro0aM6fvx4jgdwMasbnN0gghEnAC7PHp7ajcp+vvg5afUkS0sCAFfm9IjTe++9pxkzZqhfv36lUQ/g1uwNIghOANxGh9HZIWrFRGnJGEnmf63LAQAOTgenc+fOKTY2tjRqAdweLckBuKX2oyTZpBUvSkvGShnnpAbXW10VgPIuNFryCbC6iiJzOjgNGTJEn332mZ599tnSqAdwa/YRp6QzSTpx5oRC/EKsLQgAiqr9k9kjT8tfyA5QK160uiIA5d3gJVLElVZXUWROB6czZ87ogw8+0JIlS9SkSRN5e3vneP31118vseIAd1PBu4JqBtTUwdMHtfPETrWo3sLqkgCg6No9kf2vv79MkTLPWV0NgPLOs1h96izjdLVbt25V06ZNJUl//PFHjtdsNluJFAW4s5iQGB08fVC7TuwiOAFwP60fyH4AAHJwOjgtX768NOoAyo26IXW16sAq7nMCAAAoR5xuRw6gYI7Oesl01gMAACgvijzidMsttxRpvW+++abYxQDlgb2zHi3JAQAAyo8iB6fg4ODSrAMoN6KDoyVJx84c07EzxxTqF2pxRQAAALhQRQ5O06dPL806gHKjgncFhQeG68CpA9p1YpdCqxOcAAAA3B33OAGlgC/CBQAAKF8ITkApqBNSRxL3OQEAAJQXBCegFDDiBAAAUL4QnIBSYG9JvvvEbosrAQAAQEkgOAGloE5wHdlk0/Gzx5WUlmR1OQAAALhABCegFPh7+Ss8MFwS9zkBAACUBwQnoJRwnxMAAED5QXACSon9PidGnAAAANwfwQkoJfbgxIgTAACA+yM4AaXEPlVvV/IuGWMsrgYAAAAXguAElJLo4Gh52DyUfDZZSWforAcAAODOCE5AKfHz8lOtwFqSmK4HAADg7ghOQCmiQQQAAED5QHACShEtyQEAAMoHghNQihhxAgAAKB8ITkApOn/Eic56AAAA7ovgBJSiqOAoedg8dPLcSR1JO2J1OQAAACgmghNQinw9fVW7Ym1J3OcEAADgzghOQCmrE1xHEvc5AQAAuDOCE1DKaBABAADg/ghOQCmzN4ggOAEAALgvghNQys4fcaKzHgAAgHsiOAGlLDo4Wp42T51MP6nDqYetLgcAAADFQHACSpmPp48iKkZIYroeAACAuyI4AWXg/C/CBQAAgPshOAFlwHGfUzIjTgAAAO6I4ASUAUacAAAA3BvBCSgD9hGn3Sd201kPAADADRGcgDIQFRQlL5uXTqWf0qHUQ1aXAwAAACcRnIAy4O3prdpBtSUxXQ8AAMAdEZyAMnL+F+ECAADAvRCcgDJCgwgAAAD3RXACyggjTgAAAO6L4ASUEfuI064Tu+isBwAA4GYITkAZqR1UW14eXkrNSFXC6QSrywEAAIATCE5AGfH28FZUUJQk7nMCAABwNwQnoAzVCa4jifucAAAA3A3BCShDdNYDAABwTwQnoAzZO+vtPrHb4koAAADgDIITUIYcnfWSdynLZFlcDQAAAIqK4ASUoYigCHl5eCktI43OegAAAG6E4ASUofM769EgAgAAwH0QnIAyRoMIAAAA90NwAsqYvUEEI04AAADug+AElDFGnAAAANyP5cFp8uTJio6Olp+fn5o3b65Vq1blu+4333yj6667TlWrVlVQUJBat26tH3/8sQyrBS6cfcQpLjmOznoAAABuwtLgNHv2bI0YMUJPP/20Nm/erLZt26pbt27au3dvnuuvXLlS1113nRYsWKCNGzeqQ4cO6tGjhzZv3lzGlQPFF1ExQt4e3krLSNOBUwesLgcAAABFYDPGGKsO3qpVK11xxRWaMmWKY1nDhg118803a+LEiUXax6WXXqo+ffroueeeK9L6KSkpCg4OVnJysoKCgopVN3Ches3rpR3Hd+jtjm+rfUR7q8sBAAC4KDmTDSwbcTp37pw2btyozp0751jeuXNnrVmzpkj7yMrK0smTJxUaGprvOmfPnlVKSkqOB2A1+3Q97nMCAABwD5YFp6NHjyozM1NhYWE5loeFhSkxMbFI+3jttdd0+vRp9e7dO991Jk6cqODgYMcjIiLiguoGSoK9QQSd9QAAANyD5c0hbDZbjufGmFzL8vL5559r7Nixmj17tqpVq5bveqNHj1ZycrLjsW/fvguuGbhQtCQHAABwL15WHbhKlSry9PTMNbp0+PDhXKNQ/zZ79mwNHjxYX331la699toC1/X19ZWvr+8F1wuUJPuI0+7k3crMypSnh6fFFQEAAKAglo04+fj4qHnz5lq8eHGO5YsXL1ZsbGy+233++ecaOHCgPvvsM11//fWlXSZQKmoF1pKvp6/OZp6lsx4AAIAbsHSq3siRIzV16lRNmzZN27dv1yOPPKK9e/dq2LBhkrKn2fXv39+x/ueff67+/fvrtdde01VXXaXExEQlJiYqOTnZqlMAisXTw1PRwdGSaBABAADgDiwNTn369NGkSZM0fvx4NW3aVCtXrtSCBQsUGRkpSUpISMjxnU7vv/++MjIy9MADD6hGjRqOx8MPP2zVKQDFVie4jiTucwIAAHAHlt3jZHf//ffr/vvvz/O1GTNm5Hi+YsWK0i8IKCP2+5wYcQIAAHB9lnfVAy5W9s56u5N3W1wJAAAACkNwAixiH3GKS45TZlamxdUAAACgIAQnwCLhgeGOznr7T+23uhwAAAAUgOAEWMTTw9PRIIL7nAAAAFwbwQmwkP0+JzrrAQAAuDaCE2Ahe3BixAkAAMC1EZwAC9kbRDDiBAAA4NoIToCF7CNOcclxysjKsLgaAAAA5IfgBFgoPDBc/l7+Ss9K176T+6wuBwAAAPkgOAEW8rB5KDo4WhLT9QAAAFwZwQmwmP0+JxpEAAAAuC6CE2AxWpIDAAC4PoITYDFGnAAAAFwfwQmwmH3EKT4lXulZ6RZXAwAAgLwQnACL1QioIX8vf2VkZWhfCp31AAAAXBHBCbCYh81DMcHZo05M1wMAAHBNBCfABdQJqSOJBhEAAACuiuAEuAAaRAAAALg2ghPgAuwNInYn77a4EgAAAOSF4AS4APuIE531AAAAXBPBCXABNQJqqIJXBWVkZWhvyl6rywEAAMC/EJwAF2Cz2RzT9bjPCQAAwPUQnAAXYQ9OdNYDAABwPQQnwEXQWQ8AAMB1EZwAF8GIEwAAgOsiOAEuwj7itDdlr9Iz6awHAADgSghOgIsIqxCmQO9AZZgMxafEW10OAAAAzkNwAlyEzWZTnZA6kpiuBwAA4GoIToALoUEEAACAayI4AS4kJpgGEQAAAK6I4AS4EEacAAAAXBPBCXAh9pbk+07u07nMcxZXAwAAADuCE+BCqlWoporeFZVpMhWXHGd1OQAAAPh/BCfAhdhsNr4IFwAAwAURnAAXYw9O3OcEAADgOghOgIthxAkAAMD1EJwAF2MPTruTd1tcCQAAAOwIToCLsbck33tyr85mnrW4GgAAAEgEJ8DlVPWvqoo+FZVlshSfHG91OQAAABDBCXA5NpuNL8IFAABwMQQnwAXRIAIAAMC1EJwAF8SIEwAAgGshOAEuiBEnAAAA10JwAlyQfcRp38l9OpNxxuJqAAAAQHACXFBlv8oK9g2WkVFccpzV5QAAAFz0CE6AC7LZbIoJzp6ux31OAAAA1iM4AS7KPl2P+5wAAACsR3ACXBQNIgAAAFwHwQlwUbQkBwAAcB0EJ8BF2UecDpw6oLSMNIurAQAAuLgRnAAXVdm/sir5VpKR0e7k3VaXAwAAcFEjOAEurE5IHUnc5wQAAGA1ghPgwrjPCQAAwDUQnAAXZr/PafcJpuoBAABYieAEuDBGnAAAAFwDwQlwYed31ktNT7W4GgAAgIsXwQlwYaF+oQr1C5UkxSXHWVwNAADAxYvgBLg4+6gT0/UAAACsQ3ACXFxMcHZwoiU5AACAdQhOgIujQQQAAID1CE6Ai7NP1WPECQAAwDoEJ8DF2UecDp4+SGc9AAAAixCcABcX4heiyn6VJTHqBAAAYBWCE+AGuM8JAADAWgQnwA1wnxMAAIC1CE6AG3B8l1MyI04AAABWIDgBbsA+VY8RJwAAAGsQnAA3YB9xSjydqFPnTllcDQAAwMWH4AS4gWDfYFX1rypJ2pXMqBMAAEBZIzgBbqJOSB1JTNcDAACwAsEJcBO0JAcAALAOwQlwE/b7nHaf2G1xJQAAABcfghPgJhhxAgAAsI7lwWny5MmKjo6Wn5+fmjdvrlWrVhW4/k8//aTmzZvLz89PderU0XvvvVdGlQLWso84HUo9pJPnTlpcDQAAwMXF0uA0e/ZsjRgxQk8//bQ2b96stm3bqlu3btq7d2+e68fFxal79+5q27atNm/erKeeekrDhw/X119/XcaVA2UvyCdI1fyrSaJBBAAAQFmzGWOMVQdv1aqVrrjiCk2ZMsWxrGHDhrr55ps1ceLEXOs/+eSTmjdvnrZv3+5YNmzYMG3ZskVr164t0jFTUlIUHBys5ORkBQUFXfhJXABjjNLSMy2tAe7loeX36dfEX9T7ktt1RbUWVpcDAABQbG3CWynEL8TSGpzJBl5lVFMu586d08aNGzVq1Kgcyzt37qw1a9bkuc3atWvVuXPnHMu6dOmijz76SOnp6fL29s61zdmzZ3X27FnH85SUlBKovmSkpWeq0XM/Wl0G3IhvNS/5VJa+3PGFvtzxhdXlAAAAFNvU6z5Wq5pXWF1GkVkWnI4eParMzEyFhYXlWB4WFqbExMQ8t0lMTMxz/YyMDB09elQ1atTItc3EiRM1bty4kiscsNC5E63k4XtE8jhjdSkAAAAXJMA7wOoSnGJZcLKz2Ww5nhtjci0rbP28ltuNHj1aI0eOdDxPSUlRREREccstUf7ento2vovVZcDt3GV1AQAAABfM39vT6hKcYllwqlKlijw9PXONLh0+fDjXqJJd9erV81zfy8tLlStXznMbX19f+fr6lkzRJcxms6mCj+XZFQAAAEAhLOuq5+Pjo+bNm2vx4sU5li9evFixsbF5btO6detc6y9atEgtWrTI8/4mAAAAACgJlrYjHzlypKZOnapp06Zp+/bteuSRR7R3714NGzZMUvY0u/79+zvWHzZsmPbs2aORI0dq+/btmjZtmj766CM99thjVp0CAAAAgIuApfPE+vTpo6SkJI0fP14JCQm67LLLtGDBAkVGRkqSEhIScnynU3R0tBYsWKBHHnlE7777rmrWrKm33npLvXr1suoUAAAAAFwELP0eJyu40vc4AQAAALCOM9nA0ql6AAAAAOAOCE4AAAAAUAiCEwAAAAAUguAEAAAAAIUgOAEAAABAIQhOAAAAAFAIghMAAAAAFILgBAAAAACFIDgBAAAAQCEITgAAAABQCIITAAAAABSC4AQAAAAAhSA4AQAAAEAhvKwuoKwZYyRJKSkpFlcCAAAAwEr2TGDPCAW56ILTyZMnJUkREREWVwIAAADAFZw8eVLBwcEFrmMzRYlX5UhWVpYOHjyoihUrymazWV2OUlJSFBERoX379ikoKMjqcuAGuGbgDK4XOItrBs7imoGzXOmaMcbo5MmTqlmzpjw8Cr6L6aIbcfLw8FCtWrWsLiOXoKAgyy8cuBeuGTiD6wXO4pqBs7hm4CxXuWYKG2myozkEAAAAABSC4AQAAAAAhSA4WczX11djxoyRr6+v1aXATXDNwBlcL3AW1wycxTUDZ7nrNXPRNYcAAAAAAGcx4gQAAAAAhSA4AQAAAEAhCE4AAAAAUAiCEwAAAAAUguBUyiZPnqzo6Gj5+fmpefPmWrVqVYHr//TTT2revLn8/PxUp04dvffee2VUKVyFM9fMN998o+uuu05Vq1ZVUFCQWrdurR9//LEMq4UrcPbvjN3PP/8sLy8vNW3atHQLhMtx9po5e/asnn76aUVGRsrX11cxMTGaNm1aGVULV+DsNTNr1ixdfvnlqlChgmrUqKG7775bSUlJZVQtrLZy5Ur16NFDNWvWlM1m07ffflvoNu7wGZjgVIpmz56tESNG6Omnn9bmzZvVtm1bdevWTXv37s1z/bi4OHXv3l1t27bV5s2b9dRTT2n48OH6+uuvy7hyWMXZa2blypW67rrrtGDBAm3cuFEdOnRQjx49tHnz5jKuHFZx9pqxS05OVv/+/dWpU6cyqhSuojjXTO/evbV06VJ99NFH+vvvv/X555+rQYMGZVg1rOTsNbN69Wr1799fgwcP1p9//qmvvvpK69ev15AhQ8q4cljl9OnTuvzyy/XOO+8UaX23+QxsUGpatmxphg0blmNZgwYNzKhRo/Jc/4knnjANGjTIsWzo0KHmqquuKrUa4VqcvWby0qhRIzNu3LiSLg0uqrjXTJ8+fcwzzzxjxowZYy6//PJSrBCuxtlr5ocffjDBwcEmKSmpLMqDC3L2mvnPf/5j6tSpk2PZW2+9ZWrVqlVqNcJ1STJz584tcB13+QzMiFMpOXfunDZu3KjOnTvnWN65c2etWbMmz23Wrl2ba/0uXbpow4YNSk9PL7Va4RqKc838W1ZWlk6ePKnQ0NDSKBEuprjXzPTp07Vr1y6NGTOmtEuEiynONTNv3jy1aNFCr7zyisLDw3XJJZfoscceU1paWlmUDIsV55qJjY3V/v37tWDBAhljdOjQIc2ZM0fXX399WZQMN+Qun4G9rC6gvDp69KgyMzMVFhaWY3lYWJgSExPz3CYxMTHP9TMyMnT06FHVqFGj1OqF9Ypzzfzba6+9ptOnT6t3796lUSJcTHGumX/++UejRo3SqlWr5OXF/wVcbIpzzezevVurV6+Wn5+f5s6dq6NHj+r+++/XsWPHuM/pIlCcayY2NlazZs1Snz59dObMGWVkZOjGG2/U22+/XRYlww25y2dgRpxKmc1my/HcGJNrWWHr57Uc5Zez14zd559/rrFjx2r27NmqVq1aaZUHF1TUayYzM1N9+/bVuHHjdMkll5RVeXBBzvydycrKks1m06xZs9SyZUt1795dr7/+umbMmMGo00XEmWtm27ZtGj58uJ577jlt3LhRCxcuVFxcnIYNG1YWpcJNucNnYP65sZRUqVJFnp6euf415vDhw7kStV316tXzXN/Ly0uVK1cutVrhGopzzdjNnj1bgwcP1ldffaVrr722NMuEC3H2mjl58qQ2bNigzZs368EHH5SU/aHYGCMvLy8tWrRIHTt2LJPaYY3i/J2pUaOGwsPDFRwc7FjWsGFDGWO0f/9+1atXr1RrhrWKc81MnDhRbdq00eOPPy5JatKkiQICAtS2bVtNmDDBZUYP4Drc5TMwI06lxMfHR82bN9fixYtzLF+8eLFiY2Pz3KZ169a51l+0aJFatGghb2/vUqsVrqE414yUPdI0cOBAffbZZ8wfv8g4e80EBQXp999/12+//eZ4DBs2TPXr19dvv/2mVq1alVXpsEhx/s60adNGBw8e1KlTpxzLduzYIQ8PD9WqVatU64X1inPNpKamysMj50dMT09PSf8bRQDO5zafgS1qSnFR+OKLL4y3t7f56KOPzLZt28yIESNMQECAiY+PN8YYM2rUKNOvXz/H+rt37zYVKlQwjzzyiNm2bZv56KOPjLe3t5kzZ45Vp4Ay5uw189lnnxkvLy/z7rvvmoSEBMfjxIkTVp0Cypiz18y/0VXv4uPsNXPy5ElTq1Ytc+utt5o///zT/PTTT6ZevXpmyJAhVp0Cypiz18z06dONl5eXmTx5stm1a5dZvXq1adGihWnZsqVVp4AydvLkSbN582azefNmI8m8/vrrZvPmzWbPnj3GGPf9DExwKmXvvvuuiYyMND4+PuaKK64wP/30k+O1AQMGmHbt2uVYf8WKFaZZs2bGx8fHREVFmSlTppRxxbCaM9dMu3btjKRcjwEDBpR94bCMs39nzkdwujg5e81s377dXHvttcbf39/UqlXLjBw50qSmppZx1bCSs9fMW2+9ZRo1amT8/f1NjRo1zJ133mn2799fxlXDKsuXLy/w84m7fga2GcOYKQAAAAAUhHucAAAAAKAQBCcAAAAAKATBCQAAAAAKQXACAAAAgEIQnAAAAACgEAQnAAAAACgEwQkAAAAACkFwAgAAAIBCEJwAAHBCVFSUJk2a5Hhus9n07bffWlYPAKBsEJwAAG5j4MCBstlsstls8vLyUu3atXXffffp+PHjVpcGACjnCE4AALfStWtXJSQkKD4+XlOnTtX8+fN1//33W10WAKCcIzgBANyKr6+vqlevrlq1aqlz587q06ePFi1a5Hh9+vTpatiwofz8/NSgQQNNnjw5x/b79+/X7bffrtDQUAUEBKhFixZat26dJGnXrl266aabFBYWpsDAQF155ZVasmRJmZ4fAMA1eVldAAAAxbV7924tXLhQ3t7ekqQPP/xQY8aM0TvvvKNmzZpp8+bNuueeexQQEKABAwbo1KlTateuncLDwzVv3jxVr15dmzZtUlZWliTp1KlT6t69uyZMmCA/Pz99/PHH6tGjh/7++2/Vrl3bylMFAFiM4AQAcCvff/+9AgMDlZmZqTNnzkiSXn/9dUnS888/r9dee0233HKLJCk6Olrbtm3T+++/rwEDBuizzz7TkSNHtH79eoWGhkqS6tat69j35Zdfrssvv9zxfMKECZo7d67mzZunBx98sKxOEQDggghOAAC30qFDB02ZMkWpqamaOnWqduzYoYceekhHjhzRvn37NHjwYN1zzz2O9TMyMhQcHCxJ+u2339SsWTNHaPq306dPa9y4cfr+++918OBBZWRkKC0tTXv37i2TcwMAuC6CEwDArQQEBDhGid566y116NBB48aNc4wIffjhh2rVqlWObTw9PSVJ/v7+Be778ccf148//qhXX31VdevWlb+/v2699VadO3euFM4EAOBOCE4AALc2ZswYdevWTffdd5/Cw8O1e/du3XnnnXmu26RJE02dOlXHjh3Lc9Rp1apVGjhwoHr27Ckp+56n+Pj40iwfAOAm6KoHAHBr7du316WXXqoXX3xRY8eO1cSJE/Xmm29qx44d+v333zV9+nTHPVB33HGHqlevrptvvlk///yzdu/era+//lpr166VlH2/0zfffKPffvtNW7ZsUd++fR2NIwAAFzeCEwDA7Y0cOVIffvihunTpoqlTp2rGjBlq3Lix2rVrpxkzZig6OlqS5OPjo0WLFqlatWrq3r27GjdurJdeeskxle+NN95QpUqVFBsbqx49eqhLly664oorrDw1AICLsBljjNVFAAAAAIArY8QJAAAAAApBcAIAAACAQhCcAAAAAKAQBCcAAAAAKATBCQAAAAAKQXACAAAAgEIQnAAAAACgEAQnAAAAACgEwQkAAAAACkFwAgAAAIBCEJwAAAAAoBD/B+nQTqDnfR/MAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import csv\n",
    "from whoosh.fields import Schema, TEXT, ID\n",
    "from whoosh import index\n",
    "from whoosh.analysis import RegexTokenizer, StopFilter, Filter\n",
    "from whoosh.qparser import QueryParser\n",
    "from whoosh import scoring\n",
    "from nltk.stem.isri import ISRIStemmer\n",
    "import os.path\n",
    "from nltk.corpus import stopwords\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import nltk\n",
    "\n",
    "nltk.download('stopwords')\n",
    "\n",
    "# Charger la liste des mots vides en arabe de NLTK\n",
    "arabic_stopwords = set(stopwords.words(\"arabic\"))\n",
    "\n",
    "class ArabicStemFilter(Filter):\n",
    "    def __init__(self):\n",
    "        self.stemmer = ISRIStemmer()\n",
    "\n",
    "    def __call__(self, tokens):\n",
    "        for t in tokens:\n",
    "            t.text = self.stemmer.stem(t.text)\n",
    "            yield t\n",
    "\n",
    "# Créer un analyseur personnalisé avec la liste des mots vides de NLTK\n",
    "analyzer = RegexTokenizer() | StopFilter(stoplist=arabic_stopwords) | ArabicStemFilter()\n",
    "\n",
    "# Créer le schéma en utilisant l'analyseur personnalisé\n",
    "schema = Schema(\n",
    "    title=TEXT(stored=True),\n",
    "    path=ID(stored=True),\n",
    "    content=TEXT(stored=True, analyzer=analyzer, vector=True)  # Ajouter \"vector=True\"\n",
    ")\n",
    "\n",
    "# Créer le répertoire d'index s'il n'existe pas\n",
    "if not os.path.exists(\"Index-2\"):\n",
    "    os.mkdir(\"Index-2\")\n",
    "\n",
    "# Lister les fichiers dans le répertoire \"Corpus\"\n",
    "corpus_dir = \"E:/M2/RI/Corpus/Corpus\"\n",
    "file_names = os.listdir(corpus_dir)\n",
    "\n",
    "# Créer l'index\n",
    "ix = index.create_in(\"Index-2\", schema)\n",
    "\n",
    "# Indexer les documents avec prétraitement\n",
    "with ix.writer() as writer:\n",
    "    for file_name in file_names:\n",
    "        with open(os.path.join(corpus_dir, file_name), 'r', encoding='windows-1256') as file:\n",
    "            content = file.read()\n",
    "            writer.add_document(title=file_name, content=content, path='')\n",
    "\n",
    "def calculate_interpolated_precision(results, recall_levels):\n",
    "    recall_values = [result[3] for result in results]\n",
    "    precision_values = [result[4] for result in results]\n",
    "\n",
    "    interpolated_precisions = []\n",
    "    for recall_level in recall_levels:\n",
    "        max_precision = 0\n",
    "        for recall, precision in zip(recall_values, precision_values):\n",
    "            if recall >= recall_level:\n",
    "                max_precision = max(max_precision, precision)\n",
    "        interpolated_precisions.append(max_precision)\n",
    "\n",
    "    return interpolated_precisions\n",
    "\n",
    "def search_vectorial(query_str, relevant_documents, limit=20):\n",
    "    with ix.searcher(weighting=scoring.TF_IDF()) as searcher:  # Use scoring module for TF-IDF\n",
    "        # Définir un analyseur et un parseur liés à l'index\n",
    "        query_parser = QueryParser(\"content\", ix.schema)\n",
    "        query = query_parser.parse(query_str)\n",
    "\n",
    "        # Recherche Vectorielle (utilisation par défaut de Whoosh)\n",
    "        results_vectorial = searcher.search(query, terms=True, limit=limit)\n",
    "\n",
    "        # Marquer les résultats en fonction de la liste des documents pertinents\n",
    "        marked_results = []\n",
    "        true_positives = 0  # Count of relevant documents retrieved\n",
    "        false_positives = 0  # Count of irrelevant documents retrieved\n",
    "        total_relevant_documents = len(relevant_documents)\n",
    "\n",
    "        if total_relevant_documents == 0:\n",
    "            recall = precision = 0\n",
    "        else:\n",
    "            for rank, r in enumerate(results_vectorial, start=1):\n",
    "                is_relevant = r['title'] in relevant_documents\n",
    "                mark = 'YES' if is_relevant else 'NO'\n",
    "\n",
    "                # Calculate recall and precision for each document\n",
    "                if is_relevant:\n",
    "                    true_positives += 1\n",
    "                else:\n",
    "                    false_positives += 1\n",
    "\n",
    "                recall = round(true_positives / total_relevant_documents, 3)\n",
    "                precision = round(true_positives / (true_positives + false_positives), 3)\n",
    "\n",
    "                # Append information to the marked result\n",
    "                marked_results.append((rank, r['title'], mark, recall, precision))\n",
    "\n",
    "        return marked_results\n",
    "\n",
    "# Define the queries and relevant documents for each query\n",
    "queries = [\n",
    "    (\"الكلاسيكية OR الحلول\", []),\n",
    "    (\"أوبك\", [\"D493 A.txt\", \"D330 A.txt\"]),\n",
    "    (\"التفاضلية OR المعادلات\", [\"D508 A.txt\", \"D081 A.txt\", \"D105 A.txt\", \"D162 A.txt\", \"D329 A.txt\", \"D552 A.txt\", \"D464 A.txt\"])\n",
    "]\n",
    "\n",
    "# Interpolated Precision Recall Curve\n",
    "recall_levels = np.arange(0, 1.1, 0.1)\n",
    "interpolated_precision_all_queries = []\n",
    "\n",
    "# Rechercher et marquer les documents pertinents pour chaque requête\n",
    "for query, relevant_documents in queries:\n",
    "    marked_results_query = search_vectorial(query, relevant_documents, limit=20)\n",
    "    interpolated_precision_query = calculate_interpolated_precision(marked_results_query, recall_levels)\n",
    "    interpolated_precision_all_queries.append((query, interpolated_precision_query))\n",
    "\n",
    "# Plot Interpolated Precision Recall Curve for all queries\n",
    "plt.figure(figsize=(10, 6))\n",
    "for query, interpolated_precision_query in interpolated_precision_all_queries:\n",
    "    plt.plot(recall_levels, interpolated_precision_query, label=query)\n",
    "\n",
    "plt.title('Interpolated Precision-Recall Curve for Queries')\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Interpolated Precision')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "a6d05d07",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Results for Original Queries:\n",
      "Average Precision: 0.061\n",
      "Mean Average Precision (MAP): 0.061\n",
      "Average F-measure: 0.000\n"
     ]
    }
   ],
   "source": [
    "# Define the original queries and relevant documents for each query\n",
    "original_queries = [\n",
    "    (\"الكلاسيكية OR الحلول OR حل ORجذر  \", []),\n",
    "    (\"أوبك OR منظمة OR الدول OR المصدرة OR للنفط \", [\"D493 A.txt\", \"D330 A.txt\"]),\n",
    "    (\"التفاضلية OR المعادلات\", [\"D508 A.txt\", \"D081 A.txt\", \"D105 A.txt\", \"D162 A.txt\", \"D329 A.txt\", \"D552 A.txt\", \"D464 A.txt\"])\n",
    "]\n",
    "\n",
    "# Interpolated Precision Recall Curve for original queries\n",
    "recall_levels_original = np.arange(0, 1.1, 0.1)\n",
    "interpolated_precision_all_queries_original = []\n",
    "\n",
    "# Rechercher et marquer les documents pertinents pour chaque requête originale\n",
    "for query, relevant_documents in original_queries:\n",
    "    marked_results_query_original = search_vectorial(query, relevant_documents, limit=20)\n",
    "    interpolated_precision_query_original = calculate_interpolated_precision(marked_results_query_original, recall_levels_original)\n",
    "    interpolated_precision_all_queries_original.append((query, interpolated_precision_query_original))\n",
    "\n",
    "# Plot Interpolated Precision Recall Curve for original queries\n",
    "plt.figure(figsize=(10, 6))\n",
    "for query, interpolated_precision_query_original in interpolated_precision_all_queries_original:\n",
    "    plt.plot(recall_levels_original, interpolated_precision_query_original, label=query)\n",
    "\n",
    "plt.title('Interpolated Precision-Recall Curve for Original Queries')\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Interpolated Precision')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "# Calculate average precision, MAP, and F-measure for original queries\n",
    "average_precisions_all_queries_original = []\n",
    "f_measures_all_queries_original = []\n",
    "\n",
    "for _, interpolated_precision_query_original in interpolated_precision_all_queries_original:\n",
    "    average_precision_query_original = calculate_average_precision(interpolated_precision_query_original)\n",
    "    f_measure_query_original = calculate_f_measure(interpolated_precision_query_original[-1], recall_levels_original[-1])\n",
    "    \n",
    "    average_precisions_all_queries_original.append(average_precision_query_original)\n",
    "    f_measures_all_queries_original.append(f_measure_query_original)\n",
    "\n",
    "# Calculate overall average precision and MAP for original queries\n",
    "average_precision_original = np.mean(average_precisions_all_queries_original)\n",
    "map_value_original = calculate_map(interpolated_precision_all_queries_original)\n",
    "\n",
    "# Print the results for original queries\n",
    "print(\"Results for Original Queries:\")\n",
    "print(f\"Average Precision: {average_precision_original:.3f}\")\n",
    "print(f\"Mean Average Precision (MAP): {map_value_original:.3f}\")\n",
    "print(f\"Average F-measure: {np.mean(f_measures_all_queries_original):.3f}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3ff98648",
   "metadata": {},
   "source": [
    "---------------------tf-idf-transformated querries-----------------------------"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "40f10665",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to\n",
      "[nltk_data]     C:\\Users\\meriem\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import csv\n",
    "from whoosh.fields import Schema, TEXT, ID\n",
    "from whoosh import index\n",
    "from whoosh.analysis import RegexTokenizer, StopFilter, Filter\n",
    "from whoosh.qparser import QueryParser\n",
    "from whoosh import scoring\n",
    "from nltk.stem.isri import ISRIStemmer\n",
    "import os.path\n",
    "from nltk.corpus import stopwords\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import nltk\n",
    "\n",
    "nltk.download('stopwords')\n",
    "\n",
    "# Charger la liste des mots vides en arabe de NLTK\n",
    "arabic_stopwords = set(stopwords.words(\"arabic\"))\n",
    "\n",
    "class ArabicStemFilter(Filter):\n",
    "    def __init__(self):\n",
    "        self.stemmer = ISRIStemmer()\n",
    "\n",
    "    def __call__(self, tokens):\n",
    "        for t in tokens:\n",
    "            t.text = self.stemmer.stem(t.text)\n",
    "            yield t\n",
    "\n",
    "# Créer un analyseur personnalisé avec la liste des mots vides de NLTK\n",
    "analyzer = RegexTokenizer() | StopFilter(stoplist=arabic_stopwords) | ArabicStemFilter()\n",
    "\n",
    "# Créer le schéma en utilisant l'analyseur personnalisé\n",
    "schema = Schema(\n",
    "    title=TEXT(stored=True),\n",
    "    path=ID(stored=True),\n",
    "    content=TEXT(stored=True, analyzer=analyzer, vector=True)  # Ajouter \"vector=True\"\n",
    ")\n",
    "\n",
    "# Créer le répertoire d'index s'il n'existe pas\n",
    "if not os.path.exists(\"Index-2\"):\n",
    "    os.mkdir(\"Index-2\")\n",
    "\n",
    "# Lister les fichiers dans le répertoire \"Corpus\"\n",
    "corpus_dir = \"E:/M2/RI/Corpus/Corpus\"\n",
    "file_names = os.listdir(corpus_dir)\n",
    "\n",
    "# Créer l'index\n",
    "ix = index.create_in(\"Index-2\", schema)\n",
    "\n",
    "# Indexer les documents avec prétraitement\n",
    "with ix.writer() as writer:\n",
    "    for file_name in file_names:\n",
    "        with open(os.path.join(corpus_dir, file_name), 'r', encoding='windows-1256') as file:\n",
    "            content = file.read()\n",
    "            writer.add_document(title=file_name, content=content, path='')\n",
    "\n",
    "def calculate_interpolated_precision(results, recall_levels):\n",
    "    recall_values = [result[3] for result in results]\n",
    "    precision_values = [result[4] for result in results]\n",
    "\n",
    "    interpolated_precisions = []\n",
    "    for recall_level in recall_levels:\n",
    "        max_precision = 0\n",
    "        for recall, precision in zip(recall_values, precision_values):\n",
    "            if recall >= recall_level:\n",
    "                max_precision = max(max_precision, precision)\n",
    "        interpolated_precisions.append(max_precision)\n",
    "\n",
    "    return interpolated_precisions\n",
    "\n",
    "def search_vectorial(query_str, relevant_documents, limit=20):\n",
    "    with ix.searcher(weighting=scoring.TF_IDF()) as searcher:  # Use scoring module for TF-IDF\n",
    "        # Définir un analyseur et un parseur liés à l'index\n",
    "        query_parser = QueryParser(\"content\", ix.schema)\n",
    "        query = query_parser.parse(query_str)\n",
    "\n",
    "        # Recherche Vectorielle (utilisation par défaut de Whoosh)\n",
    "        results_vectorial = searcher.search(query, terms=True, limit=limit)\n",
    "\n",
    "        # Marquer les résultats en fonction de la liste des documents pertinents\n",
    "        marked_results = []\n",
    "        true_positives = 0  # Count of relevant documents retrieved\n",
    "        false_positives = 0  # Count of irrelevant documents retrieved\n",
    "        total_relevant_documents = len(relevant_documents)\n",
    "\n",
    "        if total_relevant_documents == 0:\n",
    "            recall = precision = 0\n",
    "        else:\n",
    "            for rank, r in enumerate(results_vectorial, start=1):\n",
    "                is_relevant = r['title'] in relevant_documents\n",
    "                mark = 'YES' if is_relevant else 'NO'\n",
    "\n",
    "                # Calculate recall and precision for each document\n",
    "                if is_relevant:\n",
    "                    true_positives += 1\n",
    "                else:\n",
    "                    false_positives += 1\n",
    "\n",
    "                recall = round(true_positives / total_relevant_documents, 3)\n",
    "                precision = round(true_positives / (true_positives + false_positives), 3)\n",
    "\n",
    "                # Append information to the marked result\n",
    "                marked_results.append((rank, r['title'], mark, recall, precision))\n",
    "\n",
    "        return marked_results\n",
    "\n",
    "# Define the queries and relevant documents for each query\n",
    "queries = [\n",
    "    (\"الكلاسيكية OR الحلول OR حل OR فك ORجذر OR بقاء ORلبث OR مكوث  \", []),\n",
    "    (\"أوبك OR منظمة OR الدول OR المصدرة OR للنفط \", [\"D493 A.txt\", \"D330 A.txt\"]),\n",
    "    (\"التفاضلية OR المعادلات\", [\"D508 A.txt\", \"D081 A.txt\", \"D105 A.txt\", \"D162 A.txt\", \"D329 A.txt\", \"D552 A.txt\", \"D464 A.txt\"])\n",
    "]\n",
    "\n",
    "# Interpolated Precision Recall Curve\n",
    "recall_levels = np.arange(0, 1.1, 0.1)\n",
    "interpolated_precision_all_queries = []\n",
    "\n",
    "# Rechercher et marquer les documents pertinents pour chaque requête\n",
    "for query, relevant_documents in queries:\n",
    "    marked_results_query = search_vectorial(query, relevant_documents, limit=20)\n",
    "    interpolated_precision_query = calculate_interpolated_precision(marked_results_query, recall_levels)\n",
    "    interpolated_precision_all_queries.append((query, interpolated_precision_query))\n",
    "\n",
    "# Plot Interpolated Precision Recall Curve for all queries\n",
    "plt.figure(figsize=(10, 6))\n",
    "for query, interpolated_precision_query in interpolated_precision_all_queries:\n",
    "    plt.plot(recall_levels, interpolated_precision_query, label=query)\n",
    "\n",
    "plt.title('Interpolated Precision-Recall Curve for Queries')\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Interpolated Precision')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "8f350814",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Precision: 0.061\n",
      "Mean Average Precision (MAP): 0.061\n",
      "Average F-measure: 0.000\n"
     ]
    }
   ],
   "source": [
    "# Calculate average precision, MAP, and F-measure for all queries\n",
    "average_precisions_all_queries = []\n",
    "f_measures_all_queries = []\n",
    "\n",
    "for _, interpolated_precision_query in interpolated_precision_all_queries:\n",
    "    average_precision_query = calculate_average_precision(interpolated_precision_query)\n",
    "    f_measure_query = calculate_f_measure(interpolated_precision_query[-1], recall_levels[-1])\n",
    "    \n",
    "    average_precisions_all_queries.append(average_precision_query)\n",
    "    f_measures_all_queries.append(f_measure_query)\n",
    "\n",
    "# Calculate overall average precision and MAP\n",
    "average_precision = np.mean(average_precisions_all_queries)\n",
    "map_value = calculate_map(interpolated_precision_all_queries)\n",
    "\n",
    "# Print the results\n",
    "print(f\"Average Precision: {average_precision:.3f}\")\n",
    "print(f\"Mean Average Precision (MAP): {map_value:.3f}\")\n",
    "print(f\"Average F-measure: {np.mean(f_measures_all_queries):.3f}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d8160bd9",
   "metadata": {},
   "outputs": [],
   "source": [
    "--------------------------tf-idf with transformated querries +validation "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "61dd6efa",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to\n",
      "[nltk_data]     C:\\Users\\meriem\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import csv\n",
    "from whoosh.fields import Schema, TEXT, ID\n",
    "from whoosh import index\n",
    "from whoosh.analysis import RegexTokenizer, StopFilter, Filter\n",
    "from whoosh.qparser import QueryParser\n",
    "from whoosh import scoring\n",
    "from nltk.stem.isri import ISRIStemmer\n",
    "import os.path\n",
    "from nltk.corpus import stopwords\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import nltk\n",
    "\n",
    "nltk.download('stopwords')\n",
    "\n",
    "# Charger la liste des mots vides en arabe de NLTK\n",
    "arabic_stopwords = set(stopwords.words(\"arabic\"))\n",
    "\n",
    "class ArabicStemFilter(Filter):\n",
    "    def __init__(self):\n",
    "        self.stemmer = ISRIStemmer()\n",
    "\n",
    "    def __call__(self, tokens):\n",
    "        for t in tokens:\n",
    "            t.text = self.stemmer.stem(t.text)\n",
    "            yield t\n",
    "\n",
    "# Créer un analyseur personnalisé avec la liste des mots vides de NLTK\n",
    "analyzer = RegexTokenizer() | StopFilter(stoplist=arabic_stopwords) | ArabicStemFilter()\n",
    "\n",
    "# Créer le schéma en utilisant l'analyseur personnalisé\n",
    "schema = Schema(\n",
    "    title=TEXT(stored=True),\n",
    "    path=ID(stored=True),\n",
    "    content=TEXT(stored=True, analyzer=analyzer, vector=True)  # Ajouter \"vector=True\"\n",
    ")\n",
    "\n",
    "# Créer le répertoire d'index s'il n'existe pas\n",
    "if not os.path.exists(\"Index-2\"):\n",
    "    os.mkdir(\"Index-2\")\n",
    "\n",
    "# Lister les fichiers dans le répertoire \"Corpus\"\n",
    "corpus_dir = \"E:/M2/RI/Corpus/Corpus\"\n",
    "file_names = os.listdir(corpus_dir)\n",
    "\n",
    "# Créer l'index\n",
    "ix = index.create_in(\"Index-2\", schema)\n",
    "\n",
    "# Indexer les documents avec prétraitement\n",
    "with ix.writer() as writer:\n",
    "    for file_name in file_names:\n",
    "        with open(os.path.join(corpus_dir, file_name), 'r', encoding='windows-1256') as file:\n",
    "            content = file.read()\n",
    "            writer.add_document(title=file_name, content=content, path='')\n",
    "\n",
    "def calculate_interpolated_precision(results, recall_levels):\n",
    "    recall_values = [result[3] for result in results]\n",
    "    precision_values = [result[4] for result in results]\n",
    "\n",
    "    interpolated_precisions = []\n",
    "    for recall_level in recall_levels:\n",
    "        max_precision = 0\n",
    "        for recall, precision in zip(recall_values, precision_values):\n",
    "            if recall >= recall_level:\n",
    "                max_precision = max(max_precision, precision)\n",
    "        interpolated_precisions.append(max_precision)\n",
    "\n",
    "    return interpolated_precisions\n",
    "\n",
    "def search_vectorial(query_str, relevant_documents, limit=20):\n",
    "    with ix.searcher(weighting=scoring.TF_IDF()) as searcher:  # Use scoring module for TF-IDF\n",
    "        # Définir un analyseur et un parseur liés à l'index\n",
    "        query_parser = QueryParser(\"content\", ix.schema)\n",
    "        query = query_parser.parse(query_str)\n",
    "\n",
    "        # Recherche Vectorielle (utilisation par défaut de Whoosh)\n",
    "        results_vectorial = searcher.search(query, terms=True, limit=limit)\n",
    "\n",
    "        # Marquer les résultats en fonction de la liste des documents pertinents\n",
    "        marked_results = []\n",
    "        true_positives = 0  # Count of relevant documents retrieved\n",
    "        false_positives = 0  # Count of irrelevant documents retrieved\n",
    "        total_relevant_documents = len(relevant_documents)\n",
    "\n",
    "        if total_relevant_documents == 0:\n",
    "            recall = precision = 0\n",
    "        else:\n",
    "            for rank, r in enumerate(results_vectorial, start=1):\n",
    "                is_relevant = r['title'] in relevant_documents\n",
    "                mark = 'YES' if is_relevant else 'NO'\n",
    "\n",
    "                # Calculate recall and precision for each document\n",
    "                if is_relevant:\n",
    "                    true_positives += 1\n",
    "                else:\n",
    "                    false_positives += 1\n",
    "\n",
    "                recall = round(true_positives / total_relevant_documents, 3)\n",
    "                precision = round(true_positives / (true_positives + false_positives), 3)\n",
    "\n",
    "                # Append information to the marked result\n",
    "                marked_results.append((rank, r['title'], mark, recall, precision))\n",
    "\n",
    "        return marked_results\n",
    "\n",
    "# Define the queries and relevant documents for each query\n",
    "queries = [\n",
    "    (\"الكلاسيكية OR الحلول OR حل ORجذر  \", []),\n",
    "    (\"أوبك OR منظمة OR الدول OR المصدرة OR للنفط \", [\"D493 A.txt\", \"D330 A.txt\"]),\n",
    "    (\"التفاضلية OR المعادلات\", [\"D508 A.txt\", \"D081 A.txt\", \"D105 A.txt\", \"D162 A.txt\", \"D329 A.txt\", \"D552 A.txt\", \"D464 A.txt\"])\n",
    "]\n",
    "\n",
    "# Interpolated Precision Recall Curve\n",
    "recall_levels = np.arange(0, 1.1, 0.1)\n",
    "interpolated_precision_all_queries = []\n",
    "\n",
    "# Rechercher et marquer les documents pertinents pour chaque requête\n",
    "for query, relevant_documents in queries:\n",
    "    marked_results_query = search_vectorial(query, relevant_documents, limit=20)\n",
    "    interpolated_precision_query = calculate_interpolated_precision(marked_results_query, recall_levels)\n",
    "    interpolated_precision_all_queries.append((query, interpolated_precision_query))\n",
    "\n",
    "# Plot Interpolated Precision Recall Curve for all queries\n",
    "plt.figure(figsize=(10, 6))\n",
    "for query, interpolated_precision_query in interpolated_precision_all_queries:\n",
    "    plt.plot(recall_levels, interpolated_precision_query, label=query)\n",
    "\n",
    "plt.title('Interpolated Precision-Recall Curve for Queries')\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Interpolated Precision')\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "6cd8a2fa",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Precision: 0.061\n",
      "Mean Average Precision (MAP): 0.061\n",
      "Average F-measure: 0.000\n"
     ]
    }
   ],
   "source": [
    "def calculate_average_precision(interpolated_precisions):\n",
    "    return np.mean(interpolated_precisions)\n",
    "\n",
    "def calculate_map(interpolated_precisions_all_queries):\n",
    "    return np.mean([calculate_average_precision(query_precisions) for _, query_precisions in interpolated_precisions_all_queries])\n",
    "\n",
    "def calculate_f_measure(precision, recall, beta=1):\n",
    "    if precision + recall == 0:\n",
    "        return 0\n",
    "    return (1 + beta**2) * (precision * recall) / ((beta**2 * precision) + recall)\n",
    "\n",
    "# Calculate average precision, MAP, and F-measure for all queries\n",
    "average_precisions_all_queries = []\n",
    "f_measures_all_queries = []\n",
    "\n",
    "for _, interpolated_precision_query in interpolated_precision_all_queries:\n",
    "    average_precision_query = calculate_average_precision(interpolated_precision_query)\n",
    "    f_measure_query = calculate_f_measure(interpolated_precision_query[-1], recall_levels[-1])\n",
    "    \n",
    "    average_precisions_all_queries.append(average_precision_query)\n",
    "    f_measures_all_queries.append(f_measure_query)\n",
    "\n",
    "# Calculate overall average precision and MAP\n",
    "average_precision = np.mean(average_precisions_all_queries)\n",
    "map_value = calculate_map(interpolated_precision_all_queries)\n",
    "\n",
    "# Print the results\n",
    "print(f\"Average Precision: {average_precision:.3f}\")\n",
    "print(f\"Mean Average Precision (MAP): {map_value:.3f}\")\n",
    "print(f\"Average F-measure: {np.mean(f_measures_all_queries):.3f}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "754bc6f2",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}