From 4d79ee17857e9be3f219b9529c8639a7b38a7559 Mon Sep 17 00:00:00 2001
From: pmarin72 <75830392+pmarin72@users.noreply.github.com>
Date: Thu, 30 Nov 2023 18:04:53 +0100
Subject: [PATCH] some improvements on ex 2

---
 TD2 Deep Learning.ipynb | 329 +++++++++++++++++++++-------------------
 1 file changed, 176 insertions(+), 153 deletions(-)

diff --git a/TD2 Deep Learning.ipynb b/TD2 Deep Learning.ipynb
index d4132b0..ad0ccd3 100644
--- a/TD2 Deep Learning.ipynb	
+++ b/TD2 Deep Learning.ipynb	
@@ -33,7 +33,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 2,
    "id": "330a42f5",
    "metadata": {},
    "outputs": [
@@ -55,7 +55,7 @@
       "Requirement already satisfied: charset-normalizer<4,>=2 in c:\\users\\marin\\anaconda3\\lib\\site-packages (from requests->torchvision) (2.0.4)\n",
       "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\marin\\anaconda3\\lib\\site-packages (from requests->torchvision) (3.4)\n",
       "Requirement already satisfied: urllib3<3,>=1.21.1 in c:\\users\\marin\\anaconda3\\lib\\site-packages (from requests->torchvision) (1.26.16)\n",
-      "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\marin\\anaconda3\\lib\\site-packages (from requests->torchvision) (2023.7.22)\n",
+      "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\marin\\anaconda3\\lib\\site-packages (from requests->torchvision) (2023.11.17)\n",
       "Requirement already satisfied: mpmath>=0.19 in c:\\users\\marin\\anaconda3\\lib\\site-packages (from sympy->torch) (1.3.0)\n",
       "Note: you may need to restart the kernel to use updated packages.\n"
      ]
@@ -76,7 +76,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 3,
    "id": "b1950f0a",
    "metadata": {},
    "outputs": [
@@ -84,34 +84,34 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "tensor([[ 0.5435, -0.4544, -0.1205,  0.4673,  0.1775, -0.7653, -0.0387,  0.0637,\n",
-      "         -0.1312, -1.8514],\n",
-      "        [-0.7451, -0.0433,  0.5996, -0.1318,  0.7856,  0.5377, -1.0717,  0.4718,\n",
-      "         -0.4624,  1.1403],\n",
-      "        [ 1.4221,  0.4673, -0.1187,  0.9117, -1.7922, -0.2721,  0.0516,  1.3558,\n",
-      "         -0.8833, -1.0339],\n",
-      "        [-1.4286, -1.3892, -1.4825,  1.7989,  0.1235,  1.3108, -1.5860,  0.6306,\n",
-      "          0.3286, -0.5756],\n",
-      "        [ 0.2860, -0.4381, -0.8007, -0.9251, -2.4581,  0.6307, -0.9971, -1.0066,\n",
-      "          0.8453, -0.6403],\n",
-      "        [-1.7802, -0.5362,  0.5685,  0.0599,  0.1256,  0.2542, -0.4363, -0.9823,\n",
-      "          0.4746,  1.6888],\n",
-      "        [-1.6597,  1.0951,  0.9582, -1.5032,  1.1591,  0.8159, -1.4805, -0.5566,\n",
-      "          0.4475, -1.6350],\n",
-      "        [-0.4390, -0.5932,  0.6092,  1.7203,  0.4294, -2.0137,  1.5183, -0.0681,\n",
-      "          2.6924,  2.2244],\n",
-      "        [-0.7404,  1.5136,  0.6477, -0.4592,  1.6904, -0.4243,  1.2477, -0.7878,\n",
-      "          0.4548,  0.4966],\n",
-      "        [-0.5082,  0.0487,  0.4923, -0.0613, -1.0030,  0.3108, -0.7571, -0.3653,\n",
-      "          0.4734,  0.3244],\n",
-      "        [-1.5140,  0.9956, -0.5122,  0.2580, -0.4591, -0.2065,  1.3851,  0.2364,\n",
-      "          0.5900, -0.7037],\n",
-      "        [ 1.0271, -1.3211,  0.0545,  0.5302, -0.2711,  0.6698,  2.2225,  0.2634,\n",
-      "         -0.2574, -1.6689],\n",
-      "        [ 0.3319, -1.2073,  0.3785,  1.5544,  0.4043, -1.0159,  0.1956, -1.7744,\n",
-      "          0.3340,  0.8643],\n",
-      "        [ 2.8115, -0.5446,  0.5140, -0.3576, -1.2501, -0.2065, -0.3383,  0.2077,\n",
-      "         -0.2065, -1.2150]])\n",
+      "tensor([[-0.3740, -0.1337, -1.1278,  1.0594, -0.2462, -1.1751, -0.1005,  1.1031,\n",
+      "          0.5354, -0.1985],\n",
+      "        [ 0.3067, -1.0501, -1.3315, -2.7529, -1.8386, -1.0362, -0.8983,  0.4816,\n",
+      "         -0.7046,  0.2330],\n",
+      "        [-0.1206,  0.8951, -0.6436, -0.3075,  0.9056, -0.6875, -0.7694, -0.2017,\n",
+      "          0.9787, -0.5610],\n",
+      "        [-2.8668, -0.3878,  0.7541, -1.1662,  0.4237,  1.1266, -0.3558,  0.1105,\n",
+      "         -1.0558, -1.8606],\n",
+      "        [ 0.3507, -0.8552, -0.9354,  0.3753,  1.2805, -0.3248, -0.4088, -0.5620,\n",
+      "          0.1417,  1.0160],\n",
+      "        [-0.7317, -3.4209, -0.4999, -0.1847,  0.1923,  0.7617,  0.2245, -1.9357,\n",
+      "          0.0595,  2.1604],\n",
+      "        [ 0.1924, -1.1935,  0.9019,  1.2187, -1.7188, -0.7759, -1.3686,  0.2335,\n",
+      "          0.3900,  0.9486],\n",
+      "        [ 1.4248,  0.9080,  0.3575,  1.9698, -0.3119,  1.1467,  1.9559,  1.9424,\n",
+      "         -0.1275, -0.0842],\n",
+      "        [ 0.9739,  1.7380, -0.3301, -0.3293,  0.0384, -0.9268, -1.0350, -0.6020,\n",
+      "         -1.3752,  0.6666],\n",
+      "        [-0.3124, -0.3678, -1.8143, -0.0260, -0.6726, -0.6671, -0.1143,  0.5844,\n",
+      "         -0.8527,  0.7353],\n",
+      "        [ 1.0313,  0.3691,  0.8323, -0.4683, -1.4537, -0.5249,  2.0043,  0.0210,\n",
+      "          0.7745,  1.2210],\n",
+      "        [ 0.7038, -0.3010,  1.8068,  1.0899,  1.9105,  0.3594,  0.7311,  0.8623,\n",
+      "          0.5980, -1.0860],\n",
+      "        [ 1.0422,  1.6860,  0.1746, -0.7042,  0.9685,  1.8207,  0.5156, -0.8631,\n",
+      "          0.8923,  0.5413],\n",
+      "        [-1.7911,  0.4318, -0.6459, -1.6303, -1.9783,  0.9335, -0.2233,  0.9090,\n",
+      "         -1.0225, -0.0549]])\n",
       "AlexNet(\n",
       "  (features): Sequential(\n",
       "    (0): Conv2d(3, 64, kernel_size=(11, 11), stride=(4, 4), padding=(2, 2))\n",
@@ -181,7 +181,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 4,
    "id": "6e18f2fd",
    "metadata": {},
    "outputs": [
@@ -215,7 +215,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 5,
    "id": "462666a2",
    "metadata": {},
    "outputs": [
@@ -296,7 +296,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 7,
    "id": "317bf070",
    "metadata": {},
    "outputs": [
@@ -358,9 +358,17 @@
     "Loss function and training using SGD (Stochastic Gradient Descent) optimizer"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "083fc25a",
+   "metadata": {},
+   "source": [
+    "We add a counter to do an early stopping if overfit occur. If validation loss doesn't decrease for 3 consecutives epochs we stop the training.  "
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 49,
+   "execution_count": 8,
    "id": "4b53f229",
    "metadata": {},
    "outputs": [
@@ -368,33 +376,34 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Epoch: 0 \tTraining Loss: 43.971403 \tValidation Loss: 38.781364\n",
-      "Validation loss decreased (inf --> 38.781364).  Saving model ...\n",
-      "Epoch: 1 \tTraining Loss: 34.860234 \tValidation Loss: 32.887555\n",
-      "Validation loss decreased (38.781364 --> 32.887555).  Saving model ...\n",
-      "Epoch: 2 \tTraining Loss: 30.713754 \tValidation Loss: 30.105906\n",
-      "Validation loss decreased (32.887555 --> 30.105906).  Saving model ...\n",
-      "Epoch: 3 \tTraining Loss: 28.186929 \tValidation Loss: 28.076022\n",
-      "Validation loss decreased (30.105906 --> 28.076022).  Saving model ...\n",
-      "Epoch: 4 \tTraining Loss: 26.351580 \tValidation Loss: 26.367747\n",
-      "Validation loss decreased (28.076022 --> 26.367747).  Saving model ...\n",
-      "Epoch: 5 \tTraining Loss: 24.947239 \tValidation Loss: 26.368494\n",
-      "Epoch: 6 \tTraining Loss: 23.778135 \tValidation Loss: 24.579198\n",
-      "Validation loss decreased (26.367747 --> 24.579198).  Saving model ...\n",
-      "Epoch: 7 \tTraining Loss: 22.707515 \tValidation Loss: 24.203169\n",
-      "Validation loss decreased (24.579198 --> 24.203169).  Saving model ...\n",
-      "Epoch: 8 \tTraining Loss: 21.805590 \tValidation Loss: 23.090124\n",
-      "Validation loss decreased (24.203169 --> 23.090124).  Saving model ...\n",
-      "Epoch: 9 \tTraining Loss: 21.043298 \tValidation Loss: 22.905686\n",
-      "Validation loss decreased (23.090124 --> 22.905686).  Saving model ...\n",
-      "Epoch: 10 \tTraining Loss: 20.250996 \tValidation Loss: 23.170775\n",
-      "Epoch: 11 \tTraining Loss: 19.621161 \tValidation Loss: 22.586260\n",
-      "Validation loss decreased (22.905686 --> 22.586260).  Saving model ...\n",
-      "Epoch: 12 \tTraining Loss: 18.967947 \tValidation Loss: 22.084914\n",
-      "Validation loss decreased (22.586260 --> 22.084914).  Saving model ...\n",
-      "Epoch: 13 \tTraining Loss: 18.294242 \tValidation Loss: 22.170993\n",
-      "Epoch: 14 \tTraining Loss: 17.742965 \tValidation Loss: 22.235334\n",
-      "Epoch: 15 \tTraining Loss: 17.191995 \tValidation Loss: 22.188067\n",
+      "Epoch: 0 \tTraining Loss: 41.928474 \tValidation Loss: 36.149142\n",
+      "Validation loss decreased (inf --> 36.149142).  Saving model ...\n",
+      "Epoch: 1 \tTraining Loss: 33.406021 \tValidation Loss: 31.535990\n",
+      "Validation loss decreased (36.149142 --> 31.535990).  Saving model ...\n",
+      "Epoch: 2 \tTraining Loss: 29.986645 \tValidation Loss: 29.026595\n",
+      "Validation loss decreased (31.535990 --> 29.026595).  Saving model ...\n",
+      "Epoch: 3 \tTraining Loss: 27.796806 \tValidation Loss: 28.266311\n",
+      "Validation loss decreased (29.026595 --> 28.266311).  Saving model ...\n",
+      "Epoch: 4 \tTraining Loss: 26.246026 \tValidation Loss: 26.360779\n",
+      "Validation loss decreased (28.266311 --> 26.360779).  Saving model ...\n",
+      "Epoch: 5 \tTraining Loss: 24.983256 \tValidation Loss: 25.554680\n",
+      "Validation loss decreased (26.360779 --> 25.554680).  Saving model ...\n",
+      "Epoch: 6 \tTraining Loss: 23.907650 \tValidation Loss: 24.931439\n",
+      "Validation loss decreased (25.554680 --> 24.931439).  Saving model ...\n",
+      "Epoch: 7 \tTraining Loss: 22.924509 \tValidation Loss: 24.198110\n",
+      "Validation loss decreased (24.931439 --> 24.198110).  Saving model ...\n",
+      "Epoch: 8 \tTraining Loss: 22.127981 \tValidation Loss: 23.928127\n",
+      "Validation loss decreased (24.198110 --> 23.928127).  Saving model ...\n",
+      "Epoch: 9 \tTraining Loss: 21.372160 \tValidation Loss: 23.599568\n",
+      "Validation loss decreased (23.928127 --> 23.599568).  Saving model ...\n",
+      "Epoch: 10 \tTraining Loss: 20.641422 \tValidation Loss: 23.933316\n",
+      "Epoch: 11 \tTraining Loss: 19.919589 \tValidation Loss: 22.942826\n",
+      "Validation loss decreased (23.599568 --> 22.942826).  Saving model ...\n",
+      "Epoch: 12 \tTraining Loss: 19.266797 \tValidation Loss: 22.437601\n",
+      "Validation loss decreased (22.942826 --> 22.437601).  Saving model ...\n",
+      "Epoch: 13 \tTraining Loss: 18.638509 \tValidation Loss: 22.854837\n",
+      "Epoch: 14 \tTraining Loss: 18.039349 \tValidation Loss: 23.089719\n",
+      "Epoch: 15 \tTraining Loss: 17.516967 \tValidation Loss: 22.803356\n",
       "Early stopping after 15 epochss.\n"
      ]
     }
@@ -496,15 +505,14 @@
    "metadata": {},
    "outputs": [
     {
-     "ename": "NameError",
-     "evalue": "name 'train_loss_list' is not defined",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
-      "Cell \u001b[1;32mIn[9], line 3\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mmatplotlib\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpyplot\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mplt\u001b[39;00m\n\u001b[1;32m----> 3\u001b[0m plt\u001b[38;5;241m.\u001b[39mplot(\u001b[38;5;28mrange\u001b[39m(\u001b[38;5;28mlen\u001b[39m(train_loss_list)), train_loss_list, label\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtrain_loss\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m      4\u001b[0m plt\u001b[38;5;241m.\u001b[39mplot(\u001b[38;5;28mrange\u001b[39m(\u001b[38;5;28mlen\u001b[39m(valid_loss_list)), valid_loss_list, label\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mvalidation_loss\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m      5\u001b[0m plt\u001b[38;5;241m.\u001b[39mxlabel(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mEpoch\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
-      "\u001b[1;31mNameError\u001b[0m: name 'train_loss_list' is not defined"
-     ]
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
@@ -529,7 +537,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 51,
+   "execution_count": 11,
    "id": "e93efdfc",
    "metadata": {},
    "outputs": [
@@ -537,20 +545,20 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Test Loss: 21.919417\n",
+      "Test Loss: 22.254374\n",
       "\n",
       "Test Accuracy of airplane: 67% (679/1000)\n",
-      "Test Accuracy of automobile: 78% (782/1000)\n",
-      "Test Accuracy of  bird: 59% (596/1000)\n",
-      "Test Accuracy of   cat: 39% (398/1000)\n",
-      "Test Accuracy of  deer: 55% (554/1000)\n",
-      "Test Accuracy of   dog: 48% (482/1000)\n",
-      "Test Accuracy of  frog: 67% (678/1000)\n",
-      "Test Accuracy of horse: 60% (605/1000)\n",
-      "Test Accuracy of  ship: 71% (713/1000)\n",
-      "Test Accuracy of truck: 68% (687/1000)\n",
+      "Test Accuracy of automobile: 75% (753/1000)\n",
+      "Test Accuracy of  bird: 37% (377/1000)\n",
+      "Test Accuracy of   cat: 35% (356/1000)\n",
+      "Test Accuracy of  deer: 63% (638/1000)\n",
+      "Test Accuracy of   dog: 46% (465/1000)\n",
+      "Test Accuracy of  frog: 73% (732/1000)\n",
+      "Test Accuracy of horse: 69% (691/1000)\n",
+      "Test Accuracy of  ship: 73% (734/1000)\n",
+      "Test Accuracy of truck: 66% (663/1000)\n",
       "\n",
-      "Test Accuracy (Overall): 61% (6174/10000)\n"
+      "Test Accuracy (Overall): 60% (6088/10000)\n"
      ]
     }
    ],
@@ -638,7 +646,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 31,
+   "execution_count": 13,
    "id": "43fff7d9",
    "metadata": {},
    "outputs": [
@@ -686,11 +694,11 @@
     "\n",
     "\n",
     "# create a complete CNN\n",
-    "model2 = Net()\n",
-    "print(model2)\n",
+    "model = Net()\n",
+    "print(model)\n",
     "# move tensors to GPU if CUDA is available\n",
     "if train_on_gpu:\n",
-    "    model2.cuda()"
+    "    model.cuda()"
    ]
   },
   {
@@ -703,7 +711,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
+   "execution_count": 14,
    "id": "40638ce8",
    "metadata": {},
    "outputs": [
@@ -711,50 +719,54 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Epoch: 0 \tTraining Loss: 45.992308 \tValidation Loss: 45.734079\n",
-      "Validation loss decreased (inf --> 45.734079).  Saving model ...\n",
-      "Epoch: 1 \tTraining Loss: 42.861547 \tValidation Loss: 39.263893\n",
-      "Validation loss decreased (45.734079 --> 39.263893).  Saving model ...\n",
-      "Epoch: 2 \tTraining Loss: 36.771661 \tValidation Loss: 34.267659\n",
-      "Validation loss decreased (39.263893 --> 34.267659).  Saving model ...\n",
-      "Epoch: 3 \tTraining Loss: 33.214925 \tValidation Loss: 32.355083\n",
-      "Validation loss decreased (34.267659 --> 32.355083).  Saving model ...\n",
-      "Epoch: 4 \tTraining Loss: 31.095436 \tValidation Loss: 30.169813\n",
-      "Validation loss decreased (32.355083 --> 30.169813).  Saving model ...\n",
-      "Epoch: 5 \tTraining Loss: 29.427925 \tValidation Loss: 28.723804\n",
-      "Validation loss decreased (30.169813 --> 28.723804).  Saving model ...\n",
-      "Epoch: 6 \tTraining Loss: 27.979727 \tValidation Loss: 27.368319\n",
-      "Validation loss decreased (28.723804 --> 27.368319).  Saving model ...\n",
-      "Epoch: 7 \tTraining Loss: 26.560045 \tValidation Loss: 26.531596\n",
-      "Validation loss decreased (27.368319 --> 26.531596).  Saving model ...\n",
-      "Epoch: 8 \tTraining Loss: 25.233062 \tValidation Loss: 25.151196\n",
-      "Validation loss decreased (26.531596 --> 25.151196).  Saving model ...\n",
-      "Epoch: 9 \tTraining Loss: 23.933243 \tValidation Loss: 24.118065\n",
-      "Validation loss decreased (25.151196 --> 24.118065).  Saving model ...\n",
-      "Epoch: 10 \tTraining Loss: 22.824302 \tValidation Loss: 23.136555\n",
-      "Validation loss decreased (24.118065 --> 23.136555).  Saving model ...\n",
-      "Epoch: 11 \tTraining Loss: 21.638535 \tValidation Loss: 22.276559\n",
-      "Validation loss decreased (23.136555 --> 22.276559).  Saving model ...\n",
-      "Epoch: 12 \tTraining Loss: 20.643672 \tValidation Loss: 21.490677\n",
-      "Validation loss decreased (22.276559 --> 21.490677).  Saving model ...\n",
-      "Epoch: 13 \tTraining Loss: 19.723929 \tValidation Loss: 20.878862\n",
-      "Validation loss decreased (21.490677 --> 20.878862).  Saving model ...\n",
-      "Epoch: 14 \tTraining Loss: 18.881656 \tValidation Loss: 20.151909\n",
-      "Validation loss decreased (20.878862 --> 20.151909).  Saving model ...\n",
-      "Epoch: 15 \tTraining Loss: 18.077398 \tValidation Loss: 22.761932\n",
-      "Epoch: 16 \tTraining Loss: 17.244630 \tValidation Loss: 20.172645\n",
-      "Epoch: 17 \tTraining Loss: 16.708238 \tValidation Loss: 19.282629\n",
-      "Validation loss decreased (20.151909 --> 19.282629).  Saving model ...\n",
-      "Epoch: 18 \tTraining Loss: 16.049521 \tValidation Loss: 19.141060\n",
-      "Validation loss decreased (19.282629 --> 19.141060).  Saving model ...\n",
-      "Epoch: 19 \tTraining Loss: 15.306451 \tValidation Loss: 18.852022\n",
-      "Validation loss decreased (19.141060 --> 18.852022).  Saving model ...\n",
-      "Epoch: 20 \tTraining Loss: 14.730358 \tValidation Loss: 18.173483\n",
-      "Validation loss decreased (18.852022 --> 18.173483).  Saving model ...\n",
-      "Epoch: 21 \tTraining Loss: 14.233885 \tValidation Loss: 18.950112\n",
-      "Epoch: 22 \tTraining Loss: 13.598710 \tValidation Loss: 19.136868\n",
-      "Epoch: 23 \tTraining Loss: 13.122754 \tValidation Loss: 18.415266\n",
-      "Early stopping after 23 epochss.\n"
+      "Epoch: 0 \tTraining Loss: 45.923728 \tValidation Loss: 45.169865\n",
+      "Validation loss decreased (inf --> 45.169865).  Saving model ...\n",
+      "Epoch: 1 \tTraining Loss: 41.546925 \tValidation Loss: 38.793506\n",
+      "Validation loss decreased (45.169865 --> 38.793506).  Saving model ...\n",
+      "Epoch: 2 \tTraining Loss: 36.476450 \tValidation Loss: 35.295533\n",
+      "Validation loss decreased (38.793506 --> 35.295533).  Saving model ...\n",
+      "Epoch: 3 \tTraining Loss: 33.383830 \tValidation Loss: 33.026579\n",
+      "Validation loss decreased (35.295533 --> 33.026579).  Saving model ...\n",
+      "Epoch: 4 \tTraining Loss: 31.333926 \tValidation Loss: 30.475611\n",
+      "Validation loss decreased (33.026579 --> 30.475611).  Saving model ...\n",
+      "Epoch: 5 \tTraining Loss: 29.614357 \tValidation Loss: 29.033134\n",
+      "Validation loss decreased (30.475611 --> 29.033134).  Saving model ...\n",
+      "Epoch: 6 \tTraining Loss: 28.001916 \tValidation Loss: 28.396793\n",
+      "Validation loss decreased (29.033134 --> 28.396793).  Saving model ...\n",
+      "Epoch: 7 \tTraining Loss: 26.720634 \tValidation Loss: 26.933168\n",
+      "Validation loss decreased (28.396793 --> 26.933168).  Saving model ...\n",
+      "Epoch: 8 \tTraining Loss: 25.416131 \tValidation Loss: 25.584761\n",
+      "Validation loss decreased (26.933168 --> 25.584761).  Saving model ...\n",
+      "Epoch: 9 \tTraining Loss: 24.044020 \tValidation Loss: 25.319066\n",
+      "Validation loss decreased (25.584761 --> 25.319066).  Saving model ...\n",
+      "Epoch: 10 \tTraining Loss: 22.966510 \tValidation Loss: 24.229066\n",
+      "Validation loss decreased (25.319066 --> 24.229066).  Saving model ...\n",
+      "Epoch: 11 \tTraining Loss: 21.892974 \tValidation Loss: 23.542444\n",
+      "Validation loss decreased (24.229066 --> 23.542444).  Saving model ...\n",
+      "Epoch: 12 \tTraining Loss: 21.116193 \tValidation Loss: 22.082797\n",
+      "Validation loss decreased (23.542444 --> 22.082797).  Saving model ...\n",
+      "Epoch: 13 \tTraining Loss: 20.105640 \tValidation Loss: 21.183286\n",
+      "Validation loss decreased (22.082797 --> 21.183286).  Saving model ...\n",
+      "Epoch: 14 \tTraining Loss: 19.280560 \tValidation Loss: 21.298658\n",
+      "Epoch: 15 \tTraining Loss: 18.506203 \tValidation Loss: 20.982773\n",
+      "Validation loss decreased (21.183286 --> 20.982773).  Saving model ...\n",
+      "Epoch: 16 \tTraining Loss: 17.824514 \tValidation Loss: 20.400616\n",
+      "Validation loss decreased (20.982773 --> 20.400616).  Saving model ...\n",
+      "Epoch: 17 \tTraining Loss: 17.116147 \tValidation Loss: 19.853483\n",
+      "Validation loss decreased (20.400616 --> 19.853483).  Saving model ...\n",
+      "Epoch: 18 \tTraining Loss: 16.443002 \tValidation Loss: 20.076364\n",
+      "Epoch: 19 \tTraining Loss: 15.684523 \tValidation Loss: 19.175958\n",
+      "Validation loss decreased (19.853483 --> 19.175958).  Saving model ...\n",
+      "Epoch: 20 \tTraining Loss: 15.149510 \tValidation Loss: 19.180701\n",
+      "Epoch: 21 \tTraining Loss: 14.661278 \tValidation Loss: 18.647935\n",
+      "Validation loss decreased (19.175958 --> 18.647935).  Saving model ...\n",
+      "Epoch: 22 \tTraining Loss: 14.030736 \tValidation Loss: 18.935119\n",
+      "Epoch: 23 \tTraining Loss: 13.360740 \tValidation Loss: 18.588247\n",
+      "Validation loss decreased (18.647935 --> 18.588247).  Saving model ...\n",
+      "Epoch: 24 \tTraining Loss: 12.932477 \tValidation Loss: 19.555995\n",
+      "Epoch: 25 \tTraining Loss: 12.393839 \tValidation Loss: 19.220007\n",
+      "Epoch: 26 \tTraining Loss: 11.902525 \tValidation Loss: 19.395309\n",
+      "Early stopping after 26 epochss.\n"
      ]
     }
    ],
@@ -762,7 +774,7 @@
     "import torch.optim as optim\n",
     "\n",
     "criterion = nn.CrossEntropyLoss()  # specify loss function\n",
-    "optimizer = optim.SGD(model2.parameters(), lr=0.01)  # specify optimizer\n",
+    "optimizer = optim.SGD(model.parameters(), lr=0.01)  # specify optimizer\n",
     "\n",
     "n_epochs = 30  # number of epochs to train the model\n",
     "train_loss_list = []  # list to store loss to visualize\n",
@@ -777,7 +789,7 @@
     "    valid_loss = 0.0\n",
     "\n",
     "    # Train the model\n",
-    "    model2.train()\n",
+    "    model.train()\n",
     "    for data, target in train_loader:\n",
     "        # Move tensors to GPU if CUDA is available\n",
     "        if train_on_gpu:\n",
@@ -785,7 +797,7 @@
     "        # Clear the gradients of all optimized variables\n",
     "        optimizer.zero_grad()\n",
     "        # Forward pass: compute predicted outputs by passing inputs to the model\n",
-    "        output = model2(data)\n",
+    "        output = model(data)\n",
     "        # Calculate the batch loss\n",
     "        loss = criterion(output, target)\n",
     "        # Backward pass: compute gradient of the loss with respect to model parameters\n",
@@ -796,13 +808,13 @@
     "        train_loss += loss.item() * data.size(0)\n",
     "\n",
     "    # Validate the model\n",
-    "    model2.eval()\n",
+    "    model.eval()\n",
     "    for data, target in valid_loader:\n",
     "        # Move tensors to GPU if CUDA is available\n",
     "        if train_on_gpu:\n",
     "            data, target = data.cuda(), target.cuda()\n",
     "        # Forward pass: compute predicted outputs by passing inputs to the model\n",
-    "        output = model2(data)\n",
+    "        output = model(data)\n",
     "        # Calculate the batch loss\n",
     "        loss = criterion(output, target)\n",
     "        # Update average validation loss\n",
@@ -829,7 +841,7 @@
     "                valid_loss_min, valid_loss\n",
     "            )\n",
     "        )\n",
-    "        torch.save(model2.state_dict(), \"model_cifar_exo1.pt\")\n",
+    "        torch.save(model.state_dict(), \"model_cifar_exo1.pt\")\n",
     "        valid_loss_min = valid_loss\n",
     "        patience_counter = 0\n",
     "    else:\n",
@@ -850,13 +862,13 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 33,
+   "execution_count": 15,
    "id": "206bc2a1",
    "metadata": {},
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjMAAAHFCAYAAAAHcXhbAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAB2z0lEQVR4nO3dd1yVdf/H8ddhb2TIUkQcONHcqzT3ztSy0krbVvbLynamTdv33X1bdje0pWm2y5wp5hY37omigiioLGVevz+OHCEFBYHDgffz8TgPD9f1Pdf1ORyRt9f1HSbDMAxEREREbJSdtQsQERERuRYKMyIiImLTFGZERETEpinMiIiIiE1TmBERERGbpjAjIiIiNk1hRkRERGyawoyIiIjYNIUZERERsWkKM2KzvvzyS0wmk+Xh4OBA7dq1ueeeezh27FiZnisrK4uxY8cSHByMvb091113XZkeX4o2Z84cmjVrhqurKyaTiS1btly23c6dO5k8eTKxsbGX7Lvxxhtp3rx5+RYKjBkzBpPJRLNmzcjNzb1kv8lkYty4ceVex5Xk/8yMGTPmsvtfffVVS5vLfT9La8yYMdStW7dUr73xxhu58cYby6wWqVoUZsTmzZgxgzVr1rB48WIeeOABvvvuO2644QbS09PL7BzTpk3jf//7Hy+++CIrV67km2++KbNjS9FOnjzJXXfdRf369VmwYAFr1qwhIiLism137tzJK6+8Uqa/fEtr586dfPnll9Yuo1ienp7MnTuX1NTUQtsNw+DLL7/Ey8vLSpWJlJzCjNi85s2b07FjR7p3786kSZN45plnOHToEL/88ss1HzsjIwOA7du34+rqyrhx4+jUqRORkZHXfOxz585d8zGqur1795Kdnc2dd95Jt27d6NixI25ubtYuq1ju7u7ccMMNTJo0qVJ/xkOGDMEwDGbPnl1o+9KlSzl06BC33XablSoTKTmFGalyOnbsCMDhw4cB8/80P/74Y6677jpcXV3x8fHhlltu4eDBg4Vel38r4u+//6Zz5864ublx7733YjKZ+Pzzzzl37pzl0nv+/7rPnz/P888/T3h4OE5OTtSqVYtHH32UM2fOFDp23bp1GTRoED/99BOtWrXCxcWFV155haioKEwmE7NmzeLZZ58lODgYDw8PBg8ezIkTJ0hNTeXBBx/E398ff39/7rnnHtLS0god+6OPPqJr164EBATg7u5OZGQk77zzDtnZ2Zd9f9HR0dxwww24ublRr1493nrrLfLy8gq1PXPmDE899RT16tXD2dmZgIAABgwYwO7duy1tsrKyeP3112ncuDHOzs7UrFmTe+65h5MnT17V5/Tbb7/RqVMn3Nzc8PT0pHfv3qxZs8ayf8yYMVx//fUA3HbbbZhMpiJvM3z55ZfceuutAHTv3v2Szynf1bz3lJQUJkyYUOgzHT9+fImu9L399tscO3aMDz/88Iptr+Z8t956K82aNSv0usGDB2MymZg7d65l26ZNmzCZTPz+++9XPK+3tzdDhw5l+vTphbZPnz6dLl26FHkFbPr06bRs2RIXFxd8fX0ZOnQou3btuqTdl19+SaNGjXB2dqZJkyZ8/fXXlz3etf49EgHAELFRM2bMMAAjOjq60PYPP/zQAIxPP/3UMAzDeOCBBwxHR0fjqaeeMhYsWGDMmjXLaNy4sREYGGgkJCRYXtetWzfD19fXCA0NNf773/8ay5YtM5YvX26sWbPGGDBggOHq6mqsWbPGWLNmjZGYmGjk5eUZffv2NRwcHIyJEycaixYtMt577z3D3d3daNWqlXH+/HnLscPCwozg4GCjXr16xvTp041ly5YZ69evN5YtW2YARlhYmDFmzBhjwYIFxieffGJ4eHgY3bt3N3r37m1MmDDBWLRokfH2228b9vb2xmOPPVbo/T7xxBPGtGnTjAULFhhLly41/vWvfxn+/v7GPffcU6hdt27dDD8/P6Nhw4bGJ598YixevNh45JFHDMD46quvLO1SUlKMZs2aGe7u7sarr75qLFy40Pjxxx+Nxx9/3Fi6dKlhGIaRm5tr9OvXz3B3dzdeeeUVY/Hixcbnn39u1KpVy2jatKmRkZFR7Gc3c+ZMAzD69Olj/PLLL8acOXOMNm3aGE5OTsaKFSsMwzCM/fv3Gx999JEBGG+++aaxZs0aY8eOHZc9XmJiovHmm28agPHRRx8V+pxK8t7T09ON6667zvD39zc++OADY8mSJcaHH35oeHt7Gz169DDy8vKKfV+jR4823N3dDcMwjKFDhxo1atQwkpKSLPsB49FHHy3x+T755BMDMI4fP24YhmFkZ2cbnp6ehqurq/HAAw9Yjvf2228bDg4ORkpKSrF15tfx119/GYCxc+dOwzAM4/Tp04aLi4sxffp049133zUA49ChQ5bX5X+P77jjDmPevHnG119/bdSrV8/w9vY29u7da2mX/7M5ZMgQ4/fffze+/fZbo0GDBkZoaKgRFhZmaVeSv0fdunUzunXrVuz7kupLYUZsVv4/mGvXrjWys7ON1NRU448//jBq1qxpeHp6GgkJCcaaNWsMwHj//fcLvTYuLs5wdXU1nnnmGcu2bt26GYDx119/XXKugr+k8i1YsMAAjHfeeafQ9jlz5hQKU4ZhDjP29vbGnj17CrXNDzODBw8utH38+PEGYPzf//1foe0333yz4evrW+T3JDc318jOzja+/vprw97e3khOTr7k/a1bt67Qa5o2bWr07dvX8vWrr75qAMbixYuLPM93331nAMaPP/5YaHt0dLQBGB9//HGxNYaEhBiRkZFGbm6uZXtqaqoREBBgdO7c2bIt//szd+7cIo+Xb+7cuQZgLFu27JJ9V/vep0yZYtjZ2V0SkH/44QcDMP78889iayj492T37t2Gvb298dRTT1n2/zPMXO359u/fbwDG119/bRiGYaxcudIAjGeeecYIDw+3vK53796Fvn9Fya8jLy/PCA8PNyZMmGAYhmF89NFHhoeHh5GamnpJmDl9+rTh6upqDBgwoNCxjhw5Yjg7OxsjR440DOPi59u6detC4S82NtZwdHQsFGZK8vdIYUaKo9tMYvM6duyIo6Mjnp6eDBo0iKCgIObPn09gYCB//PEHJpOJO++8k5ycHMsjKCiIli1bEhUVVehYPj4+9OjR46rOu3TpUoBLRoTceuutuLu789dffxXa3qJFiyIv3Q8aNKjQ102aNAFg4MCBl2xPTk4udKtp8+bN3HTTTfj5+WFvb4+joyN33303ubm57N27t9Drg4KCaN++/SV15d+SA5g/fz4RERH06tWrqLfOH3/8QY0aNRg8eHCh7+t1111HUFDQJd/Xgvbs2cPx48e56667sLO7+E+Qh4cHw4cPZ+3atZa+SmXpat77H3/8QfPmzbnuuusKva++fftiMpmKfV//1KhRI+677z6mTp3KkSNHLtvmas9Xv3596taty5IlSwBYvHgxkZGR3HnnnRw6dIgDBw6QmZnJypUri/3c/il/RNM333xDTk4OX3zxBSNGjMDDw+OStmvWrOHcuXOX/H0PDQ2lR48elr/v+Z/vyJEjMZlMlnZhYWF07tz5kvdf2r9HIgUpzIjN+/rrr4mOjmbz5s0cP36cbdu20aVLFwBOnDiBYRgEBgbi6OhY6LF27VpOnTpV6FjBwcFXfd6kpCQcHByoWbNmoe0mk4mgoCCSkpKu+ti+vr6FvnZycip2+/nz5wE4cuQIN9xwg6V/xooVK4iOjuajjz4CLu1k7Ofnd8m5nZ2dC7U7efIktWvXLrJWMH9fz5w5g5OT0yXf14SEhEu+rwXlf18u9/0ICQkhLy+P06dPF3v+0ria937ixAm2bdt2yXvy9PTEMIxi39flTJ48GXt7eyZOnHjZ/SU5X8+ePS2BYcmSJfTu3ZvIyEgCAwNZsmQJq1at4ty5cyUKM4Clf8qbb77Jpk2buO+++y7b7kqfW/7+/D+DgoIuaffPbdfy90ikIAdrFyByrZo0aULbtm0vu8/f3x+TycSKFStwdna+ZP8/txX8n+SV+Pn5kZOTw8mTJwsFGsMwSEhIoF27dqU+9tX65ZdfSE9P56effiIsLMyyvai5WK5GzZo1OXr0aLFt/P398fPzY8GCBZfd7+npWeRr80NFfHz8JfuOHz+OnZ0dPj4+Jai47Pj7++Pq6npJp9iC+0siODiY8ePH89Zbb/HUU09d0/l69uzJF198wfr161m3bh0vvfQSAD169GDx4sUcPnwYDw8PSwf4qxUaGkqvXr145ZVXaNSo0SVXT/Jd6XPLrzW/XUJCwiXt/rntWv4eiRSkMCNV2qBBg3jrrbc4duwYI0aMKNNj9+zZk3feeYdvv/2WJ554wrL9xx9/JD09nZ49e5bp+S4nPyAVDGWGYfDZZ5+V+pj9+/fn5ZdfZunSpUXechs0aBCzZ88mNzeXDh06lOj4jRo1olatWsyaNYsJEyZY3kN6ejo//vijZYRTSeV/D65lOPSgQYN488038fPzIzw8vNTHKejZZ5/l008/5bnnnrum8/Xs2ROTycTEiROxs7Oja9euAPTq1Yunn36aw4cP07VrVxwdHUtc41NPPYWrq6tlRNjldOrUCVdXV7799ttC7Y4ePcrSpUu55ZZbAPPnGxwczHfffceTTz5p+XwPHz7M6tWrCQkJKfT+S/v3SKQghRmp0rp06cKDDz7IPffcw4YNG+jatSvu7u7Ex8ezcuVKIiMjefjhh0t17N69e9O3b1+effZZUlJS6NKlC9u2bWPSpEm0atWKu+66q4zfzeVrcHJy4o477uCZZ57h/PnzTJs27Zpu04wfP545c+YwZMgQnnvuOdq3b8+5c+dYvnw5gwYNonv37tx+++3MnDmTAQMG8Pjjj9O+fXscHR05evQoy5YtY8iQIQwdOvSyx7ezs+Odd95h1KhRDBo0iIceeojMzEzeffddzpw5w1tvvVWquvNn+P3000/x9PTExcWF8PDwy95eKu69//jjj3Tt2pUnnniCFi1akJeXx5EjR1i0aBFPPfVUiX/penl58eKLLxYKvKU5X0BAAM2bN2fRokV0797dEvh69epFcnIyycnJfPDBByWqLV+fPn3o06dPsW1q1KjBxIkTeeGFF7j77ru54447SEpK4pVXXsHFxYVJkyYB5s/3tdde4/7772fo0KE88MADnDlzhsmTJ19ym+la/h6JFGLV7sci16CoodmXM336dKNDhw6Gu7u74erqatSvX9+4++67jQ0bNljadOvWzWjWrNllX3+50UyGYRjnzp0znn32WSMsLMxwdHQ0goODjYcfftg4ffp0oXZhYWHGwIEDL3l9UaN1inpvkyZNMgDj5MmTlm2///670bJlS8PFxcWoVauW8fTTTxvz58+/ZGRPUe9v9OjRhUaYGIZ55Mrjjz9u1KlTx3B0dDQCAgKMgQMHGrt377a0yc7ONt577z3LuT08PIzGjRsbDz30kLFv375LzvNPv/zyi9GhQwfDxcXFcHd3N3r27GmsWrXqqr4/Rfn3v/9thIeHG/b29gZgzJgxo8TvPS0tzXjppZeMRo0aGU5OToa3t7cRGRlpPPHEE4WG8l9OUX9PMjMzjfDw8EtGM5X0fE888YQBGG+88Uah7Q0bNjQAY9u2bcXWl+9ydfzT5YZmG4ZhfP7550aLFi0stQ4ZMuSyQ+Y///xzo2HDhoaTk5MRERFhTJ8+/bLf76v9e6TRTFIck2EYhnVilIiIiMi102gmERERsWkKMyIiImLTFGZERETEpinMiIiIiE1TmBERERGbpjAjIiIiNq3KT5qXl5fH8ePH8fT0LJfp5EVERKTsGYZBamoqISEhhRalvZwqH2aOHz9OaGiotcsQERGRUoiLi7vi4rdVPszkL1QWFxeHl5eXlasRERGRq5GSkkJoaOhVLTha5cNM/q0lLy8vhRkREREbczVdRNQBWERERGyawoyIiIjYNIUZERERsWlVvs+MiIhcu7y8PLKysqxdhlQhjo6O2Nvbl8mxFGZERKRYWVlZHDp0iLy8PGuXIlVMjRo1CAoKuuZ54BRmRESkSIZhEB8fj729PaGhoVecvEzkahiGQUZGBomJiQAEBwdf0/EUZkREpEg5OTlkZGQQEhKCm5ubtcuRKsTV1RWAxMREAgICrumWkyK2iIgUKTc3FwAnJycrVyJVUX5Azs7OvqbjKMyIiMgVaW07KQ9l9fdKYUZERERsmsKMiIjIFdStW5d///vfZXKsqKgoTCYTZ86cKZPjiToAi4hIFXXjjTdy3XXXlUkIiY6Oxt3d/dqLknKhKzPXYPWBU6Rn5li7DBERKQXDMMjJubp/w2vWrKnRXJWYwkwpTflzFyM/W8d//tpn7VJEROQfxowZw/Lly/nwww8xmUyYTCa+/PJLTCYTCxcupG3btjg7O7NixQoOHDjAkCFDCAwMxMPDg3bt2rFkyZJCx/vnbSaTycTnn3/O0KFDcXNzo2HDhvz222+lrvfHH3+kWbNmODs7U7duXd5///1C+z/++GMaNmyIi4sLgYGB3HLLLZZ9P/zwA5GRkbi6uuLn50evXr1IT08vdS22SGGmlNqH+wLwxcpD7ElItXI1IiIVwzAMMrJyrPIwDOOq6/zwww/p1KkTDzzwAPHx8cTHxxMaGgrAM888w5QpU9i1axctWrQgLS2NAQMGsGTJEjZv3kzfvn0ZPHgwR44cKfYcr7zyCiNGjGDbtm0MGDCAUaNGkZycXOLv6caNGxkxYgS33347MTExTJ48mYkTJ/Lll18CsGHDBv7v//6PV199lT179rBgwQK6du0KQHx8PHfccQf33nsvu3btIioqimHDhpXoe1UVqM9MKfX0Ocl/g+bzfWItJv7iw5yHOmrooohUeeeyc2n68kKrnHvnq31xc7q6X1ve3t44OTnh5uZGUFAQALt37wbg1VdfpXfv3pa2fn5+tGzZ0vL166+/zs8//8xvv/3GuHHjijzHmDFjuOOOOwB48803+e9//8v69evp169fid7XBx98QM+ePZk4cSIAERER7Ny5k3fffZcxY8Zw5MgR3N3dGTRoEJ6enoSFhdGqVSvAHGZycnIYNmwYYWFhAERGRpbo/FWBrsyUVsz3DD7zDbc6rGR9bDI/bDxq7YpEROQqtG3bttDX6enpPPPMMzRt2pQaNWrg4eHB7t27r3hlpkWLFpbn7u7ueHp6WqbnL4ldu3bRpUuXQtu6dOnCvn37yM3NpXfv3oSFhVGvXj3uuusuZs6cSUZGBgAtW7akZ8+eREZGcuutt/LZZ59x+vTpEtdg63RlprQi+sGqD+njFIN9Vi5T5u+mV5NAfNw1S6aIVF2ujvbsfLWv1c5dFv45Kunpp59m4cKFvPfeezRo0ABXV1duueWWK64S7ujoWOhrk8lUqsU4DcO45Mp+wdtEnp6ebNq0iaioKBYtWsTLL7/M5MmTiY6OpkaNGixevJjVq1ezaNEi/vvf//Liiy+ybt06wsPDS1yLrdKVmdKq3R5cauCSc5ab/I6RnJ7FOwt3W7sqEZFyZTKZcHNysMqjpLfynZycLMsxFGfFihWMGTOGoUOHEhkZSVBQELGxsaX8DpVc06ZNWblyZaFtq1evJiIiwrJekYODA7169eKdd95h27ZtxMbGsnTpUsD8mXTp0oVXXnmFzZs34+TkxM8//1xh9VcGujJTWvYO0LA3xMzl6fBYfk6qw3fr47i1bSit6/hYuzoRkWqvbt26rFu3jtjYWDw8PIq8atKgQQN++uknBg8ejMlkYuLEiaW6wlJaTz31FO3ateO1117jtttuY82aNUydOpWPP/4YgD/++IODBw/StWtXfHx8+PPPP8nLy6NRo0asW7eOv/76iz59+hAQEMC6des4efIkTZo0qbD6K4NKc2VmypQpmEwmxo8fb9k2ZswYy5C6/EfHjh2tV+Q/RZg7eYWcWM4tbWoD8OLP28nJrbgfAhERubwJEyZgb29P06ZNqVmzZpF9YP71r3/h4+ND586dGTx4MH379qV169YVVmfr1q35/vvvmT17Ns2bN+fll1/m1VdfZcyYMQDUqFGDn376iR49etCkSRM++eQTvvvuO5o1a4aXlxd///03AwYMICIigpdeeon333+f/v37V1j9lYHJqATjt6KjoxkxYgReXl50797dMpZ/zJgxnDhxghkzZljaOjk54evre9XHTklJwdvbm7Nnz+Ll5VW2hWckw7sNwMjl9AMbuPHzQ5w9l83EQU257/rqc69SRKqu8+fPc+jQIcLDw3FxcbF2OVLFFPf3qyS/v61+ZSYtLY1Ro0bx2Wef4eNz6e0ZZ2dngoKCLI+SBJly5+YLoR0A8DkWxbP9GgPwwaI9JJw9b83KREREqg2rh5lHH32UgQMH0qtXr8vuj4qKIiAggIiICB544IErDnvLzMwkJSWl0KNcRVzo1b93Abe3C6VVnRqkZ+Xy2h87y/e8IiJSKY0dOxYPD4/LPsaOHWvt8qokq3YAnj17Nps2bSI6Ovqy+/v378+tt95KWFgYhw4dYuLEifTo0YONGzfi7Ox82ddMmTKFV155pTzLLiyiHyyZBIf+xi47nddvbs7g/65kXkw8I/aepFtEzYqrRURErO7VV19lwoQJl91X5t0dBLBimImLi+Pxxx9n0aJFRd6Hve222yzPmzdvTtu2bQkLC2PevHkMGzbssq95/vnnefLJJy1fp6SkWKawLhc1G0GNMDhzGA4tp1njgYzpHM70VYeY9Ot2FozviksZzY0gIiKVX0BAAAEBAdYuo1qx2m2mjRs3kpiYSJs2bXBwcMDBwYHly5fzn//8BwcHh8vODRAcHExYWBj79hW9uKOzszNeXl6FHuXKZLKMamLvAgCe6N2QQC9nYpMymBZ1oHzPLyIiUs1ZLcz07NmTmJgYtmzZYnm0bduWUaNGsWXLFstEQQUlJSURFxdHcHCwFSouhqXfzCLIy8PTxZGJg5oCMC3qAIdOVa/VS0VERCqS1cKMp6cnzZs3L/Rwd3fHz8+P5s2bk5aWxoQJE1izZg2xsbFERUUxePBg/P39GTp0qLXKvry614OjO6QlQMJWAAZGBnNDQ3+ycvN4+dft1W4FUxERkYpi9dFMRbG3tycmJoYhQ4YQERHB6NGjiYiIYM2aNXh6elq7vMIcnKF+d/PzvebVZE0mE68NaY6Tgx0r9p1iXky8FQsUERGpuirVcgZRUVGW566urixcaJ1l5ksloh/s/sMcZm58DoC6/u48cmN9/r1kH6/+vpNuETXxdHG8woFERESkJCrtlRmb07CP+c/jmyD1hGXz2G71qevnRmJqJh8s3mul4kREpKTq1q1rmZEezFfcf/nllyLbx8bGYjKZ2LJlyzWdt6yOUxJXem+VncJMWfEMhJBW5uf7Flk2uzja8+qQ5gB8tTqW7cfOWqM6ERG5RvHx8WW+5tGYMWO4+eabC20LDQ0lPj6e5s2bl+m5qjKFmbL0jyHa+bpG1GRgi2DyDHjpl+3k5akzsIiIrQkKCipywtayZG9vT1BQEA4OlaonSKWmMFOW8odoH1gGOZmFdr08qCkezg5siTvD7Og4KxQnIlJ9/O9//6NWrVrk5eUV2n7TTTcxevRoDhw4wJAhQwgMDMTDw4N27dqxZMmSYo/5z1sx69evp1WrVri4uNC2bVs2b95cqH1ubi733Xcf4eHhuLq60qhRIz788EPL/smTJ/PVV1/x66+/YjKZMJlMREVFXfY20/Lly2nfvj3Ozs4EBwfz3HPPkZOTY9l/44038n//938888wz+Pr6EhQUxOTJk0v+jbsgJiaGHj164Orqip+fHw8++CBpaWmW/VFRUbRv3x53d3dq1KhBly5dOHz4MABbt26le/fueHp64uXlRZs2bdiwYUOpa7kaCjNlKagleARBdjrEriy0K9DLhSd7RwDw9oLdnErLvNwRREQqN8OArHTrPEowxcWtt97KqVOnWLZsmWXb6dOnWbhwIaNGjSItLY0BAwawZMkSNm/eTN++fRk8eDBHjhy5quOnp6czaNAgGjVqxMaNG5k8efIlSxjk5eVRu3Ztvv/+e3bu3MnLL7/MCy+8wPfffw/AhAkTGDFiBP369SM+Pp74+Hg6d+58ybmOHTvGgAEDaNeuHVu3bmXatGl88cUXvP7664XaffXVV7i7u7Nu3TreeecdXn31VRYvXnzV37N8GRkZ9OvXDx8fH6Kjo5k7dy5Llixh3LhxAOTk5HDzzTfTrVs3tm3bxpo1a3jwwQcxmUwAjBo1itq1axMdHc3GjRt57rnncHQs38EvuoZVluzsIKIPbPraPKqpQc9Cu+/uFMYPG4+yMz6FKX/u5v0RLa1UqIhIKWVnwJsh1jn3C8fByf2qmvr6+tKvXz9mzZpFz57mf4vnzp2Lr68vPXv2xN7enpYtL/4b/Prrr/Pzzz/z22+/WX5pF2fmzJnk5uYyffp03NzcaNasGUePHuXhhx+2tHF0dCy0VmB4eDirV6/m+++/Z8SIEXh4eODq6kpmZiZBQUFFnuvjjz8mNDSUqVOnYjKZaNy4McePH+fZZ5/l5Zdfxs7OfF2iRYsWTJo0CYCGDRsydepU/vrrL3r37n1V37OC7+3cuXN8/fXXuLubv99Tp05l8ODBvP322zg6OnL27FkGDRpE/fr1AWjSpInl9UeOHOHpp5+mcePGllrKm67MlLWC/Wb+8b8IB3s73hjaHJMJftx0lHUHk6xQoIhI9TBq1Ch+/PFHMjPNV8JnzpzJ7bffjr29Penp6TzzzDM0bdqUGjVq4OHhwe7du6/6ysyuXbto2bIlbm5ulm2dOnW6pN0nn3xC27ZtqVmzJh4eHnz22WdXfY6C5+rUqZPlygdAly5dSEtL4+jRo5ZtLVq0KPS64OBgEhMTS3Su/PO1bNnSEmTyz5eXl8eePXvw9fVlzJgxlqtZH374IfHxF+dSe/LJJ7n//vvp1asXb731FgcOlP+yProyU9bCu4G9s3nhyVN7zQtRFtCqjg93tK/DrHVHeOmX7cz7vxtwclCmFBEb4ehmvkJirXOXwODBg8nLy2PevHm0a9eOFStW8MEHHwDw9NNPs3DhQt577z0aNGiAq6srt9xyC1lZWVd17KuZ1f3777/niSee4P3336dTp054enry7rvvsm7duhK9D8MwCgWZgucvuP2ft3JMJtMlfYZKe76CxwSYMWMG//d//8eCBQuYM2cOL730EosXL6Zjx45MnjyZkSNHMm/ePObPn8+kSZOYPXt2uc7er9+iZc3ZA8JvMD//x6imfM/0bYSfuxP7EtOYvupQBRYnInKNTCbzrR5rPIr4BVsUV1dXhg0bxsyZM/nuu++IiIigTZs2AKxYsYIxY8YwdOhQIiMjCQoKIjY29qqP3bRpU7Zu3cq5c+cs29auXVuozYoVK+jcuTOPPPIIrVq1okGDBpdcpXBycrrswsr/PNfq1asLBajVq1fj6elJrVq1rrrmq9W0aVO2bNlCevrFdQVXrVqFnZ0dERERlm2tWrXi+eefZ/Xq1TRv3pxZs2ZZ9kVERPDEE0+waNEihg0bxowZM8q8zoIUZspDw/yFJy8/g3ENNyeeH2C+v/jhkn0cPZ1RUZWJiFQro0aNYt68eUyfPp0777zTsr1Bgwb89NNPbNmyha1btzJy5MgSXcUYOXIkdnZ23HfffezcuZM///yT9957r1CbBg0asGHDBhYuXMjevXuZOHEi0dHRhdrUrVuXbdu2sWfPHk6dOkV2dvYl53rkkUeIi4vjscceY/fu3fz6669MmjSJJ5980tJfpiyNGjUKFxcXRo8ezfbt21m2bBmPPfYYd911F4GBgRw6dIjnn3+eNWvWcPjwYRYtWsTevXtp0qQJ586dY9y4cURFRXH48GFWrVpFdHR0oT415UFhpjxEXJgN+MhayEi+bJPhrWvRPtyXc9m5vPL7zgosTkSk+ujRowe+vr7s2bOHkSNHWrb/61//wsfHh86dOzN48GD69u1L69atr/q4Hh4e/P777+zcuZNWrVrx4osv8vbbbxdqM3bsWIYNG8Ztt91Ghw4dSEpK4pFHHinU5oEHHqBRo0aWfjWrVq265Fy1atXizz//ZP369bRs2ZKxY8dy33338dJLL5Xwu3F13NzcWLhwIcnJybRr145bbrmFnj17MnXqVMv+3bt3M3z4cCIiInjwwQcZN24cDz30EPb29iQlJXH33XcTERHBiBEj6N+/f6GO0OXBZFTx5ZxTUlLw9vbm7NmzeHl5VdyJP+oIJ3fB8C8g8pbLNtl7IpUBH64gJ8/g87vb0qtpYMXVJyJyFc6fP8+hQ4cIDw/HxcXF2uVIFVPc36+S/P7WlZnykj+BXhH9ZgAiAj25/4Z6AEz6bQcZWTlFthUREZHLU5gpL/lDtPcthtyiQ8r/9WxArRquHDtzjqlL91dQcSIiUl3MnDkTDw+Pyz6aNWtm7fLKhIZml5fa7cDVB86dhqPREHbp/AMAbk4OTBrclAe/2chnKw4yskMdavuUbPihiIhIUW666SY6dOhw2X3lPTNvRVGYKS/2DtCgN8R8b77VVESYAejTLIj24b6sP5TM/JgEHuharwILFRGRqszT0xNPT09rl1GudJupPEUUP0S7oIGRwQAs3JFQnhWJiIhUOQoz5alBTzDZm0c1nY4ttmnvCyOZNh45zclULUIpIpVLFR/4KlZSmhmKL0e3mcqTqw/U6QiHV8HeRdDhwSKbhtRwpUVtb7YdPcuSXSe4o32dCixUROTyHB0dMZlMnDx5kpo1axY5zb1ISRiGQVZWFidPnsTOzg4nJ6drOp7CTHmL6HshzCwoNswA9G0WxLajZ1m4I0FhRkQqBXt7e2rXrs3Ro0dLNN2/yNVwc3OjTp061zyTscJMeYvoB4tfhtgVkJlmXrupCH2bBfLuwj2s3p9E6vlsPF2qRi9zEbFtHh4eNGzY8LJT7YuUlr29PQ4ODmVytU9hprz5R4BPXXOfmYNR0GRQkU3r1/Sgnr87B0+lE7XnJINbhlRUlSIixbK3t8fe3t7aZYhcljoAlzeTqcAEesWPajKZTPRpFgRoVJOIiMjVUpipCJYh2ovgCj23+zQzj2qK2nOSzJzil4UXERERhZmKEdYFnDwgLQESthbb9LraNQjwdCYtM4fVB5IqqEARERHbpTBTERycoX538/MrTKBnZ2eyXJ1ZtONEeVcmIiJi8xRmKkp+v5liVtHO16epud/M4p0nyM3TRFUiIiLFUZipKA16m/88vhlSi+/c27GeH54uDpxKy2TzkdMVUJyIiIjtUpipKJ6BENLa/HzfomKbOjnY0bNxAACLdupWk4iISHEUZiqS5VbTlReeLDhEW2uiiIiIFE1hpiLlD9E+sAxyil9MsltETZwc7DiclMGeE6kVUJyIiIhtUpipSMEtwTMYstMhdmWxTd2dHbihgT+gUU0iIiLFUZipSCYTNOxjfn4Vt5r6ajZgERGRK1KYqWgFh2hfoS9MzyYB2Jlgx/EUjp7OqIDiREREbI/CTEWr1w3sneHMYTi5p9imfh7OtK3rC+hWk4iISFEUZiqakzuEdzU/v4oJ9HSrSUREpHgKM9ZgWXjyKoZoNzUvbRAdm0xyelZ5ViUiImKTFGasIb8TcNw6yEgutmmorxtNg73IM2DJLt1qEhER+SeFGWvwCYOApmDkwoGlV2yef6tpkW41iYiIXEJhxlost5quYuHJC6to/73vFOmZOeVZlYiIiM2pNGFmypQpmEwmxo8fb9lmGAaTJ08mJCQEV1dXbrzxRnbs2GG9IstS/hDtfYsht/iA0jjIkzq+bmTl5PH33pMVUJyIiIjtqBRhJjo6mk8//ZQWLVoU2v7OO+/wwQcfMHXqVKKjowkKCqJ3796kplaB6f1rtwNXHzh/Bo6uL7apyWSi74WrM1p4UkREpDCrh5m0tDRGjRrFZ599ho+Pj2W7YRj8+9//5sUXX2TYsGE0b96cr776ioyMDGbNmmXFisuInX2B2YCv5laTud/MX7tOkJ2bV56ViYiI2BSrh5lHH32UgQMH0qtXr0LbDx06REJCAn369LFsc3Z2plu3bqxevbrI42VmZpKSklLoUWmVYIh26zo++Hs4kXI+h7UHk8q5MBEREdth1TAze/ZsNm3axJQpUy7Zl5BgHrkTGBhYaHtgYKBl3+VMmTIFb29vyyM0NLRsiy5L9XuCyR5O7obTscU2tbcz0avJhVtNmg1YRETEwmphJi4ujscff5xvv/0WFxeXItuZTKZCXxuGccm2gp5//nnOnj1recTFxZVZzWXOtQbU6WR+vnfRFZtbhmjvTCAvr/h1nURERKoLq4WZjRs3kpiYSJs2bXBwcMDBwYHly5fzn//8BwcHB8sVmX9ehUlMTLzkak1Bzs7OeHl5FXpUaiUYot25gR/uTvacSMlk27Gz5VyYiIiIbbBamOnZsycxMTFs2bLF8mjbti2jRo1iy5Yt1KtXj6CgIBYvXmx5TVZWFsuXL6dz587WKrvs5Q/Rjl0BmWnFNnV2sOfGxgGA1moSERHJZ7Uw4+npSfPmzQs93N3d8fPzo3nz5pY5Z958801+/vlntm/fzpgxY3Bzc2PkyJHWKrvs+TcEn3DIzYKDUVdsroUnRURECrP6aKbiPPPMM4wfP55HHnmEtm3bcuzYMRYtWoSnp6e1Sys7JtPFqzNXcaupe6OaONqbOHgynf2JxV/JERERqQ5MhmFU6Z6kKSkpeHt7c/bs2crbf+bAMvjmZvAIhCd3g13xGXP09PUs33uSp/s24tHuDSqmRhERkQpUkt/flfrKTLUR1gWcPCDtBBzbcMXmWnhSRETkIoWZysDBCRoPMj/fMvOKzXs1DcBkgq1HzxJ/9lw5FyciIlK5KcxUFq3uNP8Z8yNkZRTbNMDThdZ1zEs/LNZaTSIiUs0pzFQWda83j2rKSoWdv16xuWXhSc0GLCIi1ZzCTGVhMkGrUebnm7+9YvM+Tc39ZtYeTOJsRnZ5ViYiIlKpKcxUJi1HAiY4vBKSDhTbtK6/O40CPcnJM/hrt67OiIhI9aUwU5l414IGPc3Pt8y6YvM+utUkIiKiMFPptLrL/OeWWZCXW2zT/CHay/ee5Hx28W1FRESqKoWZyqZRf3D1hdTjcGBpsU2bhXhRq4Yr57Jz+XvvyQoqUEREpHJRmKlsHJyhxW3m55u/KbapyWSid9MLt5o0RFtERKophZnKKH/Omd1/QnpSsU3zbzX9tesEObl55V2ZiIhIpaMwUxkFNYeQVpCXDdvmFNu0XV0ffNwcOZ2RTXTs6QoqUEREpPJQmKms8q/ObP4WilkL1MHejp5NzLeaFmqtJhERqYYUZiqr5reAgwsk7oDjm4ttmn+rafHOE1TxRdBFREQuoTBTWbnWgCaDzc+v0BH4hob+uDrac+zMOXYcTyn/2kRERCoRhZnKLH/OmZgfil180sXRnm4RNQHdahIRkepHYaYyq3sD1KgDmSmw+49im/Ztrn4zIiJSPSnMVGZ2dnBdfkfg4m819WgUiIOdib0n0jh0Kr0CihMREakcFGYqu+suLD556G9IPlRkM283RzrW8wNgka7OiIhINaIwU9nVCIX63c3Pr7D4pGXhSc0GLCIi1YjCjC3In3PmCotP9mlqHqK96chpElPPV0RlIiIiVqcwYwsaDQSXGpByFA5GFdksyNuFlqE1MAzznDMiIiLVgcKMLXB0gRYjzM+v0BG4T/7CkzsUZkREpHpQmLEV+XPO7J4HGclFNsufDXj1gVOknM+uiMpERESsSmHGVgS3gKAWkJsFMXOLbNYgwIP6Nd3JzjWI2nOyAgsUERGxDoUZW5J/deZKt5ouXJ1ZsD2+vCsSERGxOoUZWxJ5C9g7Q0IMHN9SZLOBkcEALNxxgv2JaRVUnIiIiHUozNgSN19oMsj8fPO3RTZrXsub3k0Dyc0zeG/hngoqTkRExDoUZmxN/pwzMd9DdtFzyTzTtxF2JliwI4FNR05XUHEiIiIVT2HG1oR3A+9QOH+22MUnGwZ6cmubUADe+nM3hmFUVIUiIiIVSmHG1tjZX1iviSt2BB7fuyHODnasj01m6e7ECihORESk4inM2KLrRpn/PLgcTh8uslmwtyv3dAkH4O0Fu8nN09UZERGpehRmbJFPmPl2EwZs/a7Ypg93q4+3qyN7T6Tx06ajFVOfiIhIBVKYsVWWOWdmQl5ekc283Rx5tHt9AD5YvJfz2UUvVCkiImKLFGZsVZNB4OwNZ4/AoeXFNr27U11CvF2IP3uer9fEVkx9IiIiFURhxlY5ukKLW83Pi5lzBsDF0Z4nekcA8NGyA5zN0JpNIiJSdSjM2LL8OWd2/Q7nip9LZljr2kQEenD2XDbTlh+ogOJEREQqhsKMLQu+DgKbQ24mxPxQbFN7OxPP9msMwIxVh4g/e64CChQRESl/CjO2zGS6eHXmCnPOAPRoHED7ur5k5uTx78X7yrk4ERGRiqEwY+ta3Ab2ThC/FeK3FdvUZDLxbH/z1Zm5G+PYdyK1IioUEREpV1YNM9OmTaNFixZ4eXnh5eVFp06dmD9/vmX/mDFjMJlMhR4dO3a0YsWVkJsvNBpgfn6FjsAAbcJ86NsskDwD3tEilCIiUgVYNczUrl2bt956iw0bNrBhwwZ69OjBkCFD2LFjh6VNv379iI+Ptzz+/PNPK1ZcSeXPObNtTrGLT+Z7um9j7EyweOcJNsQml3NxIiIi5cuqYWbw4MEMGDCAiIgIIiIieOONN/Dw8GDt2rWWNs7OzgQFBVkevr6+Vqy4kqrfHbxqwfkzsOfKYa9BgAe3tbuwCOV8LUIpIiK2rdL0mcnNzWX27Nmkp6fTqVMny/aoqCgCAgKIiIjggQceIDGx+AUTMzMzSUlJKfSo8gotPnnlW00Aj/eMwMXRjg2HT7NklxahFBER22X1MBMTE4OHhwfOzs6MHTuWn3/+maZNmwLQv39/Zs6cydKlS3n//feJjo6mR48eZGZmFnm8KVOm4O3tbXmEhoZW1Fuxrvwwc2ApnIm7YvMgbxfuvbAI5TsLdpOTW/SSCCIiIpWZybDyPYasrCyOHDnCmTNn+PHHH/n8889Zvny5JdAUFB8fT1hYGLNnz2bYsGGXPV5mZmahsJOSkkJoaChnz57Fy8ur3N5HpfDlIIhdAd1fhG7PXLH52XPZdHt3GWcysnlneAtGtKsmwU9ERCq9lJQUvL29r+r3t9WvzDg5OdGgQQPatm3LlClTaNmyJR9++OFl2wYHBxMWFsa+fUXPkeLs7GwZHZX/qDYsc858W+zik/m8XR0Z170BoEUoRUTEdlk9zPyTYRhF3kZKSkoiLi6O4ODgCq7KRjS5CZy94MxhOLzyql5yZ8cwatVwJSHlPF+uji3f+kRERMqBVcPMCy+8wIoVK4iNjSUmJoYXX3yRqKgoRo0aRVpaGhMmTGDNmjXExsYSFRXF4MGD8ff3Z+jQodYsu/JycoPmw83PN115RmAwL0L55IVFKD9etp8zGVnlVZ2IiEi5sGqYOXHiBHfddReNGjWiZ8+erFu3jgULFtC7d2/s7e2JiYlhyJAhREREMHr0aCIiIlizZg2enp7WLLtyy59zZtdvcO7MVb3k5la1aBzkScr5HD6O0iKUIiJiW6zeAbi8laQDUZVgGPBxJzi5CwZ+AO3uu6qXLdudyD1fRuPkYMeyCTdSq4ZrORcqIiJSNJvqACxlzGSC1heuzqz811UN0wa4sVFNOoT7kpWTx78W7y3HAkVERMqWwkxVdN0o8AmHs3Hw5UA4c+SKLzGZTDx3YRHKHzcdZXdCNZhsUEREqgSFmarItQaMmQe+9cwjm64y0LSq40P/5kEYBry7QItQioiIbVCYqaq8a8HoPy4EmiNXHWgm9G2EvZ2Jv3Ynsu5gUgUUKiIicm0UZqoy71oFrtBcCDSnDxf7kvo1CyxCuUCLUIqISOWnMFPVeYVcCDT1LwSaQVcMNON7NsTV0Z7NR86wcMeJCipURESkdBRmqgOvEBjzhznQnL1yoAnwcuG+6y8sQrlQi1CKiEjlpjBTXZQw0DzYrR4+bo4cPJnO3I1HK7BQERGRklGYqU7ybzn5NbgQaAbC6djLN3VxZFyPhgD8a/FezmVpEUoREamcFGaqG69g8ygnvwYX5qEZBMmHLtv0zo51qFXDlcTUTKavunwbERERa1OYqY68ggtcoYmDrwZfNtA4O9gzoa95EcpPog6QnK5FKEVEpPJRmKmuPIMuBJqGxV6hGdKyFk2CvUjNzGHcrE1k5uh2k4iIVC4KM9WZZ5C5U7B/BKQcvRBoDhZqYmdn4t1bWuDuZM/qA0k8+f1W8vI094yIiFQeCjPVnWcQjP69QKAZfEmgaV7Lm0/uaoOjvYl52+J55fcdmkxPREQqDYUZuRBoir9Cc0PDmrw/4joAvlpzmI+jDlihUBERkUspzIiZZ+CFQNMIUo6ZA01S4cByU8sQXh7UFIB3F+5hTvSV13oSEREpbwozcpFn4IU+NEUHmnuvD+fhG+sD8PxPMSzZqeUORETEuhRmpDCPgIuBJvX4ZQPNM30bcUub2uQZ8OisTWw8nGylYkVERBRm5HLyA03NxpcNNCaTiSnDIunROIDMnDzu/XIDe0+kWrFgERGpzhRm5PI8AsyjnCyBZmChTsGO9nZ8NLI1rerU4Oy5bEZPX8/xM+esWLCIiFRXCjNSNI8Ac6fgmo0hNR6+vQXSkyy7XZ3smT66HQ0CPIg/e567p6/nTIZmCRYRkYqlMCPF86gJd/8G3nUg+QB8dztkX7wC4+PuxFf3tifIy4X9iWnc+2W0FqUUEZEKpTAjV+YZCHf+AC7ecHQ9/PQA5F0MLLVquPL1fe3xcnFg05EzjJu1iZzcPCsWLCIi1YnCjFydmo3g9u/A3gl2/Q6LXiq0OyLQk+lj2uHsYMdfuxN5/qcYzRIsIiIVQmFGrl7dLnDzNPPztR/Dmo8L7W5b15epI1tjZ4K5G4/y7sI9VihSRESqG4UZKZnIW6DXK+bnC1+Anb8W2t27aSBThkUC8HHUAWasunQlbhERkbKkMCMl1+VxaHc/YMBPD8KRdYV239auDhP6RADw6h87+W3rcSsUKSIi1YXCjJScyQT934GI/pBz3jzC6dT+Qk0e7d6A0Z3CMAx46vstrNx3ykrFiohIVacwI6VjZw+3fAEhreFcMswcDmknLbtNJhMvD27GwBbBZOcaPPTNBrYfO2vFgkVEpKpSmJHSc3KHkd+DT104HQvf3QZZ6Zbd9nYmPhjRks71/UjPymXMjPUcTkov8nAiIiKloTAj18ajJoz6EVx94NhG+PH+QnPQODvY87+72tA02ItTaVnc9cV6TqZmWrFgERGpahRm5Nr5N4A7ZoO9M+z5E+Y/AwXmmPF0ceTLe9tRx9eNI8kZjJmxntTz2VYsWEREqhKFGSkbdTrC8M8AE0R/Dqv/U2h3gKcLX9/bHn8PJ3YcT+GBrzco0IiISJlQmJGy03QI9H3T/HzxyxDzQ6Hddf3dmTGmPe5O9qw9mMytn6wh/qxW2hYRkWujMCNlq9Mj0OFh8/NfHobYVYV2R9b25rsHO+Lv4czuhFSGfrSancdTrFCoiIhUFQozUvb6vgFNBkNuFsy+A04WXtagRe0a/PxIZxoEeJCQcp5bP1nN8r0niziYiIhI8RRmpOzZ2cOwz6B2ezh/Fr69BVITCjUJ9XXjx4c706meedj2vV9GM3v9ESsVLCIitqxUYSYuLo6jR49avl6/fj3jx4/n008/LbPCxMY5uppHOPnWh7NHYNYIyEwr1MTb1ZGv7m3PsFa1yM0zeO6nGN5duJu8PK22LSIiV69UYWbkyJEsW7YMgISEBHr37s369et54YUXePXVV8u0QLFh7n5w5w/g5g/xW2HuGMjNKdTEycGO90e05P96NgTgo2UHGD9nC5k5uZc5oIiIyKVKFWa2b99O+/btAfj+++9p3rw5q1evZtasWXz55ZdlWZ/YOt96MHIOOLjC/sUw78lCc9CAeemDJ3tH8O4tLXCwM/Hb1uPc9fl6zmRkWaloERGxJaUKM9nZ2Tg7OwOwZMkSbrrpJgAaN25MfHz8VR9n2rRptGjRAi8vL7y8vOjUqRPz58+37DcMg8mTJxMSEoKrqys33ngjO3bsKE3JYk2125rXcTLZwaavYMV7l212a9tQvrynPZ7ODqyPTWbYtNUcScqo4GJFRMTWlCrMNGvWjE8++YQVK1awePFi+vXrB8Dx48fx8/O76uPUrl2bt956iw0bNrBhwwZ69OjBkCFDLIHlnXfe4YMPPmDq1KlER0cTFBRE7969SU1NLU3ZYk2NB5pX2gZY+jps+e6yza5v6M/chzsR4u3CwZPpDP14FVvizlRcnSIiYnNMhmGUuLdlVFQUQ4cOJSUlhdGjRzN9+nQAXnjhBXbv3s1PP/1U6oJ8fX159913uffeewkJCWH8+PE8++yzAGRmZhIYGMjbb7/NQw89dFXHS0lJwdvbm7Nnz+Ll5VXquqSMLJp4YXZgE3R9Gm58zjz66R9OpJzn3i+j2XE8BRdHO/59Wyv6NQ+q+HpFRMQqSvL7u1RhBiA3N5eUlBR8fHws22JjY3FzcyMgIKBUx5s7dy6jR49m8+bNuLi4UL9+fTZt2kSrVq0s7YYMGUKNGjX46quvLnuczMxMMjMvLmSYkpJCaGiowkxlkZcH8582L3kAUPcGGP4FeAZe0jQ9M4dxszaxbM9JTCZ4aWBT7rs+vIILFhERayhJmCnVbaZz586RmZlpCTKHDx/m3//+N3v27ClxkImJicHDwwNnZ2fGjh3Lzz//TNOmTUlIMM9LEhhY+JdcYGCgZd/lTJkyBW9vb8sjNDS0hO9OypWdHQx83zwPjaM7xK6AT66Hg8svaeru7MBnd7dlVIc6GAa89sdOJv+2g1wN3RYRkQJKFWaGDBnC119/DcCZM2fo0KED77//PjfffDPTpk0r0bEaNWrEli1bWLt2LQ8//DCjR49m586dlv0mk6lQe8MwLtlW0PPPP8/Zs2ctj7i4uBLVIxWkxQh4MAoCmkJ6InxzM0S9DXmFh2Q72Nvx+s3Nea5/YwC+XB3L2G83ci5LQ7dFRMSsVGFm06ZN3HDDDQD88MMPBAYGcvjwYb7++mv+85//XOHVhTk5OdGgQQPatm3LlClTaNmyJR9++CFBQeb+Ef+8CpOYmHjJ1ZqCnJ2dLaOj8h9SSdWMgPv/glZ3gpEHUW/Ct8MgrfDSBiaTibHd6jN1ZCucHOxYvPMEt3+6hpOpmUUcWEREqpNShZmMjAw8PT0BWLRoEcOGDcPOzo6OHTty+PDhayrIMAwyMzMJDw8nKCiIxYsXW/ZlZWWxfPlyOnfufE3nkErEyQ2GfAQ3TzPPRXMwynzbKXblJU0HtQhh1v0d8HFzZOvRswz9eBX7EzWyTUSkuitVmGnQoAG//PILcXFxLFy4kD59+gDmqyYluRLywgsvsGLFCmJjY4mJieHFF18kKiqKUaNGYTKZGD9+PG+++SY///wz27dvZ8yYMbi5uTFy5MjSlC2V2XUj4cFl4N8I0hLgq8Hw93vmDsMFtK3ry0+PdCHMz42jp88x7OPVrD2YZKWiRUSkMihVmHn55ZeZMGECdevWpX379nTq1AkwX6UpOPLoSk6cOMFdd91Fo0aN6NmzJ+vWrWPBggX07t0bgGeeeYbx48fzyCOP0LZtW44dO8aiRYssV4WkigloYg40Le8w33Za+hrMuhXSC4eVcH93fnq4M63r1CDlfA53fbGO76PVN0pEpLoq9dDshIQE4uPjadmyJXZ25ky0fv16vLy8aNy4cZkWeS00z4wNMgzY/C38OQFyzoNnCNw6A+p0LNTsfHYuT8zZwvzt5n5Vt7SpzWtDmuPqdOm8NSIiYlsqZJ6ZfEePHsVkMlGrVq1rOUy5UZixYSd2wPejIWkfmOyh58vQ+f/Mw7svyMsz+DhqPx8s3kueARGBHnw8qjUNAnT1TkTElpX7PDN5eXm8+uqreHt7ExYWRp06dahRowavvfYaef/o4yBSaoHNzLedIm8FIxeWTILvboeMZEsTOzsT43o0ZOb9Hanp6czeE2ncNHUVP28+asXCRUSkIpUqzLz44otMnTqVt956i82bN7Np0ybefPNN/vvf/zJx4sSyrlGqM2dP8wR7g/4N9s6wbyF8cgPERRdq1qm+H3/+3w10ru9HRlYuT8zZyvM/beN8tuajERGp6kp1mykkJIRPPvnEslp2vl9//ZVHHnmEY8eOlVmB10q3maqQ+G0wdzQkHwQ7B+j1CnR6FApMopibZ/Cfv/bxn6X7MAxoEuzFRyNbUa+mhxULFxGRkir320zJycmX7eTbuHFjkpOTL/MKkTIQ3AIeXA7NhkJeDix6EWaPgnOnLU3s7Uw80TuCr+9tj5+7E7viU7hp6ir+2HbcioWLiEh5KlWYadmyJVOnTr1k+9SpU2nRosU1FyVSJBcvuGUGDHgP7J1gzzz4X1c4uqFQsxsa1uTPx2+gfbgvaZk5jJu1mYm/bCczR7edRESqmlLdZlq+fDkDBw6kTp06dOrUCZPJxOrVq4mLi+PPP/+0LHVQGeg2UxV2fDPMHQOnY82jnW58Dq5/EuwdLE1ycvP4YPFePo46AEDzWl58PLINdfzcrFOziIhclXK/zdStWzf27t3L0KFDOXPmDMnJyQwbNowdO3YwY8aMUhUtUmIhreChv6H5cPNop2VvwJcDzOHmAgd7O57p15gZ97TDx82R7cdSGPjfFSzYXvTK6yIiYluueZ6ZgrZu3Urr1q3Jza08l/J1ZaYaMAyImQvznoLMFHDyhAHvQsvbC3UOPn7mHI99t5mNh819bO7pUpfn+zfByaFUmV5ERMpRuV+ZEalUTCZoMQLGroQ6nSArFX4ZCz/cU6hzcEgNV2Y/2JEHu9YDYMaqWG793xrikjOsVbmIiJQBhRmpOnzCYMw86DHRPHR7x88wrQsc+tvSxNHejhcGNOHzu9vi7erI1rgzDPzPChbvPGHFwkVE5FoozEjVYmcPXSfAfYvAtz6kHIOvboJFEyEn09KsV9NA/njselqGmherfODrDbz55y6yczWDtYiIrSlRn5lhw4YVu//MmTMsX75cfWakcshKh4UvwMYvzV8HRcLwL6Bmo4tNcvKYMn8XM1bFAtAmzIf/3tGKkBquFV+viIhYlNtCk/fcc89VtatMI5oUZoTd8+DXcXAuGRxcoM/r0O7+Qp2D58fE88wP20jNzMHHzZH3R7SkR+NAKxYtIlK9Veiq2ZWdwowAkJoAvzwCB/4yf92gNwz5CDwvBpbDSek8OmsT24+lAPDADeE83bexRjuJiFiBRjOJ/JNnEIz6Afq/Y16wcv9imNYZ9sy3NAnzc+fHhzszpnNdAD5bcUijnUREbIDCjFQfdnbQ4SF4aDkENoeMU/Dd7fD7eHP/GsDZwZ7JNzXjf3e1wcvFga1xZxjwnxX8GRNv3dpFRKRICjNS/QQ0gQeWQqdx5q83zoD/dTMvj3BB32ZB/Pn4DbSqU4PU8zk8MnMTL/0Sw/nsytO5XUREzBRmpHpycIa+b8Ddv4JnCCTtg897wYoPIM8cWGr7uPH9Q50Y260+AN+uPcLNH63iwMk0a1YuIiL/oDAj1Vu9G+HhVdB0COTlwF+vwIwBcGApGAaO9nY8178xX97TDj93J3YnpDL4vyv5adNRa1cuIiIXaDSTCJjXd9oyC+Y/A1kXrrzUbAIdx0LkCHBy40TKecbP3sKag0kADG9dm1eHNMPd2aGYA4uISGloaHYBCjNSImeOwJqPYPO3F0ONqw+0uQfa3U+uZwhTl+7nw7/2kmdA/ZruTB3ZmibB+rslIlKWFGYKUJiRUjl/FjbPhHWfwJnD5m12DubbUR0fYW1WOI/P3syJlEycHOx4eVBTRnWog6nARHwiIlJ6CjMFKMzINcnLNc9Fs+4TiF1xcXuttqS2eoDx2+rw117zytwDI4OZMjwSLxdHKxUrIlJ1KMwUoDAjZSZ+mznUxMyF3CwADM8QomsO45HdLTiV50GorytT72hNy9Aa1q1VRMTGKcwUoDAjZS4tETbMgOjPIT0RgDx7F36nK1MzehFrF8qz/Rpz3/Xhuu0kIlJKCjMFKMxIucnJhO0/wdqPIWGbZfPfuZFMz+2HQ8PevDuiFT7uTlYsUkTENinMFKAwI+XOMODIGlj7McbueZiMPAAO5AXzk+MgbhzxOO0ahVq5SBER26IwU4DCjFSo04dh/afkbvgK++xUAI4a/ixt9ia3DbsFZwd7KxcoImIbFGYKUJgRq8hMJXPDt2RE/Quf7BPkGHbMcr2ddne9QZNavtauTkSk0ivJ728tZyBSHpw9ce7yMD5PbeB4nZtwMOVx9/lZpP+vH98uXEluXpX+P4SISIVSmBEpTy5ehNz7DSn9P+KcyY22dnu4afWt/OfDd4hLzrB2dSIiVYLCjEgF8OpwJy7jVpFUowVepgyeOPsm6z8cyU9r9lDF7/SKiJQ7hRmRCmLyq4ffY0s52/Zx8jAx3LSMlvOH8Ppn33EqLdPa5YmI2CyFGZGKZO+I96BXMe7+nTTnQOrbxfPssXF8+/5TLNp+3NrViYjYJIUZESuwr3cDHo+vJaVuf5xMuYw3vsF1zq289t1fpJ7PtnZ5IiI2RWFGxFrcfPEa/R3ZA/5Ftp0zN9hv55Hdo3ntgw9YfyjZ2tWJiNgMhRkRazKZcGx/L44PryTdpyl+plTeyXqTXdMf4t0/NpOZk2vtCkVEKj2FGZHKoGYE7o9GkdVuLACj7RcxeP1d/N+Hs9gVn2Ll4kREKjeFGZHKwsEZp4Fvw6gfyXT2o7FdHP9JeYI5H7/MtGX7NdGeiEgRrBpmpkyZQrt27fD09CQgIICbb76ZPXv2FGozZswYTCZToUfHjh2tVLFIBWjYC+fH1pIV3hNnUzaT7WfQYOkDPPDJAk20JyJyGVYNM8uXL+fRRx9l7dq1LF68mJycHPr06UN6enqhdv369SM+Pt7y+PPPP61UsUgF8QjA6e4fMfpOIdfOkd72m5hyYiyT/z2V2euPaKI9EZECKtVCkydPniQgIIDly5fTtWtXwHxl5syZM/zyyy+lOqYWmhSblxBD9px7cDy9D4Cfcq9ndZ2HmDCiD0HeLlYuTkSkfNjsQpNnz54FwNe38KrCUVFRBAQEEBERwQMPPEBiYqI1yhOxjqBIHB/+m7w29wIwzH4lbxy9h8X/uod5a2N0lUZEqr1Kc2XGMAyGDBnC6dOnWbFihWX7nDlz8PDwICwsjEOHDjFx4kRycnLYuHEjzs7OlxwnMzOTzMyLU8OnpKQQGhqqKzNSNRzbRMafL+F2bBUAKYYrS33voMtdE6n5j/8EiIjYspJcmak0YebRRx9l3rx5rFy5ktq1axfZLj4+nrCwMGbPns2wYcMu2T958mReeeWVS7YrzEiVYRjk7vuL5F+fp2b6XgBOUoOEVuOJHDQO7B2tXKCIyLWzudtMjz32GL/99hvLli0rNsgABAcHExYWxr59+y67//nnn+fs2bOWR1xcXHmULGI9JhP2Eb2o+dQ6jvb4D/F2gdTkDJGbJ3PirVakb/4RKsf/UUREKoRVw4xhGIwbN46ffvqJpUuXEh4efsXXJCUlERcXR3Bw8GX3Ozs74+XlVeghUiXZ2VG762j8nt3G0vCnSDI8CcyOw/3Xezn7365waMWVjyEiUgVYNcw8+uijfPvtt8yaNQtPT08SEhJISEjg3LlzAKSlpTFhwgTWrFlDbGwsUVFRDB48GH9/f4YOHWrN0kUqDSdnF3qMfpnjd6/ha6fbSDec8U7eBl8NIvvr4ZCw3dolioiUK6v2mTGZTJfdPmPGDMaMGcO5c+e4+eab2bx5M2fOnCE4OJju3bvz2muvERoaelXn0NBsqU7OZ+fyybzV+G38kNvtluJoysXAhKnFbdD9BfAJs3aJIiJXxSY7AJcXhRmpjqJjk/nXnPmMTPuaQfZrATDsnTC1ewBueArc/axcoYhI8RRmClCYkeoqIyuHt+bvZsvapTzn8B2d7Xeadzh7QZfHoePD4ORu3SJFRIqgMFOAwoxUdyv3neKZuVtokLae5xxm09TusHmHRxDc+Cxcdyc4OFm3SBGRf7C5odkiUn6ub+jPgie7EdR6IAOz3uDxrEeINwVAWgL88QRMbQObv4XcHGuXKiJSKroyI1KNLN19gmd/jOFsahqjHP7iKdc/8MhONu/0rQ83PgfNh4OdvXULFZFqT7eZClCYESnsTEYWk37bwa9bjuPKeR5xj+JB+99wzjpjblCzMdz4PDS5Cex08VZErENhpgCFGZHLW7gjgdfn7SQu+RzunOMFv+Xclv0LDlkp5gaBkebh3I36QxHTKIiIlBeFmQIUZkSKdj47ly9WHuKjZfvJyMrFi3TeC11Fr7M/YJeVZm4U0hq6vwgNeirUiEiFUZgpQGFG5MoSzp7nnQW7+WnzMQBCnDKYWncVreLnYMrOMDcK7WAONfW6WbFSEakuFGYKUJgRuXqbjpzmld93sjXuDAAtfbP4sPZywg5+hynnvLlR3RvMoSask/UKFZEqT2GmAIUZkZLJyzP4efMx3l6wm8TUTAAGhZt4zW8hPrtmQW6WuWH9HtD9JajdxorVikhVpTBTgMKMSOmkZebw8bL9fL7iEFm5edjbmXi0lROPOvyKc8wsyLswL01EP3NH4eCW1i1YRKoUhZkCFGZErs2RpAze+HMnC3ecAKCGmyOTurgxJGUmdttmg5FnbthkMHR9WqFGRMqEwkwBCjMiZWPV/lO8+vtO9pxIBaBRoCdvdnOlzaH/QcwPwIV/SureAJ0fgwa9NU+NiJSawkwBCjMiZScnN4/v1h/h/cV7OZORDUDfZoFM7mBP8LapsONnMHLNjf0joNOj0OI2cHS1YtUiYosUZgpQmBEpe2cysvj3kn18s/YwuXkGTvZ23HdDOI+1dsFty+ew8SvIvDD5nps/tLvf/PCoad3CRcRmKMwUoDAjUn72nkjl1d93snL/KQCCvV2YfFMz+tR3xbT5W1j7CZw9Ym5s7wwtbzdfranZyIpVV2HpSfD3OxB5K9Rua+1qRK6JwkwBCjMi5cswDJbsSuTVP3YQl3wOgF5NAph8UzNqeznBrt9gzVQ4tvHiixr2gU7jILyrZhUuK3l58O0wOLgM3Pxg7ErwCrF2VSKlpjBTgMKMSMU4l5XL1GX7+PTvg2TnGrg62vN4r4bcd304jnYmOLLWHGp2z8PSWTgo0hxqmg0DByer1m/zVv4Llky++HVYF7j7N7B3sFpJItdCYaYAhRmRirXvRCov/rKd9YeSAfOopzeGNqdtXV9zg6QDsHYabJkJ+UsleAZDh4egzRhw9bFO4bYsbj1M72fufH3DBFj3CWSlQbdnzXMAidgghZkCFGZEKp5hGPy46RhvzNvJ6Qujnm5vF8pz/RtTw+3CFZiMZNgwHdZ/CmnmOWxwdIdWd0LHh8E33ErV25hzp+GTrua+Sc2Hw/AvzEPlf7ofMMHdv2o9LbFJCjMFKMyIWM/p9Czemr+bORviAPB1d+LFAU0Y1roWpvy+MjmZ5l++az6CxB3mbSY7aDwQ2j2gfjXFMQz4/m5zvySfuvDQCnC58O/cr+Ng8zfgEQhjV2kkmdgchZkCFGZErC86NpkXf45h74k0ADrW8+X1myNpEOBxsZFhmDuvrp4KB/66uN2vIbS9F667Q7eg/in6C5j3JNg5wn2LoFbri/uyMuCz7nByN9TvCaN+0CSGYlMUZgpQmBGpHLJy8vhi5SE+/Gsv57PzcLQ3MbZbfR7t3gAXR/vCjU/shOjPYNv35r4fAA4u5tsobe8z/9Ku7ldrErbDZz0gNxP6vAGdx13aJnEXfNodcs5Br8lw/RMVXqZIaSnMFKAwI1K5xCVnMOm3HSzdnQhAHV83Xru5Od0iLnMbJDMVts2B6OkXb0GBef2ntvdB5C3g5F5BlVciWenmkHJqj3mY+x1zir7qsulr+O0xMNnDPfOhToeKrVWklBRmClCYEal8DMNg4Y4EJv+2k4SU8wAMahHMy4OaEuDlcrkXQNw6c4fhHT9DbpZ5u7O3eSK+tvdCQOMKfAdWZukPEwQPrwJ3/6LbGgb8eD9s/wG8Q+Ghv8HNt+JqFSklhZkCFGZEKq+0zBz+tXgvM1YdIs8AT2cHJvRtxJ0dw7C3K+I2UnoSbPnWHGxOx17cHnY9tLsXGg+u2nPWxPwAP94HmGD0b+YO0ldyPgU+7QbJB6HRQLh9pm7TSaWnMFOAwoxI5bf92Fle/GU7W+POANCitjdv3BxJZG3vol+UlwcHl5pvQe2dD0aeebt7TWh9t3nOmhp1yr32CpV0AP7XDbJSoesz0OPFq3/t8S3wRW/zVa1+b0PHseVWpkhZUJgpQGFGxDbk5hnMWn+EdxbsJvV8DiYT3NQyhCd6RVDX/wr9Ys4eNS9uuelrSEu4sNFk7k/S7j5o0Avs7Is9RKWXk2UOI/FboE5nGP17yWf3Xfc/mP8M2DuZRz+FtCqXUkXKgsJMAQozIrYlMfU8b8zbxa9bjgNgb2fi1ja1eaxnQ2rVcC3+xbnZsOdP85DlQ8svbq9RB1reAU1ugsBmtnmLZeGL5uUgXH3M6y551y75MQwD5twJu/8An3Bz/xkX/bsolZPCTAEKMyK2afuxs7y/aA/L9pwEwMnejpEd6vBI9/oEeF6mk/A/ndoHG2aYl004f+bidp+60GSwOdjUamsbc6/sXQizRpif3/4dNB5Q+mOdOw2f3ABn4y7OGGyL4U6qPIWZAhRmRGzbxsPJvLtwD2sPmtd6cnW0Z3TnuoztVu/i0gjFyT4HO381Pw4shZzzF/d5BJlnGm4yGOpeD/aO5fQurkHKcfjkeshIgg5jof/b137Mgms5Df4PtBl97ccUKWMKMwUozIjYPsMwWH0giXcX7mHLhU7Cns4O3H9DPe69vi6eLlcZQjLTYP8S2PU77FsEmSkX97nUgEb9zcGmfg9wvMItrYqQlwtfD4HYFRDUAu5fAg7OZXPslf+GJZPMkxE+sAwCm5bNcUXKiMJMAQozIlWHYRj8tSuR9xbtYXdCKgA+bo6M7VafuzvVxdWpBJ18czLh0N/mdY12/wkZpy7uc3QzdxpuchNE9AGXYkZVlaeotyHqTfMCnA/9Df4Nyu7YeXkw8xbz0hH+jeDBZdVzAkKptBRmClCYEal68vIM/twezweL93LwZDoAAZ7OjOvRgNvaheLsUMKRS3m5cGSt+YrN7j/M/Uny2TmaV51uMtg8R0tFLdgYuwq+GmQecj70f+bJActa2knzLay0BPNq5UM+KvtziJSSwkwBCjMiVVdObh4/bz7Gh3/t4+jpcwDUquHK4z0bMqx1LRzsS9G51zDMw593/Q67/jAvGWBhgjqdzMGm+XDwDCyT93GJjGSY1gVSj0PLkTB0WvmcB8xXp766CTBg6KfQ8rbyO5dICSjMFKAwI1L1ZeXkMSf6CP9dup/E1EwA6vm7M753BIMig7Erajbhq3Fyz8UrNsc3X9xusoeGveG6URDRr+xmHTYM+O4O80SAfg3gweXg7HHl112LZVNg+VvlcztLpJQUZgpQmBGpPs5n5/LNmsN8HLWf0xnZADQO8uTJ3hH0bhqI6VqHIJ85ArvnwfYf4Wj0xe1ufhA5AlqNgqDIazvH2k9gwbPmie3u/wuCW1zb8a5GoY7GkXDfEnC8iuHvIuVIYaYAhRmR6ictM4fpKw/x2d8HSc3MASCyljePdm9An6aB13alJt/JPeY5bLbOhrQTF7cHtzRfrYm8teQLOhZccqD/u9DhwWuv82qlxMMnXcxDwNs9AAPfq7hzi1yGwkwBCjMi1deZjCw+/fsgM1bFci47F4CIQA8eubEBg1oEl65PzT/l5phHBG3+FvbMhzzzFSHsnaDRAHPH2vo9rrycQmaqed2l5APWWwxy32LzCCeAEd9A05sq9vwiBSjMFKAwIyJJaZlMX3WIr1cftlypqePrxthu9RneplbJRz8VJT0JYuaaV/VOiLm43TPYPBrpujuL7o/y00OwbTZ41YaxK0p+VaesLJoIq/8Dzt4w9m/zjMkiVlCS399Wncd7ypQptGvXDk9PTwICArj55pvZs2dPoTaGYTB58mRCQkJwdXXlxhtvZMeOHVaqWERskZ+HM0/3bcyq53vwdN9G+Lo7cSQ5gxd+jqHbO1F8sfIQGVk5134idz/zatRjV8JDK8wz9rr6Qmo8rPwXTG0DX/QxL4p5vsCEfVtmmYOMyR6Gf269IAPQ82Wo3Q4yz8IP95nXuxKp5Kx6ZaZfv37cfvvttGvXjpycHF588UViYmLYuXMn7u7myZvefvtt3njjDb788ksiIiJ4/fXX+fvvv9mzZw+enp5XPIeuzIjIP2Vk5fDd+jg+/fsAJ1LMo5983Z24t0td7upUF2/XMlzWICcT9i6AzTNh/2LzvDEADq7QdIh5cr7f/w+yM6DHS9D16bI7d2mdPgz/uwHOn4XO/wd9XrN2RVIN2extppMnTxIQEMDy5cvp2rUrhmEQEhLC+PHjefbZZwHIzMwkMDCQt99+m4ceeuiKx1SYEZGiZObk8tOmY0yLOsCR5AzAvEzC3Z3DuLdLOH4eZbR0QL7UBHOH4S0z4dTewvvCu8Jdv1y5b01F2fkbfH+X+XmDXtBlvHn9Ki1KKRXEZsPM/v37adiwITExMTRv3pyDBw9Sv359Nm3aRKtWrSzthgwZQo0aNfjqq6+ueEyFGRG5kpzcPP7YFs9Hy/azLzENABdHO+5oX4cHu9Yj2LuM12kyDDi6ATZ/A9t/Mi+XcP8S8Aou2/Ncq2VT4O93Ll5NqtXGHGoaD6w8oUuqLJsMM4ZhMGTIEE6fPs2KFSsAWL16NV26dOHYsWOEhIRY2j744IMcPnyYhQsXXnKczMxMMjMzLV+npKQQGhqqMCMiV5SXZ7B41wk+WrafbUfPAuBob+KWNrUZ260+YX7lsHZRbrY53JTVpHtlLfkgrJ5qvpqUv+K4XwPo/Bi0vKPsFr4U+Qeb6QBc0Lhx49i2bRvffffdJfv+OdGVYRhFTn41ZcoUvL29LY/Q0NByqVdEqh47OxN9mwXx66Nd+Pre9rQP9yU71+C79XF0fy+Kx2dvZs+FBS7LjL1j5Q0yAL71YNAHMH67uT+PSw1I2g+/Pw7/jjR3bD5/1tpVSjVXKa7MPPbYY/zyyy/8/fffhIeHW7aX5jaTrsyISFmKjk1m6tL9LN970rKtT9NAHuxajzZhPtc+q7CtyUyDTV/Bmo8g5Zh5m5MntL0HOj5S+W6Vic2ymdtMhmHw2GOP8fPPPxMVFUXDhg0v2R8SEsITTzzBM888A0BWVhYBAQHqACwiFWr7sbN8tGw/C3YkkP+vZsva3tx7fTgDIoNxLIsJ+GxJTpZ5WYdVH8LJXeZt9k7Q4jbo8jj4Nyz+9SJXYDNh5pFHHmHWrFn8+uuvNGrUyLLd29sbV1dzh7u3336bKVOmMGPGDBo2bMibb75JVFSUhmaLiFXsT0zls78P8fOWY2TlmDvGBno5c3enuoxsXwcf90p8y6g85OXBvkWw6t9wZM2FjSZzJ+Eu4yG0nRWLE1tmM2GmqMuzM2bMYMyYMYD56swrr7zC//73P06fPk2HDh346KOPaN68+VWdQ2FGRMpDUlomM9cd4es1hzmVZr617eJox7DWtbm3S10aBFz5P1tVzpF15lCz58+L28K6mENNw94a1i0lYjNhpiIozIhIecrMyeWPrfF8sfIQO+MvzurbLaIm914fTteG/tWvX83JPbDqP7BtzsW1qgKamm8/NR4ETu4KNnJFCjMFKMyISEUwDIN1h5KZvvIQi3edsPSraRjgwT1dwhnWuhYujtVsbpazx2Dtx7DxS8hKu7jd3sm8zIOb74U/ff7xtS+4/mObqw/YO1jtrUjFU5gpQGFGRCra4aR0vlwdy/fRcaRnmVfr9nFzZGSHOtzdqS6BXi5WrrCCnTsNG6bDuk8hLaH0x3H2Lhx83PzBN9w8703+w9mj7Oq2dXl5cC7ZPPN0WgKknjD/mXbSvIBo5C3g7m/tKoukMFOAwoyIWEvK+Wy+j47jy9WxHD19DgAHOxODWgRz3/X1iKztbeUKK5hhQFa6+RfsudOQkWx+nnG5rwv8WZJ5bDyDzaHGv+GFgNMQ/OpDjbCqc2UnNxvSEgsHlMv9mZ4IecUsoGrnABH94LqR0LCPec6jSkRhpgCFGRGxttw8g8U7E5i+Mpb1scmW7e3q+nDf9eH0bhqEvZ36kBQpN8ccaP4ZdNITIemAeRK/U/sg41TRx7BzvHAVpyH4NygQdBqYr05U1j486adg87cQu8J8hSU1ATKSgBL86nbzA48g8Aw0/+nuB7Gr4PimAm38zcPqrxsJQVc3wKa8KcwUoDAjIpVJzNGzTF91iN+3Hicnz/zPb60artzZMYzb2oXiW92Gdpelc6cLh5ukfRe/zl+K4XJcvM3BJrglNB8OdTqBnRXnDTIMOLzKfGtu528XO1EXZOcA7gEXA0pRf3oEFH3F5cRO2DoLts4xB8N8QS2g1Z3Q/BZz8LEShZkCFGZEpDI6kXKeb9YcZua6w5zOMP+ycnKwY3CLEO7uFEbL0BrWLbAqycszz1acH24sQWc/nInjkqscXrUhcjhEjqjYqxQZyeZV1TfOKLyqekhr8zpYvvUuhhQ3v7ILXLk5sH+Jef2tPfMvhic7R2jUD64bZV45vYJvQynMFKAwIyKV2fnsXH7fepyv1xwm5tjFviEta3tzd6e6DGwRXP1GQVWk7HOQfMgcHvYvNl8Jybw4xJ6ApuaOss1vAZ+wsj+/YcDRaNgwA3b8dPEKkqM7tLgV2twDIdeV/XmLkpEMMXPNwSZ+68Xt7gHQYoQ52AQ2rZBSFGYKUJgREVtgGAZb4s7w9ZrDzNsWT1aueXZhX3cnRrQNZVSHOoT6ulm5ymog+zzsW2j+hb53IeRmXdwX2tEcMJoOvfbbL+dTIOZ7c4g5sf3i9sBI8zpXkbeCi5V/ZyVshy2zzPMFFeyPFNLKHGqaDzePKisnCjMFKMyIiK05lZbJnOg4Zq49zPGz5v+p25mgR+NA7u4UxvUN/LFTh+Hyd+4M7PrdHDoOrcByO8rOAer3NAeOxgPMkwBereNbzH1hYn6A7HTzNgcXczBocw/Ublv5OiPnZpuXrNgyC/YuuDhCyt4JGg0wB5v6Pcp8tJjCTAEKMyJiq3Jy8/hrdyLfrDnMyv0X/2dcz9+dOzuGMbxNbbxdK9dw2iorJd68sGbMXIjfcnG7o7t5HarIW6F+98v3K8lKN792w4zCI4j8G5mvwrS83TwpoC1IPwXbvjffhip4Ran5LXDLF2V6KoWZAhRmRKQq2J+YxrdrD/PDxqOkZZr/Z+zqaM/NrWpxd6cwmgTr37cKc3KvOdTEzIXThy5ud/ODZkPNHYdD20PiLnNn3q2zL/bDsXeCJjdB23shrHPluwpTEvHbzKFm2/cw8D3z1aUypDBTgMKMiFQlaZk5/Lz5GN+siWXviYtLBLSv68vdncPo2ywIR3srDiuuTgwDjm00h5rtP0L6yYv73PwuzAdzgU+4+SrMdaMq9ay7pZKTZQ5lZTzaSWGmAIUZEamKDMNg7cFkvlkby8IdJ8i9MGdNgKczozqEMapjHfw9nK1cZTWSmwOHlpuDza7fzWtRmezNt6Da3gvh3aw7d40NUpgpQGFGRKq6+LPn+G7dEWatj+NUWiYATvZ2DG4Zwj1d6tK8VjVbNsHasjLMV2z8G4JnkLWrsVkKMwUozIhIdZGVk8f87fFMXxXL1rgzlu3tw325t0tdLZsgNkVhpgCFGRGpjjYdOc2MVbHMj4kvtGzC6M5h3Na2Dt5uGgUllZvCTAEKMyJSnSWcPc83a2OZte6IZdkEV0d7hrepxZjO4TQI8LByhSKXpzBTgMKMiIh52YRftxxjxqpYdiekWrZ3i6jJPV3q0rVhTU3EJ5WKwkwBCjMiIhcZhsGag0lMXxnLX7tPkP8boH5Nd8Z0rsuw1rVxdy7bmVxFSkNhpgCFGRGRyzuclM5Xqw/z/YY4y0R8Xi4O3N6+Dnd3CqO2j9aCEutRmClAYUZEpHhpmTn8sCGOL1fHEpuUAZjXgurTNIi7O4fRvq4vDpqITyqYwkwBCjMiIlcnL88gam8i01fGFloLysvFgRsa1qRrhD9dI2oS7O1qxSqlulCYKUBhRkSk5PaeSGXGqlj+jInn7LnsQvsiAj3oFlGTbhEBtK3rg4ujvZWqlKpMYaYAhRkRkdLLzTPYdvQMy/eeZPnek2yNO0Negd8aLo52dKrnR9eImnSLqEm4vzsmW148USoNhZkCFGZERMrOmYwsVu4/xfI9J/l730lOpGQW2h/q60rXhuZg07mBPx4aGSWlpDBTgMKMiEj5MAyDPSdSLcEm+tBpsnLzLPsd7Ey0CfOhW6OadG1Yk6bBXprLRq6awkwBCjMiIhUjPTOHtQeT+PvCLan8kVH5/D2c6dk4gMEtQ+hYTyOkpHgKMwUozIiIWMfhpHRLsFl9IImMrFzLPn8PJwZEBjO4ZQht6vjoio1cQmGmAIUZERHry8rJIzo2mXkx8cyPibesEwUQ7O3CoBbmYBNZy1sdiAVQmClEYUZEpHLJzs1j1f5T/L41nkU7Eki9MPswQJifG4NbhDC4ZQiNgjytWKVYm8JMAQozIiKV1/nsXJbvPcnvW4/z165EzmVfvBXVMMCDwS3NwSbc392KVYo1KMwUoDAjImIbMrJy+GtXIr9vPU7UnpOFRkY1r+XF4BYhDGwRrDWjqgmFmQIUZkREbM/Zc9ks3nmC37ceZ+X+U+QWmKmvTZgPg1sEM6BFMAGeLlasUsqTwkwBCjMiIrYtOT2L+dvj+X3rcdYdSib/t5adCbpG1OS2tqH0bBKIk4OGelclCjMFKMyIiFQdJ1LOM29bPL9vO87mI2cs2/3cnRjWuha3tQulQYA6DlcFCjMFKMyIiFRNh06l8/2GOH7ceJTE1IvLKrSuU4Pb2oUyqEUI7lpOwWYpzBSgMCMiUrXl5OYRteckczbEsXR3oqV/jZuTPYNaBHNbuzq0rlND89fYGIWZAhRmRESqj8TU8/y06RjfR8dx8FS6ZXuDAA9uaxvK0Na18PdwtmKFcrUUZgpQmBERqX4MwyA69jRzouOYF3Oc89nmYd4OdiZ6NQnktnahdI2oib2WUai0FGYKUJgREaneUs5n8/vW43wfHcfWo2ct24O8XLi1bW1ubRNKHT/NXVPZlOT3t1XHsf39998MHjyYkJAQTCYTv/zyS6H9Y8aMwWQyFXp07NjROsWKiIhN8nJxZFSHMH4ddz0Lxt/APV3qUsPNkYSU8/x36X66vruMkZ+t5ceNRzmdnmXtcqUUrNrNOz09nZYtW3LPPfcwfPjwy7bp168fM2bMsHzt5ORUUeWJiEgV0zjIi0mDm/Fc/8Ys3nmCOdFxrNx/itUHklh9IAk7E7QN86VHkwB6NQmgfk0PdRy2AVYNM/3796d///7FtnF2diYoKKiCKhIRkerA2cGeQS1CGNQihKOnM/hh41EWbE9gd0Iq62OTWR+bzFvzdxPm50bPxoH0ahJAu3BfHO01MV9lVOkH4EdFRREQEECNGjXo1q0bb7zxBgEBAdYuS0REqojaPm6M7xXB+F4RHD2dwV+7EvlrdyJrDyRxOCmD6asOMX3VITxdHOgWUZNeTQK5sVFNarjpTkFlUWk6AJtMJn7++Wduvvlmy7Y5c+bg4eFBWFgYhw4dYuLEieTk5LBx40acnS8/tC4zM5PMzIuTJ6WkpBAaGqoOwCIiUiJpmTms3HeSJbsSWbY7kaQC/Wnyb0f1bBJAzyaB1K/prttRZcwmRzNdLsz8U3x8PGFhYcyePZthw4Zdts3kyZN55ZVXLtmuMCMiIqWVm2ewJe4Mf+06wV+7EtlzIrXQ/rp+bvRsEkjPJgG0q6vbUWWhyoYZgIYNG3L//ffz7LPPXna/rsyIiEh5i0vOYOnuRJbsOsHag0lk5178Verp4sCNjQLo2tCfzg38qVXD1YqV2q6ShJlK32emoKSkJOLi4ggODi6yjbOzc5G3oERERMpCqK8bozvXZXTnuqRl5rBi74XbUXsSSU7P4vetx/l963HAfNWmU31/Otf3o1N9P81AXA6sGmbS0tLYv3+/5etDhw6xZcsWfH198fX1ZfLkyQwfPpzg4GBiY2N54YUX8Pf3Z+jQoVasWkRE5CIPZwf6RwbTPzL4wu2o0yzdncjqA0lsO3qW2KQMYpOO8N36IwA0DvKkU30/Otf3p0M9X7xcHK38DmyfVW8zRUVF0b1790u2jx49mmnTpnHzzTezefNmzpw5Q3BwMN27d+e1114jNDT0qs+hGYBFRMRaUs5nE30omdUHkli1/xS7Ewr3tbEzQWTtGnSu70fn+n60DfPF1cneStVWLjbZZ6a8KMyIiEhlkZSWydqDyaw+YJ6o71CBxTABnOztaFWnBp3r+9O5gR8ta9fAyaF6diZWmClAYUZERCqr42fOsebC7MOrD5wi/uz5QvvdnOxpV9eX6xv407dZULVaQ0phpgCFGRERsQWGYRCblGG5arPmQBLJ/1grqnktLwZEBjMwMpgwP3crVVoxFGYKUJgRERFblJdnsOdEKqv2n2LZnkTWHEgir8Bv7KbBXgxsEcyAyGDC/atesFGYKUBhRkREqoKktEwW7jjBnzHxrDmYRG6BZNMk2IuBkUEMiAymXk0PK1ZZdhRmClCYERGRqiY5PYtFOxKYFxPP6gOFg03jIE8GRJqv2DQIsN1gozBTgMKMiIhUZafTs1i0M4F5MQms3n+KnALBplFgfrAJomGgpxWrLDmFmQIUZkREpLo4k5HFop3mW1Er9xUONg0DPMydh1sEE2EDwUZhpgCFGRERqY7OZmSzaGcC87cnsGLfyULrR4X6utIx3I+O9fzoUM+X2j6Vb8i3wkwBCjMiIlLdnT2XzZILV2xW7DtFVm5eof21fVzpEO5Hx3q+dKznR6iv9cONwkwBCjMiIiIXpWXmEB2bzLqDyaw9mETMsbOFOhAD1KrhSod6vparN6G+rphMpgqtU2GmAIUZERGRoqVl5rDx8GnWHUxi7UHz4pg5/wg3wd4u5ltS4eYrN2F+buUebhRmClCYERERuXoZWfnhxnzlZuvRM4X62wAEebmYr9xcCDjh/u5lHm4UZgpQmBERESm9c1m5bDpymrUHk1h3MJktcWcu6XNze7tQ3hreokzPW5Lf3w5lemYRERGpUlyd7OnSwJ8uDfwBOJ+dH26SWXcwic1xZ2hWy9uqNSrMiIiIyFVzcbSnc31/Ote/GG7yrHyTR2FGRERESs3F0d7aJWBn7QJEREREroXCjIiIiNg0hRkRERGxaQozIiIiYtMUZkRERMSmKcyIiIiITVOYEREREZumMCMiIiI2TWFGREREbJrCjIiIiNg0hRkRERGxaQozIiIiYtMUZkRERMSmVflVs40Ly5KnpKRYuRIRERG5Wvm/t/N/jxenyoeZ1NRUAEJDQ61ciYiIiJRUamoq3t7exbYxGVcTeWxYXl4ex48fx9PTE5PJVKbHTklJITQ0lLi4OLy8vMr02HL19DlUDvocKgd9DpWDPodrZxgGqamphISEYGdXfK+YKn9lxs7Ojtq1a5frOby8vPSXtRLQ51A56HOoHPQ5VA76HK7Nla7I5FMHYBEREbFpCjMiIiJi0xRmroGzszOTJk3C2dnZ2qVUa/ocKgd9DpWDPofKQZ9DxaryHYBFRESkatOVGREREbFpCjMiIiJi0xRmRERExKYpzIiIiIhNU5gppY8//pjw8HBcXFxo06YNK1assHZJ1crkyZMxmUyFHkFBQdYuq8r7+++/GTx4MCEhIZhMJn755ZdC+w3DYPLkyYSEhODq6sqNN97Ijh07rFNsFXalz2HMmDGX/Hx07NjROsVWYVOmTKFdu3Z4enoSEBDAzTffzJ49ewq10c9ExVCYKYU5c+Ywfvx4XnzxRTZv3swNN9xA//79OXLkiLVLq1aaNWtGfHy85RETE2Ptkqq89PR0WrZsydSpUy+7/5133uGDDz5g6tSpREdHExQURO/evS1rpEnZuNLnANCvX79CPx9//vlnBVZYPSxfvpxHH32UtWvXsnjxYnJycujTpw/p6emWNvqZqCCGlFj79u2NsWPHFtrWuHFj47nnnrNSRdXPpEmTjJYtW1q7jGoNMH7++WfL13l5eUZQUJDx1ltvWbadP3/e8Pb2Nj755BMrVFg9/PNzMAzDGD16tDFkyBCr1FOdJSYmGoCxfPlywzD0M1GRdGWmhLKysti4cSN9+vQptL1Pnz6sXr3aSlVVT/v27SMkJITw8HBuv/12Dh48aO2SqrVDhw6RkJBQ6GfD2dmZbt266WfDCqKioggICCAiIoIHHniAxMREa5dU5Z09exYAX19fQD8TFUlhpoROnTpFbm4ugYGBhbYHBgaSkJBgpaqqnw4dOvD111+zcOFCPvvsMxISEujcuTNJSUnWLq3ayv/7r58N6+vfvz8zZ85k6dKlvP/++0RHR9OjRw8yMzOtXVqVZRgGTz75JNdffz3NmzcH9DNRkar8qtnlxWQyFfraMIxLtkn56d+/v+V5ZGQknTp1on79+nz11Vc8+eSTVqxM9LNhfbfddpvlefPmzWnbti1hYWHMmzePYcOGWbGyqmvcuHFs27aNlStXXrJPPxPlT1dmSsjf3x97e/tLUnViYuIl6Vsqjru7O5GRkezbt8/apVRb+aPJ9LNR+QQHBxMWFqafj3Ly2GOP8dtvv7Fs2TJq165t2a6fiYqjMFNCTk5OtGnThsWLFxfavnjxYjp37mylqiQzM5Ndu3YRHBxs7VKqrfDwcIKCggr9bGRlZbF8+XL9bFhZUlIScXFx+vkoY4ZhMG7cOH766SeWLl1KeHh4of36mag4us1UCk8++SR33XUXbdu2pVOnTnz66accOXKEsWPHWru0amPChAkMHjyYOnXqkJiYyOuvv05KSgqjR4+2dmlVWlpaGvv377d8fejQIbZs2YKvry916tRh/PjxvPnmmzRs2JCGDRvy5ptv4ubmxsiRI61YddVT3Ofg6+vL5MmTGT58OMHBwcTGxvLCCy/g7+/P0KFDrVh11fPoo48ya9Ysfv31Vzw9PS1XYLy9vXF1dcVkMulnoqJYdSyVDfvoo4+MsLAww8nJyWjdurVlKJ5UjNtuu80IDg42HB0djZCQEGPYsGHGjh07rF1Wlbds2TIDuOQxevRowzDMQ1EnTZpkBAUFGc7OzkbXrl2NmJgY6xZdBRX3OWRkZBh9+vQxatasaTg6Ohp16tQxRo8ebRw5csTaZVc5l/sMAGPGjBmWNvqZqBgmwzCMio9QIiIiImVDfWZERETEpinMiIiIiE1TmBERERGbpjAjIiIiNk1hRkRERGyawoyIiIjYNIUZERERsWkKMyJS7ZhMJn755RdrlyEiZURhRkQq1JgxYzCZTJc8+vXrZ+3SRMRGaW0mEalw/fr1Y8aMGYW2OTs7W6kaEbF1ujIjIhXO2dmZoKCgQg8fHx/AfAto2rRp9O/fH1dXV8LDw5k7d26h18fExNCjRw9cXV3x8/PjwQcfJC0trVCb6dOn06xZM5ydnQkODmbcuHGF9p86dYqhQ4fi5uZGw4YN+e2338r3TYtIuVGYEZFKZ+LEiQwfPpytW7dy5513cscdd7Br1y4AMjIy6NevHz4+PkRHRzN37lyWLFlSKKxMmzaNRx99lAcffJCYmBh+++03GjRoUOgcr7zyCiNGjGDbtm0MGDCAUaNGkZycXKHvU0TKiLVXuhSR6mX06NGGvb294e7uXujx6quvGoZhXol47NixhV7ToUMH4+GHHzYMwzA+/fRTw8fHx0hLS7PsnzdvnmFnZ2ckJCQYhmEYISEhxosvvlhkDYDx0ksvWb5OS0szTCaTMX/+/DJ7nyJScdRnRkQqXPfu3Zk2bVqhbb6+vpbnnTp1KrSvU6dObNmyBYBdu3bRsmVL3N3dLfu7dOlCXl4ee/bswWQycfz4cXr27FlsDS1atLA8d3d3x9PTk8TExNK+JRGxIoUZEalw7u7ul9z2uRKTyQSAYRiW55dr4+rqelXHc3R0vOS1eXl5JapJRCoH9ZkRkUpn7dq1l3zduHFjAJo2bcqWLVtIT0+37F+1ahV2dnZERETg6elJ3bp1+euvvyq0ZhGxHl2ZEZEKl5mZSUJCQqFtDg4O+Pv7AzB37lzatm3L9ddfz8yZM1m/fj1ffPEFAKNGjWLSpEmMHj2ayZMnc/LkSR577DHuuusuAgMDAZg8eTJjx44lICCA/v37k5qayqpVq3jssccq9o2KSIVQmBGRCrdgwQKCg4MLbWvUqBG7d+8GzCONZs+ezSOPPEJQUBAzZ86kadOmALi5ubFw4UIef/xx2rVrh5ubG8OHD+eDDz6wHGv06NGcP3+ef/3rX0yYMAF/f39uueWWinuDIlKhTIZhGNYuQkQkn8lk4ueff+bmm2+2dikiYiPUZ0ZERERsmsKMiIiI2DT1mRGRSkV3vkWkpHRlRkRERGyawoyIiIjYNIUZERERsWkKMyIiImLTFGZERETEpinMiIiIiE1TmBERERGbpjAjIiIiNk1hRkRERGza/wNoVNb5qNIJiAAAAABJRU5ErkJggg==",
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjMAAAHFCAYAAAAHcXhbAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAB0FklEQVR4nO3ddXhUZ97G8e/EPRBPIIQgCe5OcadQpAq0hQr1vtWlQlukQmXb3e5SFyhbKNSg1LBCcAnuLgmQkCAh7uf9Y2BImiAJSSYT7s91zZXJOWfO/GYy7dw85xGTYRgGIiIiIjbKztoFiIiIiFwPhRkRERGxaQozIiIiYtMUZkRERMSmKcyIiIiITVOYEREREZumMCMiIiI2TWFGREREbJrCjIiIiNg0hRmxWdOnT8dkMlluDg4O1KxZk/vuu48TJ06U6XNlZ2fzyCOPEBwcjL29PS1atCjT88vlzZkzh8aNG+Pq6orJZGLr1q3FHrd7924mTpzI0aNHi+zr3r07TZo0Kd9CgTFjxmAymWjcuDF5eXlF9ptMJp544olyr+NqLv43M2bMmGL3T5482XJMce9naY0ZM4batWuX6rHdu3ene/fuZVaLVC0KM2Lzpk2bxtq1a1m8eDFjx47lu+++o0uXLqSlpZXZc3zyySd89tlnjB8/nlWrVvG///2vzM4tl5eYmMg999xD3bp1WbBgAWvXriUiIqLYY3fv3s2kSZPK9Mu3tHbv3s306dOtXcYVeXp68sMPP5CSklJou2EYTJ8+HS8vLytVJlJyCjNi85o0aUKHDh3o0aMHEyZMYNy4cRw5coR58+Zd97nT09MB2LlzJ66urjzxxBN07NiRpk2bXve5MzIyrvscVd3+/fvJycnh7rvvplu3bnTo0AE3Nzdrl3VF7u7udOnShQkTJlTqv/GQIUMwDIPZs2cX2r506VKOHDnCnXfeaaXKREpOYUaqnA4dOgBw7NgxwPwvzY8//pgWLVrg6upK9erVue222zh8+HChx128FLFixQo6deqEm5sb999/PyaTiS+//JKMjAxL0/vFf3VnZmby0ksvER4ejpOTEzVq1ODxxx8nKSmp0Llr167NoEGD+Pnnn2nZsiUuLi5MmjSJqKgoTCYTs2bN4oUXXiA4OBgPDw8GDx7MqVOnSElJ4aGHHsLPzw8/Pz/uu+8+UlNTC537o48+omvXrgQEBODu7k7Tpk159913ycnJKfb1RUdH06VLF9zc3KhTpw5vv/02+fn5hY5NSkriueeeo06dOjg7OxMQEMDAgQPZu3ev5Zjs7GzeeOMNGjRogLOzM/7+/tx3330kJiZe099p/vz5dOzYETc3Nzw9PenTpw9r16617B8zZgw33XQTAHfeeScmk+mylxmmT5/O7bffDkCPHj2K/J0uupbXnpyczPPPP1/ob/r000+XqKXvnXfe4cSJE3z44YdXPfZanu/222+ncePGhR43ePBgTCYTP/zwg2Xb5s2bMZlM/Prrr1d9Xm9vb4YNG8bXX39daPvXX39N586dL9sC9vXXX9O8eXNcXFzw8fFh2LBh7Nmzp8hx06dPJzIyEmdnZxo2bMiMGTOKPd/1fo5EADBEbNS0adMMwIiOji60/cMPPzQA4/PPPzcMwzDGjh1rODo6Gs8995yxYMECY9asWUaDBg2MwMBAIz4+3vK4bt26GT4+PkZoaKjx3//+11i2bJmxfPlyY+3atcbAgQMNV1dXY+3atcbatWuNhIQEIz8/3+jXr5/h4OBgvPrqq8aiRYuMf/7zn4a7u7vRsmVLIzMz03LusLAwIzg42KhTp47x9ddfG8uWLTM2bNhgLFu2zACMsLAwY8yYMcaCBQuMTz/91PDw8DB69Ohh9OnTx3j++eeNRYsWGe+8845hb29vPPnkk4Ve7zPPPGN88sknxoIFC4ylS5ca//rXvww/Pz/jvvvuK3Rct27dDF9fX6N+/frGp59+aixevNh47LHHDMD45ptvLMclJycbjRs3Ntzd3Y3JkycbCxcuNH766SfjqaeeMpYuXWoYhmHk5eUZ/fv3N9zd3Y1JkyYZixcvNr788kujRo0aRqNGjYz09PQr/u1mzpxpAEbfvn2NefPmGXPmzDFat25tODk5GStXrjQMwzAOHjxofPTRRwZgvPXWW8batWuNXbt2FXu+hIQE46233jIA46OPPir0dyrJa09LSzNatGhh+Pn5GR988IGxZMkS48MPPzS8vb2Nnj17Gvn5+Vd8XaNHjzbc3d0NwzCMYcOGGdWqVTPOnDlj2Q8Yjz/+eImf79NPPzUA4+TJk4ZhGEZOTo7h6elpuLq6GmPHjrWc75133jEcHByM5OTkK9Z5sY6//vrLAIzdu3cbhmEY586dM1xcXIyvv/7aeO+99wzAOHLkiOVxF9/jESNGGL///rsxY8YMo06dOoa3t7exf/9+y3EX/9scMmSI8euvvxrffvutUa9ePSM0NNQICwuzHFeSz1G3bt2Mbt26XfF1yY1LYUZs1sX/Ya5bt87IyckxUlJSjN9++83w9/c3PD09jfj4eGPt2rUGYLz//vuFHhsbG2u4uroa48aNs2zr1q2bARh//fVXkecq+CV10YIFCwzAePfddwttnzNnTqEwZRjmMGNvb2/s27ev0LEXw8zgwYMLbX/66acNwPi///u/QtuHDh1q+Pj4XPY9ycvLM3JycowZM2YY9vb2xtmzZ4u8vvXr1xd6TKNGjYx+/fpZfp88ebIBGIsXL77s83z33XcGYPz000+FtkdHRxuA8fHHH1+xxpCQEKNp06ZGXl6eZXtKSooREBBgdOrUybLt4vvzww8/XPZ8F/3www8GYCxbtqzIvmt97VOmTDHs7OyKBOQff/zRAIw//vjjijUU/Jzs3bvXsLe3N5577jnL/r+HmWt9voMHDxqAMWPGDMMwDGPVqlUGYIwbN84IDw+3PK5Pnz6F3r/LuVhHfn6+ER4ebjz//POGYRjGRx99ZHh4eBgpKSlFwsy5c+cMV1dXY+DAgYXOFRMTYzg7OxsjR440DOPS37dVq1aFwt/Ro0cNR0fHQmGmJJ8jhRm5El1mEpvXoUMHHB0d8fT0ZNCgQQQFBfHnn38SGBjIb7/9hslk4u677yY3N9dyCwoKonnz5kRFRRU6V/Xq1enZs+c1Pe/SpUsBiowIuf3223F3d+evv/4qtL1Zs2aXbbofNGhQod8bNmwIwM0331xk+9mzZwtdatqyZQu33HILvr6+2Nvb4+joyL333kteXh779+8v9PigoCDatWtXpK6Ll+QA/vzzTyIiIujdu/flXjq//fYb1apVY/DgwYXe1xYtWhAUFFTkfS1o3759nDx5knvuuQc7u0v/C/Lw8ODWW29l3bp1lr5KZelaXvtvv/1GkyZNaNGiRaHX1a9fP0wm0xVf199FRkbywAMPMHXqVGJiYoo95lqfr27dutSuXZslS5YAsHjxYpo2bcrdd9/NkSNHOHToEFlZWaxateqKf7e/uzii6X//+x+5ubl89dVX3HHHHXh4eBQ5du3atWRkZBT5vIeGhtKzZ0/L5/3i33fkyJGYTCbLcWFhYXTq1KnI6y/t50ikIIUZsXkzZswgOjqaLVu2cPLkSbZv307nzp0BOHXqFIZhEBgYiKOjY6HbunXrOH36dKFzBQcHX/PznjlzBgcHB/z9/QttN5lMBAUFcebMmWs+t4+PT6HfnZycrrg9MzMTgJiYGLp06WLpn7Fy5Uqio6P56KOPgKKdjH19fYs8t7Ozc6HjEhMTqVmz5mVrBfP7mpSUhJOTU5H3NT4+vsj7WtDF96W49yMkJIT8/HzOnTt3xecvjWt57adOnWL79u1FXpOnpyeGYVzxdRVn4sSJ2Nvb8+qrrxa7vyTP16tXL0tgWLJkCX369KFp06YEBgayZMkSVq9eTUZGRonCDGDpn/LWW2+xefNmHnjggWKPu9rf7eL+iz+DgoKKHPf3bdfzORIpyMHaBYhcr4YNG9KmTZti9/n5+WEymVi5ciXOzs5F9v99W8F/SV6Nr68vubm5JCYmFgo0hmEQHx9P27ZtS33uazVv3jzS0tL4+eefCQsLs2y/3Fws18Lf35/jx49f8Rg/Pz98fX1ZsGBBsfs9PT0v+9iLoSIuLq7IvpMnT2JnZ0f16tVLUHHZ8fPzw9XVtUin2IL7SyI4OJinn36at99+m+eee+66nq9Xr1589dVXbNiwgfXr1/PKK68A0LNnTxYvXsyxY8fw8PCwdIC/VqGhofTu3ZtJkyYRGRlZpPXkoqv93S7WevG4+Pj4Isf9fdv1fI5EClKYkSpt0KBBvP3225w4cYI77rijTM/dq1cv3n33Xb799lueeeYZy/affvqJtLQ0evXqVabPV5yLAalgKDMMgy+++KLU5xwwYACvvfYaS5cuvewlt0GDBjF79mzy8vJo3759ic4fGRlJjRo1mDVrFs8//7zlNaSlpfHTTz9ZRjiV1MX34HqGQw8aNIi33noLX19fwsPDS32egl544QU+//xzXnzxxet6vl69emEymXj11Vexs7Oja9euAPTu3Zt//OMfHDt2jK5du+Lo6FjiGp977jlcXV0tI8KK07FjR1xdXfn2228LHXf8+HGWLl3KbbfdBpj/vsHBwXz33Xc8++yzlr/vsWPHWLNmDSEhIYVef2k/RyIFKcxIlda5c2ceeugh7rvvPjZu3EjXrl1xd3cnLi6OVatW0bRpUx599NFSnbtPnz7069ePF154geTkZDp37sz27duZMGECLVu25J577injV1N8DU5OTowYMYJx48aRmZnJJ598cl2XaZ5++mnmzJnDkCFDePHFF2nXrh0ZGRksX76cQYMG0aNHD+666y5mzpzJwIEDeeqpp2jXrh2Ojo4cP36cZcuWMWTIEIYNG1bs+e3s7Hj33XcZNWoUgwYN4uGHHyYrK4v33nuPpKQk3n777VLVfXGG388//xxPT09cXFwIDw8v9vLSlV77Tz/9RNeuXXnmmWdo1qwZ+fn5xMTEsGjRIp577rkSf+l6eXkxfvz4QoG3NM8XEBBAkyZNWLRoET169LAEvt69e3P27FnOnj3LBx98UKLaLurbty99+/a94jHVqlXj1Vdf5eWXX+bee+9lxIgRnDlzhkmTJuHi4sKECRMA89/39ddf58EHH2TYsGGMHTuWpKQkJk6cWOQy0/V8jkQKsWr3Y5HrcLmh2cX5+uuvjfbt2xvu7u6Gq6urUbduXePee+81Nm7caDmmW7duRuPGjYt9fHGjmQzDMDIyMowXXnjBCAsLMxwdHY3g4GDj0UcfNc6dO1fouLCwMOPmm28u8vjLjda53GubMGGCARiJiYmWbb/++qvRvHlzw8XFxahRo4bxj3/8w/jzzz+LjOy53OsbPXp0oREmhmEeufLUU08ZtWrVMhwdHY2AgADj5ptvNvbu3Ws5Jicnx/jnP/9peW4PDw+jQYMGxsMPP2wcOHCgyPP83bx584z27dsbLi4uhru7u9GrVy9j9erV1/T+XM6///1vIzw83LC3tzcAY9q0aSV+7ampqcYrr7xiREZGGk5OToa3t7fRtGlT45lnnik0lL84l/ucZGVlGeHh4UVGM5X0+Z555hkDMN58881C2+vXr28Axvbt269Y30XF1fF3xQ3NNgzD+PLLL41mzZpZah0yZEixQ+a//PJLo379+oaTk5MRERFhfP3118W+39f6OdJoJrkSk2EYhnVilIiIiMj102gmERERsWkKMyIiImLTFGZERETEpinMiIiIiE1TmBERERGbpjAjIiIiNq3KT5qXn5/PyZMn8fT0LJfp5EVERKTsGYZBSkoKISEhhRalLU6VDzMnT54kNDTU2mWIiIhIKcTGxl518dsqH2YuLlQWGxuLl5eXlasRERGRa5GcnExoaOg1LTha5cPMxUtLXl5eCjMiIiI25lq6iKgDsIiIiNg0hRkRERGxaQozIiIiYtOqfJ8ZERG5fvn5+WRnZ1u7DKlCHB0dsbe3L5NzKcyIiMgVZWdnc+TIEfLz861dilQx1apVIygo6LrngVOYERGRyzIMg7i4OOzt7QkNDb3q5GUi18IwDNLT00lISAAgODj4us6nMCMiIpeVm5tLeno6ISEhuLm5WbscqUJcXV0BSEhIICAg4LouOSlii4jIZeXl5QHg5ORk5UqkKroYkHNycq7rPAozIiJyVVrbTspDWX2uFGZERETEpinMiIiIXEXt2rX597//XSbnioqKwmQykZSUVCbnE3UAFhGRKqp79+60aNGiTEJIdHQ07u7u11+UlAu1zFyHvfHJnErOtHYZIiJSCoZhkJube03H+vv7azRXJaYwU0pv/Lab/v9eyfQ1R61dioiI/M2YMWNYvnw5H374ISaTCZPJxPTp0zGZTCxcuJA2bdrg7OzMypUrOXToEEOGDCEwMBAPDw/atm3LkiVLCp3v75eZTCYTX375JcOGDcPNzY369eszf/78Utf7008/0bhxY5ydnalduzbvv/9+of0ff/wx9evXx8XFhcDAQG677TbLvh9//JGmTZvi6uqKr68vvXv3Ji0trdS12CKFmVJqU7s6AD9sjCU7V7NiisiNwTAM0rNzrXIzDOOa6/zwww/p2LEjY8eOJS4ujri4OEJDQwEYN24cU6ZMYc+ePTRr1ozU1FQGDhzIkiVL2LJlC/369WPw4MHExMRc8TkmTZrEHXfcwfbt2xk4cCCjRo3i7NmzJX5PN23axB133MFdd93Fjh07mDhxIq+++irTp08HYOPGjfzf//0fkydPZt++fSxYsICuXbsCEBcXx4gRI7j//vvZs2cPUVFRDB8+vETvVVWgPjOl1KthIP6eziSmZLFkzykGNr2+2QtFRGxBRk4ejV5baJXn3j25H25O1/a15e3tjZOTE25ubgQFBQGwd+9eACZPnkyfPn0sx/r6+tK8eXPL72+88QZz585l/vz5PPHEE5d9jjFjxjBixAgA3nrrLf773/+yYcMG+vfvX6LX9cEHH9CrVy9effVVACIiIti9ezfvvfceY8aMISYmBnd3dwYNGoSnpydhYWG0bNkSMIeZ3Nxchg8fTlhYGABNmzYt0fNXBWqZKSVHezvuaFMTgFnrr5zeRUSk8mjTpk2h39PS0hg3bhyNGjWiWrVqeHh4sHfv3qu2zDRr1sxy393dHU9PT8v0/CWxZ88eOnfuXGhb586dOXDgAHl5efTp04ewsDDq1KnDPffcw8yZM0lPTwegefPm9OrVi6ZNm3L77bfzxRdfcO7cuRLXYOvUMnMd7mpbi4+jDrHq4GmOnUkjzFc93UWkanN1tGf35H5We+6y8PdRSf/4xz9YuHAh//znP6lXrx6urq7cdtttV10l3NHRsdDvJpOpVItxGoZRZPK4gpeJPD092bx5M1FRUSxatIjXXnuNiRMnEh0dTbVq1Vi8eDFr1qxh0aJF/Pe//2X8+PGsX7+e8PDwEtdiq9Qycx1CfdzoWt8fgO82xFq5GhGR8mcymXBzcrDKraSzxTo5OVmWY7iSlStXMmbMGIYNG0bTpk0JCgri6NGjpXyHSq5Ro0asWrWq0LY1a9YQERFhWa/IwcGB3r178+6777J9+3aOHj3K0qVLAfPfpHPnzkyaNIktW7bg5OTE3LlzK6z+yqDShJkpU6ZgMpl4+umnLdvGjBlj6YV+8dahQwfrFVmMke1rAfDjJnUEFhGpTGrXrs369es5evQop0+fvmyrSb169fj555/ZunUr27ZtY+TIkaVqYSmt5557jr/++ovXX3+d/fv388033zB16lSef/55AH777Tf+85//sHXrVo4dO8aMGTPIz88nMjKS9evX89Zbb7Fx40ZiYmL4+eefSUxMpGHDhhVWf2VQKcJMdHQ0n3/+eaHrjxf179/f0hM9Li6OP/74wwoVFmPv7zDrLnplLibA05nTqdks2h1v7apEROSC559/Hnt7exo1aoS/v/9l+8D861//onr16nTq1InBgwfTr18/WrVqVWF1tmrViu+//57Zs2fTpEkTXnvtNSZPnsyYMWMAqFatGj///DM9e/akYcOGfPrpp3z33Xc0btwYLy8vVqxYwcCBA4mIiOCVV17h/fffZ8CAARVWf2VgMqw8fis1NZVWrVrx8ccf88YbbxSarXHMmDEkJSUxb968Up8/OTkZb29vzp8/j5eXV9kUDbDyffhrMkTezPu+E/jv0oN0rufLzAcrV8uRiMj1yMzM5MiRI4SHh+Pi4mLtcqSKudLnqyTf31ZvmXn88ce5+eab6d27d7H7o6KiCAgIICIigrFjx161p3hWVhbJycmFbuWibk/zzyMruLNVECYTrD54hqOnb6yJikRERKzNqmFm9uzZbN68mSlTphS7f8CAAcycOZOlS5fy/vvvEx0dTc+ePcnKyrrsOadMmYK3t7fldnGSpDIX1BzcfCE7hZppu+gWcbEjsIZpi4jcyB555BE8PDyKvT3yyCPWLq9KstplptjYWNq0acOiRYsskxVdbVGwuLg4wsLCmD17NsOHDy/2mKysrEJhJzk5mdDQ0LK/zATw4wOw80fo8jyLgsby0P824ePuxNqXeuLsUDZDCEVErEmXmUouISHhslcFvLy8CAgIqOCKKq+yusxktXlmNm3aREJCAq1bt7Zsy8vLY8WKFUydOpWsrCzLkLSLgoODCQsL48CBA5c9r7OzM87OzuVWdyH1epnDzKGl9Ow+niAvF+KTM1m06xSDm4dUTA0iIlKpBAQEKLBUMKtdZurVqxc7duxg69atllubNm0YNWoUW7duLRJkAM6cOUNsbCzBwZVk6YA6Pcw/T27BIfMcd7Q1X9LSjMAiIiIVx2phxtPTkyZNmhS6ubu74+vrS5MmTUhNTeX5559n7dq1HD16lKioKAYPHoyfnx/Dhg2zVtmFeQVDQGPAgMPLuLNtKHYmWHv4DIcTU61dnYiIyA3B6qOZLsfe3p4dO3YwZMgQIiIiGD16NBEREaxduxZPT09rl3dJ3QutM4eWUaOaK90jzU2Ls6M1I7CIiEhFqFRrM0VFRVnuu7q6snChdVZmLZF6vWDtVDj0FxgGI9vVYuneBH7cdJzn+kaoI7CIiEg5q7QtMzajVkdwcIGUOEjcS/dIf4K8XDibls2CnZoRWEREpLwpzFwvR1cIu7B0+8G/cLC3484LHYE154yIiO2qXbt2oalCTCbTFWekP3r0KCaTia1bt17X85bVeUriaq+tslOYKQsXZwM+ZF7B9K525o7A6w6f5ZA6AouIVAlxcXFlvubRmDFjGDp0aKFtoaGhxMXF0aRJkzJ9rqpMYaYs1Otl/nlsNeRkEOztSs8G5o7A32mYtohIlRAUFFQh85jZ29sTFBSEg0Ol6tZaqSnMlAX/BuAZArmZELMWgBHtagHw0+bjZObkWbM6EZEbzmeffUaNGjXIz88vtP2WW25h9OjRHDp0iCFDhhAYGIiHhwdt27ZlyZIlVzzn3y/FbNiwgZYtW+Li4kKbNm3YsmVLoePz8vJ44IEHCA8Px9XVlcjISD788EPL/okTJ/LNN9/wyy+/YDKZMJlMREVFFXuZafny5bRr1w5nZ2eCg4N58cUXyc3Ntezv3r07//d//8e4cePw8fEhKCiIiRMnlvyNu2DHjh307NkTV1dXfH19eeihh0hNvXSlISoqinbt2uHu7k61atXo3Lkzx44dA2Dbtm306NEDT09PvLy8aN26NRs3bix1LddCYaYsmEyXLjUd/AuA7pEBhHi7cC49h4W71BFYRKoIw4DsNOvcSrD6zu23387p06dZtmyZZdu5c+dYuHAho0aNIjU1lYEDB7JkyRK2bNlCv379GDx4MDEx19aanpaWxqBBg4iMjGTTpk1MnDiR559/vtAx+fn51KxZk++//57du3fz2muv8fLLL/P9998D8Pzzz3PHHXfQv39/4uLiiIuLo1OnTkWe68SJEwwcOJC2bduybds2PvnkE7766iveeOONQsd98803uLu7s379et59910mT57M4sWLr/k9uyg9PZ3+/ftTvXp1oqOj+eGHH1iyZAlPPPEEALm5uQwdOpRu3bqxfft21q5dy0MPPYTJZAJg1KhR1KxZk+joaDZt2sSLL76Io6NjiesoCbVhlZW6PWDrt3DI/B+OvZ2JO9vW4l9L9jNzfQxDWtSwcoEiImUgJx3estJyLS+fBCf3azrUx8eH/v37M2vWLHr1MncF+OGHH/Dx8aFXr17Y29tb1gUEeOONN5g7dy7z58+3fGlfycyZM8nLy+Prr7/Gzc2Nxo0bc/z4cR599FHLMY6OjkyaNMnye3h4OGvWrOH777/njjvuwMPDA1dXV7KysggKCrrsc3388ceEhoYydepUTCYTDRo04OTJk7zwwgu89tpr2NmZ2yWaNWvGhAkTAKhfvz5Tp07lr7/+ok+fPtf0nhV8bRkZGcyYMQN3d/P7PXXqVAYPHsw777yDo6Mj58+fZ9CgQdStWxeAhg0bWh4fExPDP/7xDxo0aGCppbypZaas1OkBmCBhFyTHAVhmBN5w5CwHE9QRWESkIo0aNYqffvrJsvjwzJkzueuuu7C3tyctLY1x48bRqFEjqlWrhoeHB3v37r3mlpk9e/bQvHlz3NzcLNs6duxY5LhPP/2UNm3a4O/vj4eHB1988cU1P0fB5+rYsaOl5QOgc+fOpKamcvz4ccu2Zs2aFXpccHAwCQkJJXqui8/XvHlzS5C5+Hz5+fns27cPHx8fxowZY2nN+vDDD4mLi7Mc++yzz/Lggw/Su3dv3n77bQ4dOlTiGkpKLTNlxd0XQlrAyS1weBm0GEmQtws9GwSyZM8pvtsQw6uDGlm7ShGR6+PoZm4hsdZzl8DgwYPJz8/n999/p23btqxcuZIPPvgAgH/84x8sXLiQf/7zn9SrVw9XV1duu+02srOzr+ncxjVc8vr+++955plneP/99+nYsSOenp689957rF+/vkSvwzCMQkGm4PMX3P73Szkmk6lIn6HSPl/BcwJMmzaN//u//2PBggXMmTOHV155hcWLF9OhQwcmTpzIyJEj+f333/nzzz+ZMGECs2fPLteliNQyU5bqXhjVdKHfDMCo9uoILCJViMlkvtRjjdtlvmAvx9XVleHDhzNz5ky+++47IiIiaN26NQArV65kzJgxDBs2jKZNmxIUFMTRo0ev+dyNGjVi27ZtZGRkWLatW7eu0DErV66kU6dOPPbYY7Rs2ZJ69eoVaaVwcnIiL+/K3w2NGjVizZo1hQLUmjVr8PT0pEaNsu/C0KhRI7Zu3UpaWppl2+rVq7GzsyMiIsKyrWXLlrz00kusWbOGJk2aMGvWLMu+iIgInnnmGRYtWsTw4cOZNm1amddZkMJMWbrYCfjwMriQhrtG+FOjmitJ6TmaEVhEpIKNGjWK33//na+//pq7777bsr1evXr8/PPPbN26lW3btjFy5MgStWKMHDkSOzs7HnjgAXbv3s0ff/zBP//5z0LH1KtXj40bN7Jw4UL279/Pq6++SnR0dKFjateuzfbt29m3bx+nT58mJyenyHM99thjxMbG8uSTT7J3715++eUXJkyYwLPPPmvpL1OWRo0ahYuLC6NHj2bnzp0sW7aMJ598knvuuYfAwECOHDnCSy+9xNq1azl27BiLFi1i//79NGzYkIyMDJ544gmioqI4duwYq1evJjo6ulCfmvKgMFOWQtuBkwekn4H4bcDFjsDmGYFnac4ZEZEK1bNnT3x8fNi3bx8jR460bP/Xv/5F9erV6dSpE4MHD6Zfv360atXqms/r4eHBr7/+yu7du2nZsiXjx4/nnXfeKXTMI488wvDhw7nzzjtp3749Z86c4bHHHit0zNixY4mMjLT0q1m9enWR56pRowZ//PEHGzZsoHnz5jzyyCM88MADvPLKKyV8N66Nm5sbCxcu5OzZs7Rt25bbbruNXr16MXXqVMv+vXv3cuuttxIREcFDDz3EE088wcMPP4y9vT1nzpzh3nvvJSIigjvuuIMBAwYU6ghdHkzGtVz4s2HJycl4e3tz/vx5vLy8yv8JvxsB+/6AXq9Bl+cAOJWcSae3l5KXb7D4ma7UD6xEq36LiFxBZmYmR44cITw8HBcXF2uXI1XMlT5fJfn+VstMWbPMN7PUsinQy4VeF2YEnqX1mkRERMqUwkxZuxhmYtdDVopl88gLHYF/3nxCHYFFRKTCzJw5Ew8Pj2JvjRs3tnZ5ZUJDs8uab12oXhvOHYWjqyDSvChZl/rmjsAnkjL4Y0ccw1vVtGqZIiJyY7jlllto3759sfvKe2beiqIwUx7q9oSNX5tX0b4QZuztTIxoF8o/F+1n1voYhRkREakQnp6eeHpW7b6ausxUHoqZbwbgjjah2NuZ2HjsHPtPpRTzQBERESkphZnyEN4FTPZw9pD5ctMFAV4u9G54oSOwhmmLiA2p4gNfxUpKM0NxcXSZqTy4eJvnnIlZa77U1OZ+y66R7cNYuOsUP28+zosDGuDiaG/FQkVErszR0RGTyURiYiL+/v6XneZepCQMwyA7O5vExETs7OxwcnK6rvMpzJSXuj2LDTNd6vkR6uNK7NkMft8ex62t1XdGRCove3t7atasyfHjx0s03b/ItXBzc6NWrVrXPZOxwkx5qdsLlr0Jh1dAXi7Ym99qOzsTd7WtxXsL9zFrQ4zCjIhUeh4eHtSvX7/YqfZFSsve3h4HB4cyae1TmCkvIS3AtTpknIMTm6DWpWFxt7epyb8W72fTsXPsi08hMqhq9zIXEdtnb2+Pvb0ui0vlpA7A5cXOHup0N98/VHhUU4CnC30aBQLwnWYEFhERuS4KM+Xp4mzAh5YW2XVxRuCfNh8nI1szAouIiJSWwkx5uhhmTmwyX24qoHNdP2r5uJGSmctv209aoTgREZGqQWGmPHnXBL9IMPLh8PJCu+zsTNzVLhTQ4pMiIiLXQ2GmvNW7MBvw3/rNANze2jwj8JaYJI6eTqvgwkRERKoGhZnyZuk3swz+NoOmv6cznev5AfDrNl1qEhERKQ2FmfIW1hnsneB8LJw+UGT34GbBAMzfdlLThYuIiJSCwkx5c3KDWh3N94sZ1dSvSRBODnYcSEhlnxafFBERKTGFmYpwhX4zXi6O9Ij0B2D+Vl1qEhERKSmFmYpwsd/M0VWQm1Vk9+DmIQD8ul2XmkREREpKYaYiBDYB9wDISYeYdUV292oQiLuTPbFnM9gam1Tx9YmIiNgwhZmKYDJdcTZgVyd7y/IG8zWqSUREpEQUZirKFfrNwKVLTb9tjyMvX5eaRERErpXCTEWp08P8M34HpCYU2d2lvj/ero4kpmSx/siZCi5ORETEdinMVBQPfwhqZr5/aFmR3U4OdgxsGgRoAj0REZGSqDRhZsqUKZhMJp5++mnLNsMwmDhxIiEhIbi6utK9e3d27dplvSKv1xX6zQAMbma+1PTHjniyc/MrqioRERGbVinCTHR0NJ9//jnNmjUrtP3dd9/lgw8+YOrUqURHRxMUFESfPn1ISbHRyeUs/WaWQn7RsNK+ji/+ns6cz8hh1cHECi5ORETENlk9zKSmpjJq1Ci++OILqlevbtluGAb//ve/GT9+PMOHD6dJkyZ88803pKenM2vWLCtWfB1C24OjG6QlQELRFiZ7OxM3N72wvIEm0BMREbkmVg8zjz/+ODfffDO9e/cutP3IkSPEx8fTt29fyzZnZ2e6devGmjVrLnu+rKwskpOTC90qDQdnqN3FfP9g8aOabmlhvtS0aPcpMrLzKqoyERERm2XVMDN79mw2b97MlClTiuyLj48HIDAwsND2wMBAy77iTJkyBW9vb8stNDS0bIu+XlfpN9MytBo1q7uSnp3H0r1FRz2JiIhIYVYLM7GxsTz11FN8++23uLi4XPY4k8lU6HfDMIpsK+ill17i/PnzlltsbGyZ1VwmLvabiVkL2WlFdptMJsucM/O3najIykRERGyS1cLMpk2bSEhIoHXr1jg4OODg4MDy5cv5z3/+g4ODg6VF5u+tMAkJCUVaawpydnbGy8ur0K1S8a0H3qGQlw3Hir9cdsuFMLNsXyLJmTkVWZ2IiIjNsVqY6dWrFzt27GDr1q2WW5s2bRg1ahRbt26lTp06BAUFsXjxYstjsrOzWb58OZ06dbJW2dev4NIGl+k30yDIk/oBHmTn5rNo16kKLE5ERMT2WC3MeHp60qRJk0I3d3d3fH19adKkiWXOmbfeeou5c+eyc+dOxowZg5ubGyNHjrRW2WXjKv1mCl9q0qgmERGRK3GwdgFXMm7cODIyMnjsscc4d+4c7du3Z9GiRXh6elq7tOtTpxuY7OD0Pjh/HLxrFjnkluYhfLB4P6sPnuZMaha+Hs5WKFRERKTyMxmGUaVXNUxOTsbb25vz589Xrv4zX/aG49Fwy3+h1b3FHnLL1FVsP36e14c24Z4OYRVcoIiIiPWU5Pvb6vPM3LDqXhjVdJl+M3BpeYNfNYGeiIjIZSnMWMvFfjOHoyC/+MnxBjUPxmSCDUfPcjIpo+JqExERsSEKM9ZSozU4e0NmEpzcUuwhwd6utK3tA8Dv2+MqsDgRERHboTBjLfYOUKer+f5lRjUBGtUkIiJyFQoz1nSV+WYABjYJwt7OxI4T5zlyuuiMwSIiIjc6hRlrqndhcc3Y9XBqd7GH+Ho4c1M9PwB+VeuMiIhIEQoz1lStFjS8BTBg+TuXPazgpaYqPpJeRESkxBRmrK37i+afu+fBqV3FHtK3cSBODnYcTEhlT1xKxdUmIiJiAxRmrC2wMTQaYr5/mdYZLxdHekT6A/Drdl1qEhERKUhhpjLodrF15pfLts7c0rwGYO43o0tNIiIilyjMVAaBjaDRUPP9qLeLPaRngwDcnew5fi6DzTFJFVaaiIhIZacwU1l0ewEwwZ75EL+zyG5XJ3v6NAoENKpJRESkIIWZyiKwETQear6/vPjWmVtamEc1/b4jjrx8XWoSEREBhZnKxdI68yvE7yiy+6Z6/lRzcyQxJYt1h89UfH0iIiKVkMJMZRLQEBoPM98vpu+Mk4MdA5oEAbrUJCIicpHCTGVzsXVm728Qt73I7osT6P25M57s3PwKLk5ERKTyUZipbAIaQJPh5vvFzDvTPtyXAE9nzmfksGJ/YgUXJyIiUvkozFRGhVpnthXaZW9n4uZmwYAm0BMREQGFmcrJPxKa3Gq+H1W0deaWC5eaFu8+RUZ2XkVWJiIiUukozFRWF1tn9v1epHWmRWg1Qn1cSc/OY8meU9apT0REpJJQmKms/COg6W3m+38b2WQymRjczNw6o1FNIiJyo1OYqcy6jgOTHez7A05uLbTr4gR6UfsSOZ+RY4XiREREKgeFmcrMPwKaFN86ExnoSf0AD7Lz8lm4K94KxYmIiFQOCjOVXbcLrTP7/4STWyybTSaTpSOwLjWJiMiNTGGmsvOrD01vN9//W+vMxQn01hw6w+nUrIquTEREpFJQmLEFF/vO7F8AJzZbNtf2c6dZTW/y8g3+2BFnxQJFRESsR2HGFvjVg2Z3mu//rXVGl5pERORGpzBjK7r+A0z2cGAhnNhk2TyoWQgmE0QfPcfxc+lWLFBERMQ6FGZshW/dYltngrxd6FjHF4D//HXAGpWJiIhYlcKMLen6/IXWmUVw/FLrzHN9IwH4YdNxdp44b63qRERErEJhxpb41oXmd5nvR02xbG4dVp3BzUMwDHj9t90YhmGlAkVERCqewoytudg6c3AxHN9o2fxC/0icHexYf+QsC3dpvSYREblxKMzYGp860HyE+X6B1pma1d0Y26UOAFP+3ENWrlbTFhGRG4PCjC3q+tyF1pklEBtt2fxo97r4ezpz7Ew6M9Ycs2KBIiIiFUdhxhb51IEWRVtn3J0d+Ec/c2fg//x1gDOaFVhERG4ACjO2qsvzYOcAh/6C2A2Wzbe1qknjEC9SsnL515L9VixQRESkYijM2Cqf8GL7ztjZmXh1UCMAZq2PYV98ijWqExERqTAKM7as68XWmaUQs96yuUMdX/o3DiLfgDd+11BtERGp2hRmbFn12tBipPl+gdYZgJcGNsDJ3o6VB04TtS+x4msTERGpIFYNM5988gnNmjXDy8sLLy8vOnbsyJ9//mnZP2bMGEwmU6Fbhw4drFhxJXSx78zhZbB/oWVzmK8793WuDZhbZ3Ly8q1UoIiISPmyapipWbMmb7/9Nhs3bmTjxo307NmTIUOGsGvXLssx/fv3Jy4uznL7448/rFhxJVQ9DNqONd+f+zCcuzQk+/Ge9fB1d+JQYhoz12motoiIVE1WDTODBw9m4MCBREREEBERwZtvvomHhwfr1q2zHOPs7ExQUJDl5uPjY8WKK6k+kyCkJWScg+/vhZxMALxcHHmmTwQA//7rAOfTc6xZpYiISLmoNH1m8vLymD17NmlpaXTs2NGyPSoqioCAACIiIhg7diwJCQlXPE9WVhbJycmFblWegzPcMQNcq0PcVljwomXXXW1DiQz0JCk9hw+1qraIiFRBVg8zO3bswMPDA2dnZx555BHmzp1Lo0bmocUDBgxg5syZLF26lPfff5/o6Gh69uxJVtblJ4ObMmUK3t7elltoaGhFvRTrqlYLbv0SMMGmabB1FgAO9na8MqghADPWHuVQYqoVixQRESl7JsPK43azs7OJiYkhKSmJn376iS+//JLly5dbAk1BcXFxhIWFMXv2bIYPH17s+bKysgqFneTkZEJDQzl//jxeXl7l9joqjah3IOotcHCBB5dAUFMA7p8ezdK9CfRuGMCXo9tauUgREZErS05Oxtvb+5q+v63eMuPk5ES9evVo06YNU6ZMoXnz5nz44YfFHhscHExYWBgHDlz+comzs7NldNTF2w2l6z+gXh/IzYQ590BGEgAvD2yIg52JJXsSWHXgtHVrFBERKUNWDzN/ZxjGZS8jnTlzhtjYWIKDgyu4KhtiZwfDPwfvWnDuCMx7DPLzqRfgwd0dwgDzUO28fE2kJyIiVYNVw8zLL7/MypUrOXr0KDt27GD8+PFERUUxatQoUlNTef7551m7di1Hjx4lKiqKwYMH4+fnx7Bhw6xZduXn5gN3fAP2TrDvd1hjbul6und9vF0d2RufwpzoWCsXKSIiUjasGmZOnTrFPffcQ2RkJL169WL9+vUsWLCAPn36YG9vz44dOxgyZAgRERGMHj2aiIgI1q5di6enpzXLtg01WsGAd833/5oMR1ZQzc2Jp3vXB+D9RftIztRQbRERsX1W7wBc3krSgajKMQzzZaZts8DdHx5eQY57EP3+vYLDiWk83K0OLw1oaO0qRUREirCpDsBSjkwmuPl9CGwCaYnwwxgcyeOVm80BZtqqo8ScSbdykSIiItdHYaaqc3IzT6jn7A2x62Hxa/SIDKBLfT+y8/KZ8ucea1coIiJyXRRmbgS+dWHYJ+b76z7GtGsur9zcCDsT/LkznvWHz1i3PhERkeugMHOjaHAz3PSM+f78J4m0P8mIdrUAeP333eRrqLaIiNgohZkbSY9XoHYXyE6FOffwbLcQPJ0d2HkimZ82H7d2dSIiIqWiMHMjsXeA274Gz2A4vQ/fv57nyZ51AXhv4T7SsnKtXKCIiEjJKczcaDwC4PbpYOcAu37mfsdFhPm6kZCSxWfLD1m7OhERkRJTmLkR1eoAfd8AwGHJq7zT1jw8+7MVhzmRlGHNykREREpMYeZG1f4RaDwc8nNpv+k5+tSyIys3nxd+3E5OXr61qxMREblmCjM3KpMJbvkv+EViSonj3w7/wcMRVh08zWu/7KKKTwwtIiJViMLMjczZA+78Hzi6435yDb82Xo7JBN9tiOGLlYetXZ2IiMg1UZi50flHwpD/AhC+9zM+b5cIwJQ/97JgZ5w1KxMREbkmCjMCTW6F9o8C0PvAZB5t44lhwNNztrItNsm6tYmIiFyFwoyY9ZkEgU0wpZ/hH1kf0T3Cj8ycfB74ZiPHz2kxShERqbwUZsTMwRmGfQb2TtgdWMCnTffSIMiT06lZ3D89muTMHGtXKCIiUiyFGbkkqAn0GA+Ay+LxzBgeSICnM/tPpfL4zM0asi0iIpWSwowU1ulJCO0A2SkE/PUsX93bGldHe1YeOM2E+RqyLSIilY/CjBRmZw/DPgFHdzi2iqaxM/nPiJaYTDBrvYZsi4hI5aMwI0X51IF+b5rv/zWZPn5neeXmRsDFIdvxVixORESkMIUZKV7rMVC/L+Rlwc8PcX+HEO7pEHZhyPYWDdkWEZFKQ2FGindxuQPX6hC/HdOK95gwuBHdI/01ZFtERCoVhRm5PM8gGPQv8/2V7+NwchNTR7ayDNl+YPpGDdkWERGrU5iRK2s8DJreDkY+zH0YD1M2X49pS4CnM/tOpWjItoiIWJ3CjFzdwPfAMwTOHoIlEwip5spXo9tqyLaIiFQKCjNyda7VYehH5vsbPodDS2la05sP72phGbL95coj1q1RRERuWAozcm3q9oS2Y8335z0OGefo2ziI8QMbAvDWn3s0ZFtERKxCYUauXZ9J4FMXUk7CH+MAeOCmcA3ZFhERq1KYkWvn5A7DPweTHez4HnbNxWQyFRqy/eAMDdkWEZGKpTAjJVOzDXR5znz/t2cgJR4Hezv+O6IlDYI8SUzJ4r5p0SSmZFm3ThERuWEozEjJdR0HQc0g4xzMfxIMA08XR74e05ZAL2cOJKQy4ot1JCRnWrtSERG5ASjMSMk5OJkvN9k7w4FFsPkbAEKquTL7oY4Ee7twMCGVuz5fR/x5BRoRESlfCjNSOgENoder5vsLXoaz5qHZ4X7uzHmoIzWquXL4dBp3fr6WE0kZVixURESqOoUZKb0Oj0PYTZCTBnMfgfw8AGr5ujHn4Q6E+rhy7Ew6d362ltiz6hQsIiLlQ2FGSs/ODoZ+DE4eELsO1vzXsqtmdTfmPNSR2r5uHD+XwZ2freXYmTQrFisiIlWVwoxcn+ph0P9t8/1lb0L8TsuukGquzHm4I3X83Tl5PpM7P1vH4cRUKxUqIiJVlcKMXL+Wd0PEAMjLhrkPQ+6lYdmBXi7MfqgD9QM8iE/O5K7P13EwIcWKxYqISFWjMCPXz2SCW/4Dbr5waicsfb3Q7gBPF757qAMNgjxJSMnirs/XsS9egUZERMqGwoyUDY8AGPyh+f6a/8Lvz0FejmW3n4czs8Z2oFGwF6dTsxnxxTp2n0y2UrEiIlKVWDXMfPLJJzRr1gwvLy+8vLzo2LEjf/75p2W/YRhMnDiRkJAQXF1d6d69O7t27bJixXJFDQdD74nm+9Ffwv+GQfpZy24fdydmjW1Ps5renE3LZuSX69h54rx1ahURkSrDqmGmZs2avP3222zcuJGNGzfSs2dPhgwZYgks7777Lh988AFTp04lOjqaoKAg+vTpQ0qKLlFUWjc9A3fNMo9wOroSPu8OCXssu6u5OfG/B9rTIrQaSek5jPxinRanFBGR62IyDMOwdhEF+fj48N5773H//fcTEhLC008/zQsvvABAVlYWgYGBvPPOOzz88MPXdL7k5GS8vb05f/48Xl5e5Vm6FHRqF3w3ApKOmYPNrV9C5ADL7pTMHO6bFs3GY+fwdHZg+v3taB1W3YoFi4hIZVKS7+9StczExsZy/Phxy+8bNmzg6aef5vPPPy/N6QDIy8tj9uzZpKWl0bFjR44cOUJ8fDx9+/a1HOPs7Ey3bt1Ys2bNZc+TlZVFcnJyoZtYQWBjGLsManeB7FRzsFn5PlzIzp4ujnxzfzvahfuQkpXLvV+tZ8ORs1c5qYiISFGlCjMjR45k2bJlAMTHx9OnTx82bNjAyy+/zOTJk0t0rh07duDh4YGzszOPPPIIc+fOpVGjRsTHxwMQGBhY6PjAwEDLvuJMmTIFb29vyy00NLSEr07KjLsv3DMX2jwAGPDXZPjpQcgxL2/g7uzA9Pva0qmuL2nZeYz+egNrD52xbs0iImJzShVmdu7cSbt27QD4/vvvadKkCWvWrGHWrFlMnz69ROeKjIxk69atrFu3jkcffZTRo0eze/duy36TyVToeMMwimwr6KWXXuL8+fOWW2xsbInqkTJm7wiDPoCb3wc7B9j5I3zdH86fAMDNyYGvx7SlS30/MnLyuG/6BlYdOG3lokVExJaUKszk5OTg7OwMwJIlS7jlllsAaNCgAXFxcSU6l5OTE/Xq1aNNmzZMmTKF5s2b8+GHHxIUFARQpBUmISGhSGtNQc7OzpbRURdvUgm0fRDumQeuPhC3Fb7oAbHRALg42vPFvW3o2SCAzJx87v8mmqh9CVYtV0REbEepwkzjxo359NNPWblyJYsXL6Z///4AnDx5El9f3+sqyDAMsrKyCA8PJygoiMWLF1v2ZWdns3z5cjp16nRdzyFWEt4FHloGAY0g9RRMvxm2fgeYA80nd7eiT6NAsnPzeWjGJhbsvPzlRBERkYtKFWbeeecdPvvsM7p3786IESNo3rw5APPnz7dcfroWL7/8MitXruTo0aPs2LGD8ePHExUVxahRozCZTDz99NO89dZbzJ07l507dzJmzBjc3NwYOXJkacqWyqB6bXhgEUTeDHlZMO8RWPQK5Ofh7GDPRyNbMbBpENl5+Tw2cxOzN8RYu2IREankHErzoO7du3P69GmSk5OpXv3ScNqHHnoINze3az7PqVOnuOeee4iLi8Pb25tmzZqxYMEC+vTpA8C4cePIyMjgscce49y5c7Rv355Fixbh6elZmrKlsnD2hDu/hai3YMV75hmDE/bCbV/h5OLNf+5qiafzTuZsjOXFn3dwLj2HR7rVuWJfKRERuXGVap6ZjIwMDMOwBJdjx44xd+5cGjZsSL9+/cq8yOuheWYquZ0/wbzHITcDfOvDyDngWxfDMHh34T4+iToEwNgu4bw0oCF2dgo0IiI3gnKfZ2bIkCHMmDEDgKSkJNq3b8/777/P0KFD+eSTT0pzSrlRNbkV7v8TvGrAmQPmjsGHlmIymXihfwNeubkhAF+sPMI/ftxOTl6+lQsWEZHKplRhZvPmzXTp0gWAH3/8kcDAQI4dO8aMGTP4z3/+U6YFyg0gpKV5gr2a7SDzPHx7K6yZCnk5PNilDu/f3hx7OxM/bT7OI//bRGZOnrUrFhGRSqRUYSY9Pd3Sb2XRokUMHz4cOzs7OnTowLFjx8q0QLlBeAbCmN+gxSgw8mHReHi/AfwxjluDTvHZqFY4O9jx194E7vlqPeczcq5+ThERuSGUKszUq1ePefPmERsby8KFCy1LDiQkJKhfipSegzMM+QgGvAvu/pB+GjZ8Bl/0pPfSQfzVZgORLmeJPnqOOz9bS0JKprUrFhGRSqBUHYB//PFHRo4cSV5eHj179rTMBTNlyhRWrFjBn3/+WeaFlpY6ANuovFw4vAy2zYa9v5s7CF+w2dSQ77M7s92rO5882JMwX3crFioiIuWhJN/fpV41Oz4+nri4OJo3b46dnbmBZ8OGDXh5edGgQYPSnLJcKMxUAVkpsOdXc7A5sgIwf2SzDAdW2bWhfp8HqdVuCDg4WbdOEREpMxUSZi46fvw4JpOJGjVqXM9pyo3CTBVz/gTs/JHcLd/hcHqPZXOuczUcmt4Kze+Cmm1Bc9KIiNi0ch+anZ+fz+TJk/H29iYsLIxatWpRrVo1Xn/9dfLzNXRWypF3Dej8FA5PrCPlvih+dbuVU0Y1HLKSYONX8FUf+G8riHobzh62drUiIlIBStUy89JLL/HVV18xadIkOnfujGEYrF69mokTJzJ27FjefPPN8qi1VNQyU7Vl5uTx5MxoMvZHcav9KgY7bcQh71L/GlqPgQHv6RKUiIiNKffLTCEhIXz66aeW1bIv+uWXX3jsscc4ceJESU9ZbhRmqr7cvHxe+GkHP20+jiuZfNQqjp5ZS+HQUsCAsM5wx//A/foWQRURkYpT7peZzp49W2wn3wYNGnD27NnSnFKk1Bzs7XjvtmaM7RJOBi7cvzmct/3ewhj5PTh7wbHV8GVP8/pPIiJS5ZQqzDRv3pypU6cW2T516lSaNWt23UWJlJSdnYnxNzfixQHmkP3p8kO8uD2I3PsWmlfqPncUvuwN+xdZtU4RESl7pbrMtHz5cm6++WZq1apFx44dMZlMrFmzhtjYWP744w/LUgeVgS4z3Xi+j47lxZ+3k2/ATfX8+Nfgmvj/8RAcWwUmO+j7BnR4TCOeREQqsXK/zNStWzf279/PsGHDSEpK4uzZswwfPpxdu3Yxbdq0UhUtUlbuaBvKJ3e3xsXRjlUHT9P/890sb/8ZtLrXvFTCwpdh/pOQm23tUkVEpAxc9zwzBW3bto1WrVqRl1d5FgJUy8yN62BCCk/M2sLe+BQAHuoSzrhqy3BY8oo51KhjsIhIpVXuLTMitqBegCfzHu/M6I5hAHy+8gjDtzTn1KAZ6hgsIlKFKMxIlebiaM+kIU34/J7WVHNzZPvx8/T8xZG/On17qWPwV33gwGJrlyoiIqWkMCM3hL6Ng/jzqS60C/chLTuPB/5M5VX/D8mr1RmykmHWHbD2Iyi7q64iIlJBStRnZvjw4Vfcn5SUxPLly9VnRiqtvHyD/y49wH/+OkC+AfV9nZhT80d89s02H9DqXhj4vmYMFhGxspJ8fzuU5MTe3t5X3X/vvfeW5JQiFcrezsTTvSPoVNePp2Zv4cCZTDokDWFG45q0P/ABps0z4MxhuGOGOgaLiNiIMh3NVBmpZUYuJyk9m3E/bmfR7lMAPFXrKE8lvYVddqq5P82IORBQdKZrEREpfxrNJHINqrk58dk9rXl9SGOcHOz4MKY2I/NfJ9MjVB2DRURsiMKM3NBMJhP3dKzN/Cc6Uy/Ag3WpgXQ6M55Yz5aXOgavmaqOwSIilZjCjAjQIMiLX5+4iRHtQjlreNEz8RkWu/QzT663aDz8eB9kpVi7TBERKYbCjMgFrk72TBnejI9GtsLFxYWxSfcyhfvINznArrnweQ9I2GPtMkVE5G8UZkT+5uZmwfzxf11oWas6n2X24bbMVzjvGABnDsAXPWHbHGuXKCIiBSjMiBQj1MeN7x/uyGPd67LZiKBHymQ2O7SEnHSY+xD8+jTkZFq7TBERQWFG5LIc7e0Y178B39zfDpO7H7elPsdHxm0YmGDTNPi6H5w7Zu0yRURueAozIlfRLcKfP5/qQvs6/ryXNZwx2eNIs/eGuK3wWVfYv9DaJYqI3NAUZkSuQYCXC98+2J5nekew0mhOn7TX2WMfAZlJ5uHbf02G/MqzjIeIyI1EYUbkGtnbmXiqd31mPtiBXM8a3JL2Ct/m9zPvXPk+/G8opCZYtUYRkRuRwoxICXWs68sfT3WhY0Qwr2SP5snsJ8gyucCRFfBpFzi21toliojcUBRmRErBz8OZ6WPa8kL/BvxBZwZmvs5RU01IjYfpN8Oa/2rWYBGRCqIwI1JKdnYmHu1el+8f7kCGV10GZkxmfn5nMPJg0Ssw527IPG/tMkVEqjyFGZHr1DrMh9//rwudGtbi/7If45Wc+8jBEfb+Bp93h/gd1i5RRKRKU5gRKQPV3Z344t42vDqoMXPoy61ZrxFv8oezh+HL3rBxGqSd1qUnEZFyYDKMqv1/1+TkZLy9vTl//jxeXl7WLkduANtik3jiu82knE3g306f0N1u66WdLtXAtx741Tf/vHjfpy44ulirZBGRSqck398KMyLlIDkzhxd/2s6fO07ysP1vPOAahV/uKUxc7j83E1QLvRBw6l8IO3XN971qgJ0aUUXkxmIzYWbKlCn8/PPP7N27F1dXVzp16sQ777xDZGSk5ZgxY8bwzTffFHpc+/btWbdu3TU9h8KMWIthGHy7PobXf9tNdm4+darZ8dnA6tS3jzcvWnn6IJw5aL5/pY7CDq7mYONXH1qMgvp9Ku5FiIhYic2Emf79+3PXXXfRtm1bcnNzGT9+PDt27GD37t24u7sD5jBz6tQppk2bZnmck5MTPj4+1/QcCjNibbtOnufRbzcTczYdZwc73hjahNvbhF46wDDM/WkuBpvTBy7cPwhnj0B+TuETNhoC/d8Br+CKfSEiIhXIZsLM3yUmJhIQEMDy5cvp2rUrYA4zSUlJzJs3r1TnVJiRyuB8eg7PfL+VpXvNMwSPbF+LCYMb4exgf+UH5uVC0jFzsDm0FDZ8YR767eQJvV6Ftg+C3VXOISJig0ry/V2pLsSfP29uav97q0tUVBQBAQFEREQwduxYEhI0ZbzYFm83R768tw3P9onAZIJZ62O449O1nEjKuPID7R3Ml5gi+sGAd+Dh5VCjDWSnwJ/j4IuecHJLxbwIEZFKqtK0zBiGwZAhQzh37hwrV660bJ8zZw4eHh6EhYVx5MgRXn31VXJzc9m0aRPOzs5FzpOVlUVWVpbl9+TkZEJDQ9UyI5VG1L4Enpq9lfMZOfi4O/HfES3pXM/v2k+Qnw+bpsGSSZB1Hkx20O5h6DkenD3Lr3ARkQpkk5eZHn/8cX7//XdWrVpFzZo1L3tcXFwcYWFhzJ49m+HDhxfZP3HiRCZNmlRku8KMVCaxZ9N55NtN7DqZjJ0Jnu8XyaPd6mIyma79JCmnYOHLsPNH8++eITDgbWh4C5TkPCIilZDNXWZ68sknmT9/PsuWLbtikAEIDg4mLCyMAwcOFLv/pZde4vz585ZbbGxseZQscl1Cfdz46dFO3N66JvkGvLtgHw//bxPJmTlXf/BFnoFw21dw989QPRxSTsL398KsO+HcsfIrXkSkkrFqmDEMgyeeeIKff/6ZpUuXEh4eftXHnDlzhtjYWIKDix/J4ezsjJeXV6GbSGXk4mjPu7c1Y8rwpjjZ27Fo9ymGTF3NvviUkp2oXi94bC10/QfYOcKBhfBxB1j1b8grQTgSEbFRVg0zjz/+ON9++y2zZs3C09OT+Ph44uPjycgwd4pMTU3l+eefZ+3atRw9epSoqCgGDx6Mn58fw4YNs2bpImXCZDIxol0tfnikIyHeLhw5ncbQj1bzy9YTJTuRoyv0fAUeXQ1hN0FOOiyZAJ91g5j15VO8iEglYdU+M5frHzBt2jTGjBlDRkYGQ4cOZcuWLSQlJREcHEyPHj14/fXXCQ0NLfaxf6eh2WIrzqRm8X+zt7D64BkA7utcm5cHNsTRvoT/5jAM2DrLvHJ3xlnzttZjoNcEcLu2+ZlERKzNJjsAlxeFGbElefkG7y/ax8dRhwBoE1adj0e1IsCrFOs2pZ2Bxa/B1m/Nv7v5Qb+3oNkd6iAsIpWewkwBCjNiixbtiue577eRkpWLv6czH41sRbvwUraqHF0Nvz0Dp/eZfw/tYJ5sr+FgLW4pIpWWwkwBCjNiqw4npvLIt5vYfyoVezsTLw9syP2da5ds+PZFudmw5j+w4j3IzTRvc6kGze6EVvdCUJMyrV1E5HopzBSgMCO2LD07lxd/2sH8bScB6NkggEm3NCbUx610Jzx/Arb8D7Z8C+cLTFsQ0socaprcCi7670RErE9hpgCFGbF1hmEwfc1R3vpjDzl5Bi6OdjzZsz5ju9TByaGUAxLz8+DwMtg8A/b+cWkxS0c3aDwcWo+Gmm3Vt0ZErEZhpgCFGakqDiakMH7uTtYfMY9QqhfgwRtDm9Chju/1nTg1EbbPNgeb0/svbfdvYG6taXYXuF/nc4iIlJDCTAEKM1KVGIbB3C0nePP3PZxJywZgeKsavDywIX4eRdcqK+HJIXa9OdTs/BlyLyyCaecIDQeZg014d7CrFBOHi0gVpzBTgMKMVEXn03N4d+FeZm2IwTDA29WRcf0jGdG2FnZ2ZXBpKPM87PjRHGzitl7aXq0WtLwHWowC7xrX/zwiIpehMFOAwoxUZVtizvHKvJ3sOpkMQIvQarw5rAmNQ7zL7knitsHm/8H2782rdAPYOUDfN6H9w+pXIyLlQmGmAIUZqepy8/L537pjvL9oP6lZudiZYEyncJ7tG4GHs0PZPVFOBuyeD5umQcxa87bWY2DgP8HeseyeR0QEhZlCFGbkRnEqOZPJv+3m9+1xAAR6OfPaoMYMbBpUurlpLscwYO1UWPQqYEDtLnDHDC2VICJlSmGmAIUZudGs2J/Ia7/s5OiZdAC6Rvjz+pDGhPm6l+0T7V8IPz4A2SlQvTaMmAMBDcr2OUTkhqUwU4DCjNyIMnPy+CTqEJ9EHSI7Lx8nBzse716PR7rXwdnBvuyeKGEPzLoTko6Bsxfc9jXU71N25xeRG5bCTAEKM3IjO3I6jdd+2cnKA6cBCPdz5/UhTbipvl/ZPUnaGfj+Hji2Gkx20PcN6PCYOgaLyHVRmClAYUZudIZh8Nv2OF7/bTcJKVkADG4ewqs3NyzdatzFyc2G3581L5UA5uHbN38ADk5lc34RueEozBSgMCNilpKZw/uL9jNj7VHyDfBwduC5vhHc0yEMB/symAjPMGDdJ7BoPBj5UKsT3Pk/cC/DViARuWEozBSgMCNS2M4T5xk/byfbYpMAaBzixRtDm9CyVvWyeYIDi+HH+yEr2TzJ3og5ENiobM4tIjcMhZkCFGZEisrPN5gdHcs7C/ZyPiMHkwlGtKvFuH6RVHMrg0tDifvMHYPPHQEnD7j1K4jsf/3nFZEbRkm+v7XIisgNyM7OxMj2tfjruW7c2qomhgGz1sfQ6/3l/LjpONf9bxz/SBi71DwHTXYqfHcXrP6P+VKUiEgZU8uMiLD+8Ble/WUn+0+lAtCutg+vD21CZJDn9Z04Lwf++Id51mAwr+k06F/gcJ2LYopIlafLTAUozIhcm5y8fL5edYR/LzlARk4eDnYmHrgpnP/rVR/361kWwTBgw+ew4EVzx+DQDnDnt+DhX3bFi0iVozBTgMKMSMmcSMpg8q+7WLjrFAAh3i68Nrgx/RoHXt+yCAf/gh/uMy9W6V0LRnwHQU3KqGoRqWoUZgpQmBEpnb/2nGLC/F0cP5cBQI9Ifybd0oRavm6lP+npA+aOwWcPgaM7DHgbmo8E+zJcEFNEqgSFmQIUZkRKLyM7j4+WHeSzFYfIyTNwdrDjiR71eKjbdSyLkH4WfhgDR5abf/etB91ehCbDwa4Ml1oQEZumMFOAwozI9TuUmMqr83ay5tAZwLwswqPd6zKkRUjpQk1ejnmCvVX/goyz5m3+DaD7S9DwFrDTQEuRG53CTAEKMyJlwzAM5m87yeu/7eF0qnlZBD8PZ+7pEMbdHWrh61GKEUpZKbD+U1jzX8g8b94W2BR6joeI/lrfSeQGpjBTgMKMSNlKzsxh1voYpq8+SnxyJgBODnYMb1mD+28KJyKwFMO5M5Jg7Ufm1prsFPO2Gq2hx8tQt5dCjcgNSGGmAIUZkfKRk5fPHzvi+GrVEbYfP2/Z3qW+Hw/cFE63CP+Sj35KPwurPzQP5c5JN28L7QA9X4HwLmVYvYhUdgozBSjMiJQvwzDYeOwcX608wqLd8eRf+D9K/QAP7r8pnGEta+DiWMJ+NakJsOrfEP0l5JkvaRHeFXq8ArXal2n9IlI5KcwUoDAjUnFiz6YzbfVRvt8YS2pWLgA+7k6Mal+LezqGEeDpUrITJsfByvdh03TIzzFvq9cbeoyHGq3KtngRqVQUZgpQmBGpeMmZOXwfHcu01Uc5kWSep8bJ3o7BzUN44KZwGoWU8L/FpBhY8R5smQlGnnlb5EBzn5qgppeOy8s1dyTOTDLfMgr+PF/Mtgvbc7PBrz4ENoHAxuabfwNwLGH4EpEyozBTgMKMiPXk5uWzaPcpvlp1hE3Hzlm2d6zjy4NdwukRGYCdXQn61Zw9DMvfhe1zzEsjAPg3NC9mmZF0qfNwWTDZm+fACboYcC789KqhDskiFUBhpgCFGZHKYUvMOb5adYQ/d8aTd6FjTZMaXkwY3Ji2tX1KdrLE/bD8bdj5M1DM/8KcPMClGrhWM/908b50/+8/XbzBZAeJe+HULji103zLOFf0vGA+vmALTmATCGgITu4lew0ickUKMwUozIhULieSMpix5iiz1seQcqFfzaBmwbw0sCE1qrmW7GRnDsG5I+BS/UI48Tbf7B2vr0jDgJS4AuFml/l2ej/k5xbzABP4hJs7KTcaCrW7aIkGkeukMFOAwoxI5XQmNYv3F+9n9oYY8g1wdrDj4a51eKR7XdycKmkQyM0yB5q/h5zUU4WPc/OFBoOg8VCo3VXBRqQUFGYKUJgRqdx2n0xm8m+7WHfYvKxBkJcLLw5owJAWIde3SndFSk2Ek1tg72+w59dLSzQAuPpAw0HQaAiEd7v+ViORG4TCTAEKMyKVn2EYLNgZz5t/7LGs0t2yVjUmDG5Mi9Bq1i2upPJy4ehK2D3PHGzSz1za51odGtwMjYZBHQUbkStRmClAYUbEdmTm5PHVqiN8tOwg6dnmIdjDW9Xghf4NCPSywWHSeblwbBXsmnch2Jy+tM+lmvlSVKMhUKc7ODhZqUiRyklhpgCFGRHbcyo5k3cX7OOnzccBcHOy57HudXmwS52SzyZcWeTnwbHVF4LNfEhLvLTPxRsibzb3sanTQ8FGBIWZQhRmRGzX1tgkJv+6i80xSQDUqObK+JsbMqBJkO30pylOfh4cW2O+FLV7PqQlXNrn7AX1+5gvR9XrAy76/5bcmEry/W1XQTUVa8qUKbRt2xZPT08CAgIYOnQo+/btK3SMYRhMnDiRkJAQXF1d6d69O7t27bJSxSJSkVqEVuOnRzvx4V0tCPZ24URSBo/N3Mydn69j18nzVz9BZWVnb1448+b34bm9MOYPaPcQeARBVjLs/Al+vB/erQPf3gobv4aUeGtXLVJpWbVlpn///tx11120bduW3Nxcxo8fz44dO9i9ezfu7uYJqN555x3efPNNpk+fTkREBG+88QYrVqxg3759eHp6XvU51DIjUjWkZ+fy6fLDfLb8EFm5+ZhMcGebUJ7rG4m/p7O1yysb+flwYiPs/d08MurMwcL7a7Qxt9g0GAT+EdapUaSC2OxlpsTERAICAli+fDldu3bFMAxCQkJ4+umneeGFFwDIysoiMDCQd955h4cffviq51SYEalaTiRl8Pafe/l120nA3J9mRLtaPNglnGDvEk66V9kl7jeHmn1/wPHowvt860ODgeZgU6MN2JWyoT0jCZKOmde/OnfhZ9Ix86SBNdtBl2fBK+S6X4pISdlsmDl48CD169dnx44dNGnShMOHD1O3bl02b95My5YtLccNGTKEatWq8c033xQ5R1ZWFllZWZbfk5OTCQ0NVZgRqWKij57l9d92s/24+XKTo72JoS1q8HC3utQL8LBydeUgJd4cavb+DkdWQF72pX0egRA5wNyJOLxr4QUys1IuBJS/hZWkY3AuBrKucrnO3hnaPgA3PQMeAeXz2kSKYZNhxjAMhgwZwrlz51i5ciUAa9asoXPnzpw4cYKQkEv/MnjooYc4duwYCxcuLHKeiRMnMmnSpCLbFWZEqh7DMFhx4DQfLzvI+iPmiepMJujXKIhHu9elua3NUXOtMpPh4BJzsDmwyNzP5iInDwhtB+lnzcGl4AR+l+PuD9XCoFotqH7hp4s3bPgSYtaYj3F0g3ZjodNT4O5bPq9LpACbDDOPP/44v//+O6tWraJmzZrApTBz8uRJgoODLceOHTuW2NhYFixYUOQ8apkRuTFtOnaOT5cfYvHuS0sLdKrry2Pd69G5nq9tj366ktxs81w2e3+HvX9Aysmix7hW/1tYCbv0e7Va4ORW/LkNAw4vg6VvmvvygDksdXgUOj5hXg9LpJzYXJh58sknmTdvHitWrCA8PNyyvTSXmf5OfWZEbiz7T6Xw6fJD/LL1pGV17qY1vHm0e136NQ7C3q6KhhowdyCO2wJx28wjoy6Glesd3m0Y5hagpW9A/HbzNhdv6PgkdHgEnK8+GEOkpGwmzBiGwZNPPsncuXOJioqifv36RfaHhITwzDPPMG7cOACys7MJCAhQB2ARuaLj59L5cuURZkfHkJmTD0AdP3ce7laHoS1r4Oxgo5PvWZNhmGcyXvYWJO4xb3P1gZuehrZjL9/CI2UrcR+s+xgwQXAzCGoOgY3AsWp1gLeZMPPYY48xa9YsfvnlFyIjIy3bvb29cXU1/1HeeecdpkyZwrRp06hfvz5vvfUWUVFRGpotItfkTGoW36w5yvQ1R0nOzAUg0MuZB2+qw4j2tfBw1orWJZafB7vmQtSUS8PH3QPMI59a31e4A7KUnZRT5vd88www8grvM9mDX8SFcNPsws+m5kuM5SknE1LjwdEdPPzL9NQ2E2Yudw172rRpjBkzBjC3zkyaNInPPvuMc+fO0b59ez766COaNGlyTc+hMCMiAKlZuXy3PoYvVx3mVLK5X523qyOjO4YxulNtfD2qyFw1FSkvF3Z8D1Fvm0dHAXiGQNfnoeU9WpahrGSnwZqpsPpDyEkzb4u8Gfzqmy/7xW0vvO5XQdVqXQg3zS+FHM9gc0/5K8nJMI+gSz1lHqafcvFnvDm8pFy4ZSaZj+/xCnT7R5m9ZLChMFMRFGZEpKCs3DzmbTnBp8sPc+S0+YvBxdGOW1vV5P6bwqnrXwWHdZe3vBzYOhOWvwfJ5vW0qFYLuo6D5iPAXq1fpZKfB1u+NV/WS70wA3SN1tDndajd+dJxhmEOGnHbL4SbbeafSTHFn9fN71LLjUfQhcBSMKTEQWYJZth2cDF3CO/1aulfazEUZgpQmBGR4uTlGyzcFc/HUQfZeeLS0OYekf48cFOdqj0CqrzkZsGmb2DlP81fkGCep8Z0cUI/w/zFW+j+hd8vdx/A3gnCOkPkQIjsbw5KVZlhwIHFsPi1S32TqoVB7wnQePjVW1UuyjgH8TsuhJwd5oCTuK/oJarLcXABzyBz4PEscLP8HgyegeYV4MvhvxWFmQIUZkTkSgzDYO3hM3y96ih/7T1l+a5tEOTJ/Z3DuaVFiO2u1G0t2emw8StY9S9IP1P25w9scmGSwAEQ3LL0sx9XRnHbYNGrcGS5+XeXatBtHLR9EBzK4FJoTgYk7L7UipNxzjzp4sVw4hFY7iHlWinMFKAwIyLX6sjpNKavPsIPm46Tnm3+16ufhxOj2odxd4ewqrMGVEXJzTJfssB04Uvxwhfj1e5bvkAv3E87bR4avu9PiF0HRv6l5/AIhIj+5labOt3KfkRPbjacPQyJe82tGvk55sszwS3MLURl9WWfFGse+r59DmCYW6PaPwxdniv/TryVlMJMAQozIlJS59NzmB0dwzdrjnLyfCYATvZ23NIihAduCqdhsP5fYjXpZy8Emz/g4F+QnXppn4Mr1O1hbrGp38/cunCtcjLhzAFzYEnceym8nD0M+bnFP8a1urljbXAL88+QFlA9vGQBJ/M8rPwA1n0CeRcmfG16O/R81TzB4Q1MYaYAhRkRKa2cvHwW7Iznq1VH2BqbZNneqa4vD9wUTo/IAOyq8iR8lV1uFhxdZW6x2ffnpc7HF9Voc+lyVEAjc8jISoXT+wuEln1weh+cO1q4xacgJ0/wjwT/BuZLWnHb4NRucyvN3zl7mzvXBjeHkJbmnz51i14Ky82GjV/D8ncuLTkRdhP0fR1qtLrut6YqUJgpQGFGRMrC5phzfLXqCAt2xltmFq7j5859nWtza+uauDlpxI5VGYa5k+v+BeZWm5NbCu/3vtBp+PxlRviAuY+If4NLweXiT6+Qoq0tuVmQsAfitprDzcmtcGrXpdaVgpw8LwWc4Bbmc0VNMbf6gHl+mD6TzZfL1OncQmGmAIUZESlLJ5Iy+GbNUb7bEEPKhUn4vF0dGdGuFvd0DKNGtao1C6vNSo67EGz+hMNRhUOGu3/xocXd//rCRF6OubXn5FZzwInbCvE7ITej+OPdA6DHS9DyXg1fL4bCTAEKMyJSHlKzcvlxYyzT1hzl2Jl0wPw9eFM9P+5qW4vejQK0ZEJlkZ0GMWvNK3/7RVbsqt95uebLWgVbcFLioPld0OlJrWt1BQozBSjMiEh5yss3WLo3gelrjrD64KVhyNXdHBnWsiZ3tg0lMkhfWCIlpTBTgMKMiFSUmDPp/LAplh82Hic+OdOyvUVoNe5qG8qg5iFaC0rkGinMFKAwIyIVLS/fYMX+ROZEx7JkzylyL3QYdnOy5+amwdzVLpRWtaprhmGRK1CYKUBhRkSsKTEli7lbjjM7OpbDiWmW7XX93bmrbS2GtaqBnxa5FClCYaYAhRkRqQwMw2DTsXPMiY7lt+1xZOSYZxh2sDPRp1Egd7QNpWt9f+w1b40IoDBTiMKMiFQ2KZk5/LY9jjnRsYUm4wv2duH21jW5s10tDfGWG57CTAEKMyJSme2LT2FOdCw/bzlOUrp5Rlk7E3SPDGBku1r0aBCg1hq5ISnMFKAwIyK2ICs3j8W7T/HdhphCQ7xDvF24s20t7mwbSpC3ixUrFKlYCjMFKMyIiK05cjqN7zbE8MPGWM5daK2xtzPRs0EAI9vXUt8auSEozBSgMCMitiozJ4+Fu+KZuT6GDUfOWrbXrO7KiHa1uL1NTQI81VojVZPCTAEKMyJSFRw4lcKsDTH8tOk4yRfWhLo4Empk+1p0ruunFbylSlGYKUBhRkSqksycPH7bHses9cfYHJNk2R7m68aIdrW4rXVNzVsjVYLCTAEKMyJSVe2NT2bW+hjmbj5BSpa5tcbR3kS/xkHc1bYWner6qrVGbJbCTAEKMyJS1aVn5/LrtpPMXB/D9uPnLdtrVHPl9jY1ua11TWpWd7NihSIlpzBTgMKMiNxIdp44z3cbYpi/7SQpF/rWmEzQua4ft7epSb/GQbg42lu5SpGrU5gpQGFGRG5EF0dCfb8xttC8NV4uDgxpUYM72oTSpIaXFruUSkthpgCFGRG50cWeTeeHTcf5cWMsJ89nWrY3CPLkzrahDG1Rg+ruTlasUKQohZkCFGZERMzy8g3WHDrN9xuPs3BXPNm5+QA42dvRp1Egt7epSRdNyCeVhMJMAQozIiJFJaVnM3/bSb7fGMvOE8mW7cHeLtzaqia3t6lJmK+7FSuUG53CTAEKMyIiV7br5Hl+2HiceVtPWBa7BLipnh9jOtWmZ4MADfGWCqcwU4DCjIjItcnKzWPJ7gTmbIxl5YFELn47hPm6MbpjbW5vUxNPF0frFik3DIWZAhRmRERKLvZsOt+uO8Z3G2Isyye4O9lze5tQRneqTbifLkFJ+VKYKUBhRkSk9NKzc/l58wmmrznKwYRUy/Yekf7c1zmcLvX9NLxbyoXCTAEKMyIi188wDFYdPM301UdZui/Bcgmqrr87YzqHM7xlDdydHaxbpFQpCjMFKMyIiJSto6fTmL7mKD9uOk7qhTWhPF0cuKttKPd2rE2oj5ZOkOunMFOAwoyISPlIyczhx03H+WbNUY6eSQfAzgS9GwZyX+dwOtTx0SUoKTWFmQIUZkREyld+vkHU/gSmrT7KygOnLdsbBHlyd4cw+jYKJMDLxYoVii1SmClAYUZEpOIcOJXC9DVH+XnzCTJy8izbm4dWo0/DAHo3CiQy0FMtNnJVCjMFKMyIiFS88+k5fL8xlt92xLEtNqnQvprVXendMJA+jQJpF+6Do72ddYqUSk1hpgCFGRER60pIzuSvvQks2X2KVQdPk3VhTSgwdxzuHhlA74YBdI8MwNtVk/KJmc2EmRUrVvDee++xadMm4uLimDt3LkOHDrXsHzNmDN98802hx7Rv355169Zd83MozIiIVB7p2bmsOnCaJXtO8deeBM6kZVv2OdiZaBfuY2m10aioG1tJvr+tOilAWloazZs357777uPWW28t9pj+/fszbdo0y+9OTlqmXkTEVrk5OdC3cRB9GweRl2+wNTaJJXtOsWT3KQ4kpLLm0BnWHDrD5N92ExnoSe9GAfRuGEizmtW0mrdcllXDzIABAxgwYMAVj3F2diYoKKiCKhIRkYpib2eidVh1WodV54X+DTh6Os0cbPacIvroOfadSmHfqRQ+WnaI6m6O3FTfn671/egW4a/RUVJIpZ+uMSoqioCAAKpVq0a3bt148803CQgIuOzxWVlZZGVlWX5PTk6+7LEiIlJ51PZz58EudXiwSx2S0rOJ2pfI4j2nWLEvkXPpOfy67SS/bjsJmId9d4vwp2uEP21qV8fZwd7K1Ys1VZoOwCaTqUifmTlz5uDh4UFYWBhHjhzh1VdfJTc3l02bNuHs7FzseSZOnMikSZOKbFefGRER25STl8/W2CRW7E9kxf5Etp84T8FvLldHezrU8aFrhD/dIvwJ93PX0O8qwGY6ABdUXJj5u7i4OMLCwpg9ezbDhw8v9pjiWmZCQ0MVZkREqoizadmsPJDIiv2nWXEgkcSUrEL7a1Z3pWuEP13r+9Opni9eLhohZYtspgNwSQUHBxMWFsaBAwcue4yzs/NlW21ERMT2+bg7MaRFDYa0qIFhGOyNTzG32hxIJPrIOY6fy2DW+hhmrY/B3s5Eq1rV6Frfn35NgogI9LR2+VIObCrMnDlzhtjYWIKDg61dioiIVAImk4mGwV40DPbi4W51Sc/OZd3hM+ZWm/2JHD6dRvTRc0QfPcf7i/fTKNiLYS1rcEuLEALVibjKsGqYSU1N5eDBg5bfjxw5wtatW/Hx8cHHx4eJEydy6623EhwczNGjR3n55Zfx8/Nj2LBhVqxaREQqKzcnB3o2CKRng0AAYs+ms+JAIsv2JrB8fyK745LZHZfMW3/uoVNdX4a2qEH/JkF46lKUTbNqn5moqCh69OhRZPvo0aP55JNPGDp0KFu2bCEpKYng4GB69OjB66+/Tmho6DU/hybNExERgKT0bH7fEce8LSeIPnrOst3ZwY4+jQIZ1rIGXSP8tbxCJWGTHYDLi8KMiIj8XezZdH7ZeoK5W05wKDHNst3H3YlBzYIZ2rIGLUOraVSUFSnMFKAwIyIil2MYBjtPJDNv6wl+2XqS06mXRkaF+boxtEUNhrasQbifuxWrvDEpzBSgMCMiItciNy+fNYfOMG/LCRbsiic9O8+yr3loNYa1CGFQ8xD8PDRitiIozBSgMCMiIiWVnp3L4t2nmLvlBCsPnCYv3/xV6WBnomeDAO5oE0r3SH8c1L+m3CjMFKAwIyIi1yMxJYvftp9k7pYTbD9+3rLdz8OZW1vV4PY2NakXoPlryprCTAEKMyIiUlb2n0rhh42xzN1ygtOp2ZbtLWtV4442oQxqFqxh3mVEYaYAhRkRESlrOXn5LNubwPcbj7NsX4LlMpSLox0DmwRze5tQ2of7YGen0VClpTBTgMKMiIiUp4SUTOZtOcH3G49zMCHVsr2Wjxu3ta7Jra1rUqOaqxUrtE0KMwUozIiISEUwDIOtsUl8v/E4v247SWpWLgAmE9xUz4/b24TSt1EgLo72Vq7UNijMFKAwIyIiFS0jO48/d8bxw8bjrD18xrLdy8WBW1qE0D0igLa1ffB2U/+ay1GYKUBhRkRErCnmTDo/borlx03HOXk+07LdZIIGQV60D/ehfbgPbcN9NIdNAQozBSjMiIhIZZCXb7Dm0Gn+2BHH+sNnOXw6rcgx9QI8aHch3LQP9yXI+8Zd2VthpgCFGRERqYwSUjLZcOSs5bY3PqXIMWG+brSr7UP7Or60D/ehZnXXG2a9KIWZAhRmRETEFpxLyyb66FnWXwg3u06eJ/9v39Ah3i60C/ehXbgvXer7EerjZp1iK4DCTAEKMyIiYotSMnPYeOwcG46cZf3hM2w/fp7cv6WbBkGe9GkUSO+GgTSt4V2l5rVRmClAYUZERKqC9OxctsQksf7IWdYdOsOmmHOWyfoAAr2c6dUwkD6NAulYx9fmh4ArzBSgMCMiIlXRubRsovYnsHj3KZbvSyStwCrf7k72dI3wp0+jQHpEBlDd3cmKlZaOwkwBCjMiIlLVZeXmsfbQGRbvPsWSPac4lZxl2WdvZ6JNWHX6NDK32oT5ulux0munMFOAwoyIiNxIDMNgx4nzLNl9ikW7TxUZJVU/wMPcz6ZRIC1qVqu0/WwUZgpQmBERkRtZ7Nl0luw5xeLdp1h/5GyhfjZ+Hk50re9P1wh/utT3w7cSTdqnMFOAwoyIiIjZ+fScQv1sUi6sHwXmGYmb1vCma31/ukX60zK0Gg72dlarVWGmAIUZERGRorJz89l07BzL9yeyfH8ie+KSC+33dHGgc10/ukWaW24qeuVvhZkCFGZERESuLiE5kxUHTrN8fyKrDiRyLj2n0P56AR50izAHm/bhPuU+9FthpgCFGRERkZLJyzd3Il6+L5EVBxLZEnOu0GzEzg52dKjjS9cIf7pF+FPX373Ml1lQmClAYUZEROT6nE/PYdXB06y4cEkqPjmz0P672oby9q3NyvQ5S/L97VCmzywiIiJVjrebIzc3C+bmZsEYhsH+U6mWYLPhyFma1axm1foUZkREROSamUwmIoM8iQzyZGzXOqRn5179QeVMYUZERERKzc3J+lHCegPIRURERMqAwoyIiIjYNIUZERERsWkKMyIiImLTFGZERETEpinMiIiIiE1TmBERERGbpjAjIiIiNk1hRkRERGyawoyIiIjYNIUZERERsWkKMyIiImLTFGZERETEpll/qctyZhgGAMnJyVauRERERK7Vxe/ti9/jV1Llw0xKSgoAoaGhVq5ERERESiolJQVvb+8rHmMyriXy2LD8/HxOnjyJp6cnJpOpTM+dnJxMaGgosbGxeHl5lem5b2R6X8uP3tvyo/e2fOh9LT+V/b01DIOUlBRCQkKws7tyr5gq3zJjZ2dHzZo1y/U5vLy8KuUHwdbpfS0/em/Lj97b8qH3tfxU5vf2ai0yF6kDsIiIiNg0hRkRERGxaQoz18HZ2ZkJEybg7Oxs7VKqFL2v5UfvbfnRe1s+9L6Wn6r03lb5DsAiIiJStallRkRERGyawoyIiIjYNIUZERERsWkKMyIiImLTFGZK6eOPPyY8PBwXFxdat27NypUrrV2SzZs4cSImk6nQLSgoyNpl2aQVK1YwePBgQkJCMJlMzJs3r9B+wzCYOHEiISEhuLq60r17d3bt2mWdYm3I1d7XMWPGFPkMd+jQwTrF2pgpU6bQtm1bPD09CQgIYOjQoezbt6/QMfrclty1vK9V4XOrMFMKc+bM4emnn2b8+PFs2bKFLl26MGDAAGJiYqxdms1r3LgxcXFxltuOHTusXZJNSktLo3nz5kydOrXY/e+++y4ffPABU6dOJTo6mqCgIPr06WNZy0yKd7X3FaB///6FPsN//PFHBVZou5YvX87jjz/OunXrWLx4Mbm5ufTt25e0tDTLMfrclty1vK9QBT63hpRYu3btjEceeaTQtgYNGhgvvviilSqqGiZMmGA0b97c2mVUOYAxd+5cy+/5+flGUFCQ8fbbb1u2ZWZmGt7e3sann35qhQpt09/fV8MwjNGjRxtDhgyxSj1VTUJCggEYy5cvNwxDn9uy8vf31TCqxudWLTMllJ2dzaZNm+jbt2+h7X379mXNmjVWqqrqOHDgACEhIYSHh3PXXXdx+PBha5dU5Rw5coT4+PhCn2FnZ2e6deumz3AZiIqKIiAggIiICMaOHUtCQoK1S7JJ58+fB8DHxwfQ57as/P19vcjWP7cKMyV0+vRp8vLyCAwMLLQ9MDCQ+Ph4K1VVNbRv354ZM2awcOFCvvjiC+Lj4+nUqRNnzpyxdmlVysXPqT7DZW/AgAHMnDmTpUuX8v777xMdHU3Pnj3Jysqydmk2xTAMnn32WW666SaaNGkC6HNbFop7X6FqfG6r/KrZ5cVkMhX63TCMItukZAYMGGC537RpUzp27EjdunX55ptvePbZZ61YWdWkz3DZu/POOy33mzRpQps2bQgLC+P3339n+PDhVqzMtjzxxBNs376dVatWFdmnz23pXe59rQqfW7XMlJCfnx/29vZF/iWQkJBQ5F8Mcn3c3d1p2rQpBw4csHYpVcrFEWL6DJe/4OBgwsLC9BkugSeffJL58+ezbNkyatasadmuz+31udz7Whxb/NwqzJSQk5MTrVu3ZvHixYW2L168mE6dOlmpqqopKyuLPXv2EBwcbO1SqpTw8HCCgoIKfYazs7NZvny5PsNl7MyZM8TGxuozfA0Mw+CJJ57g559/ZunSpYSHhxfar89t6VztfS2OLX5udZmpFJ599lnuuece2rRpQ8eOHfn888+JiYnhkUcesXZpNu35559n8ODB1KpVi4SEBN544w2Sk5MZPXq0tUuzOampqRw8eNDy+5EjR9i6dSs+Pj7UqlWLp59+mrfeeov69etTv3593nrrLdzc3Bg5cqQVq678rvS++vj4MHHiRG699VaCg4M5evQoL7/8Mn5+fgwbNsyKVduGxx9/nFmzZvHLL7/g6elpaYHx9vbG1dUVk8mkz20pXO19TU1NrRqfWyuOpLJpH330kREWFmY4OTkZrVq1KjTMTUrnzjvvNIKDgw1HR0cjJCTEGD58uLFr1y5rl2WTli1bZgBFbqNHjzYMwzzMdcKECUZQUJDh7OxsdO3a1dixY4d1i7YBV3pf09PTjb59+xr+/v6Go6OjUatWLWP06NFGTEyMtcu2CcW9r4Axbdo0yzH63Jbc1d7XqvK5NRmGYVRkeBIREREpS+ozIyIiIjZNYUZERERsmsKMiIiI2DSFGREREbFpCjMiIiJi0xRmRERExKYpzIiIiIhNU5gRkRuOyWRi3rx51i5DRMqIwoyIVKgxY8ZgMpmK3Pr372/t0kTERmltJhGpcP3792fatGmFtjk7O1upGhGxdWqZEZEK5+zsTFBQUKFb9erVAfMloE8++YQBAwbg6upKeHg4P/zwQ6HH79ixg549e+Lq6oqvry8PPfQQqamphY75+uuvady4Mc7OzgQHB/PEE08U2n/69GmGDRuGm5sb9evXZ/78+eX7okWk3CjMiEil8+qrr3Lrrbeybds27r77bkaMGMGePXsASE9Pp3///lSvXp3o6Gh++OEHlixZUiisfPLJJzz++OM89NBD7Nixg/nz51OvXr1CzzFp0iTuuOMOtm/fzsCBAxk1ahRnz56t0NcpImXE2itdisiNZfTo0Ya9vb3h7u5e6DZ58mTDMMyr/D7yyCOFHtO+fXvj0UcfNQzDMD7//HOjevXqRmpqqmX/77//btjZ2Rnx8fGGYRhGSEiIMX78+MvWABivvPKK5ffU1FTDZDIZf/75Z5m9ThGpOOozIyIVrkePHnzyySeFtvn4+Fjud+zYsdC+jh07snXrVgD27NlD8+bNcXd3t+zv3Lkz+fn57Nu3D5PJxMmTJ+nVq9cVa2jWrJnlvru7O56eniQkJJT2JYmIFSnMiEiFc3d3L3LZ52pMJhMAhmFY7hd3jKur6zWdz9HRschj8/PzS1STiFQO6jMjIpXOunXrivzeoEEDABo1asTWrVtJS0uz7F+9ejV2dnZERETg6elJ7dq1+euvvyq0ZhGxHrXMiEiFy8rKIj4+vtA2BwcH/Pz8APjhhx9o06YNN910EzNnzmTDhg189dVXAIwaNYoJEyYwevRoJk6cSGJiIk8++ST33HMPgYGBAEycOJFHHnmEgIAABgwYQEpKCqtXr+bJJ5+s2BcqIhVCYUZEKtyCBQsIDg4utC0yMpK9e/cC5pFGs2fP5rHHHiMoKIiZM2fSqFEjANzc3Fi4cCFPPfUUbdu2xc3NjVtvvZUPPvjAcq7Ro0eTmZnJv/71L55//nn8/Py47bbbKu4FikiFMhmGYVi7CBGRi0wmE3PnzmXo0KHWLkVEbIT6zIiIiIhNU5gRERERm6Y+MyJSqejKt4iUlFpmRERExKYpzIiIiIhNU5gRERERm6YwIyIiIjZNYUZERERsmsKMiIiI2DSFGREREbFpCjMiIiJi0xRmRERExKb9P940zkL/XpQXAAAAAElFTkSuQmCC",
       "text/plain": [
        "<Figure size 640x480 with 1 Axes>"
       ]
@@ -877,38 +889,47 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 16,
    "id": "b0fbfa80",
    "metadata": {},
    "outputs": [
     {
-     "ename": "NameError",
-     "evalue": "name 'model2' is not defined",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
-      "Cell \u001b[1;32mIn[10], line 1\u001b[0m\n\u001b[1;32m----> 1\u001b[0m model2\u001b[38;5;241m.\u001b[39mload_state_dict(torch\u001b[38;5;241m.\u001b[39mload(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m./model_cifar_exo1.pt\u001b[39m\u001b[38;5;124m\"\u001b[39m))\n\u001b[0;32m      3\u001b[0m \u001b[38;5;66;03m# track test loss\u001b[39;00m\n\u001b[0;32m      4\u001b[0m test_loss \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0.0\u001b[39m\n",
-      "\u001b[1;31mNameError\u001b[0m: name 'model2' is not defined"
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Test Loss: 19.006048\n",
+      "\n",
+      "Test Accuracy of airplane: 72% (729/1000)\n",
+      "Test Accuracy of automobile: 79% (791/1000)\n",
+      "Test Accuracy of  bird: 60% (601/1000)\n",
+      "Test Accuracy of   cat: 48% (484/1000)\n",
+      "Test Accuracy of  deer: 64% (641/1000)\n",
+      "Test Accuracy of   dog: 61% (610/1000)\n",
+      "Test Accuracy of  frog: 72% (727/1000)\n",
+      "Test Accuracy of horse: 71% (716/1000)\n",
+      "Test Accuracy of  ship: 87% (875/1000)\n",
+      "Test Accuracy of truck: 77% (775/1000)\n",
+      "\n",
+      "Test Accuracy (Overall): 69% (6949/10000)\n"
      ]
     }
    ],
    "source": [
-    "model2.load_state_dict(torch.load(\"./model_cifar_exo1.pt\"))\n",
+    "model.load_state_dict(torch.load(\"./model_cifar_exo1.pt\"))\n",
     "\n",
     "# track test loss\n",
     "test_loss = 0.0\n",
     "class_correct2 = list(0.0 for i in range(10))\n",
     "class_total2 = list(0.0 for i in range(10))\n",
     "\n",
-    "model2.eval()\n",
+    "model.eval()\n",
     "# iterate over test data\n",
     "for data, target in test_loader:\n",
     "    # move tensors to GPU if CUDA is available\n",
     "    if train_on_gpu:\n",
     "        data, target = data.cuda(), target.cuda()\n",
     "    # forward pass: compute predicted outputs by passing inputs to the model\n",
-    "    output = model2(data)\n",
+    "    output = model(data)\n",
     "    # calculate the batch loss\n",
     "    loss = criterion(output, target)\n",
     "    # update test loss\n",
@@ -1043,6 +1064,8 @@
    "source": [
     "import os\n",
     "\n",
+    "model = Net()\n",
+    "model.load_state_dict(torch.load(\"./model_cifar_exo1.pt\"))\n",
     "\n",
     "def print_size_of_model(model, label=\"\"):\n",
     "    torch.save(model.state_dict(), \"temp.p\")\n",
@@ -1221,7 +1244,7 @@
     "\n",
     "fig, ax = plt.subplots()\n",
     "r1 = ax.bar(x - width/2, accuracy_initial, width, label=\"Initial Model\")\n",
-    "r2 = ax.bar(x + width/2, accuracy_quant, width, label=\"New Model\")\n",
+    "r2 = ax.bar(x + width/2, accuracy_quant, width, label=\"Quantized Model\")\n",
     "\n",
     "ax.set_xlabel('Classes')\n",
     "ax.set_ylabel('Accuracy')\n",
-- 
GitLab