Project_7_LLMs.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Applied Machine Learning Homework 4 Question 1 \n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Q1) From Huggingface (https://huggingface.co/Links to an external site.) select and download a pre-trained CLIP model (you can use your own computer, Colab, Kaggle... to store the model). Describe the model you downloaded - what is its architecture (e.g. CNN/ViT), number of layers, parameters per layer - breakdown the parameters and explain what they are doing (e.g. are they parts of K, Q and V matrices, bias, feature maps, dense layer...)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## CLIP (Contrastive Language-Image Pre-training)\n",
    "\n",
    "Source: openai/clip-vit-base-patch32\n",
    "\n",
    "### 1. Architecture\n",
    "\n",
    "- **Image Encoder**:\n",
    "    - **ResNet**:\n",
    "    Includes variants like ResNet-50, ResNet-101, and scaled versions (e.g., RN50x64).\n",
    "    Features ResNet-D enhancements, antialiased pooling, and attention pooling.\n",
    "    - **Vision Transformer (ViT)**: The model uses a Vision Transformer as its backbone for processing images. Uses variants such as ViT-B/32 (86M parameters) and ViT-L/14 (307M parameters).\n",
    "\n",
    "- **Text Encoder**: \n",
    "    - The text encoder is based on the GPT-2 architecture.\n",
    "    - The input text is tokenized and embedded into vectors using a learned embedding layer.\n",
    "    - Positional Encoding: Like the ViT, positional encodings are added to the token embeddings to retain the sequence information.\n",
    "\n",
    "\n",
    "- **Shared Embedding Space**:\n",
    "    - **Projection Layers**: \n",
    "    After the vision and text encoders process their respective inputs, the resulting embeddings are projected into a shared space using separate linear projection layers for images and text.\n",
    "    - **Contrastive Learning**: \n",
    "    During training, CLIP uses a contrastive loss to align the image and text embeddings in this shared space. Matching pairs (image and its corresponding caption) are brought closer together, while non-matching pairs are pushed apart.\n",
    "\n",
    "### 2. Number of Layers and Parameters\n",
    "\n",
    "- **Vision Transformer (ViT)**: \n",
    "    - The Vision Transformer used in CLIP has 12 layers with 512 hidden units and 8 attention heads.\n",
    "    - Typical parameter count: ~23M for ResNet-50.\n",
    "- **Text Encoder (GPT-2)**: \n",
    "    - The text encoder used in CLIP has 24 layers.\n",
    "    - Typical parameter count: ~63M.\n",
    "\n",
    "### Parameters Breakdown\n",
    "\n",
    "- **Embedding Layer**: Converts input images/text into embeddings.\n",
    "- **Transformer Layers**: Each layer consists of multi-head self-attention mechanisms and feed-forward neural networks.\n",
    "- **Parameters**: The parameters include weights for the attention mechanism, feed-forward networks, and normalization layers.\n",
    "- **K, Q, V Matrices**: These are part of the self-attention mechanism, where K (Key), Q (Query), and V (Value) matrices are used to compute attention scores.\n",
    "\n",
    "### 3. Functionality\n",
    "\n",
    "- **Image and Text Embeddings**: The model learns to map images and text into a shared embedding space.\n",
    "- **Contrastive Loss**: The model is trained using a contrastive loss function, which encourages the embeddings of matching image-text pairs to be closer together, while non-matching pairs are pushed apart.\n",
    "- **Zero-Shot Transfer**: After pre-training, the model can perform tasks without additional training by leveraging natural language descriptions.\n",
    "\n",
    "### 4. Training and Objectives\n",
    "- The model is trained on a dataset of 400M image-text pairs using a contrastive loss to maximize similarity between correct pairs and minimize similarity for incorrect ones.\n",
    "- Zero-shot transfer is achieved by encoding class names or descriptions as text embeddings and comparing them to image embeddings.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Q2 a)The dataset contains images from the following eight categories: airplane, car, chair, cup, dog, donkey, duck and hat. Each category contains images in five different conditions: realistic, geons, silhouettes, blured and features. Evaluate the model for each condition separately. For each image in the dataset, feed the image into the model together with a text label of a particular category (for each image, evaluate labels of all eight categories). If the model outputs highest correlation for the correct label, consider that as correct classification and otherwise as incorrect classification. Quantify model accuracy for each of the five conditions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import torch\n",
    "from transformers import CLIPProcessor, CLIPModel\n",
    "from PIL import Image\n",
    "from tqdm import tqdm\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from sklearn.manifold import TSNE\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from matplotlib.offsetbox import OffsetImage, AnnotationBbox"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n",
    "\n",
    "model_id = \"openai/clip-vit-base-patch32\"\n",
    "model = CLIPModel.from_pretrained(model_id).to(device)\n",
    "processor = CLIPProcessor.from_pretrained(model_id)\n",
    "\n",
    "data_dir = \"./dataset\"  \n",
    "categories = [\"airplane\", \"car\", \"chair\", \"cup\", \"dog\", \"donkey\", \"duck\", \"hat\"]\n",
    "conditions = [\"blurred\", \"features\", \"geons\", \"realistic\", \"silhouettes\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of images per category:\n",
      "Condition: blurred\n",
      "  airplane: 6\n",
      "  car: 5\n",
      "  chair: 6\n",
      "  cup: 5\n",
      "  dog: 5\n",
      "  donkey: 4\n",
      "  duck: 6\n",
      "  hat: 5\n",
      "Condition: features\n",
      "  airplane: 6\n",
      "  car: 5\n",
      "  chair: 5\n",
      "  cup: 5\n",
      "  dog: 5\n",
      "  donkey: 4\n",
      "  duck: 6\n",
      "  hat: 5\n",
      "Condition: geons\n",
      "  airplane: 8\n",
      "  car: 8\n",
      "  chair: 7\n",
      "  cup: 8\n",
      "  dog: 7\n",
      "  donkey: 7\n",
      "  duck: 7\n",
      "  hat: 7\n",
      "Condition: realistic\n",
      "  airplane: 5\n",
      "  car: 5\n",
      "  chair: 5\n",
      "  cup: 5\n",
      "  dog: 6\n",
      "  donkey: 5\n",
      "  duck: 5\n",
      "  hat: 5\n",
      "Condition: silhouettes\n",
      "  airplane: 5\n",
      "  car: 5\n",
      "  chair: 5\n",
      "  cup: 6\n",
      "  dog: 4\n",
      "  donkey: 5\n",
      "  duck: 5\n",
      "  hat: 5\n"
     ]
    }
   ],
   "source": [
    "def count_images_per_category(data_dir, categories, conditions):\n",
    "    category_counts = {condition: {category: 0 for category in categories} for condition in conditions}\n",
    "\n",
    "    for condition in conditions:\n",
    "        condition_dir = os.path.join(data_dir, condition)\n",
    "\n",
    "        for category in categories:\n",
    "            category_images = [f for f in os.listdir(condition_dir) if f.lower().startswith(category)]\n",
    "            category_counts[condition][category] = len(category_images)\n",
    "\n",
    "    return category_counts\n",
    "\n",
    "\n",
    "image_counts = count_images_per_category(data_dir, categories, conditions)\n",
    "\n",
    "print(\"Number of images per category:\")\n",
    "for condition, counts in image_counts.items():\n",
    "    print(f\"Condition: {condition}\")\n",
    "    for category, count in counts.items():\n",
    "        print(f\"  {category}: {count}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(10, 6))\n",
    "\n",
    "category_totals = np.zeros(len(categories))\n",
    "for condition in conditions:\n",
    "    counts = [image_counts[condition][cat] for cat in categories]\n",
    "    ax.bar(categories, counts, bottom=category_totals, label=condition)\n",
    "    category_totals += np.array(counts)\n",
    "\n",
    "ax.set_xlabel('Categories')\n",
    "ax.set_ylabel('Number of Images')\n",
    "ax.set_title('Total Images per Category (Stacked by Condition)')\n",
    "ax.legend()\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Evaluating blurred/airplane: 100%|███████████████████████████████████████████████████████| 6/6 [00:00<00:00,  6.31it/s]\n",
      "Evaluating blurred/car: 100%|████████████████████████████████████████████████████████████| 5/5 [00:00<00:00, 10.97it/s]\n",
      "Evaluating blurred/chair: 100%|██████████████████████████████████████████████████████████| 6/6 [00:00<00:00, 10.41it/s]\n",
      "Evaluating blurred/cup: 100%|████████████████████████████████████████████████████████████| 5/5 [00:00<00:00, 11.68it/s]\n",
      "Evaluating blurred/dog: 100%|████████████████████████████████████████████████████████████| 5/5 [00:00<00:00,  8.09it/s]\n",
      "Evaluating blurred/donkey: 100%|█████████████████████████████████████████████████████████| 4/4 [00:00<00:00,  9.05it/s]\n",
      "Evaluating blurred/duck: 100%|███████████████████████████████████████████████████████████| 6/6 [00:00<00:00,  9.99it/s]\n",
      "Evaluating blurred/hat: 100%|████████████████████████████████████████████████████████████| 5/5 [00:00<00:00,  9.92it/s]\n",
      "Evaluating features/airplane: 100%|██████████████████████████████████████████████████████| 6/6 [00:00<00:00, 11.78it/s]\n",
      "Evaluating features/car: 100%|███████████████████████████████████████████████████████████| 5/5 [00:00<00:00, 11.43it/s]\n",
      "Evaluating features/chair: 100%|█████████████████████████████████████████████████████████| 5/5 [00:00<00:00, 11.10it/s]\n",
      "Evaluating features/cup: 100%|███████████████████████████████████████████████████████████| 5/5 [00:00<00:00, 12.30it/s]\n",
      "Evaluating features/dog: 100%|███████████████████████████████████████████████████████████| 5/5 [00:00<00:00, 14.23it/s]\n",
      "Evaluating features/donkey: 100%|████████████████████████████████████████████████████████| 4/4 [00:00<00:00, 10.20it/s]\n",
      "Evaluating features/duck: 100%|██████████████████████████████████████████████████████████| 6/6 [00:00<00:00, 10.73it/s]\n",
      "Evaluating features/hat: 100%|███████████████████████████████████████████████████████████| 5/5 [00:00<00:00, 10.75it/s]\n",
      "Evaluating geons/airplane: 100%|█████████████████████████████████████████████████████████| 8/8 [00:00<00:00,  8.45it/s]\n",
      "Evaluating geons/car: 100%|██████████████████████████████████████████████████████████████| 8/8 [00:00<00:00,  9.34it/s]\n",
      "Evaluating geons/chair: 100%|████████████████████████████████████████████████████████████| 7/7 [00:00<00:00, 10.59it/s]\n",
      "Evaluating geons/cup: 100%|██████████████████████████████████████████████████████████████| 8/8 [00:00<00:00,  8.60it/s]\n",
      "Evaluating geons/dog: 100%|██████████████████████████████████████████████████████████████| 7/7 [00:00<00:00,  8.51it/s]\n",
      "Evaluating geons/donkey: 100%|███████████████████████████████████████████████████████████| 7/7 [00:00<00:00, 10.93it/s]\n",
      "Evaluating geons/duck: 100%|█████████████████████████████████████████████████████████████| 7/7 [00:00<00:00, 12.82it/s]\n",
      "Evaluating geons/hat: 100%|██████████████████████████████████████████████████████████████| 7/7 [00:00<00:00, 12.30it/s]\n",
      "Evaluating realistic/airplane: 100%|█████████████████████████████████████████████████████| 5/5 [00:00<00:00, 17.93it/s]\n",
      "Evaluating realistic/car: 100%|██████████████████████████████████████████████████████████| 5/5 [00:00<00:00, 22.92it/s]\n",
      "Evaluating realistic/chair: 100%|████████████████████████████████████████████████████████| 5/5 [00:00<00:00, 33.02it/s]\n",
      "Evaluating realistic/cup: 100%|██████████████████████████████████████████████████████████| 5/5 [00:00<00:00, 23.59it/s]\n",
      "Evaluating realistic/dog: 100%|██████████████████████████████████████████████████████████| 6/6 [00:00<00:00, 26.76it/s]\n",
      "Evaluating realistic/donkey: 100%|███████████████████████████████████████████████████████| 5/5 [00:00<00:00, 34.40it/s]\n",
      "Evaluating realistic/duck: 100%|█████████████████████████████████████████████████████████| 5/5 [00:00<00:00, 27.50it/s]\n",
      "Evaluating realistic/hat: 100%|██████████████████████████████████████████████████████████| 5/5 [00:00<00:00, 19.66it/s]\n",
      "Evaluating silhouettes/airplane: 100%|███████████████████████████████████████████████████| 5/5 [00:00<00:00,  8.01it/s]\n",
      "Evaluating silhouettes/car: 100%|████████████████████████████████████████████████████████| 5/5 [00:00<00:00, 10.17it/s]\n",
      "Evaluating silhouettes/chair: 100%|██████████████████████████████████████████████████████| 5/5 [00:00<00:00,  8.51it/s]\n",
      "Evaluating silhouettes/cup: 100%|████████████████████████████████████████████████████████| 6/6 [00:00<00:00, 10.47it/s]\n",
      "Evaluating silhouettes/dog: 100%|████████████████████████████████████████████████████████| 4/4 [00:00<00:00,  8.11it/s]\n",
      "Evaluating silhouettes/donkey: 100%|█████████████████████████████████████████████████████| 5/5 [00:00<00:00,  8.54it/s]\n",
      "Evaluating silhouettes/duck: 100%|███████████████████████████████████████████████████████| 5/5 [00:00<00:00,  9.56it/s]\n",
      "Evaluating silhouettes/hat: 100%|████████████████████████████████████████████████████████| 5/5 [00:00<00:00,  8.77it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy for blurred: 69.05%\n",
      "Accuracy for features: 80.49%\n",
      "Accuracy for geons: 54.24%\n",
      "Accuracy for realistic: 100.00%\n",
      "Accuracy for silhouettes: 92.50%\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "def evaluate_model(data_dir, categories, conditions):\n",
    "    accuracy = {condition: 0 for condition in conditions}\n",
    "\n",
    "    for condition in conditions:\n",
    "        correct = 0\n",
    "        total = 0\n",
    "        condition_dir = os.path.join(data_dir, condition)\n",
    "\n",
    "        for category in categories:\n",
    "            category_images = [f for f in os.listdir(condition_dir) if f.lower().startswith(category)]\n",
    "\n",
    "            for img_name in tqdm(category_images, desc=f\"Evaluating {condition}/{category}\"):\n",
    "                img_path = os.path.join(condition_dir, img_name)\n",
    "                image = Image.open(img_path).convert(\"RGB\")\n",
    "\n",
    "                inputs = processor(text=categories, images=image, return_tensors=\"pt\", padding=True)\n",
    "                inputs = {k: v.to(device) for k, v in inputs.items()}\n",
    "                outputs = model(**inputs)\n",
    "\n",
    "                logits_per_image = outputs.logits_per_image\n",
    "                probs = logits_per_image.softmax(dim=-1)\n",
    "\n",
    "                pred = torch.argmax(probs).item()\n",
    "                if categories[pred] == category:\n",
    "                    correct += 1\n",
    "                total += 1\n",
    "\n",
    "        if total > 0:\n",
    "            accuracy[condition] = correct / total\n",
    "        else:\n",
    "            accuracy[condition] = 0\n",
    "\n",
    "    return accuracy\n",
    "accuracy = evaluate_model(data_dir, categories, conditions)\n",
    "\n",
    "for condition, acc in accuracy.items():\n",
    "    print(f\"Accuracy for {condition}: {acc * 100:.2f}%\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(10, 6))\n",
    "\n",
    "conditions = list(accuracy.keys())\n",
    "accuracies = [acc * 100 for acc in accuracy.values()]\n",
    "\n",
    "bar_width = 0.5\n",
    "x = np.arange(len(conditions))\n",
    "\n",
    "bars = ax.bar(x, accuracies, bar_width, label='Accuracy (%)', color='blue', alpha=0.8)\n",
    "\n",
    "ax.set_xlabel('Conditions')\n",
    "ax.set_title('Model Performance Across Conditions')\n",
    "ax.set_xticks(x)\n",
    "ax.set_xticklabels(conditions)\n",
    "ax.set_ylabel('Accuracy (%)')\n",
    "ax.legend()\n",
    "\n",
    "for bar in bars:\n",
    "    height = bar.get_height()\n",
    "    ax.text(bar.get_x() + bar.get_width() / 2, height + 2, f'{height:.1f}',\n",
    "            ha='center', va='bottom', fontsize=10)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## b)In which conditions is your models better/worse/similar to babies (Fig. 3A in this paper: https://osf.io/preprints/psyarxiv/83gaeLinks to an external site.)?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Comparison of Model and Babies' Performance Across Conditions\n",
    "\n",
    "## 1. Blurred\n",
    "The model achieved an accuracy of 69.05%. Babies perform moderately in this condition as they rely on the overall shape and structure for object recognition. The model performs reasonably well, demonstrating robustness to blurring effects, which aligns with the moderate performance observed in infants. The performance is similar to that of infants.\n",
    "\n",
    "## 2. Features\n",
    "The model achieved an accuracy of 80.49%. Babies' performance in this condition is low, as they often require holistic visual context and struggle with isolated features. The model significantly outperforms infants. This result is expected since models like CLIP are trained to leverage isolated discriminative features for classification. The model performs better than infants.\n",
    "\n",
    "## 3. Geons\n",
    "The model achieved an accuracy of 54.24%. Babies' performance in this condition is low to moderate, as geons are abstract representations and can be challenging without prior experience. The model shows moderate performance but likely outperforms infants, who find abstract representations more difficult. The model performs better than infants, though not as robustly as in other conditions.\n",
    "\n",
    "## 4. Realistic\n",
    "The model achieved an accuracy of 100.00%. Babies perform highly in this condition, as realistic images closely resemble the natural stimuli they encounter in their environment. The model achieves perfect accuracy, reflecting its ability to generalize well with naturalistic images. Infants are also highly adept in this condition, so the performance is similar or slightly better compared to infants.\n",
    "\n",
    "## 5. Silhouettes\n",
    "The model achieved an accuracy of 92.50%. Babies perform moderately in this condition, as they may rely on edge-based recognition but often require additional details. The model excels with silhouettes, leveraging edge-based and outline information to outperform infants, who struggle more with such limited cues. The model performs better than infants.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Q3)From the output of the final layer of the vision encoder (before the projection layer into the shared embedding space) take embeddings for each image in the dataset. Apply t-SNE to those embeddings and plot them in a 2D space - plot little images of objects in the embedding space (something like Fig. 4 in this paper: https://osf.io/preprints/psyarxiv/83gaeLinks to an external site. but no need for color coding of bounding boxes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Extracting Embeddings blurred/airplane: 100%|████████████████████████████████████████████| 6/6 [00:00<00:00,  7.92it/s]\n",
      "Extracting Embeddings blurred/car: 100%|█████████████████████████████████████████████████| 5/5 [00:00<00:00, 10.58it/s]\n",
      "Extracting Embeddings blurred/chair: 100%|███████████████████████████████████████████████| 6/6 [00:00<00:00,  9.09it/s]\n",
      "Extracting Embeddings blurred/cup: 100%|█████████████████████████████████████████████████| 5/5 [00:00<00:00,  9.92it/s]\n",
      "Extracting Embeddings blurred/dog: 100%|█████████████████████████████████████████████████| 5/5 [00:00<00:00,  6.59it/s]\n",
      "Extracting Embeddings blurred/donkey: 100%|██████████████████████████████████████████████| 4/4 [00:00<00:00,  6.74it/s]\n",
      "Extracting Embeddings blurred/duck: 100%|████████████████████████████████████████████████| 6/6 [00:00<00:00,  9.43it/s]\n",
      "Extracting Embeddings blurred/hat: 100%|█████████████████████████████████████████████████| 5/5 [00:00<00:00, 10.10it/s]\n",
      "Extracting Embeddings features/airplane: 100%|███████████████████████████████████████████| 6/6 [00:00<00:00, 10.86it/s]\n",
      "Extracting Embeddings features/car: 100%|████████████████████████████████████████████████| 5/5 [00:00<00:00, 12.31it/s]\n",
      "Extracting Embeddings features/chair: 100%|██████████████████████████████████████████████| 5/5 [00:00<00:00, 11.03it/s]\n",
      "Extracting Embeddings features/cup: 100%|████████████████████████████████████████████████| 5/5 [00:00<00:00, 10.50it/s]\n",
      "Extracting Embeddings features/dog: 100%|████████████████████████████████████████████████| 5/5 [00:00<00:00, 11.91it/s]\n",
      "Extracting Embeddings features/donkey: 100%|█████████████████████████████████████████████| 4/4 [00:00<00:00, 10.81it/s]\n",
      "Extracting Embeddings features/duck: 100%|███████████████████████████████████████████████| 6/6 [00:00<00:00,  9.70it/s]\n",
      "Extracting Embeddings features/hat: 100%|████████████████████████████████████████████████| 5/5 [00:00<00:00, 10.19it/s]\n",
      "Extracting Embeddings geons/airplane: 100%|██████████████████████████████████████████████| 8/8 [00:00<00:00,  8.19it/s]\n",
      "Extracting Embeddings geons/car: 100%|███████████████████████████████████████████████████| 8/8 [00:00<00:00, 10.44it/s]\n",
      "Extracting Embeddings geons/chair: 100%|█████████████████████████████████████████████████| 7/7 [00:00<00:00, 11.55it/s]\n",
      "Extracting Embeddings geons/cup: 100%|███████████████████████████████████████████████████| 8/8 [00:00<00:00, 10.65it/s]\n",
      "Extracting Embeddings geons/dog: 100%|███████████████████████████████████████████████████| 7/7 [00:00<00:00, 12.68it/s]\n",
      "Extracting Embeddings geons/donkey: 100%|████████████████████████████████████████████████| 7/7 [00:00<00:00, 14.10it/s]\n",
      "Extracting Embeddings geons/duck: 100%|██████████████████████████████████████████████████| 7/7 [00:00<00:00, 13.63it/s]\n",
      "Extracting Embeddings geons/hat: 100%|███████████████████████████████████████████████████| 7/7 [00:00<00:00, 14.26it/s]\n",
      "Extracting Embeddings realistic/airplane: 100%|██████████████████████████████████████████| 5/5 [00:00<00:00, 16.87it/s]\n",
      "Extracting Embeddings realistic/car: 100%|███████████████████████████████████████████████| 5/5 [00:00<00:00, 22.67it/s]\n",
      "Extracting Embeddings realistic/chair: 100%|█████████████████████████████████████████████| 5/5 [00:00<00:00, 35.60it/s]\n",
      "Extracting Embeddings realistic/cup: 100%|███████████████████████████████████████████████| 5/5 [00:00<00:00, 32.14it/s]\n",
      "Extracting Embeddings realistic/dog: 100%|███████████████████████████████████████████████| 6/6 [00:00<00:00, 35.85it/s]\n",
      "Extracting Embeddings realistic/donkey: 100%|████████████████████████████████████████████| 5/5 [00:00<00:00, 42.98it/s]\n",
      "Extracting Embeddings realistic/duck: 100%|██████████████████████████████████████████████| 5/5 [00:00<00:00, 31.89it/s]\n",
      "Extracting Embeddings realistic/hat: 100%|███████████████████████████████████████████████| 5/5 [00:00<00:00, 20.10it/s]\n",
      "Extracting Embeddings silhouettes/airplane: 100%|████████████████████████████████████████| 5/5 [00:00<00:00,  8.00it/s]\n",
      "Extracting Embeddings silhouettes/car: 100%|█████████████████████████████████████████████| 5/5 [00:00<00:00, 10.85it/s]\n",
      "Extracting Embeddings silhouettes/chair: 100%|███████████████████████████████████████████| 5/5 [00:00<00:00,  9.06it/s]\n",
      "Extracting Embeddings silhouettes/cup: 100%|█████████████████████████████████████████████| 6/6 [00:00<00:00, 10.41it/s]\n",
      "Extracting Embeddings silhouettes/dog: 100%|█████████████████████████████████████████████| 4/4 [00:00<00:00,  7.03it/s]\n",
      "Extracting Embeddings silhouettes/donkey: 100%|██████████████████████████████████████████| 5/5 [00:00<00:00,  7.95it/s]\n",
      "Extracting Embeddings silhouettes/duck: 100%|████████████████████████████████████████████| 5/5 [00:00<00:00,  7.64it/s]\n",
      "Extracting Embeddings silhouettes/hat: 100%|█████████████████████████████████████████████| 5/5 [00:00<00:00,  8.50it/s]\n"
     ]
    }
   ],
   "source": [
    "def extract_embeddings_and_labels(data_dir, categories, conditions):\n",
    "    embeddings = []\n",
    "    image_paths = []\n",
    "    labels = []\n",
    "\n",
    "    for condition in conditions:\n",
    "        condition_dir = os.path.join(data_dir, condition)\n",
    "\n",
    "        for category in categories:\n",
    "            category_images = [f for f in os.listdir(condition_dir) if f.lower().startswith(category)]\n",
    "\n",
    "            for img_name in tqdm(category_images, desc=f\"Extracting Embeddings {condition}/{category}\"):\n",
    "                img_path = os.path.join(condition_dir, img_name)\n",
    "                image = Image.open(img_path).convert(\"RGB\")\n",
    "\n",
    "                inputs = processor(images=image, return_tensors=\"pt\").to(device)\n",
    "                \n",
    "                vision_outputs = model.vision_model(pixel_values=inputs['pixel_values'])\n",
    "                embedding = vision_outputs.last_hidden_state.mean(dim=1).squeeze(0).cpu().detach().numpy()\n",
    "\n",
    "                embeddings.append(embedding)\n",
    "                image_paths.append(img_path)\n",
    "                labels.append(category)\n",
    "\n",
    "    return np.array(embeddings), image_paths, labels\n",
    "\n",
    "def apply_tsne(embeddings):\n",
    "    tsne = TSNE(n_components=2, random_state=42)\n",
    "    reduced_embeddings = tsne.fit_transform(embeddings)\n",
    "    return reduced_embeddings\n",
    "\n",
    "embeddings, image_paths, labels = extract_embeddings_and_labels(data_dir, categories, conditions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Adding images to plot: 100%|███████████████████████████████████████████████████████| 223/223 [00:00<00:00, 2232.08it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABEYAAAMWCAYAAAD8kYNsAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOz9eZzkV13o/7/O+eyfWnqZfbIR9k0RRCJEbwIIISQihkVZJIGAXgH9IqjI4i8BWcWLcBVBLhBAQMQHYREEASHxXgkS4AoXFAXJnvRMT2+1ffZzfn98qqqrunsmk8lMZia8nzyK7qqu5Xw+VZ3HnHe/F2WttQghhBBCCCGEEEL8GNLHewFCCCGEEEIIIYQQx4sERoQQQgghhBBCCPFjSwIjQgghhBBCCCGE+LElgREhhBBCCCGEEEL82JLAiBBCCCGEEEIIIX5sSWBECCGEEEIIIYQQP7YkMCKEEEIIIYQQQogfWxIYEUIIIYQQQgghxI8tCYwIIYQQQgghhBDix5YERoQQ4nZ89atf5fLLL2d1dfWwH2Ot5aMf/Sg///M/z86dOwnDkFNPPZXzzjuP97znPVP3VUqhlOJNb3rTpud5//vfj1KKb3zjG+PbLr/88vFjtrpcf/31W65pcXER3/f51V/91YOuu9PpEMcxT3rSkwA499xzOffccw/7uI+V0TFPOtZre8Mb3sAnP/nJTbdfddVVKKW46qqrjtlrHw1/9md/xr3vfW9830cpddDP71afsbujQ/3OXHLJJUftde7K87nV78XBKKW4/PLLx9dH6zzYfy+EEEKIHyfu8V6AEEKc6L761a/ymte8hksuuYTZ2dnDeswrXvEK3vzmN/OCF7yA3/u936PVanHDDTfw5S9/mU996lM8//nP3/SYN73pTfz6r/868/Pzh/Uan//855mZmdl0+549e7a8/44dO3jSk57EJz/5SVZWVpibm9t0n49+9KMkScKll14KwF/8xV8c1lqOh2O9tje84Q089alP5clPfvLU7Q972MO45ppreOADH3hMX//O+Nd//Vd++7d/m+c///lcfPHFuK5Lq9U63ss67p761Kfyspe9bNPtO3bsOA6rOb4uuOACrrnmmoP+90IIIYT4cSKBESGEOMqSJOFtb3sbz3nOc3j3u9899bNLLrkEY8ymx/zCL/wCV111Fa9//ev5H//jfxzW6/z0T/8027dvv0Nru/TSS/n4xz/Ohz/8YV784hdv+vn73vc+du3axQUXXABwQm/+j9fa2u02P/uzP3tcXvtwfe973wPgBS94AY94xCOO82pOHLt27Trh37u7yo4dO34sA0JCCCHEVqSURgghDuHyyy/n937v9wA488wzx6n3hyqj6Pf7ZFl20L/Ear35P733u9/9uPTSS3nHO97BDTfccFTWvpXzzjuPU089lSuuuGLTz/793/+df/mXf+E5z3kOrlvHzbcqV3nnO9/JQx7yEJrNJq1Wi/vf//688pWvHP/8YOn9W6Xu/83f/A2Pf/zj2bNnD1EU8YAHPIA/+IM/oN/v3+6xbFzbJZdcctBSiVEJQZqmvOxlL+OnfuqnmJmZYX5+nkc+8pF86lOfmnpupRT9fp8PfOAD4+cYvdbBSmk+/elP88hHPpI4jmm1WjzucY/jmmuumbrP6Nx873vf4xnPeAYzMzPs2rWL5z3veaytrd3uMUMdvHrIQx5CGIbMz8/zy7/8y/z7v//71Hl59rOfDcBZZ511RKUil1xyCc1mk+9///ucd955NBoN9uzZMy73+trXvsbP/dzP0Wg0uO9978sHPvCBqccvLi7ywhe+kAc+8IE0m0127tzJYx7zGP73//7fm17r5ptv5qlPfSqtVovZ2Vme9axnce2116KU4v3vf//Ufb/xjW/wpCc9ifn5ecIw5KEPfSgf+9jH7tCxHetjH1lZWeG5z30u8/PzNBoNfvEXf5Ef/ehHm+73pS99icc+9rG0223iOObss8/mH//xHzfd77Of/Sw/9VM/RRAEnHnmmfzJn/zJlq/b6XR4wQtewLZt22g2mzzhCU/gP//zPzfdb6vfx3PPPZcHP/jBXHvttfz8z/88cRxzz3vekze96U2bArrf+973ePzjH08cx+zYsYMXvehFfPazn930u/F//+//5cILL2Tnzp0EQcDevXu54IILuPnmm7dcvxBCCHE8SGBECCEO4fnPfz6/9Vu/BcCVV17JNddcwzXXXMPDHvawgz5m+/bt3Pve9+Yv/uIveOtb38r3v/99rLW3+1qXX345juPwh3/4h4e1tqqqKMty6lJV1SEfo7Xmkksu4Vvf+hbf/va3p342CpY873nPO+jjP/rRj/LCF76Qc845h0984hN88pOf5Hd+53cOK5CxlR/84Ac88YlP5L3vfS+f//zneclLXsLHPvYxfvEXf/EOP9cf/uEfjt+f0WUUIBhll2RZxvLyMr/7u7/LJz/5Sf76r/+an/u5n+Oiiy7igx/84Pi5rrnmGqIo4olPfOL4uQ5VuvORj3yEX/qlX6LdbvPXf/3XvPe972VlZYVzzz2X//N//s+m+z/lKU/hvve9Lx//+Mf5gz/4Az7ykY/wO7/zO7d7jG984xu59NJLedCDHsSVV17J29/+dr7zne/wyEc+kh/84AdAXWL06le/Gqjf02uuueawP1OTiqLgoosu4oILLuBTn/oU559/Pq94xSt45StfycUXX8zznvc8PvGJT3C/+92PSy65hG9+85vjxy4vLwNw2WWX8dnPfpYrrriCe97znpx77rlTm+Z+v8+jH/1ovvKVr/DmN7+Zj33sY+zatYtf+ZVf2bSer3zlK5x99tmsrq7yrne9i0996lP81E/9FL/yK7+yKYByMNbaTb8zZVlu+v28M8c+cumll6K15iMf+Qhve9vb+PrXv86555471evlQx/6EI9//ONpt9t84AMf4GMf+xjz8/Ocd955U8GRf/zHf+SXfumXaLVafPSjH+Utb3kLH/vYxzYFOK21PPnJT+av/uqveNnLXsYnPvEJfvZnf5bzzz//sM4PwMLCAs961rN49rOfzac//enxsX/oQx8a3+e2227jnHPO4T/+4z945zvfyQc/+EG63e6mLLR+v8/jHvc49u3bxzve8Q6++MUv8ra3vY3TTz+dbrd72GsSQgghjjkrhBDikN7ylrdYwF533XWH/Zivf/3r9vTTT7eABWyr1bIXXnih/eAHP2iNMVP3BeyLXvQia621r3rVq6zW2n7729+21lp7xRVXWMBee+214/tfdtll4+fdeLnXve51u2v70Y9+ZJVS9rd/+7fHtxVFYXfv3m3PPvvsqfuec8459pxzzhlff/GLX2xnZ2cP+fyj9W00OpaDnUdjjC2Kwl599dUWGJ+Dgz3nxrVt9LGPfcwqpewrX/nKg96nLEtbFIW99NJL7UMf+tCpnzUaDXvxxRdvesxXvvIVC9ivfOUr1lprq6qye/futT/xEz9hq6oa36/b7dqdO3faRz3qUZuO44//+I+nnvOFL3yhDcNw02dj0srKio2iyD7xiU+cuv3GG2+0QRDYZz7zmePbtvrcHMxW97344ostYD/+8Y+PbyuKwu7YscMC9lvf+tb49qWlJes4jn3pS1960NcYnefHPvax9pd/+ZfHt7/jHe+wgP3c5z43df/f+I3fsIC94oorxrfd//73tw996ENtURRT973wwgvtnj17ps79Vg72OwPYv/qrvzpqxz46n5PHaa21//zP/2wB+7rXvc5aa22/37fz8/P2F3/xF6fuV1WVfchDHmIf8YhHjG8766yz7N69e22SJOPbOp2OnZ+fn/q9+NznPmcB+/a3v33qOV//+tdbwF522WWb1jn5+3jOOedYwP7Lv/zL1OMf+MAH2vPOO298/fd+7/esUsp+73vfm7rfeeedN/W78Y1vfMMC9pOf/KQVQgghTmSSMSKEEEfIGHPQbI2f+Zmf4Yc//CGf//zneeUrX8kjH/lI/vEf/5HnPOc5POlJTzpoBsnv//7vMz8/z8tf/vLbff0vfelLXHvttVOXraaobHTmmWfy6Ec/mg9/+MPkeQ7A5z73ORYWFg6ZLQLwiEc8gtXVVZ7xjGfwqU99igMHDtzu6x3Kj370I575zGeye/duHMfB8zzOOeccgKnykDvq6quv5td+7dd49rOfzetf//qpn/3t3/4tZ599Ns1mE9d18TyP9773vUf8ev/xH//Brbfeyq/92q9NlUk1m02e8pSn8LWvfY3BYDD1mNHUn5Gf/MmfJE1T9u/ff9DXueaaa0iSZFNZzGmnncZjHvOYLcsv7gylFE984hPH113X5d73vjd79uzhoQ996Pj2+fl5du7cuakE7F3vehcPe9jDCMNwfJ7/8R//ceo8X3311bRaLZ7whCdMPfYZz3jG1PUf/vCHfP/73+dZz3oWwNTv3ROf+ERuu+02/uM//uN2j+npT3/6pt+Za6+9duo4j8axA+O1jjzqUY/ijDPO4Ctf+QpQN3VeXl7m4osvnjoeYwxPeMITuPbaa+n3+/T7fa699louuugiwjAcP1+r1dqUWTV67o2v/cxnPvN2z83I7t27N/Wl+cmf/MmpY7z66qt58IMfvKnPz8b37d73vjdzc3O8/OUv513vehf/9m//dtjrEEIIIe5KEhgRQogj9NrXvhbP88aXe93rXlM/9zyP8847j9e//vX8wz/8AzfddBPnnnsun/nMZ/jc5z635XO2221e/epX8/nPf368yTmYhzzkITz84Q+fujz4wQ8+rLVfeumlLC0t8elPfxqoSy6azSZPf/rTD/m4X/u1X+N973sfN9xwA095ylPYuXMnZ511Fl/84hcP63Un9Xo9fv7nf55/+Zd/4XWvex1XXXUV1157LVdeeSVQN7E9Et/73vd48pOfzM///M/z3ve+d+pnV155JU9/+tM55ZRT+NCHPsQ111zDtddey/Oe9zzSND2i11taWgK2nga0d+9ejDGsrKxM3b5t27ap60EQAIc+5tt7ndHPj5Y4jqc24gC+7285Ncn3/anz99a3vpXf/M3f5KyzzuLjH/84X/va17j22mt5whOeMHWMS0tL7Nq1a9Pzbbxt3759APzu7/7u1O+c53m88IUvBDisIN2OHTs2/c48/OEP33RMd+bYR3bv3r3lbaP3aXRMT33qUzcd05vf/GastSwvL7OysoIx5qDPN2lpaQnXdTd9vrZ67MFsfCzUn88jed9mZma4+uqr+amf+ile+cpX8qAHPYi9e/dy2WWXURTFYa9JCCGEONZkKo0QQhyhX//1X+fCCy8cXx9tbg9m27ZtvOQlL+Gqq67iu9/97qa/Uo/85m/+Jm9/+9t5+ctfzm/+5m8e1TWPXHTRRczNzfG+972Pc845h8985jM85znPodls3u5jn/vc5/Lc5z6Xfr/PP/3TP3HZZZdx4YUX8p//+Z+cccYZ4w1llmVT52TjxvXLX/4yt956K1ddddU4SwSY6sFwR91888084QlP4PTTT+fjH/84nudN/fxDH/oQZ555Jn/zN38z1SA2y7Ijfs3RRvK2227b9LNbb70VrfWWo5GP9uvc0QlFx9KHPvQhzj33XN75zndO3b6xr8S2bdv4+te/vunxCwsLU9dHx/aKV7yCiy66aMvXvN/97ndnlnzUbTyG0W33vve9gfVj+rM/+7ODTsrZtWsXRVGglDro803atm0bZVmytLQ0FeDY6rF3xrZt28aBnUOtB+AnfuIn+OhHP4q1lu985zu8//3v57WvfS1RFPEHf/AHR3VdQgghxJGSjBEhhLgdB/tr/t69e6f+6vwTP/ETQN248WB/vR+VEezdu/egr+f7Pq973eu49tpr+du//dujcQibhGHIM5/5TL7whS/w5je/maIobreMZqNGo8H555/Pq171KvI8H4+Ivcc97gHAd77znan7/93f/d3U9VFgYmNA6S//8i/v0DpG1tbWOP/881FK8fd///e02+1N91FK4fv+VFBkYWFh01Sa0boOJ2vlfve7H6eccgof+chHpkqk+v0+H//4x8eTau6sRz7ykURRNNUEE+pg0Je//GUe+9jH3unXOFqUUpve1+985zubpvScc845dLvdTRlUH/3oR6eu3+9+9+M+97kP3/72t7fM+Hj4wx9Oq9U6NgdzhD784Q9PXf/qV7/KDTfcMJ5udPbZZzM7O8u//du/HfSYfN+n0WjwiEc8giuvvHIqM6Xb7W76nXr0ox+95Wt/5CMfOarHds455/Dd7353U2nMxvdtklKKhzzkIfzpn/4ps7OzfOtb3zqqaxJCCCHuDMkYEUKI2zEKeLz97W/n4osvxvM87ne/+x10I7a2tsY97nEPnva0p/ELv/ALnHbaafR6Pa666ire/va384AHPOCgf/UeecYznsGf/MmfHLTkBuCb3/wmMzMzm25/4AMfuGVQYKPReOC3vvWt3P/+9+dRj3rU7T7mBS94AVEUcfbZZ7Nnzx4WFhZ44xvfyMzMDD/zMz8DwBOf+ETm5+e59NJLee1rX4vrurz//e/npptumnquRz3qUczNzfHf//t/57LLLsPzPD784Q9vmpZzuJ75zGfyb//2b7z73e/mpptumnq9U089lVNPPZULL7yQK6+8khe+8IU89alP5aabbuKP/uiP2LNnz3iqy8hP/MRPcNVVV/F3f/d37Nmzh1artWVWgtaaP/7jP+ZZz3oWF154Ib/xG79BlmW85S1vYXV1dTzm9c6anZ3lD//wD3nlK1/Jc57zHJ7xjGewtLTEa17zGsIw5LLLLjsqr3M0XHjhhfzRH/0Rl1122Xh6yWtf+1rOPPNMyrIc3+/iiy/mT//0T3n2s5/N6173Ou5973vzuc99jn/4h38Apkdb/+Vf/iXnn38+5513HpdccgmnnHIKy8vL/Pu//zvf+ta3DiuIuG/fPr72ta9tur3dbm/ql3FnfeMb3+D5z38+T3va07jpppt41atexSmnnDIu/Wk2m/zZn/0ZF198McvLyzz1qU9l586dLC4u8u1vf5vFxcVxxs0f/dEf8YQnPIHHPe5xvOxlL6OqKt785jfTaDTGE4AAHv/4x/Pf/tt/4/d///fp9/s8/OEP55//+Z/5q7/6q6N6bC95yUt43/vex/nnn89rX/tadu3axUc+8hG+//3vA+vv22c+8xn+4i/+gic/+cnc8573xFrLlVdeyerqKo973OOO6pqEEEKIO+W4tn4VQoiTxCte8Qq7d+9eq7WemrqwlSzL7J/8yZ/Y888/355++uk2CAIbhqF9wAMeYH//93/fLi0tTd2fiak0k77whS+Mp2Yc7lQawH7xi1887ON66EMfuuWUlJGNk18+8IEP2Ec/+tF2165d1vd9u3fvXvv0pz/dfuc735l63Ne//nX7qEc9yjYaDXvKKafYyy67zL7nPe/ZNAXjq1/9qn3kIx9p4zi2O3bssM9//vPtt771rU0TSQ5nKs0ZZ5xx0HMyOY3jTW96k73HPe5hgyCwD3jAA+z/+l//a8vn/9d//Vd79tln2ziOLTB+rY1TaUY++clP2rPOOsuGYWgbjYZ97GMfa//5n/956j6j11lcXJy6/fYm9kx6z3veY3/yJ3/S+r5vZ2Zm7C/90i9tmg5yNKbSNBqNTfc955xz7IMe9KBNt59xxhn2ggsuGF/Pssz+7u/+rj3llFNsGIb2YQ97mP3kJz9pL774YnvGGWdMPfbGG2+0F110kW02m7bVatmnPOUp9u///u8tYD/1qU9N3ffb3/62ffrTn2537txpPc+zu3fvto95zGPsu971rts9zkP9zkxOY7qzxz46n1/4whfsr/3ar9nZ2dnxNKEf/OAHmx5/9dVX2wsuuMDOz89bz/PsKaecYi+44AL7t3/7t1P3+/SnPz1+308//XT7pje9acvP7erqqn3e855nZ2dnbRzH9nGPe5z9/ve/f9hTabY6xq3et+9+97v2F37hF2wYhnZ+ft5eeuml9gMf+MDURKnvf//79hnPeIa9173uZaMosjMzM/YRj3iEff/737/pNYQQQojjSVl7kNEIQgghhBDHwRve8AZe/epXc+ONN3Lqqace7+WIw/Trv/7r/PVf/zVLS0v4vn+8lyOEEEIcNimlEUIIIcRx8+d//ucA3P/+96coCr785S/zP//n/+TZz362BEVOYK997WvZu3cv97znPen1enzmM5/hPe95D69+9aslKCKEEOKkI4ERIYQQQhw3cRzzp3/6p1x//fVkWcbpp5/Oy1/+cl796lcf76WJQ/A8j7e85S3cfPPNlGXJfe5zH9761rfy//1//9/xXpoQQghxh0kpjRBCCCGEEEIIIX5sybheIYQQQgghhBBC3GXucY97oJTadHnRi150XNYjpTRCCCGEEEIIIYS4y1x77bVUVTW+/t3vfpfHPe5xPO1pTzsu65FSGiGEEEIIIYQQQhw3L3nJS/jMZz7DD37wA5RSd/nrn/QZI8YYbr31Vlqt1nE5gUIIIYQQQgghThzWWrrdLnv37kXru2f3iDRNyfP8eC9jirV20548CAKCIDjk4/I850Mf+hAvfelLj9ue/qQPjNx6662cdtppx3sZQgghhBBCCCFOIDfddNPdcvR7mqaceY8dLOzrHe+lTGk2m/R602u67LLLuPzyyw/5uE9+8pOsrq5yySWXHLvF3Y6TPjDSarWA+kPfbreP82qEEEIIIYQQQhxPnU6H0047bbxXvLvJ85yFfT1u+N5v024dOhvjrtLpZpzxoP+5aV9+e9kiAO9973s5//zz2bt377Fc4iEd08DIG9/4Rq688kq+//3vE0URj3rUo3jzm9/M/e53v/F9rLW85jWv4d3vfjcrKyucddZZvOMd7+BBD3rQYb3GKNWm3W5LYEQIIYQQQgghBMDdvtVCuxXQbp8YgZGRO7ovv+GGG/jSl77ElVdeeQxXdfuOacHV1VdfzYte9CK+9rWv8cUvfpGyLHn84x9Pv98f3+eP//iPeetb38qf//mfc+2117J7924e97jH0e12j+XShBBCCCGEEEKIk5Y9wf53JK644gp27tzJBRdccJTPzh1zTDNGPv/5z09dHx30N7/5Tf7bf/tvWGt529vexqte9SouuugiAD7wgQ+wa9cuPvKRj/Abv/Ebx3J5QgghhBBCCCGEOA6MMVxxxRVcfPHFuO7x7fJxl7boXVtbA2B+fh6A6667joWFBR7/+MeP7xMEAeeccw5f/epXt3yOLMvodDpTFyGEEEIIIYQQQpw8vvSlL3HjjTfyvOc973gv5a5rvmqt5aUvfSk/93M/x4Mf/GAAFhYWANi1a9fUfXft2sUNN9yw5fO88Y1v5DWvec2xXawQQgghhBBCCHECM1jMEZawHG1Hso7HP/7xWHtirP8uyxh58YtfzHe+8x3++q//etPPNjbF2Wr+8cgrXvEK1tbWxpebbrrpmKxXCCGEEEIIIYQQd393ScbIb/3Wb/HpT3+af/qnf5qaI717926gzhzZs2fP+Pb9+/dvyiIZCYLgsEb+CCGEEEIIIYQQQtyeY5oxYq3lxS9+MVdeeSVf/vKXOfPMM6d+fuaZZ7J7926++MUvjm/L85yrr76aRz3qUcdyaUIIIYQQQgghxEnreE+hORpTaU4UxzRj5EUvehEf+chH+NSnPkWr1Rr3FJmZmSGKIpRSvOQlL+ENb3gD97nPfbjPfe7DG97wBuI45pnPfOaxXJoQQgghhBBCCCHEsQ2MvPOd7wTg3HPPnbr9iiuu4JJLLgHg93//90mShBe+8IWsrKxw1lln8YUvfIFWq3UslyaEEEIIIYQQQghxbAMjh9NhVinF5ZdfzuWXX34slyKEEEIIIYQQQtxtnEglLCfKOo7UXTaVRgghhBBCCCGEEOJEI4ERIYQQQgghhBBC/Ni6S8b1CiGEEEIIIYQQ4ugx1mIOo33FXeFEWceRksCIEEIIIYT4sXKoPnhKqbtwJUIIIU4EEhgRQgghhBA/FkYBEWstxpjx7aNgiNZ6fB8JkAghTnR2eDkRnCjrOFISGBFCCCGEEHd71tpxQKSqKowxGFNvK7TWmy7WWgmOCCHEjwkJjAghhBBCiLu1yaBIURTjS1mWADiOg+d5eJ6H4zg4joPW9YwCCY4IIcTdnwRGhBBCCCHE3d4oKJKmKUmSkCQZeVFgAd/zCMOQMPQJfB/P83BddxwcAQmQCCFOPAaLOUGKWE6UdRwpCYwIIYQQQoi7rVGmSFmWpGlKr5/Q7Xbp9XokSYK1ljAMabValGWDMjSEQYXv+7iuO84ekdIaIYS4+5LAiBBCCCGEuFsaldBURUmWZQwGA7rdHp1Oh06nQ7fbp6oqoiiiqizWQlmWVKVPWZYEQTAOkEhwRAgh7r4kMCKEEEIIIe42RsELay0YiykLyiwh6Sf0ej0G/T5Jv0+/12N1ZZUsz4mbTRzHQSlFkWcUqUccBJRxjG02AKaCIyClNUKI488O/3ciOFHWcaQkMCKEEEIIIU561lgYxipMZcBaMAVllpP2+vTX1uj3+iSDAXmWkqcZSdJnMEgxxhCHIa7rkqcOiYYs8Gm22yhAK4VWCjW8AJI9IoQQdyMSGBFCCCGEECc/BRiLtQZMhTWGKs/Je10GK8v0V1ZJkoSqNKg8xykyVJZSpX1yrciTAZnrUtqKtCzIfY8qTVFK4ToOrtJoCzgOaCVBESGEuBuRwIgQQgghhDipGWOwlQFTYssKW5V1CU2SkK0sMdi3n8HKKnlRgtY41uBmKV6e4qRpPX0mGVApKLIM0oTS0VDkeJ5H6Hp4SqHCAMf30K4PjpKsESHEcWXqisETwomyjiMlgREhhBBCCHFSGvX7sJXBFjlVlmKyHFNkmCyj6HdJF/eTLNxGutahQuH6Po5S+P0BQZoQ5gmeo3GTAVQlVb+P6fXA0aTGkPgBievhW4NuNiEKIQSNj3ad43wGhBBCHA0SGBFCCCGEECclpRSmMpgypxj0KTtdzKCLSVOqZEC2tsZgcR/J/n0U/QHW8yCK0VrhDBL8QY8oy3CVxuv10I6D6nZQvR44DpVS5H5AqhRBWaCyHNtu4QPacbBWS8aIEELcDUhgRAghhBBCnJSstRhTUaQZ2eoq2eJ+zNoq5aBH1euRd1YZHDhAurJEWVTYMMTNc7SjUUmKO+jjFSWOBdfROMbidLvowQDtevWUGqVIqhI3TTGDPiabB63RfoDruiCBESHEcWKHlxPBibKOIyWBESGEEEIIcdKx1taXsqAY9Okf2E96842UBxapul2qfo+80yFdXSEbDMiVQjVbFJXBOg6mKLBpgi0qVFlhyxJVFjjdPmRZHRgpC6osI+0PsGsdss48RVagwgiv1cYJLCjpMyKEECc7CYwIIYQQQoiTk7FURUnW7dBfWKB/w/WU+/ZjemtUvT5lt0va75OUJWkYotFYwHc9qrKkyHPyosJkBf5ggE4z1GCAzQtwXcwgoVzrMlhZIVlt43Y75MbgzcwQ7diGF8fSZ0QIIe4GJDAihBBCCCFOOkopDBaTZ2Rrq/T3LdC56WbKfQuYXrcOjPR75FlGqh3yVhPtBhgU1vMorSEpDf3C4BQ5TpajBwOcJIGywCoX62hwXKrAp+y0IcnIPY94zx5msnyctSIZI0KI48FgMSdIEcuJso4jJYERIYQQQghx0rHWgqmo8oxsbY3e/v109i2Q799P1etRDQaYJKEoCnLPp8JB62492tf1KFxNX7t0rUVnBbrTxel28JMEVVb1iyhFpTS57zPIcioU1Uyb+W6XMsvAVoB3XM+DEEKIO08CI0IIIYQQ4uRkDVWek3fX6C0fYHV5hWS1Q5UMqNIUO0jqqTWewSoXRztQFBjfp4giBnGTRDlgwEsSgm4PkgGqNGgDlbJUGpIwpuN6lM0WziAjSRKKssSak/svpEKIk5sFzPFexNDJ/l9DCYwIIYQQQoiTjwVjLCbPyft9Bp0OK90e3SShSDNsmkNZ4pQGx1hcnWKVosgyTOCTK03RmKHwfKzuk5YVaZqhkxRdVSgLVkGuFIlyGVhLiSLGkhlLZQx2w1ZAymqEEOLkJIERIYQQQghx8lHUwZGyokhT+oOElTRhNcvIixJjKhwDkbWEpiLIM5Q1lHlOWYXkjRal46D8gMrzKdDkZYlXlTilrZ8bS641qbWkrgthQBWEWNfFKlUvwg6/2DpIYszB/347CpoopSSIIoQQJxAJjAghhBBCiJOTtVhbUZYVSZ7TyUuWiorMWBzt4LsWtMZF4aOgMhQmx2hNYcC4HgQheD6VdiiVosTBOgYsVFgKpckch9IPcOMmbqOBE4YoxwXUOChijFkfIWzXM0kmgyGjy/ryJTgihDhylhOnhOVEWceRksCIEEIIIYQ4aVkUFZBbGJiKvjWUWuM6Hk4QgrUoY9EYMBZjLRmaQkHlOCjXAc/DBh4mijDWouwwY8RaCiDzfUo/wG/E+O02fiPG8TwYBjXKssQYQ1VV46+joMfoorXGcRwcx0Frjdb6uJ43IYQQ6yQwIoQQQgghTk5aoRynzvjwfQrHo3Q9rOPhRiGR7xGhCU2FW1booiArcnLHJVN1YMJojfJcCCNoNsH1qNsZKqwxlJWhCCOqwMeJY6J2m6DVxg181DC4URRFPf0mz8nznLIsx1kjo4CI67r4vo/v+7iuOw6SSNaIEEIcfxIYEUIIIYQQJyWlNHg+Km6gmy10s4ELuEHAzMws2xpNmloRVxVuUVClKWmvT14ZUtdDo3AA5Xk47TYOFl3mONSBClNVkOcY5UAU48YxUbtF1GrjhxHKqQMjg8GALMtI05R+v0+SJBhjpjJFwjAkjmPiOCaKInzfH//8WBsFXyZLfLb6+SQJ1ghx4jMoDCfG7+qJso4jJYERIYQQQghxQtq4kZ/c3CulQGmcIMCbmSHYvoN4x06ysEPYaLJz+w62zbRpaBevLDFZRtbvka51sEmCCSNwNI6j8eKIYNscfiPCt+BpDVrhlCXJYIBTlNh2i7DZpNFuETVitOuOgwe9Xo80Ten1eiwvL9PtdinLEo1CORrXdYmiiHa7zdzcHFVV0Ww2x2U1xzJrZHS+NvY/2fh1Yy+UyTVJkEQIcXcngREhhBBCCHFSmNzEW2txA5+5U07lZ045lZ+54BePeAP/mle8imajQQNDrBxcp+4BUlYlZa9PMkgoG01aczM0W23COMb1vXEQodPpkGUZa2tr7Nu3jwMHDlAV5TgjRDmaOI6Zn5+nLOvbXdfFdd1j0m9kMqA0GRAxxkw1iR39fLIp7Gg9WzWLBQmSCCHuniQwIoQQQgghThgbN/XTP4PNsw8UG/fqd3TzPrdjO02taGqP0HNwHQftaCpj0P0+ptuj8nzmtm2jOdMiiEIcxxm/Tr/fpygK+v0+y8vL7Nu3jzIv1gMeWhGGIWmaorUmDEPCMCQIAtyJzJOjFXQYBY4mgyFlWY6bw46CI5NGwZBR6c/k16O5NiHE0WPt6L+Lx9+Jso4jJYERIYQQQghxQtgqs2F0e02v3294i1IKrRRKK7TiiDbyu/bsoeF7RJ5L4Lg4w0wOYy1hkhD0B1RaM7t9G+12myAIprI80jTFGEOWZfR6PbprHaqyxNGayloqa8Y9RdrtNkmSkKYpWZbhed5UkOVomDxvxphxc9jJ4Mjk/baanOO6Ltba8fXR/TYHq7YeTSyEECcTCYwIIYQQQojjbhQUqapqvJGvN/EWUNjJoMjwMtrQjzbyrqvxPRfHuWPBkb2nnkLo+fhuHQRwhgECCzSzjFaWU1lLo9Wk1Wrhed5UYKQoCgCqqqpH924oVdl4XGVZkuc5WZYRBMGm5ztaRoGRqqrGrz8aJbxV/xat9ZblNsBUec1WvUom7zN6PiGEOFlIYEQIIYQQQhx3o+DB5GSXLC+H2Q0alMLYepCuNYpRzojrODiuRxi4xGFIHIWEwR0LjuzatQvXdfEcdxwgGAUqqqoiLwustQRBQKPRGJe/jJ67qqrx/V3XJYxCMHY8jreyBsdxiON4nDlSVRVZllEUxTiD42ifz5GtGqtuFdgYBUUmAyPGmHEvlPXHGqypxvfTWqMdbzyGeGPzViHEsWGGlxPBibKOIyWBESGEEEIIcVyNNuBZltHpdFheXmVldZUkSagqi1Ua5bhY5WBRlGV9f+04eK6LH3jEUcRsq8GsMSgVEQbeuATk9szPz4839Bsvo0ABgOM4+L6/6XlHwQWtNY1Gow6UUDdYBTBYXNel2WzSbDbHZSplXlDmxTjAcDR7jIy+1kEecB2FNQosKAsoux7kmAiUGKUwlcZUDlW53mtEKYUyFmMqKlNiqgpjDVppPM/HCyOUCo/5lB0hhDgWJDAihBBCCCGOm9GmvCgKer0BiweWufnmm1lcXKTf71NaUNpBORHK86mMoqosxoB2NJ7nEYYB7VaDJMmwxuIMgxqBz2FNfWk0GpsmsEx+P9mLY7Ih6fi+jsYq8AKf2dlZgiAYl/eMsidGTVcbjQae52GNpSzLcXnLsVAfk8XRGhRYZQFDRQm2oqoMyhjs6IKtGygOD12NvrH1z01VYqqSssyxpkJrBy+IiBotmlpjPW/TlBshhDgZSGBECCGEEEIcN5MlNCsrK9x66y1cf/31LCws0B/0KQ0oN8LxQpQbUCmHstJYU2dwBEFAGPnMNJtkeYFS4Li6HrmrFJ5Xv86hgiO+7wOHV3azVQ+NUQAkDEPUDLRarXEAZfS6o+CINwwemKqi0nrcDHVjz4+jw6IAbQ0GizIlqsqhyLBlia2qeh1VSVmVmKrOjqmGFzPsmVKVJVVZUJU5VVlgqwqlwQ9imu05PMehihtT50WyRoQ49ixq+Ft+/J0o6zhSEhgRQgghhBDHxShbJM9zut0++xeXuPHGYWBkcT9JkmCsRrsxXhiDF2PxKY3FUAcZfN8jikKSNKcyFa5WeK6D5yi0BqWCcTDiYBv1yaahW61xo433HfUNGZXaKMtUac5oXG45bLyaZRkMs1nKspzq6XF0gwnDtSuLshVUObbIsEWGKTKqIqcscqqqpCjz4RoryrKiKAuKPCfPMvI8o8gzqrLAVCUKVfdbac8SBgHG1LcdLONGCCFOdBIYEUIIIYQQx8V6tkjO8soat956KzfefDO3LOxjefkAeV6C0mg3xwlLtFditU9lHZTVOH6dMVKWddaFRuE5Cs/V+I7CczTuMHPj9qaljJqhdjod+r0eaZbR7/fpdjukaYbruuPmqaP7G2Pqkh3Px7LebFWrOltl9FqjKTSDJKGqqnETV0drKlONJ8UcfYpxXQzDPiK2wlZFHRzJU8oipSxyijynKPJ6ck5ekOcZWZaTZQOyNKPMMsqyBGtxPRdskzBu1Bk6joPjaimhEeIuZlCYEyRT40RZx5GSwIgQQgghhDhuyrKk1+tx4MABbrltgX379rG8vEKvN8AYi9IOunTQ+OjKBQ1GOSgcPEDrYlyy0nU0nqvwXYfAdQh8F9/3h1NVzFQ5zcbshlGfk3SQkAwSBmnCysoKBw7UmStBEDI3N8/MzAye5w1LZRxQCq3AWIXGgK4DI0rp+vbhuNz6GPeTZRlxHGPMDnzfp6yOTWCkPiaFUqDQoDRKa1D1GOJRw1tbFpgqpypSyjyvs1rynDLLKNJkeEkp8mwcCNI6RFuF47p4foDrBTiOd8jMGyGEOJFJYEQIIYQQQtzlRhvzsizo9gYsLh5g//4lVlb7JGlOWQ0nvRiNUQpVWYy2KEUdjBgGQ7RSYC3GWIq8pDdIWe30aEQB7UaDRmNQl9IAvueitdo0OWW0lqIoyLKMNE3p9XosLS2ycNttDAZ9gjDCWvA8j2azieO41MNdLNV4qosFq6mURau6mWlVVSRJwvLyMvv27SNJBsRxA63rDJQ8z8fZJ0ebUgpGE3a0i3Y9cAqU48GwIStYbDXMIqkyTJlTlTmmzOvbTAW2HgqqlEUpjaM12nfxghDXj3D9AMepR/oiWSNCiJOQBEaEEEIIIcRxUfcXqbMpllfXWF3rkCQZplIovDrbQdWZGRoFSuMoF+14OK6L63l4ro/nesMGqC6mgkFa0u2lrHT6NBoxjuNijKUMfALfxXXXy15Gm/hRH5CsyEnzuoxmeWmJxf0L9Htdgigej9z1fR/Pqx9rhn08xhkf1qBUHcwBKIt8HGRZ3L9AkgxoNJpEUcTc3BzFsO/IsSmlgWE3WlRVgfbQjltfd1yUVthhtoy1FdaUWFNiqgJjy/VgjVJoXY8YVlrj+AGeF+IHUR0c8QKU64JyT/JkeiFOLsNw7AnhRFnHkZLAiBBCCCGEOC6shaIoSJKEbrdHkmT1CFnl4LoaC2jloCcmvDiui+v7w8arfj0a1/Pq667GcTTWKNK8pNNLWFrpAlAUJc04oIpD4igYloToibWsj8+tRwd3WVlZYmV5kSRJ8JOEKGqybVuXOI6xanoCy/pXhWJYUmIhKzK63S5rq8t0VlfIsgRbVXQ6HZIkqae+DEtpjs0kFw1UoByU4wwDIg7K8VDaAzUc5TvMmjHDaTVVVWFZL/GxMA6QONrF9UNcL8TzAhzPrUuLJCoihDhJSWBECCGEEEIcF9bW42HzPKcYNfcEHMfDGjsuy9COi3YcHEfjuhrf8/DDsB7VG4b4vofrODhagao38llWsNbpoqwhzxOSpEEx06QyTbSy434YapwxUZe9VFU1DNYMGPS6ZEmfLB1QFAWdzir9fp8sz1CO3nAs638vnQxu5FlGr9cbNnEdYKqMLHPJs4SyLI5dpsjkOpSmjlpo0G6d/aEcrHbQjofSDmgH0FilMNQZNJMXa+vz5Dga5Xh4XoDnh3WPEccblu0cs0MRQohjSgIjQgghhBDiuBk1JzWVAVOhtcJxFOi61MXqYY8MQNkKjcHRFs/RRIFPGHqEvofWCmPq7IsiM/TLhCobkA16JP2QtB9TZjNYk+O7Gt/3x+U0sD46eBQgybKMLEspqwJtLWWZMxj0SZIBeZajh+N4R7YKjFhrSbOMJOmTpimmyuufY7DWrE90mRjte2yoif8Ho+pmrFZ7w4sP2sM6Pqgci8ZahamG03eqClPa+navDlQ5foAbhDj+qPGqg1LSfFWIu5JMpTl6JDAihBBCCCGOi1EgAmNBGTR1wofWdXaCssOsDizGFJhSUZUptvJwVITvVMSeJfAqwJLmKUWSUuQ5xlR0laXruwyaIUXSxJYDHF0RRQFxHON53ricZlRaM7peZ0kYPEeD61CVFWVZN2fN8xztOpuCAKPgyGRgJE8zsiyjKotxIMTzHHzfJxxmvUyu41hYX2fdE2W4SpRysMoB5WKUh1EulfIwlFg0lYWqspSVxVQlaLc+T26AO8oWcb36XGjnmK1fCCGONQmMCCGEEEKIo2ZjacjGPhyTt4dhyOmnn8rpp5/KEy94/EGf8xee8FRMCVVVApbKURA6aKNxAQ+HqiqosoS026E/SCjyAjD4riZrhdiyjWMTYl8zO9Mmn50jiqLxWrTWuK47HsVbX1xczwegsnXGxSg4op16JO84zGDBbviDad1ctp46o1C4rovrOoRBRDOOiYcX3/ePaWBkdIww0cNEaywKqxyMdkB5WOoSGwNUth73W9m6Ka0xdTWO1fU58YIIPwjxfL/OnjmmqxdCiGNLAiNCCCGEEOJOmyxFmcycmCwRmSxbGd12OCUktsgpymGD0irH0ZbM11SRQoVlPV0lLyiTLoPeGr21PnlegDF4vsaWEYHKCB1DqxGT9LoUeTY1Ildrjed5w54ldVPXKIpIkxBH1wEE1/PGDWOdzLndYMaooSuA7/uYKsT3XKK4SaPVotlsjgMjd0UJilKj7BgHg8YqsDgYXCrUMEvEobKaylqKqqIsDaWpG7O6jotyvHoqjR/i+CGu66FkTK8Qx4WlDmCeCOxJHh6VwIgQQgghhLhTppqXFiXWmHqj7Ohx/wxgopFnHRhxhtNmNk6I2agsUqqioDIGU+Y4SlH4UCSW0s8ojUOR5xSDLnmvQzoYBkasxZQOni5IAksauWT92bp3yLDx6WgtSqmpgEgcx7RabYo8JcsGOAEEYYMgCKYatt7eedFaE0UhjVYbz3PwXE17ZoaZmRkajRZBEOC67vg8HAuTASo1dZ6HfVysorKa0lqMgcrU5TOlMRTDhrSVBY0eBkZCvCDADwKU66HU5rIiIYQ4mUhgRAghhBBC3GlVVdX9NJKEqihRWuP6Hm7gjwMJo3G4VVXhOM44EOG69T9JD5Y9UuR1IKOsKoxT4GhIPUsaQOIWUCjKsiRL+hRZQlVk2GGmRkFBkUGe+hR5RlUV2GFwZjKzZTJjJIoi2u02c9u2A4Y0TTBG4YcNZmZmCcMQz6snsfzqr/7qXXaO7wylFGhdBzeUAhR1exdFZSyVWf++Do5AVVZUVZ0xgtLgOPWo5CDCC2Jc1x+O6T22ZUBCCHGsSWBECCGEEELcKdZayrxg0OvTXV4iTzMcxyNsNfDjBsrR5EVBkg5IkgFlUeF5Ho1mTLvZIoqi8ZQYYFNwpEgHw2asFmydQZIlmkHPEuocEziYoiRLMsoyx1Yl1lb12iqFKetgDIDnOuM+IpOBmFEGSxDUjVlnZ+fJs4zQ98iyDIMiCGMajRZhGI7XerTP4yiINPo6cmcyMpSqy2Tqchc9zBoZZr1sGMtbWUM1nBRUDoMlxiq00jiOj+dHBGGEH4Z4nj+VESSEuGvJVJqjRwIjQgghhBDiiFlb958o85ze6jJLC/tIu338IKKZzxMUJZVy6A+6rK2t0el0yPMc3/eZm5tj27ZtzM3N0mo2NmWPjF/D1NkfWimUspgiJ08VA6fCp8AELsYYsqSkzPK6XGeYFYKGqtJYa/E8jyhubNnwdDJrpNFoUJYlSlmazSZ5nqKUg+eHhGGI67qHPV73YOU2B2tKO7pujLndccB3xOj41HCs7ug5rbUw/DoOjhhTZ47Y4esqjXI9vCDED2OCKMYPQrTngXJQWvqLCCFObhIYEUIIIYQQd4opK5JkwNryKku37iPt9ojjFlY7ZMaSVAXLq6ssHtjH0vIyWZYTBCHz8/N0Bwl5llFu387MTItoIvAw4jj191rX/3Q1xlDkKaku6ZFT5S4WS5JZiqKiqtabwBpTf3Uchyhu0p6dp9lqEwTBpr4mo8BIHMfjniNpe0BZlsOfBUcUFJkMaoz6JCpgONCmnmxj6xvtMCuGDYGQja852cD2UCYzUFAOytFo7UyNJrbWgKmwlcEUJVVZYqqyzhqxgNZ4fkAQxYSNJmHcwPfDcW8UIYQ42UlgRAghhBBCHLFR75CsP2Cwskb3wAHy7gA9C0GrRQqsJH1u27/Avn37WF5eIkkzPM9jdW2NQZKQZRn5MJihZhmP0R1tutvtNkVRDAMdDEfgluRpRUKJrRzQmqLUWLv+ODXMMPF9nyhs0JqZHQdGfD/cMjAyylgZldUURRNjzDiQMFlCc7DAxGRAZBygwU49ZuNj67KWjZN9DMqOsj3c8Rpur+nrVkaP1boeQ+y4Xt1I1XGHGSQWY0uMqajKYjiit85acV2fMGrSaM3Sas8QN1oEYSRlNEIcZ3ZY6nYiOFGm4xwpCYwIIYQQQogjMt70lyV5kpJ2OxTdPjYtoCwp85K0P2BpZYl9+/axb98+1tbWSNIUx3HoDRKSPKcoSipjUbru8+E4DmEYjjMd9u7dS6fTIU1TsiynqipsadBagdIYpXCUh+c5uIGL0u544o3WEMchM9vmmZ3fzszsHFGjgTtsnjoZqBhNkRn1IPE8bzxFB6YDLocyei5jDFVV1CUqWJRysKOSlg3ZH1MXU9WXYfaIozWOY4ZBGW8i02M62HJ761FK4bgejl8SlBFlkVPmAV7mo10frVyUmi4r8vyQMGrQnt3O7LadtOe202i28IIAx/XRUkYjhLgbkMCIEEIIIYQ4YsYaqrKgynJsmuOUBheNp12MtfSTASurq6ysrtHr9RgMBqRZhrWQ5AVlaTAVGKVwXZfIDwjDYDz1RSnFve59L/Yt7GNlZYVOp4OjABvgBw5R6NeNWz0f7Xq4jo/nBeNmrlrXGSi7du1ix849tGZm8YbZIltt6CfLVhzH2TI74/YCAevjiwvyPKcsS6wCq1WdpTF8baUUdRVNnS1irMHaClNVVMagqvq6ozWuUx9n/fzu7Y443soo28T1AkxoCKvGsJ9I3WzVmAJjS0Dh+iHWgBP4NBozzMxvZ/vOvczObydqtvH8USnSXRMUub1eLUL8OLIo7AnS9PREWceRksCIEEIIIYQ4cpWte1KUOdpC6Lo4rsb3fEprSdKM/iAly7I6QDAMGhRlPfUE1UMNSzuiIKDdbNBqNYjCYFy2cp/73pcwjghuqQMeSZLUQZQ4IIpCQj/gLf/zHYe95I2lLgfr3bFVb4/DZYyhLEvyIiXLUipjwXFxXBfH8Tb0+Bj1QymHWSZ1cMSWdfmQoxVBYMYxCK0NWjuAOw52HE7WCDBRCrReruQ4Do7r1iVHcZM0GVBVJUo5+EFIo9mmOTNHe2aORnuWMIpwXe+gwaWjbXK08sYmtKNjlwCJEOLOkMCIEEIIIYQ4IuNykbJu3Om5DnEc4eDg+D6pqUjSjCyvJ8WgFGrUJ0NVVGVFmmV0+z385WUaccTc7Cw7ts0zM9PED3y01Zxx5r3wvIDAj2g2m2RZhu/7NJtN4kaEHwSHvd71IIQ57MDIZADgcDfg1lZ1xkiaMkh6lKaqs0W8Oqul7s+hGT3daD3jyTBlji0rLAZnOEVGa41Wo14ndvj4w88cmQyOjI7NdV3cICAMI7LWDHmWjPuLaK3Hk3iCKCYMI7wgrHuUHINxxVuZfL+qymCtwVpQarJvipYAiRDiTpHAiBBCCCHEMTTdTLN2ZzIRThR1L4w6K6LIM0pToj2N34hxlIPxNHlRkGYpxTDzYZyd4LiosgIFVVWR5zn9fp/l5WWWlpZY6XTZnubEkcFqy969e8fNUGdn60asUVQHScI4JhyWmNzueq2lqiqKoqAoCsqqwhrL+tuw/n5orcbNWB3HxXWdw+4xMmKqiqLMyNIBWZGjtIvrBXWfD8epsz70MDhi64sxFcZU2LKiMiUYixn2Xamn4+Q4ti4zMqY67IyR8RFOBEe01lSOg+sFBEFAXFV1/5Zh8GH0funhezYKiNxVn931gEhV96EZTsoZ90txnDqwM5yOMyp9Opl/r4S4IwwKc4KUsJwo6zhSEhgRQgghhDhGRhvxqqo2TPlwhxvuk3uqh6kMRZYw6A9I05RSg9PwcPAoLCRZRpIk5MOMEaU0erSxRoOyw+epgyO9XpeVtVXWOh3SJKVqV3iex+z8HFprgiBgfn6eylT4YUAjbhCEIZ53+/+kHb0XaZrS7yf0ej2SJKEqK0Ch9HBNQ45bNx8Nw4AoioiiuoxnNLXmcMf1VlVFkSWkaYpV4LoBrufjON56kEFr1HCigwWsMRhbYSoDw43+VLaLrTBG12N2j8Dk2l3XrfuODL9OBvEmA3gbG9Uea6OgSFmW5HlOltWXyd8jz/PwfZcgCPC8Ojh2R4NXQggBEhgRQgghhLhTNmaD1FvbelNmjCHPcwa9Lt1elyyrx9S2Wm0azRZBEJyUwZHRprUoCga9Pt1hMKMqckLPQWmHLCnoJQn9ft1fJC/NeMqKHU8yGZ43A2WRkaQOnW6XTqdDkiTjniRRFNUb4cAnm53FAp7nEoThOMB0OOvNsoxut8/+/fvZv+8AnbU1ijwHpdBK16U+rGeKRHFIo9lkZmaG2bk27XaTKIpwXfewGrDWx1ZnPJR5RmkNpVvilgWO9tGuroNFejobZfyZMtN9NdZfc+Nn7shMTuK5vee7q4Ihk9+XZUmaZgwGCf1+n2QwnEpkCrR2CAKfKIqI45goKrF2/fNwV/U/EULcPUhgRAghhBDiCIz7VJgSU1aYqlj/C75SKOVQWctgMGBp/yL7Fm6l2+0QhBG7d+9l5549aD170m3gRpv2sizJ04TeWoeVxQOsLC2hsahmjHXr/iL9dEBv0GeQ5BRFPba2wmKNYrwHVmCwdfZJUZAkST25Jk3rUpdhNsMoIFHFDSx2XEZxe+dvlLWRZRm93oD9+/dzw3U3cuONN7K4/wBZlgGjfhV1eYvjOPi+TxyHtNttdu7cSZruGpdpRNF649JDnSOAUU5HVVWUpqKq6kCJ61boykFrF+1olHKmn8/W/+c4owk0ztTzHy13ZRbIVjYey6iMp37Pcnq9Pmtra3TWevR6vfpzUZZorYmiiEajQbudU7abGGMJAn+c2XOy/W4JcUdJKc3RI4ERIYQQQog7yBiDqQqKrO4fkSd9irSPqQo0oF0P7frgeAySjLWl/ey75SZWVlYIgxgFw79yN8bZByfDBm6yvCFLUrqra6wuHmB18QCd5VUC3yfwfSwFgyyjlyQM0oS8yKkqi7GWqhw2GJ3YD4+KSIwxlHlOmqb1Y0wdGBll1Wilsf7WZR5brRUme5gkLC0tccvNt/Ff//Uj/uuH/8X+/fvJshxrGff8cB0Xz/UIg4A4jmnNtOh2uxRFMdFzxCEMw4P2sxj1wjDD9VdQj+u1FmsrbFnfz3E8jAVtNVqvN35VqHG7E4ud2uDfXXpoTAWPjBkGytabzxZFQa83YGVlhaUDK6ysrNLtdselWVrr+v1ptYa3FZRlSaMRU1VmPPJZgiNCiMMhgREhhBBCiDtgNDEkG/Torq6wuryf/soiWb8DVY6rwfV9vLCJFzYorSbrrWHzATbtk5c5g84MyaBHnicEYXDCb3Y3Ni5N05TO2hpL+/axuLCPtcUV8v4ArwE2NxQUDNK0DooMy2EMFmPAGKiMxViDsRaHOsihHYWjNYpRlke9QYbpIMjGspJDnbfRBrvfT1heXuHWWxa4/rrrue5H13H99dezuHiALMuHGRl1wMN36rG1QVD3Fml32hTD+4RRQNyIxpvugwVlyrIgLzLyss6SGY1RscNj08MAE8rgqPW+HuPsCQUWhTOcOGNVHSCpqgqogziO3phpcWJ/hkYmx+6OgiD1xQ4bzxqKou4Fs7a2xuL+JRYW9rGyskKv16PfTyiKfJwx0ppp0U8GpHlGURRkWUar1Ry/jgRHhBCHQwIjQgghhBCHyVpbZ4qkA/qrSyzechMLN13P8v4FyqSLS4Xv23rCR6NJ0JzBDWKc0tLyDCZyqCyoIiHr98iTAVUc36WTPu6oqSyRLKPf79PpdFjcv8jizTezdMutJGsdXDSe8nGtQ5aVJElOmufjzfyop4jBUg03w3WggHFD2tAflkE4DnqUOTHhcM7RZBZCHcTJWFtbY99t+7jx+hu57kfXceONN3LbbbextrZGWVSg1LgZ6mRgJE0z8rwObriBR6vVYnZujna7RRgG40asG2V5QpalZFlCXmbro4qVQg17maDq0phqQymJmrgw3NArpTDWUJpymIVSl4m4lYvjmMMe13simBy/Ozlppqrq8qxRs9Ver8fSgRVuvfVWFhb2sbbWpd8bkCR98qJAaU0YhnR7fXq9Qd3HJsuZS2bJsoyqqsYjfU/U3y0h7qzhMKsTwomyjiMlgREhhBBCiMNQj6ctsWVOmXQZdA7QOXArB267heXF/ZiiIPY1jdhFG0OuLFpZHJPjuT7NAGykSbOKKuvR7yzR784RxU0c1z8h/6o92sDmec5gMKDT7bC8ssL+/ftZuG2B/bfcSnpgBa+A7e1ZwqiB6wYM8owky8jynNIalNZoC9VEY1GDHf6rXo17eox6RkRRiO/7OEe44R+NeM3zkn5/wOryGrfdusCNN97IzTffzL59+1hbW2MwGGCMRSmN41Ro7VC67niSUF0OY3AcTbwYs7Rthc7aGkmyjXa7OZ6OsvF9S9OENB2QZgPyPKWy1XqTU1WHPaxWWO2g9TBIogCthkGT4QVg2Bh1NI1Fo9ansrg+xpRYe3I08J0ev1tnd2RZMQ6IFHndCyZJEjqdDvv37+fmm29h//4D9HsJSZqSJglFUaAcRRCE9HoD+v0+WZaRpim9bpdBf37Yh0SNe42cTMEjIcRdTwIjQgghhBCHyRqLLQtskaLyAbpMcGyOozSuW2cZhHFAI/YJAhffdeuyiSLFFjlVkTIYJBRJTql8vLhFGDfwgvB2/7I9CijclX/9ttbWk2cGAxaXD7B/335uW1jg1ttuY+HWW1k9sIxKC+bjFvO+hxvHEAQUeU5SFCTDMb12tOkfXqyCr37xs0e8pq1umzwnWutxsKXdbnLqqXt5+CMeetDn3LX9NIypx/ZWVUk+vF0NS1kGg2DcFDYZZBR5NS7z2UqWJWR5nTFSFQVmmDUzek60i3LcOmCk1zfudeBED+MjdcbMZDDBmBJl4Z5n/uQdPm931MbzfDQ+c5Pjd5Okzj7q9/vkWTlxez1KeXV1lf37F7nltltZXV4jS4vh2N6MsixBK1w3wfM8+v0+aZqSpgM6rRa9Xo+8zHAcZzjK1zsppz8JIe46EhgRQgghhDhMylqwFRqD70Az8pmfaeIYQ1Uqoiik2YwJIg+lLXmZk3frSRr9QcLyWp+VzoDMaJq9DOuFBHELP4pxhk09D7Z5y7JsvHneOJ72UJvYI90MTmaLLK+ucOstt3LdDTdw0y03s7CwwIEDB0i6fQJcHDcgwZIOx/AOypJ+npPkBVlZoB0PpdjcS+MOrGX0davLZL+RjQ1ZDyeQNGqAO5pIo7SuMzrGz2+xWg2DV6N1qKl1jV4LIM9ziiKnKstxA1kApYcjgR2NchyUdkHr4cUBrdGjdY9XZ4ZZLaMMm2OfsD7O6rG2Lus5CpNrRj1q0jSl36+DH2urHTqdLmmaUhTlsEluXaq1srLCgQMHWDywSK87oCzqoEpRFMMymeF5VArP8+pgVJaxGq/SWVsjz3M8z6PZjGk0GuOeMELcndRTaU6MbCiZSiOEEEII8eNCgXJctOsRRDHtdpsqS/AchzIr0a6H73uAotNPWF1dZWVlhW63S28woNNL6SU5hXFodjOUH9Foz9FszxKE8SEn1KysrIx7X/i+Pw6OjJpXVsPRtkrVpSn1lJX6H8xHsiEcbWSTpJ7mct0NN/CfP/wBt956K4vL9TGVWU7DjwmSPkvpgDjtEynLapYyyDKysqSsLErVG9nRWg83MDLZj2LyOCe/TgYlJgNHk+dg9LODabVaKDUa17v+HJ7nEUURcatJs9mk0WgQBAGO46yPa96i6WlVFRhTYsZBIINWFmNVHRzRGu049SheR6OHo3r1+L23KGPrTrXGYpQCy12y7ZjOUKmzYkbn4/bO4+09bz1ppsfKyhprq12WlpaGJU11ecyot8ja2lr9u7O6wurKKlmWDRv31sGRyc+PpW5GW5Y5eZ7jBy7dYWDED1zm5+eZn58nDEMppxFCHJQERoQQQgghDpfWKMfD8WOcaIagXdCqDFr7pGmCNZbKWHrJgP2LK9y8bz+L+w+w2ukwSBLSrCAvDJXSNLMCP24xv2MX23fsoj07QxiG44DHxs32dT/8AY1Wk9nZedozMwRBMNxs5qRJ3ezTVBbX9QjjiCiKxpv4rZ7vcBhjxoGRm2+5hZtuuonb9u+j2+2TZhlYMFYRDHosrC6hg5BmnNIZ9OgXBRUGoxhPnjFWUdn16SoHMzlqd9x/YrhxHl2KopjaJI8CQr7v43leXdYU1r1KRk1SDxZ02r177/jnk4ER13WJooiZmRm2b9/OzMwMURTVvU+MqifsVNUWmT51cAMYPp9T9wdRehgYcXAcD+06OHoYHNHDoJgFa0qUMqDrp6kDRHX5yLHMelhvilqf66IqwVo81xuf1yOdfjPKPlpd7bB/34FxRsjqat3rJc3qRreDfh0YWVtbo9vtMhgMKIpiKkhWv+cWM5z57Hj1+1tVFSjDmuOQ5zlhGLJ3715OPXUv7Xb7pJncI4S460lgRAghhBDiMCntolyLEzbxDBjrYFWdQeL3+xR5Uveg6FX0s4zVzoADqz1W1/r0kgF5UVKUBpRDbhTN5RWWl5fp9XpkaTqV9bHRd7/9HWbn59i7dy+7du8ibtbNPwfdHisrK3Q6HaqqIopj5rftYPvOHczM1MGWI90MGmPIy4Jev8/K6goHVpZZ63RI05yqMiilycqCTjIgWF0Bx6PVbFJWFYYK3/fRxqDUdHDm4N05pkcD53k+LL2oe1EMBgPSNCVJMrIsq8tVqhKoAxCu647H7DabTdrtNq1WiziODxkkOuOMM8a3bwyOBEFAs9kcB0bqzAOnDtrkJdj6n9OT5U1aOyjl4qgKq3WdGaKgMqAcB0bZLK6L47jrQZFhZohlOOIXBdaiMONmrIrpMqqjrarqBqjdfo80HWCtJQximo0GAL7vH3EG0mAwYHFxkVtuuYXl1VVWV1frbKpej8FgQJ4WJEldStPr9UiSwXDCjBk/xyiLZbIky6vqaUYqh6LIMcN+JVEUsbBwOt1uj+3bt+N53lE6S0KcGIytA84nghNlHUdKAiNCCCGEEIdhtBnUjgc++ErXZTWeh+cHhNEK6aALbo8wy/FGmQpKU1SGvDAkWVUHFDR4eUWWZqRpRpbVpQT1hJGty0y+/93vMLdtnkF3jWTQZ2amTVEUrCwtsbCwwNLKCqYyzG3bzhlnDOrxt2GI53l3roTAgrETU1rKajx+dtRMtawqBllKL+nieC6e79JoRgShT1kZyspSlKYODlTF7W5QR6OBR004V1ZWhpvoOkCSJPWEknxiHLDjKjynDow0Gg1mZ2eZn59n586dzM/PA0xl5Ew6/fTTAbYsFXFdlziOmZlp0WjUARat1XCySlmfIOWNM1ZGo4fLyqGaCLSAC8oO+4kMb1ejMp/hGF8YzS+ue5FYxTiMZEErhbvF+o+GyWyRfr/P8vIBOp0O1loajQZVuW2qpOaOBEdGga5+v8/CwgI33XQTnV6PJElIk5ROp8fa2jJpkpJlGYNBHQQbNVqdDIJMfh1N51kvQYOiKEj6ffI8p9FosLCwQLfbO2TQUQghjmlg5J/+6Z94y1vewje/+U1uu+02PvGJT/DkJz95/PNLLrmED3zgA1OPOeuss/ja1752LJclhBBCCHHElFLDZqIKpV0c18PzPErfw/FcjIWoNxiWcXjrWQo4eNrgKw/fdWkGMU0vwFe63gxXBnuI/hs33Xg9ndVlTJ5Spn3a7TZ5nrO4f5Fbbl1gaXkF7Wh2nXIKcRyzfedO8mwbtmGPeEOotcJzXeIwpNFoEMcxvaiHQWFR6GHpShgF9fQP3yeKA9rNFr7vYS3kRUGaFaRZPfLXdx18d+vN/eTmvNfrceDAARYWFti3bx/Ly3V2QX8wmMgYyab6YLiexnc94jhmdnaWnTt3jsswXNcdj24dvY8ju7ZvG2ZuOONeKGVZjpt8RlFAM4qJwgDPq59jNEK3bgJa4TjrQRXfD6nKnMpxsFUJ1BNmUMMCG2tRtgLrgFX1fZRGUWfpUFVgSjB1s9+6AapGKT0uCzoSB/tsjc5FVVXkacba2gr79u1jcXkRW1a027OYssLzvHF/mzv6eaoqQ7fb47Zb93HzzTeT5jlauZRlSZIkrK11SAYDTFWRZemwIWsx1Udm4/rXxyWD53njgFue55RlycrKCksH6myq0edAiLsTO/xv8YngRFnHkTqmgZF+v89DHvIQnvvc5/KUpzxly/s84QlP4Iorrhhf933/WC5JCCGEEOJOG41c1V49WtVRFo3BVAX+oI/rOuh6hAgAnuPQ8H18P6LhBzSjkGazydzcHPNhRKAU2AprDNitC006S4uQJzQcQ2BS0maTPM/Zt3iAfQuLLHd6uEFM3GiSJn3KPMdU05vKO7qZ1dohDENmZubYtWMnS8vLGGPo9ROMqfuZBFEdNJltz7Bt2za2z29j+9w8cRwD9Sa1lyR0el36vYRev0eSpAd9zVEpx9raGgsLC9xwww0sLOyvm9j2emRZMe4xUm92DWDr8hdHEXgeYRjW5UlZhlIK3/dpNBqEYYg7nP4zqdGMCPxoGDhxKYqCLEvJkgRrLZ6r8QOXwPfx/PXRr6PgiHYVxrjjcxx4PpUfYsocUxbDbIUKjMEMA2HWVNiqxDh1oEWhsJNNV20F1qIBR2u06+A7Hp4XHNZ7N9WgdItsC5juuTIKBvWTLsvLB1i47WZu27dAnhXMzsxgK0McT094OdzP0+g8dbs99u/fz8K+fSitaTba40BUHQxJ6nNTVuPpOxtfY1hgVH8dZuOMmuSOgl9QZx2laUq322Ew6NcjfoUQ4iCOaWDk/PPP5/zzzz/kfYIgYPfu3cdyGUIIIYQQR129YdPgumgb4HgBrh/Uf1X3XELfIfJd4sCnjJu4kWIuiNneaDETNYjikKDdohWHhMqiKoOtKqzdetJJkSaUrqXoOmQBuFmXLC/IVlfJuqvkgxRjIUuS8V/MjS2mNsF39Pi01kRRxLZt2zj11FMZpAle4NPrDeqmo65PFEW0my3m5ubYvm0bO7ZtZ262TaPRQFtIi5zecDxrZ9grJMuyLV9zchLO6uoqCwsL3HLLLezbt8hapzMsrTDjhqzrU2GGTTgdRe5osiwjz3OAYWCnDtq0223CMNz0ulorgtAjDGK01uS5A6bC5GkdAFIaR2lc1xkHVkYb+vUyj/VARBCEdTCkrDMeTFlhqJuqGlPVo36VArU+NWdy5oyyBs1wPLDj1pkijltPJfIPHRjZOBL5YCOOJ9/j0bHkaUZ/rcvy4iK33XILN956C1mW0WnN4jku83NzzM3NEcfxliVJh1JVJb1ej+XlZVZXVuuAWtyq+9AMM3CqqsJT9dQex3HGU32srUMh6+uun9MYU08PCgKiqIHv+fT7fTzPG2f7FEVJluXjPiVCCLGV495j5KqrrmLnzp3Mzs5yzjnn8PrXv56dO3ce72UJIYQQQhwWpepNrtIaRznjBqDNRsx8u02yPafhNwi1y/aoxfZGg7Yf4HkaG3h4rkJXJVWRUZXF1GZ7MpihbYlnSlyTo8sMXSqcqsIxBY4poaoo82LYo2FAMhEg8X3/DmeMjHpmBEHA/OwcZ5x2GtZa2u02vd6AoipxXJc4ajDbajMzO8vc3Axz7VnarQZhGKCUqhuoJjn9ZMAgyUiGJRJbGW2Osyyj2+2yvLzMgaUVlpaXSZI6S6WqhmN7jRlnFdRZBJaytPVY1yqrMz08b9zIMx02tx2V3kyqyhxHKfzARSsXZQ2Fo8gUdUaKNTiAVqOmrOuBDGvtcHLM+vN5fkhoDJiyfrwxmKTEFGArMw7qTJ7r0df6Ujdn9dwA7arhmOaIMIwJwuiQ79vG8caTI5Inr8P6GF5HOxhrSJI+a6urLC0ucmD/PvYv3Eaa5GTNHo0gYO+u3WSnnnrQMcWHWtMo4FX3Dklx/br3jeOvj592tIM36hfiOjjF5v4i6+dIjUtpoigiDOoJRFEUEUUx1tYZLuvZ6BIYEXc/BoU5QUpYTpR1HKnjGhg5//zzedrTnsYZZ5zBddddxx/+4R/ymMc8hm9+85sEwdbR8CzLpv7K0Ol07qrlCiGEEEJsbTRKdJjtoZUmCHxa7SbbSkurZYndkNm4yYwfEWqLMhWVrkAbTJVRFSllOd1TYVLoKCJPE3ouke8RenUmQei7+J6DoyxZUTcsXVtdZW1tjUG/T5HlVEEw/qv8HQ2OuK5Lq9Vit92N53lsm5+nPxhQVhbXrUttmo0mrUaDZrNBHNeb1NEGt6oq8mZBO29R5AVFUW45rnd0zKMeI0lSl930hk066wwQNQyOVFTWDnuUKrRVoDR22CR21Bs2TeteFaMmrVsFRQDKMsdYM8ygYFirYereH1UJphqeD8bZKZPnCLWexQB1w1aCCDt6TuqyG23r47OUmGoYHLF2XBZSv76L64LrevieRxCERGGTKG4ShBFBMB0Y2fhZGZWtjEqNRiOPR8c/+h7qSTqe59XrNZbu2hrLy0ssLS2ysrzM2uoq2SCDNGel1aKzcoBs0KeqCpS6Y+XvVTXdt8V1XTyvDor4vk8QhpiqInDqjByvLCk2Nl4dvjWTGTvGGMIwJAgCfD8gjhvMzM6AhdnZGVqtFkHgrze3FUKILRzXwMiv/MqvjL9/8IMfzMMf/nDOOOMMPvvZz3LRRRdt+Zg3vvGNvOY1r7mrliiEEEIIcUjj8oSqoizr3hd5kVNWBu35NGdmUE5IHDdoN9rEno9rLSYbQDHAOgpUhakKbJljywrrbw6MtEKXdhzSajYJG02CuIEqCoK0IvD7uHpAmhf0ez1WlpZZWlqis9Zhfj4lasRTjUcP1yhrJAxD5oa9OmZmZsiyOiNj1N8hCOrmq+GwlMhxnPFrWVv3kIvCcLw5rw7SR2XynNZTX8x4c2+MqXuRGrueVaMVDgql6wac1urxZB9rIRj397idcqKqnAiqmHEPk7IoMFWBW/lYU46rZSZLZ9YzGNaf33HqJqAhDRyl0dpBO854NK/j9HEdl6LIJ0qn6nNWN/MNCcKQMIqJ4xmiuFEHRcIYZ6L56mTQYHR9NNGn3++PG5iOModGAZNRv406u8kjDAKstXRWVllaWmR1eZVup0ue5BRZToYi7ayRdDrkaYKtzHh4zlbv3cbP0PAndf+VIKDRaBLH8bBBcZ3l0Wq18ByH0PNxHYe8NBRVsSmYNfpMaq3HQbRoOI7Z90NazVbdjNVxmZ+fY/v27cRx4w6X/gghfrwc91KaSXv27OGMM87gBz/4wUHv84pXvIKXvvSl4+udTofTTjvtrlieEEIIIcQWhtkiVUVVFGRpSpKk5IVF+yGtOCJszNCenaPVnCEMAlRZkA86pN0VyiLB8TTYcripr7ZswNqOfNqtmEarRdCcRYcRpDk6yPCCiMDt08sS0kGflZUVDhw4wPLKGtt39Wm2W3e4YebI5DhUx3GIo6gONhhblw9pPe67MRkQmXwdx6lLNjZu5De+zlZf6xIQi7EKawylqShHo1eNAtdF2Xrqi7HDUpphhoepgOG6R2vb6vgdz0MpPR5JPBj06fd7JNkALPi2xGCxjJ5/InNkok/H5LGMgiN1n5A6KOK4Pp4X4AUhRZ5RFuuBG63rMh3X9XB9nzBoEEQRUdSo398gRDsujuNOnBczzv4YndvRuN2VlRW63e4w26akKLJxFkme5xhj8H2PKAyIoggHxerqKvv3L7K6tkaaphhj6mANCsoSmyWYIh/2wtmcqbKV9dvrwNrs7Ax5nhM3mzQaEV4YUTWbzM/PUzQaBMOymrIyW2b5aD0dGMmyHOXoYdZIiONCq1mXcm3fvp29e/cyMzMzbMp6cqf6C7GRTKU5ek6owMjS0hI33XQTe/bsOeh9Rn+REEIIIYQ4kVjqLId8uPnEcYmDmHh2B7PbdtCemyeOm7iuiylz0u4ag8gl7a4A9WQbW5X1tJIteozEvkMURXhRjPUjEnwSU1HggNJoDcpasixlbW2NAwcOsLS0RLfbZW5ujjAMx8GJO1pOA4yDC1VVYbHjugbFemBgY1Bj0h3JVpnsI1FfNJaSyhjKYWnI1HG4DEuF6mwUa0CpYR8NR0+N6t1qbVHcxHXqhp1pmtLr9egNemRpiuu6VAasUvWcoWEpz2hz7jh6y+Ovv/XqEcBa47o+jufj+yFh1KIoMqqipDLrx6K1i+O6eF5dXuIGIYEf4nouWnvj4Mno81GW5bhMaBSgS9OM1dVVDizuY2VllV5/QJrm5Hk6nLZTN6a11uI4ijiKaEZ1M9Vut8vi4iKdTpeiqHC1gxt6NMOQyPdxATVsEMzEZ3Rjo9fR9xuDY81mk507d+H7AV4Q0IhbOK6Lq0BrMFWJPyztMcOsoc2BET31WUzTlLwsCfyIIAjQThPPcWg2m2zbto29e3czMzMzDAwe9kdQCPFj5pgGRnq9Hj/84Q/H16+77jr+9V//lfn5eebn57n88st5ylOewp49e7j++ut55Stfyfbt2/nlX/7lY7ksIYQQQoijaLgh1g44bl0S4Xi4vkfYnmdu1x7md+ym0W7he2H9l+4yx3cdNCXaVpR5grIWOyx38O3mPiPK9XC8AByfzGqSrKSf5PSzksLUkzuqqiQrcrpraywtLXPgwAHW1tZIkoRGozHeUN7hIxxu+q21uK47tbZDBUM2PsehbJyaYiamkFTYeqNs6yyCYqJ5ad38VOE4wxIcYzDWoG39z1xHe7ebMRI32nhBMA40pGlKmqWUVYl2XZTroHWdNWMnSmi04+B6Dq6rt8iUGZXXOKiJchrPCwiqqu4nU5XYUSNTFMrRw5KbOjgw+n4UEJlUVRV5no+n/NTlRYY0TVlaOsDCwgKLi4v0eglJmpBl2ThbZFSe5DiKMAiIgzpLI88yVlZW6PW6lEWJ63q0Gg3mGg1m2m1iP8CxFms2Z/6MslfG7yH1erUCz6sDUzMzM5xyyim02+362Ny6T06jEdJq1r1TRqVYo2PcWAY1OheO42CMYTAYMEhSLBD4AWEU0Igi2u02c/OzbN++jXa7Oe5LIoQQWzmmgZFvfOMbPPrRjx5fH5XAXHzxxbzzne/k//2//8cHP/hBVldX2bNnD49+9KP5m7/5G1qt1rFclhBCCCHEUaOUQmkHxwvw/Ag/buIPEoxrhn0iIsIowg+icTmLccCaiCBtUCYhtkyxVU6ZJ1RZQpXHOI6HndjIO36A4wcoP6BCk5VFPa43z8nykqwsKMqSLCswvT4rK8ssLi6ytLTE7t27abfbeJ530ODA4R7r5NeRUVPPdJAw6PXorq3SW+vQ73XI0oQ8GZClA9J0QJqmJEnCf3/5/2/qOUZ9RfI8Hzfbr6e3rPfzmCwfGY9jHWaUjJ/HDLM69HTvjcmpLBuDQ0EQ4jg+Vbne5NPigOMP39cQ1/emSoZc18X1nalMlI3nZWqqkK4DJI7jYUyJZ/xhUGcYZJjIkBndd/J5J59rdCxZlg2n7nQoirofR5qmHFhcZN+tt9SBkf6Afpqv900Z9hmpqgqtwXUcQs/FcTRVWdWBhl4fUxSEUUyr2WJ+fo5t821azSauUzdqnYyLTL53ZVli0Vg7OpYNgZFT9zDoz9WPmch8scMsnNE5rt+HzYGR0YjhcWCkX2f4ZEWB53k0mw1mWm3aM03a7TYzM23pMSLutoxVGHtiBPxOlHUcqWMaGDn33HMPWm8I8A//8A/H8uWFEEIIIY6pOotCoxwX7YW4cYuw0aaRJtAfoCmxZUZVFWDMeFOHddCOj+MHaNfDYCmzFPpraD/CCSKU66L0+uY4as0Qz84Rt9pUOsColCIZkKi610JelKRZHSQpGLC2ssrygQMcWDzA6urqVDnN0fzL+WTpxChoUZYlVZnXk1eKHFMVlGW9aS7KnKLMp55jHFhJU7rdLitrq6ytdun3+xTFKDgykREyMdLYDJuxGqeucDfjjIV6tG+e56R5NhwTm9X9NDZkjyjHwypdN3PVHn4QEJQF1ljiZoOo0SCIYrxwfYqK6zvD7Ib1UqKDlSltDCiNymEmz9/0/ZgK9mz1nKPAyNraCktLS/U44rJkMOhz4MAi+xduY2V5hcEgrQNo5WhCzah3R4kCtDV0tRoO4jH11KCsQGHRkaIZx8xv386O7fO0ZuZxPb/utrLhfR9nr+RF3Q9G6WGvFU2rWTf/bbeb7Nq1q85wqSzZMJhjK1MnXDnTgaaNJWUbS6zq0qGUzlqPJKvLntrNFjOzLVqtFnEcEkV1iY1kjAghDuWE6jEihBBCCHGyUUphlcJxPfywSdBoEw56FHlKUaYUgx550qcM42GTz3q0rHYdtOujnIDKOmR5j6wyGB3gBDGeH1K5/ngTuG33Hrbt2k1jZg6jfaLeANcaTNql4yjMcKpKURZgFd1ul6WlJRb3L7K8vMrOnX2azea4kerR2iRObmDrTXI1Do6MLqNshcnLyMbsh32L+7nl5ttYWFhgdXWNQZ5uOcJ4qq/F+LoC7Hq/l7yg1+uxvLTKgeUl2u02vu+Pp+mMNuBPuOAxm47rf/3Fe9FaEzUaNNttmq1GnfkT+vjhKHtEjUs77mgp0dEoPRoFI1ZXVuh0u6RJn16vx+ryEkuLB4aZJBVVZer+LKhxCZJWoKypPzdpOc44oaoDFa7WYA2+6zM7N8/2XXtoz8/jBdH6eR4GLaqqIkkSVldX6Q9SSjSoYW8Vx+HU3TtwXZc4jgEoy4qyrIb9UQxKMWw86w4DGFtPvNkYILHWkmU5cRyTDDK01rTaDdrtFlEU1r1MPBfPcyUoIoQ4JAmMCCGEEELcWcpBuaC9AC9qEkQN0kGHcpCSJx2SXhM/jHE9f/yXazUczaqDGLyQkj5VXqGzjHA4ZtWd6KOx+7TT2L5zD412C4VHv9kn0JYi7bC6FI03/Ni6N0MyGLC6usri4n4OLC3R6fSYnZ0d//X8jjZhPZiNjTdHgZGNwZFRCcdkYGQUFCmKOoCxsH8fN1x/I9dffz033nIzS6urpMl0YGRqss2mAI8dTqaps0oGWcrq6iq33XYbzWZzPB72UOUvI9t27MBxHKI4oj0zQ3tmhkYjJgwDfN9dfx+PcNLPnTXZgDVNU7qdNdbWVlhbWWFldZnu6hplloEFpV20owm0xnourlJorVAYyqIgZUBVjPqegLZgSoPNS5SGRqPBzLYdtGbncMMQpfVocvE426fXG7C4tERnrUeuNMrx0drBdTRK3Qet68kxWjvjz0ie171i6swSNZXJUwc+NgZINErZ8e3WWsKwwPcDskaGUoo4jmk06t+HUdnT0QwECnEikak0R48ERoQQQggh7iSlFMaCGo5kdTwfpR1MmZH3LTpoEMRt/CjGHW7YlHZxgpCg0SJqb6dEUxQFThCj3Prxk5u5XXtPY27bdsK4gVIOcSNGK0vS77KyvMTi0gHCMMDJcqrKkmc53W6X5eVllpeW6HQ6pGlKo9EYZ40cLdNlFcOSGmPqDbCp6hG74wDJdGBkNFlkeXWFW265hR/+6Dr+6/rr2LdvkV6vNx4ZO5mVMj7vE+cfGGeOGGuxxpAkCUura7jebfi+TxB6RFFUT/cZNvk8WIBox66deJ5HGAY0m01arQZhtB4UuSNTdo6F0dja9eMwFFlGv9+js7JK2uvWmR+uTxC4eL6HG9S9UnzXwXcUCktZ5HS7GoeKrqkorEFV9YQjU5ZoC0EY0ZydJZ6ZwQsjtOPUzWKVoiwrkjRnZXWNfQv7ObC6RoUDw0bDnrPelHYUkBq9755Xjt/f9YsCNLDV+1JnqkwKggDP88jzAKUUQeBPlM7UzydBESHE7ZHAiBBCCCHEnWUBNCiNVRqDpjKQ5yWZKVG9LlF7QCMvqML1Ua+eHxI226AcgkaLsixxXJc4buJN9EVQSjG/cw+NZhvfrzeArudTlRX9nR1WVlZYXF5muTOgl+ZUaY4xBemgT7/bZW1trW5QmeXjSR9Ha2O/cdNpLeOMjaoylBPNUsuypMjrEoqROtugx+LiIjfdeAs33nQjN9+6wNrqKnbYKHV0HjZOrhn14Zgu5TGUZT1S1hhDr9etG4AGPo1Gg9nZWbbNzdNsNtezbLawe/cuXNfFD3yCwBtmm7ib+pMcL0ppgiCg0WjQbrfp97v0ux2UBVsVlEWJBlwsoe8SxxFxq0kQBgSeS+A7eAqKMqcTe/ja4qDoDXLKrAJTok2dPeK5HmHUwB+WYjmuh9L18adZzmCQsrKyym37F1k8cIBS+2hv2MtlODZ51F9nFBgZNVAdBabG/XeY/kyNg14WRjOiRz+e7EEyCvaNRjMfaTaPEOLHkwRGhBBCCCHuLAX139+hsorcKLIKBqWlKA26KCnLOjgwfoiq/+IfRg1cLyBqNtfLClx/ot9CvbFrzW3D94fNU4ejXRvtGea272Db6io7V1dY6QzoJRm56VFW62UWg36fwWBAnhdTjUuPZinN5u8VxtpxP4miKMfjYrM0nwhilPT7fZaWlrh13wL79u9jZXmZNE1xtMb31/usTK6dLRqAjpq/GlPV15WCLKPfd1hbW2NlZYW1tbVhU9di3G9jK/e+zz3GwZhDHffx2HiPgghBEDAzM1s3Xq0qyjyjs7pCbzWiShOo6qBIMwyYaTVpz7aJmg0agUsUuLiOpsxz1iIPjwplK0CRUmJs3fPDcx0838MLA4IoxHE81ETGTGeQsNrpcmC1w+KBJfYtLmKUj44adfBjYrzxZINarfX4/ZwMYBzsfNY3b+7TYq2dyv45UQJXQtwV7Ak0lcbewXXccsstvPzlL+dzn/scSZJw3/vel/e+97389E//9DFa4aFJYEQIIYQQ4ihS2kH7ETps4TVyVGUI4jZeEKInmkBO/kVba431vPXn2OKv3VHcRjsOGgcwoB2ChqU1u52du/usdXusdgb0Bhl5aRlkBZ52wFiqshiXsmzVyPRomwx6VNUoIJKR5/l4HO/ofkVR0B/0WV1dZWVlhdW1NZIkoSxL8LzhWNl6o1tNBEZGvVZHARFgGBSZDJ6MB/1CtZ65Mrrf7R3DxkyUjdkNW2U23FUcx8H3fdrtGaqqRCuFLQuSfp+838epKkxREgchM40G8zNN5ubbNJsxcRTUGSMayiIn9CyqyijygrJUOBRUqqQZxTQaDYIgwPU9HK/OYmKiPGVtrcfSWo/l1TWWVrqsrKxRaQ8vL3EcjWL6PK9ngGzOWrozY6RvL7AihDhxrKyscPbZZ/PoRz+az33uc+zcuZP/+q//YnZ29ritSQIjQgghhBBHiXbqPg6N1hyVAS+ewRpLGDdozcwRhPGmsaGHm+7v+nXTS7BgHRzj4AeWuJUzu20nu3p9Ot1kGBip0J0+juvjeQ6e46Cxw8aVRzcostWmdNRDoixL8iKnKPKpoEiapVP3y7J6nO5gMCDPsnG5zygYUWfSsDmosyFoUV+GU1eGpRpRGNCMYprNJs1mc6q/yKGUZTnOUplsIDtZrjFZ5nNXZimMzrnv+8RxPM4+wlqKPKNMB3hYin5CFPjMtdrMtVvMz7RptRtEoUfoaRwFZZ7ha4MtMpIkJc0NmgIbWmZn5pidnaPRaOJ7PtqbblqrlGLf4gpLKyssrXZZ6/Xp9hOsygmtxvWGQbyDHANwpzOXthqTLMER8ePiZG2++uY3v5nTTjuNK664YnzbPe5xj2OwqsMngREhhBBCiKNgtDn2/BBaGjcIaA2bjLqui+eHUyNi7yjtuoCugwHKorRC4RPGLVqzBduThF5vQH8woCgKPDegQtFuxESBf6de+3CoDTNWx8GRYQnNaGxvmmf003x8H2MM1oy+mmGKx7D8gumMEOzGiTT1fdeDJXUphTcMXNSNUxtsm5tn75497D1lNzu2bafVao0bgY7WsfG8pGmK4zjr5UiDAVlWlwAFgU8YhgRBgO/X53b0fHdVU9ZRMCQIgvWRwUCVp1RZSqAUaaeHrzUzrZiZuRlmZ2doNmOiwMF3AFNRufV0mjzLmOsn9FNw3RJtPbZt28H2nTtptmdw/QA90Qdk5NZ9+1hd67HS6dJPcpIsQ6HRrotSIVoZLrrkuVgL2hmWiGmHOA6475n34KyHPZT73vPM8SjpyRIypRT7ltcwxtIIQ6LQG065uWOjj4UQd51OpzN1PQgCgiCYuu3Tn/405513Hk972tO4+uqrOeWUU3jhC1/IC17wgrtyqVMkMCKEEEIIcZSMNqtKBbiuRzjRh2K8eT3ScgE98c82a7FW4zhg/ZCo0WZmLmHHzp30+32yNMfxfEoDc3NztNtt4jg+JsGRyV4b2qk3xNp1wNGAxlooK0teVmRlSZIVpHmx/phhZofv+0SjQIPvURZVHRmZygiwU8EMrTWO1nWJkda4joPrugRBQBRFtFsN5mfn2LljJ7v37uS0U05l9+7dtFqtcUDhYLrdLr7vU1UVvV6P1dUOvV6PqqrqtUYRcRwSRRGNRoM4jgmC4C5r+Dl6jclGo9gKU+SYvMDXisHqKtoYGnFAe7ZJs92i0YgIPI1WBmUqirzuzdFstZgdFAxKj6BhCbyY7Tt2sWPPHprtFu7wszP52gC37VukN+iz2u2TFTllUWEp8cqCwPr1Z8JoLAarKhytiQOfmUZEK4xxtZrKxtn4e7K43Kl71bQrPLeFI1NmhDihnXbaaVPXL7vsMi6//PKp2370ox/xzne+k5e+9KW88pWv5Otf/zq//du/TRAEPOc5z7kLV7tOAiNCCCGEEEfRKDhye6UasN7HYlR2UuYFZZVDtd7PYmNzyuGMD7SyGFwcx+D5PnGjRXtuju29HlmS4/oBeWlpzc0xOz9H3Gjgj0YF38mN5cZynNH6RuNjfb8emeqHPu7Ar7NdtEtlFaWx5NV6MMXzPOI4pt1uMzc3z+zsEnlRkKY5TDTqHPf2GJ5WBTiuizt8Tc/zxgGRZiNmtt1mbm6O7du3s33HPDu372D79u3Mz8+PsxMmG4JutLa2QqPRwBgYDAZ0u11WVlZI0xQ9bAobBPUo3/n5WbZv3z41dvauDI6sT+1pMDM3T1UWuFoxaLWhKAgDl6gREjcjwijE1aCosFWBdlyUdqmsZrbUFE6DVukQRTNs37mTHbt202g1cX1v6jVHFvcvMihS+v3+sKdMhTUlVV5gIlMHqzwfsChHE3gBzajBbLNBsxngus64XGlyUg3Un7Ob9y9hrMVYiKMQz3M5voOShThxGBTmBCmlGa3jpptuot1uj2/fmC0CdSbgwx/+cN7whjcA8NCHPpTvfe97vPOd75TAiBBCCCHEj6NROUlVVVSmHJeNKDYHICyMy1Usw4CJdnBcD98PabVazM5vIy9K3DAiLQxRa4Zme5YobuB5wZ3asI/WMypvMcYQBME4e2OcrdFo0UjTuoSmqihNRVEZBkWJlxX4WVkf47BfR6PRYPv27ezevZtevwdoer0eRVmOgyOTF2eYGeJ5Lr4fjAMirTim1WrSbreZnZ1lbn6G+bl55mZnabfbtFotoig6rADR2tpa3QAWGAxS0rTugdLpdEjTFDMcJdxut9m7d/c4WDIKDt3V/UYAfD8gbjQw5XZcpRk0m1DkuI7CD3z80KtHDiuLNRVVVaC9AuXlGDdgRoWosKCwHnFjhvltO5jZNk8UNw8a7FnrrFBUhqLMhp8PgykLyiKDytQZIqGP64HjuASBT7vRZLbdptVo4Dru+ud/Q1Ncay03LBzAWIvjeszPtGhFIdx+zFEIcZy02+2pwMhW9uzZwwMf+MCp2x7wgAfw8Y9//Fgu7ZAkMCKEEEIIcRxMTjupqoqiyCiHI2QVCmeLyTRqeLEaMKoeE6wUSms83yOKm8zNl4AmiNsM8hLHj2i02vi+P5V5caTrHZU9ZFlGmg7Ys+eUqYCFH9ZZFJhhBoDn4gYBbhjiBDFuFBM0mkDdINXzPFqtFrt27KR3Rh9jDFEYs7y2Ng5A1O1L9Hq5jOcRej5hGBCGAXFcT09pNpu0Wi1a7Qaz7RlarRbNZpNGo0EYhuPAxeGchwMHDtDr9VBKURR1Oc1gMBhPzxkMBoBhbm4OaysajQatVoswDHFd96iNQz5cozKXIIigbXEdTRzFmLJAKYPraBzPwXWd+u+6VUlVFdiqoKoqnKBA+S28vMKqgLDRYqY9S2tmBm/is7NRWeRYLJ6GwHPIHI2xFVVRUJkCrSPiRkgzCodZNj6tZoPZVptWq4Hnrud/TE4BgjoAd+PCEhZoNpqcsSfHmKM3aloIcXycffbZ/Md//MfUbf/5n//JGWeccZxWJIERIYQQQogpk2n8CsYZGpMbsY0buNHGfeP9DmVy7OuonGRk/623wEQDysmpL8N4CEor6pm1daYGfkjUsCjl4PgBfnPAIMkx2qHRaOAOJ7Eczvomz8FGruviuS5hGI5HK442qqNeITSHU2F8jyCOiNstWrOztOe7bOv16Xa7Uz0y4jhm27Ztdf8OL2B2dpbV1TUGg4SiKlF2NGVGD7NE6uanYRgSx/Gw30dMoxESD0fMRlE0bpA6mkJzRwJDN954Y30sAGjKsqTb7bG8vMTi4n7W1tbQWlPmBe1mi87uDoPBgJmZGYwxd1kT1klKKRzXJQgjtNIEQYApy3puhQalNVortDVUpkRXFdYYHFPhVBVOXBKVFut4BEFM3GzWk5QmxkxvFAT1Z8rV4GIwWYjJE7SqwJR4rkMrjtg2P1u/J0FIcxTIiqfLuza+N1VVcdviEtrx2LVjQJaXx3zUtBAnk5N1Ks3v/M7v8KhHPYo3vOENPP3pT+frX/867373u3n3u999DFd4aBIYEUIIIYQYGmdxVBXGDMs9tELr9aalo54gVZlTVRVQ93hwXP+w+oocDuU6aMfF3biR1wo9/senRRmLtQoz7GnieB5eEOHFDcJGxiBNKY0lDGPiOJ6axLLVRncy4DMqlbGVmSpvYLyBrb8Pw3B8XkblNJPZI1HcpNXOmJvbRpIkpGlKmqbjpxuVoMzMzOA4DmEYMjc3R6/XI0mScTnLqP/Eeg8TnzD0CYNwPPXA9/3xlJiNk2IOFRDZ6vbrrrtu+Lj1HiZZktJdW6Ozuka/18X1HLJWgyxPyJK6dGj0mTge6kCbAtdDKY3xXDCjEcZ2/X3Doo2tP+MT77dbGYytG/26foDvBTgTI3q3Mj/TRDsuVVWRJiG6yqHIKYzF9xyi0Gd2ps2u7TtoNBp12VMYEvjBuGHtKHA1GSCBOjCyvNbFDyKStC7LkrCIECe/n/mZn+ETn/gEr3jFK3jta1/LmWeeydve9jae9axnHbc1SWBECCGEEGKoDooUFFlGkadYLI52cYO6PEJpF2tKyjwnTfrkWYZSCj+ICBugVHBUGm/6vj/e1I+zRVATdTSAqoMV1th6c2sMvjWYyhDGFVEzp5nXzTBd1x/21nCnAjxbZcGUZUlVlBR5Rp6l5Fk+LvGxth4TrNEop87g2HvGPerlTARHRpkwo54jo4DBaGxvUaxPpYE6aySKovH42ZmZGZIkoSiqcaCh7keiJ/qLeOPxrpNjXkeb61GQ6khLh2644QZc18X3QxqNBo0oRtlhw1IMrlYEroPvOLjaQen1LfvhZjUci5KQuuTIoJSLdnQdGMFM38mCsgZrvansIM9aDHVpltJ1Y1t1O5/nvbt24Qz7hGRZhq8smpJBktJoxsy1W2yfn2fHjh20Gs1x4Gr9azgOPE1m9UAdGOklGU3HX29CfFTPlhDieLnwwgu58MILj/cyxiQwIoQQQgjB+ma2qir+/+z9eZxka17Xib+f7Wyx5Vbb3fr2BiJL091KD4ssCmg7oo7IT0ERFRSGbRQdkAEUtAUbfioqOMo4Doi2yyioKA3dIiLQzdYtIEI3vd+1lqzMjPVszzJ/nIiorKyqe6vqZlVm3fu8X6+sysiIOOd7TkRkxvOJ7/fzaZuScjbB1iVSG5K8R5LmSKlwzlKVcxaTMU1TYdKC3hBMlqHN8dSSZsVtj310Iz9+/X0IoAMkzlE4t15QrsSEW3WMrNNxmpZqMWM2mTIdHzCbTCgXnQjkrCWEbiRDKYVJzFoYWW3zqDgCrD03VuLK4Q6Uo7GzK5FknXBySGQ4nPpyNAHmOhHpGCJzp9PpUkzqFu5pZlDB0BYZTZ2TGEmWZQyHA/q9vOt80OqOhTHvw3IU6/iW/N1YVzdWhTwq1KyEEI/k+utCCEgEQhxKAXqekaBHHz6/jpJum5rMCJTszl9WFJzZ2em+tne6ka61gKWXX2r9vDwqjHjvadtrXVla3Z+0n0jkQcEHgQ+n4zVxWuq4W6IwEolEIpFIJHIE7zxtXVLOxnhrkSZBJxlSKJqmZjGbUpULhFIMN6AYbBzrgu1wqsmdbPfwp/8rMWLFcwkta7NL65bHN2e8d5WrVy6zf3mXyXif+XRGXZa0bdOJL8aQZhmf9ft+/3XbOlr34X3eyrvkcOzsqtPEGHPT291su3dzrp4PrTWDwYCdnS12trcZ9vp415IsvTSatiHJe2xtbzIcDimKgiS5fR+XFSH4tT/NcXL4nEspD53L1bmS60trr5zVtXdwPi+cP79+brVtS2I0SsB4OiNNUx46d5YzZ86wublJlmVIqW8QtFbn7GZRx46AlIrEKLTsBJ9IJBI5bqIwEolEIpFIJML13Q5Kdws115QspmNc2xKEwDpYLErmiwUE6G/t0B9tLxf06pbCw51yt+M4R0djnu82hwlLTxHXNl3iTFlSzRfMF1Om4wnj/avMxhOqct6lmChFXvRuq56j+3wur4/Dj8PzcVxCyM1GWrY2N7lw4QIXLlzg7JkzFGlGXS3QeIToRo6yImdja5vRaLT2z7gTYaTzcQEh7l3KyuHH4JpZ8K0ej67j6Ga3uRVnd86A7B4311qM0mgpmE6nKKM5e/Ys25tb9PtdMtLhLpTVPm4m2h2+LKVESYH3HmstRt8ooEQikcgLIQojkUgkEolEIksOf4KtlEYAtipZzPapq5qqbpktaha1RScFqhhg6Qw6WX4Cf3iRfXgheqsuiVvVcRzHcruEEPCHooPdkREcbbqRhxDC2ifEGIP3d2Y0+lzdKjcm/dyYUvJ8x/bpn/LJJMaglQYRsK2jbpvOWFRrNjY2ednLHufjPuET+LiP/zgef/nL2dzcvC7KeMWFCxd49NFHefjhh9nZ3iYxhvlkTLDt+jxlecFgOFin4GiTPmdnztHjDoF1gs39iKB9/u2vzFlvzmrcqmka2raLzh0Oh+tthxCWrx3JaDRHa8XGxjbD4cY6xvhmddzqfCml6IZ7Aq5tqKoFZZmjlViLLJHISxmPwJ8S553TUsfdEoWRSCQSiUQiEY4u0IHQfTpdNyWzyYT5fE5dNVSNo3ECj6apappFSVmWJHl2ndfFWgwJEIJjaWsJCMSR7pLTsMDrvD8c3nnwFiFAGU2a5OR5TpX1SPIFaduibEOSJOS9wQve79EugcPf36rD4bmOQUiJSdSyK6GmblmKPZ7ZbM7e3h4XL17kzJkz9AcDlFIMBoPrYmMBHnvZy3jsscc4f/48o9EIIUBLQTWfY9umE4fygqLok2XFTQ1zn79ecC5wTGFG95zV83t1rg6LfYcNfZVS1PUQIQT9fp88zzFGr71Lbrbdo9uCbpwJH7C2YT6bcXVvj2GqMUsTXuBEYpEjkciLjyiMRCKRSCQSiSzx3mOdpa4r5vMZ08mU/b0D9q4eMJ/PaRuHQ+CFwso54719LuVPd5/4+5b+YNiNU0gFYimyeN8JDkvD0W5Ux6BMglRqbVx5UuLIWhDyAbxbRhV7QgCpFDoxmCwlyVOyvOi6SVxGkqX0hn3+0Xf+DTZ3ttk8c5adM2fZ2DnDaHODrOijk85U89ve9E0MhxtsbGyyvbXFaLBJr9dbiwmrlJLHH38F3nuapmE8HvP000/zoQ98kA9+8IM8+eSTXLlyhfF43D0WdY1b1uoBJQTGXEukWflVrJJ2vIcQZlzZ3aV46unOCDQx6y6I4XBIlmVrceSRRx7h4YcfZnv7DHme412Lq1t6wwFt29I6R5Im5EVBkmdoc2f+ImJZrzEPiCqy5LAXyNEOn8PX53kOsI7j7QSM64XAm4khh1FK8bP/+Nvu5eFEIpEIEIWRSCQSiUQikTXee2xjWczn7O8fcPnKLs9cusrelT3m8wWNcwgkUip0UjKbViwWC6r5hGpxwM7ODv3+ALMUBELweOcIbbseOxGqi/9Nsh46zdFpjtLJLdNi7gcrAWc1RtOJOIfSbHSKyVLSLMd7S/CBNM/o9QYURUGW5Z3IsYzQ7QSfwx4RuhvBqBvmszkiKNq2vS6NZCWMrISNPM/Z2tqirmtc8GhjyPOcq1evMh6Pmc0WVFVnBtu2DoK7ZuwpNODxbpky1DqstTRtC1KSPPsMyfIxWo0NHU5FATh37gJnzpxjOByitaZpGtKipOj1l4k5HpMkJHlBlmUkSXbHxqsPIrfq8FmxShcypotoupY0c2Nnx/OdqxfaDXIn42uRyINIQBBOyQjLaanjbonCSCQSiUQikcghvF8u4Odz9vcnXN0bs3t1zGyxoGkcAdaCQZKkzGYzqsUUW06x07OMNjdI0hQpBAQP3hO8JXiH9wGkQiUZSTEgH2ySDzfJiiE6SdfjAfeLw+NDh/1FvHUE77txHyWRWnVxqUZjkgwhAkXeJ8tzkjzDJOlSGOm8SIRUCCkRAYQUbG1tkSYZed6N5WRZtu7OWHVslGW5rkspRZqmjEYjAJIkYTAYsLW9xZUrV9jb22MynjCZTFnMF5RlSdu2CBHIUoM2Buea648tBGQIOBeo6ob9/X2efvrpbiQoz+n1ehRFgVl2foxGIwaDAVmWrUc9utv2luJRQCcJSdp10hyOQn6xc1RgONw1sholOvqz49jP3XCn41iRSOSlSRRGIpFIJBKJRJYIIRBSIpUCJG45hlG3DWXVUpYVjbWARClJlpiuo8E2CN/g2wWL6ZA07T4tDyEgQycwBO/xzuMCoDQmH9Df2Gbr7ENsnPXLyN/snqRtHP7kvPv+WvqI9x5nLa6taeuGtqmwtl13UqxDXpeeKEoptJaYtBMFjDGoxKC0RmqD1NcSeoLohidGoxFpkpGuRYicXjHAGINzjqqqaNt2vR/oIot7vR5aa7IsYzQacfbsWQ4ODtjf32c8njIej5lOp0ynUxaLBbapEaETd6pyTlN3JrHGWJCSNC8YDocMhp0QtRrZmRyMu/tbuz5XnS+GWRuGKqU6ASjPyEOX4KONIUkz8kPCyEtxAX6zLpL7bSB8lNWYj3PuOsHmVrc9rv1GIveTaL56fERhJBKJRCKRSGTJauFvjCHNUpLEoIxGKIn1jrptaVtHCGI5huHxASAg8QRnWcymGK2xy+4L4QNSChDLhZoNOCDJCkbbZ2hajzQpSXL9SM1xcLQjJDiPD7YblfEe7wPedWMmdVkxX0yp5ou1UOGcA9d1vFzzSOHa2IlWSKXWKT5q+f3RxXHTNEv/lm6kSEqxTBzJINyY2LO678rMdCWSbGxscPbsWRaLBfP5nNms++q+nzGbTljMJkwnE8YHewB4AlprPJK812dra4vhsIuOVaKrbXW8nbdKV8tRM9aujoQ0TZcxu13STZJmawHluOKaTytHBbbDhqurn3WdR6vni7huROl+np+VKOKcvaXp6+p2AN53IqCSL97HLxKJ3JoojEQikUgkEoksEUKsDUdVlqKKDFGk+ETjlKQFGhcQwSPX3RQtfrW4cp75fIHSkraxNE173SfWcK1jw2QZO1VLkg8Z7ZxntNWSBH9sx3I4/ta1La1tsOuOENuZrAbfeYtYR11VVItymb6zvI11WNslsPjW4t21rg4huiBVueyyUarrtLnZAvjKlUvXBKc0ZTId0+/1ybJ8vWhW6trb0sP3XV2/um+/3+/MT9uWpmmo65qq6rxeDg4O2L96hd3LV9jN07Wp66KsCUjyXsFwtEme590i3tn14v6wkSiwFmXWzwspl0JIuvZhuRbtfM3s9cXMzVKDjopgSkmkTA6NsADIThy8T6wEmrZtcM6iteFmy55VjdZ5Gtd9n2qFkvc+OjkSiZwuojASiUQikUgkwmph1y2IhDGoLEf2esh+AXmBT2Y4JXB4CAHlu9u3ohM+5ovOI2NRVkgpaNuWumlpbXd7KQVKCoToug+KQZ+8N6Jxdtl1IuCYW5G7RJaWpqmo5gsW8ynlbEZbN/hg10alq+SWxWJBuVjQ1CXedov/lfBQ1yW26RaakoBrDc42eO8QfiX+rMaRrq9jNpusxQVjDBNjyNKUNMlJ05Q87zEcDm/xmFxbhK+SUIy5libj/TJWua7Z3NxkNOzTy4vOz6QoGAxHLBYl1lqkNiRp5+VinUWErh6TJjd4hBhjbohU7sSdFJdYrLVwKKHlpbiQPnrM3eVrr6Nb3e5eca37o3tOtG1DCOE5vXuc95StY17V3WOZZ6gXucAVefEQgiCE0/G757TUcbdEYSQSiUQikUhkjUBKjVYJSZaRFwVZ3ictclRikEri6RJcghBIIQkBfAg0bUCWFU2jkIC13eK5ddeiepUQKKXJckOBIsl7FL0RWdHrtn/MC7IQAsF5bN1QLhZMDw6YTaa0TYWUkjRNSdJ8PTIihMB7i226bgzbtl03xnTCfDahLiuCs4SlOGK0IssTvB10Zq0shRGu7xjxwSKCgCBog8PZbttt2uJ9fz228pyPzBGRBFibtyZJQpIkSz+RrrOkNxywsbnJ+OBg7UHSNA3OOWrbgu9GbFbGq2maXtclshqNWe3zcNeINuY675XDNb7Yx2luhxNLVlp+daJITdu218UE3+z2beuYzBdcnVVkSUJuNKk2L/nHMBJ5qRGFkUgkEolEIpFDrLwtUpOSJQWpSkiFJkEiEYgAIXjw4A+1RoQQqFuHWhp4dj4UjuADznddI2HZWaBkTtHrs7G9zebOGfqDEUly/MarAghh1TnS0NYNTV3Tts06TlUbidFdvK0SAbzDtSX1ohupqcqKxWxGs1hQlWU3llOVuLbpxJU8x9rlONFKFDhyCHLtNaHR2ixHXFLyrGAwGLC5uc3m5tbtH9dNOhVWHR5a627kZjBia2uH6XTMZDJhMpkwn0zX3wtfIbVZJ+QkyfX+LoeFkdU+1sKITpbn1V/XMXLYayMurO8NNzu3h41WuxGamrqusdaSpunaDPgo1nnmVcPl8ZxL4wUb/R7bgwww9+loIpHIaSEKI5FIJBKJRCKHEKuujiSnl/bIdUGmUnKdUZgUa1paFwgEBGGZCaCWYkjoYnmXqSVi2dkggSAEgs6DIc0zhhsbbO+cZbS5Sd7voe7Fp9RCcHiTXY2W0HajNF4LfCVwWuKbGrc4wM/3aKdj7HyGrRt83SCaCuEasDWubfC2Ez+SvKC1Lc6HtXAgpURItT6XAMZkFHkXh9tF9hbkeUGv16PfHzAcjBgMbhylub1DvHaAh81a8zxnOByyWW2xWCyYTaaMJ/vsXt0lu7zLeDxGCEGe513U8PK+17wy1HVCx2pfK/EFupGNw8d9s5oiL4ybGfPejGvjMzVN09A0zfqxOfxYrrYZvKdqHXvziqeuTrkyLVFaL/2C4uMXeTCIqTTHRxRGIpFIJBKJRJZc88FIKIoeG6Ntzp27QFs1KKFJdEqiLjE5mNDUzSHhIQAeKQApwHVdEqv3iVJKpBJorciynF6vz2AwYDAc0esPlqMk92YEY72u9B5vG2w1p5xOkaHFzjRNolEiYNuaajalnEyw81kniFiHcg7tHNo36NDiguveAHsHwXXnYJVQIw2oG8eBtrfOsr29zebmJoPBsBNHst66WyPLMtI0f0HHeXjcJYRuTCZZxgP3+32GwyGj+QbDwYjhYMj+/j7WWgb9Ef1+f91xstrW4S6Qw+k+1y+2A0Lcn8SVo+kvR81PX6wcFqWO/uzwZeccTVNR1/U6Uemw4LW63Wo7rQtMyopn96c8eXXCvLZc2Bo+4Eu7SCRyt0RhJBKJRCKRSOQISinyos/OubMorRltbHLm3Dm2tj7Ckx/MeUY9xfTgAGu7VBMjQWm5XIQtF+gIhLzmt9F1okiSLKPoD8h6A7K8143QKM29+JQ6hEAQXfeKbVuauqScjpntXaEt50haEilQOHCWtq5o65KmbvDO4bzHeYF3Ht96aBvwIIVCCo+SoKXE6GUkq7omEBxeyG5sbHL27Dl2ts8yHA6XoyspWpvlaIp+Xo+RO2G179V2V74jeZ5TFAX94ZAzOzOatsUkhtFoRJ7n6/GZmwkOqyjfa+k1fp3OcytR5FbdDncrZhyNyz38/2kSSFajZIuy4sreAQDDXoEQYi1aeO+p6oYPP/kk73nf+9BC8BVf+icB1nHIR7mZoWvni9Ml0FRVRVWXnUGwXMVHmxtEFes9s7rhymTBh69OeHp/Rqo1Rkm07ASv2DUSiby0iMJIJBKJRCKRyCFWHQFpmqI2t8nyHqOtTba2thn0+igh8b5B4lnMF/jgUUqRZRm9XkG/3yfLMkySorQmBE+z9PZwzqGMoej3yLJibbgqxL1LNQk+4FyLay11VVEuSmaTKfVsjHcVKoAUFuEszrb4plrGnHq8D/ggcD7QeGi8ACQyURijSRKNMQlaKbRUa1HoKHVdU1fXonU7zxO3NEu9Zpq6GlE5Dm5minp4zKbZ3MQ5d10U8GET2vX5OyJGHBZFDm/7ZuLIzRb3h2t7vuSWmwkhz3Wcp4W1H4tS9LIUgDxLUUrifdqJTEBdtyyqislkgncO6M5Z6zrj4qMClUB0XVkrsVFKEHSGum1L1dRUdQVAlph1jPKK1eNXt569ecUTe1M+fGXC7rTika0BvdRg1M3NWiOR00hMpTk+ojASiUQikUgkcoTDi2itddfhYAwET1UuWEwntFW9TJ9xSK0YDoecOXeOcxfOs7m1Q38wRGlNVVbs719l/+pVFosFIQSyvIdJk6Ufxz0SRNYL6QB+uaAXAqk0Ms1R1iJd0vmkBItrahpbUjtLXTe4xuK878ZmpEYojUpTVJqRZj2KwYBiuEFWZOgkReqVj8ONAsGzzz6DtZayWjAcbJDn+XUJMt24Szdic9wcrmW1UFZKkabp+hytPEJuJxVo9dxY+YscHrs5StO0y1Gm6zsQDneZXOfNckQoWXWprLpYjnqd3OwYTxNKSbIsAwHG6HWKEEAgoJVke2PE7Nw5WttF67bWU9adyLE+xwSkAHnkfAUhQEicc8ybmnnT0lqHkpJUCpAKT2d+vDpD1nkmZc2z4wUfvDLlib0588qSGk0/M+ibjIJFIpEXP1EYiUQikUgkErkJh/0ltNYUvQFb22eYXhizd/Uq08mYtu1ibZMkYWNrmwuPPMYrXv1qzj30MP3BECEE0+mES089g9KGg/1u/CbPcxKToZTuTFmF4Oji+RiPBGTnAZJkBcVoA6k13p1BSYmUAu8sTblgPj5gOj7Ajw+wixmhbZFKY/KcrBiQ9QekvT5pVpAVOXnRozfaIC0KpNFIefNxmPl8TpIcYIxBCoEPHrMco3nd637b8x7BUY+PuxECjvqH3Mn9VvtdjfushJHV9SvB5Whtk+nsOkHjcNfHSqDRSpEk18aJDgsk1trralgJBSuB4WadJ6eFVZ1pYhAClOrGzFaEEBBG0O8XnD+zQ9M0eB8om5aD2YLZYo7zDkL3qpACtFTdyJqSCKVAKpwQtN6zqCqqusFbT2YkOghS7xFNi5YWuTyHtQ3szio+cnXOh3fnXJlWJASKxFCkCeqYk6EikciDQRRGIpFIJBKJRJ6D60Yw+j02ts6wc+YsV3d3WZQVcrEgSRIGo022z57lzIWHOf/Qo+T9XjeqoQzlbM5sNiOErosgSVKyIidJ087Xgq7TYsXhBfRxLNKEEKgkIct7uJEj7w8xSUKW5yQmJXhHuZgx2dtl7/Jl9q9cIp0c0FpLkiQUo02Gm9v0NzfJewO0MV33iZCkWbaMGl4l0dBpPIfY2dlZR+KmWUav6HUxqkuh4Wg3xOH/V9d57yGAXH6ify8Wr9PpeB3Hqw7VBqzFlFt1b9wsqvfgYNLVvkxBOXy/7jnVda4URUGeXb+9EAJN01zXybIa/TnuWOd7hRACY7rlxtExq9X3WZqysTnEWot1ofMl2Z+xP5nQthYfQC47RoxSJFqRpgphNE5rGqFofKBqGuqmRQVPHhRWO5ysyawjkRotgRCY1o6n9+Z8eHfKxUlJWVvyTNNLNJlW3KMGrkjknrAc7DvpMoAbfu0/cERhJBKJRCKRSOR5WIsjScZgMGC0vc3G5jaTyQSQXfpJr0evN6TX75MWOUmS0TTV0gtBLMdGcpQypHlG3u+TZQVGmxvGabxfdQqo62q405q7Rbi85vmQGLK8hxCCvFfQGw7J8gIhoFosyPsDTNoJNtPxEGctaZ4z2txi48w5BhvbpEXejTw0DbZpO9HlcNTtTebMX/WqV9M0DUZ34zNJktDv90nTzn9iJXyE4JFCdt4RSzo/EottGwIBY9K1Sepx0wk818e7Hk6lOZxucvRcH/5/RVXXuOBxzuGcu04YMVKRJAlCdPG/iVE3eKysupFuNjpzK1+T08QqIQhu/vwVQmASQ49ibcg6L1su7024uHtAWbUEFyAEpOxGb4pEk+UalaXUScJCKqogqK3FOYsm0EvgwApGraPQmlwJEikRwTNeWJ7YnfPs/pyqtqRKMDCawkiMFCAe9OVdJBK5G6IwEolEIpFIJHIbrBew+VIc2Rgx3B8SgkdrQ14UpHmGMSlKagLdKETbVLRN240FAFKrrnsjK8jyHGWSTgw4tHC09aIbf5EaISVS6nUNd871IxdSSpTR6DQhSXOyokAqhdbJMvbUQgiYNMXalizLGW5uM9rZob+xRZIknV9Iuei2HsJ1E0CBzj/iMC977OWMJ/tUVYX3nropyW1Or9dbiyJtU3XmtEpdS4dZbr+pa+rFnCAFFBxrgs1hjDHr83Sz7oY7jcfNs+5cWQmNWJqxhoASCqUExqhlcsqNpq9CiBuEkcOizYMijjxXbUIIJNe8fOqmM1GdzEouX50xmVXY1oEPCAJaCTKjyIsUM8ios5SJ1MwDNN5jQ0ALKLRklLYMEkXfKPpKkktQAcal5eJeyaJs6GlBnqScLQx9s+wWibpIJPKSJAojkUgkEolEIrfJShzJsox+r89wOLzmQVL0SbKs+9RfLj0hnO8MTAUIKRFKdqKDTknSFJN2IxtHO0Zm06solWCSFG0y0GEZ6XtnIxRdB8fSvURcS/Lovj/USaISZCZwtodtNvDWIZSkrSuSJCUteqRpTpIkaJMSECjVeWJ47+HQaMfNqjtz5gxaaw7Ge5RlSdM0LMrF2mzVOUtdltT1Aq00JknX55AA5WLGYjZFmQRjEnyS3RNx5PAYyy3PJ7cvUG1sDrGtp2ka2rZdm6mu/UW07saLlh4jR0dk2ra9TjhZXX+cozQ3O9b7KbRc7/nikQiCh7ZxlPOaurF461i62KKVoCgaktbTFp6JVEx8oHSBJnSyXK5EJ4okkr5W9KWgryWpgLIOzBYNufDsDBM2UsVWphlkcYwm8uDRSYan44l7Wuq4W6IwEolEIpFIJHKbdAv1sBwBUBRFgfcepTV50SWtHI7flUqSZQX9wQbWBqROsd7RHw7oDQekSX7TBf7BlafRSUaa9cmKIXnRR4i8E1Zut9bl/0II5NIIU2mFcgq1NPrsRBm5NP0U5EW/G2mwXXdLUy064URr5PL23eVuJMalKd47kmU0MUoiJEvPlGv0+0O892vD0rZtCcGtI1m9c1TlnOnB3nLsKMOL0G3HB+bzKdViRjEY0R+M7u7Buw1u5m/ivcda24k5VUm5qKmbhuCvxTTneUaWJmtxY3XfftHDWktlJG2jscvzerj7Q5ubd4wA6w6a1fUvxHz2KIe9W1bfH/Yzud/iCICUkKcJo17ORi+nqSwyVFghulEkFwg20DQO1VpoHUE4bOuorad0ARs8tZS4TNEoSaUFpVLURlFISQiQScFgkHJ2I2erMPS1ZJgb1IO9rotEIi+AKIxEIpFIJBKJPA/XFpEO2zQ0VYW1dh1HKqReGokKAtcWmWlWEIYgtCLvFcznJc55iqJgNNokzbObfvq/e+lJ0rQg74/wtu0+yRYKLUQ3XnMbi1ZBZ4QqlUAmBpPkpLlFma5LJc2KboRBmWXUrgIRyGwPO2hw3q4X42oZtSqWJrRqOeqiTYL3Dq0TsizDGIOQ+oYOmDRN6PcHnYikFG3brjtvALx3NHXNdLxHXVUgBNa7dcRwW9WA7zpJll0B95Nrj//Sa6SLWSFJkuWXWY+DHO46WYkM2rulKHS9uNFF0V7vp3KUlRgF1ws3dzrWc/R4oBvtadt2KVQFlFLd43tIjLmfaCXp5QnntkY0VUtuNPOyxjYtjbW0rcO2HmUUaZEiMoVyntA6hLdI14kjiZIkTmAkSCe6F4KXaKNIlGKj0AyKhPMbfTaLBCNhkKfcIlQpEjm1+CDwN/F1OglOSx13SxRGIpFIJBKJRG6DEALOtlRVSV2VOOtQSmNSICxTQ3w3OgOrRA6DGgxIsoxef7iMJPUYY9ZCws2iY2cHu9RpRtvWAGiToJMcbZI7qlkgkEqTmgzR8xijsdYipSQxBpNnKHNthENJjTEGk6QkSYaz3fiH1qvFskYajV4auaauM2KVSqKVQS0X1YLrOw60NmQZSNl12VjbiS4r89WVn0i5WLB35QqzcsGiXOBDwKiuvuFwyM7ZgFBdR869ZiWGrL6sdwR/TfRaxdGuInePxvXe4AOCBNx626v/vfc4L9E3EUWOCi2r7R7+/25xzlHXNfNyQVmW2NaTGEW/6HUjTgn3XRyRUtDLM85tD5ACRv2cqmporV127Vjq1hEAnSqChqKqSENNIWAqobSQGslGYSiyBCMlqZL0E8MoT+glCb00ZaOfsTUo6GcaASRKoqMyEom8ZInCSCQSiUQikcjzcH1Eq0AaQ9EfoExnrtk613mGmASp9HUL4tXIRJIk13lMPFfsamsbnOsElsSk1L0RWdtiUn/bn2oLIUBKFCBSgVCSJLN475deI52vxbXFb0AogdQKkyQkWUpYjr+YNEEvo4W11iidIPDdfHtYdqbIQ+MgR45ptQ+lunjartuBa54ZSnexrkIwHo955plnOJhOcM6RasNg2CXYSG1QKrnni/XDnRve+3ViinP+uoSam/l/HOXaOblx+6vHwvvrR1pW20nT9NhHW66NBzUsqjmTyZiDyYymbEiNwg43YHNEIQb3PRZYSkmaaDZGPRKjGA1ymqbFOY+13ehV0zqs8wThqW1DMnVoJ+lpxcIqShvI0oSdjR79Ius8a5SilyYMsoR+nlKkhn6W0ssSMqMJOATinqUdRSKR008URiKRSCQSiUSeh9ViVemErNdja/ssRa+PtY62bWmaFiklg40RSZrdsnvgsJ/Icy3AnAcfOl+Lpmk6L45lJ8qdjFAcXrArpQghua7z4Dq/itDFAyulSdMcAihj1qk7aVagkwSlEpRUCKlv2NetjmtVx2H/jcO36wSTnCQraKxlb2+PK1eu0LYtaZpi7RkeesQtxRcOiVT31gvjcLeIba8JJNcdk+K6rp+j3R1d8sq1c726/1qgWv6ss68J1+17NWp03HhvqZuSqpwxnY05GE+o5iUpEsoKFTx6lYBj7l/qTfc8FeSpwShFnqXrc74SppzztM5TNRUH0wltrXCJpNASKyRV48nShDOjnOFgQJYkGKPJkoQiNeSJITGGzGiMVnS2Pfq6TqBI5EEhmq8eH1EYiUQikUgkEnke1uktWtPrD0nTHO9s90m26zwaAJI0pyiKmxqq3smCa7Zo0TIgVYILEIJgNb59pwu3w6MYR+97+HJg2QVhTHcfrUizvDObVcuIX5Us/UjuPCb28G2P3k9Khcly8t4QbRLatmW2TK9p25bhYLD2wmiaBpM0hGCWYk94zm3fDSuxYtUt4mxnwGqtxTqPOtTFsYpaPlrDYZGju54jQsi17hC/HME6LPgAJMmdjU49HyEEgvOEpsFWC+pqQtPMCa7GNjW+bpnWFamgG7VKE6Tu3feuEQClJNpIgr9+rMn7gHWe6UywWMzRSpAnhkIIUJLaWLSWbOSKrUFKkeWkJiFNExJjSLRCSYmSAimfO4UoEom8dIjCSCQSiUQikchtsOq6kEsD0puleqyuf6GfPD996SrDXo4yPYJQCKm6bokXWP/tXK+URgiJ1N0xihBAXh/vezvbu9PajDFkeUGaF7BKaaFLEWnblnI+Z7y/R38wwntPni9TgFRnHHvU0+N2j/tm3OAB4jqBxDqPO9oNcqQz6OhxHe4W8u76bpeV+NJ9hRsW6GYpUh0HIQTwAVyLq0tcNYV2ghE1uQEShW1q2nLBdO8qWZGTDXqoJLuvXSPQiSMhBLRSBHn9OQkhYK2nqjrvGSUleZIgpcQLwDkkgUQKcq3IE02WGNLEYJLOH0dIed1rKXaJRCKRKIxEIpFIJBKJ3CaHF91Hxx5W1x/+/2556qlnOXtmm/5wEyEN0iQIKQn3eAF3eCF/dJF+XMd2q/1KZTAmwSQpRqcoadAalBJYaxnv7/HMUx8hINhezBmONiiKgiRJEUuvFKmWApKQ143u3K04EkLAc00A88ufr8xlDwtgR7tyVpdXXiQmBEhAtN35tdau93NYILlZdO6xESyhqQnlDFGO0XZOoRwm1WRWU7WapppRjsdM8oJitEla9FFarY9pVXP37b17Pt7s+bYeaVNh7e+SJTlGaULw1G2LCKCkQAmJlmCUQClQEvQJxRFHIvcKj8CfkhGW01LH3RKFkUgkEolEIpG7oBuBsARvwTtCcAihOvEiAMHjbINrS+rFhOn+Fca7zzLe36WuFgjvERKcDexNZnzk6ct86IlL7O5PybKUIs+x1oNQKGkQchmp+wIIIWDbhrac0yzm2KbCLubM9veZTaeoLGPj/KNsnb9A1u+j1P17q7jqqliZma4Xrj7QViV7u1cQUjGflUzO7bO9s8NwNCTNe5jEoHWCMSkq6WJzV+k6Wpu76uA56g8ipUQROkPbQ8axtxopOhzLuzL17DocHG0rr+s2OiyIHB3BOTYC4ALUNWIxR1VTMj9DiYATmkwKpHDU3rIo5/irhnxrk2JjA52Y68bDVibEcPeRwXfDYbFEa0WR5d25c56yrqgahxBd4lKWqOtSn+6lsBeJRB58ojASiUQikUgkchesFr7h0OUQui4A7ztRxNsa25a0dYkPFqEU0iRI2+Jsi7OWxWLBeDxhsj9hPp7QTBdIH7Blg7PdAlTI44mo9cHjmobFwT7TS09RXr1EOd5nsr/PfFGTbZ1F6JTexgZJUSDl/Vv4CgRItU71Ca7FNhUoSVVZ9q46yqpmfDDm4OoVts+cYXNrk6LokWQFSZaRpBlpmpGkKUXRozcckRcFSZLdtTjSecvItdiivD80MnW9MHIzn5F18s4qxcY5EA3Oqc6/xLlrogjXdyEd67gSgeActC2yrjFtA1TIUOPKQJgHwqylmpaMpxUz60k3NuhtbqKSLl5aKdkJgC7gZZdsdBJCQzd6pRn0CpLEYJ0ljGFetog00O/n9Ho90mWSUjRVjUQiz0cURiKRSCQSiUTugm7hKxBC4fBAt2j0zmLbGlsvaOs5rm2wbY33vku1KfqAoCwX2NqxKFvm8wVNVSGsJSGQ+ID0HuE8wa0WzMewaA4QvKWZTxg/8yT7H/oN5lcvU85rGpnQBMFoNqGta4IL9+2dohBdnK3SCqMTpFTY1tJUNWgFLtDUFbPZgsnBhMnBAeOrV+mPhhRFQZrnpHlOnuekWUFeFPRHG5w5e57tnbPIgcAk2bV93U49XIvk7WKK3Toh5Vrc8nNv77A4cvgx006tu2NW1wOHH+JjJ6z+DSCFQIVOfPLNGDcpaa7WLK62jPdarswtYV4hej2y0QiVGGCTNNEIbwnOI5VEaAji+HxQbhchBIlRFEVG6tIuxrduqTILecrmoMegPyA/khAVxZHIi40uzeqkq+g4LXXcLVEYiUQikUgkErlrBIhOHAl0C2fbVLTVjKqc0pQzfNvgvcMhQClMVgAKF6Aq6/UogA5QaIM0gVRrDJLgWpxrca1ddhY4oBtpuBtxpEtG6UZVXFOy2N9ldvlZLJLQ2yaEQNs0tE2NdRYd7t+idzVykmQpSaIhBKxtwbcEL5cmqIHFYkG5WDAbj8nylDTLuq88J8sysiwn7xVsbG1j66rzltCdkKF0csdJOkII1DKR6LAworVGSfW83QhHu0hWYktiFGDWo0MmuTZGdPR+x4ZUoDXBpCA1oWwJ0wnt/j7V5SnjZ0t29x0XK4nNJjRKYIoe0iw7Q4oM5VtC2yCNRudDlBRwAp0jSimytFuMNa1k0OvhQkAJwWhQ0O/1rjOvjV0jkUjkuYjCSCQSiUQikchdcM1wVSGExyMIzmPbmqqcUU4PqMop1jaEAEpqhE5Q0hASQdK2KDNFqM44NE00gzzFBIE2Bk3A25amrmjqObapcUm69OIwd1WzEAKlNSYvMHmB0gahJEZnmOGQrNdDCLBNS7AtIWTHecqeE7kURrKl0KGNQgmBoBOOrLNYu4yabVvKct6Nt2iNSTRpmpIkCWmakvcKpuMDIJBlhixLMbpLsOm6fO6sa6RLIpJA2gkjSpKYm/ih3Mb2VgIQsO4YWX2/GtE5auR6PHQiHsYQjMGGQFvW1OMx1f4+8719Dq7M2Lvq2K0kiyRnETwyNSgt0ALYHJKEGtGW6KRTJaSUkAhA3VcRDUBrvT5/vV6BUp3olGcJyRFflBXHf14jkciLgSiMRCKRSCQSidwF16XTLL1GvPe0TUNZLpgv5tTljLapIQiEUugkI0kyCJIgu58pLTFGk2UG30uQUgACpKeu55SzA+bTA/JeD60VqyQQpcwdL/KEkAitSLKCfLhJNtygGl8FnZAN+iRZiiLg2xpnW0LwhHCfPmlfmq+maWc8W+QFdT4neIuzdpna4hBC0LYtTdMAoKXCGEWS6KUJqyaf5zjnyLKEzUHBYNAnK3roNEPK23v7u1pwr7pDus4RtyxV3CBk3P5hinXk8+GF/c1Sbo4VAUEKnBTU3lOWC+r9A5r9MdXBmPl4wmw8YXpgOSglYzVn7h3CdM9PIyAstujLGtPOSLKcEDxCKrRQSHOjB869fN4c3rZSkjxL0Fouu3vkdd0396umSOR+45F4jjm96i45LXXcLVEYiUQikUgkEjkWBC54Wmtp2oaqKikXC9qmwjmLFBqT1rSZRSlN2zRY23YeE1qgU4WxBo/HefA0lOWMycEuvasDkjRFraNo+8i7GF8IISCFRKcZ+WiDfHOLxdWLOOfQMmAkCNfimhLXNgTvQF1bvK+P9B4tLqXSpGlGvz9gtDEiuJa2qVgsFjStRYiw9srw3oMPWOkBD8ITggfnqaWkWpTMp2P2di+zubXJcLRJXgxue5zmcNfGSgBZdXoAdx37ergT5eh5PXqb48Z5KJuW2WzMYvcSdvci7f4+0/GU2WzBYlEyLxvmC8lMNNQykKSGLNEkeMJ8izOmIQ9zsjzHeo+XCalK0YjrUowO+3rca4Gke2w6sfBm+41dIpFI5PmIwkgkEolEIpHIC6Qbp1kmxwSBs562bamqkqqc0badAKJNTpZXKJ3gbUNdlrRtjfMtKDCpBilorCPIQF0vmE4PKPZ3yfOcLC1Isx4mLe7SY0QQpEQlCWl/SD7awhQ93PQA11R4W+GamrYsaasS37Z4IQmEpaii6Q7x2rjHynFPLDtd7nYB2i1wFWmWMRxtsLOzgxYwn01wzrEoaxCBZTXLsNhuOkQvfTukUgi9jDcWgrpumU4nHOztsrkzpr+xg0mLOx5/Ofr9rW5zJ8d6t/e9W0IItG3LdHLA1WefoXzmCcTlZ/HTPRazObNZyaKuqRtLYwVWCkTbiSi7lzthRM6v4DPLhqrIioyq9dQyoydSsmE4JDqJQ8a16joD1HvBShw5+rObfR+JRCI3IwojkUgkEolEIi8QIbpIXSkVQmuEUvgQaJqWsiypFiXOOaRekCwWpGlGCIGqXFCWC1prEUqjiwTpQbuAR6ATA0ha23SmqLbBOdt1RtxlhIkQEqUNphiSb2yT9IaU4zFtuUAtFqhkTjWfUE4OMFnWeUk4RyCgTEKSZqAUwXceKME7hJQok6C0AXn3XhNSSrIsZ7SxwfbOGaTsvDdm8xKlSpwLsDz2zvTzmi+JSVOMMZgkIc0STJLhEcznCw729xmP9xktStKiQcrbT6i5k9udVkIInbi0WHDl0hWe+fCHsU98hHx8GVlNmFYt00XDvGlpQgC5GvHq/HMWixlXr3jS8ippz0LSkqYpsg4kPqdvE/qbFSbLYWnuq7XpxqKKnDRNbznaclw86I9RJHI3xFSa4yMKI5FIJBKJRCLHgBACqQ1aZ2idd10hsBZHmqYhCEmS1KRZjpSCpi6p6wrrHEolpGlOkvUR0gACqRV5XlD0N0jyHsaknWGqkLyg7gxlMHlBNtomGW0TrlyiLkvC5ICgDF4ZpDJYZzFJSvCdGJHmPbL+EKU1TV1TzyfgHCbLKUabpL0+KulquxuUUqRZTn84YrAxomlqFosF0iSgJEgBXiIISLUSUjJ6/T5ZkZOkOUprlEnQxuC8ZLao2N8/YH9/l83JAVkxALqxGCnFstPnxb+ods4xnU559ulneOLDH0E89QwbzT6pr5lVlvnCUdmAF6CNplcY8kFKr0hQUtDWC+beMWsbMmNZGE1TCkKdkVeCwfaYNOuDUATReagMh0O2t7fZ2BiS53mMzI1EIqeWKIxEIpFIJBKJvACueVEotDJddGzRI8n6mKSHVGM8gtZ5gg8gLFK2SCWxrcdai7UWrTOyfMBo6yx5f4Q2GVp3nhtF0aM/GlIMN0mSfP3p+92O0wgpMGlGNtqi2DqLyZ+knC1oDvapXSApK5qqoirn5P0NuumIQJrm1IsZCEmzmFPNphAsxWgDqRUmy5AhWe7nzs+jEAJlNCZN0SbFC4ENXZqKFAqQIDwCkEJiEkWv12O0uUlvOMIkadetA9jW09iadu7Q5oDh5SsMty6hk5yBd50IlSRLgeTFLY6sxmgmkxnPPnuJp55+hvTyPogFmWiZtbBoAw2CYDR5mmEGBYONjKJICUEQ6gblWtqmpnItrhJM51dYlAY1h/zKASbJCULjESRpzs7ODm3jMMaQJMmL/jxHIvebgFgOFp48p6WOuyUKI5FIJBKJRCLHgJSSoDU6ySh6I/rDbRbzOVVVU9sAIsEFSJKMLM0RMgCCuqoQ3nWjNlKQFz22ts/SG26R5jlJkpEkCVmaYbICk6bIQyaTd4MQAmkSssEG/TMPkW8/yWTvKtVsRlXX6NmcqlzgnCN4jzYJoW2pxQHluBuv6RJ4KmTwtMkCV1d455bGH92x3Q0hBHwI1G3LfLGgqmta7wlifQMQEikFxqTk/R7DrR2Gm5uYJOkMRquG1s2o6s5IlmC5XFwiLYY4FJtVyWC4Qb/fpyj6GGNe1Iv21SjNfD5nd2+PK1cPyKYlSlp6OlB6KIPCak2SpST9Hv3Ngo3NnCIzeBtoS4+qFMJB7TxN4xiHBZN6n9AY9HgGKsEGhQ+KJMuYTmdkacbW9gbDYXeeI5FI5DQShZFIJBKJRCKRF8iqe0NKjTYpad5nMDqDsw6kxqQ9yqqEACbJSJOEEDyzg6t4a5nYFucsTVPhbIs2msFgSH+wQZIXXVeDUkhpUEreYDR55/V2fihJ0aN/5hyjh17G5OqVLkVnMcc2LT6ELuVDSpIswzcNIQTywYDB1hnywZDE9fD1AiFEF6m79D8RLCOH74AQAiF0HTSLxYLxeMJkMmG+6ASarnBACITozF5VYkiLHoPNDTa3zyC1oaxrKjuhtZayLHF1ifQte3tXUWlBGyTz+ZzN7Tk7O2eRQqDU8AWf09OO9566bZgvKsaLhqoMGCVpDFgpqXWC0ClFUdDbzNneztkYJmRG0raO0iicUohSU5c1lfXUztEKh28aKj/HhgWNEwQ0SZaRpimzWWc+7L0/6VMQiUQityQKI5FIJBKJRCLHRBftakizZWqMFOg0ZzDcpmlqBAKdZiilcE3NQd4jhIC1lqrson3LxZSmLgl4VGJI0hS9TPuQLzD55bpapUQnGb2NbTYfeoz57mXKyRi7W+NtuxyZ6QSfpOh1moQymLyHynKy4QbBtVQHDtvW2LrCNW3nR6LUXdXkfTfysZjPmc/nlGVN27YASKVQUoEQaNWZriZpQV4MGAw3GGxuIaUmzKaI8XQ5otTVE4KjrUqmkwkiuYz1XdxvojRFUZDlPbRW3G2Xy4NB97zxCCovqL3CoGlVF5PcaoMqehQbKTtbGWc2FaMioERLrTyJd1Re0jhNUykqbwnSkBd95HCIlYaqBR0ECEOSZeR5hkle3N04kchJ4ule06eB01LH3RKFkUgkEolEIpFjYNU10vl/pGuRJMsKmo0G59zyZwqBpGlr0iwjBI/3jsn+ZQCcd9i2S5+RsI49Xe3jOOsVSpEWfYZnzrPx8GNMdi9RTseU0zn4Bd75TkCoatAaITRBacxgo/veW6rxAa6tO5FlqyLznhc0RmO7saIQAkor8rzHP/pnP3Rsx32zfd7s+xfbQl5KidGdYCGSjFIkTJA4JEpIlE7IM8Ogp9nowWbm6ZlAcA4RGrwIeA1WC5yUeJmgTY/ezhkG5x/F6Zym9XgkUqWkacrm1hZntncoimL9HI5EIpHTSBRGIpFI5JRQLWb85I/9EP2D9/LdP/Iu/ulbfpg8z0+6rAcO+/Qe9j1PE2Ylbv8jqO2Xo7YHJJ/y0SddWuQlwLWRGokxKVIqTJKQe79edIvlyIZtW7TWsDRQTbOCtqkoekNMkqOUBinXEsO9WKivvUaGIwZnL1DsnEc+8wxNc4CzC4SqqJeeIy2KgGR8cMB8OmNvYwMlBb4uMUajTcLg3AV6zl5nCnsnBrFdso8iy3OGGxvrbpp7RTe+03WPrEY9pLw2qnR6xJFwQxTmndTWxecqil7BcGODfLjBYjJh4VqCEiRKkmmFNJJEezLZooND2ECwlmAdwgcUEqkgKE0wmnQwYufsBc69/FXooo/1IKRG64TEJPT7fXbO7NDvF/c8rjcSiUReCFEYiUQikVOCloFhc5H5Ys6XfcYr+J/f+Fn85//ycydd1gPHxe//QbK0hx1KwjO7LH7x3fhEk79ti0e+9c+fdHmRlwCHF39aG5TSRzoROv9QrfX6ttIk9AYb1NUCYzJGW1vkRb8boZH37u1al1AjSdKcYrRFf+ssMu/TOChnC8AjZzOC0jih8SiSLKWezZhvjrpUFwFZntPb3KKtKpyzGLrjXYkOt9stsIrf3dra5pFHHmNzc/u6c3ecrGqz1tLUFdZ2IzvaJCRJeirSalbHHoJjpYwEWMY13754s+pU6vf7nDl7lp3z5yjnJW01w4mA1eCNQ0gHvsU20FYSaQTBemzjCF5A8OvkIEVKb7TJ9tkLPPzoY+TDDQICqQxad+JImiYURU6e5yd+LiORFyMhCEI4Ha+r01LH3RKFkUgkEjkl/L1/+A/49MdH/OT7fp6Pf/QMX/95n8Df/nvfy5//mq866dIeKBZXx8wWT9Mupigr8akkS7apJ3P2fvJn2fqsTz3pEiMvEQ4vAo8uCIUAIQxpBkJsYYyh3x/RNjVCSvKiT94boI05tsXkqnPjaAeHEAKhNUnRIx1ukg23UMUAqs5rxLYO11i8UKgkBVKklKRpTtHvo5XAJAYhJN62eOcIPhBEWMfE3o4w0nmoSLK8YOfceZIkpWnqeyqMOOcoFzOmB2MWiyneQ6/XZ7AxougNMMd4/u++Tod3Ld62hOC7RB5lkFIhlbntjhylNINBj4ceusBjjz9O07bM9scEVyNpSUyF0jXBt9RNYKEC3gsIAhc0DgVKoU3A5IJeMWC0c46tM2fZ3jlLf3MLIfV69Gv1v9Z6fTkSiUROK1EYiUQikVPC13711/HeH/+/efmnfAF//3vezDf80U8lTfonXdYDxeR978fkPZrFGKUkziikaAnzEtHX1B959qRLjLwEea5Fa2fU2i0g07yHcxYQaG3QxqB1cleftB8dZVn9v/q6IepXSpQxpL0+ve0zbD70GPnGNr5tqOYT6sWcICRJ1mewtcXOQw9x5pHHGGxuIrynLRfoxEDoPCkOz33cSRqJEII0zdjY3CLP82tpNMfMqlukbSqmB/tcvvgs472rOO/ojzY5255fLuyHLygW+YXWCIHgWmw9o6lmXVeLMJikIMlyNNyWOLIy7i2KgnPnz/LKV74SgN0rV6gXM4QtScSCnlpgVE3QjlaAousoQimUNAgkwXr6uUamO2w/9DCbZ84zGG1Q9PvIpTDSiSDL9KDl+TtpgSkSiUSeiyiMRCKRyClBKcWf/Na/w9ve+p8Iz/4qXsyZt/fmk9LbIYSABwgeKa590nea39xe+idvZXhmk/LqRYIwyBAQPkFv5tjKc+4Lf/9JlxiJrFm9lpTSCCHROlkuhgVCimMdPXDO4Vq7FBpCt89l7K8QAgIIodBJRrGxxfbjy1Gf4JnuXmK8exnnPFnRZ7C9w84jj3L+Za9gtL2Db2tmu5dpq5IQwFuH957Vb4076RTovDA0eZ6TJOk96xYRQnTxtYuS6cE+e5ef4erly7RtQzHeByDJC0ySkWXZif3eC8HjXUNbj5lPrlKXC4JMyPMB3m+SF0uR6zbGrZRSpGnK9vYWL3/Fy9Bas729zXx2gKtnKLcgCQtMqFBYjHJoJVBCIrVB6pSAwPlAEIZscJatCy9nY+cMWVFgTHqDj8hp/nsRibwYCAjCKUmDOS113C1RGIlEIpFTxNf/pW/mnT/2bzh7bofFlYrPeMOrT7Sef/N/fTPv+eFf5Jt/9McfiD93Z974KVz61/8Oj0TIhFY5TNOw2D8g3T6LTJOTLvGueN3rXouzkl/51Xetf/ZxH/cJvPvdv0SSPJjHFOlYe4xICVICKyFAXHf93bLqEHHWUZUzytkC27adOWyaYJIElRikEDR1g/UenaQMNjboDUcoKTBa4Z2jtZZiuMlwZ4fBdjc6UWxs4KqKdjHDNjW2aWibBu8s3uu1t8Wdng+l9KHE3+MXR7z3eNdSVQsWswPKyQHVdI+yLqmqEpPm9AcbFL0BWusTGakRAoIPS2FkRjPfZTEd49G4ZoSUDqNlJ3JJSQjPLaStRKfBoM/582cxxrC1vcFiNsXWc7ALpKsJrka4GoJFCRBSoHSCTlIEEh88UqcUg202zjzCcHOLJMmvE0WiIBKJRB40ojASiUQip4j/+Y1/kPkTv8zOR3/ySZeCEILhf/oVRr2ruKZEJ/mpf7O78Smv5+Bn3wVPPEFLTWpt19qdZ7zqzf/H896/eXaPyb//SXa+/PPvQ7W3z8HBGCkTvu4vfAN/62++Ge89Bwf7vOY1r+c3fuO/n3R5kWPg2mvrHr3GQsC2lnpRMhsf0NQlBNCJ6TojtKata8Z7u5TzGfiAsxbnPU1d4zxIk5GPNhmdvcBw+wxJr48QCh8C1gXqpsWGBelsSjYYoZIEKfUdx7Teq7GLcGS8Z2W62pRzXFMifINva0rrOdjbY+/qVfobmyRJsvbLuP+/A0PnK+JavJvhmgOc97QsaIzHZgkmyZE6RdxGY46UkiRJGA6HGGMYDvvUdY1vKpytwTVLP5OG4O3SbFWitEGZZGkcHLpo396I3mCTfNBHGR3HZSKRyANNFEYikUjkFJHnOfkpEEWge/O7/fpP4Qs+Q/EjX/T7+V/+9dvvKHbzpHj8f/+zAMx/88M0ly6x8WmfdFs12+mcD/z1v8fLPvN33esS75iyqtHK8bYffxv8zTfz8pe/Eh9gOt0/6dIip5y12aqSJHlGWjcsZhOqecnsYB9rbdcRoBTeeepqgasrtOm6SJy1zGcLrIfeaMBo5wwbZ86S93vgLIvxAYvphOlkQrmoMFmgbRqcswQfQJ1898DaW8V7/Cqa11masqSpFjRVSXA1SnqMEjgPVVUxnkw4ODigKIp1gtD9SFY5Oj4kACEDSliMWCBCg7AtvjX4ZhPvtvDBIoPi+UpbdfBkWbYeWbLW4l1LcI7g3bLbx+O9Be9BdBG83ciXoBvFUpgkw2Q5aumDE4lE7j9++XUaOC113C1RGIlEIpHILfn4r/7foPo1RPIOQggP1Jvf3kc9Tu+jHr/t2z/xjX+bl/3Oz+VDP/NWPvYPf9q9K+wu8D5gg2fv6hUuXHgIgsAHR3iAHo/I/eWoz4PWGrICsSkQLMUCb5nsH1DOZ7imxVpLcB5tJCbJsJ5u0ewDUhu0SVBK0zYt04MDvHW0dYWta9qmQqYp2WCDtOithYSTjrpdJc8452jbGtu22Nbi2pbFfMJ4f4/FbEJTVzjn0FqTiBQpFU3TsJjNmc/n5Hm+ju+91zVfh5CIpZmpVoLEeGSokSIg3BxvZ93oi/dwm6WtBB5jDEopkiRZnisPwa+NaVn9bFnH4WNfCSxdOk6M4Y1EIg8+URiJRCKRyE0RQpD1etB7A//LP/93J13OPWf02Fk+8Os/w6XPeC3Fe36Dl/+WjznpktaE4HDBAwIpOqHEec/mcHjSpUUeAFaL1tVCWCmNSRN6gwGD/X3m4wmL6ZhqUeK9X3YDJCilCARUlqGlJAjFoizxe1cRSEToEmNMYuhvbJEPBvSHQ3rDDUxWIJU6kS6zlSDSjcu0NHVFOV8wn02pyjl1XdFUDXU5Z7x3mYPdXarplOAtShuyLCHJExItwXemtdbae2YEe7TuVRpN1ysSQKhlx4ZEqYCQHkSL9DXBd/G6ITjE+j7Pzw1JMcu7hhA6QQQO1QFdwoy44f6r7yORSORBJwojkUgkEokA737c4LJzKD/nfb/8DhpqBlsPY+uGRx956ETf/L/qla/mgx/8IAGQQuGwCOD97/vNE6sp8mBx2OQ1zXK0NmT9HqPtbapFSTmfUS0WtE3TjVTQGbbWVUmSF9jWopOUNMtI0hSlFVobkjQj7xXkgx5Z3ifJM7RJkEs/jpMQRbz32NbSNCXlbM5kcsDB/h7jvatMJ2PKxYK2rWmrinI+pZodYJsSLSW9Xp9e5ukZRT815Kkm0fevI6LrcGm7VJ9DiUErcUSI5bhMCAQceIt3lhD8srvjzs75+rbi2uUQbp4tsbruhvtGIpETIwRBCKfjtXha6rhb7qkw8l//63/lu77ru3jXu97Fs88+yw//8A/zB//gH1xfH0Lg277t2/i+7/s+9vf3ecMb3sD3fu/38rEf+7H3sqxIJBJ50XL5yv76zermxgCt78z08KWMyAsK36I3tviUz3wjuxcv8ss//1PM5xf58LlX8emf9XtOrLZ3vONn2DlzjkQnBAISsCeX5Bx5gFl3ChiD0pokzSmGDmdbbNvibNcd4azD2QbbWJxt8cEjhEJpjdIarTXaaEySoNMUrROUVuvF/El2ibRN1Qki43329/bY273M7uVn2dvdZToZU1cVzlq8bXG2wTddAkuiNLhAmqZI31BoQZFo0kRjlsd2r+t3rqVtms77RYhOYPIO58B6gfOdBwp4hOtGgpyrcbYlSIeU4qbdHCF4grcE3/mJdONFgSAUxiRok97WYxfFkEgk8mLlngoj8/mc17zmNfypP/Wn+PzPv9Hh/zu/8zv5W3/rb/H93//9fNRHfRRvetOb+JzP+Rze+973MhgM7mVpkUgk8qLj6t4EJSRSCZwPTCbT7pM/qdnajL9Tn4/P/rw/ct3lnfMX+OzP+8MnVM2N7F65xPnzXedKVVeMx9F4NXL3HB6lUEqhjSHJro1yBO8JfnXZd91KyKURp0AsF9FHF9MnIYhAlzLjnKOpu9Sdvau7XL10kd3LF9m9fIm9vatM9/aYL2ZY28JyZATfjaD83X/xo/e8xqPcbDQlSTJm0wPqsqR1vkvVoqUt59TzhmphcXWLFGB8g9MLbDqjZYZKJVon65Gpa0JO5yfj25K2mdLUM6qypGkCQuf0epvk/RFJmkbhIxJ5wAhB4E9Jp0bsGHkO3vjGN/LGN77xpteFEPju7/5uvumbvok/9If+EAA/8AM/wLlz53jLW97Cl3/5l9/L0iKRSORFxf7BAW9683cxr0rKsiIET900PPvEE/ye3/t5/JHP/wO8+hWPn3SZkRfIxYvPnHQJkRcZN3YWhPXlwwv65/IKOYnF9OEOkW50pqYpKyaTfXYvX+TSU09x6dmnubp7henBAbPZlHIxo2kaQvAIOjNpBUhx79qvVjU677qUHrpxptXXzc5dXc7Z27vKYlFjfcAHi6/HNPM92vkU1yyQeNIskJY5adkj6YHORmRpTlH0KYqCZOkTA55gK1y9Rzm9yGxyhfF4zLwM6HSL7bOPo0yKNubEDXMjkUjkpDgxj5EPfehDXLx4kc/93M9d/yxNUz7jMz6Dd7zjHbcURuq6pq7r9eXJZHLPa41EIpHTzsZoxM7ZcwybmkE/58LZs5w5e44zm5u89hM/4aTLi0QiDwhHuxhu9v1Jsx6ZaVtsW2PrhnIxYzI+YPfSRZ55+iM88+ST7F65wnQ6pWkafNtgXUPwXbilEAIZQEjQ4t6MyIQQsNbSti1l3dDUDUII0lSTJmk3snMTcaRcTDjYv8KV3T0m85K2bXHNHFvv4ap9vC0RWEyyIM0tab9FZXuYdMBwOGJrc4fNzU2Kok+aJkgBuBnN/DLz/Q+zf+XDXLlylclcUvQfRqseg+EGWV4shZRIJBJ56XFiwsjFixcBOHfu3HU/P3fuHB/5yEdueb/v+I7v4Nu+7dvuaW2RSCTyoCGE4Jv+wtfyD//JD/DbX/MaXveaTwTg/R/4EE8/e5GHL5w/2QIjkUjkGFiJIk3TiSGL6ZjpZMJ8vM/VK5e4+MzTPP3UU1y6dImDyZSmDcuoW1Ci655Y9sYgBCgl0PL43w5f8ztpOZhMuLK/x3S6QAjBaNjnzNbmUiS5cXylquccjHe5eOkiu3v7zOcLbLPAuRJs2aXnCIeUM3QyRyYHSNXHZD02N7c5f/4hZvOzDAcbZFmG0QLhF7jyIvO9p7m6+wRXL+8ym2uaOmFje0xTNwTvjv08RCKRe0tAcHO75PvPaanjbjnxVJqjfwyeL9btG7/xG/m6r/u69eXJZMKjjz56z+qLRCIvftq2MxfM8vykS3nB/LE//AX88q/8yvpyCIGf/blf5P/3v3zesWzf1hVSa6Q68T8fkUjkJYpzjrpacHB1lyuXL7J/+RKT/V32L1/i2UvPcunyFa6OpywqT1AJaZrxQ//239+wnRACtm1pmorJ3hW892hjMGmG0gal9F13yqy8W4peQdEreOiQOP3v3v6f8IBKEpTRmCPjK9ZaqqpiNpuwf7DHbDalrRcQajQOIQLgIXhgThAHBJGRZAWL+YS6KZnNJgwGI7IsI0sNRjaIdhc732c+n9BUc5xNCbYhrFJtCCcSrxyJRCKngRN7Z3v+fPcH4uLFi1y4cGH988uXL9/QRXKYNO1aDyORSOQ4OBjPKKczNrdGVFVFlmUnXdILol8UfNonf/L68qtf9Qpe/apXHMu2/+lf/QL2fvxDfMKXfQWf+ae+jNI1hHlJMRwdy/YjkUjkuTic3FIuFuxd3eWZJz7C7sWnme7vMtk/4Or+HtPxhKpusRjAYMON4yHddjqBZTbeZz7Zx9mGLO8z2NgiH4wQQt6T0ZL3P/sUab9PfzSiyDK0VNeJEUqnmKTApBlZltG2NUpYBAEtLDJYgl8mCbka5yVSOUKQ2HbBdHpACI75YkqaZGSZIU8gVQtUaxFCkSQGqVJ6/YQsS9D6ZOKVI5FI5LRwYsLIy1/+cs6fP8/b3/52Xvva1wLQNA0/9VM/xZvf/OaTKisSibzE+L63/ASf+4ZXsrmzgfcx//S5uPCqz+WX3/od7JQL/tzHfAwv394gEYr/9ad/5qRLe8mzsbGJlJLBYMBHPvLhky4nckpYJ8wcMlE9yQSZ46AzWm2pFzNm4wMO9nY52L3CbLzHfDHHNhVSQZ6maAweg7hpzG7A2YbFbMLVS09x9dknsHVJf7SJsy9Dqq4z7lYGqS+EJ6/usnN+n4fKszTDASnX/+3Je5ts7VygbgNZMWA+m9DWU3ALhJ8RmiltPaVtWqyDIDQq65Pk26TFJibtLeN3Ow+VQAChUTohVX1SMSBRgUDOYHPIcKO39DuJ/iKRyIOGX36dBk5LHXfLPRVGZrMZ73//+9eXP/ShD/HLv/zLbG1t8dhjj/Hn/tyf49u//dt59atfzatf/Wq+/du/naIo+KIv+qJ7WVYkEoms+Yc/+C/5tv/jJ9g69xCjnY/iV37qLdF87hb8ri/6M/zcz7+Vb/prb+KPnj2Ld4FgBO/4nr/Hp3z115x0eS9Z+r0+ykiU1DRNc9LlRE4J3nu8szhnCdYBAaEUUimkMkgpH6ixiXUKjWtxTY2ta1xbgrN41+JD95ZcaU0WIBUaKzSNBevqG7blfaBpGubTA3YvPsmVJ9+HLafUWzskxlD0N0jyAq3NsZ+j8XzM/mzCrC5prV0f22o/g8Em5y9A0RtybjalLOc01QzXHGDLS9SzZ6hngaYOWGeRekg2ukAxeoy0dwapim7kUWiUUhhjSI0g0w3GZ1BDW2UImdLb2GawMSDJkkPxvpFIJPLS454KI7/0S7/EZ33WZ60vr7xBvuRLvoTv//7v5+u//uspy5Kv/MqvZH9/nze84Q287W1vYzAY3MuyIpHIi4gX8sa+tY6L7383Jnhyo/nmr/5Crhzsc35755irfPHwjd/9L/mmv2NOuoxTzd7eVYqid9/GsqQSEATOB5R/0D+viRwH3cLf4pqaulpgm4rgXDeikeeYNAOdItWDtRDuhBGHbVuCd6TaMChyqtEQ7y0BCCKgjEIgcEFQYql9c8N2nLM05YLpwR7Tq5eY7z0L9YRaWqrpWerFlGC3r+u2OS5sPaWtFrRNva77ML3eAGMMw8EGTdNQNxVNvaBeXKUcZyz2LXXR0Nbd42zSIcXmWQY7j5MPH0GZPkJ2/ihSKpSSKAmKBuwIt8hw1VWEMmT9c/T6I5IkRcgHQySLRCKRe8E9FUY+8zM/8zn/oAgh+NZv/Va+9Vu/9V6WEYlEXqQ8e/kJXGu5fOUKr/vEN9zx/S9d2eNNb/pO/sgbX8/OzhZBShaLxT2o9MXDU2/9eh77vX/7pMs4lbzvfb/JV3711xCcIEjHT7zt7fd8n7/0S78Eomv1DyFgXRRGIsvuCuexbU0zH1PNpvi2QSUpWbuBGG4ghSTIBHjARmoCSClIkoTRaIR3D6ESQ5Ln7I+voiYJdTnDOQetRbcOJ9ojmwjYtqFczFnMJtSLCaKdk9CQhBplK4Ktcd7SNYcfbxdhQujcTwRIIRBH3ionSYIxmqIIOOextqWq5ywmCh1mKLtHT0/BQqBBJ316Gz0GZzbJRmcx6SZSJdeNTAkhEKHBtz1sUeDqM4DA5FuYbIhQGiHUg/VciEQihCAJ4XSI3KeljrslxgpEIpEHlieeegazXBD+6n9/F5/w8a+/o/v/+q/+KhsDxVvf/ov8sS/6PeA8rrX3qNo754kPfojHXvHyky7jOp66ustjJ13EKeTL/8yfYXdvn62NTZwPOOvuy5jC13/DN7B0EIDgYLm/B2lEInJvEMETnMU3NXY+pi0XKJ0gXUBrg1Km6yrQD9booFASk6T0BiO01mT9Ab2NDQZbO/T3djnYv8L0YJ/FfMpiMccT8OJaDG0IgeAD1lratsHbBi2hnyXkwtPLNakWaBFuECyOi2GaMcoyijTDLJNvrjNfVYqVGKN1Fzfctg2gcCFB6R4mG2FCQIgabRLSTJInkKeaJM8ROj0kdARAEEIKJsGaDJdvdfvSOTq51mESiUQiL1WiMBKJRE6E93zk/QyKlIfP3F3c9hNPP4sS3Rw9dPGGd8qnfuobeOc7380v/Lf3Md4bU+TqOf1FjnOx+VM/9ZN84D3v5U9/+VcA8O53/wKve90nAfDvfvhf8Pf/z3/M/sE+P//zv3Cq3qx+yhf/4EmXcOrw3jMrK7TWJIlBqZS2Le/PzkNYrnkEIYC1jq//C1/Hf/jRt/Prv/Hf708NkVOKQAIyeISz+HJBkA2NNJiswKQ5SmvCcpzmNP2euRVCCJROSHLQOiHr9SjaDfqbOwy2p2yM9zk42GW8t8vBwVUmB/tMpxPmi+nNt6U0WZ6zMRriwwaJn5LlGUmiUer4TVdXbOZ9RkWPIjFodesujVUnmPeetm2ZL2qaBpAFaTrC4FFhgVIKLTzCVRBahAhL09hrx9D9/TIEI9FKo/2AEDxCGoQyCPFgCWSRSCRy3ERhJBKJnAi/5WWvuuv7/sJ/+3lECAg6wzpjDJ/4mt9+x9vp9ft89ud8Op/9OZ9+W7c/zjfJ/+MX3827f+1X+dmfeQev+qhX8INv+df8jk//dJL5jP/+ofdz6eoV/u7f/hsPxGLldnj6PT+GLt9HOP/5nL/w0EmXc6x88Rf/CaazKYtFiXdgQ4uWx2/YeDO+67u+i8/6rM8ieI8UCgG85V/8v2R5fs/3HTm9dN4SEqEUShuUVAg8oSmxZYKtSlzT4LMc4QNCnf7fM6vXk5QSYVKCTjDBk3hH1nMUgxHDjU22ds4ynRxwMN5lcnDAeLLPYjq5YVtaa/I8ZzjcQC02ae0+oWpQSiHl0qT2HpmRpigyBMqDCNwwdn70d4f3Hmst1jmCMCTpiER5jAtgA2I5OuXaCt8s8FmN1BkcEjuujdUoghAEaRDrnz/Y7e+RyEuZsPw6DZyWOu6WKIxEIpEHjpe//DEyHfBB0IZAEurnv9Mp4yv/4l/gL//Fv8SkqTg4WPDbXvMJzCcTmkzz9OVL/PW/8g18zhs//6TLPBZ2f+6ryLyj19tEXvke5pcsTiiKj/1raP3g/xkKwTGbjZmVLba1KKXYGCb3Zd+vf/3rSbMeRhusa0mSFCkUaXJ/jF8jp5cgBUJppDbLhBJJ6yp8XeLKOa6p8daizIPzVvbGmGGFCgalPdqkpFlGrz9kONpkc/sMi/mU+WzG4lDHyGpsZZXWYnQnKLZtiy1rHAmJDwTunVhUH4ypZzPqakHb1qRp+pwdiV29hjTroaUgVQUpCdQWV9Z4W+NcwLU1rlngbQ3Ggrw2knNoa8vukOsjnCORSOSlzoP/jjQSibzkeM/lKcEFghDkGvqp4reOTrqqO+fVH/9q3vHOX+QXfuGd5CbBescn/rbfjlCSb/vO7+ZN3/k9/Oqv/upJl/mCyRMJQmDtlMaW+KBoQmD6c1/D+U/+eyj1YP8petmjD/PBJ54ED1orsjTl7W/78fu2/yzrsVhMUEoRgsMYwz/9Z//krrcX/UleHAghuq4HbZDagJKdIWtTY+sK11QEZ8Fbjttc9F5yw3NTgAwSKWU3UqITkiQjL3oMhpu0TUXTXBPP17G/1tGUc+YHe+xdvsT44mVsOaa3oTBVi3O+6+RY3v6W+78LxleusH/5MtMLF6i3z5DnveXoy43bXok4eZ4jRMC7Hlq0SGtomeObGdZ6aIG6RdULdLNAJUOUSoCbv57jazwSiUSu58F+NxqJRF6S/O7//AyVFYQgENITRCB85StPuqw75ou/5Ev5/X/wD/P5f+D3M5tM2Thzhg9/5CO8/nVv4J+/5Z+edHnHxrw84MxH/3nq8ZjLH/zbtE5ifYv3gnO+AtU/6RJfEF/xlV/F237iv5IXBQ9duMC/+pf//L7u/8kn3s/58w8hpEAJRb/f47Wv/cS73t7K1yAunB5chBAEJEIqpDJIY5BSdaJA29A2FbaucdbiQ1iOJgruYZPEPeVo+sqqGyRNc7wf4Ny1VJoQAq5tKOcT9q9c5OKTH+LiRz7I/rNPgW8YkZLNSkaLBVm56M6hSZBSIaTsRtbk0c6VO+Ng9wqXLz7NhYcusLNzll6vh9b6lq87rTVZlmGMIQSHDBbfBLBj6nKMLS3WgasdqqowbY1xLSqEB/YxjUQit0c3WH46XuinpY67JQojkUjkgWP+5Z9B+g9/gtYZQmMRVfv8dzqljEYj/tN/+amTLuMe0KUgALzvN36F3/K7fgf/6u++jJ2z2wgpCChaK9DmwRZFAF72spfzrl/6hROt4eLFZ+76vjdbjL3t7W/ld3/u732hZUVOEtH5SUijlx0jBgQ4Z3F1jW1qXNsQnAMfCGopjjzg3EwkWY3shRBwzlKXcw52L3PpqQ9y8SPv5+JTH2G6dwWEoNUDst1dsovP4IWh6G9gshSTZJgkRSUpWhuk0usujzsVSKaTMftXLnP52ac5c+48w40NTJphxI3eRCu/GGMMxujOb9k3OAaodBNhJgRlcdYSrMK0XbyvX0cNRyKRSOR2iMJIJBJ54BBC8NvHT/IVr+rzlsuW9z5zBXjjSZf1kqZtLf/1tQm/9QsuMNq5QF5s4n7HX0S/6nfz+i/4j/zAxdfTT1NC8CiZ4Fzgwkd/4UmXfQNf+js+jc/50j/Lhy9+gL/0l77tpMu5b4QQmE6n/MqvvJvxwZjf93l/4KRLirxgBKx8RkyKNBqk6Iw82xpXV/i2xS9HRgQ3moDecssPSDfRUU+S7tgbZpMxexef4uqTH2Dv4hNM93dZLBbdJ6/7ByRPP0mQmun4gLw/JM17pMWAvD8k6w3I8h5Z3kNnKVonqGWyzO2el8VszP7uRa48u83u+fNs7eyQFz30LcYKD287hEAQBkyOyTZIewuaFjwLEBKpUq6ZqT4Yj1MkErl7lhN/p4LTUsfdEoWRSCTyQPKz3/CnqFvHP/4HP8IHv+trj337V6/us729eezbPcoHP/hBvv/v/0Mef9Ur+Y6/+V187Md/HP/2h374nu/3uNk/OMA/nPGfv+NZPufrpixSSfJvvorhN7yfbPQon/fNl3nymx7mmY+/QNMs+KQv+CWEPH2+BmfS87znP7yd/3DxN/mcT3sdr/+0F79AIITgn//gP2b6wZ/DPvtrmO3HIQojDzyrTgOkRmiDMCkogw8Vrmlo64q2rknaGmU0eA1CrN/ZBvwhf85D73ZFFwOLEAgpX/BYyf1iHXtblUwPdjm4/CSTK0/TTK7imxJvLTZAOR2zf/kZXFszvvIMJi1I8h5pMSQfbNAfbdPb2GQw2qI33KDXH5HkGdqknUCyHEl6rvNRLRaM96+ye/Fpdi8/xPn9R9jY3CLNilt6jazorlMIlaDTEdnAEoJGp3MCkOZ9kqyHVAkvTmFk9VwUhy6/GI8zEoncb6IwEolE7op/+Y/exB/5sm8+sf3/yI/+GL//a98KoxS+5g8e23Y/+ZN/F7t7V8B53vf+Xzu27d6KH/43/5Yv/5I/QW/7DO/4xXfyn37iJ3ns8Ufx3vPUE0/f8/0fF2fP7FBZy1tr2P6hGa/8nwRbjyv8wVPIjUcAePSvP82jJ1znc/G3v/zP8NZn/wefvfWZ7Jx7hFc+cvakS7pvfOEX/2kme7+Pf/W3/g/cA26GGzmM7KJndYJMuohb58HVDXI+Q0/GSGWwdYtUyy4Df8hwNARC8IggCIS1GCKNRpkUlZpu27oTOU+9OOItTV2xmOwz379CO9tHuBItA0ZBsAFcTTvfpww17ewKQWiCTBBJTloMKYZb9Da2GWydYWPnHBvbZxlt7tAbjkizDKWTzghWqlsKJG1TUs4Vk4NdxlcvM5vs01QVvt/CbSZ1CalRSY8UgVQZaVsSQkDrFJP1UTrthK4XEV1Hkz/8g+Uxik6sO5bt3x6n/bkeiUTunPjuJxKJ3BVtdfXE9v32n/0JQib54e/53YTK8ppP+uP8yi+8cLPSj371a9i5cIbzZx9mc2PrGCp9ftp2woWP+xiu7l7lkY/7WD7pYI93/Pwv4gmcPbvD5cu7d73tn/6Zn+QNb/hUEnN/omMPfqnld0rBr783cOmDMNjc541f5sg37svuXzCPzRV//JVv4N8/++u8YusC/+8/fgdf+q1v6D51fwkw3DrLl73pH510GZHjZJVMYwzK5Aid0iJoygW126WqHbO9A5I0RQmJCAG8B+8JzhNW3y/Xi0IKZJJgej3S4YhkY0jaH6CzHKnVqV8srkZp2nKOa2bI0JAoyI0AJ9EqoFSgUJaUCmEbauuprKD1AnSGTnvo3pCsv8lg6wxbZx5i58JD7Jw9z3Bzm16vj8lytEnRuvMhOfo7pG1rbCWpyxl1OaOpS6y1dzjKpJAKTNpD6xTnWkIInRCmDFIZhHjhj8nKg+hoMs9q/Op+iS8hOAiO4B0Ey7pTRCiEVAT0Mob4xvpvb/uBsBQAV6MJ3X2vHZ8Uy9eAiFHHkdODDwIfTsf7FB8e7NdDFEYikchdsT+v+czPfSN/5Zu/hc/69E+5r/u+sLnJM5MFraqQheEv/aU//IK3+amf+jm8+mNezR/9wi/kZ3/yl5hO9njnO36RT/6U334MFd+ar/hfv5r/8pP/hb/1d/4mOM/WubMopVB4hDEvaNs/8h/+Df/qh/4F0/GM7/+//9kxVXxrPvld7+PXf/Y/8Zp3/RWa916h3YX8kZfd8/2+UEIIfOzrP5b/+dzL+Llnp/zTN3wabxvX/OmFwE/nyNHgpEuMRO6Y9YJNyq6LIc2QWYYXhnl5gN2fIC5eJlEJiVToIBAhIL3vukacJwRP8KHrFgGEUug8I93cpDh3jqF9uOsg0RqhTscb81vRLXLDcmHt0HhSAyJV6JCilcD5gNGKIjMkicYHgW48uAbvPE1bU5UldjzGyUvo4mmGm09z/tIFzj30KDvnLnTdI6NN+oMRxaBPkhY3+I94ZwnhkDlqOKQv3OZkSLdNTRCSIDVSp9eMlIW4qUhw5+drWc8RE4OwjABeDV0JISBw7XlyzGJBCI7gLcHXBFvifUnwLUJopMwQukCoFCTr417V78NK9GBdmwCEWAohIeBDWOqBDusDbvXl/LIpRSAlaCUxWmKUQit4iWjmkchLhiiMRCKRu6Lnn+biU+/lW7/tW/jUH/sxkpss4j/8X78J2+6z9fgb2Xrl5x3bvr/zzd/LV/5vX4ZMeszq5lg6Es6eP0OWjnj3O3+NxWQf17T8yL//sXsujGxs7vAt3/KN7B/sYxvHY03JZ376p/Jzv/gLWPvCEgVmiwrrHJPpmO/7v/4Bf/bPfMUxVX1zXvWqV/GqV72K9ou+lP/yLZ/Nb/sj33JP93dcfMxrXoNOEgq9we+78DD9JOdL3/I3+M0PPM1vOQFR5E1/48284+d+nh/9tz903/cdefEhpELpFJP3SHpDdK+AA0VdL7CLCtEGjA0oDyqAWH6t7x+630NCQJASlWdkTUUwkmTYJx0NSVwPFZJTHfPcCQkSqRQmMaRpQp6lmDYlUZ4s7fxBjFFkiUYphXMeI+m6ZrzHt46mLanaikULfn/M/OCAZrZHPdlldvUZhltnGG6dZ+fCI2yfe4jRllp3jazOjVIBaTRpnpH3CrKiQGu9FDXu+MiQUl/r4GAlshzD47Aaq/KdQIY/FP8r5bKFQhAOCz+di+81v5pj6FgJIYBv8e0U1+zjm328qxAyQSYbqLCJYgRIhFLr+zgPzgXapcEwgJICuazXe9f5zlhHaz2tDdTW0djQXXYe7wNCChIlyVJFLzP0Mk2WGLQQSE7vcz4SidwZURiJRCJ3hfJz/uSX/BkQLZ/7u9/Af/nP777u+j/+1V/Bj7zlB8EH/vgfN3zv9xyfMPJPfuD/PrZtrfjhf/OWY9/m7fIf/+OP8rs+6zOwsuI//vsf5fu+7+/za7/2q7zrv/36C9rudDqmqUGq5DlFkXe+8538iT/xh3nf+47H08QYw+f8jdMbQfxx/9MnUjYtH3j3/wCgaRyVnfLD7/8Af/orv4X+V30eH3nmgI9+xcP3ta43velN/OiP/xhPX7yM9/DYK1/FEx94/32tIfLiYiUGCK1IsoxsMKC/tUHTzBHC00zmhMYhW5BOIAGFWHc4CMTSwkHQ9QM4hBaITIMOBNF1lXSdD6c/jkAIgTFJlzLTG9AWfZQrSYxAeNd1BBiNXnqmtK0DAtY5rPM0zqLwqGVnTQiW0ATaiWcuKny5z3z/EvPxVbxrSBJDnmdL0SNZ15D3MgajIRvb59jcOc9gtMm5hx+/7eM4LECJ5WNzOLXm2EQR7wmNxTcNtLYbrQoBZPecEkYhlAHVmfF2HjQroQTCUiS5W+FgJXB41xJsRdtMcItL2OoiwS5ApqisRAdBIEVilqazXUu/c56qtTSNp3G+e/ylQGmBCL4TTdqWsmqp65aq8ZStp7LQ2EDjAj50YkqWKIaFwXvQUmKURykVfV8jJ07g9Pz2PS113C1RGIlEInfFb3vdo8jH/xBCBP7ff3XjJ9vv+rn/jETyyo96hNnsCq/8mMf5wG98+Dm3eeXX/x/ml/8b//5Hv4+v/c7qHlV++tjY2OZd/+2a0evXfO1f5Gu+9i++4O3+1b/8Zl7+8lc87+1e85rXPLBv7qy3aHl7f8o++dM/hYP5DKkMEPjcP/C7edu/+3E++J7/wWMf81vwOfyj7/92/uSf/b28/KGNa4mX94HgA//kn/9z8jRDGU3PJMxnU/73b/jf+a43f9f9KyTyokMI0Y3naU2WZ/RGffDbFL0U33hwCoJBBoEIAolASnEoeaYbPQjeYW2F8xUoSIY9VKGQhrVwctqRUmPSnHywSW/zLHa+S+NrqCUKS6I1iVEoKQjBUyuLd57aOpLGkShLqgLBBIySCGVIE0luHMbPYdHS+IpSeqphj3q8STMakqZp1zGiDCEEtnZ2OP/wYzzy+Cs5/8jjjDZ3bvsYVgak3WTQ4a5CsTYiPTw2cjeEsBJFGtyiws/m+LIiNJ0PilASkSbINEWmaSeQSA1KIqQELTuxRC7/v4v9dylCAeccbdvi6gVtOaZd7OGrXUI7J8gU1SiM66FdH5MlmKRYbQTnA23rKRvbCSNAK0RntBsC1lqqsmIxL5kvaqrGUjpovMZ6iQvdOdVKIoOhSSTWeeyyvmMToSKRyKkgCiORSOSu+MAHf4Lp9LNZzKb8j9+4Pr3lG7/x6/mKL/5yvvmvfAs75zb4mXf8LOI2ki7ml9/NO3/xP5DfH6/QFz07Ozs8e/EZ+v0+/d7glm+Si6LA6BfmZ3JSPPszX8Sjn/6vnvd2X/Jnv5DJZBenE2xV4r3hNz98rUPmid94D3/ze76X7/+BH+T//Dt/k2/8+q+/l2XfwOMf/WqEkqQ+I9GSxGjaNONH3/p2vuvN97WUyIuO0BlXYkE4TCIphhl5v0DqAmP6SJV1fg1c60K49tUtIp1taKopdXmAbedIo1CJRMgA4oWN/d0PxNKI1qQZ/dE29ZlHcfWcBYJ2fhXlSpT0aCXRsvO18EGgjUVrjVYNRgt6KPJUoE2CSRK07oQUpRRSSVQCmfYYX+GbKW05xdZ91MoMVUoef9VH88hjr+DxV38s5x95nMHo+c2+14kswRO8XRqRrkxIl94ah746D5C7TAvygeA8rqxx4ynt3hh3MMFXFQSH0AaZ5ciiQBedd43UGrRBGI1INCHp/pd0I1irWOfbOU7nAnbZzVHVDVVVUi3m2MUCW5b4uibYGkRApAt0NceUc7JeRn+wee0xh/VzWAQI3uPwBOvxraWua+bzOYvZnEW5oLEOhwaZoKRBC4NUmsQY+qmgn0hyo0i0XIqHURSJRF5MRGEkEoncFb/23j1+7u3/jPF4nz//v/2p667rmWf5zfc+w8c/MuBd73ovrm6fN0rvQx/6EOXgd1LN/p8HwtBs5f7vvTu1b47+/Dd+NcGC0orv+57/57rrrO3e7HcE9H1Krjl2/O0tyH76Z3+N1gbSVCACeGlRRzpN/sJXfxV/4au/6l5U+bzU1jPsDZlXNXlmSIRC9wc8+shDJ1JP5MVDCB7vLK5tcG2Fdw1SKVTWI+ttkRabmCRHyqXHBcuvQ3YXIXhsW1HPU8QsUM894EB2oksIp18YgW6RrE1CPhgxOvsIBI8yKeV+H7c4INgFjgaBRfiw9gZZ+YMoJdFakqYJRZGTpknn74EgSIk0BWlvQDbYYjDokxiFwBGc7Yxs8QgkH/MJr+PhR1/JI4+9gu2zF8h7/eese90l4i3BNeAWOFuCK8E7ELITRlSKUBnIBKlSkKZLbFkmVtz236oQ8G2NK0vagzHNpSvYq3u4RQnOgTGoPEP3+/jeAFXkqCRFJAkyTRBpgigSyFN8ZpBGE5C3jC8+fJzOBVrrWJQts0XDeFoxnc0o5zPaqsLXntAYgk9BGNAg0xqTzch7CY8tM+GF7B4vowWZV8jgaRuPa2rKqjPRLRcl5XxKVU5xthsvS5RBqgypUpRKSWRKlhT0Mk0vFwwySaYFSq3SaU7n3//IS4eAIJyS5+FpqeNuicJIJBK5K974x3+EH/+mPwmi4K/91b9/3XV2/n5++yf8Hh7Z2EepmrLyTKZ7eO+Q8uZO+e9617vY2dnmZZ/xr/npf/uX78MRvDA+8bWfxG/9ra8lTVOapjnpcm5KPyt4/Ws/iS/+Y3/6husuXnqaf/h//V2yLGXQO8to+NxvzE8rHokPLVI8T8eLFvTTgtY6UukRpIQXaG57nBR5jq0rgm0BReM9F86d6eJTI5G7JCyNMzthpMa2VefXgECZBJPmJFmOSYtuFALWCSOr97fdyIAjyIByKapJkLXCWof3Fu9dF+37ADxXu64RTZL3GGydQ2mDyXvMB1uU48u42VVCPcE2M2hLnPC4VTStkhhj0FozHPQYDnpkeQ+hEywGTI7OB2T9DbLhJvlwm2J0lqQYocxylGZ5Uj/m417L9tmH2dw+S68/xJj0hlqvO5+rVBZX4ZoJrt7HN2OCnYFvliNPGlSG0H2EGSD1EGl6SJ0hZNetcrvmuME6fN3iZgvagwnN1T3aK1dwiwWhdQitUUWOm5X4XokuBvg8QyUpPksQRY5sMoTzrBpFhNHPO1YTQtcpUtaWg2nDlf05u3tz9iczFosKW3t8m4IbQMgIQRJkhtABnZYUvQm/Y7ktKUArQWIULI1z66ZkNp4x2Z8wHc+YL+ad3w4zMl2SJ45EKkxIUDZFu4TUF2RyQC+z5EGSSoORZmmrErtGIpEXE1EYiUQid8XDj/8W3vVLl3n8lRn6yBjGdLzHu975IwyyGVvbfQrZUicpbVOTZsVNt/dzv/A2djYGGJOy8Yo/dD8O4a7RKuXH//NP851/46/zxJMfIk361M3spMu6ge/+//+DW173yMMvo7EOO53RNA0//dPvvuVtTzPBO97z9r/Mb/3c73jO2332Jz/M29/xEbz31LWDpKInXljizHQ24xNe9wm4IDi3uYNUCT//zp++q201jcPZmswYBllBmvVp24a3/uiPvqAajxMfPLUNfMnf+Wl++Tf2sXVFIlPe809O9+v1pco1nwaLtw22LXFtTfAOoVKkTlA6RSqNkOr6Rd6htZ4APKpLdJGq6yyRq0hUjw8OH/wD1TWidEJaCJTSmKxH3t9iMTlLNb5CPb2Cne7iFgfAAaIOKO3QJoAWpFlOf2NIf7RB3h8hsz4kfVQ2JCmGJL0Nst6QpBhg8h4m62OSHKENLDsnX/byj6IYbJDlPYxJr1tcrwUR343NEDoBaiWK2PIK7eIytrqCb2cQmu7zYqUQOgU9RCSb6HQHne2g01EnkKjk/2PvvuPbqs4Gjv/OuUPLM7azdwIBQiDsDWUVKLtAGS2jpe1LC7SsF+h6C5QyWkYHBQqFltGWQgelQNl7hzCyyd6J95Is6Y5z3j9kmwwnsR3bspPz/Xz0sSXd8UiW5HsfnfM8nU6O6MBHpb1cYqSxEb+xCb+pGZVKo8MQaVnorIfOhpD10ZkAHYujotFc3ZGMh/I9pA7aO9fI1vojehNdXHKvV0EQ+LSkfRqa01TWpqiqTdHQmKUl7RMGEhVG0aGVS9YhyA0Zkdi2TzSZbv8bt42ykShUEJBOtVBX3UDl2loqqxpoaEiS9VqwrAzFiQxuURrH8kk44GAhAwvh29hEsbxCUCmEDBEOuZoqtkC3JuRNcsQwtg0mMWIYRrcUFBSRiMVZuyK90X1VNVm0jLHTnmfy4jMPUFYocWL2JpMiADPmvs2lJ0xkRPkSRBgC1/Zi9LB8+XIWVS7l8H0O7fK6L70xjbWrV/LL23+LH/gctM/uvRBh77v5hjv4+c0/JFAhnufhugNvOs2dL5/C0qY4/+PM40uH77TJ5e6797/suNvuCEeRFQ7FMiQpuz/S58MPp3HG2eeisbG0IO35RGOSvfY7kOnvv9Pl7a1cOI899z+QgmgMtOKBB+9m4vgduh1fTxp64u2UVNtkExGcomJkQjJsUBwRKSDp57p12N0osGj0Pq0VhCFBkCUIsoRBBq3BkjZ2a3IkN1JEbLqramtdVSFka3JEtv6ea8eqlGqfTqO1yk3p6Mc+r53iIqSFdFzcSIxoYQmZ4jIyTRVkm2rINlXhNaxFYxOGoK0oyokSLSyhsGQQhaVlRAtLcRIlWNEinHgRTrQQNxrDisSxHTdXV8R2cs9Z+1QlKBk0GNuNtNYd2XjUgVYBOvDRfhblpdFBGuUnCbN1eOlKMsk1ZNK1hH4KoX2krXOdVmwHYScQTilWrAk3yBJRCofWYrrrtAzeHJX1CdMtBM0pwuYWwlQLKuMRej5C597zlga0zJWXCTXaD7G8ABnxEJ6PDAMkGmwLnLYONlauve8mkjNKhYRhSDbr0dLSQiqVJJVM0ZLO4nkBYaBRykIpgWqdRqkBITR+6G80aklrjef5JJMpqqsaWba8iuUralhT2UhjsgWET3Ghh+tkkTJDPOqTiCocDWRBBxqyNioTJ/BTZJWXa88rrdxUnYiETtRPM4zepHNdtfuF/hJHd5l3s2EY3RKLxVm2eCXRaGyj+/baMcGfX80QfecezjiygA/frWbJqs1PdVi7tpIf3b2Ih3+2B3orTlg7Q2vN1358I5XNn/DZvz7o8vpnnXY8P/npzXw2bx533XUHlt1/Twov+ubXuef+Bzs8CJVS8pMf3dLp4dW9rXnVyzQvfZyYkytoGN3lJzjRss2us9YbxF6TyzebFGkzf8anPRUqZ557HpYFSjvo0KMp2YwtbSxbcsSRR/LKyy93eZsfvPUGwDq1X/qHNU9dwWu/+BHe/Hc56Ff/paAwmu+QjE7RoBWhClC+R+hl0UqhhUQ6UWwngmXZiPaT5U1nRoTIfUYIaSGlg2ydmpHrgNLanUO37nOAEELkRsC4FpblYEeiRKJxYgUlZIvLyTaWk4kVEBEWrrAoCDUUDiJSMpREaTnxolKiBSU4sSKcaAIrksB23NwIHMvKTRsVAtFaW2NdbnTdmi4bUArt+wTpFEGyniBZg0rXo7INKL8OP1NNS7qGlnQj2SADUuFEBLbb+tlhRxFOI1Y2jR/kCsgibaR0kY5ECGeLn/lhJkuYShMkWwhTaVTWQ/sBQoVoNOgAHcpcHZK2TjitBVulHyACH6F8LAt0zEEkIojABUdtcjptblpWQBikUX4z2m9EBg04OkXU8rEtRSByXWpCFaJVrpVurq107v+Zu05SLtd1RpFuyVJX28TKVTUsWlbNkuV1VNalyPgB0WiIE/XB8nEjAbFYQCIWYqHB1gQiIFSgUyn8xgzKD9Aq93wKy0YIGytioaUZNWIY24L+dfRlGMaAIYToMCkCcOnNH/PGFw/Eiu/Cw88tpqbB4ztnbL7q/oxpNQDMe/ECgvSiHo93XZcfcSbO2BLGFoTdWv+M08/Fjjj89jd3sba6tt8OIb/++utpqEnz7TPP5f7HH93kcv3hgO6TJ7+I5SZIt3hEIlFCadE850xGTb2W8bsetcn1/vbL4/swyhytNSWlRSQbMzi2AlWA7UrSvoelHHRTS7e221MJkZ5OdAW+T8Eek7H3qsByJJ7v49ibOKkz+o1cx9UAFfqEys99wy5tHMvFcWNYbrT15K4zf8fW6RDSQtoOlhPBCrzcN+e0Tf/QA+7bwrbHLq1ctxjLsnCcCG4kTtSNECUgk2kiprKE0sYuG407eCyR4nLcWAFuLIHlxnIjPyyrdUrS56MyNvXcbiopkksyKVTgE6QaydSsJlO5lKBhNTpTjQgbCMImPL8Zz0/j6RBla4K4jR0TWI6NsF2wW5CeT6gkyFzdEWnHcaSDEFv+m/vZDEE6V3xV+QE6DMn9hSVatxYcb02EqCAAL5sbCQIIFSJxQWrCtI2diUI2hhtEILTRIpfE2PBvoMMsyksRZpsQfj1OWEtCNBBG0sRESOBolIJAKcIgROnc6y03U0ciBDiWaH8elVL4vk8y2UJ1TRMrVzewbHUjK6qTNDR7KKkodTQQ4tghrhviugrb1dhSoy2FlLp9RKWf9AgaQpSS4OTaFFtuBGFbSDnwRlsahrExkxgxDKPHWZbLzy4rRauVeH4MFVTw3zn7dmpdrcF2tq72w+a3r2msrmLMxNW8+dzije73fR/H2fzoloohQ3n1pVcYO3YMN/zsRk447qDeCner1K9YznnnnUZBrP+OaGkjpU3N6lXEigtZu6ISP/DZcZcpLF7tMX7Xrm3rjWkrGT+ygJHDSnolViEEO++4Mw31jSxesgrtKCKRQspKbLK+oiXT9/VmtNbU19cy/8M5TN5v9/Zh5WVlFVu97X2PupaxiSYOdz/hiVs/4LNsIbMopLnJ4ysn7M6vbvxmDzwCoye1F0zVOlccVYO0HVwpsZwokVghjhv/vIXsFk6Uc6MrJMKysZ0oTiSO0grQuRESG9TIGGhJs/ZEhpS56R6OjXRbu6tEIzjxOLhx7LLBRCqG4xRWYEdiCNteLxnS2ce9+eU0OgwJslkyDXWk1qwis2YhunkNlmpEiDSh9NCoXFcgB/yMRZCwkVELbbtIO4PMKkLlglWM7Q7CiZRgOwm0zo102Jwg6xF4PqFSKHLtdrGtXDFo1Zp8aG3Bq1sTIoFWSBXmap6EoEOBDF3CwEP7WfA8bFsidQTLttq7/Xy+zya8lnrCVC1kaomoBorsJtyoTxgRKGGhFIRhSBCEhEqBprVtrkSikXz+JUUYKrIZn+amFqprG1lb3UhlTTMNTVla/BDbAY1CyhApQxwrxJIKKRRIEDbICMiERmQDVMojbFGEoYNwC7ETJdiJQqxIFG3p3DoD7HVvbBtMV5qeYxIjhmH0inEH3c6St68gFisl61Xykxt+36n1dv7in5j/8qW9Ftcdt9zCkVefw7N/uJOa5vXro1ixCOMnjGPBrHmb3cbrb7/NjOmvk2xu4dhjjyEa6Z8dXX71hwfyHUKnZJKVrF0xn0T5EOxonIJSB21J5n82l0F7bn6k0bo8P+R7P3+FMAjZZ5dyvn3O3r0W86N/+hM1NVV85+LLyXohbtRChQGepykNi3ptv5uSayXqgtvMqlUriBcU0tzUxBv/vJ9Tv/XDrdp22cSJ/OL8abiRKYRCkEonSb2+luzMemKffAyYxEh/lZviIpCWg+3GkFJiu3HcWBHSjqxX92JLhMhNOXHcKGGYQJCr82A7UaTlsOkiJf1f+4gXpVDKR/tpdJBBhj62tCASx4oVYbcXVY0jHLe9nkqPnhC3zmrSCJQS+NmATFOasLYJma3HogWsACUV2haoqCbISrRno6M2uA5YaayIJlRxpFNHJN5ENJEijBTkCqcKAAl0XHPEVwGBDFGOREVsdCyCRqE9D0KVO/WRAi0ttG3naojYFqEDyhZgKbQVImRIqH104KG9NKElsNEobWPbuVEWba2Q/Uw9fks1QaYWnW3ACZuIWxncmEBYLtJy0EISBiGe7xEGIYhcvRwhBUIrVOi1d2LSYYiXzdLc3EJDfZKGhiTJVIZ0JouvdK6rjFJIAiQBEKIJcq8F1fo3sATCASIa7XgEgU+QdRBuNW7pYCKlg3ALi5CuDWyhM5phGOu57rrruP7669e7bciQIaxduzZPEZnEiGEYvSQ+aDcmn/gSy967mp33f6JL634y/Q12PLJ34qp+/k0+GLGQGCVYlt3+7ebDD9xPTEjWrq3Z4jZmfPgmM+ct5ZLvfoPvX/JtamrX9E6w24lowRAsGUUKWDZvFmVDxlC1rIpho8ZRNqi009u5/KZXQdgIR3LUIRN6MeKc8vLBPPG3P7dfVyoExHrfgvYthaipYU3oIllFXeUKVq1cstVbfenB77Jm4R/QzR8QkZKCeAT/pFKyx49k/N53b3kDRp60fpNuu1gqhrAcpLRx3Fiu8Kdtt5+UdnqLUiLtCJFoEZZ0UTpESgfbdlunh/T/0WmbpFWuvkeQ66iiAw+ERMYKcWwXmSjGSZRit09B2vJIm64SQqCxELaDFYnlRiUUD8YuriNsSRF6WcKMB2GWUHsEToifBT+UhKGFykqIOGD72L6DlkmceAov00zgNaO8OEpqUC5SugjLQWOxYYIklKAcizAWQRVGc3VqbIH2HYRSiFzFU4SVKyqrbBtsG21LtGOhbAtciXRAW+SKsIYByssQonFaC6cKEWkfaRNmkyg/hQqyIHIjkSwniiVsLDeO48ZB2gRBSDabxfcDhMhNQZRSgAoI/QwASmuCIMRLZ0k1p0g2pmhuaiGdzhD4Xq78jtU6FSgMUKFH6PuoICR02gr0tjYRssglR2yNIiDINiOb6/Ga6vFTKYJsFjsag/5dc9gw+qXJkyfz0ksvtV+3rPy+kUxixDCMXjVm/190eZ2vXN1zRTI39MmqGkYPs7n0l3dyVTTefjA45S9PMTkeZ5bY8jSI6prVjBg+Ft8PqK5Z3Wuxbk8qV35GtLmQ8sETSDcsx7EkKxcu5KCzO9eZJdO4kCv2vwuFJPRaGFHxeC9HvLFNFRXsK8XFJUw+7jSee+BqVPFUapfNpqW5qUe2PWziNzEjQwaO3ImdbD+5FNICrXN1NGwHadmbLv65mW1KaYNNa8KlrYinbC3iukHL3wEpdzYsLIl0IlAwCOHmamnJSAKr9bropfe61rq1ta2DkygkVjGUMPARlsSOuWQro/i1y/GTCt/z8cMQPwwJfEGYlShyf2dpWyBDVKjQKoQwg/bqCTMaqeNoJ4G2CpBOAcKKIuT6xXeVJVCxCDoRIfRjhDokdAT4ASiF0BqBAJmrzYIt0VLmRrBIgbZsRMTGdnOjSXLDM8h1MPKzrSPcLLSt2x+3Vj5gIe0Eto4Q6ASSAC0klhvHjhXmiv4GAYHIoKwAANuxsSwrN1pEtNV30ujWBEomlSaVaiGTTuN7Xq4VMaADReiFBFkfP+vjeT6er4m6EilELtmiRe5pkQpha7QdoFUWP5PETzYRtKRQnocKQixLgzWQX/vGQKUQuSLL/YDq4lQa27YZOnRoL0XTdSYxYhjGduPLp57CoV87hR/837XtJ7Ft7f3GJSI8cNBuPDx0NLW1lZSVDdnstlatXtrb4W5Xzvm5zz9uHs6qpQspKBpKUzrDXqc92enRF40r/kuIRRD44LXw3t0HcdgVM3s56v6nIFHAF865jn8/cjcZr4Vk3ap8h2TkyeeJDIm02qZ8fN5yt7tyCRULqR1ycw4+r68xUJMibSM1sDQSFyEk2opgRQtzI0mERNguwokibSd3vRcea/s2LQc7liBSCsKycOJR3KIEqcIYybgmW+XhJzNkQ4/QhcDK1QJBy1zSQjpYTgzXTeA6NrbIIoIaVLoJ33cRTjFWpBxLK2y3bWjEOgVjpQDHQsVdlB8l0B7KAh0EoDRCt874yVU+RVsy91MIlABtS6QrsSMOMuJiR1wsOzfVRKtcZxmlVGur59aaNMJCOjFsKw62wJMe6AxaabDiCCuRa0dMmCswq3yEIJeEsSVSBBDa6+xD4Wc8MukMmXSGrOcReh4qDNBowkARZAMyGZ+WFo90OiCb0XiOhSBXiFcIQYhGSw1ObuSIlgHKT+OnkwQtKYJshjAMka09Uwfqe8Aw8mHBggUMHz6cSCTCfvvtx0033cT48ePzFo9JjPSB9lZ2dDwP1XyIGkbXaa1ZuGAhO+zYuREFAC899xyP/e1xYo7N/3zrW/z+/vvb76t49t9ERECyan9WpJopY/OJkW3V9//nK/z6930/2gLgtB/kRt8sm/1Pxkz+cpfWHTz5EhrkoSye9SHDvIc59PIZvRHigFBePpgLL7+O6upaKio23+7Y2Ha1HVvkhiZbHd7X3W1+vv62NX+gbdSLtmyEE2ntOJYbfZNrfyJ7LSmyXhyWRIoIbkIgHQsr6iAiFqHMkNa1qLCKwKrDz7YQCI1qK/6JQAo3lxSJFBKJJohEHFwriwxrIRsSejbYJYRhgI0N0sWWTm5aTOtUKC1Ea8FVBxVxCH2HQCsIRO7kX+f2lcun5EaDaJFr3RwKwJLYjgOOi+1EsF0X23EItcoVTUW113Vpf8zCQloxLNtGCQsZZkAKBAqEg5Y2QroIGaJFCDK3tpROa1IGwMlNk2nrShMqMl5INuvj+yFhmGsJrLVGCY3nh2QyIelMQDrtk8lqYpEQKaz2hI1SCi00wgHhaLAUge/hZVJk00m8dAsR30NFo1gM4KlkxoDVmpPrF9riaGpaf7RqJBIhEomsd9t+++3Hww8/zI477khlZSU33ngjBx54ILNnz6asLD/HLuYd3MvaPpxzVbQDgiDI9WBvvd6WLTcMY/NmzfkU3/fJZDLsNmUPiovLuP7mWzu17owZrSfJMs7BU/YkQPKTn/5f+/1aa6IignYc9J/h/fdf7o2H0O99/O9fMX1m/qcGdTUpArkTtUm77M4JX7mQvb72ukk4g0mKGOuN5Bjoozp6W+55yY1cENKB1hEiwomA7YD8fLpQ78eRG40hHBsrEsWKJZDxOCIRR8Ui+FEbz5FkLYEvIFAQ+JogABVaaJ1ryyuFwNYelk4igwbw69F+LSpbi8rWofxmdJBBqxCtPz8e1a2zSEKpCC1B6EgCR+DZAt8WBI7MXWyL0JGEjoVv5e7zLUkgJYGQaEsiHCvXDtl22ov9dvg6FCLXFUg4IC0UVq7PjAYtJELnaqEoaE3gtG5jw22JXMIkRBBqCDUECJRuHaES5n6GKiQMNZ7fegkUnh/i+37rJUsQeCgVolAoodGuQNkKjywtmRTNqSZSLUky2Qxhaycwc0xvGDBq1CiKi4vbLzfffPNGyxx33HGcdtppTJkyhaOOOopnnnkGgIceeqivw21nRoz0orakiOd5eF6uUrZl5VqUtX1wWpaF67pdLoBmGNuTKXsdysL5n+KlmgEHN+pwwjn/w+HHda4F8G677caVJ32Nsy/+IcuXv8MxFYUMGzacVCoFQOWq5WSUh60t/E+TFI7b+hanA82nb/ybI8/9AXsdfDI33Xo9P7zmp/kOqV/Zd9JYmpqaSKd9UlkfpCSSiLGqqrbH9/XY43cwcvy+SNcmDEMO2aN/toM2jG1NvzsO0xqtAsIwQ+AnyWSbyXoteL5HNggJQtBKotG55IESaBSW7eNH0gSxRvyoIHBiuEiwBQILTRrCNDpIo5SHVgFau7mpKa3Hp7l0gCYkJBCaQGhCqRAKhNBIIRGirV0uBErmEhFKtX5tLAjJTa/5/Hltm2qy8fQrrULCUBEgCH2J8lunvWhNrtZrLglirbOtXBYkzGVx1mnV254wsSykYyMcO1d3xQItw9yoEqUJlUYpQRBKQi0IlSYMNUEIoHL1U5C524TCtzQZW5HWHjrbTLapDt3cgEwnkfE40nLap+AYxvZsxYoVFBV93qFvw9EiHUkkEkyZMoUFCxb0ZmibZUaM9CKtNZ7nkUwmqampoaqqmrq6OhoaGqitraW2tpZkMtmeNDEMo2OfTnuNI/fakbEjhzJu7GC+cPb/kaxfwaO//V2nt7HPSfvDolfZZ/IheH4Td975a+LxOABDhg/FdsB1I4jMXsz8aG5vPZR+65UXHmbap+8TdwSLlmZ49C9/yndI/Y7jOLmuHEKglSKbzXLxJf/TqXU7+xmvtSbiFFG/7FMc225/jRqGsX1oH7WhfFSQxs80kG5aQ6phBS0Ny0k3VeJlmtEq1xHIsqNYVhQpXNAOOtD46QxeqpFMUzWZpiqyqXr8bAs69BFaIXSI1gE6bEs+hOuNxc/VoMmNftACQq0IUARosjrAUyG+DgmFzjW5FRAIQTZUZIPcxQ8DAhUSqtwI6dD3UEHQOkVFtxcIhlwiQ4UhKvAJsy14XguB7xGGYW75dZ6fUOvcFCetQQVo5aF1gNAqVw+mtWOOZVm4kQixRIzCgijxApdIzM4N/pEhwlYgdetjgFDnkiOeFniBJu1rWrKKZCYg6YU0X8W6HAABAABJREFUBZpmrakXmhp81vgpVqbqWdVUQ01THcl0Ei/IorQZCW70NYHuJ5e2WkVFRUXrXTqTGMlms8ydO5dhw4b18vO1aWbESC+SUhKLxYjFYlRUbH/fQBtGT5FS8vRr0/jj7T/jhTffZfobf0Tjkc1mO72Ns755CZ988C7zXn+CRcvXcOWVl7ffl21cw/Q3/sOM96fBRws489yze+Nh9Gs6qGD0yJ3497//yp5T9+HovSbjBx6O7eY7tH5BCIHWFrFYjGzWQ9oOLVmf1198tVPrv/Pa+wgryoGHTt3scu+98yLFRcU0Ndey8NP32WFPM1rEMLY3WoeoIIufbiDdvJqW+sUkaz+jpWExfqYKS2SJxF2ksNs7yiilCAOBFhLbtpDaRwcZlOcQeDa+L7AsibRtEApUiNY+Wvm5xAif18KzHRvpWa2jmTVCgNKKQPmoUOWKkurc2BPLthEKAqVzU1CCEEtKAttG+QGB5+M5WSwVAmBJC+GsP2KkbQpKEGTxA/B9iecFqDBEWnZ7JRKtNaJ1Wa1CtAhQSkIY5BIlIgABUggc1yZeEKV0UCHlQ0ooa2qi2UsRiDQtGQ+BRtsaJTU+mpYAkoFAepp02DpqRuvcVJwQsllFKqtpDqFJh6RVBjfdQKqhEtlYiltYiBuN5VpXr5PwMQyjY1dddRUnnngio0ePpqqqihtvvJGmpibOP//8vMVkEiOGYfRrWmvuuOM23nvxYYYPnsLh++zKxzNmYTkWvlBb3sA6brnvzx3e/odH/8Mll17KxKmH0/C//8vgocN7IvQBZdigEl576Hben7mcjz6ZBsCy5YsYM3pCniPrH4RwiMUk9fX12LZNqAWJmEssVvB5V4XNiMbiTN1n8hb3E4mCZZfguC71qRSr16zglZYURfEE0o1h2za7jer/f5N0VSVPH3cgp34wH9vatgp0GkZvyY00UOjQI/CSZNNVpBuXkmpYSLpxEb63BstKESuU2DKO7doI20IgUEqgQoUKASGxHRs36uJEc8mNMAzxgwCLAC1CZOhDGOQSI+vUGBFC4DgRIpGQIPBbp4ADIkTpAKVDtNLYQuba2CoAizBQaD8AP0RbAuVLQj9N4MXwLJFLGNhWLjEiZGs9kXU74eSmrajAJ/Rba/GFGilAhbnEiwxC/FARBCFBqBA6wNIQaAFoVOshgZQSx7EpKIhTMbiEES1JmrNNpGkisJOQ1LnH5mpCR5NC0eAr7LQmqcGyQOncVBtfgR9qMj6kM5qkD0kNWR3geCmyyTqiDdUUFg+iKFFEPBLF7oNaNIYx0K1cuZKzzz6bmpoaKioq2H///XnvvfcYM2ZM3mIyiRHD2M41NjXy5LOvEC0axJlfOizf4WxECMHfHvktI8uKmbn4U6busz9eKLGsgGg02iP7uPiSSwBwXZeCW25B/+VJ/nfeR/zyF7/oke0PBGdffQuP3/ETzvzyl5g1/W123eugfpUUueCSLxBWZ/jyKQdz/uOzSQ8eyei992bRtzo3lWVrKSTRqENhUQlO4NHSkuHhJ/7BYV/o3Htmr/1369RyiaLBra1DBcUSsn4zixZVwrARxLFyrSv7uU8eOI3Fv5+Ba0fwG+qxy8rzHZJh9HufT6FRqNAj8JrJtlSTSa0mm1pBGFRjyRR2AdhWFMd1cJzWmhbSAiVys0uUAC1zNUAcC9t1ELZEofHDAF97oDJYIovtZLFaW+e2719r3GgMjSIMfQLPJWMLLKmwtJcbZRLqXJeYUIPQCG0hQoVUATLwkVoiQwGhT+ClcRwHJSWWsLEsC9u2sSx7veSB5RQgfQ2WRBGgNPhKI3wQvkZlQoQE3w/IZHOdZgSKINTYVq4zThi0Jak1ti1JJKKUlRXR4g8iFTSR1PVkrEZ0fZJ0NkSIkCCqSQpFlafJJBWuR+sImdbRIgp8JfADTdYXpH1BVkOAwgmyWJlm6lqaaEg10uKl8cMAVyukqVZg9JH1+zvlV1fieOyxx3otju4yiRHD2I6lM1kee+oFvv+bJxhaLPjO/91G3Yf/yXdYG6lp1lTXr2HEsJH87Bc/46dXXs3Vd9xBYbxn/hWse3AWi8X474hSUi9V9si2B5KqlcuJH34EzUtmwF79awrHDy+/lWDhs8QLLIYMHsyiRXNZvGYubEVipHL1Sn73i58RBAHZrEdTYyOJomJ+9YdHNlr2g7nzmDp+DNF4AWurklhadDop0iVWBMd10bZFzHFADGXYkEloLbBsTSj698H2orfPITIhiq8ktqX50+EH8p0Z8/MdlmH0e7kpJQqtA8LAw/eb8TP1+JlaQr8BKdJEYgopLBzHziUXbAvLcpCtozC0Ekis3ImSbm29K3NJES/wEEFutAiOhSMy4HoEoY/dXjBVARaO6yJUSBjx8CMOriPJygBfZhFkkWjQFmgHoSIIJJbWuVofItfu1pEaS4AlJVLmRow4joPjuti23Rrz5/973WgJChtHZ7CUh85k8FWuwGxGaaxsgBCqtTtdiO/nqhrYFthW6/9xbbc+lxIpIeLaFBfFyAbFJP1BJP1BtFBP6CSRqSyhCtBSkbGgPtSkMmD7bYkRgdKgNAQaQpXr+OMpCIUGYYEQeCokE2TxAi9XS0WpfnOSahhG15jEiGFsx155axq3PPo8u06YRHmZzWuf9M+io4119biOxbJVawiyIWuWLqWMkBdefL9X9nfcMcdw7rlf4e6t2EZnplf0N5fc8RBaa4btdki+Q9lIoqiEWHkJxUPHkXnhXnR1Fbtceka3tzdz+gc8+8Rf8H2fvfbch9tuv42999mHuupqzj3pizzy1AsbrfPJ4mU0NDRQWlrKvvt0riNSV+04bidmLZxD1I6hpcKxFEpppJQESlAWT/TKfntKJpMl2aJpzmZxIzbFUXdAvhcMo6/lRmzk6mco7aP8DIGfIgxSQBYhA4QAy5IgQQuN0mGuE40OEcJCCCvXS0YLlNKEWqCDXLpDk2t3qwkQoUUoMsiYhxOGaNpO5nPvU8t2UK7CCaK4kQiOY+M6El+qXF0Pkduf1CCQSCRCKITUaBtsxyYacYlGI0SiUSKxeO5nJIYdiWI7n3djbPtssKOFRKRLKLJkdBormybIpGn2AoK0RmsfpTWeH+D5PmGgEGgsCbatsaTMjRxpJVoLsEYiEYoKEwzOFtOcLSMZNBCINKLRJ+2RGxUjFWkEXiCQ6vPPqlyCKfdTaZ0rMClBCollOUTdKNFonKgTxXEcpJUrzm0YxsBkEiN54vs+juPkOwxjO/fqO2+RXvA62alfpHDwAZx87I75DqlDf/7rE5x59umUuh7xkhLu+fPD3H/LDYwcNbrX9llT1bhV6+993FlMf+5vPRRN3xFCIKyO/zXk8wR3RMUkqJgEwIolp2z19p7613/IeNDSnOTtN98kCAVlxSW0JJM0NG36b19SUtKrHQeEEHz89O1YTgWReAI3WoydiOLYhdgyzujjTum1fX/h2CuZO38Fl/7PKfz4mnO6tQ2tQ4T0KCh2CTOQpoUVdWsYXbb91e0xjK5qK0KKClHKRymPMPRQoY9SYXs3GSklYdhWvDQ38iJX8DPXClfrXEHTsDVBorRAIwEbZARLRdBOiBvo3EAR3fa5nvspZW66i+M4uG6UaCRO4MYIIxFQWbTItcqVAiyhEFggJdoB4Tg40TjxgkLiBcXEC0qIJgpw3BiW62A7LpZlt3a++ZztJkC6RESUmIrS4kWw0zY6nSHteWSyIV4Q4PuKINCEYe7zUkqJY1u4rkXc/fy4uj3hIi0S0ShlxaWkvTQZP01AiHSgMdVIxs8ShrkCtFpo2vIibf/qBKL1WZUIBJawsS2biBOlMJagoqSCwaXllCZKiEdiOJZtkiNGn8q93/vHa66/xNFd/XtM7jZKCIHrujzwwAMAHHzwwebbNCMv3luWZsdjv0XJyB2oqW9gQXV9vkPq0LHHHkfUsomJGD/eYyyhtnlz4eo+2/+5F13WpeUnf/FsXLWGU//n8i0vPIBcdNIhHDllB2697qf5DmWrpFtSRHQSW7UQi8Xwgix77jGZ5uZ61tTV8dxb0ygpKclbfOdd/iD1lYupqZzB2pWfsHzuOyyZ/QKH91JSZMwu32Di5PNYumQVERS//u0fGTH+FP748PNd3lY06pKIRdj/jsPwVcBvxo3g1B9cgBe29ELkhrFtESJ3Gq5FW4Ijl+wIlSYMQnzfI5vJkEmnSLckc5dUkpZkkmQqSSrVTDLZRDLZQHNzI6nmRlKp3HKZdAuZTBov6+P7CqUEWjgg7NbRG593UmlLONiWixtJEI0WEYuVEI+WkIgWEYkkiDgxXDuCY0VxbRfXjRONF5IoLKGgeBAFxWUkikqJFxYRjRXgxqI4bnS9pMi6x77ScrCd3KiSeDxBYVEhJcVFFBUmiMfjrSNXIti2i+1EcNw4jhMn4sZwI3Gi0QTxgsKNnk8pJa4bpTBRyOBB5YwePJLxw0YxduhoRlYMZ0hJOYOKBlFSWEphvJhEtIiCaBGJaDGFsRKK4iWUFAyirKiMwSWDGVo2lFEVIxg3dDQTR0xkwrBxjB48ioqSMgqiBdi2096VxjCMgcWMGOmGtoz+htnurvrmN7/JhRdeyNtvv92+XZMgMfpSvHQYdsQik87gK4hH+u8opsqaRr599BR+9sZsDt5jf3528y/7bN/3/erWTi879cQLkZZCO8Pws14vRgXNNWsoLO+7fu/vDClm6ZASFv/uV1xz3fV9tt+e9p/HHqKpsZGWdJpsNks61cKKNWt5edosTth/HzJZj3Sm862ge8MlP3sC38tyzy3f4uvfv4PC4t4pYLpkySpi8VICqYgoiQ6yRChFp5u48Y5/cO45R2Hbne8qk9UFLFwxjzv/9QlNuw4mFoviK49DLj2e9+/uXGtjw9ge5WqMCBACKRykHcWy4wgZByIoJQi8EK18UMFG6+YSKYDOFWEN0Wgl0UIgWhMgwhbYrsASUaSVwHIKW/fh5jrFrHsIKiwsx8EJo8QTxWgVYCFx7Ri+l0aFIVoLpLBzF+kgLQfHjROJx4kUlhArLMKNJrAjLtJy1jtu3vB4tz0ZY9vEYhYagRASx3FIJDKk0j6ZjI/X2pFGK0CAY0lsRxKLOBTEN24vL4TAlhYxN0ppQSkgcGybRDRBbWFprmhqtoWsnyUIc1132tazhIUlJZaVGyXi2i5xN0YimqAglqAoXkxpQTGlhSUUxYtbO9LY600RMgxj4DCJkS5qS4qEYditxMiGQ7DXy5ZL2atDtA1jQ279Ipprq5h00PGMHFTcpROgfLjvxZl52W8sGtniMnf95TnemP4pOgyRlk0yk+aKU77UazE11lVx3smHcvFJu+LHx3L8pXf22r7a/Ps7F/Ob3z3IAwfvxJSRQ5m5cm2v77M3HHniabz/9pt42SylpWXU16Z47YNZfPtLx/L4a69y+GFH8OLLG9cY6WuOG+F7//dwr+5j3LgRzPvwdqZ+9yj82iwNy5P4NUVEIyOwYkWM2+NrrJj5105vb/IhD3LVf4/D2qGQhK9Q2sYClMrw63/fw/dP/k7vPRjDGMByx3+5qTHSimC7hTixQTjRcuxoNX62Ge1nUSpAh6B18PnUG9YZwt5WdFWDws4VCZUaadvYIoZ0SrGjg3FiQ4jEBuFECrHsCEK2TcXJkVKCdHCcKMQBLbCtCE6kkMBLo/zc/oWWCMtuTYo4WG4CN5ZLjjixOJbrYNnu+q15NyFXF0QCmkQ8gm1JYlGX0uIY2WxA1g8JAtXaRSc33UVIgWNJHNci6m58WtO2P8dySEQTSCmJOC6FsULKi8poTje3J0a8IECpXEUWKSSWtHJddKSFYzu4doSYGyUejZGIJIhHY8QjcWLRKBE7gmPlugQZRl9SrZf+oL/E0V0mMdINSqlu1wjZUjJl7dq1DB06tLuhGUaXHLD3rjz0yJ9Y9PBNeNkW4rFCfnj+9HyHNSDNmvcJbpjm4D134r2ZM8Cx+copx/fa/m745rFcfvZhPPXcO1Q2z+yTxMiSFasJdJzDq0NWu/Fe319vKasYAq0j9NKZFv76wst85/QTufuJJ2la/C7lOx+2yYP3tm8CcwfP2w7XcnEqBMVDEgRa44p6LF2P6kZH7P/e9F8ArrrnKj5ZvpSvHHIS3/7SeT0csWFsW3IjRjRCSqTtYkeKcBODiRU1E/gtrSM0LHxqCUmjwyxKB6BbW+1qyBVvhVyRVQstbITIjTyxIiW4iQpihSMpKB1LQekYYgVDcKOFSMtFCGvjWCyJRaT9c892IrjRAkIviw4ClNIgWlsD2za242I7DlYkih1xsZzcKJLOJEXW3bdl5Y6XLekSiWiCwEEpRRgqQqVRrf1JRW6ADZbMjTaxrI6PsdtrkWAjowlcyyYeiVNaWEzay5D1PTzfJwhzBV7RGiEFUuSSIpaVuziWQ8R2cJwIrp37PTeaxMFqLSbb2cdpGEb/YxIjXdQ2WsTzPOLxrp0Y1NXVbXGZYcOGmVEjRp/5xhkn8uxr79LSWEsmVU+2xdQB6K57b7iWM75xIa9/WoOWNrbbu98avT1rNXuN9Dlh3+G8NWMVv7zqfP73tod6dZ8/vvpnVNYlGbTjCMKwd6cJ9bZJk6cyd84Mqqsb0Drk7r//m5gdZ9CEg1Fyywe2P/q/n/DzG37WR9H2vstOOJ97n/89UiqiFKCVIrTEel0euuq279zWgxEaxvYhN2LExXUL0bEh6JLc1A4po1hOEdmWavxMI6GfIgyyqDBAq7bRI7kuMbliqC7CimA5CexICW68nGjhMOJFIygoGUWiaBjRgjKcSBzLdjr8zBO5rANSRFqnrESwg2guSROGoFTbsA2ElFi2g7QsLNtGSBthdS9Z0JaIEUJjaY1j56YJfT46BmhviitaEySCze2iLTkitMByYzhOhHgkShiG+CokDMNcoVpUa9JFINept2IJ2ZqskUhp536K9TvrmISIYQxsJjHSDUopPK/rJwVlZWWdWq4tU28YPamjGjbDygdhRRIkSiDiRvBjqfwEt4144sEH8LwMl/3oRu7+5Y29uq+GrMXvX63h6Z8fwtiEw51vfNKr+wNYtKaOopjNzqXlPPLOjF7f34YmHngodUtqsLVL1dpPtmpb377yWv7+yINcde0POeHYYygdORjXifPi3x5nac3GSewND3xv/vnPufWmX7L37vvw3vQ3tyqW/uCcY87kDy8+hNIKDw/bsXEsh99fYpIbhtFX2j5fpLTAjuDGSnI1RywX2y3GSQzGS1aTzdQTZJOtyREPFXpo3VZ3w0JIB2m1jRQpxo2VEk2UEysYTKxwMJFEGdFYEbbblhTZuBhq+6gRIRCWQEgXLRTSsdFKgfp8Go8QAqRAtBZyzf2+9XU21t1G21Sjzjx/W9qe1hqJxJYW2tZEtEaz7rQkvd5nvqC1O00HSRCTDDHyzXSl6TkmMdINQgjCMOzSOtvasGtjINF8+45/85OvHcaowaUb3au0QlgOVjTKqV86Mg/xbVtcN9rrSZF9Jg7jwgPjZDI2yxcupTBRwO2Pv9+r+wSoam7u9X1sypfO/Rrjxg9hyLAyRMJh3wOP5YFf3MCUg/ft9jZPP/cbnH7uN9a/8Xf3drisEA4QUFFRTnVVDWXRQgYVFeHVV3PcQUfz7FsvDPgD5Jd/+QxHXXMClgiwpYsbiTBx+Lh8h2UY2yGBtBxA4AoLaUdw3BIiiSF4xY34mWa8bJIwSLeOGvHR7XU3ZC4xYkewnQROpBAnWkgkVowbLcaJtLbOdSLI1kKh0PEJ/rpJCSEEwrYAq722iWj9Dk+LzxMIrJN86NFnpAc/X9d9zLkkSX7iMAyjfzGJkW5oG1bXFTfccEOXlp89ezaTJ0/u0jqG0ZFLf/0MlbV13PPk69z07VM2un/SyMHMWbAQwpBrLvp63wdodNm0hWsAmHPv+ey45+68/fxTjI50oxhEHwqCAMuyePXlf1NWPprdp+7ZpfWffvhBpHC5+MeXk5y2gAOHBvzt+ivZ9fnXW4sG9q5/PftPGhtbKC0o4cffvxSRTWNLi1Q6xYvvdL/byjU/vATHcbEsC6Fh1OjRXPj1S3sw8s4TQvDyrc9w+P8ey/O3/iMvMRiG0Xby3ZockbK1lW0CN1ZCEGQI/QyhnyUM/dakSJAbMdKaGEFYWLaLZblYThS79SJtF8tyENLebIeYjuP5fCrLhs0CBOuPMOnMNvuLgRKnYWySFrlLf9Bf4ugm02i7i9oLOHWx8OoTTzzRpeWvvfbaLi1vGB256LYnWFu9lkxLPR/NntfhMg/c/AOChioevbNryTsj/8aNG4O158Uc9pPX8h3KZrWkk5x83GH87dEHWL58Je+//TqrVi7t0jZE67+r3914J5dcehLSSlCQ9PnunlN7PuAO/PtPf2WHQSVULZ5LDI2UFr7nM2/tSj6f6951t950F5FIBCkEfhhwxmnn9lzQ3SHg1duey28MhmG0k8LCsiPYbhw3WkQ0UU68aBgFpaMoGjSW4vLxFFdMpLhiR0oG70BJxURKB0+guGwsBaUjSRQPJVZQhhMtxHHjWLbbnhTp6nSXDaeSrHtpu80wDGOgMiNGuqgtMeK6HfdK39TQwTlz5nRpP6+//vom7/vpT3/K9ddf36XtGduf16fPZuWSuXh+Fj+bRW2mhdy0l/7Th5EZPSV2zMBIZr33zht8NH8+RU89QVlhhIa6Gp554m7+/dqCTa6z7jePGp1rOdlqnxO+zd7Hf4s7jzuYwlq/1+MHeOGZZ3ji738nEo0QeFkkFiGSYYkSykoHbdUJwU9+eDNaa6647DsUFZX0XNCGYQxo609lkQhLImkbuaFbp7R0NFV73YSF2GjqiElgGIZhbMyMGOkGIUSHI0ZmzZrVY/9smjcxl//ZZ5/t8rQcY/t02F6TKfTrULUryVYvptBvyHdIxnbolTee4+abbiJZH/DEf95g0co6Fs+fzZw5Sza73qRRg9t/FwhO2nX91sdCCK547m0aulEIuzsW11YTi0QpjcWJuxEKCxMMLyuhIBEDufUdiIQQ3PnrjuubGIaxfdtwdEbuSzoLKS0sy+ngYrffb0Z1GMa2TfWzy0BmRox0gxAC2974qWurCSKE4Mknn+Tkk0/u8f0C+H7ffENqDHx/ve9Xrb9pnnzmv/kMxdgOhaFixx335q77HqWxsYkDDzyA516bjmWFNFQ3bHK95R9MZ96Kl9a77d+fPonSIVVP3srQU3/Yfvv//fvu3gp/Pa7rMGPhXC786rksr6ykqqaaVLqFYUNizPxsfp/EYBiGAV3riGISIYZhGJ1jRox0w+bmZO61114AnHLKKe3L7bPPPt3aTzab5fDDD19vf6NGjeowKWMYmyc45fgv5TsIYztjWZKIIyhMlDBqxHCWL1/Ob+++Cy0cCgoKOlxn2fRPGfHha+j356G9VZ/fMfMi+NcFxJrn8Me9J/DOcy/QUL+M4VP6rpPS0GHDeeaVl5k5exaVlZXUNNQx87NVW14R05nMMAzDMAyjPzOJkW7aVFeaDz/8sFO3dUY0GuW1115b77bly5d3a1uGYRj5UFFWxrAhxaRb0jQ3N3Pxhd8gbEl1uGwYphhx/4Poz2oRd78E4dL2+8TUB9FrJZnZlUyZOpi5113E6ImHIuXASBR3tZPZ1vAy6T7bl2EYhmEY+aMR/eoykJnESC/orcKoPd0T3jAMoy8IIRg/bhSTJozd5Gg7pQJE8xzEzhWISECYakHEDvh8G4D41v1Yu01kzicrWLG2kabaZX30CPq/f37zC/ztrH0JVYa/fvUQzhqV4PIpxSybOyvfoRmGYRiGYfR7Qg/ws+2mpiaKi4tpbGykqKgo3+G06+k5nWEY9uk3joZhGMbA8JPDx7PP1D1YOuNdVlZmuXVGJT/YrYJlDRmGl8a5fWYNDIBvcZYtW8aYMWPyHYZhGIaxDeiv54g9pe3xXf/Bw0QL4vkOB4BMsoWf7nvegH3OzZl2L+nJ+eT777+/SYoYhmEYG/nZCfti+wF1NVXMq4tz0mETWDvnQ65+5g0SEZuM5zEQkiJ77703Y8eOzXcYhmEYhjGw6H52GcDM2XYvEUJwyCGH9Mh23n333R6IyDCMrnj+zZns84VT8h2GYWzWj//zPlU1TaxcWcnEMSNZXt3A9aecTOmYyYwfFENquO2co2isqc53qOuZOHEi2WwW3/cRQjB9+vR8h2QYhmEYxnbMJEZ60RtvvLHV29jakScDfKaUYeTNP198nZ32/RJfO+/SfIdidNPsd57n3z+/HKUUr//PTqy48QDCp77KzF+eSjrdku/weoQQgpLCKKtWV7HTiBhPfpRi1JACbjvzBOanEzgSVn34DoGXzXeo61m4cCHRaBTXdfMdimEYhmEYhkmM9LatSUxsbVLjw/c/5NBD+66VpWFsK5TSKN+jrMAnHDqVW37xK/be/8R8h9VrPnz3dUYMKmDN8kX5DqVHJYaM5R8vPMtVJ+5OS1MGLWx+9ttXiYo6ItFYvsPrMddPW0smmeKH9z3PVw6bSHPQwksvvsqgQSWMGVJMi6coGz4y32F2SkNDQ75DMAzDMIwBQyH61WUgGxh9Dgc4pVSXa4RsTVKkpqqaC77xLWpra3n33Te7vR3D2F41JZtZ/OkHRBIRooVDmRGMYOGst/IdVq957Pqv8+V9hzN01Ph8h9Kjxk6YxNyFDdx4XDkwCB9JXa2mYUU9socLZOeTDaxJuuwzBB79x/ucsEuEKXsU0zB5D5rXrmH3XcvzHeIWaa055JBDKC0tNSMdDcMwDMPocyYx0geEEIRhiGVZnVq+OweF1113Pa+++gbZbJZYNEqqJcUHH5jaJIbRHcWFhYyecgTZpkakCDhon105+sAnqa6to6JsUL7D61GXnvwFVqxpYumSWp59+B6OP/+7+Q6pRy1qccikkliqmn+8GeXC40ewdFENCxYsYocdJmxyvU8/ms7ue+61yfunTpnCquVrOeiwA9hlh/HcdPuveiH6zjtwpEN6l6OZ//qLfFYd0uTDso/+wu3PvsTkAw7Oa2yb8tJLL3HUUUcBcN999/Hmm2/2eEc3wzAMwzCMzjDtevvYpg76hBBbVU9kypSpOLaN7UTwfY+PP57W7W0ZhgG/fuwNHEti2zZSgN+SRWuLllQDu48r5OijD8t3iD0iqH2LDII1L/+TQQXF/PWRh5g0dS+OvubxfIfWI7RS/P3sSUwYGWOvO+bz2iXjWbIiyQVPLt/kOjddeRmfzF9ESWkRjz7xJC3p1Hr3n3nyCaypqiIhbIbtOJHDDj6Q3ffal6l77NnbD2eLNJqWhgbixSUDIsmwboxaa4QQZsSIYRiGsdUG2jliV7U9vp++92i/atd7/f5fG7DPuakx0se01h1etrbIqutGkNKhqanBJEUMowd8/6xDSTb5NNS3UF+XJplVtHgBoV2wzSRFANJiB+b/8z7mLl7KZ82SS/66aJtJigAIKfnyo7Np9kOeu2gcjS3hZpMiAJW19cTiEVpaWrj2sov53+9f3H7fl/bbj4zvcfyXjmPqfnvT0tJC1dq1/SIpAiAQJEpKB0RSZFPmz5+f7xAMwzAMw9jOmKk024g1a1Zh2zbLly/NdyiGMaBMn7OKf700E8uyAY3WihsuOQaAqy88nNVV9Tz14nTKSqOc8aX+OSVhaxQOGsKe33qI/nFa3zssx+Ww38zu9PLRRIIix0JakjVr15IJPC656Fvcde/9iNISJu0wEekrRg4bzpDhI0h5mV6Mftt2/PHH88wzzwBQXV3NZZddxqRJk8yoEcMwDMPoBK1zl/6gv8TRXWbEyDaipKTEJEUMo4seeOSf/PlvLzCsWLDLaIcDJhdxxF4VvDttRvsywweXctFXj9omkyLddciB+3LCkYdz0pGHc/Sh+7H/bhNJJpvzHVaP2XnKruD5uFKSzWZZuGglvucBUOA6pLNp3OI4n374AXUrVnHFFVfzwt/+kdeYrzz9FE4dkkArNaCSCk8//XT774MHD+bOO+9c7/7nn38eIQTxeHyjxyWE2OppqIZhGIZhGGBGjGwz5syZle8QDGPAeem1d1k99wNO+tHZFFUMZ9ac2Uyd2EhNcicGFRxE3TZ0sr+1tNac9cWDcOKFnHLcKVSMqMCJRkGF+C0prvrWBSRTzQweMpg77n803+FulQsu+i7zPpvHuFGjqf9oOrvvNoWLLs5Np/nKV7/Ks08/RZBsYf7Ket6euYi0vJZZr7zBF888LW8xV342j7izTn0OlZumKe3OFf3ub3bZZRfmzp3LEUcc0V57pG160BVXXMGdd97JpEmT+Oyzz7rc9c0wDMMwDGNDJjFiGMZ2SSlFS1M9mbSHrn+RBTWHEo3GWLZwNaXjdwMd5jvEbpv9/qtM3u/wHt3ml/ebTMnIcZSPGEZl1XIef+IRRo4cSUX5UAoKCnBKB1G1ciUozWWnHcKv/jGwW4UnEgnKhwzGjUV59c9/5a577wHgtDPPZuG8OTQ2pairWUvEibN41mz+O+PjPo+xLQkihKBy7VpGRENQCiFAKQ0WhIHGsvv3v/q2xAd8Xox17ty5640QiUajtLS0kEgk2tdZd3nDMAzD2B7p1kt/0F/i6K7+fbRkGIbRS5TSvP/yk+w2dW8eeSHKvNmP0NTcwuSdJzBs9HN87ZuX5DvEbov5Taz81nBG3r+6x7a5NpmlcfFikuk0i5YsJNXQTLK6kmWFCYaOHI3j2KimWnRBIZ+t7Ln99jXf97n6+9+nobGRl15+lWwgmN3csN4y1/z0ZwCUlpZy70OP8o+XpuchUkCFICw0msp0gDNsHP/9y5+Y9eprfPLhNNYsXkLCtnimqf/XQBkzZgzLli0DYNGiRUyYMAEhBMuXL+fiiy8mk8mQSCRQSplkiGEYhmEYPc4kRvJsh4mTaGisRyvQhGgNUki00AgEUgi0ANeJoMKASDTG0qWL8x22YQwo++9zMAtmfQjC4oSTTuWhxx7FsiSDykYyduKOTN55R3baZU8CzyesW8Y785aSamzKd9jdNjL7CX9cIfjqLUdQcO0rPbLNogLJypWriTgOicIi/JYsEvDTKSJCI+uXU9PYyGFf2I/a+oIe2Wc+OI7DnXff3allL//xT7j8xz/p5Yg2bbLtMMgFadvsWODiNNfy7O8fILbLTlx0y21M2WkiETfKrBeeZ9cvHpO3ODtj6dKl7QmPCRMmoJRCSsno0aMJggDLstabTrOuFStWMGrUqL4O2TAMwzCMbYhJjOTZgoWfobVi50mTSKUzgEBKiR/4oAQaxfgxY3nznbfMPGrD6CKtNXvvvifCshk1djQ77jyFBYtXUlFYTHVzIxVDSvj7I/fxfLQA6QguvvQykomR7Da5kONPPjnf4Xdb2NLAnEaN19yDyZ1ogn0mj8fzNVYYpQHwgoCs5+PaAuFEGFFUyKvPvU7FqGE9t19jk257+M8MGVTCoOJSxhywL9LauJ6I1pqdR4zJQ3Rd98orr3DEEUcArPf/zrIs3nvvPSA3dSabzeK6LgDHHXccxx9/PDNmzNh4g4ZhGIaxjdMINP1jJGV/iaO7TGKkHxBCMm/+gnyHYRjbnOKYi60lEddl+OgxxKIuEydNYtXSBRzzxeN5/a3X2G3izkQjENYu4vrr/o9jjjuGhmSSW2//Rb7D77SFf7sUu2k10ap6ipw0r0+r4oaTB1N59K0M6qF93HHPn7j47FOJWBbRosG40QhVa+sYVFRMYVEpS1csJ2UJlBb87aVpPbRXY3OOPOPLHDeijKynuP6P93HkSeegAx9sC2Fb6KyPagmwyxL5DrVTDj98/bo49957LxdddNFGo0QikUhfhmUYhmEYxnZA6IHU168DTU1NFBcX09jYSFFRUb7DMQyjHylyLYL27hySa6/5ASsqa3n+H/9GhVlW1NcCkM2kueP6qxmz696c89Xz8xx1/9bUUM+XD9uf8sHlLF25goLiQewxeRfG7bIz370yf9NKtlczP5jGpWefRdDYwCANWcejyI1guy62G/DInKp8h9glO++8M/PmzevyegP8UMYwDMPoYdv6OWLb4/vxu38mWhDPdzgAZJIt3HjAVwfsc25GjBiGsU3yPY/igiIy6Sxe4KG15o3XXqW0YgSZbJqo+/nHXyQa4wc3/zaP0Q4cRSWlvPTpZ/kOw2g1Zd99eG3RovbrfjaLFoKatasZPnps/gLrprlz53aquKrjOHieB5jONIZhGMb2S+vcpT/oL3F0lylaYRjGNslxXVbU1VOdbqGspIyYE+fQo06lvqaOMPTJZNL5DtEwepwTieC67oBMiqxLKUVhYeF6t91+++1AbnRIW1Kkje/7fRabYRiGYRjbHpMYMQxjm7e4upKysnLu+sX1zJ8zA6WCXIFjwzD6JSEETU1N7VNkrr76aq644goAnn322fWW3X333fnLX/7S5zEahmEYhrHtyHti5Lrrrmtvwdd2GTp0aL7DMgxjGzNn+RLmLF1MOp3EsSzqstl8h2QYRie99957LF6ca1VfULB+O+gPP/yQCy64IA9RGYZhGEZ+tXWl6S+Xgaxf1BiZPHkyL730Uvt1q4OWg4ZhGFurrLyCqmRLvsMY0Opqq/n0/bfROmTuvNmUlZVjRWKIMMMDDzzKf199O98hGv3YWWedxd/+9jfuvPNOLrvssk6v98Ybb/DWW291WGTVtnOHMueddx4PP/xwT4VqGIZhGMZ2pF8kRmzbNqNEDMMwBoCiwgIuvvwKjj78EObOmgVRh0ULFjJ82FgGFZVw1oknIG34y7+ezneoRj/02GOP8dhjjzF48GAWL17Mb3+bK3psWRbxeJwgCIDPC6qOGjUKgEWLFjF+/PgOt3nooYcCucKthmEYhmEY3ZH3qTQACxYsYPjw4YwbN46zzjqrfbhsR7LZLE1NTetdDMMwjN6308gE++2xF0cefTSuG2HKlClMGjOa0046kbLyQgoLHYpLohTEE9xz5235Dtfox6qqqtpHegCsXLmSpqYmWlpa0Fq3X5YvXw6wXlJEa83zzz/fPv128uTJaK2ZNm1anz8OwzAMw8intq40/eUykOU9MbLffvvx8MMP8/zzz3P//fezdu1aDjzwQGpraztc/uabb6a4uLj90vZtkmEYhtF7Dtx3PCMGj6GiEFZ9Nptlq1dT31hDUXEZ77z7AW6gqa+qYcWSpTQ0VnPXr3/ZK3HsP3V3Ju+4Az+/7or20QXGwHTHHXcQBAFaa+6+++6N6o0JIYjFYgDr3SalZMaMGe3Jk3vuuSfPj8QwDMMwjIFO6I4m7OZRKpViwoQJ61WgX1c2myW7TtHEpqYmRo0aRWNjI0VFRX0ZqmEYxnajqWYZpx59OEnfIZ1uIeJGkJaksLCIVCpJTFoEgceQIcNYtGwxg4oSvPTRZz22f88LOPPofaitaqaktIL5a1cxa/7i9UYdGNsWrXX7lBrDMAzD6IqmpiaKi4u32XPEtsf3g7f+SrQgnu9wAMgkW7j54LMH7HPe744oE4kEU6ZMYcGCBR3eH4lEiEQifRyVYRjG9q2ofAwvf5yb5lhdtYby8iEI2TeDDpPJJEIIvnfO0TzwxLtcdvFlXH79DWSzKaQsQsr8nzxX19RyyrFfZPHylcxftJDCwsJ8hzTgmaSIYRiGYRh9Je9TaTaUzWaZO3cuw4YNy3cohmEYRgcqBg/rs6RI4GWZ869bqaurZ+oBX+BHl57ItOkvc8COYzhwl51RSvVJHJtz4Xnn8NGM2fzw579k9NByDtzvIL50zLH5DsswDMMwDMPopLyPGLnqqqs48cQTGT16NFVVVdx44400NTVx/vnn5zs0wzAMI89sN0J5cTFF9TMQhYOIlY1h6sQm9ppQzIH7jeLrJx/Bg0++hGVZyD5K1qzrB5d/D1tI/v7AfThRyaP/ehq0xnHcPo/FMAzDMIzti2699Af9JY7uyntiZOXKlZx99tnU1NRQUVHB/vvvz3vvvceYMWPyHZphGIbRD4w/6ar234vG7c/Yg89sv37KlfmI6HNr6uqZsvNOLF20mBnz5vONr53DxHGj+OVd9/X6vn3fp662HtuWuJEIruNg2Ta+5xOEIQUFCTMdxTAMwzAMoxPynhh57LHH8h2CYRiGYXTLnx56hE/nzOfff/otJQUJqmsbmD5jDlVrV1A+qKTH9zdv3lyEkEgpsW0bIQSWZeEHAb7rkojFcSMuMctabz2tNAhTt8MwDMMwDKMjeU+MGIZhGP3fKYfsTqHt8PHCKiaNGc5nq9eSCRWD4jYfzF2c7/DyavdddmT0D2/kK2ecxg6jBtOYzLB8dQ277NJz+9BaM2vmTKKxGFZr0iNXdFagVEgQCLTWSCGIosF125cDEK0FahsbmyguHniV4g3DMAzD6IAWaN1PvvToL3F0k0mMGIZhGJuVSTWz18gS/ju3hurmJHr1KjJpD5TALXDyHV6/UFpSzIsvvtRr20+lAxzXRQjRPuojDBVaa8IwRAgfx8n9LaSQoEG7zkbtjPOdFLn99tu55ZZbaGxsxPf9je63LItEIsHZZ5/Ntddey9ixY/s+SMMwDMMwtjv9riuNYRiG0X8EqWZ+d9X5zF7djBKgtKaytgnP85myy3jemr0o3yFuFwriDjvttBMjR45EaUXg+yilCFWI0oowDAjCgCAI8YMApRRBEBKG4Xrbee6/T/Vp3I7jIGVu6o8Qgquuuoqampr1kiJtyR4hBEopmpubue+++xg/fvx669577719GrthGIZhGNsPkxgxDMMwNslOFHLlPf/kjIuvYunilYRhyK2/+AUraut48tV38h3ediWbSbJy1nQmjJ/ADjvuyIQJExhSMRgBrYkQnyD08XyPIAwRWhMEAfPnz+flV17gsUf/xEEHHdpn8UopCcMQrTVab7pW/ebuayOE4Lvf/W5eOg8ZhmEYRn+l+9llIDNTaQzDMIwtOu0r53DaV87p9vpKeSya8QEjR5RiOXGeeeIPnPqtn/dghNu+bN0ihk2cQN2ij6hrrGHVZ9MIoiMYssMeRKJxAt+nsaGeymwLyaYmmpobqFuxBNm8ChVmwXIpLLogr49hw+KvnUmKbKisrIza2tqeCskwDMMwDMOMGDEMwzB6j9aaoOozhF9PTdVS7GgR7//jTs74zi8RltWtE+PtlY6PRSBY8sx/GT5mF0ZPPpjyoSNIr5pN8/xXEU0riPgNtCyfweppz1D78TPUL/mUIJ1E2gWQTvdpvJMmTQLWnyqzOZ15LWitTVLEMAzDMIweZxIjhmEYRq/RYR1L531MauUS7rzxWtZ+9iGJspHMf/0RfvG9E0nP/PdG6/zu97/ljof+yrP/upuG5W/zxlO/5ssHj85D9P1LQWEhQe0iyg//AjJWRElBjJLMLIoKogwaNoGI65JJp6ivXIqXTaNVQDbTwtqqWgh9tBvp03hnz569xWXWTZh0lDhpm4bTdtmwmKxhGIZhbNfyPXdmG5pLY44wDMMwjF7TsnQOS5fOoXLtAoaXuCRbUhSVj4Qwy7fO+yZNyRpiykdIhyDwOffb3+dvf7oXQe7/67P//BUjBw9l5IRdeOGxX/HFsy7L8yPKH8uSSDdG3E3QVLmKoopxOLFiAm2BHaNmxTxqV35GY81Sli9ai+vGKCsvQkiJ31xDWDG+T+PtTD2QTY0SEUJ0eF9HnWwMwzAMwzC2lhkxYhiGYfQO7fPJ2/+mvHwIa9es4fSzv8qTTz+NjBdRMWQEn81+HaEDsrVLAbjw7GP425/uRWuN0hoQvP3aqyxbOodrbr6LI8+4JK8Ppz+QBUPQWZ+6xkayGR/cQmTo0bDwNaorV9NYX4Of9hg5ZgwVwweD7SJjxYhEKed86+p8h7+RjqbZbCopYhiGYRiG0VtMYsQwDMPotttOncraZYs3ca/NQefdQKqxmuUrliPdQUz78F2e/sefmTV/DtlYKUuqVhOt2AGA4eVRJk0aDoBAUlDsUrl8Jc0NKaKxGJZlBjlKJ0F88DCyDStYvXgulUs/Y9Wct2jUg6heNpemmkqE5WAlBiGsCDr0CNI1+MkqSkoH9Xm8m0pwbHj7utdNUsQwDMMwOkdr0a8uA5lJjBiGYRjdcvOJO5NWFrCJf4RCQCbJDvsfx5nfvY59TvwmI4tczvv+9bz3/jssX1PLex99nlQpjkQoiwiGlEfZqSjKN/fZn4eemUlBaQWrlszjvdef6ZsH1o8JO0I2k6WktIza1YtZvXoVNXWNrP7o79Qs+ZBscyU61CAkVkEF0ikFLKoXNeYl3jPOOKPD29cdIdLZRMgDDzzQIzEZhmEYhmFsyCRGDMMwjC5TSlGaSLDLkV+iublpk8uJ2GD8UFK9YiGW7fK1r59PpsXjsCOOwbYTXPazP7Uvm1zagl6TpqBFUJRI4GdD7v7fq2lqrGflwk9xI/E+eGT9W3MyhZCSwTvuS5htorF6NXV1VaTCEmThUGyZwIpWoNIZGpevIPAbCXSc0h3H5iXeP//5zz22rW984xs9ti3DMAzDMIx1mcSIYRiG0WUX7DWSFVVJHvznqzius9llR0zch90POwspI+xz0o8YOnZX9jziAs666Ib1lrvxyec5/pBj2WXkMIYUFaKxUUuX4DVVEy0op2Tw8N58SANCUVEh0177L64bo2Kng0EKPGWR9jyUcqBoKEoE+F4LkcIYa2YsRkYLqRi3Q17idZzNvzbMtBnDMAzDMPoDkxgxDMMwumzKvnsy6ZgvcdnXT8aVne8U0lFL1nX98O+PctDIcZTJCK4UVNauYcQOe7P7vkcxfvykrQ17wHNdl6l77cXiT95myLARRBMJQstFIfBS9Whp4Wc9pBXFSzdStvsupFONnJnHwqs9kfyIRPq21bBhGIZhGNsXU8nOMAzD6LKjvnI+QboJL5vGVz1bbOual1/q0e1ta9zi4TQunM643Q9k3G4HIRctoHbVArTWZFuaCcM0gdJ4KooVSvxMGsuy8h12hzrbgaa5ubkPosmPiooKampqNrvM6aefzhNPPNFHERmGYRgDhUagN1XrrY/1lzi6y4wYMQzDMLpsjyPPoGRQKaN3nIJsWpHvcLYrzas/Y9SuR1JdXcnQkRMYPGIc5aMnUzJ6d6xoDGW7ZFMNNDU3kk0n8fzOj+jpDdOnT1/vendGkGxpSs5A1NameEtJEYC///3vCCFYvnx5H0RmGIZhGNsfkxgxDKPbpn/0EYcffhinnfllVqwwJ8fbm4kHnIoTKyZSMirfoWxXIgXFRKIFLFu0kGwmjS2BsJnS8sEUDp6AGy3DBxKuTTaTRseK8hrvnnvuud71LU2n2tZprbv9HIwZM8bUZTEMwzCMXmCm0hiG0S3JZJLnn3+ORHEht9x4K57n5Tsko48JIRg6cSqrF3ya71C2K1mZIEwlSSeTrFy+kiHDR1JQXEF1TSXJZIpMNqSxZjUfL1gE2udXD9+d75C3KhkwY8aMHo4mv6Tcuu+kpJQmOWIYhmHk6NZLf9Bf4ugmkxgxDKNbli5bxsczP+HXt/8ayB2sL1q0iAkTJuQ5MqOvDd9h93yHsF2pGDyUlcsW0/jOczRM2o2WlklEonGa6qvRwkKHGaxIgsmTdyGVqmf4yNH5DrlDnT25nzJlSi9HYhiGYRjG9s5MpTEMo1suufhibG0jhMCyLKSUW/1NqGGYb8K3LBaNUjqonC9ccR2pyjUsnjWNBdNfZ83iuVQumUEm49Pia7ItLThOQb7DBWDq1Kn5DmGbccUVV+Q7BKOLWlpauOuuu7Btu722zLoX27a54YYbqKury3eohmEY2y1zFmMYRpelUil++9vfMmzkECzLwnVdpJTbfe0Ao3uWL57LDy48iu+edYB5DXVSNB5n9fxVhPPeJtVcR9XaJVRWrWHJp69Tu2YRLZXL0Y5LKpXJd6gAvPbaa+td72wC7KKLLuqFaAa2n/3sZ/kOwdiCp556ar3ERyKR4NJLLyUMw42SIgBhGPLTn/6UsrKy9tsnTpyY50dhGMZAoPvZZSAzU2kMw+iyRCLBW2+/x4KFi3jrndf56+OPYQuHrOdx2MGHcu21P8h3iMYAcetVZ1OzdiWKDNL3+e2N3+KSH91nEiRbEI/FKRhawnE3PsJz//wjKz98jWQqQyaTJS0KsAuKaKyu4+6/v5DvUAEoLi7u1nr33HNPD0cy8MXj8XyHYGzCCy+8wDHHHNMj21q0aFGn21kbhmEYW8+MGDEMo1uak41YlsWzz78ASlJdVUWqqZmyiop8h2YMINVVi6lvqKGmqo7mZC3zPnqXfz34f/z57p/kO7R+TUrJ6NGjibpRhowaj50QEItiF5eQzTbhDpnUb5IibZRSXVo+k+kfo116mplWtO3qTFJk3URHZ5IeRx555FbFZBiGYXSOGTFiGEa37LbbFPbffz8KCgpwHAcpWqfSmC/6jS6ozUqECsmks1iWIGuneeeVZ5HRYl46aXf++JTpeLM5RSVFHHP8GRxz/Bn5DqVTlFKk0+nNjnq48sorue222/owKsPov1555ZWNRtCZUSSGYbTrT3NY+ksc3WQSI4ZhdEthQSFuxM0VXiWXENFaI4UZiGZ03h//9i5KKS4+bS+CoAUhIVFYRirjMWXH8fkOz+gFsVhsuz2x01rzySef5DsMY4BqS5CYKTaGYRg9z5zBGIbRLfF4HNeNIqVEI1EhZL2AoUOG5js0Y4CRUvK7f0zHLRyElgmqG1MUlg3mOzf8Od+hGQPcu+++C+RGqrz//vt5jSUIgh7p3GXq7wxsW/P3WzcZYl4HhmEYPcuMGDGM7VgmkyEajXZr3YgbpW3MnNagtEYKQWlpSc8FaGw3pJT87i/vorVGCEE2myES6d5r09g+rdvlY1PWPbEcOnQoa9as6e2w2jmO02Pb6sqIgbb3lJF/PZEUMX9LwzDWpbVA6/7xudBf4uguM2LEMLZjF3z70q1YW6Fz82fQgBAaIc3QXmPrtB30m6SI0dvWrl2LEIIgCHp9X71xMtupbZrpFv1WV6bCbKpg64gRI3o0JsMwjO2ZSYwYxnZq+YpV+GEWgMOOP4GK8eO7dKAWaoVWqvVnkDsANwfhhmHkSWcSBR0t89lnn/VGOH1iS5/ZZrRI/7S19UFSqRQAq1ev7olwDMMwDExixDC2W0+/+CoFiRL8IKCytpYjv3AMf/jjnzq9/pTJu6J1uF4yZPfddu+FSA3DMHpGR4mC4447DiEEUspeSSKcfPLJPb7NNgsWLNjCEiYpMtDNnDlzo9s219XJMAzD6B6TGDGM7dSH06ZTVFDIMV/+Cn8/aF9+XG7z96efprK6lt/94Y+EodriNnbeeWcaGptpaEqy665T+iDqgau8vLy9BsKDDz6Y73AMY5vTmW/hN0x8CCFYuXJlr3b5+PDDD3tluwATJ07stW0b/cOuu+662ftXrlzZR5EYhmFs20xixDAGuAu+ewWfLVrEoqVLWbayptPrrVixgoryEu4q0DS9/CfE3OnMmfEpv3/wYZqbM7z4yhtb3IZt2xxy8IEcctABmNHaHTvvvPMQQlBbW9t+24UXXogQgvfeey+PkRnGwFdRUbFe0VWtdacubdquK6V6LTHS3QLXndETXW6MnrOlArsbvv56wk033dSj2zMMY4DR/ewygJn/qIYxgN0xpIhT/nsPK4/eiZXfOoLV39qfaUd0rhjbDrvtxt+feZbpf3uGt+pLmB4p5Ufnn8dOKsXBQSMTa5b2bvDbgdmzZ/PII49s8v4DDjjAzP83jK3w1ltvbfU2lNry6Lit8fvf/75XttupETLSfL70pYKCgj7b15gxYwD417/+1Wf7NAzD2JaZdr2G0QVhGGJZVr7DaHfomScx7PATiY0aCfEi7KZVfPrjm5hxzTXsduutm1336sYlnDrtbVYqwelXXIU1cihuQRlWohA7oYmXlPbRo9g2KaXYddddNxqiv+432+veppQySRLD6KJJkyZt1fqWZbW/74Ig6JX34FFHHdWj27vjjju4/PLLe3SbRs+oq6vr1nrded2dccYZ3Hbbbaxdu7Zb+zQMwzDWZxIjhtEF/SkpArD3bx7d4JYpHPLSsZ1ad+zv/si79/6JL955O9kxI4mWDkZIiS0UzTiUjjVz17vrjTfe4LDDDms/2O3ooLctYdJ2n5TSJEcMo499+umn7b87jkMmk+mV/Wxtd5hx48Yxd+5cIpFID0ZlDGRmGpVhGEbPMokRwyB30Dp37lySySTRaJRUqoX6hnoiEZdBg8pQYUgQ+BQUFDBu3Hii0QggBvyByZGzp5F5432mf/c7hGuqCGwLN9SoIYIxa/x8hzdgpNNphg0bRmNjY/tt3TkJklL2Wp0DwzA2tuuuu66XkOzNxIPWmpdeeomjjz56i8smEgn++Mc/csYZZ/RaPMbAVlxc3OHtGc+jWWWxpUUQBPi+T+D7hH4aP4TRZYOJxhJ9HK1hGEb/ZxIjxnbvheef49MZM/ngk2lc8j8XU9dYx3U/uY7xo8Yh0Gg0S1YuR2uNZVncc889rF6zmqFDhjJ16tR8h79JKgiR9uZHuAzeaW/+svM+uEIQopFBgA98dY05Oe+MSy65hN/97ncb3d7ZpMjmRpIYhrF5PTW6qi8T3EcdddRG7++tHU1ibJ9+9KMfAbDjjjuud3taCCwh8X0fVO61poMsYRDiZzPMX/wZntfAmKFjqRg2vs/jNgzD6K8G9tfdhrGVPv7kY35x+22cfPLJ3HTdzxk+bDj3/eH+1m8QNUKAbK3/EAQB40aM4sILL+T55/9LJpNl1sxPt7yTPAlTSQB836eiYhiHH3pYh8udozWnK8U+v7+fc7Tmq+akvFOEENx9993tHTHWvXRk3Y4YSqmNLh11zDAMY2O1tbW9lkg45phjemW7m2OSIsbW+M1vfrPedR18Ph0sUCFBEBCgCVSIp0O8ME3MLmTOjP57/GIYRudpBFr3kwsD+/+ZSYwY26XKykpmzpzJTb+4md/f83tc18WyLCzLompNFQACCQgEEq1y3+jNmj8PpRRf//o3SSTiICx+dsMN+X0wm+AUFbH3kDImTt2PbKKAmQsWcKHlUvdpxwdDO377m30c4cC1uQTIhkyywzB6zsqVKykvL++17b/wwgs8/fTTvbZ9w+gJ6/5P2TCZ56czpJNJMi0tqEwW7fk4gcRVNq5vIUObxoZ6iiqGsnTpXNbWraEh1UxzS7avH4ZhGEa/YqbSGNsNrTXz53+GlBI/CEmnM0ScCO+89zbPvfgc1ZXVWNLCz3rYts3cRfMoKirkxJNO5prrfkwk4lJdWUVRUdF6nUVO/fKX8TwP13Xz/AjXV/nc09z9zCt8PGcWu0/emW9/70oeqark0KlTOd+cqHfb2rVrN0qKtB2kbmpqjGEYPWPUqFE9sp1NvYcBTjzxRJPMNDrU9ro56KCDeOutt1i6dCknn3wyM2fOBODAAw/skRbSW7K56V/LFy9ov3/D13kYhvh+QCgtIokS6irXsPvIiUjL5r2P3uaAvQ7u1bgNwzD6M5MYMbZZvu+zdOlSwjAkm83S2NhIWVk5EDJt2jSGDhvBN7/5LYoKC9lj6t60Hz8IQOfGikDu4DjUmjVrqygtKgQ+P4jWQHrGTGZnPfbYa8++foib9Y9TTiH43UNc9d1vo5TH/AXLOO7UU+HT6dwvJd9SKt8hDki77bbbRrdtbvpMZ5c1DKNvmDo+Rnes+9n99ttvd/hZvuHtvfE6C8Ow/feOCvn6QRpLRhDi8wLxbXGEYYhSIUhJS7KWvfc7AgClQkqKB/d4rIZh9AFN2+lK/vWXOLrJJEaMbU5TUxOrV6+mqampvf2iZVm4rkvWy5LNZth3331b58Pl6ohorcl6AXWNDZSVlmBZEpDQlgDRISVFBa2/69z7XmuUVsR23ZVstndaPG4Njyg//dG1DB05koKyMj6ZOYfqplqC1oKyRvdUV1dvVXHV3vbJJ5/066LAhtFdDQ0NPbIdkxQxukp18EXCuiNHN6U3knC2/fmh+wsvvLDR/WEYgg7XawnfVttK46MUNNWt5ZATv9q+zrtvPMHYsfv0aJyGYRgDjakxYmxTGhsaeOgP91FfV0cikSASibRPcUkkEliWRbqlhSAI0FoBqvWnxrYlQ8oGEfh+a/ZVIdrTsAI0ZDIZaupqqamtyRVk9QP8wMftxRaP3fV6wuWTaR9z3te/wZOP/5Ppc2bSvHotnhQEA7w4Ur715xOr3XffPd8hGEavKCkpyXcIxnaosrISy+q4w1tn/he8+eabW7V/IQSWZW1U2+q6667rcPkwDEEEuSMXIQgFaCnRlkRLFyvisteeB677KCgetiMjxpoONYZhbN9MYsTYpvz0uuvY58CDSRQUorXm5zfdyJdPO5WvfOUr3HnH7aA1YajJeh7JVIrGpiQNjU3UNTTR0NhEQ2Mz9fUNNDQ00dDQRH3rT8e2SWeynHzSyXzjgq9z/rnncd555xIq8ANFJuvn+6FvJDNlV+598AFGjx7L7x76PTM//JiCRAJfa/x+fGI/UHTUcrOjS097/fXXO+yE03aZNm1aj+/T2L5s+Jr6yU9+0n779mDYsGH5DsHoJ4QQDB06dKu2ceihh251DB39L/npT3/a8QpWBC8UhGh8net6FurPR7xYlk1TVpNsagBgwZy3iUQGgfnCxDAGpNxXuKKfXLrv5ptvRgjBZZdd1kPPTNeZxIixTfn61y8kFksAguZkko8+/IgLzvgCpx67J2+99TYNTY3U1FbT1JQklWohk8mQyWTxMhmymQyZbG40SRiG611Wr6lk9qzZlJSWsN/uozh0nzGk0xkam+qpq6+lvqF2q2Nf90R63Z/dPbl++YNZnPXl0zjwoH3518NPsGj5EiypkVrTLDv+9svovk0lKnp6H1/4whc2u8x+++3XK/s2tg8fffRR++9tr6Of//zn7a8nIQQXX3xxvsLr8udhd94Lq1ev7tLyxrZp5MiRPbatnvw83m233Tb7PlDoXG2Rdd6zAhCWBCmwpUDrDKtr1zJ/4Vx8u5QJo3umqLFhGEZ3TJs2jfvuu6/DOn59ySRGjG3KwgWfsWr1CmpqqjjhhC+x68Ri3vtwHk3NGQ7YawLnnHUWfhDg+Zncxcvi+Rmynofn+fhebvhpJpsh66XJehk8P4sfZPnRj67FESlee3cun8xZw5kn7sOVV1yG52VJp9Pdjlkpxd/+8W8qKobhRouIxBLEEmXES4cwbsfdOPmcC3ntrfcIWpM0nbXTLjvy7IvPM2/mAr539RUcvP8+1NbUYgvJENX57Rg9oydGkWxqVEpfjVYxtn177bXXFpe5++67Of744/sgmo7dfvvtW1ymLSGybseoziQtzfvGyGazDB48mFWrVm12ub56rYwfP57x48ezcuVKtNZ8+umnm11eItCaXC0xKVovklBCJBLBibiEgY/y07S0NFNcOoQP3n+FWbM/7JPHYxhGD9P97NJFyWSSr371q9x///2UlpZ2fQM9yBRfNbYplVWVNDY3s8fUqRQkCnj5rVl85fSTKCkQaKUIwpDi4mIymQy0HjhorUF/Xo5USokQn7e6k1KQTDaTSCRAhHzz3BMoLYqwcNFS/EwL2UyWB+6/j1NOOrnL8SqluOmXdzFqZAXvvvcedsRl/JjRWDGbRbPmIIXEdixqa5Kc9tXz+NdfHu30tn96/f/xla9dwOSdJ5NMp7n7zl/xh9/dwxzlE5ohs1tt2LBhrF27tlPLmpMtYyDr6PX77LPPctJJJ/HUU0/1eTxXXHEFV1555WaXWTfmzr7/zPvU6G8j7ZYtW8bo0aO7tI5yJK6SICXalmjAkjYxaaFCRWnFMCwZ4rhxtNQsn/MRSkDopVkwexo7TDZFWA3D2DpNTU3rXY9EIkQ2UY/x4osv5vjjj+eoo47ixhtv7IvwNsmMGDG2KaPHjGHKrlPw/YBMJs0Rh+3NYQfsjpcNGFJRjCUloHEcB8d1iEQiRKNRorEYsdZLNBYjEnGJRFwcx0YIQSQSxfN8DtlvN8aMGc3Tr8xhzJhRzJu/hK+ccQbPv/Bil2NVSvHQX/7OeeeeyYEHHoSwLYIgYMHiJSyaO5edDjuUPzz6EI5jM2JkGVopEJogCPntffd1WCV/Xb+65RauvvxK5n76MVd//2LKSws48ohDOOqxf3CXGTHSbW2jMTYcbn/77bfz+uuvo5Ta6DLQR3I0Njb2WEcQY2DoTMeN//znPwRB0FchrUdrnUtW95AtfZ4a2762ESKe56G15tZbb93ksl1NoNxyyy3diqmrSREAgUS5EmwLS9rYjott28TcOEWDK/CFYvGSJSxbNJtl82fih4poNAJa0pxOdStOwzCMdY0aNYri4uL2y80339zhco899hgfffTRJu/va2bEiLFNOeH4EwCoq6vjttvu4ON3/stfnnyDRQsW0ZxKccopp/Lyiy9vcv3NnQSccMIJvPbaC6yt9dhxTBF3PfgsZaPu54gjjmDChAmdjjE3SgVO/fLprGmo56AD92sdmQJgobXmD398kKHFhaxavJpMJkMsmkDICCCYetihTBwzkdNPPIGhQ3NFAjs6SHvzrTcZt8Mk7vnD7zh0v0OYtMME3n/vHXbbbbd+963YQHXTTTfxwx/+cJs8qfrrX//KOeecs8n7s9lse8cnY9vSlc8Hx3HyluxLJpO8//777L///lu1nYGYrDS2jlIqV4djHePH57qyOI4DwOWXX84111zTI/u74YYbemxbW+JEHGzpgAC77csd1yZZtYbm5VVE3ASxRAJpRUimGknEynAcG+1n8UPV3uLXMAyju1asWEFRUVH79Y5Gi6xYsYLvf//7vPDCC0Sj0b4Mb5NMYsTYpixdupQPPviA1atX09jYSJpiPv74NZTSfOlLx2Pb67/k1y0ouKnb1r1v6p778+qrrxKGIV//+tcZM2ZMlzsYJGur+PTsM3GGTeSvt9+JQKClQCuBFIqFixfxwOOPE4u6DK6o4J13PqK4uJgRw4Zx0pe/yjPnX0DNH37I1y5q4IuHH0ssFuWSb5230UEewJIFnzF16h58e8YFjB0zirFjx/abD59twbXXXsu1117bI9tqampihx12oLKycqu2s2zZMsaMGbPV8WwqKdL2fohEIuaEchvzwx/+MN8hdNl+++3X7RO5p556ihNPPLEXojL6u1WrVjFq1PoFRzcc/bTh8cKGNhxVteFrcN3XZTqdbv99p512Yu7cuVuMsaP/6Z0Rj8ZBCGzbQQgb13HIKkFtw6cIYeNn0zQnG4lEopSWDyUScQmCgMC2ierQJEUMw9hqRUVF6yVGOjJ9+nSqqqrWq20WhiFvvPEGd911F9lsdpOt0nuLSYwY25SxY8cyduzY9W6755578hNMq7apFEqA1KCkhTfnXS5a8Q7Tj/ojQyZMRCSi2OkW3PLBxD5axf2L1mJ/YU+84eWowjirq2tYtWIFh1Qv5JHv/B0/6rDbF6byh/vu4YgvHs2yGXMYuctElOXiyvULC37yycd5fPRGZzzwwAN885vf7JFtdaVA76Zs6sC4o9v/9Kc/ccEFF2z1Po38a2uV11Udffve19qmPnQ2UWmSel03ceJEFi1a1H59ID+H5eXlG93mOA7ZbLb9emfeC1uqZdN2m5SyPVEyb968TbbgbTN8+PAtFn/dlFiiKFcnTUs8KySZzrBiznRGj5iI4wpsK4plWViWRClN1k9hWy4608KECVO7tU/DMIyuOvLII5k5c+Z6t339619np5124pprrunzpAiA0AP5Pxu5b1mLi4tpbGzcYmbKMPpKGAZkBSx+9I9EHIuPrvk2o75wNOO/fgXWoOFIrwXfy6A8jdIhEgGJOFRXUfXSf4mkVlJ2yjdo/nQa/sKZKGWjBCSrGhh0wulEBg8hWlwGxQXEoi5EIRKN0lzbRMFOU4hGzRSHgaKjbxk7I51Oc+CBB/LJJ59scpnufrx3JjHSneKWRv/WUbeWzvxtpZQ9kpDrSeXl5dTWrt9GvbCwcKOCcEbnbfjaGMjHXWEYbnTQPWHCBBYvXrzea35zn4Vd+dxTSiGEoKamhsGDB29Ub6ptP9OnT2fPPffsykPZyIdLP0NIB0sL/MCjrqGRsUNKcS0nN0JVKULfRyiFlpKWlnpiBYOw0z7jJu2yVfs2jP5kWz9HbHt8l7/4dyI9WHNra2RTKe48+vRuP+df+MIXmDp1Kr/61a96PrhOMCNGDKOHZcKQZ/adQmzxQnb912ukXn+KkaNHEr7zEgtefZ4gFOCBDjW+kFgVFaQXVDLp9FNQE4YRL67A3mkKLR99imhqguggpBMlWlBEfO8RWFP3pGBQgmDlctb84SFWvvIyQWUde111LeLYY3k/HuNLYWCGww4Am6rQvSW99bctKCggldp08b1NDRmfNWsWU6ZMMQmS7VR/rLFTU1OT7xC2eSUlJf3yb99dp5xyCnfccUenRkB19bOu7TOzvLyckSNHsmLFCpqamtpPHHrys1NlPGxL4USiDBs6nGFDhmApCHVAEATY2QxB6CNsiZdsYenM2Rx76jm4ETPN1jCM7ZtJjBhGD9Ja8aPvfZfJa1YTdSXvnXAofosiLV20UkigCEHWEkSkzbwbfsMpZx5HQntEYlGCj2Zh+YqxZ5yA69vMe/Ep5lx0IS2LVqMtEEqTQWPpXEspLUAjiWjFtBv+jy8eeghN7Y3ETWKkv/N9v8vfPPZm8mFzSZHNmTJlSg9HYvS1M888E1i/LoJJdBlt7r777o1uG8ivj46GaN92223ccccdWJbVq49t2bJlSCnbi7x2xi+uOobBwyYyYuejGDZmPCKqibqFyKiLJS2kdEBKAqEZNnQkQmmwFL4KEMonkwkQviLIZqhes5LqNYtpaFjKmmXLuPLGh3rtsRqGYXTFa6+9ltf9m8SIYfQgpTSLquCQ6iSZUKE1CAlxAgS5xIgi5AmR4PURwyl+6LeEupGZN/+YqYFkUpnHLyptxKARFIksJx17ClcohSUg0BqpNRJBQOtBmwaNwgFekS5fOeII7nTAV1lcK5bPp8LopLYT0c4eiPdGLYdZs2bx7W9/e6u3Y06o17fuyJqZM2ey6667drhcf+gC8fjjj+d1/0b/9Mtf/pKrr756k/cLIbBtG8/z8v4a3lp9Fb/neQDEYp3/H331bc+jwpDbrj6G+cWlxOJxIpEyIvFBxIoqcOJxIpECJBE0ijCbwQ/SeOkWfD+Nl60nyHiEKkUYaFpa0pz1PzcyqGJobz1MwzD6Stv3of1Bf4mjm0xixDB60Ce/u4VLXryP4oMm01JfQ0tdA+lkFpnWpLICD03WElxme1y+ZhGROCQfuJXjdxmHvfc+VBQX8NC5V/HXJ/7JIw/cw6e6GrQiAAIELrl2v220kIRaE6B5pLAUkWpmru9z1+gJXLFqdd6eB2PLKisr85pAmDt3Lrvs0jvzybs6CmZbtG4CSwjRPqqmo+dl9erVjBo1apualmAMbF1JEgRB0P56/+ijj9hjjz16K6w+I4TYqEZNT3nhhRe6tZ60LK6+/SW01lx5/gGMGDqCaGwVTjyK1hopJJbUSCsOoY8WEi1thNIEYZZs1qcpVc8+B32No086q4cflWEYxsBnEiOG0ZOWLCIcOpLK1atxYiXERg+ntKQIL9OIs3IRLZUtyKxmTjogg0ZnBE5dA3phDfZ784gIl4qDTmPc2DF89dvf5Vd3/BZVuxxfa2wtyKCxgZDWpKzO/e5Ykm9fdBFTp+zMted8lW+WbVxx3+hfEhsUyupMMqGsrKxH9m1ZVqdOwjs6OdqwaKCxsbbndsM24FprPM/DddcvjjxixAgg99x29Lf5xz/+wemnn95+vaef/47alZu/8fZra0ZO7LnnngRBkJduAj1h3dFbPfV5u6Ezzjhjq9YXQnDHw+/hBx5fPuYgyhMhQ4YUEE+4OJaLZWtU6ACKbErjWxI7PoTV1RHu++PTPfMgDMMwtkEmMWIYPUjXV/LEgrUkCwv5W+MSkAKhNELaoDRKhOwYjfIlV1HmB9g6Ny1GSgiVxdqCOI3zFzJm6GA++vBDRpRFCGsECkFaaCQylxTRGi1Aak1GwD+dGIOrVzFjVki1sNjvksvy/VQYWxCPx9t/7+xJaF1dXY/su7tJka7oD9ND8sWyrA4fuxCCSCRCZWUls2bN4ogjjmi/T2vd4TSprj6HiUSClpaWTr2m7rzzTq688srt9u9k9I7bbruNa665Jt9hdFvbZ1dvjXxbtyXw1nBsl/+8PK39+gfvvsvjjz5AsqkZ0EQKCvnaeRewzwGH9Mj+DMPop7TIXfqD/hJHN5nEiGH0oNpVa4mokC/F0zzeCFppNAKtQizLxpEWJ8UKqGlsYghgi5DCGNiWIJVSfDZhLHsEPjMWLiT0fL53+TWkLvombcVUc6XaNClsmqSiQYGj4dXQ4a5DjuCB+35PQnsc+60L8vgsGJ3R0NDQpeWXLFnSI/sdPXr0Zu/f3ElyV04Sstks0WjXuxy4rts+B39bNWTIkPbfbdvmkUceab8uhEBKuVErz7brv/71rzvcZleTGx2NEumujz/+eKvW70/WrFnD1772NV555RWEEOyxxx48++yz6/3NjM17/vnnB3RiBHKfdZ999hk77bRTj28X4MYbb+zR7QLse8AB7HvAAT2+XcMwjO1Fz1fxM4ztWGHRIE6dPJg1VSEXiyhHOAl2jMaosGwiOiTma8oamthBaGwUy0WEtWmH1c2CFmXxzLyFVJRVUFpUxDtvvUll9VpsNDEkQggWAWuFBZZiTKjYD80uMkSrFINLBtHclKJCWgjTkabfGzRo0EbDzTd3gnrDDTf0yH5XrFjRI9vZkk0VGt0cIQRBEPDJJ5/0fEB9aHN/xw3vC4KAs88+e7PLtZ1MpVIpvve973Vpf52JYWu/FZ86depWrd8XpJS4rss999yz3u033XQT+++/f/sIgeHDh/PKK68Aueflo48+YujQoey33375CDuv2p6TDX/fkpdffrk3w+ozkyZN6rFtFRYWtic9AX70ox/12LYNwzCMnmFGjBhGD0kmm9n3iRf4UvFQ9nejDJowjh9M3oXyCaPQjfXUTX+NlTMXk00L6pA0YFME+DY4EYnnC15642Uuv/LHXPQ/3+Dv/3ySk485jNsBIcDSiokIMihCLagXNj4KB4nQAR+89zZfPf98Ljz7X2Zo/AARBEH771satj158uS+CKnHLFq0qEvLl5aWtj8HmzvRHgiFXbc0jag778/efsxt299WPzuqqqpYtGgRY8eOJZVKEY/HEULwv//7v1x88cWUlpZSVVXVfuJqWRaWZVFQUJDnyPvOhiOV1p1S0tXtALz//vvsu+++PRrjQJVMJtt/7++fX4ZhGNsrkxgxjB5SUFCI0ppFY0bwyuwZhLNmwKwZADz77LMc89v7oK01K+QqxUuBTUhEuuDGuX6/A/jdg4/ixmPM/mQ6YajJ/vw2dtl7N0aNG09BNEpoCxzpIIRm2LDhxONFSK0ZO34so0aNZsgQ035vILrooou49957N3l/Xw3l39RBe2+eMDc0NNDY2AhsXP9k3URIW8Ha/lwgdMPYeuJ562qSaWt0tTaMbQ+Mw4jy8nLKyzcuSu04DkVFRWitO7x/e7Dnnnt2ajpUV99vbaNs+uP7tK/ts88+fPDBB/kOwzAMw9gMoQf4f6ympiaKi4tpbGykqKgo3+EY27AFi5cwfcYiFi2vQqdqSWUDykftTOAUkEkFtGTS3HLZsYRKESsfTdCwFlD8/Oc/53vf+x62bSOlbP9WLpPJULl2LWvXVlJbV8Wpp34FCLEdl2Rzili8gAf+cC8jRoxoPzlMJpOEYUh5eTkTJ0xg+PDh2E4EKSOEKo0QgjAIttlvfbd1TU1Nm/wc2/Db3M7oaPnuvDa6kohYd9mOlm8rDhqGIUIIli9fzrhx49rvj8fj6327uqkipuvG889//pNzzz13vfXyIQxDbNvm0ksv5Te/+U23vm1fV9vj29LzvqnuQV1ZvrvaRl8MNJsqdtvme9/73ibruWxr2l4PTzzxBKeffvom68909vNnw6TgTjvtxNy5c3sq3D7XE4WoDcPoe9v6OWLb47v8uX8Q2aDTYb5kUynuPPa0AfucD4yvegwjj9auWcNOO+5M6bBx+C3NYGfINAWU7XwEu+68Esu2WVPn09wccuvlx2FJSbY2V8eh7dtXrTW1tbUsX76c+fPnU11dTSaTQSlFOp0mk8nw3Uu/y3333Evg+0SjLjfffDPV1dXU1tbiOA5KKVzXxbZtLMti3mefMXfePP7zn38TBAGlpaUMHz6cdDpNLBYzyZEBaHP/RHzfHzDfzmutOfzwwze6/fjjjyedziXwNvVYWlpa1jth3dKUFK01X/7yl4FcYiKfbUIty+rxk6Ctbe3ZXTfffDM/+MEPtrjcQEyKdOaz8Te/+Q2/+c1vePvttznwwAP7IKr8uOyyy4D1T94PPfRQ3njjjfbb160zsuGyG1q3FXXbuvPmzevpsPvU1nTY6kwHMMMwDKN/GBhH2YaRR5f88E4m7Dye8rHjWf7xZwwZX0J9dYqR40twIrkTocICQfXaFmDTReoGDx5MRUUFe+21V/ty69Ja89tfbbnjxKYO0rbn9qjbg3ye8HfHF7/4xfWud2b0RHdew+su/+ijj3L++ed3af3e4vv+Rrd1Z3rN448/vsVlejoh1BbntddeSxAEOI7Tfl9/nsbUGV19fR100EED9rF2xj333MPYsWPXu+3111/v9v+SwsLC9a4rpTY7MmegaHsNjBkzhuXLl29x+T/+8Y9ccMEFvRyVYRiG0ZMG/n8rw+hFJ5x9KS88fjd2tIWaVbOpmDABZJwVi+r5bG4VvpelxQ8QSiPtLR9Itp0cbmp6wKYuGy63qW0b266B1hL12muv7fN9vvTSS32+z01Zt+jkhifWGxa47Oh+rTVz5szp1L66mhTZ1In+j370o43us227PZ4nnniC0aNHc9111/HGG28MuITBwQcf3K31tuXPVs/z+PDDDzu1bGf+3rW1tetd39aeu2XLlrW/H7TWLFiwgOrq6vVu01qbpIhhGMYAZEaMGMYmPPjg/fhNdYwaP5zAtxBhBhvFooUNxAoSDC6L0ZzKUtcQkPUVpWUl+Q7Z2IbtueeePbKdRCJBKpXqkW111+ZOsLZ08jVQTsbbWg5vWCdkSzUb1r1t55137rX42vbz0ksvcdRRR3VqndNPP53TTz+912LqbW+//Xa31z311FP517/+1eX1Hvjyuew1KM7a4w+ifMSuFJaU4PkZhNLsvMtOPP30E/zwZ79j9odvdju2rbVo0SLKysrar7/66qvtv/d0a+dtzcSJE/MdgmEYhtFDTGLEMDqw265TKRmxE7PfeZGCshKcwpGsXTafQ3Yo4f03l+Na8P/s3Xd8FGX+wPHPM2VLNj0hEHoHqSqIDbtYUFHs/fRO76fYy50Nz3a2u7OcvZx6Z6+nnl1sWBFBEBDpvaa37TPz/P7YJKSStskm4Xn7WrOZnXnmu0t2d+Y7z/N9tq9dy9a8Hkgb3L5k8vz5iQ5b6cbiNU1tRUVFs052PvjgA4499tjqx2tus2zZMvr3709xcTEZGRmUl5fXmjWnsa7zu/pwr50lRZr6vb00NynS1a1cubJN27/zzjut2u7BDT/hXe/lhGI/pz1wJIZuU5JXRlZGDyZOmow/UoEuNQaNHkefXrl8+/knbYqzNfbee+9af2+HHnpo3Npeu3Zt3NpSFEVRlPakhtIoSgPSesWuAk047CQCRUG8GdlojmTLqjUcOeUwhg0dgZBh0tJ9DB87jNycFIxIcYKjVrqz9iriF4lE6g3tcByHo48+mr///e/11p8/fz7Dhw/H4/GQm5uL2+0mOzubwYMHV28/Z86cRve3K1xxbmo2oKZeg+nTp8c9pl3d8OHD29zGBx980KL1A4FShJ3F/446naP7DiHX1ujXK5eJ48Yy/dSTGDIsi7TkVA496gz8IT8jR0ziiJOmd+h7ZMSIEUB835e2bVffr/pcUBRFUdqJ7GS3LkxN16soDTj4iFNwJ3lYu2QJQ/c6GCtQjmaYaLpJYWEZ+Ldz0FHTWLRkNd5UD8KdRlJSBpHybWT2GozQDBzHxu3x4EtJoqy0HLcBE8f3Z9P6PJKTXJx01J70zMlO9FNVupC+ffuyefPmZq+/s4/3SCSCEKJWYc2GPPbYY1xyySUADBs2jBUrVjS6btXJv+M4za6N01Id9ZVVWFhIVlZWrbiPPfZY3nvvvSa3/fXXXxkzZkz17y0ZjvDuu+8ybdq0FsXa0ul6d0Xx+PvLzMysV0NjZ+48+RQ2b89jfgTunHgwg/50FYP7pXPQsSex+7gJfP3Z2+Tk9GD7lvUkZw5HiBJs4SUtJ4fjDtyTSy6+vM0xN0fV50AkEqn+vS2eeuopLrzwwuq2dF3Hsqw2x5l4kr33OJjV61ZiaAYet6f6uQXCAfaeuC/RaBQpZfVwxaoZuHRd59LLL2H6KVMT+xQUZRfS3c8Rq6fr/aiTTdd7dMdN17t27VoGDRoUt/ZUjxFFaYAug2Qke+k7dBjhsu3Y0SBpFPHz7I/IzPSRkdOD5SuWkZqRQt/Bu5OVmYtpGKRk9Ma2bWw7gs8jGNw/hx5pSQwb1ItB/XtRWhElPScL3ZPEk699zz//M4tb7noq0U9X6SI2bdpUXdwvPT29TW25XK4mkyIAM2bMqN7nzpIisKNoaFMnVp1xOE1qaioDBgxg0KBBCCHIzs6uF+f777/frNl1qq7Ct5SUssVJEaXjFBUVtWj9+ds38V2gjNl7TyElqjG4Xzq3/uNJNF2nJG892T1zcBsOpkunoHQzLlcyUVvg0t288sZ3PPnsE+30TOpraBal1nr66aeBHUPqukdSBK646AaWrVyMZUcJRQL06dOPkvISRo4YjWVZfDfnGyzLIhwOo+s6uq4jhCAajRKNRrn/Hw/w+istr1OjKIqiNGzo0KEccsghvPjii4RCoTa3pxIjitIAHdCJIhwbJxxGhkspCMDoCXtjh8qQQlCwbQMr53/L1pVzCZVsIiUtjeyeueT26UXfPrlkZvWgvCKAPxCirCJCcWmI/IIAefkBCovDmJ5kghED0gZw+2Mf87dnZlXvX13tVZpSXFwc15OZjtaZkiOWZVFeXs6GDRtYt25dm9szjNrlu2oWYG3ovZ2Zmdmt3/OJ/rdO1GurO+n8Y/Th/L1kA3v95SocYNu2Lew1djgpqR7SM7PQTTdD+w7E0N0EQkFKSgowvG5y+vTkoy9+7JA4f/vtt7i2V/X3353+ps897WJWrlqBZUdAxv6mN26IzVDz86KfqpNAi379Bcuy0DQN27bRNA2Xy1X9Hnjs4ScT+TQURemOEj10JoFDaX755Rf22GMPrrnmGnr16sX//d//MXfu3Fa31ykSI4899hiDBg3C4/EwYcIEvvkmcdXZFcVxHHr3H0BZeQjpyNhVH92DJkGXFk4kgEuGSU1No9fAfoQjfpb88CmLvnyNgo3LqCgPECgPUFFSStjvR7dC5KREGdXfzV4jU9hvbCaj+2sUrZjNh8/dzgdP3cT/nvkrP719f/WJbqJPJJSuoWoq1a6k6m+7M8XdEa9jQ+0feOCBSClbNDyjuZoz9XdzHHjggfXa0DStwbZt247LPuMtnv+2ze39sH3dr9x3zPHsZyYxJakvj/zvCzQgOd2HrklSvQapbklqVga2LxnHcWGaGiN3G0lpWSklRSsxDZOrrr8ubrE3ZuTIkaxZsyZu7b3++usA/Pzzz53qfd4aL/77be646X40DbZu3Yqm6YCDlOAPBxBC4DgO/fsORAiBZUVYsWoZ4XAY0zSr3y+apqHrOlJKph9zZqKflqIoSrcwZswY7r//fjZv3sxzzz3Htm3bmDx5MqNHj+b+++8nP79lE2MkPDHy2muvceWVV3LTTTexYMECDjjgAI4++mg2bNiQ6NCUXZSmaSz4/kt8XgNdRElPNUlJScFxHKTjgNQIhm2QNrph4M/fzoDhI8jK7c2GZfP4+pnLWfTxw4iCxaz/+kny5r3ATRefzgWnn8DZJxzJ9MMnc97Jx/P0o3/n11/n88B1Jjf+6VyGDvQyb26saGVXP5hUOtazzz6b6BBapOrv+69//Wu7tL+zk/HGEgY1t4lHfZSdFcu1bRspJbNnz25xux2tJRcq6vaUqemjjz6KRzitEs/kTEVFRbPW+2TWT2zffwBJj1zEfyJ5XLYpj2A0SlaKh0hFiI2bi8CVRN+cHnhcJnYoiktzs3XbVratXEt2Wib5+WF+Xta22XSaK55jtPv27QvAHnvsEbc2EyE/v4hn//UvVq5axsZNG1i59leE0CpvIKUVOy6QEqq/swWWbbFy9XKysrJITU2NDVs0TNxuN7quU1pawmknnJfIp6YoitKtGIbB9OnTef3117n33ntZvXo11157LX379uXcc89l69atzWon4YmR+++/nz/84Q9ccMEF7Lbbbjz44IP069ePxx9/PNGhKbuwF195nuW/LiEaClJWWsrGVb9iamFceoRRo0di2xaRYJBP/vs033/1FpqmYwXLSU718YfLZnDY/n2Y99Ms7rxoGUefcRFO1MGJBrEiNr1ze2NZYTTNQFgO5cmXUlBQzKqVxSxduhRQiRGlZc4///zq+w8++GCjQzYSqao3VM24brrppha1MXny5OreCo2J10lw3YRJS3s/NDS9ctVJVGPTGXc2bZ3itqYLLrggbm0lUkZGRrPWO2jSYDLT0wgUrQWfBwZk8cgTL1FeWoQtvKRkuNFsB83QsEN+hNsk5K8gLSmZnOwczKRMcvtI9FD997HlODjSasXNbiDS+try2dFdjt2klFxy4ZVI6bBl6xbmL5yDlE7lYw5SOgg0IPY+37B5PT2ze+3oEYfk2++/ZuHCWK+ZgQMHYpomlmVh2zaBoD+Bz05RlO5FdLJbx5s3bx4zZswgNzeX+++/n2uvvZbVq1fzxRdfsHnzZo4//vhmtdP45Z0OEIlEmD9/Ptdff32t5UcccQTff/99gqJSdnVSSgYMHsbkA/dn/drVbN+WhyfJSzQUBpfJvO++okfPHqSlCDQJLpeJv6wQt8eHlJJN67dx7LRjeOaF69juvZXtqxZh20G8San07TsIU7fJGjOGyYcfRShssfDX7UhbZ9Skw1ixah1AlzlxUjqPzpYIqcswDGzbRtf1WstbGrdt2xiGQXZ2NgUFBbWG5iQnJwNw8MEHN7p9JBLB5XK1LPgaWpp46ez/Lo15+OGHufzy+M2KsmXLlri11VLxHs5TNWNRY1Yu/JHLL/4rjzx2MeFIEg/d/kfWBrL58ao/E3UM0lLDRAIWyRlpbNpcyta8CkwTjCQDPSkdLZDPlnWbcbvKsLUsKvylJPvSajwfEOjVU2vHDkRlg0nD2PCO2PAfXQpout4yEPu7ff311znttNOatb7P52t2b5rObNkvywlHgkTCNmXlFXg8HtauW0PUClF1wC+RIAWabuBYEaqSI+FIGK87iVAkiJSSqBVFIFi0aAELf/kZQzdITU0nEChnyOC2Tx+tKIqyq7v//vt57rnnWL58OVOnTuX5559n6tSp1edRgwYN4sknn2TkyJHNai+hiZGCggJs26Znz561lvfs2ZNt27Y1uE04HCYcDlf/XlZW1q4xKrueE044k4jl0LuHm6W/LKHCX0pmdk/S0rNwuVx4fW5KSkrxJeeQ6vHhcafwzcevcdjRJ9JzwFAct5cPZi3ntN/9kfc/X0t6RjY33no3jmMjK68Yl5dXsGTJUjatWYu/tITpQws5emSA0172cu+9dzdrZg9F6WpqJvxamzCoSqwUFhZy5pk7xupfd9111VNkfvnll41ub5omUkq2bt1K7969WxVDV+Tz+apfn6ZUJbG6kw0bNtC/f/+4tJWdnb3Tv99bLruFQ3oN4Nwr/smHL92NmZrNmsf+S9QFm9evYPVaSY+e6WT0zCTiRCi3HAwhEUJD4hDw+0nOzaI032bQkFSuufmvPHn/32vtQ0qJdGwWLttEqjeNAf2TsaKxop8CKnMlGkgHKW0cBx5/bxmXndr84S2nnnoqp556autepC5oyfzlBPxlFBYV4Aidq/90Gf93wUX0zMll8zYXUtasLyMre47s+K4uKStm97F7smz5UsLRcCxZEg2jCQ3d0KkIlTNixG5s3y74/NsPEvMkFUVRupHHH3+c3//+95x//vn06tWrwXX69+/PM88806z2EpoYqVL3BHBnJ4V33303t912W0eEpeyi3n33FQ497ARMVwYRK0JFOIhWVk6kvBjDcCNdyaRnprJ9m4bQBJMOPZy5X37G17PeRSAYOnQ0A4YO5GTfJ4wbEuL0x4cjHElu3z4s+W0xE/fYgwl77096sofPl//CiaedzuKtm/l89lbuvOtCQBVfVbqvePSgKC4uJiMjg1deeaV62d/+9jeAZvcGyc3N7TYJSCkll1xyCY899liT6zXF4/F0u6QIQL9+/bjmmmu477774tJeTk4OeXl5DT52WIaXRY6NNzWN12f9xoVn9ecQ0811G1cjJfiSfYSCUYrLghCxKS8sZeSo/mzcVEifHA8uXxL+vDLCjo85c+fRo2cDB3tScsuj3/P+r6W4DB0XDiHHwTQERUE/PZMceif34IfVK6B0PSklPSn1uVqUGNnVhCzJ9z9+w5V/urp62d///jfm/ryEdRvX4jgOkUgQR9qVRVV1IIKUNhCbmnfh4p/RtVjyNjak0UY3TIYMHMa6DWvZsj2fBx56ODFPUFGU7ikBs8E0qgPjsCyLs846i7PPPrvRpAjEjgt/97vfNavNhCZGsrOz0XW9Xu+QvLy8er1Iqtxwww1cffWOL62ysjL69evXrnEqu5533n6eieP3wIqGcZtuQhEI2hLHKgMnD39REh6vj1FjxtK/p485jmTfg6aiaZCcnESvnj352jqPzzeHGDKyiILtW/nm+68YNmgIfQYMZuWK5WxYt47dho9g8uT9eP+ha5m3LNQtTtIUpb2lp6eTmpraYI/B1NTUFrXV0uRIZx0a8+KLLzb6WM3pgpt6rjV7ZO5MZ5xdqCn/+Mc/4pYY2Vml+2e35DFsWA4z5DCmrw5z493/ZtSYSUwc0ZvZP28hLTkVofkJaQaBiAQJGzZtJSo9CGGSO2QI3875kT+ceBDLV/bntw2b6u1DaBq3zNiPW4WInajjAnb0aKiujSMPJRqNEo1G2XTFRCB+w6O6EyklkXAZ2XWGSOWVVLDPXuM4YPIz+EOSmdddzfaCjXhMH4P6D6b/gIEsXDCfQKACEBiGSWpqGhnpGaxes5KIFcG2I4watyd33/cgWqgCw+1m6YI1oElGjRvS7CH53SWRqyiKEg+GYXDfffdx3nnnxa/NuLXUCi6XiwkTJjBr1iymT59evXzWrFmNFklxu9243e6OClHZRaWmpvKnP1/D9X+6ik+/+IpjjjgaW0YJE8VxbAr9ZbiCAc77v/PZXlCEpgu+/fwDBvQdgter0d+/itE9JHO1MwlWlDFgYH8W/vIzq9esIWpH2X3CBDJ79KCssJhrL7+cZI9XHfAoSgu8//77HHjggfWWFxQUtLgtXddr9ZLY2QlIQ0VVO4PS0tJG64JUPZedzZQDzZuBRn1ONW1s74HclJpL77HjeWblYm7uM4Cxf/srHl3w++MPIDMjg0+//ZXt+SH23H0A61b9QkFhBX0GpePoDu+8P4u/XHkV69dvJKhX0KNXw8m+qlmADBH7KWXs546ZUiBiWWhS4ugWweMndswL0EXp0iElJaXWshSv5N4HHmDmDdfRM8vHf559Gt00SfKlk5qWhqZrbC8oJz+/iFAohCc5hX59epDkcREOhgiWFxONBAlFo8hoOVIIxu81li3LV+HLyGbjyi1I20b3GvQZmLvT+CSJKmuoKIrSOR122GF89dVXcUuOJHwozdVXX80555zDxIkT2XfffXnqqafYsGEDF110UaJDU3ZxF148gwsvngHAtpISpJT0zcokGg4TjkawHZucHmksXbGWcNBP3169yeyRiSfJgz5iCv949XWSkr6jwu9n5aIo6d4UNMNH1ErhlwWrWL1mETqQ6vEyfi91wKooLdFQUqRK1cl7IBDA6/U22ZZlWS0a0tlZXXbZZVx22WVAwwkMTdN2mtTZ2WvamJr76YwJo7o2bdpUPZ1se/l75lDEIzezZFUef5hxAgCDvv2IleuX8tS7P5CUZHD2CfuwdX0pm1avxMFARCoQUqe0oJR999mLQHkRpZaNGRa8+0LjvYGg9usedQI4jollOyx5/ilWPP4v/pd8GANT19I/fRN7Tt9JQ7s4lyeZ8pLyWsuSPclcf9kV/OXmGykqruCBBx+ld490rHAFRUUhTCFxoTEoxwPShy0cCBQTCQgsxwYp0TSDaCTEtdf/mfKK7fxpxlUcOvUIVqxYTk52Lpk9MgiWVbDy19UMGz2k0fi2//Y8uaOa1x1cURRlV3D00Udzww03sGTJEiZMmIDP56v1+LRp01rUXsITI6eddhqFhYXcfvvtbN26lTFjxvDhhx8yYMCARIemKLUIIdhcVAzAgbuP57flK5l5w1/5y19vYta7/yMcDvPjz3NAwuxvvqLqfCE2eZWOqRloWhQzEkY3PCRrOo4msKTDx199nbDnpSjdQUM9OZKSkqrvn3DCCbz99tuNbi+lxO/3V89sU7Wsq534V6mK9a677mrxtMiNaSh5VPfxzv4a9enTp0WFaBuzs+c57vvXWOjczLjhvZCFYUSmB29qJknuVLwpqWgEee/H1Qzo0YeSihD77zGar+b8wIb1W+iX05fNpeVkTplMf9Pih23NO0yzLCv292obfH7AeLY4Nkv1Qfw3+yzcmsMm4XD/uU0nCXdlySnJhCKBWsuELkCHu+64C7fppjQQ4JIrLqKgoBQ0m9S0TMaNmcj48buTkZ6B49hs3rqNeT/N49eF8whaIUxdcNVVV/LPe+8hEPQTicKI3YYzYrfYzDRzv/qJnr16U7y9qrdbw31Dohv+i9ztHIRQs9YpiqIAXHzxxUBsdpq6hBAtrpmW8MQIwIwZM5gxY0aiw1CUZvt64S+MGzyU/MJCFv+2HoRGIFCBBmgidlijETu00QBdd+iR5eO/H33KmN0nVLezcN5cdp84qfr3rniVWlES4b333uO4446rtWxnJ+bvvPNOk236fL5a27/99tu89NJLvPXWW/XWrXqftnaa0o5KuNx4443ceOONhEKhJnvP1B1SVFfV51NXHGpUU0VFBX369GnVFMKO4zT5Ge3LyuHgyQcy+63/krS+ALHfKAqKNxNxHLKTfWjSQLfCSL/Nui0FrFy/Cs1wk5aejeF2kLbB0pVreOe991i/eGm99qWUODX+DYQQSEfy4tVX8OPcYj53TUUm9UV3SVI0F44R4H83g0Ua2365GCnBcQQ4FmKDzZqX12GEyyBiUZyazmEvvIduehucNr67fj8JITA1E0OrM5+x0HAcBweJo2tkpqfz2AOPobtc6JqO0HWcaBSJiH3v6zqTtD2ZfsxUHBx+WbiAR598ir0m7oNjWziAptd+j006eC8Wzl1CKBIFYN3KVWhuL1Y0ClJj8JABIMAyTTWcRlGU+nbR4qvQ9BDhllJpZ0VppUVrVpGWms47L7yAY4WROMya9SkBWxK0JX5bUmFLymxJccRhxdbCWkkRoFZSBLrvQaeixNuxxx7b7vuYPn06b775ZuXsEg3fWpMUSQSPx9NkwqJuIfSmCCFq3aDr9KrZvHkzUkqefPLJRtcRQnDMMcfU+vduzme0QKL5kjj6D+fzwpoNvPDGJ9j+MB6XRnlxPsV+i5SkHLJ6e0hJ9aFJjR7pOaSn+LAtg6zsnixfXkjf3IaH/OhCQ9d1NE2LTc8rBF/+/Q7u+SWFz5JHgZ6F2xAYuEFAsualwglhejzouommCVwuDbfXDYNc9LtxKCFdoLnc+EpKCJcVE5Z2l/m3jAeJxBESt6t2DTspBJojERKsiM0Vl1/OlTNmsKXAHzv+F7FaL27TwON24TbdeIQL3RBoOKRmZNM3NR3QQEocoKE/oXAkSigSK3w858v3+fztF5n7zftU+PP54IM3AXBsiw//E5uVservUVEURYkflRhRlDZYtnEtm0uLOOzwg8jzBzng0CmJDklRdlldtcdVZ4k5Ozu72evWPDGrmThoTLyv6sTLH//4x0aTXo7j8P7777e4TSHDCMsmLG1eeONfvPz6E5SFyzE9JmmZqXhNk/ScNAqKA0TKg/i8LpJTkoiE/RhuB93lYUvheoaN3r3h9uv8vYQqyrjkQwsR1TDCUXQ7RNiKIqVNJBpgv37r8PuDSNtBEzqG4QI0hNBxmy5cmk6/P48jYFpgC+acegx533+yS514CwShSJicXjm89eKr1culZeFIAVJy65+vYvyeAznuzLN4/LYree3RGViWAwIcBAgtlmCxoghHYjuS4pIyTj//dL77fg7CkhgCcOp31ratKJFoBIDT/3gV511xHSmZPVi68CcKiwpASu785ByemjeUU2a8wmmXvsSZl79MqT/UUS+RoihKpzR79myOO+44hg4dyrBhw5g2bVqzisk3RCVGFKWNDMPk9Q8+TXQYiqJU6iyJhq6opcNam3PyPHbs2F3q3yRsWdi6jWY7FEfKKa2wyOqRTYYOVsTC49LZY/fd8Jf4Md02UujklxTi2AahQBmB8ihEIjz99zubtb+b9j4OTerYhgtbaEhdR5cCaQUxbcmBe2wmyeMjasVOogUCXdOwLQvHsTF1g+RknSE3TiTildhIVvzlWsqjESzLqrWv7pws6dMvh94DRuD2+vj8o88BWLx0BRqCcn+E3QclES7awKdvPMuQEZkc/YcHmHnFuUgp0BBgOzhSYusOjiO5+rLLyKxYSG7vAWxc+gNLFy9BkwLLrp8kNHWDSDBS/btAcMy003FEFBmWSKBCujH0NCzDR1gkExRJnHfdR5xx5RsEQla9NhVF2UVI0bluHejFF1/k8MMPJykpicsvv5xLL70Ur9fLYYcdxssvv9zi9lRiRFEURekWqupbdMWeI9FoNNEhAPDoo4/u9LVr6Ynx73//exYtWtTWsLoUw3RhCQM0jRRPGqYUfPu/d/n80y/wmUFMJ5/XXnmXilA5wuMizZuKv6SUjCwf4AHK0IXANMymdkXhxo28l74PEgvhRBBU9naxgghN4EuSTBjbF69bYJomjuMQCgcIhSqIRAI4jo3jOFiWJCyC5B43EBmJYtkmcw4b12BPn+6aHElLTyEaDpGVkY0AZn/yFXmbN2NrFkkeg6BLI9+vc/wkDytXFLJq1atsrbD4/NOZRB07VmdESiQ6kSj85dppJOWO4IvLryV73ircTgBHM4hEg/X2rWs6jqz/Wo8cdQjJPgMhICvVTa8sH31zfPTtlUKvzBSyM5IxXCmcP/MTpN09/10URVEac+edd/K3v/2N1157jcsvv5wrrriC1157jXvuuYc77rijxe2pxIiiKIrSJVVNw9bQiXzVyVvNx6pqYVTVZqiqz1D3fmOJgYyMjHp1NVp7i5clS5bUa3vIkB1TfrZmv47jkJ6e3ubYpJQ888wzbW6nqzENLxmeJEwDZr/7Bt/Oig3HEULwybsfMev9WRx9yF4MyEnChZcyTSNr0BD69RtCVnoy/lA5tt74SW7NxMTTl96KEwWJhsBECg0pQDixk+3yUDkeI4TbHRs6Y9lRBE71FNWWHcGyIuBIkvQksvcdQJlpEgpblEQljiM77TCo9jBuwkgiAvr2G4yma/T3lSOlhtThhDOvZ+3GEh77vIwjT5vOGy8tZ0j/TCZM+jNC03CwEdLGkTZg8+Z/v8KbkoQc1JvBl1/EgIn7xv5t9PoJL0c3CVv1EyaFhcXEXn0Rm42matgOAk3XME0dX5KJz6PjD3aO5KqiKEpHWbNmTb1C/BA7Ply7dm2L2+sUs9IoiqIoSku9++67Tc6SUvOxmieUdRMnzellss8++/Dxxx/HI/R6TLPp3gEN+eWXX+otW7NmTZuTL8XFsanJH3vsMS655JJmbzdhwgTmzZvXpn13dVdfcgnnnn7yTte5+U9/bvuOJCzYkIchUrD0JBxNIGwJwsImgrRM3ElpaHoAOxIlHPHjiChCxhKBlmWh6zoIGyF0NAGhcBneNJtohYNt2Xx1/aUcct/juOtcR+uKvbKa66BD9uHLz39ERiHozQYHNAvMJC8DengoLgpx//3/on+PdEQ2eJKTEZW9PRx0ygN+PC4D05fBo08+Sx+zB9n9h/Hw/dcy47K7GDN+ZK39rV+7Hl2DY6cdVS+Wdat/IjPdW9kTzkHaooHirbF58O559Bv+et1h7fOiKIrSaQk6z2xVHR1Hv379+Pzzzxk6dGit5Z9//jn9+vVrcXsqMaIoiqJ0C4118a+7vKEESV1HHnkkn3zySa1lH330EVJKPv30U4444gg+/fRTjjqq/slMSw0ePLhV20kpOfvss9u8/52ZMWNGrbojjuOwceNGiouL0TSN9PR0+vfv364xdDVNJUXiRSIpScpFhEDIEFhm9dS9OhAJVBCVHkrLtpNkurGdMLZto+tuLDuEoetIx8DBrrxvYUXDWIYHm3JwG5R+PhvNjiAMs9sOoalPcMhh+wDw9Q/fgpAIAbqmccvtj+JokqceuJELLr6RzOy+CC02fGnWJ6+xYWMB2b1HktWrD4dP/yOfffQvUt0ruOvOazhh+hl4TRcDB9Y+WN+6cTvBMKSlp9aLxLFKENKHEALHcZCS6mmUpYwldkOWQyAYQuT42v+lURRF6USuueYaLr/8chYuXMh+++2HEIJvv/2Wf//73/zzn/9scXsqMaIoiqJ0efE4aat5FfzTTz+t7o1SkxCCI488EoglTzZs2NDmxMDq1atbtV3VCVJH0jSNAQMGMGDAgA7ft1Lb+nk/scFKxZF+tKiJwMFxbASgCQ3bjmBFysgvyKd3Vg5JyZmEwhUIITBFKqbpIhoNI4SJEBLHtikqKiLrqCzyn6rAKzXM5CiOHatN0l17iOxM/wF9EFIiBXz3w5c4+Dj04IO44oYH0XUNiQNSIDWdKUedxsq1a7nz7n8SDJRgRy3GjpUs3+YiNymJfffeH5/Hy/Ily9GFhm1FiTo2pulh0Ije9fZtOxJfUirTTv0DAJYtEZrEQaAJgaF5cWyJl3KSw6u47dKLOvrlURRFSaiLL76YXr16cd999/H6668DsNtuu/Haa69x/PHHt7g9lRhRFEVRuqTOcKLWr1+/WsmTlsbUloROVlYWhYWFrd5e6dq+vPF6QoHhGKYAGUWTBlJG0LTKWiIOOF6b5KQUSkqLcISOz5e8YwgNDi6XC9t2CAQrsB0HtyeZlJEm28wNCBvCwgW6xLZtDGPXO2Sc/c3XHHvksTgC9t37ANxuD0QjfPP52xRFvBx1xNGkpCWj4SClztCBQ0nSyhmdm4Se5AZ/hG2hADfedS8CQVZ2Nm6hEdU0SktKcKJRXF4XPXvWnipbIvnva4/gcvvQK1/3G/Z/GE3o6JqBg8QGEALh2ERDYRxxVWU39liypjN8PiqK0gFk5a0zSEAc06dPZ/r06XFpSxVfVRRFUXZJJ554Yr1lbe15UjUrzvjx43e6Xl5eXpv25ff7W5UUEUIQDodbvV+l8yjbtAW3y4sh3ABIzUboOlKArmuEwn5EpIT3Z+ezcUse27Zvpqh4G1IKYn96gkg0RCBYDDg4TgiPy42mGSS7BRKJYcP6T9unrk5XsNee+3DPAw9gO05l8VOB4XZz0JGncdKJ00lL8cXGtKAhhEQIyf0PP8P6YBmeVB9/mHkf195wF++8+1/C0Sj+qEN+uZ/NW7YzZo/hTNh7DGPHDa+1z3fe/DfPP/Q3AmUlCLGjp46m6UjdhV1ZgFUTsYN46VgIGWLJyxP59c0DWf3h8Sx47kAcJ0LnOVtSFEXp/FRiRFEURemSAoEAAIMGDWLbtm3Yts2KFSvqrafrOk899VR10qLq9tZbbzXY7tatW6vXaa2FCxdWb3/kkUdWt+fz+XjiiSfo0aNHq9sGGDFiRKu39Xg8bdq30jk40sKRsZlIdF1H2jqOBMtxiFohdOkQLrZ59M10yqIam7YUsGbtRjZvWU1ZeSEV/mKCwfLKRInEsmyEEBiaw8A/DMeWoKHzxY1/qqxvseudZF96+YVcd+Xl3HDjn/j061/A0dE0A1uPDa9BA6HpsZ4bgCM0Fi1exfTTLsbXczAl2zeQ1TODY6aeyIwrL2fFr0sZt8dQ9p08rtEeHdGiAJruxyV8nHzWjsLHrmH3kTnuIdJ2fwT3yPsI5d7COnke7328GWmFWPZbgFWLi1j5yxqkLGXDF6fhWFbHvFCKoigJkJGRQWZmZr1bVlYWffr04aCDDuK5555rdnu7Xr9IRVGUTuTDDz9k6tSpiQ6jS/J6vfVO1oYNG9aiE7j2PtkLhUK1EhEVFRVtbjMcDrN58+Y2t6N0bUEzAxmNxk7SbRvLjqJJHWk7aKaJJzUTw5+P4/LRu4cfU9iUlJYTCIQIBrfQu3dvTNPEcSSRSARdN5DSIhKxCeYKSssDmMluzCQDxxIIV+0T+e48M01Np5x5Eobbg9vt45rrf8/mjRUcfuRpnHXmaWSmJKEbEl2HqNTQJOy5x3CizgiKR+7Bw48+wcL5n3P89Ck8eO+DTD3uwCb353eKGDJ0PJOPOLHW6xt2ouh2BBMd6USRjoVtWWzN08FI55Q7vkdKh/efuYL1G2YjWUnkg+MYOu0DNKG350ukKIqSEH/5y1+48847Ofroo5k0aRJSSn766Sc+/vhjLrnkEtauXcvFF1+MZVlceOGFTbanEiOKoigJNHXqVEaOHMmyZcsSHYrSDtxuN5MmTWpwyuDW+vHHH9vcxqxZs5gyZUqb21ESJ+T2YocthCOxnRBSCnTNwDRNpOFGGh70rCFogVIM00Ov7BR8Ph8VFUFMU6OsrJTk5FSi0SiRSATDiNUo0XUTGTZJSzaRQpKGgx4owXJl7JJ1RgSCE086FiklHvczeFxuDF2nPFjKu++/xLff/0D+9kJsR2K6PeRkZ3LgAftw0IEncfMNlyPtS0nJSGXcuKFN7ww476KZDS6v6nXmOA62HevdIwC/kcmWgjx2ExoIjRMufBTHcXj/r7uT0TsaGwKkKIrSDX377bf89a9/5aKLaheffvLJJ/n000956623GDduHA899JBKjCRKJBKlrKSAaCRAakYuXq8nIbMHKIrSNSxbtgxd17Esa5e4AruriUcio6aDDjqozW0cffTRWKqbfZdW4kpHDwsidhAZjeA4EA36MV0mGhqay48eDWKkDsQw3ITDYdyGgZbkwbJtQoEASW43OpCa5CUQiaLrOroucLtskpLdRPyCYDDK3tefx9wH3sROEpWFW3c9QggOm7IfEEtSLPttA8cecz4nTf8jui4QaEgpsO0oFf4gAwbnMmhQn7juv2rfEEuQICUOSRx6waf11j3imi9Y/PLB9CnbjCe1b9ziUBonpcTvz+f5549i0+YFGAa05CvdccC2Ya+JF3D88U8ghKaOCZQmicpbZ9DRcXzyySfce++99ZYfdthhXHPNNUDsAuT111/frPZUYiSOpJSsn3c/afZaTMNFsu5BK4aogDxrBDm7nY7b7U50mIqidEJVVwB31WkxlY5l23aiQ1DaKJiSjRN1oYcEhtvAljZWxCYc8oO08AoN4XZhiQp8XptAeRlpaSkAGFIjKSkFTQOBgZSStNRUHClxHAvdNBh8x2i+mTGXJ8f3RTcsDrz+OL5+4CMcIdA0rdMciCeCEILdRg0AOm7aak2reZJcOZW4BCHr/0sIITC96ZSU+tj286MMPPjuDotzV1JSsoG//2MAeqzMDFUdAoUA02x5e5oWuy385V8s/OVftR6LRuGWv5ThdqfEIXJF6R4yMzN57733uOqqq2otf++998jMzARixepTUpr3vlGJkTgKh8OECr4kKTWbZHc6Ej/hsIXtaPiYT9nCeZRnXcagIcPUiY+iKPVUjdnfFYscKs1TVXA2HqZOncqHH34Yt/Z2FW+88Qannnpqg4+tXr2awYMHd0gcT3/yLP97/zv+fOt/sTSBaUt0VwS3k44dKaeivICMlOHYrjR6+bKRGX1AlwhMkDbRaAWhihKEYeB2e6kIWYQqSin1+1m4rown3t+GZ+JA3C4PUnOQQrJg9RL2HD4uFoA6julwQgikYyFlLJGOroEMYlk2pln7kF7XDALRMI5/aYKi7b5uv8OHbQeQEvbZGxYvBqFBkheKimDVShg6LL77NE246+5UAPr0PoI//vGT+O5AUbqgm2++mYsvvpgvv/yyetjy3Llz+fDDD3niiSeA2NDh5va0VYmROLFtm+Ci6+id1RfQMbJOpmjzv3GiZURtC9sxsSyBU3E3v0WuYbfdRqvkiKIo9VQlR9asWcOgQYMSHY7Syfh8vri19dFHH8WtrV1Bc76zhwwZAsQulLhcrvYOiWnH7s8xR+3NmD8cTGirTdnmEHool+TU/iRl7gH+IqLhTWzYtIGKSAB0gePYVJRFqSi3WJpnsWJzgO3eNNxpmQhNYGkmusyh117ZRGwNxwY7amGaJtc+dgOf/P0dgF2y3kgi7fj70wANTWgYpo6IFPLDj3M5cPJ+9bbxGRqaqXqHxdOtt4laPUPmzYecHvD++x0Xw+Ytn3LjTYK77lQXURRis3J3lj+FDo7jwgsvZNSoUTzyyCP897//RUrJyJEjmT17NvvtF/tMrBpS0xzqW60Nqrq8CyF478Xr2LvfesK2zTlX/czcX5/m+/+dg1sUIBDYjo7tODiajlh9NwU9HmzzdI2KonRPNWd7UL1HlPa0q8ws0hZpaWmUlZW1aBu32813331XfWDWnjTdwJecTNLQKD13SwMcHLkBw16Ho0fxuDSu/1nHlGloLgMZttA8AsfSiZo2xiANr6bhOJXT9Vo20gFHariFRdBwYxg6MmrjiChvfv8uZx7QcI+ZXV3s/QQbV61i1S8/4vV6KQ0U4vWksfv+h5GantXGnjaich8SiUTX3aQaQT54/6N6iREpHSyhE7HVd0g8DRkMq1bXXuYkIPe0YX3H71NROqP999+f/fffPy5tqYqgrbT02zdZer6b/Dv7U/KvIzjBWUSvzQ4lBRuZt3Q7EaKEyuYRtSykI3GIYskI0tKIhEpZ9cNtscJZiqIoOyGEoKSkpPr38vJyhBAd1l2/s9M0jdmzZ9dbXnXCr2lai26dPUlw3333xbW9uXPnxrW97kYI0eKkSJV4Hag1RQhIy0wlyaXh0hySTDcpbheax8AwPei6joaBo+lEozaWJrAsDTSByzBxpIxN8YsGUuAgcATouo70uNEr3xvS0DBdBk9/8G80s3O/TxIl9vkh6Dd0GIecdDb7TD2JI0/+IwceexqpGdltS4oI0JCxs3AnluwQmkZ2mkk0WlIvie44Nj4jhCtJfVfE05dfQFIS5ObuWBaOxgqn9uoFe02CSXu33/6FgHffgZdear99KEpXsnr1ambOnMmZZ55JXl4eAB9//DG//vpri9tSiZFWWvvKP9hmmfzw9CYi6+dTvHkh5Zt+IXXTdj7+m49v/tkbEdWACFEZRkgQjo3tRJFaFDu8kdLS0kQ/DUVRuoCMjIzqwqypqbExxmvXru30J/Ed5fjjj6++X1XEVtf1Vr0+Qgj8fn88w4urq6++utEhDK15vueee25bQ+q24vH+ao/3aENtfnrrKxiGEUtgECEibUzTRNM0DN0N0qhMFOoYhomuGwihYdsOIKpnzqvuQWS4Y23ZBkKAZUXRKi+LSyJxf05K03RNgtQRugspJIamYxgGPQbtScGG32qtK6XkjVv3ISXNoPeEmxMUcfdk21BQANu2gccD48bCXhPhgAOh/4B2KL0jITMDtm+D116Dp56Epb81vZmyC5Gd7NaBZs+ezdixY/nxxx956623qKioAGDRokXccsstLW5PDaVppfE3vsX3B/flyw3g/0cRh1ziw/CEyEkbSE/TS0n+BgIrf6Ng7yEgdBxLB91AOkEc3YVtWWzevJmMjIxEPxVFUbqIhqbJFEIwYcIE5s2bl4CIEk9KSVlZGfvttx8+n4/PP/+80ZPR5g4biWcdj3iTUnLppZfy4IMPNvhYS61YsSIOUXU/8UxobNy4kX79+sWtvYbouk6KO43ycClCaLg0sKXE7XYj7QiapuE4Al2PndgJ4aDrWnWvqtjfTmymmdiQmthwYce20TQwhI6mR5BCZ/KIQ1RStplW/e8gFi8s4viZi9C0Ha/ZxdOGM/m4/2PYqIn06t2T5OTU6po0/mCQ4uJiVq9Zi98fxOPxMKBvLinp6QQKZpOaOwHN1hC6hpDQa+ThZPxwB/6KAMkpsc8uiSTZEyUtpze6Lzshz727+s/zkmv/tOPfcu062LwJ0lJjM8oA6DqMH5/LxInTsG03W7Z+x+ZNSykqClFeDpYVG3LlcoHHI3C7BLYD5eUOW7bA2rVQUgKRMPgDEAiC3x+b9cZxYjc1zFZR4Prrr+evf/0rV199da2ZZw455BD++c9/trg9lRhppV49eyJdBhNdFitDkP22H8MDux27Ac2TwfY521g8W5DVbyOOk0xURjENL5oukcKHrYeQW1bDmDGJfiqKonRCI0aMYPny5c1ad/78+bvsbDaO46DrOnPmzGly3Z2dzHWVWhtCCGbOnNlgYqS1XnrpJc4666y4tafU1r9//7i+Nxv7W/3fna9yxPXHxqbHQOAQBQwsK1Y41TRdODKCdEAzYvXRqnofxZIgGo50AA3hVA5FEw66piOMMNLREbbgtrNujNtz6c6kBNOVjNdTxBsPnM2pV7+IEIJFS5aQ6nWhp2YgNA1px3q62bYFCETl2a9EYNuy8jEbTUq8Zhah0jyEkQaWjcvjRktKITnJ4vlXXmXGH/8AwBfP30iyz0KaLnbtiZXbxz/+Hns/3313rBCrlFS/ylXT9i5cuJUFC56sftxxQDoQtWPJSdsCy44lSaIRSdSCaDg2LCcSjrVVtW3NK/Hbtu163/OK0pjFixfz8ssv11veo0cPCgsLW9yeSoy0km3b9NjDR89BZQwvlJQXQ9lW+PDBUkJOKVjQ6yhJ/pZyrGgFthObOk26vOh6AAwvvpHtX7G+vUkpq7vgejwegsFggiNSlO5h2bJlQMuuXDeWHDn22GN5vyNL5negqufclqRG3destLSUtLS0tobWbo499tgGl7c2OXb66ae3NaRupaGeWW0Vz8RbVDq4RMMxfnL3e0y5cToaDh7dhSSCy5uE40ikFOiaB80EISS2ZVcP0YuduDmYpomUsaEyVcPRHMfBcLl54uJ/0qtHf3Q1CLtZhIDsvR8ndcN0ilYtrF7+zTff4fE4eHUPEonjWEgpK2c6kbEkleNgW1Z1UiQaiRBY8S+c1IHoUYmr94E4MoLQNFzCTbLbw/p1G6r3kbfsPUYO85Ez8dEukfDtqvLzYchQSE0DXUBxcWWSJE4vueNAWRnkF8DmzRDovKM8FSUh0tPT2bp1a71ZHBcsWECfPn1a3J5KjLSS2+1m8rN5FBQUsHz5cnL+cyyhjQFyimKZ3mge5A/VEI6DYejo0sayLeyKMMIdRfdA796xrrVlZeX485eR1XccLrc7wc+sZaqSIkIIQqHQLnvVWlHaS9X7qbkHt3Xfg9u2beODDz7ANE2i0Wi7xJhoNRO0rdm27v2kpKS4xNVe6vaOaevnbnskArqy9iiMnpmZSXFxcVzaMoWGZVnVtXRqEkLw2d3v8NhrT/K/Xz9CODqODKPpAoG7cjY9cByJYRi1/m40LXaSrusGhuYQtqMcPmIKfz7vCqSM9VyoKlKsNI83rTfeZA1Nl9Uny/369KXYTMIWEidqIUXs3yCWHJFELJuoXZkcsW0cByK2Q0grIT17Lyo2fIxhRwENaYeJOmHM9D61ZjIzBWgeN76Mlp8YKM1nWbFeIEjweGHQQDBM0OucXUkZ6yESjUIkCtEIRCKx7aMWhEOxITOhUCz5UU5svYi1Y/tEzHyjKJ3dmWeeyXXXXccbb7xRncj/7rvvuPbaa1tVQ00lRtrA5XLRu3dvcnNzWT/oV1ZPGQQpUHo0WKbA1DR06cHRQwhMbMdElyC9guTkDAYOGkT+9m14NlxLCjaB4jSiu93Xqce313TllVdW3697cqGuUChKfNm2zdtvv83JJ5/c5Lo1T5RzK0vne73edo2vs6ibIGhq+ExDTNOMa0ztqWaPmZrPp+ZJUt3XQCWvO17NmaXaqmoIjJSSaDQam3WmTrJixmn/xwz+j2g0ypkzf0+BVkKKGcUxBEjQHImOwMbBkZBXWEa0KMAlJ57P76efV9lzZMcJe81hN0rzaZrBuDN+YOlvk3CkRBOCadOmMv+HT7HDUULRCOFgBLc7itC06jNgy4riVL7+th373dN7BrbmwsrYBxwHOxImEgkhHYlv4MH88feXVe5VENY1xp/+FUKoJFZ7cruzkbIAqBz20sT6mojVH5HGjlohVA2XqWqjzn1ZOQTHdmIJVkVRdrjzzjs577zz6NOnD1JKRo0ahW3bnHnmmcycObPF7alvuTgQQtCvXz/mn9cLJ1wM0sHUJBINpI2mpaLpFi4h0A0XhiuJ4dM+xjRNUrbcSmmoiFAkSjBSSEngM/Y+YFqXSCw0VtRmRyE3RVHiRdM0TjrpJGbNmsWUKVOaXL/uZ8hvv3XvMvaNfeY0lqjd2WdUJBLBNM3q+iWdSd24m/NZu7Nk9SWXXBKXuLqLqor2XYEQotEkXtXfha7rvH7P8w1u25z2lbbThcFZd/xc/XtZRZClRXuy9ou1HH14Em6PB9PjqX69I1Zs+Iy0ZeXJs4Nti1gvoYiGIVIpLvMTCkcpKA4yd8EGlqxxcf4fYz2OhYCz71ickOe6q8nN3QMpZ1UnM1o9I4esfbdmXZGq5Ihjd9wU4EoXk4DZYBrVwXGYpslLL73EHXfcwc8//4zjOOyxxx4MGzasVe2pxEic6LqOY2cSLC1AChtTFwhN4AgLodkIPAhTIIxcDjj9E3r06IG56VZkchKpTi6SbVh2FL3oESKRo3B3sSE1dakhNUoivfHWO5xy0gmJDqNdHH744a3arjVjLbuSxmb+qNlzorncbnf1du0xrKIt/vOf/9Q6Ya3bS6Sh3iN116spEAi0T6Bd1OTJk9ut7Y4czlb1N6KSG51PcpKb5IxMLCuHD+c7yJ/LQQRjJxQidjwZjdjY+IBkpCMxNAtEPrG5g0ATEoEG0kRog+k9UOL2JPZ57YoGDZrEhg2z6i1vTp2RRr+R5I7kSHXh1sqfzbko0hrBkiDBkiCaoeHL8mF6u06vSWXXdvvtt3PttdcyePBgBg8eXL08GAzy97//nb/85S8tak/ILn72WlZWRlpaGqWlpaSmpiYsDsdxePYyEyvg7KhMrWtouoMjdWwHNMPF5MvmkpsaJd16H2HaoLkR4WJCZWUUlRZQXOGnx8T/kJOTk7Dn0lx1D7gmT57MN998w4YNGxgwYAAA33zzTbseaCrKrmjdunU4jsOQIUOavU0X/6hvlvY4Cexsr1t6ejqlpaVxbbOzPcdE8vl87ZYsUq+zUiUSjXDm1e9jSxvDdGPbDrZl4xDrbWw7NpYtY/UrsBFo6JpA1zU0YvVhdE1g6uA24OFbjyLFpzIjHW3Bgvf56qvjyMyG5CTwuHfUGKn5dVRVY8S2YrPSWFEIR2K1RsJhCAVjU/IGQ1BRAeXlUF4Wm7K3sBC2b4cNG2DhwoWMHz++1fHeNfwuSlaWtPi70pEOh8w8hKl3TG31vhOhs5wjtpeq53fN22/j7iRlGMJ+P/dNn95hr7mu62zdurXeeXNhYSE5OTnYdsuK86geI3EgpWTT8u8IbnJwpABdIojNIx9FIIktN02b7/6+B7YBKZkmp55zEZpuYCSl4Ek2yJA6FoXV/5idxRlnnMGrr77a5HrffvttvauzRUVF7RqbouyKBg4cWG/Zzq5GtySB0pW1tr5ReXk5Y8eOZf369SQlJTFjxgx+++033nvvvXaIsm3inRRRaguFQnFvU83YptTlMl288s/jOPXC5wgHvViAlFpshiDHifUYcBzQROXFNoEjBGgaEgmWQGgamtfk6btOxDBULZFEGDhwTxwJtKFjoaz8yqo5hKZmDrVm3ZHRo0e3uP0Xz3iRxa8trtxHbF5hBweN5v/NaEJj9p2zmX3nbKSU/E3+rcVxKEp7aOy475dffmlVTR6VGGkjKSXrfvmYLx85FisEsanWiH2waZKoEESlREhwXFEkOrplEypyeOO5JzjujAswowK3y8DlTibJDRuL4lO5vi3aMssD7OjSrQpFKUr7aWi4RNX71rbt6voYK1as6PDYEuWyyy7j4Ycfbvb6Va/h2rVr2yukuLn33nu5/fbbq3+PV6Hrl19+mTPPPLPN7XQH7dGroz2SLUrXZ2om//3XhQgktnS4+9EP+fzLVUgdXKaJpploEmL1Ux0c28EJ2/TI8PDAX08nKzN2NVYNl0qcjIzeOE7t5EVLNJgIqfFY1TAap3JhSwsg/1n8OXY8juRQeSjzxXzKKMOUBl/zDfuJ/VocsxCC67TruNe5t8XbKkq8ZGRkIIRACMHw4cNrfQ7atk1FRQUXXXRRi9tViZE2sG2bb5/bl9L8jUgNLF0QtiT9BnmYfPJhQJR1a9cx542VRARggwgJHF2iC5NDTzyDqB0mHLQIWSZuzUQTGviXAC3/sIoHy7LiNiOD4zhqWj9F6UBVCckFCxZUJ0WklHi9XhYuXMiIESMSHGH7e+ihh5g5cyY9e/as95iUEo/HQzgcrl4mhODbb7/tEkXtrrvuukSH0O01tz5Wa2rXKEpdsb8jgS40Zl42jZmXNbmJ0sm0NinSeIPsKLpao/BqawghaiUwjuKo+MSodC67YPHVBx98ECklv//977nttttIS0urfszlcjFw4ED23XffFrerEiOtJKWkeN61FBduwQ4GSMrwghHkgOOPwNQc0KNoeBg9agLjbt8zNl2vNHGEQNeM2HhDJ0TU1rBlBMuRREUYkOiRXxIy5a3L5WpTYbi6hf+ys7PVUBpFSYA99tgDKSXp6enV78fvv/+e+++/n6uvvjrR4bW7nJwcpJTVJ601P0tDoVCD9ZEmTJjAvHnzOjROpfNxuVxN9vBo8fj8TlbAV1GU+Kl+ezdjut6m1Cy4KuWO6Xxb+xEiOsvJsqLE2e9+9zsABg0axH777Re3i/oqMdJKBVvXM+eH/+FYfhwnhBAaU049GiFBYICwERg4MoRwPOjCxtAMUn0pRKNhQlEL00wFTae0vBBBEE238HkzcAt/h04TWfcKakvUvLpW96c6GFSUxCopKcHn81W/T/fdd1+ysrIoLCxMdGgdorET2IYSz/Pnz++IkJROLiUlpdlDX3bWW6S0tJRAIEBubq4a6qAo3Vj1UJq2NlSngVoz0rTycDpMpK1RKUqndtBBB1XfDwaD9S7wt7QArBrn0EpFv94NMoq0I/QePoxJhx2AFQ1i2TZRO0DUgogdwrIcotEojiMwNSgt244/EiZsRbGtCvIKNxCJWgjNhW1rrMvbTt6mZe2aVCgpKakelyWE2GlSpOZ6NQ/umtONuKSkJG4xK0pHuOqa61i4oHv1GvD7/UgpiUQi6LpOYWGhOlGj/mfXgw8+mJhAlE7lqKOa7mpeszdSQ8ullKSmptKrV6/2CFFRlE5EOjVqhbS6kcofNXqLVE/b67S+x4jTlqqwSpchZOe6daRAIMCll15KTk4OycnJZGRk1Lq1lOox0gq2beOPFNNn2AiEtDEdh1A4BBoIEUbDQOKgaQYWITRpoBtplActJDrYAUrKyygpLsWTko4jgxQHihF4Kd66gZwBQ+Ja66OmNWvWoOs6L774Iuecc0718oa6nDek6sqzGletdEfffvsFvy5ZyOTJB3LkcdPxlZQz5uC9Ex1WXLhcLhzHoU+fPur9W0m9DkpdTz31FC+88EKLt4tE6l+ZXbJkSTxCUhSlE6tbfHWnXyuNHGJXl4iQtRfUGlLTCsMmD2vdhoAdtdm2dBslm0owvAY9R/QkvU96q9tTlPbwpz/9iS+//JLHHnuMc889l0cffZTNmzfz5JNPcs8997S4PZUYaQXHcQgF/EhCCCdIGIGux3pUSMdGMzzoIogQBgIJwg26H11zsz1/E7ZlIfQkklIzKNy+AY83hYJt+fQZPITeA3djw7oVjGmnYTSDBw/m119/5eyzz26w10djtU1aegKhTjiUruinH39KdAjtbvPmzYkOQVE6rQ0bNrRqux49etTrJXn++efHISJFUTozx2ndUJd6iZQaP6sSJbF6hGC1MjGy76XNKz7554w/Ey2JIhrL3DRCIrlm4TX0Hd+3NeEpSpu99957PP/88xx88MH8/ve/54ADDmDo0KEMGDCAl156ibPOOqtF7anESCsFgiUIuwARddCFjuESONJBCB2MMBo2SIHQXKCF8ZdXEA6XIBwXuu4mUJZH8dYiPOk+vMnpZPZwEwqF2bZ1KZort8VTcjVXzaRHY71EahZQbQ2VFFEURVG6okmTJrVqu9LSUt58801OPvnk6mVVxXwjkQgulysu8SmK0rnY9o4kRq0kRytGrNYtvlpdY6SVh9W7n7h7o49tXriZ+/a4rzJc2eKkCIBAcP/u9wOQMSaDmxff3Ko4lbYStOoPrl10bBxFRUUMGjQIiNUTqZr0Y/LkyVx88cUtbk/VGGkFXdcx9BRkWCMcDOAPllNRVkK4vIJQRQnBshLCFWEi/hChshLCZcVEo8V4ktLwl5RQkL8eTVh40g2kFiVv4yoMt872Ddvo0bcv0YzTGh3SEgqX89ILD7c69hEjRqBpGm63u3pZzXHRdQuotsTWrVtVUkRpNttu5fxziqIo7aQttbFOOeWUBmtyud1uhBC89dZbcYhQUZTOxLbjM2Vvre1rTtUrWz9dr2423Pv8mqRruG+P+ziAA5CVXVWyyWYFK1q3I6B4STGXiEtavb2itMbgwYNZt24dAKNGjeL1118HYj1J0tPTW9yeSoy0gqZppA88C6G7Yh9ekQiRcIhAKES4IkiwvJyK8iLKyyvwV5QRLPNTVhAkb/0mLDsKjgchUtiyaiOBgnJSe2TjL1yHN0njh1nfsvvEyY0mRh6+43Y+e+djDpk0nnJ/y2eWWLZsGbZtEwqFaiVDJk+e3OrXY8qUKUgpVaE5pUnX3fUv9jnnch74z5vc8th/SJtyXqJDUhRF6RAnn3wy3333XaLDUBQljiyrfo2Q5uZIZJ31aw6vqe4pUll8tUePHvEKGRmM7WgRixjIQBaxiM/5nHLK29SuG3fTKylKHJ1//vn88ssvANxwww089thjuN1urrzySv70pz+1uD2VGGmlgaOPw/BmoLuT0bxeNM0D0saSNlbUIhyEaHmIQLAcf6CCUDiIv6ScUCSCFQlRuGUdLp+HQDTIhmUrsIWHjWu2MXjKM/Tv37/R/W5cv4Zlq9by0N3Xcdj+revy25BvvvmmOklyxhlnNLn+4Ycfjm3bSCn59NNP4xaH0j1JKRk77f/4dNkahK83/1u4je/WBNl74gGJDk1RFKXDzJw5M9EhKIoSR5a1Y2aaqiE1rVEzKVJ3KI1tw5AhQ+IW81CGVt+32dEdxYOnTe2+wztt2l5pJdnJbh3oqquu4vLLLwfgkEMOYdmyZbzyyissXLiQK664osXtqcRIK6Wnp5O1x2O4XV48ugfDZWK4dAyXB9PlwXTbSK8LQ9fRdBMhoiRlmAgpsSMCiYbH58KJQlqPTMIVIbL3f5SjjjoKfSeFVyfvNYE7rr2Yvz38KCmGTihUGvfn9vLLL9cbXlNziMyll17KrFmz0DT156PsjOTZ939k9KnXcNzMZ+k3egKUF7D1xzfRNs1jYGYSe/br2WQroVCo3rTRmZmZrYqof//+1W00NIuEoijKnDlz2q3tMWPGtFvbiqJ0PCld1UmRWrPKNHv7+qtXdhSJJVycWAHWo48+Oh7hArCa1YxlLBVUsJGNjK/8bzSjW9VeEkm8wRusZW3cYlSUnfniiy8YNWoUZWVltZb379+fww47jDPOOINvvvmmxe0K2cWLQpSVlZGWlkZpaSmpqakdum/HcZj741z8i86tnEvLxnGiOI6DtB0cx67sCicADV1IXF43VlQjHI71ttB1jYISD+NPfJWxY8c2OUXvd288yqqVS0nzevlq3lxWFyfzvw8+qDXDTFNT7raWEIJXXnmFM844A8dxqvdj2/ZOkznKrufKh95nQ0EBQX85DrGBsra/jPyCPCwrTGaSi1CoAidYwfyPGx93P3ToUFavXr3Tfbndbo455hiuueYa9ttvv52uW3fs//Llyxk2rPXT2SmK0n298cYbnHrqqXFts6ioiIyMjLi2qShK4txwwwD69t1AWjqk+MDlBsMArcZhsZRgW7GeH1ELrCiEwnDWmYk5BZt11yzm3DSHJJIwMRmpjWSls5LVrG6yEKtE4sNHT3oyj3lsYhPFFJNHHhudje12DtIaiTxH7AhVz+/aN9/B7fMlOhwAwn4//zj5hHZ/zadNm8YhhxzCVVdd1eDjDz30EF9++SVvv/12i9pVl/zbQNM0Ju09idHHf015zl3kby8nHJFIR4LuwjBNTNPEMDQ0XWILk0C5RUVFkMLiAIuX9yd7/9c4+br57LHHHk0mRQAqSvLI7ZHGW598ytpSkzCp/PXuO6ofb+8PpKphNpqmVZ9kVlXeVxSA6++4n19//JDCZd9SvmYe/o2LKVz6LbbhIV0LkGna2CE/JmC4vQ22MWfOHIQQTSZFAMLhMP/973/Zf//9mxVfzVzw8OHDm7WNoii7nlNOOaW6x2RKSkpc2lSz0yhK95KZOWDH0JdEB9NMU26cwkAGVhdeXeGsQCAYWvnfQAbSj370pjc96EE66fjwYWDg4LCd7cxhDqWUIiv/K6Osib0qSvz88ssvHHXUUY0+fsQRRzB//vwWt6um620jTdPo1asXx594CuFjppGfn09eXh5bt2xly4ZVbFuxgKh/O253Cil9RjFgtz0YPKA/vXv3JjMzE9M0W5TMWLd+BcHyQu751+us3bSRW2Y+REm5yTnnXMQLLzxOe06TpGkajlN/MvW99967Te1KKflt2WrscISRo4djmurPsiu75+ara/2+asNWhvbPZcbll5A5MofjD5nIUx/M45RTz2XKpN1qrTtw4EDWr1/f7jHWTI7cdNNN3Hnnne2+T0VRuq663XUB9tlnH3788UfS09MJBoOEw+Em2+lMV1MVRWm7yy//kKee3JE47Sr98G1sRjKSJJJYz3qCBKsfE5X/Vd2vqSqZIpFkkYWDQymlODjq803pMNu3b99phwLDMMjPz29xu+oMNE6EEHg8Hvr160e/fv1gAvVqc1R9YLTlgyOn3wS2b9/MKw8/zEGnnsTnn73HoYdO5qLp5/D7C67h2X/d3+bn0ph4T6/qOA7jd5/Alvwy3C43AodkI4Xlq3+K636UxBraPxeAxx56tHrZC/MKeOSNj3nwxXf58OEbOOCAA/j222/bPRYhRL0ppe+66y6VGFEUpcVq1iJJSUmpToxccsklPPLII4kKS1GUDuR2J2M7OlLaXSYpAlT39IBYMVYdHYHAqfzPqvwvSpQQIQIECBLEj5+yyv+KKa5up2YRV0Vpb3369GHx4sUMHTq0wccXLVpEbm5ui9tViZF2VLeeQTwES4qQQT/77TGa4p9/5oFXnueLL75FSskpcd1T+/niy+/4/J2X2FQSpE/fHFat30ZZaRlJbjcrNy9IdHhKBzjtqCnsP24IxaWx4sGG0TEfRfvvv3+tBExDiRJFUZSWqvldf/PNNycwEkVROtqVV1osWHgz69b+tUP2Z9s2S7dtY1NJCUmGwbCePemdnt6iNhycWsmRKjV7iEhkdaKk6veaPUZq/qcSIwm2ix3KTp06lb/85S8cffTReDy1Z1MKBoPccsstHHvssS1uVyVGOgkpJVedPY3FS39l4MBh9O3Vk8l7T+CqO+5m7KABzFu9nqHZGWSkp7C9pASXrjE4x8vI3rk8dM3vuewfTyNE1yiA+tebzmNbnoE/HMFNBCscJS01ienHHElSUsM1J5Tu5dRzz2fzwq/JSEsD4Msvv6z1uM/nw+/31+pl5XK50DSNQCBQvWxn3n77baZPn15r2RlnnFErMaKSIoqixENDQ20URdl17LH7Heyx+x2sWfMAq1df3fQGzZRx1TWURCLQ0gutUvLL9X9mXL9+DT5clexoi6qkSM3kiaJ0hJkzZ/Lf//6X4cOHc+mllzJixAiEEPz22288+uij2LbNTTfd1OJ2VWIkwYqLtnLZmSeSnTuA4uISUrxJOBE/C39dyroN6+jXI5Oho8YgdYd0XwbRQAk+Hd745Evuuuk6stOyGDiyqMskRQDySlPpmSWwDQ/zF21lzOA+/PzrQjU2cRdy5P4Td/q43+8HaNNQtNdee61eYuSwww6rbkMlRRRFURRFiafBg69i8ODYTBlSOsz+OpNotLT2Sk0cfizatJnx99xbua5seVIEQAjG3/t35CMPNfhwYz1GmqOhniMqMZI4QsZunUFHxdGzZ0++//57Lr74Ym644YbqY3ohBEceeSSPPfYYPXv2bHG7KjGSQP/6+w0sX7WF3YaNIRAJ4uk9gL65vfjsuzm4k1IY0r8vaWlpuDWJjoseKSZh4WHs+EP56vu5fPrx/zj1tNP4/OdlrNz8V6ZdODPRT6lZ9hqRxMrVRfy0Kp/FP3/D8BEjEh2S0sGefbTltXDmz5/Pnnvu2ezEyDvvvMOiRYsYO3Zs9Ta9evUCVE8RRVEURVHi5+B164iEwziOg7+sjMV7740QGgcfVNKidpIuv4KgI2H//eDb72ILszNh7lwY3HA9hdawsWslN1qjZlKktW0oSmsNGDCADz/8kOLiYlatWoWUkmHDhpGRkdHqNtV0vQly+yXnsGLJar799juSfS7WrlnJxrztbN2ex9gRI9lz1DDcPh+27eCvKKdXTiYRdw8qAmE2rV3OQfvsQV5+CZork1EjBhOovMLeFTgBP088/SRzP3pWJUWUZnEch4MOOgjYUdS47i07O7vWNuFwmPHjx9eaWjq9cgxuQ8mVKVOmNBlHfn4+aZXDf4QQqsCioiiKoig4joNjWTi2jdBaf3oVdCoTDEuWwMD+8Oti+OoriPNwvbb0GIHaNUZUbxElkTIyMthrr72YNGlSm5IioHqMdLgX//UPPnvnfaKBciaN3o0xwwfy6kcfMmLwQEwriuNy4Q/6Kc8rIcuXTmaPTDatX0eyYZGaA4P2PZRIoITS/Dw8Ho1fli4hSXPYsHVrXOMM+FfjfvJD/KdMJ7VfX1Z89x3i+wdx9+3DZw+8wG/rK9hjrz58s9bPE0vzye45gGvOmsSsBWtZsLQM06lgZJqbGw724C8qZnt5mKKAw0ubMlm6biUCwZr1i+Mas9J9lZSUUFFRsdN1CgoKqu9bltVkmzWH09QcpjN16lTuuecexo4dW2+bnJycWutfdtllXHrppc17EoqiKIqidDujn3mGFYaB43bjuFyg6xgffURKWhr9c3MZ5fMx0DRxpGRdKMRvxcVsWreOomOOqd/YkMGwek3sfs3ZIOsUmGyrmj1GIJboqDs1b0PqJlPaklxRlM5GJUY6QF7+Rv5w0kmcfewB+DzJZHo1Fm4p5ckPP6U4ECFsOfy6Lh8hBWtXrOS0aYeQ7TXJC23l11XLGTW4H4NGTqBw01oKtqxgxOgJOLZFYb7G9z/9xOCBA+i725C4xuzxDkS7+jJSK3/f9OyF7LbnSIoiJUw6aBCr31rDo1/ls/8QQY7HB9KiIpTHicccyu3XpLBl5TpAkDL2AI454hJunyQIBMBfVo60omBoDOo3Jq4xK91XZmZmk8Nfaj5uGAaapuE4DV/F2FlbH374IR9++GH17zfffDO33347CxaoGZMURVEURalt6apV0KMH+HzgdoPLhQyHCUWj5FkWIieHbWlpSNumuKCA/M2bCdW4mFPLmjUwZjT8tgxKSmHMuHaJ2cJqU0JD9RhRuiOVGGlnJx51APn5Wzli7wn89PNCisoi5JcWs62kjG3lARxLY7+jr8IKhdA8Lg474jD0SBlFZSYuESE1I4OIlOw+YQ9WZKQRKCqnwrEIu7wYeiFegixe+BmP3Dg/rnFrWu1irk5ZmCVfLGLgXkOJZPnoP0xDK9N4ZZmFowuOOHAAy7Zkc+TehWgFGxH+LZiam4plH/Cfmw9g7hqNoyaOYePDLxAo3EzQZdAjo39cY1aUmkpLS0lJSWn08eYWYL3jjju444474hmaoiiKoijdRSQS691RNUxXxlIOEnCkxHYcLNtGWha2bePspKDqHVOncvP330PPHDBdaCOG46xcCWvWNl2EVUrwJUFODvz8M2zZ1HjIRGolN2pqTs8RUIVXle5HJUba2Zsfzcbt8lBWUsTp006kpGQ5tp7OyEHJXDB2HPe++A5FJQUYSWn4S8uQoULuvuJ8nnljFr1yM6kIhkjN6kvR1vVMHD2ELXllLF29inCgmG3bNjJoaC69ska3+/PYuDmfgw8Yz9Iv5uBPzsVrCPbJcDC9JqbQWFOQzF3nJJFiBCnIy8fjBKgoLSFQVoyI2AxOimClDGFRqcbqRcsYP2Vqu8es7NqSk5PrzWrz9NNPc+GFF9Za1hUKsXaVOBVFURRlVyPvuw8A8eKL0K8fZGQgvV7CLhcFhoE/KYnU5GQMIQikpFCRnEykkZohM48+ipvz8mHzJkDirFoFmgZDh8QSH44DVhRsB2wrlpSJRiEShXAY8vJg/XoI+Hc6+01IhGI12lo5K03N+2ooTYJVZeE6g84SRyupxEg704TG3y45g3+/+zHrV69l4r7788JbbzJy3BhKS7bx1fv/ZeqJJ5KW4iIjJZkBw4dy48MvcPCkMaQkpzBnwa8MGRzlW+GwdPF8luWXYcoIPXKy6ZGbwuqVS3n5w1Xt+hw+euEZUpJNSrdsIhoxyNu8kfQUk82aTrnLg6/XIO69eDS9PUX4C8vQ7DB2KEQkFEEKiEqDsD/Ilu8+Zr8+tkqKKO0uGo1immatZVWJhbVr13LXXXfVekwIgeM4SCmxbbv6Z9UtGo0C1Cvw2pHJCpUcURRFUZROrIHvaEmsMKtt27FjjcpjjJ1ybBg7FrxJsGY11JxgQYid9ByRO+LokbPTXaT2TUVubNkxRUMJkaoeIz169GhRW4rSGanESDv66sWH2bB6I17TjdflobBoGwP7D2BEbirBYDkbt29i2/oVTBw+kOWb1tM3M41fV6/DTEplQ56fNQuWM3TwEIYN6UdFKIBMySAp6EcLOwQjIUqKN7d7UgTggktn8vdxIZb/FiIt10u/FC/zl/j5cZvGkVdfzNpNftKdLWjBCDhhTNtCd2zskINlWAiXC9vxUBZ1+OPjL7V7vErXIaVs9vS7LVE3KVLTnXfeWS8xUrMIq2EYjbZRN96ysjJSU1PrrRdvVfv1+Xz4O2gGqnvuuafW3PCKoiiKouxEI9+XVRdbal6EaWxdINYrpOrxkSNAMyovjjixITuODVE71lskFIJgkFghPz+Ul0FZKRSX7PTq/TUbruFV8WqLe3s0lBxxcNi+fXuT2xpCYKljCqUTU9P1toNt21fw2t/uYtXK9fTqlcOwQYO44cIzKNHCPPXGw2wp85ObbjJu9O489OA9pKclE4k6VAQCTNxtKD2zUhk5rC/hYIAtRUWsy9tIXt46Pvv8c7auW8PmTRvYnLeel99b0SHPpzwS5cPVNr9ts/lhcRk/LPIzYmgaQ3bvxwsPPc59x+cRCjkEykqQ/nIcxyEciWIh8aZlI4x0fNk9WLE1yh77HNwhMStdQ3skRZryl7/8pcEYmpsAWLduXfX9jkiK1BQIBPjss89avf3+++/f7HVvuOGGVu9HURRFUXY5VQmPOscTsrLOSHV9kaY4VW3IhpMbkh3DaqRT54E6sexElGjlVvXXq1tnpLHhMxLJ2TPObtbxnN3kGkpriE5268pUj5F20KvncE778431lk+75i72nZSDlBZz10uyU1LpPXQIi5at4KYrL+bHBQv4/vvvOeuUI/lm7nz69c1kYP8MfluzhuGD+zJwUC+2b9vKhVffzKmnXtZhz0cTGmsyx9DXWsv64ghJbi9fFaWQnJxBuifI0mUbiAYkKRTiRMHQTSoiXgoqIvQbOJio7aVw3Xr+9WnHJHIUpTErVqzgkEMOqS6mWrMHiK7rjc5iU9OAAQMS0ouiKtYpU6bgOE6rkkrfffdd9RWrnW2fiISVoiiKonRl8ne/a3KdlLKy5g2lkY0kRRr9epY7flYlTproDXKOPIcPxYc7XafxvcX+u/GFG5ly9pRmbbOz3ryK0hmoxEgHEkJjzk+x6bk2bfyNc08/BG37RjKz05n721wWLZ/PyDF9efWTLygvjZCZbFBevo3eA4cgDQ+l/jxmzcnr8LgrIhFCmiRkuBnWL5lVRVCqm3w+5zd6pWeyauVWIoEwunAwNTe+NJPN+SGiMo3hAw7h7LPO7vCYFaUuKSUjRoxg2rRpnH/++Tz33HP1Hu+MaiYp3G434XAYTdNaHW9VgqU521900UWt2oeiKIqiKPU15wIMja7TjIsWsvp/TfYYAZgqY3X/vhXfNrC3xvdnCYuXnZebjqeGSCTSovUVpaOpoTQJ0rffbnzx3TZmfbcNw6exbOkiTKGzbs06CosKyUzRMExBfnmIO/7xBE8++xXvzVqTkFglku2FBSyI9qLCcrHClUWfvQ4lHImwtbSIstIoxWUhSiokeWUhSq0UvD1G8vvbn+AklRRROgkhBLfeeivvvvsuzz77bL3HAH7++edEhLZTo0aNqr6/YsWK6ljb2qtDCMEZZ5yx03Uef/zxNu1DURRFUZQdpJTN6DHiNGsoTKN5i6oeIy24gDJZTmZfuS9jvhsD3oaH13h7e7lGXsPt8nYed9TxgdL9qB4jCSaE4K03f6u3fOaUscxa8CsHn7QHwwZNTEBkO1z4+/P4z/MvsMmVREh4KAxFOWXqMbz0/BuMH5xEflkAOwRunwtvRh+OuOA2Jkw6UHXHVzqdW265pfr+d999V6/mxoQJEzpdz5Fff/21uofIgAED+P7779lvv/2A1s1UM3Vq7OrQwQcfzH777cf111/PPffcE/e4FUVRFKUr2XPt2nrL5vTOxeX2xG0fVT1GdvrNXTWUppqkRdUbquuTtPx4JnW/VCYHJrd4O0XpDlRipJMq8ibxQ77FjFFZiQ6FJ576J0/961nyissIpSdjhR2Onn4mN994NavmPE8g7GXA6D2484k3VDJE6TKqkgvQsuElHWX27NmcdNJJFBYWkpKSwo033siiRYvYd999a8XakrinTZvGRx99BMCXX35ZvbyzPXdFURRF6Wi5AwfUW5a/cRN9+veP2z6a1WPErpEYqbduE8fZ1ds1tK2iKDujEiOdVA+7jGDAzzargiunHUi6HiXfSWbkXntw2cy/dXg8tl2GpiUTiVjM+uBlDjn8eL5672Um7TmBe59+pcPjUboXy7JYPfta3KYbR0p0KRGmj/7739ru+647Be8LL7zAOeec0+773RkhRK0hMxUVFdx4441IKXnllfrvNyEEL774ImeddVajbQ4ePJi1lVfD6h6USSlZt24dAwcOrF6WnJwch2eiKIqiKEqVZqUqLLuZQ2Eqj11qJlGqa7C2rseI0gU1XWe343SWOFpJJUY6KTMjg0d+fzSbevRn1LpfWdZjCCViM9ef90RC4oldUfZj2zbvvf8Rtl2RkDiU7skwDHQhcaSDJgzQHXDC7b7fukkRgIKCgnbf784sW7as+n5DV5Xqxlu1ztlnn83ZZzdd06ex4mc1kyIAFRXqPa4oiqIocdWcZEUk0visMs3pmd2GoTSKsitTxVc7qeP+9DeG77Ubfz70CLKOncrxe0+ESDn9+g5JaFy6rnPC8ccmNAalexIygHDKY70l0Dss6SylJCMjAyklqampXHXVVR2054aNHDmyOq6GVHXDbc3Ql5kzZ6rp8hRFURQlQZrz/S1CoR29P2rOMrPzhuv/LpsxA46iJMjjjz/OuHHjSE1NJTU1lX333bd6uHeiqB4jndT48ZMZP7528aMzeCxB0ShK/Ni2w/QTj+Xrb77j3PMuJByyOPesU+ijCZ7+QOP/TtCROOiiskBZA7066mrOOjtTWFjY6m3bQ9VB0xlnnMEnn3xCOBwmEAg0ul5TcnNz2bJlS4vjcBwHTVP5c0VRFKV7EjNvgj59ICMTfEloXi9JqWkMHj6MvdMzeL5/f7YD84qL+G3BQkq3b8MuK4OyMigtg+Ii2J4PW7fAmrWwbetOv5ub863dNzWFjU5VVqSlF0FkjaSKRAweTFZZWZO9Yb8TGslIPICL2JVzDQFCIDSBZujYQNSRhB2bsCMJS0kYCFbeAkAFUA6UAEXAdmA9sACwVA+W9tFFh9L07duXe+65h6FDhwLwn//8h+OPP54FCxYwevTodgpw51RiRFGUDvXbrCu45+770XXBDTdeQygS4eFHHyTZ56PcKmP36f8GB8753Qz47GbG7zaSD15/nTfffKfB5MfWZe8RyfsMEAw48MGOfjrtqqF6InU1lhA64ogj+OSTT1q1XyklBx54ILquq6KsiqIoyq6l8ntVA/TKn9So/dX4ds1ouxnfqRuuvRbxyqutzotU35GA45Cfn9+SDWupekpSyh31XJt4DrLGz6qb6rui1HXcccfV+v3OO+/k8ccfZ86cOQlLjKhLgYqidKj3PvoPb//vJZ79z7/ZvHE9P8+bw6zPPuXdd99hv90nYugGZ0w7jkW/zGHjooXcetO1bN++nj/84QxefvVVQqFQ9XR3AOFtH4MDH370eAKfVdvEK/nwxBNPVLf16aefsmbNmla39fXXX6ukiKIoitK9OU5sFhjbjhU9tW2wLGQ0SjQSIVL504lEkJa1Y13H2bFt1f1mDF3Zkbdo4vs1Gt3JOk3NTEN1UuSPp5/e7B61srLpunvdMYQ3tlZjUe1suTqa2LWUlZXVuoXDO68baNs2r776Kn6/n3333beDoqxP9RhRFKVDlVXk8NVHL7B5vZ8xY/oyYvhuZKb5yEhOIiU1mWsvnkog2oPtP3zPLwsXYZg6q9du4eqrZ5LdI4dly5bhcrmIRKKMGDkC6Ug2bFmJ1omnii4qKSMzPbXRx9syDCgUCvHFF1+wefNmLrjgAmDH0KIhQ4aw5557Mn/+/Fa3ryiKojTs54yfyYjEalQN8g9q02e5kiAeD/TvB737QGYGji+JgNfL5tR0shG4keQh2JKTQ8XQIdiZGVBaCkVFkJ8Phg6hEGwXUFyCZVk7318zLzjIc89BvP9h5S8QG9Uiam9e98+tuh5JbDjNc9dfz3lHHdW8/TVyv27cspl1S1o7EEjpHvr161fr91tuuYVbb7213nqLFy9m3333JRQKkZyczNtvv82oUaM6KMr6VGJEUZQO1XPseWz5/HEys1P4bcUG0jNM0nxu9hjoIb1XHpGKYlLdqzl/el/KQoMpi+oEI4L3P3gaw9ubnKwc0lJTSUvLYM2a1fTqdTYbgp+BPSvRT61RW9ev4djzHub7d56Je9tut5ujjz66+nePx1OdmZ83bx4TJ06sPJhShyeKoijxMkVM4R9p/2h1MWxFaYo8dioA4ptvai5tcjsRtXBeebll+2qi+dgImliGRkoH2axxQzu2VfXKdi0bN24kNXXHBUG3293geiNGjGDhwoWUlJTw1ltv8bvf/Y7Zs2cnLDmiEiOKonSoP/zhCq68cmb175MmjeTM085HbLsbwSqy0tfy31kllIUgu6ebgG3it0BKneLiRWzdFGDU7kPISPdg6gZHH3AY4agbu88dHH/c0RhzvuG1rSUYRuf5eBs9fncG9cts9/18/fXXtborTpgwobr3iEqOKIqixI9EoktdJUaUFmlV2ZADDgDgm9JSjly0mGADM9Dker1suej/4hJj7aZjQ2eEA4hYUkRU7rPWenXu16wz4oCaEa89daauOZVxVM000xSXy1VdfHXixIn89NNP/POf/+TJJ59szygb1XnOHBRF2SUkJyfX+n3u3GUA/DTLwOf8zPzFw3GSi9iev5aj9y3i6vsrWL5kDhs3LqeiopzVKxaxPW8Lm7ZspqCkiMf+9SBPzhyObYXZ53cGxh8P6FRJkSrTJu/LmMOnseSz/7XbPg488MAGD9CrpiIWQhAKhRrN3CuKoijNIxDoTmWB6s5yUqJ0fm1Ioh2QlkbggMlNr9hKO4tMiho9Rpr5B1/zfF0lRpTmkFI2WY+kPXW+swdFUbq1xsZg7zXlan597xAm7Kaz58RTiRT+h63bRyCEYOTYfRk5tvFiTMtn/QE7sh4hQ9giudH1Eum0007ks/mLyCssIScrvcP3X1ZWRlFRER6Ph6lTp/LBBx90eAyKoijdiSY1lRhRWq4T9jBqbkQtDb0qOaISI0pdN954I0cffTT9+vWjvLycV199la+++oqPP/44YTEldMDXwIEDq7t4V92uv/76RIakKEqC/LRgEaY5CHfqYYjAp7hdKQzoE2nWtkMPfRrTPQjdTMbj7dnOkbbeuEE5nD7jTwnbf2ZmJlJKPvzwQ1UkUFEUpY00qVXOiiqbN1WronRyjeY9nKp+Ijsvvlp3qt6q+x6Pp+3BKQ0SnezWXNu3b+ecc85hxIgRHHbYYfz44498/PHHTJkypXUvRBwkvMfI7bffzoUXXlj9e91u9oqi7BomjB/D4g+CZGeMhPAarMg6HCOpWdvqukY0/UhcxZ8y9NBH2jnSljvwhBMoLtzMOSdMZfjAfhx8yhV89cY/ExaPlJLnnntO1R1RFEVpA+EIHBxieRGVGVG6A0FD6ZHYhDetO16QQO/evdsSlNINPfNM/CckaKuElwhOSUmhV69e1TeVGFGUXc+2bVu58657mLt1b4LbP0M3fAhXLqbemy2/vtCsNkZNPJmhU55q50hbJxzMRxMR/nT1rRy5zyTSUzXCkeb1hmkv559/PrZtI4QgkuBYFEVRuiKBqCxOqRLMSte2o4dHA3/LldMFx4gmx93U7S0igcGDB8chSkVpXwlPjNx7771kZWWx++67c+eddzZ5gB4OhykrK6t1UxSlaxs9ZgR7Dh9Ej2DsgoQmBC4jB8OVTiT/p0SH12Y/fvId6T1249ATDuOgA3ZnYG4Wx52fuCE1VTQtNj7e5XIlOhRFUZQuRziVZ4sqL6I0U4amkaFp9DUM9v7pJ+5duRLL2fnwlPb2Yo2htfH8U65KijjA66+/HseWlVpkJ7t1YQkdSnPFFVew5557kpGRwdy5c7nhhhtYu3Yt//rXvxrd5u677+a2227rwCgVRWlvViTCMaeeCcDqz2YAoOs6wsxF4E9kaHERCoXITZegZZGZ3YtemRksW7uRA0+7kq9fezDR4SmKoigtJCqvnKvhiEpLFFZOTdqZnC0lX+207litLiM7Xavu7xK47I47Wh+conSguPcYufXWW+sVVK17mzdvHgBXXXUVBx10EOPGjeOCCy7giSee4JlnnqGwsLDR9m+44QZKS0urbxs3boz3U1AUpYOVloWq71e4K2efkTbYJThWhHAomKDIWi8SCXHBFdfgyu6L1+vltWff4LV/vUlqejqnTTmB/cePQMgK5i9dn+hQFUVRlLZQuRGliztYSvQe2Q0+VlVfJJboaP4fuwTel5KLZ86MR4iK0u7i3mPk0ksv5fTTT9/pOgMHDmxw+T777APAqlWryMrKanAdt9uN2+1uU4yKonRe4w84h3Vf/oSua0gZRTdTKVjyBH0mXpXo0FrE7fYy88YZXH7lDQTCkkjUxrFtIpEQt/77ef755z/w9fxruOzGW+jTezhvPHZjokNWFEVRmqmqx0j1rDSK0sWNz8sHwAkE2JiVBaHwjgdl5f+a6DjiACe88w7Djz++vcJU6upMQ1g6SxytFPfESHZ2NtnZDWccm7JgwQIAcnNz4xmSoihdjmTJ2ghfLnJz5amgiYSXQ2qVN76aT27PfKxQBZaQ7DvxUGSyh1A4wkEnHcdzDz7Mdfc9yap1vyU6VEVJCEdKFmzahGXbjOvdG6+qd6M005v778+hq1Zh6DqpW7YkJgiVFFG6IS0piQHB2j11JRBZuJDImjVYxSXIJC/uUbvh3W0UmvrcVrqJhNUY+eGHH5gzZw6HHHIIaWlp/PTTT1x11VVMmzaN/v37JyosRVE6gTJ/iNVbvJw6oj8b9P3JDX+U6JBabOOmNew75WBWlJWQ7NHRdIMPPn0VKW2k41BWXMKxp0zj6Uee5qGX/5focBWlwxQHg2Q+/zxkZEDVAXU0CoEAlJbCtm38b/p0jttrr8QGqnRqttuNdBwcTeP03r15NRHJEVnnp6J0UwJw77477t13T3QoitJuEpYYcbvdvPbaa9x2222Ew2EGDBjAhRdeyJ///OdEhaQoSicxduqTpCVfw0ufb6JoxWvMmFqa6JBarG+fQWysrB8SDgVZvHgeH33yDj/89BPbigrQ0w1WzN/KscdO46mnH01wtIrSMcTTT8OAATBuXCwpommxqagiEQiFoKICsrKY9s038O9/Ix9V7w2lYXpycmyqXMehJE4zFFohiwu9F9ZaVtUjpG7PEBOzagVFURSlG0hYYmTPPfdkzpw5idq9oiidmNA0eu9/D+eafwXp8Os6ncGJDqoN3B4vE/c6gIl7HVBr+czbruLOO/7JW198y4UXzEhQdIrSMcTrr8PYsZCWBh4P6DoIAbYd6zESDoPXC2537DFdR1x6KfKRRxIdutIJOW43jpQIx4kl1+Lgj94/xmqH1FD39yo55MQSMyozoiiK0i0kdLpeRVGUxpimm777x6Z46zu56fWrpkwUO51yrnOZNGF/Ruy7kKXLliU6FEVpV+Kll2DYMMjMhORkMM1Y8kNKcJxYYsQ0Yz1IhIgtsywIhRDnnot8/vlEPwWlkzFSU2NDaeL0mb9t3jZu8txUPQWvlLL6voNTu2dIZW2RMOHq5f8Q/+BaeW1cYlEURWk2VXw1blRiRFGUbqErJUSqTDv2ZDJ8Xr78+tPqZY7jcMw55+E1dP77n+cSGJ2ixMcNn38O/frFaoqkpsZ6i7hcCE1DANK2IRJBChFLlFT+TjAY613Sqxdi4kTkvHmJfipKZ+L1Vg+liQdXtouSaEn1gX29niCy5t0av1R+9URkJC5xKIqiKInRNad6UBRF6SYOOOQY/nLLPwF49qV/M2LSGNYs+ARbc3PVX+5PcHSK0nb3FBbGEhxVw2TcbnSPB5/LRbbLRbbXizclBc3tjvUaqbq53bEkSnIyDBqU6KehdDKu1FQcKXHiNJTGne6uHhojd2RHak3JW+uxOkKE2hyDoiiKkjgqMaIoitIJTBjXmzsfeoTSgMXypdt597kn2Lh1Fa+8MyvRoSlK26SmxgqtVg6f0QyDDGA/4HfAqcBoTSM5ORmtaiiNpsVuuh7bzudj2qmnJvZ5KJ2L14t0nMpkRtu5kmtMOVqja/rOkiE11w8S3Pk6iqIo7UB0sltXpobSKIqidALzF9WfavKNpx7ltP+bwbLV67jtmgsb2EpROrchf/sb7Lln7BcpQUoMTaM3cCxwNrCN2DnoVsPAD7H6InV7AWga782d28HRK52ZNyMjVv9DSkQceoyYhtn4MJrG1FhNJUa6pmv23Rd6944lcJO8sWF+pokX6I2kB+BHsidQmpaKbZqxdbOzY9sNHgyjR8PEiXDAgV1yWK+iKDEqMaIoitJJCSF4/anHOevSq7jjof9w8+W/S3RIyi4ualm4jjkm1pujOfr0gT322FFgNRpFRiKE3G62Ar8BBUAhEK2anabmLRKJFWG17VjvEUWppCclxQqkxqnGCLQuIVIlQCBucSgd5x9Tjmj+ypnZkNl+sSiKklgqMaIoitLJvfTIA1x201/4+xPP86eLzk10OMouamk4zCPbt/Pn557DZVn4AJ9h4LFtkg0Dr6bhkZIkw8ANeHQdj65jAm5dxxSi+mpqkZQUCEEAiAAHABNCIcyUFHr4fPTp14++pkmmYeDVdbymmbDnrXROZnJy9WxkMk7T9e5UE7tQiRFFURJCzUoTNyoxoiiK0gU8fOft3HbvA9z8j6e449o/JjocpQNJ4G+ahiklQaAIOOKxRzny4hkdGscot5vH+vePS1s9gBF1F/p8sZuiNIM7NTWWEGnvoQvNPNDvjomRqoSTGh6iKMquQBVfVRRF6SJuue4qeiQLnnn1f4kORekA0YoK3hSCj4RgrJQMBgYDw4H5My7hMiGYlp2V4CgVJTEMrzd2obSydk1c1ZiNpnmrS/yxCjndxqmnzkfTPkfTPkeIzxBiVuXtB4RYhRDzEOI+Nm0qSHSoiqIocaF6jChKgjTU9VddlVGacvlFF1JQoA5Eu7t/C8EAoB+gAzaxIScBYl/c1Z8UhUUcJQQfd8RQAkXpRNw1htLETQuaq1uPJEw4vrEkmGnWPR6p+t1feQP4Ea9XFQZXlIRSQ2niRiVGFKWDSSljlfQrpxmEWEJEqyxmqJIjSlOys7MTHYLSjl4TgpFAClBVWcMGQsS6ecrK36OVy/oABwjBNyo5ouxCdI8HKqfqzbFtzmngu/OFdnhPNFagNUIk7vtKpKZrHcdeB7dbnUooitI9qE8zRekgNYvE2bZNNBrFsiwATNPENE10NeuCouzS3p80iUFAOuBFoFeehlnUTopEgCCQDKQCOcBIIVimkiPKLmL5mDFIwAGmQ/V9p8b9eGnObDXS6F7vvfo9RhrmcqnCyIqidA8qMaIojajZRbfm/Zo9OnbWu6Pu9lU327YJh8ME/QHC4TC6rpPkS0LTNNVrRFF2cek//UQa4APcSHRiJ3hW5eNVPUXcgKfylkQsQZIDjB87ll8WL+7wuBWlo+2m6whNQxD7zhRCQOVPIQRt/RZtLBnS0HKJJKNvRhv32Lm43c27UGMY6oKOoiSaOmuID5UYUZQG1BzuYts2jhO79lQ15EXTtB0HX40kMaq2r2rDtm1syyYSDhMIBAhUVBCJhPB4POiGhtvjwTAMlRRRlF3Uu0IwmFhSxAu4iB3syMqfNrGhNUblT1eNm7dyu01LlnR84IqSABIQUiKFqC6+WvV+oZWz1YySo1odz6d82uptOyPDaOr1iz1edUFHURSlq1OJEUWpoyqhYVkWkVCYcDiMFY0ipUTTdQzDwHSZ6IaBruv1enpUJVUsyyIaiRIJh4lEIkQrf4ZCIfz+CsLBYGzbDAHSUdleRdnFZbEjIeIilvwQQiCFQCCxJZhSYhD78tYrf1YlSjzEeo4MGTKE1atXJ+Q5KEqHqUyK1E2OKPHR9FAaddSiKEr3ohIjilJDVVIjGo0SqPBTUVZGeWkZwWAQKSWmaeLyuPF4PLjdbky3G9M0dyRHKg/OLMsiHA4TCgTw+8sJBYIEg0FCoRDhQJBIJILAITU1BZGZhqHpaEJTvUUUZRf1Wq9chhEbIlOVFDEFCMMEjwsXAjsUImhZaFIiiCVGtMqfeuU2bmDlmjUJehaK0sEaSI4IUImSOGi6x4jqKaIonYKalSZuVGJEUeqwbZtQIEhpcTF5W7dTlL8df8APEtwuL26vB29SEu4kL97KBIluGtXDa2K9RaIEKirwV1QQKPcTDPoJBUOEQmHCoRA4Fr4kL0leD5ojaw3JUcmR7qesIsCSZb8QRmKGQxQHoljBKP/3p//wyzdP0qtHOpYVG64lkbhMVcxuV5O0fRsudiQ5NARCE5CWAv36g5SkbNpERUkJmmVVX6sVxE5PqhIkJrGaI4qyK6g5nKahniNrPZ7qGiQAm3v2ZL9165rd/rd52zk0FJuGt7G6Yw1NGexEIsgRI1r2ZDoZt7upxIdKjCiK0r2oxIiiVKoaQhONRgn6/RTk5bN14wby8/IIB4IAuNwe3F4vHq8Xj8ezo+eIaYBW1ePDwbFtggE/wUAF4UCISDg2JCc2LCeCJjRcQmKHQ9iWhW3tqGMipVTJkW4mNTmJfSfswzdzfyDocqFbNsLr4/3X7yA3JxMA01QHmbsyDzuSIrETORAuF3LQYDjiSIhGYNYsTH8AaVmNtiOIfbG/+eabnHzyyR0Su6IkRNVMbzWSIqJy+t6atXlqpi20tLQW7cIU2k6TII3+3g16q5hmw0VVpTy8xm8HdUwwiqIoHUAlRhSlBiklViRKRUUFJQX55G/Po7iggGg0iqZpuFwhIqEQ4YCfgGnicrlwuVzouo4jqirjgyYd7GiUaCSEHbWwLAsnGkVGLaRlg+ZgRy2igVBsdppQCNu2kY6DUFP2djur129m29ZVpHo8+B0N3euhIBRlfd4a9pIjVCJsF2dHItUJkR0E0u2F3XaDU06GsjJYvhxj9WpkCBxZe1rSmj1pNeDiiy9WiRFl11AnKVI1pLXeT0C0NDFS+X3cVEKk3rJukBhRNVUVRdnVqMSIotRQVTQ16A9QXl5ORVkZgYAfaUt0w6gucu84NlYkihUKETF0NE3gAAiBpgkMTQPHRjoWjuMgbQfpOGhCYmgaGg7SihIOhQj5A0RDYRwriiPdqnNqN+Tgp1fPnvzw3Vw++XQ2x59yPP17ZbEdySsfvsMZU6e3ZgIFpZvYOHt29ZVtp/om0Q0d0tKhb19Efj7S40FqGg4CB4lNbKYap8atKkFSUFCQmCejKB2kKvUgYEdypOZ9WWdiXSFa3GPEqBweW73PRhIe9ZZ3g8RI08VXW058/XXstenRA4qKYOEvePx+bp68P2eNGcOAFv77KIqixJM6B1OUGqSUOLZNNBqrBRKNhIlaUWw7Co4DtgTbAdvCsSJErTDRSJhIOIwViWBFYj+jlUNmbCvWQ0TaNpp00ITAqJzJZsfMNRGsaBjHsRP99JV2snJTMSs2FUJGLmZ6P0469jgCjo/c1CR6pqfy5vvvJDpEpYOFS0ur72/8+uvqxIZV40Y0Ctu2wbx5yAU/Q14eISta/bhdZ/0oEKn82RPYXwj2F4LJDdwOy8jg/eee68BnrChxVjWUpsbvte5X/V51X0qMjIwW7cJV+V1ddasfQsPLuwOPp516rwoBBQWxY6pxYwntuw832TYDn36apMMPb3p7RVFqk53s1oWpHiOKUkdVrRHHdrBtC8e20SQ4joXjaDiOwLFBIJFSQxoSoWkgY8USRawvL8hYL5GqgyahG7gQSGnjWDK2nm2BbVf2KJGx/vFKtzP9j/8mUm6T1M+D/7uH+HXVCpb+PJeKigoOOnQ/SHF4+X//44zjjkUIla/eFRQsWUKf/fcH4IA77kDedhtWfj7hL79EfjUbZ+lS7II89G1b4e23EX4/bN1K0HQhs7y40tLJ6J1L1siRTNhzT3L33Yeeu43CEaL6c0dKGRueh0Q6MjaiwKk8kXNi/UvKi4pIycxM7IuhKK1QVXgVIarriVQnR+r0HqFyXaOFf+uNHSQ3mQzpgskSx5H06vUP8vPDdR8B9qfdz3iKi0lyudp3H4qiKDuhEiOKUkfdr34NWfmfAzJ2koHjVE4H4YCtI4glRzQ0dCHRRGx2EUfI2FS+LhPTcKNpGtFomEgwhGNFK69oOYl4mkoHKpz7AHkFJfTNzUZKia47HHb4ZG6+7m4WLVrM7y68hGG9k/jyu89JzxzEnqOGULfihNK9+Hr1qvW70DTMnj0xTz8dTj+90e2GtHdgitLV1E2OVC6rmRCBWJHW1vQYqb2rZiYHnK7xve7xPEA47G9iLQ34ofK+BNqpV0dJCcltTIxURCKsLC7m5i+/5IMVy2O9U3ZWLMWyQNPYp29fLttrEocPHkxOkprXS1F2VSoxoih1CCFiyQwNdF1D03WwbQwBhi5iNwGakGgShHRACjTA0DQMXcPUNIQwEJpAd7lx+1JweT1ICcEKPxWymHBIoGta9TS/Qgh1LtxNJfs8JPtiJ8JSSpZvKCYciZDeJ5sH7r0bX5IHKSXzl6xg8+YNZCR7GdS/T4KjVtpT+hCV4lCUuKmR/IAdvUdqLZMSIzm5Rc2amtbioTJSyi7TYyQcHgMMBwLAW83Yoh0PUkpLSfH5Wr25WL0SkpPB44FjpwJTIRyOFa7esAHWrIH8AtB1cLvBtiEYhPJy5vz0E3M++ggKCrA+nYVuqNMjpQvpTENYOkscraTe+YpSh6Zp6LqO6TLxuAyihg6GwGMaeDwmbpeBoWsIGTtAkFR22dViCRGPW8fjMjEME93twe1LwZueiTvJh23bVBSXxnqXlJdjmiaGy4NumrFhOKoC5y5AcM7lr+DH5vEr98eX5IktFYKJY0fA2BEJjk9RFKXzq1V8tXrhjt4j1Y/V7DHSiuKrzY6nS85KI4CVlfcPBEYCC4Efqf3KSiCXWGWjdlJRgS8np/Xb5+XFbg1xuWD0aPYcNJifp08Hy6axiucqKaIouy717leUGqp6i5imicfjxuPxYCcFMYRGks9Lss+D12UgAMdxsG0H25E4TuwL1jR1krwufL4kvElJuJNT8KRl4U3NwvTGEiOeJB/C0DHdHiTgSkri/9k77zg5yvqPv5+Z7ddLLr03kpDQe0mogoAUJQKKIiAdRcrPCoJIU0FAUFBpIr0jhBZa6CSBQBLS212Sy/W2fWfm+f3x7O71y93lavK885rM7JTn+c7u3O48n/kW0+3BMFyDXWjVdAIh4Hc/msXue0zj20cf0t/maDQazaCmWQgNNIoSbQgk7tzcLrU9JBAgPG4cCBDJhyCpvtpbVl0PxoccceBr1JmcAgwB3gRqgMXA0uR+F/ZO95EIRV0MdeoSts26sjKVQF8IFe40tAj22ANyc5UXSVVV7/Wv0WgGPFoY0WiakBJGXC4XHo8Hn88DdgZer5fcvBxyczLwe1xgJ0jEE8TjFvGERdyysR2J1+UiEAiQnZNFRm4e/uw8vNn5eDNzcHl92LaNy+1CCBOX10ciYeHPzMTl80IypEaz8/N/V53f3yZoNBrNoKapINJKHIE2BRJXVlaX+/F1lKNip0QAZcnpG2B933QbjzOioHcTQcsmF4l46kkeGjeec757mqr+ZSfzwvz++l61QaPpaVoKs/3JQLGju2hhRKNJIoRoVnrPEAZujxchDDJzsigcWkR+Xg4+t8COx4hFI0SjMWKxBNG4hR1P4DLBGwjgzcoikJOrQmgycnH5AgjTjSklYIJQCVljsTgerxePz4/pdmlhRKPRaDSabtBmaA00E0jMHchhoellLIvRRUN7uZPGq0OWbuXS+nrYWprcpO+/NJpdHS2MaDRJUqKI4zhYlkXCdsB04/P6yS0somDYCPLycnCZAjseJRqJ4InEiMVieKNRouEw0ooi3G6Ey4vhDeD2ZeD2+hAud7oMq8fnVU+3TBeJRALTNPH5fZimqYURjUaj0Wg6gYR0TpFW62lbIDG83t43TNM9pGTc0N4VRmSLi0K2rB40aHLDaDSa3kALIxpNE6SU2LZNPGFhWbYSRjKzyMkfQnbBEDKyszEMsC0Ld0YcbyxGPBojFKxHUkk06JCwBQnHwBFuMN1guhDCSHukGIaB1+fFMA0cx0EIgcvlSgsjWhzRaDQ7A7bt4HL9G0Q2apgaA+pBlhIMXktGhq+fLewdPjn/fBY98ADlwJRf/5qzb765v03aKZlmt5EI1HGQ8ThOfT1OOIxTXY1dX49dXo5TWUnOvvv2vaGDmj4UCgSMbVHGvKdJnY385JPGlU2XNRrNLo0WRjSaJKkQGsdxsB0HC4Hp9uLLzCIjJwd/RhYefwamIXCkjW3beC2LWCSKIx1C9Q3ErQaEtPHZYCMQwgWY6T5SoodpmhhNygCmBBEtimg0mp0BIe4DMRHE7qjvQAcljIRA5JGZdS/Ir5Hykf41tAdZOnIkDVu3EgbGAHlA3S238PtbbsF/6qn86vnn2z32y1//mm9uvZUfQpfLw+6qiLZyfxgGwuXCCATU6wkT+tYoTRtIxM23Ks+enFzYfXo7+xmMHDKk+92EQvDZ542vPR7Ydx9VvjdtiUaj0bSPFkY0Ow0pUaG7N5VNRQkhBKbpBp9BIDMDXyCAx+fFdClBQ0gT0wW2y0Y6EmG6sRxJNG5jmKoSnJQmUgiVzb5F29tb1mg0msGKEE+B2AvIAjwobxEbJYz4kutMEAIhfoKUD/WfsT3EFiHIQElAPsAPBIBMIAeoe+EFrheC6mnTuPubbwDYPG8e751wQjrk44f9YLdG0xtYto371tsgJV6l7m/qauGjj9Wy2wX77994kCEo7GLVIAARrIeKCijdpirN+P3g90FDEF54CaZOgb33AronjLwjBB7Ajfr7dgEGMBT4INlm08lJTk2XbeB8LXhqNAMeLYxodir22GOPHW9ECBAmpsuFYZp4vAE8Xm/aywNIe3sYhoHEwJGQSNjELRsDgcRAmAZCmOkQmpbChxZCNBrNzoYwHgUxHchFyQMpjzmLxiFFSiiJAxMQ4iSk/F8/WNszhM47Dz9q4ORGyT4ewIsSSQJABpANDFmxghuEYA1wPI15MM5OLjt68KQZ5Ij6Bpj36vaTmSasRpHEcWDPPcn+2z3MGTWSd88+u/MdLl2qcoPMmsmvzj6bW485BoJB0n9dHnejMNLFP6+Pp0whB/XNZQIzjjmGF956S4kdQrBU/71qBgIpVW4gMFDs6Ca7Wg0yzU7G5s2bm4WhLFq0qEfaTdalwTTdmC4z7SkCrQUNJ5mXxHIkthQIw8QwTUzTnXb11SKIRqPZ2Vm1uhqYhAoiyUJJAn4afShScz9KKsgCkQ9M4oILftMvNvcEWx98EBsliqS8RLJQQkgOSiIqAIqA4cB44EBgLfAKjZ4iWhTR7BQs+RJGjICzzoRzftzoNdIRLhdMmwaLFvLeffd1vU8hIBzmjoULIRiiWerd6ur0Ylf/wkpjMTYAa4CVQHjTJr7qunUajWaQoD1GNP1KS8Hg0Ucf5Yc/7Nih2DRNnJaZxJO4XDt+SUtUdRobiSOdZiV8W+2b3GZLcKQADAzDjen2IEwTYei8IRqNZtdgt90+BSagRA8/KQ8RISRgJb9HbRr9KfwoGSGPf/1rCf/8Z//YvaO8CBwOjEJJP17UmTf1IPHStgdJATAVWNbXRms0vU1xsZofegjMmAErV8I77zb3JMnwgyPhtdcgHt/xPtu6B6yrSy/KLkoj7oIC4qnzoI0qNhqNZqdCCyOafqMtweDss8/m7KQL5amnnsr9999PQUEBhmFQWlrKiBEj+sKyZolQHSRSilbiSCpExnGcpMeIo3KKmKkwGi2KaDSaXYOSkjrUMD+AkgDcCOHGNE1cLpDSwbIktp2guWzgBREAWcgPfnABjz02+NSRa6TkdCH4AcpXZiJK+PDRWiBpKpI0zUNSANwgBGuBR7XnyC6LlJLb8vMxTbPZejMnh1+sWdN2wtkdwmF7zuNSPr1jXdg2fP21imP57qlQNAxefAHKyuDFl1rFtwQ8nm53Jc02zqWhoXF7F9vzFBbSTK5pYqv+K9Vodj50KI1mwPLCCy9QVFSULmO7PVGkJzL5q9K5Jm6vB7/fj9fnw2V60mE0bfXnOA62JbFtiSUNHATSER16mmg0Gs3OxJgx/0MN8z2ACyFcuN0mBQUwZgwMG2YQCGRiGKkUhqk0hi6UTJDB449/3W774uqrev8kuknFJ59wMhACNgPvA6+iBk5uGgOHksFDZNEYYpOfnN4AFgG1wElaUN+lqC0u5rbcXP5cWMifhwxRokiLa8Cuq+MvRUPVPoWF3JKZScmXX+5w31IeS0GBuf0dewIhoKwcln4NmZnwxhttJv0YkpHR7S5kW8JRONK4vYu3ZJ68vObtd9NjRP9FazSDA+0xoukXLr300h06vjcEByEEHo+HwsJCCgsLO7W/2+3GQeIIkMIAQ+BIgSUdbNkYaqM9RzQazU6NKCBdbQaVeDo7G2bNgt12U0UjFi8WlJR4iEUFEoEaLqREEjdKKmgb+Zfbe/8cukCwuJgXxo5NP11KfcM7wBDDIMc0mWfbHO04aQ+R1Fk29SD5G9AkTWSa2ro6cnNyev08NP2HlJLi8eOpcRyGer2U23bjdZC8b2h+r9O4nBUI8Oyxx1Jm29zaJIdGd6isPAIAId5iIAzhR+bnd/tYp61bw0Qivdhlj5EeEkY0ml5FJ1/tMbQwoukX/v73v/e3CT2Gabpxub14A358lo3b68Yw3UDvCDgajUYz8EgN/ZXgYZqCoiKYPRuOOQa++QaqqqCi3EssJpI3T6lBmEHKcyQajeLz+frnFDrBYy4Xjm2TZxjsk53Nl+EwWywLSePZlDoOOA4SWAWMQeUTcQuBKWXaV+Zq2r+H1KJI/6LCZCV33TWf669/EceRuFzteVaoT1EIP4bIQOJQXX3jdvt449RTGeM45BoGxwQCAESkZGsiwZJ4nFiL+wcTcBsGNbZNSTxOpW1Tads7cprNz0Iew623buDXv17XY2120Fu7W0YVFe1Yq1I297jZgffIN3Ros9ft5bdrz5amgqlGoxn4aGFE0y8MHz6c0tLS/jajR3C7XWRmZZBfUKBCb1wuMjIycLvd2lNEo9Hs9Hzx5VYaI3PVoyvThOxsmDwZ9tlHVePMywOXSyVjldJBDRdSj7pU8upZs/Zg9epV/XMiHfAnIVqJH9/U16e3t/ymrwBmP/wwd5xzDnOBmcAQKfEJgRu4XrafBrLnhrqarlJfH2OPPUoBGyktpBxBbu65SCeM49QTj6/CceqROEiZQMo4kFDLTg2WLAMs3nxzPscee3SHfcVGjWLN559z8tatzdZPQiXz7Q9+9avx/OpX4xHiTXo32r79e6PhXRVGSkpg9GigjeSqy5bBww+mX3Y1+aq/hTDSkceIBL6HeteKgaVAISrvkLtLvWo0mv5CCyOafmHq1KltCiOD0cPC7XaRlZWJIUxycnIQQuAPePH5vOn8KBqNRrOz8tSTy1C3/jZK7LCRUhKPC6qqVHGKbaUQCoEjE0hpJfe1mxyjBhwlJcVtdzIAaPlNLoA12dncvGEDBXl5bX7Xn/jjH3P80Udz4NtvcxgwRkqGTZxIybr2n8rPG4S/gzsL06d/hcuVhxreejCEF4xs5a6BQBgZ1Nc/nLyGraQ4YpEWR2QCSPDaa69tVxiZ9t3vsuKFF3r7lLqFlMcSDtuMG/dJ1w92nE6U6G3/Gh87dFiXupNzzwAg47hvEb7uWigthcJCuOfuLrXTFi1DabbnMbK6yfKehsHHjkPqG+2kHbZGo2mbVGDqQGCg2NFdtDCi6Rfefffd/jahx3C5XPj94Ha7sW0bIQSmaeJ2m1oY0Wg0Ox3PPbeE733vnzTeQvhAHAuoASIkcJwYVVU+Pv0UgkH1UHfzZohGg8l9rORkN1m2sKyB6S9Rd9pp3PTcc9069rX58wH4nhCcAizvQBT5nxZF+hXbrgGiqIG7icAFwkRgIowAtl2JZW1FCSEW7QVJHHzwgdvta8yBB/LNAP68AwGT8vJru36gZcF2K8u0f180qmhI1/sEQq+/gfjkI3jy8Xb36eq7vWHRoubHNxVGtvPZtQyF0mg0Ax8tjGg0O0hK/HC5GhOtNp00Go1msOI4DtnZVxMKWS22NL19iIKMg1CiCCSw7SjV1R4WLjTYuFFVzNy6NUY8HkqGH8ST+zadx8nKyuyDs+o63RVFmvKslIwvKuKYiopW2xzgVT2QGgCkwrsEyvPJTo5/JaZIAA1IGdpuK3vtted297ln8mQm7ZCtA5TOCCMdXOs7kmNku7TT7bNC8Bmt5RoT2Kfp4dsJpWlKXP89azSDDi2MaDQ9QFvlfAEtjGg0mkFNWVmUUOhgYMF29nw9OTKYAOJUHEcSiTiUlPgpL3dhWREikTpsOwzEUEJIrMkU5eijp/PWW2/05un0OxvKywGYIwQZQL3LxYJ4XP9WDBCEsFHDYzWoTeeVSX8+nRvsjh07brv7ONFoJzwrBh9FQlC+A8cPH9I9j5HO0F6OEQkckFz+lMbsKi1lkM4mX5WobziNpk/QVWl6DC2MaAYtQggsy8I0GzPF90dpXH1Dq9FodlaGDw8g5ffIzi6goeFFtn/Xsx6kKq1rJSCYgGCDA7gQxgVIGQFiIKO4XDEuvngqd9/90949iQHIe/pp8oDEMJIeI0lFJCWRCJK/9Z383Nzu7afbnOr2sCq+8w2f9/B6eWsHji/Ize0pU7pFKghqNVDbYltXyvXGesgejUbTd2hhRDNgSQkOLROyTp06ldWrV3Puuec2E0WaHqPZdTjooIP45JNuJIjTaDSdpr7+COAIhHgfeJbtp1iTqFuMCPPnX8JRR83qbRM1mh3GNCW23UQOERIhG/WQzskiktzcywGJYeRTVXVDm/cm070e9vMP3NLU3eXNOXM6t+Pcub1qR7u08SEaHg+yhUg1JTmPAP7UoV0QNBMdd6nRaAYgvVmLS6PZIZ5++mkAjjjiiGY5O1avVnm/H3jggS6198wzz2AYBkIIZs6c2eP2avqHTz/9lEAg0N9maDS7BFLO5vLLLwAOa2OrgUqmGqG09A9I+VekvE+LIppBQV1dmEQiBqQ8T5PDWdEkkqbdIa7E690Hr3c//P6jAYltVxOLLWLFitVtHrE8HmdRNNqj56DpDK0/Q1cHXip+VCDhJlQojUWjJ1HTlgxgYZPX2mNE01cIObCmwYz2GNEMWE4//XQA3nvvvVbbuqLat/WkZtmyZe16pGgGD5WVlQBEIpFWYVRer5d4PK4/X42mh7n77pncffdMhCgEXgOCGEaCePy+Vl58Gs1A5aWXvuCMM+7H620e9iKEQODF5z8Kw8hQ7iLJnxb1e6K8SYTw4vHsjhA+BG5spw7pVJNIrEPKOqRUFZaKigrb7F9KiTPoi1sOLtq7HfAOGUKivP3MKIcn5+/X17M3TXLPtMAHLAUWA4uAbwHjgaM//LDbNms0mr5DCyOaAU1JSQmjR49utT41AK6uriYvLw/Lsqirq6OgoAApJaWlpYwaNapTg2IhhB48D1IKCgrSy+0lwNVoNL2DlEcg5RwdwqgZFBx33J188smapPChEoj4vJ7miUSSSGJEIvNSL5BIDOFFYuP17othZIGU2E4ttr0Fx6nHcepRZaebk5ub06Y9Pfl3U19fzhtv3M78+XfjdnsBsG0LIQTTpx/N7NkXMHbsXmRnD+uxPgcl7dzrBYYNI7h8+XYP99DoJdLRXeM+NFazkcC0adM6b6NGo+k3tDCiGdCMGTOmw+35+fk90s+kSZNYu3Zts3VCCLxeL1Ht6jpg2d6NpRa8NJreRYsimsHA8OEr8Pm+S36+jW3XkkisIBr9uHGHJpEz6ddNlpWQkoGQMRxZRyKxEcdpQDoNyO3UH3G52rnV7ubv07PPjmXFiiAVFTaO09x0t9uLlOD1CjIzvdi2ZMOG91i54l0iUYd//7tu1/6bbefUs3ffnfK33+61bgt76F5Vo9H0LloY0QxoVq9ezeTJk3u9nx//+Mdtro/FYtqjZJBSU1PT3yZoNBqNph8pKvqI8vJDcJxyLKsew8jAMLLweg/B5zsUKRM4Th2JxGqikQ+U40hSOFBiSB5SRnCcOqSzFkcGlSAidzyDxMZLLsF9991dOmbZstvYf3+DAw7ITq+zbYllSdastnjzrRDRqKS21qGszCYek0RjDrGYJBrtn8p9A4p2buXy9t67b+3QaDQDEi2MaAY0kyZN6pN+ZsyY0eF2LY4MPnL7ueSfRqPRaPoXn68IAMuK4/EYOE6sseCMyMIw/JhmLqZ5MD7fwUmhJEJDw0PYTrCJGBJMlpruPFLGSSTWN1tXUHAD0agkFLqeK669Fq69tkttrlv3HIWFCdxukZzAZQpMF0RjknXrLOJxSSwuicdkd51Seo2Skld59NGLqKnZgsuljGup06RstixBUdF4Tjvt30yceESn2pcHHcKQv95B5YEHtLdHm2tz99yzU+1rNAOSlpmA+5OBYkc30cKIZkAwa9Ysli5d2q740NvCxC9/+UtOO+20Xmtfo9FoNBpNXxMCwO+PoVJjJu/bJUjZgOM0JNd4lTeJ8LLfvqN5/Y21SBlCylCXe3ScMInEBhynrtU208zD49mAbdvdTFQcwLLAtiTRiFTjISlxuQRlZTaVlXaHR/e3t8iGDSdyyCFgmJCfD3m54HZBPAGvvw7xOIRCUF0N1dWS4k3rWbToSJ56qvP3fxW/uBKAgr/8mepDDm6+sZ37yMwpU9pcr9Fodi20MKLpdzIzMwmF2r/5CAQCRCJde1LTVVrmF9FoNBqNZrASTUTwuf0AhGIhLMfCkQ6GMMjxNyYDbYg24DJM4nYCj8tDzI6RYQZwudz9PojuGdRAeNo0LytXdjS4jiOdOBaSTz9bi+O0X6GkPWy7nkRiLVK2n5csGHwBEN1OFi5EBtD0oWwqNEYMmie1QoB0oKpSTSmKN8HKVS137n4/VVdfA4D31luIz07WlWlHGDF9PqQQiIHmYqPRaPoUXcZB0+90JIoALF26tFf793pVBvc777yz3X1su+OnMJqBxxVXXNHfJmg0Gk0/0Xh710zgaDnQFOo/gfLKFIg2dhq8pMa5Rx5ZSKrSbursmus+somu0BVPDoltVxONfk48vqxDUaTpMd0VnUxT5RZpPDq1JDuli1x4oZ/XX7+deDzcrf57ilWrBXaLAj6JXrrNiv3q18iDDoH3F3SY8PZ0x2HcBRf0jhEajWZQoD1GNAOeCRMm9GoYTUNDAx6Ph1/84hftDqZ1KdiBSUeC1V133dWh2KXR9AZSSowVBt6RHmKb4yxwL+CwKYf1t1maHeSYJccwv3x+s3XuEjeJ2gSEUREbDUAdUAOcBTjbabQAGAosBbbQ5qOq4rOLGT2qdcn67WE2/c2STYfRzQfk6lVqoJ7K+bDzCCOpczr++FHce++KpBoiVSWXVExNhyVp2sPBsiqw7Y04TldH9Nu7MNrH7c4DWlotAaNDj5GMDMHUqR4KCkxeffVGHnroep56qqHbduwQAqZOUflP7rkHLrtMrbYSai4lmCYMHQqBgAqtKS1dyvDhM3eoW/mrXwOQ+OCDdvfZ9/772ftvf+P55AMzjUaza6GFEc2g4KuvvmL//fcnGo32yE1bKBQiI0O5pLrd7nSm9g0bNjB+/Phm+95444073J+md6ioqOhwu+M4WtTS9BlfVH3B+oJlHDz9ABxs7ByH6/gN7/39Q+Ql2kV7MNNSFAFIjEnARiAI1NMoipxD49hXAEVAPrAc2EyjALK5SWMtv6bi8OPhP+6WKKK6be1TIGgtjEgAITCFQcadmfhz/DScX9+tPgcye+45hGj0AdzuKbjMgmbuIkK0dCQwaC2WpEhgWWVYVkm3Htj86U+ncdlll3fpmIaGSm64YS+i0QZME046KYDbbTQRR9T/La0pLDQYM8ZNXp5Bba1DVZXDmjUJGhq6L8z0BEI2vrNN38JE0oPkttsgIwCOA1XVsHQpPPXULDIyVA6SyVOe4Nhjzuh2/263u8PthsfD96Tk2Z1KINRoNJ1BCyOaficcDhMIBDrcZ9asWUSjykW1+0nLGhkxYkSrdVJKwuHW7qW//e1vd6gvTe8xfPjwDrebpqmrCWn6jIqCjdzJ33FwcJA4ONRRz6xLZjDi4hFs/cfW/jZR05NIcJ3hwrrVUoLI5SgPkEVAOY2KREmTY1oKIAKG1Qzj9QteZ4+Je/SYaYZhkLDiIGHGf2aw5OwlyGTITLP9EAT+FFDeK3kQcSJk3JtJ5PLezevVdzR+/1dW/grHcfB5TyY373jc7rEIkZlSh5J7tX6PAKSMYVmlZGaGiURqu21NZ0SRaDTIb3+7G7FYUFmUKh8sBI4DL73U+NlICaNHmxx2mB+QjB/vIi/PJD/fpK7OpqrKZv36BHV1zoCpUNOeGanQmpImfy9SwpQp8NabEExGXRcW7nip5M7wPSl5RrR1NWg0A5AB8vc92NHCiKbf8fl8XdrfMAyqqqooKCjo1P5nnXUWjz32GACXXXYZ9957L3V1rbPFA20KNDuXW/Guhy61rOkuj1U+xuXvXw4LQJZKlXqgja8DKSXexzxMMSYmRZHW//x3e8j+fna7fUkkeAET/BE/5U92PfmjphexaTP1hLXVgm1AFfBFkw1t/WxE4Z7d7+HCUy/E5erd2y+BwDRcyhMytJGc+3LVBlsS+1mMhIyT+ccsGA4UJV0m4srGWCiGbdsIIQjGg0oAMl3Yjo0jJbmBnA567ns+/HAlxx33y+Sr5t/1+Xm/b/baMAziif8BUF1dzcSJN+H1zsQ0RyBEMnwilcwUiZQhLGsbY8dmsm5d7wubJSXjCYclhx3mMH9+yrEl5RvSGiFg82abxx9vID/fYMgQF9XVNsXFFjU1NoMpPVpbtgoBhgHRJlrIjBl795lNp0vJ6zNmEPzmmz7rU6PR9B9aGNH0O6Zpkp2dTX1959138/Pzm72urq5Or/P7/W16fgDcc8893Hvvva3aqq6u7qLVmsHEjTfeyLXXXtvfZmgGGX7hp2ZkDZzZzg5NAv0PMw7Cwm5HFpF43B4aChtgJR2NcwDwZHh6/Fw0O8a5mefyYOTB1hskKiSmAiWc2ICAIZuHUPrXUkz3jnk3dhfLsXAcG6OlmmMKvDd5YQgwGmW/JSGGypWSzJNiGEYzQTm1PNCeExx11IssXVpGRsZ3ELhAmICJEAZgkrA2Afu0eWx+fj41NbcDMGXKiVRXH4rLNRHTzELKBhKJLcycOYolS8r67HxKSiyysw1mzfKw7z5ebFtSW+ewcGGMJUvird7//HxBMCiprnbYssWittYhFhsYDwIcx+LGGwuIxdS9nW3DCSe0v397Io5pCiyr8ZwmTJjck2Zul+OWLye0aROvjRvXp/1qNJq+Rwsjmn5nn3324fPPP+9yLojDDz+cBQsWAJCXl5de39nSvoZh4DgOq1a1rA+n2dm47rrrtDCi6TJ5Rl7HO6QGKV6wsHBw0tKI3UokkbgvdJP4RWK7uR39Hn+P2K/pOR447AEefLMNYQSgFJVf5JcMGO80t+kGU+VSkFd0PaeEnbBJyISqVtPEC2OgBRYcdtjJDB2qBtW2DZaVnBKShGVjJaztNwKsXv1Kejkr6ydcdNER/ZK827Kgutoh9awmM1OQlSk45pgAxx0XIBaT1NU5/Pe/9dTW2pSWSurrHYLBzl13UkpKSixef31Jr9h/4x8E4UjbAloqAlrKtre3J4y4XM2PCQS65mXcE2SMHdvlvCNfAX/46itmzZrVe4ZpNKi8PWJg/PQMGDu6ixZGNP3OwoULu3Xc+++/3yzM5T//+Q8/+tGPOn18IBAgGAwyZMiQbvWv6X/uu+8+Lrrook7tm5WVRUNDP2Xh1wxKssysTu231x6z2vAVka2WxswcyTo2brc9n6vvb/w12yevOo+a/JrWG5KRmQNFFOkJjnQ4r6MAAI5ySURBVH/+eF497dV0CtKUODLQhJGXXtpMNFqNlBLHcdRnIMGRSgxSn8n+XWqzoeGhXrC0c6hEsI2uaMEgBIMOUtr4fILMTEFOdiErVlRSXy87nTckkZCsX5+grCzWq2Fcs+c0Li9dCps3txZB3n3X4IgjWot17QkjhtleIty+53tS8rQQbRWQAuAD4NEtWxjZRh47jUYz8NHlGjQ7DT/4wQ+6tH8wGOxw+4wZM3bEHE0fcOGFF3Z63+193hpNSzJdmZ3ab6hZ1G4ITWpSAzfZqft7LYwMTKrPqG79+cnGymY7E2+Vzsdzr5fs+3MZ88AY3EJ5nwy0UBrpgHRkWhCRUuJIJZBIOVjFqlR+k+TL5DwalVRVOqxYuY3a2s6JIpGIw1dfxfjsszBVVXav57ZpysyZcPzxcNxxsPvuIm3vkUc66dwhP/954/7thtIMsMpyc6Uk7+CD069fBH4diXCDlLwjpRZFNJpBzMD6ttFodoCuhOKkSvV2xPLlyzu1n2bw0DI3jUbTEbsFdkMeKJEHSuJ7x/lBww+gDc/8BoId+Isk/0k1Udni4Dicu/+5yPkS+baalv13WZ+cn6bryG9JstYlPYkk/HHlH4nH4/1rVG8ioT7egP8VPzmrcsl6q/0Ewv2B7TQRQpICJCjPlkEpijQxuTHxqkoiLoRAAoaxfXWqttZm4cIoX34ZIxTq/7L1o0ZJjj+++TopVUneFE47EV+mIQacIHfURx+xt+PwByn5SsouFxHQaDQDEx1Ko9mlKCoqAlR1mttuu227+Ujaq16jGZzU1LThBq/RdAK3x81/j/kv/+W/6XWO5fDH9//Iy7zQgb+IVE+0k9PhIw/n9edexx/QeUQGK/UXN0kU/q3+s6NXGdJCVIgCyyWu+v5JJtsejqPsTHtkJZdT7gmDTRz54MMGZs4MkJPT9H1Wgk9KHGhP45ASystVed7Bdt4AyY+Pa64RnHSS5NDDwBCAEOkcI0uX9quJzdAVCzWanQ8tjGh2KqLRaIfK/Zo1a5g/fz5nnXUWsP1SwaY5sG4CNTvOTTfdxG9/+9v+NkOzE2C4DK476jqe4+lkHoY2wmhkyltE4kjJUw88pUURzaDl+Mxv97cJzXCSYTQqnKZpXpHByW9+UwvAt7+dz0knuZgwwY3fl8qxIdIhKE1xHNi61QKGUVxc0scW9xzJj49x4yRLlyoRxLIEhYVxjj3235x//nn9a6BGo9np0cKIZqfC6/UC8MYbb/Ctb7V+lJeTk8N3v/tdYrHYdtvqb9dTTe/wu9/9Tgsjmh6lKlpNri+3tTAinUZvkaS7/7ARw/rbXM0g444VdxD6JsS13+3/ylqXzry0v01ohsolItPJYdsSRcaO/SS9rbj44FbbByLz5lV3uP3YY/vIkD6krVAal0tSW4sWRTSajmjiJdfvDBQ7uokWRjRdwnH6P1a1JX5/49PXDRs2AHDppZeydu3aHWrXbi8TmGbAse+++7Jw4ULt2qrpF7Z8VUr2AVnpzCIy5SniKGEkJZDYjv5O0XSNIRVDqPRUwlC47h/XYV9kY4je/w2+IHgBvzn6NwwdPhRb2ghhIJB4B1guBcuysKy48haRjhIgcQAHpGyynFqvGai0N57SP+sajaav0MKIpk1umXsgv37601brT9p7Cq8u2THBoT266v46fPhwhBCEw+H0ugkTJgDssCiiGVyMHTtWiyKafuPx2OPcwHUpSUSFzbQQRRwp2XZfGVzZ39ZqBhOV/krwAAEgD8zXTVgA8pbefSp3/4X392r7PUVd3ZvY9jaktIAEyASOTAAxpIwjZQzHiSFlBMtqAE7qZ4s1XWWQP4DWaDSDCC2MaNpk1pxDeOSBv/Hj8y5PrytZ9CyHnXBiP1rVnNLS0lbrHn30UX74wx/2gzWa/uS5555TGftlk6oEHQglQgj23ntvFi9e3FcmanZizjz8TH5R/3OysjNxWogh6bl0WHH8iv42tcscfvPhfOD5AAqADMAEbBCVAucS/QS+14mihJHM5DwDKATxhOD08tN5+udP96t5/U1FxcAK7dnVSSVJ7cljrTYqgWk0Gk1vMLBiIjQDhrFT9mL2fpMBFT6z4qtFvPvKy/zqpjv717DtoEWRXYuU+LHXXns18zj66KOPWu37u9/9rtnrL774Il0CMTX95Cc/6V2DNTstI+4Y1UIUcZCOkxZFSp8sY7dpu/W3mZ0m/+x8xF8FH0z+APYEpgFTgKnADJAHS8RHgtFPjO5XO3dmNoQ2QDkQAmzAC+QBY4C94Zk9nkH8Q7Bu67pWx7659c0+tVWjAZg9W3L44bLd0rvt0ZFXyM5cDVuj0QwstMfITsRVP72EUCzEd884k2O+fdwOtSWKZnDL767hO8e+g/C4+MOtjzNtdA4/ur5nbNVoeoLs7GwmT57cyvPjkEMOaTM068Ybb2TFihVMnz69zfZuu+22XrFTs/PzxfVfIJ4QjPneaJXnIO0pIin/uhLnJ33rXSGl5IP7P+CD33xAwBvgF6W/6PSx4kcCDkYNwAuBbFQohwflMSJQA/UYbM7azKTnJ7H2NB2+2NOMzxgPq4HRKI+dTJQ4EgDcpMNrJi2ZBA+B/K3kHyv/wSWLLwFA/kDHIAw6HIfnS0oQjoNbCDyGgd/lwi0EBmAaBm7TZFZhYX9b2iFz5qhkkO+9b7RbXrgl7XmMdCJXvkazSyOkmgYCA8WO7qKFkUGOlJLKyiruv/XPZGf6GF6Yy7LFSyhZuYbKhjryhw7h/Isu7HK7M2btxTW/u4iSFSuw7Rj5eRlkTJrZC2eg0XSfurq69PL2SjWnmDZt2qAu56gZuMgzJeL/BBkXB8gdloNt2ZQ9WI7zM6fPcuBUlVVx17C7mMQkhjCEOczBT+fLA5uHmHAqMAooQnkoZAE+lDCSGuRYydcC1o1o7bGg6SHeBfYBxqM+jxwaPwsXSijJAgpAPCXgPeDQ/jFVs+OIhx/m2HPPxYXSIFPzlstPLlpEeN99+8/QziAEc+YocXjBAmMHQmy0c7tGo+kbevXb5qabbuLggw8mEAiQm5vb5j7FxcWcdNJJZGRkUFhYyM9+9jPi2m+u0wghuPeGP1JTXUk8HicatYk1BKkqr8Brughuq+I/9z/CM/96kIfv/Fun222oraJ6SzGLP/uKeS99ygUX/4iirEAvnolGs2N0RhTRaHob+SdJcHyIDeYmSjO3IX8u+0wUeeDOB7hr2F1MYxpjGcuo5L/RdD7cxdndUV4iOUA2iGyBP99PUU4RI/0jyfXm4va6EW7ROEpzw82f3NxLZ7Vr49zpwKvAEmADUAY0ADGU544fJV6NAmYBc5Pbq8CROgfMoCMSwYL0lGhnkoMo2bgQIh1i09EzifbOKBbTJc41Gk3f0KvCSDwe5/TTT+fiiy9uc7tt25xwwgmEQiE+/PBDnnzySZ577jmuuuqq3jRrp6J08zZi8SiOZWEID263IBKN4vJ4ELbEECa1WzcTDYUJNYS332Cq3aXz2VZajT87gzknzmHlV6vIDZhUV1X04tloNBrNzoHb4+7zSkn/+8X/yCGHAAF8+AiIAMNzhzPkwCGdOv7yyy5X3gjJR9PCJRgyZAjnuM/hGZ7hDu7gIA4iQECN2uJABAjCB8s/6L0T24URQiCfkvAEsBBYB2wFaoEw4KDCarJQgtZ4YH81N282tXfcYCMYxEb9eb1//fXMAY4EjgGe/uijtDAyWEu1dEYgSbFsWTZPPSVZsGBL7xum0Qxm5ACbBjG9Gkpzww03APDwww+3uf3NN9/km2++oaSkhBEjRgBw++23c84553DTTTeRnZ3dm+btFNx70024XC5MBJFoCMeOkZlVgGGAMCBh2/jdEA7XkdUFj4/KshrefG8JG7ZsYfjaSvaeOZWKigpOOGwOI0Zk89z8T5rtf9NVl7P7yEx8w3fjW2f+uKdPU6PRaDQdIKUkSJAGGqinngwyCBgBXAUuOLbtY15c9SLfffm7OCLpWTAOWI5K9hkGERX48ZNDDjXU8BmfsZrVhGpDyKCEIFAFlMDr5uuIN0Rj/guAOORsy6F2bm2vnvuugPxYkjsyl7or65QYtRbYi3T+F+EG6acxvCYDyAfjCQPxvsC5X3uPDDTmoj4qT5OJYBALNbYQDQ28TqNjFsEgCZQWNtiZPVuNnt5/XzSG2CTn27btyfvvf9k/hmk0ml2afs0x8sknn7D77runRRGAb33rW8RiMRYvXswRRxzR6phYLEasSSam+vr6PrF1oBKLhZESLAOyPW5sAbF4GK9P4PN7EPEEDQ1B8nOz6Gya8FAozJOvzEd4/eRkZlJeUcP/3niPbF+AWbtPYv+JuZR9PQ8ESGmytaSCc885hdqtpYTC9Syc9zD7ffuc3j1xjUaj0aQRQrCEJRRSiEQSIUK9XU/Fugqy/pjFfjfsx/df+D5Pr326uc96S6eWGahQjK/A+dJh04pN/HWPv2J4DRJ2AjtmI+NSiSd1QKnal4tozHuRatOCuqI6xPsCOXuQP0bqZ0579zTq/lTXfOWXqBHzASAzQPgEuEGaUolTfiAb5FCJuE9w+/DbufLkK/veeE2bLHEc3IaRFj4MgIYG7OR20+1OiyQSQAgslDAymEJpOqKpQDJ1t19z4x90SJ5Go+k/+lUY2bZtG0OHDm22Li8vD4/Hw7Zt29o85pZbbkl7omgARxJ3bISEeCKBy+vDbYAVixOyHaSUZGX6sWyQ8UinmszICLD/ntN5+LF55GW4MPFQH7OpDVbjDfhZX5fLZ++9zcQZ+7Nx1VKmTJ1IpKYU2w6TiIXxeVxsXPwi4/Y5pXfPXaPRaHZRHMehdE0pmz7bBMD4A8dT46rhXetdqqlmLGMpoohccgk4Ab5+8GueXvg07AdUt9OoRI3OfoTy5f8GuAlil8XUQDtVjSYK1AOVwHrg/4D85LEi2U4qQQJADmwLb2NYYOfIFSAeF6pKTAhVTnctyL/0jvBTn6hn75V7kvDH8QTcxIMJ9R6nxsU28DGwL8gsifALlf/FB9IllVgVAHLgqq1XcdWtV5G4JoHL3HVy7+/90ktUjhsHtgOODa7Gc5c00QZTZVGkRNq2CoUzTb6cMIGCXvBgjlsWpsfTzJZUKI0ETJ8vvSwBXC4sejkGvp+YPVsye3Z/W6HRDE6a/iT0NwPFju7S5V/G66+/frvCxMKFC9m3k9my24rBlrL9ZHW//vWvufLKxice9fX1jB7d+cRyOxt/fvjBHm9TSgcjHmK38UP4ZtUGQnHIyw0QDkWxbcnaDWV8+5DpVNXW8sWX6xgxaRaJulI2bdjIV0s3MmbcSCrLyviNFkY0Go2mR7h25rUMWzaMIorIIgsfPjx4cOHCwGArW/mMz/gH/+BBHmQkIxnBCPLII0CAF859gfPOPY+rr7qa27ferpIWVKNGXEcAU4CVwOfAUGBPYAKwW3KdN2mIhcptUQ0Uo0SRkTQKIk5yH2gmkFz1wVU89q3Hevtt6nXEPKGqxCRQ4pAbcIG4VrD6gtVMHj25R/vLdmezbuMG9cIF5LbYIanHeNa5iY9MIHMlIgPwiHSCXOmRKndMMrzG/bQbFoK8Y9fw4gmccALThCCRSJCwLKxEAsu2cWwbu8nkOFKV2k6W2ZaORCKZ/tRTlP30pz1uVywex9tSGAmH02KI4fenhRAJYBgk2DmFEY1GoxkIdFkYueyyyzjjjDM63GfcuHGdamvYsGF89tlnzdbV1NSQSCRaeZKk8Hq9eL3eNrdpegYhDMKRGDMmjmLTlnKidTGisTh+n4+yqlqkAS++/RFHzzmU9du2ceUvb2RMUQ6jx43Dm5PFK/M/55gj90Q6DqKzBew1Go1G04orzSvZz9mPOcwhgwz8+PHgwY0bFy7M5D8DA4HgluQ/gAM5kIUsbNbeX27/C3/hLzzz9DPMfW4uXAa73bMbKx5eoZJ27g9clnxo8TUwAlX1xI8SP2KoEJqNICIiPWqTjlTeC6nMkQmUZ0ky5Oacief0/pvVy2yq3gTTgExUng8HdZ4xIApT7pxC9JYoXk8P36PUoD6Dtkg+Q4rbCdikbJF5QJZE+JLiiBclkLil+hwzgUIQ/xKM/GIkm/+xuWftHWDUVlaybetWlYhWSuWBkVx2WmQBbZmsVkqJE432il3xWAw7M7N5rsJYrNFjJCMDCxVmIwFMs1Eo2UlCaTQajWYg0WVhpLCwkMLCwh7p/KCDDuKmm26itLSU4cOHAyohq9frZZ999umRPjTdo2jSODYvXkieV1KFTSjhIj87QEM4SGYgk81lIdYsW4fbcCMMF+V1YSZ4vWzZUkw4GmHN2hK6XbReo9FoNNwsbuZwDieffDLISHuJuIUbN27chhuP6cHn9uFxezDcRjJLI2DB+uh67JhN3Iq3avv0uacj5yaHZMe37js1QBTHJQWSGmAvEE8JzEkmhjBwcHAcB2nJ5rVFY6Sr1VALbIFjjj6mR9+b/mDc6+PUe+VBjVT9qPMMoJKgFoLvfB/yPz3riSF/1H57Ke8G8zsmHIDy5hkJxEFmS5WY1atyj6TFEW+jzVuGbkHcJ5AX7bzeI8oTRJ1fWhShRfGENsqkpEUS2261rSdIRKPNEqlKwHCctPhhZGQ0eo9AOpRmIN5Z1dYqq1LvYsqBbONGwV577gzpYjWaAcxAqgYzUOzoJr0aZFpcXEx1dTXFxcXYts2SJUsAmDRpEpmZmRx77LFMnz6ds88+mz//+c9UV1dz9dVX89Of/lRXpOlnvvODq+EHcMl29gsF68nI1J+VRqPR9CRzp87laI7GlfyZdnCwsbGFjSlMRtojO9WOiYkff7ftkK9v5y7HABEXqnRvjOaiSB1QDuIzAYO8WJl4UMBolKjgRglA7uTr1JQBDIHcQ3Kp/ai2b+wSQpX0fUUi9hDwLZQ4MgrlPZIjIROEX2C6DKQPpOnguJPhNXnAHSCeEMj3B/kdbTukwmPUsmzTKyQ1F6kcIyRLJUvZ6cT1XcWKRJSwaBjpcY3hcqXFkMKf/KSZcLLvQQcNWI+R8oq211dV7pzXlEaj2TnpVWHkuuuu45FHHkm/3muvvQB49913mTNnDqZp8uqrr3LJJZdwyCGH4Pf7Oeuss/jLX/7Sm2ZpukBH+V7+9/wT7L3fAe0KI45jIyWYptnm9o5wHAdDh+FoNJpdlLWr17Iv+5JJJgKBjU2CBB7pwYNn+w30JSHS4SRpUaQelZh0NTh/H1hPjC3H4if/+gm7D9ud2dNmM2P0dALeDCUyJJ/HN/3dEw8LlXOlGjUqdSUnd5O5h7Q4Uje8RfWYPkJ+JRk+fjjbjt4Gs1CfRQxkQnn0OAEwPAaGx0S4JI7LRv6QxqS5OyltiSFtkfrM03smF2QvCSORZcso2bpVeaTYNtJxkPF4WgxJiyU0fkSp5YTjIH70oy73+cB553FuL2U5jcUhM6PxdSLRa842Go1G0yv0qjDy8MMP8/DDD3e4z5gxY3jllVd60wxNN5HS4emnn2fu3O+2Ekdu/NWv+fjzT3jtnTPbPd4wOi+INBVgtCii0Wh2dZ5a/RS3TLkFB4dCCskmGz9+vHjxyAEmjNSjRJEwjZ4iyYo1zi0DRxT55Ze/5E+r/qTyoARQYT4fonKFOKj1TpPJl5z2SM6DyX1AjU5N1F1Uqt5qyoMkGzaWbGTc6HF9cl5NKd1QyrKly5h5/UzYG5VANwYyLpVAkgGG38B0mxzs348PjE/73Ma+JiVsSCnT+k9aKOlAMJEpZaSXRvfylFNarTv9tdf4ksZqOSlxxKExCasD7L3PPoj//KeZptWyMkXT5dRZ/gv4J9Abn7p0lBjSFEsLIxqNph1uueUWnn/+eVauXInf7+fggw/mtttuY+rUqf1m065Tr03TDQRZfh/333sPF112OQ89eC8/+vGFnHXGGXyzchVLvvyi2d7pmPRuuHimXFaFEFoU0Wg0uzyTJ0/mcdfjnGCdwAQmUEABWWSlc40MKKpRXiMNpMNnWA/y7wPHjT7wSYCIjKgQk9ToMjW19XoM8CUwE5VLJAhsQYk/qdyqTR/jp8QRl9o+/sjxyDWSO1+9kytOuKJPzjHF7jN3Rz4nEQcLOBT12YwEYuAkkjlhApIP5u78oghAoqwcq6YGmUggbRssK72cnsfj4DhIy1KhM5al7mlsG2vzZrJ/9CMwTY7Zd1+eu/TSXrN1fFERi2n0FEkJI02LPhnJuWhjgo6df3r1L7KNxi2r9TqNRqMBeP/997n00kvZb7/9sCyL3/72txx77LF88803ZGRkbL+BXkALI5p2EUKQiDTwzZIl/OGGG3j5pZf5YN4CVq9ZxdnfPwnT5W7zmLaoqCjllZdfo76+gVConsrKWkwTbrn1FlxuT4fHajQaza5INBHlzjvv5MFfPMgUpjCUoeSQQwY9d8NgWTFeeeU8qqo2MG7ciRx11K+73kgpShSpAbYBq0E+O3BEkU3hTUTyI80fv6cEkKbrUvPRwIvAgTR6ipQBn6FKGwdoPmJNTSmhxAQKVN+/qPsFv7j4F8h/9P37IT+WiBwBc4HdaQytiUushK3Cn1LsxAPYafvsrfIC2zaWZWEnJytVpte2Va4PKZGOg+0kkwo38TRJVbH5UsK0Sy5hxd//3iu2jh8xAttxMIRACpEWR0STeXuiSFvCSNpDpsm6XrsS27iF08KIRtMHDNLkq6+//nqz1w899BBFRUUsXryYww8/vIcN6xxaGNnJ6CgnSHc4+ftn8tmnX1JX28D4sWMJFBUR/GoR51/8s1b7puNzW9jw2huv8dG7H+Dx+YglYjQ0NLBmzRqQcPzx3+all58nEGidp+RHp8wmZMW59W8PMHn89B47J41GoxksXHHFFVxxxRWceuqpPP/i84xnPEMYssPtPv64oKAA3G7IygKvF2prP+bBB39DWRnY9tH87ndvda6x9UAVsBmyl2RT91Xf5NgQ//ufGnmtXYu85pp29xtXMU4JHE1vHlNzp8lcou6KlqMq8cyhURQpBoYI2ChVwlIfpLNk0qJNaKwONArYG4y5Bs7TfR9WJOskr7/+Osf/63jYi7Q4YvoN7CapPW87/7Y+t62v+Gbx4qSwAcIQaaEjnVwVkiE1AkmT+5ekEtHUG1ZKSXVpaa/ZOm74cCXKCKHEEcNIa24tdbiORJG27gL7Y9yUSMDqNcqaSRMl2iFYo9G0R12dunfIz8/vNxv0V9ROQKr8nMQmFg3y34f/RTTSQCJhpW8Ampao6ypZQwr5fOHHVJSXYUsL0+3ikNlH8Zebb+GTT95VNkhJONKAlOrmIR6P8cXCj/nr7Xfy5cIv8PiU77HX42XTxk1UVlZSXVNNWUUZhx86p+2OozF8MYtrzvkRCxd/wMCRQzUajaZveeGFFyiVpXwsP+awPx/W7XY+/PAKPvpIMHMmjB0LI0bAsGFQVKSmYcNgzBjw++dzwQWdFNk/B14B+YjsE1FEvD0fsWolTJsGE8bDjN0Rv/99+79xqdUtQ15cqKSpTavLDAPmJfdPiSIlwEZgk1TJGbaiPGTiKE8Lu40plWshV7Up95Dc8/g9PXH6Xea4447DedaB+cBSYD3Yxc1Fmv875//6xbY+RTRPxNpUFGlVwFeqHCON+zTun9WL3q0jhg1THiy2je042Mm5JWWbl5nVzamvsJrkHNGiiEaz61BfX99sisViHe4vpeTKK6/k0EMPZffdd+8jK1ujPUZ2AhwnyktP/Zd/3nkHmVlZxBMJnvrPP/F5fbg8PmKxKMed/D3mzj2T3IKiLrdfUlJMXmYO1XXVVFVVMWXydJYu+5K///t+Xpz3KrvPmsm+++xLTVUVI4uGsXbdOjAEQwoLyc/JRRQUsHLlN6xevZotpZupqKpGSonjWJimifT6ePiRv3POjy9J37QYhkHZlm2YbhPbNLn55+dzzClzuPjq+3fm5PkajUazXa6++upuHff007szZcpy8vLA51MDFceBeFx5jLhcIETjutGjYe5cwdNPdyxKyyf7TrQWH38EU6eqE7BtZXB2AwwdinHggcjPPmt9UITGch6pCdp+1L4uuT+oXCmbUaLIemA18CYwEZiB8hqJoQSSRHKKoxLRpvShDCAbGAqXv305l5112Q6/B91BCIF8TyKmCTiyxcZ+zo+7bNsyjv3mWBzLSbtFCCGafTapakHNPq+keGFLm9fGv8Y+E/Zpt4+mpXibrAQhWjkRqSStLe40mnSf5+9++evtMbSgAMdxVIUkKRGG0ViSOTlX1ZO27ynSVvLVlsu9TTyx/X00Gs2OIaSaBgIpO0aPHt1s/e9//3uuv/76do+77LLL+Prrr/nwww970brto4WRQYyUEsuKc/Zpx1BZUYPhGGzYXA5IcvPyicRC2LEaPB43Tz3wb9566Qmeeb3rydbiiQaq6uupD4bxlpZSXFqCaZokEgk2lWxky7YtFBcX4/J68AgTKSVZ2dl8tnghhUOGsGXLFgwEFZVlJJKxvabLxLEdHMtGIFn61bL0zUAq+arX68UwBbvtsSdbSys4auZM/vPAjfz4vGt7+J3UaDSanRspHSZNWk5REWRmKhEEwLYFpikRQukMKZHE71f7DRkCM2Zksnx5sH9PAJi+YAFMmgiZ2WAI5afvjoHLrYweNpSFX37Jfnvt1fzAKho9RFLeIi2FEoESOtY3Oe4bVGLZjcBaYBmIfQXyZanaGZHcL4wSUyKoRKd1yf1BeaH4UeLIMLjujuv4w5V/6Lk3pYvIFZKioUVUmBWN6xb07x31ccuP49KjLuXF8hdxHAfbsrEdlfdD5fqQSOngOBJHOumwGKn+A+CkZSexdcLWNttP7Q9NS/G2PuemYcASVcHGSVeyEUmPExiW3Tr0t6fIzclpFEZS4khTQUQIaCGQIESHITQt6ctPu2WVGo1Gs2tQUlJCdpPvSq/X2+6+l19+OS+//DILFixg1KhRfWFeu2hhZJAz74UnidU3IKRNeSiObatnCFtKKzBNgcvlJifXhyMtirfUdrrdpjcIE8ePobaqjoQVo6qmGikElmUlk5WBlbBZuWoVw4qGUhdswLFtMgMBcLlYuWYNIMjJzmTY0FGUbNmEMAwsy07fmMTjca6+8opW1WiyMrOJRiNMmTqZiPSzpbiYw0YU9tA7p9FoNLsOr71mMn26yifi84Hb7UIpAQKXK4ZtJzBNiWmCaSpvEtNUesPQoaH+Nh8pJStGj1KiiM+jRnfpMqoqMSami/0vvhj5aYsHACWoU02FzXhoFEqaltvNRpUZTrEelVB2DUoUyRPIfZPDynnAwahQGZvG5LPVwGaQdnK/ZJUa/GrfG9+4sV+FEYDysnLKy8sZetpQHr340X61BcDINHh23bNUl1Ynw1Ukskk+GNlS1Ugl26DxtVPffl3YdJvJfVPtCUQz0USondNdNA25kcltUkoO3nPPHTndDvFlZuIEg8obpIkoQgtxJC2ItBRHmoT5tBRJUm+fLWWPx7W0F8VmaWFEo9klyc7ObiaMtIWUkssvv5wXXniB9957j/Hjx/eRde2jhZFBzj1/+xtV5dU4QhCJ2BiGQUFBAfF4nNzsHELhMLFolNraWvLy8rrVxwH778/qbzZx5ve+z13/vA+ZzORumi4cJ3nj4DhsKyvFNF24PB4agkF8Ph8+jwePx092dhZVNZVEwspH2TRNbEe143a7qa6pZ3hSJEyJMrZt4/K4qaioBCPBw699TZE/zF9O2AVioQcYTRPVNZ1rNJrBQSp8xu0Gt9sHTAP2BYL4/YuJRouRMobjSGyb9CSlEkj6G+PTT2HKFCWKmG5wkpkSHKkSsMZjyt2lpqbVsZ5VHuKBuBIn/Cihoi2hxEvrBAx5wEHAAU0G6KDCZZ4B9gQyUSE1VcA6GFk6snG/VJUaT7LvHc+b2yMUFRUhPxwgvtc2JOIJpCPTGhekvDZE67Is0HwdINvXRYh8/bWKD3Mc1bhtI20HHDtZvtdG2la6lK+TSODE48h4HDsWR8aiyEQCw7IwHYffP/dcb70TAM09RlACSZteIu2IIulf5pa/0UmBx4IeF0amTmn7WpoyuUe70Wg0OxGXXnopjz/+OC+99BJZWVls27YNgJycHPy9GLLYEVoYGdQIauoaiFuOCrPO8RMKJrDsKPWhMNuqKvB6vGT5A0QiEZwuOFA2HfTOOep4YtU2Q4fk8/f77ydi28kwngQul0d5j2AgHBshHGKRCIZhEAyF8PoCGC6TcChEbW0Njm0hTHXZGSZIaeBImwUL5jNj5r7NbHAHAkRjMSLROBPGT2VZXQMTx4/kf688y0knfq9n3kJNp0hdD47jYJpG8n7LQUrRbLtGoxl4bNu2DNNUY0LHEajR/veBy4CFwJ8RooxEIkY8DrFY4xSNqumDDz7gsMO6n/R1hxk6VLmvmG4VRuMAjq189aMxCIehvg4CgVaHxs+Pwy+AHCALJWQEaC6UeIDa5OuWtJGDY8GVCzjs0MO46uqruOOJO1R7FXDfH+7jwgsvbNwxNWBPhfAEoKKigiFDBohCMhBI50Bt4rohkx4daqHZ+pbHAR3mSSn/6U97zNS+QDqOyn3SVBxp8ZoO5tvDlrIxlk6j0Wj6iX/84x8AzJkzp9n6hx56iHPOOafvDUJXpRnkSN758BMqy2uw7QTCVj+oNdVBrFiCTI8flyOoKCsnHo8Tj8a71YthuPjWaScyes8ZfPuEYxESpCOToTQqUSqOlX7KASTjhC0S8Si11ZVUVlYQi0aVS6rtkEjECPgD2LaNECb3//uJVv2OnD6L/Y45jszho/EEAjiGh4QpefnJ+3foXdN0H8O0SdSuQSY2E9+6EDu8Bek0ABCLRrZztEaj6Qtqazdxyy0Z3Hyz4JZbBA89NJNwGCIRNakMoauA/wFvAZuoro4RCkmCQQiFoKEB6uuhrk45YXz++W3U1PRemdKOOOq99yDgV+4uRnLgZzsqs2MspgyurYOKSsS0aa2Ot0Za3HLBrfAS8AGqis4SVFnelah8IBtQoTMZnbPpsEOVSHT7X25HrpfIZRJZJpuLIqA8SVKlgJNVcZ58+skunf/Ojsoh0roCTFMPnXRYS1ooaRRRJFJ5m+wkOFLiSIlM5lhxpEznW+losh0nXdGm3f2SVW40Go2mv0lXTm0x9ZcoAtpjZNAT8GcwevRQAj4v3kCAtZtX4PJ6EQ54HRtL2pgZXjyWB2F34GvagpahEo5t8c9HHiI7+ZRLAIYwsCwLn8+XFElUaIyKD3YQwsRKxLGsRDpLsQSEcHAccOFgmi7mHLI7b7y9ECkd7ISFy+MBYOKU6QDEYzHefns+AY+XN99fzr67dS8kSLMjJG9KY5WYRoLwxo1E7QSbPn6BSbPP5q3X53HqDy/HcZxWuWI0msFIPG7z61+/T2amww03HN3f5nTIv/+9N+XlX7Z6WNz09SefqMm2JfvsU83hhz9PRsZ8amujlJXVU1sbp64OamuVEFJTA9XVUFEBpaXg97/Kv/89gtpa2LgRHnus7wai73g9rUWRRDJ8JhRS6k1lJWzegtxvv1bHm8LkV8f8kl9V/BKAuro6cvNzYQJQhHKgyWoy5aHyhbTDskuXdd74EI1Va2zAgdUrV3f++F0BpzFMKV0Rpq2SKrKxOk06qialk+xEY/2mD5maeoggBI4QaZ1NCIFHCLKAAsNghBCMNwymGgaTDYOxQjBUCHKaJGfVaDQ7IU1yMvU7A8WObqKFkUGMEALDdLO1soZJ44ZxyD570NBQR3F5LYbPhbAlXtNNKB5LeqB2L5QGYM+9Z2KYJjX19eRkZxOOhLEsB8MQgINpCmxbYtuJpAeJRBo2jqW8SoRpYKRuaCSMHz0EQ5g0RBJcfc31vPXed3jg33dz/k+vSIsybp+HcDhKTSiKz5/JxnWrqKythMq1TJs2jKVfb8bl1pdwb6Oe3CWgroRYbQmrl31IvKGKZcu/orre5ulX3ueqm/5OcNtyfNn5eDL6P3mSRtNdYjEbn+99lOuAF6jlD3+4H1iOlHf3r3Ft8OmnglNPhaefVqLG9jBNKCmRPPlkHevX17FiBVx2GQSDykuktlYJImVlShDZvFm99vlUKg+/H4qK4JRTBC++2Ed3QD6/ir0ElSfCSkAsCuGIMrqqCrZsgWBDp8pg5OTkNCZHBU749gnMe2ceDEWV5D0CFfrSxrOEWZ/NYsbfZ3Te9upkO6mqNWHBqd8/tfPHt4N4QsA2kL8Y2Heh2WdsJ/kekHtRTnPvkCbVZhrL5KaqwkhVmrLpUF+o0KqsM7KYXDSZL+7+omdPoo+pLyjobxM0Go1ml0SPKgc1Esex2Vpdz+F7T6WssoKfXX4BNXUW8995m2AkgbRjNNSGqAk2EAx3v7LAm2+9ydHHHsfkiePIDuSwcs1qwMblciEMlXPCSlgIITANA4/XTSKewDYdvC4fPq8Xl9uF6TIIeAwuOucUHn/uHbJDMaZMm8lNv/8lN/3lLs7/6RVNnpS4CcUaeO9/TxKKWNSUbuIvv7+AS69/lvO+uwfjhuex516TqIk28NEHazs+gQGElJLf//731NWGuOvu21ttsyyLr5Ysw+ULMGn8KAKBQFKA6nvCoQb8gUxi9Vtw6ktp2LqKxZ++Q2VlNdN2n0Vp9WbueOobvv/9d3jh6Uf4/jlXMGO2FkY0gxMlinwFTEJl5YyjElIYQBQhfoiU/+1PE1tx4IGSH/xgJOecs5WZM5WI8eKLjd4iUioxo6BACR0rVijnivfeg4cfhj33VJpCKnymtlZ5iWzbpjxDVq+G//xH5TWVTRJj+nzwyivPc+KJp/X+SZqG6tiyQTgqfCYShoZ6JYqUboW169RJR7oe0vfqvFfTy7ZtK8F9LuqjT90lbYOyf5ZRVFTUtcZLkm3Uo3KYlEuOPPLILtvYFPEfoRLCZoB4VfBf73/5wdE/2KE2e4PsC7Ix8gyEW4AHNfeC4TUQPoHwCoRPkNiSoLayVr32CLWfG4QpwEzOm5ZWBoQhEE1+F80ik8DhAapFNeNWjiP0dZCKuZVt2qXRaDQaTVtoYWRQo0rzGgj2mjoWx/BQUV7PkmXLmTpxPKfP/T5e06GhIYw0BNWV3b9JGD5iHLfcdCMnnHAKX37xGT+94CIyswTxmIXX7yVhOSTiCUxTleLNyg7g93mQgM80yc3NxhQCjymorK3l+O9ewLz5n1NVXw/AD39yKX//532EQg1kZGQhJVRUlnP7rXew57Rx2MEq6kMOlg0H7J7J+uJa6iJh3np/OaMLszn/tGP413NvDvgkoI7jcNxxxxEK11JZUcfuU1/EExjCp58twLIs3G43QggmT5nA6AlTVHWhQCb//ve/OeywQ8nOCfToOUq7GmHmt7vdH8jADm/DjRfyR/HnP/yWWKKes+eeQSjSwOGnXkbuyH/jyyjgi1WVTP50ATNmn9tj9mk0fYnXayLl3i3WjgX2AI7vB4s6x2OPbQHg6af/g8fzY045BYYMgSefhHXrYNMmWLVKaQhnnw2TJsHs2Wq9ZTWm6WhoUCE05eVKLFmzRukRQihHjHi8cUok4LLLvsuJJ/aBx4JlKy+RhBukkxRFglBVrdSedeth1UqYPk0lYN0BTNPs2XwVS1BJX2PAJuDLHWtOvCBgNio0pxDYD34474fMtebidrl3rPEe5OUtL3PFP39OlBhhwkSS/6JOlJgVI27FicfixBNxEokEVsLCsixsK5kfw1aTTObXwE6WQFZ1dNN5SYB0ThIgXcXGZQyc90Kj0Wg0gwMtjOwERMJhln36Js/852H+t+BFpszYjT/c/ncMQxCN1LNqyfvIujDhyo1dbjsV1iKE4ORT5gKw3wGHMXX6ZEL1tUQjUaQQRMJRYh4Ty1ZJWPfbb08iNdVI28btcZORmQFIguEYoa3b+OkF52NHI8TjNobhxjQNPv3sG4INtek+r776lzz99OP86ORDeealhSR8Hp6bv5CqGovSsk00hJSf89bi9Vx2zvcYDFG05/3kHLyGQ71tU19fj7BizHvqOYKhGnzeDBKJBLFYjFAoxMKPF7DP3gfS0FDP3nvPxLItxo2dzKbinvOOkbWliIL2hRGcSt556Dr2PfVyQhs+4Y+PfMqdV53A1ENPxcgsRBo+Lrrm9yx66V4mTJ7JqiUf9ZhtGs2uzPz512EYIWbPvgXT9HTqmLlzfwT8CCklN95o8Kc/Cb75ZiNjxoxJ77PPPoJRoyArS6XtcBwldKSEkeqk1lBRoQq8PPCA2paqThMOqykU6pZzRvcIJw2QqBrCDQ3K7WXbNli/HpZ+DVIinngKOXeAVSx7GJiOeo6xEmR590SXsoYyhi0dBocCucDXKIemQmBP8FzhQd4zcMJqXix5kXc3vNssd0hTpNNY+h0kQhhKAGlajjdFi2o0kmQOtGSikXRJX0CKZMnfxMB5LzQajUYzONDCyCDH5XIRlzYz9zuY5x97mCH5ufzzgUdxu9WNdEZmAXsfehqO4zCrC8lXARzHxkjFdbfgicee55dXnseqtevx+zxY2Vk4GERCEYJOBCcSxOf2IN0WLhOOP+E7HHvCXIo3reMHZ55GXXU148aOYUt5BbYdwzRVjcSsbJVYNSWORBpq+HLpcv79/LP8eO53GFuUwaMvr6FpZrYRYybw/DsDP6b4Z5ecSyJWRzwRRiRsIpEIoVCIeCKEx+OhIViLx6O8bCzLIjenkBUrv8ayLBqCQV5/43UqK7dx5BHf4p1339hhe+KRKmzLarM6JajPoGL5Eg49/QIqy7fy1jsf8YdLvs0PTjsdkVVIw8qPaAiFCZpDmHnQCRxU5WHv/a/aYbs0ml2Zn/1MMGMGjBoFubmwaNEdhEIwe3YU0/R2qg0hBNddJ7nuutbbFi+WmKZg+HDIyADDUF4j4bDKYRpKRlw+84yaB4OkK9qEQmoKBtW+FRU9c87bpbwccnLBl3Rbqa2GrVth7Vr4agmsWUskFsO/Zi0890IfGdU5ZMWOD9CNOw3k0VIJLBmoXCXlqMSxPmAYsAcs+HQBhx94+A731xNMy5jGO3XvNPPmaCVwpCvNgJQqpagq0ZtamzxANlFLhECkXqc2p4+Q6W1aGNFoNBpNV9HCyE6ACTSEo7y14HNq4xKXy9dqH8MwulwtpD1RJNVeVuFoimrKqayKMGr0CGxHUry5jNEjhxEO1vPtE0/h9B9ciBBGOvxj9JiJjBwxgpVrNzNzZgaOA2+89iInfueMZiEiqeWlK4qxbQvTdPHKO58k9ZCB7xnSklNOOYX66jIO2H8vrGiILZvLGFKUi11qMWbMWMLhMKYUnHzyybz4zHPU1dUh8BCJhnj00UcZN34UF15wKW63m7LyzTQ0BMnKytwhm0y3g3vIqFbrw6E6Ahk5COGwcfXXTB9+EjVbF3PU0d+mdMsmHAFUFfPgGyt4+e3P+Ojd13n72TuYOmtf3n7qHqbufewO2aXRDBbmr1vH+JwcJhYW9kh7P/yhYL/9YMwYGDFCCSMejwpb+eorH3vsEeu090hH2LbktdcELhcc2+TP9YQT4NprVZneVDLWpmJIqoRvba1y2Kiu7qB0Sw+yX3ExCzOzICMAkRiUl6k4n8WLYcEHzap4IAfXgFjcsp3fMwmcCQxB5QJuANYBZShvET8qHc5ImP2H2ch5A+P8Z+XOgtrWokXTxKqyuRtIY6UZmiyLVNL21EGymYdIU7Glqddo0+S6Go1GszMjHDUNBAaKHd1FCyM7AaYwycktYMKILKYdcCyO42CarUUN27H4+Pn7+Off/8nWUJB5H6zGZQrOOvlkPl60gFUbNhPwN88g37Jsb1MuvuACbrj+Nwz1Wdx9zyM88/RD/PGmP0NWJo889WYzj5NUGVchBEXDxvDJwq9JREMEwzH+8McbOOnkM9vtyzSTl+kAzx/SHs8//wwvvPACe+41k9v/ej+ZmT7qQzFm7TGNutowNTXV5Ofns6FkC3+9426CsRj5+flcc801/PT8C3nwgUepqSmnsLAQv8/H9b/5Nd856TTefe/NHTMs7iD9XnWjmXZphkBGDgDh0q+xhGTRW0/wu1v/wZOPPsKkCQKfx8XGlV/yyH+fZ8kXn2OaJs/+bz4/Pec8Pv70My7Z0TdMoxnArKquZrePP1buFqmwjtJSWL4cee+93W535kzB0UdDXp4SRPLz1bLLpYQRIeCjj7wcfviOD/jq6jaz114wdGgAKScTjxdTUlJHTY1Dfb3yEAkGm0/19c1FkU2bIDc3d4dt6QyfX/ELxM03w9AiVYmmpBi++ALefodQKDTgc0t1yF7b30XMM5A/daAKWAMsQAkCuShhJAeVKHZ0r1nZaRzpMHn9ZN4qeCsdJiPTJXbbEUWaoH6GmogpTUQQQfJ3SjTZ1qQt0eR/rB4/NU0nsOzzcJl7AgmqqsqRxJFOAkggZRwpEwwdmoEkwaqV/8duu03tZ4s1Go2mka65EGgGLFKYNFTX8/b/nsdp51O1gQf+8S8qqitxHDDdAsMwyKacPYfn4fO1fhLZ0Q3nkKJRxC0bf2Y2hmnwvbk/IRiJs7m0DCklhmE2JkQTAsexkVJy3DFzMA2Dv/7tEULBEPPfeH+7fQ1WKioqOPmUk5DE+fzzT/nxeWdTH4ph2zZCCLL9bs7+/pkqtlrCoXOOZvPmYq655hrWrFnJlVdeSTBUQzyRoKamBl/AR0VlGWVlZa1itruKGRiKVVcONM8lAxIp6/HmjWHV0s9ZtWIZw3O9VNfWsrlkI3FMykq3MHK0ElBs22bN0q+oqa/hl3e9tKNvmUYzYLngnXfYbdUqmDIFxo9Xrh2jRsHo0TB1KuLc7iUerqqqIjNT5fxwu8HrhZwcH5mZk/H5MvB61bqsLIhGa3f4PD7/fDRLl4IQ9wDTqa2N09Ag014hTaeUGFJbqxKzVlaqEr4ff7xwh+3oCvI3v4HHHoMXXoDHHsd8732klAQCgcadrO2X6h1wjNr+JBc6ykPkK+B14D+oajclwGagAiUEZKnqcP3J0vKlTBs9jWGBYciUZCGblN+lqfdH49S8XG+TfR2VaBWncdmxHGRC4sQc7LCNE3awgzZWg41VZ2HVWlibB+G1sBPg2BuRfMry5VdRULCOwoINDBlSzJDCzRQVbWXo0G3Y9moEa/B62wvk1Wg0mv5Be4zsLAgozPTTULoNx7aRptFKaHBh8P0fnMFf77ib+YvXYArlzTF5j/0IBT/Fkl48XdQmPvhgIUcecwSgRJa4beGlUWBpboN62jN02BjAQAqBZTtk5hR075wHOlKSmeXFsixM06Sqqorzzz+fBx94FNM08aCSzSUSCdatW8dx3/oO8ZjN4YcdxbCho3lr/jyCDUHOOv0U3KNHU1q6GduykCQ47qjZSOkgRPvhTp3h4/deZvbJlyOEO2lyiERDGeG6GrJGTuRHv7iTP/7fGYwo9PCv+/7CJZf/mpKqEAnLIU9W4/ObTMzJ4utVNfgCWcQiO1YRQqMZyPxLCBg+HHJylCtHPK6SdESjys1jyBDEoYciP/ywS+0WFBTwySftC50eDxQVqakn2bTpXMaMmUB5eZi6OkldXaNnSGqqq1NTylNkyxaoqhLsu+++PWtMJ5DvvNvh9srx4yks2YxY/k3rjYaA666D+W8ja6p7ycJukJucd/TbeyiwCHgfeAZiG2J4T/KqfCMuVMUbC7Chtq6Wwh4K7eoOeb48EokE/oAfq8oCKymEOChxw5aqwkxK7Ei9tmR6GUu9Jg4y7uDEJU7UwQ5b2CEHERH4HB/5nnyG5wxnVP4oxg8fz+ihoxkxZASjho1i31P6/vrUAFgIrKSDb/KilDYIG7CR2AihLobc3Lx+tVSj0WhaooWRnQKHbcXrWbFxM2PHjGWPyUV8s74MQfNEfQlibP1mEbbXh2iSbyQjYJCwGujWENuxKSnemH4C5HO5sOxEs9CMpuJINBLk1j/dgiMl8UgQ02g/wetg5tTTTiTaUMujjz6B4fVg2zZbt24lGo2qhLnxONFYkEQiQTAY5Pzzz8ft9hKzoyD8GKbNT37yEx555BHiCYETjZKXl09dbR3Z2VnU1FQjxI47fM055UoSsQoMbz6G3UCkfCsVW9fj8nhxZ+RQW13N/932NFUbviRn3N78+9afce3d/+OJu36OnbDJxcMwd4DhI4cSqqlg/Iz9e+Dd02gGHuLee+Hgg1UITWYmGW43juMQiUaVcuF2q3l2dochiAOJNWsgGNxIZaWktrZRAGkqitTXq0o1VVUq3+m2balEmQOPguxsaKiHrOzWG1esgGef7XujtkdnHpqPB54Enga7xlb5wj5AJV6VQBiVbKwa3O7+LVOb5cnCTiiPyE/3+5SAL0BWZhZeX+cSB2sGOzaqlnRyWVrJpAMWEgeBjcRBYpOREeigHY1Go+l7dCjNToFkyIgxjJsxg3+8+AY3//nuNvPPue0Y73yymIlTpjUmOJOS9QvfJ54wEDLe5Z4zfV6WLVtLyvlVCBOP2XhjlhJI1OTg8Qb46quV5GX7eOGZh3AbrkExgOgKpaWl3HXXXVSU13D5ZZdTU1PDaaedRiwWQ8Zi/PzSc8n0msSiNo7jEIxHKS8vx7JiuN3gdwkmTJyI3wXBmgoO32MSATd4PG5+fM5czv7JpTz02H977H1z4mGEjNBQWcbmTaupqSnDdiw2r1lGMBwhWFdGxvApNNRXsu9Bh3LDz47lmj89x7j4SCbl5+D3+Tlx/BRWLf+UVatX9YhNGs2AIztbJfoQggy3m1uBywwDj8ulco1YlppLSVFPu3b0ElLC1q0O5eWkp8pKVW2mokK9Li1Vgsj69SlRZGAntZQHHgh1tc1XFm+Cn/4UgAkTJvS9UR0hOzFlA4+ArJPpJOoyLJUXyTfJaQWwGnJycvr8FJqS5c2ivqYegDGjxlBYWKhFkV0K5fKjbk8SICyUKGIjUh4kOICNx7PzPRTTaPoDIQfWNJjRHiM7BSam6fDIC29junyc9r0ftblX2HazYEUpp+55dHqdEILf3f8C02fNZNSwQrZWNABgWXFcru1XPxg5YgwVK5YipUAI8LggHncay/A1y11hICVYToI/3nwX11x9JW7vzqjNSWoqKlm1vpjRE8dz2mmnsnbNRuxonFC4gauvupoFCxZQUVGBx+PBMAQjRozi+utvACAvLxuX6cftdvPcQ/8gHo/zvUP3Z1F5LSd+++Qet3bD+hJGDYsTqq+hIVRLKNiAL6OBsvItFIzJRggDGY9QsvYbcovGMW7MFC7+foDjxs1lyHtD+HrdckqLN7Hvt49hS0lxj9un0QwIYjGVlTQUQuTkYLvdRKFVUo6hzzyDFIKiBvVdKoUqPyqTQmZVRgYA7733N+bMubxbprz6quCEE7p/97FpE4wdq5aLi5t7hzSdV1crgWTzZpg5cw9qapZ0u8++RB58MKddczUvFBbCho1cYprcW1XV32a1TbQT+9gg460/b7lFIjwCJqI8Rpb3tHFdx2W4+GjER/1thqbfSHqJJJeVIGIjkmKITMZUqdcajUYzsNDCyE6CYbjx+l3NytW1xGvFOfHoOdSXNX+qnz1kDHtMGEJmZmNccmdEEYDZhx/A8rWrkw9SBW+9Po/Djjgq7c2QSMRxuxvbqizfwrePOxonVgrEWLR45/MwCEfq+fyLxUQiEcaNG8eoUaPYY489QUpyc/KpqKzg+ede5qijZxONWMRjDtIx+N3vfsvf/nYXQgg8Hg/77LMPLz6bz+jRo4mZbm687lqGj+x82QEpJU7y3kMIMIy2r42J0w+gYv0X1AfrCYbCBOuqycgqIJGwqNlWwshhozCkxdDhY9iw7msy8gqxN65lWcNbZJVtY7I/kw82b+JgXxZuSz8Z1OykpFwpXC6CkQi/CgSwQyESpaVQVpZ2tyibPh02b+bPf/5zqyby8vIgmaC1vv5nQNeFkfffF4wfv2Oncv75krfeEoTDsHJlYxWapslWq6uVt0g0SvNyuIOE5//8l/42oXO0rHjclt7VQVXktgST/qa/w3k0/YmjvEQESKy0IEIyfAYchJDJdXuj8pA4yn1NDABlT6PR7NJoYWQnoiNRBMDMyOXvT86DtAeHwhAuXvt4Y7f6nDp9N4YNb3QbHzl2CqbhTpfqTYkilhXHcSS5+YWc+b0TOXD2d7jrvnvJyszqVr8DmaVfL+Wrr77CNE0ikQivvvImgQw3D/37IRZ++gnBhiix6LZ0AQWfz8fIkSM579yz+en5P+VfD/ybffbZByEEhxx6KO+9+hwVZRW8+MxzXPKLKzvsW0oH2wbbtpDSJh5pwO3LxDTdeDxt36y63W7C0TCRSD2W4xB3JFKAFamndOsWCoaNJssfwIpF8XoyKavfhO3OYHNtkMwjR7Pkzfc5+SfHsuLr5eTleZAyxqKPP2C/Q45usz+NZlCyYQMEAspzJDOTqBBKNaiuVrEmJSVqqq3lZ5dfztVXXdVhcyNHdt2Ed94RjB6tqtPsKEcf7TB9ukFenhqTRKNKHKmpUflEYHAKIoOOMtoWQ5qu29BHtmg0O0wqwSrJ0Bkn7SVC0nNEpsWSpEAinWQeEo1G0z1ScZcDgYFiR/fQwsguhCEEtFHFpOmNbyr0pbPJA4PBMOFwpNm+LhM+/3QBBx58RLodx3Fwu70IITj4iFPVeieO+gMSzfoe7Lzzzvtc84ufc/H55/Lya2+wefPm9ABjv4MP573XXsQf8PCf/zzCd049HcMlOfvM07nvnnsZWlDAnMMbBYWhIybg8/nwZGRRXlHJ0hXLOGr4qDb7tSybSO0WNi94guryLZSUFBOOxMjMysQWbs68/vF2bXb5MrGqqzDdbsLhCMFYmNqGWuqrqli58gtm7rYH69YvJ1xVTlVVOT5/JrFYA3W1tUyfMoYt69cyZMJMvvliAeu+eI0hGV5e/u+f+c4Pr+nx91ej6Q3yb7+dmsxMlUBVCJUzJBaDUEi5UCxcCKaplgOBxmo0NTXKY2TTJpWIo6GBu+66a7v95eZ2zb4nnhjJnnuqgjiBHshZKIRgxYrGG5hIJILfr8tn9jkbaVGfto3lJX1ljEazoziqKg3QGDqT9BRJCyRNprQoooURjUbT/2hhRNMMJUy0FijaEy3efOdtqqprmu0XlyaX/uwKFi/6Kp1rxOPxNetDCMFbb69po+/Bzztvv8u8J5/mn089ypuvzGP48OHpczNNEyseZdPGVYybdSCWIyERYljRKP718H+ZN29e88akzbjhIwk1xDjp1B8yZOiQVv05jiRhWWx5627O+tX9VFRUc+zhB5HhszlyWh7bKqrxbMe1eUtZDX6Xi4qyYkzTZPnizykaMoSG8Gbiq79hyoTdqKmuxI5GqK2txW2YSMsmasVxbEEEk3deeYr83Cz+c99fGFI4glA0roURzYBm1n//y9K8PMjMhEMOUWKHlOA4KpFqIqHEkUgEpk2De+9Vrh6ZmY3CSF2dCqMpKVHzTtJVcWPkyK1kZanjfL6e/67Uokg/sYI2xBD1O5xe/3ZfG6XRdBdVllcIGkNn2hJE0qKI8hyR0mYnuQXUaDSDGC2MaFohJa1+oNoTLVauXMmhhx7WrFKB5TgcccSRSNmYQX9XYsEHC1jyxcd89O77nHHGGYwcN6nZ9oZgCCsB8996DbfbTdHQIuYcq7xELr744mb7zpw1g9LVi9ha28CUGbtTGdzWbLttS2w7Qfm7/8Dn9lNeUcP0cePZLW84DdEIn6xPMG3UCOrL1rVrr5SSQw4/llWL3sJJ2ISCQaKJGDX19UhpsHlDCY5tEA1HCQUbcLlcRMIRTJcLGXdYvm4NtQ0hXBkFNNSH2EI58YTg1oc/6KF3VKPpWaSUGO+9BzNnKqXB7U5XnMFx1GRZaorHlQASjcK118KFFyq3DdNU2xoalCBiWem2O4PP1/mKDPX1m0k5s7hc4HLpHA4DASkl1UY1+eRjYxEiQimlDGM4BlBJFd+wnP/wCE/LZ9LHiaNEc8/nVnPZbN3Ju/d80m2NpjeQ0kGivgtToTTNp1T4jEx7iqgQYBuXHpFoNN1DR9L0GPprSNMKIQSWZeFyudI3+W0JI47j8I97/s70Wfult9fVlHPAjMncfNOtOI5DefkWhg8f06f29zdlZWU8+8LLXP+b3/H763/LhZdd0Wz7hOl78c2H8/GY6s+vI/HI4/Fwz4MPc/EPf8SI8SMo+7I0vc22JYlEgq1v/ZVAdh4+r599Zozjky9Xc9See/PdM8/kgzfm8Z9587nke4e020c6Ua6ZTVb+cCoqygiHIzRU1xEJx6lrqKW4eC3Eo4SCQeLxOC7TzUefLmXeRyp5riMlH773FocfcWx337Y0TQeWzasaaTQ7TnkkwtBVq2DCBPD7ldLQ9G8w5S2SKsGbSKj9olE1f+QRla30O99p1u5ZZ53FY4891mk7srIyO73vM88cyX77pQRrAeR2+lhNz/HhHR9i+ly4vAZUCab+cir55GNhESJMLbXUU0cuOXjwYGKQQQaTmdK8oXe61u+L8sUu2/r66ypJ77p1MGdOLYFA/5bx1QweIpF6LvhpDgiBIQwMw4XpcmGaLoQwcJke3G4fLpcb0+XFNFy4XB7cngDXXKPK8qrvqlRpXklKGJE4iKQgcvjhGzAMME2JEDB/fj+etEaj0aCFEU07uJLSfUcDUsMwmLnXgenXtm3z0Ydvc9sdf8HjUdVJdjVRBCBeU8qFF17IpuJ1HLz/3q22Fw0byislpRSNHc+wYcNwd/D0V0rJqSefztb6agBqamrTwoHjWCTqt+IJZJJRMBqrqpizTjiEr1ZvZlVFGceE6vn2d0/H43WYc8bF7faRYve9DmDTmk8xTReR+gZCdQ1Uhxqoqqzj9ttu5qKLLyK8YR2JWIJ4Is7wsRPSxxpC9Igo4jgOX331NVMmjsWXkZ1er4URTU8gpWTounUwbJjyFHG5wDTT4psARDKURloWMpFQk9utvErc7pTLBrz8sspSumULfPVVl0QRAJcrb7v7RCI13H//dITYltZolGOKrv7UH4y6ahQy+TislFKyyCJBghAhaqmljDJqqSOffDLIwEHixkMBBRRvLGbMOPV72Fmvoh1h3DgYNUppeS+/nMv3v68T6Wo65pNPDmbLlq3YtsO3jhtBIuEQCUsc6WDbDomEQzTiIGUY2wlhJRxicYllOUSiEtuSSDkOsFm1ajxz536NIUAYYBgSw1Aa9McfTwAcFiyw+/mMNRqNpjlaGNF0mpZ5Rlq+Nk2Tb590Zn+YNqCI1ddTmJ+L48mjqKio1fac7BxqIkHG5hRi2XH23ru1eJJCCEFObg4e1zB1bE52OjkugCEtfNlD1GDOMDl8nxlce9Gp5PrHULK5hI0LP2La1D2I1Gwje/TM7dp+wtyfc++t15A/UuIJ1NBQXEJBrsG0/b/F9867lhsvO4FEPEI8Hsdv1fPSC89w8qmnd/OdUqSFHiQ/Of4Q5i1aRWFmgBfeeJspkyfpm3lNj2F88gmMHw8ZGeDxIFwuDNPEC/iBAJDKhhQHIkAkkSAaCmFHIjipa1HK5rlIYjHEkUci3+miK0ALotE67rlnAolEdatt5eWQna2ervp8tdrtvB8opCDtJRwkRJw4IULUUUcZZWxkIwkSFFGIiYmNhYFBAD+nHXAai8oW9ZmthgFeLwwZAlOmwA03GFx//SD3cdb0KgceFEfQvFyWbJUAp+lEMn9Iy/U2J54Y4MQTd1MtyEYvEQAp7WSode+ej0azy6BDaXoMfWulaUV7iVaFECQS8XQJXj1gbZurb76Dm395MSE8lJRWtNru8XjJLchnypTJOE6CSy65pMP2TL9XDb6AvLwCQL33pmniLRhHrOQrbDtB5sgplC6ax7fn7MejL37IKG8m+x19HAXj8rDDNR110YxLf/VnNpds4pZrz+M3t/6HSZN3S2/71llX8587fkGoIYjbsCld+hp3L5/Pz353f6fbbwspJd//1mEsXbmJkVkByhviXPCD03l/4RJ9nWl6jvz8xkQdpolpmmQARcAEYCIwAvAANUAxsMHtZktuLlVuN1EpsW0baVng86kcI5mZUFAAU6d2yZS6ulqysyW33ZaFlKHt7v/ll2pSOaAa+PWvu3ryGoDQshDuWd279fEIT+MLCUGC1FFHOeVsZCOLWEQBBYxgBB68CAwkDi7ceMr71ssnFlM/Gx4PFBXB9OnwxhtP861vze1TOzSDCYEk0cF2iWiWEKfplKoq0yiUSByElAjRGErjOA5SShzH0cKIRqMZcGhhRNOKjgaiKVFE0z4PPfgQzzx0D9XRGOef/9NW24UQZOfkMHXqFGJRh3333bfD9rKysshMVvUpKChIr0/lJsmYeRJbFr9Kcckynn/tCy776XS+d+y+XPuXR8mbPpWc/CiVxavJmXVCp89h1Oix3Ptw64Dfgw47mr/9vh5p25g+H199+Qk5Q8d2ut22+Mm3Dqe4uo5wLIxtO5QGw7iEi8MPO0yLIpoeY01trUqYmvT0EFJioDxFhgIzgMOAvYBCoBpYCnwKfA6sycigPB4nEovhmCbSNJXA4vGoeIUu1t8tK2tg8WKD88+H1avh/fc7d5wQkJcHDz0k2LwZli6FSy9VaU8aGqC+HqqrVZRPaamqIrx8ORQWwpVXqurD9fVQUaH6ffPNQf54p4tU/LWC4XLYDrdjY1NLLeVUsImNLGYx9xr38l3nu4xjHL7kPytZujSTjB03vgtIqa6J7GwVNTZsGLzwwve1MKJpF0EGUJ9+1fajX6eN5ebzZoKIUEKJlI2CSGqu0Wg0Aw0tjGg0PczESZM484IriEWjTJm6W6vtwhCMGDaarKwsgsHgdiv3HLD/oXz4jho1tSypaRgGbreLMfueiGVZHGpn4ckbQTy6mpuuvZTS8hqqamoRmcN77Pwef2cTl8/dh0QixtjxU9hWWUk4FCKQ0b0b/++f9X02b1jBCy+9RrkjyM3IYERRNr/5w81dbiuRSODeTmliza7Jg998o0aJ8bgSSFC38TG3m3qgDNiMEkkEKpTGjRJOMlEhNm7DIIZKNkzTPBGGoQSSLlBba3PkkZKf/Uxw8slw3nlKyHj11dZVwaSEESNUfpFNm2DRIqithXffhVdeUWE2LYWRigrV3pYtSiS56y4liqRMNwzIylK5oUyz8xVyBjvOgw4W2x+UbU+SjROjjBDFFPMFX/AP8Q8idoQP3v+Qj+Z8SAYZZJNNjBgJLOxO9NmTDB2qrgevV13u2dkwdsc0bM1OTwFQ28b6ln76TV83iiFIFVojBK0EESWGNH2thRGNpqcQbP83q68YKHZ0Fy2MaDS9wNix49rdJh2HESNG4DgOiURHbquK3NxcRk4YhyPVIKYlhmEgXGo++9tn4EiI1GwjLt0MK8zFl51HwgjsyOm04q4nF3LB3P1YvaGYiTP37bYoAjD/v/ex234HcdJuBawtXU844aOyPkrcMeiq1VoU0bTHb/bem1u/+EIJGFIiLQs7Hifo9VLs9xMVglJgIWp4YKKenZYCW4C6RIJ4KIQTi6ksqPF4YzZU21b5RrpAXZ2a3323GmD87GcuvvMdm3PPhWAQnnlGJc8MBmHjRiWGlJfDwoVw991K2DjjDCguVmET4bDat65OCSNlZVBSoqqSTJrUWGSn6SQljBmTyZYtkZ56mwc8BgYJYrR/+9bWetFqbZQYmylhCUu4j/sIOkEADpt9KHfyV7LJZhjDcHAIEaKKyh48i+1TW6uui0CgMd9Ifj4cemghH37Yt7ZoBgvDgVUdbG/+HddaDGkMq0mJIU0nLYxoNJqBjhZGNJo+xnEkOTk5LF26FE8nnzL7hZcVy78mJzePUaNGt9ouDENlfsdESkne9COoWvU5Lo8Ll5Ogomw9w2Yd0WPnYBgG/352cbePHzukkO/sPoxhuQEm5mdQvvgT9tpvElO/KePNFeVkZw5jxUdvcuDxp+pwGk2PkOXzwbZt6vG5bYPXi3S7sSIR6oNBIh4P27xelnm9eIXAJOlREo0SDYWIhUIkQiGcaBQZiajyvakpElFTF6iqav767rstAC6+2OTUUx0qKuDzz5XHx7ZtoykuLmb//QVz58I336joHcNQekw8rrpvaFDCSNMwGoA//EF5i6T0nFiscR6NRnf8zR0klD1SjoVNjHi7+4g2XolmW9Q8SoQlLOGf3E+9rG921HPyOS4QFzKDBB48VFCBbx8ffcnGjUoMycxsvFYyMmDEiKrtHqvZNZGMAaz2NiavfJkUQJKvhZPeISWApJaVECLTQkhLoUSj0fQQLb1Y+5OBYkc30cKIRtPHxONxcLsYmp/H9Tf8plPHrCpeysknnc7ixYsYOXJUh2KBEAIzZxgFE/fEqVhJ3DHJ8Q6sm2HTANu2ePDjjVx+4EhyR48jEYmxd1E2nxfXU1JWy29+dx2P73EwI0b1XBiQZhdn2TIljMTj6lG614t0uZCGQcwwiBsGwbbK9loWMh5HxuPq2JQYEgw2qhHVrSvJdERZWdvr//EP5RX2s5/txsqVK5tt+/xzyfjxgjFjVGhEykEqkVAmNTQoT4HycqhJ5lt+5hkliigRpHEKh9VU33xMv1OTcY6fMrbhQoUOiTaFjxRtJCBvshwlyr/4F7Wyrs2+zv/sPO484E6GMIS1rOXTRZ/usP1dYdkyFTqTlaXEM8tS10uTNFUaTTME42npFdK4sa2yFymxA5oKI6l1jWJIaplm4olGo9EMNLQwotH0MVI6+EwXth3n0EMP2+7+QggmTpiCEIK8vDwcR7C9lACGMDAKxhLZthzT48eXlddD1vcMjh0jL8fLARPzGT19EiN224/5Tz3FWxsbsCybRDyB1+NhyNDC/jZVsxMhr7sO8ec/qxiV3Fz1KN3tVklUDQMpBFKI5uV4U6NKy1IKREphSCX1qK1VCT22bu2SLSUlHW9vKYqk2LBBVQ0rKFDeACkHmGhUiRwpx5W5c+Gss5R209K5JRxuTMI6duzELtk9mAkToZhNTGAibYXHtF4nmq1vKqCECVMj26/2tf/++/OY8xhDhw6lrD0VrAkVFaJTTkdtaeI+n7qMq6pUaNWyZfDRR8pjJCdHHZN6iOfrW8cVzaBiNNAyXLdRwEh5eTQKIal5e+JIR+u1MKLRaAYeWhjRaPqYhG0TDoeJ4+Byde5P0O1S2TZcbhfqKY3YboiJTJZXNuNluHy5O2h1z3LlkbtRWhli39HZHPiTP2J63Iz47EMOCIbZ0hCjOlzFF6s3Mn/e/zj+5NP621zNTsRHxx/PIc8+q8p0ZGUpzxG3WykMptk48kwJI7bdmKAjFX+S8hapr1ej0ZISaEfIaI81a7p/DlJKXnpJ8M47Kt9IU958U51OKKRMTEX5NBVEUo4u1dWwevWK7hsyiAiXhqmgnHmBefw4fE4bXiIt14hmskjLb9sw2y+xLISgvLx8u/tZVh2dTY/U1td+SYnyINq8WV2Gn34KL72kKkjn5qpjXC6l6+kSqZr2kHIYUrbOY9ZUxEgJHan12xNHGvdrLZZoNBrNQEMLIxpNH7Np4yayc0ZgyWoE279LlVKSl58DgGM7wPZFEQCBRDoS05+NYccpefcRSmuDrPnmaxJGgLv/+TIFfh/jRo3iz989k+DmrWwI1oA/g6mjxxJavpz/Jtax55zvMG7KNCZPmYYvK6dL5yqlpL6hluOOmMPKVevYc0wOuw3JZlSGyejhOYj8ERimgYVJ5pAhuMVK9hmTRzASpqY+wnvvvquFEU2PcvDuuzOvro5vP/64Ekeys9VjdI9HCSOG0Tj6bOox0pa3SE2NSuaxejV7dLFc72efdf8cpHSYOROOP97grrtGIGWELVvqqKmxqK9XpqWEkNQ8lZw1JYrU1Cgnl10lYXHioAQb2MhNi27m8+mf09w7RDQJqxEt/ETUUuP/an09DT1mWyyWC3RftFi/HoqKVPnlRYvghRfAshzOOMMgL09d1n5/Y54ZjaZtstpM8J6ipSjSct5VkUSj0WgGGloY0Wj6mMzMTAqGZFNZuf1SvSkCyTK9GZkZSbdouR1xJJUZ3iaGiTuQjctbz6YN6/HmDqdsSwUHHbQ7+x5yGPsfdDAVbi+h0HjGPvE6CxYvZIbLy/XrP+LAmRPYvOQtcguHEIlEuiWMVJdXsbF4K5k+D9J2cCXibKtJMHpMEXOu+BMby7fxw59cytsvP8Hyjz/HGwwxPDdAQyTGgvfex7KsTnvWaDSd4fhDDsE5+GCMk05SYTXZ2Wrk6PGkw2oANRJo6S2SUhhqa1WikHXr2HPkSL788ssu2bBtW/cHBq+/bjJhAng8hwLjse15hMNOWgBJeYY0nRoaGsv5pqJ/TjvtjG7bMNjI2ZjDiZwIQDnlSQGkUQwBMJIBMy3/by2cCGpoP4ymq5SXq7CoLhY2SrNpk/pcv/hClXsOh9W19eWXKqdIqlzvrpZTRtNVXDjtXIQtvTzaEkZS846WU6/b60ej0Wj6Ez3a0Gj6EMdxMJ04HmGRmZmJZbWTAT6Jbdts27YNx7bIysomHKpDFBZ1wmNEIB0Hy7Iwo2Fqwgmqgw7L15fi99eydsMWho+byIyZ+3DXLTcxceo0ps/cm/rvHMG4U45iafEijvfui9flonBoPjXVVQRDQXKlgxBde6z5u+uvB8chO8NHKBKhPGyQiUPJljKw4/zhhj8SqQ+y/xEncepIm4qaIImohWmarCkpJxwOk52d3aU+NZrtIYRAvvIKF/3859z/6adQWKjKdni9zUNqmgojqRCalKfIli398uTT5VK5JCZMWI/bPZrKyhDBoGwlhjQNm2kpimzZAk888USf2z4QWM96jLTEYbT6vzP/ytjWI7Y4jsVrr8GJJzbmAenqJbV5s7oe5s+H0tLGAeeqVQ5HHGFgGOryjsVUYSaNpm28SDkNKauBUHKKJUvrWihhRCBl6v7DSC6L5DUrOhBFoKk3iaPL9Wo0PcdAymc8UOzoJloY0Wj6iOVffYHfZ1JTW8WkaTOpqlpOKBSE0e0nP9y4diVVdWFmzZwJgGF3/hsnWpUctAkXTjxKQ10de+4+jXXrN5Hh91JXXs7a1au56JrfMXTYSPyBDIRhYCcS3P3cf5kzcwrSscnPy2Tjxg0kEgdiWwlcbm+n+pdSYts2X36xBLfHRyyeIBBwsWprLTOGZuJ3u4klIuy1x0EsW76Wv93zd168/nxiCYv6YJB43AFChKIWWhfR9Bb33XUX9wGWZeEdOhRn6FCVd8TlUuJIqh5uquxLRQVeIQZEmduFCzez//7vU1ERob5e0tDQKISkQmrq69VUV6f0nIoKlZNiy5Yt/W1+vxE6JsTCtxZhYmDiwkz+c2FgYGIk5yYiOTeSS2puYLCa1T1iS1mZm3/9C2bPVhpcW3QklEiphJG1a+Gll5Y3E82FEKxZMwIhtjJ8uNL3li3rEbM1OyFC+PD55ve3GRqNRtNvaGFEo+kjTLfJwoULyfBnAPDC84+zz74HdnjMhKkzmOgABtTV1zD/rTc496KpnepPyAQul5todTHLln7Dso11rC4pZejwsZx14VWMmTCZzOwcfB4PpstEGOqG2k5YXHnDX5j3wF0Mz3bjFhbxhmqEkAixnXI4TZBSEg6HMR0Hl7TwuKGuIYRbSoTLy9bKMA31DZRXl5ObYZKf6aIuHKMiHEUkSwOaQnLRuefy4v9e6FReFY2mu7hcLuyq5mWtpZTU1dWRk5MzIK+/YBBWry6lulpSV9cogDT1EEl5iaREkc2bYfLk3RkxYkR/m99v/OnNP3VqPyfpdRePx4nH4wSDQTZs2MBXX3/Ffr79esSW/HyYMuX71NQ8RcqBsC0hpKN1K1dCJDKJ6dOnt9pny5YtBAIBhg6NkEjAhg3xHrFbo9FoNJqdDS2MaDR9xG7T92C36XsA6un0h+8vJBaGw444qd1jBJDKz5qTnce5F/2s0/25MwuJe/xEYxb1EZuqumoOOeYUTjxtLrn5BbhcbYfEmG4XWTm5nHjR//Hynb9l7MixBESYeDyOdDrvsSKlxHEcfG4Hd6YfQQykG8d2mDxtHIeceCojR4/F7UTweXx88MiduDwuKhsSxCVkud3E7ThffLEYy4rj7qSnikbTUwghyO1iUtW+RAioqrKprlaCSEocaSmK1NZCZaUSRWzbw9KlS/vb9EGBYRh4PB48Hg8A+fn5jBkzhtmzZ/dI+1JK1q6Fp556kueff4pYrOX2zi0vXy6w7fbLHIXD4R6xV6PRaDQDDyHVNBAYKHZ0Fy2MaDT9gMvl4tlX3uzVPoxAHpZtkzd8FEM2rcdj+hg2ejwenx/T7Pjpt2GaeFwuxo8biRvBhGG5lG4uYfTYibg8natiIaXkTzf/Eb8vgMtnEqyX5OQYNDSEOfzbJ1A4dDihTYu5+MxjMX9wIo/ffhOGlcBtSoJxSY7XTbYvQMKKE4nEcbk8A/KpvUbT16gEzGq5rEyJHm2FzzT1FNmyBUzTTTwe67BtTd8hhODTT2HGDPX5hEJtCyDtzVPLmzYV943BGo1Go9HsxGhhRKPZSRFAYOyeRDZ9yYzpMwiHw3z0zlvM3HMvMjMDHYoMQggMw8Xi4iBjhg4nOzuX5V++y4w998Pn82G4Og6pkVJi2RZnfP97ZHoFSz7+CI9w8JBgr93GMmXvI2hoqMafk0umLx+ZaGD/k05gyR334fOaWFGLUDSB6U7gxGz+/thL/N+FZ2lhRKMBpk5dxcqVKqRu/frmZXhTwkhdHVRXK1GkqgoqKiooLCzsZ8s1LTnvPKVyrF+vBC4pmwsgnVkeOXJkH1ut0Wg0Gs3OhxZGNJqdGG/+WELb1uA3BPneBHn2ZpYs/Izso44mKyuzw2MNA06b+yMqPnwET5afA6aN5KN3XufoE07FH/BjGI15SVoipaS+rpacTB9TJ47kk3ejjB8zhsptmzjq+KORLg+JhATDC8LG9GYza6/9yM56ALNSMLbAz5baKAIDx3R48O6/8n8XnoXjOJ0ucazR7KyMHj2FN95QyTRXrVK5YcNh0slXU54iZWWpQfQg923dBQgGD+C99z5TFTucRvGj6ev25low1mg0ml0YXZWmx9DCiEazEyME5E07korPnyG3YCh7JQxefPlZps/cg0Ago8OQGsMwyB0xEmfi7tSXbyARD5MdLOPj97I56PAjMd0mgYyMNo+VUpKTk0dmViYHuV1s3VRMfV0FLiymH/wdbMsip3AopnBjxeOYho3l2ORk+HG7PARjFv7/b+/e46sq7z2Pf5+19jUJIUCEkCMBFHRUbgIWgkeLtSKUoedYi9oLg1NRq1Iv4PFowYIXtNbL2GIVPXWgjtNR287Yei1U8VbUKgeUShVRNFEIEG4JgSR7r/XMHzEpMSEkYV/Zn/fr9UDW2uvy23mtrKz89u95noCjXbV7JQVUX3nw/vNALpo1y2rQoAGKRD6T60qxWFNXjD17mv4vKCiQ79emO0x00n/8xxvpDgEAgJxGYgQ4wjky6j362wpHo2pseE1fGVStv617R/1KShTNixz000Yn4Mp1Ha3Z8Kn6F4X1ycaPtad2h/J2NqjymCEaMHBwh+c1xsj6Rvk9emni6f+s119/TaedNkHRnn0Ub6yXa1wFgnkyjXtlAlEVFA/SuRf9N731k8Wqje2TrKdoyFWjJ7mOq9q6fSrIiybjWwRkpU8+qUx3CAAAAEcEatKBLOP7ftd2cIxkPcXjVm6PIg36pxJ9sPKPevR/PqT9+xs6LLMPhYI6ffoP9dtn39Bntb5GnnKq8nsWqXbPtk7FaYyRQgUq6luqE4cN10mnnCUjV4FID5lgRJ48BQJROY5RQ0Othg4v17FHFys/kq9wOKxYzFPIcVQQzVPt/vquvW8AAAAA6AQqRoAs050xNhzX1d6anQqG8lTUJ6xTRwf1WsUuNTQ0KBgMKBhs/1bguq4KCwq0cPEy7a2r0Rsvr9SQCado0DEnKRA4+Ow0vu/Ldd0v/neUVzxAJ/Q+WsZxJK9RjgnIBgIyfly+sZJxVNCzv/x4vYYNO17bat/Tvvr9ioZDclxHMtL6DZtUMr6oy+8dAAAAADpCYgTIAVaSH8hT3Naofk+13ny3QvsbHe2s3q78g4wT0swNuMovyFN+QZ7+5bzvyPr2oIOuSv8Y6DEYDMpxHNXU1Mv36uWYPBmvUca4ssGgjG++GKTJynd8GdeVGyzUqHET9GHlHlVt36Vg2FU8tl+9ekQUDh+82w8AAACQa4wkkyGDnmb7UzpdaYAcEI/F5Cmgl1Zv1IP/71WFIgH9n+df1/59+2Stle1C95yOkiJS09gigcA/cq5eY73cUE/J+E13TDfQlDzx4rKOkXEDcqwrxw3JGKPRXzlNw0eeqBEnnaSCUEDFhYWKhMNyrE9iBAAAAEDCkRgBckAwGNBWP6zPdtSptGyw3v5or66+6koNPmaIQqFAUxeXBGpOYFhr5Vu1JDWMG5WcoFwnKAVDknHlO0bWSJ5xZa1VzPdUPmGCjh5wtAKOr0heWL17Fimal0diBAAAAEDC0ZUGyAFuIKDxp07WyWO/pgcWL9SdN/6b8vMLFAyFkppsaKjfp0AwLN+3Mn6jbMiVo6bqlKaxQxx5nidrHLnGl3UDKiz+J+Xv3K6+pX01dtQIbd2zT0XF/XTcwFISIwAAAEAza5taJsiUOLqJxAiQIwKBgAKBgK69/mcpO2dDQ0yO68o6kvUD8vy45HsyoaCMb2WNZBWXa4ysjGy8UdY4CjhB7dpaJRMKqaS0SMeMmqCA66YsbgAAAAC5g8QIgKTx/biCoZB831fMCcj14jK+L3kx+TIygYCMXMmVrO/LenEpEFZ+r34qOfpojT12qNb87QP16l8qOdmdhQYAAACQmRhjBEBSWGsVj8cUa6yX7/uS9WTly5cv68VkjZGslXGsPM+X40syjuprq+XFGnTscUM1fMQYHXPSWL38p+WSpRsNAAAAgMQjMQIg4ay1isVi2la9XRUffaB44z7Jl6wv+Z4nK8kxVta38j0rY315sb3yYzG5xsh1Axr6X4YpWpCvSCigzyo/UWPMa0qwAAAAAEACkRgBkFC+76u+vl719Q36pOJzvfTC8/IaGmQCjhSMyAmGm7aLx2Ss33QTijfKi3lyXEcmEFEo2kM9+5bKMY7ceEybPv5E+xsbFIt5TVP9AgAAAECCMMYIgIRyHEfRaFTRqPTNqf9VmvpfD+t43zlmqL7z/e8kKDoAAADgCGG/aJkgU+LoJipGAAAAAABAziIxAgAAAAAAchZdaQAAAAAAyDLmi5YJMiWO7qJiBAAAAAAA5CwSIwAAAAAAIGfRlQYAAAAAgGzjf9EyQabE0U1UjAAAAAAAgJxFxQgAAAAAAFnJpjuAIwIVIwAAAAAAIGeRGAEAAAAAADkrqYmRRYsWacKECcrLy1NRUVG72xhj2rQlS5YkMywAAAAAALKbzbCWxZI6xkhjY6OmT5+u8vJyPfzwwwfdbunSpZo8eXLLcs+ePZMZFgAAAAAAgKQkJ0ZuuukmSdKyZcs63K6oqEglJSXJDAUAAAAAAKCNjBhjZPbs2SouLtYpp5yiJUuWyPezfBJkAAAAAACSKt19Z46cvjRpn673lltu0ZlnnqloNKoXXnhBc+fOVXV1tebPn9/u9g0NDWpoaGhZrqmpSVWoAAAAAADgCNPlipGFCxe2O2Dqge3tt9/u9PHmz5+v8vJyjRo1SnPnztXNN9+sO++886Db33777erZs2dLGzBgQFffAgAAAAAAgKRuVIzMnj1bF1xwQYfbDBo0qLvxaPz48aqpqdHWrVvVr1+/Nq/fcMMNmjNnTstyTU0NyREAAAAAQE4xtqllgkyJo7u6nBgpLi5WcXFxMmKRJK1Zs0aRSOSg0/uGw2GFw+GknR8AAAAAAOSOpI4xUlFRoZ07d6qiokKe52nt2rWSpCFDhqigoEBPPfWUqqqqVF5ermg0qpUrV2revHm65JJLSH4AAAAAAICkS2pi5Cc/+Yl+/etftyyffPLJkqSVK1dq4sSJCgaDuv/++zVnzhz5vq9jjjlGN998s6644opkhgUAAAAAQHbLpMlgMiWObkpqYmTZsmVatmzZQV+fPHmyJk+enMwQAAAAAAAADqrLs9IAAAAAAAAcKZJaMQIAAAAAABKPWWkSh4oRAAAAAACQs0iMAAAAAACAnEVXGgAAAAAAsg7T0iQKFSMAAAAAACBnUTECAAAAAEC2oWAkYagYAQAAAAAAOYvECAAAAAAAyFl0pQEAAAAAINvQlSZhqBgBAAAAAAA5i8QIAAAAAADIWXSlAQAAAAAg69CXJlGoGAEAAAAAADmLxAgAAAAAAMhZdKUBAAAAACDb0JMmYagYAQAAAAAAOYvECAAAAAAAyFl0pQEAAAAAIMsYa2VsZvRhyZQ4uouKEQAAAAAAkLNIjAAAAAAAgJR55ZVXNG3aNJWWlsoYoyeffDKt8ZAYAQAAAAAg29gMa11QV1enkSNH6r777uvGG088xhgBAAAAAAApM2XKFE2ZMiXdYbQgMQIAAAAAQNbpRqlG0mRKHN1DYgQAAAAAABy2mpqaVsvhcFjhcDhN0XQeY4wAAAAAAIDDNmDAAPXs2bOl3X777ekOqVOoGAEAAAAAINtkYE+ayspKFRYWtqzOhmoRicQIAAAAAABIgMLCwlaJkWxBYgQAAAAAAKTM3r17tXHjxpblTZs2ae3aterdu7fKyspSHg+JEQAAAAAAsk0GdqXprLfffltnnHFGy/KcOXMkSTNnztSyZcsSGFjnkBgBAAAAAAApM3HiRFmbKVkdZqUBAAAAAAA5jIoRAAAAAACyjbVNLRNkShzdRMUIAAAAAADIWSRGAAAAAABAzqIrDQAAAAAAWSeLp6XJMFSMAAAAAACAnEViBAAA4CDef/99XXbZZekOAwAAJBFdaQAAANphjGn5esmSJbJZPuI+AOAIQ0+ahKFiBAAAoBOi0Wi6QwAAAElAxQgAAMBBGGNaKkXq6+vTHA0AAF+S5ZUamYKKEQAAgCOMMaZVO/HEE9MdEgAAGYvECAAAQDuqq6vTHUK3HDg2SrO///3vMsbohBNOSENEAABkNrrSAAAAtKNPnz6SWnenyXS1tbUdvv7++++nKBIAQLIZa2Uy5PdTpsTRXVSMAAAAHERzQiRbkiPjx48/5Dbnn39+CiIBACB7kBgBAAA4iEWLFqU7hC5Zv379Ibd54oknUhAJAADZg8QIAADAQfz4xz9u+XrNmjVpjAQAACQLiREAAIAONHehGT16dJojSZyhQ4emOwQAADIGiREAAIAORCIRSU3jjJSVlbWZCrejdtxxx6U5+vZt3Lgx3SEAAJAxSIwAAAB04Lvf/a6stbLW6rPPPutSYmTjxo1yHKfdKXQTrTmB0xnNMRljlJeXpxEjRmj16tVJjA4AkHDWZlbLYkzXCwAA0AHf91sSGwfOUtOeg81ck4rKkYaGBpWUlKiqquqQ28ZiMUWjUTU0NLSsM8bI8zw5Dp+bAQByC7/5AAAAOvDEE0+0VIzU1tbKWivP81RfX6/6+nr5vt/SmhMjnue1SpJ88MEHKYl1y5YtHU7H+84778haK2OMfN9v8/qmTZuSGR4AABmJxAgAAEAH9u3bp8LCQklSQUGBpKbqilAopFAo1O4+O3bsaPk6Fd1oJKlv3756+umn9dhjj2n69Okt53ZdtyWR06tXL3meJ2OM4vF4q/1DodBBK14AABko3V1njqCuNCRGAAAADuGKK67o0vbr1q1LUiQH98knn2jatGkyxui3v/1ty3rP8+S6riKRiMrKyhQIBFqSNUOHDm1JhvTq1UvhcDjlcQMAkG4kRgAAAA5h8eLFXdr+xRdflCTl5+cnI5x2RaPRNusOVQFy4OCwW7du1YABA5IVHgAAGYvECAAAwEFs2LBBkrR3794u7dc8w0tpaWnCY+qItVbFxcUpPScAIE1shrUsRmIEAADgIKZNmyap6+OENHel+cMf/pDwmA5l+/btqq6u7vJ+Xx5zBACAXEFiBAAA4CA2bNigYDAoSbr00ks7tc/AgQP1+eefS5JOOOGEtAxo2qdPH1lrtX///g63GzduXMtsOq7rpig6AEBipLtE5MgpGSExAgAA0IFVq1ZJku6///5ObX/OOee0WVdbW5vQmDorGo22TDXcrHnZWqs33ngjZbPmAACQqUiMAAAAdOCxxx6TJDlO5x6bzjzzzDbrZsyYkdCYOovpdwEAODQSIwAAAO1o7oby85//vEv7DR06tM26p556KiExdUZ+fr6MMTrzzDP10ksvyfO8lJ0bAJBC6e45c+T0pCExAgAA0J7LL79cUtOgpF2pvDj++OPl+36rdV9eTpZNmzZp3759stbqhRde0AMPPKBAINCqu4wxpk37wQ9+kJL4AADIRCRGAADAEcFaK2OMTjrpJJ1zzjn66U9/qrffflsNDQ0tr3fF0qVLu11tka5xOwYPHqw9e/a0LF922WVttjnzzDNbjTNirdXSpUs7dfwvj1cCAMCRIJDuAAAAABLhzTfflCStX79e69ev15NPPqkbbrih1TZd/aO+eVyRlStXdnqf5gRNuhQWFkr6R3Kmurpaffr0aXndGKOSkhJVVVW1rIvFYp06dv9/6iPXDchxXAVcV6ecMlqPPvpbhYLhBL4DAECnWL+pZYJMiaObqBgBAABHhPPOO0933313u8mP7lY5vPXWW5Kk008/vdP7fDkpcscdd+hnP/tZt87fHTfddJOMMZo0aZKsta2SIlLT92Lr1q0qLy9vWRcIdO6zsqrNO/V55TZVfrpFH39UqTfffEvDhx+T0PgBAEg1KkYAAMARobKyUldddVWrdV//+tf16quvqrq6ulvHLCsr07x58zo9I017rrvuum7v2x0LFy6U7/sdVq14nifXdbVt2zb17du3W+cxxsjIke/zORsAILvxmwwAABwxXNdttfznP/9ZF1xwgR588EGVlZV1+Xj9+vXTzTff3KV9vlydksoxOebNm9ep7RzH0YgRI9SvX7/DOt9Xxo1X8EvfcwBAiqR7FhpmpQEAAMhsDz30kCRp2bJl+slPfqKrr766W8c53PFCUjneyKJFizo9xsm7777b5eN/Oclz6SWXKcj4IgCALEdXGgAAcMT5/ve/ryeffLJl2ff9lul3u6qriY10DrzaFd2pZLlh/jUKhwKKRqOae82NuuTSi9TYsC8J0QEAkDpUjAAAgKxXV1fXann8+PF69tlnW60Lhw+/suHVN/6q/3x3nSRp+cpX9bunnu/WcdLZ3eawGF91DTHV1tXpuh9fpe/PmK6Kim3pjgoAcpO1mdWyGBUjAAAg6911110aMWKEJOkvf/mLbrzxRklSNBpN6HnGjhgux216fPqn0n4ym6sOsUf7DqwqSff0vl3hOI4iAUfBQFBugau4H1dTx/LsiB8AgPaQGAEAAFlv4cKFWr58uSTpRz/6kXbt2iVJWrFiRULPY2UVDgclSccOLNPmqu7NdnMgY0zWJEfCwZD2NezXvtq9emPVKr38yn/KGAqQAQDZjd9kAAAga8ybN0+BQKBpqtgDmiSdddZZkqQ1a9ZIkoYOHapTTz1Vvu8nJZZIJKLt1TsS0g0mG5IikvTsM89ozMhxuvuuB/WXVesUDIbSHRIAAIctaYmRTz75RBdddJEGDx6saDSqY489VgsWLFBjY2Or7SoqKjRt2jTl5+eruLhYV155ZZttAAAAfN/Xbbfdpt27d8ta26q1Z/78+ZKk7du3JyyGeDzeann37t06VDcS60u+lRptXJ4fU9zGZCV5XlOznlVjLKb6/fvUuH+f6mprVFe7R/t371Zd9Q7V796j+j17VO/FZWMJeytNsVmr0qOLZK3Vt789RePKh2nEyCEadfIQXT77Im3dtqXV9m++8Z7Om/79VgkpAACyXdK60rz//vvyfV8PPvighgwZor/97W+6+OKLVVdXp7vuukuS5Hmepk6dqqOOOkqvvfaaduzYoZkzZ8paq8WLFycrNAAAkIUcp+nznL1796qgoKDN68YY3XTTTS3LM2fO1MyZMyVJrusmNIZmr77yki77wffVUXLkiennKGRcyfMUU1xBX7KeLz/uNSUXbNMnVcYYxa0v94tjxa0v10rGcdRgPTV6njYEA7r1uT8l5L002/zZbknS7373nL7ylWHa8lmlTpt4qu6/7+GEngcAkGCZNOhppsTRTUlLjEyePFmTJ09uWT7mmGP0wQcf6IEHHmhJjCxfvlzr169XZWWlSktLJUl33323LrzwQi1atEiFhYXJCg8AAGSpTz/9VCUlJW3WW2tVVVWlhx9+WBUVFa1emzNnTkLOfWC3HOv7uuUn8w9ZOVFRUSk3EFBBOCxXRmHjyg27cuJWUlO1SHPixnfclioY4zT971hHni/VNNQrsCemuLUKJKha48uxW1lF86PqXzogIccHACAbpHSMkT179qh3794ty6+//rqGDRvWkhSRpLPPPlsNDQ1avXp1u8doaGhQTU1NqwYAAHLHwoUL9cgjj7QZY8QYo/79+6uiokK33nqrJLVM2Xv22Wcf9nm/PEDqrl21nd7Pi8e1r75eDfGY6o0vL24VdxzVe74aYvGmr31fvqzkGMV9TzKu4p6VlaOY70my8n0rdTMpEo/HddakU3X6xFM06ex/1oX//bta8uBibd/+j+l2454n141qy+at3ToHAADZKGWz0nz00UdavHix7r777pZ1VVVV6tevX6vtevXqpVAopKqq9qe/u/3221uVyQIAgNywf/9+SdLzzz+v559/vsNtm8cX+cY3viFJGj16dEJiCAaDLV/X7K1Rfn7eIffZF4ipKObI8V3VO54CntV+x1cwEFDMGsUapECsQQHHke/F5DiOXGNl5Snmx2Q9X42urw111brnpXfl+L7kdP2zrQmnnazqbTv18Uefy1pfGz7coH+77kd67LFH5PtWe/bs0ebN2xQMBrX23b/pK+NHyhgr35esb+R5cTkBo4FlZbr0kit09qSpXY4BAJBA9ouWCTIljm7qcmJk4cKFh0xMvPXWWxo7dmzL8ubNmzV58mRNnz5ds2bNarVte+WnHU1Zd8MNN7Qqh62pqdGAAZR7AgBwpItGo93eNxHdc40xikQiLcuDBrZ+/ohZT0/c/M+64ILv6O5HVunfFj0i2aAWvP5Ou8fqiifunqFZg4fr7eeu0Ljp/9Hl2K21OuHEkdpS+PEX53d0/HH/RX98coV839eok4cqFourqGeBAsGAZKTq6u26b/ESfe2MSS3v2/M81dXt1WWXX6Lr/n2u3lnzfpdjAQAg03Q5MTJ79mxdcMEFHW4zaNCglq83b96sM844Q+Xl5XrooYdabVdSUqI333yz1bpdu3YpFou1qSRpFg6HFQ6Huxo2AAA4Qnzve9/Tzp079dxzzx1y21/84he68sorUzKDyo6PVquoZ1gvvPiKrr31f8tXSG6CTru5xlGf2p3a9MFGjT3PdLkvtDFGv3740Vbr9uzZpcnf+Jrq99ersbFRjz/+R40YPkq+72vY8CHqWdhD1//7NXr33Y9a9nFdV4WFPbVr11adcMLxCXhnAACkX5cTI8XFxSouLu7Utp9//rnOOOMMjRkzRkuXLm0zknt5ebkWLVqkLVu2qH///pKaBmQNh8MaM2ZMV0MDAAA54NFHH22z7suJD8/zWp47rrzyStXV1alHjx6dPofv+22eWw7GykoyWvn/btfQQSfqzy/8RWcplNCB3Kac8x2te/F+lfY/WgE/Jjmhbh/LWquzp05Uza5a7a3bo3XvfNTq++e6rv6+fpPmzLlMr7+5SuPGj9Qbr69ttc2zz7x0OG8HAJAI1m9qmSBT4uimpA2+unnzZk2cOFEDBgzQXXfdpe3bt6uqqqrV2CGTJk3SiSeeqBkzZmjNmjV64YUXdO211+riiy9mRhoAANAl9oupAq21bZIaH3/8cZeO1dmkSDPPejpxcF/tqq5UsPfwhFeoHHvSaXIVVUFeVHW7Pu/mUazOPKtcU6ZO1K4dO/X6qtX627sfHzTWe+55QH9e/hc5rqPTzxgrm+UPvQAAHEzSEiPLly/Xxo0b9eKLL+roo49W//79W1oz13X1zDPPKBKJ6NRTT9V5552nf/3Xf22ZzhcAACARtm3bduiNuslYXysfna99dbWKh/pqzqJfdWn/5oROhwJ52rVnt+r21+uNlf+3W3H6vlX9/gZdfeV1euvNdZ1K3uTnF+j551bKi8d15lnlnYsVAIAsk7TEyIUXXihrbbvtQGVlZXr66ae1b98+7dixQ4sXL2YMEQAAkFBbtyZv+tm4ddWw+305IUdvvrFaxnbtOaYzCQpfUu3enfL8mNb99cUux2it1VVzZ2nEmJPk+fVd2rdnYZF+vfQJ+Z6vte+81eVzAwCQ6VI2XS8AAECyHKqS4bPPPkvOeWVlYnsUdaVdu/aqoGig5HRtxr3OcGVUdFSJHBNUUc/aLu9vjHT7op9rw4a/6eSR47u8/9p3VssNurr++qv1p+dXdXl/AAAyWdIqRgAAADLF++8nZ1pZY6Unf3G5AiFXQbdIc3/6+/a3O8wxRxxZRcJ5clxp7+5d8v2ujvdhVJDXQ6NHlXcrlpGjRsmRtGN3dZf3BQAg01ExAgAAskZ9fde6gUidHMOjm/Yb6aijjGTztO4/39EZxihZEwPvrPHVp2+ebKwxSWc4uOOGnKgVy99I+XkBAB2wtqllgkyJo5uoGAEAAFkj08Yh2/Dab7R3/y7t3VejgmPPkGPcpJ1r9Onny7ghFfc+KmnnAAAgF1ExAgAA0s5aX8Yc/uc15eXlCYjm0HxZGWu09qXfqaykSEd95QZNGT5MySoXMcboK1/9ltaHGxTXc5LfKDkh8RkXAACHj9+mAAAgbSaOPEEj+/fVt48tS8jxVq1K3cCgnh/XwGP76qjjz9GwEcMPexyRQ/EdafDwUxV0rHZVvidld9UyAOBw2QxrWYzECAAASJuX3vm7/vrxJ6qKe+kOpUsca/XYkn9TJFar4yecm5JzunIULRio+v0N+sNTv5VveYwDACAR6EoDAADSKhzNk0xXZ1lJM2P0vcvv+eLL5FaKtD6t0Tdm/y5pXXYAANnDWitrM+P3ZzIHOk8FEiMAACDtYllX/WCUwnzIl08NAAASKNueQgAAwBEobUkGAACQ86gYAQAAafXVQX01vCQ/3WEAAJBdMmnQ00yJo5tIjAAAgLTKcwL6vLpBZx1TIs/3VW8k3xg1xGO6YtIoFRZEta9uvzZ/vl3GddVvQKkioYCqt2xR1PXlBEJ66M8faMRJx+jzrdUqCAcUNtL/+uvGQ557ztA8jRw9RDMffzcF7xQAAGQiEiMAACCt3IDkOFbyXUWMo16SGq2V3LCefPl9uUZq8Hw1SDLGkfdelb4+pEiNsbiWrt2uo1wjBYxi+/fq6J758mQVNJ3rLfy1783WrvV/Tur7AwAAmY3ECAAASCvXdVUjaeWGynZfv+fay/XiH59UniMF5Euuqw17pc+37Vdx0JU1VrOv+G+adePPJUnjBh2lU4YO6NS5p/7kp7p1XH9Z38o4DHQCAMgi1ja1TJApcXQTg68CAIC0Mq7kddA5ec5d9+va//FL9e3bR2d9f5aiwYDywyE9/c77Wrttp9Zu3dWSFJGkYMBVwD30Zz+PXv8jyRi54TxmegEAIIeRGAEAAGkV9yT5HX/SdPKpE7V3f70unn+r9sc9Oa6jvPwe7W4bCoUU70Sio/CoYi349rfUp3dUv7/+om5EDgAAjgR0pQEAAGkVs4cu2OhR2FM7avZJkoLGyHTw2U5BtED+IRItkvT0vT9Vvz4hranxNOCzHdIdXYkaAIA0oytNwlAxAgAA0sr3fTm249SIMUZymx5bGmXkyT/otpFIWF4nHtAeqtyvW9buUcmw8YozvggAADmLxAgAAEgvI8U7SHRITYmR59Y3Dc4al+nwg6keRUWKxb1On37e759VLObLZvmnXQAAoHtIjAAAgLQy1srxOp+UCDiSZw+eSOlb0l976+s7dSxrrQKBoHpEXNkvBoC1vu1UVxwAANLLZljLXowxAgAA0sqzVnsaO5fIkKR8x8h2kLi4/eHfdOo4frzpQc5aaXdtTBXr3tbAk8ZKxmkaDJaPjwAAyAn8ygcAAGn1wodbtLZyV6e3dxxXvn/4Y4IYz5c8yVjJ9WN69he3yFqT9QPIAQCArqFiBAAAZBVrrRq9jsck6Yx92/fIBANyAhFd9ewa3f3NcqnBKu5LjrFSgZuAaAEASBJrDzndfcpk+YcKVIwAAICs8j9Xf6RHX1t32MfxY/tkPCunIa4ePfrrtQ+rdfmEIYrF6xTLlAdNAACQdFSMAACArGJMYqbWdZ083XT5uXrz7XdUHDb694d+pUlnf1teY0iWJyQAAHIGv/YBAEDGiO1pkBN35Nu45PuS5ykea5Tvx/S7pT/Xr+79uXrmRZUXCeviBbdobPnXFZRRIBRQMBSRdQOS8WTCAZ1zTIn+8Pke2d3V2l/XoEAwpBeffkKP/3KxdtTUqHfElQ3n6X//8U9q3N+ggj4lMsGAIgVByaFiBACAXEFiBAAAZAwn5MqPS40xo1h9ozzfV14gLDcQ0Ld/cJ3OmzVPxljVxRp0+Xe+oesuvVxlRSG5AVd+MCw/ZhX34qqLOwpFeug7Y09Qvo2puna3duyu177GuCLHDtOMq67XiSccr5LefdVowgr36aFIfr5kg03dpJ3EVKUAAIDMR2IEAABkjPheX7FYTNZ3ZZyAIsGAZHw5rpXrBOX7vmKxRpnGBt374P9SrH6vdu2q0vZtW/T+ho9VsXWPjiobqtGjxqqs/yAV5IUVCBgFQiFFo1E5jitjrKzny3MkR0au68gEjIyMTMhKQUfWHP7grgAAJJW1mTPoaabE0U0kRgAAQMYIFgQVkJG338hvMJJ8xeOeAsGg/LivRt9XoxeSdRz5obACod7qGe2vXkeP07BTrIKhsALBkAKO5DiSMa7csCsjKxlH1reS4ygQdBRwrGzAyEiSkYwrWcfIyJfjMD49AAC5gsQIAADIGE7USArKjUr/eEyJtrweTvL5zQH/AgCA3EBiBAAAAACAbENXmoShThQAAAAAAOQsEiMAAAAAACBn0ZUGAAAAAIAsY62VzZAuLJkSR3dRMQIAAAAAAHIWiREAAAAAAJCz6EoDAAAAAEC2YVaahKFiBAAAAAAA5CwSIwAAAAAAIGfRlQYAAAAAgGxDV5qEoWIEAAAAAACk1P3336/BgwcrEolozJgxevXVV9MWC4kRAAAAAACyTXPFSKa0Lnj88cd19dVXa968eVqzZo1OO+00TZkyRRUVFUn6ZnWMxAgAAAAAAEiZe+65RxdddJFmzZqlE044Qffee68GDBigBx54IC3xZP0YI/aLzFRNTU2aIwEAAAAApFvz34a2i1UM2aa+sT7dIbRojuXLf5eHw2GFw+FW6xobG7V69Wpdf/31rdZPmjRJq1atSm6gB5H1iZHa2lpJ0oABA9IcCQAAAAAgU9TW1qpnz57pDiPhQqGQSkpKdNNvbkh3KK0UFBS0+bt8wYIFWrhwYat11dXV8jxP/fr1a7W+X79+qqqqSnaY7cr6xEhpaakqKyvVo0cPGWPSHU7Wq6mp0YABA1RZWanCwsJ0h4M043pAM64FHIjrAc24FnAgrgccKJ3Xg7VWtbW1Ki0tTel5UyUSiWjTpk1qbGxMdyitWGvb/E3+5WqRA3152/b2T5WsT4w4jqOjjz463WEccQoLC/mFhhZcD2jGtYADcT2gGdcCDsT1gAOl63o4EitFDhSJRBSJRNIdRrcUFxfLdd021SHbtm1rU0WSKgy+CgAAAAAAUiIUCmnMmDFasWJFq/UrVqzQhAkT0hJT1leMAAAAAACA7DFnzhzNmDFDY8eOVXl5uR566CFVVFTohz/8YVriITGCVsLhsBYsWNBhXzDkDq4HNONawIG4HtCMawEH4nrAgbge0JHzzz9fO3bs0M0336wtW7Zo2LBhevbZZzVw4MC0xGPskT6HEQAAAAAAwEEwxggAAAAAAMhZJEYAAAAAAEDOIjECAAAAAAByFokRAAAAAACQs0iMoMWiRYs0YcIE5eXlqaioqN1tjDFt2pIlS1IbKJKuM9dCRUWFpk2bpvz8fBUXF+vKK69UY2NjagNFWgwaNKjNfeD6669Pd1hIkfvvv1+DBw9WJBLRmDFj9Oqrr6Y7JKTBwoUL29wHSkpK0h0WUuSVV17RtGnTVFpaKmOMnnzyyVavW2u1cOFClZaWKhqNauLEiXrvvffSEyyS6lDXwoUXXtjmXjF+/Pj0BAt0gMQIWjQ2Nmr69Om67LLLOtxu6dKl2rJlS0ubOXNmiiJEqhzqWvA8T1OnTlVdXZ1ee+01PfbYY/r973+vuXPnpjhSpEvz1GrNbf78+ekOCSnw+OOP6+qrr9a8efO0Zs0anXbaaZoyZYoqKirSHRrS4KSTTmp1H1i3bl26Q0KK1NXVaeTIkbrvvvvaff1nP/uZ7rnnHt1333166623VFJSorPOOku1tbUpjhTJdqhrQZImT57c6l7x7LPPpjBCoHMC6Q4AmeOmm26SJC1btqzD7YqKivhU6Ah3qGth+fLlWr9+vSorK1VaWipJuvvuu3XhhRdq0aJFKiwsTFWoSJMePXpwH8hB99xzjy666CLNmjVLknTvvffqT3/6kx544AHdfvvtaY4OqRYIBLgP5KgpU6ZoypQp7b5mrdW9996refPm6Vvf+pYk6de//rX69eun3/zmN7r00ktTGSqSrKNroVk4HOZegYxHxQi6bPbs2SouLtYpp5yiJUuWyPf9dIeEFHv99dc1bNiwlqSIJJ199tlqaGjQ6tWr0xgZUuWOO+5Qnz59NGrUKC1atIhuVDmgsbFRq1ev1qRJk1qtnzRpklatWpWmqJBOH374oUpLSzV48GBdcMEF+vjjj9MdEjLApk2bVFVV1epeEQ6H9dWvfpV7RY566aWX1LdvXx133HG6+OKLtW3btnSHBLRBxQi65JZbbtGZZ56paDSqF154QXPnzlV1dTVl9DmmqqpK/fr1a7WuV69eCoVCqqqqSlNUSJWrrrpKo0ePVq9evfTXv/5VN9xwgzZt2qRf/epX6Q4NSVRdXS3P89r87Pfr14+f+xw0btw4PfLIIzruuOO0detW3XrrrZowYYLee+899enTJ93hIY2a7wft3Ss+/fTTdISENJoyZYqmT5+ugQMHatOmTbrxxhv1ta99TatXr1Y4HE53eEALKkaOcO0Njvbl9vbbb3f6ePPnz1d5eblGjRqluXPn6uabb9add96ZxHeAREn0tWCMabPOWtvuemS+rlwf11xzjb761a9qxIgRmjVrlpYsWaKHH35YO3bsSPO7QCp8+Wecn/vcNGXKFJ177rkaPny4vv71r+uZZ56R1NRlApC4V6DJ+eefr6lTp2rYsGGaNm2annvuOW3YsKHlngFkCipGjnCzZ8/WBRdc0OE2gwYN6vbxx48fr5qaGm3durXNJwPILIm8FkpKSvTmm2+2Wrdr1y7FYjGugyx1ONdH8+jyGzdu5JPiI1hxcbFc121THbJt2zZ+7qH8/HwNHz5cH374YbpDQZo1jyVRVVWl/v37t6znXgFJ6t+/vwYOHMi9AhmHxMgRrri4WMXFxUk7/po1axSJRA46pSsyRyKvhfLyci1atEhbtmxpeehZvny5wuGwxowZk5BzILUO5/pYs2aNJLV6AMaRJxQKacyYMVqxYoXOOeeclvUrVqzQv/zLv6QxMmSChoYG/f3vf9dpp52W7lCQZoMHD1ZJSYlWrFihk08+WVLTGEUvv/yy7rjjjjRHh3TbsWOHKisreWZAxiExghYVFRXauXOnKioq5Hme1q5dK0kaMmSICgoK9NRTT6mqqkrl5eWKRqNauXKl5s2bp0suuYQ+gkeYQ10LkyZN0oknnqgZM2bozjvv1M6dO3Xttdfq4osvZkaaI9zrr7+uN954Q2eccYZ69uypt956S9dcc42++c1vqqysLN3hIcnmzJmjGTNmaOzYsSovL9dDDz2kiooK/fCHP0x3aEixa6+9VtOmTVNZWZm2bdumW2+9VTU1NZo5c2a6Q0MK7N27Vxs3bmxZ3rRpk9auXavevXurrKxMV199tW677TYNHTpUQ4cO1W233aa8vDx997vfTWPUSIaOroXevXtr4cKFOvfcc9W/f3998skn+vGPf6zi4uJWCXYgI1jgCzNnzrSS2rSVK1daa6197rnn7KhRo2xBQYHNy8uzw4YNs/fee6+NxWLpDRwJd6hrwVprP/30Uzt16lQbjUZt79697ezZs219fX36gkZKrF692o4bN8727NnTRiIRe/zxx9sFCxbYurq6dIeGFPnlL39pBw4caEOhkB09erR9+eWX0x0S0uD888+3/fv3t8Fg0JaWltpvfetb9r333kt3WEiRlStXtvucMHPmTGuttb7v2wULFtiSkhIbDoft6aefbtetW5feoJEUHV0L+/bts5MmTbJHHXWUDQaDtqyszM6cOdNWVFSkO2ygDWOttalNxQAAAAAAAGQGZqUBAAAAAAA5i8QIAAAAAADIWSRGAAAAAABAziIxAgAAAAAAchaJEQAAAAAAkLNIjAAAAAAAgJxFYgQAAAAAAOQsEiMAAAAAACBnkRgBAAAAAAA5i8QIAAAAAADIWSRGAAAAAABAziIxAgAAAAAActb/B6aJajcNpGwKAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1200x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def plot_tsne_with_images(X, images, labels=None, min_distance=0.04, figsize=(12, 8), image_zoom=0.1):\n",
    "    \n",
    "    tsne = TSNE(n_components=2, random_state=42)\n",
    "    reduced_data = tsne.fit_transform(X)\n",
    "\n",
    "    if labels is not None and isinstance(labels[0], str):\n",
    "        label_encoder = LabelEncoder()\n",
    "        labels = label_encoder.fit_transform(labels)\n",
    "\n",
    "    fig, ax = plt.subplots(figsize=figsize)\n",
    "    ax.set_title(\"t-SNE Visualization of Image Embeddings\")\n",
    "\n",
    "    if labels is not None:\n",
    "        scatter = ax.scatter(\n",
    "            reduced_data[:, 0], \n",
    "            reduced_data[:, 1], \n",
    "            c=labels, \n",
    "            alpha=0.6, \n",
    "            cmap=plt.cm.viridis\n",
    "        )\n",
    "        plt.colorbar(scatter, ax=ax, label=\"Category\")\n",
    "\n",
    "    resized_images = [img.resize((250,250)) for img in images]  \n",
    "\n",
    "    neighbors = np.array([[10., 10.]])  \n",
    "    for index, image_coord in enumerate(tqdm(reduced_data, desc=\"Adding images to plot\")):\n",
    "        closest_distance = np.linalg.norm(neighbors - image_coord, axis=1).min()\n",
    "        if closest_distance > min_distance:\n",
    "            neighbors = np.r_[neighbors, [image_coord]]\n",
    "\n",
    "            if resized_images is not None:\n",
    "                image = resized_images[index]\n",
    "                imagebox = AnnotationBbox(OffsetImage(image, zoom=image_zoom), image_coord, frameon=False)\n",
    "                ax.add_artist(imagebox)\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.show()\n",
    "\n",
    "\n",
    "embeddings = np.array(embeddings)\n",
    "image_list = [Image.open(path).convert(\"RGB\") for path in image_paths]\n",
    "labels = [label for label in labels]\n",
    "\n",
    "plot_tsne_with_images(X=embeddings, images=image_list, labels=labels, image_zoom=0.1)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Observations from the t-SNE Plot\n",
    "\n",
    "### Cluster Formation\n",
    "Images group by category (e.g., airplanes, dogs, cups), showing the model's ability to capture semantic similarities.\n",
    "\n",
    "### Condition-Specific Groups\n",
    "Silhouettes cluster distinctly, while realistic and feature-based images are more dispersed but aligned within categories.\n",
    "\n",
    "### Inter-Class Separation\n",
    "Most categories (e.g., airplanes, ducks) are well-separated, but visually similar objects like chairs and cups show slight overlap.\n",
    "\n",
    "### Cluster Density\n",
    "Categories like cups and hats have dense clusters, while donkeys and dogs are more dispersed.\n",
    "\n",
    "### Silhouettes and Geons\n",
    "Silhouette images form clear clusters, reflecting the model's reliance on edge-based features.\n",
    "\n",
    "### Summary\n",
    "The model effectively groups images by category and condition but shows slight overlap for visually similar objects.\n"
   ]
  }
 ],
 "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": "3.12.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}