Newer
Older
{"cells":[{"cell_type":"markdown","metadata":{},"source":["<h2> Imports </h2>"]},{"cell_type":"code","execution_count":1,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T18:52:18.410005Z","iopub.status.busy":"2024-01-03T18:52:18.409290Z","iopub.status.idle":"2024-01-03T18:56:01.555478Z","shell.execute_reply":"2024-01-03T18:56:01.553732Z","shell.execute_reply.started":"2024-01-03T18:52:18.409894Z"},"trusted":true},"outputs":[{"name":"stdout","output_type":"stream","text":["Collecting tensorflow==2.5\n"," Downloading tensorflow-2.5.0-cp37-cp37m-manylinux2010_x86_64.whl (454.3 MB)\n","\u001b[K |████████████████████████████████| 454.3 MB 11 kB/s s eta 0:00:01 |█▌ | 20.7 MB 7.4 MB/s eta 0:00:59 |█████████████████████████▋ | 363.2 MB 62.2 MB/s eta 0:00:02 |██████████████████████████████▏ | 427.8 MB 46.2 MB/s eta 0:00:01\n","\u001b[?25hRequirement already satisfied: opt-einsum~=3.3.0 in /opt/conda/lib/python3.7/site-packages (from tensorflow==2.5) (3.3.0)\n","Requirement already satisfied: astunparse~=1.6.3 in /opt/conda/lib/python3.7/site-packages (from tensorflow==2.5) (1.6.3)\n","Requirement already satisfied: absl-py~=0.10 in /opt/conda/lib/python3.7/site-packages (from tensorflow==2.5) (0.12.0)\n","Requirement already satisfied: typing-extensions~=3.7.4 in /opt/conda/lib/python3.7/site-packages (from tensorflow==2.5) (3.7.4.3)\n","Requirement already satisfied: keras-preprocessing~=1.1.2 in /opt/conda/lib/python3.7/site-packages (from tensorflow==2.5) (1.1.2)\n","Collecting h5py~=3.1.0\n"," Downloading h5py-3.1.0-cp37-cp37m-manylinux1_x86_64.whl (4.0 MB)\n","\u001b[K |████████████████████████████████| 4.0 MB 57.1 MB/s eta 0:00:01\n","\u001b[?25hCollecting gast==0.4.0\n"," Downloading gast-0.4.0-py3-none-any.whl (9.8 kB)\n","Requirement already satisfied: protobuf>=3.9.2 in /opt/conda/lib/python3.7/site-packages (from tensorflow==2.5) (3.17.3)\n","Collecting keras-nightly~=2.5.0.dev\n"," Downloading keras_nightly-2.5.0.dev2021032900-py2.py3-none-any.whl (1.2 MB)\n","\u001b[K |████████████████████████████████| 1.2 MB 42.3 MB/s eta 0:00:01\n","\u001b[?25hRequirement already satisfied: wheel~=0.35 in /opt/conda/lib/python3.7/site-packages (from tensorflow==2.5) (0.36.2)\n","Requirement already satisfied: google-pasta~=0.2 in /opt/conda/lib/python3.7/site-packages (from tensorflow==2.5) (0.2.0)\n","Requirement already satisfied: six~=1.15.0 in /opt/conda/lib/python3.7/site-packages (from tensorflow==2.5) (1.15.0)\n","Requirement already satisfied: termcolor~=1.1.0 in /opt/conda/lib/python3.7/site-packages (from tensorflow==2.5) (1.1.0)\n","Requirement already satisfied: flatbuffers~=1.12.0 in /opt/conda/lib/python3.7/site-packages (from tensorflow==2.5) (1.12)\n","Collecting grpcio~=1.34.0\n"," Downloading grpcio-1.34.1-cp37-cp37m-manylinux2014_x86_64.whl (4.0 MB)\n","\u001b[K |████████████████████████████████| 4.0 MB 56.1 MB/s eta 0:00:01\n","\u001b[?25hCollecting tensorflow-estimator<2.6.0,>=2.5.0rc0\n"," Downloading tensorflow_estimator-2.5.0-py2.py3-none-any.whl (462 kB)\n","\u001b[K |████████████████████████████████| 462 kB 61.4 MB/s eta 0:00:01\n","\u001b[?25hRequirement already satisfied: numpy~=1.19.2 in /opt/conda/lib/python3.7/site-packages (from tensorflow==2.5) (1.19.5)\n","Collecting tensorboard~=2.5\n"," Downloading tensorboard-2.11.2-py3-none-any.whl (6.0 MB)\n","\u001b[K |████████████████████████████████| 6.0 MB 38.8 MB/s eta 0:00:01\n","\u001b[?25hRequirement already satisfied: wrapt~=1.12.1 in /opt/conda/lib/python3.7/site-packages (from tensorflow==2.5) (1.12.1)\n","Collecting cached-property\n"," Downloading cached_property-1.5.2-py2.py3-none-any.whl (7.6 kB)\n","Requirement already satisfied: markdown>=2.6.8 in /opt/conda/lib/python3.7/site-packages (from tensorboard~=2.5->tensorflow==2.5) (3.3.4)\n","Requirement already satisfied: google-auth<3,>=1.6.3 in /opt/conda/lib/python3.7/site-packages (from tensorboard~=2.5->tensorflow==2.5) (1.30.2)\n","Requirement already satisfied: requests<3,>=2.21.0 in /opt/conda/lib/python3.7/site-packages (from tensorboard~=2.5->tensorflow==2.5) (2.25.1)\n","Requirement already satisfied: google-auth-oauthlib<0.5,>=0.4.1 in /opt/conda/lib/python3.7/site-packages (from tensorboard~=2.5->tensorflow==2.5) (0.4.4)\n","Requirement already satisfied: setuptools>=41.0.0 in /opt/conda/lib/python3.7/site-packages (from tensorboard~=2.5->tensorflow==2.5) (49.6.0.post20210108)\n","Requirement already satisfied: werkzeug>=1.0.1 in /opt/conda/lib/python3.7/site-packages (from tensorboard~=2.5->tensorflow==2.5) (2.0.1)\n","Requirement already satisfied: tensorboard-data-server<0.7.0,>=0.6.0 in /opt/conda/lib/python3.7/site-packages (from tensorboard~=2.5->tensorflow==2.5) (0.6.1)\n","Requirement already satisfied: tensorboard-plugin-wit>=1.6.0 in /opt/conda/lib/python3.7/site-packages (from tensorboard~=2.5->tensorflow==2.5) (1.8.0)\n","Requirement already satisfied: cachetools<5.0,>=2.0.0 in /opt/conda/lib/python3.7/site-packages (from google-auth<3,>=1.6.3->tensorboard~=2.5->tensorflow==2.5) (4.2.2)\n","Requirement already satisfied: rsa<5,>=3.1.4 in /opt/conda/lib/python3.7/site-packages (from google-auth<3,>=1.6.3->tensorboard~=2.5->tensorflow==2.5) (4.7.2)\n","Requirement already satisfied: pyasn1-modules>=0.2.1 in /opt/conda/lib/python3.7/site-packages (from google-auth<3,>=1.6.3->tensorboard~=2.5->tensorflow==2.5) (0.2.7)\n","Requirement already satisfied: requests-oauthlib>=0.7.0 in /opt/conda/lib/python3.7/site-packages (from google-auth-oauthlib<0.5,>=0.4.1->tensorboard~=2.5->tensorflow==2.5) (1.3.0)\n","Requirement already satisfied: importlib-metadata in /opt/conda/lib/python3.7/site-packages (from markdown>=2.6.8->tensorboard~=2.5->tensorflow==2.5) (3.4.0)\n","Requirement already satisfied: pyasn1<0.5.0,>=0.4.6 in /opt/conda/lib/python3.7/site-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard~=2.5->tensorflow==2.5) (0.4.8)\n","Requirement already satisfied: idna<3,>=2.5 in /opt/conda/lib/python3.7/site-packages (from requests<3,>=2.21.0->tensorboard~=2.5->tensorflow==2.5) (2.10)\n","Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.7/site-packages (from requests<3,>=2.21.0->tensorboard~=2.5->tensorflow==2.5) (2021.5.30)\n","Requirement already satisfied: urllib3<1.27,>=1.21.1 in /opt/conda/lib/python3.7/site-packages (from requests<3,>=2.21.0->tensorboard~=2.5->tensorflow==2.5) (1.26.5)\n","Requirement already satisfied: chardet<5,>=3.0.2 in /opt/conda/lib/python3.7/site-packages (from requests<3,>=2.21.0->tensorboard~=2.5->tensorflow==2.5) (4.0.0)\n","Requirement already satisfied: oauthlib>=3.0.0 in /opt/conda/lib/python3.7/site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<0.5,>=0.4.1->tensorboard~=2.5->tensorflow==2.5) (3.1.1)\n","Requirement already satisfied: zipp>=0.5 in /opt/conda/lib/python3.7/site-packages (from importlib-metadata->markdown>=2.6.8->tensorboard~=2.5->tensorflow==2.5) (3.4.1)\n","Installing collected packages: grpcio, cached-property, tensorflow-estimator, tensorboard, keras-nightly, h5py, gast, tensorflow\n"," Attempting uninstall: grpcio\n"," Found existing installation: grpcio 1.32.0\n"," Uninstalling grpcio-1.32.0:\n"," Successfully uninstalled grpcio-1.32.0\n"," Attempting uninstall: tensorflow-estimator\n"," Found existing installation: tensorflow-estimator 2.4.0\n"," Uninstalling tensorflow-estimator-2.4.0:\n"," Successfully uninstalled tensorflow-estimator-2.4.0\n"," Attempting uninstall: tensorboard\n"," Found existing installation: tensorboard 2.4.1\n"," Uninstalling tensorboard-2.4.1:\n"," Successfully uninstalled tensorboard-2.4.1\n"," Attempting uninstall: h5py\n"," Found existing installation: h5py 2.10.0\n"," Uninstalling h5py-2.10.0:\n"," Successfully uninstalled h5py-2.10.0\n"," Attempting uninstall: gast\n"," Found existing installation: gast 0.3.3\n"," Uninstalling gast-0.3.3:\n"," Successfully uninstalled gast-0.3.3\n"," Attempting uninstall: tensorflow\n"," Found existing installation: tensorflow 2.4.1\n"," Uninstalling tensorflow-2.4.1:\n"," Successfully uninstalled tensorflow-2.4.1\n","Successfully installed cached-property-1.5.2 gast-0.4.0 grpcio-1.34.1 h5py-3.1.0 keras-nightly-2.5.0.dev2021032900 tensorboard-2.11.2 tensorflow-2.5.0 tensorflow-estimator-2.5.0\n","\u001b[33mWARNING: Running pip as root will break packages and permissions. You should install packages reliably by using venv: https://pip.pypa.io/warnings/venv\u001b[0m\n","Collecting rank_bm25\n"," Downloading rank_bm25-0.2.2-py3-none-any.whl (8.6 kB)\n","Requirement already satisfied: numpy in /opt/conda/lib/python3.7/site-packages (from rank_bm25) (1.19.5)\n","Installing collected packages: rank-bm25\n","Successfully installed rank-bm25-0.2.2\n","\u001b[33mWARNING: Running pip as root will break packages and permissions. You should install packages reliably by using venv: https://pip.pypa.io/warnings/venv\u001b[0m\n","\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n","yellowbrick 1.3.post1 requires numpy<1.20,>=1.16.0, but you have numpy 1.21.6 which is incompatible.\n","tensorflow 2.5.0 requires numpy~=1.19.2, but you have numpy 1.21.6 which is incompatible.\n","s3fs 2021.6.1 requires fsspec==2021.06.1, but you have fsspec 2023.1.0 which is incompatible.\n","pdpbox 0.2.1 requires matplotlib==3.1.1, but you have matplotlib 3.4.2 which is incompatible.\n","matrixprofile 1.1.10 requires protobuf==3.11.2, but you have protobuf 3.17.3 which is incompatible.\n","kornia 0.5.5 requires numpy<=1.19, but you have numpy 1.21.6 which is incompatible.\n","imbalanced-learn 0.8.0 requires scikit-learn>=0.24, but you have scikit-learn 0.23.2 which is incompatible.\n","gcsfs 2021.6.0 requires fsspec==2021.06.0, but you have fsspec 2023.1.0 which is incompatible.\n","allennlp 2.5.0 requires transformers<4.7,>=4.1, but you have transformers 4.30.2 which is incompatible.\u001b[0m\n","\u001b[33mWARNING: Running pip as root will break packages and permissions. You should install packages reliably by using venv: https://pip.pypa.io/warnings/venv\u001b[0m\n","Note: you may need to restart the kernel to use updated packages.\n"]}],"source":["!pip install tensorflow==2.5\n","!pip install rank_bm25\n","%pip install -Uq sentence-transformers faiss-cpu accelerate hdbscan bertopic evaluate kaleido datasets>=2.11"]},{"cell_type":"code","execution_count":39,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T18:56:01.558915Z","iopub.status.busy":"2024-01-03T18:56:01.558368Z","iopub.status.idle":"2024-01-03T18:56:32.924438Z","shell.execute_reply":"2024-01-03T18:56:32.919977Z","shell.execute_reply.started":"2024-01-03T18:56:01.558854Z"},"trusted":true},"outputs":[],"source":["import gensim\n","import numpy as np\n","import nltk\n","from nltk.corpus import stopwords\n","from nltk.tokenize import word_tokenize\n","from scipy import spatial\n","from nltk.tokenize.toktok import ToktokTokenizer\n","import re\n","tokenizer = ToktokTokenizer()\n","stopword_list = nltk.corpus.stopwords.words('english')\n","import pandas as pd\n","from tqdm import tqdm\n","tqdm.pandas()\n","from rank_bm25 import BM25Okapi\n","from bertopic import BERTopic\n","from bertopic.vectorizers import ClassTfidfTransformer\n","import numpy as np\n","import pandas as pd \n","import os\n","from nltk.stem.porter import PorterStemmer\n","import string\n","from nltk.stem import WordNetLemmatizer\n","import matplotlib.pyplot as plt\n","import pickle \n"]},{"cell_type":"markdown","metadata":{},"source":["<h2> Load datasets and model</h2>"]},{"cell_type":"code","execution_count":30,"metadata":{"_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","execution":{"iopub.execute_input":"2024-01-03T18:56:32.931742Z","iopub.status.busy":"2024-01-03T18:56:32.929029Z","iopub.status.idle":"2024-01-03T18:56:33.045085Z","shell.execute_reply":"2024-01-03T18:56:33.043862Z","shell.execute_reply.started":"2024-01-03T18:56:32.931643Z"},"trusted":true},"outputs":[],"source":["queries = pd.read_csv(\"./data/cisi-csv/queries.csv\")\n","docs = pd.read_csv(\"./data/cisi-csv/docs.csv\")\n","rels = pd.read_csv(\"./data/cisi-csv/rels.csv\")\n","\n","full_doc = docs['text'].to_list()\n","full_query = queries['text'].to_list()"]},{"cell_type":"code","execution_count":31,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T18:56:33.047676Z","iopub.status.busy":"2024-01-03T18:56:33.046992Z","iopub.status.idle":"2024-01-03T18:56:34.042871Z","shell.execute_reply":"2024-01-03T18:56:34.041391Z","shell.execute_reply.started":"2024-01-03T18:56:33.047624Z"},"trusted":true},"outputs":[{"name":"stdout","output_type":"stream","text":["There are 36 queries without a groundtruth.\n","Remaining queries: 76.\n"]},{"data":{"text/html":["<div>\n","<style scoped>\n"," .dataframe tbody tr th:only-of-type {\n"," vertical-align: middle;\n"," }\n","\n"," .dataframe tbody tr th {\n"," vertical-align: top;\n"," }\n","\n"," .dataframe thead th {\n"," text-align: right;\n"," }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n"," <thead>\n"," <tr style=\"text-align: right;\">\n"," <th></th>\n"," <th>id</th>\n"," <th>text</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>1</td>\n"," <td>What problems and concerns are there in making...</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>2</td>\n"," <td>How can actually pertinent data, as opposed to...</td>\n"," </tr>\n"," <tr>\n"," <th>2</th>\n"," <td>3</td>\n"," <td>What is information science? Give definitions ...</td>\n"," </tr>\n"," <tr>\n"," <th>3</th>\n"," <td>4</td>\n"," <td>Image recognition and any other methods of aut...</td>\n"," </tr>\n"," <tr>\n"," <th>4</th>\n"," <td>5</td>\n"," <td>What special training will ordinary researcher...</td>\n"," </tr>\n"," <tr>\n"," <th>...</th>\n"," <td>...</td>\n"," <td>...</td>\n"," </tr>\n"," <tr>\n"," <th>100</th>\n"," <td>101</td>\n"," <td>.T Parallel Computations in Information Retrie...</td>\n"," </tr>\n"," <tr>\n"," <th>101</th>\n"," <td>102</td>\n"," <td>.T The Measurement of Term Importance in Autom...</td>\n"," </tr>\n"," <tr>\n"," <th>103</th>\n"," <td>104</td>\n"," <td>.T The Selection of Good Search Terms .A van R...</td>\n"," </tr>\n"," <tr>\n"," <th>108</th>\n"," <td>109</td>\n"," <td>.T Author Cocitation: A Literature Measure of ...</td>\n"," </tr>\n"," <tr>\n"," <th>110</th>\n"," <td>111</td>\n"," <td>.T Document Clustering Using an Inverted File ...</td>\n"," </tr>\n"," </tbody>\n","</table>\n","<p>76 rows × 2 columns</p>\n","</div>"],"text/plain":[" id text\n","0 1 What problems and concerns are there in making...\n","1 2 How can actually pertinent data, as opposed to...\n","2 3 What is information science? Give definitions ...\n","3 4 Image recognition and any other methods of aut...\n","4 5 What special training will ordinary researcher...\n",".. ... ...\n","100 101 .T Parallel Computations in Information Retrie...\n","101 102 .T The Measurement of Term Importance in Autom...\n","103 104 .T The Selection of Good Search Terms .A van R...\n","108 109 .T Author Cocitation: A Literature Measure of ...\n","110 111 .T Document Clustering Using an Inverted File ...\n","\n","[76 rows x 2 columns]"]},"execution_count":31,"metadata":{},"output_type":"execute_result"}],"source":["# TODO: this is not really necessary I think? because\n","#remove queries where we don't have a groundtruth for:\n","queries_wo_gt = [36,38,40,47,48,51,53,59,60,63,64,68,70,72,73,74,75,77,78,80,83,85,86,87,88,89,91,93,94,103,105,106,107,108,110,112]\n","print(f'There are {len(queries_wo_gt)} queries without a groundtruth.')\n","print(f'Remaining queries: {len(queries)-len(queries_wo_gt)}.')\n","\n","queries = queries[~queries['id'].isin(queries_wo_gt)]\n","queries"]},{"cell_type":"code","execution_count":32,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T19:04:27.585430Z","iopub.status.busy":"2024-01-03T19:04:27.584998Z","iopub.status.idle":"2024-01-03T19:06:04.031449Z","shell.execute_reply":"2024-01-03T19:06:04.028796Z","shell.execute_reply.started":"2024-01-03T19:04:27.585396Z"},"trusted":true},"outputs":[{"ename":"FileNotFoundError","evalue":"[Errno 2] No such file or directory: '../input/googlenewsvectorsnegative300/GoogleNews-vectors-negative300.bin'","output_type":"error","traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mFileNotFoundError\u001b[0m Traceback (most recent call last)","\u001b[1;32m/home/hanna/Documents/air-23/initial-retrieval.ipynb Cell 7\u001b[0m line \u001b[0;36m1\n\u001b[0;32m----> <a href='vscode-notebook-cell:/home/hanna/Documents/air-23/initial-retrieval.ipynb#Y125sZmlsZQ%3D%3D?line=0'>1</a>\u001b[0m model \u001b[39m=\u001b[39m gensim\u001b[39m.\u001b[39;49mmodels\u001b[39m.\u001b[39;49mKeyedVectors\u001b[39m.\u001b[39;49mload_word2vec_format(\u001b[39m'\u001b[39;49m\u001b[39m../input/googlenewsvectorsnegative300/GoogleNews-vectors-negative300.bin\u001b[39;49m\u001b[39m'\u001b[39;49m, binary\u001b[39m=\u001b[39;49m\u001b[39mTrue\u001b[39;49;00m)\n","File \u001b[0;32m~/.local/lib/python3.10/site-packages/gensim/models/keyedvectors.py:1719\u001b[0m, in \u001b[0;36mKeyedVectors.load_word2vec_format\u001b[0;34m(cls, fname, fvocab, binary, encoding, unicode_errors, limit, datatype, no_header)\u001b[0m\n\u001b[1;32m 1672\u001b[0m \u001b[39m@classmethod\u001b[39m\n\u001b[1;32m 1673\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mload_word2vec_format\u001b[39m(\n\u001b[1;32m 1674\u001b[0m \u001b[39mcls\u001b[39m, fname, fvocab\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m, binary\u001b[39m=\u001b[39m\u001b[39mFalse\u001b[39;00m, encoding\u001b[39m=\u001b[39m\u001b[39m'\u001b[39m\u001b[39mutf8\u001b[39m\u001b[39m'\u001b[39m, unicode_errors\u001b[39m=\u001b[39m\u001b[39m'\u001b[39m\u001b[39mstrict\u001b[39m\u001b[39m'\u001b[39m,\n\u001b[1;32m 1675\u001b[0m limit\u001b[39m=\u001b[39m\u001b[39mNone\u001b[39;00m, datatype\u001b[39m=\u001b[39mREAL, no_header\u001b[39m=\u001b[39m\u001b[39mFalse\u001b[39;00m,\n\u001b[1;32m 1676\u001b[0m ):\n\u001b[1;32m 1677\u001b[0m \u001b[39m \u001b[39m\u001b[39m\"\"\"Load KeyedVectors from a file produced by the original C word2vec-tool format.\u001b[39;00m\n\u001b[1;32m 1678\u001b[0m \n\u001b[1;32m 1679\u001b[0m \u001b[39m Warnings\u001b[39;00m\n\u001b[0;32m (...)\u001b[0m\n\u001b[1;32m 1717\u001b[0m \n\u001b[1;32m 1718\u001b[0m \u001b[39m \"\"\"\u001b[39;00m\n\u001b[0;32m-> 1719\u001b[0m \u001b[39mreturn\u001b[39;00m _load_word2vec_format(\n\u001b[1;32m 1720\u001b[0m \u001b[39mcls\u001b[39;49m, fname, fvocab\u001b[39m=\u001b[39;49mfvocab, binary\u001b[39m=\u001b[39;49mbinary, encoding\u001b[39m=\u001b[39;49mencoding, unicode_errors\u001b[39m=\u001b[39;49municode_errors,\n\u001b[1;32m 1721\u001b[0m limit\u001b[39m=\u001b[39;49mlimit, datatype\u001b[39m=\u001b[39;49mdatatype, no_header\u001b[39m=\u001b[39;49mno_header,\n\u001b[1;32m 1722\u001b[0m )\n","File \u001b[0;32m~/.local/lib/python3.10/site-packages/gensim/models/keyedvectors.py:2048\u001b[0m, in \u001b[0;36m_load_word2vec_format\u001b[0;34m(cls, fname, fvocab, binary, encoding, unicode_errors, limit, datatype, no_header, binary_chunk_size)\u001b[0m\n\u001b[1;32m 2045\u001b[0m counts[word] \u001b[39m=\u001b[39m \u001b[39mint\u001b[39m(count)\n\u001b[1;32m 2047\u001b[0m logger\u001b[39m.\u001b[39minfo(\u001b[39m\"\u001b[39m\u001b[39mloading projection weights from \u001b[39m\u001b[39m%s\u001b[39;00m\u001b[39m\"\u001b[39m, fname)\n\u001b[0;32m-> 2048\u001b[0m \u001b[39mwith\u001b[39;00m utils\u001b[39m.\u001b[39;49mopen(fname, \u001b[39m'\u001b[39;49m\u001b[39mrb\u001b[39;49m\u001b[39m'\u001b[39;49m) \u001b[39mas\u001b[39;00m fin:\n\u001b[1;32m 2049\u001b[0m \u001b[39mif\u001b[39;00m no_header:\n\u001b[1;32m 2050\u001b[0m \u001b[39m# deduce both vocab_size & vector_size from 1st pass over file\u001b[39;00m\n\u001b[1;32m 2051\u001b[0m \u001b[39mif\u001b[39;00m binary:\n","File \u001b[0;32m~/.local/lib/python3.10/site-packages/smart_open/smart_open_lib.py:177\u001b[0m, in \u001b[0;36mopen\u001b[0;34m(uri, mode, buffering, encoding, errors, newline, closefd, opener, compression, transport_params)\u001b[0m\n\u001b[1;32m 174\u001b[0m \u001b[39mif\u001b[39;00m transport_params \u001b[39mis\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[1;32m 175\u001b[0m transport_params \u001b[39m=\u001b[39m {}\n\u001b[0;32m--> 177\u001b[0m fobj \u001b[39m=\u001b[39m _shortcut_open(\n\u001b[1;32m 178\u001b[0m uri,\n\u001b[1;32m 179\u001b[0m mode,\n\u001b[1;32m 180\u001b[0m compression\u001b[39m=\u001b[39;49mcompression,\n\u001b[1;32m 181\u001b[0m buffering\u001b[39m=\u001b[39;49mbuffering,\n\u001b[1;32m 182\u001b[0m encoding\u001b[39m=\u001b[39;49mencoding,\n\u001b[1;32m 183\u001b[0m errors\u001b[39m=\u001b[39;49merrors,\n\u001b[1;32m 184\u001b[0m newline\u001b[39m=\u001b[39;49mnewline,\n\u001b[1;32m 185\u001b[0m )\n\u001b[1;32m 186\u001b[0m \u001b[39mif\u001b[39;00m fobj \u001b[39mis\u001b[39;00m \u001b[39mnot\u001b[39;00m \u001b[39mNone\u001b[39;00m:\n\u001b[1;32m 187\u001b[0m \u001b[39mreturn\u001b[39;00m fobj\n","File \u001b[0;32m~/.local/lib/python3.10/site-packages/smart_open/smart_open_lib.py:363\u001b[0m, in \u001b[0;36m_shortcut_open\u001b[0;34m(uri, mode, compression, buffering, encoding, errors, newline)\u001b[0m\n\u001b[1;32m 360\u001b[0m \u001b[39mif\u001b[39;00m errors \u001b[39mand\u001b[39;00m \u001b[39m'\u001b[39m\u001b[39mb\u001b[39m\u001b[39m'\u001b[39m \u001b[39mnot\u001b[39;00m \u001b[39min\u001b[39;00m mode:\n\u001b[1;32m 361\u001b[0m open_kwargs[\u001b[39m'\u001b[39m\u001b[39merrors\u001b[39m\u001b[39m'\u001b[39m] \u001b[39m=\u001b[39m errors\n\u001b[0;32m--> 363\u001b[0m \u001b[39mreturn\u001b[39;00m _builtin_open(local_path, mode, buffering\u001b[39m=\u001b[39;49mbuffering, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mopen_kwargs)\n","\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: '../input/googlenewsvectorsnegative300/GoogleNews-vectors-negative300.bin'"]}],"source":["model = gensim.models.KeyedVectors.load_word2vec_format('./models/GoogleNews-vectors-negative300.bin', binary=True)"]},{"cell_type":"markdown","metadata":{},"source":["<h2>Initial retrieval with bm25</h2>"]},{"cell_type":"code","execution_count":33,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T18:57:01.149061Z","iopub.status.busy":"2024-01-03T18:57:01.148640Z","iopub.status.idle":"2024-01-03T18:57:01.166124Z","shell.execute_reply":"2024-01-03T18:57:01.164536Z","shell.execute_reply.started":"2024-01-03T18:57:01.149026Z"},"trusted":true},"outputs":[],"source":["def data_clean(text):\n"," pattern = r'[^a-zA-Z0-9\\s]'\n"," text = re.sub(pattern,'',' '.join(text))\n"," tokens = [token.strip() for token in text.split()]\n"," filtered = [token for token in tokens if token.lower() not in stopword_list]\n"," filtered = ' '.join(filtered)\n"," return filtered\n","\n","# just the same code as above to clean the df texts for bm25\n","def data_clean_df(text):\n"," # Regex pattern to keep only alphanumeric characters and spaces\n"," pattern = r'[^a-zA-Z0-9\\s]'\n"," text = re.sub(pattern, '', text)\n"," tokens = [token.strip() for token in text.split()]\n"," return ' '.join(tokens)\n","\n","\n","#function is needed to get the texts of the relevant documents from initial retrieval\n","def get_texts_from_df(doc_ids, df):\n"," return df[df['id'].isin(doc_ids)]['text'].tolist()\n","\n","def embeddings(word):\n"," if word in model.key_to_index:\n"," return model.get_vector(word)\n"," else:\n"," return np.zeros(300)\n"," \n","def get_sim(average_vec_query, average_vec_docs):\n"," sim = [(1 - spatial.distance.cosine(average_vec_query, average_vec_docs))]\n"," return sim\n","\n","#some queries have a .T in the begining we want to remove this\n","def clean_query(text):\n"," pattern = r'^\\.T\\s'\n"," tokens = [token.strip() for token in text.split()]\n"," return ' '.join(tokens)\n","\n","#special pre-processing for bm25, because for embeddings we don't want to pre-process that much\n","def data_clean_for_bm25(text):\n"," # Lowercasing the text\n"," text = text.lower()\n"," # Removing digits\n"," text = re.sub(r'\\d+', '', text)\n"," # Removing punctuation\n"," translator = str.maketrans('', '', string.punctuation)\n"," text = text.translate(translator)\n"," # Whitespace normalization\n"," text = \" \".join(text.split())\n"," # Stopword removal\n"," stop_words = set(stopwords.words(\"english\"))\n"," word_tokens = word_tokenize(text)\n"," filtered_words = [word for word in word_tokens if word not in stop_words]\n"," # Lemmatization\n"," lemmatizer = WordNetLemmatizer()\n"," lemmas = [lemmatizer.lemmatize(word) for word in filtered_words]\n","\n"," return lemmas"]},{"cell_type":"code","execution_count":34,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T18:57:04.266231Z","iopub.status.busy":"2024-01-03T18:57:04.265732Z","iopub.status.idle":"2024-01-03T18:57:10.812646Z","shell.execute_reply":"2024-01-03T18:57:10.811193Z","shell.execute_reply.started":"2024-01-03T18:57:04.266183Z"},"trusted":true},"outputs":[{"data":{"text/html":["<div>\n","<style scoped>\n"," .dataframe tbody tr th:only-of-type {\n"," vertical-align: middle;\n"," }\n","\n"," .dataframe tbody tr th {\n"," vertical-align: top;\n"," }\n","\n"," .dataframe thead th {\n"," text-align: right;\n"," }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n"," <thead>\n"," <tr style=\"text-align: right;\">\n"," <th></th>\n"," <th>id</th>\n"," <th>text</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>1</td>\n"," <td>[problem, concern, making, descriptive, title,...</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>2</td>\n"," <td>[actually, pertinent, data, opposed, reference...</td>\n"," </tr>\n"," <tr>\n"," <th>2</th>\n"," <td>3</td>\n"," <td>[information, science, give, definition, possi...</td>\n"," </tr>\n"," <tr>\n"," <th>3</th>\n"," <td>4</td>\n"," <td>[image, recognition, method, automatically, tr...</td>\n"," </tr>\n"," <tr>\n"," <th>4</th>\n"," <td>5</td>\n"," <td>[special, training, ordinary, researcher, busi...</td>\n"," </tr>\n"," <tr>\n"," <th>...</th>\n"," <td>...</td>\n"," <td>...</td>\n"," </tr>\n"," <tr>\n"," <th>100</th>\n"," <td>101</td>\n"," <td>[parallel, computation, information, retrieval...</td>\n"," </tr>\n"," <tr>\n"," <th>101</th>\n"," <td>102</td>\n"," <td>[measurement, term, importance, automatic, ind...</td>\n"," </tr>\n"," <tr>\n"," <th>103</th>\n"," <td>104</td>\n"," <td>[selection, good, search, term, van, rijsberge...</td>\n"," </tr>\n"," <tr>\n"," <th>108</th>\n"," <td>109</td>\n"," <td>[author, cocitation, literature, measure, inte...</td>\n"," </tr>\n"," <tr>\n"," <th>110</th>\n"," <td>111</td>\n"," <td>[document, clustering, using, inverted, file, ...</td>\n"," </tr>\n"," </tbody>\n","</table>\n","<p>76 rows × 2 columns</p>\n","</div>"],"text/plain":[" id text\n","0 1 [problem, concern, making, descriptive, title,...\n","1 2 [actually, pertinent, data, opposed, reference...\n","2 3 [information, science, give, definition, possi...\n","3 4 [image, recognition, method, automatically, tr...\n","4 5 [special, training, ordinary, researcher, busi...\n",".. ... ...\n","100 101 [parallel, computation, information, retrieval...\n","101 102 [measurement, term, importance, automatic, ind...\n","103 104 [selection, good, search, term, van, rijsberge...\n","108 109 [author, cocitation, literature, measure, inte...\n","110 111 [document, clustering, using, inverted, file, ...\n","\n","[76 rows x 2 columns]"]},"execution_count":34,"metadata":{},"output_type":"execute_result"}],"source":["queries_cleaned = queries.copy()\n","queries_cleaned['text'] = queries_cleaned['text'].apply(data_clean_df)\n","queries_cleaned['text'] = queries_cleaned['text'].apply(clean_query)\n","\n","docs_cleaned = docs.copy()\n","docs_cleaned['text'] = docs_cleaned['text'].apply(data_clean_df)\n","docs_cleaned\n","\n","queries_cleaned_bm25 = queries.copy()\n","queries_cleaned_bm25['text'] = queries_cleaned_bm25['text'].apply(data_clean_for_bm25)\n","\n","docs_cleaned_bm25 = docs.copy()\n","docs_cleaned_bm25['text'] = docs_cleaned_bm25['text'].apply(data_clean_for_bm25)\n","queries_cleaned_bm25"]},{"cell_type":"code","execution_count":35,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T18:57:10.815397Z","iopub.status.busy":"2024-01-03T18:57:10.814883Z","iopub.status.idle":"2024-01-03T18:57:10.918743Z","shell.execute_reply":"2024-01-03T18:57:10.917514Z","shell.execute_reply.started":"2024-01-03T18:57:10.815350Z"},"trusted":true},"outputs":[],"source":["corpus = docs_cleaned_bm25['text'].to_list()\n","bm25 = BM25Okapi(corpus)"]},{"cell_type":"code","execution_count":36,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T18:57:25.428988Z","iopub.status.busy":"2024-01-03T18:57:25.428280Z","iopub.status.idle":"2024-01-03T18:57:25.442445Z","shell.execute_reply":"2024-01-03T18:57:25.440524Z","shell.execute_reply.started":"2024-01-03T18:57:25.428941Z"},"trusted":true},"outputs":[],"source":["def initial_retrieval_bm25(query_id, query_text, bm25, k):\n"," query = query_text\n"," document_ids = docs_cleaned['id'].to_list()\n"," tokenized_query = query.split(\" \")\n"," doc_scores = bm25.get_scores(tokenized_query)\n"," doc_scores_dict = dict(zip(document_ids, doc_scores))\n"," #print(doc_scores_dict)\n"," most_similar_init_k_documents = {doc_id: [score] for doc_id, score in sorted(doc_scores_dict.items(), key=lambda item: item[1], reverse=True)}\n"," most_similar_k_documents = {}\n"," counter = 0\n"," for id, score in most_similar_init_k_documents.items():\n"," try:\n"," if counter == k:\n"," break\n"," most_similar_k_documents[id]=score\n"," counter += 1\n"," except:\n"," break\n","\n"," return most_similar_k_documents "]},{"cell_type":"markdown","metadata":{},"source":["test initial retrieval"]},{"cell_type":"code","execution_count":37,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T20:29:57.136014Z","iopub.status.busy":"2024-01-03T20:29:57.135199Z","iopub.status.idle":"2024-01-03T20:29:57.164581Z","shell.execute_reply":"2024-01-03T20:29:57.163070Z","shell.execute_reply.started":"2024-01-03T20:29:57.135955Z"},"trusted":true},"outputs":[{"data":{"text/plain":["{1399: [11.776525442346998],\n"," 166: [10.781292227767148],\n"," 1071: [10.763945625290491],\n"," 1096: [9.519484116329025],\n"," 523: [7.7636491859711825],\n"," 145: [7.635257862877457],\n"," 374: [7.63466463549917],\n"," 810: [7.1141474648611664],\n"," 778: [6.9228491481657555],\n"," 1054: [6.886386163257785]}"]},"execution_count":37,"metadata":{},"output_type":"execute_result"}],"source":["initial_retrieval_bm25(2, queries_cleaned['text'][1], bm25, 10)"]},{"cell_type":"markdown","metadata":{},"source":["retrieve documents for all queries"]},{"cell_type":"code","execution_count":38,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T19:43:48.581787Z","iopub.status.busy":"2024-01-03T19:43:48.580745Z","iopub.status.idle":"2024-01-03T19:43:53.300653Z","shell.execute_reply":"2024-01-03T19:43:53.299119Z","shell.execute_reply.started":"2024-01-03T19:43:48.581700Z"},"trusted":true},"outputs":[],"source":["initial_retrieval = dict()\n","initial_retrieval_with_scores = dict()\n","for index, row in queries_cleaned.iterrows():\n"," query_id = row[0]\n"," query_text = row[1]\n"," retrieved_documents = initial_retrieval_bm25(query_id, query_text, bm25, 100)\n"," initial_retrieval[query_id] = list(retrieved_documents.keys())\n"," initial_retrieval_with_scores[query_id] = retrieved_documents\n","\n","\n","with open('saved_dictionary.pkl', 'wb') as f:\n"," pickle.dump(dictionary, f)\n"," \n","with open('saved_dictionary.pkl', 'rb') as f:\n"," loaded_dict = pickle.load(f)"]},{"cell_type":"markdown","metadata":{},"source":["<h2> Re-rank with word embeddings and cosine similarity </h2>"]},{"cell_type":"code","execution_count":21,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T19:06:04.035877Z","iopub.status.busy":"2024-01-03T19:06:04.035379Z","iopub.status.idle":"2024-01-03T19:06:07.936834Z","shell.execute_reply":"2024-01-03T19:06:07.935286Z","shell.execute_reply.started":"2024-01-03T19:06:04.035826Z"},"trusted":true},"outputs":[],"source":["# Average vector for each document\n","out_dict_docs = {}\n","n = 1\n","for sen in full_doc:\n"," average_vector = (np.mean(np.array([embeddings(x) for x in data_clean(nltk.word_tokenize(sen)).split()]), axis=0))\n"," d1 = {n: (average_vector)}\n"," out_dict_docs.update(d1)\n"," n +=1 "]},{"cell_type":"code","execution_count":22,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T19:06:07.940189Z","iopub.status.busy":"2024-01-03T19:06:07.939757Z","iopub.status.idle":"2024-01-03T19:06:08.212281Z","shell.execute_reply":"2024-01-03T19:06:08.210836Z","shell.execute_reply.started":"2024-01-03T19:06:07.940125Z"},"trusted":true},"outputs":[],"source":["# Average vector for each document\n","out_dict_queries = {}\n","n = 1\n","for sen in full_query:\n"," average_vector = (np.mean(np.array([embeddings(x) for x in data_clean(nltk.word_tokenize(sen)).split()]), axis=0))\n"," d1 = {n: (average_vector)}\n"," out_dict_queries.update(d1)\n"," n +=1 "]},{"cell_type":"code","execution_count":23,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T19:06:08.214490Z","iopub.status.busy":"2024-01-03T19:06:08.214177Z","iopub.status.idle":"2024-01-03T19:06:08.264818Z","shell.execute_reply":"2024-01-03T19:06:08.263685Z","shell.execute_reply.started":"2024-01-03T19:06:08.214460Z"},"trusted":true},"outputs":[{"data":{"text/html":["<div>\n","<style scoped>\n"," .dataframe tbody tr th:only-of-type {\n"," vertical-align: middle;\n"," }\n","\n"," .dataframe tbody tr th {\n"," vertical-align: top;\n"," }\n","\n"," .dataframe thead th {\n"," text-align: right;\n"," }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n"," <thead>\n"," <tr style=\"text-align: right;\">\n"," <th></th>\n"," <th>id</th>\n"," <th>text</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>1</td>\n"," <td>[0.053237017, 0.07729071, -0.07088695, 0.08220...</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>2</td>\n"," <td>[-0.019184113, -0.014770508, 0.012769063, 0.02...</td>\n"," </tr>\n"," <tr>\n"," <th>2</th>\n"," <td>3</td>\n"," <td>[0.0105957035, -0.0440918, 0.12583008, 0.15257...</td>\n"," </tr>\n"," <tr>\n"," <th>3</th>\n"," <td>4</td>\n"," <td>[0.03293185763888889, 0.05387708875868055, -0....</td>\n"," </tr>\n"," <tr>\n"," <th>4</th>\n"," <td>5</td>\n"," <td>[-0.013476203, 0.03181727, -0.039708756, 0.028...</td>\n"," </tr>\n"," <tr>\n"," <th>...</th>\n"," <td>...</td>\n"," <td>...</td>\n"," </tr>\n"," <tr>\n"," <th>107</th>\n"," <td>108</td>\n"," <td>[-0.013725492689344618, -0.03991911146375868, ...</td>\n"," </tr>\n"," <tr>\n"," <th>108</th>\n"," <td>109</td>\n"," <td>[0.016849247276360262, -0.006489138231209829, ...</td>\n"," </tr>\n"," <tr>\n"," <th>109</th>\n"," <td>110</td>\n"," <td>[0.0070692516508556544, 0.009782482328869047, ...</td>\n"," </tr>\n"," <tr>\n"," <th>110</th>\n"," <td>111</td>\n"," <td>[0.06036170054290254, 0.027709379034527276, 0....</td>\n"," </tr>\n"," <tr>\n"," <th>111</th>\n"," <td>112</td>\n"," <td>[0.022738986545138888, 0.027384086891456886, 0...</td>\n"," </tr>\n"," </tbody>\n","</table>\n","<p>112 rows × 2 columns</p>\n","</div>"],"text/plain":[" id text\n","0 1 [0.053237017, 0.07729071, -0.07088695, 0.08220...\n","1 2 [-0.019184113, -0.014770508, 0.012769063, 0.02...\n","2 3 [0.0105957035, -0.0440918, 0.12583008, 0.15257...\n","3 4 [0.03293185763888889, 0.05387708875868055, -0....\n","4 5 [-0.013476203, 0.03181727, -0.039708756, 0.028...\n",".. ... ...\n","107 108 [-0.013725492689344618, -0.03991911146375868, ...\n","108 109 [0.016849247276360262, -0.006489138231209829, ...\n","109 110 [0.0070692516508556544, 0.009782482328869047, ...\n","110 111 [0.06036170054290254, 0.027709379034527276, 0....\n","111 112 [0.022738986545138888, 0.027384086891456886, 0...\n","\n","[112 rows x 2 columns]"]},"execution_count":23,"metadata":{},"output_type":"execute_result"}],"source":["average_vec_queries = pd.DataFrame(list(out_dict_queries.items()), columns=['id', 'text'])\n","average_vec_queries"]},{"cell_type":"code","execution_count":25,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T19:06:08.323471Z","iopub.status.busy":"2024-01-03T19:06:08.322893Z","iopub.status.idle":"2024-01-03T19:06:08.382600Z","shell.execute_reply":"2024-01-03T19:06:08.381085Z","shell.execute_reply.started":"2024-01-03T19:06:08.323410Z"},"trusted":true},"outputs":[{"data":{"text/html":["<div>\n","<style scoped>\n"," .dataframe tbody tr th:only-of-type {\n"," vertical-align: middle;\n"," }\n","\n"," .dataframe tbody tr th {\n"," vertical-align: top;\n"," }\n","\n"," .dataframe thead th {\n"," text-align: right;\n"," }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n"," <thead>\n"," <tr style=\"text-align: right;\">\n"," <th></th>\n"," <th>id</th>\n"," <th>text</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>1</td>\n"," <td>[0.00830413818359375, 0.00127777099609375, -0....</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>2</td>\n"," <td>[0.05055378758630087, -0.02922665795614553, 0....</td>\n"," </tr>\n"," <tr>\n"," <th>2</th>\n"," <td>3</td>\n"," <td>[0.07028712, 0.0061023016, 0.02422772, -0.0077...</td>\n"," </tr>\n"," <tr>\n"," <th>3</th>\n"," <td>4</td>\n"," <td>[-0.0112152099609375, 0.007620472019001589, 0....</td>\n"," </tr>\n"," <tr>\n"," <th>4</th>\n"," <td>5</td>\n"," <td>[0.028684964743993617, 0.040372458837365593, 0...</td>\n"," </tr>\n"," <tr>\n"," <th>...</th>\n"," <td>...</td>\n"," <td>...</td>\n"," </tr>\n"," <tr>\n"," <th>1455</th>\n"," <td>1456</td>\n"," <td>[0.006353525, 0.023849096, 0.0082289865, 0.072...</td>\n"," </tr>\n"," <tr>\n"," <th>1456</th>\n"," <td>1457</td>\n"," <td>[0.019117838, 0.03813685, 0.039766844, 0.10477...</td>\n"," </tr>\n"," <tr>\n"," <th>1457</th>\n"," <td>1458</td>\n"," <td>[-0.014762384, 0.016318252, -0.0022795142, 0.1...</td>\n"," </tr>\n"," <tr>\n"," <th>1458</th>\n"," <td>1459</td>\n"," <td>[0.087884314, -0.021434652, 0.05227727, 0.1237...</td>\n"," </tr>\n"," <tr>\n"," <th>1459</th>\n"," <td>1460</td>\n"," <td>[-0.045013427734375, 0.01713788067853009, 0.05...</td>\n"," </tr>\n"," </tbody>\n","</table>\n","<p>1460 rows × 2 columns</p>\n","</div>"],"text/plain":[" id text\n","0 1 [0.00830413818359375, 0.00127777099609375, -0....\n","1 2 [0.05055378758630087, -0.02922665795614553, 0....\n","2 3 [0.07028712, 0.0061023016, 0.02422772, -0.0077...\n","3 4 [-0.0112152099609375, 0.007620472019001589, 0....\n","4 5 [0.028684964743993617, 0.040372458837365593, 0...\n","... ... ...\n","1455 1456 [0.006353525, 0.023849096, 0.0082289865, 0.072...\n","1456 1457 [0.019117838, 0.03813685, 0.039766844, 0.10477...\n","1457 1458 [-0.014762384, 0.016318252, -0.0022795142, 0.1...\n","1458 1459 [0.087884314, -0.021434652, 0.05227727, 0.1237...\n","1459 1460 [-0.045013427734375, 0.01713788067853009, 0.05...\n","\n","[1460 rows x 2 columns]"]},"execution_count":25,"metadata":{},"output_type":"execute_result"}],"source":["average_vec_docs = pd.DataFrame(list(out_dict_docs.items()), columns=['id', 'text'])\n","average_vec_docs"]},{"cell_type":"code","execution_count":29,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T19:12:12.457922Z","iopub.status.busy":"2024-01-03T19:12:12.457418Z","iopub.status.idle":"2024-01-03T19:12:12.558607Z","shell.execute_reply":"2024-01-03T19:12:12.556953Z","shell.execute_reply.started":"2024-01-03T19:12:12.457878Z"},"trusted":true},"outputs":[],"source":["def basline_reranker(initial_retrieval, query_id, average_vec_query, k):\n"," retrieved_document_ids = initial_retrieval\n"," retrieved_document_embeddings = average_vec_docs[average_vec_docs['id'].isin(retrieved_document_ids)]\n"," similarities = dict()\n"," for index, row in retrieved_document_embeddings.iterrows():\n"," document_vec = row.values[1]\n"," document_id = row.values[0]\n"," similarity = get_sim(average_vec_query, document_vec)\n"," similarities[document_id]=similarity\n"," similarities = dict(sorted(similarities.items(), key=lambda item: item[1], reverse=True))\n"," \n"," most_similar_k_documents = {}\n"," counter = 0\n"," for id, score in similarities.items():\n"," try:\n"," if counter == k:\n"," break\n"," most_similar_k_documents[id]=score\n"," counter += 1\n"," except:\n"," break\n","\n"," return most_similar_k_documents \n"]},{"cell_type":"markdown","metadata":{},"source":["test for one query"]},{"cell_type":"code","execution_count":73,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T20:31:31.280564Z","iopub.status.busy":"2024-01-03T20:31:31.279777Z","iopub.status.idle":"2024-01-03T20:31:31.332206Z","shell.execute_reply":"2024-01-03T20:31:31.329630Z","shell.execute_reply.started":"2024-01-03T20:31:31.280499Z"},"trusted":true},"outputs":[{"data":{"text/plain":["{429: [0.7788516433824191],\n"," 523: [0.7654931162749747],\n"," 68: [0.7621192073077985],\n"," 381: [0.756619393825531],\n"," 145: [0.7549756090477774],\n"," 421: [0.7507395596452094],\n"," 492: [0.745815041875989],\n"," 1078: [0.7430375974739672],\n"," 1054: [0.7420839667320251],\n"," 202: [0.7335939972659942]}"]},"execution_count":73,"metadata":{},"output_type":"execute_result"}],"source":["basline_reranker(initial_retrieval[2], 2, average_vec_queries['text'][1], 10)"]},{"cell_type":"markdown","metadata":{"execution":{"iopub.execute_input":"2023-12-28T18:34:19.926855Z","iopub.status.busy":"2023-12-28T18:34:19.926354Z","iopub.status.idle":"2023-12-28T18:34:19.963082Z","shell.execute_reply":"2023-12-28T18:34:19.961716Z","shell.execute_reply.started":"2023-12-28T18:34:19.926815Z"}},"source":["re-rank documents for all queries"]},{"cell_type":"code","execution_count":42,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T19:47:00.276459Z","iopub.status.busy":"2024-01-03T19:47:00.275796Z","iopub.status.idle":"2024-01-03T19:47:01.953982Z","shell.execute_reply":"2024-01-03T19:47:01.952820Z","shell.execute_reply.started":"2024-01-03T19:47:00.276401Z"},"trusted":true},"outputs":[],"source":["baseline_reranker_retrieval = dict()\n","for index, row in queries_cleaned.iterrows():\n"," query_id = row[0]\n"," query_text = row[1]\n"," query_embeddings = average_vec_queries['text'][query_id-1]\n"," retrieved_documents = initial_retrieval[query_id]\n"," basline_reranker_documents = basline_reranker(retrieved_documents, query_id, query_embeddings, 50)\n"," baseline_reranker_retrieval[query_id] = list(basline_reranker_documents.keys())\n"," #print(list(retrieved_documents.keys()))"]},{"cell_type":"markdown","metadata":{},"source":["## Initializing BERTopic model"]},{"cell_type":"code","execution_count":14,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T18:57:52.932829Z","iopub.status.busy":"2024-01-03T18:57:52.932037Z","iopub.status.idle":"2024-01-03T18:57:52.940868Z","shell.execute_reply":"2024-01-03T18:57:52.939770Z","shell.execute_reply.started":"2024-01-03T18:57:52.932765Z"},"trusted":true},"outputs":[],"source":["from umap import UMAP\n","\n","model_name = 'sentence-transformers/all-MiniLM-L6-v2'\n","\n","# UMAP is stochastic, so re-produce results you need to set the random_state for umap and pass this umap model to BERTopic:\n","# umap_model = UMAP(n_neighbors=15, n_components=5, min_dist=0.0, metric='cosine', random_state=42)\n","# HOW TO SET UMAP MODEL: topic_model = BERTopic(umap_model=umap_model)\n","\n","topic_model = BERTopic(embedding_model=model_name, ctfidf_model=ClassTfidfTransformer(reduce_frequent_words=True))"]},{"cell_type":"code","execution_count":15,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T18:57:55.031840Z","iopub.status.busy":"2024-01-03T18:57:55.031137Z","iopub.status.idle":"2024-01-03T19:00:05.623076Z","shell.execute_reply":"2024-01-03T19:00:05.621204Z","shell.execute_reply.started":"2024-01-03T18:57:55.031784Z"},"trusted":true},"outputs":[{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"3e55827322034c1d843039710fe4041e","version_major":2,"version_minor":0},"text/plain":["Downloading .gitattributes: 0%| | 0.00/1.18k [00:00<?, ?B/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"5fbf40e5b9b04bc5b73da81c1409943d","version_major":2,"version_minor":0},"text/plain":["Downloading 1_Pooling/config.json: 0%| | 0.00/190 [00:00<?, ?B/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"66be26f4c7f4484e894a4f96a26324ac","version_major":2,"version_minor":0},"text/plain":["Downloading README.md: 0%| | 0.00/10.6k [00:00<?, ?B/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"eac12f55c119482b9e8fc286dae5c709","version_major":2,"version_minor":0},"text/plain":["Downloading config.json: 0%| | 0.00/612 [00:00<?, ?B/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"5b48033f288844fd99b4d0b227b6f96b","version_major":2,"version_minor":0},"text/plain":["Downloading (…)ce_transformers.json: 0%| | 0.00/116 [00:00<?, ?B/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"f96a99058b554cd8ba61b9a70634b956","version_major":2,"version_minor":0},"text/plain":["Downloading data_config.json: 0%| | 0.00/39.3k [00:00<?, ?B/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"21bf0cf5777742a2abe36c0495b0f091","version_major":2,"version_minor":0},"text/plain":["Downloading pytorch_model.bin: 0%| | 0.00/90.9M [00:00<?, ?B/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"d8805775a7574c1cb58542c47333e259","version_major":2,"version_minor":0},"text/plain":["Downloading (…)nce_bert_config.json: 0%| | 0.00/53.0 [00:00<?, ?B/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"120157781f3a42b695d3625eef6a193e","version_major":2,"version_minor":0},"text/plain":["Downloading (…)cial_tokens_map.json: 0%| | 0.00/112 [00:00<?, ?B/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"4f5b544f3c8c4435ae83de863ad60f73","version_major":2,"version_minor":0},"text/plain":["Downloading tokenizer.json: 0%| | 0.00/466k [00:00<?, ?B/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"e3117d7067574f428b7f8ed00672fe4f","version_major":2,"version_minor":0},"text/plain":["Downloading tokenizer_config.json: 0%| | 0.00/350 [00:00<?, ?B/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"615371625e25447fa1377fb1dc4e9fb1","version_major":2,"version_minor":0},"text/plain":["Downloading train_script.py: 0%| | 0.00/13.2k [00:00<?, ?B/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"f9d8faee1348468bad7cc8bd5803f910","version_major":2,"version_minor":0},"text/plain":["Downloading vocab.txt: 0%| | 0.00/232k [00:00<?, ?B/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"73638937ef0f4e18bab90ca6e6c0d3bf","version_major":2,"version_minor":0},"text/plain":["Downloading modules.json: 0%| | 0.00/349 [00:00<?, ?B/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"text/html":["<div>\n","<style scoped>\n"," .dataframe tbody tr th:only-of-type {\n"," vertical-align: middle;\n"," }\n","\n"," .dataframe tbody tr th {\n"," vertical-align: top;\n"," }\n","\n"," .dataframe thead th {\n"," text-align: right;\n"," }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n"," <thead>\n"," <tr style=\"text-align: right;\">\n"," <th></th>\n"," <th>Topic</th>\n"," <th>Count</th>\n"," <th>Name</th>\n"," <th>Representation</th>\n"," <th>Representative_Docs</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>-1</td>\n"," <td>269</td>\n"," <td>-1_document_system_indexing_retrieval</td>\n"," <td>[document, system, indexing, retrieval, on, an...</td>\n"," <td>[PRECIS: a manual of concept analysis and subj...</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>0</td>\n"," <td>384</td>\n"," <td>0_scientific_science_journals_social</td>\n"," <td>[scientific, science, journals, social, scient...</td>\n"," <td>[Recent Growth of the Literature of Biochemist...</td>\n"," </tr>\n"," <tr>\n"," <th>2</th>\n"," <td>1</td>\n"," <td>253</td>\n"," <td>1_library_libraries_university_academic</td>\n"," <td>[library, libraries, university, academic, pub...</td>\n"," <td>[Undergraduate Library The development of the ...</td>\n"," </tr>\n"," <tr>\n"," <th>3</th>\n"," <td>2</td>\n"," <td>90</td>\n"," <td>2_chemical_compounds_notation_search</td>\n"," <td>[chemical, compounds, notation, search, titles...</td>\n"," <td>[Experiences of IIT Research Institute in Oper...</td>\n"," </tr>\n"," <tr>\n"," <th>4</th>\n"," <td>3</td>\n"," <td>54</td>\n"," <td>3_automatic_indexing_classification_document</td>\n"," <td>[automatic, indexing, classification, document...</td>\n"," <td>[What Makes An Automatic Keyword Classificatio...</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>"],"text/plain":[" Topic Count Name \\\n","0 -1 269 -1_document_system_indexing_retrieval \n","1 0 384 0_scientific_science_journals_social \n","2 1 253 1_library_libraries_university_academic \n","3 2 90 2_chemical_compounds_notation_search \n","4 3 54 3_automatic_indexing_classification_document \n","\n"," Representation \\\n","0 [document, system, indexing, retrieval, on, an... \n","1 [scientific, science, journals, social, scient... \n","2 [library, libraries, university, academic, pub... \n","3 [chemical, compounds, notation, search, titles... \n","4 [automatic, indexing, classification, document... \n","\n"," Representative_Docs \n","0 [PRECIS: a manual of concept analysis and subj... \n","1 [Recent Growth of the Literature of Biochemist... \n","2 [Undergraduate Library The development of the ... \n","3 [Experiences of IIT Research Institute in Oper... \n","4 [What Makes An Automatic Keyword Classificatio... "]},"execution_count":15,"metadata":{},"output_type":"execute_result"}],"source":["# training bert model\n","docs_for_bert = docs[\"text\"]\n","topic_model.fit(docs_for_bert)"]},{"cell_type":"code","execution_count":77,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T20:37:47.548151Z","iopub.status.busy":"2024-01-03T20:37:47.547570Z","iopub.status.idle":"2024-01-03T20:37:47.601583Z","shell.execute_reply":"2024-01-03T20:37:47.599861Z","shell.execute_reply.started":"2024-01-03T20:37:47.548107Z"},"trusted":true},"outputs":[{"data":{"text/html":["<div>\n","<style scoped>\n"," .dataframe tbody tr th:only-of-type {\n"," vertical-align: middle;\n"," }\n","\n"," .dataframe tbody tr th {\n"," vertical-align: top;\n"," }\n","\n"," .dataframe thead th {\n"," text-align: right;\n"," }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n"," <thead>\n"," <tr style=\"text-align: right;\">\n"," <th></th>\n"," <th>Topic</th>\n"," <th>Count</th>\n"," <th>Name</th>\n"," <th>Representation</th>\n"," <th>Representative_Docs</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>-1</td>\n"," <td>269</td>\n"," <td>-1_document_system_indexing_retrieval</td>\n"," <td>[document, system, indexing, retrieval, on, an...</td>\n"," <td>[PRECIS: a manual of concept analysis and subj...</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>0</td>\n"," <td>384</td>\n"," <td>0_scientific_science_journals_social</td>\n"," <td>[scientific, science, journals, social, scient...</td>\n"," <td>[Recent Growth of the Literature of Biochemist...</td>\n"," </tr>\n"," <tr>\n"," <th>2</th>\n"," <td>1</td>\n"," <td>253</td>\n"," <td>1_library_libraries_university_academic</td>\n"," <td>[library, libraries, university, academic, pub...</td>\n"," <td>[Undergraduate Library The development of the ...</td>\n"," </tr>\n"," <tr>\n"," <th>3</th>\n"," <td>2</td>\n"," <td>90</td>\n"," <td>2_chemical_compounds_notation_search</td>\n"," <td>[chemical, compounds, notation, search, titles...</td>\n"," <td>[Experiences of IIT Research Institute in Oper...</td>\n"," </tr>\n"," <tr>\n"," <th>4</th>\n"," <td>3</td>\n"," <td>54</td>\n"," <td>3_automatic_indexing_classification_document</td>\n"," <td>[automatic, indexing, classification, document...</td>\n"," <td>[What Makes An Automatic Keyword Classificatio...</td>\n"," </tr>\n"," <tr>\n"," <th>5</th>\n"," <td>4</td>\n"," <td>50</td>\n"," <td>4_bases_data_bibliographic_line</td>\n"," <td>[bases, data, bibliographic, line, readable, s...</td>\n"," <td>[Survey of Commercially Available Computer-Rea...</td>\n"," </tr>\n"," <tr>\n"," <th>6</th>\n"," <td>5</td>\n"," <td>46</td>\n"," <td>5_relevance_retrieval_answer_relevant</td>\n"," <td>[relevance, retrieval, answer, relevant, docum...</td>\n"," <td>[On Relevance, Probabilistic Indexing and Info...</td>\n"," </tr>\n"," <tr>\n"," <th>7</th>\n"," <td>6</td>\n"," <td>38</td>\n"," <td>6_catalog_catalogs_cataloging_card</td>\n"," <td>[catalog, catalogs, cataloging, card, catalogu...</td>\n"," <td>[The Potential Usefulness of Catalog Access Po...</td>\n"," </tr>\n"," <tr>\n"," <th>8</th>\n"," <td>7</td>\n"," <td>28</td>\n"," <td>7_classification_decimal_udc_dewey</td>\n"," <td>[classification, decimal, udc, dewey, schemes,...</td>\n"," <td>[Progress in Documentation Thirty years or mor...</td>\n"," </tr>\n"," <tr>\n"," <th>9</th>\n"," <td>8</td>\n"," <td>27</td>\n"," <td>8_language_linguistics_linguistic_semantic</td>\n"," <td>[language, linguistics, linguistic, semantic, ...</td>\n"," <td>[Functional Approach The present book sums up ...</td>\n"," </tr>\n"," <tr>\n"," <th>10</th>\n"," <td>9</td>\n"," <td>25</td>\n"," <td>9_medical_health_hospital_manpower</td>\n"," <td>[medical, health, hospital, manpower, hospital...</td>\n"," <td>[Library Practice in Hospitals According to a ...</td>\n"," </tr>\n"," <tr>\n"," <th>11</th>\n"," <td>10</td>\n"," <td>22</td>\n"," <td>10_automation_library_processing_telefacsimile</td>\n"," <td>[automation, library, processing, telefacsimil...</td>\n"," <td>[HDB of Data Processing for Libraries The four...</td>\n"," </tr>\n"," <tr>\n"," <th>12</th>\n"," <td>11</td>\n"," <td>21</td>\n"," <td>11_retrieval_user_systems_isrs</td>\n"," <td>[retrieval, user, systems, isrs, system, dialo...</td>\n"," <td>[Information Retrieval Systems This book is co...</td>\n"," </tr>\n"," <tr>\n"," <th>13</th>\n"," <td>12</td>\n"," <td>19</td>\n"," <td>12_thesaurus_thesauri_vocabularies_vocabulary</td>\n"," <td>[thesaurus, thesauri, vocabularies, vocabulary...</td>\n"," <td>[Theoretical Foundations of Thesaurus-Construc...</td>\n"," </tr>\n"," <tr>\n"," <th>14</th>\n"," <td>13</td>\n"," <td>18</td>\n"," <td>13_evaluation_cost_costs_systems</td>\n"," <td>[evaluation, cost, costs, systems, scale, serv...</td>\n"," <td>[Design and Evaluation of Information Systems ...</td>\n"," </tr>\n"," <tr>\n"," <th>15</th>\n"," <td>14</td>\n"," <td>18</td>\n"," <td>14_fuzzy_classification_sets_membership</td>\n"," <td>[fuzzy, classification, sets, membership, hedg...</td>\n"," <td>[Prospects for a New General Classification In...</td>\n"," </tr>\n"," <tr>\n"," <th>16</th>\n"," <td>15</td>\n"," <td>17</td>\n"," <td>15_serials_isbd_serial_international</td>\n"," <td>[serials, isbd, serial, international, rules, ...</td>\n"," <td>[No Special Rules for Entry of Serials One of ...</td>\n"," </tr>\n"," <tr>\n"," <th>17</th>\n"," <td>16</td>\n"," <td>16</td>\n"," <td>16_compression_coding_length_grams</td>\n"," <td>[compression, coding, length, grams, error, na...</td>\n"," <td>[An Information-Theoretic Approach to Text Sea...</td>\n"," </tr>\n"," <tr>\n"," <th>18</th>\n"," <td>17</td>\n"," <td>15</td>\n"," <td>17_microfiche_microforms_microform_microfilm</td>\n"," <td>[microfiche, microforms, microform, microfilm,...</td>\n"," <td>[The Microform Revolution Librarians have trie...</td>\n"," </tr>\n"," <tr>\n"," <th>19</th>\n"," <td>18</td>\n"," <td>15</td>\n"," <td>18_network_networks_cable_television</td>\n"," <td>[network, networks, cable, television, communi...</td>\n"," <td>[The National Biomedical Communications Networ...</td>\n"," </tr>\n"," <tr>\n"," <th>20</th>\n"," <td>19</td>\n"," <td>14</td>\n"," <td>19_medlars_medline_twx_medicus</td>\n"," <td>[medlars, medline, twx, medicus, medicine, nlm...</td>\n"," <td>[MEDLARS: A Summary Review and Evaluation of T...</td>\n"," </tr>\n"," <tr>\n"," <th>21</th>\n"," <td>20</td>\n"," <td>11</td>\n"," <td>20_centers_services_systems_micrographic</td>\n"," <td>[centers, services, systems, micrographic, enc...</td>\n"," <td>[The Annual Review of Information Science and ...</td>\n"," </tr>\n"," <tr>\n"," <th>22</th>\n"," <td>21</td>\n"," <td>10</td>\n"," <td>21_marc_records_readable_pilot</td>\n"," <td>[marc, records, readable, pilot, cobol, machin...</td>\n"," <td>[The Marc II Format: A ...</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>"],"text/plain":[" Topic Count Name \\\n","0 -1 269 -1_document_system_indexing_retrieval \n","1 0 384 0_scientific_science_journals_social \n","2 1 253 1_library_libraries_university_academic \n","3 2 90 2_chemical_compounds_notation_search \n","4 3 54 3_automatic_indexing_classification_document \n","5 4 50 4_bases_data_bibliographic_line \n","6 5 46 5_relevance_retrieval_answer_relevant \n","7 6 38 6_catalog_catalogs_cataloging_card \n","8 7 28 7_classification_decimal_udc_dewey \n","9 8 27 8_language_linguistics_linguistic_semantic \n","10 9 25 9_medical_health_hospital_manpower \n","11 10 22 10_automation_library_processing_telefacsimile \n","12 11 21 11_retrieval_user_systems_isrs \n","13 12 19 12_thesaurus_thesauri_vocabularies_vocabulary \n","14 13 18 13_evaluation_cost_costs_systems \n","15 14 18 14_fuzzy_classification_sets_membership \n","16 15 17 15_serials_isbd_serial_international \n","17 16 16 16_compression_coding_length_grams \n","18 17 15 17_microfiche_microforms_microform_microfilm \n","19 18 15 18_network_networks_cable_television \n","20 19 14 19_medlars_medline_twx_medicus \n","21 20 11 20_centers_services_systems_micrographic \n","22 21 10 21_marc_records_readable_pilot \n","\n"," Representation \\\n","0 [document, system, indexing, retrieval, on, an... \n","1 [scientific, science, journals, social, scient... \n","2 [library, libraries, university, academic, pub... \n","3 [chemical, compounds, notation, search, titles... \n","4 [automatic, indexing, classification, document... \n","5 [bases, data, bibliographic, line, readable, s... \n","6 [relevance, retrieval, answer, relevant, docum... \n","7 [catalog, catalogs, cataloging, card, catalogu... \n","8 [classification, decimal, udc, dewey, schemes,... \n","9 [language, linguistics, linguistic, semantic, ... \n","10 [medical, health, hospital, manpower, hospital... \n","11 [automation, library, processing, telefacsimil... \n","12 [retrieval, user, systems, isrs, system, dialo... \n","13 [thesaurus, thesauri, vocabularies, vocabulary... \n","14 [evaluation, cost, costs, systems, scale, serv... \n","15 [fuzzy, classification, sets, membership, hedg... \n","16 [serials, isbd, serial, international, rules, ... \n","17 [compression, coding, length, grams, error, na... \n","18 [microfiche, microforms, microform, microfilm,... \n","19 [network, networks, cable, television, communi... \n","20 [medlars, medline, twx, medicus, medicine, nlm... \n","21 [centers, services, systems, micrographic, enc... \n","22 [marc, records, readable, pilot, cobol, machin... \n","\n"," Representative_Docs \n","0 [PRECIS: a manual of concept analysis and subj... \n","1 [Recent Growth of the Literature of Biochemist... \n","2 [Undergraduate Library The development of the ... \n","3 [Experiences of IIT Research Institute in Oper... \n","4 [What Makes An Automatic Keyword Classificatio... \n","5 [Survey of Commercially Available Computer-Rea... \n","6 [On Relevance, Probabilistic Indexing and Info... \n","7 [The Potential Usefulness of Catalog Access Po... \n","8 [Progress in Documentation Thirty years or mor... \n","9 [Functional Approach The present book sums up ... \n","10 [Library Practice in Hospitals According to a ... \n","11 [HDB of Data Processing for Libraries The four... \n","12 [Information Retrieval Systems This book is co... \n","13 [Theoretical Foundations of Thesaurus-Construc... \n","14 [Design and Evaluation of Information Systems ... \n","15 [Prospects for a New General Classification In... \n","16 [No Special Rules for Entry of Serials One of ... \n","17 [An Information-Theoretic Approach to Text Sea... \n","18 [The Microform Revolution Librarians have trie... \n","19 [The National Biomedical Communications Networ... \n","20 [MEDLARS: A Summary Review and Evaluation of T... \n","21 [The Annual Review of Information Science and ... \n","22 [The Marc II Format: A ... "]},"execution_count":77,"metadata":{},"output_type":"execute_result"}],"source":["topic_model.get_topic_info()"]},{"cell_type":"code","execution_count":18,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T19:00:05.785848Z","iopub.status.busy":"2024-01-03T19:00:05.785233Z","iopub.status.idle":"2024-01-03T19:00:05.894032Z","shell.execute_reply":"2024-01-03T19:00:05.892648Z","shell.execute_reply.started":"2024-01-03T19:00:05.785783Z"},"trusted":true},"outputs":[{"data":{"text/html":["<div>\n","<style scoped>\n"," .dataframe tbody tr th:only-of-type {\n"," vertical-align: middle;\n"," }\n","\n"," .dataframe tbody tr th {\n"," vertical-align: top;\n"," }\n","\n"," .dataframe thead th {\n"," text-align: right;\n"," }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n"," <thead>\n"," <tr style=\"text-align: right;\">\n"," <th></th>\n"," <th>Document</th>\n"," <th>Topic</th>\n"," <th>Name</th>\n"," <th>Representation</th>\n"," <th>Representative_Docs</th>\n"," <th>Top_n_words</th>\n"," <th>Probability</th>\n"," <th>Representative_document</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>18 Editions of the Dewey Decimal Classificatio...</td>\n"," <td>7</td>\n"," <td>7_classification_decimal_udc_dewey</td>\n"," <td>[classification, decimal, udc, dewey, schemes,...</td>\n"," <td>[Progress in Documentation Thirty years or mor...</td>\n"," <td>classification - decimal - udc - dewey - schem...</td>\n"," <td>0.781595</td>\n"," <td>False</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>Use Made of Technical Libraries This report is...</td>\n"," <td>1</td>\n"," <td>1_library_libraries_university_academic</td>\n"," <td>[library, libraries, university, academic, pub...</td>\n"," <td>[Undergraduate Library The development of the ...</td>\n"," <td>library - libraries - university - academic - ...</td>\n"," <td>0.737408</td>\n"," <td>False</td>\n"," </tr>\n"," <tr>\n"," <th>2</th>\n"," <td>Two Kinds of Power An Essay on Bibliographic C...</td>\n"," <td>-1</td>\n"," <td>-1_document_system_indexing_retrieval</td>\n"," <td>[document, system, indexing, retrieval, on, an...</td>\n"," <td>[PRECIS: a manual of concept analysis and subj...</td>\n"," <td>document - system - indexing - retrieval - on ...</td>\n"," <td>0.000000</td>\n"," <td>False</td>\n"," </tr>\n"," <tr>\n"," <th>3</th>\n"," <td>Systems Analysis of a University Library; fina...</td>\n"," <td>1</td>\n"," <td>1_library_libraries_university_academic</td>\n"," <td>[library, libraries, university, academic, pub...</td>\n"," <td>[Undergraduate Library The development of the ...</td>\n"," <td>library - libraries - university - academic - ...</td>\n"," <td>0.993856</td>\n"," <td>False</td>\n"," </tr>\n"," <tr>\n"," <th>4</th>\n"," <td>A Library Management Game: a report on a resea...</td>\n"," <td>-1</td>\n"," <td>-1_document_system_indexing_retrieval</td>\n"," <td>[document, system, indexing, retrieval, on, an...</td>\n"," <td>[PRECIS: a manual of concept analysis and subj...</td>\n"," <td>document - system - indexing - retrieval - on ...</td>\n"," <td>0.000000</td>\n"," <td>False</td>\n"," </tr>\n"," <tr>\n"," <th>...</th>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," </tr>\n"," <tr>\n"," <th>1455</th>\n"," <td>World Dynamics Over the last several decades i...</td>\n"," <td>0</td>\n"," <td>0_scientific_science_journals_social</td>\n"," <td>[scientific, science, journals, social, scient...</td>\n"," <td>[Recent Growth of the Literature of Biochemist...</td>\n"," <td>scientific - science - journals - social - sci...</td>\n"," <td>1.000000</td>\n"," <td>False</td>\n"," </tr>\n"," <tr>\n"," <th>1456</th>\n"," <td>World Trends in Library Education One of the m...</td>\n"," <td>1</td>\n"," <td>1_library_libraries_university_academic</td>\n"," <td>[library, libraries, university, academic, pub...</td>\n"," <td>[Undergraduate Library The development of the ...</td>\n"," <td>library - libraries - university - academic - ...</td>\n"," <td>1.000000</td>\n"," <td>False</td>\n"," </tr>\n"," <tr>\n"," <th>1457</th>\n"," <td>Legal Restrictions on Exploitation of the Pate...</td>\n"," <td>-1</td>\n"," <td>-1_document_system_indexing_retrieval</td>\n"," <td>[document, system, indexing, retrieval, on, an...</td>\n"," <td>[PRECIS: a manual of concept analysis and subj...</td>\n"," <td>document - system - indexing - retrieval - on ...</td>\n"," <td>0.000000</td>\n"," <td>False</td>\n"," </tr>\n"," <tr>\n"," <th>1458</th>\n"," <td>Language and Thought This book considers the b...</td>\n"," <td>8</td>\n"," <td>8_language_linguistics_linguistic_semantic</td>\n"," <td>[language, linguistics, linguistic, semantic, ...</td>\n"," <td>[Functional Approach The present book sums up ...</td>\n"," <td>language - linguistics - linguistic - semantic...</td>\n"," <td>1.000000</td>\n"," <td>False</td>\n"," </tr>\n"," <tr>\n"," <th>1459</th>\n"," <td>Modern Integral Information Systems for Chemis...</td>\n"," <td>2</td>\n"," <td>2_chemical_compounds_notation_search</td>\n"," <td>[chemical, compounds, notation, search, titles...</td>\n"," <td>[Experiences of IIT Research Institute in Oper...</td>\n"," <td>chemical - compounds - notation - search - tit...</td>\n"," <td>0.805337</td>\n"," <td>False</td>\n"," </tr>\n"," </tbody>\n","</table>\n","<p>1460 rows × 8 columns</p>\n","</div>"],"text/plain":[" Document Topic \\\n","0 18 Editions of the Dewey Decimal Classificatio... 7 \n","1 Use Made of Technical Libraries This report is... 1 \n","2 Two Kinds of Power An Essay on Bibliographic C... -1 \n","3 Systems Analysis of a University Library; fina... 1 \n","4 A Library Management Game: a report on a resea... -1 \n","... ... ... \n","1455 World Dynamics Over the last several decades i... 0 \n","1456 World Trends in Library Education One of the m... 1 \n","1457 Legal Restrictions on Exploitation of the Pate... -1 \n","1458 Language and Thought This book considers the b... 8 \n","1459 Modern Integral Information Systems for Chemis... 2 \n","\n"," Name \\\n","0 7_classification_decimal_udc_dewey \n","1 1_library_libraries_university_academic \n","2 -1_document_system_indexing_retrieval \n","3 1_library_libraries_university_academic \n","4 -1_document_system_indexing_retrieval \n","... ... \n","1455 0_scientific_science_journals_social \n","1456 1_library_libraries_university_academic \n","1457 -1_document_system_indexing_retrieval \n","1458 8_language_linguistics_linguistic_semantic \n","1459 2_chemical_compounds_notation_search \n","\n"," Representation \\\n","0 [classification, decimal, udc, dewey, schemes,... \n","1 [library, libraries, university, academic, pub... \n","2 [document, system, indexing, retrieval, on, an... \n","3 [library, libraries, university, academic, pub... \n","4 [document, system, indexing, retrieval, on, an... \n","... ... \n","1455 [scientific, science, journals, social, scient... \n","1456 [library, libraries, university, academic, pub... \n","1457 [document, system, indexing, retrieval, on, an... \n","1458 [language, linguistics, linguistic, semantic, ... \n","1459 [chemical, compounds, notation, search, titles... \n","\n"," Representative_Docs \\\n","0 [Progress in Documentation Thirty years or mor... \n","1 [Undergraduate Library The development of the ... \n","2 [PRECIS: a manual of concept analysis and subj... \n","3 [Undergraduate Library The development of the ... \n","4 [PRECIS: a manual of concept analysis and subj... \n","... ... \n","1455 [Recent Growth of the Literature of Biochemist... \n","1456 [Undergraduate Library The development of the ... \n","1457 [PRECIS: a manual of concept analysis and subj... \n","1458 [Functional Approach The present book sums up ... \n","1459 [Experiences of IIT Research Institute in Oper... \n","\n"," Top_n_words Probability \\\n","0 classification - decimal - udc - dewey - schem... 0.781595 \n","1 library - libraries - university - academic - ... 0.737408 \n","2 document - system - indexing - retrieval - on ... 0.000000 \n","3 library - libraries - university - academic - ... 0.993856 \n","4 document - system - indexing - retrieval - on ... 0.000000 \n","... ... ... \n","1455 scientific - science - journals - social - sci... 1.000000 \n","1456 library - libraries - university - academic - ... 1.000000 \n","1457 document - system - indexing - retrieval - on ... 0.000000 \n","1458 language - linguistics - linguistic - semantic... 1.000000 \n","1459 chemical - compounds - notation - search - tit... 0.805337 \n","\n"," Representative_document \n","0 False \n","1 False \n","2 False \n","3 False \n","4 False \n","... ... \n","1455 False \n","1456 False \n","1457 False \n","1458 False \n","1459 False \n","\n","[1460 rows x 8 columns]"]},"execution_count":18,"metadata":{},"output_type":"execute_result"}],"source":["doc_info = topic_model.get_document_info(docs[\"text\"])\n","doc_info"]},{"cell_type":"markdown","metadata":{},"source":["<h2> Re-ranking with bertopic</h2>"]},{"cell_type":"code","execution_count":78,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T20:38:18.212878Z","iopub.status.busy":"2024-01-03T20:38:18.212296Z","iopub.status.idle":"2024-01-03T20:42:37.435531Z","shell.execute_reply":"2024-01-03T20:42:37.433913Z","shell.execute_reply.started":"2024-01-03T20:38:18.212829Z"},"trusted":true},"outputs":[],"source":["# transform queries to topics\n","query_topics = {}\n","for index, row in queries_cleaned.iterrows():\n"," query_id = row[0]\n"," qq = queries.loc[queries['id'] == query_id]\n"," topic, prob = topic_model.transform(qq[\"text\"].values.tolist())\n"," query_topics[query_id] = topic"]},{"cell_type":"code","execution_count":79,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T20:42:37.438390Z","iopub.status.busy":"2024-01-03T20:42:37.437898Z","iopub.status.idle":"2024-01-03T20:42:37.454106Z","shell.execute_reply":"2024-01-03T20:42:37.452690Z","shell.execute_reply.started":"2024-01-03T20:42:37.438329Z"},"trusted":true},"outputs":[{"data":{"text/plain":["{1: [-1],\n"," 2: [4],\n"," 3: [0],\n"," 4: [-1],\n"," 5: [11],\n"," 6: [0],\n"," 7: [-1],\n"," 8: [-1],\n"," 9: [-1],\n"," 10: [-1],\n"," 11: [0],\n"," 12: [0],\n"," 13: [-1],\n"," 14: [19],\n"," 15: [4],\n"," 16: [11],\n"," 17: [-1],\n"," 18: [2],\n"," 19: [-1],\n"," 20: [-1],\n"," 21: [0],\n"," 22: [19],\n"," 23: [10],\n"," 24: [0],\n"," 25: [-1],\n"," 26: [13],\n"," 27: [3],\n"," 28: [2],\n"," 29: [3],\n"," 30: [0],\n"," 31: [0],\n"," 32: [3],\n"," 33: [11],\n"," 34: [-1],\n"," 35: [20],\n"," 37: [-1],\n"," 39: [11],\n"," 41: [-1],\n"," 42: [11],\n"," 43: [11],\n"," 44: [0],\n"," 45: [10],\n"," 46: [10],\n"," 49: [11],\n"," 50: [-1],\n"," 52: [19],\n"," 54: [10],\n"," 55: [19],\n"," 56: [-1],\n"," 57: [-1],\n"," 58: [18],\n"," 61: [5],\n"," 62: [-1],\n"," 65: [-1],\n"," 66: [18],\n"," 67: [-1],\n"," 69: [12],\n"," 71: [-1],\n"," 76: [-1],\n"," 79: [-1],\n"," 81: [12],\n"," 82: [-1],\n"," 84: [5],\n"," 90: [0],\n"," 92: [4],\n"," 95: [5],\n"," 96: [5],\n"," 97: [5],\n"," 98: [11],\n"," 99: [11],\n"," 100: [-1],\n"," 101: [-1],\n"," 102: [3],\n"," 104: [-1],\n"," 109: [0],\n"," 111: [-1]}"]},"execution_count":79,"metadata":{},"output_type":"execute_result"}],"source":["query_topics"]},{"cell_type":"code","execution_count":58,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T20:12:01.716284Z","iopub.status.busy":"2024-01-03T20:12:01.715582Z","iopub.status.idle":"2024-01-03T20:12:01.726826Z","shell.execute_reply":"2024-01-03T20:12:01.725962Z","shell.execute_reply.started":"2024-01-03T20:12:01.716244Z"},"trusted":true},"outputs":[],"source":["def bertopic_reranker(initial_retrieval, query_id, k, lam=0.2):\n"," #print(topic_model.get_topic_info())\n"," qq = queries.loc[queries['id'] == query_id]\n"," #print(\"query: \", qq[\"text\"].values.tolist())\n"," #topic, prob = topic_model.transform(qq[\"text\"].values.tolist())\n"," topic = query_topics[query_id]\n"," topic = topic[0]\n"," most_similar_init_k_documents = {}\n"," #print(\"topic: \", topic)\n"," i = 0\n"," for id, score in initial_retrieval.items():\n"," doc_topic = doc_info.iloc[id-1][\"Topic\"]\n"," #print(\"d\", id, doc_topic)\n"," if doc_topic == topic and topic != -1:\n"," #print(\"same topic, increase score\", score)\n"," most_similar_init_k_documents[id]=[score[0] * lam]\n"," else:\n"," most_similar_init_k_documents[id]=[score[0]]\n"," i += 1\n"," most_similar_init_k_documents = dict(sorted(most_similar_init_k_documents.items(), key=lambda item: item[1], reverse=True))\n","\n"," most_similar_k_documents = {}\n"," counter = 0\n"," for id, score in most_similar_init_k_documents.items():\n"," try:\n"," if counter == k:\n"," break\n"," most_similar_k_documents[id]=score\n"," counter += 1\n"," except:\n"," break\n","\n"," return most_similar_k_documents "]},{"cell_type":"markdown","metadata":{},"source":["test for one document"]},{"cell_type":"code","execution_count":80,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T20:52:22.053797Z","iopub.status.busy":"2024-01-03T20:52:22.053248Z","iopub.status.idle":"2024-01-03T20:52:22.084399Z","shell.execute_reply":"2024-01-03T20:52:22.082985Z","shell.execute_reply.started":"2024-01-03T20:52:22.053751Z"},"trusted":true},"outputs":[{"data":{"text/plain":["{145: [15.270515725754914],\n"," 1399: [11.776525442346998],\n"," 597: [11.654505315568294],\n"," 166: [10.781292227767148],\n"," 1071: [10.763945625290491],\n"," 546: [10.405693755713079],\n"," 626: [9.676309340155163],\n"," 1096: [9.519484116329025],\n"," 728: [8.859895015786782],\n"," 1197: [8.78125771012449]}"]},"execution_count":80,"metadata":{},"output_type":"execute_result"}],"source":["bertopic_reranker(initial_retrieval_with_scores[2], 2, 10, lam=2.0)"]},{"cell_type":"markdown","metadata":{},"source":["re-rank documents for all queries"]},{"cell_type":"code","execution_count":112,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T21:23:21.816142Z","iopub.status.busy":"2024-01-03T21:23:21.815690Z","iopub.status.idle":"2024-01-03T21:23:29.221491Z","shell.execute_reply":"2024-01-03T21:23:29.220201Z","shell.execute_reply.started":"2024-01-03T21:23:21.816101Z"},"trusted":true},"outputs":[{"name":"stdout","output_type":"stream","text":["Currently retrieving for lam: 1.2\n","Currently retrieving for lam: 1.5\n","Currently retrieving for lam: 2\n","Currently retrieving for lam: 2.5\n","Currently retrieving for lam: 3\n"]}],"source":["lam_values = [1.2, 1.5, 2, 2.5, 3]\n","\n","results_for_different_lams = dict()\n","for lam_value in lam_values:\n"," print(f'Currently retrieving for lam: {lam_value}')\n"," bertopic_reranker_retrieval = dict()\n"," for index, row in queries_cleaned.iterrows():\n"," query_id = row[0]\n"," query_text = row[1]\n"," retrieved_documents = initial_retrieval_with_scores[query_id]\n"," bertopic_reranker_documents = bertopic_reranker(retrieved_documents, query_id, 50, lam=lam_value)\n"," bertopic_reranker_retrieval[query_id] = list(bertopic_reranker_documents.keys())\n"," results_for_different_lams[lam_value] = bertopic_reranker_retrieval"]},{"cell_type":"code","execution_count":113,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T21:23:35.507029Z","iopub.status.busy":"2024-01-03T21:23:35.506573Z","iopub.status.idle":"2024-01-03T21:23:35.833096Z","shell.execute_reply":"2024-01-03T21:23:35.831944Z","shell.execute_reply.started":"2024-01-03T21:23:35.506994Z"},"trusted":true},"outputs":[],"source":["# Initialize an empty dictionary to store the results\n","result_dict = {}\n","\n","# Iterate through the DataFrame and populate the dictionary\n","for index, row in rels.iterrows():\n"," query_id = row['queryID']\n"," doc_id = row['docID']\n"," # If the query ID is not already in the dictionary, add it with an empty list\n"," if query_id not in result_dict:\n"," result_dict[query_id] = []\n"," # Append the document ID to the list associated with the query ID\n"," result_dict[query_id].append(doc_id)\n"]},{"cell_type":"code","execution_count":null,"metadata":{"execution":{"iopub.execute_input":"2023-12-31T12:12:26.418730Z","iopub.status.busy":"2023-12-31T12:12:26.418435Z","iopub.status.idle":"2023-12-31T12:12:26.422457Z","shell.execute_reply":"2023-12-31T12:12:26.421575Z","shell.execute_reply.started":"2023-12-31T12:12:26.418702Z"},"trusted":true},"outputs":[],"source":["# ! When we sort here, we already use the information of the similarity ordering, so we cannot calculate measures @ k!\n","\n","# Sort each list in the dictionary\n","#sorted_dict = {key: sorted(value) for key, value in final_retrieval.items()}\n","#sorted_dict_bert = {key: sorted(value) for key, value in final_retrieval_bert.items()}\n","\n","# Print the sorted dictionary\n","#sorted_dict"]},{"cell_type":"markdown","metadata":{},"source":["<h2> Evalution </h2>"]},{"cell_type":"code","execution_count":115,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T21:23:42.614233Z","iopub.status.busy":"2024-01-03T21:23:42.613727Z","iopub.status.idle":"2024-01-03T21:23:42.637263Z","shell.execute_reply":"2024-01-03T21:23:42.635902Z","shell.execute_reply.started":"2024-01-03T21:23:42.614194Z"},"trusted":true},"outputs":[],"source":["def evaluate(predictions, k):\n"," f_1 = 0\n"," precision = 0\n"," recall = 0\n"," number_queries_evaluated = 0\n"," for prediction in predictions.items():\n"," q_id = prediction[0]\n"," predicted_documents_k_relevant = prediction[1]\n","\n"," ground_truth = rels\n"," ground_truth_k_relevant = ground_truth[ground_truth[\"queryID\"] == q_id].iloc[:k]\n"," ground_truth_k_relevant = ground_truth_k_relevant['docID'].to_list()\n","\n"," false_positives = 0\n"," for predicted in predicted_documents_k_relevant:\n"," if predicted not in ground_truth_k_relevant:\n"," false_positives += 1\n","\n"," true_positives = 0 \n"," false_negatives = 0 \n"," for truth in ground_truth_k_relevant:\n"," if truth in predicted_documents_k_relevant:\n"," true_positives += 1\n"," if truth not in predicted_documents_k_relevant:\n"," false_negatives += 1\n"," try:\n"," query_precision = true_positives/(true_positives+false_positives)\n"," query_recall = true_positives/(true_positives+false_negatives)\n"," except:\n"," query_precision = 0\n"," query_recall = 0\n"," if query_precision > 0 or query_recall > 0:\n"," #print(f'precision: {query_precision} recall: {query_recall}')\n"," query_f_1 = (2*query_precision*query_recall)/(query_precision+query_recall)\n"," f_1 += query_f_1\n"," precision += query_precision\n"," recall += query_recall\n"," number_queries_evaluated += 1\n"," else:\n"," f_1 += 0\n"," precision += 0\n"," recall += query_recall\n"," number_queries_evaluated += 1\n"," \n"," f_1 /= number_queries_evaluated\n"," precision /= number_queries_evaluated\n"," recall /= number_queries_evaluated\n"," \n"," ndcg = 0\n"," number_queries_evaluated = 0\n"," for prediction in predictions.items():\n"," q_id = prediction[0]\n"," relevant_items = rels[rels[\"queryID\"] == q_id]\n"," relevant_items = relevant_items[\"docID\"].to_list()\n"," documents = prediction[1][:k] \n"," i = 1\n"," dcg_document = 0\n"," idcg_document = 0\n"," for document_id in documents:\n"," idcg_i = (1/np.log2(i+1))\n"," idcg_document += idcg_i\n"," if document_id in relevant_items:\n"," dcg_i = (1/np.log2(i+1)) \n"," dcg_document += dcg_i\n"," i +=1\n"," ndgc_document = (dcg_document/idcg_document) if idcg_document != 0 else 0\n"," ndcg += ndgc_document\n"," number_queries_evaluated += 1\n"," \n"," ndcg/=number_queries_evaluated\n"," \n"," return {'f_1':f_1, 'precision':precision, 'recall':recall, 'nDCG':ndcg}"]},{"cell_type":"markdown","metadata":{},"source":["K's to evaluate for"]},{"cell_type":"code","execution_count":116,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T21:23:47.877918Z","iopub.status.busy":"2024-01-03T21:23:47.877437Z","iopub.status.idle":"2024-01-03T21:23:47.883332Z","shell.execute_reply":"2024-01-03T21:23:47.882258Z","shell.execute_reply.started":"2024-01-03T21:23:47.877879Z"},"trusted":true},"outputs":[],"source":["k_values = [3,4,5,6,7,10,20,30,40,50]"]},{"cell_type":"markdown","metadata":{},"source":["Scores for the initial retrieval"]},{"cell_type":"code","execution_count":117,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T21:23:52.563434Z","iopub.status.busy":"2024-01-03T21:23:52.562784Z","iopub.status.idle":"2024-01-03T21:23:53.711365Z","shell.execute_reply":"2024-01-03T21:23:53.709924Z","shell.execute_reply.started":"2024-01-03T21:23:52.563393Z"},"trusted":true},"outputs":[],"source":["initial_retrieval_scores = dict()\n","for k in k_values:\n"," scores = evaluate(initial_retrieval, k)\n"," initial_retrieval_scores[k] = scores"]},{"cell_type":"markdown","metadata":{},"source":["Scores for the Baseline Re-ranker"]},{"cell_type":"code","execution_count":118,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T21:23:54.981836Z","iopub.status.busy":"2024-01-03T21:23:54.981366Z","iopub.status.idle":"2024-01-03T21:23:56.215784Z","shell.execute_reply":"2024-01-03T21:23:56.214384Z","shell.execute_reply.started":"2024-01-03T21:23:54.981798Z"},"trusted":true},"outputs":[],"source":["baseline_reranker_retrieval_scores = dict()\n","for k in k_values:\n"," scores = evaluate(baseline_reranker_retrieval, k)\n"," baseline_reranker_retrieval_scores[k] = scores"]},{"cell_type":"markdown","metadata":{},"source":["Scores for the Bertopic-Reranker"]},{"cell_type":"code","execution_count":127,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T21:27:50.108059Z","iopub.status.busy":"2024-01-03T21:27:50.107464Z","iopub.status.idle":"2024-01-03T21:27:51.305679Z","shell.execute_reply":"2024-01-03T21:27:51.304065Z","shell.execute_reply.started":"2024-01-03T21:27:50.108014Z"},"trusted":true},"outputs":[],"source":["bertopic_reranker_retrieval_scores = dict()\n","for k in k_values:\n"," scores = evaluate(results_for_different_lams[3], k)\n"," bertopic_reranker_retrieval_scores[k] = scores"]},{"cell_type":"code","execution_count":128,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T21:27:53.094975Z","iopub.status.busy":"2024-01-03T21:27:53.094546Z","iopub.status.idle":"2024-01-03T21:27:53.107728Z","shell.execute_reply":"2024-01-03T21:27:53.106596Z","shell.execute_reply.started":"2024-01-03T21:27:53.094939Z"},"trusted":true},"outputs":[],"source":["initial_retrieval_results_f_1 = []\n","baseline_reranker_results_f_1 = []\n","bertopic_reranker_results_f_1 = []\n","\n","initial_retrieval_results_recall = []\n","baseline_reranker_results_recall = []\n","bertopic_reranker_results_recall = []\n","\n","initial_retrieval_results_precision = []\n","baseline_reranker_results_precision = []\n","bertopic_reranker_results_precision = []\n","\n","initial_retrieval_results_nDCG = []\n","baseline_reranker_results_nDCG = []\n","bertopic_reranker_results_nDCG = []\n","\n","for k in k_values:\n"," initial_retrieval_results_f_1.append(initial_retrieval_scores[k]['f_1'])\n"," baseline_reranker_results_f_1.append(baseline_reranker_retrieval_scores[k]['f_1'])\n"," bertopic_reranker_results_f_1.append(bertopic_reranker_retrieval_scores[k]['f_1'])\n"," \n"," initial_retrieval_results_recall.append(initial_retrieval_scores[k]['recall'])\n"," baseline_reranker_results_recall.append(baseline_reranker_retrieval_scores[k]['recall'])\n"," bertopic_reranker_results_recall.append(bertopic_reranker_retrieval_scores[k]['recall'])\n"," \n"," initial_retrieval_results_precision.append(initial_retrieval_scores[k]['precision'])\n"," baseline_reranker_results_precision.append(baseline_reranker_retrieval_scores[k]['precision'])\n"," bertopic_reranker_results_precision.append(bertopic_reranker_retrieval_scores[k]['precision'])\n"," \n"," initial_retrieval_results_nDCG.append(initial_retrieval_scores[k]['nDCG'])\n"," baseline_reranker_results_nDCG.append(baseline_reranker_retrieval_scores[k]['nDCG'])\n"," bertopic_reranker_results_nDCG.append(bertopic_reranker_retrieval_scores[k]['nDCG'])"]},{"cell_type":"code","execution_count":129,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T21:27:57.114992Z","iopub.status.busy":"2024-01-03T21:27:57.114467Z","iopub.status.idle":"2024-01-03T21:27:57.125581Z","shell.execute_reply":"2024-01-03T21:27:57.123592Z","shell.execute_reply.started":"2024-01-03T21:27:57.114950Z"},"trusted":true},"outputs":[{"data":{"text/plain":["[0.3560716124730853,\n"," 0.3426624624489207,\n"," 0.33741020727621196,\n"," 0.33791628338014396,\n"," 0.32865531690243627,\n"," 0.30763712181088515,\n"," 0.2571290703967697,\n"," 0.2282620325948478,\n"," 0.2060150933125378,\n"," 0.19024116999894183]"]},"execution_count":129,"metadata":{},"output_type":"execute_result"}],"source":["bertopic_reranker_results_nDCG"]},{"cell_type":"code","execution_count":130,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T21:28:00.248875Z","iopub.status.busy":"2024-01-03T21:28:00.248077Z","iopub.status.idle":"2024-01-03T21:28:01.314148Z","shell.execute_reply":"2024-01-03T21:28:01.313253Z","shell.execute_reply.started":"2024-01-03T21:28:00.248831Z"},"trusted":true},"outputs":[{"data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAABZgAAAQwCAYAAABolTvzAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOzdd3yV5f3/8dd1ThKyE/bIgIS9QhBkikYQUVFxa7VW6qDOalVataj9VdvaQlvrqFQRR2sddaBfQUHBKAgoIFM2ATJYCZC9Ts65fn+ck5BA2ElOQt5PHzw8uefnzh2SO2+u87mMtRYRERERERERERERkZPl8HcBIiIiIiIiIiIiItI0KWAWERERERERERERkVOigFlERERERERERERETokCZhERERERERERERE5JQqYRUREREREREREROSUKGAWERERERERERERkVOigFlE5AxijJlojFlUR8dKNcbcXhfHOsY5UowxmfV5DhERERE5c1R/Rj3Ws68xposxxhpjAurgnL8zxvzndI9zAuexxphu9X0eEZG6poBZRJodY8wOY0yJMaaw2p9OvnUvG2M2GWM8xpiJxzlOrDHmA2NMjjEmzxiz7nj7iIiIiIicKQ57rt5jjHndGBPu77pERKRhKWAWkebqMmtteLU/u3zLVwN3Az+cwDH+DWQAnYHWwM3A3rossi5GXIiIiIiI1KPLrLXhQDIwEHjUv+WIiEhDU8AsIlKNtfZFa+18oPQENj8beN1aW2StrbDWrrTWfla50hhzjjFmsTEm1xiTUTm62RgTZYx50xiTbYzZaYyZYoxx+NZNNMZ8a4z5uzFmP/A7Y0wLY8w0Y0y6MWavMWa6MSbkGHUZY8wLvlHVG40xY3wLrzXGrDhswweNMR8f70KNMR2NMWuMMZNrWfcbY8z7hy37hzHmOd/rnxtjNhhjCowxacaYXxzjPDXeFugbBfN0tY8vNcas8n1OFxtjkg6rI8t3nk2V1y0iIiIi9c9auweYizdoBsAYM6za8/BqY0xKtXWtjDGvGWN2GWMOGmNm+Za3NMZ86ntWPuh7HXsapd3qO8duY8zDvnN0MMYUG2NaV6vnLN85A491MGNMoDHmbd87GYMOWzfUN5LbWW3ZlcaYNb7XQ4wxS3yfj92+Z/agw8/h27ZGuzpzWDsQY0wvY8wXxpgDvmff66qtu8QYs973XJxVed0iIvVFAbOIyKlbCrxojLnBGBNffYUxpjPwGfA80Bbvg/Yq3+rngSggETgP+Bnw82q7DwXSgPbAH4BngB6+Y3QDYoAnjlHXUGAb0AZ4EvjQGNMK+ARIMMb0rrbtzcCbx7pIY0wC8DXwgrV2ai2bvANcYoyJ8G3vBK4D/utbvw+4FIj0XeffjTFnHeucR6ljIDAT+AXeEeP/Aj7xBfA9gXuBs621EcA4YMfJnkNERERETo0vBL4Y2Or7OAaYDTwNtAIeBj4wxrT17fJvIBToC7QD/u5b7gBew/suwXigBHjhNEo7H+gOXAj8xhhzgS8MT8X7zFrpZuAda63rGNcYAswCyoDrrLXl1ddba78DioDR1RbfyKHnYjfwK7zP6cOBMXjfPXlSjDFhwBe+47YDbgD+aYzp49vkVeAXvufifsCCkz2HiMjJUMAsIs3VLN/IgdzK0RKn4FpgIfA4sN03svZs37obgS+ttW9ba13W2v3W2lW+8PUG4FFrbYG1dgfwV7wPtJV2WWuft9ZW4B1JPQn4lbX2gLW2APij7xhHsw941nfed4FNwHhrbRnwLvBTAGNMX6AL8OkxjtUH+Ap40lr7cm0bWGt34m0pcqVv0Wig2Fq71Ld+trV2m/X6GpgHjDrGOY9mEvAva+131lq3tfYNvA/3w/A+rLcA+hhjAq21O6y1207hHCIiIiJycmYZYwrwto7bh3eAA3ifOedYa+dYaz3W2i+A5XgHJnTEG0bfaa096Htu/RrA99z8gbW22Pfs+we8gzJO1f/zveNwLd7g+ie+5W9w6LnY6Vv+72McJxL4HO9Ajp9ba91H2e7tynP4BmBc4luGtXaFtXap792PO/AOmDiVa7sU2GGtfa3ynZTAB3h/PwFw4X0ujvR9fk+k/Z+IyClTwCwizdUV1tpo358rTuUAvoe1R6y1ffGONl6F9wHbAHF4Hz4P1wYIBHZWW7YT76jkShnVXrfFO7JjRWUgjvfBti1Hl2WttYcdv5Pv9RvAjb4abwbe8wXPR3MTkAW8f4xtwDt6ovJhvfooDYwxFxtjlvrevpeL9yG7zXGOV5vOwEPV/mEgF+/nuZO1divwAPA7YJ8x5h3jm7hRREREROrVFb6RsilALw4953UGrj3s2e0coCPeZ7gD1tqDhx/MGBNqjPmX8baSywe+AaKrt504SdWfras/F3+MN4RNAMYCedba749xnGFAEvDMYc/ah/svcJUxpgVwFfCDb0AGxpgevpYfe3zX9kdO/bl46GGf25uADr71V+N95t5pjPnaGDP8FM4hInLCFDCLiNQBa20OMA3vA2srvA+yXWvZNAfviILO1ZbF4w1xqw532PYlQN9qgXiUbyKVo4nxBcjVj7/LV+dSoBzvCOIbOfYoDfAGtjnAf4/zUP8/IMX31sgr8QXMvgfrD/B+btpba6OBOYA5ynGK8QbqlTpUe50B/KHa5yHaWhtqra0cEfJfa+05eD+3Fvjzca5NREREROqIbwTy63if+8D77Pbvw57dwqy1z/jWtTLGRNdyqIeAnsBQa20kcK5v+dGeH48nrtrr6s/FpcB7eEcx38zxn4vnAX8C5htj2h9tI2vterxB9sUcNvACeAnYCHT3XdtjHP26ijj2c/HXh31uw621d/lqWGatnYC3fcYs33WKiNQbBcwiItUYY4KMMcF4H/QCjTHBxjcBXy3b/tkY088YE+B7+9tdwFZr7X7gLeACY8x1vvWtjTHJvrfSvQf8wRgT4evV/CDwn9rOYa31AK/g7VvcznfeGGPMuGNcRjvgl74JSK4FeuMNdSu9ibePnctau6i2A1TjwvtWuzDgzaN9Lqy12Xj72L0GbLfWbvCtCsLbuiIbqDDGXIy3/93RrMI7wtppjLmImm8ZfAW40zd5ijHGhBljxvs+jz2NMaN9gXYp3lDec5xrExEREZG69Sww1hgzAO/z7WXGmHG+Z7tgY0yKMSbWWrsb73wl/zTeSf0CjTGVQXIE3me5XN88Ik/WdqKT8LhvVHRfvPOBvFtt3ZvAROByjh8wY639C97AeL4x5lgjj/8L3I83HP9fteURQD5QaIzphff3h6NZhXckdKjxToJ9W7V1nwI9jDE3+z53gcaYs40xvX2/z9xkjIny9ZPOR8/FIlLPFDCLiNQ0D+8D7QjgZd/rc4+ybSjwEZCLd1K+zngfTrHWpuN9W9pDwAG8D4gDfPvdh3dEQhqwCO8D6Mxj1PQbvJOlLPW9le5LvKM6juY7vBOZ5ODtWXeNL/Su9G+8k33UGmofzjd5yVV424DMPFrI7LuOC6g2SsPXN++XeEP1g3hHcXxyjNPdD1yG93N6E94RF5XHWg7cgTccP4j3czLRt7oF3skQc4A9eEP2R0/k+kRERESkbvgGHbwJPGGtzQAm4B2lm4131O1kDuUQN+MdzLARb+/mB3zLnwVC8D7XLcXbHu50fI33uXE+MM1aO69avd/iDV+r2lgcj7X2KbzPqF/6AvDavI13oMQC3zsdKz2M93m4AO/giXdr2bfS3/G+83Av3jZ3b1WroQDvoI0b8I7I3oP33XstfJvcDOzw/e5wJ97nahGRemOO3TpIRETONL7Zr/cBZ1lrt/i7HhERERERfzHGLAD+a62d4e9aRESaqgB/FyAiIg3uLmCZwmURERERac6MMWcDZ+EdaS0iIqdIAbOISDNijNmBt7/0Ff6tRERERETEf4wxb+B9Jr7f13JCREROkVpkiIiIiIiIiIiIiMgp0SR/IiIiIiIiIiIiInJKzpgWGW3atLFdunTxdxlSD4qKiggLC/N3GeIHuvfNm+5/86V733z5696vWLEix1rbtsFP7Cd6bj5z6ftn86V733zp3jdvuv/NV2N7bj5jAuYuXbqwfPlyf5ch9SA1NZWUlBR/lyF+oHvfvOn+N1+6982Xv+69MWZng5/Uj/TcfObS98/mS/e++dK9b950/5uvxvbcrBYZIiIiIiIiIiIiInJKFDCLiIiIiIiIiIiIyClRwCwiIiIiIiIiIiIip+SM6cFcG5fLRWZmJqWlpf4uRU5DVFQUGzZsOKl9goODiY2NJTAwsJ6qEhERERERERERkTM6YM7MzCQiIoIuXbpgjPF3OXKKCgoKiIiIOOHtrbXs37+fzMxMEhIS6rEyERERERERERGR5u2MbpFRWlpK69atFS43M8YYWrdurZHrIiIiIiIiIiIi9eyMDpgBhcvNlO67iIiIiIiIiIhI/TvjA2YRERERERERERERqR8KmOtZeHj4cbe5/fbbWb9+PQB//OMfa6wbMWLEKZ/D6XSSnJxMv379uOyyy8jNzT3mcWbNmlVVR22mT5/Om2++edx6Tsbvfvc7pk2bVqfHFBERERERERERkYahgLmaWSuzGPnMAhIemc3IZxYwa2VWg5x3xowZ9OnTBzgyYF68ePEpHzckJIRVq1axbt06WrVqxYsvvnjM7Y8VMFdUVHDnnXfys5/97JTrERERERERERERkTOLAmafWSuzePTDtWTllmCBrNwSHv1wbZ2FzKmpqaSkpHDNNdfQq1cvbrrpJqy1AKSkpLB8+XIeeeQRSkpKSE5O5qabbgIOjU4uLCxkzJgxnHXWWfTv35+PP/74pM4/fPhwsrK817Jt2zYuuugiBg0axKhRo9i4cSOLFy/mk08+YfLkySQnJ7Nt2zZSUlJ44IEHGDx4MP/4xz9qjDau7Rh5eXl07twZj8cDQFFREXFxcbhcLl555RXOPvtsBgwYwNVXX01xcXGdfF5FRERERERERETEfwL8XUBD+X//9yPrd+Ufdf3K9FzK3Z4ay0pcbn79/hre/j691n36dIrkycv6nnANK1eu5Mcff6RTp06MHDmSb7/9lnPOOadq/TPPPMMLL7zAqlWrjtg3ODiYjz76iMjISHJychg2bBiXX375CU1m53a7mT9/PrfddhsAkyZNYvr06XTv3p3vvvuOu+++mwULFnD55Zdz6aWXcs0111TtW15ezvLlywFvO4tKRztGcnIyX3/9Neeffz6ffvop48aNIzAwkKuuuoo77rgDgClTpvDqq69y3333nfDnTkRERERERERERBqfZhMwH8/h4fLxlp+KIUOGEBsbC0BycjI7duyoETAfi7WWxx57jG+++QaHw0FWVhZ79+6lQ4cOR92ncjR0VlYWvXv3ZuzYsRQWFrJ48WKuvfbaqu3KysqOeozrr7/+iGXHOsb111/Pu+++y/nnn88777zD3XffDcC6deuYMmUKubm5FBYWMm7cuBO6bhEREREREREREWm8mk3AfLyRxiOfWUBWbskRy2OiQ3j3F8PrpIYWLVpUvXY6nVRUVJzwvm+99RbZ2dmsWLGCwMBAunTpQmlp6TH3qezBXFxczLhx43jxxReZOHEi0dHRtY6Srk1YWNgRyzwez1GPcfnll/PYY49x4MABVqxYwejRowGYOHEis2bNYsCAAbz++uukpqae0PlFRERERERERESk8VIPZp/J43oSEuissSwk0MnkcT0btI7AwEBcLtcRy/Py8mjXrh2BgYF89dVX7Ny584SPGRoaynPPPcdf//pXQkNDSUhI4H//+x/gHRm9evVqACIiIigoKDju8SIjI496jPDwcM4++2zuv/9+Lr30UpxO7+e0oKCAjh074nK5eOutt064dhEREREREREREWm8FDD7XDEwhj9d1Z+Y6BAM3pHLf7qqP1cMjGnQOiZNmkRSUlLVJH+VbrrpJpYvX07//v1588036dWr10kdd+DAgSQlJfH222/z1ltv8eqrrzJgwAD69u1bNWHgDTfcwNSpUxk4cCDbtm075vGOdgzwtsn4z3/+U6O9xlNPPcXQoUMZOXLkSdcuIiIiIiIiIiIijZOx1vq7hjoxePBgWzkZXaUNGzbQu3dvP1UkdaWgoICIiIiT3k/3v+lLTU0lJSXF32WIn+j+N1+6982Xv+69MWaFtXZwg5/YT2p7bpYzg75/Nl+6982X7n3zpvvffDW25+Zm04NZRERERBqhqd2haB8AKQCpvuVh7WDyFv/UJNLEzFqZxdS5m8jKLSFm6QImj+vZ4O/EFP/QvRcRkcZAAbOIiIiI+I8vXD7h5SJSw6yVWTz64VpKXG4AsnJLePTDtQAKGs9wuvciItJYKGAWEREREf8oO/7kwiJybFPnbqoKGCuVuNw88fE6Mg8WU9kRsbIxYvUOida3tOayIzc8kX2rN148tOzIA9fc7ujnr+0YtXV3rDrGcc5/rM8DtX0eTuAYtW5X67Uc73N59O04Rh1fbdpHqctT/UIocbn5y+cbFTCLiEiDUsAsIiIiIvXLWijcC7vXwJ7KP2vhQJq/KxNp8nblltS6PL+0gmnzNp/WsY2p9rpqmallWfXtTM2VtWxnqq00R25edY7qyw4/bI06TvAYh3Y5+r61XcvxPg9HHOsUjlFLaUdsd/i1HB4uV9qVV8qEF79lYFw0yb4/nVuH1lqziIhIXVDALCIiIiJ1x+P2Bsd71vgC5bXe10XZh7Zp2QU69IcBN8JXT/utVJEzQafoELJqCZk7RQfz9eTzgZMIhxVANikjn1lQ670PbxFAcICDd5dl8PriHQC0DA1kQLXAOTkumujQoAauWEREzlQKmEVERETk1LhKYd+P3hC5Mkze+yO4irzrHQHQtjd0v9AbKHdIgg79IDjq0DEUMIucltG92vLvpek1loUEOvn1uF4EOh1+qkoawuRxPWv0YAbvvX/6in5cMTCGCreHzXsLWZWRy6qMg6zKyOXrzdlVLTYS24R5w+Z4b+Dcq0MkQQH6mhERkZOngLmeOZ1O+vfvj7UWp9PJCy+8wIgRI+rs+BMnTuTSSy/lmmuu4fbbb+fBBx+kT58+p3XM1NRUJkyYQEJCAqWlpVx66aVMmzatjio++jmnTZvGp59+Wq/nERERkVNUfODQaOTKQDlnM1hfsBEU4Q2Rz7rZFyb3h7a9IKDFsY8b1q72Cf3C2tX9NYicYbJyS5i1ahddWodS7vawK7eUmOgQJo/rqR68zUDlPZ46dxNZuSVH3PsAp4M+nSLp0ymSG4fGA1BQ6mJtZh4rM3JZlZHLN1ty+HBlFgBBAQ76dYokOa4lyfHRDIyLJrZliEa2i4jIcSlgrjS1+9F/uZm85ZQPGxISwqpVqwCYO3cujz76KF9//fUpH+9YZsyYUWfHGjVqFJ9++iklJSUMHDiQK6+8kpEjR57Qvm63G6fTWWe1HE9FRQUBAfpSFhERqRPWQm56zTB5z1rIyzi0TURH72jkXuOhY5I3TI7uAo5TGPlW7TkrNTWVlJSU074EkebA47E89N4qPB7Lm7cOJb51qP4ONUNXDIzhioExJ3zvI4IDGdGtDSO6tQG8Ewtm5ZZ4Rzmne0Pnt77bycxvtwPQJjyoWluNliTFRREZHFiflyQiIk2QUrlKtYXLx1p+CvLz82nZsiUAhYWFTJgwgYMHD+JyuXj66aeZMGECRUVFXHfddWRmZuJ2u3n88ce5/vrrWbFiBQ8++CCFhYW0adOG119/nY4dO9Y4fkpKCtOmTWPw4MGEh4dz//338+mnnxISEsLHH39M+/btyc7O5s477yQ93fs2umefffaYwXFISAjJyclkZXn/VXvevHk8+eSTlJWV0bVrV1577TXCw8Pp0qUL119/PV988QW//vWvKSgo4OWXX6a8vJxu3brx73//m9DQUCZOnEhkZCTLly9nz549/OUvf+Gaa66pcc5ly5YxadIk3n//fXJzc7n//vspKSmpcd0pKSkkJyezaNEifvKTn/DQQw/V2X0SERFpNtwu7yjk6r2S96yB0jzfBgbadIe4oXD27YfaXIS39WvZIgIzFqWxNO0Af7kmifjWof4uR5ooYwyxLUOJbRnKpUmdAHC5PWzaU8DKjFxWpntba3y5YZ9ve+jaNrxGL+deHSIIUDsWEZFmrfkEzJ894v3F6VS8Nr725R36w8XPHHPXkpISkpOTKS0tZffu3SxYsACA4OBgPvroIyIjI8nJyWHYsGFcfvnlfP7553Tq1InZs2cDkJeXh8vl4r777uPjjz+mbdu2vPvuu/z2t79l5syZRz1vUVERw4YN4w9/+AO//vWveeWVV5gyZQr3338/v/rVrzjnnHNIT09n3LhxbNiw4ajHOXjwIFu2bOHcc88lJyeHp59+mi+//JKwsDD+/Oc/87e//Y0nnngCgNatW/PDDz8AsH//fu644w4ApkyZwquvvsp9990HwO7du1m0aBEbN27k8ssvrxEwL168uOpaO3bsyM0338xbb71FQkLCEdddXl7O8uXLj/n5FxEREZ+yAm9/5N1rDo1M3rcB3GXe9QHB0L4v9L3S1ys5Cdr3gaAw/9YtIkdYvyufaXM3M65ve64dFOvvcuQME+h00C8min4xUdw8rDMAecUuVmfm+vo557Jg4z7eX5EJePs+94+JqurlnBwXTceoYLXWEBFpRppPwOwn1VtkLFmyhJ/97GesW7cOay2PPfYY33zzDQ6Hg6ysLPbu3Uv//v156KGH+M1vfsOll17KqFGjWLduHevWrWPs2LGAtwXF4aOXDxcUFMSll14KwKBBg/jiiy8A+PLLL1m/fn3Vdvn5+RQWFhIeHl5j/4ULFzJgwAC2bNnCAw88QIcOHfj0009Zv3591Yjn8vJyhg8fXrXP9ddfX/V63bp1TJkyhdzcXAoLCxk3blzVuiuuuAKHw0GfPn3Yu3dv1fINGzYwadIk5s2bR6dOnaque8KECTgcjiOuu/r5REREpJqCvYdGI1eOTj6QBvhmdgpp5W1tMXTSoTC5dTdw6tFQpLErdbl54N2VRIUG8qerkhTiSYOICg3k3B5tObeH9x0s1loyDpSwMuMgK32tNV7/dgflbg8A7SJaVE0gODCuJUmxUYS10M8YEZEzVfP5Dn+ckcb8Luro634+u05KGD58ODk5OWRnZzNnzhyys7NZsWIFgYGBdOnShdLSUnr06MEPP/zAnDlzmDJlCmPGjOHKK6+kb9++LFmy5ITPFRgYWPWw6XQ6qaioAMDj8bB06VKCg4OPuX9lD+bt27czbNgwrrvuOqy1jB07lrfffrvWfcLCDo1wmjhxIrNmzWLAgAG8/vrrpKamVq1r0eLQZD+2cgpjoGPHjpSWlrJy5Uo6deqEtZa+ffsyb948IiIijnk+ERGRZsnj8QbHe9bUnHyveouv6M7ed10NuOFQi4vITt73OTci2cXZPLvnWfqV9KNNSBt/lyPSaE2du4nNewt5/edn0yosyN/lSDNljCG+dSjxrUOZkOydVLCsws2G3QWs8rXVWJWRy7z13gFFDgM92kccaq0RH033dhE4HY3rZ5GIiJya5hMwNwIbN27E7XbTunVr8vLyaNeuHYGBgXz11Vfs3LkTgF27dtGqVSt++tOfEh0dzYwZM3jkkUfIzs5myZIlDB8+HJfLxebNm+nbt+9J13DhhRfy/PPPM3nyZABWrVpFcnLyUbdPSEjgkUce4c9//jPPPfcc99xzD1u3bqVbt24UFRWRlZVFjx49jtivoKCAjh074nK5eOutt4iJOf4s1tHR0bz66quMHTuWsLAwRowYQXZ2Nt999x0XXHDBaV23iIhIk+cqhX3rD5t8bx24irzrHQHQtjd0u8AbJHdMgvb9ICTar2WfqOlrppNWlsb01dOZMmyKv8sRaZS+3ZrDq4u287PhnUnp2c7f5YjU0CLAWRUgVzpYVM6qTO8Egiszcvls3R7eWeadNDYsyEn/2CiS41oyMD6agXHRtIs89kAoERFpnBQwVwprV/uEfmGn9+BW2YMZvKN133jjDZxOJzfddBOXXXYZ/fv3Z/DgwfTq1QuAtWvXMnnyZBwOB4GBgbz00ksEBQXx/vvv88tf/pK8vDwqKip44IEHTilorQyJk5KSqKio4Nxzz2X69OnH3OfOO+9k2rRpFBUV8frrr/OTn/yEsjJvv8ann3661oD5qaeeYujQobRt25ahQ4dSUFBwQvW1b9+eTz/9lIsvvpiZM2fy/vvvc8899/DQQw+d1nWLiIg0KcUHYO+6mpPvZW8C6/auD4qADv1g4E8Phclte0FAi2MftxGq8FSwOns1H275EItl1tZZ3DngTo1iFjlMXrGLh95bTde2YTx6cW9/lyNyQlqGBXF+z3ac7/sHEWst23OKqkY4r8rIZcbCNCo83ne2dooKrtbLuSX9Y6IICXL68xJEROQEmOotCpqywYMH28MnfNuwYQO9e+vhq6krKCiotUXG8ej+N32pqamkpKT4uwzxE93/5qtZ3XtrIS/jUGuLyjA5L+PQNhEdD7W26NDf+6dlAjgc/qv7FJS5y9iRt4O0vDTS8tLYlruN7Xnb2ZG/gwpPRdV2gY5Arup+VYOOYjbGrLDWDm6wE/pZbc/N0rhZa7nv7ZV8vm4PH909kv6xtbf3a1bfP6WGpnzvS11uftyVz8pqrTUyD5YA4HQYeraPYKAvdB4YH01im3Acaq1RpSnfezl9uv/Nl7/u/dGemzWCWURERKQhuF2Qs6Var+TV3v+X5vo2MN6J9uKGwNm3HQqUw5vW2+CLXEWk5fpC5LxtbM/dTlpeGpmFmXisd/Inh3EQGx5LYlQig9sP5v0t71eFzC6PS6OYRQ7z8apdfLpmN5PH9TxquCzSVAUHOhnUuSWDOresWpZdUMbqaqOcP1m1i7e+SwcgIjiAAbHRNfo5twlveu/gERE5kyhgFhEREalrZYWw98dDk+/tXgP7NoDb22KKgGBo1wf6XnFodHL7vhDUdCawPVh6sMZI5G2520jLS2Nv8d6qbQIcAXSJ7EKvVr24JPESukZ1JSEqgS5RXWjh9IYBTy19CkPNkWge61EvZhGfrNwSHv94HYM7t+TO87r6uxyRBtE2ogUX9GnPBX3aA+DxWNJyCvkh3Rc6p+fy0tfbcPtaa8S1CiE5rmVV6Ny3UyTBgWqtISLSUBQwi4iIiJyOwn2+9hZrDo1O3r8N8LUhC2npDZCH3AEdB3gD5dbdwdn4H8Ostewr3ucdiVwtRE7LTeNg2cGq7UICQkiISmBIhyEkRieSEJVA16iuxEbEEuA49nWu3rcal8dVY5nL42LVvlX1cUkiTYrbY3nw3VV4PJa/X5+MU20BpJlyOAzd2kXQrV0E1w2OA6Ck3M3arDxWZXhba6zYcYD/W70LgECnoXfHyEOjnOOiSWgThjH6OyQiUh8a/282IiIiIo2BxwMHtx9qbVEZJhceGrFLdLw3TO5/3aHJ9yJjoJH/Quv2uNlVuKuqrUVli4u0vDSKXEVV20UGRdI1uiuj40eTGJVIYnQiiVGJdAjrgMOcWk/o9y9/v+q1+giK1DRjYRrfbT/AX65JIq5VqL/LEWlUQoKcDEloxZCEVlXL9uWXsjIjl5XpuazKOMj7KzJ5c8lOAKJDAw+11oiPJjk2mpZhQf4qX0TkjKKAWURERORwFWWwb33Nyff2roPyQu96RwC07QVdR1ebfK+fd7RyI+Zyu9iZv7MqPK4Mknfk76Cssn0H0DakLYlRiVze9XJvkOwLk1sHt9boL5EGsn5XPtPmbeKivh24dlCsv8sRaRLaRQYzrm8HxvXtAHjfBbBlXwGr0g/1c35+wRZ8nTXo0jq0aoTzwPiW9O4YSVBA05pEV0SkMVDALCIiIs1byUHfiOS1hwLlnE3gm3SOoHBo3w+SbzzUL7ltLwgM9m/dx1DsKmZH/o6q/siVvZIzCjJwW3fVdjHhMSRGJTKs4zC6Rnv7IydGJxIZFOnH6kWk1OXmgXdXEh0axB+v6q9/2BE5RU6HoVeHSHp1iOSGIfEAFJZVsCbzUC/nxdv2M2uVt7VGUICDvp0OtdYYGNeSuFYh+jsoInIcCpjrmdPppH///lhrcTqdvPDCC4wYMYIdO3bQu3dvevbsWbXtgw8+yM9+9jO6dOlCREQExhhatmzJm2++yQMPPMD27dspLCwkOzubhIQEAP75z38yYsSIE6pl+fLlvPnmmzz33HMntH1KSgq7d+8mODiYoKAgXnnlFZKTk0/6c3AyUlJSmDZtGoMHD67X84iISDNkLeRl1mxvsXsN5KUf2ia8gzdE7nnRoTC5ZQI4GudopryyvBoBclpeGtvztpNVmFW1jdM4iY+Mp2t0V8Z2HktidCJdo7rSJaoLIQEhfqxeRI5m6txNbN5byOs/P5tWegu/SJ0KbxHAiK5tGNG1DeCdb2B3XmnVCOdV6bm8/X06r327A4DWYUEMqBrlHE1SbDRRIYF+vAIRkcZHAfNhsouzmfzNZKadN402IW1O+3ghISGsWrUKgLlz5/Loo4/y9ddfA9C1a9eqdYf76quvaNOmDU8++SRPP/00H330EeDtTTht2jQ+/fTTk65l8ODBJx3cvvXWWwwePJjXXnuNyZMn88UXX5zQfhUVFQQENNyXV0OfT0REGjl3BeRsrhYm+wLlksqJ6Qy07gqxg2Hwz729kjskQXg7v5ZdG2st+0v3k5Z7qD/y9rztbMvbRk5JTtV2LZwt6BLZhaS2SVzR7Qq6RnclMSqR+Ih4Ap36RVikqVi0JYdXF23nluGdSenZ+L4niZxpjDF0ig6hU3QIl/TvCIDL7WHTnoJDoXNGLgs27qvap2vbMJLjWpIcH83AuGh6dogg0Nk4/zFaRKQhKJE7zPQ10/lh7w9MXz2dKcOm1Omx8/Pzadny5HozDh8+/Kgjjnfs2MGtt95KTk4Obdu25bXXXiM+Pp6JEycSHBzM8uXLyc/P529/+xuXXnppjXC6sLCQ++67j+XLl2OM4cknn+Tqq68+Zh1Tp04FoKioiPvuu49169bhcrn43e9+x4QJE3j99df58MMPKSwsxO12M3v2bCZMmMDBgwdxuVw8/fTTTJgwgR07dnDxxRdzzjnnsHjxYmJiYvj4448JCTk0isrj8XDrrbcSGxvLww8/zOTJk0lNTaWsrIx77rmHX/ziF6SmpvL444/TsmVLNm7cyObNm0/qcysiImeI8iLY+2PNyff2rofKnsLOFtC+D/S+3Dfx3gBo1wdahPu37sN4rIc9RXuqRiJX75GcX55ftV1YYBhdo7oystPIqhA5MTqRTmGdcDqcfrwCETlducXlPPy/1XRtG8YjF/f2dzkizVag00G/mCj6xUTx02GdAcgvdbEmI49VGQdZlZHL15v38cEPmQAEBzro1ymKgfHRVcFzp6hgtdYQkWaj2QTMf/7+z2w8sPGY25S7y1mbsxaL5b1N77Fx/8Zjjvjp1aoXvxnym2Mes6SkhOTkZEpLS9m9ezcLFiyoWrdt27YaLSeef/55Ro0aVWP/zz//nCuuuKLWY993333ccsst3HLLLcycOZNf/vKXzJo1C/CGz99//z3btm3j/PPPZ+vWrTX2feqpp4iKimLt2rUAHDx48PDDH7WOP/zhD4wePZqZM2eSm5vLkCFDuOCCCwD44YcfWLNmDa1ataKiooKPPvqIyMhIcnJyGDZsGJdffjkAW7Zs4e233+aVV17huuuu44MPPuCnP/0p4B2NfNNNN9GvXz9++9vf8txzzxEVFcWyZcsoKytj5MiRXHjhhVXnW7duXVXLEBERaaKmdoci78igFIBU3/KwdjB5y6HtCrNhz+qak+/t3wr4ZusJjvaORh5yx6HJ99r0AGfjeeSp8FSQUZBRI0DelruNHfk7KKkoqdquVXArEqISuKjLRSRGJ1ZNttcutJ1+YRU5A1lrmTJrHTmFZcy4ZSQhQfoHI5HGJDI4kHO6t+Gc7odaa2QeLGGlr63GqoyDvLFkJ68s3A5A24gW1Xo5R5MUF014i8bzPCIiUpf03a2a3UW7a3y8q2gXnSM7n9Yxq7fIWLJkCT/72c9Yt24dcOwWGeeffz4HDhwgPDycp556qtZtlixZwocffgjAzTffzK9//euqdddddx0Oh4Pu3buTmJjIxo01w/Uvv/ySd955p+rjo42svummmygvL6ewsLCq1nnz5vHJJ58wbdo0AEpLS0lP9/avHDt2LK1atQK8P3Afe+wxvvnmGxwOB1lZWezduxeAhISEqnB90KBB7Nixo+qcv/jFL7juuuv47W9/C8CCBQtYv34977//PgB5eXls2bKFoKAghgwZonBZRORMULTv6Mvn//5QmFy459C6qHhvmNz/mkNhclQsNJLwtcxdxo68HVWjkSsn3NuRv4OKygkEgQ5hHUiMSmRQ+0E1guSWwSf3ricRado+XrWLT9fsZvK4nvSLifJ3OSJyHMYY4lqFEtcqlMsHdAKgvMLDxj35Vb2cV2Xk8sX6vb7toXu7cAb6Rjgnx0XTo30ETkfjeG4RETkdzSZgPt5I4+zibC7+8GKsbwSUxZJfns/U86bWSS9m8LaZyMnJITs7+7jbfvXVV0RHR3PTTTfx5JNP8re//e2kznX4yKZTHen01ltvMWjQICZPnsx9993Hhx9+iLWWDz74oMYEhQDfffcdYWFhNfbNzs5mxYoVBAYG0qVLF0pLSwFo0aJF1XZOp5OSkkMjtkaMGMFXX33FQw89RHBwMNZann/+ecaNG1fjfKmpqTXOJyIiZ6hFz0LbXpCY4uuV3N/7J6RxBLBFrqJDI5HztrE919sfOaswC4/1AOAwDmLDY0mMSuTc2HNJjEqka3RXEqISCAvUzzKR5i7zYDGPz1rH4M4tufO8rv4uR0ROUVCAg6RY70SAPxvuXZZbXF6jl/Pc9Xt4d3kGAKFBTvrHRFX1ck6Oa0mHqGA/XoGIyKlpNgHz8UxfM73ql8BKHuup017MGzduxO1207p1a4qLi4+7fUBAAM8++yz9+/dnypQpVSODK40YMYJ33nmHm2++mbfeeqtGe43//e9/3HLLLWzfvp20tDR69uzJ0qVLq9aPHTuWF198kWeffRbwtsg42ihmYwxPPfUUXbt2ZePGjYwbN47nn3+e559/HmMMK1euZODAgUfsl5eXR7t27QgMDOSrr75i586dJ/Jp4rbbbuObb77huuuu48MPP2TMmDG89NJLjB49msDAQDZv3kxMTMwJHUtERBqxvCzYMg+2HGcC2cd2QaD/f9k6WHqwxkjkyl7Je4v3Vm0T4AigS2QXerfqzfjE8XSN8obIXaK60MLZ4hhHF5Hmyu2xPPTeaizw9+uTNZpR5AwTHRpESs92VZN2WmvZub+4KnBemZHLzEXbcbm9g906RgVXtdZIjoumf2wUoUFHRjezVmYxde4msnJLiFm6gMnjenLFQP2eLCL+oYDZZ/W+1bg8rhrLXB4Xq/atOq3jVvZgBu8PkjfeeAOn09tP7fAezLfeeiu//OUva+zfsWNHfvKTn/Diiy/y+OOP11j3/PPP8/Of/5ypU6dWTfJXKT4+niFDhpCfn8/06dMJDq75i/mUKVO455576NevH06nkyeffJKrrrrqqNcREhLCQw89xNSpU3nhhRd44IEHSEpKwuPxkJCQwKeffnrEPjfddBOXXXYZ/fv3Z/DgwfTq1euEPmcADz74IHl5edx8881Mnz6dPXv2cNZZZ2GtpW3btlW9pkVEpAnxuCFzOWyZC5vnwV7vPABExR17vwYMl6217C3eS1peWo0QOS03jYNlh+YrCAkIISEqgSEdhpAYnUhCVAJdo7oSGxFLgEOPVyJy4mYsTOO77QeYek0Sca1C/V2OiNQzYwxd2oTRpU1YVSBc6nKzfnd+VVuNVRm5fLbO2xbM6TD0aB9R1ct5YHw06zLzeGzWOkpcbgCyckt49EPvc5VC5uZB/8AgjY2x1vq7hjoxePBgu3z58hrLNmzYQO/ezW/25YkTJ3LppZdyzTXX+LuUOlFQUEBERMRJ79dc7/+ZJDU1lZSUFH+XIX6i+3+GKD4AW+d7Q+WtX0LJQTBOiB8G3cdC93HQrjf8v+ijH+N3eXVeltvjZlfhLrblbasx2V5aXhpFrqKq7SKDIuka3ZXEKF+I7HvdIawDDuOo87qaO3/9vTfGrLDWDm7wE/tJbc/N4h/rd+Uz4cVFjOnVnpd+etZpT+Cpn53Nl+79mWd/YRmrM729nFdm5LI6I5f8Uu8cDoaq6Y1raB0WxN+vT8ZhDA7jDbMdBhyO6h/7lhmD8f3/8O3NCWxTucw4OO4xpW7NWpnFox+urfoHBoCQQCd/uqq/QuZmpLE9N9frEBtjzEXAPwAnMMNa+8xh688FngWSgBuste8ftj4SWA/MstbeW5+1ioiISB2wFvaug81zve0vMpeB9UBoa2+Y3ONC6Dr6yP7JYe1qn+gvrN1pleNyu9iZv7MqPK4Mknfk76DMXVa1XduQtiRGJXJ518urJtlLjE6kdXBr/WIkIvWi1OXmgXdX0jI0iD9e1V/fa0SkhtbhLRjdqz2je7UHwOOxpOUUsSojl4f/t7rWffYXlfOzmd83ZJkn5FDoXEtAbQ4PqKuv923vOHJ7w2GBtuMY+5/A+cyJ1uQ4+WPWXlNtobz3HwOOtb0x8Pv/W18jXAYocbl5evZ64lqFEOh0+P6Yaq8P/9jo547UqXoLmI0xTuBFYCyQCSwzxnxirV1fbbN0YCLw8FEO8xTwTX3VeKZ6/fXX/V2CiIg0J2WFsP1rX6j8BRTs8i7vmAyjHoYe46DTQHA4j36MyVuqXp7Kv8YXu4rZkb+jqj9yZa/kjIIM3PbQA3hMeAyJUYkM6zisapK9xOhEIoMiT+p8InXtBAZm3AncA7iBQmBS9edqY0w83oEZv7PWTmuwwuWU/eXzTWzeW8gbtw6hVViQv8sRkUbO4TB0axdOt3bh/P2LzWTllhyxTZvwFvzr5rPwWG8g7bHe9l8eCx5r8ViLrXqN7+Pq6yu3t3g8VG1vqWUbT/VjnOAxfcs4YvtD9R06/tH3P3L742/j9niOe51VnyOoub/n8M/jidRUc11DyCks5+qXlpzw9gEOb+Ac4DQEVQbPAYZAx6HXAQ6Hd53vdaDTQVCAbz9HzdeBAd7jVH9defxAp/c4tb0+1rqa2ykUh8bbHqU+RzAPAbZaa9MAjDHvABPwPvgCYK3d4VvnOXxnY8wgoD3wOXDKb1m01uoLsBk6U1q/iIg0Wvu3eUcob54LO78FdzkERUDXFOj+mLf9RUSHkz5sdnE2z+55ln4l/WgT0uaI9XlleTUC5MpeyVmFWVXbOI2T+Mh4ukZ3ZWznsSRGe0ckd4nsQmig+ptK43OCAzP+a62d7tv+cuBvwEXV1v8N+KyBSpbTtGhLDjO/3c4twztzXo+2/i5HRJqYyeN61toiYcr43gzq3MqPlcnRHDfo9xw9oD58++v/tYR9BWVHnKNNeBB/vS4ZV4WHCo+Hcret9bXLbSmv8OBye6jwVHvttrjcHsoPe+1yeyhzeSh0V1DutlT4lrl821S+rty2PuOYQGdl0G0ICjgUZgc6Hd5gPMAcdcT2iY7sDqwWjAcdZZsAZ83gvfL14XUEOOo2FD+8PUpj6r9enwFzDJBR7eNMYOiJ7GiMcQB/BX4KXHCM7SYBkwDat29PampqjfXh4eFkZmYSFRWlkLkJc7vdFBQUnPD21lry8vIoKio64mtCmpbCwkLdw2ZM979xMR4X0bk/0urAclrvX0FoiXeUcnFIDPs7Xsz+1oPJi+qNdQRCPrBiI7DxpM/z7v53SStL45HZjzAkfAh7XHu8f8r3sLdiL/nu/KptA00g7QLa0SGwA8lRyXQI7ED7wPa0DWxLgPE94uR7/+zz/SeNWzP+e38iAzPyq20fRrX2m8aYK4DtQBHS6OUWl/Pw/1bTtW0Yj1ys+UJE5ORVBklVoxijQxrNKEapnTEGpwEnp59NPXZJ76P8A0OfRvGPlm5PzeD58NflFd5g2+X24Krw4PJ4w2+X+7DXR9n/RF+XujwUlnpD8cplFdWC8IoGDMUrw+agAEetAfbRRm9Xhtnejw3v/5BZa3uUqXM3+f3vf2Od5vxuYI61NvNYwbC19mXgZfBOVnL422ldLheZmZlkZWXVsrc0FaWlpQQHB5/UPsHBwQwYMIDAwMB6qkoagiYrad50/xuB/N3eUcpb5kFaKpQXgrMFdDkHejwA3ccS2iqRUCDuNE+1q3AXn2z9hG93fovF8l3Rd3xX9B0AYYFhdI3qyvlR51dNspcYlUin8E44j9V2Q5qcZvz3/oQGZhhj7gEeBIKA0b5l4cBv8I5+PlrbOWkkrLX8dtY6cgrLmHHLSEKC9D1MRE7NFQNjuGJgTHP+2dlsNfZ/YHA6DE6Hk+DApvMzrjIUL/eF3jVGdh/22lVRObLbO5q7+mvvMWq+9obZh0Z51/a6MngvcblxlR4lhHd7KCpz11r/rlpa5jS0+gyYs6j5+2asb9mJGA6MMsbcDYQDQcaYQmvtIydTQGBgIAkJCSezizRCqampDBw40N9liIic+TxuyFpxqPXFnjXe5ZEx0P9a6H4hJJ4HQWGnfSprLdtytzE/fT7z0+ez4cCGGuudxsk5Mefw+LDHaRfaTu9EEgGstS8CLxpjbgSmALcAvwP+bq0tPNbfk+O9808axuJdFcxeU8Y13QPJ2bKS1C3H3+dkNON3ATR7uvfNl+598xQN/GGYg8JCS3i4A/K2kFrXP1TkmAJ9f0KOtoHD9+e0xj0avNNzeD2UWsH+0iOHW7cKNn7/PlCfAfMyoLsxJgFvsHwDcOOJ7GitvanytTFmIjD4ZMNlEREROQHFB2DbAm+ovPVLKN4PxgFxQ2HMk95QuX1f75TVp8ljPazNWcv89PksSF/AzvydAAxoO4BJSZN4fd3rlHvKAXBbN0t3L8XpcCpclubgZAdmvAO85Hs9FLjGGPMXvL9veowxpdbaF6rvcLx3/kn9yzxYzH1fLeTsLi3588+H43TU/fc2jWRsvnTvmy/d++ZN9795eTwqq9b2KI9P6E/Kmdoiw1pbYYy5F5iLN26faa390Rjze2C5tfYTY8zZwEdAS+AyY8z/s9b2ra+aREREmj1rYe+Ph1pfZHwH1gMhrbwT83W/ELqOhtC6mSDG5XGxfM9y5qfP56v0r9hXso8AE8DZHc7m5t43c378+bQLbcdTS5/CUvNf4z3Ww/TV05kybEqd1CLSiB13YIYxpru1tnJo0nhgC4C1dlS1bX4HFB4eLov/uT2Wh95bjQX+dl1yvYTLIiIicmZrzO1R6rUHs7V2DjDnsGVPVHu9DO8IjWMd43Xg9XooT0REpHkoL4Lt33jbXmz5AvIzvcs7JME5D0KPcRAzCOqon3FJRQmLsxYzP30+X2d+TX55PiEBIYzsNJLR8aM5N/ZcolpE1dhn9b7VuDyuGstcHher9q2qk5pEGrMTGZgB3GuMuQBwAQfxtseQJuKVhWl8t/0A064dQFyrUH+XIyIiIk1UY+2/3lgn+RMREZHTcWD7oV7KOxaBuwwCw6Dr+XDer70jlSM71tnp8sry+Drza+bvnM/iXYspdZcS1SKKlLgUxsSPYUSnEQQHHH3C1vcvf7/qdWN7WBJpCCcwMOP+EzjG7+q+MjldP+7K46/zNnFxvw5cfZb/RxiJiIiI1DUFzCIiImeCinJIX3Ko9UXOZu/yVl3h7Nu8gXLnERDQos5OubdoLwsyFjA/fT7L9yzHbd20C23Hld2vZEz8GAa1H0SAQ48aItJ8lbrc/OrdVbQMDeKPV/ZXT3kRERE5I+m3PhERkaaqYK8vUJ4L21KhvACcQdB5JAy+1Rsqt+5ap6fckbeD+enzmZ8+n7U5awHoEtmFiX0nckHnC+jbuq8CFBERn798vonNewt549YhtAwL8nc5IiIiIvVCAbOIiEhT4fHArh98vZTnwe5V3uURnaDfVd5eygnnQYvwOjultZb1B9Yzf+d8FqQvYFveNgD6tu7LLwf+kjHxY0iMTqyz84mInCkWbslm5rfbmTiiC+f1aOvvckRERETqjQJmERGRxqwkF7bN907Ot+ULKM4B44DYs2H0495QuX0/qMNRwxWeClbuW8n8dG+ovLtoN07jZFD7QVzb81rGxI+hQ1iHOjufiMiZJre4nIf/t5pu7cJ55OJe/i5HREREpF4pYBYREWlMrIXsjYdGKacvBeuGkJbQ7QJv24tuF0Boqzo9bZm7jKW7ljI/fT6pGakcLDtIkCOIEZ1GcNeAu0iJS6FlcMs6PaeIyJnIWstvZ61jf2E5r95yNsGBTn+XJCIiIlKvFDCLiIj4W3kx7FjoC5W/gLx07/L2/WHk/d5RyjGDwVm3P7YLygtYmLmQ+enzWZS1iOKKYsIDwzk39lzGxI/hnJhzCA0MrdNzioic6WatymL2mt1MHteTfjFR/i5HREREpN4pYBYREfGHgzu9I5Q3z/WGyxWlEBgGiSkw6kHvSOWomDo/bU5JDl9lfMX89Pl8t/s7KjwVtA5uzfjE8YyJH8OQDkMIdAbW+XlFRJqDzIPFPDHrR87u0pI7z6vbSVZFREREGisFzCIiIg3B7fK2u9gyFzbPg5xN3uUtE2DQRG+g3OUcCGhR56fOLMis6qe8ct9KLJbY8Fhu6nUTYzqPIalNEk6H3sItInI63B7Lg++txgJ/uy4Zp6PueuOLiIiINGYKmEVEROpL4T7f5HxzYdtXUJYPjkDoPAIG3QLdx0GbbnV+Wmstmw9uZkH6Auanz2fTQW+Y3bNlT+4acBej40fTo2UPTB1ODCgi0ty9sjCN77cfYNq1A4hrpfZCIiIi0nwoYBYREakrHg/sXukdobxlHuz6wbs8vAP0meDtpZyYAi0i6v7U1sPq7NXM3zmf+enzySzMxGAY2G4gDw9+mNHxo4mLiKvz84qICPy4K4+/ztvEJf07cPVZdd/eSERERKQxU8AsIiJyOkrzYNsC30jlL6BoH2AgdjCcPwV6XAgdkqAeRgu73C6+3/M989Pn81XGV+SU5BDgCGBox6Hc2v9Wzo87nzYhber8vCIiJ2PWyiymzt3ErtwSOkWHMHlcT64YeOwQtqH2qQulLjcPvLOKlqFB/OGK/np3iIiIiDQ7CphFREROhrWQvck7QnnLPEhfAp4KCI6Cbhd42150GwNh9RPsFruKWZS1iPnp81mYuZACVwEhASGMihnFmPgxjIodRURQ3Y+QFhE5FbNWZvHoh2spcbkByMot4dEP1wIcNfxtqH3qyp8/38iWfYW8eesQWoYF1eu5RERERBojBcwiIiLH4yqBHYtg81xvP+XcdO/ydn1hxH3eCfpih4Czfn6sHiw9SGpGKgvSF7Bk9xLK3GVEt4jmgs4XMCZ+DMM6DaOFs+4nBxQROV1T526qCn0rlbjcPPnJjxSUVdS6z7Q63Gfq3E31GjAv3JLNa9/uYOKILpzbo229nUdERESkMVPALCIiUpvcDG+YvHkebP8GKkogMBQSzoNzfgXdxkJ0/fU03lO0h/np81mQvoAVe1fgtm46hHXgmh7XMCZ+DAPbDSTAoR/jItK47cotqXV5XomLx2etO6ljnco+Rzt/XcgtLufh/62mW7twHrm4V72dR0RERKSx02+mIiIiAO4KyPjuUKicvcG7vGUXOOtn3lHKXc6BwOB6KyEtN4356d5J+n7c/yMAXaO6cmu/WxnTeQx9WvVRb08RaVI6RYeQVUvI2yEymP+775xa97ns+UXsyS+tk306RYecZMUnxlrLbz9ax4Gicl695WyCA531ch4RERGRpkABs4iINF9FOb7J+ebC1gVQlgeOAOg8Agbe5O2n3KZ7vUzQB96AYl3OuqpQeUf+DgCS2iTxwFkPMCZ+DF2iutTLuUVEGsLkcT1r9EYGCAl08sjFvWgbUXtrn0cu7lUn+wQHOpg8rmcdXYlX5USClaH5pUkd6BcTVafnEBEREWlqFDCLiEjz4fHAntXeEcpb5kLWD4CFsHbQ+zLocSEkng/BkfVWgsvjYsXeFczfOZ8FGQvYV7wPp3EyuMNgbup9E+fHnU/7sPb1dn4RkYZU2f946txN7MotoVN0CJPH9TxmX+S62McC/TtF1Wn/5cMnEgSYv2Efs1Zm1ftEgiIiIiKNmQJmERE5s5XmQ9pX3lB56xdQuBcwEHMWpDzqDZU7DACHo/5KqChl8a7FzE+fz9eZX5NXlkewM5gRnUYw5qwxnBd7HlEtNAJORM5MVwyMOekA9nT3+eu8TTy/YCtzf9zDuL4dTuo4R1P7hIWeep9IUERERKSxU8AsIiJnFmshZwts8Y1S3rkEPC5oEQXdRnvbXnS7AMLb1msZ+eX5fJ3xNQvSF/Dtrm8pqSghIiiClNgUxsSPYUTMCEIC6qc3qIhIc3ff6O4s2LiPxz5cy6DOLWkTXntrjZNxtAkD63MiQREREZGmQAGziIg0fa5S2LHoUKh8cId3edveMPxub6gcNwScgfVaRnZxNgvSFzA/fT7L9iyjwlbQNqQtl3e9nDHxYxjcYTCBjvqtQUREICjAwd+uS+ay5xfx2Idr+dfNg05rklRrLSFBTorL3Uesq6+JBEVERESaCgXMIiLSNOVlegPlzfNg+9fgKoaAEEg4F0bcB90vhOj4ei8jPT+9apK+NdlrsFg6R3bm5r43MyZ+DP3b9Mdh6q/9hoiI1K5nhwgeHteDP87ZyIc/ZHH1oNhTPtbfv9xCcbmbAIehwmOrlocEOut8IkERERGRpkYBs4iINB5Tu0PRPgBSAFJ9y8PawYMbIHOZd4Ty5nmw70fvuuh4SL7JGygnjILA+h1JZq1l44GNVaHy1tytAPRu1Zt7ku9hTPwYukZ3Pa2RciIiUjduOyeRL9fv43ef/Miwrq2JOYXRxv9ZupPn5m/h+sFxDEtsxbR5m0948kERERGR5kABs4iINB6+cLnW5VO7QmkuOAIgfjiM/b239UXbnlDPYa7b42ZV9iq+3PklX2V8RVZhFg7jYGC7gfzm7N8wOn40ncI71WsNIiJy8pwOw7RrB3DRP77h1++v5t+3DsXhOPGfGZ+v283jH69jTK92/OHKfgQ4HVx51qmPhBYRERE5EylgFhGRpqHnJdDjQug6GoKj6v105e5ylu5eyoL0BXyV8RUHSg8Q6AhkeKfhTEqaREpcCq2CW9V7HSIicnriW4cyZXwfHvtoLW8u2cHEkQkntN/32w/wy3dWkRwXzQs3nkWAU+2ORERERGqjgFlERBqH7M3HXn/lS/VeQpGriIWZC5mfPp+FWQspchURFhjGuTHnMrrzaEbFjCIsMKze6xARkbr1kyFxzFu/h2c+38ioHm3p2jb8mNtv2lPA7W8sI65lCDNvOZuQIGcDVSoiIiLS9ChgFhER/7EWdiyCxc97eyv7wYHSA6RmpPLlzi9ZunspLo+LVsGtuKjLRYyJH8PQjkMJcgb5pTYREakbxhj+cnUSFz77DQ++t5oP7hx+1BHJWbkl3DLze0KCnLxx6xBahulngIiIiMixKGAWEZGG53bB+o9h8XOwezWEtoaURyH1Tw1y+l2Fu6om6Vu5byUe6yEmPIYbet3AmPgxJLdNxunQaDURkTNJu8hgnprQj/veXsn0r7dx7+juR2yTW1zOLTO/p6i8gv/dOZzYlqF+qFRERESkaVHALCIiDac0H354E5a+BPmZ0LobXPosDLgBAkNg2au1T/QX1u60TmutZWvuVuanz2dB+gI2HNgAQPeW3ZmUNIkx8WPo2bInpp4nCxQREf+6bEAn5q3fy7NfbiGlZzv6xRzq6V9S7ubW15eRfqCYN28dQq8OkX6sVERERKTpUMAsIiL1Ly8TvpsOK96AsnzoPBLGT4Pu48BR7S3Kk7dUvUxNTSUlJeWUT+mxHtbmrPWOVN45n/SCdAAGtB3Ag4MeZEz8GOIj40/5+CIi0jQ9NaEv36Xt58H3VvHJvecQHOikwu3hvrdXsjIjl3/eeBbDElv7u0wRERGRJkMBs4iI1J/dq2HxC/Djh95+y32vgOH3QsxZ9XI6l8fFsj3LWJC+gAXpC8guySbABDCk4xBu6XsL58edT9vQtvVybhE5dbNWZjF17iayckuIWbqAyeN6csXAGH+XJWeo6NAg/nxNEj9/bRmDn/6SorIKQoKcFJe7eWpCXy7u39HfJYqIiIg0KQqYRUSkbnk8sPVLWPI8bP8GgsJhyC9g2J0QXfcjhotdxSzetZj56fP5OvNrCsoLCAkI4ZyYcxgdP5pzY88lMkhvcxZprGatzOLRD9dS4nID3gnWHv1wLYBCZqk3ecUunA5DYVkFAMXlbgIchojgQD9XJiIiItL0KGAWEZG64SqFte/BkhcheyNEdIKxv4ezboGQ6Do9VV5ZHqkZqcxPn8+SXUsodZcS1SKK0XGjGRM/huGdhhMcEFyn5xSR+vHMZxurwuVKJS43U+duUsAs9Wbq3E24PbbGsgqP1dediIiIyClQwCwiIqen+IB3cr7v/wVF2dC+P1z5MvS9EgKCTumQ2cXZPLvnWfqV9KNNSBsA9hbtZUHGAubvnM/yvctxWzftQ9tzZfcrGRM/hkHtBxHg0I81kcasoNTF2sw8VmfmsSYzl9UZuezJL6112125JQ1cnTQnR/v60tediIiIyMnTb+IiInJq9m+Dpf+ElW9BRQl0uwBG3AcJ54Exp3Xo6Wumk1aWxl++/ws9WvVgQfoC1uZ43zKfEJXAz/v9nDHxY+jbui/mNM8lIvWj1OVmw+581mTmsdoXJqflFGF9g0bjW4UyqEsrijbvI6+k4oj9O0WHNHDF0px0ig4hq5YwWV93IiIiIidPAbOIiJw4ayHjO1j8PGycDc5A6H8dDL8H2vepk1N8v/t7Ptj8ARbLZzs+47Mdn9GvdT/uP+t+RsePJjEqsU7OIyJ1x+2xbN1XyOrMXN/I5Dw27snH5famyW3CW5AcF8WE5BiSYqMYEBtNyzDvOxwO78EMEBLoZPK4nn65FmkeJo/rqa87ERERkTqigFlERI7P44YN/+cNlrOWQ3A0jHoIhkyCiPanffg9RXv4bPtnzE6bzaaDm6qWO42TSxIu4Y+j/nja5xCRumGtJfNgSdWo5NWZeazLyqO43BvUhbcIICk2itvOSSQ5Loqk2Gg6RgUf9d0Glf1up87dRFZuCTHRIUwe11N9cKVeVf+625VbQid93YmIiIicMgXMIiJydGWFsOot78R9uTuhZRe4eCoMvAmCwk7r0Pnl+Xy580tmp81m2Z5lWCy9WvUiwARQYb1vl3dbN/N2zuPBwQ9W9WIWkYaVU1jGmsxcVmV4+yavyczjQFE5AEFOB306RXLtoFgGxEWTFBtNYpswHI6Ta11zxcAYrhgYQ2pqKikpKfVwFSJHqvy6ExEREZHTo4BZRESOVLAHvvsXLJ8JpbkQOwQufBp6jQeH85QPW+YuY2HmQmanzebrzK9xeVx0juzMXQPu4pLES3hz/Ztsy90G9tA+Huth+urpTBk25fSvS0SOqbCswjcJ36FWF5V9ah0GureLYEyvdgyIi2ZAbDQ9O0QQFODwc9UiIiIiIuJPCphFROSQvethyQuw5j3wVEDvS2H4fRA/9JQP6bEelu9Zzuzts/lixxcUuApoHdya63tez/jE8TUm6lu9bzUuj6vG/i6Pi1X7Vp3OVYlILcoq3GzcXeBrdeENlbdlF1ZNwhfXKoTk+GgmjuhCUmwU/WKiCGuhR0cREREREalJvyWIiDR31kLaV7D4Bdg2HwJDYdBEGHYXtO56ioe0bDq4idlps5mzfQ77ivcRGhDKBZ0vYHzCeIZ0HEKA48gfQe9f/n7Va71VXqTuuD2WbdmFrM7wtrhYnZnLht3VJ+ELIik2msuSOpEU552Er5VvEj4REREREZFjUcAsItJcVZTDug+8I5b3roOwdjD6cRh8K4S2OqVDZhVmMSdtDrPTZrMtbxsBJoBzYs5h8uDJnBd3HiEBIXV8ESJyuMpJ+NZk5vl6J+eyLiuPomqT8PWPieLWcxIYEBvNgLhoOh1jEj4REREREZFjUcAsItLclOTCite8PZYLdkPbXjDhReh/LQS0OOnD5ZbmMnfHXGZvn83KfSsBOKvdWTw+7HEu7Hwh0cHRdVu/iNSwv7CsalRy5Qjl/dUm4evdKZKrB8WSFBtNclwUiW3CT3oSPhERERERkaNRwCwi0lwc3AlLX4If3gRXESScB5e/AN3GwEmOXCypKCE1I5XZabP5NutbKmwF3aK7cf9Z93NxwsXEhMfUzzWINHOFZRWsy8qr0eoi86B3Ej5joHu7cM7v1Y4BsVEMiPNOwtci4NQn5hQRERERETkeBcwiIme6zBWw5HlY/zEYB/S7GobfCx2TTuowFZ4Kvtv9HbPTZjM/fT7FFcW0C23HzX1uZnzieHq07KG32IvUofIKDxv35LM6I5fVvnYXW/YdmoQvtmUIA2Kj+dnwziTFRtMvJopwTcInIiIiIiINTL+FiIiciTwe2PyZd+K+9MXQIhJG3AdDfgFRJz662FrLupx1zN4+m8+3f87+0v1EBEZwUcJFjE8Yz6D2g3A6NDpS5HR5PJa0nEJWZXiD5NWZeWzYlU+52wNA67AgkmKjuKR/RwbERpMUG0Xr8JNvaSMiIiIiIlLXFDCLiJxJyoth9duw9J+wfytExcG4P8FZN0OLiBM+zM78ncxOm82c7XPYmb+TQEcg58Wex/jE8YyKHUULp4ItkVNlrWVXXqlvZLK3b/K6rHwKyyoACAty0i8mip+P7EKSL0yObRmidwiIiIiIiEijpIBZRORMUJgNy16B71+BkgPQaSBcMxN6TwDniX2rzynJ8U7WlzabtTlrMRjO7nA2t/a7lQs6X0BkUGQ9X4TImelAUTmrM3NZk+HtmbwmM5ecQu8kfIFOQ++OkVw5MIYkX9/krm3DcWoSPhERERERaSIUMIuINGXZm2HJC7D6HXCXQY+Lva0wOo84oYn7ilxFLEhfwOy02SzdvRS3ddOrVS8eGvQQFyVcRIewDg1wESJnjiLfJHxrMvNY5QuTMw4cmoSvW9twzuvRjgFxUQyIjaZXR03CJyIiIiIiTZsCZhGRpsZa2LHIGyxv/hwCgiH5JzDsHmjb47i7uzwuFmctZnbabL7K+IpSdykx4THc2u9WxieOp2t01wa4CJGmr7zCw6Y9BVVtLtZk5rFlXwEe3yR8MdEhDIiL4qahnRkQG02/mEgiggP9W7Q0WsaYi4B/AE5ghrX2mcPW3wncA7iBQmCStXa9MWYs8AwQBJQDk621Cxq0eBERERFp1hQwi4g0FW4XrP8YFj8Pu1dBaGtIeRTOvh3C2hxzV2stq7JXMTttNnN3zCW3LJfoFtFM6DaB8YnjSW6brP6uIsfgnYSvyDsBX4Z3Er71u/Mpr/BOwtfKNwnfuH4dSI6LIik2mjaahE9OkDHGCbwIjAUygWXGmE+steurbfZfa+103/aXA38DLgJygMustbuMMf2AucCJz+YqIiIiInKaFDCLiDR2pfnww5vw3XTIy4DW3eDSZ2HADRAYcsxdt+Vuq5qsL6swi2BnMOfHnc/4xPGM6DSCQKdGU4oczlrL7qpJ+PJYk5nL2sw8CnyT8IX6JuG7ZXhnBsRFMyA2WpPwyekaAmy11qYBGGPeASYAVQGztTa/2vZhgPUtX1lt+Y9AiDGmhbW2rN6rFhERERFBAbOISOOVl+kNlVe8AWX50HkkXDIVuo8Dh+Oou+0t2stn2z9j9vbZbDywEYdxMKzjMO5Ovpsx8WMICwxrwIsQafwOVk7C5wuTV2XkkVPozeYCnYZeHSK5PLlTVZjcrZ0m4ZM6FwNkVPs4Exh6+EbGmHuAB/G2wxhdy3GuBn6oLVw2xkwCJgG0b9+e1NTU069aGp3CwkLd22ZK97750r1v3nT/m6/Gdu8VMIuINDa7V8PiF+DHD739lvteAcPvhZizjrpLQXkBX+78kk/TPmXZnmVYLP1a9+M3Z/+GixIuok3IsVtoiDQXxeUVrMvK97a6yMxjdUYu6QeKAe8kfIltwji3RxsGxEaTFBtF746RBAdqEj5pHKy1LwIvGmNuBKYAt1SuM8b0Bf4MXHiUfV8GXgYYPHiwTUlJqfd6peGlpqaie9s86d43X7r3zZvuf/PV2O69AmYRkcbA44GtX8KS52H7NxAUDkN+AcPuhOj4Wncpd5ezMHMhs7fP5uuMryn3lBMfEc+dA+5kfOJ4Okd2buCLEGlcXO4jJ+HbvPfQJHydooIZEBfNT4bEMyA2in6xUURqEj7xjywgrtrHsb5lR/MO8FLlB8aYWOAj4GfW2m31UqGIiIiIyFEoYBYR8SdXKax9D5a8CNkbIaITjP09nHULhEQfsbnHelixdwWz02Yzb+c8CsoLaBXcimt7Xsv4hPH0a9NPfWClWfJ4LNv3V07Cl8fqzFzW78qnzDcJX3RoIANio7mwT3uSYqNJiouiXUSwn6sWqbIM6G6MScAbLN8A3Fh9A2NMd2vtFt+H44EtvuXRwGzgEWvttw1WsYiIiIiIjwJmERF/KD4Ay16F71+Gon3Qvj9c+TL0vRICgmpsaq1l88HNVZP17S3eS2hAKGPixzA+cTxDOw4lwKFv59J8WGvZk19aFSSv8fVPLij1TsIXEuikf0wUNw87NAlfXCtNwieNl7W2whhzLzAXcAIzrbU/GmN+Dyy31n4C3GuMuQBwAQc51B7jXqAb8IQx5gnfsguttfsa9ipEREREpLmq10TCGHMR8A+8D8ozrLXPHLb+XOBZIAm4wVr7vm95Mt63/UUCbuAP1tp367NWEZEGsX8bLP0nrHwLKkqg2wUw4j5IOM/bALaaXYW7mLN9DrPTZrM1dysBJoCRMSN5aPBDpMSlEBIQ4qeLEKkfs1ZmMXXuJrJyS4hZuoDJ43pyxcAYcovLWePrl7w60xsqZxd45zALcBh6dYzgsgGdSPaNTO7WNpwA59EnwhRpjKy1c4A5hy17otrr+4+y39PA0/VbnYiIiIjI0dVbwGyMcQIvAmPxzoS9zBjzibV2fbXN0oGJwMOH7V6Mt4fcFmNMJ2CFMWautTa3vuoVEalX6d/B4udg42xwBkL/62D4PdC+T43NcktzmbdzHrPTZvPDvh8AGNhuIFOGTuHCLhfSMrilP6oXqXezVmbx6IdrKXG5AcjKLeGh91bz9Kc/klPkqtousW0Y53Rrw4DYKJLioumjSfhERERERET8qj5HMA8Btlpr0wCMMe8AE4CqgNlau8O3zlN9R2vt5mqvdxlj9gFtgdx6rFdEpG553LDh/2DJC5C5DIKjYdRDMGQSRLSv2qykooSvM75mdtpsFu1aRIWngq5RXfnlwF9yccLFxEbE+u8aRBpAYVkF/+//fqwKlyu5raWw3M2vL+rJgNho+sVEERWiSfhEREREREQak/oMmGOAjGofZwJDT/YgxpghQBBwxIzYxphJwCSA9u3bk5qaekqFSuNWWFioe9tMNdV776woocOe+cRm/h8hpXsoCe5AZrdJ7O44Bo8zGFZswG3Xsbl0M8uLlrO6eDVltoxoZzTnhZ/H4LDBxATGYA4Yth7Yyla2+vuS/KKp3n85MSUVltX73CzbW8GabDcuT+3blbo89CETV2YmKzMbtkZpePp7LyIiIiLS9DTqWaGMMR2BfwO3WGuP+NXTWvsy8DLA4MGDbUpKSsMWKA0iNTUV3dvmqcnd+4I98N2/YPlMKM2F2CEwYiohvcbT3eGkm7X8uP9HZqfN5rPtn7G/dD8RgRGM7zae8QnjGdR+EE6H3upfqcndfzmuwrIK5m/Yy+w1u0ndnE15hYf2kS24aVgcn67ZRU5h+RH7xESH6OugGdHfexERERGRpqc+A+YsIK7ax7G+ZSfEGBMJzAZ+a61dWse1iYjUnb3rvW0w1rwHngrofSkMvw/ivW/aSM9PZ/b22cxJm8OO/B0EOgI5N/ZcxieO59zYc2nhbOHnCxCpPwWlLuZv2Mfstbv5ulqofOOQeMYndWRQfEscDkNyXHSNHswAIYFOJo/r6cfqRURERERE5HjqM2BeBnQ3xiTgDZZvAG48kR2NMUHAR8Cb1tr3669EEZFTZC2kfQWLX4Bt8yEwFAZNhGF3Qeuu7C/Zz+cb3mJO2hzW5KzBYBjcYTAT+07kgs4XENUiyt9XIFJvjhYq3zQ0nvH9O3KWL1Su7oqBMQBMnbuJrNwSYqJDmDyuZ9VyERERERERaZzqLWC21lYYY+4F5gJOYKa19kdjzO+B5dbaT4wxZ+MNklsClxlj/p+1ti9wHXAu0NoYM9F3yInW2lX1Va+IyAmpKId1H3hHLO9dB2HtYPTjMPhWigODmZ8+n9krp7J011Lc1k3Plj15cNCDXJxwMR3COvi7epF6U1Dq4ssNe5m9Zg/fbPGGyh0ig48ZKh/uioExXDEwRm0SREREREREmpB67cFsrZ0DzDls2RPVXi/D2zrj8P3+A/ynPmsTETkpJbmw4jVvj+WC3dC2F0x4EVffK1mybwWfLn+G1IxUSipK6BTWiZ/3+znjE8bTrWU3f1cuUm/yS118uX4vc9bu5pvNOZS7PXSMCuanQzszPqkDA+OOHyqLiIiIiIhI09aoJ/kTEfG7gzth6Uuw8t9QXggJ52Eve57VUW34dPts5n14MQfLDhLVIorLEi9jfOJ4ktsl4zAOf1cuUi+OGioP68z4pI4MjItWqCwiIiIiItKMKGAWEalN5gpY8jys/xiMA/pdTVr/K/m0YAtz1v6VrMIsWjhbcH7c+YxPHM/ITiMJdAb6u2qRepFXcihUXrjFGyp3igrm5uGduaS/QmUREREREZHmTAGziEgljwc2f+aduC99MbSIZN/Q2/msTRyzdy9iw+Jf4zAOhnUcxt3JdzMmfgxhgWH+rlqkXuSVuPiiKlTOxuW2dIoK5mfDO3NJUkeSYxUqi4iIiIiIiAJmERFwlcCq/8LSf8L+rRREx/HlsJ8x2xbw/d652L2Wvq378uuzf83FCRfTJqSNvysWqRe1hcox0SHcMrwL45M6khwXjTEKlUVEREREROQQBcwi0nwVZsOyV2DZDMqL97Mwtg+zB47j6/xtlO9NJS4ijl8M+AXjE8bTJaqLv6sVqRd5xS7mrd/DnLW7WbQ1pypUnjiiC5f0V6gsIiIiIiIix6aAWUSan+zNsOQFPKvfYUUgzO7Uk3nOthRUFNKqdC/X9LiG8Ynj6d+mv4I1OSNVhsqz1+7m22qh8s9HJnBJ/44MiI3S176IiIiIiIicEAXMItI8WAs7v4XFz7Np+3xmR0byWZd49njKCHGWMSZ+DOMTxzOs4zACHPrWKGeevGIXc30jlQ8Plcf370iSQmURERERERE5BUpRROTM5nbB+o/ZvfhZZhfvZHZEJFtjOxJgnIzoNJRfJYwnJS6F0MBQf1cqUudyi8uZ9+PeqpHKFR5LbMsQbvWNVFaoLCIiIiIiIqdLAbOInJlK88lb9jJz177ObGcZPwQHQ3A0yW2S+G3XyxjXZRwtg1v6u0qROldbqBzXKoTbRnlHKvePUagsIiIiIiIidUcBs4icUUoPbCN14dPM3r2ERcEBVIQZEkPiua/ndVySOJ7YiFh/lyhS5w4Wlft6Ku9h8WGh8qX9O9EvJlKhsoiIiIiIiNQLBcwi0uS5PW6+W/82s1e/yvzyfRQ5HLQLDeWmuDGMT7qVXq16KVyTM87BonLm/uidqG/Jtv1UeCzxrUK5fVQi4/t3VKgsIiIiIiIiDUIBs4g0SdZa1uf8yKcrX+Lz3d+Sg5twj4exoZ259Ky7GNz1EpwOp7/LFKlTB4rKmecLlRdv24/bY+ncOpQ7zvWGyn07KVQWERERERGRhqWAWUQapezibJ7d8yz9SvrRJqRN1fKM/Aw+3TqLOZvfZ0fZAQKtZVS5ZXznsZw38lFahLf3Y9Uide+Ab6TynMNC5UkKlUVERERERKQRUMAsIo3S9DXTSStLY/rq6dw14C7m7pjL7K0fs+bAegAGl5Ryi6MlYwfdTdSAGyEgyM8Vi9Sd/YVlzP1xL3PW7mZJmjdU7tI6lF+cm8glCpVFRERERESkEVHALCKNy9TuZJfm8HFsJ6zDwf82vsP/Nr6Dxxh6lFfwq8ICLmlzFh0ufBASzgOFbHKGOFqofOd53lC5T0eFyiIiIiIiItL4KGAWkcalaB8vtW6JyxekeYBe5eX8MfsA3fteD5ffA+37+LdGkTqyv7CMz33tL5amHcDtsSS0CVOoLCIiIiIiIk2GAmYRaVT2OR18EBGOpzJUM4btgYG09Ljhihf9W5xIHcgpLOPzdZWh8n48FhLbhHHXeV25pH9HeneMUKgsIiIiIiIiTYYCZhFpNKy1/KJ9u0Phso8Hw/ToKKb4qS6R03W0UPnulG6MT+pIrw4KlUVERERERKRpUsAsIo2CtZapy6aytcWRk/W5HIZVLVr4oSqRU5dd4Gt/sWY33233hcptw7jn/G5c0l+hsoiIiIiIiJwZFDCLiN9Za/n7ir/z7w3/5qd5+fz6QC6K3aQpqi1U7to2jHvP78YlSR3p2V6hsoiIiIiIiJxZFDCLiN+9sPJ5XvvxNa7PL+DXBWW1h8th7Rq6LJETsq+glLnr9jB77W6+336gRqg8PqkTPdqHK1QWERERERGRM5YCZhHxq5dW/ZOX177C1QWFPNbjJsyFT4MvjEtNTSUlJcW/BYrUYl9BKZ+v28PsNbv5fscBrIVu7cK5d3R3xvfvqFBZREREREREmg0FzCLiNzNWv8w/V7/EhIJCnug1EceYJ6rCZZHGprZQuXu7cH45ujvjkzrSo32Ev0sUERERERERaXAKmEXEL95YO5N/rHqe8YVF/L++d+BIeVThsjQ6+/JL+czX/mKZQmURERERERGRIyhgFpEG99aPbzLth79zUWERT/e/C+e5k/1dkkiVvfmlfLZ2N3PW7mHZTm+o3KN9OPeP8ba/6K5QWURERERERKSKAmYRaVDvrn+LZ5ZP5YKiYv6YfD8BI+/3d0kitYbKPdtH8MCYHoxP6kC3dgqVRURERERERGqjgFlEGsz7G97m6WXPkFJUzF/OeojAYXf7uyRpxvbklfLZut3MWbub5TsPKlQWEREREREROQUKmEWkQcza9D9+//0fGVVcwl8H/ZrAob/wd0nSDO3JK2XO2kOhMkCvDhH86oIeXNK/I93ahfu5QhEREREREZGmRQGziNS7Tzd/wBNLf8+wklL+fvZvCRr8c3+XJM3I7rwSPlu754hQ+aGxPbgkqSNd2ypUFhERERERETlVCphFpF59vuUjfrv4d5xdWso/hj5Bi4E3+7skaQZ255Uwxxcqr1CoLCIiIiIiIlJvFDCLSL35cusnPPLtEySXlfH8sP9HSPKN/i5JzmC1hcq9O0by8IXe9heJCpVFRERERERE6pwCZhGpF19t/T8mL/ot/cvK+eeIpwntf52/S5Iz0K7ckqqeyj+k5wLeUHnyuJ5c3K+DQmURERERERGReqaAWUTq3Dfb5vDgosfoXV7OSyP/RFi/q/xdkpxBsnJL+Gztbmav3c1KX6jcxxcqX9K/IwltwvxboIiIiIiIiEgzooBZROrU4m2f8auFv6G7y8X0c/5CeJ8J/i5JmphZK7OYOncTWbklxCxdwORxPTk7odURoXLfTgqVRURERERERPxNAbOI1Jnvtn3OLxf+mgRXBa+Mmkpkr8v8XZI0MbNWZvHoh2spcbkB72jlX723Cmu96ytD5fH9O9JFobKIiIiIiIiI3ylgFpE6sTxtLvctnExcRQWvnPc3onpc7O+SpAn6y9yNVeFyJWshMjiAT+49R6GyiIiIiIiISCOjgFlETtuqtLnc/c3DdKxwM+O8Z2nZfZy/S5ImJr/UxTvfp7Mrt7TW9QWlFQqXRURERERERBohBcwiclrWpM3jzm8epr3bzYyUf9C621h/lyRNSObBYl77dgfvLsugsKyCoAAH5RWeI7brFB3ih+pERERERERE5HgUMIvIKfsxbR53fv0grTweZqQ8R9uuF/i7JGkiVmfk8srCND5btweAS5M6cseoRLbuK6zRgxkgJNDJ5HE9/VWqiIiIiIiIiByDAmYROSUbt81j0jcPEumxvHr+C7RPHO3vkqSR83gsX27Yy4yF2/l+xwEiWgRw2zkJTBzRpWqEcr+YKACmzt1EVm4JMdEhTB7XkysGxvizdBGRemeMuQj4B+AEZlhrnzls/Z3APYAbKAQmWWvX+9Y9CtzmW/dLa+3chqxdRERERJo3BcwictI2b5vLHd88RKi1vDr6RTompPi7JGnESsrdvP9DJjMXbWd7ThEx0SFMGd+b68+OIyI48IjtrxgYwxUDY0hNTSUlJaXhCxYRaWDGGCfwIjAWyASWGWM+qQyQff5rrZ3u2/5y4G/ARcaYPsANQF+gE/ClMaaHtbbmjKkiIiIiIvVEAbOInJRtWz/njoUPE2Rh5uh/EtPlPH+XJI1UdkEZ/16yg38v3cnBYhcDYqN4/icDubhfBwKcDn+XJyLSmAwBtlpr0wCMMe8AE4CqgNlam19t+zDA+l5PAN6x1pYB240xW33HW9IQhYuIiIiIKGAWkRO2fctn3L7wYRzG8OroF4lTuCy12LK3gBkLt/PRqixcbg8X9G7PHaMSObtLS4wx/i5PRKQxigEyqn2cCQw9fCNjzD3Ag0AQUNmbKgZYeti+R/QVMsZMAiYBtG/fntTU1LqoWxqZwsJC3dtmSve++dK9b950/5uvxnbvFTCLyAlJ3zKH2xdOxmMMr41+kS6dFS7LIdZaFm/bzysL00jdlE1woIPrBsdy68gEEtuG+7s8EZEzgrX2ReBFY8yNwBTglpPY92XgZYDBgwdbtSA6M6m9VPOle9986d43b7r/zVdju/cKmEXkuDI3z+a2hb+m3OFg5pgXSYw/198lSSNRXuHh0zW7mLFwO+t359MmPIgHx/bgp8M60yosyN/liYg0FVlAXLWPY33LjuYd4KVT3FdEREREpE4pYBaRY9q98f+4/dvfUOxw8uroF+mucFmAvGIX//0+ndcXb2dvfhnd24Xz56v7MyE5huBAp7/LExFpapYB3Y0xCXjD4RuAG6tvYIzpbq3d4vtwPFD5+hPgv8aYv+Gd5K878H2DVC0iIiIiggJmETmGvRs+5tbFj5LvdPLK+S/Qq7PC5eYu40Axry7aznvLMygudzOyW2ueuTqJlB5t1V9ZROQUWWsrjDH3AnMBJzDTWvujMeb3wHJr7SfAvcaYCwAXcBBfewzfdu/hnRCwArjHWuv2y4WIiIiISLOkgFlEapW9/kNuW/xbDgYE8PLoF+gbr57LzdkP6QeZsTCNz9ftwWEMlw/oxG2jEujbKcrfpYmInBGstXOAOYcte6La6/uPse8fgD/UX3UiIiIiIkengFlEjpCz7n1uW/o42QEB/Ov850lSuNwsuT2WL9bv5ZWFaazYeZDI4AAmnduViSO60CEq2N/liYiIiIiIiEgjoIBZRGo4sPZd7lj6JHsCA/nn+c+R3DnF3yVJAysur+D9FZm8umg7O/cXE9cqhCcv68N1g+MIa6EfGyIiIiIiIiJyiJICEamSt/otJn3/FBmBgbyY8iyDO5/v75KkAe3LL+WNJTv4z9J08kpcDIyP5jcX9eLCPu0JcDr8XZ6IiIiIiIiINEIKmEUEgLwf3uSO5X9ge1AQz6f8naFdxvi7JGkgG/fkM2Phdj5ZtQuXx8OFfdoz6dxEBnVu5e/SRERERERERKSRU8AsIhSsmMmdK/7M1qAg/nHe3xjR5QJ/lyT1zFrLwi05vLIwjYVbcggJdHLDkDhuHZlAlzZh/i5PRERERERERJoIBcwizVzR969w18qpbAxuwd/PncaohLH+LknqUVmFm09W7eLVRdvZuKeAthEtmDyuJzcNjSc6NMjf5YmIiIiIiIhIE6OAWaQZK/7uJe5e9TfWBQczbdSfSUkY5++SpJ7kFpfz1nfpvLF4B/sKyujZPoKp1yRxeXInWgQ4/V2eiIiIiIiIiDRRCphFmqmSxc9x35rnWRUSzJ9H/pELEi/xd0lSD3buL+LVRdv53/JMSlxuRnVvw7RrBzCqexuMMf4uT0RERERERESaOAXMIs1Q2cK/cv+PL7EsJIQ/jniKi7pd5u+SpI6t2HmAV77Zztz1ewhwGCYkx3D7qAR6dYj0d2kiIiIiIiIicgap14DZGHMR8A/ACcyw1j5z2PpzgWeBJOAGa+371dbdAkzxffi0tfaN+qxVpLko//oZHlg/g6WhITw1/P9xafcr/F2S1BG3xzL3xz28sjCNlem5RIUEcndKV24Z3oV2kcH+Lk9EREREREREzkD1FjAbY5zAi8BYIBNYZoz5xFq7vtpm6cBE4OHD9m0FPAkMBiywwrfvwfqqV+SMZy2ur57moY2vsygslN8NfZwJPa7yd1VSB4rKKnhveQYzv91OxoES4luF8vsJfblmUCyhQXqjioiIiIiIiIjUn/pMHoYAW621aQDGmHeACUBVwGyt3eFb5zls33HAF9baA771XwAXAW/XY70iZy5rcX35OyZv+Q+pYaFMGfIYV/e6zt9VyWnak1fK64t38N/vdpJfWsGgzi357SW9GdunA06H+iuLiIiIiIiISP2rz4A5Bsio9nEmMPQ09o2po7pEmhdrqZj7GI9ue5f54WE8MvjXXN/7J/6uSk7D+l35zFiYxierd+Gxlov6deD2UYmcFd/S36WJiIiIiIiISDPTpN87bYyZBEwCaN++Pampqf4tSOpFYWGh7u2pspaErS/zz/LvmBsexhXRVxCTHddkPp+694dYa1mb4+bzHS7W7/fQwgnnxwUwtnMg7UILyE9bTWqav6usW7r/zZfuffOley8iIiIi0vTUZ8CcBcRV+zjWt+xE9005bN/Uwzey1r4MvAwwePBgm5KScvgmcgZITU1F9/YUeDx4Pv0Vj5ctZU5EOPcPvJ/bk273d1UnRfceSl1uPl6VxYyF29myr5j2kS34zUUJ3DgknqjQQH+XV690/5sv3fvmS/deRERERKTpqc+AeRnQ3RiTgDcwvgG48QT3nQv80RhT+X7vC4FH675EkTOUx43n4/v4fdZnfBIRzt0D7m5y4XJzd6ConLeW7uSNJTvJKSyjd8dI/nbdAC5N6kRQgMPf5YmIiIiIiIiIAPUYMFtrK4wx9+INi53ATGvtj8aY3wPLrbWfGGPOBj4CWgKXGWP+n7W2r7X2gDHmKbwhNcDvKyf8E5HjcFdgZ93FH3d9wQeREUzqP4m7ku/yd1VygrbnFPHqojTeX5FJqcvDeT3aMuncREZ0bY0xmrhPRERERERERBqXeu3BbK2dA8w5bNkT1V4vw9v+orZ9ZwIz67M+kTOO24X9cBJ/3v0V70ZFcGu/W7l34L3+rkqOw1rLsh0HeWVhGl9u2Eugw8EVAztx+6hEerSP8Hd5IiIiIiIiIiJH1aQn+RORairKse//nL/uW8RbUZHc3OdmHjjrAY16bcQq3B4+W7eHGQvTWJ2ZR8vQQO49vxs3D+9Mu4hgf5cnIiIiIiIiInJcCphFzgQVZdh3f8Y/cpbwRnQUN/a6kcmDJytcbqQKyyp45/t0Xvt2B1m5JSS0CeOpK/pxzVmxhAQ5/V2eiIiIiIiIiMgJU8As0tS5SuDdn/LPnGW82jKK63pcxyNDHlG43Ajtyi3h9cU7ePu7dArKKhjSpRVPXtaHC3q3x+HQ/RIRERERERGRpkcBs0hTVl4M7/yEfx1YyfSWUVzV/Sp+O+y3CpcbmXVZebyyMI3Za3ZjgYv7deCOUYkMiIv2d2kiIiIiIiIiIqdFAbNIU1VWCG/fwKsH1/BCqygu73o5Tw5/Eodx+LsyATwey1eb9vHKwjSWph0gLMjJLSO6MHFEF+Jahfq7PBERERERERGROqGAWaQpKs2Ht67ljfwNPNsqiosTLub3I36vcLkRKHW5+WhlFjMWprEtu4iOUcE8dkkvbhgST2RwoL/LExERERERERGpUwqYRZqaklz4z9W8VbiFaa2iuLDzhfzxnD/idGhyOH/aX1jGv5fu5N9LdrK/qJy+nSJ59vpkxid1JNCp4F9EREREREREzkwKmEWakuID8O8rea9oO8+0jmJ03GieOfcZAhz6q+wvW/cV8uqi7Xz4QyZlFR5G92rH7aMSGJ7YWr2wRUREREREROSMp1RKpKko2g9vTuDDknSeah3FebHnMe28aQQ61HahoVlrWZp2gBkL05i/cR9BAQ6uPiuG285JoFu7CH+XJyIiIiIiIiLSYBQwizQFhfvgzQl8XLqL37WOZmTMCP6W8jcCnQqXG5LL7WHO2t28sjCNdVn5tAoL4v4x3bl5eGfahLfwd3kiIiIiIiIiIg1OAbNIY1ewB964jNnl2TzeOoqhHYfybMqzBDmD/F1Zs5Ff6uKd79N5/dsd7MorJbFtGH+8sj9XnRVDcKB6X4uIiIiIiIhI86WAWaQxy8uCNy5jbsUBHmsdyeAOg3lu9HMEBwT7u7JmIfNgMa99u4N3l2VQWFbBsMRWPHVFP87v2Q6HQ/2VRUREREREREQUMIs0Vgd3whuXMd8W8ps2USS3TeaF0S8QEhDi78rOeGsyc3ll4XbmrN0NwKVJHbn9nET6x0b5uTIRERERERERkcZFAbNIY3RgO7xxGakU83DrSPq26cs/L/gnoYGh/q7sjOXxWOZv3McrC9P4fvsBIloEcOvILkwcmUBMtEJ9EREREREREZHaKGAWaWz2b4PXL2WRo5wHW0fRq1Uvpl8wnbDAMH9XdkYqKXfzwQ+ZzFy0nbScImKiQ5gyvjfXnx1HRLAmURQRERERERERORYFzCKNSfYmeONyFgd4uL91FN2iuzF97HQigiL8XdkZJ7ugjH8v2cG/l+7kYLGLpNgonvvJQC7p14EAp8Pf5YmIiIiIiIiINAkKmEUai73r4c3L+T7QyS9bR9AlKoGXx75MVAv1/a1LW/YWMGPhdj5alYXL7WFMr/bcMSqBIQmtMEYT94mIiIiIiIiInAwFzCKNwe418OYEVgQHc2/rMOIi4njlwleIDo72d2VnBGsti7ft55WFaaRuyqZFgINrB8Vy2zkJJLYN93d5IiIiIiIiIiJNlgJmEX/L+gH+fSWrQsO5u1UoHcI78sqFr9AquJW/K2vyyis8fLpmFzMWbmf97nzahAfx4Nge/HRYZ1qFBfm7PBERERERERGRJk8Bs4g/ZSyD/1zF2vBo7moVStvQtsy4cAZtQtr4u7ImLa/Exdvfp/P6tzvYk19Kt3bh/Pnq/kxIjiE40Onv8kREREREREREzhgKmEX8ZecSeOta1ke25hetQoluEc2MC2fQLrSdvytrsjIOFDPz2+28tyyDonI3I7q25k9X9ee8Hm1xONRfWURERERERESkrilgFvGH7Qvhv9ezKboDk1oGExEUwavjXqVDWAd/V9YkrUw/yIyF2/ls3W4cxnDZgE7cPiqBvp00QaKIiIiIiIiISH1SwCzS0LZ9BW//hC2t47gjugXBAcHMGDeDTuGd/F1Zk+L2WL5Yv5cZC9NYvvMgEcEB3HFuIhNHdKFjVIi/yxMRERERERERaRYUMIs0pC1fwDs3kdY2kdujgwhwBDBz3EziIuL8XVmTUVxewfsrMpm5aDs79hcT2zKEJy7tw3VnxxHeQt/SREREREREREQaktIYkYaycQ787xZ2tO/BbVEBGAyvjnuV+Mh4f1fW6MxamcXUuZvIyi0hZukCJo/ryYiurXljyQ7e+i6d3GIXyXHRvDiuF+P6tifA6fB3ySIiIiIiIiIizZICZpGGsP5jeP9WMjr247YoJx7rZua4V0mISvB3ZY3OrJVZPPrhWkpcbgCyckt46L3VgMUDXNinPXeMSmRQ55YYo4n7RERERERERET8SQGzSH1b9wF8cAdZscncFumg3F3Gq+NepWt0V39X1ihNnbupKlyu5LaWsCAns385ii5twvxUmYiIiIiIiIiIHE4Bs0h9Wv0OzLqLPfFnc1uEodBVwKsXvkqPlj38XVmjtSu3pNblxeVuhcsiIiIiIiIiIo2MGpeK1Jcf/g0f3cneLsO4NdJJXnk+r4x9hd6te/u7skatZWhgrcs7RYc0cCUiIiIiIiIiInI8JxQwG2POMcb83Pe6rTFGjWNFjmX5TPjkXnISR3F7hIMDZQeZPnY6fdv09XdljZa1ln99vY0DxS4Ob60cEuhk8rie/ilMRESkARhjLjLGbDLGbDXGPFLL+geNMeuNMWuMMfONMZ2rrfuLMeZHY8wGY8xzRpMUiIiIiEgDOm7AbIx5EvgN8KhvUSDwn/osSqRJ++5f8Omv2N9tDLdFGPaWZPPSBS8xoO0Af1fWaLncHh77aB1/+mwjlyZ15C9XJxHjG7EcEx3Cn67qzxUDY/xcpYiISP0wxjiBF4GLgT7AT4wxfQ7bbCUw2FqbBLwP/MW37whgJJAE9APOBs5roNJFRERERE6oB/OVwEDgBwBr7S5jTES9ViXSVC1+Aeb9loM9L+KO0HJ2Fezmnxf8k4HtBvq7skYrv9TFPW/9wMItOdxzflceGtsTh8Nw7eA4UlNTSUlJ8XeJIiIi9W0IsNVamwZgjHkHmACsr9zAWvtVte2XAj+tXAUEA0GAwTsYZG8D1CwiIiIiApxYwFxurbXGGAtgjNEsWyK1WfhXmP978npfyqSQMtLzM3hhzAuc3eFsf1fWaGUeLObW15eRll3EX65O4rqz4/xdkoiIiD/EABnVPs4Ehh5j+9uAzwCstUuMMV8Bu/EGzC9YazfUV6EiIiIiIoc7kYD5PWPMv4BoY8wdwK3AK/VblkgTYi18/WdI/RP5/a5kUotituWm8fzo5xnWcZi/q2u0VmfkctsbyymrcPPGrUMY2a2Nv0sSERFp9IwxPwUG42uDYYzpBvQGYn2bfGGMGWWtXXjYfpOASQDt27cnNTW1wWqWhlNYWKh720zp3jdfuvfNm+5/89XY7v0xA2bfBCHvAr2AfKAn8IS19osGqE2k8bMWFjwNC6dRmHQdd7UoYvOBLfzj/H8wMmakv6trtD5ft4cH3l1Jm/AWvH3HULq3V9cdERFp1rKA6m/jifUtq8EYcwHwW+A8a22Zb/GVwFJrbaFvm8+A4UCNgNla+zLwMsDgwYOtWlCdmdRerPnSvW++dO+bN93/5qux3ftjBsy+1hhzrLX9AYXKItVZC188AYufo3jgTdwVWMD6nA38NeWvnBt7rr+ra5SstcxYuJ0/fraB5LhoXvnZYNqEt/B3WSIiIv62DOhujEnAGyzfANxYfQNjzEDgX8BF1tp91ValA3cYY/6Et0XGecCzDVG0iIiIiAicWIuMH4wxZ1trl9V7NSJNhbXw+aPw3UsUD/45dztzWZu9lqnnTWV0/Gh/V9coVbg9PPnJj7z1XTrj+3fkr9cNIDjQ6e+yRERE/M5aW2GMuReYCziBmdbaH40xvweWW2s/AaYC4cD/vG8yJN1aeznwPjAaWIt3wr/PrbX/54/rEBEREZHm6UQC5qHATcaYnUAR3pER1lqbVK+ViTRWHg/MeRiWv0rp0F/wS8d+Vu5dyTOjnmFs57H+rq5RKih1ce9/V/L15mzuSunK5At74nAYf5clIiLSaFhr5wBzDlv2RLXXFxxlPzfwi/qtTkRERETk6E4kYB5X71WINBUeD3x6P/zwJmUj7uV+cvh+9zL+cM4fuDjhYn9X1yhl5ZZw2+vL2LqvkGeu6s8NQ+L9XZKIiEidM8YU4B1BfMQqvIMzIhu4JBERERGRBnHcgNlau9MYMwAY5Vu00Fq7un7LEmmEPG74+B5Y/Tblox7kV3Yvi7MW8/sRv+eyrpf5u7pGaW1mHre+sYzScjev/3wI53Rv4++SRERE6oW1VjPWioiIiEiz5DjeBsaY+4G3gHa+P/8xxtxX34WJNCruCvhwEqx+G1fKIzxkcliYtZAnhj/Bld2v9Hd1jdK8H/dw3b+WEOR08MHdIxQui4jIGc0Y0+pYf/xdn4iIiIhIfTmRFhm3AUOttUUAxpg/A0uA5+uzMJFGw+36/+zdd3hUxf7H8ffspieQQEJC701qQi8WFBG8KjbsDQv2cvX+sGK5tqvi9dpFbIh6FRuIqCAqQS+g0kLvnVCSUEJ6Ntn5/bEhJiShZrNJ9vN6nn32nDlz5nx3h8Dhmzkz8NWNsPIbXIMf44HCZBK3JfJw34e5pP0lvo6u2rHW8v6czTz93Uq6NY3i3Wt70aBOsK/DEhER8baFeKbIKG+RAQu0rtpwRERERESqxtEkmA1QWGK/kPJvnEVqn4I8+OJ6WPMdBWc9zcOurczcMpP7e9/PFR2v8HV01U5BoZsnp61k4rwtDOvckP9cFk9okNPXYYmIiHidtbaVr2MQEREREfGFo0kwfwD8YYyZXLR/AfCe1yISqS5cufD5NbDuRwrPfoFH8zczffN07ut5H9d0usbX0VU7mXkF3PnfRSSuSeWWU1vzwLCOOBz6XZSIiPgfY0w9oB0QcrDMWvur7yISEREREfGeo1nk7yVjTCJwclHR9dbaxV6NSsTX8rPhsyth4yzc5/yHx/M2Mm3jNO5OuJvru1zv6+iqnZ3pOVz/wXzWpWTy7IVdubJvc1+HJCIi4hPGmJuAe4CmQBLQD8/0cmf4MCwREREREa85YoLZGNMPWGGtXVS0X9cY09da+4fXoxPxhfws+O9lsPl/uIe/zpO5G/hmwzfc1v02RnUb5evoqp3lyenc+OF8svIK+WBkb05t38DXIYmIiPjSPUBv4Hdr7enGmI7Asz6OSURERETEaxxHUectILPEfmZRmUjtk5cBH4+ALXOwF77Ns/lb+GrdV4zqOorbut/m6+iqnZ9W7ubSt+cR4HDw1W0DlFwWERGBXGttLoAxJthauxro4OOYRERERES85qgW+bPW2oM71lq3MeZozhOpWXLTPcnl5IXYi97lhZz1TFozies7X89dCXdhjOYTLumDOZt4atpKujSJ5N3rehFbJ+TIJ4mIiNR+240xUcAUYKYxZh+wxacRiYiIiIh40dEkijcaY+7mr1HLtwMbvReSiA/k7IOPLoJdy7CXTOCl7HV8vOpjrj7pau7tea+SyyUUFLp5atpKPpy3haGd43j5sgRCg5y+DktERKRasNZeWLT5hDFmFhAJTPdhSCIiIiIiXnU0U2TcCgwAkotefYGbvRmUSJXK2gMfnge7l2MvnchruZuZsGICl3e4nPt736/kcglZeQXc/NFCPpy3hVGntOLNq3oquSwiIlKCMaafMaYOgLV2NpAIJPg0KBERERERLzriCGZrbQpweRXEIlL1MlNh4vmwdwNc/ilvZa3lnWXvMKL9CB7q+5CSyyXsSs/lhgnzWbM7g6cv6MLV/Vr4OiQREZHq6C2gR4n9zHLKRERERERqjQpHMBtjRhlj2hVtG2PM+8aYdGPMUmOMbpCl5svYBRPOgb0b4cpJjM/ZyFtL3uKCthfwaL9HcZijGeDvH1bsSOeCN+awdW82713XS8llERGRipVZv4Sjm5ZORERERKRGOlwG7R5gc9H2FUB3oDVwH/CKd8MS8bIDOzzJ5fTtcPWXfJC9idcWv8a5rc/lif5PKLlcws+rdnPJuHk4DHx5W38GdYj1dUgiIiLV2UZjzN3GmMCi1z1o/RIRERERqcUOl0UrsNa6irbPBSZaa/dYa38Cwr0fmoiX7N8GH/wNMnbDNV8zMWsjLy18ibNbns1TA5/C6dCcwgd9OHczoyYuoE2DCKbcMZCODev6OiQREZHqruT6JdvR+iUiIiIiUssd7nE9tzGmEbAPGAw8U+JYqFejEvGWfZs9C/rlpMO1U/g0cx1jF4xlSIshPHvKswQ49AQrQKHb8vR3K/lgzmaGdIrjlcvjCQvSdyMiInIkWr9ERERERPzN4UYwPwYswDNNxlRr7QoAY8xp6DE/qYn2bPCMXM7LgOu+4YvsTTz7x7Oc3ux0nj/1eSWXi2TlFXDLRwv4YM5mbjy5FeOu7qnksoiIyFEyxrQ3xvxsjFletN/NGDPG13GJiIiIiHhLhQlma+00oAVwkrV2VIlDC4DLvB2YSKVKXetJLhfkwnXfMjlrM0/Oe5JTmpzCi6e9SKAj0NcRVgu7D+Ry6dvz+GV1Ck+d35lHz+2E02F8HZaIiEhN8g7wEOACsNYuRSOaRURERKQWO+ywRGttAZ4pMkqWZXk1IpHKlrIKPhzu2R75Hd9mrOfxuY8zoPEA/nP6fwhyBvk2vmpi5Y4D3PjhfA7kuHjvut6c3lGL+YmIiByHMGvtn8aU+gVtga+CERERERHxtsNNkXHCjDHDjDFrjDHrjTEPlnM82Bgzqej4H8aYlkXlgcaYD40xy4wxq4wxD3kzTqnFdi2DCeeAccDI7/g+cxNj5oyhT8M+vHL6KwQ7g30dYbUwa00Kl4ybi7Xwxa0DlFwWERE5fmnGmDaABTDGjAB2+jYkERERERHv8drEqsYYJ/AGMATPCtrzjTFTrbUrS1S7EdhnrW1rjLkceB7P9BuXAMHW2q7GmDBgpTHmU2vtZm/FK7XQjiT46AIIDIPrvuXHjA08/L+HSYhN4NUzXiUkIMTXEVYLH83bzONTV3BSo7q8d11vGkbqexERETkBdwDjgY7GmGRgE3CVb0MSEREREfGe4xrBbIzpeBTV+gDrrbUbrbX5wGfA+YfUOR/4sGj7S2Cw8TxPaIFwY0wAEArkAweOJ1bxU9sXeKbFCKoD13/PL1lbeODXB+jWoBtvDn6TsMAwX0foc4Vuy1PTVvLoNys4o2Msn9/SX8llERGRE1R073sm0ADoCJwGnOzbqEREREREvOd4RzD/CDQ/Qp0mwLYS+9uBvhXVsdYWGGPSgWg8yebz8TxOGAbca63de+gFjDE3AzcDxMXFkZiYeMwfRKq/zMzMY+rbuumr6Lb0n7gCI0k66TEWzpvGu6nv0iyoGVcEX8Gfc/70XrA1RF6BZdzSPBanFDKkRQBXNM9k/rz/+TqsMo6176V2Uf/7L/W9/6rJfW+MqYtn9HIT4Bvgp6L9fwBLgU98F52IiIiIiPdUmGA2xrxa0SEgyivR/KUPUAg0BuoBvxljfrLWbixZyVo7Hs8jiPTq1csOGjTIy2GJLyQmJnLUfbt5DnzyFEQ1IeC6b3FnbOL9X/5Fx+iOjD9rPHWD6no11pog5UAuN364gBWp2TxxXidGDmzl65AqdEx9L7WO+t9/qe/9Vw3v+4/wLI49DxgFPILnvvlCa22SD+MSEREREfGqw41gvh7PiIu8co5dcRRtJwPNSuw3LSorr872oukwIoE9wJXAdGutC0gxxswBegEbEanIxkT47+UQ1Ryum8rvmZu5Z9Y9tIlqw9tD3lZyGVi96wA3fDCf/Tku3rm2F4NPivN1SCIiIrVFa2ttVwBjzLt4nsRrbq3N9W1YIiIiIiLedbgE83xgubV27qEHjDFPHEXb84F2xphWeBLJl+NJHJc0FbgOz0iPEcAv1lprjNkKnAF8ZIwJB/oBLx/FNcVfrf8JPrsK6reBa79hfuZm7vr5LprXbc74IeOJDI70dYQ+l7gmhTv/u5jwYCef39KfLk30nYiIiFQi18ENa22hMWa7kssiIiIi4g8Ol2AeAZR7U2ytPeIz9UVzKt8JzACcwPvW2hXGmCeBBdbaqcB7eJLI64G9eJLQAG8AHxhjVuB5tPADa+3So/1Q4mfWTIfPr4EGHeCab1iUuYU7fr6DJhFNeGfIO9QLqefrCH3u49+38PjUFXSIq8N7I3vRKDLU1yGJiIjUNt2NMQcXpTZAaNG+Aay1Vo9SiYiIiEitdLgEc0R5C+sdC2vt98D3h5Q9VmI7F7iknPMyyysXKWPVt/DF9dCwK1zzNUsyt3LbT7cRFxbHu0PfJTo02tcR+pTbbfnXD6t457dNnNExllevSCAi+HjX9hQREZGKWGudvo5BRERERMQXHIc5NuXghjHmK++HInKMln8Nn18HjRPg2imsyNrBrTNvJTo0mnfPepeY0BhfR+hTOfmF3PbJQt75bRPX9W/B+Gt6KrksIiIiIiIiIiKV6nDZJlNiu7W3AxE5Jks/h8m3QLO+cNUXrMrczqiZo4gMjuT9oe8TF+7fi9elZOQy6sMFLE1O5/HzOnH9wCPOaiMiIiIiIiIiInLMDpdgthVsi/hW0n9hyu3Q8mS4chJripLLEYERvDf0PRqGN/R1hD61ZlcGN0yYz96sfN65phdndvLvZLuIiIiIiIiIiHjP4RLM3UssTBJ6yKIlWqhEfGPhBPj279B6EFz+X9Zn7WDUj6MIdgbz3lnv0SSiiY8D9K1f16ZyxyeLCA1y8sWt/enSJNLXIYmIiIiIiIiISC1WYYJZC5VItfPnO/D9/0G7s+DSj9iYvYObfryJAEcA7w99n2Z1m/k6Qp/69M+tjJmynHaxEbw/sjeNo0J9HZKIiIiIiIiIiNRyWvFLqqex7SArBYBBAIlF5c4guOxjtmTv4qYZN2GxvDv0XVrUbeGbOKsBt9vy/PTVvP3rRgZ1aMDrV/bQYn4iIiIiIiIiIlIllIWS6qkouVxGYT7bclK4ccaNFLgLeH/o+7SO9N81KHPyC7nv8yR+WL6La/q14PHzOhHgdPg6LBERERERERER8RNKMEuNsiPAyY0zbiS3MJf3znqPtvXa+jokn0nNyOOmiQtYun0/j57biRsGtsQY4+uwRERERERERETEjyjBLDXGLqeTGxrGkenK5N2z3qVD/Q6+Dsln1u3OYOQH89mblc/bV/fkrM4NfR2SiIiIiIiIiIj4IT1LL9VaqtPByIaxrA4M5MZGsaQ7HYwfMp5O0Z18HZrP/G9dGhe9OZf8Qjef39JfyWUREREREREREfEZjWCWam1cVCSLQoK5vlEsbmN4e1cKXWK6+Dosn/nsz62MmbKctrERvDeyN02iQn0dkoiIiIiIiIiI+DElmKV6Cq5LakEmUyIisMaQ6XDw8u404gOifB2ZT7jdlrE/ruGtxA2c2r4Bb1yZQJ2QQF+HJSIiIiIiIiIifk4JZql+CgsgvAHjwmMpcBQAbpyOAOaddieD+43xdXRVLtdVyD8+X8J3y3ZyZd/mPDm8MwFOzW4jIiIiIiIiIiK+pyyVVD/LvyI1fRNTAgtx4wag0BYyZf0U0nLSfBxc1UrLzOOKd37n++U7eeRvJ/HMBV2UXBYRERERERERkWpDmSqpXtyF8OsLjGvUCpd1lz5k3YxbMs5HgVW9dbszuOCNOazaeYC3rurJqFNbY4zxdVgiIiIiIiIiIiLFlGCW6mX5V7BnPfPrRGKxpQ653C6SUpJ8E1cVm7M+jYvemkuuy82km/szrEtDX4ckIiIiIiIiIiJShuZglurDXQizX4DYzrSNjSdl5zymXzSdpN+TGDRokK+jqzKfz9/Gw5OX0bpBOO+P7E3TemG+DklERERERERERKRcGsEs1ceKybBnHSt7Xc3MrT9xTadriAqJ8nVUVcbttrwwfTX3f7WU/m2i+fK2AUoui4iIiIiIiIhItaYRzFI9FI9e7sQbB5ZTN6gu13a61tdRVZlcVyH/+GIJ3y3dyRV9mvPk+Z0J1GJ+IiIiIiIiIiJSzSmDJdXDyimQtoakHpfza/JvXN/leuoE1fF1VFViT2YeV77zO98t3cnDf+vIsxd2UXJZRETEzxhjhhlj1hhj1htjHizn+H3GmJXGmKXGmJ+NMS1KHGtujPnRGLOqqE7LKg1eRERERPyaRjCL77ndntHLDTryevpS6ofU58qOV/o6qiqxPiWTGybMZ/eBXN66qgdnd23k65BERESkihljnMAbwBBgOzDfGDPVWruyRLXFQC9rbbYx5jbgBeCyomMTgWestTONMRGAuwrDFxERERE/p2GS4nsrp0Dqav5IGMEfu/5kVNdRhAXW/rmH523Yw0VvziE7v4DPbu6n5LKIiIj/6gOst9ZutNbmA58B55esYK2dZa3NLtr9HWgKYIzpBARYa2cW1cssUU9ERERExOs0gll8q2j0so1pz2v7k4gNi+WSDpf4Oiqv+3Lhdh76eikto8N5f2RvmtWv/Ql1ERERqVATYFuJ/e1A38PUvxH4oWi7PbDfGPM10Ar4CXjQWltY8gRjzM3AzQBxcXEkJiZWTuRSrWRmZqpv/ZT63n+p7/2b+t9/Vbe+V4JZfGvVVEhdxW9nPsiSDf/l0X6PEuwM9nVUXmOt5aWZa3ntl/Wc3DaGN67qQWRooK/DEhERkRrCGHM10As4ragoADgFSAC2ApOAkcB7Jc+z1o4HxgP06tXLDho0qGoCliqVmJiI+tY/qe/9l/rev6n//Vd163tNkSG+UzR62R3Tjtf3LaJpRFMubHehr6PymlxXIfd8lsRrv6znsl7N+OD63koui4iICEAy0KzEftOislKMMWcCjwDDrbV5RcXbgaSi6TUKgClAD++GKyIiIiLyFyWYxXdWT4OUFfzc7TxW7V3NbfG3EeionQnXvVn5XP3uH0xdsoMHhnXkuYu7EujUj5+IiIgAMB9oZ4xpZYwJAi4HppasYIxJAN7Gk1xOOeTcKGNMg6L9M4CSiwOKiIiIiHiVpsgQ3ygavVwY3YY39i6iVWQrzml1jq+j8ooNqZncMGE+u9JzeePKHpzTTYv5iYiIyF+stQXGmDuBGYATeN9au8IY8ySwwFo7FRgLRABfGGMAtlprh1trC40x/wf8bDwHFgLv+OaTiIiIiIg/UoJZfGPNd7B7Gd+fdicbtk7lxdNexOlw+jqqSvf7xj3c8tFCAhyGT2/uR4/m9XwdkoiIiFRD1trvge8PKXusxPaZhzl3JtDNe9GJiIiIiFRMCWapetbC7Odx1W/NW3sX06FeB4a0GOLrqCrd14u288BXS2leP4wJ1/ehWf0wX4ckIiIiIiIiIiJSqTQJrFS9Nd/DrmVM7XwW2zK3cWfCnThM7fmjaK3lpZlrue/zJfRuWZ+vbxuo5LKIiIiIiIiIiNRKGsEsVctaSHyO/PqtGLd3Ed1iunFa09N8HVWlySso5IEvlzIlaQeX9GzKMxd2JSig9iTPRURERERERERESlLmS6rW2umwaylfnHQ6u7J3cWfCnRQtVFPj7c3K5+p3/2BK0g5GD+3ACyO6KbksIiIiIiIiIiK1mkYwS9UpGr2cU68F7+xdTO+GvenXqJ+vo6oUm9KyuP6DP9mRnstrVyRwXvfGvg5JRERERERERETE6zS8UqrO2hmwM4lPO57Kntw93BlfO0Yv/7lpLxe+OYcDuQV8OqqvkssiIiIiIiIiIuI3NIJZqoa1MPs5MqOa8/7eJAY2GUiPuB6+juqETV68nQe+XEbT+qF8MLI3LaLDfR2SiIiIiIiIiIhIldEIZqka62bCjsV81GEg6fnp3JVwl68jOiHWWl7+aS33TlpCjxZRTL5toJLLIiIiIiIiIiLidzSCWbyvaPRyer3mTNy3hMHNB9M5urOvozpueQWFPPjVMiYvTmZEz6Y8e2FXLeYnIiIiIiIiIiJ+SVkx8b71P0PyQj5o24csVxZ3xN/h64iO2/7sfK55708mL07m/85qz9gR3ZRcFhERERERERERv6URzOJdRaOX06Ka8d99Szm71dm0q9fO11Edl81pWVw/YT7J+3J45fJ4zo9v4uuQREREREREREREfEpDL8W7NvwC2+fzXpse5Ltd3Nb9Nl9HdFzmb97LhW/OYX92Pv8d1VfJZRERERERERERETSCWbzJWkh8jl1RTZm0fznD2wynZWRLX0d1zL5JSmb0F0tpWi+U90f2pmWMFvMTEREREREREREBjWAWb9o4C7b/yfjW3bFYbu1+q68jOibWWl77eR33fJZEfPMovr59gJLLIiIiIiIiIiIiJWgEs3iHtZD4PNuimjB5/0pGtL+ExhGNfR3VUcsvcPPQ18v4atF2Lkpowr8u7kpwgNPXYYmIiIiIiIiIiFQrSjCLd2yaDdt+Z1z8WTgzNzGq2yhfR3TU9mfnc+vHC/l9417uPbM9dw9uizHG12GJiIiIiIiIiIhUO0owS+UrGr28Maox0w6s5dpO1xIbFuvrqI7Klj1ZXD9hPtv35vDyZfFckKDF/ERERERERERERCqiBLNUvs2/wda5vNHtDEJydnB9l+t9HdFRWbhlL6MmLsRtLR/f1Jc+rer7OiQREREREREREZFqTYv8SeVLfI7VUY34MWM9V3e6mvoh1T9R++2SHVzxzh/UDQlg8u0DlVwWERERERERERE5CkowS+Xa9BtsmcMbzdpRJ6gO13W+ztcRHZa1ljdmreeuTxcT3zSKybcPpFVMuK/DEhERERERERERqRE0RYZUrtnPsySqEYmZm7k74W7qBtX1dUQVyi9w8/DkZXy5cDsXxDfm+RHdCA5w+josERERERERERGRGkMJZqk8m+fA5t94vfMA6ruzueqkq3wdUYXSs13c+vFC5m3cwz2D2/H3M9thjPF1WCIiIiIiIiIiIjWKEsxSeWY/x/x6Dfk9ezuje40mLDDM1xGVa+uebK6f8Cdb92bz0qXduahHU1+HJCIiIiIiIiIiUiMpwSyVY8tc7KZfee2kPsQaN5d2uNTXEZVr4ZZ93DxxAQVuy0c39qVf62hfhyQiIiIiIiIiIlJjaZE/qRyJzzGnXkMW5+7i5m43ExIQ4uuIypi2dAdXvPM7ESEBfH37ACWXRURERERERERETpBGMMuJ2/o7dtNsXuvYgyaBwVzU7iJfR1SKtZa3Zm/ghelr6NWiHuOv7UX98CBfhyUiIiIiIiIiIlLjKcEsJy7xOX6pF8fKvDSe6vUUgc5AX0dUzFXoZszk5UxasI3h3RvzwohuhAQ6fR2WiIiIiIiIiIhIraAEs5yYbX9SuHEWr3foTsuQSM5tfa6vIyqWnuPi9k8WMmf9Hu4+oy33DmmPMcbXYYmIiIiIVJ6x7SArBYBBAIlF5eGxMHqdb2KSKpWancrLu16mS04XYkJjfB2OiIj4Ic3BLCcm8Tmm149jff4+7oi/gwBH9fidxba92Vz81lz+3LSXFy/pzn1ndVByWURERERqn6LkcrnleZngdldtPFLlXlv8GhvzNjJuyThfhyIiIn7Kq9lAY8ww4BXACbxrrX3ukOPBwESgJ7AHuMxau7noWDfgbaAu4AZ6W2tzvRmvHKNt8ynY8DNvtu9Cu4g4zmp5lq8jAmDx1n2MmriA/AI3E2/oS/82WsxPRERERPzQv5p43gPDISgMgsKLtg/uR0BgUXnJV2DRsTLnHFLHoannvM1ay768fezM3MmOrB3syNzBzqydJGcmszNzJ9szt5PlygJg0ppJ7MnZQ//G/ekR24PWUa1xGI0pExER7/NagtkY4wTeAIYA24H5xpip1tqVJardCOyz1rY1xlwOPA9cZowJAD4GrrHWLjHGRAMub8Uqx2n2c3xbP46trgO8Ev9Utbh5+WHZTv4+KYm4uiF8dnNv2sZG+DokERERERHfGPIk5GdDfibkZ4Er2/N+8JWV9tex/GwoSlQetYDQIyehjyZRfWgdZ/V4KrIqFLoLSc1JZWfWzjLJ4x1ZO9iVtYucgpxS54QHhtMovBGNIxpT4C5g84HNFNpCDIbfkn/jp60/ARAZHElCgwQS4hLoEduDTtGdCHJqsXMREal83vyXuw+w3lq7EcAY8xlwPlAywXw+8ETR9pfA68Yzj8FZwFJr7RIAa+0eL8Ypx2P7QvLX/8Rb7U6iS2QrTm92uk/Dsdby9q8bee6H1fRoHsU71/YiOiLYpzGJiIiIiPjUwHuOrb7bDQU5pZPQ+VmexPPBJHR+ZtlE9aF1sveW2C96YY8+DmfwXyOsi5PQh0lKlzcSu7w6PliM3FXoYlfWrrKjj4sSyruzdlNgC0qdUy+4Ho0iGtEmsg0nNzmZxuGNaRzheTUKb0TdoLoYY0jNTuXsr8+m0BYCYIu+44/P/pjNBzazOGUxi1IWkbg9EYAgRxBdYrrQI64HCbEJxMfGUzeobpV+HyIiUjt5M8HcBNhWYn870LeiOtbaAmNMOhANtAesMWYG0AD4zFr7wqEXMMbcDNwMEBcXR2JiYmV/BqlA16VP8kNUDDsLsrjQeSqzZ8/22rUyMzMP27cFbstHK/OZvb2APg2d3NQhn2UL5nktHqk6R+p7qd3U//5Lfe+/1Pcix8hVyTMIOhx/JWgrk7XgyilKTGcWJaqzyiahi0dZl6hTMqF9YMchI7EzwR7DHNOOwCMkoY92CpG/trMdDnbmpLEja2eZ0cc7M3eSmpNanPgFMBgahDWgcXhjujXoRuOWfyWPG4c3pmF4Q8ICw47q44z78FTcIU5w/LXWjNuVy7dfXc6Y21ZxftvzAdibu5fFKYtZvHsxi1MWM2H5BN6172IwtK3Xlh6xnoRzz7ieNAxvePTfp4iISJHq+uxRAHAy0BvIBn42xiy01v5cspK1djwwHqBXr1520KBBVR2nf0peSM7sRbzXpj09ok/i1mG3enUBvcTERCrq2wO5Lu74ZBG/bc/mztPbct+Q9jgcWsyvtjhc30vtp/73X+p7/6W+FzlGs56p+Fh4bNXFcSTGFCVswyA8pvLatRYK8o4wyrq8hPYhdTJ3l63jLsACBxwOdgQ42REQUPRysrNoe2eAk/3O0vNQB1hoaB00NoH0dwTTJCCORoF1aRxYl0Yh0TQMjSYwuG6J5HZRotuGQF4u2NS/EtkBIZ7vrgJLnG5chyyy7nIYkpylk+71Q+ozuPlgBjcfDEBOQQ7L05azaPciFqcsZtrGaUxaMwmARuGNSIj1TKmREJdA26i21WIqRBERqd68mWBOBpqV2G9aVFZene1F8y5H4lnsbzvwq7U2DcAY8z3QA/gZ8b3ZLzCpfgPSCnMYm3CXV5PLh7N9XzY3TJjPxtQsXhjRjUt7NTvySSIiIiIitcGWuTD3Neh5PZz3MuCHv6QxBgJDPC+OfWFvt3WzJ2dP8WjjklNX7MzcwY6sHWQfMv9xqCOQRkGRNAqoQxdnKI0dITQmgMY4aVToJsZVgNOV7Ule52ZC/n7I3/FXQrsw/xg+n+OQEdWlR1h/uWPXMX9mgNCAUHo37E3vhr0BzzzQ6/avK044L9i1gO83fQ9AnaA6xDeIL55Wo0tMF4KdmopQRERK82aCeT7QzhjTCk8i+XLgykPqTAWuA+YBI4BfrLUHp8a43xgTBuQDpwH/8WKscrR2LCZr3Qzea92WAQ170qthL5+EkbRtPzd9uIC8gkIm3tCHAW0rcSSEiIiIiEh1lpcBk2+Fei3hrKd9HU215XK7SMlOYUemZ/7jg4nkg+87s3bicpdeS75uUF0aRzSmWd3m9Gvcv3gxvUYRjWgc3pio4KgTG2BT6Dqx+a3zMyF3Pxw4dOzWIf7TBaLbQHTbolc7z35Uc3CUHnXtdDjpWL8jHet35MqTrsRaS3JmcvEczot3L+a35N8ACHQE0jm6c/HCgQmxCUQGRx7/9yEiIrWC1xLMRXMq3wnMAJzA+9baFcaYJ4EF1tqpwHvAR8aY9cBePElorLX7jDEv4UlSW+B7a+133opVjsHsF/g4ugH73XnclXCXT0KYvnwnf5+URIM6wXx2c1/axtbxSRwiIiIiIj4x42FI3wbX/wDBEb6OxmdyC3KLRxwfmjzekbWDlOwU3IfM0RwTGkPj8MZ0iu7E4BaDixfQO5hIDg+s5PmnD+UMhNAoz+tEPXGYxG7z/rBnPSz9HPIOlLh+ENRr5Uk6x7QtkYBuC+ENwBiMMTSt05SmdZpyXpvzANifu5+k1KTihPNHKz/ig+UfANA2qi0JsQmeqTXietA4vLHPnnIVERHf8OoczNba74HvDyl7rMR2LnBJBed+DHzszfjkGO1IIn3dD3zYsjWnNz2ZLjFdqvTy1lre+W0j//phNfHNonjn2l7EROjxLBERERHxI2umw6KJcPK90Lyfr6PxqgP5BzzJ4goSyHtz95aq7zRO4sLiaBzRmD4N+/w1+rjovWF4Q/+Z3uHidzzv1kJWqifZXPzaAGnrYN2PUHIEd3DkIaOe20BMO6jfhqiQKAY1G8SgZoMAT3J/edry4lHOP2z6gS/WfgFAbFhs8ejmHnE9aBfVDucho6ZFRKR2qa6L/El1NPsFPoxuQKYt4I74O7x+uSmLkxk7Yw3J+3No/PvPtIgOY96GvZzTtRH/vrQ7IYG6SRERERERP5KVBlPvgrguMOghX0dzQqy17MndUyZpXPI9w5VR6pxgZ3BxsrhD/Q6lkseNwxvTIKwBAQ4/+i9ueCxkpZRffpAxEBHrebUYULpeYYFnJPyeDUWJ53We963zYNnnpevWafRX0jm6HSHRbekV3ZZenUdCt1EUugtZv399ccJ50e5FTN88HYCIwAi6N+heah7n0IDQyv0uRETEp/zoX185ITuXsmfdD3zcsiXDWgyhQ/0OXr3clMXJPPT1MnJchQDs2J/Ljv25nHlSLK9dkYDDoUeuRERERMSPWAvT/u6Zf/eayRBQvUfiFroLPfMfZ3nmPy5ePK/ofVfWLnILc0udExEYQaOIRjQJb0LPuJ40Dv9r7uNGEY2IDonW1AsljV5XvHlcCzw6A6B+K8+r3Zmlj+Vnw75NnpHOB0c971kPK7+BnH1/1XMEQL2WOKPb0iG6LR2i23B5s3Mh/l52Gsui1MXFSefXF7+OxRLgCKBTdKfiUc4JsQnUC6l3/N+DiIj4nBLMcnRmP8970Q3Iw81t8bd5/XJjZ6wpTi6XtGpnhpLLIiIiIuJ/ln4Oq76FM/8JDat2qrry5Bfml0kal3zflbWLQlv6fr5+SH0ahzemXb12nNb0tOLk8cFF9OoG1fXRp5EygsIgrrPndajsvYdMubEe0tbDxkQo+OuXBo0Cwzknug3nRLeF6J6kNz2fJc4CFuWlsXjPSj5Z9QkTVkwAoFVkq7+m1YjtQdM6TfXLBBGRGkQJZjmyXcvZve4HJjVvxnmtz6NVZCuvX3LH/pxjKhcRERERqbXSt8P3oz0Ltw2omoW2s1xZpZLGJaev2JG5g7SctFL1HcZBbFgsjcMbEx8bX2b0caPwRpoWobYIqw9hfaBZn9LlbjccSC6bfN6xCFZOIdK6ORU4FSC8AXnRbVgZGcuioAAWuzOZuXk6X637CvAsxpgQm0DPuJ4kxCbQvl57/5r+RESkhtHf0HJks5/nnegY3A5HlYxeBmgcFUpyOcnkxlG6KRURERERP+J2w5TbwV0AF7wFFSyWlpqdysu7XqZLThdiQmMO26S1lv15+/9KGh8y+jg5M5kD+QdKnRPoCKRReCMaRTTilCanlB59HN6IuPA4Ah2BlfaxpQZyOCCqmefV5vTSxwryYO+mUonn4D0bSNj4OwlF80i7gY1BQSyq14hFJo/F2/7HzC0zAQhzhtI9tjsJcT3oEduDrjFdCQsMq+IPKCIiFVGCWQ5v13K2r/uOr5o15eJ2F9MkokmVXHbUqa14YurKUmWhgU5GD/Xu3M8iIiIiItXKn+Nh02w47xXPXLkVGLd0HBvzNjJuyTge7vswqdmpFY4+3pm1k5yC0oM5wgLCipPF3Rp0K1487+B7TGgMDuPw9qeV2iogGGI7el6HytkPezfg2LOBtnvW0zZtHZfuWQ97NrPLncvikGAWhQSzOOc33trxO9aAE8NJoQ1JiOlCj6anEt/s5CP+YkVERLxHCWY5vF9fYFx0DE5HIKO6jqqSS1prmb0mlQAHREcEs/tAHk2iQhk9tAMXJFRNgltERERExOdS18JPj0O7odDjuoqrZafy9bqvsVgmrZnEl2u/LDP/cVRwFI3CG9EqshUDGg+gcUTjUtNYRAZHas5b8Y3QKGjS0/MqyVoaZuzi7D3rOXvPOtizgYy0NSxJ38Cign0sysnh86xkPto2E+ZBC+skISiaHpFt6dGwD82b9MVEt4FAPQUrIuJtSjBLxXavZNO67/i2aROu7ng5ceFxVXLZyYuTmbUmlUfP7cSNJ7c6vhWRRURERERqskIXTL4ZAsNg+GtwmOTvU78/RYG7AACDoV29doxoN6LUNBaaTkBqHGOgbiPPq9UpANQBTgZOLnTBvi3kp65m5Y7fWbxnBYuyk5mVt5spaSmQNpf6SwrpkZtHAqH0CG9Ch+hOBMZ0gOi2EN0GoppXOOWMiIgcGyWYpWK/vsCb9WMIdoZwQ5cbquSSKRm5/PPblfRsUY+RA1pWyTVFRERERKqd3/4NOxbDJR9CnYoHeixNWcqsbbOK9y2WTembGNxisKYMkNrLGQgxbQmKaUv8SecSD1wPuK2bzakrWbTlJxbvWsCiAxv4qSAT2EFoajLdtk8jITePhNw8uhdYwuu1+ivhHN226NUOwmMO+0sdEREpTQlmKV/Katas/Y7pTRsyqtPVRIdGe/2S1loenbKcHFchz1/cDadD/6CLiIiIfzDGDANeAZzAu9ba5w45fh9wE1AApAI3WGu3lDheF1gJTLHW3lllgYt3JC+C2S9At8ug8wUVVsvMz+S2n8suwu22bsYtGceYfmO8GKRI9eMwDlrHdqF1bBdGFJWlZKewKGURi3cvZvGuPxm/fwNuLE4MHYylR8ZKErbPJiEniwaFbs9JwZGlk84xRe/120BwhM8+n4hIdaUEs5Tv1xd4I7o+dQIjuK5zxfO9Vabvlu1kxordPHh2R9rG6h9tERER8Q/GGCfwBjAE2A7MN8ZMtdaWXPF4MdDLWpttjLkNeAG4rMTxp4Bfqypm8SJXDky+BSLi4OwXKqxW6C7kwd8e5ED+gbJNuF0kpSR5MUiRmiM2LJZhLYcxrOUwwPOLmaWpSz1J55TFfJm6lI+D6gH1aBYSQ0JQDD3cThIyD9Bq6zzMss9LN1inUYnRziVe9Vp4RlaLiPghJZilrJTVLFs7jVlN4riz80gigyO9fsk9mXk89s0KujeN5KaTK14dW0RERKQW6gOst9ZuBDDGfAacj2dEMgDW2lkl6v8OXH1wxxjTE4gDpgO9qiJg8aKf/glpa+GaKZ7FzyrwyuJXmL19Ng/3fZgrOl4BoLVLRI5CRFAEA5oMYECTAYDnFzKr96xmUcoiFu1exG8pi5matw+Aeo3qE9/tNHqENyPBEUGn3GwC92yCPeth5TeQs/evhh0BUK9liaRzG890G9FtoU5DTbkhIrWaEsxS1q9jeT26HvWCIrm609VHrl8Jnvh2JRm5Ll4Y0Y8Ap6NKrikiIiJSTTQBtpXY3w70PUz9G4EfAIwxDuDfeBLOZ1Z0gjHmZuBmgLi4OBITE08sYvGKqH1LiV/yFtubnMP6bQa2JZZb74/MP/h4z8ecHHEyjXY1InGXp15mZqb61k+p709cC1rQwtGCC+IuIKUghY25G9mQt4Hlu1Yxq8DzgEigCaRlUEtax3SnTZOLaOuIJjp3H6E5yYRlJxOas4Ow5NWErv8Fpzu/uO1CRwjZYY3JDmtCTmhjssMakxPahOywxhQGhJ9Q3Op7/6b+91/Vre+VYJbSUteyYP005jaK5f+6jSI88MT+sTsaM1bs4tslO7hvSHs6NKzj9euJiIiI1FTGmKvxjFI+rajoduB7a+12c5jRcdba8cB4gF69elmNcq2GctPhzTsgui1Nr3uXpkFh5VZLSkli0oxJ9GnYh1eHvEqg469H8jWC2X+p770rLSeNxSmLWbTbM63GT3t/YoadgcM4aF+vPQmtEugRewkJsQlEhMeB2w0Hkj0jnfesx7lnA3X2rKfOnnWQOges+6/Gw2PLLjQY084zGjoguPyAxraDrJSy5eGxMHqdV74DqZ70s++/qlvfK8EspdjZL/Ba/Xo0CInm0g6Xev16+7PzeWTycjo1qsttg9p4/XoiIiIi1VAy0KzEftOislKMMWcCjwCnWWvzior7A6cYY24HIoAgY0ymtfZBL8csle2HByBjJ9w4EypILu/M3Mk9s+6hYXhD/n3av0sll0XEe2JCYxjSYghDWgwBIMuVxdLUpZ6kc8oipqyfwqerPwWgSUQTEmITSIhNoEdsD1q3Pg2HKfGUbkEe7NvsST6nrStKQm+AtTMg66O/6hkHRDUvSjq3K52ALi+5DBWXi4h4mRLM8pe09czbMI1FDRvwcPdbCA0I9foln5y2kv3Z+Xx4Q28CNTWGiIiI+Kf5QDtjTCs8ieXLgStLVjDGJABvA8OstcUZBGvtVSXqjMSzEKCSyzXNyqmw5FM47QFo2rPcKtmubO765S7yC/P5YOgHRIVEVW2MIlIsPDCc/o37079xf8Azj/PavWuLFw6ct2Me0zZOAyAyOJKEBgkkxHkSzp2iOxHUoAM06FC24dz0vxLORaOf2bMetv4O+ZlHF9z/XvbMB+0IAIezxHZFZeXVcRxFnUPKjEPzTFeVohHsqU4HExrE0OW3NGIK3RrB7g9KPL0wCCCxqLwa9L0SzP6uxB9OC7zWOI7GrgIu/vYxKFosxFt+Wb2brxclc9cZbenc2PsLCYqIiIhUR9baAmPMncAMwAm8b61dYYx5ElhgrZ0KjMUzQvmLoqkwtlprh/ssaKk8mSkw7e/QqDucOrrcKm7r5pH/PcK6/et4/YzXaR3VumpjFJHDCnQE0jmmM51jOnNNp2uw1rItY1txwnnR7kUkbk8EIMgRRJeYLvSI60FCbALxsfHUDarraSgkEpr09LxKshYydv2VcJ7294qD+elxr3zGo1JRAto4jzK5fbT7zhNoo4Lk+DEn4itIzFeFohzOuKhIFoUEMy4qkjF79mkEuz+oxk8vKMHs74r+EKY6HYxqGMuGoCCeTN1DUFaWVy97INfFw18vp31cBHee0dar1xIRERGp7qy13wPfH1L2WIntChfwK1FnAjChsmMTL7IWpt4NeZlw4Xhwlj/lxRtJb/DT1p8Y3Ws0pzQ9pYqDFJFjZYyhed3mNK/bnAvaXgDAnpw9JKUkFSedJyyfwLv2XQyGtvXa0iO2h+cV14OG4Q0PbRDqNvK8Wp1y+ATzwzvBXVD0KvS828LS+6Veh5Ydaf/QssJKarfQM32IO6tEzMfYhs+ZwyTCS+9bh5NChxOXI4B8hxOXw0mBw4nLOMh3OHAVbbuMId8YXMb8tR8exj6ng6/rRGCN4as6EbTLz6deoZvAydcRgCHQOAgAAnEQYIynDEPgwTIgAAeBBs8xa3AYg2foIZ5/nzwbpd7KPV6mbmXsH3qMw9Q9yv3jPvdw167Mz1xRnHDIRrWkBLMA8FZUJBsCA4koLOS8TO8mlwGe/W4VKRm5vH3NQIIDnF6/noiIiIj4gBaiOrzFH8PaH2DovyC2Y7lVvt/4PeOXjueidhdxTadrqjhAEaks0aHRDG4xmMEtBgOeaW+Wpy0vTjh/u+FbJq2ZBECj8EbFczgnxCXQNqpt6Xmci6Q6HYxuEMOLqUVTJECFc7jXetZ6Fk8sSjYXFuThKswj35WDqzAXV8HBl6fcVZhHfkEursL8olce+YV5nm23q+g9H1ehy7PvdpHvduFyFxTtF5DvLsBlCyhwF5JvC3DZQs/LXUi+LcRl3bgoerduXLhxWYuLPPKxFacL3UWvisTGlNotMIanY6I9OwcWHfdX6LDWk5S2FCWnIeDgdsmy4n1DgKUoaV36nEDMIW2YojpF79YUn3Mw+e2p/9d7qTILAeZgmfFc3/x13AmY4p+RoqlaiqdsKbl/uGNHu288+8VFJ9LWMca5ezlQwc++jynBLKQ6HUyuEwHGkOdwsN/p8Oof0N/WpfLZ/G3cclprujeL8tp1RERERMTHqvGjnD63bzNMfxBangJ9by23yvK05Tw29zF6xPZgTN8xGM1vKlJrhAWG0adRH/o06gNAgbuAtfvWFk+pMX/XfL7f5HmwpU5QHeIbxBdPq9ElPJbgrJSyUySEx1ZJ7G7rJv9gItbtwlXoIt+dX7xd4C7w7JdMzpaof3C7oChZW3ys8GAit/S5FbVTpq0Scbht5ec0AhwBBDmCCHQGEujwvIKcQQQGBBLoCPOUOQMJcQRSp+TxovKD5xzcLretQ+qUdyzjrf7c1CiOfMdf/yYEuS1v7kqhzs2Jpb7bg99LgS04bFlx3x0sK9ou9e4uKFWWV+gi65Cy8uodfPe2QEcgAY6A4vcAh7P4OytZXrbe0ZUdeux42q2ojaP+9/0Jz/SyZX72qwElmIW3oiIp/lG3ePUPaGZeAQ9+tYzWMeHce2Z7r1xDRERERKRacxfClNs9835e8Fa583buztrN3b/cTXRINP85/T8EVjB9hojUDgGOADpFd6JTdCeuOukqrLVsz9xenHBenLKY35J/AyCwYR3aR/Vk9b7VWFvI11H1aTfkeUICQshf+8UxJ2ddbhcFhYdPCpcsL7SFlf/5TUCZJGyQo2xyNjggmAhHRJmka8ntAEdAmWRtecncg0m+UnXLSfQGOYKOLQnoZU/VCS8z+tkCMyPCGBPdyRchHZG19q8Ed4lE9MGEd6nkdwVlZRLnR6hfUdnB99yC3KNq1+V2ef37CTBHmehuFIfbWpaFBGONYUpEOLfuT68Wo5iVYPZzqU4HU4pGLwO4HCX+gHrhei9MX82O9By+uKU/IYGaGkNERERE/NDvb8KWOZ7kclSzModzCnK4Z9Y9ZLmy+OhvH1E/pL4PghQRXzLG0KxOM5rVacbwNp41Xffl7iMpJYnFKYv5ZsM3xYlel9vF0388XWFbTuMsk6gtL6ka5AgiLDDshEbXlpsYPsJ1AxwB5U4BIuVbEhqOy1E62e1yGJJCw30U0ZEZYzx/DmrgL0uttRTawqNOXJeXtD7W5HeFSXVHABscf02v4sZ4Bonm+v57VYLZn1nLuKgoDv1djBvDuJiGjKnky/2+cQ8T523h+oEt6dVSN8kiIiIi4od2r4Sfn4SO50L3K8octtby2JzHWLlnJa+c/grt6+mpPxHxqBdSj9Obn06XmC78d/V/Sx0LcgTxwbAPiAuL+2sKBmcQASYAp0ODu2qTL29aUbydmJjIoEGDfBeMHzDGFI8wDiHEp7GkZqdy9tdnQ2EeUDRItF4Mt1483SuDRI+FfkXkz1ZOYV5ocIlJwz1cDkNSk86Veqmc/EIe+GopzeuHMXpoh0ptW0RERESkRijIh8k3Q0gknPdKmftwgPFLxzN983Tu6XEPpzc/3QdBikh1N27puDJzDFssUzdMJS48jvoh9akTVIdgZ7CSyyK1SHk/+27rZtyScT6K6C8aweyvCl3w85M0iQghJ7Qe0y+eTrAz2GuXe/HHNWzZk82no/oRFqQ/diIiIiJ+ITy2/AX9QqKqPJRqYfbzsGsZXP5fCC871uinLT/xetLrnNf6PG7ocoMPAhSRmmBJypIy88K63C6SUpJ8E5CIVInq/LOvTJ+/WjSRZZnb+T2yIfd1utaryeWFW/by/pxNXN2vOf3bRHvtOiIiIiJSzYxeV3o/PwveGgi2EPIyILiOb+LyhW3z4X8vQfzV0PGcModX7VnFw/97mG4NuvH4gMerzWJSIlL9fDn8y+JtTZEg4j+q88++psjwR3mZkPgc7zZuSZ2gOlza4VKvXSrXVcjoL5fSODKUB88+yWvXEREREZEaICgcLhwH6dthxsO+jqbq5Gd5psao2xSG/avM4bScNO6edTd1g+ryyumveHXwh4iIiEhlU4LZH/3+Fhvy9/KLyeWqk64iPNB7K42+/NM6NqZm8a+LuhIRrAHzIiIiIn6veT8YcDcsmghrZ/g6mqox8zHYuwkufAtC6pY6lFeYxz2z7iE9L53XzniNmFBfL9MjIiIicmyUYPY3WWkw5xXea96J0IBQrup4ldcutWTbfsb/uoHLejXj1PYNvHYdEREREalhTn8YYjvDN3dC1h5fR+Nd63+C+e9C/zug5cmlDllr+efcf7I0dSnPnPwMJ0XriT8RERGpeZRg9je/vsh2m8v39gAj2o8gyksLrOQVFHL/l0tpUCeYh8/RjbKIiIiIlBAQDBe9DTn74Lv7wFpfR+Qd2Xs9SfQGHeGMR8sc/mDFB3y78VvuiL+DIS2G+CBAERERkROnBLM/2bcZ5r/LhDY9McbBdZ2u89ql3vhlPWt2Z/Cvi7oSGRroteuIiIiISA3VsCuc/hCsnALLv/J1NN7x/WjISoUL34bAkFKHErcl8vLClxnWchi3dLvFN/GJiIiIVAIlmP3JrGdJCwhksiuV89ucT1x4nFcus2JHOm8mbuCihCac0dE71xARERGRWmDAPdC0j2cU84Edvo6mci3/CpZ/Cac9CI3jSx1at28dD/z6AJ2iO/HkwCcxxvgmRhEREZFKoASzv9i5FJZ+zsQOAyiwhdzQ5QavXMZV6Gb0F0uJCgvisfM6eeUaIiIiIlJLOAPgwnFQ6PJMJVFbpso4sBOm3QdNesHJ95Y6tDd3L3f9chfhgeG8cvorhAaE+ihIERERkcqhBLO/+PmfpIdGMil3G0NbDqV53eZeuczbszewcucBnr6gC1FhQV65hoiIiIjUItFtYMiTsOFnWPC+r6M5cdbC1DuhIM8zNYYzoPiQq9DFvbPuJS0njVdOf8VrTxSKiIiIVCUlmP3Bxtmw/ic+7XQ62QU53NjlRq9cZu3uDF79eT3ndGvEsC4NvXINEREREamFet8ErU+HH8fAng2+jubELHgf1v8EZz0FMW2Li621PP3H0yxKWcSTA56ka4OuPgxSREREpPIowVzbWQs/PU523aZ8krWe05qeRof6HSr9MgWFbkZ/sYSIkACeHN650tsXERERkVrMGDj/DXAEwpTbwF3o64iOz54NniR569M9SfMSPl71MV+v+5pRXUfxt9Z/81GAIiIiIpVPCebabuUU2LGYr7oMYX9eOjd1vemIpxyP9/63iSXb03lieGeiI4K9cg0RERERqcUim8A5L8K2P2Duq76O5ti5C2HyreAMhAve9CTNi/wv+X+8uOBFBjcfzJ0Jd/owSBEREZHKF3DkKlJjFbrg56fIb9CRCenL6N2wN/Gx8ZV+mQ2pmfx75lrO6hTHed0aVXr7IiIiIuInul7iWezvpyc8r5LCY2H0Ol9EdXTmvAzb/4SL34O6jYuLN6ZvZPTs0bSLasezJz+Lw2iMj4iIiNQuurupzRZNhL0b+Lbb30jJTuGmLpU/ernQbXngy6WEBjp5+oIumBIjNUREREREjokxUJhX/rGslKqN5VjsXAqz/gWdL4QuFxcXp+elc9fPdxHkDOK1M14jLDDMh0GKiIiIeIdGMNcmY9uVufEuBN5f+zmd4rrQv3H/Sr/kh3M3s2DLPv59SXdi64ZUevsiIiIiItVaQR5MvgXCouGcl4qnxnC5Xfwj8R/szNrJ+0Pfp1GEnvQTERGR2kkjmGuTckZ1zAwPY2uAg1FdR1X66OIte7J4YcZqTu/QgIt6NKnUtkVEREREyph6N6yaBnmZvo7kL788DSkr4fzXIax+cfHzfz7PH7v+4PH+j3tlmjoRERGR6kIjmGsxC7wTWZdW+S7OaH5Gpbbtdlse+GopgQ4Hz17UVVNjiIiIiIj3Lf8aFn0IziBoMRDanQXth8L7w8qfQsPb8zZvmQtzX4Oe10O7IcXFn63+jElrJnF95+s5v+353ru+iIiISDWgBHMt9ltoCGuDg3gmdU+lLyby3z+38vvGvTx3UVcaRYZWatsiIiIiIuW6fyNs+x3WzoB1M2HGQ55XRbw5b3NeBky+Feq1hLOeLi7+fefvPPfnc5zW9DTu6XGP964vIiIiUk0owVxLWeCdqEgauwo4OzOrUtvevi+bf32/ipPbxnBZ72aV2raIiIiI+Lnw2IpHIwcEQatTPa+hz8DeTZ5E8w+jqz7OGQ9D+ja4/gcIjgBgy4Et/CPxH7SKbMVzpzyH0+Gs+rhEREREqpgSzLVQqtPBrXGxrA0O4uG0vQRWYtvWWh76ehkW+JemxhARERGRynYsU1rUbwV9b66aBHM5C2oDMOkaGL2OA/kHuPPnO3EYB6+e8SoRQRHej0lERESkGtAif7VJQAgA46IiWRsUSLDbzYWZWZ7RHpXkiwXb+W1dGg+e3ZFm9cMqrV0RERERkWqtouk2slIocBdw/+z72Z6xnZcGvUSzOnrKT0RERPyHRjDXFpt+hYJcUvuOYnLaL+B2UegIJPP+9YSExlTKJXal5/LUdyvp26o+V/dtUSltioiIiIjUdP9e8G/m7JjDE/2foHfD3r4OR0RERKRKaQRzbZCfBVPvgvqtGVc3nAJ3AQDGGMYtGVcpl7DW8sjkZbgK3Tx/cTccDk2NISIiIiLVREVP7BkHZKV59dJfRYTz8aqPufqkq7m4/cVevZaIiIhIdaQRzLXBL8/Avs2kXvFfvv5zDBYLgMvtYsr6Kdza/VZiTnAU85SkZH5encKYc06iZUx4ZUQtIiIiIlI5ypu3edt8+PBc+OxKuHYqBIYcf/vpyeUWLwgJ5umY+gxsPJB/9PrH8bcvIiIiUoNpBHNNt20+/P4m9LqBt/YsLB69fJDbuk94FHNKRi5PTF1Jj+ZRXD+w1Qm1JSIiIiJSJZr1hgvHwbY/4Jvbwe0+vnZS18B7Z5Up3h7g5L7YGJq6CnjhtBcIcGjsjoiIiPgn3QXVZAV5MPVOqNsEzvwns7+9sEwVl9tFUkrScV/CWstjU1aQ4yrkhRHdcWpqDBERERGpKTpfCHs3wc//hPqt4Ywxx3b+1j/gv5dCQDCE1oecvaQ6HdwXG8N+h4NC4PUMN3WD6nolfBEREZGaQAnmmuzXFyF1NVz5BenGkleYR99GfXlnyDsYUzmJ4O+X7WL6il08MKwjbWMjKqVNEREREZEqc/K9sHcj/DrWk2SOv/LozlvzA3xxPdRtBNdMhnotAXhr3pMkrf0CgPFDxtOicX8vBS4iIiJSM2iKjJpq1zL430vQ7TJofxbvLH2HA3kHGN1rdKUll/dk5vHYN8vp1jSSUadoagwRERERqYGMgXP/A61Og6l3w6bfjnzOoo/gs6sgtiPc8GNxcjk1O5Wv130NQIAJoF29dl4MXERERKRmUIK5JiosgG/uhNB6MOw5tmVs47+r/8v5bc+nQ/0OlXaZf367kgO5Ll4Y0Y0Ap/6oiIiIiEgN5QyESyd6RjBPugpS15Zfz1rPSOepd0LrQXDdNIhoUHz4xQUvUmgLATDGnPBaJyIiIiK1gbKGNdG812BnEvxtLITV5+WFLxPgCOCuhLsq7RI/rtjF1CU7uPP0dnRsqDnlRERERKSGC42Cqz4HRyD89xLISit93F0IP9wPvzzteUrwis8g+K8p4lKzU5m+eXrxvsvtYsr6KaTlHNKOiIiIiJ/RHMw1Tdo6mPUv6HgudLqApJQkftzyI7d1v43YsNhKucT+7HwembKckxrV5fbT21RKmyIiIiIiPlevpSdx/N6ZMLaC+9z+d8KQp8BReizOc38+h9u6S5W5rZtxS8Yxpt8xLh4oIiIiUosowVyTuN0w9S4IDIFz/o0Fxi4YS4PQBozsPLLSLvPUtFXszcrng5G9CdTUGCIiIiJSmzTrffjjQ58pt3hO8pwyZS63i6SUpEoISkRERKTmUoK5JlnwHmydB+e/CXUaMmPzdJamLuXJAU8SFhhWKZeYtSaFrxZt587T29KlSWSltCkiIiIiUpMt2LWArIIs7u15Lzd0ucHX4YiIiIhUK15NMBtjhgGvAE7gXWvtc4ccDwYmAj2BPcBl1trNJY43B1YCT1hrX/RmrNXe/q3w0xPQ5gyIv5L8wnxeXvgy7eu1Z3ib4SfU9JTFyYydsYYd+3MwBuLqBHPX4LaVE7eIiIiISA1mreXlRS8TGxrLlR2v9HU4IiIiItWO1xLMxhgn8AYwBNgOzDfGTLXWrixR7UZgn7W2rTHmcuB54LISx18CfvBWjNXe2HaQlVK6bMMv8GJ7/nv2IyRnJjN+yHicDudxX2LK4mQe+noZOS7PatjWwv4cFz8s28UFCU1OJHoRERERkRpv1rZZLEldwuP9HyckIMTX4YiIiIhUO96cYLcPsN5au9Famw98Bpx/SJ3zgQ+Ltr8EBhtjDIAx5gJgE7DCizFWb4cml4vsy0lj/NLxnNLkFPo37n9Clxg7Y01xcvmgvAI3Y2esOaF2RURERERqukJ3Ia8uepWWdVtyQdsLfB2OiIiISLXkzSkymgDbSuxvB/pWVMdaW2CMSQeijTG5wAN4Rj//X0UXMMbcDNwMEBcXR2JiYqUFXx0MqqB8XFQkWa4sTnGfcsKfOXl/ToXl1eX7zMzMrDaxSNVS3/s39b//Ut/7L/W9VJnw2PIHc4THltr9duO3bEjfwL9P+zcBDi1fIyIiIlKe6nqX9ATwH2ttZtGA5nJZa8cD4wF69eplBw0aVCXBVZnEskWbAwL4vG4EI9qP4Ir+V5zwJZr8/ku5SeYmUaFUl+8zMTGx2sQiVUt979/U//5Lfe+/1PdSZUavO2KVvMI83kx6k87RnRnSYkgVBCUiIiJSM3lzioxkoFmJ/aZFZeXWMcYEAJF4FvvrC7xgjNkM/B142BhzpxdjrTFeqh9FsLXcHn97pbQ3rEvDMmWhgU5GD+1QKe2LiIiIiNREk1ZPYmfWTv7e8+8cbtCLiIiIiL/z5gjm+UA7Y0wrPInky4FDl12eClwHzANGAL9Yay1wysEKxpgngExr7etejLVGmB8SzKzwMO7Zu5/o0OgTbm9vVj7fJCXTONKzWMnO9FwaR4UyemgHLfAnIiIiIn4rMz+Td5a9Q/9G/enXqJ+vwxERERGp1ryWYC6aU/lOYAbgBN631q4wxjwJLLDWTgXeAz4yxqwH9uJJQgvAruWldt3A2Pr1aFhQwNUFJ756tbWWMVOWkZ7jYuqdJ3NSo7on3KaIiIiISG0wYcUE9uft556e9/g6FBEREZFqz6tzMFtrvwe+P6TssRLbucAlR2jjCa8EV9399ASERMLdSRBWn+82fMuq/z3Msyc/S0ib8064+alLdvD9sl3cP6yDkssiIiIiPmaMGQa8gmdgxrvW2ucOOX4fcBNQAKQCN1hrtxhj4oG3gLpAIfCMtXZSVcZe26TlpDFx5USGthxK5+jOvg5HREREpNrz5hzMcrw2/QrrZ8Ip/4Cw+mzL2Mbjcx+nfb32nNP6nBNufld6Lo9OWU6P5lHccmqbSghYRERERI6XMcYJvAGcDXQCrjDGdDqk2mKgl7W2G/Al8EJReTZwrbW2MzAMeNkYE1UlgddS45eOJ78wnzvjtQSMiIiIyNFQgrm6sRZmPgZ1m0CfmwG4f/b9uNwumkY0xWFOrMustTzw1VJchZZ/XxqP06EFS0RERER8rA+w3lq70VqbD3wGnF+ygrV2lrU2u2j3dzwLaGOtXWutXVe0vQNIARpUWeS1zLaMbXyx9gsuancRLSNb+jocERERkRrBq1NkyHFYMRl2LIbz34TAUDamb2T5Hs98zHN2zCEtJ42Y0Jjjbv6/f25l9tpUnjy/M61iwisrahERERE5fk2AbSX2twN9D1P/RuCHQwuNMX2AIGBDOcduBm4GiIuLIzEx8QTCrb0+TPsQYw3ds7rXyO8oMzOzRsYtJ05977/U9/5N/e+/qlvfK8FcnRTkw89PQmxn6O5Z7/DR/z1afNht3YxbMo4x/cYcV/Nb9mTxzHerOLltDFf3bVEpIYuIiIhI1THGXA30Ak47pLwR8BFwnbXWfeh51trxwHiAXr162UGDBnk/2Bpmzd41LPx2ITd0uYHze55/5BOqocTERNS3/kl977/U9/5N/e+/qlvfa4qM6mTRh7BvE5z5BDic7MjcwdK0pcWHXW4XU9ZPIS0n7ZibLnRb/u+LJTgdhhdGdMOhqTFEREREqotkoFmJ/aZFZaUYY84EHgGGW2vzSpTXBb4DHrHW/u7lWGutVxa9QkRQBNd3ud7XoYiIiIjUKEowVxd5GZD4HLQ4GdoNAeCxuY+VqXZwFPOxeu9/G5m/eR9PnNeZxlGhJxyuiIiIiFSa+UA7Y0wrY0wQcDkwtWQFY0wC8Dae5HJKifIgYDIw0Vr7ZRXGXKss2LWA35J/46auNxEZHOnrcERERERqFCWYq4u5r0F2Ggx5EozBWktSSlKZai63q9zyw1m7O4MXZ6zlrE5xXNSjSeXEKyIiIiKVwlpbANwJzABWAZ9ba1cYY540xgwvqjYWiAC+MMYkGWMOJqAvBU4FRhaVJxlj4qv4I9Ro1lpeXvQysaGxXNHxCl+HIyIiIlLjaA7m6iBjN8x9HTpdAE17ArBg9wLyCvN4rP9jXNL+kuNu2lXo5r7Pk6gTEsCzF3XFGE2NISIiIlLdWGu/B74/pOyxEttnVnDex8DH3o2u9krNTmXUj6PYkL6Bx/s/TmiAnvQTEREROVZKMFcHs5+HwjwY/NeUGBNXTqRecD3Oa33eCTX92i/rWZ58gHFX9yQmIvhEIxURERERqTXeWvIWG9I3EBEYwQVtL/B1OCIiIiI1kqbI8LW09bBwAvQcCdFtANicvpnZ22ZzWcfLCAkIOe6ml2zbzxuz1nNRQhOGdWlYOfGKiIiIiNRwe3P38vGqj/lq7VcA5BXmsT9vv2+DEhEREamhNILZ137+JwSEwGkPFBd9vOpjAhwBXNbhsmNubsriZMbOWMOO/Tk4HYaIYCePD+9cmRGLiIiIiNQobutm1d5V/Lr9V37b/hvL05ZjsaXqjFsyjjH9xvgoQhEREZGaSyOYfWnbfFg1FQbcBRGxAOzP3c8367/h3NbnEhMac0zNTVmczENfLyN5fw4WKHBbcvLdzFqdcsRzRURERERqutTsVEZOH0laThoZ+RnM3DKTR+c8yuAvBnP5tMt5K+ktDIbrOl9HoCOw+DyX28WU9VNIy0nzYfQiIiIiNZNGMPuKtfDT4xDeAAbcWVz8+drPyS3M5dpO1x5zk2NnrCHHVViqLL/QzdgZa7ggockJhywiIiIiUp2NXTCWhbsXctm3l7E3dy8FtoA6QXUY2HggpzY9lQGNBxAdGs1Tvz9V5ly3dWsUs4iIiMhxUILZV9bOgC1z4G8vQnAdAPIL8/l09acMbDyQtvXaHnOTO/bnHFO5iIiIiEhtMSd5Dj9s+gGA1JxULutwGcNaDaN7g+4EOEr/t2dJyhJcblepMpfbRVJKUlWFKyIiIlJrKMHsC+5C+OkJqN/as7hfkR82/UBaThrPDHzmuJptGBnCzvTcMuWNo0KPM1ARERERkZrhhfkvFG8HOAKwWHrG9Sy37pfDv6yqsERERERqPSWYq8rYdpBVzlzIL3WC0euw1vLhyg9pG9WW/o37H3PzbrelXlhgmQRzaKCT0UM7HG/UIiIiIiLV3raMbWxM31i8f3BO5Vu733rM65qIiIiIyLHRIn9Vpbzkcony33f+zrp967i207UYY465+X/PXMPKnRmM6NGEJlGhGKBJVCj/uqir5l8WERERkVrtn3P/Wabs4JzKIiIiIuJdGsFcTUxcOZHokGjOaX3OMZ/77ZIdvDFrA1f0acazF3Y9rgS1iIiIiEhNtSxtWZkyzaksIiIiUjWUYK4GNuzfwP+S/8cd8XcQ5Aw6pnOXJ6cz+ssl9G5Zj38O76LksoiIiIj4le0Z28kuyOaO+Du4tfutvg5HRERExO9oioxq4KOVHxHsDOayDpcd03mpGXncPHEB9cOCeOvqngQFqDtFRERExL9M3TAVg+H8Nuf7OhQRERERv6SMpI/tcTj4dsO3DG8znHoh9Y76vPwCN7d9vJC92fmMv7YXMRHBXoxSRERERKT6cVs336z/hn6N+tEoopGvwxERERHxS0owV5UKpr6YFNOQfHc+V3e6+qibstby+NTlLNiyj7EjutOlSWRlRSkiIiIiUmP8uetPdmTt4MJ2F/o6FBERERG/pTmYq8LG2VCYD2c+ASffW1ycW5DLpK+GclpMV1pHtj7q5j76fQuf/rmNO05vw3ndG3shYBERERGR6m/yusnUCarDGc3P8HUoIiIiIn5LI5i9zV0IPz4Ckc2h722lDn238Tv25u7l2k7XHnVzczek8c9vV3LmSbH8Y0iHyo5WRERERKRGOJB/gJ+3/szfWv2NYKemixMRERHxFY1g9rYln8GuZXDxexAYUlzstm4mrpxIx/od6d2w91E1tW1vNnd8sojWMeH857J4HA7jrahFRERERKq16Zumk1eYx4VtNT2GiIiIiC8pwexN+Vnw85PQpBd0ubjUoTnJc9iYvpFnT34WYypOFE9ZnMzYGWvYsT8Hp8MQ6DS8c20v6oQEejt6EREREZFqa8r6KbSr145O0Z18HYqIiIiIX9MUGd409zXI3AVDn4VDksgTV04kNjSWYS2HVXj6lMXJPPT1MpL352CBArel0A1J2/Z7N24RERERkWps/b71LEtbxgVtLjjsYA0RERER8T4lmL3lwE6Y8wp0Oh+a9y11aM3eNfy+83euOOkKAp0Vj0QeO2MNOa7CUmX5hW7GzljjlZBFRERERGqCKeunEGACOLfNub4ORURERMTvKcHsLb88De4COPOJUsWp2ancMvMWQpwhXNL+ksM2sWN/zjGVi4iIiIjUdi63i283fstpzU6jfkh9X4cjIiIi4veUYPaGnUsh6RPoczPUb13q0MsLX2ZP7h6a1mlKZHDkYZtpGBlSbnnjqNBKC1VEREREpCb5bftv7M3dq8X9RERERKoJJZgrm7Xw4yMQGgWn/l+pQ6nZqXy36TsAth7YSlpO2mGbalpOIjk00MnooR0qLVwRERERkZpk8vrJxITGMLDJQF+HIiIiIiIowVz51s6ATb/CoIcgtF6pQ68uepVC65lT2WIZt2Rchc1MX76L+Vv2cVanWJpEhWKAJlGh/OuirlyQ0MSbn0BEREREpFpKy0njt+2/cV6b8whwBPg6HBEREREBdFdWmQpdMPNRiG4LvW4odSg1O5VvN35bvO9yu5iyfgq3dr+VmNCYUnVTMnJ5ePIyujSpyxtX9STQqd8DiIiIiIhM2zCNQlvIBW0v8HUoIiIiIlJEmcvKtHACpK2FIU+BM7DUoZcXvVw8evkgt3WXGcVsreWhr5aRmVfAfy6NV3JZRERERATPffLk9ZPp3qA7rSNbH/kEEREREakSGsFcWXLTIfFf0PIU6HB2mcO/bf+tTJnL7SIpJalU2aT52/h5dQqPntuJdnF1vBWtiIiIiEiNsixtGRvTN/J4/8d9HYqIiIhfcLlcbN++ndzcXF+HIoeIjIxk1apVXms/JCSEpk2bEhgYeOTKKMFceX77N2TvhbOeBmNKHdqZuZNMVyYXtbuIfw74Z4VNbN2TzVPTVjKgTTTXD2jp5YBFRERERGqOyesnE+IMYVjLYb4ORURExC9s376dOnXq0LJlS8whuS7xrYyMDOrU8c7AVGste/bsYfv27bRq1eqoztH8C5Vh32b4/S3ofgU0ji9z+O2lb2Ox3NLtlgqbKHRb7vs8CYcxjL2kOw6HfnBFRERERAByCnKYvmk6Q1oMISIowtfhiIiI+IXc3Fyio6OVXPYzxhiio6OPaeS6EsyV4ad/gnHC4EfLHNp6YCtT1k/hkvaX0DiicYVNjP91Iwu27OOf53emSVSoN6MVEREREalRftryE5muTC5sd6GvQxEREfErSi77p2Ptd02RcTzGtoOslLLlb58Go9eVKhq3ZBwBjgBGdR1VYXMrdxzgpZlrOLtLQy5MaFLZ0YqIiIiI1GjfrP+GphFN6RnX09ehiIiIiMghNIL5eJSXXC6nfMP+DUzbOI0rOl5Bg7AG5Z6SV1DIfZ8nERkaxDMXdtVvhkREREREStiesZ0/dv3B+W3Px2H03xcRERF/EhFx5KmxbrrpJlauXAnAs88+W+rYgAEDjvsaTqeT+Ph4unTpwnnnncf+/fsP286UKVOK4yjPuHHjmDhx4hHjORZPPPEEL774YqW2eTx0h+ZFbyS9QWhAKDd0uaHCOi/NXMvqXRm8MKIr9cODqjA6EREREZHqb+qGqRgM57c539ehiIiIyGFMWZzMwOd+odWD3zHwuV+Ysji5Sq777rvv0qlTJ6Bsgnnu3LnH3W5oaChJSUksX76c+vXr88Ybbxy2/uESzAUFBdx6661ce+21xx1PdaYEs5es2rOKmVtmck2na6gXUq/cOn9u2sv4XzdyRZ9mnNExroojFBERERGp3tzWzTfrv6Ffo340imjk63BERESkAlMWJ/PQ18tI3p+DBZL35/DQ18sqLcmcmJjIoEGDGDFiBB07duSqq67CWgvAoEGDWLBgAQ8++CA5OTnEx8dz1VVXAX+NTs7MzGTw4MH06NGDrl278s033xzT9fv3709ysuezbNiwgWHDhtGzZ09OOeUUVq9ezdy5c5k6dSqjR48mPj6eDRs2MGjQIP7+97/Tq1cvXnnllVKjjctrIz09nRYtWuB2uwHIysqiWbNmuFwu3nnnHXr37k337t25+OKLyc7OrpTvtbJoDmYveT3pdeoE1eHazuX/ZiIzr4B/fJFEs3phjDmnUxVHJyIiIiJS/f256092ZO3g7z3/7utQRERE/No/v13Byh0HKjy+eOt+8gvdpcpyXIXc/+VSPv1za7nndGpcl8fP63zUMSxevJgVK1bQuHFjBg4cyJw5czj55JOLjz/33HO8/vrrJCUllTk3JCSEyZMnU7duXdLS0ujXrx/Dhw8/qqlqCwsL+fnnn7nxxhsBuPnmmxk3bhzt2rXjjz/+4Pbbb+eXX35h+PDhnHvuuYwYMaL43Pz8fBYsWAB4prM4qKI24uPjmT17NqeffjrTpk1j6NChBAYGctFFFzFqlGd9tzFjxjBx4kRGjx591N+dtynB7AVJKUn8uv1X7ulxD3WD6pZb56lvV5K8L4fPb+lPeLC6QURERETkUJPXTaZOUB3OaH6Gr0MRERGRwzg0uXyk8uPRp08fmjZtCkB8fDybN28ulWA+HGstDz/8ML/++isOh4Pk5GR2795Nw4YNKzzn4Gjo5ORkTjrpJIYMGUJmZiZz587lkksuKa6Xl5dXYRuXXXZZmbLDtXHZZZcxadIkTj/9dD777DNuv/12AJYvX86YMWPYv38/mZmZnHFG9bo3UmbzeITHlr/QX3gs4Bm9XD+kPld2vLLU4SmLkxk7Yw3J+3MAGHxSLL1a1vd6uCIiIiIiNc3G9I38sOkHzmtzHsHOYF+HIyIi4teONNJ44HO/FOe7SmoSFcqkW/pXSgzBwX/dDzidTgoKCo763E8++YTU1FQWLlxIYGAgLVu2JDc397DnHJyDOTs7m6FDh/LGG28wcuRIoqKiyh0lXZ7w8PAyZW63u8I2hg8fzsMPP8zevXtZuHBhcSJ55MiRTJkyhe7duzNhwgRmzpx5VNevKpqD+XiMXgdPpJd9jV7Hnzv/5I+df3BjlxsJCwwrPqXkXDQHzVmfVmUTnouIiIiI1CRPzn0SiyWnoOx/VkVERKR6GT20A6GBzlJloYFORg/tUKVxBAYG4nK5ypSnp6cTGxtLYGAgs2bNYsuWLUfdZlhYGK+++ir//ve/CQsLo1WrVnzxxReAZ2T0kiVLAKhTpw4ZGRlHbK9u3boVthEREUHv3r255557OPfcc3E6Pd9pRkYGjRo1wuVy8cknnxx17FVFCeZKZK3ltcWvERsWy2UdSw+BHztjDTmuwlJluS43Y2esqcoQRURERESqvd1Zu1mUsgiAX7f/SlpOmo8jEhERkcO5IKEJ/7qoK02iQjF4Ri7/66KuXJDQpErjuPnmm+nWrVvxIn8HXXXVVSxYsICuXbsyceJEOnbseEztJiQk0K1bNz799FM++eQT3nvvPbp3707nzp2LFwy8/PLLGTt2LAkJCWzYsOGw7VXUBnimyfj4449LTa/x1FNP0bdvXwYOHHjMsVcFTZFRif6X/D+SUpN4tN+jZR7j21HOYwKHKxcRERER8Vejfx2NxbMyvNu6GbdkHGP6jfFxVCIiInI4FyQ0qfSEcmZmJgCDBg1i0KBBxeWvv/568XZiYmLx9vPPP8/zzz9f5vyYmBjmzZt32Gscqfzbb78t3p4+fXqZ+gMHDmTlypXlxgWlF/lr1apVuW0AjBgxAmttqbLbbruN2267rXj/4Ejpkm36kkYwV5KDo5ebRDThwrYXljkeVzek3PMaR4V6OzQRERERkRpjwe4FLE5ZXLzvcruYsn6KRjGLiIiIVFNKMFeSn7f+zKq9q7it+20EOgNLHbPWEhUWWOYcX8xFIyIiIiJSXRW4C/jHrH+UKT84illEREREqh8lmCtBobuQN5LeoGXdlpzT+pwyxyfN38bqXRlc1KOJz+eiERERERGprt5d9i578/aWKXe5XSSlJFV9QCIiIiJyRJqDuRJM3zyd9fvXM/bUsQQ4Sn+l2/Zm89S0lfRvHc2LI7rjcBgfRSkiIiIiUn2t2LOCt5e8zd9a/Y3nT33+yCeIiIiISLWgEcwnyOV28WbSm7Sv156zWp5V6pjbbbn/y6UYY3hhRDcll0VEREREypFXmMcjvz1C/ZD6PNz3YV+HIyIiIiLHQAnmE5CancqF31zI1oyt3Bl/Jw5T+uucOG8z8zbuYcw5J9GsfpiPohQRERERqd5eW/QaG9I38OTAJ4kMjvR1OCIiIiJyDJRgPgFvJr3JlgNbqB9cn0HNBpU6tikti+emr2ZQhwZc1ruZbwIUEREREanmFuxawMSVE7msw2UMbDLQ1+GIiIhINeJ0OomPj6d79+706NGDuXPnVmr7I0eO5MsvvwTgpptuYuXKlSfcZmJiIpGRkcTHx9OxY0f+7//+74TbPJprnnvuuV6/TkU0B/NxSs1OZcr6KQBkuDLYk7uHmNAYAArdln98nkSQ08HzF3fDGE2NISIiIiIVM8YMA14BnMC71trnDjl+H3ATUACkAjdYa7cUHbsOGFNU9Wlr7YdVFvgJynJlMWbOGJrWacp9Pe/zdTgiIiJyvMa2g6yUsuXhsTB63XE3GxoaSlJSEgAzZszgoYceYvbs2cfd3uG8++67ldbWKaecwrRp08jJySEhIYELL7yQgQOP7hfphYWFOJ3OSovlSAoKCggIOLEUsUYwH6dxS8dBibzxuCXjirff+W0ji7bu58nzuxBXN8QH0YmIiIhITWGMcQJvAGcDnYArjDGdDqm2GOhlre0GfAm8UHRufeBxoC/QB3jcGFOvqmI/UWPnj2Vn1k6eOfkZwgI1pZyIiEiNVV5y+XDlx+HAgQPUq+e5zcnMzGTw4MH06NGDrl278s0333gul5XFOeecQ/fu3enSpQuTJk0CYOHChZx22mn07NmToUOHsnPnzjLtDxo0iAULFgAQERHBI488Qvfu3enXrx+7d+8GIDU1lYsvvpjevXvTu3dv5syZc9iYQ0NDiY+PJzk5GYAff/yR/v3706NHDy655BIyMzMBaNmyJQ888AA9evTgiy++4J133qF37950796diy++mOzsbMAz4vruu+/mzDPPpHXr1sWjr0uaP38+CQkJbNiwocLPPWjQIP7+97/Tq1cvXnnllWPriHJ4dQTzUYzECAYmAj2BPcBl1trNxpghwHNAEJAPjLbW/uLNWI9FanYq36z/hgJ3AeBZ6G/K+inc2v1W9h4I5qUf1zK0cxznxzf2caQiIiIiUgP0AdZbazcCGGM+A84Hip/RtNbOKlH/d+Dqou2hwExr7d6ic2cCw4BPqyDuE/Lr9l/5at1X3NDlBhJiE3wdjoiIiBzODw/CrmXHd+4H55Rf3rArnP1c+ceK5OTkEB8fT25uLjt37uSXXzzpwZCQECZPnkzdunVJS0ujX79+DB8+nOnTp9O4cWO+++47ANLT03G5XNx111188803NGjQgEmTJvHII4/w/vvvV3jdrKws+vXrxzPPPMP999/PO++8w5gxY7jnnnu49957Ofnkk9m6dStDhw5l1apVFbazb98+1q1bx6mnnkpaWhpPP/00P/30E+Hh4Tz//PO89NJLPPbYYwBER0ezaNEiAPbs2cOoUaMAGDNmDO+99x533XUXADt37uTHH38kOTmZ4cOHM2LEiOLrzZ07t/izNmrUiGuuuabCz52fn1+cUD9RXkswlxiJMQTYDsw3xky11paczORGYJ+1tq0x5nLgeeAyIA04z1q7wxjTBZgBNPFWrMdq3NJxuK27VJnbunkz6S3+XDCIiJAAnrmwq6bGEBEREZGj0QTYVmJ/O54RyRW5EfjhMOeWuW82xtwM3AwQFxdHYmLiCYR74rIKs3h257M0DmxMlwNdfB5PbZGZmanv0k+p7/2X+t6/ebv/IyMjycjIACDYlY+jsKDCuodLMBZUcJ7blU9eUfsVCQ0N5bfffgPgjz/+4Oqrr+aPP/6goKCABx98kLlz5+JwOEhOTmbDhg20atWKH3/8kXvvvZdhw4YxYMAAFi1axPLlyxk8eDDgmYIiLi6OjIwMXC4XOTk5ZGRkUFhYSFZWFhkZGQQFBXHaaaeRkZFBp06dmDVrFhkZGcycOZPly5cXx5eens7OnTuJiIgoLsvOzua3336ja9eubNiwgdtvv53w8HB++OEHVqxYQf/+/QFPgrdPnz5kZGRgreWcc84p/r7//PNPnnrqKdLT08nKymLw4MHF8Q4dOhRrLc2aNWP37t1kZGSQnZ3NihUruOmmm5gyZQr16tU77OcuLCzkvPPOK75eeXJzc4/6z5c3RzAfcSRG0f4TRdtfAq8bY4y1dnGJOiuAUGNMsLU2z4vxHrUlKUtwuV2lylxuFz9v/JMtyT1466oexEQE+yg6EREREamtjDFXA72A047lPGvteGA8QK9eveygQYMqP7ijj4X/m/1/ZNts3h/2Ph3qd/BZLLVNYmIivuxb8R31vf9S3/s3b/f/qlWrqFOnjmdn+EuHr/xEZIWHAm6aUeGxoKOI42AMZ555Jnv37iU3N5fvv/+e9PR0Fi9eTGBgIC1btiQgIIAePXqwePFivv/+e5599lkGDx7MhRdeSOfOnZk3b16ZtgMDAwkNDaVOnTo4nU7Cw8OpU6cOgYGB1K1bF/BMl2GMoU6dOlhr+fPPPwkJqXhK3LCwsOI5mDdt2kS/fv24+uqrCQ0N5ayzzuLTT8s+aGaMIS4urviz3n777UyZMoXu3bszYcIEEhMTi+OKiorC6XQWx1OnTh3CwsJo0qQJubm5rFu3jvbt2xMWFlbh53Y6nTRo0OCv/i1HSEgICQlH95SZNxPMRzMSo7iOtbbAGJMOROMZwXzQxcCi8pLLvhqJcWfdO6Fu6bLN6YU89Xsu/Ro5Cd2zhsTENVUSiz/Qb2T9l/rev6n//Zf63n/5cd8nA81K7DctKivFGHMm8AhwWol742Rg0CHnJnolykryw6Yf+HHLj9zT4x4ll0VEROSorV69msLCQqKjo0lPTyc2NpbAwEBmzZrFli1bANixYwf169fn6quvJioqinfffZcHH3yQ1NRU5s2bR//+/XG5XKxdu5bOnTsfcwxnnXUWr732GqNHjwYgKSmJ+Pj4Cuu3atWKBx98kOeff55XX32VO+64g/Xr19O2bVuysrJITk6mffv2Zc7LyMigUaNGuFwuPvnkE5o0OfLEDlFRUbz33nsMGTKE8PBwBgwYUGmf+0i8OgfziTLGdMYzbcZZ5R2vLiMx8goK+ddrc4iOsIwbdSpRYUfz+xc5WvqNrP9S3/s39b//Ut/7Lz/u+/lAO2NMKzwJ48uBK0tWMMYkAG8Dw6y1JVfLmQE8W2Jhv7OAh7wf8vHZnbWbZ/54hm4NujGy80hfhyMiIiKVJTy2/AX9wmNPqNmDczCD5ymoDz/8EKfTyVVXXcV5551H165d6dWrFx07dgRg2bJljB49GofDQWBgIG+99RZBQUF8+eWX3H333aSnp1NQUMDf//7340q0HkwSd+vWjYKCAk499VTGjRt32HNuvfVWXnzxRbKyspgwYQJXXHEFeXmesQJPP/10uQnmp556ir59+9KgQQP69u172KksSoqLi2PatGmcffbZvP/++5X2uY/EmwnmoxmJcbDOdmNMABCJZ7E/jDFNgcnAtdbaDV6M84S9/NM61uzO4P2RvZRcFhEREZFjUvQk3514ksVO4H1r7QpjzJPAAmvtVGAsEAF8UbTOx1Zr7XBr7V5jzFN4ktQATx5c8K+6sdby+LzHcbldPHvyswQ4qvVYFxERETkWo9d5pdnCwsJyy2NiYsqd+qFly5YMHTq0THl8fDy//vprmfIJEyYUb5d8ki4zM7N4e8SIEcUL6cXExDBp0qTDxjxo0KBSgyZCQ0NJTk4ujm/+/Pllztm8eXOp/dtuu43bbrutwngPJpwPxlnyms2bN2fFihXF55T3uSv7qUFv3tUdcSQGMBW4DpgHjAB+sdZaY0wU8B3woLV2jhdjPG5TFiczdsYaduzPwQJ9W9XjjI5xvg5LRERERGoga+33wPeHlD1WYvvMw5z7PlDxMujVQGp2KiOnj2RrxlYe7vswLeq28HVIIiIiIlJJHN5q2FpbABwcibEK+PzgSAxjzPCiau8B0caY9cB9wINF5XcCbYHHjDFJRa8TG1NfiaYsTuahr5eRXJRcBliyPZ0pi8tMlSciIiIi4vdeXPAiWzO2EhcWx2UdLvN1OCIiIiJSibz6XNpRjMTIBS4p57yngae9GduJGDtjDTmu0kP0c11uxs5YwwUJR550W0RERETEX6RmpzJ903QA9uXtY2/uXmJCY3wclYiIiIhUFq+NYK7NduzPOaZyERERERF/NW7pOBwOz387rLWMW3L4hXBEREREpGZRgvk4NI4KPaZyERERERF/lJqdyjfrv6HAXQCAy+1iyvoppOWk+TgyEREREaksSjAfh9FDOxAa6CxVFhroZPTQDj6KSERERESk+hm3dBxu6y5V5rZujWIWERERqUWUYD4OFyQ04V8XdaVJVCgGaBIVyr8u6qr5l0VERERESliSsgSX21WqzOV2kZSS5JuAREREpEZxOp3Ex8fTvXt3evTowdy5cwHYvHkzoaGhxMfHF78mTpwIQMuWLenatSvdunXjtNNOY8uWLVx44YXEx8fTtm1bIiMji8852N7RWLBgAXffffdR1x80aBAdOnSge/fu9O7dm6SkpGP67Mdj0KBBLFiwwOvXOZRXF/mrzS5IaKKEsoiIiIjIYXw5/EtfhyAiIiJVKDU7ldG/jubF016slEV9Q0NDixOzM2bM4KGHHmL27NkAtGnTpsKk7axZs4iJieHxxx/n6aefZvLkyQAkJiby4osvMm3atGOOpVevXvTq1euYzvnkk0/o1asXH3zwAaNHj2bmzJlHdV5BQQEBAVWXtj3R62kEs4iIiIiIiIiIiJywcUvHsWj3Iq9Mh3XgwAHq1at3TOf079+f5OTkco9t3ryZM844g27dujF48GC2bt0KwMiRI7n11lvp1asX7du3L05GJyYmcu655wKQmZnJ9ddfXzxS+quvvjrqOLKysrjhhhvo06cPCQkJfPPNNwBMmDCB4cOHc8YZZzB48GAyMzMZPHgwPXr0oGvXrsX1Nm/eTK9evRg1ahSdO3fmrLPOIicnp9T13G43I0eOZMyYMRQWFjJ69Gh69+5Nt27dePvtt4s/zymnnMLw4cPp1KnTMX2vh9IIZhEREREREREREanQ838+z+q9qw9bJ78wn2Vpy7BYPl/zOav3rCbQGVhh/Y71O/JAnwcO22ZOTg7x8fHk5uayc+dOfvnll+JjGzZsID4+vnj/tdde45RTTil1/vTp07ngggvKbfuuu+7iuuuu47rrruP999/n7rvvZsqUKYAnifvnn3+yYcMGTj/9dNavX1/q3KeeeorIyEiWLVsGwL59+w77OUrG8cwzz3DGGWfw/vvvs3//fvr06cOZZ54JwKJFi1i6dCn169enoKCAyZMnU7duXdLS0ujXrx/Dhw8v/uyTJk3inXfe4dJLL+Wrr77i6quvBjyjka+66iq6dOnCI488wvjx44mMjGT+/Pnk5eUxcOBAzjrrrOLrLV++nFatWh02/iNRgllEREREREREREROyM6snaX2d2TtoEXdFifUZskpMubNm8e1117L8uXLgcNPkXH66aezd+9eIiIieOqpp8qtM2/ePL7++msArrnmGu6///7iY5deeikOh4N27drRunVrVq8unVz/6aef+Oyzz4r3KxpZfdVVV5Gfn09mZmZxrD/++CNTp07lxRdfBCA3N7d49PSQIUOoX78+ANZaHn74YX799VccDgfJycns3r0bgBYtWhQn13v27MnmzZuLr3nLLbdw6aWX8sgjjxRfb+nSpXz5pWfqsvT0dNatW0dQUBB9+vQ54eQyKMEsIiIiIiIiIiIih3Gkkcap2amc/fXZWCwAFsuB/AOMPW1spczFDJ5pJtLS0khNTT1i3VmzZhEVFcVVV13F448/zksvvXRM1zLGHHb/aH3yySf07NmT0aNHc9ddd/H1119jreWrr76iQ4cOper+8ccfhIeHlzo3NTWVhQsXEhgYSMuWLcnNzQUgODi4uJ7T6Sw1RcaAAQOYNWsW//jHPwgJCcFay2uvvcbQoUNLXS8xMbHU9U6E5mAWERERERERERGR4zZu6Tjc1l2qzG3dlToX8+rVqyksLCQ6Ovqo6gcEBPDyyy8zceJE9u7dW+b4gAEDikchf/LJJ6Wm1/jiiy9wu91s2LCBjRs3lkkGDxkyhDfeeKN4/3BTZBhjeOqpp/j9999ZvXo1Q4cO5bXXXsNaTzJ+8eLF5Z6Xnp5ObGwsgYGBzJo1iy1bthzV577xxhv529/+xqWXXkpBQQFDhw7lrbfewuVyAbB27VqysrKOqq2jpQSziIiIiIiIiIiIHLclKUtwuV2lylxuF0kpSSfU7sE5mOPj47nsssv48MMPcTqdwF9zMB98vfrqq2XOb9SoEVdccUWpZPBBr732Gh988AHdunXjo48+4pVXXik+1rx5c/r06cPZZ5/NuHHjCAkJKXXumDFj2LdvH126dKF79+7MmjXrsJ8jNDSUf/zjH4wdO5ZHH30Ul8tFt27d6Ny5M48++mi551x11VUsWLCArl27MnHiRDp27HjE7+ug++67j4SEBK655hpuuukmOnXqRI8ePejSpQu33HILBQUFR93W0TAHs+U1Xa9eveyCBQt8HYZ4QWJiIoMGDfJ1GOID6nv/pv73X+p7/+WrvjfGLLTW9qryC/uI7ptrL/396b/U9/5Lfe/fvN3/q1at4qSTTvJa+9XVyJEjOffccxkxYoSvQ6lQRkYGderU8eo1yuv/iu6bNYJZRERERERERERERI6LFvkTERERERERERERASZMmODrEGocjWAWERERERERERGRMmrL1LpybI6135VgFhERERERERERkVJCQkLYs2ePksx+xlrLnj17yixseDiaIkNERERERERERERKadq0Kdu3byc1NdXXocghcnNzjykBfKxCQkJo2rTpUddXgllERERERERERERKCQwMpFWrVr4OQ8qRmJhIQkKCr8MopikyREREREREREREROS4KMEsIiIiIiIiIiIiIsdFCWYREREREREREREROS6mtqwEaYxJBbb4Og7xihggzddBiE+o7/2b+t9/qe/9l6/6voW1toEPrusTum+u1fT3p/9S3/sv9b1/U//7r2p131xrEsxSexljFlhre/k6Dql66nv/pv73X+p7/6W+Fzkx+hnyX+p7/6W+92/qf/9V3fpeU2SIiIiIiIiIiIiIyHFRgllEREREREREREREjosSzFITjPd1AOIz6nv/pv73X+p7/6W+Fzkx+hnyX+p7/6W+92/qf/9VrfpeczCLiIiIiIiIiIiIyHHRCGYREREREREREREROS5KMIuIiIiIiIiIiIjIcVGCWaoVY8z7xpgUY8zyEmX1jTEzjTHrit7r+TJG8Q5jTDNjzCxjzEpjzApjzD1F5er/Ws4YE2KM+dMYs6So7/9ZVN7KGPOHMWa9MWaSMSbI17GKdxhjnMaYxcaYaUX76ns/YIzZbIxZZoxJMsYsKCrT3/kiR0n3zf5L983+S/fNovtm/1QT7puVYJbqZgIw7JCyB4GfrbXtgJ+L9qX2KQD+Ya3tBPQD7jDGdEL97w/ygDOstd2BeGCYMaYf8DzwH2ttW2AfcKPvQhQvuwdYVWJffe8/TrfWxltrexXt6+98kaM3Ad03+yvdN/sv3TeL7pv9V7W+b1aCWaoVa+2vwN5Dis8HPiza/hC4oCpjkqphrd1prV1UtJ2B5x/NJqj/az3rkVm0G1j0ssAZwJdF5er7WsoY0xQ4B3i3aN+gvvdn+jtf5Cjpvtl/6b7Zf+m+2b/pvlkOUa3+zleCWWqCOGvtzqLtXUCcL4MR7zPGtAQSgD9Q//uFoke9koAUYCawAdhvrS0oqrIdz3+cpPZ5GbgfcBftR6O+9xcW+NEYs9AYc3NRmf7OFzkx+hnyM7pv9j+6b/ZrL6P7Zn9V7e+bA3x5cZFjZa21xhjr6zjEe4wxEcBXwN+ttQc8v5T1UP/XXtbaQiDeGBMFTAY6+jYiqQrGmHOBFGvtQmPMIB+HI1XvZGttsjEmFphpjFld8qD+zhc5MfoZqv103+yfdN/sn3Tf7Peq/X2zRjBLTbDbGNMIoOg9xcfxiJcYYwLx3CR/Yq39uqhY/e9HrLX7gVlAfyDKGHPwF6FNgWRfxSVeMxAYbozZDHyG5xG/V1Df+wVrbXLRewqe/yD3QX/ni5wo/Qz5Cd03i+6b/Y7um/1YTbhvVoJZaoKpwHVF29cB3/gwFvGSovmj3gNWWWtfKnFI/V/LGWMaFI3AwBgTCgzBM5fgLGBEUTX1fS1krX3IWtvUWtsSuBz4xVp7Fer7Ws8YE26MqXNwGzgLWI7+zhc5UfoZ8gO6b/Zfum/2X7pv9l815b7ZWKunZqT6MMZ8CgwCYoDdwOPAFOBzoDmwBbjUWnvogiZSwxljTgZ+A5bx15xSD+OZT079X4sZY7rhWZTAiecXn59ba580xrTG89v5+sBi4GprbZ7vIhVvKnrU7/+steeq72u/oj6eXLQbAPzXWvuMMSYa/Z0vclR03+y/dN/sv3TfLKD7Zn9TU+6blWAWERERERERERERkeOiKTJERERERERERERE5LgowSwiIiIiIiIiIiIix0UJZhERERERERERERE5Lkowi4iIiIiIiIiIiMhxUYJZRERERERERERERI6LEswiItWYMaalMWZ5dW9TRERERMSXdN8sIuI7SjCLiIiIiIiIiIiIyHFRgllEpIYwxrQ2xiw2xvQ+pPwzY8w5JfYnGGNGFI24+M0Ys6joNaCcNkcaY14vsT/NGDOoaPssY8y8onO/MMZEeO/TiYiIiIhUDt03i4hULSWYRURqAGNMB+ArYKS1dv4hhycBlxbVCwIGA98BKcAQa20P4DLg1WO4XgwwBjiz6PwFwH0n+jlERERERLxJ980iIlUvwNcBiIjIETUAvgEustauLOf4D8ArxphgYBjwq7U2xxgTCbxujIkHCoH2x3DNfkAnYI4xBiAImHf8H0FERERExOt03ywi4gNKMIuIVH/pwFbgZKDMjbK1NtcYkwgMxTPi4rOiQ/cCu4HueJ5YyS2n7QJKP80SUvRugJnW2isqIX4RERERkaqg+2YRER/QFBkiItVfPnAhcK0x5soK6kwCrgdOAaYXlUUCO621buAa4P/Zu+/wKKq3jePfk0ZCEkpCr6F3EiB0kADSFFEUFUUpFuTFjqKoWH5WVAQV0QAqiKIiKNioikGKgPTeew+dhASS7Hn/2CUGCC1tU+7PdXG5Oztz5pkMwdl7zz7jmcp2O4EwY4yHMaYs0Mi1fBHQ3BhTGcAY42+MuZ6ZHCIiIiIiWU3XzSIibqCAWUQkB7DWxgKdgaeNMV1SWWUW0Ar4w1p7zrXsU6CXMWYVUB2ITWW7BcAOnDM8PgaWu/YXDfQGvjPGrMb5Nb/qGXZAIiIiIiKZQNfNIiJZz1hr3V2DiIiIiIiIiIiIiORAmsEsIiIiIiIiIiIiImmigFlERERERERERERE0kQBs4iIiIiIiIiIiIikiQJmEREREREREREREUkTBcwiIiIiIiIiIiIikiYKmEVEREREREREREQkTRQwi4iIiIiIiIiIiEiaKGAWERERERERERERkTRRwCwiIiIiIiIiIiIiaaKAWURERERERERERETSRAGziIiIiIiIiIiIiKSJAmYRERERERERERERSRMFzCIiGcwY08MYM+sa1os0xrycCft/zRjzTQaNtdMYc2NGjHWFffQ2xszPzH2IiIiIiKSFMSbCGLM3g8YaZ4x5MyPGusI+Qowx1hjjlZn7ERFJSQGziOQprsA0zhgTY4w55LrIC8jIfVhrJ1hr21/Dev2stW9k5L5FRERERNLDFag6XNfLMcaYvcaYH4wxDS9azxhjnjDGrDXGxLrWm2SMqZNinXBjzG/GmOPGmBPGmPXGmLeMMYUvs+9CxpgvjTEHjTGnjTGbjTGDMvuYRUQkfRQwi0hedIu1NgCoD4QDgy9eQZ/4i4iIiEgett91vRwINAE2AvOMMW1TrPMR8CTwBBAEVAWmAjcDGGOaAVHAAqC6tbYQ0BFIBEIvs9/hQABQAygIdAG2Ztxh6TpfRCQzKGAWkTzLWrsPmA7UBnB9lexRY8wWYItrWWdjzErXjIuFxpi657c3xpQ1xvxkjIk2xhw1xnziWp7c8sE1s2O4MeawMeaUMWaNMeb8/i74ipwx5mFjzFZjzDFjzC/GmFIpXrPGmH7GmC2uWkYaY8wVDs/XGDPRNfNjuTEm1DXOQGPMjylXNMZ8bIz56Go/L2NMDWPMDmPMPam89pkxZuhFy342xgxwPR5kjNnmqme9MabrZfZxyVf6jDFRxpiHUjx/wBizwTUTZqYxprxr+WV/1iIiIiLyH9e3+p41xqw2xpx0XTf6XryeddprrX0F+Bx417V9FeBR4B5r7Rxr7Vlr7RnXN/mGuDZ/DxhrrX3HWnvINd5ua+2r1tqoy5TWEPjWWnvcWuuw1m601k5OUXctY8xs1/XyIWPMi67l+YwxHxpj9rv+fGiMyed6LcI1u/p5Y8xBYKwxxiPF9elR45yhHXSVn9mLxpgjrp9dD9eyhq46PFOsd7sxZtU1nINAY8xfrmtxc9Frdxtjll607GljzC+uxzcbY1a4rnn3GGNeu8J+Lmh5Zy5qp2eMaWKc73NOGGNWGWMiUrzW2xiz3XUNv+P8cYuIXEwBs4jkWcaYssBNwIoUi28DGgM1jTH1gC+BR4BgYBTwi+sC1hP4DdgFhAClge9T2U174AacMzoKAncBR1OppQ3wjuv1kq5xLx6vM86L7rqu9Tpc4fBuBSbhnE3yLTDVGOMNfAN0NMYUcu3XC+gOjL/CWBhj6gMzgcettd+lssp3wN3nL46N82uP7VMcwzagJc6fwf+Ab4wxJa+0z8vUcSvwInA7UBSY59o3XOPPWkREREQA57VSR6ACzuvL3ldZ/yegvjHGH2gL7LXWLkltRdc6TYEfU3v9ChYBbxlj+rhC7JRjBgJ/ADOAUkBl4E/Xyy/hnGkdhnN2dCMu/JZiCZzXxeWBvsDjOK/7W7nGOg6MvEJdJYAiOK/5ewGjjTHVrLX/4rzeTNke736ufm0d7Kp9gbX2CWutvWiVX4FqF/0M7sV5XQ8QC/QECuGcMf5/xpjbrrTPy9RRGvgdeBPnz+dZ4EdjTFHXOfwY6GStDQSaASuvdx8ikjcoYBaRvGiqMeYEMB+YC7yd4rV3rLXHrLVxOC8+R1lrF1trk6y1XwFncV68NsJ5MTrQWhtrrY231qZ2o7oEnF8trA4Ya+0Ga+2BVNbrAXxprV1urT0LvAA0NcaEpFhniLX2hLV2N/AXzgvoy1lmrZ1srU0AhgG+QBPXvv8G7nSt1xE4Yq1ddoWxWgK/AD2ttb9dZp15gHWtC9AN+Mdaux/AWjvJWrvfNRNlIs4Z4o2usM/L6YfzHG2w1ibiPHdhrlnM1/qzFhERERH42HV9dgxnoBl2lfX3AwZnqBkMXOk6qzDOvOHg+QXGmPdcs2RjjTGXtKhzeRyYADwGrDfOb/d1cr3WGThorf3Ade192lq72PVaD+B1a+1ha200zgkN96cY1wG86pppHYfzmvIl1+zss8BrQDdz5fYZL7u2n4szlL3Ltfwr4D7XMQbhnATybepDAM73EHOBSdbaVH8O1tozwM/APa5xq+C8xv3F9XqUtXaN69p6Nc4JF62usM/LuQ+YZq2d5hprNrAU5yQccP7cahtj/Ky1B6y169KwDxHJAxQwi0hedJu1tpC1try1tr/rIvO8PSkelweecV0In3CF0mVxXhSWBXa5Qs7LstbOAT7BOSPisDFmtDGmQCqrlsI5a/n8djE4Z0OUTrHOwRSPz+DsT3c5ycdhrXUAe137gBQXwa7/fn2lY8B5Ab7wCl9lxDXr4ntcF8E4Z1hMOP+6Maan+a/VyAmcbUmKXGW/qSkPfJRinGM43+iUvo6ftYiIiIhc37UlOK9LLXAC53Xqlb6NdhxnOJm8jrX2OVcf5ilAqkGutTbOWvu2tbYBzhD7B2CSK7gti/Nbcam54Fra9bhUiufR1tr4FM/LA1NSXFNuAJKA4pc7Hmtt7GXG/wa4xTXj9y5g3lUmOdwM+AGRV1gHnCF1ymvrqa7gGWNMY1d7jWhjzEmc1+tpvba+86L3Oy2Akq7jvds19gFjzO/GmOpp2IeI5AEKmEVELpTy62l7gLdcYfT5P/ldLSL2AOWuMsvBOaC1H7sukmvibN8wMJXV9uO8wAOSv1YYDOxL43GUTTGWB1DGtQ9w3nylrnH2J+5MiiD4MvrhPNbhV1nvO5wzP8rjbDPyo2v/5YExOGeiBLveWKzFGQxf7PyFe/4Uy0qkeLwHeOSic+JnrV0I1/yzFhEREZHr1xVY7goe/wTKGGPCU1vRtc5inG3N0sRaewrnt9X8cbbx2ANUvMzqF1xLA+X479oXLrzGxzVWp4uuKX2t8x4tqSnsuj6/ZHzXNv/gPNb7ufrkjTE423xMu2jMi80GihpjwnAGzSlnRX+LczZzWWttQZxh9eXuzxLLla+tv77o5+B/vo+2tXamtbYdzg8KNrpqFxG5hAJmEZHLGwP0c80QMMYYf9cNNQKBJTi/FjjEtdzXGNP84gFcN/5o7Op/HAvE45zNcbHvgD7GmDDjvCHJ28Bia+3ONNbewHWDES/gKZytPRYBuGZvTMZ5YbrE1XLjSk7jbKVxgzFmyOVWstauAI7gvAHMTGvtCddL/jgv6qMBjDF9cN1YMZUxonGG6vcZYzyNMQ8AlVKsEgm8YIyp5RqroDHmTtfja/1Zi4iIiMg1cF0DlzbGvAo8hPNeGFhrtwCfAt8Z5030fFzXw92NMYNcmz8HPGCcN9Mr5hqvDM6w+HL7e9l1TedjnDcdfBLnjOlNOO9/UtIY85TrniiBxpjGrk2/Awa7egcXAV7BObP4ciJx9no+f7Pooq57fVzJ/1x1tcQ5SWNSitfGu463Ds5e1VfzmOuYfjXG+KW2gqvV3STgfZz9kWeneDkQOGatjTfGNMI5w/lyVgLdjTHerg8EuqV47fzs6w6ua29f1/ksY4wpboy51RWCnwVi0LW1iFyGAmYRkcuwA/v72wABAABJREFU1i4FHsbZduE4sBXXzU+stUnALThvLrIbZwuKu1MZpgDOoPo4zq/SHcV5kXjxvv4AXsY56/cAzlC1ezrK/9lVz3GcMylud12knvcVzgvgq82wOF/fCaAd0MkY88YVVv0WuJEUMyysteuBD3DO7Djk2u+CK4zxMM6Zx0eBWsDCFGNNwXn38u+NMadwzoQ+35fvmn7WIiIiInJVpYwxMThDxX9xXr9FWGtnpVjnCf5rT3YCZ/uKrjj7OeO6P0kbnDdh3uxqvzADiAJGXGa/FhiLc9LCfpzXnzdba2Ostaddz2/B2d5jC9Datd2bOHsHrwbWAMtdyy7nI5wzgGcZY07jnIjR+ArrH8R5jbkf57f/+llrN6Z4fQquthvn21hciau9XF+c7yF+doXpqTl/bT3potZ8/YHXXbW/grOVyOW8jPO9xXGcvalTXqfvwXlz8BdxTgbZg/M63MP1Z4DrmI/h7PH8f1c7NhHJm4y95GalIiKS2xljyuH8mlsJ19cPRUREREQkjYwx23C2cvvD3bWIiGQ1zWAWEcljXD2ZBwDfK1wWEREREUkfY8wdOGdfz3F3LSIi7nDVm1OJiEju4eqhdghnC4mObi5HRERERCRHM8ZE4bzB9P3WWvUoFpE8SS0yRERERERERERERCRN1CJDRERERERERERERNIk17TIKFKkiA0JCXF3GZIJYmNj8ff3d3cZ4gY693mbzn/epXOfd7nr3C9btuyItbZolu/YTXTdnHvp38+8S+c+79K5z9t0/vOu7HbdnGsC5pCQEJYuXeruMiQTREVFERER4e4yxA107vM2nf+8S+c+73LXuTfG7MrynbqRrptzL/37mXfp3OddOvd5m85/3pXdrpvVIkNERERERERERERE0kQBs4iIiIiIiIiIiIikiQJmEREREREREREREUmTXNODOTUJCQns3buX+Ph4d5ci6VCwYEE2bNhwXdv4+vpSpkwZvL29M6kqEREREREREZHcS7la9pWWrOx6XG+ulqsD5r179xIYGEhISAjGGHeXI2l0+vRpAgMDr3l9ay1Hjx5l7969VKhQIRMrExERERERERHJnZSrZV/Xm5Vdj7Tkarm6RUZ8fDzBwcH6JchjjDEEBwfrEzYRERERERERkTRSrpY3pSVXy9UBM6BfgjxK511EREREREREJH2Ur+RN13vec33ALCIiIiIiIiIiIiKZQwFzJgsICLjqOg899BDr168H4O23377gtWbNmqV5H56enoSFhVG7dm1uueUWTpw4ccVxpk6dmlxHaiIjIxk/fvxV67ker732GkOHDs3QMUVEREREREREJOdTrnZl2SVXU8CcwtQV+2g+ZA4VBv1O8yFzmLpiX5bs9/PPP6dmzZrApb8ICxcuTPO4fn5+rFy5krVr1xIUFMTIkSOvuP6VfhESExPp168fPXv2THM9IiIiIiIiIiKSOylXy7u5mgJml6kr9vHCT2vYdyIOC+w7EccLP63JsF+GqKgoIiIi6NatG9WrV6dHjx5YawGIiIhg6dKlDBo0iLi4OMLCwujRowfw36coMTExtG3blvr161OnTh1+/vnn69p/06ZN2bfPeSzbtm2jY8eONGjQgJYtW7Jx40YWLlzIL7/8wsCBAwkLC2Pbtm1ERETw1FNPER4ezkcffXTBpyKpjXHy5EnKly+Pw+EAIDY2lrJly5KQkMCYMWNo2LAhoaGh3HHHHZw5cyZDfq4iIiIiIiIiIuJeytXydq7m5e4Cssr/fl3H+v2nLvv6it0nOJfkuGBZXEISz01ezXdLdqe6Tc1SBXj1llrXXMOKFStYt24dpUqVonnz5ixYsIAWLVokvz5kyBA++eQTVq5cecm2vr6+TJkyhQIFCnDkyBGaNGlCly5drqnpdlJSEn/++ScPPvggAH379iUyMpIqVaqwePFi+vfvz5w5c+jSpQudO3emW7duydueO3eOpUuXAs5p9+ddboywsDDmzp1L69at+e233+jQoQPe3t7cfvvtPPzwwwAMHjyYL774gscff/yaf3YiIiIiIiIiIuIeytWyV642fvx4Bg4ceM0/u8yWZwLmq7n4l+Bqy9OiUaNGlClTBoCwsDB27tx5wS/ClVhrefHFF/n777/x8PBg3759HDp0iBIlSlx2m/Of2uzbt48aNWrQrl07YmJiWLhwIXfeeWfyemfPnr3sGHffffcly640xt13383EiRNp3bo133//Pf379wdg7dq1DB48mBMnThATE0OHDh2u6bhFRERERERERCR7U66WuszK1dq0aXNNx51V8kzAfLVPRJoPmcO+E3GXLC9dyI+JjzTNkBry5cuX/NjT05PExMRr3nbChAlER0ezbNkyvL29CQkJIT4+/orbnO8Vc+bMGTp06MDIkSPp3bs3hQoVSvXTnNT4+/tfsszhcFx2jC5duvDiiy9y7Ngxli1blvwXvnfv3kydOpXQ0FDGjRtHVFTUNe1fRERERERERETcS7la9srVZs+efU37zyrqwewysEM1/Lw9L1jm5+3JwA7VsrQOb29vEhISLll+8uRJihUrhre3N3/99Re7du265jHz58/Pxx9/zAcffED+/PmpUKECkyZNApyf4KxatQqAwMBATp8+fdXxChQocNkxAgICaNiwIU8++SSdO3fG09P5Mz19+jQlS5YkISGBCRMmXHPtIiIiIiIiIiKSvSlXy9u5mgJml9vqlead2+tQupAfBucnLO/cXofb6pXO0jr69u1L3bp1k5uRn9ejRw+WLl1KnTp1GD9+PNWrV7+ucevVq0fdunX57rvvmDBhAl988QWhoaHUqlUrubF59+7def/996lXrx7btm274niXGwOc0/m/+eabC74G8MYbb9C4cWOaN29+3bWLiIiIiIiIiEj2pVwtb+dq5vwdF3O68PBwe75p9nkbNmygRo0abqpIMsrp06cJDAy87u10/nO+83eJlbxJ5z/v0rnPu9x17o0xy6y14Vm+YzdJ7bpZcgf9+5l36dznXTr3eVtmn3/lKtlXWrOy65Ha+b/cdXOe6cEsIiIiItnQ+1Ug9jAAEQBRruX+xWDgFvfUJBku+kw0A/8eyNBWQyniV8Td5YiIiIhIBlKLDBERERFxH1e4fM3LJUf6bNVnLD+0nMhVke4uRUREREQymAJmEREREXGPXNKqTa4s+kw0P275EYtl6tapHIk74u6SRERERCQDKWAWERERkax1aj/MGwafNHR3JZIFIldHYjAAJDoSNYtZREREJJdRwCwiIiIimS8hHtb+CN/cAcNrwZ//A/+i7q5KMln0mWh+3vozSTYJgCSbpFnMIiIiIrmMAmYRERERyRzWwt6l8NvT8EFVmPwARG+Cls/C48vhgenurlAyWeTqSBzWccEyh3VoFrOIiIhILqKAOZN5enoSFhZGaGgo9evXZ+HChRk6fu/evZk8eTIADz30EOvXr0/3mFFRURQsWJCwsDCqV6/Os88+m+4xr2WfnTt3zvT9iIiISBY4dQDmD4eRjeDztrDyO6jaCXr+Ak+uhjYvQXAl57r+xVIf43LLJUdZdXgVCY6EC5YlOBJYeXilewoSERGRHEW52rXv0525mpfb9pzdvF8l9buV+xeDgVvSPKyfnx8rV64EYObMmbzwwgvMnTs3zeNdyeeff55hY7Vs2ZLffvuNuLg46tWrR9euXWnevPk1bZuUlISnp2eG1XI1iYmJeHnpr7KIiIhbJcTDpt9h5bewbQ5YB5RrCl1GQM3bwLdA6tuluM6KiooiIiIiS8qVrDG5y+QLnvea3ouj8UeZdMskN1UkIiIimUK52gXyWq6mGcznpfZLcKXlaXDq1CkKFy4MQExMDG3btqV+/frUqVOHn3/+2bm72FhuvvlmQkNDqV27NhMnTgRg2bJltGrVigYNGtChQwcOHDhwyfgREREsXboUgICAAF566SVCQ0Np0qQJhw4dAiA6Opo77riDhg0b0rBhQxYsWHDFmv38/AgLC2Pfvn0AzJo1i6ZNm1K/fn3uvPNOYmJiAAgJCeH555+nfv36TJo0iTFjxtCwYUNCQ0O54447OHPmDOD8ZOiJJ56gWbNmVKxYMflTopT+/fdf6tWrx7Zt21i2bBmdOnW65LgjIiJ46qmnCA8P56OPPrq+EyEiIiIZw1rYuwx+G/BfC4zDG6HlM64WGDOgfs/Lh8uS53Sr2o1dp3ax9NBSd5ciIiIiGUm5WqoyK1e78cYbrzlXS+24MzpXyzvTPqcPgoNr0rbt2JtTX16iDnQacsVN4+LiCAsLIz4+ngMHDjBnzhwAfH19mTJlCgUKFODIkSM0adKELl26MGPGDEqVKsXvv/8OwMmTJ0lISODxxx/n559/pmjRokycOJGXXnqJL7/88rL7jY2NpUmTJrz11ls899xzjBkzhsGDB/Pkk0/y9NNP06JFC3bv3k2HDh3YsGHDZcc5fvw4W7Zs4YYbbuDIkSO8+eab/PHHH/j7+/Puu+8ybNgwXnnlFQCCg4NZvnw5AEePHuXhhx8GYPDgwXzxxRc8/vjjABw4cID58+ezceNGunTpQrdu3ZL3t3DhwuRjLVmyJPfffz8TJkygQoUKlxz3uXPnkn/xRUREJAudOgCrJzpnKx/ZBF5+ULMLhN0LITeAh+YwSOralW/HkCVDmLRpEg1LNHR3OSIiInKtlKtlq1xt1qxZ7Nu375pytcsdd0bmanknYHaTlFP5//nnH3r27MnatWux1vLiiy/y999/4+Hhwb59+zh06BB16tThmWee4fnnn6dz5860bNmStWvXsnbtWtq1awc4p8qXLFnyivv18fFJ7r3SoEEDZs+eDcAff/xxQT+ZU6dOERMTQ0BAwAXbz5s3j9DQULZs2cJTTz1FiRIl+O2331i/fn3ylP5z587RtGnT5G3uvvvu5Mdr165l8ODBnDhxgpiYGDp06JD82m233YaHhwc1a9ZM/gQIYMOGDfTt25dZs2ZRqlSp5OO+9dZb8fDwuOS4U+5PREREMllCPGya5mqB8ee1t8AQScHXy5culbowcdNEjsUfI8g3yN0liYiISDamXC39udrljjsjc7W8EzBf5RMRXit4+df6/J4hJTRt2pQjR44QHR3NtGnTiI6OZtmyZXh7exMSEkJ8fDxVq1Zl+fLlTJs2jcGDB9O2bVu6du1KrVq1+Oeff655X97e3hhjAGdD9MTERAAcDgeLFi3C19f3ituf7xWzY8cOmjRpwl133YW1lnbt2vHdd9+luo2/v3/y4969ezN16lRCQ0MZN24cUVFRya/ly5cv+bG1NvlxyZIliY+PZ8WKFZQqVQprLbVq1WLWrFkEBgZecX8iIiKSCayF/cthxQRYOxniT0KB0tBigHO28vkb9Ylchzuq3ME3G77h122/0qtWL3eXIyIiItdCuRqQM3O1yx13RuZq+v5iFtq4cSNJSUkEBwdz8uRJihUrhre3N3/99Re7du0CYP/+/eTPn5/77ruPgQMHsnz5cqpVq0Z0dHTyX4iEhATWrVuXphrat2/PiBEjkp+f/xTocipUqMCgQYN49913adKkCQsWLGDr1q2A8+sCmzdvTnW706dPU7JkSRISEpgwYcI11VaoUCF+//13XnjhBaKiopKPe/HixUD6jltERESuw+mDsOAj+LQJjGkDKydAlQ5w/1R4ag20fVnhsqRZ5cKVqVesHpM3T77gTZGIiIjIlShXu7LL5WoZcdxXk3dmMF+Nf7HL3+0yHc73igHnpwpfffUVnp6e9OjRg1tuuYU6deoQHh5O9erVAVizZg0DBw7Ew8MDb29vPvvsM3x8fJg8eTJPPPEEJ0+eJDExkaeeeopatWpddz0ff/wxjz76KHXr1iUxMZEbbriByMjIK27Tr18/hg4dSmxsLOPGjeOee+7h7NmzALz55ptUrVr1km3eeOMNGjduTNGiRWncuDGnT5++pvqKFy/Ob7/9RqdOnfjyyy+ZPHkyjz76KM8880y6jltERESuIiEeNk93tsDY+oezBUbZxnDLR1CrK/heYVZKBog+E82HBz+kdlxtivgVydR9ifvdWfVOXpz/IksPLVUvZhERkdxAudpluTtXy4jjvhqTW2YNhIeH24sbU2/YsIEaNWq4qSLJKKdPn061RcbV6PznfFFRUURERLi7DHETnf+8S+c+C51vgbHyW1gzGeJPOFtghHaH0HuhSOUsK+WNRW8wadMk7qp2F4ObDM6y/QIYY5ZZa8OzYD8dgY8AT+Bza+2Qi17vBzwKJAExQF9r7XpjTAiwAdjkWnWRtbafa5sGwDjAD5gGPGmvcoGf2nVzVotPjKfNpDa0KN2C9254z6215Cb69zPv0rnPu3Tu87bMPv/KVbKvtGZl1yO183+562bNYBYRERHJa04fhNUTncFy9Ebw8oUatzj7KldoBR6eWVrOntN7+GnLT1gsU7dOpV9ov1w3i9kY4wmMBNoBe4F/jTG/WGvXp1jtW2ttpGv9LsAwoKPrtW3W2rBUhv4MeBhYjDNg7ghMz5SDyEDnb/b3w6YfOB5/nMK+hd1dkoiIiIikkXowi4iIiOQFiWdh3VSYcBcMqwmzX3G2vbjlI3h2M9zxOVRqk2Xh8rmkc/y1+y+e+/s5bplyC4kO141TrIPIVVf+mmEO1QjYaq3dbq09B3wP3JpyBWvtqRRP/YErzkQ2xpQEClhrF7lmLY8HbsvQqjNRtyrdSHAk8Mu2X9xdioiIiIikg2Ywi4iIiORW1sL+Fa4WGJOcLTACS0HzJ52zlYtUydJyEh2JLDm4hBk7ZvDH7j84fe40BXwKYFPkqAmOhNw6i7k0sCfF871A44tXMsY8CgwAfIA2KV6qYIxZAZwCBltr57nG3HvRmKVT27kxpi/QF5y9+VLeidydKuaryPiV4yl3uFzyndol7WJiYrLNuZWspXOfd+nc522Zff4LFix4zb1/JWslJSVl+rmJj4+/5r9fCphFREREcpvTh1K0wNjgbIFRvbMzVK4YkaUtMBzWwcrDK5m+Yzqzds3iWPwx/L39aVuuLR1DOvLXnr+YunUqDuu4YJvIVZFZ3os5O7DWjgRGGmPuBQYDvYADQDlr7VFXz+WpxpjrujuLtXY0MBqcPZizS7/OU9tO8dL8lwioEaCb/WUA9WLNu3Tu8y6d+7wtK3owZ3afX0mbrOjB7OvrS7169a5pXQXMIiIiIrlB4lnYNN0ZKm/9A2wSlGkEnT+EWl3Br1CWlWKtZf2x9UzfPp0ZO2dw6MwhfD19aVW2FZ1COtGiTAvyeeYD4KPlH5HgSLhg+wRHAisPr8yyerPIPqBsiudlXMsu53uc/ZWx1p4FzroeLzPGbAOqurYvcx1jZjvty7dnyJIhTN48WQGziIiISA6lgFlEREQkp7IWDqz8rwVG3HG3tsDYenwr03dOZ8aOGew+vRsvDy9alGrB0w2eJqJsBP7e/pdsM7nL5OTHuXwW1r9AFWNMBZwhcHfg3pQrGGOqWGu3uJ7eDGxxLS8KHLPWJhljKgJVgO3W2mPGmFPGmCY4b/LXExiRNYeTMVLe7G9Q/CDd7E9EREQkB9JN/jKZp6cnYWFhhIaGUr9+fRYuXAjAzp078fPzIywsLPnP+PHjAQgJCaFOnTrUrVuXVq1asWvXLrp27UpYWBiVK1emYMGCyducH+9aLF26lCeeeOKa14+IiKBatWqEhobSsGFDVq5ceV3HnhYREREsXbo00/cjIiKSo50+BAtHwGfNYHQELPvKeYO++36Cp9fCja9mWbi859Qexqwew+2/3E7XX7ry+ZrPKRVQiv81+x9Rd0Uxou0Ibq54c6rhcl5irU0EHgNmAhuAH6y164wxrxtjurhWe8wYs84YsxJnH+ZeruU3AKtdyycD/ay1x1yv9Qc+B7YC24DpWXE8GemOKnfoZn8iIiKSKuVq18dduZpmMF8k+kw0A/8eyNBWQzPkxjJ+fn7Jf4FmzpzJCy+8wNy5cwGoVKnSZf9y/fXXXxQpUoRXX32VN998kylTpgDOmT1Dhw7lt99+u+5awsPDCQ8Pv65tJkyYQHh4OGPHjmXgwIHMnj37mrZLTEzEyyvr/npl9f5ERESyXOJZ2DzDOVt5y2xXC4yG0Hk41Lo9S1tgHIo9xMydM5m+Yzprj64FoF6xerzQ6AXah7TPbTfnyzDW2mnAtIuWvZLi8ZOX2e5H4MfLvLYUqJ2BZWa5KoWrEFY0jMmbJ9OzZk/d7E9ERCQHU652obySq2kG80UiV0ey/NByIldFZvjYp06donDh6/vaX9OmTdm3L/VWejt37qRNmzbUrVuXtm3bsnv3bgB69+5Nv379CA8Pp2rVqsm/NFFRUXTu3Blw3mm0T58+yZ/o/Phjqu9ZUq0jNjaWBx54gEaNGlGvXj1+/vlnAMaNG0eXLl1o06YNbdu2JSYmhrZt21K/fn3q1KmTvN7OnTupUaMGDz/8MLVq1aJ9+/bExcVdsD+Hw0Hv3r0ZPHgwSUlJDBw4kIYNG1K3bl1GjRqVfDwtW7akS5cu1KxZ87p+riIiIjmCtbB/JUwbCB9Ugx96woFV0PwJePRfeOgPCH8gS8LlY/HHmLhxIr1n9Kbd5Ha8v/R9kmwSzzR4hll3zGJ8p/HcW+NehcuSJt2qdmPnqZ0sO7TM3aWIiIhIOihXu3odGZGrhYeHZ6tcLVOjcGNMR+AjwBP43Fo75KLXbwA+BOoC3a21k13Lw3De1KQAkAS8Za2dmJ5a3l3yLhuPbbziOueSzrHmyBoslh82/cDGoxvx9vS+7PrVg6rzfKPnrzhmXFwcYWFhxMfHc+DAAebMmZP82rZt2wgLC0t+PmLECFq2bHnB9jNmzOC2225LdezHH3+cXr160atXL7788kueeOIJpk6dCjj/si1ZsoRt27bRunVrtm7desG2b7zxBgULFmTNmjUAHD9+/IrHkbKOt956izZt2vDll19y4sQJGjVqxI033gjA8uXLWb16NUFBQSQmJjJlyhQKFCjAkSNHaNKkCV26OL8BumXLFr777jvGjBnDXXfdxY8//sh9990HOD816dGjB7Vr1+all17i448/pmDBgvz777+cPXuW5s2b0759++T9rV27lgoVKlyxfhERkRwl5jCs/sE5W/nwOvDMBzU6O/sqV2wNHp5ZUsapc6eYs3sOM3bMYNGBRSTZJCoWrEj/sP50DOlISMGQLKlDcr8OIR149993mbR5EuElrm9mkIiIiGQ+5WrZK1fbtm0bEydOvKZcbfTo0Zmeq2VawGyM8QRGAu2AvcC/xphfrLXrU6y2G+gNPHvR5meAntbaLcaYUsAyY8xMa+2JzKoX4EDsgQue74/dT/kC5dM1Zsqp/P/88w89e/Zk7VrnV0mvNJW/devWHDt2jICAAN54441U1/nnn3/46aefALj//vt57rnnkl+766678PDwoEqVKlSsWJGNGy/8R+CPP/7g+++/T35+uU+AevTowblz54iJiUmuddasWfzyyy8MHToUgPj4+ORPedq1a0dQUBDgvIP8iy++yN9//42Hhwf79u3j0KFDAFSoUCH5H4EGDRqwc+fO5H0+8sgj3HXXXbz00ksAzJkzh/Xr1zN5svMmQCdPnmTLli34+PjQqFEjhcsiIpI7JJ5L0QJjlrMFRulwuHkY1L4d/LLm5mdnEs7w996/mb5jOvP2zSPBkUDpgNL0qd2HjiEdqVq4qloYSIbz9fLlloq3MGnzJF6If4FCvoXcXZKIiIhcJ+Vql8qsXK18+fLXnKvNmjWL1atXZ2qulpkzmBsBW6212wGMMd8DtwLJAbO1dqfrNUfKDa21m1M83m+MOQwUBU6ktZirfSISfSaaTj91wmKd+8Vy6twp3m/1foZ91bNp06YcOXKE6Ojoq677119/UahQIXr06MGrr77KsGHDrmtfF7/xS+sbwQkTJtCgQQMGDhzI448/zk8//YS1lh9//JFq1apdsO7ixYvx9/e/YNvo6GiWLVuGt7c3ISEhxMfHA5AvX77k9Tw9PS+Yyt+sWTP++usvnnnmGXx9fbHWMmLECDp06HDB/qKioi7Yn4iISI5jrbPlxcpvYc0kiDsGASWg2ePO2cpFq119jAxwLukcC/YtYPqO6UTtjSIuMY6ifkW5u9rd3FThJmoXqa1QWTLdHVXv4NuN3/LLtl/oWaunu8sRERGRFJSrKVe7kswMmEsDe1I83ws0vt5BjDGNAB+cd8W++LW+QF+A4sWLExUVdcHrBQsW5PTp09e0nxErRuCwF+TcOKyDEf+O4Nl6F0+wvj7na9i8eTOJiYn4+Phw+PBhHA5HqvVZa4mJiSFfvny8+eabNGnShCeffJKgoCDOnDlDYmIip0+fplGjRowdO5Z77rmHCRMm0LRpU06fPk1CQgLfffcdt99+Ozt37mTbtm2UKlWKffv2JW/bqlUrhg8fzrvvvgs4p/Jf/GlLUlISsbGxxMTE8NxzzxEaGsqyZcto3bo1H3zwAUOHDsUYw6pVqwgNDSU+Pp5z584lH9OhQ4coVKgQ8fHxzJo1i127dhETE+P82aY49rNnz3L27FlOnz5NUlIS3bt3Z+HChdx+++1MmDCB1q1bM2LECBo2bIi3tzdbtmyhVKlSF/wsUhMfH3/J3wnJWWJiYnQO8zCd/7wrL5x773MnKH5oLiUO/klA7C4cxpsjRRpzsHJbjhcOxXp4wroDwIGrjpVWSTaJzfGbWR67nFVnVhFn4/D38KdB/gbU969PpXyV8DjjwdF1R5nL3EyrI6W8cO7l8qoWrkpo0VAmbZ7E/TXv14caIiIiOUjk6shUc7XIVZEMbjI4Q/axceNGkpKSCA4O5syZM1dd38vLiw8//JA6deowePDg5JnB5zVr1ozvv/+e+++/nwkTJlzQXmPSpEn06tWLHTt2sH37dqpVq8aiRYuSX2/Xrh0jR47kww8/BFLP1c4zxvDGG29QqVIlNm7cSIcOHRgxYgQjRozAGMOKFSuoV6/eJdudPHmSYsWK4e3tzV9//cWuXbuu5cfEgw8+yN9//81dd93FTz/9RIcOHfjss89o06YN3t7ebN68mdKlS1/TWNcq625HmAbGmJLA10Avay/6WwpYa0cDowHCw8NtRETEBa9v2LCBwMDAa9rX+hPrSXAkXLAswZHAuhPrrnmM1MTFxSX/BbXWMn78eAoVKsSJEyfYsWPHBX95H3jgAZ544gmMMQQEBBAYGEhgYCD33nsv48eP5+WXXyZ//vx4eXkRGBjIZ599Rp8+ffjkk08oWrQoY8eOJTAwEG9vbypWrEjbtm05deoUo0aNomjRohds+/rrr/Poo4/StGlTPD09efXVV7n99tsvqN3T0xN/f//kOp599lk+/fRTPvnkE5566imaN2+Ow+GgQoUK/Pbbb/j6+uLj45P883rwwQe55ZZbaNasGeHh4VSvXp2AgAAAPDw8ktfLly8fCQkJBAYGJu/zhRdeID4+nv79+xMZGcnhw4dp1aoV1lqKFi3K1KlTLzie1Pj6+qb6Cyo5R1RUFBf/XkveofOfd+Xac594DrbM/K8FhiPR2QIj4kk8at9OMb/CFMvkEhzWwYrDK5i+Yzqzd83mWPwxArwDaFexHZ0qdKJxycZ4e1y+T15my7XnPo+bumIf78/cxP4TcZQq5MfADtW4rV7qb2rurHongxcMZtmhZerFLCIikoOsOrwq1Vxt5eGV6Rr3fA9mcOZqX331FZ6ezvuRXNyD+XyullLJkiW55557GDlyJC+//PIFr40YMYI+ffrw/vvvJ+dq55UrV45GjRpx6tQpIiMj8fX1vWDbwYMH8+ijj1K7du3L5mop+fn58cwzz/D+++8n52p169a9IFe7WI8ePbjllluoU6dOcq52rQYMGMDJkyeTw/OdO3dSv379C3K1jGSstRk6YPLAxjQFXrPWdnA9fwHAWvtOKuuOA347f5M/17ICQBTwdsrllxMeHm6XLl16wbINGzZQo0aNdBxFztS7d286d+5Mt27d3F1Khjh9+nSaQv68ev5zEwUNeZvOf96Vq869tXBwtTNUXv3Dfy0wQu+GsB5Z0gLDWsv6o+uZvmM6M3bO4NCZQ/h6+tKqbCs6hXSiRZkW5PPMd/WBsoC7zr0xZpm1Ns+kmaldN2eWqSv28cJPa4hLSEpe5uftyTu310k1ZI5LjKPtD225oewNDGk55JLX5cpy1b+fcl107vMunfu8LbPPf17NVXJCrpbWrOx6pHb+L3fdnJkzmP8FqhhjKgD7gO7AvdeyoTHGB5gCjL+WcFlEREQkW4mJhjU/OIPlQ2vB0weq3+wMlSu2Bs/M/xLZ1uNbmbZjGjN2zmDP6T14eXjRolQLBjQYQETZCPJ758/0GkTen7npgnAZIC4hifdnbko1YPbz8qNzpc78uPlHBjUcpJv9iYiIiOQAmfbuxlqbaIx5DJgJeAJfWmvXGWNeB5Zaa38xxjTEGSQXBm4xxvzPWlsLuAu4AQg2xvR2DdnbWrsys+rNTcaNG+fuEkRERPKexHPO1hcrJ6RogdEAbv4Aat0O+YOuPkY67T61mxk7ZzB9x3S2ntiKh/GgcYnGPFznYdqUa0PBfAUzvQaRlKbG9aao78lLlkfHFQR2p7pNt6rd+G7jd3y74VuWHFrC0FZDM+zmQCIiIiJXo1zt+mXq9Blr7TRg2kXLXknx+F+gTCrbfQN8k0E16AYheVBmtX4RERG5xAFXC4w1P8CZoxBQHJr0h7B7oVjmf6XwYOxBZu6cyYwdM1h7dC0A9YvV58XGL9KufDsFc+JWRc2l4fKVlsN/N/v7ZuM3xJyLydCbA4mIiMj1Ua6WN11vrpatb/KXXr6+vhw9epTg4GD9MuQh1lqOHj16SQN2ERGRDBMTDWsmuVpgrHG2wKh2k7MFRqU2md4C41j8MWbvnM20HdNYfng5ADWDa/JMg2foENKBkgElM3X/IpmtQ0gH3vv3PQCmbp1Kv9B++rBEREQkiylXy5vSkqvl6oC5TJky7N27l+joaHeXIukQHx9/3WGxr68vZcpcMjleREQk7ZJbYHwLW2Y6W2CUqg83DYXad2R6C4xT507x564/mbFzBosPLCbJJlGpYCUeC3uMjhU6Ur5A+Uzdv0hW2nZiW/LjJJukWcwiIiJuoFwt+0pLVnY9rjdXy9UBs7e3NxUqVHB3GZJOUVFR1KtXz91liIhIXnVwDayY4JYWGGcSzjB371ym75jO/H3zSXAkUCagDA/UfoCOFTpSpVAVzSaRXCf6TDS/bf8t+XmiI5EpW6doFrOIiEgWU66WfWW3rCxXB8wiIiIiaRJ7xNUCY4IzYPb0gWqdXC0w2mZqC4xzSeeYv28+M3bMIGpvFHGJcRTzK0b36t3pFNKJ2kVqK1SWXC1ydSQO67hg2bmkcwxfOpy3Wr7lpqpERERE5HIUMIuIiIgAJCX81wJj8wxXC4x6WdICI9GRyJIDS5i2Yxpzds/hdMJpCuUrxC0Vb6FThU7UL14fD+ORafsXyTT+xSD2cOrLL2PV4VUkOBIuWT5953Qer/84JfxLZGSFIiIiIpJOCphFREQkbzu4xhkqr/4BzhxxBl9N/g9C74XiNTNttw7rYPmh5czYOYPZu2ZzLP4YAd4BtC3Xlk4VOtGoZCO8Pbwzbf8iWWLglv8e//E/kuZ/RAu+YP4zt+F5mU0md5l8ybJV0avoN7sfvWf05osOX1A6oHTm1CsiIiIi100Bs4iIiOQ9sUedPZXPt8Dw8IbqN2V6CwxrLeuOrmP6junM2DmDw2cO4+vpS0TZCDpW6EiL0i3I55kvU/Yt4naVWuM5fxg1zq1l3f7W1C1T6Jo3DS0ayuftP+fh2Q/TZ0YfvujwBWUDy2ZerSIiIiJyzRQwi4iISN6QlABbZjtD5c0zwZEAJcOg0/tQp1umtsDYcnxLcqi85/QevDy8aFG6Bc80eIaIshHk986fafsWyTbKNsZ6+dEycQ3ztx65roAZoFaRWnzR/osLQubyBcpnTq0iIiIics0UMIuIiEjudnCtqwXGRFcLjKLQ+BEIuxeK18q03e4+tZsZO2cwfcd0tp7YiofxoHGJxjxc52HalGtDwXwFM23fItmSVz5M+Wa02bGeF7ceoX9E5eseokZwDWfIPMsZMn/e4XMqFqyYCcWKiIiIyLVSwCwiIiK5T+xRWDPJ1QJjtbMFRrVOzhYYlduCZ+b0Nj4Ye5CZO2cyfcd01h1dB0D9YvV5qfFLtCvfjmC/4EzZr0iOUTGC8tv+ZPfObcQnNMTX+3KdmC+vWlA1vujwBQ/NeogHZjzA5+0/p3Lh6w+rRURERCRjKGAWERGR3CEpAbb+ASu+ydIWGEfjjjJ712ym75jO8sPLAagVXItnw5+lQ0gHSviXyJT9iuRIlVrDbGjoWM2yXR1oXrlImoapUrgKYzuO5aGZD/HgrAcZ3W401YKqXfc40WeiGfj3QIa2GkoRv7TVIiIiIpLXKWAWERGRnO3Quv9aYMRG/9cCI/QeKFE7U3Z56twp/tz1J9N3TGfxwcU4rINKBSvxWNhjdKrQiXIFymXKfkVyvGK1cOQvwg1Ja5m/9UiaA2aAigUrMrbjWB6Y+QAPzXqIMe3HEOwbfF2BceTqSJYfWk7kqkgGNxmc5lpERERE8jIFzCIiIpLzxB6FtZOdLTAOrHK1wOjoaoFxY6a0wDiTcIa5e+cybcc0FuxbQIIjgTIBZXiw9oN0qtCJKoWrZPg+RXIdDw88KkbQav2fjN0SDR2rp2u48gXKM67DOB6c9SAPznyQxiUaJwfGj4U9xpG4I0THRXMk7ghH445yJO4IR+KPcOTMEQ7GHmTX6V0ATN06lX6h/TSLWURERCQNFDCLiIhIznC+BcbKCbBphqsFRih0eg9qdwP/jO9vfC7pHPP2zWPGjhnM3TuXuMQ4iuUvxj3V76FThU7UCq6FMSbD9yuSq1WMoPDaycQfWMfJM00omD99HwiVLVCWsR3H0nt6b2bvng3AxE0Tmbhp4iXr5vPMRxG/IhTxK0KiTcQDDxw4SEhK0CxmERERkTRSwCwiIiLu934ViD0MQARAlGu5fzHoOTVLW2AkOBJYcmAJ03dMZ87uOZxOOE3hfIXpUqkLHUM6Ur94fTyMR4bvVyTPqBgBQHOzln+2H6Fj7ZLpHrJ0QGnCi4fz+47fsVg88CC0aCj31rg3OVAu4lcEf29/jDFEn4mm00+dcOAAwIGDKVumaBaziIiISBooYBYRERH3c4XLqS7/rBl4eEFVVwuMKu0yvAWGwzpYfmg5M3bOYNbOWRw/e5wA7wDalmtLpwqdaFyyMV4eumwSyRCFymKDKtPq6Fr+2Ho0QwLm6DPRzN49G4sFnIHx+mPrCS8RnmpgHLk6Eod1XLAswaFZzCIiIiJpoXdKIiIikr11fBfqdAP/jJ1VaK1l3dF1TNsxjZk7Z3L4zGH8vPyIKBNBxwodaV66Ofk882XoPkXEyVRqTePjX/PWlgNA+r+JkFpg7LCOywbGqw6vIsGRcMEyi2XBvgXprkVEREQkr1HALCIiIu6VlHjl15v0y9DdbT6+mRk7ZjB9x3T2xuzF28ObFqVb8Gz4s7Qq04r83vkzdH8ikoqKEfj+O4ZCx1ax/0QLShXyS9dwqQXGCY4EVh5emer6k7tMvnDdpARu/+X25MfemXCjUBEREZHcSgGziIiIuEfiWWdv5QUfZfqudp/azfQd05mxcwZbT2zF03jSuGRj+tbtS9vybSngUyDTaxCRFCq0xBoPmnuuZcHWI9wZXjZdw10cGF8vb09vnmv4HP3/7M+3G7+lV61e6RpPREREJC9RwCwiIiJZ62wMLBsL/4yE0wegVH04viPDd3Mw9iAzd85k2o5prD+6HoD6xerzUuOXaFe+HcF+wRm+TxG5Rr4FoXQDWu9dx5cZEDBnhJZlWtKydEsiV0Vyc8WbdbM/ERERkWukgFlERESyxpljsHgULBkFccehwg1w22dQMQKGVk39Rn/+xa5rF0fijjB712xm7JjB8sPLAagVXItnw5+lQ0gHSviXyIADEZGMYCq2pvbeoazcuhtrwzDGuLsknmv4HF1/7sonKz7htWavubscERERkRxBAbOIiIhkrlMH4J9PYOlYSIiFajdDywFQJvy/dQZuSX4YFRVFRETENQ9/8uxJ5uyew/Qd01l8cDEO66Byoco8Xu9xOoZ0pFyBchl4MCKSYSpG4PH3e1Q9s4Ith9tQtXiguysipGAIPWr0YPz68dxV7S5qBtd0d0kiIiIi2Z4CZhEREckcx7Y7+yuv/BYcSVD7DmjxNBRPf2BzJuEMUXuimL5zOvP3zSfRkUjZwLI8WPtBOlXoRJXCVdJfv4hkrjINcXjnp3niWuZvOZItAmaAR0If4dftv/LukncZ13FctphZLSIiIpKdKWAWERGRjHVwLcwfDut+Ag9vqHcfNHsCgiqka9izSWeZv28+03dMZ+6eucQnxVMsfzHurX4vN1W4iZrBNRUEieQkXj54hLSg9da1/G/rER5okb5/IzJKoE8gT9R7gtf+eY2ZO2fSsUJHd5ckIiIikq0pYBYREZGMsWcJzPsANs8AnwBo+hg0fRQCr73vcfSZaD48+CG142pTxK8ICY4EFh9YzPQd05mzew4xCTEE+QZxa+Vb6VShE/WK1cPDeGTiQYlIpqrYmnJbZrFrx2YSkhrg7Zk9fp9vq3wbEzdN5INlH9CqbCv8vPzcXZKIiIhItqWAWURERNLOWtg2B+YNg13zwS8IWr8EjR4Gv8LXPVzk6ki2n93O6/+8TlG/oszeNZvjZ48T6B3IjeVvpFNIJxqVbISXhy5hRHKFihEA1Etcyeq9nWhQPsi99bh4enjyfKPn6T2jN+PWjuP/wv7P3SWJiIiIZFt6dyYiIiLXz+GAjb86g+UDKyGwFHR4Bxr0Ah//NA254egGftz8IxbLX3v+wtfTl9ZlW9OxQkdalG6Bj6dPxh6DiLhfsRo4/IvR4tRa5m85mm0CZoAGxRvQMaQjX679ktsq30bJgJLuLklEREQkW8oe30ETERGRnCEpAVZMgE8bww894ewp6DICnlwJTfunKVw+GHuQtxa9RfffupNkkwDwNJ7cXPFm3mv1Hm3KtVG4LJJbGYNHpda08lrHwq2H3V3NJQY0GIDFMnzZcHeXIiIiIpJtKWAWERGRqzt3BhaPgo/rwc/9wTMfdPsSHlsK9XuCV77rHnJfzD7+98//6PRTJyZtmgQp7s+XZJP4bftvHIk7koEHISLZUsXWFLInObNnNbFnE91dzQVKBpTkgdoPMH3ndJYdWubuckRERESyJQXMIiIicnnxJ5037vuwDkx/DgqUhnsnQb95UPsO8PC87iF3n9rNywtepvNPnfl568/cUeUOOlXohKe5cCyHdRC5KjKjjkREsquKrQBowmqW7Dzm5mIu1ad2H4rnL867S94lyZHk7nJEREREsh31YBYREZFLxUTDok/h38+dbTAq3wgtn4HyzdI85PaT2xmzegzTdkzD28Obu6vfTZ9afSjuX5xuv3QjwZFwwfoJjgRWHl6ZzgMRkWyvQCkcRarR8vA65m09Qutqxdxd0QX8vPx4JvwZnvv7OaZuncodVe9wd0kiIiIi2YoCZhEREfnPid2wcAQsHw+JZ6HmrdByAJQMTfOQW45vYfTq0czcORNfL1961uxJr1q9KOJXJHmdyV0mJz+OiooiIiIiPUchIjmMR6XWND46lqFbDgA13V3OJTqGdOS7jd/x8YqPaR/SnkCfQHeXJCIiIpJtKGAWERERiN4E8z+ENT84n4d2h+ZPQZEqaR5y47GNjFo1ij92/0F+r/w8UPsBetbqSZBvUIaULCK5SMXW5Fscif/hZRyJaUGRgOvv656ZjDEMajSI7r91Z/jS4Ww/tZ2hrYZe8EGZiIiISF6lgFlERCQv27cc5g+DDb+Bly80fBiaPQYFy6R5yLVH1jJq1Sii9kYR6B3II3Uf4f6a91MwX8EMLFxEcpWQ5ljjRQuPNfyz7Si3hJZyd0WXqBlck9ur3M7kLc5vXESuimRwk8FurkpERETE/RQwi4iI5DXWws75zmB52xzIVxBueBYa9wP/tM/GW3l4JZGrI1mwbwEFfArwaNij3FvjXgr4FMjA4kUkV8oXCGUacsOedUzYeiRbBswA91S/hx+3/AjA1K1T6RfaT7OYRUREJM9TwCwiIpJXWAubZ8C8YbB3CfgXgxv/B+EPgG/aQ+ClB5cSuTqSxQcWUzhfYZ6s/yTdq3UnwCcgA4sXkdzOVIqg1p4hrNqyA6jr7nJS9cPmH/A0niTZJBKSEjSLWURERAQFzCIiIrlfUiKsmwLzh8PhdVCoHNw0FOrdB95+aRrSWsvig4sZtWoUSw8tJdg3mGfDn+XOqneS3zt/Bh+AiOQJFVvjEfUO5U8tZ/fRdpQLzl7/lkSfiebnrT+TZJMAcODgpy0/aRaziIiI5HkKmEVERHKrxLOw8ltY8CEc3wlFq0PXUVD7DvD0TtOQ1loW7F/AqFWjWBm9kmJ+xRjUaBB3VLkDXy/fDC1fRPKY0vVxeAfQMnEN87ce4d7gcu6u6AKRqyNxWMcFyxIcCQxbOoy3W77tpqpERERE3E8Bs4iISG5zNgaWjYWFn0DMQShVH9q/BdVuAg+PNA1prWXu3rmMWjWKtUfXUtK/JIMbD+a2KreRzzNfBh+AiORJnt6Yii25YdMKhmw9wr2Ns1fAvOrwKhIcCZcsn71rNv9r9j+80/jBnYiIiEhOp4BZREQktzhzDBaPgsWREH8CKtwAt4+CCq3AmDQN6bAO5uyew6jVo9h4bCOlA0rzWtPX6FKpi8IUEclwpmJrym6azs6t63A46uHhkbZ/uzLD5C6TL1k2c+dMnp37LMOWDeP5Rs+7oSoRERER91PALCIiktOd2g//jISlYyEhFqrdDC0HQJnwNA+Z5Ehi9q7ZjFo9iq0ntlK+QHnebP4mN1W8CW8PBcsiaWGM6Qh8BHgCn1trh1z0ej/gUSAJiAH6WmvXG2PaAUMAH+AcMNBaO8e1TRRQEohzDdPeWns4Cw4nc1RsDUCdcytZf6ALtUsXdHNBV9YhpAMrD6/kmw3fUK9YPdqHtHd3SSIiIiJZTgGziIhITnVsOyz4yNln2ZEEdbpB86egeM00D5noSGT6jumMWTOGHSd3ULFgRYa0HELHkI54enhmXO0ieYwxxhMYCbQD9gL/GmN+sdauT7Hat9baSNf6XYBhQEfgCHCLtXa/MaY2MBMonWK7HtbapVlxHJmuSBWSAkrS4uQaFmw9ku0DZoABDQaw+shqXln4ClULVyWkYIi7SxIRERHJUmlrxCgiIiLuc3AtTH4QRjSAld9BvfvhieVw++g0h8sJjgSmbJnCrVNv5cX5L+Ll4cXQVkOZcusUbq54s8JlkfRrBGy11m631p4DvgduTbmCtfZUiqf+gHUtX2Gt3e9avg7wM8bkzubnxuBZuQ03eK1n4dZod1dzTbw9vfmg1Qd4e3gzYO4A4hLjrr6RiIiISC6iGcwiIiI5xZ4lMO8D2DwDfAKg2ePQpD8ElkjzkAlJCUzdNpUv1nzBvph91AiqwYetP6R12dZ4GH0OLZKBSgN7UjzfCzS+eCVjzKPAAJztMNqkMs4dwHJr7dkUy8YaY5KAH4E3rbU2lXH7An0BihcvTlRUVBoPI/MVO1uMmvY0R7f+S8igowT7Gu6o6k2zUtm7Pc89Be8h8nAkj099nPuK3OeWGmJiYrL1uZXMo3Ofd+nc5206/3lXdjv3CphFRESyM2th2xyYNwx2zQe/IGj9EjR6GPwKp3nYs0ln+WnLT3y59ksOxh6kTpE6vNj4RVqWbolJ4w0BRST9rLUjgZHGmHuBwUCv868ZY2oB7wIpG/32sNbuM8YE4gyY7wfGpzLuaGA0QHh4uI2IiMi0Y0iv6f8UpuaG4bTwWMvapIocjbd8vSGJmjVqclu90lcfwE0iiMCutESuiuTm0JvpWqVrltcQFRVFdj63knl07vMunfu8Tec/78pu514Bs4iISHbkcMDGX53B8oGVEFgKOrwDDXqBj3+ah41LjGPy5smMXTuW6Lho6hWrx/+a/o+mpZoqWBbJXPuAsimel3Etu5zvgc/OPzHGlAGmAD2ttdvOL7fW7nP997Qx5lucrTguCZhzkjfnHiXEUZYWHmuITOoCQFxCEu/P3JStA2aAfnX7sfLwSt5a/BY1gmtQPai6u0sSERERyXQKmEVERLKTpARY/QMs+BCObIagitBlBNS9G7zS3nL1TMIZJm6ayLh14zgWf4yGJRoypOUQGpZoqGBZJGv8C1QxxlTAGSx3B+5NuYIxpoq1dovr6c3AFtfyQsDvwCBr7YIU63sBhay1R4wx3kBn4I/MPpDMNjWuN0U9TgKw0/O/H1F0XEFgt5uqujaeHp68e8O73PnrnQyIGsDEzhMJ9Al0d1kiIiIimUoBs4iISHZw7gys+BoWjoCTe6B4Heg2FmreCum4wV7MuRi+2/gd49eP58TZEzQt2ZRHQh+hQfEGGVi8iFyNtTbRGPMYMBPwBL601q4zxrwOLLXW/gI8Zoy5EUgAjvNfe4zHgMrAK8aYV1zL2gOxwExXuOyJM1wek2UHlUmKmpPXtTy7CfINYmirofSZ0YeXF7zM8Ijh+iBPREREcjUFzCIiIu4UfxL+/Rz++RTOHIFyTaHzcKh8I6QjkDh59iTfbviWrzd8zelzp2lZuiWPhD5CaNHQDCxeRK6HtXYaMO2iZa+kePzkZbZ7E3jzMsPq06JsqF6xejzd4GmGLh3K1+u/pmetnu4uSURERCTTKGAWERFxh5hoWPSpM1w+ewoqt4OWA6B8s3QNeyL+BOPXj+e7jd8RkxBD67KteST0EWoF18qgwkVE5Fr0rNmTFYdXMHzZcOoUrUO9YvXcXZKIiIhIplDALCIikpVO7Ha2wVg+HhLPQq3boMXTUDJ9M4uPxh3lq/VfMXHjROIS47ix/I08UvcRqgVVy5i6RUTkuhhjeKP5G9z92908O/dZfuj8A8F+we4uS0RERCTDKWAWERHJCtGbYP6HsOYHwEBod2j+FBSpnL5hz0Qzbt04ftj0A+cc5+gQ0oG+dfpSuXD6xhURkfQL9AlkWMQwevzeg0HzBhF5YySe6eirLyIiIpIdKWAWERHJTPuWw/xhsOE38PKFhg9Ds8egYJl0DXsw9iBj145l8ubJJNkkbq54Mw/VeYgKBStkUOEiIm7iXwxiD6e+PAeqHlSdl5q8xKsLX2XU6lH0D+vv7pJEREREMpQCZhERkYxmLeyc7wyWt80B34Jww7PQuB/4F0nX0Pti9vHFmi+YunUq1lq6VO7CQ7UfomyBshlUvIiImw3ckvzQ7piH+aozX5UcTK9HBrqxqPTpWrkryw8tJ3JVJKFFQ2leurm7SxIRERHJMAqYRUREMoq1sHkGzBsGe5c4Z9vd+D8IfwB8C6Rr6D2n9jBmzRh+3fYrxhi6Vu7Kg3UepFRAqQwqXkQk+zHlm3PMuwRVDvyGw/EsHh7G3SWliTGGl5q8xPpj6xk0bxCTbplECf8S7i5LREREJEMoYBYREUmvpERYNwXmD4fD66BQObj5AwjrAd5+6Rp6x8kdjFk9hmk7puHl4cVd1e6iT+0+CiZEJG/w8OBQSBcabx7D1u1bqVq5irsrSjM/Lz+GtRpG99+788zcZxjXYRzent7uLktEREQk3RQwi4iIpFXiWVj5LSz4EI7vhKLVoetoqH07pDM02Hp8K6PXjGbGjhnk88xHjxo96F2rN0XzF82Q0kVEcooizXrhuWU0xxZ9C5VfdXc56RJSMITXm73OM3OfYdiyYTzf6Hl3lyQiIiKSbgqYRURErtfZGFg2FhZ+AjEHoXQD6PA2VO0EHh7pGnrTsU2MWj2K2btm4+flR5/afehZsyfBfsEZVLyISM5StEJtNnpUoeSun4GcHTADtA9pz32H7+ObDd9Qr1g92oe0d3dJIiIiIumigFlERORanTkGi0fB4kiIPwEVWsHto5z/NenrC7ru6DpGrRrFX3v+IsA7gL51+3J/jfsp5FsoQ0oXEcnJtpfqzE17h3Nu/xp8StVxdznpNqDBAFYfWc0rC1+hauGqhBQMcXdJIiIiImmmgFlERORqTu2Hf0bC0rGQEAvVO0OLAVCmQbqHXhW9ilGrRjFv3zwK+BSgf1h/etToQQGf9N0UUEQkN/GvfzcJez4mev54St/1vrvLSTdvT28+aPUBd/56JwPmDmDCTRPw80pfz34RERERd1HALCIicjlHt8GCj2DVd+BIgjrdoMXTUKxGuodedmgZo1aN4p8D/1AoXyGerP8k3at1J8AnIAMKFxHJXerXrMzfU0NptGUKOIaAh6e7S0q3Ev4lGNJyCP/3x//x9uK3eaP5G+4uSURERCRNFDCLiIhc7OBamD8c1v0EHt5Q735o/gQUDknXsNZalhxcwqjVo/j34L8E+QbxTINnuKvaXeT3zp8xtYuI5EKBvt6sKNyRtiffhh1/Q6XW7i4pQzQv3ZxHQh8hclUk9YvVp2uVru4uSUREROS6KWAWERE5b/dimD8MNs8AnwBo9jg0eRQCi6drWGstC/cvZNTqUaw4vIKifkV5vuHz3FH1Dn0lWkTkGvnUvIlTCz/Ed/l3+OSSgBmgX91+rDy8krcWv0WN4BpUD6ru7pJERERErosCZhERydushW1/wrzhsGs++AVB68HQ6CHwK5zOoS1/7/2bUatHsebIGkr4l+Clxi/RtUpX8nnmy6ADEBHJG5pWK83v8xpz58Zf4Vws+Pi7u6QM4enhybs3vOvsxxw1gImdJxLoE+juskRERESumYe7CxAREXELhwPW/wyjW8E3d8Cx7dBxCDy9FloNTFe47LAO/tz1J3f/djePzXmMY/HHeLXpq0zrOo3u1bsrXBYRSYOwsoWY7nEDXklnYOPv7i4nQwX5BjG01VD2x+znlQWvYK11d0kiIiIi1yxTA2ZjTEdjzCZjzFZjzKBUXr/BGLPcGJNojOl20Wu9jDFbXH96ZWadIiKShyQlwIoJMLIR/NATzsZAl0/gyVXQ5P/SNSMuyZHEjJ0z6PZrN56KeorYhFjeaP4Gv3b9lW5Vu+Ht6Z2BByIikrd4e3rgU6E5B00x581Xc5l6xerxdIOn+WP3H3y9/mt3lyMiIiJyzTKtRYYxxhMYCbQD9gL/GmN+sdauT7HabqA38OxF2wYBrwLhgAWWubY9nln1iohILnfuDKz4GhZ8DKf2Qok60G0s1LwVPDzTNfT5YHn06tFsP7mdCgUr8E7Ld+gY0hEvD3WjEhHJKM2qFGPStmY8tv0XzOmDEFjC3SVlqJ41e7Li8AqGLxtOnaJ1qFesnrtLEhEREbmqzJzB3AjYaq3dbq09B3wP3JpyBWvtTmvtasBx0bYdgNnW2mOuUHk20DETaxURkdwq7gT8PRQ+rAPTn4NC5aDHZHhkHtS+PV3hcoIjgalbp3Lrz7cyaN4gPIwH77d6nyldptC5YmeFyyIiGaxllSJMSWqBsQ5YM8nd5WQ4YwxvNH+DkgEleXbusxyLP+bukkRERESuKjPf+ZYG9qR4vhdonI5tS1+8kjGmL9AXoHjx4kRFRaWpUMneYmJidG7zKJ37vC2959/73AnK7P2F0vum45V0hqNBDdhddQAnC9WCfcC+uWkeO9EmsjhmMbNPzeZo4lHKeJfhoaIPUcevDh47PZi3c16axxb97udlOvdyNZWLBRATUIGdnjUIWTURmj3u7pIyXKBPIMMihnHftPsY9PcgPrvxMzzT+U0bERERkcyUo6dWWWtHA6MBwsPDbUREhHsLkkwRFRWFzm3epHOft6X5/J/YDQtHwPLxkHgWat0GLZ4muGQowems6WzSWaZsmcIXa7/gYOxBagfX5n+h/+OGMjdgjEnn6HKefvfzLp17uRpjDC0qF2HixmY8f+gLOLQOitdyd1kZrnpQdV5s/CKvLnyVUatH0T+sv7tLEhEREbmszAyY9wFlUzwv41p2rdtGXLRtVIZUJSIiuVP0Jpj/Iaz5ATAQ2h2aPwVFKqd76PjEeCZvnszYtWM5HHeYsKJhvNb0NZqVaqZgWUQki7WoUoQ3VjTkOb+vMKu+h/ZvuLukTNG1cleWH1pO5KpIQouG0rx0c3eXJCIiIpKqzAyY/wWqGGMq4AyMuwP3XuO2M4G3jTGFXc/bAy9kfIkiIpLj7VsO84fBht/A2w8a9YWmj0HBSzorXbczCWeYtHkSY9eO5Wj8UcKLh/N2y7dpVKKRgmURETdpXrkIxynA7uBmlF8zCW58Ld03a82OjDG81OQl1h9bz6B5g5h0yyRK+OeumxqKiIhI7pBpAbO1NtEY8xjOsNgT+NJau84Y8zqw1Fr7izGmITAFKAzcYoz5n7W2lrX2mDHmDZwhNcDr1lrd4UJERJyshZ3zYd4HsP0v8C0INwyExv3AP72NMCA2IZbvNn7H+HXjOX72OE1KNmFo3aGElwjPgOJFRCQ9ihfwpUqxAH6xN/D46b9hx1yo1MbdZWUKPy8/hrUaRvffu/Ps3GcZ22Es3p7e7i5LRERE5AKZ2oPZWjsNmHbRsldSPP4XZ/uL1Lb9EvgyM+sTEZEcxlrYPAPmDYO9S8C/GLR7HRr0Ad8C6R7+1LlTfLvhW75e/zWnzp2iRekWPFL3EcKKhaW/dhERyTAtqhRhzJKqPJa/AGbVxFwbMAOEFAzh9Wav88zcZxi2bBjPN3re3SWJiIiIXCBH3+RPRETyiKREWDcF5g+Hw+ugUDm4+QMIuw+8fdM9/MmzJ/l6/ddM2DCBmIQYIspG0K9uP2oVyX03jhIRyQ1aVC7C2AU7OVS2EyU2/ArnhoGPv7vLyjTtQ9pz3+H7+GbDN9QrVo/2Ie3dXZKIiIhIMgXMIiKSfbxfBWIPA647vUa5lhsPsA4oWgO6jobad4Bn+v8Xdiz+GOPXjee7jd9xJvEM7cq3o2/dvlQPqp7usUVEJPM0rhiMl4dhjk9r7k2Y6OzDH3q3u8vKVAMaDGD1kdW8svAVqhauSkjBEHeXJCIiIgIoYBYRkezEFS5fwjqg+7dQtRN4eKR7N0fijjBu7Th+2PwD8YnxdAzpyMN1H6ZK4SrpHltERDJfQD4v6pUrxA+HAri3UDlY/X2uD5i9Pb35oNUH3PnrnQyYO4AJN03Az8vP3WWJiIiIkP536SIiIlmh+s3pDpcPxR5iyJIhdPyxI19v+Joby93I1Num8l6r9xQui4jkMM0rF2HV/tPE17gTtkfBqQPuLinTlfAvwZCWQ9h6fCtvL37b3eWIiIiIAAqYRUQku9i/IvOGjtnPm4vepNNPnZi4cSI3VbiJX2/7lbdbvk3FghUzbb8iIpJ5WlQugrWwOPBG5zdd1kxyd0lZonnp5jwS+ghTt05lypYp7i5HRERERC0yRETEzeKOw59vwNIvM3zoPaf38MWaL/h5689goGvlrjxY50FKB5TO8H2JiEjWCi1biIB8Xsw6FEir0uGweiI0f8LdZWWJfnX7sfLwSt5a/BY1gmvo3gEiIiLiVprBLCIi7uFwwIoJMCIclo2Fxv0ybOidJ3fy0vyXuGXKLfy67VfurHYn02+fzitNX1G4LCKSS3h7etCkYhALth6B0O5waC0cXOvusrKEp4cn797wLgXzFWRA1ABOnzvt7pJEREQkD9MMZhERyXoH18Lvz8CeRVC2Mdw8FUrUgbU/pn6jP/9i1zTsthPbGL16NDN2zsDHw4d7a9xLn1p9KJq/aMbWLyIi2UKLykX4Y8Nhkv56B0+AyOYXruBfDAZucUdpmS7IN4ihrYbywIwHeGXBKwyLGObukkRERCSPUsAsIiJZJ/4URL0Di0eBXyG4dSSE3vvfzftShABRUVFERERc07Cbjm1i9OrRzN41G18vX3rV6kWvmr0I9gvO+GMQEZFso0WVIgB4xh1NfYXUPrTMReoVq8dTDZ5i6NKhfLbyM2YdnEXtuNoU8Svi7tJEREQkD1HALCIimc9a5+zkmS9BzCEI7wNtXob8Qekadv3R9YxaNYo5e+bg7+3PQ3Ue4v6a91PYt3AGFS4iItlZpaIBFC+QD865uxL36VmzJysOryBydSQWS+SqSAY3GezuskRERCQPUcAsIiKZK3qTsx3GznlQqh7c8y2UbpCuIVdHr2bU6lH8vfdvAn0C6R/an3tr3EvBfAUzqGgREckJjDG0qFwU1ru7EvcxxvBk/Sf5c/efAEzZMoV+of00i1lERESyjAJmERHJHGdj4O/34J+R4BMAnYdD/V7g4ZnmIZcfWs6o1aNYuH8hBfMV5Il6T9C9encCfQIzsHAREclJWlQJztMBM8A3G77By3iRaBM55zjHm4ve5MPWH7q7LBEREckjFDCLiEjGshY2/AIzXoRTe6HefXDj/8D/2mdSRZ+J5sODH1I7rjbBvsEsPbSUyFWRLDm4hCDfIAY0GMDd1e4mv3f+TDwQERHJCZpXytszdaPPRPPz1p9JtInJy/7c/Sezds6ifUh7N1YmIiIieYUCZhERyThHt8G0gbDtTyheB7p9CeUaX/cwkasj2X52O68tfI3T506z/PByivoV5bmGz9Gtajf8vPwyoXgREcmJihXw5ZgpRJA9cemL/sWyvJ6sFrk6Eod1XLJ84NyB+Hj6EFE2IuuLEhERkTxFAbOIiKRfQhzMGwYLPgQvX+j4LjR8CDyv/38zh2MPM2XLFCyWuXvnUsSvCC82fpHbq9xOPs98GV+7iIjkeJ/Un86ExbtY9Wp7fD2BkY3Bwwv+b4G7S8t0qw6vIsGRcMlyH08fnvrrKd5u8TY3VbzJDZWJiIhIXqGAWURE0mfTDJj+HJzYBXXugvZvQGCJ6x7GWss/B/7hhb9fSH6j7Gk8aV22NfdUvyejqxYRkVykRZVgvlywg+W7jtOschGIGAQ/PgjrpkCdbu4uL1NN7jI5+XFUVBQREREAxCbE8vicxxk0bxCxibHcWfVON1UoIiIiuZ2HuwsQEZEc6vgu+O4e+O5u8PaDXr/BHWPSFC4vPbiU3jN688jsRzh29ljy8iSbxC/bfuFI3JGMrFxERHKZxhWCMUDfr5dRYdDvtPitEKcCK0PUEEhKvOr2uZG/tz+ftv2UlmVa8vo/rzN27Vh3lyQiIiK5lAJmERG5Poln4e/3nV8/3j4X2r0O/eZDhZbXPdSq6FU8POth+szsw57Te6hfrD7eHt4XrOOwDiJXRWZU9SIikgvNXn8IYyDmbCIW2HvyLINPdIGjW2Dt5Ktun1v5evnyYesP6RTSiWHLhjFixQiste4uS0RERHIZtcgQEZFrt22O8yZ+R7dCzVuhwztQsPR1D7P+6HpGrhzJ33v/Jsg3iGfDn+Xuandz37T7LukjmeBIYOXhlRl0ACIikhu9P3MTjoty018T6vO4VwhVooZA7TvA0zv1jXM5bw9v3mn5Dvm98zN69WhizsXwfKPn8TCaayQiIiIZQwGziIhc3cl9MPNFWD8VgirBfT9B5bbXPcyW41v4dOWn/LH7Dwr4FODJ+k9yb/V7ye+dH7h8H0kREZEr2X8i7pJlFg/eO3sHY45/AKu+h/r3u6Gy7MHTw5NXm75KgHcAX63/itiEWF5r9hpeHno7KCIiIumnKwoREbm8pARY9ClEvQs2CdoMhmZPgFe+6xpmx8kdfLbyM2bsnIG/tz/9Q/tzX837CPQJzKTCRUQkLylVyI99qYTM6wObQ/AcmPse1L0bvHzcUF32YIzhmfBnCPAJYOTKkZxJPMOQlkPw8cy7PxMRERHJGAqYRUQkdTvnw+/PQPRGqHYTdBwChctf1xB7Tu8hclUkv23/jXye+XiwzoP0rtWbgvkKZlLRIiKSFw3sUI0XflpDXEJS8jI/b08GdqwOAS/BhG6w8hsIf8CNVbqfMYZ+of0I8A7g3X/f5UzCGYa3Ho6fl5+7SxMREZEcTAGziIhc6PQhmDUY1vwAhcrBPROhWsfrGuJg7EFGrR7F1C1T8fTw5L4a9/FA7QcI9gvOpKJFRCQvu62e834A78/clDyTuUfjcs7lthSUaQh/D4XQe8Hb152lZgv31bwPf29/XvvnNfrN7scnbT/Rt4pEREQkzRQwi4iIU1Ii/Ps5/PUWJMbDDc9BywHgfe2zmqLPRPP5ms+ZtHkSFku3qt14uO7DFMtfLBMLFxERcYbMt9UrTUKSg5s/nsf0tQcZ0L4q+X28oPVL8PVtsHw8NO7r7lKzha5VupLfOz+D5g3iwZkPMqrdKAr7FnZ3WSIiIpID6dbBIiICe5bAmAiY8bxzllf/RdDmpWsOl4/FH+ODpR9w0083MXHTRLpU6sK0rtN4qclLCpdFRCRLeXt68HbXOuw7EcdHf2xxLqwYAeWawbwPIOHSXs15VYeQDnzc+mO2n9xOnxl9OBR7yN0liYiISA6kgFlEJC+LPQI/PwpftIMzx+Cu8XDfjxBc6Zo2P3n2JB8v/5iOP3Zk/PrxtA9pz6+3/cprzV6jZEDJTC5eREQkdeEhQXRvWJbP5+9gw4FTYIzzg9OYg7D0S3eXl620LNOSyBsjOXjmIL1m9GLP6T3uLklERERyGAXMIiJ5kSPJ+QZ7RANY9T00fxIeXQI1b3W+Cb+KmHMxfLbqMzr92Ikxa8bQqkwrptw6hbdavEXZAmWz4ABERHImY0xHY8wmY8xWY8ygVF7vZ4xZY4xZaYyZb4ypmeK1F1zbbTLGdLjWMfOqQZ2qU9DPmxenrMHhsBDSAircAPOHw7lYd5eXrYSXCOeL9l8QkxBDr+m92HZim7tLEhERkRxEAbOISF6zbzl8fiP89jSUqAP9FkC71yFfwFU3PZNwhi/WfEHHnzry6cpPaViiIZNvmcz7rd6nYsGKWVC8iEjOZYzxBEYCnYCawD0pA2SXb621day1YcB7wDDXtjWB7kAtoCPwqTHG8xrHzJMK5ffhpZtqsGL3Cb7/1zUrt/VLEBsNS8a4t7hsqFaRWozrMA6A3jN6s+7oOvcWJCIiIjmGAmYRkbwi7jj8NgDGtIFT++D2z6HXr1Cs+lU3PZt0lq/Xf02nnzrx4fIPqVOkDt/f/D0ftfmIakHVsqB4EZFcoRGw1Vq73Vp7DvgeuDXlCtbaUyme+gPW9fhW4Htr7Vlr7Q5gq2u8q46Zl91evzRNKgYxZPoGok+fhXJNoFJbWPARnD3t7vKyncqFK/NVx6/w9/bnwZkPsuzQMneXJCIiIjmAl7sLEBGRTOZwwKpvYfYrEHcCmvwfRLwAvgWuumlCUgI/bfmJ0atHczjuMI1LNOaxeo8RViws08sWEcmFSgMpG9zuBRpfvJIx5lFgAOADtEmx7aKLti3tenwtY/YF+gIUL16cqKioNB1ATtSllIN/dyTyxNi/eKSuL4EFO9Jg259s//55dpe/y93lZaiYmJgMObf9Cvbjk0Of8PDMh3mo6EPU9NOk+Owuo8695Dw693mbzn/eld3OvQJmEZHc7OAa+P1Z2LMIyjaBm4c622JcRYIjgV+3/cqoVaPYH7uf+sXqM+SGITQs0TALihYRydustSOBkcaYe4HBQK8MGHM0MBogPDzcRkREpHfIHOWgzyY+nrOVx26qTYPKERAzh4q7f6PiXW+DXyF3l5dhoqKiyKhze0P8DfSb3Y8xR8bwbst3aR/SPkPGlcyRkedechad+7xN5z/vym7nXi0yRERyo/hTMH0QjGoFR7fCrZ9Cn+lXDZeTHEn8uu1Xbpt6G68ufJUg3yAib4xkXMdxCpdFRNJvH5DyTqhlXMsu53vgtqtse71j5kn9W1emfHB+Bk9dS3xCErR+AeJPwqLP3F1athXkG8QXHb6gTpE6DPx7IFO3TnV3SSIiIpJNaQaziEhuYi2smQyzXoKYwxD+ALR9GfwKX3Ezh3Uwe9dsPl35KdtPbqda4WqMaDOCVmVaYYzJouJFRHK9f4EqxpgKOEPg7sC9KVcwxlSx1m5xPb0ZOP/4F+BbY8wwoBRQBVgCmKuNKeDr7cmbt9Xm/i+WEDl3G0+t6OZ8Ye4Q55/z/IvBwC2pD5IHBfoEEnljJE9HPc3LC14mNiGWHjV6uLssERERyWYUMIuI5BaHN8K0Z2HnPChVH+75HkrXv+Im1lqi9kQxcuVINh3fRMWCFfmg1QfcWP5GPIy+5CIikpGstYnGmMeAmYAn8KW1dp0x5nVgqbX2F+AxY8yNQAJwHFd7DNd6PwDrgUTgUWttEkBqY2b1seUELasUpUtoKT79axtPeR9OfaXYyyzPw/J752dEmxE8//fzDFkyhJhzMfSt21cfQIuIiEgyBcwiIjnd2Rj4+z34ZyT4BEDn4VC/F3h4XnYTay0L9i9g5IqRrD26lnKB5Xin5Tt0CumE5xW2ExGR9LHWTgOmXbTslRSPn7zCtm8Bb13LmJK6wZ1r8NcmhcjXy8fTh/dbvc+rC1/lk5WfEJMQw4AGAxQyi4iICKCAWUQk57IWNvwCM16AU/ug3n1w4//Av8gVN1tyYAmfrPyEFYdXUMq/FK83e51bKt2Cl4f+lyAiIrlbsUBfnutYHWa4u5Kcx8vDizeav4G/tz/j1o0jJiGGwY0H64NpERERUcAsIpIjHd0G0wbCtj+heB3oNhbKNb7iJisPr+STFZ+w+OBiiuUvxstNXqZr5a54e3pnUdEiIiLu16NRuSsHzGdPQ77ALKsnJ/EwHrzQ6AUCvAMYs2YMsQmxvNXiLbw9dC0hIiKSlylgFhHJSRLiYN4wWPAhePlCx3eh4UPgefl/ztcdWccnKz9h/r75BPkG8XzD57mz2p3k88yXdXWLiIhkEx4eV2nr8Hk7uOdbCKqYNQXlMMYYnqj/BAE+AQxfNpy4hDiGRgzVdYWIiEgepoBZRCSn2DQdpj8HJ3ZD3buh3RsQWPzyqx/bxMiVI/lrz18UzFeQpxs8Tfdq3cnvnT8LixYREcl+4vMF43v26CXLz3kXwCfmIIxuDXeOhUpt3FBdzvBA7QcI8A7gzUVv0v+P/nzc5mP8vf3dXZaIiIi4gQJmEZHs7vhOmD4INk+HotWh9+8Q0uKyq28/sZ1PV33KzJ0zCfQO5NGwR7mvxn0E+ARkXc0iItdh6op9vD9zE/tOxFF60RwGdqjGbfVKu7ssycXams/ZFx93yfLSvn4seLgCfH8vfHMHtH8TmvQH3cwuVXdVu4v83vkZPH8wfWf15dMbP6VgvoLuLktERESymAJmEZHsKvEsLPgY5g0F4+mcsdzk/+AyPZN3n9pN5KpIft/xO76evjxc52F61eqlN3oikq1NXbGPF35aQ1xCEgD7TsTxwk9rABQyS6bZf+LScDl5eVAFeHA2THkEZr4IB9dA5w/B2zdri8whOlfsTH6v/Dw791n6zOzD6HajKeJ35RsOi4iISO7i4e4CREQkFVv/hE+bwl9vQtWO8Ni/0PyJVMPl/TH7eW3ha3SZ2oXZu2bTq2YvZtwxgyfqP6FwWUSyLWstO4/E8tov65LD5fPiEpJ4f+YmN1UmeUGpQn5XXp4vAO76GiJehFXfwbib4NT+LKwwZ2lTrg2f3vgpe0/vpfeM3hyIOeDukkRERCQLaQaziEh2cnIfzHwB1v8MQZXgvp+gcttUVz185jCjV4/mxy0/YjDcXe1uHqrzEEXzF83iokVErs5ay86jZ1i0/SiLtx9l0fZjHDwVf9n1LzfDVCQjDOxQ7YKZ8wA+nh4M7FDtv5U8PCDieSheyzmbeVgtwHHpYP7FYOCWzC86m2tSsgmj242m/5/96TmjJ2PajSGkYIi7yxIREZEsoIBZRCQ7SEqARZ9C1LtgHdBmMDR7ArwuvSP70bijfLH2C37Y9ANJjiS6VulK37p9KeFfwg2Fi4ikzlrLjiOxLNp+zBkq7zjKoVNnASgSkI8mFYNoUjGYj//cwuHTZy/Z/nIzTEUywvn2K+/P3MT+E3EYA+WC/FJvy1KjMwTNhs+apj5Y7OFMrDRnCSsWxtgOY+k7uy+9ZvRidLvRVAuqdvUNRUREJEdTwCwi4m475sG0ZyF6I1S7CToOgcLlL1ntRPwJxq0bx7cbv+Vs0lluqXgLj4Q+QtnAsm4oWkTkQtZath+JZZFrdvLi7UeTg+OigfloUjE4OVSuWMQf47ppWkA+r0tmkvp5e144k1QkE9xWr3RyoDzm7+28NW0DK3Yfp165wpeuXLxmFleXc1ULqsa4juN4eNbD9JnZh0/bfkpYsTB3lyUiIiKZSAGziIi7nD4EswbDmh+gUHm4ZyJU63jpaudOM379eL5e/zVnEs7QsUJH/i/0/6hQsIIbihYRcbLWsi36fKB8lMU7jhHtCpSLJQfKzlC5QopA+WIpZ5LuOxFH6UJ+DOxQTTf4kyx1T+NyfPLXVj6L2sbonuHuLifHq1CwAuM7jefhWQ/Td3ZfPm7zMU1KNnF3WSIiIpJJFDCLiGS1pET4dwz89TYkxkOr56HF0+B94dfBzyScYcKGCYxbN45T505xY7kb6R/WnyqFq7ipcBHJy5yBcgz/nG95sf0YR2KcgXLxAvloVik4OVQOCc5/2UA5NednkkZFRREREZFJRyByeQH5vOjVtDwfz9nK1sOnqVws0N0l5XilAkrxVaev6Du7L/3/6M8HrT6gdbnW7i5LREREMoECZhGRrLR7Mfz+DBxaA5VvhE7vQXClC1aJT4xn4qaJfLn2S47FH6NVmVY8GvYoNYJruKloEcmLrLVsPRzzX8uLHUc5EnMOgBIFfGlR+b9Aufx1Bsoi2VGvZiGMnredz6K288Fdode38ZIx0PAh0O/BBYr4FWFsh7H83x//x9NRT/NWi7e4ueLN7i5LREREMpgCZhGRrBB7BGa/Ciu/gQJl4K6vocYtF7wRPZd0jsmbJ/P5ms+JjoumacmmPFrvUUKLXuebXBGRNLDWsiU5UHbOUD4a6wyUSxb0pWWVosk9lMsFKVCW3Cc4IB/dG5bjm0W7GNC+KqUvvtGkf7HUb+jn6eO8l8K+ZdB5+CXfSMrrCuYryJj2Y3hizhO8MO8FYhNiuavaXe4uS0RERDKQAmYRkczkSIJl4+DP1+FcDDR/Clo9Bz7+yaskOBL4eevPjFo9ioOxB2lQvAHv3fAe4SXUA1JEMo/DcVGgvOMYx1yBcqmCvrSqWjR5hnLZID8FypInPHxDRb5ZtIsxf2/ntS61Lnxx4JbUN3I44O/3IOodOLQO7v4m1Zv15mX+3v58euOnPBP1DG8seoOYhBgeqP2Au8sSERGRDKKAWUQks+xbDr8PgP0rIKQl3PwBFK2W/HKiI5Hft/9O5KpI9sbspW6Rurze7HWalGyiIEdEMpzDYdl8+DSLtv3X8uL4mQQAShfyo3W1YjSuGETTisGUKaxAWfKm0oX8uDWsNN//u5sn2lYhyN/n6ht5eEDEICgZBj/1hdGt4I4voHLbTK83J8nnmY/hrYfz0ryXGL5sODHnYni83uP6t0ZERCQXUMAsIpLRzhyDOW/A0rEQUNz5JrP2HcntMBzWwcydM/l05afsPLWTGkE1GNl2JC1Lt9SbLBHJMA6HZdOh0xfMUD7hCpTLFPajTfXiyS0vygbld3O1ItnH/0VU5Mflexm3YAcD2le7+gbnVesIff+CiffBN3dA25ehxQD1ZU7B28Obd1q+Q37v/IxZM4bYhFieb/Q8HsbD3aWJiIhIOihgFhHJKA4HrPoWZr8CcSegyf9BxAvgWwBw9jeds3sOI1eNZMvxLVQuVJnhEcNpW66tgmURSTeHw7Lx4H+B8pKd/wXKZYP8uLFGcZpUDKZxhSAFyiJXULlYIO1rFuerf3bRt1UlAvJdx1um4Erw0B/wy+PO9lj7lsNtnyVfCwh4enjyatNXCfQJZNy6ccQkxPC/Zv/Dy0NvTUVERHIq/V9cRCQjHFwDvz8DexZD2SbOdhglagPOYHnevnl8suITNhzbQEiBEN5t+S4dQjrg6eHp5sJFJKdyOCwbDp5i0fZjzkB5xzFOxjkD5XJB+Wl3PlCuGESZwgqURa7H/0VUYtb6Q3y/ZDcPtax4fRv7+Du/vVQ6HGYNhjFtoPuEC9pk5XXGGAY0GECAdwCfrPyEMwlnePeGd/HxvIaWJCIiIpLtKGAWEUmP+JPw1zuwZBT4BcGtn0LoPeDhgbWWRQcW8cnKT1gdvZrSAaV5s/mb3FzxZs3SEZHrluSwbDhwyjVD+RhLdhzlVHwiAOWD89Oh1vlAOZjShfzcXK1IzlavXGGaVgxmzLzt3N+0PPm8rvMDYWOgaX8oWRcm9XaGzLd9CjVvzZR6cyJjDI+EPoK/tz/v/vsuj895nOERw8nvrQ/EREREcholHCIiaWEtrJnknJkUcxgaPghtBoNfYQCWHVrGJys+YemhpRTPX5xXmr7CbZVuw9vT282Fi0hOkeSwrN9/isU7/uuhfNoVKIcE5+emOiVpXDGIxhWCKaVAWSTD9W9difu/WMLUFfu4u2G5tA0S0gL6zoUfejr/NH8K2rwMnnobdt59Ne/D39uf1/55jX5/9GNk25EE+gS6uywRERG5DrqyERG5Xoc3wrRnYec8KN0A7p0IpeoBsCZ6DZ+s/ISF+xdSxK8IgxoNolvVbuTzzOfmokUku0tMcrD+wCkWn295sfO/QLlCEX861y1J4wrOlhclCypQFslsLSoXoXbpAkTO3U63BmXx9Ejj/RIKloY+02D687DgQ9i/ArqNBf/gDK03J+tapSv5vfMzaN4gHpz5IJHtIgnyDXJ3WSIiInKNFDCLiFyrszEw911Y9Cn4BEDnD6F+L/DwYOOxjYxcMZKovVEUzleYZxo8w93V78bPSyGQiKQuMcnBuuQZysf4d8cxTp91BsoVi/jTuW4pmrhmKJco6OvmakXyHmMM/SMq03/CcmauO8hNdUqmfTCvfHDLh84Ppn9/Bka3grvGQ+n6GVZvTtchpAP5vfLzdNTT9J7RmzHtxlDcv7i7yxIREZFroIBZRORqrIX1P8PMF+HUPqh3P9z4P/APZuvxrXy66lNm75pNoE8gj9d7nB41euDv7e/uqkUkm0lMcrB2/ykWb3e2vPh353FizgfKRf25JawUjSsE0aRiMMULKFAWyQ461CpBxSL+fBq1lU61S2BMGmcxn1f/fihey9ku48uO0HkY1LsvY4rNBVqWaUnkjZE8Nucxes3oxZj2YygbWNbdZYmIiMhVKGAWEbmSo9uc7TC2zYESdeDOcVC2ETtP7uSzv99j+o7p5PfOzyN1H6FnrZ4U8Cng7opFJJtITHKwZt9JFu9wtrxYmiJQrlTUn1vDStG4YjBNKgRRTIGySLbk6WF4pFVFnv9xDfO3HqFllaLpH7R0fegbBZMfgJ8fhX3LoOMQ5yxnIbxEOF+0/4J+f/Sj1/RejG43msqFK7u7LBEREbkCBcwiIqk5dwbmD4MFH4GXL3R6D8IfZF/cISIXvMyv237Fx9OHPrX70KdWHwr5FnJ3xSLiZgmuQHnR9qMs3n6MpTuPEXsuCYDKxQK4rV4pmlQMplGFIIoFKlAWySluq1eaYbM38+lf2zImYAbwLwL3/QRzXndeaxxc42yZUaBUxoyfw9UqUouxHcbSd3Zf+szsQ+SNkdQqUsvdZYmIiMhlKGAWEbnYpukw/Tk4sRvq3g3t3uCgh2XMknf4aetPeODBPdXv4cE6D1LEr4i7qxURN0lIcrB6rytQ3uEMlM+4AuUqxQK4vX6Z5EC5aKBmJorkVPm8PHm4ZUXe/H0DK3Yfp165whkzsKcXtHsdStV3zmQedYPzm1IhLTJm/ByucuHKfNXpKx6e9TAPznqQT9p8QniJcHeXJSIiIqlQwCwict7xnTB9EGyeDkVrQO/fOVK8Ol+s+YIfNv2AAwd3VLmDh+s8rJvOiORB5xIdrNl3gkXbnS0vlu06nhwoVy0eQLcG/wXKRQIUKIvkJt0blWPEnK18FrWN0T0zOOSsdRsUrQ4Te8BXXaD9m9Dk/yC9/Z5zgbKBZfmq41f0nd2Xfn/0Y3jEcFqWaenuskREROQiCphFRBLPwoKPYd5QMJ7Q7g2Oh3Vn7IZv+G7+0yQ4Eri18q30rduX0gGl3V2tiGSRc4kOVu89kWKG8nHiEpyBcrXigdyZIlAOVqAskqsF5POiV7MQPv5zC1sPn6ZyscCM3UGx6vDwXzClH8x8wdmXucvH4KObBhf3L87YjmPpN7sfT/z1BENaDqFDSAd3lyUiIiIpKGAWkbxt6x8wbSAc2w61unIq4gW+2juLb6beQlxiHDdXvJl+of0oX6C8uysVkUx2NjHJ2fJimytQ3nWM+AQHANVLBHJ3w7I0qRhEowrBBPn7uLlaEclqvZuF8NlfW7llxALiE5IoVciPgR2qcVu9DPrw2bcA3P2N8x4Qc96Ewxvg7q8huFLGjJ+DBfkG8UWHL3jsz8d47u/nOJNwhq5Vurq7LBEREXHJ1IDZGNMR+AjwBD631g656PV8wHigAXAUuNtau9MY4w18DtR31TjeWvtOZtYqInnMyb0w4wXY8AsEVyb2nu/4Jn4PX83uw+mE07Qv357+Yf2pVEhv6kRyq7OJSazac76HsrPlRcpAuXvDcskzlBUoi8jfm6NxQPI3GfadiOOFn9YAZFzI7OEBNzwLpcLgx4dgdGu4YwxU1YzdQJ9AIttF8vRfT/PKwleITYjlvpr3ubssERERIRMDZmOMJzASaAfsBf41xvxirV2fYrUHgePW2srGmO7Au8DdwJ1APmttHWNMfmC9MeY7a+3OzKpXRPKIxHOw6FOY+x5YB3ERL/B9UDBfrniHE2dPEFE2gkfDHqV6UHV3VyoiGexsYhIrdzt7KJ8PlM8mOjAGqpcowD2NXIFySBCFFSiLyEXen7mJJIe9YFlcQhLvz9yUcQHzeZVvhL5RMPE++PYuiHgBbnjOGUDnYX5efnzc5mMGzRvEu/++S0xCDI/UfQSjftUiIiJulZkzmBsBW6212wGMMd8DtwIpA+ZbgddcjycDnxjn1YEF/I0xXoAfcA44lYm1ikhesGMe/P4MHNnE2aqdmFStGZ9v/Ymju47SvFRzHg17lDpF67i7ShHJIPEJSazc4+yhvGj7UVbsPpEcKNcoUYAejcu7Wl4EUSi/AmURubL9J+Kua3m6FQ6BB2fDb09D1DuwbzncPhr8CmXO/nIIH08f3rvhPV5d+CojV44k5lwMz4Q/o5BZRETEjTIzYC4N7EnxfC/Q+HLrWGsTjTEngWCcYfOtwAEgP/C0tfbYxTswxvQF+gIUL16cqKioDD4EyQ5iYmJ0bvOojDr3PmePUWnbWIof/ptTvsX5vPpdTEncyIm1Y6iSrwr3F7+fSt6VOLruKFGkf3+SMfS7n3el9dyfS7JsO+Fg47EkNh5LYttJB4kOMEC5Ah5ElPagerA3VQt74u+dBByG6MOsjM7oI5C00u+9ZGelCvmxL5UwuVQhv8zbqbcf3PYZlG4AMwbB6AjoPgGK18q8feYAXh5evNH8DQK8A/hq/VfEJMTwcpOX8fTwdHdpIiIieVJ2vclfIyAJKAUUBuYZY/44Pxv6PGvtaGA0QHh4uI2IiMjqOiULREVFoXObN6X73Cclwr9j4J+3SUyMZ0qDOxl1djf7YhcRVjSMx+o9RuOSF3/uJdmFfvfzrms99/EJSSzffZxF24+xaPtRVu45wblEBx4GapYqQO9awTSpGEzDkCAK5vfO/MIl3fR7L9nZwA7VeOGnNck9mAF8PD0Y2KFa5u7YGGj0MJSoAz/0gs9vhC4joE63zN1vNudhPBjUaBD+3v6MWTOG2IRY3m75Nt4e+vdeRET+n737jo6q+to4/j0z6YX0XqhphBB67713CEUFpCiCoCgqKip2ARFEkCYCiobeewtFBOkEAqFD6L23lPP+McEX/aGgJrkp+7PWXWbu3Jk8eCGZ2XPu3iKrZWaB+TQQ9MjtwPR9jzvmVHo7DBcsw/46Asu01snABaXUL0AZ4ChCCPE0Tm6Bxa+Rej6eZYXK8a2DiRNXtlDUoyjvVBhElYAqcimlENnMvJ2nGbo8kdPX7hKweQ0D6of9oa/pveRUdpy4mt7y4oqloJxqKShH+rvQuWJ+KhTyoEwBd1zspcAghMhYD38eDV2eyJlrdzGZFC4OVjQu7pc1AYIrwAvrYGYXmN3N0jKj7mAw592fd0op+pbqi5ONE19t/4o7KXf4svqX2FnZGR1NCCGEyFMys8C8FQhRShXEUkhuj6Vw/KgFQGfgV6ANsEZrrZVSJ4FawA9KKUegAjAiE7MKIXKL25dg5fuk7fqR1Z6BjI4ow5F75wixCWFkuZHUDKophWUhsqF5O0//YWXg6Wt3eWvOHg6ev4nZpNh89DK7k67/XlAuFuBCl8oFqFDInTIF3Mlnl3cLLEKIrNOiZMDvheY1B87z/ORtTP31BN2qFMyaAM6+0HkhrHgXNo+Gs7ug7WRw8s6a759NPV/seZysnfh488e8tPolRtUahaO1o9GxhBBCiDwj0wrM6T2V+wDLATMwSWu9Tyn1IbBNa70A+A5LEfkwcAVLERpgNPC9UmofltaJ32ut92RWViFELpCWCtsno1d/yDpzMqNDojiQcp2Cto4MLTeUegXqYVJ5e/K6ENnZ0OWJf7jsHOBechpj4o5gNimKBbjQtXIBKhTyoHQBNykoCyEMVzPMm+qhXoxYdZAWJfzxcLLNmm9stoaGX4B/KVjYD74MA50GQA3g93ESjt4w4FDWZMoG2oW1w9HakXc2vkOPFT34ts63uNi6GB1LCCGEyBMytQez1noJsORP+9575Ot7QNvHPO7W4/YLIcRjnd6OXtyfX68e4BvfQOLVAwLtnPm0xJs0KthIBr4IkY2lpml+O3blsYOzHtr1Xl2cpaAshMhmlFIMahJBgxEb+HLlQT5tGZW1AaJjwDsCxlV9/P23L2RtnmygcaHGOFg58Pq61+myrAsT6k3A097T6FhCCCFErpddh/wJIcST3bkCqz9k676f+cbTkx2+3vg5evBB8RdoVqSZDHkRIptKS9NsO3GVxXvOsGTvOS7evI8C9GOODXC1l+KyECLbKuLtzHMVC/D9pmN0Kh9MpH8Wr5j1K5613y8HqBlck9F1RtN3TV86L+3MhHoT8HfyNzqWEEIIkatJgVkIkfOkpcGuaexaN5hv7BVb/LzxsvPg7egXaB3SGhuzjdEJhRB/kpam2Zl0lUV7zrIk/iznb9zH1spErXBvGhf34879FN5fkPCHNhn21mYG1A8zMLUQQjxZv9ohzNt1mg8XJhDbs4LMesgGKvhVYEK9CfRa1Yvnlj7HhHoTKOiSRX2yhRBCiDxICsxCiJzl7B72Le7L6ORTbHC3x90mH68X70lMWIxMDBcim9Fasyvp2u9F5bPX72FjZaJGqBeNi/tRJ8IHR9v/fyliY2Vm6PJETl+7S4CrPQPqh/0+TEsIIbIrFwdrXqsXyjtz97Ik/hyNi/sZHUkA0V7RfF//e3qu7EmXZV0YV3cc4e7hRscSQgghciUpMAshcoZ71zm44i3GnFrJakd78tm50a94TzpGdMLB2sHodEKIdFpr4k9fZ9Gesyzec5bT1+5iYzZRLdSTNxqEUSfC5y9bXrQoGUCLkgHExcVRo0aNrA0uhBD/Qfuywfzw6wk+XbKf2hHe2Flnk/kPv02Ast0hj66qDnMPY0qDKfRY2YPnlz3PmDpjKOFdwuhYQgghRK4jBWYhRPYyNARuX+Ci2cRkL0+KbbjEDZOJb11dWe5oj6OTMy8VfZZnivfA2cbZ6LRCCCxF5X1nbliKyvFnSLpyFyuTomqIJ/3rhlKnqA8u9tJHWQiRe5lNivebRtJhwmbGrz9K39ohWffNHb0fP9DPbANLXoezu6DRl2CdN6/0KuBSgKkNptJjZQ96ruzJyJojqehf0ehYQgghRK4iBWYhRPaS/gZprKsLO+xs6errzUlra2y1plvBZnSp8CYutlk8QEcI8T+01uw/e5PF8WdYvOcsxy/fwWxSVC7iycu1QqhX1AdXB+mHLoTIOyoW9qBRlC9j4g7TpnQg/q72WfONBxz6/cs/XAGSlgZxn8L6oXBhP8T8CPny5rA7Pyc/JjeYTM+VPem9ujfDqg+jVnAto2MJIYQQuYYUmIUQ2c5Fs4k5zk5opThubU3bm7foffU6Hl0/NTqaEHle4rmbLN5zhkXxZzl68TYmBZUKe/Ji9cLUj/TFzVGKykKIvGtgwwhW7b/AF8sOMLJ9SWPDmExQ613wLQ7zesG46tBuKuTPm6t3Pe09+b7+97y06iX6x/Xno8of0bRwU6NjCSGEELmCFJiFENlOf29PUtJ7BVoBJsAjLc3QTELkZYcv3Py9p/KhC7cwKahQyINuVQrSINIXDydboyMKIUS2EOTuwAvVCjFqzWGerZCfMgXcjY4ERZuBZyjEdoQpTaDhF1CmW57sy+xi68L4euPpu6Yv72x8hzvJd4gJjzE6lhBCCJHjSYFZCJFt6NRUhrm5ssvu/3sEpijFPCdHXrx2HU8DswmR1xy9eIvFe86yOP4sB87dRCkoV8Cdj5pHUr+YL97OebOXpxBCPEmvGoWZue0UgxcmML93ZUymbFDI9Q6HHmtgTg9Y/Bqc2QWNvwSrvPcBoaO1I2PqjOH1uNf5eMvH3Eq+RbeobkbHEkIIIXI0KTALIbKFtJT7DJnRhGmu+VBaox9ZVZOGYqyrC+8amE+IvOD4pdssjj/Loj1n2X/2BgBlC7jxQdOiNIzywyefFJWFEOJJHGyseKthOK9M38WsHadoVybI6EgW9q7QIRbWfgobhqX3Zf4hT/ZltjXbMrzmcN7Z+A4jdozgVvIt+pbsi8qDq7qFEEKIjCAFZiGE4VLuXef9GY1YoG/glprGVbPpD/cnmxS77B0NSidE7pZ05Y6l/UX8GfaethSVSwW7MqhJURpF+eLnkkVDqoTIgZRSDlrrO0bnENlP8xL+TP31OEOWJdKwmC/OdtZGR7IwmaH2IPCLhrkv5um+zNYmaz6r8hmO1o5MjJ/IrQe3GFh+ICZlevKDhRBCCPEHUmAWQhjqwc1zDJjVmDWmB/T2rswLDb79ffXIHyahCyEyzOlrd1m85wyL95xl96nrAEQHufJu4wgaRvkR4CpFZSH+jlKqEjARcAKClVLRwAta65ee8LgGwEjADEzUWn/+p/v7A92BFOAi8LzW+oRSqibw1SOHhgPttdbzlFKTgerA9fT7umitd/3HP6L4j5RSvN80kuajf+GbtYcZ2DDC6Eh/VLQZeIY80pd5CJR5Ps/1ZTabzLxX4T2crZ35ft/33Em5w+BKg7EyydtkIYQQ4p+Q35xCCMPcuXyIfvPasNkqjbeCm9Kp5qdGRxIi1zpz7S5L4i09lXeevAZAVIALAxuG0yjKjyB3B2MDCpGzfAXUBxYAaK13K6Wq/d0DlFJmYDRQFzgFbFVKLdBaJzxy2E6gjNb6jlKqFzAEiNFarwVKpD+PO3AYWPHI4wZorWdlyJ9MZJjoIFfalA5k0sZjtC8bTEHPbHY1lncE9FgLs7vD4v5wdhc0Gpbn+jIrpXi19Ks42TgxaucobiffZki1IdiYbYyOJoQQQuQYUmAWQhji+ultvLS0C/us4JOwLjSr8LrRkYTIdc7fuGcpKu85y7YTVwGI9M/HGw3CaBzlR36PbFbsECIH0Von/alfa+oTHlIOOKy1PgqglIoFmgO/F5jTC8kPbQaeeczztAGWSmuOnOGN+mEsjT/LJ4v3M7FzGaPj/C97V+g4HdZ+Ahu+hPMJebIvs1KKnsV74mjtyOe/fU6f1X0YUXMEDtby4asQQgjxNKTALITIchcPr6RnXD9OWJn4smR/akc/b3QkIXKNCzfvsWzvORbtOcvW41fQGsJ9nXm9XiiNovwo5OVkdEQhcoOk9DYZWillDfQD9j/hMQFA0iO3TwHl/+b4bsDSx+xvDwz/075PlFLvAauBt7TW9//8IKVUT6AngI+PD3FxcU+IKzJKowImZu4/zzezVlHMM3Pfft26devfnVtzNTwjrYjYP5KUbyqxL/JNbrhks7YeWSCAADp5dOKnsz/RYXYHXvB+AQdTzigy/+tzL3I8Ofd5m5z/vCu7nXspMAshstSp+Fh6/vYhl6zMjKn4ERXCWhgdSYgc79Kt++lF5TNsOWYpKof6OPFK7VAaF/eliLez0RGFyG1exNJLOQA4jaVdRe+MenKl1DNAGSy9lR/d7wdEAcsf2T0QOAfYAOOBN4EP//ycWuvx6fdTpkwZLTMOsk7FKqn89tV65p808ULLqlibM2+I3H+bX1EDLrTEHNuRUrsHQcMv8mRf5hrUoNTxUry54U2+v/094+qOw93O3ehYTySzS/IuOfd5m5z/vCu7nXspMAshssyRLaPpuXc098xWTKw1iuLB1Z/8ICHEY125/YDl+yxF5V+PXCZNQyEvR16uFUKT4n6E+khRWYjMkN5LeaTWutM/fOhpIOiR24Hp+/78/HWAd4Dqj1mJ3A6Yq7VOfrhDa302/cv7SqnvAek5lc3YWpl5p1EEPX/YzrTNJ+hSuaDRkf6adwT0WAOze+Tpvsz1CtTDwdqBV9e+SpdlXRhfdzy+jr5GxxJCCCGyLSkwCyGyxN6179Pr2EyszTZMbjCZEN8SRkcSIse5ducBK/adZ+GeM2w6cpnUNE0BDwdeqlGEJtF+hPk4o/LYSjMhsprWOlUplV8pZaO1fvAPHroVCFFKFcRSWG4PdHz0AKVUSWAc0EBrfeExz9EBy4rlRx/jp7U+qyz/+FsAe/9BJpFF6hb1oUoRT4avPEizEgG4O2bjAXL2bpa+zGs+ho3D4cJ+aPcD5PMzOlmWqhJQhbF1x9J7dW86L+3MxHoTCcoX9OQHCiGEEHmQFJiFEJlLa7Yu7k2fi+tws7JnQpPpBLkXMTqVEDnG9bvJrNh3jsXxZ9l46BIpaZpgdwd6VitEk+J+FPXLJ0VlIbLeUeAXpdQC4PbDnVrrP/dG5pH7UpRSfbC0tzADk7TW+5RSHwLbtNYLgKGAEzAz/d/1Sa11MwClVAEsK6DX/emppymlvAAF7MLSvkNkM0opBjUpSqOvN/DVyoN81KKY0ZH+nskMdd4Hv2iY9xKMr24pMgf/Xdvw3Ke0T2m+q/8dL658kc7LOjOu7jhC3EKMjiWEEEJkO1JgFkJkntQU4uY+w2s39xJk7cz4FnPxdspbq1+E+Ddu3ktmZcJ5Fu85y/pDF0lO1QS62dOtakGaRPlTLECKykIY7Ej6ZgKeuh+N1noJsORP+9575Os6f/PY41h6Pv95f62n/f7CWGG+zjxTPpgfNp+gU4Vgwn3zGR3pySJbgGcoxHaEyY2h0RBLX+Y8JNIjkskNJtNjRQ+6Lu/K2DpjKeaZzT8gEEIIIbLYUxWYlVKVgQ+A/OmPUYDWWhfKvGhCiBwt+S4Lp7diUEoSEXYefNtiHq72bkanEiLbunU/hdX7z7Noz1nWHbzIg5Q0/F3s6FKpAI2L+xMd6CJFZSGyCa31YACllFP67VvGJhI5xat1Q5m/+wyDFyTwU4/yOePnuk9R6LkWZneHRa/CmV3QaGie6stc2LUwUxpOoceKHnRf0Z1RtUZR1res0bGEEEKIbONpVzB/B7wKbAdSMy+OECJXuHuNn2Ob8qnpGuUcg/i6xSwcrR2NTiVEtnPnQQqr919g8Z6zrE28wP2UNHzz2fFM+fw0Lu5HySBXTKYcUHwQIo9RShUDfgDc029fAp7TWu8zNJjI9lwdbOhfN5T35u9j+b7zNCiWQwbH2btBxxl/6ss8NU/1ZQ5yDmJKgyn0XNmTXqt6MbzGcKoFVjM6lhBCCJEtPG2B+brWemmmJhFC5Ar6+hkmzmjK1zYPqOkSztCmP2JrzjsrXIR4krsPUlmbaCkqrz5wnnvJaXg529KhXDCNi/tROthNispCZH/jgf5a67UASqkawASgkoGZRA7RsVwwP24+wSdLEqgR5oWdtdnoSE9H+jLj4+jD5AaTeXHVi/Rb04/Pqn1GgwINjI4lhBBCGO5pC8xrlVJDgTnA/Yc7tdY7MiWVECJH0hcPMXxOKybbQVOvsnzYYDxWJmn1LsS95FTiEi+yaM8ZVu+/wN3kVDydbGhbOojGxf0oW8AdsxSVhchJHB8WlwG01nFKKblURzwVK7OJ95tG0mniFr7beIzeNXPY8OP/6cs8FMp0NTpVlnGzc2NivYn0Wd2HN9e/yZ3kO7QKaWV0LCGEEMJQT1v5efixdJlH9mlAhooIIQBIPb2djxY8w2wHKzoE1uGtWl9iUiajYwlhmHvJqWw4dIlFe86wKuE8tx+k4u5oQ8tSATQp7kf5gh5SVBYi5zqqlBqEpU0GwDPAUQPziBymchFP6hX1YfTaw7QpHYhPPjujI/0zD/syz+oGi16Bs7ug4ZA805fZ2caZsXXH8mrcq7y/6X1uJ9/m2aLPGh1LCCGEMMxTFZi11jUzO4gQIudKPryKt1a+xAoHW14o0pbelQbljKE1QmSwBylpbDh0kcV7zrIy4Tw376fg6mBN02h/mhT3p0Ihd6zM8sGLELnA88BgLFf3aWBD+j4hnto7jSOoO3w9Xyw7wPB2JYyO88/Zu0GnmbDmI9j4FZxPgJgfwDmH9JX+j+yt7BlVcxRvbniTIVuHcOvBLV6MflFeAwshhMiTnqrArJRyAd4HHk4xWAd8qLW+nlnBhBA5w5090+m/aRC/ONjyerGedC79stGRhMhSyalpbDx8icV7zrJi3zlu3Eshn50VDYr50iTan0qFPbCWorIQuYrW+irQ1+gcImfL7+FIt6oF+TbuCM9WyE/JYDejI/1zJjPU+eD/+zKPq24pMgeVMzpZlrA2WzOk2hA+2PQBY3aP4VbyLV4v87oUmYUQQuQ5T9siYxKwF2iXfvtZ4HtAmk0JkYfd+PUb+uz5mt12tnxY5k1aRj5jdCQhskRKahqbjlxm8Z6zLNt3jut3k3G2s6JeUV+aFPejchFPbKykqCxEbqWUWgm01VpfS7/tBsRqresbGkzkOL1rFmHW9lMMXpjAnF6Vcu6Q18iW/9+X+ftGeaovs5XJig8rf4iTjRNTE6ZyO/k2gyoMwmzKIcMbhRBCiAzwtAXmwlrr1o/cHqyU2pUJeYQQOYHWXFr9Pr2OTuewnR1Dq3xKvcJNjE4lRKZKSU1jy7ErLNpzlmV7z3L1TjJOtlbULepD4yg/qoZ6YmslbyaFyCM8HxaXwbKiWSnlbWAekUM52VrxZoNwXp+5m3m7TtOqVKDRkf49n0josRZmd0/vy7w7vS+zjdHJMp1JmXiz7Js4Wjsyfs94biff5tMqn2JttjY6mhBCCJElnrbAfFcpVUVrvRFAKVUZuJt5sYQQ2VZaKmcX9qHHxbVcsLXjm1qjqBxU7cmPEyIHSk3T/HbsCovjz7Bs7zku3XqAg42ZOhE+NC7uR/VQL+yspagsRB6UppQK1lqfBFBK5cfSi1mIf6xVyQB++PU4ny89QP1IXxxtn/YtWjbk4P7HvswXEqDd1DzRl1kpxcslX8bJ2onh24dzJ+UOX1b/EjurHDbAUQghhPgXnvbVSy9gSnovZgVcAbpkVighRDaVcp+js56l5+147tg4ML7+d5TwKWl0KiEyVFqaZtuJqyzec4Yle89x8eZ97K3N1IrwpkmUHzXDvaWoLIR4B9iolFqH5bVxVaCnsZFETmUyKd5rGknrbzcxJu4wA+qHGx3pv3nYl9m3OMzvnef6Mnct1hVHa0c+3vwxvVb1YlStUTjZOBkdSwghhMhUT1Vg1lrvAqKVUvnSb9/IzFBCiGzo3g0SprfhxdRTKFtnvm/0A2HuYUanEiJDpKVpdiZdZeHusyzde5bzN+5ja2WiVrg3jYv7USvcGwebHLyiTAiRobTWy5RSpYAKWFYuv6K1vmRwLJGDlc7vRsuSAUzYcIyYMsEEezgYHem/K9bK0pd5eidLX+bGw6B0F6NTZYl2Ye1wsnbi7Y1v02NFD76t8y2udq5GxxJCCCEyzd++W1ZKPaO1/lEp1f9P+wHQWg/PxGxCiOzi1gW2/9ScPtY3cLZzY0KTn8ifL7/RqYT4T7TW7Ey6xuI9Z1kSf5az1+9hY2WiRqgXTaL9qR3unbMvUxZCZLj0VhjXtNbXtdaXlFK3gRZAmFLqG631A2MTipzszQbhLNt7jk+X7Gfss6WNjpMxfIul92XuBgv7wZldeaYvc6NCjXCwduC1uNfourwr4+uOx8vBy+hYQgghRKZ40jtnx/T/Omd2ECFENnX1OBt+bk5/+1T8HH0Z33gavo65v4+eyJ201uw5dZ3F8WdZvOcsp6/dxcZsolqoF282CKd2hDfOdjKQRwjxl2YALYHrSqkSwEzgMyAaGAN0Ny6ayOl8XezoXbMww1YcZNPhS1Qq4ml0pIzh4A6dZsHqD+GXEXmqL3ONoBqMqTOGl9e8TOdlnZlQbwIBTgFGxxJCCCEy3N8WmLXW49L/Ozhr4gghspVz8Syb2ZaBzlaE5CvA2EZTcbdzNzqVEP+I1pp9Z26waM9ZFsefIenKXazNiqohXvSvG0qdoj642EtRWQjxVOy11mfSv34GmKS1/lIpZQJ2GRdL5BbdqxYidmsSHy5KYNHLVbAym4yOlDFMZqg7GPyKw/w+ML4GtPsBgsoanSzTlfcrz4R6E+i1qhedl3ZmfL3xFHIpZHQsIYQQIkM91SsWpdQQpVQ+pZS1Umq1UuqiUuqZzA4nhDDQ8V+YNb0FbzhbEe1elO+a/CzFZZFjaK1JOHODocsPUHNYHE1GbWTihqMU8nRiSJvibHunLpO6lKV16UApLgsh/gn1yNe1gNUAWus0Y+KI3MbO2sw7jSI4cO4mP29NMjpOxivWGrqtBLMNTG4E26cYnShLRHtF833970lJS6Hrsq7sv7zf6EhCCCFEhnra5pL1tNZvKKVaAseBVsB64MfMCiaEMNCBxUxa1puv3Jyp5lOWL+uMwc7KzuhUQvwtrTUHz99i8Z4zLIo/y9GLtzGbFJUKe/Bi9cLUj/TFzTH393wUQmSqNUqpGcBZwA1YA6CU8gOk/7LIEA2K+VKhkDvDVyTStLgfrg657HeXbzHoGZfel7kvnN0FDb7I9X2Zw9zDmNxgMj1W9qDb8m6MqTOGEt4ljI4lhBBCZIinvebqYSG6MTBTa309k/IIIQymt09hxPKX+MrNmYZBtRhRb5wUl0W2dvjCTUasOkjdr9ZTf8R6vll7GN98dnzSshi/vV2bH7qVp325YCkuCyEywivAHCwLLqporZPT9/sC7xiUSeQySineaxLJ9bvJjFh1yOg4meNhX+bK/WDbJJjSFG6eNzpVpivgUoCpDabibu9Oz5U92XRmk9GRhBBCiAzxtCuYFymlDgB3gV5KKS/gXubFEkJkOa1J2/Aln+z+hhmu+WhbpCXvVHwfs8lsdDIh/sfRi7csPZX3nCXx/E2UgnIF3OncPJIGxfzwcrY1OqIQIhfSWmsg9jH7dxoQR+RiRf3z0aFcMD9sPkGn8sGE+OTCmesmM9T9EPyi0/syV4eYHyGwjNHJMpWfkx+TG0zmhZUv0Gd1H4ZWH0rt4NpGxxJCCCH+k6daway1fguoBJRJX6lxG2iemcGEEFkoLY3kpW/yVvwYZuRzplvRLgyqNFiKy8IQ83aepvLna+iy7DaVP1/DvJ2nATh+6Taj1x6m4cgN1PpyHcNXHiSfvRWDm0WyZWBtpr9QkWcrFpDishBCiFyhf91QHG3MfLgoActnG7lUsdbQbYWlL/P3DWHHVKMTZTpPe08m1Z9EhEcEr8W9xsIjC42OJIQQQvwnf7uCWSlVS2u9RinV6pF9jx4yJ7OCCSGySMoD7s17kdcurme9kyOvlOxHt+LdjU4l8qh5O08zcE48d5NTATh97S6vz9zNsOUHOHXNcuFMqWBX3mtSlEZRfvi6SPsWIYQQuZOHky2v1Anlw0UJrN5/gTpFfYyOlHl8oyx9mWc9DwtehjO7oMHnubovs4utCxPqTqDvmr68vfFtbiffpn14e6NjCSGEEP/Kk1pkVMcyvKTpY+7TSIFZiJztwW1uxXakz50EdjjYM6jCINqFtTM6lcjDhi5P/L24/FBKmub8zfu82ziChlF+BLjaG5ROCCGEyFrPVszPtC0n+HhxAlVDPbG1ysVXlz3sy7x6MGz6Gs7vg3ZTwTn3FtYdrB0YXWc0r8e9zidbPuFW8i26R8lCDyGEEDnP37bI0Fq/n/7fro/Zns+aiEKITHHnClemNOH5ewfYbe/AF9WGSHFZGEZrzZajlzl97e5j709J1XSvWkiKy0IIQymlmiulej9ye4tS6mj61sbIbCJ3sjabeK9pJMcv32HyL8eNjpP5zFZQ7yNo/R2c3Q3ja8CpbUanylS2ZluG1xxOo4KNGLljJCO2j8jdLVGEEELkSk/Vg1kp9alSyvWR225KqY8zLZUQInNdS+Lc9/Xoos5x1M6BkbVH0bBgQ6NTiTzo+p1kJm08Rt2v1hMzfjPqL47zl8KyECJ7eANY8MhtW6AsUAPoZUQgkftVD/Widrg3o9Yc5sLNPDJnPaoNdF9pKTh/3xB2/GB0okxlbbLm0yqf0ja0Ld/t/Y5PtnxCmk4zOpYQQgjx1J6qwAw01Fpfe3hDa30VaJQpiYQQmevCAU5Ork9nu7tctHNiXP2JVAusZnQqkYdordl58iqvz9xNuU9X8eGiBJxsrRjapjhftC6OvfUfL/+1tzYzoH6YQWmFEOIPbLTWSY/c3qi1vqy1Pgk4GhVK5H7vNI7gfkoqw5YnGh0l6/hGQc91kL8SLOgDi/pDygOjU2Uas8nMoAqD6BrZlemJ03l347ukpKUYHUsIIYR4Kk/qwfyQWSllq7W+D6CUsseyYkMIkZMk/Ubi9Bhe8HAkzTYf39WfSFGPokanEnnErfspzN91mmmbT5Jw9gYONmZalw6kY7lgigW4/H6cjZWJocsTOX3tLgGu9gyoH0aLkgEGJhdCiN+5PXpDa93nkZteWZxF5CGFvJzoWrkgEzYc5ZkK+Ske6Gp0pKzh4A6dZv9/X+YLCdB2Sq7ty6yU4tXSr+Jk48SonaO4nXybodWHYmPOvcMOhRBC5A5PW2CeBqxWSn2ffrsrMCVzIgkhMsWhleya25WXvN1wsPdkfIPvKORSyOhUIg9IOHODaVtOMG/naW4/SCXCLx8ftyhG8xL+ONtZ/8/xLUoG0KJkAHFxcdSoUSPrAwshxF/bopTqobWe8OhOpdQLwG8GZRJ5RJ9aRZiz4xSDFyYw68WKKPVXjaVymYd9mf2iYX4fS1/mmB8hsLTRyTKFUoqexXviaO3I5799Tp/VfRhRcwQO1g5GRxNCCCH+0lMVmLXWXyildgN10nd9pLVennmxhBAZavd0Ni1/lVe8PfB2DmR8/e/wd/I3OpXIxe4lp7Joz1mmbTnBzpPXsLUy0aS4P50qBFMyyDXvvCkWQuQ2rwLzlFIdgR3p+0pjubKvhVGhRN6Qz86aAfXDeHN2PAt2n6F5iTx2dU9UG/AMhemd4PsG0Hg4lHrW6FSZplNEJ5ysnXhv03u8sPIFRtcZTT6bfEbHEkIIIR7raVcwA+wHUrTWq5RSDkopZ631zcwKJoTIIJu+YeXGj3nDx4vCrkUYW28CnvaeRqcSudThC7f4actJZm1P4sa9FAp7OTKoSVFalwrA1UEu7xRC5Gxa6wtAJaVULSAyffdirfUaA2OJPKRN6SB+2HyCz5ceoG5RHxxs/snbuVzAr7ilL/PMLpa+zGd3Qf3PwCp3vsZoXqQ5DtYOvLH+Dbot78bYOmPxsPcwOpYQQgjxP57qFYlSqgfQE3AHCgMBwFigduZFE0L8J1rDqveZu3siH/h4UdyzOKPrfisrH0SGe5CSxvJ955i25QSbj17B2qxoUMyPTuWDKV/QXVYrCyFyDaVUWcBTa70UWPPI/obABa31dsPCiTzBbFK83zSStmN/Zey6o/SvG2p0pKzn4A7PzIHVH8CmUXB+H7SbCk7eRifLFHXz1+WbWt/wytpX6LKsCxPqTcDX0dfoWEIIIcQfmJ7yuN5AZeAGgNb6EJA7f4MLkRukpsD8PkyNn8R7Xh5U8KvIuHoTpLgsMtTJy3f4YtkBKn62mpd/3snpa3d5s0E4vw6szagOJalQyEOKy0KI3OYLIOEx+xOAoVmcReRRZQu40zTan3HrjnDq6h2j4xjDbAX1PobW38GZXTCuOpzKvZ/vVA6ozNi6Y7l09xKdl3bm5I2TRkcSQggh/uBpC8z3tdYPHt5QSlkBOnMiCSH+kwd30LGd+Ob4QoZ6uFE3f11G1f5GBoOIDJGSalmt/Nyk36g+bC3j1h2hdH43pjxfjnWv16RXjcJ4OtkaHVMIITKLs9b6xJ93pu+T/lMiy7zVMByl4LOlB4yOYqyoNtBtBZis4PuGsPNHoxNlmtI+pfmu/nfcSblD52WdOXT1kNGRhBBCiN89bYF5nVLqbcBeKVUXmAkszLxYQoh/5e5V0n5syeeXtzDOzYVWIa0YWm0oNubc2ZdOZJ2z1+/y1cqDVPliLS/8sJ2D527Sr3YIv7xVi/HPlaF6qBcmk6xWFkLkem5/c598kiuyTICrPS9WL8ziPWfZcvSy0XGM5VccesZBcAWY3xsWvw6pyUanyhRFPYoyucFkTJjourwr8RfjjY4khBBCAE9fYH4TuAjEAy8AS4B3MyuUEOJfuHGWlO8b8u7dw/zk4kznop35oOIHmE1mo5OJHCotTROXeIEeU7dR+fM1fL3mEGG+zox/tjQb36zJK3VC8XOxNzqmEEJkpVVKqU/UI/1/lMWHPNKTWYis8EK1wvi72DF4YQKpaXn84lJHD0tf5op9YOsEmNIMbl0wOlWmKOxamCkNp+Bk7UT3Fd3Zem4rF+9cZMS5EVy6e8noeEIIIfKoJw75U0qZgX1a63BgQuZHEkL8Y5cOc/+Hlgywf8BaewdeLvkyPaJ6SP9b8a9cvHmfmduT+Pm3kyRduYuHow0vVC9Mh7LBBHvIAj0hRJ72GjAROKyU2pW+rwSwFehuUCaRR9nbmBnYKIKXf97JjG1JdCgXbHQkY5mtoP4n4FcCFrwM42tAzA8QUNroZBku0DmQKQ2m8MLKF+i1qhflfMtx9P5Rxu4ey7sVZB2YEEKIrPfEArPWOlUplaiUCtZayzQBIbKb0zu4/VMb+rnYscXWhrfLv02H8A5GpxI5jNaazUevMG3LCZbvO0dyqqZiIQ/ebBBOvaK+2Fg97QUvQgiRe2mtbwMdlFKFgMj03fu01kcNjCXysCbF/Zj663GGLU+kUZSf0XGyh+JtwSsMYjvBpIbQ5Cso2cnoVBnOx9GH7xt8T7cV3dhwegMA8w7P48XoF/G0l5bwQgghstYTC8zp3IB9SqnfgNsPd2qtm2VKKiHE0zmylusznqWXtxsJ1iY+rfwxTQs3NTqVyEGu3XnA7B2nmbblBEcv3sbF3prnKhagQ7lging7GR1PCCGynfRh15FAePourZQ6qbVOMTCWyKOUUrzfNJImozZS+fPV3LqfSsDmNQyoH0aLkgFGxzPOw77Ms7rA/Jfg7C6o/ymYrQ0OlrHc7Nwo5lGMw1cPo9E8SH3A5799zrDqw4yOJoQQIo952gLzoExNIYT45/bO4cL8F3nB34+TVma+qv4lNYNrGp1K5ABaa3acvMZPW06yaM8Z7qekUSrYlS/bRtO4uB921tK3WwghHkcpFYCl1/JZYCeggCbAl0qpmlrrM0bmE3nT4Qu3MJsUt+6nAnD62l0GzrEMf8vTRWZHD3hmLqx6H379Bs7thXZTwMnb6GQZ5uKdiyw5tgSNpQe3RrP8+HK01rxe5nX8nGRVuxBCiKzxtwVmpZQd8CJQBMuAv+9kdYYQ2cBvE0ha8RY9A4O4Ym3Dt7W+oZxfOaNTiWzu5r1k5u06w7TNJzhw7iaONmbalgmkY7n8FPXPZ3Q8IYTICT4BvtVaj3h0p1KqL/AZ0NmIUCJvG7o88X+G/N1NTmXo8sS8XWCGXN+XeeyesaTptD/sM2Fi1YlVxCXF0SG8A92juuNq52pIPiGEEHnHk1YwTwGSgQ1AQ6Ao0C+zQwkh/oLWEPcZhzcNp2dQMA9sHPiu7jiKeRYzOpnIxvaevs60LSeZv+s0dx6kEumfj09bRtGshD9Otk97IYsQQgiggta6y593aq2/VkolGpBHCM5cu/uP9udJxduCVyjEPmPpy9x0BJToaHSq/2z3hd0kpyX/YV8aaRR2KUwxz2JMTZjKnENzeD7qeTpFdMLeyt6gpEIIIXK7J1UWimqtowCUUt8Bv2V+JCHEY6WlwuLXiI+fRq+gIGzt3JhcdzxF3IoYnUxkQ3cfpLJwzxmmbTnJ7qRr2FmbaFrcn04V8hMd6IJSyuiIQgiRE/1dxe5OlqUQ4hH+rvacfkwx2WxSbDt+hTIF3A1IlQ35Rf9/X+Z5veDMLsvq5hzcl3lWs1m/fx0XF0eNGjX+cH/nyM58veNrRu4Yyc/7f6ZXiV60KNICK5MsMBBCCJGxnvSb5fePQ7XWKVKQEMIgyfdgTg+2HFtO38BA3B19mVBvAoHOgUYnE9nMofM3mbblJLN3nOLmvRRCvJ14v2lRWpUMxMUh576BEkKIbMJFKdXqMfsVIL2GhCEG1A9j4Jx47ian/r7PxqywtzHTZuyvtCsTyFsNI3B3tDEwZTbxsC/zyvdg82g4vxfaTgEnL6OTZYoQtxBG1R7F9vPb+Wr7Vwz+dTBT9k3hlVKvUCu4liw4EEIIkWGeVGCOVkrdSP9aAfbptxWgtdbyQlqIzHbvBsR2ZM2FbQzw9yfYpQDj647HyyF3vhAW/9z9lFSW7T3HtC0n+e3YFWzMJhpG+dKpfH7KFnCTNw9CCJFx1gFN/+K+9VkZRIiHHvZZHro8kdPX7hLgas+A+mHULerD16sP8d3GY6xIOM9bDcJpVyYIkymPvy4wW0GDT8G/RHpf5uoQ8yMElDI6WaYp7VOaHxr+wJqkNYzcMZJX4l6huFdx+pfuT2mf3NGPWgghhLH+tsCstTZnVRAhxGPcugA/tmbh7WMM8vEm0iOSMXXG4GLrYnQykQ2cuHybn347ycxtp7hy+wH5PRwY2DCcNqUD8XCyNTqeEELkOlrrrkZnEOJxWpQMoEXJgP9pkzCwUQStSgUyaN5e3poTz4xtSXzcIkqG+wIUbweeoTD9GZjUAJqOhBIdjE6VaZRS1A6uTfXA6sw/PJ8xu8bQZVkXqgdWp1+pfoS4hRgdUQghRA4mzZeEyK6uHIMfWjJNX+dzTzfK+5Xj65pf42DtYHQyYaDk1DRW77/AtC0n2HDoEmaTom6ED50qBFO5sKesShJCiCyglLIFWgMFeOT1tNb6Q6MyCfFXwnydmf5CBWbvOM2nS/bTZNQGulQqyKt1Q3C2y+Pts/xLWPoyz+wC816EMztzfF/mJ7EyWdE6tDWNCjVi2v5pTIqfROsFrWlWuBm9S/TGz8nP6IhCCCFyoEwtMCulGgAjATMwUWv9+Z/utwWmAqWBy0CM1vp4+n3FgXFY+tmlAWW11vcyM68Q2ca5ePSPrRlvB984O1ErqBZDqg/B1iyrUvOqM9fuEvvbSaZvS+L8jfv4udjRv24oMWWD8MlnZ3Q8IYTIa+YD14HtwH2DswjxREop2pQOpE6EN0OWJ/L9pmMsjj/DoCZFaRzll7fbaTl6wrPzHunLvA/aTs61fZkfsreyp3tUd9qEtGFi/ER+OvATS48tpWNER7pHdZcrJoUQQvwjmVZgVkqZgdFAXeAUsFUptUBrnfDIYd2Aq1rrIkqp9sAXQIxSygr4EXhWa71bKeXBIwMHhcjVjm9E/9yBYW75mGpvolnhZgyuNFimPedBqWma9QcvMm3LCdYcuIAGaoR68UmL/NQI88LKbDI6ohBC5FWBWusGRocQ4p9ydbDh05ZRtC0dyLvz9tLnp51MD0licLNICnk5GR3POA/7MvtFw8K+ML4GtP8R/EsanSzTudq58nrZ1+kY0ZHRu0YzZd8UZh+cTbeobnSK6ISdlSxkEEII8WSZWZ0oBxzWWh/VWj8AYoHmfzqmOTAl/etZQG1l+fi8HrBHa70bQGt9WWudihC53f6FpP7Qivc9PZhqb6JTRCc+qvyRFJfzmAs37zF67WGqDVlL18lb2ZV0nV41CrN+QE2+71qOOkV9pLgshBDG2qSUijI6hBD/VslgNxb0qcLgZpHsOnmNBiM2MHxFIveS8/hbrugYeH45KAXf1YddPxudKMv4O/nzSZVPmNVsFqV8SjFixwgaz23M7IOzSUlLMTqeEEKIbC4zq1YBQNIjt08B5f/qGK11ilLqOuABhAJaKbUc8AJitdZD/vwNlFI9gZ4APj4+xMXFZfSfQWQDt27dyhPn1u/MCgoc/JbX/YOJs0mloUtDyt8uz/p1eXcofV459wBaa/ZfSWNtUjI7zqeSqqGoh4kWJWwp6W3GynSOI3vOccTooFkoL51/8Udy7vOuHHTuqwBdlFLHsLTIUIDWWhc3NpYQT89sUnSuVICGxXz5ZMl+vl5zmHm7zjC4eSQ1w7yNjmecP/dlPrsL6n2cq/syPyrULZRvan/DtnPb+GrHV3zw6wdMSZhCv1L9qBVUK2+3UxFCCPGXsuuySCssL9zLAneA1Uqp7Vrr1Y8epLUeD4wHKFOmjH50YrLIPf48DTvX0Ro2fMmdQ2PoWzCMTdzhzbJv8kzRZ4xOZrhcf+6Bq7cfMHvHKX7acpKjl+7h6mDN81WC6VAuOG9fqkreOP/i8eTc51056Nw3NDqAEBnFO58dI9uXpF2ZIAbN30vX77fSsJgvg5oUxd/V3uh4xvi9L/Mg2DwGzu3NE32ZH1XGtww/NvyRNSfXMGLHCF5Z+wrRXtH0L92fUj6ljI4nhBAim8nMAvNpIOiR24Hp+x53zKn0vssuWIb9nQLWa60vASillgClgNUIkZukpcHygVzfOo4+hSLYo+/wUaWPaFGkhdHJRCbSWrPj5FWmbT7JovizPEhJo0x+N16uXYSGxfywszYbHVEIIcTf0Fqf+DePe4oB2P2B7kAKcBF4/uH3UkqlAvHph57UWjdL318QSys6DyxDB59Nb08nxD9SuYgnS/tVZcL6o4xac5h1By/ySp0QulYuiHVebM1ltoIGn6X3Ze6Xp/oyP6SUonb+2lQPqs68w/MYs2sMnZd1pkZgDfqV6kcRtyJGRxRCCJFNZOYrha1AiFKqoFLKBmgPLPjTMQuAzulftwHWaK01sByIUko5pBeeqwMJCJGbpDyAOT24tHU8zxeOZB/3+bL6l1JczsVu3Evmh1+P03DkBlp/+ysrEs7TvmwQy16pyqxelWhZMlCKy0IIkUs9MgC7IVAU6KCUKvqnw3YCZdJbbcwCHm0Rd1drXSJ9a/bI/i+Ar7TWRYCrWIZoC/Gv2FqZ6VMrhFX9q1OhkAefLjlAk683svX4FaOjGSe6/f/3ZZ7UAHbHGp0oy1mZrGgT2obFrRbTr1Q/tp3fRuuFrRn0yyDO3T5ndDwhhBDZQKYVmLXWKUAfLMXi/cAMrfU+pdSHSqmHL4q/AzyUUoeB/sBb6Y+9CgzHUqTeBezQWi/OrKxCZLn7t+DnGM4cmEvnIkVJIpnRtUdTJ38do5OJTBB/6jpvzd5D+U9WM2j+PqzMis9bRbHl7dp82LwY4b75jI4ohBAi8z1xALbWeq3W+k76zc1YrgD8S+nDsWthKUaDZXh2i4wMLfKmIHcHvutchnHPlubmvWTajv2VATN3c/nWfaOjGeNhX+bAsjD3BVj6FqQmG50qy9lb2dM9qjtLWy3lmYhnWHx0MY3nNGb4tuFcv3/d6HhCCCEMlKk9mLXWS4Alf9r33iNf3wPa/sVjfwR+zMx8Qhji9mX4qS1HL8bTo2AYd5VmQt0JRHtFG51MZKA7D1JYuPsM07acZM+p69hbm2kW7U+nCsEUD3Q1Op4QQois9zQDsB/VDVj6yG07pdQ2LO0zPtdaz8PSFuNa+sKOh88Z8Lgnk+HYeUNGD8q0Bd4va2L+EWvm7DjFkj2naBtqQ7VAK0x5cNibCn6FwsmuBG75lquJG0go+gbJNi5GxwKyfkhqWcpS2K8wi68tZvK+ycTuj6VevnpUc66Gjckmy3KIHDUgV2QCOf95V3Y799l1yJ8QudO1JPihJfvunKVX/sKYrGz5vu44wtzDjE4mMkjiuZv8tOUEc3ac5ub9FEJ9nBjcLJIWJQNwsc8b08eFEEL8N0qpZ4AyWNrEPZRfa31aKVUIWKOUigeeesmgDMfOGzJrUGb9OnDo/E3enbeXyfuusOuGAx+3KEaxgOxRXM1SterA7ljcFvaj8r53IOZHywpngxk1JLUVrUi8ksjXO79m/qn5/PrgV/qU6EPTwk2xMkm5ISvkoAG5IhPI+c+7stu5z4PTGoQwyIX98F09tj24TLcAf+xtXZjacKoUl3OBe8mpzNt5mrZjN1F/xHp+/i2JOkV9mPliRZa/Uo3OlQpIcVkIIcTTDMBGKVUHeAdoprX+vR+B1vp0+n+PAnFASSzDsV3TZ5b85XMKkRFCfJyJ7VmB4e2iSbpyh2bfbOSDBfu4eS/vtYqw9GVeBlrDpPqwe7rRiQwV5h7G6NqjmVR/Er4Ovry36T3aLGjDmpNrsIxYEkIIkdtJgVmIrHByC0xqwHprzYvebvg4+TO14VSC8wUbnUz8B8cu3ebTJfup+NlqXpm+i4s37/NOowg2v12br2JKULaAOyoPXj4qhBDisZ44AFspVRIYh6W4fOGR/W5KKdv0rz2BykBC+nDstViGZYNlePb8TP+TiDxLKUWrUoGsea0GHcsHM+XX49T+ch0Ldp/Je4VE/5KWvswBZWBuT1g2EFJTnviw3Kysb1l+bPQjX9X4ilSdSr+1/Xhu6XPsvLDT6GhCCCEymVyzIkRmO7gcZnRmqYcvbztCqFsIY+uMxc3Ozehk4l9ITk1jVcJ5pm05ycbDl7AyKepF+tCpfH4qFvLAZJKCshBCiP+ltU5RSj0cgG0GJj0cgA1s01ovAIYCTsDM9A8oT2qtmwERwDilVBqWBSKfa60T0p/6TSBWKfUxsBPLEG0hMpWLgzUft4iibekg3p23l74/72TG1iQ+bB5JIS8no+NlHScveG4erHgXNo+Bc/HQdjI4ehqdzDBKKerkr0ONoBrMPTyXb3d9y3NLn6NGUA1eKfUKhV0LGx1RCCFEJpACsxCZadfPML83MwJC+Nj6LqW9SzOq1iicbPLQC+9c4tTVO8T+lsT0bUlcvHmfAFd7Xq8XSrsyQXjnszM6nhBCiBzgKQZg1/mLx20Cov7ivqNAuQyMKcRTiw5yZV7vykzbcoKhyxJpMGIDL1QvRO+aRbCzNhsdL2uYraHhF+BXAhb2g/E1sk1fZiNZmaxoG9qWxgUbM23/NCbtnUSrBa1oXrg5L5V4CV9HX6MjCiGEyEBSYBYis/zyNawcxMSC0YzkKtUDqzOs+jDsrKQYmVOkpmniEi8wbctJ1iZarlSuFeZNpwrBVA/1xiyrlYUQQgiRx5lNiucqFqBBMV8+XbyfUWsOM2/XaT5sVoya4d5Gx8s6JTqAdzjEPmPpy9z0a4iOMTqV4RysHehRvAdtQtswIX4CsQdiWXJsCR0jOtKtWDdcbPPgoEghhMiFpMAsREbTGla+h970NSNCyjIp5TyNCjbi4yofY22SQW85wYUb95i+NYnYrUmcvnYXb2db+tQsQkzZIALdHIyOJ4QQQgiR7Xg72zGifUnalQ1i0Ly9dJ28lfqRPrzfNBJ/V3uj42WNh32ZZ3ax9GU+uxvqfghmedvtZufGG2XfoFNEJ0bvHM3kvZOZdXAWPaJ60CG8gyzCEUKIHE5+0wmRkVJTYGFfUndN45OISsy8d4qYsBjeLv82JiUzNbOztDTNpiOXmbblBCsTzpOSpqka4smgJhHUjvDB2iznTwghhBDiSSoV9mRpv2pM2HCUUWsOUWf4OvrVDuH5KgXzxuuph32Zl78Dm0fDuT15vi/zowKcAvi06qd0juzMyB0jGb59ONP2T6N3id40K9wMsymPtFYRQohcRgrMQmSUB3dgVleSDy7j7ciqLLtzgh5RPXi55MukD+oR2dCV2w+YtT2Jn7ac5PjlO7g5WNOtSkE6lAumgKej0fGEEEIIIXIcGysTvWsWoVm0P4MX7uOzpQeYveMUHzUvRvlCHkbHy3xma2g0xNKHeeEr0pf5McLcwxhTZwxbz23lq+1f8d6m95iaMJW+JftSI6iGvH8SQogcRgrMQmSEu1fhp/bcPfUb/aOqsfHWcV4r/RpdinUxOpl4DK01205cZdrmEyyJP8eD1DTKFXDn1bqh1I/0zTtDaYQQQgghMlGQuwMTO5dlZcJ5Pliwj5jxm2ldKpCBjcLxdLI1Ol7mK9ERvMJhenpf5majoHg7o1NlK2V9yzKt0TRWnVzF1zu+pu/avpT0LsmrpV+lpHdJo+MJIYR4SlJgFuK/unEGfmjFzatH6VOsMjtvneCDih/QOrS10cnEn1y/m8y8naeZtuUEB8/fwtnOio7lg+lYPphQH2ej4wkhhBBC5Ep1i/pQuYgHo9YcZsL6o6zaf543GoTRoWwwptw+NDmgFPRcBzM7w5wecGaX9GX+E6UUdfPXpUZQDeYemsu3u7/luaXPUTOoJv1K9aOwa2GjIwohhHgC+a0mxH9x6RD80JIr96/zYnhpDt05zdDqQ6lfoL7RycQjdiddY9qWEyzYfYZ7yWlEB7owpHVxmkT74WAjPwaFEEIIITKbg40VbzYIp1XJAAbN38s7c/cyY9spPmlRjGIBLkbHy1xOXvDc/P/vy3w+HtpMBsc80C7kH7A2WdMurB1NCjXhx/0/MmnvJFotaEWLIi3oFd0LX0dfoyMKIYT4C1JZEeLfOr0dprXlnMlEj8IRnLt3iVG1RlEloIrRyQRw+34KC3afYdqWE+w9fQMHGzMtSwbQsVx+ogJz+ZsYIYQQQohsKsTHmZ97VGDertN8sng/zb7ZyHMVC9C/Xij57KyNjpd5HvZl9ouGRa9a+jK3/9FyW/yBg7UDPYv3pG1oW8bvGU9sYiyLjy6mU0Qnni/2PC628lpeCCGyGykwC/FvHFkDsc9wwtmDHr5e3Ey+ybi64yjlU8roZHnegXM3mLb5JHN3nubW/RTCfZ35qHkkzUsG5O43LUIIIYQQOYRSipYlA6kV7sOw5YlM+fU4i+PP8m7jCJpF++fuAW8lO4F3OEx/Fr572Je5rdGpsiU3OzfeLPcmnSI6MXrXaL7f+z2zDs6iR1QPOkR0wNacB/p4CyFEDmEyOoAQOU78LJjWjkSPYJ7zduW+TmVS/UlSXDbQveRU5uw4RetvN9FgxAamb0uiXqQPs3tVZGm/qjxbsYAUl4UQQgghshkXe2s+alGM+b0r4+diR7/YXXSauIXDF24ZHS1zBZSGnnGW/sxzultaZ6SmGJ0q2wp0DuSzqp8xo+kMoryi+HL7lzSZ24R5h+eRmpZqdDwhhBBIgVmIf2bLOJjdnZ1B0XTNp7CxsmNyg8lEeEQYnSxPOnrxFh8vSqDCZ6vpP2M3V28/4N3GEWwZWJvh7UpQOr977l4BI4QQQgiRCxQPdGXuS5X5qEUx4k9fp+HI9QxdfoC7D3Jx8dDJ29KXuVxP+PUb+LEV3L5sdKpsLdw9nLF1xvJdve/wtPNk0C+DaLOwDeuS1qG1NjqeEELkadIiQ4inoTWs/QTWD+WX0Oq8knYGX3s/xtcdj5+Tn9Hp8pQHKWmsTDjPtC0n2HTkMlYmRf1ivnQqH0zFQh5SUBZCCCGEyIHMJsWzFfLTINKXz5bsZ/TaI8zfdYbBzSKpHeFjdLzMYbaGRkPBr8QjfZmngV9xo5Nla+X8yvFT459YeWIlX+/8mj5r+lDKuxSvln6VEt4ljI4nhBB5khSYhXiStFRY3B+2T2ZFZAPevHeQIq5FGFtnLB72Mvk5M8zbeZqhyxM5fe0uAZvXMKB+GKXzuxG79STTt57i0q37BLrZM6B+GG3LBOLtbGd0ZCGEEEIIkQG8nG0ZHlOCtmWCGDR/L92mbKNeUR/ebxZJgKu90fEyx8O+zLHPwHf1oPk3ENXG6FTZmlKKegXqUTO4JnMPzWXMrjE8u/RZagXVol+pfhRyLWR0RCGEyFOkwCzE30m+Z+mLtn8hc0q2ZPD1nZTwKsGo2qPIZ5PP6HS50rydpxk4J567yZZLIk9fu0v/GbtI02BSUCvch04VgqkW4oXZJKuVhRBCCCFyo4qFPVjStyrfbTzGyNUHqfPlOvrWDqFblYLYWOXCTo8BpeGFdTCjM8zuBmd2Qp3BYJa37H/H2mRNu7B2NCnUhB8SfuD7fd/TckFLWhZpSa/oXvg45tLV70IIkc3kwt/MQmSQe9dhWhvYv5Ap5drz/rXtVPSvyNi6Y6W4nImGLk/8vbj8UJoGZzsrNr5Zi4mdy1AzzFuKy0IIIYQQuZyNlYleNQqzqn91qoR48sWyAzT+egObj+bSXsVO3tB5wR/7Mt+5YnSqHMHB2oEXol9gSasldAzvyPwj82k8tzEjto/gxoMbRscTQohcTwrMQjzOzfMwuTH65K98XaEjwy5uon6B+oyqOQp7q1x6aV42ceba3cfuv3UvBf/celmkEEIIIYT4S4FuDkx4rgwTnyvDnQeptB+/mf4zdnHp1n2jo2W8h32Zm4+Gk7/C+Opwdo/RqXIMdzt33iz3JgtbLKRO/jp8t/c7Gs5uyJR9U7ifmgv/vgghRDYhBWYh/uzKUZhUj7TLR/i0fDsmnN9I65DWfFH1C6zN1kany7WSU9P4Nu4IfzX/WYrLQgghhBB5W52iPqzqX52XahRm4e4z1BoWxw+bT5Ca9levIHOwks9A12WQmmLpyxw/y+hEOUqgcyCfV/2cGU1mEOUZxbBtw2gytwnzD88nNS31yU8ghBDiH5ECsxCPOrsHvqtP8r0bvFOmKbHnNtK1WFfer/g+ZpPZ6HS51o6TV2k6aiNfLDtAVEA+7P7UV8/e2syA+mEGpRNCCCGEENmFvY2ZNxqEs7RfVSL9XRg0by+txvxC/KnrRkfLeIHpfZn9S1r6Mq9411JwFk8twiOCsXXHMrHeRDzsPHj3l3dps7AN60+tR+tc+MGEEEIYRArMQjx0bANMbsx9sw39i9dg0blf6VeqH/1L90cp6febGW7cS2bQvL20/nYT1+8mM/7Z0ix8uSqfty7++5TwAFd7PmsVRYuSAQanFUIIIYQQ2UURb2d+6lGeke1LcPraPZqN3sh78/dy/W6y0dEylpM3PDcfyvaATaNgWmvpy/wvlPcrz8+Nf2ZY9WE8SH1A79W96bq8K7sv7jY6mhBC5AoyklYIgIQFMLs7t93y07dgGFsvbOfd8u8SEx5jdLJcSWvNsr3neH/BPi7duk+XSgV4rV4YTraWH0ktSgbQomQAcXFx1KhRw9iwQgghhBAiW1JK0bxEADXCvBm+IpEfNp9gSfw53m0cQfMS/rlnkYiVDTQeBv4lYNGrlr7M7X8C3yijk+UoSinqF6hPreBazDk4h293f8szS56hdnBt+pbqSyGXQkZHFEKIHEtWMAuxfTLM7Mw1v2J0Dy7A9st7+azqZ1JcziSnr92lx9Rt9Jq2A08nW+b1rsz7TSN/Ly4LIYQQQgjxT7jYWzO4eTHm965CgKsdr0zfRccJWzh84abR0TLWo32ZJ9aVvsz/krXJmpjwGJa0WkLvEr359cyvtJrfig82fcCFOxeMjieEEDmSVHRE3qU1rB8Gaz/mfOEavOCkOXXjGCNrjqR6UHWj0+U6KalpTN50nOErD6I1vNMogq6VC2Blls+5hBBCCCHEfxcV6MKclyrz828nGbLsAA1HbqBH1UK8XCsEe5tcMk8lsDT0jIOZnS19mWd3A6AGQFz6MY7eMOCQIfFyEgdrB16MfpG2oW2ZED+B6YnTWXx0Mc8UfYauxbqSzyaf0RGFECLHkMqOyJvS0mDpG7D2Y5KKtaCzw33O3b3At3W+leJyJog/dZ0WY37h48X7KV/QnRWvVqNHtUJSXBZCCCGEEBnKbFI8UyE/q1+rQdNof8bEHaHO8HWsSjhvdLSM4+wDzy346/tvyyrcf8LD3oO3yr3FghYLqBVci4nxE2k0pxFT9k3hfup9o+MJIUSOINUdkfekPIA53eG38Rwq8xzPpZ3kdvJtvqv3HWV9yxqdLle5fT+FDxcm0Hz0Rs7fuM/ojqWY1KUsQe4ORkcTQgghhBC5mJezLcPblSC2ZwUcbMx0n7qN7lO2cerqHaOjZQwrG6MT5DpBzkF8Ue0LZjSZQaRHJMO2DaPp3KYsOLKA1LRUo+MJIUS2JgVmkbfcvwU/x8De2eyp0psuN3dgUiYmN5hMpGek0elylVUJ56k7fB2TfjlGh3LBrOpfncbF/XLPsBUhhBBCCJHtVSjkweK+VXmrYTi/HL5EneHrGBN3mAcpaUZHy1y7foYHt41OkSNFeEQwru44JtSbgJudG+9sfIe2i9qy/tR6tNZGxxNCiGxJCswi77h9CaY0haPr2FxrAN3Pr8LF1oWpDadS2LWw0elyjfM37tHrx+10n7oNJzsrZveqyCcto3CxtzY6mhBCCCGEyINsrEy8WL0wq16rTvVQL4YsS6TR1xv49chlo6NlnnkvwrBQmN8bTvxqmT8j/pEKfhX4ufHPDK0+lHsp9+i9ujfPL3+ePRf3GB1NCCGyHSkwi7zh2kmYVB8uJLC63kBeOjGHQOdApjacSoBTgNHpcoXUNM0Pvx6nzpfrWHPgAgPqh7Ho5aqUzu9udDQhhBBCCCEIcLVn3LNl+K5zGe4lp9JhwmZenb6LizdzYZ/dLkugaAvYOxe+bwCjSlsGnF8/ZXSyHMWkTDQo0ID5zefzTvl3OHr9KJ2WdKJ/XH+OXT9mdDwhhMg2rIwOIESmO58AP7aC5DvMr/sG7yVOJcozitG1R+Ni62J0ulxh/9kbDJwTz66ka1Qp4snHLYpRwNPR6FhCCCGEEEL8j9oRPlQq7MnotYcZt/4Iq/af5436YXQsnx+zKQe1c3P0fvxAP0dvKFDZsjX8AhLmw66fYM1HsOZjKFwTSnSC8CZgbZf1uXMga7M17cPb06xwM6YkTGHy3smsObmGViGt6BXdCy8HL6MjCiGEoaTALHK3k5vhp3Zg7cC0mn35/MBkKvpVZETNEThYy6C5/+rug1RGrj7ExA1HyWdvzVcx0bQoESB9loUQQgghRLZmb2Pm9fphtCwVwKB5exk0fx8ztp3ik5bFKB7oanS8pzPg0O9fxsXFUaNGjf89xtYJSnaybFeOWnoz7/4ZZncDOxco1hpKPAMBpUBewz+Rg7UDvaJ70S60HeP3jGfGwRksPLKQZ4s+S9diXXG2cTY6ohBCGEJaZIjc6+BymNoC7eDBt5W78PmBKdQJrsM3tb+R4nIGWH/wIvVGrGPsuiO0KhXA6v7VaVkyUIrLQgghhBAixyjs5cS07uUZ2b4E527co/noXxg0by/X7yYbHS3juReCWu9Avz3w3HwIqW9Z2TyxFoypAL+MhJvnjU6ZI3jYezCw/EAWNF9AzeCaTIifQKM5jZi6byoPUh8YHU8IIbKcFJhF7rTrJ/i5A2leoQwp3ZwxidNoUaQFQ6sPxcZsY3S6HO3izfv0i93Jc5N+w9psIrZnBYa0icbNUf6/CiGEEEKInEcpRfMSAax+rTqdKxZg2pYT1P4yjrk7T6Fz43A8kwkK1YDWE+D1g9BkBNjmg5XvwfAI+CkGEhZAihRKnyQoXxBDqg1hepPpRLhHMHTbUJrObcrCIwtJTUs1Op4QQmQZKTCL3OeXkTCvFykFqvBeREV+PDybZyKeYXClwViZpCvMv5WWpon97SR1hq9jafw5+tUOYWm/qlQo5GF0NCGEEEIIIf6zfHbWfNAskgV9qhDg5sCr03fTfvxmDp2/aXS0zGPnAmW6QveV0HsrVHoZzuyCGc/Cl2Gw9E04u8folNleUY+ijK83nvF1x+Ni68LbG9+m3aJ2bDi1IXd+SCGEEH8iBWaRe6SlwYp3YeV7PCjaggFBBZh/bAkvlXiJN8q+gUnJX/d/6/CFm7Qfv5m35sQT7uvMkn5VebVuKLZWZqOjCSGEEEIIkaGKBbgwt1clPm0ZxYFzN2k4cgNfLDvAnQcpRkfLXF6hUHcwvLoPOs2CgtVg2yQYVxXGVoHN38Lty0anzNYq+lcktkksQ6sN5U7yHV5a/RLdVnQj/mK80dGEECJTyXJOkTukJsOCl2H3z9wp+zyvWN/i16S1vFXuLTpFdDI6XY51LzmVMXFH+DbuMA42VgxpXZy2ZaTPshBCCCGEyN1MJkXH8sHUi/Th86UH+DbuCAt2neH9pkWpF+lrdLzMZbaCkLqW7c4V2Dsbdv4Iy96CFYMgrIFlMGCROpZjxR+YlIkGBRtQO7g2sw7NYuzusXRc0pG6+evSt2RfCrgUMDqiEEJkOPltIHK+B3dgZhc4tJzr1V7npbsH2HduH59U+YRmhZsZnS7H2nTkEu/O3cvRS7dpUcKfd5sUxdPJ1uhYQgghhBBCZBlPJ1uGtY2mXZkg3p0XT88ftlMnwpv3m0YS5J4HBoc7uEO5Hpbt/D7YOQ32TIf9C8HRG6JjLMVm73Cjk2Y71mZrOoR3oFnhZkzdN5Xv933PmpNraB3SmhejX8TLwcvoiEIIkWGkZ4DI2e5cgR9awKEVXKr/EV1v7WT/lf18WeNLKS7/S1dvP+D1mbvpOGELKWmaH7qVY0T7klJcFkIIIYQQeVa5gu4s7luVtxuFs+nIZep+tY7Raw/zICXN6GhZxycSGnwKrx2A9j9BYFlL24wx5WFCLdg6Ee5eNTpltuNo7UivEr1Y0moJ7cLaMefQHBrPbcyonaO49eCW0fGEECJDSIFZ5FzXT8P3jeDMTk43H8FzZ5dx6uYpxtQZQ+3g2kany3G01szZcYraw9cxb+dpXqpRmBWvVqNqiHyyLoQQQgghhLXZRM9qhVnVvzo1Qr0ZujyRhiPXs+nwJaOjZS2zNYQ3hg4/Qf8DUP9TSL4Hi1+DYWEw63k4vBrSUo1Omq142nvydvm3md9iPjUCazB+z3gazmnIDwk/8CD1gdHxhBDiP5ECs8iZLh6ESfXh+imOtPyG5w79wPX715lYbyIV/CoYnS7HOXbpNs98t4X+M3ZTwMOBRX2r8EaDcOysZYifEEIIIYQQj/J3tWfss6X5vktZHqSm0XHiFl6J3cmFm/eMjpb1nLygYm/o9Qv0jIPSnS3F5R9bwYgoWP0hXD5idMpsJThfMEOqDyG2SSzh7uEM2TqEZvOasfDIQtJ0HloRL4TIVaTALHKeU9stxeWUe+xrNYoue79Bo5ncYDLFvYobnS5HeZCSxjdrDlF/xHr2JF3noxbFmPViJcJ98xkdTQghhBBCiGytZrg3K1+tTt9aRVgSf47aw9YxZdNxUtO00dGynlLgXxIaDYXXD0LbyeBdFDZ+BaNKwXf1YcdUuH/T6KTZRqRHJBPqTWBc3XHks8nH2xvfpt3Cdmw8vRGt8+DfISFEjiYFZpGzHF4NU5qCXT62Nv+Sbju+wNHakSkNphDiFmJ0uhxl2/ErNBm1gWErDlI3wofVr1Xn2Qr5MZmU0dGEEEIIIYTIEeyszfSvF8bSV6oSHeTK+wv20Xz0RnYnXTM6mnGsbCGyJTwzC15NgDofwJ3LsOBlGBYKc1+EY+shTVbrAlTyr0Rsk1iGVBvC7eTb9FrVi+4rurP30l6jowkhxFOTArPIOeJnwU8x4F6IdQ0H0+u3j/B18GVqw6kE5QsyOl2Ocf1uMgPnxNNm7K/cvp/Kd53LMLpTKbzz2RkdTQghhBBCiBypsJcTP3Qrx6gOJblw4z4txvzCu/PiuX4n2ehoxsrnB1VehT5bodtKKN4ODiy2LBr6ugTEfQ5XTxid0nAmZaJhwYYsaLGAgeUGcvjaYTos7sBrca9x4ob8/xFCZH9WRgcQ4rGGhsDtCwDUAIhL32+yZnGtfry7eTDh7uF8W+dbXO1cDYmY02itWbTnLIMXJnDl9n26VynIq3VDcbSVHwNCCCGEEEL8V0opmkb7UyPMi+ErDzJl03GWxp/j7UYRtCoVgFJ5+EpBpSConGWr/xkcWAQ7f7QUmOM+gwJVoeQzENEMbByMTmsYa7M1HSM60rxIc6bsm8LkfZNZfXI1bULb8GL0i3jaexodUQghHktWMIvsKb24/GfTHW0ZuPkjSvmUYmL9iVJcfkpJV+7QdfJWXv55J34udizoU4V3mxSV4rIQQgghhBAZzNnOmvebRrLw5SoEezjw2szdxIzfzMHz0n8YsBSQi7eDzgvglT1Q8x24ngRzX7C00FjwMpzcDHm4D7GjtSMvlXiJJa2W0Da0LbMPzqbRnEZ8s/Mbbj24ZXQ8IYT4H1JgFjmCBia65ONjT3eqB1VnTJ0xOFo7Gh0r20tJTWP8+iPU+2o9vx27wntNijKvd2WKBbgYHU0IIYQQQohcLdLfhdkvVuKzVlEknrtJo5Eb+Gzpfu48SDE6WvbhGgzV34CXd0KXxVC0GcTPtgx1/6YMbPgSbpwxOqVhPO09eafCO8xvMZ/qgdUZt2ccjeY0Ytr+aTxIfcDFOxcZcW4El+5eMjqqECKPkwKzyNYumk108fXmE3c3Rrq70uTWbYbXGI6t2dboaNne7qRrNP3mFz5dcoDKRTxZ1b86z1cpiFmG+AkhhBBCCJElTCZFh3LBrHmtOi1LBjBu3VHqfLmO5fvOofPwCt3/YTJBgSrQYgy8ngjNR4OjN6z+EL6KhB9awd7ZkHzP6KSGCM4XzNDqQ4ltHEuoWyif//Y5zeY1Y+DGgRy9f5Sxu8caHVEIkcdJgVlka9+6urDdzpbpLs60v3GTTy5extpkbXSsbO3mvWQ+WLCPFmN+4crt+4x9phQTniuNv6u90dGEEEIIIYTIkzycbBnaNpqZL1bE2c6aF37YTrcp20i6csfoaNmPrbOlH/PzS+HlHVD1NbiYCLOehy9DYVF/OL09T7bQiPSMZEK9CYyrMw57K3u2nN2CRjP70GxO3TxldDwhRB4mBWaRbV00m5jj7ARKYdaanteuy1/YJ1i+7xx1h69nyq/Hea5Cflb2r06DYn55e6CIEEIIIYQQ2UTZAu4s6luFdxpFsPnoZeoMX8c3aw5xPyXV6GjZk0dhqPWupVfzs3MhpB7smgYTasGYirBpFNx6/Pye3EopRaWASpT0LolZmQFISUuh+bzmDN82nKSbSQYnFELkRVKvE9nP2d0AfOThzsOXWSYN41xdLJdJif9x9vpdek7dxgs/bMfVwZo5vSoxuHkx8tnJam8hhBBCCCGyE2uziR7VCrH6terUjvBm2IqDNBy5gV8OSx/dv2QyQ+Fa0HoivJYITb4CG0dY8S58GQ4/tYf9CyHlgdFJs8TFOxdZcGQBqfr/P5hI0SlM2TeFxnMa02tVL9afWk9qmnxwIYTIGlZGBxDiD1Luw9xeHMvny1pHm993J5sU89w8ebH1MjwNjJfdpKZppv56nGHLE0nVmrcahtOtSkGszfLZkRBCCCGEENmZn4s9YzqVZm3iBd6fv49OE7fQLNqfdxtH4J3Pzuh42Ze9K5R53rJdTLSsaN4dCweXgoMHRLWDkp3AN8ropJlm7J6xpOm0P+wzKzMNCzYkwDmAWQdn0Xt1bwKcAmgb2pZWIa1ws3MzKK0QIi+QKpTIXtZ9ARf20TfA/3/uStNpMrzgEfvOXKfVmF8YvDCB0gXcWflqdV6sXliKy0IIIYQQQuQgNcO8WfFqNfrWDmHZ3nPU/nIdk385Rkpq2pMfnNd5hUHdD+HVBOg40zIocOtEGFsFxlaFLePgzhWjU2a43Rd2k5yW/Id9yWnJJF5N5KUSL7G8zXKGVR+Gv5M/I3aMoM7MOry94W12X9wtwyWFEJlCVjCL7OPUdtj4FSsi63P8zv7/uTs5LZldF3Zlfa5s5s6DFEasOsR3G4/h5mDN1x1K0rS49FkWQgghhBAip7KzNtO/bigtSwbw3vy9fLAwgZnbT/Fxi2KUDJaVp09ktoLQepbtzhWIn2lZ2bz0DVj+DoQ1tAwOLFzbcmwON6vZrN+/jouLo0aNGn+439pkTf0C9alfoD5Hrh0h9kAsC48uZOHRhUS4R9A+vD0NCzbE3koGwQshMkbO/8kqcofkuzDvRS7k8+fDlNMU8yjG1EZTsTZZP/YXZl619sAF3p23l9PX7tKhXBBvNYjAxUH6LAshhBA5gVKqATASMAMTtdaf/+n+/kB3IAW4CDyvtT6hlCoBfAvkA1KBT7TW09MfMxmoDlxPf5ouWutdmf6HEUJkioKejkx9vhyL48/y4cIEWn27iQ7lgnmjfhiuDjZPfgIBDu5Q/gXLdm6vpdC8ZzrsXwBOPlA8xlJs9gozOmmWKOxamHcqvMMrpV9h8dHF/HzgZ97f9D7Dtg2jeeHmxITFUMClgNExhRA5nBSYRfaw9hPSLh1kUMl6PLh1gs+qfoa1SQqnD124eY/BCxNYvOcsRbydmPFCRcoVdDc6lhBCCCGeklLKDIwG6gKngK1KqQVa64RHDtsJlNFa31FK9QKGADHAHeA5rfUhpZQ/sF0ptVxrfS39cQO01rMQQuQKSimaFPeneqgXX608xORNx1i+9xwDG0XQulSAXLn4T/gWgwafQZ3BcGiFpdj862jY9DUElIESHaFYa0tf51zO0dqRdmHtaBvalh0XdjD9wHRiD8Ty4/4fqehXkZjwGKoHVsfKJGUiIcQ/Jz85hPFOboZN3/BzZB02XTvAoAqD5BPUdGlpmp+3nuTzpQe4n5JG/7qhvFC9ELZWZqOjCSGEEOKfKQcc1lofBVBKxQLNgd8LzFrrtY8cvxl4Jn3/wUeOOaOUugB4AdcyP7YQwijOdta817QorUsHMGjeXl6fuZsZW5P4qEUxwnydjY6Xs1jZQEQTy3brAuyZYSk2L+4Py9+G8CaWwYAFq4Mpd7/XUkpR2qc0pX1Kc+nuJWYfnM3MgzN5Ze0r+Dr6/j4U0NPe0+ioQogcRArMwlgPbsO8XhxxD+Kr+8epFliNtqFtjU6VLRw8f5OBc+LZfuIqFQt58EnLYhTycjI6lhBCCCH+nQAg6ZHbp4Dyf3N8N2Dpn3cqpcoBNsCRR3Z/opR6D1gNvKW1vv+Yx/UEegL4+PgQFxf3T/OLHODWrVtybnOpPhGaDflsmJF4hUYj11OvgDXNC1tjZ2VZzSzn/p8qBhGf4hx4GN9zq/E+sBTrvbO4Z+vJeZ+anPOtzV0HP6NDPpX/eu7DCGOg50D23t3LhpsbGLVzFGN2jiHaIZpqztUoZFtIVs1nY/JvP+/KbudeCszCWKsGk3zlKAOjKuOQfIvBlQbn+V9e95JTGbXmEOPWHcXZzophbaPlUjghhBAiD1FKPQOUwdJb+dH9fsAPQGetdVr67oHAOSxF5/HAm8CHf35OrfX49PspU6aMlvkWuZPMLsndagEv337AF0sPMH1bEruvmKlfzJcV+85x+poiwDWNAfXDaFEywOioOUhNoAck34PEJdjtmkb+I7PJf3ImBFeyrGou2hxss++K8Yz6d1+b2vSjH8euH2NG4gzmH57PjvM7CHELoX1Ye5oUaoKDtcN/DywylPzcz7uy27k3GR1A5GHH1sNv4xgTWZP9t5L4oNIHef4ynF8OX6LBiPWMXnuE5iUCWP1aDdqUDpTishBCCJHznQaCHrkdmL7vD5RSdYB3gGaPrkRWSuUDFgPvaK03P9yvtT6rLe4D32NpxSGEyKXcHW34ok1xZr1YEa013/9ynNPX7gFw+tpdBs6JZ97O//nRIp7E2g6KtYJnZsOr+6D2+3D7AszvDcPCYG4vOL4R0tKe/Fw5XEGXgrxZ7k1WtV3FBxU/wKzMfLT5I2rNrMWnWz7lyLUjT34SIUSeIyuYhTHu34T5vdnhVZBJd4/RKqQVtYJrGZ3KMJdv3eeTxfuZs/M0BT0d+al7eSoVydvFdiGEECKX2QqEKKUKYikstwc6PnqAUqokMA5ooLW+8Mh+G2AuMPXPw/yUUn5a67PK8ml0C2Bvpv4phBDZQpkC7ijT/y5CuZucypDlB2QV83+Rzx+q9ocqr0LSb7DrR9g7F3b/BG4FILojlOgArsFGJ81UDtYOtA5tTauQVuy5tIfYA7HMOjiLnw/8TDnfcsSExVAzuCbWJmujowohsgEpMAtjrHiXWzdO83ZYNP5WdrxR9g2jExlCa83M7af4dMl+bt9P4eVaRehdswh21rl7sIQQQgiR12itU5RSfYDlgBmYpLXep5T6ENimtV4ADAWcgJnpVy+d1Fo3A9oB1QAPpVSX9KfsorXeBUxTSnkBCtgFvJh1fyohhJHOpq9c/rMz1+4xdPkB2pUJIr+HYxanykWUguDylq3BF7B/oaXYHPcpxH0GBatBiU4Q0RRscm/rCKUU0V7RRHtFM6DsAOYemsuMxBm8tu41vO29aRPahtahrfF28DY6qhDCQFJgFlnv8GrYPpnPi1bh7L1TTKk5BUfrvPfC58jFW7wzN57NR69QJr8bn7WKIsQn+/b2EkIIIcR/o7VeAiz50773Hvm6zl887kfgx7+4L+9eAiZEHufvas/pa3f/Z7+tlYlv444weu0RKhX2IKZsEPUjfWURy39h4wDRMZbt2knY9TPsmgZze8KSfBDZEko+A4FlLYXpXMrdzp1uUd3oEtmFjac3EpsYy7e7v2XcnnHUCq5F+7D2lPUtKy0ehciDMrXArJRqAIzEskpjotb68z/dbwtMBUoDl4EYrfXxR+4PBhKAD7TWwzIzq8gid6/BgpdZ5VuE+XdP0rN4T0p4lzA6VZa6n5LK2LijjF57GDtrE5+1iiKmTBCmx1ziJoQQQgghhBCPM6B+GAPnxHM3OfX3ffbWZj5rFUX5Qu7M2naKGduT6Be7Cxd7a1qWDKBdmSCK+uczMHUu4BoMNd6EagPgxC+w6yeInwk7poBHCJToCNEdIJ+f0UkzjdlkpnpQdaoHVSfpRhIzDs5g7uG5rDyxkkIuhYgJi6FZ4WY42TgZHVUIkUUyrcCslDIDo4G6wClgq1JqgdY64ZHDugFXtdZFlFLtgS+AmEfuHw4szayMwgDL3+binQsM9g0hMl8IL0bnras4fzt2hYFz9nDk4m2aRvszqEkE3s52RscSQgghhBBC5DAP+ywPXZ7I6Wt3CXC1Z0D9sN/3v1w7hN41i/Dr0cvEbk3ipy0nmbzpOMUDXYgpG0TTaH/y2Un/3H/NZIKCVS1boyGwb55lVfPqwbDmIyhc21JsDmtkGSKYSwXlC+K1Mq/Ru0Rvlh9fTuyBWD777TNG7BhB00JNiQmPIdQt1OiYQohMlpkrmMsBh7XWRwGUUrFAcywrkh9qDnyQ/vUs4BullNJaa6VUC+AYcDsTM4qslLgMvWsag8LLcS/lGp9V/SzPDAS4ducBny05wPRtSQS62TO5a1lqhEmPKiGEEEIIIcS/16JkAC1KBhAXF0eNGjX+536TSVG5iCeVi3hy9fYD5u06zfStSbwzdy8fLUqgcZQ/MWWDKFvATdoa/Be2zlDqWct2+YhlVfPun2FWV7Bzhai2lmKzf8lc20LDzsqO5kWa07xIc/Ze2sv0xOnMPzKfGQdnUMq7FO3D21MnuA7W5rxRAxAir8nMAnMAkPTI7VNA+b86Jn3wyXUsw0vuAW9iWf38+l99A6VUT6AngI+PD3FxcRkWXmQsq+QblN3al5898/PL/XO0dW/LiZ0nOMGJJz721q1bOfbcaq3ZfDaVnw/c51YyNCpoTfPCCs4mEHc24clPkMfl5HMv/js5/3mXnPu8S869EEJkHjdHG7pWLkiXSgXYc+o6sVuTWLj7DLN3nKKQpyMxZYNoVSoQL2dbo6PmbB6FofYgqPk2HFsHO6fBjqmwdQJ4F7UMBiweA05eRifNNMU8i1HMsxivlX6N+UfmMz1xOm+sfwN3O3dah7SmbWhb/JxybwsRIfKi7Drk7wPgK631rb/7FFVrPR4YD1CmTBn9uE9sRTYxqxtHucs3Lu5U8SvPoNqDnvoT8r/6ND67O3n5Du/Mi2fDoUtEB7nyWcso6Xf2D+XUcy8yhpz/vEvOfd4l514IITKfUoroIFeig1wZ1CSCxXvOMn1rEp8tPcDQ5YnUjvAmpmwQ1UK8sDKbjI6bc5nMULiWZbt7DfbOtqxsXvEOrHofQupZis2h9SGXrup1tXOlc2Rnni36LJvObGL6gelMjJ/Id3u/o0ZgDWLCY6jgVwGTkr9nQuR0mVlgPg0EPXI7MH3f4445pZSyAlywDPsrD7RRSg0BXIE0pdQ9rfU3mZhXZJaE+STvncXAsBLYqzQ+rPRhrr78Kjk1jYkbjjFy9UGsTCYGN4vkmQr5McsQPyGEEEIIIUQ24mBjRdsyQbQtE8ThC7eYsS2J2dtPsXzfeXzz2dG2TCDtygQR5O5gdNSczd4VynazbBcOWHo175kOiUvAwdOyorlER/AtZnTSTGFSJqoEVKFKQBVO3zrNrIOzmHNoDmuS1lAgXwHahbWjWeFmuNi6GB1VCPEvZWaBeSsQopQqiKWQ3B7o+KdjFgCdgV+BNsAarbUGqj48QCn1AXBLiss51K2LsOhVvg0KJeHBFUbUGIGXQ+69FGjHyau8PSeeA+duUj/Sh8HNiuHrknsHOgghhBBCCCFyhyLeTrzdKILX64Wx5sB5YrcmMXrtYUatOUzlIh60KxNE/Uhf7KzNRkfN2bzDod5HUPt9OLIadv4Iv42HzaPBLxpKPANRbcDB3eikmSLAKYB+pfrRK7oXK06sIPZALEO2DuHrHV/TqFAj2oe1J8IjwuiYQoh/KNMKzOk9lfsAywEzMElrvU8p9SGwTWu9APgO+EEpdRi4gqUILXILrWFxf3bpe3xndZ8WRVpQO39to1Nlihv3khm6LJEft5zAN58d458tTb1IX6NjCSGEEEIIIcQ/YmNlokExPxoU8+PMtbvM2n6K6VuT6Be7Cxd7a1qWDCCmbBARftL+7z8xW1naY4TWh9uXIX6mZWXz0gGWNhphDS3F5sK1LMfmMjZmG5oUakKTQk3Yf3k/0xOns+TYEuYcmkNxr+K0D2tPvQL1sDVLT3AhcoJM/SmltV4CLPnTvvce+foe0PYJz/FBpoQTmW/vbG4fWMjAIkXxs8vHW+XeMjpRhtNas2zvOd5fsI9Lt+7TpVIBXqsXhpNt7nsBIIQQQgghhMhb/F3t6Vs7hD41i7DpyGVit57kpy0nmbzpONGBLsSUDaZptB/Odrmzh3CWcfSACi9atnPxlsGA8TMgYT44+UJ0e0u/Zq9Qo5NmigiPCD6o9AH9y/RnweEFTE+cztsb32bo1qG0DGlJ29C2BDoHGh1TCPE3pAomMsfNc7D4Nb4IDuFM6m0mVx2No7Wj0aky1Olrd3l//l5W7b9ApH8+JnYuQ/FAV6NjCSGEEEIIIUSGMpkUVUI8qRLiydXbD5i78zTTtybx9tx4PlqUQOPifrQvG0Tp/G65et5OlvCNgoafQ90P4dByS7F50yj4ZQQElrUUmtd+ArcvAlADIC79sY7eMOCQIbEzQj6bfDxT9Bk6RXRiy7ktTD8wnSn7pvD93u+pFliNmLAYKgdUlqGAQmRDUmAWGU9rWNiP1dZpzDXdo0exHpT0Lml0qgyTkprG5E3HGb7yIFrDu40j6FKpgExYFkIIIYQQQuR6bo42PF+lIF0rF2D3qetM33qSBbvOMGv7KQp5OdK+bBCtSgXi6SStDf4TKxuIaGrZbp63rGjeOQ0WvfLXj7l9IcviZSalFBX8KlDBrwLnbp9j1sFZzDo4i3Wn1hHoFEi7sHa0LNISVztXo6MKIdJJgVlkvF0/cenICgYXLEKES0F6RfcyOlGGiT91nYFz97D39A1qhXvzYfNIAt1korIQQgghhBAib1FKUSLIlRJBrrzbuCiL488yfWsSny45wJBlidSJ8CGmXBDVQrwwm2RV83/i7AOVXoaKfeDMDphQy+hEWcbX0Zc+JfvwQvEXWH1yNbGJsQzfPpxvdn5Dg4INaB/WniivKKNjCpHnSYFZZKzrp9DL3mJQUBHuoPm86udYm3N+P67b91P4csVBJm86hoeTLaM7lqJRlK9c/iWEEEIIIYTI8xxtrWhXJoh2ZYI4fOEm07cmMXvHaZbtO4efix1tSwfStkwQQe6yOOc/UQoCSv/9MUsGQEQzyF8JTOasyZUFrM3WNCjYgAYFG3Do6iGmJ05n4ZGFLDiygEiPSGLCYmhYsCF2VnZGRxUiT5ICs8g4WsOCl5lhb8VGdY+BpQdSyLWQ0an+s1UJ53lv/l7O3rhHp/LBDKgfjot9zi+aCyGEEEIIIURGK+LtzDuNizKgfjir958ndmsSo9Ye5us1h6lSxJOYskHUi/TB1ir3FD+zlR1T4bfx4OgF4Y0txeaC1SAXLPx6KMQthHcrvMsrpV5h0dFFxB6I5b1N7zFs2zBaFmlJu7B2BOcLNjqmEHmKFJhFxtk+mWMn1zMsKJjKfuXpEN7B6ET/yfkb9/hgwT6W7j1HmI8zozqWonR+N6NjCSGEEEIIIUS2Z2NlomGUHw2j/Dh97S6ztp1ixrYkXv55J64O1rQsGUBM2SDCffMZHTV3GXAEDq+EhAWwZyZsnwx2rpZic9HmUKgGWOWO/thONk60D29PTFgM285vY3ridKbtn8aUhClU9q9MTFgM1QKrYc5FK7mFyK6kwCwyxtUTJK94l4GBBbC1tufDyh/m2PYRqWmaaVtOMGRZIsmpaQyoH0aPqoWwsZIhfkIIIYQQQgjxTwW42tOvTggv1yrCL0cuEbs1iR83n+D7X44THeRK+7JBNI32x8lWShRPxdH78QP9HL3B1gkiW1q25LtwZI2l2Lx/EeyaBrb5ILQBFG0GReqAtX3W589gSinK+palrG9ZLt65yKxDs5iVOIu+a/vi5+j3+1BAD3sPo6MKkWvJT2/x36WlwfzejHO2Zx8PGF7pc7wdvI1O9a/sP3uDgXPi2ZV0jaohnnzcohj5PRyNjiWEEEIIIYQQOZ7JpKga4kXVEC+u3H7A3J2nmb71JAPnxPPRogQaR/nRvlwQpYLdcuyCpSwx4NDvX8bFxVGjRo3HH2dtb1m5HN4YUh7AsXWQMA8OLIb4GWDtCCF1LSubQ+pZitM5nJeDF72ie9E9qjvrktYRmxjLyB0jGbNrDPUK1KN9WHuivaLl75cQGUwKzOK/2zqBXWe3MMHfj2aFm1I3f12jE/1jdx+kMnL1ISZuOIqLvTUjYkrQvIS//NIRQgghhBBCiEzg7mhDtyoFeb5yAXYlXWP61iQW7j7DzO2nKOzlSPuywbQsFYCnU+5o52A4KxtLMTmkLjQZAcc3wv70lc0J88DKDgrXthSbwxqAnYvRif8Ta5M1dfLXoU7+Ohy9fpQZiTOYf3g+i48uJtw9nJiwGBoVbISDtQyeFCIjSIFZ/DeXj3Bn1fu8HRiEn5MXA8sNNDrRP7bu4EXenRdP0pW7tCsTyMCGEbg52hgdSwghhBBCCCFyPaUUJYPdKBnsxqAmRVm85yyxW0/yyZL9DFl+gDoRPsSUDaJqiBdmkywAyhBmayhc07I1GgYnN1uKzQkLIHExmNLvj2hmWf3s4G504v+kkEsh3ir3Fn1L9mXxscXEHohl8K+D+XLblzQv0px2Ye0o5FLI6JhC5GhSYBb/XloqzHuJIe4unCaVSVU+xckm51xSc/HmfT5alMCC3Wco5OVIbM8KVCgkPZmEEEIIIYQQwgiOtla0KxtEu7JBHDp/k+lbk5iz8zRL957D38WONmWCaFs6kCB3WXWaYUxmKFDZstX/DE5vt6xo3r8ADq2Ahf2gYFXLyubwJuCUM9thAjhYO9A2tC1tQtqw++Jufj7w8++DAcv7lad9WHtqBNXAyiSlMiH+KflXI/69zWNYe3k3s3286B71PKV9Shud6KmkpWlmbEvis6UHuPsglVfqhNCrRmFsrWSyrBBCCCGEEEJkByE+zrzbpChvNAhn1f7zxG5NYtSaQ4xac4gqRTyJKRtE3aI+8j4uI5lMEFTWstX7GM7uhoT5lm3Rq7CoP+SvZCk2RzSFfP5GJ/5XlFKU8C5BCe8SvHH3DeYensuMxBm8Gvcq3g7etA1tS+uQ1ng5eBkdVYgcQwrM4t+5mMiltZ/wQaAvEe6hvBT9ktGJnsrhCzd5e85efjt+hfIF3fmkZRRFvHPOqmshhBBCCCGEyEtsrEw0ivKjUZQfp67eYdb2U8zcdoo+P+3EzcGaliUDiSkbRJivs9FRcxelwL+EZav9HlxIsLTQSJgPS9+wbIHloGgzSysNt/xGJ/5XPOw96B7Vna6RXVl/aj3TE6czetdoxu0eR+38tWkf1p7SPqVlPpMQTyAFZvHPpaag577A+55u3DaZ+bzq51ibrY1O9bfuJacyJu4I38YdxsHGiiGti9O2TKD8khBCCCGEEEKIHCLQzYFX6oTycq0QNh6+xIytSfyw+TiTfjlGiSBX2pcNokm0P062UurIUEqBT6RlqzkQLh6E/fMtBecV71o2vxKWlc1Fm4NHYaMT/2Nmk5mawTWpGVyTEzdOMCNxBvMOz2P58eUUcS1CTFgMTQs3xdHa0eioQmRL8lNX/HO/jGDmzUOs93TnrTL9KeSavZvhbzpyiXfn7uXopdu0KOHPu02KyiRiIYQQQgghhMihzCZF9VAvqod6cfnWfebuPM30rUm8NSeeDxcl0KS4HzFlgykV7CqLijKDVyh4DYBqA+DKsf8fELh6sGXzjkwvNjcDr3BLgToHyZ8vPwPKDqBPyT4sO7aM2MRYPtnyCV9t/4qmhZsSExZDiFuI0TGFyFakwCz+mXN7Ob5xKMMCfKnkX4EO4R2MTvSXrt5+wCdL9jNr+ynyezjwQ7dyVA2RHkpCCCGEEEIIkVt4ONnSvWohulUpyI6T15ixNYmFe84wY9sping70b5sEC1LBuAhi4wyh3tBqNzPsl0/BfsXWtpoxH0GcZ+CZ6ilhUbR5uAblaOKzfZW9rQMaUmLIi3Ye2kvsYmxzD00l+mJ0yntU5r24e2pHVQ721/RLURWkAKzeHopD0ie9wJve3tiY+PER5U/wqRMRqf6H1pr5u48zceL93PjbjIv1ShM39oh2FnL8AchhBBCCCGEyI2UUpTO70bp/G4MalqUxXvOELs1iY8X7+eLZQeoW9SHmLLBVCniidmUc4qcOYpLIFToZdlunoMDiyzF5o3DYcMwcCuQPiCwOQSUyjHFZqUUUV5RRHlF8XqZ15l3eB7TE6czYN0APO09aR3SmjahbfB19DU6qhCGkQKzeHobvmTCvSTi3Vz4suL7eDt4G53ofxy7dJt358Xzy+HLlAp25dNWUYT75jM6lhBCCCGEEEKILOJka0VM2WBiygZz8PxNpm9NYs6OUyyJP4e/ix1tywTRtkwggW4ORkfNvZx9oWx3y3b7EhxYbGml8eto+GUk5Av8/wGBQeXBlP0Wrz2Om50bXYt1pXNkZ345/QuxibGM3zOeifETqRlUk5jwGMr7lpfWLCLPkQKzeDpndrFny0jG+3nRrHAz6hWoZ3SiP3iQksb49Uf4es1hbM0mPm5RjI7lgjHJJ9NCCCGEEEIIkWeF+jgzqElR3mgQxqqEC8RuPcnXaw7x9ZpDVCniSfuywdQp6o2tlVzxmmkcPaF0Z8t29yokLrOsbN76HWweA06+ENHEsro5uBKYs3+pyqRMVA2sStXAqpy6eYqZB2cy59AcVp1cRYF8BWgf3p6mhZuSz0YWvIm8Ifv/qxXGS7nPnbkvMNDbEx8HX94q95bRif5g2/ErvD03noPnb9E4yo/3mxbFO5+d0bGEEEIIIYQQQmQTtlZmGhf3o3FxP05dvcPMbaeYuS2J3j/twM3BmlalAokpG0Soj7PRUXM3ezco0cGy3b8JB5dbis07p8HWieDgAeGNLcXmgtUhB/Q3DnQO5NXSr/JSiZdYcXwFsYmxfP7b54zcMZJGBRvRIbwDYe5hRscUIlNJgVk8WdxnDE07T5I5H99X+xxnm+zxC/f63WQ+X3qAn387SYCrPZO6lKFWuI/RsYQQQgghhBBCZGOBbg68WjeUvrVD2HDoIjO2JTH11+N8t/EYJYNdaV82iCbF/XG0lZJJprJ1hqg2lu3BbTi8ChIWwN45sGMq2LlAWGNLK41CNcE6ey8kszXb0rRwU5oWbkrC5QSmJ05n8dHFzD40mxJeJYgJj6Fe/nrYmG2MjipEhpOfluLvJW0lbsc4Zvl48nyxrpT2KW10IrTWLNpzlsELE7hy+z49qhbklTqh8stfCCGEEEIIIcRTM5sUNcK8qRHmzeVb95m78zSxW5N4c3Y8Hy5MoElxf2LKBVEyyFV66mY2G0fLquWizSH5Hhxda1nZnLgYdv8ENs4QWt9yf5E6YJO9+2cX9SjK4EqD6V+6PwuOLGB64nQGbhjI0K1DaRXSirahbfF38jc6phAZRipy4q8l3+Xy/Bd538uDcNcQ+pToY3Qikq7cYdD8vcQlXiQqwIXJXctSLMDF6FhCCCGEEEIIIXIwDydbulctRLcqBdlx8irTtyaxYPcZpm9LItTHiXZlgmhVKhB3R1l9mums7SCsoWVLeQDH11uKzQcWw95ZYO0AIXUtAwJD61tWQmdTLrYuPFv0WTpFdGLz2c1MPzCdSXsnMWnvJKoFVqN9WHsq+lfEpHLGkEMh/ooUmMVf0qs+5APTNW6Z8/FdtSFYZ3Hvo3k7TzN0eSKnr93Ff/NqyuR3Y2XCBUwK3mtSlM6VCmCWIX5CCCGEEEIIITKIUorS+d0pnd+d95pGsmj3GWK3JvHx4v18sewA9Yr6ElM2iCpFPGWofFawsrGsWC5SBxp/BSd+gf0LYP9CS9HZbAtFaluKzWENwd7V6MSPZVImKvlXopJ/Jc7eOsvMgzOZfWg2cUlxBDkHERMWQ4siLXCxlQV0ImeSArN4vBObmJ0wlThPd94s8xpF3Ipk6beft/M0A+fEczc5FYAz1+6x4NpZIv3zMeG5Mvi72mdpHiGEEEIIIYQQeYuTrRXtywXTvlwwieduMn1rEnN2nmJx/FkCXO1pWyaQtmWCCJD3p1nDbAWFqlu2hkMg6TdLkXn/AkhcAiZry31Fm1t6Nzt6GJ34sfyc/Ohbqi+9onux8sRKpidOZ9i2YYzaOYqGBRvSPqw9kZ6RRscU4h/5v/buPD6q6v7/+OvMZCWErCSQSSBhCwQEIgk7AZWt1n1fvnWpy0+72Q23olXb+m2/tFVstdQNrVrFFXFFQEPQiibsEPY9CSRhCQgEsp3fH3dIwmYlJJkk8376uA9mzr1z7xkPGa/vnPkcBcxyooqDbJt1J/8XE83Q+Eyu63Nds3dhyuy1teFyfWWHKhQui4iIiIiISLNK7RTOgxemcc/3UpmTX8yM3O08Pnc9U+etZ1TPjlyTmcTYPvEEBajUQbNwuaHrMGeb8CgULa4Lm2f9FMzPIXmEEzb3vhDC433d4xMEugM5v9v5nN/tfNbuWcvra1/nvU3vMXPDTM6KPYurU69mQvIEQgJa9uKGIqCAWU6i6pMHuC+onMDAKH4/6n99UguoqKz8FO2Hm7knIiIiIiIiIo7gADcX9E/ggv4JbN9ziDcWFfBG3nZ+9MpiosOCuCzdw9WZSfSMb7l1gdsclwsSM5xt3COwcznkz3IC5w9+BR/8GroMg7SLoM+FEJHo6x6fIDU6lQeGPcDPB/2c9za+x4y1M5j8xWSm5E3hsh6XcWXqlSSFJ/m6myKnpIBZjrUpm2c2vMHyqEimjHiY+DDf/JYvPCSA/YerTmjX7GURERERERFpCZKi2/HLcb2467yeLFhfyozc7bzwny08+/lmBnWN4uqMJL7fvzNhwYpemo0x0HmAs507GUrXOEFz/iz4+F5n82Q4M5vTLoKoZF/3+BjhQeFc1+c6ru19Lbk7c3lt7Wv8K/9fvLDqBUZ4RnBt72sZkTACt8tN6aFSHt/5OP3K+xEbGuvrrouf06ecv5vSEw6W1D5dERTEPxPiueBwFROTJ/qkS4/PXcf+w1W4jaHa2tr20EA3kyak+qRPIiIiIiIiIifjdhnGpMYxJjWOXQeO8PbiAmbkbufut5bz8HuruGhgAldlJDEwKRJjtDBgszEG4vo425h7YdcGWO0Nm+c84Gyd+nvD5oshtqeve1zLGMPgzoMZ3HkwxQeLeWv9W7y57k1+PO/HeNp7uLLXlWzet5lNRzYxbdk0Jg+d7Osui59TwOzvvOFyqdvFr+JiKXW7iauu5v6dO5q9K9ZaHpuzjic+3cDlZycyonsMf5mzjsKycjyRoUyakMol6Z5m75eIiIiIiIjIdxHbPpjbs7pz26huLNq6lxm525m5pIhXv95Oanw4V2UmcWm6h+iwIF931f/E9oBRv3K2vVtg9XvO7OZPf+dscWnQ5yInbI7r4wTULUB8WDw/Gvgjbut/G59t+4wZa2fw+OLHa/e/te4tLup+Ef079vddJ8XvKWAWAKZFRrAkOBiA53eWEF5v5nBzsNbyf7PX8o/sjVydkcT/XnYWLpfhskGJZGdnM2bMmGbtj4iIiIiIiEhDGWPISI4mIzmaBy9M471lO5iRt53fvZ/Pnz5aw/i+8VydmcSI7rG4XC0jyPQrUckw/KfOtq/QCZtXz4L5f4L5f4SYHk7Q3Ocip9xGCwibA12BjE8ez/jk8UyaP4lPtnxCDTVU2Squ//B6ukd0Jysxi1GJoxgYN5BAV6Cvuyx+RAGzUOp28U54ezAGt7WkVFY26/WttfzvR2t4OmcT1w3pwu8v7qf/wIqIiIiIiEibEB4SyHVDunDdkC6s2bmfGbnbeWdJIe8v30FiVChXDkriyoxErTnkKxEeGHqHs31TDGved8Lmzx+HBX+ByK7eBQIvBs8gZ1FBHyo9VMpn2z+jhpratgATQERwBC+tfonpq6YTHhTOiIQRZCVmMdIzkqiQKB/2WPyBAmZhamQkRyNll3VmM0/evbdZrm2t5ZH385n+xRZuGNaVhy/qq5pUIiIiIiIi0ib17tSB317Yl3sm9uaT/GJez93OY3PX8fi8dYzu1ZGrM5I4r088QQG+DTH9Vng8ZN7ibAd3w9oPnTIaC6fBf/4GHTzQ50JndnPSEHC5m72L05ZPo8bWHNNmjKFnVE+eGvsUXxZ9SU5BDjkFOXy85WMMhv4d+zM6cTRZiVn0iuql3EUanQJmP1fidvFeeFjt1z0qXYaZ7cO4o2wfTb0GqbWW385axb++3MrNI5J58II0fciJiIiIiIhImxcS6OaiAQlcNCCB7XsO8Ubedl7PK+DOVxYTExbEZWd7uDoziR5x4b7uqv8Ki4Gzf+Bs5WWwbrYTNudNh6+mQVicN2y+CLqOBHfzRGzLSpZRWXPsN88raypZWrKUsMAwxnYdy9iuY6mxNazevZqcghzmF8zniSVP8MSSJ4hvF09WYhZZiVkM6TyE0ADNnJczp4DZn616h/tjY6g5LtStwTAtthNNuQZpTY3lgXdX8spX27htVAr3n99H4bKIiIiIiIj4naTodvxyfCp3je1FzrpSZuRuZ/oXW3hmwWYyukZxdWYS3+/fmXZBinB8JjQSBlztbEe+gfWfQP4sWPYq5D0HodHQ+/vOzOaU0RDQdIs4vnnRm7WPv23NKpdx0Te2L31j+3LnwDvZVb6LBQULyCnI4YNNH/DGujcIdgeT2SmzNnD2tPc0Wb+lbdOnk7/atZ6d7/2U3E4n1uGpdBmWevo22aVraiz3v7OC13K3c+eY7tw9IVXhsoiIiIiIiPg1t8twTu84zukdR+k3R3h7cQEz8rYz6c3lPPxePhcOSODqzCQGJEbw7tIipsxeS2FZOZ6FnzJpQiqXpCscbBbB4dDvcmerOAQb5zlh86qZsOQlCI6A1O85YXP3cyEwxNc9BiA2NJZLe17KpT0vpbK6krzivNpSGo9+9SiPfvUoPSJ7MCpxFKMTRzOg4wACXIoN5bvR3xR/VHEQO+N/eDC6A8EBIbx10dskdUhqlktX11jueWs5by4q4Kfn9uCX41T7R0RERERERKS+juHB/L/R3bk9qxt5W/fy2tfbeWdJAa9+vY3OHYLZdbCCymoLQGFZOfe9vQJAIXNzC2rnlMnocyFUHYFN2U4ZjTUfwPLXIKg99BzvhM09x0FQmK97DECgO5BhCcMYljCMewbfw5Z9W5ywuTCHl1a9xPSV0+kQ1IERHu9CgQkjiQyJ9HW3pQVTwOxvrIX3f8GMw4V82T6KBzInNVu4XFVdw6Q3l/POkkJ+PrYnPx/bq1muKyIiIiIiItIaGWPITI4mMzma316UxnvLinho1qracPmo8spqpsxeq4DZlwKCodcEZ6uuhM05sHoWrH4fVr0NAaHQcyz0udg5JqSDr3tcKzkimeSIZG7oewMHKg7w5Y4vmb99PgsKF/DR5o9wGRcDOg4gKzGLUZ5RWihQTqCA2d8sms7W/Lf4S5ckRiQM5cpeVzbLZauqa/jF68t4b1kRvx7fi5+c27NZrisiIiIiIiLSFnQICeT6IV2Z/M7Kk+4vLCvnpYVbGdsnjs4RWrjNp9yB0OM8Z/v+X2Hrf5ywOX8WrH4P3EFO+Yy0i51yGqEnli/1lfZB7RnXdRzjuo6jxtaQvzuf+QXzySnIYeriqUxdPJVOYZ3I8jh1mwd3HqyFAkUBs18pWkLVR/dwf5duBAUG8cjwR5rlN06V1TX8/LWlfLBiB/dM7M2dY7o3+TVFRERERERE2qKEyFAKy8pPaHe7DA/MXMkDM+EsTwTj0uIZ2yeePp3DNdvUl1xuSBnlbBP/BAW5ThmN1bNg3cfgCoCULCds7n0BhMX6use1XMZFv9h+9Ivtx48H/pjSQ6UsKHQWCnxv03u8vu51gt3BDO40mNGJo8lKzKJz+86+7rb4gAJmf3FoD7x+Ay/ExLGcw/zf0EeIaxfX5JetqKrhp68uZvaqYn5zfh9uy+rW5NcUERERERERaasmTUjlvrdXUF5ZXdsWGujm0Uv7cVZiBJ/kFzMnv5jH5q7jr3PWkRgVytg+8YxLi2dwSjSBbpcPe+/nXC7oMsTZJvwBihY7s5rz34X37oL3fwFdR9SFzR1aVljbsV1HLut5GZf1vIyK6gryivNYULCA+QXzWfDVAvgKekT2qA2b+3fsr4UC/YRG2R/U1MDMO1lTXsqT0Z2Y2HUc30v5XpNf9khVNT9+ZTFzV5fw4AVp/HBkSpNfU0RERKSlMsZMBKYCbuBZa+0fj9v/S+BWoAooBX5ord3q3XcjMNl76O+ttS962wcBLwChwIfAXdbaYwtziohIm3K0zvKU2WspLCvHExnKpAmpte094sL50ZgelHxzmE9XlzB3dTGvfr2NF/6zhQ4hAZzTO46xfeIZk9qR8JBAX74V/2YMeAY529iHoHilEzTnz4IPfw0fToKkIZB2EfS5CCK962dN6QkHSwAYA5DtPV9YHExa32zdD3IHMTxhOMMThnN35t1s2e9dKLAghxdXvchzK5+jQ1AHRnpGOgsFekYSERzRbP2T5qWA2R988RgV6z7mvt7pRLlc/GbIb5r8kocrq7nz5UV8traURy7uyw3Dkpv8miIiIiItlTHGDTwJjAMKgFxjzCxrbX69w5YAGdbaQ8aYO4H/A642xkQDvwUyAAss8r52L/AP4DbgK5yAeSLwUXO9LxER8Y1L0j1cku4hOzubMWPGnPSYuPAQrhnchWsGd+FQRRUL1u9ibn4x89aU8O7SIgLdhqHdYmpLaSREqo6uzxgDnc5ytnMnQ8mauprNs+93toSznZnN3nD5BKdqbwbGGFIiUkiJSOHGvjfyTcU3/KfoP+QU5PB54ed8uPlDXMbFwI4DGZU4itGJo+kR2UOlW9oQBcxt3eYc+PT3/L3XYDYc2clT5z1FZEhkk17ycGU1t7+0iJx1pTx66VlcN6RLk15PREREpBUYDGyw1m4CMMa8BlwM1AbM1trP6h2/EPgf7+MJwBxr7R7va+cAE40x2UAHa+1Cb/u/gEtQwCwiIsdpFxTAhL6dmNC3E9U1lsXb9jLXW0rjwXdX8eC7q+jn6VBbSiOtcweFf74U19vZRt8Nuzd6w+Z3Ye5vfd2z7yQ8KJwJyROYkDyBGlvDyl0ra2c3H10osHNYZ7ISvQsFdhpMSECIr7stZ0ABc1u2fwe8+UMWx3Xjhcpirux1JaMSRzXpJcsrqrntX3l8sXEX/3d5f67KTGrS64mIiIi0Eh5ge73nBcCQbzn+FuqC4pO91uPdCk7SfgJjzO3A7QDx8fFkZ2efRteltThw4IDG1k9p7P3XmYz9sHYwLAOKDoSytKSKxSUHmDp3P4/PXU9MiGFgnJuz4wJIjXYR4FLY7Fvp0Cud4C4lDFt42ymPWvPqA+zv0ItD7TxgWlat7X70o194P/aF7mNV+SpWlq/knXXvMGPtDAJNIKkhqfQN7Uvf0L5EBUT5urstXkv73FfA3JbUq8Nz1EFjuD86GE/7Lvw649dNevlDFVXc8kIeCzfvZsoVA7hiUGKTXk9ERESkLTLG/A9OOYzRjXVOa+3TwNMAGRkZ9lRfp5bW7du+Ki9tm8befzXW2F/n/bP0myN8tqaET/KL+XxDKfO2HSY8JIBzUuMYm+bUbe6gus2+9S0Bc++1TzgPgiPAkw6JmeDJgMQMCIttpg7+dxdzMYCzUODOPOYXzCenIIcZe2YA0CuqV+3s5v6x/XG73L7sbovU0j73FTC3JSept/Pn6EiK3IYXRz1Ku8B2TXbpA0eq+OH0XPK27uGxqwbWLi4gIiIiIgAUAvW/2pXobTuGMWYs8BtgtLX2SL3Xjjnutdne9sTj2k84p4iIyHfVMTyYqzKTuCozifKKahasL2Xu6mLmrS5h1rK6us1j+8QzNi0ej+o2tyw//hoK8qAgFwrzYMFfwVY7+6KS68JmTwZ07g8BwT7tbpA7iOGe4Qz3DOfewfeyef9mcrbnkFOYw/SV03l2xbNEBkcywjOCLE8WIzwjtFBgC6WAuQ3LCQ3hzQ7h3Fy2n/S49Ca7zjeHK7lpei5Lt5cx9Zp0LhyQ0GTXEhEREWmlcoGexpgUnBD4GuomjAFgjEkH/glMtNbWnzkwG3jUGHP0+6LjgfustXuMMfuNMUNxFvm7AfhbE78PERHxE6FBbsb37cR4b93mJdv2Mme1U7f5t7NW8dtZq+ibUFe3uW+C6jY3i7C4ky/oFxYHHVOdLf16p63iIOxY5gTOBXmw7UtY+aazzx3kLCp4NHROzICoFGfBQR8wxtAtohvdIrpxU7+b2F+xn/8U/YcFBQtYULCADzZ9gNu4GdBxAKOTRpPlyaJ7ZHf9nWshFDC3QaVuF7+Ii6UgIIAeFRX8ZG9Zk11r/+FKbnjua1YW7uNv16Zz/lmdm+xaIiIiIq2VtbbKGPMTnLDYDTxvrV1ljHkEyLPWzgKmAO2BN7z/s7TNWnuRN0j+HU5IDfDI0QX/gB8BLwChODWbtcCfiIg0OrfLkJEcTUZyNPd9rw8bSw/ULhL4xKfrmTpvPQkRIYxNc8LmISkxBAW0rBrAbcak9bUP/2uZhKAw6Drc2Y7av8OZ3VyQCwWLYMnL8PU/nX3tYsAzyBs6D3Ieh/qmHnKHoA5MTJ7IxOSJVNdUs3J33UKBjy16jMcWPUZCWELdQoGdBxPs9u2MbH+mgLkNmhYZwbLgYAwwrbiUoCa6zr5Dldzw/Ffk79jPk9efzYS+nZroSiIiIiKtn7X2Q+DD49oerPd47Le89nng+ZO05wH9GrGbIiIi/1X3ju3pPro9/290d3YdOMKna0qYk1/M63nb+deXWwkPDmB0akfGpcUzJjWOiFDVbW4xOnSGDhdCnwud59VVULqmrqxGwSJYPwewzv6Ynt6yGoOcms7xfcHdvOPpdjkzlwd0HMBP03/KzoM7WVC4gJyCHN7d+C6vrX2NEHcIQzsPZVTiKLISs+gUpoyqOSlgbisKFwHO7OW3w9uDMbitJba6ukkut/dgBT94/ivW7TzAP64fxNi0+Ca5joiIiIiIiIi0XLHtg7kqI4mrMpy6zV9s2MWc/GLmrSnm/eU7CHAZhnSLZpy3bnNiVNOtDyUN4A6ATv2cLeNmp+3wfiha7JTVKFwEG+bCsledfQEh0HlgXVkNTwZEJDZraY1OYZ24steVXNnrSo5UHyF3Z27t7ObsgmwAUqNSa2c3nxV7lhYKbGIKmNuCkjXw8uVgXPw+Oooqb7OxzmzmyYcb9zdLew5WcP2zX7Gx9AD//MEgzukd16jnFxEREREREZHWJzTIzdg0J0iurrEs3V7GnPxi5uTv5KH38nnovXz6dO7AuLR4xqtuc8sV0gG6jXE2AGuhbFvdDOeCXPj6Gfjy787+9vHO7GbPICd0TkiH4PBm6WqwO5iRnpGM9IzkvsH3sWnfJnIKcphfMJ/nVz7PMyueISo4ihGeEYxOHM1wz3A6BHVolr75EwXMrd3eLfDSJeAOYtUPXuPTBb+u3VXpMsyMiuWOyz8mtpEut+vAEf7n2a/YvOsgz9yQweheHRvpzCIiIiIiIiLSVrhdhkFdoxjUNYp7v9ebTaUHmOtdJPDvn67niXnr6RwRUrtI4NBuqtvcYhkDUV2drd/lTltVBRSvdGY4H11EcM373uNd0LF33QznxAzneRPPIjbG0D2yO90ju3Nzv5vZd2QfXxZ9yfyC+Xxe+Dnvb3oft3GTHpdeO7u5W0Q3/ZKjEShgbs2+2Qn/ugQqyym/YSZ35PzshENqbA3Tlk1j8tDJZ3y5km8Oc/0zX7F97yGeuzGTkT0bK7YWERERERERkbasW8f23N6xPbdndWd3vbrNby4q4KWFW2nvrds8Pi2eMb3iiGinus0tWkAQeM52tsG3OW2H9kDh4rp6zvmzYPG/nH1B7Z2ZzfVD5/CmrZMcERzBxJSJTExxFgpcsWtFbSmNvy76K39d9Fc87T1kJWYxOnE0GZ0ytFBgAylgbq3K98JLl8GBEuwN7/Lb9a9SdqTshMMqaypZWrL0jC9XvP8w1z6zkB1lh5l+02CGdY8543OKiIiIiIiIiP+JaR/MlRlJXJmRxOHKurrNc1eX8IG3bvPglGjGpcUztk88SdGq29wqtIuGnmOdDZzSGns21c1wLsyD//wNarzFXSOS6spqeDIgYSAEhjZJ19wuNwPjBjIwbiA/O/tn7Dy4k5yCHBYULOCd9e/w6ppXCQ0IZUjnIYxOHM0ozyjiw7Te2HelgLk1qjgIr1wFu9fDda/zXNkyPtryEXedfRe3nnVro19ux75yrnvmK4r3H+bFHw5mcEp0o19DRERERERERPxPSKCb8/rEc16feGpqLHGbUZgAACVmSURBVEsLnLrNc/OLefi9fB5+L5/encIZ763tfJYnQiUNWgtjIKa7sw24xmmrLIcdy731nL2hc/5MZ58rAOL71s1wTsyE6O7gavzSKZ3COnFV6lVclXoVh6sOH7tQ4PZsAHpH964tpdEvpp8WCvwWCphbm6oj8Nr1zg/glS+SE+zmic+f4HvJ3+OWfrc0+uUKy8q59umF7DlYwUu3DGZQV4XLIiIiIiIiItL4XC7D2V2iOLtLFPdM7M3mXQeZm1/MnNXF/P2zDTzx6QY6dQhhbFocY/vEM6x7DMEBCv1alcBQ6DLE2Y46UFIXNhfkwvLXIe85Z19IhDPL2ZNRt5BgWON+qz4kIIRRiaMYlTiK++39bCzbyPyC+eQU5PDciud4evnTRAVH1R4zPEELBR5PAXNrUl0Fb90Kmz6Di59iU0I/7vnwenpH9+bhEQ83+m/wtu85xLXPLGTfoUpeumUw6V2iGvX8IiIiIiIiIiKnkhIbxm1Z3bgtqxt7Dlbw6ZoS5uYX8/biQl5euM2p29yrI+PS4jknVXWbW632cdD7fGcDqKmGXeuc0Lkg11lIcMGfwdY4+6NS6mY4ezKgUz8IaJzaycYYekT1oEdUD2456xb2HdnHF4VfkFOYw/yC+czaOIsAE0B6fDpZniyykrJI6ZDi97PqFTC3ZFN6wsGSE9uD2rMv7QJ+9uH1BLmDeOLcJwgNaNwaNdt2O+HyN4crefnWIQxIimzU84uIiIiIiIiIfFfRYUFcMSiRKwYlcriymv9s3MWc/BLmri7mgxU7cLsMg5Odus3j0lS3uVVzuSGuj7Od/QOn7cgB2LG0rp7zls9hxRvOPncQdOrvBM6JGc4s56hkp0THGYoIjuD8budzfrfzqa6pZvmu5eQUOGHzXxb9hb8s+guJ7ROPWSgwyB10xtdtbRQwt2QnC5eB6ooD3JNzD4UHCnl+wvN0CmvcVTe37DrItc8spLyymn/fNpR+nohGPb+IiIiIiIiISEOFBLo5t3c85/aO5w81/Vh2tG7z6mIeeT+fR9536jYfXSTwLE8ELpd/zzBt9YLbQ/JIZztqX6F3hnMeFCyCRS/AV/9w9rWLrVs8MHGQEzqHnFm+5Xa5SY9LJz0unbvOvosdB3awoHAB8wvm89b6t/j3mn8TGhDKsM7DyErMYlTiKOLaxZ3RNVsLBcyt0GPRkXxR9AUPDXuI9Lj0Mz7fzCWFTJm9lqKycjqGB3O4sgq3y8W/bx1KWoJqyoiIiIiIiIhIy+RyGdK7RJHeJYq7J/Zm6+6DzMkvZk5+MU9+toG/fbqB+A7BjO3jLBI4XHWb244Ij7P1vcR5Xl0JJfnees6LnD/XfVx3fGxq3QznxAyI6wvuhkejndt3PmahwK93fl27UOCn2z8FoE90n7qFAmP74TKNv2BhS6CAuZWZ1T6MFyM6cG3va7m81+VnfL6ZSwq57+0VlFdWA1DyzREA7p7YQ+GyiIiIiIiIiLQqXWPCuHVUN24d1Y29R+s2ry7mnSWFvPLVNsKC3IxO7cjYPvGc2zuOyHb+V86gzXIHQucBzpZ5i9NWXgZFi50ZzoXewHnpK86+wHbQeaB3hrO3pnOEp0GXDgkIqQ2SrbWsL1tPTkEOCwoW8MyKZ/jn8n8SHRLNSM9IRieOZljCMMKDwhvlbbcECphbkRVBQTwcE83g8sNMypzUKOecMnttbbhc3ysLt/GjMT0a5RoiIiIiIiIiIs0tKiyIywclcrm3bvOXG3czZ3Uxc/OL+XDFTtwuQ2ZyFGP7xDM+rRNdYlS3uc0JjYTu5zobgLWwd0vdDOeCXPjqn1D9N2d/eOe6Gc6JmU4AHdz+tC5pjKFXVC96RfXi1rNupexwGV8UfcH8gvlkb8+uXSjw7Piza0Pp5A7JrXqhQAXMLdWBuvrLpW4XP4/rSGGAm47V1fy5ZBeBrsZZGbWorPy02kVEREREREREWpuQQDfn9I7jnN5x/P7ifiwv3MdcbymN33+wmt9/sJrU+HDGpsUxLq0T/VW3uW0yBqJTnO2sK5y2qiOwc2W9es55sOZ97/EuiEvzhs7eRQRjU8H13UtdRIZE8v1u3+f73b5PVU0Vy0uXM79gPjkFOfw578/8Oe/PJIUnMTpxNKMSR5ER3/oWClTA3BJVVcCMH9Q+fSoyguXBQbiB1wt3EhUa22iXig4LYvfBihPaEyJDG+0aIiIiIiIiIiIthctlGJgUycCkSH49IZVtuw8xZ3Uxc/J3Mm3+Jp78bCNx4cGc1yee8WnxDOseQ0ig6ja3WQHBTpmMxEF1bQd31c1yLsyD/Jmw+EVnX1A4eNKdwNmT4YTO7b/bYn4BLmfm8tnxZ/OLQb+g6EBRbd3mN9a9wcurX6ZdQDuGJXgXCvSMomO7jseco/RQKY/vfJx+5f2IbcSM8Ew0acBsjJkITAXcwLPW2j8etz8Y+BcwCNgNXG2t3WKMGQf8EQgCKoBJ1tpPm7KvLYa18OGvYftCuOJ5SlJG8fZb48DW4HIFEv3LNdBIf3neXVrInoMVGMDWaw8NdDNpQmqjXENEREREREREpCXrEtOOW0amcMvIFMoOVfDZ2hLm5Bcza2khr369jXZBbrJ6dmRcmlO3OSqsdc0ulQYIi4VeE5wNoKYGdm+om+FcmAdfTIWaKmd/RBdvWY0MJ3Tu3B8C//vkzYT2CVzT+xqu6X0N5VXlfL3DWShwfsF85m2bB0BaTBpZiVmMThxNWkwa05ZPY9ORTUxbNo3JQyc31b+B09JkAbMxxg08CYwDCoBcY8wsa21+vcNuAfZaa3sYY64B/gRcDewCLrTWFhlj+gGzgYZV2W5tcp91fiMy8pfQ73J+88lt1Nia2t2N9ZfnX19u4bezVpGZEs2l6Qn8/dONFJWVkxAZyqQJqVyS7h//ukVEREREREREjopsF8Sl6Ylcmp7IkSpv3eb8YuauLubjVTtxGchIjmZ8Wjzj0uLpGhPm6y5Lc3C5oGMvZxt4ndNWcQh2LKsLnQtyYdXb3uMDIL5fXVkNTwbEdHdKdJxCaEAoo5NGMzppNJPtZNbtXceCwgXM3z6fp5c/zbRl04gKjmJfxT4slpkbZnLHgDtaxCzmppzBPBjYYK3dBGCMeQ24GKgfMF8MPOR9/Cbwd2OMsdYuqXfMKiDUGBNsrT3ShP31vc0L4ON7oddEOPcB8nfls3DHwtrdlTWVZ/yXx1rL43PXM3Xeesb2iefv16UTEujm2sFdG+tdiIiIiIiIiIi0esEBbsakxjEmNY7fXdyPlUX7mHNc3eaece0Z5w2bByRGqm6zPwlqB12HOdtR3+ysm+FckAdL/w25zzj7QiLrwubEDKeuc7vok57aGENqdCqp0am1CwV+XvQ5Ty15ir1H9gJQY2tazCzmpgyYPcD2es8LgCGnOsZaW2WM2QfE4MxgPupyYPHJwmVjzO3A7QDx8fFkZ2c3WuebW0h5MYMW/YqKkE4sjruRqvnzebDwwROOq6qu4oEPH+DqmKtP+xo11vLK6grmbatipCeAa5O+YeEXCxqj+03qwIEDrXpspeE09v5N4++/NPb+S2MvIiIiLZXLZeifGEn/xEh+NT6V7XsO1YbN/8zZxFPZG+kYHszYPnGM7RPPiB6xqtvsj8I7QZ8LnA2gphpK19TNcC5cBBv+RG2x2ujux4bO8f0g4MQSLJEhkQzpNISHyh+qbWuMiaiNpUUv8meM6YtTNmP8yfZba58GngbIyMiwY8aMab7ONaYjB+C58RDgIvCWWYyK6c7ra1+nbFvZCYdWU01pYCmn+14rqmr41RvLmLetiNuzunHf93pjvmVafkuSnZ192u9X2gaNvX/T+Psvjb3/0tiLiIhIa5EU3Y4fjkzhhyNT2Heosl7d5iJe/Xo7oYFusnrFMi6tE+f2jiNadZv9k8sN8X2dbdCNTtuRb6BoiRM4FyyCTdmwfIazzx0MnQfUzXBOzITILmAM017MoibEDfVmyddUHmbaC6OYfOfq5n9v9TRlwFwIJNV7nuhtO9kxBcaYACACZ7E/jDGJwDvADdbajU3YT9+qqYGZd0Dparj+TYjpzrb92/hz3p8ZnjCcaWOnnXEQfKiiijteXkzOulLu/V5v7hjdvZE6LyIiIiIiIiLi3yLaBXJJuodL0j0cqapm4aY9zMnfydz8EmavKnbqNneNZlxaPGPT4kmJVd1mvxYcDilZzgZgLewrqJvhXJAHec/Dwqec/WEdwZPBMncNla5jo9xKl2GpuwZfa8qAORfoaYxJwQmSrwGuO+6YWcCNwJfAFcCn1lprjIkEPgDutdZ+0YR99L2cKbD6PZjwKPQ4j+qaau7//H4CXAE8PPzhMw6Xyw5VcPMLuSzbXsafLj+LqzO7NFLHRURERERERESkvuAAN6N7dWR0r4787mLLysL9zMnfySf5xfzhw9X84cPV9PDWbR7bJ570JNVt9nvGQGSSs/W7zGmrroTild56zk7o/Obunb7t57dosoDZW1P5J8BswA08b61dZYx5BMiz1s4CngNeMsZsAPbghNAAPwF6AA8aY44WIh5vrS1pqv76xOr3IPtRGHAdDP0RANNXTWdZ6TL+OOqPdArrdNqnnLmkkCmz11JUVk58hxCstew9VMlT1w9iYr/TP5+IiIiIiIiIiJw+YwxnJUZwVmIEv/TWbZ672qnb/HTOJv6RvZHY9nV1m0f2VN1m8XIHQkK6s3Gb0/ZQhE+79G2atAaztfZD4MPj2h6s9/gwcOVJXvd74PdN2TefK14Fb/8/p4j3BY+BMazds5Ynlz7J+K7jOT/l/NM+5cwlhdz39grKK6sB2Ln/MAA/Oqe7wmURERERERERER9Kim7HzSNSuHmEU7c5e10Jn+QX8/7yHbyW69RtHtUzlrFp8ZzXO46Y9sG+7rLId9KiF/lrU6b0hIMnmYC9dwsEhlBRXcF9n99HZHAkk4dOblBpjCmz19aGy/W9u6SIuyf0bkCnRURERERERESksUW0C+TigR4uHuihoqqGhZt2Mye/mLmri/kk36nbPKhrFGP7xDMuLZ5uHdvXvvbot9cLy8rxLPyUSRNSuSTd48N3I/5OAXNzOVm4DHBoFwBPLn2S9XvX8+R5TxIVEtWgSxSVlZ9Wu4iIiIiIiIiI+FZQgIusXh3J6tWRRy7uy6qi/czJd0pp/O9Ha/jfj9bQvWMYY9PiCQlw8XTOJsornYXdCsvKue/tFQAKmdu6sLiT54thcc3fl+MoYG4BlpQs4YVVL3B5z8vJSsxq0Dn2lVcSHODicNWJK0cmRIaeaRdFRERERERERKSJGWPo54mgnyeCX4zrRcHeQ8xbXcKc/GKeW7CZqhp7wmvKK6uZMnutAua2btL62ofZ2dmMGTPGd305jsvXHfB3h4zhN5//hs5hnZmUOalB51hVtI+L/v45R6pqCHQfW1ojNNDNpAmpjdFVERERERERERFpRolR7bhxeDIv3zqERQ+MO+VxhWXlvPTlFlbv2E/NSUJokaakGcw+9pfoSAq+KeD5Cc8TFhh22q+fkbuNB95dRXS7IN68cxjb95QzZfZaisrKSYgMVR0eEREREREREZE2ICI0EE9kKIUnKYXqMvDAu6sACA8JIKNrFBnJ0WQmR9M/MYKQQHdzd1f8iALm5rD965M2LwgN4fUO4dzU90YyOmWc1inLK6p58N2VvLGogFE9Y3n86oHEtA9mUFfV3BERERERERERaYsmTUjlvrdXUF5ZXdsWGujm0Uv7kZEcTe6WPeRu2Uvelj18tnYtAEFuF/0TI7yBcxQZXaOJaBfoq7cgbZAC5qa2eyO8eg0YN9i6H/59Lhe/jY2mR1UNP0n/yWmdcvOug9z58iLWFn/DXef15Gfn9cTtMv/9hSIiIiIiIiIi0modnVQ4ZfZaCsvK8Rz37fWk6HZcdnYiAHsOVrBoqxM2527Zw3Ofb2LafKd8Rmp8OBnJUWQmR5OZEo1H63fJGVDA3JQO7oZXrgRr4Se5ENO9dtcf5t/N3q1zePL7/ybYHfydT/nhih3c/eZyAt2G6TdlMibV9ytFioiIiIiIiIhI87gk3cMl6Z7/utBbdFgQ49LiGZcWDzjfhl9WUOYNnPcya2kRr3y1DYCEiJDaGc6ZKdH0igvHpcmM8h0pYG4qlYfhtetgXwHc+F5tuFx6qJRbPrmFzfs289P0n9Inps+3nmbmksLamsrtgt0cPFLNwKRInrz+bP12SUREREREREREvpPQIDdDu8UwtFsMANU1ljU795O3ZS+5W/bw1ebdzFpWBECHkAAGees4D06J5iyP6jjLqSlgbgo1NfDO/4PtX8GVL0CXIbW7Hlv0GJv3bSYmOIYf9vvht55m5pLCY+rqHDxSjdtl+J8hXRQui4iIiIiIiIhIg7ldhr4JEfRNiODG4clYaynYW15bxzn3FHWcB6dEMaiL6jhLHQXMTWHeQ5A/E8b/HvpeUtu848AO3t/0PgDfVH5D2ZEyYkNjT3maKbPXHlO0HZzfLj02dz1XZCQ1QcdFRERERERERMQfGWNIim53yjrOX5+kjnNmilPHOSNZdZz9mQLmxpb7HHwxFTJvhWHHLt7343k/xuL8EFos05ZNY/LQyac8VWFZ+Unbi07RLiIiIiIiIiIi0lhOVcc5d/MecrfuZeaSIl5eeFwd5xSnlrPqOPsPBcyNad1s+PDX0GsiTPwTmLofopfyX2J92fra55U1lczcMJM7Btxx0lnMLy/cesrLJOg3QiIiIiIiIiIi0sxOVcf5aOC8cNOxdZwzkqPJSHZmOauOc9ulgLmxFC2BN26GTv3hiufBXfevdmnJUqbkTsFgamcwA9TYmhNmMdfUWKZ8spZ/ZG8krXM4m3Yd5HBlTe3+0EA3kyakNs97EhEREREREREROYX6dZxvGpFSW8f56817yNvq1HL+dE0J4NRxHpDk1HHOTFYd57ZEAXNjKNsO/74a2sXAda9DUFjtruKDxfwi+xcEuAKorKk85mWVNZUsLVla+7yiqoa731zGzKVFXDekC49c1Jf3l+9gyuy1FJWVkxAZyqQJqVyS7mmudyYiIiIiIiIiIvKd1K/jfPmgujrOeVv2kLfVWTjwmZxN/CPbYoxTx/noDGfVcW69FDA3xJSecLDkxPbQaAiPr316uOowd312F4cqD/H6Ba/TI6rHKU+5/3Ald7y0iP9s3M2kCan8aEx3jDFcku5RoCwiIiIiIiIiIq1SdFgQ4/t2YnzfToBTx3np9jLytpxYx9kTGUpGchQZydEMTo6mZ1x71XFuBRQwN8TJwmWA8j21D621PPTlQ6zavYqp50z91nC5qKycm6fnsrH0AH+9akDtSp0iIiIiIiIiIiJtSWiQm2HdYxjWva6O8+od+2sD5y837ubdpSfWcR6cHM1ZiREEB6iOc0ujgLmJvLDqBT7Y9AE/GfgTzu1y7imPW71jPzdPz+XgkSpe/OFgRvQ4ccE/ERERERERERGRtsjtMvTzRNDPU1fHefuecnK3nKSOc4CLAYkRtTOcz+4aRUSo6jj7mgLmJrCgYAGPLXqM8V3Hc3v/20953BcbdnHHS4sICw7g9TuG0adzh2bspYiIiIiIiIiISMtijKFLTDu6xNTVcd594AiLtu49ro7zxhPqOGcmR5OgOs7NTgFzI9u8bzP35NxDanQqvxvxO4ypqxMzc0lh7YJ9ke0CKTtUSa/4cF74YSadI/SXX0RERERERERE5Hgx7YNPWcf56y17eGdx4Ql1nI8Gzqrj3PQUMDei/S7Dzz79GQGuAKaeM5V2ge1q981cUsh9b6+gvLIagL2HKnEZuGlEV4XLIiIiIiIiIiIi39HxdZyrqmtYs/Mbp47zlmPrOEeEBpLR1Vk4MDM5SnWcm4AC5oYIizthob9q4J5OCRR8U8Az458hoX3CMfunzF5bGy4fVWPh759u5NrBXZu6xyIiIiIiIiIiIm1SgNt1yjrOR7d59eo4D0yMrJ3lrDrOZ04Bc0NMWn9C09RFf+XzldN5YMh9ZHTKOGF/UVn5SU91qnYRERERERERERE5fd9WxznXO8v56ZxNPFWvjnNmcnRt6Kw6zqdHAfMZKD1UyqScSUxInsD0ldO5qtdVXJV61YnHfXOEALehstqesE9/YUVERERERERERJrWqeo4H53h/PbiAl5auBVw6jhnJh8tq6E6zv+NAuYzMG35NBYXL2ZJ8RIGxQ/i3sH3nnDMptID3Dj9a7AQ5HZRUV1Tuy800M2kCanN2WURERERERERERG/d6o6zrlb9pC3ZS9fbNzNzJPUcR6cEkU/j+o416eAuYFKD5Uyc/1MrPef3wz5DYHuY+u1LNq6h1tfzMNlDG/eOZzNuw4yZfZaisrKSYgMZdKEVC5J9/joHYiIiIiIiIiIiAgcW8f55np1nL/esse7eOAp6jinRHN2F/+u46yAuYGmLZ9GtXUW7QswAcxYO4PJQyfX7v945Q7uem0pCZGhvHBzJl1jwhiQFKlAWURERMRPGWMmAlMBN/CstfaPx+3PAh4H+gPXWGvf9LafAzxW79De3v0zjTEvAKOBfd59N1lrlzbh2xARERHxC/XrOF9Rr45z3ta93sD55HWcM1OiyUyOonOE/5TFVcDcAKWHSnl3w7u1AXOVrWLmhpncMeAOYkNjmf7FZh55P5/0pEievTGT6LAgH/dYRERERHzJGOMGngTGAQVArjFmlrU2v95h24CbgF/Xf6219jNgoPc80cAG4JN6h0w6GkaLiIiISNOJaR/MhL6dmOCt43yoooql28vI27L3lHWcncA5mh4d224dZwXMDTBt+TRqbM0xbTW2hn8snYbZfRnPfr6ZCX3jmXpNOiGBqsciIiIiIgwGNlhrNwEYY14DLgZqA2Zr7RbvvpqTncDrCuAja+2hpuuqiIiIiHwX7YICGN49luHdY4Fvr+Mc2a6ujnNmctuq46yAuQGWlSyjsqbymLbKmko+WPclO1enc9PwZB64IA13G/2thIiIiIicNg+wvd7zAmBIA85zDfDX49r+YIx5EJgH3GutPXL8i4wxtwO3A8THx5Odnd2AS0tLd+DAAY2tn9LY+y+NvX/T+LdcKUCKB65IcFNaHsq6vdWs21vDym2lzF3t1HEOcEG3CBe9otz0inLRPdJNWOB3yxJb2tgrYG6ANy96k5lLCmsX7OsUEUJIoIvNuw7xm/P7cOuoFIxRuCwiIiIijccY0xk4C5hdr/k+YCcQBDwN3AM8cvxrrbVPe/eTkZFhx4wZ09TdFR/Izs5GY+ufNPb+S2Pv3zT+rVP9Os5fb9nLx1v28f4mizHQu1MHMpPrZjkfX8f5aB5ZWGbwRNYwaUJqi1jvTQFzA8xcUsh9b6+gvNKpwbxj32EAbhjWlduyuvmyayIiIiLSMhUCSfWeJ3rbTsdVwDvW2tqv0llrd3gfHjHGTOe4+s0iIiIi0rL8tzrOby0q4F9fOnWcE6NCyUyOJiM5iv3llTwxbz3llU41tcKycu57ewWAz0NmBcwNMGX22tpwub55q0t45GIfdEhEREREWrpcoKcxJgUnWL4GuO40z3EtzozlWsaYztbaHcb5+twlwMpG6KuIiIiINJNvq+Ocu2UPn2/YxTtLTj4vobyymimz1ypgbo2KyspPq11ERERE/Ju1tsoY8xOc8hZu4Hlr7SpjzCNAnrV2ljEmE3gHiAIuNMY8bK3tC2CMScaZAT3/uFO/YozpCBhgKXBHs7whEREREWkSAW4X/TwR9PNEcPOIFKy1bNtziNFTsk96fEvIIxUwN0BCZCiFJxm8hMjQkxwtIiIiIgLW2g+BD49re7De41yc0hkne+0WnIUCj28/t3F7KSIiIiItiTGGrjFheFpwHunydQdao0kTUgkNdB/TFhroZtKEVB/1SERERERERERERNqqlpxHagZzAxytazJl9lqKyspJiAxtMas2ioiIiIiIiIiISNtSP48sLCvH04LySAXMDXRJuqdFDKCIiIiIiIiIiIi0fUfzyOzsbMaMGePr7tRSiQwRERERERERERERaRAFzCIiIiIiIiIiIiLSIAqYRURERERERERERKRBFDCLiIiIiIiIiIiISIMoYBYRERERERERERGRBlHALCIiIiIiIiIiIiINooBZRERERERERERERBpEAbOIiIiIiIiIiIiINIgCZhERERERERERERFpEAXMIiIiIiIiIiIiItIgCphFREREREREREREpEEUMIuIiIiIiIiIiIhIgyhgFhEREREREREREZEGUcAsIiIiIiIiIiIiIg2igFlEREREREREREREGkQBs4iIiIiIiIiIiIg0iAJmEREREREREREREWkQY631dR8ahTGmFNjq635Ik4gFdvm6E+ITGnv/pvH3Xxp7/+Wrse9qre3og+v6hO6b2zR9fvovjb3/0tj7N42//2pR981tJmCWtssYk2etzfB1P6T5aez9m8bff2ns/ZfGXuTM6GfIf2ns/ZfG3r9p/P1XSxt7lcgQERERERERERERkQZRwCwiIiIiIiIiIiIiDaKAWVqDp33dAfEZjb1/0/j7L429/9LYi5wZ/Qz5L429/9LY+zeNv/9qUWOvGswiIiIiIiIiIiIi0iCawSwiIiIiIiIiIiIiDaKAWUREREREREREREQaRAGztCjGmOeNMSXGmJX12qKNMXOMMeu9f0b5so/SNIwxScaYz4wx+caYVcaYu7ztGv82zhgTYoz52hizzDv2D3vbU4wxXxljNhhjZhhjgnzdV2kaxhi3MWaJMeZ973ONvR8wxmwxxqwwxiw1xuR52/SZL/Id6b7Zf+m+2X/pvll03+yfWsN9swJmaWleACYe13YvMM9a2xOY530ubU8V8CtrbRowFPixMSYNjb8/OAKca60dAAwEJhpjhgJ/Ah6z1vYA9gK3+K6L0sTuAlbXe66x9x/nWGsHWmszvM/1mS/y3b2A7pv9le6b/Zfum0X3zf6rRd83K2CWFsVamwPsOa75YuBF7+MXgUuas0/SPKy1O6y1i72Pv8H5j6YHjX+bZx0HvE8DvZsFzgXe9LZr7NsoY0wi8H3gWe9zg8ben+kzX+Q70n2z/9J9s//SfbN/032zHKdFfeYrYJbWIN5au8P7eCcQ78vOSNMzxiQD6cBXaPz9gverXkuBEmAOsBEos9ZWeQ8pwPkfJ2l7HgfuBmq8z2PQ2PsLC3xijFlkjLnd26bPfJEzo58hP6P7Zv+j+2a/9ji6b/ZXLf6+OcCXFxc5XdZaa4yxvu6HNB1jTHvgLeDn1tr9zi9lHRr/tstaWw0MNMZEAu8AvX3bI2kOxpgLgBJr7SJjzBgfd0ea30hrbaExJg6YY4xZU3+nPvNFzox+hto+3Tf7J903+yfdN/u9Fn/frBnM0hoUG2M6A3j/LPFxf6SJGGMCcW6SX7HWvu1t1vj7EWttGfAZMAyINMYc/UVoIlDoq35JkxkBXGSM2QK8hvMVv6lo7P2CtbbQ+2cJzv8gD0af+SJnSj9DfkL3zaL7Zr+j+2Y/1hrumxUwS2swC7jR+/hG4F0f9kWaiLd+1HPAamvtX+vt0vi3ccaYjt4ZGBhjQoFxOLUEPwOu8B6msW+DrLX3WWsTrbXJwDXAp9ba69HYt3nGmDBjTPjRx8B4YCX6zBc5U/oZ8gO6b/Zfum/2X7pv9l+t5b7ZWKtvzUjLYYx5FRgDxALFwG+BmcDrQBdgK3CVtfb4BU2klTPGjAQWACuoqyl1P049OY1/G2aM6Y+zKIEb5xefr1trHzHGdMP57Xw0sAT4H2vtEd/1VJqS96t+v7bWXqCxb/u8Y/yO92kA8G9r7R+MMTHoM1/kO9F9s//SfbP/0n2zgO6b/U1ruW9WwCwiIiIiIiIiIiIiDaISGSIiIiIiIiIiIiLSIAqYRURERERERERERKRBFDCLiIiIiIiIiIiISIMoYBYRERERERERERGRBlHALCIiIiIiIiIiIiINooBZRKQFM8YkG2NWtvRzioiIiIj4ku6bRUR8RwGziIiIiIiIiIiIiDSIAmYRkVbCGNPNGLPEGJN5XPtrxpjv13v+gjHmCu+MiwXGmMXebfhJznmTMebv9Z6/b4wZ43083hjzpfe1bxhj2jfduxMRERERaRy6bxYRaV4KmEVEWgFjTCrwFnCTtTb3uN0zgKu8xwUB5wEfACXAOGvt2cDVwBOncb1YYDIw1vv6POCXZ/o+RERERESaku6bRUSaX4CvOyAiIv9VR+Bd4DJrbf5J9n8ETDXGBAMTgRxrbbkxJgL4uzFmIFAN9DqNaw4F0oAvjDEAQcCXDX8LIiIiIiJNTvfNIiI+oIBZRKTl2wdsA0YCJ9woW2sPG2OygQk4My5e8+76BVAMDMD5xsrhk5y7imO/zRLi/dMAc6y11zZC/0VEREREmoPum0VEfEAlMkREWr4K4FLgBmPMdac4ZgZwMzAK+NjbFgHssNbWAD8A3Cd53RZgoDHGZYxJAgZ72xcCI4wxPQCMMWHGmNOZySEiIiIi0tx03ywi4gMKmEVEWgFr7UHgAuAXxpiLTnLIJ8BoYK61tsLb9hRwozFmGdAbOHiS130BbMaZ4fEEsNh7vVLgJuBVY8xynK/59W60NyQiIiIi0gR03ywi0vyMtdbXfRARERERERERERGRVkgzmEVERERERERERESkQRQwi4iIiIiIiIiIiEiDKGAWERERERERERERkQZRwCwiIiIiIiIiIiIiDaKAWUREREREREREREQaRAGziIiIiIiIiIiIiDSIAmYRERERERERERERaZD/D5PvG1V2NWtqAAAAAElFTkSuQmCC","text/plain":["<Figure size 1440x1080 with 4 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["# Creating a subplot with 2 rows and 2 columns\n","fig, axs = plt.subplots(2, 2, figsize=(20, 15))\n","\n","# Plotting F1 Score\n","axs[0, 0].plot(k_values, initial_retrieval_results_f_1, marker='o', label='Initial Retrieval')\n","axs[0, 0].plot(k_values, baseline_reranker_results_f_1, marker='s', label='Baseline Reranker')\n","axs[0, 0].plot(k_values, bertopic_reranker_results_f_1, marker='^', label='BERTopic Reranker')\n","axs[0, 0].set_title('F1 Score by k values')\n","axs[0, 0].set_xlabel('k value')\n","axs[0, 0].set_ylabel('F1 Score')\n","axs[0, 0].legend()\n","axs[0, 0].grid(True)\n","\n","# Plotting Recall\n","axs[0, 1].plot(k_values, initial_retrieval_results_recall, marker='o', label='Initial Retrieval')\n","axs[0, 1].plot(k_values, baseline_reranker_results_recall, marker='s', label='Baseline Reranker')\n","axs[0, 1].plot(k_values, bertopic_reranker_results_recall, marker='^', label='BERTopic Reranker')\n","axs[0, 1].set_title('Recall by k values')\n","axs[0, 1].set_xlabel('k value')\n","axs[0, 1].set_ylabel('Recall')\n","axs[0, 1].legend()\n","axs[0, 1].grid(True)\n","\n","# Plotting Precision\n","axs[1, 0].plot(k_values, initial_retrieval_results_precision, marker='o', label='Initial Retrieval')\n","axs[1, 0].plot(k_values, baseline_reranker_results_precision, marker='s', label='Baseline Reranker')\n","axs[1, 0].plot(k_values, bertopic_reranker_results_precision, marker='^', label='BERTopic Reranker')\n","axs[1, 0].set_title('Precision by k values')\n","axs[1, 0].set_xlabel('k value')\n","axs[1, 0].set_ylabel('Precision')\n","axs[1, 0].legend()\n","axs[1, 0].grid(True)\n","\n","# Plotting nDCG Score\n","axs[1, 1].plot(k_values, initial_retrieval_results_nDCG, marker='o', label='Initial Retrieval')\n","axs[1, 1].plot(k_values, baseline_reranker_results_nDCG, marker='s', label='Baseline Reranker')\n","axs[1, 1].plot(k_values, bertopic_reranker_results_nDCG, marker='^', label='BERTopic Reranker')\n","axs[1, 1].set_title('nDCG Score by k values')\n","axs[1, 1].set_xlabel('k value')\n","axs[1, 1].set_ylabel('nDCG Score')\n","axs[1, 1].legend()\n","axs[1, 1].grid(True)\n","\n","# Adjusting layout and displaying the plots\n","plt.tight_layout()\n","plt.show()"]},{"cell_type":"markdown","metadata":{},"source":["<h2> Sources: </h2>\n","\n","* https://www.geeksforgeeks.org/text-preprocessing-in-python-set-1/\n","* https://www.kaggle.com/code/namansood/document-ranking-ir-system-word2vec-embeddings\n","* https://pypi.org/project/rank-bm25/"]},{"cell_type":"code","execution_count":124,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T21:24:23.876178Z","iopub.status.busy":"2024-01-03T21:24:23.875511Z","iopub.status.idle":"2024-01-03T21:24:29.656502Z","shell.execute_reply":"2024-01-03T21:24:29.655288Z","shell.execute_reply.started":"2024-01-03T21:24:23.876120Z"},"trusted":true},"outputs":[],"source":["scores_lam = dict()\n","for lam_value in lam_values:\n"," bertopic_reranker_lam_scores = dict()\n"," for k in k_values:\n"," scores = evaluate(results_for_different_lams[lam_value], k)\n"," bertopic_reranker_lam_scores[k] = scores\n"," scores_lam[lam_value] = bertopic_reranker_lam_scores"]},{"cell_type":"code","execution_count":125,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T21:24:49.722093Z","iopub.status.busy":"2024-01-03T21:24:49.721617Z","iopub.status.idle":"2024-01-03T21:24:49.741793Z","shell.execute_reply":"2024-01-03T21:24:49.740575Z","shell.execute_reply.started":"2024-01-03T21:24:49.722055Z"},"trusted":true},"outputs":[{"data":{"text/plain":["{1.2: {3: {'f_1': 0.02137001771910353,\n"," 'precision': 0.011315789473684215,\n"," 'recall': 0.19736842105263155,\n"," 'nDCG': 0.3567336087776632},\n"," 4: {'f_1': 0.029268432519206498,\n"," 'precision': 0.015789473684210534,\n"," 'recall': 0.20723684210526316,\n"," 'nDCG': 0.3387887500965494},\n"," 5: {'f_1': 0.03783657003471244,\n"," 'precision': 0.020789473684210538,\n"," 'recall': 0.21842105263157896,\n"," 'nDCG': 0.3254128326590611},\n"," 6: {'f_1': 0.04468892820286009,\n"," 'precision': 0.025000000000000015,\n"," 'recall': 0.21929824561403505,\n"," 'nDCG': 0.3158658688374358},\n"," 7: {'f_1': 0.05273536007200668,\n"," 'precision': 0.030000000000000016,\n"," 'recall': 0.22744360902255642,\n"," 'nDCG': 0.308625281389308},\n"," 10: {'f_1': 0.07406262545006927,\n"," 'precision': 0.0442105263157895,\n"," 'recall': 0.24057017543859646,\n"," 'nDCG': 0.28300583381381733},\n"," 20: {'f_1': 0.1132536014591653,\n"," 'precision': 0.07631578947368418,\n"," 'recall': 0.24656228274649325,\n"," 'nDCG': 0.24020967725033257},\n"," 30: {'f_1': 0.12628539760607768,\n"," 'precision': 0.09342105263157892,\n"," 'recall': 0.23542462058342276,\n"," 'nDCG': 0.2187536976058229},\n"," 40: {'f_1': 0.13881667308622633,\n"," 'precision': 0.11052631578947364,\n"," 'recall': 0.23716928643641721,\n"," 'nDCG': 0.19742563005153},\n"," 50: {'f_1': 0.14556567488821576,\n"," 'precision': 0.1228947368421052,\n"," 'recall': 0.23680116546834148,\n"," 'nDCG': 0.18633380335568112}},\n"," 1.5: {3: {'f_1': 0.02137001771910353,\n"," 'precision': 0.011315789473684215,\n"," 'recall': 0.19736842105263155,\n"," 'nDCG': 0.36761259805323043},\n"," 4: {'f_1': 0.028781103084508642,\n"," 'precision': 0.015526315789473692,\n"," 'recall': 0.20394736842105263,\n"," 'nDCG': 0.3522630905573355},\n"," 5: {'f_1': 0.03735810113519091,\n"," 'precision': 0.020526315789473698,\n"," 'recall': 0.2157894736842105,\n"," 'nDCG': 0.34057203643295536},\n"," 6: {'f_1': 0.044219003390830014,\n"," 'precision': 0.024736842105263175,\n"," 'recall': 0.2171052631578947,\n"," 'nDCG': 0.33648246659831077},\n"," 7: {'f_1': 0.052273679554924506,\n"," 'precision': 0.029736842105263173,\n"," 'recall': 0.22556390977443613,\n"," 'nDCG': 0.3273528742202902},\n"," 10: {'f_1': 0.07406262545006928,\n"," 'precision': 0.0442105263157895,\n"," 'recall': 0.24057017543859646,\n"," 'nDCG': 0.30027573618182124},\n"," 20: {'f_1': 0.11357254720520335,\n"," 'precision': 0.07657894736842104,\n"," 'recall': 0.24683065472539154,\n"," 'nDCG': 0.25271751792123504},\n"," 30: {'f_1': 0.12714793205359917,\n"," 'precision': 0.09421052631578944,\n"," 'recall': 0.2360219399307421,\n"," 'nDCG': 0.2304155597260177},\n"," 40: {'f_1': 0.139247256443902,\n"," 'precision': 0.11105263157894732,\n"," 'recall': 0.237016374943478,\n"," 'nDCG': 0.20663999242512765},\n"," 50: {'f_1': 0.14541447319529882,\n"," 'precision': 0.1228947368421052,\n"," 'recall': 0.23599731372837882,\n"," 'nDCG': 0.19022281905314956}},\n"," 2: {3: {'f_1': 0.02137001771910353,\n"," 'precision': 0.011315789473684215,\n"," 'recall': 0.19736842105263155,\n"," 'nDCG': 0.35217579785511344},\n"," 4: {'f_1': 0.028781103084508642,\n"," 'precision': 0.015526315789473692,\n"," 'recall': 0.20394736842105263,\n"," 'nDCG': 0.34827045924291733},\n"," 5: {'f_1': 0.03735810113519091,\n"," 'precision': 0.020526315789473698,\n"," 'recall': 0.2157894736842105,\n"," 'nDCG': 0.33882964121847015},\n"," 6: {'f_1': 0.044219003390830014,\n"," 'precision': 0.024736842105263175,\n"," 'recall': 0.2171052631578947,\n"," 'nDCG': 0.3363461610398492},\n"," 7: {'f_1': 0.052273679554924506,\n"," 'precision': 0.029736842105263173,\n"," 'recall': 0.22556390977443613,\n"," 'nDCG': 0.3308458515733193},\n"," 10: {'f_1': 0.07450122194129735,\n"," 'precision': 0.044473684210526346,\n"," 'recall': 0.2418859649122807,\n"," 'nDCG': 0.30593150319079393},\n"," 20: {'f_1': 0.11400395359001787,\n"," 'precision': 0.07684210526315786,\n"," 'recall': 0.2480268269741954,\n"," 'nDCG': 0.2567146283321391},\n"," 30: {'f_1': 0.12757933843841368,\n"," 'precision': 0.09447368421052628,\n"," 'recall': 0.23721811217954594,\n"," 'nDCG': 0.23066769026982406},\n"," 40: {'f_1': 0.13967866282871652,\n"," 'precision': 0.11131578947368415,\n"," 'recall': 0.23821254719228185,\n"," 'nDCG': 0.2063959838017688},\n"," 50: {'f_1': 0.1458458795801133,\n"," 'precision': 0.12315789473684202,\n"," 'recall': 0.23719348597718257,\n"," 'nDCG': 0.1905707583249521}},\n"," 2.5: {3: {'f_1': 0.02137001771910353,\n"," 'precision': 0.011315789473684215,\n"," 'recall': 0.19736842105263155,\n"," 'nDCG': 0.3583505179343602},\n"," 4: {'f_1': 0.028781103084508642,\n"," 'precision': 0.015526315789473692,\n"," 'recall': 0.20394736842105263,\n"," 'nDCG': 0.34455822114959667},\n"," 5: {'f_1': 0.03735810113519091,\n"," 'precision': 0.020526315789473698,\n"," 'recall': 0.2157894736842105,\n"," 'nDCG': 0.34078361540846763},\n"," 6: {'f_1': 0.044219003390830014,\n"," 'precision': 0.024736842105263175,\n"," 'recall': 0.2171052631578947,\n"," 'nDCG': 0.34092607470195774},\n"," 7: {'f_1': 0.052273679554924506,\n"," 'precision': 0.029736842105263173,\n"," 'recall': 0.22556390977443613,\n"," 'nDCG': 0.3313893347571242},\n"," 10: {'f_1': 0.07450122194129735,\n"," 'precision': 0.044473684210526346,\n"," 'recall': 0.2418859649122807,\n"," 'nDCG': 0.30814484944827053},\n"," 20: {'f_1': 0.11362801374039382,\n"," 'precision': 0.07657894736842104,\n"," 'recall': 0.24736893223735326,\n"," 'nDCG': 0.25702637613212626},\n"," 30: {'f_1': 0.1272503910699926,\n"," 'precision': 0.09421052631578944,\n"," 'recall': 0.23677951568831784,\n"," 'nDCG': 0.22942427360812398},\n"," 40: {'f_1': 0.1393862651678978,\n"," 'precision': 0.11105263157894732,\n"," 'recall': 0.2378835998238608,\n"," 'nDCG': 0.20674307592426877},\n"," 50: {'f_1': 0.14558272168537648,\n"," 'precision': 0.1228947368421052,\n"," 'recall': 0.2369303280824458,\n"," 'nDCG': 0.19068412896415185}},\n"," 3: {3: {'f_1': 0.02137001771910353,\n"," 'precision': 0.011315789473684215,\n"," 'recall': 0.19736842105263155,\n"," 'nDCG': 0.3560716124730853},\n"," 4: {'f_1': 0.028781103084508642,\n"," 'precision': 0.015526315789473692,\n"," 'recall': 0.20394736842105263,\n"," 'nDCG': 0.3426624624489207},\n"," 5: {'f_1': 0.03735810113519091,\n"," 'precision': 0.020526315789473698,\n"," 'recall': 0.2157894736842105,\n"," 'nDCG': 0.33741020727621196},\n"," 6: {'f_1': 0.044219003390830014,\n"," 'precision': 0.024736842105263175,\n"," 'recall': 0.2171052631578947,\n"," 'nDCG': 0.33791628338014396},\n"," 7: {'f_1': 0.052273679554924506,\n"," 'precision': 0.029736842105263173,\n"," 'recall': 0.22556390977443613,\n"," 'nDCG': 0.32865531690243627},\n"," 10: {'f_1': 0.07450122194129735,\n"," 'precision': 0.044473684210526346,\n"," 'recall': 0.2418859649122807,\n"," 'nDCG': 0.30763712181088515},\n"," 20: {'f_1': 0.11362801374039382,\n"," 'precision': 0.07657894736842104,\n"," 'recall': 0.24736893223735326,\n"," 'nDCG': 0.2571290703967697},\n"," 30: {'f_1': 0.1272503910699926,\n"," 'precision': 0.09421052631578944,\n"," 'recall': 0.23677951568831784,\n"," 'nDCG': 0.2282620325948478},\n"," 40: {'f_1': 0.1393862651678978,\n"," 'precision': 0.11105263157894732,\n"," 'recall': 0.2378835998238608,\n"," 'nDCG': 0.2060150933125378},\n"," 50: {'f_1': 0.14558272168537648,\n"," 'precision': 0.1228947368421052,\n"," 'recall': 0.2369303280824458,\n"," 'nDCG': 0.19024116999894183}}}"]},"execution_count":125,"metadata":{},"output_type":"execute_result"}],"source":["scores_lam"]},{"cell_type":"code","execution_count":null,"metadata":{"trusted":true},"outputs":[],"source":[]},{"cell_type":"code","execution_count":95,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T21:00:40.206275Z","iopub.status.busy":"2024-01-03T21:00:40.205513Z","iopub.status.idle":"2024-01-03T21:00:40.229921Z","shell.execute_reply":"2024-01-03T21:00:40.227362Z","shell.execute_reply.started":"2024-01-03T21:00:40.206205Z"},"trusted":true},"outputs":[],"source":["lam_2_f1 = []\n","#lam_4_f1 = []\n","#lam_5_f1 = []\n","lam_1_2_f1 = []\n","lam_1_5_f1 = []\n","lam_2_5_f1 = []\n","lam_3_f1 = []\n","#lam_10_f1 = []\n","\n","lam_2_nDCG = []\n","#lam_4_nDCG = []\n","#lam_5_nDCG = []\n","lam_1_2_nDCG = []\n","lam_1_5_nDCG = []\n","lam_2_5_nDCG = []\n","lam_3_nDCG = []\n","#lam_10_nDCG = []\n","\n","for k in k_values:\n"," lam_2_f1.append(scores_lam[2][k]['f_1'])\n"," # lam_4_f1.append(scores_lam[4][k]['f_1'])\n"," # lam_5_f1.append(scores_lam[5][k]['f_1'])\n"," lam_1_2_f1.append(scores_lam[1.2][k]['f_1'])\n"," lam_1_5_f1.append(scores_lam[1.5][k]['f_1'])\n"," lam_2_5_f1.append(scores_lam[2.5][k]['f_1'])\n"," lam_3_f1.append(scores_lam[3][k]['f_1'])\n"," # lam_10_f1.append(scores_lam[10][k]['f_1'])\n"," \n"," lam_2_nDCG.append(scores_lam[2][k]['nDCG'])\n"," # lam_4_nDCG.append(scores_lam[4][k]['nDCG'])\n"," # lam_5_nDCG.append(scores_lam[5][k]['nDCG'])\n"," lam_1_2_nDCG.append(scores_lam[1.2][k]['nDCG'])\n"," lam_1_5_nDCG.append(scores_lam[1.5][k]['nDCG'])\n"," lam_2_5_nDCG.append(scores_lam[2.5][k]['nDCG'])\n"," lam_3_nDCG.append(scores_lam[3][k]['nDCG'])\n"," # lam_10_nDCG.append(scores_lam[10][k]['nDCG'])"]},{"cell_type":"code","execution_count":102,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T21:02:40.020032Z","iopub.status.busy":"2024-01-03T21:02:40.019397Z","iopub.status.idle":"2024-01-03T21:02:40.336791Z","shell.execute_reply":"2024-01-03T21:02:40.335971Z","shell.execute_reply.started":"2024-01-03T21:02:40.019976Z"},"trusted":true},"outputs":[{"data":{"image/png":"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","text/plain":["<Figure size 1440x864 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["plt.figure(figsize=(20, 12))\n","#plt.plot(k_values, lam_2_f1, marker='o', label='lam=2')\n","#plt.plot(k_values, lam_4_f1, marker='s', label='lam=4')\n","#plt.plot(k_values, lam_5_f1, marker='^', label='lam=5')\n","plt.plot(k_values, lam_1_2_f1, marker='*', label='lam=1.2')\n","#plt.plot(k_values, lam_1_5_f1, marker='D', label='lam=1.5')\n","#plt.plot(k_values, lam_2_5_f1, marker='h', label='lam=2.5')\n","plt.plot(k_values, lam_3_f1, marker='.', label='lam=3')\n","#plt.plot(k_values, lam_10_f1, marker='x', label='lam=5')\n","\n","\n","plt.title('F1 Score by k values for lam values BERTopic')\n","plt.xlabel('k value')\n","plt.ylabel('F1 Score')\n","plt.legend()\n","plt.grid(True)\n","plt.show()"]},{"cell_type":"code","execution_count":98,"metadata":{"execution":{"iopub.execute_input":"2024-01-03T21:01:22.782430Z","iopub.status.busy":"2024-01-03T21:01:22.782006Z","iopub.status.idle":"2024-01-03T21:01:23.036331Z","shell.execute_reply":"2024-01-03T21:01:23.034616Z","shell.execute_reply.started":"2024-01-03T21:01:22.782393Z"},"trusted":true},"outputs":[{"data":{"image/png":"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","text/plain":["<Figure size 720x432 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["plt.figure(figsize=(10, 6))\n","plt.plot(k_values, lam_2_nDCG, marker='o', label='lam=2')\n","#plt.plot(k_values, lam_4_nDCG, marker='s', label='lam=4')\n","#plt.plot(k_values, lam_5_nDCG, marker='^', label='lam=5')\n","plt.plot(k_values, lam_1_2_nDCG, marker='*', label='lam=5')\n","plt.plot(k_values, lam_1_5_nDCG, marker='D', label='lam=5')\n","plt.plot(k_values, lam_2_5_nDCG, marker='h', label='lam=5')\n","plt.plot(k_values, lam_3_nDCG, marker='.', label='lam=5')\n","#plt.plot(k_values, lam_10_nDCG, marker='x', label='lam=5')\n","\n","plt.title('nDCG Score by k values for lam values BERTopic')\n","plt.xlabel('k value')\n","plt.ylabel('nDCG Score')\n","plt.legend()\n","plt.grid(True)\n","plt.show()"]}],"metadata":{"kaggle":{"accelerator":"none","dataSources":[{"datasetId":6763,"sourceId":9801,"sourceType":"datasetVersion"},{"datasetId":576263,"sourceId":1043323,"sourceType":"datasetVersion"},{"datasetId":4135603,"sourceId":7160356,"sourceType":"datasetVersion"},{"datasetId":4137237,"sourceId":7162602,"sourceType":"datasetVersion"}],"dockerImageVersionId":30120,"isGpuEnabled":false,"isInternetEnabled":true,"language":"python","sourceType":"notebook"},"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.7.10"}},"nbformat":4,"nbformat_minor":4}