diff --git a/CNN.ipynb b/CNN.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..cc63fad3a2436e70a42b2d10a9924d88d2e1ee56
--- /dev/null
+++ b/CNN.ipynb
@@ -0,0 +1,2153 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "30b5b598",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import torch\n",
+    "import torch.nn as nn\n",
+    "import torch.nn.functional as F\n",
+    "import torchvision\n",
+    "import matplotlib.pyplot as plt\n",
+    "import numpy as np\n",
+    "import torch.optim as optim"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "ad273f17",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "class NeuralNetwork(nn.Module):\n",
+    "    def __init__(self):\n",
+    "        super().__init__()\n",
+    "        self.conv1 = nn.Conv2d(1, 6, 5)\n",
+    "        self.pool = nn.MaxPool2d(2, 2)\n",
+    "        self.conv2 = nn.Conv2d(6, 16, 5)\n",
+    "        self.fc1 = nn.LazyLinear(120)\n",
+    "        self.fc2 = nn.Linear(120, 84)\n",
+    "        self.fc3 = nn.Linear(84, 3)\n",
+    "    def forward(self, x):\n",
+    "        x = self.pool(F.relu(self.conv1(x)))\n",
+    "        x = self.pool(F.relu(self.conv2(x)))\n",
+    "        x = torch.flatten(x, 1) # flatten all dimensions except batch\n",
+    "        x = F.relu(self.fc1(x))\n",
+    "        x = F.relu(self.fc2(x))\n",
+    "        x = self.fc3(x)\n",
+    "        return x"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "6eb55c23",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/user/anaconda3/lib/python3.8/site-packages/torch/nn/modules/lazy.py:178: UserWarning: Lazy modules are a new feature under heavy development so changes to the API or functionality can happen at any moment.\n",
+      "  warnings.warn('Lazy modules are a new feature under heavy development '\n"
+     ]
+    }
+   ],
+   "source": [
+    "net = NeuralNetwork()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "1545454a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "if torch.cuda.is_available():\n",
+    "    net = net.cuda()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "1d846b72",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "criterion = nn.CrossEntropyLoss()\n",
+    "optimizer = optim.SGD(net.parameters(), lr=0.01, momentum=0.9)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "fd6eaa1b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from torchvision import datasets, transforms\n",
+    "from torch.utils.data import DataLoader, random_split"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "845c288e",
+   "metadata": {},
+   "source": [
+    "train_transforms = transforms.Compose([transforms.Grayscale(num_output_channels=1), transforms.Resize(512), transforms.CenterCrop(511), transforms.RandomRotation(30),transforms.RandomHorizontalFlip(), transforms.transforms.ToTensor()]) "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "57cae1a0",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def UploadData(path, train):\n",
+    "    #set up transforms for train and test datasets\n",
+    "    train_transforms = transforms.Compose([transforms.Grayscale(num_output_channels=1), transforms.Resize(512), transforms.CenterCrop(511), transforms.RandomRotation(30),transforms.RandomHorizontalFlip(), transforms.transforms.ToTensor()]) \n",
+    "    valid_transforms = transforms.Compose([transforms.Grayscale(num_output_channels=1), transforms.Resize(512), transforms.CenterCrop(511), transforms.transforms.ToTensor()]) \n",
+    "    #test_transforms = transforms.Compose([transforms.Grayscale(num_output_channels=1), transforms.Resize(512), transforms.CenterCrop(511), transforms.ToTensor()])\n",
+    "    \n",
+    "    #set up datasets from Image Folders\n",
+    "    train_dataset = datasets.ImageFolder(path + '/train', transform=train_transforms)\n",
+    "    valid_dataset = datasets.ImageFolder(path + '/validation', transform=valid_transforms)\n",
+    "    #test_dataset = datasets.ImageFolder(path + '/test', transform=test_transforms)\n",
+    "\n",
+    "    #set up dataloaders with batch size of 32\n",
+    "    trainloader = torch.utils.data.DataLoader(train_dataset, batch_size=32, shuffle=True)\n",
+    "    validloader = torch.utils.data.DataLoader(valid_dataset, batch_size=32, shuffle=True)\n",
+    "    #testloader = torch.utils.data.DataLoader(test_dataset, batch_size=32, shuffle=True)\n",
+    "  \n",
+    "    return trainloader, validloader #, testloader"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "65a32a3e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "trainloader, validloader = UploadData(\"/home/user/research/CXR_Covid-19_Challenge\", True) #, testloader"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "c530bc6c",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'covid': 0, 'normal': 1, 'pneumonia': 2}"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "trainloader.dataset.class_to_idx"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "c716ed31",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.image.AxesImage at 0x7fbe7a11ff10>"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "#Show Image in Train set\n",
+    "train_images, labels = next(iter(trainloader))\n",
+    "trainImg = train_images[0].numpy()\n",
+    "plt.imshow(trainImg[0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "edebd50c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import time\n",
+    "  \n",
+    "def convert(seconds):\n",
+    "    return time.strftime(\"%H:%M:%S\", time.gmtime(seconds))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "3550121f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from tqdm import tqdm"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "id": "5ecfe7ee",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "playsound is relying on another python subprocess. Please use `pip install pygobject` if you want playsound to run more efficiently.\n"
+     ]
+    }
+   ],
+   "source": [
+    "from playsound import playsound"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "a7feb171",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "min_valid_loss = np.inf"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "id": "c762f9cf",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "  0%|          | 0/562 [00:00<?, ?it/s]/home/user/anaconda3/lib/python3.8/site-packages/torch/nn/functional.py:718: UserWarning: Named tensors and all their associated APIs are an experimental feature and subject to change. Please do not use them for anything important until they are released as stable. (Triggered internally at  /pytorch/c10/core/TensorImpl.h:1156.)\n",
+      "  return torch.max_pool2d(input, kernel_size, stride, padding, dilation, ceil_mode)\n",
+      "100%|██████████| 562/562 [05:42<00:00,  1.64it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1 \tTraining Loss: 0.9029228049877276 \tValidation Loss: 0.8944440715842776 \t time: 00:06:31\n",
+      "Train Accuracy : 57.63447952270508 \tValidation Accuracy : 59.32400894165039\n",
+      "Validation Loss Decreased( inf ---> 96.59995973110199 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:53<00:00,  1.59it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 2 \tTraining Loss: 0.8856767352365514 \tValidation Loss: 0.9128290994299783 \t time: 00:06:48\n",
+      "Train Accuracy : 58.620113372802734 \tValidation Accuracy : 56.75990676879883\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:02<00:00,  1.55it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 3 \tTraining Loss: 0.8471003622346925 \tValidation Loss: 0.9120684911807379 \t time: 00:06:52\n",
+      "Train Accuracy : 61.18721389770508 \tValidation Accuracy : 57.808860778808594\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:56<00:00,  1.58it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 4 \tTraining Loss: 0.8119809344888158 \tValidation Loss: 0.8293078966714718 \t time: 00:06:45\n",
+      "Train Accuracy : 63.35337829589844 \tValidation Accuracy : 62.150352478027344\n",
+      "Validation Loss Decreased( 96.59995973110199 ---> 89.56525284051895 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:52<00:00,  1.60it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 5 \tTraining Loss: 0.7705369327628315 \tValidation Loss: 0.8124735769298341 \t time: 00:06:39\n",
+      "Train Accuracy : 66.14321899414062 \tValidation Accuracy : 63.05361557006836\n",
+      "Validation Loss Decreased( 89.56525284051895 ---> 87.74714630842209 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:04<00:00,  1.54it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 6 \tTraining Loss: 0.741771535623116 \tValidation Loss: 0.8236339531011052 \t time: 00:06:49\n",
+      "Train Accuracy : 68.24813079833984 \tValidation Accuracy : 64.16084289550781\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:51<00:00,  1.60it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 7 \tTraining Loss: 0.7327183050723263 \tValidation Loss: 0.7604639452916605 \t time: 00:06:43\n",
+      "Train Accuracy : 68.82726287841797 \tValidation Accuracy : 67.83216857910156\n",
+      "Validation Loss Decreased( 87.74714630842209 ---> 82.13010609149933 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:56<00:00,  1.58it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 8 \tTraining Loss: 0.7158604902198731 \tValidation Loss: 0.7314019556398745 \t time: 00:06:48\n",
+      "Train Accuracy : 69.6681137084961 \tValidation Accuracy : 69.17249298095703\n",
+      "Validation Loss Decreased( 82.13010609149933 ---> 78.99141120910645 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:00<00:00,  1.56it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 9 \tTraining Loss: 0.6923646768323043 \tValidation Loss: 0.7229070735198481 \t time: 00:06:45\n",
+      "Train Accuracy : 70.92103576660156 \tValidation Accuracy : 68.88111877441406\n",
+      "Validation Loss Decreased( 78.99141120910645 ---> 78.07396394014359 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:27<00:00,  1.72it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 10 \tTraining Loss: 0.6785714580707278 \tValidation Loss: 0.7380137079291873 \t time: 00:06:11\n",
+      "Train Accuracy : 71.2718505859375 \tValidation Accuracy : 70.39627075195312\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:33<00:00,  1.68it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 11 \tTraining Loss: 0.676262518105982 \tValidation Loss: 0.676862828709461 \t time: 00:06:17\n",
+      "Train Accuracy : 71.97905731201172 \tValidation Accuracy : 72.17366027832031\n",
+      "Validation Loss Decreased( 78.07396394014359 ---> 73.1011855006218 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:34<00:00,  1.68it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 12 \tTraining Loss: 0.6698277538039082 \tValidation Loss: 0.8797201442497747 \t time: 00:06:22\n",
+      "Train Accuracy : 72.0180435180664 \tValidation Accuracy : 60.75175094604492\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:48<00:00,  1.61it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 13 \tTraining Loss: 0.6627295674272279 \tValidation Loss: 0.7323361265438574 \t time: 00:06:38\n",
+      "Train Accuracy : 72.8477554321289 \tValidation Accuracy : 69.23077392578125\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:04<00:00,  1.54it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 14 \tTraining Loss: 0.6527160725347512 \tValidation Loss: 0.7401043899633266 \t time: 00:06:54\n",
+      "Train Accuracy : 73.28766632080078 \tValidation Accuracy : 69.66783142089844\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:45<00:00,  1.63it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 15 \tTraining Loss: 0.6403792584281799 \tValidation Loss: 0.6886368589820685 \t time: 00:06:36\n",
+      "Train Accuracy : 73.85565948486328 \tValidation Accuracy : 71.47435760498047\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:00<00:00,  1.56it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 16 \tTraining Loss: 0.6318943491769007 \tValidation Loss: 0.7204805187605046 \t time: 00:06:50\n",
+      "Train Accuracy : 74.20091247558594 \tValidation Accuracy : 70.54196166992188\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:03<00:00,  1.55it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 17 \tTraining Loss: 0.6226259863461464 \tValidation Loss: 0.7040614678903863 \t time: 00:06:53\n",
+      "Train Accuracy : 74.68537139892578 \tValidation Accuracy : 71.99883270263672\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:20<00:00,  1.48it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 18 \tTraining Loss: 0.6235986035383468 \tValidation Loss: 0.6673177524849221 \t time: 00:07:15\n",
+      "Train Accuracy : 74.70207977294922 \tValidation Accuracy : 72.7272720336914\n",
+      "Validation Loss Decreased( 73.1011855006218 ---> 72.07031726837158 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:12<00:00,  1.51it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 19 \tTraining Loss: 0.603867971520619 \tValidation Loss: 0.6556077698866526 \t time: 00:07:03\n",
+      "Train Accuracy : 75.72669219970703 \tValidation Accuracy : 73.42657470703125\n",
+      "Validation Loss Decreased( 72.07031726837158 ---> 70.80563914775848 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:02<00:00,  1.55it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 20 \tTraining Loss: 0.5996408582370052 \tValidation Loss: 0.6324033361894114 \t time: 00:06:51\n",
+      "Train Accuracy : 75.86033630371094 \tValidation Accuracy : 75.29137420654297\n",
+      "Validation Loss Decreased( 70.80563914775848 ---> 68.29956030845642 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:38<00:00,  1.41it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 21 \tTraining Loss: 0.5935023639921192 \tValidation Loss: 0.6485145138921561 \t time: 00:07:34\n",
+      "Train Accuracy : 76.26127624511719 \tValidation Accuracy : 73.31002807617188\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:00<00:00,  1.56it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 22 \tTraining Loss: 0.5775027493830254 \tValidation Loss: 0.6033395292858282 \t time: 00:06:52\n",
+      "Train Accuracy : 76.95177459716797 \tValidation Accuracy : 76.45687866210938\n",
+      "Validation Loss Decreased( 68.29956030845642 ---> 65.16066916286945 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:01<00:00,  1.55it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 23 \tTraining Loss: 0.5658614781032253 \tValidation Loss: 0.6150209567061177 \t time: 00:06:55\n",
+      "Train Accuracy : 77.76478576660156 \tValidation Accuracy : 75.81584930419922\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:12<00:00,  1.51it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 24 \tTraining Loss: 0.5587423649546939 \tValidation Loss: 0.6809103475124748 \t time: 00:06:59\n",
+      "Train Accuracy : 78.14344024658203 \tValidation Accuracy : 72.98950958251953\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:57<00:00,  1.57it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 25 \tTraining Loss: 0.5468475903745648 \tValidation Loss: 0.6453374088914307 \t time: 00:06:46\n",
+      "Train Accuracy : 78.58335876464844 \tValidation Accuracy : 74.09674072265625\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:42<00:00,  1.64it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 26 \tTraining Loss: 0.5547165346209265 \tValidation Loss: 0.6095044458353961 \t time: 00:06:27\n",
+      "Train Accuracy : 78.1768569946289 \tValidation Accuracy : 75.64102935791016\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:43<00:00,  1.63it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 27 \tTraining Loss: 0.5380605110812442 \tValidation Loss: 0.6276791799399588 \t time: 00:06:28\n",
+      "Train Accuracy : 79.0733871459961 \tValidation Accuracy : 75.14569091796875\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:50<00:00,  1.61it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 28 \tTraining Loss: 0.5361319128887934 \tValidation Loss: 0.6042463669070491 \t time: 00:06:41\n",
+      "Train Accuracy : 79.27942657470703 \tValidation Accuracy : 76.42774200439453\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:52<00:00,  1.59it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 29 \tTraining Loss: 0.5231574617883065 \tValidation Loss: 0.6972870423837945 \t time: 00:06:40\n",
+      "Train Accuracy : 79.44648742675781 \tValidation Accuracy : 71.32867431640625\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:50<00:00,  1.60it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 30 \tTraining Loss: 0.538012137139395 \tValidation Loss: 0.5839059344596333 \t time: 00:06:38\n",
+      "Train Accuracy : 78.90633392333984 \tValidation Accuracy : 77.41841888427734\n",
+      "Validation Loss Decreased( 65.16066916286945 ---> 63.061840921640396 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:46<00:00,  1.62it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 31 \tTraining Loss: 0.5200198032624781 \tValidation Loss: 0.6576835181978014 \t time: 00:06:31\n",
+      "Train Accuracy : 79.82514190673828 \tValidation Accuracy : 73.89277648925781\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:36<00:00,  1.67it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 32 \tTraining Loss: 0.5140829452469255 \tValidation Loss: 0.7197631487139949 \t time: 00:06:23\n",
+      "Train Accuracy : 79.9476547241211 \tValidation Accuracy : 70.97901916503906\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:51<00:00,  1.60it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 33 \tTraining Loss: 0.5107236489706617 \tValidation Loss: 0.6104853100798748 \t time: 00:06:35\n",
+      "Train Accuracy : 80.11470794677734 \tValidation Accuracy : 75.58275604248047\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:42<00:00,  1.64it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 34 \tTraining Loss: 0.4948225080489688 \tValidation Loss: 0.6035054706864886 \t time: 00:06:28\n",
+      "Train Accuracy : 81.0446548461914 \tValidation Accuracy : 74.6212158203125\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:42<00:00,  1.64it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 35 \tTraining Loss: 0.48714314199639386 \tValidation Loss: 0.5258092273164678 \t time: 00:06:34\n",
+      "Train Accuracy : 81.13375091552734 \tValidation Accuracy : 79.22494506835938\n",
+      "Validation Loss Decreased( 63.061840921640396 ---> 56.78739655017853 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:49<00:00,  1.61it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 36 \tTraining Loss: 0.48779024150607 \tValidation Loss: 0.594145884944333 \t time: 00:06:37\n",
+      "Train Accuracy : 81.06136322021484 \tValidation Accuracy : 76.36946868896484\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:52<00:00,  1.59it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 37 \tTraining Loss: 0.5009896781733028 \tValidation Loss: 0.5729684631029764 \t time: 00:06:37\n",
+      "Train Accuracy : 80.40984344482422 \tValidation Accuracy : 76.95221710205078\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:53<00:00,  1.59it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 38 \tTraining Loss: 0.4772794395685196 \tValidation Loss: 0.5359699957900577 \t time: 00:06:39\n",
+      "Train Accuracy : 81.89664459228516 \tValidation Accuracy : 78.8170166015625\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:10<00:00,  1.52it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 39 \tTraining Loss: 0.48806979586751437 \tValidation Loss: 0.5759746941427389 \t time: 00:07:01\n",
+      "Train Accuracy : 81.21171569824219 \tValidation Accuracy : 77.36013793945312\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:56<00:00,  1.58it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 40 \tTraining Loss: 0.4783191208154281 \tValidation Loss: 0.5715127614913164 \t time: 00:06:42\n",
+      "Train Accuracy : 81.46229553222656 \tValidation Accuracy : 77.53496551513672\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:03<00:00,  1.55it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 41 \tTraining Loss: 0.4732402967308976 \tValidation Loss: 0.5820767724955523 \t time: 00:06:55\n",
+      "Train Accuracy : 81.56253051757812 \tValidation Accuracy : 76.63170623779297\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:06<00:00,  1.53it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 42 \tTraining Loss: 0.4650362822694498 \tValidation Loss: 0.6060397622210009 \t time: 00:06:59\n",
+      "Train Accuracy : 82.19178009033203 \tValidation Accuracy : 75.08741760253906\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:57<00:00,  1.57it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 43 \tTraining Loss: 0.4605097197977249 \tValidation Loss: 0.5849179564112866 \t time: 00:06:47\n",
+      "Train Accuracy : 82.54259490966797 \tValidation Accuracy : 77.47669219970703\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:06<00:00,  1.53it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 44 \tTraining Loss: 0.45593101589599117 \tValidation Loss: 0.5505736610955663 \t time: 00:06:58\n",
+      "Train Accuracy : 82.46463775634766 \tValidation Accuracy : 79.77855682373047\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:58<00:00,  1.57it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 45 \tTraining Loss: 0.4556735833424054 \tValidation Loss: 0.6444224847687615 \t time: 00:06:49\n",
+      "Train Accuracy : 82.25859832763672 \tValidation Accuracy : 75.72843933105469\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:23<00:00,  1.47it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 46 \tTraining Loss: 0.4516571824873045 \tValidation Loss: 0.594107069351055 \t time: 00:07:16\n",
+      "Train Accuracy : 82.7430648803711 \tValidation Accuracy : 77.01049041748047\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:01<00:00,  1.56it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 47 \tTraining Loss: 0.44326862740559086 \tValidation Loss: 0.5104974389628127 \t time: 00:06:55\n",
+      "Train Accuracy : 82.88784790039062 \tValidation Accuracy : 79.05011749267578\n",
+      "Validation Loss Decreased( 56.78739655017853 ---> 55.13372340798378 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:13<00:00,  1.51it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 48 \tTraining Loss: 0.44160879766601685 \tValidation Loss: 0.5294558451407485 \t time: 00:07:05\n",
+      "Train Accuracy : 83.23309326171875 \tValidation Accuracy : 80.12820434570312\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:58<00:00,  1.57it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 49 \tTraining Loss: 0.42966303054336125 \tValidation Loss: 0.5046062285977381 \t time: 00:06:45\n",
+      "Train Accuracy : 83.60618591308594 \tValidation Accuracy : 80.24475860595703\n",
+      "Validation Loss Decreased( 55.13372340798378 ---> 54.49747268855572 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:50<00:00,  1.60it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 50 \tTraining Loss: 0.4365461183135196 \tValidation Loss: 0.49573717942392387 \t time: 00:06:46\n",
+      "Train Accuracy : 83.43356323242188 \tValidation Accuracy : 80.88578033447266\n",
+      "Validation Loss Decreased( 54.49747268855572 ---> 53.539615377783775 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:00<00:00,  1.56it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 51 \tTraining Loss: 0.4393774249799735 \tValidation Loss: 0.542883567236088 \t time: 00:06:47\n",
+      "Train Accuracy : 83.1941146850586 \tValidation Accuracy : 78.84615325927734\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:09<00:00,  1.52it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 52 \tTraining Loss: 0.4241394173764885 \tValidation Loss: 0.52265090384969 \t time: 00:07:04\n",
+      "Train Accuracy : 83.80108642578125 \tValidation Accuracy : 79.13752746582031\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:27<00:00,  1.45it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 53 \tTraining Loss: 0.41579384702088784 \tValidation Loss: 0.5224103926232567 \t time: 00:07:17\n",
+      "Train Accuracy : 84.31896209716797 \tValidation Accuracy : 80.7109603881836\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:00<00:00,  1.56it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 54 \tTraining Loss: 0.4219729055607446 \tValidation Loss: 0.5424923496665778 \t time: 00:06:52\n",
+      "Train Accuracy : 83.9570083618164 \tValidation Accuracy : 79.16667175292969\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:59<00:00,  1.56it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 55 \tTraining Loss: 0.4344543345484657 \tValidation Loss: 0.5544744128430331 \t time: 00:06:51\n",
+      "Train Accuracy : 83.56163787841797 \tValidation Accuracy : 79.4871826171875\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:52<00:00,  1.59it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 56 \tTraining Loss: 0.42892047239208986 \tValidation Loss: 0.6630421954172628 \t time: 00:06:40\n",
+      "Train Accuracy : 83.89018249511719 \tValidation Accuracy : 75.75757598876953\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:04<00:00,  1.54it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 57 \tTraining Loss: 0.4104134016189711 \tValidation Loss: 0.5150271084297586 \t time: 00:06:54\n",
+      "Train Accuracy : 84.22429656982422 \tValidation Accuracy : 79.63286590576172\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:54<00:00,  1.59it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 58 \tTraining Loss: 0.3950271703488462 \tValidation Loss: 0.5091825410447739 \t time: 00:06:41\n",
+      "Train Accuracy : 85.00389862060547 \tValidation Accuracy : 80.5361328125\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:59<00:00,  1.56it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 59 \tTraining Loss: 0.4049163169469486 \tValidation Loss: 0.5659814819141671 \t time: 00:06:49\n",
+      "Train Accuracy : 84.37464904785156 \tValidation Accuracy : 78.11771392822266\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:06<00:00,  1.53it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 60 \tTraining Loss: 0.4085990673467994 \tValidation Loss: 0.6131980065946225 \t time: 00:06:57\n",
+      "Train Accuracy : 84.45260620117188 \tValidation Accuracy : 77.91375732421875\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:12<00:00,  1.51it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 61 \tTraining Loss: 0.4266822505543453 \tValidation Loss: 0.5203534404712694 \t time: 00:06:58\n",
+      "Train Accuracy : 83.90689086914062 \tValidation Accuracy : 80.97319793701172\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:55<00:00,  1.58it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 62 \tTraining Loss: 0.4057566263842201 \tValidation Loss: 0.5222419394111192 \t time: 00:06:46\n",
+      "Train Accuracy : 84.88695526123047 \tValidation Accuracy : 79.42890930175781\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:48<00:00,  1.61it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 63 \tTraining Loss: 0.3956257048000008 \tValidation Loss: 0.5357174716751885 \t time: 00:06:39\n",
+      "Train Accuracy : 85.03173828125 \tValidation Accuracy : 79.72028350830078\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:03<00:00,  1.54it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 64 \tTraining Loss: 0.3930586477827771 \tValidation Loss: 0.5213135158022245 \t time: 00:06:54\n",
+      "Train Accuracy : 85.09856414794922 \tValidation Accuracy : 79.72028350830078\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:57<00:00,  1.57it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 65 \tTraining Loss: 0.38868296031717514 \tValidation Loss: 0.5845458168122504 \t time: 00:06:48\n",
+      "Train Accuracy : 85.43824005126953 \tValidation Accuracy : 77.36013793945312\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:00<00:00,  1.56it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 66 \tTraining Loss: 0.3743342116103902 \tValidation Loss: 0.47417158150562533 \t time: 00:06:50\n",
+      "Train Accuracy : 86.14544677734375 \tValidation Accuracy : 82.2552490234375\n",
+      "Validation Loss Decreased( 53.539615377783775 ---> 51.210530802607536 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:04<00:00,  1.54it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 67 \tTraining Loss: 0.3814578014309932 \tValidation Loss: 0.6384360061751472 \t time: 00:06:56\n",
+      "Train Accuracy : 85.88929748535156 \tValidation Accuracy : 77.85547637939453\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:03<00:00,  1.55it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 68 \tTraining Loss: 0.3842689093348182 \tValidation Loss: 0.48013403570210494 \t time: 00:06:49\n",
+      "Train Accuracy : 85.60530090332031 \tValidation Accuracy : 82.51748657226562\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:56<00:00,  1.58it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 69 \tTraining Loss: 0.37951515866874375 \tValidation Loss: 0.5551777308185896 \t time: 00:06:48\n",
+      "Train Accuracy : 85.79463195800781 \tValidation Accuracy : 78.52564239501953\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:04<00:00,  1.54it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 70 \tTraining Loss: 0.37778037140855164 \tValidation Loss: 0.6369037684743051 \t time: 00:06:53\n",
+      "Train Accuracy : 85.70553588867188 \tValidation Accuracy : 76.8939437866211\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:03<00:00,  1.55it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 71 \tTraining Loss: 0.3850887486929889 \tValidation Loss: 0.5161193176000206 \t time: 00:06:50\n",
+      "Train Accuracy : 85.5997314453125 \tValidation Accuracy : 80.76923370361328\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:02<00:00,  1.55it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 72 \tTraining Loss: 0.3706632720379431 \tValidation Loss: 0.4723347807648005 \t time: 00:06:51\n",
+      "Train Accuracy : 86.20670318603516 \tValidation Accuracy : 82.98368835449219\n",
+      "Validation Loss Decreased( 51.210530802607536 ---> 51.01215632259846 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:02<00:00,  1.55it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 73 \tTraining Loss: 0.363397472170165 \tValidation Loss: 0.5204819073831594 \t time: 00:06:58\n",
+      "Train Accuracy : 86.26239013671875 \tValidation Accuracy : 79.45804595947266\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:58<00:00,  1.57it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 74 \tTraining Loss: 0.35750560526109676 \tValidation Loss: 0.4980683869647759 \t time: 00:06:51\n",
+      "Train Accuracy : 86.69673156738281 \tValidation Accuracy : 81.81818389892578\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:03<00:00,  1.55it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 75 \tTraining Loss: 0.35306707385171776 \tValidation Loss: 0.5271965176970871 \t time: 00:06:49\n",
+      "Train Accuracy : 86.94731903076172 \tValidation Accuracy : 80.59440612792969\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:00<00:00,  1.56it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 76 \tTraining Loss: 0.35285318319377523 \tValidation Loss: 0.46226683879892033 \t time: 00:06:52\n",
+      "Train Accuracy : 86.84151458740234 \tValidation Accuracy : 83.24592590332031\n",
+      "Validation Loss Decreased( 51.01215632259846 ---> 49.924818590283394 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:59<00:00,  1.56it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 77 \tTraining Loss: 0.3440603395771514 \tValidation Loss: 0.4683791854315334 \t time: 00:06:50\n",
+      "Train Accuracy : 87.11994171142578 \tValidation Accuracy : 82.7505874633789\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [05:57<00:00,  1.57it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 78 \tTraining Loss: 0.33785627478753544 \tValidation Loss: 0.5578625021433389 \t time: 00:06:46\n",
+      "Train Accuracy : 87.38723754882812 \tValidation Accuracy : 77.79720306396484\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:09<00:00,  1.52it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 79 \tTraining Loss: 0.34567932541286606 \tValidation Loss: 0.48196811063422096 \t time: 00:06:58\n",
+      "Train Accuracy : 86.9417495727539 \tValidation Accuracy : 81.06060791015625\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [08:04<00:00,  1.16it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 80 \tTraining Loss: 0.34494500493239677 \tValidation Loss: 0.4852829767322099 \t time: 00:08:57\n",
+      "Train Accuracy : 87.29813385009766 \tValidation Accuracy : 80.27389526367188\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:34<00:00,  1.43it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 81 \tTraining Loss: 0.3418734817835254 \tValidation Loss: 0.4769098451016126 \t time: 00:07:31\n",
+      "Train Accuracy : 87.08096313476562 \tValidation Accuracy : 82.19696807861328\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:20<00:00,  1.48it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 82 \tTraining Loss: 0.3200561340725931 \tValidation Loss: 0.4822914483094657 \t time: 00:07:20\n",
+      "Train Accuracy : 88.06102752685547 \tValidation Accuracy : 82.89627075195312\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:07<00:00,  1.53it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 83 \tTraining Loss: 0.3378799657247881 \tValidation Loss: 0.496175997649078 \t time: 00:06:56\n",
+      "Train Accuracy : 87.55429077148438 \tValidation Accuracy : 81.87645721435547\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:18<00:00,  1.49it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 84 \tTraining Loss: 0.32990650342797256 \tValidation Loss: 0.4860944132562037 \t time: 00:07:12\n",
+      "Train Accuracy : 87.9552230834961 \tValidation Accuracy : 82.37179565429688\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:06<00:00,  1.53it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 85 \tTraining Loss: 0.33767457954002233 \tValidation Loss: 0.5501075481513032 \t time: 00:06:56\n",
+      "Train Accuracy : 87.74362182617188 \tValidation Accuracy : 80.3613052368164\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:23<00:00,  1.47it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 86 \tTraining Loss: 0.315956815653 \tValidation Loss: 0.43915262694160145 \t time: 00:07:13\n",
+      "Train Accuracy : 88.2670669555664 \tValidation Accuracy : 82.634033203125\n",
+      "Validation Loss Decreased( 49.924818590283394 ---> 47.428483709692955 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:12<00:00,  1.51it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 87 \tTraining Loss: 0.3222351762999737 \tValidation Loss: 0.5052717746821819 \t time: 00:07:00\n",
+      "Train Accuracy : 88.03875732421875 \tValidation Accuracy : 82.43006896972656\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:05<00:00,  1.54it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 88 \tTraining Loss: 0.32136212186589574 \tValidation Loss: 0.48401865404513145 \t time: 00:06:59\n",
+      "Train Accuracy : 88.0777359008789 \tValidation Accuracy : 83.47901916503906\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:29<00:00,  1.44it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 89 \tTraining Loss: 0.3022051092098033 \tValidation Loss: 0.46346047189500594 \t time: 00:07:19\n",
+      "Train Accuracy : 88.84062194824219 \tValidation Accuracy : 83.88694763183594\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:13<00:00,  1.50it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 90 \tTraining Loss: 0.30220048422544027 \tValidation Loss: 0.5105993631123392 \t time: 00:07:06\n",
+      "Train Accuracy : 88.97427368164062 \tValidation Accuracy : 83.24592590332031\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:10<00:00,  1.52it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 91 \tTraining Loss: 0.3055765822171847 \tValidation Loss: 0.5244620022950349 \t time: 00:07:02\n",
+      "Train Accuracy : 88.66799926757812 \tValidation Accuracy : 82.95454406738281\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:10<00:00,  1.52it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 92 \tTraining Loss: 0.30953261457065456 \tValidation Loss: 1.0839747182197041 \t time: 00:06:56\n",
+      "Train Accuracy : 88.37286376953125 \tValidation Accuracy : 74.56294250488281\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:13<00:00,  1.50it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 93 \tTraining Loss: 0.3242695900840267 \tValidation Loss: 0.45126490432907035 \t time: 00:07:07\n",
+      "Train Accuracy : 87.93852233886719 \tValidation Accuracy : 83.07109832763672\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:21<00:00,  1.47it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 94 \tTraining Loss: 0.30310345478886175 \tValidation Loss: 0.47017240883023653 \t time: 00:07:16\n",
+      "Train Accuracy : 88.6401596069336 \tValidation Accuracy : 82.634033203125\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:07<00:00,  1.53it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 95 \tTraining Loss: 0.29942518796148676 \tValidation Loss: 0.4341367942591508 \t time: 00:06:59\n",
+      "Train Accuracy : 88.92972564697266 \tValidation Accuracy : 84.38228607177734\n",
+      "Validation Loss Decreased( 47.428483709692955 ---> 46.88677377998829 ) \t Saving The Model\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:08<00:00,  1.52it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 96 \tTraining Loss: 0.30434063478667966 \tValidation Loss: 0.46439169802599484 \t time: 00:07:05\n",
+      "Train Accuracy : 88.56776428222656 \tValidation Accuracy : 80.85664367675781\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:24<00:00,  1.46it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 97 \tTraining Loss: 0.3168919574058353 \tValidation Loss: 0.6083490194545852 \t time: 00:07:21\n",
+      "Train Accuracy : 88.36730194091797 \tValidation Accuracy : 78.05944061279297\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:10<00:00,  1.52it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 98 \tTraining Loss: 0.29728707446454894 \tValidation Loss: 0.4969119317829609 \t time: 00:06:58\n",
+      "Train Accuracy : 89.08007049560547 \tValidation Accuracy : 82.83799743652344\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:22<00:00,  1.47it/s]\n",
+      "  0%|          | 0/562 [00:00<?, ?it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 99 \tTraining Loss: 0.28510440849480334 \tValidation Loss: 0.5444494509310635 \t time: 00:07:10\n",
+      "Train Accuracy : 89.38077545166016 \tValidation Accuracy : 81.00233459472656\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 562/562 [06:07<00:00,  1.53it/s]\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 100 \tTraining Loss: 0.28271967525431535 \tValidation Loss: 0.47462088207679765 \t time: 00:06:56\n",
+      "Train Accuracy : 89.64249420166016 \tValidation Accuracy : 83.42074584960938\n",
+      "total time :  11:29:30\n"
+     ]
+    }
+   ],
+   "source": [
+    "loss_train_list = []\n",
+    "loss_valid_list = []\n",
+    "acc_train_list = []\n",
+    "acc_valid_list = []\n",
+    "epochs = 100\n",
+    "total_time = time.time()\n",
+    "for e in range(epochs):\n",
+    "    start_time=time.time()\n",
+    "    train_loss = 0.0\n",
+    "    right_train = 0\n",
+    "    total_train = 0\n",
+    "    for data, labels in tqdm(trainloader):\n",
+    "        # Transfer Data to GPU if available\n",
+    "        if torch.cuda.is_available():\n",
+    "            data, labels = data.cuda(), labels.cuda()\n",
+    "         \n",
+    "        # Clear the gradients\n",
+    "        optimizer.zero_grad()\n",
+    "        # Forward Pass\n",
+    "        target = net(data)\n",
+    "        _, predicted = torch.max(target, 1)\n",
+    "        # Find the Loss\n",
+    "        loss = criterion(target,labels)\n",
+    "        # Calculate gradients\n",
+    "        loss.backward()\n",
+    "        # Update Weights\n",
+    "        optimizer.step()\n",
+    "        # Calculate Loss\n",
+    "        train_loss += loss.item()\n",
+    "        correct = (predicted == labels).float().sum()\n",
+    "        right_train+=correct.float()\n",
+    "        total_train+=len(predicted)\n",
+    "     \n",
+    "    valid_loss = 0.0\n",
+    "    right_valid = 0\n",
+    "    total_valid = 0\n",
+    "    net.eval()     # Optional when not using Model Specific layer\n",
+    "    for data, labels in (validloader):\n",
+    "        # Transfer Data to GPU if available\n",
+    "        if torch.cuda.is_available():\n",
+    "            data, labels = data.cuda(), labels.cuda()\n",
+    "         \n",
+    "        # Forward Pass\n",
+    "        target = net(data)\n",
+    "        _, predicted = torch.max(target, 1)\n",
+    "        # Find the Loss\n",
+    "        loss = criterion(target,labels)\n",
+    "        # Calculate Loss\n",
+    "        valid_loss += loss.item()\n",
+    "        correct = (predicted == labels).float().sum()\n",
+    "        right_valid+=correct.float()\n",
+    "        total_valid+=len(predicted)\n",
+    "    ftloss = train_loss / len(trainloader)\n",
+    "    fvloss = valid_loss / len(validloader)\n",
+    "    ftacc = float(right_train*100/total_train)\n",
+    "    fvacc = float(right_valid*100/total_valid)\n",
+    "    loss_train_list.append(ftloss)\n",
+    "    loss_valid_list.append(fvloss)\n",
+    "    acc_train_list.append(ftacc)\n",
+    "    acc_valid_list.append(fvacc)\n",
+    "    print('Epoch',e+1, '\\tTraining Loss:',ftloss,'\\tValidation Loss:',fvloss,\"\\t time:\",convert(time.time()-start_time))\n",
+    "    print(\"Train Accuracy :\",ftacc,\"\\tValidation Accuracy :\",fvacc)\n",
+    "    if min_valid_loss > valid_loss:\n",
+    "        print(\"Validation Loss Decreased(\",min_valid_loss,\"--->\",valid_loss,\") \\t Saving The Model\")\n",
+    "        min_valid_loss = valid_loss\n",
+    "         \n",
+    "        # Saving State Dict\n",
+    "        torch.save(net.state_dict(), '/home/user/research/CNN/cnn_model_new.pth')\n",
+    "print(\"total time : \",convert(time.time()-total_time))\n",
+    "playsound('/home/user/research/audio')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "id": "1bb67e73",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.plot(loss_train_list)\n",
+    "plt.plot(loss_valid_list)\n",
+    "plt.annotate(\"min loss valid\",(95,min(loss_valid_list)))\n",
+    "plt.title('Training and Validation Loss during Model Training')\n",
+    "plt.ylabel('loss')\n",
+    "plt.xlabel('epoch')\n",
+    "plt.legend(['train', 'valid','minimum'], loc='upper left')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "id": "58eae872",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plt.plot(acc_train_list)\n",
+    "plt.plot(acc_valid_list)\n",
+    "plt.annotate(\"max accuracy valid\",(95,max(acc_valid_list)))\n",
+    "plt.title('Training and Validation Loss during Model Training')\n",
+    "plt.ylabel('loss')\n",
+    "plt.xlabel('epoch')\n",
+    "plt.legend(['train', 'valid'], loc='upper left')\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "id": "5534455f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "f1 = open(\"/home/user/research/CNN/loss_train.txt\",\"w\")\n",
+    "f2 = open(\"/home/user/research/CNN/loss_valid.txt\",\"w\")\n",
+    "f3 = open(\"/home/user/research/CNN/acc_train.txt\",\"w\")\n",
+    "f4 = open(\"/home/user/research/CNN/acc_valid.txt\",\"w\")\n",
+    "for i in range(len(loss_train_list)):\n",
+    "    f1.write(str(loss_train_list[i]))\n",
+    "    f1.write(\",\")\n",
+    "    f2.write(str(loss_valid_list[i]))\n",
+    "    f2.write(\",\")\n",
+    "    f3.write(str(acc_train_list[i]))\n",
+    "    f3.write(\",\")\n",
+    "    f4.write(str(acc_valid_list[i]))\n",
+    "    f4.write(\",\")\n",
+    "f1.close()\n",
+    "f2.close()\n",
+    "f3.close()\n",
+    "f4.close()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "id": "e99936d8",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "using GPU\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/user/anaconda3/lib/python3.8/site-packages/torch/nn/modules/lazy.py:178: UserWarning: Lazy modules are a new feature under heavy development so changes to the API or functionality can happen at any moment.\n",
+      "  warnings.warn('Lazy modules are a new feature under heavy development '\n"
+     ]
+    }
+   ],
+   "source": [
+    "net = NeuralNetwork()\n",
+    "if torch.cuda.is_available():\n",
+    "    print(\"using GPU\")\n",
+    "    net = net.cuda()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "id": "2d86405c",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<All keys matched successfully>"
+      ]
+     },
+     "execution_count": 26,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "net.load_state_dict(torch.load(\"/home/user/research/CNN/cnn_model_new.pth\"))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "id": "f44a6760",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "100%|██████████| 108/108 [00:31<00:00,  3.38it/s]"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "84.38228607177734\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "right = 0\n",
+    "total = 0\n",
+    "for data, labels in tqdm(validloader):\n",
+    "    if torch.cuda.is_available():\n",
+    "            data, labels = data.cuda(), labels.cuda()\n",
+    "    outputs = net(data)\n",
+    "    _, predicted = torch.max(outputs, 1)\n",
+    "    correct = (predicted == labels).float().sum()\n",
+    "    right+=correct.float()\n",
+    "    total = total+len(predicted)\n",
+    "    #print(correct*100/len(predicted))\n",
+    "    #pred = predicted.tolist()\n",
+    "    #correct = labels.tolist()\n",
+    "#     for i in range(len(pred)):\n",
+    "#         if(pred[i]==correct[i]):\n",
+    "#             right+=1\n",
+    "#         else:\n",
+    "#             wrong+=1\n",
+    "print(float(right*100/total))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "d1112aed",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from torchviz import make_dot\n",
+    "train_images, labels = next(iter(trainloader))\n",
+    "y = net(train_images.cuda())\n",
+    "\n",
+    "make_dot(y.mean(), params=dict(net.named_parameters()), show_attrs=True, show_saved=True).render(\"attached\", format=\"png\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.8"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/CNN_old.ipynb b/CNN_old.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..5d3de74fe5bf10f75c765154447a28bbe972ebc9
--- /dev/null
+++ b/CNN_old.ipynb
@@ -0,0 +1,615 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "57ee40ee",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import torch.nn as nn\n",
+    "import torch.nn.functional as F\n",
+    "import torchvision\n",
+    "import matplotlib.pyplot as plt\n",
+    "import numpy as np"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "a6837b5b",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import torch.optim as optim"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "b67b6732",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "class NeuralNetwork(nn.Module):\n",
+    "    def __init__(self):\n",
+    "        super().__init__()\n",
+    "        self.conv1 = nn.Conv2d(1, 6, 5)\n",
+    "        self.pool = nn.MaxPool2d(2, 2)\n",
+    "        self.conv2 = nn.Conv2d(6, 16, 5)\n",
+    "        self.fc1 = nn.LazyLinear(120) \n",
+    "        self.fc2 = nn.Linear(120, 84)\n",
+    "        self.fc3 = nn.Linear(84, 3)\n",
+    "    def forward(self, x):\n",
+    "        x = self.pool(F.relu(self.conv1(x)))\n",
+    "        x = self.pool(F.relu(self.conv2(x)))\n",
+    "        x = torch.flatten(x, 1) \n",
+    "        x = F.relu(self.fc1(x))\n",
+    "        x = F.relu(self.fc2(x))\n",
+    "        x = self.fc3(x)\n",
+    "        return x"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "950fcadb",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/lns/anaconda3/lib/python3.8/site-packages/torch/nn/modules/lazy.py:178: UserWarning: Lazy modules are a new feature under heavy development so changes to the API or functionality can happen at any moment.\n",
+      "  warnings.warn('Lazy modules are a new feature under heavy development '\n"
+     ]
+    }
+   ],
+   "source": [
+    "net = NeuralNetwork()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "id": "0d818e23",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "criterion = nn.CrossEntropyLoss()\n",
+    "optimizer = optim.SGD(net.parameters(), lr=0.01, momentum=0.9)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "31cd8b26",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import torch"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "57df6949",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "using GPU\n"
+     ]
+    }
+   ],
+   "source": [
+    "if torch.cuda.is_available():\n",
+    "    print(\"using GPU\")\n",
+    "    net = net.cuda()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "5c2af0ec",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from torchvision import datasets, transforms\n",
+    "from torch.utils.data import DataLoader, random_split"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "e26f46d9",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def UploadData(path, train):\n",
+    "    #set up transforms for train and test datasets\n",
+    "    train_transforms = transforms.Compose([transforms.Grayscale(num_output_channels=1), transforms.Resize(512), transforms.CenterCrop(511), transforms.RandomRotation(30), \n",
+    "                                         transforms.RandomHorizontalFlip(), transforms.transforms.ToTensor()]) \n",
+    "    valid_transforms = transforms.Compose([transforms.Grayscale(num_output_channels=1), transforms.Resize(512), transforms.CenterCrop(511), transforms.RandomRotation(30), \n",
+    "                                         transforms.RandomHorizontalFlip(), transforms.transforms.ToTensor()]) \n",
+    "    test_transforms = transforms.Compose([transforms.Grayscale(num_output_channels=1), transforms.Resize(512), transforms.CenterCrop(511), transforms.ToTensor()])\n",
+    "    \n",
+    "    #set up datasets from Image Folders\n",
+    "    train_dataset = datasets.ImageFolder(path + '/train', transform=train_transforms)\n",
+    "    valid_dataset = datasets.ImageFolder(path + '/validation', transform=valid_transforms)\n",
+    "    test_dataset = datasets.ImageFolder(path + '/test', transform=test_transforms)\n",
+    "\n",
+    "    #set up dataloaders with batch size of 32\n",
+    "    trainloader = torch.utils.data.DataLoader(train_dataset, batch_size=32, shuffle=True)\n",
+    "    validloader = torch.utils.data.DataLoader(valid_dataset, batch_size=32, shuffle=True)\n",
+    "    testloader = torch.utils.data.DataLoader(test_dataset, batch_size=32, shuffle=True)\n",
+    "  \n",
+    "    return trainloader, validloader, testloader\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "66e7ea96",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "trainloader, validloader, testloader = UploadData(\"/home/lns/research/dataset\", True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "5c9e6023",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.image.AxesImage at 0x7f307d53a550>"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQcAAAD8CAYAAAB6iWHJAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAEAAElEQVR4nOz9abBkW3bfh/3WHs45mXnvrXr1pn79uhuNbgCNGYIAgaJAWiQ4iKIogiFREsCQLIUUgbBFTxH+IMoRDn1iWP7CDw7bciDClKiwaZqhwZQjZNEmKYqWTAmcBQIgMTV6fHNV3Skzzzl78Ie19z4nb9XrbnT36y4AtSPeq6q8efPmPbnP2mv91///X5Jz5vl6vp6v5+vuMt/sN/B8PV/P17O5ngeH5+v5er6eup4Hh+fr+Xq+nrqeB4fn6/l6vp66ngeH5+v5er6eup4Hh+fr+Xq+nro+sOAgIn9ARP6hiPyyiPyJD+rnPF/P1/P1wSz5IHgOImKBXwR+H/B54G8AP5lz/vmv+w97vp6v5+sDWR9U5vAjwC/nnH815zwBfw748Q/oZz1fz9fz9QEs9wG97uvA51b//jzw297vyZ30eWD3Ab2V3zpLjCEPPeMLhouLPQA5CwkhZSEj5AwpCwLk8vX1qnmkrP4tgEjWx06fTk08jWScSYC+fn3MkNufnQlYEqm8SEYwZEQyKQshW6Zs9XvL9wnL9zpS+S59H0ImI0h51/WtxfK7sjy7vU5CiPpTqc+qr1N/3/Vrpmza+33yOgnrvDtl4frzZ8jV/mkfzzd9XfPo3Zzzy1/p8z+o4PDk1eTkOiIiPwX8FMDAlt8mv+cDeiu/+ZfZbrn6Q9/Po3/xhh/7ll/CSGqbOmbhED1TchyDZ0qWMTpSFmIyTNG2m9mWm7uuGjhEMt4krEkYWT7G+n05a/A560bO/Ni+du5HNnbGkPlw/5gPd48wJPap1/ctmZ0ZOSbPu+GcN6Z7HKLHi74PZ2L7WS/5G17zj/ES6CQymBlDYs4OK4mYDYOZeNles089x+yZsuWYfXuNnRmx5PZ4yqa9TsIQs2HOllgSai+Bfeq5jhti2dK2bOOIkLJpf9Zr/n/6G/8kn/of/yx5HL9On+7Xb/2l/B995tfz/A8qOHwe+Ojq3x8Bvrh+Qs75p4GfBriQB88FHl/Fch96lcf/g2/l8ieu+ec/+f8lZsNN7MtXE6mcwr2JWNHTd0qOkAxTchyC3jhzMi0QmHKK5yxICQR+FRTWwWH9fMnCGHU7DXZmsIGdnXAmsjUTD9xNu5nr93USiFk4Zs/DsCMkS2/CEz8jZeHRvKU3M/fsgWOG29RjJWFIWMl4CVhyu1G3MuKzhQRzdszZ6veQsXIaBJ+2YhYsBi8RLwGyawGiBgQowWL1ev/cD/xt/u7v+EHcX/5bX9mH+AyvDyo4/A3g20XkW4EvAD8B/LEP6Gf9llvuEx/nV//lD/M9v+8X+d6z/5bBzMRcT7vInG3bvABGEg5wAl4SyQoxT1x4wzE6jtFzO3fMUc9FI5mYpaT1enNak1rZAJr2I/qnNXpypyzMSQPSC92BjZ0wknnJ3zCYmYSezMtraLC4DFvG5Nr7fyI7Ef1dHs07vETOzRGAKWsGZCQziDDIzJvxHlO2XJgjg8y8aG84Zs8xe25Tf/rzVzd1vX5GEmSI6PMsqVxT17KG+ryTAAEgiRf9Lb/6k8Kn/itHDuHX/+E+Q+sDCQ455yAi/xPgLwIW+NM555/7IH7Wb5klgv32T/DpP/YqP/T7f55//eJn2Ufd7HNaUuFUUnxY6nb9/tTKjJQNBiFJxkiiM5HBzuxDx5wsc7QNk6hrjU1UnKDeFEYyOReMQDKdiVpOSOaePXDPag0+ZUcsp/EgEwD71LNPHQC9mU8CUH1tU157zpZ353N8F/ES2++ZsnCk483QMSZPRHgoM4OZuW/3PLA3eIlYMrepb69tya2kqKuTyARYDFO2WMkMZtbrXLIeSyYiJwGilR2S+Me+89PcfvR1wqd/XVn8M7c+qMyBnPN/DvznH9Tr/1ZZZrtl/2Pfyxv/hOVHf+zv83vP/g5GEmPyZcPqCVeDQsSQssGZWB4zrXYPyYKAlXgSJHoXOZfEC91BT/9s2IeOY/TMUTEKa1I7pZ+2ZFV2bJ2WE5bEuT3iJZDQzMKSsCQ6iRyzZ586xuToTWi/g5dIxGDRQGFLXLJSSoyw4yV3jZGMJbXAqD8/EcuNHJPhNvVM2fIhd8nWKA5wlQY6ieU100mWFfNpAy9moZPYsof1Wn/f+u8/dO+z/Ic/8fv4yP/meXB4vj6AJb7j+Pt/gLf+tSP/5nf/RbwEjtnjJbbAkLIwJndy03qJJw1qI3ozOpPoTSg1dCSVTgHQUvq7a86W29hzEzpCsoRsOEZ/kp1UYLOCmVs38cpwjZfImR3pSzkxZdfeY635j8lzTL7d5LV8WN/s+jvk9rNqBnGdBu7Zg75eCRANWCWflAwPwxmWzMvuinNzKD9DiNkUzCAQc9eeb8kkUistNNsJwGnWUbGGinPUlbLh2//ArzD/2Y8SPrNu2v3GWs+DwzO23Gsf4uHv/jiP/sgtf/x7/t9szUjEMCYNDHrSLllDSJaE0Elgaya8idyEvuEOOzdybo/6tXJaVkBtzlaxAiMNM9Cva+pfv8dJLEHI8EKnJcIheo7lv9rx6GzkleGG3gTO7ZEH7oZOQsMGvMQGIlYcAFgyBBJj0iDytAzFm7gKaJ69RAaZ72AUp4HFFlzjvXgGwIfcYx7YG67TwIyCjBWknLJdlUoJSlkBlOwhtMxMX1uD0AxQOhd1/fALn+HP/cSP8fr/9nlweL6+xuVe+xBf/Oc/wf0f/wJ/5EN/mRfc7clNYyRxEweOq8CQsuBMpDeBwcxsjdbxN/SkbBq6r8FBg8ycLQaYgV4Sc7bYXADFcprW1waDkVxSfsO4uvHO3MSZm5iTZUyWY/Tc80fOnQaiF9wtWzMyl8AQMZBhMDMA++jZl86KN5GYDWNSTCJmsDmBoQW0+uf67/vYk4xha8YWUDzhicwDNJi8xxleAi/aG3YycSxtTYCtjBg8CYMhQXYKop4Au5m4ilkVc6hZRKwcDUmYbPjYH/g1+NMvE99556vfGN/E9Tw4fDOXCOb7PsVbP/oC8gff43te/Hl2buQ6DtyzewYTWv99LCl4JeTUFLuXiTO79NRjNjiTcGYsmYQi5rV//7TefEPqyyavaX8lDVnRVmafZg6pU55Ea3NqcPrI8JiX/A0AWzO27kS9UTXAZY7JYyW19qKVBNkwt5IgPXFz1/ejWUcu7ye17Gef+pMAQdbn332dMXkexy2dRM7NQb+n8B2mbBlkbuBkfY+xZBJrvgQsnY51gEgFqKzf/333v8hf/9EfYfP/eB4cnq+vcIlzpH/se/jln9jw+3773+MH/c+zT1rXX7hDywBA6/KI3tC1TRlZWn7n9thO+zlribE1E4OZ2wl7zI7rODBny5gcY3INyFyDejVTGZOnN3MrL+65A95GjM0k9uxjxz51DfDc2CVAndkj9+1t6Qx0hRFZuik5YUU5CtdxOOEbVBZlWnVF6vtpz1kFBi0FlmzimD2DJvjt9chPBpp96rlKSnwazExHJJbXnLOFrDe2p3QlRLOp6zSsmJ/pBGeof/ryX0RIZB64W976ySOf+C8G0vH4690m3/T1PDh8A5f4jvQj380v/WTPP/6Dv8hvP3ubb+3fYcqOR2HXbuCb2LcbZ13/j2Ta/SKwtSPz6kavp/C5PbYNvE8dj+YtY9KTT4OD4hH7oJjBflYwztuIL52Nylkwojfk1k3c647c8wd6o/iGkczWTJxb3fi9mdmVjsCcLZ0o+crnUE5hzST2qWtZQ8ymAaQpC2EVpCIGwxIAEoJlqfXXvIOIMGfXbtSYT4PLGjDcp55OIikZZgkMMmsXRRJHfPl+LS+0BHMtmEzZaTCVpXVcl5USjLLBSsRI4ie+82/xV/7p38H2P/3vvspd881bz4PDN2DZFx/w1h/9FI9/55E/+j1/h+9Lmvq/5G6YsuMmDoButNvQa81abrx6w83ZYklszcgxe8bkW1pdOw+GzNYe6Utffh0YajmiQcPzeNrw6LjhMGm7MiXB2sTgA4NTlmJIpt1+l+PAu4czehfYuJkLf+T1zWO2dsRIYhB9n3oD2SfS8F1rI26eDGhoJhTF0K/KoLpaxnCnC1F/H33OshpOICVA3CHzx1JGGBIRZWhaMufmwK7wL6JosLlKQynnKj9j5u6q76GWSjGbhu/cswfe/WN7Pv4Xt6T9s6m5eL/1PDh8gMu++IA3/4VP8ZGf/DT/01f/Y6bsuIwbbuJAL0E3UhIu40ZbjCYuuogs7FPXevODzHhbug3RMIue/jVbAL3R+gL4jclzHYeTtDpm4b1xx+W0YQyOKbgWGGI0zLNlnh2hn9h0c2tTxiSkZDgCB+uZO8vWTTyeNxyi50V/y7f077YyJpVTt9bvZnWC23IC16Dma3eA3JieCYF0StU2PNm9WAeKdbaxXvFOJlL/nLNrN/qcHbcloG1lXFqtpaMSC+5Tb/r1z7OSMFl5G5bUMJ6KY3iJ/Cuf+hn+2nf/Y/A3//6X3jDP2HoeHL7eSwTzA9/FZ/7Z+3zsd3+G/9Frf0EzgtRjSdyziik8Cjsu46bhC7ULoaBaPCE4wbKx62ODBI44KBsQaK3HCl4eoucQO0I2vDfueHjYEpIhJkPKkIqmIqXlhgqzZV96/s6m9vVY0mcvmcHNTZ+hOMmR67hpKbeXgJWMRclWtR1YMwojiXtuzz72jNnRSyivvbQqvY1LoACepoe4yy8AWnlQwVcjp3iKfl9hit55/Jg8GBhQTKICtbGRy6RpSdKKA6GfSyoaDFt+d9V87GSk6wK/8i+e84m/bSFFfqOs58Hh67jM934nv/Sv3efHf/fP8AeH9zgWLMFLpC8AoTLySsswWY549rFrG+/Mji0AzOVm8jm0U3ip0TUlTrJoKnqjp/1l3PAobFtgeGN/wTu3O8JKgQlKiY7RsPb7EZNISQjJtOzBmoRIZutnXt7c4AqAd+6OvOhveeBuMZIKSFlATHvLzkxNd6G/S3fSGvQmkJK+n8HMJ5RkL6m1Olv3pOAM67bm0wJE+zzW2olV1mDvUKbXz6/lRlcu05zdkgGtWKD18+oknLyf9ee2Jmb90z/2N/ml7/520t//B099r8/ieh4cvsZltltuf7/Sm3/sd/1dft/2ZwDtEFREfalJl2xgMHPrHHiJzNi26UIyheBk6GWLdaltRFsYj01LIUsnQ0uRnkfzlut5YM6Gt0pgmIJTL4dk1JuhbHTnYmsBGpMwAilD5yJnflKthI1s3dTk1wAbO7fAADTcBIrOIKo/x4U9sjNja/HdkhadgkS8rR0YaQCjL2WHBtJyQrcsQDB35NNPdDWe0qlYXzPlK5gSaGnBoGYK2oKtMu7KpFyVMJKK6Eyfc5v6pumoOFAjmBU86MId+dWfeIGP/68FfoNMmXseHL7KZc7PufoD382jf+mWP/LJ/7YReS7jhl5CI/sorqDgnDeBc3MsJ+teT5jkGwX68bzlvXELgDMqiJqzxZvAPXsomzi2GnguMmJfAtCcXWNQArx9OOftmzPmaInRkJLgXKT3AWdTwzCqRHtdbtyOmh7v/KTCpuA5Bo8zkXNfadHCu+GsEZhMESk1CnfckTAMRQRVv1Y7C8fkGxhYg4fiLQvvwpBLYIgn3AxAORnkk7JhjVOssYa7pUX9bAZql6dkFYXBCVpm7FPf2r3r0qa+93Xp18qYEnhqSVSDxad+56cJ3/ltxF/4pa9h533j1vPg8OtcZrvl+p/+Ph7+5C3/zCd+ptT47mSTzGIbD6BuSm8iNmsG4EsGcM8e2sY5RM/lPDTtQhUwVYLNhTk0wK5uzJqdrPkMx6Q38Bx6jqHoIJJgTMK5zOADvdMaf40/wJPmLr0LeKsZyTE5OhuVBemPGMlchg0hWYwkQlaB1s6NjfNgTeZx1GDnJXLf7hsI6CXgbWDIcwuSEcMMC+vQhOLHYFsbE2hZRspG9Q3ZlG7CIpGuWdh6tcxh1R6NFdSFFhiqKKsCl+RFyk15fzWY+RK09MLdyS7uuEh96vwt/p8/8Qm+5d95Hhx+8ywR3Mc/xq/9xOu43/aIH3nt73PhVMBzE/sTFSRAkju9MwOeerpbBlkwgkr8mZJjsDMvdHum5JiT5b4/8GJ3U9iSi2cDLDdQ3ZxztlyGbWuHvn04J2Vh081kFOBLyRCiYQqLExPQwMYqlxbJpbyQQtFOnLmR+92Bc6eZzyF6VXmW99IVlD4UolVN3aGc0GZmTJ57dt/amsvlSa0rc8xdwy7qsquyzGSDN+GJAFEziMqOfD/1qDpkqchryhafHbAElU7iSXdl/X1r0pOBJ4NCZZ+uAsJas2Ik47//MfaFF4iPHj31/T1L63lw+DLLffxjfOZf+ggPfuwNfuTezzbbNSMJX4A5qDeSKWSeRSZdVZDaIw+cmwO20Ijrug4Dc7K81N/wyeEdjCQuw1ZLConsU4+PoZ1oU3NTSk2+XYlFh+h553jWXJl23URnI4fZE+5IrivuIOXxWmqsvRw6EznzIw+6W7xErsNALAGoajtUxxFOb4oCxl2HAUOmt4Fzd9RMyVrO7bGl+euW5yATg5045o6bOJBWrVp9rjAn11qGeu3NCVvyrmuVIa4YoauuRWWe1u+jYhq0TORp7dL6+62p0xVMnVefa6Wep8JRiRh+90d/iZ/5Z36Ie/+X/5ZnfT0PDk9Z4hzhd34/n/5nOz703W/znbtf1I0+9wxWN+WUHJjQNpae3qbdKFUmXTsUyiQ8NB8DoPkdbOzMzk58rH/YAgMUNiNOQS1J7MzYTrXamQBaWl5vxilZYjJ4G+mt+iR0pZSoASEmwYj6RlqzcBr2Y8dx8syzZbOZ2L2gmMPDaceULFMJOo1URcaZyFSyiCk6Dk5Lh8fjhutJsxRBs5edn3jQ3/I9Z2/wqr8sGEw4IRANZmZgonOBx3Fhdy6AoQYIigR9LdF+msflXd+HNfMyZUOS05KkBqvFx8E+ERjqWhiZCwh51xOi4lHkxMbO3Pxz1zz4zy6IV1fvuwefhfU8OKyW9D3ht38Pv/pHPd/5PZ/je0xiipbraaB3odXSjVpMLspBdVnqTWBjl5all9g2yj51+LghGgXHqpR5a0a+Y/tmq2NrFjAYlSKPybGPnf4sp2SoToIi5WJ4WADB+r6qvZu3i91aZyIxmRYgFNNQj8ichZtjz+HoSdGSi61C18+8sD2wdRMhG6bgCNk8WcsLDXdQB+rIMXiu557bqWtgJ8AULcfguJl69qHjleGGB/6Wj/Xv8aK9Uc+K5BsHIWbDebF7q6a0tY6vGRqrblD93epqbcuqCl2t+lilQceCWxjSKXZAYsbypVa6m9mULKWRwpCT9/VjH/tFfuYPP/vZw/PgANhXX+GdP/hJHn5f5pXveodvcYGbqW83l7fqaRCyaQi6KQHBSm6sxI1VRF43+HKCaPtRgbR96vWUEqUUV4fjd8I5l2FbnJOWoFKzh4ieqBgasFkp0vW0rTiAk4STxS16RvA2YkvgGIPjECzj7BmPnhgMJAHJuD5wth0565WjcDUNJ9ZtFYdYG7BoW285tR+PG66OizFKM681CSPCnAw3s17fh9OWzx5e4GObR3zH8EaTSU/FE3/KrnV3qqVcyra1VBsNu3wmNSCs1Zn69adzIeZihz8UMDJhmviq6imO0T8luKyyDdHMId39zFvJubSpLYkzO/L4D99y788+26So39LBwb78Mu/8oW/j8e878MqDt7g/O8bZMQWLEXBWhUOamhtCgs5We7GyOZPFieVFf1tkvvbk5Gj/Nuo/oL4LY0mHE8fkixnKQpn2ErkMG3oT6E3gEDtuQ8+1GZovQ8w0TELTXv23M5HOav0fkmm8CSj9/WQ4zI7j5EnJ4HxETMLaTOcDm25GJHMMCirmQoLqSzlVORFzXDQd3kYSen2u54FH+02bdwHL7AuSIZYH93NHZ5URecTzD0LHw3nHt23e5iV3deKlUN2izu0BL6EoJGu2pD4TvZlhxYg0xJMAdrfcMFIBzBUzcqWbWAx7w0lJcrc8uSvuqlnjei1S+OWx3/6xX+ONf/S7yM8wpfq3XHAQ57Avvcg7/9QnePd3T9x/8JAhWh7dbBWlNzqcJUsGLN7PzQKtyorX8xRAT/eb2LeyY82Yq5tnThZjMgOztiGjb0arOzNyI0WROff0RnUXqdxJlQdxE3seONccierP2pqJOdtW1oRk6YzqG2spEJNhHzTwzdEiAtYmjEls+oSziZiEw+RbKWBMnR9xWs/XUzskQ2ciIRmMzVxPGx7tN8RCtEIytjhZV5ZlygqATtEyhsWbskN4+3jGIXpeH3Z8pHvIA3ejMmyjGUQtw4xkLuPmJJ2fs9USY3XN73pQwoIXPBkwFHuo+ol6g1fc4S4RqmENVHOcVfuV01KjL4Fubajzsc1D/tq//D182zNMqf6tExyK5uHX/sh94nfe4P0VbnJc32zaRjYmA0lbkVmwJtK7gDPaFXjCGXk1FamapKqAam798/WGAMUe9kkxhJf8dUljE2f2SG92HGLHIXYkhI2ZiBhCNozRcRUGHtpdk2RXoK43Mz36Mw9RdRF149esYU5Ggca41M85QwiWZAxzzKQisIolKFlrOAC91w3vzEKcavgDgpfEIXj2s2fTzWxL+1SA3gVieQ9S3ZpKcD3zI86kluGA2s+9cbynWEvX82H/SANCShxzxzF3dKKksBoglo6RIa3mUlRsp652g4tp5RFisBgiS+s1IkV2ra/zfqVF/ZltP5yI4OIJ67N2U2zZN3O2/Nhv/1k+90PfTf4bP/vULfvNXr/pg4M4h3z3t/GZH39A+r5rjLkiT47DrdM622SMVTpxLpuhbmy3YhCuZb9eEq7Yr5/Z8aQjURmAda2dl9amLFbU9u2Bu1GhTtb6tjfqtgRwSB2Ppw3vHXckhH3o1B+hX06arR0ZZOYy6uwHZ+JJS3FOlv3s2Y8dc9DAIKVtGZOyJufZYExuwYKsYGVadQC8jXQlWALsJNIVgtTN3HOYPX4VPOcSEPazb5iDNamVY65oTDZ2puJ91Uq/OlFfxg1GEi+7ay0zkhLAJtxJgIA15iCQHZ5F89B4CyuK9HrV1nC141uTq/TzDieg6NO0HM3OrmYJJcCsGZ21BATFJD48POav/LEd3/53ns0ZF79pg4O9uOD6934XX/gxsA9GRG6JkyMH/aBypt0EOYNImQdZbdBN0klPZcPqB6ytwY2debm7ZmvHdjLM2XJd9AW1U1E3UdVYeIn4QripBKaKPfgi4HEmQYIxOq5Dz6Nxy2EuDs2SeDhtebm75qwAl4OoY9OaN+EKWBokMUfLzbFnmhw5Cdal1s4MwZBiuR4lczKSsT7ivWZMZ/3E1k8MNpxkTyEZQlI36oeHLdMq8BjRVmksfhB2lb7XwORt5BA8IZcSSHTeReVL1LbgPvY8JPPR7j0Ge8O1bBpO00nggb3lsrAw61KugmnU66bULCXDibpyVV7E2q2AJUDU/STp5Eavz6mEqPWqGcTTINAWWAq4/QM/9CuML79EeOPNpzz7m7t+0wUHe3HB9e/5Lr7we6B7ZY8NljhZclq33zQI1ABBGdJizCJIskZPS1ltbEPm3B15we+b+1ENCsqU1Hbm1k5tYwKksrF8qYmrcrGyAlVtaAoi33FLz6HSqMmNwTgnyxR1iM25PzbNQOMBYArzMnAdeh15NytnoQaBebYIluT1pBOTsTaxGyacjU8M1tX3lwnZkJKFWP6dTMtKpqAiMSnPnYvMO+ei/KxYRbm+GpAdIpnbuWMyVaquI/uGEoBDtoxW8RwjiY96bXneSt90DV4CZ/aoGcQag6j7gZVN3Wo9Qa0un8nyO58GgtPvfdLh+uS1msDrKU5UJejUsvM7z9/iv/gjP8rL/97z4PDBrDIN6nN/+FVuPhHwLxyRYBhveoiidYJRTkINDMiCp4voTVJPToCNXyzPUxYGN/NSf8OL/raZlRyT5715x8NJFYjeRK5ZUuZ6UxkyGzuzsVMLHFs7QqLNouhkcZB+swqPSjr/wnDAGbVqe9Dt2Vr1gKglSg0OlkQoKfIxeg7BtxPdlWAQ5mruIuw2U1Nidi40IlRa4iHWpDbibvm6/m7XY0coGMaaVakirzs3UPGGgDr5O7chNXOyy2le0MNQSGVzssUERzOt1/3DAsiKDsvJBivjyXXQ91p0GjUoZFaZ3KnHQ8Ue1mSnlBex1rpTMXPHvv4rWE8LNFV74U1k++NvYf+TV4hvvf0Vvd43av3GDg41KPz4q0w/fIMx13DTMV/2S0AQnh4YSkDQ16lgpK7OaW1dASkniVdWgQEUWHxnOufdcaeovY2kKHrClnS1pp3OLGPnq1YBYBbX0tUHVluhZ/bI/eLTOCbHh4YrPtRd8ShsuY09hsx1HFrZkir3osifr8KGMblGONp02m2pN7VIJkbD0M94FwnRcAyWMegIvM5FqmzblMO1/j65tEVjFsbZYU2m94GYBGty+9OYpDqOKg+H1v4EmgIUIsfgGAroW12uQcuWfe44Nowi8q45p5PAy+6KjqjeC5LwRF52V7wTLtinrtGZ5+RbxtZ0GHCCH63FWDEb5Ww0WvQ6mJxmGk/ziniac/aXW4bM7/rQL/Gf/3O/g5f/vefB4euyxHe89VM/zM2P7nH+mjA60mRhNEgQsi0BwWonQoz+t/6MazioMx6bx4GNbaPqBKebJrQC7Yu/M53z9lEHpdSbfSqGrKACJC0LhG2TbxtInoi2/zZ2bm3IS7bNEOYlf01ytUTQ771ZTZ46lDboPbfHyGINV7UdN7MqMudoOe9HOmu5GTtyMgyFx5CzMAd9j+ofmXE20rnQsgABzrqRzkbGwpBMWUoLMtO5wHFWuznQkiUGzUxy0oytYjyYEphraWEUf+j6mThMDazceKVyD2XWppHMFG3LzkABxvt2TyfqLlW7AusAUQHg2OjRkWoWs3aIAhqLsoqxltlXdX9o2felVsqGJ9GHJ0uQu8tWdu0ffvuZyx5+wwaHHKOiz7MhHN1SPmQKXQ9wqQWFuhlrAMhZyCVbEJPbDcPq61s/8mJ/S1d4Bz2BmA1XYeDt4xlTcrhisnKMnjGqJ+Nh9oSop+a2m9uoe0Omd6HV1UYS9/2hmah4iXRm5IG95WHcUUfZ7VfToRM6Au8qDGzNpMQrFOCsLdJK160cAm9jay9aya2tCTTBVec0MHSFdu1EgcvBBkI23Mwd+7FTnsSk2yZGQ44lEJTOD/XvUX0gEVpArte7/oldA5XKoDzMrpU5QinvCij8eN40YHYoxruxdAYqIexld8XDcMZ1UnC4BojWZrwTIOpacx6eyATyMnfD5qV9vW6h6veddqnW64kOVv2e8uc/8cqn+as//tt46aefB4evfeXEq//NJVffdk7exmVjosg7PiE2NyxBTHqiI1EBs1pS1MDhTeJef+CVXoe0hGRbmbFPHZ87vKBj6IrQ6XrsG3koJVnRhTOdU0UkLBlGMKbJnfeh40PDNVsztQ00mJnzfGSfeo74BpRduJHH86J23NiZB+6WOTlSAbl6MzPYma23xZwlsXFz62DM0dLbSHSLTsJIZnCB3qrXQ8rCzdRzeVAFZgi22cnFYMmFB5Gj3jg1MIjVzCxnkGggArZ8HkL7fFoAz/rXEGxrpaZkMCYRov6et6Oe44/chl03cdVtCFtLb2Y+0b3NYCaOqSOhN2BXfCMArtPQSoTafVjfsusMogmxMFjiE0KrhSm5cBdat+KOOnO9ahv75LGVgKyu3gQe/nDg5T/Tk8dTSfs3a/3GCw4iuI+8zvE7XuXN39aThtIfzkpcqieVuNT4C8amE6R8bbewduxSp6TERX9sMyETwq5wGeZseWu84GbuuRoHbseuuTfX709JyimoacxcMgi7Kl0WSXFq4qr35p3Krr09GWoDSnI6s2PxSUhMydOjhKHZ2hMthjOJe/7Y0Picha1TEPQYHUE0KOy6qX1dCkXcmcTt3PFov+HqeksK0gIsVEZlJkY5DQwlQ8izWf6tBFOyZCQLmYxE/Xyy5aRNWINyDQzrDC5lIQNxdszBcjt1PB43vDvtGC88nxre4NwcuE4b6nBcoBDFcuNKzKmqSXPLHvT1Kxfh6RPEF2PdSqU+fc7TPCzXZjDrLKF9fZ1FtJEBmd/7Az/Pr/7o9+L+yt964n18M9ZvnOBgLO6jH+bhj77Ouz8ghAcB8eOyQUPdUAI+YbxSg43Jq8CwlBa2nJqZUmJkIWcYOp3JAPrBuTJZykriMmx463jOFy7vMU6uBYX6ujUw+NIZMCbhbaK3EWtUDFX1BM4s3o2hMCzrtOt79qDdDFSP4SVybo9chQEvials4jGp89PWTY1gBRp0zn0xT4mezgS6LsCkhKGcdUZF/fkpC2N0XB4G3rvcMe87vZZZyKlkZAK5izivwGsOLKBNLtc/cTJAIktGyvdKxSBqSZE1xchRyFZakDbl+2tZpEQtDbAJNEBk4TPxBQzqRvX928/xwN4wY7lNfQmglkEmrE08jAtesbAfF0BXHzfNd9KgOoi74OLiECUnMvEaDNZdjCrEurvW5eFJuzsLL3fX/KWfNHzqrz0bpKhnPjiI77AffpX3fsfrvPNDkF6YC0agX6/iIFxJc8kYH58IDCcchlWA0LaabmRvI/f644r0lNhZne40Jsebxwu+cHXB4ejJq4Ev9RwQyfgutEARo+FQavvORowrgOdKm2HIRbyV2LlRqcSlndebuXlCelFG5pzHkm3o5jtEz9H4Nvw2JMN9f+AQfZNvgw6+DckW7wXXlKX7oLMvr449jy93pNGWtF8QlxRQnA24RJoNcQ3qlhseWP5MJRAkEAr24HIL4lJKilyeW9G/ysOIksk547tZAzaKbYhksIUWnQSwvHU4p6pTv2/7OV5x11yYo44BKI8PZuZluVZpe3njsaQvnifbkdq+VDp1pVrH0nlqxLaV6GtNdFtnC+318qlV3MnnvrIWrD/79//A3+dXfsf3Y//q337ie77R65kNDtL3yHd+gnf/0fs8/hSEl2akU/WgPqFG6dxq/FS+Zl3E2tyEQy1jYAGeKgApkrXLKZnzflp1GwybFWPv88f7fPHmnqoZ7+gTKPyIfpjpXGB/7IlRWgCbgsWapCh/a4+WANFaZwoiepMxq9S1ti4rB2K2S1C4DjraDuDCKSmqkrAezxs1pEGzhwfdnp2bSKEn5UjIlsvCvrwdOw6HTrs97YbPDVsACsAoJJM1YDT+CIotFEGV/l0UbEzlcwqyqiJKhoHeYzkaUszEYDToR4MxsXVAFIeopZqQCwELktLCg+fd6Yx/IB8mbd7kZXelNnRp4YEMMvPA3fA4bk9S/nkVIJ4gMmXBCkuAoBjlysw+9yfPfRrmcDdA3F0npUVePCDO7Min/3DHt/3VJ77lG76ereAggr13Qf6WD/POj9zn6pMQtok8JMSpBsLYeAIsWpuWWj8JORmcS/ReT/BYTFQFmjJw+XElc8jCWT8xuFk5CVlwktiUbsAXj/f5wu195mSU4LO6aSrDsO9nBJiCK608g6ywDlBfiFwUiclouw40k6nYQ91UVtTMxZvAGF2bsH0dBh5PG/ZBxVnJCofUccGRe/bQaMKwaBWO0XE5b7jnD6QsTFEdmq6OPcfJMx67Jj4jSysjTu6XCvbWzKI+J2tWIOWaSNRAEH250VflRy5dpNz4J4oNiUk4n9WC3yzXrJV+NYjXYF/eUiz0bSYKsUxxlAf2hqG0iOvUqm3pbOxTj1mBhNUWrh4K7+f4BNo2HYyqateS8fo6azxiXbJUKX973ZWoa3m+Br2I4aPf+ybuWz5K+Mzn+GauLxscRORPA38IeDvn/L3lsQfA/x34OPBrwL+Yc35UvvZvA/8Geob8z3LOf/EreiOf+Dhf+Gc+zM0/fiAFwb0ByeciY8sFYIw4l04IS1CBMikYQGLoZs76id4FDJl3bndNGfi0VWc0NEYkgjeqc3hnOuezNy+ovXsymg2sBFu+Cwxd4THEVY+fUmbYyK6b2biZnR+VEhxUgj0VI1aiY0TNYgJrHsNBFZcSuIxKbroJXeFTqDdhxQtAQbiI8CjslELN4r9wOQ2tzXk9D9yMPftjzzw5UslyFko5BcNRHEemBehtV76SzNYfRVJ+SfaAVSFbckJ2SUsLaF0NU/gnQ8m2YjIteBtZFJwpC85GvFXzGkFB14pH1BLwED3vzmeFop540d3o3M5oSWi34ry4fZPdCSDa9tFTs4flF6zEqKdhCScu1F9mLXM/S4dENJtKIsQMn7r/Fn/9j/4gr/2pz58i5t/g9ZVkDv8B8L8H/sPVY38C+Ms5539XRP5E+fe/JSLfDfwE8D3Ah4G/JCLfkXP+koL1+UM77L9/5H/44n/Bh/0j3gz3+N/9zR+DG69kJpsbfmAKA29dIrRldYRbHfbau8CZG0kIb1+ftY0HnFCEd93ULNirnqErDMXP7+9zO3elDM84F5k7TR77Yab3c2MOqoCrtE8l0XeBi2Hkoj9iyNzOPYdQbNCy4E3C245zP6oN/SrVrHb3Juv8RRs19b1wIykbpqR6CWdUIVqZm7Xt2ZvAuT8yFS+IkA1f2N/j8XHDHA1zXPgKjeOxXjVIULpAGS0VTF66QjYjTh/L03JjiMtYnzSpSKJxxZQyRfR1c1ZMIkbDhCNG0/wltOW72Nhp4MhQBhCDslZ7H9i6iXM3lpNXnbEexjOsZF60NxzFM2Wa29N9u+e9eAb5tKNQHamrxHuhXa9mhZSWZsWClm2XT/6+DjHVYLZiDFWUV9c+dtzEvmWNYyrzQn7vW7j/6CPf1OzhywaHnPNfE5GP33n4x4HfVf7+Z4C/Cvxb5fE/l3MegU+LyC8DPwL89S/1M15/6T3+zdf/y+IEHPmQu+Rf/8H/H//pZ76/vIflJF7/3ZrcjFjqABaRXPwXckurX91c897t9qRFVoOEtZFtyRpSFrpyUwG8ebzgalQyjTeJrZ+bE5QzCe8iMclJO7OCZ5t+5pWzG6xJXE/qp3iYfONBVHVi59Tf0ZnY3JZiVmOX3oQmCNpafY8Rw4U/MkZH39/iJLYRenr9Ei+4WyVcmcBVGDD03MaOY/CMsxK15rmkuZIxFuKkhCaylmft8zeZ3CW9qbuE7VWMlqIghZKe5hUtOQg56gkorb0JuYKSggaTqGVXNCobT1F9JJyr3Qtp5ZiRTCiAZbKyBNcyiUufk5aRg2RuU8/OjCpuS32bQVGngV8XqXdd627FOuVPqGVdc/7Krk09vwsoLhO/hZhtKy0MmQgnLlExGx6FLe+OZ7xzOCvmvZapGOAMLvDFP/RRXvk/PMPB4X3WqznnNwByzm+IyCvl8deBtWvm58tjTywR+SngpwBe+bAr0mOt2waZ+e27X+Izr7zI5Tww2LmZuM7Jsg++CYuqZmDnJ14cbjn3R/zqw93YiYfTrt2Ey8/XD3twgc4sN/zWKafgkDqmaNn6U86B9Ql/Hrkee0I9gcuN5lxiO0xcDCO9Vazg9qCswhBNa5mmJPT9jCuqzzkZrqeBe16dqKtxTF3eqMAIc7rB1LZeT7A5W217OiUB1YG6AI/nDSEZxgKkGpPoukwseA1ArGBkISPpxYn4zQxVZZmKD6RNiOjvE2ejOESqIGTJMkarWV9GM41ZwCXNPKJAMGQP2SfIC0lN99QqOGUprlzLZwYFxCvmv0tLVsuMan5zm3petDdgRq7Tpl2nc3NUifvK4n+9noYJ3NVNDDK/bxlhyW04T3usBQ7NIN6bd7w7nvH5m/u8c3WmWVxtH2flyaRvS7zqO/I8PfXnfNDr6w1IPtmzOa1Klwdz/mngpwE+9f1DvjsoZCcTfdH5b+zMhTuq5fsKcLuJPdfzwG3s2NiZc3fkzKrHYkiK3J9ZTTk/tLsmZFOEQ6aBji8Ot+3E7kzg5e6Gc3vkC+N9zgpXYEp62lqTuN8fuJoGHqehuSztNiMbH7g/qFvzo3Hb5lPq70obR0cWTGnJqRBJaYSjUf+G+37Pi50yM+tYekvGmxmf1UhmKKnvnC29hGY0W12SBz9zbg6t/dabyFEyvY0Ea3GoQ5ORzM3YcRg7bVsmbV+KKC5gV63DNC9UaetLQAlmIT5JKTmkBAep3QpOsYkGeBZGZVmp4AfZCLlEp8perZvKyJOlZMiGKarJzW3o8CbyaF54DYPM3C+itutCiALFZ+o0rvUovabEvBMgmm/kUyXtT8MvlGx14hRVAsOjsOXt8ZzPX9/n7cdnyiuBdu1zEsJsYBdJP/xdyF//e0+8/jdifbXB4S0Rea1kDa8BlRD+eeCjq+d9BPjiV/KCHYWUwoqzjvDouGWKloP3XLiRjZ2aHdsr/ppXvJquVAHTPnVchm0rN+pA2ovucAIu1UErL3fXTRZcZ1xex4Hb0HOMakYyR4u3kfvdoZQekVfPbhijY+OUruyMcga+cHOP62PPPLtGz84ZtX2vmJxNTdKcTcIazSD2oeOqGMvuk4qs7vt9a2P2RSpe52Nuy9CVd+fzJhwas2Ofeu5ZlTUnhJ0buQ49g5sJpW6vuouzftKa3iZCt5i/xGhUji1oUIjKWchitGWcUZyhliCGEhwKiGxXfIjSBqXeWLWVGYUwWsSY9nBKRgMHILVktIr1QESKhR93fS1FiWVzwR2YFyzAS9D2ptWyQEu1U+PYu0Hnbmlx4mQtmVQMer5SFaYht67Fw2nL56/v8+bDC2Lt/uRSgrm0BNAMX/wnd3z07wyk4/Er+jlfz/XVBof/DPhXgX+3/PkXVo//WRH5Uygg+e3Az3y5FxNyOw0rinyV1fCkd4GQbZkhqaj0xs5ao5vAA3fbBqNAxz174J49tKnHdYPcNRZNWTQbkaDGr6Ve1fH2oo7TaLvR9pl7/oBFfRsfdPvCM9B+d0iGN28vuJ06jvPCnIzBLh4GiXLqLr4GnQv0PrSgFZLhNnRcOFd634aH846dHZmt5aViJlunW9XZFnVid/VG2KeOrdGAORolSTWxV1+JWDqTIxqjJi+A95mpBLOaAOTKcCzAMKA8B8mIT4141tqgWZAgEJRYZopKtsrkAWQWskVbwhlyt9yYsYKeQsOHokvM1jLZhC3/eRvpXDxx1l6uo2UW2yzkHsoZXcEirMxE0UG+WzMu8zCesp7u5bR8bU1br+sudRoWNiQUVmv0XB174qHcfjWIBiFHq9e5XJL9RyPxBz/1TckevpJW5v8NBR9fEpHPA/8OGhT+vIj8G8BngX8BIOf8cyLy54GfRwcQ/vEv16kAjfDnZrEFN2gq+/rwmFf66+KKpOYfoNLnlIStmZoO4Tqdmo0aSWylTnYO3HP7FY9e/zw3x3ISL5Ogj8nzgtMPU70BUvNWqPV9QsfAdSbyzuGMR8fNApglYZ4tuWYGaQWoFgGY96qA3JQWqCkgagU7xzIMF9T9aTZ10KwwyMQsamwyl3kIdSBvtXCP2XDMnp0Z1SUpbAqXInHmQ5NBY+F2tliTmYK0boyCqpGcdPBLToLpIjkKKWg7txGhoiCzYCZR0WMsLMhU4kHVWWQNCBWojxbMVIhVNik+AdpCtWjrs5QsKYkGoiTEYBHJjAWjGAfNfC561ZPc88dGTZ+T5WHetc+6zuisQcLYVCaVPzmBu3k7PCVA1KE7tnhJ3AUmYVFhrkVWdXzAFB3j6FdZ1yprCaL/rWLOGz+64/W/+Y3HHr6SbsVPvs+Xfs/7PP9PAn/y1/MmLJleFmr+mOFX5pd54G6bYetYBsXMZVCrN1E1CGZslmFztsyl518HwMRUJiVTmXDVz1FLCJsXcM9L4JaesRCO6oZ5b1ZX6EpZ9sVcdh88V1PffBHWiWnrjMhCq+660Ig9MekUKWdPdXzVAckRy0m4OCDXLGiQWW3UMQwyMuJV1Vnq5FSuly2n29ZOLXPYuamIsjpcTk1OnpJuSGsTMdbug55gprBN50OHHDQNtuXGBjBBg0Cy7bAvxCilUudmtqP/jn1uz6k/Q8uRAlY2ILO8VhRysmRXlLYCUm7I4+RxNjEnw63ruJoGLrojL/W33HcKzN7EgVvbMZgJT2xtSsUj9ify7vWKq8NmXZLGLNQBOJ0UB63CpQCaCnTtPL4GkW9DR5hKQElwSiABmQt5rGAzx5cz8t2fJP+9X+AbuZ4JhmRAuEyWORtus+f/dfUD/PeXr3NR0PudG6n+jL0J3HMH7tk99+1ezVVjCQ7JNa5AlcqO2XDMjjm5kzpya+C2+CRYEhg1cXl3PudRmVZtJHMVBi4L2p+y0NlIT+CL+3sFAEtEFxlnV3wTFURLSdNs6yK7zYSzafFabP17wRppLtdONE2eksNLYixAm05JOrbpWCrEOnBMnkFmjuLpjTQW5TEtwVIt77UDU1ueh9ixsxNzCXSDC8zOErO6VM0CoDdhFIhHS7rpcaO0IAAaEDQAKPBvQskGBFIhPWl80wAhCVJXAmYCREhd1uCRBSZpAULWLdAm9TYKVppMTrlpVaZgG28lFm/Lm7ln687ayL196rmf921YTiwUzXOjJj63qXsqnN5G5LEiSRVLuRogNFuj7bsvteasfh+5BuAaj0p2JVnaNTMz2NEgCfYfPWf4BlcWz0RweHu+4D94+E8A8A+vX+WdvaLNj82mjKJTKfHGz7w43PKJ7bu85K6ZsuU6bRYPRUltquGJ+3NebNZBMwhfUP66YjZcp6ExEc/syE3suQ19cVq2dFYZl1/Y32u+irbMcoiN5+CIQdP9YTuyLaXDFPTcrxqPesQuYjAtdyqoNhujIGJp487Jsadvv9s9q1nVlK1Sg5Nha0cuw1ZFVxiORrOtc3vkle6afep4b6ruVZE56lAaa5RQtJ881mhpMU1KTEo3HnttsVPBE2tSkcBMYIKQfFaSWCkl7irSpJCekoNswcRV88JrViFBsYrUp/Z3SQWXqxwJ9LVzEDBCIpE9rUVduxkmGWbRAHGMni/KPa52A9uzkQtzxIpSoGuAGEppeszLnM42Gu991pf7+npfpaIE9UXh20YFnmRINQAu168mKxLg6lscu5deJL773pf9mV+v9UwEhzE5/sa73wJQvAUUva8bFzSVO8weBtjaiXfDOZdBAacKylVAcfENLBJcSfjVz1ucoMuNifoX1iCzPmFBwbvq3PRw3DW6ctVlCOpWnVIuabnQdZHBB+ao5UNlAHYuKNuPOulaiVw5C49HDYaXwMYNXE89Oz/hJXEZN83tyFgd8DLIhI6Krxb3uWUMFX2fCxnnVX/JF6cXmieEsiw106q/R0qG671TSnXhLsgkDSeQIrBSKnQBWs1yw1c1JrAQqcrpn3zZ7FWkVbKLbLMGgll/jojRP6tWo2gzGmdiLQxLhtQZksmISCOkSdlHSkc23CbDr/EiGzuTMGoSwzpA0MRat7lrmcJdvGGZqG20zMipmMN8+WG7dd2GnhAsYhPIykWrXCv9pZ/8vtjD8R/5OP4v/RYLDupEtHgGrh2KQNPts87yQr/nBy+UMfaZ44vtJjWS2TlNm7dmal4IT/6cU3yB7NrA1DrVuYKSt7FvisfOKFbw1uG8/UwoRJwKPFZTEBs5u9AyqBKfAJxTrcbgA75Mear7qTpJxWhwLtK7yN55Ht9uGLqZrZt0SG+3eEXO2dLJ4pOodnJdo+GaEhxr0Btk5p7bs3OjWtZHLXEMeq3nYDkeOuLeLTf4bJofQ8HnWjAgg4mlXAiKQWS7NC2AOhmOWPg9ZizenuX1klcDmJopAJi5ZFTVQKaenlGKUrqUGdWWbnXta6k2lQurGZ2Wg4fgeXs8ZzAvY0h8stPuew0QU7aN+TitMsq72gpYCFF3M4enlRTVZbryc94tXpjGZo0JNdaJLO5Y8mR8kAxXH+945cUHxPcePvFzPoj1TASHzgS+40I/rJQNt7HTGQ3Fim2wgRf7Pd+xe5NBAm9M9wBaEAElKl1l4UZ6zlLH1kwtm9DnLrr79cBUg3Lnb5NTJmISLtOGMTqdLlVarO+MZ81CHQpBpwSyKhrqXWjj3+ZkWgejzr8YfGBbDFQBxmh5fLNlnpy20QWCM+Re2I+ecfQYk7kcN+zcxEWy9IW/ULMcXzwsr9PAvmU6sQVXU2TE1QL/JX/D544vMJWWmpHM7eS5utpothBW4FjxgdRTWzdvoyp0mThrUGjKS6M3fVNhWg0AJzhf1sdjX0DK8vNSl5c0OwlSsgotO+p/RaNhCxgi2kKtsu4oUuT0kWgEk6oXg2CT4dG45Z4/8Fl5iQt75GV7BWi2UGnS5/bA47hrZKf65zpIrFebrVkPGGRpZVK/NzXR2zH4xbawGluYJTPSwPBkkMkCyUP85OvwWyo4SOT1/nG7kedseRh23IYeZ3RDf2v/DpbEZ6aXiJgyKDa3D8AVVVzKcIMKWXRuZWwElJp+689YfvWKQQwS2OdePSNtaJ4Lb4/nAPQ2YMtpHQr4qCWFcuG9UUl2ElXzdU7rep1Urdbsbx4Gbh5t240no2YQVcacrCMcfTuCR+O4mnrOu54HvtCqkw5tmbPlnt1r5iWhDeAF2tFTqdcVeH3JXXPtBx5NW+ZouZpUnZkLRwEpKX1N66t5S4kZqcvLqe5Aiil36srvUFjJyWdV1RotK2QWUi9kn8nbiPGxtUNr8DBe26c5aedHivLVGMU0KsehEsu8j816397JOGs5Os2+Ua2P0Sk/xQx8bnqRoVcm6VA6CnN2WJkYZOY23/WgPjWerX6SaxLU2i+yZXQlMGgG13E99/p+rYGcSUYWfpUsn9vdWFQhs9uPbjn7O9+YtuYzERzqvIZmeGFmjKjD8Avulk90b5Mw/MPja7w7n+lJUDoM1ah1vVKWVgPO0bYbw5llTF3KpjEOAfap52HY8dZ0wcNpi5OkVNxxq62nUj7UNDUDx+KU7Ip1+z57NYOp6s6ySUMy7I8dYXbEK49MphF9sruzC1RqoO06m3AuMc6Oy2nDW/aCQ+rYmIkzN9KX0XqdBAarBrVrenktr2oL2EtkZ0a+tX+Ht8dz3t6fczt2akxjStdg1iggBSuQpBhADQ6SS4kR9bSrTOVU2pMSS3Dok7YxS4sygwaKTVTOxNqlOqLPyaUGLyQr/XdW1iA0zkWdP6J+HYI1tGu9MTq/YyjTxnd+aoE8Zyly+MRDs+NNew/vdQ7mLG5VWk7cpu6kU7GmTcc7+EI6+ZpQRwQ8bbV9UXUtd2kUNesqpResQN4M05nBfOtHib/4K099/a/neiaCg1AtufTflsSH/WOO1vO6f0RE+NXxFR4G3Yk1jbMUXX+L3sU+vPzpi9TWFrVdTf/3UdFiHVs/MzDzMJzxxfEen7t9gTGqz0LDAoocG4ode1apMVkYNhMibplQXR2NJLMdpjZEZrztIJgTU5RWW6aFD5D7jCnmuL4L3NsdmiTcm8jGTE22rHRcx7kowaeTyGOZC7FHN2lta9oV/XcwMy91N/xKfkl/fr0pCw/BHI0aOG0SMWX8dckcbElvN2mh+QLMBTcoWYAGFVEHKAeZ0n4sHg5pr6ME6rVQbKGUH2O9I0pb06Jtv6LFSCXDqbqPlAxTgGhMI5HdvTG3Tq+ZE5W8OxMZk+NhOGNrRj7kLhlkZsq2AYy7EiDuroozrMFKNepZIO+nOT/FbE4mhzVxm9STAJqbVn08L4EaahYH+08+YPjVz3zgPpPPTHAYyth60LrvwhzZ+pHHccuvTq/wKOwaRbgOXwUa0SdlIb0PYlyHp4KOrFufrmvK9PU8ND/Hy3Hg8nZTLMoW4KvqJAD6YWrTndaj643JDN2MNZmr24F5dMr2K5mCFNpw2Q9tE1SiUAoGsbGZwMZCmpqT5Sb2bJgVgKTDSOI29ZybQ+N9VP7GTSVzFU7/TRxaj//Mjry0ueFm7HSzVnyhZjIJZBZwmTiU99Vn8hCxm2Iy2/wmy/cVoDLbaiybmx9H83MYy2eUWVhvdeUSJO2SLeh1BzGoMKxY8nkfm7MW6A1a7zfN8k6BY5Nyk3dvzVQEeYlj8lyngXNzbO1koDlJ3QUnUz5lT6730t1VS4q7666OQye8Lx2g9nguHaLKNs0gORO2Bvv6ax+418MzERxyIe4AbM3IK/aac3Pkvbjj7x8+2lD4Kblm9rle6l2wgIzrOQT2ToSHJbIbSY1SexnUes2byBgdN8eeUKTYVUWZYiU5JYaNbrQ6c0Et1U2zOTtOXluCRc24tnBfswgBqmt2LsYpIrRT2UomFdDw01cPeHlzS2cDr/Q3zb9gNhPH7OlKF6aTErzMrPhE2eB16O+YHXOyvNjveTjsuD30JFOCV8qkbdLN6jSCRVsBDGUodv2sPIjZQDV58ZlskgZBUCm2pzg/FQ1GzRRcKpLt8vsOkW4IyquwiwqzZgtr5y/rIl0XGAoNfI4GHd8nkAzHrHhP5T5UE5+YDDe557xkDrX9PWfHMXXsCtawxgqqk9TpXtMMtwYIOC0rgCdYke2xp/AipH7/nRZFM+nNqyBRSgtJkM53T7zW13s9E8FBR647vAQ+2b3ND/eRz4eJv3H4Vt6ez09ucCtKf641YNNTlLmI7YOqGW8dS1fKjF5CE8JUS7GYhUPsCFlLj3dudoxH306lGIucOGsWsSmBYQ6WEAym+Eh6r5vucOiYr7tFQFQ+aDMaTZ3TEiBql6K+X9NHvSEkM8+2zZKMxnCcHY/NhrNu5LHZcN+rk9AL7lZLiOxbN8YUTObu3ISqwaiM01e21zzcb4hBj/2YWRiLxdItF2cccXpjpCQMw8whqpRbS6RUxFmRHLTbsIwhhFg3d3X2GuaWBdSBOSksmAK5mMlkzQhyFC21SsZQuSGVM2KLwU+9v+ZVcKifW8o6LWyTLAc81zK0tvZgOu6bPVH0OgKNfRoxp12LWqaWTsUi6ltRqBH8KmswlMnkdfBweaO5/m/1n6z/ZKne6j6RnElnHWa7Je33739jfY3rmQgOdW3NyLmZcHQYFCQMyZJEN7ITdU02JSdfB4ixUIE7E1SVtzotU7sjT+vR2tI85q49/u5+y/HQFe8FfV5ORoGvLtEP+j0hmqZBWKe2+31P2LulF1/1Amj0N7MUpuGycbIvLTufdGwfNEDuej/gz2/LzaAj75UgFni1v+IFf6vB4A4DdJAECUbxYPQ6rINEDaLnbuTB9sBc1KTGRsLkFuWlTYij+Uum2TCPDmsnvZELkCldwhY/gmhN84NQKnnxg3Do5hbIwZCluE4fLKFm1bWdKRqksjGITyWjSoxHzyF3ajabBesSrozyG7wO5vE2tpswZSFgcAWf0j1h6KGJ6YwkbMxNfEeitSW3xSjGFixLnaIyEJiza/urti+NJGLJgushVDOJm9g/aW14J+so31gCRNk3dyoOiSAhYV588FsjOGzNyMe7d/EkxhzYldNqzmoRD0W9WMqFyspJmYI3CCnqh2EkEZJ6MFgSSRZOQv2wtnZsvotzCThX48DNfljGx9dNnLWt1vUzOcNUbqRULOGcD4jA9dVGPQ7i6liop+CsAGYW2pEQNwmGpEYrk4GgN5Ht9HfU8XOG/dgxdDOxWLRXY9UX/L7xOWApl2xpnXU2MmenwqLiX6hf12tSr+9LmxvmpDM25qhl0nT0zZq+4gV6AitAmJLBD4FQDWMl48ownxgM8dZrFyQI5ii4ItNuztOgA48rNTor1iBmFZR96ehEAZ+YD1a5Dn0kJR3LN5uM8YnROQ4Fp6hl3rabOe/HZubjRIfdbPJy2h/LTJBb6XknXPBh/6iUG4sR7Tpz9VJcucQwZxq/oQaIOlmr2vutV+2sLZPXVoFClv+qklUK5ruOH7l8bhIS+WL3gWYPz0Rw8BL5ZPc2g8x8JrzAh93jkjmo4UmQJeqv5xNS/h4KZ0FPBMGJOvvWgFA/2/oh++KLUNNClT8bHu03J2ktAJKxXWIYZjVAmctk6tKVsC4xz5b5um+EHqkAo5SsIoOZS91oVGOQNgmzmzWVDoZqhpKjNMu0CsCNs1NTXcmEaDGik797o8DkMXmiSMEc1gNz1FB1zraoOPV36k1o1vY66CbxyXvvcjVtePP2nIOo1+V8UIGQsbGw99SlqBrpDn1gLirOMFnCZBGTSaPFXFvsqFLu8lFpp8JpR6MRp0LxdrAUoDY39mNONJBTDlY5EX0iHwsOUMufrNldCNq9qIHs6Duuu57eq2/GC8OBMz8+4dxUKedXacM2jty3+ybrrtqVfZmkVVfMUqj5WUu41SSrCgKvl5bOGpDaWMQSxJI8BUgv++duaSE5IzHrZ2EF88pLpF/77Je+wb7K9UwEB0fkvjnwOG14M9zjYXyPl61OZRqLfft6DsDT5hPGrBtduw3LTeIlkkROQKOtmegktJN2nzreOlwsuAJQqbvGJ7bbsXk/1snSdV5FDJZ05TUzqFlC9TAwtHq9aHxIXSZtIrhMmq0GrixId/r7WKuneyX9jLPHmsQ0O7yNnLsjc3LM4lrdmwrvoY5jGyQxmIkH7oYYlOswQyOEHbIvegPhlf6aV3u10ntPtlo2zbYEgyWDEJ8Kz6DMBCmBIe1dYVCWa1BOvpoSa2DU/szTyIZV3YmRcr9LQ+oRZWEmh75gWoKvOMVolBOhmEWuGIY1TKNnni1703F9GBgv1L3r1eGavg/tWtTS9HHctvIiFTq9zanNqjjBH2qZgZrb3m1nrgN1zKYYFgmmYI/VqVxWnZZ1cyO39ubp12pwAEj3dpjhg3GKeiaCQ0a4yj2/Mr3CLx4/xCe6t3nZamspZFNIRQuxZN0esqLj4g1CElOebzAmPDGgpKaSdzsb785nHIKn94Gj9au5E7DdajdjnFRtSSkxxGbV5F97zFxOx9WEXslo/7/shGyK0GgbtexosyCyIv2ZJsCJs8U5JUBNk2W+7dpIOoDDuW/Brs5+HLNyOKIoJbda7mnPfuRYhs1WN+Tqr9l7ZXbOyWIFXt8+bnVxSoZ5LkKmUuOHWfuts7g2ri5NttGZZbJkUc5Ci8cVdE2rnn0sMy5WhM4aVFNXSoryfKEoOnsFWUlLadIyrozSjjM6TKhQvlMUBIOIZjhXx4G99dzOHZfzwCvDDS9317zk1Ldzzk47PyuPIi+RC3PkvXjWpNq695K+31ULvc6/rFOsqlvUmhhXZ3LU7SI2Q1hnqyxGOZSH63+5BIe20QS5dwG/WYODAI/jlp/df4S3j+e8s7vAdNOKv7AEhTmbJ0EdFI9YOzbXZaUQo7J2MrwJbTZB7VG/ebxQLYVRcGueHMaqkzTAYfTaxkxaJliXCEeHuXT6AdZgAgvQmKT4HWhdnXaxpcGihgmNGyAmLwGpCJPmyam/4q3DFNZidkLeaqfmNvZNPWrL77NulekplzGYk7H0lRcCSobqi4nMu+mM9+YdGzvzke1jQjIcJ08Mii+Eg4eD1YnZfSRkwXaRFLSdKaFQnifBpIUS3eLzXRS+fIYSWdiAq92YBfB5GbtX2rwVn8BkcHU2RvlRScjBYFaYDWh5llB3q5RpU9AejVvV5MwDj7otL/pbjE9FZ2FPvB+UXj+fMCVbWxPT/BxqOaGj9pYAUzk6pgWGU9yhWmPW3/1pOCXlspl4Z/+/9ALy3sOvOynqmQgOVhIf9w/5g/f+HvM9y3f6d/Gy4VV/CSxdCXVyKsGilBZOTmdLggqPTnQVpa0HtPblnB2DmXlrvsfVNLQPrnORebb0/YwxicPYEYKm12L1Z9XA0DIGSlAXFjBS0JOxy+RNXNLhtQWYgBn0AxVHoxSnG09KqCoyLhCIBE3xrWTG6Jp+xJpTngdUrodK1dVebmbXfAty+dO1lLo3odnx3fd7Xttc8ei44fp2aJRvM+tNnI0hS1ombqMmLzloOWFmIVkWOXdVZ3Yq3TaBdgoCjVVZvGUhaRmRRV+v8ZDq9a1ZVrnW2Sx4DSVAVPMXsRkhNzC1ej7UzCtl4RgdD6cdD6cdV/2gZYWfG9hY19oYZi6K3gUE1qA7r4DfuirHIWR7JyAU3KFwWervWLkNbd0JFDLfef3BYV9+ifDGm3w91zMRHFIWvsVlvssfy8exIZE4NwecpKaIqymxKdkA8MSNDyxA5FOWsiGXD/QybE7mWWz7Ce+0j16JUBW1NyYRjh7z2GFCzQxoUtuaGueiTkxdhrMZ5xMpCulYfCAqA7AAlsZm4miRg1U/xlkWT0UKUFdafHI0JXPo2uAbX7CGlHXgS50MDQ5LppOx1NZKkDqWsmOo34cG6I2duQ093qhF/qvbG9584wXcpdXYVoRUZBCX2A4j3gdukpAPXRNkSU2S6k0veoPHbtn0NWOgXrP6u5YAk6Msr2EL0zKqv4R+GJCpGgxb6NzQ5m7UWZ6JZoxbV6XR3w0QIVlCvs/GzkqC8jo4d6K2K9X34Zg9aVVa1AARkTugpWllxiEupaDqQDIxsQSKSi+fl/3ztCURzFQNNpaokV6+j7zz7tc1e3gmgsM+d/yFm4+2ltx13HDMnrfni/acGhCacOWkvVRNXmoQOUWLvYlKQipEqCk7umLrfhWWyUd1RN297S3v7nfK2CvMNWMyYXRw5dqw2Mpak0RzWK61dNigN//BEWat12UyugESzSsx32hb1N2apUmSoR4h1TSl2a/1udnNHWLHjQkt/bUNZ4gtoFYJsVkx9iyJzoRiWmswpR//0eERnzu+wNU8MBrHC/2e4XxkunS4W8VPatDK0XB9s8H5iFQRmVWsQAoVm5oeF02GmTVrqN4NYlBGZAYTpPX3syz7Xm8UNYRpGhRB2Zg1ZatLVv+uTNP63sp2COUzrYN656jGOCZnQjYcg+fhvOOL5gUMie/o3qIjclz5N+zMyFUc2r87CY2Alsj0EjiKb1lEKmD5GhRf4w6VKNY4I3dp1AWLyevfL9XrWrKn3mPuXXxdvR6eieAQsuXnDh/hOgwcoseQue/3fHR4qP3zdX/+ffQTak1e6z0PiScwiMHMdBKYSjq9Tz1XoT95jTrhyhcRT07SiDxc+aU1B9pzD3qy649S89TkAcmaas96ktc0Ove5EaPsjcUehFRP1LzcHM3KPS/gVBYgKSvQGVWNpiyNyFO9J1R4tgTHY+qIEqiuUedWB95M4oi5GNYara0v3JHLecN+UvDywfktbzzwpKlf+u9ljsW89wRnyaNVBD4AogEk9fqeTRYonYgmtCqnYipDdysEITVTkKVMy4XCTVScIXUlU5R8OhtjtKvH9LmmW9KTjN6QoXBE6kEzRtdGKIZkwMDjacO5O2LnF9iZiY/69/BoAJiyZSsjk7HN1j7mpaVpShtzMHPz19BRfMvPrW95XV60Jas/VxhEfUxi6fYYTjIHAF568HX1enhGgoPhte4xv/v8DX72+BG+vX+L98IZv3P7K/zq4eVm11ZJTPXvwElqaGUhAukQWuXP1xytZg0pK8D0MJydjM4bnA5mvZ7LfEwXiV6/N1z32FGKenKpoyUvKXLqM2GjG1ZPx/L32gYvH7iMBn9tFnfm1ZyCWnPmXGzTyn/Z1dfQ+tlJUocsN+owHgktM7pLigJl/A1lZsMxdaRyjcZCAqrDYXsT2NiZY/Dchg4BXn3lkjfTfdw73RKkbMb4qKc/UTHZteK0WMmDUUA1lPdeZ7j4rK1Jp6IjiVJav7U+Kzd6LWN8UpB3BfjWD6BxUypoCRiXViQu1WRYm8qM1eUzr2a/deZH25PJsE8dn5sfsDMjD+wNXgJTtsxYLspIvXWJqhwHs9gQSm5U9ZDsidOZuRsYShaRWUqKE1ByhdGQTjGHGiTSxQZ7/x7x8SVfj/VMBIechZs48O3+Pc7NgXMz4SVwzJaPD+/xC7evtefenS9QV8S00uH9Mow6P7Ii9Ep+WTj4514t1Kp3w6ab2XQz7z0+KzwGGvU5y4rcYyFs82KEsj4I0nIaIhl7a3C3pgBunHzotWYnlT1uaFyBZBb3pP3omZIlFKSuzuao16dOdDKSTsors5JBtvFurR6wjd23sTPXRb1ar8XLr1zxTr6HHKpFvLYLlbBlFCeMCto2sFAg9YnsZYXaQgVr6VSPUSdySxl9V29Y64oEOwl2NZ6v4gXTtIwblK2WfhXwWw9cBhi6mcGHdjA7k+itOndVU59asobif3FMnn3s+ez8IlszLlqLbPAF4K229lAOrKwM1Rqk97FrXB39DJagsO5UCLT0aZ1IZNM+sqev1deyAK+9ApdXS6vma1jPRHCoGcEDY/iYy7wbA0d7zefCfb53+By/enj5JNrWnnH93kVsZSArNtGXeQLNlLWcqnXVD79SWnd+wkhiDr6dJM4kxqCYQC4CJCkMvuqXiMB0P5G7hDmWUxI9AVOXl7ZbEvyVwR6lWampocoqM1gFA9JygiSvVOvsMpKF8dixDx30t4De4DrmTU1wzu0Bi2kBA+poeG3RWUnYnFbzL/Q61Zbxzo6ce702gwsqYkqG+y/ecHM7qBqz3O/eJawNjEdPcku1HAvrM0cLEZ3U7XMxkynPCkUCb0pGkKV0GdRfIs7LLM7ZZeygEdXYqDTyaJYWsMnknLBVEX6nvgfavNNTX0hpp3/1fTh3YytJx+y4iQNfnF/go/49LDoxe8Y2U5hGQlu3OYtcu4q9prQwKJcRfndKC020VFPSSBCnZUV5EZ6Qu9frfj5g7118XbKHZyY4WEk8TonbfGDM8KIZ+ZAdGUT4m/1j3iuWQ14iM4u7MiwRv74WVKt3e8JUM8XXoK4xq/uyNYnBzmU2pmnAJMDV7aD0ZptJfcKM2tIjLTToaoN20sESTZ3pI3Jr8TeCHWXBFUppkmrqLEtwSMWQNVnAltO3W4C+eON4dNzwsd0j5mQZRS3ljST2qW8DcExVC2KId9py1WlLBVkLs8+Lpt8P/C1zsoRsuJn6RU1oEmJUMemcemN6G0mdNH5ISoLrtP6YgzIXm6DKFUA207ILpT9rEM1R74ZcWZCrlmUK5feZl/erLeZSQjh1jqoZB5RukKGYzwIYMAmHlrOSMp0LK/attBt4MHMbVPxePOO+3TdTmDm71h5+vDIdNoWQF0sQNpLV5buSoO603WumVFmnWfKq1Z1bZKjch+QErKhP4J2lbV9B7t+D3zTBobQU/6vDJ/jM+BKvdY/5uH+Hv3P4OK/5R9xzex6F7ck05Pr3NfGnui3DagxZljtqRL35qyzXkLnojjpMJtoTH4CmtWg/QHkLCcUU0qAZg8xGiU2ZJndOVYF47fDXtT1JSxmTz5ig2oPYay1eg0vqMnGbCuC5qq0rKSgJ713t+PzmPm4XW+b1vq7bq8CwDJE1zPVa5qxEKkMbDHTmRuZ824IlwGXxrghZIGXlf5SvDd2Ms0kneU0WaxVhDZ2OuhObGwbQuAq181EEZZisgbh+rbYjjeIPtVyQ6iNZWZkV7V+l6jlLU4V2LpyAfrY8Tx2ZpJVnrmSnc9KpanUfDOVweRy3fMhd0knkWMhRg5nwqSPm7ql+DdWsp17DWk5YyURZE6I4yRyWEnZVWpTnJG+xxxXYLnKCT8QXz5E3evL49P3wla5nIjgIepP/3P51xuR4FLb8Qz6Ekdys4U4yg7xw2qvC8P2wiMpUO7fH8v1L8Jirm3MWbuZdc5e2pZw4FJck7bXpjZklkzcZ2QYVANWNXpaWB3pjm4OhuzSYkeLEzEkqmJy29+ykxqupy6Qhk11B5OvJGlnKk3ICjzc9nx/u6dCbzVpLEgrXYX0N3Orvp1hMJ4FZ7HL9VszGC3fkNvRMVq/L3Ct/ImchzEv2Vn0VvJ25Oa66PybhukBOBjFJpeCzZhHVG6LS0VMw5Mlql8GlZjIL+QSUN06Ln5RQivR6X0guuMNyE65reyN6szuTWrenWu6pTb9VAp1ZMIhlBqnqK9TFO7TswUsZhLzaA7Wl2f4tueiB9DFnUlPWSnvPS1sztwAhp50bI2Xk4JfAE1Ime4t75WXC5z7//s/7CtYzERxq6+eYvJJ61spCWXEbWt6uTk/N+u2OwOUuQ60SpyrjTfEHjbwbOzeCSv3AnElcHoZm+yZGwbMcDLiM386apnZCHk074dqHalHgcS9IYNEPVO5K1r5+dgttOlmI5yrIIgPFRKUKs8h5sU4rDkzXNxs+5++zK3Mt6o1vcqbLccmiSqRYJo6rgtNmvUHr9duKDnZpOI3JbOzMpQwMNtBt9JWugSMqDut9wXaCZehmOhf0rdqknIJkivLULTRrV4hKNfj1sUjBgZKpSMFplKeQmwIWyaSkw3yrVL6ullnU0tIu+ELVv2nnwDSHqLVxcPVzrid99WFIsgSJJ7KHrF2g6zzocJvi3xkLd0RdKauNgMMZLYGM5BO8QWqHpvzXulO1c1FLTwvJlYIxV3Q3n7Y1Uya+cg95+52vKXt4JoKDIdGXtPbuzW3QyUXV3jvmagC3ZBN317qMWJiUuaV4dVVzmJtQpmdnwdvIzdSxP3bkEiySUb6C2ES/LZ4OB6+nIJBd0lkKmknjbmwLDCctykKFru3AuIH5vJyPvqL8tNMw+6RO1RWsqllMaeuFW8/Npufzt/ebp+a5PSohqigGtYRYcJjFnKQAtBkGmdinns6kJ8bS3/d7bmPHbeg4d2Nh/OnmnoJt1/M4+XZKp2SYkmEaHelocdtAtlmDbRBkNMjRNN1FWk18MpPR4FGcpKi0cp8Io2Ye4stpX7avMakAklJk0BlnE11hutbPdb3maAmi4wCrKQsoDhFKWZHEaIaRY3Paukob7ud9GUEojUbtJbYBOXdtDKstXd1joB0XmwzWZHLOJJuIwRSDmwJKGlk6Wu3wgdTd6cTd4TtIRrOHD73yNflMPhvBoWQFJudWeNcP666T8Prfd809rSTm0k/WAKPdC/0ZlUi1BJStmdhLp4N4k6V3eoPdjl0zcqnGL8YvvpHT6HQzSonYJoNNEAVz4xbmZM0U1mxB9PHUaZmRvLauJIBMom5JxcMAdJM0f4jJaPZggNlgDobbwzmffqnjMHs+df9tUi/ccwc9/XPAi5ql6szGVRszLxlDnfkwZccgE70xmJwgaZv09eExh+gZk+fF/lapxtFye+g47jttXwI5d3RdoPNBp30VJ6wX7t0yuMC71zvGgydde53XAWVQLssNUKnSlc8QpOglBKmW9QUcjbP+HhHAFezBJoLNjMCxYA4iuWQ1y3jFmj1UTMWZBMk0jUNIFTcpmVQJEHO2Tda9k4lbaDqdY1wPXVxWbwKdia1Nbo2CttVOPxYT44adVNdtV3ghrde5ZBCSK00czaxYcAcly2XigwvkC+6rplQ/E8EBdKPu6U7GyNd1ki2wBJC7aw1C1nysBoo2bIQyAclkzuyxzcFIWeht4N39jmkqN3/ZnyK5SbePo2/KzFzkwkgmTxZzYzGjvu+wyRhXRtWHJWNAVGMQt3npTniUYBVKBgF6czjFGHIVawVBjov+QgKY2TDlgbdL16C3ga1Vw1mfIwNzMS6JRBKT2GZyYwqTcn3dkpg2JiCJKU7NRy5ly3uzYjQP+j3XU7+AaGjnohrBuuLxeDQdOeqp7W0kBEMaLZKqlb1yI5JR0BJfMrBaWlVj3lXg0DfaPl7qmHpmISNkY5qeIhqQYl0XgsF7LV9qgHArf8ldGY7T20A1bImlDWmNZlwDc8sgpmzpJLYA4Qtukwqb5MRiQLJmD6uDybCAkaqz0Iwn1rLCajuzlROyBIawsXSiJaeYTF4foOq2Wz4U8zVRqp+J4CDAmT1yHYeTG9+0GpkWMAx3VJir51eG27rcuKuc0z613gDqvagYw85PHILnej8QZre0FYCun3VwTfFvbO+vsADTcQkMZqaxHrODaHIbDKuBoOouSvpYgkFaHzrF+MWU9Dk7Ie8ddr/Mvag6BAB/I0x9z7vDjpuzDtMvJ14NiobEVDo0i1nJcl06CSemqRYtMYA2Sm9jZ8akG/28GzluFbQVyQ1jiNEgQOdiww2uDgPvHD3zrUeqNX3VS9QgUAKLdMXpqZTTFKt88aVdGYvlXiqldtFanLhvtenVpfthdXzAOJriqGWx5f2C8iWOk46p81b9F879ka2rs0krtqV+nLepVy2FpGboG8W0srW2MykDjVWxmTRAFNC7MjVtMtRzvU5g1xueVVAseAyKTWWHBsekv+MTmENZWYAPvQwPH31VpKhnIjhkVMyytWMbOFPXOoswq9p5+WbTgKO7q2YQkQWlr0Bk7fNXEVKywvXUt2udo6as1ie2/cxxdoRgiMGWfafhPE2L54Lom9TaOQi5TIVShqTqCOKQS8uTdkI0Hn35r7b10tE2mnAlWNV7pp4kgrY73WPL7Xbg+LLnplxD7yJTtgUgU3Q9rcqLkpBjJLGVkWPyHMsYuIrC10AyF+8IgDM3cey0+/NGES55F0nZN0GRNWqtd8zCYd8Tb51mPpMs/AUglexICrchT4I5lJLDl5vEFuC22uCXLApbhEq1Y1R/rVqSCMrk9IuKlyxk1PFaM4pCTooGUyzvHpsNh1nniJ51I52JvLq5Ojl09qlXg+KyF5XzMDKmZXL3murf20AXI0dZSHtmda2sEYKYxiOp5VU2q+zBKI8hdgZEkJxOswbd9CfZQ9p22AcvfFXZw5ciZur1FfmoiPyXIvILIvJzIvI/L48/EJH/j4j8UvnzhdX3/Nsi8ssi8g9F5J/6St6Il8A9e9BoXPGBcuq/H/4QV6kb8ESfeU2ImrM9cYOqEd+biJfEzk3cTkXDYXLj8PfDXFD31c8ugSGOFtlbVUyulHPZZd30UqZIRyEOMN+PpE1U4pQvPIby9eyysggBeeThsS8S7pLyb8rXTuNiW3YUeOy5mXvem3fsy7SmOpfhqVOYig6gTY+W3IxwmmZAlAy0NYv5jiHTmchFd+Te5kidKF6ne83RMhY7O2OSXqexdCpW8unUlb/XkXhBMHvbboTyRpRVORkNkAeDjMVCzpSWcq98k+wTuY/N/l5Wn+OaMdl+fslyrK1dELUBjEkYg2NOhqtx4OFxyy9fvcw/uH6VzxwfcB0Hjtm3kYt1j923tyes1Pq4igc186oiwrVj+amn5KLQpHliZN1X5d+xU8Bx+UV438ygZQ9PySy+3PqywQGVzPwvc87fBfzjwB8Xke8G/gTwl3PO3w785fJvytd+Avge4A8A/0eRpzlorn6BcoNfmAP9avJVxR/q5KDasbi7qg3X6XyLp+MSdVRe/frWaBtwjK65O1dwaNhNbLqZkAwhWGKwjc2WSjvRjNKszNoVLfiCRPUmiJtM2qRlQK1bfYNBg0Kv9nHu0uKvpZUQMpfv6aNmHauuB9R++HKCvnN7xuW84fG8ZV+ovddpaKj66XWT5ouYqh6jlBh1rTf7YGa2VmnmviDwG6cl13Hy7UbLGcbZMUfbHKmVQSotGMJC8pGMOmIVUljzk4yCZFkwipKRNcn8qNwImQyNEFCuvwrUlFWZo7TPzbrUsJGcq5lvbgG/Pp4yiw1eaXNfTht+7fpF/t7l63xmfKlQ1lfkMmIrxe6uijvUqVvrNqYCwjVQ0LpTS/bAijVZ+A79+yT9tWdbZ41kzR7Mdvv053+J9WWDQ875jZzz3y5/vwZ+AXgd+HHgz5Sn/Rngj5S//zjw53LOY87508AvAz/ypX+GtoSMpJY9nL7JZUr2qX9keiJgtM5HeXzNmKxp8SDLhOLezFy4A/vQnRipOhfZlVmX7T+jJ0xOym+o8y5ri0lRYk4mU4dNVgv6tNwgTCtGpUtIH6HMqAR0/By0Vp+UkzN1SbsbEcxUbiBTaNxeN9PtsePxuGFMqgkYk2cqvoiGdKLYrArC2vYEnXhuGqy2dHeqiaqXyIU7srEznQls3MzFoAa84+QK+5B2Aw7dTLebip3eUk4p2aNcrwBElsDQAgGtFJG5gJOr6yuFWi21FFkfngXMrC3O2nVKxTA3Rt1JOZk2ySxlHTdQVwWqa4AAGKMONf6569f4pfFV9rlvowAAXnQ3bIvjVtuHhaznRQFPZ9IT7XYtLZJOTJNKEstNst4yB4HoIXnT8AZpf7KAkfrCLeaZV1/m17u+ksyhLRH5OPCDwH8HvJpzfgM0gACvlKe9Dqybq58vj73vyiwgWc0eYDn9KxBZ/12ziBP0VxZhVn1udeGp5UdNtdfCm62Z2NqJKdmGXDsXOd+OhdCjtmzOJazNupluPHK0mIOoSUkx6dCLtASKeZeJZ8pJqIIsYskKgjTwQN7rcO947FFaG7TNrCwgQ1WFpl6lzjX4pEKWSV0mD4kQVAtxiJ6b0HMdh8UhCs0EtmZsGcGaPWnJqj40M11pc1b/xJ0Z2ZmxiYk6o5jE4GbOu5F7uwMpCVOwbSKVNYlNN/PaC1fEe1FLpyESz5K+3xXeIuUaLTJ13dgmaMfH1FJCalZG+bfeQNmgwWMyOqU7i3YqWpZQ9/BCPKuZQg0ArAhUIdo2mbvur5AMc5ladTlu+MzhJf7h8TVuc9ccxiy5lBe1Tbw+uFIpyRbnrvraJ6WFLe/b5RP6dAsQDsLOaqmwPkdXpUW+U0aEl8759a6vGJAUkTPgPwb+FznnK3n/GuZpX3iiIBKRnwJ+CuDBh3vm7Oiy0oAf2NsyS8I+kRnc5TbUIbhPG1h6F6OYk8PY3GTNoPTh3sxqaW8TISm1dtdNjKupSfXDi5M9TX9h0T9A826YL6KKpaZiBJMEUT0vze4pg39kcTdCLpyHJsxKsgiw6q/m9dviUDa8gTgooyr3CXs2433kMKsKsGYDGgBGqht1JzqH4phO+/LqNZnYycht6vU1ygxOJVRF7rl9w29qDa1Txgy3xW9TJHDWT8zRMgUHLrB9sGf/aKOBcjuTB0MeDdlpkJMgquospUEuQaDOpZC6s/KCu0jhPiiBLC83Sq3Zoc3+EIEULTmV8XqlhExJVm7jilVUQxh1jVrtWZRJKSYxRsfDadv22A9vf7VMxxroJPLA3vBWvtc6Fs4kLbNMZLCBfVBh4LKvcxHP5QXzqqWFKwdCkOJirqBkthXrqifIrx9X+FLrKwoOIuLRwPB/zTn/J+Xht0TktZzzGyLyGvB2efzzwEdX3/4R4It3XzPn/NPATwN87HsvckTlxEZUcnydBh6FnaomKTf6GoMoQFoFLE0JElBYlSVgzA2IMicajIooe4ncxIF73YHrscfbyFk/aS98NUS19sqriWp1ks6p7MN6MiX1dshdbuSeJsWuQ1tKTeluDe62nmJgZiF2xTZtEiTlE2Q/Z4hbVWimQdt3ShjSwLDZTI0JeDP32E2ZAo2WbVO2K9+HrCVEaceZ0trVVu+yyRKGY1ZOhJXMzozMzkKAZIXZWHbo5t7PHe/e7AhJgT5vI4dJqelDN3PwPXlSs17XBxgKFRq9SePOqPtzkWmTBGYhFZxBr2XGHs0CzCZgSEopr2CnyS1rWOZYFLqyUbv6ym7VD027GilYHTkQbJkbkjFZ8EVQVvkRGZ3FeRN0vN1neJF7ds8/MnwWT+Q29+zMxNZMzSNjwboW7OF27tsU7xMjmDrsxiaymAZEZkeT9sdOwBgkRsXs1mSoVEDN1Voly1/x+rLBQTRF+D8Dv5Bz/lOrL/1nwL8K/Lvlz7+wevzPisifAj4MfDvwM1/u56QszDhszs15Zx97xgL6PA1gNKsgAZzYu9j2HL0qdX7AXeHRIKrpcCZx3o9c03PRH5u3IChZJkRLiktrUf1RtP6vA1iyQDjToTVSMIlq9pI6fW4FJu2tQYIauFQ8zASVdce+bOSShUgNPnMpRbpl44tVcNO5SDUutSaxnzuuwoYLdyzXQzMwLQ1SM4KJ2TQG5Xrp83M7OXXkm/793BzB6TV9FLbKMJTEK9tr9rNXdiTqrFXl730fSff3zFHLjv2xK9lY0jaiyfTnCuZV9y1b5N8h6GchJtP3OnlsfzVAFvxmxvmozMvZrhiVppV66wM1VvCyrgIA5oI1CBqEU1KpN7JM7I5Jr0FVc07RMlnL1TzwDw6v8cDd8CF3iU1qJPSivWGfOny2jNlhSWXgko4qPMaEycuQHC0rElK0I2JVhJedkK00n4dsIWxEXaen8L6dinIDnOIQv471lWQOPwr8K8DPisjfLY/9r9Cg8OdF5N8APgv8CwA5558TkT8P/Dza6fjjOa8mhDxlVf2deu2pwGVnJs7tkXmVPdTNfDdQeIknp11bRVNfMYel75y4OzUrZWHnJnZuorOB98KuAFJ6fUM0pIqyV1/Dkg3kAqyFXSafBT3FKrchA7lYrZd+NVlraamS42LFvmgvNIOAemKoL2XNWMQmXBcxhR6cs7Dpp0YPFpQefDkP3Pc9VhLH7DCp5zx1BY9ZMqiqLlTRUFoAtvfZU16CBgjUE+OYPBs7cRs6tn7mNnctBd/6mWNQluG2nxhc4PXdY/7eW68zFicnMTrARyQzjV5LN5uoNhOmjAfsutAs5jfno043N4k5WHwfSD4udPfSmah1fJvk7UoQMJkUDdZFUjGdtTbR93PDpFyxlZuDLXJ07VoElaRwM/V0NhKS4V3O+Dn3ES52R87NgXfiBZbEy+6azyffSgsSjYUZnM7PUO2QZl/eJlJKxMK5oJQV2WVyMRmu4wNTb/UQrEzJzNe1tPiywSHn/F/zdBwB4Pe8z/f8SeBPfqVvQtBeegUPJ1blRRxa9rD8gLoBVljEKh2u5cbTevt1VfpwdQeuI/de6A4nNuJ1xPs8r1h7RvvOUiZmZ5fJ2zKfovD9a/CoABoZzCjYo86QbAzHWu8mMGX8m0EDR+zRk+texO1mDU6jpr73L/bNpcmXydKggqL63veh4yoM9CZgyIxkblPHRZGvV+mxlmZ6PWKpgUxB2GdOMy0o9bHMOmfU+BIcZnZuYusn9rNnDI5oI/fcjPEKYm79zMbNeEl86PyaQ9DrfChtzxiNio+iaETOoo7fJWuoowhzNGCWmz8VSL4SsIzVYFKzqRrjtFOxAH/V7yEWQLTvZ/piYNPZyM4rUE2nN/QYdcrXMTjG2TFFyxgcwehMii8e7/Mr/hW+s/+iZqTZM8jMmT0uU6+KiFCxGqWVT4U1uUi406q0WMRYa7A7dUug+0rWU2geX3Y9EwxJYRk2k2ojrUhht3ZkDmphloqUGJ4upKrrLjkKlg5GxHBMvp0OXlSLYMjNV3BMOk9xihZjI5f7jboZlcCwlBbKLUhdQs6CouRJsQAoHYUVgGZKu87MsLAkl/eYpeAXpV2XBeYLnU/pfCRZIRqVKg8uNGWk8RlbxENrw5qbqefaD20OAwmu04YLeywg4/vvmIrrmLzKrlbNrUq57s3c5PYbO3HRHbkcB/Zj12r0nVtax4Ob2bmRlzc3zTrtZu558/q8eUIShZwN81gCcg3GVeAVBOmyBsvqyiqVt6DBtZYkWaRl3Skt3YgYTTOcTcVhvAYG9OPhEDR16Z22H2sCbE1i42dlPdqo5UW0PJo2fHZ8kXNz4L7dF5WmWUrkqiItPBFlTQZmYxlPWJOUAGF0polRo5/KltSSVpjPLL1IEWHVE+nrt56J4JApQKFoTRezIYqScu7bfcMeamvSlClNJz4OSMsY1oFhDVZais1cdoAqFmM29KKOy66YfIRsGOyMGTI3U09KOkR38RcoN31WNyjZhtaDVyFomV9RQEsJRXMRBLPcJ0/Vj9XWZDYFv+gT7NUKrt/O2H7mYqusxNvJU4ey5izcTF3rGADEJGz9Rk90M2FM5jJsOTcHvFsGCc8FqFwHjApYJqm+k5mqcK3W90r6mZitbS3Oe/7AfqNlRTVSWZN/FuOTRMo6SKYzkW03MwVHcInUCXkyy/WufhkFOHTbuDg5lYna1ibloZQVgjkxfalTy0x1kUp6inunOEZKyox0KwZn9XWwxQhmSqdZlJHM1k1sHc0X4t3xjM/al7D9Oy17sJI5t0fNgKs1ftZWfMUeJCm2IRQ5t2jwCiE3mnjTVJQMIvaG7AwSyu9dQEltw+RTQtRXgTs8E8FhzraJe+oNP2fbxD9r7KF2IurHdLeUuJsx1LUecNqZxDF3qliUxNaOLTisS4qtm3SGZklTa0mXXYZJyC5hz1RymabCUHEJgqEms9V+vvbma+eirvpvUwBNUT4L4SwTzpPiEaOQx45jNGzvHRjcyjjWJGIW9se+teCgdDbKTMgxLeDqnC3vhgsGo1qAjkisWEyJbhoMojqAs2RZa34ILNnDIOEUDd4ohnMz9VxPPZfjwNbP3OsOGMlczz0pC/vQcTP1ZOCdx2ekaElhaS22EYKm9nfBuKyDdCQTZts+/zplu2YM9WIaAyLLa9bAIIWqnAohqmI3OYvayrFALjnrfIv1zAvdc2oW40xisKHZ+j+at/TmPp/o3tE9moV79pbLuGGmZL8GXDaYqFT0aAyHlTtUNYBZSgtWOp0CSvZC9haZI5UMlcuhpYS8OwHi17meieCQUX++B/ZGqaQUR50yMfrcHtin7ql9eWAxB70TGNbB5C5fIhVWJoxszcSF0zmIU3J0ZvFlrAIc3+kwnDQXvm+XsEPQzRgNYpOy2jK06XoZ2C93Te08ZFgNrylfK6dBMirnTk45EpV9aWYh7y35QqXl2pHw7Cevg27LSXm+PZJB6d7JcDN3bKYNV/7AC36vQRJpwiGzYkfCCodA8MRWMuj1LmrWQlKrTkiNYm2rTX7k2g+8d9gSk2EOlptjz3XX88ruppGAxqiO2THpzRkmOSmOZTSNw0FmaVVSuglFbSmSGmiZ4orxkgVjIjGqxL5Su9s0rbyUGpXwVsuKMTic1X0wJ9Mo1BXbWQ/FhQJylw9zY2du4sB1Ggqd2pUAoT4bMRlsXhiTIRumpM7Y1ci3dp2MUc1KnUje2ppGmbSpd5jbsWWtX8/1TAQHAW7iwNZMDCx5dx0nVkGduYBnwBNg49MEWq07cQeXMKXnfEyeaHTD33OHxqD0JmJy4hg9h9k1EpTvAkE0QJg+sNlMq9kJZTJWMBokQHvpa0zBaqCqVGKgGcJkoxTocFYCQxSdbhR1Q8RBhUW1Tj4Ezzg7jkcPlR4smf3Y4W1kjIaUMofJM/WWt0dlyF244yrLWkDZpwnefMkeYmma1w5H9U+MWQfEbs3InCM9dTCOx5nIRTdyO3fa2szCYfK8Z7a8MBwYo+MY9NoNLsDuyBUwXhVVri8ujKXboyhcq2y0vVno7EZym2sxpaKOleolCdYquFdv7JBUWl7FdPWaOqvP8SYRjZYfqXy2MZmWUdRyY06WKbpSjjqcJA54nOnZmonbEoCVnm4LM1WzhyQq7XZFmNWZyGxsa6E3cNWoOW/rWlhpNOpkIewc7tEKdyj+Fmvzl692PRPBAfTmvowbOhfaho3ZcEQR33NzYG86lcQW85e7QQBOs4d1prAMOzWKtpddVl18zqxOMKpIshN4+3DeBDG1Z+58wA6J3TAxzk7T92AwTj/kbCnai8KJKG8tW8qU6lwmTwsryITs1DIuV6MT0Jt+WIm0ukTnAtdTz37stBVYgaxihzbPls5pRhNmZSjGZNiHjoeyY2PnE75HV0xg6jWyBUuoP9MTsWV47DH1LUBXivXO3GgGl+CYdO6o6i9G0uaGjet5M523G/Lm2DcpeIj6Xk3tDmxGwuxIQTAuq02eQJpNwyDSZKmzdMUueMs8OcYypq+yC61LxFhUlxjGOlMkS+lcqAZk6OZ2Ulc9hTUJX/4dVxlF/WjacOc1YSwLczZMhfh0Gba8aG9Kq13h9nv20PaZl0gygdFYgjH0NjBHy1zLl5OuRSIbQ7JaWmUrZCfE3pRa96vjMnyp9cwEB4B97Hkskft2D9DMR+r0oEECM465lRMKhN21llv//S44uciTU7tJBtFU2BZDjuYCVOYRVNRdRBCRspk0GPjSW29af4EcjY6sL3MqsqGRpKoRTLLKvqu1ZNjldiLo6yyvp4g9aip7O5A2GpjCrHMeTTn5JAveRzbdrB6JQXv7dR7k0TgO0fPA3VIZqVsZWzZazUvWY6+r3P1Y6Oxr45xaXtTWcsqm0d43dio3iuWl7S3vsuPm2DOOjivpmWcdFnTvTDGUOv8C4GY/kKIp2AIEsYS56FGC1RPfJUINvFknlQMtMEhtYWYhBqsOS+XfrP7u/URKyqCMVZRVMoiKJ8QsK6Pa0u3Ii/FQG7JUPruYhdvQc2kCj+OWl9114ZFY+iJ/v4l9yx56EwkmEoobesoeqkN1MadRhDdTp62p0SxMZ4atFWROEHNRpa5qp7p+owKSUqLonC37qKlY3XjL1KrAPXvLMWtP/GmtSjh1ogbFyWrvHniifTdl20xfvMRm/nI56ymogcEgEpmjpXOJwQemoOWGMZlhmJuXQRgdHA3mWGTGyxtEM//cyE7ZZubzrB2PtHgdSGH5pT7BJsJosTcGCY6wiYS+CHdiad1ZTmY0AEVApjX5e7dbXjq7ZbCBQ1Ts5qwY0epw3cWEVsVBuqo5zPIrVGZlnVau5YWlPhaIeSAh9CZwZkeuZcCJalWuDz0xWI4HBQHTaLlEadZnfmq28btu5vrY64mZDLdR1FS2mr0EgWRIkwFfdAg2twChxLOScdShOA3YpNGljdPP7Ha27V5a+A+pZTg5C4NfQGBXgpiIjs5zQEhFlr3iLY/JcRm3DGbWA8+MxKhZ1z51LQCbcig5E/FWA+oc3CLnLnb90RqSA9vamYUpue2wl4qZSS62cTVAVFDyNyogCYvj05x1enG9yfWEc/gc6URbZ2PyJ6KV2t68u9Zsyho89PSTEyBOXY/0eb3RCH8oZqG2RPAQLQJs++nEQbjc8YRomCfX0OIWFSrZKaNOyiU7EFRolc6ingar952TIfcZczGz2U7sr3u46TSgHC1Trx0U61JLj63VoOVtZOtVKfn4uOHh1ZZxP9C5wHk3coiefdLybDJWg6PMdwhlqfkk1lU7SYsDuLTnVqDSS8Sb0DKujZ15Zbjmct5gJPGu3RWx0zLGbrrpeDtewIMrXu8vOfMjIRku+iO3c8cxOIxJ3JrMZDJ575R8VoYGUXwoYxW/FWwC6telWfxnl5s/ZRZBJDEdFLPJaVFxSknlT/wdkpQMA5zTzGJwAbfqTacsBVx07d/71HFb9vN9e9s8KL1EZtHORRCrlOqSPXQmqtcnGoBscaZOVt9/skrtTq74kW4c9lEuLMmCO1Tw+2vAHZ6J4JBZbN4SwnUc2uTnKpQ6Jt/ab3vTQeLke+7So9WKXmnSd+nWdRT9XO7IuyYdXmLzd4jJNHDIu6inSXmsmpXGagYzG2VIphoEKrBIAyGlBIuwzaT7s3o7TLZwItANbTNmG1QzcPTkgyP1pc/tFKzyTg1bnS2nnWS2fi49cr2B95MGODGZw9hxCJ4zpwHiaD371De2ZL3Oc3aq3iw+D/o1dc2qszan0resxrVkmg3+WqocsxrUWq9S5Y/ef8znuM84OaJYKMEtBsPlfsO9/shrmysOqKPVXGTTKRmcSzDMTBkyrgCVNJIUwSx2ceU6tond+ma121HxClH36pZZ6EZsU7kySpevZsIhKIMzBZ1vYZziETeuZ+Nntl7nbBrJjezkRKeWj9ZzzdDk8lO23JiBMTkCVThY/0sN72h7rwCvmOrMXduaQuqEsLV05svgDr9Ry4q7S9OxDS81zwEtI2xORT0YdABsvfFrS83MJ4GhkggqOr82fhlkXvlFFDammbiWgTlbVRZKbtlB7wPn/dhaTUa0JViBqlSt2Eu5UElS+mSa/kKSRvu0i7qhC91asrTNLL2CkNN1h33ky2Qn9Hl1FqWNdJ0Gxwe7PXMyjNHSl/f27s2Om8uNGqwKBB85zJ6p0+ncVXZd25bk061QA0JdSyA2J6zU1sbM4E3iPns9HZNnn3XC9Jg8Ozfy2kbr8zeuLpBhwhVC0n70On7wuKEOMbqZeqYy+FY1F65Yz5eJYGVuZlNv1iWclmflRqpO3qDXEFjxKWgSb7E6a9O5om0oz2suYFkxpSyZcXJMs2MKltsSiDsXOesmeqvXpTehDGuauY4brM1c2CP7tOcybMpe0vejE7gsJin+ot4RpVNUMprskg61WeMO55aNt0qGWvEd1GcS8gcovPqGrLsmsfvYsTc9WzNq3Su1fFDNxZwtUZT0dNfjYa3UrP9er7lIw+serxOgKlj0+fGFYhunnY0kmbNuwpuoU7dLHVp1F8fJLMSbXDakW06pDM29KHXFACbK4r5clmQh+4i4RDpa7COPnaC9UIIgHfPOcpUMF+d7zW5KELo+DBxtJGZhnp2mykcNZsE5VX+X51Zn6ho4kXSC19jS6KxdjSrEqriDvoZbMIuq8JSZB+6G67hhzK4FdieZjZ14fXsJwOU4cJi8EotM5jj5Vt8fJo+zkcFr67BzkWEzcTz6kv5n8IF8cIsDVANyUe9Fnxeb/8qPmFSTgdFSTOxq6lRRYNoSOKbJaefDrkhPSf9Nzo3KDVWtKcU/M3Jz7MlZuL9VtuiFO7C1I15i07bct3veted6fXPBerI0f4y5lLFGaLhD83jwWZWaQWN6GIQ0eOz18ZTv8D6u1F/pemaCw90V0dZm7bXP2RV2WWglx1wYIXcDRFqJsKoga63FmLMrSsTV2D3UIHSfOpwsRJeYhd5GNm5WMVEhv7SpyFklxW0ArMnN+zBb3aASytBXV8RZmUYJrv+16cpJSLdOnabqoVwosyaCuxFicASX8ffVmGYM6n85jY5gTatRiQXkTMBkmubCFezArDKnamVWr/3JZ7HKugaZmbFlHuRiZd+uoyQGZrBwzI597PA5tuds7MT97qCZQXCMRz1x6/zN22v1yLMuYc4PXAxHTLRIrxTocCw3tqF4cRo1+K0dHVhs/pOoka+v7WG95rlkazlnzLACGl3pjsxWadnVbUuElcQE4xNdtwC/pnAi6n6IUUlp78UdV8eeQ/D80IPP8nr/CItrJfIr/op97HRWaTmEannhbcQn7ZjEov0wkkkuka1plgHJCmGAtHHYyydxh69lPTPBoVKf1wNW1CrdtdS1uhtaopqOZNvAyTUDsmUGZd01MImlVVkdjnTzZy7MkWtzZExKaJkLWl3nKia0+7CYgmYuD0PZRCVKFy2AprIF1Iplk1ZArKa7Rp8jTgMKGTWtndTroekJQlHhdahqc4Z0sOwnz66f6F0gUxSZSXA24TbaootThuPSI3Um0ZtAL6FpJE4+hyJ6Q2jDbWI2rcVp0RutEqjuBpKKQxhJPLAGOm1Rj8mV1xZeGa6ZkoqVxqMn3HpVWQ5RU/ZZ6VaX5TVf3N7SFbbiI+C47zQYG3TSeXGTohrxtDeTGyiJzToPpKB0YnKbO1LLwVSCfs0QXBcLi9IgzcELjI3Fir/yJYoi1NCEXKCZxjh63uCcv8XHiA8MnxzeVtOdHNXvoet4Y7pPaDNXqlt1WohRYpaWtTEntvUIxEGYLjz+7YI71NLiaxRiPRPBIfMk9bmmcjcFnKyn/FRqYy9RTzEpPZ07665fw911zB2dHE6coTzKsWhaggIw7ZxKdyu3Ppf3NxdqcK1NFTQS3YQl1ZWD0WAwqJFsngo5qo61SxTTlkwug2alej1A6VvrayWXMZNgj0I6CMej6h7ubw+cdyOdVVyhWsX3w8RsHBU5EMk6lVsiW6sA7Jwdt6lnZ2it3LQKEB2xnci+BIMOpU7DqcdGXD1mUdo7VOWstji9RI7J85HtY21TR8PD8Rwm7QR024kwO+KtYz46LtkQouH+9sDWzxy6mZSU9NQMXS6CgoZHt1Rpk2mS+YZJOA3cUqZzpyCrjA9yTsTZKuHKQLKr/SO0Vmndn0YyY3TMs2k6CGuLg3SWcp9qgHjkN/zMu99CelH4R3afba/zsrtuGNeYXDt0alvTxdSIeLaM+sOqbZwVxRySg/HCsO0cMoWltCgA5VfLlnwmgkNdVRhVT6OUhX3s6CVwbo8FS5DmxaCbOTeXp7tGMKnU0ouDcp22nRsgt66XvWhG8sDf8ln7AiEbNl7NX6oBbVPqSeZ27HR03hrQKilmngyMpozrysigrEPxiewKqS3IAqwVFyh1qF7aUGoeC6lPzasyWzWICZNDtmNTBzqJ5Lxr3pfWzBwEkEycLVOwhKRgZMUc5uzaTVyvhS/Tr2rg7IitQ1GXslilBa67dHYFjhPJmJbJqZguaxdKEt+6K8BgFi6vtnqjdrDdHYmDgsAh6TSqNyfP0M0Yoym9KhYtU/a4LtB1sZnEuC4yTUOTzOeq7Jz1vWqFIRq0M3qhbZnQXQRYTELGnIwRSCSMVeJbSKbNuVgk4baVRykt1yZjOM7qBfE33/sYWzvxQ9tPN3Pfl901Y9JZpCl6nCQCBle0F6O1xCTNkCaVDDQ5PUiSV9whdw45zqelRVERn2RUX+F6JoJDqh4LZTNWUUtd+9ThTeBcjnpir6ZX9WZWh54vYewCi++DtjZjQ9QHO7f3UFPlc3vk5eGGtw7nre8OevIqSJQ5Bsd+3xMrTboQVVIwemrl1XyGXVCr9nKjkuuJpd8no9bNMkvDGShVSi4DVXU25mKFD5APlpuu52IY2+9W3+vWz0XWDNO0iHlCNjyedYaBkYV8ZpwGBWtyIUU9SRqrQdtIoivvsTE4W6KTTgL0YGa2edQhyWXVIBGz4WO7R6RseLebefdqpySpJPR9YNvNZOBgkt6MWRhcZNfNDVi97ns23axeFpPOFmn28gmtu7u1DJZSj9+5W0zGWJ33OY524UqUr9XPP0WlqANFybk8LUW1uY/BtD0hqJ1fjNoCvaHnv37nk7zw2i3f3X+hDd99yV2zj9rdUVJUOqFxj3X/iTqBpVpaWAAhboRw5umuDk+WFl8lMPlMBIe65mxPgkJ9LEVhMJ5BAr3MLeJ6Ca20qK3NutYlxRqcXH/9mH0ZkLqIimrX4twdiYN+XyiCLFeyAmsSh8kvDJOy4dLBaWsy0lLxfBbxQ9ATpmQJed2PT8vfBRrOQAkC2SqgaaJiD9mp+3Qu4OW877ja9ryyVQIOFHAvGa6PPTc3gw6v9TobcooOvAbgQ+zadeqLVsJK4n4BfVNWV64KVj6BL9SukKQi/ZaGT6yt7/V1R+2iZKd4hklEp2UGZ9CVeQ5ffHjBfNNBFqZOZ5lak3E2cJwdt2OnlvjdzNbP3L9/aIEi3hMeXW911qlLMFvFfuY6Y+ROQKjX3mTlECTDPMtp0Ct4hQga+IFQ2qq18msqz1Ii1vYxGcWTMmX8HkwBHu03/JV3v5OPvvZQLeXCBV4ir3WXXIWNSuw5BSc7F4mTOSktmkqzkKHG+57uzRIMvg4qzWciOAh5ZfZSb0jTAKyE5TJsdKhKDiVFdRyzjm/zEhnx7zvlCp50jKqYxFRPMU4dqXsTeG24YkyuRXcrCaz+WUVXretQN1JCs4agXAY7BEJxXG6bsZYSmTavobHxivlRhVGkmMhkQetigTTULAQ4GjVJKXjIWTeynzsOs9eJ4DW1ngy3hx7uwc6Nq9LNcIi+MSbnMgDnLoh797o97dpWMLIGhtvUn7AsE0tnw0tsPpQAU+/Yh47z3ZHHwRJmy+XNBu91pN620wxvv+/JUdh3Hbf9zP2Nvka9kXabkclHZJc57HvCaPV6z4Kk5W7JhQJNua4iEG48cjTLIVu7HaNt7u94LTPUZEaf05yuc7Gwq07YVgOKCIUnsVyzL1xd8F/tPsUfuv932RkdBbA1Ix/qLxnTi+U6LwGizgGpeotYJ2EV3CE7mLfF32EMS2nxNYCSz0RwuLuWoadLwBiT49G8w3exsfWqIUxtbY6lh3XXQLZu9EaOYsks1FT19ALuzNgChEXnaO5DzR4CV+PAPJcbvg6zmUyrcZFM6jN2G0hTAbhgCSJlNUYfLFqnxn5jyS7KrqqGMWYU4rYAE5220vahoysbvi+uz0M/0/cz1+/tkBvHZNVoJSTLzo3qEVmytWPyrbzTqdFlsNCKJQnpZK9pB+N0NmQNLqDBoF3rzIkfpZfQvB8smdQpw/XMj7yzOXI7dYzzauCOSfQuMrlIMtodmCbH1Nlmj9euq6jl38X5nrBVf8bx6Fv2lpNg+9i2WI5l5MCon2FjVpahMjKVeaZlqFE2UOdxUhS4Wcpj68nhBaDOSdmYOnpPo1HKwn//8HU+sXmHb+/fbDqVV/wVj+YtY7KkkiGb0nc48XhwaaFSS2lpbhfc4euxnpng0LKELwGr1hFv3p26RemY8+XGXxOgTtiSnGIORlJxvF5mC4Bar5/ZsXVQzt2RubM8npTBd33s1VMSFrVs5fQXcEs2oZw2+TR4l0ygzsGU+j1CmwCVhtyegxpNLUEkl+9NZZBNHxlHz3v7HRfDsRFyXDlpHj/eIbdqahv3Suyay7VwJuFW/IapcD+O2TPkuTEkdUq3tAyrXjtLwkObiTFj27VUBupM9QWN5JKlLVlHxY6MJHoJzXAnZeHcjxyj4xg8V8ee4+zwVl2qj4cOkajgXNkvXWEUKotYOE4e52KZvqWnb4hKc6/6CGNysb43hKMnb3UIr9ikgjajLMl0LAeBKzyCQo5qhKpZOSWEMravZnmYhm3k0UIfCh0cQLg8DPzXD7+Nl1+5ZmcmHhdNyuv9Y67ChqOo6rMqVjU7NIsQq0twrLb1KsKa7/X0a9whKmHqq1nPRHD4Uv3YdbkRMa282JqJWOzlBgptOj+ZCmsdbE+CB9A2N9BKC3BEmVUB6vbclKGVRjL3/QEjibf2F1pSQJEFl3JhxXPIQ8T6pNZxoFwGYaHrIkubLa98HUoNKVNRdDar+vL1+mdB4N0QSFGYbjsel+nQ1iysv3F2cO1xt6bRt9XVyKyGrKjuISGMydOJektWQdbda9kRm5+CPmZOAkPFbiowaYqIC1FwMiUV0sWsmVhV3aoYTCXlx5VV3+Bmjk7dnqPoDeu7QKzDcdES9Lwf2UrieurbVPScRRmGJnE2jG0wzX7UmRnqeK3zSMQlbKfU6fVEbpOX6efWqU9lirY8Tz+/1Onz02zIB8eJ4ikur5WiBSLW6r07zo4v3Nzjvxm+nd917xeU7FfKi5e7a65D38DJOeVGJTem3DE2k7y2t7PVrsV04eitWXAHyb/BAcm8GHRWbv0a/KoEkZSVrXb5/2/vz6Nly/K7PvCz9z5TRNzpzfky8+VUlYWQBBYSCFnQjFpoAFuSaWgxGJlWW9BN9zI0k2i7cbNsvMBe7mV6LYMRBUJCQgMIUWUBkjUVMqhUGkqlGlRVqimrKjNf5pvvEBEnzjl77/7jt/c+O+LGy3xPqqp8Wfl+a9117404EXEi4uzf/g3f3/c7TNMIrfOanuJU1HBKpCWEr9FiH19oY8ZdzHpNpSy7eonzmmMlDsKhOCiXLKoqdS18ihoIrTAPpaOcdXIhxIXqhLwkIegWhQjUwKjQHUJRNZCKkxEIBcFpIAGQqz1qNsiFuqyg17hSseoLmqoHrzg8mqGNFVl6pYVPopDzPh5qJqYTynrlk6MoA3gpDVkRHESYwowOtvIxmlgXCUqsUvg1stpIKKNxOKUDN0SRhHmnumOlS+E4iAN0IQKQ70rJLu5UYt4yxocpSaG2L7Tj7GRBqV1qeUbSWKOliNwECPxutZLBtL4UAqHQZgTC6L1KdHM+0MiZVFeIeAbRJSV8RUVpsTq0GcPzAFLz8EqEkK1CKZ3IdQAWq4oP3rnEQbHgy2bP0fqK3hsuVkfc7GfcdDMAijAqXxhHb8WJ2UIIZlPLu5ToAa1hsHK+aSO6/17mA+Ec4qi2limRLMUYCWMTeYvXnNiaRk9G6rgAYJL0wSS05SZyMrfRmajESSBRSJf4JBZO8uKlLSmUIPQGZ9ibthwrWC6q8QmtRA3ltCfKqOeSZLYLUm+LArU0AvkNFssKaiBpbXoDFKODcAbR3nTgz3bs7C0T1JjSJRj3Sgkyz7YGVyi5L6Iy/UhScjw0DEEpW3QtbMCZ6DSuXWZEOGygTmMqAePcRRSTBdJ9a3T2G5FbvK1BIsGp7ljqkr1qmTaExvQMXifo+nxZp4Ub5yD6wXDjaMZOtWJadihV0g0m4VHi8JIJ6taRoMU6TWEcMCTdjMRaHcfxE35ldO46vK7APPxIdjuEaV0NZRTgCYI9zglGQWpV4Wr2IphzuGx49+0rXKyOeKQ45KbdweC5UJ0wH6RGNCjhr7ReJAitFvYxGwVvBvBa0e0Kv4M+XNxltd27PRDOIfq02JmQv2WB9+EiKbVNDmJwhmPbUOt+LWLIkZSJ5IVxYEuH3S8eP9YcDKUWoNPc1ZTapkVR655aDymXXw4lZVBv8l6Qls7Ktm52RJptiBiG+OZ8QEA6JQxRwQnEdqUKytprw5FOHILJWakV2B3H2TMn8tQ5dXvopQ+dnIuuJDzWhcPOHLoV1J9WI+GuzXZm1Mj61IWiYuP7wOM5jmK3sUgWvFiMFMpIOutKegwWI3RzwcoQO8SRZSkEj8C1WACeGBG9ORpqpoXoiexXS4YyCMjMDEfzhr4tGEVgPH1XcGsp/JRyfqJS1vXCBzGkBS0j96u+oA8U/lF6oCwt3ktdItWOYxcpICsjL+XQm5G1GhKxjCrDrp4mOjVDLw6lLgd2Jyu6wbBoRUJgGDQrVXD1aI93T57gdx/8Ko3qWfia/WLBXjkTdK4fYf5R8FlrcQ6ulE3EFWBrRbdf0RwuwEmbK01p3qc9GM7BK4ZYnMpC/4hGdCE0yneu+VCza1qpPWR1hpEkRq3BecXZ5Ak8qUjZupJd3couEwa8qtCSE6hxh1aeF9qDoDwtr1cYRzPtWOkKZ+XiEnTciJpTGlRY4XZeSNsyvnxWhIxDUsoKmk35kGnU4hWUA90r3L7j6HjK0EpuqwuHa2WC0AYuCUqPaQaizKDe6fG2kucGOmsoVJF0OoAwZKYTKKr3BQtfJ+XzzdbmGk0cI+kLjI5Avk8pSMZhrUb1zHRH631itda4JI4z1R0nvqbWUjSeBJDa4KU2Mil72komTofOcPPWDkVlcU7R9gWHqsF5RRf5F6yoUXUgrd3wvZhQKLThe3POyMxG+G5NIAl2NkNNIsNZhXF0FJK2BMCTj7olpWAaXGfQpUtDWRF7YpxnWvUUxtH2RWIOHwbDxw7P83hzh7dMXkq0cheqY476hk45hsD1UBhLYTW2cAyFEA/7zuALj60UdmKILFDK+lRvul97MJxDFiHkkUB0DL0zWKUo1Shb13vNjX6Hx+rbCbAT82KBU59+azZu1dGio0i9fZd2tZleBAdhk85AHyHUIZf1iGMrqyFM9HmGvg65ZbgKA6vQ0BWpOCX5oRPUYxjlltZlQEh6OUZbEosRwLDr0aUNAjuM1F/Ko1YmvBUpgkTaOKVkVx0K8Tgeod9nCEVZNXI39l6QgbXuhVnaa+a+QnsBOcGIaZDpy4E2RgCsz1qUobAZUw5LSP8SPFvJZxpTk9CSjn39Wg+snDBy1WZgGCoKLdOxyzK0Jb1iWBZ0KwNWceIm9BPBnwyhDeoGRUSi2s4k59A7hWks2liGwVAUTnb5uTxu6HSq96halD1cKHRG5x8JfyIgjRAd+sDd6QP4CUAbwrWjwth/6BaVA9Rwsqw5Wja8587jnCnn7AdI+1R3nK9PaG1B4bUUJrPowRQCivIBTu1K6HY1s8JAH8d6X8cFSQ8sbRku1nWKeRmRzuclxj7xsTNcU3tcrg7HhZ/PUmzwPGzaSDYrw0AHZpFCXR1C3fjTe0Nri1PtVuFqtOw1K6Gxb3r6fhzG0qG67VdmnI3QGRy6CFT0odvqCqkNqAAh0KFW4EpR7zYa7NKgViYpMKvSwVKe39eip9HUPYWxdEPB8qQUDYyKoDiliHLwIC1ibXySAogzF7234KrAQC2fR4VLrc28vRltjRMiqyvk7eVTXRCv1iZq46YQVb4jI3jrCuGZNFYKr8iCc1baiX5l6I2E0K41iTtSOZU6DqndjEQNrpK6giukC+KsMFxjJQX0hZdmlHEB56DoI8O1V6Ht6cGMyufp+upGfVWrBZfhCiEo1qEuFQvw+7Ml1mlutxN+df4IX7TzQhKIPlMsOCwnDM5QBgGcQUsXQxsZSRc0rfx0Owo3KdFdP7Y0fw15xYPhHLxI0GmvGCBFByZcwDYWJxM4ilSovNNPmZouFSc1CmJn4hWcg83qD7HyDjE8lgJpozuw68XLiFYbbCHAFOMSSs8EgtfYH9fhgrCtQbU6oSkjgazXpB8Iu34f0go7OgavhbaeymFXBtUaVCdO0geuO+XGKEKrUTa+XVYUdwrUAP25IRXiRLxnziTjQIyOMH4eEcwkNQRNwzrgKTnXDR6IvCs0ok7D2H14zLqDKGhdEYiELXNfy7SjEwm6OMYcUYNNEKCJQ062K2TcOwaCVgWl8/DZ9Eqo+MJUJiAt59onKvs0eq2R+lD8PoLaVlLMCvUdH3Q8vTJQh86A1fggxEPEQcTihVeJhHjVl4kI2HmdcBjTekVtBg77hmvdHherI/msdM/Zcs58qBi8zlqa0jXRpcWVYdS9DNRxOxXV4WK9pXmf9kA4B4B2KBOy8c5qQjsUnJ0s2KtaKi0XdR88Z7+24DXXu13Kel1KL1qph5HtCNZqDimH9oFSXZtsarOgUT2VGlhQjfUL7dChheaBWd0JEYwtgtMIuhhh3l9rEcL1pUZ1YdMKeAjVK+lRa4hisa72mEV0CnKbnXh8HXamMNYtUGov/A9LLcpxAZUplXJYdQXDylAOYDpFX3pmZUfUmexcQR+6FWYjKXVeiXPQUDobUJNqbeGniC6bwbCMACkrpBVr9R+QDkapBqpQm1i4OlHaC4DLMkE2gNaKnmYspOZIyLKwzKYr7pxUWK9odley0SiDDQuSyolP7lQahvLGCwGMA/rQsrSKNsygqKBzQTGIoxjGc4/gJx/nNYxP8GrlVGIVT1B6DRRSe4BQXxs03heJg9TogVKLwE106hHwN9UdDtgpVlTa0iqf2ppaSQ2kNwWucviVRDpDo+h3CiqlwAoq0+v7H7R4IJyDUnIhLPuaayc7LNoKOxhuHU052F3y6M4RlRkSiUZuLrTnDs2E/WKJ3hjbzltmMV3YpJCzKDSKIzfhwMyDGreiBM6aE1pX8oIVJaZCScyitZTc9iqZiByCw1BhUMh5Ab3YwBegJgO+DjtaQNMlIJQCV3kyXyW/EcfgagelkwtVhdZVIZGEOjEUS0W/42FnCBV8yXWHvgiFzvBkIU9uzJCKkbHLs2NWqWsQnaPMnkg9JnYvqrvMr+Qs1KfvGwvB0kp06XccnjsmsH5py5SOFcWpblVjeganwyIR7gMKxPkuDCtdMd1rmTYdR15hj4VERjUWb3Tg1cjaPyuJwNSySGxSvvTYHStoyCLWFKKDkMcp41HTQW5zoFYG3aqgSyIXtCs9mADFRkvqkVlU3DJGoqG9nWNpMfc1h6sJ+2XLTrFK4+61GtgrW076OkXNESZelJbBmLHuUEG3p5kZwTvIF731a3tFeyCcQ+w7f/r2Aau2TJBk2xtuHwoI5Px0nqk2u7WuhvOGW92MWg9rOW96fjWyULstaUbcDWOVvsoWSKN6dgNDcxzAEmiysD1Pip7lUKZ5kChoXAWkXULbeTWGtCFsRZNSJeXVutaFBlsFx2B8Gu4BJH82gVNQSwuLQmC/pnCp5hERfN5Iu8s0ogtRaCt6oAi3QqltcgwxrZAX0vSI821dRatEWGhziM0EHIQMr0XKOU+cxcgJfjdJYSI+YqpXqc5zzIhMBSkEi4K3fA8nvahmdUE6TlcWd1zAcUFblUz3O2azlqWxDCshhSHS5cW/E1u1So5TWwHZ6YXB1cISZXu/dgxeS/oSZ18Ca7irQqdpiHKHEqW4EnxjZfjOK4rSBr0MeW1rYb6sWO0UXJoe4VDM+4rjoU7KWLErdFAsuKZ3AsVfFsFpJ12rwuMqjysV/UThZjX6zhysE2DUfdoD4RyMctxYzFgeNZIzl2Nxx1kl03lB+KRTRvrfyq99QHNbcaefcq46SZ41xzmcSi9YB+W4sBCEFWmVFogGkaxXkibM+ypRx+3VcrE6xloECMai1I7joUjEswlJ6SEf0xYdBYVqY7GSlO+6SXACtaWehDkHXcgOWITedeUYSgeVw7cFtrRSAA1RiZDLGFwFVd2zU66o9JDUlaKjlbw+/DDWEwQUVWG8T9qPzcawFUCjRMYwIU03YOxJTYvgTNLtdiSZCaxTsTBqvabWA0tbsnQVGs+ZasHgNEtTBn7FwM4V0KX2qOIwKIDJ56BSJ4EY7nuFarW0hkuPrxEwUdZmVkNI9wxrUHcpFivUUgUSYYXqVKCrE5m65CDCxaCMaKgyaPpQNFWFLOiomXHU1VyYaHYK4Q9ph5KlFa7JeM3WemC/bFkMVWKnjhqtphg5HmztGaaKYbeiujOXgqS7/9DhgXAOCjhp61BJFo68vPNiB8PRshGYaliAlbHZRS0X3a1umkBLpbYSkmpSqzPaCAIa6xDxwl25Ms0ViBS99ObPlydURlpagzVMSlmsXYJ2qzA5J65p2Zd0K9FXSKPdIN4mgBh8wDeoPnBGhnA37jYYqVeYQtShndWY2mI9qMKNvBCxlWYV/bxC11ZSkDAZ6AuPNZ5JEaMFG05F3vMIHPMJKbpyZaq/2HDujeoFJGZsODYv6KpAOjveJs/piDoh0XJMRIzoGtUnZGut+9RSlWNmwpKEKGk9OpFCnfcz2qGgqgcWxmMWGr/Q9E1J0fTYTkvqYEO3Igjc+NLJtOWgJEoN14+duHGyNtwfi/x6JZR0voo1C9L8i1dyf2xBo8bMRfUKtwj8EnHOYhCmMJvxjN4+nnI0nXOhPqE1QhNwPDScKRfpmgUZt58WIqxUahPmR4RfkkIIjV0Jw5SgZxHmLH4N9qqxhlKqUUr9nFLql5VSH1BK/Y1w+1ml1I8ppT4Sfp/JHvPXlFIfVUp9WCn11a/2Gp6ARQ9gjTgr70MF2XuhCl90pTAw9RWroaBzhtaWdK5I7a87/ZSVE9ZjIaAtEkV6HklEIZx42zjCLYpbMEYWkQDmQn2S2n8xnVj0VeJS0GocIVt25ZqyU5zUo5D6QapmEwrJ+UablJmku+GcomtL+jZIpDUWHdMKq9AnBnVSCFCssollSmnwg04tzroQyTUJ30VhLNYIkqI2OokVR+xItD4Muo3TlKcvuljDya3ciDQSiS1j1yiqaengjEXQyCfkZK375MwKbbnQnHBQL6mMpSoGmahUSIE3EOoq4wU+rr3wNPQKtZKowRuPnUnaY5aa4kRj5kIO42cW31iZRYkTsCZEeLE2EWcoQrShe6H5M8uQHoaN2rQKM9eopUmYFWVlPFzPDeqoRN8pWN2acPV4jygqXGjHcV9zs9sJXbhw3RlhvaqMTJxGbZXYtRCUrTiI1b6BwqR25v3avSQiK+D3ee//A+BLgK9RSn0F8G3AT3jvnwV+IvyPUuoLgW8Cvgj4GuDvKqVekSRbK89Os0pvDhuAJEE1OdKEr3qZzpt3JfO+Su2u1SCetrOGua04sXXiSYwisK0r13QUoiVhmyyyOLaT1MaT83Ps6iWzYsVBs2RS9tShFdU7TTsU9KHKXAfV6L4XBitlBMU4wh43nEX4Mr0ev1Rfh88hkM66ZZHGgodFEZSYwtMMEsaaVqFbCTOLesANMl8B8rzlbses6qi1pYycFhu1g94XY8Sg1pmpR8eQ8TVEvMAactKnomYeXcTPOHJXxv8NPhUm8+giDoHF54u1kvi6pXLsVUvOT064MJszPVhiJ1LUVZ20OIvSCgYkIE+FFVwWsuoCi3PAmfgC9ADFoUn6FmoanA5kEZ98zipECsCampkOk7SRCNhDKggnBGxoV+PkNZVV6IXmxvMHfPjORSZh5iXxmAxTHErEpHUvDkLZ1LUw2qfUgkLSG9t4VnsKu1PHD5/7tVd1Dl7sJH5n4ccDXw98Z7j9O4FvCH9/PfB93vuV9/4TwEeBL3+l1yiUZa9aUdZDygv9IGEywTFEWbTBaXortObtULAaClZWfjonjEjHfZP0CmN7LFeK7gPJKmwhRg0739zVSROy94Y901Iqy17ZcrYJoZ4fhXKl1SoDQofLRmjBMok7FXaeCMJR0WHosfvgK4/fGcZiIyTKehVk0OhGwhLf61hAl/kMhcxYDAIIUguDWspu2Ew6Lk6PiTLx8b3GYmTuTON9cWHmzN+xLRkH3uA0y1Z83Gna+9PdJpDITGoZ0jqW53CpUBkFYQxjQVmmSi3Tome3armwO6d8dM6wZyV1mBeBmk2iB1e7cfzdSQs5phu+ltHnMKiLmWtozUjrF/kmDfJ96eAIwvBcLCIHGZXT3wnhO1/psd60WdQMr/vpT57nueOzKRIdvGY+1Nzup2mCdbdoU+QgfBUScanUNvfYGvodqTso92uLHO6p5hB2/l8E3gz8z977dymlLnnvrwJ4768qpS6Gwx8DfjZ7+PPhts3n/FbgWwHOP1pxaXrEjcWUflXg+pJEGhAJUMJwTAQX9YMZ9QKUTwSkzisoOk6Gir0i6GWGb6j3Jsiee7T3YTTY0TvZNvJCXOxcgITajerZK1pu9TMenR5yq5sy7ysp6IXXjXT183mDXWYDUHEhh+6B0i6jjVMjKKqRnc6HCU+lRG5dmdHJxNQr5sXCGhXaZhEibTWq05iFvJdhz7LbrJgVHYW2LG1JGUJ4m+3i/UYbWNq5foOgpUgcGvlgFURiHb8WUeQAMyCpa6Xx78A92flRsGhzy9rVLa6Q8fkTK0jOqelC2ijCs03RsztdYQfD0OQEsGE2onayHpdZG7tVOHSqIThFavmpXuFLUIV8vp4AZVdIOzNiGgjfixVn4M0YSaTZGQc6bApJ6zK0PSMNYPQT5tjw3PPn8Y8pDuolK1ugTc/SVRzbhkvlIcdFw61uJoVlI5tlYSzGGIFSa6k9DFPF6qCg1go+WwVJ770FvkQpdQD8kFLqi1/h8G3bwym35b3/duDbAd7ymxr/pukNXpgcsGhr2kWBjiFx7fFhMaVJt+AFIztvnLTLX0T4CVzSLIx3LgNZ7KxYZceOPX8gXKxF2kXjrrZfLJiZXbRyMj2orQjJhMVQGstxW+MilsEp+Tu9kDg675TwOWQphp9YzHRIbTKlfWI6cr1OeXSMPPyg0bXFKdFfEPLZ0Jfv9DjA1cvrnmmWaYjJeY3Wo2p268osvTISHiakpEJnBcXWlcyCRGEERZXpsWprqpF/thAji/WWaBUBbGq8JI9tw8LVqV4U6w+ymDxnigWdG1XIemeYt4HOr7TsTOQ7jrTw3aqgNxXmyKT0Qi8VbuKCwBCJ9h8HqjVj1BDbnSvBM+hOagwALjxGWVCrED3o+PX60Qn44GA0CTIfUwuvY1dU4Q9LPl2eYTivuTiVdKrwihNbc7YouFQecVQ1dG5PRIqKQYbpCsE7uEJa4LZRLM9pdqc1+mTk67xXu69uhff+jlLqHUgt4WWl1OUQNVwGroXDngeuZA97HHjxlZ7XKM/l6g6Pze5w42RGq0PIpsAGXoM4+ebQ+NRozjZU5U+9nYiqq7Tg9HP3MTgNehxTxoELebYLiEn0UuTS/UClLGfNCReqKSe2Zo7noF7iQ3szUokn4tm8dRZl4JUaIblxoSvw04Fy0mN7k1IQU1jRYrCazgZREg/UWdXbQznp6V3QyVBSzMWOOS0gmhkIXqHWAwdl1KmwW0fi431j7WBdKq91JZ0yNDG+JgKd2ECvRv5OElVfbGPaEALF4mSku5/pFdYpju1kjQq/dSULV6UUceUK9oslZ8t5gF072rrgeFJzFAhaujCSPQyGSd2x06zod5bcmc5wS6ndxIIihcdbElJVudCS7McWc6xdKIdojIQhOdOrFNHpoKjuAiu0WUoEsrY7hglb5ZD5GasSSQxOHJC9VfGS2mdWdkyLPg0iLlzFvlnwSHXE0lZ0VkbXS2PpByOqWIXD9wpXe/odzXBmQvVrcA730q24ECIGlFIT4KuADwFvB745HPbNwNvC328HvkkpVSulngaeBX7ulU/C81h5i0v1MbvNKn3Iupcvgbww6VXCDcQ0I2paDlaYmBd9yXIoOeoa5kPF0opI68oV9JEnItYinFnrXPQB4x95C2L3ovOGme7YDbUHHZzNfr1kr24xYTS3jyI3ceMMqUAS1g2QXdUpkb3rJW2wg8GtjEQaXgBgQooaMP+Vw+z2UqsIHlFwICZV5aMUPQ6p0HtwJRT1QKEc86HGec2sWCWtUdFJ8NQhpI/4gnwmQhTFdepURFq4zZrDJv9nnjqYUHQEUh2nj10kVIosdABQRVBWrH3EusPSVsyHmuvdLreGGSe25sZqh7mtxkp/4IeM2BNrRTTneFnTlAOPnD+kOWjROz3s9aJfGuUK4wxF6UMUITWFqCviFZhOIjJlw3U6gO7kdrMKt/ey8E2nKOaKci4Tt7oPP4NsgDKAJ3/H4bv4mvak4FO3znB9uSPXcUgtQMYCHm3u0BQiEFRqR1nYUXA3UMj1u9Dtlfxa7F4ih8vAd4a6gwZ+wHv/w0qpdwI/oJT6FuBTwB8B8N5/QCn1A8CvAAPw50JaclfzwDk956nmBh+oL/OCltxNrxRq8OhO40KY5pRD8MEa7yM7j1Rrc12CFmSxBrDPTrmiCJFBqRwY0N6P/A9eCOPy9mbrSyIH4sILOGrHtCxcRa0HnFEsbSnPrR0vHO2NnQRPAtwA0nlQofUVdw0Lbhby9Jh++KjdqLDOo7RKMF1vkEJjp6EHX2gZSVZAY9GTAdeb8PxyIQ/7lloLLLcpeqIkXWksR0OTilxNcBYuDD8BI4w6KwSK3keVEJOlWo84clsDPWUFzNRSZoS1x4gi59FI/BIAXrNrWpxX3OxnuFCoW7mCm6sZ86HieFXTB2RoYWSxNIWoZi0Wmm5Vcmwcs7pjd9pyjLCIU4aqVG2xvR4JY5WkhtoGZ+CEbk/Z2GUgFTm9Fu0IF8r1tgndp7BqzFK+pxRxgESJoTahbEg3kYKnRM6GVVNyvRaUcMTnPF7dSt/bI80Ri0Fmf4w14hS1F0BU6RkmnvlFw+yj9+8gXtU5eO/fC/yWLbffBH7/XR7zN4G/ea8n8bGji0x1z4XiiJ1yhW4GXGnQK/GsviMw7Cq8kh9nQpFHS61FkGIOGxyHFGjHDsLKFgK2MQNNIbl3oURIV8vUTKrCm4D9n7uaA7OQIqOrmZoVB2bO7WEWqPCLVOzcK1ue9/uBc5CMWj5LIyKgQXkJNStQEysLOq6vXuF7kzARftCo1oBFNBAGhVlqbC27ml5pUfLes0JLN4jCN4j4DSHCmncV5xpFEZCesaVYmH7UzQz1mJhmifrVCCCLUQSwNm8R6w6lEratuzGIb4riRIuzLDHSaFRPq0qsUgmO3VKiUewXS/oAl59bOd9p0XHci2OYL2tWx7WE1vuirzmtepHO6wpWq5LSiGqWn6y4009lxqG0FIGXsTdCnkNjsbWBOwVFB6Yl8W0oJ2lDP5HPeZh62O/RlQ2SfQFoFrgsCZfC4nACnaY4lJmYYkGiBoSQhhQgtQrFcFTS7RYcrhrmfUU7KXlycoNd3bKi5GJ5zK1qyuB2qI1loYUEV6CZEgF1B5rh7Oxel2OyBwIhWd+UCvRMr3h0csivzDoWdYVfCO+B7sEVCl+GFCOQnDgtFFjC6edTu5PA/9dDYhy2oe14TE1TDHRVpBPzrBip6bXyoMRJ3B5mAUqtA6xaooep7hIKswqw7NYK7XtRDwyrgtTP0qTiZOoyeAUWaVvGCCN2JKITibeHFEQIRTy6VamnjgqhaKgt2E4LtNpKWuFreR47GPqQfpXBIa7cKEi8bd7EhWJk3G8iEjXWHqQ1XNGrTiIRRtyEUT6kCptphqAl1zoiXicyWkGmOlFQxwT2rkHqPvE8lWO/WLK0FUeDzFjUZmBS9MxNJbySCjguOQ4pwqzuKAuh73NWs1yJUzE6ULq1ZaJ98066GyowPKvKCn+DE7CCX8nidZWnP3CovQ5TOKb1wLTuEoq2KSSVk9cR7oVCWeYHNfO+orOGk2XNyVGDPiqobmtMK5cNAxgvr+OXmlVbMq07Fl3JJ9qznKkf5XfufyS1ii83R5z0NdaJvmg/hK6FktRimMLqbMZ3eo/2QDgHtVjxNz71n/CXrvwIT9S3ePzgDh++PcHNdcjTlDDplCJu6kHyQxxoYQoWwZDYsRkdhEPwH6mKHujE2kEUlvaqmsb0ibuw0JYSKwIsTnr7UfzlyDZcLI4TRLtQAVCkXCKCMcZDZenbED7EIRwfKtahgo1GIoqAVUiRQ+GkfZbrIAAoj241ZiVsP2m4J5hbCQGM6sV5uGkQ3o3sU6GT0HtDbyV1sGpMq0b9UUkT0nxKhmCMFhd368vEFCV1Ap8eEz/v1JGIqYVy2AxjkkcNlnGuYle3kuo5jVVSh5FzUzil2SuWOBSroPXQmD6gJS1tZXErgU4v5g1dV1AUooDdW2Gwdk4xqfvEluUHiTbpdRpiiwxSlA5XaVwJ7oJFzQbqSc9uObDTrJIydm2G9LnVZkht3UoPCe5fGcuZZkFnDd2sYL4v6mTHy5r57QnFjZLyRGFaSVFUA/1JyaKpqMueZVvygZuP8Kbpdc4UEsXumpbGDLS2pCkG2kLk+uJsSD/zLC68Ig5xqz0QzsE7x8tvfRr3/9E8Wd3gyuwOn9o/Qzs3FEsTGJHAdyqsM7l8vRcGXhmF1DjlspkMjY/DWS6vT6igkCz94d4ZITB1hlnRxYcmB7FwFQdmkdCBIDtghPpG1F5rZTJTa4fTiqK2Qo2SagmEHT8oJ01tyGtJYKgoraZLh10WJLEbLS0wyVelgOmKkFaFh+OCIwn4B4g1G6EmK4yc73yo0cozqeZMIrM0cZR6rB2s/a1cwnHEY3PJuxIZp14X3Y0LfnQSPWP0YFFr9PUwtjNBlL0b1ePCNKHxDu19UjWbGvmurrtd0ZZUnknR05VGWsCRaAWJnERgV27QJkRTxRjB+BjdhVkXrBmlCKcOd9DT7HRUwSE0hUQrTYAzAxnWRs45XhuVFtj6EDQwixBxdk5QjruVIG/b3TnHlypu395Bv1xT3VHyHbaaxVFDV8sGtFhV/Nztp/h95z+ERqLus9WCk6EWxi5jhU9EGxnxLz3Li/cChl63B8I5AJz9F+/lv/8/fw1/8Ykf5c3Ta3x49yKfnkxwpaboAilKEYpyYaYAwKvYQxojiGhaj5G8yi5ugD6Qhiy0LOrODQxes1usmEByEIfDNGk62nBR75olz3dnMbik/TDvpVpeGBdmKjxFbbHa4xfhYw71Sd+Ec4zTgp7EG6B0mOCLZLGEwpcdC1kJkOMJgzZepNEqJ8XbxoUIQtpZEX9facvtbsKF5iSRuUZBGXnL8vmUeqB3RQjjbUolhKV6nbMhUtn33lDh0m6Z62daAvBM2ijYUMGL7c2YUqR0wwsnfxMGsFrKFI1EzIPGMzUde8WSW/0sdSYqYylKSx92fdGZCBoXoViptKBTu85gjPAzACKHN4yfsxqkJex2PWfOnbA/aamNRAGNEcdQa+GgSGxlIWKK11psEfdeMygXgGZC2FIGvEyhHKswJ7RXl+zWHbf3JhzfmcKRcE24k4K+16jK0mnD1eM9Pji5zG+cXeVGv8tBueAlvUuhS6rCCtlx4YQ+TyN8H/dpD4xzcPM5d/7Rb8b+15qn62s8tXeTq7N9bG1kmKUHv5JCpFISOUBI050oFmlDSCUi4+9IXa71uPt5rxK0thsMS1UmUR2QfHtCDxpObM2xa0TNO+TZe7plalYcDc24m8aiE8LOA8JRqIwgH+lEfITSjmlDhEiHFEMhFWu/LIIUXth1I1pUITBfE8oSpWhy+jrOboALjkd30jqzJRSVpQx6mp0TTECtRJC4tQ0rV7BjVpQZ1X+crZDFPSpkxfvMWv2hSHWBBnsKJQniWCqcAF9zIZy7pBkEsplG90z9ivmwlwhoYmelVI4z5SI4jBGhqgPBCyqoY0c28NDqdTZEmhhcQLDGL0+vFJE5z0499lzPmbMnPLJ7zOXJUSqCTkzPzHSnajFJ+DYjv7EBROYCkGlzQrjSAzOEUb1QMkLfFD1npkuOz1UiaXizRrcG12ioRXH8kydn+Y2zq8nJX2xOEteFMYEcKMDzbb29SPxK9sA4B4DzP/lJfumvPMVvmTzHW2bX+ND+JW7cqqVXP0i1WK6jAIaKGFXjQQeJ9ZBihM57II4J3jwUJsU5KMCsSZ+tdHGKJ8Jox+1hxuXyTuKXBLhQHHOj300RSFwMsTJtjKQ9LnTEfCklbl0K2lNBoCMbzQduQhWcgR5k90q4hiiX5wmEszJogyawGmkZTfZjGjLMwsSe8iyHUWi4yXAOy4ARyHEFJpHryjaa2psqskWNgLE4lNX4PkUPuY2M1R7jfZLUy0e38yJl7FrEBdcEiUKtSvBSR0r4B91xrpzjvOakrylj5Kg9RSkOIilcq8Ck1Wm5ZpYGu5IaA4MA1QS7oOhnnvqpYx49OGJS9OyVLQflgt2ypXNFqDfZMULIUqKcdGjlC7SPuA1h1Y4tyZUr1mo9dSUOu9CWqTMMleb8RHGzWXF9tsPi6g56qeiXJcZ4rp3s8On9s1xpbnF7mHG2nPNSsUs7SEdGa4+Lszvl6zhyABheeJG3/quv4h/+0b/H5fIOT+zd5sbeLu5I41vpN5tWjYKzEXFIaHF6LRReOrSoQ3SgjSgOKaVC3QEiC49SisEaBuPobcgJw8RbdDHaevpC5NtcaHHu6pZd07JyBcd9k96D96MzUgqRpPPAoCknfQBv6QTkEpGUQGgaZiIiCi+ShiSUtRJgjTfgdiyqtoIc9WFYyIOtWQuN0SGs9orOGSaFFFNhxBto5QQLoCp2lRNi3XB7vvvFYagIfY6AKKP8GlNUXq/YHMqKUGt8rEtIDSK+3mbnJDqPOJzVUkoHJVPgmpoVs6Jip1wx72uqYmBBnYbf/CCUbCpwJ6SZloAmZRD8iQpIx9UFy6VnbvD03i0pUKsRHbrJNpYg3RsWI6+SkKKFTSUWezWyu0c8B8DCVewVop+ysvKYQjl2yxUXJie8ON3nxZcPUCclqxB1vv/Oozx26Y7wkwDn6gXHfUNpHKawDFqWuNevc+cA8Ow/fJl3/sFn+Y3NCzw5vcWHdy+yaCrKE5XAJ6aNhTzZsX3pBBlIoBpxKuTvSO8/Yg/CoJYxDqpBxqqdWpsJcAjxarExqHJj2OFyeUdwETgqZblYHnFsG264gvlKoL3C3zg6B5DFaaYddTXQ9YV4ECVMxFtZgcPOrzxpyi/Bp7W00XQzSDh8XASyVJWGfGKHw0WuSYKQjVecqRfrVHCMu97KFZSqDJyS7RpRb0Q4Rscw1gl86uYI9qGnVCOjdZ5ijJ2LmDrooGURkZPjbItECHK/Q6cxcuNd2IWHlGY4r5nqjrPVgjvdNGmGuFVWoY8OtnD43kh9wYZalhFSX+UUq0d73vzMS+yHVne0ielTKzjK6sEY8cSIKnKFgNRjat1LJyHTUYmfXx/ayZEGLrKVlcpyxISVK6j1wExbzpZzHmmOuDg95v0vXqafV/S94dZyyseXF3i0vsPCVcyKlUzaakdVDfRlgWs12dd9z3b/JczPstmPfJy3/quvYk+3PNXc5PzOHDdxuIKAaReoquoDrDUPveOI98aPc1o0CoIgiUIQdEUxph3WqbXJzs4WDGFSsXeG2/00kcBEurlz5oSzxRyQAme7KnFOJWn3PqhfASFkFMcUw12dUeG5QZB5QjUWCpdKUHZxus+VHjuV8WM3L+GoFCKRLsNUjOMO44IICzSShCRuxrD4BcKcT6OacSEwqnHHbkKEO+d1B0BwD75ItZ3N9GKT9yFyRsp9OkUiNitSxkVYZWG8VsKruGuWY+oR2ss7xYoqCAxJUTdGCoQo0idnqnqFXkFxIuzdqytdcgyVtoKfMH1oS7pTEcNUdwl2nkSIdZ9kGuNsig6t2SYcG+9rwnGbjvpMOedCdcyF6pi9Ypne98T0PD27ye995iNcefwmdd0zWM1Hj8+PLVQtaOBSC7mPcDz4tbb3vdoDFzkAPPsd1/mZr3uWZ+prvHnvBp/aOYurDF4pwa8r8IUU9JwSxqNQwxprEDFPDy1ElMrYqwUYVQRNwxjqWy+aDr0yaQx8/IgqbvQ7nC1O0rBWo3sulYd8SD8CCKOwHQza2LGlGke0kVaXTJGGuoeX6CY6sdRaMUI+kpxDLW1PoVcHPegRiy+lE4HtBo2FhL/qpJU5DAZbDJTGMjH9OIEZQ13l134DIUKSi7IK/JC5Gtk2pukYPXSsmDGMeInwWURh20gnly+KrcS/4fUqGc/CKvluQOoQWjmmehUk6WDXtOyWLft1y81yhqNMHR+8wrWBrs2H2Yaw1oeZo3hizm+8cJOz9UKYp8yQxJNjMbHWwzjzgU9dnU2h5gjYSsXd4GBrFSHq4+xKjCJyHRUgEbuApH8rX4ij1LCnWn77hef49M4ZXl7sps1rp1hxMtScqRbcbGcUxlJWA0NVwOp13MrMzf7qx/hHP/r7+Dvf+B08ObnJhXPH3Hi5DuSdHhPl4xJ6SAmAxZNqEGhBDsYjXJK8dmkn1WH3diFqGKyEyUp5Sq2SXkJUfD6xdeJQtF6hvVycF+vjUZfAwWBFQUopWaxVPaQ6hEcilaocsEbTdQZlCDMBKo32piJSIJj1KyEnjCQlEddvZ14Gh+KkZmhxuhKU9YJzGOS5Sy21lIWtUggsRcKsGKhc4m6EcGGHNGpzAefOIprUY5Zo1Sd1rfH40ylUdDSblsvuAVRAo7vEQpWndwdmwVxJVLdXtOxVLbNJx51FlaLJmKoRGaDC2LQ9Z7n02G2e2r/FxPTsFuNgXawlGOXow7WQGLoDjV6a6kVYkOKMSIwyNgWRyjCc0XtDH97/NrKcnNdUh3Qqwtr7oBFyZXKbi/Ux11a7rFzJGSUkRDtmxazsWA2FgMIKJ+3u+7QH0jngPY/+O8v862sul3d4dOeQl88cYG9XlECxzByAzhwEJG8gtQePJxQhHTKT4RQqYSMI+WlQT7IaozWFloshjmE7JWH08dBwOEw5Z05kzjDkh5erO5ybLVgu6lTzsAH5WM064Tj0Y23De1GAdk5TBCCOLZ3ob0bS2Py9LE0ib1GQdkNbB4m8OmyBymOPKomghsiETKiF+CRnP7f12oUu72897CyVpQp6l1UIheNCtl6tAZZy670JDqJLuO5NItp1h+EwXuoKmyplsWMhtwckpy5oXZmNeksnY6bFkfWl4ahq2J+0LGYV/aqQekwvlPJRGcyf7ZjtL3nmzC0enR4yCZD43CHEzwEQlCaj88y7DKUfazg5+1XecbFxt9r4jOP7yqOJTV2VOKpZIvWrhatSp2Nieh5r7hBlHOVzUUyLjiPdyABaNbAqXqcIyW02+7H389/+ytfx337R23h29zof3HuEYVbiDxWq9eIgdOA5yCOIkJ8rBGpN4eQ3onyk0XhtwxyGAKWkF64YlEYHB2G0E12EuFKDP3mp2+Px6iaAzALongOz4MrOba4f77A4qcepzNBJGJwWsdpApR71LJJ8fB8Uk4yIuWI1UQ8BL+CcyG8RR0e8DmlUjJBMkMHTPoxr65RaKaA0gu9f2SJQ0+skoCuTj+aUg4iLtPOGiohJ0KG7sO4c8k7Dwtf0fr7GoTEW6dbrDglJmboYshDiqHf+3DoUIiOcOlqlLL2X8H7fLHi0PuR41nDU1px4GX8XnIMUqg/Oznny4DaXJkcclJLT75o2DH0NmSM8nfaMrd7Y4h1kn1rDdoyLe1tUlHNYxJRDejY6zY9EZaNt6VajJTU8GWoGJ5uZdePQYOcrJqYX0Zswwt7X2535K9kD6xzcYsGZt+6y+B9rnqhv8sTZ23xkf4q9aSgWMhNvltnVj+AclApheci9GYKDCEUAH1qc4MhZu5UK6YU19NonebLBawpsQAfC8dBwy+5woTgW0hNv2NUtO0XH+Z05L64KBleI0wrjsyB09gkwFbkNUXSdZugK6b1vI4ZRPrQ0Fbbxkk50AgaLgjeuM0JHX4YFFqnKkoMSxqwYBQ3OsFQ5tsGPKuf4U2GuQ9N5aMLF3BOVsSP6MQKAXCrWHruG0izW5jS28T3Ihy/RA4QWYGqfBrr7nF8CS79RxJPnktsaLYxdF5tj9HnP1cUeR23DvK2oy4FLu8c8uXOLR+ojdkyb8WQO6TeQoPIjhb5Kt+uNqKJKkMoALceuOYWx8FusRRb5/ZFQJ77/XEk0d0xaCa1hrQYoYGGr7Nwk2tVOipeTopc5GmPpys8j5wAw+dH38N/86a/j//XFP8Kze9f56P4FhumE6hD0IGQmaHBGhYo+srtmRTCPXx8+UmHOG4j1h9ix8F5J29tqBhMiCO8YnKHQNpHK3Oh3OWuEc7d1JXum5VJ1xMfMeXZ3ltyxM/wgqE1jxpyzCyxRWosyd9cVDG0pFHCOkSUZApbDizcL+XEi39UhOppYeSyImKsKx6xCXdIpUA7voBtGzcmlLQNFvU8OImEZcjBS7O/HcNXFSEI4HnLLHYRFMfcVjeuZBor5Mewe6z3hCxILNY3E+ZA5iPg/hChhQ9A3yhxWgMVzYBacL0+o9cCl+ijocGp2zIq9QtCtu7pdE4zJ368MkA0SIflwe7iemmzZbn4GJnyWvR9H8Ef6vfGajJ9TrFmsv9+QvuQOWo1CQ5ZRkbxmoCxEId5ZlTpMMRWZFh2LvqIMcOr7tQfaOfi+Y+8Hd+GL4ZH6kIvnjrgxa7CVolh6dA8FQqjhjQpQRCDuVvn1F7oYXqkIdA3wWo0OPJRR5NRonaTWSm1xjMVJ7T1HQyMqUHGIyGvOFnNm5YqTsmJvd8mhnVJUds3xwIi16HsZBvKRrzAWWXUW9YQpTh9FX0NU4QrwtUwOemGtCS1RSUeig0F59KDoMyFY5xUDmqUtJb82I5AnV6uKF7POLtjWlQkaLBgEn5CSuUWSmLmvKL1NxbU4yr1pufMYz1NqNnHBxDBcK8eMjnk4Ll/csijk/M8Xx+yb5aldPvGCauEF7ULJ9dQ5KdE7lZ3cJbX3eF9udiMiGouWcnulBjqKUzWH+Jbje9xMQfJUbazHZNgT5TCQ0sOTUEsS1mppXZfGYqyhMKcjllezB9o5AJz9qef4wT/7pfzOsx/jsZ1DXrpwBnu9wh8rlHOYFRRFJmef1SC8D47AA9GZq6AloHzoYEgkkTsI6zTWSZQQR4IrYyW/U0K3dmIbdkybevKXyjtMix4F1OXAdHeVwDjRBIAV0pfBYFeikaDiruJCxKB9mgNI0UOiU2d0IHFwS0l9Iu7ssSArjEKkHndrC3Rgay6U404/wXlNUwlZrpT5fFqYcUeOGARA1KsVlNnu3XsheY0LKJHF+oLWFzRqtY6azD8XdZp3Mtr6zqogTIPGac5IOBNfPz53pSzoFSK5wojAxNGEBZkXVPWY8adzM7jwGWQ1kg1cx3j7em1hdJhxildasWPHLEZf2W7us8JlZlGeUFKPoCgWPpPeFZShgLpfCLht5YpAVqRS27rUjsLcf+TwwIGgNm24+hIf+6FnWbiKadFRTnv63dCq89KrLkKBsmhHkFRcSCrm8OFv4VhUMqEXAFIRHJVec5Bx7lg8tNm0nUO4KG8NswQWiozMT0xuURpJPyZVz/6kDWnLuBikEBqo4IIgDZYsvA4ru5PePYHoVPcEFuPgACs30pkFOjo3yO+og6EHGfmNziHyI0RrbcmdfoINCMN9s6TUw9rwlUOfqhXkFrsXuXWBEm5cvAEwpXxgi8ocBSN79Xjb+P/mYrEBLdkoSVdi+N36ikM749aww027w8LJANJMdwGXEIlmfOBfGIFIje4Df2aoCYX3Ex1R7jBi9ALrnQlBjrq1c9/8X4e2sDipjkoNVOFxsfazOfEqrc6Ntmb4e1MYOgLBdHiPhRLQm9GOqvg8dA4Aj3/fx/nZW0+zV6xENzKqF2uF7j2m8xQrj1l6zIr0EycToypUkr0fRDTHB+YfF+TQvR9h1sMgziGiJgenGbxJ/x8NTRg6Kmi9IAOfam4kkg+jHbOyY69ZJYfg3KhUtdayDCAuFKhBj6PDYVGrIdYP5FjXhGGrxEIdW3XhPer4XEJ6SmidLvvAhVB0AhO3Ba0tuNNP0aFFlyMWc4sXahxL7nM4cLgYY/QApGJt68s1wE88ftNBjMNW49/j/dmCwCVswAjnlsUW6x0RjHXHTpmHcfQ4m5ET3UbUZ6N6ZnqV3e7T+48dkjx9yslvorOI55wjKUcGb7cWLSUHxeiE5VyG5CDK0DWJkdvm62ziQOLr7ZqWqenS51tpEUveTIXuxV4XzmG4+hKfftvTPFIf8vjZO3gji1/Ugz1m5TCdx/QhiggRhGmDg+hBD2MkEWnbckZr7yIGQnZ6azWrQUaBe2vWcAqD1yyDFqdMNpbMXc2eXnJxcgyQFLdnpVCiA1kLk+QM4i5PKCCqXmGCArTAe2UGIG3OnhH05JVEGg4hnY3djsCUrAdwBVDIaPpxWzPEyne2AJdWhHPjIswv/rmr1xZKfjHGHdaGQm20uD/2AdXXemH93lZvsHEIaeNiX2erHtGGkVMDROxGFvX6YmwCiU2sjbTh+4nHVOEnLs6YUsz0iplepSlQIB0bH9eoEWFaqXVB4c2/o8WWZR6FRItRTfyJDiI6iTz6GNuf43eX1ySiY9o3SyYmDs/58F1/njoHgMvvOKR1Jc/s3qB/fIXXUJ44YafuPcXSUSylBqH7EEEEUlDTBvrvEEGkKMJK9BArvcC4eBG9gyGI1jiic5Cdcz5UHNoJwsZcCHJSWR5v7lAZuxa+T8ueWdOtpRg+4v5hFFjtg2pzQPDhSexOysp7cQVjKqF9mKUIqckgjsS0SgbVwgQnnabvDauuoLeGxVCJUEqQUYvCP8LJMKYVrStTWJujJqPluT6MOXmTMUxFrofe6zVnkptD3dPFu+ksDJ4DvaTRXVjE45xDmncI7cH4/lLXJLUN3biDZ9FIFDKK95cBQj7WJtZ3/W0/6XPBZ+c0Rjvx84pOI0YTMd3I51PyEflt7dD4OiDOKHZrKi1drPLzseYQzb/nV/ied30FTza3ePyR25w8HirBt1cU8wHTOopWUozoIHTQERCtgDDV2QcHERdYJtTrs9B/xD3IRT0EAFN0EoPX3Ox2ACkMndiG1pc81dzgoFkKriDr79fFwGyyQmvhF6CX1EEvNXo56hfEbkScj4jXQVybPqpvQ8J+xfeTOxLdi2OIDNSR08AGxxZp7QrlqHTs7ZvUUhsr6LJwG9WlRa+VXxvWiksMGPP4sPuO05oygrzpGKLlu24uobeZt1fYoIMhtQqDT23JuLBkcXcBLOTTOZUBLNV56VBEfYx8McfbqizCiI4iToPmTiX/uZvF5481DBjTlTFdcMlh5FFEdHg530a0bfD1/P4L1bGQIGfDdvdjD3y3Ipn3PPtdPYdfPuGLz17l31w5z+KTBfVNKG4v0NMKr+ow0RiTeE8KA0IXw9U+0zwM60yH6U0luwIBVu1DSqEHqfgOymMC05DznnlILaKk3B075UJxxBPT23QhFRm8INi8VzTlgFawWspAUE7kksRUAmgqRjqxkaGtYph41gr7noSkVMOow4jPEJQg6tszleT1Bqc56hrO1gtmRcdUd0nQZwEBJegTkcvKlTKmvKU4ua0NCOPkZkRVChmMPG+JOM5NvYvoHDZjFLl9E5YcOCrQVASOCeUxjLR2d9tpxyJmmaKAqPYNp3fM0XGMuI/Yls1bqWWAmycLH1XuOGTRr7dg5ViX7jfEwbbQDg2XcnTcMvk6pnr55yLnYemVoHfPlnNurmanhsPuxV43kQOAftf7+YGf/2082dzkySevM39U0Z2RXNLcOqG62WJC9KAHHxSzfJZeIDRgUZAkphi9KFfH9CInbLE2EtGOSLkhpBZLW3I4TNOXsnAVC1fzSH3I2XqROheRgk4hTqesZafOuykCmAyCJmEcHUgQald6XBURhfE3UmNwJEVn1QdshCFFIMUiyOUhSE3rZAy9cyaoVQ9pgKh3RagRjP30qPolYKJC0ongKLpQV4gLIF8waWdD03lDt1nkvEsNIi+GrrFTRSh2HOnOaiEz1aWiYi7ll9cYYo0gt1gXibY1RdhAjcYooMQGZXC7lmaM5z6sFR2jJYe0JQWJ7ydPPUYHNrJsbY6Qx+8ppoFNSMEuVsfsFKvP88gB8MPAm7+r5/jLG37Dwct88s3nWX66oLpdU7Qd5tYRtQavpvjAWuyV7FBehUWjIUKuXbhdKTkuieagBWDkBV49DIbOGMqQo2vvGbxBO8/1bof9Qqbheie1hzPFnPO1ICh7K4vQ+YCvCJ0QJla4H1ZIaqOCQyg8Joi1+sBhoaykBz7CHuIH0quxVYtECLpHdpmJgMP0gEgKdmE03UuRajUUtEUANalM0CfstpF0NiL+NCObdGSBivm3gJVkpxNsgh25HwKculdSvG2UpQ8hfQ6tzi0qb68V99Q4/ShDTGMEEesIsUZwN3OMrNniDIuUJsTbcieQC+2sgZHCfSNWwian6NApCoh0e+m9qGyoLCz4ka+C9HnF54+gLpnrdymCEKcT5zAyZKkfKfTESUpL+WJzzMlQ3/VzuZu9riIHAPOz7+efvve3caW5zSOP3uboaU13psZNharN3Diiub6knDuJGqKmYaxBdGFnjmQxPVLMCwIyPrQ1hWKO1OLshoLejfiH2Lk47htObJMusJUTJqVz5ZzdsqUu1gtLQKoZ+NJhZw5be5Fc86SFHusNyopDi7gOvJyrWmnpZITURHmSWjMRDKbBLJUUZ5eavgv6n04QoABdxmPo/LjrL6wAveKiyPPadSTl6VQjMkRHk/ZisTUtyU0q9GPxLn98mghFb9w2pi8xp09RwsZuHh1B/D8vLEaLGA0YOwAxncjblae7K36tPpE7qtzJxfPaFmnE//NoJdUjsggidjJi9JA7ySg4PJLaavaLJTuZqvy92usqcgCJHs7+dM2Z3zrnC85c4x2PnWP+SEl1WKHaFartMNcPaZQC1bCmaqs8tPHiFDcslfLAs6LBD7KqfGUTx2OMHtquxNQeozXKebQyDN5xu58yraV1FNub+8WCo6FhWnQMTrP0JQpYdqWAlVx2ToVnmJBargQnoTs5zX7HBSo8lfAOKbMIayhlAbEg6RSDFqejOzALlcbE806K9SpxFURVcutFESteYNPAKZmihrBLbauaj3mxWdv9Qbofjemp/Pa+e9S2gNPTiFo5Yla1rSVaYROqM0U0jBqccQePr6M5HZbLcWE3Tw5kfP3xPMdZi8378sfEqOtUfWFzTw4OIB4XVcHW0rMsgpDXFzKe9FkpmafogRPbsLA158vj9B4vNsen3uur2esucgC4+IMf4l9e/RLeNL3OhSu3WTyi6A5qqCswGqzDXD+kvt5Ke7Pzo8Lx4KX+0IEJGAgVOxh96F64+LP+8fTW0FlRjEpfpFccDyPBrCyskkb1nCvnTEzPTrmiNkNqJ/leSytVMXYetNQUXCmCNVITkYEr13hU6G7Ew30sqGrWipTSnZHR5Mh0hJeCputGEFdlBEIdi45JbwHPKtQcIvpzm1iNvPcRc7AtnB+fW6dFK7v8CILSGUIyPlcemcDYyTjVpgw7ZqMGpqFDkuMX8t+a7bWDvDMRXyPfzXPWq7xzkv/e7Kjk98dOSR7JbOtu5J2TdL4baY6csw8R0pC9xkg/J48Xzocb/S5TLdD1MyH1vR97XToHe/s2t7/3cWrd80XnXmL+5MDiYoHdrfFVKdv9YCmuH9FEB7Hyo/T54NEJSZmBpKJ83KASY7GzkX9BUow+AKNiobFzBfNBKOTyOYDeF+wXC/ZL4SNsjFCPCW2dSvMSm2vKBeEeFzQp+j3ZpfQwphkqqnQPYzph2pAuWSUtzArpWsSW6AAMotcBMCl7KjOk9CgP+VtXJi1N51UoRN6dLCSF+hspQ9QYjaH6NqRkPtC0aXExj87BpfZlbjEdaZRlqldraUVeEIwFwsggteks8hQh/r4bpkDuX3c42xxI3jkpN9Kd3Gmlz2wjyoipUgR1pXPdKFLGc8md58JVHLuGqVm94vd3N3tdOgeQ6OFfvfibeKy5w9krdyR62K/wMXqIDuLaEc21FWWIINQw4gB0hoFIv4MYre9l9kKKiCRYdR8jh4h7CIvrRrczzu2HPN7gOFMsAt5dcO4+H92NrcdgvhSBGl8IPHyYefxE4OL5d6ucCnwN4mDMUoay9GrklVReHJ9y4mxMJyjKoTcitOo0nRXk4sqNoj69NyxtyRDSIxBnMVbCx4Lc3RZOF4hi566m9WWKGFpX0SGOokefGriKzmlEOg4yi6FGTEOOb8j/1kiLtAqApeSItmAYXrE7wHpNIbdtk5ObDi+30bGNTiIBrLJoIq9N3K0WkZ9nBE1FB3GKJyPrZJzY5lQN517tdesc7O3bXPu3j3K+POY3nrvGyTMD7bkCNy3xZQE6ln9DBHFtlUBSMb2QSn5wEGlQixRBxOEsH4azQKKHtpfi5OBl9mJwsqBu91NcwPb3XvL4XdNytppTBGYp71UQogltBxcmLBnbmngBO9mJCw5EpO7szIkDMaOzKBaKol2PLHzhx8giXmMeVKcYeqmdLPuS5VAyOCMM2wHU1HvD4I3oJvjxvdh0TAR2iVOJtYHWV2HGRHg4e1+wcFVglC7GYSxXsgg/OWoSSE4g/sSIIC5+nTmBUjkaZRHX49cGu/JWZv6zrVaxDQORzzbkNG9rlG/5tZhFRa90+2bqkUcTeZqRRxb5bduKlJEHVKuxWBrTi0b3LGzF4TBNHbX7sdetcwB48odv8+n2LE9MbnFw+YjjK5p+L0YPJjkIFSKIycst5cJjej+mFx0USynapfpDXKSDUMaP6cXYiuyGItUeBi/V/zv9dI3uvQ/CK2eKBbuh19z3RiZDo3CuHyHdalDoVklRUiEU6nA6uqiEnj5CqmP6kFPax41MhYlPwU8o/KLAWkU3SIu1c2NIsrQl86FOCzaqYUX4eO8NC1tvjRoSpX+AVLd+dBzy+DC0hQ4M1YJ76CNWYS08FweQOwWQNKRSjkq55BQ2W6EaT4VQzI0tvSFFA3kakCMTo20uZoMPiMz1v+9mmxFETt6S/4zPKYs5okoblZH73iUyy18jj3g2Z2Pq0La9PUzfWGkFgHvvh3jbT30558sT3nT2BotHHcsLpUQPhSHISQGgrKO4fkxzo0tEMcoiXYcwvRinOOPQU5q/GNZ5H71XrAYBRg1pclNzPNQsgpZi3Gljget8fSKEtb1GLTUqMDj5wNeQdvoAanKlCNeoGGFE0Kf2UHpJLQK60jbgaiGctROfGLESCjTI6JlWhrJ8OOfIyhSnLZe2YmlLOieygHlU0bqSY9tIsdLJT+tKWhcl9sZ2besrTmyTLsgkogtpjFsiiJrWm+QgYLzwNxd9pLTPd+P8mJzPoY6DUti1xXi3cettlu/S+XltOpnTzmDjvBlh6Pl9+fPmdQmpnfSUGdfDZvFyzYEpf8qJbH4+gzNc6/Ze8f1us3t2Dkopo5T6JaXUD4f/zyqlfkwp9ZHw+0x27F9TSn1UKfVhpdRX3/dZ3at5z7P/+A7Pted4084NmsdOWFzS9AehMBmAUNFJqMFSvnRIc7OnaJ04hIEwvDUWKFWURxsUDIH3weaj3QEYNZi16GFweq32ENmptXLsG+k1eyftSN3KXIUaFOgMFu1CB6J2qWuSC5Ko2qYUQdkxYrBVVqcIKleukMKkreW4iLx0TqWhKxfOfW4r5kOVRHxitDA4k9Km46HhZKiF/dgXrFzBKkB588LloZ1IlyO8/0Z1RA6FSI4TUZP5jpanDvEn51i4W+ju1nZ7Ob7MHEG0RvXJSdzN4q676UhyfMPm/ac6CptdkSwdydOyPFLadBCVsszCPMvIRrWOt0iPSY9dH+e22fU3t59dENR/AXww+//bgJ/w3j8L/ET4H6XUFwLfBHwR8DXA31VK3X9Mc4/m3v8h/vU7voyzxZwvuPgy8yuO1b4Zo4fclEJZR3X1iObmMHYwAkgqQqxjByO1NwMuwdlRGCcWJ4cAjOqt5OpLW3Inqz1EIo9SD1xuDjGVkwUdZiCkCKrWd3rtUcE5EIFPcY5CgZobdBiyQku70+swzRkEfbxBuh37nn5POh8Rfu0DbdzgNK0tEtahtQVH3YTFUJIPSUWCm5UtUrrUO6lRaDytL1i4Os1hLKxMqE6NjEDfnUFpnJGQ/8faQbTeh2G3cD6b4+Ex9YgmHItjWhJvi6aVWwvpN6OAbR2LbVFCFdOezGnkx2wrVI5OYrujy9OM/DMqiWPjQ0J4xvcyfm6nZywiYKrW6/ql92r35ByUUo8DfxB4a3bz1wPfGf7+TuAbstu/z3u/8t5/Avgo8OX3fWb3YZf/vRRmnpjdpnxizvyywc7Ksa0ZzGuBSKt+oH7pmOb2QLEaU4xYg4hYgagYpaJUnVO4yO3gVJi70Kk42YVx6FvdlJOhTvoK8QKZmJ79vbnIk8VzUgHw1AYn4cHuSMEypRTRrMK3JpyXdC0izHrYs7gdi68dyikRbGl8iCgkqoizFpEuv4tRT4wUQhTR2pL5UOFCRHEySLrRB3DUfKgTDsKhUncmkq1MzYqzxQkHZpFacJsdgE0QklHriyuPBnJ6+/xnsx6RO468OLm+QLO6xgYKc9sxuW06gE2ncrcBr03LHcg2R5EzTsUoItYjEsIynYPbcJZZtJOiHE+h779bca8Iyf8J+CvAbnbbJe/9VQDv/VWl1MVw+2PAz2bHPR9u+6zZzo++j+/9U1/GH3vTL/LE2dt87MoOOy+WFCclftWh+uhtJb3wGlTb0bxwjDN7eGUS8ay24DvphsY5DKeCY1EapxxKaZR2gWJOIodIy+W8orUFt/qZyK17JSIo1jDVHQeTllsTi1oV0klQ6cxkEtPD4MCvDLoZoAwRRJwkDePZOmAcZMDKj3qITiIIVwQMxEqNBcsidGJWhrYVxKZRXrgllU+pkfeK1pZ0VqjUOyuU59HmYTefGYHk5gIvsU2XOwUY23Q6IAFjdKDVOEyU27adbnMBb052JjaqbPE6Rk2MtLAjahKPDdOVr7RLrmMhRvZs59VpBxAdxClU5Pb3se015Jj4epnD8Zoon95hwBdbS6PxffeMbU2nTr/mq9mrRg5KqT8EXPPe/+I9Pue2szj1TSulvlUp9QtKqV/ouX/cd25usWDve/cwyvHs3nV4tGX+iBERnLoaaw/ji0sEsWiZvHBMfWjHFmc/TnEWy1Cg7GJ6oRM4ihA9DBkZbedMopObDxV3BhHfbVSXkHJn6gXltBNG6c1PR4+7u+o0bmUETZkdqlstQsJxBsMSVr7cV5yYNGsRyWMgwKuVTyAq7zSDlejhsJuwGCoGp1lZSR+OVg3HfU3nCjpXsAgcEIuh4qSvudVNwyl7pmbF+fKYc+ZExGHCbhZZo4G1vHlzBPqVLHYkNpWyknoWai0NqfLwGokeqi3hfO5Aqi3pRX7MXc9tjYxl/fc63Pr0c0bMhLSPi4T/GHUzt+McJOIZsjRjneMyP7YM6USOe7gfu5fI4XcA/7FS6uuABthTSn038LJS6nKIGi4D18LxzwNXssc/Dry4+aTe+28Hvh1gT529/4Row/b+zQf4nj/+W/mjz/wSb3rkOs89coXZ1ZLipMB3JoseSKmGNxp90tJcNXgzY7UXwFOENkCMGLRH9yGCUApvFE4HmvhATmudxiqP14FvUmmO+oZb5YzzxZEwKTlE6bqy9DOLH4pTo9l24kOHIXwkvQ609SJuYzr5Edi0gKi8CW3RiGtwERbpE+W9COSMCMthZVDKswiU5b0xSeWrD7T8JhP2iTT9hVI0Rc/MdJwvTzhfHIvitV4SKdk22ZghR/T59H+0uy3COL256RheddECRslrtOHcKz86iHRcFgWUuPQ6Oht9HRf7WBvZGumsjcuGx2bzHMBaNBG7GPJYl9q88f+8xToK66x/prnqeXTG8fGEiU004EjiSvdjrxo5eO//mvf+ce/9U0ih8Se9938SeDvwzeGwbwbeFv5+O/BNSqlaKfU08Czwc/d9Zvdp7viY3e/Zo1SWN+3dYHiqZXFBM+xWUBZSjHTZB6RUiiDM4ZzJi3PBQEQUZVZ/MG1kagqtzQw9aW0gXI2MUeELHLxhMZRc74QJOTIU7RYrGdWuLG7iUpERAM94WxmGe7Iio5lL1JDecyltTBWiBjWoNHmq+jFfcZUXdazcVoahN3RdwUlbM+9KOmsE4DUYejteGpOiZ79acmlyzOPTOzzW3OGJyS0ulYdcKI7Y1ctT2pl5cW0zT07jyhF2HS7uzp/mmdxczJu3rb+mT44h3aZiTSAK+Lj0f3zO9TrEWPnftPy2zVRj8zzyv/NoYr146dhWhJSib5FhJE4fk8PChSC3C3iOcWIz1lUa3TM1n9upzL8F/IBS6luATwF/BMB7/wGl1A8AvwIMwJ/z3t9/TPNrsL1/9T5+5Fu/kN918aNcuXibq1cuM3u5pDgu0f2AGuyIhV7LwTzm9pxpaVg8OknSeSiP7xSFAm+i0pTCa40zsrPawWALyZtj7WGccNTc7qYc2ilTfRutPJfqI4rCsqJENRbrwMyN1A8KBBEZo4a8W0GEfYcOQkxLwv2RSBcvqYTxYCMrVOnHbcAHJGWAiPeqwA6eYRCVcY8sqL1mxW614qBaUCrHxPTsFUumpqPWPbu6ZarlgosIv1INdMqssS/F8DdGALH2ENtsEEBSKougwjk479d383CsVuNzbdYmHGAQxWvRWBZOhdabBNfeTAVyW3s+dVrCL9rdHNTdLC9Cxkhh/fXHSU8bZ1ECZ0beoVg/V5k8lXBRxIRx0FMITkLlUcf91xzuyzl4798BvCP8fRP4/Xc57m8Cf/O+z+bXaW6x4KV/9xj7f+R9PL13k09ePs/iYklzo0Qvexg2fFSWXijrMDeOaQqNezSMeseipJYIwhfgiqBhOWh84cBA3xu0dmin1pWylNC/v7A64FJ5iMEx1R1NOTBH/BSNE9Zrq2URl2E4wjFSzxdeCpF+jBbNUv63jWeYhgXvMgi4B5fpYSR+STdGFc4WuFrjSodvFGVp2Zu04hTqJeereWIxrvXAVHecLYTEJmIkUOMuZn3kZbRrNYYSS+RszNOJ3hvBHWSV9LzImBZq+Bi2XeA5OCqa9eIY8lgp1h/WUhSVpQ0bjiJ2QpxSW0V38teMf+cOLEYgzm9CqON72ojkfFjZ4bxiCiFOolzviIQR7ZGBS6fPXHg7izWOz1+rve74HF7NnvknL/Lur36Cp6Y3eeTybW5ducjspRJzUqC7HmUdHjfCq1NxWTAQ5ctHTArN4lKV6g0ohS7C7mwALV0L1wvbn0OLOC7ylJGMFqTGcGO1w53JlKnuOF8es1uvuK1n6eL1jcXaUCeIjiECn8JshbJ6XNxhLUWHgE9F7ID6BJzgNYZGODNVYIRSgZrOrKTd2ZWgSkdV95ybLbg4PabSNsjR90kDYWpW7JsljRJnkWs7APSYxKwk97tTziCvO8R8vMPQ+AGTRQPRTJbHx8WXL2aT/R+Pifc5oM/We/5cOfFvznx9yikFy0V34jmeOte1yMCvO4y7pPub+AtNdAouMVXF5+kCzLXK6hibhd0SJFrwPT3beCTuz17X8OltNnzik7z733whu6blC85co3ukZ3FRcA+U4guV8+Bc5EwLUYKWCGKwVC8eMrnRC0iqW68/FAGPoHoFfeB98CrVHiJbdaSGAxi85jDg2xvV8czuTZR2KDPujH5i8TsWZUJ4WYQoIlpML0L7M7ZBXZDD03ZUxdJh6tS04e+VzGuYhUrMUHolUUdzYcn5c8dcPjjikdkR+2ULEDoEAnIS3L9g+KOAT8T0y7F6jahkm+WjxxHME9OKLiAH8zRhbSCLMGyFW0NOjgtHr2EcHBI9bOIIxvbp+m9YR1rm3Y8crQl3x0DE58gtL2Zu/r4buEo+q8gtcVp9rPVleu58PLtRvfBoqo493XJOz7lojjkwCw7Mgj3dbv1eXsk+7yIHvOfp736B933tYzw2ucOFy4fML1xgdrXCzHupPVhHIohEjRGEUuIg+oH66jGu3Gd53kidwQCrAEMOwr3OaHzvccoxKKF+p4BuCLgH7XDe0dmC28OUXdNSKsubp9d4V/ME897IKYRro5x2aO3pVwWuj2OXoUAWuhTKjxFpwjIE8JSy4gxUILZx4b7YxTA9OAPD1NOdc8wePebJM7ep9MC06NmLjsGLzmKpbNhZhdnp2Alt3I5pw67mEuFsAkF5g0HT+6gzYRLvAMTiWtx5w+MxzNSQ2pbbdDU3UweALuAt8seNIKrT/BLpOfFjSqHG29Lz+wwzkcG3czBW/vcr1R/GNEJ+5Q6pVNvGre042OaF62l8jiF93iYMaEU4+AjGWo+uBKmrOeL+4dOff84BiR7+/ft+G3/2K9/Bmw9u8M6nDli+UFDfLtGrHmyHchs071ohRQYvjczliub5Y1y5hzOGOOVkQnszYRKMDgK+ngFpD/bKECfGNR5deA77CX1lOLENZ4o5B5OW+UkzEr4oIJDKlPVAr8B1BqU9vtfiABwpgvAmgJqcQtnQ0SBoc2Sj2pFqzpWwOvAMuw59tmN/Z8nlvSPO1nOM8uwWbULTVTrKv4W8OxKXht09MkTlhbLOG1zY1aK+hVzEIwWaCTtidCLysYdWqtfUyq4hHvMFm35nkO5oyTFspCbbiGTuRoef/x21Se4GOc5vz6OKzfPKp0Y3YdUpekggLH+qPiGPGwuVOQgq3ieRg4yvx9fJL+sX7ZT/23v/BAdv3UVwjPdun5fOAeDpH/S0X1FyuTnk7OVD5o+dZ/ZyiZkXEKMH50JKkdx6iiBQCj1fMv20xhW7dLsaH3AMJnQtCgNDYKt2SmL+LpC4GuVRylDoAqMdx70MLJXKsmNant2/zku3d3HaBFwCFKVNmppae6isCNEs9VrKgB+vE69BezU6ulB/UM6nrottPKvzluKgY9r01GXPpBw4qJbUob5wplykizyyQK3vcnLx1UF4dtNyJWtJR9ZrDj1mTEPWKvVyW4dOPA3xAs9Ldutye+sL8NQxd9nIN6OAbczX8TkjRDtGEfnj89c6Lb7r1v4/dQ53K0BsHJNHDybiJdU6K3fuGEp8wHaI/fjiGf5/v/p72f32fR798ffi2odpRbLqJ9/Dd733t/Mt/8HP8OT+bd7z2BnmLxSUhyXFqgfrQvQQUovYSjNKogcto9P6cM7seYN7aipphJb7ihbQ0ta02uG1FqCi9/TZx9obGzQiCha2Yr9YsnAVX7BzlffMHuX2qsBbA8ZjTGhNgbQ4i+AcXElirFaMIWpwDt6QRHNNB9r6xFi9Ouux53rqnRXn9uYSoAB7dcvE9ExMx17RsmNaDF4k8PS4a24i6zb/jyQoUmz06w4gRA8g9QbrNVapU7toDIktSlqQ4WuQw/xaPWBzp84tRhDWq4SUTIs3YCk26wmbj91237aFniMxgVAIzLoSITKIqc/dLG/Vbi1ebsykNKyjMkvlkmMA+NH5m/lbb/9Gnn3ry1x87hP4vrsHHOp2+7x1Dn4YePKfGE6+uObp2U0+/MhFlhf3mb1cYRadRA8u1B2cH5EzEQehw/5hPebmMbPK4J5oQgcDaW8G7UpXBM0Lo1Da46xiGDRaC6io1FJ9vtNPxTnYml3T8tjeESeLhm7QoPya0rcphFJOa4/yKgjuKkGOQKo9uALpTPSk4qkaYNhTtJcc5pEF06Znr1kxKzvqYqCzhmnRUWjLjlmJMrPuRvht+FXqYQ3As8ljGM2i1xxD7OOPiz/2dOTYUvVr0cMo6yZhdRlbglu+19xJ5Lt2vrjXHQBrx29yIcTHb6tzrL1Hv+5cxgJqNL/1cVVon94t8pH3v/649eggG9Pe6NBEJ3rLlvz75Zv5O2//Q7z5rS/xzEffyWcCWPR56xwA6p9+P9///i/jT/3md/Hs+Ru877FdVp82VHdKTNuDG0JRMrOAb5AkWy4ZNViK68dMG8PJo1VANcpEpFmqQAobaw+AkaEsaz2DNrRDgVKem6spjzZ3ZHdDca6eU1UDQ2WSU4gphQQyXlqcQSFceB7kNURlPJyyC47BSuRgKzh50lE9NmdnsmKnFvbrylimRUdjZKeamJ4dsxLsRShwRYoxiLJ4o6BLzi0AIZXwLqUUPUJuo7PoAcDi0WjRrmRdoi6HBMcIQdqQ6xOWeQEwHX8X3EO0uHhs8P05QDamDPlzvtJYc+5I8nrIaXzEdqe25jg2XmYTBbqmOp45hBEWLnbsNH/5E3+YG//gSc795HM8ffUz4xSifV47B9e27P7chOZLeh6f3uEjV445+dQ+k+slelGAjVNJ/jRyMn7LIYpQXU999Rhb77M8X6zXHwr5xmyAYxO0N61WWCNj3YWWce6ToWanEDbgx5o7vL+8zELVVHUvsw3e4JzoWioFzmr0gBQrNUTBClcJVDoVHkOdwVZw/BRMnjhmf7qkKYT1elp0VNoyMWO9oNYDje6pdZ9Qi43qWUnHPE0rRqx/o/u13dN5cRBdEOWsVBTkLU7BfXPbVKqSx5gQNTi61BaNQ0g2/b9peTdjM593EGoG8nWWCrSXxdZ6sx1wBafSj/j8kSsifjZRxzJ/3fsN4bc5pLs5BINMay684i8994e5/vef4uBfvIf99irbMZS/Pvu8dg4Aj/3zj/Nv/5Nn+e1nn+PKmTt85Mouu8+XFPMKsxpQ/SBUbc7LN7AGrfZp/kL5MMX5YoGtd3HFWH8wbRzQUjitE3AqRg/O6ESocjRMmJg+pRaXd4+4fTRlGAyTuscYiRYkLfEMvUEHQJTyjKPboWORQE9IkXT+mEc/PefC7kmKFqRV2VFrK6rLemBwMkK+LVUolZXag9e40F2421TfKOc26kLKEJChCs+VqzjlJq3EQPgSbusD6tJsLNbcNnEOmzWM/PnzOYsIqe79eicB7l7kvJvFc9rmSOK55Oe39vtVZjE2HQKAUYqfX53j//7jf4pn/0mHec9H2Jv/7K+5nnAv9nnvHIarL/HSv/xKpv/5h3lm9yYfvXSBk0enNDfH6EFZFyr7aj2CCAI5eXqhD+dMXzC4csZqVwWggSzYWI/wWqepSRunNK2h146bqylnq7lwASjH5ckhn5yeYdnKbq2UR2uwg0EZiw/cDYKVVqMkXmp/ioNwBlZnYXh8xZNnD9mvWgodiD5UYHTWlkgAMi26wLsge471I+16Gab5NhmJWlemx/R+vHTygqQc68N0ZqBP5+6goU0Y8uZ9uW3rMGwurC7jirDh+5ToZ+RIKBU4XIoe8ufWG8+37XziAk8OYltEE+sfr9A6Hd/XerqQOwRAnMJP/Kf8hv9lwVt++RfB2c+qU4j2ee8cAB7//o/xb7/xLXzZmU/xpks3+MSlK7QvlpRHJWo1SO0htjXN3R2EMEY7zK0Tmp0KW5bSygy1hkKpQOaqhbJNa5wGax3WKHqnOe4b5kPNxHT03jAxPRd3T3jR7iXyWpC2pjEOVkZa2z46oCxaUKHeMEB3BrqnWg4O5lyYnFAol9h/am2ZmG4d+29GmTsIUFwPKE25EaTGxR8r8mVYC9GhGKTuYLKMV9IUhUGPeO/0fOvIwH4D8xDPMS++xcW4VohEURLk+eLzZos236HX5ywI78GmFEZu34hWNtb8Noe1dn8WRXRep2LkZsqT4xC2RQgaaL3nb1/73bzt57+U3/D357zlPb+A36yPfZbtDeEchpde5qUf+kq+8ls+zlO7N/nYk+eZvziluVFRLXtU6Fx45UY4dW5aqlreyPyEGiz1S8fYZp+lLoLoTFi8i6A4FQqUTnkGoxkKjQlcCUdDLexQTgp4u2XLtJ6glLQzrdVYq1Aq0NQFEZvIy2ACB4RyElAME2ivdOzuL3li/w67mWiqFB67hHhcuYKof6lVxCSE8eENBxE5DBrdh4U6ajPmoX5SvX6F/Uzjso7Hq1/ka+PRGziDbYs/LuYyi0TibeuLnnCbWGxJvtLuvzbU9UrpwcY5RweRW3SsuUMoIwsZcOw8/8O1r+Kd//hLufz9H+YtN37uHj6tz469IZwDwOM/+Ek+9scucKk+4sqF27zw2ISdFwvK40AlFyHVeVsTRkcRJziz+kNzrcI2s1R/ABnQKpYKX0hB0muNNYa+l11tZQ13uikH5RLtpWc9LXoOJsuQfmg6CrwXVWzVq3EgC1IaobK0ot/zTA+W7DUrpoVECLUecCgKJSAn6RKEnrr2p3ZpaTEOAb1YbZ8/CDtj4hnIYNHrArakNONuvIrxeccx6u0qU5rThcbNdRwXazy/Kitqjuce3mv4DKODcNlz3M1OTX1uSW22OZd1Vqrxt2FMGaJ9pJ/w1mu/m/d892/i8vd+iIs3f+Yz2nn4tdgbxjnYl6/zk+/+Ev7s/+GnuLJzm08+eo7FxZrJjQK9KGG5GkFRMa3ILaYXcf5isJJeTEpsWRMZpIyS3coVcpzT4I3HhuhhsEYo2OoGXS4x3jErVsyKms4Z5qYi0l/YwVDMJXLwAcGtouBNaGe6AvpHOs5NWyZlL4K92grzFLKrx4UYlbNhbCEmrUWV1Qz8SMQyLv4xz04M0aGNGVOO3JK+Y9b63Pq9RCRg/JhjyzK8zlgTEItHrrc5RzxBfE45Vq0hHMvgTB2yQEsFK3/3galtacQ2R7CtKJpjIPLUITqFeN/Hh4o/+a5v4cn/2WDe+T4uDq+9U4j2hnEOvu94y3e13PiKHS43Rzx68Q7Xrlxi52qJORrHuVPtQflTUUOav3ChBuFkxLuensEVhVDIBYCUKwL+oVD4QVqa1mo6ZTC64OXlLpUemJieWg9UZqApeo5WDdYJUtDX0BcTVBvqkROfKOljILE665nutTTFwG7ZSq0hRAsQQUc2/b0JVNqmqiQTl+OkZW45X2EfWp75JOZIdbaut7CuubBe9Lvb4NLpEW4SajI+xCbM5/bnGtuhHuszhxDqyHK2fj3SyJ5i8xxejRw27zbAdqfQe8/Pri7yl37x/8iTf8/w9P/+XnD2NUsf7mZvGOcAwM99gH/+rt/Gt/yOn+apvVu8cOks84sl9c0K3XZjwSfDPXilUPH2CJDyaqw/rDqaqye4ag+vjRQtQ3vTlQpfKmwhzFFDb9Da0w0heqgaSuWEWSnUCXaqFR6wTrNoqzWgk1kF5uyQWgzNGDXUZqAJUUMcoALQyO7dB9m7Qq8LyUrNYdyrZJSaRORi0QGOLMeM6Md155EUxr2kJwJ06tfo44Q3MZto9NnCfQWIsTx/WMjBF1g41S1I9YOAmExDTjk2g4B3QBatC8/3SmrfW89n45w3owQYnUIe9XzH4W/mO773q3n6uz/NMy9+CN9nvH8PmL2xnIOzvOU7W258+Q6X6iMuXb7D0aWLdFdLzEkl49zOj7gH5dPX75UaL4W1+oNGHy1oXi6x9TS0MSW98IWAlVwpzmToZcpSKc9q8NxaTZmYHq0cu2XL4Aw75Yp2CBJz2jOEDkVsX5oug05XUE56mmKg0I7aDBTBMciunVXIQx0iYhuiEtKmViTIhV/pPjENGdYRjVFwN0Yia5LySjgjc6Rl1KjY3GEjgWz8P55nbjZLK/KzjIszb222oeux2RodC5qn24UC6SYdt6l9kT9f/pwgxc/4vPG5tqUOrfd8150v47u///fz9D9+jisv/MxnBbT0mbY3lnMA1C/8Cv/7i8/wjU++l2f2b/LOJw9Yfrqkul2i2wJW/Rg55BFDbhvTm3hPceOYyaTAlrWAoYzUBIqFOAcbahCDNhjj6JThcNVwUC/TjMO06DgqaqZlx3IoKQpLbxDdzPCaaiWTmV5Dv+OZNX3CMsSOBESg0LicYpfCKKEsv9ucBJDSAZkKLEaUFWMkYYMEXmR5htN1hkhYYhCtymhuo014N8cQuQ6sUrjU9pRFJ6H79jHn9DVtIA3T8278veZ0tgCaNs+T9Pp3jxIAXrSGP/PBP8HRT1/iqe96/TiFaG845+CHAf0vzsFfgIvNMecfOeLkynmmL4sIDkMERW3pXOSmVYA0S/cCaylfOqaeFvhzJuAdVNLf9CG18EbRh7HulS64vtxhZjp6bThjFpytFnSuoNKWO3qSuhJegYqrInQr7JmeupSWY2WELCXXUohCvlGmvVRSqDQbEcXdioaVsomWLHYjxGTSMrI75WSyORXctue8G1Zg0zHk3Yze68RXAKS2n0QP20FU28z6MUpINQu/eUwWJWzURuB0+rDpFAyK687zpz/4n2L+3nkOfvS97LUfe105hWhvOOcAcOFffogf/iNfzNc+9itc2bvNLz12wOLTBdWdcZx7LXrYrDvAOlAqCvW2KyZX57h6B2cMOkxSmkUoTpYaP3icFrZnW2jmXcXtbkKhZRHvF0uOTYPGsz9pOS48BNHcBJNWATpdeCojkOgmpCdAACXF8eUR9VhmdYPUpWB0DJGlKS7wkSCWtZ1eVLKrdEylrHAcwlbHEIlR80nMpI+5pUAJp1ODGDGM941dCulAZIQvW2oHfTZFaTLHsOm+7oaS3OYU5H+VUp4fmT/Jf/fer+XSd03Y/ZF344ePf06QjJ8te0M6B3v7NvaffQHlX7A8MbvNRx85z+KRM0yvVZhln8a589rDqdRiszgZHIk+nNNcK7FlI/iHAqFnW0n9wRdSpMQrEcJxmlvtjGnRsyoLzpRzdstW8A9lh505zHIUBI4R+TCBoumlFBHYjmPUEKMEkAJhjCASuWtYkFXOKrSxc5ehkAl52C9OJXIW9r4YHUgGw87Tmc2UIg5XvRrlu8uKjSWsAYrW8n4FcYnnXJGbLFJRpTs6k/FR4+M2R7ej3a37EB3Gjy6e4L9++x/lLd/+Mk9+5H3AaVqG16O9IZ0DwIW3f5h/9X/6Iv7A5Q/xlnPXefdje7SfNlS3S4plB/YVag+hzpDfFrsXDJbixgnNpMBWJT6yVneSYrhaQVDJGgZNpw2FMdxoZ5yt5lysjjhTLGQwquhE22IiNHERERnbmnUtrM2FFqh0EklJhUC/1jPPpeSjdb5g5cq1Y2Z6xSPFYYok4u2xgwGjLuYpCx0OURgfHxt/b0MdbjqKXLwm8hjE7kMsAm7CocU2aOLugnCMUYdlpJ0vM5xHfGx83lKdjhQW3vN3rv8u3vaTv523/INrvOkjP4v9HMObP9v2hnUO9uYtrr3nNzB9/Jd5dHLIr145Yv7JM0yvC5WcClRyHsYoIf5N9rdG4tNsFkOtOqrrc4bpHrYy6BLR4exAt1pUr7XHKsNgHIM1tEPJS+0ej9V3mJoVE1OzW65EGaspsL1HHwZeBwXDzDHTjjJMXcIIeBpJT1yaWzBpcUeuQRmkWjhRAwchd8FLR6JRPbtmKRwOW/bBvJi5Hll4cQxbgE3RtjmDvK0Z7xeWI5cKj31glxoLkmO3wcFapPCKMxAxBQk1i7X3EM/5LunDofP81U99A8///Tdz5ofex5vmn1kOhQfJ3rDOAeDN332bX/29j3CpPuLRvSN+9Yk9dl4oKI8qVLs5cxEuk227w2b0YGV6s7leYesJrtToKkQPHbg+oCeVZxgMnXEUxnC4mnCj30nRw9zWaOWxxmMrj26Efl4PMmNRFjbBr9dx/W57eJzVJFpfsnIlrSuTA+lteI8G5q7GooNqdkfprfBAeulA5EpM46IeKw3bIMd5SpOfXt5xSNGCygezdLovMkXlCzePX3IeyRK/NefPB6Sa+JlspA3jZzk6hb/43B/m6luf4dzbPsD+0Wd3XPpBsDe0c3Af+DDveMdX8H/5uh/nyZ1bfPzCeeaXZ0xfLtEn5Sif570UKY0+XXuAsbXJiH3AOYprR0xqw1DX2DowQ68UZqmxYWrTWi/zFINhqUvmtmZha6ZmxeP1bcpqoKcW1GXtsd14FcdOg/zrszZiCKezYmTcKePCa51h4SrBK4S2ZMQtTM2K1pdYp2kpaZJWgqQkTQA3WdTWyGIbLmDNMbDendgmkpunE/E5BeWosPjECXy3XTt3AGuYCD/WV+J05rY2pAOs97x7dZb/60/+KS7/pGH/f30fB5/HkcKmvaGdA97z5n9ym4/+notcqI558uItnrs4ZXmhpDwa1bnXipM5IUw+gxFL1nkHox8oXzqmmRXYOoKhxEG4Kox1G0XfB+yDddxYzdgtWoxynCnmPHpwxEevzeR0lQCflAeqmMcLuKmIyMiwuEo9hJ08VvHX6w0uFCojcnLlRHuijo/zGocDJcrZFp2IUGd6xSzoZAphy4YWZvZ3Dn7KlaDic92t3vBKFPJCI0cSBVs7LnQrcth0vD06Bjkv+coapTAodPgNYPG8q93jz/zEf8YX/L05b3nPz6fXfSPZG9s5AO5XPsI7furL+ZNf+295YnabTzxxjsULUybXa+pFJ/qaa4XJjZoDZKS0EC9rFW4X9qg5/WwXW2tcJeI0rtXY0uO0HNz3nk47brYzHp0c4vyEWvc8u3edjxaPwHJ8aW9AVxajpUNRZJDpuNOKIM16QTC3/DaLTkVJrXzgaVDEMmWkkGt9iXPBqXjDrm5FI1Q5DKMTGFODvJ250R5UG7dvZEF3Y2zOYdDRV6/NXECCWMu3EX/7NKSVRwu5Y2i95W9f/5287X/7Cp75Z0e85d2/gPs8KzLej73hnQPO8qYfPMF9reJsNZdx7kcn7IbaQ1TnTiI48YqE092LvL0Z0wsc+vYxsxcrhskEWwm60QTcg9MepwQ5udIF865KZDDXuj2enlxnenZBe7yb0cEJGUxTDBQqjkaPkOlUfGQsEt5tJBokXYjEMDE1ye8Dwri3OI2FqwKng7QtG0Zy2ohlyFWYcrubXkQUuY22Sa8Wrfcam+EZJCLYUjBdq8FAuXbfCG/WSuG857+78R/yo9/xlTz6PR/i6Zvv/LxoRf567aFzANQHPsY/+8hv4Y+/5Re4PD3iU1fOMv9UI1Ryy3I9epBHvMoTroOjsI7y6h2mOwVDU+JKhS4VZhlYowBrPIMxzFeVsFRPBo6Ghh2z4ise+yQ/ef0LKZYmqWh7pyi1ZadcMQm08jlkep3ReIQygyz0pEqFqIJPdScFyEA4u40zUitHHwqYK0raIIPXqB6jl5CxSm9CohP4ibvDnfPjctt2fFSk2kbymnce8jZkTqpiUHx8KPjW9/9J7I+d59Hv+RCXHgAOhQfJHjoHwC0WnPunM9q/LgpZl88fcu3JhtlLFeaoQ3dGogfr8EVWy85rDnmhchMcpRV0PZNPH9HvnMHWBldKBGAWGquFb3LQIqd3db7HuXqBVo6joeELdq7yS48+xvHRWfRKkJJDZ1K9If7IS4/hfO9Eji6nacvD+x3TpqKl0ZI6iH6FTwjKOH0Z+R20cuhQy+hdwTENN+1OoK2XCGITZ/Bqtu34V1Kk6gMcPEUVIY+L3YmYapSMUUK0EsWH+5o//Yv/GY/93ZIL/+79+P5XHzqFLXZ6YP8Najv/+pf5wY98CTvFiid2bzM80jF/RGN3KlHnzusLbnsoC5yKGrzRwv2gFOp4wezTC+pDL0rXARatWwW9xq0MdhijB4ATW9O7gi+79DzDuT6RvuAVlbbUZkQ5xt2+dwW9K2SXDsXF3AyOSg2hTdkHhKNl17TUug+FS59AVTGlSI/fSF9aVzL31SkAUq4mHe2VuBty67NLc7Obsa2QKeclkUKpQqExdCBqJc/18aHiK3/2z/Bffct/zpN/7IOYn3r3Az0y/Vrbw8ghmGtbzn/vlP7/bbjYHHPp0h1uPxLIYE7K7ercr2Sbyt3eB/RkrD9UuFKKkygC7gFsoRkGw7XFLmeqJVp5bvYzLtVHnL14xJ2jsxRLjW9NGrYCCddj4TC1/7xGR9XmjQIjQKN7et+xcBUwzl9E3EPEMyQgFV6KnNmsxCiEW9D6MnUocpxCtLsxLq19bBsRQ168zJ8jPn8sNJb4FCnkrrD3nu86/CLe+r1fwxM/esQTv/h+oljyQ3tle+gcMtv995/gQ8eX+MK9l3h67xYvPXHA8pOVkMHk6tx5WzPaK8xepPTDaOh6mhePGXYOsHUh05uaJKfnC+F9OG5rbncTLjQnHA0Taj3w9MEt3n12h66twUFn17++NNfgNStfjCmDl9Rik9hF45jqVRKUibMT6xoOgYQ2K3Km+5SjUQNGiRhP60txMPcQpMex7TK83qvRsW0rTpapliIpRKkUpYo0eJ7vOPxi3vr9X8PT3/FJrjz/Mw8dwn3aPaUVSqnnlFLvU0q9Ryn1C+G2s0qpH1NKfST8PpMd/9eUUh9VSn1YKfXVn62T/0ybffkaH/43zzLVHefrEx555A4nTyj63QJfFaJmtTlvAespxzZ5vc3pzaM5s+dOaG66oG+pErSaQWMHTdcXXJ3vpQV5NEx4084NdveX9AcWJjIq3Qe8QowYooOIku7REnYhcDFELGOje6Z6leoM0SLdvEOnx26aRBpDiDYK5q5Obc61z3UjYogYBzjdsszTiZiSbAKjSjWqcdeZYzAoblnLf/SBP84f+Ot/kZ/46i/gyn/zMwzPv3C3r/yhvYLdT+Twe733N7L/vw34Ce/931JKfVv4/68qpb4Q+Cbgi4BHgR9XSr3FR9bUB9ye+t7neffXXeGL9q7y2M4hLz12lpPL2Ti36+8ePcCWoSw/phbxmMGibx6x80JDP6txpSI2532pcaZgMJ7jZcMLi32uzG6ztCWzYsVvuniVdx43oGTBDwHEFBd94rAnG2AK4X+eWkRiWBOiBxhrFrEIGSOAvK0oQKsIcBrWuhoCrCrQel3u/VQ7U52uQWw6ADnH0+K4m2PXcp/iulX8mQ//Cfi7F5j+6/cw6T/xuuRQeJDs11OQ/HrgO8Pf3wl8Q3b793nvV977TwAfBb781/E6n1MbnvsUH/w3b6FWA49P73Bw8ZjFI4ruoJLowYSPLEYJbiNS2Iwc9Bgx+Pi3EXh1/albzK5ZioXMS+heYZYKeoXtNX1vePlkh8N+Qu81tzqpPTx28Q5VPTAEIpTN4pxDre3emySyNvvaLZpKWZrAVt36Mu3+q2zuAkgOYVe3MpQV1K+sl9frQnritpzTpr1SJ8OxTv821hRCpKBGNqgfXzzDV/yLv8i3/oW/wOwbX2bytp97WGT8DNm9OgcP/G9KqV9USn1ruO2S9/4qQPh9Mdz+GPDp7LHPh9vWTCn1rUqpX1BK/ULPavPu19Se+t4XePfhFXbMijedvUF72bI8X2CnlXQeAJV3LF4JRafUmoNY+1m27HzkkOaWTwNVJsxe+NC5WPUlzx8fALByhhNb85vPvsgjB0c4r+icBH9xweezElFyb31UOgwxbaQJws+wnhJEwFOsSUQrM11MKViO9Y3eF2lAKn+eTcsjhrsVJhtlaZSjxKdORLQfXzzD7/n+v8wPfP3v4tk//y6mP/Qu3GJx9+/hod233Wta8Tu89y8qpS4CP6aU+tArHLttyzj17Xvvvx34doA9dfaBqhUNz32KD/z4f8iX/OHnudicsPv4EYtPBTKY+QrVh4D1XjoXG7WJRCtntMCr7xyz/7GGfmdKt6dQA5ilwpUihtMZw7GuuVHvcLZeMB9qDsolX3TwEje6WRhlNhR5vWAjUnAh3RDwj7QmK70Onzah8Nh7k7oQXwAqzQAABjhJREFUBs/C1kISoz2l7gNEe0RHrr1OdDwbe06eMmxzFLlO5d3k5g1wyxn+/fJN/A//69fz7D94mTd95I0zBPVa2D05B+/9i+H3NaXUDyFpwstKqcve+6tKqcvAtXD488CV7OGPAy9+Bs/5s2/ec/mdHYdfP+GgWPDY/iEfenKPnecN1Z18nJvTA1mbFmsQm61NCFSMlvKFW+yerbnzZvk6vFGYVmMNDGVBZzw3FjOmRSfckv2Eg3LJbrHieKjTSwnT02hr3A7ZDp1Qi5kGRaN7ShfmI7JIwwWClZEnUqIGKYYWa44gUs45r7FKneo2bNom6KmKDmzDKSw8/NVPfQOf/gfPcv7HPsEzVx86hc+FvWpaoZSaKaV249/AHwDeD7wd+OZw2DcDbwt/vx34JqVUrZR6GngW+LnP9Il/tq3+6ffzz3/5S6n1wGPTQ6oLC04e0/Q7Jb4qx0Xv7zG9iLaZWmhpb84+dpvmZkgvelCDgKP8yjD0hsWq4rCb4FAMXnM01JTaUijHyhUpgogkslG/IqYVwBq4KeeNBEkVDsycHdOyq1umepU0NSNb9UyvmOoVM93hkEGshavCrIUiUsYBp7obeaqx5qiCU2iUSwCm+Mhjp/l/fPIb+OP/1V9m/tULDr7rnQxXX/r1fK0P7T7sXiKHS8APKZX2un/qvf8RpdTPAz+glPoW4FPAHwHw3n9AKfUDwK8AA/DnXi+ditxc2/LU9yn4UjhbzXny3G0+dmlGe66kunO3ce4tkUPEOOTRA6x3L7xBHS/Y+8SSbncqsOpOVolbGGxtGQbNzfmUSdHTmJ7WSt+h1kIV13tD6wRnUMdWZrYYIxV9TiM3OgYht41j2JEZqvc26Gd6GtUx0x2lGuh9QRfmKkbAlBQ110ars5blpuUpRLklUviqH//zvPk7Bopf/hgHx+98w41LPwimPtey3ltPQqnrwBy48WrHPgB2ntfHecLr51xfL+cJr59z3XaeT3rvL9zrEzwQzgFAKfUL3vvf+lqfx6vZ6+U84fVzrq+X84TXz7l+Js7z4eDVQ3toD22rPXQOD+2hPbSt9iA5h29/rU/gHu31cp7w+jnX18t5wuvnXH/d5/nA1Bwe2kN7aA+WPUiRw0N7aA/tAbLX3Dkopb4mjHZ/NEx3vtbn84+UUteUUu/PbnvgxtOVUleUUj+llPqgUuoDSqn/4kE8V6VUo5T6OaXUL4fz/BsP4nlunLNRSv2SUuqHH+Rz/axTKXjvX7MfBPPyMeAZoAJ+GfjC1/icfhfwpcD7s9v+e+Dbwt/fBvzt8PcXhnOugafDezGfo/O8DHxp+HsX+NVwPg/UuSKDJzvh7xJ4F/AVD9p5bpzz/xP4p8APP6jff3j954DzG7d9xs71tY4cvhz4qPf+4977Dvg+ZOT7NTPv/U8DtzZufuDG0733V7337w5/HwMfRKZfH6hz9WIn4d8y/PgH7TyjKaUeB/4g8Nbs5gfyXO9in7Fzfa2dwz2Ndz8A9usaT/9sm1LqKeC3ILvyA3euIUx/DzKc92Pe+wfyPIP9T8BfYV3g6kE91884lUJurzWH5D2Ndz/A9pqfv1JqB/hB4M9774/UtsnQcOiW2z4n5+pltuZLlFIHyJzOF7/C4a/ZeSql/hBwzXv/i0qp33MvD9ly2+fy+/+MUynk9lpHDq+X8e6Xw1g6D9J4ulKqRBzD93jv/8WDfK4A3vs7wDuAr+HBPM/fAfzHSqnnkBT39ymlvvsBPVd8RqUArFEpfCbO9bV2Dj8PPKuUelopVSHck29/jc9pmz1w4+lKQoR/CHzQe///fVDPVSl1IUQMKKUmwFcBH3rQzhPAe//XvPePe++fQq7Fn/Te/8kH8VzV54JK4XNZBb5LxfXrkEr7x4D/8gE4n+8FrgI94m2/BTgH/ATwkfD7bHb8fxnO/cPA134Oz/N3ImHhe4H3hJ+ve9DOFfjNwC+F83w/8NfD7Q/UeW4579/D2K144M4V6fD9cvj5QFw7n8lzfYiQfGgP7aFttdc6rXhoD+2hPaD20Dk8tIf20LbaQ+fw0B7aQ9tqD53DQ3toD22rPXQOD+2hPbSt9tA5PLSH9tC22kPn8NAe2kPbag+dw0N7aA9tq/3/Ad1/knFtLImVAAAAAElFTkSuQmCC\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "\n",
+    "#Show Image in Train set\n",
+    "train_images, labels = next(iter(trainloader))\n",
+    "trainImg = train_images[0].numpy()\n",
+    "plt.imshow(trainImg[0])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "9342d2db",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "playsound is relying on another python subprocess. Please use `pip install pygobject` if you want playsound to run more efficiently.\n"
+     ]
+    }
+   ],
+   "source": [
+    "from playsound import playsound\n",
+    "import time"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "id": "b0e43c6a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import datetime"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "id": "564a52d5",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch  1 \t Training Loss:  0.8935222659206045  \t Validation Loss:  0.9074592170891939 \tTime taken:  0:06:25.599870\n",
+      "Validation Loss Decreased( inf ---> 98.00559544563293 ) \t Saving The Model\n",
+      "Epoch  2 \t Training Loss:  0.8259349410300669  \t Validation Loss:  0.8810660772853427 \tTime taken:  0:06:31.011904\n",
+      "Validation Loss Decreased( 98.00559544563293 ---> 95.15513634681702 ) \t Saving The Model\n",
+      "Epoch  3 \t Training Loss:  0.7864946653337582  \t Validation Loss:  0.8560434800607187 \tTime taken:  0:06:42.054107\n",
+      "Validation Loss Decreased( 95.15513634681702 ---> 92.45269584655762 ) \t Saving The Model\n",
+      "Epoch  4 \t Training Loss:  0.7327645833509556  \t Validation Loss:  0.85228816833761 \tTime taken:  0:06:44.607239\n",
+      "Validation Loss Decreased( 92.45269584655762 ---> 92.04712218046188 ) \t Saving The Model\n",
+      "Epoch  5 \t Training Loss:  0.6791089947564878  \t Validation Loss:  0.7294051167037752 \tTime taken:  0:07:20.984154\n",
+      "Validation Loss Decreased( 92.04712218046188 ---> 78.77575260400772 ) \t Saving The Model\n",
+      "Epoch  6 \t Training Loss:  0.636842123300269  \t Validation Loss:  0.6813744753599167 \tTime taken:  0:08:34.964242\n",
+      "Validation Loss Decreased( 78.77575260400772 ---> 73.588443338871 ) \t Saving The Model\n",
+      "Epoch  7 \t Training Loss:  0.5975492863875368  \t Validation Loss:  0.6167861642660918 \tTime taken:  0:11:04.116870\n",
+      "Validation Loss Decreased( 73.588443338871 ---> 66.61290574073792 ) \t Saving The Model\n",
+      "Epoch  8 \t Training Loss:  0.5665440737823213  \t Validation Loss:  0.7120587097273933 \tTime taken:  0:11:23.060953\n",
+      "Epoch  9 \t Training Loss:  0.5627748251393221  \t Validation Loss:  0.6692608294111712 \tTime taken:  0:07:31.104611\n",
+      "Epoch  10 \t Training Loss:  0.5312202539888845  \t Validation Loss:  0.6212516344255872 \tTime taken:  0:06:54.422804\n",
+      "Epoch  11 \t Training Loss:  0.5294764937320049  \t Validation Loss:  0.6790190057622062 \tTime taken:  0:07:05.191500\n",
+      "Epoch  12 \t Training Loss:  0.5147726689397857  \t Validation Loss:  0.6835516048250375 \tTime taken:  0:06:54.515692\n",
+      "Epoch  13 \t Training Loss:  0.516053875638307  \t Validation Loss:  0.5850699772989308 \tTime taken:  0:06:51.785147\n",
+      "Validation Loss Decreased( 66.61290574073792 ---> 63.18755754828453 ) \t Saving The Model\n",
+      "Epoch  14 \t Training Loss:  0.5054195006580456  \t Validation Loss:  0.5986003837099781 \tTime taken:  0:07:15.805469\n",
+      "Epoch  15 \t Training Loss:  0.4909070910265048  \t Validation Loss:  0.5399058399101099 \tTime taken:  0:07:06.906177\n",
+      "Validation Loss Decreased( 63.18755754828453 ---> 58.30983071029186 ) \t Saving The Model\n",
+      "Epoch  16 \t Training Loss:  0.47156357611327065  \t Validation Loss:  0.5632316759890981 \tTime taken:  0:06:35.463655\n",
+      "Epoch  17 \t Training Loss:  0.47529219876488915  \t Validation Loss:  0.5946385694874657 \tTime taken:  0:06:48.524029\n",
+      "Epoch  18 \t Training Loss:  0.45477662398817315  \t Validation Loss:  0.5385274368303793 \tTime taken:  0:06:36.054347\n",
+      "Validation Loss Decreased( 58.30983071029186 ---> 58.16096317768097 ) \t Saving The Model\n",
+      "Epoch  19 \t Training Loss:  0.4467847437422345  \t Validation Loss:  0.5110538527369499 \tTime taken:  0:06:57.678132\n",
+      "Validation Loss Decreased( 58.16096317768097 ---> 55.19381609559059 ) \t Saving The Model\n",
+      "Epoch  20 \t Training Loss:  0.43970482947601786  \t Validation Loss:  0.578308700411408 \tTime taken:  0:06:55.490158\n",
+      "Epoch  21 \t Training Loss:  0.4437054166084398  \t Validation Loss:  0.5535954915814929 \tTime taken:  0:06:42.015812\n",
+      "Epoch  22 \t Training Loss:  0.44214661373500375  \t Validation Loss:  0.5116364767567979 \tTime taken:  0:06:50.197071\n",
+      "Epoch  23 \t Training Loss:  0.428848498608863  \t Validation Loss:  0.6346474328526744 \tTime taken:  0:06:44.003828\n",
+      "Epoch  24 \t Training Loss:  0.42443272973532264  \t Validation Loss:  0.5160020545676902 \tTime taken:  0:06:38.077509\n",
+      "Epoch  25 \t Training Loss:  0.4128420905790467  \t Validation Loss:  0.5394650613544164 \tTime taken:  0:07:03.717844\n",
+      "Epoch  26 \t Training Loss:  0.4160873369462248  \t Validation Loss:  0.5195161462933929 \tTime taken:  0:06:47.996162\n",
+      "Epoch  27 \t Training Loss:  0.4038385886941915  \t Validation Loss:  0.524651152116281 \tTime taken:  0:07:07.090472\n",
+      "Epoch  28 \t Training Loss:  0.4079473443005396  \t Validation Loss:  0.48507266533043647 \tTime taken:  0:07:16.085398\n",
+      "Validation Loss Decreased( 55.19381609559059 ---> 52.38784785568714 ) \t Saving The Model\n",
+      "Epoch  29 \t Training Loss:  0.40127678457107663  \t Validation Loss:  0.5480059128668573 \tTime taken:  0:06:58.621673\n",
+      "Epoch  30 \t Training Loss:  0.4079972279082606  \t Validation Loss:  0.4860136133653146 \tTime taken:  0:06:45.081846\n",
+      "Epoch  31 \t Training Loss:  0.40153468269314885  \t Validation Loss:  0.5275546344066108 \tTime taken:  0:06:29.638256\n",
+      "Epoch  32 \t Training Loss:  0.38982377127082884  \t Validation Loss:  0.5190505493018363 \tTime taken:  0:06:41.280194\n",
+      "Epoch  33 \t Training Loss:  0.38694217512249085  \t Validation Loss:  0.5106402661789347 \tTime taken:  0:06:40.476959\n",
+      "Epoch  34 \t Training Loss:  0.3934135176916269  \t Validation Loss:  0.5252610400870994 \tTime taken:  0:06:56.501683\n",
+      "Epoch  35 \t Training Loss:  0.3762604439280171  \t Validation Loss:  0.4691933031987261 \tTime taken:  0:06:55.941128\n",
+      "Validation Loss Decreased( 52.38784785568714 ---> 50.67287674546242 ) \t Saving The Model\n",
+      "Epoch  36 \t Training Loss:  0.3657908062113152  \t Validation Loss:  0.5094884762937134 \tTime taken:  0:06:14.850790\n",
+      "Epoch  37 \t Training Loss:  0.3734221094544383  \t Validation Loss:  0.6241087309188313 \tTime taken:  0:06:16.002782\n",
+      "Epoch  38 \t Training Loss:  0.36661212104440166  \t Validation Loss:  0.5721067921430977 \tTime taken:  0:06:38.790656\n",
+      "Epoch  39 \t Training Loss:  0.3647138514704462  \t Validation Loss:  0.46293201159547875 \tTime taken:  0:06:31.337280\n",
+      "Validation Loss Decreased( 50.67287674546242 ---> 49.99665725231171 ) \t Saving The Model\n",
+      "Epoch  40 \t Training Loss:  0.3529694011515897  \t Validation Loss:  0.4971495535638597 \tTime taken:  0:06:54.011661\n",
+      "Epoch  41 \t Training Loss:  0.3624838686318717  \t Validation Loss:  0.49220840232791724 \tTime taken:  0:06:39.659764\n",
+      "Epoch  42 \t Training Loss:  0.3560715395876247  \t Validation Loss:  0.5051267051172478 \tTime taken:  0:06:34.931163\n",
+      "Epoch  43 \t Training Loss:  0.3451506820088927  \t Validation Loss:  0.5052853872378668 \tTime taken:  0:06:38.050483\n",
+      "Epoch  44 \t Training Loss:  0.35159600940465496  \t Validation Loss:  0.4847101875477367 \tTime taken:  0:06:40.725714\n",
+      "Epoch  45 \t Training Loss:  0.34435510416717635  \t Validation Loss:  0.4940365508750633 \tTime taken:  0:07:10.671743\n",
+      "Epoch  46 \t Training Loss:  0.3474206558658161  \t Validation Loss:  0.522217466323464 \tTime taken:  0:07:03.929293\n",
+      "Epoch  47 \t Training Loss:  0.3459917993446731  \t Validation Loss:  0.5019334826480459 \tTime taken:  0:06:59.124492\n",
+      "Epoch  48 \t Training Loss:  0.343043835590715  \t Validation Loss:  0.47634170011237814 \tTime taken:  0:07:07.514332\n",
+      "Epoch  49 \t Training Loss:  0.3417059610584292  \t Validation Loss:  0.48328716933934224 \tTime taken:  0:06:49.670400\n",
+      "Epoch  50 \t Training Loss:  0.33828138229369686  \t Validation Loss:  0.4837376345638876 \tTime taken:  0:07:05.494005\n",
+      "Epoch  51 \t Training Loss:  0.3368406971727592  \t Validation Loss:  0.5092699004820099 \tTime taken:  0:06:52.169459\n",
+      "Epoch  52 \t Training Loss:  0.32966294400798885  \t Validation Loss:  0.49807466428588937 \tTime taken:  0:06:46.435914\n",
+      "Epoch  53 \t Training Loss:  0.3330573174238637  \t Validation Loss:  0.5112950452775867 \tTime taken:  0:06:37.646492\n",
+      "Epoch  54 \t Training Loss:  0.3176659231843508  \t Validation Loss:  0.5276876427923087 \tTime taken:  0:07:10.544489\n",
+      "Epoch  55 \t Training Loss:  0.3167355079271331  \t Validation Loss:  0.48752967323418017 \tTime taken:  0:06:53.239792\n",
+      "Epoch  56 \t Training Loss:  0.33106483089859073  \t Validation Loss:  0.5366714083486133 \tTime taken:  0:06:50.363107\n",
+      "Epoch  57 \t Training Loss:  0.3157502224371917  \t Validation Loss:  0.49652552604675293 \tTime taken:  0:06:50.219071\n",
+      "Epoch  58 \t Training Loss:  0.32312700518177473  \t Validation Loss:  0.49135159977056364 \tTime taken:  0:06:52.234396\n",
+      "Epoch  59 \t Training Loss:  0.32884922479211853  \t Validation Loss:  0.5520870255099403 \tTime taken:  0:06:40.685577\n",
+      "Epoch  60 \t Training Loss:  0.32655611036318366  \t Validation Loss:  0.5200985421047166 \tTime taken:  0:07:17.081857\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch  61 \t Training Loss:  0.30881113818158273  \t Validation Loss:  0.4709427464167001 \tTime taken:  0:07:04.777596\n",
+      "Epoch  62 \t Training Loss:  0.31899434945586586  \t Validation Loss:  0.4534462762099725 \tTime taken:  0:07:24.489515\n",
+      "Validation Loss Decreased( 49.99665725231171 ---> 48.97219783067703 ) \t Saving The Model\n",
+      "Epoch  63 \t Training Loss:  0.30296595127362275  \t Validation Loss:  0.5448673636549048 \tTime taken:  0:07:21.031577\n",
+      "Epoch  64 \t Training Loss:  0.3128937078518388  \t Validation Loss:  0.47087927217836734 \tTime taken:  0:06:59.824642\n",
+      "Epoch  65 \t Training Loss:  0.2997312688309213  \t Validation Loss:  0.5260913667303545 \tTime taken:  0:06:54.770228\n",
+      "Epoch  66 \t Training Loss:  0.29212546828405367  \t Validation Loss:  0.48078452706061026 \tTime taken:  0:06:48.639146\n",
+      "Epoch  67 \t Training Loss:  0.29864214170400216  \t Validation Loss:  0.48277139180788287 \tTime taken:  0:07:46.980863\n",
+      "Epoch  68 \t Training Loss:  0.30616738184260717  \t Validation Loss:  0.5491416105241688 \tTime taken:  0:07:09.436473\n",
+      "Epoch  69 \t Training Loss:  0.30911092676114343  \t Validation Loss:  0.5284606596385991 \tTime taken:  0:06:46.755442\n",
+      "Epoch  70 \t Training Loss:  0.2899289741045863  \t Validation Loss:  0.4915556306088412 \tTime taken:  0:06:30.837177\n",
+      "Epoch  71 \t Training Loss:  0.28148552666728693  \t Validation Loss:  0.476806804123852 \tTime taken:  0:06:14.940365\n",
+      "Epoch  72 \t Training Loss:  0.2830918931969158  \t Validation Loss:  0.4757958865827984 \tTime taken:  0:06:29.769445\n",
+      "Epoch  73 \t Training Loss:  0.2834681684245774  \t Validation Loss:  0.4960245700484073 \tTime taken:  0:06:28.163800\n",
+      "Epoch  74 \t Training Loss:  0.271076404283741  \t Validation Loss:  0.49762628875948767 \tTime taken:  0:06:19.063000\n",
+      "Epoch  75 \t Training Loss:  0.2651180992429347  \t Validation Loss:  0.4562846920280545 \tTime taken:  0:06:23.142275\n",
+      "Epoch  76 \t Training Loss:  0.2780000296947749  \t Validation Loss:  0.5817143093380663 \tTime taken:  0:06:24.102482\n",
+      "Epoch  77 \t Training Loss:  0.26203114487419743  \t Validation Loss:  0.5183601475976132 \tTime taken:  0:06:19.916551\n",
+      "Epoch  78 \t Training Loss:  0.2656295466358247  \t Validation Loss:  0.4517332358216798 \tTime taken:  0:06:12.285122\n",
+      "Validation Loss Decreased( 48.97219783067703 ---> 48.78718946874142 ) \t Saving The Model\n",
+      "Epoch  79 \t Training Loss:  0.25902939332730096  \t Validation Loss:  0.5275237707904091 \tTime taken:  0:06:12.527802\n",
+      "Epoch  80 \t Training Loss:  0.27003029295904696  \t Validation Loss:  0.43247401852298667 \tTime taken:  0:06:06.624470\n",
+      "Validation Loss Decreased( 48.78718946874142 ---> 46.70719400048256 ) \t Saving The Model\n",
+      "Epoch  81 \t Training Loss:  0.25004229357506597  \t Validation Loss:  0.4975293082771478 \tTime taken:  0:06:34.292878\n",
+      "Epoch  82 \t Training Loss:  0.25978407818500115  \t Validation Loss:  0.4732779983293127 \tTime taken:  0:06:26.887846\n",
+      "Epoch  83 \t Training Loss:  0.25056894296996185  \t Validation Loss:  0.48900712055533574 \tTime taken:  0:06:20.755738\n",
+      "Epoch  84 \t Training Loss:  0.2545999112687465  \t Validation Loss:  0.507621691458755 \tTime taken:  0:06:27.582362\n",
+      "Epoch  85 \t Training Loss:  0.25078019064486673  \t Validation Loss:  0.49945354489264665 \tTime taken:  0:06:12.671544\n",
+      "Epoch  86 \t Training Loss:  0.26112554750094813  \t Validation Loss:  0.5432884284743557 \tTime taken:  0:06:17.521409\n",
+      "Epoch  87 \t Training Loss:  0.24813495000716354  \t Validation Loss:  0.4381025697760008 \tTime taken:  0:06:45.409486\n",
+      "Epoch  88 \t Training Loss:  0.2407905622317955  \t Validation Loss:  0.4595319861546159 \tTime taken:  0:06:38.403619\n",
+      "Epoch  89 \t Training Loss:  0.2420460439314121  \t Validation Loss:  0.5119537124065338 \tTime taken:  0:06:21.750875\n",
+      "Epoch  90 \t Training Loss:  0.23983753497536847  \t Validation Loss:  0.42354016266418276 \tTime taken:  0:06:33.566486\n",
+      "Validation Loss Decreased( 46.70719400048256 ---> 45.74233756773174 ) \t Saving The Model\n",
+      "Epoch  91 \t Training Loss:  0.24144316095507878  \t Validation Loss:  0.4537632310831988 \tTime taken:  0:06:51.406086\n",
+      "Epoch  92 \t Training Loss:  0.24106762742461718  \t Validation Loss:  0.5368691544151969 \tTime taken:  0:06:40.923380\n",
+      "Epoch  93 \t Training Loss:  0.2403287360464911  \t Validation Loss:  0.519628349415682 \tTime taken:  0:06:21.729799\n",
+      "Epoch  94 \t Training Loss:  0.23408343208332857  \t Validation Loss:  0.5315946434383039 \tTime taken:  0:06:44.949518\n",
+      "Epoch  95 \t Training Loss:  0.24215694125134335  \t Validation Loss:  0.526571866814737 \tTime taken:  0:06:35.568442\n",
+      "Epoch  96 \t Training Loss:  0.227954283937056  \t Validation Loss:  0.5385689740931546 \tTime taken:  0:06:44.634303\n",
+      "Epoch  97 \t Training Loss:  0.23911816895892168  \t Validation Loss:  0.5085669070896175 \tTime taken:  0:06:57.621963\n",
+      "Epoch  98 \t Training Loss:  0.2287096957055231  \t Validation Loss:  0.5074349546598064 \tTime taken:  0:06:54.685269\n",
+      "Epoch  99 \t Training Loss:  0.22803339804185258  \t Validation Loss:  0.4975540955999383 \tTime taken:  0:06:43.416768\n",
+      "Epoch  100 \t Training Loss:  0.23149734836719607  \t Validation Loss:  0.529600771350993 \tTime taken:  0:06:38.450953\n",
+      "Total time Taken :  11:27:26.481190\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Training with Validation\n",
+    "epochs = 100\n",
+    "min_valid_loss = np.inf\n",
+    "total_time = time.time()\n",
+    "for e in range(epochs):\n",
+    "    start_time = time.time()\n",
+    "    train_loss = 0.0\n",
+    "    for data, labels in trainloader:\n",
+    "        # Transfer Data to GPU if available\n",
+    "        if torch.cuda.is_available():\n",
+    "            data, labels = data.cuda(), labels.cuda()\n",
+    "         \n",
+    "        # Clear the gradients\n",
+    "        optimizer.zero_grad()\n",
+    "        # Forward Pass\n",
+    "        target = net(data)\n",
+    "        # Find the Loss\n",
+    "        loss = criterion(target,labels)\n",
+    "        # Calculate gradients\n",
+    "        loss.backward()\n",
+    "        # Update Weights\n",
+    "        optimizer.step()\n",
+    "        # Calculate Loss\n",
+    "        train_loss += loss.item()\n",
+    "     \n",
+    "    valid_loss = 0.0\n",
+    "    net.eval()     # Optional when not using Model Specific layer\n",
+    "    for data, labels in validloader:\n",
+    "        # Transfer Data to GPU if available\n",
+    "        if torch.cuda.is_available():\n",
+    "            data, labels = data.cuda(), labels.cuda()\n",
+    "         \n",
+    "        # Forward Pass\n",
+    "        target = net(data)\n",
+    "        # Find the Loss\n",
+    "        loss = criterion(target,labels)\n",
+    "        # Calculate Loss\n",
+    "        valid_loss += loss.item()\n",
+    " \n",
+    "    print('Epoch ',e+1, '\\t Training Loss: ',train_loss / len(trainloader),' \\t Validation Loss: ',valid_loss / len(validloader),\"\\tTime taken: \",datetime.timedelta(seconds=(time.time()-start_time)))\n",
+    "     \n",
+    "    if min_valid_loss > valid_loss:\n",
+    "        print(\"Validation Loss Decreased(\",min_valid_loss,\"--->\",valid_loss,\") \\t Saving The Model\")\n",
+    "        min_valid_loss = valid_loss\n",
+    "         \n",
+    "        # Saving State Dict\n",
+    "        torch.save(net.state_dict(), '/home/lns/research/MODEL.pth')\n",
+    "print(\"Total time Taken : \",datetime.timedelta(seconds =(time.time()-total_time)))\n",
+    "playsound(\"/home/lns/research/mixkit-small-group-cheer-and-applause-518.wav\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "id": "9949cf3f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "torch.save(net.state_dict(), '/home/lns/research/MODEL_100.pth')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "id": "93a18166",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'covid': 0, 'normal': 1, 'pneumonia': 2}"
+      ]
+     },
+     "execution_count": 28,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "trainloader.dataset.class_to_idx"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "id": "97ef3efa",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'covid': 0, 'normal': 1, 'pneumonia': 2}"
+      ]
+     },
+     "execution_count": 29,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "testloader.dataset.class_to_idx"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "id": "9c58dee7",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "{'covid': 0, 'normal': 1, 'pneumonia': 2}"
+      ]
+     },
+     "execution_count": 30,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "validloader.dataset.class_to_idx"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "15afee52",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<All keys matched successfully>"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "net = NeuralNetwork()\n",
+    "net.load_state_dict(torch.load(\"/home/lns/research/CNN/MODEL.pth\"))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "94d7c2a1",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "using GPU\n"
+     ]
+    }
+   ],
+   "source": [
+    "if torch.cuda.is_available():\n",
+    "    print(\"using GPU\")\n",
+    "    net = net.cuda()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "778b7112",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "84.64452214452214\n"
+     ]
+    }
+   ],
+   "source": [
+    "right = 0\n",
+    "wrong = 0\n",
+    "total = 0\n",
+    "for data, labels in validloader:\n",
+    "    if torch.cuda.is_available():\n",
+    "            data, labels = data.cuda(), labels.cuda()\n",
+    "    outputs = net(data)\n",
+    "    _, predicted = torch.max(outputs, 1)\n",
+    "    pred = predicted.tolist()\n",
+    "    correct = labels.tolist()\n",
+    "    total = total+len(pred)\n",
+    "    for i in range(len(pred)):\n",
+    "        if(pred[i]==correct[i]):\n",
+    "            right+=1\n",
+    "        else:\n",
+    "            wrong+=1\n",
+    "print(right*100/total)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 40,
+   "id": "7c770792",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "'attached.png'"
+      ]
+     },
+     "execution_count": 40,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "from torchviz import make_dot\n",
+    "train_images, labels = next(iter(trainloader))\n",
+    "y = net(train_images.cuda())\n",
+    "\n",
+    "make_dot(y.mean(), params=dict(net.named_parameters()), show_attrs=True, show_saved=True).render(\"attached\", format=\"png\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.8"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/acc_train.txt b/acc_train.txt
new file mode 100644
index 0000000000000000000000000000000000000000..3d69c8c4410e6c7c8f8b1a23056784a40d70f049
--- /dev/null
+++ b/acc_train.txt
@@ -0,0 +1 @@
+57.63447952270508,58.620113372802734,61.18721389770508,63.35337829589844,66.14321899414062,68.24813079833984,68.82726287841797,69.6681137084961,70.92103576660156,71.2718505859375,71.97905731201172,72.0180435180664,72.8477554321289,73.28766632080078,73.85565948486328,74.20091247558594,74.68537139892578,74.70207977294922,75.72669219970703,75.86033630371094,76.26127624511719,76.95177459716797,77.76478576660156,78.14344024658203,78.58335876464844,78.1768569946289,79.0733871459961,79.27942657470703,79.44648742675781,78.90633392333984,79.82514190673828,79.9476547241211,80.11470794677734,81.0446548461914,81.13375091552734,81.06136322021484,80.40984344482422,81.89664459228516,81.21171569824219,81.46229553222656,81.56253051757812,82.19178009033203,82.54259490966797,82.46463775634766,82.25859832763672,82.7430648803711,82.88784790039062,83.23309326171875,83.60618591308594,83.43356323242188,83.1941146850586,83.80108642578125,84.31896209716797,83.9570083618164,83.56163787841797,83.89018249511719,84.22429656982422,85.00389862060547,84.37464904785156,84.45260620117188,83.90689086914062,84.88695526123047,85.03173828125,85.09856414794922,85.43824005126953,86.14544677734375,85.88929748535156,85.60530090332031,85.79463195800781,85.70553588867188,85.5997314453125,86.20670318603516,86.26239013671875,86.69673156738281,86.94731903076172,86.84151458740234,87.11994171142578,87.38723754882812,86.9417495727539,87.29813385009766,87.08096313476562,88.06102752685547,87.55429077148438,87.9552230834961,87.74362182617188,88.2670669555664,88.03875732421875,88.0777359008789,88.84062194824219,88.97427368164062,88.66799926757812,88.37286376953125,87.93852233886719,88.6401596069336,88.92972564697266,88.56776428222656,88.36730194091797,89.08007049560547,89.38077545166016,89.64249420166016,
diff --git a/acc_valid.txt b/acc_valid.txt
new file mode 100644
index 0000000000000000000000000000000000000000..6fd12b612c5643f7b451065e575b1c514909a934
--- /dev/null
+++ b/acc_valid.txt
@@ -0,0 +1 @@
+59.32400894165039,56.75990676879883,57.808860778808594,62.150352478027344,63.05361557006836,64.16084289550781,67.83216857910156,69.17249298095703,68.88111877441406,70.39627075195312,72.17366027832031,60.75175094604492,69.23077392578125,69.66783142089844,71.47435760498047,70.54196166992188,71.99883270263672,72.7272720336914,73.42657470703125,75.29137420654297,73.31002807617188,76.45687866210938,75.81584930419922,72.98950958251953,74.09674072265625,75.64102935791016,75.14569091796875,76.42774200439453,71.32867431640625,77.41841888427734,73.89277648925781,70.97901916503906,75.58275604248047,74.6212158203125,79.22494506835938,76.36946868896484,76.95221710205078,78.8170166015625,77.36013793945312,77.53496551513672,76.63170623779297,75.08741760253906,77.47669219970703,79.77855682373047,75.72843933105469,77.01049041748047,79.05011749267578,80.12820434570312,80.24475860595703,80.88578033447266,78.84615325927734,79.13752746582031,80.7109603881836,79.16667175292969,79.4871826171875,75.75757598876953,79.63286590576172,80.5361328125,78.11771392822266,77.91375732421875,80.97319793701172,79.42890930175781,79.72028350830078,79.72028350830078,77.36013793945312,82.2552490234375,77.85547637939453,82.51748657226562,78.52564239501953,76.8939437866211,80.76923370361328,82.98368835449219,79.45804595947266,81.81818389892578,80.59440612792969,83.24592590332031,82.7505874633789,77.79720306396484,81.06060791015625,80.27389526367188,82.19696807861328,82.89627075195312,81.87645721435547,82.37179565429688,80.3613052368164,82.634033203125,82.43006896972656,83.47901916503906,83.88694763183594,83.24592590332031,82.95454406738281,74.56294250488281,83.07109832763672,82.634033203125,84.38228607177734,80.85664367675781,78.05944061279297,82.83799743652344,81.00233459472656,83.42074584960938,
diff --git a/accuracy_plot_100epochs.png b/accuracy_plot_100epochs.png
new file mode 100644
index 0000000000000000000000000000000000000000..6db1bad8e81b703d7c47f6604cd4a8c7886366f1
Binary files /dev/null and b/accuracy_plot_100epochs.png differ
diff --git a/attached.png b/attached.png
new file mode 100644
index 0000000000000000000000000000000000000000..8a1d38b88bd29f7b8ac28a323519f4b531e6ab80
Binary files /dev/null and b/attached.png differ
diff --git a/loss.png b/loss.png
new file mode 100644
index 0000000000000000000000000000000000000000..e2aa8c8b1c8d9172dc80ca827be0c9db5d8caf86
Binary files /dev/null and b/loss.png differ
diff --git a/loss_train.txt b/loss_train.txt
new file mode 100644
index 0000000000000000000000000000000000000000..c68b09a676b3a86f65f03a4c96b2ccd5ed6c91da
--- /dev/null
+++ b/loss_train.txt
@@ -0,0 +1 @@
+0.9029228049877276,0.8856767352365514,0.8471003622346925,0.8119809344888158,0.7705369327628315,0.741771535623116,0.7327183050723263,0.7158604902198731,0.6923646768323043,0.6785714580707278,0.676262518105982,0.6698277538039082,0.6627295674272279,0.6527160725347512,0.6403792584281799,0.6318943491769007,0.6226259863461464,0.6235986035383468,0.603867971520619,0.5996408582370052,0.5935023639921192,0.5775027493830254,0.5658614781032253,0.5587423649546939,0.5468475903745648,0.5547165346209265,0.5380605110812442,0.5361319128887934,0.5231574617883065,0.538012137139395,0.5200198032624781,0.5140829452469255,0.5107236489706617,0.4948225080489688,0.48714314199639386,0.48779024150607,0.5009896781733028,0.4772794395685196,0.48806979586751437,0.4783191208154281,0.4732402967308976,0.4650362822694498,0.4605097197977249,0.45593101589599117,0.4556735833424054,0.4516571824873045,0.44326862740559086,0.44160879766601685,0.42966303054336125,0.4365461183135196,0.4393774249799735,0.4241394173764885,0.41579384702088784,0.4219729055607446,0.4344543345484657,0.42892047239208986,0.4104134016189711,0.3950271703488462,0.4049163169469486,0.4085990673467994,0.4266822505543453,0.4057566263842201,0.3956257048000008,0.3930586477827771,0.38868296031717514,0.3743342116103902,0.3814578014309932,0.3842689093348182,0.37951515866874375,0.37778037140855164,0.3850887486929889,0.3706632720379431,0.363397472170165,0.35750560526109676,0.35306707385171776,0.35285318319377523,0.3440603395771514,0.33785627478753544,0.34567932541286606,0.34494500493239677,0.3418734817835254,0.3200561340725931,0.3378799657247881,0.32990650342797256,0.33767457954002233,0.315956815653,0.3222351762999737,0.32136212186589574,0.3022051092098033,0.30220048422544027,0.3055765822171847,0.30953261457065456,0.3242695900840267,0.30310345478886175,0.29942518796148676,0.30434063478667966,0.3168919574058353,0.29728707446454894,0.28510440849480334,0.28271967525431535,
\ No newline at end of file
diff --git a/loss_train_old.txt b/loss_train_old.txt
new file mode 100644
index 0000000000000000000000000000000000000000..fbc7ce5c44b0f46ce191f5d2f90a78358f1a464c
--- /dev/null
+++ b/loss_train_old.txt
@@ -0,0 +1 @@
+0.8441031010470529,0.8497029065653898,0.8115832861782848,0.7646360919419406,0.719308659920226,0.6882924675077632,0.6532480342556601,0.6339848736382049,0.6106657782380563,0.5946273674405571,0.5670546730575354,0.5607690397865962,0.5532481207417838,0.5381492998019077,0.5263509273906981,0.5081568959344557,0.508065088233654,0.49568649116849556,0.4950284741252013,0.4833472721494626,0.4815491572076428,0.4774224379950244,0.47000723394254845,0.4641701220062332,0.44393498013200966,0.45194079627053463,0.43756033331695676,0.43203936309378216,0.4270447193091546,0.42842536836700595,0.42243523628491425,0.41496449160942994,0.4083881352258765,0.40162385214606056,0.40554388316915085,0.40995896380880603,0.3954985524541226,0.3878848762453898,0.3920577052116826,0.39466455229220615,0.3875494974705836,0.38377476972190366,0.37667413472967304,0.38178481239879475,0.36876419488934503,0.37044498636184825,0.36230879056982807,0.3642571686321627,0.36602410338009184,0.3572306552146008,0.3604762159368914,0.3578577160268374,0.34325537194862316,0.3643265035327362,0.34488123731579684,0.3327634610991547,0.34904009547840426,0.3381150304677262,0.32971405522708874,0.339250985155071,0.3404549351097017,0.33967561034512694,0.33515092588799156,0.3294831759876747,0.33311509473276313,0.32233433610341256,0.32315585012480186,0.3126404246153391,0.31820504642024205,0.31651253203281027,0.32192365567375353,0.305074136548986,0.3147186248063825,0.30745019732326595,0.3091972486750371,0.30780382845820725,0.3116833662371273,0.300667689849987,0.3002038650970528,0.30369554264335963,0.28773324064019584,0.29643661152922374,0.29064890814513183,0.2886795782538104,0.2908970232432087,0.28397169749697915,0.31978519640161074,0.2970599836426909,0.2949577732808024,0.2793661203181398,0.3123172843504859,0.2924956642023787,0.2867284516291018,0.27161509993796545,0.2700741979347515,0.2816199139907848,0.28585991819483647,0.26357225005420437,0.2669716270039857,0.26669022520545166,
\ No newline at end of file
diff --git a/loss_valid.txt b/loss_valid.txt
new file mode 100644
index 0000000000000000000000000000000000000000..fb193055b8d2fab2773294318fb9bebd8274845f
--- /dev/null
+++ b/loss_valid.txt
@@ -0,0 +1 @@
+0.8944440715842776,0.9128290994299783,0.9120684911807379,0.8293078966714718,0.8124735769298341,0.8236339531011052,0.7604639452916605,0.7314019556398745,0.7229070735198481,0.7380137079291873,0.676862828709461,0.8797201442497747,0.7323361265438574,0.7401043899633266,0.6886368589820685,0.7204805187605046,0.7040614678903863,0.6673177524849221,0.6556077698866526,0.6324033361894114,0.6485145138921561,0.6033395292858282,0.6150209567061177,0.6809103475124748,0.6453374088914307,0.6095044458353961,0.6276791799399588,0.6042463669070491,0.6972870423837945,0.5839059344596333,0.6576835181978014,0.7197631487139949,0.6104853100798748,0.6035054706864886,0.5258092273164678,0.594145884944333,0.5729684631029764,0.5359699957900577,0.5759746941427389,0.5715127614913164,0.5820767724955523,0.6060397622210009,0.5849179564112866,0.5505736610955663,0.6444224847687615,0.594107069351055,0.5104974389628127,0.5294558451407485,0.5046062285977381,0.49573717942392387,0.542883567236088,0.52265090384969,0.5224103926232567,0.5424923496665778,0.5544744128430331,0.6630421954172628,0.5150271084297586,0.5091825410447739,0.5659814819141671,0.6131980065946225,0.5203534404712694,0.5222419394111192,0.5357174716751885,0.5213135158022245,0.5845458168122504,0.47417158150562533,0.6384360061751472,0.48013403570210494,0.5551777308185896,0.6369037684743051,0.5161193176000206,0.4723347807648005,0.5204819073831594,0.4980683869647759,0.5271965176970871,0.46226683879892033,0.4683791854315334,0.5578625021433389,0.48196811063422096,0.4852829767322099,0.4769098451016126,0.4822914483094657,0.496175997649078,0.4860944132562037,0.5501075481513032,0.43915262694160145,0.5052717746821819,0.48401865404513145,0.46346047189500594,0.5105993631123392,0.5244620022950349,1.0839747182197041,0.45126490432907035,0.47017240883023653,0.4341367942591508,0.46439169802599484,0.6083490194545852,0.4969119317829609,0.5444494509310635,0.47462088207679765,
\ No newline at end of file
diff --git a/loss_valid_old.txt b/loss_valid_old.txt
new file mode 100644
index 0000000000000000000000000000000000000000..b41a82c1bce7ac850ad8ce4b6017b808301ca071
--- /dev/null
+++ b/loss_valid_old.txt
@@ -0,0 +1 @@
+0.8674659265412225,0.858217479178199,0.8849059243996938,0.7732666281086428,0.7553680486701153,0.7712281315966889,0.6456847621334924,0.6636026140164446,0.6464473497536447,0.6138652024997605,0.6630741401954934,0.6868172595622363,0.6112280119624403,0.5809305067415591,0.5705836318709232,0.5947056367165513,0.5841603911033383,0.5833484555284182,0.5582468676622268,0.5394910522909077,0.5768332649712209,0.5272311384755152,0.5224805935113518,0.5676913777435267,0.5637580210136043,0.5722048529596241,0.5721317226136172,0.5327197342283196,0.5241014608354481,0.5545626297869064,0.5789179211413419,0.6381655058099164,0.6047858819365501,0.5051082772789178,0.5353965535070058,0.5032828458481364,0.52293625542963,0.5478894364226747,0.569466141362985,0.5664596587971404,0.5089240012069544,0.5031804796653213,0.5074075361092886,0.49775540945982494,0.49973885732254497,0.49891040025761835,0.5088043982783953,0.5234297762314478,0.5290909115638998,0.5501965762598923,0.5543567792133048,0.4903783861685682,0.5274432320147753,0.5330567766946775,0.527078941050503,0.5486167269172492,0.511558510363102,0.5387480808077035,0.5009127659378229,0.5604935434681398,0.5344267963535256,0.47556024444875894,0.4781123712244961,0.5586284661182651,0.49362038600224034,0.4850667109368024,0.5388349147030601,0.5314383802038652,0.5445920531810434,0.5347450921932856,0.5099611518283685,0.45728773844462856,0.5727233766681619,0.5070217447010456,0.4620397361340346,0.507594204335301,0.5185853809946113,0.47204770768682164,0.5617581164395368,0.5066448463885872,0.6637866491520846,0.4840451491375764,0.49376404088818365,0.5022408930515802,0.5062403042835218,0.5060996104169775,0.483770119647185,0.491223679786479,0.5039577412384527,0.46893033336985995,0.5174671373916445,0.5316194277946595,0.5071840846428165,0.5141122157650965,0.4986010122768305,0.48562143922404005,0.526746118993119,0.5276217797288189,0.5089392931786952,0.49559183460142875,
\ No newline at end of file
diff --git a/train_loss_100epochs.png b/train_loss_100epochs.png
new file mode 100644
index 0000000000000000000000000000000000000000..33f529dab239f74fb16b8835893857c68496d2e9
Binary files /dev/null and b/train_loss_100epochs.png differ