diff --git a/Warehouse/Classification of clothing through images/notebook.ipynb b/Warehouse/Classification of clothing through images/notebook.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..8d8fff4e5c2fd852024b6c5f34ff4ad28d0aac34
--- /dev/null
+++ b/Warehouse/Classification of clothing through images/notebook.ipynb
@@ -0,0 +1,1835 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# 1. Business Understanding"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Many online mail order companies have a high return rate\n",
+ "(of up to 50%), with 97% of all returned products being able to be restocked\n",
+ "and can be sold. In order to resell the goods\n",
+ "goods, they must be identified, labeled, and restocked accordingly.\n",
+ "again.\n",
+ "Assuming that in 2020 185.5 million orders (Statista, 2021) with\n",
+ "6 items each (acceptance) would be received, then a return rate of 50% would mean that\n",
+ "of 50%, 556.5 million items would have to be re-identified and re-categorized.\n",
+ "To support this process and to facilitate identification of the\n",
+ "returned garments, image recognition software is to be developed that will\n",
+ "the associated categories of the individual garments on the basis of images.\n",
+ "of the individual garments on the basis of images."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# 2. Read Data"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## 2.1. Import of Relavant Modules "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Requirement already satisfied: tensorflow-datasets in c:\\users\\ar\\anaconda3\\lib\\site-packages (4.9.6)\n",
+ "Requirement already satisfied: absl-py in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (2.1.0)\n",
+ "Requirement already satisfied: click in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (8.0.4)\n",
+ "Requirement already satisfied: dm-tree in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (0.1.8)\n",
+ "Requirement already satisfied: immutabledict in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (4.2.0)\n",
+ "Requirement already satisfied: numpy in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (1.24.3)\n",
+ "Requirement already satisfied: promise in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (2.3)\n",
+ "Requirement already satisfied: protobuf>=3.20 in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (4.25.3)\n",
+ "Requirement already satisfied: psutil in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (5.9.0)\n",
+ "Requirement already satisfied: pyarrow in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (11.0.0)\n",
+ "Requirement already satisfied: requests>=2.19.0 in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (2.31.0)\n",
+ "Requirement already satisfied: simple-parsing in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (0.1.5)\n",
+ "Requirement already satisfied: tensorflow-metadata in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (1.15.0)\n",
+ "Requirement already satisfied: termcolor in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (2.4.0)\n",
+ "Requirement already satisfied: toml in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (0.10.2)\n",
+ "Requirement already satisfied: tqdm in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (4.65.0)\n",
+ "Requirement already satisfied: wrapt in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-datasets) (1.14.1)\n",
+ "Requirement already satisfied: etils>=1.9.1 in c:\\users\\ar\\anaconda3\\lib\\site-packages (from etils[enp,epath,epy,etree]>=1.9.1; python_version >= \"3.11\"->tensorflow-datasets) (1.9.2)\n",
+ "Requirement already satisfied: fsspec in c:\\users\\ar\\anaconda3\\lib\\site-packages (from etils[enp,epath,epy,etree]>=1.9.1; python_version >= \"3.11\"->tensorflow-datasets) (2023.4.0)\n",
+ "Requirement already satisfied: importlib_resources in c:\\users\\ar\\anaconda3\\lib\\site-packages (from etils[enp,epath,epy,etree]>=1.9.1; python_version >= \"3.11\"->tensorflow-datasets) (6.4.0)\n",
+ "Requirement already satisfied: typing_extensions in c:\\users\\ar\\anaconda3\\lib\\site-packages (from etils[enp,epath,epy,etree]>=1.9.1; python_version >= \"3.11\"->tensorflow-datasets) (4.7.1)\n",
+ "Requirement already satisfied: zipp in c:\\users\\ar\\anaconda3\\lib\\site-packages (from etils[enp,epath,epy,etree]>=1.9.1; python_version >= \"3.11\"->tensorflow-datasets) (3.11.0)\n",
+ "Requirement already satisfied: charset-normalizer<4,>=2 in c:\\users\\ar\\anaconda3\\lib\\site-packages (from requests>=2.19.0->tensorflow-datasets) (2.0.4)\n",
+ "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\ar\\anaconda3\\lib\\site-packages (from requests>=2.19.0->tensorflow-datasets) (3.4)\n",
+ "Requirement already satisfied: urllib3<3,>=1.21.1 in c:\\users\\ar\\anaconda3\\lib\\site-packages (from requests>=2.19.0->tensorflow-datasets) (1.26.16)\n",
+ "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\ar\\anaconda3\\lib\\site-packages (from requests>=2.19.0->tensorflow-datasets) (2024.2.2)\n",
+ "Requirement already satisfied: colorama in c:\\users\\ar\\anaconda3\\lib\\site-packages (from click->tensorflow-datasets) (0.4.6)\n",
+ "Requirement already satisfied: six in c:\\users\\ar\\anaconda3\\lib\\site-packages (from promise->tensorflow-datasets) (1.16.0)\n",
+ "Requirement already satisfied: docstring-parser~=0.15 in c:\\users\\ar\\anaconda3\\lib\\site-packages (from simple-parsing->tensorflow-datasets) (0.16)\n",
+ "Requirement already satisfied: googleapis-common-protos<2,>=1.56.4 in c:\\users\\ar\\anaconda3\\lib\\site-packages (from tensorflow-metadata->tensorflow-datasets) (1.63.1)\n",
+ "Note: you may need to restart the kernel to use updated packages.\n"
+ ]
+ }
+ ],
+ "source": [
+ "pip install tensorflow-datasets"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import numpy as np\n",
+ "import pandas as pd\n",
+ "import matplotlib.pyplot as plt\n",
+ "import tensorflow as tf\n",
+ "\n",
+ "import tensorflow_datasets as tfds\n",
+ "from tensorflow.keras.models import Sequential\n",
+ "from tensorflow.keras.layers import Flatten, Dense, Conv2D, MaxPooling2D, Dropout, Layer\n",
+ "from tensorflow.keras.utils import to_categorical, plot_model\n",
+ "from sklearn.model_selection import train_test_split\n",
+ "from sklearn.metrics import classification_report"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "'2.16.1'"
+ ]
+ },
+ "execution_count": 3,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# Should be 2.5.0\n",
+ "tf.__version__"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## 2.2 Read Data\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "The training- and test-data is already labeled and split up into two datasets"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "csv_file_train = \"https://storage.googleapis.com/ml-service-repository-datastorage/Classification_of_clothing_through_images_fashion-mnist_train.csv\"\n",
+ "csv_file_test = \"https://storage.googleapis.com/ml-service-repository-datastorage/Classification_of_clothing_through_images_fashion-mnist_test.csv\"\n",
+ "df_train = pd.read_csv(csv_file_train) \n",
+ "df_test = pd.read_csv(csv_file_test)\n",
+ "df_train.to_csv('df_train.csv', index=False)\n",
+ "df_test.to_csv('df_test.csv', index=False)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## 2.3. Data Analysis"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "<div>\n",
+ "<style scoped>\n",
+ " .dataframe tbody tr th:only-of-type {\n",
+ " vertical-align: middle;\n",
+ " }\n",
+ "\n",
+ " .dataframe tbody tr th {\n",
+ " vertical-align: top;\n",
+ " }\n",
+ "\n",
+ " .dataframe thead th {\n",
+ " text-align: right;\n",
+ " }\n",
+ "</style>\n",
+ "<table border=\"1\" class=\"dataframe\">\n",
+ " <thead>\n",
+ " <tr style=\"text-align: right;\">\n",
+ " <th></th>\n",
+ " <th>label</th>\n",
+ " <th>pixel1</th>\n",
+ " <th>pixel2</th>\n",
+ " <th>pixel3</th>\n",
+ " <th>pixel4</th>\n",
+ " <th>pixel5</th>\n",
+ " <th>pixel6</th>\n",
+ " <th>pixel7</th>\n",
+ " <th>pixel8</th>\n",
+ " <th>pixel9</th>\n",
+ " <th>...</th>\n",
+ " <th>pixel775</th>\n",
+ " <th>pixel776</th>\n",
+ " <th>pixel777</th>\n",
+ " <th>pixel778</th>\n",
+ " <th>pixel779</th>\n",
+ " <th>pixel780</th>\n",
+ " <th>pixel781</th>\n",
+ " <th>pixel782</th>\n",
+ " <th>pixel783</th>\n",
+ " <th>pixel784</th>\n",
+ " </tr>\n",
+ " </thead>\n",
+ " <tbody>\n",
+ " <tr>\n",
+ " <th>0</th>\n",
+ " <td>2</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>...</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>1</th>\n",
+ " <td>9</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>...</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>2</th>\n",
+ " <td>6</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>5</td>\n",
+ " <td>0</td>\n",
+ " <td>...</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>30</td>\n",
+ " <td>43</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>3</th>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>1</td>\n",
+ " <td>2</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>...</td>\n",
+ " <td>3</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>1</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>4</th>\n",
+ " <td>3</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>...</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " <td>0</td>\n",
+ " </tr>\n",
+ " </tbody>\n",
+ "</table>\n",
+ "<p>5 rows × 785 columns</p>\n",
+ "</div>"
+ ],
+ "text/plain": [
+ " label pixel1 pixel2 pixel3 pixel4 pixel5 pixel6 pixel7 pixel8 \\\n",
+ "0 2 0 0 0 0 0 0 0 0 \n",
+ "1 9 0 0 0 0 0 0 0 0 \n",
+ "2 6 0 0 0 0 0 0 0 5 \n",
+ "3 0 0 0 0 1 2 0 0 0 \n",
+ "4 3 0 0 0 0 0 0 0 0 \n",
+ "\n",
+ " pixel9 ... pixel775 pixel776 pixel777 pixel778 pixel779 pixel780 \\\n",
+ "0 0 ... 0 0 0 0 0 0 \n",
+ "1 0 ... 0 0 0 0 0 0 \n",
+ "2 0 ... 0 0 0 30 43 0 \n",
+ "3 0 ... 3 0 0 0 0 1 \n",
+ "4 0 ... 0 0 0 0 0 0 \n",
+ "\n",
+ " pixel781 pixel782 pixel783 pixel784 \n",
+ "0 0 0 0 0 \n",
+ "1 0 0 0 0 \n",
+ "2 0 0 0 0 \n",
+ "3 0 0 0 0 \n",
+ "4 0 0 0 0 \n",
+ "\n",
+ "[5 rows x 785 columns]"
+ ]
+ },
+ "execution_count": 5,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "df_train.head()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Describe the dataframe, not really helpful in this case, but is shows that the data is not corrupted by evaluating:\n",
+ "- Label must be between 0 and 9\n",
+ "- Pixel values must be between 0 and 255 (non-negative)\n",
+ "- Count must be 60000 (train) 10000 (test)\n",
+ "- Maximum number of pixels must be 784 for all rows"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "<div>\n",
+ "<style scoped>\n",
+ " .dataframe tbody tr th:only-of-type {\n",
+ " vertical-align: middle;\n",
+ " }\n",
+ "\n",
+ " .dataframe tbody tr th {\n",
+ " vertical-align: top;\n",
+ " }\n",
+ "\n",
+ " .dataframe thead th {\n",
+ " text-align: right;\n",
+ " }\n",
+ "</style>\n",
+ "<table border=\"1\" class=\"dataframe\">\n",
+ " <thead>\n",
+ " <tr style=\"text-align: right;\">\n",
+ " <th></th>\n",
+ " <th>label</th>\n",
+ " <th>pixel1</th>\n",
+ " <th>pixel2</th>\n",
+ " <th>pixel3</th>\n",
+ " <th>pixel4</th>\n",
+ " <th>pixel5</th>\n",
+ " <th>pixel6</th>\n",
+ " <th>pixel7</th>\n",
+ " <th>pixel8</th>\n",
+ " <th>pixel9</th>\n",
+ " <th>...</th>\n",
+ " <th>pixel775</th>\n",
+ " <th>pixel776</th>\n",
+ " <th>pixel777</th>\n",
+ " <th>pixel778</th>\n",
+ " <th>pixel779</th>\n",
+ " <th>pixel780</th>\n",
+ " <th>pixel781</th>\n",
+ " <th>pixel782</th>\n",
+ " <th>pixel783</th>\n",
+ " <th>pixel784</th>\n",
+ " </tr>\n",
+ " </thead>\n",
+ " <tbody>\n",
+ " <tr>\n",
+ " <th>count</th>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>...</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.000000</td>\n",
+ " <td>60000.00000</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>mean</th>\n",
+ " <td>4.500000</td>\n",
+ " <td>0.000900</td>\n",
+ " <td>0.006150</td>\n",
+ " <td>0.035333</td>\n",
+ " <td>0.101933</td>\n",
+ " <td>0.247967</td>\n",
+ " <td>0.411467</td>\n",
+ " <td>0.805767</td>\n",
+ " <td>2.198283</td>\n",
+ " <td>5.682000</td>\n",
+ " <td>...</td>\n",
+ " <td>34.625400</td>\n",
+ " <td>23.300683</td>\n",
+ " <td>16.588267</td>\n",
+ " <td>17.869433</td>\n",
+ " <td>22.814817</td>\n",
+ " <td>17.911483</td>\n",
+ " <td>8.520633</td>\n",
+ " <td>2.753300</td>\n",
+ " <td>0.855517</td>\n",
+ " <td>0.07025</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>std</th>\n",
+ " <td>2.872305</td>\n",
+ " <td>0.094689</td>\n",
+ " <td>0.271011</td>\n",
+ " <td>1.222324</td>\n",
+ " <td>2.452871</td>\n",
+ " <td>4.306912</td>\n",
+ " <td>5.836188</td>\n",
+ " <td>8.215169</td>\n",
+ " <td>14.093378</td>\n",
+ " <td>23.819481</td>\n",
+ " <td>...</td>\n",
+ " <td>57.545242</td>\n",
+ " <td>48.854427</td>\n",
+ " <td>41.979611</td>\n",
+ " <td>43.966032</td>\n",
+ " <td>51.830477</td>\n",
+ " <td>45.149388</td>\n",
+ " <td>29.614859</td>\n",
+ " <td>17.397652</td>\n",
+ " <td>9.356960</td>\n",
+ " <td>2.12587</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>min</th>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>...</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.00000</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>25%</th>\n",
+ " <td>2.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>...</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.00000</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>50%</th>\n",
+ " <td>4.500000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>...</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.00000</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>75%</th>\n",
+ " <td>7.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>...</td>\n",
+ " <td>58.000000</td>\n",
+ " <td>9.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.00000</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>max</th>\n",
+ " <td>9.000000</td>\n",
+ " <td>16.000000</td>\n",
+ " <td>36.000000</td>\n",
+ " <td>226.000000</td>\n",
+ " <td>164.000000</td>\n",
+ " <td>227.000000</td>\n",
+ " <td>230.000000</td>\n",
+ " <td>224.000000</td>\n",
+ " <td>255.000000</td>\n",
+ " <td>254.000000</td>\n",
+ " <td>...</td>\n",
+ " <td>255.000000</td>\n",
+ " <td>255.000000</td>\n",
+ " <td>255.000000</td>\n",
+ " <td>255.000000</td>\n",
+ " <td>255.000000</td>\n",
+ " <td>255.000000</td>\n",
+ " <td>255.000000</td>\n",
+ " <td>255.000000</td>\n",
+ " <td>255.000000</td>\n",
+ " <td>170.00000</td>\n",
+ " </tr>\n",
+ " </tbody>\n",
+ "</table>\n",
+ "<p>8 rows × 785 columns</p>\n",
+ "</div>"
+ ],
+ "text/plain": [
+ " label pixel1 pixel2 pixel3 pixel4 \\\n",
+ "count 60000.000000 60000.000000 60000.000000 60000.000000 60000.000000 \n",
+ "mean 4.500000 0.000900 0.006150 0.035333 0.101933 \n",
+ "std 2.872305 0.094689 0.271011 1.222324 2.452871 \n",
+ "min 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "25% 2.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "50% 4.500000 0.000000 0.000000 0.000000 0.000000 \n",
+ "75% 7.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "max 9.000000 16.000000 36.000000 226.000000 164.000000 \n",
+ "\n",
+ " pixel5 pixel6 pixel7 pixel8 pixel9 \\\n",
+ "count 60000.000000 60000.000000 60000.000000 60000.000000 60000.000000 \n",
+ "mean 0.247967 0.411467 0.805767 2.198283 5.682000 \n",
+ "std 4.306912 5.836188 8.215169 14.093378 23.819481 \n",
+ "min 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "25% 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "50% 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "75% 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "max 227.000000 230.000000 224.000000 255.000000 254.000000 \n",
+ "\n",
+ " ... pixel775 pixel776 pixel777 pixel778 \\\n",
+ "count ... 60000.000000 60000.000000 60000.000000 60000.000000 \n",
+ "mean ... 34.625400 23.300683 16.588267 17.869433 \n",
+ "std ... 57.545242 48.854427 41.979611 43.966032 \n",
+ "min ... 0.000000 0.000000 0.000000 0.000000 \n",
+ "25% ... 0.000000 0.000000 0.000000 0.000000 \n",
+ "50% ... 0.000000 0.000000 0.000000 0.000000 \n",
+ "75% ... 58.000000 9.000000 0.000000 0.000000 \n",
+ "max ... 255.000000 255.000000 255.000000 255.000000 \n",
+ "\n",
+ " pixel779 pixel780 pixel781 pixel782 pixel783 \\\n",
+ "count 60000.000000 60000.000000 60000.000000 60000.000000 60000.000000 \n",
+ "mean 22.814817 17.911483 8.520633 2.753300 0.855517 \n",
+ "std 51.830477 45.149388 29.614859 17.397652 9.356960 \n",
+ "min 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "25% 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "50% 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "75% 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "max 255.000000 255.000000 255.000000 255.000000 255.000000 \n",
+ "\n",
+ " pixel784 \n",
+ "count 60000.00000 \n",
+ "mean 0.07025 \n",
+ "std 2.12587 \n",
+ "min 0.00000 \n",
+ "25% 0.00000 \n",
+ "50% 0.00000 \n",
+ "75% 0.00000 \n",
+ "max 170.00000 \n",
+ "\n",
+ "[8 rows x 785 columns]"
+ ]
+ },
+ "execution_count": 6,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "df_train.describe()"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "<div>\n",
+ "<style scoped>\n",
+ " .dataframe tbody tr th:only-of-type {\n",
+ " vertical-align: middle;\n",
+ " }\n",
+ "\n",
+ " .dataframe tbody tr th {\n",
+ " vertical-align: top;\n",
+ " }\n",
+ "\n",
+ " .dataframe thead th {\n",
+ " text-align: right;\n",
+ " }\n",
+ "</style>\n",
+ "<table border=\"1\" class=\"dataframe\">\n",
+ " <thead>\n",
+ " <tr style=\"text-align: right;\">\n",
+ " <th></th>\n",
+ " <th>label</th>\n",
+ " <th>pixel1</th>\n",
+ " <th>pixel2</th>\n",
+ " <th>pixel3</th>\n",
+ " <th>pixel4</th>\n",
+ " <th>pixel5</th>\n",
+ " <th>pixel6</th>\n",
+ " <th>pixel7</th>\n",
+ " <th>pixel8</th>\n",
+ " <th>pixel9</th>\n",
+ " <th>...</th>\n",
+ " <th>pixel775</th>\n",
+ " <th>pixel776</th>\n",
+ " <th>pixel777</th>\n",
+ " <th>pixel778</th>\n",
+ " <th>pixel779</th>\n",
+ " <th>pixel780</th>\n",
+ " <th>pixel781</th>\n",
+ " <th>pixel782</th>\n",
+ " <th>pixel783</th>\n",
+ " <th>pixel784</th>\n",
+ " </tr>\n",
+ " </thead>\n",
+ " <tbody>\n",
+ " <tr>\n",
+ " <th>count</th>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>...</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.000000</td>\n",
+ " <td>10000.00000</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>mean</th>\n",
+ " <td>4.500000</td>\n",
+ " <td>0.000400</td>\n",
+ " <td>0.010300</td>\n",
+ " <td>0.052100</td>\n",
+ " <td>0.077000</td>\n",
+ " <td>0.208600</td>\n",
+ " <td>0.349200</td>\n",
+ " <td>0.826700</td>\n",
+ " <td>2.321200</td>\n",
+ " <td>5.457800</td>\n",
+ " <td>...</td>\n",
+ " <td>34.320800</td>\n",
+ " <td>23.071900</td>\n",
+ " <td>16.432000</td>\n",
+ " <td>17.870600</td>\n",
+ " <td>22.860000</td>\n",
+ " <td>17.790200</td>\n",
+ " <td>8.353500</td>\n",
+ " <td>2.541600</td>\n",
+ " <td>0.629500</td>\n",
+ " <td>0.06560</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>std</th>\n",
+ " <td>2.872425</td>\n",
+ " <td>0.024493</td>\n",
+ " <td>0.525187</td>\n",
+ " <td>2.494315</td>\n",
+ " <td>2.208882</td>\n",
+ " <td>4.669183</td>\n",
+ " <td>5.657849</td>\n",
+ " <td>8.591731</td>\n",
+ " <td>15.031508</td>\n",
+ " <td>23.359019</td>\n",
+ " <td>...</td>\n",
+ " <td>57.888679</td>\n",
+ " <td>49.049749</td>\n",
+ " <td>42.159665</td>\n",
+ " <td>44.140552</td>\n",
+ " <td>51.706601</td>\n",
+ " <td>45.128107</td>\n",
+ " <td>28.765769</td>\n",
+ " <td>16.417363</td>\n",
+ " <td>7.462533</td>\n",
+ " <td>1.93403</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>min</th>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>...</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.00000</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>25%</th>\n",
+ " <td>2.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>...</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.00000</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>50%</th>\n",
+ " <td>4.500000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>...</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.00000</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>75%</th>\n",
+ " <td>7.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>...</td>\n",
+ " <td>55.000000</td>\n",
+ " <td>6.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>1.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.000000</td>\n",
+ " <td>0.00000</td>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th>max</th>\n",
+ " <td>9.000000</td>\n",
+ " <td>2.000000</td>\n",
+ " <td>45.000000</td>\n",
+ " <td>218.000000</td>\n",
+ " <td>185.000000</td>\n",
+ " <td>227.000000</td>\n",
+ " <td>223.000000</td>\n",
+ " <td>247.000000</td>\n",
+ " <td>218.000000</td>\n",
+ " <td>244.000000</td>\n",
+ " <td>...</td>\n",
+ " <td>254.000000</td>\n",
+ " <td>252.000000</td>\n",
+ " <td>255.000000</td>\n",
+ " <td>255.000000</td>\n",
+ " <td>255.000000</td>\n",
+ " <td>255.000000</td>\n",
+ " <td>240.000000</td>\n",
+ " <td>225.000000</td>\n",
+ " <td>205.000000</td>\n",
+ " <td>107.00000</td>\n",
+ " </tr>\n",
+ " </tbody>\n",
+ "</table>\n",
+ "<p>8 rows × 785 columns</p>\n",
+ "</div>"
+ ],
+ "text/plain": [
+ " label pixel1 pixel2 pixel3 pixel4 \\\n",
+ "count 10000.000000 10000.000000 10000.000000 10000.000000 10000.000000 \n",
+ "mean 4.500000 0.000400 0.010300 0.052100 0.077000 \n",
+ "std 2.872425 0.024493 0.525187 2.494315 2.208882 \n",
+ "min 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "25% 2.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "50% 4.500000 0.000000 0.000000 0.000000 0.000000 \n",
+ "75% 7.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "max 9.000000 2.000000 45.000000 218.000000 185.000000 \n",
+ "\n",
+ " pixel5 pixel6 pixel7 pixel8 pixel9 \\\n",
+ "count 10000.000000 10000.000000 10000.000000 10000.000000 10000.000000 \n",
+ "mean 0.208600 0.349200 0.826700 2.321200 5.457800 \n",
+ "std 4.669183 5.657849 8.591731 15.031508 23.359019 \n",
+ "min 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "25% 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "50% 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "75% 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "max 227.000000 223.000000 247.000000 218.000000 244.000000 \n",
+ "\n",
+ " ... pixel775 pixel776 pixel777 pixel778 \\\n",
+ "count ... 10000.000000 10000.000000 10000.000000 10000.000000 \n",
+ "mean ... 34.320800 23.071900 16.432000 17.870600 \n",
+ "std ... 57.888679 49.049749 42.159665 44.140552 \n",
+ "min ... 0.000000 0.000000 0.000000 0.000000 \n",
+ "25% ... 0.000000 0.000000 0.000000 0.000000 \n",
+ "50% ... 0.000000 0.000000 0.000000 0.000000 \n",
+ "75% ... 55.000000 6.000000 0.000000 0.000000 \n",
+ "max ... 254.000000 252.000000 255.000000 255.000000 \n",
+ "\n",
+ " pixel779 pixel780 pixel781 pixel782 pixel783 \\\n",
+ "count 10000.000000 10000.000000 10000.000000 10000.000000 10000.000000 \n",
+ "mean 22.860000 17.790200 8.353500 2.541600 0.629500 \n",
+ "std 51.706601 45.128107 28.765769 16.417363 7.462533 \n",
+ "min 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "25% 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "50% 0.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "75% 1.000000 0.000000 0.000000 0.000000 0.000000 \n",
+ "max 255.000000 255.000000 240.000000 225.000000 205.000000 \n",
+ "\n",
+ " pixel784 \n",
+ "count 10000.00000 \n",
+ "mean 0.06560 \n",
+ "std 1.93403 \n",
+ "min 0.00000 \n",
+ "25% 0.00000 \n",
+ "50% 0.00000 \n",
+ "75% 0.00000 \n",
+ "max 107.00000 \n",
+ "\n",
+ "[8 rows x 785 columns]"
+ ]
+ },
+ "execution_count": 7,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "df_test.describe()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Both test data and train data seems to be valid and uncorrupted"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Define human readable names for the 10 categories"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "class_names = ['Top','Trouser','Pullover','Dress','Coat',\n",
+ " 'Sandal','Shirt','Sneaker','Bag','Ankle boot']"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Split the dataset an check the distribution of each class (split could be also done later)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "54000 train examples\n",
+ "6000 validation examples\n",
+ "10000 test examples\n"
+ ]
+ }
+ ],
+ "source": [
+ "df_train, df_val = train_test_split(df_train, test_size=0.1, random_state=365)\n",
+ "print(f\"{len(df_train)} train examples\")\n",
+ "print(f\"{len(df_val)} validation examples\")\n",
+ "print(f\"{len(df_test)} test examples\")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Show the distribution for each set"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\n",
+ "TRAIN DISTRIBUTION\n",
+ "\n",
+ "Sandal : 5429 or 10.053703703703704%\n",
+ "Coat : 5421 or 10.03888888888889%\n",
+ "Pullover : 5407 or 10.012962962962963%\n",
+ "Dress : 5405 or 10.00925925925926%\n",
+ "Ankle boot : 5404 or 10.007407407407408%\n",
+ "Shirt : 5397 or 9.994444444444445%\n",
+ "Top : 5396 or 9.992592592592594%\n",
+ "Sneaker : 5395 or 9.99074074074074%\n",
+ "Trouser : 5384 or 9.97037037037037%\n",
+ "Bag : 5362 or 9.92962962962963%\n",
+ "\n",
+ "VALIDATION DISTRIBUTION\n",
+ "\n",
+ "Bag : 638 or 10.633333333333333%\n",
+ "Trouser : 616 or 10.266666666666667%\n",
+ "Sneaker : 605 or 10.083333333333332%\n",
+ "Top : 604 or 10.066666666666666%\n",
+ "Shirt : 603 or 10.05%\n",
+ "Ankle boot : 596 or 9.933333333333334%\n",
+ "Dress : 595 or 9.916666666666666%\n",
+ "Pullover : 593 or 9.883333333333333%\n",
+ "Coat : 579 or 9.65%\n",
+ "Sandal : 571 or 9.516666666666666%\n",
+ "\n",
+ "TEST DISTRIBUTION\n",
+ "\n",
+ "Top : 1000 or 10.0%\n",
+ "Trouser : 1000 or 10.0%\n",
+ "Pullover : 1000 or 10.0%\n",
+ "Dress : 1000 or 10.0%\n",
+ "Bag : 1000 or 10.0%\n",
+ "Shirt : 1000 or 10.0%\n",
+ "Sandal : 1000 or 10.0%\n",
+ "Coat : 1000 or 10.0%\n",
+ "Sneaker : 1000 or 10.0%\n",
+ "Ankle boot : 1000 or 10.0%\n"
+ ]
+ }
+ ],
+ "source": [
+ "def get_classes_distribution(data):\n",
+ " # Get the count for each label\n",
+ " label_counts = data[\"label\"].value_counts()\n",
+ "\n",
+ " # Get total number of samples\n",
+ " total_samples = len(data)\n",
+ "\n",
+ "\n",
+ " # Count the number of items in each class\n",
+ " for i in range(len(label_counts)):\n",
+ " label = class_names[label_counts.index[i]]\n",
+ " count = label_counts.values[i]\n",
+ " percent = (count / total_samples) * 100\n",
+ " print(\"{:<20s}: {} or {}%\".format(label, count, percent))\n",
+ "\n",
+ "print(\"\\nTRAIN DISTRIBUTION\\n\")\n",
+ "get_classes_distribution(df_train)\n",
+ "print(\"\\nVALIDATION DISTRIBUTION\\n\")\n",
+ "get_classes_distribution(df_val)\n",
+ "print(\"\\nTEST DISTRIBUTION\\n\")\n",
+ "get_classes_distribution(df_test)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "This is already helpful, we can see that it is quite evenly split.\n",
+ "Print the data as a pie chart, to make it even nicer."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ "<Figure size 1000x700 with 1 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk4AAAJDCAYAAAD9x5srAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAADfSElEQVR4nOzdd3hTZfvA8W9Wk+69KS17yRRoEZUhiANQEVFQEMWFe4/39zpwi76KEyfi3rgVF6hMWbJXge69Z3bO749CobRAgSYnbe/PdfWCnJzzPPfpSO48U6MoioIQQgghhDgmrdoBCCGEEEK0FpI4CSGEEEI0kyROQgghhBDNJImTEEIIIUQzSeIkhBBCCNFMkjgJIYQQQjSTJE5CCCGEEM0kiZMQQgghRDNJ4iSEEEII0UySOAmvdNFFF+Hr60t5efkRz7n88ssxGAwUFBQ0u1yNRsMjjzxS//jPP/9Eo9Hw559/HvPamTNnkpSU1Oy6DvXaa6+xcOHCRsfT09PRaDRNPqcWd9ynu4wcOZKRI0fWPz7R7+fHH3/MvHnzjuuapup65JFH0Gg0FBcXH1dZR7N9+3YeeeQR0tPTGz13Mj8rIcSJkcRJeKVZs2ZhsVj4+OOPm3y+oqKCr7/+mvHjxxMdHX3C9QwaNIhVq1YxaNCgEy6jOY6UUMTGxrJq1SrOP/98t9bvKZ5OnA53ot/PE0mcPPWz2759O3PmzGkycXrwwQf5+uuv3Vq/EKIhvdoBCNGUc889l7i4OBYsWMCNN97Y6PlPPvkEs9nMrFmzTqqeoKAgUlJSTqqMk2E0GlWtv63xxPfT6XTicDi84mfXpUsXVesXoj2SFifhlXQ6HVdeeSXr169ny5YtjZ5/9913iY2N5dxzz6WoqIgbb7yR3r17ExAQQFRUFKNHj2bZsmXHrOdIXXULFy6kR48eGI1GevXqxfvvv9/k9XPmzCE5OZmwsDCCgoIYNGgQ77zzDofunZ2UlMS2bdv466+/0Gg0aDSa+u6VI3UtLV++nLPOOovAwED8/Pw47bTT+PHHHxvFqNFoWLp0KbNnzyYiIoLw8HAmTZpEbm7uMe/dk/dpsVi46667GDBgAMHBwYSFhTFs2DC+/fbbZsWpKApz584lMTERk8nEoEGD+Pnnnxud19T3s6ioiOuuu46EhASMRiORkZEMHz6c33//Hajr7vvxxx/JyMioj1uj0TQob+7cuTz++ON06tQJo9HI0qVLj9otmJWVxaRJkwgKCiI4OJgrrriCoqKiBucc3m186Pdx5syZQN3P55JLLgFg1KhR9bEdqLOprjqLxcIDDzxAp06d8PHxIT4+nptuuqlRt3dSUhLjx49n8eLFDBo0CF9fX3r27MmCBQuO8FMQQoC0OAkvdvXVV/P000+zYMECXnjhhfrj27dvZ82aNdx///3odDpKS0sBePjhh4mJiaG6upqvv/6akSNH8scffzQYA9McCxcu5KqrruKCCy7gf//7HxUVFTzyyCNYrVa02oafNdLT07n++uvp2LEjAKtXr+aWW24hJyeHhx56CICvv/6ayZMnExwczGuvvQbUtYwcyV9//cXYsWPp168f77zzDkajkddee40JEybwySefcOmllzY4/5prruH888/n448/Jisri3vuuYcrrriCJUuWeM19Wq1WSktLufvuu4mPj8dms/H7778zadIk3n33XWbMmHHUWOfMmcOcOXOYNWsWkydPJisri2uvvRan00mPHj2Oeu306dPZsGEDTzzxBN27d6e8vJwNGzZQUlIC1HUvXnfddezdu/eI3V4vvfQS3bt357nnniMoKIhu3bodtc6LLrqIKVOmcMMNN7Bt2zYefPBBtm/fzj///IPBYDjqtYc6//zzefLJJ/nPf/7Dq6++Wt+lfKSWJkVRuPDCC/njjz944IEHOOOMM9i8eTMPP/wwq1atYtWqVQ1+9zZt2sRdd93F/fffT3R0NG+//TazZs2ia9eunHnmmc2OU4h2RRHCi40YMUKJiIhQbDZb/bG77rpLAZTdu3c3eY3D4VDsdrty1llnKRdddFGD5wDl4Ycfrn+8dOlSBVCWLl2qKIqiOJ1OJS4uThk0aJDicrnqz0tPT1cMBoOSmJh4xFidTqdit9uVRx99VAkPD29wfZ8+fZQRI0Y0uiYtLU0BlHfffbf+WEpKihIVFaVUVVU1uKdTTjlF6dChQ3257777rgIoN954Y4My586dqwBKXl7eUWP15H0e7sDPaNasWcrAgQOPem5ZWZliMpka/SxXrFihAA3qa+r7GRAQoNx+++1HreP8889v8p4PlNelS5cGv4NHquvhhx9WAOWOO+5ocO5HH32kAMqHH35Yf+zw38UDEhMTlSuvvLL+8RdffNHgd/RQV155ZYO4Fy9erADK3LlzG5z32WefKYDy5ptvNqjHZDIpGRkZ9cfMZrMSFhamXH/99Y3qEkLUka464dVmzZpFcXEx3333HQAOh4MPP/yQM844o8Gn/tdff51BgwZhMpnQ6/UYDAb++OMPduzYcVz17dq1i9zcXKZNm1bfXQOQmJjIaaed1uj8JUuWMGbMGIKDg9HpdBgMBh566CFKSkooLCw87vutqanhn3/+YfLkyQQEBNQf1+l0TJ8+nezsbHbt2tXgmokTJzZ43K9fPwAyMjK86j6/+OILhg8fTkBAQP3P6J133jnmz2jVqlVYLBYuv/zyBsdPO+00EhMTj1nv0KFDWbhwIY8//jirV6/Gbrc3K95DTZw48bhaig6PdcqUKej1epYuXXrcdR+PA62MB7r6Drjkkkvw9/fnjz/+aHB8wIAB9a2IACaTie7dux/1d0eI9k4SJ+HVDnT9vPvuuwD89NNPFBQUNBgU/vzzzzN79mySk5P56quvWL16NWvXruWcc87BbDYfV30Hum9iYmIaPXf4sTVr1nD22WcD8NZbb7FixQrWrl3L//3f/wEcd90AZWVlKIpCbGxso+fi4uIaxHhAeHh4g8cHumKOVr+n73PRokVMmTKF+Ph4PvzwQ1atWsXatWu5+uqrsVgsR732eGJtymeffcaVV17J22+/zbBhwwgLC2PGjBnk5+cf89oDmvp5HM3hcen1esLDwxv97FpaSUkJer2eyMjIBsc1Gg0xMTHH/N2But+fE/ndFaK9kDFOwqv5+voydepU3nrrLfLy8liwYAGBgYH1A2YBPvzwQ0aOHMn8+fMbXFtVVXXc9R14I2nqTfXwY59++ikGg4EffvgBk8lUf/ybb7457noPCA0NRavVkpeX1+i5AwO+IyIiTrj8Azx9nx9++CGdOnXis88+a9DCZbVaTzrWY61jFBERwbx585g3bx6ZmZl899133H///RQWFrJ48eJmxX9ozM2Rn59PfHx8/WOHw0FJSUmDRMVoNDZ5/yeTXIWHh+NwOCgqKmqQPCmKQn5+PkOGDDnhsoUQdaTFSXi9WbNm4XQ6efbZZ/npp5+47LLL8PPzq39eo9E0Gmy9efNmVq1addx19ejRg9jYWD755JMGM8YyMjJYuXJlg3M1Gg16vR6dTld/zGw288EHHzQqt7mf4v39/UlOTmbRokUNzne5XHz44Yd06NCB7t27H/d9Hc7T96nRaPDx8WmQgOTn5zdrVl1KSgomk4mPPvqowfGVK1ced5dSx44dufnmmxk7diwbNmw4Ztwn6vBYP//8cxwOR4OJCklJSWzevLnBeUuWLKG6urrBsea0IB5w1llnAXWJ6qG++uorampq6p8XQpw4SZyE1xs8eDD9+vVj3rx52O32Rms3jR8/nl9//ZWHH36YJUuWMH/+fMaNG0enTp2Ouy6tVstjjz3G+vXrueiii/jxxx/56KOPGDNmTKPul/PPP5/q6mqmTZvGb7/9xqeffsoZZ5zR5Iy5vn37smnTJj777DPWrl3b5BILBzz11FOUlJQwatQovvzyS7777jvOO+88tm7dynPPPXfcrR/ecJ/jx49n165d3HjjjSxZsoT33nuP008/vVldYKGhodx99918/fXXXHPNNfzyyy+8/fbbTJky5ZhddRUVFQwaNIjnnnuOH374gb/++ovnnnuOxYsXM3bs2AZxFxYWMn/+fNasWcO6deua8208okWLFnHvvffy22+/MW/ePK6//nr69+/PlClT6s+ZPn06P//8Mw899BB//PEHL7/8MrNnzyY4OLhBWaeccgoAb775JsuXL2fdunVHbJUaO3Ys48aN47777mPOnDn8/vvvPP/881x11VUMHDiQ6dOnn9R9CSGQWXWidXjxxRcVQOndu3ej56xWq3L33Xcr8fHxislkUgYNGqR88803jWYcKcqxZ9Ud8PbbbyvdunVTfHx8lO7duysLFixosrwFCxYoPXr0UIxGo9K5c2flqaeeUt555x0FUNLS0urPS09PV84++2wlMDBQAerLaWpmlqIoyrJly5TRo0cr/v7+iq+vr5KSkqJ8//33Dc45MKtu7dq1DY4f6Z6a4qn7VBRFefrpp5WkpCTFaDQqvXr1Ut566636WWjH4nK5lKeeekpJSEhQfHx8lH79+inff/+9MmLEiKPOqrNYLMoNN9yg9OvXTwkKClJ8fX2VHj16KA8//LBSU1NTf11paakyefJkJSQkRNFoNPUxHSjv2WefbRTT0WbVrV+/XpkwYYISEBCgBAYGKlOnTlUKCgoaXG+1WpV7771XSUhIUHx9fZURI0YoGzdubDSrTlEUZd68eUqnTp0UnU7XoM6mflZms1m57777lMTERMVgMCixsbHK7NmzlbKysgbnJSYmKueff36j+zr8eyqEaEijKIe00wshhBBCiCOSrjohhBBCiGaSxEkIIYQQopkkcRJCCCGEaCZJnIQQQgghmkkSJyGEEEKIZpLESQghhBCimSRxEkIIIYRoJkmchBBCCCGaSRInIYQQQohmksRJCCGEEKKZJHESQgghhGgmSZyEEEIIIZpJEichhBBCiGaSxEkIIYQQopkkcRJCCCGEaCZJnIQQQgghmkkSJyGEEEKIZpLESQghhBCimSRxEkIIIYRoJkmchBBCCCGaSRInIYQQQohmksRJCCGEEKKZJHESQgghhGgmSZyEEEIIIZpJEichhBBCiGaSxEkIIYQQopkkcRJCCCGEaCZJnIQQQgghmkkSJyGEEEKIZpLESQghhBCimSRxEkIIIYRoJkmchBBCCCGaSRInIYQQQohmksRJCC8xcuRIbr/99vrHSUlJzJs3T7V4hBBCNCaJkxAtZObMmWg0GjQaDQaDgc6dO3P33XdTU1OjdmhCCCFaiF7tAIRoS8455xzeffdd7HY7y5Yt45prrqGmpob58+erHdoJsdls+Pj4qB2GEEJ4DWlxEqIFGY1GYmJiSEhIYNq0aVx++eV88803zJw5kwsvvLDBubfffjsjR45sdtmZmZlccMEFBAQEEBQUxJQpUygoKABg165daDQadu7c2eCa559/nqSkJBRFAWD79u2cd955BAQEEB0dzfTp0ykuLq4/f+TIkdx8883ceeedREREMHbs2BP7RgghRBsliZMQbuTr64vdbj/pchRF4cILL6S0tJS//vqL3377jb1793LppZcC0KNHD0499VQ++uijBtd9/PHHTJs2DY1GQ15eHiNGjGDAgAGsW7eOxYsXU1BQwJQpUxpc895776HX61mxYgVvvPHGSccuhBBtiXTVCeEma9as4eOPP+ass8466bJ+//13Nm/eTFpaGgkJCQB88MEH9OnTh7Vr1zJkyBAuv/xyXnnlFR577DEAdu/ezfr163n//fcBmD9/PoMGDeLJJ5+sL3fBggUkJCSwe/duunfvDkDXrl2ZO3fuSccshBBtkbQ4CdGCfvjhBwICAjCZTAwbNowzzzyTl19++aTL3bFjBwkJCfVJE0Dv3r0JCQlhx44dAFx22WVkZGSwevVqAD766CMGDBhA7969AVi/fj1Lly4lICCg/qtnz54A7N27t77cwYMHn3S8QgjRVkmLkxAtaNSoUcyfPx+DwUBcXBwGgwEArVZbP87ogOPpwlMUBY1Gc9TjsbGxjBo1io8//piUlBQ++eQTrr/++vpzXS4XEyZM4JlnnmlUTmxsbP3//f39mx2XEEK0N5I4CdGC/P396dq1a6PjkZGRbN26tcGxjRs31idWx9K7d28yMzPJysqqb3Xavn07FRUV9OrVq/68yy+/nPvuu4+pU6eyd+9eLrvssvrnBg0axFdffUVSUhJ6vfzpCyHEiZCuOiE8YPTo0axbt47333+f1NRUHn744UaJ1NGMGTOGfv36cfnll7NhwwbWrFnDjBkzGDFiRIOutUmTJlFZWcns2bMZNWoU8fHx9c/ddNNNlJaWMnXqVNasWcO+ffv49ddfufrqq3E6nS16v0II0VZJ4iSEB4wbN44HH3yQe++9lyFDhlBVVcWMGTOafb1Go+Gbb74hNDSUM888kzFjxtC5c2c+++yzBucFBQUxYcIENm3axOWXX97gubi4OFasWIHT6WTcuHGccsop3HbbbQQHB6PVykuBEEI0h0Y5fOCFEEIIIYRoknzMFEIIIYRoJkmchBBCCCGaSRInIYQQQohmksRJCCGEEKKZJHESQgghhGgmSZyEEEIIIZpJEichhBBCiGaSxEkIIYQQopkkcRJCCCGEaCZJnIQQQgghmkkSJyGEEEKIZpLESQghhBCimSRxEkIIIYRoJkmchBBCCCGaSRInIYQQQohm0qsdgBCidbM4LNQ6arE5bei1evQaPXqtHoPOgF6jR6fVqR2iEEK0GEmchGjn7C47hbWFlFvLqbBW1H81eGyroNpWjdlhptZRS629llpHLRaHBafiPGr5GjR1CdX+L4PW0ODfA4mWXqvHpDcR4RtBpG8kkX6Rjf4N8gny0HdFCCGaplEURVE7CCGEe1kcFrKqsuq/Misz6/6tyiS/Jv+YyY+3MOnqEqsov6gj/hvjH4O/wV/tUIUQbZQkTkK0EdW2ajKrMsmsyiS7KpvMyrr/Z1VlUVRbhEL7+VOPD4inW2g3uod2r/9KDEpEq5FhnUKIkyOJkxCtUHZVNpuLNrOpaBPbS7aTUZlBmbVM7bC8mklnoktIF7qHdqdHWI/6hCrYGKx2aEKIVkQSJyG8nNlhZlvxNjYVbWJT0SY2F22mxFKidlhtRpRfVIOWqe6h3ekc3FkGtQshmiSJkxBeJqsyi41FG+uTpNSyVByKQ+2w2hV/gz8DogYwJHoIQ2KG0Du8N3qtzKURQkjiJISqau21bC3eyubizWwq3MTm4s2UWkrVDkscxk/vx8CogQyOGcyQmCH0Ce8jiZQQ7ZQkTkJ4WEFNAX9m/cnS7KWszVuLzWVTOyRxnF4xdWeEywhdR0PXMRCapHZIQggPkcRJCA/YWbqTpVlL+TPrT3aU7GhXM9zaGr1Gz/KcIvytVQcPhnWBrmdBl7Og0xngI8shCNFWSeIkhBvYXXbW5q/lz6w/+SvrL3JrctUOSbSQ/kFd+HDT0iOfoPOBjinQayL0mQT+4Z4LTgjhdpI4CdFCKm2VLMtexp9Zf7I8ZznV9mq1QxJucH1wX27e+GPzTtbqofMo6HsJ9DwfjAHuDU4I4XaSOAlxEnKqc1iauZSlWUvZULBBZr+1AwtckQzJWH/8Fxr8oPs50G9K3bgonaHlg/NCGo3mqM9feeWVLFy40DPBCNECJHES4jjV2mv5Jf0XvtnzDRsKN6gdjvAgX52JFWnpGJwnOaDfNxR6X1DXEpU4HI6RXLRm+fn59f//7LPPeOihh9i1a1f9MV9fX4KDDy5CarfbMRi8L6m02Wz4+PioHYbwArL/gBDNtC5/Hf9d/l9Gfj6Sh1Y+JElTOzQoMOnkkyYAcxmsXwgLz4cX+sCv/4W8TSdfrheKiYmp/woODkaj0dQ/tlgshISE8PnnnzNy5EhMJhMffvghLpeLRx99lA4dOmA0GhkwYACLFy+uL/PPP/9Eo9FQXl5ef2zjxo1oNBrS09MByMjIYMKECYSGhuLv70+fPn346aef6s/fvn075513HgEBAURHRzN9+nSKi4vrnx85ciQ333wzd955JxEREYwdO9bt3yvROshCJEIcRX5NPt/u+ZZv935LVlWW2uEIlaU43bCaeGUOrHy57iuiB/SdXPcV1rnl6/JS9913H//73/949913MRqNvPjii/zvf//jjTfeYODAgSxYsICJEyeybds2unXr1qwyb7rpJmw2G3///Tf+/v5s376dgIC6MWZ5eXmMGDGCa6+9lueffx6z2cx9993HlClTWLJkSX0Z7733HrNnz2bFihVI54w4QBInIQ6juFzULFtG2Sefkhpk5pU+JzCeRbRJyUUZ7q2geBcsfaLuK3E4pNwIPc4DbdvuHLj99tuZNGlS/ePnnnuO++67j8suuwyAZ555hqVLlzJv3jxeffXVZpWZmZnJxRdfTN++fQHo3PlgIjp//nwGDRrEk08+WX9swYIFJCQksHv3brp37w5A165dmTt37knfn2hbJHESYj9HaSnlX35F+eefY8/OBiAuJBi/XgZqtXaVoxNqC/EJpmfaVs9VmLGi7iusMyTPhoGXt9n1oQYPHlz//8rKSnJzcxk+fHiDc4YPH86mTc3vzrz11luZPXs2v/76K2PGjOHiiy+mX79+AKxfv56lS5fWt0Adau/evfWJ06FxCXFA2/4YI0Qz1K5bR85dd7NnxEiKnn++PmkCUMormFnSS8XohLcY6t8BjRoLl5bug5/vged7w28PQ2XbWxPM379xQnj4bDxFUeqPafe3wB3afWa3N/xwc80117Bv3z6mT5/Oli1bGDx4MC+//DIALpeLCRMmsHHjxgZfqampnHnmmUeNSwhJnES7pLhcVPz4I/suuJCMK6ZT+eOPKPamW5VOX2/2cHTCGyVbVV5qwlIOK+bBvH7w1bWQu1HdeNwkKCiIuLg4li9f3uD4ypUr6dWr7kNMZGQkUDdW6YCNGzc2KishIYEbbriBRYsWcdddd/HWW28BMGjQILZt20ZSUhJdu3Zt8CXJkjgWSZxEu6I4nVR8+y37xk8g9667sR4yLfpI9Bt3MsAW44HohDcblp+qdgh1XHbY8jm8OQLePR92/gRtbODyPffcwzPPPMNnn33Grl27uP/++9m4cSO33XYbUDf2KCEhgUceeYTdu3fz448/8r///a9BGbfffju//PILaWlpbNiwgSVLltQnXjfddBOlpaVMnTqVNWvWsG/fPn799VeuvvpqnE6nx+9XtC4yxkm0C4rdTsW331L85lvYMzOP82KFGXvj2Ngr/9jnijYp3i+ahLS1aofRWMbyuq+wLpAyGwZcDj5+akd10m699VYqKyu56667KCwspHfv3nz33Xf1M+oMBgOffPIJs2fPpn///gwZMoTHH3+cSy65pL4Mp9PJTTfdRHZ2NkFBQZxzzjm88MILAMTFxbFixQruu+8+xo0bh9VqJTExkXPOOae+G1CII5EFMEWbpthslC9aRMmbb2HPPfGxIdqIMKZdU4NNI59G26NJoX2Zs6GZ26yoyTcUTr0Kkq+HQGklFcIdJHESbZLLaqX8s88peecdHAUFLVLmL9cP4J0wD86qEl7jGWNnztv5p9phNJ/OBwZeASMfgIAotaMRok2RxEm0Ka7aWso+/YySdxfgLCo+9gXHwTb0FK44a2eLlim8nwYNS4tqCa8uUjuU4+cTAMNuhtNukQ2GhWghkjiJNsFZXUPZRx9RunAhzrIy91Si1fLIndFsN7TCN1BxwroFJLBoywq1wzg5/lEw8j4YNBN0MrRViJMho+BEq+aqraXotdfYe9ZZFL3wgvuSJgCXi5lpCe4rX3ilFH2o2iGcvJpC+PEueC0Ztn+ndjRCtGqSOIlWq+KHH9l77nkUv/QyzooKj9TZaXk6OtruTvaisZTKUrVDaDkle+Dz6fDO2ZD5j9rRCNEqSeLkQTNnzkSj0aDRaDAYDERHRzN27FgWLFiAy+VSO7xWw7JrNxnTZ5B7990tNvC7uZSCQqaWy0ri7YVeo2dw9ma1w2h5Wf/AgrPh08uh2EvWpxKilZDEycPOOecc8vLySE9P5+eff2bUqFHcdtttjB8/Hoej6ZWJD99KoL1yVlaS//gTpE2aRO1a9dbUGdv87bJEK9c3KAk/a7XaYbjPzh/gtRT4/naoLlQ7GiFaBUmcPMxoNBITE0N8fDyDBg3iP//5D99++y0///wzCxcuBOr2aHr99de54IIL8Pf35/HHHwfg+++/59RTT8VkMtG5c2fmzJnTINl65JFH6NixI0ajkbi4OG699db651577TW6deuGyWQiOjqayZMne/S+T4aiKJR98QV7zzmXsg8/BJVX9vVds50ujjBVYxCekYyv2iG4n8sB69+FlwbC0qegLSeKQrQASZy8wOjRo+nfvz+LFi2qP/bwww9zwQUXsGXLFq6++mp++eUXrrjiCm699Va2b9/OG2+8wcKFC3niiScA+PLLL3nhhRd44403SE1N5ZtvvqFv374ArFu3jltvvZVHH32UXbt2sXjx4gYbWXoz8+bNpE+5lPwHH8JZ6iVjTRwOrs5IUjsK4QEppe1otXhbNfz1dF0Ctf69NreNixAtRealeomePXuyefPBsRTTpk3j6quvrn88ffp07r//fq688koAOnfuzGOPPca9997Lww8/TGZmJjExMYwZMwaDwUDHjh0ZOnQoAJmZmfj7+zN+/HgCAwNJTExk4MCBnr3B4+QoKaHw+eepWPS1V76Ad1+RhaYzKDJOvM3y1fvSL6MNjm86lppC+P5W2PwZTHgRIrqpHZEQXkVanLyEoihoNAffhQcPHtzg+fXr1/Poo48SEBBQ/3XttdeSl5dHbW0tl1xyCWazmc6dO3Pttdfy9ddf13fjjR07lsTERDp37sz06dP56KOPqK2t9ej9NZficFD6/vvsPedcKr5a5JVJE4CSk8clVT3VDkO40akBiRhc7Xh8YcYKmD8c/noWnO34+yDEYSRx8hI7duygU6dO9Y/9/f0bPO9yuZgzZw4bN26s/9qyZQupqamYTCYSEhLYtWsXr776Kr6+vtx4442ceeaZ2O12AgMD2bBhA5988gmxsbE89NBD9O/fn/Lycg/f5dHV/LOGtIsmUfDkU7iqqtQO55jO3SINtm1ZilNeHnFaYenj8MaZkOWFmxwLoQJ5ZfACS5YsYcuWLVx88cVHPGfQoEHs2rWLrl27Nvo6sJu3r68vEydO5KWXXuLPP/9k1apVbNmyBQC9Xs+YMWOYO3cumzdvJj09nSVLlnjk/o7FWVFBzt33kHnllVhTW8/UaP9V2+jgCFY7DOEmKYWZaofgPQq31y1f8NM9YPX+DzVCuJN8ZPYwq9VKfn4+TqeTgoICFi9ezFNPPcX48eOZMWPGEa976KGHGD9+PAkJCVxyySVotVo2b97Mli1bePzxx1m4cCFOp5Pk5GT8/Pz44IMP8PX1JTExkR9++IF9+/Zx5plnEhoayk8//YTL5aJHjx4evPOm1axeTe79D+DIb4WDcO12rsnuwiNJG9SORLSwMGMI3dO2qB2Gd1FcsOZN2PkTTHwJup6ldkRCqEJanDxs8eLFxMbGkpSUxDnnnMPSpUt56aWX+Pbbb9HpdEe8bty4cfzwww/89ttvDBkyhJSUFJ5//nkSExMBCAkJ4a233mL48OH069ePP/74g++//57w8HBCQkJYtGgRo0ePplevXrz++ut88skn9OnTx1O33YjLZqPg6WfIvOrq1pk07ddnVa7aIQg3GOIXjwbvHF+nusps+HASfH+bLF0g2iXZ5Fd4nGX3bnLvuRfrrl1qh9IiPry5F98Ftp4uRnFsD/v1YPK239QOw/uFdIQLXoNOZ6gdiRAeIy1OwmMURaFk4ULSJ1/SZpImgInb28Eiie1Mcl7b+f10q/JMeG8C/HQv2Lxzpq4QLU0SJ+ER9oICsmbNovDpZ1BsNrXDaVHBK7cT5fI/9omiVYj3iyahVAaGN58Ca96A10+HzNVqByOE20niJNyucvEvpE28gJqVq9QOxS0Ui4Vrc7qrHYZoISnGKLVDaJ1K98K759at+yQjQEQbJomTcBtndQ259z9Azu2346yoUDsct+q/ukjtEEQLSampUTuE1ktx1a379MllYC5XOxoh3EISJ+EWtRs2kHbhhVR8843aoXjGnnTG1XZWOwpxkjRoGJq9Ve0wWr/di+HNEZDXDresEW2eJE6iRSkOB4Xz5pExfQb27Gy1w/GoSTsC1Q5BnKTugR0JqylWO4y2oSwd3jkb/v1I7UiEaFGSOIkWYy8oJOOK6ZS8/gY4nWqH43Fhy3cQ6pIZdq1Zsk5Wgm9RDjN8e2Pdmk8Oq9rRCNEiJHESLcK8cSPpkydj3rhR7VBUo9TWck2++quxixOXXFmidght0/qFsGAclGepHYkQJ00SJ3HSyr/8kozpM3AUyQDpwWvK1A5BnCC9Vs/gLBmT4za5/9ZtFrznD7UjEeKkSOIkTphit5P/6KPk/fdBFLtd7XC8gmbHXkaYE9UOQ5yAfoFJ+NlkRp1bmUvho8nw11xZskC0WpI4iRPiKC0l86qrKfv4E7VD8TpTdoepHYI4ASnI+DSPUFyw9An4+FIwSwutaH0kcRLHzbJjB2kXT6Z23Tq1Q/FKUct2Eugyqh2GOE7JJbJhs0el/gJvyJIFovWRxEkcl6olS0m//AoceXlqh+K1lKoqri7qqXYY4jj46f3omyPrN3lceQa8MxY2fqx2JEI0myROotlKFi4k++abUWplM89jSVlbpXYI4jicGtARg0vG6anCYYFvZsPfz6odiRDNIomTOCbF4SBvzhwKn34GXC61w2kVdFt2M8zSQe0wRDOlOOSlUHVLHoef75dB48LryauFOCpndTVZN8ym/JNP1Q6l1ZmaGql2CKKZkgvT1Q5BAPwzHxZdB05p/RPeSxIncUT2nBwypk6lZvlytUNplWKX78bPZVA7DHEMYcZQuhfsUjsMccCWz+GTqWCTIQHCO0niJJpk3buX9MumYk3do3YorZZSXsHMkl5qhyGOIdkvHg3SPeRV9vwG718gyxUIrySJk2jEsms3GTOulJXAW8Dp681qhyCOIdliUzsE0ZTsNbDgXKiUZSKEd5HESTRg2b6dzCuvxFkie3a1BP3GnQywxagdhjiKlPzdaocgjqRoB7wzDoql5Vt4D0mcRD3z5s1kzLwKZ3m52qG0HYrCjL1xakchjqCDXwzxpZlqhyGOpiKzboPgnA1qRyIEIImT2K92wwYyr56Fq7JS7VDanI7L9+Cj6NQOQzQhxSgzH1uF2mJ4bwLs+1PtSISQxElAzT9ryLrmWlzV1WqH0ia5ikuZXiaDxL1RsvzOtx62avjoEtj2tdqRiHZOEqd2rnrFCrKuvx6XrAbuVqP+dagdgjiMBg3Jss1K6+K0wZdXw9p31I5EtGOSOLVj1X/9RfbsG1EsFrVDafN81m2nt126hbxJj8COhNbIJIhWR3HBj3fC8hfUjkS0U5I4tVNVv/9O9s23oNhkKrZHuFzMTEtQOwpxiBRdsNohiJPx+yPS8iRUIYlTO1S5eDHZt9+BYpdtDTyp0/J0dGjUDkPsl1wprU2t3k93w5Yv1Y5CtDOSOLUzFd9/T85dd4NDxtx4mlJQyNRyGSTuDQxaA4OyNqsdhjhZigu+vgFSf1M7EtGOSOLUjpR/tYjc++4Hp1PtUNqtsZvUjkAA9AtMws9Wo3YYoiW47PDZdMhYpXYkop2QxKmdqPzlV/IefBBcLrVDadd812ynsyNU7TDavWTFqHYIoiU5zPDxpZAnrYjC/SRxagdqN/xL7r33StLkDRwOZmV2UjuKdm9YaZ7aIYiWZq2ADydByV61IxFtnCRObZw1LY3sG29EsVrVDkXs1315FhpF7SjaL3+9H6fkbFE7DOEONUXw/oWyMbBwK0mc2jBHSQlZ110ve895GSUnj8lVPdQOo90aHJCI3iWTI9qsisy65Km2VO1IRBsliVMb5TKbyZp9I/asLLVDEU04b4uP2iG0W8kOWRKizSveBR9eDNYqtSMRbZAkTm2Q4nSSc9fdWDbLQElv5b9qKx0csgCjGlIK09QOQXhC7gb4ZCo4ZJiCaFkaRVFktEUbk//oo5R9/InaYYhj2D51CI8k/eux+mp21VD8UzHmDDOOcgcdb+lI0KlB9c8rikLhN4WU/VWGs8aJb2df4mbEYYo3HbFMS46FwkWFmNPN2EvsxEyNIWJcRINzyleWk/9lPopVIfSMUGIui6l/zlZkI/25dLo80gWdr67lb/ow4cZQ/twpa0K0Kz3Oh0s/AK37f79E+yAtTm1MydtvS9LUSvRZ5dmZXS6rC1NHE7FXxDb5fPFPxZT8UkLsFbF0ebgLhmAD6c+m4zQfed0vl9WFT6QP0ZdEow/WN3reUeUg590cYi+NJfGuRMpWlFG18WD3Se77uURfEu2RpAkg2S/eI/UIL7LrR/j2ZpA2AtFCJHFqQyp+/JHC/z2vdhiimZT0bCZWdfNYfYH9Aom+OJrgwY27CBVFoeTXEiInRBI8OBhTBxPx18bjsrqoWF1xxDL9OvsRc1kMISkhaPSNxw7ZimzofHUEJwfj19kP/17+WHLrNpUuX1WORq9pMh53SbHI3ozt0qaPYcljakch2ghJnNqImjVryHvgP/KpqpWZuN1X7RAAsBfZcVQ4CDgloP6Y1qDFv6c/tXtqT7hcY7QRl81V1z1Y7cCcZsaUYMJR7aDw68Ijtn65S0reLo/WJ7zIsv/B1kVqRyHagMZt66LVse7dS/Ytt6LY5NN0axO8cjtRQ/wp1Kq7/Yejom56vj6o4UuCPkiPveTEN4PW+evocG0Hst/KRrEphJwWQmDfQLLfySZsTBj2YjuZL2aiOBWiLowieIj7Wp86+sUSm/aP28oXrcC3N0FEN4jpq3YkohWTxKmVcxQVkXXtdbgqjtydIryXYrFwbU5fnkjw3CDxozq8t60FGjCDTg1qMAi9ekc11mwrcVfEsfu+3STckIA+WM/eR/fi38O/UfLWUpKNEcc+SbRt9lr4ZBpc9yf4h6sdjWilpKuuFXOZzWRdfwP2XFkltzXrv7pI7RDqB3YfaHk6wFHlaHLQ94ly2V3kfZBH3JVx2AptKE4F/57+GGONGGOM1O498W7BY0mpkTV9BHULZH4+A5yyCKo4MZI4tWL5j8zBsn272mGIk7UnnXG1nVUNwRBpQB+sp3pbdf0xl8NFzc4a/Lr6tVg9Rd8VEdA3AN8kXxSXAodsn6g4Gj5uSVqNlqFZW91TuGh9MpbD4vvUjkK0UpI4tVLli76m4ttv1Q5DtJBJOwLdXofT4sScYcacYQbAVmzDnGHGVmJDo9EQfnY4Rd8XUbm+Eku2hZy3c9AatQSnHBx3lP1mNvlf5Nc/djlc9WUqTgVHmQNzhhlrQeNFBy05FirWVBA9KRoAY6wRNFD6VylVG6uw5lnx7eyewfI9AjoSIltwiEOtfRtl/XtqRyFaIRnj1ApZ9+wh/zGZWtuWhC3fQehAX8q0ZrfVYU4zk/5Mev3j/E/qEqCQ4SF0uLYDEedF4LK5yH0/t24BzC6+JN2d1GCNJVuJrcE4KEeZg70PH9yNvnhxMcWLi/Hr4UfnBw62oimKQu67ucRMjUFrrPu8pvXREn9NPHkf5KHYFWKnx2IINbjl3lN07k9MRetiD+7MbX/ruC6qnAEJIWqHI1oRWTm8lXGZzaRPmYI1dY/aoYgWtvbKwTwbt1HtMNqkNzRxnLZvtdphCC9RFDeaibkzyLP4EBds4odbzyDMX/aPFM0jXXWtTP5jj0vS1EYNXlOmdghtkkFrYGC27NsoQNFoWZlwHUPTZpFnqUuUciss3Pbpv7hc0oYgmkcSp1ak4ttvqVgkC7i1VZodexlhTlQ7jDanf2ASvjb3zdYTrYPLFMK8yEeZljoSRWm47say1GLm/b5bpchEayOJUyth3bePvDmPqh2GcLMpu8PUDqHNSVGMaocgVGYJ68Xlmqd4MfPIs1dfWbqHVXtLPBiVaK0kcWoFXBYLObffgVIrn5rbuqhlOwl0yRt9S0ouzVE7BKGi7A7nM6zoAVaVHX1VepcCd36+kfJa2YFBHJ0kTq1AwRNPYN0tzcjtgVJVxdVFPT1Sl6PawY5bdmAr8uwbRd6neeR+6JlFWwMM/vTNlvWb2iNFq2dxh9s4fc/llNmbN4E8r8LCvV/KeDhxdLIcgZer+P4Hyr/4Uu0whAelrK3mxfHur6fohyICBwTiE3lwNlHZsjKKfynGlm9D56cjaEgQcdPj6p+3ZFnI/TAX8z4zOn8dYaPCiJwYiUZTN2akekd1gyUPDuj2ZDeMcXUtaZHnRbL73t1EjItoULc7DPbviE7Z4dY6hPdx+UUyx3Qv7+2JP+5rf91ewAerM5ieIuMNRdMkcfJi1rQ08h9+WO0whIfptuwieWwi/xjd18Xksrko+7uMpDuT6o8dWIMp5tIYfLv4otgVbIUHW6OcZifpz6bj38ufuIfjsOZb6xbJ9NEScW7DfeC6Pd0Nrelgg/ah+8/pg/QE9AmgdGkpMVNi3HaPACmOwzffE21ddeRAppbfyJZS/xMu44kft5PcKYzu0bL+l2hMuuq8lMtqJeeOO3HJuKZ2aVpqlFvLr9pchUanqd9OxVnjpGBRAR2u60DIsBCMUUZM8SaCBh7cnLd8VTkuu4v4a+IxdTARPDiYyPGRFP9SzOHLwekD9RhCDPVfGm3DBCZwYCAVq92/MXVy4T631yG8x+6ES0jOvYstVSeeNAFY7C5u+fhfLHZnC0Um2hJJnLxUwZNPYd25U+0whErilqdiUtzXIFy7qxbfpIPbm1RvqwYX2MvspD6Qys47dpL5ambdSuH7mfeY8e/pj9Zw8GUjoG8AjnIH9mJ7g/L3PLyHnbftJO2ZNKp3VHM4386+2Evt2IrdN74q0hRG1wIZG9geKDojn8ffx9mpF1HjbJm3tV0FVTzxo3TzisYkcfJClb/+Svlnn6kdhlCRUlbOVcW93Va+rdiGPuRgYmYrtIECRd8XETMtho43dcRZU9c153LU7bxrr7A36HKDg11wjoq6neYNIQbiZsbR8eaOdLylI8ZYI+lz06nZVdPgugNbqxyecLWkob5xxz5JtHqOwHjuCniGe/f2b/GyP1idwfLU4hYvV7Rukjh5GWdFBfmPyj50As7YYHFb2YpdadBypCgKilMh9opYAvsG4tfVj4QbErAV2KjZcUjSc4whQ8ZYI2Ejw/BN8sWvqx9xM+II7BdI8c8N33wO1O2yuVrsng6XbGm80fCJ+jvDwYRPaon7XxWaOZV8s7NhwqcoCo/8aSHuf1X4PlHJyIU1bCs8ejfPoh12Br9ZTcjTlfg/WcmA16v5YFPDFriPNttJeKGKsGcquefXhr8P6eUuur9cTaW1/a54XR5zGmfXPMaiAvd1bd/31WZqrA63lS9aH0mcvEzBU0/jLJZPOAL0/+5ggM09g6d1ATqctQff2A0hdS1ApjjTwfqD9OgCddhL6pIEQ7ChvmXpAEdl3WN98JG7FX27+GIraJgQOGvq6tYHuq87cljurhYrq8am0D9ayyvnmZp8fu4KG8+vsvHKeSbWXutPTICGsR/UUnWUpCbMV8P/nWFk1Sx/Nt8QwFUDDFz1rYVf9tR9T4trXVzzvZnnxpr45Qp/3ttk58fdBxO22T+aeXqMkSBj+xwAvzFhBkMyb2JfbdM/k5aSU27m6Z9l2IQ4SBInL1K9bDkV33yjdhjCWygKM/a6p7vJN9EXa87BFhm/bnWDxK35B485qh04q5wYIuqSKt+uvtTsqqnvugOo3lqNPkRff05TLJmWBt2CAJYcCxqdBmO8exb7TPSPJaY8u8XKO7ebgcdHm5jUq/F9KorCvH9s/N8ZRib1MnBKlI73LvSl1q7w8ZYjd0WOTNJzUS8DvSJ1dAnTcluKkX7RWpZn1iVO+8oUgo0aLj3FwJB4HaM66dheVPe9/3iLHR+dpsl42jrFx5+3Yx7mwtRzsLs8kzR++E+GrCou6kni5CVcNTWy9IBopOPyPfgouhYvN+CUACy5lvqWH2OMkcCBgeR9lEdtai2WbAs5b+VgjDUS0DMAgJCUELR6LTlv52DJtlC5vpKiH4qIGBdRv45T8S/FVK6vxJpvxZJjIf+LfCrXVRJ2VsOtZGp21eDX3Q+tj3teglJ8Io59UgtJK1fIr1Y4u8vB5NCo1zAiSc/K7ObNylIUhT/2OdhV4uLMxLpyuoVpqbUr/JvnpNSssDbHSb9oHaVmhYeWWnjlXPe2tHgjW0hnrjE8w+PpPTxar6LUddnV2qTLTsg6Tl6j8PkXsOd6ZjVl0Xq4ikuZXjaAd8JadvVrU4IJ3yRfKtZUEDaqLqnpcF0H8j7OI/2FdDQaDf49/Um8KxGNvi4p0vnpSLonidwPctn7yF50/joixkUQfk54fbmKUyH/s3zsZXa0PlqM8UYS70gksH/D9XAq/qkg6kL3jUtJrq50W9mHy6+uawWKDmjY+hHtryGj4uhjuCosCvHPV2F1gk4Dr51vYuz+BCzUV8N7F/oy4xszZrvCjP4GxnXVc/W3Zm4Z6kNauYuJn9Zid8IjI41M7t22W58K485iYs508q3uXTT1SDJLa5m7eBePTOyjSv3Ce0ji5AVqN2yg7JNP1A5DeKlR/zp456yWLzdqYhT5n+UTOiIUjVaDzldHh1kdYNaRrzElmOj8nyNvlBp5XiSR50Uetd6qjVVotBqChxx977ATpdVoGZq1xS1lH83hnUaKcsyx9AQaYeMNAVTb6lqc7vzFQudQLSOT6l6aL+pl4KJDuuP+THewpdDJK+eZ6PpSNZ9c7EtMgIahb9dwZqKOKP+214mgaLSs6HAt0/eciaKoO57rvVXpnNc3lqGdZDPu9qzt/ZW1Mk6Hi7SPfgKX+2YXidbNZ912etuPnoyciMD+gYSODMVe5r4lAZrisrmInxWPRueeN8GeAR0JNpe7peymxATUvYzmVzccCF5YqxAdcPSXWK1GQ9cwLQNidNx1Wl2r0VPLm54NaHUo3PijhTfG+7Kn1IXDBSOS9PSI0NE9XMs/zewWbE1cphCej3yMK1JHqJ40QV0yfP9Xm7E55PW6PZPESWX//prJX7WnkT31GVyB8ilGNMHlYmZagluKjjg7Ap9wz3Z9BA8Nxq+Ln9vKT9F5dpuMTiEaYgI0/Lbv4PgXm1Phr3QHp3U4vvFpigJHmvn+2N9Wzu2qZ1CsDqcLHK6DiZrdCc42tiqBJbw3U3malzM7qR1KA/uKa3jz771qhyFUJF11KqosNrP+53QUBXbnBZAz6il6Vy8jcMmHaofWaqyrrWVBaQnbLFaKnA5eiotnTODBN05FUXi1pJgvyiuodDnpZzLx3+gYuhmPPJsr1WrlleIitlks5Doc3B8ZxYywhknt95UVvFBURK3LxcXBIdwTdXC8To7dxjVZWXyRmESArmUGdndano6uuwYnbezd0Q2Sy4tavMxqm8Ke0oOtDGllLjbmOwnz1dAxWMvtyT48ucxKtzAt3cK1PLnMip9Bw7S+B7vZZnxtJj5Qw1Nj6gZ1P7XMyuC4uhl1NqfCT6kO3t9sZ/75jQd9byt08tk2Bxuvr9tKpGeEFq1GwzsbbMQEaNhZ7GJIXMtPIlBLVofxTMyYQpndO9+iXlm6hwsHxtMh1H0fAIT38s7fynbi709347AffDGuqXSwlmEkTEuh01/z0OfsUTG61qHW5aKH0cRFwSHcltt4U9x3Skt5r6yMJ2NiSfLx4fWSYq7JyuKnzp3w1zb9RmNxuehg8GFcYBBPFxY0er7M4eCh/HyejImlg8HA7Jxshvr5MSKgbvbZnIIC7oyMarGkCUApKGRqeT8+DNneYmW2RT5aHwZltvz4pnW5Tka9d3DfyDt/tQJWruxvYOGFvtw73AezQ+HGnyyUmRWSO+j4dbofgYessZRZ4UKrOdjIX2OvOz+70oWvHnpG6PjwIl8uPaXhIG9FUbjuBwsvjDPi71NXnq9Bw8ILTdz0kwWrA145z0R8UOvvQFC0ehbH3czsPUPVDuWoLHYXc77fzlszBqsdilCBRjl8d07hEXv/LWTxG0eeKeVj0tHbuJvQ719E42p7YxfcofeunQ1anBRFYcTePcwIDeOa8LqZXzaXizP27uHOyEguDQk9Zplj9l9/aIvTZrOZm3KyWda1GwB35ubQx2RiVlg4P1RW8HNVFa/Gd2jx+zOf1o8rR0jidDRDg7vzzsbf1Q5DnACXXySPmO7j/dzWs1XOuzOHMKqnezfkFt6n9X9EaYXsVifLP0896jk2i5ONFV3Yesmr2HoP81BkbUu23U6x08lp/gd3SvfRahns58dGs/mEy0308cGiKGy3WCh3OtlqsdDDaKTc6eTl4mL+GxXdEuE34rtmO50dx0722rNkRZ2p6uLkVEcOZIL9yVaVNAE8/N02LHb5YNveSOKkgjU/pFFd1rx9tIoKnKyMmU7upY/j8g1wc2RtS7GzbpRthL5hl1mETkex48Rf7IJ1Op6KieWBvDwuzUhnYlAQp/sH8GxhIVeEhpJjtzMpPY2Jafv4paoF1xNyOJjlZQNlvU1KSePuWuHddiVcQnLuXWyr8j/2yV4ms7SW1/+SgeLtjYxx8rCSnGo2/5F1XNe4XAo7C0LJGfc8fYoW47dikZuia5s0h62mo3Ds9XWOZUxgYINB6Gtqa0i1WflvdDTn7NvHc3FxROh1XJqRwWBfP8L1LfOn1n15FppO4AUzs71OoCGAPuktu1CocB9Fb+LzqNu5L7Wf2qGclPl/7mXSwA50DJeB4u2FtDh52Kqv9+JyndiwsqpyO6sNZ7Fv2jyckS0/hqatidDVJStFjobzu0ucTsL1LTdw2+Zy8WhBAY9Ex5Bps+FEYYifH518jCT5+LDZcuLdgodTcvKYXOXZ7SZai8EBHdEp0m3SGjgC47nD/2nu29e6kyYAq8PFkz/tUDsM4UGSOHlQ7p5yMrae/EaR6bkG/hnyX8rPu7EFomq7OhgMROh0rKqpqT9mUxTW1dYywNe3xeqZX1LCGf7+9DaZcAKOQ+Zb2BWlxdfXOW+LjONpSrLdc/NcSmpdRD1bRXq5ZxdCvPtXC7f+bPFonS2tLGY4Z9c8xjcFbWdQ9eJt+azPKFM7DOEhkjh50OpvWq4v3FLrZENtH7ZfNh9714EtVm5rU+NyscNiYYel7s0kx25nh8VCrt2ORqNhRmgYb5aW8HtVFalWK/+Xl4dJo2V8UFB9Gffn5fJ8UWH9Y5ui1JdpV6DA4WCHxUKGzdao/lSrlZ+rKrklom5l784+Pmg1Gr4qL+ev6mrSbDb6mlp2M1b/VVvp4HDPdiWt2bCCfR6r66nlNiZ015MUcvAldOFGG/3mV2N6vJKY56q4+aeDLY0Wh8LMb8z0nV+N/tFKLvy0tqlisToU/u8PC4nzqjA+XkmXl6pY8O/B37t7h/vw7kYbaWWtc+XqfxOuZGjmjeyrbXsbFEurU/shY5w8JGNbCXl7Klq83Px8F0VJ19JzYD5R385FY2vdn0aP1zaLmZlZB8eMPbM/AbowKIgnY+OYFRaGRXHxaEE+lS4X/Uwm3k5IaLCGU57d3uATRJHDzsUZ6fWP3y0r5d2yUob4+vJex8T644qi8Eh+PvdHReOnrSvBpNXyZEwsjxXkY1MU/hsVTbShhTdftduZldOFOYkbWrbcVizKFE7ntH89UpfZrvDOvzZ+mnZwTMvzq6z8b5WNZ8eaSI7XYXEo7DskuXG6wFcPtw714asdR97iZsqXZgqqFd6Z6EvXMC2FNUqDFcKj/LWc3UXP6+tsPDO29SQfio8/b4fdzROpbbebeX1GGYu35nHOKbFqhyLcTNZx8gBFUfjiqXUUZVa5tZ6QcAO9Mhfhu3axW+sR6tMkdeCSqflqh+E1JoT25ckNP3qkrkU77Fz/g4Wie+omB5SZFeKfr+L7qX6c1fnYn0VnfmOm3KLwzWUNBxMv3uPgsi9r2XdbIGG+Rx79/95GGw8utZJ5h2e3ljlRtpAu3GC/kyUlbX8pjc4R/vx6x5noddKZ05bJT9cD9m4ocnvSBFBeYmdVwAQypz6HM7TtjB8QjSnp2Uys7qZ2GF4j2ey5lta/M5wMjjv40vnbPgcuBXKqXPR6tZoOz1cx5YtasiqOrzvtu112BsfpmLvCSvzzVXR/uZq7f7VgPmzs1tB4HVmVChkeHl91IgrjzuLMsgfbRdIEdfvYfbImU+0whJtJ4uRmLpfCmu89N/YCBfbk+bJ2+GNUjr3Kc/UKj5u4veUGuLd2yXk7PVZXermLuICDL537yly4FHhymY1540x8OcWXUrPC2A9qsR3HzIB9ZS6WZzrZWuji60v9mHeOkS+327npp4ZJ4YGtVTw9MP14KBotyxJuIDntavKt7Wsyw4t/pFJ9pJ2aRZsgiZOb7VyVR1l+0wNB3am22sE6+2B2TXsVR8eeHq9fuF/wiu1EuVrfooEtLck/nphyzy18aXYomA7pkXMpYHfBS+eaGNdVT0oHPZ9c7EtqqYulac1fHsGlgEYDH03yZWi8jvO6GXh+nImFG+0NWp1899dd68FZhMfDZQrluYjHmZ56Jko7XHCsuNrGG7IoZpsmiZMbOR0u1v6YpmoMObmwsudtFF90H4pO5gK0JYrFwjU53dUOQ3XJPmHHPqkFRfhpKLMcTFpiA+qSg96RB19OI/21RPhpyDyO7rrYQC3xgRqCTQeTjV4RWhQgu/JgOaVmpb4Ob2MO78NlPMWrWUlqh6KqBcvTKKtpPAtXtA3e95fXhmz9O4fq0uZtreJODpuLzWUd2TzpVaz9zlQ7HNGCBqwuUjsE1Q1ryW1tmmFgjI7tRQcTmeEd62Zo7ipumNwU1yokhjT/JXZ4go7cKoVq28GkbHeJC60GOgQdLGdroQuDFvpEetfLd2aHCaQU3sea8qBjn9zG1dicvL3cg0M0hEd5119eG2K3Oln/c7raYTRQUuRgZcSl5Fz2FK6AtrkOULnTyel7Usmxe/bT3tzCQp4oKPBonQDsSWdcbWfP1+sltBotQ7I3e7TOcV30bCtyUba/5ad7uI4Leui5bbGFlVkOthY6ufIbMz0jtIxKOrjsxfYiJxvznZSaFSqsChvz6x4fMK2vgXA/DVd9a2Z7kZO/Mxzc85uVqwcY8DUcbIValungjERdg2NqUrQGfupwO2fumUqFXVq1D3hvZQYVtUdeekK0XrIcgZus+ymdf77z3k8cAcF6+pQvxf+vT9UOpUXNLSyk0uXk8ZiDa6l8XVHOe6VlpNttBGq1jAsM5L/RMfXP77ZaeLyggC0WC8E6HVOCQ5gdHo5GU/fGtL62lueLithns2JRFOIMBqYEh3Bl2MEuohKHg3H79vFNUhIdfDw7GLbs7FO5/tRNHq3TW5wS1IlPNv3l8XqHvVPDzP4Grh9c97OutCrcsdjCop12tBoNIxJ1vHiOiYTgg59Nk+ZVkVHR+OVWefhgC83OYie3/GxhRaaTcD8NU3obeHy0sUGS1OOVauaMNHLZKS28PtgJcPpH8YjxHj7IjVc7FK906+iu3Hl22127qr2SxMkNLDV2PnxwFdZa759ZkRjnJHHp8+jz0tUO5aRZXC5G7t3D6x0S6rdUWVhaysKyUu6OjKSfyReropBttzEqoG4NnGqnk3PT9pHs58f14RGk22z8Jz+PG8PDuSosHIDtFgtpNhvdjUb8tBrW15qZU5DPfVHRTAkJqa//tpxsOvr4cFekZ5eC0Pj5cd0tOsq0LbcnXmsxK6Qvt//rmfWbDvVTqp27f7Wy9UZ/tBrPtfz8uNvOPb9Z2TzbH71W3Ran6qhTuaR0NjuqZXPbIwk06Vl+32iCfdVPckXLka46N/j314xWkTQBZOTqWN3/fkon3I7iwTcAd1hWU4NOo6lPmiqcTl4qLuKpmFjGBwXT0ceHbkZjfdIE8ENlJTZF4cmYWLoZjYwNDOS6sHDeKyvjwGeK3iYT5wcF0c1oJN7gw8TgYIb7+7Pe3HC25KiAQH6s9Ox4GwCltpZr8tvnp9qUcnXGeJ3XzcD1pxrIqfTs584aO7x7gUn1pGlXwqUMybldkqZjqLI4WLBc3QlCouVJ4tTCaiqsbF6arXYYx8VmcbKxqhvbpszH1nOI2uGcsHXmWvocsi/cypoaXEChw8H4tH2M2ruHO3JzyLMfHHew0WJmsJ8fPtqDfwrD/f0pdDjIsTc9PmG7xcK/ZjNDfBu+afQ1mcg/ynXuNHhN+9tg1KgzMjDLs+ObDnVbirFBV5wnTOljILmDeuOIFL2JT+IeYFzqBZidumNfIHh3RRqVFhnr1JZI4tTCtizNxmHz3oXpjqawwMnK+KvInzIHxdj6FlfMsduJ0h98U8m223EpCm+WlnB/VBTz4uKpcDq5JjsL2/7WpGKHgwhdwzeACH3d42JnwzV4Ru3dQ//du5iSkc60kFAmH9JNBxC9v+5cFRInzY69jDAnHvvENmRAQCJGR/vam1FNjsAO3O7/NA/s66t2KK1KpcXBwhXpaochWpAkTi3IYXeybXmu2mGcFJdTYXthBOvPe5HaYRPVDue4WF0KxkO6G10oOID/REVzun8A/X19eS42jgybjTW1NYdc2bDb48Cov8M7Qz5I6MgXiYk8HB3D+2WljbrljPtbrcwudRLnKbs9u56R2lKU9rUitZrKYoYzpvpRvi2QrZxOxMKV6VjszV8MVXg3SZxaUOraQizVbaNJtrLMzmrjONKnvYAzvHXs9h2i01HpPJi0RO5vAepyyCy3ML2eUJ2OPHvdGLQIvZ5iZ8PxaCX7W5rCD2uJ6uDjQ3ejiUtC6mbUvVpc3OD5iv3XhenV6cKIWraTQJdRlbrVkFycpXYI7cKGjjMZnDGbdLPp2CeLJpXW2Fi0wXOr2wv3ksSpBW35s3WNbWqOfbk+/JPyCOXnXK92KMfUy2Rkj+3ggqOD9o9BSrMdXNOp3OmkzOkkzlA3y2WAyZf1tbX1XXdQNzYqSq8n3nDkmTCKAjalYctSqtWKHujqo07yolRVcXVR+9heJ9AQQJ/cbWqH0aYpPgG8Hv0Ik3afjVORt4qTtWBFGjKJvW2Qv4YWkrennKLMKrXDcAtLjYMNln7smDofe+d+aodzRKf7+7PXaq1v+Uny8WF0QABPFRbwr7mWVKuV/+Tl0snHh6F+dUnV+UFBGDQa/i8vj1Srld+rqniztIQrQ0Pr13H6uKyMpdVVpNtspNtsLKoo592yUiYENVxEdL25llP9/DBp1fuzSllbrVrdnjQkoCNapXWOJWwNbCFdmGV4hqczZEuflrKnsJo/d8tK/22BJE4tpLXNpDsReXkuVnWZTeHF/4dL733jS7obTfQxmVh8yBYcT8fE0s/Xl9nZ2czIzECv0fBmhwQM+5OiQJ2OdxISyHfYuSQjnUcL8rkyNIyZoQfHC7lQeKGoiIvT05iSkc5HZWXcGRHJLRERDer/qbKKS4JDPHKvR6Lbsotka9tfjDDFJp/c3aUgbgynlz3EkpJQtUNpc95ZJksTtAWyAGYLqC6z8sH/rcTlaj/fytAIA732fYFpw29qh9LAX9XVPFtUyHdJnTy6MOGBer9J6oRe5fWw8iYM4bZT/lU1Bnf7ttpA5yLZgb4lKRotf3e4npl7TkdRWveabt5s8e1n0DNG9vNrzaTFqQVs/Tu7XSVNAGXFdlYFX0jW1Lm4gsLVDqfeiIAALg0JocDh2QVIa10unoiJVT1pAohbnopJabt7hkWZIiRpamEuUyhzIx7nytQzJGlyM2l1av0kcTpJTruL7a18CYITpSiQmufPmhFPUnXWDLXDqTc9NIzYowzsdodzg4Lo7+sda18pZeVcVdxb7TDcJsU35tgniWYzh5/CFOVp5mclqR1Ku/DtplyKqqzHPlF4LUmcTlLqugLMVW1jCYITVVvlYK0zmdRpr+Do0E3tcARwxoa2uzBkirnt3punZXSYSErhvayrCDz2yaJF2BwuPlmT6ZG6NBrNUb9mzpzpkTjaGkmcTlJ7GBTeXFm5Glb1uZOSC+9B0cp2DGrS/7uDfrZotcNwi+TcHWqH0OopWgM/dLiDEXsuo8Ledrt1vdVna7M8MrwjLy+v/mvevHkEBQU1OPbiiy+6PYa2SBKnk9CWlyA4UXari03lSWyZ/BrWU05XO5z2S1G4cl/bm13XyT+eqIo8tcNo1Zz+0fw3+Clu3tN696Vs7XLKzSzbU3zsE09STExM/VdwcDAajabBsY8//pguXbrg4+NDjx49+OCDDxpcr9FomD9/Pueeey6+vr506tSJL774wu1xeztJnE7C5ja44GVLKS50sCpqGrmXPYHLX2aQqCFx2R58lLbV8pdi8J6JCK1RVdRgxluf4KO8OLVDafc+9VB33ZF8/fXX3Hbbbdx1111s3bqV66+/nquuuoqlS5c2OO/BBx/k4osvZtOmTVxxxRVMnTqVHTvad6uvJE4nqLrMyr4NspjZ0bhcCjvzQ1g/9jlqzpisdjjtjqu4lCtKe6kdRotKri5XO4RWa2fCpQzNuY0d1X5qhyKA33cUUFyt3iDx5557jpkzZ3LjjTfSvXt37rzzTiZNmsRzzz3X4LxLLrmEa665hu7du/PYY48xePBgXn75ZZWi9g6SOJ2gbcty2t0SBCeqqtzOP7pR7Jv2Is6ojmqH066M2ujZZRncSafRMTRri9phtDqK3peP4x7gnNQLMDvbVgtka2Z3Kny5Xr1eix07djB8+PAGx4YPH96oNWnYsGGNHkuLkzhuLqeLbctkw8bjlZ6rZ/Wp/6Fs/M0oXrDeUXtgXLednvaIY5/YCvQOTCTQUqF2GK2KIyiBW/2e5j/7+qodimjCZ2vV3ahac9jrsKIojY4157r2RhKnE5C1s6zdL0FwoqxmJ/9W92L7pa9h6zZI7XDaPpeLq9PaRitfisZf7RBaldKY0xldNYfvCyPVDkUcQVpxDav2lqhSd69evVi+fHmDYytXrqRXr4bd+6tXr270uGfP9rGZ+JHIPNQTsGd9odohtHoF+S6KE6+h54ACor59Bo1N1uZxl07L09F11+CkdXctJ5cXqB1Cq6CgYUPCTKbsOQunIp+Nvd1nazMZ1sXzkx7uuecepkyZwqBBgzjrrLP4/vvvWbRoEb///nuD87744gsGDx7M6aefzkcffcSaNWt45513PB6vN5G/quPkdLhI2yiDwluC06GwrSiKDeNfwjz0fLXDabOUgkKmlrfuQeImnZGBMr7pmBSfAF6PfoSLU8dK0tRK/LKtgBqr58ciXnjhhbz44os8++yz9OnThzfeeIN3332XkSNHNjhvzpw5fPrpp/Tr14/33nuPjz76iN692+7OBM0hm/wep/Qtxfz46ma1w2h7NNAl1kKHxXPRlUrLQkszn9aPK0dsVzuME5YS0oO3/vWuDaW9jS20G9dZb+fP0lC1QxHH6YVL+3PRwA5qh9GIRqPh66+/5sILL1Q7FK8iH0mOk3TTuYkCe3NNrDntUSrPnqV2NG2O75rtdHa03jfUZKeMKjia/PixnF76X0maWqlv/m2f+522VpI4HQen3UXaJvev9tqemasdrLMNYtfU13Aktu/m4BblcDArs5PaUZywYSWy2GxTFI2WPxNuJGXvVRRaPbuxtWg5y/cUq7qmkzg+kjgdh8ztJdjMbWddHG+Wk6ewssctFE16AJfeR+1w2oTuy7PQtMKO+SCfQHrlblM7DK/j8g1jbsQTzEyVrY1aO6dL4ect3reVkKIo0k3XBEmcjkPqOumm8ySHzcWW0g5svuhlLANGqR1Oq6fk5DG5qofaYRy3of4d0SoutcPwKuaIU5jseor5WYlqhyJayPebvS9xEk2TxKmZHHYn6Vukm04NpUUOVoVOJnvq07gCQtQOp1U7b0vra71LtjnVDsGrZHSYyND8e9lQEah2KKIFrUsvpbBSlmVpDSRxaqaMrSXYLfICrhZFgd15gawd/QzVI6epHU6r5b9qKx0cwWqHcVxSCvaqHYJXULQGvu9wFyP2XEaVQwbLtzUuBX70wu460ZgkTs0ks+m8Q02lgzUMZ8+0l3HEdVY7nNbHbmdWThe1o2i2GN9IkookcXL6R/N/wU9zy55T1Q5FuNHirflqhyCaQRKnZrDbnKRvUWdZfNG0zFwtq/vdS+kFd6BoZePS43HKytYz9TnZFKN2CKqrihrM+dYn+DgvVu1QhJutzyijola28/J2kjg1Q8aWEhxW6abzNjaLk40VXdl6yavYeqWoHU6roaRnM7G6m9phNEtyba3aIahqe8JUhubcxs5qP7VDER7gcCn8uVt6N7xdiydO+fn53HLLLXTu3Bmj0UhCQgITJkzgjz/+aLE6kpKSmDdvXouVdyx71stK1t6sqMDJyrgZ5E15DJdJNoJtjonbfdUOoVlScneoHYIqFL0vH8X9h/NSJ2B2Sotqe/L7DkmcvF2LjjBMT09n+PDhhISEMHfuXPr164fdbueXX37hpptuYufOnS1ZnUfYrU4ypJvO67mcCjsKw8g59wV6F/yE38pv1A7JqwWv2E7UYH8KtTVqh3JEXQI6EJm2Uu0wPM4R1JHblbv5YV+E2qEIFfy1qxC704VBJx1C3qpFfzI33ngjGo2GNWvWMHnyZLp3706fPn248847Wb16NQCZmZlccMEFBAQEEBQUxJQpUygoONiis3fvXi644AKio6MJCAhgyJAhDXZrHjlyJBkZGdxxxx1oNBo0Gk1L3kIj6VuKcdhlDZnWorLMzmqfsaRNewFnRLza4XgtxWLhmpzuaodxVMn69rd9SGnsGYyueoQfiiRpaq8qLQ7WppWqHYY4ihZLnEpLS1m8eDE33XQT/v6Nu0tCQkLqVyEtLS3lr7/+4rfffmPv3r1ceuml9edVV1dz3nnn8fvvv/Pvv/8ybtw4JkyYQGZmJgCLFi2iQ4cOPProo+Tl5ZGX597pm5nbpLWpNUrL9eGfoQ9Sft5stUPxWgP+KVI7hKNKqa5QOwSPUdCwNuFqhqRfT6bZpHY4QmXSXefdWixx2rNnD4qi0LNnzyOe8/vvv7N582Y+/vhjTj31VJKTk/nggw/466+/WLt2LQD9+/fn+uuvp2/fvnTr1o3HH3+czp0789133wEQFhaGTqcjMDCQmJgYYmLcO+sme2eZW8sX7mOpdbKh9hR2TJ2PvcsAtcPxPqnpnF3jnUs66DQ6hmRtVjsMj1CMgcyPfoRLUsfgVKR7RsAfO2VcrTdrsb9SRanbBOtoXWc7duwgISGBhISE+mO9e/cmJCSEHTvqBoHW1NRw77331h8PCAhg586d9S1OnlSWX0N1mWy82Nrl5blY2fl6Cic/iMvHqHY4XmXSziC1Q2hSn8AkAiyVaofhdrbQbszUP83cjNYxy1F4RkZJLXsKq9QOQxxBiyVO3bp1Q6PR1CdATVEUpcnE6tDj99xzD1999RVPPPEEy5YtY+PGjfTt2xebzdZSoTabtDa1HU67i63FMWya8BLmwePUDsdrhC/fTqjL+2bYJWva/vT7/PizGV7yX/4qaX9jucSxLU+VLb68VYslTmFhYYwbN45XX32VmprGM3XKy8vp3bs3mZmZZGVl1R/fvn07FRUV9OrVC4Bly5Yxc+ZMLrroIvr27UtMTAzp6ekNyvLx8cHpdP+6Slk7ZIBeW1NW4mBV4EQypz6LK1gG4Cq1tVxT4H0b/w4rb7tjPBSNjqUJN5KydyZFNoPH67dkbaXwyzlkvzqDjGfGU7t7VcP4FIXy5R+R/eoMMv83ifyP78dWlHHMcl2Wakp+nU/2K9PJeO4ict66AfPetfXPV29bSvZrM8l68TLKli5ocK2jooCcN6/DZW3f63YdatU+GV/rrVq0Q/21117D6XQydOhQvvrqK1JTU9mxYwcvvfQSw4YNY8yYMfTr14/LL7+cDRs2sGbNGmbMmMGIESMYPHgwAF27dmXRokVs3LiRTZs2MW3aNFyuhrPakpKS+Pvvv8nJyaG42D1ZuculkJta7payhcoU2JPnx5ozn6BqzEy1o1Hd4H+8q2XVV2eifxsd3+TyDePp8Ce4KvV01WJQbBYMUZ0JG3NDk89X/vMVlWu/IWzMDcTMeB6dfyiFnz941KRGcdop+OxBnBUFRFz4APHXvkH4ubegCwwHwFlbQenilwkddTVRUx6leusf1B6SVJX88hqhI2aiNbb9lsbm+ietFJdLUTsM0YQWTZw6derEhg0bGDVqFHfddRennHIKY8eO5Y8//mD+/PloNBq++eYbQkNDOfPMMxkzZgydO3fms88+qy/jhRdeIDQ0lNNOO40JEyYwbtw4Bg0a1KCeRx99lPT0dLp06UJkZGRL3kK9woxKrLUOt5QtvENtlYO1jiHsnvYq9o7e1+riKZodexlhTlQ7jHoDAxPxcba9sYW1EX2Z7HqKN7I7qhqHb5fBhJ45Hb8epzV6TlEUqtZ9S/CwS/HrcRo+kUlEnH8nLruVmh1/HbHM6s2/4bJUETnpv5g69EYfHIWpQx98ouomHzjK89EY/fDvdSbG2O6YOvbDXlw3brVm+59odPom42nPymvtbM9r++P8WqMW32I7NjaWV155hVdeeaXJ5zt27Mi33357xOuTkpJYsmRJg2M33XRTg8cpKSls2rTp5IM9iuwd3vUpXLhPdi4U9Lqd3qemEfbd82ic7S9hnrI7jL/6H7s7xhOSnS3+sqS6tA4XMjF9ElUO7743R0UBzpoyfDsNrD+m0RswJZyCNWcHgQPObfK62j3/YIzrSelv86lN/QedXxD+vUcSlHwxGq0OfVg8it2KrWAvuqAobHm7Ceg7Bqe5ivJlHxE99UlP3WKrsmpvCafEB6sdhjiMzH09gpzdkji1J3ari01liWye9CqWfmeoHY7HRS3bSaDLO2YcJhd7fgatuyg6H76Nv4tRe6Z4fdIE4Kyue93T+oU0OK7zD6l/rimO8gJqdq1AcbmIuuQRgoddSuWar6lY9Xnd9aYAIs6/g+Ifnif//TvxP2U0vp1PpWzpOwSeOh5HRQG5795K7js3UrNzudvur7WRcU7eyfv/klXgcikUpEkTaXtUUuRgVcRldL/sXGJ/eAZtO1mEUamq4uqiU3kx2r0tuccS7BNEr/TtqsbQUpz+Mfyfzz18ujdW7VCO3+GznxWl8bEGz7vQ+YUQfs7NaLQ6jDFdcVaXUrlmESHDpwLg1/00/Lof7I6zZG7GXpRB2NgbyH3zOiIm3IPOP5S89+/ElHAKOv8QN9xY67ImrRSH04Vetl/xKvLTaEJxVhV2q/tn7QnvpLhgV34w686aS/WIS499QRuRsrZa7RAY6t8BrdL6tziqjBrCuZYn+DSvdSVNuoC6pRFcNQ1bl5y1FUdNZHQBYRjC4tBoD25IbAhPwFlThuK0Nzpfcdgp/XU+YeNuwlGWh+JyYurYF0N4Bwxh8VjzdrXMDbVy1VYHW3Lax4e31kQSpybIbDoBUF3hYI3mTPZOewlHtPcMnnYX3ZZdJFvV3d8vxdb6k6ZtCVMZkn0bu2u8b32sY9EHR6PzD8Wc/m/9McVpx5K1FWN8ryNeZ4zvhb0sD+WQpNdeloMuIAyNrvGSC+UrP8XU+VSMMV3rPqm4Dn5QVVwOcLX+34OWsnqfLIvjbSRxakLeHsnwxUEZuTpWD3qAsvG3orh5U2m1TUuNUrX+5PxUVes/GYrelw9i/4/zUydgdXnvS6vLZsZWsA9bwT6gbkC4rWAfjspCNBoNgYMvoGLVF9TuXomtKJ3iH+ehNRjx7zWivoziH/5H2V8L6x8HDjwPl6WKst/fxF6aQ+3etVSs+oLAgec3qt9WlEHtzr8JOf0KAPRhHUCjpWrTr9TuXYu9JBufWFlJ/YCNWTLe1tvIGKcm5O0tVzsE4WVsZif/0oPoKfPptvEdfHatPfZFrVDc8lRMffRYNJ6fWRjrG0li2nqP19sS7EGJ3KbcxU9p3r+oqi0/lYJP/lP/uGzJ2wD4n3IWEeffQVDyxSgOK6W/zsdpqcYY14OoKY82WGPJUVkEmoPJoT4okugpj1L6x9tULbgZfWA4QYMnEpR8cYO6FUWh9JdXCB19LVqfus2MtQYj4efdTulv81GcdsLG3oA+0Pu/j56yMatc7RDEYTTKgU3mBFC3P93Hj/yjdhjCi2l1GnqGFxL97TNorGa1w2lxS68ZxPxIzy9AeWFoXx7b8KPH6z1ZJbFnckHeVWRbvGNWomh7Vj9wFjHBJrXDEPtJi9NhpJuuzp7czfy+6TMyi1OprC3h2rPn0L/TwdWOFUXhp/Xvs2LHj5itVSRG9eLS028lNizpiGXmlabzw7qFZBXtprS6gIuH3ciofg0/ka5N/Z1v/3kbm8PCsB7nctGw6+ufK6nK55Uf7+XeSfPx9fFv8XtuLpdTYXthJDnnv0iv3O/xW/29arG4wxkbLMxXYTu/lNrWtd2Ggoa1CVczdc8onIr3ds2J1m9jVjnnBMeoHYbYT/7aD1OQLssQAFgdZuLDuzBl+C1NPv/7pk9ZuvlLpgy/hXsmvUaQXygv/3gvFtuR3/xsDgsRgbFMTL6GIL+wRs9Xmyv4+K//cVHK9dx03tP8s/tXtmasrn/+s2XzuGDotaomTYeqKLWz2nQO6dOexxXWdl7U9P/uoJ8t2uP1Jue0nmUIFGMgr0Q9ypTUsyRpEm63Kbtc7RDEIeQv/jCluepPyfYGfTomM2Ho1Qzo3HgxSEVRWLplEeMGTWNA5zOIC+vE9FH3YXdYWLfnjyOWmRjVk4uGXc/grqPRaxvPtCmuysPk48+pXUeRGNWT7nEDyC+rW816beof6LSGJuNR275cI6uHzaHinOvUDqVlKApX7vPs7LquAR2IqCrwaJ0nyhranSt1z/C/zC5qhyLaiY2Z5WqHIA4hidNhSnNr1A7B65VU5VFZW0rPDoPrjxl0PnSN7c++gm0nXG5UcDx2h5Ws4lRqLJVkFO0iLrwzNZZKfly3kCmnN9365Q0sNQ7WW/qzc+pr2DudonY4Jy1x2R58FN2xT2whKYbGLZDeKDf+HE4v+T/+Lg1ROxTRjmzNqZANf72IjHE6RFWpBZtFFr48lsrauumxgb6hDY4H+oZSWn3irQZ+xkCmj7qP95c+g91hZWj3sfROGMKHfz7LiFMupKQynzcWP4jT5eC8wTMY2HnEsQv1sNw8hcKuN9FrUBbh3z6H1mFTO6QT4iou5YrSASwI3+qR+pIrvXvKtaLRsaTDbGalyka0wvOqrA72FlXTLTpQ7VAEkjg1UJonrU3HQ8PhaxopTRw7Pv07nd5gEPru3I3klqYxZfgtPPLpDK466/8I8gvj2a9vomtsv0bJmzdw2F1sKYkn7MKX6bn3c0z/Hrn70puN2uhgwVnur0ev0TMky/Oz+JrL5RvO0/738GZqR7VDcQunuZLct2cTO+N59MGeG9tWtuQdFJeDsDHXH/tkwZacCkmcvIQkTocozZHEqTmC/OqSlUpzKcH+4fXHq8zlBB62OejJsDttfL7sRa4c/QBFlTm4XE66xfUHICq4A+kFO+ib5L0tAKXFDlaFTKLr1HHE//A02qrWtQKwcd12ep4ZxU5DsVvr6ROYiL91n1vrOFG1Ef24oupmNmQHqB2K21Su+gLfLkMbJE3VW36ncu032Etz0Jr88e8xnLCxswGwl2RT8uur2IuzcFlr0AeE4dd7JCHDp6LR1b2lFP/4AjVbG39gMIR3JO6a1wAISr6YnDevJXDwBRhC2s7kCnfZXSDjb72FJE6HKM2TX8zmCA+MJcgvjJ3Z60mIqFvh1+G0sydvExckX9ti9Sxe/yG9Ow4lIbI7WcWpuJSD3ahOlwNXK9jTTFEgNc+f3FFP0bt6GYFLPlQ7pOZzubgqvSP3dXNv4pSi8Tv2SSrY1+EiJqRPosbhubFenuayW6ne/CtRlzxSf6xyzddUrv2a0FFX4xPbA8Vhw1GRf/AinZ6APqPxiemC1hiArTCNksUvg+IidMSVAISNuY7QETPrL1FcTvLevQW/nsMPFuMfgm/SQKo3/kzoyKvcfautXmpBldohiP0kcTqEDAw/yGo3U1SRU/+4pCqf7OI9+BkDCQuMZlTfSfz678dEBXcgMjieX/79GIPexOCuB/t23l/yNMH+EVyQfA1Ql1wdmCXncDkorykmu3gPRoMvkcENZ3HllaazYe+f3D/5DQCiQzqi0WhYufMngnzDKCjPJDGqh7u/DS2mptLBWoaRMC2FTn/NQ5+zR+2QmqXz8nR03TQ4cd/A1OSy/GOf5EGKzodvY27l9j2D1A7F7Sz71oNWV78PndNSTfmyD4m8+EF8kwbUn+cTeXCvRkNITIMWIn1wFJbMzVizD04M0Rr9wXhw2ZDa3atwWaoJ6Du2Qf2+XZMpX/ahJE7NsLtQEidvIYnTfoqiUJrfuhbgc6eMol289P1d9Y8XrZoPQHL3s5k+6j7G9L8Mm8PGZ8tfpNZaRVJUL24+/xlMPgdbD0qr6/a+OqCitoSnvzo4nuGPzZ/zx+bP6Rrbn9snPl9/XFEUPvn7eSadNhujoW6jVB+9kStG3svny1/C4bQzZfgthPhHuu3+3SUrV0NB37vpNTiVsO/noXF592QEJb+Qy8r78lHIDreU76szMSBzi1vKPhHOgFj+Y7iHz/a2j64jS9ZWfGIO7gtnSfsXRXHhrC4h560bUGxmjPG9CB09C31Q039v9rJcLGkb8O1+5G7z6s2/YkoagD644V6IxtjuOKuKcFQUNnpONJRdZqbW5sDPR9621SZbruxXWWzmg/+uUjsM0Y5ERunpvu09jNtWqh3KUZmH9+fKM098mYmjGR7Sk9f//dUtZR+vyuihTC6+nt01vmqH4jGFix5Hawok4rzbAKhY/QXlyz5CHxJD2FnXojX6U77sAxxVJcRd/TIa3cH11/I/uBtrwV5w2gnofw5h425Eo2m8wo2jupSc12YSMeEe/Hs1XIfNZa0la94Uoqc+haljX/febBvw7U3D6Z8QonYY7Z6s47SfdNMJTysqdLAq+gpyL30cl6/3Dj72/WcbnR3umb2Y4vSO8UNbEy5nSNat7SppAlDsVjR6n0MOKOByEDbmOnw7n4oxvicRE+/FUZaLJaPhzMeIC+4jduaLREy4B/PetVSuWdRkHTVbfkdrCsCve0qj5w7UrditLXdTbdhuGefkFSRx2q9EVgwXKnC5FHYWhLJ+3PPUnH7xsS9Qg8PB1Rmd3FJ0clGGW8ptLsXgx/ux/2V86vlYXe3v5VDrF4TLcvC1T+dflyAbwg8uvaDzC0brG4SjsqjBtfqgSHwiOuLfewQhI2dSsfwTlMO6nhVFoXrLb/j3GdWgteoAl6VqfxzBLXZPbVlqobxPeYP290pxBLKGk1BTVbmdf/Sj2TftRZyRHdQOp5EeK7LQtHCnfohPMD3z3DN2qjnsQYncaHqah9J6qxaD2nyiumAvyax/bOxQ972wl2bXH3Oaq3CZK48+BklRUFyORoetWVtwlOUR0G9sExeBrSgDtHoMEW1zjayWJi1O3kESp/2kq054g/RcPauH/Jfy829SO5QGlJw8Jle17CzGof4d0Lhxtt7RlMSOYFTlI/xcFKFK/d7Ct/Mg7MWZOPe3OhnC4vHtlkLZH29iyd6BrSidkh+fxxDWAVPHfgBUb1tKzY5l2IuzsJfnU7NzOeV/vYd/zzPQaBt2vVZv/g2f2B74RCY1Wb81exumhN5oDUa33mdbsa9I3qe8gQzPBxSXQpnMqBNewlrrZAO9ibnsdbqtewPDnn/VDgmA87b48MXwY5/XXMnWxi0U7qagYU3CLKbtGYlTkc+NPpFJ+MR0pXbnMgIHnAtAxPl3UvrHWxR9+QhotBg7nkLUlDn1i1tqtDoq//kSe1kuKAr6oCgCB51P0JALG5TtstZQu2sloWcdeW23mh1/EzJ8mrtur83JqzDjdCnotCe3Q4M4OTKrDigvqOWjh1erHYZXq7ZU8PhnV3PPpFcJD/TcVO1Fq17H6XJwyfCbPVanN9HpNfQMzSfq27lobBZ1gzEYuPP2ILL1FS1S3E+VWhJK0lukrOZQjEG8EnIP/8vo4rE6WwPz3rWULV1A7KxXm5wV5y61e9dSvnQBsVe/0qilShzZivtHEx/SviYxeBtpcULGNzXHr/9+Qt/ElAZJ0+pdi1my+UsKK7Lx9QlgYOczmXL6rfXP55Ts44sVL5NRuBM/YyCn9x7POYOm16/ttDdvC9/+8xb55ZnYHVbCAqMZ3ms8o/tNri9j7IBLeeST6YzqezERQbGeu2Ev4XQobCuKJmfCS/TK+gbfNT+pF4zdzqycLsxJ3HDSRcX7RZOQtrYFgmoea2gPZllvZ3mGDEI+nG+XIdjLcnFWlRxxrSZ3UGwWws+7XZKm45RdWiuJk8okcQKqSlT+JO/lbA4rq3b+zOxzn6w/9sfmL1iy6QsuTLmepKheOJw2iitz658322p45cd76R43gHsmvUZheTYf/jkXH72Js/pPAcDHYOLMUy4kPqwzPgYTe/O28OmyefjoTZzeezwAgb6h9OwwmOXbv+fClOs8e+NepLzEzir/8+k6bTTxP89FV1aoShynrMyFxGOfdyzJRs8tdpgbfw4TsqZSYms8q0vUCRp8gcfrPHxNJ9E8WWVmktUOop2TTn6gttKmdghebXvmGnRaHZ1j+gBQa63ih7XvMmPU/QzpdhaRwXHEhiU12HB3XeofOJw2rhh1L3FhnRjQ+QzOHjiNJZu/5EDvcEJENwZ3HU1sWBLhgTEM7T6WXh0Gsze/4UrSfROHsX7vUs/dsLdSYE+uL2uHP0bl2KvVCSE9m4nV3Y594jEk17q/lVfR6Pi9w82ctneGJE2izcguk/G4apPECaitksTpaPbkbaZjZPf6xzuz16MoLspri3nss6v474eX8s5vj1JWfbAVJK1gO11j+2PQHVxcr1fCYCpqSyipanpvsqziVPYVbKNbbL8Gx5OielJWXUhpVUEL31nrVFvtYJ39VHZNexVHx54er3/i9pPrJtCgITnbPSuRH+DyjeDxsCe5Zs+RtwERojXKLjOrHUK7J4kTYJbE6ahKq/MJ9guvf1xcmYeiKPz678dcfNqNzBr7MLXWKl758V4cTjsAleZSAn0brjZ94HFlbWmD4//98FJuf+sc5i66kTP7XMBpvc5v8Hywf92U8SMlXO1VTi6s7HkbxZPuR9F5rtc9eMV2olz+xz7xCLoGJBBeXXTsE09QbUR/Jjmf5J2cBLfVIYRapMVJfZI4AWbpqjsqm8OG/pCWI0Vx4XQ5mHzazfROGEKn6N7MPOv/KKzIYXfuxvrzNI1mzCr7jzd84vaJ87hn0mtcdsbtLN3yFev2LGnwvI/OuD8OGYt2OIfNxebSBDZPehVr/5EeqVOxWLgmp/uxTzyCZL37BmjvS5jEkPy72VjpvVvYCHEyskqlxUltkjghY5yOJcAUTK3t4FL/Qftbn2JCD44SDvQNIcAUVN9dF+QbRmVtWYNyqszl+89t2BIVERRLfHhnhvc6n9H9JvPTuvcaPF9jrayLwzekRe6nLSopcrAy/BKyL3sKV4D7Z44N+OfEW4yGVZYd+6TjpOh8WBR/D6NTJ1PjkFlaou0qrJIPkGqTxAkwV9nVDsGrdYjoSn7ZwT3FDgwSLyzPqj9WY6mk2lJJWEA0AJ2ie7Mnb3N91x3Azux1BPuFH3UdKEVRGlwDkFeajk6rJzY0qSVup81SXLA7P4h1Z82leuRU91aWms7ZNZ2P+zK9Rs/g7M3HPvE4OAPiuDfwae7cO7BFyxXCG9mdChVmec9SU7tPnKy1dpwOl9pheLVeHQaTV5ZOrbVun6TokAT6JZ3GlytfZV/+NnJL0/hg6TNEhyTQPW4AAIO7jkavM/DBn3PJLU1jU9pyfvn3E0b3m1zfVffX1m/Ykr6SwopsCiuyWbVzMX9s/oIh3cY0qH9P/ha6xPTFRy/bMjRHdYWDNZzO3mkv4Yx1z+a8AJN2Bh33NX2DkvCzttxGpZXRyZxjfowv8j23KKsQaiutkV4SNbX7dZyktenY4sM70zGiOxv2/snpvScAMH3U/Sxa+Rrzf/4PGo2GbrH9uem8p9HtH6Tsawzg5vPn8vnyl5i7aDZ+xkBG953M6H6X1JeroPDdmncoqcpHq9URERTLBUOvYfj+NZwOWL9nCecNvtJzN9xGZOTqyOt/H72H7CT0+xfRtPAmAeHLtxM60JcybfPHXCTTcgv3bUm4nMl7z8Xqavef/0Q7U1JtpVPEiU/QECen3W+5kptaztf/O/mVkNu6bZn/8PWq1/nPlHfQenBbhq0Zq/lm9Rs8cMnb6GSF4RMWFa2j++Z38dnxT4uWu3bmYJ6N3djs8xc6Izk1c/1J1akY/FgYcRdz0nqdVDlCtFZvTD+VcX2klVUt7f6jmgwMb54+HZMZ3ns8FTXFHq3X5rBwxch7JWk6SYUFTlbGXUn+pY/iMrXcJ9XB/5Q3+1xfvS/9TnJ8kz04idmmZyRpEu1aSbW8b6mp3SdOsoZT843qezGhAZ7bKgNgUJeRJEXLm2RLcDkVtheEs+HcF6gdNrFFytTs2MMZlo7NOvfUgEQMrhPvGi+OG8nIiodZXBR+7JOFaMNKa6xqh9CutfvESVqcRHtTWWZntXEc6dNewBl+8hsnX7areYlMivPEXm4UNKxKuJYhadeSY5EJAkKUyOBwVbX7xElanER7tS/Xh39SHqHi3BtOqpyoZTsJdB07oUkuzDzushVjEC9FPcbU1FEoSqMVVYVol6SrTl3tPnGSFifRnllqHKw392XH1PnYO/c79gVNUKqquKro6HvmhRlD6JG/47jKtYb14ArtM7yQefzrRQnRllVaZDa4mtp94iQtTkJAXp6LVV1mUzj5v7h8jr87bNi6o6/NNMQvHg3Nn8CbE38upxX/hxVl7l8FXYjWxmxzqh1Cu9buE6daWcdJCAAcdhdbi2PZNPFlLIPGHte1us27GGqNP+LzKVZHs8pRNDp+63ALw/dOp8RmOK4YhGgvzHZJnNTU7hMnh1V+AYU4VFmxnVXBF5I1dS6uoObPYLs89cgzLpPzdh3zepdvBI+HPcm1e4Y1u04h2qNaaXFSVbtPnNr5+p9CNElRIDXPnzUjnqTqrOat2h63PBWT0ngzgni/aBJKjz4wvCZyABc6n+SdnIQTileI9kS66tQliZNsUyfEEdVWOVjrHMruaa/i6NDtqOcqZeVcVdy70fEU49HX/tqTcDFD8+5ic2XAScUqRHtRa2te17dwD0mcpMVJiGPKzoVVfe6k+MJ7UI6yivsZGyyNjqXU1DR5rqIz8lX8PYxJvZgah6wML0RzyRgndUni5JLESYjmsFtdbC5PYsvk17D2PaPJc/T/7qCfLbr+sQYNQ7O3NjrPGRDHPYFPc9fegW6LV4i2ymJ34ZL3LtVI4iS/e0Icl+JCB6sip5J72RO4/IMaPqkoXLnv4Oy67oEdCTtsf8OK6BTOMT/Gl/nRCCFOjLQ6qUcSJ8nahThuLpfCzvwQ1o15jpozLmnwXOKyPfgodV1vybqG6zBtTriCwVm3kFrj67FYhWiL7E4ZoKuWdp84uaTJSYgTVl1h5x/dSPZOexFnVN1mv67iUq4orduYObmyBADF4M+C2AeZmHoedpdsnSLEyZLP/Opp94kTkrQLcdIycvWsPvU/lI2/GUWjYdRGB3qtnsFZm7EHd+J64zM8mtZL7TCFaDPkQ7962n3iJL98QrQMq9nJv9W92H7pfDQVJi7U9aEmYihnVjzMr8VhaocnRJsi713qabxaXXsjv3tCnDCNRsHH147Bx4Hex45Ob8OltbEneQRDfXTc1HkwcV00xKkdqBBtjU66vNXSrhMnmc4pRB2dwYmPyY7ex4HBYEOrs6PVWEFjAcWKy2nG6TDjtJuxW2uwW2qw1lZjt1gwH1aW0eTP+FNvZr11KxcGzOc1rmePWf7WhGhJGn277zBSTbtOnGTxS9GmaBR8jA4MJjsGHzs6nR2NzlaXAGFBcVnqEiC7GYetFrulBpu5BmttDS6ng6aXqTw+4eEJjOk8A0exnW3lRZzVqZCHrFfwS/BcPqmIlSGFQrQQnUZanNTSvhMnaXESXkhncGIwOvAx2tHp7Wh1NnRaG+xPgFxOCy5HbX3rj81Sg622BpvFjEXFDwPduiRzqt8YlGIHxQk2XEUuiot7EBDwL+eV30w//wt5zXklGVZJn4Q4WXpJnFTTvhMnyZuE2ygYTA4MRgcGgx29wY5Wa0WjtQI2FJf5YPeXrRa7pRabuRqruQaL3a528MctZeAkEqu7o1TX7aGVb6oAYOsWA8NPD8LprKRDzTfM0f7FD8HP8FVFuAwvFOIkyCZF6mnfiZO0OIlj0Opd+BjtGIyOuuRHZ6tLgDTWurE/LjMuR10CZLfu7/6qrcZqrlW19cdjNBrOPW02QbmBHDrTIttcCIDdDlptMk7nbwAYXGVcVH4dAwOm8Yr9EvJs0vokxInQSouTatp34tQO3tcEKCj4GPcPfjY40Bnqur40Whv1Y39cZlx2Mw77gdafusHPTruNWrVvwEuZTAGMP/UmdLkNj9t8XRSUFNY/3rc3loSODc9Jqv6YJ3W/81XQXH6oDPRAtEK0LTKpTj3tOnHSy6yEVkWjc+FjcmDwsWMwONDqrWi1trrWH6wozrruL8chg5+ttdXYzLVYXdKy0ZIiI5M4K/EKlNzG3YpFUVaUgoOfStLTFbp164rFuqfBeT7OQqZWzOTUoGt41XI+xXb5GQnRXDI4XD3tOnHSGbRo9RpcDml68iS9jwMfY926P3ofe13yo7WiwYqiWFCcFpyOWpy2g4OfrTXVOGzWRlPfhed17zqMQabRKCVNj8XK8ylvdKy6uj96w57GJwPdK9/maf2vfBr0FL9XmloyVCHaJA2SOKmpXSdOAD4mPZbq1jcYV21NLXzYoPVn/9gfh71u8LPNun/ml7kGi1N29W6tThs0mYSqrig1jiOek1WV3+jYli3+DDrVB0WxNXmNryOTqyouZ0jwbbxWO5IKh7Q+CXEkIXoZGq4mSZx823fi1JILH4q2S6PRcu5pNxKY68/Rlts3B7ooKS9tdLy6WsHHZwhW64qj1nNKxYs8a/iZDwIfZVmV4WTDFqJNCjW0+7duVbX7776PqQ1k7l6w8KFou0y+QYwfeCO63GN3aRdGmiG36edyczoRHnH0xAnA376bG+yXMTTkAV6vHkKNU7rShThUqKENvG+1YpI4mbznW9BaFz4UbVdUVCdGJ1yOkte8Vtk8bdkRn9uxQ8Po0THY7I278poyqPwpnjX2513Tf1lbIxM5hDhAWpzU1e6/+z6+Lf0taF8LH4q2q2e30xngcyZKafN/LzMr8o74nKKA3TEY+KHZ5QVbN3GbdQqrQ+bwdtUpWGTtNSGkxUllkjj5Nv0LKAsfivZs+KmX0qGiE0pt8wfy14Q6qayqPOo5O3eE06OnhqONkzqcBoVh5Q/Rw5TCm7p72CILa4l2LkxanFTV7r/7AcEZhEaskoUPhaBuEPj5p92Ef64fx5PcABSE10L20c8pLHTRf0B/LJaNxx1bmGU192ku5a/gJ1lY2RW7fEAR7VSYvt2/dauq3X/3FWc5ean/qh2GEKrz8wvm/AE3os09saUAcpSSZp1XUtITf/+NJ1SHRnEwsvxeevuO5nXNzewyS/Ik2h/pqlNXux9x6RcUrHYIQqguJqYrE3vdgjbvxJImRaOQVXrk8U2H2rrFgE53ctusRJmX8H/my5kRnCObnYp2RwaHq0sSp2BJnET71rvHmYwMm3Jcg8APVxnhpNbcvM5tm61u49+TpVPMjCu/laf93qWTqd2/lIl2RFqc1NXuX218pcVJtGNnDJlKX+dpKOaTW829IPj4VgNL2xd3UvUdKq7mBx62zOTS4CJkEwrRHsjgcHW1+8RJuupEe6TV6phw+m3EFXeEFlhgMsdVfFznp6UpGI1dTrreAwxKBRPLb+AJ/8+I92n3L2uijZPESV3t/hVGWpxEe+PvH8rFKffil9MyG+q6tApZxUdYLvwoamsGtEj9h0qs/pzH7NdyQXBFi5cthLeQrjp1SeIUEIhG0+6/DaKdiIvtzoQeN53wIPCmlEc7sNma3rz3aLZs9Uejafn96IyuYqaUX82cgB+INMjftmhb/HRajFr5vVZTu//ua7RaTIEnN8NHiNbglF4jOTN0MkpZy65Qnx9QdULXVVXWbfzrLl2r3uVp582MC5IV2UTbEaqX1ia1tfvECSA4MkrtEIRwqxFDr6CPLeWkB4E3JdtWdMLX5uV2asFIGjM5c5hRMZ3/C1xCqF5e7kTrFy7jm1QnryRAaGy82iEI4RY6nZ6Jp99OTFE8uGGfN4deIae4ees3NWX7dg0Gg/s/uPSufJVnuJsRgcffpdheuGprqHrlWYouO5eCc1IovflK7Du3HfF828Z1FIwe2OjLkZlWf4513WqKZ1xA4YQzqHj6QZRD9uN0VVdRPOMCnAUn/vvTHiX6GtUOod2T1BVJnETbFBAQynl9Z6PJaflWpgNKY+w4i0+8fEXR4HQM5Xg2/j1R/va9XGefytDge3i95jSqnC03zqstqHzuURxpewh+4HG0EZFYfvuJsntuIHzBV+iO0iof/t43aPz96x9rg0MBUFwuKp78D/5Tr8I4+DTK59yD+cdF+F14KQDVb72I74TJ6KJj3XtjbUxXP0mc1CYtTkBIbMutKSOEN4iP68n4bjehyXdf0gSQ73v0TX2bY+fOMPDgCkwDKp5lru5BkgPc+71pTRSrBevffxB4/e349D8VfXxHAmbegC4mDvN3Xxz1Wm1oGLqwiPovja5uDI5SUY5SXobfBVPQd+qC8bQRODL2AWDbuhH7ru34TZrm9ntrazpL4qQ6SZyAMGlxEm1Iv95ncUbwJJTylh0E3pQsc8FJl1FQoGAy9W2BaJovyLqVW6ou5dbgzfhqZdlMxekElxN8fBoc1xiN2LYefS/Pkusuo2jyWMruuh7bv2sPXhsSijY8AuvaVShWC/bNG9B37oZit1M170mC7vi/+iRLNF8XSZxUJ4kTECotTqKNGDl0Br2sQ1As7m9NsZtc5JecfOIEUFbaq0XKOR4aFJLL5/Cs/kkG+rfvzYK1fv4Yevej5oO3cBYXojidmH/7EfuOrbhKml7cVBsWQeCdDxLyyHMEz3kOXUIiZXdfj23TegA0Gg3BD82l5sO3KL7qYvTdeuJ77gXUfLIAn4FD0fiYKL1lJsUzLqT26089ebutWhcZ46Q6jaIo7fsVY7/Xr59OTXmZ2mEIcUJ0Oj3jh92CKcfn2Ce3kLyOFn4sXNEiZRmNGoadtgins7pFyjteikbPsuDHeLeyB7Z2+pLoyMmi8tlHsG/eAFod+m490SckYk/dQcS7i5pVRtl/bgMNhD7xYtN1ZGVQ/p9bCHvzU8pum4Xf5GkYhwyneNZkQp99HUOX7i15S21OhEHP1tNPUTuMdk9anPaTAeKitQoMjODiIfd6NGkCyDO23OrcVqvSIhv/niiN4uDM8gd41jiPXr7ts+tOH59A2Lx3iPpxJRGf/Uz4/A9RHA50Mc1/bTT07oszJ6vJ5xRFofL5xwi44U5wuXDs2YnpzDFoQ8Pw6Xcq9v0tVeLIZGC4d5DEaT/prhOtUUJ8H87vcgOaAs8PdM6qyW/R8tLT1P8bjDD/zQPmacwMzkTfPvMnNL6+6MIjcVVVYlu7EuPwkc2+1rFnJ9qwiCafM//0NdqgYEzDR4Krbkaj4nDUPel0oLhkluOxyMBw7yCJ037S4iRam/59zmZ44AUoFe4fBH44q7+LotLj29j3WPbtA6PRvQtiNodOsTC2/A6eMb1FF1P7eYm0rl2Jdc0KnHk5WNetpuzOa9ElJOF7zkQAqt56iYqn/lt/fs2XH2FZvhRHdgaOtL1UvfUS1r//qF9u4FCuslJqPnybwJvvA0AbGIQusRO1X32MbdsmbBvW4NOnv2dutBXr4tcy+0uKkyPrOO0niZNoTUanzCSyKBrFpc6U+sIoC7hh3cLa2oHodGnHPtEDYmoX85BmOYuD5/JpRTRtfeSTUlNN9Vsv4ywuQBsYjPGMswiYdRMafd1+gq7SYpyFh7QyOuxUv/4CzuJCNEYj+qQuhDz5EsaUMxqVXfnKXPynzGiwHlTwvY9S8cxD1H79CX6XzsDQS8buHIsMDPcOMjh8v5LsLBbeNVvtMFodi93BL1t3sSWngGqrlfiQIC4Y2IeOYSFHvGZFajor9qRTWmsm1M+Xs3p1ZXBSh/rnd+cXsWjDNqqtVvrERXPJ4H7odXWf/M02Oy/+voLrRyQT6u/r7tvzOjqdz/5B4Op+5vmnWw5bsna2eLnBwVr6D/gYRfF8K9rRZAVM5lXH5WRZpTtJqGfZ0J5085dWJ7W1n3boYwiJiUEjO04fty/WbWZ3QTFTk/tz99ln0j06kjf/+oeKWkuT56/ck8FPW3Zxdp/u3DNuBGf36c7XG7ayLbduWrtLUfjon40M69KRm0efRlZpOf/sy6y//sfNOxnWpWO7TJqCg6KYPOQe1ZMmgKyKlh3fdEBFhQujz6luKftkJFR/yWO2WUwKlpm3Qh06DSRJi5NXkExhP53eQJBs9ntc7A4nW7LzOb9fT7pEhhMR6M+4U7oT5u/Hyr0ZTV6zPiOblC4dGdAxjvAAPwZ2jGNopwSW7twLQI3VRo3VxmldE4kJDqR3XDQFlXVT1NOKS8kuq+CMbuqPg/G0jh1O4dzO10KBQ+1QqA12Ul5Z7rby8/K7uq3sk2FwlXJx+TU8HvAN0T7y0ik8q6PJB4Ms1uoV5K//EDLO6fg4FQWXomA4bPVfg05LWnFpk9c4XC4Mh7XsGXQ6skrLcbpcBBh9CDIZ2Z1fjN3hJK24lNiQQBxOF4vWb+XiU09B285ePAb2PYfT/CegVKifNAEUhJvdWv62rWAwRLq1jpPRqeoDnnTcyHlB6qw5JdonGRjuPSRxOoQsSXB8TAY9ieEh/LY9lQqzBZdLYX1GNpkl5VRZrE1e0yM6kn/SssgurUBRFLJKy1mTloXTpVBjtaHRaJg+bBC/bU9l7i9/ER8SzNBOCSzZuZeuUREYdDpe+WMlz/z8J8tT0z17wyo4a9jVdK/pj2LznrE1udqmk+KWoiganM4hbq3jZJmceVxecSUPBv5GmF5eRoX7ycBw76H+YAkvEh7fUe0QWp2pyQP4fO1mHvv+D7QaDfGhQQzsGEd2edObv47t3Y0qi5WX/qhbcTrA5MPgpA78uWsfGk1dS1KnyDBuH3t6/TVFVdVsyMjmjrFn8NrSVZzRvRM9YiJ57pe/6RwZRlxIkPtv1MP0eh/Gp9yKMcf79vLKKst1ex27d0XStZvbqzlpPStf5xn9Yj4OfJKlVfLGJtynq7/8fnkLSZwOEduth9ohtDoRAf7cOGoYVocDq91BkK+JD1ZtIOwIg7cNeh2XDu3P5MF9qbJYCTKZWL0vE6Nej7+x8crXiqLwxbotTOjfGwWFnPJK+nWIxUevo0tkGPuKStpc4hQcHM05Pa+BHO/omjtUZbiT6poat9eTl+filL59sVi2uL2uk+XnSOeaymkMDb6T+TVnUOn0ntbB4+GqKKf4qkmEv/YhuhjPtb5XzX8exWEn6Jb7PFZna3RqkL/aIYj9JHE6RGTHJHx8/bCZa9UOpdUx6vUY9XpqbXZ25Rcxvt/RN23VabWE+NUlVxuzcukdF4VW03js0pq0LPx9fOgTH02trW6KutPlAnQ4XQquNraYRmLH/gwLHY9S6H1JE0BBaDW4P28CoKysN76+3p84HdCv4nme9fmJ93znsLK69b201ny8AOOwM+uTpoLRAxudE3j7f/CbeAkA1Qtfp+b9NxoXZDIR/dMqAJwlRVTPfx777h04czLxu2gqgTff0+B0/8tmUnzFBPwnX4FOxpk2KVCnpacsQ+A1Wt9ftxtptFpiu/UgY/O/aofSauzKL0JRFCIDAyipruGHzTuJDAxgSKe6dZl+2ryTCrOFqckDgLput8zScjqGhWK22fl79z7yK6q4bGjjVYOrLFZ+376Hm0efBoCfj4GooACWpabRPTqSPYXFnNXLO2dgnYhB/c6ju7U/SqV3Jk0AuS73jm861JbNPgwb5ofT1Xo+yATYdnKT7VKSQ/7LG1WDqG0lmb1itWD++RtCnnq5wfGge+fgM/S0+sda/4D6//tdOgPfiZMbnF921/UYevY5eMBuRxMSiv8Vs6j98qMm69aGhuEzOIXa778k8LrbWuBu2p5Tg/yb/GAp1CGJ02HiuveSxOk4mO12ft68i3KzBT8fA307xHDuKT3Q7Z85V2mxUlZ7cBaWS1H4a1caRVVb0Gm1dIkM5+bRpxHm79eo7G//3c6IHp0JPmQ2yWVD+vPpmo0sT01nRI8udAwPcfs9esLYYdcQVhCOonhvN49Lq5BV4v7xTQdYraDTp+C0LfFYnS1lcPnjdDMO5B3D/7G+xvvf8Kz/rACdrtG2J5qAQHRH2HtO6+sHvgf/bu17d+HM2EfQHf9Xf0wXE0fQzfcCYP752yPWbzxtBNXvzpfE6QgGB0s3nTeRxOkw8T16qx1CqzIgIY4BCUceD3F4S1J0UCB3nt14S4amXDGscVdBx/AQ7j135HHF6M0MBhPjk2/BJ8f7Z2ZVRDqwVDS9sKm7pKfFE9dKe2+Crf9yu/VSVoXM4Z3K3li9eJMG++YNGJp47at66Wkqn3sU3f+3d9/hUVXb38C/Z3qmJpn03jsQ0oDQSwhVQAQRKZFyFeyoYFe8NuCHvteCBVEQvWJHQECUKtK5QAglSE1I78mkTKac94/IQEghhMycKevzPHlIZs45e02YzKzZe+29vX3gNHI8nMZMbHOh4PpffwbfLxCi7gm33b4wKg7G4kIYCvMtWl9lK5IpcbIqlDjdxDsiEgyPRzt1E7NzdvZGesQsqywCb02RUgNUWbbNCxeA4JBAaLWtL6hq7XgwoG/li4h26otPeAuQZaWjjoaifPDUzdfOkj0wH6KEFDBiCRr/dxA1H78DY1Ul5NPntjifbWxEw/YtkN33QKfa57l5mOKgxKk5PgMkKlv2yBPuUOJ0E5HECW4BQSi5fJHrUIgdCw7qiV7KUWBLbCNpAoCrulJO2q2vSwCPb5uJ0zWu9X9hEXMYu1RvYU11EPRW1vnEarXgiZpPd78xQRKGNc04rl27stXEqeHP7WDr6iAZPqZT7TPiprbZBsv2aNqCKJkEcoH1LUviyKx/fIADvpHtzwgj5E4k9xiLXqIRYGtsJ2ky8FnklRVw0nZWlgIMY/uf8XhsI4ZUPoVlko8Q7mRddU88lTPYmtbXXrtGGNMdbK0GhvKyFvc1bF4Pce/+bdZD3YqxuqltnrNLp863Z8kq+a0PIhZFiVMrfKjOiZhJeuq/EFITA1ZnW0PB5Z466HQ6TtquqmIhFlvfxr+d5VH3B16qn4ZpqgKreQEWhEVBf6X9Xnb932cBkRg8uaLZ7YaCPDQePwynUeM73b7+8nlAIIAgKLTT17BXyTRMZ3Ws5e/WqvhGUI8T6VoioRPu7rcQzgUugJUN03REoaz93gizt2+lG/92Fp+tw8jKR/C29EsEirl/GRYl94H+8kUY/+l10u7bjbpNP0F/6Tz0ebmo+/UnaD7/EE5j7gYjar5Qbf2W9eC5ukGU0rfVa+vOZ0N3PhtsfR2MVRXQnc+G/vKF5sdkHoOoWwIYMa1VdDOaUWd9bL//2wyU7h6Qu6qhaaVLmlxXq23E0q278fiwvq0uJ2AuG4+fhsHIYnxC7K0PtgKurr5IC8uwmSLw1uRpSzht/1QWD4MGq6HT2dffpG/tL1jM24NNqiX4sUrNWU4tDAmHMDIaDbu2QTr2HkAgQP2G76D5aDlY1giBtx/kGfPgNH5ys/NYoxH1v22E04i7wPBbr8Mp/9cU0/f6c2fQsH0LeJ7ecP9ms+n2hh1bIc94yDwPzoZ5igQIpD3qrA4lTm3wiYzBuf1/ch2GVdtx5jxifTxMSdPT3/3a4pi7E+KQGhZo+vl4bj52nLmAkhoNZGIx+oYFYnDU9e7588Vl+HjXgRbXWThiIDyUTWP9g6JC8fbmnegfEQy13Lq7sUODk5AsT7epIvCb6UUs8ksLOY3BaASMxl4ANt/yWFsjNFZgQuW/0FM+FR/oJqGAow2dZdPmouaTd+E0+m6IU/pC3EYP0o0YHg/u325t9xjPHe2vi6c98CfA50E8cNhtxesIqLfJOlHi1AbfiChKnNqh0xtw6FIuZvdPaXb7vcndEel1fVqzk1Bo+v5MQTH+e+A4xveMRaSXO4qqa/D9kZMQ8vnoFx7U7DqLRg6EWHD96SkXX//UpZCIEeHpjv0XrmBMD+sdVk2JH4/gumiwGttNmgCgzEsLYzH3NVnnst0Ral8jds0Eaf6LN/l/4EflUmyqVtz6hC4m7t0fhrxcGEuLwffwsli7bEM9VAsXg+HT29HNkml/OqvE/eC6laIC8fadLSwGj8dDkFvzWTBOIiGUThLTl/CGabT/u5KHOF9PpIYFQi2XIsbHE4OjQrHz7AWwNy0OKBeLm12Hx2s+CynG1xPHcy23ivVtYRiMSH0IwdWRgI0VgbcmX8xtfdM1+flGSCS2MTzbWSJDMe6rysArii1wF1r+5Vk6capFkyYAkAwaDmF0N4u2aStSqMfJKlGK3waPoBAIxRLotLSuSGsulpTDz0XV4vaf/3cK3x3OhKtMipRgf/QODTDtsaQ3GJslUgAg5PNQVd+Airr6ZnVS7/z+J/QGIzyVcgyLCUOYR/NpzgGuKlTWNaC8ts6i9VW3IhJJMTb5UQisNKfrjKt1RVyHYFJZGQuJ5BTXYZhdRPVneEuwDeuUb+GPaiqYdkQSHoNuCut5bSPXUY9TG3h8PrzCIrgOw2qV19ZDdVPR4oi4CEzvk4AHB/ZCfIAPNp44gx1nzpvuj/Ryx8mrhfi7qBRGlkVJjQZ/nrsEAKiu1wIAlBIx7knshpmpiZiZmgh3hQyf7DqICyXNi4JVTk1vJhW19bAWarU/7u6xwK6SpkYnI4rKirkOw+RkpgQ8nmO8mTjpc/BA1f14TrkbKgG9VDuaRKUMQp51rfdFmlCPUzuC4xOReyqT6zCsks5ggIDX/JPwsJhw0/e+//RG/XH6b9PtvUL8Uaqpxaq9h2E0shALBegfHoRtp/429Up5KOWmInAACHJzQWVdA3ZnX0Sou9p0u/CfGTyNBoN5HuBtCg/thUTpMLCl3Kx1ZC4lng0At3XhzTQ0sBAIeqGxcSfXoVhMXNV7WCbcirWK1/BnjfDWJxC7MEyt5DoE0gZKnNoRltIHe77+guswrJJMLEL9LRZEDFQ7o0GnR02DFgqJGAzDYEyPaIzqFoWaBi1kYhH+Lm7axsNF5tTudf53Ja/ZbXWNTW3LxaLWTrGo3j3vRqAmwuaLwFuTL7Tw5nQdkHPFD17eXEdhWTLdOTykm4IU1bP4pDYFGoMNLgZGbkuaGyVO1or6f9vh4uUDtV8A12FYJV9nJYqqNe0ek1dRDQGfBydh8/ycx2Ogkkog4PNwPCcfgWpnKCRtr1WSV1kNhVPz3q3CqhrweQy8lJaffWTCMBiZOg+BleGwus3HukhuDTfbrLTn778Bscgx/y4Tqt7GMv6rSJbZ/qQD0rYgJxHCpFTbZq0ocbqFsOTeXIdglSK93FFYVWPq+TmVX4QDF3JQUFWDUk0tDl7MwdasbPQOCYDgn2G1Wm0j9p2/guJqDfIqqrD+2CmcuFqAcfHXZ0rtOXcJWXmFKKmpRWFVDTZnnsXJq4Xod8NaUEBTcXqwm2uLYnNLkUjkuCd1IZQF9vupsEFhRHllBddhtKq+IYHrEDij1Gbicc1kPKI6BQnVwNglGqazbjRUdwthSb1x8OfvuA7D6ng7K+HvqsKJ3Hz0CQ0En2Gw78IVbDxxGkYWUMulSI+NaLb4JQAcvXIVmzLPgGWBILUz5g3qjQC1s+l+g9GIjSfOoKq+AUI+H15KOWb3T0a0t0ez6xzPzcfwWG6K993dgjA0aBrYfPuqZ7pZkXs9YKWF7qeylOjenQ8W1lHjZmkMWPSpfBmRkt5YyX8GmXVcR0S6Upq65YxlYj0Y9uYFdEgzLMvi04cfgKaslOtQrM6ZgmJsPHEGT6cPMBV3W8Lp/CJsyjyLp4b3B59n2U7TiLDeSJAMBVtrf/VMN9sflotTV89xHUab0tLOokF7mOswOMcyAuxRvYkvqsOgo5dzmyfj83CmXxxEFn5tIx1H/zO3wDAMQhN7cR2GVYr29kCfkABU11t2ratGgwH3Jne3eNLUJ+Ee9GQGOUTSBAA5ldZX33SjoqLwWx/kABhWj4GVC7FM/D6inGjoztYNdFFQ0mTl6H+nA6jOqW39I4LhLG17Rpw5xPv7IFDtcusDuwjD8DCq78MIqAi12yLwm2lcDajR1HAdRrtOneJDIHDlOgyr4V6/E8/X34+ZqqvgpvKPdIV0N+sepnv11VcRHx/f5v2rV6+Gs7PzHbWRkZGB8ePH39E1zIkSpw7wj+kGsYyWvndEEiclJvZZCEW+/NYH25Eil1quQ7glgwEAS73BN+Kz9Rhe+Tjeln6BYAm9vNsaAQOkm3kZgn379oHP52PEiBFmbccWDBo0CE888cRtn0d/WR3AFwgQHJ/EdRjEwjw8gjE+9jHwCxyjl+lG+SjnOoQOOXfO/dYHOSCf2k14pSED96pKQIN3tiPVWQ5noXnnbH3++ed49NFHsXfvXuTk5Ji1LXtFiVMHhSX34ToEYkFR4f0wxP0+sGX2PXOuNSzDIqfcSqfT3SQvj4VEQhtyt0bIVuGuyofwpmwdfEX0Um8LRrs7m/X6tbW1+O677zBv3jyMGTMGq1evbnb/rl27wDAMtm/fjqSkJEilUqSmpiI7O7vNa166dAlhYWGYN28ejMbW1xfbuHEjEhMTIZFIEBISgsWLF0Ovv3Wt6OLFi+Hh4QGlUokHH3wQjY2Npvu0Wi0ee+wxeHh4QCKRoF+/fjh8uPlkkd27dyMlJQVisRje3t549tlnTe1mZGRg9+7d+M9//gOGYcAwDC5fvnzLmABKnDosuGci+ELa7sAR9E28Fz3YfmDrHHOqe7WbAfX11rMH4K1UVcbe+iAHFqD5Hv/WzcU4lfWtAk+u4wEYaeb6pm+//RaRkZGIjIzEtGnT8MUXX6C1ifUvvPACli9fjiNHjkAgEGDWrFmtXi8rKwt9+/bFpEmT8NFHH4HXSlH7b7/9hmnTpuGxxx7D6dOn8cknn2D16tV444032o11+/btOHPmDHbu3IlvvvkGP//8MxYvXmy6f+HChfjxxx+xZs0a/O9//0NYWBjS09NRXt7UW56Xl4dRo0YhOTkZJ06cwEcffYRVq1bh9ddfBwD85z//QZ8+fTB37lwUFBSgoKAA/v7+Hfo9UuLUQSKJEwK7xXMdBjEjhuFhTN9H4VceBDjwlhaFqvZXhLc2J7OcwONZdoKCrREbSzG5chYWyzfCQ0gv+9YoWSWDh9i8H85XrVqFadOmAQBGjBgBjUaD7du3tzjujTfewMCBAxETE4Nnn30W+/btQ0ND89nT+/fvx8CBA7FgwQK89dZbbbb5xhtv4Nlnn8XMmTMREhKCtLQ0/Pvf/8Ynn3zSbqwikQiff/45YmNjMXr0aLz22mt47733YDQaUVtbi48++gjLli3DyJEjERMTg5UrV8LJyQmrVq0CAKxYsQL+/v744IMPEBUVhfHjx2Px4sVYvnw5jEYjVCoVRCIRpFIpvLy84OXlBT6/Y9Mq6C/oNoQm0ew6eyWVKnFPn0WQ5Uu5DoVzeYYyrkO4LfV1LIQCKhLviLCa1XjL8AjSlbRiprUZ5W7e3qbs7GwcOnQIU6ZMAQAIBALce++9+Pzzz1sc2717d9P33t5NG0MWFxebbsvJycGwYcPw4osv4umnn2633aNHj+K1116DXC43fV3r5amra/t52KNHD0il11+P+/TpA41Gg9zcXFy4cAE6nQ59+/Y13S8UCpGSkoIzZ84AAM6cOYM+ffqAuWGNwb59+0Kj0eDq1avtxnwrtHL4bQhL6oU/Vn4IlqV9ouyJl2coBvndB7bA8eqZbmbks7haZhv1TTfKyfGHpxfXUdgGiSEPM6qmI1n5MD6sH4YKPb2ecY0BMMrM9U2rVq2CXq+Hr6+v6TaWZSEUClFRUQEXl+tLvAhvKEu5lnjcWL/k7u4OHx8frFu3DrNnz4ZS2fZMQKPRiMWLF+Puu+9ucZ9Ecvv78TEMYxpeZG5aeJllWdNtN35/4/2tnXe7qMfpNkhVzvCJjOI6DNKFYiIHYJD6XrDllDQBQIWHvlkBpq04dw4QiTpWn0CaRFd/iCV4GgMVtvf/bW/6OsvhLxGZ7fp6vR5ffvklli9fjuPHj5u+Tpw4gcDAQHz99de3dT0nJyds2rQJEokE6enpqKlpe823hIQEZGdnIywsrMVXazVR15w4caJZreWBAwcgl8vh5+eHsLAwiEQi7N2713S/TqfDkSNHEB0dDQCIiYnBvn37mtVw7du3DwqFwpQ8ikQiGAy3X8tKidNtiu43iOsQSBfpn3wfuhlSwdY7ZhF4awrl1r3oZXsatY678W9nyXQX8K/q+/CM8gAUfHo74Mr9PmqzXn/Tpk2oqKjA7NmzERcX1+zrnnvuMdUF3Q6ZTIZff/0VAoEAI0eOhEbTem3kyy+/jC+//BKvvvoqTp06hTNnzuDbb7/Fiy++2O71GxsbMXv2bJw+fRpbtmzBK6+8gkceeQQ8Hg8ymQzz5s3DM888g61bt+L06dOYO3cu6urqMHv2bADA/PnzkZubi0cffRRnz57FL7/8gldeeQULFiwwJWxBQUE4ePAgLl++jNLS0jZnBd6M/lJuU1TfgRCIxFyHQe4Aj8fHmL6Pw6c0wKGLwFtztbH41gdZqVOnVGBozexOia9ahmW8F9FLTh8iLM1FwDd7fdOqVaswbNgwqFQt25k4cSKOHz+O//3vf7d9Xblcji1btoBlWYwaNQq1tS0Xzk1PT8emTZvw+++/Izk5Gb1798Y777yDwMDAVq543dChQxEeHo4BAwZg8uTJGDt2LF599VXT/W+//TYmTpyI6dOnIyEhAefPn8dvv/1mGnL09fXF5s2bcejQIfTo0QMPPfQQZs+e3Sxhe/rpp8Hn8xETEwN3d/cOr2tFm/x2wpYPluP0nzu5DoN0gkzmglE9HgKvgOo6bmYQsvhStLtTXdfWIm34GTQ0HOE6DJvFgsFh55fwaU086o301mAJc/3c8O9wP67DILeBepw6IW5wGtchkE7w8Y7A2MiHKWlqQ5lno00nTQBQTBv/3hEGLFIqX8MywZvoKaPEyRKmept3mI50PUqcOsEvphucPb25DoPchtioQRjgcg/YCioCb0uBtJrrEO7YqVMCCASW2wDaXrloj2CBZgoeUp2D+A5nIJG2JSqliJbTGmS2hhKnTmAYhnqdbMjAlPsRp+tNReC3cLXOduubrtHrQRv/dhEe9Ohf+RyWit9FDC1vZhb3U2+TTaLEqZNiBw4F085USsI9Ho+Pu/o9Aa8SP4DqNdrVKGFRWFbEdRhd4u+/PbkOwa641f+JZ+vuxwOqHAio86nLyPk8jPN05joM0gn0zt9Jclc1guMTuQ6DtEEud8HEXovglEczIDui1FPb6p5VtujqVSMkElpvrSvx2QYMq3wSSyQrESah7KkrjPdwgayDW3wQ60KJ0x3oPmwk1yGQVvj6RGFM+MPgFdLQXEcViCq5DqFLVVd34zoEu+RVtxUvNczAfaoievO4Q1N9XLkOgXSSzT33MzIyMH78+A4ff/nyZTAMg+PHj3d5LCE9k6B0p2EBa9Itegj6q+4GW0lF4LfjqsY+humuOZnpBB7v9rdzILcmYDUYUzkfb8m+hr/Y5t5CrEKMTIIEpYzrMEgndfpZX1xcjAcffBABAQEQi8Xw8vJCeno69u/f35XxWTWGx0OPNOp1shaDUmYgpjEFbAP1NN2OBrkRJRWlXIfRperqWAgFKVyHYdf8ND/h342zMUFVwXUoNmeqmVcKJ+bV6cRp4sSJOHHiBNasWYNz585hw4YNGDRoEMrLy7syPqsXNzgN/Bs2RCSWx+cLMK7fk/As8aYi8E4odq+/9UE2KDc3gOsQ7J7QWI57KufgdfnP8BJR71NHSHgM7vGkJTNsWaee6ZWVldi7dy+WLFmCwYMHIzAwECkpKXjuuecwevRoAMA777yDbt26QSaTwd/fH/Pnz2+2l83q1avh7OyM3377DdHR0ZDL5RgxYgQKCgpMxxgMBixYsADOzs5Qq9VYuHBhiwLWrVu3ol+/fqZjxowZgwsXLnTmYXWKVKlCZO9+FmuPNKdQuGFi8kJI8sy3Qaa9KxBUch2CWWRnAyKR760PJHcsuOYrvKl/CKOUre9XRq4b7e4MZ6GA6zDIHehU4iSXyyGXy7F+/XpotdrWL8zj4b333kNWVhbWrFmDHTt2YOHChc2Oqaurw//93/9h7dq12LNnD3JycvD000+b7l++fDk+//xzrFq1Cnv37kV5eTl+/vnnZteora3FggULcPjwYWzfvh08Hg8TJkzo8GZ9XaHH8NEWa4tc5+cbg9FhD4EpoqG5O5FTVXDrg2wSg8bGJK6DcBhiQxHur5qJlxS/Qy2k3qe20NpNtq/Te9X9+OOPmDt3Lurr65GQkICBAwdiypQp6N69e6vHf//995g3bx5KS5tqKVavXo0HHngA58+fR2hoKABgxYoVeO2111BYWAgA8PHxweOPP45FixYBAPR6PYKDg5GYmIj169e32k5JSQk8PDxw8uRJxMXF4fLlywgODsaxY8cQHx/fmYfaIWuffRzFlyzX0+XoesSmIdqQDFZLSdOdqHU24psG+913Ua1mEBP7FQDaZseS6gRB+Eb6JnbU0HIgN0pQSrE5MYLrMMgduqMap/z8fGzYsAHp6enYtWsXEhISsHr1agDAzp07kZaWBl9fXygUCsyYMQNlZWXNdk+WSqWmpAkAvL29UVzctHpxVVUVCgoK0KdPH9P9AoEASUnNP0FeuHABU6dORUhICJRKJYKDgwGgw7scd5WEkXdZtD1HNqTXTEQ1JFDS1AWK1XVch2BWZWUsJJKenLVfV2fEig9LMfW+HIwaeQmPPZqHs2cb2jx+6ZJiDBt6scXX7Fm5pmOOHqnDzBm5uOuuS1jydjF0uuuffTUaI2bOyEVRkd6sj+tWpPrLmF09Fc8q90LFp96na54IpFnY9uCOntESiQRpaWl4+eWXsW/fPmRkZOCVV17BlStXMGrUKMTFxeHHH3/E0aNH8eGHHwIAdLrr08SFNxVVMwxz24vwjR07FmVlZVi5ciUOHjyIgwcPAgAaGxvv5KHdtuh+g2hpAjPj80UY128B3Iu9qAOhi+ShjOsQzK60JJKztpcvL8HRo/V49jl3rPzMD4lJTli4sAClJa0nNvMfdsN33weYvr5ZFwCFgocBA5umrhuNLN56qxhjxyrw3nu+yM7WYvOv1/cY/GxlGcaOVcDT0zpqaLpVvYulvOeQKuc2kbMGMTIJ0tRKrsMgXaBLPwrExMSgtrYWR44cgV6vx/Lly9G7d29EREQgPz//tq6lUqng7e2NAwcOmG7T6/U4evSo6eeysjKcOXMGL774IoYOHYro6GhUVHAzNZbH56PXhEmctO0IVEoP3JP8DCR5NIOxK+VW2Gt903UnT/IhEDhbvF2t1og/99Ri7r/U6N7dCb6+Qsyc6QpvLyE2bGx9Q2W5nAdXV4Hp61y2FhqNESNGKAAAVVVGVFYacdc4JYKCROjTR4orV5o+jGZlNeDcOS0m3K2y2GPsCHnjWTxccy+eVP0PUp7jrjr+WKAnGNow2S50KnEqKyvDkCFD8NVXXyEzMxOXLl3C999/j6VLl2LcuHEIDQ2FXq/H+++/j4sXL2Lt2rX4+OOPb7udxx9/HG+//TZ+/vlnnD17FvPnz0dlZaXpfhcXF6jVanz66ac4f/48duzYgQULFnTmIXWJ2IHDoHBz56x9exXgF4eRIXMBjocf7E21mx61dbW3PtDG6fUAYPk1nQwGwGgERKLmb5YiEYOsrLaH6260ZUsNEhKc4OnZ9IHB2ZkHtZqPI0fqodUacfJkA0JCRNDpWPzn/5Xi8Sfcwedb55tzUuUb+D/Bv5Eoc7zu4lAnMe7ycOY6DNJFOj2rrlevXnj33XcxYMAAxMXF4aWXXsLcuXPxwQcfID4+Hu+88w6WLFmCuLg4fP3113jrrbduu52nnnoKM2bMQEZGBvr06QOFQoEJEyZcD57Hw7p163D06FHExcXhySefxLJlyzrzkLoEXyBAr/HU69SVenYbgVTZWLBVlDR1tSJn+0+arjl/3svibUqlPMTEiPHVVxUoLdXDYGDxx+81OHtWi/KyW9fnlZXpcehQHUaOUphuYxgGL77kia++qsTsWVcRFibGiJEKrPumEj0TJBCLGTz+WB4yZuZi/foqcz68TlFpj+EJzRQ8rDoDsQP1vjwS6AGeAz1ee9fpWXWkdQa9Dqse+xdqykq4DsXmDe09C27F7gA9Q81iV8hFnM+/xHUYFpOWdhAN2nMWbTM/X4f/W1aCzMwG8HhAeLgYfn5C/P23Fp9/4d/uuf/9bwV++L4K334XCKGw7Tfdq7mNeOGFQnz8iR+efCIfd09UITlZirlzcrF0qTdCQq1zZlu5U198wluALPuenwA/iRAHesVA4MDDlPaGpjt0Mb5AiORxE7kOw6YJBCKM7/cU3IooaTIXI49Fbtnt1R3aupqa1pdKMScfHyHeedcHGzcF4Zt1AfhwhS/0BhZe3u0Xb7Msi9+21mBYmrzdpIllWbzzbikefEgNoxE4f74RAwbI4OLCR/fuTjiR2bEhQS641v+FRXX3YbbqEoR23BvzcIAnJU12hhInM+g2JB1yV1rkrDNUKk9MTHwG4jzrmBVkryo9DG0uXmuvTp6UgsfjpvfFyYkHtVqAmhoDjhyuR2pq+xu8njjRgLw8PUaObH8W1pbNNVAqeUhNlcH4z3ZD+n9GtfV6FhZcB7hTeGwjhlQ+jaWSFQh3sr/kwkMkwH1erlyHQboYJU5mIBAKkXwX9TrdrsCAHhgZNAcotnw9k0Zbh1f/eA+9P5qEsOXDMH7tPBwvONPuOav/9xMGr5yGsOXDMHDl/fgha2uz+/dcOowBn05FzLsj8eSvb6DRcH0pjmqtBgM+nYq86iKzPJ5bKVLUcNIul2prWQiFli0SP3y4DocO1aGgQIejR+rw9FMF8PcXmmbJffZZOd5+u7jFeVu31CAqWozg4La3EqqoMODrryvxyMNuAACFgo+AACF++qkKp0814NixesTGWucw3c086v7AS/XTME2Vb1dvSg/5e0BC61jZHfofNZPuQ0dA5kKfNDoqofso9HEaBbaamyLwZ7YuwZ+Xj+D/jXkBv89ajQHByZi6bgEKalqvVfvy2Hos2f0pnuz3ALbP/hIL+s7Ci7+/i9/P/wUAMLJGPLbp35gWPw4/T1uB4/ln8N8TG03nv7XrY0yLHwdfJTdrf+XpHLMG72puoEXbq6014v33SjHrgVwsWVKCuDgJ3l7iDYGgqXelvEyP4ps+KGg0Rvz5Zy1GjlS0dkmTFR+WYtJkFdzcr/fOLlzkjl07NXjhhUJMnuyMqChJ1z8oM+GzdRhZ+SiWSFcjUGz7b00uAj5m+tDIgz2i4nAzOvrrL9j15Uquw7B6aX3mwLVIzVk9U71Oi+h3R2DVxDcxNPT6SvXpX8zC0NA+WDhgbotzxq+dhyS/bnhx8HzTba/+8R4yC7Px07QPUVpbgZ4fjMPfT/0OiUCMN3d9hNrGerwxfAEOXz2JV7a/h43TPwafx7fIY7yRQcBirWQP9HpHnKnIYuiw7WhstP/1q2yZjueCTcol+LFKbbNljs8EeeGpYMvP5iTmZ/tpvRXrnjYCUpUz12FYLaFQggn9noFrIXdJEwAYjAYYWAPE/ObDIhKBGIevnmz1nEaDruXxQjGOF5yBzqCHWuoMD7kaey4dRr1Oi0O5mYj2CEWjQYfnty3HW8Of4iRpAoAKT52DJk0AbfxrG4TGCkyo/Bdel/0IH5HtvU0p+DzM8XPjOgxiJrb3jLQhQpEYyWPv5joMq+Ts7I27E56CKI/7p6BcLEWiTyz+s28NCmtKYTAa8NOpbTiWfxrFta1vSTIgOAXrMjchszAbLMviRMFZfJu5GTqjHuX1lWAYBh+NW4z/7PsSQ1dNR6xnOO7tNhorDnyNvoGJkAjEmPDVfAxceT9WH/3Roo+3QNr6qtWO4sxpF9BLn20I0vwXb+gfxFilbT1nM3zdoBLSBBd7RUN1ZqbTNmDlI7NRX219i9FxJTioJ3opR4GtsZ5ej8sVeXh6y9s4mHsCfIaPOK9whLj4I6voHHbMWdvi+HqdFi/+/i5+OvUbWBZwk7ng7tg0fHTwGxx75Be4yVxanHOxPBczf1iIrRmrcM9/H8XspEkYFNwLwz6fiW/ufRfRHqEtzjGHLf6nkVfi2ENVacOz0NBwjOswyG04p5yDFQ2jUaKz7qmCKgEf+3pFQy2ixMle0ccuMxOKJUgaM+HWBzqIpO5j0Es0wqqSJgAIcvHFD1PfR/aTv+Hg/O+xacan0Bn18Fd5t3q8k1CM5aOexbkFv2P/Q9/i4Lzv4afyhlwkhau05V5hLMti0dZleGnwwzCyLLKK/sboyEFwk7mgt388DuQeN/MjbKITsygo42YmnzUpLY3iOgRymyKqP8NbxseRpqznOpR2LQjypKTJzlHiZAE908dQrROA4X3+hVBNLFgr/sQoFTnBU+6GyoYa7Ll0GMPD+7V7vJAvgLfSA3weHxvObMfQ0FTwmJZ/Vusyf4WLkxLDw/vBwDY9fp1Rb/rXYLz1FhxdocxDC6O1L+5jAVknBRDwaad6W+Okz0FG1TQ8r9wNlcD63r7CpGLM8qX9Su2d9T3z7JBQIkH/+2ZyHQZnREIn3N1vIVwKXax2JfBdFw9h58WDyKnMx55Lh3HvN48jxNUfk7uNAgC8vfsTPLHpDdPxF8tz8dOpbbhUnotj+acx/5dXkV1yCYsGtpyBV1pbgff2fYnFwx4HADhLFAhXB2LV4e9xNC8Lf105ikTfOIs8zgInGjIGAJ0OYHi9zdpGVZUB90y8jMJC3a0P7kKffFyGDz4otWiblhZb9R6W4Rn0V1j2d3srr4b5QkirhNs96k+0kNhBw5C5fSsK/s7mOhSLcnX1RVpYBpBnXUNzN6vRavD2nk9RWFMCZ4kCIyMHYuGAuRDym/5EijRlzRarNBgN+PTQt7hQngMhT4A+gT2xftqKVof2Xtn+Hh5MuRfeiuufRJePeg4Lfn0Tnx/9EQ+l3IeePjHmf5AAcmtpmO6aC+e9EGDGZZ3WfVOJ3r1l8PISAgCGDb3Y4pjHn3DD2LFNPV9r1pRj7ZeVLY6RSBhs+jXY9PMv66vwyy/VKCzUw8NDgKn3O2P48OtrPk2+1xkzpudg4kQVvL2FXfyorIdMdx4P6aagl+pZfFybAo2B209lQ1wVGKamXkxHQMXhFlR08Ty+fn4BWNYxhkpCg5OQLE8Hq7HupMlRNEqN+NK4k+swrEpa2gE0aP/u8utqtUbcOzkHb77phZjYpkUohw29iGeecUdyipPpOJmMB/E/iz3W1xtRX9/8teGZpwsQGSnGwkUeAIANG6rx2coyPLnAHZGRYmSf1eKdd0rw/PMe6HPDNi6vvloIXx8h5v7LMRZgrBZ3xxeiF3FIw80SH0KGwY7kSITLbGfBUdJ5NFRnQZ4hYeg+LJ3rMCwiJX4ckoRplDRZkWIPx9qbriNqND3Mct1Dh+rB58OUNF0jl/Pg6iowfYlvWCHbyan5fRUVBly5omu2gvgfv9dg9BglBg+Ww8dHiMFD5Bg5UoF131Y2aye1jww7dmrM8tiskVKbicdq7sUjqixIOBgqe8DXjZImB0KJk4X1nTIDEoUdd+cyDNJTH0RwdRRgxUXgjqhAWMF1CFYn66QUDNP2fnCddTKzHhERLfeJe//9Utw94TLmz8/Dxo3Vpo15W7N5cw38/ITo1v16D5VOx0Ikap4YiMQ8ZJ/VQq+/fq3IKDFKig0oKrKuGiBzYsCiT+Ur+D/hEvSQWq5dVyEfTwVxs3US4QYlThbmJFeg/5QZXIdhFiKRFBNTF8K5wNlqi8AdWW51IdchWB2NhoVIlNzl1y0s1EOtbl5CmvGAC15+2RNLl3lj8CAZPvm4DP/9b2Wr5zc2stixXdNiv7qkJCm2bK7BuXNasCyL7Gwttm6pgV7fVIx+jZubwBSHo3FpOIhn6u7Fv1TnIWLM3/u0KNibFrt0MPS/zYFuQ4Yjc/tvKLrY9bUVXFGr/ZEWMhNsvuN8wrUl9Uojyquox6k1+XnBULv91aXX1Da27BmaNu36oqhhYU29UWu/qmh2+zV799airs6ItOHy5teY7ozyCj0efSQPLAu4uPCRni7Ht99WgXfDEJVY3PS9tsExP8EwrB4DKxchxmkwPmYexdl68/weYmQSTKONfB0O9ThxgOHxMHTWQ4AFPg1ZQnhICtJ8ZoAtpaTJWhW71XEdgtU6c4aBSNS1m7GqVHzUaNpfmys6RoK6WhYV5S17hbZsrkbv3lK4ujb/bCsW8/DMMx74dXMwvv5vAP77TQA8vYSQShmoVNdfzmuqm9pWOXNTLG0t3Ot34vn6+zFTdRXm+E28Fu4Lvp28jpOOo8SJI97hkYgbNIzrMO5Y7553I4E/xOaLwCvqqxD//l3IrbLsViT/3vEhXv7jP2ZvJ49HvU1tYVlA19i1w3VhYSLkXGn/g8T581qIRAxk8uZv6QUFOhw/3tBimO5GAgEDd3cB+HwGu3Zq0Ku3tFmP06XLOggEQFCQ/S5H0FF8th7DKx/H29IvECzpure80e4q9HNp+/+I2C8aquNQ/6kZOH9oPxpqbXD2C8NgZJ+HoCxQwh4Kmj7Y/xWGhaaa1mHyXzKgxTFvDn8K03uOM/288cwOfHDgK1wsz4Va6oyMhLvxUK/7TPfvzzmGyd883uI6O+esRZi6aQGheb2mot+nUzAnaRICnH26+mGZ5FY69t50t3LmjAuiohl01XM5KUmKVZ+Vo6bGAIWCj/37alFeYUBMjARiMYPjx+rx+eflGD1a0WJIb+vWGri68pGc0rLC+WpuI86e1SIqWgKNxoAfvq/CpUuNWLjIt9lxJ0/Wo1s3SbNZe47Op3YTXmV2Y7NqKb6r8rij/2kxj8HLoeb7eyXWjRInDkmVKqTeOw07Pv+Y61Bui0Qix5jEh8HP5zqSrlGv0+LbzF+xZtLSZrcvH/UcBgWnmH5WiK/Xm+y8cACPbfo3Xhv2BAYEJ+N82RUs3LIEEoEIGYkTm11n99yvIRddfxNUS51N37vJXDAgKBlfHf8Fzw+a18WPrIlGbUCNpsYs17YXJSUs4nv2QEPD8S65XkiICBERYuzeVYsxY5XgCxhs2FCNjz8qA8sCXt4CZMx0xbjxzWfYGo0stv1Wg/R0Bfj8lkNABiPw/Q9VuJpbCr4AiO/hhPfe9zEtsnnNzh21mDmzZe2UoxOwNbirch7i5ZPwgW4q8ho7N/P3QT93BDq1nDVJHAMlThzrkTYSJ3dsQ8nllqsKWyN3tyAMDZpmV0Xguy4eAJ/Hb7HtiVIsh4e89cLPH09tQ3p4f1MPVKCzD+b1mooVB/+LmQl3g7mh7kEtdYZK0naXflpYX/zfn6vMljgVutQCtWa5tF0pK4uCTHa8y643bboLPvm4DKNGK5CSIkVKKz1IN+PxGHyzru3lzAMDRfjkE792r3HgQB14PGDAQFm7xzmyAM33eJ23E+tVS/BLlfPtnSsR4XFafsChUeLEMR6Pj6EPPIR1ry5qKrawYhFhvZEgGQq2zH6SJgA4mHsC3b2iWtz+0u//Dwu3LoW/yhtTuo/G/fFjTRv4Nhoa4SRsvuCdRChGQU0JrlYXNtt6ZeTq2dDqGxHuFoTH+sxAamBCs/PivaORX1OMq1WF8FN1bZEyAOSz5V1+TQDQarXYuXMnzp49i9raWnh5eWHEiBHw9fVt85xDhw7h8OHDqKyshEqlQv/+/dGjx/VFKC9cuIDNmzejtrYWUVFRGDt2LPj8phqghoYGrFy5EjNmzIBKperyx3MyU4i+/ZQwGKq75Hq9ekmRd1WH0lIDPDws91Lb0GDEMwvdW+2xIteJjKWYXDkbCYoMfKgdh+IOrDvHAHgnyh8yvmMX3Ts6GgC3Ar5RMYjpP5jrMNrVJ+Ee9GQGga217SLw1uRWF8Lzpp6lp/vPxsfjF+Obe9/BXdFD8O+dH+L9/WtN9w8MTsGWc3uw9/JRGFkjLpbnYtXh7wEAxZoyAICHTI0l6c/gk/H/xqcTXkeIawCmrHsSB3KPN2vL65897HKrun6dJZZhkVtmnjHVjRs34uLFi5gwYQLmzZuH0NBQrF27FtXVrScehw8fxvbt2zFw4EDMnz8fgwYNwubNm5Gd3bR/I8uy+Omnn5CUlIRZs2YhLy8PR48eNZ3/xx9/ICkpySxJE9C08S+Pl3LrA2/D3RNVFk2aAGDQIDmio2kV644Kq1mNtwyPYITy1t2y033UVBBOqMfJWgycPhtXTh5HbYV5egc6jWEwKnU+FPly2EMReGsadFqI5c1Xj348dabp+1jPcADAf/5aY7p9ao+xuFKZh4wfF0FvMEAulmJ24j14568vTL1SoeoAhKoDTNdJ9I1DQXUxPjm0Dr394023SwTif+Jo6PLHVuWuR311fZdfV6fT4fTp05gyZQoCA5uGlgYNGoSzZ8/iyJEjGDJkSItzMjMzkZiYiLi4piFRFxcXXL16FX/99RciIyNRV1eHuro6JCcnQyAQICIiAiUlJQCAnJwc5OfnY9SoUV3+WG506aIP/PzN2gSxQhJDHqZXzUCSaj4+rEtDhb5l75OfRIhXqCCcgHqcrIZUqcKIhx63qrWdJE5K3NNn0T9Jk/1ylapQ1dD+zMYEn1jUNNaipLYpsWUYBs8PmofsJ3/D/nnf4X+PrEe8dzQANBumu1lPnxhcLr/a7LbKhup/4nC+g0fRuiKVeYqbjEYjWJaFQND8s5dQKEROTk6r5xgMhhbHCwQC5OXlwWAwQCqVQi6X48KFC9DpdMjJyYGnpycMBgN+/fVXjBkzBjyeeV+yLl1iIRaHmbUNYr2iq1ZgKRZgkKKxxX3vRgZAJqAhOkKJk1UJik9EzxFjuA4DAODhEYzxsY+BX2CfvUw3ivUIx99ll9s95lTROYgFIijFzZNIPo8Pb4U7RHwhfjmzHYk+sXCTtT2b6VTR3y0KzrNLLkHIEyDCLbjTj6EtV/WlXX5NABCLxfDz88OePXtQU1MDo9GIzMxMXL16FRpN60loaGgojh07hvz8fLAsi/z8fBw/fhxGoxF1dXVgGAaTJk3Cnj17sGLFCnh5eaFnz57Yu3cvgoODIRAI8Pnnn+ODDz7AoUOHzPK4AKC21jwb/xLbINVdwtzq+/CM8gAU/Ka3yOk+avR3pSE60oSG6qzMgKkPIDcrE6W5VziLISq8L+JFg+yuCLwtA0NSsGTPp6hsqIGzRIHfz/+FEk05EnxjIRGIsT/nGJbu+Qz39xgLsaBpSK+8rhK/Zu9Gn4B4aPWN+O7kZmzK3onv73vPdN3PDn8HP5U3It2C0GjQ4+fT27D53G58Ov7fzdo/dPUEUvy7w0nYtdObjXwWV0vNt2bEhAkTsGHDBrzzzjtgGAbe3t7o1q0bCgpaXzNqwIAB0Gg0WLVqFViWhVwuR48ePbBv3z5TT1JAQADmzp1rOqesrAyZmZl48MEH8cUXX6B3794ICwvDihUrEBgYCE/Prp/dlJUlQ8+eQrCsYzz/Seviq5ZhmSgWm5Wv0BAdaYYSJysjEIkw6tGn8fULC2DQWf6Fu2/iZPhVhYCts78i8LZEu4eiu1cUNp3dgWnx4yDgCfDlsfV4becHMLIsAlTeeKr/LMxMmNDsvB+ytuL1nSvAgkWiTyy+v+899PSJMd2vM+rx+s4VKNSUQCIQI8ItGGvuWYIhoX2aXeeXM9uxoO+sLn9cFZ466MrN9xxydXVFRkYGGhsbodVqoVAo8MMPP8DFpfUeN6FQiHHjxmHMmDGora2FXC7H0aNHIRKJIJW2nKrPsiw2btyI4cOHg2VZFBYWIiYmBkKhEEFBQbh8+bJZEqea6qaNf7XafV1+bWJbFI2n8XZgI+Q0REduQImTFXIPDEa/KTOwe+0qi7XJMDyMSn0Y8nwp7LUIvD2Pp87A6ztXYGqPsRgc0guDQ3q1e7yr1Bm/TP+o3WPm9ZqKeb2mtnvM9gv7wWd4GB018LZjvpUCWQ1ggbkGIpEIIpEI9fX1OH/+PNLS0to9ns/nQ6lsWvjx1KlTiIiIaLbu1TXHjh2DVCpFZGQk6uubCtwNBgOEQiEMBgNYMy7fUZAfAlc1JU6OLsB/Flxd+tz6QOJQqMbJSiWOHo+AbvEWaUsqVWJin0X/JE2OaUhoH9wffxcKa0os2m5dYz2Wj3oOAl7Xf4bJazTvYzl//jzOnz+PiooKXLhwAWvWrIGbmxvi4+MBNC0f8PPPP5uOvzbsVlZWhry8PPzwww8oLi7G0KFDW1y7trYWe/bswYgRIwAATk5OcHNzw4EDB5Cbm4tLly7B3998099On2YgFNIih45MLotEaOhTXIdBrBD1OFkphmEwYv4T+PKZR9Fgxu0yvDxDMcjvPrAFVM8xO2mSxdscG91y2n5XMAhZ5Jd2/bpQN9Jqtdi+fTuqq6vh5OSE6OhoDBkyxLRgpUajQVVVlel4o9GI/fv3o7S0FHw+H0FBQZg1axacnZ1bXHvr1q1ITU019UwBwPjx47F+/XocOnQIqamp7S60eadYFjDokwFsMlsbxHrxeCLExr4LHo+2VSEtMaw5+7vJHTt3YC82vvu2Wa4dEzkA3fn9wNYbzHJ9wp0ify02luzlOgyb5unJICJyLRxx6NrRhYU9h8CAOVyHQawUDdVZuYje/RA7cFiXX7d/8n3oZkilpMlOFUiqbn0QaVdREQuJpBvXYRALc3HpgwD/2VyHQawYJU42YMgD/4LKs2v2MOPx+BjT93H4lAYABvokba/y6ou5DsEulJdFcx0CsSCBwBkx0UtbnaxAyDWUONkAkZMUox55CswdrposlTpjYu+FkOXTPlb2rFHCorCMEqeukJUlBp9v3yvnkyYMw0e3uPcgkdCaTaR9lDjZCJ+IaPS++95On+/tFY67oh4Br+DWO4AT21bi2WDWqfqORKtlwee1vzQFsQ+hIU/B1bUv12EQG0CJkw3pPXEKvCOibvu82KhBGOg6CWwFzZxzBIUiqm/qSpcumW/2HrEOHh6jERj4INdhEBtBiZMN4fH4GPvks5A5t70X2s0GJE9FnL43FYE7kFyNeZchAIC6ujosW7YMlZWVZm/rRtu2bcOWLVss2ubFiyzE4hCLtkksRy6PQkz0Eq7DIDaE1nGyMQpXN9z11PP4bvFzMOjb3haFx+NjdJ9HIc0Xg6ZTO44GuRGlFWVmb2fv3r2IiIgwrcG0ePHiFseMHj0aSUlJpp9PnTqFP//8E2VlZZDJZEhOTkbfvteHRs6cOYMjR46gsLAQer0eHh4eGDhwIMLCwkzH9O3bF++99x569+7d5tYu5lBX1xN8/kWLtUcsQyBQoXu3j8DnO3EdCrEhlDjZIJ+IaAyZNQ+/f/p+q/fL5S4Y2W0eePnUy+Roit3rgdb32O0yOp0Ox44dw9SpzbeTGTduXLMkRyy+vnjg33//jZ9++gkjR45EaGgoSkpKsHHjRgiFQqSkpAAArly5gpCQEAwZMgQSiQTHjx/HN998gzlz5sDb2xsAIJPJEBoaiiNHjtxya5eulHVShnja+NfO8BAX+//g5BTAdSDExtBQnY3qPjQdPYaPbnG7r08UxoQ/DF4hJU2OKJ9fYfY2zp8/Dx6P12LLE4lEArlcbvoSCoWm+zIzMxEVFYWkpCS4uLggIiICffv2xV9//WUqZB8xYgT69u0LX19fqNVqDB06FGq1GufOnWvWTkREBLKyssz+OG9UXc1CJEq69YHEZoSGPAW1egDXYRAbRD1ONmzwzLkoy72Cq2ea3kS6RQ9BLHqDraRPxY4qt8rM3U1o6hny8Wk5ZXvz5s3YsGEDXFxc0LNnTyQmJprWw7m2Oe+NBAIBqqurUVVV1eq2KyzLQqvVwsmp+TCKr68vqqurUVlZ2ep55lJYEAIX1/0Wa4+Yj4fHKAQFPcR1GMRGUY+TDeMLBBi74Dko3NwxqNd0xDSmgG2gniZHpXE2oKqm2uztVFZWQi5vvrbR4MGDMWnSJMyYMQOxsbHYtm0b/vzzT9P9oaGhOHPmDC5evAiWZVFWVoaDBw8CAGpqWt+Lcd++fdDpdIiNjW12+7X96yxdmH7qFAOh0N2ibZKuJ5NFUDE4uSPU42TjpEoVpj6/DFWfZIM10hpNjqxEXQ/kmb8dvV4PgaD5S8eAAdeHPLy8mla537Nnj+n2hIQElJeX45tvvoHBYIBYLEavXr2we/du8FpZ2PXkyZPYvXs3pkyZAplM1uy+a23rdJbtWWVZBgZDCoBfLdou6TpNxeAfg8+Xch0KsWGUONkBua8b+FMYlK09TRPoHFgezD+bDgCkUikaGhraPcbPzw9arRYajQZyuRwMwyAtLQ1Dhw6FRqOBTCbDxYtNs9RuHm7LysrChg0bMGnSJISEtFwGoL6+HgBaJFSWkH1WjfAIizdLugQPcbHvQioN5DoQYuNoqM5OOMWooRpFa804KpZhkVuRb5G2vLy8UFJS0u4xhYWFEAgEkEiab+/D4/GgVCrB5/ORlZUFPz+/ZgnQyZMn8csvv2DixImIiGg9QykuLgaPx4O7u+WHzQoLaeNfWxUa8iTU6oFch0HsAPU42RFFf1/oy+pRe8D8BcLEutS4GVBbU2eRtkJDQ7F9+3bU19fDyckJ2dnZ0Gg08Pf3h0AgwOXLl7Fjxw4kJCSYhtXq6upw+vRpBAUFQa/X49ixYzh9+jQyMjJM1z158iTWr1+PESNGwM/PDxqNBgBaJGA5OTkIDAxsUWxuKRUVMXByOslJ26RzPD3vQlDQfK7DIHaCEic743xXKPTlDdCeM/+0dGI9ClW1QOs11l3O09MTPj4+OHXqFJKSksDn83HkyBFs27YNLMvCxcUFgwYNMq3PdM2JEyewbds2AE1DeTNnzoSv7/XtTI4ePQqj0YjNmzdj8+bNptt79OiB8ePHm37OysrCoEGDzPoY23MyU4Q+qTIYDLWcxUA6Tq0ejJjoZVyHQewIw9JuoHbH2KBH8UcnoC+yTA8E4d6ukIs4n3/JYu39/fff2LZtG+bPn29acsASzp07h99//x3z5s1rtajcUoYOzUOjbgdn7ZOOcVYlIz5+Nfh8ya0PJqSDqMbJDvEkArg9EAueQsR1KMQCjDwWuWWWqW+6Jjw8HImJiaiuNv/yBzfS6XQYN24cp0kTAFy+TBv/WjuFPBY9eqykpIl0OepxsmO64jqUfJoJo4YWxLRnFV46/Fi5h+swHM6wtD+h1V7mOgzSCqk0BIkJ6yASqbkOhdgh6nGyY0IPKdzndANPSqVs9qxQoeE6BIdUX9eT6xBIK8Rib/SMX0NJEzEbSpzsnNBLBrfZ3cBIKHmyV1d1xVyH4JCyshRgGPq7siZCoSt6xn8JiaTllkCEdBVKnByAyFcO99lxYMR8rkMhXUwvYJFXWsh1GA6pqoqFWEwb/1oLPl+O+PgvIJPRenbEvChxchAifwXcHogFI6L/cntS4dkIvV7PdRgOq7CA3qStAY8nRo/uK6FUxHEdCnEA9C7qQMRBKqhnxoIR0n+7vSiUWmjxJtKqU6f4EArduA7DoTGMAN3iPoCLS8qtDyakC9A7qIORhDpDPT0GEFhu7R1iPrkNVN/EJaMRMBroDZs7DGKil8LNbQjXgRAHQomTA5JEuEA9LQbgU/Jky3RiFgVU38S57GzqceJKRMQr8PIax3UYxMFQ4uSgnKJcoZ4aBfAoebJVpZ5a0DJs3CsoYCGRxHIdhsMJC3sW/n7TuQ6DOCBKnByYU6wbXKdE0rPARhWIq7gOgfyjspISJ0thGD6io95GYMBcrkMhDoreMh2ctLs7XCZFAtTxZHOu1hZxHQL5x8lMCfg8Kddh2D0eT4S4uPfh4zOJ61CIA6PEiUDW0wMud4dT8mRDtFIWxeUlXIdB/tHQwIIv6MV1GHaNz5ejR/dV8HBP5zoU4uAocSIAAFmyF9TTommpAhtR7FnPdQjkJlcu+3Edgt0SCl2R0HMtXF1TuQ6FEEqcyHVOsW5wm9MNPBltI2HtCgSVXIdAbnL+PCAWBXAdht0Ri72RmPAtlMruXIdCCABKnMhNxIFKuM+LB18t4ToU0o7c6gKuQyCtqG9I5DoEuyKVhiIp8TvaRoVYFUqcSAtCNyd4zOsBob+C61BIK+pURlRUVXIdBmnFqSwlbfzbRRSKbkhMWEcb9hKrQ4kTaRVfLoL73G6QRLtyHQq5SbG6jusQSBsqK40QixO4DsPmubj0QULPryES0esPsT6UOJE28UR8qKfHQNbbm+tQyA3yeRVch0DaUVQYxnUINs3dfTjie6yCQCDjOhRCWkWJE2kXw2PgMj4MyhFBtFyBlcipyOc6BNKOU6f4EAiop6QzfLwno1vcB+DxxFyHQkibKHEiHaIc5A/XeyNpfzuO1bjqoanVcB0GaYfBALAsrel0u4KCHkZ09FtgGD7XoRDSLkqcSIdJ4z3gNisOjIRe2LhS5Er1Tbbg73MeXIdgM3g8J8TFvofQkAVch0JIh1DiRG6LJNQZHvN6gK+irnQu5BvLuA6BdEBenhESSQzXYVg9icQXSYnfw9NzNNehENJhlDiR2yb0lMHj4R4QelPxpiWxDIuccqpvshVVlXFch2DVnJ17ITlpPRSKaK5DIeS2UOJEOoWvFMNjfg9Ikzy5DsVhVHkY0NDQwHUYpIMyM8Xg8Zy4DsMq+fpOQ8/4L2m5AWKTKHEincYI+XC9JwIu90aCEdFTydwKlVQUbksaGgAhbfzbDI8nQlTkG4iKXAwejxYKJbaJ3u3IHZP19IDHIz0h9JJyHYpdy9OXch0CuU05Of5ch2A1JBI/JCZ8B1/fKVyHQsgdocSJdAmhhxQeD8fT0J2ZGPksrpZSfZOtOXcOEIsoeVKrByEl+Rcold24DoWQO0aJE+kypqG7yRE0dNfFyj110Ol0XIdBOqGhwZG3YOEhJPhJ9Oj+GYRCZ66DIaRL0Lsb6XKyBE94PNITAk8auusqhbIarkMgnXT6tAoMHG/tM6HQFT3jVyM4+BEwDC2cS+wHJU7ELExDd4k0dNcVrmqLuQ6BdFJ5OQuxxLF6nZxVyUhJ3gBX175ch0JIl6PEiZgNT8SH66QIuEyiobs7oRexyC8t5DoMcgeKi8K5DsEieDwJwsNeQELCfyGR0ObgxD7RuxkxO1kiDd3diVLPRhiNRq7DIHcgK4sPgcCF6zDMSqnsiZTkjQgImAWGobcWYr/o2U0sgobuOq9QUsV1COQOGQwA7HTjXx5PhNDQhUhK/BYyWQjX4RBidpQ4EYu5NnSnnhEDnlLEdTg242pdEdchkC5wzg43/lUo4pCc9AuCAh8EwzheATxxTJQ4EYtzilHD68lEWvOpAxolLIrKS7gOg3SBvDwWEnEU12F0CYYRIiT4CSQl/gi5PILrcAixKEqcCCd4TgK43hMBt9lx4LuIuQ7HapV4NoBlWa7DIF2kutr2F4CUy6OQnPQTgoMfpW1TiEOixIlwShLuAs8nEiHr4w3QUi8tFIgquQ6BdKGTJ53A40m4DqNTGEaAoMD5SE76GQpFDCcxZGRkgGEY05darcaIESOQmZnJSTzEMVHiRDjHE/PhMi4M7g92h8CDdpO/UW4NLUNgT+rqWJvc+FcmC0dS4vcIDX0KPB639YkjRoxAQUEBCgoKsH37dggEAowZM4bTmIhjocSJWA1xkAqejydAmR4IRkhPzXqFEWWV5VyHQbpYbm4A1yF0GMMIERjwr3/2mevOdTgAALFYDC8vL3h5eSE+Ph6LFi1Cbm4uSkqaagEXLVqEiIgISKVShISE4KWXXmqxXdHrr78ODw8PKBQKzJkzB88++yzi4+M5eDTEFtG7E7EqDJ8H5eAAeD6ZCEmkfa97cysl7vVch0DMIDubhUjky3UYt6RWD0avlM0IC1sEHs866xA1Gg2+/vprhIWFQa1WAwAUCgVWr16N06dP4z//+Q9WrlyJd99913TO119/jTfeeANLlizB0aNHERAQgI8++oirh0BsEMNS5SmxYnUnS1C18SIM1Y1ch2JxB8KuIutqNtdhEDPoP6AawC9ch9EqmSwc4WEvQK3uz3UoLWRkZOCrr76CRNJUJ1ZbWwtvb29s2rQJCQmtb2uzbNkyfPvttzhy5AgAoHfv3khKSsIHH3xgOqZfv37QaDQ4fvy42R8DsX3U40SsmrSbOzyfSoS8r4/DPVtzqgq4DoGYyelTzrC2J7RQ6IrIiMXolfKrVSZN1wwePBjHjx/H8ePHcfDgQQwfPhwjR47ElStXAAA//PAD+vXrBy8vL8jlcrz00kvIyckxnZ+dnY2UlJRm17z5Z0LaQ3NJidXjiQVwHhsKWS9vVG+7jPqsMq5DMrtaFwOqa6q5DoOYSVmZERJJTzQ0HOU6FDCMEP5+MxAU9AiEQiXX4dySTCZDWFiY6efExESoVCqsXLkSY8aMwZQpU7B48WKkp6dDpVJh3bp1WL58ebNrMEzzKbw08EJuByVOxGYIPaRQT4tB49UaVG29DO35Sq5DMpsi1zogj+soiDmVFEdAoeQ2cXJzG4bwsGchlQZzGsedYBgGPB4P9fX1+OuvvxAYGIgXXnjBdP+1nqhrIiMjcejQIUyfPt1027VhPEI6ghInYnNEfgq4z+mGhvMVqNp6GbqrGq5D6nJ5oNl09i4rS4D+A5yh11davG25LBLh4S/A1bWvxdu+U1qtFoWFTct0VFRU4IMPPoBGo8HYsWNRVVWFnJwcrFu3DsnJyfj111/x888/Nzv/0Ucfxdy5c5GUlITU1FR8++23yMzMREgI7bNHOoYSJ2KzJGEukDzigrqTpajedhn6EvuYhcYyLHLL87kOg5iZXg8AvQD8ZrE2hUJXhIQ8CV+fe212b7mtW7fC29sbQNMMuqioKHz//fcYNGgQAODJJ5/EI488Aq1Wi9GjR+Oll17Cq6++ajr//vvvx8WLF/H000+joaEBkydPRkZGBg4dOsTBoyG2iGbVEbvAGlnUHS1C9R85MFRpuQ7njlS56/F9zW6uwyAW4B/AQ1DQGrO3w+OJ4ec3HcFBj0AgUJi9PVuTlpYGLy8vrF27lutQiA2gHidiFxgeA1myF6TxHtDsz0fNrlwY6/Rch9UpRapaoIbrKIgl5OYYERUViYYG8yw7IRAo4Os7DQH+GRCJ3MzShq2pq6vDxx9/jPT0dPD5fHzzzTf4448/8Pvvv3MdGrERlDgRu8IIeVAM8IMsxQs1e65CszcPbKOR67BuS56hlOsQiAVVV3eDSNS1iZNQqEaA/wPw85tGPUw3YRgGmzdvxuuvvw6tVovIyEj8+OOPGDZsGNehERtBQ3UEQNOLyc8//4zx48e3ev+uXbswePBgVFRUwNnZ2aKx3QmDphE1O3KhOVQI6K0/gTLyWHwl34vGRsdb8NNRyWQMkpK/hdF450PMEokvAgLmwsd7Evh829xMmBBrZ10rsBGzKS4uxoMPPoiAgADTXk/p6enYv39/h85PTU1FQUEBVCpVu8dlZGS0mXxxgS8XwfmuUHg/lwJleiD4Sm43KL2VSk89JU0OpraWhVB4ZwswSqVhiIlehj69d8DfbzolTYSYEQ3VOYiJEydCp9NhzZo1CAkJQVFREbZv347y8o5NexeJRPDy8mrzfoPB0GJROWvClwmhHBwAxQB/1GeVQLM3H4251ldIVCivASq4joJY2tXcQLh7/Hnb5ymVPRAU+BDc3NKs+u+PEHtCPU4OoLKyEnv37sWSJUswePBgBAYGIiUlBc899xxGjx5tOq60tBQTJkyAVCpFeHg4NmzYYLpv165dYBgGlZWVAIDVq1fD2dkZmzZtQkxMDMRiMR544AGsWbMGv/zyCxiGAcMw2LVrl4UfbfsYPgNpDw94PBwP9/k94NTDHeBbzxvO1cYSrkMgHDh7FhCJfDp8vItLKnrGf4nkpJ/g7j6ckiZCLIh6nByAXC6HXC7H+vXr0bt3b4jFre90vnjxYixduhTLli3D+++/j/vvvx9XrlyBq6trq8fX1dXhrbfewmeffQa1Wg0vLy80NDSguroaX3zxBQC0ea41EAcoIQ5QwlClheZAAWoPFnA6E08vYJFXSvvTOSYGusYkABvaPoLhw81tKAIDH4JK2cNyoRFCmqEeJwcgEAiwevVqrFmzBs7Ozujbty+ef/55ZGZmNjsuIyMD9913H8LCwvDmm2+itra23UXhdDodVqxYgdTUVERGRkKlUsHJyclUQ+Xl5QWRyLprigCArxJDlR4E7+dS4HJ3OASeUk7iKPfSwWAwcNI24d7p0y5o7SVZIvZBcPATSE3dg+7dPqKkiRCOUeLkICZOnIj8/Hxs2LAB6enp2LVrFxISErB69WrTMd27dzd9L5PJoFAoUFxc3OY1RSJRs3NsHSPkQ5biBa8nE+E2Jw6SaFfAgiMghU5VlmuMWJ3SUiMkkngAAMMI4OY2DD26f4bU1N0ICX4UEnHbNYaEEMuhoToHIpFIkJaWhrS0NLz88suYM2cOXnnlFWRkZAAAhEJhs+MZhoHR2PYUficnJ7utrZCEuUAS5gJ9WT00BwtRf6LE7CuSX21oO0kljqG6KhlRUQPh4zMJYrEn1+EQQlpBiZMDi4mJwfr167v0miKRyK6GmwRqJziPCoZqZBAar1Sj7kQJ6k+WwqjRdWk7OokRBaVFXXpNYhtEIhFiY2PRo0cPBAYG2u2HEULsBSVODqCsrAyTJk3CrFmz0L17dygUChw5cgRLly7FuHHjurStoKAg/Pbbb8jOzoZarYZKpWrRk2WLGIaBOEgFcZAKzmNDob1Q2ZREZZWBbbjzgvJSj0awxbQWraNgGAYhISHo0aMHoqOj7eJvhBBHQYmTA5DL5ejVqxfeffddXLhwATqdDv7+/pg7dy6ef/75Lm1r7ty52LVrF5KSkqDRaLBz507TruX2guExkIS7QBLuAna8EQ3nKlB3ogQNZ8o6vb1LgZjqmxyBt7c3YmNj0b17dyiVSq7DIYR0Am25QkgXMTYa0HCmvCmJOlcO6Dv+p/WLzwmUlNMedfZGKBQiJCQEERERCA8Pp2SJEDtAiRMhZmBs0KM+qwx1mSXQnq8EjG3/mWllRqw17LRccMSsVCoVIiIiEBERgaCgIBqGI8TOUOJEiJkZG/TQXqhCw98VaPi7Aoayhmb35wbX4beCju0ZSKwPwzDw9fU1JUvtbU1ECLF9VONEiJnxJAI4xarhFKsGAOjL6tHwdyUa/q6A9kIl8gW0OZ2tEYvFCA0NNQ3ByWQyrkMihFgI9TgRwiHWwKIwvwAXrlzE5cuXkZOTg8bGRq7DIjcRiUTw9fWFr68vQkJCEBgYCD6fz3VYhBAOUOJEiBUxGo0oKCjA5cuXceXKFVy5cgVarXkX3iTNMQwDT09P+Pr6ws/PD76+vnBzcwOPRxstEEIocSLEqhmNRhQXF6OoqKjZl0aj4To0u6FSqUy9SX5+fvD29raJPRYJIdygxIkQG1RbW9sioSouLoZef+eLcdozsVgMHx+fZr1JCoWC67AIITaEEidC7ITRaER5eXmLZKqiwrGKzxUKBVxcXODi4gJXV1fT9y4uLpDL5VyHRwixcZQ4EWLntFotqqqqUFNTY/rSaDQtfraV3iqBQABnZ+cWSdG1L1o3iRBiTpQ4EUIAAPX19a0mVhqNxpRYGY1GGAwG01dbP98Kj8eDSCSCUChs9q9EIoGTk1OLr2u3K5VKKBQK2giXEMIZSpwIIV2utcQKgClBoqn8hBBbRYkTIYQQQkgH0cIkhBBCCCEdRIkTIYQQQkgHUeJECCGEENJBlDgRQgghhHQQJU6EEEIIIR1EiRMhhBBCSAdR4kQIIYQQ0kGUOBFCCCGEdBAlToQQQgghHUSJEyGEEEJIB1HiRAghhBDSQZQ4EUIIIYR0ECVOhBBCCCEdRIkTIYS04fLly2AYBsePH+c6FEKIlaDEiRBilYqLi/Hggw8iICAAYrEYXl5eSE9Px/79+7kOjRDiwARcB0AIIa2ZOHEidDod1qxZg5CQEBQVFWH79u0oLy/nOrQ7otPpIBQKuQ6DENJJ1ONECLE6lZWV2Lt3L5YsWYLBgwcjMDAQKSkpeO655zB69GgAAMMw+OyzzzBhwgRIpVKEh4djw4YNza5z+vRpjBo1CnK5HJ6enpg+fTpKS0tN92/duhX9+vWDs7Mz1Go1xowZgwsXLrQZl9FoxNy5cxEREYErV64AADZu3IjExERIJBKEhIRg8eLF0Ov1pnMYhsHHH3+McePGQSaT4fXXX+/KXxUhxMIocSKEWB25XA65XI7169dDq9W2edzixYsxefJkZGZmYtSoUbj//vtNPVIFBQUYOHAg4uPjceTIEWzduhVFRUWYPHmy6fza2losWLAAhw8fxvbt28Hj8TBhwgQYjcYWbTU2NmLy5Mk4cuQI9u7di8DAQPz222+YNm0aHnvsMZw+fRqffPIJVq9ejTfeeKPZua+88grGjRuHkydPYtasWV30WyKEcIIlhBAr9MMPP7AuLi6sRCJhU1NT2eeee449ceKE6X4A7Isvvmj6WaPRsAzDsFu2bGFZlmVfeukldvjw4c2umZubywJgs7OzW22zuLiYBcCePHmSZVmWvXTpEguA/fPPP9lhw4axffv2ZSsrK03H9+/fn33zzTebXWPt2rWst7d3szifeOKJTv4WCCHWhnqcCCFWaeLEicjPz8eGDRuQnp6OXbt2ISEhAatXrzYd0717d9P3MpkMCoUCxcXFAICjR49i586dpt4ruVyOqKgoADANx124cAFTp05FSEgIlEolgoODAQA5OTnNYrnvvvug0Wiwbds2qFQq0+1Hjx7Fa6+91qyNuXPnoqCgAHV1dabjkpKSuvaXQwjhDBWHE0KslkQiQVpaGtLS0vDyyy9jzpw5eOWVV5CRkQEALYqsGYYxDbMZjUaMHTsWS5YsaXFdb29vAMDYsWPh7++PlStXwsfHB0ajEXFxcWhsbGx2/KhRo/DVV1/hwIEDGDJkiOl2o9GIxYsX4+6772419mtkMlnnfgGEEKtDiRMhxGbExMRg/fr1HTo2ISEBP/74I4KCgiAQtHypKysrw5kzZ/DJJ5+gf//+AIC9e/e2eq158+YhLi4Od911F3799VcMHDjQ1EZ2djbCwsI694AIITaHEidCiNUpKyvDpEmTMGvWLHTv3h0KhQJHjhzB0qVLMW7cuA5d4+GHH8bKlStx33334ZlnnoGbmxvOnz+PdevWYeXKlXBxcYFarcann34Kb29v5OTk4Nlnn23zeo8++igMBgPGjBmDLVu2oF+/fnj55ZcxZswY+Pv7Y9KkSeDxeMjMzMTJkydp9hwhdooSJ0KI1ZHL5ejVqxfeffddXLhwATqdDv7+/pg7dy6ef/75Dl3Dx8cHf/31FxYtWoT09HRotVoEBgZixIgR4PF4YBgG69atw2OPPYa4uDhERkbivffew6BBg9q85hNPPAGj0YhRo0Zh69atSE9Px6ZNm/Daa69h6dKlEAqFiIqKwpw5c7roN0EIsTYMy7Is10EQQgghhNgCmlVHCCGEENJBlDgRQgghhHQQJU6EEEIIIR1EiRMhhBBCSAdR4kQIIYQQ0kGUOBFCCCGEdBAlToQQQgghHUSJEyGEEEJIB1HiRAghhBDSQZQ4EUIIIYR0ECVOhBBCCCEdRIkTIYQQQkgHUeJECCGEENJBlDgRQgghhHQQJU6EEEIIIR1EiRMhhBBCSAdR4kQIIYQQ0kGUOBFCCCGEdBAlToQQQgghHUSJEyGEEEJIB1HiRAghhBDSQZQ4EUIIIYR0ECVOhBBCCCEdRIkTIYQQQkgHUeJECCGEENJBlDgRQgghhHQQJU6EEEIIIR1EiRMhhBBCSAdR4kQIIYQQ0kGUOBFCCCGEdBAlToQQQgghHUSJEyGEEEJIB1HiRAghhBDSQZQ4EUIIIYR0ECVOhBBCCCEdRIkTIYQQQkgHUeJECCGEENJB/x/6RzEdU5PtngAAAABJRU5ErkJggg==",
+ "text/plain": [
+ "<Figure size 1000x700 with 1 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ "<Figure size 1000x700 with 1 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "def func(pct, allvalues):\n",
+ " absolute = int(pct / 100.*np.sum(allvalues))\n",
+ " return \"{:.1f}%\\n({:d})\".format(pct, absolute)\n",
+ "\n",
+ "def plot_pie(title, data):\n",
+ " # Creating plot\n",
+ " fig, ax = plt.subplots(figsize =(10, 7))\n",
+ " plt.pie(data, autopct = lambda pct: func(pct, data), labels = class_names)\n",
+ " ax.set_title(title)\n",
+ " \n",
+ " # show plot\n",
+ " plt.show()\n",
+ "\n",
+ "plot_pie(\"Train data distribution\", df_train[\"label\"].value_counts())\n",
+ "plot_pie(\"Validation data distribution\", df_val[\"label\"].value_counts())\n",
+ "plot_pie(\"Test data distribution\", df_test[\"label\"].value_counts())"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Print a single image"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ "<Figure size 640x480 with 2 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "# Make copies of the data to allow easy exploration\n",
+ "df_train_exp_copy = df_train.copy() \n",
+ "y_train_exp = df_train_exp_copy.pop('label').to_numpy()\n",
+ "x_train_exp = df_train_exp_copy.to_numpy()\n",
+ "\n",
+ "\n",
+ "# Take a single image, and remove the color dimension by reshaping\n",
+ "image = x_train_exp[0].reshape((28,28)) / 255.0\n",
+ "\n",
+ "plt.figure()\n",
+ "plt.imshow(image, cmap=plt.cm.binary)\n",
+ "plt.colorbar()\n",
+ "plt.grid(False)\n",
+ "plt.show()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Print one image from each category to see how they look like and how they differ"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ "<Figure size 1000x1000 with 10 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "plt.figure(figsize=(10,10))\n",
+ "i = 0\n",
+ "for index in range(len(x_train_exp)):\n",
+ " label = y_train_exp[index]\n",
+ " image = x_train_exp[index] / 255.0\n",
+ " if label == i:\n",
+ " image = image.reshape((28,28))\n",
+ " plt.subplot(5,5,i+1)\n",
+ " plt.xticks([])\n",
+ " plt.yticks([])\n",
+ " plt.grid(False)\n",
+ " plt.imshow(image, cmap=plt.cm.binary)\n",
+ " plt.title(class_names[label])\n",
+ " i += 1\n",
+ " if i == 10:\n",
+ " break\n",
+ "plt.show()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# 3. Data preperation"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## 3.1. Test and Train Data"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "df_train_copy = df_train.copy() \n",
+ "y_train = df_train_copy.pop('label').to_numpy()\n",
+ "x_train = df_train_copy.to_numpy()\n",
+ "df_val_copy = df_val.copy() \n",
+ "y_val = df_val_copy.pop('label').to_numpy()\n",
+ "x_val = df_val_copy.to_numpy()\n",
+ "df_test_copy = df_test.copy() \n",
+ "y_test = df_test_copy.pop('label').to_numpy()\n",
+ "x_test = df_test_copy.to_numpy()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## 3.2. Feature Scaling"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "x_train = x_train / 255.0\n",
+ "x_val = x_val / 255.0\n",
+ "x_test = x_test / 255.0"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Convert the image shape from 784 to 28x28 (only if load as CSV with 784 columns)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 16,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "IMG_ROWS = 28\n",
+ "IMG_COLS = 28\n",
+ "IMAGE_SHAPE = (IMG_ROWS, IMG_COLS, 1) \n",
+ "x_train = x_train.reshape(x_train.shape[0], *IMAGE_SHAPE)\n",
+ "x_val = x_val.reshape(x_val.shape[0], *IMAGE_SHAPE)\n",
+ "x_test = x_test.reshape(x_test.shape[0], *IMAGE_SHAPE)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## 3.3. Convert Labels"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 17,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "y_train = to_categorical(y_train, 10)\n",
+ "y_val = to_categorical(y_val, 10)\n",
+ "y_test = to_categorical(y_test, 10)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Check the data shapes to get ensure that the data is in the correct format"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 18,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "(54000, 28, 28, 1)\n",
+ "(54000, 10)\n",
+ "(6000, 28, 28, 1)\n",
+ "(6000, 10)\n",
+ "(10000, 28, 28, 1)\n",
+ "(10000, 10)\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(x_train.shape)\n",
+ "print(y_train.shape)\n",
+ "\n",
+ "print(x_val.shape)\n",
+ "print(y_val.shape)\n",
+ "\n",
+ "print(x_test.shape)\n",
+ "print(y_test.shape)\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# 4. Modelling and Evaluation"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Define how the model will look like. Below some descriptions for different Layer types.\n",
+ "The first trial of this model was just a DNN with simple dense layers, but the results can be improved by using a CNN.\n",
+ "As a start architecture the LeNet-5 implementation was chosen and then altered.\n",
+ "The hyperparameters could also be optimized with a Keras Optimizer which tries out several defined combinations.\n",
+ "The current parameters got chosen by exploration.\n",
+ "\n",
+ "- Dense: receives all inputs from previous layer, creates dot product \n",
+ "- Dropout layer: removes noise for overfitting, drops at specific rate\n",
+ "- Reshape layer: changes the shape of the input, not used\n",
+ "- Permute layer: alter shape of the input, not used\n",
+ "- ReapeatVector layer: repeats the input for fixed number of times, not used\n",
+ "- Flatten Layer: flattens the matrix\n",
+ "- MaxPooling2D Layer: reduces number of input\n",
+ "- Conv2D Layer: convolves an input"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 19,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "C:\\Users\\ar\\anaconda3\\Lib\\site-packages\\keras\\src\\layers\\convolutional\\base_conv.py:107: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
+ " super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
+ ]
+ },
+ {
+ "data": {
+ "text/html": [
+ "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential\"</span>\n",
+ "</pre>\n"
+ ],
+ "text/plain": [
+ "\u001b[1mModel: \"sequential\"\u001b[0m\n"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/html": [
+ "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”┳â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”┳â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”┓\n",
+ "┃<span style=\"font-weight: bold\"> Layer (type) </span>┃<span style=\"font-weight: bold\"> Output Shape </span>┃<span style=\"font-weight: bold\"> Param # </span>┃\n",
+ "┡â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”╇â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”╇â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”┩\n",
+ "│ conv2d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">320</span> │\n",
+ "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+ "│ conv2d_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">9,248</span> │\n",
+ "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+ "│ max_pooling2d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+ "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+ "│ flatten (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6272</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+ "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+ "│ dropout (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">6272</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+ "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+ "│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">802,944</span> │\n",
+ "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+ "│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1,290</span> │\n",
+ "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
+ "</pre>\n"
+ ],
+ "text/plain": [
+ "â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”┳â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”┳â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”┓\n",
+ "┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
+ "┡â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”╇â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”╇â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”┩\n",
+ "│ conv2d (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m320\u001b[0m │\n",
+ "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+ "│ conv2d_1 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m9,248\u001b[0m │\n",
+ "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+ "│ max_pooling2d (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
+ "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+ "│ flatten (\u001b[38;5;33mFlatten\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6272\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
+ "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+ "│ dropout (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m6272\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
+ "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+ "│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m) │ \u001b[38;5;34m802,944\u001b[0m │\n",
+ "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
+ "│ dense_1 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m) │ \u001b[38;5;34m1,290\u001b[0m │\n",
+ "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/html": [
+ "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">813,802</span> (3.10 MB)\n",
+ "</pre>\n"
+ ],
+ "text/plain": [
+ "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m813,802\u001b[0m (3.10 MB)\n"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/html": [
+ "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">813,802</span> (3.10 MB)\n",
+ "</pre>\n"
+ ],
+ "text/plain": [
+ "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m813,802\u001b[0m (3.10 MB)\n"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/html": [
+ "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
+ "</pre>\n"
+ ],
+ "text/plain": [
+ "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "model = Sequential()\n",
+ "model.add(Conv2D(filters=32,kernel_size=3, activation='relu', padding='same', input_shape=(28, 28,1)))\n",
+ "model.add(Conv2D(filters=32,kernel_size=3,padding='same', activation='relu'))\n",
+ "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
+ "model.add(Flatten())\n",
+ "model.add(Dropout(0.40))\n",
+ "model.add(Dense(units=128, activation='relu'))\n",
+ "model.add(Dense(units=10, activation='softmax'))\n",
+ "model.summary()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Compile the model"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 20,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "model.compile(\n",
+ " optimizer=tf.keras.optimizers.Adam(), \n",
+ " loss= tf.keras.losses.categorical_crossentropy, \n",
+ " metrics=['accuracy']\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Train/Fit the model"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 21,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Epoch 1/10\n",
+ "844/844 - 9s - 10ms/step - accuracy: 0.8485 - loss: 0.4227 - val_accuracy: 0.9016 - val_loss: 0.2939\n",
+ "Epoch 2/10\n",
+ "844/844 - 7s - 9ms/step - accuracy: 0.8973 - loss: 0.2795 - val_accuracy: 0.9141 - val_loss: 0.2329\n",
+ "Epoch 3/10\n",
+ "844/844 - 7s - 9ms/step - accuracy: 0.9124 - loss: 0.2361 - val_accuracy: 0.9312 - val_loss: 0.2121\n",
+ "Epoch 4/10\n",
+ "844/844 - 9s - 10ms/step - accuracy: 0.9236 - loss: 0.2059 - val_accuracy: 0.9156 - val_loss: 0.2440\n",
+ "Epoch 5/10\n",
+ "844/844 - 9s - 10ms/step - accuracy: 0.9330 - loss: 0.1798 - val_accuracy: 0.9234 - val_loss: 0.1941\n",
+ "Epoch 6/10\n",
+ "844/844 - 9s - 10ms/step - accuracy: 0.9403 - loss: 0.1591 - val_accuracy: 0.9312 - val_loss: 0.1952\n",
+ "Epoch 7/10\n",
+ "844/844 - 9s - 11ms/step - accuracy: 0.9474 - loss: 0.1394 - val_accuracy: 0.9234 - val_loss: 0.1963\n",
+ "Epoch 8/10\n",
+ "844/844 - 9s - 11ms/step - accuracy: 0.9535 - loss: 0.1258 - val_accuracy: 0.9266 - val_loss: 0.2069\n",
+ "Epoch 9/10\n",
+ "844/844 - 10s - 12ms/step - accuracy: 0.9601 - loss: 0.1071 - val_accuracy: 0.9312 - val_loss: 0.2307\n",
+ "Epoch 10/10\n",
+ "844/844 - 9s - 11ms/step - accuracy: 0.9645 - loss: 0.0956 - val_accuracy: 0.9458 - val_loss: 0.2156\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "C:\\Users\\ar\\anaconda3\\Lib\\contextlib.py:155: UserWarning: Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches. You may need to use the `.repeat()` function when building your dataset.\n",
+ " self.gen.throw(typ, value, traceback)\n"
+ ]
+ }
+ ],
+ "source": [
+ "# Determine the maximum number of epochs\n",
+ "NUM_EPOCHS = 10\n",
+ "BATCH_SIZE = 64\n",
+ "\n",
+ "# Fit the model, \n",
+ "# specify the training data\n",
+ "# the total number of epochs\n",
+ "# and the validation data we just created \n",
+ "history = model.fit(\n",
+ " x_train,\n",
+ " y_train,\n",
+ " batch_size=BATCH_SIZE,\n",
+ " epochs=NUM_EPOCHS,\n",
+ " validation_data=(x_val, y_val), \n",
+ " validation_steps=10,\n",
+ " verbose =2\n",
+ ")"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "We want do know how especially the loss changes over time"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 22,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ "<Figure size 640x480 with 1 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/plain": [
+ "<Figure size 640x480 with 0 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "acc = history.history[\"accuracy\"]\n",
+ "val_acc = history.history[\"val_accuracy\"]\n",
+ "loss = history.history[\"loss\"]\n",
+ "val_loss = history.history[\"val_loss\"]\n",
+ "\n",
+ "epochs = range(len(acc))\n",
+ "\n",
+ "plt.plot(epochs, acc, \"darkgreen\", label=\"Training accuracy\")\n",
+ "plt.plot(epochs, val_acc, \"darkblue\", label=\"Validation accuracy\")\n",
+ "plt.plot(epochs, loss, \"lightgreen\", label=\"Training loss\")\n",
+ "plt.plot(epochs, val_loss, \"lightblue\", label=\"Validation loss\")\n",
+ "plt.title(\"Training and validation accuracy\")\n",
+ "plt.xlabel(\"Epochs\")\n",
+ "plt.ylabel(\"Precent/100\")\n",
+ "plt.legend(loc=0)\n",
+ "plt.figure()\n",
+ "\n",
+ "\n",
+ "plt.show()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "We can see that the model is performing quite well, but after the second epoch it starts to overfit. To prevent that we could try with different train-validation splits, add more dropout or restructure parts of the model."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Evaluate the test-data"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 23,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "test_loss, test_accuracy = model.evaluate(x_test, y_test, verbose=0)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 24,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Test loss: 0.22. Test accuracy: 93.17%\n"
+ ]
+ }
+ ],
+ "source": [
+ "print('Test loss: {0:.2f}. Test accuracy: {1:.2f}%'.format(test_loss, test_accuracy*100.))"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "The model performs quite well with an accuracy of > 90% on the test-data. The loss is acceptable."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Show results by class"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 25,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "\u001b[1m313/313\u001b[0m \u001b[32mâ”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”â”\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step\n"
+ ]
+ }
+ ],
+ "source": [
+ "predicted_classes = (model.predict(x_test) > 0.5).astype(\"int32\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 26,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ " precision recall f1-score support\n",
+ "\n",
+ " Top 0.91 0.85 0.88 1000\n",
+ " Trouser 0.99 0.99 0.99 1000\n",
+ " Pullover 0.89 0.89 0.89 1000\n",
+ " Dress 0.92 0.97 0.94 1000\n",
+ " Coat 0.93 0.86 0.89 1000\n",
+ " Sandal 0.99 0.98 0.99 1000\n",
+ " Shirt 0.81 0.82 0.81 1000\n",
+ " Sneaker 0.96 0.98 0.97 1000\n",
+ " Bag 0.98 0.99 0.99 1000\n",
+ " Ankle boot 0.98 0.96 0.97 1000\n",
+ "\n",
+ " micro avg 0.94 0.93 0.93 10000\n",
+ " macro avg 0.94 0.93 0.93 10000\n",
+ "weighted avg 0.94 0.93 0.93 10000\n",
+ " samples avg 0.93 0.93 0.93 10000\n",
+ "\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "C:\\Users\\ar\\anaconda3\\Lib\\site-packages\\sklearn\\metrics\\_classification.py:1517: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in samples with no predicted labels. Use `zero_division` parameter to control this behavior.\n",
+ " _warn_prf(average, modifier, f\"{metric.capitalize()} is\", len(result))\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(classification_report(y_test, predicted_classes, target_names=class_names))"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.11.5"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}