Training working, multiple bug fixes

42 files changed, 1119 insertions, 594 deletions

diff --git a/notebooks/00-scratch-pad.ipynb b/notebooks/00-scratch-pad.ipynb
index 0350727..a193107 100644
--- a/notebooks/00-scratch-pad.ipynb
+++ b/notebooks/00-scratch-pad.ipynb
@@ -2,18 +2,9 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 1,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "The autoreload extension is already loaded. To reload it, use:\n",
-      "  %reload_ext autoreload\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "%load_ext autoreload\n",
     "%autoreload 2\n",
@@ -33,8 +24,295 @@
     "\n",
     "from text_recognizer.networks.transformer.vit import ViT\n",
     "from text_recognizer.networks.transformer.transformer import Transformer\n",
-    "from text_recognizer.networks.transformer.layers import Decoder\n",
-    "from text_recognizer.networks.transformer.nystromer.nystromer import Nystromer"
+    "from text_recognizer.networks.transformer.layers import Decoder"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "True"
+      ]
+     },
+     "execution_count": 2,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "torch.cuda.is_available()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "loss = nn.CrossEntropyLoss()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 51,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "o = torch.randn((4, 5, 4))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 52,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "t = torch.randint(0, 5, (4, 4))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 53,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "torch.Size([4, 5, 4])"
+      ]
+     },
+     "execution_count": 53,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "o.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 54,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "torch.Size([4, 4])"
+      ]
+     },
+     "execution_count": 54,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "t.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 55,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "tensor([[0, 1, 3, 2],\n",
+       "        [1, 4, 4, 4],\n",
+       "        [1, 4, 2, 1],\n",
+       "        [2, 0, 4, 4]])"
+      ]
+     },
+     "execution_count": 55,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "t"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 56,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "tensor([[[ 0.0647, -1.3831,  0.0266,  0.8528],\n",
+       "         [ 1.4976,  0.4153,  1.0353,  0.0154],\n",
+       "         [ 1.4562, -0.3568,  0.3599, -0.6222],\n",
+       "         [ 0.2773,  0.4563,  0.9282, -2.1445],\n",
+       "         [ 0.5191,  0.3683, -0.3469,  0.1355]],\n",
+       "\n",
+       "        [[ 0.0424, -0.3215,  0.5662, -0.4217],\n",
+       "         [ 2.0793,  1.2817,  0.1559, -0.6900],\n",
+       "         [-1.1751, -0.3359,  1.7875, -0.3671],\n",
+       "         [-0.4553, -0.3952, -0.8633,  0.1538],\n",
+       "         [-1.3862,  0.4255, -2.2948,  0.0312]],\n",
+       "\n",
+       "        [[-1.4257,  2.2662,  0.2670, -0.4330],\n",
+       "         [-0.3244, -0.8669, -0.2571,  0.8028],\n",
+       "         [ 0.9109, -0.2289, -1.2095, -0.9761],\n",
+       "         [-0.0156,  1.2403, -1.1967,  0.6841],\n",
+       "         [-0.8185,  0.2967, -2.1639, -0.7903]],\n",
+       "\n",
+       "        [[-1.0425,  0.1426,  0.1383,  0.9784],\n",
+       "         [-1.2853,  1.4123, -0.2272, -0.3335],\n",
+       "         [ 1.5751, -0.7663,  0.9610,  0.5686],\n",
+       "         [ 0.9697, -1.5515, -0.8658, -0.5882],\n",
+       "         [-1.2467,  0.0539,  0.1208, -1.0297]]])"
+      ]
+     },
+     "execution_count": 56,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "o"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 57,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "tensor(1.8355)"
+      ]
+     },
+     "execution_count": 57,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "loss(o, t)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 63,
+   "metadata": {},
+   "outputs": [
+    {
+     "ename": "TypeError",
+     "evalue": "unsupported operand type(s) for |: 'int' and 'Tensor'",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
+      "\u001b[0;32m/tmp/ipykernel_9275/1867668791.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mt\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m2\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m2\u001b[0m \u001b[0;34m|\u001b[0m \u001b[0mt\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m2\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
+      "\u001b[0;31mTypeError\u001b[0m: unsupported operand type(s) for |: 'int' and 'Tensor'"
+     ]
+    }
+   ],
+   "source": [
+    "t[:, 2] == 2 | t[:, 2] == 1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 60,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "torch.Size([4, 1])"
+      ]
+     },
+     "execution_count": 60,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "torch.argmax(o, dim=-1)[:, -1:].shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 43,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "class LabelSmoothingLossCanonical(nn.Module):\n",
+    "    def __init__(self, smoothing=0.0, dim=-1):\n",
+    "        super(LabelSmoothingLossCanonical, self).__init__()\n",
+    "        self.confidence = 1.0 - smoothing\n",
+    "        self.smoothing = smoothing\n",
+    "        self.dim = dim\n",
+    "\n",
+    "    def forward(self, pred, target):\n",
+    "        pred = pred.log_softmax(dim=self.dim)\n",
+    "        with torch.no_grad():\n",
+    "            # true_dist = pred.data.clone()\n",
+    "            true_dist = torch.zeros_like(pred)\n",
+    "            print(true_dist.shape)\n",
+    "            true_dist.scatter_(1, target.unsqueeze(1), self.confidence)\n",
+    "            print(true_dist.shape)\n",
+    "            print(true_dist)\n",
+    "            true_dist.masked_fill_((target == 4).unsqueeze(1), 0)\n",
+    "            print(true_dist)\n",
+    "            true_dist += self.smoothing / pred.size(self.dim)\n",
+    "        return torch.mean(torch.sum(-true_dist * pred, dim=self.dim))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 44,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "l = LabelSmoothingLossCanonical(0.1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 45,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "torch.Size([1, 5, 4])\n",
+      "torch.Size([1, 5, 4])\n",
+      "tensor([[[0.0000, 0.0000, 0.0000, 0.0000],\n",
+      "         [0.0000, 0.0000, 0.0000, 0.0000],\n",
+      "         [0.9000, 0.9000, 0.0000, 0.9000],\n",
+      "         [0.0000, 0.0000, 0.0000, 0.0000],\n",
+      "         [0.0000, 0.0000, 0.9000, 0.0000]]])\n",
+      "tensor([[[0.0000, 0.0000, 0.0000, 0.0000],\n",
+      "         [0.0000, 0.0000, 0.0000, 0.0000],\n",
+      "         [0.9000, 0.9000, 0.0000, 0.9000],\n",
+      "         [0.0000, 0.0000, 0.0000, 0.0000],\n",
+      "         [0.0000, 0.0000, 0.0000, 0.0000]]])\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "tensor(0.9438)"
+      ]
+     },
+     "execution_count": 45,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "l(o, t)"
    ]
   },
   {
diff --git a/notebooks/03-look-at-iam-paragraphs.ipynb b/notebooks/03-look-at-iam-paragraphs.ipynb
index 76ca6b1..ed67e9c 100644
--- a/notebooks/03-look-at-iam-paragraphs.ipynb
+++ b/notebooks/03-look-at-iam-paragraphs.ipynb
@@ -2,24 +2,10 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "id": "6ce2519f",
    "metadata": {},
-   "outputs": [
-    {
-     "ename": "ModuleNotFoundError",
-     "evalue": "No module named 'loguru.logger'",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
-      "\u001b[0;32m/tmp/ipykernel_3883/2979229631.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     16\u001b[0m     \u001b[0msys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpath\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'..'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     17\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 18\u001b[0;31m \u001b[0;32mfrom\u001b[0m \u001b[0mtext_recognizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0miam_paragraphs\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mIAMParagraphs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     19\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0mtext_recognizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0miam_synthetic_paragraphs\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mIAMSyntheticParagraphs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     20\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0mtext_recognizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0miam_extended_paragraphs\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mIAMExtendedParagraphs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m~/projects/text-recognizer/text_recognizer/data/iam_paragraphs.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     20\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0mtext_recognizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0memnist\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0memnist_mapping\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     21\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0mtext_recognizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0miam\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mIAM\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 22\u001b[0;31m \u001b[0;32mfrom\u001b[0m \u001b[0mtext_recognizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmappings\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mWordPieceMapping\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     23\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0mtext_recognizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtransforms\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mWordPiece\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     24\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;32m~/projects/text-recognizer/text_recognizer/data/mappings.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      6\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mattr\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 7\u001b[0;31m \u001b[0;32mimport\u001b[0m \u001b[0mloguru\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogger\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mlog\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      8\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      9\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0mtorch\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mTensor\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
-      "\u001b[0;31mModuleNotFoundError\u001b[0m: No module named 'loguru.logger'"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "import os\n",
     "os.environ['CUDA_VISIBLE_DEVICE'] = ''\n",
@@ -62,42 +48,12 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": null,
    "id": "c6188bce",
    "metadata": {
     "scrolled": true
    },
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "2021-07-30 23:09:28.009 | DEBUG    | text_recognizer.data.mappings:__attrs_post_init__:90 - Using data dir: /home/aktersnurra/projects/text-recognizer/data/downloaded/iam/iamdb\n",
-      "2021-07-30 23:09:28.117 | DEBUG    | text_recognizer.data.mappings:__attrs_post_init__:90 - Using data dir: /home/aktersnurra/projects/text-recognizer/data/downloaded/iam/iamdb\n",
-      "2021-07-30 23:09:28.277 | INFO     | text_recognizer.data.iam_paragraphs:setup:103 - Loading IAM paragraph regions and lines for None...\n",
-      "2021-07-30 23:09:47.357 | DEBUG    | text_recognizer.data.mappings:__attrs_post_init__:90 - Using data dir: /home/aktersnurra/projects/text-recognizer/data/downloaded/iam/iamdb\n",
-      "2021-07-30 23:09:50.514 | DEBUG    | text_recognizer.data.mappings:__attrs_post_init__:90 - Using data dir: /home/aktersnurra/projects/text-recognizer/data/downloaded/iam/iamdb\n",
-      "2021-07-30 23:09:50.612 | INFO     | text_recognizer.data.iam_synthetic_paragraphs:setup:67 - IAM Synthetic dataset steup for stage None...\n",
-      "2021-07-30 23:10:02.137 | DEBUG    | text_recognizer.data.mappings:__attrs_post_init__:90 - Using data dir: /home/aktersnurra/projects/text-recognizer/data/downloaded/iam/iamdb\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "IAM Original and Synthetic Paragraphs Dataset\n",
-      "Num classes: 1006\n",
-      "Dims: (1, 576, 640)\n",
-      "Output dims: (682, 1)\n",
-      "Train/val/test sizes: 19959, 262, 231\n",
-      "Train Batch x stats: (torch.Size([1, 1, 576, 640]), torch.float32, tensor(0.), tensor(0.0026), tensor(0.0239), tensor(0.7412))\n",
-      "Train Batch y stats: (torch.Size([1, 451]), torch.int64, tensor(1), tensor(1002))\n",
-      "Test Batch x stats: (torch.Size([1, 1, 576, 640]), torch.float32, tensor(0.), tensor(0.0372), tensor(0.0767), tensor(0.8118))\n",
-      "Test Batch y stats: (torch.Size([1, 451]), torch.int64, tensor(1), tensor(1003))\n",
-      "\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "dataset = IAMExtendedParagraphs(batch_size=1, word_pieces=True)\n",
     "dataset.prepare_data()\n",
@@ -107,21 +63,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": null,
    "id": "55b26b5d",
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "1006"
-      ]
-     },
-     "execution_count": 5,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "outputs": [],
    "source": [
     "len(dataset.mapping)"
    ]
@@ -161,7 +106,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": null,
    "id": "0cf22683",
    "metadata": {},
    "outputs": [],
@@ -171,146 +116,52 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": null,
    "id": "8541e6ee",
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "torch.Size([1, 576, 640])"
-      ]
-     },
-     "execution_count": 7,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "outputs": [],
    "source": [
     "x.shape"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": null,
    "id": "40447ce6",
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "tensor([1002,   59,    6,    1,   54,    7,    2,   41,   36,   15,    4,    3,\n",
-       "         842,    2,   46,  230,   65,  439,   97,  784,  779,    7, 1003,    1,\n",
-       "         218,   18,   12,   11,    1,   20,   26,   54,   23,   36,    4,    1,\n",
-       "         511,  679,  352,  324,    4,   43,  172,   33,   14,   81,   84,    1,\n",
-       "          47,  281,   59, 1003,  890,  350,   14,   49,   33,   14,   81,   84,\n",
-       "           1,   20,   15,   95,   23,   21,    2,   24,   21,   59,    1,    2,\n",
-       "           7,   31,   54,    7,   15,   20,   54,   13,   33,    3, 1003,  784,\n",
-       "          68,  409,  196,  663,    2,   42,    1,    9,   41,   31,   89,   14,\n",
-       "        1003,  827,   89,   35,    1,   54,    7,   15,   23,   54,    7,   16,\n",
-       "           7,   21,   15,    4,   14,   42,    1,   24,   31,  247,   26,   89,\n",
-       "          28, 1003,    1,   31,    7,   21,   15,   54,    7,    2,   33,    3,\n",
-       "         867,  166,    2,   96,   15,    2,   10,  928,    2,   88,   16, 1003,\n",
-       "           3,  842,    2,   46,  230,  115,   52,   26,   52,   89,   53,  105,\n",
-       "         170,    1,    9,   41,   31,   89,    1,   17,    7,   26,   20,   54,\n",
-       "          15,   16,    7,   21,   15,  201, 1003,    3,  252,  176,   44,    1,\n",
-       "           9,   41,   31,   89,   28,    1,   20,    2,    2,   24,   31,   23,\n",
-       "          20,   15,   23,   24,   21,  201,    3,  108,   23,  216,    2,   62,\n",
-       "          13, 1003,  608,   30,   16,  105,   28,    1,    9,   41,   31,   89,\n",
-       "         663,   14,   82,   26,   58,   15,   97,    2, 1003,   10,    1,   26,\n",
-       "           2,   13,   31,   47,   24,   36,   24,   46,   13,    4,    1,    9,\n",
-       "          41,   31,   89,   14,   87,  664,    1,    2,   31,   23,    7,   21,\n",
-       "          31,    7,  201,    1,   33,   33,   33,   33, 1000, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001, 1001,\n",
-       "        1001, 1001, 1001, 1001, 1001, 1001, 1001])"
-      ]
-     },
-     "execution_count": 8,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "outputs": [],
    "source": [
     "y"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": null,
    "id": "016e8c81",
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "451"
-      ]
-     },
-     "execution_count": 10,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "outputs": [],
    "source": [
     "len(y)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": null,
    "id": "7aa8c021",
    "metadata": {
     "scrolled": true
    },
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "torch.Size([1, 576, 640])"
-      ]
-     },
-     "execution_count": 12,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
+   "outputs": [],
    "source": [
     "x.shape"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": null,
    "id": "7ef93252",
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAFfMAAAMRCAYAAADmM+iiAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOzcd7ilVXn///cHhiq9CA7CICAIGEUNIoqCgYggiLGggihYMVFjFCUiKiogMYoNjcQCEaQpSkBQ0Z+hCdgBg0ikDQNDHRiYoUm5f38863zZHE7Z+0w5M8P7dV3nmr2ftda97lWe/ddcd6oKSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZI0uqUmOwFJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkhZ1FvOVJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJGkcFvOVJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJGkcFvOVJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJGkcFvOVJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJGkcFvOVJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJGkcFvOVJEmSJEmSJEmSJEmSJEnSfJNkhyQ3LIR59k1ywYKeZ4z5D0ly/GTNv6hJsneSsyc7jyeKJF9P8rH2eaG8c5IkSZIkSZIkSZIkSZIkSbKYryRJkiRJkiRJkiRJkiRJkhagJNcl2WkeY2yYpJJMmV95aeJGOo+q+m5VvWwy8+rHknKXqmr/qvr0ZOchSZIkSZIkSZIkSZIkSZL0RGMxX0mSJEmSJEmSJEmSJEmStMRKss447WskWWZh5bMoWdyLmWrBmd93w7smSZIkSZIkSZIkSZIkSZKkJYXFfCVJkiRJkiRJkiRJkiRJ0hIlyWpJ3p3k18Cx7dmBSW5MMifJlUl2bN3/HrghyeeTPHMe5109yY+S3Jbkzvb5qT3t+ya5puVwbZK9B4z/vSQ3J7kryXlJthxw/L5JfpnkC0lmAYckWS7J55Jcn+SWJF9PskLrv1Zbw+wkdyQ5P8lSra2SbNIT+9gkh44w53HABsAZSeYm+fAgOfc4r/07u8XZtmeOz7X9vjbJLj3PV03yrSQ3tbM/NMnSg0ya5M1JpieZleRjSa5LstMofUc9nyS7JvlTO/sbkxzQ0/bhluPMJG/v3du2hu+0OzU9ycE9ZzA9yfPa573buC3b97clOW2Ade6Q5Ib2ntwMHJNkqST/muTqtv5TkqzRhjzuPNr9uqAnZiX5pyR/Af7Snu2W5JJ2py5M8qye/tcl+VCSy5Lc085unSQ/bvv28ySr9/R/QYsxO8mlSXboaTsnyafbfZ+T5Owka42W+wD7tF3PnDOS7Nuej3VOhyQ5vifGhm1vpvSR61hzjvjODbCWFdL97k1vd/aCPPruj3WXj03ytXYuc1ve6yb5Yrr38M9JntPTf2qSU9veXJvkfRPNWZIkSZIkSZIkSZIkSZIkaVFgMV9JkiRJkiRJkiRJkiRJkrTYa4VHX5bkRGA68DLgMOCVSTYD3gNsXVUrAzsD1wFU1cnAjsAjwNlJfpPkH3uLhg5gKeAYYBpdAdv7gKNafk8Cvgzs0nJ4IXDJgPF/DDwdeDLwe+C7E8hxG+AaYB26/TkC2BTYCtgEWA/4eOv7QeAGYO3W/yCgBpmsqvYBrgd2r6qVquqzE8gZ4CXt39VanIt61nMlsBbwWeBbSdLajgUeolvXc+juxNv7nTDJFsDXgL2BpwCr0u3PaMY6n28B72pn/0zgF22OlwMfAHZqee4wLOZX2rwbAdsDbwb2a23n9vTfnu5cX9Lz/dw+lzpkXWANuvv7TuC9wKtarKnAncBXW9/RzmO4V9Gd0RatwOu3gXcBawJHA6cnWa6n/2voCmxvCuxOt6cH0d3BpYD3ASRZDzgTOLTlfABwapK1e2LtRbdXTwaWbX0Gyf0xkkxr+Xyl5bMVj77DY51TP0bMdZw559XngOfR/RatAXyY7ncQxv+t2RM4mO69ewC4qPVbC/g+cGTLfyngDOBSundnR+D9SXaeT2uQJEmSJEmSJEmSJEmSJEla6CzmK0mSJEmSJEmSJEmSJEmSFmtJ3kNXnPcIuqKSG1fVP1TVf1fVg8DDwHJ0BUWXqarrqurqofFV9b9V9SFgfeAQugKp1yY5Kckq/eZRVbOq6tSqureq5tAVy92+p8sjwDOTrFBVN1XV5YOss6q+XVVzquqBluezk6w6SAxgZlV9paoeAu6nK9r6L1V1R8v5cOANre+DdEVsp1XVg1V1flUNVMx3IZheVd+oqoeB/6LLd50k6wC7Au+vqnuq6lbgCzy6tn68Fjijqi6oqr/SFTkedf3jnM+DdPdvlaq6s6p+357vCRxTVZdX1b1tHABJlm75fqTFvQ74PLBP63Iuj96vFwOf6fk+kWK+jwCfqKoHquo+YH/go1V1Q8+aXptkygAxP9Pu1n10d+3oqvpVVT1cVf9FVwj2BT39v1JVt1TVjcD5wK+q6g9VdT/wQ7qizABvAs6qqrOq6pGq+hnwW7ozH3JMVf1fm/sUukK482Iv4OdVdWJ7H2ZV1SV9nFM/Rst1xDnncR1DRXbfCvxzVd3YzuPCds79/Nb8sKp+13Mu91fVd9p7eDKPntPWwNpV9amq+mtVXQN8g8HeQ0mSJEmSJEmSJEmSJEmSpEWKxXwlSZIkSZIkSZIkSZIkSdLi7mnA6sAlwKXArN7GqroKeD9dUcpbW5HeqcODtEKUf2wx7gCeCSzTbxJJVkxydJLpSe4GzgNWS7J0Vd0DvJ6uQOpNSc5M8owBYi+d5IgkV7fY17WmtfqN0czo+bw2sCLwuySzk8wGftKeA/w7cBVwdpJrkvzrgHMtDDcPfWjFcAFWAqbRnd1NPWs7GnjyALGn0rNfLf6skTr2cT6voSs0Oz3JuUm2HWmOYZ/XamuY3vNsOrBe+3wu8OIkTwGWpisC+6IkGwKr0r0Pg7itFWcdMg34Yc/+XUFXGHudAWL2rmca8MGheC3m+nR7MOSWns/3jfB9pZ5YrxsWazu6Ys5Dbu75fG/P2IlaH7h6hOfjnVM/Rst1tDnn1VrA8iPF7vO3ZpBzmjrsnA5isDskSZIkSZIkSZIkSZIkSZK0SLGYryRJkiRJkiRJkiRJkiRJWqxV1QeBjYH/Bb4CXJvk00me3tPnhKrajq64ZAH/NtSWZKUk+yb5BfB7uiKcr6+qZ1bViMVbR/FBYDNgm6paBXjJ0BQth59W1d/TFRz9M/CNAWLvBewB7ERXqHXD3tgDqJ7Pt9MV3tyyqlZrf6tW1Uot3zlV9cGq2gh4JfCBJDu2sffSFQIesm6fc07UoDFmAA8Aa/WsbZWq2nKAGDcBTx36kmQFYM1R+o55PlX1m6rag66Y8Gl0hXcfNwdd8dYhtwMP0t3ZIRsAN7aYV9Gdw3uB86rqbrqisO8ELqiqR/peaWf4Hs8AdunZv9WqavmqunGEvv3EnAEcNizeilV14oB5DsU6blisJ1XVEQPmNOicG4/wfMxzAu6h/3el3znn1e3A/aPEnl+/NdDlf+2wc1q5qnadQCxJkiRJkiRJkiRJkiRJkqRFgsV8JUmSJEmSJEmSJEmSJEnSYq+qbq2qI6vqWcBrgNWAi5J8O8lmSf4uyXJ0BSzvAx4BSPJyYCbweuBoYL2q+seq+s0E0li5xZ6dZA3gE0MNSdZJskeSJ9EVmZ07lMMAsR8AZtEVBj18Avk9Riv2+g3gC0me3PJcL8nO7fNuSTZJEuAu4OGenC8B9kqydNvD7ceY6hZgo94HSa5Lsu8A6d7W5t5ovI4AVXUTcDbw+SSrJFkqycZJxspzuO8Duyd5YZJlgUMYvaDpqOeTZNkkeydZtaoeBO7m0X08BdgvyeZJVgQ+1rOGh1v7YUlWTjIN+ABwfM+85wLvaf8CnDPs+7z4ept7WlvH2kn2aG0DnUfzDWD/JNuk86Qkr0iy8gRyO57ubHZud3D5JDskeeq4IyeWO8B3gZ2S7JlkSpI1k2zVxzldArwkyQZJVgU+Mq9zDpj347R3/9vAkUmmtj3ctv1Gzs/fml8Dc5IcmGSFNs8zk2w9r2uQJEmSJEmSJEmSJEmSJEmaLBbzlSRJkiRJkiRJkiRJkiRJS5Sq+l1VvReYSleQdDngCOB24GbgyTxaUPNK4BlVtUtVnVxVD8zD1F8EVmjzXAz8pKdtKboCnzOBO+iK3757gNjfAaYDNwJ/avHnhwOBq4CLk9wN/BzYrLU9vX2fC1wEfK2q/qe1/TOwOzAb2Bs4bYw5PgMcnGR2kgNaYdw1B1lDVd0LHAb8ssV5QR/D3gwsS7dfd9IV533KAHNeDrwXOAm4iW4fbqUrdDrceOezD3Bd2+P96faMqvox8GXgf2jn0PoPzfFe4B7gGuAC4AS6IqxDzqUrvnreKN/nxZeA04Gzk8xpuW3T8h74PKrqt8A7gKPozuMqYN+JJFZVM4A9gIPoivPOAD5EH/83eoJ3iaq6HtgV+CDdO3wJ8OzWPOo5VdXPgJOBy4DfAT/qZ74+5pxXBwB/BH7TYv8b3f7Nt9+aVuh4N2Ar4Fq638ZvAqvOQ96SJEmSJEmSJEmSJEmSJEmTKlU12TlIkiRJkiRJkiRJkiRJkiTpCSLJdsA/VdUbJzuXQSRZia548dOr6toFNMfmwP8Cy1XVQwtiDkmSJEmSJEmSJEmSJEmSJEkTZzFfSZIkSZIkSZIkSZIkSZKkMSR5J7DRON2urKpjJhD7dcDzxul2a1UdOUaMA4HVx4lxcVWdNmB6C9z8WP+CnDPJ7sD/BwT4PLAN8Nyaj/8BN8k/AGcBKwL/BTxSVa+aT7EX2N1dUkzGHVxQkqwOHNhH1xOq6rIFnY8kSZIkSZIkSZIkSZIkSdKSyGK+kiRJkiRJkiRJkiRJkiRJY0hyDrD9ON1+WlUvn0DsY4G3jNPtyqp6xhgxrgOmjRPj6Kraf7DsFrz5sf4FOWeSbwKvpSvm+1vgH6vqyvmVS8vnJ8C2wMPAuW2Om+ZT7HNYQHd3STEZd3BBSbIhcG0fXd9YVSct4HQkSZIkSZIkSZIkSZIkSZKWSBbzlSRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRpHEtNdgKSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJC3qLOYrSZIkSZIkSZIkSZIkSZKesJJ8PcnHJnH+fZNcMMGxeyc5e37nNMZ8ByX55sKab1BJNkgyN8nS7fs5Sd4+2XktatoebTRG+3VJdprPcy6yZ5FkwySVZMoikMshSY5fCPMsMmuWJEmSJEmSJEmSJEmSJEla3FjMV5IkSZIkSZIkSZIkSZIkPWFV1f5V9el++iY5NsmhCzqnflXVd6vqZQtxvsOrapEsyApQVddX1UpV9fD8iJfkKUlOTzKzFT7dcFj7ekn+O8kdSW5Isv+w9kpyTyueO7e3EHKS5Voh6Vva+DOSrDc/8h5P26NrWh6L1J2WJEmSJEmSJEmSJEmSJEmSFnUW85UkSZIkSZIkSZIkSZIkSdJiLcmUBRD2EeAnwGtGaT8euBZYB3gFcHiSlw7r8+xWPHelYYWQ/xnYFngWMBW4E/jK/Ex+cbGAzm6+SMf/by1JkiRJkiRJkiRJkiRJkqT/x/9cKkmSJEmSJEmSJEmSJEmSnrCSHJvk0PZ5hyQ3JPlgkluT3JRkv9b2TmBv4MNJ5iY5oz3fPMk5SWYnuTzJKyeWRo5KcleSPyfZsadh1STfarncmOTQJEu3tn2TXNDTt5Lsn+QvLZ+vJklrWzrJ55PcnuTaJO9p/fsupJrkkCTHt88btvFvSXJ9i/vRARe9epIfJbktyZ3t81N72s9J8pkkv05yd5L/TrLGsPnfluR64Bc9z+ZLcdiquqWqvgb8ZoTcVwJ2AA6rqger6lLg+8Bb+wz/NOCnbY77gZOBLSeaa5L9hu5k+/6XJN/r+T4jyVbtcyXZZLQ73WyV5LJ2J09OsvyA+fx9u8t3JTkKSE/bvkl+meQLSWYBh0xgvc9PclG75ze192fZnvbx3oXPtTt7DV0h5t7Y5yQ5LMkvgXuBjZI8I8nPktyR5Moke7a+T2vxl2rfv5Hk1p5YxyV5/4DLWzbJd5LMab8pf9sTb9Tfm/Zb8Z32Pk1PcnBPXmOuuR9j7Xk6X0j3u3l3kj8meeagc0iSJEmSJEmSJEmSJEmSJC0OLOYrSZIkSZIkSZIkSZIkSZL0qHWBVYH1gLcBX02yelX9J/Bd4LNVtVJV7Z5kGeAM4GzgycB7ge8m2WzAObcBrgbWAj4B/GCoaC1wLPAQsAnwHOBlwNvHiLUbsDXwLGBPYOf2/B3ALsBWwHOBVw2Y42i2AzYDdgQ+nmTzAcYuBRwDTAM2AO4DjhrW5810BXKfQrcPXx7Wvj2wOY+uc2HJsH+HPg8vYHpekpuT/CDJhj3PvwW8KMnUJCvSFdX98Tzkcy7w4iRLJZkKLAtsC5BkI2Al4LLeASPd6Z7mPYGX0xUdfhawb7+JJFkL+AFwMN2dvhp40bBu2wDXAOsAh/Ubu8fDwL+0+NvS3b9/HNZnrHdhN7r36W+B144Qfx/gncDKwG3Az4AT6N7zNwBfS7JFVV0L3N1iAbwEmNvzHmxPdzaDeCVwErAacDrtnejj9+YrdL9dG7V53wzsN8CaxzPWnr+Mbu2bthz2BGZNYA5JkiRJkiRJkiRJkiRJkqRFnsV8JUmSJEmSJEmSJEmSJEmSHvUg8KmqerCqzgLm0hWrHckL6IqkHlFVf62qXwA/At444Jy3Al9sc54MXAm8Isk6wK7A+6vqnqq6FfgCXTHR0RxRVbOr6nrgf+iK90JXXPNLVXVDVd0JHDFgjqP5ZFXdV1WXApcCz+53YFXNqqpTq+reqppDV9R1+2Hdjquq/62qe4CPAXsmWbqn/ZC2N/fN60IG0fL9JfCxJMsneS7wGmDFnm7bAxsCzwBmAj9KMqW1/QWYAdxIVwx2c+BT85DPNcAcuvN+CfBTYGaSZ7Q8zq+qRwYI+eWqmllVd9AVkN1qgLG7ApdX1fer6kHgi8DNw/rMrKqvVNVDEzm7qvpdVV3cxl8HHM3j785Y78IXq2pGW99nRpji2Kq6vKoeoitqfF1VHdPm+wNwKvC61vdcYPsk67bv32/fnwasQvdeDOKCqjqrqh4GjuPRd2rU35v2TrwB+EhVzWl78nm6osT9rnlM4+z5g3SFj58BpKquqKqbBp1DkiRJkiRJkiRJkiRJkiRpcTBl/C6SJEmSJEmSJEmSJEmSJElPGLNaAc8h99IV0BzJVGDGsCKp04H1BpzzxqqqYTGmAtOAZYCbkgy1LUVXBHY0vUVTe3OfOmzcWDEGMdp840qyIl1x4pcDq7fHKydZuhUyhcfmOZ1uP9bqeTa/1jERewNfbTlcAxwPbDnUWFXntY9/TfLPPFq0949t3HLAmsA9wIeBHwPbzEM+5wI7AJu0z7Ppiq1u274PYvi5Th1g7GPuWlVVkuHnNE/nlmRT4Ejgb+kKKE8BfjesW7/vwvQRpuhtnwZsk2R2z7MpdIV2odvbVwI3AOcB59AV0b2fwYsoj5T38q0I9Fi/N2vRvRvTR2iD/tY8prH2vKp+keQouns9LckPgAOq6u5B55EkSZIkSZIkSZIkSZIkSVrULTXZCUiSJEmSJEmSJEmSJEmSJC0matj3mcD6SXr/P+YGwI0Dxl0vPdV6W4yZdMU3HwDWqqrV2t8qVbXliFHGdhPw1J7v608gxvz2QWAzYJuqWgV4SXveuxe9eW4APAjc3vNs+JksNFU1vap2q6q1q2obuoKqvx5rCI+ubSvg2Kq6o6oeAL4CPD/JWqMN7sNQMd8Xt8/n0hXz3Z7Ri/kuiP27iZ5za3d7+H2b13n/A/gz8PR2dw7isfem7/zo7tVwvfnNAM7teQdXq6qVqurdrf1cuj3foX2+AHgRY+/7RIz1e3M73bsxbYQ26G/N4xlzz6vqy1X1PGALYFPgQxOYQ5IkSZIkSZIkSZIkSZIkaZFnMV9JkiRJkiRJkiRJkiRJkqT+3AJs1PP9V8C9wIeTLJNkB2B34KQB4z4ZeF+L8Tpgc+CsqroJOBv4fJJVkiyVZOMk208g91OAf06yXpLVgAMnEGN+Wxm4D5idZA3gEyP0eVOSLZKsCHwK+H5VPTzoREk2TFJJNhxw3PLAcu3rcu37UNvmSVZOsmySNwEvA45sbVsm2SrJ0klWAj5PV1j1ijb8N8Cbk6yaZBngH4GZVXV7G39skmMHXOa5wEuBFarqBuB84OXAmsAfRhkz/E7PD2cCWyZ5dZIpwPuAdefzHCsDdwNzkzwDePc4/XudQve+PTXJ6sC/jtP/R8CmSfZp7+gySbZOsjlAVf2F7h6/ia7o7910+/oa5m8x31F/b9o7cQpwWLuT04APAMdPcM0jGXXP235s0+7yPcD9wCMTWaQkSZIkSZIkSZIkSZIkSdKizmK+kiRJkiRJkiRJkiRJkiRJ/fkWsEWS2UlOq6q/0hXT3AW4Hfga8Oaq+vOAcX8FPL3FOAx4bVXNam1vBpYF/gTcCXwfeMoEcv8GXWHgy+gKu54FPAQMXBh3PvoisALdui8GfjJCn+OAY4GbgeXpCsNOxPrAdLqCuoO4D5jbPv+5fR+yM3AN3bnsD7y8qm5rbesAJ9MVP70G2BDYraoebO0H0BU8/QtwG7Ar8A/D8v3lIIlW1f+1XM9v34fm/uUYBZAfc6cHmW+MPG4HXgccAcyiu9ujriXJ3kkuH3CaA4C9gDl0d/vkAcZ+A/gpcCnwe+AHY3Wuqjl0hZrfAMyku4v/xqNFnqEr2jurqmb0fE+LP1/08XvzXrpCutcAFwAnAN9ubQOteRRj7fkq7dmddO/ZLODfJzCHJEmSJEmSJEmSJEmSJEnSIi9VNdk5SJIkSZIkSZIkSZIkSZIkaSFKsgvw9aqaNtm5jCbJOcDxVfXN+RDrYOC2qjp6nhNbwJIsS1d09Vk9xX8lSZIkSZIkSZIkSZIkSZIkLQKmTHYCkiRJkiRJkiRJkiRJkiRJ81uS1YED++h6QlVdtgDmPxBYfZxuF1fVafN77pEkWQF4KXA2sA7wCeCHre2IPkL8uKrO7XOudwIbjdPtyqo6pp9480NVHTr0eVE7m+Gq6q/A5gBJnge8ro9hh1bV3AWZV5KdgJ3G6fZgVX1sHuaYr3dxUbEor2tRzk2SJEmSJEmSJEmSJEmSJGlRZDFfSZIkSZIkSZIkSZIkSZK0JFqV/or5XgLM92K+wLuBaeP0ORo4bQHMPZIAnwROBu4DzgQ+3tr62afZQL/FPPcCth+nz0+BhVbMd5hF7WzG8jf0dz5fBBZoMV9gO8bP5QFgwsV8+4gPg93FRcWivK5FOTdJkiRJkiRJkiRJkiRJkqRFTqpqsnOQJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJGmRttRkJyBJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJ0qLOYr6SJEmSJEmSJEmSJEmSJElAkg2TVJIpExw/N8lG8zuvxUGSfZNcMNl5LExJzkny9snOYyKSXJdkp1HaLk+yw8LNSIuTJIckOX6y85AkSZIkSZIkSZIkSZIkSZoMFvOVJEmSJEmSJEmSJEmSJEmaD6pqpaq6ZrLzkHolOTbJof32r6otq+qcBZiSxjFWseWFLckOSW6Y7DwkSZIkSZIkSZIkSZIkSZIWFRbzlSRJkiRJkiRJkiRJkiRJWsCSLD3ZOSxJkkyZ7BzGsqjnp8mzMO6GvzeSJEmSJEmSJEmSJEmSJEkLjsV8JUmSJEmSJEmSJEmSJEnSE0qSFZJ8Psn0JHcluSDJCj1d9k5yfZLbk3y0Z9zzk1yUZHaSm5IclWTZnvZKskn7fGyS/0hyVpJ7gJcmuS7Jh5JcluSeJN9Ksk6SHyeZk+TnSVbvife9JDe3HM9LsmVP27FJvprkzDb2V0k2HnAf9ktyRRt/TZJ39bTtkOSGJB9Mcmtb73497WsmOT3J3Ul+DQw0d4txTpJDk1yYZG6SM1rc77a4v0my4bD9/ackfwH+Muh8LcaE9zTJ3yf5cxt7FJCetn2T/DLJF5LMAg5JslySz7W7dEuSrw/dsyTnJnlN+/yitrZXtO87JrlkfqwryTuBvYEPD+3xCGM3T3Jtkje279cl2al9PiTJKUm+0/bk8iR/2zP2wCQ3trYrk+zYni+X5ItJZra/LyZZrrWNebf6XO/GSX6RZFa69/S7SVYbL68+Y6+V5Efp3vM7kpyfZKnWNjXJqUlua3v2vp5xSyc5KMnVbd7fJVm/tT3u7ibZLcklbZ4LkzyrPT8O2AA4o53Zh9vz8e5u7+/NB9qdW7qnz6uTXDrgPj8J+DEwteUyN8nU1rzsGPdi1H2SJEmSJEmSJEmSJEmSJEla3FnMV5IkSZIkSZIkSZIkSZIkPdF8Dnge8EJgDeDDwCM97dsBmwE7Ah9Psnl7/jDwL8BawLat/R/HmGcv4DBgZeCC9uw1wN8DmwK70xXKPAhYm+7/dfYWvfwx8HTgycDvge8Oi/8G4JPA6sBVba5B3ArsBqwC7Ad8Iclze9rXBVYF1gPeBnw1jxYb/ipwP/AU4K3tbyLeAOzT5tgYuAg4hu5crgA+Maz/q4BtgC0mON+E9jTJWsAPgIPpzv9q4EXDxm4DXAOs08YdQXfOWwGbtDV+vPU9F9ihfd6+jXtJz/dz58e6quo/2+fPVtVKVbV776B23j8F3ltVJ44S+5XAScBqwOnAUW3sZsB7gK2ramVgZ+C6NuajwAvo1v5s4Pl0ezdkrLvVjwCfAaYCmwPrA4f0kVc/PgjcQPdOrkP3flYr6HsGcGnLe0fg/Ul2buM+ALwR2JXunXorcG9P3FfR7m6S5wDfBt4FrAkcDZyeZLmq2ge4Hti9ndln2/jx7m7v781XgFnAy3ra9wG+M8A+UFX3ALsAM1suK1XVzNY82r0Yb58kSZIkSZIkSZIkSZIkSZIWaxbzlSRJkiRJkiRJkiRJkiRJTxit0ORbgX+uqhur6uGqurCqHujp9smquq+qLqUrSPlsgKr6XVVdXFUPVdV1dAU4tx9juv+uql9W1SNVdX979pWquqWqbgTOB35VVX9o7T8EnjM0uKq+XVVzWm6HAM9OsmpP/B9W1a+r6iG6wp5bDbIXVXVmVV1dnXOBs4EX93R5EPhUVT1YVWcBc4HNkixNV5T441V1T1X9L/Bfg8zd45iWw110xUqvrqqftzV9j579aD5TVXdU1X0TmWwe9nRX4PKq+n5VPQh8Ebh5WPiZVfWVNvZ+4J3Av7R85wCH0xULhq5Y79DdeQldYdqh7wMX8+1jXSN5MV0R1jdX1Y/G6HdBVZ1VVQ8Dx9HeB7ri1svRFaddpqquq6qrW9vedHfn1qq6ja5A8j49MUe8WwOs96qq+llVPdDiH8mj+zdWXv14kK5I9bSW3/lVVcDWwNpV9amq+mtVXQN8g0fP9O3AwVV1ZXunLq2qWT1xe+/uO4Gjq+pX7Tfov4AH6Aogj7bm8c54+O/NfwFvAkiyBl1R4xMG2IfxjHYvxtsnSZIkSZIkSZIkSZIkSZKkxZrFfCVJkiRJkiRJkiRJkiRJ0hPJWsDywFjFPXuLtN4LrASQZNMkP0pyc5K76YqzrjVGnBkjPLul5/N9I3wfmmvpJEckubrNdV1P/mPm2a8kuyS5OMkdSWbTFaztjT+rFaYdPsfawBQeu77pg8zdo6/96DHSnvZlHvd0au/crbjr8Fx6v68NrAj8Lsnstr8/ac8BLgI2TbIOXcHg7wDrJ1kLeD5w3nxe10j2By6sqnPG6Td8T5ZPMqWqrgLeT1dY9tYkJyWZ2vpN5bF3Ynp7NmS0u9WXJOu0+W5saz6ett5x8urHvwNXAWcnuSbJv7bn04CpQ+fZzvQgYJ3Wvj5j/6703o9pwAeHxVqfx+5R73r7OePh9/F4YPckTwL2BM6vqpvGyG9QI94Lxt8nSZIkSZIkSZIkSZIkSZKkxZrFfCVJkiRJkiRJkiRJkiRJ0hPJ7cD9wMYTGPsfwJ+Bp1fVKnQFKjNG/5rAHEP2AvYAdgJWBTZsz8ear29JlgNOBT4HrFNVqwFn9Rn/NuAhuuKjQzaYH3n1YbL29CZ61pskPHb9w3O7na4Y8ZZVtVr7W7WqVgKoqnuB3wH/DPxvVf0VuBD4AHB1Vd0+H9c12p7tD2yQ5AsDzPUYVXVCVW1HV8C1gH9rTTPbsyEbtGfzy+Ftvr9p7+Kb6DnHMfIaV1XNqaoPVtVGwCuBDyTZka5Y7rU957laVa1cVbu2oTMY+3el9xxmAIcNi7ViVZ04Ql/o7+4+ZkxV3UhXNPrVwD7AceOvfty8+zHePkmSJEmSJEmSJEmSJEmSJC3WLOYrSZIkSZIkSZIkSZIkSZKeMKrqEeDbwJFJpiZZOsm2rbjteFYG7gbmJnkG8O4FmOrKwAPALGBFuuKl89OywHK0wrxJdgFe1s/AqnoY+AFwSJIVk2wBvGU+57cgzMuenglsmeTVSaYA7wPWHa1zu2ffAL6Q5MkASdZLsnNPt3OB97R/Ac4Z9r1f463rFmCjEcbNAV4OvCTJEQPOSZLNkvxde3fupyte/EhrPhE4OMnaSdYCPg4cP+gcY1gZmAvclWQ94EN95jWuJLsl2aQVbL4LeLiN/zUwJ8mBSVZovx3PTLJ1G/pN4NNJnp7Os5KsOco03wD2T7JN6/ukJK9IsnJrH35mE7273wE+DPwN3Ts7EbcAayZZtc/+4+2TJEmSJEmSJEmSJEmSJEnSYs1ivpIkSZIkSZIkSZIkSZIk6YnmAOCPwG+AO4B/o7//U3kAsBddEdRvACcvqATpinBOB24E/gRcPD+DV9UcuoK0pwB30q3r9AFCvAdYCbgZOBY4Zn7m148kc5O8eIAhE97TqrodeB1wBF1B1acDvxxn2IHAVcDFSe4Gfg5s1tN+Ll2R1vNG+d6v8db1LWCLJLOTnNbbUFWzgb8Hdkny6QHnXY5uP26nuwdPBj7S2g4FfgtcRveu/b49m18+CTyXrtjumTy2UO1YefXj6XRnNRe4CPhaVf1PK2K9G7AVcG2L/01gqMjtkXTv09l0Rb+/Baww0gRV9VvgHcBRdO/fVcC+PV0+Q1cMeXaSA5j43f0hMA34YVXdm2SD9t5s0Od4qurPdMWZr2n5TB2n/3j7JEmSJEmSJEmSJEmSJEmStFhLVU12DpIkSZIkSZIkSZIkSZIkSZKk+SzJ1cC7qurnk52LJEmSJEmSJEmSJEmSJEnSkmDKZCcgSZIkSZIkSZIkSZIkSZI0PyRZHTiwj64nVNVlCzqfyZBkI+CdfXT9alXNWBLmT/I64HnjdLu1qo6cH/MtLEvqusaS5EBg9XG6XVxVp00g9juBjcbpdmVVHTNo7EVVktcABfxihLYn3H5IkiRJkiRJkiRJkiRJkiTND6mqyc5BkiRJkiRJkiRJkiRJkiRpniXZELi2j65vrKqTFnA6kyLJDsD/9NF126q6eEmYP8mxwFvG6XZlVT1jfsy3sCyp6xpLkuuAaeN0O7qq9p9A7HOA7cfp9tOqevmgsRdFbb1bAPtU1U9HaX/C7IckSZIkSZIkSZIkSZIkSdL8YjFfSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZLGsdRkJyBJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJ0qLOYr6SJEmSJEmSJEmSJEmSJEmLmSQbJJmbZOn2/Zwkbx9g/IZJKsmUBZflgpfk60k+Ntl5TJYkL05y5WTnsThKskOSGyZh3oOSfHNhzzs/JDk2yaGTnYckSZIkSZIkSZIkSZIkSdJkspivJEmSJEmSJEmSJEmSJEnSYqaqrq+qlarq4cnOZWFJsm+SC3qfVdX+VfXpycppslXV+VW12dD3JNcl2ann+xJRtHlxNVLB4Ko6vKr6LrwtSZIkSZIkSZIkSZIkSZKkRYvFfCVJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJGofFfCVJkiRJkiRJkiRJkiRJkvqQZIUkn08yPcldSS5IskJre2WSy5PMTnJOks17xlWSTXq+H5vk0PZ5hyQ3JPlgkluT3JRkv/HmTLJhiztlHpe1d5Lrk9ye5KM98z4/yUVtPTclOSrJsq0tSb7Q8r07yR+TPLPfCZNsnOQXSWa1eb+bZLWe9vWT/CDJba3PUW0/vw5sm2Ruktkj7OUVSXbriTOlxXhu+/6CJBe2NV2aZIdBNirJ0kkOSnJ1kjlJfpdk/db2wiS/aWf0myQv7Bl3TpJD29xzk5yRZM227rtb/w17+leSf0zylzbPp9ueXdj6n9JzFjskuaF9Pg7YADijzfNh4LwWdnZ7tm3r+9a2X3cm+WmSaT3zPyPJz5LckeTKJHv2tO2a5E8trxuTHDDgHo63T59O8ssW/+wkaw0Yf2qSU9u5X5vkfT1tK7T7cmeSPwFbDxs76nvavu+R5JJ2BlcneXl7vl/byzlJrknyrvb8ScCPgalt7+e2/A5JcnxP3LF+O65LckCSy9qenZxk+QH35EtJZrS8f5fkxT1th7T79J2W/+VJ/ran/TlJft/aTgYGmluSJEmSJEmSJEmSJEmSJGlJZDFfSZIkSZIkSZIkSZIkSZKk/nwOeB7wQmAN4MPAI0k2BU4E3g+sDZxFV1B12T7jrgusCqwHvA34apLVx5pzfiym2Q7YDNgR+HhPIdGHgX8B1gK2be3/2NpeBrwE2LTlvScwa4A5A3wGmApsDqwPHAJdwVzgR8B0YEO6PTmpqq4A9gcuqqqVqmq1EeKeCLyx5/vOwO1V9fsk6wFnAofS7eMBwKlJ1h4g7w+0+LsCqwBvBe5NskaL/WVgTeBI4Mwka/aMfQOwT1vPxsBFwDEtlyuATwyba2e6c38B3Zn/J/Amur165rB1AlBV+wDXA7u3Pfos3TkBrNaeXZRkD+Ag4NV09/V8ur0bKkD7M+AE4Mkt768l2aLF+RbwrqpaueXxi343r8992gvYr829LN059Rt/KeAM4FK6fd4ReH+SnVuXT9Dt/cZ0+/uWAWI/H/gO8CFgNbp9va413wrsRncn9gO+kOS5VXUPsAsws+39SlU1c1jcfn479gReDjwNeBawb795N78BtqK7aycA3xtWEPiVwEltXacDR7XclgVOA45rY78HvGbAuSVJkiRJkiRJkiRJkiRJkpY4FvOVJEmSJEmSJEmSJEmSJEkaRysU+lbgn6vqxqp6uKourKoHgNcDZ1bVz6rqQboCvCvQFeDtx4PAp6rqwao6C5gLbDbOnPPLJ6vqvqq6lK4I6rMBqup3VXVxVT1UVdcBRwPb9+S7MvAMIFV1RVXd1O+EVXVV26sHquo2uqKuQ7GfT1fk90NVdU9V3V9VF/QZ+gTglUlWbN/3ohWppSuEe1ZVnVVVj1TVz4Df0hXm7dfbgYOr6srqXFpVs4BXAH+pquPafp0I/BnYvWfsMVV1dVXdBfwYuLqqfl5VD9EVSX3OsLk+W1V3V9XlwP8CZ1fVNT3jh/cfxP7AZ9q5PQQcDmyVZBpdUdrrquqYtpY/AKcCr2tjHwS2SLJKVd1ZVb8fYN5+9+n/quo+4BS6IrT92hpYu6o+VVV/raprgG/QFSSGrijuYVV1R1XNoCsq3K+3Ad9u9/aR9j7+GaCqzmxnW1V1LnA28OI+4/bz2/HlqppZVXfQFSveaoC8qarjq2pW2/PPA8vRFfAeckF7Lx6mK9z77Pb8BcAywBfbb9P36QoDS5IkSZIkSZIkSZIkSZIkPaFZzFeSJEmSJEmSJEmSJEmSJGl8awHLA1eP0DYVmD70paoeAWYA6/UZe1YrqjrkXmClceacX24eYV6SbJrkR0luTnI3XcHXtQCq6hfAUcBXgVuT/GeSVfqdMMk6SU5KcmOLffxQbGB9YPqw/ehLVV0FXAHs3gr6vpKuwC/ANOB1SWYP/QHbAU8ZYIr16eP8m+k89vxv6fl83wjfVxo2ftD+g5gGfKlnH+4A0vKdBmwzbJ/2BtZtY19DVwB5epJzk2w7wLz97NOI97FP04Cpw3I/CFinZ/4Zw+bu12hnT5Jdklyc5I425648ep/H089vx7zsCUkOSHJFkrtafqsOy294/OWTTGm53VhV1dM+yJ5JkiRJkiRJkiRJkiRJkiQtkSzmK0mSJEmSJEmSJEmSJEmSNL7bgfuBjUdom0lXSBSAJKEr/nlje3QvsGJP/3Xpz1hzLmj/AfwZeHpVrUJXFDVDjVX15ap6HrAFsCnwoQFiHw4U8Dct9pt6Ys8ANmjFRIerEZ4NdyLwRmAP4E+twO9Q3OOqarWevydV1RED5D2DPs6/2YBHz39hGr5HI+3ZDOBdw/Zihaq6sLWdO6xtpap6N0BV/aaq9gCeDJwGnDJAbgt6n2YA1w7LfeWq2rW130T3XvbO3Wus93TEs0+yHHAq8DlgnapaDTiLR+/zeHd2vN+OeZLkxcCHgT2B1Vt+d/XkN5abgPVaTkOG75kkSZIkSZIkSZIkSZIkSdITjsV8JUmSJEmSJEmSJEmSJEmSxlFVjwDfBo5MMjXJ0km2bcU8TwFekWTHJMsAHwQeAC5swy8B9mpjXg5sPx/mXNBWBu4G5iZ5BvDuoYYkWyfZpq31HrqCw48MGHsucFeS9XhsIeBf0xURPSLJk5Isn+RFre0W4KlJlh0j9knAy1q+J/Q8Px7YPcnObR+XT7JDkqcOkPc3gU8neXo6z0qyJl3x1k2T7JVkSpLX0xU5/tEAseeXW4CNer7fRnc2vc++DnwkyZYASVZN8rrW9iO6teyTZJn2t3WSzZMsm2TvJKtW1YN092OQc1/Q+/RrYE6SA5Os0M75mUm2bu2n0K179Xbu7x02/hJGf0+/BezX3vGlkqzX3otlgeXo9vmhJLvQ3b8htwBrJll1lJzH++2YVysDD7X8piT5OLBKn2MvamPf1+7Bq4Hnz6e8JEmSJEmSJEmSJEmSJEmSFlsW85UkSZIkSZIkSZIkSZIkSerPAcAfgd8AdwD/BixVVVcCbwK+AtwO7A7sXlV/beP+uT2bDewNnDavc87jOvqddy9gDvAN4OSetlXaszuB6cAs4N8HiP1J4LnAXcCZwA+GGqrqYbq92gS4HrgBeH1r/gVwOXBzkttHClxVN9EVIX1hb85VNQPYAziIrrDpDLoiwoPs5ZF0xVfPpitk+y1ghaqaBexGV4h1FvBhYLeqGjHHBewzwMFJZic5oKruBQ4DftmevaCqfkh3j05Kcjfwv8AuAFU1h64Y7RuAmcDNre9QAel9gOvauP3p7nNfFvQ+tbuzG7AVcC3du/hNYKiQ7ifp7uu1dGd43LAQo76nVfVrYD/gC3T39lxgWtuv99Hdizvp3pnTe8b9GTgRuKbt/9RhOY/32zGvfgr8BPg/urXfT3f3x9VyeDWwL91vz+vpeVclSZIkSZIkSZIkSZIkSZKeqFJVk52DJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmLtCmTnYAkSZIkSZIkSZIkSZIkSdJkSrI6cGAfXU+oqssWdD7zIsnrgOeN0+3WqjpyAcx9ILD6ON0urqrT5vfc8yLJEX10+3FVnbvAk1kMJdkJ2Gmcbg9W1ccmGP95wOv66HpoVc2dyByLmwW955IkSZIkSZIkSZIkSZIkSRpdqmqyc5AkSZIkSZIkSZIkSZIkSZo0STYEru2j6xur6qQFnM48SXIs8JZxul1ZVc9YAHNfB0wbp9vRVbX//J57XiTp5z/TfqSq+in6+4ST5BDgE+N0e6Cqlp9g/H2BY/ro+pSqunkicyxuFvSeS5IkSZIkSZIkSZIkSZIkaXQW85UkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZIkaRxLTXYCkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiQt6izmK0mSJEmSJEmSJEmSJEmSJJIclOSbC3G+FyX5S5K5SV61sOYdVJK9k5w92XksrpJUkk0mO48lxcJ6T5Ock+TtC3oeSZIkSZIkSZIkSZIkSZKkxU2qarJzkCRJkiRJkiRJkiRJkiRJ0kKUZAfg+Kp66iTm8P8Bp1fVlyYrh+GSbAhcCyxTVQ9NcjoDWVRzT1LA06vqqsnOZXEzme9pknPa3AutwLckSZIkSZIkSZIkSZIkSdLiYKnJTkCSJEmSJEmSJEmSJEmSJGlJl2TKQpwrSRaH/yM6Dbh8IgMnup/z+xwW5rlKkiRJkiRJkiRJkiRJkiRJmnyLw3/UliRJkiRJkiRJkiRJkiRJWuQkOTbJ15P8LMmcJOcmmdbTXkn+KclfgL+0Z7sluSTJ7CQXJnlWT//rknwkyZ+S3JnkmCTLt7bVk/woyW2t7UdJntoz9pwkhyX5JXAvsFGS/ZJc0XK7Jsm7Wt8nAT8GpiaZ2/6mJjkkyfE9MV+Z5PKW6zlJNh+W6wFJLktyV5KTh3Ltc++uBjYCzmjzL9dyOD3JHUmuSvKOnv6HJPl+kuOT3A3s2+c8OyS5IcmBSW4GjkmyVJJ/TXJ1kllJTkmyRhtyXvt3dstr2yT7JrmgJ+ZEzvVDba/uSfKtJOsk+XE7m58nWb2n/wtajNlJLk2yQ0/bOUk+neSXbezZSdYaLfd+9miEPds4yS/a3tye5LtJVhu2nlHPvq31piQzk7x1AvMvleTgJNOT3JrkO0lWbW0btv1/S5LrW34fncAc72h37I5256b2tG2Z7p2+I8ktSQ5qz5dOclC7N3OS/C7J+j05TemJcU6St7fP+7bzOqrt15+T7NjTd79MwnuacX5TJEmSJEmSJEmSJEmSJEmSNDKL+UqSJEmSJEmSJEmSJEmSJE3c3sCngbWAS4DvDmt/FbANsEWS5wDfBt4FrAkcDZyeZLlh8XYGNgY2BQ5uz5cCjgGmARsA9wFHDZtrH+CdwMrAdOBWYDdgFWA/4AtJnltV9wC7ADOraqX2N7M3UJJNgROB9wNrA2fRFd5dtqfbnsDLgacBz6LPArsAVbUxcD2we5v/AeAk4AZgKvBa4PAkf9czbA/g+8BqPH6fx7IusAbd3r0TeC/duWzf5roT+Grr+5L272otr4tGifkqBjvX1wB/T3emu9MVaT2Ibm+XAt4HkGQ94Ezg0JbzAcCpSdbuibUX3Xk+GVi29Rkk9/EE+Azd3mwOrA8cMqzPiGef5OUtn78Hng7sNIH5921/L6Ur+LwSj7/r2wGbATsCH+8tYDuedqc+09bwFLp35aTWtjLwc+AndOvfBPj/2tAPAG8EdqV7p95KVzi7H9sAV9P9TnwC+EEeLSA9We9pP78pkiRJkiRJkiRJkiRJkiRJGsZivpIkSZIkSZIkSZIkSZIkSRN3ZlWd14rRfhTYNsn6Pe2fqao7quo+ukKyR1fVr6rq4ar6L+AB4AU9/Y+qqhlVdQdwGF3xUKpqVlWdWlX3VtWc1rb9sFyOrarLq+qhqnqwqs6sqqurcy5wNvDiPtf1+ra2n1XVg8DngBWAF/b0+XJVzWy5ngFs1Wfsx2l79iLgwKq6v6ouAb4JvLmn20VVdVpVPdL2s1+PAJ+oqgfauP2Bj1bVDe3cDgFem2TKADEHPdevVNUtVXUjcD7wq6r6Q1XdD/wQeE7r9ybgrKo6q63zZ8Bv6QrIDjmmqv6vzX0K87DvI6mqq9q5P1BVtwFH8vi7NtrZ79ny+99WjPaQCaSwN3BkVV1TVXOBjwBvGHY+n6yq+6rqUuBS4NkDxv92Vf2+nf9H6N7bDemK6t5cVZ9v93BOVf2qjXs7cHBVXdneqUuralafc94KfLG9lycDVwKvAJis97TP3xRJkiRJkiRJkiRJkiRJkiQNYzFfSZIkSZIkSZIkSZIkSZKkiZsx9KEVHr0DmDpSOzAN+GCS2UN/wPpj9J8+1JZkxSRHJ5me5G7gPGC1JEuPMpYkuyS5OMkdba5dgbX6XNfUNv/Q2h5p8dfr6XNzz+d7gZX6jD3afHe0oqJDpg+bbwYTc1srmjtkGvDDnjO4AngYWGeAmIOe6y09n+8b4fvQ3k0DXjcs1nbAU3r6z899f5wk6yQ5KcmN7a4dz+PvzWg5TOXxd3hQj7l77fMUHns+87IHw+/2XGAW3V1bH7h6lHFjtY3nxqqqnu+97/akvKd9/qZIkiRJkiRJkiRJkiRJkiRpGIv5SpIkSZIkSZIkSZIkSZIkTdz6Qx+SrASsAczsae8t4DkDOKyqVuv5W7GqThwpHrBBT6wPApsB21TVKsBLhqYdaa4kywGnAp8D1qmq1YCzevr35jWSmXSFZYfipeV24zjjJmomsEaSlXuebTBsvvFyHs3wcTOAXYadw/JVdeMAcwx6rv2aARw3LNaTquqIAXOaF4e3WH/T7tqbeOw9G8tNPP4OD+oxd6/FeIjHFkCeF8Pv9pOANenu2gxgo1HGzQA2HuH5Pe3fFXuerTusz3rtHRqyATBzkt/Tfn5TJEmSJEmSJEmSJEmSJEmSNIzFfCVJkiRJkiRJkiRJkiRJkiZu1yTbJVkW+DRwcVXNGKXvN4D9k2yTzpOSvGJYAdt/SvLUJGsAHwVObs9XBu4DZre2T4yT17LAcsBtwENJdgFe1tN+C7BmklVHGX8K8IokOyZZhq7w5wPAhePMOyFtzy4EPpNk+STPAt4GHL8Apvs6cFiSaQBJ1k6yR2u7DXiE0Qu6jqSfc+3X8cDuSXZOsnTbix2SPLWPsY/LPcmGSSrJhgPksDIwF7gryXrAhwYYewqwb5ItkqzI+Pd0JCcC/5Lkaa1A9uHAyVX10ARijRZ/vyRbtWK6hwO/qqrrgB8BT0ny/iTLJVk5yTZt3DeBTyd5ejvnZyVZs6puoyue+6Z2Zm/l8UV/nwy8L8kySV4HbE5XtHcy39NBf1MkSZIkSZIkSZIkSZIkSZKExXwlSZIkSZIkSZIkSZIkSZLmxQl0RTDvAJ4HvGm0jlX1W+AdwFHAncBVwL4jxDsbuAa4Gji0Pf8isAJwO3Ax8JOxkqqqOcD76Ip93gnsBZze0/5nuqKm1ySZnWTqsPFXtrV8pc25O7B7Vf11rHnn0RuBDYGZwA+BT1TVzxfAPF+i24uzk8yh289tAKrqXuAw4JdtX14wXrA+z7UvrajxHsBBdAVeZ9AV0x33//yOkvv6wHS6YrP9+iTwXOAu4EzgBwPk/2O6u/oLun34xQDzDvk2cBxwHnAtcD/w3gnEGVG7Ux8DTgVuoiu8+4bWNgf4e7r7fjPwF+ClbeiRdO/T2cDdwLfo3knozv9DwCxgSx5fTPdXwNPp3qXDgNdW1axJfk+/yAC/KZIkSZIkSZIkSZIkSZIkSeqkqiY7B0mSJEmSJEmSJEmSJEmSpMVOkmOBG6rq4PkU7zrg7QuogK2egJIcDNxWVUdPdi5PVEn2pXuvt5vsXCRJkiRJkiRJkiRJkiRJkjTvpkx2ApIkSZIkSZIkSZIkSZIkSYuSJKsDB/bRdZUFncviJslOwE7jdHuwqj42j/O8E9honG5XVtUx8zLP4q6qDh36nORAYPVxhlxcVactiFwW9PzeCUmSJEmSJEmSJEmSJEmSJC0MFvOVJEmSJEmSJEmSJEmSJEl6rFXpr5jvA8CfFnAui5vtGH/vHgDmqZgvsBew/Th9fgpYuPVR7wamjdPnaOC0xXR+74QkSZIkSZIkSZIkSZIkSZIWuFTVZOcgSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSdIibanJTkCSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSpEWdxXwlSZIkSZIkSZIkSZIkSZKeYJJsmKSSTFmAc2yW5JIkc5K8b0HNsyRJclCSb052HsMl+YckM5LMTfKcyc5ncZZk7yRnLwJ5+H5KkiRJkiRJkiRJkiRJkiRNQKpqsnOQJEmSJEmSJEmSJEmSJElaYiTZELgWWKaqHprkdABIch3w9qr6efu+IQs4xyTfAu6uqn9ZEPGfyJKcAxxfVQul8G+Sq4EPVNV/L4z5FlVJ9qV7j7brs/+GLGK/BUN8PyVJkiRJkiRJkiRJkiRJkiZmqclOQJIkSZIkSZIkSZIkSZIk6YkmyZTJzmEhmAZcPpGBC3p/niD7Pz9N+CyHJFl6PuXi+c0fY57p/DwvSZIkSZIkSZIkSZIkSZKkJYnFfCVJkiRJkiRJkiRJkiRJ0hNWkhWSfD7J9CR3JbkgyQqt7QVJLkwyO8mlSXboGXdOkk8n+WWSOUnOTrJWaz6v/Ts7ydwk2ybZt/X9QpJZwCFJlkvyuSTXJ7klydd75l4ryY/a3HckOT/JhP7fZ5LjgA2AM1o+H+5p3rvNf3uSj/aMWSrJvya5OsmsJKckWWOAOX8BvBQ4qs25aZJVk3wnyW1tvw8eWtMo+3Nskq8l+XGL8csk6yb5YpI7k/w5yXN65pya5NQW/9ok7+tpOyTJ95Mcn+RuYN8B1rJHkkuS3N324+U9853ezueqJO8YNt8pbb1zklye5G972g9McmNruzLJjj3jjm+fl2/5zmr34DdJ1klyGPDinr09Kp0vJLm15fnHJM8cYI1LtfOY3mJ8p53XcknmAksDlya5eoCYxyb5jyRnJbkHeGmfZ3Ry25ffJ3l2T/t1bd8uA+5JMiVjv6P7Jrmmxbo2yd49bW9NckW7Rz9NMq2nrZLsn+QvLe5X2/5uDnwd2Lbt++zW/xVJ/tD2fUaSQ3q2YbTfggva2P9I8rlh+/bfST7QPo+1X89P8ts27y1JjhzgbEZ6P0c6r83T/dbNbnf4lcPOt+/3U5IkSZIkSZIkSZIkSZIkaUlhMV9JkiRJkiRJkiRJkiRJkvRE9jngecALgTWADwOPJFkPOBM4tD0/ADg1ydo9Y/cC9gOeDCzb+gC8pP27WlWtVFUXte/bANcA6wCHAUcAmwJbAZsA6wEfb30/CNwArN36HwTURBZYVfsA1wO7t3w+29O8HbAZsCPw8VawFOC9wKuA7YGpwJ3AVweY8++A84H3tDn/D/gKsCqwUYv7Zrr9GzJ8fwD2BA4G1gIeAC4Cft++fx84ErpitMAZwKV0+7gj8P4kO/fE36ONWQ34bj/rSPJ84DvAh9q4lwDXteaT6M5oKvBa4PAkf9cz/JWtz2rA6cBRLeZmwHuAratqZWDnnpi93kK3X+sDawL7A/dV1Ud57N6+B3hZy23TNmZPYFY/a2z2bX8vpTuflYCjquqBqlqp9Xl2VW08QEzo3pHDgJWBC+nvjL5H986dAJyWZJme9jcCr6Db03UY5R1N8iTgy8AubY9fCFwCXXFmuvfp1XTv1/nAicPy3g3YGngW3V7uXFVX0J3BRW3fV2t976G7y6u13N6d5FWtbbTfgiEnAq9Pkpbb6nRneVIfd/pLwJeqahVgY+AU+jTK+wmPPa9ftfnPpvuNey/w3XZ/h/T1fkqSJEmSJEmSJEmSJEmSJC1JLOYrSZIkSZIkSZIkSZIkSZKekFqxzLcC/1xVN1bVw1V1YVU9ALwJOKuqzqqqR6rqZ8BvgV17QhxTVf9XVffRFdLcapwpZ1bVV6rqIeB+4J3Av1TVHVU1BzgceEPr+yDwFGBaVT1YVedX1YSK+Y7jk1V1X1VdSlc09Nnt+f7AR6vqhrYfhwCvTTJlIpMkWZpubR+pqjlVdR3weWCfnm7/b3/angL8sKp+V1X3Az8E7q+q71TVw8DJwHNav62BtavqU1X116q6BvgGj+4ndEVYT2vneR/9eRvw7ar6WRt3Y1X9Ocn6wIuAA6vq/qq6BPgmXVHXIRe0+/MwcByP7u3DwHLAFkmWqarrqurqEeZ+kK6I7ybtbv6uqu4eJc8H6QqwPgNIVV1RVTf1uUaAvYEjq+qaqpoLfAR4w0TPu8d/V9Uvq+oR4G8Y/4x+V1Xfr6oH6QrBLg+8oKf9y1U1o53feO/oI8Azk6xQVTdV1eXt+f7AZ9oePUT33m2VZFrPPEdU1eyquh74H8Z4t6vqnKr6Y8vhMroCvdv3uT/n0xXpfnH7/lq6ezqT8e/0g8AmSdaqqrlVdXGfc46l97y2oivqfESb/xfAj+gKKg/p9/2UJEmSJEmSJEmSJEmSJElaYljMV5IkSZIkSZIkSZIkSZIkPVGtRVcsdKRCqtOA1yWZPfQHbEdXYHfIzT2f76UrfDmWGT2f1wZWBH7XE/8n7TnAvwNXAWcnuSbJv/a3pIGNtoZpwA97cruCrgjtOhOcZy1gGWB6z7PpwHo932fweLf0fL5vhO+9+U4ddl4HDct3pPjjWZ+R78dUYKgI85Dh6xm+t8snmVJVVwHvpyuQfGuSk5JMHWGO44CfAiclmZnks0mWGSnJVmj1KOCrLeZ/JlmlrxU+up7hZzOFiZ/3kN49H+iMWkHZG1puo8Ub8R2tqnuA19MV7r0pyZlJntEz7ks9Y+4AwthnN+q7nWSbJP+T5LYkd7U51xqtf69WoPskHi2Quxfw3Z48x9qvtwGbAn9O8psku/Uz5zh693cqMKOdw5Dhd7zf91OSJEmSJEmSJEmSJEmSJGmJYTFfSZIkSZIkSZIkSZIkSZL0RHU7cD+w8QhtM4Djqmq1nr8nVdURfcStPp7fTlfscsue+KtW1UoAVTWnqj5YVRsBrwQ+kGTHvlfWf06jmQHsMmz9y1fVjROc/3bgQboCpUM2AHrjDZpjrxnAtcPyXbmqdp3H+DMY+X7MBNZIsnLPs+HrGVVVnVBV29HtRwH/NkKfB6vqk1W1BfBCYDfgzUPNI/T/clU9D9iCrsjrh/rJpWc9w8/mIR5bnHUievPs54zWH/qQZCngqS230eKN+o5W1U+r6u/pCnD/GfhGz7h3DRu3QlVdOOB6hpwAnA6sX1WrAl+nKw48Wv/hTgRem2QasA1wak+eo+5XVf2lqt4IPJnu/nw/yZP6mG8svfnOBNZv5zCk7zsuSZIkSZIkSZIkSZIkSZK0pLKYryRJkiRJkiRJkiRJkiRJekKqqkeAbwNHJpmaZOkk2yZZDjge2D3Jzu358kl2SPLUPkLfBjwCbDTO3N8AvpDkyQBJ1kuyc/u8W5JNkgS4C3i4xZyoW8bKZwRfBw5rBUZJsnaSPSY6eVU9DJzSYq7c4n6Abp/nh18Dc5IcmGSFdmbPTLL1PMb9FrBfkh2TLNXO6BlVNQO4EPhMuxvPAt5GH+tJslmSv2v37H66os6PO9skL03yN0mWBu6mK4Y81O8x55lk6yTbJFkGuKfFHeS+nAj8S5KnJVkJOBw4uaoeGiDGePo5o+cleXWSKcD7gQeAi0eJN+o7mmSdJHu04rYPAHN5dD++DnwkyZYASVZN8ro+13AL8NQky/Y8Wxm4o6ruT/J8YK+etn5+C/5AV+z6m8BPq2p2axpzv5K8Kcna7bdkaMy8/EYM9yvgXuDDSZZJsgOwO3DSfJxDkiRJkiRJkiRJkiRJkiRpsWMxX0mSJEmSJEmSJEmSJEmS9ER2APBH4DfAHcC/AUu1Yq17AAfRFeScAXyIPv7vZVXdCxwG/DLJ7CQvGKXrgcBVwMVJ7gZ+DmzW2p7evs8FLgK+VlX/A5Dkx0kOGnCdnwEObvkc0Ef/LwGnA2cnmUNXUHWbAecc7r10hWavAS4ATqArpjzPWrHg3YCtgGt5tDjqqvMY99fAfsAX6IoqnwtMa81vBDYEZgI/BD5RVT/vI+xywBEtx5uBJwMfGaHfusD36Qr5XtHmPq61fQl4bZI7k3wZWIWuOPSdwHRgFvDvAyz12y32eXT7dz/dec03fZ7RfwOvp1vHPsCrq+rBUeKN9Y4uRVcseibde7098O427od07/lJ7b37X2CXPpfxC+By4OYkt7dn/wh8qr0nH6crWj2UY7+/BScAO7V/h8aOt18vBy5PMpfuPryhqu7rcx3jqqq/0hXv3aXN/TXgzVX15/k1hyRJkiRJkiRJkiRJkiRJ0uIoVTXZOUiSJEmSJEmSJEmSJEmSJEl6AktyCLBJVb1psnORJEmSJEmSJEmSJEmSJEmSRjNlshOQJEmSJEmSJEmSJEmSJEma35KsDhzYR9cTquqyBZ3PgpDkecDr+uh6aFXNXVznXBCSvBPYaJxuV1bVMQsjnwUlyYHA6uN0u7iqTusz3hL/Xi3OlpT3U5IkSZIkSZIkSZIkSZIkaVFmMV9JkiRJkiRJkiRJkiRJkrQkWpX+io5eAiyuRUf/hv7W+EVgfhXunIw5F4S9gO3H6fNTYLEu5gu8G5g2Tp+jgdP6jPdEeK8WZ0vK+ylJkiRJkiRJkiRJkiRJkrTISlVNdg6SJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJEmSJC3SlprsBCRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJWtRZzFeSJEmSJEmSJEmSJEmSJD2hJLkuyU6TOH8l2WSy5tfIklyeZIeFON+hSW5PcvN8jHlOkrfPr3gLwmS8f0k2SDI3ydILc96e+fdNcsFkzC1JkiRJkiRJkiRJkiRJkqT5y2K+kiRJkiRJkiRJkiRJkiRpiZXk2CSHTnYeWrSMdC+qasuqOmchzb8B8EFgi6pad4IxDkly/PzNbMkwvGBwVV1fVStV1cOTmddkmOx7YvFySZIkSZIkSZIkSZIkSZK0pLGYryRJkiRJkiRJkiRJkiRJkiZNkimLcrwFZANgVlXdOpHBC3qN6fj/jBdzi8m7IEmSJEmSJEmSJEmSJEmStFjxP9lKkiRJkiRJkiRJkiRJkqQlUpJ3AnsDH04yN8kZPc1bJbksyV1JTk6yfM+43ZJckmR2kguTPGvAeddMckaSu5P8JsmhSS4Ype8rkvyh9Z2R5JCetuWTHJ9kVsvlN0nWaW1PS3JekjlJfp7kq0mO7xn7yiSXt3HnJNm8Pd+vdx+S/CXJ93q+z0iy1QBrXSPJMUlmJrkzyWk9be9IclWSO5KcnmRqT1sl+ackfwH+0vPsfUmuSXJ7kn8fKiibZKkkByeZnuTWJN9Jsmpr27CNfVuS64FftOffS3JzO+PzkmzZno94L5Jcl2Sn9nm5JF9s65rZPi/X2nZIckOSD7Zcbkqy3wB7thPwM2Bqm//YoZjD+vXmc0iS77f7cDewP3AQ8PoW49KeodOS/LLdjbOTrNUT8wXtTs9OcmmSHXrazklyWJJfAvcCG/W5nqWS/GuSq9tdPSXJGj3t+7Rzm5Xko8PGHpvk0J7vj9mHJOsn+UGS29r4o9rzjZP8oj27Pcl3k6zW2o6jK5Z8RtubD/fckSmtz9R2J+9od/QdPXMe0tbwnbaHlyf52372oifGmi3+3Ul+DWw8rP0ZSX7W5r8yyZ7D9uTrrX1OknOTTOtp/1K69/TuJL9L8uJhuY97T9pZH9ruwtx0v1drtn0c+t3acIB8v5rkzJbvr5Js3NrOa90ubfO8vj2fnWS7QfZUkiRJkiRJkiRJkiRJkiRpUWExX0mSJEmSJEmSJEmSJEmStESqqv8Evgt8tqpWqqrde5r3BF4OPA14FrAvQJLnAN8G3gWsCRwNnJ5WyLVPXwXuAdYF3tL+RnMP8GZgNeAVwLuTvKq1vQVYFVi/5bI/cF9rOwH4dXt+CLDPUMAkmwInAu8H1gbOoitsuixwLvDiVoB1KrAssG0btxGwEnDZAGs9DlgR2BJ4MvCFFuvvgM/Q7fNTgOnAScPGvgrYBtii59k/AH8LPBfYA3hre75v+3spXZHZlYCjhsXbHtgc2Ll9/zHw9JbX7+nuwnj3YshHgRcAWwHPBp4PHNzTvi7d2awHvA34apLVR4jzOFX1c2AXYGabf99+xtHtx/fp7sq3gMOBk1uMZ/f02wvYj27dywIHACRZDzgTOBRYoz0/NcnaPWP3Ad4JrEx3Zv14L91Zbg9MBe6kewdIsgXwHy3uVLr7+tR+giZZGvhRy2NDur0eukOhu19T6c58fbr3gKraB7ge2L3tzWdHCH8ScEMb/1rg8HZnh7yy9VkNOJ3H37XxfBW4n+7uv5VH7zFJnkRXzPkEujN6A/C1tldD9gY+DawFXEK7u81v6O7lGi3G99JTjJz+78kb6M5lPbpiwxcBx7S4VwCfGCDfNwCfBFYHrgIOA6iql7T2Z7f5T27PV6uqEQucS5IkSZIkSZIkSZIkSZIkLeos5itJkiRJkiRJkiRJkiRJkp6IvlxVM6vqDuAMuuKY0BUyPbqqflVVD1fVfwEP0BV2HVcrQPoa4BNVdW9V/Qn4r9H6V9U5VfXHqnqkqi6jK8K7fWt+kK746SYtl99V1d1JNgC2Bj5eVX9tRTFP7wn7euDMqvpZVT0IfA5YAXhhVV0DzGnrfQnwU2Bmkme0ec+vqkf6XOtT6IrS7l9Vd1bVg1V1bmveG/h2Vf2+qh4APgJsm2TDnhCfqao7quq+nmf/1p5dD3wReGNPvCOr6pqqmtvivSHJlJ6xh1TVPUPxqurbVTWnzX8I8Owkq/aztjbfp6rq1qq6ja5Q6T497Q+29ger6ixgLrBZn7En6qKqOq3dlfvG6HdMVf1f63MKj97tNwFnVdVZLcbPgN8Cu/aMPbaqLq+qh9rd6cf+wEer6oaevX5tO5vXAj+qqvNa28eAvu4XXQHlqcCH2rneP1QAtqquavf7gXY+R/LoezOmJOsDLwIObDEvAb5JV1R7yAVtnx6mK1j97MdHGjX+0G/Ax1ve/8tjfwN2A66rqmPaPv8BOBV4XU+fM3v27KN07876be3HV9WsNvbzwHI89u4Nck+urqq76ApfX11VP6+qh4DvAc8ZIN8fVtWv29jv8uidkyRJkiRJkiRJkiRJkiRJWuJYzFeSJEmSJEmSJEmSJEmSJD0R3dzz+V5gpfZ5GvDBJLOH/oD16YqK9mNtYAowo+fZjFH6kmSbJP+T5LYkd9EVRl2rNR9HV2z3pCQzk3w2yTItlzuq6t5R5pgKTB/60orzzgDWa4/OBXagK+Z7LnAOXSHU7dv3fq3f8rhzhLbhOcwFZvXkMDznkZ5N59F9f0y89nkKsM5IY5MsneSIJFcnuRu4rjWtRX9Gmq/3DsxqhUuH9N6hBWXUezTMWHf7dcPu9nbAUyYwR69pwA97Yl4BPEx3NlN7Y1bVPXT3oB/rA9OH7TMASdZJclKSG9v5Hs9gZ3tHVc3peTadx97N4Xu4/LDC0WMZ6Teg9y5NA7YZdg57A+v29Onds7nAHS1vkhyQ5Iokd7Wxq/LYtfd7hrf0fL5vhO+992a8fEe7c5IkSZIkSZIkSZIkSZIkSUsci/lKkiRJkiRJkiRJkiRJkqQlWQ3YfwZwWFWt1vO3YlWd2Of424CHgKf2PFt/jP4nAKcD61fVqsDXgQBU1YNV9cmq2gJ4IbAb8GbgJmCNJCuOMsdMugKcACRJa7+xPRoq5vvi9vlcJlbMd0bLY7UR2obn8CRgzZ4cYOSz6V3HBi3O4+K1tod4bAHS3nh7AXsAO9EVO91wKJUx5h41/2G5LAj3AP/vPJMsTVcUttfwnCdyt48bdrefVFVHzEPMobi7DIu7fFXdSHdX/9+Ztju7Zs/Yx6ybxxe03WCUIrqHt1z/pqpWAd7Eo2c73jpm0t3blXuebcBj7+a8GPoNGH6Xh8wAzh22XytV1bt7+vTu2UrAGsDMJC8GPgzsCaxeVasBdzH22idypr36yVeSJEmSJEmSJEmSJEmSJOkJw2K+kiRJkiRJkiRJkiRJkiRpSXYLsNEA/b8B7J9km3SelOQVwwp/jqqqHgZ+ABySZMUkz6ArwDualYE7qur+JM+nK0ILQJKXJvmbVtj1buBB4JGqmg78ts2xbJJtgd17Yp4CvCLJjkmWAT4IPABc2NrPBV4KrFBVNwDnAy+nK7L6h37W2dZ6E/Bj4GtJVk+yTJKXtOYTgf2SbJVkObriq7+qquvGCfuhFmt94J+Bk3vi/UuSp7XipocDJ1fVQ6PEWbmteRZdsdjDh7WPdy9OBA5OsnaStYCPA8ePk/u8+D9g+XbXlgEOBpYbZ8wtwIZJ+v3/wMcDuyfZOcnSSZZPskOSp447cmxfBw5LMg2g7dkere37wG5JtkuyLPApHvv/ly8Bdk2yRpJ1gff3tP2arhjwEe09XD7Ji1rbysBc4K4k6wEfGpbTqOdbVTPo3oXPtJjPAt7GfDrfEX4DtgDe0tPlR8CmSfZp78wySbZOsnlPn1179uzTwMUt75XpCgXfBkxJ8nFglXFSGvSeDNdPvuPN/5izSFJJdphgPpIkSZIkSZIkSZIkSZIkSZPKYr6SJEmSJEmSJEmSJEmSJGlJ9i1giySzk5w2Xueq+i3wDuAo4E7gKmDfAed8D7AqcDNwHF1h2AdG6fuPwKeSzKErGHtKT9u6dMVQ7wauoCvCe1xr2xvYlq5Y7aF0RW8faGu4EngT8BXgdrpCv7tX1V9b+//RFUI9v32/G7gG+GUrRDqIfeiKDP8ZuJVWjLWqfg58DDiVriDrxsAb+oj338Dv6Iq8nkl3fgDfbms/D7gWuB947xhxvgNMB24E/gRcPKx9vHtxKF3B5MuAPwK/b88WiKq6i+4ufLPlfA9wwzjDvtf+nZXk933MMQPYAziIrhjsDLoiuPP6/4m/BJwOnN3u8cXANm3Oy4F/Ak6guwd38th1HQdcClwHnM2jxZuHiuLuDmwCXN/Gvb41fxJ4LnAX3T35wbCcPkNXjHl2kgNGyPmNwIbATOCHwCfanZ1f3gOsRPcbcCxwzFBDVc0BXkb3Psxsff6NxxZvPgH4BHAH8Dy69xngp8BP6Io/T6d7D2aMk8tA92S4PvMdyyHAf7Wz2LMV6p5D915JkiRJkiRJkiRJkiRJkiQtdlJVk52DJEmSJEmSJEmSJEmSJEnSEivJvwHrVtVbFuAcJwN/rqpPLKg5FrQkBTy9qq6a7FykyZLkWOCGqjp4snNZEJK8Cdiyqj4y2blIkiRJkiRJkiRJkiRJkiRNxJTJTkCSJEmSJEmSJEmSJEmSJGlQSVYHDuyj6wlVddnCnA/4K7As8Edga+BtwNvnNYdh+WwN3AFcC7wM2AM4Yj7F3gh4Zx9dv1pVM+bHnEuKJAcCq4/T7eKqOm0hpDNPkvRzn35cVecu8GQWAQv7N2dJVVXHT3YOkiRJkiRJkiRJkiRJkiRJ88JivpIkSZIkSZIkSZIkSZIkaXG0Kv0V1rwEmB+FNQeZ72rgRGAqcAvweeC/50MOvdYFfgCsCdwAvLuq/jCfYm9Af2s9DbCY72O9G5g2Tp+j6fZuUdfPHZgNPCGK+bLwf3MkSZIkSZIkSZIkSZIkSZK0CEpVTXYOkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiRJkiQt0paa7AQkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSZIkSVrUWcxXkiRJkiRJkiRJkiRJkiRpDEmuS7LTZOcxqCSHJDl+svOYV0kqySYLec4XJ7lyIcyz2J5RksuT7DDBsQtlfwcxGfesX0kOSvLNhTznIndGkiRJkiRJkiRJkiRJkiRJiwKL+UqSJEmSJEmSJEmSJEmSJDVJjk1y6GTnsbhIsmGS6yY7j3k1vJBrVZ1fVZtNZk6LuqrasqrO6afvora/Sc5J8vbJmn9QVXV4VS3UfIef0eJa1FySJEmSJEmSJEmSJEmSJGl+s5ivJEmSJEmSJEmSJEmSJEnSEibJlEUxlqRFR5KlJzsHSZIkSZIkSZIkSZIkSZKkxY3FfCVJkiRJkiRJkiRJkiRJkoAk7wT2Bj6cZG6SM3qat0pyWZK7kpycZPmecbsluSTJ7CQXJnnWAHM+rY1bqn3/RpJbe9qPS/L+9nlqktOT3JHkqiTv6Ol3SJLvJzk+yd3AvsPmWSbJiUlOTbJsi3VqktuSXJvkfWPFSvL8JL9NcneSW5Ic2e8aR1jzckk+l+T6FuvrSVboaf9QkpuSzEzy1mFjz0ny9p7v+ya5oOf7lkl+1vboliQHtefPT3JR2+ubkhyVZNnWdl4bfmk799cn2SHJDT1xN29zz05yeZJX9rQdm+SrSc5MMifJr5Js3NP+pSQz2t79LsmLJ7p3Ld6/Jrm6zfWnJP/Q07ZJknPbPb09ycnteZJ8IcmtLY8/Jnlma1s1yXfaXZie5OCh+9ja35Hkip75ntueX5dkp0Vgf3+U5F8H2L/DgBcDR7V8jupp3inJX1oeX02SnnFvbftwZ5KfJpnW75zD96t9PyTJ8e3zhkkqyVvae3F7ko+O0vfHSd4zLPalSV7dPj+j5x24MsmePf2OTfIfSc5Kcg/w0iS7tnOdk+TGJAe0vv/vjJIcB2wAnNH27MPtPN47LI/Leu+jJEmSJEmSJEmSJEmSJEnSkshivpIkSZIkSZIkSZIkSZIkSUBV/SfwXeCzVbVSVe3e07wn8HLgacCzaMVykzwH+DbwLmBN4Gjg9CTL9TnntcDdwHPao5cAc5Ns3r5vD5zbPp8E3ABMBV4LHJ7k73rC7QF8H1itrYOW4wrAacADbR0PAWcAlwLrATsC70+y8xixvgR8qapWATYGTmn5X1dVG/az1h5HAJsCWwGbtBw+3nJ9OXAA8PfA04GdRg7xeElWBn4O/IRujzYB/r/W/DDwL8BawLZ0a/7HtoaXtD7Pbud+8rC4y9Dt19nAk4H3At9NsllPtzcAnwRWB64CDutp+01b6xrACcD30lMMegKupitGu2qb8/gkT2ltn255rg48FfhKe/4yuru1aRu3JzCrtX2lPduI7r69Gdivrf11wCHt2SrAK3vG9Zq0/a2q3arqiNG367Gq6qPA+cB7Wj69hXF3A7ame8f3BHZuOe4BHAS8Gli7jT+x3zkHsB2wGd3+fbznd6DXicAbh74k2QKYBpyZ5EnAz+ju2ZPp9u1rrc+Qvej2b2XgAuBbwLuqamXgmcAvhk9YVfsA1wO7tz37LPBfwJt68nj2/8/en4dbltbl/f/9WcOepzPWXN1MooCI8lXUL2onYjRqokkERQXbCTE/k3wTB6LR2M7ExNk44YATgholDiEmig2iCEQEgoBAd1dT3VV16gx7nofn98caep9T51Ttc/pUVXfzfl1XXX3O3muv51nPeu6164++7lKU5T8+2qUDAAAAAAAAAAAAAAAAAAA8PlDmCwAAAAAAAAAAAAAAAAAAcGM/6Zy75JzbUVQ8+pz49ZdJ+nnn3Nucc1Pn3K8qKs391EOc+02SPsvMTsa//278+5MUFai+28zOSfp/Jb3COTdwzr1L0i8qKllNvNU593rn3Mw5149fqygqt71P0lc756aKykrXnHPf65wbOeful/QqRcWfB51rLOmpZrbqnOs45/76ENeXMjNTtGb/1jm345xrS/rBubFfJOlXnHPvdc51FRXJLuoLJV1xzv1IvEZt59zbJMk59zfOub92zk2ccxcUlS5/1oLn/VRJJUmvjNfrjZL+SHOFqpJ+3zn3dufcRFH58XOSN5xzv+Gc247H/hFJWUWFrUfinPudeC/O4mLcD0n6lPjtsaJi19PxGrxl7vWypI+VZM659zvnLpuZr2jtvz1erwuSfkTSS+LPfZ2icut3uMiHnXMP7jOn27a+x+yVzrmGc+4jkv58bpyXS/qheN0mivbsc8zsjmMe/3ucc33n3LsVlW1/wj7H/P6esb9C0u8554aKMnDBOfcr8b34W0n/TdIL5z7/351zfxnvn4GivfEMM6s45+rOuXcuONc/kPQxZva0+PeXSHqdc250qCsGAAAAAAAAAAAAAAAAAAB4nKHMFwAAAAAAAAAAAAAAAAAA4MauzP3cU1Q+KkXFqd9sZo3kj6Rzkk4f4txvknSXpM+U9GZJ9yoqQv0sSX/hnJvF50vKbxMPSjoz9/vFfc79qZKeraik1M3N+fSeOX+HpBPXOdfXSvoYSR8ws3eY2Rce4vrmrUkqSPqbubH/Z/y6FF3n/NjXFMdexzlFpcXXMLOPMbM/MrMrZtZSVMa6uuB5T0u6GN+H+XnNr/1B+0Nm9i1m9n4za8bXWz3E2Ncws5ea2bvm1u9Zc+f7Nkkm6e1m9ndm9jWSFBfk/rSk/yrpqpn9gplV4s+F2r3O89d24JrumdNtW99jdr2c/8Tcmu8oWuczOl43vM74GfDHeqQA+8WKCo6TeT5vT7a/QtLJuVPszfa/kPT5kh40szeZ2actMtG4CPh1kr7SzLx4Hr++yGcBAAAAAAAAAAAAAAAAAAAezyjzBQAAAAAAAAAAAAAAAAAAeIS78SG7XJT0A8652tyfgnPutw5xjjdJ+gxFhb5vkvQWSf+vojLfN8XHXJK0bGbluc+dl/TwDeb+vyT9kKQ/M7OkrPeipAf2zLnsnPv8g87lnPuQc+7FktYl/SdJv2tmxUNcY2JLUl/SM+fGrjrnktLSy4oKZOevcV5XURlwYm9J6ZMPGPdnJX1A0tOccxVF5cW24JwvSToXF5bOz+vhA45PmdlnKCrYfZGkJedcTVLzEGPvPd8dkl4l6ZskrcTne29yPufcFefc1zvnTkv6Bkk/Y2ZPjd/7SefccyU9Q1Ex87cquh9jRSWw+13bRUlPWWBqt2V9H4Wj5Pwb9mQm75z7q0Oc43p797B+S9KL4+LdnKQ/n5vnm/bMs+Sc+8a5z+7N9jucc1+kKNuvl/TbB4y535r9qqKy4M+W1HPOvfXIVwQAAAAAAAAAAAAAAAAAAPA4QZkvAAAAAAAAAAAAAAAAAADAIzZ0cCHsfl4l6eVm9jyLFM3sC/aU7l6Xc+5Digpuv1JREWcrnse/UFzm65y7KOmvJP2QmeXM7NmSvlbSbyxw/h+W9BpFhb6rkt4uqW1mrzCzvJn5ZvYsM/vkg85hZl9pZmvOuZmkRvzybNFrnJvLTNGa/ZiZrcfnPmNmnxsf8tuS7jazZ5hZQdJ37znFuyT9czMrxCW1Xzv33h9JOmVm/5+ZZc2sbGbPi98rS2pJ6pjZx0r6Ru12vfv+Nkk9Sd9mZqGZ3SXpn0h67QKXXJY0kbQpKTCz/yipssDnDlJUVKq6KUlm9tWSnpW8aWYvNLOz8a/1+NiZmX1yvEdDRaWyA0kz59xU0Zr/QLxed0j6d3pkX/2ipG8xs+fG+/up8TH7XeftWF+Z2b1mds8ixy44n/38nKRvN7NnxmNWzeyFhxzzXZK+LL7G/0fSlxzy8/P+h6IC5u+V9Lo4V1KUgY8xs5fE44Txvf+4/U5iZhkz+wozqzrnxoru4UG5vmbN4vLemaQfkfTrj+J6AAAAAAAAAAAAAAAAAAAAHjco8wUAAAAAAAAAAAAAAAAAAHjEL0l6hpk1zOz1NzrYOfd/JH29pJ9WVJ76YUl3H2HcN0najkt7k99N0jvnjnmxpDslXZL0+5K+2zn3p4uc3Dn3fZJeL+lPJVUlfaGk50h6QNKWotLW6nVO8XmS/s7MOpJ+QtKXOef6i4y9j1coWqe/NrNWPKenx/N8g6Qfl/TG+Jg37vnsj0kaKSoW/VVJvzl3jW1Jn6OoCPaKpA9J+gfx298i6csltRWVCb9uz3nvkfSr8X1/0fwbzrlRfM5/rGitfkbSS51zH1jgWv9E0v+U9EFJDyoq0b143U9ch3PufYqKU9+qaA0+XtJfzh3yyZLeFt+nP5D0b5xz9ysqEH6Voj36oKRtSf85/sy/UlTwe7+ktygqfv7leLzfkfQD8WttRXtoeZ+p3a71laRze9ZgET8h6UvMrG5mP3mjg51zvy/pP0l6bbxn3xvP9zC+S9JTFN2D71G0pkfinBtK+j1JL5g/T5yBfyTpyxQ9J67E885e53QvkXQhvq6XS/qKA477IUnfGd/Db5l7/dcU7cMbFosDAAAAAAAAAAAAAAAAAAA8EZhz7nbPAQAAAAAAAAAAAAAAAAAAAAAOxczOSvpt59yn3+65fLQys5dKeplz7vm3ey4AAAAAAAAAAAAAAAAAAAC3AmW+AAAAAAAAAAAAAAAAAADgCc/MliS9YoFDX+Oce8/jbbzb7aPteo+bmT1X0gsXOPT7nXOdmz2fxyMze7Kkly1w6H91zl18vI75WGJmBUlvlPQzzrlfu93zAQAAAAAAAAAAAAAAAAAAuBWC2z0BAAAAAAAAAAAAAAAAAACAW6Cqxcpm3yXpOMpmb/V4t9tH2/Uet4/XYuv345Io893feS22hq+XdFzFurdjzMcEM/tcSb8n6U8lveY2TwcAAAAAAAAAAAAAAAAAAOCWMefc7Z4DAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAACPad7tngAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAI91lPkCAAAAAAAAAAAAAAAAAADguszsgpm94Cae/9Vm9v036/zXGffvzOyuWz3uY4WZnTezjpn5t3suN8tx7F0zc2b21OOaEwAAAAAAAAAAAAAAAAAAAB6/KPMFAAAAAAAAAAAAAAAAAAA4JmZ2t5m95XbPA9farzDYOfdM59y9t2lKt51z7iPOuZJzbipJZnavmX3d/DEU2d4+ZnZnvP7BbRr/tpRsAwAAAAAAAAAAAAAAAAAAPJZR5gsAAAAAAAAAAAAAAAAAAHALmZl/u+cw73YVhQI4GLkEAAAAAAAAAAAAAAAAAAB4bKLMFwAAAAAAAAAAAAAAAAAAYA8zy5rZfzGzj5jZhpn9nJnl4/f+h5n9yNyxrzWzXzazj5P0c5I+zcw6ZtaI33+1mf1s/LmupH9gZl9gZn9rZi0zu2hm99xgPvea2feb2V/F5/5DM1sxs9+Mz/EOM7szPvZOM3PzZaDx578u/vluM/tLM/sxM9uWdI+ZPcXM3mhm22a2FZ+39ijX8AvN7F1m1ojn/ey59z7RzN5pZm0ze52k3Nx7d5vZW/acy5nZU+Of82b2I2b2oJk1zewtc/fmd8zsSvz6m83smfHrL5P0FZK+LVm/+PULZvaC+Oesmf24mV2K//y4mWXj9+4ys4fM7JvN7KqZXTazrz7Cmny9mb0/vu73mdknxa9/XHyPGmb2d2b2T+c+82oz+xkze0M89780s5Px/Opm9gEz+8S54y+Y2bea2XvMrGtmv2RmJ+LPt83sT81sKT423Stm9gOSPkPST8fj/LSZvTk+7bvj1750gXt72sz+m5ltmtkDZvavD7lGS2b2R/Hn6/HPZ+fev9fMvi9eh7aZ/S8zW517/yXx3tg2s/9wqBv0yDm+Nb7Hl8zsa/a8d71nQ7JPvsOiHF0ws6+Y++yBuZ+7F19rZh+R9EZJyfo34vX/NNud34aZ3W9mnx6/fjHen191yPles6/t4My8wswejtf+783ss4+yxgAAAAAAAAAAAAAAAAAAAI9XlPkCAAAAAAAAAAAAAAAAAABc65WSPkbScyQ9VdIZSf8xfu9rJL3EzP5hXNT5KZL+jXPu/ZJeLumtzrmSc642d74vl/QDksqS3iKpK+mlkmqSvkDSN5rZF99gTl8m6SXxXJ4i6a2SfkXSsqT3S/ruQ1zf8yTdL+lEPC+T9EOSTkv6OEnnJN1ziPPtEpfL/rKkb5C0IunnJf1BXCyakfR6Sb8ez/13JP2LQ5z+v0h6rqRPjz//bZJm8XtvkPQ0SeuS3inpNyXJOfcL8c8/HN+bf7LPef+DpE9VdM8/QdF9/c65909Kqipa/6+V9F+TUtxFmNkLFa3pSyVVJP1TSdtmFkr6Q0n/K573v5L0m2b29LmPvyiey6qkoaJ7/87499+V9KN7hvsXkj5H0R7+J/G6fIekNUX///A1BbvOuf8g6S8kfVO8Rt/knPvM+O1PiF973Q3urRdfy7vjdfpsSf+fmX3uousUz+9XJN0h6bykvqSf3nPMl0v6akXrlZH0LZJkZs+Q9LOKcnI6nt9ZHYKZfV58vs9RtJdesOeQ6z0bpGifrMavf5WkX5i7l4vk/rMUZfBzJSXrX4vX/63x78+T9J74+l4j6bWSPjmez1cqKmQuHWK+1+zr/TITX8c3Sfpk51w5nuOFa1cRAAAAAAAAAAAAAAAAAADgiYsyXwAAAAAAAAAAAAAAAAAAgDlmZpJeJunfOud2nHNtST+oqExXzrkrkr5R0q9K+glJL42PuZ7/7pz7S+fczDk3cM7d65z7v/Hv75H0W4pKPK/nV5xz9znnmorKWe9zzv2pc26iqBD3Ew9xmZeccz/lnJs45/rOuQ875/63c27onNtUVA57o/lcz8sk/bxz7m3Oualz7lcVldB+avwnlPTjzrmxc+53Jb1jkZPGZbFfo6g8+eH43H/lnBtKknPul51z7fj3eyR9gplVF5zzV0j6Xufc1XgNvkdRKWxiHL8/ds79D0kdSU/f5zwH+TpFxajvcJEPO+ceVLQeJUmvdM6NnHNvlPRHkl4899nfd879jXNuIOn3JQ2cc7/mnJtKep2uvfc/5ZzbcM49rKig923Oub+d+/xh9spe17u3nyxpzTn3vfG13C/pVYqzswjn3LZz7r8553pxrn5A1+7FX3HOfdA515f024qKaiXpSyT9kXPuzfEe+C49UvS8qBfF53+vc66ruVLrGz0b5nxXnKU3Sfrj+JxaMPf3OOe68bUd5AHn3K/M3f9zivbm0Dn3vySNJD11wfkeZl9PJWUlPcPMQufcBefcfdeZJwAAAAAAAAAAAAAAAAAAwBNOcLsnAAAAAAAAAAAAAAAAAAAA8BizJqkg6W+iLkxJkkny5475Q0k/JenvnXNvWeCcF+d/MbPnSXqlpGdJyigqyPyd+L2fk/SV8aE/6Jz7wfjnjblT9Pf5vbTAPA6azwlFxcSfIaksyZNUP8T59rpD0leZ2b+aey0j6bQkJ+lh55ybe+/BBc+7Kikn6ZoCUTPzFRW/vlDRPZzNfaa5wLlP75nHg/Frie24ODnR0+HW/Jz2mXc8xkXn3Hzp7IOSzsz9fth7f5x7Za/r3duppNNm1ph7z1dUKLwQMytI+jFJnydpKX65bGZ+XF4rSVfmPjJ/H05rbm8757pmtr3o2HPn+Ju53+f3xCLPhnpcAjz/+dPS9XM/56JubO/9lHNuv3u8yHwX3tfOuQ+b2f+nqOD4mWb2J5L+nXPu0gJzBgAAAAAAAAAAAAAAAAAAeELwbvcEAAAAAAAAAAAAAAAAAAAAHmO2FJVhPtM5V4v/VJ1z8wWXPyDp/ZJOmdmL516fL6jVdV5/jaQ/kHTOOVeV9HOKSjblnHu5c64U//lBHV5SJFqYe+3kDebzg/FrH++cqygqEzYd3UVJPzC3fjXnXME591uSLks6Y3PtopLO75l/Onczm5/7lqSBpKfsM+aXS/oiSS+QVJV0Z3KK+L8H3ZvEJUVFtfNzOs6S0ovaf96XJJ0zs/n/r/e8pIePcexF3WiNpOvf24uSHtjzXtk59/mHmMM3S3q6pOfFe/Ez49cX2Y+XFZUmRx+IioFXDjH2NefQ7r25yLNhycyKez6f7KMDcz/HHfDzUSwy3+u5Znzn3Gucc89XlBUn6T89yjkCAAAAAAAAAAAAAAAAAAA8rlDmCwAAAAAAAAAAAAAAAAAAMMc5N5P0Kkk/ZmbrkmRmZ8zsc+OfP1PSV0t6qaSvkvRTZnYm/viGpLNmlrnBMGVJO865gZl9iqIi2uOa/6aiItivNDPfzL5G+5fI7p1PR1IzvpZvfZTTeJWkl5vZ8yxSNLMvMLOypLdKmkj612YWmtk/l/Qpc599t6RnmtlzzCwn6Z65a5tJ+mVJP2pmp+Pr+zQzy8bXMJS0ragMeG8R8oakJ19nzr8l6TvNbM3MViX9R0m/cfQluMYvSvoWM3tuvCZPNbM7JL1NUk/St8XrcZekfyLptcc49qL2W6O9r13v3r5dUtvMXmFm+fj+PMvMPvkQcygrKqBtmNmypO8+xGd/V9IXmtnz4wx+rw7//0v/tqS7zewZcRlwOv6Nng1zvsfMMmb2GZK+UNLvzF3bYXK/KWmm6+/bAx1ivgfZde/N7Olm9g/jvA0U3afZUeYGAAAAAAAAAAAAAAAAAADweEWZLwAAAAAAAAAAAAAAAAAAwLVeIenDkv7azFqS/lTS082sIunXJH2Tc+5h59xfSPolSb9iZibpjZL+TtIVM9u6zvn/paTvNbO2otLY3z7m+X+9okLebUnPlPRXNzj+eyR9kqSmpD+W9HuPZnDn3P+J5/DTkuqK1vLu+L2RpH8e/74j6Uvnx3POfVBRCeufSvqQpLfsOf23SPq/kt4Rf/4/Kfp/Yn9N0oOKiozfJ+mv93zulyQ9w8waZvb6fab9/ZL+j6T3xOd/Z/zasXDO/Y6kH5D0GkltSa+XtByvxz+R9I8lbUn6GUkvdc594LjGPoSfkPQlZlY3s5+MX7tH0q/G6/aiG9zbqaLy2udIekDR9fyipOoh5vDjkvLxZ/9a0v9c9IPOub+T9P9TtMaX4/k9dIix5Zx7QzyHNyq6tjfuOWTfZ8Pc+1ficS9J+k1JL5+7l4fKvXOup2jP/GW8/p96mGtZcL7XszczWUmvVHRvrkhal/TtR5gTAAAAAAAAAAAAAAAAAADA45Y55273HAAAAAAAAAAAAAAAAAAAAADgcc3M7pL0G865s7d5KgAAAAAAAAAAAAAAAAAAALhJgts9AQAAAAAAAAAAAAAAAAAAgFvFzJYkvWKBQ1/jnHvPzZ7P442ZvUzSk29w2N87537lVsznscDMXiFp6QaH/bVz7vW3YDqPWbd779zu8QEAAAAAAAAAAAAAAAAAAPDEQJkvAAAAAAAAAAAAAAAAAAD4aFLVYmW+75JEme+1vlzSZ93gmD+R9NFUiPqNku64wTE/L+n1N38qj2m3e+/c7vEBAAAAAAAAAAAAAAAAAADwBGDOuds9BwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHtO82z0BAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAe625Jma+Zfb+ZbZnZlVsx3mGZ2V1m9tARP3u3mb3luOd0q92u6zCznzOz77rV4x4HM7vXzL7uNs+hY2ZPXuC4O83MmVlwk+fzuL2fAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAwPXc1GJPSTKz85K+WdIdzrmrN3s8PHaZ2d2Svs459/zkNefcy2/fjB7/nHOl2z2HedxPAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAPFF5t2CM85K2KfIF8NHEImsLHnviGMZb+ByPdrxDXtuymYWPcryqmWUXPPaWraWZZc2s+ijHCs1secFj18zMHuV4t2yfHOYcZlY2s/yjHCtvZuXjnNdxnYPMHc85yNzxnYPM3fDzZE5k7jjPQeZu+HkyJzJ3nOcgczf8PJkTmTvOc5C5G36ezInMHec5yNwNP0/mROaO8xxk7oafJ3Mic8d5DjJ3w8+TOZG54zwHmbvh58mcyNxxnoPM3fDzZE5k7jjPQeZu+HkyJzJ3nOcgczf8PJkTmTvOc5C5G36ezInMHec5yNwNP0/mROaO8xxk7oafJ3Mic8d5DjJ3w8+TOZG54zwHmbvh58mcyNxxnoPM3fDzZE5k7jjPQeZu+HkyJzJ3nOcgczf8PJkTmTvOc5C5G36ezInMHec5yNwNP0/mROaO8xxk7oafJ3MiczfknHvUfyRVJf2apE1JD0r6TkVFwS+Q1Jc0k9SR9OpDnPOCpG+X9D5JdUm/IikXv7cq6Y8kNSTtSPqLeLxvlfTf9pznJyX9RPzzcnyeS/E5Xx+/fpekhyR9s6Srki5L+uobXV/83t2S3jJ37KdLeoekZvzfT59770mS3iypLelPJf1XSb8Rv/fHkv7Vnrm/R9I/exT35WMl/e94jf5e0ovm3luR9AeSWpLeLun7kuuQdKckJymYO/5eSV839/vXS3p/fC3vk/RJ8ev/XtJ9c6//s/j1j5M0kDSN90Ijfv3Vkr5/z3k/HM/5DySdnnvPSXq5pA/F9/6/SrJDrsnvSLoS3583S3rm3Huvjs/5x/H83ybpKXPvf46kD8Sf/WlJb5pfkwXGzkn6DUnb8fzfIelE/N7p+Hp34uv/+rnP+ZK+Y25d/0bSubk1eWr88xdI+tv4nl6UdM/cOa65pwvM91Mk/Z/4fBuSfnTuvedL+qv4Oi5KuvuA+/mFkt4VH/dXkp69J+PfomifNyW9TnHG4/e/KP5sK772z5vL4y8pyunDkr5fkh+/dz4e6/xxPNuOmLsnS/qe+PpeMff6p86t2bsl3TX33s8qysu3Sjp5iLEKkl4i6Y2S3rcnrz+kKNstSf9d0vLc++34tS+WFD6aa5vbWy9T9Gy9LOlb5j7zpfH++RFJzzrEWMl3yGvi+Z6JX79H0u/G+6Ut6Z2SPmHuc2+Ir/vlkmqHGK8m6Rvjz75hzz496LvoTLy+vxnP1TvEeM+K12RD0pfGr92l6LvoOyRtxWN/xdxnXhG/9j2SnnSIsUJJ/0zRM6Y99/qrJf2cou+JtqJn2h1z779P0p9J+kpJhUOMdzLey++T9LNzrztJ/1rS/fH1/Wc98l36afH6/rykTz1k5j41/lxd0qfFr90t6S8VPaubip7dn03myJzIHJkjc2SOzJE5MkfmyNwFkTmJzJE5MieROTJH5sgcmSNzZI7MkTkyR+YkMkfmyJxE5sgcmbtHZI7MkTkyR+bIHJkjc2SOzJE5MkfmvlhkjsyROTJH5sgcmSNzZI7MkTkyR+bIHJkjc2SOzJE5icyROTInkTkyR+ZqInNkjsyROTJH5sgcmVt03MN+4IDJ/1q8WcvxZvugpK+dX7AjnPOCpPdKOqeohPcvFZeExsH5uXjRQkmfIckknZLUTTaVpEBROe9z49//ON6US/HnPmtujhNJ3xu//vmSepKWFri+u/VICe5yvAFfEo/94vj3lfj9t0r6L5IyispQW3qkzPdFkt42d/2foKj0NXPEe1JUVLL61fFcPjHesM+I33+tpN+Oj3uWolLUhcp8Jb0wPv6T43V/arIp4/dOK3o4fGl8P07tXas9Gzy5r/8wnuMnScpK+ilJb5471ikqca4pKm3dVFzweoh1+Zr4PmYl/bikd+2Zy7aiEttA0UPjtfF7q4oC+CXxHvm38Z45TJnvN0j6Q0VfCr6k50qqxO+9WdLPKCr8fU58bf8wfu9bJf1fSU+P1/sT5vaU0yNlvndJ+vh47Z+t6MH1xQfd0wXm+1ZJL4l/Lin+opV0R7wWL47XYkXSc/a5n5+oKH/Pi6/3qxTlOjuX8bfH+2VZUTn0y+P3PkXRF/HnxNdzRtLHxu/9vqIv8KKk9fgc33Acz7Kj/onv6Usl/Xm8h35W0vPm3j8Tv/758fV8Tvz7Wvx+8mX66/F1/4GiL4R9v/wV/QXoVYqeL/9b0ldIyu/J68OKsl2U9N8UP2vi92uKvoDfGu+TH5X08Ue8tmRv/VY81scr2r8vmDvmWYr+AndJUYn1v1T8fN1nvCcrehY/qKjo+ZsVl17H798jaaxHsvgtkh5I1ip+7YvjfdJU9BeUz9E+X/7xuv+jeO7N+DNfNL/uus53Ufz+yXgO/zee8/dKevIB17YUX/s74rX4z9pdKH6XoufKjyp6Rn2Womfo0+eO+dT4HmzH9+QlOuDLP74XP6ooh29V9Ayq7XnmtSV9ZjzeT2h3OX1B0V8u/reivfYLiv/SvM9YoaR/rugZ11S0lz97ft3jffLn8TqeV/RdOl8S/6T4/t6v6HnwbYq/P/YZ71T8/vsVlX1/t+b+0qXo+2ai6FkdKvo+air+i7bIHJkjcxKZI3NkjsyROYnMkTkyd6fIHJkjc/eKzJE5MkfmyByZI3NkjsyROTJH5sgcmSNzZI7MkTkyR+bIHJkjc2SOzJE5MkfmyByZI3NkjsyROTJH5sgcmSNzZI7MkTkyR+bIHJkjc2SOzJE5MkfmyByZI3NkjsyROTJH5sgcmSNzZO6AzO07/qIHHniCqKRzpLgkNn7tGyTdO7dgDx3hvBcUF3vGv3++pPvin79XUbnuU/f53BskfX388xcqbtBW9ACY7bep4zn2tbu89mp8I290fXfrkRLcl0h6+55zvzU+5ny8cQpz7/2GHinzzcU38Gnx7/9F0s88ivvypZL+Ys9rP6/owecrCujHzr33g1q8zPdPJP2bBefxLklftHet9mzw749//iVJPzz3Xime553x707S8+fe/21J//5RrFEtPmd1bi6/uGfPfSD++aWS/nruPVPU6v11hxjvayT9laRn73n9nKSppPLcaz8k6dXxz3+frOE+53TaJwfxez8u6ccOuqcLzPfNilrJV/e8/u2Sfv+Az8zfz5+V9H173v97PVKifUHSV86998OSfm5ur/7YPuc/IWmo3V+mL5b050fdB4/2j6RflLQj6X8oKuXO7nPMKyT9+p7X/kTSV+1zbDneK29W9Bz6vrn3XqToXxr4e0WN8mcPmNO9kl459/szFD3H/H2Ofbqi/F+U9H8Ul0gf4tqSvTX/PPlhSb+0z7G+pC9QlN2GolLxpND6E+J5X5X0k5I+6YBru0e7s+gp+hcKPmOfY1cV/YsQ75T0EUnfNPfeN8WvvTM+ZvWA8S7ogO+ifY59bjz3q/G1fEL8eiW+1kZ87Z9/wL24S9H3RHHutd+W9F37HJuN78n/iO/R/LPrH8b38mJ8bz/mgPm+WnFhefx7SdGz6Nw+x56L99zfx3vwRXPvfV98zW9WXJh+wHhOcwXsiv6y9Wf7HGeK/nL1y4q+F/9I0vn4vfPx7/X4/c+UZPuc425Ff4mzudferrignMyROTJH5sgcmSNzZI7MkTmROYnMkTkyJ5E5Mkfm7haZk8hcQ2SOzJE5MkfmyByZI3NkjsyROTJH5sgcmSNzZI7MkTkyR+bIHJkjc2SOzJE5MkfmyByZI3NkjsyROTJH5iQyR+bInETmyByZI3NkjsyRuYbIHJkjc2SOzJE5MkfmyByZI3Nkbi5zB/3x9OitKmpnfnDutQclnTmGc1/cc87T8c//WdKHJf0vM7vfzP793HG/qqjdWPF/fz3++ZykHedc/YCxtp1zk7nfe4oW+zDXd3rPcfPHno7H7+13fc65gaTXSfpKM/MUFZT+uo7uDknPM7NG8kdRo/hJSWuSAl27vos6p6gR/Rpm9lIze9fcmM9StIaL2LV+zrmOokbs+bW+Mvdzco8WYma+mb3SzO4zs5aiB4f2zO+g85/W7vvltHv9FvHrigpcX2tml8zsh80s1CN7oz137PweO3C955nZ88zsz81s08yaitrgF137/XytpI+R9AEze4eZfeFh5qNoD37znj14To/kWDp4vQ8a4w5Febw8d86fl7S+2CXdFM9S9OX9Lknvcc4N9znmDkkv3LMWz1dUMr5LvA/eE58vVPQXgMRZRfviXZLerd3rt9fefIfafz88GJ/rvZKeqt1ruci1HTTe6b0HOOemilr1363oS/FZ8bykqFz7YxU9298d//eGYznnZoqKta8ZT9HzI1nLJUX/AkTiSfFr74rH215kPB1wbbEPzc39YxVdkxRd47MUXfO7JL03Xov91J1z3RuNF9+L5NpG8fkT64ru5Xvj+XzkgLGk3WvZiee43/Vdjsd7t6I9eHbuvacrusZ3Kdon7Ws+vc94OvjanKT3xWM9JOmZiv4lCsX/fWb8+rslvT8+fj8P73nvoPHIHJkjc2TuXSJzCTJH5ha9NjJH5vZD5sgcmSNzZI7MkTky9y6RueScZI7MkTkyR+YWGE9kjswtMJ7IHJlb/NrIHJnbD5kjc2SOzJE5MkfmyNy7ROaSc5I5MkfmyByZW2A8kTkyt8B4InNkbvFrI3Nkbj9kjsyROTJH5sgcmSNz7xKZS85J5sgcmSNzZG6B8UTmyNwC44nMkbnFr43Mkbn9kDkyR+bIHJkjc2Tu+pnb13GU+W5JGisqq0ycl/TwMZz73J5zXpKiADrnvtk592RJ/1TSvzOzz46Pe72kZ5vZsyR9oaTfjF+/KGnZzGqHnMNhru/SnuPmj70cj1+Ye+/cnmN/VVHh7mdL6jnn3nrIuc67KOlNzrna3J+Sc+4bJW0qaqTeu76JZHPPz/XknnM/Ze+AZnaHpFcpauFecc7VFG1wiw856MGb2LV+ZlaUtKLj2UuS9OWSvkjSCyRVFbWca25+13NZc+tlZqZr7991OefGzrnvcc49Q9KnK9qfL1V03ctmVp47fH6P7bve+3iNpD9Q1PRdlfRzWuzaDprvh5xzL1b0sPpPkn43vieLzueipB/YswcLzrnfWvCz+41xUdJQUbN7cs6Kc+6Zi13V8XPOfaqkf6CoIPuNZvZ2M/smM1uZO+yipF/fsxZF59wrkwPM7KyZ/Xsze5+iZvlNRW3zL5ob60cVPdz/TNJ/kPSQmf2YmX3iPlPbm++xoueZLPIZZvYqRfvvayX9mqSTzrnXHvLaDhrv0ty1lczsbjN7o6J2/jOSvtQ59yzn3HY81psUfWm9UtG/KPARM3uNmX2emfkHjRWXn5/dM97TzOz7JD0g6ScU/aXmyc65b567tm9WtMfeK+mnJD1gZt9nZk875LX5ZvaPzey3FH2Rf4GkH1L0Lzm8KR5r2zn3LElfGs/1nWb2xnhNSnvGWopzdtB4K/E9eLukNyr6Fxj+QXyvkmt7raJn9q8rureXzOxVZvb8611bPJflPeN9opn9mKK/xH2HpP8t6Uy8F5PxXiTpOYr21+vM7O/M7BVmduZ64+1zbVkz+xIz+0NFf1l7rqJ/yeHJzrn3x2O9X9KT49efK+lDZvYH8eeye8Y6Ez+rDxqPzJE5icyROTJ3o2sjc2TuoGsjc2SOzJE5MkfmyByZI3NkjsyROTJH5sjcI2OROTJ30LWROTJH5sgcmSNzZI7MkTkyR+bIHJkjc2TukbHIHJk76NrIHJkjc2SOzJE5MkfmyByZI3NkjsyROTL3yFhkjswddG1kjsyROTJH5sgcmSNzZI7MkTkyR+bIHJl7ZCwyR+YOujYyR+ZuSuYO5Jx71H8k/Yak35dUVlTG+gFJXxe/d5ekh45wzguKNsXZ+ILfIukH4/e+UFETssWLczle4OSzr1LUavzGPef8Y0WFp0uK2ps/86A5xuO/YIHru1vSW+KfVyQ1FJXGBoo2U0NR8agk/bWkH5aUkfRpkpqSfmPPuB+M5/4f97x+r6R7DrF+ZUUt0y+JrzWU9MmSPi5+/3WKHmIFSc+IN89b5j7/kKR/qWjzfo2ih1NyzS9UVE763PgePDVel2dIGihqN/clfbWi0uDkc58Xr2tmbpxXS/r++OcXKHqoPkdSVtGDYX5OTtJT9/vsgmvyLxU1a1cUtbz/zPw5955vfl8oallvS/rn8b39N/PXtuD4/0DSx8drs6yodfur4/f+QtJPS8pJerakDT2y/7413hNPi9f72YrKknetiaSrkr4q/vlT4t9/I/79zvjY4BDz/UpJa3P3ZiApr+iB15b0ongtViQ9Z5/7+f/E++R58byLih6+5b0Zi3+/Z26+n6IoO5+tqHT8jKSPjd/77/HeqMTvPUXSZ+25zjuP49l2hOeWH1/j78Rr9LXx6+cUNft/bnxMLt5fZ+euvSnpVyR9piRbcLynK/oye0jSn+15XjykKJOFeD6vmXv/fkXPmv+QzOFRXFuy5r8Zj/XMeO/9o7nctyS9QdEzMbvgeKuKcvYuRV9463NrNdYjWfx38V4K4/d/WdFfpH5K0icd4t49N/7MlqRfnnv9gg7+LlpX9P3zt/FcVxccKxuvxRvitfm8uWfORNJ/UfQ98RmKytWTvf+18dr/Tnwv/AXHOxff6w9Jum/u9VfH4z8/Hu/HJP3l3PtvjPfRD0n6mAXHMkmfFe/lpua+t+J98meKvoPPKfoufVn83rMV/UsFfxFfZ3nB8crx8W+JP//s+PW747X8N4q+/14YX2vy7LxHZI7MkTkyR+YkMkfmyNw9InNkjsyROTJH5sgcmSNzZI7MkTkyR+bIHJkjc2SOzJE5MkfmyByZI3NkjsyROTInkTkyR+YkMkfmyByZI3NkjsyROTJH5sgcmSNzZI7MkTkyR+bIHJkjc2SOzJE5MkfmyByZI3NkjsyROTJH5sgcmSNzZG7RNT3sBw6Y/JKiwttNReWd/1GSN7dgRy3z/XZJ71NU6vmrkgrxe/82fr8bX/x37fns8+MN/9V7Xl+Oz7MhqS7p9w6ao3aX+V7v+u7W7sLZ50v6G0UPi7+R9Py5956i6IHVVvRg+wVJv7Rn3O+M5/7kPa/fJ+lzDrmGT1dUYLwpaTveLM+J31uT9Efx5nq7pO/bcx3/WFHTdkPSj0h6k+aKayW9XNLfS+ooatv+xPj1H1D0YN2S9KPzn4s38B8n789t8O/fc9774mP+SHMPPz36Mt+SoiLYtqKi45fOn3OfuezaF4oelB+M7+1P712TBcZ/cbxm3XgP/qTicl1FD68/iq/7Pkkvn/ucH++LB+K5vyNZlz3z/5L4utrxuX5aj67M9zcUfVF0JP2dpC+ee+8zJL0t3j8X9UiJ8N41/Lx4vg1FD+Lf0QJlvvHv/0xRiXFb0oclfW78elXSzyrKflPRw/3L5uZ1QfGXzSGutSPpM+bO0Zl77zskvWHu9zdI+o4Fzrks6Rlzvz8v3jM7ijL5x5LOx+89R1LxMHPeM5Yn6dPmfr9X0ZfC2+N79Iea+/LT3HPpiOOl1za3t16m6C8CVyR929yxT5J0+lGO9wmSSnP75HcVFZK34/v/SXPHformCsOPMFZG0qfM/X5BB38XlRT9iw6P5tpOS3pS/PNd8b7+D4qeoR+R9JK5Y58haflRjjf/nfRqST+n6F8A6Eh6czKX+P1PU/xdd8Sxioq/c+LfnaJ/WeN+Rd9JP6L4L0mK/nWDpx51rPgcT1X0L15I0XfzXyp6DjYVPbv/0dyxZO7645E5R+YWGI/MOTK3wHhkzpG5BcYjc47MLTAWmdtn3Y84HplzZG6BscjcPut+xPHInCNzC4xF5vZZ9yOOR+YcmVtgLDK3z7ofcTwy58jcAmORuX3W/YjjkTlH5hYYi8zts+5HHI/MOTK3wFhkbp91P+J4ZM6RuQXGInP7rPsRxyNzjswtMBaZ22fdjzgemXNkboGxyNw+637E8cicI3MLjEXm9ln3I45H5hyZW2AsMrfPuh9xPDLnyNwCY5G5fdb9iOOROUfmFhiLzO2z7kccj8w5MrfAWGRun3U/4nhkzpG5BcYic/us+xHHI3OOzC0wFpnbZ92POB6Zc2RugbHI3D7rfsTxyJwjcwuMReb2WfcjjkfmHJlbYCwyt8+6H3E8MufI3I3+WHySxxwzu6CoKPVPj/DZ84qav08651rHPbfjYmavk/QB59x3z732UkWN5c+fe+2spN92zn36bZgm8LhhZt8padM59/O3ey63k5ndq6gY+RdvwVh3KiqaDp1zk1sw3j2K/vL3lTd7rHi8Czrid9ERxrpL0X07e7PHisd7taLC8u+8ReM5SU9zzn34Fox1t6L79vwbHXtM490rMndc410QmTuu8cjc8Yx1p8jccY11l8jccY11t8jccY13j8jccY33apG54xrvXpG54xrvgsjccY1H5o5nrDtF5o5rrLtE5o5rrLtF5o5rvHtE5o5rvFeLzB3XePeKzB3XeBdE5o5rPDJ3PGPdKTJ3XGPdJTJ3XGPdLTJ3XOPdIzJ3XOO9WmTuuMa7V2TuuMa7IDJ3XOORueMZ606RueMa6y6RueMa626RueMa7x6RueMa79Uic8c13r0ic8c13gWRueMaj8wdz1h3iswd11h3icwd11h3i8wd13j3iMwd13ivFpk7rvHuFZk7rvEuiMwd13hk7njGulNk7rjGuktk7rjGultk7rjGu0dk7rjGe7XI3HGNd6/I3HGNd0Fk7rjGI3PHM9adInPHNdZdInPHNdbdInPHNd49InPHNd6rReaOa7x7ReaOa7wLInPHNR6ZO56x7hSZO66x7tITKHPBDQZfkvSKBc7zGufce250kJk9V9ILFzjf9y9wzEFjeJL+naTXusdYka+ZfbKkHUVh/EeSvkjSK+feL0j6l5J+Zv5zzrmHJH16fMwLJL3gBkONnXPfdXwzf2wzsxdKeu4NDrvqnPvRJ+L4h2Vmr7zxUXqDc+5NN30yx8w5lz47Hm/3BQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAI9t1y3zlVTVYmW+75J0wzJfSR+/4Pl+fIFjrmFmRUkbkh6U9HlHOcdNdlLS70lakfSQpG90zv2tJJnZ58bv/amk11znHM/XjddwKOmjpsxX0hdI+qobHPP3km5WaevtHv+wFslgQ9Ljrsx3j8fbfQEAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAMBjmDnnbvccAAAAAAAAAAAAAAAAAACPM2b2eZJ+QpIv6Redc6+8zVMCAAAAAAAAAAAAAAAAAAAAAAC4qSjzBQAAAAAAAAAAAAAAAAAcipn5kj4o6XMkPSTpHZJe7Jx7322dGAAAAAAAAAAAAAAAAAAAAAAAwE3k3e4JAAAAAAAAAAAAAAAAAAAedz5F0oedc/c750aSXivpi27znAAAAAAAAAAAAAAAAAAAAAAAAG6q4HZPAAAAAAAAAAAAAAAAAADwuHNG0sW53x+S9LzrfcDM3E2dEQAAACDJOWe3ew4AAAAAAAAAAAAAAAAAgCcuynwBAAAAAAAAAAAAAAAAADeFmb1M0stu9zwAAADw2GIW9e06t/i/92Bm8jxPzrlrPneY8wAAAAAAAAAAAAAAAAAA8GhQ5gsAAAAAAAAAAAAAAAAAOKyHJZ2b+/1s/NouzrlfkPQLkmRmNKwBAABA0tHLd5PPmZnMTLPZjCJfAAAAAAAAAAAAAAAAAMAt5d3uCQAAAAAAAAAAAAAAAAAAHnfeIelpZvYkM8tI+jJJf3Cb5wQAAICPMp7nycxu9zQAAAAAAAAAAAAAAAAAAB9Fgts9AQAAAAAAAAAAAAAAAADA44tzbmJm3yTpTyT5kn7ZOfd3t3laAAAAeAwyMznnjvz+vL3HmVn6ZzabPap5AgAAAAAAAAAAAAAAAACwCFv0f3oDAAAAAAAAAAAAAAAAAOCozIz/WQ0AAABHZmbX/D5fBJyU+Trn7JoPAwAAAAAAAAAAAAAAAABwTLzbPQEAAAAAAAAAAAAAAAAAAAAAAIDDSEp8k5LfvWW/AAAAAAAAAAAAAAAAAADcDMHtngAAAAAAAAAAAAAAAAAAAAAA3AxmJs/z5Pu+giCQmWk8HmsymWg2m93u6QE4wHwxb1Lam/x3/pi9rwEAAAAAAAAAAAAAAAAAcLNR5gsAAAAAAAAAAAAAAAAAAADgCcXMFIahcrmc8vm8stmsMpmMJGk4HKrb7ardbms8Ht/mmQLYz35lvgeh0BcAAAAAAAAAAAAAAAAAcCtR5gsAAAAAAAAAAAAAAAAAALAgM6M0EHiM831f+XxelUpFpVJJQRBoOp3KzBQEgYrFoorFoiSpXq+T6ccoM5PneXLOaTab3e7p4BZLcrlIkS/fzQAAAAAAAAAAAAAAAACAW4kyXwAAAAAAAAAAAAAAAAAAgAVRFgg8tvm+r3K5rFqtplwup8lkokajoeFwqNlspjAMValUVK1WVS6X1Wq1NJlMbve0H7eCIFAQBBqPx5pOp8dyTjNTJpNRsVhULpdTu91Wt9t9whT6Ujy7mButkZnJzNKfWVMAAAAAAAAAAAAAAAAAwK1CmS8AAAAAAAAAAAAAAAAAAACAxz3P81QqlVSr1RQEgdrttlqtlvr9vqbTqZxz8jxP/X5fxWJR+XxeQRBQ5ntEYRiqVCrJ9301m81jKfPNZDIqlUoqFovKZDKSpH6/n5a2Pt6FYaggCNTv92/3VJ4w5ot8KfQFAAAAAAAAAAAAAAAAANwKlPkCAAAAAAAAAAAAAAAAAAAAeNzL5/Oq1WryfV+NRkOtVkuj0WjXMbPZTL1eT8PhUIVC4UjjeJ6nMAw1m800Ho+PY+qPO2amYrGoarWqfr+v2Wz2qM9XKBRUq9VULpc1m83U6XTU6/XU7/efECWtmUxGlUpF4/GYMt9jML8nnHNPiD0CAAAAAAAAAAAAAAAAAHh8oMwXAAAAAAAAAAAAAAAAAAAAeIIwM/m+L0maTCa3eTa3jpmpXC4rDEO1Wi01Go0Dr985p8lkolKpdOg1Skpsy+WyOp3OR22ZbxiGKpfLyuVyqtfrmk6nRz6X53kqFApaWVlRuVzWaDTS1taW2u22JpPJE6KkNQgCVSoVlUolbW5u3u7pPKFQ5AsAAAAAAAAAAAAAAAAAuNUo8wUAAAAAAAAAAAAAAAAAAMDjkud5MjPNZjPKHGNhGCqfz2symXxUlfl6nqdsNqvBYHDdIl9J8n1fQRDI87xDr1E2m1WtVlOxWFSv13u0035cSsp3s9msxuOx+v3+ozpXqVTS6uqq8vm82u22tre31ev1NJvNjnHWt4bv+/I8T5I0m83SkuN8Pq/l5WWNRiMNBoPbOcXHNTO75vf5Z//e3wEAAAAAAAAAAAAAAAAAuBko8wUAAAAAAAAAAAAAAAAAAMDjThAEymazkqTBYJCWZn40MDP5vi8z03Q6TUtPfd9XoVBQsVhUq9W6zbO8tTKZTFose6OC3mw2q1wup0ajcagxfN9XsVhUoVDQaDRSv9/fdS/my1ufyMIwVKlUUhAEajQaGo/HRzqPmalUKml9fT29H1tbW4/bslvP85TP55XP5zWbzdTr9TQYDOR5nmq1mgqFghqNxrGVFCel1LPZTJPJ5AlfYmtmNyzrfaKvAQAAAAAAAAAAAAAAAADgsYEyXwAAAAAAAAAAAAAAAAAAgCOYLxakQPB4eZ6XlqMetLaFQkGlUkn9fl/9fv+Wze1GZZI3e+wgCJTL5ZTJZCRJ/X4/LT/N5/MqFouaTqfq9Xq3ZY63i5lpPB5rNBpdtyzV8zyVy2VlMhldunTpUGNks1mVSiU551Sv1zUej5XNZlUsFhUEgUajkdrt9g3LhB/vksLo2Wymfr9/5DwUCgWdPHlShUJB29vbj+siXynaW0nJeL/fT4udgyDQysqKJKnX6z3q54eZKZPJqFaraXV1Vd1uV1evXt11L3zfT58Ro9HoCVEyvd+68d0LAAAAAAAAAAAAAAAAALgdKPMFAAAAAAAAAAAAAAAAAAA4gnw+r1KppE6n86gKLbFbEASqVCryfV+DwUC9Xu+aIspMJqNisShJGg6Ht2ztgyBQJpPRaDS6LYWtuVxOS0tLyufzmk6n6XVPJhN5nqdKpaJ8Pq+dnZ0nRHnnYZiZJpPJDa87m82qWq2q1+tpZ2fnUOfP5/PKZrMaDodpWXIYhsrn88rlchoMBhoMBk/oMl/f91UqlZTL5dRoNI5cvpvJZHTq1CmVSiVtbW1pa2tLw+HwmGd7a02nUw0GA02nUw2HQ41GI0nRta6srOjBBx981GXFyfqvra3p5MmTqlarqtfr6na76bMwk8moUqlobW1Nw+FQV65cSUuEH4/fU2YmSbvmn7w2f4xz7raWrQMAAAAAAAAAAAAAAAAAPnpQ5gsAAAAAAAAAAAAAAAAAAHBIYRjq/PnzOnXqlD74wQ/elHJXz/M0m82O9ZyPdWEY6uTJkzpz5ox6vZ46nY42NzfV6XR2lTjm83nl83mNRqNdpbbHwcwUBNH/YjuZTNJzB0GgcrmsTCZzqBLY41IoFHTy5EkVCgU1Gg11Oh35vq/pdCrP89Jy6dls9rgvRT2KIAjked4Nj1leXlYul9OVK1fU7/cPdf5isaggCFSv1zUej+Wc02AwkHNOQRBoPB5fs/ZmJt/3NZvNnhB5ni/b7na7Go/HRzrP0tKSTpw4oY2NDW1tbaXreBRhGEqKynSTNTaz9Bl6M8pdPc9TJpORc07j8TgdZzAY7CoY9zxvV/nxo/meSPbvnXfeKTPT/fffr+FwqBMnTmgwGKTPxpMnT6pUKqlUKqnf72tra0tmlq7J/DpJ0VqFYSjP825pOfqi9ptPUtwrUeQLAAAAAAAAAAAAAAAAALj1KPMFAAAAAAAAAAAAAAAAAAA4BM/zVKvVtLq6qvF4nJYkHrcnQvHnYfi+r+XlZRWLRV28eFHj8ViFQkGZTCYtoJSUFtfmcrljL1CWpFwup1KppOl0qna7rfF4LM/zVCgU0nLMmzHujeZ07ty5tIS2Xq/L8zyFYajpdCozU6lUUiaTUaPR+Kgs800KXQ8q80xKVU+ePKlGo6GdnZ1rjk0KWieTyTX3OJ/Pq1AoaDKZqNvtpvkcj8cHFtr6vq/19XVVKhVdvXpVzWbzcZ/rpDR6MBjsWofDyGazOnfunPr9vq5evap+v3+kElYzU7lc1p133qler5eWAgdBoDAMlcvl1G631e12j7Xk1fd91Wo1LS0taTKZaHt7W91uN/0umN87YRhqZWVFo9FInU7nyGNms1mtra3p9OnTGo1GunDhgnq9nqrVqrrdrobDoSqVis6fP68gCLS1taXJZKJ+v6/hcCgzU6FQUDabVbfbVa/Xk/TI87RSqWg6nWp7ezt91i6yDkmh+M0qTT7I3u9dSnwBAAAAAAAAAAAAAAAAALcSZb4AAAAAAAAAAAAAAAAAAACHkMlkVK1WNRqN1Gq1JEWFiDg6M9Py8rJqtZp2dnbUbDaVy+WUz+evOdb3fRWLxbTI9jjLUc1MxWJRlUpFvV4vLd/MZDKqVCryfV+9Xu+WFkeGYaizZ88qn8/r0qVL2tnZ0WQyURiGms1mms1mKhQKKhaLkqROp7NvGWeyRx/vZbIHSUqNnXPyPE+e58k5l65FLpfT2bNnFYahNjY2NBqN0s8mJb4rKys6e/asHn74YV29ejU9xsxUq9WUy+W0s7Oz67PXs7y8rKc85SkaDAZqt9vp8+LRSEqcR6PRNfswuf69rwVBkBZi71dEvd/n9pPkIwxDNZvNhddhXhAEOnPmjIrFot7//ver0+kcKU++72t1dVXPeMYzdPbsWV2+fFmrq6uq1+sKw1ClUklBEOj+++/XYDBIr/vR5iCbzWp5eVkrKyvKZDLqdDoKgkDZbFa5XE6S0tJm3/dVLpdVrVa1vb19pPVKisTX19e1tLSkRqOhhx56KC0tHo/H2traUqlU0pOf/GTNZjM9/PDDWllZUbFY1NWrVzWZTJTP57W2tiYz27Wvc7mcVlZWVC6Xtb29LeeczExhGCoIAg0Gg11r5XmefN9XJpNRqVRSPp9Xv99Xq9XSYDC4Zc9G51w6VlLsm8ydYl8AAAAAAAAAAAAAAAAAwM1GmS8AAAAAAAAAAAAAAAAAAMAhlEolFQoFNZtNTadT+b4v3/cpEXwUstms1tbW1Ov11Gq1NB6P0zLOvaWbSZnpdDrVaDQ69nLabDarbDarXq+n2Wwmz/OUz+fTouGjFHIele/7uuOOO1QsFvXAAw/sKuqdL2UtFArK5XLq9Xrq9/u79qHv+6rVaioUCrvW94lmNBopCKL/NTqXyymXy2kwGKjX6ymTyWh9fV2nTp3Shz70obSkOSlKrVQqCsNQq6urOnv2rIIgSAtKfd9XqVTS6dOn0+LStbW1dP/1+30Nh8Nr5pPP53XHHXeo1+vp4sWLkrTv8yHZz8k1JIWkmUxGnuep3+/vOnZ5eVnr6+v64Ac/uGsPZLNZFQoFdbvddI8GQaBKpaKlpSX5vp8WZSf3P7n+6XS6a5yDZLNZlcvldK5mJt/35ZxbKIdmpkqlovX1dW1sbKSlt4cVhqFOnTql06dPq9Vq6f3vf7/K5bKcc+r3++r1evI8T6PRKM1Msqb5fF7T6VTdbjd9bTab7XsP9869WCzqxIkTKpVK6na72tzcTIuC8/m8yuXyrpL3IAhUrVYlSZcvXz70tXqep0qlohMnTiifz2tzc1NXrlxJ134ymajT6ahUKun8+fOazWZ68MEHVS6XVavVVK/X1el0ZGZaWlpSuVxWvV5PrzUpG67VappMJur1epKi58nq6qomk4k2NjY0m81kZspmsyqVSspms/J9Py0RT/bcaDTat0j8uJlZ+vPeUl8AAAAAAAAAAAAAAAAAAG4FynwBAAAAAAAAAAAAAAAAAAAW5HmeSqWSJKnX66lYLO4q4KTMd39mpjAMFQSBRqPRrhJSM1O1WpXneWo0GmlBb1JWORwOd5Vg+r6vXC6373vHIQzDtCx4Op0qk8moWq0qk8mo1WrtmvvNZGY6deqUzp07p/e85z1qtVq79td8gWWpVFIul7umbNj3fa2trenZz362xuOxNjc3NRwOd5X5hmGoXC6nQqGg2WymVquV7unb4ag5Sq7LOZfutWTfJSW9m5ubunz5sqSoaHVtbU0rKytyzqndbmtjY0POOa2vr+vEiRMqFAoKw1C1Wk2rq6saDocKwzAttJ3NZppMJqrX62o2m+lcwjDU+vq6fN/Xhz/84bQkde8+LhQK6Vi9Xk8XLlxQu91WNptVtVrVbDZLS3bNTOVyWXfccYcGg8Gu0tRsNquzZ88qn8/rwQcf1Hg8ViaT0fLysk6dOqVyuaxut5uuUavVkpmpVqtpfX1djUbjhmW+SSn0+vq6yuWyJpOJMpmMJpOJBoOB6vW6ut3udc+Rz+d19uxZSdLGxsaRSqWTa11aWtLDDz+sfr+vO+64Q+PxWJcvX9bOzo5qtZpGo5Hq9Xpabp3JZLSysqJyuZwW/JbLZVWrVbVaLV25ckVS9IzPZrOSpMFgkD7bK5WKzpw5o3w+r+3t7TRLs9ksLQnO5/NpUW5SlLyysqJ+v79rfyzC8zwtLS3p5MmTCoJAV65c0fb2drpmSUay2azuvPNOZTIZPfDAA5pOp2kR79bWliaTiQqFgpaWljSZTNRut9NnRC6XU7VaVRiGqtfrGgwGymQyWltb0+rqqi5duiQp2s9JKXRSdD0cDuV5njKZjAaDQZq9W8nzvF1F0s659DUAAAAAAAAAAAAAAAAAAG4mynwBAAAAAAAAAAAAAAAAAAAW5HmeqtWq+v2++v2+qtWqJpOJnHNPuEJfz/PkeZ5ms9mjKsxNynrPnj0rz/P04Q9/eFchbiaT0dmzZ9XtdtXtdtP1KxQKMrNrSiKT8tl2u50WW/q+L0m7Ck6PIimnlJTOsVgsanl5WYPBIC1lvdmSotUnP/nJunz5sprN5oH7KpfLKZvNajAY7Fo/3/e1tLSkO++8U7PZTBcvXkyP7XQ6u8pckzUdj8cyM+3s7FxTWux5Xrq/j7tAWYrKdUulkkqlkra3t29YLrvXZDKRmSmbzaYlvKPRSMViUSdPntRgMNB9992n0WikTCajkydP6uTJk2q1Wrp8+bLG47GWlpZULpcVBIEKhUJaLD2bzdRut7W9va3Lly9rOp2mBae1Wk3OuV1lreVyWUtLS9rY2FC73U7XKykXLpfLKpfLyuVyacF1EATpuQqFgpaXl7Wzs5Oe0/d9nThxQqVSSRcvXtxV5nzixAmdPHlSm5ubGo/HymazWl5eVrlc1s7Oji5evCgpylqSlUwmo/X1ddVqNW1vb++6D2EYajKZpPkKgkCVSkXnz5/XiRMn5JxTPp9P173VaqnVal33/phZWqZ75coVdTqdQ++jpLh2aWlJFy9e1NWrV1Uul9VsNtXtdhUEgZaWllSpVNTtdtVut9N9USwWVavV5HmexuOx8vm8Tp48mRYdm5kkqVKp6OzZs+m+mEwmKpVKOnfu3K5S3eFwmN6DTCajUqmU3svZbKYgCFStVtOC5aTkdxFJefCpU6eUzWbTkuK9mUyencViUffdd59arZbOnTunWq2mBx54QO12O90f+XxeV69eTcuNk3taqVTU6/XUaDQ0m81ULpd15swZNZtNtVothWGotbW19Bm4s7Oj8XiscrmsQqGgVqulnZ0d9fv9m/Jc2E9S2rvfugEAAAAAAAAAAAAAAAAAcCtQ5gsAAAAAAAAAAAAAAAAAALCgMAxVLBZ3lWYmRb6PN9VqVZLU7XZ3lUQmhahJ8Waz2VSj0bimXDVZi0wmo+FwqH6/f03xru/7qlarOn/+vNbX17WxsbGrcDd5f2lpSVeuXEnnkclkVCgU1O12dxVEBkGgYrEo3/dlZioUCmn56mg0UqvV0mg0OvKa5PN5ZTKZtHS1VCqpUqnI8zzV6/VbVlZZKBR05swZtVotPfjgg9eUeCY8z1OxWFQYhhoMBhoMBunrlUpFJ0+e1Gg00jvf+U4557S2tpZeQ7Va1VOe8hQFQaCtrS31+31Np9O0lDqRrHmxWFQQBBoOh2q1Wrv2g+d5CsNQQRBoPB5fsw8WEYahlpeXVSgUdhXjHoZzTtlsVsViUZ7nyTmnpaUleZ6nhx56SN1uV5lMRmfOnNHJkye1tbWlhx56SJ7n6eTJkzp//ryKxaKuXLmiS5cupWWoQRBoNpvp6tWr6T7IZDLKZDJaWlpSGIbpePl8Xuvr65rNZtra2tq1Z5Ji4WRvb25uyjmn5eVllUqldB2q1ap839+1xoVCQevr62q327uKc0ulkk6dOqXhcKidnR3NZjOtrKykhbbb29uaTqeqVCoKw1Ddblee56lcLqtWq6ndbqvT6UiKsr+2tqZsNqutra1dZb65XE75fF79fl+bm5u6cuWKBoOBnHMaDoc3LLr2fV8rKyvqdrva2to6cE9f7/Nra2taWVnRQw89pKtXr6alrhsbG/I8T6urqyoWi3LOqd1up4W7SbFuoVBQvV7XaDRSqVRSGIbq9/vp9efzeZ09e1ZPecpT9N73vldSVCB87tw55XI5Xbx4UfV6fdf+9jxP+XxexWJRw+EwLe3NZDJaXl7WZDLRzs7OoYrGc7mczpw5o1wup42NDW1tbV3zXPM8T0tLSzp16pTuv/9+NRoN+b6fFlRfvXo1LbM+ffp0um9Go1H67FxdXVUul9NDDz2kfr+vMAy1urqqcrms+++/X7PZTKdPn9apU6fSAuYgCLS+vq5sNqvt7W1tbm7e9CLf+edRsu7zBdn7HQcAAAAAAAAAAAAAAAAAwM1EmS8AAAAAAAAAAAAAAAAAAMACzEy5XC4tjjUzmZlms1laLLhfgWlSMPhYKvzNZrM6efKkptOpJpPJrmLNbDarEydO6NSpU+l7ZqbRaKTpdCozS0tJs9mszCwt9P3IRz6iTqcj55wymYxqtZrW19fTYtF6vZ6WWialwcvLyxqNRmq325KikspqtapcLqd6vZ4eGwSBSqWSVldXValUlMlkVCqV0qLVpJD2MGW+nucpCAJ5nqdsNqvV1VXVajXNZjPl83l5nqdcLqfRaJQWfh4kKchM1mIwGByp4DKTyejUqVPK5XK6//77rylRnpesie/7GgwG6b5MCmVzuZwefvhhtdttLS8vazabaTqdKggCra6uKpPJ6MKFC2npbBiGcs6l9zmfz6tWq6lQKMjzvHQvjMfjtMg1KS2tVqsyM00mE7VaLTUajfReJIWnZpaWBu/l+75yuZycc4+qGDSTyaQ5rVQqkqSNjQ3t7OzI8zytra3p3LlzajQa2tzcVKFQSPe7mem+++7T5cuX1e12NZ1OVavVdhVWz2fd9/204NjMVK1Wdfr0aa2ururChQvpGklRefK5c+cUBIG2t7fTUuAkH71eT81mU5lMJs1Lsn5mpqWlJeXzeV24cCEtjDUznThxQtlsVg8++KDa7XZaMJzch+FwmJbNep6n8XisMAy1tLSUfi45Xzab1fnz59VsNnfdo8lkok6no+FwqPF4rIceeigtNE/W4kby+bxWV1f1wAMPqN/vH+p5aGYql8s6ceKEHn744bT42zmnnZ2dtEQ5KRjf2NhQu91Or6FYLKpWq2k8HqfPlGw2q9FopF6vp36/L9/3tbS0pFwup16vp3q9LjPT6dOntba2pvvvv187OzvpuImkGDmfz6vRaGgwGKTPyGq1qosXL97w2THP8zydOHFCpVJJm5ub2tjY2PeZFoahzp8/r8FgkJZCr6ysaGVlRW9729vSPbm2tqZKpaJLly6lhcvJHqlUKhoMBup0Ounzem1tTb1eT71eT8vLyzpz5oyuXr2q7e1tra2tpaXTV69e1ebmpkaj0U3/brve+eff21tEDgAAAAAAAAAAAAAAAADAzUKZLwAAAAAAAAAAAAAAAAAAwAJ830+LU3O53HULfKVHChOXlpYkKS3wnC/OXVRSmJvJZDSbzTQcDo90nkShUFCxWNRwOJTv++nrSTlsPp9Xq9VSu92W7/vKZDLyPE/OORWLRZ0+fVq5XE5Xr17VdDpVuVxOi02DIEjLQpPix6QcMikwDcNQuVxO1WpVpVIpLcGUopLKarUq55x6vV667ktLSzpx4oTOnTunXC6nra0t1et1dToddbtddTqd9ByL8H1f5XJZxWIxnffS0pLCMFS9XtdwOJTnecpkMur1emlRbmL+vodhqPX1dX3Mx3yMMpmMLl68mJboHqaY1vO8tFB4a2tLrVbrusdns9l0jZPy4DAMVavVVC6X1Wg00rLTYrGoUqmUlpQWCoX0Ho/H47TE1zmXFpQuLy8rk8mo3++nhaTZbDYt9Q3DUCdPntTy8nK6/tVqVSdPnpSZaXt7W5PJRPl8XqdOnUoLaveW+ZqZPM+T53nq9/tH3ttJaXBSBt3tdncVzxaLRZ09e1bj8ViNRkPValXLy8vyPE+dTkdbW1u6fPnyrrLZSqWiIAjUbDbT0lvpkfJhSep0OruKguv1elr8KkVFtk960pMkRcXC9Xpdnufp7NmzqtVqajabunLlinq9XlrCfPXq1XS8MAx16tQp9fv99H4m5z19+rR6vZ4ajYYmk0lauLqxsZEWQRcKBeVyOe3s7Gg4HKpQKGh9fV3dbleNRkOz2Uxmlu6b+++/f9e1zl9Hu91Wt9tN98yiSqWSCoWC6vV6WgK8iKRU+vz58zIzXb58eVe5bZKvMAy1srIiM1O3203n7Pu+VlZW0nLcwWCgXC6n8Xis4XCYXn+yRqPRSDs7O+r1eqpUKrrjjju0ubm5q7w4key3Uqmk6XSqVqul2WymXC6XFklvb28v/AzwPE9LS0vpvdna2tq3yNf3fS0vL6tSqeiDH/ygJpOJqtWqzp8/r1arpU6nIzNLy88Hg4H6/X76WrVa1dLSkmazmTqdTvrcX1lZke/76na7ymQyOnnypFqtli5evCjpked48sw9zH28GTzPk6Rd6/tYKs0HAAAAAAAAAAAAAAAAADxxUeYLAAAAAAAAAAAAAAAAAABwHWaWlukuLS2pXC6rUqloNBrJ87y04HW+kLRUKuncuXM6ffq0zCwteex0Omlh6qKS8y0vL6tQKGgwGKher6vZbB6qKHZeLpdTJpNJC2rNTM45ZbNZ5fN5jUYjNZtNTSaTtPQxmXOpVFKxWNTOzo52dnY0m83U7Xbl+35a3Fqr1ZTJZDQcDmVmmkwm6nQ6Gg6HCsMwLaLM5/PKZDLa2tpKS2RLpZJyuVxaqJuU7Calqt1uV1evXtVHPvKRtCx2Op1qMpkcujg3CAJ5nqfpdJqeo91ua2dnR/1+XysrK8pkMtrZ2dF0Ok0LiAeDQXpPzUzr6+tpyXC321U+n1c+n1e32114TmamYrGo9fV1dTodXb169YaltsViUcViUd1uV4PBIC3rXFlZ0Xg81vb2tkajUVrwWygUJEXll9PpVL7vy/f99P4752RmKpfLWltbUxAEajQaajab6X4fj8fpvqnValpeXla73dbGxkZakLq2tqalpaW0BDiTyWgymajf719T5Jtce3Iv5otYDysIAlUqFa2tranf76clrKPRKC1APXnypC5cuKBSqaRsNqter6fNzU2trq6q0WhoOBymez0pNZ5MJtfMK5vNqlwuazweq9vtyvM81Wo1FYtFfeADH1Cv15MUFa/ecccdKhaL+vCHP6xGo6EgCHTu3DlVKhXt7OxoY2MjLVAtl8vKZrNqt9uaTqfpOp8/f14f+MAH0sJqM9OpU6d08uRJvfvd71a/35fv+zp9+rRms1la0lwoFFQul+WcU7PZlHMufZ68973vTQt/wzDU2bNn1el00oLbvXstDEP1ej0NBoNDPcOSQt7xeHzoz2YyGa2urqYlw/uV24ZhmO65zc1NDYfDNJvJtSbPoKS0ejweK5PJaDweKwgCZbNZDYdD5fN5NRoNZbNZnT17VtPp9JoCYSnaa7lcTsvLy8pms2nBre/7KhaLqlQqafHxopKczmYzNZvNdA/tXctk7yUlvZVKRadPn1apVErLgwuFQprhfr+flh1nMhn5vi/nnCaTiYIg0MrKisIwTJ+DkrS0tKRisagLFy6kmbh69Wq6fnv3R/L9lzxHbnapbvK9S3kvAAAAAAAAAAAAAAAAAOB2oMwXAAAAAAAAAAAAAAAAAADgAPMFvYVCQbVaTcPhUKPRSIVCQbPZTLPZTJ7nSVJaxlmr1eR5ni5fvqzJZCIz03A43LfI9HqSctJaraYwDCVFhY9hGMrzvCOV+SZlk6VSaVchsed5ymazymazaWGr7/vyPE+9Xm9X2WtS6poU6SbFsMm1S1Kz2ZTnecrlchoMBmmRZqVS0fr6upxzyuVyms1m6vV66TrWajU559RqtdL1qtfrarfbWllZkXNOjUZD9Xpd/X4/LRo+bKnjZDJRq9VK55XL5VSr1TQej9XpdOT7vvL5vCaTiXq9nnzf14kTJ3TixAlduHAhPU8ul9OJEyc0GAz0wQ9+UGamlZWVtGxyUZ7naX19Xb7v6+rVq2lp6/WUSiWZmTqdjgaDQVqUXCgU9JGPfETtdlvOubTIttFopGtdr9d1+vRpVatV9ft9DYdDSVI+n99V5Fuv1zUajeScSwtSkyLTU6dOqd/vp2W0zjnV63XNZjOdOnVKxWJRw+FQhUJB/X7/umW+mUwmLZI+bE4SuVxOa2trymazunjxoq5cuZJeV1LOWi6X5fu++v2+tre31Wq1NBqNVC6XFQTBrr3k+75yuVxaRJ3kLdnX+XxeOzs7mkwmKhaLqlarGg6HaflzshfOnDmjCxcuqNVqqVAo6PTp08pms9rY2NDW1pb6/X66B3O5nPr9vrrdblrw/KQnPUme56VrK0mVSkXnz59P8+OcS0u0r169qn6/r0wmo1qtlt7L2WymUqmk9fV1SUqLZpP9sbS0pA996EPXlCkHQZDem16vd+iy5TAMlclkDlVunTynTp48qfX19bTEfC/P81StVnXixIm0sDhZ+zAMtbS0pEKhoE6no/F4nJZXB0H0v9Eney8pWU+ed7VaTaVSSZcuXUpzNH89+XxepVJJ5XJZ0+lUnU5Hk8lEYRiqVCopDENtbW0daq3my9R7vd41OUjmWqlUVKlUJEnLy8u7CtMHg0FaOj+bzdRqtXYVXDcaDbXb7bTMOjlPu92W7/uqVqsKgkDLy8vyPC8t7nbOaTweXzPnpAg5m81qNptpMBikmTuq+VLg66HIFwAAAAAAAAAAAAAAAABwu1DmCwAAAAAAAAAAAAAAAAAAsA8zS4trgyBQrVbT8vKyms1mWnY4Go2UzWaVyWQUhqEqlYqy2WxaPJkUdHqep2azqclksnABYTJGrVbTaDRSvV5Pyz6DIJDv+4cu1ZSiosRSqaRSqZSW1s4XFpdKJTUajbQMczabpaW5ktTv9yVFpa+ZTEaDwUBmlq7PcDhUs9nUYDDQ8vKygiBISyaTctFcLqft7W2Vy2V1Op30nLlcTtVqNS0jdc5pOp2mZbJJaWWv10vLkY9a6Oic02g0Sn/PZrPpXEejUVoC2mg0NBgM0mLRfD6/q9i1UqmoWq3qvvvu0/b2tk6dOiUz02AwOFTZclIK/PDDD6vZbKaf9X1fYRjK9/10LZJS3VwuJ+ecBoOBJpOJlpeXtby8rMFgkJbMmplKpZKWlpb0kY98JF3LRqOhSqWipaUldbvd9NjV1VVls1nt7OyoXq+nxZzzRaLValWz2Uy1Wk3ve9/70iJfKSpJbrfbWlpaUqVS0XA4VKlUum6xqe/7aaHqUYtAfd9XpVJRoVBQs9nUpUuX0n3l+75KpZKq1aq2trZ0+fJltdttDQYDjcdjeZ6nTqejbDabljCbmfL5vCSlxyWSctuk+NXzPBWLRQVBkJbzSlKhUNC5c+c0Ho/VbDZVLpfT8tWNjQ3t7OxoMBjIOacwDFUsFtPy4Mlkokwmk5bsNptNdTodSdFerdVqymaz6na7Gg6HCoJA1WpVzrn0uGq1qmKxqMFgoHa7rWw2m96XXq+Xlnnn83mtrq5qNBqp2Wxek6kwDNNS7sPu62Qtfd9PC8JvdB/z+bzK5bLK5XJaNpuUSu89b7FYTAvE2+22Op1OWgyey+V2la5ns1lNp9P0mj3PS/fddDpNy8Wn06nK5bIGg8Gufev7flpeG4ahgiBQEAQajUZpBnK5nEqlkobD4a7y5UUk55vPvnNOQRAok8kol8ulc0i+S5Jn6Ww2S7Pp+76azaYajUb6vJekXq+XFmqPx+O0hL3dbqvf76tUKmk8HisIAhWLRXmep3K5fE0Rt+/76XMpl8ulRb5JWfNxlOwedI4kn8n3wXxperJeRy0DBwAAAAAAAAAAAAAAAABgUZT5AgAAAAAAAAAAAAAAAAAAHCAIAlUqFVUqFT3taU/TiRMn0qLFkydPpoWgScmuJLVaLbVarbRgU1Ja9rhosaPv+1peXtba2poGg0F6vkKhoEqlIjNTs9k8UvGp7/uqVqsKgkC5XE75fD4t/5y/jmw2q3K5rF6vlxaOSlEhZLvdVrVaVafTkZmpWq2qWq1qMBio0Wio2+2mZahBEKTFu9VqVcvLy2kBaSaTSX82M62vr8vMVK/X0wLNpFTY87y00Lbf72s8Hh9LaWQik8nI87y01LZYLCoMQ3W7XXmep+XlZeVyOY3HY41Go3Rea2trcs6p2WxKkmq1WloAvOj9DoJAS0tL8jxP29vbaRllNpvV2tpaWhQ7Ho+1ubmpdru9q9RzOp2mBa9BEOjSpUvp3sjlclpbW9NkMkkLkiVpOBxqa2tLZ86c0crKigaDgcIwTEuem81mWp6alNdWq1XVajXVajX1er20uHe/+5AUjSYF10lxaBiGmk6nu4qtk0LpvaW5h71/tVpNZqatra1rCpFzuZym06kefvhhbW9vazwep+/PZjP1er1dJaBJAXCybvNFxGEYKpvNajQapetWrVY1nU515cqVdA8tLS2pVqupXq+rUqkok8loMploe3tbrVZrV0l1NptVqVTSbDZTvV6XFBVmFwoFjcdj9ft9DQaDtIg2CAK1222FYZiudbFYTOdaKBRUKBTScu2k1DmXy2kymWgymaSFt0tLS2meB4PBNWsbhqHCMNRgMEjnfBjzx+dyOfV6vWvOkZT4JoXMSdl2kqNut3vN8ZlMJr1Hw+EwLbZOypWTwl7nnLLZrLLZrPr9frqfk9wk5y4Wi7ty0+/306xnMhkVi8W0UHw0GqlUKsk5p263m5Y6J+W2zWZTvV7vUOuUXG+xWNTa2prK5XK6P2azWfr+ZDJJy9GHw6G63a6CIFCr1VI+n1e9Xle9Xk/vZXJ98+XnSYGvmaWvm1la8p2UAK+vrysMw7RgOsnq3nknRcFHze+86z03j/OZDwAAAAAAAAAAAAAAAADAUVHmCwAAAAAAAAAAAAAAAAAAsA8zU6FQ0NmzZ7WysqIzZ86o3+9rZ2cnLSwcDAYaDAYqlUoKgiAtEe31eprNZprNZmlx6XxR6H5jJSWFSenm6dOnd5XjJuWiSXnsokWxe2UyGZXL5bRAt1gsqtVqaTKZKAgCZTIZZTIZ+b4vM1On09lV0Dgej7W1taVcLqczZ85oeXlZnuep1Wqp0WjsKhMtlUryPE/j8VhhGGppaUn5fF7b29uSopLYpGA2KQ/e3NxUv9/XbDZLS06TEl0z03A41Gg0OvL17ycpBh2NRmkJczabTUs7k2tJymZns5k8z1OhUNDq6qra7bYmk4lqtZpms5m2t7fTItwbSUqPq9WqdnZ2NBgM0us+ceKE8vl8WqCalNWOx2P5vi9JaYlpqVRSoVBQr9dTo9GQc05hGKZl1EnBcLLPptOpOp2O2u12ekwmk5EkdTodjUYjeZ6XrkWxWEz/5PN5TSYTNZvNfcs7M5mMgiBI55wUtRYKBWUyGfV6vbSMOilFTsqKj1rWWSgUVKvV0uuan5eZKQiC9P5Op9Nr9s/8vU2uoVgspntgPp/ZbFb5fD4tds7n82nBcb1el3NOQRBoZWUlXYtcLpeWJCdFyEmBalIaXiqVNBqN1Ol0lMlklM/n5ZzTeDxOC3BzuZzK5XJa7OqcS8trc7lcWvo9f4/H47Gy2azMTN1uN70PSeFvuVyW7/tpFuefR5LSZ8FRy3wnk4m63a5KpZKWlpY0Ho/Tkt4gCFQoFFQsFpXNZtNC3l6vl661pF3Fx0EQqFqtpveg1WppZWUlLeKtVqtpkXG/31ehUNBsNkufHUlp+XA4TF/PZrMKw1A7Ozvp3k+ej0lBbxiG6vV66eeCINB0OlW3200zWSwW5fv+NXtwEYPBQDs7O+mzOCnrTu51Ui7snEuLeGezWbpHLl26JM/z1Ov1du3l+ftlZpKUfjcla5rc342NjbQsvFarqVKppM9tSfI8L92jSYFvsi+SudxsyTXsvTbnHGW/AAAAAAAAAAAAAAAAAIBbgjJfAAAAAAAAAAAAAAAAAACAfZiZfN9PyzPDMNSHPvQhPfjgg/I8TydOnNDOzo4ajYZOnTqVFiwOBoO0JHEymaSlndcrGUwKFSUpDEOtra0pm83qypUr6nQ6mk6nymQyymaz8n0/Pe9hJYXAYRhqc3NTg8FAuVxOS0tL6c+5XE6VSkWdTict59071263q263q6c97WlyzunixYtqNBrq9XqaTqdpCWcQBGlpZ6lUUrlclnNOo9FIYRhqPB5rOp0qCAIVi0X1ej1tbm6m58jlcqpWq5pOp2mh6Gg0Stf0uIRhqDAM1e/3NRwO0zLjpBw0m81qPB6na57cr1KppEwmkxaVJkXF7Xb7uuXNB92Tq1evajqdyvM8VatVFYtF1et1tdtthWGofD6vYrGYFi5LUQGo7/sql8vyPE87OztpIXE+n1c+n5fv++p2u2lZa2I0Gqnb7apcLqeFrkmBb6FQSH+XovLPyWSS7lPP89TpdK4pxU3KiZ1zGg6HWlpaku/7qlQqmkwmCsNQktJS1GS8+fU+rGSv5HI5TafTXaXFyf1KMpkclxRuJ/YWn4ZhqEwmk5b8Jq8nxcRJOe90OlW5XFapVEozlaxPuVyWFBUuNxoNNZvNNA9JCW9yLwuFQroHJalUKqXr5Pu+ZrOZ8vm8KpWKzEzNZlNLS0vKZrNp8W+pVFI2m9VsNkvz6JxL70en00kLaguFgpaXlyVJ3W43LWNN1nA+Y8k+6Pf7hy6oTda2Xq8rl8upVCrtute+7ysMQ81mMw0Gg3SfjkYjZTIZTSYTFYvFtGg8DMO0yLfb7ab3QIpympSL93o9dbvd9Hk8mUzUaDTSe2lmmkwmaXF1LpfTYDBQq9XScDhUq9VSNptVpVJJP5+cczgcyvf9tCy52+1qOp2m+yI572ELxyeTier1uvr9flqGPZ1ONR6P02flQabTqVqt1kLfNQf9nGS81+vJ8zy1222Vy2Xlcrm03Hg6narf76dFvklp8M0q0d1bLJ0UD+833nEWvAMAAAAAAAAAAAAAAAAAcD2U+QIAAAAAAAAAAAAAAAAAAOyRFAaOx2PV63UtLS1pNBrpypUrajQaKpVKGgwGaSHpcDhMS0KDINB0OtVsNjtUyaFzLi09rdVqqtfrarVamkwm8jxPuVxOYRhqOByq2+0eqbgwDENVKhWNRiNtbW2p1+upVCqlf2q1WnoN3W5XzWbzmgJJM1MYhgqCR/431Ol0ml5zch25XE7OOY3H47SwNpvNpmWmYRhqNBrJ87z0va2trbTkMwzDtGC22+0qm81KUlrme5ySa07mlpRXJqWds9lMrVZLxWJRnuelrxWLRZmZCoVCWka6s7Oj0Wi00LhJ6XG1WpWktHw1ud9SVNY7HA41mUzU7/fTkt1MJpOWb4ZhqEKhoOl0qm63mxbq5vN5SUoLOvcWsc5ms7TgNylUlZQWBs9mM41GIw0Gg7RUNTlfUuq5d38HQaBcLqd+v58W3UrRHknKQX3fVxAEaUFoEAQys3T9D8v3fWWzWXmel15jUkKcjJ2UsJZKJa2urqrRaGg4HGo2m8nzvHQuyVonpcPJ7wnP89Ki6qQQONkXOzs76fjJdQwGA9XrddXrdY1GozQfSYmv53npa/P7ajabqd1uy/d9SVImk9GpU6cUBIEuXbqUFtnWajVVq1WVSiUtLy+r2WxqNBqlhdmFQkHj8VjtdluDwUCe56nZbKparSqbzarZbGpnZ0e5XC6dV7IP5/fpbDbTcDhcuKR6r06no83NTdVqNYVhmO7vZF8nBeFJqXBSuNvr9VSr1bS2tqZSqZTe1263q52dnfTeJNeUyWTUbre1s7Ojfr8vM1O/30+zknz26tWr6ZjT6TR9JibP1o2NDVUqlTQDSXltUgptZmlp93yGfN9Xp9PZNxuLSErD5wtrFz3PUcts9yv1TcqOe71ems9kHyT5vxXluXuv/XplxUnROAAAAAAAAAAAAAAAAAAANxtlvgAAAAAAAAAAAAAAAAAAAHOS0kIpKlZstVoajUZpqadzTplMRuPxWM45TadTNZtN+b6vcrks55za7fau8tRFyxhzuZxWV1c1mUzSYtukJLJYLCoIAvX7/WtKWReVnKfT6ajZbKrT6aQFp+VyWWEYajweq9frpWWne9cml8upWq3K8zxdvHhRQRAoCAKVSiVNJhONx+O00DVZn1wup2w2q+l0qsFgkJaXzmYzlctlTadT9Xo9dTodOefS8/m+n5aQJgWn4/H4wELR5L4dpkTT9/20lHc8HsvzPOXzefm+nxbr9vt9SUpLW5Ni4Xw+r0KhoGq1qk6no0ajkRaHLiIp8y2VSur1ehoMBpKiUs6kRDMpKE2KSQeDgabTqYIgSNcjDEOFYZiWjebz+XS9kuOTa/B9f9f6jcdjZTKZtJxaispvJ5OJhsOhBoOBRqORCoWCMpmMnHMajUZpAeq8pJhZklqtVnrvZ7NZujbZbFaFQkFhGMrzPGUyGRUKBXmep16vd6Sy2EwmozAM0/knvycZnc1m6nQ6unr1qtbX17W6upqu6Ww2Swtzk6LoyWSSru/evCWFt57nqd1uS4r2Rb/fV7PZTO/9dDpVo9HQ8vJympGkuDhZy9lspm63m96jJJ+StLm5me6BVqulSqWiXC6nS5cuqd1uazwea2trKy3CDoIgvTfJnJJy4m63m15rci+SQuV6va5utyvf95XP5/ct7E32UTLPo0jWYzAYKJPJpPlKioeTgth5k8lErVZLmUxG+Xw+LTNvt9tqNpu7SpY3NjbSvZoUGifPgb3l38kzJbnXktTv99P8SFE5cJK55Lm2d01arZZms5mm02n6PGu322nJ71Fdr7D2Vkme3dPp9MjfNwAAAAAAAAAAAAAAAAAAPFFR5gsAAAAAAAAAAAAAAAAAAHCA+ULFbreblnLmcjlJj5SeNptNOedUKpVUKpUUhqEymcyuos4blTN6nqdqtapSqaSdnR11Oh3NZjOFYahcLiff99Oy173llIvKZDLKZrPqdDrq9/vq9XppYWtSYJrNZtXtdneVYUpR8WyxWNTy8rIymYzq9bp6vZ5yuZyWlpZUrVY1mUzU6XTS8s2krDVZq9FolF5TPp9XPp/XZDJRo9FQp9NJSzGTstO9Bb+j0UiDwWDfslwzUyaTkZml4ywimYtzTuPxWNlsNl1vM0uLcD3P02QyUalU0tLSksbjsWq1mmq1WjrPg+Z2PZ7nKQgC9Xq99L4mxafT6VTLy8s6c+aMtre3FYahOp1OWq45mUw0Go0UhqHMTJ7npXNPSqXny0iTAtfxeKzJZCIzS9d7MBio3W6na5fs7aSUNNmHg8EgLbSdLyw1M1UqlbTYuNvtyjmnK1euaDabpeft9/sql8sqFArq9XqqVCpaXl5Wq9VK9/dhFQoFSdLVq1dVKBSUz+dVKpU0Ho/TQt/BYKArV65oMpmoUqmkhbxSlPOkrDu5piAINBgMdhW8SlGZr+/7aUluUgSc5CHhnNPGxoZyuZyWl5fleV66N5K93O121e1207We3+edTid97eGHH0734eXLl9NsttttPfjggyqXy+keSnIgRc+snZ0d9Xq9NMvza+F5Xro+Ozs7aQHy3iLawWCger2ufr//qEpmk9LupCQ3Ka096JzOOfV6PV2+fFlBEP1v78le3vuZTqejXq+n2Wx2wzkmJbXz9j5TkwLjgyTPi71zHY/H15Sg49a43QXIAAAAAAAAAAAAAAAAAICPHpT5AgAAAAAAAAAAAAAAAAAAzHHOycwkRQWl+XxeZqZGo5H+nhTVzmYzzWaztAi01+spn88rl8upWq0qk8mo0WikxabXEwSBarVaWg48nU4VBIGKxaKy2ayGw6H6/b583z9ymW82m1UYhtrZ2dFoNEpLLSeTSVruury8nBbcJqWuZqZsNqv19XUVCgVtb29rZ2dHk8lEg8FAnudpZWUlnX8QBMrn8yoUCmkpZqfTked5KhQK8n1fq6urymQy2t7e1ubmpvr9voIgSNet2+2mhaa+76fFo8PhcN/C3CAIVKlU5JxTo9FYuFQ3l8spk8mo1WppPB6nhcqe52k6nWowGGg0Gsn3fXW73bTA1/O8tDg4KdEtlUrpOjrn1O/300LW/TjnNJlM0gLS+ePG47EajYYkqVaraWlpSZlMJr134/F415o459I5jEYj1et1dbtdhWG4q1C6XC6nxcpJgXSxWNTGxkZaXrt37ZJS5iAItLOzoyAItL6+npYDJ7lYWVnRZDJRu91O9+h8wa0UFcwmpchmlt7ver2+qxx1UUnJ9HA4VKPRUKlU0vLyspaWljSdTtNi6qRI9tKlS6rX68rn8wqCIC1l7ff76Tp5nifnXPrZebPZLC3pHg6HCsNQW1tbaSnvvGazqUuXLml1dTXdG71eLy3+TcZL9tbGxoZGo5E2NjbSwtrpdKqtra203HlvgWyz2VSr1ZLnedre3lalUpHneWl+9s5/ft/N26/gNpFk4KjPnb1jH6Z01Tmn0Wi073XsPe6g+d8KyXMuySIAAAAAAAAAAAAAAAAAAHjioswXAAAAAAAAAAAAAAAAAABgj6SM0cxULpc1mUx05coVOedULBZVqVSUyWTS42ezWVqwmhT6JiWpSXHqcDi87phhGKpUKmlnZ0f9fl+e56lUKqlcLms0GqnX66lUKsn3/SOXReZyOXmel5bMzs9/MBio2Wwql8upUChoZWVFYRhqOBymJbyrq6tpeWhSNjoej9VsNlUul1UqldRut9NSTd/30/LTpNQ1KTouFApqt9u6cuWK+v2+zEyVSkVra2va2dnZtxw1KRTdr6g3DENVq1UNh0PV6/WF1iMpoZ3NZur3+xqPxwqCIC1P7na76vf7aRFuo9FI35ekra0tNZtNLS0taX19XZVKZddcL126tGs99krWbzAYpOWy88fOZjN1Oh1tbGzoxIkTaRnseDzWaDRK92Cn01Gz2VSlUtFsNkvvUVLa2mw2NR6P02LkZE/ncjktLy9rNpup3W4fWJSclDtLUrvdVhAEWlpa0tLSUlq0nJQEb21tXbd4NSkaXllZ0erqqiTp6tWrajabR9rXmUxGZpbusSRnpVJJ1Wo1LXlN1jUpG+50Omnp8t5xzSwtwt1vD25ubkpSui+uXLmy7z0ej8fa3t5Wr9dTEASaTCYaDocaj8fpZ6XoPjcaDbXbbc1ms2sKe5Oy7YMk19jpdNTtdtPXjqtU9npFv9jtuNY8KZSnGBgAAAAAAAAAAAAAAAAAgMceynwBAAAAAAAAAAAAAAAAAAD24ZxLi08Hg4G2t7dVLBZVKpWUz+fleZ7G43Fafpr8dzqdajabKQzD9PhFymWDIFAYhur3+5rNZioUCqpWqxqPx7uKbT3P02w2O3TJY1Jcm5xvbzloUgYaBNH/XlosFhWGoQaDgWazmYIgkOd5mk6nyuVyacmomcnMNBwOtby8nJYFNxoN9ft97ezsqNFoaDKZpMW1pVJJvV5PmUxGnuepWCwql8tpZWVFnU5HW1tb6fySwtXZbKbhcLhvqamZKZPJqFAopOW7i65JEARpEex0OlW/308LWHd2dtJi1aTw+MqVKwrDULPZTNPpVNvb21pZWVGlUknLkpPS3WQvXM94PNbOzo5OnDih5eVl1et1TafTdG7lclnFYlFbW1vp+klK10+S+v2+Ll26pKtXr6bF0cn+SAp2fd/XdDpVq9WSmSmbzSqTySgIAj388MNqNpv7rluyb/L5fHpdQRBoe3tb5XJZmUwmLR1O7vn19uZ0OlWj0dB4PE7P2Wq1rltWe6P71+/31e12NR6PNR6PtbW1pX6/r2w2m+6dva5XdpuUG+93zN5i3aTIez/OOQ2Hw13v7zdmUuo8X+J7FMdZ4AsAAAAAAAAAAAAAAAAAAID9UeYLAAAAAAAAAAAAAAAAAABwgEwmo3K5rAceeEC9Xk9mplqtlhaYJn+SAs3kv0kxZ1Iyu0jBZlJMmpTbrqysaDabaWdnR/1+X2aWFokuUhK7l+d5CoJAnU5HvV5v38+Px2PV63WNRiMVi8V0zNFoJM/zlMlk0uLdwWCQlsaWSiVls1mNx2MNh0MNh0NdunRJktIS36RUdTKZqNvtqt/v64477tAdd9yh8Xgsz/O0ubmpy5cvX1Nq6nmeGo2GhsPhvoWzvu8rl8tpNpup1+sttB7JvUkKakejkZxz2tnZUbvd1mQyuaaYdr/S1clkouFwqO3tbQVBsGvNkuu+nqTMtlAo6NSpUyoWi+r3+wqCQLlcTmamVqul7e3t9N5L0mAwSEtxp9Ppda97fs6dTkeDwUCZTEZLS0uaTqe6dOnSgSW8ydomxc6j0Uij0Uibm5vqdDqSpOFwqG63e2Cp7X7zaTabarVaj6qA1szS4t35AuNkbyZF0Ic9/3GX4lKwi8NizxxNknkAAAAAAAAAAAAAAAAAAG4mynwBAAAAAAAAAAAAAAAAAADmmJkkKQgClUol5XI5tdttTadTDQYDOeeUy+XU6XSUz+cVhmFa6pt8NikB9n1fzWZz3wLavabTqfr9vk6dOqXRaKTBYKArV66o2+1qNpul5+r3+wuXps7zfV9bW1tp4e5BknLZbrebFiMm408mExUKBWUyGYVhqGw2K8/zlMvlNBqNdOXKFXU6nbTAV9I1ZbiTyUTOOTUaDTnnVKvV0tLig9ZqOp2q1WrtO18zUzabVTab1Wg0Ur/fX3hNzCwtFk7G7ff7hzqHcy4tuE3OeZgyydlspm63q4cffljlclmlUkmlUiktVm61WhqNRteUL+8tFT6M5B54nqd+v6/BYHBgObTv+8pms5KkXq+XFkm3Wi212+30Gg7rOApzZ7PZgffquAt5ATw2JN+z++WbMl8AAAAAAAAAAAAAAAAAwK1AmS8AAAAAAAAAAAAAAAAAAMAeSUHt6uqqBoOBer3ersJeM1Ov15PneVpeXk4LdqfTqTKZjKrVqkqlktrttra2thYqOx0MBrpw4YIKhYK63W5a4poUE06nU9XrdZnZQuXAeyVlu0k57/Ukpbt7X0tKfoMgUBAE8jwvLVZMSoJns9kNyxRns5kmk4l2dnZUr9flnNN0Oj1SCaPn/f/Z+XNfSbIEz/f72b64+e5+l4iMqKyq7unBtDLgKBRGeAAVkgo1AlT4QBB4Cv8AUqNKmQqBJxAkFQLUSIEK8QCC0siD7mk0+lVVLhEZd/F9MbfdKETZ6XtjybyxZWRkfj9AICL8upudc+yYuXLxteX7vizLMmv2EG3bKs9zM+aP5X3mUNe1jsej0jTV9fX1Bx3rodq2leM49yLUb9JFm6uqMmHp7vNEMwH8XLpQ7489d3gmAQAAAAAAAAAAAAAAAAB+DsR8AQAAAAAAAAAAAAAAAAAAXmHbtpIkUb/f13q9vhep7aKm2+1WaZoqjmONx2O5rmtig3mea71ea71eK8uyBwUGy7LUzc3NW3/+psDuu3qfCPDd80tSVVWqqsqEfLvXX/37bbpwbLee3f/fN8LoOI4sy9LxeNR+v3+nMO/HjPh+DD9niLJpGrN2r17Lu+I4VhzHOp1OStP0ZxsfAEi6FxvvYvrS68/vH4uSAwAAAAAAAAAAAAAAAADwMRHzBQAAAAAAAAAAAAAAAAAAeIXjOOr1erIsS5vNRmVZqm1blWWp7Xarb7/9VqvVSofDQWmaKggCue7LX8ssikKn00mn00lVVf2sgdaf04eGcB8a//0pdV0rTVM1TaOiKD7oWL81aZpqOp3Ktu03/tyyLCVJojAMtVqtlOf5zzxCAL91Dw3EAwAAAAAAAAAAAAAAAADwcyHmCwAAAAAAAAAAAAAAAAAAcEfbtrIsS7Zta7fbablcqq5r8/pyudR6vVZRFCZoezcm2DSNef3XGvL9EB97Teq6Vl3XH/WYvxW73U7n5+dKkkTH4/G1dQyCQP1+X5J0OBw+OOAMAB/qx75DLMviexcAAAAAAAAAAAAAAAAA8MkR8wUAAAAAAAAAAAAAAAAAAHhFXddarVZaLBbabrcmDtg0jbIsuxcL7P5NRBBfmqqqtNls1O/3td1ulaap2cO2bStJEjmOo/1+rzRNP/NoAeD179q7z6y7YX0AAAAAAAAAAAAAAAAAAD4VYr4AAAAAAAAAAAAAAAAAAACvqKpK6/VaTdOormvz+o/Fegn5vplt25Jerg9r9MtSVZWOx6N835fv+8qyzOx3z/MUx7GKotB2u1VRFJ95tADwdnzHAAAAAAAAAAAAAAAAAAB+LsR8AQAAAAAAAAAAAAAAAAAAXtG2raqqevD7LcsynwO+FE3TKE1ThWEo3/fleZ4JYiZJYmK++/1eTdN87uECwBu/Z7vvYAAAAAAAAAAAAAAAAAAAfg7EfAEAAAAAAAAAAAAAAAAAAF5xNxZoWdZPRnqJ+L4dEdhfrrZtlWWZTqeT4jjWaDTS4XCQJI3HY7muq9VqpTRNP/NIAeDtuu9gor4AAAAAAAAAAAAAAAAAgJ8DMV8AAAAAAAAAAAAAAAAAAIAH6CKBhHvxa1LXtdI0Va/X0+PHj3U8HiVJvV5P2+1W2+1WVVV95lECwE/j+xkAAAAAAAAAAAAAAAAA8HOwP/cAAAAAAAAAAAAAAAAAAAAAfsm6OGDbtoQC8auUZZm2260sy9JwONR0OlWe51qtVjqdTux7AAAAAAAAAAAAAAAAAAAA4K8sfrkWAAAAAAAAAAAAAAAAAPCpWZbFL6sBwC+Y67oajUZKkkSu62qz2Wi73aosy889NAB4K8uyXguOt21rfabhAAAAAAAAAAAAAAAAAAB+A4j5AgAAAAAAAAAAAAAAAAA+OWK+APDL5ziOPM+TZVkqy1JVVX3uIQHAa94U8L2LmC8AAAAAAAAAAAAAAAAA4FMi5gsAAAAAAAAAAAAAAAAA+OSI+QIAAOBDdBFfYr4AAAAAAAAAAAAAAAAAgM/J/twDAAAAAAAAAAAAAAAAAAAAAADgU7JtW5ZF4/VL8rbr9WMhXwAAAAAAAAAAAAAAAAAAPjVivgAAAAAAAAAAAAAAAAAAAACAX60gCNTr9eT7/uceCt7Bq9Het0V8iTQDAAAAAAAAAAAAAAAAAH5OxHwBAAAAAAAAAAAAAAAAAAAAAL9KjuMoSRL1+315nvfazwnBfvneFvkFAAAAAAAAAAAAAAAAAOBTcD/3AAAAAAAAAAAAAAAAAAAAAADg18a2bVmWpbZtzR/8vCzLku/7CsNQtm2raRrzM9u25Xme2rZVURSfcZQAAAAAAAAAAAAAAAAAAOBLQswXAAAAAAAAAAAAAAAAAAAAAD4Cy7Lkuq48z1MQBLIsS5JU17VOp5PKsvzVRH27WHHTNL/YOXUxX9u2lee5yrKU9HLsURQpjmPlef6zx3wty1IQBCYk/EtdPwAAAAAAAAAAAAAAAAAA8DpivgAAAAAAAAAAAAAAAAAAAADwgbpAbJIkiqJIvu+raRrZti1J2u122m63yrLsk8dbHceR67pq2/aTBIRd11UURXIcR6fTSXmef9TjfyzdOG3bVlVVappGkuR5nobDocIw/NlDvpIUx7HG47FOp5M2m43quv7ZxwAAAAAAAAAAAAAAAAAAAN6P/bkHAAAAAAAAAAAAAAAAAAAAAODXwbIs2bZtAra/RLZty3Xde69ZliXLst77mJZlKY5jzedzTadTtW2rzWaj9Xqt7Xaruq41Go00mUwUhuEHneshY+n3+zo/P9dkMpHv+x/9HFEUaTqdfrLjfyxBEKjX68lxHBPz7aLLg8FATdMoy7KfdUy2bWs2m2k6nUrSJw87/1p8ynsGAAAAAAAAAAAAAAAAAIB34f70WwAAAAAAAAAAAAAAAAAAAADgx7muqzAM5fu+iaRalqW2bVUUhZqm+dxDlG3b6vf7sm1b+/1eVVUpCAJ5nqeyLJXn+Xsd1/d9TSYTBUGgxWKh3W6nsizVtq0cx1GapppOp+r3+6rrWnVdqyiKjzy7lzzP0/n5uc7Pz7VarXQ8Ht97Xm9iWZaiKNJoNFJVVb/YyKplWer1eoqiSIfDwVwP13UVx7F6vZ7W67WqqvpZx+V5nubzuU6nk06n0y/ivgAAAAAAAAAAAAAAAAAAAA9HzBcAAAAAAAAAAAAAAAAAAAD4DXo1wtq27Xu/PwgCTadTjUYjDQYDVVWl1Woly7JU17U2m432+73quv54E3gPYRjq8vJSaZpqv9/L8zzNZjPZtq3VavVe0dsubuv7vpbLpdbr9b15Nk1joreXl5fq9XpK01RVVX2SkGsQBOr1eorjWLvd7qMf37ZtBUEg3/eVZZnKsvzo5/gYHMfRaDQy4+ziya7rKkkSOY6jw+Hws+/JJEnU6/X0/PlzZVn2yc5jWZa5Z38NweCfej4BAAAAAAAAAAAAAAAAAPBzIeYLAAAAAAAAAAAAAAAAAADwBbNtW5ZlqWkaYod4MM/zFIahXNeVbduqqkqn00llWd7bR7Zty3VdOY4j13Xluq7Zb3meK89zE6mdTCba7XY6HA6SpNPppCAIdHZ2Jtu2VRSFTqfTZ9untm2r1+tpNBppvV6rbVuNx2PN53NdXV29V8hXehlN9X1fp9NJ2+32jXHYtm11PB612+00nU4VRZHSNP1kMd+maZSmqcqy/Ojn8DxPcRzL8zyVZWkiuZ+KZVnvtWdc19VoNFKWZUrTVHVdy7IshWGoOI6VpqnSNP0EI347z/M0n89VlqV2u90nuf62bcv3fQVBINu2792rfEcAAAAAAAAAAAAAAAAAAPDhiPkCAAAAAAAAAAAAAAAAAAB8oTzPU5Ikcl1Xu93uvWOkv2WWZZlAbVVVnySu+SrbtuV5nizLUlmWb4y/fiqO46jf72symSiOYxPolaQffvhBi8XCxFnDMNRoNFKv15NlWbJtW5JU17VOp5PqulZRFHJdV48fP9Z+v9fNzY0JSxdFocvLS41GI202G7Vt+1ljopZlaTAYKIoiHY9HOY6jx48f63Q6ablcvvd1sCxLjuOoKAqVZfnW99V1re12q7OzM4VhKMdxVFXV+07nrWMZDAZq29aElrvr9rGEYajhcCjXdc0++BRs21YURQrDUJvN5p3OY1mWer2emqbR4XAwz0bf99Xv9+W6rhaLxVv3o2VZkvRR96tlWUqSRKPRSNvt9p0jyJZlmbDx28YVBIFGo5Hm87niOJYk5Xmum5sbLZfLTx5eBgAAAAAAAAAAAAAAAADgt4CYLwAAAAAAAAAAAAAAAAAAwBfItm0NBgM9fvxYeZ5rv99/7iF9VF0g1bIs1XX9ySK7QRDo4uJCbdtqtVrpeDx+8qBvF9wsikKbzeaTnacLf3Zc19X5+bkuLi5MiLUsSyVJoouLC51OJxNNjaJIw+HQhHy799Z1rTzPVVWV8jxXXdfyPE/j8VjX19dK01R1XattW/V6Pf3+979XlmVarVafNDbtOI6JCLuuqyAI1Lat0jQ17/E8T/P5XHmeK8syDYdDnZ+f6z/9p//0oxHeh5w7CIIHHaMoCu33e3meJ9d1P/qaRFGkwWCgJElk27ayLDOx5o/B8zz1+32FYag8z1UUhRzHkeM4kqSyLD9KANeyLEVRpOl0Ktu2td1u32ucRVHocDioqipZlqUgCJQkidq21fF4fO1zjuMoiiJFUSRJOp1OStP0g54JXYTXcRyNx2M5jqPNZvPgkLNt24rjWMPhUFEUqaoqbTYb7XY7M64uXvz1119rNptpt9vp5uZGZVnK8zwTEP9Y1wcAAAAAAAAAAAAAAAAAgN8yYr4AAAAAAAAAAAAAAAAAAAAfiWVZCsNQruuqKAoVRfHJwomu6ypJEoVhqOVy+UEx0rteDcB+LmEY6vz8XJK0XC4/WWT37OxMf/zjH5XnufI8vxd//RRs21av11Ov19N+v1dd15/kPK7rqtfrmXCtZVmazWb6N//m3+j29lbffvutiqKQ7/sm6FtVlZqmURRFevr0qZqm0WKx0PF4VFmWJpb76v5wHEdt28r3/Xsh37//+7+Xbdv653/+Z202m09y/SzLUpIkZj2zLNNgMNBsNlNRFPr+++9V17WJX0dRpG+//VZBEOjy8lKHw0GHw+GDzu95nvn3Q95fFIV6vd5HjexKL/dWv983z4WyLE1E9mPoArvD4VCO4yjLMjVNo9FopPF4LNu2dXt7q+12q6ZpZFmWmWNVVe/0XHEcR2EYyvM8pWn64PBtJ4ois6e7degi2kmSKE3TeyHl7uez2Uzn5+fq9XpyHEer1UrffPONdrvdvfE/5DnZBYW7eHDTNJrP58qyTLvd7ifncHdMjx8/Vr/fv7dfq6oy/w7DUH/zN3+j2Wym77//Xs+ePVOWZWYtXNf9ZM8aAAAAAAAAAAAAAAAAAAB+a4j5AgAAAAAAAAAAAAAAAAAA/JVt23IcR3Vdv1d41Pd9PX78WL7v6/r6WpvN5pMFFD3Pk+d5Wq1WWq1WH+U8nucpiiI1TaPj8fhZo75VVSlNU/m+/8nOEUWR/vCHP8j3fTVNI9u2P+qcXddVFEWK41iSTCh4MpkoiiIVRfHRznWX53maTqd6/PixifYOBgN9/fXXur291Z///GdZlqU4jmVZlk6nk3a7nQ6Hg2zb1nQ6le/7WiwW2mw2PxlSdRxHm81Gk8lEYRgqCAL9zd/8jWzb1n/5L/9Fy+XynWOs7zLX8XisOI6VZZlOp5PqulZRFOae6CLbk8lERVFovV4rSRJNJhP98MMPH3zvVFWl7Xb72hwdx5FlWW+ce7fP7kZhuxjw++5B3/c1Ho/Vtq0Wi4UZ26u6yG7btu90XRzH0WAw0HQ6lW3bOp1Ocl1Xo9FIk8nEjMH3fe12O/X7fV1cXMi2ba1WK93e3up0Opnj2bYt13XVNM0b187zPLVte+8zP8W2bQVBoPPzc81mMx0OByVJIs/zFMexzs7OFEWRdrudOYck9Xo9XVxcaDabmXUbjUZyXVfL5VKn00m2bSuOY4VhaKLMp9NJWZaZiHU3hl6vp+l0ql6vJ8uyzGcHg4H+9Kc/Kc/zn7zO3Tym06kOh4O+++47nU4n9Xo9jcdjHQ4H80x5/Pixer2e/uVf/kXX19cqy9Ls6zRNZVnWrybm+6H3CQAAAAAAAAAAAAAAAAAAH4qYLwAAAAAAAAAAAAAAAAAAgF4GGLvA5+l00mazUZ7n73QM3/c1m83UNI1evHjxXkHgh7AsS0EQyHVdrdfrd4pd/hjXddXr9UxA83PGH8uy1O3trWzbVtM0Hz3c6Lquzs/PNRqNtFgsTJDzY7FtW77vazAYKAgCFUUhz/Pk+76iKLoX//yYupDoo0ePNBgM1DSNgiDQbDaT67q6uroy46iqSlmWKQxDFUWhsizV6/XU6/W02+0eHKOuqkrH41FnZ2d68uSJCcr+6U9/0nq9/mQhX0n35lJVlQlRd0HuLtKcJImGw6H2+71Op5POz88lSdvt9oPO37atCQffvd89z9NkMlFZllqv1+Zat21rAr9t28rzPAVBoH6/L8/zlKapNpuNyrJ857H0+31NJhPlea7D4aAwDO+Fcm3blud56vf7SpJEm81Gm83mR4/p+74J2Pq+r0ePHqnf72u32ynPcxOr7v7f6/X01VdfSXoZy5ZexqvPzs5UVZXyPDfr1MVyD4fDa/eD67oKw1CS3il6/ejRI11cXOjrr7828+3CxUmSaDAYmABuN744jnV5eSnbtvXDDz9ovV6rLEuNx2Odn5/L931dXFyYNbgbfM/zXNvtVrvdTkVRyHEcDYdDnZ2dqW1bLZdLE9N+/Pixqqp6UCA7CAJdXFxoMplou93q+vpaWZaZ/dM9u7r9O51OlWWZFouFyrJU0zQmFP2pvocAAAAAAAAAAAAAAAAAAPitIuYLAAAAAAAAAAAAAAAAAABwRxeOfdfQqmVZCsNQQRBou92qKIpPEmuVXgY2e72efN9/r+jn29R1rcPhIMuyPtoxP0Tbtp8sKByGoc7OzpRlmQmRfsy1bJpGRVFouVzKsizVdS3XdTUYDOR5no7H4yeJbLquq9FopOFwqN1upzRNTRi1qiolSaKiKLTf73U8HuW6ruI4NnMfDoeybdtEbx+yh8uyVFmWiuNYf/d3f6fnz5/r22+/fXAM+EP4vi/XdZVlmVnPpmlM+FT617Cu53l69uyZLMvSaDTS4XDQ8Xj84DG0bftanLULxL548eLe603TmBisZVmaz+d6/PixPM/T9fW10jR9r31o27YJz3ZB1+5ad+OJ41iDwUDz+Vyu6/5kyDhJEp2fn6vX68myLDmOo/PzcwVBoMViYd7juq5++OEHHY9HDYdD/f73v9fTp091Op307NkzbTYbjcdjE8KVXl6Ts7MzHY9H7ff7e+e1LEu+7ysIAmVZ9k7r0QWd8zzX999/r6urK+V5Ltu29eTJE9V1rR9++EFXV1eq61r9fl+PHz827z8cDmYfFUWhfr9vQsSHw8FE3tu2VRRFGg6HkqQ0TVVVlfr9vi4vL1XXta6urnQ8HmXbtsIw1Gg00r/8y79ot9v96H1l27Zms5mm06n2+72urq7uhXxt29Z0OtVutzNjPR6PGo1G8n3/wfftl+rXPDcAAAAAAAAAAAAAAAAAwJeBmC8AAAAAAAAAAAAAAAAAAPhNsizrXhSwbVudTieVZammaV6Lc/4Uz/PU6/VUVZWyLHvnzz+UbduKosgENt90nlfn9lBlWZr46qeOsH5uo9FIvV5P6/VaeZ6b+X7MkHFVVfeuTxiG6vV6sm1bp9Ppk8R84zjW2dmZbNvW7e2tiqKQZVk6nU5yHEeO45iIbdM0JjRaFIWCIFCSJMqyTEVRyHEc+b5v9lNZlm/d147jaDwe67vvvtO3336r9Xr9Seb3qjAM5bruvbG5rivP88y9HASBptOp0jTVfr9XFEWKokgvXrxQlmWfZFxdPLcLv97l+76KolAYhnry5Ilms5mePXumm5sbpWn6XucLw1BfffWV9vu99vu94jhWkiQqy1Kz2Uy2bct1XTVNI8uytN/v753Ltu1716vf7+sPf/iDPM8zUegwDGXbtizLUpZlchxH/X5fRVHocDioLEulaSrHcZRlmRaLhTnn4XAwwVtJGgwGGo1GWq/XKsvytTXqzlMUxTs9S1erlSaTiU6nk66vr7VYLNS2rbnvsizTzc2NTqeTkiTRxcWF6rrWt99+q8PhIEkmmus4jglj/+M//qO5n7o17K5xt98cx9FsNpNlWbq6ujJh4C7iLUm3t7eqquqtz+fuPprP58rzXNfX1ybOa1mWgiDQcDhUWZZarVaSZM6XJIl+97vf6bvvvtNut/tk30EAAAAAAAAAAAAAAAAAAPzWEfMFAAAAAAAAAAAAAAAAAAC/Sa/GFNu2fS2++lBdZDEMQxO0/FQhU8dxFIahfN9XnucqisL8zPd9RVEkx3F0PB6V5/k7Hbtt219kxNf3/XvB1vcJFd/luq5ms5kkabvdmnDnxwz5vonv+wqCQEVRmEDnx+Q4jgaDgfr9vtbrtVarlZqmUZ7n+uGHH+Q4jsqyVFEU5tzd3j0cDiZOWhSF+v2++v2+WRvbtrXdbrXdbl+7R/r9vkajkSRpuVyaUPDPIQgCszfqujZr4HmeNpuNmqYx83r+/LmqqtJ8Ppfv++bnrx6vi8i+7xwsy9JwOFTTNG+M83Yx3+l0qvl8rvV6re+//17H4/G99oRlWZrNZprP51osFvJ9X5eXl3ry5Il2u52Ox6M2m42Ox6PCMNTZ2Zmurq5M6DmKIiVJotvbW7VtK8/z9Lvf/U79fl9//vOfzTqdn5+raRqdTidlWSbP8xRFkfb7vaqqMnspiiL95S9/0fX1tVz35a+rl2Wp9Xqttm3l+77m87mKolCapuaZEwSBer2e2rY1keamad5pTVzXVRiG2mw25tiu6ypJEtm2rd1upzzPFYahptOpXNfVN998o+Px+Npxzs/PNZ/PdX19reVyqSzL7o2lu7e6CHYXdF8ulybk261JkiS6urrScrl863y6QPBsNlPbtloul0rT1Ly/iydPJhNzXbs9utvt9M033+js7ExPnz7VYrHQarVSnufvvIYAAAAAAAAAAAAAAAAAAODHEfMFAAAAAAAAAAAAAAAAAAB4g3eJH9q2rTiO1ev1XgtUdhzHke/78jxPZVmayKL0MuL40PN1x7FtW1mWqWka2bYt3/c1nU41Ho8lSS9evLgX87VtW0EQKAgCtW1rQsB3x3B3HK7rmrGWZfngtfgYHMeRJDO20Wgky7JMNPRDBUGg+XyuNE11OBw0Go3Utu1br4HjOAqCQL7vq6oqnU6nt0aPu/dKUlEUZrxdNNX3fR0OB2VZ9sHzeFUYhhqPx3JdVzc3NyYk2wVY37TPuvHmea7hcKggCBTHsaqqUpZl92Kltm3rcDjcuwb9fl9PnjxRmqa6vb2VbdsfNeRr27bCMHxjFLebs/Ryreu6VhRF6vV68jxPWZYpiiKdnZ2ZOY1GIz158sSsyV2e5+nrr7828eO7oex34XmeZrOZVqvVa/u1i9vatq3xeKwsy3R1daXdbvdeIe0uHPz1118rCAKNx2MFQaDpdKosy/Ts2TN9//33Op1OchxHw+FQlmWZ9QyCQBcXFxoMBlqtVmrbVmdnZxqNRnr27JkWi4XKsjTPrqIotN/vVRSFBoOBmqYxEWLf9zUcDiVJNzc3Wi6XZpx1Xauua1mWpV6vpyRJdHNzo6IozPPz/PxcSZKYGHoXWH/o89G2bSVJItd1tVwuzfOv279VVWm328m2bY1GIyVJosVioe12e+/+737+u9/9Tk3TaLFY3Atg31377pqORiMTld5ut2aurutqMBgoDEO9ePHiR+9713XNPbjdbrXf782e6OY2n88VBIG+++67e3urLEstl0tVVaXZbKbz83MFQWCeA7+moK9t25L0xufMu3yXAgAAAAAAAAAAAAAAAADwvoj5AgAAAAAAAAAAAAAAAAAA3GHbtlzXVV3X92KJvu9LehlNfFPUMUkShWGozWZjIqBdIDcMQ52dnenp06dyHEe3t7f6/vvvtd/vfzQi+7bx+b4vx3HUtq2CIJDneUqSRL1eT5LU6/Xkuq6JTfq+rziOlSSJgiCQbdvabrfabrcmeBlFkVzXVVEUJhBqWZbW6/VPxny7wHDTNCrL8oNjroPBQHme63Q6yfd9TSYTE8js1spxHDmOYyKhD9GFHpMk0Xg81vX1tU6nk87OzkwM9tWxd4HUwWAg13XVNI1ub2+1Xq/Ne7ugbNM0Oj8/12AwkGVZ+uGHH7TZbNQ0jSzLUhzH8n1faZq+dyj2bRzH0WQy0dnZmbIs03K5fG0ub9pnXaQ1yzITHz0cDtpsNkrTVGVZKooixXGs0Wh0LxIdhqGePHkiy7JMXHQ6nZp99KEsy1K/39d8Ptef//zn1+bThXGzLDPR1y5oPBwOzR5+/PixuUd931e/39e333577/52XVeXl5d6+vSp1uu1Cc2+K9u2NRwOTVD21TWPoki2bavf7ysMQ61WKy0Wi/eOVIdhqN/97nf64x//qMPhoOvra6VpqrZtZdu2uY5N0yiKIiVJYu6tLkJ7eXmpPM9l27aiKNJ8PtfhcDAhX+nlPun1eirLUrvdzryW57mOx6Okl/fKYDBQlmU6nU6qquq1a+Z5nsbjsZqm0eFwUNM0SpJEs9nMhHi7Z0nbtu8UofU8T4PBQEVRaLvdqixLsydc11WapjqdTorj2LxvvV7fe6509+njx481mUxeu9c7lmWZOHr3TO3Om+e5LMsykeLpdKqiKLRYLN76bOwix8PhUGVZ6nA4mPvMcRwTO57P59rv99rv9+Zz3dibptF2u1VRFDo7O1Ov11O/3zfx+F+Tu3ui+5571+9SAAAAAAAAAAAAAAAAAADel/25BwAAAAAAAAAAAAAAAAAAAPBL0cUs+/2+PM8zr7uuq9lspjiOTTjw1c91n8nzXFVVqW1bE4a8vLw0QdE8z5Ukifr9vgkEv+sYgyBQFEUKw1BJkmg4HCqKIu12Ox0OBzmOo6qqzLi6wGwX2x0MBiYqKr2MYI5GI43HYxPknM1msizrXmTUtu3X5h8Egb766iv9h//wH/Rv/+2/1WQyMZHV9+G6rsbjsYlUxnGsOI6V57kJi9q2rel0qsvLSxMwfogucDqbzRSGoQ6HgzmH4zgmHNpFbeM41mw202w2U9M0StNUvu/r8ePH5to5jmPisZeXlxqPxyrLUmEYajAYmH3k+76iKDLHeWiA+CFs29ZgMND5+bnCMNRyuXxQiLaLjeZ5bmKjf/nLX0yE+HQ6qSxLE2juQqvd+ozHYyVJouvra223W202G1mWpSiKPsq8uthskiRv/Hm3tl001nVdE6Xu9m0XWv7uu++0Xq9VFIUOh4Ourq5MWNVxHI1GI81mM7Mub7rPHzrm4XCo0+lkYt2dLsra3W9dVLuL774Px3HU7/clSVdXV1osFkrTVGma6nA4KE1T8yyI41hBEGi/35tI82w2UxRFJu47n88VhqFub291Op3MesRxrCiKzHFd15XrujqdTiqKwswtDEPt93tlWfbWeHSSJErTVFmWmf9LL69jd91s21ae5+8UoQ3D0NzXp9PJ7Ileryfbtk2wejwey/M8bTYbHY/He4Hd7ueXl5dqmkY3Nzevvac712g0kud55rmRZZkJFHfrMRqNTOT9eDy+9Trbtm3em2WZsiwzx+kizL7vq21b7fd7sy53A7bd8/14PGq5XEqS2W9dWP7X4NXAMxFfAAAAAAAAAAAAAAAAAMDP7f1/UxoAAAAAAAAAAAAAAAAAAOBXxLIsE9l0XVdpmpr4Ya/X09nZmb755hu1bSvHceR5nqqqUl3Xcl1Xg8FAkkzM17Is+b6v2Wym+XyuzWajzWYj27Y1n8/V6/VUFIVWq9W9YO5DxtkFhxeLhTzPUxzHJuQbBIEkqSgK9Xo9PX36VNLL0OfpdNJgMDCx07vB0/F4rLZtdTweNRwOFYahjsejic52cdrtdqs8z++FirsIpeM4Oh6POhwO7zSnu7qxrFYr+b6v4XCoqqpUFIWJ8Q6HQz158sRcp4ewbVue52k4HOrp06dq21a+72s8HptIaXdde72eoigyodEsy7RYLCS9jKdeXFxoMBioKArFcazhcKimaRTHsYng3g1wVlVl/n06nT4o3vqqbj+cnZ1pOp0qyzJtt9t78VHbtt8YvOz2TlmWqqpKp9NJp9Pptfc5jiPbts017a79cDjUbrfTer020d8u0PoxdOfowqbdXCzLUtM0CoLAjKv7f3ctb29vZVmWZrOZrq+v9fz5c+V5bmKzu93OHK/f72swGKiqKhM2fjXe+i5jTpLE3Cd3ua6rIAjMvdRFZz8k7BwEgeI41vF4NLFaz/PkOI6yLFNZluZZ1IWvt9utJKnf7ysMQ1VVpePxqDAMNZlMlKapttut2QfdGrmuq81moyzLNJlMZFmWTqeTiQUHQSDP83Q8Hk3Q/FXdfbDdbk0cuq5r7fd7s8/atlVd10rT1AS8H6LX68n3fa1WKxOz9jxPnuepKIp79+XhcNB2u31t7R3H0Xw+V7/f17Nnz7RYLMyzp9M9N7rwruu6CsNQRVGY74Moiu49a7fbrcqyfOt97zjOve+QpmnkeZ6SJDER8CzLzH1a1/W9e7qLkEsvY7dFUci27XthdYK3AAAAAAAAAAAAAAAAAAB8HMR8AQAAAAAAAAAAAAAAAAAA9DJY2cUsu3ioZVmybVvT6dTEGi3LUr/fN5HQLMsUx7F6vZ5Op5P5nOu6Gg6HmkwmKopCV1dXOhwOcl3XhBpns5n2+/2Dw7e2bcv3fRPOlaQoilSWpXa7ndq2VRAEyvNcruvq8vJSw+FQz5490/F4lG3b6vV6StPUhHodx1EcxwrDUHmem9hlURTK89xEb6fTqcbjsQ6HgznOo0eP1Ov1tN/vlaapiVG+L8uyFASBXNeV4ziKokij0eheHDiOY52fn2symWi/3/9kdNW2bUVRpF6vpyAI9PjxY11cXGi9Xpv1a9vWxEejKNJ0OlW/31fbtsrzXIvFQofDQY7jmJBmkiTa7/caj8caDofK81xpmmq1Wpkgqeu65poNBgPZtv3GyOuH6ALFo9HIHL8L8t4Nu3bB3rt831ccx2bub4p9duvXhUvLsjTX33EcLZdLEzttmkZlWcr3fVmW9UHx0C4ga9u2uVZdBFeS9vu9oiiSZVkmkhrHsYIgMDHaJEnUNI0Wi4V2u52SJFEYhlqv1yb2GgSBCfl2e7i7j9+V53maTqdyXfeNodgwDE1w2HVdc9+8bzjYtm2Nx2M5jqPvvvtO+/1eruuq1+vJ8zzt9/t78+z1esrzXIfDQb7vK4oiVVWlsiyV57kGg4F839fV1ZW597s9NBwOVZalttutqqpSEASyLMsEdx3HMdHguwHeLjArvYwZT6dTSS/Dsr1eT2VZ6nA4qGkac480TaO2be9FnB+yFv1+X3Vd37t+3fMkTVPVda04jmXbtjabjZnjq9doPp8ryzI9e/bs3jOmW4vBYKAoinQ8HrXf7xXHsXkuD4dD8/xvmkan00lhGD5oT3meJ8uyZFmWer2eec4XRaE0TU2Y903H6T7Xxb1ns5niODZhY0K+AAAAAAAAAAAAAAAAAAB8PMR8AQAAAAAAAAAAAAAAAADAb14XX+z1eoqiSLvdTnVdm5DpeDw2YckoivT48WN99dVXkqSiKHR2dqbJZKLlcqkoihRFkQlt+r6vFy9emCBtXddar9caDAaaTCaK41in0+lB4/Q8T/1+30RMu8BqFyeNokie5ynPc00mE81mMy2XS202G7muq/F4bAKPaZqa+G+v15MkVVVlQqNdtNNxHPX7fc3nc1VVpaZpFIahzs/P1ev1tN1uTTi1i1a+GjF9l+sQhqGJDEdRpCAItFqt1DSNfN834VpJJi77Y8fr9Xp6/PixwjBUVVUawygDLQABAABJREFUj8eqqkrff/+91uu1xuOxiRs3TaM4jjUajczcNpvNvZhwXdeqqsrEQ+fzuaIo0uFw0O3trdI0VZIkJugryQSSq6q6d6wP5TiOBoOBBoOBHMdRmqba7/fKssyEVeM4Vl3XJup6VxAESpJEh8PBhEBfjX56nmfCvZvNRkVRmPuirut7MdqqqsxxbNt+730gvQwNn52dqaoqHY9HEyjtxlLXtYIgUF3XJo7chVqLojD7pWka7XY7lWVpItH7/d5Eqvv9vnnP5eWlHMcxoet3Ydu2hsOhZrOZDoeDdrvdvZ9blqXBYGDWM8uye/fh+67RkydPlKapnj17pqqqFIaher2ebNtWnudmr3UR4cViYWLUTdOYa1aWpc7PzyW9DCV383ccR71eT0mSaL1emxBwEARqmsZEciWZ0LPv++Ya2bYtz/MUBIEkmVBud8+kaarT6STbtuU4jglgd4Hhh16HbpxdzLrbk1EUSZLyPJdt2wqCwNyHd6O43Tj7/b5c19Vms9FmszFzcF3XxJh939fxeNThcDDfC2EYKgxDs5/SNDXh8+45/WMx37Zttd/vNRgMNBwOTWS5CwZXVSXP87TZbMxadde228d3I/L9fl+73U6bzeZHn5EAAAAAAAAAAAAAAAAAAODdEfMFAAAAAAAAAAAAAAAAAAC/eV2AMooitW1rAptdyDaKIl1dXcmyLA2HQ3311VdKksREZS8vL2VZltI0VVmWJtiZJIlOp5M2m40J4Uq6F3P1PO9BY7Qsy8RXXdc1Qdn9fq/j8SjpZbCzC6nOZjMVRaGbmxu1bavxeGyixF3ctPtMF7yUpDAMTYyyC552EeDnz59LkiaTiYbDobbbrRaLhQlddgHLD4n5dtFP13U1GAxMrNW2bSVJYkK5p9NJaZr+ZBg3SRJNJhPlea7T6STHcXR9fa1nz55pt9tpOBwqyzITVO7O0batttutdrvdvXO47stfv+3WdDqdKk1TE/2t61qe58n3fR0OB7VtqziO5fu+siwz0eCPoQsPu66rNE3lOI4Oh4OJqoZhqDiOXwuX3l2bIAi02+1kWdYbzxFFkXq9nuq6NiFi13XNPnMcx4SLq6oy1+9tx5Nexke7gOybxmVZluI41nQ61Z///GfVda3JZCLHcZRlmWazmZlfVVUqisKEoLu96ziOWfPD4WDOK8n8fDQayfM87XY7E2Hu7v93vUZhGGo2mykMQ3333XfKsuzezz3P03A4NJHqxWLxwaHV4XCos7Mz/eM//qO2260JGHfrUJaleU4EQaAwDFUUhYk8H49HhWFo4rlJkqgoChNHdhxHSZJoOBzK8zylaWqi4d2e68K5lmWZe7KLGhdFIdd1zX7JskxBEKgsS9V1bcLTXby7C1BXVWVitg8VBIFs29Zut1NRFGrbVq7ryvM8c6xuHEVRqKoqtW0ry7LMNenu08ViYZ59cRyb55vnearrWqfTSafTSXmem/2yXq/V6/VkWZbyPNd+v1dRFArD8EEB76ZptFgszPUqy1KHw8Hcz5LuPVvrujYR6Lvhc8/zTHi7i0V/rOcNAAAAAAAAAAAAAAAAAAB4iZgvAAAAAAAAAAAAAAAAAAD4Tetijndjl13M0vM8DQYDExH1PE/z+VxxHGu1Wul0OikMQ7Vtq+Vyqevra223WxPdlaT1em2Cl118sQtFNk3z4PBtF7p1XVe73U6n08kEQZumked5iuP4Xtjx+fPnKopCw+FQk8lEVVWZcbdtayKfvu+b+KXv+yqKQqfTyQR0Z7OZqqrSbreT7/uazWaq61qr1Up1XWs6ncp1Xa1WK2VZ9t7xSMuyFEWRiZImSaLj8ag8zxVFkYbDoYlldhHNn4pklmWp5XKpw+FgwsnX19fa7/fmPHVdmwBuv99XEARaLpfa7Xb3oqy2bSuKItm2rbqu9fjxYxN63m63JhDaRTXzPFfTNArDUNLLGGeWZWYffAjf9zWdTk1E2fM8eZ6n4/Fo1sRxHIVhqDzP5bruvTiqZVkajUaq61pFUci2bTOvjuu6Gg6HiqJIm83GhIibptHpdFIURRoMBmY+3R4sy/Ktc3QcR5PJRG3bar1eqyiK195jWZa5DpvNRlEUyXEcrVYrSdJsNpNlWUqSROv12lw713VNaHU8HisMQ93c3CjLMhN2lV7uieFwqMFgoMViYWLYvV5PeZ6/8x52XVfj8ViDwUBZlmm73d5bx7vz6V7fbDavhVa7mK3v+z8ZfXZdV/P5XG3b6vb21uzTbi0kmWvQRWm7YG6/31fbtua51J3Htm3znPI8z9xzcRybZ0LTNHIcR1VV6XQ6mT3V3Y9XV1caj8c6Ozsz+zBNUxM6v76+VlEUJjRb17V5BndjkKSiKB58DRzHUa/XU5ZlJkIu/WsoPcsyVVVlYrx1XZv94rquCa+3bavj8ajnz59rMploNBqpLEuzbl2Mu3vudOcpy9IEoSWZGHEXrH5bTPuupmlM3NhxHPNMunuePM/Nd9Pd41VVpf1+b6LnWZYpy7K3xrIBAAAAAAAAAAAAAAAAAMCHIeYLAAAAAAAAAAAAAAAAAAB+cyzLMv/uIqZBECgIAh2PRxPE9H1fw+HQRBSTJNF4PFZRFLq+vtZisdB8PldZltput9rv96rrWmEYKgxDZVmm/X5/L8govYxP+r6vqqreGDN9ky44XFWVrq6uTJS3LEsT+h0MBiZyejgcdDgclCSJ5vO56rrWYrEwEVvLssznuvBsF7fc7XYqy9JESofDob799ltlWaZer2ciqE3TaDAYKI5jpWlqPve+HMcxAd8wDOV5ng6Hg6qq0mQyMbHWIAjUtq3yPP/RGHLbttpsNjocDmrbVn/84x9lWZZWq5WKopDruuY6VVVl1rBpGq3X69eCqt34uvjtfD7X4XAwwdi78dS70eE4jk089W3r47quoij6yYir9K8h3sFgoMPhoM1mY9YnyzITSO32RpIkKstSp9PJ7EXHcTQej7Xf75WmqVmLsizNmg4GA43HYxPe7UKidV1rt9spiiJFUSTP89Tr9VRVlQlfv20OnudpNpuZe+ZtazEej8084jg218PzPLVtq36/b+Lb3bUrisKsXxeo7vb73bVLkkT9fl95npufd/fBu+7hLkA9nU4lScvl8l4AWpKCINB4PFae54rjWKfT6Y3nuXu/PXv2zESG3ySOY00mE11dXZlnjGVZKoriXoC2e7aFYSjbttXr9VTXtZbLpQn3nk4nOY6jNE1NDLz7W3oZ1j0cDiqKQm3bqixLrVYrE6lt21ZN0yjLMl1dXSnPcxM5z7LMxMfbttU333xjQsB3790u4iu9jNO+S1DZ8zyFYWiCtm3byrIsE+/t7rtuPo7jmAi17/smtr3ZbLTf7+W6rolwW5ZlxtuFxV995nRr8ur17F7/qeB4996qqsyel3QvAN/9/0260PrdYwEAAAAAAAAAAAAAAAAAgE+HmC8AAAAAAAAAAAAAAAAAAPjN6qKJjuMoDEMTi+2CimEYKo5j3d7eqm1bDQYDeZ6n7Xar7XarsiwVBIEsy1Kapibi2YUz0zQ1AdTufJZlyfd9JUmizWaj0+n04LG6rqu2bXU4HEx0uG1beZ5nAqWn00lRFOnFixeyLEuz2Uy2bevm5ka73e5eWLiL+TqOI0km5ns6ndQ0jaIoMoHgu5/tQsez2Uyj0cjEN1+NFr+rLmi7XC41Go0kScfjUa7ryvd97fd7FUWh+XyuqqpUluVPBj+rqlJVVSaUmue5drudmqYxodMuSpokiXq9nvb7vdbr9b3QchcHDcNQt7e3sm1b/X5ff/rTn7Rer02ws3ufJOV5bvZRnuevBUw7tm0rSRJ99dVX+tOf/vSTeyIMQ00mE+V5ruVyaSKid+PQXUh1s9loMBhoMBgoiiKdTicTWw3DUN9++632+72ZW3cMx3E0n88VBIFub2+13W7NWncB2JubG/V6PSVJoiRJTDD6eDy+cR9YlmXev9vt3vqeOI41n8+VZZlGo5GWy6UJMndx1fl8rqZpTLy3O3eWZWb99/v9vbBut05dkPvm5sZcjyAIlOe50jR9pz3suq56vZ5831eapiZyfffnw+FQYRgqTVONRiMTt31173bH6vf76vV6b435Wpalfr8vy7L07bffmshud23yPJfv+5Je7i3HcUws1/M8E611HMfs0aZpdDgcNBwONZlMzP65O4aqqsyaF0VxL7jbxWi7oG4XXe4it9377o71rqZpTHB6t9u9dQ+9TZqmKori3v1lWZayLDMx327MXQi5e17lea7tdqs0TdU0jaqq0mKxMFHuLpD80LjwXT8WG3+T9zkH8V4AAAAAAAAAAAAAAAAAAH5exHwBAAAAAAAAAAAAAAAAAMBvThcF7f52XVdhGEqSCTo6jqM4juW6rg6Hg3zfVxiGJmTaRUOjKFJVVTocDiqKQrZty/M8E718NQLpOI4Gg4GCINButzOB0YewbVt1XassS9V1bSKOXTC0i3J24dsuNPzDDz9os9ncC2natm3mLskEb23bVlEUcl1X/X5fruuayKb0MlB7OBw0m800n8/Nz7o45nq9fqc53eW6roIgUFVV6vf7JtTaBYO32608z5Pruibg+tCQped5GgwGur29VZqm5nxdHFaSuS4vXrx4LSbaBYyzLNNut9P5+bnKstTt7e296K/neSZkWte1giCQbdvKsuxe2Pku27YVx7Fms5meP3/+ozHfIAjuBZqPx6OCIDAx1Lvx0G7NqqrSaDTScDhUVVU6nU4Kw9DEfo/Ho3q9nqIoUpIkJvLseZ6Wy6Vubm7MGnXqutbhcDCx1C762sWC38R1XSVJIkmvhVfvrkWSJPJ9X3memzHejdV2++329tYEkrv17WLAXZy2G3dd1zoej9psNsqyTDc3N6+Nc7FYaLPZPDioejfwXFWVdrudDofDvZ93ke3dbqe6rk3wNsuy1/ZuXdfa7/fmfW/T3bvX19daLpdmHdu2NTFa6WUYtovsdjHk9XptAtDdMyrLMhVFofV6bcaUpqkZc/dsbJrGfKZt27dGebMsuxdRvutt92tZllqv1zoej69F0H9KnucmbH73+FVVqa5rFUVh1qB7v+M4KorCnO/uM6ttWxPw/bExf2xEeQEAAAAAAAAAAAAAAAAA+DIQ8wUAAAAAAAAAAAAAAAAAAL9pXdDW8zwTBZVeBjPDMJRlWcqyTL1ezwRv8zxXXdcmTHo4HHQ4HFTXtVzXNYHcoijuhUEtyzJR4NvbW+12uwcHHOu6Vp7nJkDbHde2bUVRpMlkot1uZ0LEXdj02bNnJsrZvb+bs2VZ9wLBXbSzC6p2wdckSWTbtmzbNvMdj8daLpdaLBaybVuz2UyPHj1SWZbabrcPDqLeXZsgCNS2rVzXNbHNwWCgMAx1fX2tqqoURZFZj4eunWVZ6vf7sm1bi8VCdV2b85VlqTzP5fu+hsOhicXeDc12YeckSbTb7VRVlcIw1Gq1eu0aBkFg1k2S4jhWVVU6Ho9vDbR2oeCmacwav0kXbe71eubcTdPIsiwVRXEv1iy9jIOWZan9fm/GNh6PNZlM1DSNVquVCU53sdd+v68gCJTnuVarldbrtYkfv6o7/na7Nef4set+N359Op3e+F7LsuQ4jo7Ho1arlW5vb+/NqYuzdmHeu8HVtm3N+nVB2e6zdV1ru92acOurwenT6aTvv//+tXv2x3T3keu6Op1OZm90uvuybVutViv5vq+bmxsTQX5VURRaLpdarVY/Ooa6rnVzc6ObmxsTse2UZamqqkzEt2kapWmqFy9emHW5u0/2+70sy1JZlloul9rv9yZm2+2t/X5vgtRd0Pdjq6pKm83GPNveJWzbhbNffa2LYnfjLctSm81G2+3WxNx/bC7EdQEAAAAAAAAAAAAAAAAAwJsQ8wUAAAAAAAAAAAAAAAAAAL9JbdvKsixJMkHOqqpM5NN1XcVxbAKh/X5f+/1ejuNIksIwlOd5KsvSBEK7GGsX8+2OL/1rsHUwGEiSbm5u3hp3fdt40zSV4ziybVuO46iua4VhqMlkItu2tdlslCSJ8jyX4zhar9f3oqWu6yoMQ/m+b87dNI3yPDcx1KZp1Ov11LatCY62bavxeCzP85QkiZIk0XfffaerqytlWWbW7+LiQuPxWGmaKs/zd7oetm0rCAIdj0czt269ugBqFxftArwP5bquRqORsiwzsVLHcRSGofI8V1mWiuNYYRjqeDzqeDya/dGFfCeTicIw1HK5NPNdrVavjSMMQ9m2re12q7Zt5XmeCcu+LRzazb0Lrb7tPf1+3+zDuyHbuq6Vpulb91Nd19rv96qqSnmem7W4ubkxYdeyLLXb7cwe62KwD92jPxV4tSxLvV5PYRjqdDqZ6/mmsS6XSx0Ohzfuoy76ejweTUD71XGkaaosy1772Zsivp0uhvyuunvxdDqZeKz0Mlx8dnamKIr04sULM6Yu6vzq2O7O/yHeNtbuOdTdy12Y903vb5rmtajwq9e7bVsdDgfz708ZuP3YkeA3He9TzwEAAAAAAAAAAAAAAAAAAPz6EfMFAAAAAAAAAAAAAAAAAAC/WV2s1bZtFUWhNE1NkDcIAhO9HQwGyrLMxH7jONZ0OlVZlqqqyvysO2ZZlrIsS7ZtmyhsEASaz+eaz+e6urpSmqbvNNYuylnXtXq9nuq6VpZlGg6HGgwGWiwW2m63CoLABFvvxk6DIFC/31ccxyqKQpvNxkRzu9CsZVlyXVe9Xk+bzUbb7VZVVWm73Wo2m2kwGKgoCt3c3Gi5XOp0Opno636/12g0UhzH8n3fRGIfqouibjYbtW1rYsnPnj3TbrdTVVWyLEuWZZn17v7/U+cJgkCDwUDb7VZpmpqYbxAEKstSdV2r3+/Lsizt93vleS7P82RZluI41nw+VxRFWi6X2u/3GgwGJir7ppjszc2NdrudpJch4SzLTDD5Vd2aB0Ggw+HwxuhqF8Lt9txms7m3vt2++LFQaV3XOhwOyrJMV1dXappGZVnee3/TNO8VtH0I27Y1mUzkOI4Wi8W98O1dXTy2C8i+yWaz+dFz1XX94Cjuh+r2Up7n5hlg27bm87kmk4l++OEH7XY7NU2jpmneKeD9Prp7sWkanU6njxLI/diRXQAAAAAAAAAAAAAAAAAAgC8ZMV8AAAAAAAAAAAAAAAAAAPCb5jiOXNdVVVUm9GrbtjzPM8HXOI714sULua6rsizV7/eVpqmKolBRFMrzXE3TmJhqWZZyXVeO48hxHHmep/F4rMlkov1+r9vb2/cOZEZRpNlsptFopM1moyRJVNe1Xrx4odPppOPxqDRNdXZ2ZoKeXZTW933t93vtdjtlWaYoipRlmdq2Ne+VpNPppO12q+PxaGK3Xfg2TVOlaaqyLM2YmqYxMdMuyvuQyO5dtm2rLEvd3NzI8zyt12sdDgc9f/7cjLFtW2VZpuPxaOLDDzmP7/tyXVeLxcKM2/d9xXGsNE3luq5Go5GklzHUXq+nXq+nIAgURZEJ0N7e3pro87Nnz+7Fkjv7/d6s23g8VtM0JhD8JpZlyfM8SdJqtXpjTNfzPD169EiWZenm5ua1SOu7xGurqvrkQdk3ieNYQRBovV6/dZ5fmm5Pep6nfr9vAtmz2UxPnjzRixcvtFgsftb1bppGP/zwgyT9bEFj4FOwLEuSzPO9e9Z3r9/9GQAAAAAAAAAAAAAAAAAAPydivgAAAAAAAAAAAAAAAAAA4DetC8J2Ud4u0CnJhGnTNNV+v1ccx1qv12rbVovFQp7nqSxLFUVh4qpdWDWOY43HY9m2rSiK1O/3tdvt9MMPP7xXyLRtWxVFYcK2o9FIYRgqTVNdX19rt9upaRptNhtdX1/r0aNHury8VF3X5nOr1UrL5dIcI89zrddrM/Ysy+S6rgn2dnPpQrRd9Lf7992oYhfJrarKhI3fRVEUur6+Nuu+Wq1U1/W9aLAkHY9H1XUt27bN9XrIsW9ubrRYLMz4oyiS67rK81xBEChJEp1OJx0OBxNx9n1fZVnq6upK6/XarEOapjqdTm+Mpd69to7jmMDyq/O4+544jpXnubbb7WvzsSxLo9FIjuPo+fPnOhwO7x2C/pz6/b4Oh4Nub29NNPrXoKoqnU4nTadTVVWlMAwVhqFevHihFy9efJZwMhFf/Nq8GvB911g8AAAAAAAAAAAAAAAAAAAfEzFfAAAAAAAAAAAAAAAAAADwm9WFe6uqUl3XJnJb17XSNNXt7a0k6erqygR7v/32WzVNozzPNZlMVJblvYBtnufabDYKw1Cz2Uyz2Uyn00k3Nze6vb1VlmXvPd6yLHVzc6Msy5Qkidq21Waz0fF4VFVVsixLp9NJ33//vTabjZIkkWVZOh6P2u12yrLsXmg3z3MVRWHiiN3Ymqa5F+29Gzh+df06lmXJdV3tdjuVZfnOocXuOvyUuq51PB7f6djH49Fct25cbdtqtVppv9/rD3/4g1zX1WKx0IsXL9S2rVz35a/Z5nlu1q3z0JhuURQ6nU5mX72J67qKouitkVvf9xVFkb7//nsdDocvMmDpuq4sy9JqtVKapl/kHN6kbVuVZanb21s1TaMwDM09ut1uieoCH+Duc6L7jrob9QUAAAAAAAAAAAAAAAAA4HOxfi2/DAsAAAAAAAAAAAAAAAAA+OWyLItfVsMvlud5CoLAhHjrupZlWXIcR1EUqa5r5Xmutm1l2/a9wGCSJLJtW1mW3Yviuq6rOI4VBIGaplGapsqyTGVZfrRx27Yt6c1h2e5nnYfGZ7tQ4qu/X2pZ1k8GWCeTif7dv/t3+vOf/6zb29uPOtdPwfM8OY4jx3H0H//jf1TbtvqHf/gHXV1dPXi9forjOGrb9q3Hs21bcRyr3+/r+vr6je+zLEu2bX/RYVjXdeW6rsqy/KLn8TaO4ygIAtm2rbIslef55x4S8Kv0asz3bd9LbdtS/QUAAAAAAAAAAAAAAAAAfDLu5x4AAAAAAAAAAAAAAAAAAADA51RVlZqmkWVZJjTatq2qqtLhcLj33rqu78UDj8fjG2OtVVVpu92aY32sOOxdP3bMd4n33p3Pj4QRf/I4g8FAVVUpyzJVVfWg839OVVWprmvN53PZtq3b21sdDoePeq0eEq7N81xFUbz1vG3bfvEB3Kqqvog98b7qulaapp97GMCvxpvC8ne/rx4SmAcAAAAAAAAAAAAAAAAA4FMh5gsAAAAAAAAAAAAAAAAAAH7zuuBu27YmIti9Lr0MB3bxwLsBwaqq3hobvPv5X6qPNb4kSTQcDrVer3U6nX7x85Zezt3zPI3HY5VlqdVqpSzLftYxNE3zSULPAPAle1Os90v6bgUAAAAAAAAAAAAAAAAA/LoR8wUAAAAAAAAAAAAAAAAAAL95bwuq/lQw8Mdig78VrutqNBrJsiytVisVRfG5h/RgURQpSRKlaar9fq+qqj73kADgN+/VcD4AAAAAAAAAAAAAAAAAAL8kxHwBAAAAAAAAAAAAAAAAAMBv2kODvL+lsKBlWfI8T5ZlqSzLt8aOJWkwGGg6nSrLMh2Pxx997y/NYDBQHMdarVY6nU5f1NgB4Nfqt/R9CwAAAAAAAAAAAAAAAAD48hDzBQAAAAAAAAAAAAAAAAAAwD2O46jX66ltW9V1/dbIbRAEuri4kG3bWq1WyvP8i4kw2rat4XCopmm03W6V5/nnHhIAAAAAAAAAAAAAAAAAAACAXzj7cw8AAAAAAAAAAAAAAAAAAAAAvyy2bcvzPPm+L8dx3vge13V1cXGhwWCgxWKh7Xaruq5/5pG+P8dxlCSJ0jTVfr//osYOAAAAAAAAAAAAAAAAAAAA4PMg5gsAAAAAAAAAAAAAAAAAAPAT2rY1f34LmqZRXdfyfV9BEMi27//KqeM4uri40OPHj7XZbLRcLlUUxWca7fuxLEtpmur29lan00lN03zuIQEAAAAAAAAAAAAAAAAAAAD4hXM/9wAAAAAAAAAAAAAAAAAAAADwy9I0jfI8VxiG6vf7sm3bBG89z9PZ2ZkuLi602+10dXWlLMu+uNBxXdd69uyZTqfTFxciBoBfG8uyJOmDvkssy/rivosAAAAAAAAAAAAAAAAAAF8ei19WAwAAAAAAAAAAAAAAAAB8apZl8ctqwBfG930lSaIoiiS9jCS6rqs4juX7voqi0IsXL7Tb7VTX9Wce7fuxbVtt2xJ/BIDP7GOEeLtjtG1rfaRhAQAAAAAAAAAAAAAAAADwGmK+AAAAAAAAAAAAAAAAAIBPjpgv8OWxbVue5ykMQ/MniiLZtq3j8ajNZqPD4fDFhnwBAL9OxHwBAAAAAAAAAAAAAAAAAJ8SMV8AAAAAAAAAAAAAAAAAwCdHzBf4MlmWJcdx5HmegiCQ53mq61ppmqooCjVN87mHCADAPcR8AQAAAAAAAAAAAAAAAACfkv25BwAAAAAAAAAAAAAAAAAA+Lwsy/o/W5Z1Y1nWP9x5bWJZ1v/Hsqx/+evf47++blmW9X+0LOu/tyzrP1uW9T/4fCMH8Km1bauqqpRlmfb7vVarlTabjbIsI+QLAAAAAAAAAAAAAAAAAAAA4DeHmC8AAAAAAAAAAAAAAAAA4P8i6X/8ymv/O0n/Xdu2fyvpv/vr/yXpfyLpb//657+R9H/6mcYI4DNq21Z1XauuayK+AH4VLMv63EP4KCzLkm3bchxHtm3/auYFAAAAAAAAAAAAAAAAAMAvlfu5BwAAAAAAAAAAAAAAAAAA+Lzatv3/WZb19Ssv/88k/Vd//ff/VdL/V9L/9q+v/9/atm0l/SfLskaWZV22bfviZxougM/Msiy9fAQAwJfHsiy5rivXdVXXtYqi+NxDejDLsky81/M8eZ4n13XlOI4kqaoqlWWpqqpU17WqquJ5DQAAAAAAAAAAAAAAAADAR0bMFwAAAAAAAAAAAAAAAADwJud3Ar1Xks7/+u/Hkr6/875nf32NmC/wG0EYEsCXzLZt9Xo9JUmiw+Ggsix/0c+1V+O9vu/L930FQaAwDOW6rqqqUtM0KopCVVXJtm2laardbvdFxYrfh2VZ5t+/5OsIAAAAAAAAAAAAAAAAAPj1IOYLAAAAAAAAAAAAAAAAAPhRbdu2lmW9cx3Nsqz/RtJ/8wmGBAAA8F48z9NgMFCSJErTVJZl/SIjsJZlyXVdBUGgOI6VJInCMJTnebIsS4fDQdvtVqfTSXmeq2ka2bat4XCoP/zhD3r+/Ll2u93nnsYn17btvaAvAAAAAAAAAAAAAAAAAACfGjFfAAAAAAAAAAAAAAAAAMCbXFuWddm27QvLsi4l3fz19eeSntx531d/fe01bdv+t5L+W0l6nxgwAAAfk+M4kqS6rj/zSPC5WJYl3/fl+77SNNXpdFLTNJ97WG/k+77G47FGo5F831eWZVqv10rTVFVVKcuy1/ay7/tKkkSStF6vVZbl5xj6z+6XGGMGAAAAAAAAAAAAAAAAAPx62Z97AAAAAAAAAAAAAAAAAACAX6T/l6T/+q///q8l/T/vvP6/tF76H0ratm374nMMEACAt7EsS5Zlmf+7rqvpdKp+vy/b5leof6ts21av11Ov19Nut9PpdPrcQ3ojy7IUBIHCMFSWZfrmm2/0zTff6Pr6Wvv9XqfT6bWQr2VZiqJIk8lEq9VKeZ4TuQUAAAAAAAAAAAAAAAAA4BNwP/cAAAAAAAAAAAAAAAAAAACfl2VZ/3dJ/5WkmWVZzyT97yX9HyT9PyzL+l9L+lbS//yvb/9/S/qfSvrvJaWS/lc/+4ABAJDkOI7atlXTNPdej6JIYRiqKAqlaSpJiuNYf/u3f2tiqPjt6WK3vV5PVVX9YkO+nTzPtV6vVde10jQ1+9xxHNm2Lcuy7gV9XdfVeDyWJC0WC5Vl+VnGDQAAAAAAAAAAAAAAAADArx0xXwAAAAAAAAAAAAAAAAD4jWvb9n/xlh/9j97w3lbS/+bTjggA8Kl4nifP85Rl2WsR3C+Jbdt68uSJdrudttutiZqGYajf//73Go1G+vOf/6zj8SjHcRSGoZIk0YsXL+4FUPHb4TiOer2eXNfV4XBQnuefe0hv1batiqJQURTm/9LLYG+SJPJ9X7vd7t5eHg6Hmk6n2m632m63X/T9DQAAAAAAAAAAAAAAAADAL5n9uQcAAAAAAAAAAAAAAAAAAAAAfCls25ZlWT/LuRzH0XQ6leM4H+14jx8/1u9//3vZ9pf7a8TdPP79v//3evTokTzPk/Ty2oxGI52fn8t1Xe33e0kvA6hxHOt4POrm5uZzDh2fURAEiqJItm3reDz+4mO3bduaP5JkWZaCINB0OlW/37/3HPI8TxcXF3IcR8vl8jcbrP65ns0AAAAAAAAAAAAAAAAAgN+2L/e3cAEAAAAAAAAAAAAAAAAAAIC3eN+o4499znVdTadTnZ+ff/JopOM4evLkif7Df/gP6vV6H3w8y7LU7/f16NEjRVEk3/c/wih/frZt6/z8XH//93+vtm3luq4JE9u2LcdxtN/vdXNzoyzLZNu2oijScDjU6XRSmqY/eQ7Lsr7o2PH7chxHvV5P0+lU4/FYQRC89z63bVue58nzvM++lt31jKJIcRyrKAqdTqfX3tNFn2ezmYbDoVzXfesxbdu+Ny/LsuR5nnzf/2jx7TedM4oijUYjVVWlqqrMuWezmc7Pz7VarbTb7UwA+MeO5fu+giCQ67q/mgjuT80bAAAAAAAAAAAAAAAAAICP4e2/YQgAAAAAAAAAAAAAAAAAAABZlkUg8AtiWZbCMFQcx9rv9yqK4kGfs21bvV5PruvqdDopy7J7P3ccR5PJRF9//bU2m42ur68/xfCNJEn09OlTOY7zUUKbXdB0s9koTdMvdk/3ej199dVXchxHV1dX2mw2Jmrquq7CMFRVVdrv96rrWr7vazAYKAxD3d7eqmmaHz2+67qKokiSdDqdzLEf6m4IuK7rd/5sd73ruv7JsX5Mvu9rNptpNBopiiKVZanr62ut1+sH30PSy/Xr9Xoaj8eK41h1XWu9Xmu9Xqssy3vvs21bbdu+01zjOFYQBDoej2ZcXeTWsixlWfbGa+a6rpIkke/72m639+bk+76iKFK/31eSJJrNZlqv1/r22291PB7v3SuO42gwGCiKItV1rePxqKqqlCSJhsOh8jzXcrnU6XSSZVkmILzf75Xn+YPX8U26OXieZ/Z3FxH+3e9+J0laLBb3Ir/S64HbLko+n88lScvlUuv12ozPsqyfde8BAAAAAAAAAAAAAAAAAPClIeYLAAAAAAAAAAAAAAAAAADwFrZtazabybKsTx5vxYezLEtxHGs+n6vX66mu6weHSIMg0Pn5uZIk0c3Nja6vr02MtYtyzudz+b6vzWbzSWO4XTDUcRzd3t6+cxT2TWzbVhAEqutaeZ5/sbHO0Wgk13V1dXWl4/GoNE3NXDzPk+u6KopCaZqasPNoNFJZllqtVj95fNu2FYahWaeHsixLvV5Ps9lMX331ldbrtf7yl78oTdOf/JzneUqSROPxWEmSqCxLEyr+GNepiwQ7jiPbtlXXtcqyNHvYdV09evRI5+fn2u/32u/38n1fnuc9OCTdxbDn87niOFbTNMrzXG3bKo5jHY9H1XWtMAw1Ho8VhqFc11XbtlqtVtputyb263me4jhWHMdq21abzUZ5nisIAj19+lQXFxf6h3/4By2XS7muq7/927/V7373O11fX+uHH37Qcrm8Fw5u21ZRFClJEjVNo91up7Is793X0+nUxHlXq5Wapnlt7R3H0dnZmcbjsaqqUpZlKopCQRDo7OxMvu/rcDiYmG4QBHr06JGSJNFf/vIXFUXxQc8Nz/M0mUzkOI6Ox6MZ32Aw0JMnT/Sf//N/1uFwMNc7iiI1TaPT6WSeIZ7naTwe6+/+7u90eXmp3W4n13WVpqnZ71/qswEAAAAAAAAAAAAAAAAAgJ8LMV8AAAAAAAAAAAAAAAAAAIC3iKJIl5eXyvNci8Xio0RV8em4rqskSUwQ9V3CmV30NUkSZVmm3W6nNE3Vtq2JdY5GI93e3mq9Xn/CWbyMgHYx4t1u91H2XRAEGg6H8n3fBFrvhlo/ZZz4Y/E8T48fP1ZRFCrLUmEY3ptDFEXyPE+n00l5nstxHCVJotFopKurqzeGdS3LMnO3LEtN05jwbBdl/Sld5LXbIxcXFyrLUnEcK89zc/16vZ6iKJLjOErTVFVVKY5jDYdD2batsix1PB41Go0URZG22+0Hr5nruur1ehoMBur1erJtW3me6+bmRofDQW3bajgc6uuvv9azZ8/0/PlzFUWhyWSitm0ftPds21aSJLq8vJTnedrtdibA20Vl67rWdDrV+fm5XNfVZrNRmqYaDAaaTqcqy1KHw0FRFOni4kKj0UhhGMrzPP3zP/+zlsulCSV3191xHM1mM00mExPx7fV6r0WbbdvWcDhUr9fTcrnU6XQyoefLy0s9fvxYq9VK3333nYqikOM4sizLxIill3tjMpno/Pxcu91Oy+VSVVWpbVsNBgM1TaPlcqnVaqWqqmTbtqIo0mg0Utu25r0fch1Ho5H6/b4Wi4WJEXuep7OzMy0WC11dXcmyLDPWy8tLrVYrXV1dqW1bua6r8Xis8Xis1Wql6+trc/+0bSvbtj9JyPfuPQYAAAAAAAAAAAAAAAAAwK8BMV8AAAAAAAAAAAAAAAAAAIA3cBxHvV5Pnudps9l8tOPatq1+v69+v6/VavXGwCjenW3bCsNQvV5PTdMoy7IHx1ill6HYsiyVZZnCMNRkMjFRz9FopEePHqmqKi0Wi3c67vsYj8fq9XpK01SHw0FlWX7Q8SzLku/7SpLEREWbplEcxwrDUIfDQXmef6TRfzr9fl+Xl5f6p3/6J/V6PVmWZaLNtm1rMBgoDENtt1uVZSnf9zUYDOR5npbL5b117O7vMAxNgLWL7Xafb9vWRF/btr0Xd+1YlqVer6cnT57ocDjo22+/1fPnz3U4HFQUhdq2NVHXyWRiIsGu6yqKIgVBoP1+r81mo6qqNJ1O5bquqqr64LBqGIaazWaazWaSpCzLZNu2xuOxiqIwsep+vy/HcfTs2TNlWSZJOh6PD475BkGgyWSiKIr0/fffa7fbmbW2LMuEfC8vL7XdbrXdbs1zr9frKUkSBUEgSXr8+LH6/b6Wy6WOx6PiOJbjOOr3+3r69Kkcx9F+v9dwOFSWZYrjWP/0T/+kNE0VBIFc1zX3rW3b5vpMJhNZlqXT6STbtuX7vubzuc7Pz7VYLPT999/rdDopCAINBgNtt1uz/t0z4PHjx8qyTIvFwqxPHMcKgkBZlplrKMncb5K02Ww+6B62LEtRFJlI+e3traSXz7zRaKTZbKb1eq3T6SRJSpJEFxcXcl1X+/1eTdPI8zxNp1NNJhNlWabVamWi4Wmamvv/btz3Y/kYx/qpcX2KcQMAAAAAAAAAAAAAAAAA8DbEfAEAAAAAAAAAAAAAAAAAAN7A8zwNBgMVRaHNZvOgqOVDua4r13Xl+/4vKuZrWZYsy1Lbtl9cFNG2bUVRpCiKVJalyrJ8cHTXsiw5jqOqqpSmqSzL0mAwUBzHkl5GUT3P0+3trQ6HwyddG8/z9OTJE/m+L9u270Vduzjpu+7FLiAdx7EJ10rS5eWlZrOZ/umf/kllWZrzdKHZOI5V17WOx6PyPP/guOyHcBxHT548URAEWq/Xms1m2u/3KorCXL8unno6ndS2rQmz5nmuzWZjrpvv+5pMJvr66691OBy0Wq1MsDUMQ+12O/PeKIo0mUzeGjx2HMdEhP/lX/5Fx+NRkkwIOEkSPX36VG3bar/f63A4yLZtTadTWZal5XJpYq++79/bh6/uM9u2H3QNuvjr2dmZRqORDoeDFouF8jw3EePuXN0cXg0Vd2v4U7owruu6StP0Xsi3O/Z4PNajR4+02Wx0dXVlrtlwODTXx3Eczedz+b6v6+tr3d7eqqoq7fd7RVGk6XRqQstFUWg2m5kAcJqmqutatm2rrmu1bSvf901odzqdmoCx53kajUaK41iTyUSr1UrPnz9XmqZyHEej0Ujj8Vj7/d6sZZIkevLkicIw1IsXL3Q6nUwgt9/vKwgCHY9HZVmmtm3Nd8d0OpUk7Xa7DwqAO45jwtvb7dY8g1zX1Xg8luM4Wq/Xkl7eu91adWN1HEeTyUTz+Vx1XSvLMhPyrapK2+3WrNuXjKAvAAAAAAAAAAAAAAAAAODnYn/uAQAAAAAAAAAAAAAAAAAAAPzSWJalMAyVJInSNNXpdHrnY9i2LcdxTGCw0zSNdrudbm5uTPjzIcfyPE+u6752PNv+OL8O2sVwe72eXNf9KMe8e+wgCBSGoQmIfmxd8NKyLDVNo7qu3ynmezduuV6vdTgcVBSFsixTWZZq21ar1UpFUXyS8XfjGI/HOjs702azURRFyrLMBFxHo5GGw+GD17ALrYZhaEKeVVWpKArZtq3Hjx/r8vJStm2bAGb33qdPn+r3v/+9/vCHP+js7Eye5907tu/76vV6bz13GIYfbW924dc//OEP2mw26vf7Go/H8jxPURQpSRKNx2MTX7YsS71eT9PpVL7v6+bmRnVdm9efPHmiP/7xjwqCQLe3t2qaxoR/sywz+8bzPE0mEwVBYALXlmWZqLJlWfI8T0EQmD1X17WKolBVVQrDUE+fPpXrurq5udHNzY0Oh4MZ426303q9NpFgz/Nk27Y2m42yLDPzd11Xs9lMX3/99WvX4U1rFcexLi4uTMj36urqXvg4iiI5jqOmadS2rQlfj8djc+8/NIja3W9VVamqqnt7swv5XlxcKMsyLRYLM6/hcKiLiwsTix4MBhoOh0rTVKvVytx3ZVmaaK0kbTYbFUWh+Xyus7MzHY9HlWUpx3EUBIFZy36/r/Pzc3311Vd6/Pix+v2+Cfl+9dVXuri4UFVVur6+NjFg13XV7/fN2nRr30WRN5uNDofDvb00GAxkWZZOp5OqqjLPkslkosFgoMPhoMPh8N4h7O56jUYjua5rwsCO4yiKIo3HYxNR7tZ1Op1qv99rvV6raRqzFq7rKssyhWGoyWRi1rOLEN+NuL/6PfO5/VhgvrsXpY/3fQgAAAAAAAAAAAAAAAAAwI/5uL9pDQAAAAAAAAAAAAAAAAAA8Ctg27b6/b6SJNHt7e2Do7DSvwYsz87OZNu2lsulFouFyrI07+nimQ8NZvb7fQVBoDRNlaap2rY18d0gCLTf7+8d/33EcazhcKiqqu6FRD9EF7x8/PixptOp0jTV7e2tbm9vP3oUt4vLduHdLMtU1/U7jTOKIq1WK61WKzVNY6LOvV5PWZZpv9+ba+b7viaTiaqq0nK5fPC1/DFBEOjrr79WmqY6HA46nU5mncIw1MXFhbbbrQl3vk0X+uzmNBwOdXZ2ppubG7Vtq6qqNBwOTbw2iiLt93u5rqvz83P1ej1ZlqUsy+S6rqIoUhRFyvPcxEqHw6HKsnxjkLoLKzdN897XuQvlDgYDTadT/e53v9OTJ0/0/Plz/d3f/Z3Ozs6Upqkmk4natlWSJBoOhzqdTjo7O5P0MmzaRZhd11WSJHr06JEmk4n2+72ur6+1Wq1kWZZGo5G537tob3fMLhjr+76iKJLrujqdTiby3QVtoygysd4wDHV5ealer6fvv/9e2+1WZVnKtm01TaMsy0w4uG1bOY6jMAwlSfv93uxdz/M0n8/1t3/7t8rzXM+fP//RNYui6F7I9/b21jwzuojraDQyzzXbtpWmqbIs09nZmeq61mazefDzpGkalWVpIrF378Fer6fLy0u5rqsXL17odDrJdV1Np1Odn5+raRptNhv5vm9Cvuv1+l5c1vM8jcdj9Xo9HQ4H7ff7ezHibt+ORiN5nqfj8aiqqlSWpQ6HgxzHMfH0zWYjx3E0n8/Vtq2Wy6WOx6OappFt2ybgvtvtzBx839dsNlNRFFoul8rzXG3bynVdDYdDRVGkzWaj0+lkrmOSJJpOpyqK4oMD4N3xuvlvt1vz/E+SRFEU6cWLF8qyTJ7n6eLiQo7jaLVaKc9zua6ri4sLJUmi/X6vMAzl+76yLNN6vdbxeFRd1298flmW9VGeax/i1TF00d5Xx9Xds597vAAAAAAAAAAAAAAAAACA3wZivgAAAAAAAAAAAAAAAAAAAK9wHEf9fl9t22q/36tpmgd9zvd9zedznZ+fazAYqG1bE6x8NY75puigbdtyHEe2bassSzVNI8dxNJ1OZVmWTqeTGUsXC/V93wREHcdREAQmLNnxPE+e56mua1VV9Vrk1nEcDQYD9Xo9LRaLd4oX/5h+v68nT57o4uLCBIz7/b7quv7gyKVlWbJt2/wdx7GiKNJut1Nd1+907CAITKy2C5vWdS3P85QkiTzP03q9VlmWsizLBD5ns5murq4+SkTScRyNRiP1+309e/ZMlmWpKAoTGu33+4rj+F442HVdeZ5n4qWSTBh1Pp9rMBhoNBrp8vJSTdPo9vZWdV3Ltm0TwV2tVkqSRGmaqt/vazwea7/fm2Bwr9eT67pmX/q+r/F4rDiOdX19/db17Pbh+7JtW+PxWL/73e/U6/U0GAx0fX2tf/7nf9ZwOJTnedputybSO5vN5Pu+9vu9qqqS53mSZEK84/FYjx8/1nA41GKx0A8//GD2ShiGmk6nappGeZ6bdZxMJnJd18Stx+OxZrOZ2rbV9fW1siwzMduqqhQEgSzLMsHY8XisxWKh1Wpl4rBN06hpGtV1bfZuXdfyfV9hGKquax2PRxOynU6nevTokeI4/sl70/M8nZ2dqd/vm5Bvd6wuPjufz1UUhfb7vSSZIO5qtdJ8Ptd8PpckbTabBwXH27Y1UWfHceT7voIgUF3XZg9eXV3pcDjI931Np1NNp1OVZanlcinbtjUYDMxzsovrSv8ahR6Px2qaRtvtVmmaynEc86wbjUaaTqfyPE+LxcLcu6fTyUSasyzTarVSmqaazWaybdvEfbvr4LquxuOxwjDU8+fPzTr3ej1zT6Zpasbm+76SJFHTNGaPdfHvyWSiKIr0zTffmD32voIgUL/flySt12uzP7vvqKZptF6v1batxuOxBoOB8jw3QeRer6eLiwvVda0oipRlmbbbrQkQvxry7WK53bX93LpI70+95+7fAAAAAAAAAAAAAAAAAAB8asR8AQAAAAAAAAAAAAAAAADAb5JlWSY6alnWvQCi67qK41in00lpmj7oeGEY6uzsTIPBwAQTu6BoFEX3gquu674WUfR9X6PRSFEU6XA4aLfbqW1bhWGoKIpMMFL61/DrbDYzQcpXY7BdjDIMQ81mM7muq+PxqCzLdDqdzM8tyzLByrZtzRp8qDiOdXZ2pjAMdX19rd1uJ8uyNJ/PFcexiqLQarW6dz2knw4ydjHdOI7l+75Zx36/rzAMdTgc7q31T7FtW0mSyPd9HY9HpWlq4puu6yqKItm2re12K8uylCSJptOpiYQeDocHx57fNh/bttXr9XR2dqY8z7VerzWdTlUUhdmP/X5fVVWZ69PtF9/3td1uVVWVCe3OZjMFQaCiKJTnuWzbNnPNskxhGCpJEt3e3mq73SqKIl1cXCiOY1VVpd1up91uZ8KsXbC2C4gOBgMTPX6TKIrM3n9fbdua+LTv+yqKQt99952eP38u3/e12WzuBUnjONZ2u9UPP/yg4/FoArCn00lJkpiQ7+3tra6urrTf7811jqJI/X5fm83GxFL7/b76/b6yLFOe5wqCQOPxWOPxWFmWmcBxd46qqkyMu4t6Z1mm29tb5Xn+WnC0e193jCiKFMexjsejiRF31zIMQx2PRy2Xyx8Nw47HY02nU2VZpsViYebYXbezszPFcazVamXmKUlFUWixWJh5T6dT1XWt3W73oLB3XddmTeI4Vq/XU13XJiK93+8VRZFms5niONbhcNB6vVaapnr8+LF839fNzY12u50JWEsyx+vuzc1moyzLNBgMVFWViqLQdDpVmqZarVb3xtvFuLuo9PF4lOd5CsNQWZZpuVya69I9Ay4vL1WWpfb7vXl9NBrJ8zzt9/t7gfA4js3zpounB0GgyWSi2Wxm1vTV59Dd55xt22rb9q3PPMuyTKQ8TdN7e7YbWxcuHg6HCsNQRVHIdV0TKJ5Op5rP59psNtput1oul9put8rz/N5z600R319KHPeXMg4AAAAAAAAAAAAAAAAAADrEfAEAAAAAAAAAAAAAAAAAwG+O4zhKkkSDwUBBEKhtWxPR7MKaXWDybsyyCwBLuhcyDcNQFxcXGo1GStNUi8VCeZ5rMBhoOp3q7OxMbdtqu90qCAL1+33tdjtzjCiKdH5+bqKdWZaprmtzvrqulee5qqoygcfxeCzXdbXf7yVJvV5P5+fnJtwqvQwEz2YzzWYzpWlqwqiSdDwe1TSNPM8z67Ddbk2w0rZteZ4n3/eV5/m9kOVPCYJA0+nUBFYXi4WaptF8PleSJPeCwV2cN0mSeyHNN7FtW3EcazqdajgcmuvQNI2Gw6GJY74aqvwxrutqNBrJtm0dj0czz27tu/Dw6XRSEASaz+eaz+fK81yLxUKn0+nB63JXtwd7vZ7CMNRkMtHFxYU2m416vZ4mk4n2+70ZRxfz7fbAYDAw0dD9fm/ivpPJRG3b6vr6WkVR6HA4qN/vm7hvlmUajUZyHEfX19eqqkpJkmg+n6ssSz1//lx5nsvzPPV6Pdm2rdPpZIK2w+FQlmVpv9+/MfTa7c+2bT8ocizJBGyfPn2q/X6v6+trM7Yu0Ho4HOR5no7Ho3a7nTabjRzHkeu+/DVpz/P01VdfaTababFY6Pnz59rtdmqaxtwrw+HQ3EtdLHU0Gsl1XaVpaqLaYRjK8zxlWSbLsmRZlompFkWhMAzNdQjDUN98842Ox+Nr+7muaxOidhxH0sv7NwgCrddrSVKSJBqPxyrLUkVRqKoqHQ6HH91P8/lclmVps9mYqHV3z4zHY7OXu31lWZaJyZ5OJ93e3qooCl1cXGg8Ht+Lfr96rm7u3dyqqlKapuZ50rat5vO5bm9v1e/35fu+PM/TZrPRer02cd3xeKyiKLTdbs293IVuwzA09/lqtdJ+v1fTNAqCQHme65tvvjHn7WLnbdvKsiw5jqPxeGz2ap7nZhzL5dJEuLvnz2QyUb/f11/+8hcTOu4C6d3ad/O1LEtRFMl1XWVZpqqqFIahed7HcawXL16YZ2ynG5frumasb3rWdWFd3/fV6/XUtq02m42Je9u2Ld/35fu+sixTr9dTr9fT4XCQbdu6uLjQo0eP1Latnj59KklaLpcmYt0Fk38s2vtzBXQfGnF/9TM/FkAm/gsAAAAAAAAAAAAAAAAA+NTsn34LAAAAAAAAAAAAAAAAAADAr4fruhoOh5rP5+r3+3Icx4RLXdeV4ziK41iS7oVaLctSkiS6uLhQv983EULP8zSdTjWdTlUUhZbL5b3oZJIkiuNYlmXJdV31+30lSSLbtk349PLyUkmSmHhpmqZqmkau6yqKIjVNYwK1nudpNBppOByqLEsdDgcTpO33+yaI6bquBoOBkiTR6XTSer1W0zQmINsFOeM4VpIkaprGnNeyLIVhqLOzM52dnZkA8EM4jqPhcKjhcKg8z7VcLk24s1vv/X5v1tbzPM3nc52fn5tQcsfzPIVhKMdxZNu2Wf/xeCxJZrzD4VCDwcCs0+l0enBItgvllmWpNE1VlqWZRxdnPZ1Oqutao9FI8/lcnudpt9uZKOz76GKls9lMT58+1R/+8Af1+33Ztq2zszNdXFyYNZtOp+ba1nUt13VNNLQLUHchUdu2tVqttFgsTDC6KArtdjsTq51Op8rz3ERfJWkymSgIAhMZnkwmiuNYZVmaIG2v19NgMNDpdFKapnIcR/1+X1EUmfuh27NdfPpDdHMLw1Db7daEdYMgUJZl5hy+76tpGp1OJ5Vlae67IAg0m800n8+VZZmeP3+u7XarqqruRV/Pz89NsLVpGjmOo+l0qqZpVJalXNc1oeimaVQUhfI8V13XquvaRJJ7vZ5839fZ2Zm557o48F1dzNdxHLVta54/lmUpyzL5vq/BYKAsy7RarWTbtrIsuxcQf1UYhmaeu91ORVHcC1J3x66qyoSZ7+qCvqvVSmmamlh4d13v7tsgCOR53mufL8tSWZbJ8zzNZjPFcawwDM3z5fb2VtfX1yaMmySJoijS4XBQmqZmrbpncHftpJdh5zzP5TiOfN/X4XDQzc2NXrx4ofV6rTzPTcS1e0ZOJhNlWabj8WjWqGkac/67IfEwDJVlmdkfXQS51+u9dvwuyNvtlX6/r/F4bKLZVVXd22ed7nnbhdtfXdu7a9md33VdHY9HbbfbezHhIAhk27Zc19V4PFae59rv91oul1osFvJ9X8PhUGEYarVa6erqStvt1kSEO296fr1rDPdt83io94nv/tjaAQAAAAAAAAAAAAAAAADwqRHzBQAAAAAAAAAAAAAAAAAAvxldeHE2mykMQ6VpaoKnXXjUdV3Fcay6ru/FM13XNdHZLoRr27ZGo5Fms5nqutZyuTSB1y5i6bqu0jRVnucKw1D9fl9N06iqKgVBoMvLS41GI223W93c3Oh0OpkgYRiGiqJIVVWZyGwURRoOh/J9X9vtVkVRKAxDTf7/7NzXjuRIgqbtj1q59pBZXTU9O4sBFnv/l7HAHkwvsDPdWyIzQ7qm1v9B/mZtziA9IrJSVdb3AIXIcKcwGo2Mk8K7WAAAdrud3G42m6GqKtzf32O/30PTNIzHYxkuFnFh13VlkBQAbNvGfD7HmzdvZGD2JTRNg+d5WCwWMtobx7GMWAZBIIOhZVlC13WMx2NcXV3BNM2jYKau61gul1gsFnIezs7OEAQBoijCzc0N7u/vkWUZXNdFEASoqgppmp6Mnna5rgvXdRHHMeI4lsFMEVjVdR1pmkLXdZyfn8tA6Ha7lffkYzRNgyzLkCSJjHPe3d3h7u5OzkFZljKMahiGjBf7vo/5fC4DtmrcN4oiuS7EdTiOgziOcTgc5JqNogh5niPPczmmsizlGhUR48PhgDzP5VoRIWMRhv7hhx8wmUxkXFPEVkVU+mOJaOlyuUQcx0exV8/zUJaljKWKZ0ScMwgCzOdz+L4Pz/NQ1zXu7u7kuhPzK+bn7OxMvgfatoXrupjNZjKKbdv20T3Z7/dHwei6rpEkCTzPk+tfPZ5KRJxN05Tj930fjuOgLEv57miaRsaAPc+Ta7OPpmnymRb3VZzHdV05PvGea5oGtm3LeLNKBML7vlPvb993bdvK9+hsNkNZlgjDENvtFg8PD1iv1zKMLdZT27aI41i+DxzHkdHo2WwG13WRpqlcp5ZlwTAMRFGEsizl8yrmQQSwF4sFdF1HGIbI8xy2bcM0TZRliaIoYFmWDJlbliWfaREZ13Ud0+kUhmFgs9kczb2u62iaRq616XQKx3FQ1zWqqkKSJDIgrEZnDcPAYrHA9fU1PM+Drusy6i7+EyzLgud5aNtWxpnFHIsxiLkWYW4R3r65ucHt7S12ux3CMMR+v5fv1u56VM8pxvDS97163z/Gx0aA+66DiIiIiIiIiIiIiIiIiIiIiIiI6Esyv/YAiIiIiIiIiIiIiIiIiIiIiIiIiIi+BBG8XC6XsG0bh8NBRhodx0HbtjKc6Xke8jyXAUUAMiYrwpht2yIIApyfn8M0TRnMLcsSpmkiCAKMRiOUZYndboemaTCdTuF5HjabDdq2xWKxwPn5ObbbLR4fH4/CqpqmYTQawbZt7HY7FEUBXdcxmUywWCxQVRXCMAQAjMdjTCYTbDYbpGkqo5GO4+Dh4QFRFMlrGI1GMvhrWRYmk4kMDhdFAcMwEAQBzs7OMJ1O8f79exmNfMkcz+dzjEYjHA4HhGGIqqrgeZ6Mjd7f38sYrG3bOD8/x3Q6xT/+8Q8Z0gSAIAjwww8/yCDwdDrFZDJBGIZYrVYygiuuS9M0HA4HRFE0GD3tEwQBNE2TsdOmaWSUeDQayfstgrmO42C73SKO498VlKzrGnEcwzRNXF5eIs9z/Prrr7i/v4dlWXh4eMDNzQ2yLJMR5CRJAADz+Rye52G73SLLMsxmM0ynU6Rpiu12K8OdIhJsmiY2mw3iOMZsNoPneQjDUK5V3/flGkzTFEVRIE1TxHEs16TneZhMJijLEkmSyODz9fX10VxYlvUkzPwxxD24vr7G27dvEYahjGiLEG5d1zAMA5ZlIYoiuX5HoxF835drx7IsrNdrGTgWxFo3TROHwwFVVcEwDEwmE3ieB8MwZAhZzGVd1/Jc4lhN08jrnU6nMg7bvX7DMGTkW4ScRdBahHhd14WmadjtdsjzXL5fTq03wzBk8DfLMpimCc/z4LqujMuKeKwYl2EYmM1myPMcaZqiqipomgbbtmHb9uD9MwzjKEDbt404/n6/x93dnZx7Ed7VdV1GpkVc2fM8GU4W3wNAlmXY7XbIskzGejVNO5pfEaC1bVsGeh3Hke8gEakW8WXxTDiOI4PYwHEgV6yjLMvk+0q932maIk1TOQdinanhZHV+xBhd1z255sVPMQ/iuVePI85/d3eHqqpkJFnMfRzH8vfFYiHXg4icd2O43ZhwVVXy+fpcPjbkO3Qsxn2JiIiIiIiIiIiIiIiIiIiIiIjoS2LMl4iIiIiIiIiIiIiIiIiIiIiIiIj+FGzbxmKxQBAE2O/32Gw2Mo4qgpIi2GnbNjabDcqyhK7rAIDZbAbf9/H27VukaQpd13F2dobxeIz1ei3jm8CHoKkISt7f38to62w2k4FGwzBwfX0NXdexWq2OQr4AZFRU13UZwnRdF+fn55hMJri7u0OSJLAsC4vF4ijQKrbb7XaI4xhN08CyLMxmM1iWhaIo0DQNfN+H53moqgppmsrA7mQywWg0AgAZ5H3pHC+XS1RVJQOcbdvKgGlZlthsNsjzXM71xcUF0jTFfr+X8UjDMOR13t/fwzRNuZ24b03TQNd1mKYpY66r1ao3ojpExJGrqkIURSjLUn4ugrAPDw/I8xzT6RSj0QhpmmK9Xsttf4+maeS9ORwOMsDpeR72+z3CMJSRUnFOy7JwdnaGtm0RRZEMKBuGIedGrGURNy2KAqvVCm3byiCviB67rgvP83B/fy/DzWVZyphn27ZHcdj1en00P8A/nx9d1+H7/icJgdq2jbOzMwDAzc0N8jyXQdYkSeTzIoKsYk0HQSDDtnd3dwCAyWQivwcg59R1XUwmExwOh6NY7Hg8BvDhOa6qCkVRYDwew7bto3eFqm1bpGmK0WgkI+AiMiruRRAEsG0bTdPg4eEBcRwjCAJMp1PYtg3TNOG6Lvb7PeI4lmNMkuQodN0lQqxZlsF1XRiGAcMwUJYl8jxHWZbwfR9N0yDLMtR1jaqqMJlM4Ps+XNdFlmXyveX7vnyfdaOv4qdlWU+iv+J5FGMpy1LGf0WIWYSADcNAXdeo6xrT6VRGbquqQpZlMAwDAJAkCQ6Hw1FYvW1bGV4WcysCvWKMeZ7LqG5d1yiKAlmWySi1WOciWG2apozdimvRdR273U6+GwURdBb3VkTAl8slTNPEbreT7wcxN23bomkaRFEkQ8BqIFjdVtzPsiyRZdmTa2+aBmEYIk1TGaEXEXJBBH+jKMJsNsNisZAx57qu5bbiOsXfuaIo5N+M7vg+pb5n6GP1PYtEREREREREREREREREREREREREnxNjvkRERERERERERERERERERERERET03dM0DZPJBLPZDFmWPQk0ip8inqnrOqIokhFbwzBwdnaGqqqw3W5R1zWCIMDl5SXyPD8K12qaBt/3MZvNAECGKMfjMSaTiQzOep6Hq6sr/PLLLwjD8EmA0HEczGYzhGGIoihktPXy8lLGa8uyRBAEMiIpIrej0Qiz2Qxv375FWZYyeHl+fi6jlsCHaKzjODKOqus6giBAEAQygBnH8bMxXxFTnM1mGI1GuL29lRFgwzAwnU4xHo+x3W4RhiGapoFhGHKcf/vb3xDHsTyW4zh48+YNyrJEmqaYz+dwXRd3d3dH47EsSwZlHx8fsdvtXhWRtW0bo9EIcRzLuRPH9TxP3j8A8H0fpmni7u6u9359DNM0ZfT55uYGAOC6LkzTxOFwgKZp8vuqqqDrOsbjMYIgwHa7RZ7nGI/HcF0XURQdXb9pmhiPxzI2HcexXFP7/V4ebzQayTCwCL+KiK+4RsuyZEg3z3OMRiO4rouqqhDHsVxPjuPA8zwZCv5YYh2en5/j9vYW2+0WhmHIOHGapvJ5U4O7IoANAHd3d1itVlgsFgAgA7/AP++7ZVmwLOvoGRNB2jzPkWUZoihCURQydiqOlWWZjLhqmgbTNOWcpmkqI8ki1jwej+F5HpIkke8LcSwR+RWxVTUUCwDr9foo6NolIr1VVWE2m6EoChwOB4RhiCRJZBhXzJOIygKQ60sEm23bRpqmeHx8fHJOEZIFIN+TIjoLQB6jbVs8Pj7KMLhYNyIiK2LCInJuWRbqukYYhgjDEHVdYzaboaoq+Zk4b1EUKMsSjuNgPp/LsQBAmqaI4xiu66IsS3lf67pGnufyWdZ1XUaCxX1I0/QoiiuCuSJKrD7vIpQrAtBiHsW7KEkSVFV1FAAW8y+i2uL7vlgyABnUVp9HcQ/EfajrGrquP/lOHCdNU9zf38MwDBltVqPWItgr/hPPvzouEaT+HMRYf+85GO8lIiIiIiIiIiIiIiIiIiIiIiKiL40xXyIiIiIiIiIiIiIiIiIiIiIiIiL67tm2jcvLSxiGge12iziOe2OjInrYtq3cRsRlz8/P8fj4KIOgy+US4/FYxl3F8QzDwHw+x2QywW63O4rBep6Hw+GAuq4xnU7h+35vNBP4EKgMggCr1QpVVcF1XSwWCwRBgDiOEYahjL2qoVVd1zGZTGSgtW1bGTkVgd88z+W+mqYhyzLUdS0DpHVdI01TADiKHp9iWRbOzs7keUUcNQgCLBYLOI4jA7+6rsN1XZydnUHXdazX66P5m81mWC6X+Mc//gHHcXBxcSHnUoR8TdPEdDrFZDJBURTYbrfPRoe793o6ncK2bTw8PKCqKhlz9X0ftm0jiiIkSQLbtjEej5EkCe7u7n5XqFYQ62o8HiPLMjw+PqJpGozHYxl1BT6sm6qqUFUVLMvCbDaTsWkRvRWRabGPiDePx2MYhoGHhwfUdY35fC7XVNM0cv+yLGXkU4Q81dCm4zhwHEfOu+d5KMsStm3LELSIWBuGIYOlH8s0Tcznc+i6jpubG3lfy7LEaDRCWZYysOo4jgy8ep6H0Wgk5zNNU7nv+fm5DJ+KEG2SJAD+GdwWgdQoivD27Vus12v5nDVNI5/T8XgMANjv98jzHJZlyWcpDEM5jqurKxkcbtsWYRhiv9/LELFhGDBNU8Z1t9st9vu9fN6qqpLR7lNru65r7Pd7ed+jKEIYhnKORHA3z3N5nCzLZKxbRI1F+FaEnfvuoQjM+r6P0WgE27aPYs5BECBNU6xWKzkP4h2VpqmM4BZFIaPDmqbJ46pBZDH/4h2kaZqMIV9cXODy8hJlWSKOY2y3W6RpCsMwYFkWkiSRgV6xdrbbLQ6HA9q2leFeIUkSef0itiveCyKaK4LOYhsxJsMw5LtTRIrFd+octm07OK8qEWcW+4txqrFfVdu2T6K+4jgi+j2bzTAej2FZlpzvLMuQpinyPJf/iXfNlwrkqvHhT3EMhn2JiIiIiIiIiIiIiIiIiIiIiIjoS2DMl4iIiIiIiIiIiIiIiIiIiIiIiIi+e0EQYLlc4vb2Fvv9HmVZPtlG0zQZ5SzLUgYqgQ/xzslkgr///e+oqgqO4+Dq6gpZluFwOMjjGYaB8XiMs7MzGIYhv7NtG77vo65rZFkG27afxH5VpmliNBqhrmsURSEDvY7jIE1TGeoUAUl1rK7rYjweY7/fyzDmeDzG1dUVqqqS4WHHceD7PoqikEFSz/MAfAh9uq6LLMteHMj1fR9BEGC/3yNNUzmfk8kEvu8jSRIZQhYh3uVyic1mgzzP5T1wXRdv3ryR45jP5zBNE5vNRkY9TdNEEASYz+dwHAf7/V4e+6VEzBcA8jyXgU4R5myaBkmSoKoq+L4P13U/6jxDRNDYtm1st1tEUQTLsnB+fi7DzOI+iesejUaYTCbI8xxFUcA0TTiOgyzLjgKipmnK2GqapnKNBUEAAIjjWAYwLcuS0VfgnzFMXddhmqY8VhAEcBwHrusijmOkaYrpdIqyLNE0DSzLguu6R/f5Y4h48Gg0wmazwXa7lXFhy7IQBIEM3Oq6Dtu2UZaljBWbpon1ei2j2fv9HkEQyOenLEskSYLNZoOiKDAajWS4WwR7Hx4eZGRbhFzTNMV+v4dhGNB1HdPpVD4juq5jPp8jiiI8Pj7C931Mp1OMRiPkeY4kSbDf7+V6EnMvAr8///zzk3cJABRFIcPLz0VKRZRVzJVKrGURyxXnaJpGBnZfSuwj3lHiHWQYhozQ7nY75Hkuo+O2baOua+R5Lscggrnqu0uMvWkaHA4HRFGEqqpkzFaN7EZRBE3T5LMg9jUMQ6558bnYt67rwRC3uP9iexE0F5FY13Xls6ZuJ+6l67qwLAvb7RZJkrwogD5EjWm/ZNu+7USYXISikySBrutHgWDxU/wn9vvSQdxPcT41Ps6gLxEREREREREREREREREREREREX1ujPkSERERERERERERERERERERERER0XfNsizM53PUdS0juH0Mw4DneTBNE7vdToZRPc/D+fk58jxHGIYyjut5Hh4fH2VAU9d1GfEEgDAMkWWZPEYQBDIa6rqujJX2jUdES7MskzFgy7JQFAXiOEZZlnAcB5ZlwTRN5HkuQ42u68pr0DQNvu9juVzC8zw5JgDwPA+2bcvxiyBnmqbyGGmaviiMKM7Ttq0MloooruM4KIoCaZrK+XAcB7PZDLZt49dff5XBYDG3s9kMcRzL4G4URTJWaxiGDBEbhoEwDGU89DVEpDbLMjkHYh1YloW2bZFlmQw0u66L9XqNoiieXPvHxCNN04TrumjbFuv1Gnmey1huWZYoy1Kunbqu4fu+DKFmWYa6rmHbNjRNk+tE13U5767rQtd1hGEo50aEgdWAtBi7WE8idmrbNlzXlf+2bRu6riPLMoRhKNdakiTQNA1BEMCyLGw2mxcHoIc0TYP1eo3dbncUnhXx4yiKUBSFXP9RFKFtW5imiTiO5XpomkbGoieTCWzbRhzH2O/3cr7FfIiYs1jDgqZpaJpGPnfAP6Pdo9FIPu9lWeLh4QGHwwFJksh4ap7nMmLbF9nd7/dH5+96zVwOBWTbtkVRFE/W7scqyxKHw0HOoViHIuArYr0iRj40PvV61edIRHuHVFXVG0EHPgSQHx8fe+f7Y7mui7OzMxn4VYPJmqbBtm2MRiP5vA0Fg4FPE67tO54asxW60eG6ruU89/33Ocb3e732/cqYLxEREREREREREREREREREREREX0JjPkSERERERERERERERERERERERER0XdLBEdHo9GTaGuXaZpwHAdN08jgrWEY8H0fo9EIYRiiaRrouo7RaIS2bRGGIaqqgq7rCIIAo9EIAJBlGWzbRtu2cBwH4/EYALBer9G2LcbjMRzHwcPDQ2/4UcR7RUxWhCKbppFB2/F4jKZpkCQJHMdBXdfQdR2macpxzmYzLJdLWJYlA6Wapsnga9M0MgorQpVlWcLzPDRN86qYr4jO5nkOTdMwGo1kHLmqqqOg7Hw+x9nZmQzmivmfTqeYz+fQdR1RFGE8HkPXdRl11XUdtm3L683zXI5bzONLY44iVpskiQy/apoGy7JgWRbyPEeaprBtG9PpFLquy+iq4DgOgiBAmqZI0/TZc6rqupbhVxGjFde32WyQZRnatpXx4vPzc/z2229yvuq6xng8lvMLfIiOep4Hy7LkWhGhX3H8zWaDJEnkGtnv95jP55jNZk/WflmWMsJ8OBxkKDXLMhnWzbIMnufhzZs38np+DxHTTdP0KP4q1qphGEiSRK7boijk9YixRlEkg7JN0+BwOMj4q1gbInQqrudUCFfTNDkm4EN0Ok1TxHEsn5XD4SDfESLeq8ZU+44r7uMfUVVVqKpKRouBf8Z61WDsS4gI7aeIsDZN89HR4r7z27aNxWIB27axWq3ksyaiuYZhIAgCOI4jw80fez3d7dVg8HP6tuk+P33n+JRz/zn83jkkIiIiIiIiIiIiIiIiIiIiIiIi+hwY8yUiIiIiIiIiIiIiIiIiIiIiIiKi75rneXAcB3EcoyiK3tifCNxaloWiKJCmqQwB+74vA52GYUDXdbiuK+PAIuQbBAHyPEdRFHAcBwBgWRam0ymCIJDRWQDwfR9lWT4Jn4ogowiyFkWBIAhQVRXSNJUBVd/3Udc17u/vkec5JpMJAMiQr6ZpmEwmWCwWsCwLNzc3yLJMRjcty4Jt2wA+BGmrqsJut0Mcx3IM4pqHIqcqXddlzBcAZrOZDOKK+KU4r4geq6FZwzAwmUxwcXGBsiwBfIicuq4rg7IA5L4ilNo0jQyKVlX1orEKhmGgKApst1sZzgU+xDt1XUeSJMiyDEEQwPM8JEmC/X4v57dtW0ynU5yfn+Pt27dHx3iJqqpwOBxkLFbXdWiaBsdxEIahvMbVaoXxeIy7uzv5b8dx5DyJgLPneXLdHA4HOT9FUch5Wa1WiOP4KJS72+1we3uL5XKJ2WyGoijketjv9yiKAnmeyzkW11mWJYqiwGKxgOM48DwPv/7669H9/lhDIdj1eo2mabDb7eS1rVYrOaaiKFDXNaqqOgrKiqivmGtxDrEGn7tvagBY07SjdWcYhjyWGrM9FfL9nrRt+yRI/DHX/S3OlQi3j0YjbDYbxHEsxyl+mqYJ3/ehaRrCMDx63j4Vdd2qP/vCvOo+agxYXffq5596rF+Len1EREREREREREREREREREREREREnxtjvkRERERERERERERERERERERERET0XRPR2izLUFXVYMxX13VYloWmaVDXNRzHwXw+h6ZpiKIIo9EIpmkebT8ajeA4DizLQhRFiKIIpmlC0zQEQQAASJIEmqYhyzIZ1LVtG2EYYr/fH42jbVv5veM48H0fURTh/v4eYRjCsixkWQbTNLHdbrHZbGAYBsqylEHhNE1R1zXOz8+h6zrevn2Lh4cH1HWN8XgM4EOkUkSJq6pCURSIoghlWcogsIj5vpSIWl5fXyNNU6zXa4RhiCAIYJomTNPEcrmEbdtynq6uruA4Dq6urnBxcYGHhweEYYi//vWvaNsWvu/LOZpOpxiPx8jzHIfDAbquy6hqnuevjjimaYr379/LcK4IZoqYb57nKMtSRna32y3SND06T13XyPP81SFfQd1HBKV1Xcd+v0dd19A0Dbe3t9hutyjLEmVZoqoqeW+TJMFiscDZ2RlGoxGqqsJ6vcbhcJBRXxFwLssSNzc3R+cUAdybmxuEYQjHcaBpGuI4RhzHMoorrlFVVRXu7+/xww8/wDRN/PLLLwjD8NVz8Bp5nuP+/l5eQ1EUMnosrrGPGiztXn/3sz59+4h/D8VQGRV9uW91rkTQXTznarBZsCxLvkfFOxT4MtekBn3F36RurFdQw9bqZ9+yvmDxEDV4TERERERERERERERERERERERERPS5MeZLRERERERERERERERERERERERERH9oQ8E/Efar61oGWUUsVQQZxe8ihCj+E+FZ13Xx+PiILMsQBAHG4zHSNMV+v8doNMLFxQUOhwNWqxV2ux3yPIdlWYjjGKZpIooi1HUN3/dlNLeqKrx79w5pmiKO4yfjFtcymUwAAHd3dzgcDiiKAuv1GkmSAADiOEZd1zAMA5vNBuPxGMvlEnEco21bhGGI1WqFd+/eyRDrer2Gpmny3GKM6/UaaZqiaRq0bYvD4YCqqhCG4YtiinVdY7vdom1bVFWFh4cHGaStqgqu62I6nSKOY7x9+xbr9Rp5nuPq6gr//u//jjRN8csvv+Dx8RFN02C9XsuxeZ6Hn376CaZp4vb2FqvVSt4PERweiriekiTJkzgv8M/4s4jEpmmKd+/eYb1eoyiKo20PhwPSNH3y+Wtomib/a9sWu91OXo+maajrGnEcQ9M0eb3r9RoAZEQ4CAIURYHtdosoimTY8nA4wHGco4hnN+Yp/ttut/KZ6UY/+9R1fbRuoij66Dl4jb7xv2af7uefIwD6rUdS6eUsywLQf09F7Nc0TWRZhiiK5Dv0JV4Tq1WJQHc3Ut33PDRNc/R++SN5brzdmPEf7fqIiIiIiIiIiIiIiIiIiIiIiIjoj4sxXyIiIiIiIiIiIiIiIiIiIiIiIiL6LolQZxRFKIoCk8kERVFgv9+jaRpYlgXTNFGWJZIkkbHd2WyG2WwG13Vxd3eHMAyRJAkMw0DbtsiyDHEcI89z6Louf6+qCsCHCOz9/T1WqxXqusZoNMJ4PJb7NE2D1Wp1cuyHwwEPDw/Y7Xa4u7tDmqYyVigisyJc2DSNjOCKcadpivfv32O73aIsS7nt4+OjPEdZlthsNijLElVVyTBkXdeIokjOyUsCiU3T4O7uDpvNBnVdoygK1HUNTdOwXq/RNA1s20YYhtjtdjLY+7/+1/+C7/s4HA5yDnVdx9///nfMZjPkeY7FYgFN02QwWYRu27ZFURQoiuLFUde+NaKybVuGMkXkOY5jpGkq76+qLMuPCgl3xyDW6nq9xnq9lscU51RjnHVdI01Tub8INXfnoCgK7HY7uU/f9apjGPr+1LyK+f8jYwCUhrRtK+Pu4r2g0nUdvu/DMAzs93v5LL6GGtB+6ZiapoGu6zAM4yjo2x23evy+7f4ITkV6+ewSERERERERERERERERERERERHR16Lxf2IjIiIiIiIiIiIiIiIiIiIiIiKiz03TNP7PavTV6LqOi4sLXF1dAfgQPxXx2jRNcTgckKYpDMPA2dkZ/uVf/gVlWWK73eLt27coyxKapsE0TWiaJkOrIpIInA4lmqYJy7JkNPclxPmapnlybBFp7P4/oKZpAgAMw0Bd1zL42ze2vkCi+vupgOJLxt53LZqmPbkeEakV5xff6bouY7qapsnrUfdxXRe2bSPPc8Rx/EnCjo7jYDabwTRNhGGIKIq+aASze2/Fz+6c9lHnUd1P3Ev+P8NEr6NpGq6vr3FxcYFff/0Vh8Ph6H3gOA7++3//7zBNE+/evcN2uz0KrT937FPbdb/vRn/F+149n67rT96vf+Rn/7V/h8T77v+fk+dfmkREREREREREREREREREREREREQfyfzaAyAiIiIiIiIiIiIiIiIiIiIiIiIi+pyapsFms0FZlvB9XwZ50zRFmqYoikIGA/f7PX7++WcAwOFwkPHdtm1RVdVRWPClkcGqql4c8VWPLaLBXXVdH0VwgQ8RQ3EONTY8NMbPGXhs2/ZJWFYEkfvOORQnFvMtjtU9Zp7ncm4/1bWUZYndbifn80uGfIGnEV/gOMjb/a5v377Ph777PdHmP7LXXPefdY7ogzAMcXZ2hvF4fBQN1zQNnufBcRzkeY48z1/8Xn1JnFts1z2e2LdpGhk8V7cR79OXhny7keAv7bm/Uy/VNxdEREREREREREREREREREREREREnwtjvkRERERERERERERERERERERERET03SvLEtvtVoZa27ZF0zRH4b+maZAkCbIsk7+rvqVIoIg0qmHHvm1+z/F/j+7+p2Kyfd+L6+lGbLv3qyzLT3pfmqY5ijt/C14a/hzSdy1fO+D5Ob3k2l5z3UMx1e9x7uhY27YoigKHwwGTyQRFUWC73aIsSxiGgSAIUNc1oiiSMd8+vycyqz7/akhY/A0b2kds8y0Hbn/vu61L/ZtIRERERERERERERERERERERERE9Dkx5ktERERERERERERERERERERERERE371uCPaUoUDit+hbjTT2UUOU6s++7zVNg+M4aNsWVVUdhZe7Pz/XOL8V3fVoGIZcz8+N9Vu7li/hz3jN9PmUZYnD4QDXdTGbzaDrOqIogq7r8H0fRVEgiiJUVTW49vri5n3R2VOhaPWzvkiv+L1tW+i6Pnju58b2JX3KMPZr/sYTERERERERERERERERERERERER/V6M+RIRERERERERERERERERERERERER0RchQr2qvtClZVl48+YN4jjGfr9HWZao6/pLDvWreEmMsm3bo8Dva4OY3QgovR7n78+jaRpEUQTTNLFYLHB9fY00TdE0DWzbxn6/RxzHnzQCPxT6FetOxHq7gd/uv7/1wO2nHN+3fJ1ERERERERERERERERERERERET0/dG/9gCIiIiIiIiIiIiIiIiIiIiIiIiIiOjT6otBfm1iTGp8tmma3gjjaDTCxcUFPM97sl9fEPjPQly7ev2vCWL+WeeN6PeoqgqHwwGHwwGapmE2m2E2m6GqKkRRhLIsj7Z/7h01FOt9zfanznPq3fpHw3cWERERERERERERERERERERERERfYvMrz0AIiIiIiIiIiIiIiIiIiIiIiIiIiJ6HU3TToYa27aV2/Rt+9z+n0o3xKj+roZpm6aRn+u6Dtu2AQBZlskoZdu20HX9VfHaPzI1eqwSvz93D/v2/zPMG9Gn1rYtiqLAdrsFAEwmE7Rti91uhziOUdf1q4/3ObcXdF2X79Zv7dl/aaT3Wxs3EREREREREREREREREREREREREQBo/B/ciIiIiIiIiIiIiIiIiIiIiIiI6HPTNI3/sxrRF6bGEr+lmK8IDA+NUdd1eJ6H0WiE3W6HqqqOgpR/lpgv8DTIe+qeDu37km3/6L7Uev7S56Jvi67rME0TjuOgaRqUZYmyLAfXw1CQ+yVOxW77ouZiXarvUU3TUNf17xrH5/IpxySi8ML/H4B/WS2YiIiIiIiIiIiIiIiIiIiIiIiI6CMw5ktERERERERERERERERERERERESfHWO+RH9eari3LyoswpQAZOhXhCirqpLbfs8R39cEYkW4UsxHX/Sz+/n3Om+vxRAvfQqfOo479G4Uz3H3e/X86jbis77v+8bafXd86WfjUzyPfdfOmC8RERERERERERERERERERERERF9bubXHgARERERERERERERERERERERERER0cdinPPbpoYWxe9d3Uhl27ao6/qrhya/tI8JhKqhZLGvGkYW3/3ZnxM1fvy9+bPf26/hS8z3UNxX/Ltvm753aVff++BrrJ++vwUf62tfCxEREREREREREREREREREREREf356F97AEREREREREREREREREREREREREREH0PX9U8aBPwYX+P83UDut+TUuMS4T43/VJTxuX3/yF4amu3OQdu2vSHf7rH/7L7nOfier+3P5NR7re/Z7oZ9+/bpHn9o26/1Tv3Ua/d7DXYTERERERERERERERERERERERHRt8f82gMgIiIiIiIiIiIiIiIiIiIiIiIiIvpYXyvcJ+KHagTxS45FDbh+zL6fY6wfG4Tsi04+Nz6xze+Zh2/NqdCmuFZ1m+72p+bge5gfoj+Dvneapmlomkb+Ww35qu+GvnepiN6L0G3TNE+O/aX93nN+j0F3IiIiIiIiIiIiIiIiIiIiIiIi+mNhzJeIiIiIiIiIiIiIiIiIiIiIiIiI/pBE3PBrUYO+aiyx+z1wHGb8vRFasc/Hhnw/h27cWFyjruu926lhSrH9S87RnV/1vN35/1qG5vjU2NQ11N2/u476vhefdyOf9BTnhr62U+Hu7r+777hT2/X9W92n+zfza4Z8P/YZ7Ps78HuOR0RERERERERERERERERERERERPQxGPMlIiIiIiIiIiIiIiIiIiIiIiIiou/ap4h3DgVW1YhqX2BQ/BRR26ZpZORWxBWbpnn1+H5PCFGNOv6e43THIq6pbVt5nX2653zNGPru5aeMs546Vl8cU/z8vWHpvrkSa0s9dt+8DwV+v6XA8Zf0kntI9C0YCr73fa9S/+6oz3p3G/UdMfQ36ku/Hz71+RjzJSIiIiIiIiIiIiIiIiIiIiIioq+BMV8iIiIiIiIiIiIiIiIiIiIiIiIiot+hLy7b/d0wDAAfgrdt28q4rwj71nX9u4Ow3XMOxQ1fG3RUj9UXlxQxSXFN3e2G9ukbx9B36vm7Y31pmHjonC+NQPZFc7v7f2xYUo1yPrcOumHOU/PzZ4z6PnedL1knL9mO6GOIZ/LU94K6Boci6M+9n/v2Vc/1R1vnf7TxEhERERERERERERERERERERER0feJMV8iIiIiIiIiIiIiIiIiIiIiIiIi+q59qvifGvlUo6vic/Wn2EbX9aOgqhpy7P67KIpPMk4xxtd8rjoVmjy1TzdCK66vG0cdik+qn4kwcF8oV92/+7N77qGIrfp997O+43XH2Pf7x86bOM5LotBD+z4X5fwjRjs/h1NRaKJPre+dIP4mNE1ztB773m+GYchtm6aR70b1XXsqJj4UA/6egtXdvyXfwzURERERERERERERERERERERERHRt48xXyIiIiIiIiIiIiIiIiIiIiIiIiL6bp2Kdr4k6DkUfVV/qtt1t+nqxkTVY3cjjZ/CqXH0fde9hr7oLoDe2O4QNSzZtq3ctxvy7c5HN3AJQEYt1c+eC+/2RTK796u7Xd/a6EaE+8K+IrYpvn8usNv376Ftxbw9t0aG7ps6rj+yvufxY65paD81uv0tzxXjpd+uU8+2WFfd95igrj/LsmAYBqqqQlmWT47ZfYeK4G9f8Lx7ju+BOg99105ERERERERERERERERERERERET0uTDmS0RERERERERERERERERERERERETfpb7g4Wv3FwHUofir+l33d7FNN7ioUuOqnytE2BfN7RvLc/t2Pz8VQtV1/Sj4243Liu+6x+wLJDdNc7R/d1913rvXpYaSh8bdnZ9u/Lcvtnnq++74hwK6fYHovu+64x7ary+w3H0GRPDzW4ldfkyQ9mPH/pII8GvG8yVjuup9NE0ThmHId1OWZYOB5285bPo54uXfmr53Q/f564uOA4Bt21gsFnAcB/v9HrvdrvddKo4hPhdR208V7v5a0ehTfzfF99134re61omIiIiIiIiIiIiIiIiIiIiIiOj7w5gvEREREREREREREREREREREREREX03hgK+fbFUNbzbjQGq0Vg1QNsNJHa3GQq7qucTQdVuvFbXdRmv7YtcfkyYc2gefm/0sO8aunPYvQY1jNw3lqFocl8UWP2pzqs6vj5941X3UY/XPdap+G7fcdQxvfTePbfNUDy4O26xlvritSKi+rXDl77vIwgCFEWBw+EwGOwU13Qq/Do0v6ei031R1FP3qXs/+94Hn5KmaTAMA4ZhwPM8eJ6HIAjg+z4AIM9zhGGI1WqFoiiO9jVNE7ZtQ9M0lGX55PuvzXVdTCYTxHGMJEm++lr8lIbi4Opnz12vYRgYj8eYzWbyWT21llViXX5Kn+p4z0XLX3ruvr8T39MaIiIiIiIiIiIiIiIiIiIiIiIiom8fY75ERERERERERERERERERERERERE9IfTFzbshnn7oqdiGxGiVXVjsGpc9lSkV5xzKDDYDciKEK4Irqrfid+HQqEfE/Ttu76P2acvetwXv+3uOxRM7h5D3VaEW/vu5UuCtyr1/N2As7gHdV0fba8GMU9FX09Forv7nYpSqscTx9B1/WgdiP1F3Fj9d1+c91SU9mOCvr9n7XU5joOzszP4vo/Hx8eTxz31XffZUbfvPo9D9+FUdLlv3T43rtes1b59LcuC7/uYTqcYj8dwHAd1XSPLMiRJgv1+jziOURRFb/jbtm0sFgvMZjOUZYnVaoWqqk7GkPvG8bniqKZpYrFYYDKZoCxLpGn6XYdY+wLmfZ+r69WyLIzHY5imid1uhzzPn+wvdP+29H3+e8f/KTwXzX7JO3Mo9C62fc0aJyIiIiIiIiIiIiIiIiIiIiIiIvo9GPMlIiIiIiIiIiIiIiIiIiIiIiIioj+kvpBn3zZDEVVBjaSqUVcR3FWjr+r51PDq0LjEccS2dV0PBmr74rGvjRX2RRBPhRNfE3x8LqqqRni719EXmVTP3TQNDMOQxxBzDuDJNasx5r5Ya/eaxL1T51fTNBiGgclkAsdxsNvtkGXZk+P0hTjFePrW0FBscyjoqc5d3/Wq0emXhGu786+eW8xrNyAqtv1SQVXLsnB2dobJZIIwDBGG4cl1eGotvzbsrHouQN13vu7nHxOb7mOaJoIgwHQ6lWuyLEscDgfc3d0hiiKUZfmiALPrupjNZhiNRgjDEIZhyHjzS32utaBpGkajERaLBaqqQlVV8nPTNGHbNtq2RZ7nR+9isc23HP3tRsvVz9TfgX8+m+rfFXU7z/Pg+z6qqkKapkfzBBy/I7prsRsT/xY99755ydjFu1ENr3/r101ERERERERERERERERERERERETfD8Z8iYiIiIiIiIiIiIiIiIiIiIiIiOgPRw0gqsHT7jZDod2+z4aiwN1Io/hdhAT7gqvdMfadXwQZRYxQjQQ/dy1qxFANI6phw1P6grJD53su+KsGJruRym6gWB2rYRhH/+6LzIprUvXFVdX56/uu77p838f19TUAIMsyGUtVzzO0frrx1u78vSRK2/28G+Mc2v9UoFrM1VDQc2it9/3e3e/U/X9pQNMwDEynU5ydnaGua4RhiLquB9djXxi1b40/tzafu65Tc/7cs/LScfR9p+s6fN/HfD6X8db1eo04jlEUBcqylFHbl8yzaZpwHAdpmiJN0ydr+jU+RxjVMAzMZjO4rovHx0cZqTUMA2/evMH19TWiKMLbt29xOBxOrvlTa6Tvmk+9xz7mWtVQdve5656zu4/YvruWxbEWiwUmkwlWqxWyLDuKOKtRd8Mweq/hVOT7c/m96+U1+/b9PSEiIiIiIiIiIiIiIiIiIiIiIiL6khjzJSIiIiIiIiIiIiIiIiIiIiIiIqI/FF3XB+OJ4nvxXV+QVuxvWRZs20bbtjLmOhQGVOOo3bhsN97YDQx2I4wirChCi2qsUwRuxWdDkVH1uOq5m6Z5EtJVI7tD+3Y/ey7s2vVcQFMdR/eedLdVr13MtTi2eo5uZHnouroBXhHMdF0XpmkiTVNUVXUUzByizqm6DtQQczek2xeeVddhd5tu1PjUNZ2KNw+FUPv+3ReLfm5eTxmKr9q2jbOzMxlzzbLsaPu+4KqYo6HQtfgOwJMYc3ebU1FYNdJtGAYsy4JlWWjbFnmey/dD3zy9dN10iWsT8V5d1xHHMZIkkaHb10RSbdtGEASoqgplWQLAi9Z116lg8cfSNA2e5+Hi4gK6riNN06MQehRFeHx8PHo2xLvg1DH7nqE+p9Z63+dD5xsKYotnse892LetruuYTCbwfR9lWSKKIhRFAcuyMJvNoOs6kiSRMd/uNXdD7uq1nHoOnru+j43jfkyMt++86j3trt3ngs6fI0BNRERERERERERERERERERERERE1IcxXyIiIiIiIiIiIiIiIiIiIiIiIiL6wxCh2ucCusBwsFbTNMznc1xdXcF1XYRhiM1mg91u9yTo2w19qucDhqOJalxQbK+GfNXt1CCr+Ck+U/ftjkMNRHbju0Oh25dGOrvxSPXf6vnVcauBWfVa1OjkUAi4G6Hsu/a+61PnqjuHfXMj4qGmacLzPDRNgzAMn4RaTwWUu/d+KELbnY9uuFkN2A4FRrvH7M5bd17V+erGhrvXpq4p9XhD1HEOxTiHAs1i+/F4jMlkgizLsNvtkOd577m760z92bdehtaVqi+yLZ4/cUwReZ7NZgiCAIZhIM9zbLdbuU4+pbqukaYpNE3DYrGAYRiI47h3TT0XKdV1HbPZDL7vY71ey7Ve1/VHBU4/dRTVMAyMRiOcn5/j9vYWaZqirmsZxg7DEEmSQNd1lGX55J0n1ohpmtB1XV7feDzGfD4HAKxWK0RRdBSD7Yvr9l3rqdizrusYj8fwfR+GYSBJEkRRdPTe6AvKdgO14t0zn8/xww8/4Pz8XI47TVPs93uYponpdIr9fo8kSQAc/z3oBm6fC4d/i05FysU8nYpQ9z3vL4lqExEREREREREREREREREREREREX0qjPkSERERERERERERERERERERERER0R/OULhVUGOB6r9t28ZsNsPFxQV830dRFNA0DY7jwDAMlGX5JALbF2rshmr7vhv6THyu/jd0fd2oqxrK7cZau8dVP++bo66hIHDfcbrX3Rd6HQq7qvuIYzdNA8Mw5PdqjLNvnH0By757od5L9Vi+72M2myHLMiRJgqqqYNs2DMNAVVWoqupVYcy+uerGdrux2+7v3Qhyd8zdsajH7Z5DRDG7sdtu/Fe9d2qc+DW6a8U0TViWhaZpjqKsuq5jsVjANE08PDwgTVN5DBHuVM9tmmZvLLovGtz3fJ36vbuPGP9oNML19TV830eWZUjTVF6PrutH9+hThkNnsxmm0ym22y2yLPuoY9u2jYuLCxRFgaIo4DgO6rpGVVWD+zwXcP4U1HUhgsUinK6ubxHePRX+FqFXx3GwWCxwdnYG0zRRVRUcx4Hruvj5559RVdVgxLjv/dT3nhNjdhwHy+USi8UCtm3DNE0cDgc8PDxgv98fzW/fO0B9Hi3Lwps3b/DTTz9B0zSsVivsdjsURQHbtrFYLDAej2GaJpIkQZ7ng38/ur+/JFr8nM+1Fp6b625M+dTzdervMxEREREREREREREREREREREREdGXwpgvEREREREREREREREREREREREREf0hiJAj8DQE2A00dn/XNE3GJM/OzlDXNX777TcZgazrWm43FIcdiuF244Ldz8W/dV0/Op6Ig3bP1/29G14dGoOmabAsC5ZloaoqGSbui8iK+ew7Rl+QuO/fgrgOEeLsHlNco/hPvY8qNUY5FL3troG+8/VFItVtDcPAfD6H67rYbDbI8xzAh8Cv67pI0xRxHKOqqsFwcff61bnr2+a5wHF33Yk451Ck8rl10Z2HvvvfvTd9+/WddyiQKmKlV1dXmM/nSJIEv/32G8qyRNu2cBwHFxcXSJIEYRiiruvBCKkYV/c7ca1i3260uxs3fknAWtM0GIYB3/fxr//6rzBNE/f399jv96jrGtPpFE3TIAgCuK4LAIiiCFEUDQZUn5tHdT1PJhMsFguUZYn1en0Uh31JpNQwDBlKHo1GePfunfy8KIreqO1L5sU0Tdi2Ddu2oes68jxHlmXyXaluK+bQsiwAkNuJNSZCyfP5HOv1GmEYyjGK+6Wu9771rOu6DN6+efMGAHB/f48oimAYBn788UcZBBdhYHXfvndbH7GWDcPAaDTC5eUlJpMJ8jyX0V3XdeH7PtI0fRKaVsevvucdx8H19TV++OEHhGGIu7s7RFGEoijQti2CIMBoNMJ4PEaSJDgcDnKuTwWO1fMO3cvfE7vte3e81tB9HTqP+P25GPNLA8BEREREREREREREREREREREREREnxpjvkRERERERERERERERERERERERET0zVPDnH1BX6EbORW/67qO6XSKi4sLAMBqtcJ+v4dlWdA0DVVVPQlzDkUD1fOLf3fDq30xVzUs20fEddXzDEWLu3Mi9p9MJvA8D9vtFlVVPTlXN3IqfhqGAdd1YRgG8jxHnueDodLunPTdj+7YTkWF1f26UeCh4DDwdM67seS+44tI6dnZGaqqQhzHMphZ17WMeHav67mIcnc++wLP3VBzl67rg6HUvvWjnrd77X3jVImIat967LtXakRZRI6HAqZlWSIMQyRJgqZp5D2czWZwHAf39/dIkqQ3wN1dM4ZhoGmao+sU41fndehau0HgvvujaRqCIMBPP/0EXdfx9u1bHA4HNE0jzzOZTGDbNjzPg6ZpWK1WSNNUjq0v+tx3j9XtdF2H67pYLpcwTROPj49IkkSOe2h/8dOyLHieh+l0CsMwsFgsZMTb8zy0bYs8z+Wa0nUdpmnKezR0fDGu+XyO2WwG0zTRti2yLMPj46MMMQOQY/B9X8aw27bF+/fvcTgc5DFt28bZ2RnG4zEeHx8xmUwwnU4BQMadsyx7cq9EwNf3fTiOg/F4DN/3sV6vsd1ukWUZAGC5XAIA9vu9XEdijfetkVPvajXk+8MPP2A0Gsn7Ld4bZVkC+LAOgyCAbdsoigJpmsqIsbq+TNPE+fk5Li4usF6vcXNzgyRJZGhZ0zSUZQnbtgEA2+0WURTJ74f+1nXX3HPX+lpD7/ePdepv4GvG3ff3RUS+iYiIiIiIiIiIiIiIiIiIiIiIiL4ExnyJiIiIiIiIiIiIiIiIiIiIiIiI6JunRjn74rrd6G33M9u2cX5+DtM0sVqtEIahjJJmWYa6rmU4UegGGbsh3+426uenPjsV+lW36zuf+r0IrIrvHMfBfD5HnucyNjmkOw7P8zCfz9G2Ldbr9eB+3RiwOlb1uEOB225seWhMfffYMAw4jgPTNNE0DdI0RZ7nvQHn7jHU6KrruphMJvjtt9+QpqncPssy6Louw85qQLYb1FWDwkPXcSrm2jdPQ+MXoV41qKtGcvvG1bfm2raVcVp1TtR7q8aTT92LvvsKfAgi7/d7+XyJc9i2jfl8jjRNZaS0ex7TNOW1VlV19L0a9VXHoMaXm6ZBXddPIs/qOdQ5E9fk+z7evHkDz/Pw7t07GfIV68TzPBlarqoKruvCNE15b/rmu+9ed5mmiel0islkgjRN5XlPEceaTCa4vr7GX/7yF8xmM+R5jizLcHd3B13Xoes64jhGVVXQdR2WZWE8HsPzPERRhN1u1xt/FhHbs7MzzGYzAECapjIe7Lou0jQFAPi+j9lshiAI0DTNk/ttGAYMw4Dv+zg/P8e///u/w/M8xHF8FDQWoeT1ei2fZ+DDO3s2m2G5XMLzPFiWhaqqcHt7i+12K99xpmliMpmgKAqEYSjHIOZBrIuhmHiXrusIggA//PADJpMJHh4esNlsMJ/P4TgOoihClmXIsgyGYWA2m8GyLOx2O/kOsSwLpmnKoPpiscDFxQUOhwNubm4QhuGTSK+IFadpiv1+j6IonoxN/Rt46hqe89z+6jw9tyZfa+h4pwLLfc9PN9Tctw0RERERERERERERERERERERERHR58KYLxERERERERERERERERERERERERH94fRFMruhQ/FvTdMwm80wn8+x2WxwOBxQliXatpWRRxEAfS44q0ZUu+fpUqOF3eBqN/ja/b27vWEY8jPTNGFZ1lFgFACCIMD5+Tn+/ve/y8+GQrNqGFbTNIzHYwRBgMPhMLhv3/yr8Vc1OquezzAMOVY1qtm9fnE/+kK3nudhuVxiPp/DMAxkWYbtdovtdossy+SYxDjUc6shXxFQ1XVdhj/V86dperQW+u6vOq6++G33/nX3UbdR91ejxWLc3XGYpimDvOLed+dKXUN997ovxgtARnH74sDqvRL3uUtsW5bl0TZt28L3fXiehzAMZQxWHMuyLPi+j+l0Ct/3Eccx7u7uUNc1XNfFYrGAbdvY7/dyfYq47Hg8xng8hqZpCMOwN4LajV6ra87zPLx58wbj8RiPj4/Y7XYAgPF4jMlkAsdxkGUZ4jg+mj81It23ntX56LvHmqYhCAIsFgs0TYP9fo8kSU6+D8Sxzs/PcX19DQD47bff8PDwgB9++AFFUSBNUxiGgaIokGUZmqaBZVmYz+e4vLxEVVWIoqg3Im0YBiaTCS4vL2WcNo5jGaj1PA9FUcAwDIzHYywWC7iuiyzLjkK2rusCAJbLJUajkXwvOY6D29tb3NzcIMsy+S47OzuD7/sIwxBFUaBtW7iui+VyidlshqqqsNvtMJvNsN/vsd/vj2Llvu/DcRwZAxbPiVg3h8MBYRj2vnv6ouyO48hI8t3dHVarlZz3sixxOByQ5znyPEcQBBiPxyjLUv5NsW0bi8UClmXJfa+vr9E0De7v7xFF0ZOQtmmauLy8xHg8xrt375AkydF7q/t3ru/5e03Y97ltPzYS/DG6z82p8/f9Pes+I0RERERERERERERERERERERERERfAmO+RERERERERERERERERERERERERPSH1g2nduOdhmHg/Pwcpmliv9/LkKTYt2maJyHgvpivbdsYj8dHAUs1Kim278ZU1e/6qLHPrm7U0zRNTCYTTKdTNE2DKIoQRRFM08R8PodlWTJuq+7fjbGqn7uui9FoBAAyZquOV8zvUJR0KBhsmiaCIMB8Pofv+0iSBA8PD4ii6Oi4fdepRofH4zGurq7geR7SNMXhcICu6/B9H0VRoCxL1HUt9zMMA7Ztw3EcmKaJNE2RZRksy8JkMsHV1RWA4/izuIZuQFeMQdd1NE0jQ7LqtYrzua4LwzBkAFWNS3bnT51fNabreR6urq4wHo/x8PCAx8dHucY8z8N8PkcQBCjLEtvtFnEcy4CorusyKirus67rcBwHTdMgz/OjuK/63DiOA8MwUJYlqqqSIVE18Cs+U++9mC9xHk3Tjp4JNVyr6zqiKEJZltA0DbZtIwgCjEYjmKYJ0zThui5c15Xz/ObNG/i+L9dMXdcIwxCu6+Li4gKe58EwDJimibZtkSSJDO2q61SMRb0ntm3j+voai8UCm80Gq9UKTdMgCAJMJhPoui4ju0VRYDQayZB0kiRyjk3zw/+OXZblk4D3UHxU13VMp1O4rovD4SCvz7Is+Z8IOmdZhjzPZVD24uICcRxjs9kgTVNMJhPM53OEYSgjymmayqixCGG7rou7uztkWSbnX9wvEfT+8ccfAQCr1UpGz0VgW9zXyWRy9D4V67Cua1iWBc/zZKBXrEfHcbBarfDLL79gv9/LsK5t2zg/Pz96NsRnZ2dnSJIE6/UabdvCsiy5ftT3xHQ6leuiqioYhoHlcol/+7d/Q5IkSJLkSey6+2yK78XfirOzM9ze3mK1WqEsS1xdXWE6neKXX36R66FtW0ynUziOgzAM5fPl+z7Oz8/lPC8WC8xmM/zjH/+QAfHu36vpdIoff/wRbdvKQLn4u6RGiIdCvn9UfRHsvm36tu3GstVnj4iIiIiIiIiIiIiIiIiIiIiIiOhzY8yXiIiIiIiIiIiIiIiIiIiIiIiIiP6Q1ACgGoPsBmbH4zHm8zkOhwOSJOmN1ar7imOp8dvRaITpdCqjo5PJBHmeo6qqozGpscVuSFeNC6vUkKz6e/c6RIQVgAyuihCn53mYTqfI81x+po5DxGi70UTDMBAEAVzXRRRFyPNcnqsbRO3GSUV8VoRP4ziW82vbNubzORaLBUzTRNM0MtaaZRmqqjoZuxXbT6dTvHnzBoZhYLVayYCy7/vwfR+GYcjoqa7rsCwL4/FYhlKBD4HiJEngOA4uLi7w5s0bZFmG8XiMNE2R57mcm7Zt5THF9YrAbJZlOBwOR/PnOA5ms5kMItu2jfV6jcfHR6Rp+iQ8ORR0tixLztdoNDoKRldVhdFohIuLC9i2LbcvigJ5nsN1XcxmM1RVhe12K++/aZqYzWYYjUbY7XYoikLGkQHI+KmYl6qqsNvtUNc1DMOA67pwHAdVVR3FiS3Lgu/7qOsaaZqirms4joPlcoksy7Df74/WsogoF0Uho6++72M+n8PzPFRVhbqu4boubNtGlmWYTqfy36vVCkVRYLFYIEkSpGmKy8tLeV1JksC2bRRFIddVN+zZnX8Rxj07O8N+v8fj4yPyPJfXres60jRFFEUoikKuAxFrFs+JCCynaYrdbnf0zIjzdWPjYt1cXFzAsiwZc53P57BtW8aB8zxHlmWo6xpVVWE2m+Gnn37CZrPBw8ODDG+L8K+4bjFHdV3Ld9VoNEKaptjv9yjLErZtY7lcoq5rrFYrOI6Dv/zlL3BdF+/evcN2u5UxYOBDRFnXdcxmM1xdXUHXdWw2G6zXa6Rpiqqq5LnOz89hGAZ838d6vUaSJNB1Hbe3t9jtdjJyK541sdbFMzybzXBxcYG2bbFerxFFERzHQdu2KIri6N7ato2rqyvc3NwgyzLouo7FYoGLiwtMp1OEYXgU/1XvTd/7JwgC/OUvf0EURbi/v0ee5/B9X4Z2D4cD8jyXa3gymaCqKkRRJOd7NBphNBohDENYloWrqysURYHNZiPf24Ku6zLGbNs27u/vZURZ/ZvRXU/fk77r6fvb3I2/i5/f23wQERERERERERERERERERERERHRHwNjvkRERERERERERERERERERERERET0zVPDtt0gqhq57UYPRdBU13UZOu0eT4Q2RaTU931UVSXDraPRCL7vAwCSJJHhVhHAVUOd3bhgN8irfq6OV/1MDemq4xNE4LNpGhRFAcMwMB6P4bou1us16rruDQJ3o8IAYJomgiAAgN44cXe+xVzbto3xeIzlcilDsrvdTh5nOp1iuVxC0zRsNhtkWQYAT8Y2RERnr6+vYZom7u/vsd1uZUTVMAwZCRaxYtd1sVgssFgsYBgG0jSFruu4vLyU4ePpdCrHOJlMkGXZUbhU13V4ngfXdeVaOTs7w2QywcPDg4zRAh9CouK7PM9lLDcIAhlGruu6N8ipMk0T8/kc5+fnKMsSt7e3sCxLhj9938dPP/0E13WxWq1k3FfMg+d5Mu4r5lfM3w8//ICmabDdbqFpGnzfx/X1NbbbLcIwhKZpWC6XODs7w8PDw1HA+OzsDKZpyoCpON9oNMJyuZT3VQRJZ7MZbm9vn9zLIAjg+z622y2yLINt2zK2KuZdzHkURaiqCovFQkZr4ziG67q4urqS4eDr62us12tsNhukaQrTNNG27ZNwqxqIVj/zPA8//PADsizD/f09kiSR+4lnyzAM+Szato0gCNA0jVwD4v5fXl7i559/7n2Wh94HZ2dnuLq6QpqmiOMYbdvK84ogd5ZlMnw9Go3w008/oSgKPDw8yPHquo75fC4DwIfDQY5P3Efx/hPhXREn/+mnn/D+/XsYhoHz83PM53O8e/cOm80GRVHId4Wu60fPo2VZWK1WWK1WMvIsYtR/+ctfYJomHh4ekOc5NE1DEASoqgqbzUbeHzE/4jkWQWdxHNM0cXd3h8Ph8CTWqkbWp9MpDMPAfr8HABlFBiDnsCzLo4BuXwBWvP8XiwV838evv/4q16qIPt/e3spIsmEYmM1m8DwP6/VaXqv42yDWou/7CIJAxqLV84lnSYSrkyTBZrORcXE1MP69B2uH/hZ0/1b0Beb/LHNERERERERERERERERERERERERE3xbGfImIiIiIiIiIiIiIiIiIiIiIiIjom3cqijsU8gUA3/cxm80QxzHCMJQhVvVYmqbBNE1Mp1NcXl5iPB4jSRIZyHRdF1mWIYoiFEWBs7MzGZEV5xKRW/UzdUzd4O/Qdei6DtM0oeu6DNCqIcq2bVEUhQzNqteoadpR/FIdjwhDqudv2xau68pwrYgEd8OI6rhFrHY2m8G2bZimiaIoUBQFbNuG53kwTROLxQKmaWK1WmG9XqOqKhiGIY8t5mhoLoIgwJs3b2DbNu7u7rBer49CrVVVybBm27ZwHAfL5RLn5+cAgO12iyiKsFwuEQQB4jhGHMeYzWYyEjqZTOD7PqIokscW68B1XYRhCF3XMZvNMB6PcTgcZEBYxDxnsxmyLMNut0Nd1wiCAPP5/CgwPbR2gQ8x0+l0iqurKxRFgfv7e8RxDMdx5Hz/9a9/xXg8xt3dHXa7HWzbljFQcQ+yLEMYhijLEpqmwfM8XFxcYDab4bfffpPR3cvLS4xGI9zf38tI7Y8//oiqquT9F9d8cXEhA6NiDTqOg7OzM3ieJ48hYrBt28pos+u6Mmi6WCzgOA7yPJfb/vTTTzAMA2EYoigKGXtN0/QoGBuGIdq2xXg8hmmaSJIEhmHAdV0kSSLDp+I5EeNU17767Ig5XS6XGI/H+PXXX2VMt21bVFWFOI5lKDbPc+R5Dt/3ZWxYBHGn06kca5qmR8+p+tyIILD4XcSMRWR5t9shTVOUZYm6rlHXNaqqQlVV8hm9vr6G4zj4r//6L0RRJI/vOA6urq5QVZUMNIsgtwjjep6Hw+GA3W4nj3d2dobRaIQ8zzEajXBxcYEwDPHw8HAU8hWCIMDV1RVs28ZqtcLj4yPSNJVjmEwmuLq6wng8xvv377FerxHHMZbLJRaLBQ6HA5IkOZojXdfh+75cJ+K5F/O83W6P3rNVVR3FmT3Pw/n5uYx8i3CuWBPivSTOORR8Fe9c8V7bbDbY7XYwDANBEODs7AwAsN/v5d8P8c41DEPOl4jBO46DqqpkPLuua+z3+6P3r2ma8H0fnufJdRvHMQ6Hw9FaVsf8Z4jV9l1j92+10P27SERERERERERERERERERERERERPQlMeZLRERERERERERERERERERERERERN+0oeCrMBTyE1FS13WxWq2QZdlRAFAcwzRNTCYTnJ2dIQgCaJoG27axWCyQ5zniOMZ+v0ee5wA+xDhFKFL8HgQB6rqWUU6VGmQUkdfumDVNO4pviljpZrN5EsFs21b+bpomPM+TAeIkSZ4cXz1/N4zoeR4syzoKo3ajk5ZlydCo67q4uLiA67rY7XY4HA6o6xq2bWM6ncogZxAECMMQh8MBeZ6jbVsZwuyLaqrhW8MwZHT27u4Om81GBjPFHNZ1Le+nYRiYz+cyKHt/f4/D4QDXdeF5HsIwxO3tLYAPAci7uztEUYTJZALXdWUgVNyD0WgEwzAQRREcx5HBYtd1YZom6rqG53m4vr5GURTYbDaI4xiapmE0GgHA0RpQQ83da/Y8T0ZSb25uEEWRDJG2bYurqytcX1/jl19+wf39vQy11nUNx3FkRDSOY2RZJgO98/kcFxcXKIoC6/UadV1jNBrhxx9/lOdp2xaTyQSXl5f429/+JteZuMeO4+Dh4UGGakXoeD6fy+dJhJQvLi7w//7f/0NZljJAXJYlkiTBv/7rv0LXdSwWCywWC7x580YGZsuyRBzH2G63iOMYQRBguVzi8fERYRgC+BAGvrq6wuFwQJZlsG0bYRjCdV1YlnUU8e2uKzWiK6LbYs73+70M3KprUkSHl8slLMuSsdq2bWX41/d9+b5Yr9dI07T3OVfPK+657/vwfR93d3d4+/YtoihCVVXy+VDHbZqmjEa/f//+KAprWRbOz88RBAH+/ve/Y7PZyBCs+k4DPsSty7KUa+P8/FxGrMXz/Ouvvz55R4rzLJdL+L6P7XaLx8dHGYe2LAuTyQTn5+eYTCZIkgSr1Uq+S8bjMSzLwu3t7ZOQunjWRDBZxH0BIAxDGYYW24u1L6Lb8/kc4/EY7969QxAEsCwLWZahKApcXFygLMujmK94L3bfh7quy/eXaZq4ublB0zTwPA+e5wEAoiiSz6Z457quK2PilmXBNE04jiOfT8MwoOs60jSV70ARPxbR8zzPURQFJpOJDEVXVfUk5PtncOpa+8Lv3SD8n2muiIiIiIiIiIiIiIiIiIiIiIiI6OtjzJeIiIiIiIiIiIiIiIiIiIiIiIiI/lBEHFUNeKoxWEGEMEWEU4QXVbquy4CobdtYr9fI81zGIdM0xXa7lSFfx3FkaDfLMmiahtlshsVicRT8FecX8VcRd+0LEwMfwpYXFxe4uLhA0zRwHAe2bcvYY1mWME1ThmxF8NEwDPi+D8dxcH9/L88/dB6VaZoyPptlGaqqOgojGoaB8XiMIAiw2WxQVZWMdm42G+x2OxkPXSwW0HUdVVXB8zwYhiHHLuZc/Xkq0CzioKvVSoZB1fimbdvQdR1lWaJtW4xGI1xcXMA0Tfz222/YbDbQNA3n5+fQNA3v3r3DdrvFv/3bvwEA9vu9DIWKIK56H1zXRVEU0DQNruvKQLOIe9Z1jcVigeVyib/97W+Iogi6rmM6nWI6ncoQqRoRVWOugm3bWC6XmM1mePv2Lfb7vdxHBJJ//PFHHA4H3NzcyDFHUQTf97FYLGDbNqqqOloPo9EIZ2dnsCwLd3d3SNMUlmXh+voalmXh/v5erqfLy0s0TYPtdouqqqDrOiaTCSaTCXa7Hfb7PcqylMddLpcwTRObzQZ1XcuoquM4CMNQxoQvLi7kmC4vL7HdbrFYLDAej+Vavb29xX6/lwFTsd5M08RqtZLPgbi3t7e3aNsWVVXh7u4OZ2dnWCwW2G63yLIMdV0fhazV+yrugWVZmE6nCIIA//f//t+jNa9ua5qmfM+MRiOMRiMZyxYB2/F4jDzPsVqtUNe1DLCqY+geW0R2q6rCarWSQWMRbu1GrUejEebzOeI4lnMCQM7V9fU1kiTB3d2dXHMiEn1+fo7xeCzDyGKfy8tLOI6Du7s7GSoW91qNeYtxjEYjTKdTJEmCx8dHpGkKwzBg2zZs25ZR5aqqZNhYBJbF+y+O4yfHdl0Xo9FIRnfFMYuiQJIkR3Fj9b0nnsuzszN5r3zfR5IkCMMQjuPA9335XHSvp+/eiJhvVVVI0xRBEMD3fTRNg6qqEEXR0TvBNE00TSOfgdFoBF3X0TSNDE4LVVXBNE05T77vQ9d1xHGMOI4xHo8BQMaqxd8pxmk/6D5TwMti/kRERERERERERERERERERERERESfE2O+RERERERERERERERERERERERERPTNUyOXQ6HabqRRhDMPhwPSNO2NyDqOg9lshtFohM1mg4eHBxnMDIIAcRzLIK0IyZqmiTiOZaTx4uIClmVhvV7LAKVhGJhMJgiCAFEU4XA4yGPoun4UazQMAxcXF3jz5g3iOJZjME0TRVEAgIzF2raNMAwRx7G8xiAIAHyI1FZVdTQX6pyp/wEfwrWj0QhpmiJNUxlDFeFE3/dxeXkJXdex2WwQBAH+8pe/YLVa4fHxEXmey+uZTqeo61qGYw3DgGk+/7+pdu+HZVn44YcfAAC3t7fyvok5sCwLi8UCdV1jv9/Dsiwsl0sEQYCbm5ujyKxlWYiiCJvNBpZl4fz8XMZjRRTWMAwZGdV1XYY6syyDZVkwTRPb7RZlWcL3fSyXSznGpmkQxzFs25ahWxF3VaPKfaFpsT6Wy6W852p41DRNnJ2dYTQa4X//7/+NNE2P5qyqKjnWzWaDsixlmHqxWMDzPIRhiM1mA8MwsFgscHFxgffv38v48Hg8xvn5+dE8i0hqXdfYbrdI01QGTJfLJcbjMXa7HZIkkdHY5XKJKIqQJIkc//v37+G6LqbTKaIowvv37wF8CLiu12v8+uuvMiAsrsnzPMxmM2RZhiiK5Dlnsxlubm7kmq/rGpvNRj7fIriqxl+7a0z8J573PM+x3W57nxPLsjAej+X9mEwmck00TQPXdeG6Ltq2RRiGMmKsHkuN84qQs7j+yWQiw7Miht19d4lxzGYzOI6Dh4cHlGUpPxcRXtd18fj4eBS/Fc/I2dkZyrLEfr9HXdfwPA/z+RyTyQRRFCGKIozHY7Rti4eHh6PwtiCebXWdiViuCPWK8UdRJM8lxjmdTlEUhVxf4jkTc+F5Hg6Hg3yXApDxcjEPIhhclqV853ieh8lkIqPju90OYRiirmsEQYAgCJDnuRxf956o90k8b57nyVC253ky4gtARnY9z5Oh4KIo5DvUMAw5p1VVYTKZyOMbhoHpdArHceS79XA4yL8hlmWhKArs9/vee/C1dAO6n8tLzjEUZP5W5oqIiIiIiIiIiIiIiIiIiIiIiIj+fBjzJSIiIiIiIiIiIiIiIiIiIiIiIqJvmhpf7Pv30E/btjEej3F/fy8jiSLmKCKbk8kEs9kMRVFgvV4jTVPouo48z+G6LhzHgWmaMi7pOA6appFx1fF4jMVigbu7OyRJgqZpZBT2/PwcQRDIbUVMNAgChGEoQ6aj0Qj/8i//gt1uh59//hlZlsnopa7rqOsaruvi8vISABDH8VFc2PM8GUAV4UpBjVeKaxa/+74Px3Gw3W6P4rNi7haLBZbLJW5vbwEAy+UStm3LkK8Y42g0wmQywd3dHYqikNHT0WgE3/dRFMWTcfWNT0Rnl8sl/vGPfxwFmHVdh2EYmM/nuLq6wna7RZIkGI/HmEwmCMNQBnHF/S2KAqZpwnEcTKdT6LqO1WqFpmngOI6M+YrrNk0Tvu/L2KgIuB4OBwDA4XDAbDbDcrnExcUFbm9vMR6P4XkeXNdFHMfYbDZH968vPi2iu8vlErqu47fffpPzKXieh8vLS+z3e+z3+6N50nUdjuPA8zyYpom2beU1TSYTTKdTAEAURUjTFEEQ4OLiApqm4ebmRm57cXEBXddxd3cn520ymWA0GuFwOMg4qmVZ8jlpmga73Q51XcMwDIzHY4zHY6xWK7mei6LA+/fvMZlMYNs23r17h/fv32M+nyOOY6xWK+z3e/lMtW0Ly7IQBAF830cYhtA0Defn55hMJthsNlitVkfzk6YpHh4ecHFxgdFohCRJjsLXahxXXf+2bcP3fRwOB+R5Lp9XQawB0zQRhqG8V1VVoSxL2LYNy7JQ17WMy6rhWfWnusbF52Kt7Ha7o3eSGLcYr2EY8H0f8/kcZVkiDEMAH0K9vu/L0HSWZdhsNnLuDcPAaDTC5eUlLMvCarVCmqYyOB0EAdq2xW63Q9M0mE6nSNMUURTJsajPnLj3uq7LeLJpmtA0DVmWyfekpmmIoghZlsnxi/v58PCALMuezJHjODJkK0LE4n5YlgXXdeX9Eu8pEQo2TRO6ruNwOGCz2eBwOBx957ruURy3+98Qx3GwXC7l3wPxDlEjxlVV4XA4yOC0ZVky6pymqQxri3d3URRyjg6Hg4zLi3tm2zbSNJVx3z96oPZzR4DF+lSfqz/6nBEREREREREREREREREREREREdEfD2O+RERERERERERERERERERERERERPRNU4Oc3Uin+nt3W8dx4LousixDVVVHQVkR9pzNZjBNE3d3dzJo2TQNoiiS4cyyLLHb7WS4VkQsTdOUQVQRBwVwFLitqgpJkshg6Xw+x8XFBf7rv/4LVVXBtm1cX19D0zT88ssvMmALAHVdo2kamKaJxWIho6kijGkYBhzHgaZp2G63KMvyKPoqiGtSP7csC6PRCHmeI01TGdMUMcz5fI7Ly0vUdY39fi/js/f393KeRJz0/PwcSZLIoGae51itVpjNZphOp8iyTM6BmB8xJvG7ONabN28QRREeHx+PQqsivPzmzRvYto08z+X9A4DVaiXnXwQ7syzDYrHA+fk5fN+XYd62bWHbtoyhinGIQK74vWkaea/btsVms5FRZRFUvbi4QFmW2O/32G63R0HZbkhWcBwH8/kco9EIu91OxnoFca22bePt27cyFiuIMKvrujKerGkaXNeVIdqmaZAkCXRdx2w2w3w+x3a7RZqmMAwDk8kEZ2dniOMYURQB+BAQPjs7Q9M0R2Frx3EwGo1gWRbiOEaapjLYOp/P0TQNNpuNHF/TNMjzXN7n9Xotnxc1WqoyDAOe58FxHIRhiOVyCd/3sVqtsFqtZKhVvb9Jksh77Louoig6Wsfq+hdEBFasYUENbYsxJEmCq6srOW7LsqDrOsqyRJZlGI1GKMtSPltirXbDompAXMTBi6KQcyD2E8+eeK5FWPnu7g5t28rPHMfB4XCQxwnDEE3TyDmcTCYyIh5FEUzThOd5sCwLTdPIfcQ93263R/dDhLNFFFcN+TqOI9dMHMdyrKZpIooiVFUlzyeeze7x1ecAAJIkQVVVMpA8Ho+xXC7lsaqqwna7lbFyEcX9+9//ju12K+PmYt4Nw5D3uCzLJ38r+oh3g/h78fj4iDAMYRgGiqKA4zjy2bi/v0eSJPLe1nWN3W6HOI7RNI18b9u2DV3XEcexjB/HcSzf0+KetW2LMAyfBL2/to8ZS/dvz+c4BxEREREREREREREREREREREREdG3gDFfIiIiIiIiIiIiIiIiIiIiIiIiIvqm9QVqhb7ILwAZIc2yTMZQ1bCqrusYj8cy3KkGVdu2RZ7n2G63mE6nGI/HSJIEnudhOp3i/v4emqZhsVjg8vLyKOQIfAjlTiYTmKaJ/X6PoiigaRp835dByLquZbR1Pp/j3bt3RyFfcc1im8vLSxmEFGFMEc1s2xb7/f5JyFQcQz2mCOTatg3P81CWpQzQikDraDSSAdzVaoWqqnB+fg5N0/D4+CjPIwKXruvi5uYGWZbJuVutVjJe6vs+8jw/CrKq9xaADMP6vo+ff/5ZHkts77ouLi4u4Hke1uu1jKn6vo84jrHf749izU3TyBio67owTVMGM4EPEVNd12UYWcSdRYATgAwx13Ut47EiRPvw8ID7+3sZt03T9Gge1Z8qEXqezWYoyxLr9VquT3G9hmFgsVjIc4owsAiWep4H0zRRliXCMERRFDBNU0Z867qW4dAgCDAejwEAh8MBTdMgCAIsFgv4vo9ff/0VRVHAsiycnZ1hPp9js9nI2K3jOLAsS46vKAo5XhGXFfPSXbvT6RR1XcuQs+u68r50I56macL3fXieB9d1kSQJbm5uEIahXDcirF0Uhdw/yzLoui7nSJy/G+oVY1IDqiIUK4K0i8UCtm3jcDjgcDgAAEajkXwObdtGFEXyeoqiQJ7n8pkS609dt32xUnGfxL0W61/Exw3DQF3XSNNUruH5fA7XdeXaK4oCP/30E6qqQlmWMAxD3msRLK6qCo7jyPh1kiRwXVcGeQ3DkNcm4r3ivWnbtryO1WoFx3HkvQzDUD6fruvK+ypC0eIeGoYhtxfPZNu28rxiHsS7QQSzq6qS17rf7xHH8VHoVrzv1OOK+yyeXRHiFhF3dV81rixkWYaff/4ZpmnK8YjjbjYbGXW+vb2V0fLD4SBD2Or2bdsijmP57IrP1G3EZ03TYL/fy1DxH92XjvOq71qGgYmIiIiIiIiIiIiIiIiIiIiIiOhLYsyXiIiIiIiIiIiIiIiIiIiIiIiIiP4Q1GDfqVgmABmVFGFGdR8AsG0b4/FYxhRF5FUQsc4sy2RkczqdQtM0Gfa9uroCAGy3WxkY1TQNnufB930URYHD4YC2bWHbNiaTCVzXxd3dHaqqgm3bR1HXbsxRHGs+n8PzPGRZJqOWIpTreR6qqjoKAZ+aO7GNCG7GcYyyLOXnIqpq2zbSNEUYhjAMA77vI01TxHEs528+n+Py8lIGbkUgs65rRFEE27axWCwwHo+R57kMX4r5FfFNTdNkAHm322G73cpjicjuaDSSkdfdbgfgQ2hV3D8R0hXXKo4bRRFGoxFGoxFWq5WMKPu+L4OfYiwi5isCq3meH0VExfzkeY53797h/fv3KMuyN+Kq6/pgyHU0GsEwDNzf38s5EfdG7OO6LsqylNHpqqqQZRls24brujIEmmWZDMGKdS5Cv2LdqtHUyWQig69VVeFwOMAwDBmMtixLhqlHoxEsy0JZljLgK86h67oMub5//14+P2qceTweY7vdIk1T1HUN0zTld+r8iDU3nU4RBAH+3//7f7i5uZHPlAjpXl5eIgxDbLdbGbQW4xDzoepGwEVoum1bjEYjTKdTlGUprzMIAqzXaxnGtixLPgtlWWK32+FwOKAoChkFFve/e6/7AuTq/RLh3aZpYFkWxuOxfMY3mw0OhwNM08Td3R1msxlM08Tj4yPW67U8tmmayLIMdV1jNBrh6upKBmaLooDneWiaBpvNBpvNRj4z4n5WVYUwDDGZTBDHMeq6hm3bcm2EYSifN3GNagxXjSiL6wiCAE3TIEkSaJqGw+Eg3y/dORIBaBFEbtsWURTJQLPYRwSq1XlUjyfWiHj2i6LA+/fv5XupaZon96gb9BXx5G7kt2karNdr5HmONE2RJIlce5qmHYW41b9N4vjq357uuMXx1fn9PV4atP3WwrdqgPml+p43IiIiIiIiIiIiIiIiIiIiIiIioi+JMV8iIiIiIiIiIiIiIiIiIiIiIiIi+mZ144oi/AfgKJ7YZRgGXNeV4c3uMX3fPwrh9oUERZAxz3MZqhQxRxHmjeMYURShrmsZ2BTh2SiKZNRSxHyrqpJB3PF4jNlshsPhIEOS4pp0XYfjOLBtGwCQZRmiKJIxUhFwdRwHcRwjy7InY+8GRcVc6rqOIAigaRqyLJNxStM0Ydu2DGKKa9d1HaZpIo5jGIaB0WiE5XKJ8XgsY56u68qwaNu2KMsSh8NBBmVFqLSqqqNxiTGJAPLd3Z0Mw4p5830fdV2jLEsZ1PR9H67rPrl/hmHIKG9ZllitVpjNZqjrGmEYomkaedyiKGRoVgSXXddFXddomgZFURxFRE3TxHg8Rtu22G63R2Hl7vyq61MlArEAZEBVpeu6vBbP87BYLPDw8IA0TQEAl5eXcj7F9fq+jyiKUJalvIcikBtFEXzfx2w2w2g0wnw+l/FpYblcyrVUVRXquj5a32mawnVdGWwVsVfbtmUsthvMtm0blmVht9vJc+33e1xdXWGxWMj5NU0Tk8kEZ2dnME1TrjsR/hUxY7HGZrOZDKsCwNnZGRzHQVEUR2tLjX2r9yHLMtzd3eHq6gqGYaAoCriui91uh7dv3yJJErl90zQ4HA6wbRsPDw9Yr9cyMFwUBbbbLZqmkbHavkhq9/0VhiFub28xmUwwm82gaRqqqkKSJFiv19jtdvI5yfMcv/32G25ubtC2rbw+cY/LsoRlWTg/P8d8Psd6vcbj4yMsy8Jms4Hrunh4eMBqtUKe5zBNE2EYyvWfpinu7u5weXmJxWIhzxmGoQycD0VSxfWK++g4DhaLBeI4RhiG8lhqkFfsJ9bKw8MDAMjtxDZlWT6JMKvhXnVM3XtdVRXW67W8N+p51b8X3WdUfY675xLv+D7dSHp3f/UzEexV10lfhPpjvTRo+6XDt6fiwc9F+Z/DiC8RERERERERERERERERERERERF9LYz5EhEREREREREREREREREREREREdEfTjfi1w3XAh/iq2pQUhDxWMuykOc5qqrqDQ6qgVPXdeG6Lu7u7mCaJkzThKZpMrArjiuivbquI01TlGUJXdfh+z5s20YYhkiSBLquYz6fwzAMGZ5UY5MiLCtCu03TIE1TGUbVdV2Gd6MoktfQvQ71MzV0almWDG42TQPHceD7PnRdl2FU8e+2baHrOmazGYIgQBAESNMU2+1WBpN1XYdhGEf3IMsyxHEM13XhOA4Mw5DXI8bUNA1M08RisYBlWTgcDjK4Ox6PMZlMZPzY8zzs93s5pyL+WpYlgH+GcoMgQBzHMj4sjiFCrSL0LMKpIi4qArR1XSPLMhlpFXRdh+d5aNsWeZ4/WW/ivolxqfuKYKdlWXBdF23bHkWI1RDwaDSC7/uYTCb4j//4DxwOB+i6jqurK5ydneGXX36B67pyjvb7PcIwhGVZ8j4eDgdst1tsNhuYpokffvgBZ2dnuL+/x8PDA3744Qc4joPlcgkAuL29xXQ6xXw+x8XFBYqiwGq1wuFwQFEUyLJMBnTFfBmGgf1+L++PGlYVoewoiuQ13t/fQ9M0TKdTTKdTeawoinBzc4OqqnBxcYHr62vMZjOUZYn9fo8kSfD4+IgkSfA//+f/xP/4H/9DRrLH4zHevXsnQ819AWv1HpRliZubGyRJgtFohKqq8PDwgDAMj4LEIkj9yy+/AICMgovnqK5rGX3uPm/qv7tR7bIscXt7i/1+L6PIeZ7L95CIKQvidzXyrUZmf/zxR5imiZubG6xWKxm9/s///E85v2L/pmmw3W5lfFzErvf7PTRNQ1mWTwLWpzRNgyzLsFqt5HswiiIZFu/OpyDmT6ybLjFXauRWDWaLcLoaglW/F8c/FSIW75hu9LkbERbnOzVOdXv1c/WY3SDxn0VfaL/rY2LG3efsUwWRiYiIiIiIiIiIiIiIiIiIiIiIiF6KMV8iIiIiIiIiIiIiIiIiIiIiIiIi+mbpug5d13vDmEB/KBLAUQR3KCio6/rgeR3HwWKxgGmaSJIEV1dXqKoKTdPIWK+u68iyDFVVoW1b2LaNxWKB5XIpI50AMB6P8ebNG8znc2w2GzRNI8OweZ4jjuOjEORsNoNlWYiiCGVZwrZtAJDnAj6EVEUMNI7jk9epzpWYI9M00batjPbOZjM4jiPjrU3TwLZtNE2Doiiw2WxwdnaG5XKJMAzx9u1bNE2DN2/eYLvdAgCWyyWiKJJx4bZtURQFbNuGYRh4fHyUY+nGND3Pg2mayLIMhmFgsVjg/Pwct7e3OBwOOD8/h2EYMjQs5lBEfcfjMXzfh+M4CMMQYRjKCLEI54oQsoj5qvMpuK4rw78i5qvOa1mWGI1Gcu2IawiCAIvFAmma4nA4PAl2iu2apjlam+pa9H0fZ2dnGI1GuLm5gW3bePPmDYIgkGvil19+QRzHyLIM8/kc2+1WRok1TcN+v0dd13h8fMR6vUZd11itVvjb3/6GqqrkZ/v9Hnd3d6jrGvf394iiCHd3d9B1XUZfd7udnLMkSfDu3Tu5DnVdx263w263611bruvKuKyQJAnev3+P9XoN0/zwvzAXRSFDtgAQhiFub2/hui7yPMd+v5ffxXGM//iP/8D5+TkWiwWyLMN//ud/YrPZyHGq8yriyOpndV2jKAo8PDzg4eHh6L4I4j3Tti2iKJLHUa+xe1/7Aq3duLA4T5Ik8n51w7NqcFudVzE+8TNNU7x9+xar1QpVVSHLsqOwcDduKgK2dV3LELUIFov47mu1bSuj1+I5U8fQF1jvxpXVz/oC7erP7rHU/cWzrkbC1XN275O4v2K9d8956l2qhnmHtusG5P+sTs3B0HPT/Vw8w+qcq88FQ75ERERERERERERERERERERERET0NTDmS0RERERERERERERERERERERERETfNDWY2I0xqsFONYBZ1zXSNMVisTiKPAIfAoB5nqNpGpimebSfpmlwHAfL5RJBEOD+/h5t22I8HuPx8RG+78v4pQjC1nUNwzBwdnaGIAhgGAbyPEdZljAMA0EQwHEcGIYho5e6rsNxHBmnBQDTNHFxcQHTNPHw8IA8z+G6LsbjMbIsk+cS2zqOg6qqkKbpk3lRiVhoN0hqmiYsy8JyuYTv+zgcDoiiCAAQRREmkwkWiwWiKJLRy9vbW7x79w5xHGOxWKBtW4RhiKZpcHZ2hqurK2RZhsPhgDRN4fs+LMs6ivB2Y5umacpt37x5A9/3YRgG3r9/j91uJ8cexzGKokDbtsjzHEmSwHVd/PWvfwXwIRb78PCAKIrkHJimiTiOZTBZ0zR5D9M0lfddhJF3ux02mw2yLHsyj03T4HA44OzsDP/2b/+G7XaLpmkwGo3g+z52ux2iKJIRZ3FsNWpa1zWyLMNkMsGbN28QhiEsy8JsNoNpmjgcDvjHP/4hI8o//PADxuMx1us1fvvtN8RxDMMw8PDwgM1mgyiK5HoUgVc1fA0AeZ7j5ubm6Hr2+z3+z//5PzLoDACr1UreyzzPjyKzRVHIuLG4DjVarD6jpmnKuequyyzLZHi3G2tVn82+sGvTNIiiCHEc49dff4WmaSjLUq6p7toX8Vz1XdEXEB0Kr/a9d9TfXxqh7buOuq6fhHrVuRx6ntVz5nkuw7zindfdrnvs5675tcT1DJ1HvZ5uwHsohNt3LHW/vvOI4xuGcfS9eq6+QGz3GH2G/v6o33XvHUO+p3XvvWEYg3PXnePu86x+R0RERERERERERERERERERERERPSlMOZLRERERERERERERERERERERERERN+0vmBoN57YDfa2bYv1eo2//OUv8H0f+/3+aJ8oipCmKYIgwGKxAPAhIul5HsbjsQzqhmGI8XgM3/cRBAHiOMZut8PZ2RlM00RZlkdB3DAMEQQBLMvCaDRCEATwPA+r1UoGJEV0MkkSXF5eYjqdQtd1zGYzlGWJ9+/fyyCsaX74Xz1FHFiM3zAMVFUlA79Dwc++OKII3I5GI1xfXyPPc+x2O+x2O5RlKcO5m81GXnvTNLi7u8N2u5UBUV3XcTgc0DQNyrLEer1GnueYTqf413/9V1iWBQDY7XZYrVZHsVExTnHP1us13r59K8O1q9UKRVHI4DIAVFUlryVJEtzd3WGxWMA0Tez3e2w2GxnoFcfXdR3r9RphGKJtWxmN7AZ7m6bBdrtFmqZIkgRFUTwJTpZlic1mA9d1EQQBJpMJyrJEFEVybkSodkhRFNhut7BtG9fX1zg7O0Oaptjv99jtdkjTVMZM3759i7u7O7Rti7quZZhWRIfV9SzmUcyxWGvq86Bej4gFq3NVlqVcYy+Jv4pxdoOaZVliv9/L+K+6Xzfk3Bff7fuu7x0gxqDuox5LDX6qoVARlVXnqBtp7b5Pusc+NS9Dz+LQdXSPp15PN0DcNI0cWzd6OxSYPTW+T6Fv/tXx942t+2/VUAy3+z5T71H3ndc3vqH461A8dshLQs7d452ao+/dUEBbJZ7J7nfq/TVN88mzynAyERERERERERERERERERERERERfS2M+RIRERERERERERERERERERERERHRN+0lIcxurBT4EHzd7XZYLBZYr9dH0duyLLHdbmFZFhaLBVzXRVVVaJoGVVVhs9kgjmNomgbf92HbNizLwnq9RlEU8r+//vWvAD7Edu/v75FlGcbjMa6vr/Hf/tt/w3q9xsPDA4APIVfHcaBpGuq6xuPjI6bTKf793/8dcRzj7u4O6/UaVVUdXct6vUaSJEehVk3TkGUZ4jiWkVfxnRo4VKOlImbZNA3iOJaB48PhgMPhgLIsZTxRnM91XTn2LMtQVZUcV5ZlWK/XMiKbZRmKokCSJAjDEJZlyditGqlV76dhGGjbFvv9HmEYAngaB23bFmVZIssyeT1VVWG9XmOz2Rwds0vcFzF3TdMgiiLEcSxjs+Lzw+Ego799xyvLErvdDkVRwPd9GXOO41hGlbu60VARBA7DEIZhyPUm1p7YTtwHEdzt3tOhGOlQZHRoXGrg9NS2Yo11t+3bv65r+ax1Q6pqVPc1Y+7bXpxHPbbYtvuu6L4zuuFvNZJ76txD4+2bG/U78b16DvU+i+/UZ1ns1w0zq+vs1Ji/pcjpUJxXvRd90Vd1vajbd4/bFzA+dbyXjvVUGP1UqPi5a/8zGLpXfdt1v39JTHzofUhERERERERERERERERERERERET0JWj8n9eIiIiIiIiIiIiIiIiIiIiIiIjoc9M0jf+zGn0U0zTlv/vCgCJ02f1e0zQ4joPr62ssFgvc3d1hs9mgKAr53Wg0gud50DQNVVWhLEvkeY40TWXo1TAMLBYLXF1dYbVa4eHhAXVdIwgCXF5ewvM8pGmK7XYrQ7BBEGA8HgMADocDkiSB4ziYz+cwTRMPDw9I0xSapsGyLOi6jrquj+Kygq7rMtArwqimacL3fRiGgSRJkKbpYFCyGzoUgUXDMGRMVoRXX/L/lKpRTXFvyrI8CvCK44tzqxFl9RwiRDoUlFWvyTRNmKYp47fduGY30No3F6/RN171PGpg9bmw7tB3fRHKvvv4MbHeru5YxP0Sa6tPN3bbd371HortRKC5ez19Md++cw/dNxHcFWPpi1h3z6kGcdX5Vt8b4jhq9LppmqP4dV+Y9CXXIH6qz4d6rr7zD8WDxfWrx+ub175zfi1i3tUQOYAn6+klTgWTxbHEOqjr+miOu/Mtjjd0ju466bvXfc/ixz6fL/FHitYOxbWHtu36FIHktm0//g8AERERERERERERERERERERERER0TMY8yUiIiIiIiIiIiIiIiIiIiIiIqLPjjFf+lgiZisMxWn7IraGYWAymeD8/By6riOKIlRVJY9ZVRXSNEWe52iaRv6nhmJF+NeyLOR5jqIoAACGYcC2bZimibIsj4K2IpQropIiwmlZFjRNQ1EUMnDbDR6K+OUphmHANE20bSsDwENB0aEw6NDn6nfdoKIadVWjnN2Q79C5+sbz3P1VP++GZ4fOe+qzU9fdt+3Q9qeCs+L7oSBv3/7qnD+372sCuH3UOT4VCO2ug6Hzqve5qioZch7aTmyrrh3xXXf9q/MxFGHtG2c3MKyuMXE+8Zl4FsUzOxQS7Qsf90V91eCz+EwND4tt+q6n7/npjrv7HKrH6pufr6l777q/d7cFnr53huLEQ2use67uelDvydCY1X8/N5/f0nx/TS95379kv+e2625/4v3FmC8RERERERERERERERERERERERF9NubXHgARERERERERERERERERERERERERUR8RxFSjgGoAV2yj/rsbew3DEAAwmUzguq78PEkSRFGELMtQ17XcpxvybdsWWZahKIqjIGdd10jTtDcuLAK+6vHEZ93xq2HJbth3aE7quj4ac3ceTkURT4V6++KYp6hzpW6vhky73w8FUfvOqYY5++buSxua475opbqGngu09lGvdSjK/Hsios9FjPvOpQZx1evrjkcNp6rX0o1Xi227YVpxbMMwnpxP/V29BvXZ7M63aZryPOqYuutTHZe41r45Vq/nlG5UtnvMvmeh7/jqdt3Phu7j710fn1J3HEMR3eei0kO660Dorifx39D5h87dXdfdMb303fu9G3qvn4r1vmadDs05ERERERERERERERERERERERER0dfAmC8RERERERERERERERERERERERERfZP6Qr1qPLQbCez72bYtDocD8jyHZVnQNA1VVaEsS1RVdRT4HDq3ekz1uK8JovaN79S23fDnS4KHrwlJfurYZ/e+dCPBfZ8NHQP4EOUUMU71OxE9fu5YL9UXrj31+UuPcSo6LfZ5aUC5G5ztxkO7c9Q3pt9DXYOnYsyapsEwjN5xDMWO+4LH4vuqquTvfeupe271uF1qSFqEsIdio0PPvnhX9I25+845Nca+91o3LKwGfdVwd9/z8Jq1+qW95F13KgQ9pG/NqMfXdf3J3LzmmXhNWPjPHPB9zudai6/5u0hERERERERERERERERERERERET0OTDmS0RERERERERERERERERERERERETfrKHY56ntm6YB8M8gblVVqOtaxmH74qinYqnqv7vRSeCfgdnu/n3BUnEuMUaxrWEYvecX23fjvqeuXz2GpmnyXN1rfW2MWMQx1WvoXttQxLQvvNy3nRrjPBVZHYq29l1n33V/zghkN9J7Kjj7XNhYDck+9xwMhVPVAO/Q/i9dV33zrh5b3Dd1zXW3Va+j+zyeet661zEU8+4blzjWc89q3/Wqn3Wf2yEvib4OxYRPHUMYmt+X7Put6Qv5Dr2P1e26n6n35jVzOTSe7vFfui19oL4LnntXAy+by27s+XO/y4mIiIiIiIiIiIiIiIiIiIiIiIiGMOZLRERERERERERERERERERERERERN+kbhx2KKjYF89V9UV1+77vC92eCq2KfbvjHdpW/UyNhaqRYfFZX3B06HinQq2nordD0dluRLXvWtVAY9/4mqYZDJ/2zfNzcdvu8ftiwH3b9X1/KuSsrpG+sfTFI18Tozx1PUPj7q7bl4Snu5+/NIbdNzZV9xq7UdlTIWx1bXS36Qu5inivGjQWP9W49XOh4m7At083Djp039Ux9m3XHXP3HH37vCZM/TnDpZ87jHpqboTnYtHP6UaiX+olzzWjsa+nvrueC1C/Rl/0mYiIiIiIiIiIiIiIiIiIiIiIiOhLYsyXiIiIiIiIiIiIiIiIiIiIiIiIiL5ZItrZpYYd1VhnX0D2Y0Km3f26sV/x2XOeC1cahtEb8n0uEPuSwGVfaFWNID4XvOwGTtXjPDeObqS1q+8a+6KrQ8fvi/r27TcUlO1uN/TZc9fR3X5oLN1I7EuO9dx3zwWLhyLHQ/f1pU7N2an49VDAGYAMWg/Fgfue6+41norkqtf6mhjwa2LefZFo9f3Vfca743/uWe8zFH/+FnxM5PrUGhGfd+dMzHH33faSeLB6rr7jd8dFL6c+sx/zvntue94TIiIiIiIiIiIiIiIiIiIiIiIi+poY8yUiIiIiIiIiIiIiIiIiIiIiIiKib9Zzsc6uU+HAU8fsBnXVn+K7bmj01DmHIpLdeOhLIodqoPK5OeiLUb4mpDgUo33JmF8S/VW/U+ObauBU1/WjEOdz4+8Lf3bDqS/xMXHblwae+0K+L42vnhrXqfjoULy4L5r83JhfOj5x/14Td+6umefmR1073c/VY/ZFq/vCveqz1X1u1XX4XBy5O1Yx1+IY6rmGrr/vmEPUIPGn8LniqH3zeur86r0YCpurxxt67rufPXfe1+5HL/Oad/BzPtVxiIiIiIiIiIiIiIiIiIiIiIiIiH4vxnyJiIiIiIiIiIiIiIiIiIiIiIiI6JvTjX6+NACqhjO72740YDq0jxrZ7P5b7KPGQ4dis91IpQh+9m3bN6YuNc7anTd1PN1wcd+1nNINa7400DpEHEOMvxu7VY/zXCx5KPD60rH0RZ37zjv02UuOPTSevvul/rtvvfR5aZy5e67uWuuL4qr3SB1Td813A83AP6Oz6nj65mQo3Dq0boeuQd1vKJTbneeh7Z/7XB3HEPVc4t511/tLYrd9x/2Wnbo3z+136r69ZJ9vfW7+bD71/eD9JSIiIiIiIiIiIiIiIiIiIiIioq+NMV8iIiIiIiIiIiIiIiIiIiIiIiIi+ia1bdsbAhXf9X3+Med4Tex16N9q8LR73L5Ya98xXhK+HIqtdudKfNc35qHrGophdu9B91hq4HUopvuS8/QdV2yjzm13v6HrGTr+c8HeU3P13P3pi9EOjWco7ty3fTfGrH6vjqt7rL54bV9ceui6h+5p371Qtx0Ksg4df+i+9QWx+wLE6k91+1Mx5u74Tt3bvnvfvd5Ta7wb2x66hufG8dLxfiteM8evOc5LvvuYc31MfJiexzklIiIiIiIiIiIiIiIiIiIiIiKi7w1jvkRERERERERERERERERERERERET0zVEjtd04qbrNEDWa+ZLzdPftO/5QBLZ7jL6Qb18kcigQ/NxYXhI4BSADuH1j6M5N3/hPRXtfEsDt2/b3xjxPzVlfELYbOH7u+H3HUT87FRF+zbV1o7fd6O5zAWN1PvrWWve7vjXTd31DhtZhd3xN0zyZ86Zp0LYtDMM42rb7jHeP2xeq1jQNuq6/OMB9Klg8FOdV1624B33B4OdCtaeeu+71f+9e83x8zLPUF0j+mDDvn+FefEmfK47M6DIRERERERERERERERERERERERF9bYz5EhEREREREREREREREREREREREdE3qRvkfUm8tRv5Gwr+9cU5T23XPZ8aIO0LdZ46Vne/5wKU3fN246cviRafGufQ5925V88hgqrd4GxfUFaN0/Zdowi09o1XfH4qlNq9vpcYCrQO6dtm6LO+ddl3XjEf3e3EPW6aRgZw+yKw3etRf3aPq26vRp6fW6/qmPrWrbou+p479Rjd+/9cyFZEgJ8LLfd9/pLrUudAhHu7Y1eP/ZL3Snc8pyLFLznOl/Q5A6kfE7pWf39JlF31kvehuu23dB++F0Mh++fC6KfwPhEREREREREREREREREREREREdG3gjFfIiIiIiIiIiIiIiIiIiIiIiIiIvrm6LoOXdcH46R9nw3FNoeCr6dind19+8Kl3cBtXxy1Gxh+aWy4O271GsV/Yn7Ez+41n5q7U2HkbuS0L9bb3a7vWENhW3Xsz0VXT0VqP9Zzc/4xcd9u2Fidg745V9fF0H3qW5d94zsVyhy6R13q2PtCvN1zi+3Fz24sWN23G37uhny7gdvuNYn9uxFeVd9zdSrA3J3/vvGo+3zs+vuS8dHfE0n9PfsN+ZSR3L5Y76ljP/fMqN/R53HquWFAmYiIiIiIiIiIiIiIiIiIiIiIiP7oGPMlIiIiIiIiIiIiIiIiIiIiIiIiom+OYRjQdR3A0yBu16lAb/f7odhpN+rZFwxVg599UdA+Q2Hbl8SGh+Kc3WP2xV270V0x9qZpYBjGk2s5FbVUt+vek74g7XORxqFAr4jDqmNWf/ZFX7vz0Tc/fdfS3fe5a39pKHUowNvd5rl7q0aa67p+cq+763XoHvTN9XORbHWfU8fu/uxeX13Xcj+xbrr79QWOn7tnAOQ6Gfr91HvgJXHjvkBx93gf46Uh79f63qK0p96pIu6sbgu8PNLdfY7o8zkV1P698WkiIiIiIiIiIiIiIiIiIiIiIiKir4kxXyIiIiIiIiIiIiIiIiIiIiIiIiL6pr02gqt+NxTh7O7bFy3tEjFcEYMU/+6OQcQm1f/Uc/YFg0+FZ7vjU2OU4vq6kdCmadA0zZPwpdjnVEjxVOhV6DvuqbCwiNOK61Djruo8qNup26vfd8/z3NydutahoGx3+278sxvJ7TvOqXjsc6HY59ZF37G7/+4eZ+i4p9b8a8On6hi6AV8RERbrU72/3e26Uefu8U9Fnfu8ZK32neslke7utmrwunvuzxkw7Qaw+777WoZiukNOfS/eXy8NGDMW++0ZelZOYfyXiIiIiIiIiIiIiIiIiIiIiIiIviWM+RIRERERERERERERERERERERERHRN6cvrtsXxhWfC2rocyjS+nvGpB6jrmsZKO2OsS+e2jTNUeDTMIyj8Q/FX/uONxQTFRHUU+NW9YVi1XP0BTj7orN9Y+qepy/C2z1GNwCrxpO72/bFjZ+77qG47ZBugLnvu75zdePLQwFa9TN1LYl9m6aR60ycU9f1o3lSr6dvDk7pi+mq16Gu2e5+3Rh139rvRovVsHPfsfvupWmaaNsWdV0PXlc3ANwXCj6lL1r8kjkeWv+nIqVibM9FtV87ZnWtdef2awRQTz0vYn763lcvPXZ3/T1HrAux/ceemz6d1zyjr4lBExEREREREREREREREREREREREX1ujPkSERERERERERERERERERERERER0TflVKRRxP/U0GY3lvracG83kKuety8G+1wk81R0sC+U26cbQRXXNDSubghYnYNu7PjU933R4FPBxZfGcPvGMXTN6rm7seS+Yw1FntXvT41z6F78nvizGg5Vx9uN9w6d57UR4r6YcHe7vvPUdd27HkQQVl173UCteA67UdS+9aZpGuq6RtM0TyK5ffsMzcFQVFkd51Doui98e+oZVOPAQ+Ptzpn6/dC7QWxrGIYcgxoqfgkRBO4b01D4+nuh6zpM04SmaSjLUn7evZ99QWc1oExf3/e0LomIiIiIiIiIiIiIiIiIiIiIiOjPhTFfIiIiIiIiIiIiIiIiIiIiIiIiIvqmPBfBFbpRzv+PnT/3sSRL9Dy/n+273d2XWDKrXjX7oVujNAIVApQ4ymijERyCwCikQIACB/wLRiIwEoEGKHAAAiQBEiAFKgQBChRIoQeNbqB7qt7Ll5kVi7vffbF9uxSizim75nave0RkZkRm/T5AItzvteXYsWPmJRS+50Kf3fDruRBu3/nO7fOcY3QDu+Kz7rbtSOpTxz0X4e07djdmem6/S+MWPiWO3P793DkvRXiBv8Y323PYPmY34PqcaG93Ps6N+Tnja29/bt12I6vnQtTniOir0Bcv7osvd499ab10I9jdtdS+ByJy241nd+9T+7t2gLu9v/i3HV7VNO3ROLux23Nz3Z6nvufqUri669zzeClQ3RfUbc+Z2FfEhz/Gpehz+163A9jH4/GTzvVzOBcf/hiqqiIIAkynU1RVhfv7e1RVJb8/dw/6zvvUc0c/rZ8qLs17RkRERERERERERERERERERERERF8DxnyJiIiIiIiIiIiIiIiIiIiIiIiI6KvVF988931f1LUb+HzqHO19298959x9+/SFY8+dqx3gFOOu6/pRGPSpWG7790vh0XP7donQ6sfoC2X2xXfbAddz13Yu5ntuzH1h327ks299PLXW2sfTdR2WZUFRFOR5jqIoeufh0vrpu9/deVNVVYZgxbYicNyO1fYduxu5bkeEu2Poi153xwWcRmjba7t7f7oh3+49akeaRXxWHLcdvn0qzt291vY1tefmYwLWfcHtvm37PCcy/dTx+3TvdfeZ1TQNlmXBcRw4jgPTNFFVFZIkQRzHyPP85F3yS/spQr4AYBgGptMpbm9vcX9/L+elu7bFc/IUBn1/Xufml/NOREREREREREREREREREREREREv3aM+RIRERERERERERERERERERERERHRV68vLtqNsF4K9z4nJikCgyICeS5+K0KkmqadPVb7M03T5HHbYxUR027Yth1rPRfjbY+pjwjwtmOqYn8xnr59zh2/L/raF3C9FNzti+52Q8Ht8Kz4txtrfc48dNdGO2p7aU7but+pqgrHcTCZTDAcDmGaJnRdR5IkePv2Ldbrde989R2vHeNtrzfxc/u79vW3151Yi23PCZi2x9ON/orj9q39boS2ff+63507dnes3XXRDReL79trtu+Y3XnrXmdf6Pfcsfr27+57bq7a3z33965zge9z2/q+j6urK9zc3GA2m8F1XRRFgSiKsNvtMJ/PsVgsEMcxqqq6eO4v7dKzqWmajBVXVYUsywDgZA12j9MXyO57Z51bIz91cFZVVaiq+kXDyj+3c/P3W71eIiIiIiIiIiIiIiIiIiIiIiIi+tvEmC8RERERERERERERERERERERERERfVXakVIRP+wLyLZjn+K77nEMw4Bt29A0DWVZIs9zFEVxNqwqnIv9ds/dPW83OHzu+sR1tc/VjrX2BVKfQ5yzL+qqKApM00QQBIiiSM7DU+PtHrsvYNr+/rnR4XPx3+72fce7FFPtO+9TodqnzmdZFsbjMcbjMTRNQ5qm2O/3ME0TYRjid7/7Heq6xm63OzsWcS5VVeG6LjzPQ1mW2Gw2j+5XN0LcjTv3xXMvXX/3Z3G+c3N9KZ7cvp7uHLefo0v3U1XVRzHfvnnvhlb71ms7AN2NC3cDrZeeye51P+f3vlDvufB13/PcDTp3dd9x7f0URcFwOMS3336Lm5sb6LqO1WqFP/7xj9hut9A0TYanq6pCVVWPouI/t48J4/a9v8VaVRQFuq4jCAKYpimfv/Y1iTlpR7EBoK7rJ8fUfuc+J/D8KXRdx+3tLRzHwdu3b5Gm6W82cPtzXddTgXAiIiIiIiIiIiIiIiIiIiIiIiKiXwpjvkRERERERERERERERERERERERET01WkHF9vxyUvB2XYM17ZtBEEA3/dhmiZUVUVVVUiSBOv1GlEUnQRNu8fvi2iKbbrh3+64+8YoYqztgGk3ltoNE7fjqH3b9B2/vW13vkzTxNXVFXzfR1EUMoTZ3a47px+rL7B7bkzdMOq5AO258XT3OXc93fvaN/d9RHh3Op0iDEPEcYzNZoMkSVDXNXRdx8uXL3Fzc4PRaIQoih7FQ7t0XYfv+xgMBtjtdr33XfysaZpcp5eCvd011Xesc9uL56s79+eiyU/dl6fCwue2OXfsc+NqH6MdxW1vcy4UfOn4fd9dirxeeiaFS+d/6hnrrlVN06CqKgaDAQaDASzLwn6/x48//ojlcikj3ZZlQdM0DIdD2LYN13VRVRXKsvwqQ6jdNdimKAps28bt7S2KopBRdrF2DcPAv/gX/wJpmuLNmzfIskwe89y52sS6ubTP5xBh5ZcvXwIAVqsVsiz7Ku/D5/o5r4kRXyIiIiIiIiIiIiIiIiIiIiIiIvpaMOZLRERERERERERERERERERERERERF8VVVVlWLEd8hUuxTcNw8BoNMJoNIKu6zLgCwC2bWM4HEJRFNR1jTRNz0Zdz4U4LwWA+/brBoDb19YOjj4VKDwXDD035r59Pc/DaDR6VsT3uZ4TST0XQO07/1Nz0Q1uPjX+p6LJl46haRo8z8NsNoNlWVgul1iv1zKWCgBVVSGKopPA87nrFue1LAu+70PTNOR5/uh6TdOEbdswDAMAkCSJDJc+NeZLod3nrJPnBJXbxxXPZzs8/ZTj8SiDx92QcXu8fWMT33e3aYeszwWI2++U9n7tSO9TUd5L350797nr6IsDi3FeutfH41HGYQeDAQzDwH/4D/8Bm80GVVXJ7ZqmQZ7nWK1Wci3HcYyiKM5ey3PfR33X0+fSu/s55xDXapomgiCAYRiIogiHw0G+x1RVheM4GAwGME0Tm80GRVFcjGr3rbWfKxKraZqMd4u/OXVdM0pLRERERERERERERERERERERERE9CvGmC8RERERERERERERERERERERERERfTX64q7duGhfnFREUkXgMo5jvH//HmmaomkaKIqCIAhwc3MDz/PgOA7yPJcx0nOR2HPBRbGfiIH2BTfPHacv8ts3B30h2nNz1D1v9ziqqsoY5m63O4l+9o37U53b/2MCoW2WZUHTNFRVhbIsT+LO5+ahb84+ZsziOLZtYzwewzAMLJfLR7HU7j5Zll28fgAyPGpZFvI8fxTzFWvp+voaL168gKZp+O6773B3d3eybrrh2O61dCO5fetTrPtza7e9bzcu2zeO9vHENk3TQNf/+n9X7t4/x3EwHo+hqirm8znKsnxWqLn73aV10Tc33WvXNE0GVvsCu33z0j1X37x0t1NV9VFkthtR7pt/8bn42bIs3NzcQFEU/Pt//+/lM91+r6iqijzPkSSJjOG6ros0TXvnsy8ufEl7n4/xVPy3+44Xz8x4PJb3KUkS+V4/Ho84HA741//6X8MwjN5YcXceBV3X5Vwdj0f5jjm3/cdSVRW2bSMMQzRNg91uB13XzwaViYiIiIiIiIiIiIiIiIiIiIiIiOjXgTFfIiIiIiIiIiIiIiIiIiIiIiIiIvpqHI9H1HUtI7fPjSo6joPpdArLsjCfz7Hb7VAUxUm0M0kSVFUF27ZhGAZUVT2JlHb1RXXb41AUBZqmnWzbF1kV/3bDwd1jnQv7dsOj3WN0t+uLgVqWheFwiOPxiCRJHgVFL7kUVb003kvHOXcvxeemaeL29hYvXrxAnueYz+dYLpcnAea+fbufd6Oq4rOnrknTNARBANu2sd1uZci3L+jqui5UVUWSJGdDsIK4D47jYLvdPop6iqDo/f09FosFmqZBmqYXI8J989CNzD419yJMfe77drj1XMwXgHx2u2u/b12aponJZILr62vs93v5LLbH0Y4Dt5/Vc89j+/iX3hft43b36T7D52LPYru+5/acvmOde2779hEB5JcvX0JRFPzxj3/EdruVEdr2tR2PRxmYNk0TqqrCsiwZFL4UcW7TNA2apsl3szjPpTk6d42XtukLqquqCsMwMBwOEYYhkiRBlmWoqupknTRNgyRJoGnao3nQNA2macIwDHkNiqLAMAxYlgXDMKAoCsqyxH6/RxRFj+a8e31iLsV8Nk2Duq5RFAXKspTzZhgGxuMxPM9DHMfQNA1ZlsltPjVy/rX4KYLHRERERERERERERERERERERERERL9GjPkSERERERERERERERERERERERH9jVMU5TWA/xrANYAjgH91PB7/K0VRxgD+TwB+B+AHAP/p8XjcKB/qbf8VgP8YQALgPzsej//Nlxg7/XY9JwgrtEO19/f3MlbbF/zsC5Y+FZnsGxvwIdbYFzN8zvHawdO+/fqO1w2j9m3fDmG2//M8D8PhEIvFAmmanoRTbdtG0zQoiuLZkd9LEcePif/2cRwHf//3fw9d17FYLLDZbGDbtgyK9sWNhXYYtS9Weima2uZ5HjzPQ5qm2Gw2j9aToOs6ptMpdrudjPleOrbjOHBdF1VVIY7js/MdxzGapoGqqr2B6PbP5+LR7f36wqt9x+oLVz813119+3TjsaqqwnEc+L4PAHJNtq+5aZqTe97Vju/2XfdTz5P4uX2cdii6/Yy2A7HieM+NJJ/7rnsdfeNr0zQNlmXh9vYWg8EA3333HVarlQzDinkT+4v4tKZpJ3MqIrmqqkLXdRnArarqJFitaRpc18VgMIDneSjLErvdTq7bdtj30rV156nNsiwEQQDP83A8HhFFEeI4ltek6zrCMMSrV69wdXWFoigQBAFGoxG22y3W6zWSJJHHa49H0zR4nofpdIrhcAhd1+UaE9tWVYWiKOR8iCiweC77ninTNDEYDDAajWCaJqqqQlmWqKoKSZIgiiJUVSXHPhqNZEhe0zQ5f8+NIX/Nfq3jJiIiIiIiIiIiIiIiIiIiIiIiIvpcjPkSERERERERERERERERERERERFRBeB/eTwe/xtFUQIA/1pRlP8ngP8MwP/reDz+l4qi/BcA/gsA/ysA/0MA/52//PcfAfjf/uVfos8mAprd/4DHkUtFUWCaJsIwRF3X2G63J6HabrRTVVUYhiEDjh8TIjwXQf2Y7buxzc+J3j61b/faB4MBbNvGbreToUwAME0Tt7e3qKoK8/kcaZo+65rOaV9zO5x8LkjbPq6iKLBtG//yX/5L2LaN//a//W9xOBzQNI2Mh3b3747nUiC573r6ttF1HY7joK5r7HY7qKqKMAyRZRmyLDu5vslkAtu28ac//UkGa8+FqA3DgOd5AIDtdnsSIW1rx2g/Zs2dO44IunaDsSJc237OROxV7H9OXwi4b/zt47SvxTRNBEEAXdex3+9xOBzk/RXjOh6PMnqqKAo0TTv5Xqyv9jtCPPuapsmIrbiusixPrq89/r511D6uOFffPLRjwO11L2K5qqqehHLb8/Gc94E4pqZpGI/HGA6HmM/nWCwWqKrq7LjFteq6DsMw5Ge6rkPXdXieB9/3oSgK8jw/WZOGYWAwGGAwGMB1XZimKaO3hmEgjmOkafroue7Gjk3ThO/7sCwLeZ7LUK+iKAiCALPZDI7jAPgQHzYMA6qqIooiNE2DMAzx7bff4ptvvgEA7HY7bDYbAMCLFy9wPB6RZRmqqnoUi55MJri5uYFt2/LZHQwG8H0fu90Od3d32O/3aJoGtm3j+vr6ZB7bxPVYloXr62uMRiNomiaPK0K9pmnCNE00TQPDMDCdTuV8ifCvmLffoue8O4iIiIiIiIiIiIiIiIiIiIiIiIh+CxjzJSIiIiIiIiIiIiIiIiIiIiIi+ht3PB7vANz95eeDoij/AcBLAP8JgP/+Xzb73wP4f+NDzPc/AfBfHz/U2v6/iqIMFUW5/ctxiD6LCHZeClu2A6Wu68IwDKzXa+R5/ihY2maaJgzDwOFwQJZlMvz5HN2oajtceW67vu+7n7fDm93tPkXffrZtYzKZoCgKGTsW5xXRU+BDxPZjg72X4o19Qdfuz+35NAwDt7e3GA6H+OMf/4jtdisDuWmaPgqX9gVYu+e5dD/6orsi2KnrOrIsQ1mWCIIArutitVrJmC8AWJaFMAyxXq+x3W4fXVP3uI7jwHVdNE2DOI5P4sTnYrKqqsJxHDRN0xsB7UaTz81Ld1ztuekbc3fe+n5vh4C75zwXbxb7eJ6HIAhwPB5xOBxkRLp9vHbIV1VVeZ1i/XZj0cBfA7qO48hYra7riKIIy+Xy5B0hYr8iDnxuLrr6rks8S5Zlwfd9DAYDmKYpxx9FEdbrNZIk6Z2r9rX3EXN2fX2NKIqwWq3k+mkfQ8yJeG5EzNc0Tfm553kYjUYYDAaoqgp5nsO2bSRJgizLYNs2ZrMZgiBAlmVYrVYAAM/z4HkeqqqS0d++d5kYhziPCBoPBgOs12scDgeMx2NMJhNUVYXtdivfx4ZhQNM0GIYBXdfx4sUL3NzcIE1TrNdr/Pjjj9hut9A0Db///e/h+758f7XX43g8xjfffIOqqnB3d4fD4QDP82QoeLfbYb/fI47jk2e+KAr5LhIRccEwDFxfX+PFixcn97MoCgAf3gfiOLquYzAYYDab4e7uDmVZYjAY9AaYn3rfPued/LX4tYyTiIiIiIiIiIiIiIiIiIiIiIiI6HMx5ktERERERERERERERERERERERESSoii/A/DfBfD/A3DdCvTeA7j+y88vAbxp7fb2L58x5ku/GBHPNE0TaZoiyzIZcxSB2rIs5faqqsK2bVRVhTRNUZal3P5cXFOcp+/c577rfn8uXNv+TIRJPyeEeG4suq5jOBwiDENsNhsZBRWxyqqqsFwuoSjKo8hkV9889cVwu9s/Jw6saRrG4zFevHiBh4cHLBaLk5CmuF/nrrsbp22fW8SS+8K0XWJNNU2DPM+h6zqCIICmaSfXq2kagiBA0zS4v7+XQc9zNE2D7/uwLEuu13Pa6yQMQ7x8+RL7/R7v378/uUeGYcB1XViWhaZpkCQJ8jw/mbdLYVrxrAAfArntoLSYq3OxVrGPOIaI7wridxGXbTNNE2EYwnEcRFGENE3lvuKY3XG2g9fnrkkEZAeDAcIwhOu68jzL5VKGV4/HI0zTxHg8lufvu772tZyLVYv9LMvCcDhEEAQy4prnOe7u7nA8HjEej5GmKfI8l/Hgvvh0dxztSO+LFy9gWRbevXsn1093bbfHKz5zHAe2bSPLMliWhdFoBM/zsNlssNvtcDweMRwOUZYlDMPAy5cvMRwOsVwusdlskKapXO+GYaCu65N3a3eNqKoK3/cxnU5lhLqua1xdXWE4HMK2bUynUxkKjqJIHs+2bRluDoIA4/EYqqpivV5jsVhguVyiKApomoaqqmBZ1km4WFEUBEGA3//+96iqCvf399hsNqjrGqZpIkkSJEmCKIpQ17X83HVdKIqCLMug6zomk4m8dnFNo9EI33zzDZIkwXK5RBRFqKpKzrVhGHJb27ZxfX0N0zSx2+1gGAYAXHzuz2Egl4iIiIiIiIiIiIiIiIiIiIiIiOjrw5gvERERERERERERERERERERERERAQAURfEB/F8A/C+Ox+O+E+U8KoryUUU5RVH+cwD/+U87Svqt6wvEduOr7XBlURQy5AtARiANw8BqtZKfG4YBx3GQpinSNJVh0I8JJYr45qXA73M/717bzxFsFJHR6XQqI5zD4RCqqsI0TVRVhc1mgyRJ0DQN6rr+5HF0I759kdL2tt2Qqeu6uL6+RpZlePv2rYzjnov1ds/Z9twYc3t78Zmu6zIEXVUVPM+T0VkRHBWB5CAIkCQJ9vv9yTH6xuk4DkajEVzXxfF4xNXVFeq6RlEUSJIEcRzL4x+PR6iqKmO+k8kEWZbJz0SMdjab4eXLlzBNE1EUYT6fY7VayTBu+zlph3oFz/PgeR4Oh8PZyGg7Cts3j6qqwnVdFEXRu37acV4xftd1MRqNYJomsixDnuePznkuFN1dV+I/XdcxGo0wmUygKAryPEeSJNhutwjDEKZpQtd1OebhcIibmxt8//33j+aqfc1989Z+H4nrv7q6gu/7qKoKdV2jqirsdjtst1soioLXr1+fXJcIvopodDtI2zfHYRji+voab968kSHadhRYxMC7c6frOjzPg2mayPMc19fXCIIADw8PmM/ncl2JsPdkMsHLly9xf3+P5XIpQ7yu68L3fble2+Hm9ljF+V6+fImqqjCfz+X9LYoCNzc3qKoKh8MBi8UCcRyjqip57SJWbZomfN+H53k4Ho84HA7YbDYyxiyi21mWyblTVRWWZeHFixfQdR3v3r3DdruVz1We5zAMA1VVyaiyeEfato3dboemaTAcDjGbzbDf7+W4XNfF7e0tAGCxWOBwOKCuaxiGAU3TTmLUAGTUebfboa5rhGF4Ev+lv+oL6hMRERERERERERERERERERERERF97RjzJSIiIiIiIiIiIiIiIiIiIiIiIiiKYuBDyPf/cDwe/69/+fhBUZTb4/F4pyjKLYD5Xz5/B+B1a/dXf/nsxPF4/FcA/tVfjs9SGz1LXzz0XJy1aRokSSIDiYqiwLZtTKdTlGWJ9XotQ59hGMIwDBwOBxmE/BjtkPCl78/9fsm5SO25MYr5ERHEbsCzvZ/v+7i6upLBStu2UVWV/D1NUxnnFCFKRVFkLNM0TTRNgyzL0DQNNE2DqqqoqkpGMj9GX7BUVVWMx2O4rov379/jcDiczEN3325g9anIb18M+hwRfM3zHMfjEbZtQ1VVuc50Xcd4PMZkMkGSJIiiCADgOA4cx0EQBNjv9zL2qaoqHMfBq1ev8OrVKxkdNQwDx+NRRlQ3mw2Wy6UMn4qgr+M4UFUVmqbJEKlhGLi6usJwOERd1zgcDjLwq2naybWIiGt3zgzDwDfffANd1xHHsfxc3N+6rmUkVawVEUEV4VIRsp1MJnh4eJBz236G++7heDzGaDTCfr9HmqYoy/Jk7XXvZ3eNd7/XNA2TyQSvXr1CXddYLBbY7/dyrgDg6uoKRVFAURQYhoEXL15A0zT5GQBYlgVN0+R89AWk29ehqqoM14ZhiO12i+12K0PZ2+0Wh8MBg8FAzp24r0EQ4Pb2FrvdDsvl8tHz246TW5aF29tbGcdtz1d3jtpjE+89z/Og6zqCIIDjOFgsFnh4eECSJHLfNE3h+z5evnyJNE1xd3cno+divKZpYr1eP4qni3XXNA1s28bt7S3CMMQPP/yAOI7lO1jXdUynU9zd3WE+n8t1axgGgiCAoiiIoghFUcjYsWEYyLIMh8NBbi/ulRiP+Ew8m6PRCMvlEpvN5uQdVRQF8jyX7zexPhzHAQDsdjtYloXJZCLfcWK9TCYT+L6P+Xwuj2tZFgaDAQDIKHGWZTBNE8PhEACw2Wzgui5UVZXvhEvrqvsMPPW34Nfq3N93IiIiIiIiIiIiIiIiIiIiIiIiol8LxnyJiIiIiIiIiIiIiIiIiIiIiIj+xikfqmr/OwD/4Xg8/m9aX/3fAfyPAfyXf/n3/9b6/H+uKMr/EcB/BGB3PB7vfsEh02/UuQBoWzv81zQNmqY5CR8GQQDP804CmUEQIAxDlGWJOI5lpLR7vL4orKZpMuiaZZkMB1+6hr7PRJAVgAyl9u1nmiYsy5IBXRF/FETUVYRHm6ZBnucyUNyeC8uycHV1hfF4jDiOkSQJDocDsiyDpmlwHOfRPIh4ZRAEGA6HME0TRVEgyzI5PjEX3bDnuXl4KoBsmiZmsxmKosBmszkJlYpjm6YJx3GgaRryPEee5733QlXVR/FnEc1tB1PPBX9FZLcsSxkzzvMccRxDURSEYYjb21sURYE4jmXAF/gQGH3x4gXW67W8v0EQYDAY4JtvvoFlWVitVthsNkiSREZOfd/H7e0tyrLEYrE4iZPato3dbofxeIz1eo2qqjCbzXB9fY3D4YD7+3tUVYXhcIimaeR5NU2D53nwPA/z+fzkWlVVxdXVFX7/+9/jz3/+M6qqkp8PBgOoqor9fi9DrpPJBJPJBLvdDrvdDvv9Xo5vMpnAcRz5TIk51nVdhmvba0vca9u2ZTC2u2779K2xdqj79evXMAwDb9++xXq9RlmWUFUVvu9jMBjI50mEaWezGf7hH/5Bxl51XcfV1RWqqpIR23ZAuBtYFbHZ6XSK8XiM7XaLxWKBNE1laFasT9d1sd/v5Zhs28br169xdXUl47Kqqp5cpzivYRgYjUYYj8d4+/btydja89Bd2+L+D4dDGIYhw86r1epk3gXTNPHixQtYloXvvvsOcRzL89i2jeFwiKIoEEWRXC8inj4cDqFpGrIsg+/7GA6HyLIMURTJc4jt6rrG/f099vs96rqGpmkIw1DGsZMkkeMXIebD4YDdboeiKOQ6dV1XhqzFdXueh8lkgizLMJ/PZcRafF/XNaqqku/Yuq7hOA48z0NVVSiKAoPBAI7jYLlcyphvEASYTqfIsgzL5VLGhn3fh+/7SNMUdV2jLEs5J57nyXejbdsyXP2xfzt+654T8j33riYiIiIiIiIiIiIiIiIiIiIiIiL6khjzJSIiIiIiIiIiIiIiIiIiIiIiov8egP8RgH+nKMq/+ctn/2t8iPj+nxVF+Z8C+BHAf/qX7/4fAP5jAP8IIAHwP/lFR0u/We2Qb194tRvU7O6r6zqGwyGOx6MMSXqeh/F4DFVVcTgcUBSFDOGWZYmiKB6FdVVVhWEYMpr7+vVr6LqOd+/eYbVayZilGKNpmlAUBVVVnRxLfCeiqiLwmWWZjOqKYKoIk15fX8O2beR5juVyiTRN5fF0XZfX4/s+jsejjGZuNhsZJlUUBY7jYDqd4vXr11BVFXd3d7i7u8PhcEDTNBiNRmiaBmVZngSDRRx1NpvBdV2UZQnXdWHbNtI0RZ7nAIDxeIyqqpDn+UmMte+eirk6973v+xiPx/jHf/zHk+sVXNfF1dWVjCrneY71ei3DouLYIrwswsMiCFqWpRyn0A3+tteQiERblgXDMLDb7VBVFYIgwOvXr2HbNubzOUzTxGg0kpHooihQFAUmkwmqqpJryHEcGfL9/vvvZdhV3PvRaIQ//OEPGI/HWC6Xciyu68I0Tbx9+xa///3vMZvNYJomXr58iTiOsVgsEEURPM+TMVURILVtG6PRCABO4sgiTvrtt9/CsizM53MZIbUsC9fX18iyDHEcy/DseDzGaDTC8XhEVVXY7/cnUdz5fC7vg1h7juPI6K9YV7quYzweYzqdoq5rGWfVNE2uoW7Mtk1V1ZOQraqq8DwPL1++hOM4JyFf4EOYeTAYwHVdOUbbtjGZTAAAm81GzslgMMB0OsX9/f3ZMbTXiWmaGA6HGI1GSNMU9/f3SJJEbqtpGlRVheM40HVdxsVt28b19TWur69xPB5RFMXJNbXfb5qmwXVdjMdjlGWJ+XyOuq4fham770Rx3tFoBM/zoOs6VFVFFEWYz+ePnjHDMDCZTDAcDnF/f4/1en0SCA7DUN5PEaQVoeHBYIDJZIKmaWCaJsIwRFVVWK/X8v4qioLxeIzhcIiHhwesViv5nnQcB8PhELquI89zee9EcFcEdOM4Pnnf+r6PKIoQx7FcX5PJBLZty3shziH2E8+1eM4dx0EYhjBNE7vdDpZlYTQaoSgK7HY7OTfj8RimaeL+/l7Gl13XxWAwgKZpKIpCrn/LsjAcDmHbNuI4hmmaiKJIxov7/n51343d91Q3Jv1rdum6n9pPzMFvYR6IiIiIiIiIiIiIiIiIiIiIiIjo148xXyIiIiIiIiIiIiIiIiIiIiIior9xx+Px/wNAOfP1/6Bn+yOA/9nPOij6m6Sqam+0ry8I2xf0E3HVxWKBLMtkiNG2bRmENAwD0+kUuq5jv99jt9vJ6GM7oBmGITRNg67rePHiBXRdR1VViOP4JCwpQrSapmGz2eBwOKCuaxlUFfFS0zTRNA2qqpJR0eVyiSzL5PWNx2NcX19jv9/DNE2Ypok8z9E0DTRNQxiGmE6nMtKapikMw0AYhojjGFVVyfFPJhPMZjMMh0Os12u8ffsW+/0eVVVB13W4rgsAMioLQJ5jNpvBtm25/e3tLXzfx3a7lVHSv//7v4fv+/L3tr4Q6zmapmE4HMIwDKzXaxlKFhzHwYsXLzCdTnE4HOT1ibBskiQyiuu6LjzPQ13XsCwLtm1D0zRst1uUZfkoBNk3zqZpUNc1NE2D4zhQFAVJksBxHNze3mI4HGI+n0NRFEynU+R5jtVqJddbkiQIwxBXV1dI0xS73U4GkTebDfb7/UloVMSJxZoQ4xPrVIz//v4er169wtXVFQDghx9+QJIkMh6qKIoMFhuGgSAIMBgM8PDwcBKYtiwLV1dXGAwGmM/nMrgrzheGoQyPijkAIMesqqqMxYrnaLPZnIR8r66uZDxWHMfzPIRhiG+++QbD4RC73Q5hGAIADocDdrudjDB3w7ZinrrvBNu2T65luVzKORDrxPd9FEWB5XIJTdMQBAHG4zE2m40MU1uWhdlsBl3XZQRcXE97LOL9JILX4/EYAPDw8HAS8tV1HYqiyOCxiG6L+O10OoWiKFiv10iS5ORc7fWoaRp834fjOFitVr2hazEX4l9VVeH7PiaTCTzPk++0NE2xWCyQpunJucQ7bDqdIssyGT0WxDsVgIxQi/1c18VwOIRpmsiyTEa/9/u9jIYDgG3buLm5QV3XeP/+PYqikPHy0WgkQ8Eizm0Yhoyfi7UhxiTG63mejJ2LiO54PEZRFPId3L6P4tkWYxKhdt/3kec5iqLAYDCAZVm4v7+Xcy3uWZIkOBwOJ9fted5JlL397jUMAwCQZRm2263cpnvf2vf+XKT2Y96nv3UM+RIREREREREREREREREREREREdHXgjFfIiIiIiIiIiIiIiIiIiIiIiIiIvrqdEO+57ZphyxFTFIEesfjMUajEXa7nfxsNpthNBqhaRrkeQ5N0+QxRDzz+voajuNgt9shSRK8efMGQRAgCAJomibPZ9s2rq+vMRwO4boudF1HWZbIsgymaWI6ncr46sPDA+I4BgAZz9R1HfP5XIYwb25ukCQJ3r9/D9u2Afw1cBwEAW5ubmDbNtbrtYyfuq6L8XgMy7JQVRV835cxXhGKXSwWMtCqKApM04TrujLWK+Knvu/j6uoKvu9jtVphv99jMBigaRosFgvM53Psdjs5B7qun8QV2/ei69xnIsIZRdGjWKlpmri+vsZ0OsVyucR8Ppfx5Hb0WWwbBIGMFPu+j6qqkKbpSdizaRq5b994xBhs24Zt2zJ+PBqNMBqNsNlsUBQFbm5ukKYp7u7uUBSFDHmapinPeXd3hyzLcHNzg6IoUBQF6ro+iXqKAKtYN+KzIAgwmUyQJAmqqsJut8M333yDFy9e4J/+6Z/kNQZBIAOqYn3Zti1DuVEUnazvwWAgj3t3d4eqquT5ptOpjJa2g8h1XSNJEuR5jiRJYBgGhsMhhsMhttutnDPTNHF1dYWrqyu8e/dOPp+6rsNxHMxmM9zc3EBRFDRNI8duGAayLENRFPK5Fv+2fxZzJf6dTCaYTqfY7Xa4v78/CeOapokwDGFZFna7HdI0lWvNNE0ZOdY0DdPpFLPZTMa1+3TfMyISLp5FcU8VRYFhGGiaRq5J8e4R7xDDMJCmKbbbrbzmvvPpui5jwN1ztLUj6J7nyedfBHPF/iJU3D6HCBw7joM3b96cRInFtfq+j8PhgDRNT2K4g8EAvu/LILgIlsdxfDKPg8EA4/EYb9++xXa7leFbETYW0WsRV3ZdF4PBAKqqPjqviH+rqnoSVbcsC5ZlnQR+2++HdhBevLvF2tvtdnJNZ1km51qsI8dxsF6vcTwe4fs+LMuSfwvyPJehcHGMMAwRRRH2+z02m40Mjp9bU92xtr/vC1v/mn3sdbTX+m9lDoiIiIiIiIiIiIiIiIiIiIiIiOi3gTFfIiIiIiIiIiIiIiIiIiIiIiIiIvoqnIusAn8NHp6LHqqqisFgIMOjnufh6uoKWZZhsVigKAo4joPxeIyiKBBFkYxHiv09z8PNzQ2GwyEWiwWWyyXKskSe5/j222/heR7quoZhGFBVFbPZDK7rYrlcygiu4zhQFAXj8Riz2QxVVeHu7k4GInVdh6IoeP36NUajkYytXl9fwzRNvHnzBnmey8CupmkyamtZFubzuRyX+D7LMhmSnM1m0HUd6/VahjLX67UM2orr1DQNSZKgrmuoqgrHcXB9fS3jo5vNBqZpwvM8rFYrGTau6xqu66IoChnsVFUVhmFAURQZ9mzft/Z9av8sAsKu68rIp7i/hmHI+OtiscC7d+9kMFREUJumgaqqAD4EbEWU1LZtZFmG1WolxyxCvt311B7j8XiUIWbHcaDrOqqqwng8xmAwQBRFiKIIo9EImqbJ++B5HjzPkyHpLMugKIoM3zqOg6IoEMexDH8KqqrCsix5bhGDnU6nME0T33///UnsM8sy6LqO6XQqP99ut9hutyjLEoZhyKh1FEUn9ygIAozHYxiGgf1+LwOlIjyt6zru7+/lPo7jyHipuN8iijsajQAAq9VKxlmHwyFev34tg6jiOkXgtWkaRFGEN2/eyPMoiiJDx911Iq65+9w3TSND2iI0LQKu4rkR16ooihyLCBaL8QEfgsivX7+GruvYbDYybqwoioxft8fTDbau12uUZXkydsdxoKoqfN+Xz5njODKwLCK1feuh/bNhGPA8D2maIo7jR/HXNvFci5Dvfr+H7/uwbRuHwwGHw0GGu8V/4trEOEXsVxxbVVWEYQjHcTCfz1FV1cm5xuMxdF1HFEUyGpznOaIokucyDAMvX77E8XjEw8MDiqKQz8R4PIbjOCcBXsMw5HMsIr9ifkXgeDAYyHi1GKthGDgejzgcDifbi/eDuC5N0+T5Pc9DkiQnIfjlcin/LgCAZVkwDAMAEIYhDMOQ3zdNc3KdQRBgNBpBVVU8PDzIwLR4r4kxiffQufsutuu+n/r0/T38Lej73wLtuDcRERERERERERERERERERERERHRl8aYLxERERERERERERERERERERERERF9FdphwnbAtBuv7PvZNE34vo/dbgdd12Xo8/379zLQOp1Ooaoq1us1DoeDjDEqigLLsmS0dbVa4e7uDnmew7IsGbR89+4dFEWRwcbBYIDdboc4jmHbtgyfuq6L2WyGpmlwf3+P3W4no7kAoGkaFEWR0dSXL1/i9vYWb968kVHLoiig6zosy8L19TXCMMRiscBqtZKxX0VRUFWVDLwGQQBd1/H27VsAgOd5iKII2+0WdV3LmLBt2yjLElmW4Xg8wrIsTKdThGGINE2xXC5RVRWCIEBd19jtdjIGKuYqTdOTEOZkMgEAPDw8yPE9da9N08RkMpHj8X0feZ7LgO7NzQ02mw3evXsnj7ndbuG6LlzXRRiG0DQNdV3D8zx5/ev1Gvf394jj+GRd9QUiu2vueDzCtm14ngdd12GaJlRVxW63QxRFmE6n8DwP9/f30HUdNzc3ME0TRVFgs9kgz3MMBgPYti3XhGEY2G632O/3MogqtO/Hfr8H8CEwe3V1hcPhgO12i+PxiNFohKZp8Mc//lGu56IoHh3Xsiz4vo+mabDb7U7u2XA4hOu6Miabpqlcz6PRCKvVCvv9XkZsPc/DYDCApmmI41gGeT3PQxiGWK/Xco5N08TLly/hOA5+/PHHkzivWGOe52G32+Ht27dYr9dyzCK03L5P3XBnNwo9Go3geR7evXuH/X4vw87tZ1mMUcSuB4MBPM+T8V+x/sbjMd6+fYv9fi/PI0Kw3XE4jiNjzuv1Wj6v7ffQcDhEnucYj8d4//49qqpCGIbIsgyWZckod5qmMjTdPq84nq7rMAwDDw8PKMtSrt+maU7mSVEUGTc2TRPL5RJJkmAymeB4PGK328lxtudS/KuqKnRdP4ngivOLa82yTL7DxFoSoeGqquC6Luq6PlknYg29fv0ab968weFwgGVZCIIAtm1D13UZVheRdLG2qqrC4XCQ0W4xLvFO3G63yPMcdV3LazFN8+T3bjRbhNE1TYOu6wjDEHmew3EcVFWFpmlkfBf4EOstyxJJksjri6IIu90OpmnCMAwZBrZtG7e3txgOh1iv13j79i2iKDoZS3feuz+37/9zw7Xd5+Jri90+Z0znAv7ntvvarpGIiIiIiIiIiIiIiIiIiIiIiIj+NjHmS0RERERERERERERERERERERERERfXDtMCUCGb9uRy3PRP0VR4DgOdF1HmqYy2vvu3TvEcQzTNDGbzTCZTGRsUoRsAciw42AwQBzHMkgrYp+z2Qzr9RqbzQau62IwGAAAiqLA4XA4CS+Ox2PYto2maTCfz+X3IjQqYp+apmE+n8O2bbx8+VLGYNvHUlUVs9kM0+kU2+0Wm80GVVXJY6mqirquZWwSAA6HA/I8x2Qyga7reHh4OAmOinhskiQypjoYDDAajVBVFZbLJeI4hq7rMhbcNI2MAYtosogBm6aJ29tb3N7e4uHh4eSenbtXmqbBtm3MZjP8/ve/lxHmLMuw3W4BAC9evEBRFDLkK1RVhSRJZDRZVVU0TYPpdIogCDCfz/H+/XsZmW2ft+/n7u+apsHzPPi+D9u2UVUVVqsV1uu1jBaLoO1wOERZllgul1iv1zJgq2kaDMNAGIaYTCYoy1LGR7vxTRHfTdMUaZpC0zRcX19DVVXc39+jKApomobZbIYsy2S8VuiuF3HO7XaLOI6hKAoMw5BRVsuyoCiKDEzbto1vv/0Wuq5jv9/LaxDB1jAMsdvtsNvtkCQJTNNEGIYyil2WpYy+3tzcYLFYYLfbnTy3Iujq+z6+//577HY7lGX5KErb/k+sExG7bd8nRVHw4sULpGmKzWaDsizlMybWt+u6OB6PMk7seR5s24aiKFitVjgejzK6rKoqNpsNiqK4GEdVFAWDwQCu654EwQUR+LZtG8fjUca0gyAAABmD1jRNPqftyHT72sV9E9fRjtS2/1VVFaZpYjweQ1VVvH//HkmSQNd16LqOPM8RRZGco+6zeTwekWWZXLNiPxHWFe8F8Qw6joPBYCCD0WVZQtM0+U6K41gGpMV713EcxHEM3/flu2O/38NxHBnvNk1TBrRFZF3MTzvmK96tSZLI8auqKqPmrusiiiIZExbryLIsGeYWcyHWVhzHKMsSURTJ62iaBk3TIIoi/PnPf4au64iiSJ43SRL4vo/JZCKD157nIcsyzOdzxHH8aE13/dRR2q8xcvuxQWKh/eyJ+9gOWRMRERERERERERERERERERERERF9aYz5EhEREREREREREREREREREREREdEX1431dn8+F/IFPsRTRXxVBEDn8zmiKIKiKPB9H1dXVzJYmue5DERqmoYgCGRk8/7+HmmawrIsTCYTXF9fI4oizOdzmKaJwWAAy7JQ17UMtBqGIeObnuehKAqsVqtHsU8RbxWBzN1uh+vra+i6jjdv3iCOYxntFOFQEXFdLpcyyitiuLquoyxLpGmKuq5lyHQwGMiI7Gq1QlmWcg5d14VpmoiiSI5dhF6Xy6UMrQIfwrm6rsOyLKiqiuPxCMuyMBgM5LUNh0Pc3t5CVVVEUSQDwX33UVVVGcS9urrCZDLBcDjEd999hz/96U8ywPvP//k/h+M4+Md//EekafrofldVhTiO4boubNuGaZoYjUbIsgz39/cypNk+f3c9dYOQ4nvTNGHbNhzHgaZpWK1WWCwWqOsaV1dXME0Tq9UKWZZhv99jv98jiiIZh9Y0DXmeQ1EUTCYThGGIH3/8UUZO2+fSdR1BECAIAiyXS9R1Dcdx8OLFC8znc2w2Gzmm4XCIH3/8Ua7dbujyeDzCMAwEQQDTNLHdbmUIOAxDBEGA3W6H8XiMN2/eYL/fQ9M03Nzc4MWLF3KMIiY7nU5xdXWFpmmwWq2w3+/RNA0cx8FkMkGaptjv9zKkPZ1Ooeu6nBtxjeI5mk6nyLIMm81GrmNBBFvb0VYAch2KbTVNk/N2dXWF7777DnEco2kaqKoKy7LgeZ6M0op1IsLPiqLIALeId2+3W7mdOB8AOZb2PLdD3GJ+22vbtm0Mh0Psdju8fPlSxo9F+Ph4PGI2myGKopM1KsKz3ZivCFUbhiGfPwCo61qOJwgCuK6Lsizx5z//Wa4x0zShaZpcK+3r0DQNuq7L6zscDthut5hMJvId2g7n5nku35NifYnYrqZp8lnJ8xxZlp0Ehn3fx5s3b04C0NvtFsfjUe7v+z5UVUWWZSiKQsZ8i6KQv4tr8n0fVVWhqir5rqzrWq6tMAxxOBzkfRFzJN7L+/0eZVnCMAz5nhbBd/G8tec6z3MZKW/f68PhIK+9fa6yLE/ubVf770GfX2uotu+d+jn6Qr7noshEREREREREREREREREREREREREXwJjvkRERERERERERERERERERERERET0xbWjq91gnwhcnvvMsiyMx2PYto3tdov7+3tEUSS3tW0brutiv9/L8K0gQrCe52Gz2SCKIliWhdlshul0KkO+IubqOI6MAidJAuCvIc8gCFCWJfb7PXa7HYqiOAm4qqoKx3Hg+z7m8zkAnERO2+NSVRXD4RCqqsoAKvAhZBqGISaTCeq6xmKxkBHM/X4PVVURhiHG4zHm8zl2u52MhaqqislkAtM0kWUZ6rqWocssy+SYj8cj6rpGWZYy/Nmea9/3sVqtYFkWwjCE67pYLBaIoujkGtoMw8B4PMaLFy/k/BVFgbqu8ebNm5M48PX1Nb7//ntsNptHMeTuMW3bhud5sG1bRmrbUdSuvhhkezvXdWFZFgzDQFmWMths2zZ830eWZfjxxx+RJMnJ/W2HWcU9HgwGAIA0TU/Cr4qiQNd1OI6DIAiQ57kMiY5GIwyHQ/zbf/tvkee5vA7TNOU5z12POG9RFNjtdvKzyWSCKIpkPPj+/h5VVclYtG3b8tgiCvzq1SsYhoEffvgBDw8Pci0EQYAwDPHHP/4RRVHIwK2qqlAUBfv9HnVdy0ir53lyjh4eHhBF0aMYcZu43+2gqthORHtd1wUARFGE4/Eog9NivgHIc4q1K8LVWZZhMBhgMBhgv9+jqipst9uTkK+Y125QVKyzPM9PItMiou37PqIogqZp8vkzTRP39/fIskxGszebDcqyRF3Xj9a3qqry+ouiwHq9xvX1NXa7nXynift4c3MDVVXx/v177HY7ec3iGEVRyLkxTVPOlQj9ZlmGLMuw3W5l9Nx1XRRFgTiOkWUZPM+DruuYTqdyHne7nYyJu64rI8viukTU2HEcvH//Xq6pNE1xOBxQFAUMw0CSJLAsC7ZtY7fbIUkSuK4rx980jVyTIvicJIkcd3v9p2mKzWaDq6srvHz5EkVRQFVV6LqOpmlksFiEtpMkkcdUFAVZlqGqqkfRWLFNdx2maSpjxABkTN62bXkPqqp6FK3+OX3J6O1zztd+1p96D7d/74a1iYiIiIiIiIiIiIiIiIiIiIiIiL4GjPkSERERERERERERERERERERERER0RfXjfiKiOlTQVYR9xyNRthsNvjzn/8so5cijun7vgx5lmUp9xffhWGI4/GIKIrgui5msxnCMMR2u8XDwwOyLJPnEPvv93sZADUMA47jwLZt7Pd7bLdbZFl2cg1inLe3t2iaBqvVSkaG7+/vZXhThBg9z8NsNsN+v0cURaiqCpqmwXVdjMdjeJ6H3W4nx3A8HpHnOaIowu3tLQzDwGq1QlmWcgy6rmMymch5AIAgCKCqKqIokoFLRVFQ1zWSJEGapjBNE6qqQlVVjMdjGaoMgkAGK7fbrQxvtu+p4Ps+JpMJiqLADz/8gOPxiD/84Q8n4VlN03Bzc4Pj8Yj5fP4osNomwpm6rkPXdWRZJqOzl2K+3TUk1p24R57n4Xg8IssymKYJy7LkPiIMutlsZDi1e62e58kwtIi8pmkq56k9JtM0oSgK5vO5PN9wOERZltjtdidB0SzLZOS0jxi7ruvYbreo6xqWZcl1cjgc8M/+2T/DZrPBdruVEes4jhFFkVwnjuPg9vYW0+kUq9UKy+VSBlB938doNEJd1zgcDvKZEEFTEV81DENGiff7vbw/q9XqJM7afubF/IuI7/F4lIHg7j1zHAdJksjxtkPLURTB8zxomgbDMOD7Prbb7cm6ubq6ktfmeR4WiwU0TYOqqjI03H0faZoG27blc1EUhYwsh2EIz/MQRZF8/sQ9fHh4QBzHUBQFlmUhTVPEcYyqqk6ixt24cdM0iOMY6/UaL168wN///d8jSRIZDgeA+/t7LBYL5Hl+cixxHw6HA4bDIcbjMXRdl+ukqioZzwU+RH/n8zkWi4Ucy/F4hOu6iONYvj+3262MBldVJWPUYRgiSRI0TSPj3qqqYr1ey3UsIrhijCKifDgc5PHE8yOuoSzLk1h2lmV4//69/L4d2a3rGg8PDzgcDvA8D4ZhoKoqJEmCPM/lfPc9s+Jn8Yy277vYpx3Kbd+jqqpkeDnLMliWBdd14Xke4jh+9E78OX2JiO+nuDRGMbe/husgIiIiIiIiIiIiIiIiIiIiIiKiv22M+RIRERERERERERERERERERERERHRF9cO97bDieL3c9sCHyKrVVXh7du3iONYhkBVVZXRUk3TkKYp6kcjRsMAAQAASURBVLoG8CFqKeK4ruuiKAr4vo8gCKBpGpbLJRaLBYqiAPAhemvbNqqqkuFGMRbbtjGZTKAoCrbbrYyjivCjiLre3t7i5uYG79+/R57nePXqFZqmwcPDw0n0sWkaeJ4H0zSx3W5luNR1XTiOg6qqkOc5AMjrEbHMsizhui72+z02m81JFNEwDARBgOVyiSzLUNc1DMPA8XhEHMcnIVzgQ+Rzv99jMpmgrmtomiaDxr7vI0kSZFkGRVHkNbd1Q63iPIfDAUEQwLIsRFEk74vruri5ucH9/b0MMvexbRvj8RjH4xFFUSAIAhntbK+Pbgi2Gxruri/HcaCqKvb7PZIkwdXVFWazmZxbsdZs20YURScRYN/34fs+iqLAer3G8XiEpmkIw/AkIC3GoWkafN+HrusyeCriuPP5/CQWrKoqttstJpMJ3r9/jzRN5ToR34ugbVmW2G63MuTbNA12u52Mrt7f30NVVQyHQ9R1LWPRjuPg6urqJNgr5trzPDiOg9lshtFohCzLYBgGPM9DURSIogi+76Oua7x8+RK6rqOua2w2G9R1jaurKxwOh0eB6/YaEc+s67owDAOHw0E+e115nmO73eLbb7+VAevlcondboemaeT+iqJgtVrJ9Sae291uh8VigaqqkKapjF7XdY2maR6tn/a/IiJr27aM2Yp1vN/v5Xz/m3/zb7DdbuX7SATA//SnP8mAbTde3qYoCsqylIHaIAjg+z6OxyPu7++x3++R5/mjGLCQZRnm8zmOxyPCMEQYhvKdkCTJyXtD7N8NGadpivv7exlCF9FgAEjTFFEUYTabwTRNFEWB4XAoI7aLxQJRFKGu65M4c3vdivdam5ibuq7lO0mo6/ps7PV4PMrIdBRFj7YT19QXG39OQLbvOGKNNE0j15F4th3Hkev9c8K03Tn7WN0o9aXthC8V0u1G2MVcis8Z+CUiIiIiIiIiIiIiIiIiIiIiIqKvCWO+RERERERERERERERERERERERERPTF9cUzRQC2T/tzEe6NokjGOAURI82yDIfD4SQOKUK8juPIbdM0xXq9loFV4EPAdTgcyt+7kUkR9by/v8dqtUIcxyfXoWkahsMhhsOhHAcAhGGI1WqFJElOYoWqqmI0GiFNU8RxLMOvpmnKaK1pmlAUBXVdy8ilpmkyArxYLFCW5UmgczAYoKoq7HY7FEUBVVVlnLhvfsX4D4cDjscjfN+H4zgyeno4HDCZTGRI89y9UhQFVVUhSRKUZQnDMGDbNuq6luFbVVXheZ4McrbnrxskDsNQHm80GkHX9UfhYuCvAcynop3H41FGeouiwOFwkNczGo0wnU5xPB5hGAZUVcXV1RWapkFd17AsC4PBAI7jYLvdYr/foygKGUBtmuYk1CzWgwgzZ1mGOI6hqiocx8HxeMTd3d3JfSvLUs61iANXVSWD1IZhYDQaIQxD1HUN0zQxHA7RNA02mw2apsFgMICiKNjtdhiPx7BtG8vlUoZvR6ORDLHe3d0hTVMMh0P8/ve/l+td0zRUVQVd13F9fY3tdiuDvbvdDr7vYzKZYL1eYz6fI4oiXF1dwTRNzOdzOQ99AVpFUeA4DqbTKZqmkXPSjsCK7ff7Pf7hH/4BlmWhrmvkeY6iKFBVFRRFwWazwX6/l4FaAIiiCD/++KM8tggsF0WB7XZ78hx1A7NifYvzeJ6H169fy8/W6zWiKJIBZrGmxe/iePv9HgBQVZW8v+3r785JXdcy3C0CtWLfdkhW7N+NZ8dxLIO53Xlvb9v9XoxFvCv6Yurt2PR+v8fDwwOiKEKWZSeh4fZ/Yqx95xfzEcexjOCK4HB7XtrvpXP6wq9976ZuOPbStufOLc4l5ipNU1iWJd+rnxOgFSF4ACdB8J9a3/393HFf2v+p79uh5L4IMxEREREREREREREREREREREREdHXgDFfIiIiIiIiIiIiIiIiIiIiIiIiIvriuiHJdmTxXGAT+GvIN89zGX8Un6mqisFgAOBDYLMoipP4pWEY0HUdTdMgiiI8PDwgjmMZohTHcBwHg8FAnrMdm3VdF9fX1zJMK0KUgmVZcF0XmqZhvV6jaRpUVSXDtXd3dzKOKwKWIsjbNA08z8NwOAQAbDYbRFEE13XldbTnyPd9vHr1CpqmncRCNU2DbdsIggDb7RZpmqJpGhiGIeOgIpgqQqGO4yAMQ7iui81mAwAYDAawbRtv377Fw8MDmqaRYd32eNrjEmOr6xplWaKuaxiGAdd1cTwekSQJmqaRgVsRMG3fY8E0TXieh+PxiNVqBdu2Yds2yrKU8dZPjT/qug5VVZFlGfI8R1mWMlBq2zY0TYNpmhiPx7i9vZVhZBEgXSwWJ7HopmmQ5zmyLHsUS26aBmmaYj6fy3OJ0O6/+3f/DlEUncRL8zzHcrnEaDTC73//e9i2jf1+L++TbdsYj8fwPA9lWWI4HGK73WK1WiHLMhiGIdeb7/soigKLxQJRFEHXdbx//x5BECCOY2w2GxRFgTiOURQFTNNEmqYy2pskCVzXxX6/x2KxkOt9Pp8jSRIAkPsej0cMh0MZSBbBXfHstNeJrusIggBhGOL+/v4kRA1Ark9xf+M4PolmtyPeIswr9hNB5O122xuebodjxTG667mqKmy3WwBAEAQAgDRNcTgckKbpo227sd6+NXkpVi7GIL4X66q9fXeM7edNvL+6z4M457nQdfe+iPO046piO/Fee3h4wA8//IAsyx4FeNv/tT/vRnS7oeH2OM69+5/rXMS7O06x7XNDs93P67pGmqZyPfS9Ez+GiE1/Slz3Y96DP3Uo92NiyH1EjLodq1ZV9dEzRURERERERERERERERERERERERPQlMeZLRERERERERERERERERERERERERF+FdiSxHSJsB/y6MUAR/UuS5NH2juNgNBrJCGk3gKtpGgAgyzIcDgfsdrtH24jIqGVZKMsSVVUhz3M0TQNd1xGGIXzfR5IkiOP4JKJpWRY8z4Ou69jv9xgMBjJSOh6PAXwIn4rArzh3XdeI4xhhGOLly5dYLpcyNNw0jYyVipirqqrwPA+j0QiKomC73cqxiDnyPA+GYWC73aKqKhl7FHFV0zRh2zaqqoLruhgOhzAMA7vdDnEcyzn+x3/8RzkWz/NgWRbiOEZZlmfjpe2gp6ZpcBwHrusiz/OTEKqu6yiK4lFsVUSZwzCUUdaiKBAEATRNk3Mv9ru0XrraEc+yLOU9FnOc57mMS5qmiSiK4Ps+VFVFVVWIogiHw0EGoIWmaU7uV/ecIvrZnqMoiuT9b0dTRSwXAG5ubvDy5Uu8fv0aRVFgt9ths9nIILKYj+Vy+SgsraoqfN/HDz/8gP1+L6/x7u4Od3d3J3O3Wq1kMLiqKlRVBVVVkSQJNE1DnucncWxxPe1xizW1WCxO1mN73sW2lmUhDEM0TSOf5fbciHUqntmmaWSEWkQ/RfC3fU/bMdD2Omw/5+0YsPi8fQxxD/I8x2KxwHq9Poml9gVGu++Rdjy3O86uvs+7IdS+8Xe37RtbNzDbF6Xte/d2tzMMA2EYQlEUbDYbGQg/F+ftXtenPKPdn5/rqTjvc7d97vGeigc/1091nK/JU9fSft6eimETERERERERERERERERERERERERfUmM+RIRERERERERERERERERERERERHRF6WqqoxyAn8NmAp9sUfxmQiYdsOlx+MRjuPA932UZYksyx5FVcUxRAhVVdWTIKwIz06nUxnzTZJEHsfzPAwGA9R1jTRNZYhWVVUEQYAwDAEAh8MBwIcAZlVV0DQNYRgiiqKT44nxlGWJd+/eYbVayXO2Y795nsvY72QyQVVV0HVdRomDIJDbi0ipbdtomgaHwwF1XaNpGhRFgSRJYNs2wjCEbdtyDvM8l0HXsiyhKApWq5UckwgQ53n+aHxi/ttE5FLXdRiGIcO4RVHIe5BlGRzHQRiG2O12qOsatm1jPB7D8zwZrhX3Os9zrNdrZFmGOI5P7mtfVLS7ltrbVVWFNE3l3LTHLWKsdV2jKAqs12tomobj8YiiKHoDwmJ+++ZCXG97e7EGRBBV1/WT8x+PRzw8PGCxWMCyLGiahqIoZETY8zy5lvb7PbIsk8dSVRXL5RJ1XcsYrbhffSFY4ENQuKqqR4FiEdUVc9KN5IrzqaoK27ZlyLooCrlv9/6I8K/jOHh4eJCRY7FdO4TbXtPd2Gk3VNsd+7lI7lM/tz8T60Nco7jmvgBpdzztiG83ztsXL700VqE9l+35aIeOu8foC86K4/RtL/5tv3ODIIBpmjgcDtjv973x3r4x9ul7V/wWwq1PXfdTfuo5+Nzx/FLEWuv+74Hu328iIiIiIiIiIiIiIiIiIiIiIiKiL4kxXyIiIiIiIiIiIiIiIiIiIiIiIiL64roBv77vxM+CiGk2TYMkSR6FCk3ThGmaWK1WiOMYZVmeHKuqKhnXtW0bjuPIWKdlWfB9H0EQwPM8uY+qqjBNEwBwc3MD0zSRJAnKsoRpmhgMBrAsC47jIMsy7Pd7pGkKz/OgqiryPIemaVAUBbvdToZx2wHLsixRliX2+/3JfIg4aJZl2Gw2aJoGjuPgeDzKyK7rujgejzJeLI5bVRVWqxUOh8PJPBwOBxmDNU1TBn6jKJKhWDHvWZbJOQA+BH/fvXuHNE1PtgUehyPFNRqGAVVVYVkWlsvlSRx2tVphOp3im2++wWQykeM+HA64u7tDFEVy7MfjEdvtFofDQYZ220FZMQZxbBE3FXMofhZrTqyFS7oh2b44pjhX31rui7eKtfCc9S+usz0HQhzHSNNUjlOEL8Uc3t/f4+HhQc5TO0rbnStxrnact32tYg7FtbbH0r7+sizx5s2bk4B035xYlgXXdWWcWUSEu8989zztfz/HpYhuN9Z7bvu+6G3fedrx5EtjPxfX7X5/KfDbjiCf2777zPYF1bvrXNd1eJ6Huq6x3W7ls98+p3g3/y3oBsKBLx8kfm4M+mvSfT+Kz9rrioiIiIiIiIiIiIiIiIiIiIiIiOhrwZgvEREREREREREREREREREREREREX1VnhsdbIdqRWi2/blpmtB1HXmen4RaxfdFUeBwOMho76tXrxDHsYyciqioruuo6xqapuF3v/sdfN9H0zTwfR/39/domgZhGGI2m+H6+hpxHGO5XMrQJYCTmKwIFIoA8aVYYV/Qs2kaGdBtB41FIFhEhMUx67rGarVCXdfymsSx8zyX4WBVVVHXtQyvdiOmAGSMVFEU1HWNzWYjt+9qB0SrqkJZlgA+hD51XZfBYSHLMvzpT3/C/f09dF3H8XhEmqbyWrsRVDHW7jx1w6rd8Z+b5+7Y22vl3PbtwG17bsRn3blob9uO5PaNuf3fUwHYdnRXRIfFXAOQAeDufarrWo6lPbbnBG7PfS+kaYqiKFBV1aPxiznSNA2u60LTNKxWK6RpKq/llwiifurxz92T7r3rCzg/pbtPe520A73dfS6tjXPrvu+aLh1LURQ4jgPXdZFlGXa7nXyndAOs556p9vm+Zt2gcd/37X/7otP0fO25bq/5c+F0IiIiIiIiIiIiIiIiIiIiIiIioi+JMV8iIiIiIiIiIiIiIiIiIiIiIiIi+qJESBQ4DaB245B9MUgRLhVByXb0T1VVaJomo7HdGGBVVdhut1AUBcPhELquw3Vd5HmOLMsQxzFc14Vt2yiKAuv1Go7jwLIs1HWN+/t7LBYL1HWN/X4P27YBAEmSIIoilGUp45hFUSCKIui6jtVqhcVigcPhcBLXbetGQcVn7Wtvx2yFxWIBRVFkOFfI8/xRgFbMt4j89unu0w1XivO0Q6B96rqWYVfbtuXv3WtPkuQkUiz2FfevG9DsjrEbsu2Ou298fSHgS84FPvtinu0IaztM2Q5Wirk/F4HtG1f7WO2x9IVl22ukL47Zd+xL520HZfuCrWIsVVWdhKH7zqFpmgxu73a7R8Hjnyvg+TH3W2x36X63j9n33cfou7dPzcWlGHXf2hfH7Itwn3vnAB/u12AwgGma2G63MkjeN4buWNrb/BrCrJci3uJvVjf6/Wu4rp/Lxz5TTx3nXFCeiIiIiIiIiIiIiIiIiIiIiIiI6GvAmC8RERERERERERERERERERERERERfVEi3ClCrJfild24ZFVV2O/3MhzaPmZVVUjTFPv9HnVdPzrm8XhEmqbyGLZtQ1VVFEWBoihQ1zWGwyEsy0IURbi/v4eiKDAMA2VZYr/fI89zNE2DNE1l4LEb2T0ej8jzHA8PD1BVFXEcy+3aY2pfW19k8xJxnG4cuO/nbmi2Lx7a3a8bauzGTfsind3QpZgT27ZRlmVvYFnMiwg5iv/aAdm2vusTEd/uWPrW1LlYZHuO+iKVzwl3dq9f/Cwivt3Ib/cc3X3b1/DUPFwKa/YFjcX8nou4nhvLuTH3Rbn7qKqKuq5lQPvn1rdmn+tcMLdvm+6a+iniwZfG0nfO7hifs37b900Ep9u/W5YF3/flOzPP82eNsf2u+TVqj7s9J+cC3t19LrkUVv41uRTdbb+7nnondQPUv+Z1Q0RERERERERERERERERERERERL9djPkSERERERERERERERERERERERER0Rcl4oiXgojt37uBvzRNe/fdbDaoqgqHw+FROLYdYSyKAmVZIk1TGVVshyvrukaapjgcDqiqCsCH6GVZls+ODFZVhTiOH0Ub2/ufC9Y+x1Pj6Pu+bz76Qr6feq5uJLgd2xUR5OeMt30M8bvYpi9eeu68lyKST0Vs2z93r68vCNwX9BRjEjFfEa/uzkM7sHwpMtz97LnhZrG+2593P+vOdzvG+VRwuzuWc98pioKyLBHHMaqq6p2Hn9rHhlY/5phCX8T23O+XiO3EfH9KfLgvovop7xfgw3MbBAEcx8F2u0UURfK92j3+c2POvzbd5+5SPLkbahch776/YT/1evwSc/655+x7VrrRcyIiIiIiIiIiIiIiIiIiIiIiIqKvBWO+RERERERERERERERERERERERERPRF1XUNTdPk7+fifSKCKIif67p+9P3xeMR+v0ccx8jz/CSQ2ud4PMpQLwBomgbbtuH7PqqqQpIkKMvyJOB7LrTa1Y3PPhXV/KmiiN3z9gVg29u3t2uHXS9dz7nPz4VtNU2Dpmkn9/dSBPc5cyX+E5Hcp7a9tE33ep6KbT61rtrbtK9DXH93rs5FlS/Fd8/N0bkIsfi971kSYe3u590xtP/rhrLb+zwVbK6qClVV/eShznMx5Z8rCHru2fopYq190dhz24mxnBtP3/ZiH/HvuZi3pmkYDAYAgCiKkOf5k+dof38udPu163uuPnb8fXHsn+v6u+/+X9LHxr371umvaW0QERERERERERERERERERERERHR3y7GfImIiIiIiIiIiIiIiIiIiIiIiIjoq9CO+LUjoN2IaPszoS+aKeK7nzoW27bhOA6KosB+v0dd1xfjpH3j6V7DU1HYc9ufCxxeCh9+bBCxe/5LQdRLgdH2mPpCxqqqnp3LblQYgAzXto91LnbbPs6luRHnFvHavhhvX1D40v3oixn3BXXb8WkR9e2LfYoxXgqs9l1f9zNVVXvnQ1XV3ufs3LG6833uuB+z7i5t252Hvu8vrf1za/hjx3hJ+z4/NxL9czoX1L60Xdu5+LGqqrAsC67rIo5jRFEk1/Glc3zse/Br9dw49LlIdvfnbsz8XEz5U8faDrp/ybhv27n3XF/svL1P+18iIiIiIiIiIiIiIiIiIiIiIiKirwVjvkRERERERERERERERERERERERET0RfVFMPt+7273VACw7zzP+QwANE2D7/vQdR37/R5xHD875NvnOdHVbuzwqeM8N/z41HVfClCei30+FRDuO0ZVVSiKAmVZIs9zGQPtBh1FvPfS+Puu59K4+6KW3eDluXN1xyLGd+4cl+aoG0J+KjZ87vNu7PfSsfpcimI/JwjbvdZzx//cCOfn7H/uefqpQ74/x7E/1VPn74bRn/tOUxQFhmEgyzLsdjukaXoS3e4LsX/sXPwUIdufSzcwfinqfe698NQ23Z8/d+0LIlj+S85tX6i3bwznvv/SUWwiIiIiIiIiIiIiIiIiIiIiIiKipzDmS0RERERERERERERERERERERERERfVDs2+JzwbFs7itoNCH5qEFBRFGiaBtd1AQC73Q5JkvSGL5+KNIrPuhHeblT2nI+9juds341Lngsnt4/Z3u6p+W3fi7aqqnA4HLBcLnE4HFBV1dnx9UV32/HaS6HW9n/ttdXepxuHvjQflwKT7fNcClc+N6T5nHXRFwK+9NycCwd3w8Ldfc7FersR2HPbnPvuuX6K+OgvHYf9WmO0Xc+J/nbXYFmWWC6XiKIIRVH0ru9uHFsc66nzfM261/mxf1cuhXvbvz8nov3c8/VFm3+JeT4XCT+33XPeYZ8T0SciIiIiIiIiIiIiIiIiIiIiIiL6uTDmS0RERERERERERERERERERERERERfTDvg1zTNSXy1+71wLqDb3rYd9j13vktUVYVlWWiaBofDAWVZ9m53KZB4LqbaHVf38/a/l6Ko3XDmuehx33HOzVnfsZumgaIo8t50r7F7rnZ0t3t/oyhCnueoqkrGfM8FHdvnbn/WF9QUx2hHK7vXLNZXN+TbPca5MHT7e3Gevohvd/76YqB991X8rKrq2eN2t23//Jy10Hetz9F3De3rb+tbJz+Hj42Tfk7g+7nH/5pdej+c27797CRJgjRNUVUV6ro+2ea5AdrnRLP/FnRjyX0R5M/RfV5/ibk9d46nQufn4sZEREREREREREREREREREREREREXzPGfImIiIiIiIiIiIiIiIiIiIiIiIjoi+kLGJ4Ltra/E0Q49FyU9lPDgKqqwrZtGaCtqqo3EtsXC74Ux+y7nr7fz4Udz0V5u2HIc4Hg7j7t/box2O5x2mMQ894O93avozsmACcR3z7tczZNI2O+3f+6c9I3VrFtOwh8Llh8aT67n3evuRu07a6LS2uwb+xPhZcvRT/74rpt5+boY4Of7XvU91076vy36KcMs37uOPp+PxdWvRQcF0Hzc+/rvuP1/f7U51+7n2LcfXHstp9q/XypOf6YcHb3XdkNRP9a1wkRERERERERERERERERERERERH9tjHmS0RERERERERERERERERERERERERfVDu22jTNSWT1XNhU6AZZ+6KtH0tRFGiaBk3TUFUVkiQ5CZN2Y7jn9EUuRQS3L/h6Loop9ml/1j2P+PfcdT8VROwL1z71c180+VLA9Vzg9tzYuvHebsi3G3zs2649Hk3Tzu7fPmd7Pttx3KfitM+Z477IcV/IF/hw38Xz0D1O93zdeTr3vPSNqb1d+1n8VF9zgPOXGtfXcP2XQr5PRcOfE+D9WoLFX5un5uXcfH9MAPfcec8d/zn7fs69PBfl/5TjXgpDExEREREREREREREREREREREREX0tGPMlIiIiIiIiIiIiIiIiIiIiIiIioq/CpWhpO3Iooqp9Qdhz+5zbro+qqjAMA1VVIYoiVFXVe5x2aLUb3D0X5n1OCLgv+NrWF6Ttfn5pbkT49Tkx4u6xuuNrx24/JUjZDuS2g7TiON04c/f44j60ta9NBHnb19y+tm64t71/+/v2ve6b276wbt/c9l1T+1jdcHDf9fath75A61Prpnvuc9/Rr1NfPPvce+ljjvWx+9J5T8XqP8an7vu58e5L+3cj688ZY/ddz3VGREREREREREREREREREREREREXyPGfImIiIiIiIiIiIiIiIiIiIiIiIjoi2lHW7uRyXYwtrsP8Dia+jFRwnOBwHZ4MEkSHA4H1HV9Mtan9m+HZMUYu0HDblSzvV3fPHSvr+943etrfyd+7o5ffC5itd2x9h1f/N40jdz2XGBZzF3fd2K/vvt8Lpx7LhjcF4zshoL7wr/dsYgxXIovn4tEd2PE52K+QjsQ3P7s3LEvxZ8v/X4OI5mnLgWYf+1z9VS4+7l+C3PxqZ577U9tc+4ddWn75xz3U33ucZ8b6H3OvLTf46qqPmt+iIiIiIiIiIiIiIiIiIiIiIiIiL4ExnyJiIiIiIiIiIiIiIiIiIiIiIiI6ItpB1JVVX0UTe0Ly3YDuX3H6osHnvusvS/wIbCYpinevXuHOI5ldPVcSFZEazVNOwnjiu/6or3tc7ePcy5u3B1j9/O+6O1T+577rDsn3TBwO9z7MWPtBm/7rv3c9bXntR3cfU7gVtyf7vy2t2tHibuf9UWIL4WXuyFgQaxv4WNClV9ryPNvwc8dU/01ee4ciGf0txRj/ZR10H1HfExw/rfiOQH89r/tz/+W49FERERERERERERERERERERERET0dWPMl4iIiIiIiIiIiIiIiIiIiIiIiIi+qHYgVugGV7tx3KZpHkV9Rai1HXt9zrm7v9d1jSiKkCTJSQT2XLhVBAdFCLYbu22PrXuubjBXVdWT7do/i+OKfbrXeSkgLOarG+btnqM9/vY27RDtpWO0txHXcmmbc/dMnK+u65PxdOPOl2Kh4ljt+9H+vO9Y3W3ba6yu60fHPxfq7Ft73WP/0n6LUcxfIvZ5KRpNj+m6Ds/zoOs69vs9yrL80kP6STy1BtrvCk3T5Duj/V75lHX0t7r2zoXfiYiIiIiIiIiIiIiIiIiIiIiIiL4GjPkSERERERERERERERERERERERER0VejL1zYjrZ2Q7Pt4J/4V4Rr2/u3f74UR+yGe9vh4G4ctjvebvy2u23783P7Nk1zsk037tvWDUT2hW3b4xaBXfF59zq6cV4RJxa/C5qmAcCjuG038NuO73aP3d62G2cGIAOg56K43WsS8d9z4d+29tppb9sdX9+4u+Hk7nGf4281zvlz+Cnn8tyxuuuDztM0DUEQ4Pr6Gk3TIE3T33zMtx3xNQwDQRBgMBhAVVXsdjtsNhsURXHx2F96fXVD5z+1bsy+e73n/sbxmSMiIiIiIiIiIiIiIiIiIiIiIqKvFWO+RERERERERERERERERERERERERPRFdAOC3XDfpXBqe5tzIcBz24tz98VvRbxVRG+75+4Ga7vh4O55uuq6fhT97W7bHlf72N34b9852yHap/br+7kdphTX+tT1deO37XGfm+O+a+jb9tw19u1/7rNLxxLa19y3zaXPu2FKBijpb5GiKDBNE4PBALquY7/fP+uZfuqY3Xfu16IdC1cUBbZtYzQaYTgcwjRNNE2Doiiw3++fjOVeCgV331OfOw99cyreq6qqfvY9656r72+oIGL1fff3a7znRERERERERERERERERERERERERAJjvkRERERERERERERERERERERERET0xfWF/M6FHPvCqiIKKL7vi/VeOgbwISbbjsV2Q73dcXRDiN1znQvBimjic2KFdV2fjRW3Q4mXQrTtf0WkuDuuc4HhbtT4UnixPYan4pXdYPFTwcpuALL9fdM0vQHKvjGcO243Etw3r+cixH370W/Hx8TC/1YpigJd1+F5HsIwxHa7xX6/R1mWn3xMTdMQhiHCMMRms8HhcPiqnq/23x3TNDGbzTAejxFFEVarFVzXRV3XqOtabv9TnPOcp965YpvnxnO7f1+eGsNToeH2ecXfwO4xNU3je5SIiIiIiIiIiIiIiIiIiIiIiIh+FRjzJSIiIiIiIiIiIiIiIiIiIiIiIqIv4mPChO1Q4HNCf+0g4Lkg4bmY7fF4lHHYbiT2XNSwG7Ptxoi7xHEvXUs7UNx3zj7tWOK5c/fNY3eez0Ufzx2vu213Pp67/6Xrao9P/HzpuH37XnJufV2KSx6PRxnrpM/z1Br4Ur7GMX1tFEWBYRgIwxCmaWK/3yOKok9+NhRFge/7ePnyJVzXxWaz+erugwjSapqGyWSC6XSKLMtwf3+Ppmmg6zryPP+s98PH/I187t/F524rtKO73X0/NnR97u9E+9jPfV8TERERERERERERERERERERERERfUmM+RIRERERERERERERERERERERERHRV6sd9rsUp22a5mSbdki2aZreIOGlQGuf58QFLx23e6ynIrbimrpj6ovktqPD3aBw95x9oePPuZb29uf2f87cfUzMtTsfuq5D13UYhoHj8Yg8z1EUxcXxtT/rjo8hyV+OaZqwbRvH4xFxHD8KaNPXT1EU2LaN8XiMNE1xOBw+K2JrGAam0ylmsxk2mw3KsvwJR/vTEO9h3/cxnU6haRqWyyXSNIWu64iiCFmW/aSx708J+HY9J57ePUf3b2t7LGKb5/7d6467LyT/McckIiIiIiIiIiIiIiIiIiIiIiIi+hIY8yUiIiIiIiIiIiIiIiIiIiIiIiKiL+Jj461PBX3bx1VVFaqqoq7rk4jtuX3OnVNQVfUkMtg0zbPisN1QYZ+PmQdxzL5rEvPTF6LtxhMvhYwVRfnooKrY/jn3p+1cRPfSnHSv2bIsjMdjzGYz6LqOw+GAh4cH1HX97JBm37nOrRv66RiGgSAIYBgGdrvdFw/5KoqCIAigqiqSJHkUhKbHFEWBYRgIwxCmaeLu7k5+/inPjqZpCIIAQRCgaRocDgdUVfVTD/uzHY9HGIaB0WgE27ax2WzkGi7LEmVZyiD7p8zDpfCt+Llvm+ceu/3ufc5zd+5d2Hd93eCv+Jspovp97/32mPjeJSIiIiIiIiIiIiIiIiIiIiIioq8dY75ERERERERERERERERERERERERE9EV8TEz3Odu3o5JhGAIAttstoihCVVWfFfMVUUZVVXtDg09FaD81TPjcGHB3exFN/JiAcDdE3N62e45L19N3bDGevmhkN5r73PilCPleX1/jm2++gaqqWCwWAADbthHHsTwfw5BfH03TMJvNEIYhlsslsiz7ouNRVRVXV1f4wx/+gNVqhc1mg/V6jbIsv+i4vnaKosC2bXieh6IoEEXRJz9viqLAcRyMRiO4ros4jhFFkYwsH49HxHH8VTzPiqLA9334vo80TbFer1EUhQzXAufj7peO2Y70PnccwnPnRfw9e+4+fe/tc1H59nfPGVdfyPdTgvJEREREREREREREREREREREREREvyTGfImIiIiIiIiIiIiIiIiIiIiIiIjoq/MxsUYR2bVtG7PZDDc3N9B1HVVVoa5r5Hl+EvN9Tijx3Pmbpjkb61VVtTf2+1OEJ/tiieKzvnBiXdePIrztUGTfPmIeVVWVYWTDMGBZFnT9w//ltCgKZFmGNE1R13XvOMV52z+fizN25+q5EUtFUaDrOq6vr/Htt98iz3P88MMPyLIMw+EQvu9juVyejKFvHi/5GoKhv1WKosDzPIRhCFVVkSTJF59v3/fx6tUreJ6HzWYD3/fPhsB/jT4l+vocmqbBdV0YhoHtdos0TZ8Mfp+jqio8z8NgMICiKNhut0iSBLZt4+rqClVVIc/z3sCyqqqwLAt1XaMoiovnaYfZPyUaK0Liw+EQmqZhtVrJiPHHhs7F2AGcvNPF+Nr7Cu3t2u/0j/Hc7T9nrXzqGhB/L34Lzx0RERERERERERERERERERERERH9NjHmS0RERERERERERERERERERERERES/WiLKGAQBXrx4geFwiCRJsFqtYJomyrJ8FAX82EDgc0KwqqpiOBzCdV3EcYw4jlHX9Uk096cME547Vjvq2D5vN6gr/hWfq6oKwzDgOA5834dlWbBtG6qqomkaVFWFsiyhqipM08TxeDwJsPYFgtvnfyqY+TGhUTHewWCAm5sbAMCPP/6IzWYDwzAAQMaH28drR4afwojk5xP3qS/KaRgGBoMBLMvCbrdDnudfaJQfmKaJwWCAoijwxz/+EU3TQNM0GVn9NRPha8uyoGmaDHL/VGvcNE24rgsAOBwOnxU/1nUdnufBsizkeY4oipDnOTRNkxHfvmMrioLBYABd17Hf788eX1EU2LaNMAxhmibiOMbhcOiNA1+iKAp830cYhijLEvv9HkVRPBpb3/um710paJomv+/udyk+3HeOvs9/DufO0RfP77vuc5F9voOJiIiIiIiIiIiIiIiIiIiIiIjoa8aYLxERERERERERERERERERERERERH9aomQ7+9+9zt4noeHhwcsl0ukaQrTNKEoSm9k8Tm6UcVzIVhVVeE4DsIwRFVVKIpCBi0vRXeBy8HC54ZnxbHOhXS75+qGfHVdRxAECIIAtm3LEG5VVfJ6iqKQYWQRBe4GgoVzAWMRdxWfPydK2ed4PMI0TUwmE5imieVyid1uh6qqoCgKDodD7z1nHPKXo2kahsMhHMdBmqZIkgR5nst77jgOXNdFVVU4HA6o6/qLjtf3fXieh81mg/V6jfF4jLquP2pcIrbaXfOO48CyLJRliTRNf5JrVRQFpmlCVVX5nPatb03TYFkWxuMxrq6uoCgK3r17J/cREe/j8Sif7/Y5+p7v7jhEADzLMsRx/Oj6dV2X47x07YqiwHVdhGEIXdexXq8Rx7Hcb7Vaoaoq+e5pn8f3fYxGI7kNgJMQs7gGy7Jwe3uL169f43g84v7+HkmSfFLMdzAYwLZtbLdbJElyMk+apj2a1/aYRbRXfG5ZFoIgwGAwgGmaSNMUq9UKURTJKHxfFPuXci5K/NxocDfq3RdWb0eOv+S1EhERERERERERERERERERERERET0HY75ERERERERERERERERERERERERE9KukaRp838fLly9hWRbevHmD9XqNPM9R17WMu16KUZ7TDuMK5+KChmFgOBxC0zTsdruTaOk5z43WXhpfN4TY/Vl83/63HVMEPkQWXdeF7/swTRNlWSKKIpRlibIsZUizrutH1+R5HmzbxuFwQJZlj87VnUMRpDw39ufOia7rCMMQg8EASZJguVzKORfjT9P0k+47/TQURYFhGAA+hFRFkFR8J9aOuFdfkqZpmM1m0DQNh8MBqqrCsixEUXQ2ktsmYqyDwQBRFCFJErlOZ7MZrq6uAABZlmG5XGK/3z8rHtsXBwb+GqQVsfLdbofVanVyTBHR9X0fg8EAvu/L50M8m6ZpIgxDeJ6Huq5lBB348E4LggBpmiLLsrNzoGmavJfi3SvOr2kagiDA9fU1AOBwOMhn9dw8DgYDjEYjlGWJOI6R57mcAzE2x3EwGo3ksTRNw/X1NcqylFFdx3HkOzmOYxkFDoIAYRhit9thu90+6/720XUdw+FQvm/E3xoxd8PhEJ7nIUkSxHEs30eapsn4sa7riKIIuq7LMHld17AsC57nIc9zJEnyrPiziOUKnxPC7XsfP/dvzVNj7P5NEvuLuLH4ne9uIiIiIiIiIiIiIiIiIiIiIiIi+tox5ktEREREREREREREREREREREREREvzoiInl9fQ3HcXB3d4f5fH4StPycIOBzY4iqqsJxHHieJ8OXz4kvfq6+0GI3knuOCCYqigJVVaEoCrIsQxzHKMtSjl/EF9ufiX1N08R0OoWu64jj+Mkxtj/rfv4x4UkRTh0Oh1AUBdvtFofDQd7rpmlQVRWqqpLbf07Ykj5NXdeIoggAkOf5STRVxJgNw0CWZc8K2/6cXNfFdDrF3d0diqLAcDgEACRJItfRJZqmYTqd4ubmBg8PDzJW++LFC8xmM1RVhcPhIOO6Yl7EviKGWxSFDMOKyHYYhlgulyiKAsCH5+53v/sdDMPAarWCYRgYj8ePwsOapmE8HmM0GqGqKiwWC/kc13UN0zQRBAHG4zFs24amadhsNgA+PDOj0QgvXrzAjz/+KIO6Qvv9Yds2fN+XIWQRtTUMA7PZDJPJBIqioCxLzGYzNE2D+Xze+460bRvT6RS+7+Pu7g6Hw+EkAg38NT5+e3uLw+GAoigQhiFs28Z8PkdVVfA8D99++y1GoxGiKIJt21AUBWmawvM8AMBiscB6vYaqqs+6x22qqsL3fViWhd1uJ+PNImAchiFGo5GMWIvPTdPEYDCA67rydzEnRVFgt9uhqiqMRiPo+l//L/59Yfm2p77/uZwL+XbHc2l87YDvc/chIiIiIiIiIiIiIiIiIiIiIiIi+how5ktEREREREREREREREREREREREREvyoi6Hh1dYUwDLHdbrFcLlGWpQwLfk7I92OIQKeqqsiy7CSo+Us7F89tE3Fb8e/xeESe5yfBVREY9TwPWZbJkKjYXlVVDAYDDIdD7Ha7R8HNn5OiKPB9H77voyxLGfTs0jQNlmUBAMqyPAl2Mu57SlVVGIYBXdehqirqukae509GqUXIOgxD5HmO/X4v5/l4PCJJEgCPn0XLshCGIeq6RhzHJ/dGVVVYloUgCGCaJnRdx93dnQzk/tREuHY4HOJPf/oTTNOU75Q0TZ/1HjFNE8PhEKZpYjKZYL/fIwgCvHr1Cuv1GsvlEmmaIgxDGZMFPrw7RHDXcRxUVYX5fI7lcgnTNDEej/HixQvs93uUZQlVVTGZTDAYDPDnP/8Zm80GjuNgPB6fxF81TcNgMJAh4fV6LcO4mqbBcRwMh0PYti3nII5j+Ry5roubmxt4nncSVVUUBaZpwjRNqKoKXdcxHA4xGo0AfLjnuq7L0PdsNkOWZdhsNqjrGuPxuPeYqqrCNE1cXV3h5uYGvu/L82iahqZp5DtL13UZPwY+rKXBYIAkSZAkCQzDwGQygW3b2Gw2iOMYpmnCcRw55wBkQLcvLv4UwzAwGAxksFrEjsXcep6H4/GI9XqNOI6hKApc18VkMpEx5ziOoes6BoMB9vu9HKu4rjiOH0WU29rv8e47/XNcCvSKz8S52tt1x9M+Vne77nHb19DdnoiIiIiIiIiIiIiIiIiIiIiIiOhrxZgvEREREREREREREREREREREREREf1qKIoCx3Fk1DKKIiwWCxlmBH65kC/wISZp2zaKokCWZU8GUH8uz40fdoOMdV3LWCbwYX5FTFXEfLv7aJqG2WwGXdcRRdFJjPXnpigKBoMBLMvCarVCkiQn91uEnn3fl1HN5XKJKIrQNI2Mh4r7JH4X19eOe54LW/4cNE2DpmkAcBKl/hiqqsoo5nOeARFnFQFSTdOgqiqOxyMWi4WM8fbtJ9ZIEAQIwxD7/R5FUSCKInk9feNQFAWe58E0TWw2G6RpKscuIrODwQBhGKKqKpimiTzPcX9//7PcB8MwMB6PUVUVmqbBcDhEVVXY7XbPWteapsG2bWiahvV6jeFwiD/84Q/wfR9JkuD+/h5RFMn7K+bDMAwMh0NcXV3heDyiLEvYto0wDHE4HGBZFsbjMcIwhKZpMmJ9fX0t4+UirN0N0tq2jel0Cl3XsVqtsN/vZchXRGXDMERZlgCA/X6PxWIh53s6nWI4HCKKIhn4FWtlNBphMBjItRKGIYIgwH6/h+u6ME0TQRBgMpkgyzI8PDxgv99D0zToug5d1+XasG0bruvK5/Xly5cYj8fI81yGjo/HIzabjbwXuq7DcRz5LA8GA9i2jfl8DgAymLxarbBer6HrOkajkRyveNfpui7ffx9DhHnDMJTr27ZtGIYBVVXheR4sy0Icx9jv9zgej/B9H1dXVwiCAFEUYbvdoixLGSTf7Xby/WSaJgDISLA4Zzt22xe/PTdW4Pnvr3Pbt98r7e8vxXe70d7257/ke5WIiIiIiIiIiIiIiIiIiIiIiIjo58KYLxERERERERERERERERERERERERH9KoiI6Gg0wmQyQZ7nWCwWiOP4i0R0VVVFEAQwTRPr9Rp5nv+iIeHP1Y4yilCviCUPBgOUZSkDxe1tbNvGbDbDw8MDoij6Rede0zQEQYCqqrDf709iw7quw/M8DIdD+L4P27YBAGmaIo5jAIDnedB1Hfv9HoZhwDRN6LqOqqqQ57kMnOq6Dsuy0DQN0jSV4VRN0+Q2T3kqWqnrOmzbhud5cBwHtm2jqiq8f//+bEj33HlEVFrTNOR5/ihy3D6nuC5d1xGGISzLgqIoMubqui72+7287jYRS55Op7AsC3Vdy/CoWD+2bWMwGCCOYxwOh5P9DcPAYDAAAOR5Ls/nui5GoxE8z5PjFxHY5853d05EQFdEXKuqOlmrIpA7Go0QRRFc14WmaVgul8jz/FmxUcMw4Ps+FEXBfr+H53n4u7/7O9R1jfl8jqqq4DiOjGMfDgccj0d4noerqyuoqorlcommaTCZTOT9cRwHruvKazcMA5PJBK7r4scff0Rd1zLEHMexHK+maTJou9/vcTgc0DSNvG+z2QzX19eo6xpZliHLMmy3W3mfgiDAdDoFABmdFfFbEd0djUYwTRN1XcMwDBlxFoHi4XCIoihwf3+P3W6Huq7l2hDPqBin4zhyHQ4GA6Rpiru7O+z3exmMbgfDxXrJ8xy2bWM0GqGqKiRJIgPITdNgu90iz3N5T7MsQ1EUSNNUPieGYTw7RC7GblkWZrMZRqORDPLqui6fJ/EMp2kqQ7+z2QzD4RC73Q7z+RxJksig8Xa7lbFly7Lgui6KokAcx6iqCrquwzAMAB+el6qq5FjEubIs633Wnwr+dt9Pfdv2RXm7x70U8m1/1z2W+L07DhFXJyIiIiIiIiIiIiIiIiIiIiIiIvqaMeZLRERERERERERERERERERERERERL8KIgApYpPL5RK73e7ZQcafYzzD4RDAh2Dsp4xD0zTYtg1VVZFlWW+4VNM0GVA8Ho8/aTC4G2IUsU3P8/Du3TsZCRXbiQBqGIb405/+9OzoqaIoMtLYNM2z9uk7hud5sG0bh8PhJCRsmiYGgwHG4zFUVUWapiiKArquy1ikruu4vr6WxwuCQIZexfZinGEYIggCpGmKPM+haRp834eqqlitVhfHr2kaXNeVodOiKGQUVXwvgsmDwUAGhUVUd71ePzvmq2maDBi7roumaXA4HFCWJfI8l9upqgrTNBEEAa6urmQEWVEU5HmONE1R1zWm06lcb33zb9s2ptMpxuMx1us1NpvNybV5noebmxtcXV3hzZs3iONYrtf282vbNnzfR9M0JxHYzWaDt2/fIooiFEVxMTTaR1EUGIYB27bhOA4Mw5Ch4jRN5XWK8YzHY4RhiIeHBxiGIcO2z3nGxJoYj8cyOlzXNTRNQxRF0DQNV1dX0HUdpmkiTVMcDgeoqorhcAjHcbBYLLDdbmGaJvI8h6qqMupqWRbSNIWu63AcB8PhUN4rEbUWYy6KAsCHNR4EARRFQZqmKMtSxp6n0ylevXoF27Yxn88RRRE2m42MNhuGgdFohCAITiKz4nnwPA9N0yDLMqiqClVVUVUVttstVquVjJurqorFYiHfzWLdjEYjqKqKsixlvDpNU0RRBNM0EUURdrsd3rx5g+PxiOFwKM8j1rDrujBNE4fDAUEQQNd1rNdr1HWNyWSCMAzlmrQsS4aW8zyX8+t5HkzThGmaJzHwc8Sz6XkefN/Hy5cvEQQB1uu1/N40TYRhCFVVMZ/PURQFDMPAeDzGdDrFfr/HfD6Xz4OIG2dZJkPY4l2w2+3kuIIggO/7KIoC2+1WhpXFu265XMrnT7zLxPu6/d7uPiNPhcb7tP8GdN8P7Rh8+7O+0K/YVvze3fbcuImIiIiIiIiIiIiIiIiIiIiIiIi+Joz5EhEREREREREREREREREREREREdFXT4RcRQj0/fv32Gw2Hx3QFbFMTdNQVRWKovikOK4IVIZhiOVy2XscEYY9F68Vsc+rqytUVYXlconVaiWvSVEUmKYJ3/dh2zaKokAURc8KUH4qcT5FUbDf71FV1UlcUdd1TCYTHI9HGft8imEYcF1XxkCjKDqJqj6XrusYjUYwDANJkshQqYiQXl9fwzAMvHv3Dvv9HrZtwzAMHI9HaJoGy7IwHA5RVZUM6O73+0eRV9u2MZlM4LousiyT8cyrqyskSYLNZnMydhGoBP4aeJ7NZgCAOI6RJAniOJZx4TAMMR6PEQQBAGC73SKOYxncTdP02fMRBAHG4zEcx5HrWXwnYr4iBuv7PgaDAV6+fIk8z/Hw8IDFYiHvYxAEsG1bRqX7Qs+DwQDD4RBZlmE+n5+EYF3XxXA4xPX1tQzkapom595xHLx8+RKz2QxVVUHXdRl3LcsSRVEgSRIZUP7YNa4oClzXxWAwgOu60HVdrl/HceB53smxLcvC69evZXR5v99jt9s9e11aliVjwCJma1kW5vM55vM56rqG67pQVRX7/R6r1QpJkiAMQzmH+/1ehl/F9dq2LUO1IgwtQraLxQKqqmI0Gsnr2e/38t0j3hki+t0O+b548QKe52G1WuHh4QGr1UrOhZg7EUrfbDaI4xjH4xGmacr1LEKzrutiPB5D13UcDgekaYrRaATbtuV7rB2ZHQwGGI1GWK1WqOsauq4jSRKsVisoioKbmxukaYr5fI7dbgfP82TEt/1siXkpyxJhGGK9XmO/30NVVcxmM7l+RajZsiz5fIlQepZl8p0ax/HFvyHifRiGoQwJh2GIKIpwf3+P3W6Hpmnkc6hpGrIsQ9M0cF0Xs9kMRVHg4eHhJD7e/psg1udkMkEcx3JMlmXJa5rP5zLkGwQBvv32W+i6joeHB3nfu+Hr9vPT931fsLu7bfv37meXjtEm1pc4hriv7e8Z8yUiIiIiIiIiIiIiIiIiIiIiIqJfG8Z8iYiIiIiIiIiIiIiIiIiIiIiIiOirZ5omRqMRRqMRttstFouFjJU+l67r8DwPk8kEjuPgcDjIwObHxANF+PL6+hrD4RD39/ePgpAiwmmaJuI4RpZlj8YyGo3whz/8AUEQYLPZQNd1RFEkw5wiwjmbzRAEAXa7Hd69e4eiKD46hPscuq7D931YliXH3A4riqDoeDzGcrl8dE1d7dDuq1evMBwOEcexjJ1GUfTssYk5H41GMlorgsqDwQC3t7ewLAvv37/Her0GANR1LaOgSZLAtm0ZcTVNE/P5HIvFQsY329foui6qqkKWZXAcB7PZTMZD2/Nh2zYAyACv53m4vr6G67pYLpcAgMFgIK8hCAJcXV3BNE2sVius12sZxBWeE6jWNA1hGGI2m0FRFCwWC2y3Wxn8FOtDBHgHgwGappEBWbHW9vs9yrKErusYDofQdV1GXsV5RJDacRyMRiPouo77+3s5btM0EQQBhsMhFEVBnudIkgRVVcm5DIIAYRji1atXUBQFDw8P+POf/yzXQJ7ncv0Nh0MZ9/0YIp4ahiGqqsJms8F+vwcAhGGI0WgE3/dRVRWqqoLneZjNZlgsFnj37h222+2z4+DtZyHPc2w2GziOA8dx8P79e9zd3UFVVRkUzrJMXs9gMICu69hutydrT8S/LcuCbdtIkgS6rstYs+/7mM/nmE6n8p2xXq9lpBb4ayRWHCcIAgRBgFevXiEIAtzf3+PHH3/EbreToWNxn6fTKUajEe7u7rDZbGSMV4Rqxbsyz3MURQHf99E0DeI4lhHrsiyxWCxO1rSu65hOp/Kad7sdLMuCoiio6xqDwUAGkTebDfI8RxiGUFVV3ivgw9+A4XAITdPkvIpocBAEuL29lYH30WiEMAzlHIl3WVmWSNMUvu/D930ZLO9jGIacEwDyHhZFgfv7ezw8PCBJErkeqqqSMV8xH57n4fvvvz8Jo4v7JKLiADCZTBCGofy7Ju7HZDLBfr/H4XCQz+D19TWurq7wT//0T0iSRN5zsX5EyFnoxnmFbtRX/PtUyLcd5b0UBlYU5eTY3YBxe7924JchXyIiIiIiIiIiIiIiIiIiIiIiIvo1YMyXiIiIiIiIiIiIiIiIiIiIiIiIiL5qIlw6mUxQliXev3//USFfRVFgWZYMhY7HY9i2DcdxUNc1qqpCXdcyWlrX9cWgoGVZuL29xd/93d8hDEPYtg1N02RUU1VV+L6P29tb6LqO9+/fn4RvRcj3m2++QZZl+O6776CqKm5vb+E4DvI8h2EYuL6+xmQywfF4RBzHci76womfS8yR53nQNO1R7FNcVxiGME3zJGAsoo7dORMh39evX2M2m8kwp23bMmL5XJqmYTQaYTgcyvkZDocIggCz2Qy2beP+/h6r1UrevzzP5b1I01QGZcuyxNu3b3F/fy/XUTs86jgOjscjoihCXdfyPHEcY7vdomkaKIoCz/NwdXWFLMtkjFaEP9++fYvdbofBYADHceT1hmEIAHh4eMBisZDB1I+9V+31NZ/PZXwVgLxvuq4jDEPc3Nygrmvc398jSRJkWYbBYABFUaDrOsqyhG3bCIJAxn6rqpL767ouY8i+76MoCmRZBlVVYRgGBoMBRqMRmqbBarWC7/vI8xxN0+D29lbuU1UV4jjGZrPBjz/+KOO5iqLImPVsNsN0OkWapnKun8OyLFxfX8vo9WKxOIl0i/PPZjMYhoGyLKEoCsqyxPfff4/VavXskC/wYT36vg/P8/Djjz8iyzLc3t7ieDxisVggiiIZWG1fg2VZuLq6QlEUiONYRpcNw4Cuf/i/dZumCVVV8fDwgDAMMZ1O8fLlSyRJguFwKO+lWD9tTdMgz3NMp1OoqoqiKGT89+7uDt999x32+/2jeTVNE99++y0A4O7uTs6druu4urqCruvY7/cyPizeQWmaIk1TDIdD+L6Pu7s77Ha7k0Cx4ziYTCZYr9cy1rvb7TCdTuH7Pm5ubnA8HrFarXA4HHA8HqFpGoAPkWdxr3RdRxAESNNUxrjFOvQ8D4PBAO/evcNgMIDruliv1zKO247otp8B13Vl6LdNHPP29hZVVeHNmzdI0xRXV1cAgDiOURQFmqaBruswDAOqqqIsS9R1DdM0MRgMkGUZ1uv1ydoScydi17ZtYzaboSgKGQceDAa4ubmBpmnY7/coigKGYSAMQ1xfX6MoirPHfc77pBvsPbdf+163dQPB7W3Euhfz2A5Gi78T3agvERERERERERERERERERERERER0a8NY75ERERERERERERERERERERERERE9NUSMcjRaARFUTD//7PzLy+SZQu+5/fb7/e2t/kj3DMis+pUqc7lVkkIBJo2aKyJRDcCoYGg/wD9A9JAkx5pIiHR0IOWJhfRICQEmqlBCDS4dFXdU11Vp+pkREZGePjLXttsv98axFnrmHm4e0RkRmZEnvP7gJPhbttsr7322ssmyff2FnEcf3DkU1EUuK6L8XgMy7LQdR2Wy6WMKIof0zRh2zbyPMd2u5UBSREjFOfTdR3D4RCTyUQGJ8MwhOM4Mu5oWRZmsxnG47EMMQoiAnp0dARFUfD8+XNkWSZDoKqqQtM0jMdjTKdT7HY7XF9fo65raJqGrut+kgCioiiwbRuu66KqKjnH++cSsdq2bbHb7dD3PQzDgGVZaJrmIIopjj0+Poaqqnjx4gWyLINlWTJa+iE0TYNhGPB9H6enpxgOh4jjGKPRCKPRCNPpFJZl4fr6Gre3tzIi2/c96rqWsdMgCHB8fIwgCPDixQssl8uD+7I/B47joK5rpGkqY7WqqmK1Wsn4r6ZpODo6wtHREb7//nsoioIgCOD7PqIokqFNsUb2A6V5nr+zLj6GaZo4OzuD7/u4vr5GFEUHUWBxLjFnmqbJyGrf91iv11BVFa7rwrIsVFWFIAigqiqiKEKWZVAUBcPhEMfHx6jrWoaRdV2XcVTXdTEajWQEWIRTwzDEbrdDXdcYDAbI8xw3NzcYjUZI0xTL5RJxHMt70/c9iqLA7e0tmqbB8fExjo+PZfD2fetdVVXM53OMRiNsNhvc3t4iz/OD99V1jfV6jb7v4fs+1us1FEVBVVXYbDYfFfIV98D3fSiKgvV6DQAYj8e4uro62J/ujl3sRa9fv0ZRFGjbVsaZxbNn2zaKosDNzY387L/6q7/CbrfDarXCzc0NkiR5cGxFUcBxHLn/iPt6cXEhY7l3528wGODs7Az/4T/8B6zXaxlxdV0XJycnWCwW8p5pmibj1CJS7nkeFEXBarWSe6fYHyaTCSzLwtXVFaqqQt/3yLIMdV1jOp0iDEMZ+q3rGqZpymdFhKk1TYPjOHKOqqpCFEUyvj0ajaBpGiaTCbIsw+vXr+VYgMP4rNirBoMBBoMBoiiSUeX9+zudTtG2Lb7//nvEcQzTNGVMOM9zee/EWjAMA2maymfLtu2DMQiKokDTNKiqKteN7/t49eoVuq6D53kYjUawbRtxHCPLMqiqCsdx4Ps+mqZBHMfI8/zguvq+l3FsEabfX4N3A7zib4KI8D70vCmK8ujrd4+7+7n3vXc/+ixeZ+CXiIiIiIiIiIiIiIiIiIiIiIiIfgkY8yUiIiIiIiIiIiIiIiIiIiIiIiKiL5au6wiCQIYNl8vlR4U3bduWEdf1eo31eg1d12VIsaoqGdc1TRNXV1fIsgxt28LzPDiOgyzLkKYpAMBxHIRhiLZtZWhU13V4noc0TdG2LcIwxGAwkKFQEYBVVRWWZWE6ncK2bVxdXWG328FxHAwGA9R1jTzPoes6ptMpNE3DbrdDURQyagm8Gwj9FDRNg23b0HUdWZYhy7KDYLKISYZhiDiOEccxgLch4+l0KoOyTdNAVVVMp1OcnZ1BVVXc3Nxgs9nAtm2EYXgQ2X2MiGdOJhPM53P82Z/9GZIkkeHewWAA27ZxfX2N77//Htvt9mDMXdehLEsZan7y5AmWy6WM/t4lQs2macrY7nA4xHA4RBRFiKJIRk4dx8H5+TnyPJdzcXR0BMMwkCQJ6rqWc6aqqpybruugquoHhzHvM51OcXJygsvLSyyXy4OIsmBZFo6PjzEYDPD8+XMZ8gXervnlconz83M4joOyLDGZTFCWJeI4Rtu2CIJABoOvrq5QliUMw5Cx0TAM4XkeTNPEZrORIV/TNOF5Ht68eYMkSbDdbmW8eTgcynPcDRl3XYc8z7FcLqFpGr766itUVYWXL18++rwrioLBYIDT01NEUYTr62tkWXbvseIc0+lU3uuiKJDnudwPPiQSrqoqgiDAcDjEbrdDlmUYDodwXVc+Bw+9bzwew3EctG2Luq7RdZ2MaIv9wfM8fP/998jzXEaIr6+v8Zvf/AYvX758Z4wi3moYBsIwxHA4RNM0yLIMcRzDMAwZgL0b6FYUBZ7n4enTp6iqCm/evJFhW8uycHR09E7I2jAM2LYt9yvHcWDbNrIsQ1EUMuRqGAZGoxGOjo4QxzGiKJLn1DRNBqGbpkEURfLzxfMqjhHB3MFgANM0UVWVXJO+7+P8/BxnZ2cyxv7ixQtEUST3IhG2vRu93Q8H78d8VVWF7/uwbRuXl5fY7XYAIGPCIqQr7oNlWfB9HwCQZRn6vpfheHHc3UitZVkyzi6eizRN4bqu/GmaRkapxXcRADnvIqq9v77F9d3nvr9/zP6zH+i97/e7n6soyjv7sTifeN7EvIjP+qF7IhEREREREREREREREREREREREdHPTf3cAyAiIiIiIiIiIiIiIiIiIiIiIiIiuo8ITY5GI6iqijiOURTFB79f0zTMZjNMJhNkWYbNZiNDpSI+mWUZDMOAaZooikJGRi3LwunpKebzOXRdl/FVEd0EgKIoZFxY13Xoug7HcTAcDqHrOpIkQZqmMmioKIoMBxdFgSiKoGkajo6OYFmWDLMqigLDMFCWJeq6Rt/39wYhPyURJFZVFVmWoaqqg3OpqgrXdeF5HrbbLeq6hq7rOD4+hu/7qKpKXqdt2zg7O4PrulitVlitVmiaBq7rwrIspGn63vvoui7Ozs7wq1/9CuPxGMDbUObz589l6NRxHDRNg9vbWxkq3SfGHwQBptMpfN/H69evZbBWzKc4zrZtTKdTqKqKJEmg6zpmsxlc18Vms5Fj1jQN8/kcs9kM6/UaWZbJuajrWsZ1XdeVYdWyLOX60nUdtm3LdfQxVFXFyckJ6rrGcrmUsc+793I4HOL4+BjL5RKLxeIgVgq8DYJqmgbP8+SaFEFVy7Lkc1NVFaIoQlVVqKoKdV3j5OQET548ga7ruLq6wuXlpZwbwzCg6zryPEee57i8vESSJBiNRgjDEGmaHoRQ76qqCovFAtfX1zg7O0MYhg9GQxVFgW3b+Oqrr1AUxaMh3/35E7HsyWSCPM8xGo0wGo1kePkxIuQ8mUxgGAaur6+hqio8z5Nz9BDHcTCdTtF1nYxzi+dKVVUZBi/LUoapxTVWVXXvvIm9IgxDHB0dYTQaIY5j/Pa3v8XNzQ2CIIDneTIMvL9WNE2T1xIEAa6vr5GmKRRFgWVZGI/HmEwm2G63SJIEXddB0zS4rgvbtg+CuKqqIk1TNE0jxzQajfDkyRPYto3dbiejzGKPNE1TRoeTJJHvFZ+3HwcXwdw8z3F9fS3npygKrNdrFEWBly9fynB127by+jzPk/FcEYvtuk7+3L3n4p7UdY04jmVs1vd9qKoq97++7+UzJI7f7Xboug66rsv5uft8ivslYre2bSNNUxlJFmMS7xPP6P74xPP2UGD3h35PPBT8vbtXinPtn/++c+9Hy8Xa3Y+Z372GxwLBRERERERERERERERERERERERERF8SxnyJiIiIiIiIiIiIiIiIiIiIiIiI6Isjoo7T6RSu6yJJEhlp/FCWZeHk5ARt22K1WqGuawwGA0wmE3RdhyiK0LatDPmKYGvf9/B9H7PZTIZYBRGa1HUdaZpiuVwehGx934fjOEjTFJvN5iDAaZom5vO5jMOWZYkgCPDkyZOD6G1ZlqiqCqZpwjTNnyVwaJomHMdB13XY7Xao6/rgdRExNQwD2+1WxnmfPHmC3W53EC0OggDD4RBJkmC9XqMsS/i+L8O26/X6nc/fp+s6fvWrX+Hp06e4urrCP/zDPyCOY9R1jSiKkOc5PM9DGIbyvolQ6P54VVWF4zhwXReTyQRlWeLm5gZlWco48v7xIupa1zXKspQxUgAyHipCn+fn52jbFovFAl3XwbIshGGIPM/lmjo6OkLTNNhsNjKQnOc5gLdr86Gg7WMcx8FoNMJqtbo3YCzu03A4RFVVeP369b3PjGVZcBwHAPDkyROUZYk0TaFpGkajEVzXRVmWcr5VVZWxURFlvri4wHK5lJ+vqirCMERd13LdZ1mG7XYLz/OgaRqKonj03vd9j7Is5fWdn58/GNgVUWXf97Fard4b8tU0Ta7D8/NzhGEoo7P7ke73fUYQBPB9H1mWIYoiGdzN8/yddSgoioLBYIDRaITdbockSdC2LQzDkPdBRE+XyyWSJJFRXM/zUJblO/dRVVWMRiN88803MhD++vVrvHnzBsvlElEUyTHv71Fibx0MBphOpzIWLPYfy7IwnU4xn89hGIaMX+8HyxVFQVmWMrYLAF3XwTAM+L6P+Xx+EG9umga2bcv9V1EUuWfkeY6yLGUc17IsAECSJDJu7roufN/Hzc2NfOburpUkSeQzLe7J0dERzs/PcXR0BN/3YZomLMuSQeL9edmfV8MwkKapnHNFURAEAZqmkX8X0ePBYADDMBBFkYz53l0zIl5r2zbm8zkGgwGKopAR9clkguFwKOfUtm0ZMN4Pz+u6DlVV5R62//OxsfeP/V65+/l391DxmfuRXrGm9+/L/muqqspw8X3BYCIiIiIiIiIiIiIiIiIiIiIiIqIvlf65B0BEREREREREREREREREREREREREdFff9zJKKgKweZ5/VOgvDEOEYYjf/e53yLIMnufh9PQUXddhvV4jyzIMh0P4vo+u62REV4Q3XddFFEWoqgp93x/EBx3Hwe3tLbIsk3FgRVHg+74MuIoorniv4ziYzWZI0xTb7RamaeLrr79GFEW4vLyUkdOmabBcLvGXf/mXGAwGMmT7U0YOwzCE67pYr9cHYV5BVVUMh0PUdY3tdou+7zGdTmEYBhaLBaqqAvA2XDmbzaBpmgzBuq4r/3ZxcSGDtvcRoVHLsvDdd9/h6upKhke32y3iOJbhXRFkFue++zm2bWM8HqPrOpimiTdv3sho6F2O42A+n6OqKiRJAsdxMJ1Ooes6oiiS829ZFsbjMSzLQpqmaJpGBl7btpVh39FoBF3X8ebNGxn4VFUVbdvKAKqmaQ/GX+8j4q3iebjvunVdlwHQ5XKJoiju/Rzf99G2rbyv19fX6LoOnufBMAx0XYeiKOQ1iri1uN9iHvfnUtd1jEYjGWHt+x5N06Cua7RtK8OuIvpqGAaapnnnOkSkNU1T2LYNTdPuDRKL/WG73cp79BBN0zAYDHB+fo7ZbCZjxRcXF7i+vkZRFB8UChexWkVR5J5kWRZ0XZfXLAKmd983m82g6zqurq5QFAU0TcNwOITnefK4oihk7FoET8U87VMUBdPpFH/zN3+DruvwT//0T1iv1zKwKqLgZVlC0zS53nRdh2VZGA6HCIIAVVXJe5QkiXxmXNdF0zRomkZGa8W6NU1THi/2zL7vZdxb0zQZ2e26Dq7rwnVd/Nmf/ZkMRG+3W/i+D1VVkWWZfL40TYPjONA0DXmeo65rqKoKRVGQJAmurq7e2T9EfFtElquqkmFpEcsNwxC+7yNJEui6jjAM0XWd3Fvv0nX9IA4snvE8zw/Wim3b8H0fZVnKeLumaciyDL7vw3VdGTMW+4Jt21iv14jjGK7rykj51dUVdrsdhsOhvGdJkmCxWCBNUziOI59vseb2Y8MPfT/c99qPDcTvR3r3A753P3s/0Cvec/fYu2NjzJeIiIiIiIiIiIiIiIiIiIiIiIh+CRjzJSIiIiIiIiIiIiIiIiIiIiIiIqIvjqIoMnR5c3ODLMveCcw+RlVVHB8fo2kabLdbuK6Lr7/+Gqqq4uLiQgZpRXBRRGzruobv+zg+PkbbttjtdmjbVoZ8DcOAaZowDOMg5qqqqgw75nmOJEnQNI0cs4hUuq6L7XYLRVFwfn4OAHj+/PlBoFJRFKxWK7x58wbT6RS73Q5lWT4aK/0xbNuWMdokSd6ZZxEiDoLgYN6Ojo5wfX2NNE1lgNFxHHieh6qqUFUVLMuSEdjb21tEUfToWPq+R9d1WK1WMj46GAxg2zZubm7k/XEcB0VRPLguNE2D7/sH92m9Xt8bilRVVZ5ju92irmuEYQjLsrDb7RBFkQwCB0EAz/OQ5zkMw5ChURHlHAwG8DwPZVni+voacRzL4KaIWu6viY+J+VqWBd/3ZSj6vvisaZrwPA9d12G73b7zugizWpaFJElkLHs/rtq2LTRNk/fQ932Mx2Pouo6Liwu5zmezGYqiQFmWB+NLkuRgraqqCgDY7XbI81zGsMW57qPrOlzXlaHY+4xGI1iWhdvb23vDxvvX7Hkenj59itlshiiKUBQFbm5u8Pr160ffe5dhGDJsu91u0bat/BFzcJemaRiPxzg7O5Ovi4jydDqV60F8ZpqmAHAQQ9Y0Da7ryrCv7/v48z//czRNg+fPn2Oz2bwTdhWhXsuyEIYhJpMJ6rqG53mwbRubzQbb7Rbz+RyGYcB1XZyfn6NtW9ze3kJVVRl1FmFf8UzleY4sy1CWJbbbrRyf2EO22y2yLIOqqri6upLh4DiO5TMxHA4BAHmey/us6zp0XUee53KNd12HxWIhA9Z3n/e6rvHy5Uv823/7b/H1119jOp0iSRKsVissFgtcX19jMplgOBzi7OwMmqYhjmNcXV0hiqJ715cI1O7va77vY7lcyiC12GNs20YURTLe3jQNoiiCbdsYDAYIggCqqqJpGhRFcbAPij1CzCUARFGE3/72t+i6TgaTxTiur6/lHN0X5P3QOO4PDebed867gV7xuwhLi3PtR3/3I8D7c/0x3/FEREREREREREREREREREREREREnxNjvkRERERERERERERERERERERERET0xdF1HZ7nAQCqqnow+nkfRVEQBAGGwyE2mw0sy8Lp6Sk0TcPr16+xXq/RNI0MzwJAmqaoqgq6riMMQ7iui5ubm4PQp4jLimBpWZYyrmmaJmzbhqZpB3FKMR7TNOH7PjRNAwAZ0Pz+++9lxFEQIc7VaoXRaITxeIwsy2RI91MSoV5d15GmKdI0fSeoqGkagiCArusysjubzeA4Di4uLmSQVlVVBEEAx3FQliUURYHv++i6DtfX19hsNh8Ur+26DmVZwnVdOI6DyWQCTdOQpinatoXnedA0TQaO786JCCuLyOZ0OkWWZfL8d48X91ycV9zLtm0RxzHSNJUh3+FwiLIsUZalDDS3bQvf92UE+s2bN1gsFojj+GAuxfppmgZ5nn/UmgYgzxFF0YPvFQFRRVEOYtIAZHTY9325dofDIdbrNQzDgGEYiOMYmqZBURSkaQpN0xCGITRNw83NDeI4lqFjz/PkvRbzaBgGFovFwfgMw4Cu64iiCHVdQ1VVaJr2YLzTsiycnJzAdV1cX18/GLG2bRtN0yDLskfnUoSBTdPEy5cvkec5vv76a6zX64++B33fI01TxHGM3W4ng7u73Q7j8RjD4RDb7VbOied5CIIA8/kc4/EYo9EIVVUhz3P4vo+mabBYLOQ+I/YOoSxL3N7e4quvvsJf/dVfIQxDqKqKMAxRVRXW67XcW0SMfH9ORUDWcRx89dVXqKoKSZLg6upKxp7LspRB6yRJcHFxgSRJ5HjEc9H3PSzLQt/3Mj7bti2yLENVVTAMA13Xoa5rOa9d1+H29lbGhsXzpygKdF1HVVXy/onwaxzHMtQt5vx9seY4jvEP//APcF0XZVkeBKWLokAcxzLC3vf9o3F0MWf7sekgCGAYhgz2Am/D2ZPJBH3fY7PZyLC4eHaurq4QhqGcFxFA3v9uyPP8YC8Qz21RFAf3URy/XC5llLxpGjlv4r37x+7Pz4d8b3zMcXePv28M+38T0fP9oO/dzwPwaOCbiIiIiIiIiIiIiIiIiIiIiIiI6EvCmC8RERERERERERERERERERERERERfXFM04SmaWjb9t7g50NUVYVlWZhOp7AsC03T4Pj4GKqq4uLiAuv1WoZYPc+DbdsoikLGK23bhu/7qKoK2+32ICyoaRo0TYNhGFiv16iqCpZloW1b6LoOXdex3W4RxzH6vpcBQ+Bt0NTzPLiui5OTE3Rdh5cvXyKO43vjhSKYGUURhsMhwjBEnufvhH9/LFVV4TgOVFVFmqYoiuLeOK7ruuj7HkVRwPM8OadlWR7cH8uyZODY8zx5Dftxzfep6xp5nsuArrh2Ef0UQWQAMg4pxqyqKnzfx2w2g6IoKIoCg8FABj0fiseKWLHneei6Dk3TyPCoYRiYTqcYDAZI0xRRFMlA7LNnz9A0Db755hs4joPvvvsOr1+/fudcrusiCAKYpoksy7Db7T4obLxvMpkAwEFQdJ8ITXueh+12i6Io5GtiXubzOfI8RxzHePbsmVy3pmlitVohTVMYhoG6rlHXNWzblutaxHgVRcF2u8VwOISu/+F/RRbrY7PZHKwh13UBvA1mi2vWdR2WZclIc13X8pk8OTnBbDbD69evsVwuH50TEY+9L0IqItpBEMhwbpqm0HUdqqpiu91+1N4CvI2v3tzcyMAt8Db+ulqt5NyfnJygKAqYpom+72W4tygKvH79Gpqm4dmzZ9hut7i8vJRxcTGH+5qmwc3NDf72b/8W33zzDb766ivkeY5//dd/xfX1NVRVxXg8xtHREZqmOQgJD4dD9H2P7777DpvNBpqmoa5rOd/7gVhxrvV6fbC2xL4IQO59IjYr9q2+7+V6uavve7Rt++AeV9f1wR4i9riu6z44XC7CsdvtVgaK7zumKIqDZ+IhqqqirmuYpinD00EQYLPZIM9z9H0PTdMwHA7hOI58NsQ1iOva7XbIsgyKoqDrOvn63Zjtfvh2/5pF9Fgcf99n3BfU/aE+5P13o70fMob9se6/71OH6YmIiIiIiIiIiIiIiIiIiIiIiIh+Toz5EhEREREREREREREREREREREREdEXR9M0GeoUUdXHKIoCTdNg27YMp+q6juFwiCzLcHl5idVqhaqqALyN9o7HY1iWheVyibIsZdRVBH7FseLzRdix6zpst1vUdQ1VVWXMt6oqGee1bRtt28px27aNMAyhKApWqxVev36NNE0P4ogiMto0Ddq2RVmWiONYRodN00RVVfcGFX9oGFHTNDiOg67rkGXZg8FdEVZWVRVhGMJxHJRleRDp7PseXdfJeU+SBMvlEnEcf3DIV3xOWZYyMuv7Pq6vr2UQuKoqOV+GYch7oGkafN/H8fExdF3Hzc0NdF1HEARYLBb3hoqBP4SWHcdBlmUyYOr7vgzWNk2DKIpwe3t7EB+1bRsAEEUR2rbFer1GnufydTEXk8kEbdtis9k8GON9H13XZSh0P0Irgsy+78P3fXRdJ9eeqqpyDqfTqQy+1nWNIAhgGAYMw0AURYjjGE3TIM9zuQYtywIAVFUlzymOOT09PZjPJEnw4sWLd4KqnudB0zS5XkSoejAY4OzsTEZmxVirqsK//Mu/YLFY3BuBFYqiwLNnz3Bzc4PdbnfwmljXIsKcJAmiKJKB1q7rDp6/DyUiz/vE9bx69Qqj0QiWZcE0TTnXfd/D930sFgs8f/4cjuNA13Usl0tsNpv3roWqqnBxcYGrqytomnYQmgYgn43T01McHx/LAHPTNFgsFri+vkaWZQ9ea1mWuLy8vPf1/b+pqiqDuD9k/e5TVRVZlqGqqoPP2g/Wfi4ivquqqny2xDMi9l/xXZMkCVarlYz87uv7/p1Is6qqB8fdDfveZz+Uux/O/ZD37r//Y+2f633juvu+u8e8b6zi9ceedyIiIiIiIiIiIiIiIiIiIiIiIqIvCWO+RERERERERERERERERERERERERPTFEUHHtm0fjTuKqKvjOHAcR4Y6RVCz6zpcXV0dRDMty8JwOMRwOISiKCjLEqZpwvM8WJaFvu/lDwAZdVQURcZGi6IA8DYaqqoqgLdRTBEI9X1fBn0Nw8DR0RHm8zk2mw1ev36NKIoOgq+O42AwGKBpGqzXa3nddV0fRGv3r9uyLNi2jaqq7o1JfghN02BZFtq2RVEU98YURZRS13WMx2Nst1tkWQZN0w4Ck7quw7ZtTCYTGf0UYdiPJYK5vu/DcRzEcSyDskmSoCgKeJ4H3/dRFAW6roPneRgOh1BVFZeXl4jjGMPhEG3bYrvdHsSZ97Vti7ZtEccxbm9vEUURFEWR0dW6rrHZbHBzcyPnebfboSgKmKYJRVGwXq/x61//GmEYYjweo21beJ6HwWAATdOw2+1k2PiHBivjOMZ4PD74m2maGAwGCIIAVVVhs9nAsiyEYYjBYIAwDOH7PizLQhzHuLm5kSHYxWIBz/Pw/PlzrNdruR7zPMd6vUZd1+j7HpPJBNPp9CCy7DgOTNOUzwEA1HWN9Xp9MD5FUeB5Htq2PQgC53kuxzIYDKAoCrIsQxzH2G638p4+RFEUGXeez+coyxJ5ngP4Q5x5MBhA13UsFgvEcYwwDDGZTKCqKqIo+tFB2n193yNNU1RVBcMw5Hy0bSvj0svlEsvlUj6n4vUPIdboQ+eO4xjPnz+HZVlwHAfA22eoKIoPiqF/yN4hgtb7kdofqus6RFEkP/dLczcQrigKdrudXJNd18kw92MRdODduX3f78J9UVyx3z72vsf8mPfeHdNDr++fQ0TwH/usD40SExEREREREREREREREREREREREX1JGPMlIiIiIiIiIiIiIiIiIiIiIiIioi+OCFD2fS9juvv2I76DwQCe5yHPc2w2G3lMURQyzipii7ZtIwxD2LYNVVVlpHY0GsmgqPhdVVWYpgnDMNB1Hbqug67ryPNcxmZt24brumiaBk3TyEBnURTwfR+j0QgAkKYp1us1+r5H13VQVRWGYcjob9M0sCxLjlNRFGiahiAIYJom1uv1QfTSMAzM53McHx/j9evXKMvyB0VidV2HaZrymu6LPIqIq23bMmxb1zX+7M/+TMZyNU3DeDzGdDrFdruFbduYTqeI4xhd16FpGqiqCl3XoWkaqqp68HzA23hpXdc4OjqCbdsylNr3PaqqwmKxgGVZOD4+hud5AN4GOJMkwfX1NbIsQ9/3MjC7WCwejPnmeY5Xr16hKArsdjtUVQVFUXB9fY3tdou2bZGmqYzbivHleY6iKGQQ+ttvv8V4PMazZ8/k2trtdliv1zJA/Fig9n3SNIVhGBiNRmjbFoqiYDKZwDRNeR4RnH7y5AmePn2Kuq5RFAVubm4QRZFcQ4qi4Ntvv4Wmacjz/GBcbdvKKK9Y447jYD6fo21bWJYFy7JweXmJsiwfHbNhGHK97M9/13XIskzeSwDyGfqQddz3PZIkwbfffoujoyN8/fXX8tnRdR1d12G32+Hq6gppmkLTNPn8qqoqn8VPSazN/et0HAez2UwGoKuq+sEx5/cR85dl2cGYPpWu6x58hj6WeDa/RGLOqqqCrusYDAaoqgpJksjnpK5rbLdbefxD8/xYuPdD3nNf6Fa8LiLqP8SHRH3vvnY3vnt3nPs/d8cu5kh8rz3kx+yPRERERERERERERERERERERERERD8nxnyJiIiIiIiIiIiIiIiIiIiIiIiI6IvTti3KssRoNMJwOERZlsiyTIZwLcuC7/vwfR8AsN1usdlsUJYlVFXFYrFAEAQyIAsAlmXBdV0Z7S2KAkEQYDgcIs9zLJdLJEkCXdcxm80QBAE0TUNZloiiCLquw7IsbDYbFEUB4A9x4CiKUBQFmqZB13WI4xhVVcEwDBkkrusaf/7nf46/+Iu/wJMnTwDgID5r2zZmsxmAtyFJ3/fx9OlTbDYbLJdL5HkO4A/hRBEyvri4+MERRBFKLsvywVBn27bYbrd4/fo1oijCarVCnucYj8c4OzvDdDpF3/coigKXl5e4uLjAeDzGbDbDr371K2RZhqZpZBAzjuMHQ777ocg8z6HrOna73cHxXddhvV6jbVsMBgNomoYsy+Rni+ivuM//8A//cBDivKuqKlxeXsrQsjhPlmXI8/y9sU5x7SIcK2KyVVXJ4G/btj86rJplGW5ubjCfzzEajeQzcXt7iyRJZBz1+voau90Ouq7LmK+IT++PW6zh+65JBGfTNEXbtnBdF6ZpQtM0NE2DKIqw2+0OAtP3adsW19fXMs5897W2beX9+lhVVeHm5gZZlsHzPOi6LvcNMe91XaPrOrmuVFVF3/eI4/ijz/dDBEGA0WiEOI4fXYOf0qeOFP+pEntsGIYyBr6/P/yQeyn2pft+/9Ag8Pv+ftdjcd2P9dBY9wPD4r/7x+4fL0Lk9x3DtUtERERERERERERERERERERERES/FIz5EhEREREREREREREREREREREREdEXp21bbDYbjMdjHB0dwfd9ZFmGqqqgqipM00RVVYjjGHEcI01TNE0jQ6TL5VIGeffjrnVdI0kSNE0jA6UiThpFEeq6RhRFsG0bTdOgaRoZozVNE4ZhYLvdomkaKIoCVVXhui5ub29lMLXve9R1LY8Rx4nXXNdFkiTY7XZIkkR+/m63w3w+x69+9Ss0TQPbtmVEN45jGY8UAcSyLN8J3X4MRVGg67q8zseikUmS4LvvvpPR377v8fLlSxlTrqoKaZrK4HIcx4iiCJZloe97GbZN0xRlWb43AivOWxQFoig6CL72fY+yLLFarRBFEYC3YU0RUt5/f13X2G63Mk77kPvG8zFxSREfFutThIE/RcRXKIoCL1++hOu6ck7LskRd1wfXVxQFqqqCoihyHD9U27YyfK1pmlx7Yr4/5P0iiPyQHzM/VVVhs9lgt9sdXO9+dFWcQ1EUmKYpr+enpqoqRqMRLMvCq1evkGUZY6W/AH3fw/d9OI6DJEnkfx/bI3/IOe77N/CH4PSniNuK7x1BPCM/9HPvxojFZ+7/96FI+91r7vsemqbxmSAiIiIiIiIiIiIiIiIiIiIiIqJfJMZ8iYiIiIiIiIiIiIiIiIiIiIiIiOiLIgK4SZLg+++/x2g0gm3bUFUVmqYhz3Nst1skSYKiKNA0zTux1izL8Pr1a4RhCMuyUJYl8jxHkiSoqgoAsFqtUBQF6ro+iOImSYI3b95AVVVUVSVDso7joK5rFEWBrutgGAbatkVZlojjGHVd3xssVBRFBm6///57GIYhP3c/9hrHsQzkKoqCqqpk8Hf/+kSwOIoiFEWBNE1/8Dz3fY80TVEUxaNRxaZpEMexvKaqqrBcLrHdbgG8jba2bSsjqk3TIM9zaJomXxf36X3xRvG6CDaLePLdee26DnVdv/c63xfy/VRERPan/Pw0TZHnufz9obn8lOMQ6+2HzuNPHev80HlXFAWGYcjY8U9N13WMRiMURYHlcvmznJM+jSAI8PTpU1xcXMj9/VPtI+97Hn7Ms3s3mns3bv5Dn8X9gO9DnyP+th8MFt/lD33OpxgbERERERERERERERERERERERER0efAmC8RERERERERERERERERERERERERfVFE2K+ua6zXa8RxDF3Xoaoquq5D0zTy56HwYdM0WC6XiOMYqqqiaZqDeK6iKIjjGGmayvisiAk2TYMkSQ6ihLquwzTNgxiwoihI0xQ3NzfYbrcPxh7F57ZtiyzLZNz3rrqusVqtsN1u5TF3A8FC27aI4/id0O/HKooC6/VaBoofc/d1cQ/uI+KvIt74oaFGcZyqqgiCAMDbMPNPGcn9JRFhXfo4IuRrWRaKovhZ5tDzPIxGI1xeXiKOY67hXwgRkq/rGq7r4vXr119siPluvPenCuLuh3rvnv+hMe0Hffc/QwTexd8Y8SUiIiIiIiIiIiIiIiIiIiIiIqJfIsZ8iYiIiIiIiIiIiIiIiIiIiIiIiOiLI8KXImgLvBsOfF8EsKoq+d67x/Z9/2hc825kUFEUtG2L1WolP7NtW2w2GyRJgjiOPygQ+r54oYjgfsjn/Nggadd1KIoCVVX9ZKHRHxpqVBQFvu+jaZoPCg0TPUZRFOi6Dk3TfnQA+0P5vg/XdbFcLlGW5U9+Pvo0+r7HcrmE67roug6LxUJG4L8k4vvwbtD3vmM+5djvC/iKv++PCXj7HbMf9BV/uzuuhwL3RERERERERERERERERERERERERF8qxnyJiIiIiIiIiIiIiIiIiIiIiIiI6BfhhwQJP1XEsG1bGe1tmubgb8AfooW/NF3XfZERRVVV4fs+qqpCnudf5Bjpl6WqKmw2G6Rp+pM/q4ZhYDAYoCgKbLdbrt9fkK7rEMcxvv/+ewD4WdbLx/jYqP1PNYa7sfv9n/vGtB/wFYHfvu+hqqr8OxEREREREREREREREREREREREdEvBWO+RERERERERERERERERERERERERETv0XUdsiyTIUKhbdvPOKo/bpZlIUkSlGXJ0CP9KG3bYrvdomkalGX5k5/Ptm24rovlcvmznI8+nb7vUVUVoihC3/dfbIj5Q/bEx455KLr7sUTA975z7r/2WIT4U42FiIiIiIiIiIiIiIiIiIiIiIiI6OfCmC8RERERERERERERERERERERERER0QdguPfTeyzi2LYtiqJAXdc/86joj03btkiSBGma/ixxVtM00bYtoihC0zQ/+fno0+r7/ovd7/u+fyeM+3Oee9/dkO/d41RVfec1cby4DvH7lxpNJiIiIiIiIiIiIiIiIiIiIiIiIroPY75ERERERERERERERERERERERERERPTFuby8RBRFX2xUk35Zfs511HUdVqsVoih6MFZN9EN9ijX1Qz9j/337IV/x97tx3/t+3w8S933PZ4SIiIiIiIiIiIiIiIiIiIiIiIh+cRT+z29ERERERERERERERERERERERET0U1MUhf+zGhF9MFVVYRgGmqZhzJd+cUzThK7rKMuS65c+ORHEfej3n3Mc+7Fe8bsI9Irf7wZ97+r7Hk3TfNKx9X3/+EmJiIiIiIiIiIiIiIiIiIiIiIiIfgTGfImIiIiIiIiIiIiIiIiIiIiIiOgnx5gvERER0Y8jwrifK957H1VV5ev7Md/3vU/o+/6TR68Z8yUiIiIiIiIiIiIiIiIiIiIiIqKfkvq5B0BERERERERERERERERERERERERERERERI+7G8p9yPsCup/y/Pu/333tsXF8jiAxERERERERERERERERERERERER0aegf+4BEBERERERERERERERERERERERERERERHRpyEiup8ymHtftPe+WO/HxH37vkfXdZ9sjEREREREREREREREREREREREREQ/B8Z8iYiIiIiIiIiIiIiIiIiIiIiIiIiIiIj+SPycgdz9cLD4EX+7Lyq8/9qnjA0TERERERERERERERERERERERER/VwY8yUiIiIiIiIiIiIiIiIiIiIiIiIiIiIiogPvC+6KMK8gjr37nv3fVVU9eP/PGR4mIiIiIiIiIiIiIiIiIiIiIiIi+hTU9x9CRERERERERERERERERERERERERERERER/Sh4L+e6/LqK+iqLIn337f9v/t6qq7z0HERERERERERERERERERERERER0ZeGMV8iIiIiIiIiIiIiIiIiIiIiIiIiIiIioj9Sd+O6n+oz98O8+2Ff8e++7w/O/VOMg4iIiIiIiIiIiIiIiIiIiIiIiOjnpn/uARARERERERERERERERERERERERERERH9kt0Nmt5H13UoioK2bdF13c81tA+i6zpUVUVVVZ97KPSJ7cd1P+Vnip+7n9113UHY9+5/+76X6/9Lew6IiIiIiIiIiIiIiIiIiIiIiIiIPgRjvkRERERERERERERERERERERERERERETvoaoqNE2DYRgwTROGYUDTtINjqqrCdrtF27YHfx8MBjg7O0NRFHjz5g2Kovg5h/4oXdcxn89RVRWWy+XnHs6jVFWFZVmwbRuqqqLve+R5jqqq3pnzL5lhGACAuq4fPe5ThHh/ipCveBZ0XUfbtmjbFn3fv3MuEfHd/128/77jiYiIiIiIiIiIiIiIiIiIiIiIiH4JGPMlIiIiIiIiIiIiIiIiIiIiIiIiIiKin9V+0LPrus89nAdpmgbTNOF5Hnzfh23b0HUdqqqi6zrUdS1jsnVdo67rd65H0zQ8efIE0+kUURS9Ezj9nFRVxWAwQBiGX3TIV1VVhGGI+XwO3/ehqiqapkHf92jbFtvtFqvVCnmef9GBWEVRMBgMMB6PsdvtsNlsHo0Qf6nXous6hsMhTNNEHMcH8973/UG09+7f9n2p10dERERERERERERERERERERERET0GMZ8iYiIiIiIiIiIiIiIiIiIiIiIiIiI6GclIr77kc8vKeprmiZ834fv+3AcB4ZhQNM0NE2DoihQVRXKspT/bppGXs9+oFRRFIRhCNu2sd1uUZblFxUwdRwHg8EAcRxjt9t97uFIIvTc9z0sy8JwOEQQBEjTFBcXF6iqCsDbUPJwOMR0OoXruri6ukKSJF/UWtrnOA6Ojo5gWRaiKPqi1sJDFEV5Z5ye52E6naIsS8RxLI+7+z4AB5Ff8e+2bd95VoiIiIiIiIiIiIiIiIiIiIiIiIh+KRjzJSIiIiIiIiIiIiIiIiIiIiIiIiIiIgBv46jiR1VVqKoK4A8hzrZtUdc12rb90efq+x6maSIMQ+i6jsVi8Uk+98dyXRfj8Riu60JVVdR1jTiOURQFyrKU1y+CsY8FSS3LQhAEMvpbVdUXE5o1DAOj0Qi2bWO1WslArqqq0HUdbdt+1vuhKAoMw4Dv+1BVFbe3t0jT9GBMbdtis9lA13VMp1OMx2PUdY2iKL64UKyqqphOpwiCALe3t19EdPi+UO9d4nVxrK7rGI/H0HUdq9VKBqrF6/tR37uf/aXdEyIiIiIiIiIiIiIiIiIiIiIiIqIfgjFfIiIiIiIiIiIiIiIiIiIiIiIiIiKiP2G6rsOyLLiuC9d1YRgGdP0P/5tx13UyOtp1HcqyRJZliOMYTdP84PNqmgbf9zEej2VI9nPTdR22bQMAoihCnueoqgpN03xQvPcu27ZhGAayLEPXdUiS5IsIFgOQodyyLJHnufy74zhwHAdZliHP888SYO26DqqqQtM0dF2H7XaLoijeid/2fY+qqrBYLGDbNjzPg+u6qKrqi5lnwbIszGYzZFmG9XqNuq4/63gURfmgmK84VvzX930cHx9juVwiz3O0bStj3wBk0Fd89n4MeJ+qql/cPSIiIiIiIiIiIiIiIiIiIiIiIiL6EIz5EhERERERERERERERERERERERERER/YwURYGmaVBVFXVdf5ZYqmDbNkajEYIggGmaAN7GONu2RdM0aJpGxjr3Y5y6rsMwjB8V8zUMA2EYwnXdd0KtiqJ8lthn13VI01RGd7uu+8H3R1VVhGGItm3lT1mW7wRpPwcRUrZtG5vN5uA+KooC0zRRluUHx15/SmVZ3hvy3VdVFdbrNZ4+fQrbtr+oUKyiKDAMA9PpFEEQ4PLyEkVRfPZ53Q/tPubucz+bzeA4DtI0lfvXfc/J3c9XVRWKovygKDYRERERERERERERERERERERERHRl4QxXyIiIiIiIiIiIiIiIiIiIiIiIiIi+mxUVYVlWTIMW5blFxPh/Cmoqgrf9zEajdC2LRaLBcqy/CxjMQwDw+FQBmc3mw2KokBd1zI+ux/kVBRFRnYB/KiQLwB4nofxeAxFUZCmqTyPpmkYDofo+x673e5Hn+djdF2Hqqo+SWjUdV2Mx2OsVitomoa6rn/Wa3mMruuYTqfoug5JkhyEcvM8h2manz063Pc9mqZB13UfNJY8zwEAlmUdBGg/F0VRoOs6HMeB7/uYTCYoigK73e4Xucfpug7XdTEcDrHb7ZCmKZqmOXhW9uf9vmfoQwPCRERERERERERERERERERERERERF8yxnyJiIiIiIiIiIiIiIiIiIiIiIiIiH4iInyqqiratv3sccwviaqqcF0X8/kc4/EYABDHMdbrNaIoQl3Xn3mEPw3DMDAejzGdTrFerz9r2FLTNCiKgvV6jTzPUdc1+r5H13Uy5HufTxFKFVFj0zSxXC6Rpqn8+2AwwPn5Od68efNZnpn9qLDjOGiaBlVVffRYhsMhfN/HxcUFhsMhyrJ8J376qYl789g5FEWBbdsYj8d49eoViqI4eL1tW2RZhq7rPuv67Pv+YB2qqiqjyPcR4zYMQwanPzUxBkVRHnxOVFWFaZqwbVuOxTRNWJaF5XKJoih+cUFb8V0WhiFM08RisUBVVfI18d/HYr1938sguPidiIiIiIiIiIiIiIiIiIiIiIiI6JeIMV8iIiIiIiIiIiIiIiIiIiIiIiIi+iKJMOCXRlEUmKYJx3GgaRqapkFRFKjr+iD2KYKZjuOgbVskSfIZR/1lMQwDw+EQ8/kchmEgz3MZcw2CAEVR/OTR08/Ftm0MBgOYpomyLNG27Wcbi1iXZVmirmvoug7f91GWJfI8f/B9n+K+GIYBz/MAAFEUyTCoruuYz+fo+x5Jkny2ALaqqhiPxwjDEOv1Wo7vY94/nU7RNA26roOu6yjL8ie7HlVVYRgGbNuGoigoiuLBALF4/oC3c3/fGqyq6tEw610/1X4txmAYBgaDASzLwtXV1YPzmOc5XNeFqqoH0en3jU3EagHcGzFWVVXu56ZpQlEU1HWNLMtQFAXatoWiKDAMA47jyO+Huq7Rti1s20bbtthsNveOXVVV6LoOXdfR9z2apvnke+CPvUemacrwepqm6Lru3s8Uv9+Nfou/i3v6x7i/ExERERERERERERERERERERER0Z8GxnyJiIiIiIiIiIiIiIiIiIiIiIiI6IsiIriqqqKqKtR1Lf8u4oCfK/Cp6zrCMMRwOITrutA0DX3fY7vdYrVaIcsyOTbTNBGGIRzHwW63Y7jw90zTxHQ6xcnJCdq2xXK5xHa7RVEUsG0bk8nknQjkHwtN0+B5HnzfR9u2SNMUfd/L6/2510jTNHLN9n0P27ZxcnKC6+vrR2O+n4Jt23BdF13XyWivoihwXReDwQCvXr1CURQ/6Rge4zgOjo6OZIz3ffdGVVWYpgnDMGS8NQxDLJdLGIYhg9/A25iuoijysz+Uoigy9tq2rVw7pmliMBggDEPouo6maZAkiXyu9seuqqqc4ziOURQFVFWFqqpyP7sbJt8///6zqWkaDMOAYRhQVRV936PrOnRdh7quH/ycj9H3vRzzcDh8NHArxibmBHi733RdJ+f/7nvFnFqWBdu20fc90jQ9WHti3xfzC7z9DjJNU153WZYyUC0+RwSVfd+H53nIsgxZlr0zbsMwEAQBRqORXCtpmsrQ9n4E90Pt3ydd1w8CxB9yX/bnUoSOfd9HEATI8xx5nt8b8X1sjAz4EhERERERERERERERERERERER0R8LxnyJiIiIiIiIiIiIiIiIiIiIiIiI6ItimiYmkwn6vkcURWiaBsDb+KdlWajrGnme/+xBX8MwMJ1OMR6PoSgKuq5D0zSwbRuO48hAJ/CHIHEQBOj7HlVVfbYA8ZfEMAxMJhOcn5+jbVtcXV1hvV7LyGnXdXAc5yBg+cdEBJ5d18VqtZIh1SAIoCgKNpvNz3rdIgoLQIZ0fd+Hqqo/6Xk1TYPv+zBNE0mSoKoq+cyI5yuKos+2BkS8Vdd1bDabe8OlgqIosCwLYRjC930AQNu2GI1GsCwLWZbBtm0ZLBZroOs67Ha7D94XbNvGbDbDbDZDmqYyduw4DubzuQzdJkmCtm2haRpM00RVVfIei2sLggCmaWK5XELXdbiuexAijqLoIGYrQsWWZcmYrfibrusy9rp/LXVdY7fbIUmSH30fDcNAGIYwTfPRdSEivpZlYTqdwrIsOI6DPM+xWq0QRdHB3qKqqtynHceBruvQNE0Gkdu2hWEYGI/Hcl0mSSLXg23bME1TvlcEgauqQpqmqOsamqYhDENomobdbie/z8R3hW3bGI1GCMMQtm2jaRpYliWPCYIAuq4jyzLEcXxwL++7/v25EfMxnU7h+z6qqsJ2u0Ucx/eGjfepqgrDMAC83Zd1XcdkMoFhGLi+vpZ7ttgr9oPXIgD8vvv+x7jHExERERERERERERERERERERER0Z8GxnyJiIiIiIiIiIiIiIiIiIiIiIiI6IsigqL7AU3TNDEcDuF5HlarFfI8/1nHpOs65vM5jo+PUZYllsulDAp7ngdVVVFV1UEk0vM82LYtA5J/6jRNw2AwwMnJCRRFwevXr7FarWTcEngbQRVx1z82qqrC930Mh0MoiiJjn4ZhYDAYwLIsGdn8HHRdh+M4qKrqvaHPH8uyLARBcBC0FZHZMAw/6zOjKAo8z4Pnedjtdoii6MH1qKoqXNfFeDyWQeaiKGDbNk5OTtA0DTRNQ9u2yLIMiqJgOBxiNBpht9vJoOuHjOf4+BjPnj3DaDTCcrnEZrOBoig4Pj7GeDxGnue4ublBHMfQdR22bb9zDzVNg+M4GAwGMso7Ho9hWZaMEiuKgrquZfTWMAz4vi/jyyLcKwLmeZ7LdSteM00TnufJ+fgxa1oEd8MwRFmWSJLkwWN1XcdgMMBgMIDneXL+dF1H27YoyxJ1XaNtW3kNo9FIRnTLssRoNEIQBIjjGF3XYTqdYj6fo2karFYrbLdb1HUNVVVlgNpxHBndLYoCSZLIGLKIBSdJgjiO0fe9jN26rouTkxOMRiOkaYrNZoM0TWGaprxmz/NgWRZubm7kd87dYK9t2wBwEG7u+14+U8PhUB5v2zaKojjYd++uE/GZw+FQ7se+7+P4+BhN08iIvW3bUBQFTdMgz3MZLxYx4jzPD+LD4toZticiIiIiIiIiIiIiIiIiIiIiIqJfOsZ8iYiIiIiIiIiIiIiIiIiIiIiIiOiLoaqqDE3meS4jmrZtYzAYQNM01HX9s8UARXRxPB7j/PwcVVXh+voaURTJSGFVVdB1HWVZynGJAKau60jT9CCc+KdIhGyPjo7gOA5evXqF5XJ5EHoUiqL45CFZVVUB4LNGJC3Lwng8hu/7SNMU2+1Whjnruoau63J9fw4izpkkyU86BhExFVFREbW1LAthGEJVVdze3j54r0ScVQRaxc+nIiKmABBFkYyy3qVpGnzfx3Q6heM4MuJaVRVOTk5gmiaqqoJt29hsNqjrGrZtYzabQdd1rNdreY1inwEO16iYqydPnmA6naIoCnz33XcoigK6ruP4+BjT6RS73Q6Xl5dIkgRd18kgr9h3VFWVsebZbIajoyOUZQnbtuVx4jMnkwnW6zVUVYWqqhiNRhiPxzJUW1UVDMOAYRiwLAtRFMngsYj5ijHbtn2wpjVNA4CPul+apskw+nq9fnBtqqqKMAwxn8/RdR02mw02mw3atsVkMoFpmjAMQ64d0zQRhiFs20aWZdhut2iaRp7Lsiy4rovj42MAwHK5xHq9RtM0cBwHqqrK51fEjquqOpgLERe2LAuXl5domgaGYcio8tHREU5OTuR+IN4bhiHCMITv+9A0DWVZomkaKIoC0zQPvgMdx8F8PkdVVVgulwdrSkSGkyRBkiRwXVfOlWEY8hpEvFvsu5qmyfsu4sJnZ2c4OTnB1dUVwjCE67pyzYo9o21bGbcGgOvra7mfizUu/v1TxsKJiIiIiIiIiIiIiIiIiIiIiIiIfmqM+RIRERERERERERERERERERERERHRF8M0TUwmE5RliaIoZPRxPzT6UFzzp6AoCnzfx9nZGXRdx/Pnzw8inMDbmG/TNHKsiqLI8XZdhzzP/6TDhSIqOZvNEAQBVqsVbm9vHwx6fsrgrohf2rYNRVGQZZkMRP+cRJR1NBpB13XEcYwkSWTIN4oipGn6SaO0H8v3fZimiTiOD+LTmqZB13U51h+7lkWcVdM0RFGELMtkADYMQ2RZhiRJ7n2voigIggDz+RyqqiJNU8RxjCzL5JhFUHY/GNr3/QetK03TMJlMEAQBNpsNsiyTcVxFUWSMdn9N+76PKIqwXC5RliUMw4DjOOi6DrZtwzAMuQeMx2MMh0Os12vkeY62beXaME0TZVnKY8U+cnJygtlshuVyicvLSxRFAcdxMBwOMZ/PEUXRQchXjE+M23EcOaYgCHBycgLHcbDZbJCmKXa7nZy/IAhgmibatoWmaRgMBpjP5wCA7XaL7XaLqqpgWZYM0uZ5LiO/4tyapsn9WlyLCAPXdY04jj/4Odd1HUEQoO/7R9+n6zrOz88RBAH+8R//EW/evJFhXMMwUFUV2raFbdvwPA+maULXdRnRbZpGxm1FyPfs7Axt2+LNmzcy5CvWSNu2SNMUiqJgMBjAdV0sFgv0fQ/XdaFpGizLwtnZGWzbhq7rGA6HMvI7n89xdHSE7XaLi4sLrNdrVFUFTdNkENeyLKxWK1xdXaGua7iuC8MwsFqt0Pc9NE3DdDrFkydP8Pr1a3nPgT9EfgHIse/HfMMwhGVZyLIMURQdPAMiaC1Cv0EQ4Pz8HIqiyCBx13Xoug5N0xzEhg3DwGAwgKIocj7uYsyXiIiIiIiIiIiIiIiIiIiIiIiIfukY8yUiIiIiIiIiIiIiIiIiIiIiIiKiL4IITzqOgyRJZDjUMAzYtg1VVVGW5c8aPBUxzSAI7g35Am9DoftjEjFERVGQpulBGFUQcVDg8XitOK7v+/fGD/c/80OO/yFEiBIA6rr+oPeYpikjpnme4+rq6pMHdVVVhaqqB/dCxFAnkwl83wcALJfLe+/hQx6b0/1orAhbPja+IAgQBAGapsFut0NZlgDe3v8sy+Q5fsyYfihVVTEYDGCapoyEin97ngfXdZHn+UHoV4Q7RdDzfeMVY7UsC0EQAADSNEVd1zAMA57nQVVVbDabB9eW53l49uwZnj59ivV6Ddu2oWkamqZB13Vy3IZhQNM0KIqCtm1RFAXKsnzvfXddF6enp9jtdjKOa5omNE2DYRgwDEPGaYfDIQaDAVarFW5ublAUhQzzBkGAsiwxGo2Q5znyPJcBV8MwkCQJyrJE3/cwTROz2Qye58nP6fteHn92dobFYoFXr14hyzI578fHxyiKAm/evJHRXAByLsUc6Lou47siYHtxcYHnz58jTdODALBt2wCAPM9hGAZOT08BAFdXVzLkK56xJElk3HV/XkWYWYSh67qWUfRnz57J8PCHEDFux3EOQsf3CYIAz549QxRFuLm5kXHi2WwGXdexXC5RVRUGgwGm0ymapsF6vcZms5F7hthHDMPAdDqF67r453/+ZyyXSxnKdV0XR0dHWK1WSJIEpmnK/UWMQ9d12LYN3/fx1VdfYbvd4vT0FHVdoyxLqKqK8XiMJEnw7bffYrVayVCw7/uYTCbwPA9VVWG32yGOYxn4VVUV6/VaHjufz9G2LbbbrbwOwzAwmUwwHo9lCNiyLPkdKqL5lmXJ+yPmW8R4TdPE1dWVvHe6ruPFixe4urpCnueo6xpN06BpGlRVha7roKoq6rqWa0O4754x5ktERERERERERERERERERERERES/ZIz5EhEREREREREREREREREREREREdEXQcQJRRBQhAlN04RlWei6TgZ+fw66rmM4HGIymWCxWOD29vaDQsKmacJ1XQBvo5h3A6Ii9Og4Dtq2RZZlMuy5f4y4bl3XUZYlsix7MAwsYqOmaQIAqqqSscW7RLCy67qPjsHquo4wDNF1HaIoeu97NU3DYDDAZDJB13W4ubk5CI++71ymacr7/tDcG4YBx3Gg6zqKopDBTxHkFLFL27bRdR2SJJHhVVVVoWkaVFVF27Zo21beLxE/FVHboijk+lNVVd5DVVVRFIV8/T6WZWE8HsO2bSwWCzkH+/fhLlVVAfwh2Csio+JHURTUdY2iKNA0zQ96LsT1W5aF+XwO4G2MVFVVuXbG47EMe+6fQ7xnt9thu93K1/aDz2KtaZom16Pv+/B9H1mWIU1T9H0Px3EwHo9R1zXiOJbBXF3X5ZxqmoYnT57AcRz867/+K66vrzEajWAYBnRdh+u68H0ftm3L8wOQ90zEkx8zmUwQhiEuLy/R9738PBH0bdsWVVXBdV1MJhMZjs3zXF67eM9ms4FpmlgsFqiqCpPJBLPZDJvNBrvdTl6X4ziYTqdyrCKKGoYhzs/PURQFvvvuOxl8VhRFhlr//b//90jTVP5dRIa/+uorOI6Duq6RpqmMEzuOA9M0cX19/c5zKNZ0nueoqgrj8Rij0Qj/+q//is1mc7DG2raV9/NuQFvEbbMsQxRFAN5G0c/PzzGZTPDy5csPXquqqsoI7Wq1uncPEAHak5MTeJ6H58+fQ9M0TCYTjEYj6LqON2/eYLVawbIsDAYDWJaFKIoOPlNEiD3Pg+M48H0fFxcXB/FvwzAwHA7lehbfWbquI01TGSlXVRW6rmMwGKCua2w2GxRFgTiOUZalDJu/ePHiIOQbhiHm8zls20bTNNhut9jtdlBVFZ7nwfd9OaeWZWE4HMI0Tdze3srQs6Io8nOyLMN2u4WmaXBdF23bou97hGGIMAzf+e4Rxw2HQzRNgzRNZdx3t9vhxYsXiKJIzofYm8QeJsLkt7e3UFVVzoc4bj9ETkRERERERERERERERERERERERPRLxpgvEREREREREREREREREREREREREX0RRLhRBDNFwNSyLBl1FXHAn5qiKHBdF/P5HE3T4PLy8oPPbdu2DCfux3xFNDUMQ5yenmIwGCCOYywWC9zc3KAoCvn+4XCI4XCI8XgMXddxe3uLy8tLGToVVFWV8Uzf92Gappy/OI4Rx/FBrFKEb8VxRVGgLEsZbhWxxbsBYkEEKuM4/uC5ENewXC6x2Ww+KIgsAqcnJyfY7Xa4urpCkiQH7xUBY9/3EYahDNuK14IgwGQywXa7xWKxwLNnz+C6LizLQlVVME1Thjs1TUOe5zK22fc9XNfFYDCApmky1CsCnqPRCJPJBJqmyXUpwpv3BX09z8N4PEbf90iSBKqqYjgcyt/3Q7MiqhqGIdq2RZIkck727zPwNhZ9e3t7ENN9HxExNgwDruvCcRwEQYDj42Psdjs4joO+71FVlYwQZ1mGLMsO1lIYhvj666/x29/+9uDzgyBAGIbo+17OjwjEAsBoNIKmadhut8jzHIqiwPM8BEGA77//HlVVyXjpZDLBq1ev0HUdBoMBRqMRvv32W2w2G/R9D9u2MZlMYFkWRqMRXNdFkiTys8V4RST3MaZp4uzsTL5PrFsAaJoGeZ6jLEvYto3BYIC2bXF5eSlDvmLdietrmgYvXryQMdb5fI4gCPC73/0OWZbJ+RkMBrBtG1EUySCr4zg4OjqC7/v4u7/7u4Nz2LaNk5MTbDabg/1AURRYloXZbCbnP0kSLJdLrNdrGYsVe8Pd9SKejZubG5imiePjY+R5jt1uJyOw+2uoaRq0bXsQZ9U0DZ7nwfM83NzcoG1bWJaF09NTPHnyBNvtFmmavvdeCCJQXpblvXuOCEUPh0Ocn5/LiO7p6Sl0XZf7axzH0HUdo9EIQRAgyzKs12v5rIogehiGMrq9Xq9xdXUl90axJofDIdI0RZ7nsCxL7vVRFGG9XqOqKhiGga7r4Ps+FosFXr9+jaZpZMzatm0sl0usVit0XScjwdPpVIbrxX7UNA2CIMBoNIKqqkjTVM7zdDpF13VYr9dyrdu2jePjY+i6jvV6DQAyOFwUBUzTRBAE6LoOcRwfhM1FINiyLCyXSzRNg+FwiDAMsV6vEcfxQXxcRHrvBn2bpoGiKPfuSSLqS0RERERERERERERERERERERERPRLxpgvEREREREREREREREREREREREREX0RNE2D7/to21bGNxVFgW3b0HVdxjR/DoZhYDqdwrZtGdH9EJqmwXVd6LouI62KosAwDIxGI8xmMxkIff36Neq6hmVZcBxHRiCfPn2KyWSC3W6Hy8tLaJom48b7gURVVeG6LsbjMYbDIdq2xW63kxFJRVFkmFTM7XA4hOM4sG0bTdNgt9ths9kgTVN53cDD8VNd1+G6Lrbb7XvnQlEUObYkSWTo8n1c18VXX32F8/NznJ2d4fr6GlVVoWkaGUEVYVvP82DbNvI8x3K5RJ7nUFUVvu/j6OgIWZbh8vISZVkiTVMcHx/LWO5oNILv++j7XgaR+75HXdfQdV1GLUWAVIR8x+MxvvrqKxRFgeVyKe+h7/tynHeDy6PRCKPRCHVdw7ZtnJ6ewnVdKIqC58+f4+bmRt5f13Xx7NkzPH36FNvtFldXVzJanCSJjL4OBgMcHR0hTVPsdruD82mahr7vZYh0n+M4ch14ngdd1+H7PgzDkGHpLMvQdR3Oz8+haRp2u51cI2KdiGsS9wR4G8T99a9/jbqusVqtkOc5sixDURRomga2bcP3faRpiu12i7Zt4bouhsMhdF3HZrORx3311VfwfR+vXr2Crus4OztDURTYbrfouk5ep+M4GI/H8DwPt7e3uLm5uTeo/BhVVTGbzRAEAb777ruDWLIIt7ZtC8dxMJvNoGka3rx5czAniqLA933MZjO0bYssy/Dy5UsZBhYx29VqJWOn4l50XYftdoumaeTaG41GuL6+xmq1Onjmh8MhXNfFb3/7W9R1LV8TUdvhcIg8zxFFEYC3cVfP8+QaL4rinedQnFNRFCRJgjAMEQQB3rx5c+8zK6Lfd/8tzlHXNTabDXRdx3g8xunpKQDg9vb2g6PoIjitaZqMbN8dg4g5f/PNN3j27JkMUEdRhOVyid1uh6ZpZIg3CALUdY3b29uDqLD4rNlshtFohDiOcXV1Jde22MMnkwlM08TNzQ2qqkIYhjAMQwak94PjmqYhSRJcX1/L/dKyLEwmEwDAcrlE3/cymD0ej9E0DdI0xWAwkHuWCJMbhoHNZoMsy+T9sm0bV1dX8vtJ7DXj8Vjut4PBAI7jIE1T9H0v1896vUYURajrGoqiQNd1GQ0WEWex3sR4xd52N+Z7N/Ar5nz/Xgn77yciIiIiIiIiIiIiIiIiIiIiIiL6pWLMl4iIiIiIiIiIiIiIiIiIiIiIiIi+CLquw7ZtFEUhI5UiHqvrOsqyvDdOetd+8PaHEOHH2WyGOI6xWq1koPFDrsHzPABAmqaoqgqmaWI2m+H09BRZluH777+XIUnbtvHkyRMZlRThy9evX+PNmzcoigKKokDTNHRdJ48Tgcf5fA7f97Fer3Fzc4OiKKCqKo6Pj2EYhgwiDwYDnJ6eyiiypmkA3oZzsyxDnucwDENGhR8KbmqaBtu2ZVz4MZZlyfCpCNB+yHv+4i/+Apqm4b/9b/9b/OY3v5EhSxEJ7fseT548wfHxMXRdx2KxwGKxQJ7nAN4GRc/PzzEYDPCb3/wGWZZB0zSkaQrXdaGqKk5PT2GaJtbrNbIsk/FUy7JgGAYGgwE8z0MURVitVjIeOxqN8M0336AoCrx+/RpxHENRFFiWhdFoBNd1kSQJAMj7FgQBzs/PMR6PsVwuURQFVqsVPM/DkydPYJrmQcj317/+NTzPwz//8z+jqip5vpcvX2Kz2cj4pmVZct72g65BEMho63K5PJhfEakej8eo6xppmqIoChRFAU3T8PLlS6zXa/R9LwOwaZpis9nItSci0qPRCNvtVs67qqo4OjrCdDrF3/3d38mx7od+xRyLOKthGAjDUAaiy7KEaZqYTCaYz+f49ttvAQCj0QiTyQTffvut3ANM04TrujKyK0LHHxvyBd4Gjo+Pj2UEN01Tud+I6zZNE4PBAL7vY7fbyXUhaJqGwWCAMAyxWq1wfX0t49Kj0Qi2bePFixfyOdB1HaPRCEEQyCCziLsOBgMoioKbm5uDPc91XcxmM2RZJqPGYq2J+VVVFcvlEqvVCoZhwLZtDAYDWJYFVVVlNHifeG9RFCjLEsPhEGVZIoqigzm4q+s6tG17sNZt28Zut0Nd1/A8D57noes6RFGE9Xr9wXupqqowTROKoiBN03fGLGK/4rnZbDb4zW9+g5cvX6KqKrRte/Bc+L4PXdex3W4RRdHBOEQcdzaboe97XF1d4fb2Vh6jqioGgwHG4zGSJEEcx9A0DWEYQlVVJEkiQ77iXti2LY9t21ZGxgeDAVarFeq6xmQywWAwgGEYiOMYu90OR0dHACCfEc/z5PMs7ofneXIsy+XyICA8mUxQVRV2u91BuDeOYxkNFp8lnl1N0+B5Hk5OTuD7PhaLhYyUj0Yj7HY7bLfbd0K8fd8fzPPdtbEfehbH7/+XiIiIiIiIiIiIiIiIiIiIiIiI6JeKMV8iIiIiIiIiIiIiIiIiIiIiIiIi+uxEnFQEVsuyRN/3MjJomibqun4nBCkityKG6vs+LMtClmUoiuLBCOVjNE3DeDyG67r47rvvZPBQjFMEJJumeWc8hmHA8zy0bYs8z6EoCiaTCc7Ozt4J+SqKIgPGIkA6mUyQZRlub2/lHPR9f3AdqqoiCAI8efIEjuPg4uICi8VCjqXvexm+bNsWYRjim2++Qd/3eP36NbIsw3Q6haZpaNtWntv3fRl+fCi2qGkaTNN8b1RZURSMRiMMBgPsdrt34pn3MU0TX3/9NSaTCf72b/9Wxk1XqxWGwyGGwyHquoau65jP5xgOh7i+vsabN2+QZZmcmzAMcXx8jMViIf8u5qWua5ydnSEIAvzud7/DarWSc2yaJnRdRxAEGAwG6Pseu91OhnzDMMTZ2RmapsF3332H3W4nrxV4G9+0LAuapslI7Wg0wvHxMc7OznB7e4t//Md/xMXFBYC3AdmmadD3PVRVheM4+Prrr+F5Hv7+7/8eaZri6OgIXdfhxYsXWK/Xch3oug7TNFGWpQwvizjskydP4Ps+Li4u5PWJcaqqijRNsd1ukWUZmqaBZVkIggBZlslIswjT6rqOKIpkoFicJwxDOI6Dq6sruRZ838eTJ0+wXq9lDNTzPBwfH8t49HA4hGEY8voNw8BsNoNpmri4uEDXdfA8D0dHR6jrGpvNBrZt4/T0FH3fY7vdynDsYDCAbdtomga6riNJEpRl+egae2itjsdj+L6P6+trrNfrd2LWiqLAcRwZlBZ71P7rtm1jMpkgz3N8//332G638jrn8zl2ux2ur69l4Ni2bcznc9i2jdVqhaqq5D7oOA6yLDuYd0VREAQBgiDA9fU1iqKQr4n5EM+vWPv7cWrbtmUwdv/5VhQFnufBdV0sl0s0TSPjsWIPemjeuq6TcWnXdTGdTgEAWZbBcRwZW9Y0Dcvl8uB5fB+x1wC49762bYs4jqGqKjRNw+XlJa6uru7d9z3Pw3A4hKIoSJLkIPisqips28Z0OoVpmtjtdjKMvv8dJK4tiiKUZYnRaIQwDAEAeZ4ffGdZlgVFUWSQV3w/iXhzkiQ4OzuTEd/FYoEkSWDbtpz7NE1h2zaOj49h2zYuLi6Q5zl0Xcd4PEYQBLi8vESapvI6RHB6t9thOp3CMAxcXl4iiiK5b3Zdh9vbW7l/6boOy7IwHA4xmUywXq/lvud5HizLwnfffXdwD+6un4fWyP7ec997iYiIiIiIiIiIiIiIiIiIiIiIiH6pGPMlIiIiIiIiIiIiIiIiIiIiIiL6E6coig3g/wPAwtv/r+y/6vv+f60oytcA/h2ACYD/BsD/vO/7SlEUC8D/GcB/H8AKwH/c9/3LzzJ4+kXTdR2apskY43A4hGVZaNsWhmHAtm04jgPP86CqqgyNAm/Dha7rIggCGcJ0XRd/8zd/g67rZNjxY+OeIqY5m81QVRWiKJIhRhEmHQ6H0HUdm80GFxcXyLJMjstxHNi2LYOPYRji9PQUqqri1atXSJJEHqvrOgaDAcqyRFmWUBQFeZ5jOBxCVdV7x6eqKgaDAU5PT2EYBl6+fHkQeRVRUdu2UZYlXNfFn//5n8O2bfzTP/2TDHnudjvM53MURYGqqmBZlpznh0K9mqZB0zQZHX2MbduYzWYAgO12+2gUVHz2yckJvv76a/zzP/8zttutvKau6+R9HA6HCMMQrutis9ngzZs3MkoJvA0Ci8DmfuBYBDtd14Vpmvj++++x2WwOosdN08hQsmVZ2O128lodx5FxzFevXsl5NE0Ts9kMhmFguVyi6zocHx9jMBjAMAwZVo6iCK9fv8bV1RXyPIfv+zBNE23byvk/OTnBdDrFt99+iyiK4DgOHMeR4c39QKlpmrAsC3meoygKGRuez+fwfR95nh+sS+BtRFOstX2WZcnQq5gPx3EwHA7RdR3SNJXx0/1orQgti0jwfD6H53l4/vy5PP6rr77CX/zFX8jg6GAwQFVVqOsatm3LdScisyLuGwQBbm5u0LYtPM9DEARYLpfyc8T4qqqCYRgoigJ5nr83GH0fXdcxm83Qti3W6/VB6HWfbdvQdR1pmiJN04O5tSwLx8fHmM/nuLq6kmthP2q9//zrug7XdeE4DgAcRJ0ty4JlWe9Ehfu+lyFkcW/FfjWfz3F+fo6yLLFer5FlGdq2Rdd16LoOpmnCNE1kWfbO/RchYFVV5X5n2zY2m8298yki5GJPzvNcPjdBEOD29haWZUFVVZRlKQPr2+32vRFwQVVVGIYBVVVRFMVBuHifYRg4PT2Ve64Y393rG4/HGA6HiKJIxtkVRZHrWYR8y7LEdruVz7emaTLubVkWkiRBHMdo21aGd+M4lvMt3mPbNoqiOHhuVVWVQXLxTOV5Lv/ddZ0MaBdFIUPvuq6jaRqkaQpVVTGdTjGfz9H3PZIkkesMeBstns/nUFUVm80Gr169QlEU8pmdTCa4urqS99kwDPlda9s2AODy8lJ+H00mExnV3o/FfwwRS3/fdwYRERERERERERERERERERERERHRLwljvkRERERERERERERERERERERERFQC+I/6vk8URTEA/H8VRfl/AfhfAfjf9X3/7xRF+T8B+F8C+D/+/r+bvu//XFGU/wTAfwbgP/5cg6dfHhFH/eu//mtomgZd16HrOqbTKXzflwHMoihkaFKERkUYUcQzh8Mh1us1DMPA+fk5fv3rX+Pv//7vZbjxY2O+uq4jDEN4noc3b96gbVuYponRaITpdApd15FlGUzTxNHRERRFwffff488z2EYBoIgQNd1Mi58dHQEy7LeCfmKiKTneTLGqygKdrsdjo6OMJlMkOf5wfjFXJydncEwDLx58wabzUZGQEX8cTQaYTQaIY5j+L6P4XCIf/mXfzkIS4owo4g4ijjsfrj1LhEabZrmvWFGz/Pgui7SNH1vxFNRFHldm80GNzc37wQjRRByNBrJ8OrNzY2MUorxua6L8XiMqqoQx7F8v2VZmE6nCMMQr169wvX19cGYTNOEpmkyjtp1HYqikMHLwWAA3/exXq+x2WzkuEejEY6OjhBFEYbDIabTKZqmQRzHSNMUpmnCcRwsl0tcX18jyzK5/n3fl2HbMAwxnU7lcSIK2rYt0jR9J6rqeR50XZcxz8FggCAIoCgKiqJAmqZIkuTRe7T/WQDkfVIUBcPhEK7rIo7jg3VrGIZcX6vVSp7DcRycnZ2hKAqsVisZ79xut3jx4gXSNMVwOESWZVitVliv12jbFrPZDLquI4oi1HUtg7AiKq0oilyb4lziXoqQrHhWiqI4CB5/KBE6vbm5QRzHD8ZKRcxaBHLFGnBdF8fHxzg9PYWmaVgul6jrWr52cnIi46p1XcMwDBmPjaJIBlw1TUPTNNA0DaZpoqqqg+tRFEWux6ZpZIR5Pp/LWO9qtUKWZei6Dq7rwnVdGUxXVfXe4LGmaTg6OpKRYhFn7vte7ktiTjRNkxF1ANjtdqjrGo7jYDAYyOdeURSsViuUZYmvvvpKBsTfF4JVFAWGYci5FgHq+/YkTdNwfHwMx3Hw29/+FicnJ7BtW4aWxRoUsV7xvOyvExHjns/nWCwW8v1N08hnVDynYm8UUXkxtyL+u79eDcNAFEUH4XPDMGBZlnzGRcBXzLH47rFtG+PxGABkyFjc68FgAAC4vr7G2dkZwjBEnufQNA2j0QjffPMNBoMBvvvuO7x69QqapsnAt67rqKpKrjHHceC6LnRdlzHzzWYj9ygR214sFu/Mv5jf/fv2IcRxHxsEJiIiIiIiIiIiIiIiIiIiIiIiIvrSMOZLRERERERERERERERERERERET0J65/W1UT1Ufj9z89gP8IwP/s93//LwH8b/A25vs//v2/AeC/AvC/VxRF6Vlno4+gKAo2mw0Mw0Acx3AcB4ZhYLlc4uXLlzJgGgQB0jRFnudomkZGToMgQBiGyLIMbdvC8zyMx2P87ne/w+XlJYbDIVRV/egxiSBv27ZyXCIyvNvtDuKVX331FSaTCdbrNaqqkuFWEYocjUZwHAebzQar1eogHmvbNk5PTzGfz2W0uGkaJEmC9XqN4XCI7XYrg4+qqsrgrWmauL29xWazkWFY13VhmiZUVX3nd3H+/YiniDfqug7XdeE4DjRNQ13XD4YW90O3j82tCEGKgGee54/GGzVNw2w2g6IoePXqFcqyfCcUKaLDo9EItm3j8vISm83m4FgRqBTrqCgK+X7f9zGbzZBlGd68eYM0Td8JsoqgqmEYMqIpXhsOh6iq6uA+2raN2WyGpmmQZRnm8zn6vsfNzQ12ux0A4NmzZ9B1HYvFAtvtVsahxfmWyyX6vsdsNkNd17i8vJThzLZtYRiGDA2L+2eaJqbTKYC3sU+x7pqmQVEU0HUdu93ug0PWo9FIrj0RJB0MBtA0DVmWyaixWGuz2Qy2bcs5VhQFQRDA9318++238rx93+P29lZGpB3HQZqmuL29RZIksCwLuq6j6zoZyPZ9H9vtFp7nyQCtbdvQdR15niMIAjx79gxxHOPy8hKWZcl5eCwY/RjXdeF5HlarFaqquveYvu8PnhfDMOR7xZpM01Q+b33fw7IsnJ2dYTweoyxLJEkCXdcxn88RBAEWiwWSJJGR1jiOZZy66zqYpglFUeSPruswDAPj8RjffPMN6rqGrut48+YNbm9vZcS673t4ngfHcWQsWsyjeH2fWOM3NzcoigKGYSBJEoRhCMdx5PUbhgHP8zAcDtG2rQwHi7UhNE2D9XqNNE2haRps25Zr87F9QFVV+L4vn7+6rmWk+b5jJ5MJnjx5gu+++04+z6enp7AsC0VRoG1bqKqK6XSK2WyGqqrkPHRdJ6PQk8kE19fX2G63CIIAtm3j6OjoYP9M0xSe50FRFHRdB8MwZKx6t9vJ8LJ4Xuu6Poinq6oKwzBQVRXKskRVVfJ5VhRFRpANw0DbtijLEtvtFl3XybD20dERdrsdbm5u5HU8ffoUT58+lRHvm5sblGWJMAzxb/7Nv5GB3ouLC9i2DcMw4Ps+zs/PZbA8iiK0bSv3f7H/OI6DsiyxWCzkdYiI792Y7/5aeij0+6HBXyIiIiIiIiIiIiIiIiIiIiIiIqJfAsZ8iYiIiIiIiIiIiIiIiIiIiIiICIqiaAD+GwB/DuD/AOA5gKjve1FHvADw5Pf/fgLgNQD0fd8oirIFMAGw/FkHTb9oWZbhxYsX0HUdbdtiPB5js9kgjmPc3Nwgz3OoqgrTNJGm6UFo1LIsBEEAwzCwWCxkjDLLMlxcXEBRFDRNI6OEH0p8tuM4MiIZhiEGgwFWqxVub29lULhpGqRpCtd1EQQBkiSRgVYR9nVdF03TYLFYHAQpDcPAbDbDN998AwAyPAkAZVkiiiIZ+MzzHFmWwXEczOdzmKaJzWaD9Xot45dBEEBVVSRJIoOig8EAtm2jbVssFot3Ip77YV8RkiyKAmVZynDjffNjmiYAvDfma1kW2rZFnufvvQ8imJwkiQxY7hNBUBH1zbIMy+USSZIcHGuaJobDIQBgs9m8E78dDAZ4/fq1jFfu3w8xV4ZhyHtYVRU0TZPzc3t7izzPAQCWZeHJkyc4OTnB7373OyRJgvF4DFVV5ZhEQLiqKhkeFnNomibyPMdms4GmafB9X16TUBQF8jyXEU4RmnVdF2EYYrFYoG1bGS+t61qGSuM4PrjGh6iqiiAIsFwuZYRXBKHF/RPXo+s6BoMBZrMZuq6TwWJVVXFyciLDpyKqqygK2rZFlmUy3LvdbuVn7odP27bFZDJBVVXYbrcYDocYj8dYLpfyWfI8D0+fPsVut8ObN29QVZWM+ZZl+YNjvpZlwTTNd9bFPkVRZOh6OByi73t5bJZl2Gw2Mt5dVRUcx8Hx8bEMK4vxn5+fQ9M0XF9fI4oiAMDFxQXG4zE8z0NRFEiSBLe3tzg6OoKu68iyTM5/0zT4u7/7Oxk5v76+RpqmcBwHWZZhOBzK9RtFEdI0lc+MqqoPBlhfv36N1WolQ6y3t7c4OzvDycmJXLeapsm53m63iOMYXdfJubm5ucF2u8V2u5WRX03TsFqt5L7xEFVV4bquDA6LSLdYG3ePHQ6HOD09xZs3b3BzcwNFUfDixQsMBgOMRiOEYYi2bdH3PUajEcqylPu1iHqHYQjXdbFcLnF9fQ1VVbFarTCZTGQoXoSnAcj733UddF1Hmqao61rOgxhb27Yypi3mu+s6tG0rvycGgwHKspTXraoq8jzH1dWV3GdEmFvs7XmeI4oi+V0i4r2apqGqKvlcic/v+x5JkiDLMrk3iL1QVVWs12tsNhsZhRZruqoquQ/GcSyj8iLU+5D3vQ7gwQgwERERERERERERERERERERERER0S8NY75ERERERERERERERERERERERESEvu9bAP9dRVGGAP5vAP47P/YzFUX5TwH8pz/2c+iPj4j51XWNuq6haRosy0Lf98iyDFVVoW3bg4iriMyKGKrv+2jbFkVRQNM0uK6Luq7RNA3CMJQx1o+hqqoMuxZFga7rZGxU/C5ihCLmOxqNoGkadF2H53kwDAN1Xcuw73q9xm63OwiiDodDPHv2DIPBAN9++y22262ck77vsdvtMJlMZLCzbVsMh0P4vo+madA0DQaDgQwwZlmGJElQ17UMIIt5iuP4nejt/vW2bQvP82BZFpIkQdu28u93ic8UY3iIoihwHAdd16Esy0ejsiL8a5qmDLTefd33fQBvo74iuLlerw8iwSLoPJvNkCQJkiRB3/dQFEUGaDVNw3K5RJZlB/FNMdY8zxGGoYxjNk0DVVXl+MT7TNPEeDzGdDqFbdtI0xRlWWKxWODo6Aij0QiGYWA4HMJ1XWw2G+R5LuchDEOEYYgoilCWJYbDITzPw4sXL96Z17ZtcXJygtVqJQOpIhIrorFVVSFJEnieh8FgIJ+LfSLEevdeOI4D27ax3W5l8NkwDLiuiyiK5LjFPQ2CQEZWRdjYNE0cHR2hruuDGLN4VsQ9FsFSEQ12HEeuJ8/z0HUdlssl6rrG9fU1/vIv/xJ//dd/jdFohOPjYywWC3z33XdYrVYyMKuqqlyPPyQSKtaUiIffJda84zgwDEPeBwCoqgpxHGO320FRFDk3p6en8DwPTdPg5uYGVVXh66+/xq9+9SskSYKLiwsZUQWA5XKJzWYDAPL+N02DLMtk6DVNU+x2Oznn+/uFOP7q6kqGWne7nXzuRdxZ7GF3FUWB58+fy3O3bYsoiuSzYRiGfD5EMHw/Dt73PcqylMHX/TVW1zUuLy/l5z7ENE14nifDwCLEe/c9mqZhNBrh6OgIcRzj6upKBna7rsNiscBqtYKqqtA0TT6Pi8UCSZLAtm1MJhNMJhMURYGrqyusVit57be3t9hsNuj7Xu4B+0FeoaoqLBYLOX/iu0F8L7RtezD2vu+R5zkWiwXCMMTR0RGapkFRFMiyTO4hd58dEYNfrVbvfKYI/t6N6Ir9b/9zxPkvLi5wdXUl18x+jFnsL33fQ9d1GXcX91qcY/98iqJAUZR37ufdMd13DBEREREREREREREREREREREREdEvGWO+REREREREREREREREREREREREJPV9HymK8l8D+B8CGCqKovd93wA4A/Dm94e9AXAO4EJRFB3AAMDqns/6zwH85wCgKMrHVxbpT4phGKiqCnmey6ChYRhwHEeGFUVo1HEcGZ8tyxKmaSIIAjRNA8MwZMzyseDsfUR4VNd1NE2Duq6Rpils20YYhmiaBkmSyMBjlmVomgZVVUFRFHieJz/LsiyUZYndbiejnbquIwxDnJ+fw/M8bLdbGfvcVxQFbm9v8fTpU4RhKIONqqpiMBjAtm0kSYIoirBer2VMUkRTRfRX13XUdY2yLN8JnYpwadM0sG0btm3LGK4IhxZF8U6QUdM0ZFl2ENK9S4xDRF/vC4juH+s4DjzPQ5qm8jr6voeqqpjNZphMJliv13AcB6qqIkmSd8amaRrCMMRoNMJyuZSxV8uycH5+DtM0AeCddeF5HqbTqYzWDgYD5HmOPM/lelIURa6rIAgwHo8BAFEUYTaboWkatG2L3W4Hy7JwfHyMs7Mzee/359IwDPi+D13XEccxuq6DbdswTfMg5imkaSqjwsDbAO7JyQnquobneYiiSM6biALf3Ny8s/bvuwcidFwUxcHci7CuuM+qqsLzPAyHQyiKIuPU4h4FQQDbtrFYLO595nRdl2t2P5JtWRZ0XYfjOGiaBhcXF8jzHH3fY7FYoK5rTCYTaJoGwzDw8uVL3N7eynNrmibvq4iKfmzQV0RGi6JAEAQoikIGrU3ThOu68DwPfd8jTVMZcAUgI8Jd18EwDJRliTRNEQQBkiSRz6Y4TqyROI4P5um+OHaapvK5B96GcB97jkQAebFYyAiuOL4sSyyXS6iqKkPK+8RzKnRdJ8+v6zo0TZPjF5Hd+9w3vrZtEcfxo/dFRIN1XUee5weh4H26rsuIdpqmuL6+Rl3XB1Fh8dO2LSzLwng8Rt/32Gw2WK1WB+H4oijkcy8+oyzLe+foLvFci/MKYp7uG39d11itVkjTFKZpous6Gcx96D0i6PuQ/ciu8FA0WXzWfZ9393OapkEURffu33fPJ/bl/fjx3djvfsxXUZRH1zIRERERERERERERERERERERERHRLwFjvkRERERERERERERERERERERERH/iFEWZAah/H/J1APyPAPxnAP5rAP8TAP8OwP8CwP/992/5f/z+9//f71//f/cfW1Ak2qOqKnRdl3FFEdUVsc62bVHXtQztep4H13URRRG6roNlWRiNRojjGEVRYLvdyuDoxxARSEGEKHVdh+u6ODo6wnQ6xeXlpRynbdsyLOp5ngz7GoaBOI5lnFhEa09PT2W0UkQORah1fxy73Q5JkmA4HCKOY0RRhDzPYZom6rqW0cv9MKMYh+/7WC6XGI1G0HX93nnYj2daliVjyABgmiZ0Xcft7e1B5FPEUu+e9y4RXxUh0/fFfMXrIvioKApM08Tx8TEmkwlubm5QliV830ccxzKSux+MtG0bQRCg73tEUYS+76HrOiaTCYbDIZbLJcIwPIhLOo6Do6MjqKqKoijguq4MOe8HlkWg8uTkRAZ3V6sVbNtGlmWwbRvb7RZ938s4r4jX1nUtg5iqqsJ1XZimiSzLkCSJvG/i/CLwKwKn4/FYxoVVVcVoNEIYhjIknSQJqqqSQWQR4b27Jd+3RYtY7Wazkc+XpmkyritCvkEQYDgcwjAM5HkOXdflvKuqCt/3AeDe8wJvn29FUWSYWLy3bVtUVYUkSXB5eYkoiuTzV5YlVqsVttst4jjG06dPYds2gD9EYy3Lgud5MlD6Q76Guq5DWZaI4xjT6VTef1VVYVmWjIyLceyvu3378VMR+hZrpW1bpGn6TmT3Me+LuN53vIjt3jc28fcPnaPHwq8f633nFPu/iLnfd07TNDGdTjEej7HdbnF7e/tgdFd8fwRBANd1sdvtsN1u5frcD8veF8P9sdf12N/LskRVVVBV9SA+/Dnsf/+IvUuM5UPvvWVZmE6nqKpK7rv7nw/gYJ/e/y8RERERERERERERERERERERERHRLxljvkRERERERERERERERERERERERHQC4L9UFEUDoAL4v/Z9//9UFOWfAPw7RVH+twD+DsB/8fvj/wsA/xdFUb4FsAbwn3yOQdMfDxEEjaIIVVXJqKCqqrBtG0VRyKisZVnwfV+GS0UYMwgC6LqOKIqw3W4PIrQfqus6VFWFoihkpLeqKqzXa+R5jul0iqdPn6KqKlxdXcnxdV0HTdNk3FUEdLMsk+FGEY61bRu3t7c4OztDVVUwDEMGQ/fDjm3bIooiTKdTGXmN41hGO+9GIPfDqpqmYbvdwvd9BEEA0zRRlqUMWYpI7H5UtG1bJEmCzWYDx3EwGAyQ5zm22y3atj241izLHp1HTdPk9bwv3Ciuv21bnJ6eygjvYDCAqqp48+YN4jjGZDKBbdu4vr5GlmUHY9d1HWEYysjtbreDoigIwxBHR0cyxJokCUzThG3bMAwDw+EQqqpitVqhKAoEQYCyLJEkiYyfdl2HNE1xc3ODvu+RpimSJEGWZVAUBUmSYD6fy/lxHAd5ngMAfN8/iLoCbyOZy+USTdOgLEtomoY0TRHHMU5PT2FZFpqmkUHgLMvw4sULua7CMETbtlgsFlgulzJQKgK7RVEchIgf4zgOAGC5XMrrFSHqpmmgKAqGwyFc14WiKHL9GYYho9FiDYu1cl+YVNM0Ge4VgdCmabDZbOQzsh/6FUSENooijMdjjMdjGd5VVRXT6RSmab4Tnf4YXdchiiI4joMgCDAajdC2rQw6R1GEJElQFIV87u4jgsxlWUJV1YNjH4rs/py+9N7+ftx2n2EY8H1fPv/b7RbL5fLBkO/+++bzOQBgs9kgTVP5DH7uubgbjf/c9iO+H8M0TXlf7oau74uJi3gwERERERERERERERERERERERER0R8DxnyJiIiIiIiIiIiIiIiIiIiIiIj+xPV9/xsA/717/v4CwP/gnr8XAP6nP8PQ6E+AoijQdR2apiHP84MIZt/3MugJvA00hmEI3/cB/CEACbwNEh4dHeG7776ToccfEiis61pGdAeDAZbLJaqqQtu2MmDquq6MwjZNg67rYJom2raVwUjxX9M0YVkWhsMhTNPE1dWVjIWKaOxwOERVVQdB0r7vUdc1mqaBZVmwbRt1XR8EE+8yTVMGafM8RxRFGAwGGAwGMpJsWRbG4zFM08Rms0FZlri9vYWqqlgul8jzHH3fYzgcYjweo6oq5Hku5x6AjNU+JssyeJ4H13WRJIkMtYqgo2macF1X3ssXL17A8zw8efIEVVXJAG+e59A0Da7rIk1TLJdLFEVxMAcicut5HtI0ha7rGA6HmM1m6Ptehpd3u52MBIvg8mq1wm63AwCUZYmbmxsZgxbRzTiO5TnFPen7HlmWYbFYwPd9jEYjNE2D7XaLrutwfn6O0WiEKIoO7pkIU3ddJ9fLdruFoigIgkAGk6uqwmKxwG63kzFgEbb9p3/6J6zXa8RxfBAJ3u128jn6EIZhyDjx/roV/x4OhzJqu9lsZERXVVWYpon/Pzv31dxGkqh5/ynv4AiCTuqemZ45JmIv9/tf7xfYs+fMO66NJIoOhC3v3gtF5iSSWSClNurWPL8IBUmgUJWVlVW4Ufwty5KB42+//RbL5dK4Nruuk6FbPZIswsxqnFmX5znevn2Lq6srXF1d4eLiQu7z/v4e6/X6SQj4YxRFgffv32O1WsHzPBnxFvfksYivSsRoj53LS1mWBcdxftR5/VaItT0ajeB5noyFJ0mC0WiEIAiQ5zmur6+x3+9fFG6Oogiz2QyPj4/YbDb/EvP4scSaFvfxxwjDEIvFAqPRCNvtFlmWveg+MQWbiYiIiIiIiIiIiIiIiIiIiIiIiH6LGPMlIiIiIiIiIiIiIiIiIiIiIiIios/K8zy0bSuDs8A/o69VVQEAgiA4COiWZQnXdWUYtm1bBEEgX1P/2bZ9ENo9pmka7HY7TKdTXFxcyOiuUBQFqqqCbduIouggkitCr7Ztw3EcnJ+fI45juK6Lqqrw8PCA3W6HKIqQ5zm6rkPf9wjDEKPRCGmaynHati3PR+zjWAjRsiwZ/d3v96iqCmmaIk1TzOdzee5hGMJxHKxWK6zXaxmOFQHTruvgOA7CMEQYhpjP59hsNoiiCNPpFFmWoSiKo3NZ1zXSNEWSJDg5OUHXdcjzXEZggyCAbdtomgZpmspQru/7sG0bdV2jLMuDCO5+v8cPP/yA1Wp1EOa0LAthGGI2myGOYxRFgcvLS/i+DwC4vb3FarVCEARYLpcIwxC+76MoCmy3WxkatixLRimrqpLBXjXsChwGosuyxMPDA9I0lWtVxIevrq7QNM2TuWrbVgZt9f1st1sZqG7bVq4n9bPL5RKPj48yIi3keY6bm5uDe+Y5dV3L81WPIda467qo61qGg6uqgmVZaNsWnufJ0Pbj46MMQ5vWp9hOD3Xr82Aigt7r9RpVVcl7oa5r5HmOoih+dKi173sZGxbPjOcCwz8327Y/27F/aSII7TiODPg6jgPgw72xXq8Pws/PxWBd18VisYDjONhut0/i3/TUxwR9Pc/DbDbDaDRClmXYbDbyO+SYvu8/KRxMRERERERERERERERERERERERE9GvEmC8RERERERERERERERERERERERERfTYixlvX9UHsU2iaBq7rYjwey7jmfr9HEASI4xjz+RyTyQRt22K73WIymWCxWMjwqGVZMhpbluWz42nbFmmaYrPZYDab4fLyUoZ3RZR3vV7Dtm2EYYiyLNG2LVzXRdM0qKpKhoZHoxHiOMZ+v8fDwwPW67UMx67Xa7iuKz8/mUwQhiHyPEdd1/A8D/P5/CDOeyx+KuLCruvK44jjJkmC6XSKruvQNA02mw12ux2KogCAJ6HaPM+x2WwAAOPxGHEcIwxD9H0vQ7LHiPiq4zgYjUY4PT2VYxdzlGXZkxhrnucH41CvyXK5lKFadR5s20aSJIiiCPv9HnmeIwxDFEWBx8dHGZpsmgbv37+H53noug5lWcq5F8crikKGXNUxDMUn27bFfr9HlmUHAVjXdbHdbuU4jl03EQzuug51XR+dVwCD25giwc8R8Wj9/LbbLYAPc5tlGdI0fRI3VkPL4voNzdOPje0CH85vt9vJuX5pDPiljoWyP4fPGRL+pfV9j7IssdlsUJYlfN+Xz20Rlhbr9LlrZFmWjIhvNhtst9ufZP19qdT5fEloV0TmRSh5u93K70ciIiIiIiIiIiIiIiIiIiIiIiKifyWM+RIRERERERERERERERERERERERHRZ2NZFlzXRZ7nMhgKQMY6u65DGIZwHAf7/R77/V4GXEejES4vL1HXNa6vr2VA9dWrV6jrGk3TyEjkS0OdXdehKAosl0t0XQfP8+B5Htq2RVmWMt46Go2exFpF0LCuaxngLYoC6/Ua+/1eRiXF/kUc2LZtzGYzzGYzLBYL2LYN3/eRJImM8pZleTSY6DgOoihC13Uy0pplGW5ubpAkCRzHkecgApnq/tSQY1VV2G636LoO0+kUYRiiqiqsVitst9tn57Lve2RZhr7vkaYpPM+DZVlo2xZVVcngsXq9X3JNTEQ02XEcvHv3Do+Pj/I8syyTxxAxWHWMerD3U2KuIuCrapoGy+VSRo1/rTHRoTCwuBcBcyT4c4Zvf8qA76/Zryks/EsQge26rmFZlnztY9ea4ziYzWZwXRc3NzfIsoyh2Wd87Frr+15+J+nfIz/1sYiIiIiIiIiIiIiIiIiIiIiIiIh+rRjzJSIiIiIiIiIiIiIiIiIiIiIiIqLPRsQai6J4Ej2tqgppmiKKIjRNgzRNsV6vYds2PM/DbDZDXddYrVZI0xRhGGIymcB1XTRNg6IokKYp0jT9qKBq0zTY7XYoyxK+78NxHBmUraoKfd/DdV34vo/tdoumaWTUMI5jOI4jI8JZlqGqKrRti77vZdQ2TVMZ0HUcRwaHx+Mx4jhG13VYrVYA8GzIVwSR4zhGVVXI81xun6YpiqKAbdsyPGsKZOphWxH0retaxozFubyE2D7Pc9i2DeAwzvlTRR2DIECSJGjbFnd3d1iv13KuxbmI33+pqGff99hut8jzHGVZ/uYCtOL6E/2SRHT7x4iiCLPZDEVRYLfbDQar6dP0fY+yLGWM/aXUWDwRERERERERERERERERERERERHRl4AxXyIiIiIiIiIiIiIiIiIiIiIiIiL6bLquQ9M06LruIMIKfIj5rlYrGcrd7/coigKWZeH29hbb7VZGY0Ugd7PZwHVdGR2s6/qTIq5N08ggsAgRiv1YlgXLshAEAaqqkvFfEdUtyxL7/R5ZlqFpmiehXPFTjcxmWSYDwEEQwLIsdF2H+XwO27bhOM5gENGyLPi+jziOcXNzg7IsD471MdFFdZx1XaNpGhkC/tgYo5ivnytma9s2RqMRkiSRc17XtYwmf05i7hiwJPplWJaF2WyGKIrw8PCALMt+sYD3v4q+79E0zUc/X/kcJCIiIiIiIiIiIiIiIiIiIiIioi8NY75ERERERERERERERERERERERERE9FnleQ7gaXS273vsdjukaQrgn3HYvu+RpinyPIdlWWjbVv4UMVex3Y9likH2fY+2bdF1HcqyRNM0qOsaZVnCtm35t/rZY2MRoeCu61DXNdI0hWVZcBwHSZIgiiIEQYCiKA4iwABkQDhJEnieh91u90nx3mNj+7livD+W53kYj8dwHAfL5RJVVf2k1/7H+jWMgehfheu6mM/naJoG6/X64HlAzxOBXs4ZERERERERERERERERERERERER0fMY8yUiIiIiIiIiIiIiIiIiIiIiIiKiz0rEek0RwaGYrAjg6p+zbftJ8PbnkOc53r17J+O5fd/LiK5+7OfGYgrQ9n0Py7KQpiniOEYYhsjzXAZ9Bdu2EUURJpMJ6rqWYeRjx/wSoo2WZSGKIkRRhLIssVqtPmt02LKs3/R8Ev3W2bYNz/Pw+PiI/X5vDLHTT4PPOiIiIiIiIiIiIiIiIiIiIiIiIvpXx5gvEREREREREREREREREREREREREX1WnxoG1OO34ucvERrM8xxv3rxBXdfyeCLo+lMdv+97FEWBPM8RhiGiKEJVVfIYlmXBdV0Z+93v9zIofCzY+yWEZ23bRhzHcF0X2+0Wu93us47ntz6fRL91Xdfh/v4eDw8PKMuS9+RHUr/H1L9/DPW75kv43iEiIiIiIiIiIiIiIiIiIiIiIiISGPMlIiIiIiIiIiIiIiIiIiIiIiIioi9G13W/yHHatkXbtvLvnyNU2Pc96rpGURTwfR9xHKMoChRFIY/lOA6CIIBlWdhut+i67tlx/FLB45+LZVkIggBxHKPrOmy3W5Rl+bmHRUSfUV3X+OGHH37Tz7Zfg59q/kQU+KeO3BMRERERERERERERERERERERERH9GtifewBERERERERERERERERERERERERERL9Gtm3LIOFL/BzBwr7vUVUVACAIAkRRBMdxAHyIJLquizAMUZYl0jR9Ucz4tx5VtCwL4/EYo9EIVVVhu92iaZrf/HkR0Y/DZ8CvBwO+RERERERERERERERERERERERE9CVjzJeIiIiIiIiIiIiIiIiIiIiIiIiI6IiXBH37vodlWR8V/33JcbuuQ57nyPMcAOB53kHMNwgChGGIoihQlqVxH7Ztw7a/nP8y6rouTk5O4Ps+drsdsixjNJKI6FeIz2YiIiIiIiIiIiIiIiIiIiIiIiL6ErmfewBERERERERERERERERERERERERERL9Wepz3lwwTdl0H27bRdR2yLIPrurBtG47jwLZtuK6LKIrg+z6yLENVVcaYsBizZVkH49f//lzUMb9kPK7rIkkSNE2D3W6Hpml+FedBREREREREREREREREREREREREREREXz7GfImIiIiIiIiIiIiIiIiIiIiIiIiIBqjBWxGd/SnDsUMhW/X1tm2RZRl830cURRiPx7BtG7ZtI45jWJaF/X6PrusGj9P3PSzLMsZ+h8b1SwVyxdhewrZteJ6HPM+Rpumz501ERL+8X0ssnoiIiIiIiIiIiIiIiIiIiIiIiOjnwJgvEREREREREREREREREREREREREf1m/RzBwGOBXTU8q0d+1d9fOiYR2G3b9ujx27ZFnucYj8eYz+fYbreo6xpJkmC5XKIoioMx6mN+7tw+t5eOpes6ZFmGb7/9Fm3bomkaRiOJiH6F+GwmIiIiIiIiIiIiIiIiIiIiIiKiLxVjvkRERERERERERERERERERERERET0m2EK5+rBQD0eaNs2+r4/GhVU96H+1OO36t860zFErFffTui6zrgf/bNd16GqKlRVhTiOcXJygv1+j7Ztsd1u0bbt0fGZXj8W/f01BRj1sTVNI+PHz11XIiL6fBj0JSIiIiIiIiIiIiIiIiIiIiIioi8RY75ERERERERERERERERERERERERE9JthigKq8Vvx87lgrslzwdtPCRKqY/vYfamftW0bXddhs9ng7u4Oi8UCjuPg4eEB+/3+2dDw0L4/Zjwf46UBR9N26nUciiyL8x06BgOSRES/nKHvOP05TkRERERERERERERERERERERERPQlsPgf44iIiIiIiIiIiIiIiIiIiIiIiOjnZlkW/7Ma/SJECPZj/3/kUAh3KMb73PbP7Vvfzrbto2MWQUTbtuH7PoIgQN/3KMsSZVmi67rBMK6I3+rjGYro/hT/t/THxHxf8nnGIYlIsCwLQRCgaRo0TfO5h/MvR3x/Hfv++6Wf133fv7xuT0RERERERERERERERERERERERPSR3M89ACIiIiIiIiIiIiIiIiIiIiIiIiKin8qPCQYOxXCH3n9JpFfdRg0Dq/vS92MaR9d1ch9VVaGua/R9j67r5LZ6TNG2beOYTHMktj22jWmMP0Wo0XS+PyYE/HN9joh+fWzbxunpKeI4xt3dHdq2hW3bCIIAAFCWJdq2/cyj/PX7Mc/yoZCv+j4RERERERERERERERERERERERHRl4QxXyIiIiIiIiIiIiIiIiIiIiIiIiL6VXlJJNe0/XPbvCQ2+JLArn48Pfg7ND51H+pYjoVsTWMWYV91HGpYV93ONB6TT43i/hRhXMuyYNv2QXDzuWv6a4pDMg5M9HlYloXpdIokSZCmKZqmgW3bODs7w2KxwGazkYHfL1kURQiCAHmeoyzLT97PpwZ9X/LdSERERERERERERERERERERERERPQlsT/3AIiIiIiIiIiIiIiIiIiIiIiIiIiIdCIUC/z4cOtQYFc9huk4eixXDeaqr6mfU/ep73/oOC8Zmwje9n2PruvkccXvQ+dqOifHcWDbtnFb/XyGxqeOyRQ+No3FtB/b/uX/K+tPHd5lsJKOcV0Xnud9UetEPEccx5HPKs/z4LruL3aelmUhjmP8/ve/h+M42O12aJoGnufh8vISo9EIbdvK+PmXynVdnJ+fI0mSn+RcP/X5qH5vMepLREREREREREREREREREREREREXzrGfImIiIiIiIiIiIiIiIiIiIiIiIjoJ/NTRftModpPPZ4puHvss5ZlPTm++vMlsUM9EjwUyhXhQ/U9dRvbtmUwV8RvRQDXdE4fE+LVjz80ft1z25qCxqY48dB10Lf7NbMsC67rfu5h/Ob82q/rT8V1XVxcXOD09PSLWidBEGA8HiOOYwRBgNPTU3z99deYTCZwHOdnP76ICV9dXSGOY2w2G9R1DcuyEAQBXNfFdrvFfr9H27Y/+3h+Lp7nIQzDo3MahiEuLi5QliWapvlRxxsK0Ju2M2G8l4iIiIiIiIiIiIiIiIiIiIiIiP5VMOZLRERERERERERERERERERERERERL8qLw3mfsr2L43/mj5n+uyx/amBXT1yOBS5VbdzHAee58mYr74vdZzPzYF6HBEqfm67lxgKJb/0c/q5mI7/aw9Dtm37qx/jpzCtz5/CZDLB2dkZxuPxLxJ+/ZziOMbr168RhuEXs0Ycx8FsNsPp6SmiKMJoNMLFxQXiOEbbtoPPlp9S3/cIwxCLxQI3NzfYbrfoug62bWOxWCBJEqRpirIsP+q7BAB838dsNvvsa9N1XSwWC7x69Qrj8VhG3FW+7+Py8hK+76Msy88wyqc+9vubiIiIiIiIiIiIiIiIiIiIiIiI6LeIMV8iIiIiIiIiIiIiIiIiIiIiIiIi+sn8lBE/U0z0U2Ovpv0MHWPo8zrbtuE4DlzXPQjuWpYl/zZFe8Xv+jbqPqMowvn5OebzOXzf/6QQqOm8hkKLQ3Hhl+7/WJR4KIJsChQDQNd1xkjwc+vgc+m67mcL334Oz63fHyMMQ3z11Vf46quvBgOlXwrLsjCZTBBFEbIsQ9u2B+/7vg/f939zcxAEAebzOYIgAADMZjMkSYL1eo08z3+RkKvjOJjP53BdF/f392iaBpZlycBvXdfY7Xao6/qjxuM4DhaLBf7t3/4NURQ9u71t2wfP/p+SZVlo2xZt2xrXiOM4mEwmWCwWeHh4+KRw8dBxnzsf/Tim7zKxLyIiIiIiIiIiIiIiIiIiIiIiIqIvkfu5B0BEREREREREREREREREREREREREv21qlNUUdH0J27bRdd3BPtX9qscR2+ux16Fjijipun+daV/qOZnG4LoukiSBbdtI0xRN0xjHfiwcbIrrWpaF8XiM09NTbLfbwSCiaZzqWEUAUsRxh7Y37dc0JtPfpuju0FiHAr+muTVdf3Ub/edz41DnRPz+U0ZHRdD32Prr+/7JGjStrV8ihvoSlmXB8zy4rouu61DXNdq2HQxCD+1DbOs4Dk5PT7FYLGBZFh4eHp5s8yVxHAez2QxN0yDLsoNzdF0Xr169QtM0eHh4QFEUn3GkL2fbNkajEXzfR1EU8H0fo9EIWZZhtVqhaZpf5FpGUYT5fI48z+XceZ6HxWKByWSC9+/foyiKo898E8/zMJlMMJ1OEYYh9vv90W2DIEDXdSjL8kms+SUcx0EYhjLYXlUVqqpC0zToug77/R5FURijxJ7nYTaboe973N/fo67rjz4+YH6WfgwR8h16/n6p9zcRERERERERERERERERERERERH9a2PMl4iIiIiIiIiIiIiIiIiIiIiIiIh+EnoUUPVczE8P0aqfe2ksdoiIDaoh1+fChWIsQxFb4EOQczKZHMQc9fEOHU/sX0SG1biubdsyElzXtXz/2BhFuHcoPDw0Bn0/pvM2xXeHQrv6NTwWBT4298eOr37WdV14noe2bVFVlYzqmti2Ddu2Pym6OTTO53ieh/F4jCiKkGUZdrvdQfR5aL9iHj9nBNPzPJyfn2M8HiNNU6zXa+x2uyfXfOhv9fwcx8GrV6/w6tUruZ5NceOPdez4n8K2bTiO88lhVHUc4/EYs9kM+/3+SeTWdV2EYQgA2G63v5mYr+d5iOMYjuMAAEajEWzbxnq9/qR47qdwHAfT6RRxHOP6+hpt28KyLCRJgsvLS/R9j81mY7zPnlsbXddhu93K58TQehIh9yAIkOc5yrJ8ss2xtWhZFoIgQJIkcp30fY/dbofNZoP9fo+6rlGWpXyuqftyHAej0Qiz2QxpmiJN02ePecyPeeaY4ujPBeuJiIiIiIiIiIiIiIiIiIiIiIiIfusY8yUiIiIiIiIiIiIiIiIiIiIiIiKiH02Pr+qBv2OBXz1Ea4q3iveHArWWZcmQpCnCq49L/5z+nhofFOPT9yuiu1mWyff189bDtmowUR2DCA5blgXP85AkCaqqkoHM5+LC+vzr56aO4blIsSn66zgO+r5HEARo2xZ1XT8bWX4umKzGlfXj6eclosfqvi3LwnQ6xWQyQZqmWC6Xxn2Z4sTqtTh2DiZD56PuU+wrDEOcn58jSRLc398jz3M0TXMQYBbRUH1fnyuAKcYfhiEWiwUmkwmCIEBd19jtdk+2PfY38CE8OpvN8Lvf/Q6+78sgsBr1fannrtGPmTPHcTAejxHHMZbLJaqq+uT9eZ6H09NTeJ6H7XYrQ9/qOHe7HVzX/UUCuD+VOI4xGo0AfAjauq6Loiiw3W5/sfMIggDT6RRlWWK9XgP48Hw4OzvDbDbDw8MD9vs92raFbdvwfR+u66Ku62evadM0WK/XSNN0MLBsWRZGoxFGoxGapjm4tuJ4cRzDsiykaYqyLJ88ZyaTCU5OTuQzoK5rtG0L13URRRGqqpLRcf0esW0bnudhNpvB933c39+j6zr5PP3YNas+bz9lvR875tB3MREREREREREREREREREREREREdFvHWO+RERERERERERERERERERERERERPSjqfHV52KxAOD7PmzblnFFNTKrh1ePRVePRWQFU1BYj+ma9nEsLAz8M9wo4qSmgO/QcU3zI/4FQYAgCLDZbFDXtTzWUDhVjym+JGRsCtrq8ypCk+rPyWSCvu+xXC4PIrSmudTjkqbrqv5zHOfgpxq7bZrmyb7CMMTZ2Rkmk4kci/gnQrF6WNq01kyxX9M8m+ZS/4weIo6iCK7rPjlPEWy2bRvb7VZe588Z8VU5joMwDNF1HfI8B/Ah3qoyzYP6nhDHMV6/fo0kSfD4+CgjqU3TfPS49OeBKcr8qUQgdTQaoaoqeV0+JY46Ho9xenqKqqqw3++fhG7rusZ6vYZt2yjL8keP/ZfgOA5GoxGCIEDf9/B9H13XYbfboSiKn+w6HGNZFpIkQRzHWK1WMqQeRRHOzs7Q9z0eHh6Q57kc49nZGRzHwWazkfFddR3pkeWqqmRMV31PrH/P8zCdTmWst6oq+X4URTg9PcVisUDTNLi5ucFyuTxY63Ec4+uvv4brulgul9jv91itVuj7HkmSwLIsuK6L8XgMx3FQVRXyPJfjdhwHSZJgPp8jyzJst1v5rHEcB1mWyeelPnfqeerU5+BPeR1/Dc8zIiIiIiIiIiIiIiIiIiIiIiIiop8aY75ERERERERERERERERERERERERE9Mn0MK0paCuirOpnTk9PEQQBbm9vURTFk9Cs6TjqT/09/fNd1z0bZR16TT+moO7Ttm3Ytg3XdVHXNdq2fdE+9Nf0OKllWYjjGACw3+9R17VxftXzNJ2LGqq0bVtup8+jHrI1BSDFvhzHQRRFGI1G2O12KMvyYE5M+xZzpe7Ltm04jgPXdeH7PsIwhOM4CIIAXdcdjLdpGpRliSzLZFxT7COOYzlXIl6pRnP1uVDP8VikWb9GQ1FMcV5DaygMQ0RRhKIo5HmIeYzjGIvFAp7nwbZtrNdrGfz8lHjsTxnMFHMrQtV5nsPzvIO4qSl+bApkB0GAi4sLXF5eoigK7HY7OI6DrutkBPUlcy1+dxzHGGD9FCKaKtabZVlomgZ1XSMMQ3ndPnb/nufJyPTd3d3B801QI8kfs/9jz7SfO6br+z4mkwkAyGffbrczxopN4xH3fN/3aJrG+JnnuK6L6XSKruuwXq9l4HY2m2E2m+Hm5gbr9Vo+N8fjMc7Pz1EUBTabjRy77/vwPA9lWcpgszgnAE8izp7nYTKZoK5rRFGEKIqw3W5RFAW6roPneYiiCEmSYDKZYDweo65rzGYz7HY7ea62bePq6gqLxQJv377FarWScWvxTArDEEmSYDqdwnVdPDw8oKoqee1938fJyQlmsxm+//57uK6LxWKB2WyGoijw/v37g9C6mHvP82TMWMTOhaHgPRERERERERERERERERERERERERGZMeZLRERERERERERERERERERERERERJ/MFETVY5v6do7j4OzsDK7r4u7uTr4vPvdcAFc9hnocEeQ8dmz17+e204+jh3RFiFUPUw5FYtXQrBo9Vl/3fR/T6RRVVcn4q/pPjUKazkFs5zjOwWvq8U3neCziqH6+73uMx2MZqR2iR1hFwDcIAkRRBN/3D/6p8V4RehXnq64Lde6jKALwIXpclqXx+Pp8q+PRz90UAzatx2MhYLG967qYTCYIwxB1XaMoChkIVY8PfAiFuq6LrusG17A6/z+3IAgwm80QBIGcV8uyUNc1bNs+WIcmYoyO42A+n+Prr7+G4zh4fHxEXdcyoirixcA/49jqdRHXXsRXwzCE53koigJ5nj8J7er3lx4WV+fPtm0EQYDxeAzf91GWJdI0xcPDA7Isg+d58jM627YP1od+L4tQs+M42Gw2MiyrP3eG5lDsX8S61TGL98T9fSwk/inU+0aPl6uRWtd14TgO8jyXoVnTs03sy/d9Ga+uqgrr9VpGjl+6pi3LQhAEmM/neHx8RJqm6Psenufh/PwcAHBzc4Msy+QzcD6fw3VdpGkqn1dBEODk5ARxHOP29lauQ8dxcHJygq7r8Pj4eDCvo9EIZ2dnWK1Wcps0TeV6nk6niKIIXddhs9mgLEtEUYQwDBEEgRy/53n4+uuvkec57u7u5FjFM0NsI47XNA1WqxWCIEAQBDKA/urVK8RxDM/z8OrVK8zncziOg7/97W9Pvots28Z8PkcURVitVkjT9Mn1FvT7yfT60LUZCn0zCkxERERERERERERERERERERERERfIsZ8iYiIiIiIiIiIiIiIiIiIiIiIiOhHMwVh1bijGon0fR9JkmCz2aCqKvm6Gjg1BVzV1/VgpCkgqYYoReBQHYcp5iuiip7noa7rg5ipGvG0bRthGD7ZrxojfS6iq3McB5PJBLPZDHd3d8bYpxoq1SOiegxZjFkEUVXHosmmQKr6u+M48H3/6PbquYtwahzH8vyCIEBRFNhsNthut8iyDFVVoW1bGbVVx6r+tCwLYRgijmP0fY+iKA7WkWnOxZjUQKk+D+pa04Ow6vH189TDv7ZtI0kSGf0U76n7K8sSq9UKXdehLMuDcKg+f+r+TetrKEqtrgvd0OuO42A2m+Hk5AR5nqNt2yf3wrHrou7f8zxcXV1hNpvh/v4eq9UK4/EYQRBgv9+j6zo4jgPHcRBFEZIkQRAEcg1kWYamaZAkiYzu2raNqqrw/v177Pd7eY6O48gosrjfq6pCXdcHa0kEcZMkwXQ6xXw+B/Ahint9fY3tdovdbgff92XkVb23Pc+D7/twXRd936NpGlRVJcPCIiA7nU6RpimyLJPj0+9PU2DadV2EYYgwDGFZlty3+lwS27mui+VyKefyp6LGtz3Pk9fy4uICk8kE+/0ecRyjKAoZHBef08/Ntm3EcYzpdIp///d/x36/R13XAID7+/uDoPNzxHULwxCr1Ure83Ecy9CuiPCKqPJsNsN+v8dqtUJd13AcB6PRCJeXlzImL843DEO8fv0a6/Uaj4+P8riu6+Lk5ARJkqAoCoxGI7x//x55nsO2bcxmMywWCzRNg/v7e1RVhdFohCiK4Lou4jiG7/tyTNPpFN9++y3yPJfnJb4TJ5OJfLY2TYOiKNB1HU5OTrBYLBAEgQwj73Y7hGGI0WiE+XyO//mf/8Ht7e2TZ1wURbi8vEQQBMiyTMZ8TddLvP4x1OenihFfIiIiIiIiIiIiIiIiIiIiIiIi+pIx5ktEREREREREREREREREREREREREPyk1IqvHUgEgCAIAQJ7nMkLpOA7CMIRt2yiKQgYfATwJqg7FSU1/qz/FsdQor74vz/MwmUwwGo2w2WzQNM3B58S+XNdFFEWD4VtxDHFcU0xYD856nofT01M4joPdbvckknks2KnGM/Wo6tA8mWKipjGq21uWhbIsZTBSD8aaQpCe58m4peM4crv1eo3r62tjwFeci37txdwnSYIoitC2rYwA6+d5LMar/q1Ho/Vw9FDcUt1WXSO2beP09BRxHGO9XsO2bTRNg7Zt5T5FaFa8ps+hiMaGYSi3b9sWjuPAdV0Z+hSfN41NHaO6/sQYRdxYXEsR2r64uEAQBHh8fJTbZ1kmw63qOIeC2LZtYzKZ4OrqCgDw8PCANE1xdnYGx3FkxFXETefzOSaTiVz7wIfAbt/3mEwm6Pse9/f36LoOs9kMruvK+Q6CAOPxGGEYIggCeb77/R7r9RpFUchxuq6L8XiMi4sLeJ6Hvu+R5zkWiwXu7+8RBAHKskSe5wfnJ+73yWQir4lYc/v9Xo45DEO8evUKvu9jvV7D8zxMp1M0TYM8z1GW5ZOArQjnhmGIyWSCk5MTjEYjeZ3X6zX6vofneQA+RF5d18XFxQWKosB+vzeuzx9DDdh2XQff9/H69WsEQSDjs0VRyPGrsXMRHxYh36urK0RRhKIocH9/jyiKMB6PsVwuPyr46vu+nEsRpBXPzCAI8ObNG1RVJdfx2dkZXNfFZrOR4Vyxj/F4jM1mI6+hiFhPJhPc3NwA+Oc9JKLAfd9jNBqhKAp5/PF4jMViAd/35fdFHMeYz+dIkgRlWSIMQ+x2O3RdhyAIkOc50jSVsWQR5I3jGE3TYLfbIQgCNE2Duq7hui6m0yl830dZlvA8D6vVCn/729/QNA1evXqFoijw9u3bg+eBeFZOp1NMp1Ps9/sngepjz+5jz3TTdsJQKJyIiIiIiIiIiIiIiIiIiIiIiIjoS8KYLxERERERERERERERERERERERERH9KHo8VsT8uq57EvUTUUwRI3UcB33fI45jLBYLNE2D+/v7g5gvgCf7MsV6xU9TtFUP45qIMOhkMoHneU/ismrY1XVdhGF4EPsV73mehziOEUURACBNU2RZdrCdKZI7Go0wmUyw2+2Q5/mTkKwaSFbPS2zjOA583weAg3isPnaVHrFVw7bq+MRP13VllFQ9vv4ZNabb9z3KskRZlnBdF57noW1b7HY7eZ3Va6YGifUIrWVZSJIE4/EYruuiLEsURfEkkKqP3RQ4Nq1b/Xz1fZjOVX3Ptm14nofFYoG6rlFVFXzfl/FesS+xntWYtXodR6MRLi8vcXJygjzPURQFyrIEABkUvb29xX6/fxL0HTpXMbbRaATP81BVFdI0lQFUAEiSBOfn53h4eECWZZjNZvA8Twa29dCxiWVZcBwHV1dXGI1GuLm5wWq1Ql3X8p4RkdKzszNcXFygLEtcX1+jKArkeQ7f9/HNN9/g5OQE79+/x/fff48syzCdTuG6rnx2RFGEs7MzGfwVIdkwDGUcVYSQRVz29evXiOMYb968wWq1guM4mM/n8H0fi8UCNzc3B+cqAt8iGptlGbIsg23biKIIs9kMbdvC8zzM53N89dVXaJoGfd/LdSqCyLe3tzJirsapR6MRFosFTk5O5HFFoDjPc9zd3SHPc7mGTk5OcHZ29uQ5qROhYAAymm26XnoMfDqd4k9/+hMAYL1eIwgCuK6LNE3l/K/Xa7kmRXi6bVtst1tYloUoivD69WssFgvc3d3hzZs3MkgtnlMv5TgO4jhGHMdYLpeo6xq2bcv7ZL1eyziw4ziYTqc4Pz/HbrfDfr+XrydJgtlsBsuy5OviO2k6naIoChlmVqPUcRzL++T+/h5VVSEMQywWCyRJIqPR4/EY0+kUURShrmsZ503TVMbS8zyXYx+NRvKe2G63yLIMwId7vO97GUy2LAtZlsnvhd1uh81mg/Pzc/i+j7dv3xqfyWEYykB8lmUHzyDA/HzUX/uUMC+DvkRERERERERERERERERERERERPSlY8yXiIiIiIiIiIiIiIiIiIiIiIiIiH4UNRSrhk5NgVcRWuz7HlVVAYCMacZxjPv7e7RtOxhLFb/r8VI9xqqHBE0RVlNoWAQ01+s18jyXQUc1gtn3vQyjpml68J7rujg9PcX5+bkMOi6XS3RdJ/enjwv4ENEU0cXHx8eDSKdlWfB9H5Zloa5rGW8VMVg10np+fo6mabBcLrHdbmXkdijYq8c+TQFGNTLrui42m42MTIr5MAVyBTFm13WRJAmCIMDj4yPSNH1yTYeiw+qY5/M5RqMRuq5DURSoqkpeAxHMNEUq1XnQz02dF3G9xc9jMV+xFlRRFGEymeAf//gHAMhwtQhpHgsre56H6XSK169fw3VdrNdrZFmG//zP/0TTNPj2229RVRWurq6QpinSNDWGfIci2tPpFFdXV6iqCuv1Gq7r4vHxUW5/enqKJEnw/fffo65rBEGAtm1lRFQPSuuxac/zAHwI0X7zzTdo2xb39/coyxKe5yEIAnm9RMC773t8++23MngtgrJBEOD29hZ//vOf0TQNHMdB27a4vr5G13WYTCb46quv4Hkebm5usF6v0bYtoiiS918cx1itVnBdF3Ec49WrV5jP5/juu+/w8PCAruvg+z7yPMdsNoPv+7i7u5PX2PM8nJyc4OLiAo7jYLlc4vHxEWVZwvd9GW4WsdvFYgHXdbFcLnF9fS3DrUmS4PT0FHVd4/b2VoZhxf6vrq4Qx7G8Z0ejEd68eSMDr9vtVq5xcR3v7++x3+/l3OtrS8RckyRB3/fYbrfymateMz3g7XkeZrOZXAer1QonJydYLpcoyxJJkmC/3+Pu7k7GfSeTCcbjMfb7vXzt9PQUi8UCDw8PuLm5QVmWiOMYvu8/GyHWifvCdV2sVisZ5z05OUEQBPjuu+9kgDmOY1xdXcG2bTw+PspnVRiGODk5QRzH2O/32O12cl3FcYwwDOW1VdfzbDZDGIbo+x5ZliFNU7nGx+MxmqZB0zSYTCaIoghd12G9XiMMQ9R1jSzL0DSNDB5bloWTkxO5nsU1LssSXddhPB4jiiKUZYn9fi/D10EQIEkS1HUtA8unp6fI81zOiR6MH41GSJJEriExR0P3sPrasWC3iem5yqAvERERERERERERERERERERERERfakY8yUiIiIiIiIiIiIiIiIiIiIiIiKiTyLCp8A/Y6siAGiK6QIfAomTyQRVVaFpGhmcnM1mMrCoxnzFfkTgU31ND7Oq7w0FhtXt1CiwbduI4xhnZ2do2xZpmh6MQ41eAh/Ctp7nHcR1XdfFdDpFkiRI0xTr9RpRFGE0GiFNU5Rl+WReHMeRkcezszPsdjvsdruDOG2SJJjP5zLAmue5/LyItCZJgvPzc3zzzTcoigKu66IoioNr4TgOuq5D27awbVt+djabwbZteeyqqp4EZ0WIGQA2m81BJFjVdZ1xrkU0dTabySCwHhY1xZbVcdi2LdeKen5iHYlgq+M4Mo6p7kuNMqvXXr2mjuPIa922rfynn6spVinmaLFYyGBpHMdI01SGfE2fV8PKs9kM//mf/4n379/jhx9+QNd1mM/nCIIA79+/x83NDXzfx9nZGXzfl/eFHmRV/3ZdFycnJzg5OYFlWTI2muc5RqORvOZRFOHi4gL7/R5ZliGOYwCQ41fvQfWY4udkMpFB6qurK5yfn8uo6fn5OcIwxHg8RlmWCMNQznVZlqiq6mCeR6MRHMfB27dv5etd12G73QL4EB3+3e9+h77v8cMPPxysSXE/imh4EASwLAtnZ2dYLBZ4//49bm9vZRDWsiyUZSmfQWIteJ6H09NTXFxcoOs63N3dYbVaoaoquRan0ylGo5EMulqWheVyiXfv3uH+/v7gfhAR6oeHBxnRns/nMtx8e3uL7XaLJElkZHm9XqPrOrluHcfBbDaD53m4vr6W95Aa+xYBasdxZPhXDV+b7jX1PmmaBrvdDn//+9+xXC5lhLksyycB2r7vZUjccRwURSHnZj6foygK3N7eyqhxFEUAIJ/zLyH2NxqNUBSFjOmK58l6vcbd3R2Af8bUJ5MJ7u/vsd1u5fNuPB7j9PQUlmVhvV7LmHgQBJjP5wCA9Xp9EK32fR+TyQSe5yHLMuz3e3RdJ59DYl6SJJHzlqapfJaIQLU4zsnJCZIkwWazwXK5xG63kxFfcdzRaATf97HdbpHnOeq6huM4CMMQvu/L2PvV1RU8z8O7d++efLeI76P5fA7HcXB7eyvnXA2Ui22PBcbV7YYMfQ/r4X0iIiIiIiIiIiIiIiIiIiIiIiKiLwVjvkRERERERERERERERERERERERET0SdSQp+k902u+78uYYV3XAIAkSRDHMR4eHgYDj2o4tOs6uK6LMAwRBAGapkFZlnJ/Inypx1rVcYioqeB5HiaTCWazmYwj6tur5yqOX1WVjDVOp1MsFgsURYH1eo26rnFxcYH5fI79fi/DkmIfURRhPp8jDEMURYEoivD999/LY9u2jZOTE3z11Vdomgbv37+X74lI63g8xnQ6RVEUuLu7w2azwXw+x2QygW3bMkoZhiGiKEJZliiKQp7vYrHAV199hbqu8f79e/R9j81mI4Oo6hx2XYfVaoU8zwevuxqGVN8XMU0R38yyzHiddSL4GoYh4jjGbDbDycmJDMwWRYGmaWSwVszVmzdvUJblQXzWRER0XdfFaDTCdDqF7/vIsgzL5RJFUcjtfN+XQeOmaQ7Wqginnp2d4T//8z9h2za+/vpreJ6Hx8dHlGWJtm0PIsjq2MT4//3f/x15nuPdu3domgZhGOLVq1d4fHzEfr+XYVJxDUSMVj8n8XoURVgsFri8vMR2u8UPP/wgxx4EAaIokttOp1PEcYz7+3u0bYskSVDXNdI0lWtcv77in+/7uLi4wH/8x3/A930sFgsZfRVR1PF4DM/zkCQJptOpXBee58l7Sbwm/hbzpZ7bbDbDH/7wBzRNg3/84x8H4+u6Tj4PRFhWHPv8/Bzb7RbX19cH1048B5qmwXK5lM+X09NTXF1doWka3NzcYL1eHwTIRZi1qirUdQ3XdWHbNh4fH7FcLuW24pmlrjfP83BycoKrqysAwPX1NVarFYAPkdqmaeB5HgAcjDWKIkwmE6RpijRN5Vydnp6ibduDNdu2LXa7HYqigGVZT+LZKjW42ratXG8iiiyCx5PJBHmeY7fbychsFEUykLvf72XYOY5jvH//HlmWwbIsxHGMMAyx3W7lWn4Jx3GQJAmiKMLt7a18zk+nUwRBgDdv3hxEjc/OzrDf73F3dyfnIggCLBYLxHGMu7s7LJdLGWVOkgRnZ2e4v78/CKUDQBzHmE6ncF0X+/0eVVVhPB4jiiJcXl7KecmyDNvtFmmaou97RFGEtm1lBFuEhMMwlPMrorzqPSVizeIeF98v4l4X3yOO48jn4GazOdiHuJYior1arbBarVDXtTGYDuAgyq/Hd8V7x67X0PcBERERERERERERERERERERERER0ZeKMV8iIiIiIiIiIiIiIiIiIiIiIiIi+iR6vO9Y0E+EZcMwxGg0kgFO8ZqIhqr7Ne1Pjdj+6U9/wmKxwPv373Fzc4PHx0cZHBQRWz1Iq+5P/d3zPIzHYziOg8fHRxniFOPQt3ddF57noaoqtG0L3/fx+vVr9H2Ph4cHpGkq46IiViqCo57nYT6fYzabYb/f4/r6GpeXl3AcB+v1Gm3bwnVdXFxc4I9//CO22y3ev3+PzWYjo6CO4xzEg5fLpQywTiYTNE2DJEnguq4MRIqxTiYTGa21LAt3d3eoqkpGP036vkfTNFiv1wcR5CFqVFJcYxFwFWPVr7X+eXGtz87O8M033yCKInieB9d1kec50jRFnuewLAtBEGAymWA8HmO328F13SfX23QtbdtGHMc4OzvD2dkZTk5O4Lou7u/vZUzWdV1MJhNMJhN5/PV6jZubGxnuXCwWmE6nmEwmCMMQt7e3aJoGTdMgiiL87ne/w8PDAx4eHmSQWcyj67oyUNv3Pb799lsZh53NZjJaqwZ8RcRYnItt2zJ83LYtuq6TMc9Xr15hu93i7du38hqLecjzHG3byjXZdR32+70MZW82Gxk4VedO/K4GRMU6FBHjN2/e4C9/+Qs2m428ltvtFsvlEqvVSr4WBAHiOD6IXVdVhTRNEYYh9vu9vH5hGOIPf/gDLMvCd999h/1+L2OneiBZvC5irW3b4vr6+iDULc4jjmPkeS5jtLPZDFdXV6iqCjc3N9hut3K+HcdBHMfwPE+eWxAEODk5Qdd1WC6XMoQrxuK6Lnzfx2azAQBMJhN89dVX6LpOhoJFqFnEakUcWJyX53k4Pz9H0zS4u7tD0zQyonx6evokUtz3PaqqQlVVB9dMv9/0e0/c6yLCPJlMUBSFDFmL+06s/fF4DNu2kaYp2rZFGIZYLBawbRu73U7O73Q6RZ7nMnSuj2mI67pIkkTGobuuk9Hetm1lzNayLFxeXuLk5AR//etfZVjXsiwZEs/zHLe3tzImLq5bkiT4xz/+cTAu27ZxenqKyWQiA8B5nmM6ncqg78PDA25ubrDZbFBVlbw/RSC873uMRiOcnZ2haRq8ffsW5+fnKIpCXquu62RAOggCTKdTbDYb7HY7NE0j11sQBDLoPpvNMJlM8D//8z8HwXH1eTCfzzEajfC3v/0NWZbJ54L6varfLyr1PT0UbKK+99JrS0RERERERERERERERERERERERPRbxZgvEREREREREREREREREREREREREX0SPQpoCuXatg3gQwDT8zxMJhOMRiMZKvV9H//2b/8mY5kioqgeQ40C2raNJEnwhz/8Af/7f/9vGQc9OzvDarV6cnxTDFgfv+M4GI1GGI1GMl6qxgv1qLDjOAiCQIZQHcfB2dkZwjDEzc2NjBI7joO+72Xg0vd9Gf31fR/v3r3DarVCFEWYz+e4v79HXdcy/vrVV1/h5uYG33//vQxTijmYTCa4uLiAbdt48+YNuq5DkiQ4PT2FZVnY7XY4Pz9HVVUIggBpmuLx8RFBEODVq1cAgNvbW+z3+4NIadu2g1Hmtm0PArLH4o3qPhzHkfHLNE1lnFW/tirbtuH7PhaLBf70pz/Btm3c3NxgOp1iNpvBsiw0TYO6rmU4U5z3fr+XAVNx/CiKYFmWjL+q43r16hWm0ynqusbd3R0syzqI6Y7HYyRJIiOro9EItm1jvV7Dtm1cXl5iMplgvV7D8zzc3d3h8fERVVVht9vJCOtisUDTNFgulwdrfTQayQDu3/72N2y3WwD/DExnWYbVaiXjn0EQoG1b1HUt46Hj8Rij0Qh1XWO9XgMATk9PcXl5ifV6je+++05GedX7IE1T1HWNOI4xmUyQZRksy5Jx291uJ49jWhPiGjZNg8fHR7iui6+++gr7/R673U4GkUejEU5OTtC2LXa7nYy8imiw53nwPE8GR0V0WH0WiFhrEAT47rvvsNvtZAjVtH5E4HgymSCOY7x9+1YeV+zPsixMp1OMRiPc39+jaRpMp1P8/ve/R9M0MtSqxqejKEIcxyjLEtvtFpPJBGdnZ0iSBNfX13h8fJRzJq6xOL+yLOG6Lq6uruC6Lq6vr7Fer+V91bYtiqJAlmVPYr6np6eYz+f4xz/+gbIsZRj98vISWZbJe1kPrx4LrKtRZpUIj0dRhCiKsF6vZaw4z3MZEj45OcF8Psfj46OMJAdBgPl8jru7O+R5LmO2WZZhs9k8iVkfY1kWwjCUazNNU3nPXV5e4rvvvpMx2ziO8ac//QkAsNls5P0aRRG++uorRFGEb7/9Vt4fIpw7n89hWZaMAovjuq6L169fyzC1CC43TYMwDGXQW0TkxWfF88b3fdi2Lc99uVzC8zxMp1MEQXBw/mK9zmYzAMB6vZb3QBAECMMQVVVhu93Ctm1EUYS2bQ++F8Q6Ec+UP/7xj9hsNjKSPfS8Vb/n9P2YgviM9BIREREREREREREREREREREREREx5ktEREREREREREREREREREREREREn6jrOhnbFJE/EQIU0cLLy0t4nifDkK9fv5bx0rIs4XkeRqMR3rx5I4OZgogJiuihbduI4xhnZ2cYj8f47//+b/zjH/9AEASYzWbwPA9N0xyMCYAxaqnGSUXU0XVdPDw8oGkaGbPUI8XAP8PEVVUBAGazGc7Pz2X8VAQ1XddFEASwLAtd1yGOY8znc7iuix9++AHb7RZ93yNJEjiOg/v7e4zHY/zhD3/AeDzGd999h3fv3qEoioNzEHMQxzFubm7QNA3G47EM+d7d3SFJEkwmEzRNg/V6jdvbW3Rdh8Vigclkgjdv3mC73crgJQA5bj3Sq86Z+pop6GuKKPu+j9FoBMdxZOTWFIlUP2PbNl6/fo2zszO8ffsWt7e3SJIEo9EIVVUhz3OUZYm2beF5HuI4RhAEaJoGRVHIawMA4/EYX331FbbbLbIsk6+PRiP87ne/kyHXu7s7FEUhA6ae52E+n8P3fTw8PGC1WsG2bex2O3nuURRhNBrh8fER2+0WJycn8ji73Q5pmsJxHDn2MAzlOhXX8tWrVwjDEN999x0eHx8P5sCyLKRpehBZjuP4IKw8Go3w+vVrGZMWcd/Xr1+jKApcX1/L4Km4HkmSoCzLgwBrFEVYrVYy/rxarQ7C1mINmCLOfd+jLEvMZjPYto2//vWvWK1WSJJE3lthGOLu7k7eIyJgKs5nuVyirmtYloWyLGUM1bZtdF0Hz/Nk4Hq1Wsm1qwbAu66D4zjwfV8e8/T0FKvVCg8PD/JZIs4pDEMZt07TFJ7n4ZtvvkEQBHj37t2TyHgcx1gsFgiCALe3tyiKAuPxGLPZDNvtFjc3N8jzXD4bxboOw1DO+Xg8xmKxwNu3b/H4+CivY9d1ch73+z0mk4mca3HuaZoiyzL5vKvrGkmS4LvvvpPrRA+p64aup5hn9dqKAHOWZbi8vJTjE+tOBNm7rpNhbbGW1us14jjG69evsd1u8fDwINfhUGBYp4Zx37x5g77vMZ/PZXw7yzL0fQ/f92Xc/Pb2Vt7n4t44OzvDer2WIV/f9zEejzEej9F1Hd6/fy/Xughkn5yc4OTkBDc3N/J7QVynuq5RFMXBM1M88+M4RhzHqKoKruvC9325VoAPod4kSeR8O44DAPKZvdlskKYpuq6D67py2/V6jbquEYYhwjCUzzn1Grqui5OTE/zHf/wHptMp/t//+38HYXN9Dejfler1H4r26t+r+utERERERERERERERERERERERERE/woY8yUiIiIiIiIiIiIiIiIiIiIiIiKiT6LHbsXvjuPIcG0Yhthut6iqCr7vw/M8vHv3Dv/1X/+Fpmnw+vVruK6L9XotA7oiEqhGfG3bRhRFWCwWmE6neP/+vQzUXl5eytCiKSZrCtIKYr9xHKMsS2y32ydRYX0/ItJbVZWMPrqui/v7+4P4qeu6MvqbZRmSJEEQBHh4eMBut5P7ms/nsG0b4/EY//Zv/4a2bfHXv/4Vd3d3B7Fd4EPU8+TkBKPRCPv9Hmmayuhk0zS4u7tD3/d49eoVJpMJ7u7uZKRWRHXbtkVVVTIeqs6NHtg1nb8p5jgUALYsC0mSII5j5HmOzWYjw6i2bcP3fTiOg77vUVUV2raFbdv4wx/+gFevXuHdu3d49+4dqqrC5eUlTk9PAeAg2BuGISaTCXzfR1EU2O/3MmBp2zYuLi5wdnaGh4eHgzjq69evMR6P5VpSo8me52GxWCBJEtzd3eHh4QFlWcK2bXiehzAM5Vj7vpdR1yiK8Pj4iCzLkGUZ2rZF27bY7/eIogiu68K2bTiOI9fOfD7H+/fvD4KhAFDXNfI8l2FSEYUNwxCPj48oigK2bePs7Azz+Ryr1Qrb7Rae5+Hi4gK+7+Pbb79Fmqby847j4PLyEq7r4u3bt3INzGYzBEEAACjLEpvNBvv9XsZh1Wivfo3Fevd9H19//TW22y3+/ve/y3DtH//4R+x2O7Rti91uh6Io0HUdgiCQMdEwDA/u/bIs0TQNZrMZbm9vZTh5PB7jz3/+M+q6PhiLGqSdTCbyngrDELZt4/r6WkZoxWd838d8Psfl5SX+9re/Ic9zXFxc4OrqCv/3//5fGdoV85YkCc7OzmS4d7/fy2OEYYj//u//xsPDw0F4WQR9R6ORvKYXFxfo+x673U6ehx5RbZoGbdsijmP0fY/z83PEcYxvv/0Wl5eXMlp7eXkJADL+OhTwFc8jx3EO5kHMmdhGvceDIEAYhnh4eJBBcvE8Go/HGI1Gco3ati33H0URuq7DbDbDdDrFfr/H+/fvD+4vdQ0di8YGQSCfzXVdYzweI0kS7Pd7efzJZILFYoHRaISu65BlGXzfl+tgMpmgKAqs12uUZYkwDGWsW8S/7+7uDp6H4t7s+x7v379HWZYH60x8R4hotIjyiu+Spmmw3+/huq787hOfE98Fvu8jz3MAH74HRqMRbNvGcrk8iBQHQYCiKGT8XcSqxfePeJaGYYiTkxOcnp4iDEOs12usVis5VrEWh6jXQTzX1PeGvheIiIiIiIiIiIiIiIiIiIiIiIiI/hUx5ktEREREREREREREREREREREREREn0SP+QoizLpcLvHw8CBjomdnZ6jrGqvVCnmew/M8JEmCsixRluVBMFAPPDqOg/F4jNlshrIscX9/j6IokCQJwjCUcVo9EGkK+apxUhHpFPHDNE2NEUM1cimCrnmeIwxDzGYzFEWB3W4nY6wiVAsAm80GRVFgNpvBsixkWSbjneLzs9kMr1+/xnq9xl/+8hfc3d3JCKM6njiOMZlM0LYt8jzHZDLBaDRCmqZyTk5PTzGfz9F1He7u7uT8N02DsizheR48z4PjOGia5smcq+cqzl+PhOrzYloXYh5GoxGCIMBqtQIAGSOO4xhBEMCyLBRFgdVqhcfHR1xcXODrr7/Gt99+i/fv38sY5nw+RxRFuLu7w3q9RlVV8DwP4/FYxozzPJeBTLFuZrMZsizDbreT57hYLPD69Wt8//33T0K+bdtiNBphMplgtVrh9vZWBlBt28Z8Ppex1a7rUJalDL12XSdDr2qYt2kaGS4W1zWOYxljFeFmPWbted7Ba/P5HKPRCDc3N2jbFvP5HPP5HG3bYrvdoq5rXF5e4uzsDO/evcNqtZKRV8uysFgs8PXXX+P/+//+P1RVJfd7dnYmI6yr1UreB/p1PhZenc/nuLi4wP/5P/9HXmsxR9PpFI7joCiKg4Bt13Uyeh1FkQwIO46D7XaLq6urg6i3CMqqIWVxfBERXywWMrDq+z4eHx+x3+8Ptg2CQK7Druuw2WwQBAFev36N+/t7PDw8yHCtCMNGUYTRaISmabBer9H3PZIkwXQ6xf39vbwmemRYxH7X6zVc10Ucx9hsNsjz/GCNqPeOOJ/xeAzHcfD69Wv0fY+LiwsZx/U8D6enpwdzarqXReh1MpnAsqyDaLQpfi7uGxHR3Ww28H0ffd+jbVuMx2O4roumabDb7WRkejwew7IszGYzRFGEyWSC7777Toanh+LfQ0SA2HVdub5c18Vms5HHF+tju91ivV7L+zaOY3RdhzRNZei46zqMRiN5D9Z1jTAMD0K5asx7MplgvV5juVwePIubpkFd1+i6DuPxGG3boigKGXnvug7r9Rp5niOOY1RVhSRJkGWZ/J4T60o8q8Tz9fHxEdvtVkakPc+TgWJxTzZNg67rEEURXr16hf1+jziOEYYh0jTF4+MjkiTBdrtFnufy/jE9q9V73BTHVrf5mGtn2g8RERERERERERERERERERERERHRl4QxXyIiIiIiIiIiIiIiIiIiIiIiIiL6ZGqsTw1+lmUpY5h938uIZd/3WK/XaNsWURQhDENsNpuDqKyIWYqAIvAhfDidTuH7PpbLJcqyhG3bSJIEnudhtVoZw4F6cFaPkrquizAM0fe9DLA+FyF0HAdhGGK32yGKIti2je12i7IsD8KJIiYp5iFJEuz3e1RVJfc1Ho9xdnYG13VRliX+8Y9/4OHh4eDcBdu2MZ1OZUByNpuhbVs8Pj5iuVzK6O35+Tlmsxn++te/4vHxUY5JhFqn0ymm0ymyLEPTNE+CoioxF+p5iYiqGkVW51adb9/3MR6PZZT25OQEtm2jqirc39+jqiq0bSsDlZ7n4X/9r/+F29tb3N7eyrkaj8eYTCa4u7vDDz/8gN1uJ2OaFxcXMnC7Xq/lfAMfYsknJyf4r//6LxRFIefx97//Paqqws3NzUFgU4z55OQErus+CZFGUYSzszP8/e9/R9M0sG0bq9UK33zzDV6/fo2///3v2Gw2B2MQxxThYhGrnUwmmE6n+P7775Hn+ZNotLhvgiBAlmVwHAeLxQL7/R5FUcgQtOu6yLLs4LW2bXFzcyP36bouJpMJXr9+jbqu5RqzLAtJkuDk5ARVVWGz2SDLsoMoLWAOeKqvhWGIr7/+GqvVCu/fv0fXdXLdbDYbfPPNN8jzXEZnxTxkWYY8zzEej5EkCR4eHuC6LmzblvMexzHyPIfjOHIeTPHZ0WiE+XwOANjv95hMJuj7Hvf39wdB4CAIMJlMZCh1s9lgv99jNBrB9328efMGTdPAcRz4vo84juE4jryHyrJElmXwfV+Ga6+vr58EyUVcfDQaIQxD3N/fy8iwCPnqoVQA8vwdx4Ft24jjGHEcAwDyPMfNzQ3qupb31X6/l/NtIsK0p6enyLLMeExxPcW1EXHloihQVZV8Rop4+Ha7lfeg7/tIkgRXV1cAPjznROxWnKdt2/K5oc/RMSJEPB6PAQDv37/HcrlEHMfY7XawbRv39/fY7XaYTqcy8P7w8CDv7YuLCxn89jwP6/Uau90OjuMgiiKkaSrvP/V7oes6vHnzRoag1aDvbrdDGIZIkgTj8ViGsMuyxG63k8H2PM9RFAUuLy8RxzGur6/huu7B/SUCveI7TKwLz/Pgui7quj54PpRlidvbWwDAZDJBkiTYbDa4vr5GlmV49eoVAMjvSTHH4vN6mN30ffcpEd6huDsRERERERERERERERERERERERHRl4gxXyIiIiIiIiIiIiIiIiIiIiIiIiL6aKZwq6BGEcX7rutiNBqhaRpkWQbgQxg1jmPc3NzIsOhoNILneciyTIYIbduW2zZNI+OVQRBgNBphv99ju90+CeDqYxPxWTUg6fs+wjCUYVE1SOi6LhzHQVVVBzFbEUet6xqTyQRN0yBN0ychYBHjrOtaHme5XMp4rud5+N3vfgfXdVEUBbbbLe7u7mTwVD0HES4+OzvD6ekpttstttstHh4eZAzZsizEcYzf//73sG0b79+/R1mWB3Pw+PiI0WiEs7MzNE2Dtm2x3+/l/IvrpweQTXHHoe3VcKoIXtZ1Lc+9bVvUdS3HLIjIruM4+Pbbb2V817IsGV/94Ycf8PDwIN+bTCaYzWaoqgrr9VpGL8X1WywWKMtSRo0ty8JiscB0OsWf//xnFEVxsB7EPLuuizS9p422AAD4/0lEQVRNsd/v5RiCIMB8PpdzJs5dhFV938fDwwOqqjqYL9d1EUUR6rqWceIwDDEajVAUhYx4mgK1Yu0AHyKpYRji8fFR3k9ivquqQtM0iKIIruvKubBtWwZpJ5OJfE+Mz/M8nJ2dyftQhIi7rjuYF3VsprU5m82wWCzw7t07pGkqt23bFlVVwXGcg2C3eC/Pc2y3WyRJgvl8jru7O3lNxD0ZRZGM+TZNgziOsVqt5Jr1fR+j0Qij0Qht22K9XsP3fXieh7quURQFbNuW1zCOY7Rti6IoMJvNZAj79PQUdV2jrmtEUQTf9+X9m+c5fN9H13UoikKGqc/Pz5HnOe7u7p6Eabuug+M4SJIEbdtis9nAtm15bmIO1TmxbRue5yFJEgDAZrNB27b4y1/+Ip8RIhTteR4AIE3Tg2effq+KMQRBgN1uZ4z56n+HYQjbtuWzQdy3lmUhyzIsl0vsdju5fVEUMoBbFAXKskQQBLi4uJDnLyLv4tluCpbrYxL3tbhPVqsV6rrGbrfDP/7xDxmM7vsej4+P8v4X94NlWbi7u0NZlnAcB/v9HrvdTsaaq6qSzyI1Ztw0Dd69e4fdbncQXxfjKooCDw8PyLJMnrdYa+r3Rdu22O12mM/nWCwWcF0XruvKdS6OKeZNrBv1/hDzJo4t1lKWZTKQrMbQz8/PkaYpttutnBv1ntXvX7HuxHejKbD8XOBebKPH8omIiIiIiIiIiIiIiIiIiIiIiIi+VIz5EhEREREREREREREREREREREREdFHE7E+ET8UET81eiviqbZty/jozc2NjBaKwG2WZTJ8OR6PAUAGf8V+oyhCFEXY7XYoigKO48jAqwgjqmMTY1KjgmqAWGwTxzFGoxF2u92TEOzJyQnG4zEeHh6w3W7lfkTwsK5rhGGIpmkOorCWZcHzPIxGI7iui6qqZGS1qioZxZzP57i6usLNzQ1msxnevXsng8L6eJMkwddff41Xr16hrmv88MMPMlIpgo2u62I6nWKxWOC77747iKqKeWmaBre3t3BdF+PxWEYg0zSV+1Gvrzi+GmkU11UN0A4FIpMkQd/3SNMUWZYdhIrVz1iWhel0iqurK3z77bfy+gPAaDTC+fk51us1VquVnGtx7YAPkWIRNRb7HI/HODs7w/39vYxhhmGIq6srGQfVw5WO4yAIAgAfIql1XcvrOZlMMBqN5HmI48RxjCRJ8Pj4iN1u92SfYp0XRSEDnGEYwnEcrNfrg2CwOo7ZbIYwDLFareR6F2tvNBrBcRy5vQgIB0EAx3FQFAWiKILnefB9X8ZUAWC328nI73g8lsHj9XotI6vqWjBFsdXfPc/D6ekpLMvC/f29XB/Ah3tFhGBd15XnLcYjAqgiRHx6eorHx0cZRK2qCq7ryoCqiO7meY6iKGTQu+s6ZFkmo9qu66JtWxkCF/Fv3/dRliWKosD5+TkAYL1ey2vW9z3G4zE8z5PhcRESdt1//rdrEfJ1HAc3NzdI03QwYBrHMcqylPdq13UIw1AGnkXU1nEcRFGEJElg27YMtvZ9L3+Kz4vnq4gSq9dL57ougiCAZVkoy/Jgu6H7NgxDVFWFLMvkMdbrNTabDfb7/UFAeLfbIcsyOSbHcZCmKWazGfq+l3Oe5znyPJfPwOeCr33fI89zXF9fy8CuOGbbtgcxYBFJFve/Kk1T5Hl+MGfip7iP9bGIcYrno5gjNb6b5znKsjx43RRJFucg9lXXtYx+69dB3YdYfyLgq56rWDe2bR8cz/M8zGYz/PnPf5bPPLGN/p2tHsv0Uw/262vl2PV7SfyXiIiIiIiIiIiIiIiIiIiIiIiI6LeMMV8iIiIiIiIiIiIiIiIiIiIiIiIi+mRq6HXobxF1DYIAq9VKBhSDIJBRxK7rkCQJJpMJ1uv1QexRBFZ930fTNOj7HkmSYDweY71ey5CmKVKpBgjVQKFt2zK4a1mWjID2fQ/HcTAajRCGIXzfx3w+l7Fby7LgOI4MJQKQcUrB8zxMp1P4vo80TWFZFqIoQlVVyPMcwIfA51dffSWDmWEY4vHxcTCgKsKmm80Gm80GDw8PB/Fhy7IQhiEWiwW6rsPt7a0xbCnikre3t+j7HqPRSI5fDejqc6hHI/VtTO8HQSDPe7vdyvnVPy8itPP5HH3f4+7uTs5nEAS4urrCaDTC27dvZUTUtm3EcYwwDJGmKZbLJYqiOIg0X15eIggCfP/99zLIOZlMMJ1O8fDwILfXiTCoCNB6nocoiuD7vgwii4htFEWYz+do2xa3t7cH60Jcl/F4DN/3sd/vZczX8zy57vT4r+u6mM1mSJIE9/f3SNMUvu9jMpmgKAqEYYi2bbHb7eB5Huq6PphbsX4vLy9RFAX2+z3yPMdkMpHzPRqNEMcxHMdBnufYbrcyLqqvveeinCLUfX9/j+Vy+eQ+FBFXx3Fwenoq49FVVSEIAnk+eZ7j7OwMV1dXuLu7g23bcBxHPi/2+z3evXuHy8tLfPXVVyjLEk3TYLfbYbfbyb8ByAjvdDrF6ekpoiiScVSxzmezGaqqwn6/R9d12O12iOMYTdOgLEvs93sZNxYxXBF6FlarFZbLpXxeiX/qHIp7vCgKtG2L9XqN09NTXF1dYTaboW1bOI6Dvu9RVRWKokBRFAehbjGXaqR8v9/j7du3B8fX157rujKErsZh1W3U/YoYr+d52O/3qOsaXdehLEvc3d3BsqyDkK74W11/tm3LY4l5qKoKVVUdxLxfQjz3xLFM1NfV8K4gxqsGpvXr1HUdbNuWf4vwrvo5U7DXFMQVcynmtm1bbDYblGUp17P+nND3Lz4r5lW/RuK81DmwLEvGzUVUXA1rD323mNasaW5NvzPYS0RERERERERERERERERERERERP+qGPMlIiIiIiIiIiIiIiIiIiIiIiIioo8mIr16ANAU8/V9H0mSyMAm8CEA6rouiqIAAPi+j+l0iiAIngQTxbYiXjiZTDAajdA0Ddbr9UHwcogaSBR834fv+6jrWoZx1XBrnuewLAtJksD3fRRFAcdxkCQJmqaR0VcREBXRWxEBzrIMu90Ovu8jjmMZlRSx39FohMfHR1iWJfcxFKys61qGWff7/UE8U4Rlx+MxTk5OsN/vsdvtDuKunuehaRpUVSUDqJ7n4fz8XEZixXv6vJmuvfh5LAIZBAGCIJBj12OY6n5c10WSJDI8K677aDTCeDxGnufY7/cyrul5HiaTCTzPw3a7xW63k5Fd3/cxm81wdnaGsizl/kTs0rIsOYemddI0Ddq2he/7ODk5Qdu26LoORVFgNBrBdV0sFgsAHyKgIny6Wq0OQqEA5HUBPgRm67qWQWhx3dV5FIFpz/NkqLptW8RxjNFohK7rUNc1ttutjPyGYQjHceC6Luq6RpqmcF0XVVVhvV7LgKsItZ6cnCCOY1RVhd1uh6qq5JoX10vETU2BbPW6idfX6zW22y2yLJNzIN7Psgzr9VrO6/n5ubz3HcdBFEVYr9dYLpdIkgRXV1d49eoVmqbB+/fvsd/vAXyIl4oIs+M4aJoGdV2jrmtUVXUQci2KAg8PD2iaRs6LCP6KeXAcB5vNRgZmt9utPHcRSFbDqUVRIAgClGUpA9XiXtODzGIcXdfh7u7u4BmzXC7RdR1835drQcSM8zyX5zT0XBOvpWkqt9cDveKneH65rovVaiXD6UMBdvGzLEsZONfjsepn1XtajdqK9aZGjk1z9DGGnvFqkFf9W4/s6sHxlwTgh46rn7ceWtbHpMaWX/p9pc+n6VxE2Fw8Q0ejEbbbLYqiePIsEtd96LxMz2ZT+Fn9W2yjf+8PzRsRERERERERERERERERERERERHRl4IxXyIiIiIiIiIiIiIiIiIiIiIiIiL6aHrQUcT89CikbdvwfR+e52G1WqGuawAfIqci6CkiqycnJ3Ac58mxRLiw73uEYYjZbIa+7/Hw8ICqquQ2pjCjiRiv7/uwbRtVVaGqKriuiziOEYYhyrJEURQyuBlFEYqigOu6CIJAxmlFzFeEfOM4RpIkaNsW+/0eVVUhjmP4vo+qqtD3PYIgwGw2Q9d1WC6XiOMYAOSxTGMXYUZxLHWuHcdBEATwfR+u68ogrghJivCtiLfWdY22bbHb7TAejzEajRCGIVzXfRK4NcWaxXjUudTHJM7TcRxkWSbP3cS2bdi2Dc/zUFWVnO/JZALHcdD3PdI0lSFWy7IQxzFmsxls25ZBWjVqPBqNYNs29vu9XHNirkRAVQSORQiz73sZbd3v93KNiiCwWH9ivsqyRFmWcF1Xxob1c0ySBGEYIssyZFkm45tt28qAcRzH6LoOURTJMYkgp4gXj8djTCYTpGmK3W4nw7O73U5euziOUdc1bm5u0DQNiqJAWZYywvr4+IgkSdD3PYqiwHq9lmHc9XoN13UP7uPnYpwiVltVFW5ubmRYVhCfL4oCb9++RV3XcBwHk8kEk8kEYRii6zo8Pj7i4eEBWZbJQK2Yh+VyKQPIYv1vNpsngVI1qCq2E/Fesa7Vtd33PTabDbbbrZwfEcUV97V6/nVdyzCyCP2K19X9qveL2IcYrzhOlmUyMiyed23byvtSjeCq+9Ofceq86NdKfM51XXieh6Io8Pj4aAxY69er6zqs12uUZfkkInvss/rzV8Sbf25DUV7T70OxdNP2KvV+0J9zpv2qYWHxftd18p94b+gchgLp6rZqTL/ve0RRhCRJsN1uZdz6WIh46LWhY+v3mPqeHvP9seFmIiIiIiIiIiIiIiIiIiIiIiIiol87xnyJiIiIiIiIiIiIiIiIiIiIiIiI6KOpEUH1p/qe+F1EXTebjYw7iphqWZbwfR/j8RhhGMr39Iil67qIogiO4yDPczw8PCBN04NooB4hPBZHVGO+IngahiGiKELTNMjzXAZAbdtGGIbyXEQwV5yL7/tIkkRGaAEgTVNkWSbHLsKyTdPI+Orj4yMeHx9l1HM2m6EoChRFga7rYNu2jH2KGGjbtjLeK373fV9Gb4uiwGq1QlEUB/MQBAGiKELXdXLsIkQq4rZDoctj86r+rYacRVi36zoURSFDqUPXVERu4zjG+fk5XNdFVVVI0xSWZaEoChmo9H0fk8kEo9EIVVXJcLHneRiNRkiSBHmey8+oQdHNZoOLiwtMp1MZuxXjFvHcPM+xXC5lgFlEai3Lwnq9ltHVPM/Rti2SJMFms0FZlk/mLAzDg3CuuI5pmmK/32M8HsNxHFRVhaZp5LoR5yqcnp4iCAK8efNGRrG7rkOWZbi/v4fruvKais+qwWVx7iLwWVWVXGdN0+Du7k6OVQRlh4Kc+vVv2xbb7VbGPvW4aV3XWC6X8rqXZYndbicju3mey2vYti3KsoTjOOi6TsaB1YCpiO2qc21akyLMLN4XUXARU769vZXXRHxerBU9xirmOsuyJ5FU9XzV56EYqxrdBSAjt5ZlHQRZxX7FT1MgW4xPD/uqc6H+7jiOfKakaWrcRtd13UHI13QMUxBW38cvZShYq9Kvj/63/ruI75pCvXrsWo/yDo1lKK6rMwW11XWibiP2L757LMuSgeqheLEeJT42Lv1aD4V+1fE+tzaIiIiIiIiIiIiIiIiIiIiIiIiIvgSM+RIRERERERERERERERERERERERHRRzHFC02BSBGr9H1fxk/F6yKmW5YlptMpXNdFURQIwxCe58nAq9guiiJEUYSiKLDb7bDb7Q4ireqxj4UE1bF5ngfLslDXNVzXheu6MhwrIqJVVcF1XcRxjCAIMJ1O4Xke8jyXAdY4jnF2doamaVDXNfb7PdI0RdM0cF0XeZ7j7u4O2+0WTdPIwOd6vcZut4Nt2xiNRnj16hWCIJCxVhE0FTHOLMtQFAXiOMZ8PkcQBAA+BEirqpJh1/v7e1RVJc+5rmuUZYmTkxPMZjNYliWvhYgUi3DxcxFGU6xVj44KIlKshmHF9QzDEOPxWEaOq6rCw8MD5vM54jjGfr/Her1G27bwPE/Oreu6GI/HcN3D/wIrgsqTyUTGYqfTqTy2GrS9u7vDZDLBdDqVc1qWpQzkimiybdvyb3GeeZ7LaK8YW5IkuLm5kWtGXWtVVWG5XMr1IIggtQjvlmWJsiyRpqncjxqhresa7969w/X1tQzQApABYBHlFP/EZ9XgZ9M02Gw2TwLcIvAs4sT6nA3FnNW1IqLXQ9uI+bRtWwaL9Yit+Ne2rZwrdT/PBWKfC7jqwd4sy56ER/Vj6KFfdf3r94IeqxbzYZpLcT7iWg2djxohFsc3jXNoLkRoua5rGdQ2Pb+PnctLrv/Q75+b+rz/WPo56WvlWOBW32YoiDsUZFafp6ZnKwAZfBfR5vF4jLqusd1uX3Qc/XjHgvzqM35oHvTzJCIiIiIiIiIiIiIiIiIiIiIiIvqSMeZLRERERERERERERERERERERERERB9FBARfwnVdhGGIuq6R5/lBMNLzPPi+D8/zcH9/jzRNcXV1hdFohP1+D9u24XkeJpMJxuOxjNBmWSZjn2r4UI2C6tQApzo2EfWdTCbo+x7b7RZlWcrty7JE13WYTCbIsgxnZ2dI01TGV1erFabTKZqmQZZlyLIMeZ7L8bVti91uhyzLDiKtRVFgs9nIqO8PP/yA8/NzRFEkz7MsS2RZhqqq0LYtsizDarWSY/c8T0Z+67qW2+33+4PwpxhDFEUYj8eIogjb7Ra2bWMymSDPcxmbHYpLmkKpYs5N0VQR8Q2CAJPJRAaSRZg5CAIZohXHvrm5kddWBH5934fjOAdrIQxDFEWBNE0RRRGm0ymCIJDrbLPZoCgKPDw8wHXdg3EVRYHr62tkWYa2bVGWpQwzq7HWoViqCP2KOfA8D33fy/3p9vs9uq6TcyvmUkQ3RVBWvC+OrQc93759i7qusdvtnlwLEb1W174puKpGOdV4q7iOQ3Fs9fqqQVI1cDoUfBXbitfEvOpxUH07/Rrof6ufFeevB0j189SfDaYgr+lcTc+U54Klz4VY1SivKYRuokdTj+1XqKoKdV0fPA+eO4+XBr1fsu2vhb5u1dee+9xQAFllWm/6c1N9/aXHd10XQRDAcRwZHDc9i6Mowmw2w8PDg4x9PxfnfW4eTPemPjdD+3/p+RERERERERERERERERERERERERH9VjHmS0REREREREREREREREREREREREQfbSjyKf4W79m2Ddd1sV6vUVWVfL9pGliWhfF4jCzLZNj25OQEk8kEAFAUBSzLguu6SNMUtm3DcRx4ngfP89B1HRzHget++O+QWZY9ibDq4UI9NBgEAYIgkPFXNRQsoqur1QqvX7/G69evEccx3r17J+Ovq9VKhlybpnkSW+z7HlVVHcyHGKuILjZNg7u7O+x2O4zHY9i2LSOzZVkeRFY3mw3KsoTv+2iaRkZygQ9R16qqjCFaER72fR+TyQS+7wMA6rrGcrmUAWB1bvRAY9d1cvymMKUeiNztdpjNZphMJnBd92BsVVVhvV4jTVMZY83zHHmeP1lDdV1jNBqhbVs4jiMDwOv1Gp7n4fT0FLvdDvv9Ho+Pj8iyDACwXC4RhuGToOt+v0ee57AsC23bygi06iUhSrE2Hx8f5bj1fYj1oEdbRYRWRFbVudUDsH3fY7lcDo5NX29D+r6HbduwbftJuNcULdUDt/rrpvtef10/D1NMVd1WzNOxfZvCwep61EPDpnkQx9KP/9JzMTHFW00h4eeunxpd1vdv+ozp2GIb8SwbOt7Qe6Y47dBxjo3h12Ro/l467qF1YAr06mtR/Zwe1dbHp75m2zbiOEYcx8jzHNvtFnVdHzzjbdvGyckJPM/Der2Wz5yh57jpHIZe08djoo77uegxERERERERERERERERERERERER0ZeCMV8iIiIiIiIiIiIiIiIiIiIiIiIi+ihqqHCIGvbr+x55nh8ERJumwcPDA+q6xtu3b7Hf79H3Pe7u7uA4DoIggGVZKIoC6/Uaq9UKl5eXmM1mOD8/h+M4aJoGjuOg73sURWEcox62VYOaIsjreR7KssR2u0VZlgcRwrZtsVwuEUURZrMZlsslVqsVmqZB3/coy/IgUiw+K8Kspvlo2xa73U5GbMVn0jSVIVo9/in21zSNjNmagpSmgKKYh/1+L38X87fdbrHZbFBV1YuipcfCserfXddht9vh+voas9kMYRgC+BAxTtMURVEcHPPY/jebDabTKaIokmuhaRo8Pj4CAFzXxW63w2q1OojqVlU1eAw1cvqx0Un1ujRNg+vra2RZNhiaBZ6GOk1BXD0gq8ZAReTWtKZN86auPz3IKuLF6lhN+xTj1IOkQ8FOUwBX/czQutLnQg+Oqse2bftgP7ZtH435qvvRY6pDgeCXxntN56bOvZhTfd8fs391fOrnXhrY/RhD+xkav/4ZfXyfk+l5+NL5Nu3rY+nff6bItCmyq35e3PciYj6bzeB5ngzBi0h8HMe4uLjAarXCdruVgfJj624ocq2OxfRsMsWJ9X2Ie5KIiIiIiIiIiIiIiIiIiIiIiIjoS8aYLxERERERERERERERERERERERERF9FD3Mqf+uRz3LskRd1wcBw6Zp8N1338FxHOR5Lt8TQVbf99F1HYqiQF3X8piu62I0GgH4EL+tqgr7/R5pmspY8FBIUI/t7nY73NzcwHEcrNdrGRRW45kiKHx9fY3NZoMsyw6Cv0NhRn0MaiCxKAoZA9aDpUNMoVjdsUijiMBut1sURQHbttG2Leq6Rtu2T2KterzRFMRVI5ViToEPEdOu61CWJR4fH7Hb7WTgsWmag+Opc2NS1zXu7u5QFAXyPD8ID2+3W6RpCuDDejoWbja9/mNjqOJa3t3dyfX9ks+oxxa/6+tgaH6OxWBNa1F/XYR8xedNx3lJ+PPYuZmYgr36eetxXT1aq45H/bwa9AXw5BzVsalBZHW9mELEehxY305dP/pY9WthmuehfZvWwzFD5/qxTPeJHqE9FnM+9tov7SXjOjZfpm2Pfa+IbdSItlhfL70+pudS13XIsgyWZWE2m2E8HmM0GiHPc+x2OziOg/l8DgB48+YN6roePOZQ3Fh//bnY9FCI2PSMISIiIiIiIiIiIiIiIiIiIiIiIvpSWfzPckRERERERERERERERERERERERPRzsyyL/1ntC+K6LhzHOXjNFP90HAdBECCOY+z3e1RVZQyTDoU99Zig67qI4xhJksC2beR5jv1+j7Isn4Q5h4Km4j3btuE4jjyPoRisHg8dOlc9zKkbimCa9qn+bXpvKBAqQopDx9fnQn2v73v5Wf08jkVG9dinun8xz/q4TeMyzduxuTd5Lsp57Pj69kNrUN1GH1fXdcZA8LHrDkB+7lhMVd1GHaNt208iynoMVL0fAcjotSnIKfYjPiOu39A91XWdPL56v+ixW3U/+tpQ156J2I+6jfjssXWtv67GVvWYr2nND+1fP4YYi3r/6OM33VtqiFgfy9CY9P2q12lofCbPRZrVKO2xc3vp8X4NnosQH/ucaV3r19F03Uzzpt/X+mfUdSf27bougiBAFEUYj8cIwxBN08B1Xbiui9vbW9zd3cl7V70u+hozfY8ce04OrcHn9vO59X1/vH5PRERERERERERERERERERERERE9CMw5ktEREREREREREREREREREREREQ/O8Z8vxyWZcHzPPm7oIb81DivGi00hVn16OdQVFANmOrbDkUW1f2oodFj52b6/FB02PTZY/s+ts1QzNcU7NU/I947FvscipKaIsT670PhXjVgqQZg9fG+dL6G4rkv+aw61mPx3KE1oO9TXa/650TAVj+OKeY7FIU1jU8fuz6n+poHDqO26vam+0EN2erHEsFP01wOxXz17fQosR6ENa0dPeL7knuk73sZJn7JvA69J8ZommdxLqZ747n/e22KJJueP8diwXooVR+3+poeSv4xMV9xPMdxZPRZPfbQM/LXSl27x56tz+3DZCjmqx73WAh3aDz6fak+Rx3Hge/7iOMYnuehaRrs93vkeX5wvdR7XA096+elPzfEMcV4TN+F+hjV++RYlPuXxpgvERERERERERERERERERERERER/Zzczz0AIiIiIiIiIiIiIiIiIiIiIiIiIvptMYU9TfFJsa0a+BuKGw4d5yXEMdVYqf558Z4psvjc8fQgpOn8Xxqifcl5fMr2QxFkPRj5XPBY/12nzp0e+TXtV7w2FII9tt1LP/ecobkxHds0L+rvQ3FL4GkQV99mKPCrhjD1AO6xNanuv21b4zVW51C9D8VY9W2HDK2j57ZXx6jf+6aIrnhPnQfTuNq2lUFf9TxNMWN9G9O4TNdEf1Y8t851euxVH5c+l6Z4sGoouq1e12P7GLpmpnlTr41pzn5LXnLeL/Gpz/iXPE9Nzz3T+gWAoihQFIV8ves6ef8PBbRFnFpdh+rPl6zBoXNXjzkUDiYiIiIiIiIiIiIiIiIiIiIiIiL60jDmS0REREREREREREREREREREREREQvpof+ngsVDkV/9QCnHjY17Ut/X92X/vtQmPVjI4PiM47jyMjlp+7PFG007csUDX7J54be198bipPqoVDTPIrfHcd5EpBUw6him+fOfSjurI/TdH7PhTWH1pEp3PtcyFcfvyn0ajoXNfCrh2P119V9ijk1xTFNcd5jc2AK0wIfro9+zZ77vH5++nw9F2PW73l1XKaIqXq++n6GYrb668fWiR4aN33eNBdDc6Buq/7U72PT82/oWn1sQPbYmtTX8NBz5SVz91ugfze85Hl9bK2bnsnHwrv6fofmGzgMbJue++q+n7tWpufb0Dl+bKx56J7+mNA1ERERERERERERERERERERERER0W8dY75ERERERERERERERERERERERERE9GK2bcvQKDAc9Rx6/8eG/p7b37GArym+qodKTYHNoXEfiyX+mPMwxSOPxVb1v4fijEOxVNN+1Nf0uKg+vq7r5LrQA6jHjvVc3PfYts+dgykOO3Q8076OjU1d/2Jb9XhDhiKfesxWHEO97iKcrB5PP7eu657Ml7qfY+dsOl+x/6HwpykybAps6/vV96PT31PX1NCz51j41rTuh+6pY/f7S943RU3FPWLaj7h3ju3fFJA1PeeOjcu0rel1/drrjt27v1amdXdszB87l0PzORT+PRbhNq0f9W813Gt6nujPgaHvh6FzG4o427YN13Xlc0jdz9B3FxEREREREREREREREREREREREdGXjDFfIiIiIiIiIiIiIiIiIiIiIiIiInoxU9jzWMzwuX3p2w/FQNXXh4KE6ntd1x2Nt5rGNhQy1M/lWAzTtL06NlMI1rT/ofjqc9HS58ZjCpvq0Uk9Ljs0V6ZAqv7v2Bj14OTHBJFNoUvdUMhXbG+a85eOR4+eDoU19THqc6NGXYF/xrKfCx0PBW+HQsOm8YjPDMVuh6gx4WP0/b40JqzHtNVx6yFjPTasrl3xGTEnejDZtHbUfQ+tqZeEfJ9jWpf6dTEZmjv9/h065tBciNfET1Ng2BT7HQpUfw7Hzl3fDvi08Ky+nofuzaG1bnquDs2p6ToMPfeG1uzQnIj9ivCv+Lz+/BCfdV0XQRCgLEvUdW08/1/LOiAiIiIiIiIiIiIiIiIiIiIiIiL6JTDmS0REREREREREREREREREREREREQvpkZcTQHBl8QsTZ/RA4lDIc2hY5jiuabXh14bev/YMfWwqIkech2KiZqCvcfO67njHNvedH7H5uRYeFXdpx6WBA7DpKZxmQKS6jamaKXjOE+Cw0NjPLY2TQHUl8z3c+HllxgK/w6FPkXk13RsdY7V903bi9dExFM9NjB8vUzRZtMxh+brWFhZv49M5yDCpqbw6FBI+dg2x8LLpjVgeuYNXWvTe6agsh5uHXomDu3DFIPVz8c0Nv38LcsyBoSPPYOPbfO5vGQsL7lPn7u2AAbXgr6OnhuTvq7EtRgap+u6T8Zgunbq+rFt+0m0d+g89fcdxwHw4blQFMVBJF+PPP+YSDIRERERERERERERERERERERERHRbw1jvkRERERERERERERERERERERERET0YiKQqcdaTYG/Y3E/NVL6XNxX/4xpn3q0U92neE8PlerBTD2iqo/RdNzn4sWm2Kwpvjn0WfV3fW6Hoold1x2EXI8FgMV+1Ujjc+NSP3csovuxkWd1rCJe+9w2nxJrVs/B9Ppz25uiqer6U+8JNbapz5cawRTRzKH9HwvEmrZVY7Z6bPi5CPLQtR2aJz1Kq9OjnyLMq5/X0D0tdF0n16n6nh4SVfejrmt9f6ZnhWkc6jGGngNDUdRj97D6uWP3qP7+c/OszrE+L6Y1rD8Xj13/Y+Hrz+XYvWz6Xjk27pe8/9JnuX4/itccx3myvRrw1cdxLDSt3uOm7zn1+Op3tv6sN52Tfm/qAWP12PpniYiIiIiIiIiIiIiIiIiIiIiIiL50jPkSERERERERERERERERERERERER0YsNRf+G3lN/N8U09W3V43xMGHAotqm+Ztv2YFj12Dl0XWcMMKpMoWDT+TwXwRXHPRaOHTo/03yYmPapxyP18OVQxNMUZX5JxFc9rim6aQrA6p8diprqnzGtieditc8dxxQJFa8PhWbF32qQ1hRpVrcVa1aNBJuOrwd6TetCj4o+d7/qa1rflz5uNUg6FBoeWq/AP58PpoCtHkUWwWpxTHUuTaFU9Zqoa06/tm3bHsyfKVZuCuOa7iN1fKb1M7R+9fWjH+vY/aCPZ+g6qMfQ46/6vtTP6mvw1xBvPTYGNTb73P1+jH6v6NdZvG56XurPTj2gq0aqTcds2/bgb/366ver+CnGpX/edD7qT3XMQ+vWtOb0fRARERERERERERERERERERERERF9yRjzJSIiIiIiIiIiIiIiIiIiIiIiIqIXGQrsDsU3Tb8PfWYoxquGEvVj6Ibiwmp80RRYNI3r2D6PBVjVzw3FUo8FTfVtTOMc+lsdlx4vfQlTBNYURlW3NR372FjVmOixiKQpJvwpwUj92utBabFvPWiqx1718erxW3WdmuLE6ueGIsFD4V9TpFYNAg8FXk2BXtN10j9vip+a1vyxwKy6b33OTfe0HtjV2bYN27ZlmPRY4FaPpYoxqFHToXvCcRzj+E3PK32O9P3qMeRjz4yh5596jI8NpR575gzta+h1/f741GjrLxl9Pfac/DH7Ml3LY9dWjVyr5296RpvuzaFwrmmMx+K6Q/ezaTv1OWyK96rr9dh9S0RERERERERERERERERERERERPSlYsyXiIiIiIiIiIiIiIiIiIiIiIiIiH60lwR7n/v8UBR3yFCoUH9Nj3qa9jEUYD0WwVTfU6OHesRTH9NLopJD8dChczV9ToxHjac+F2AdihwPHUuPO5qiskPRSnWe+r6H4zhHA5ND86rHNE3zYApSmo6jvmba57Gg5lAQV78OwrGAcNd1BwFhfTsRndXDuENrw3Rt1KDo0Po/Rt9G7EsPPuuBWz2OLI7dtu2TtTIUETYFdk0RavGeGlQVc6t/diiALN4zXS9xHBFXfu7e0e9HUyxV7EMfo/77sdfEsUyR4Of295L3PoU49x8T1f0UQ89402sfE1F+yfum9Ws61qfsX6zp555FAJ7cL0PPFf2nvn708Rz7niIiIiIiIiIiIiIiIiIiIiIiIiL6kjHmS0REREREREREREREREREREREREQvYooLmiK8z/2tvmaKWaqvmz4/FLQ0BTlfcg6mAK3ppwh2HgumPjdeUxhR3a/pHF4SVTQdZyiwOBT0PLZvU9xYj7Wa9qNGJAVxvrqXRFk/lj6fakBVP9ZL96eOdyiQKY6hhmNVH3M+ahzX9LcpkvqSe0B/TQ3l6uNU39fjyOKnHujV477qtm3bPtmPfryh/QzFqfXxHRt/27ZPzl0/F8uy5DjV7dXtnpvroTnWXzfdq/p2Yn71NWw6jvr5j1lr6mefe84dC8Lqz3Hf9zEej+G6LtI0RVEU6LruSSD5p/ZjQsgf89rQfobmyHQfPXcsdZ+m57DYn23bB88I13Xh+/7BPSd+H9rfsXMaGstz65KIiIiIiIiIiIiIiIiIiIiIiIjoS8KYLxERERERERERERERERERERERERG9iBq6FESwVLz/XHjWFGrVA4DHIp/q556LM74kSGna9ljo1RS1fen+h/b5knE+dzzTvvQIsbqNGkTV938s5KsfxxSS1GOqeoTWtK+h8za9ZloLJh8zdy+JUJpiuernxPnpc2MKCgOH91HXdQdzaFo/Q1Ho52LMprjrsTkxjct03uqxXhKBHToPfX/q+Q2tRdN+9TDs0PNHPw/9+eW6LmazGQBgt9uhqqonYzl2r5iCxEPn/pLwqX6P6fTn8dBzUY8uH5vbY/fnc/erab/z+RyvXr0CALx58wZlWcr3fqn463ORWv2e8DwPSZIgTVO5BoSPCTi/9DtJH6f4XY84m7ZV1756f9u2jdlshvPzcyyXS6zXa7Rt++RcxBgdx3nR+ahzaYprExEREREREREREREREREREREREX3p7Oc3ISIiIiIiIiIiIiIiIiIiIiIiIiICXNc9iJOqTMFK8ftLAoaCuv3HBHPV323blkFUdR963HXon7qNSo2smmKrahTUtM2xIKceWH3u3IfmWf2nbqv/bgp0DgVPhz6jj/O5sQ/FVUV0Uv+sPiZ1//q5msbxkjENjU/92zQX6nmowdshQyHeoWMOvabfT+L4agxY31Zsr0Zfj10j/b3nIqimNafuR70fTfT7T99WDwoPzZM6FrGNHq02ba9/djwe4/LyEmEYwnGcwfGbxmra90ufffq+Vfoa+9hoqmkNq+PXr/dL7hd9rENr2XVdTKdTtG0L27bRtu0vHn197lqp24l/8/kcv/vd7+A4jvEZBwxHtoeeXQCefE+ozxDT50zPTXU78VnbtmWMV/zueR5msxm6rkNd12jbdvCcn3vODX2X6+dDRERERERERERERERERERERERE9K/A/dwDICIiIiIiIiIiIiIiIiIiIiIiIqLfjmMBUOB4ZFLfruu6g2Ch+rfps2ps0Lbtg/Dgc1FU8fmhIKy6n6FA7bE5ML2uhxn190zzom6rb6Pv6yVhSlOQdyjSO3QN9Wtlet907Jc4FjY+th99HHrgVj8v9fqboqhDoeXn1rNKjeW+9HOmGKjpWonX1XWvE9fIxBT6Hbpvh4Ke4thiHOrfQ7Fl/Zz18Yu1Jd47Fi42/TTdX0Pj1qOl4rjqGG3bxtnZGVzXRVVVGI1GcF0XWZYZ14d+fvq49OMce06JeVXPQX1fne+hsK/+LDRdA/G7+FsEXsXfQRCg73vUdW0co+mY+mvqMU5OTuC6LjabDWazGYqieFEA+6dgWjOmZ5guiiJcXV2hbduD6/eSNWB67r/kPtPXp7oedPozQv9+E8bjMcIwxHK5RJ7nzz6r9f0OfX8RERERERERERERERERERERERER/atjzJeIiIiIiIiIiIiIiIiIiIiIiIiIXkREDU3h2WMhTxEZ1EOepkCgHgbV6cHEl8Q8h/Zhet0U3TWFOj8lRDkUkdSP+5IQrB4y1YOQpmilHvJV31MjokOhUH0MuqHw5EvpAUvTvKvX/9i1eu4cjr2uj8d0TH0dDM2J4zhwHEdGXcX2pnvJdMyu6+R8iP2ohgKiQ+evv6cfXx2X+GfbNlz3w385bpoGTdM8G7w2RZFNYVPTOtVfsyzryXzpP02hWzF36nNHHaPjODJkK+Y2CALM53Pc3d3Btm1cXl7i5uYGWZbJ46jX0XTfDa1D3dC1NM2b6ZkriDGJ38Xnnju2mBd1fqIowuXlJbIsw/39/ZPPqevyufvHtm0kSYL5fI40TQEAWZahqqpfJOSr+pgoreu6WCwWiKII3377rVwjpv3p+zx2TY/dk+KnKUg+9AwWr4sItn5PRFGE+XyOruuQ5znath1cl+q9M/T9a/r+Es8H0z6JiIiIiIiIiIiIiIiIiIiIiIiIvmSM+RIRERERERERERERERERERERERHRs9SQnx7804OFpsCoSg8rfkps1bSPoQChfg76+E1hw+eO+1L6mExxR9M8PrcvU6BVj0I+F7AcmjPT+IYiqs8dQx3LUHh5KHL83Jieix7rwdWhcR6bq2NrpOs6Y2hVndcoijCdTrHf72XQVD9ndU2KMKtpHPoc6veZKfj60uiw6d7wPA9JkmA6nSKOYxRFgfV6jc1mI+OxpnMZ2r+I6w4FhU1BX/V1NZgLHIaHXxKX1ccD4CAc7TgOptMpPM9DURSYzWZwHAd1XctrogeDhaEItfq+Ol9DTJFg07jV34fWoP65Y89dy7Lgui7Ozs5wdXWF6+trY0RZ/9wQy7Lg+z5OT08RBAG22y2iKMJ2uz1YOz830/fRseeBZVmYzWZYLBbY7XbYbrefPE7Ts29oG3W8ps8PjVe8pgZ1xVqezWbwfR/r9doYUDbNy1DI13T857YlIiIiIiIiIiIiIiIiIiIiIiIi+pIx5ktEREREREREREREREREREREREREL6ZHIY/FT/XfTSHL50KFLx3Hsf2p8c+h4x2Lcw6N99g41PfFP9PxXxqnHYoXm+b/WABU/K0GY48FXoc+/ymOnaPpWCJSKQKwANA0zZN4rQiRAkDbtjIWqhuKWZrGcCy6OrQfPUBrWRbm8zlmsxnKshw8Z/2z6nHUczeNQY9wmqK+z503cHhvuK6LJElwcnKC2WyG2WwGy7Kw2+3QdR2yLJNx0I8Nduuf0cdnChnr+zXNkxi/KY6q7t8URBb/fN+X18q2bcRxjNVqhTRN5ZoyRYOPBamHwr/HDO1ff88UUh56zpi208cXBAFmsxlc10Vd18bxHAtkq685joPJZILJZIIsy2DbNvq+x36/N8anXzIvn+q5gLIadh6NRri6uoLjOHh4eEBd10ev37F7f+hZPHSePyYarH7esixEUYTFYoGiKLDf7w+em/ox1c+ZYttDx9O3+yXizERERERERERERERERERERERERES/Foz5EhEREREREREREREREREREREREdGz9ECoYAr4mcKGQ/saMhR8NG2n7lP8bQrxquPRw4ZqfNEURDXt47nzEfuLoghBECBNUxlBfe6z+vmZ9q2P57mw8tCY9etlel8fz1BI9Ngxhs5DvOc4DjzPQxAEsG0bjuPAcRy5/6ZpkOf5wRxaloXxeIzJZIKmaZCmqQyGmuZAPa7p+h97X58DscbUuKWIvoo47HQ6hWVZaNv2RSHjoXWv/i2Ct0P33rH1+Vy4WQRtz87OMBqNAAD7/R51XaNpGriuK6+Jer+If8cC2MdCouoYhu5N036O3cPqPm3bhuu6sCwLVVWh67on4eAkSZAkCbIsQ5Ik6LoOm81Ghm2HwsWCel1M96M6Hn38z10j05x8SijY9Hkx3iRJEIYh9vs98jz/5OisbdtyHQFAlmUIggBVVaGqKrnN0DmI8X3M+TxHvSdMP0W8+auvvsLJyQnev3+P7XZ7EEY+9swwXb+hZ+xLz01/LpvWvWmt2LaN09NTJEmCh4cHFEXxJPBsiloP0QO/ephb7IMxXyIiIiIiIiIiIiIiIiIiIiIiIvpXwpgvERERERERERERERERERERERERET1LDRHqgcFjIUpT2FB//6VBQPVYeojyWCjVFMHU9yuChS8dt/6eHldUI6JJkmA+n+Pdu3cyZnnMS8KIx8b6XBxTn0edKXypbnvsM0ORSVO01bZteJ4H3/flvyAI4HkemqZB13Vo2xZ1XaNt2ydhTeBDfPbk5ASnp6fY7/ewbRtZlqFtW+NcmcYztI3p3NW/1XCtHuK1bRuj0Qie5x0EYT9W13UH0eCheR2KoqrjMb2nnqfneZjP57i4uEAYhkjTFNvtFmmaAgCiKIJt20djwscCwyam++ZjQq6mNafGlsW5B0GAyWSCvu+xWq3kfajOz2QykecXRRGyLJMh1CHPBYqHxvzcZ/RzMZ3jsfv8Ofp+XdfFdDqF67pYr9coy/Kjz02Nco9GI4xGI6zXa7n/zWaDruvgOI68Nvp9fex8jsV/X3rOQ8/NMAxxcXGBr7/+Go+Pj1gul6jr+mhcW4/cDl0zdex6uFuNTw/FvE3rZeh51fc9fN/H69evUZYldrsd2rZ9NuKt7lsfg+l+1L9DXrruiIiIiIiIiIiIiIiIiIiIiIiIiL4UjPkSERERERERERERERERERERERER0bP02K1KjRLqrwNPY4CmQONQsPBYIFY/pmk7NU6oHm9oP/q2OtN5HuM4DsIwRJIkcBzHuN+hELApVHwsnDgU9zwWqn0uuquGKV+yHzVya9redV24rivDvWEYwvM8uO6H/9LadR2qqkJRFKiqCmVZoq5rY/BTRFoty5Lb69HaTwlFfyxTqNjzPEynU/R9jzzPUde1vIe6rjsIxOprTpyDKQo6FFQe+ltc/+fizWrI13VdLJdLPD4+Ik1TtG0L3/fl2EUo2RRWNc3FUAxV/YxpnHq8dShKPBQ+VffnOA6CIADwYQ2KuLLYdxiGmM1mqKoKnucBAHa7HaqqMp7T0N8/ZfDU9PxzHAeO46BtW7Rta4z9Dl1j/TmtfiaKIszncwB4UQx7iAh0z+dzdF2H3W6H8XiMsiyR5zksy8JkMkGSJGiaBrvdDkVRHJzL0BjVv5+bN9O2pnVi2zZ838dsNsOrV6/g+z6ur6+RpulBFFo8u8V9O/R9d+w1/Vwcx0GSJPB9H1VVIc9zNE0zeG8MHUO9J2zbxng8xnQ6xV/+8heUZSmPqc6HbdsH95b+nNePY4oM688pIiIiIiIiIiIiIiIiIiIiIiIion8ljPkSERERERERERERERERERERERER0bOGIqnA02Co/v5Q1FINKr4kRqm+r+7juago8CHCOBTLHYr/HovSqud+LLZr2zZc1z0aPDQFII/FWXXHopW6Y3M9dMxj8WN1v+qxTe+LgG8URQiC4P9n519+JMkWPd/rZ++3v8MjIjMys2o/zlMttcSQyRVDaIHEgNtCYoR0hcQACSGYMoABI0BCcHWlOwAmLQH/R0vNFa0+3WefvXdVVmVlZrz87W5u7weDPGttcwtzj8yszKraO38fqRQR7m5my5YtM69B6gtd16EoCsqyRBiGiKLoQdizay7EvnVdh2EYSJIEURRBVVUZ/X1sPbTfE9etGah8LHR8KgwdBAF6vZ4MEgOA67rQdR1pmiJJkoOxto/XnEfxe1fQ+NQ5HbtH2vsWgdWnT59C13Xc39/j/v4ecRzLbcqylLHRoihQ1/VBkFjTtAfHaIZP22uu/dl28PRYeFa83o6qto/RvmZpmmK328GyrIN9i9/7/T5s20YYhjg7O8NyuZRR2679tUPXzbE8FlYWrzWfM13XqHmdRPjVtm2oqoqyLJEkCfb7PfI8f/Bc7Lqv2/tUVVX+NxgM0Ov1sF6vZcC5uc+ue+rYOB3HwWAwwHw+R1EUsG0bt7e3KIoClmXh66+/RhAEWK1WUBQFRVEcXE/DMOQ5NufpfdZ+cy6Pnbv4W0TFfd9Hv9+H7/uYz+dYrVbyuoj/LMuCpmlIkuRBILcZ/RWvifM5FsEXAeXnz59D13Xsdjvc398jiqIH59r1nXns+8A0TZydnclnqgilV1Ulo+htp54/zedh87ujPc9EREREREREREREREREREREREREXxrGfImIiIiIiIiIiIiIiIiIiIiIiIjopFNxWOGx+KnYj4gjapomg4BVVaGqqgdRx659NGOCIiBaluXB+yLce2w8XfHhY8fqihB3BT2PhQ3F+QLv4rNdNE2DpmlyDk7FeR+LOT7msXMQjo2hfbx2RPQY0zQRBAF834eu68jzHKvVSgZi23HYx+ZWVVUZBS6KAmVZQtd1xHF8EL3tmru2Zjyzvf66Qqan1o+iKDBNE+PxGKZpYrPZIEkSqKoK3/fR6/Ww3W5RliXSNH2v6PCpsHF7TO3PA+9CvM0YdzuU6zgOnj59Cs/z8PbtWxnybc5FVVXIsuwgtAz8KVZtmiY0TUNRFEjTVN6TzTE1483ta90cm4iGNu/15vvN+OipeK3Yd13XyPMc2+1Wrpfm+5qmYTKZyHE7joPtdnsQbe26Nu112XUvHYtvd0VV22tVjM2yLAwGAwyHQwBAkiQAgPPzc3z//ffYbrdy3sRzpDlXYn/N4LZpmrBtG5ZlQVEUnJ2dHYSx2+fRdZ7N57kIa5umicFgAMdxEIYhPM9DURSIogi6rmM0GuHy8hKbzQZFUcBxHOi6jizL5Ll6ngdVVVEUhYz9xnGMLMtOPmfEeI7Fh5vbGoYB27Zh2zY8z4NpmkjTFG/fvpX3iwj46roOz/Pgui7W6zXW67U8VpZlclwinCvubQBye+BdUFpsp2kaBoMBptMpNpsNLMuCaZoyZt61Zpq/t/8W9+B4PMbTp09RVRV6vR5830dRFMjzHLvdTs6ziCWLNf/Y92Ez6iu04+IM+xIREREREREREREREREREREREdGXhDFfIiIiIiIiIiIiIiIiIiIiIiIiIjqpK6DYDFeKsGPzZzsEqGmajB1algXLsqDruoyEJkmC/X7fGbnsCrmKfSRJchDnPEYEQrucCte2Q6NdwdtjwU5FUeT56rreGfNVFAWj0Qi9Xg+73Q6r1Qp5nj8Isp5yKiraHmczhto+3/b4gT/FWtsR1mPXumt/IvgpYsWu66KqKqRpKqOXXcc/Nb+2bcP3fQBAnudy/1mWHcRej51X27E13vXeY+Fi3/cxnU4RRRF2ux3SNJVxUgAyEtq1v6qqHsRsu8bcnPtTEVnxXjvYKX5XVRWXl5cYjUa4vr7G7e2tjMU2NY+l67qMxuq6Dt/34TgOTNNEVVV48+aNDLGK42mahuFwiCzL5H3eDK+2xyTmonku7xuwbu+vGTWOokhuL8ZgGAaGwyFev34t46ebzQZ5nj+Yy3bIWcxr87q0j9u8ns0I7rF7ULwvwriTyQTj8Riz2Qw3NzdI0xSO4+Bv//Zv4bouwjBEVVUwDAO+7yNJEmRZJiPp7XNwXRfD4RDD4RD9fh+qqsK2bex2O6zXa7lOm9HhrvUl5s5xHARBANd1ZVBWxIVd18Vms4GiKAiCAKPRCK9evcJ8PodhGDAMQ86RaZoYDodwXVd+ZzTD34vF4mBsgvisaZowDAN1XaMoChmHb8+3pmkyZCxeNwwDq9UKq9VKrmnXdWVoXDxvxLNMrPXVaiXXURAEsG0bcRwjz3M51+KaiLErigLHceD7Pu7v7xGGoXzGiufYqXB3c52L4LN45nz99deYTCZYrVYYjUYy5Jskidxfr9eDZVnYbDZYLBaIoujge7cZ6W0+R5vr+th3IREREREREREREREREREREREREdGXgjFfIiIiIiIiIiIiIiIiIiIiIiIiIjqpKxbbFeEEDqOPIgRr2zY8z4PneTAMA5qmIU1ThGGINE2RZRmyLENZlp0R1vbflmWh3+9D0zRkWfZBYxfaodT2ORz7/X2OJbZpxnxN05QR1LIs5TaGYeCrr77CdDrF999/jzAMZUT0sWMd+/t9osXHrmcz3KhpGnq9Hnq9Hqqqwnq9liHWDw0bp2mKqqrgui6CIHgQBj51fu1jKYqCfr8P3/dlfNRxHBmubJ9Xc3/t63NszF3H74pXdx1rOBzC8zzc3t5iv9/Lc12v1wdh42ZUtn1sEWJtn7cgIqRd27b/7ppfEdLt9Xp48eIFFosF3rx5gyiKHuxL/NR1HY7joNfrwTRNFEUBTdOgKIoMAH/11VeYzWaI4/hgXnzfx+XlpXxPBFVVVUWapg/irO15fiwYLeKxp2Kj4jNie/F86vf7MgQdBAFms9mDkG9zm+Y+m+8dW2vtcXYFf5uva5oGx3EwmUzw9OlTqKqKt2/fykiyOI/vv/9eRpN1XcdkMsHZ2Rmur69lTFasExG8dV0XZ2dnODs7g2ma6Pf76Pf7SNMUd3d32Gw28vl06p4WQfV+v4/RaCTvP/HabreD7/vIsgxhGMJxHIzHY+R5jrdv36IsSwyHQzl+0zTR6/UwGAxgWRbSNEWSJIjjGEEQ4Ozs7CD4Lp5BIhjc7/fR6/VkNDuOYxnSNk0Ttm3L7XRdh6qq8hr3+30oioLZbIa6rjEcDjEajZAkiYw6Z1kmr8t0OoWu69hut3LNOY6Dfr8PXdeRJAkURYFhGOj1evA87yCQbZomRqMRAOD29ha2bUPTNHlPiPCvuA7tgHTzuaCqKjzPw3Q6xeXlJfr9Pu7v7/H73/9eXksRFe73+5hMJphMJlBVFdfX14jj+CD021zPXWu3a70z5EtERERERERERERERERERERERERfIsZ8iYiIiIiIiIiIiIiIiIiIiIiIiOhRXXHMrsgj8C4waJomPM9Dr9eD7/vQNA37/R7z+RxhGCLLss6YazsS3LVv27bR7/ex3+9RFMXB+8circfChM3PHotvNomQ4vvMhfhcVVUoyxKmaT6I+VqWBU3TsFgsEEXRgzjrhzgW2D32WeAwnNpm2zZevHiB0WiE5XKJLMuw3+/luXVFdo8dq65rmKYJ3/eRJAnm8znyPP+oEKRlWTg7O0Nd10iSBKqqwjAMbLdblGUpA5td59t0LHR7bA6PxSub60HTNIzHY+z3eyyXS3mOiqKgKArsdjtomnYQI22uxebvuv6nf+Z7LN7bNQ6xn67PNs/NMAy8ePECAPDdd98hjuMHQedmyLfX6+Hi4gLD4RAA4LouXr9+jZubG0RRBN/3Udc1LMs6WB+2bWM6nUJVVRnfFrHaqqrkWhBjchwHmqZhtVp1nsOxOLFhGFAURUbBm58VYVsRyxX3omEYODs7Q5qmMi68Wq0OQsLNyK6IFxdFIZ9h7c+IWLNY911rqOuaib8dx8HFxQWm0yn2+z1++OEHubYNw4DrujBNE7vdTj7/fN/H2dkZAMhxifGI8Yno62g0wmazwXK5xGAwwF//9V/LOK24v7vWTXMubdvGeDzGcDiEoihYr9dI0xTn5+cyzmuaJhaLBQBgPB7DcRzc3NwgTVM4jgPDMOR6cF0Xo9FIBoB3ux02m42MwbuuC9d1ZSRaxIn7/T6ePn0Kx3EQhiHCMIRt2zg/P4eu69hsNhgMBuj1esiyDHEcI45jGQUejUYIgkBuOxwOcXFxgSRJsFgskCSJjAGLCLCu6zJSLCK/vu/LMYhYte/7GAwGiKJIBn5VVcV4PJbR6CzLMBgMoKoqJpOJPG4YhgdBeNM0ZfhazFnzXl8sFvJZ/urVK9zd3cm1EQQBnjx5Isd3e3sLTdNkpPjUtW4ep/ndJ+4lEQtm0JeIiIiIiIiIiIiIiIiIiIiIiIi+NIz5EhEREREREREREREREREREREREdF76YqZNn9XVRWWZaHX66Hf78OyLKRpivv7e4RhiDRND0K2zX0cC8G2jymOYRgGoihCURQfFBNsxzOPHU/8/qGRwub2uq7DsixUVYUoitDr9bBcLlEUhYx86rqOKIqgKAp2ux3SNH2wn1Nx4+b7j42pOc/NiOwxIpC62+2w3+9h2/ZB5LQdfm0fq/m3YRgYj8cYDAa4vr7Gbrc7GaY9ds6KoiAIAti2jfV6jTzP4bouAMi5MwwDuq4jSZL3iiMfO4fmWETAEsDRNayqqhzb/f29DIY2iahnMzbbjrkqigLLsmBZFsqylOfRnNuu8LQgjnnsfITRaITRaITXr18jDMPOuDYAmKaJfr+PyWQC27ZlZNVxHGRZhu12K8e32+1kULmqKqiqKmOq9/f3MnTa7/cRBIH8PPCnYPB0OkWe53K/AGRIt6oqef80r5/jOOj1eijLUj5r2uHRqqpgGAZ6vR6Gw6EMxU4mE+z3e0ynU5RleRDp1TQNtm3DcRw4jgPbtmWYfLlcYrfbyfvYcRz0+30AQBiGMrbbXMunYr4iuHt5eYnxeIz1eo3Xr18jiiI5lslkgsvLSywWC6zXa5RlCV3XMRqNoGka7u7ukKYpDMOQEfU4jqEoCiaTCfr9PpbLJW5ublAUBUzTRJIkSJJEzrd4NjTXaPMesCwLk8kEg8EARVFguVwiDEN4ngfP82SEWkRsPc/D+fk5wjBEFEUA3sWHdV2X0dp+v4/BYIC6rhGGIbbbLZIkgWEYSNMUruvK66goCgzDwGQywdXVFcIwxO9//3vs93v4vo/JZALTNFGWJRzHwWAwgOd5WK1W2O/38r50XReDwQAAsNvt4LouLi8vkSQJbm5uUJYler0eBoMBHMdBWZaoqgpVVWG/3yNJEnkM3/dRVRV2ux3yPIdlWXKNiecU8C6QfnFxgc1mg91uB8MwZPx6OBzi7u5Ornlxvr7vYzgcIk1TLBaLB98N4ntwOBxiu91iNpsdrDsRPr69vcXd3R2yLJPXKMsyGf1tP1+6nh3NZ347Rk5ERERERERERERERERERERERET0JWHMl4iIiIiIiIiIiIiIiIiIiIiIiIge1RXybb5nGAb6/T6m0yl838d2u8Xbt2+x3+9RluXJ+Giv10NRFEiSRO6v/VkRkhRhzbquEcdxZ3SwGfAUx25GbEWUta7rg/BpM1LbPm/xfjOq2hXmFFRVRb/fx9nZGeq6xm63w+XlJSaTCXa7HYB38dLLy0v0+33c3d1hu912nvexqHDXOI+Np2uOjkUbxfyJEOV2u4Vt23JOj8WcuyK1Yi5E4DlJEiwWCxm3fCyY3A49i0hpmqbY7/cwDAOGYSDPc5RlCcMwMBgM4Loubm5uZOC3OcZ23Lh5vrquy6CreF8EXYMgAAAsl0tkWfZgrCJSm+e5PEcRvBTHO3b/iO1t25brxnEcAMDd3R2ur6/lmNrbNeeqGQ9uX9/m5x3HwcXFBeI4luHSrrk3TROj0QiTyQRlWcq4rO/7MgIq5t+2bRlwbYZIJ5MJsixDGIYyfup5HsqylPewoigwTRO9Xg+u6+L6+hplWcprMx6PEQQBVqsVlsvlg4j4cDjEYDDAer0+eM+2bdi2jTzPkSQJbNuW91ye5xiPxzBNU57rbDaT11w800SgWURRDcOA4zgyLAwAruvi/PwcnuchyzJYloWiKGSsVoxTXIuu+9z3fTx9+hTD4RCr1Qpv3ryRId+6rmHbNp48eQJd12UMu65rWJaFwWAgI7hlWWI4HOLy8hJZlmG5XMJ1XYxGI4RhiPv7e2RZJs+vKAr88MMPcp/H7nFxjabTKUajEaIowmw2QxRF0DQNnufBdV0kSYLlcinjwKPRCEEQ4Pr6GnmewzRNOI6DKIoQxzE8z8NoNIJpmlgul9hsNkjTVJ63WJt5nsvQ9dXVFabTKW5vb+V+FUXBaDTCeDzG69evEccxrq6u8OzZMxRFgd1uJ6+hrusyIDybzVCWJZ4+fQrTNHF7ewvXdeH7PizLQl3XWK1WME0Tvu8jSRLs93ukaSrXnm3bMhasKIoMBc/nc+z3ezl/Z2dn8DwPL1++RFmWmE6n8tpXVYXtdnsQdLcsC6PRCK7rYrPZdMbr67rGcDjEaDTCv/t3/+7B92Ke5/LZJr7DmvPbDvS2nzHHnjPHIuFEREREREREREREREREREREREREXwLGfImIiIiIiIiIiIiIiIiIiIiIiIjoJF3/0z83FEFKADLoZ5omJpOJ/M8wDIRhKOOQQlcQ2Pd9/Mt/+S/x/fff4/r6+uDzze10XYfjOBgMBjIKK4KwbbZty0Dwdrs9CMeKSCMARFGEKIpkLPJU/LR9DqeI44xGIziOg8Vige12K0Oz4/EY6/UanufhyZMnqKoKd3d3DwKxzWjjqYhu1/FFtFjERo/FkZvnJM5fRJM9z0NRFEjTFKZpdgZsm45FkF3XxdnZGQzDwO3t7YPY5KmgrxiniPb2ej30ej3EcYwkSWAYBqqqksFMz/NwdnZ2cH7N47THqaoqXNdFr9fDcDiE53lYr9e4ublBGIbymCJgmqYpyrLEfD6X+xLrs9/vYzKZIE1TGRc9dY3EXIs4rOu66Pf7UBQFs9kMWZbh17/+NS4uLrBarZBl2cG8dsWB2+fYjjSLn+PxGJ7n4fXr1wjD8EHUWszNYDDA+fk5qqrCfD6XIV0RTrZtG77vy8Dvfr+X95thGLi6ukIQBHj16pWMlPZ6PZimie12izAM5bFEyDhNUxm8VlUVpmni8vJS3iftcxFRWgBIkkRGgC3LwmQyge/7uL+/R5Ik0HUddV3ju+++Q5Ik0DTtIAo9n8+R57lcs4PBAEmSIAxD+azo9XoyGO37Pnzfh6qqSNMUm80GlmXB87yD5+axNS7Ow3EcXF5eyrjs9fU14jiW56IoCnq9HhzHwXfffYcwDOX+Li4uYFkW3rx5gyzLoGkazs7O4LouwjCE67qYTqcoigI3NzeIogjAu+f6cDjEYrHA3d2djKk3g9zNMYr1MJlMkCQJZrMZ9vu9DAoHQQBd1/HDDz9gPp8jjmNomiY/v1qtUFWVjHFvt1sAwNnZGYIgkPO33+/ld4u4t2zblmvuxYsXeP78OX73u9/h9vZWjtN1XfzmN79BmqaYz+cIggDT6RSr1UqGkG3bxnK5hKqq8rj39/e4urrC1dUVrq+v5bWtqgr39/dYr9fQNA1//dd/DQAIwxBhGCLLMgRBIGPV6/UaZVnKuLKu65jP5/L7xbZtPH/+HGEYIo5jGTkOw1CGicW9KILSw+EQvV4P+/3+4JoLYu4nk4kMIQM4mL88zzGfz+E4DhzHkc+n9rU+9l3TfnZ2jeHU9xERERERERERERERERERERERERHRXyLGfImIiIiIiIiIiIiIiIiIiIiIiIjoUV2hVuBPEU7gXUgzjmNEUYTlcomqqk4GclVVheM4sCwLlmUdHE98Ttd1Gcf0fR+O40BVVURR9CD8qygKHMfBxcWFjDbquo6iKKBpGgaDAb7++mv4vo/dbofFYoH1eo3dboeiKE4GZcV5dEVS29tpmgbP8+C6LvI8lyFGEVh89uwZJpMJXNeFaZr4/e9/L+Olx+a8axxdwWFFUWDbNoIgQFmW2O/3SNP0IH7qOA5834eiKEiSBFEUIcsyeTwRsO33+9hut9A0DaqqdsaG28dvj1lVVfR6Pfi+jyRJsNlsDq7bsXUlNKOquq7LEHMYhjJYnGUZsiyDYRhwXReGYWC9XsuQMQCYpgnP85BlGaIoQl3XMAwDo9EI4/FYzpnjOFAURUYxgyCA53koy1LGSJvjtywLw+FQhnzH4zF2ux2ePn2K2WyG9Xp9NIIs5ubFixe4vLyU8c9vvvlGHitNU2iaJq9B15pv/17X9UGct319VFXF+fk5kiTBYrF4sE9xTXzfx8XFBXRdx/X1tYyVKoqCLMuw2WxgmiZ0XYfneTBNE7PZTO5vNBrh+fPnuL6+RhiGKMsStm1jNBqhKAqEYSivkQjyWpaF2WyGJEkOQrf9fh/fffedjNU270Xf99Hv9zGfz5EkibzPRGxX3INiG8uyZNg3z3Pc3Nzg66+/xmw2w83NDRRFwVdffYXpdIrlcombmxtst1uUZQld1+G6rjy2uOYvX77EarVCURSYTqcYDAYy6KxpmpzboijkfSPOz7IsPHnyBP1+H7e3t7i9vUWapvIeF/Pz/PlzbLdbGRwW5/7rX/8a33//vXwmep6HyWSCOI5RVRXOzs5gWRZevXolnzPiHvd9H7/73e9kYPtYgFo8V0TY+f7+HrvdDnVdQ9M0GXWO4xivX7+W0Wtx/97c3CBJEliWBcMwsNlskGUZBoMBLi4uoKoq7u7u5Dw3n1fiXrEsC+fn53jy5An+43/8j7i/v5fXQdd1vHjxArqu4x/+4R+QZRl+85vfIEkSvHnzBqZpYjQaybU/GAzQ6/UQhiGePXuG8/Nz+T3QDBXneQ5N0zCdTtHv9/H27VusViskSQLTNDGdTqGqKtbrNdI0lcF28ewUAW4R5rUsC99//z1s24brutjtdjLoe39/L58VqqrC8zz57Nnv9yiK4sG9qqoqfN+H67q4ubmR91MzZF4UhQwaB0GAKIqQ53nn95l4PnSF49vf411xdCIiIiIiIiIiIiIiIiIiIiIiIqIvBWO+RERERERERERERERERERERERERHRUMwooNCOAwLuIr6Zp8H0faZpivV7LGGpzm6qqZIBVEEFZTdNkXLMrKqsoCvI8R57nMAxDRgub8UnLsjCdTuE4Dm5vb2UsVFVVDIdD/Pa3v8V0OsV8Psd+v5dx2CzLkOe5DCW2z7krqnvqb8uyZCw3DEPEcYw8z7Hb7VCWJYbDIUajEUajEZbLJe7v7x9EGj3Pg2VZiOMYSZIcDSI3r4UIjZ6dnUHXdWy324NzUVUVg8EAL168wPn5ObIsw+3tLe7u7mRcE4AMrpqmiaIoYJomqqpCkiSd53uMoigysFvXNdbr9UGkVdM0GIYBXdehKArKskSWZfLa+r6PFy9eIAgCbDYbRFGEs7Mz2LaNOI4BvIv6pmmKqqrgOA4GgwE0TUMURTIKqmkaLi8vMR6P8fr1a8RxDMMwcHl5idFohDRNcXd3h7dv38LzPNR1DV3XcXFxIcPA6/UadV3D931UVQVd11GWJS4uLjCZTLDf7+G6LpbLpQxnPnv2DKqqYrlcyrltxkl7vR6ePn0Kz/Ow3W7hui7KsjxYC4qioCgKGYEV17MZdgbwIDLcXB9irsVxdV1Hr9fDzc0N0jSFqqooy/JgO3H+ruvi7u4Oi8VCBmTFfkW42zAMTKdTAMB+v5fbX11dwbZtzGYzGYudTqdwXRe3t7fyNbHmBoMBoijCYrGQ41EUBZPJBAAOwsPiPhBxVsMwZLhanK/neVBVFfP5XEZVAaDX66GuaxkVzrIMnufh5cuXyPMcT548wdOnT/H27Vu8fftWhoBN08R4PJZrpigKvHjxAkmSYLlcIs9z2LaN6XSKzWYjI9rN50fz+olxPn36FKPRCPP5HNfX1wdhcbF+z8/PMR6P8W//7b+V52IYBn71q1/BNE0Z+NV1XT4v4ziWcev5fI71ei2P6fs+xuMxVqsVNpvNgzh4e551Xcd4PIau65jNZjLkK545g8FAzrW4rqqqIggCqKqKPM/R6/VgWZZ8JorArnhONp9BgvheqKpKBnW/+eYbzGazgzkaDocYj8f44x//iO12i+l0Cs/z8Pr1a6zXaxiGgbIsUZYlVFVFURTYbDaYz+fo9XpQVRV//OMf5Tpvrj/XdTGdThFFkYz8KoqCfr+PXq+HKIqw2+3kfXV+fg5d12X8GngXEx8Oh8iyTAaXoyiC53kIggA3NzeIokjeC67rwrIsKIqCNE0P4vXN9STmsCgKLBaLzth8VVXIsgxpmsI0TRiG8SCWLJ4l7eve3ldXuJcxXyIiIiIiIiIiIiIiIiIiIiIiIvoSMeZLRERERERERERERERERERERERERCdVVXUy4KooChzHgW3bCMNQxjPFe83PdW1bVRUMw4CmacjzXO5vNBrJ0GJRFHAcB67rIk1Tuf9m+Hc6naLX6+H+/l6GcwFgPB7j6uoKmqbhn/7pn3B7e4uyLOW+NU2DqqqPRgmb73fFEIF38ckgCOB5HgAgyzIZcKyqCvv9HkmSIIoiGeEUkVzBMAw8ffoUmqbJ4GrzPEWYsjnHmqZhMBjI8Ont7S32+72MM2qahslkgouLC/i+jyRJsN1uUde1jOOKYKyISYooZxAE2G63MpQqjtk1X+3r7boufN+X4xXrxHEceT1FsHK73eL+/h5FUSAIAvzt3/4tiqLAq1evsNls4HkeXNfFV199Bdu2kaYpVqsV8jyXgdLhcIjVaoUwDGWg1PM8XF5eoixLGY5+9uwZJpMJ5vM5ZrOZjGWu12u4rivn6f7+HovFAlmWwbIsuK4LADLo6TgOiqJAHMdQFAXff/89VqsVPM/Dr3/9a1xcXCDLMmw2G5RlKSOnw+EQT58+RV3X+Oabb+QY1+v1wTWzbRvz+VyGYYF3Ac9er4fdbockSVBV1YPAdnN9ivfEa5ZlIQgC/O53v5NjEsfU9Xf/tFhESdfrNe7u7h6EfIF3z4WyLGHbNp48eYJ/+qd/kqHZfr+PyWSCzWYjz8kwDDx58gRxHGO73crniq7rCIIAlmXhzZs3B+tdBLpfvXp18LrgOA4mkwnW67W85qqqwvM82LaNLMsQRZHcLk1TvHnzRl7L+XyOIAjkPel5Hl68eIHNZoObmxsZfg6CAIPBAK7rIooibLdbDIdD6LqOm5sb+Xw6Pz9HWZa4v78/mHNxjZpzqOs6JpMJxuMxttutvNebzxQR4L66usLNzY1cR6Zp4unTp/jqq6/w+vVrGVH2PE/eb1VVwTRNGaEtigK6rsuAbJ7n2Gw2BwH1rnC5oijwfR9BECAMQ3ntTNOU96Su60iSBGEYHgRhHcdBVVXy3lwul4iiCIqiIAgC+bwSgd9mVFqsf9u2Zdh9tVrh7u7u4Bi9Xg9fffUV8jzHYrFAr9fDxcUFoijCer2Wz600TWHbNnRdRxzHcl2LYLWYo2YU2zAMXFxcwLIsvHr1CrvdDgAQBAHOz8+hqqo8Z3E+rutit9vJz4r5EzFk27ax2Wzkvquqwm63Q1VVcF0XrusexLlFbL7r3g6CAL7vY7lcIo5jef+2g9Cqqsrv2uZzofkcbz5D2nHn5jbt5wARERERERERERERERERERERERHRl4gxXyIiIiIiIiIiIiIiIiIiIiIiIiJ6L8fCvJZlod/vA4AMjLa3UxQFhmHAcRwAQBRFKMsSaZrKwKFpmjKuOJ1OYZqmDFcOBgN4ngdFUXB9ff0gbhgEAUajEbbbLTabjYwRDgYDPH/+HFVV4ZtvvsF6vUae51BVVUZg2+HCU+Fi8X47ZCj+1nUdvu/DcRwkSfIgxChilaZpQtM0hGF4ELAEAN/3MRqNZES2ue/pdArXdbFer7FcLlGWpQzWPn36FJZl4YcffpD7FaHfi4sLfPXVV9jtdvjd734nA6eDwQCqqsrYo6qqME0ThmFgu93CcRzouo79fi9jj81Iafv820QEtKoqTKdT1HUto7rb7Raz2QxpmqIoCuR5jqqqYNs2/v7v/x6GYeCbb76R0eEgCLDZbPDNN99guVxiuVzK8/Q8D5PJBL7v4+XLlzKKKsK5tm3L6OnZ2Rl+9atf4eXLl3jz5o2MIovrEwQB+v0+drudDOmqqgrbtjEajQ7WXxiGcF0XQRBgv99jNpvJa353d4cXL15gPp9ju93KtdPv93F1dQXTNPH73/9ehoBF1FNEa03ThK7r2G63MrRsGAZevHiB0WiE3//+90iS5EHIt3k9RBC2+ZrrusiyTEZIRVRXrAHDMHB5eYmqqrBareQxAByEhkXMVcRjV6uVXIviXn379i3KsoSqqhiNRjBNU17zqqqgqqqMoBZFcTBPuq7j7OwMuq5jNps9COJaloWzszOoqorNZoMsy6BpGjzPk9c8DMODsGmWZbi/v8dgMJAR5n6/j9VqhbIsMZ1OMRgM8Pr1axmc9n0fruuiKAosFgtst1tYlgXHcRBFEXa7HXq9HgaDATRNw5s3bw5C2sei1yIIm2UZ5vM54jiW26iqKgO2T548gaZpuL+/R1mWMsz9q1/9ClmW4c2bNzKY2+v1YNu2DPcahoH9fi+jy+I5K0Ld7UBy11jF81fMc5qm8tqbpomqqpBlGXRdR1EUB+doWRYMw5DbhmEoI99nZ2cIggBpmiJJkoNtAciIcq/Xg2maWK1WePv2rbz3DMOA53m4urqC67p48+YNyrLE+fk5+v0+fvjhB0RRhKIo5HNezIGIBzuOI591IoAsxm4YBs7OznB2dobtdovVaiXneTKZyLhxURQygK6qKsqylHH0uq5hGAaGwyE8z5PPLFVV4bouTNPEbreT41IUBVEUybB01/0svqtEBLsoCqzXa/nMsCxLfvcA7wLPZ2dnMgwuvnub+27+J86/az00vyfFc4NRXyIiIiIiIiIiIiIiIiIiIiIiIvoSMeZLREREREREREREREREREREREREREeJeF877teMuXqeB8dxEMfxQShQBCl934fv+7BtG0mSYLlcyjBnM5RqWRZs28bZ2RmqqsJyuYRpmjK+WJYl+v0+iqKQ4UcRNRyNRiiKAvP5XEYMbdvGr371K6iqiu+++w7L5fIgGCnGKX4eC9Seivy2Q8C6rsN1XRiGgSiKkOf5g0ilpmmwbRtlWcpIbnO+XdcFAKzX64PYpuM4GI1G0HVdzqGYZ8dxEAQBZrMZ1uv1QaxxOBzit7/9LXa7Hd68eYP1ei2jss3Qa1VV8nqJIKht2zLQ3BVtbAZL20FkXdfR6/VQ1zVmsxkWi4Wcj6IoZEi2uV/DMPC3f/u36Pf7+Pf//t/LMLOIJCdJgt1uh9lsJq+nCMIOh0NsNhsZ3RRzeXFxIWOzpmniq6++wna7xc3NDdI0fbCOzs7OEMcxbm5uEMexHJfneUiSBJvNRm4TRZGMan733XfyWhdFgdVqJQPDlmWhLEs4joOzszOYponr62uEYQgASNMUq9UKaZrKe6LX66GqKuz3e6iqKiOu/X4fVVWhLMujcWVxXZrrWOxD1/WDgLCYK/Ffr9dDr9fDdrvFdrs9uMZiX5qmod/vy4CtiLyK8K1pmkiSBOv1GgAQBAEuLi5QFAXCMESe5zL8O51OEQQB7u7u5PyJ+O+zZ88eBIWBd4HZ0WiE8/NzGeUV11usX+BdNLw5/rIsZVy4eW6LxUIe03EceW+IdbrZbLBerxFFEVRVRb/fh+M4yPMcvV4PhmHIULCIJHfdL+I10zQxmUxgGAbm8zk2m428Ds0gsWEYUBQFaZrKCLlY66qq4v7+Hvv9Hr7vwzRNGcYWEWdFUWSMV5z7breT1/BYDLo5XsuyZAC6KAq4rivDsyJaK9arruuwLEvGt0XoebfbIc9zWJYF0zTR7/cxHo8BQD4TmmMQsdvBYADbtmXwOIoiKIoC27blNdJ1HbvdTt73tm1ju93KcLu4HwHIdRGGIbIskxH65joRz2exllRVxXq9Pjh3MTe6rmMwGCDLMkRRhDAM5fsiYhwEAQaDAQBgs9kcjGG1Wslnwm63w3K5xH6/R57nMAwDVVXJ2H3zmQlAvr5erxGGIeq6RpZlODs7w3Q6lfsQ1+D29har1eogLt8VnG4+x9tO/X8AERERERERERERERERERERERER0ZeEMV8iIiIiIiIiIiIiIiIiIiIiIiIiOqoZ+ewK+KmqKoOH+/0eWZbJ8Gq/35dB0jAMcXd3h/1+fxBuzPMcWZZhMBhgOp3KcO/9/T2SJMFkMkFRFEjTFJqmwTAMGbgUY3EcB4PBALPZDPv9HlVVQdM0XFxcYDAY4B//8R8fhHyrqkKWZVBVFXmeH8RPxXlpmibjmiKCqqoqkiRBkiQyvtlk2zYcxwHwLtAqPtfet+/7iOMYSZLI86jrGrZt4+LiArvdDpvN5iCMLCKpURQhjmM5h4ZhoN/vwzAMrNdrGTMGAMuy8OLFCxRFgZcvX2K73aKua/R6PVxcXCBNUxnaVBQFlmXBsiy5fZIkMpLZDn6KAHBzDprXxXVd6LqO9Xotr70Ih3bFQzVNw3Q6xWQywTfffIP5fI6yLA+CpMC78OpsNpNzJ8KeiqLg5uZGnr9pmjIUe319jSRJMB6Poes6/vjHP8rtm+toMpkAAGazmYy9imP3ej1sNpuD+RXjFrHg5jxVVYUoigBARk6HwyFM08R8Psf9/b08vzzPsd1u5byKWG2WZTK+adu2DLaK/YtjdcU42++J62WaJrIsg2EYB9uJmHOv1wPwLjwq1lnzcyJeLEKoiqJgvV7L0GlVVfA8D2EYIo5jmKaJXq8Hy7IQx7EMVIt5HQ6HMvIrIrRBEODrr7+G53n47rvv5HFEIHw4HMrYrYjn+r4PXdflGiyKAlmWPYgQm6YJANhutzAMA0mSHBxbBJVFvFeMWTy3HMeRgXIRcd1sNlgul0iS5CCkLbTnUIRioyiSEee6ruX8D4dDlGWJzWYjo62GYcg4tDi/MAwRBAE8z8Nms0GSJBiNRgAgn2+macJxHKxWKxldFrFlEaturp32vSmC6iJeWxQFkiTBdruVEWUxzyLoK+ZaxLjrusZgMJDz6Ps+DMOQz0BxXUSwVtzTnuehrmtsNhtst1soioIgCBAEgZz7Z8+eYb1eI45j6LoOTdOwXC4PAr3N+6cZqRZRbPG9JUK+uq4jSRJ5XiIkr2kaoijCer1Gr9eTkd3NZoPNZgPHcdDr9eB5nnyWiBi0ZVnyO2S1WsnvS7GOwjA8iOHHcYzdbiej2bquY7/fy/PSNA11XctAr6IoKMsS8/n8IPp9c3Mj10ZRFCjLUs5L+9ktAtDttaCq6oN10fwsERERERERERERERERERERERER0ZeGMV8iIiIiIiIiIiIiIiIiIiIiIiIiem8iFCoYhoEgCFBVFfI8R6/Xk5HDsiyxWCwQhiGSJEGe550R191uh7/5m78BAHz77be4vb2VgUcRFFRVVYY6dV2HYRgy8BsEAXRdx2azkZFCz/NwdXWF5XKJxWJxEPIF/hRXBQ5DheI4g8EAo9EItm1DVVWoqoqiKBBFEZbLJdI0fRAy1HUdnufJQHCapjLEKmiaBt/34TgO7u7uUJblwbFHoxE8z8Pr168PxmyapowmR1Ekg7IirDoYDJAkCXa73cH+xuMxgiDAy5cv5Xsi/qvrOna7HaIoQlEUUFUVlmXBMAyUZYmyLLHb7ZAkiQxytq9/8+/mT3GeAGTUtXmuzTGKbQaDAa6urnB/f4/r62t5/iLg6TgOdrvdQahTzLkIo67Xa9R1LdfF+fk50jTFfD6XoeQwDGXMs3ldBoMBXNfFer3Ger1GWZYwTRPD4RDn5+coy/JgOxEAtixLhlwFET6t61oGXkXguigKLJfLg7UhwrxiPj3Pg23bWC6XMnyrKAqSJEGv15P7FAHOrthye32KmK/4rOM4ctuqqmSc0/M8ZFkmr5nYVlEUGZsdDAYyCioCrOIaiCDqYrGAqqrynqjrWq41sZ5d10VZlkiSRIaPLcvCeDxGr9fDarWSsV6xrsbjMSzLkuujKAr0+33keS5j3p7nyWdScz5UVYXjODJk67ou4jiW91MYhjK0PJvNkKbpwbWxLEvGdnu9HmazGdbrNebzuQzyinWtaRpUVZWx2Ob1FXMfRZG8vwzDgO/7sG0bu91Onrfruri6usLV1RXyPMfNzY0MXF9eXmIwGOD6+hrb7RZ5nmO9XsNxHHkudV3LsLiYszzPEYbhwZiPEdv4vg9N07BYLGToWaxBcSzP8+T+d7sd4jjGcrmUAek0TeUzq65r7HY7bLdbGSQWke6zszOMRiN5/ZIkkfFn0zSxXC6x3+9hGIacY8dx4DgOrq+vZSxX3KviupdlifV6jSRJ5PyL+940TZimiTzPsdlskOc5BoMBNE2T4WoRJVcUBbPZDJZlYb1eY7vdIssyFEWB29tbuK6LoihQFAX2+71cx+I7ZL/fy1iz+F18XqyRLMuwWq3kfSHC3mI9ZlmG29vbg+C2oiiIoghv3rzB7e0tAMhnubhfm9e7Geg/Fllvvt78Pm6/R0RERERERERERERERERERERERPQlYcyXiIiIiIiIiIiIiIiIiIiIiIiIiI5qB0GbwdZmZDXLMhlvFTFEEfA9FvEVQUpN0wC8CzYuFgsZahWRRgAyZhpFkQw3xnEM27YRBAHiOEaapgDeBYaHwyEsy8L3338vQ52Cpmno9Xo4OztDkiTY7/cHMUQR+nUcB4ZhyBhmGIYIw1COT3xeME0TrusCeBdeFPHFZghR0zQ4joO6rhGG4cG8OI6DwWCA/X4vxwS8C1G6rgvf91GW5UFkVcRVbdvG3d3dQbzUcRxMJhNEUYTVaiX35zgOfN+XoUkREzVNU8Z8RXizOY73WR/isyJMqqoqsizrDPk2fzdNE9PpFIqi4O3btzKMK8Knk8kEdV0/iHRaloV+vw/DMDCfz2XYUoSRTdPEzc0Ndrsdrq6u4HkeXr169SDubFkWgiBAlmUyzClC1aPRCI7j4P7+Xs6VGNdwOERVVVgul/KaNMctopzAuxiqrusyRCoCteIaiwC2qqqYTCYAgO12K0OjImiq6zriOJbnINbusUinII4nos4iGCzuKbGNrusyaq3rOqqqgqZpME0TnufJ+z3LMgyHQ3mP3N7eYrfbYTgcQtd1aJqG8/NzJEmCLMvkfR4EAUzTlGMSkVEA6PV6CIJAvifOU1VVmKaJfr+Puq4xm83Q6/VkhLkoCmy3W+x2O/lMqarqQaxWBHa32618XZw/AMxmM7iuK+83EWRuPxfyPEcURTKQLN4X4W8ReRbPwmY4WsyxmAPDMGDbNjzPk2FcET1WFAWLxQKO4wAAlsslNpsNdF2H7/uoqgovX77E/f29DA/PZjNomibntfmfIJ7Lbc25EmsniiLMZjOEYYjdbidDt837OUkSzOdzubZESLkoChk+F58/Pz+Hbdsy9Cuu2XQ6hWmaqKoKtm3LCPz5+Tmm06l85ottmut5PB4DANbrNcIwRJ7n8pqK7xkA2Gw2WK/Xci72+z3W6zVc15WB5zAM5T273++x2WxQ1zWWy6UMJgPAzc2NjLwXRSHPW8SOgT+FdAHItSheE/dsc9v2dRDfIeJ+EN87zedqezuxP7F+xOvNz3XFe9vPZzG/zWdI+/PtdU1ERERERERERERERERERERERET0pWDMl4iIiIiIiIiIiIiIiIiIiIiIiIg6KYoCVVU7Xwf+FJIF3kUPRQQxTVNkWSbDg8eIgONoNMJ+v0dRFJ1hQhE6FQFcAHBdF7vdDrZtQ9M0GZgE3oVZe70edrsd1uv1wRhUVYXneTg7O0Ov13sQGq7rGnmey6CrqqqoqkqeV1EUDwKKYj5EaBX4U8y3HTsUUdQ4jhHH8cF7vu/DdV0sl8uDbVVVhe/7ME0TYRgeBGtFALgoCqxWK7mNoigYDAZwHAeLxUJGaAGg3+/Dtm2sViskSSJjk5qmwXVd6LqOzWaD/X7/IEYsNMOSYozNORTnWRTFwVoQccj2unBdF47jYDabyXimoiiwbRsXFxcYj8fYbDbY7XYyYqtpGvr9PkajEcqyxHq9Rp7nUBQFnudhOp0iSRLc398jz3P0ej0URYEwDB9cF8uyZDA3z3PYtg3DMGBZlgza7vd7lGUJRVFgWRZs25Yx3GaYWUSbfd9HnueI41ieizh+M6QqPm8YBoqigK7r6Pf72O12MtoqIqIiZtsMSh+LcrZ/F9dtu91itVrBdV0Mh0O5XhVFgWEY8noMh0OUZYk0TWV8VlEUJEmCJEmgqirquoZhGDJ62px/RVGwWq2wWCzkOhBrLwxDrFYr+dnhcChDp9vtVgZXHceBqqoYDAYYj8dI0xSz2UwGYsUx7+/vsVqtZEBWzG97nhVFkSFeXddlMFast81mg/l8jsFggH6/LyO6Yi7LssRut5MR7H6/j+l0Ctd1ZaS1GWeN4xhZlsn9i+uRJImMHbuuiyzLEEURNpvNQVwYePds/eGHH2RUXARj/8N/+A9yvM3nRZIk8lzFMZv7+1BZluHu7g6qqiLP84NnTHNu0jSVz4tmBFY8L1VVhWVZGA6HKIoCURRhu90ijmPMZjOUZSmf58vlEovFAmVZIssyGbCOokjOj6IoKIoC19fX8h4R3yPNtS8Cy1EUIY5jJEkizyGOY9ze3kLXdZRlKSPAYtvdboeXL1+iLEskSSKvrQgYi3loBnBFyLgdez8WNG/OofjZftaK18XvzVBze3/iu6H5+fZzoh1tbkZ/u8bXvObHAsFEREREREREREREREREREREREREXxLGfImIiIiIiIiIiIiIiIiIiIiIiIioUzvW2PxdxEld10Ucx1iv19hsNjLIeywKKPap6zomkwkGgwFWq5UMA4oYrvisqqrQdR2GYSDLMhmqFK/ruo40TbHZbGTg0XEc2LaN7XaLNE0PIqu2bSMIAnieh6qqZByzGSkUwdX9fi+jiO0wYtdcifCrCLiKfTc/YxiGnLMsy+R7hmFgMplAURRst9uD+KamaTKOGkWR3E4EY0XIUgReAcgAKgCs12sZf3QcB+PxGEVRYLfbyQioCCYHQSCjps3QbztYKeajHYFsX7c8zw+Cqs19iH3quo6zszPUdY3lcik/b9s2zs7OYNu2jNA2w6siymzbNu7v72Xo2TAMjMdj9Pt9/OEPfzgIQO/3+wfXRaxHER/2PA+2bSPLMrmey7JEHMfyvCzLQlVVMAwDm83mQZhZxH7n8zniOJbhX7Eu2scOggBFUSBJEhnJXq/XGI1GMqysKArOzs5wf38vw6LNQOj7yrIMb968wZMnTxAEAYDDAKyICJumiV6vJ2O0ZVliv99jv9+jqipomobZbIbdbofb21t5XmmaIk1TGWkVoePb21t4nocsy7DZbBCGIVRVlSHXNE2x2+1kzHk2m+Hy8hLPnz+HqqrYbDZYLBZyrqMowrfffiuvjVg3Ir7dnpe6rpFlmbwf6rrGdrtFURRyLWZZJudXhI0ByHUsIt1izuI4hmEY8rMiWiyC5iIA3haGId68eQPLspAkiZxXcc81g64iJN1cs/v9Hi9fvnwQnxXn2Q7CfozmM1yEa4Xm81L8DfwpMqtpGjRNk/Mq9tXr9RAEAeI4lvdzURTyWW1Z1kEwWlEUGW5urvfmOYq11IzBN59PIsS7WCxkjFi83wziNucOePcMFde7a37bIfP2d96pIK74u/38bP/d/Ez7uoj5PnZd2pHvru2bx+zS3Je4loqiPBrqJyIiIiIiIiIiIiIiIiIiIiIiIvpLx5gvEREREREREREREREREREREREREXUS0T7xe/N1AHAcB7quy/BtM5zb3qZJVVWMx2NMJhPEcYzNZgNd12HbtgxAim1VVYVpmtA0DWmaYr/fw7ZtAO+CqY7jyAiqIPYThqGMHaqqCsuy4LouVFWVYchmzFeo61oGPsV/j8UxRahX13UURYH9ft8ZFBVh4uVyeRDKHQ6HmEwmWK/XiOP4IBTpui5c1z0IWorzHA6HsCwLYRgeRHM1TZPRzCiKZDR3NBqh3+/j7du3cn5EiHgwGCAIAhlVbYYijwUqm9e5HaS0bfvBemhGkZvjvLy8xGKxQBRFctt+vy/ncjQaIY5jGUa1bRuO48A0TVRVhTAMkWUZVFWF7/sYj8dQFAWLxQJFUUDTNNi2jeVyeRBRFqqqQlEUcr2FYYgoimCapgw6F0UB13VhGAbyPJdx6NevXyPP84NzVVUVYRhit9shz3NYliWjn814tGma8H0fqqpitVrJmLAIFc9mM6xWK2RZhl6vB8dxjgaJ35eI2Iq5AiDnUcRsxT0JvAvUipBtM05bVRVub2+hqqoM7CqKgjiO8f333yOOY6xWKzlX8/lchnRFRFpcIxGrFRFcVVVxc3Mjjx+GIdbr9UHUNs9zrFarzvuyGTJtx0/bMdfm86aqKrnOmjFaEXRuxl/F80Nc2zzPZcxX3KfN0G7zZxzHuLm5kc8LsW07Gtscd1NRFJ2R4PYcfE7N74b287EZ0xXjMAwD/X5f3l/i+SPmVESZxb7FPsV5NoOy7eO056I5HhEjFs/jZpy9PfZT0d7m68fG056PU8/N9jHa231MkPmx4zw2BjH+Y/dU1/kSERERERERERERERERERERERERfYkY8yUiIiIiIiIiIiIiIiIiIiIiIiKik9qRRuBPkVlN05BlWWe49hjXdXF1dYUkSTCbzRBFEUajETzPk/FUEXNUVRWapskgY5IksCwLuq7DNE04joP1ei1jjqqqyv/yPJdxVsuyYJom6rqW0WER3mxGHbuih11B4645sm1bxo2b4dnmZzRNg67rBxFh3/cRBAEcx8Ht7a2MeiqKAtd1EQQBDMOQ0UoR5vV9H4PBAKqqYrvdHsR8gXfxzO12i6IoZPh4MBjAsixEUSQjq5ZlIQgCBEEAXdexXq9lALgdpGyGj0/Nh4gEi+Cu2F7sU9M0GIZxcO5JkqCuawRBAN/3UZYl1us1giBAXdcH+/R9X649VVWRpilUVZXn2AyGinUkrmMzgivOIUkSzOdzAMB+v5fBXF3XUVUVVFWF53nQNA15niPPc/i+L8Os7bCuiOKK98qyRFEU8pyTJIFhGDBNE4qiYLvdYr/fy+tWVRXyPMdsNkOSJFBVFUEQIM9zGQhuxjc/NLBZliU2mw3iOIau6zJwnee5DMeK18Xci3uyuS7EOlFVVb6Wpilubm4OYrbi/hXroDn3IuDbXGN1XWO9Xss1Ie6nZhxXaB67vf92PFpRFLmOjgV/xdw376euZ4MI8EZRdLA/Eac9FVtuhmv/HOOo7ehtO/be9axwXRe9Xk9eT3FvnArmNvfZdc2a4dmuGG3zujW3OxXoFfvoiry315ZY+23N9diOGp+ax2Ovtffdpfl6M1LdHN+ptXYs4Ns+n8fGQURERERERERERERERERERERERPQlYMyXiIiIiIiIiIiIiIiIiIiIiIiIiDqJGGA75Cciio7jyHinCIB27UOEd7MsQ1mWMk775s0b7HY7lGWJ7XaLJ0+ewDAMGR8UxzEMA2VZIkkSeSzf91EUBWzbRpIk8vgi0qtpGhzHQa/XAwAZfk2SBK7rQtd1JEnyaIT4VLSyfZ4iGJzn+UGst7kPMR/iPdu2MRwOZaC3KAoZATYMA71eD4qiHIRFNU2DbdsIggCe56EsS0RR9CAemue5jMfqui6PKz5nmqacJ8uyALyL0Ipw6mOBzeZrIvoqzrEoCux2OwwGA4xGIxlHVVUVhmFA13W5tvI8RxiG8H0fl5eXMvy63W6RJAls20ae57BtG4PBALquoyxLpGmKJEnkuQ0GAziOA8Mw5PHFXGqahjAMYdu2XDMiKgy8i6suFguUZSkjreJ1Eby1LAtpmmK/36OuaxkMFlHWpizLkOe5nOuiKLDdbuH7PobDIZIkkXFmEQ8WYdAsy/Dy5UvMZjOsVit5LN/3sdvtEMfxjwr5CmKtifh1O0BbFEVnLPXYWmhu134eiH10hbGPRUxFJFwEX8Xr7Rhwe913hV+b+22HZ9vHFmuiud+uKOyx+GnXcbv8OUZ8T+mKf4uYrKqqMiQexzHCMJRR9cfmqxnNfZ+AbPMaP7bvY/tr31/teG/X90LX/SHW0PuEhE8Ff9vjPPbZ9znv5v7aP//S1iQRERERERERERERERERERERERHR58KYLxERERERERERERERERERERERERF1ElHdpmbszzAMVFUlg5ttIvjr+z4URcFyuZT73Ww2WK1WMpoqwquWZT2I+ZqmiTiOkaYpsixDURRwHAeapkFV1YO4aVVVCMMQ+/1eHjeKIiRJgv1+j6Io4Pu+jKbmeX40YFhV1cEctCOV7bmyLAuapqEsy84Yrti2qio4jiOjxCKILGKqVVVB13U5zv1+D8uyEASB/LxhGHL+4zg+iP2K46xWK/T7fUwmE2RZhv1+j81mg/F4LCPH4vxExFfXdRnZbJ9z1zy0r7d4vSgKzGYzaJqGIAhkTFlsUxQFoiiSEeI3b96g3+/Dtm2s12usVitkWQbgXZR4t9uh1+vBsizs93tEUSSPAwCDwUDGaYuiQFVV8jNiTOv1GsPhEKPRCLvdTsZBi6LojBgrioIsy7Ber5GmKaIown6/R5Zl0HUdWZZhNpsdzNexiGye51gul8jzXEabkyRBFEUH909d19jtdnj58uXBeERMuBko/ljt6ydiuafin6eiuF1h0Xb09ljstPm5dvT6WOj1VNC1GXw9FUntWt/iXtA0TR63eV3aYzkWcf2Sg6jHzt0wDIzHY+i6jsViIddxM9Tc3MexULKiKDKE3nWtgfcL03btvx3C7VqDze+mY9u0vwu71mHXGv2Q8beDxV3n0g5xt+e6fcxT91l7vr7kNU5EREREREREREREREREREREREQEMOZLRERERERERERERERERERERERE/0xRFA3A/xfA27qu/5WiKF8D+DcAxgD+GwD/k7quM0VRLAD/DwD/LQALAP95Xdff/0zDps/oWFRRhBSBd1HWPM8PAoaKokDXdXieh16vB13XsV6vZXB2u90iz/ODAG2apthutzBNE5qmydd1XZfvp2kqw6ti/7PZTEZfhf1+j7dv38LzPCRJgt1uJ8OomqbJmGqe5yjLUo69GT8U5//YXIj3TNOEqqrI81zGWbsilUVRIE1TjEYjWJaFoiiwWCzQ7/eRJImMAjuOA8dxEEUR4jiGaZrwfV9GePM8R1EUCMNQRmKbxynLErPZDMC7WOtut8N2u4Vt29hsNnAcR4aAwzCU2xdFgf1+fzTWeCxWKwKb4v08z7FarVDXNYIggGEYctwirJwkCYqigKIouL6+xmq1QlmWiOMYZVnKYyRJgtlshiAIkCQJttst4jiGYRhIkkQGcqMowna7hWVZ8joIWZZhuVxC13U4jiOvhQg6i0h0cx3XdY0sy7DZbLDdblGW5cG4NpsNqqqSQeiusKd4rSgKuQ5F8LO59prHFNHq5hyLNRuGoQz8fkw4tmsdH9tPV+BTfP7Yfo/dM10h3659dgWz2xHU5vy2x9tcg+2gcPvYzfu+fc4iqt0eV3sszXjrlxw4PRWJFUH24XCILMuw3W5lvLwr4Cz21xW/FdelHVlv32/HxthcU01dz/72WjsW622f/7HXT63Frr+Pffee2o8Yr6qqB7HjLu3t2/PYDBcfC3ETERERERERERERERERERERERERfakY8yUiIiIiIiIiIiIiIiIiIiIiIiLhfwHgdwB6//z3/wHA/7Gu63+jKMp/CeB/CuD/9s8/V3Vd/0ZRlH/9z5/7z3+OAdPn1Y4mAn8K+VqWBQAyLCpomgbTNBEEAYIgQFmWmM/nCMNQfm632z0IxtZ1je12i36/D13XZRAYeBdzjaJIHksEcXVdx2KxQFmWB/vJ8xyz2QzL5VIGWEWA0TAMKIqCJEmQpmlnrFMEE8W5N6OfXSFDMSdxHGO32yEMw87ApxibiOrmeY7FYoEkSdDr9ZBlGVzXxWg0gq7riOMY+/1ehmANw4BlWYjjWMaJi6LAarV6cLyyLLHZbJAkiQzEis/c39/Dtu2DqK7YZr/fH8SI2+HUrvPvCrACkGNcr9dQVVVeu2YQV+xfxJq7ApVinsR5ivOoqgrL5VJeRzFXvV4PSZKgqioYhiFj0+v1GmVZwrIsZFkm14CI8XbFNauqklHg5pirqpJruDn3x4Kix/bVXnMixNwMcYqA6HK5xG63k9Hhj41ptsf5PiFSMY732XfXNo/FfNua8dTmvfjYWB8LjXbFqNsR1zRNHx1f13ZfomPXV1wvRVGgaRp834fnedjv99jv9/IZ8L6xaPF61z3W/ry4f47to/17XdcyIC/229z+sev8PnHrY98dp/bR1h5TMyzcvg4irt7l2L3fjGG3w+RdxyQiIiIiIiIiIiIiIiIiIiIiIiL6kjHmS0RERERERERERERERERERERERFAU5QrAfw/A/x7A/1J5V3L77wD4H//zR/7vAP63eBfz/R/88+8A8P8G8H9RFEWpv9Sa4V+wdsiv+VOEdlVVhWEYMAwDmqbBtm34vi9jscvlEnEcP4i3tpdLVVVIkgS2bcuQqQgHhmGIKIpkyDVJEiwWC2w2G2w2m4P9iZ95niPP8wfxThGUDcNQRmBPhUZPLevme0VRYDabIU1TRFH04LNiHPv9Hm/evIGiKMiyTIaIoyjCfr9Hr9dDVVWYz+dYrVYyNLvb7RBFkZwTRVEQx7HcT9ecithvM7KbZRnm8zkURTmIHIvPt69RVyzysdim+HxVVcjzXO63+ZnmenqfR4e4bu1rFUXRwXyrqiojxUVRQNd15HkO4N2aWC6XcmyPnUvzvLsCls2IdHOdtSOaXWvsWNyzeZzm61EUyThyM4b8sX7M9l3XrrlmuuLPzeO257YrrNucTxGHbV+DdmS06zlw7Fq0o61irTfjwceuZXvs/OrrJuZW0zQMBgMAwHq9RhRFR5+5jz0bTr3XFWnuuobHtNdy+5jHwrztZ95jQekPdWz8Xeu9uU37HMSa7tq+a9/NnyLK3PRjguJEREREREREREREREREREREREREf+4Y8yUiIiIiIiIiIiIiIiIiIiIiIiIA+D8B+F8DCP757zGAdV3XosL5BsDTf/79KYDXAFDXdaEoyuafPz//yUZLP5mu8KiIFhZFAcdxMJlM4Ps+dF2HZVkoigLr9RqbzQZ5nstIZjuA2QwzlmWJLMsQhuHBMcuyRBzHiKJIhkyjKML19TUAdIZsj/0tjiOiuEmSyCBhV4y1HW88FYZMkgRpmp6MxIoYcVcEcbvd4tWrV/A8D2EYYrvdylisCI22w7pFUTwaquwKXortusbRtb/3Ce4+Fs08FcoU5yXin13j6Tp+83ji/aqqEMcx5vM5VFV9EABuXu/munyfKGVX2LUduT4V8uxaF11rVUQzq6qSAU4x7k8R8v1Q7fNuz1lXVLT9vOhaZ13Xr70/QdwLXWusfX2PXZNj67PrXLrizV2B1/b29Cfi2huGAd/3EccxVquVvCcfi9E211k7qt0ML7evb9e9+Nh77bGIoHPzuO8T6e7yvjHh9n7a90jz+X/sWSLe73qWNsfS/m479SwU8yDmSuy76x4hIiIiIiIiIiIiIiIiIiIiIiIi+lIw5ktERERERERERERERERERERERPSFUxTlXwG4r+v6v1EU5T/7hPv9LwD8F59qf/TTasYUhWbQNo5jbLdbTKdTXFxcyBDvZrPBer1GHMcoy/IgntoMBbYDh1VVIYoi5HmOJElkvDOKIiRJgjAMZci0LMuDqOmpkGY70lgUBcIwBICD8XWNqyv0CPwpOtt8rRnebW7f3OepMG1ZllgsFpjP5w/GcSx8emo+u8Kj7e2OBYy7Xu+KaoqwYzug2hVE7oplNkO+zRjvqVBqVzS4vQ6qqsJ+v39wzu1jPxYuPna92ud4LPIpQrztMbdjncfWX3NfzfDyTxmObV/X5rjaY/mQcTXXUdfzQezvsYDqKcfCwuK6iOM3/z527GPvf2io9S/NsWvRfl4WRYGbmxus1+ujUer2fdB8zh5bb8fupXYEtytA3X7/2N8fsuYemwsx1vfZ32PfS137bu+/K1R8bM22vzOOPWePxYSJiIiIiIiIiIiIiIiIiIiIiIiIviSM+RIREREREREREREREREREREREdF/G8B/X1GU/y4AG0APwP8ZwEBRFL2u6wLAFYC3//z5twCeAXijKIoOoA9g0d5pXdf/FYD/CgAURWH17c9MM37YFZbN8xyz2QxpmkLXdWRZJsO7IpILQP4EDiOnTeIYYRgeRArLssRyuZT7ORUzbcdom8dr/l1VFfI8fzCm9jl2vd+OK3aN41i8tSs2+T6x0K79t38/FfJtjv3YcY6dw7HPN99rx3i7Ir7HwpNdx2jPazuseeo8m+faPudmXPfYsdqByvb2zfNr7qN5nu34bXt+uuboVGi06+9TIc9PrT2XzdeOfbZNnF/7Hj61Zpr77Pr9sTF/yOfKsnx0jYpxfsx4/lz9mHXWvOfSNMXbt2+RJAmiKDpYB+1tmprfHe33u54vVVUdhHtPbdf+jjh2DmK/x577j/19ar8f4tjzp3ncrt+7zq/5rGpHtU+N89SzkoiIiIiIiIiIiIiIiIiIiIiIiOhLo/Af0hEREREREREREREREREREREREZGgKMp/BuB/Vdf1v1IU5f8F4P9T1/W/URTlvwTwH+q6/r8qivI/B/Av6rr+nymK8q8B/A/ruv4fPbJf/mO1PzOapgHAQXxXRABVVT342QyXiiigiASWZQkAB+HA9k+hKx58SlfkVEQGmyHVdrSw67PNY2qaJt8vy/IgBNveV/Nc2kHFZoBRbN+MXLYDv22PRXqFZhy5HZasqurBsbtCl6eO1XytGaft2rYdpu2am2P7b18jcW7t+HNXGFns89hxmtfwVIz22O/ienWNuysg2nzv2DGa2zTns309xf4fi21+Lsfu02PXsB0m7orCds1Xc65Oha67xvY556J9nbqO+5f277Hfd/7fZz+qqkLTNPn90KUdtj72XvuZ3r6fxfOuvW3zO+xYHLu5zanXPud9+Nh6bkZ42wHerudw8+/md3MzbN713H2fcZ36/C9BXdfv9z8SRERERERERERERERERERERERERB9B/7kHQERERERERERERERERERERERERL9Y/xsA/0ZRlP8dgP8fgP/6n1//rwH8PxVF+QbAEsC//pnGR59RV/S0HfQToduuCO+xOOP7hlfbnz82PhFvbEZ1xevHYrPt17vipM347TEiitj8fHt/7fNvn2NXkLLrXLuCou19Nj/bHOOx/QGHkcv2/rpCzafG195P+7ybMc7m/puvNWPH4u9moLN5nGPjaYdvu8bUdb2b89F87dg66LoO7XhwM5rZDva212LzXgLwIIzddcyfUnt8be1g92Oh2/a5n/rMqTF9qI8JAL/PM+kviaqqD2KxH6K9jrtCzV3PwmPfD133YNdzo7m/5rk0P/O+1/LYc6PrHNu6guaP7e/U9u0xdT37Tn3fHfs+7PreOTW2U+dMRERERERERERERERERERERERE9CVR+A/qiIiIiIiIiIiIiIiIiIiIiIiI6HNTFIX/WO3PTDPK2RUeFd4nNtoO2jZfb3++GZBtj6V5jHacsHmMqqpQluXBNs1I67GxtoOuXQHJY+HDrjGKuWnOX3O+mjHbY8He9rw3z/FYDLXrerXDl+3PdcUwm9dCzPmpUG577rrm6dS2zdebgeauc+rapvk58Vp7HR073/ba7jovEeY99W9v28HeY+cp3hNB7K5zaY5bHPtYwPhzet/4bVe4+lSstOmn+vfMHxPybW/btZ+/tH+Pres6iqL4LPvuinh3rYeu7wLx2VPz/T7R8+bnuqLuj30HNff5MdHbx4K5XYH599m2vY/2c6b9XXkqOtx+prbDwD/Hs+hD1HX9l1/dJiIiIiIiIiIiIiIiIiIiIiIiop+N/nMPgIiIiIiIiIiIiIiIiIiIiIiIiIh+udqB2a5wb1f879j7XbHa5s927LX5uogHnoqvdkUGHwvQtsOJZVmejOp2aYd722PoCiF3ne+pAKWI3Lbf6xpX8zURXzw2Z815PTam9mea2gHIrijxY9FTEattj/+xdSE+0/y7OffN3zVNOzru9vm2990VAW5+pjmu9lx0xaHb1+RYOLQdHH5sHX5qHxK/fd+o6bEw6qcaxynvs33X/P9c8eGfy+cK+QLHn1ddz4yueHrXvXLsdXH/N4PYp4Lgzet9au12rYtT3zEf4lhc99j7Xa+1n0ni3NvPs65z6zqXrmcbERERERERERERERERERERERER0ZdM4T+mIyIiIiIiIiIiIiIiIiIiIiIios9NURT+Y7U/I82on4jTdoUOm783g6/HAoNdrzVfF7Ha9utdMdp2OLj5+WaMt+3YmI5FO49t0xxDcxsRjG3OxfvES4/9e852lFH8V1WVPEYzhCvmr7lNcw6FdhT4VIi1fZz2ubXjxe3Q7odEXtvzeixk2V6D7fe6oryapj2IE7e1z+NYmLp5nPb5tufnWAC4GclsH/tz+aljwF3HB06v91/iv20+FosFGDb9VB57Vp56ZgCH90/X/X/qu0e8Jn4eC/+2g7Zdz6f3CU8/tn4+5D7o+o445Vgo/UOC1X8Oa76u6w+vhhMRERERERERERERERERERERERG9J/3nHgARERERERERERERERERERERERER/bK0g4TtkGk7NNgVi+2K3bb32X69HTg8tt/m/quqkuHFUwHEx87h2DjfZz/t7duB12OxxmMRWhGl1DRNxmHbwWFVVQ8ivs2wbnP+uojgrzhOV9xWVdUHr52a22OByFO61tGxeesKcVZV9SD6DOAg2lvX9YPAcnNNtaOd7TG156AdLhZjaP7XPK4YV9fa/jl86uN+yPUWnzsVGxXx5LIsf/zgPqH3CbP+OQROf8lOBdXF+8eeyR8z98dC4c19Nu/d94myd30fHDvW+3xffYiu74D2vtpR+GPPWCIiIiIiIiIiIiIiIiIiIiIiIiI6jTFfIiIiIiIiIiIiIiIiIiIiIiIiIjrwWISwHVE99rnm7814ajt629xHO1jbtb92iFCMpxn/7YrcNiOsXftrnvv7Bjrb8cSyLDvjj+L3rjE056GqKvmZruCi0PxMc3/Nz4qxiHlsz3XzOjbDwMfOuyt42xXgPBXoPbX/9nk057G9j7qu5XmJOev6XNd+25/tOl57LrvWw2Nr41S49lRU+pfm2LX92Ijtsc9/riju59hv1zzQj9cVBhev/9j9nnq9fayyLB9c4674+y/h3m1+tzb/PubUc7L9vXnsfSIiIiIiIiIiIiIiIiIiIiIiIqIvHWO+RERERERERERERERERERERERERNSpHSNtB1y7QojN17ve6/pMU/PzIszbPmZXJPexyGAzontqLO0w4jHtEOyx/Yl4cVf4sfl3c3xdceCuuPGxeT8VkW2+3w7adkWPxXja8dnm/psRyfcND7fXVVfsuD1n7WvdXB/t+OSxUO+xa3BqHZ0KW7ZfL4riwf5OBZfb16JrvJ87Gnpq/6qqyjVcVRXKsjwY+6fWvKafyucOkDJw+nn8mHl9n22PhblPPTvF8+mx74f38anXTTPODhyG6dvP71PH7jr/9n6bzwEiIiIiIiIiIiIiIiIiIiIiIiKiLxFjvkRERERERERERERERERERERERET0QDui+yHbPKYZbD0Wnu0KwDb/7gq0Hgvmtl97nwCwCCOeirs29/1Y0LgZCRbbNUOt7Sht87Xmz/b27ahkV7S4HYjtmodj1/tYlLb9fns/Xe8fI2KxIhDZjh+3Q7/NSGVzbO9zrPZ27/P++5y7+Nyx6PFjQd6u+f+Ye/BTEgFpXdehqiqKovgssV3hzymK+74RcfppfOh1eOzzx+Ldx7Y/9j3xMWP7VE6FwI+Nrev7RwS9gdNRfCIiIiIiIiIiIiIiIiIiIiIiIqIvCWO+RERERERERERERERERERERERERHTgfeKD7cDpsW2ORW27PtMOBn5InPVTBROb+2qP/X21o4/HArvi56mxi+OK4G3z9fcd06l4cNd4u+K4p671Y9HZx667CPmeCjSfCmk2x908ZvscTl2H9mvtcRzb9th1aIeAT83RY699yvXd5dR1E+dWluVnDfn+ueE8/HJ8jmtxLOT9PhHfx957n/v5U93zp0K+x45zapvP/SwiIiIiIiIiIiIiIiIiIiIiIiIi+nPBmC8RERERERERERERERERERERERERfTKPBWab0dVjsdjmZ5sRwa7jtPf1PrHBdqCx/V6biJoeC8yeOuf2/h4bX1VVnefSFYNtR2ebY6qqSsaRu0K3XXHcrtceCzafClYe+7v93mOR2Ob4m+Nqr4Fj430smNw+1mO/H9vmfYLWH+PnimeKELX4jxFP+kvV9azter58Cu8by/9Q7e+pH7PvY+fOmC8RERERERERERERERERERERERHRO4z5EhEREREREREREREREREREREREdFHORZ3PRbmfZ99dEVyPyQe+DHR1vfVDud+qFNB2vb77YjxsWMfCwmfev19Q8TtMT12zu19ve81b55j13x0rYP2HIo5U1X1IGh8bL91XR/Ejo9Fo8Vrp87lLzVwKWK+f4nnRn85PsX99yHbt8Pkv7T7o/28PBaoF06NXzw7m8/SX9r5EhEREREREREREREREREREREREf0cGPMlIiIiIiIiIiIiIiIiIiIiIiIiog/yWGT3Y4K6XRHgD40Gfs7I4KkwYjs6+9h5NOOz7blshyLbx24Hatth3/b27RDjsbE3x9V+/3MGkpvbiTkR4+06fnOM7fM9Nt4PuXZdx3osdnnqtZ8rfCnCxj82xstwJ/2S/RTP/K6/3+fZ8HM4Fts99p39PuN/LHZORERERERERERERERERERERERE9CVizJeIiIiIiIiIiIiIiIiIiIiIiIiIPtjniBj+UqOjPyYq/D7h42aAsRn5FVHbuq6hqqrc37EQcPt47bCvrusPQrldnz12Lj+Fx8K5XYFj8V57+2Nhy2PHasel38djx/g56LoOx3Fgmia22y2KovjFjfGn8Eu8Np/CY8+U9mf/Eufgp/YxEdyP9bEx8Gbs/dh+T8Xjm5/7kOcmERERERERERERERERERERERER0ZeIMV8iIiIiIiIiIiIiIiIiIiIiIiIi+iCfK+b3S4oEfkgEU3y26/PHIpBiG8uyEAQByrJEGIbIskx+risu2/557Fhi+7quoWmaPE6apoiiSAZe33d/vwTtsVZV1Rm+PBYmPrY/sa9jTq2FruveHtNPGVRVVRWe52E8HkNVVYRh+JMc95dEURQYhgFd11EUBfI8/8Wt5Y9hmiaCIICu64jjGPv9HmVZ/tzD+uI0nzMfElb+2P1/aNBX6NpejLf5nXXsedl+va5rlGX5F3EvEREREREREREREREREREREREREf1YjPkSEREREREREREREREREREREREREf3ERHD0+fPnuLi4wGKxwP39Pebz+YPgb5eu0G/78+J327ZxcXGBwWCA9XoNRVGw2+1OBmx/Lo+FIo/FkR/77McGKEXQUlVVqKoK4F3491QEWWzXjg3/FAzDQL/fx2AwwHa7/UVe48/NsiyMRiOYponNZoPtdvtnH701DAOTyQSXl5fI8xz7/R55niOO45PbfY7wquM4UFUVaZqiKIpPvv8/J58z1P1j99sVHRbEa+/zfPgcwWIiIiIiIiIiIiIiIiIiIiIiIiKiP3eM+RIRERERERERERERERERERERERHRz+aXGgo8FWn9sWNVFAWapsH3fZydncFxHGiaBtd15f67jvFYyLb5t5hXXdfR7/dxfn4O4F3o1PM8RFH0iwpx/th5bcYqP3Q/j61BTdNgWRYcx0Fd14iiCGmangwtH7uGn5OiKLBtG77vQ1EU7Pf7g/Dwl0BRFPR6PVxeXkJRFMRx/JMHlZtj+RRzr2kagiDAYDBAWZZIkgS+70PTtE8wyg+jqipGoxEMw8BqtcJ2u/2i1lfTL+G8j62x5jOt/Wxs3w9dMXgR+xX//RLOlYiIiIiIiIiIiIiIiIiIiIiIiOiXhDFfIiIiIiIiIiIiIiIiIiIiIiIiIvpZ/TmFAt8nsvsYRVFgGAZs28Z8PscPP/wgA4vHwqMfM0eKosBxHARBgP1+j9VqBV3XoWnazxY4PeXHRCO7IpUfun3XOBRFgWVZmEwmCIIARVHg7u4OeZ6jLMvPMpaucbwPEYR2XRdJkiCKIjmWP6d77MdQVRVBEKDX62G73SJJElRV9ZOOQayZsixRFMWPjlS7rouLiwvs93v88MMP0HUdX3311U9+XsC7+bUsC67rIoqiL2ptdekK4X6s99lHV3i8/ZoI8IrXxOuqqh6Mu/lf17Gb+3nf8RERERERERERERERERERERERERF9KdTHP0JERERERERERERERERERERERERE9OmIIKRlWb/IqOznJEK+nudBURTc3d1htVoBALIs+6TH0nUdw+EQmqbh+voai8UCRVEgz3M5ll+SHxuKbIYrf8xxm1FMTdPgOA4GgwEsy4Jt2+81b58ievkh+xDh5tFoBMuyEIYh4jgG8G4d6Lr+i7ven1ozkl1VFXa7HeI4/smjt7qu4+zsDEEQPAiifggRBR6PxwiCAMvlUl5TVVU/yXmpqgpd12Ga5nutkbqusVqtsFgskKbpjzq2uL+akdkP2fYvyYeEfMXv7YivqqpQVbUz7tue46qqUFXVgyhwMyovxtQM/hIRERERERERERERERERERERERHRO/rPPQAiIiIiIiIiIiIiIiIiIiIiIiIi+nIoioJ+v4/xeIw8z3F/fy8DlV8CVVXhOA76/T72+z2SJIGu69A0TUZ2PxXTNDGZTBCGIaIogqZpMAwDaZrKEGg7/Phz+bmP30XEYR3HQVmW2O/38H0feZ53hjBF8LLr9c9N0zQMh0NMJhPkeY7dboeiKKCqKvr9vnztlzjPn4IIwwZBAM/zUBQF9vs9iqL4ycdimiY8z4NpmgjD8KPva8MwMBgMcH5+jvV6LZ8VnuchyzKUZfnRY2yubdd1oaoqsizDfr9HFEUPPivWcVVV2G63CMNQ/v0YTdMA4OCeUVUVnufBtm2kaYr9fv/o+YgxG4YBVVWRpuknD6B/rE8RIhc+5pnRju02Q77Au7lvx3rbx+gK+bbH9pf6/CAiIiIiIiIiIiIiIiIiIiIiIiL6UIz5EhEREREREREREREREREREREREdFPxjRN/NVf/RUuLy9xe3uL/X5/EJf9S2cYBsbjMWzbxmKxQFEUMqaZpuknjSX6vo/BYIDb21uUZYnBYAAAiOP4IJz5KY+pKAp0XUdVVT8qNvopnQoWn3pPVVW4rgvP85AkCeI4Rq/XQ5ZlH7ReP3fQV1EUOI6Ds7MzuK6L169fyyCrZVk4Pz9HkiQIw/CzjeHnpCiKjGSPx2Pouo71ei2Ds6eoqgpN06AoCvI8/yTXSYSfLcuCZVkfFfNVVRW2bWM8HkPTNCwWC9R1DdM04TgOoij66PtLrOvBYIDJZALP85CmKaIogmEYyLIMRVHI+Leu61AUBUVRIM9zlGX5XscWzwLXdWEYBqIoQhzHqOsavu/j2bNnCIIA8/kcNzc3iKLo6PyL/YgQvK7ruL29xe3tLYqikAFaVVWhqirqukZZlifv048NmX/u+/nYc0qcVzPc2/y9GRQX77Ujv8d+7zp+c7/N/RMRERERERERERERERERERERERERY75ERERERERERERERERERERERERE9BMSUcjb21sZRRUhxr/0UKCiKDBNE5PJRMZGy7KE4zgoigJxHKOqqk8yD5qm4ezsDKqqYrfbwbIs9Pt9zOdzGQL9HPPteR7Oz8+x2WywXq9RFMUnP8bHOHauzQBm8zOapsFxHPT7fZimie12C03TsN/vH4RM29u2g5mfe10bhoHRaIThcIiyLLHdbmXgtK5rxHEM0zRhGMbB2H+ue645P5/i+HVdQ9M0DAYDjEYjVFWF3W53Mo6taRo0TZNR26qqcH9/f7CNeC4B+KB4c57n2Gw2CIIAhmF81Dzruo4gCBAEAbbbLcIwhKIo8H0fuq5jPp9/cABdxHU9z8NkMsF4PJbx3vV6DdM0Yds2LMuCYRgIggC+70NVVRRFgSzLEEUR0jRFURQnY7m6rkPXdfi+j8lkAl3XZby8LEsEQQDHcQBAPiM0TUNd1wfPQEVRYBgGBoMBnjx5guFwCNu25bpeLpcoy/LgWpqmiTAMsdlskCRJ59zrug7TNFFV1UdF1H+Oe6d5vMeeZ+K/5mvN79n29s33mt/Jqqo+ekwiIiIiIiIiIiIiIiIiIiIiIiKiLxFjvkRERERERERERERERERERERERET0k9rtdsiyDIPBAEVRfLKA7S+dqqqwbRtBEODt27eo6xqWZcH3faxWKxm1/DGhSBFh9DwPFxcXWC6XUFUVo9EIcRxjs9kgz/NPeVqSqqrwPA/Pnz+XQdnNZvOzX9tTx2+GfEW4VVEUWJaF0WiEfr+PKIqQZRl830cYhg/2145sNmO1n0sztimCqZqmYbFYyDHWdS3Dsp7nwTRNJEkitz+2z2bA9lNfO1VVZehVjO9TBJ9t28ZoNILjOFitVoiiqHOdi6C27/twXRfT6RTPnz/HcrlEHMfyeSTCtiJ2G8cx8jzvjKCqqiqjpyJwG0URqqpCURQfFYo1TRNBEKCqKiyXSzmm0WiENE0RhuEHz5FhGBgOh7i4uIBhGFgsFlgsFojjGHVd4+LiAnVdw3Vd9Pt99Pt9KIqC7XYrXzcMQwZwwzCU2zbHblkWgiCQ90IYhvB9H5ZlQdM0GfmN4xiKoiCOYwDAeDwG8O77Qbxm2zam0ymePn2Koijwxz/+EWEYwjAM6LouPzMcDnF5eYmzszNUVYXNZoNXr16hKIoH60DTNPi+j9FohP1+LyPDH+Jj7osf81w/FuDt2nfXMRRFkfezWLMi3Ns8xrH9/9zPcCIiIiIiIiIiIiIiIiIiIiIiIqJfGsZ8iYiIiIiIiIiIiIiIiIiIiIiIiOgnoSgKdF1Hr9fD3d0d6rpGlmV/0aHAZiDWtm1MJhMAQBiGMsQKAOv1Wm7zY+dD0zT0+32Ypok8z2Xg8u7uDlmW/ah9n9KMx1ZV9ZNEbT81sUaDIMBwOISqqgjDEKZpQlVVRFH06PVpx30/Zgx1XR+EbxVFQVEUMg6rqioMw4Bt27i8vMTFxQU0TUNVVdB1XUZoy7LEfr9/EO4UxxE/xb5M04SiKMjzHEmSIE3TB9t9DHGMyWSC6XQK27ax3W4xm82wXq9/VGBaPFP6/T6qqpKR2a7ztW0b5+fnmEwmqOsahmFgvV5jv9/LObUsC0+ePMFwOISiKFiv17i7u5MhbBFENQwDruvC8zxomoaiKLDdbpEkiQz6NiOszflu/hSfEZ9TVRWu62I8HmO322G32wEAgiCAbdu4v7//4PtY13WMRiM8f/4cqqrihx9+wHK5lM9fVVWRJAlM08SzZ8/gOA7m8zlWqxV2u52M/TqOA9d1sdlsEEXRwTHE82wwGEBRFCyXS+R5jiiKMBgM5OeGwyEMw0BRFDKU7Hkenj17ht1uJ6+FuA4vXrzAarXCq1ev5FyoqirD07/61a8QBAF2ux3+03/6TyiKAk+ePIGu6/JzzWvQvH7innlfP+aZ9jm/507tu/0s7jqH5vyIMLV4XczZX/L3NBEREREREREREREREREREREREdGHYsyXiIiIiIiIiIiIiIiIiIiIiIiIiD47ESbt9XrQNA2WZX2yUOipY6qqiqqqPutxuogoooixmqYJz/Pgui7W6zXSNIWmafA8D1EUoSzLT3Lcuq6h6zqGwyE0TYNt20iSBPf394ii6GAemmHLZvDyY6ONdV1jt9vh22+/RVVViOP4FxWAFNdEnGtzLjRNg2EYMuY6mUxg2zbW6zWyLIPrujJwK/bV3O+nWl9ijI7jYDgcYjwe48WLF1BVFd9++y3evHmDNE1lvHY0GuHZs2cIggBhGGI0GkFVVczncyyXS8RxjKIoEIYhyrKU4xZx4rquYds2er0eTNME8Kf7JgxDrNfrB+vmYziOg6urK/z617+GpmmYzWYwDAPD4RBZlmG328lj6LqOuq4fvSc0TUMQBPB9H7/61a8wHo9xd3eH9XqNJEkerD1d13F5eYnf/OY3uL29xdu3b2Xo2PM8VFWF4XCI6XSKy8tLFEUhY9giPFuWJaqqkmHu8Xgs131d19A0DY7joKoqOQZxfxmGgV6vB9u25bNhvV5ju92iKIqDcQ4GAwyHQ7x58wZZlkHXdZyfn2OxWGC73X7QfaUoClzXxWQygWEY+MMf/oDlctkZO37y5Als28a3336Lu7s7GY9WFAVxHMN1XRkqbt7f4tl+cXGBoihwc3OD/X4PwzBweXkJ0zSxXq9hmiYuLy8RhiF0Xcd8PkdVVbi8vEQQBJjP58jzXK6XZ8+eYbVa4dtvv5Xx4Gbs+urqCr/97W/xu9/9Dq9evUJVVfA8T55TO6QsXtvv93jz5s0HP59+iudZ81n8GHFtul4X++oK+XbFebv+/nMMshMRERERERERERERERERERERERH9FBjzJSIiIiIiIiIiIiIiIiIiIiIiIqKfhG3bmE6n6Pf7SNMUy+XyIGL5qaiqCsuy4HkeLMtCkiTYbrfI8/yTH6uLiOgOBgMMBgMYhgFd1+G6LhRFwZs3bwAA/X4fmqZhvV7LcOJjEUcRZxSRxaqqDrYRgdOLiwuoqorNZoPr62uEYdi5bxEZtiwLZVkiDMPOEOr7KopChk1/CSFfEcc1DEOeq2ma2O/32O12KMsSlmWh1+uh1+vJ4OpoNEIYhoiiCKqqAoCcQ8MwUJblg/hx87p0hTE1TZNxZwDIskzGUsX2ruvi8vISV1dXKIoCSZIgDEP0ej0Z9RX72O12yLJMru/Xr1+jKAqMx2M8f/4cmqbh5uYGaZrKtW+aJobDIQaDAYqikPtKkgS73Q5pmkJVVVxcXGA0Gskx/JiYr+u6+Ju/+Ru4rou7uzvM53MURYGvv/4ao9EI+/0e+/0eiqKg3+/j7OwMb9++RRiGR/dp2zaurq4wnU6hqip+/etfI0kSLJfLgzBwcx34vo+vv/4am80G3377rYyJi7U6nU7x29/+FrvdDt999x2WyyVM08SLFy9g2zZ0XYeqqnBdF+fn55hOpwjDEHd3d9jtdjIU/fz5c8RxjFevXsm10+/38eTJE+i6LtfUYDCAoihIkkQ+BxVFgeM46PV6yLIM6/Uauq5jPB7DsqyDwO77EoFhXdexWCywXq8fRL2DIMDz58/hOA7+6Z/+CbPZ7CCmrCgK0jQFACRJchAgNwwDg8EA0+kUAHB/f48kSeA4Ds7OzuB5Hm5vb1GWJZ49e4Z+v488z7HZbJBlGXq9HgaDAXa7Hfb7PUzTlPO73W7x3XffyWNXVSVj04PBAF9//TWur6/x+vVrGaaeTqcoyxJxHKOqqs7nalEUB989HxLQ/ZxOxXObY/yY8bafT+3nl6ZpB68fe5YREREREREREREREREREREREREREWO+RERERERERERERERERERERERERPQjicCoZVkwDAN1XSNNUxnLFBzHwWQyged5KIoCy+XyR0VCu+i6jn6/j8FgAMdxYNs2kiSBoihYr9cP4sGqqkLTNGiaBkVRUFUViqI4CFl+CE3TZJDUcRwZWvU8D4PBAHmeI45jOI4Dz/NkwPLU8RRFgaZpMAwDlmXBcRwZlBXBWRGLdV0XV1dXGI1G+Pbbb/H27Vvs9/sHQUZVVeF5Hp4+fYrJZALDMLDf73F7eytjnF1j0HUddV13zpGmabAsC3VdI89zGbwVUUhFUaDrOhRFQZ7nn/zat4lr2+v1EASBDDy7rov1ei3HLwLTVVUhz3OMx2MMBgP5GdM0kaYpiqLAZDKBruu4vb1FlmXy/FRVxXg8luusGY5WFEXGnc/Pz/H06VNUVYXf/e53uL6+llHdXq+H3/zmN9B1HdfX17i7u0OapqjrGqPRCP/iX/wL+L6P+/t7zOdzpGmKv//7v4eqqpjNZnjz5g2SJMFms8GLFy/w9OlTzOdzORbXdfHs2TNcXl6iqirMZjP86le/wsuXL3F9fX1wzQ3DwLNnz+T1/FiGYeCv/uqv0O/38fLlS8znc5RlifPzc/i+j6IokOc56rqWa9e2bbx8+fLoPj3Pw9dffw3f9/H69WtomoarqyssFgssl0s5Z02maWI8HsNxHPzDP/wD0jSV61NVVfT7ffz2t79FHMf47rvvZKzWsizc3t7Ctm2Ypol+v4/pdIrBYIDNZoM3b95gv9+jqioZ3R0Oh9hut/KeOTs7w5MnTxBFEW5vbxFFETRNk/eliACLcfZ6Pdi2jcVigbIsZVz61atXD+7L9yXmw7ZtBEGA7XYr17/runj+/Dk8z8Mf/vAHeY0Esb5N04RpmtjtdnJNaZoG3/cxHo/lus2yDMPhEMPhEIqi4ObmBvv9HmdnZ3j27BmyLEOapthsNtA0TQaZxf12dnaGi4sLZFmGH374AWEYyuiyiNhaloXLy0t5fcRrFxcXMh4srsux9SvmXFVV1HV9EOg+9nnxHfFTa0d+28/V9ueaY+yK87ZDviI63d7Xj4kHExEREREREREREREREREREREREf0lY8yXiIiIiIiIiIiIiIiIiIiIiIiIiD6aYRgYDocYjUawbVvGXvf7Pe7v77HZbFBVlYzHBkGAMAxxf3+PKIo+6Vg0TcNgMMB0OkWWZZjNZnAcB6PRCIPBAEmSHAQ0bduG53nwPA+2bQMAsizDZrPBdrs9CLK+DxEFvbi4gKqq8vxFBLEsS+x2O2RZBtd1kaYpwjB8EBhuMwwD/X4f4/EYo9EIw+EQg8EARVHgm2++wdu3b1HXNRzHge/7ePr0KZIkwZs3bxDHcWfIt9fr4enTp3j+/DmKokAcx7BtG71eD6vVClmWySCkCBQPBgPouo6qqrDdbmV8U3zu7OwM5+fn2Gw2uL+/RxzHAP4UnrRtG6PRCABwe3v7WaOYqqrCdV2Mx2OMx2NEUYQwDLHZbNDv9zGZTLBerzEej2EYBpbLJTabDQzDwNnZGYqiwP39PcqyhKqqCMMQpmniq6++krHe5rxqmoa/+7u/w93dHVarlYxfirj0kydP0Ov15PyJeK1lWTBNE9PpFC9evMBms8Hvfvc7GZNVVRWGYeDJkycIwxB//OMfsdvtUFUVfN9Hr9dDHMeYz+eIoghVVWG322G9XuPrr7+G53lIkgS+7+PXv/41RqMR7u/vMZvNkCQJJpMJVFVFVVXyeqiqKmPLImD6MUzTxPPnzzEcDvHq1SvMZjOkaQrXdeE4DuI4xnq9RpqmsCwLw+EQQRDgzZs3nfeeuKa/+c1vYNs2Xr16hSiKcHV1hTAM8cMPP8jnTXs7x3HQ6/WQJMlByFdRFHiehxcvXqCqKvzxj388CN2qqoo4jqHrugxBu66LxWKBm5sbeX+J50m/35dBZcMwMJ1OcX5+jvV6Lee8rmv0+3157YA/hWJFbBcAVqsVPM/DaDTCer3GarX6qOtQliXiOEYYhhiNRri4uEBZltjv93JtiViwWPNNzXtX13UkSYI8z+Xr/X4flmXJZ93FxQV830eSJJjP5wjDEK7r4uLiArqu4+3bt5jP58iyDGdnZ+j3+4iiCHEcw/d9nJ2doaoq3N7eygC7uKYikCzmZblcIgxD2LaN8/Nz2LYt78HmPdoM2Qq6rsu1GEURdrtd57k3w7efO2h7bP+njtsM+wJ/ii83x9t8vx3ybc5N+35nxJeIiIiIiIiIiIiIiIiIiIiIiIioG2O+RERERERERERERERERERERERERPRRdF3H5eUlnjx5Il9zHAfAu4CkaZrQNA1VVcnor2maeP36Ndbr9SePBPq+j/Pzc+z3exnO1DQNruvC931YloU4jqEoCgaDAcbjMRzHQVVVMqjrui4AIM9zGIYhw5/vw7ZtPHnyBLqu4/7+HsvlElmWQdd1GUjcbrfQdR22beP+/h5pmp7cp2mauLi4wJMnT6CqKqIogqqq8H0fjuPgyZMnUBQFSZKgKAoZBxbh2nYoWFEUWJaFq6sr+L6Pm5sb3N/fo6oqPH/+HLZtQ1XVg+OLSG+WZdjtdlAUBaPRCKqqYrlcoq5r2LYtw62bzUbGUgXxmV6vhzAMP2vIV9d19Ho9XF1dIQgCvH37FrPZDHmewzRN9Ho9GR3VNA1v3ryRodTLy0skSYLb21usVitomoY4jlFVFYIggOd5+OGHHx6s3eFwiF6vhzdv3sh5FmHgJ0+eyDipOHdN0+ScjEYjXF1dYbFY4Ntvvz2IXBuGgRcvXiAIAtze3mK/36MsS+i6juFwKEPE2+1WRoLLskSe54jjGK7rIkkSPH/+HJPJBH/4wx9wc3ODoihgmiZubm5khFuEO1VVhWVZqKrqIGqr6zpUVT0IPR+jaRouLi7w7Nkz3N/f4+7uDmmawjAMDAYD2LaNzWaD2WyGLMvgOA6CIECappjP5w/2JeK3/X4fvu/j22+/xXq9hud5cBwH8/kcy+WyMwIsArCKoiAMw4P7Wdd1nJ+fw7IsfPfddzLoKkLGmqbJcx8MBlAUBcvlEre3tzLkqyiKjDb7vo/VaoWyLHF+fo6LiwusVivc3NwgyzIAkJFpwzAOorPiWeV5ngzYBkGA/X6Pu7s7+blTcWVxjcS1E0FX8UwEAM/zMBwOZVD67OwM6/Uad3d3D551qqrCNE0ZEI/jWN4Puq7D9334vo+yLKFpGs7Pz6FpGjabDVarFeI4hqqqMqq9WCxwd3eHKIpkxFrXdWw2GwDAdDqF4zjy/ut69qqqiiAIZLjXMAxMJhMoioK3b98exJjFmm6HbcVz8Pz8HJ7n4fXr1wefF459R+n6u3+C34xgf4iuuPCHap+P+F08B45RVfVgXruCv819ijX3vt+DRERERERERERERERERERERERERF8CxnyJiIiIiIiIiIiIiIiIiIiIiIiI6IMpioKzszNcXV0hyzLM53MZng2CAMC7cG0URfB9H5PJBFdXV6jrGsvlEkVRfNKYr2maOD8/R5qmMhwqAo673Q4vXryArusyTHl2dgZFUbBer7HdbmV01/M8eJ4nY5GvX78+iBgqivIghgi8C44+efIEvu/j9vZWhnxFHNU0TaiqiiRJ4Ps+drsd9vs96rqWkdDVanUwJ6ZpygjrdrvF/f094jhGEARIkgTj8RibzQbz+Rzb7RZ5nsO2bWiaht1u1xkKNk0TT58+xWAwwPX1NWazGdI0heM4SNMUmqbJ83RdF9PpFJPJBLvdDrPZDFEUQdM0jEYj9Pt9hGEIABiPx7BtG1mWIUmSB/MjQr6maSJJks8S81UUBaZpYjQa4enTp7AsC2/evMH9/b2M104mE0wmExiGAcMw8P3332O9XiPPc/i+jyAIkOc5oiiSAcskSWDbNvr9PpIkkSHqZsh1MBggTVNst1sA7yK8Iua63+8xn8+x3+9lbFaEW8/OzjCZTHB3d4fXr18jSRJ5LoZhYDwew3VdhGEIXdehaZqMEgdBgCiKcHd3J9cS8C7WqSgKiqKA67qwbRvj8Rjff/89rq+vkee5DH7O53N4nncQgdV1Ha7roq5r5HkOVVXheR76/T7KssR8Pn80Qt3r9fDkyRPEcYybmxv5eTHHVVVhvV7LOHSv14Nt25jP54jj+GBfvu/j2bNnuLy8RFEU+O677+S90u/3oes65vO5jCC3g6xNURTJyK2maRiPxxiNRpjNZjL4LIh5cF0XQRBA0zTM53MZoxVrWFVVeS3zPMd6vUa/38fFxYW8PuL8xTFd18V6vZZxYREOFjFfsR53ux0WiwWSJJHndOzcHMdBv99HURQySixUVYXdboeqqnB+fi6D1kEQQFEU+Rxo7l/TNDiOA8dxMBwOoes69vs9siyDoihwHAeDwQC9Xk+u2zzPMZ/PsdvtkGWZnJurqyuUZYnr62vs93sAQL/fl/Hv3W4H27Zh2zZ2ux2Wy6V8houwbDM66ziODAmLMPT9/T2iKHoQMG8Tz3ARVTcMQz6r3+c7Sdd1jMdjqKqKzWYjo86PHVMEpR8bX9e2ItLbpa7r93qeirlsj1XsvxnvFd9bzUgwEREREREREREREREREREREREREf0JY75ERERERERERERERERERERERERE9MEcx8GLFy8AADc3N1gul9A0Dfv9HpeXl3jy5An+7u/+DvP5HHmeY7/fY7vdwvd9hGH4SWOuIq7oOA5evnx5EL5UVRVZlsl462QywdnZGdI0xf39vYzgis+KsKZhGCjL8iCIqaoqhsOhjHqKc9A0TUZit9stFovFQYjSNE2Ypok0TaHrOtI0xXq9RlVVcF0XT58+xWazkZFYRVFg2zaePn2K6XSKu7s73N3dychpXde4u7tDGIZYr9dYr9coikJGZcuyxHa7fTDHuq5jMpngyZMnWC6XWCwWiKJIzlUz0GoYBobDIabTKbbbLd6+fYs4jlFVFRRFwXa7haZpME0Tuq5jNBpBVVXs93ukaXpwbMMw0O/3MRgMAOBBrPVTEbHeZ8+eAQC+//57rFYrlGUJy7IwnU4xGo2g6zoMw8D19TVWqxXyPIeiKPA8D5ZlIU1T5HmOsizldQyCAL7vy7CqIKLHo9EI2+0WSZJA13VMp1NcXl4iSRLc39/LaKuYaxHYHQwGmM/neP36tZzf5n5935eBYNu24TgOsiyD7/swTRPz+VzGiJuxU03T5Ho1TRPL5RJv375FlmVy7FVVIQxDqKoKTdOg6zqyLINlWXIesiyD4zg4OztDv9/HYrF4NFxqGAYuLy+haRp++OEHeb/bto3JZIJerydDvkVRYDAYYDweoygKeV8I4hy++uoruK6Lf/zHf8T9/T2qqkK/38fZ2RmKokAYhsjzXJ53+94V+xRrXFVVOI4jY+R3d3cHzw1xL4l1a9s2lsslZrMZ9vv9QRDYtm1Mp1O4rov/fzv/EmNJvuD3fb+I8z558lnVVd339p0ZGhxI5MIaEgOCgglBomCBlAVTC0GQYEMEQWA2WkiADYPSxrABLbyxHrBBQKBkjwzLNEGbFmEYgghKgLwRLdKkRFozw5m5j7m3u+uRlc/zfoUXVRFzMquqs7q7uqtu9+cDJCpPnDhx/vH/R0RuCt/f+73fS1VV+fDDD7NarfLo0aMm9loURQ4PD/Pxxx9nPB43wdt6zdrtdhPkHY/HWS6XOTs7u3GPvk59PpvNJtPp9Mb1sDsH0+k05+fnGY1GuXfvXvb395tncx14rscyHA7T7Xaba3E2m2U8HjfPmoODg9y7dy/dbjcXFxdNlHexWGS9XjcR6F/8xV/MRx99lN/4jd9o1vfw8DCj0SidTifz+by5DsqybCLH6/W6ed7cPs/Dw8Pcu3cvP/3pT/P48ePmPq73eVWYdzdMWz/jR6NRVqvVnXHq28f54IMPsre3lx/96EfNdVNfC0VRZDabNX9L6vvx+9//fvM8eNNw8O2Q8evs3vuv27b73u5xb4d66+27Y/yiAWIAAAAAAAAAAAD4thPzBQAAAAAAAAAAAL6Qbrebjz76KN1uN48ePcrFxUUTQ6wjqFVV5eDgINfX1zk7O0tVVbl3716m0+lbDfnWwciHDx8mSRNf3VUURRaLRXq9Xvb397PdbvPs2bNcXl42AcjkeexyvV438dezs7MbQdDRaJSPPvoo4/H4RiDx4OAgDx8+zGw2y9OnT5t4Z5Im8DgcDpvjnJ+fZ7lcptVqZW9vL+12uwnE7oZ8T05O8uTJkzx+/PjGvK3X60yn02w2m0wmkybcWc/FfD7PeDy+MQ91SPT73/9+lstlnj592hyjKIpst9tmDNvtNt1uN4eHhynL8kbwOHkeiVwul5lOp0mSg4ODJpI7mUyyXC5vnP9oNMrx8XF6vV6ePXt2Iyj7trRarezv7+fDDz9MWZb56U9/mvPz82y32wwGg9y/fz/7+/vNvufn503AtV7vg4ODdDqdZk7rqO/+/n7u3buXzWaTZ8+e3YjyFkWR+/fvZzAY5Kc//Wk2m02Oj4/z8ccfZ7Va5fHjx7m6umo+U1VVer1ejo6OcnR0lKurq3z22WdN/LOOgvZ6vXzve99Lkjx9+jRlWabf7zdh6Pv376csy4zH45eu+frzdfz32bNn+fTTT2/EmOt1Xi6X2Ww2abVaTXR0f3+/OXZVVRkOh9nb20uSZr4+z97eXh48eJCnT5/m7Ows6/U6VVVlNBrl4cOH2Ww2mc1mzT15//797O3t5Sc/+UlzTdWGw2F+8Rd/MQ8fPszv/u7v5tNPP81yucxwOGxCxRcXF809NxqNmqh4PZ/1OQ8Gg1xeXma73aYsy+aa+M3f/M0b90s9B/UzZW9vrwnr1set96uvrYcPH+by8jLj8TjHx8c5ODjI7/7u7+bq6qo57mg0yoMHD7K/v5/T09NmLuvxdbvdJhz8e7/3e03A+02ir3WUtY6W18+E25+t5365XObk5KQJ926327RarSbQXcedZ7NZs8/FxUXzzKhDuO12O6enp3n06FETES6KImVZ5vDwMPfv38+9e/eSJI8ePcpyucze3l6Ojo6yWCya868jvcvlsgn51mPfDeXeu3cv9+/fb85xtVrl/Py8uQfqn/oavz0Hdai2vp96vV4TQ9/dp/7s68zn8+zt7TX3TX193b9/P+12u4mf18fb29vLhx9+mPPz8zx58qS5fupn7+vivm8a/b39mV1lWTa/78bC63nYdTsGXM/lFx0DAAAAAAAAAAAAfNuJ+QIAAAAAAAAAAABvbDfSeHV1lWfPnmU+nzeRwOVymaurq/R6vSwWi1xcXOTy8rIJqY7H47ca8+31evnggw/ywQcf5PHjxy9FB+sQ5Hw+T6fTSbfbbaKTdURzN1rY6XSacGkdY0ySTqeTk5OT9Pv9PH36tPme4XCYhw8fpt1u58mTJ7m+vr4RO2232xkOhxkOh1ksFrm+vm4CnWVZZjgcZrlcNt/V6XTy8OHD3L9/PxcXF00c+PacbTabzOfzJnpZh1j7/X5ms9lLEdBut5uHDx+m2+3m008/bQKztbIsU1XVjaBsHXocjUZJnoeSF4tFE76cz+cpyzInJyc5ODjI1dVVE+Ksx3N4eJjj4+Mbx9gNZ74t3W43Jycn2d/fz6NHj3J+ft7EXe/du5der5fpdJoPP/wwy+Uyn332WSaTSTOvw+EwR0dHSdLM32q1SrvdzuHhYfb39/PkyZMmBlsHPLvdbj7++ONUVZXr6+v0er384Ac/SFEUzf51WDV5fj2cnJzkwYMHWS6XefLkScbjcTPvu/N2//79/PCHP8x0Om2u03a7nYODg4xGo1xeXt4ILe9GSu/du9fERh89epTLy8sb670bSq1tt9t0Op0cHR1lvV5nuVym0+mk3W6nqqosFos779+iKJpIbB1ursf14MGDjEajfPLJJ7m+vk6SHB8f5/79+1mv17m4uLgRje50Ovnwww/zve99L1VV5Sc/+UnG43Ha7XYTg53P57m4uGjCuO12O4PBIOv1+qXIcafTSVmW2Ww2KcsyH374YWaz2Y1odx1evXfvXhM7brfbN+Lf9f0xGAxydHSUg4ODDAaD/PCHP8xms8lHH32U6XTahIzraPfJyUmWy2XW63XW63UTca2PdXJykuFwmNlsltPT0xtB4rtUVZXZbJa9vb0cHx+n1Wrl+vr6RhS3vv76/X46nU7G43GS5OjoKCcnJ831XhRFJpNJxuNxVqtVRqNRttttEwGuw79lWWYymeTp06fNWOtn6O79lDx/7i+Xy+zv7+fw8LBZr8lk0oR8y7Js5qVeizpMvb+/n6Ojo3S73VxdXeXi4iKdTif379/PBx98kGfPnjXXTj2Geny710F9LXa73YxGoxRF0VzTr7onbl/bdRz67Ows8/k8k8mkuW/rY7ZarXQ6nSaOnjx/vl5fXzfXfavVysHBQfr9fs7Pz2/s+7btBrxrt+dj9/xunzMAAAAAAAAAAADwMjFfAAAAAAAAAAAA4I3VodA61lmHSGt17HC1WmU+nzfxx26320Qi31bMt45k3r9/P2VZZjqdvnTssizT6/WyXC6bn3a7nb29vSyXyyaIm6TZ3u/3b4RFi6LI/v5+7t+/n9Vqlel02oRVP/jggxweHub8/DyXl5dNvLT+XK/Xa6KRFxcXTfyx1WplNBrl5OQkl5eXWSwWKYoi9+7dayKjp6enmU6nN+a3PqfdsGT9fa1WK91ut4lu7qpjouPx+EZktSiKJgDb6/UymUyyXq+z2WxyeXnZxDXb7XYzt3t7e008tCzLjEajdLvdrFarJjxcB2kPDw/Tbj//76p1gPhtxpxrdQx1u93m4uIiVVU1a1aWZU5PTzMYDNLv9/PDH/6wWd86Znl0dJTBYJBnz57l6uqqCYDW11dVVTk7O2vmtZ63/f39PHz4ML/1W7+V2WyWk5OTHB8f58c//nFOT0+zWCyamGYdwv7www/Tbrfz6aefNnO8e8zDw8P80i/9UlarVRPhraoqk8kkw+GwiVLXYejd4GsdRt0Nn9ah01fFSutrvr536lDw06dPm2Bxfdz1ep35fP65cdmiKPLgwYPMZrMmGF0URfr9fr73ve9lsVjk2bNnWSwW2dvby9HRUXq9XsbjcXNfFUWRdrud+/fv5w/+wT+Ybreb8Xic8/PztFqt5j5cLBaZTqdNeLkODu/t7eXg4KCJZNfnsVgs0ul0UlVVOp1Oer1enj171oyxDgEfHh5mvV7n+vo6o9Eoy+WyidEmz6PA/X4/g8Ggid9Op9NMJpP0er0MBoM8ffq0CfmORqMMBoNmPJeXl+n3+819U8d1y7JswrK7YdV6De6K+i4Wi1xdXTUB7TpcXt/ru1HhoihuRMkPDg6yWCyyXC6bkG99XZZl2Ty7duPh2+22uU/Ksky73U63202v10tZlpnNZmm1WtlsNlmtVnnw4EFarVYzB/V9Wcd1N5tN8/el0+kkeR7prp8fk8kkz549a5473W43v/ALv5Bf/uVfzkcffdSsUR1Mnk6nr5yn+h4ZDAaZz+fNfnfN7+55X19fNwHj7Xbb3B91rHc3WF6HkD/99NPm/qnjv61W68a98ybh5vp6riP0X2T/V53T66633ZDvm3wPAAAAAAAAAAAAfJeI+QIAAAAAAAAAAABv7PDwMEdHR5lOp7m6umqCt7uqqspms2nCiknS6/WSJNfX128l5lrHEA8PD5sIbR3C3FVHM+fzea6urppg6cHBQcqybKKM6/U6/X4/h4eHTTi1Dhj2er08ePAg+/v7+elPf9qc097eXh4+fNiEb3dDtXXI9/DwMKPRqAmz7oaDP/jggwwGg3z66aepqir9fj/f//73s91u8/jx41xdXd2IQiZpIpb9fr8JxdbfV8d8Ly4ubnyuLMucnJyk1Wo1geDdz/X7/ezv76fT6TQhzO12m9PT01xfX78UDu71ejk4OGiiuXt7e02IdD6fpyiKjEajHBwcZLPZZLPZZDQavVEM9svqdrvNOifJyclJhsNhlstlLi4uMpvN8vHHH2c+n+f09PRGzLjX6+Xhw4dNCLgOVPf7/dy/fz9HR0d59OhRc/3Uccxut5sHDx408dblcnkjdF2HjWvtdjv37t3L3t5eHj9+nLOzsxvB6FarlYODg3z/+9/PgwcP8pu/+ZtN4LaOJ3/88cdJktVqlclkciPSuxuP7na7TVy0Dv6+as4ODg7SarUyn8+bqHF9rDpQ2uv10m63c35+/srj7Or1eun3+3n06FEWi0XzPR988EGGw2F+9rOfZTqdNtdcr9e7EXmuo7rHx8f5wQ9+kOPj41xdXaUsy+zt7eXevXvZbrc5OztLp9PJfD7ParVq5nk+nzdB3zrKu/teq9Vq5nqxWKTb7WY4HKaqqgyHw/T7/azX6yaivL+/36xRfW51mLyOlldVlaurqyyXywwGg6zX63Q6nRwcHDTB1tls1lybp6enOTg4yL1795rx70ZlT05Osr+/34y/Dr7uHuNVNptNxuNxqqpqosx7e3vNmrXb7bTb7VRV1Vx/RVE00d31et1EkncD0/P5PI8ePboRFt9ut00YvT7Pel7r4PlqtcpgMMh4PE673c5oNMrFxUVzPxZFkSdPnqTb7TbnMBqNMhqN0m63m/HM5/PMZrMsFotmvouiyGeffZbFYpH9/f2UZZn5fN78LBaLZm53FUWRTqeT0WiUXq+Xq6urzGaz5r03fTbdPm59f9bPyDpyXKtDyfW1uN1um/XevYffRH3+9Zh3t7/qOHVsuN6nVn92929B/W9Zls3x3jQaDAAAAAAAAAAAAN8lYr4AAAAAAAAAAADAG2m1Wjk5OUm3221ina8L81ZVlfV6nfV63URGk7y1mGtRFBkMBk2scjweN+HT3X1Go1H6/X7Oz88zmUyaUGUdm+33+00MuN727NmzZpytVitHR0f56KOP0mq1Mh6Pm3O6d+9eer1ePvnkk5fCu3VEuN/vN+e9G3ftdrv56KOPMh6PmwDn4eFh9vf386Mf/agJ8t6eq3a7nb29vXS73UwmkyauWUdQ68jvbnS1LMvcv38/i8WiGWcdhKzPbzgcZjweN8es45SLxaKJ19brWseB79+/n/Pz8wwGgyTJeDxOkiYkul6vm3ktyzLL5fJzY6RfVlmWKcsyrVariU3X47+6umriscfHx/nkk08ym81uhHxPTk5ycHCQs7OzZn46nU4ODw9zcnKSsizz7NmzGwHgTqeTk5OTfPTRR7m8vMx4PG4CrpPJ5KXrvCzLDAaDJjj89OnTG8HPel2Pj49TlmU2m00Tvq6/r9/vN3P9+PHjl4KhZVmm3++n3+/fCJpuNpsb92l9rezv7zex1cVikeFwmHv37jX7rVarlGWZ0WiUJDcC168zGAxuhErr4PbR0VEWi0Vms1n6/X46nU7a7XYT8q3jrnVotQ7EzmazXFxc5IMPPsgv/MIv5OLiIo8ePWrCvNPp9MZ9V0eOe71e9vb2mtdFUdyYh+12m+vr6+zv7+fBgwfNPNfR7+l02oRpl8tlhsNhTk5Omjj1bDbLer1Ot9vNZrNprpvNZpPpdJq9vb2cnJxktVo19/hqtUpRFE1MOkkzJ/Va1uPo9/s5OTlp4tq31/p11ut184waDAZNfLiOudZjqUPQSZqxJC8HYauqaq6lOlRbf+b6+jplWTbPnfV6neVy2Vz/9Vyfnp42seKLi4smnp6kOUY993X4t44F1xHcOthcz08dD18sFnn27Flzz6xWq2Yddsdba7VaGY1G2d/fz2KxaMZ2e7/XhX1ftwb137vd2PGu2/ffZrNp7qcvE7evP1Ov66vGdzv0Wz/HX7XP7rrX87s7LjFfAAAAAAAAAAAAuEnMFwAAAAAAAAAAAHgjnU4nx8fHTXRxN6K5qyzLdDqdJlpYRzqrqnrjKOVd6vhpHR2cTqdZLBYvjePo6ChlWWY+n2e1WjUx38VikYODgwwGg5ycnGQ4HDYh1PF4nOVymaIoMhwOc//+/RweHub8/Dzj8Tjb7TbdbjcffPBBJpNJTk9Pb5xXr9fL4eFhut1uE7CtQ5d1IHhvby9HR0f52c9+1gRz63jnbjh2V6vVynA4TLfbbSKgtyORrVariUrW8cZOp5O9vb1XBmRHo1EGg0Gm02nOz89fClvW8cyiKJrvmk6nGY/H+fjjjzOfz9Pr9ZpA7PHxcYqiuBEVPT4+TpIb8eG3qd1uN8HS4XCYvb29PHnyJBcXF835DIfDlGWZ6+vrZg263W729vayv7/fxF2Xy2UT/Dw6OspgMMhkMmnCukmaaO+DBw8yGAzy5MmTJhDd7Xbz7NmzZt86jFlHausgdB2ert+rw6v19bIbUm632xkOh+l0OlmtVun1es3c7q5VfU+02+1cXV01QdNWq9XM+258td1uZzKZNNf0/v5+RqNRE71eLpdN6HkymWQ2m33uOtQR2vr+ODw8bGKy9XU5Go1SFEWm02mWy2UTXK3XqL4Pk+dh2qdPn+bx48eZzWaZzWZ5/PhxLi4ust1us1gsmnhrrY44T6fTnJyc5OTkJO12u4k918+I5XKZ09PT5lpYr9eZTqdNiLYOWl9fX6fX6zXR1NVqldlslvl83oRRJ5NJLi8vm2M8ffq0CSrX81vHaIuiyPX1dfP5zWZz436tr7M62lwHbev1eBN1UHg+n6fT6aTT6dyI3d5+ttwVk33V+6vVKtfX103QuKqq5vm6XC5v3OfPnj1r4uT1e/Ux66BwURRZrVaZTqfNvNbx3lcFhuv3NptNMy/1ttepY9d1KPr09DTn5+fN9fNV/y696efr/d72s7Be493v2Q2x34777tqN/Naf2Y3/fpngMAAAAAAAAAAAAHybifkCAAAAAAAAAAAAb6SOn56enr42zFoURXq9XgaDQRPF7Ha76ff7qaoqy+WyCWN+lUBgHThdr9fZbreZTCZNGLLW7/dzdHSU5XLZhEO3221ms1lWq1Xm83mOj49zcnKSDz/8MO12O6vVKpPJJMnNKG9VVZnP500MtA6P/uhHP8p0Om1Cnd1uNwcHB+n3+004tx5jHezs9Xq5d+9eqqrK5eXljTm6vLx8KdKbPA/59vv9DAaDLJfLXF9fv3S+dRy0LMsmIlvPQ5Im4lkHZPf3928cbzdYWx+v3W43sdP6vc1mk8VikU6nk+FwmOVymV6vl9FolCS5urrK1dVVlstlBoNBer1elstlM09vUx0rbrVamc1mTcCyXo92u93MQR397Pf76XQ6TWS2jj3P5/Nmbff397O3t9fEWlerVRNLrgPAnU4ny+WyWcNOp5PtdpuyLNPr9Zq5brfbabfb6fV6zfzVY+71eul0OkmSxWKR6XSa/f39Zm3rEHM9h3Vkuo6i7gY4+/1+c0+cn583wd79/f3MZrNmLHUcdTqdZjabZb1ep9VqZTAYZLVaNeu32WzS6/Wy3W6bwPXnqY+7Xq9zcHDQxEXr+O12u21CxOPxOElyeHiYTqeTk5OTJvhdr8V6vc719XWePn2a09PTJnRcX0P1et2+puqo7sHBQe7du5fBYJDFYpFut5uzs7PmXjw/P89sNkur1WqeB7tzWgeet9tt82yoA8L1tnqs9fNwOp3myZMnabfbzXrtHrMe33q9buKq9bzVz8fNZtPMTx2+/aKh2aqqstlsstlsmnl6m+ow+2q1uvE8vz3Wei12P3c7zru7vf6bcjvem7w6Rrt7vM+bo/pZsL+/n+FwmKurqzx79uyVcfnd7/mm7MZ0v4j6efcm79/+jtvfV5bljXn9smMCAAAAAAAAAACA7wIxXwAAAAAAAAAAAOBOdSy0LMtcX183Ydjb6shur9fL1dVVyrLM/v5+Dg4Oslwus7+/n16v10QeXxUEvksdQW23203I83Yott1u5+DgIEmasOztyORkMklZljk6Osr+/n7KsszTp0+TJIPBoAk/TqfTTKfTTCaTJuJ5eHjYhH/rSG8dO+52u81nhsNh85115PX4+DjHx8e5uLjIfD5PVVXpdrtJ8spgan2+/X6/iaPOZrOX9ivLsgnr1gHf9XqdTqeT9XrdjG+73TYx4uVymYuLi0yn06zX6xvH6na7TYB29716HlerVTabTZ48eZLhcJjZbJaLi4tcXl420dl6neoQ65uow5t1dPUudXT1xz/+cfr9frrdbkajURP4Xa/XabfbWSwWTXC4DuLWcebFYtHMSx3BTX4/XFxVVTqdTgaDQUajURNcrQO49RzNZrP0+/08ePCg+e46yFofr35/vV6nLMusVquMx+Pmemy1WtlsNhmNRs1Y5vN58z2z2awZ7+6c1ddQfV3du3cve3t7uXfvXiaTSaqqasLXddB6955YLBb59NNPc3Z2lslk0lw3V1dXubi4uHENvM50Os3FxUW63W42m02urq4ymUyaEPJyuczp6WnG43FarVYuLy8zGAyaa+38/DyXl5dNYHexWDTzejva+7owdB0Evr6+zsHBQUajUQaDQRMl3g3n3hUoruPG9XFf9QypA8b1v3UM/FXPx9246qvGXx/jbYVU60Dr1xFl3Q0Gf94+deD3Vef0unF9kfHevg9e9fn679fe3l5zDdZr91W++21+9qt85+uul92Qb73v6/bb3aeOoX9d1w4AAAAAAAAAAAD8vBPzBQAAAAAAAAAAAO5UlmX6/X7W6/VLEdBaq9XKaDTK/v5+2u12er1ePvzww3zwwQc5ODjI5eVlDg8PM5lMMp/PM5/Pv1TMtz72bvR1N8pZlmUGg0GGw2HG43Gm0+lrA4atVitlWSZ5HoXsdDo5OTlJURTpdDqZzWZptVpZLBZNULTX62U0GmU+n2exWKQsyyb0Wn+mjgsPBoMkSafTyf7+frrdbk5OTrLdbnN1ddWcf6vVavbrdDpNTLGOnLbb7VRVlfF4/FLINfn9sOZ8Ps9wOMzBwUFms1mWy2UTQx0Oh3nw4EE2m03KssxiscjV1dVLId/dMHGr1WqCxbfXYD6fZzKZZDweJ3kekd2NxHY6nSYGPJ1O7wyn1gaDQfr9/mujxbfPez6fZ7lc5vz8PP1+P6PR6Mb1sdlsMp1Oc3193czvbDZrwq7tdru5drvdbhMeruO5STIcDtPpdNLv97PdbnN5eZnNZpPDw8Mkz6+56XSa8/Pz7O/vZzQapSzLXF5e3piT+trs9/vN/F1fXzf3Qj2vR0dHuX//fubzec7OznJ5eXkjrnv73imKIuv1OpeXl8161uOrr/36vG6vd/L82j87O8tms8lyuWyur4uLiyTJ9fX1G0U9J5NJHj161Nwz9Tjra7cO6G6325RlmdPT08zn8ya8PJlMmvjys2fPUhTFjRjym1qv1zk/P09VVTk5OUmv18v19fUbn8ftuXmVOgj8qu2fZzfo+3n7vA11zPfLPGfflvpcvuq8vO4Z/iZzVQfR6+dEff++L77ser/uc7vzWId63yTke3v+xXwBAAAAAAAAAADg1cR8AQAAAAAAAAAAgDvVcdvlcplWq/XK6GK3283h4WGOj4/T6/UyGAyaSOpms8nTp0/z+PHjzGazbDabLxVTrAOtdeC2Dgx3Op10u91mHPV3Xl5epizLDIfDVFWV9XrdRFX7/X729/eTJOfn52m32+l2u3nw4EGm02kuLy9zfX2dBw8eZLlcZjqdJknzXcvlMkVRZG9vL3t7eynLMpPJJJPJJKvVKmVZNt/X7/fz8OHDtFqtbDabnJ2dZb1eN/O4Wq2yWq0yGAxyeHiY+Xx+I1pch053Q6u7ttttZrNZrq6umjjtarXKZrPJZDLJkydPcnR0lE6nk81mk6urqyYKe/t4nU4ne3t76Xa7TXj5dtBxMBhkvV5nOp1mPB5nvV5nvV7fOFa73c5gMGhiurcDsq9b33v37qUsy8zn8ze6JupY6Xa7zWQyaQLMdaS5noPVapVOp5P5fN4EduvrqL5mdoPEe3t7OTw8zN7eXh4+fNhEo8/Pz5uwcx1kruPIn376afb29pKkiRHXa1ZHlVutVnMt1tHdet42m02ePHmS4XCYJHn69GmePXvWBHHPzs6addtdk+1228Rq62Mtl8smxrtarbLdbl8b56yqKpPJ5Ma2Og78RWKw9WeKorhxLdTx7t3vrtelDjbX46t/f/LkSTOvX8Zyuczp6WlzTywWixuB5m/Sm0Zn37aftxjrFx3rF9m/Dll/2Yj816n+O/BV1ur23+TdSO/uvXVbvf12APj2+wAAAAAAAAAAAMDvE/MFAAAAAAAAAAAA3thms0m3201ZljeCiHUAtv7p9/vZbDY5PT3NYrFogrK7sdEvoyzLdDqd9Pv9tNvttNvtzOfzjEajdDqd5v0kubi4SFmW2d/fz2g0SqvVynK5TPI8+FuPczwe5/T0NL1eL/v7+1mtVjk/P8/FxUUT4l2v101ctizL5uf4+Djtdrv5vjq4mzwPKM7n81xeXqbf72c4HObq6iqnp6dpt9vpdDpNNHE+n+fZs2d58OBBEw+uqirz+TxXV1cZj8efGyKtqirj8TifffZZ2u12E62sqiqLxSKfffZZLi4uUhRFFotFFovFK4OW7Xa7WcPJZJLpdPrSfkVRZDgcNuf3qiBwkvT7/fT7/SyXy9fu86rv/+CDD5rY8RdRh0u3220z5t0Q5enpaRNVrf9dr9e5vr5uQszPnj3L9fV1qqrK1dVV9vf302q10u12c3FxkbOzs0wmkyaO+dlnnyVJE+W9vr5uQr/1Tz2u5XJ5Yw3rcdwe5/X1df7hP/yHWa/XzbWUPL9GfvaznzWR5tvnvjtfRVFkvV5/5XjtF12Deiy3A6CfFxJ91XVYz9dXVV+j75tvKpD6vkVr35XVatVcy+9jnPZtjal+ntTPt9fdh7fDv/Vzsw4A7z6r38f5AgAAAAAAAAAAgHdNzBcAAAAAAAAAAAC402azyWKxyHa7zWAwSK/Xy2azSVVV6XQ6OTg4yGg0ynq9zuXlZcbjcR4/fpzJZJLRaJTVatV8/qsoyzLdbjedTiftdjvL5TLj8ThlWWY4HKYoiiyXy1xdXWU6nabVaqUoigwGgwyHwwyHwybOmiSTySSXl5dZLBZptVo5PT1tznW9Xmc4HGY0GmU2mzWB3dVqleVymeFwmKOjo4zH45yfn+fq6uqlWOtiscjp6Wkmk0nKsmyO0+l0MhwOm8DkZrPJ48ePs1gsmu3j8Tiz2Syr1eqNgoqbzSaXl5evfK8O+N6l3+/n8PAwSTKdTl8Zcy3LMv1+P2dnZ5+7pv1+P2VZZj6fZ7FYvNE57O3tZTgc5kc/+tGXDtG+KpBbv6631f+u1+s8evQoz549y2w2a6LDZVnm4uIi2+023W430+k0V1dXWS6Xzfmu1+ucn5+n1WrdCAnXAd/XjeGu19vtNtPp9JXn9qZhWgHO94v1ePe+jWtQlmWSm8+83Vj57WhvrZ6Lsixfiv7Wz6D62AAAAAAAAAAAAMDLxHwBAAAAAAAAAACAO1VVlevr66zX6xwdHWU+n+fi4iJVVeXg4CDD4TDz+Tzn5+dZrVYZDodZLBYpiiLtdjvT6fSNQ6SfpyzLdDqdJjR4dXWVR48eZbPZpCzLbLfbrFarJkK73W5zcXGR6XSa4XCYbrfbxFknk0kmk8mNQOuueuyDwSBPnz5tYriz2Sw/+9nPMhgMslwuM5lMslgsstlsXjrGdrvNYrFowrT1PpvNJqvV6kb4dbVa5enTp02I8ZvWarWaKPPu+d7W7XbT6/VycXGR5XL5yn3Kssze3l6SNEHiuxRFkYODg4zH41xfX3/pOfgin9tsNrm6umrmvA5gbrfbzGazZg622+1L67sb8N1ds6+6dt/G6Cjw9r0uylv/PatfF0VxI+z7tp5VAAAAAAAAAAAA8F0l5gsAAAAAAAAAAAC8kclkkk8//TQff/xxPv744zx48CCbzSbr9Trn5+c5OzvLarVKVVUpyzKtVitFUaTf739u+PWLKMsy7fbz//5YB4Ynk8nnRgk3m01ms1nm83kzpjrCWntdQLfT6aTf7zch4/p45+fnub6+fmXk9baqql4Zgq2P96r934XBYJDhcJjtdpvpdPrKwHGr1crh4WHm8/nnBnf7/X76/X6Wy2Vms9krj3Vbr9fLYDDIp59++tq5+TrsRjDr18nzIObtKObnHQPgq/oyMffdKPztY93+/VXPuvr93f0BAAAAAAAAAACA1xPzBQAAAAAAAAAAAN7Ier3O06dPs1qtcnx8nG63m/F4nMvLy4zH4yyXyxRFkdlsluFwmH6/n8VikU6nk+Vy+VYCra1WK51OJ51Op4n5vmn48FUB3ToOXBRFFovFjffa7XYGg0FWq1Xm8/lLx1qtVl/tZN4jnU4no9EoZVnm8vIys9nslfv1+/0cHBzk2bNnr40zl2WZ0WiUdrudq6urN444d7vdTCaTnJ+ff+nzeJsEeoFv2uc9d3aDvLvh3d0g+e3t9bbb7+++3n2/LMuXAucAAAAAAAAAAADATWK+AAAAAAAAAAAAwBtbLBY5PT3N5eVlyrLMer3Oer3OdrtN8vuR2/l8nsFgkHa7neVymcVi8ZXjgGVZptPppNvtpiiKTKfTlwK8X1S73c5wOGxCv5vNpnmv1WplMBhkOp1+q8K9r9Lr9dLv97PZbDIej2/MQ62O9BZFkWfPnjVrflun08n+/n6SfKG5WywWTSwagJt2I76vivbu/r77/u33bv8tftVndgO/AAAAAAAAAAAAwE1ivgAAAAAAAAAAAMAXstlsXhl7rW232ybye3BwkMVikfl8/pW/tyzLtNvtlGWZ7Xabq6urrNfrL328oijSbrfT7XZTVVXa7XZzXkVRpNvtNjHfzzvfn3dFUaTf76fb7WY2m702vNzr9bK/v5/ZbJbJZPLa0ONgMMje3l7m83lms9lro7+3fdUwM8C33aueu3WE9/Z7RVGkLMsmzFs/i3f3f1UIGAAAAAAAAAAAAPh85bseAAAAAAAAAAAAAPDtUlVVE9ntdruZz+dZrVZf+bhlWabb7abX62W73X7lyG4dN6yqKmVZptVq3fiufr+fTqeT8Xj8rY751vNaz+mr1qrVauXk5CTtdjuXl5efu577+/tpt9uZTqeZz+cCkQBv2avCvXW8tw74tlqtFEWRzWbT/K3b/Vz9mfrn9vE9uwEAAAAAAAAAAODVxHwBAAAAAAAAAACAt6qOBrZarbRarbcW822329nb20u/389qtcp8Pv/Kx9xsNqmqKr1eL71erwn6FkWRwWCQVquV8Xj8rY8almXZzOl2u73xXlEUGQ6HuXfvXq6vrz93PsqyzOHhYbbbbSaTyVtZdwBu2o3R169rdZh3s9k0f+N23999/aoI8O5+AAAAAAAAAAAAwMvEfAEAAAAAAAAAAIC3arvdZrPZNJHA1Wr1UiD2yyiKIp1OpwkEL5fLr3zMzWaT7Xabbrebvb29tNvtJM+jtIPBIOv1OuPx+K2M/3232WyyXq9fCjh2u908ePAgm80mp6en2Ww2rz1Gr9fL0dFR5vN5ptPpd2LeAL5pt6O79U/997cO/VZV1fwt3v3s7msAAAAAAAAAAADgixHzBQAAAAAAAAAAAN66OgjbarUym82yXq+/8jGLokir1UpZlplOp2/lmNvtNovFIklyeHiYbrfbfM9oNGqitN9mRVE00d2yvPlfS3u9Xk5OTjIYDPLo0aNsNpvs7+830eNdZVnm+Pg4VVXl+vo6q9XqGxk/wHdRHeS9HWDffb9+ptf71Nvqn/oYrwr+AgAAAAAAAAAAAK8m5gsAAAAAAAAAAAB8bcqyzGKxyGaz+crHarVaabfbWa/XGY/HTYD2q1qtVpnNZimKInt7e2m1Wun1emm325nP52/te95XZVmmqqq02+0MBoO02+20Wq0Mh8Pcv38/h4eHOTs7y9XVVTqdTvb393NwcPBS+HcwGOTo6CiTySSTyeStrDkAL3tdwDf5/TDvq/apqiqbzSbb7fZG4Pd2yPf28x0AAAAAAAAAAAD4fe13PQAAAAAAAAAAAADg26WOAq7X60yn0yyXy698zLIsU5Zllstl8/O2bDabTKfTDAaDDIfD9Pv9dLvdrNfrzGazt/Y976vtdpv1ep29vb0cHx+n3W6nKIrs7++n3W7n+vo6Z2dnWa1WKcsyRVHk6Ogo2+02l5eXqaoqrVYrx8fHGQ6HOT8/z2Kx+NzYJABf3m5893XqUHu9/+4zuf79drS3qqrXhoABAAAAAAAAAACA58R8AQAAAAAAAAAAgLeqqqqs1+ucnZ3l+vr6rQVxV6tVzs7Osl6vs16v38oxk+cx29VqlfV6nX6/n4ODg/T7/Wy327caDX5fbTabzOfz7O3tZTgcptfrpSiKbDabXF5e5uLiIovFIsnzQPN8Ps9oNMqHH36Ybreby8vL9Hq93L9/P+v1OpeXl9+JeQN4F3ZDvneFd2+/X3/2ddvq7WK+AAAAAAAAAAAA8HpivgAAAAAAAAAAAMBbt1qtcnV1laIo3kp4d7vdZjabZblcpqqqbDabtzDK53aPd3h4mNFolLIss1wuvxNR2qqqMp/Pc3FxkdVqlU6nk+12m8lkkvF43Mx58jz8Ox6Pc3BwkD/0h/5QfvCDH+S3f/u3MxwO0+/389lnn2U8Hr/V9QHgZbtR391tr4rz7u6/u2273Taf2Q35ivkCAAAAAAAAAADA64n5AgAAAAAAAAAAAG9dVVVvJeK7a7PZfG2R2PV6nfV6nV6vl+FwmKqq8pOf/OQ7EfNNnseXr6+vM5/PU5ZlttttVqvVK+d7sVg04d/RaJSPP/44SXJ6epqzs7PvzJwBvAt1aPd2zLeO8tZh3vrf22He3c/V75dl+dI2AAAAAAAAAAAA4NXEfAEAAAAAAAAAAIDvvM1mk/F4nLOzs5RlmaIocn19ndVq9a6H9o2pY8l1APJ1qqrKeDzO7/zO7+Tg4CBlWWY6nebRo0eZTqfZbrff4KgBvjvqEG/9jN59Vu9uq/fb3f/29tuf3Y0Bf13hfAAAAAAAAAAAAPg2KD7vP1sDAAAAAAAAAADA21AUhf+sxnuv0+nk4OAgx8fHKcsyn3zySSaTybse1nur1+ul3++n0+lkPp8L+QJ8zXbjvGVZvhT1rd9PkrIskyTb7falQHsd7d393O5nf95jvlVVFXfvBQAAAAAAAAAAAF+OmC8AAAAAAAAAAABfOzFffl602+10u90URZHpdPpSAJGbyrJMURSvjEUC8HbtRnfrZ2/yPOb7qkBvVVWvfTbvhoFbrVbze5Kf+zC7mC8AAAAAAAAAAABfJzFfAAAAAAAAAAAAvnZivvy82Q0iAsA34a6/Pbsx3+RmdHd3+xf9+7X72S/z+feNmC8AAAAAAAAAAABfJzFfAAAAAAAAAAAAvnZivgAAfBPEfAEAAAAAAAAAAPg6le96AAAAAAAAAAAAAAAAAAAAAAAAAAAAAPC+E/MFAAAAAAAAAAAAAAAAAAAAAAAAAACAO4j5AgAAAAAAAAAAAAAAAAAAAAAAAAAAwB3EfAEAAAAAAAAAAAAAAAAAAAAAAAAAAOAOYr4AAAAAAAAAAAAAAAAAAAAAAAAAAABwBzFfAAAAAAAAAAAAAAAAAAAAAAAAAAAAuIOYLwAAAAAAAAAAAAAAAAAAAAAAAAAAANxBzBcAAAAAAAAAAAAAAAAAAAAAAAAAAADuIOYLAAAAAAAAAAAAAAAAAAAAAAAAAAAAdxDzBQAAAAAAAAAAAAAAAAAAAAAAAAAAgDuI+QIAAAAAAAAAAAAAAAAAAAAAAAAAAMAdxHwBAAAAAAAAAAAAAAAAAAAAAAAAAADgDmK+AAAAAAAAAAAAAAAAAAAAAAAAAAAAcAcxXwAAAAAAAAAAAAAAAAAAAAAAAAAAALiDmC8AAAAAAAAAAAAAAAAAAAAAAAAAAADcQcwXAAAAAAAAAAAAAAAAAAAAAAAAAAAA7iDmCwAAAAAAAAAAAAAAAAAAAAAAAAAAAHcQ8wUAAAAAAAAAAAAAAAAAAAAAAAAAAIA7iPkCAAAAAAAAAAAAAAAAAAAAAAAAAADAHcR8AQAAAAAAAAAAAAAAAAAAAAAAAAAA4A5ivgAAAAAAAAAAAAAAAAAAAAAAAAAAAHAHMV8AAAAAAAAAAAAAAAAAAAAAAAAAAAC4g5gvAAAAAAAAAAAAAAAAAAAAAAAAAAAA3EHMFwAAAAAAAAAAAAAAAAAAAAAAAAAAAO4g5gsAAAAAAAAAAAAAAAAAAAAAAAAAAAB3EPMFAAAAAAAAAAAAAAAAAAAAAAAAAACAO4j5AgAAAAAAAAAAAAAAAAAAAAAAAAAAwB3EfAEAAAAAAAAAAAAAAAAAAAAAAAAAAOAOYr4AAAAAAAAAAAAAAAAAAAAAAAAAAABwBzFfAAAAAAAAAAAAAAAAAAAAAAAAAAAAuIOYLwAAAAAAAAAAAAAAAAAAAAAAAAAAANxBzBcAAAAAAAAAAAAAAAAAAAAAAAAAAADuIOYLAAAAAAAAAAAAAAAAAAAAAAAAAAAAdxDzBQAAAAAAAAAAAAAAAAAAAAAAAAAAgDuI+QIAAAAAAAAAAAAAAAAAAAAAAAAAAMAdxHwBAAAAAAAAAAAAAAAAAAAAAAAAAADgDmK+AAAAAAAAAAAAAAAAAAAAAAAAAAAAcAcxXwAAAAAAAAAAAFIUxY+Lovj7RVH8vaIo/vaLbSdFUfyNoih++8W/xy+2F0VR/HtFUfxOURT/bVEUf/Tdjh4AAAAAAAAAAAAAAODrJ+YLAAAAAAAAAABA7Z+qqupXqqr61Rev/0KSv1lV1S8n+ZsvXifJn07yyy9+fi3JX/zGRwoAAAAAAAAAAAAAAPANE/MFAAAAAAAAAADgdf5Mkl9/8fuvJ/nnd7b/R9Vz/1WSo6IoPnoH4wMAAAAAAAAAAAAAAPjGiPkCAAAAAAAAAACQJFWS/6woir9TFMWvvdj2sKqqz178/ijJwxe/fz/JT3c++7MX2wAAAAAAAAAAAAAAAL612u96AAAAAAAAAAAAALwX/kRVVZ8URfEgyd8oiuI3d9+sqqoqiqL6Igd8EQX+tTt3BAAAAAAAAAAAAAAA+DlQvusBAAAAAAAAAAAA8O5VVfXJi3+fJPlrSf5YksdFUXyUJC/+ffJi90+S/GDn4x+/2Hb7mP9+VVW/WlXVr36dYwcAAAAAAAAAAAAAAPgmiPkCAAAAAAAAAAB8xxVFsVcUxX79e5J/Jsk/SPLXk/zZF7v92ST/yYvf/3qSf6V47o8nuayq6rNveNgAAAAAAAAAAAAAAADfqPa7HgAAAAAAAAAAAADv3MMkf60oiuT5/yv7j6uq+k+Lovivk/yVoij+fJKfJPkXX+z//0ryzyb5nSTTJH/umx8yAAAAAAAAAAAAAADAN6uoqupdjwEAAAAAAAAAAIBvuaIo/Gc1AAC+dlVVFe96DAAAAAAAAAAAAHx7le96AAAAAAAAAAAAAAAAAAAAAAAAAAAAAPC+E/MFAAAAAAAAAAAAAAAAAAAAAAAAAACAO4j5AgAAAAAAAAAAAAAAAAAAAAAAAAAAwB3EfAEAAAAAAAAAAAAAAAAAAAAAAAAAAOAOYr4AAAAAAAAAAAAAAAAAAAAAAAAAAABwBzFfAAAAAAAAAAAAAAAAAAAAAAAAAAAAuIOYLwAAAAAAAAAAAAAAAAAAAAAAAAAAANxBzBcAAAAAAAAAAAAAAAAAAAAAAAAAAADuIOYLAAAAAAAAAAAAAAAAAAAAAAAAAAAAdxDzBQAAAAAAAAAAAAAAAAAAAAAAAAAAgDuI+QIAAAAAAAAAAAAAAAAAAAAAAAAAAMAdxHwBAAAAAAAAAAAAAAAAAAAAAAAAAADgDmK+AAAAAAAAAAAAAAAAAAAAAAAAAAAAcAcxXwAAAAAAAAAAAAAAAAAAAAAAAAAAALiDmC8AAAAAAAAAAAAAAAAAAAAAAAAAAADcQcwXAAAAAAAAAAAAAAAAAAAAAAAAAAAA7iDmCwAAAAAAAAAAAAAAAAAAAAAAAAAAAHcQ8wUAAAAAAAAAAAAAAAAAAAAAAAAAAIA7iPkCAAAAAAAAAAAAAAAAAAAAAAAAAADAHcR8AQAAAAAAAAAAAAAAAAAAAAAAAAAA4A5ivgAAAAAAAAAAAAAAAAAAAAAAAAAAAHAHMV8AAAAAAAAAAAAAAAAAAAAAAAAAAAC4g5gvAAAAAAAAAAAAAAAAAAAAAAAAAAAA3EHMFwAAAAAAAAAAAAAAAAAAAAAAAAAAAO4g5gsAAAAAAAAAAAAAAAAAAAAAAAAAAAB3EPMFAAAAAAAAAAAAAAAAAAAAAAAAAACAO4j5AgAAAAAAAAAAAAAAAAAAAAAAAAAAwB3EfAEAAAAAAAAAAAAAAAAAAAAAAAAAAOAOYr4AAAAAAAAAAAAAAAAAAAAAAAAAAABwBzFfAAAAAAAAAAAAAAAAAAAAAAAAAAAAuIOYLwAAAAAAAAAAAAAAAAAAAAAAAAAAANxBzBcAAAAAAAAAAAAAAAAAAAAAAAAAAADuIOYLAAAAAAAAAAAAAAAAAAAAAAAAAAAAdxDzBQAAAAAAAAAAAAAAAAAAAAAAAAAAgDuI+QIAAAAAAAAAAAAAAAAAAAAAAAAAAMAdxHwBAAAAAAAAAAAAAAAAAAAAAAAAAADgDmK+AAAAAAAAAAAAAAAAAAAAAAAAAAAAcAcxXwAAAAAAAAAAAAAAAAAAAAAAAAAAALiDmC8AAAAAAAAAAAAAAAAAAAAAAAAAAADcQcwXAAAAAAAAAAAAAAAAAAAAAAAAAAAA7iDmCwAAAAAAAAAAAAAAAAAAAAAAAAAAAHcQ8wUAAAAAAAAAAAAAAAAAAAAAAAAAAIA7iPkCAAAAAAAAAAAAAAAAAAAAAAAAAADAHcR8AQAAAAAAAAAAAAAAAAAAAAAAAAAA4A5ivgAAAAAAAAAAAAAAAAAAAAAAAAAAAHAHMV8AAAAAAAAAAAAAAAAAAAAAAAAAAAC4g5gvAAAAAAAAAAAAAAAAAAAAAAAAAAAA3EHMFwAAAAAAAAAAAAAAAAAAAAAAAAAAAO4g5gsAAAAAAAAAAAAAAAAAAAAAAAAAAAB3EPMFAAAAAAAAAAAAAAAAAAAAAAAAAACAO4j5AgAAAAAAAAAAAAAAAAAAAAAAAAAAwB3EfAEAAAAAAAAAAAAAAAAAAAAAAAAAAOAOYr4AAAAAAAAAAAAAAAAAAAAAAAAAAABwBzFfAAAAAAAAAAAAAAAAAAAAAAAAAAAAuIOYLwAAAAAAAAAAAAAAAAAAAAAAAAAAANxBzBcAAAAAAAAAAAAAAAAAAAAAAAAAAADuIOYLAAAAAAAAAAAAAAAAAAAAAAAAAAAAdxDzBQAAAAAAAAAAAAAAAAAAAAAAAAAAgDuI+QIAAAAAAAAAAAAAAAAAAAAAAAAAAMAdxHwBAAAAAAAAAAAAAAAAAAAAAAAAAADgDmK+AAAAAAAAAAAAAAAAAAAAAAAAAAAAcAcxXwAAAAAAAAAAAAAAAAAAAAAAAAAAALiDmC8AAAAAAAAAAAAAAAAAAAAAAAAAAADcQcwXAAAAAAAAAAAAAAAAAAAAAAAAAAAA7iDmCwAAAAAAAAAAAAAAAAAAAAAAAAAAAHcQ8wUAAAAAAAAAAAAAAAAAAAAAAAAAAIA7iPkCAAAAAAAAAAAAAAAAAAAAAAAAAADAHcR8AQAAAAAAAAAAAAAAAAAAAAAAAAAA4A5ivgAAAAAAAAAAAAAAAAAAAAAAAAAAAHAHMV8AAAAAAAAAAAAAAAAAAAAAAAAAAAC4g5gvAAAAAAAAAAAAAAAAAAAAAAAAAAAA3EHMFwAAAAAAAAAAAAAAAAAAAAAAAAAAAO4g5gsAAAAAAAAAAAAAAAAAAAAAAAAAAAB3EPMFAAAAAAAAAAAAAAAAAAAAAAAAAACAO7Tf9QAAAAAAAAAAAAD4Thgn+a13PQg+1/0kp+96EHwua/R+sz7vP2v0frM+77+fhzX6xXc9AAAAAAAAAAAAAL7dxHwBAAAAAAAAAAD4JvxWVVW/+q4HwesVRfG3rdH7zRq936zP+88avd+sz/vPGgEAAAAAAAAAAEBSvusBAAAAAAAAAAAAAAAAAAAAAAAAAAAAwPtOzBcAAAAAAAAAAAAAAAAAAAAAAAAAAADuIOYLAAAAAAAAAADAN+Hff9cD4E7W6P1njd5v1uf9Z43eb9bn/WeNAAAAAAAAAAAA+M4rqqp612MAAAAAAAAAAAAAAAAAAAAAAAAAAACA91r5rgcAAAAAAAAAAAAAAAAAAAAAAAAAAAAA7zsxXwAAAAAAAAAAAL5WRVH8qaIofqsoit8piuIvvOvxfFcVRfEfFkXxpCiKf7Cz7aQoir9RFMVvv/j3+MX2oiiKf+/Fmv23RVH80Xc38u+Goih+UBTFf1EUxX9XFMX/ryiKf+3Fdmv0niiKol8Uxf+nKIr/5sUa/a9ebP8DRVH8rRdr8X8tiqL7YnvvxevfefH+L73TE/iOKIqiVRTF3y2K4v/54rX1eY8URfHjoij+flEUf68oir/9Ypvn3HuiKIqjoij+alEUv1kUxW8URfGPWx8AAAAAAAAAAAC4ScwXAAAAAAAAAACAr01RFK0k//skfzrJH07yLxdF8Yff7ai+s/6PSf7UrW1/IcnfrKrql5P8zRevk+fr9csvfn4tyV/8hsb4XbZO8j+rquoPJ/njSf7VF/eKNXp/LJL8yaqq/rEkv5LkTxVF8ceT/G+S/NtVVf3BJOdJ/vyL/f98kvMX2//tF/vx9fvXkvzGzmvr8/75p6qq+pWqqn71xWvPuffHv5vkP62q6h9N8o/l+b1kfQAAAAAAAAAAAGCHmC8AAAAAAAAAAABfpz+W5HeqqvphVVXLJH85yZ95x2P6Tqqq6r9McnZr859J8usvfv/1JP/8zvb/qHruv0pyVBTFR9/IQL+jqqr6rKqq/++L36/zPKD4/Vij98aLuR6/eNl58VMl+ZNJ/uqL7bfXqF67v5rkny6KovhmRvvdVBTFx0n+R0n+0ovXRazPzwPPufdAURSHSf6JJP9BklRVtayq6iLWBwAAAAAAAAAAAG4Q8wUAAAAAAAAAAODr9P0kP915/bMX23g/PKyq6rMXvz9K8vDF79btHSqK4peS/JEkfyvW6L1SFEWrKIq/l+RJkr+R5HeTXFRVtX6xy+46NGv04v3LJPe+0QF/9/w7Sf4XSbYvXt+L9XnfVEn+s6Io/k5RFL/2Ypvn3PvhDyR5muT/UBTF3y2K4i8VRbEX6wMAAAAAAAAAAAA3iPkCAAAAAAAAAAAAqaqqyvPIIu9QURSjJP+3JP96VVVXu+9Zo3evqqpNVVW/kuTjJH8syT/6bkdErSiKfy7Jk6qq/s67Hguf609UVfVHk/zpJP9qURT/xO6bnnPvVDvJH03yF6uq+iNJJkn+wu4O1gcAAAAAAAAAAADEfAEAAAAAAAAAAPh6fZLkBzuvP36xjffD46IoPkqSF/8+ebHdur0DRVF08jzk+3+uqur//mKzNXoPVVV1keS/SPKPJzkqiqL94q3ddWjW6MX7h0mefbMj/U75HyT5HxdF8eMkfznJn0zy78b6vFeqqvrkxb9Pkvy1PI9ie869H36W5GdVVf2tF6//ap7Hfa0PAAAAAAAAAAAA7BDzBQAAAAAAAAAA4Ov0Xyf55aIo/kBRFN0k/1KSv/6Ox8Tv++tJ/uyL3/9skv9kZ/u/Ujz3x5NcVlX12bsY4HdFURRFkv8gyW9UVfW/3XnLGr0niqL4oCiKoxe/D5L8D5P8Rp5Hff+FF7vdXqN67f6FJP95VVXVNzbg75iqqv6Nqqo+rqrql/L8b81/XlXV/yTW571RFMVeURT79e9J/pkk/yCec++FqqoeJflpURT/yItN/3SS/y7WBwAAAAAAAAAAAG4o/J9TAAAAAAAAAAAAvk5FUfyzSf6dJK0k/2FVVf/Wux3Rd1NRFP+XJP9kkvtJHif5Xyb5fyT5K0l+IclPkvyLVVWdvQjL/u+S/Kkk0yR/rqqqv/0Ohv2dURTFn0jy/07y95NsX2z+N5P8rVij90JRFP/9JL+e58+yMslfqarqf10UxX8vyV9OcpLk7yb5n1ZVtSiKop/k/5TkjyQ5S/IvVVX1w3cz+u+Woij+yST/86qq/jnr8/54sRZ/7cXLdpL/uKqqf6soinvxnHsvFEXxK0n+UpJukh8m+XN58byL9QEAAAAAAAAAAIAkYr4AAAAAAAAAAAAAAAAAAAAAAAAAAABwp/JdDwAAAAAAAAAAAAAAAAAAAAAAAAAAAADed2K+AAAAAAAAAAAAAAAAAAAAAAAAAAAAcAcxXwAAAAAAAAAAAAAAAAAAAAAAAAAAALiDmC8AAAAAAAAAAAAAAAAAAAAAAAAAAADcQcwXAAAAAAAAAAAAAAAAAAAAAAAAAAAA7iDmCwAAAAAAAAAAAAAAAAAAAAAAAAAAAHcQ8wUAAAAAAAAAAAAAAAAAAAAAAAAAAIA7iPkCAAAAAAAAAAAAAAAAAAAAAAAAAADAHf7/rqGhntVOHHEAAAAASUVORK5CYII=\n",
-      "text/plain": [
-       "<Figure size 864x864 with 1 Axes>"
-      ]
-     },
-     "metadata": {
-      "needs_background": "light"
-     },
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
     "_plot(x[0], vmax=1, title=dataset.mapping.get_text(y))"
    ]
diff --git a/notebooks/05c-test-model-end-to-end.ipynb b/notebooks/05c-test-model-end-to-end.ipynb
index b652bdd..e3e92e2 100644
--- a/notebooks/05c-test-model-end-to-end.ipynb
+++ b/notebooks/05c-test-model-end-to-end.ipynb
@@ -2,10 +2,19 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 4,
    "id": "1e40a88b",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "The autoreload extension is already loaded. To reload it, use:\n",
+      "  %reload_ext autoreload\n"
+     ]
+    }
+   ],
    "source": [
     "%load_ext autoreload\n",
     "%autoreload 2\n",
@@ -25,7 +34,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 5,
    "id": "d3a6146b-94b1-4618-a4e4-00f8e23ffdb0",
    "metadata": {},
    "outputs": [],
@@ -45,32 +54,14 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "mapping:\n",
-      "  _target_: text_recognizer.data.mappings.WordPieceMapping\n",
-      "  num_features: 1000\n",
-      "  tokens: iamdb_1kwp_tokens_1000.txt\n",
-      "  lexicon: iamdb_1kwp_lex_1000.txt\n",
-      "  data_dir: null\n",
-      "  use_words: false\n",
-      "  prepend_wordsep: false\n",
-      "  special_tokens:\n",
-      "  - <s>\n",
-      "  - <e>\n",
-      "  - <p>\n",
-      "  extra_symbols:\n",
-      "  - \\n\n",
       "_target_: text_recognizer.models.transformer.TransformerLitModel\n",
       "interval: step\n",
       "monitor: val/loss\n",
-      "ignore_tokens:\n",
-      "- <s>\n",
-      "- <e>\n",
-      "- <p>\n",
       "start_token: <s>\n",
       "end_token: <e>\n",
       "pad_token: <p>\n",
       "\n",
-      "{'mapping': {'_target_': 'text_recognizer.data.mappings.WordPieceMapping', 'num_features': 1000, 'tokens': 'iamdb_1kwp_tokens_1000.txt', 'lexicon': 'iamdb_1kwp_lex_1000.txt', 'data_dir': None, 'use_words': False, 'prepend_wordsep': False, 'special_tokens': ['<s>', '<e>', '<p>'], 'extra_symbols': ['\\\\n']}, '_target_': 'text_recognizer.models.transformer.TransformerLitModel', 'interval': 'step', 'monitor': 'val/loss', 'ignore_tokens': ['<s>', '<e>', '<p>'], 'start_token': '<s>', 'end_token': '<e>', 'pad_token': '<p>'}\n"
+      "{'_target_': 'text_recognizer.models.transformer.TransformerLitModel', 'interval': 'step', 'monitor': 'val/loss', 'start_token': '<s>', 'end_token': '<e>', 'pad_token': '<p>'}\n"
      ]
     }
    ],
@@ -85,6 +76,20 @@
   {
    "cell_type": "code",
    "execution_count": null,
+   "id": "5e6b49ce-7685-4491-bd0a-51487f06a237",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# context initialization\n",
+    "with initialize(config_path=\"../training/conf/mapping/\", job_name=\"test_app\"):\n",
+    "    cfg = compose(config_name=\"word_piece\")\n",
+    "    print(OmegaConf.to_yaml(cfg))\n",
+    "    print(cfg)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
    "id": "9c797159-845e-42c6-bd65-1c976ad627cd",
    "metadata": {},
    "outputs": [],
@@ -98,6 +103,405 @@
   },
   {
    "cell_type": "code",
+   "execution_count": 6,
+   "id": "764c8736-7d68-4261-a57d-face10ebbf42",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "callbacks:\n",
+      "  model_checkpoint:\n",
+      "    _target_: pytorch_lightning.callbacks.ModelCheckpoint\n",
+      "    monitor: val/loss\n",
+      "    save_top_k: 1\n",
+      "    save_last: true\n",
+      "    mode: min\n",
+      "    verbose: false\n",
+      "    dirpath: checkpoints/\n",
+      "    filename:\n",
+      "      epoch:02d: null\n",
+      "  learning_rate_monitor:\n",
+      "    _target_: pytorch_lightning.callbacks.LearningRateMonitor\n",
+      "    logging_interval: step\n",
+      "    log_momentum: false\n",
+      "  watch_model:\n",
+      "    _target_: callbacks.wandb_callbacks.WatchModel\n",
+      "    log: all\n",
+      "    log_freq: 100\n",
+      "  upload_code_as_artifact:\n",
+      "    _target_: callbacks.wandb_callbacks.UploadCodeAsArtifact\n",
+      "    project_dir: ${work_dir}/text_recognizer\n",
+      "  upload_ckpts_as_artifact:\n",
+      "    _target_: callbacks.wandb_callbacks.UploadCheckpointsAsArtifact\n",
+      "    ckpt_dir: checkpoints/\n",
+      "    upload_best_only: true\n",
+      "  log_text_predictions:\n",
+      "    _target_: callbacks.wandb_callbacks.LogTextPredictions\n",
+      "    num_samples: 8\n",
+      "criterion:\n",
+      "  _target_: text_recognizer.criterions.label_smoothing.LabelSmoothingLoss\n",
+      "  smoothing: 0.1\n",
+      "  ignore_index: 1002\n",
+      "datamodule:\n",
+      "  _target_: text_recognizer.data.iam_extended_paragraphs.IAMExtendedParagraphs\n",
+      "  batch_size: 8\n",
+      "  num_workers: 12\n",
+      "  train_fraction: 0.8\n",
+      "  augment: true\n",
+      "  pin_memory: false\n",
+      "logger:\n",
+      "  wandb:\n",
+      "    _target_: pytorch_lightning.loggers.wandb.WandbLogger\n",
+      "    project: text-recognizer\n",
+      "    name: null\n",
+      "    save_dir: .\n",
+      "    offline: false\n",
+      "    id: null\n",
+      "    log_model: false\n",
+      "    prefix: ''\n",
+      "    job_type: train\n",
+      "    group: ''\n",
+      "    tags: []\n",
+      "lr_scheduler:\n",
+      "  _target_: torch.optim.lr_scheduler.OneCycleLR\n",
+      "  max_lr: 0.001\n",
+      "  total_steps: null\n",
+      "  epochs: 512\n",
+      "  steps_per_epoch: 4992\n",
+      "  pct_start: 0.3\n",
+      "  anneal_strategy: cos\n",
+      "  cycle_momentum: true\n",
+      "  base_momentum: 0.85\n",
+      "  max_momentum: 0.95\n",
+      "  div_factor: 25.0\n",
+      "  final_div_factor: 10000.0\n",
+      "  three_phase: true\n",
+      "  last_epoch: -1\n",
+      "  verbose: false\n",
+      "mapping:\n",
+      "  _target_: text_recognizer.data.word_piece_mapping.WordPieceMapping\n",
+      "  num_features: 1000\n",
+      "  tokens: iamdb_1kwp_tokens_1000.txt\n",
+      "  lexicon: iamdb_1kwp_lex_1000.txt\n",
+      "  data_dir: null\n",
+      "  use_words: false\n",
+      "  prepend_wordsep: false\n",
+      "  special_tokens:\n",
+      "  - <s>\n",
+      "  - <e>\n",
+      "  - <p>\n",
+      "  extra_symbols:\n",
+      "  - '\n",
+      "\n",
+      "    '\n",
+      "model:\n",
+      "  _target_: text_recognizer.models.transformer.TransformerLitModel\n",
+      "  interval: step\n",
+      "  monitor: val/loss\n",
+      "  max_output_len: 451\n",
+      "  start_token: <s>\n",
+      "  end_token: <e>\n",
+      "  pad_token: <p>\n",
+      "network:\n",
+      "  encoder:\n",
+      "    _target_: text_recognizer.networks.encoders.efficientnet.EfficientNet\n",
+      "    arch: b0\n",
+      "    out_channels: 1280\n",
+      "    stochastic_dropout_rate: 0.2\n",
+      "    bn_momentum: 0.99\n",
+      "    bn_eps: 0.001\n",
+      "  decoder:\n",
+      "    _target_: text_recognizer.networks.transformer.Decoder\n",
+      "    dim: 96\n",
+      "    depth: 2\n",
+      "    num_heads: 8\n",
+      "    attn_fn: text_recognizer.networks.transformer.attention.Attention\n",
+      "    attn_kwargs:\n",
+      "      dim_head: 16\n",
+      "      dropout_rate: 0.2\n",
+      "    norm_fn: torch.nn.LayerNorm\n",
+      "    ff_fn: text_recognizer.networks.transformer.mlp.FeedForward\n",
+      "    ff_kwargs:\n",
+      "      dim_out: null\n",
+      "      expansion_factor: 4\n",
+      "      glu: true\n",
+      "      dropout_rate: 0.2\n",
+      "    cross_attend: true\n",
+      "    pre_norm: true\n",
+      "    rotary_emb: null\n",
+      "  _target_: text_recognizer.networks.conv_transformer.ConvTransformer\n",
+      "  input_dims:\n",
+      "  - 1\n",
+      "  - 576\n",
+      "  - 640\n",
+      "  hidden_dim: 96\n",
+      "  dropout_rate: 0.2\n",
+      "  num_classes: 1006\n",
+      "  pad_index: 1002\n",
+      "optimizer:\n",
+      "  _target_: madgrad.MADGRAD\n",
+      "  lr: 0.001\n",
+      "  momentum: 0.9\n",
+      "  weight_decay: 0\n",
+      "  eps: 1.0e-06\n",
+      "trainer:\n",
+      "  _target_: pytorch_lightning.Trainer\n",
+      "  stochastic_weight_avg: false\n",
+      "  auto_scale_batch_size: binsearch\n",
+      "  auto_lr_find: false\n",
+      "  gradient_clip_val: 0\n",
+      "  fast_dev_run: false\n",
+      "  gpus: 1\n",
+      "  precision: 16\n",
+      "  max_epochs: 512\n",
+      "  terminate_on_nan: true\n",
+      "  weights_summary: top\n",
+      "  limit_train_batches: 1.0\n",
+      "  limit_val_batches: 1.0\n",
+      "  limit_test_batches: 1.0\n",
+      "  resume_from_checkpoint: null\n",
+      "seed: 4711\n",
+      "tune: false\n",
+      "train: true\n",
+      "test: true\n",
+      "logging: INFO\n",
+      "debug: false\n",
+      "\n",
+      "{'callbacks': {'model_checkpoint': {'_target_': 'pytorch_lightning.callbacks.ModelCheckpoint', 'monitor': 'val/loss', 'save_top_k': 1, 'save_last': True, 'mode': 'min', 'verbose': False, 'dirpath': 'checkpoints/', 'filename': {'epoch:02d': None}}, 'learning_rate_monitor': {'_target_': 'pytorch_lightning.callbacks.LearningRateMonitor', 'logging_interval': 'step', 'log_momentum': False}, 'watch_model': {'_target_': 'callbacks.wandb_callbacks.WatchModel', 'log': 'all', 'log_freq': 100}, 'upload_code_as_artifact': {'_target_': 'callbacks.wandb_callbacks.UploadCodeAsArtifact', 'project_dir': '${work_dir}/text_recognizer'}, 'upload_ckpts_as_artifact': {'_target_': 'callbacks.wandb_callbacks.UploadCheckpointsAsArtifact', 'ckpt_dir': 'checkpoints/', 'upload_best_only': True}, 'log_text_predictions': {'_target_': 'callbacks.wandb_callbacks.LogTextPredictions', 'num_samples': 8}}, 'criterion': {'_target_': 'text_recognizer.criterions.label_smoothing.LabelSmoothingLoss', 'smoothing': 0.1, 'ignore_index': 1002}, 'datamodule': {'_target_': 'text_recognizer.data.iam_extended_paragraphs.IAMExtendedParagraphs', 'batch_size': 8, 'num_workers': 12, 'train_fraction': 0.8, 'augment': True, 'pin_memory': False}, 'logger': {'wandb': {'_target_': 'pytorch_lightning.loggers.wandb.WandbLogger', 'project': 'text-recognizer', 'name': None, 'save_dir': '.', 'offline': False, 'id': None, 'log_model': False, 'prefix': '', 'job_type': 'train', 'group': '', 'tags': []}}, 'lr_scheduler': {'_target_': 'torch.optim.lr_scheduler.OneCycleLR', 'max_lr': 0.001, 'total_steps': None, 'epochs': 512, 'steps_per_epoch': 4992, 'pct_start': 0.3, 'anneal_strategy': 'cos', 'cycle_momentum': True, 'base_momentum': 0.85, 'max_momentum': 0.95, 'div_factor': 25.0, 'final_div_factor': 10000.0, 'three_phase': True, 'last_epoch': -1, 'verbose': False}, 'mapping': {'_target_': 'text_recognizer.data.word_piece_mapping.WordPieceMapping', 'num_features': 1000, 'tokens': 'iamdb_1kwp_tokens_1000.txt', 'lexicon': 'iamdb_1kwp_lex_1000.txt', 'data_dir': None, 'use_words': False, 'prepend_wordsep': False, 'special_tokens': ['<s>', '<e>', '<p>'], 'extra_symbols': ['\\n']}, 'model': {'_target_': 'text_recognizer.models.transformer.TransformerLitModel', 'interval': 'step', 'monitor': 'val/loss', 'max_output_len': 451, 'start_token': '<s>', 'end_token': '<e>', 'pad_token': '<p>'}, 'network': {'encoder': {'_target_': 'text_recognizer.networks.encoders.efficientnet.EfficientNet', 'arch': 'b0', 'out_channels': 1280, 'stochastic_dropout_rate': 0.2, 'bn_momentum': 0.99, 'bn_eps': 0.001}, 'decoder': {'_target_': 'text_recognizer.networks.transformer.Decoder', 'dim': 96, 'depth': 2, 'num_heads': 8, 'attn_fn': 'text_recognizer.networks.transformer.attention.Attention', 'attn_kwargs': {'dim_head': 16, 'dropout_rate': 0.2}, 'norm_fn': 'torch.nn.LayerNorm', 'ff_fn': 'text_recognizer.networks.transformer.mlp.FeedForward', 'ff_kwargs': {'dim_out': None, 'expansion_factor': 4, 'glu': True, 'dropout_rate': 0.2}, 'cross_attend': True, 'pre_norm': True, 'rotary_emb': None}, '_target_': 'text_recognizer.networks.conv_transformer.ConvTransformer', 'input_dims': [1, 576, 640], 'hidden_dim': 96, 'dropout_rate': 0.2, 'num_classes': 1006, 'pad_index': 1002}, 'optimizer': {'_target_': 'madgrad.MADGRAD', 'lr': 0.001, 'momentum': 0.9, 'weight_decay': 0, 'eps': 1e-06}, 'trainer': {'_target_': 'pytorch_lightning.Trainer', 'stochastic_weight_avg': False, 'auto_scale_batch_size': 'binsearch', 'auto_lr_find': False, 'gradient_clip_val': 0, 'fast_dev_run': False, 'gpus': 1, 'precision': 16, 'max_epochs': 512, 'terminate_on_nan': True, 'weights_summary': 'top', 'limit_train_batches': 1.0, 'limit_val_batches': 1.0, 'limit_test_batches': 1.0, 'resume_from_checkpoint': None}, 'seed': 4711, 'tune': False, 'train': True, 'test': True, 'logging': 'INFO', 'debug': False}\n"
+     ]
+    }
+   ],
+   "source": [
+    "# context initialization\n",
+    "with initialize(config_path=\"../training/conf/\", job_name=\"test_app\"):\n",
+    "    cfg = compose(config_name=\"config\")\n",
+    "    print(OmegaConf.to_yaml(cfg))\n",
+    "    print(cfg)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "9382f0ab-8760-4d59-b0b5-b8b65dd1ea31",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'model_checkpoint': {'_target_': 'pytorch_lightning.callbacks.ModelCheckpoint', 'monitor': 'val/loss', 'save_top_k': 1, 'save_last': True, 'mode': 'min', 'verbose': False, 'dirpath': 'checkpoints/', 'filename': {'epoch:02d': None}}, 'learning_rate_monitor': {'_target_': 'pytorch_lightning.callbacks.LearningRateMonitor', 'logging_interval': 'step', 'log_momentum': False}, 'watch_model': {'_target_': 'callbacks.wandb_callbacks.WatchModel', 'log': 'all', 'log_freq': 100}, 'upload_code_as_artifact': {'_target_': 'callbacks.wandb_callbacks.UploadCodeAsArtifact', 'project_dir': '${work_dir}/text_recognizer'}, 'upload_ckpts_as_artifact': {'_target_': 'callbacks.wandb_callbacks.UploadCheckpointsAsArtifact', 'ckpt_dir': 'checkpoints/', 'upload_best_only': True}, 'log_text_predictions': {'_target_': 'callbacks.wandb_callbacks.LogTextPredictions', 'num_samples': 8}}"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "cfg.get(\"callbacks\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "216d5680-66bf-4190-9401-1a59dbbc43af",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "pytorch_lightning.callbacks.ModelCheckpoint\n",
+      "pytorch_lightning.callbacks.LearningRateMonitor\n",
+      "callbacks.wandb_callbacks.WatchModel\n",
+      "callbacks.wandb_callbacks.UploadCodeAsArtifact\n",
+      "callbacks.wandb_callbacks.UploadCheckpointsAsArtifact\n",
+      "callbacks.wandb_callbacks.LogTextPredictions\n"
+     ]
+    }
+   ],
+   "source": [
+    "for l in cfg.callbacks.values():\n",
+    "    print(l.get(\"_target_\"))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "c1a9aa6b-6405-4ffe-b065-02340762476a",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2021-08-03 15:27:02.069 | DEBUG    | text_recognizer.data.word_piece_mapping:__init__:37 - Using data dir: /home/aktersnurra/projects/text-recognizer/data/downloaded/iam/iamdb\n"
+     ]
+    }
+   ],
+   "source": [
+    "mapping = instantiate(cfg.mapping)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "969ba3be-d78f-4b1e-b522-ea8a42669e86",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "network = instantiate(cfg.network)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "a23893a9-a0da-4327-a617-dc0c2011e5e8",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "OmegaConf.set_struct(cfg, False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "a6fae1fa-492d-4648-80fd-1c0dac659b02",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "datamodule = instantiate(cfg.datamodule, mapping=mapping)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "514053ef-fcac-4f3c-a7c8-72c6927d6798",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2021-08-03 15:28:22.541 | INFO     | text_recognizer.data.iam_paragraphs:setup:95 - Loading IAM paragraph regions and lines for None...\n",
+      "2021-08-03 15:28:45.280 | INFO     | text_recognizer.data.iam_synthetic_paragraphs:setup:68 - IAM Synthetic dataset steup for stage None...\n"
+     ]
+    }
+   ],
+   "source": [
+    "datamodule.prepare_data()\n",
+    "datamodule.setup()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "4bad950b-a197-4c60-ad89-903124659a98",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "4992"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "len(datamodule.train_dataloader())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "7db05cbd-48b3-43fa-a99a-353126311879",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "mapping"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "f6e01c15-9a1b-4036-87ae-78716c592264",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "config = cfg"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "4dc475fc-31f4-487e-88c8-b0f445131f5b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "loss_fn = instantiate(cfg.criterion)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "c5c8ed64-d98c-47b5-baf2-1ba57a6c882f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import hydra"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "b5ff5b24-f804-402b-a8ab-f366443025ca",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "    model = hydra.utils.instantiate(\n",
+    "        config.model,\n",
+    "        mapping=mapping,\n",
+    "        network=network,\n",
+    "        loss_fn=loss_fn,\n",
+    "        optimizer_config=config.optimizer,\n",
+    "        lr_scheduler_config=config.lr_scheduler,\n",
+    "        _recursive_=False,\n",
+    "    )\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "99f8a39f-8b10-4f7d-8bff-52794fd48717",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<bound method WordPieceMapping.get_index of <text_recognizer.data.word_piece_mapping.WordPieceMapping object at 0x7fae3b489610>>"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "mapping.get_index"
+   ]
+  },
+  {
+   "cell_type": "code",
    "execution_count": null,
    "id": "af2c8cfa-0b45-4681-b671-0f97ace62516",
    "metadata": {},
diff --git a/poetry.lock b/poetry.lock
index f8a6de3..76ea763 100644
--- a/poetry.lock
+++ b/poetry.lock
@@ -244,7 +244,7 @@ python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*, !=3.4.*"
 
 [[package]]
 name = "charset-normalizer"
-version = "2.0.3"
+version = "2.0.4"
 description = "The Real First Universal Charset Detector. Open, modern and actively maintained alternative to Chardet."
 category = "main"
 optional = false
@@ -658,7 +658,7 @@ test = ["pytest (!=5.3.4)", "pytest-cov", "flaky", "nose", "ipyparallel"]
 
 [[package]]
 name = "ipython"
-version = "7.25.0"
+version = "7.26.0"
 description = "IPython: Productive Interactive Computing"
 category = "dev"
 optional = false
@@ -814,7 +814,7 @@ traitlets = "*"
 
 [[package]]
 name = "jupyter-server"
-version = "1.10.1"
+version = "1.10.2"
 description = "The backend—i.e. core services, APIs, and REST endpoints—to Jupyter web applications."
 category = "dev"
 optional = false
@@ -876,7 +876,7 @@ pygments = ">=2.4.1,<3"
 
 [[package]]
 name = "jupyterlab-server"
-version = "2.6.1"
+version = "2.6.2"
 description = "A set of server components for JupyterLab and JupyterLab like applications ."
 category = "dev"
 optional = false
@@ -1542,7 +1542,7 @@ six = ">=1.5"
 
 [[package]]
 name = "pytorch-lightning"
-version = "1.4.0"
+version = "1.4.1"
 description = "PyTorch Lightning is the lightweight PyTorch wrapper for ML researchers. Scale your models. Write less boilerplate."
 category = "main"
 optional = false
@@ -1936,7 +1936,7 @@ python-versions = ">= 3.5"
 
 [[package]]
 name = "tqdm"
-version = "4.61.2"
+version = "4.62.0"
 description = "Fast, Extensible Progress Meter"
 category = "main"
 optional = false
@@ -1994,7 +1994,7 @@ python-versions = "*"
 
 [[package]]
 name = "urllib3"
-version = "1.25.11"
+version = "1.26.6"
 description = "HTTP library with thread-safe connection pooling, file post, and more."
 category = "main"
 optional = false
@@ -2007,11 +2007,11 @@ socks = ["PySocks (>=1.5.6,!=1.5.7,<2.0)"]
 
 [[package]]
 name = "wandb"
-version = "0.10.33"
+version = "0.11.2"
 description = "A CLI and library for interacting with the Weights and Biases API."
 category = "dev"
 optional = false
-python-versions = ">=2.7, !=3.0.*, !=3.1.*, !=3.2.*, !=3.3.*"
+python-versions = ">=3.5"
 
 [package.dependencies]
 Click = ">=7.0,<8.0.0 || >8.0.0"
@@ -2025,11 +2025,11 @@ psutil = ">=5.0.0"
 python-dateutil = ">=2.6.1"
 PyYAML = "*"
 requests = ">=2.0.0,<3"
-sentry-sdk = ">=0.4.0"
+sentry-sdk = ">=1.0.0"
 shortuuid = ">=0.5.0"
 six = ">=1.13.0"
 subprocess32 = ">=3.5.3"
-urllib3 = {version = "<=1.25.11", markers = "sys_platform == \"win32\" or sys_platform == \"cygwin\""}
+urllib3 = ">=1.26.5"
 
 [package.extras]
 aws = ["boto3"]
@@ -2037,7 +2037,7 @@ gcp = ["google-cloud-storage"]
 grpc = ["grpcio (==1.27.2)"]
 kubeflow = ["kubernetes", "minio", "google-cloud-storage", "sh"]
 media = ["numpy", "moviepy", "pillow", "bokeh", "soundfile", "plotly"]
-sweeps = ["numpy"]
+sweeps = ["numpy (>=1.15,<1.21)", "scipy (>=1.5.4)", "pyyaml", "scikit-learn (==0.24.1)", "jsonschema (>=3.2.0)", "jsonref (>=0.2)", "pydantic (>=1.8.2)"]
 
 [[package]]
 name = "wcwidth"
@@ -2111,7 +2111,7 @@ multidict = ">=4.0"
 [metadata]
 lock-version = "1.1"
 python-versions = "^3.9"
-content-hash = "91db4ec12db098a730fcdd63a7590cab62f0be1072c65229ae52fc35c58875a7"
+content-hash = "78ee5b3911c60380b6d7a487f61df807d2d623cdb8f9848dee260b581ec06460"
 
 [metadata.files]
 absl-py = [
@@ -2300,8 +2300,8 @@ chardet = [
     {file = "chardet-4.0.0.tar.gz", hash = "sha256:0d6f53a15db4120f2b08c94f11e7d93d2c911ee118b6b30a04ec3ee8310179fa"},
 ]
 charset-normalizer = [
-    {file = "charset-normalizer-2.0.3.tar.gz", hash = "sha256:c46c3ace2d744cfbdebceaa3c19ae691f53ae621b39fd7570f59d14fb7f2fd12"},
-    {file = "charset_normalizer-2.0.3-py3-none-any.whl", hash = "sha256:88fce3fa5b1a84fdcb3f603d889f723d1dd89b26059d0123ca435570e848d5e1"},
+    {file = "charset-normalizer-2.0.4.tar.gz", hash = "sha256:f23667ebe1084be45f6ae0538e4a5a865206544097e4e8bbcacf42cd02a348f3"},
+    {file = "charset_normalizer-2.0.4-py3-none-any.whl", hash = "sha256:0c8911edd15d19223366a194a513099a302055a962bca2cec0f54b8b63175d8b"},
 ]
 click = [
     {file = "click-7.1.2-py2.py3-none-any.whl", hash = "sha256:dacca89f4bfadd5de3d7489b7c8a566eee0d3676333fbb50030263894c38c0dc"},
@@ -2626,8 +2626,8 @@ ipykernel = [
     {file = "ipykernel-6.0.3.tar.gz", hash = "sha256:0df34a78c7e1422800d6078cde65ccdcdb859597046c338c759db4dbc535c58f"},
 ]
 ipython = [
-    {file = "ipython-7.25.0-py3-none-any.whl", hash = "sha256:aa21412f2b04ad1a652e30564fff6b4de04726ce875eab222c8430edc6db383a"},
-    {file = "ipython-7.25.0.tar.gz", hash = "sha256:54bbd1fe3882457aaf28ae060a5ccdef97f212a741754e420028d4ec5c2291dc"},
+    {file = "ipython-7.26.0-py3-none-any.whl", hash = "sha256:892743b65c21ed72b806a3a602cca408520b3200b89d1924f4b3d2cdb3692362"},
+    {file = "ipython-7.26.0.tar.gz", hash = "sha256:0cff04bb042800129348701f7bd68a430a844e8fb193979c08f6c99f28bb735e"},
 ]
 ipython-genutils = [
     {file = "ipython_genutils-0.2.0-py2.py3-none-any.whl", hash = "sha256:72dd37233799e619666c9f639a9da83c34013a73e8bbc79a7a6348d93c61fab8"},
@@ -2666,8 +2666,8 @@ jupyter-core = [
     {file = "jupyter_core-4.7.1.tar.gz", hash = "sha256:79025cb3225efcd36847d0840f3fc672c0abd7afd0de83ba8a1d3837619122b4"},
 ]
 jupyter-server = [
-    {file = "jupyter_server-1.10.1-py3-none-any.whl", hash = "sha256:b3eef770ffa34595ed26a6e4460866eaf0f4ff710eccc7648f701bb8c1d0443c"},
-    {file = "jupyter_server-1.10.1.tar.gz", hash = "sha256:fe6b589bd8d8fe08f608e90ce7da1e6bbfd020d99897453b45149a7853e9188f"},
+    {file = "jupyter_server-1.10.2-py3-none-any.whl", hash = "sha256:491c920013144a2d6f5286ab4038df6a081b32352c9c8b928ec8af17eb2a5e10"},
+    {file = "jupyter_server-1.10.2.tar.gz", hash = "sha256:d3a3b68ebc6d7bfee1097f1712cf7709ee39c92379da2cc08724515bb85e72bf"},
 ]
 jupyterlab = [
     {file = "jupyterlab-3.1.1-py3-none-any.whl", hash = "sha256:a181184b1000a550c38da35471dcf91ce11e96750de56430be3fc93ca01dde1e"},
@@ -2678,8 +2678,8 @@ jupyterlab-pygments = [
     {file = "jupyterlab_pygments-0.1.2.tar.gz", hash = "sha256:cfcda0873626150932f438eccf0f8bf22bfa92345b814890ab360d666b254146"},
 ]
 jupyterlab-server = [
-    {file = "jupyterlab_server-2.6.1-py3-none-any.whl", hash = "sha256:58d4b660fce8da4e90f0433ac54f462436fe5fbe731e3a281e15adcdecddb0eb"},
-    {file = "jupyterlab_server-2.6.1.tar.gz", hash = "sha256:73279d1ffdcd3426f716bf5538cf1fdd2eb8a340ac25c5688f3c192c5bd3afc9"},
+    {file = "jupyterlab_server-2.6.2-py3-none-any.whl", hash = "sha256:ab568da1dcef2ffdfc9161128dc00b931aae94d6a94978b16f55330dcd1cb043"},
+    {file = "jupyterlab_server-2.6.2.tar.gz", hash = "sha256:6dc6e7d26600d110b862acbfaa4d1a2c5e86781008d139213896d96178c3accd"},
 ]
 jupyterlab-widgets = [
     {file = "jupyterlab_widgets-1.0.0-py3-none-any.whl", hash = "sha256:caeaf3e6103180e654e7d8d2b81b7d645e59e432487c1d35a41d6d3ee56b3fef"},
@@ -3202,8 +3202,8 @@ python-dateutil = [
     {file = "python_dateutil-2.8.2-py2.py3-none-any.whl", hash = "sha256:961d03dc3453ebbc59dbdea9e4e11c5651520a876d0f4db161e8674aae935da9"},
 ]
 pytorch-lightning = [
-    {file = "pytorch-lightning-1.4.0.tar.gz", hash = "sha256:6529cf064f9dc323c94f3ce84b56ee1a05db1b0ab17db77c4d15aa36e34da81f"},
-    {file = "pytorch_lightning-1.4.0-py3-none-any.whl", hash = "sha256:41fb26e649b830019ecdffb6dc6558266e1317963f7bf2cddb1f1ed862245928"},
+    {file = "pytorch-lightning-1.4.1.tar.gz", hash = "sha256:1d1128aeb5d0e523d2204c4d9399d65c4e5f41ff0370e96d694a823af5e8e6f3"},
+    {file = "pytorch_lightning-1.4.1-py3-none-any.whl", hash = "sha256:4a06723a66296a2ac94cdf353335d64e7ae76c37202b2a4c38a845063e3fe386"},
 ]
 pytz = [
     {file = "pytz-2021.1-py2.py3-none-any.whl", hash = "sha256:eb10ce3e7736052ed3623d49975ce333bcd712c7bb19a58b9e2089d4057d0798"},
@@ -3569,8 +3569,8 @@ tornado = [
     {file = "tornado-6.1.tar.gz", hash = "sha256:33c6e81d7bd55b468d2e793517c909b139960b6c790a60b7991b9b6b76fb9791"},
 ]
 tqdm = [
-    {file = "tqdm-4.61.2-py2.py3-none-any.whl", hash = "sha256:5aa445ea0ad8b16d82b15ab342de6b195a722d75fc1ef9934a46bba6feafbc64"},
-    {file = "tqdm-4.61.2.tar.gz", hash = "sha256:8bb94db0d4468fea27d004a0f1d1c02da3cdedc00fe491c0de986b76a04d6b0a"},
+    {file = "tqdm-4.62.0-py2.py3-none-any.whl", hash = "sha256:706dea48ee05ba16e936ee91cb3791cd2ea6da348a0e50b46863ff4363ff4340"},
+    {file = "tqdm-4.62.0.tar.gz", hash = "sha256:3642d483b558eec80d3c831e23953582c34d7e4540db86d9e5ed9dad238dabc6"},
 ]
 traitlets = [
     {file = "traitlets-5.0.5-py3-none-any.whl", hash = "sha256:69ff3f9d5351f31a7ad80443c2674b7099df13cc41fc5fa6e2f6d3b0330b0426"},
@@ -3618,12 +3618,12 @@ typing-extensions = [
     {file = "typing_extensions-3.10.0.0.tar.gz", hash = "sha256:50b6f157849174217d0656f99dc82fe932884fb250826c18350e159ec6cdf342"},
 ]
 urllib3 = [
-    {file = "urllib3-1.25.11-py2.py3-none-any.whl", hash = "sha256:f5321fbe4bf3fefa0efd0bfe7fb14e90909eb62a48ccda331726b4319897dd5e"},
-    {file = "urllib3-1.25.11.tar.gz", hash = "sha256:8d7eaa5a82a1cac232164990f04874c594c9453ec55eef02eab885aa02fc17a2"},
+    {file = "urllib3-1.26.6-py2.py3-none-any.whl", hash = "sha256:39fb8672126159acb139a7718dd10806104dec1e2f0f6c88aab05d17df10c8d4"},
+    {file = "urllib3-1.26.6.tar.gz", hash = "sha256:f57b4c16c62fa2760b7e3d97c35b255512fb6b59a259730f36ba32ce9f8e342f"},
 ]
 wandb = [
-    {file = "wandb-0.10.33-py2.py3-none-any.whl", hash = "sha256:84f111e31cc4d6e95dcb62028c0c2a9fed7cdf0f8c563d86438aeadcf6d5f495"},
-    {file = "wandb-0.10.33.tar.gz", hash = "sha256:ee69d4e251ae55e73d7d8b1a88b5629a588c820cce8dc8d5f5da15ac298556a7"},
+    {file = "wandb-0.11.2-py2.py3-none-any.whl", hash = "sha256:7bd00153873b0c1ceb31ae45852991bb08c1785f9c89d30dec0c569378ea3020"},
+    {file = "wandb-0.11.2.tar.gz", hash = "sha256:324ee38bcc1baea13cf914d5b28b21519237e17ab13dc7cac0870e0291930a2e"},
 ]
 wcwidth = [
     {file = "wcwidth-0.2.5-py2.py3-none-any.whl", hash = "sha256:beb4802a9cebb9144e99086eff703a642a13d6a0052920003a230f3294bbe784"},
diff --git a/pyproject.toml b/pyproject.toml
index 6c5a2a0..7d81365 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -15,8 +15,6 @@ click = "^7.1.2"
 boltons = "^20.1.0"
 h5py = "^3.2.1"
 toml = "^0.10.1"
-torch = "^1.9.0"
-torchvision = "^0.10.0"
 loguru = "^0.5.0"
 matplotlib = "^3.2.1"
 tqdm = "^4.46.1"
@@ -24,7 +22,7 @@ opencv-python = "^4.3.0"
 nltk = "^3.5"
 torch-summary = "^1.4.2"
 defusedxml = "^0.6.0"
-omegaconf = "^2.0.2"
+omegaconf = "^2.1.0"
 einops = "^0.3.0"
 gtn = "^0.0.0"
 sentencepiece = "^0.1.95"
@@ -33,8 +31,10 @@ Pillow = "^8.1.2"
 madgrad = "^1.0"
 editdistance = "^0.5.3"
 torchmetrics = "^0.4.1"
-hydra-core = "^1.0.6"
+hydra-core = "^1.1.0"
 attr = "^0.3.1"
+torch = "^1.9.0"
+torchvision = "^0.10.0"
 
 [tool.poetry.dev-dependencies]
 pytest = "^5.4.2"
@@ -50,7 +50,7 @@ flake8-import-order = "^0.18.1"
 safety = "^1.9.0"
 mypy = "^0.770"
 typeguard = "^2.7.1"
-wandb = "^0.10.30"
+wandb = "^0.11.2"
 scipy = "^1.6.1"
 flake8-annotations = "^2.6.2"
 flake8-docstrings = "^1.6.0"
diff --git a/text_recognizer/criterions/label_smoothing.py b/text_recognizer/criterions/label_smoothing.py
index 40a7609..cc71c45 100644
--- a/text_recognizer/criterions/label_smoothing.py
+++ b/text_recognizer/criterions/label_smoothing.py
@@ -6,37 +6,31 @@ import torch.nn.functional as F
 
 
 class LabelSmoothingLoss(nn.Module):
-    """Label smoothing cross entropy loss."""
-
-    def __init__(
-        self, label_smoothing: float, vocab_size: int, ignore_index: int = -100
-    ) -> None:
-        assert 0.0 < label_smoothing <= 1.0
-        self.ignore_index = ignore_index
+    def __init__(self, ignore_index: int = -100, smoothing: float = 0.0, dim: int = -1):
         super().__init__()
+        assert 0.0 < smoothing <= 1.0
+        self.ignore_index = ignore_index
+        self.confidence = 1.0 - smoothing
+        self.smoothing = smoothing
+        self.dim = dim
 
-        smoothing_value = label_smoothing / (vocab_size - 2)
-        one_hot = torch.full((vocab_size,), smoothing_value)
-        one_hot[self.ignore_index] = 0
-        self.register_buffer("one_hot", one_hot.unsqueeze(0))
-
-        self.confidence = 1.0 - label_smoothing
-
-    def forward(self, output: Tensor, targets: Tensor) -> Tensor:
+    def forward(self, output: Tensor, target: Tensor) -> Tensor:
         """Computes the loss.
 
         Args:
-            output (Tensor): Predictions from the network.
+            output (Tensor): outputictions from the network.
             targets (Tensor): Ground truth.
 
         Shapes:
-            outpus: Batch size x num classes
-            targets: Batch size
+            TBC
 
         Returns:
             Tensor: Label smoothing loss.
         """
-        model_prob = self.one_hot.repeat(targets.size(0), 1)
-        model_prob.scatter_(1, targets.unsqueeze(1), self.confidence)
-        model_prob.masked_fill_((targets == self.ignore_index).unsqueeze(1), 0)
-        return F.kl_div(output, model_prob, reduction="sum")
+        output = output.log_softmax(dim=self.dim)
+        with torch.no_grad():
+            true_dist = torch.zeros_like(output)
+            true_dist.scatter_(1, target.unsqueeze(1), self.confidence)
+            true_dist.masked_fill_((target == 4).unsqueeze(1), 0)
+            true_dist += self.smoothing / output.size(self.dim)
+        return torch.mean(torch.sum(-true_dist * output, dim=self.dim))
diff --git a/text_recognizer/data/base_data_module.py b/text_recognizer/data/base_data_module.py
index fd914b6..16a06d9 100644
--- a/text_recognizer/data/base_data_module.py
+++ b/text_recognizer/data/base_data_module.py
@@ -1,12 +1,12 @@
 """Base lightning DataModule class."""
 from pathlib import Path
-from typing import Dict, Tuple
+from typing import Dict, Tuple, Type
 
 import attr
 from pytorch_lightning import LightningDataModule
 from torch.utils.data import DataLoader
 
-from text_recognizer.data.mappings import AbstractMapping
+from text_recognizer.data.base_mapping import AbstractMapping
 from text_recognizer.data.base_dataset import BaseDataset
 
 
@@ -25,7 +25,7 @@ class BaseDataModule(LightningDataModule):
     def __attrs_pre_init__(self) -> None:
         super().__init__()
 
-    mapping: AbstractMapping = attr.ib()
+    mapping: Type[AbstractMapping] = attr.ib()
     batch_size: int = attr.ib(default=16)
     num_workers: int = attr.ib(default=0)
     pin_memory: bool = attr.ib(default=True)
diff --git a/text_recognizer/data/base_mapping.py b/text_recognizer/data/base_mapping.py
new file mode 100644
index 0000000..572ac95
--- /dev/null
+++ b/text_recognizer/data/base_mapping.py
@@ -0,0 +1,37 @@
+"""Mapping to and from word pieces."""
+from abc import ABC, abstractmethod
+from typing import Dict, List
+
+from torch import Tensor
+
+
+class AbstractMapping(ABC):
+    def __init__(
+        self, input_size: List[int], mapping: List[str], inverse_mapping: Dict[str, int]
+    ) -> None:
+        self.input_size = input_size
+        self.mapping = mapping
+        self.inverse_mapping = inverse_mapping
+
+    def __len__(self) -> int:
+        return len(self.mapping)
+
+    @property
+    def num_classes(self) -> int:
+        return self.__len__()
+
+    @abstractmethod
+    def get_token(self, *args, **kwargs) -> str:
+        ...
+
+    @abstractmethod
+    def get_index(self, *args, **kwargs) -> Tensor:
+        ...
+
+    @abstractmethod
+    def get_text(self, *args, **kwargs) -> str:
+        ...
+
+    @abstractmethod
+    def get_indices(self, *args, **kwargs) -> Tensor:
+        ...
diff --git a/text_recognizer/data/download_utils.py b/text_recognizer/data/download_utils.py
index 8938830..a5a5360 100644
--- a/text_recognizer/data/download_utils.py
+++ b/text_recognizer/data/download_utils.py
@@ -1,7 +1,7 @@
 """Util functions for downloading datasets."""
 import hashlib
 from pathlib import Path
-from typing import Dict, List, Optional
+from typing import Dict, Optional
 from urllib.request import urlretrieve
 
 from loguru import logger as log
diff --git a/text_recognizer/data/emnist_mapping.py b/text_recognizer/data/emnist_mapping.py
new file mode 100644
index 0000000..6c4c43b
--- /dev/null
+++ b/text_recognizer/data/emnist_mapping.py
@@ -0,0 +1,37 @@
+"""Emnist mapping."""
+from typing import List, Optional, Union, Set
+
+from torch import Tensor
+
+from text_recognizer.data.base_mapping import AbstractMapping
+from text_recognizer.data.emnist import emnist_mapping
+
+
+class EmnistMapping(AbstractMapping):
+    def __init__(self, extra_symbols: Optional[Set[str]] = None) -> None:
+        self.extra_symbols = set(extra_symbols) if extra_symbols is not None else None
+        self.mapping, self.inverse_mapping, self.input_size = emnist_mapping(
+            self.extra_symbols
+        )
+        super().__init__(self.input_size, self.mapping, self.inverse_mapping)
+
+    def __attrs_post_init__(self) -> None:
+        """Post init configuration."""
+
+    def get_token(self, index: Union[int, Tensor]) -> str:
+        if (index := int(index)) in self.mapping:
+            return self.mapping[index]
+        raise KeyError(f"Index ({index}) not in mapping.")
+
+    def get_index(self, token: str) -> Tensor:
+        if token in self.inverse_mapping:
+            return Tensor(self.inverse_mapping[token])
+        raise KeyError(f"Token ({token}) not found in inverse mapping.")
+
+    def get_text(self, indices: Union[List[int], Tensor]) -> str:
+        if isinstance(indices, Tensor):
+            indices = indices.tolist()
+        return "".join([self.mapping[index] for index in indices])
+
+    def get_indices(self, text: str) -> Tensor:
+        return Tensor([self.inverse_mapping[token] for token in text])
diff --git a/text_recognizer/data/iam_extended_paragraphs.py b/text_recognizer/data/iam_extended_paragraphs.py
index ccf0759..df0c0e1 100644
--- a/text_recognizer/data/iam_extended_paragraphs.py
+++ b/text_recognizer/data/iam_extended_paragraphs.py
@@ -1,6 +1,4 @@
 """IAM original and sythetic dataset class."""
-from typing import Dict, List
-
 import attr
 from torch.utils.data import ConcatDataset
 
@@ -15,7 +13,6 @@ class IAMExtendedParagraphs(BaseDataModule):
     augment: bool = attr.ib(default=True)
     train_fraction: float = attr.ib(default=0.8)
     word_pieces: bool = attr.ib(default=False)
-    num_classes: int = attr.ib(init=False)
 
     def __attrs_post_init__(self) -> None:
         self.iam_paragraphs = IAMParagraphs(
diff --git a/text_recognizer/data/iam_lines.py b/text_recognizer/data/iam_lines.py
index 1c63729..aba38f9 100644
--- a/text_recognizer/data/iam_lines.py
+++ b/text_recognizer/data/iam_lines.py
@@ -22,7 +22,7 @@ from text_recognizer.data.base_dataset import (
     split_dataset,
 )
 from text_recognizer.data.base_data_module import BaseDataModule, load_and_print_info
-from text_recognizer.data.mappings import EmnistMapping
+from text_recognizer.data.emnist_mapping import EmnistMapping
 from text_recognizer.data.iam import IAM
 from text_recognizer.data import image_utils
 
diff --git a/text_recognizer/data/iam_paragraphs.py b/text_recognizer/data/iam_paragraphs.py
index 6189f7d..11f899f 100644
--- a/text_recognizer/data/iam_paragraphs.py
+++ b/text_recognizer/data/iam_paragraphs.py
@@ -17,7 +17,7 @@ from text_recognizer.data.base_dataset import (
     split_dataset,
 )
 from text_recognizer.data.base_data_module import BaseDataModule, load_and_print_info
-from text_recognizer.data.mappings import EmnistMapping
+from text_recognizer.data.emnist_mapping import EmnistMapping
 from text_recognizer.data.iam import IAM
 from text_recognizer.data.transforms import WordPiece
 
@@ -50,11 +50,9 @@ class IAMParagraphs(BaseDataModule):
         if PROCESSED_DATA_DIRNAME.exists():
             return
 
-        log.info(
-            "Cropping IAM paragraph regions and saving them along with labels..."
-        )
+        log.info("Cropping IAM paragraph regions and saving them along with labels...")
 
-        iam = IAM(mapping=EmnistMapping())
+        iam = IAM(mapping=EmnistMapping(extra_symbols={NEW_LINE_TOKEN,}))
         iam.prepare_data()
 
         properties = {}
@@ -83,7 +81,9 @@ class IAMParagraphs(BaseDataModule):
             crops, labels = _load_processed_crops_and_labels(split)
             data = [resize_image(crop, IMAGE_SCALE_FACTOR) for crop in crops]
             targets = convert_strings_to_labels(
-                strings=labels, mapping=self.mapping.inverse_mapping, length=self.output_dims[0]
+                strings=labels,
+                mapping=self.mapping.inverse_mapping,
+                length=self.output_dims[0],
             )
             return BaseDataset(
                 data,
diff --git a/text_recognizer/data/iam_synthetic_paragraphs.py b/text_recognizer/data/iam_synthetic_paragraphs.py
index c938f8b..24ca896 100644
--- a/text_recognizer/data/iam_synthetic_paragraphs.py
+++ b/text_recognizer/data/iam_synthetic_paragraphs.py
@@ -21,7 +21,7 @@ from text_recognizer.data.iam_paragraphs import (
     IMAGE_SCALE_FACTOR,
     resize_image,
 )
-from text_recognizer.data.mappings import EmnistMapping
+from text_recognizer.data.emnist_mapping import EmnistMapping
 from text_recognizer.data.iam import IAM
 from text_recognizer.data.iam_lines import (
     line_crops_and_labels,
@@ -47,7 +47,7 @@ class IAMSyntheticParagraphs(IAMParagraphs):
         log.info("Preparing IAM lines for synthetic paragraphs dataset.")
         log.info("Cropping IAM line regions and loading labels.")
 
-        iam = IAM(mapping=EmnistMapping())
+        iam = IAM(mapping=EmnistMapping(extra_symbols={NEW_LINE_TOKEN,}))
         iam.prepare_data()
 
         crops_train, labels_train = line_crops_and_labels(iam, "train")
diff --git a/text_recognizer/data/make_wordpieces.py b/text_recognizer/data/make_wordpieces.py
index 40fbee4..8e53815 100644
--- a/text_recognizer/data/make_wordpieces.py
+++ b/text_recognizer/data/make_wordpieces.py
@@ -13,8 +13,6 @@ import click
 from loguru import logger as log
 import sentencepiece as spm
 
-from text_recognizer.data.iam_preprocessor import load_metadata
-
 
 def iamdb_pieces(
     data_dir: Path, text_file: str, num_pieces: int, output_prefix: str
diff --git a/text_recognizer/data/mappings.py b/text_recognizer/data/mappings.py
deleted file mode 100644
index d1c64dd..0000000
--- a/text_recognizer/data/mappings.py
+++ /dev/null
@@ -1,156 +0,0 @@
-"""Mapping to and from word pieces."""
-from abc import ABC, abstractmethod
-from pathlib import Path
-from typing import Dict, List, Optional, Union, Set
-
-import attr
-import torch
-from loguru import logger as log
-from torch import Tensor
-
-from text_recognizer.data.emnist import emnist_mapping
-from text_recognizer.data.iam_preprocessor import Preprocessor
-
-
-@attr.s
-class AbstractMapping(ABC):
-    input_size: List[int] = attr.ib(init=False)
-    mapping: List[str] = attr.ib(init=False)
-    inverse_mapping: Dict[str, int] = attr.ib(init=False)
-
-    def __len__(self) -> int:
-        return len(self.mapping)
-
-    @property
-    def num_classes(self) -> int:
-        return self.__len__()
-
-    @abstractmethod
-    def get_token(self, *args, **kwargs) -> str:
-        ...
-
-    @abstractmethod
-    def get_index(self, *args, **kwargs) -> Tensor:
-        ...
-
-    @abstractmethod
-    def get_text(self, *args, **kwargs) -> str:
-        ...
-
-    @abstractmethod
-    def get_indices(self, *args, **kwargs) -> Tensor:
-        ...
-
-
-@attr.s(auto_attribs=True)
-class EmnistMapping(AbstractMapping):
-    extra_symbols: Optional[Set[str]] = attr.ib(default=None)
-
-    def __attrs_post_init__(self) -> None:
-        """Post init configuration."""
-        self.extra_symbols = set(self.extra_symbols) if self.extra_symbols is not None else None
-        self.mapping, self.inverse_mapping, self.input_size = emnist_mapping(
-            self.extra_symbols
-        )
-
-    def get_token(self, index: Union[int, Tensor]) -> str:
-        if (index := int(index)) in self.mapping:
-            return self.mapping[index]
-        raise KeyError(f"Index ({index}) not in mapping.")
-
-    def get_index(self, token: str) -> Tensor:
-        if token in self.inverse_mapping:
-            return Tensor(self.inverse_mapping[token])
-        raise KeyError(f"Token ({token}) not found in inverse mapping.")
-
-    def get_text(self, indices: Union[List[int], Tensor]) -> str:
-        if isinstance(indices, Tensor):
-            indices = indices.tolist()
-        return "".join([self.mapping[index] for index in indices])
-
-    def get_indices(self, text: str) -> Tensor:
-        return Tensor([self.inverse_mapping[token] for token in text])
-
-
-@attr.s(auto_attribs=True)
-class WordPieceMapping(EmnistMapping):
-    data_dir: Optional[Path] = attr.ib(default=None)
-    num_features: int = attr.ib(default=1000)
-    tokens: str = attr.ib(default="iamdb_1kwp_tokens_1000.txt")
-    lexicon: str = attr.ib(default="iamdb_1kwp_lex_1000.txt")
-    use_words: bool = attr.ib(default=False)
-    prepend_wordsep: bool = attr.ib(default=False)
-    special_tokens: Set[str] = attr.ib(default={"<s>", "<e>", "<p>"}, converter=set)
-    extra_symbols: Set[str] = attr.ib(default={"\n",}, converter=set)
-    wordpiece_processor: Preprocessor = attr.ib(init=False)
-
-    def __attrs_post_init__(self) -> None:
-        super().__attrs_post_init__()
-        self.data_dir = (
-            (
-                Path(__file__).resolve().parents[2]
-                / "data"
-                / "downloaded"
-                / "iam"
-                / "iamdb"
-            )
-            if self.data_dir is None
-            else Path(self.data_dir)
-        )
-        log.debug(f"Using data dir: {self.data_dir}")
-        if not self.data_dir.exists():
-            raise RuntimeError(f"Could not locate iamdb directory at {self.data_dir}")
-
-        processed_path = (
-            Path(__file__).resolve().parents[2] / "data" / "processed" / "iam_lines"
-        )
-
-        tokens_path = processed_path / self.tokens
-        lexicon_path = processed_path / self.lexicon
-
-        special_tokens = self.special_tokens
-        if self.extra_symbols is not None:
-            special_tokens = special_tokens | self.extra_symbols
-
-        self.wordpiece_processor = Preprocessor(
-            data_dir=self.data_dir,
-            num_features=self.num_features,
-            tokens_path=tokens_path,
-            lexicon_path=lexicon_path,
-            use_words=self.use_words,
-            prepend_wordsep=self.prepend_wordsep,
-            special_tokens=special_tokens,
-        )
-
-    def __len__(self) -> int:
-        return len(self.wordpiece_processor.tokens)
-
-    def get_token(self, index: Union[int, Tensor]) -> str:
-        if (index := int(index)) <= self.wordpiece_processor.num_tokens:
-            return self.wordpiece_processor.tokens[index]
-        raise KeyError(f"Index ({index}) not in mapping.")
-
-    def get_index(self, token: str) -> Tensor:
-        if token in self.wordpiece_processor.tokens:
-            return torch.LongTensor([self.wordpiece_processor.tokens_to_index[token]])
-        raise KeyError(f"Token ({token}) not found in inverse mapping.")
-
-    def get_text(self, indices: Union[List[int], Tensor]) -> str:
-        if isinstance(indices, Tensor):
-            indices = indices.tolist()
-        return self.wordpiece_processor.to_text(indices)
-
-    def get_indices(self, text: str) -> Tensor:
-        return self.wordpiece_processor.to_index(text)
-
-    def emnist_to_wordpiece_indices(self, x: Tensor) -> Tensor:
-        text = "".join([self.mapping[i] for i in x])
-        text = text.lower().replace(" ", "▁")
-        return torch.LongTensor(self.wordpiece_processor.to_index(text))
-
-    def __getitem__(self, x: Union[str, int, List[int], Tensor]) -> Union[str, Tensor]:
-        if isinstance(x, int):
-            x = [x]
-        if isinstance(x, str):
-            return self.get_indices(x)
-        return self.get_text(x)
diff --git a/text_recognizer/data/transforms.py b/text_recognizer/data/transforms.py
index 3b1b929..047496f 100644
--- a/text_recognizer/data/transforms.py
+++ b/text_recognizer/data/transforms.py
@@ -1,11 +1,11 @@
 """Transforms for PyTorch datasets."""
 from pathlib import Path
-from typing import Optional, Union, Sequence
+from typing import Optional, Union, Set
 
 import torch
 from torch import Tensor
 
-from text_recognizer.data.mappings import WordPieceMapping
+from text_recognizer.data.word_piece_mapping import WordPieceMapping
 
 
 class WordPiece:
@@ -19,8 +19,8 @@ class WordPiece:
         data_dir: Optional[Union[str, Path]] = None,
         use_words: bool = False,
         prepend_wordsep: bool = False,
-        special_tokens: Sequence[str] = ("<s>", "<e>", "<p>"),
-        extra_symbols: Optional[Sequence[str]] = ("\n",),
+        special_tokens: Set[str] = {"<s>", "<e>", "<p>"},
+        extra_symbols: Optional[Set[str]] = {"\n",},
         max_len: int = 451,
     ) -> None:
         self.mapping = WordPieceMapping(
diff --git a/text_recognizer/data/word_piece_mapping.py b/text_recognizer/data/word_piece_mapping.py
new file mode 100644
index 0000000..59488c3
--- /dev/null
+++ b/text_recognizer/data/word_piece_mapping.py
@@ -0,0 +1,93 @@
+"""Word piece mapping."""
+from pathlib import Path
+from typing import List, Optional, Union, Set
+
+import torch
+from loguru import logger as log
+from torch import Tensor
+
+from text_recognizer.data.emnist_mapping import EmnistMapping
+from text_recognizer.data.iam_preprocessor import Preprocessor
+
+
+class WordPieceMapping(EmnistMapping):
+    def __init__(
+        self,
+        data_dir: Optional[Path] = None,
+        num_features: int = 1000,
+        tokens: str = "iamdb_1kwp_tokens_1000.txt",
+        lexicon: str = "iamdb_1kwp_lex_1000.txt",
+        use_words: bool = False,
+        prepend_wordsep: bool = False,
+        special_tokens: Set[str] = {"<s>", "<e>", "<p>"},
+        extra_symbols: Set[str] = {"\n",},
+    ) -> None:
+        super().__init__(extra_symbols=extra_symbols)
+        self.data_dir = (
+            (
+                Path(__file__).resolve().parents[2]
+                / "data"
+                / "downloaded"
+                / "iam"
+                / "iamdb"
+            )
+            if data_dir is None
+            else Path(data_dir)
+        )
+        log.debug(f"Using data dir: {self.data_dir}")
+        if not self.data_dir.exists():
+            raise RuntimeError(f"Could not locate iamdb directory at {self.data_dir}")
+
+        processed_path = (
+            Path(__file__).resolve().parents[2] / "data" / "processed" / "iam_lines"
+        )
+
+        tokens_path = processed_path / tokens
+        lexicon_path = processed_path / lexicon
+
+        special_tokens = set(special_tokens)
+        if self.extra_symbols is not None:
+            special_tokens = special_tokens | set(extra_symbols)
+
+        self.wordpiece_processor = Preprocessor(
+            data_dir=self.data_dir,
+            num_features=num_features,
+            tokens_path=tokens_path,
+            lexicon_path=lexicon_path,
+            use_words=use_words,
+            prepend_wordsep=prepend_wordsep,
+            special_tokens=special_tokens,
+        )
+
+    def __len__(self) -> int:
+        return len(self.wordpiece_processor.tokens)
+
+    def get_token(self, index: Union[int, Tensor]) -> str:
+        if (index := int(index)) <= self.wordpiece_processor.num_tokens:
+            return self.wordpiece_processor.tokens[index]
+        raise KeyError(f"Index ({index}) not in mapping.")
+
+    def get_index(self, token: str) -> Tensor:
+        if token in self.wordpiece_processor.tokens:
+            return torch.LongTensor([self.wordpiece_processor.tokens_to_index[token]])
+        raise KeyError(f"Token ({token}) not found in inverse mapping.")
+
+    def get_text(self, indices: Union[List[int], Tensor]) -> str:
+        if isinstance(indices, Tensor):
+            indices = indices.tolist()
+        return self.wordpiece_processor.to_text(indices).replace(" ", "▁")
+
+    def get_indices(self, text: str) -> Tensor:
+        return self.wordpiece_processor.to_index(text)
+
+    def emnist_to_wordpiece_indices(self, x: Tensor) -> Tensor:
+        text = "".join([self.mapping[i] for i in x])
+        text = text.lower().replace(" ", "▁")
+        return torch.LongTensor(self.wordpiece_processor.to_index(text))
+
+    def __getitem__(self, x: Union[str, int, List[int], Tensor]) -> Union[str, Tensor]:
+        if isinstance(x, int):
+            x = [x]
+        if isinstance(x, str):
+            return self.get_indices(x)
+        return self.get_text(x)
diff --git a/text_recognizer/models/base.py b/text_recognizer/models/base.py
index 8ce5c37..57c5964 100644
--- a/text_recognizer/models/base.py
+++ b/text_recognizer/models/base.py
@@ -11,6 +11,8 @@ from torch import nn
 from torch import Tensor
 import torchmetrics
 
+from text_recognizer.data.base_mapping import AbstractMapping
+
 
 @attr.s(eq=False)
 class BaseLitModel(LightningModule):
@@ -20,12 +22,12 @@ class BaseLitModel(LightningModule):
         super().__init__()
 
     network: Type[nn.Module] = attr.ib()
-    criterion_config: DictConfig = attr.ib(converter=DictConfig)
-    optimizer_config: DictConfig = attr.ib(converter=DictConfig)
-    lr_scheduler_config: DictConfig = attr.ib(converter=DictConfig)
+    mapping: Type[AbstractMapping] = attr.ib()
+    loss_fn: Type[nn.Module] = attr.ib()
+    optimizer_config: DictConfig = attr.ib()
+    lr_scheduler_config: DictConfig = attr.ib()
     interval: str = attr.ib()
     monitor: str = attr.ib(default="val/loss")
-    loss_fn: Type[nn.Module] = attr.ib(init=False)
     train_acc: torchmetrics.Accuracy = attr.ib(
         init=False, default=torchmetrics.Accuracy()
     )
@@ -36,12 +38,6 @@ class BaseLitModel(LightningModule):
         init=False, default=torchmetrics.Accuracy()
     )
 
-    @loss_fn.default
-    def configure_criterion(self) -> Type[nn.Module]:
-        """Returns a loss functions."""
-        log.info(f"Instantiating criterion <{self.criterion_config._target_}>")
-        return hydra.utils.instantiate(self.criterion_config)
-
     def optimizer_zero_grad(
         self,
         epoch: int,
@@ -54,7 +50,9 @@ class BaseLitModel(LightningModule):
     def _configure_optimizer(self) -> Type[torch.optim.Optimizer]:
         """Configures the optimizer."""
         log.info(f"Instantiating optimizer <{self.optimizer_config._target_}>")
-        return hydra.utils.instantiate(self.optimizer_config, params=self.parameters())
+        return hydra.utils.instantiate(
+            self.optimizer_config, params=self.network.parameters()
+        )
 
     def _configure_lr_scheduler(
         self, optimizer: Type[torch.optim.Optimizer]
diff --git a/text_recognizer/models/transformer.py b/text_recognizer/models/transformer.py
index 91e088d..5fb84a7 100644
--- a/text_recognizer/models/transformer.py
+++ b/text_recognizer/models/transformer.py
@@ -5,7 +5,6 @@ import attr
 import torch
 from torch import Tensor
 
-from text_recognizer.data.mappings import AbstractMapping
 from text_recognizer.models.metrics import CharacterErrorRate
 from text_recognizer.models.base import BaseLitModel
 
@@ -14,14 +13,14 @@ from text_recognizer.models.base import BaseLitModel
 class TransformerLitModel(BaseLitModel):
     """A PyTorch Lightning model for transformer networks."""
 
-    mapping: Type[AbstractMapping] = attr.ib(default=None)
+    max_output_len: int = attr.ib(default=451)
     start_token: str = attr.ib(default="<s>")
     end_token: str = attr.ib(default="<e>")
     pad_token: str = attr.ib(default="<p>")
 
-    start_index: Tensor = attr.ib(init=False)
-    end_index: Tensor = attr.ib(init=False)
-    pad_index: Tensor = attr.ib(init=False)
+    start_index: int = attr.ib(init=False)
+    end_index: int = attr.ib(init=False)
+    pad_index: int = attr.ib(init=False)
 
     ignore_indices: Set[Tensor] = attr.ib(init=False)
     val_cer: CharacterErrorRate = attr.ib(init=False)
@@ -29,9 +28,9 @@ class TransformerLitModel(BaseLitModel):
 
     def __attrs_post_init__(self) -> None:
         """Post init configuration."""
-        self.start_index = self.mapping.get_index(self.start_token)
-        self.end_index = self.mapping.get_index(self.end_token)
-        self.pad_index = self.mapping.get_index(self.pad_token)
+        self.start_index = int(self.mapping.get_index(self.start_token))
+        self.end_index = int(self.mapping.get_index(self.end_token))
+        self.pad_index = int(self.mapping.get_index(self.pad_token))
         self.ignore_indices = set([self.start_index, self.end_index, self.pad_index])
         self.val_cer = CharacterErrorRate(self.ignore_indices)
         self.test_cer = CharacterErrorRate(self.ignore_indices)
@@ -93,23 +92,24 @@ class TransformerLitModel(BaseLitModel):
         output = torch.ones((bsz, self.max_output_len), dtype=torch.long).to(x.device)
         output[:, 0] = self.start_index
 
-        for i in range(1, self.max_output_len):
-            context = output[:, :i]  # (bsz, i)
-            logits = self.network.decode(z, context)  # (i, bsz, c)
-            tokens = torch.argmax(logits, dim=-1)  # (i, bsz)
-            output[:, i : i + 1] = tokens[-1:]
+        for Sy in range(1, self.max_output_len):
+            context = output[:, :Sy]  # (B, Sy)
+            logits = self.network.decode(z, context)  # (B, Sy, C)
+            tokens = torch.argmax(logits, dim=-1)  # (B, Sy)
+            output[:, Sy : Sy + 1] = tokens[:, -1:]
 
             # Early stopping of prediction loop if token is end or padding token.
             if (
-                output[:, i - 1] == self.end_index | output[: i - 1] == self.pad_index
+                (output[:, Sy - 1] == self.end_index)
+                | (output[:, Sy - 1] == self.pad_index)
             ).all():
                 break
 
         # Set all tokens after end token to pad token.
-        for i in range(1, self.max_output_len):
-            idx = (
-                output[:, i - 1] == self.end_index | output[:, i - 1] == self.pad_index
+        for Sy in range(1, self.max_output_len):
+            idx = (output[:, Sy - 1] == self.end_index) | (
+                output[:, Sy - 1] == self.pad_index
             )
-            output[idx, i] = self.pad_index
+            output[idx, Sy] = self.pad_index
 
         return output
diff --git a/text_recognizer/networks/conv_transformer.py b/text_recognizer/networks/conv_transformer.py
index 09cc654..f3ba49d 100644
--- a/text_recognizer/networks/conv_transformer.py
+++ b/text_recognizer/networks/conv_transformer.py
@@ -2,7 +2,6 @@
 import math
 from typing import Tuple
 
-import attr
 from torch import nn, Tensor
 
 from text_recognizer.networks.encoders.efficientnet import EfficientNet
@@ -13,32 +12,28 @@ from text_recognizer.networks.transformer.positional_encodings import (
 )
 
 
-@attr.s(eq=False)
 class ConvTransformer(nn.Module):
     """Convolutional encoder and transformer decoder network."""
 
-    def __attrs_pre_init__(self) -> None:
+    def __init__(
+        self,
+        input_dims: Tuple[int, int, int],
+        hidden_dim: int,
+        dropout_rate: float,
+        num_classes: int,
+        pad_index: Tensor,
+        encoder: EfficientNet,
+        decoder: Decoder,
+    ) -> None:
         super().__init__()
+        self.input_dims = input_dims
+        self.hidden_dim = hidden_dim
+        self.dropout_rate = dropout_rate
+        self.num_classes = num_classes
+        self.pad_index = pad_index
+        self.encoder = encoder
+        self.decoder = decoder
 
-    # Parameters and placeholders,
-    input_dims: Tuple[int, int, int] = attr.ib()
-    hidden_dim: int = attr.ib()
-    dropout_rate: float = attr.ib()
-    max_output_len: int = attr.ib()
-    num_classes: int = attr.ib()
-    pad_index: Tensor = attr.ib()
-
-    # Modules.
-    encoder: EfficientNet = attr.ib()
-    decoder: Decoder = attr.ib()
-
-    latent_encoder: nn.Sequential = attr.ib(init=False)
-    token_embedding: nn.Embedding = attr.ib(init=False)
-    token_pos_encoder: PositionalEncoding = attr.ib(init=False)
-    head: nn.Linear = attr.ib(init=False)
-
-    def __attrs_post_init__(self) -> None:
-        """Post init configuration."""
         # Latent projector for down sampling number of filters and 2d
         # positional encoding.
         self.latent_encoder = nn.Sequential(
@@ -126,7 +121,8 @@ class ConvTransformer(nn.Module):
         context = self.token_embedding(context) * math.sqrt(self.hidden_dim)
         context = self.token_pos_encoder(context)
         out = self.decoder(x=context, context=z, mask=context_mask)
-        logits = self.head(out)
+        logits = self.head(out)  # [B, Sy, T]
+        logits = logits.permute(0, 2, 1)  # [B, T, Sy]
         return logits
 
     def forward(self, x: Tensor, context: Tensor) -> Tensor:
diff --git a/text_recognizer/networks/encoders/efficientnet/mbconv.py b/text_recognizer/networks/encoders/efficientnet/mbconv.py
index e85df87..7bfd9ba 100644
--- a/text_recognizer/networks/encoders/efficientnet/mbconv.py
+++ b/text_recognizer/networks/encoders/efficientnet/mbconv.py
@@ -11,9 +11,7 @@ from text_recognizer.networks.encoders.efficientnet.utils import stochastic_dept
 
 def _convert_stride(stride: Union[Tuple[int, int], int]) -> Tuple[int, int]:
     """Converts int to tuple."""
-    return (
-        (stride,) * 2 if isinstance(stride, int) else stride
-    )
+    return (stride,) * 2 if isinstance(stride, int) else stride
 
 
 @attr.s(eq=False)
@@ -41,10 +39,7 @@ class MBConvBlock(nn.Module):
     def _configure_padding(self) -> Tuple[int, int, int, int]:
         """Set padding for convolutional layers."""
         if self.stride == (2, 2):
-            return (
-                (self.kernel_size - 1) // 2 - 1,
-                (self.kernel_size - 1) // 2,
-                ) * 2
+            return ((self.kernel_size - 1) // 2 - 1, (self.kernel_size - 1) // 2,) * 2
         return ((self.kernel_size - 1) // 2,) * 4
 
     def __attrs_post_init__(self) -> None:
diff --git a/text_recognizer/networks/transformer/layers.py b/text_recognizer/networks/transformer/layers.py
index ce443e5..70a0ac7 100644
--- a/text_recognizer/networks/transformer/layers.py
+++ b/text_recognizer/networks/transformer/layers.py
@@ -1,5 +1,4 @@
 """Transformer attention layer."""
-from functools import partial
 from typing import Any, Dict, Optional, Tuple
 
 import attr
@@ -27,25 +26,17 @@ class AttentionLayers(nn.Module):
     norm_fn: str = attr.ib()
     ff_fn: str = attr.ib()
     ff_kwargs: Dict = attr.ib()
+    rotary_emb: Optional[RotaryEmbedding] = attr.ib()
     causal: bool = attr.ib(default=False)
     cross_attend: bool = attr.ib(default=False)
     pre_norm: bool = attr.ib(default=True)
-    rotary_emb: Optional[RotaryEmbedding] = attr.ib(default=None)
     layer_types: Tuple[str, ...] = attr.ib(init=False)
     layers: nn.ModuleList = attr.ib(init=False)
-    attn: partial = attr.ib(init=False)
-    norm: partial = attr.ib(init=False)
-    ff: partial = attr.ib(init=False)
 
     def __attrs_post_init__(self) -> None:
         """Post init configuration."""
         self.layer_types = self._get_layer_types() * self.depth
-        attn = load_partial_fn(
-            self.attn_fn, dim=self.dim, num_heads=self.num_heads, **self.attn_kwargs
-        )
-        norm = load_partial_fn(self.norm_fn, normalized_shape=self.dim)
-        ff = load_partial_fn(self.ff_fn, dim=self.dim, **self.ff_kwargs)
-        self.layers = self._build_network(attn, norm, ff)
+        self.layers = self._build_network()
 
     def _get_layer_types(self) -> Tuple:
         """Get layer specification."""
@@ -53,10 +44,13 @@ class AttentionLayers(nn.Module):
             return "a", "c", "f"
         return "a", "f"
 
-    def _build_network(
-        self, attn: partial, norm: partial, ff: partial,
-    ) -> nn.ModuleList:
+    def _build_network(self) -> nn.ModuleList:
         """Configures transformer network."""
+        attn = load_partial_fn(
+            self.attn_fn, dim=self.dim, num_heads=self.num_heads, **self.attn_kwargs
+        )
+        norm = load_partial_fn(self.norm_fn, normalized_shape=self.dim)
+        ff = load_partial_fn(self.ff_fn, dim=self.dim, **self.ff_kwargs)
         layers = nn.ModuleList([])
         for layer_type in self.layer_types:
             if layer_type == "a":
@@ -106,6 +100,7 @@ class Encoder(AttentionLayers):
     causal: bool = attr.ib(default=False, init=False)
 
 
-@attr.s(auto_attribs=True, eq=False)
 class Decoder(AttentionLayers):
-    causal: bool = attr.ib(default=True, init=False)
+    def __init__(self, **kwargs: Any) -> None:
+        assert "causal" not in kwargs, "Cannot set causality on decoder"
+        super().__init__(causal=True, **kwargs)
diff --git a/training/conf/callbacks/wandb/image_reconstructions.yaml b/training/__init__.py
index e69de29..e69de29 100644
--- a/training/conf/callbacks/wandb/image_reconstructions.yaml
+++ b/training/__init__.py
diff --git a/training/callbacks/wandb_callbacks.py b/training/callbacks/wandb_callbacks.py
index 6379cc0..906531f 100644
--- a/training/callbacks/wandb_callbacks.py
+++ b/training/callbacks/wandb_callbacks.py
@@ -1,11 +1,10 @@
 """Weights and Biases callbacks."""
 from pathlib import Path
-from typing import List
 
-import attr
 import wandb
 from pytorch_lightning import Callback, LightningModule, Trainer
 from pytorch_lightning.loggers import LoggerCollection, WandbLogger
+from pytorch_lightning.utilities import rank_zero_only
 
 
 def get_wandb_logger(trainer: Trainer) -> WandbLogger:
@@ -22,31 +21,27 @@ def get_wandb_logger(trainer: Trainer) -> WandbLogger:
     raise Exception("Weight and Biases logger not found for some reason...")
 
 
-@attr.s
 class WatchModel(Callback):
     """Make W&B watch the model at the beginning of the run."""
 
-    log: str = attr.ib(default="gradients")
-    log_freq: int = attr.ib(default=100)
-
-    def __attrs_pre_init__(self) -> None:
-        super().__init__()
+    def __init__(self, log: str = "gradients", log_freq: int = 100) -> None:
+        self.log = log
+        self.log_freq = log_freq
 
+    @rank_zero_only
     def on_train_start(self, trainer: Trainer, pl_module: LightningModule) -> None:
         """Watches model weights with wandb."""
         logger = get_wandb_logger(trainer)
         logger.watch(model=trainer.model, log=self.log, log_freq=self.log_freq)
 
 
-@attr.s
 class UploadCodeAsArtifact(Callback):
     """Upload all *.py files to W&B as an artifact, at the beginning of the run."""
 
-    project_dir: Path = attr.ib(converter=Path)
-
-    def __attrs_pre_init__(self) -> None:
-        super().__init__()
+    def __init__(self, project_dir: str) -> None:
+        self.project_dir = Path(project_dir)
 
+    @rank_zero_only
     def on_train_start(self, trainer: Trainer, pl_module: LightningModule) -> None:
         """Uploads project code as an artifact."""
         logger = get_wandb_logger(trainer)
@@ -58,16 +53,16 @@ class UploadCodeAsArtifact(Callback):
         experiment.use_artifact(artifact)
 
 
-@attr.s
-class UploadCheckpointAsArtifact(Callback):
+class UploadCheckpointsAsArtifact(Callback):
     """Upload checkpoint to wandb as an artifact, at the end of a run."""
 
-    ckpt_dir: Path = attr.ib(converter=Path)
-    upload_best_only: bool = attr.ib()
-
-    def __attrs_pre_init__(self) -> None:
-        super().__init__()
+    def __init__(
+        self, ckpt_dir: str = "checkpoints/", upload_best_only: bool = False
+    ) -> None:
+        self.ckpt_dir = ckpt_dir
+        self.upload_best_only = upload_best_only
 
+    @rank_zero_only
     def on_train_end(self, trainer: Trainer, pl_module: LightningModule) -> None:
         """Uploads model checkpoint to W&B."""
         logger = get_wandb_logger(trainer)
@@ -83,15 +78,12 @@ class UploadCheckpointAsArtifact(Callback):
         experiment.use_artifact(ckpts)
 
 
-@attr.s
 class LogTextPredictions(Callback):
     """Logs a validation batch with image to text transcription."""
 
-    num_samples: int = attr.ib(default=8)
-    ready: bool = attr.ib(default=True)
-
-    def __attrs_pre_init__(self) -> None:
-        super().__init__()
+    def __init__(self, num_samples: int = 8) -> None:
+        self.num_samples = num_samples
+        self.ready = False
 
     def _log_predictions(
         self, stage: str, trainer: Trainer, pl_module: LightningModule
@@ -111,20 +103,20 @@ class LogTextPredictions(Callback):
         logits = pl_module(imgs)
 
         mapping = pl_module.mapping
+        columns = ["id", "image", "prediction", "truth"]
+        data = [
+            [id, wandb.Image(img), mapping.get_text(pred), mapping.get_text(label)]
+            for id, (img, pred, label) in enumerate(
+                zip(
+                    imgs[: self.num_samples],
+                    logits[: self.num_samples],
+                    labels[: self.num_samples],
+                )
+            )
+        ]
+
         experiment.log(
-            {
-                f"OCR/{experiment.name}/{stage}": [
-                    wandb.Image(
-                        img,
-                        caption=f"Pred: {mapping.get_text(pred)}, Label: {mapping.get_text(label)}",
-                    )
-                    for img, pred, label in zip(
-                        imgs[: self.num_samples],
-                        logits[: self.num_samples],
-                        labels[: self.num_samples],
-                    )
-                ]
-            }
+            {f"OCR/{experiment.name}/{stage}": wandb.Table(data=data, columns=columns)}
         )
 
     def on_sanity_check_start(
@@ -143,20 +135,17 @@ class LogTextPredictions(Callback):
         """Logs predictions on validation epoch end."""
         self._log_predictions(stage="val", trainer=trainer, pl_module=pl_module)
 
-    def on_train_epoch_end(self, trainer: Trainer, pl_module: LightningModule) -> None:
+    def on_test_epoch_end(self, trainer: Trainer, pl_module: LightningModule) -> None:
         """Logs predictions on train epoch end."""
         self._log_predictions(stage="test", trainer=trainer, pl_module=pl_module)
 
 
-@attr.s
 class LogReconstuctedImages(Callback):
     """Log reconstructions of images."""
 
-    num_samples: int = attr.ib(default=8)
-    ready: bool = attr.ib(default=True)
-
-    def __attrs_pre_init__(self) -> None:
-        super().__init__()
+    def __init__(self, num_samples: int = 8) -> None:
+        self.num_samples = num_samples
+        self.ready = False
 
     def _log_reconstruction(
         self, stage: str, trainer: Trainer, pl_module: LightningModule
@@ -202,6 +191,6 @@ class LogReconstuctedImages(Callback):
         """Logs predictions on validation epoch end."""
         self._log_reconstruction(stage="val", trainer=trainer, pl_module=pl_module)
 
-    def on_train_epoch_end(self, trainer: Trainer, pl_module: LightningModule) -> None:
+    def on_test_epoch_end(self, trainer: Trainer, pl_module: LightningModule) -> None:
         """Logs predictions on train epoch end."""
         self._log_reconstruction(stage="test", trainer=trainer, pl_module=pl_module)
diff --git a/training/conf/callbacks/checkpoint.yaml b/training/conf/callbacks/checkpoint.yaml
index db34cb1..b4101d8 100644
--- a/training/conf/callbacks/checkpoint.yaml
+++ b/training/conf/callbacks/checkpoint.yaml
@@ -6,4 +6,4 @@ model_checkpoint:
   mode: min # can be "max" or "min"
   verbose: false
   dirpath: checkpoints/
-  filename: {epoch:02d}
+  filename: "{epoch:02d}"
diff --git a/training/conf/callbacks/wandb/checkpoints.yaml b/training/conf/callbacks/wandb_checkpoints.yaml
index a4a16ff..a4a16ff 100644
--- a/training/conf/callbacks/wandb/checkpoints.yaml
+++ b/training/conf/callbacks/wandb_checkpoints.yaml
diff --git a/training/conf/callbacks/wandb/code.yaml b/training/conf/callbacks/wandb_code.yaml
index 35f6ea3..35f6ea3 100644
--- a/training/conf/callbacks/wandb/code.yaml
+++ b/training/conf/callbacks/wandb_code.yaml
diff --git a/training/conf/callbacks/wandb_image_reconstructions.yaml b/training/conf/callbacks/wandb_image_reconstructions.yaml
new file mode 100644
index 0000000..e69de29
--- /dev/null
+++ b/training/conf/callbacks/wandb_image_reconstructions.yaml
diff --git a/training/conf/callbacks/wandb_ocr.yaml b/training/conf/callbacks/wandb_ocr.yaml
index efa3dda..9c9a6da 100644
--- a/training/conf/callbacks/wandb_ocr.yaml
+++ b/training/conf/callbacks/wandb_ocr.yaml
@@ -1,6 +1,6 @@
 defaults:
   - default
-  - wandb/watch
-  - wandb/code 
-  - wandb/checkpoints
-  - wandb/ocr_predictions
+  - wandb_watch
+  - wandb_code 
+  - wandb_checkpoints
+  - wandb_ocr_predictions
diff --git a/training/conf/callbacks/wandb/ocr_predictions.yaml b/training/conf/callbacks/wandb_ocr_predictions.yaml
index 573fa96..573fa96 100644
--- a/training/conf/callbacks/wandb/ocr_predictions.yaml
+++ b/training/conf/callbacks/wandb_ocr_predictions.yaml
diff --git a/training/conf/callbacks/wandb/watch.yaml b/training/conf/callbacks/wandb_watch.yaml
index 511608c..511608c 100644
--- a/training/conf/callbacks/wandb/watch.yaml
+++ b/training/conf/callbacks/wandb_watch.yaml
diff --git a/training/conf/config.yaml b/training/conf/config.yaml
index 93215ed..782bcbb 100644
--- a/training/conf/config.yaml
+++ b/training/conf/config.yaml
@@ -1,8 +1,9 @@
 defaults:
   - callbacks: wandb_ocr
   - criterion: label_smoothing
-  - dataset: iam_extended_paragraphs
+  - datamodule: iam_extended_paragraphs
   - hydra: default
+  - logger: wandb
   - lr_scheduler: one_cycle
   - mapping: word_piece
   - model: lit_transformer
@@ -15,3 +16,21 @@ tune: false
 train: true
 test: true
 logging: INFO
+
+# path to original working directory
+# hydra hijacks working directory by changing it to the current log directory,
+# so it's useful to have this path as a special variable
+# learn more here: https://hydra.cc/docs/next/tutorials/basic/running_your_app/working_directory
+work_dir: ${hydra:runtime.cwd}
+
+# use `python run.py debug=true` for easy debugging!
+# this will run 1 train, val and test loop with only 1 batch
+# equivalent to running `python run.py trainer.fast_dev_run=true`
+# (this is placed here just for easier access from command line)
+debug: False
+
+# pretty print config at the start of the run using Rich library
+print_config: True
+
+# disable python warnings if they annoy you
+ignore_warnings: True
diff --git a/training/conf/criterion/label_smoothing.yaml b/training/conf/criterion/label_smoothing.yaml
index 13daba8..684b5bb 100644
--- a/training/conf/criterion/label_smoothing.yaml
+++ b/training/conf/criterion/label_smoothing.yaml
@@ -1,4 +1,3 @@
-_target_: text_recognizer.criterion.label_smoothing.LabelSmoothingLoss
-label_smoothing: 0.1 
-vocab_size: 1006 
+_target_: text_recognizer.criterions.label_smoothing.LabelSmoothingLoss
+smoothing: 0.1 
 ignore_index: 1002
diff --git a/training/conf/datamodule/iam_extended_paragraphs.yaml b/training/conf/datamodule/iam_extended_paragraphs.yaml
index 3070b56..2d1a03e 100644
--- a/training/conf/datamodule/iam_extended_paragraphs.yaml
+++ b/training/conf/datamodule/iam_extended_paragraphs.yaml
@@ -1,5 +1,6 @@
 _target_: text_recognizer.data.iam_extended_paragraphs.IAMExtendedParagraphs
-batch_size: 32
+batch_size: 4
 num_workers: 12
 train_fraction: 0.8
 augment: true
+pin_memory: false
diff --git a/training/conf/lr_scheduler/one_cycle.yaml b/training/conf/lr_scheduler/one_cycle.yaml
index 5afdf81..eecee8a 100644
--- a/training/conf/lr_scheduler/one_cycle.yaml
+++ b/training/conf/lr_scheduler/one_cycle.yaml
@@ -1,8 +1,8 @@
 _target_: torch.optim.lr_scheduler.OneCycleLR
 max_lr: 1.0e-3
 total_steps: null
-epochs: null
-steps_per_epoch: null
+epochs: 512
+steps_per_epoch: 4992
 pct_start: 0.3
 anneal_strategy: cos
 cycle_momentum: true
diff --git a/training/conf/mapping/word_piece.yaml b/training/conf/mapping/word_piece.yaml
index 3792523..48384f5 100644
--- a/training/conf/mapping/word_piece.yaml
+++ b/training/conf/mapping/word_piece.yaml
@@ -1,4 +1,4 @@
-_target_: text_recognizer.data.mappings.WordPieceMapping
+_target_: text_recognizer.data.word_piece_mapping.WordPieceMapping
 num_features: 1000
 tokens: iamdb_1kwp_tokens_1000.txt
 lexicon: iamdb_1kwp_lex_1000.txt
@@ -6,4 +6,4 @@ data_dir: null
 use_words: false
 prepend_wordsep: false
 special_tokens: [ <s>, <e>, <p> ]
-extra_symbols: [ \n ]
+extra_symbols: [ "\n" ]
diff --git a/training/conf/model/lit_transformer.yaml b/training/conf/model/lit_transformer.yaml
index 6ffde4e..c190151 100644
--- a/training/conf/model/lit_transformer.yaml
+++ b/training/conf/model/lit_transformer.yaml
@@ -1,7 +1,7 @@
 _target_: text_recognizer.models.transformer.TransformerLitModel
 interval: step
 monitor: val/loss
-ignore_tokens: [ <s>, <e>, <p> ]
+max_output_len: 451
 start_token: <s>
 end_token: <e>
 pad_token: <p>
diff --git a/training/conf/network/conv_transformer.yaml b/training/conf/network/conv_transformer.yaml
index a97157d..f76e892 100644
--- a/training/conf/network/conv_transformer.yaml
+++ b/training/conf/network/conv_transformer.yaml
@@ -6,6 +6,5 @@ _target_: text_recognizer.networks.conv_transformer.ConvTransformer
 input_dims: [1, 576, 640]
 hidden_dim: 96
 dropout_rate: 0.2
-max_output_len: 451
 num_classes: 1006
 pad_index: 1002
diff --git a/training/conf/network/decoder/transformer_decoder.yaml b/training/conf/network/decoder/transformer_decoder.yaml
index 90b9d8a..eb80f64 100644
--- a/training/conf/network/decoder/transformer_decoder.yaml
+++ b/training/conf/network/decoder/transformer_decoder.yaml
@@ -18,3 +18,4 @@ ff_kwargs:
   dropout_rate: 0.2
 cross_attend: true
 pre_norm: true
+rotary_emb: null
diff --git a/training/run.py b/training/run.py
index 30479c6..13a6a82 100644
--- a/training/run.py
+++ b/training/run.py
@@ -12,35 +12,40 @@ from pytorch_lightning import (
     Trainer,
 )
 from pytorch_lightning.loggers import LightningLoggerBase
-from text_recognizer.data.mappings import AbstractMapping
 from torch import nn
+from text_recognizer.data.base_mapping import AbstractMapping
 
 import utils
 
 
 def run(config: DictConfig) -> Optional[float]:
     """Runs experiment."""
-    utils.configure_logging(config.logging)
+    utils.configure_logging(config)
     log.info("Starting experiment...")
 
     if config.get("seed"):
-        seed_everything(config.seed)
+        seed_everything(config.seed, workers=True)
 
     log.info(f"Instantiating mapping <{config.mapping._target_}>")
     mapping: AbstractMapping = hydra.utils.instantiate(config.mapping)
 
     log.info(f"Instantiating datamodule <{config.datamodule._target_}>")
-    datamodule: LightningDataModule = hydra.utils.instantiate(config.datamodule, mapping=mapping)
+    datamodule: LightningDataModule = hydra.utils.instantiate(
+        config.datamodule, mapping=mapping
+    )
 
     log.info(f"Instantiating network <{config.network._target_}>")
     network: nn.Module = hydra.utils.instantiate(config.network)
 
+    log.info(f"Instantiating criterion <{config.criterion._target_}>")
+    loss_fn: Type[nn.Module] = hydra.utils.instantiate(config.criterion)
+
     log.info(f"Instantiating model <{config.model._target_}>")
     model: LightningModule = hydra.utils.instantiate(
-        **config.model,
+        config.model,
         mapping=mapping,
         network=network,
-        criterion_config=config.criterion,
+        loss_fn=loss_fn,
         optimizer_config=config.optimizer,
         lr_scheduler_config=config.lr_scheduler,
         _recursive_=False,
@@ -77,4 +82,4 @@ def run(config: DictConfig) -> Optional[float]:
         trainer.test(model, datamodule=datamodule)
 
     log.info(f"Best checkpoint path:\n{trainer.checkpoint_callback.best_model_path}")
-    utils.finish(trainer)
+    utils.finish(logger)
diff --git a/training/utils.py b/training/utils.py
index ef74f61..d23396e 100644
--- a/training/utils.py
+++ b/training/utils.py
@@ -17,6 +17,10 @@ from tqdm import tqdm
 import wandb
 
 
+def print_config(config: DictConfig) -> None:
+    print(OmegaConf.to_yaml(config))
+
+
 @rank_zero_only
 def configure_logging(config: DictConfig) -> None:
     """Configure the loguru logger for output to terminal and disk."""
@@ -30,7 +34,7 @@ def configure_callbacks(config: DictConfig,) -> List[Type[Callback]]:
     callbacks = []
     if config.get("callbacks"):
         for callback_config in config.callbacks.values():
-            if config.get("_target_"):
+            if callback_config.get("_target_"):
                 log.info(f"Instantiating callback <{callback_config._target_}>")
                 callbacks.append(hydra.utils.instantiate(callback_config))
     return callbacks
@@ -41,8 +45,8 @@ def configure_logger(config: DictConfig) -> List[Type[LightningLoggerBase]]:
     logger = []
     if config.get("logger"):
         for logger_config in config.logger.values():
-            if config.get("_target_"):
-                log.info(f"Instantiating callback <{logger_config._target_}>")
+            if logger_config.get("_target_"):
+                log.info(f"Instantiating logger <{logger_config._target_}>")
                 logger.append(hydra.utils.instantiate(logger_config))
     return logger
 
@@ -67,17 +71,8 @@ def extras(config: DictConfig) -> None:
         # Debuggers do not like GPUs and multiprocessing.
         if config.trainer.get("gpus"):
             config.trainer.gpus = 0
-        if config.datamodule.get("pin_memory"):
-            config.datamodule.pin_memory = False
-        if config.datamodule.get("num_workers"):
-            config.datamodule.num_workers = 0
-
-    # Force multi-gpu friendly config.
-    accelerator = config.trainer.get("accelerator")
-    if accelerator in ["ddp", "ddp_spawn", "dp", "ddp2"]:
-        log.info(
-            f"Forcing ddp friendly configuration! <config.trainer.accelerator={accelerator}>"
-        )
+        if config.trainer.get("precision"):
+            config.trainer.precision = 32
         if config.datamodule.get("pin_memory"):
             config.datamodule.pin_memory = False
         if config.datamodule.get("num_workers"):