From 26b2597cb1843e7ac4cdf1b779decce462413c40 Mon Sep 17 00:00:00 2001
From: Khalil <medkhalilbabay@gmail.com>
Date: Fri, 1 Dec 2023 15:54:28 +0100
Subject: [PATCH] commit

---
 TD2 Deep Learning.ipynb        | 1404 +++++++++++++++++++++++++++++---
 dog.png => images_test/dog.png |  Bin
 images_test/husky.jpeg         |  Bin 0 -> 50279 bytes
 images_test/toilet_paper.png   |  Bin 0 -> 10660 bytes
 4 files changed, 1284 insertions(+), 120 deletions(-)
 rename dog.png => images_test/dog.png (100%)
 create mode 100644 images_test/husky.jpeg
 create mode 100644 images_test/toilet_paper.png

diff --git a/TD2 Deep Learning.ipynb b/TD2 Deep Learning.ipynb
index 2ecfce9..c44f423 100644
--- a/TD2 Deep Learning.ipynb	
+++ b/TD2 Deep Learning.ipynb	
@@ -33,10 +33,35 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 3,
    "id": "330a42f5",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Requirement already satisfied: torch in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (2.1.1+cu118)\n",
+      "Requirement already satisfied: torchvision in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (0.16.1+cu118)\n",
+      "Requirement already satisfied: filelock in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (from torch) (3.9.0)\n",
+      "Requirement already satisfied: typing-extensions in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (from torch) (4.8.0)\n",
+      "Requirement already satisfied: sympy in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (from torch) (1.12)\n",
+      "Requirement already satisfied: networkx in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (from torch) (3.0)\n",
+      "Requirement already satisfied: jinja2 in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (from torch) (3.1.2)\n",
+      "Requirement already satisfied: fsspec in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (from torch) (2023.4.0)\n",
+      "Requirement already satisfied: numpy in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (from torchvision) (1.26.2)\n",
+      "Requirement already satisfied: requests in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (from torchvision) (2.28.1)\n",
+      "Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (from torchvision) (10.1.0)\n",
+      "Requirement already satisfied: MarkupSafe>=2.0 in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (from jinja2->torch) (2.1.3)\n",
+      "Requirement already satisfied: charset-normalizer<3,>=2 in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (from requests->torchvision) (2.1.1)\n",
+      "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (from requests->torchvision) (3.4)\n",
+      "Requirement already satisfied: urllib3<1.27,>=1.21.1 in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (from requests->torchvision) (1.26.13)\n",
+      "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (from requests->torchvision) (2022.12.7)\n",
+      "Requirement already satisfied: mpmath>=0.19 in c:\\users\\lenovo\\anaconda3\\envs\\new\\lib\\site-packages (from sympy->torch) (1.3.0)\n",
+      "Note: you may need to restart the kernel to use updated packages.\n"
+     ]
+    }
+   ],
    "source": [
     "%pip install torch torchvision"
    ]
@@ -52,10 +77,72 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 4,
    "id": "b1950f0a",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "tensor([[ 0.6005, -0.8262, -1.1444,  1.8317,  0.2427,  1.2135,  0.5529,  0.3090,\n",
+      "          0.1768, -0.0671],\n",
+      "        [ 1.5132,  0.6618,  1.2110,  1.8297, -0.4285, -0.4998,  0.4473, -0.2253,\n",
+      "          1.4177, -0.0405],\n",
+      "        [-0.0883,  0.6764,  0.3432, -1.2304,  2.6615,  1.1792,  1.4577,  1.5665,\n",
+      "         -1.7107, -0.2310],\n",
+      "        [-1.2376,  0.3868,  0.4568, -1.3576,  1.7694,  1.4209,  2.0126, -0.2384,\n",
+      "         -0.9737, -0.5757],\n",
+      "        [ 0.1787, -0.5972,  0.9511,  0.0971,  0.0702,  0.6209, -0.5282,  0.1848,\n",
+      "         -0.4190,  3.4575],\n",
+      "        [ 0.3149,  2.1750, -1.6734, -0.0107, -0.2639,  0.3729, -0.3992,  1.0509,\n",
+      "         -0.1983, -0.5388],\n",
+      "        [-0.4991, -0.0539,  1.3420,  1.1376,  1.0812, -0.9487,  0.2711, -0.1039,\n",
+      "          0.6608,  0.1926],\n",
+      "        [-0.8309,  0.3424,  0.7537,  1.1209, -0.6249,  0.9049,  0.0279, -0.9683,\n",
+      "          1.5207,  1.3997],\n",
+      "        [-0.8293, -0.9169, -0.8743,  1.1780, -0.6684,  1.0099, -0.9231, -0.9749,\n",
+      "          0.6912, -0.9500],\n",
+      "        [ 0.4564,  0.5852,  1.0717, -0.9455, -0.6503,  1.3128,  0.3559,  0.1450,\n",
+      "          0.0160,  1.7602],\n",
+      "        [-1.3473, -0.6786,  0.2884,  1.3234,  0.6250, -0.0373,  0.5641,  0.5425,\n",
+      "          1.3457,  2.0471],\n",
+      "        [ 0.7055,  0.2446,  2.1138,  0.1851, -1.1334, -1.0468,  1.2929,  0.5373,\n",
+      "          0.9560,  0.1556],\n",
+      "        [-0.6760,  1.7166,  0.8109, -0.9636, -0.9815, -0.4989,  0.1184, -0.7103,\n",
+      "          0.3448,  0.4121],\n",
+      "        [ 0.6817, -1.2063,  0.8555, -0.7569, -1.3178, -0.9650, -1.0587,  1.2064,\n",
+      "         -0.0845,  0.6830]])\n",
+      "AlexNet(\n",
+      "  (features): Sequential(\n",
+      "    (0): Conv2d(3, 64, kernel_size=(11, 11), stride=(4, 4), padding=(2, 2))\n",
+      "    (1): ReLU(inplace=True)\n",
+      "    (2): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
+      "    (3): Conv2d(64, 192, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))\n",
+      "    (4): ReLU(inplace=True)\n",
+      "    (5): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
+      "    (6): Conv2d(192, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
+      "    (7): ReLU(inplace=True)\n",
+      "    (8): Conv2d(384, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
+      "    (9): ReLU(inplace=True)\n",
+      "    (10): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
+      "    (11): ReLU(inplace=True)\n",
+      "    (12): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
+      "  )\n",
+      "  (avgpool): AdaptiveAvgPool2d(output_size=(6, 6))\n",
+      "  (classifier): Sequential(\n",
+      "    (0): Dropout(p=0.5, inplace=False)\n",
+      "    (1): Linear(in_features=9216, out_features=4096, bias=True)\n",
+      "    (2): ReLU(inplace=True)\n",
+      "    (3): Dropout(p=0.5, inplace=False)\n",
+      "    (4): Linear(in_features=4096, out_features=4096, bias=True)\n",
+      "    (5): ReLU(inplace=True)\n",
+      "    (6): Linear(in_features=4096, out_features=1000, bias=True)\n",
+      "  )\n",
+      ")\n"
+     ]
+    }
+   ],
    "source": [
     "import torch\n",
     "\n",
@@ -95,10 +182,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 5,
    "id": "6e18f2fd",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "CUDA is available!  Training on GPU ...\n"
+     ]
+    }
+   ],
    "source": [
     "import torch\n",
     "\n",
@@ -121,10 +216,19 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 6,
    "id": "462666a2",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Files already downloaded and verified\n",
+      "Files already downloaded and verified\n"
+     ]
+    }
+   ],
    "source": [
     "import numpy as np\n",
     "from torchvision import datasets, transforms\n",
@@ -193,10 +297,25 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 7,
    "id": "317bf070",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Net(\n",
+      "  (conv1): Conv2d(3, 6, kernel_size=(5, 5), stride=(1, 1))\n",
+      "  (pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
+      "  (conv2): Conv2d(6, 16, kernel_size=(5, 5), stride=(1, 1))\n",
+      "  (fc1): Linear(in_features=400, out_features=120, bias=True)\n",
+      "  (fc2): Linear(in_features=120, out_features=84, bias=True)\n",
+      "  (fc3): Linear(in_features=84, out_features=10, bias=True)\n",
+      ")\n"
+     ]
+    }
+   ],
    "source": [
     "import torch.nn as nn\n",
     "import torch.nn.functional as F\n",
@@ -242,10 +361,58 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 8,
    "id": "4b53f229",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch: 0 \tTraining Loss: 43.400863 \tValidation Loss: 38.021109\n",
+      "Validation loss decreased (inf --> 38.021109).  Saving model ...\n",
+      "Epoch: 1 \tTraining Loss: 35.353367 \tValidation Loss: 32.136121\n",
+      "Validation loss decreased (38.021109 --> 32.136121).  Saving model ...\n",
+      "Epoch: 2 \tTraining Loss: 30.913728 \tValidation Loss: 29.369840\n",
+      "Validation loss decreased (32.136121 --> 29.369840).  Saving model ...\n",
+      "Epoch: 3 \tTraining Loss: 28.601452 \tValidation Loss: 27.490179\n",
+      "Validation loss decreased (29.369840 --> 27.490179).  Saving model ...\n",
+      "Epoch: 4 \tTraining Loss: 26.915667 \tValidation Loss: 26.303214\n",
+      "Validation loss decreased (27.490179 --> 26.303214).  Saving model ...\n",
+      "Epoch: 5 \tTraining Loss: 25.459954 \tValidation Loss: 25.887691\n",
+      "Validation loss decreased (26.303214 --> 25.887691).  Saving model ...\n",
+      "Epoch: 6 \tTraining Loss: 24.229395 \tValidation Loss: 24.873652\n",
+      "Validation loss decreased (25.887691 --> 24.873652).  Saving model ...\n",
+      "Epoch: 7 \tTraining Loss: 23.164625 \tValidation Loss: 23.284470\n",
+      "Validation loss decreased (24.873652 --> 23.284470).  Saving model ...\n",
+      "Epoch: 8 \tTraining Loss: 22.206993 \tValidation Loss: 23.291942\n",
+      "Epoch: 9 \tTraining Loss: 21.421621 \tValidation Loss: 23.335741\n",
+      "Epoch: 10 \tTraining Loss: 20.651834 \tValidation Loss: 22.401363\n",
+      "Validation loss decreased (23.284470 --> 22.401363).  Saving model ...\n",
+      "Epoch: 11 \tTraining Loss: 19.985618 \tValidation Loss: 24.326989\n",
+      "Epoch: 12 \tTraining Loss: 19.291143 \tValidation Loss: 21.739429\n",
+      "Validation loss decreased (22.401363 --> 21.739429).  Saving model ...\n",
+      "Epoch: 13 \tTraining Loss: 18.624268 \tValidation Loss: 21.852032\n",
+      "Epoch: 14 \tTraining Loss: 17.965765 \tValidation Loss: 21.446357\n",
+      "Validation loss decreased (21.739429 --> 21.446357).  Saving model ...\n",
+      "Epoch: 15 \tTraining Loss: 17.394854 \tValidation Loss: 22.545597\n",
+      "Epoch: 16 \tTraining Loss: 16.794884 \tValidation Loss: 22.042855\n",
+      "Epoch: 17 \tTraining Loss: 16.227497 \tValidation Loss: 22.688590\n",
+      "Epoch: 18 \tTraining Loss: 15.704653 \tValidation Loss: 22.186874\n",
+      "Epoch: 19 \tTraining Loss: 15.155005 \tValidation Loss: 22.500867\n",
+      "Epoch: 20 \tTraining Loss: 14.652102 \tValidation Loss: 22.332764\n",
+      "Epoch: 21 \tTraining Loss: 14.156440 \tValidation Loss: 22.655204\n",
+      "Epoch: 22 \tTraining Loss: 13.740071 \tValidation Loss: 22.890621\n",
+      "Epoch: 23 \tTraining Loss: 13.230510 \tValidation Loss: 23.827616\n",
+      "Epoch: 24 \tTraining Loss: 12.847678 \tValidation Loss: 23.486764\n",
+      "Epoch: 25 \tTraining Loss: 12.306719 \tValidation Loss: 24.269633\n",
+      "Epoch: 26 \tTraining Loss: 11.897007 \tValidation Loss: 24.318152\n",
+      "Epoch: 27 \tTraining Loss: 11.547802 \tValidation Loss: 24.906728\n",
+      "Epoch: 28 \tTraining Loss: 11.044322 \tValidation Loss: 25.903962\n",
+      "Epoch: 29 \tTraining Loss: 10.720178 \tValidation Loss: 25.918305\n"
+     ]
+    }
+   ],
    "source": [
     "import torch.optim as optim\n",
     "\n",
@@ -253,7 +420,8 @@
     "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",
+    "train_loss_list = []  # list to store training loss to visualize\n",
+    "valid_loss_list = []    # list to store validation loss to visualize\n",
     "valid_loss_min = np.Inf  # track change in validation loss\n",
     "\n",
     "for epoch in range(n_epochs):\n",
@@ -297,6 +465,7 @@
     "    train_loss = train_loss / len(train_loader)\n",
     "    valid_loss = valid_loss / len(valid_loader)\n",
     "    train_loss_list.append(train_loss)\n",
+    "    valid_loss_list.append(valid_loss) # I added this list to plot it later with the train loss\n",
     "\n",
     "    # Print training/validation statistics\n",
     "    print(\n",
@@ -324,16 +493,42 @@
     "Does overfit occur? If so, do an early stopping."
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "---\n",
+    " Answer to the question: \n",
+    "During the training process, the model's parameters are adjusted to minimize the training loss.\n",
+    "If the model is trained for too many epochs, it may start fitting the noise in the training data, leading to a decrease in training loss but an increase in validation loss.\n",
+    "Overfitting is often identified when the training loss continues to decrease while the validation loss starts to increase, indicating that the model is becoming too specialized for the training data. \n",
+    "To adress overfitting, we perform an early stopping when the validation loss starts to increase. \n",
+    "---\n"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 9,
    "id": "d39df818",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
    "source": [
     "import matplotlib.pyplot as plt\n",
     "\n",
-    "plt.plot(range(n_epochs), train_loss_list)\n",
+    "\n",
+    "plt.plot(range(n_epochs), train_loss_list, label='training loss')\n",
+    "plt.plot(range(n_epochs),valid_loss_list,label='validation loss')\n",
     "plt.xlabel(\"Epoch\")\n",
     "plt.ylabel(\"Loss\")\n",
     "plt.title(\"Performance of Model 1\")\n",
@@ -350,11 +545,34 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 10,
    "id": "e93efdfc",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Test Loss: 21.956040\n",
+      "\n",
+      "Test Accuracy of airplane: 67% (670/1000)\n",
+      "Test Accuracy of automobile: 78% (786/1000)\n",
+      "Test Accuracy of  bird: 57% (574/1000)\n",
+      "Test Accuracy of   cat: 46% (462/1000)\n",
+      "Test Accuracy of  deer: 54% (548/1000)\n",
+      "Test Accuracy of   dog: 46% (460/1000)\n",
+      "Test Accuracy of  frog: 73% (734/1000)\n",
+      "Test Accuracy of horse: 61% (615/1000)\n",
+      "Test Accuracy of  ship: 70% (703/1000)\n",
+      "Test Accuracy of truck: 64% (641/1000)\n",
+      "\n",
+      "Test Accuracy (Overall): 61% (6193/10000)\n"
+     ]
+    }
+   ],
    "source": [
+    "\n",
+    "\n",
     "model.load_state_dict(torch.load(\"./model_cifar.pt\"))\n",
     "\n",
     "# track test loss\n",
@@ -436,176 +654,1110 @@
   },
   {
    "cell_type": "markdown",
-   "id": "bc381cf4",
    "metadata": {},
    "source": [
-    "## Exercise 2: Quantization: try to compress the CNN to save space\n",
-    "\n",
-    "Quantization doc is available from https://pytorch.org/docs/stable/quantization.html#torch.quantization.quantize_dynamic\n",
-    "        \n",
-    "The Exercise is to quantize post training the above CNN model. Compare the size reduction and the impact on the classification accuracy \n",
-    "\n",
-    "\n",
-    "The size of the model is simply the size of the file."
+    "---\n",
+    "Here is the new network\n",
+    "---"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
-   "id": "ef623c26",
+   "execution_count": 11,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Net2(\n",
+      "  (conv1): Conv2d(3, 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
+      "  (conv2): Conv2d(16, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
+      "  (conv3): Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))\n",
+      "  (pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
+      "  (fc1): Linear(in_features=1024, out_features=512, bias=True)\n",
+      "  (fc2): Linear(in_features=512, out_features=64, bias=True)\n",
+      "  (fc3): Linear(in_features=64, out_features=10, bias=True)\n",
+      "  (dropout): Dropout(p=0.5, inplace=False)\n",
+      ")\n"
+     ]
+    }
+   ],
    "source": [
-    "import os\n",
+    "# define a new CNN architecture\n",
+    "\n",
+    "class Net2(nn.Module):\n",
+    "    def __init__(self):\n",
+    "        super(Net2, self).__init__()\n",
+    "        # Convolutional Layers\n",
+    "        self.conv1 = nn.Conv2d(3, 16,3, padding=1)\n",
+    "        self.conv2 = nn.Conv2d(16,32, 3,padding=1)\n",
+    "        self.conv3 = nn.Conv2d(32,64,3,padding=1)\n",
     "\n",
+    "        #MaxPool Layer\n",
+    "        self.pool = nn.MaxPool2d(2,2)\n",
     "\n",
-    "def print_size_of_model(model, label=\"\"):\n",
-    "    torch.save(model.state_dict(), \"temp.p\")\n",
-    "    size = os.path.getsize(\"temp.p\")\n",
-    "    print(\"model: \", label, \" \\t\", \"Size (KB):\", size / 1e3)\n",
-    "    os.remove(\"temp.p\")\n",
-    "    return size\n",
+    "        #Fully connected Layer\n",
+    "        self.fc1 = nn.Linear(64 * 4 * 4, 512)\n",
+    "        self.fc2 = nn.Linear(512, 64)\n",
+    "        self.fc3 = nn.Linear(64, 10)\n",
     "\n",
+    "        #DropOut Layer\n",
+    "        self.dropout = nn.Dropout(0.5)\n",
     "\n",
-    "print_size_of_model(model, \"fp32\")"
+    "    def forward(self, x):\n",
+    "        x = self.pool(F.relu(self.conv1(x)))\n",
+    "        x = self.pool(F.relu(self.conv2(x)))\n",
+    "        x = self.pool(F.relu(self.conv3(x)))\n",
+    "\n",
+    "        x = x.view(-1, 64 * 4 * 4)\n",
+    "\n",
+    "        x = F.relu(self.fc1(x))\n",
+    "        x = self.dropout(x)\n",
+    "        x = F.relu(self.fc2(x))\n",
+    "        x = self.dropout(x)\n",
+    "        x = self.fc3(x)\n",
+    "        return x\n",
+    "\n",
+    "\n",
+    "# create a  CNN\n",
+    "model2 = Net2()\n",
+    "print(model2)\n",
+    "# move tensors to GPU if CUDA is available\n",
+    "if train_on_gpu:\n",
+    "    model2.cuda()"
    ]
   },
   {
    "cell_type": "markdown",
-   "id": "05c4e9ad",
    "metadata": {},
    "source": [
-    "Post training quantization example"
+    "---\n",
+    "Here we train the model2:\n",
+    "---"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
-   "id": "c4c65d4b",
+   "execution_count": 12,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch: 0 \tTraining Loss: 45.857789 \tValidation Loss: 44.560349\n",
+      "Validation loss decreased (inf --> 44.560349).  Saving model ...\n",
+      "Epoch: 1 \tTraining Loss: 41.778772 \tValidation Loss: 37.877069\n",
+      "Validation loss decreased (44.560349 --> 37.877069).  Saving model ...\n",
+      "Epoch: 2 \tTraining Loss: 37.083519 \tValidation Loss: 32.782693\n",
+      "Validation loss decreased (37.877069 --> 32.782693).  Saving model ...\n",
+      "Epoch: 3 \tTraining Loss: 33.402934 \tValidation Loss: 30.173893\n",
+      "Validation loss decreased (32.782693 --> 30.173893).  Saving model ...\n",
+      "Epoch: 4 \tTraining Loss: 31.229537 \tValidation Loss: 28.297551\n",
+      "Validation loss decreased (30.173893 --> 28.297551).  Saving model ...\n",
+      "Epoch: 5 \tTraining Loss: 29.413370 \tValidation Loss: 27.164286\n",
+      "Validation loss decreased (28.297551 --> 27.164286).  Saving model ...\n",
+      "Epoch: 6 \tTraining Loss: 27.889698 \tValidation Loss: 25.733314\n",
+      "Validation loss decreased (27.164286 --> 25.733314).  Saving model ...\n",
+      "Epoch: 7 \tTraining Loss: 26.490659 \tValidation Loss: 24.033053\n",
+      "Validation loss decreased (25.733314 --> 24.033053).  Saving model ...\n",
+      "Epoch: 8 \tTraining Loss: 25.185638 \tValidation Loss: 22.838416\n",
+      "Validation loss decreased (24.033053 --> 22.838416).  Saving model ...\n",
+      "Epoch: 9 \tTraining Loss: 23.968391 \tValidation Loss: 21.469535\n",
+      "Validation loss decreased (22.838416 --> 21.469535).  Saving model ...\n",
+      "Epoch: 10 \tTraining Loss: 22.870672 \tValidation Loss: 20.855845\n",
+      "Validation loss decreased (21.469535 --> 20.855845).  Saving model ...\n",
+      "Epoch: 11 \tTraining Loss: 21.748044 \tValidation Loss: 20.084015\n",
+      "Validation loss decreased (20.855845 --> 20.084015).  Saving model ...\n",
+      "Epoch: 12 \tTraining Loss: 20.823984 \tValidation Loss: 19.152056\n",
+      "Validation loss decreased (20.084015 --> 19.152056).  Saving model ...\n",
+      "Epoch: 13 \tTraining Loss: 19.860081 \tValidation Loss: 18.863754\n",
+      "Validation loss decreased (19.152056 --> 18.863754).  Saving model ...\n",
+      "Epoch: 14 \tTraining Loss: 19.011136 \tValidation Loss: 18.424995\n",
+      "Validation loss decreased (18.863754 --> 18.424995).  Saving model ...\n",
+      "Epoch: 15 \tTraining Loss: 18.079763 \tValidation Loss: 17.192690\n",
+      "Validation loss decreased (18.424995 --> 17.192690).  Saving model ...\n",
+      "Epoch: 16 \tTraining Loss: 17.376203 \tValidation Loss: 17.144983\n",
+      "Validation loss decreased (17.192690 --> 17.144983).  Saving model ...\n",
+      "Epoch: 17 \tTraining Loss: 16.608376 \tValidation Loss: 16.288037\n",
+      "Validation loss decreased (17.144983 --> 16.288037).  Saving model ...\n",
+      "Epoch: 18 \tTraining Loss: 15.988603 \tValidation Loss: 16.175762\n",
+      "Validation loss decreased (16.288037 --> 16.175762).  Saving model ...\n",
+      "Epoch: 19 \tTraining Loss: 15.412869 \tValidation Loss: 15.441982\n",
+      "Validation loss decreased (16.175762 --> 15.441982).  Saving model ...\n",
+      "Epoch: 20 \tTraining Loss: 14.778294 \tValidation Loss: 15.939022\n",
+      "Epoch: 21 \tTraining Loss: 14.146509 \tValidation Loss: 15.906133\n",
+      "Epoch: 22 \tTraining Loss: 13.569586 \tValidation Loss: 15.517570\n",
+      "Epoch: 23 \tTraining Loss: 13.081360 \tValidation Loss: 16.112445\n",
+      "Epoch: 24 \tTraining Loss: 12.587562 \tValidation Loss: 15.454825\n",
+      "Epoch: 25 \tTraining Loss: 12.073642 \tValidation Loss: 15.664324\n",
+      "Epoch: 26 \tTraining Loss: 11.623898 \tValidation Loss: 15.579594\n",
+      "Epoch: 27 \tTraining Loss: 11.123427 \tValidation Loss: 15.694448\n",
+      "Epoch: 28 \tTraining Loss: 10.639247 \tValidation Loss: 15.685701\n",
+      "Epoch: 29 \tTraining Loss: 10.170218 \tValidation Loss: 16.205130\n"
+     ]
+    }
+   ],
    "source": [
-    "import torch.quantization\n",
     "\n",
+    "criterion = nn.CrossEntropyLoss()  # specify loss function\n",
+    "optimizer = optim.SGD(model2.parameters(), lr=0.01)  # specify optimizer\n",
     "\n",
-    "quantized_model = torch.quantization.quantize_dynamic(model, dtype=torch.qint8)\n",
-    "print_size_of_model(quantized_model, \"int8\")"
+    "n_epochs = 30  # number of epochs to train the model\n",
+    "train_loss_list = []  # list to store loss to visualize\n",
+    "valid_loss_list = []\n",
+    "valid_loss_min = np.Inf  # track change in validation loss\n",
+    "\n",
+    "for epoch in range(n_epochs):\n",
+    "    # Keep track of training and validation loss\n",
+    "    train_loss = 0.0\n",
+    "    valid_loss = 0.0\n",
+    "\n",
+    "    # Train the model\n",
+    "    model2.train()\n",
+    "    for data, target in train_loader:\n",
+    "        # Move tensors to GPU if CUDA is available\n",
+    "        if train_on_gpu:\n",
+    "            data, target = data.cuda(), target.cuda()\n",
+    "        # 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",
+    "        \n",
+    "        \n",
+    "        # Calculate the batch loss\n",
+    "        loss = criterion(output, target)\n",
+    "        # Backward pass: compute gradient of the loss with respect to model parameters\n",
+    "        loss.backward()\n",
+    "        # Perform a single optimization step (parameter update)\n",
+    "        optimizer.step()\n",
+    "        # Update training loss\n",
+    "        train_loss += loss.item() * data.size(0)\n",
+    "\n",
+    "    # Validate the model\n",
+    "    model2.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",
+    "        # Calculate the batch loss\n",
+    "        loss = criterion(output, target)\n",
+    "        # Update average validation loss\n",
+    "        valid_loss += loss.item() * data.size(0)\n",
+    "\n",
+    "    # Calculate average losses\n",
+    "    train_loss = train_loss / len(train_loader)\n",
+    "    valid_loss = valid_loss / len(valid_loader)\n",
+    "    train_loss_list.append(train_loss)\n",
+    "    valid_loss_list.append(valid_loss)\n",
+    "\n",
+    "    # Print training/validation statistics\n",
+    "    print(\n",
+    "        \"Epoch: {} \\tTraining Loss: {:.6f} \\tValidation Loss: {:.6f}\".format(\n",
+    "            epoch, train_loss, valid_loss\n",
+    "        )\n",
+    "    )\n",
+    "\n",
+    "    # Save model if validation loss has decreased\n",
+    "    if valid_loss <= valid_loss_min:\n",
+    "        print(\n",
+    "            \"Validation loss decreased ({:.6f} --> {:.6f}).  Saving model ...\".format(\n",
+    "                valid_loss_min, valid_loss\n",
+    "            )\n",
+    "        )\n",
+    "        torch.save(model2.state_dict(), \"model_cifar.pt\")\n",
+    "        valid_loss_min = valid_loss"
    ]
   },
   {
-   "cell_type": "markdown",
-   "id": "7b108e17",
+   "cell_type": "code",
+   "execution_count": 13,
    "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
    "source": [
-    "For each class, compare the classification test accuracy of the initial model and the quantized model. Also give the overall test accuracy for both models."
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "plt.plot(range(n_epochs), train_loss_list,label=\"training loss\")\n",
+    "plt.plot(range(n_epochs),valid_loss_list,label=\"validation loss\")\n",
+    "\n",
+    "plt.xlabel(\"Epoch\")\n",
+    "plt.ylabel(\"Loss\")\n",
+    "plt.legend()\n",
+    "plt.title(\"Performance of Model 2\")\n",
+    "plt.show()"
    ]
   },
   {
    "cell_type": "markdown",
-   "id": "a0a34b90",
    "metadata": {},
    "source": [
-    "Try training aware quantization to mitigate the impact on the accuracy (doc available here https://pytorch.org/docs/stable/quantization.html#torch.quantization.quantize_dynamic)"
+    "----\n",
+    "We note that validation voss are less than the first model. lets see the accuracy on the test data now:\n",
+    "\n",
+    "---"
    ]
   },
   {
    "cell_type": "markdown",
-   "id": "201470f9",
    "metadata": {},
    "source": [
-    "## Exercise 3: working with pre-trained models.\n",
-    "\n",
-    "PyTorch offers several pre-trained models https://pytorch.org/vision/0.8/models.html        \n",
-    "We will use ResNet50 trained on ImageNet dataset (https://www.image-net.org/index.php). Use the following code with the files `imagenet-simple-labels.json` that contains the imagenet labels and the image dog.png that we will use as test.\n"
+    "Here we apply the model with the lowest validation loss value:"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
-   "id": "b4d13080",
+   "execution_count": 14,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Test Loss: 16.039122\n",
+      "\n",
+      "Test Accuracy of airplane: 79% (799/1000)\n",
+      "Test Accuracy of automobile: 82% (820/1000)\n",
+      "Test Accuracy of  bird: 58% (584/1000)\n",
+      "Test Accuracy of   cat: 55% (552/1000)\n",
+      "Test Accuracy of  deer: 72% (723/1000)\n",
+      "Test Accuracy of   dog: 55% (551/1000)\n",
+      "Test Accuracy of  frog: 80% (807/1000)\n",
+      "Test Accuracy of horse: 76% (766/1000)\n",
+      "Test Accuracy of  ship: 86% (862/1000)\n",
+      "Test Accuracy of truck: 79% (790/1000)\n",
+      "\n",
+      "Test Accuracy (Overall): 72% (7254/10000)\n"
+     ]
+    }
+   ],
    "source": [
-    "import json\n",
-    "from PIL import Image\n",
-    "\n",
-    "# Choose an image to pass through the model\n",
-    "test_image = \"dog.png\"\n",
-    "\n",
-    "# Configure matplotlib for pretty inline plots\n",
-    "#%matplotlib inline\n",
-    "#%config InlineBackend.figure_format = 'retina'\n",
-    "\n",
-    "# Prepare the labels\n",
-    "with open(\"imagenet-simple-labels.json\") as f:\n",
-    "    labels = json.load(f)\n",
+    "model2.load_state_dict(torch.load(\"./model_cifar.pt\"))\n",
     "\n",
-    "# First prepare the transformations: resize the image to what the model was trained on and convert it to a tensor\n",
-    "data_transform = transforms.Compose(\n",
-    "    [\n",
-    "        transforms.Resize((224, 224)),\n",
-    "        transforms.ToTensor(),\n",
-    "        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]),\n",
-    "    ]\n",
-    ")\n",
-    "# Load the image\n",
+    "# track test loss\n",
+    "test_loss = 0.0\n",
+    "class_correct = list(0.0 for i in range(10))\n",
+    "class_total = list(0.0 for i in range(10))\n",
     "\n",
-    "image = Image.open(test_image)\n",
-    "plt.imshow(image), plt.xticks([]), plt.yticks([])\n",
+    "model2.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",
+    "    # calculate the batch loss\n",
+    "    loss = criterion(output, target)\n",
+    "    # update test loss\n",
+    "    test_loss += loss.item() * data.size(0)\n",
+    "    # convert output probabilities to predicted class\n",
+    "    _, pred = torch.max(output, 1)\n",
+    "    # compare predictions to true label\n",
+    "    correct_tensor = pred.eq(target.data.view_as(pred))\n",
+    "    correct = (\n",
+    "        np.squeeze(correct_tensor.numpy())\n",
+    "        if not train_on_gpu\n",
+    "        else np.squeeze(correct_tensor.cpu().numpy())\n",
+    "    )\n",
+    "    # calculate test accuracy for each object class\n",
+    "    for i in range(batch_size):\n",
+    "        label = target.data[i]\n",
+    "        class_correct[label] += correct[i].item()\n",
+    "        class_total[label] += 1\n",
     "\n",
-    "# Now apply the transformation, expand the batch dimension, and send the image to the GPU\n",
-    "# image = data_transform(image).unsqueeze(0).cuda()\n",
-    "image = data_transform(image).unsqueeze(0)\n",
+    "# average test loss\n",
+    "test_loss = test_loss / len(test_loader)\n",
+    "print(\"Test Loss: {:.6f}\\n\".format(test_loss))\n",
     "\n",
-    "# Download the model if it's not there already. It will take a bit on the first run, after that it's fast\n",
-    "model = models.resnet50(pretrained=True)\n",
-    "# Send the model to the GPU\n",
-    "# model.cuda()\n",
-    "# Set layers such as dropout and batchnorm in evaluation mode\n",
-    "model.eval()\n",
+    "for i in range(10):\n",
+    "    if class_total[i] > 0:\n",
+    "        print(\n",
+    "            \"Test Accuracy of %5s: %2d%% (%2d/%2d)\"\n",
+    "            % (\n",
+    "                classes[i],\n",
+    "                100 * class_correct[i] / class_total[i],\n",
+    "                np.sum(class_correct[i]),\n",
+    "                np.sum(class_total[i]),\n",
+    "            )\n",
+    "        )\n",
+    "    else:\n",
+    "        print(\"Test Accuracy of %5s: N/A (no training examples)\" % (classes[i]))\n",
     "\n",
-    "# Get the 1000-dimensional model output\n",
-    "out = model(image)\n",
-    "# Find the predicted class\n",
-    "print(\"Predicted class is: {}\".format(labels[out.argmax()]))"
+    "print(\n",
+    "    \"\\nTest Accuracy (Overall): %2d%% (%2d/%2d)\"\n",
+    "    % (\n",
+    "        100.0 * np.sum(class_correct) / np.sum(class_total),\n",
+    "        np.sum(class_correct),\n",
+    "        np.sum(class_total),\n",
+    "    )\n",
+    ")"
    ]
   },
   {
    "cell_type": "markdown",
-   "id": "184cfceb",
    "metadata": {},
    "source": [
-    "Experiments:\n",
-    "\n",
-    "Study the code and the results obtained. Possibly add other images downloaded from the internet.\n",
-    "\n",
-    "What is the size of the model? Quantize it and then check if the model is still able to correctly classify the other images.\n",
-    "\n",
-    "Experiment with other pre-trained CNN models.\n",
+    "---\n",
+    " Models comparaison: During the training, it's noticibale that the validation loss values are less than the one in the first model. It gives a first idea that the second model is performing better than the first one. This is confirmed by the test accuracy of model2 which is 73%, higher than the one of model1, 61%. So, the addes layer to the cnn net and the other modifications improved the accuracy of the classification.\n",
     "\n",
-    "    \n"
+    "---"
    ]
   },
   {
    "cell_type": "markdown",
-   "id": "5d57da4b",
+   "id": "bc381cf4",
    "metadata": {},
    "source": [
-    "## Exercise 4: Transfer Learning\n",
-    "    \n",
-    "    \n",
-    "For this work, we will use a pre-trained model (ResNet18) as a descriptor extractor and will refine the classification by training only the last fully connected layer of the network. Thus, the output layer of the pre-trained network will be replaced by a layer adapted to the new classes to be recognized which will be in our case ants and bees.\n",
-    "Download and unzip in your working directory the dataset available at the address :\n",
-    "    \n",
-    "https://download.pytorch.org/tutorial/hymenoptera_data.zip\n",
-    "    \n",
-    "Execute the following code in order to display some images of the dataset."
+    "## Exercise 2: Quantization: try to compress the CNN to save space\n",
+    "\n",
+    "Quantization doc is available from https://pytorch.org/docs/stable/quantization.html#torch.quantization.quantize_dynamic\n",
+    "        \n",
+    "The Exercise is to quantize post training the above CNN model. Compare the size reduction and the impact on the classification accuracy \n",
+    "\n",
+    "\n",
+    "The size of the model is simply the size of the file."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": null,
-   "id": "be2d31f5",
+   "execution_count": 15,
+   "id": "ef623c26",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "model:  fp32  \t Size (KB): 251.342\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "251342"
+      ]
+     },
+     "execution_count": 15,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import os\n",
+    "\n",
+    "\n",
+    "def print_size_of_model(model, label=\"\"):\n",
+    "    torch.save(model.state_dict(), \"temp.p\")\n",
+    "    size = os.path.getsize(\"temp.p\")\n",
+    "    print(\"model: \", label, \" \\t\", \"Size (KB):\", size / 1e3)\n",
+    "    os.remove(\"temp.p\")\n",
+    "    return size\n",
+    "\n",
+    "\n",
+    "print_size_of_model(model, \"fp32\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "05c4e9ad",
+   "metadata": {},
+   "source": [
+    "Post training quantization example"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "id": "c4c65d4b",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "model:  int8  \t Size (KB): 76.65\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "76650"
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import torch.quantization\n",
+    "\n",
+    "\n",
+    "quantized_model = torch.quantization.quantize_dynamic(model, dtype=torch.qint8)\n",
+    "print_size_of_model(quantized_model, \"int8\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "7b108e17",
+   "metadata": {},
+   "source": [
+    "For each class, compare the classification test accuracy of the initial model and the quantized model. Also give the overall test accuracy for both models."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [
+    {
+     "ename": "NotImplementedError",
+     "evalue": "Could not run 'quantized::linear_dynamic' with arguments from the 'CUDA' backend. This could be because the operator doesn't exist for this backend, or was omitted during the selective/custom build process (if using custom build). If you are a Facebook employee using PyTorch on mobile, please visit https://fburl.com/ptmfixes for possible resolutions. 'quantized::linear_dynamic' is only available for these backends: [CPU, BackendSelect, Python, FuncTorchDynamicLayerBackMode, Functionalize, Named, Conjugate, Negative, ZeroTensor, ADInplaceOrView, AutogradOther, AutogradCPU, AutogradCUDA, AutogradXLA, AutogradMPS, AutogradXPU, AutogradHPU, AutogradLazy, AutogradMeta, Tracer, AutocastCPU, AutocastCUDA, FuncTorchBatched, FuncTorchVmapMode, Batched, VmapMode, FuncTorchGradWrapper, PythonTLSSnapshot, FuncTorchDynamicLayerFrontMode, PreDispatch, PythonDispatcher].\n\nCPU: registered at ..\\aten\\src\\ATen\\native\\quantized\\cpu\\qlinear_dynamic.cpp:662 [kernel]\nBackendSelect: fallthrough registered at ..\\aten\\src\\ATen\\core\\BackendSelectFallbackKernel.cpp:3 [backend fallback]\nPython: registered at ..\\aten\\src\\ATen\\core\\PythonFallbackKernel.cpp:153 [backend fallback]\nFuncTorchDynamicLayerBackMode: registered at ..\\aten\\src\\ATen\\functorch\\DynamicLayer.cpp:498 [backend fallback]\nFunctionalize: registered at ..\\aten\\src\\ATen\\FunctionalizeFallbackKernel.cpp:290 [backend fallback]\nNamed: registered at ..\\aten\\src\\ATen\\core\\NamedRegistrations.cpp:7 [backend fallback]\nConjugate: registered at ..\\aten\\src\\ATen\\ConjugateFallback.cpp:17 [backend fallback]\nNegative: registered at ..\\aten\\src\\ATen\\native\\NegateFallback.cpp:19 [backend fallback]\nZeroTensor: registered at ..\\aten\\src\\ATen\\ZeroTensorFallback.cpp:86 [backend fallback]\nADInplaceOrView: fallthrough registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:86 [backend fallback]\nAutogradOther: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:53 [backend fallback]\nAutogradCPU: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:57 [backend fallback]\nAutogradCUDA: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:65 [backend fallback]\nAutogradXLA: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:69 [backend fallback]\nAutogradMPS: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:77 [backend fallback]\nAutogradXPU: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:61 [backend fallback]\nAutogradHPU: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:90 [backend fallback]\nAutogradLazy: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:73 [backend fallback]\nAutogradMeta: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:81 [backend fallback]\nTracer: registered at ..\\torch\\csrc\\autograd\\TraceTypeManual.cpp:296 [backend fallback]\nAutocastCPU: fallthrough registered at ..\\aten\\src\\ATen\\autocast_mode.cpp:382 [backend fallback]\nAutocastCUDA: fallthrough registered at ..\\aten\\src\\ATen\\autocast_mode.cpp:249 [backend fallback]\nFuncTorchBatched: registered at ..\\aten\\src\\ATen\\functorch\\LegacyBatchingRegistrations.cpp:710 [backend fallback]\nFuncTorchVmapMode: fallthrough registered at ..\\aten\\src\\ATen\\functorch\\VmapModeRegistrations.cpp:28 [backend fallback]\nBatched: registered at ..\\aten\\src\\ATen\\LegacyBatchingRegistrations.cpp:1075 [backend fallback]\nVmapMode: fallthrough registered at ..\\aten\\src\\ATen\\VmapModeRegistrations.cpp:33 [backend fallback]\nFuncTorchGradWrapper: registered at ..\\aten\\src\\ATen\\functorch\\TensorWrapper.cpp:203 [backend fallback]\nPythonTLSSnapshot: registered at ..\\aten\\src\\ATen\\core\\PythonFallbackKernel.cpp:161 [backend fallback]\nFuncTorchDynamicLayerFrontMode: registered at ..\\aten\\src\\ATen\\functorch\\DynamicLayer.cpp:494 [backend fallback]\nPreDispatch: registered at ..\\aten\\src\\ATen\\core\\PythonFallbackKernel.cpp:165 [backend fallback]\nPythonDispatcher: registered at ..\\aten\\src\\ATen\\core\\PythonFallbackKernel.cpp:157 [backend fallback]\n",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[1;31mNotImplementedError\u001b[0m                       Traceback (most recent call last)",
+      "\u001b[1;32mc:\\Users\\LENOVO\\Desktop\\deeplearning\\td-2-deep-learning\\TD2 Deep Learning.ipynb Cell 36\u001b[0m line \u001b[0;36m1\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X50sZmlsZQ%3D%3D?line=10'>11</a>\u001b[0m     data, target \u001b[39m=\u001b[39m data\u001b[39m.\u001b[39mcuda(), target\u001b[39m.\u001b[39mcuda()\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X50sZmlsZQ%3D%3D?line=11'>12</a>\u001b[0m \u001b[39m# forward pass: compute predicted outputs by passing inputs to the model\u001b[39;00m\n\u001b[1;32m---> <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X50sZmlsZQ%3D%3D?line=12'>13</a>\u001b[0m output \u001b[39m=\u001b[39m quantized_model(data)\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X50sZmlsZQ%3D%3D?line=13'>14</a>\u001b[0m \u001b[39m# calculate the batch loss\u001b[39;00m\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X50sZmlsZQ%3D%3D?line=14'>15</a>\u001b[0m loss \u001b[39m=\u001b[39m criterion(output, target)\n",
+      "File \u001b[1;32mc:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torch\\nn\\modules\\module.py:1518\u001b[0m, in \u001b[0;36mModule._wrapped_call_impl\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1516\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_compiled_call_impl(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)  \u001b[39m# type: ignore[misc]\u001b[39;00m\n\u001b[0;32m   1517\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[1;32m-> 1518\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_call_impl(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)\n",
+      "File \u001b[1;32mc:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torch\\nn\\modules\\module.py:1527\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1522\u001b[0m \u001b[39m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[0;32m   1523\u001b[0m \u001b[39m# this function, and just call forward.\u001b[39;00m\n\u001b[0;32m   1524\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m (\u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_pre_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_pre_hooks\n\u001b[0;32m   1525\u001b[0m         \u001b[39mor\u001b[39;00m _global_backward_pre_hooks \u001b[39mor\u001b[39;00m _global_backward_hooks\n\u001b[0;32m   1526\u001b[0m         \u001b[39mor\u001b[39;00m _global_forward_hooks \u001b[39mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[1;32m-> 1527\u001b[0m     \u001b[39mreturn\u001b[39;00m forward_call(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)\n\u001b[0;32m   1529\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m   1530\u001b[0m     result \u001b[39m=\u001b[39m \u001b[39mNone\u001b[39;00m\n",
+      "\u001b[1;32mc:\\Users\\LENOVO\\Desktop\\deeplearning\\td-2-deep-learning\\TD2 Deep Learning.ipynb Cell 36\u001b[0m line \u001b[0;36m2\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X50sZmlsZQ%3D%3D?line=18'>19</a>\u001b[0m x \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mpool(F\u001b[39m.\u001b[39mrelu(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mconv2(x)))\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X50sZmlsZQ%3D%3D?line=19'>20</a>\u001b[0m x \u001b[39m=\u001b[39m x\u001b[39m.\u001b[39mview(\u001b[39m-\u001b[39m\u001b[39m1\u001b[39m, \u001b[39m16\u001b[39m \u001b[39m*\u001b[39m \u001b[39m5\u001b[39m \u001b[39m*\u001b[39m \u001b[39m5\u001b[39m)\n\u001b[1;32m---> <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X50sZmlsZQ%3D%3D?line=20'>21</a>\u001b[0m x \u001b[39m=\u001b[39m F\u001b[39m.\u001b[39mrelu(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mfc1(x))\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X50sZmlsZQ%3D%3D?line=21'>22</a>\u001b[0m x \u001b[39m=\u001b[39m F\u001b[39m.\u001b[39mrelu(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mfc2(x))\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X50sZmlsZQ%3D%3D?line=22'>23</a>\u001b[0m x \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mfc3(x)\n",
+      "File \u001b[1;32mc:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torch\\nn\\modules\\module.py:1518\u001b[0m, in \u001b[0;36mModule._wrapped_call_impl\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1516\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_compiled_call_impl(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)  \u001b[39m# type: ignore[misc]\u001b[39;00m\n\u001b[0;32m   1517\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[1;32m-> 1518\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_call_impl(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)\n",
+      "File \u001b[1;32mc:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torch\\nn\\modules\\module.py:1527\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1522\u001b[0m \u001b[39m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[0;32m   1523\u001b[0m \u001b[39m# this function, and just call forward.\u001b[39;00m\n\u001b[0;32m   1524\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m (\u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_pre_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_pre_hooks\n\u001b[0;32m   1525\u001b[0m         \u001b[39mor\u001b[39;00m _global_backward_pre_hooks \u001b[39mor\u001b[39;00m _global_backward_hooks\n\u001b[0;32m   1526\u001b[0m         \u001b[39mor\u001b[39;00m _global_forward_hooks \u001b[39mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[1;32m-> 1527\u001b[0m     \u001b[39mreturn\u001b[39;00m forward_call(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)\n\u001b[0;32m   1529\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m   1530\u001b[0m     result \u001b[39m=\u001b[39m \u001b[39mNone\u001b[39;00m\n",
+      "File \u001b[1;32mc:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torch\\ao\\nn\\quantized\\dynamic\\modules\\linear.py:54\u001b[0m, in \u001b[0;36mLinear.forward\u001b[1;34m(self, x)\u001b[0m\n\u001b[0;32m     51\u001b[0m         Y \u001b[39m=\u001b[39m torch\u001b[39m.\u001b[39mops\u001b[39m.\u001b[39mquantized\u001b[39m.\u001b[39mlinear_dynamic(\n\u001b[0;32m     52\u001b[0m             x, \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_packed_params\u001b[39m.\u001b[39m_packed_params)\n\u001b[0;32m     53\u001b[0m     \u001b[39melse\u001b[39;00m:\n\u001b[1;32m---> 54\u001b[0m         Y \u001b[39m=\u001b[39m torch\u001b[39m.\u001b[39;49mops\u001b[39m.\u001b[39;49mquantized\u001b[39m.\u001b[39;49mlinear_dynamic(\n\u001b[0;32m     55\u001b[0m             x, \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_packed_params\u001b[39m.\u001b[39;49m_packed_params, reduce_range\u001b[39m=\u001b[39;49m\u001b[39mTrue\u001b[39;49;00m)\n\u001b[0;32m     56\u001b[0m \u001b[39melif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_packed_params\u001b[39m.\u001b[39mdtype \u001b[39m==\u001b[39m torch\u001b[39m.\u001b[39mfloat16:\n\u001b[0;32m     57\u001b[0m     Y \u001b[39m=\u001b[39m torch\u001b[39m.\u001b[39mops\u001b[39m.\u001b[39mquantized\u001b[39m.\u001b[39mlinear_dynamic_fp16(\n\u001b[0;32m     58\u001b[0m         x, \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_packed_params\u001b[39m.\u001b[39m_packed_params)\n",
+      "File \u001b[1;32mc:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torch\\_ops.py:692\u001b[0m, in \u001b[0;36mOpOverloadPacket.__call__\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m    687\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m__call__\u001b[39m(\u001b[39mself\u001b[39m, \u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs):\n\u001b[0;32m    688\u001b[0m     \u001b[39m# overloading __call__ to ensure torch.ops.foo.bar()\u001b[39;00m\n\u001b[0;32m    689\u001b[0m     \u001b[39m# is still callable from JIT\u001b[39;00m\n\u001b[0;32m    690\u001b[0m     \u001b[39m# We save the function ptr as the `op` attribute on\u001b[39;00m\n\u001b[0;32m    691\u001b[0m     \u001b[39m# OpOverloadPacket to access it here.\u001b[39;00m\n\u001b[1;32m--> 692\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_op(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs \u001b[39mor\u001b[39;00m {})\n",
+      "\u001b[1;31mNotImplementedError\u001b[0m: Could not run 'quantized::linear_dynamic' with arguments from the 'CUDA' backend. This could be because the operator doesn't exist for this backend, or was omitted during the selective/custom build process (if using custom build). If you are a Facebook employee using PyTorch on mobile, please visit https://fburl.com/ptmfixes for possible resolutions. 'quantized::linear_dynamic' is only available for these backends: [CPU, BackendSelect, Python, FuncTorchDynamicLayerBackMode, Functionalize, Named, Conjugate, Negative, ZeroTensor, ADInplaceOrView, AutogradOther, AutogradCPU, AutogradCUDA, AutogradXLA, AutogradMPS, AutogradXPU, AutogradHPU, AutogradLazy, AutogradMeta, Tracer, AutocastCPU, AutocastCUDA, FuncTorchBatched, FuncTorchVmapMode, Batched, VmapMode, FuncTorchGradWrapper, PythonTLSSnapshot, FuncTorchDynamicLayerFrontMode, PreDispatch, PythonDispatcher].\n\nCPU: registered at ..\\aten\\src\\ATen\\native\\quantized\\cpu\\qlinear_dynamic.cpp:662 [kernel]\nBackendSelect: fallthrough registered at ..\\aten\\src\\ATen\\core\\BackendSelectFallbackKernel.cpp:3 [backend fallback]\nPython: registered at ..\\aten\\src\\ATen\\core\\PythonFallbackKernel.cpp:153 [backend fallback]\nFuncTorchDynamicLayerBackMode: registered at ..\\aten\\src\\ATen\\functorch\\DynamicLayer.cpp:498 [backend fallback]\nFunctionalize: registered at ..\\aten\\src\\ATen\\FunctionalizeFallbackKernel.cpp:290 [backend fallback]\nNamed: registered at ..\\aten\\src\\ATen\\core\\NamedRegistrations.cpp:7 [backend fallback]\nConjugate: registered at ..\\aten\\src\\ATen\\ConjugateFallback.cpp:17 [backend fallback]\nNegative: registered at ..\\aten\\src\\ATen\\native\\NegateFallback.cpp:19 [backend fallback]\nZeroTensor: registered at ..\\aten\\src\\ATen\\ZeroTensorFallback.cpp:86 [backend fallback]\nADInplaceOrView: fallthrough registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:86 [backend fallback]\nAutogradOther: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:53 [backend fallback]\nAutogradCPU: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:57 [backend fallback]\nAutogradCUDA: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:65 [backend fallback]\nAutogradXLA: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:69 [backend fallback]\nAutogradMPS: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:77 [backend fallback]\nAutogradXPU: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:61 [backend fallback]\nAutogradHPU: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:90 [backend fallback]\nAutogradLazy: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:73 [backend fallback]\nAutogradMeta: registered at ..\\aten\\src\\ATen\\core\\VariableFallbackKernel.cpp:81 [backend fallback]\nTracer: registered at ..\\torch\\csrc\\autograd\\TraceTypeManual.cpp:296 [backend fallback]\nAutocastCPU: fallthrough registered at ..\\aten\\src\\ATen\\autocast_mode.cpp:382 [backend fallback]\nAutocastCUDA: fallthrough registered at ..\\aten\\src\\ATen\\autocast_mode.cpp:249 [backend fallback]\nFuncTorchBatched: registered at ..\\aten\\src\\ATen\\functorch\\LegacyBatchingRegistrations.cpp:710 [backend fallback]\nFuncTorchVmapMode: fallthrough registered at ..\\aten\\src\\ATen\\functorch\\VmapModeRegistrations.cpp:28 [backend fallback]\nBatched: registered at ..\\aten\\src\\ATen\\LegacyBatchingRegistrations.cpp:1075 [backend fallback]\nVmapMode: fallthrough registered at ..\\aten\\src\\ATen\\VmapModeRegistrations.cpp:33 [backend fallback]\nFuncTorchGradWrapper: registered at ..\\aten\\src\\ATen\\functorch\\TensorWrapper.cpp:203 [backend fallback]\nPythonTLSSnapshot: registered at ..\\aten\\src\\ATen\\core\\PythonFallbackKernel.cpp:161 [backend fallback]\nFuncTorchDynamicLayerFrontMode: registered at ..\\aten\\src\\ATen\\functorch\\DynamicLayer.cpp:494 [backend fallback]\nPreDispatch: registered at ..\\aten\\src\\ATen\\core\\PythonFallbackKernel.cpp:165 [backend fallback]\nPythonDispatcher: registered at ..\\aten\\src\\ATen\\core\\PythonFallbackKernel.cpp:157 [backend fallback]\n"
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "# track test loss\n",
+    "test_loss = 0.0\n",
+    "class_correct = list(0.0 for i in range(10))\n",
+    "class_total = list(0.0 for i in range(10))\n",
+    "\n",
+    "quantized_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 = quantized_model(data)\n",
+    "    # calculate the batch loss\n",
+    "    loss = criterion(output, target)\n",
+    "    # update test loss\n",
+    "    test_loss += loss.item() * data.size(0)\n",
+    "    # convert output probabilities to predicted class\n",
+    "    _, pred = torch.max(output, 1)\n",
+    "    # compare predictions to true label\n",
+    "    correct_tensor = pred.eq(target.data.view_as(pred))\n",
+    "    correct = (\n",
+    "        np.squeeze(correct_tensor.numpy())\n",
+    "        if not train_on_gpu\n",
+    "        else np.squeeze(correct_tensor.cpu().numpy())\n",
+    "    )\n",
+    "    # calculate test accuracy for each object class\n",
+    "    for i in range(batch_size):\n",
+    "        label = target.data[i]\n",
+    "        class_correct[label] += correct[i].item()\n",
+    "        class_total[label] += 1\n",
+    "\n",
+    "# average test loss\n",
+    "test_loss = test_loss / len(test_loader)\n",
+    "print(\"Test Loss: {:.6f}\\n\".format(test_loss))\n",
+    "\n",
+    "for i in range(10):\n",
+    "    if class_total[i] > 0:\n",
+    "        print(\n",
+    "            \"Test Accuracy of %5s: %2d%% (%2d/%2d)\"\n",
+    "            % (\n",
+    "                classes[i],\n",
+    "                100 * class_correct[i] / class_total[i],\n",
+    "                np.sum(class_correct[i]),\n",
+    "                np.sum(class_total[i]),\n",
+    "            )\n",
+    "        )\n",
+    "    else:\n",
+    "        print(\"Test Accuracy of %5s: N/A (no training examples)\" % (classes[i]))\n",
+    "\n",
+    "print(\n",
+    "    \"\\nTest Accuracy (Overall): %2d%% (%2d/%2d)\"\n",
+    "    % (\n",
+    "        100.0 * np.sum(class_correct) / np.sum(class_total),\n",
+    "        np.sum(class_correct),\n",
+    "        np.sum(class_total),\n",
+    "    )\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "---\n",
+    "comments here\n",
+    "\n",
+    "---"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "a0a34b90",
+   "metadata": {},
+   "source": [
+    "Try training aware quantization to mitigate the impact on the accuracy (doc available here https://pytorch.org/docs/stable/quantization.html#torch.quantization.quantize_dynamic)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "class NetAwareQuantized(nn.Module):\n",
+    "\n",
+    "    def __init__(self):\n",
+    "\n",
+    "        super(NetAwareQuantized, self).__init__()\n",
+    "        #Same architecture as the previous model \n",
+    "        # Convolutional Layers\n",
+    "        self.conv1 = nn.Conv2d(3, 16,3, padding=1)\n",
+    "        self.conv2 = nn.Conv2d(16,32, 3,padding=1)\n",
+    "        self.conv3 = nn.Conv2d(32,64,3,padding=1)\n",
+    "\n",
+    "        #MaxPool Layer\n",
+    "        self.pool = nn.MaxPool2d(2,2)\n",
+    "\n",
+    "        #Fully connected Layer\n",
+    "        self.fc1 = nn.Linear(64 * 4 * 4, 512)\n",
+    "        self.fc2 = nn.Linear(512, 64)\n",
+    "        self.fc3 = nn.Linear(64, 10)\n",
+    "\n",
+    "        #DropOut Layer\n",
+    "        self.dropout = nn.Dropout(0.5)\n",
+    "\n",
+    "\n",
+    "        #adding the quant and dequant \n",
+    "        #QuantStub converts tensors from floating point to quantized\n",
+    "        self.quant = torch.quantization.QuantStub()\n",
+    "\n",
+    "        #DeQuantStub converts tensors from quantized to floating point\n",
+    "        self.dequant = torch.quantization.DeQuantStub()\n",
+    "\n",
+    "    def forward(self, x):\n",
+    "\n",
+    "        x = self.quant(x)\n",
+    "\n",
+    "        x = self.pool(F.relu(self.conv1(x)))\n",
+    "        x = self.pool(F.relu(self.conv2(x)))\n",
+    "        x = self.pool(F.relu(self.conv3(x)))\n",
+    "        x = x.view(-1, 64 * 4 * 4)\n",
+    "        x = F.relu(self.fc1(x))\n",
+    "        x = self.dropout(x) \n",
+    "        x = F.relu(self.fc2(x))\n",
+    "        x = self.dropout(x)\n",
+    "        x = self.fc3(x)\n",
+    "        x = self.dequant(x)\n",
+    "\n",
+    "        return x"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "c:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torch\\ao\\quantization\\quantize.py:309: UserWarning: None of the submodule got qconfig applied. Make sure you passed correct configuration through `qconfig_dict` or by assigning the `.qconfig` attribute directly on submodules\n",
+      "  warnings.warn(\"None of the submodule got qconfig applied. Make sure you \"\n"
+     ]
+    },
+    {
+     "ename": "RuntimeError",
+     "evalue": "Input type (torch.cuda.FloatTensor) and weight type (torch.FloatTensor) should be the same",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[1;31mRuntimeError\u001b[0m                              Traceback (most recent call last)",
+      "\u001b[1;32mc:\\Users\\LENOVO\\Desktop\\deeplearning\\td-2-deep-learning\\TD2 Deep Learning.ipynb Cell 40\u001b[0m line \u001b[0;36m3\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X54sZmlsZQ%3D%3D?line=28'>29</a>\u001b[0m optimizer\u001b[39m.\u001b[39mzero_grad()\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X54sZmlsZQ%3D%3D?line=29'>30</a>\u001b[0m \u001b[39m# Forward pass: compute predicted outputs by passing inputs to the model\u001b[39;00m\n\u001b[1;32m---> <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X54sZmlsZQ%3D%3D?line=30'>31</a>\u001b[0m output \u001b[39m=\u001b[39m model_qat(data)\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X54sZmlsZQ%3D%3D?line=31'>32</a>\u001b[0m \u001b[39m# Calculate the batch loss\u001b[39;00m\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X54sZmlsZQ%3D%3D?line=32'>33</a>\u001b[0m loss \u001b[39m=\u001b[39m criterion(output, target)\n",
+      "File \u001b[1;32mc:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torch\\nn\\modules\\module.py:1518\u001b[0m, in \u001b[0;36mModule._wrapped_call_impl\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1516\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_compiled_call_impl(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)  \u001b[39m# type: ignore[misc]\u001b[39;00m\n\u001b[0;32m   1517\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[1;32m-> 1518\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_call_impl(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)\n",
+      "File \u001b[1;32mc:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torch\\nn\\modules\\module.py:1527\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1522\u001b[0m \u001b[39m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[0;32m   1523\u001b[0m \u001b[39m# this function, and just call forward.\u001b[39;00m\n\u001b[0;32m   1524\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m (\u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_pre_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_pre_hooks\n\u001b[0;32m   1525\u001b[0m         \u001b[39mor\u001b[39;00m _global_backward_pre_hooks \u001b[39mor\u001b[39;00m _global_backward_hooks\n\u001b[0;32m   1526\u001b[0m         \u001b[39mor\u001b[39;00m _global_forward_hooks \u001b[39mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[1;32m-> 1527\u001b[0m     \u001b[39mreturn\u001b[39;00m forward_call(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)\n\u001b[0;32m   1529\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m   1530\u001b[0m     result \u001b[39m=\u001b[39m \u001b[39mNone\u001b[39;00m\n",
+      "\u001b[1;32mc:\\Users\\LENOVO\\Desktop\\deeplearning\\td-2-deep-learning\\TD2 Deep Learning.ipynb Cell 40\u001b[0m line \u001b[0;36m3\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X54sZmlsZQ%3D%3D?line=30'>31</a>\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mforward\u001b[39m(\u001b[39mself\u001b[39m, x):\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X54sZmlsZQ%3D%3D?line=32'>33</a>\u001b[0m     x \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mquant(x)\n\u001b[1;32m---> <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X54sZmlsZQ%3D%3D?line=34'>35</a>\u001b[0m     x \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mpool(F\u001b[39m.\u001b[39mrelu(\u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mconv1(x)))\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X54sZmlsZQ%3D%3D?line=35'>36</a>\u001b[0m     x \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mpool(F\u001b[39m.\u001b[39mrelu(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mconv2(x)))\n\u001b[0;32m     <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X54sZmlsZQ%3D%3D?line=36'>37</a>\u001b[0m     x \u001b[39m=\u001b[39m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mpool(F\u001b[39m.\u001b[39mrelu(\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mconv3(x)))\n",
+      "File \u001b[1;32mc:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torch\\nn\\modules\\module.py:1518\u001b[0m, in \u001b[0;36mModule._wrapped_call_impl\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1516\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_compiled_call_impl(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)  \u001b[39m# type: ignore[misc]\u001b[39;00m\n\u001b[0;32m   1517\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[1;32m-> 1518\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_call_impl(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)\n",
+      "File \u001b[1;32mc:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torch\\nn\\modules\\module.py:1527\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1522\u001b[0m \u001b[39m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[0;32m   1523\u001b[0m \u001b[39m# this function, and just call forward.\u001b[39;00m\n\u001b[0;32m   1524\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mnot\u001b[39;00m (\u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_backward_pre_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_hooks \u001b[39mor\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_forward_pre_hooks\n\u001b[0;32m   1525\u001b[0m         \u001b[39mor\u001b[39;00m _global_backward_pre_hooks \u001b[39mor\u001b[39;00m _global_backward_hooks\n\u001b[0;32m   1526\u001b[0m         \u001b[39mor\u001b[39;00m _global_forward_hooks \u001b[39mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[1;32m-> 1527\u001b[0m     \u001b[39mreturn\u001b[39;00m forward_call(\u001b[39m*\u001b[39margs, \u001b[39m*\u001b[39m\u001b[39m*\u001b[39mkwargs)\n\u001b[0;32m   1529\u001b[0m \u001b[39mtry\u001b[39;00m:\n\u001b[0;32m   1530\u001b[0m     result \u001b[39m=\u001b[39m \u001b[39mNone\u001b[39;00m\n",
+      "File \u001b[1;32mc:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torch\\nn\\modules\\conv.py:460\u001b[0m, in \u001b[0;36mConv2d.forward\u001b[1;34m(self, input)\u001b[0m\n\u001b[0;32m    459\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39mforward\u001b[39m(\u001b[39mself\u001b[39m, \u001b[39minput\u001b[39m: Tensor) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m Tensor:\n\u001b[1;32m--> 460\u001b[0m     \u001b[39mreturn\u001b[39;00m \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49m_conv_forward(\u001b[39minput\u001b[39;49m, \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mweight, \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mbias)\n",
+      "File \u001b[1;32mc:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torch\\nn\\modules\\conv.py:456\u001b[0m, in \u001b[0;36mConv2d._conv_forward\u001b[1;34m(self, input, weight, bias)\u001b[0m\n\u001b[0;32m    452\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mpadding_mode \u001b[39m!=\u001b[39m \u001b[39m'\u001b[39m\u001b[39mzeros\u001b[39m\u001b[39m'\u001b[39m:\n\u001b[0;32m    453\u001b[0m     \u001b[39mreturn\u001b[39;00m F\u001b[39m.\u001b[39mconv2d(F\u001b[39m.\u001b[39mpad(\u001b[39minput\u001b[39m, \u001b[39mself\u001b[39m\u001b[39m.\u001b[39m_reversed_padding_repeated_twice, mode\u001b[39m=\u001b[39m\u001b[39mself\u001b[39m\u001b[39m.\u001b[39mpadding_mode),\n\u001b[0;32m    454\u001b[0m                     weight, bias, \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mstride,\n\u001b[0;32m    455\u001b[0m                     _pair(\u001b[39m0\u001b[39m), \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mdilation, \u001b[39mself\u001b[39m\u001b[39m.\u001b[39mgroups)\n\u001b[1;32m--> 456\u001b[0m \u001b[39mreturn\u001b[39;00m F\u001b[39m.\u001b[39;49mconv2d(\u001b[39minput\u001b[39;49m, weight, bias, \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mstride,\n\u001b[0;32m    457\u001b[0m                 \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mpadding, \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mdilation, \u001b[39mself\u001b[39;49m\u001b[39m.\u001b[39;49mgroups)\n",
+      "\u001b[1;31mRuntimeError\u001b[0m: Input type (torch.cuda.FloatTensor) and weight type (torch.FloatTensor) should be the same"
+     ]
+    }
+   ],
+   "source": [
+    "import copy\n",
+    "import torch.quantization.quantize_fx as quantize_fx\n",
+    "Aware_model=NetAwareQuantized()\n",
+    "\n",
+    "Aware_model.train()\n",
+    "model_to_quantize = copy.deepcopy(Aware_model)\n",
+    "model.qconfig = torch.quantization.get_default_qat_qconfig(\"qnnpack\")\n",
+    "model_qat = torch.quantization.prepare_qat(Aware_model, inplace=False)\n",
+    "\n",
+    "# quantization aware training\n",
+    "model_qat = torch.quantization.convert(model_qat.eval(), inplace=False)\n",
+    "n_epochs=30\n",
+    "train_loss_list = []  # list to store loss to visualize\n",
+    "valid_loss_list=[]\n",
+    "criterion = nn.CrossEntropyLoss()  # specify loss function\n",
+    "optimizer = optim.SGD(model_qat.parameters(), lr=0.01)  # specify optimizer\n",
+    "for epoch in range(n_epochs):\n",
+    "    # Keep track of training and validation loss\n",
+    "    train_loss = 0.0\n",
+    "    valid_loss = 0.0\n",
+    "\n",
+    "    # Train the model\n",
+    "    model_qat.train()\n",
+    "    for data, target in train_loader:\n",
+    "        # Move tensors to GPU if CUDA is available\n",
+    "        if train_on_gpu:\n",
+    "            data, target = data.cuda(), target.cuda()\n",
+    "        # 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 = model_qat(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",
+    "        loss.backward()\n",
+    "        # Perform a single optimization step (parameter update)\n",
+    "        optimizer.step()\n",
+    "        # Update training loss\n",
+    "        train_loss += loss.item() * data.size(0)\n",
+    "\n",
+    "    # Validate the model\n",
+    "    model_qat.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 = model_qat(data)\n",
+    "        # Calculate the batch loss\n",
+    "        loss = criterion(output, target)\n",
+    "        # Update average validation loss\n",
+    "        valid_loss += loss.item() * data.size(0)\n",
+    "\n",
+    "    # Calculate average losses\n",
+    "    train_loss = train_loss / len(train_loader)\n",
+    "    valid_loss = valid_loss / len(valid_loader)\n",
+    "    train_loss_list.append(train_loss)\n",
+    "    valid_loss_list.append(valid_loss)\n",
+    "\n",
+    "    # Print training/validation statistics\n",
+    "    print(\n",
+    "        \"Epoch: {} \\tTraining Loss: {:.6f} \\tValidation Loss: {:.6f}\".format(\n",
+    "            epoch, train_loss, valid_loss\n",
+    "        )\n",
+    "    )"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "plt.plot(range(n_epochs), train_loss_list,label=\"training loss\")\n",
+    "plt.plot(range(n_epochs),valid_loss_list,label=\"validation loss\")\n",
+    "\n",
+    "plt.xlabel(\"Epoch\")\n",
+    "plt.ylabel(\"Loss\")\n",
+    "\n",
+    "plt.legend()\n",
+    "plt.title(\"Performance of the aware quantization Model \")\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# track test loss\n",
+    "test_loss = 0.0\n",
+    "class_correct = list(0.0 for i in range(10))\n",
+    "class_total = list(0.0 for i in range(10))\n",
+    "\n",
+    "Aware_model.eval()\n",
+    "# iterate over test data\n",
+    "for data, target in test_loader:\n",
+    "    \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",
+    "\n",
+    "    output = Aware_model(data)\n",
+    "    # calculate the batch loss\n",
+    "    loss = criterion(output, target)\n",
+    "    # update test loss\n",
+    "    test_loss += loss.item() * data.size(0)\n",
+    "    # convert output probabilities to predicted class\n",
+    "    _, pred = torch.max(output, 1)\n",
+    "    # compare predictions to true label\n",
+    "    correct_tensor = pred.eq(target.data.view_as(pred))\n",
+    "    correct = (\n",
+    "        np.squeeze(correct_tensor.numpy())\n",
+    "        if not train_on_gpu\n",
+    "        else np.squeeze(correct_tensor.cpu().numpy())\n",
+    "    )\n",
+    "    # calculate test accuracy for each object class\n",
+    "    for i in range(batch_size):\n",
+    "        label = target.data[i]\n",
+    "        class_correct[label] += correct[i].item()\n",
+    "        class_total[label] += 1\n",
+    "\n",
+    "# average test loss\n",
+    "test_loss = test_loss / len(test_loader)\n",
+    "print(\"Test Loss: {:.6f}\\n\".format(test_loss))\n",
+    "\n",
+    "for i in range(10):\n",
+    "    if class_total[i] > 0:\n",
+    "        print(\n",
+    "            \"Test Accuracy of %5s: %2d%% (%2d/%2d)\"\n",
+    "            % (\n",
+    "                classes[i],\n",
+    "                100 * class_correct[i] / class_total[i],\n",
+    "                np.sum(class_correct[i]),\n",
+    "                np.sum(class_total[i]),\n",
+    "            )\n",
+    "        )\n",
+    "    else:\n",
+    "        print(\"Test Accuracy of %5s: N/A (no training examples)\" % (classes[i]))\n",
+    "\n",
+    "print(\n",
+    "    \"\\nTest Accuracy (Overall): %2d%% (%2d/%2d)\"\n",
+    "    % (\n",
+    "        100.0 * np.sum(class_correct) / np.sum(class_total),\n",
+    "        np.sum(class_correct),\n",
+    "        np.sum(class_total),\n",
+    "    )\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "201470f9",
+   "metadata": {},
+   "source": [
+    "## Exercise 3: working with pre-trained models.\n",
+    "\n",
+    "PyTorch offers several pre-trained models https://pytorch.org/vision/0.8/models.html        \n",
+    "We will use ResNet50 trained on ImageNet dataset (https://www.image-net.org/index.php). Use the following code with the files `imagenet-simple-labels.json` that contains the imagenet labels and the image dog.png that we will use as test.\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Experiments:\n",
+    "\n",
+    "Study the code and the results obtained. Possibly add other images downloaded from the internet.\n",
+    "\n",
+    "What is the size of the model? Quantize it and then check if the model is still able to correctly classify the other images.\n",
+    "\n",
+    "Experiment with other pre-trained CNN models."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "id": "b4d13080",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "c:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torchvision\\models\\_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n",
+      "  warnings.warn(\n",
+      "c:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torchvision\\models\\_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=ResNet50_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet50_Weights.DEFAULT` to get the most up-to-date weights.\n",
+      "  warnings.warn(msg)\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "test_photo is dog.png Predicted class is: Golden Retriever\n",
+      "test_photo is guinea_pig.jpg Predicted class is: guinea pig\n",
+      "test_photo is husky.jpeg Predicted class is: husky\n",
+      "test_photo is toilet_paper.png Predicted class is: paper towel\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "import json\n",
+    "\n",
+    "import matplotlib.pyplot as plt\n",
+    "from torchvision import datasets, transforms, models\n",
+    "from PIL import Image\n",
+    "\n",
+    "\n",
+    "\n",
+    "\n",
+    "# Configure matplotlib for pretty inline plots\n",
+    "#%matplotlib inline\n",
+    "#%config InlineBackend.figure_format = 'retina'\n",
+    "\n",
+    "# Prepare the labels\n",
+    "with open(\"imagenet-simple-labels.json\") as f:\n",
+    "    labels = json.load(f)\n",
+    "\n",
+    "# First prepare the transformations: resize the image to what the model was trained on and convert it to a tensor\n",
+    "data_transform = transforms.Compose(\n",
+    "    [\n",
+    "        transforms.Resize((224, 224)),\n",
+    "        transforms.ToTensor(),\n",
+    "        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]),\n",
+    "    ]\n",
+    ")\n",
+    "\n",
+    "\n",
+    "# Download the model if it's not there already. It will take a bit on the first run, after that it's fast\n",
+    "model = models.resnet50(pretrained=True)\n",
+    "# Send the model to the GPU\n",
+    "# model.cuda()\n",
+    "# Set layers such as dropout and batchnorm in evaluation mode\n",
+    "model.eval()\n",
+    "\n",
+    "\n",
+    "# Load the images\n",
+    "# a folder named images_test is created to add more photos from the internet\n",
+    "\n",
+    "test_photos=os.listdir('images_test')\n",
+    "for i in range(len(test_photos)):\n",
+    "\n",
+    "    image = Image.open('images_test/'+test_photos[i]).convert('RGB')\n",
+    "    plt.imshow(image), plt.xticks([]), plt.yticks([])\n",
+    "\n",
+    "    # Now apply the transformation, expand the batch dimension, and send the image to the GPU\n",
+    "    # image = data_transform(image).unsqueeze(0).cuda()\n",
+    "    image = data_transform(image).unsqueeze(0)\n",
+    "\n",
+    "    # Get the 1000-dimensional model output\n",
+    "    out = model(image)\n",
+    "    # Find the predicted class\n",
+    "    print(\"test_photo is\",test_photos[i],\"Predicted class is: {}\".format(labels[out.argmax()]))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We traied the model with 4 photos from the internet and it worked perfectly"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "184cfceb",
+   "metadata": {},
+   "source": [
+    "---\n",
+    "Size of the model is 102523.238 kb :\n",
+    "\n",
+    "---\n",
+    "\n",
+    "    \n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "model:  fp32  \t Size (KB): 102523.238\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "102523238"
+      ]
+     },
+     "execution_count": 25,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "\n",
+    "print_size_of_model(model, \"fp32\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "---\n",
+    "Quantization of the model:\n",
+    "the size is lower as expected :))\n",
+    "size is decreased by 9%\n",
+    "\n",
+    "---"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "model:  int8  \t Size (KB): 96379.996\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "96379996"
+      ]
+     },
+     "execution_count": 26,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "quantizedModel = torch.quantization.quantize_dynamic(model, dtype=torch.qint8)\n",
+    "print_size_of_model(quantizedModel, \"int8\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "c:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torchvision\\models\\_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n",
+      "  warnings.warn(\n",
+      "c:\\Users\\LENOVO\\anaconda3\\envs\\new\\lib\\site-packages\\torchvision\\models\\_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=ResNet50_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet50_Weights.DEFAULT` to get the most up-to-date weights.\n",
+      "  warnings.warn(msg)\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "test_photo is dog.png Predicted class is: Golden Retriever\n",
+      "test_photo is guinea_pig.jpg Predicted class is: guinea pig\n",
+      "test_photo is husky.jpeg Predicted class is: husky\n",
+      "test_photo is toilet_paper.png Predicted class is: paper towel\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "quantizedModel = models.resnet50(pretrained=True)\n",
+    "# Send the model to the GPU\n",
+    "# model.cuda()\n",
+    "# Set layers such as dropout and batchnorm in evaluation mode\n",
+    "quantizedModel.eval()\n",
+    "\n",
+    "\n",
+    "# Load the images\n",
+    "# a folder named images_test is created to add more photos from the internet\n",
+    "\n",
+    "test_photos=os.listdir('images_test')\n",
+    "for i in range(len(test_photos)):\n",
+    "\n",
+    "    image = Image.open('images_test/'+test_photos[i]).convert('RGB')\n",
+    "    plt.imshow(image), plt.xticks([]), plt.yticks([])\n",
+    "\n",
+    "    # Now apply the transformation, expand the batch dimension, and send the image to the GPU\n",
+    "    # image = data_transform(image).unsqueeze(0).cuda()\n",
+    "    image = data_transform(image).unsqueeze(0)\n",
+    "\n",
+    "    # Get the 1000-dimensional model output\n",
+    "    out = quantizedModel(image)\n",
+    "    # Find the predicted class\n",
+    "    print(\"test_photo is\",test_photos[i],\"Predicted class is: {}\".format(labels[out.argmax()]))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "After Quantization, the model is still able to classify the images used for the test"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "5d57da4b",
+   "metadata": {},
+   "source": [
+    "## Exercise 4: Transfer Learning\n",
+    "    \n",
+    "    \n",
+    "For this work, we will use a pre-trained model (ResNet18) as a descriptor extractor and will refine the classification by training only the last fully connected layer of the network. Thus, the output layer of the pre-trained network will be replaced by a layer adapted to the new classes to be recognized which will be in our case ants and bees.\n",
+    "Download and unzip in your working directory the dataset available at the address :\n",
+    "    \n",
+    "https://download.pytorch.org/tutorial/hymenoptera_data.zip\n",
+    "    \n",
+    "Execute the following code in order to display some images of the dataset."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "be2d31f5",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -696,10 +1848,22 @@
   },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 1,
    "id": "572d824c",
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "ename": "ModuleNotFoundError",
+     "evalue": "No module named 'matplotlib'",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[1;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
+      "\u001b[1;32mc:\\Users\\LENOVO\\Desktop\\deeplearning\\td-2-deep-learning\\TD2 Deep Learning.ipynb Cell 49\u001b[0m line \u001b[0;36m5\n\u001b[0;32m      <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X66sZmlsZQ%3D%3D?line=1'>2</a>\u001b[0m \u001b[39mimport\u001b[39;00m \u001b[39mos\u001b[39;00m\n\u001b[0;32m      <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X66sZmlsZQ%3D%3D?line=2'>3</a>\u001b[0m \u001b[39mimport\u001b[39;00m \u001b[39mtime\u001b[39;00m\n\u001b[1;32m----> <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X66sZmlsZQ%3D%3D?line=4'>5</a>\u001b[0m \u001b[39mimport\u001b[39;00m \u001b[39mmatplotlib\u001b[39;00m\u001b[39m.\u001b[39;00m\u001b[39mpyplot\u001b[39;00m \u001b[39mas\u001b[39;00m \u001b[39mplt\u001b[39;00m\n\u001b[0;32m      <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X66sZmlsZQ%3D%3D?line=5'>6</a>\u001b[0m \u001b[39mimport\u001b[39;00m \u001b[39mnumpy\u001b[39;00m \u001b[39mas\u001b[39;00m \u001b[39mnp\u001b[39;00m\n\u001b[0;32m      <a href='vscode-notebook-cell:/c%3A/Users/LENOVO/Desktop/deeplearning/td-2-deep-learning/TD2%20Deep%20Learning.ipynb#X66sZmlsZQ%3D%3D?line=6'>7</a>\u001b[0m \u001b[39mimport\u001b[39;00m \u001b[39mtorch\u001b[39;00m\n",
+      "\u001b[1;31mModuleNotFoundError\u001b[0m: No module named 'matplotlib'"
+     ]
+    }
+   ],
    "source": [
     "import copy\n",
     "import os\n",
@@ -940,7 +2104,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.5"
+   "version": "3.10.13"
   },
   "vscode": {
    "interpreter": {
diff --git a/dog.png b/images_test/dog.png
similarity index 100%
rename from dog.png
rename to images_test/dog.png
diff --git a/images_test/husky.jpeg b/images_test/husky.jpeg
new file mode 100644
index 0000000000000000000000000000000000000000..dddeed32b163343c766694bdec1d95d56f1d20cf
GIT binary patch
literal 50279
zcmex=<NpH&0WUXCHwH#VMn)Y*9R`N~4;h+VD>Bm<7<_#hv=|r|I2ah)GZ|PI7#J8C
z7#O4&K!9ZeGXoO?1A_<y14F<B7C1X)(nK(u*}#y2VbUZ78>Gp|z|h>vz{1MFT*1)T
z%D~9V(2N0_7|0Eb3o=Sd3as??^NRHnQ}UBi^^)^*^(zt!^bPe4^cigID+&^mvr|hH
zl2X$%^K8}M@7}MjkeOnu9&YMu;9KCFnvv;IRg@ZB<rk7%m7Q#vqHb@;%VkqxRgqhe
zn_7~nP?4LHS8P>bs{}UPDzDfIB&@GwXH%4tW)<Y(231j%l4h#}bxV19xn8-kUVc%!
zzM-Y1rM`iYzLAk`QA(O_ab;dfVufyAu`<*Em(=3qqRfJl%=|nBkb#Lw`K2YcN~NWl
zDOPETX{JdjDW<wfiN-0qhK4DLx``>qhPo-n#wq5eh89L?hDl0L>nm~#;MQXe3MD(6
zirfM#=c3fal2n((l2kiTkm?#(=o*-Zz=B!Nz|g=(A5EQ&KFDE64z+`b1i84`+33SV
z3KVSc&`!-uu~jcux3~L$o57iZjfItsm4%Ism5rU9je|>oi;I(!ON5V~M?hReLPA_b
zOiW5vRbEP3Sw>7uK}SJZO<hw<Q&L{nKv%;+RYOw)WC$ZWJ3AK#moOKXu!fYFlm^M*
z{{aR;4#o*gBFv0R3`~NI%z}*nk1)tHFfg)$(ljF|(SlMR10xeN3o9Et2PYTz|04`r
z1sIqZnVFebm|0m_SQr=>YZ;lC8CV2ag%k}P*@OcV*_8@Kj2b5{<WP3ncu+Lx;s+Ju
zq@pHHE-`TlNhwt|bq!4|6H_yD3rj0!7gslT4^OY)kkGL3h{&kql+?8JjLfX!lG3vB
zipr|yme#iRj?S)0lc!9bHhsp-S&J4gS-Ncbij}K2ZQinV+x8thcO5!><mj>ECr+Na
zbot8FYu9hwy!G(W<0ns_J%91?)yGetzkL1n{m0K=Ab&A3Fhjh>2=f`5zXTbW7@1gD
zm|56C{$gY*2YFnOg;mjzO~^5jJ+V+&$*7S-#A)KfjR!fEje|ajCKX-e5>qjGsQMA)
zHL%Z!^H>vEK7)G<;jdc^Jj{#?OoGgU4E7AiqGR?i^?lO(M`700X%{q3J_rf8J$*Wt
zoA&eI>cw*%&V357SKykM<nm9BH^X#MXQ7_OiX{%45{k5@%nvMDtmhq?<n<wQHMjmf
zo3K3rE0d3g?b((#B}ZN_bNQ8+y*;}-+CFaYnvm*ntT?LOFIXb)%j;^smaY{C-({>U
zH(uP~Q(xJCtaI(duwxga<!}Dp$t=q9&5`4MRI#Ua|IME>FWhpdykW0Zd&K)`YoCJu
zr7f)MB?JAoZm%~l{(GzNh}m|{__Wm9(M)w2`@PC<9`^0mVLiVzzr$%;;ZgTeKE7mT
zwyn=<w(pGDo0oc$OGazKpOvazCVojQ)m*!Iw@mW<T(kJ4|1?AC96kQWzM)gs7X*Gb
z>`~Br(o+}m<7(G`hE<PWMmyc-{$8Y&@!-@ahLu6HuNpkrA3QrEV4svuooHfGt3;ys
zr?uBT%Kl!QztsJkMU+{RL1y*)G(CS~)h25mg+Hspw{*p%vOf*i4P)|9DP3*kEVZ1a
zI=Ck_`;MsTozuTr91B)0|F9(6^mfytt!v*tHk=gy?7PYN%@6mL|I8>^Rgzk*dgpXB
ztNIgW<y&(uSV|;pUHYW|XWIR>j>S2b7`Jw0%}+GF;y3ACdPm{J#<RA%(kqyjiW!MZ
zCWlSCaqP~1=B@4<=SOk|#YWCosjgaZ)X?i43uD(6MU}7s%Qf#ei=3LbZ0YBt@6`P_
zi{9ndXPR4;*eg1L=dw+|_NrNqg0rosJeoY$eWRXEYhGCKRs;V_Yi=)^C+Su<m2q3&
z&BMo*+f6Ovl$@_>VE8gr-1^mVi5qPdo9xRN?7h;?uKvuB^@!iRertHf?98u=g6BKr
zZc<o%kc-vhgP6~QfRGPsk7->IUiGK<tk<NQ8=m$D^@P3c+4*R7tOw(*y{_4(iu}K4
zE!^7XcE|Co`)*x9n~g6VbK-LiA5AWN-F2-`<8d7O%ODev)AOd>2@2a`RroH&t(Bc!
zG)YWn$MVzpJ&zZDSvEyb<K~ZRzOT|xr)i&fkSx+txw>t&R^Y?u>$Il1&%7>T`RUcn
z@W30%k#0BdWQkwC(m!=$@En^Krov!nslH`PL#z8|PSq{Bz<IdXH&@G9?^(_IYg=M+
zW7ewfOn+6eDl|SsuTW{5OW<Xbo64tIZM9A}+<Cq<TyM>fv*rfHs?mbI2Y0TBy1D&x
z(ECptKL`C>dPC58T1(9NUCG%Ogk8eja+uoI#1uUcoaL74#K6?5%|GjqPR|~mN6T0m
zCwp|MINddLsi@;z9pt)d(YhF!bWx_0+nhK|(^Zv!x(4m=v`mdE3zMEVOUZ3=(FE2|
zje~n%X;*hP%N##`X}!+u6DkSU{LUwvy;FLke@;_Gttxb;#|`sPr-A}@5!(`}T;bKu
zm%BWdMqF+TIL>i*@~kb2LJa&Oi`tfa+ZdbbWK{e!Fsyf)RbbSoeMe(wmkVmLa=y~q
zI(I?K#NB&+E;WjMGRm@elEgf(ZRfAMBD3-xr;F{rG2tsq*}VvrN!jk&N`IEK&3-yx
z=Q-P!ykCpv1m9S&;Y7;c8z)SbaX!kQTfpnL=`?eG;KY}~TizFLozD=meECe7)*9Bp
z$$_S~)L9zC)LT^;j{W%+s=Zp`>4CJvTo*$MxlK2gzZF~~$!Api!u8JVbG(7S`R}?W
zM;YAeSah&>)-}^q)9NWZcTE&-QM21Jt?0yq)ysGz+O9BpW=)--c;I!gR{w&#r*2Ee
z^PX54(B!q@XV#@R$0Sm&@Rg-*F#EEU)9<15)Vw;6h+~{ZP7A)OH%5t7g=F)o2}(*9
zcq^zpU+1}E;WqKA-hCl|=P9I?y)dumSS7w|Rj*^y=F(qXUpBmSvpK&qY~PW|){$|u
zE%`$FC+=VB->~s%=bsl>H`XrRGjU;chkZq>o2T+t<C$wOPM?~}5+<?ID(ZL|@9v1!
z0FR<&^FHkCY0CU?u#M$o@Me+6Yb!s@j}%yY<yhbG)3fUy_j^tA*?x-iXk4bUYuc1n
z=E<{8@#;P>R!ZW%ub|G@zVhp@urCR(t(B&BYs-m0t(m{p&Y;SpIP?_91doo@eY@F=
zY>&oxRC%qP;F=j^wB);lrAxrcE5VOvN_yt;8mP~lyESr=)@@Jw*5cE`+AmHNWd~Kh
z57Ib)igT4foTu^p>3vr=-#+<t8~^jLk`p|Ic{NGXw->$gxBZf0SJJI<Yfky{%^N3J
zt@B?RtIwP=hdK0Giu;MzPODY-9ooBV;%Vj6r!|5!C!T(4ekP=AUYf9Sa&ze8ZJy%O
zp0$U02kd@vbXLX3X}2`HwLCv9E?u#Ib5hdH2^+thwJdqF?#7JOA-g8!bbAE3@E$yA
zs&!`Vaj#C3lRfpuPhVZid>Z`BBrfnu&8DcS(#4j?l@=x{)QRuSbmn?@fNhzg=z*ND
zx@!-%-dQAj&RjLdw`V~v!<Bbk;ZKh`c(vWWxAy82{fA0rH_9Hr44iPJ@x-Es8?|_Z
z!)GkgxVuHz_2;!kzTQRiK1wp~6xHT9^_*$m0n1~o`V;%C16H|d9ri8JbeR3@(BE5;
zZ@5nHbaPz&$h5b<%yPNT+CEqREv;*gy!o!7IwAC&*49%Zk_TT$pE~&O!_nYfpHALB
z8oE$*!B6F6JF`;{H*cF}`qR_-sCThWfJV5MvHzxwK#z0I!JE{l#+_L{dB4m=kCQXQ
zGam0;9qh?uXJqU?V@XYeluYQl!_!p6H|7RzpVPHo?8PyOC)aucZT?m;EiyY^o%hNr
z<jUsFoi?jW0;k?EIPv(}(n4tg_oyw3KMNIpg?cPJ@<gEUc2}lrDWB_|X-0e@DN-R$
z@)D1B%d=(*WC?r9oSw9J;o8~#jH`U!3eR0Wf&bIG9S_w_mNHE>KCeE_#5E@2a?~pY
zhNGo*3Ll~mXFqq{x$#Q!)}l`lE)8raQeSCl1<zH_6g_e2PJ_hF537$~>Qs!quO8+3
z<kJlu&-0h&t-I~_ux(v<q}qwCUF89u$L*(@sV9Eg^lxgZWIyl8)dni5esA&?2?v)u
zp04#-sdAX7FlZ*TzKwa-#os2n-y2LS&3-;fxVlsC@srq7+{#m*%=glExPAP##G>W5
z4T}U`{%6Sibp1T@ysa7Qj@`Z@cQsGqyT)hZSrdL1+Lb+9J&jjNqfArLAwsX5Ytk)8
z1;swisAs%qTMvF+ne(S{&ezwTZbDUDx87Me7+F4=^YPf~Y4)u~mH!Hqk1C5kKJ(l(
zckLvTXP!L0N}iUB9&y}kd^xLH$vQF1ddbfMY11hyw`@_pU1FQuTez8RlJDUOc2gsR
zJf=UfJR7lP=Ut;E-IqF6?~BxC%6#XZZ?k%$j{A)(J3mLOB(00}>oRBvWlQJX5q7D&
z>uI#et-jXv`WYvdObpLIe_3)x;@lascdk8eTO;+W*rb6&;pc4im2b2%gQt2f71&iR
zajW;#%czQW6|a-~9A6yYS|E1)r0VzgKYIJhC;67h3$Nn|@(6fv;z<4cb9??X98ii4
z4l~O;6IyAJ=O8!rjqT6CJ<9Ly&fEC#=IpnD56^70IH_*)x85p6BXLr7PR^$%if&bU
zMf;a&on5==md|RD`PI#nLsG7mZ22g1t9o1BjP+bqbNBpn{uvn7zO7$#@;kpv95>kW
zwmhpie>dl9b@y&Foye}H8?whN-~XuF@_S=4PyW+8vNsPk_f?1Ue=Ci6o1_!6yXSt4
z#6hNe4D0pk`z;Usn{%!7+lmQI8wH=b?FeB>w!gIg)zh3cb?4a6&%Gp6xU)b1@X|u@
z?W?*ZI%cfrInrd;v}@(zo9+`X8Cvu6-L2i2a>BZ60)uVV5p!+1`2sBcQ%cuGABtLb
zdZX4`)&9Ky46UVBdmH<{USAUQ$mRIL=RcoqJnDSQF?!+L?Bhui`X_d}yt$j5`Pkv$
zW%g^E6*qO6x!KIRpyJl@b-k&oXj-eR*dky4*_XT4niMUX!v6GVROieFh0@h5Iv0OG
zTP66)Y)W6yvxf^OXtI1hxkxkeej)pfaCf`;A~WtRkm|C|2>h_yPunHwrOJ29n|e=-
zCUo51H&t(bq1A4kjvK$ion=>5-tOY6KC+_j!OEKRVO+ECisU>Fn_d{i`E~WC=%2Ye
z{Y|bNWSri$Zj(pq&)Z#XlcT1WUfC+3vt;G7MYBA`9F|+G7JQg|N-k^f*+bJLMK|{(
zZI#MY%HOq9A!$)q<R*ug!MArzG<e)qe{wqq!<wZM_g*K)N?!K4@IhqG<tsI_Ppl4?
zi|n4jxvKEQDYN|?zTRCqK^sL*oV%}_H0zqrjr5`;)z=z6xrwr6Wj_A&_{5jBL7P}E
zx_0!-?YlB@rg!lHQ>_hL^3waZ8l90|bmh~r@2SNutut25nN_-5T;cNU<Vl7;o?>zw
zkCwf(-ry^?<h-lIs{LtqZcP8Y)OcnJqra%tq~wFo*80iZxu!ic=IoA3kEb3s)NywV
zKAg9^K>3*Ms>A!IILfa4F1*?K69eO`%#()}=dXy<4>nx0Xj0NS&6oKS_Ig<n3(9{s
zo>($rOXbhX6))clzni@7#^M7uSB!a2ag{E0h%;~x4*Y4<5b@=#%C&<FzWSd0JS(M;
zHK#|a_ii(*EpJ5Zs-3bK--9=?JSv;duhzLPMYZhAVwrN$ml~&DRZomlT=44t&0AZy
zJz_fHC4X$MyVw5Da=FKQ_@Bmo<NO@xE3)>&m6<`Q%FH`VpUIhD7CQI*oa-U;rB%E=
zw~zE3@7XBt_4H4&bnMGDB6&%Fm~Vv5=+fFE6CuO3%BYE9ndy$42cP{Xnf5<TJIJ)~
zg!9zoSwCMoUM`yut2Eo)$iqQ$`R(_oE1TKYP1RaD<LgVWuHAKc%QgjMeilsKrF<}e
z<H_$U;jWt)W2&!As(skp6Fl><;JnOTRd=Vn@z0i#^SF1!RMhw8_nN&L>%Kggx+Zhs
z3i;yVgCd8#&F#zgrr)>Ta{8$md(}=Zd4+`MJJ*QXwVR}zROwTD$EtQURLRIVB=y$K
zl00{bv%$M|r5=3h@23@$yJM+I#+1a<X11EkLc(rqH&urQy}NcQa@E|Wv;AwQ=W+g7
z?N#V0SpA~+ciPPz+ZC>Q>V>LnEjJB%^&&cEkJs;QU8f@Nu9Mt6tIC$~^unvR^>VwL
zFFadpF0f^0m~?y8u}S|NvrbALPTG_k_HyxzFI7b^Cd`?4;^nO8Z4>5CTjQwgE1~74
zJ*~-e>2|?W1;%Hy4i_aghWZ%>rcG`1+?<do{XXG(?4rY_+uBtw8mNVzh!4<i+wQ+*
z)sDPB*K?N~w_P`Dt+w`UhO+gA{~4~WTO%WHopxor&Gtuy)xEN_EY8o`l*jU)f!|iU
z`NM3xnU8h~I?oi`#=BVM&))2-f5Lai?2OJx<Z9ehe6eG#KF^ONX|u{bD|(O2xbvX&
zc>ce=<;TR!tt;QGDK^fKdbYff_5Qcc4b58;i;iFUdSIP%<R+{1PL<tVk398~f2K@x
zJDGUg-%tDQdXon^cY05Wa%-Hg`l#O}_WAG5GZ{f<r-j%IELLyVl%9F*)|cz%f+>F{
z%6vYqHeY4ds#nY*{`Z_zB6(u{93C{@pLeylH0@XEzWD`?sXKeT*prX(RxNKYSw8R7
zwaDa8hdzb6GfPPRaW%VlVB++pKepYE*fB}BQH_D^{yf`?b>WX(CLSzkvYt{lBW2Oi
zp5%LbrK3J<I=9I;ucq_kk=eyx&#$dC+r8-id;zWd3qL*n+5Egsw%z7)q?gW$NMrwp
ze|&8pY1KU5`Nj0@t)p!^IeIcRyKOerO?~#Cfw!bGH#cdba(d>uPX`_-e6e3zHEplT
z>W3@-oZtKW!Q*X?N0Tj<bH;T2T0Q5b#yJL2gO+m3HNVq7eHXV0`8K`qOayZiZ{?)T
z7FRl#l{8u=@Go^1>Dl<6y{q(0q1W!smsa_lnkd@yvF|FEP<8VPpIfs|PCaj{wRh$H
z;?I^WpDfpOwmrBPSERKn-dE(*oA&0e9lv*Nwk!K?b#pzB?uO!RYgA86x?jvV>$%FT
zQl&6?Wv(^X&8?gBxFu}0tJPO}<;waw9dR@84L>9vIA`9EHPiN{O?1e45TyI>(lh%#
zR?QK+W7TAyolsQMTJqN9<>G)jJ-@<q$}hz%yTFkj6E?*>I`_kW21D7bEm>#0a;CD(
z__;&m<eUXRqnT!GZ%h6pz#wxa=tPr&=gFm7o(n%ul5P#<`MKiOOVxt46SmH*EjT!#
z&-m7ows@^Odm7SOd3D5166a}f*WBLuXs+5K`R2<@Qw|r(oVgJ{gXhYH^DGlZCmJ01
zy6n-dO)S4=ahHhM-L4LP8(pYBZ?Crb8U9236UtW`?Brp1$m-mE>yhuQiK2I$uS`}+
zGrDn3Tj%byb63yrjr{HL+y0c><7*2Y*Ggv9vK{H-pTWl(%_k*$I%Mj_#ihm50%xty
zui&_4#9ruoe44rN#C0~C`BF^%-Z?%^G7eGARJpU__}Y+?J#iv=&o<}YPO6-_H|vvQ
z`t*Y*SE?y4KKWqQ!o110+f;12*49?5=&acDvFk<HQ_o&&?dfYnzYFf}niX`ya8kD1
zrAGng8-H%TwBEZcHHNV{@YF;v=cQ9DnPskQ`|Xn#+s2xH=&0Ve=a#asEXCX=ws{0q
zg={T9b0O|%r<woCs1;9oHauUR_2sSVP2GT#-<RClDf?r+_J^x$w#O-GHeFmNB7E8Q
z^|i+-5sz{bTqm(s&#>4L=sElHt!GmbR?WD&`)U7zRnuL0r}QiQH1POpH<kCy`ERAJ
z)^WwtlaF1zx1_h_)b1UfY;X7Nn|f(xx?t@xV>geW_;O~^UPag3{=<CQmulu4a`)uC
zjdOq0Tqw6t?~>teW!cr|j|cYdJfo$Zm2u{GILo^)E1Kn}^L|}Zkv{uukJz~iyQ}4A
z1WuKkWuKmEYhtwbbg*73v!$q<$Mnnx!Iw%6-^*-Ys&!)0pK}~X-PIj-w1ovX7`r;_
z9IpKIA*}1vr6_hookEXGq6(9=&Rm+ET{!1+xbE$o_B>AIS&NM}d-?2HV8iuhTGxx*
zmkaxiJtU^yVSFF*xVqos=g!1=+GUl&iUtgeM8Z`jSge`k60s`n-YSujw?57$zR#XC
zz6^hx<XTi<IIE_E<;R2dQTuZBOj5S4pIfqip8vh?t8+8C=HwpV8(GknTKH+=tz}bG
zE;3k{nr}{dlWyBEE#y{R)OOtlhK~Mo?4~MNNlHs5#U7FP_q{6jW#N>?8-z5~+eCS0
z6>B#|One!&T8U%d%kx#O>o1=PIq-O`_HMNskrC6K435`KwA5Pi>w4@X?%5?<97Rqg
zIevGPu#fueFBWgxH~-rXv$vLv?JQDDAI#p}8)d&W{!EJ29k$0zd)L^AWUe<}HOn*R
zi>%T8Gt)~Z3M@%i;krCKWXq<QlNx?=j~3tBw_CP%#-}Ith6l9VP3JEAQT91(n)UDB
z%eu_0r?E+@R6kx|dU=z{o%dpY?nS?E<zB~sSMqIK@JBn3XB&6>_C8aoZe*VG<=3z9
zy%}uxv?CU7^Ejlz{b@q>`8thEvF-m1F1~r(!?w#kv#h`)Xve>_DchvdCD&Edvj0)$
z)v14RGHunxiGQ3W|1(_Ztx}t&)bHP;#B!^PeZjl>n9DWZJn2y%p3O?W`^;EOc|+j&
zFYB&7_uM?|^eIEFKRYBAMm@NBvaRotL0yo?^BsS*=Y2G9U%0!+;F04A^_5>mr><SM
z`{7<M_lh`Q%ZQ0eeGK2ft?PHay69@oKc~>YZ1r{-RgIG$8kH<inkky7{`CBZ`99ZO
zAI;WozwmfRkfcxRhdqr)(<PNR?Y7SeQY^kQry+YuuG6K@Zof?;Bx9PJ8-ILRvv<Bp
zOvJLCQ~OPdwKw%H>036<qT7e3Uf<^Gi~kJ8#%tzGwODo5pwDeW56{Oqnf2^x1)pxd
z5}#-hyK&>m*P%i8GNwGY`E_kdxo2ycsCC1~=dx8HtM>-1nR!ktMtk$mTKlNIEOxc~
zyc&*t;>|uLxYo14oU3JF#is_gmtU7kny&civwF4U%(C|<j*D_!{>T`~sT<AFBcr@d
zYMKZC%&7{$Ls#xSsLcFESMEAv<M)-<&Zsm6xV<Ubw{^Ago*ljmR=j>2SuVJ7vZUga
zZ!1g1%@%D;Q;T@+qi4>SC4SgjA?*F7b%%rkmh^0p&0MunrGM(luWMu0O}O+fXot=U
zCN)NxN1Lz2%sgf0c5&jDwX?fh!x$d$O|4?-^N^gwYVq3X`wVNDGt0a?C)e3U?cCt8
z+h2Xd>WfENl9hQRuAX?f)nTc-u1m`8&8*LgXJ*JOS}6G{%cXao8HZJM*qIaE#~7b2
zNeO8#Nl4$zmz&WTaePCUQ_q?uTxBnXYS(c1giYxaGF4_S7xnt-EvkCpsOy{cg+@xB
zT)6ejegz%3W>cLk-7fN+V~t>6#mdKxOB2>MEMB<c+2kbS$Ku`v{aaepKZ~yY7O*Kl
zaME2@{TZSz!2%2~vd*?NrdV$XmXW$-=vy>pH&5)w6AKoe^}AuDe@Y}+b$ZEig~AJ0
z1J3K7yHp|f)#iMPbVmFv!^>sdRiT=?3wv58G`|nLXR^^l$m~wxoY2^_I`j8g<Xu_i
zKlMhQ+w92WLQV4xvS!_rIH~2pv*&7ts@84UgG_7g*t>NURd-!fRX-!HIP0D`zs3PA
z!^qEf4$IC9c|Y5R(P#IB!_u89Pd>S?jree{Rrm7J!|R+2R=MxXUe4=zE__?-Aw8d$
zdjyx+&fP!Ps!^rKL;cUUXS*IO&J#&j;kgp5b!;V%`!bP<zh>23*4?_{Y59YHX?ydP
z*D8KWa{SD*&hYC|AH~npC4W{ohAb)y5m#TXerc`l!HEA1<=@>U-(68ix@SK%FS>7b
zVD`Bh*X6<rmC}-ymhXJ4M7kntPHu3Isk+Yi*m9lhk%dl2q`A-Q#{Ae_ZC)%jTSAv9
zdb`iOtGtJ_1$VzMUAQhjL&tK-q(`RxXD81oT`%e#Jf(D($@|~y%(^Z#82#znGmkfT
z+Z3(EN6Owmn_XF2FsXUYO52soPro-@bm!ZuO^H>3-ZBRks2<B$YbkuCif`+lh}-Hb
z7ry&*vyb6+$kQ&D*xl8oD{>ypK2?_7d_(G>c{bOZOS`t+7oWi)v8rI-pWfRQe_nn&
z<g_q7am(+kue~c&CvI%q671{wLV9Y+{aLY_CyD>Ox>G5vj*qQk+1=X>r!HE^Tw3MT
zuFkXn)ar*}$<BJZo7;PyG+tV;T}AV=XYf32w;L;ypVdwA+ptKWE@z&W&eHd6vnSk<
zpDIyuJZjx#H;!|TZ?Zn0S)ZrG{Bgyy-$riA$~swdcQJ3+#3)mjRW90<o1(k-q;<cA
z=w_A54)@7#G&krd*;r4tJZiBoQ{#*1y3^vfoMS8EPtH>Kx45D3#x?iWKZ+}(A}8l+
zhS-;K$o%#HaeYaK0r%^WrhuY7P4&liiEZ^z=9+Qp`o&|KuF7k3Y92n~Fm1n{xMAMS
zTVcx5Ri)q8`pahj)A_J<jbQCGiM25k^d9RRtT@2JBTz3BzM#fwmUeV^<W=?=%iGvx
zx@P-c%iX%xRC#CUbY{LDBk2#enI>I6pZS~PEUv15WWBm?aao$DL&1^cXa5;&!b_R&
zWy^~#-WwEk>!#fq<u4~r+MioC`QWrgAI>j}T2sE6aci=~0rqw54`ag*zvsG}d27k`
zs0nO2rOZ9w4=!{`<ri}{UEW<%eJnxvnFD{3yZ>p^=;Lu#Rx$$4*#h%^J^iy|F>mvM
zllDtX_Y0at@7^7uSMorry6lcf-=aOY-XGXj_4QBquG1%#k3>9FQ8@No;+Unxzb$K5
z<aL$V3#{3y9ltsLu<h0-ofBDB#GK%lVXz9feDK%hc6PAyvnk!xTFNDIbIR8HUi!i-
ze3)DJPR$*U9hL7?j64pMowQ`W5zfElhw(oldx7{@(<VzKM@BX<JUqEjw?u5O?|R|*
zX#XW!%CqJh6>e;>pSecAZC2;b-Lnp*)?SpKdbnoI-p!4FvzK1o!}U3{ukYQb#}@Yr
zqU9H^H@@)Y^jqa4>?MxZ%zn1)Jn(!y`>((amu;huh4JRCE@kRewqR-UlwVu)t#Ip|
zzKGUdOUw0+Gvk9ky!^}&ym<jb*<bDRfi>F;j_@7PGyD}kzjxW>P$A1#$D{jar5Gg%
zm~l<Zbr8A1#?*HuY6g>GD)UzV%v8}M=?k8(t=#ihJeak}bO(Q)0f*ZBZw(WENo_Lu
z5^|tMFi=`RT77<2yV}gKRJJ|V|8kG~x_Z_sG*x$p@VsK59X$p&H}Hg-_P89ilH0O#
z@5AiksK<HSE0{~uqNW(1QdKr({rY|SM$Y34S8X%r`2KE*WztTW;9Sck+eN?XMNgJJ
zylVc`e#vDqG2GEj>*q|I)up{@GD}6L<>Rh-i87PdOzL_2Hu!QuQlehjpE=uKhN<-K
zJEwHc@Y-r^<1;lIxA09(_0*g^Dcd8&V@ZoyuaGwXOv7VZdpqWwc+Q@X_f_eBW$wn%
zM!CIO+0{!8ANL-4-W7APXvZv<;HjCOnFhs|)*R;(ey-yv!x*wYRApic_m*<gLbZyi
z;nI6~xxBuM_SFZN*A;pch59pG-CeDDPxkZ1v$3`Jich>g>uG)B#J-5z$4#oYm9r;G
ziG=cmTf3_*-M-h=W(BiMSh3fs8&fy%`&Ui$46!(zzN|Xrx3_htSjx=E&Xf}?m8V_d
z5wrJdyOp!jjoEH%Tt?g9Q~RgdrEPUpEnHg^l@{m3d^BR>zH=uoKA9%V8yc`_QrY9g
z)dFX~bxJOIoN-`z{?;qba$j3^ca^;M)UN!`V0LIuN}>OpRfjm@XP(oNb@H5cCpVz1
zYst=OyV{mW<y<m8yZ?uO-Iw3JejbGz9;7pro?Nk3)x2xsn&^0oOlzsF*_pZ0I`4d1
zwx2Sd^LTAp;s&9*FJDYs&2>+1vDK?S>t6anVe%ZgMn{Q1uP^4vX={htXv>LTbYFjM
z!`sj!x;8Vy3fm66d}Hqz6?o_9)AkAS(_8&^98Y{z>sz%gFs$>c#L|s*-d7!8_}hGw
zTek9!|8t}5XWsweu&(w05h}A~{j)Pq))fg_p8qjxmU+#et`i<F`0S0no=O=^65*UN
z|6535_=7vqdJcgX!b)dE97yMRe%*rk)>=u`=^c+<<JT4Y3Y{)8VolagbdP%Y$LDSN
z?@POt<COho%5A-RQj$yRox{fd-4gG=u5&)JATCB;{b|nQLK*(Vm>t_IB91LPZoy@E
z;|K4ocr7ti)ngX}W1?;J`_{e=+Pi*ke(T+5>9anDzR#9Wa*ZzSF5;h9A$MilCY>{)
z)77^+=;n#3bvG<q+7#BElj!Ci9OdTs?_2X%Srz5VbM;)By*oDEy8HX0=klVQ*m>@|
znJpy#Df&**eX;S$%9rz_Cj0js6=dl1SGaOruOqO{c;1w~ldHCwFAR#-xEikK$$V|n
z1%>I^c?=)2u8E~vcQEb>b?<yMef|&rz5d!g>!vlyefv52d(IKf$EIZxUWbmJ48CtF
zyJ?r3&BrTWSAMnZw!4#P_OEnR`>o_nTB3$#$<~{_);+Snub}Y7ly^_>?Y`=(Cq?#M
za=W@_%GA9V!i4T#&+`zC%T89FS)Lfa)aT-&w~U*tyYwU^Y_^_y@LX-JuBq5k>FDj^
z5|)oAc&rXLH}wBJLGZZ3`l`;QJzg{YJdS%s<}SJQ<EO|Mi<f@l-CGZQ(dURYnPn3+
zRl4DzLP2&`<TQ`ZJERVN2s!$tP;iF#hQ~^4vkXqGI?FG2^^5246|bWuXFXkF{&4a1
zNyb~gGOe%3J``4#sBXdXiQ)U2yXGa^b7h|A%W%CZ`Oh%<wDX;d>;<MJckQIBr_T+L
zjA4lQdf;u;sgO`%BbS|N42&n$Hl}|po6dUBZN|QPOWj;!XG>l&3=8hdp0(@Z!t9^x
z1adwHXg6MXyztl6z1p#xu5@!}HYWaIZw^~`PQ3Bjw8J)skJd@;*>Nu{O|g%i<v)Yg
zr4N2HZrM&gsNFUxc*g|8U*Y~Mgc5DiyN>iZ<Qcr*y1V*{nQ~<ApCk7AXD>Fsk2$yc
z;qHn&sYeEm$&(iRXPAF!m&N;O3g&`e*=84(%3kO4dmo??CfUuBD>_+OgG+q7<3H1<
z=GUEW2kZ(<s7r2c;H|p0>cj0DCsY$0gXdm-eBo$*`O0&%1Dkd}m6(vt6H>WyVqmQB
zjq}e=Hn_2?ZFxAi|J0t8uOde`Z*BGLvt7P(iE7Q(PmU5c&-PfFO1~=FGBN*bg7KWU
zHkVWPmHd5nxo^tR&W*p%tgyK}<GA{%of6+(2iC0+PEMP+K}KfsNzqgV<6AQ$%O+l$
z`BYPxC8bm}HQhPm#F3tYtc4tR@05ES*|pi|1h<03i|k<EKHF1~qAAC$LUb+uF?hty
zT(R%UqZEyqZ@UU-bwAq?8<YBE;+9FmaxyYZtg2_U6nnPcd+)zfI%b;0d394a?*kD{
zE$i1Wb=kTrWpCWlZjY$yR>OkDO<8Xr@^)NjoWA3|?^cf}Rl{ik?#iZqE=dQKlMJe^
zt}il=bziBqd<IAAiMSn0#5@@$J_<hXbFFobnnkB_@$d5GevOPleui_Z0#Cdx<2AN@
z`~BFQFR!!;1tpG|eNn$OZxhEl(XOZd63b&^R6ZGL<w`7`YRj@+$iwF8vGA_O{|rU<
z%`F}<hdkb2p{jMlYvn=H%I94kSDW4?pI9^J-M1^Ll>*+1$r9gI3f(iBk!op^60+^Y
zrk&-g{NH_@pR-JDm{9saI8=IXl<2Ohk2e@@&MP`@DY__ko5LnK{-uuJw!8jpn!wLC
z>(|8OZ9XD4$(OdZihSw1Q|FR*akIQm%H$I(X4K1ZH59l;R+rtH9Ievmc|f7m?*5E5
zzSVEtiyV!N6;%J_{cDzg{%P&Q=~w0-UFbDuNz9eJw}~DnO4Hx$JZ5`-Yr6N!=xy`z
zmK8sJo-wiSip4wYdG%8ZuDAVi*E#W|s&D?>@L1!(YY}eBlN_Z!?B1}-UQYYc;fKlY
zu{SSrd8vCRKI-@?{cQWTo_^!+g^Y2HizC+*9-l5`l-S`o!It4&{nW~7jSr7UFJyX~
zS=@8<Gvkl0Z@J<g)zT#mib3qQjNjKyJ1Znxl*gkfqs24V%}>nmRpGtWKd)U|GjXFi
ze~<7L`2!`QD@}L%`1l-DWPYabXqHZJp02W!edoQ+$vflBX1(RR5LjE>X1~^($7hrL
zT{H1@SD#$HUC6uX&D3+BuV;O*R}Omjc{*3j@|BOKHa$+NXp>OTeRXGIRN1E!$v;CL
zUuRl*SV~d4;`{C3eOINIPGG#RHdA+s>FOVE7na%Y%PF@A{Azz`&-ajP_cAw1N`F4b
z&a~}MYJHa5=|a^ja}DQc{bSn6aXm{ih1v4=>Nq#6;@MBSEwjw-wLJ-Udd0JyPcv<Y
zwo2i#O54JaH>o!#KYd!w9R47rdcwKg4w7p`q&>}7O7~XAE>SExQhGMD#Z%?U9LCEp
z{i2t6x2t@pV7@j@cj>2|%C~{pr@nJfIGJ_oN#))bD`k`K6?vM3?ET{OOxMLwU&rCN
z<VwZueVN=d`2v1@$Xc~2X8W{w9($|J*-XbJ`zu*KX$4<)TKvq@?r8th-$$;N?Ygld
zU8a9m<{^bQr(&kX`8tZt$y2+t`5pJU%eR)U+?(_@X?xO*eZL(HpG!QS<=rwb&0%?0
z<Qk_{)&?;jXKkLaF-MP+xk+m&=gBCCt#^`c?dU$k!Y146sVID_;T&^%mqKD8*D?P(
zUsu!a<96n~J05N6+LAZn#1*S|8jf1_R&uo;bEekSEn2#5yX@_v=2@OoO!o3=JbxOt
z;c3#mKeA_bD6KzT<h#-N@13r*J5oBkQsz7QH|3jiRrmc}t#SS7)HQC)o*%AFkX}7o
zGMVdDIgi}p`}UW1zg}suhfn*YNzu1Dr7j0i#~|jCvsF`!D?c>-t_r_+VZN=(xgEA#
zMn`NOJeFkJzDB+GT}x`)7l(tz=WMg30v62QVK=pP^}U$PU;8GmbX%1huy^n22a{Ia
zxK;7O_|AWZsB6+Umc{&dxF+VM9s8_T7w;IIXlyd$l$-DMv(3KKsx!%81^a@twc;Pe
z1Dbw2@7g~9S(nEN+t2Z;udGUs6iq!j&7eR+|MTjiZHJ=f^*{9Qm^Jl%r^yK&=A)%5
zuNs;U3l;W+-`M22Pw*9MP4fq3XXi&8@89p4kh|fcLg3u{_w078U;6cJP3A70$*Z<(
zp5rQ#*>vtr0sGfqT>7;Uf4a-RKU|v{$9(Ky)pZl@JLhN2X11@mR&zZ*Woz5P=ap*q
zCg%z~lTWVQtFOqketFa(zjsGfuk6lVuH184VM2d!*pem*fvKkij3-qrt9-Mh?T&-Y
z<Fy6%bUq4ePP^f?(Bsgy%JWOZI#@oMZJ1X-Rr=}Yt?f2kkJev(cvX4YRll27e|9kc
z@%3I3;aORxrzc^$tB3dEe}<Tke^xKK@+M)=_9oSUMnA_V{Rfh}vMz-sJT6U^y7u6X
z=az|E<SJJ$shqAUm7rm{F8YYE@6y)>4wI6ZEx1DKe|=rcwp*rP;WW;gLls&(*Bfn%
zd;G^X@U&K*we<AD6CuxAgS*zPQVcrhFY__*<eFP|a+Q4N--~3s|2$kR<LrivvW;>-
zPpr$lFe&$><d##7jT3yOg?BA&3;J-5<4H)6hf8&Gf8U<4l;9I<6=sO4Z(p%S_ta&_
zq+4e9XWkLM>#7{rqBw!^>#EPkuImbPx+KfYkDB!8eC?i!6O*kyMd$5L@U?51eCpCs
znMw1f_KTX7t$MXz&-2Twi;hpvtUT%Zf;UEJM$Cge_62c;(ZMaR^4b=?lbX8OMP#7{
z!#}CA*(!o=VacC`CwROLefQNPC{d;-a#t?1snhqsZJzA2WOAP~`etqTnyh_Hr)JZO
z(xix^r%no<N|so$^vEUS31!cAF8|0Q#45OYQU9Y<lZR_pxP>NFoLBC8*(bDTdz8pR
z<>h_1i|4LdV|d`pDx(#R-*?9CHI879m^1CL!TH$bGIe{dX6h}O<+G~h@N!Az&+9s#
zu6?bZG=*<o^jp(U2i83@IKXvo(yuE<OIKv3oMT$?J|eDqHItZ8@tplO|0@1!m3yze
zT%>o;C_<0TZtH~&B9nJ(I`Uns-?Db2$tHKcho98m%&F#&da~@wx=W9%JTC2!I1$I1
z{huNIY^Ac-@7dd))jYRt4a}DOF7aS{dg$+S6%2p<y<U6^h*r81$p1%0eB$G?IXqv~
zFYU_be)Y{<!2XuJvG{+6Y3*F>!INJeE%<fIw5M{O@FWY%e_PMJ+|_Nipi`<|OeDRb
z)_<*f##-*4z<1a0sZ1(OOlw$^e7d`@_P*@i{Y#5JN5uOqEz{CAKX>$l=l8Yp{LV77
z>Q5SYY;K$L)O)@|ZT9kwm*0j?yfcmSdyt*~i;6Wr--WJ8d-=;niFcAaPj7=XbKwj7
z^QOUJ7v>~MOWQX8UOjn!z1Mr4rP~9aF0l(^FFf(*)s+=X4LMHB?>T!~{$<GA1J7&Z
zlXZLQd<va5l_&nO556*gX<5NxDUEwiCSQ5ZGXKgo-3>|!t926DKWu0DCll_?x+CIh
zXaBtUdYWBN>lGvWPk8-j*urY$YF`-ZRKJP+>sqmGQ+g)L*%pRA-4?V{u;=*?KjWVv
z8#K8i+RxmyZ}>OWB$MgWGMBfF|3p(2?_Rbk%hRaW)1R{N_@>DozEe_y&rek@eSc}g
z){1xWmWI2kBn{_(T>En?w}jhE`NS`V^Xk2h?*Aw|=~&%4#`E8kj@3&YTeD!=np3$E
zPq*GK?Q!d7JZ5w><Uw8NjMF?5R~<8aqa7YG^Zm}^CU;3wKaC*WCA;3qwpNP9o4na(
zIFV(M??#D#VeAv%`t7T1JG;UmPwVlDISmJ%%;R|LT_bv+b%~Zpih0dN^{c|kT#_D{
z-0Ei*%Uqoyaa#P{!Mm@tp1-SjklJSa+9!5`&xDmSJs%_PN5_P@Y%o@-P>lF)5<0`h
zUqaF2t!viKwF@Fe*FNmhbezoE`E=Eh8_{b|D%@J!HObl~ROjsugX0%2zy0##RNis>
zduNMRoZM~n#(9_7U8UxEUw&uY)nh(3Uvp)iZq1Gxt9c$<tV}N5-Wz+nSn$y0kh^b=
zOj}}p;*igY<EPZ$UCis6)nRR4!d7tiWs2Iziw6$1eSL8Cd9P3Fj1>%3*G_%Eu`gJM
zea9rX=I`}}2Shd79CJdSspSWIZ|6yl`PH;lt68Yu{JDCUW!zi->5jUx3equcDrGAa
zXP!A;@vguyXiZ_~p0ll8-_H80DD0mrdHYS~ie>9~GuEX0=1SjTD=6nH_Uk+8WV7nd
z%5$zU5!W82F+6Sh+_i36vHP<b=LFlhW}Tk)c3+Eaq}Q`&Y8)yFW)_xmcVAz8n<kWb
z(ImZS>VrpnKE>*3D&IL({#a7<Vug9}%Xz_@ja`nGi=F?^uxhW(zTD+vnSQU`vQF;!
zaCE!e@vfyc7k>sv#Vwe0?_rt9e}=EcQAS$FZb<6YYRiZCxW#_5^?Y(YyX>m8T=lw_
z*SCGY{!6v>N;ga4`!>tj)p~sVabA-uz8tOG`n~;YvSG5$Oy?cB55l)+h2^if8<lmK
zyE3n4`+3g^9~{5f{oAopZ_AWSDHnM)zL2h~8s|ew16tD7e7A^i|5dj&v&8JDmPE=%
zKmRA$7gJ_MGb+CdUtL~zT<ju8)7t&3ljju_&Wh#;I_Z1xMfVQ=IBjG9vrpe#4`PVy
z`~0T((&m-#GMe^U96Bi#wz2Zedi7t2|9p;(?WqWr`FrQdb*ptOg`VQfW!Gho?|&W@
z9QSvY*%aZe;X3TeT^%!w>IGj-jE>lBZBn#cQs#QkkLdy4tJhfSDwnV@)HC1O`X}%a
zW7+O-ZO$Bab_x6EKd;+A{A*|3J#E^<DMxod)w^NemNBJWq+E`DV|BL6o9M~Y&cFO8
zdiyZ<?mwsZl%DFT|N2_odCQ?z-zLM$nMX?|<aH#O=^R?6aNhLC!i9@+_V+Hmw<jm=
z&es!b)^p!Es&_*_T<^ftskgIMe2aUw+h2O_w>Pq<dLB$#-e33idHu~F`PRO_eI_qI
zbM6xx`+tVD(V7vnGphWiKRB)Z+OhDN--?^cT1JPjFK62HwVvO7dH<bgA9kmMQwqPF
zmyde!hxd{Bj!hoxLZeb7wD`GSPQL$P-KuBzw2v~C-t$*1Yz$diXi_mbEBMwsF}s`w
zAs(HDC;m9Baz4ryX;+~fDZcL2yFHc5*3R8H@z2)y#4Y>>!uw)<_GeCD=WqT#r64O-
zDyY2B#X_w#sG=?Gi3soQjAL5czdczy-^fkfCO1!e_U8>DGZ`eF+iD#;kQ#5|Ch2+L
zIK%w<532)P$~>~z_Uzi2*jjet^{k95oGizl$R~%c`EosB!I87`JcB2muYdmaSLp5N
zYK_IJYtK)gGby6E@wW2yefIoOwu@YgTGw9cpR;Oz{yjP4A0fgvn@=2IllkJ<z`nI?
zR*Hdp-i*zik<$w;7*_q|_y2t5%Bu-(E}IT;8$WRqxA`XVu73Kp{tTJrY!fZdU70fF
z;pW?)S8sedy~y#ou+(}7$(?@~MDMd~b>r4uB^Ee$Y1RDTw}1Xk6<vEMHs++ZT5?Ro
zT@jhjKmOWkRy#bkysgi7=7Yk#kj^`K9K{pAe)*HB_~Gc+yT_C#9^ooTsp&m_qbfXZ
zyN0<61IID$ZTF&WH;46{w7axNi+?Hu7yJI7{~2CiDwX&dICptP+}X;^jT#ck2RZDI
zZPEX-SKWW}kte784Ax(m@4#y9{Zl!UdGU1XZr0!Xpa1zCv$Z#MS9bT&7&QmIs`Nkq
z8GfxS-muBY<gJkQvC2F~j|0C~?5_B->et$-tD-%=eehUV5oYtWe*tLV-PNYwrv1&D
z7JaVeEsn2h|LzV6l3X@1$Z2WF$4`?dzkZh)Q)aU~{=?I^hn7rN`Ez=L>Kpsxw{{-+
z5YqLZ;YH5I2YKn|6_UTa|L0oGwlGO;?``cpYwBJJpZu;+*q@hW>+REZ*RHarVTnxR
zoXM|WWktxSHD~9hcf7MXz4O4&@+R>Qvac_e+&s}|aan`oWskw1A8RB_(vN*j>*P3)
z|M<^qZ<p#^%V{^yPrJ9~XWrhIQHOV4QoT}`d}iv^kDsRf?0<RuWmw(i`BP8Z)ixXr
zE-RPuld>t3f9$(8wDA*X;)}x)t6VmFpX#l9kb2(9@=w+k=F<nm&;Rz(*7N>xLQzcB
z^4I6U%FBl83iFQ7)0lEe<t6Kd;Lj%~-nR+uc%Lj^F0szfSo!Zag)6Jh%&(ZgF3WbY
z@WgA<9rGn`)pIquH}8LWwAknsF9*x6$1hydeb>%6Px~&tOQvMa^Uj`f<5D5>ui4$I
z9Y5ccRZUadvDfR>!P|x&3#)7P)?2lGdiZpHwz2U1;I8$xse-ANJomOfzjoxe>!J&%
zf{JdO$XqO5tvq#h;fv~!=|>-=MJ!CaXOZu9>GB(8ouh}Qy;w9?{N3sIxpyAR7RKai
z$9`2^Z0|EGv`}P0r{SDSuf(>S`vlg7ofply^(g7~f%8>$pFSm8*3De#c5y?=M!$FG
zD&8i$9{Va=n>F*^x_*nFQ;J-(w%)%bxv=a;_{vq6Z>NUNXli)!c*mNx;^r2UpNV~c
z{gi)NSM0X6P3*EOR)wmu&9*U&*}!$<{9V~qNB7KEZhazrV$EC2xJt<@KX0(Jl>NH0
zxw2!^8oypEy?1-KcEs+Uy6VcL$4~4v^(TM#^<E|7v8cK<;b>iIUSY()+AAB}jTiQ|
z>rTA3?!deM414A|Zu9GY=@NalTqw2UaoWbay?as~Ut6_hna8&0I$C9WQiHkXPhKmy
z>9o=`hT@Z=McLVXZ)Tkf{cZN9@tkJoGtZJ8!N={R^1hU&D%8GVzO?S7WSj8QeQx&_
zZ<YIe(Uxu5B@N$Vt(bfchV#nT)|h<sShDZw!IMmz?^j2^&k<-}R=K?W?@e_@iQ?l?
z2X|^(@0BRd%D8o2GBqW3@s;4TofD4y^I)I1Y?bJz#Sv44qUZbynSAU?YlqN+l!>8|
z))qZVGoLz{y=6#wwsnuBfl;ip+#9hw>`~8S-zh67_0PVxt;S|;dZEZ~1|gy4%uCCk
zPA@9goBO`3v|nCbr9U|0y63k}uH|?7CwZQ?e;ifrclpoiLth_!UF9v<XD0cgTqb<E
za!IO7@3z1`B^$<jbCNUFlAdnMWoRhe-T!ZI)aFOLXZCB)T~MxWTH2GbVAJC{U(Pzk
zdix%|D<R~dxJz@#j3bpCo(k!&gEFtJ=rvT&GCw`P{&Uxbls0!A!_Qx{7Q8zt%(O@B
z!FQIfdE435zAMkOeEXpG&O8o<Qb|enuWRlL#O`cR37YX?f%iEht!o=sZ(R~}%tKEk
z<(E+9(c+qWw>$O94bD|O?EKHbH-*XaVy@}uob91EmKU}8d8|8ps4P?FU@ZUH&oz4(
z?t8Vnbl&?p@uQ&jE<20kTT5<*t~y@%`K``jn}odo3{u=yyk;!yPnjEUgswjQaruFF
zFDxWqusSk#Y|p!8`|;qq<fFIdXiIb#q!srq7F~2<T3UAa_o_Xs)dg;Ao(N-p{`|&A
z&(ch$o8~hF%$(EgJicr3zw&il9Hltf?BkgQg)?PZMKAkzD1`}Vd+jg1&HJNgmaycp
z7sbA*YjXQ;h4i-TD`-Ef-`Z0cw|&WdlWDI%tqhYfw5bz)E`DqJbd&swpRC*Ujz0dJ
z^+le?U-XK%ww^6#&iv0`R_|N%!uio%%dNa-5jV{>gFiL*|J(jNUh95DjQd*O+@066
zWj5q!$VdH_4%^oGcD{r{Pl^1seQ6(;O?>)emP-uB%*H?U(?748ylJOKv-9TGMH%}%
zBlyyC1m^pE{}s09+p=q-UAE6<*1cnpJ!#9Z^zxcbLfXZj9|&3VJJnkMt$p?PP4SJB
z0y`&o8~4v>+|7J{`vYz3O((b9^qIgKBc4#U%<ALSy~ckwepz)kM`1_AveRDDWfzS9
zGrW>tx%!vv8L_vz2g9bGPPZ>%{%X2x<N9goTzt>Y{b$(S_+<f?z&&T~xyRfS+}@};
zedqG`|IZMc5&g<@f93gGPH!VF<bC@5`d@L#^CnKau!+-5a{mPtaF?CmYbCL`y=H>h
zOMh2}CslJw|M+R_kPf`dbXGd`bnPec*1#M88RozAz2Y#>cEiN{oOZo4W$(TO_T6vG
zno?}JwRgI;<In8|GJW^HJc#q`nIzVy^1HE((Qe<CBfk0hQ_p<b*~@3(ckpNZJbS6G
zUoP%~VjkC$x4P7b$^HG$;D4?y_tT??+l_kW%wgbPu;%xUHNJ<vZiWOlh0mGqRKH-+
zgY4*23cX>{(^RcA9@g`G{dKK(Ywz7F6Hiten<%6wJPi3Y`ObC4bREX&IYu@uk4%-*
z<@P_Hb*DCUl0nbIK#SVtE5Dpi^m@V9eWpwBY4nPkz6Fi_A6A?Fo+nh5@<7ObTGe&y
z<yi$MeC8MOOLrWw_m5v&GQsQ8kqE^b?kqj&pBFtaT)U{3r&A*RP35uLTF+&XFV|Na
zoh@5sR(nLb@m<y%9hVdVcG;h|gVsK~a>(Ac->aM<R#0Vzq5brzrK^6F@ZXDS^>k;>
z;(s31^zu-{rqtW}vYr$x_io-G?>Oi3jy^84y<2w*#vI=w_;0I3N}cTMm7Df$k2OEF
zdydhUAhtXI84j#8)%MW4bM*H<cgbVLk-4d+TT%^99e7+Q8SGpZ?sDpdb8XIirJv2R
zn<Avb%LIOIsJS6u@F2`)qR(cdZCuY6eY4xLM!l3lhV|?FxRQ^4ZzB$#{bKU?icrLV
zhO4`FeU*F?n!VQf*^M(Nk5v_hCQVF`zWq1xZOA;=ox5}nKV@gzsoqnz`OdV{yu~`_
zJ2r2>zBTR6ohc`DcOEW0W*D{ESt>7L$-|w$!eXDU(QZmn?R=O$uQ)SyuSZ7wmiITC
z9P=No7Cb2IyK!B}1atGXA-~NYDLYSn<?3^?U1wvfPwt(Jo~5+`mo}F9Z&sA^E9?=T
zH+9~|XZ3fyj<#DWuMOH$yPb{W{K=lp%txc6CqB_~Ic)LZ!mH&w&BL$Ed${RLkLB%2
z4(|gayn5S|pD~NZJ^Om}(#n&EJ~6!ZF3nXDe19;kt~+GO-6vgFmR8E1+;d;fOt`DW
zC}v4xD%Xm8%hyJ=J$&5LDf017SGd@xBTMF(M=73|dLtuk({AGunQtaxoxysNP7`|q
zJ+H(Hnyc-f_BBpojfkVv{Fx0;F1-{x(Uexiv&k^X*yFqKnO2Wee2%hNj`_mTZcn#8
zc*679%T)GM)vFWhc&<%4*053MhF0aH88=n8AE^s_@O)`>TYC1g@X!+FwpE`d&CJ(%
z@VqQ)c6N2w-8O~3^f<1`ccgmPJ=*TPd-CoKlPeuI^{@Xd?lZpEX|mtg@6(>6Y4>+b
zD9Sdi%v0Zcv^}^>DQUK(?o^d8YuWeral7c0OWM6(n&iFU=l(sC4~mvton33OW2@)g
zhR#ik)|!i^h^6j5dUdY!lNUGNM)@^)TQp5LYWE>3<=R@FewD!1o}OLviZVY|x^C(?
z<@KrIL)Mkbk_m59R~+3o?WvK``u=H4kL=4j_9{6hVa`_*)92Gf+3W(Z+pG<}!m?e#
zou_4U&+2pulPJURzQ-oIVfreoC%m&f9vmw+KbPlLaAo0)^RH(;GMSy-A+hJ2%`@xh
zeA5^A<a?vePFZ=*dj7u+>%H}Jt;M;QC-u!&D7F4n&C?U|=~$DR$CV-@U&p7%^J`x2
zp0~TU{@0b)r`uv$BrQ&ezVNuybv68SoJGH~`nB$z%F#*uy8X{X-oynhO*x=$!@t&N
zUBBer?N9&JuGVYTzOA(;^@#;n-Mi|lFum0^y3aHfxg_4p?X{|p+Fq!1bjz{w97~1z
z`BBfdX4Y@Xzkc!`-x{}qS}kc!fmh+4Wy!A2J7?rhnE&}`)V9k{rb%40X_yfsIPWWe
z`&;E5*Cd_(z0YKsc;We<f2B7gqpi-$U3<TEW!5Q;+V+IDhm*72DpO?S?qAyS=Ix@X
zEiIaP51thN)pkD-;dfnLyU2avvN$bw#z{g;9xr$uHep@Rm2?B;)r#u!`jYFMPh>us
zIz6l1!ul9nc78-~amSvME)6MDEadx_rrthc7d!FP(Yhe6Ir4?h$!>k^c^7K73$vbQ
zPEKKbJ2Bi;S#;OreC{*u1yyx5Vbv}*n^wiCGfVs5&8pcRKH-+7<_m+h=QJ0j>Hg8J
zEs)X>IeUYZUB0(ePPRK})9t<GPs-xhU!E<9nzm6#s^RvxCAEj&tBJP1$i7sgyZX$x
z6Dkp>lPztt)~#P59W;BVf?Y|Z#H+K7Kf`+~F2<T(`*cm?Fi&MOL*vi-udi>FO3%*U
zY<#BYsasu)Or7Y1dun^yen(7k`Yv!=KFF_L?rPADsXMbhx7eqrZLRcLU8>$xTettY
zX<2o~!PqySa-ZlGpWi9N|2*VNz457xmQSvEi`CgDzkeOI$(!fe;WzV=t!fsl2(VvY
zni6o^eTK);Nv2Jvj3G4(JHE(YTXXx))G$t^%_nYJ_n+DywP@Pwykp+Gr%&8CQLf_r
zQpQCSB;FqEGkrSKuk!oXUtHx(hdfT19TSVRKB=7R`u(V<vWex9iGS`N+rct_{noJ0
zf=<@cI4_9B`TzXIbv8qB){iH~mbQ_d$}e3H-0pprEa54BI<D2=Ox@1^41(_8wr-h}
zJL}~<$3GlJjK7|(yuHYy%<Xq#@bANY{#WKT%`diHbwYVt=iH`;HG2~3VuE*chjZtP
zq*pOqsyVvTzVSc9*K4Lvrt{0)`17o@$W7(ORFipi2Zc{=*zmx(-tNYl-ODm!IxHVQ
zOOHMNVc|xRJh=~ddHa%CE7P9*XYkW@yC3xFrofr=TX=Uc{A;#fn-~{&$9P4+nLX)O
zs^?wZ{;<Vue$6E2?aPE0uFOeKUKIWFlky$2XEGhf)n6Uk<y`hbVD-j`@8{~Zj&ARZ
zEV#fD*poW>*}syjT_qD|^X}eloWOsr`$pQ4?@#A*@7(cZPGjh_w{ef{5(2DQRQ}07
z4*c8sv|lJ@o#Ba?C;u7duMJXgK9Vb=<Hwl({L}I>c3+-Hi5+)-Z?sq$^;I{d-(02o
zbJyX^_tTzN)|_2D>5b)|)yhYN`*rs0p40qEt7!rM%H;)CnWi&U^YtvA9F#F`W&V?0
z?pVJq>SuU0PrK!f+wJk^_VnEkm>1q>nD^oNlIA&o8P^>?Jn?*H!KCA=Z!R%U+9Oc6
z=c<}o-s`HEvh&ZDd}7#t=yarSg=I>=vIN8R;##h_n@<ki-Z;;-^4n^u=rb?d{Qgw^
zzCO2W$z#2YIa9^;j%sp>_q{zJ>%P?Oc0gMVYhK~K)oU*K99i7P{e9!Lxn;bEI<1l$
zw2x&jyE(aBWp`Ieyq)l#l3yV!^UkbPUU}#AId`w<)Cn7p-1zAhH1$@Z;_-8PH)WSS
zuD-EVe7B}_W>}KQbRK?%$D+<5TXs#lT6f;M@SJV7pY)1pP7zacKJa_BJ$mp?i|^?x
z4%y6DjhmC+T<sF`cywE6rTcM}o+}%oy}B+<<=Od|U3B5|LyamCUGF3mu122bWSS6W
zW)sw-H0hPQmcpl}liqml=GZf-I&`jcz~^OgZtf16Jty*{L<(EKzYe_1@%EX+%3W?2
zcVBw>`C6>)4zCi+Ps?3+q{3;3<<}K&C-JLTKalydV!hzSF!#Gkd}S+N2cAowX8KdI
z@U(`^qg9$G?=p4gJ3EX0XE-=PW6i~?6NWPJ`Z;|WYjdB(Z+4q{_@}mJ$jLYd*{mto
zi*~tubSpP}uDmoS#-+)<^r7Pexg9GXOs`$-Hc{EEa88)aq{&;4o-UR%+wCK}yHVzM
zu!^%LPx%V_O8(_*qq9#>w4eU1;3D%$(eyLf)9*f=s*rn2>!;P`P`)*bb@$IJn;@OQ
zl2EpC#_ZIIx(8AUtW(y;hI-`wP%n7>Wu?l_rrdo_xjDIchvr1xo%pKw#dnWQ0+Ti#
zvk+al?3t>47K5lHOW9h9=j>I7t=>6))_5Q2Tl{R*u>{Tib0%KvI3#_iup@f6&-=ae
zPb)82c49}pVZc;oN!hbTW)+JQ)lcnZ4&9+N&0xVMPwCzR&6nnsUhuuMUL~t8W2(g4
z)6Ai=eGk8t+&vb<ci(G&QJ~0%+?M`b)mJ0+6IuhspPmo0zOr3Hz$Ij6vetR-Ct?Sm
zTrYjNyCW`4`nGxlf7PR?jGI#=MJn4_CZEl?ad+v~BRaBO3lFo}u3pb<%6a$Asn*HX
zwLurlW(vQbcq2^b(i%bD#ghyt|9M@R9zDaHi79DQ-on_mjpu8wZ#OR1yBN9H+AX?J
zr_kf{)^*NL)<2)N+AikAQM)&lk29W4bNp7C^qgT1pVc+jm;<hENt<{!sBR3={c^Qz
zE`##fhuKL6y((+it|f>hPo7kN{%Lq(r?=8F=1GQq2c*s&-Z^{mOQV-vn<h`=F;CxD
z^=$i-?rh1$G6{KsSCaRdE{k~gByqZ1>}-FZ=0=z4ZR)W%xIB5^&i_=ewOaAco=uC7
zMLBRYH=p;fI=y>`$1PXMyc?^&n>*e<)5Cgex4&)Hmd)HtxMuLq<F7wA_n_du*K<S;
z&71$^@}sR~J*{_TnkQ6U-JNapvF9C&oJ9FbHBTYs&l677YKkP@mA(=4^LA``>9Hc)
z!W;igorC85oiBK@{mTu*c?X{SXZZTMa%u5N#rB0SYFAJF$bH<x&@HuN+CQ6##}j=y
zo<G>y9C&r#*;SlMJVj<r_FC^tWVOEgrbb8dp55`{XxjtZ{|tXi4@AG*=&h4)I7|GT
z=ya93%13*R4|D3Sh<G~Z4F6sAg@5Y*mgpY1w&!!)mG(}R$5Yv#$v3PQ|5#$^U#^nP
z>9Ab-+vA^K*T;WUnS3xWPRVYSrRZaG?&88#Cns^L7u+$N^Vc}kz^!j$#)U;XpBQ?K
zR<WypIGg|Sq@~f@h{Zjh+Y~0p%vls~rC4ruw%t1UPQ~}5@ja)v?EEamd@Qu=`A3^;
zsp6)q+iq^UZ**qQM#t9R_ph>Aw;s8@_};lsq6f}7$WQ;m{My_2*qMn@hk|7`PgD7~
zb+P33w!1>MpLWaa{IsjCX-3<Ll_K{0eHgD^-D&SQ=W2BKi&MuA8A~<ye4bO?bK%7?
zx6iuurLQ90>d!w6U8gtyw8plxB3^Yl%PJN)e%s6~(k8j-xx!S5Kfdm|drm&xb(i7E
zr={QiiSm5E_BvPnJ?EEz`_DzaEPEQ~sq|LKh^ZcB$Z|R<vYSOO<($OiJ^w^Cry0HF
z@U(txZ&uH>I=^lb|6SI|>&=rl{Ac+3GGF!1!h1m~#~f^W`j;{pmwK;PDslaEMf2sB
zaQB6eJd&pwE!=qh!gcX=_f*dHoRo{66!Q7cj<vo_FCR6Qr8<}BPCkEWvXrgZz2|Q}
zxlP^{xl{7@{?_cZX9|}+oABWC_wX-|c4|i-pDaD!x_(#X->hHvS8rPPK%(#QTF%Cu
zs+V>Y1Q>n2b@FULW|->@2A}^7Vov{Nt`ZkEjM>t)uxh2^ew~svPj^mT7S^nB^WpZq
z6Uz(drZ%~6S$9c%-7{HxuZ3$mW=qRP%QQ)=|7Te5x2aMsAZ1FEzm~{n@6|3Td<#;~
zGs_5mS@fXz+9yYrrp!G36xAKf6WsS#edT2li&)yFqsZ{Fp<ez!Luh5~Zqv4<GDSvf
zjCq$BD$lREzM>+oB4umnoz5+fr~PM;+R-_8>YQb+2g47@Y~~5m?|hRqN%u9+B#Y-c
z8N9hS&h_l@k*OEYn)OQGOiY4ZdBS0ZtA1LAn~f*!+W7dh%f>Z3`%kibo_H>6bBs>C
zI(y%vUHqaERy$QMSl3>SV_cX0rrbDl_cy^pNeh)y(a#aHOqI<~6z$m8dfRnc<>|wG
zjB$TgNiIpB{N#b=Va98Vo1)LFSXs;NG2jl2Icaov)uJZ}LRXFEbshdJt3Fw+U{h)Z
zi>^mb%0#8I=~*sD46=1wvm|YfaB#V?2j81({O0!4rdJhNx9@D8tbcdwiM4OVGCbxm
z-8f-7({$;n(-RIZbbftl>-5vJls}8)-uR;>-LNk*;^GYErkhJOjvo6cGAT8@N!cXL
zi2wV(t$Z6f3>Q6VKgY0iYm(Z{4Q+3pt#K^*xIx%OBJWs4kv4~^%4OpV3zvI+nBMeI
zy6*n;KI34`sYOh@cZ*+#KW<#F$62n?annM+qAFNViX*a1X1WElt#-z`kgYO#p%(vC
zgQ`EeI;5wlmOW?Rv1q>KqO>Cuo3}jJkW+YX%iFSOcS#Pec|Tu<D9v*UPp~cAx@%&v
z%=TNxEe?CF-u()_{5`drB{jA1x;~GiL36dthb-$wLRaPPntRl0=^7@@-Cw!AB%Q%<
z<<p`~Q;s?De7x0~)jjho%bl%8E#kqS=Bry;n)<!E_NZcisixD~w&PV-W<M`@bL4re
zET87b+s8z2nJauroYl^8>+n-I+cV_>&Vo(%izjcLZ)^M7&3)%@i#rQ8^=PVXnPzJC
z(@B(Ff%(eSNVTb<(@*d=PHet3=Sy#EU1+k|OFgm6)n3dNFZ(Y|l2jK{&ivVUAnl*j
z#Z9~U<*xRolv=lCd*&USWoWegy<Pn7L$hXRY%SwFbX?hF`}FA>7YT`(9=4EN`L1Ml
zOKR_bhKik!94y!M{d=ZeHnBkPfV7|06}ifZi&_3Gf1kY6(n)G}miei<Jsu^NJgsTg
zZEsed<6OICW5wGJ*0S`zO3@^xOAISNeP{VRtEOLSMSJ^X>HeH+TP{Vlo0RFl*zPwm
zDLjbrz;lbW)3q+9S|;DywB<DGq>HC|*naM?T>Z`LtW?jt34F&J<DQzIWjWm~w_53#
z*{AAI>lGVzKee90+*g&}H9uhgZ~M^O?{9>Oe!VSH(lhsM(>Z>xpp4x@$C*}%2RC_e
z)mruZ3t4Luc02Y)>FYQ4%cD-r4iPw{=u%|HKcn=#|Hm~qOEaEt`Yg&lfoaov2dl@X
z;*QCSXQig}PV7Iee)f%V&wqxf^E<9wGMRq9_D)-Xd(`P!k0N;zPHbFM^Q8I8^ZwU?
zCnO)At2uGC;pL<S^8Yq2F}k^XS|tYq$CRgK^Cas}tevfLS4i}5@x+POHs|D)pK{W$
znKUQiKf^rjMVEEAYn*z>^vTdyyY=+9i2|GldDxvV%WdU-c=E}^>PPF{<&HgGXT)fm
zwe~Sfsv=*N=<03FZaNz}lL~GIU%&rpopWUIvl;U|W;XuxcvWk}Z+iWX@6#=P?Fos;
zH?E6%xbLFE#6vs6e%~%|U#_%#!3mR@lO%iB-mzAmJbz9u`<Hc(y*DIE8tgw|;BZp;
zPd5LF?YlOAcA3Q1GATr+`A68d=8{Q~yvyJ9E%>%vbcI>bJgX~tLWUyNf!(5aC)l6=
zyxwe*&ApYgvv#Myd-5-#>$Od6t;nvAJ-Ztsg!V5t?Oq#w$3OR5hj_IHi@}q^8^435
zJ!o|~JZTBrF>dDVU)Dw}OS?5)F8z$*grD&zW<Ad}RlQ=i`q0Ny3z(J^7~ZM6ws-%#
zrNNS!p>2GB3crXln=va3ZTt4Y?wq;jb-Np9#kmV!1>E(N{?AYvoG;&DxuV%^qRu1_
zHcP=~Pt{8^zSS(XWGFp<WmZDE#nwLmdw1>Jb!48~$9!33zFk}As3!-1yyo`#m)1zC
z&YheU{dfY`3CDk-u5Wrid-R=HKYu!p@%+#4eVp|yd)^kyT|Veya>al1!ZIG|6-fs6
zhdNI7S9uAh%QT-?{&mH0x399B-%}kwaf|Oq6?AT&QRVt^=j;`O6ZV0Dp4AfDrl~UN
z_Fmt9Y4vT5&NG=x66c(HiodMLGiupUu<FL+>$m&9G#tqExt@^vO<#dI)@*jcy`Y9S
z7oW2vPkr7T^Y`n*-aT)ZI&I@zUFH^cXQR+!W%Vod^Pf5k-r{UJ$CGu^?2d4sM4hP8
zojBHs+^a1lnO8F3+PIkKRR^DBL<8e1t?JBk5B^vL_iT7lr+jbOQ}NjWEL(hL-|1WO
zpg(VFx$mbZiJOAA?qHm4T{rdo>P>gwwY(LweDZ1EPT5ue8E$R6qiq;s-S$MedETE;
z#;Z{=Taq&@N*`!E{uQ?UmQ)5C^W?tbbK2e8cTUaY_Sn(0?W$Jc@yD^Ty@z+TDo!uw
zy2~8UmiTH{NzomXi;ivQ&qVU^KaLe|(T_4p2oDJPd+GM)pPz%yDK|#h-agV?puT^_
z+4S2g@!4$|A})@{-oKZ34(m)`nQ>{i2lu}BuE)JzoILeK{bh_>+taq|$!`pgX>Z(;
zf46sezt=ULX$h|{cxvTG)V(};@@!%FB%l1g%N-pDeM-KIzP!FO!Li5S`#Rwx*ESU$
zyr!^RorS@D=i8Wy6V-c89IMJoy1mmQZOfA_EjbDob2Fq_CaApdwDk?xFfZlGNA(k}
z9>Le`rY?#z&ES%KZ8Ygg|E}y4fkB@8+H9Vco?Ub8^S+P2_dM<^4ys&s>tXVYoS){6
zvt~WInYhE{vuDyx$tOWs0Y7h~rAnxz?@d2(_0|jRtl;dG91|w3l(anbRpiFHo-*t7
zG>#*W`}~*sng}!t_fJUXF+6FvHSFEF>fJ58N?RC`xsPg_Y?nRx=9+4YS99SVzT#OM
zS2~+Crx>z-uFE}D6<Qtbm~E2L6WMt{wl!kb?b2hPjo0xpyx@(=To-pZY$NNQd?D7|
zX`cNHAB(&+d2f}P5czGMsC!i3qf^Bi+WiJ)GLPn6zI$uavHnSxN$d<)B3*BJy;!nl
zo2LKMB})W~?-xAlntCl$Vt4MHst0LZM^|nyNl$j<I?*tV<9W|7uATWgeKU0UL_Y6Y
zX|rw9&t(F)9gn@6vE1Yk^Gs<2jq~hl%U+7FTUxSX;;o4#90g1H+}Mh=DuYy8XTI~)
zDLO3kSKIE{->`=+@&&zj=DBb6<(U|L`kBA*Zi`br1zN#fz6!pIi934wr&d3jliFF%
z!avFKe6r*1(lnoHA=jD-6)Hie^ObK~a&LadFh9!g+Sb!~w?tH~tlhKl+uD;#a~SsW
zI@_gA{~mNwkXgm>K;p|NP1Xf{##Re!?=V?aUwNkO`EE+Wwu{Y;T>;mu-iZad8-H7w
zv&_Bp*3wN*<)S<DSL(iDOZd1+J9Y6f{%fmVRh4Xt*FLKBGxC|y)c7|mQ>!Kk1g1Xx
zc`mbPSJulHGF?_*OrLN1cQN;ES;t;MgNi>H)totxlT{8*OxAI)z1tcZ-E{7x^r{6m
z(_Xun^L(y!a+_Sn_}J!Zv2*y9xpGTq<k{?*qW`@h<lC>1Z%-wboRWCbdQg_t;p3X!
z-x?(OimE1uJy7#Fev6Y~PP%Nxa`6i~`7ijNzRKk&+Bof);rIPg_olMwRXo}`mo>&G
z`fP4jk#TI;lAAq(=iIVyA3C|KYL>&vt3sO7pF17jz&trJ=;o2Na~@wQJ-Al7^H=YJ
z!s9mATrSV(nwI47{AIYNa!A_U%-_+!4@>07RdFrY=E;1b%UD`pW6tA0AER>K9htyg
z<f;BEbazB$T7a?QlgAC(Chs?#k+t`ml->C#MJ2hR>al6oWxq*k9rJfYv`f5xA1k^?
zF*T0w;rET<OSW@s{n@W6d3;&cCBdNMUQBNz)>-$It(~!!r>2Mb%I_-^nkKm&zRb4X
zXTk9urE3Eu!wq}qYs$#<6?PrlS2iK{$(+ysOgF9&Qsl9hI+QMK?H1B$d!8laM8@ge
zr`<oTRGB2VVO`F@tIwj{m0Sw<vs8sg7wpctIg{<_u}5t6=a(jRD)*%x+9va7_ujf4
zON{1CQVBL)d;8y)6}PpX7Drq8ZjQRO+v_){&E(_ur{-<@_T7;8#g5}~J-m7g+3s4o
zvOW#4WS-np`h8jByx2!Cgmg1zdUdKLH}_9%|MB~(tep3ZGaKg5^*nIm{6E_-EBAOZ
z-C5GN_-vCBhb&ir^3RVxmz8H(6z)GQuYbmB$9ehQbTz}gglW?24UMG)x3t8yt;@XZ
zy2I6Y`^-6iUN--{Qr><l$k|w1hH(yG{ratLx8^9l6v_xOi2vGBIN$kg5ZmeVIp1!2
zn41~hKJffyi0U3S+YOEy&sR#wpWD3eO6q5Z+(u3<h7<CClLNcA3T&2?@{wKY`SQob
z0x^%&&y^iZX5N#wt=)B{?D~dwzK`#er%9gt&tS1~(Y3mQO$!QEwpu*-EgzMzqp5o7
zO^KdO51L<pU0z_i#O!-nUG5IIb_?t2Q~xu(4!0~>pR!(SwTWZ9-_5lhIVoDnolPn4
zg6GYDaz!)OaF5^}y&p?|MouZo|FGmnSKy(%Kb>WB4B0J?)ooo}yO@2ObcgOAm9Gci
z)K*=4x^AlIE{*uzo3ox2@BY)ZwNk*-QhKqE8~c8lkUM%s-b)zsR&RNn=V^CrL((>Z
zIIoVQlM9&di@p+yX3Y8>B73Ur<AE3U=P&Kxbz{zY>K=Hp&^f{CV*4$L=XnKI&lmh>
zu%GJ9voB^7Q~P^4hsPh*{Fr?z(=lhd^>mx!uF8aaQS0w&o_OJTlq2=VzO){<-QI>d
zatlgdUs_tJH1kBwO63H_jdSYfT&-r@{nTFl?Cy1yGWAh=4xe?KuuAXomYx*m*FQrQ
zv~TWO_2Fp4;l$Z0dX?+ed+c`ny?OWPG<Eg`c2|$h*t7Qb#l>}v#uI9@vkw<9KmEDD
zlR@wm=TQ^mjU7L)ZLoOiEcnj$Wsa9hW&HGY4-P*y<(pr)J*SYh;<DmiciH;6zDeg!
zZ{3h4c|51@*evF)7sH?SxoS7DRfi{MPio%yeVx4PJ>APmAMX}l+AV9jsd@b}!>(&8
zo4jA#oxErpf8nghH7=J#W)><+bE-N@{HY4hEi!6<!uw8p*44z&;HjyPpUK)s&06*1
zl*cFQ(<c+>MtHs0`EYLi@~<oHx5piNS@U9=%gnMlzvRBIUUlR4aYb&9N)_9}<54+f
z(=ztOOk@i_zg~H3&7FlBo;RL9@qKaX!hQczr6qaBl7BqDuC^=t>Oa@3xyipcY_WGL
zS7M9yj)Pj(RZ>+(Mjclct=i^)`dsD?6Fcus^;4B1i)J4<Zu4w`n%ZQ+hBI!Dw>)?@
zQ^i;3)ky{0BCRbP7vgl*IQl$jef`BW=IxeQ7q&0Vw^sI9t#k9l^R03jGd6}=yz|de
z>F~{q*eLt$`NUA&&q^nMA2K|!sL1n%ezM=W*fpF73LII!i$+O!?w<Oz-ndw3m6d#!
zN6kUW&JgX=lgskNUI~R*@J>~}HhD_PPL<W!D$#qECaoz?=y`mrtLfGElm_NGJzp2C
zzjZk(!t(Kr<@L|HGei@$^@{l3Nd!Op?NgTZpCK*G^5@mQ!l?5bB?V{KTQaYRs(U(1
zjNz@$jRW6SJ(c!ONshj?=$6jp!<@T$cb|>vsp1Kq=KEriC$q<K#^R$<`#O)^ZJaCp
zjG=F;Uv{AC5g&y^zg8Yiw=mswBc{)4@=1}RHs71K>{}kBzYg-uFPpH)UE6&J56^S9
zx%`thS$ND0;CZJo!GEh4mnZwI#nP1lew@ET-OQ_c&Aut`=!=+UGN<A3d&Bp>^L`Z{
zc39WksAO&`w&c@kHNivg84`~<ta!F*?%M|ur;b(XH(&X7d}+|#8`sS5Y)&t>tdtRR
z)xJ30G|snIF>`9UK-rv+Uys@@n0AdJD))?=M5%pX(95!UMP?t(zN%=89(=Y^d&xH4
zS8LL}`ShhvyB|1s>Jx9w6cvL59~ZA)UEF6JY&W|wmuFRr)6(s1dA<fW&b{T`)zP7@
zt8Te2KW*yOwZ(ki8K#Gi+??3-HFy2-xT$8dD~=k)stC?qZ1dUMZjN5UZ<AN=rM5YX
zTS~ms+VnEI%<yD=sBPWWqL{vqj;Ai0Z~1+6XZ)Pic5fJ(EljT)S5C`flyq$FyLw>l
zk^<*vDt3Xtmz12Vn|l4l&)=G!{GpE}4*m*X-#oj@aINI*nm%dubrmXC!VQgXO=9!<
z+<0%ztM=0|eFa$@DV1xU)jzh^_$GSc+BQ>zxA8i^H>t=Q%C=@@ewV%}*%>pzz>;aB
z$_&S<kp9^Ve}x^LDEQ=YVe;-g<DP4Q8rS=ki#1gvH)(ycoMf6)l2a!-n`fuuEXS!f
zO_ruBIVG7lDx2<c*eYOM)^}Vqz5d<pvcfr6B~lJ;c*pB){7A_0^*r19b6OWwvmfsD
zV3@$}A9ZT`W~Zag;<qoWZ;k!!+{DsQvv|&xiHo+htYdq4eBqa+J1)*$;n*?v;s%b_
zuVc2Y<uH3@(&c;Nz#L2V>&n-5Yi|!}`Ps~ICtP(^)2_A`cPjl1KYiTN_f==nwJx!D
zCS^}I9`1X5Z`t)*5dk^Jl{1-F{%4Tdx@6U{*$x5HbC%2e)x5Ro*u|+6(s#T)@lGR}
zd*!pwPRGn|N~+!vd;GrdL4TCqo8)ZPz$p%PVa)G-%U{~{pFuU*!K``5*VmEaWlL|&
z6Zw4b<*c*ItmhQA*xlX4z;KT9!Ti9;a#rr7r$24@tgFJE%ejt<PkViQMx^~s+rlrl
zt7q-rmivCgG_Fd?CF0TsTuH|ad-B+qx_x^mvia@L4wdv3XPrX7=RLhVjd|hmpZ`L2
zwBA1P*5i<Me>sck$);#WnVBpJ3?W>1E|>gBjg(H7F?XuhI^FF*^=WD1r#!~&-51_S
zCiWiVeClz$ZcF}xbxB9lb;1RWHZyQad!Ex?)OVI6ER~_Lf4yn$+lxyj1Fw90|IgLa
z?&tvqCa-tjmN`#!b6RFpt<x-W?Us_G!a3E?61A598MtyUP1#;*Th3Eg8kc!|MsWE(
zmPxjpD{{h3r;6r0NjxVaI6waLpVxiY*X)ZlTfx3yUEiw*%jO+Y{_Ga&fBDbnkahDc
z5BmM7EAjjyl|C<KxwThY;eC_0v$uJj3Y|PBZOc|}v%Yt(9w*orr_VY!&CZ|wg@sJO
z%%x{#HH*sMADLZLe0O7%!Nl*8&KeIJ-xNlOT|O(R>aHwd`R8@iof8j)+^5ONe6jU4
zkgaBW*%?3ky82$DgHBTztmYn_&Qf@A<+_tD{1P^X^1-umCuo|*G%QG$zxUGhSZNji
z`{W(!3JuN)$#P%UgmZGs?ruL7TYb%Lnuy1_B`Y*`p1+W`J@?^%hKRMFS#y|QZOuq~
z-md5Oy!V_zaqP6&Ma$bBbUl8vSn$rop4A$M`Le5y$S$1NV9W9KS6GM7%WgJ5d#i(2
zE*Q(6ZL)eRe{Rdd>6&e8-e!gL2_0#Xzo)GJVY&LkITN4nQ#s`!X5`Xy+pWH(`kKkI
zXDT<9EaSTV+Dq+S<H>sF)1<UJtEVzNE`NSDdD)`wynwS$mCbna&(+O*A8+3IbjsYj
z6Hlw;MAskRKJBx@WJ~FOm7pURb)M|5eRg}Z<+<wXW``bGZz%{rk>oIW*Wc{GT-C}&
z^_KB-K3=K5RycdImv&O#S&4TFU11u>&jyJzd|0$}mfQ62C;TpODl@ZOOIw@QU7((R
zXOhetzdfP93r=hlOW_cIXCAgzS<}*D>mrlxd^c^+XK!vTGPSRn91<k?AYH4-_-EIp
zc`sF*dIjCiF>)9Nc=QSz+&FXgLf6jA!577@UTd!T?vS!WYxMyYxu$7p=?SY2F3z7_
zYGf=Nxnoy%g2JCA|J)XB?D^^Fav*WT(dm+F?p;f{`6jYDTazPg!ph@HU7hckto5gy
zZN4rN{H#4m$4D|Qa{Hnb;T;V|hVP?v&5o6OT$eb1`qrG)-#mF9Ox(|Sd4c>x-$@~p
zQYI-KU@A|ZbyHn*=elV-^X@%3x5BBRUA0lYr+ig!W#+erbsX&PHhpaVvf`aS^Noo@
zxjMIFR=j;1dheL#+HEThPc2tn^146EuV~Wu^3Q75O_jI1rcR7jc`lJ>6)L%nvA1`>
zO-f<#4T(khN{-6B3$N}rt6%H&pCRpY<Y`H--i1{Ozc1wlIY%uvyY%PCyH7obB~NR_
zEeTn>l2zql%zmxKtA3>|c9y;?Z+Kvm|ICn*#~asjPnx*=-Q(WW0>=EP8D5^|MMBwW
zp~*6n7N0mf`A1`T&b#H!-*e8ksXt47`!=%hZDqs3kn`++vb;;KPWl>jsebEir9USe
zSgHeMyz{QrD|?(>d%ApAbIiZD?Ds4fzSu|2_eg2^bmYPF=N7lS+_ru`8FnNuJpb~4
zhV>z7E{5KDpI0(J)#3Ba*8H?D&C>VI_Tqv?{~1zO>oF=@?@+kvzU#Nirl0kD(&P^q
zceP0@*xaip6lx^A{kbHM!)mb^J9oy<Xt-aqUM4I&@A{pfU02_E^R4AQy6doV%`5Zt
zOWRF$zqqG5Ni%S|&7YfVv`<$@s#hol7@j{>`0HY(@S$ZE3y$uPJofg;{A0T^!k$c>
zx7brPdGS@zcs{--Ha#V4s;kdy-jqCaWU=b%u&Xt<{Z%(eO9p}toU9Mto-yNr{IR{3
z+x9-O*m!G=^bC(?nfE4l?S6VX%;3G)@OqZwPIE)i$z1Gt&$~|T%HJHz6vDBKK}&ku
ziY>onRm`PrE=5m!{QAeWbF=kw7p2)BDvp_dtNd8c5}BUlW!lBA6J|TCZeYB9eQk*9
zieJ*HGaaAzJel~~ch>8t%mqIq3V(!3w<|6VN#mX~zj&V3lI>iV*e87Q_xHLhlx55*
z&Cb{!pOw4YCdee>ggWz$s~jmjFI*N+J)90&?9a7fbN!OkAK%Y8F1hJ3FH`(v0_T&*
z>_2>q#S7VY`>sh=-F%E=?b;yMC!fq~cFMli5|jzDGCz0VdH?gE8)v&E4)kcgzO1zP
z%JJD^68(l8ihJvSeYstEQREBvecM%YuO5?mGO_vkTI&~Tr#1=5q$GSl9@QP(e<rbU
z%Le6w7xm|(LZ(-;+axn9fB9F^vhOPM+~xM|OFh;VZ!c@hZ9MqXgU9gv&vV&Ywc4kI
zOlm*4Z%$YE&)|P(cUfa(<PMdd=C>isw%y-Z;E+6jsW)G8SG$+y{;brZ0^`dH{~5ln
zIyQZg&cr6SO3TOBcI_zn!*c5JpZ^SBS35~N3JZzw6bBbtO8lyuB3RaX;z{uf`x9$#
zTZFM3lzC$C`1T(yk8LWEl6N>|xQch}Ket5i+VWc(6P~y~ZNBJMRL|A)=<KrIkJnfH
zXRvP#p7oRUklWJ6x--_-rq21?qN1s_q)zVY9IchM9y?EG8PDH(R62#pGx=ZL)|Ttl
zlDFIrDk%M$SNvCd&BxF_$-E?quqThV7Hk)kX(~Rr)a<fZ)U57RJSjX{lQ_<^%RdQK
z*`1Qg@Lav^*XNKB0n3+{=B;p0zC5d4&00G7^fdSMlP6XgWj210x6q#0m7;gM-S%}5
z=heA_yONoIhHA1ci_Ux@I8AR&hM`U1#{*x!d|#WXWS6|@<EIbyQL{97CNemz71lJK
zmwBMJPkOnNk=D^Usmo4(?rWD}j23>mWI@6zi<4(l?%npyHhM5$cjtpIUsrFqY^IyH
zq)X-m+nt817hi@u^f`QbE_w28-}0aJ6Jz$S*7=xs{=^T<zr4A(5B-Y?)H8Sg+{7Ok
z*4B62H%Mk%UBL~zsTU4Q1g&8_%;%QPbo7ABWJdqsciMUPl0q2R`12NRamaF%P|i2?
z4!vyiMCVWA3Awhlw@%Kx@h9{9=9ccsKOW8A95w0U^Eide_bjd)6ye?S@59L}{mHN2
z#`J#OWV2hfSI$qe$UuAh$B>sY`Foz;Zjq@uwsZNESp9dG%yu$o<}Qp7sQx$gL9%7-
z;=sEyo3_45GCH-LS>wHM!Q%~UJIZeVoadu@KQZaa1hEAtCz%WM8wF0f&F9tg;M&ro
zXQFDGmfqo$n7_l#b6r;8l;ET=P2Dp^2i{Gb!^$$hKC8<i>h`(28g<Qm^N!6=%}W)0
zP&-fYbjqJQSAV;<X)HT($m7lwNn6|35y$i+t;H71GG6;`$t;u4hc<-0+W&lQywR~w
zk@=CLaXy^Nk_^8DBPKE?ueK68CRvv_|9Nz8v9&vo<6~C83g+cmnp^kue04T;SX(Id
z{!UBr_K8Q09zJTFcem`lcH8@Y|6_Ai_s=Z~vux@(nRFnvElWe^(dIhQ)Uetk&#gPZ
zuX&#O-k@5@xn#mh@9x?+e9XoS9*ee4I&|r)SY#ykW1jNd=PzB$+~%y=S;kvEC3HE*
zv+~ZlCwRE#O%3dJxxZuk%sCT3A2Z^=5_YDL)5%|4`d6N9{jaOuL7R8)3G2CgUWm_-
zJD2^V=$Z}hmaV<<qU6fl3!cRy(=Hz4Qf~jf_RjpO;yW&V8^o0Y_S^9}b)DMw;n&}m
z+jZY8B{!*--*LY@Z{OGKe}X@4_D_9!??$jDmnZYqR2`ey7i?!LODfOUZ1CyIL(e?T
zO&n)hPc!%jv<UcSII<M1DG<4J^Vo+S=daD$RrdR+kB0Bv;DyV-M!M<8I^SH<b6#`V
zRPRQK{%1P;Oj~SDS7%CoQ{M1ocb`3D!1<+j*KGVL-m)a+U6_OWWmC7$4%IuH?%p%j
z`};U4dDha_x$7F8Zi_wIG4tJx-EBNKc$&*>dworuo*epQ?XcoOZ%?>Nb=TVb<!hH+
z;+gGV`n*(p+s_HJQ)h@;avf%v_j5%>)x@j)H(#!-dZ%64Z5Y;+W_iwV-_osJlIHwI
zW>Xj^UN${@_rBMqj>WN03t~A{G`^Xpzg;(BOJC!^1co??+S=8UvECuxJ&W>PHvCK#
z&OQ_|$Ebh8tX8f=>fRX=|K96JKRZ{;oABb-(woy>c`g!kdyw1nVDgve_EEPd2Tjxy
zlDL?xV^?RvzC&r!%B9aQ%>6p8XQAoQD~4jOioaS44Cg=p8M-pxMKD_T^wg=-b=&Ul
zzF7F$H+l9N7Ne6V57<<5b#@qTz8__z_x8TMO7f*;zh1Sh`EG8q)-g3zF{MBASjzDy
zTF<uf{rGd_dAVcEwn=Yw=GFM@^ZU2v$GY=Xd$@O`E1SN4u&O#L*EFH==$+NkCwm&-
zzV<FTX_Fv4`Ml+qwYot!<J|99+WggKxxLBlhB~w9;$0W67nW>4d!n6zslwuHnc2oQ
zf-U`x^Rinnr{uqN;XM$1@*Haw3xnsD<#%5n4K-b0WZm{@Nw~Tezy0&is}B5J6Y)Uy
zhUJTjC08XE1qvKECjE6q&Ue#QYnBLKktw{iOYA_m!J##W3vM4{vzGU&`_#<NK7aZC
zA6KU|OWvt_c&v7n_0dmjPagQ`R*)sR$j#-l%B77HXU;NvQz7c~=$GjfvpXvv7c4ux
zapu!f=E7ITKZE!5?vWAAKeef7$I^vW+7m7|@#`4y2V~B;KWCD=k5r?ghQVU3s;(Uq
zWgDI^<q9|685wY7iRGQ0ud16bTwO4`?Cp~>)6|48A47>_c3brB9%OiKxo|hb(VhY}
zt1Eg!=F%N3bCPD)SLv?3yD!u0lZ&$o*YVFwry1%5Nphzwu-C6^nb)@S)%9&1lf$xG
zLL;pdUU@tSI$5^t&gnT!?53ygtNuRx#8+$8r*}&?L{>k38Nk1lU$KSFV@Ay;?Yyvm
zQ;&TVmC8Hv`?yz1#M&o`ig#XlTB>OL>RO*~<(;+Z_9^e4_OJ({$JtGzLig;+&bDyb
z_DSKs%5C>)^>PiNv#VWuqqJ8(oN&y%Qy{_IRJiEw8{hd(Ptq<u5={?LtxHYFa@Ev%
zv%Ex8P2~Abxv4RcN4iua?@rvs^KZskZ$ZI7ANL%cT5;m1hrDy3WJs7#)LO~ax;u7k
z&XDz=7XDNvS@e0;*2&T5jVz8?h389qmgik(^HJ(OWzBeNW!AUXE2HuQY`)4H$iBYR
zvv)~OCTEGw_tS6v{O*LVo)M+F!F_9Zs&)Uh1y_6${%VD=+dP}Kc!K|ht1FVfF60%@
znc@Fw<Ih#!j&&>al*>O1N!obl;KVMWiqu!UrDnJ4Z>>GF&-1HncE+;fc}KI8WnO=N
zTz22@s&~*v8Nth<DF=0XZ@X(fNLHBlVAh3-X(<7fryh6REK9QJj!4;_!*D=fw$?9u
zd53t$QW-({GyATn%@caC(t3M|`p>YrtqWormhSF9Hp^A2DrbpXSHbV;3jMF6u6@`3
zG;wEjLfM+%-lcuMr~j>OWB6j4#&cZpfk3pyD#@MMR&#_3YO8|3a&k0mN-ncd+nkUy
zr%b))>g|d9&P#VZo|lm$b>Y)`F>iw%6Q@00rxM(|yF%Xa`sY<*8_t++nq=H>(RO!&
z-<^`<8|BxQ_AYl!xOBRpV9kxweESZ+j?_76`(~0``&s*fD*m+<o4daC1x0P1?7i(|
zNp$ev6`xwTg{_n4JD)WV`t7&8?z=Thmj#!bS-)4xuib}N7#(WlyQln)z1TTU&?d;~
zIPZa5Rr0T0J>3GI#GZI|?TFdC>2fJKHk;<_T`=5IcGO*U)6auSJCtmmw=(=Qi8;c$
zW)k~tWu65r%#&Y-u6d>?B$;;RiGW(1gKguZovLRh?X)?cR>CWj{dCEsCq`{7`gT*}
z13Jz={aLhX-tFm=kFQOS+Is1P-Q}iNPILY<h-P{+hGmL!@jqt&*|n=UG}u>U>4nlY
znwg5~$#(@m<<`_3NoT*c@}zO<Vh*Lb4X5N7EnjK7ozC(*`Y-KP;pFo^1-060-qZJH
z#JRTD=pR((IUrZJHN0iI)@RKfvnISW@|4=OTOo0d)trlA>-6^=XD_r#mOQ4hwswkG
zr|_AM+_y_6Ii2DNVCYF-XEi@-!}LHA!&s5+uNcd2@BX;vt=*-s={GAs9XT$q_~cjU
z>u!;0zMD6T7S7j^J#07U%i7S$xi;q)W>@7{^POMEdDJ&4ZNgcZF0Up@kDT;t8(j?l
zGh`Tf=k1>IXF-#J;p?8z75Q`DM_qpN*`<}U%kk9n*RS9D-}ZfV)8^dcuRH6GTI;Mc
z=bLZ-^v`Q+_w4OW<rO#5oJG=CepP=NHJjUs_vQBN2IEtoJYIHn$wn;pdbKLjzAU+>
zr^)j1>&R1!x}QyJ)Ku|ae(18zzv4~KQcurS&Dve4lA|nme1>%6hN6AXHr{)qGihUw
z#<5F1MfS{Ly6=^CrQdG(beYG`@chsBq5jhOWxu*_Z9BPWcf>=3Bg?DJb=XyEv(~)2
zbn^H4De`g8mTcQ)_&)N}sTGk%h9_URX_e(34e{na)4MyvdDhn0lf_0mZ(TVTt9QiJ
z<UsBY$KB1}uk=mLcv-eBV9UCLH`en=+g3}R**!00@6Do+W6FvRuY-<0_gY=}%<bDt
zg*y(%9ArJ$#Fhr-s-$bJy|Y<H(A{W@!SXYEn*XSqs%$v)(JndSpuF+=kMe878fGq+
zKJBrtAktmJ)-W*Uq=@xeQT5*zx%W#p^~X8>)rtw~&hWoiuExt_&Z%Wx+0J#Os%wWx
zjJxi>MWT+m2O4{d-^V=NBF???=h-J6N{LEd_8sLxdsnA!^$sjlS+nHGeFG!23!HD*
z_iC(*RnPU@w)lP6rMru=%an`%K0hw8Hal|in-_^sdD?ZVKmJ+WyK86P-*^G5s*e+_
z9`tD!A1>!tDURAdeV;*3napFGx|*<`HE(0rHOGtg#ySa9zRQ;KIB;^^v>(j!UI%6t
zonf!jzB=p53y&wI2FXRIZ#?08`$xDa@5|XqcBx@M`!~M)SF1hyYVNVi$8BBXf0Qf6
zI#rx}(0lyOmEcQd(os_T6nak9Fl}QqaI1X%pW*sSlN~8q25t-t>975-x&3v#obSN(
zcwOb{TmC1x^bDJf=lg_vi}+eR3|KtDf8nZKOSHaiQhaf`Z|2eXhdV4zta#pHw9)c_
zD9=S1>7!3pggu_amlZlO)ImX}$aYn1VzT#9&reGBN9W|)Ka4E4*6G#sH_&HomI<F2
zE%fs2;#o`Xo}T1mZa&FbL%Hb6uiUqL-Xu+&uacZ0Z(w@f>x>>t*^`IURYmfSKMtL9
zlDT1~xrXza#%=RYU)x!?JJ3IuBRD&F`z(upX9|uQ?QZiA3|&0su=Jij@j063E!I`M
z4maA<7TRLcCFm@ZBmX$;Q3Yejoaw>S_|7)TZ_WR@H1LI8xmwn79l@8Iwj8kDc)Q`?
zrrKAVmV|wKu*c?k8Jl7A<&UeX(q=Jj*qSapXa2q9iSzxw>h!&x{7ma-gqHL|+t&2A
zUwmr5skU*O&-6Px>4Xt?J^!WImCJ7X=Wp;zcFn!7ovZL$F!15?9h0+<s@&#Kw%L9m
z=;ZPxRTZC?O|i^b)xU>*ZK$P}=ay|UX~(thL@M@OS#s*evCp%fe-+ywy<}FlT0c8$
zd0Sldvt1kyc1_c`s@7k5&S1{G8>XhuW)~_+r0%o6E<Kg~lx=VPrqV@MdM!h?Yezov
zZsN}E4VH+1-k&wIcT2&}62s^33?yIQU*@pl@%CNYCmOfk;d=ifYI&a~i^M$Rod?d$
z<(IX1{%NUHP?V?ODVLO~{S&y?B&6^pm+BO}+3B>q%{uSv>-nd)D7HHEPOqP}+*1AN
z%cH@wI4>N0c0z5+<Q)r+J$&aI^=#Q}v6UiQ?<%J8+wM5cJT36nPR)f%PFCTIww~T8
zI4$(xf!huI>#~I-E{EPYEA-&2_TTJOiz`3eIpd2befk*o$a#Ce7msT+``XCu$DP9_
zuF|<FrnS3K=6d$S&(kF(S;DOov*ei<uDzX_?X>2uRg-pwc2<z-AMd3{O=s=a{MlyC
zI7jNqf{71iI4d5z!;pP(A#=bb38sfvQp+AZ+aY~B-LuP@hpBjHfv?t8QM1{v(zTZv
zpXB-d^XVqOs<64cTbnZ#Jl1e+-<o0+6LF%+fhUDc?&?Yjr;`(>Cihugn^=CM+;l;~
zirv@SS<I{jvZ4#5dE9-jR;S+C9k}^zxZ?K21M-@yH4PpstWFXBd2NsO^P9U?zI`3@
zpJ7JQ^pq}lo8M9|Oe)J|m-OD%3VZs+=5$>4WYaT6r(TQr?oqqoT+3U0Kk=&S7P;ug
zLhbfD?sgF`|D3Ig?_S_msXR?phyRqn{HdMM9rvU~Hyx7rbMdK6o|cE@^ZUQTXU5G*
zRy!xD>*toOy|98=+jUM`jZoRI>ab&T6t``U&{Hoow4S<+Cu!TGHqnifEEd?M2)+wf
z=d1NJb-T)RKl*gZzQQn;Z5RAyuG_1p@0jeJ6kVM*?agtmJu_c<1lYB&T%o`AOx8-?
zB|W=$pKkn_v@C6>uT$h&y;OfIr?Q2*J0>4E>i2zKN}~Sv_s6(3CE4fk%b)6uI>$Tx
zY^_c}`IKkcTe*T+%nIBuUt2oQ$9S{n_lhu?I*acn)BZE0h3l)FH{HR)vo89D{jaYg
z+1$sv&Tp{z=_&J|<eavuwA%GxqdV6HJ#>rWY!|#Ml~kGkI#f?-b6dtX#jxV}dydb4
z8MrG(Vq4^))c%DzZ=c_q|3o4;<nFFx;(CfFPn$cQkO_|Y<n5RjJ@4fG(@GVKf3lxC
zFv(s^TUKvz>Cbh`S^e%voK6-AoAdgo^Ub-dY<lw&=iM)`l{|N8Pe|6GOLsQ37aQDa
zb$i^^_9$g0OQ+rs0nPN_>*@<cZL<0$la44Vr_Wfp!FF|4?)L1thHqxv6xmXEYNt@!
zLDs5%|D};FM=mI7dn_+2XP#I4YR9hj^I|XM(t|2w*{r`ViC(*1M^DUS3h$XPqXVD?
z-@nba_BNhe>zSu>bLnTt`Bif?WI4aD`Tg4S!ph$UG26T*6&2r~rycxuU+JE;%qu&&
z{yf|`@oZ9-cvqC}N~LKw200UdefiHIw`EmuV0ctl(1ny;Wj`L@QB4i9+Ge|RrQK<r
ztl3Hxr}tf!W4m~~_}<)aH;!&@Rh@dycK@%d)qYcARZ_2d^g1T+#}~i6a7}C0<nMvA
z?leierJp@qV014?pMh_0)*}Uvl%k0Wj}zFReSfspSBAerQuXh>Gwb)cTdw)J_Fm*h
znG;E}+mqG*{0iH5e)`g5U#=XvV*1z0x_#TudaWeUq)g>I4CmGR7W}><zH834_fK?A
zeGoi&?9SumqMJ?o`c+y}d)4PNMl+ZD*WGW8*wJP=$5o3pE>b`(rhZH2j?8!OIMq@!
zzlRIlDVC`0dV1a=<Uhl_tu@hp$~Sb62PjFNGcue|KR0amr9Is@{v4Y)$?ZmpN{Pb!
z!&~Dz?T*Pfa-3$6E>79><x$J_XD>q6Z=Crp=H5@^_}Q}y%HFtgaF;)~>6-TJ*s<>c
zJ`VL4uQTlYS+Dn%^@-uxq@!XQ@9e(Ma;w=m|9O~B?$XUo>^y1B3OWr5+pj!+9eGrJ
z6U(G3f!AlPGVg^h__Uq3?_=}#+Ny%uSN?1_yQ`aJA}xc~JUuLCTW_^p_0ZPC{2Dq9
zc?trKf2^;*mTnckeWL7VkbuaF7gcqkYgJE&TEDU7^=_z)+aa+gQlPu1Z0*krtDbyg
z+o;!fy3gOa>+6!<z=^Be`?l=k=ug|S<cq?UUz46QeE+Kz&+XhBCwE-s-_%)gpR^Zl
zUCk4_S>s$?)aJAU$C~Fa_exu}<iu^a%0DZ=&G>0H&+x#yP9vp^BSs6JFBf?|xuf#^
zQ317rJD+{`TL(I*%l=rAdgqDh6QMhs7FG7JGsq-d)1A9vd(WD~(~{1uGu++8W*i;0
zda+NTnea+3yQKQL&N5}!1$XY)uG@In>S4pX?2~gX&aE!L^6cWNXFeY)gT&j!CZ$V!
zW_owa*5c)?XV11=U0XgQC!thb<&2;EqvF_m)240vc;a}8QW77Vqy62#8asBicy8(a
zVZr5Zx%AD;7598Deu{Yd=Iu1^jHOp~9(hJ?ND0+rlxykRTDta0Wb~W)**kcT9Su!j
zU&(!9_4G-u+1iXxj$KMHX8$*Js<xL}_|_?Qmp_?Q7rAe@&7QYr-pSfqk|(#l752ND
z(4Q@t+U0sk;|`njef6K0cKy^=nq6<Yy{5;{;P7&jm8)V}&OTbxX3BHfyLDnwf$`sW
zhh}Y0D^#-LT{}s3!@<TEwHe=&E_?Xbx<wy-Dsky)Qs3LgX>REU<PQ91{(1Sym9-pG
zr!JWkCe2;@@X{iWlNsB&D=*H@?A>`@w(@MmCzIdOGx8R-C~5vY_D9p+|8?Z&)7cTz
zwqDq~WYOWg>uF(01`TBnS=Nn3x6TyY(w{BnG4XT)<Ck~EKAWeUWck>*iJ|BcWAD8?
z1)0kp3Rx~|^i2M~!!qlVsYuYiqs#M_cE3pV*j}`zPlEUH$NvmpL#Ca#GCppzcDKEj
zG0)bptvjbZvkXdeW%X*locV7n(~=42Rx5f6D_5i)NPK0_wKO_Z<ITbIvFx6;N+*&+
zG+x~4dGh?izpY8x6AoEk5Uoi(R$=p8bnfcgo;63;&RpE*BM`yl@cd`+qF$eZmC|Qi
zHg)nVGq2zI<!riMx!UZDogJq&cVsWjJIQ`e(Xa5UXqIJ|TkCeQFv}uKhU8`LpH_V@
z%F>nXDG&R=A)jT;CiD2^y}9aVt_wx2)><a&VADT)|5l6OlIJN$PhT|ZTbZy)Y|?cZ
zxe%F0o2KsB_WVD?)#z#FzM406x4hpVmf4diD!n|g=U=GW+9mEf+`4;Ke4aQhgq{7`
z?uFL7ZO>jVESqpw?MbT0SDCck(Gkgew%j=Uo+o^EWLx*d?^C{*-uh7-#G*8B(QeIs
zJs0*b&0_Z0@!RxpZkTm=;j!%a@XN|q-rURBvD1HM)5jg5xzXOuVcUZ^JsDXHY+v3A
zeKIZFNxE;9;n%e?!7-a|NS>Z_olz)3UNYdhOtoOhv$<=Qy8gb{^I-mpwff@j3fI-M
zla+2a$Q;}4SkrgpSGWQ1+hYrI7q{$Tb+*|%r7Xyw;X3=4xt2wn`O>ev*|K|TQD3uG
zrQE%%mv5fvTW(uhn|)^OJG<k>+wKK@?0NF5s$M5NaY>evTfmX(N4sBM3364-%nCUE
zt>N#-=HpSROD}W<SQQ^!&3D>Z@A5ot>0q~xEgfYBHlL4Q>s+1H``}&vkET2EFHdVQ
zoY*5jz0CddpVj9|86|q$C4T;z;PJEn$Cq`xeXcy;^=Xm`p9N3I6REQwS(>=Fn@bA4
z)0UYdz+d;Dp=zgLKv2Jt;aZU%Ysn>B<wBnevFfzvc~{HVlr7!av*cbzSfGN!C#L@E
z_WxF$pX8~#ytE>y_Dpo!&%M7G99Jq`(S2&?ddWuc<HqNC(z5LfpD(RE<$I5J&dMyY
z&sl38y|gc7n!7LK0bi~E)1cTb=HhZC*OqZljh~~P$9|c4n`=Z!g;H_eX(Q%6iRU=%
z7RW!-l3XvU+~F-Gu=r_Aa^u(6T`?B}19L82^fhxV)G>0OuYF3opnC(){Fq>!!$!L_
z?o52>Jnyi`$`v--YgSbjJ%0L1T=wHRtI+-}j$$8$-hO7YIi{}8ER(dzXwU4OdS@<Q
zdOI`sn%S*W{(S3<IBgHUjx6uhzUP_$j8)4sljDu8ZPxl}C-Zp%wI%(+>k8HxA7Gll
z<i;7Hx3UoyOpfQgCSG6v^U<!HC6hK>%J9g^l`LPPbGz?Za)al0k<j@Y_S^58=VYOB
zqUv8rxKr1*?$2gVT`QPxMJ@5w=rKL1vHp7p^Y%x+sduvLKN>{19+?-wdUm(NskdKM
z7_VR2W&JY2?Ay<<GI#ZL&tCpf`1tbu(bR*hIMb4zMb8x6_u46%*(Rm%m(=rFo{_pd
zIg59yO}30&GOI$I)iH<RcD?-POY5Rz{R>qus~!;0Nk8qzlBYbeKmORt{MPq(&1+}n
zF335@s4}PY{8B&lnV}-$%|fg~3J+GEP@lKs`<e%HLVqbmy|H@O`$Xu-0qs)-$IGgp
z+XuL&#;BZ9>EgJRuOf6iUitP5-}uk#o=v-X&EsCZv3yC*6PbE_uS>f&+%38;=F>2-
zX-1s%>+7b${aeC?W6y9r+A%G8A|KbPpONRJdnW%#SH8CXz^14hGnwDdwEJ<7t-9p#
zR=3}8V^j>q4K0sdIrdC>#qs4LudO9hcg?wLlelME?7>Ru=@Q4ExaVnG992)<arwq(
z*^e`)>xHnU9#ELzVgEPl)O6qdUCGLK-*#IXr7!<_eyjJxznWX_8a)s*p84*C>W-9{
zbes9>pGR#nii_318L}p^O!mQsFxg+<*Q&BS+BxI-idia~95$_=BcXZ6*@ds-^?gy3
z?;AOdD|~kg?2eRX-1+5P;bqGeYgEd2S3I3?SV4MYPoJ#;yL^DX*SvSvEG6#P?%jRv
zjwFX<@Qr!@=4!3H>2_U9%k0XDYqIY|Bjqx6A6gg{94t?6?5`>PVp?`oFHE(?;<M;B
zYne$7mJC(trqVOi!kRQ%EQ9y=?KtNDpCN8t#C?_uZvJ}<Zau$Q_wx9o$~B=Ot11QS
z!VA|fohTvXbtb+{>A)>V`%7zX-Zn9OD6W5Mk{_ce@6jXOCYx=#<~)>)R}8-_w`|Q_
z$(k8bH;R7TpY}Mfc=EsMYa3Tr?>oA3UGvfnO%<1(+}3{35T(cepJCO+AZza@J6!eJ
zxi(JwJ>zZPoKQ=>yh~!cwpHBGJkr&m;n>F^%kiLV+O?E|sK~oXb8Q(pZk%(ge7-dE
z+%Kc0wda<-e%@d1;9z?Go?>Q^6?62oOS<P~a<63Nc+757_HXL+irJA3U)KlyIcc2n
zs^|37)6BQF)R^-B@mZw2!T!|B4eTt!*O%s2<k^3@P%APkFz#b4_pT`^b`COM7afyn
zxM{tYH)>bx>m83xRcAd&mslJ3WYTTTb5)=3t;{ZB6gKzM_ET`MWR_f4exzl_sg=rw
zHqRMCd~fNfyy<gXGVR!}VwsJPB^8ga-@nw|Zv6(g)^g*X6NU4v8EntL3^=YZyJD8^
zi`Daj8Tg$%pU-!&y0j=RHY@wdtz)x_de0hv5Z`8a-EMC7k$}n5CRqmUR^7I9`>y1D
znY!<ke{a~;%HFr&^ZDf>*EW|-Gi&8NmCiDGYyWz`$9;b;uV1_N-4u5XBY*L`Qyaqy
zZ&X*Ot%=@e^3!~SYZ#BY#e-im^Y&exvrzihZQdgrW!Jd*>n?OsQB2JY$v?g{FjwuW
zDep6v&cdn8&D*aeZP9%`aqa0tVGjkSrMR=UReyc?WtD#6bGO-EOl!n51B#Y^T)THo
za;~YCap4U|(Iwj_oi0p^inM>a@8h)NehdDI#N6%>OP1ygtx~f~Jn<rPVvL*erzic@
zn(B3rIOPs6JgT*+TTSo%?w$=#wr{gvn)X~W>!#7>JL~d$%>E_zay2)UeV(6FzIvrW
zuKJlHe3xpLafZL=+hPBFtNn@o&TzxeTNX1J)gJ#5@^O{bsaIl7{K}FFjjyaOtX#S#
z(mEr3+md;neI6~<)1{r$8GeU^eLrz&m&y%?f=UVY{|wi(+BWiBzGpabdDGqKl7t7V
z=R2NA(9FrbCUWf9j+y7mkIFm03ZMV--p;w2mKi59@4dLzcmI38w!gAvOp9aVoa#%`
zCOhiY-n~rfSbyrsB=>Fhe?z(j#rCCMdB(~7_j*mu;v4Coo-M3y**9Z}gU_akdl&z7
zGrkU5&3rrK%xd!;M|TJc%jC#iFF$q0((cE~)>CYM@1_3qzdUch*G#po{bzbF2;5|i
zbqn7v>UHbhr9%()o~*r;;x1u%;CS%yJc}jwjlM@+{-D3?S<{woF4uh~_n0dcCK=wA
z`)WH|N9WwMd&w(*PW3BKzqUE&l<C%!Dl@0q8>DVxIA^#;?yA~!Ym>DSDSc|J5`T{~
zJYfE_-|Etwy<d*(>9#ui=A+L?m*milV_$jLC0<|ua6ojr+)~}GvyL%*zFD>+`1X~_
z-nAvk_cJEl+T`Zce(qUPk)`^J3eoxHmJZ4j7l#I%-<;{-9>?*@(xy37)2pcLVzck8
zVu!Ol=Dw0a{~2mIj$K>v<@LpEi#tavU%w2jh<oO}Q!jV^TS4ci@47gnD}oG9ZC@Uz
z`CZJZ;6TBHk8w-QE0*<j<!)X1qRRA3G0Wl92)-$2tJuY_t>+eM`F!5=L`BM-+vf!O
z+D&iXyQ38~yG`=b<Fw-sW6k%Jn(Xa!K6>fA+2+D0qQ5Oe!VZ5)u<mNQbS@*u%;cC$
zg1NZ7=JU_z{kBfK)v<HxbI+8|5z_?wX74>$#Oj^A>dwJD!KI$omc7jk?)E<YULXE0
zH+gEl+2VHQ-C1Gh8MasKS#g-jvF`Yd&{(T)-kF@Khu$6Mol}*x`78HK)+5S^91$l3
zYA;@x#31`)wS2D8JsZW4)jVIHRMp4Tu6mgyle#i+-tziIF_PEUIm%|cy%nmMDI@SC
zd41I5J+}3_J@<T86;ufs#)i*}mwa)Tx7u*hl)Mu4tD+?{{O0R@4PvWH>^kwIRam-Z
z7V|#qBljBm#6tcv2*%o7)Tyx0y(053^lY}cj+nRR{&F4fXNwK)Zjs}E7#ZGFa!b}@
z<&lTSmKKU|1>Q7pbFfgUuJ5&eD}1kVV|<dgMc;=dA7<DvUbgwRmVd*6OSy_tVFwjY
zFz_c-2S4A^bi+wE)PY^^%z?Rw_Zfd(FZkIl>g|&vyQ9&yVV^GM-c`CGAwP4&a`uT|
z{}_kGz0J>UzxAxk(E6>ukohUc^9qk;zKbR)<`=AZzbSj_Pd0CBAwPE8U6;37r*7Xd
zO>L6xs+)YCe;<Bfzfu)?Y}V7g?^^Cnb5-BY@h^0@@Gj%eCA$57cdf7PuGeJY@+b{u
zSu8m5tMb0A-diy?(W$;iYEE73E4+UB&+n@);!XE1IW2!)$2Gz5P5t!e!R5Z&lk(0=
zw;H}n=y>pSo1Dk<jVm8sN$d&|Kil(|{fEEuUiVA`!;WQBm?cgbI31byKt9fM$t2EH
zi&jO&+dPLSZ~y$-ee1=dcR@KDBKfs0F;vR$IsaLyt9MJ}5-y+ZD!B}a4Ub>=ipH$}
z>AwEUvA??n4nCQFc2nU*ZJw!boO7SZ{0=>#y7{-Y>XJ>Tw|rXjGx%kj%zuV?*|Xau
zb@{lqPkH-E`THOC$3H_?m-SvUPjS)mhzwU~{_^}!MsHkr_QW&SdUope?raJ@c4xP?
zCbLT6mlG#nX;q{&ee$?fZ&WE*Ch>_qj*<EDoS-<5)5{t^8~%25+`>Gu@w|Qf(zN=_
zv$?fTD+8?+eH$;x6kbUvd_7M)(qd)N-lq2pv;N%PvhhSAkHJRue4V|9Jv=;e{~4n0
z_5Bl2&P;R*z18Z+dCxB9`Q@d>-IdHyF)<R2-90Hx<pTU)EtcJSWp1PF<n;DuNdFGc
zpPpZra!u^c5!q_8_Uh8BclDf4O%(gI@Mas!oWKA4EZ+wC%0BH~dwA{IofGdE7@eH(
zN2a>pKdSF?NSMw`{TUagp4N#ivXr~=Z_4Q&r4#R#^1QJ*ram?Cw%lAktF$>~scLUa
zl>PNyC{<h(oUnP7-3889Zja|jm9AdnwJ&D!&m`Z)PjzBGUY4*h)%l&&tHRZ0aDzer
z^pC1gU$^vDpZO`p(yb}6J2`e7kdObjH8ov&o4(Q2J9!gzy5b}o8Lv5n^!x2e>L`uu
zJw8$KjL<gGcplNJbjuU#>d)=3?hx~c?SA*^!0(Ox!F~Iu`t$Wn{`Wn2p11ep&#gh{
zr3L#u-u_4l3M$XjI<Ylx#oliZ&TKp?68n9sCD%<$S<4r^A<ZFu+>ccsOsq>_jJU#7
z_;L>ap`};eEEeL_)p@&Q=Z&+1y;s=m7BpEtUuJ#XcumHwHQR$KPa7O$ym7k0mgCzo
z`>DavnkB01V)<Ul`g`sOw7IUyV0l7|TlkIp*Ox&jS01jJ_;-rY^q-G<9@ty5{BHai
zF1zr><>Sk5pG)MIjya=}d*3qjmT358m1%#peJ8HBI~V5KXOZ%gWn$uD`72*6*{^Mx
zwDrrKskLXn?wFb@WAyH)QFKA>&FAt_MXPKS-*HcV`m}33%RIm8`d_Q|?p=~2<(~Um
z!ASN8i$PKm|1viID<PcQGE%o}nwln;m~>c+<K4dh3|`0YKlk(4mv--X{z8fQ?(rJ3
zvrkB_s_J4?KOIz&_#~H;Wr>h%n_Cr6$oG{~)%OR+z17WJyi8i=jl#^B^mlU@+~caQ
zdp^AJpP}XmU+3R!5rxR}GEy&1SZ~X${lak7z366Aj!5q53m<H4wXzF4{^eY8oED}$
zzwrLj$bM}(ts{K5MK&-mD~y@ml@{^Plk4xj^5n)J*F($di{~FGjA8GOs$B3}N27nI
zXUycL!*c`-Z`*#14F4fHH@@w->C%l?qRO%tOgMI9g4UCSd3Jlcu3vcfPTbJX+*Ocg
zrBa@1<>UTGFVA1v{KuMO%Vq!D>mpO89`0By?bO$`zw_Jb%x@ieHaji|y$W#Oxl?&%
z@a>O}_IT92T&X|vtjI;PmF3^M7fgIJNy(!;r(dV}^$WwG+h=(;*%W!XWXr#ooZj}N
zZr{{d-!|Vls(h%;?ZCZ|cPhUbuism><oieGe@l<OnJG~FMB#}L=gX$rRjFFNJGa=S
zZ!G@p#cSKW<kdxEnVvf4d9Kkj)-4uOxGU4;=Dh6g-5<8U3nPpquANnByI=C|@ROL_
zSCw+}w%X0i&bPUw*p^=4%zMe;xqRH*%3s$CCr^4Z(WqR=dhP^cAI^!=8jF9N-@iQm
z+M0~*yf&NHd~DIU-6VG2ufSga((IalB^UR^#l4Wu(AncL_t&rg48L9-+3cG4dKVk(
z2`x{J0{*<?{~4}F{<b*2?vP7(5qBr|yV9Pb4vXUp)jykT+x#wf+a`sKDHE?h%lq?v
zMQOy+$hS+I{PkE8e2R`wY`*?1pt|tZmK>HMe|<xvCGpCPy=#iEUteeZW%W0%S6)%m
zPi)^AWto@0Ut4He>hVyUbW4U`31L?|!nd~BE-{`zTf4F6lik+fa~W%|Eb-bXS^iF)
z?`%)uTbHKnorexRRMz3{J(4F=z?#c)Y1h@sVw3f<!ovEwO&UB_6zqB0UB*y1H(S|#
zt6{uPywT5uO$Vh~B$X%Fy4g>)%;VU#GsDr){nL@3I=goj&ilKU-86Ao*IlkHlceWK
zF41l9{HMR%YrB4>$zHF$XTHsT^J(*zMCEfL51RHJS7zIC<GAUgnRVw1X2ws+PbxgA
zwo?0ex$Y_b$ySxsF^qF2URJ*T`n~I&ZWiuuOZ2$qXP@CZEFt;2=RZSh@yeom9<yKB
zrA*S^&@Xw;GilDM-_N_I3aR&HTrp;4*~x#-__+GL&G~wHdb)Z-mOf`XnDi{<179}=
z9GY%$bCD~<{J>1k(y~Jj`tE#TT4l63xAdL$me~D$Hx9kK`MJC$sG#7+pI4D<S|<F|
zG88Lk*c%|{|8P!d%A{NBca3#67`GQK*|ah~?m>NPX5`*I-vvwxPkyr$@hP6lT0Hsk
z<;ySqICo|pG&6R+n7ekXc6gUN&+UD2tL_%}*JVX__?SB^IPpGeo>cwTSHf1&91D6Q
z8}0V8ntc-f*5tml>PE)z>7T7KbBm-sR5HE!IDUGpooTR5b=mzbJcgk<^?viC7Bbvo
zmU;I;(B~uT!)AZS?)HgG&Sd7R9Z&9IobNO1*5aql8QV9Wz2(cg_UV;JsRkRAPj(0`
zm|xfO*wip&=Qg%u5{tNZXD8&9omcq<TDhy_St_q4x#r6kbD4(C+n(oHPsqGBDZYED
zXK2nd*RZ?Wi<HwRo{z6zTT!YUbZ5EE<$X)vD|t7`-raPZ$Mea9Z(sgRT_{twlE*H8
zvP$IfX$wSuO{%)CUv2ZPXiCMh49<Hzo)c_~|JBbm%1z?mly-`Z)u>;<=D~!DuX0}l
zYj$m2IrV4kxjs*yi9au{=(9M-aCg$o7j`qguIqQc!g@tjCUIk2e(73Q5o`7tO+MN}
ziwk6rUrD~bH0R~?eco%XZRSj}N%haWDev{>P}ubex2F6%ZL93(yXmdvF*l2K#g9bx
zr(Y}C(Efc>n;T2v=IeG>w?6#q5bJneZ_neUS$AHRSshl2k6v=yS!<f4_Kt*qQ(s!F
zJa*h`+nLYR-P1g7c}%k3qM`NMGw44<d|*g&OU#yXo>xnb9bw{p{_^Yjs2d;WHeM=y
z%;a{WXQrolX^`c=(DQfB%4qLX3+=Z^XP@(YnXUb=tDmoitK3QXW}*IWTg<i{U7F<!
z1gGBKar|Ih#h*;IP=g&wn~p4)dZ*~*eY?A#v|6rbebVf+ooI0DUNoD^^Uw0V;j_M_
z+?tSl(kWVX*L?q_`nmZfF}I`wI&*KQ&sY~}*2=s*u69>wu*he_Uis={g%al^s(T*a
z|Fr7*t}TUkbk^+Icw^_eYsvT5uRI#I_o_<u?uvxe-VGsM_q84`E51IzN_EvI>5N;S
zm~Z@?m#w{=r@-S5yCu`?Y_pq-du3;zI%qB4bYz~#pI2*+J$sT9(8+UPBljFVPUfCu
z75=EbOE%Yih<2Agas0XRJT`_0g|A-+OzZkRWlislhaAm{@itFRc4^yrymMK*^MK_$
zk^c;Du7rK;-hSxty*Hu1eJ&pq%;;vAY0bzW_viT^E$+^RyDH+YNPL&Sr)Tke``6X8
z=4{qmwo!|<%Vsst<I^vj4B!4)#nJ4tT|8{!&+gAAhI^cigbz;K@%HVjs%w5)#f3X(
zfA8k{bTC`!nE7#${?D6}^H1$JwA*>|sB@m(9ran7Lb|VfHk&h4r_8hdc)zQwBE>4C
zAj@1ybyB6ugk#N&b+M}(*puv~B8`_vX+=l9a}9OPJQf$Yd*0l_pFQ*De%X~DzKB0q
zEVJA0$b>cNnL4k+&v1V)*f;0OwacEm%co22SurVp!cMJd^+~q-4{hCdOl7N(@yUYw
z)qHOv5}wP<`C?mts5?^c?AjaEC2o=Bti3e{j11?@$zw@YwwwDhxT$;Vmw)>iy3B1f
zYqO2dCM{XAuw2iLQ+*E4R~wlx%PR{yQh&R2ecUJ{EO>WE^ZVw1g%Q6p`Yt_t6dqlW
zm=HSSK_ABnNsFUapB|O9MMvn)&pExrrgsag?iHRQi#PUCuY$jO?Y4@@`?4a@T<w|W
z!v;}}<cBgT$?WTFU5h_w3B3wt>lO02!W1f<==WICZZ7+$Wf3nmw!Qhd$iK9$wn1>a
z@(Nia|HqeSEzM3l`Q)(mCt->Dsnc&KK3AVJ|HPG$XHzFdG3Rdf<^6kRy_JPkNly5w
zoYhm02QZwk+sbt_s`qzC&yEzPQftTKK3jcxCLEd6k{R`IsYQh4&G~*aAC&wGR^5F?
zPa&iH)U*W;&&&uoQp(cYe?b0qpvJzWDYI^rPTk`sD3G(^_?h>h{sHG>`~KI#C#!C!
zbevI*5mC0~D?Gn`Yx=z`{^iHcU2ywg#Gu@fSN`SXaliVh#&2_SAC|Z0OlEL;f8xN4
zsAYvyw^TFN`dXxa%6<FxqT5&L3EM3m&ygv-P!&=rdtFV$Bg&0kTRGYCg+cAB-lK0X
zPFA~^e(K@u1^w*HEbW(Co$Y=6GdxqpE}G5#_%w-f#?F5;7G2uKskYViWz*@U(%C(V
zp|uZWWY;zxm-&2KCTUL2(S7rp%-aH+)Y>x?nhvus58!-!``W7eF}mxPU6vH`ObxP`
z<UQ5aUjD$<EoR%C$|KdfA}0uVTJyQv79ZcfG+R>BY`){l=;T$G9(&uWsdk=L^L+AL
z@@>%U*3j2;uF3UR%0At5;YjE_{h3{6Pfz}f-LdlNLx#zAUte9CzGF?v!MkFzflKbR
zNWI+hMDXQf`6Hg|;;!8d3n{gXYf0O+qwvX*#R~$oBc`3`GW(ihk@Kr`El0?fn0l=y
z*IVq@Z8NW3GJC`PSL=v-orMFt<vQn(d0}(^ObNXHo~QZ!{Fh(WZ4RB7=r-}kt>vQ6
zwqGx{I;pfL>u^!;Wi0_)%cF&lo}ZMK2%YYD>9(VY#w9D0x6{-duO7Fx&5~YS$-U*0
zUq5H*4d<f13cCkoYfipb`R$@DqGeZmzrx^L<%@c+ZPPczMEmpTef#utaqUi(DY84B
ze3g9tI?g3kb9dA+&o64(6{lBk{3IAMHSvJ%ZJE-SXU%@S&C!*(cTQ5p;zaI&YQ`J#
z{~1C(d5<3ath_Hgvdmg}`cFT`W43&bYi6HUUC<OPebT0gcjLxWO;y5;C;z-$m+w94
zw%SIqY2iwmO>2aw%PUBS2+mvZ>pz3ltm}c>GAF-D&<S+k%`nN~fCA5_s^_1-tch4^
zzP+1AY*X#R`PnC%L?dJ#q;HR_U3Fo5Q>^DC%}7C!e}AtC8opy-w*R;4oZ*z%JqKe~
zi?2)Udgc48?B@ZCd+%FMN`8C%`DJjH(M+95I{aH*j@dA*o>1Q*e9pptYIy5nL2EHV
zt!-bOiumoWR-9hG!e3j+qoU|kNXZNTK+uY^xX(q-UF8>6Tvk3XbMDMLLJm)utj+V|
zKdic!l5c(XTT5?*D97Ha!uei(@1Cw(=XA%(b&tZDBpVj>4JTiJeYUhs^6bhT{ipk$
zPX8Y0sm8RdaL%t^e1EGi^~5~QSj+Karj4+qT2We`Td&QAhQ~?2m>kYJ>(A&{iitY0
zQG-KriDJX?m@hBvWv)h_w%%>=xbNW<n~CoQSEnpIS82=saMs3ji7UE^UFIPZr>yXl
zpC<X;;!VZlwb4cX6W6*gePeVzFjy=4&?&_Z^XWo=Zo9{?tNqm&CHCn+@}%gQ%v)A}
zIzDl0;hk-j#}}-*x%P}vrO;WfBkQ*ub5@hy=KlKL?7tIN*tDOixTGGl(DmdoK1+r4
zc|ZTWddRjVYRba3y(T+1h@ESS?fTE4HDk9>)#E1q%b!+n7c0w{c>I~9(?-F~>S4#M
zp6#^M$v64U$vIX10oy!_2$P!U7AhZ4oUh^gx_!xc-h1;l-SzRAq0p?LBfZ@3@jA;d
z%THXJbSIBD<K)`Mx2;b_xodOEm|f+OU@*LFGub;|X4)D7*>lHp)?V%EyK7-n98h`Y
z+ht81(I+t{*dE<&x4SpBPEbl|NsmhGk`?j${E8dz+dp6H{4zl++Uru3$9J(N5&e$7
zLypYXZTr2Vqu1+MPwKmoQ>xOp*YdHwPs!4q`@<Urgj7!4Shc78^^2;)dux8Ao!qK@
zB3HTb>^9G-*XPbSX!+%Ju&=qv%E`MoS<UCx3CuK3e!4R!Kkk*~zjTw6i&dxZE}rZy
zy7BUt$Hl@AQ!}pf{%5ec>clud_)67e|BWTHlQsGJ%-LkMUt}Ji#>E~ZI<-SNtaEx}
zU$%Vc?x|l@JQk*TT-lgY^3MOK>FU*S$1@HIPn+bL+fZs<IIpUFU0^A{*)?~kly-|U
z-tRk~hTp#OFYeBbmACJx6?@*;v}9>Z(UIRvCW!SqKG9!hTNpXrYPB@ywNj&0j<n>t
zi<kSg{I%8m{A%LtMg@5*q1YKc=^G{gW?s+D(w%U7^^`PDt>;=64<1jxex+o+m9Mp@
zvAD@OxkuJ7b)Gs)#YD@`(iFYv^@+{$eC(aNJt4XhZJ7(s9DF!MVW;NAgUY%S-2Bz$
zu3kSb<eDiy?d0~e=@}CWBK?=k8qdmpm7g!O?b?!~B6n3?Ib!DgIWBvAuU*E{%8aC+
ze#t&f&wf1>%vYZ3ac9y4cJ2$UiT7=aU&n^GKbRX<@zt5d*yzv1efJHDJQ&;K=JKYW
zF1RknG12Q@A<t%>&%15eH$M2Cw#I4O=VSXbcNCrep29tM-puFAf4&Tw{rs7$>+R$%
zAMTvpvCQq`0lTgDGv7~?*>O-_yX$kT>WPdc;S2^WQ(yR3eeqR`aa!ir-c<YgZu5or
zey^UDALVwN?Ai41Zn2K_;-tM^6AFJ_?|vn`_4F12r-dGOj(j;_q44|jwSA2rP2#gY
z?ERy$@P_+N_q8=23zh#aT)b^ZMoekq@tXN-D=nvlWU1eId?e$<xzlcUb<U*Dej?cX
zy{;ue&i%F5MZbQnbFwpp8;&by{AZA>*QzP-HrvRv$wx4``*dj#)8=}?gRhL=z6^>L
z*%|5ggjY%8mdZhez8S~Y*+)Hm+TDBk%CpUipPvcNn91_=)FuJOn+JZj%Y<4l&#u>8
zXP(Pky_=7_ia)@r@a<)jn@@I~5Im~J#_)c(!pC>)pO^n<NRIV7b>@-na_M;0&FB4g
z{Q9=K`9(-++R`nTZ1)6eW==i!Raq<f`A%E@t@#IX9yKT!p3Xngvt$EnyU%0$__aZ~
zMs6FoZtROLoo4yHW?kcr(%c)r?{56ER=R_2ezwR4jYw{jq7?yV^Vlx`nsxhO%$u8)
z6VI-VQ@g*>z2KsO?(ZqNi&Z{eUhkzPc4eWf&qlY@I}T|KpQ@6&M6OJIXL7yh*wVs$
z)2}~0WmzVR#lB)pJs)Q&^Xr?)3x$d$8t*j|4AzKSI==Hbp8rW(%+@WMZQdsX`BVQH
zf`v5#Qe^TH-(KOby10F>xK!+t&!^JnJuaA8zw7ygdga?&lezMOzV#QIeBEh!WZG}(
z`m+UVI}SW(F#F){e`=9$s9r*wWvHCS$y?51uTN^<s7hNt?cU|Yh#Lzk*_<|VPriTo
z@`jfYCl}pwEZ5vKpXb#gfwhgzFKYktYHgUiuJQiD{i!83+sc}Z^S8);_F#H<-mh@-
zm-A8U3s<P_Y2|94cr0K}!_0FG%w0<h6J^9UPDtteu5~KSDMWbwjO6>uJ0!m!wK;M!
zSFlW&_u%7CcQ(p=dBWU(XwIUaxr#e>$4m<`x_#Q(Xlldz>DhethCNK~Yn?^Utj`HQ
zSC*>6bd$IFMdey<mU*n+9YH1yPh|{PW90W9tB{Ysw5pQ3xZ-|F|LL#?j1v|J&(VBQ
zVR7Eyt7zFJKl|%5qMl}~Q*X<7v}M=UO|5m2M!WBRp4?=$J?`0}kh|rgbDEw#O;$cR
z&-KRXf6pzIZ!gW!o1U<3JDYVx%}wc?P2Sdq36>{hn)@sc95>B7<XvlBBFJp|NbS<a
zl}EDwo!^@;QFK)8i|LbBr+zKd(J3^&dia%fW9s&Y4(F}+emuu+I$dkoqHCKYZbY2e
zIN{;5pWD8@F`VZgweQgFs?{zZPZXSy`1@wg!+rI~7Q}7~u3x-Etgo)(&xV@C^A=A&
zaNd6T7Vp^=FaK6{Pn`B}o9l$LHihoill)FfOqca%-LX=wQ@!Q%GnpqEcUpvOD_?f)
zndmcZHnWRfK$~8Y?x$phdF^X$a~Ir_eP<x?jyY?AedEf4*<lkGFMH!CyLFe7Y;?_B
zH@8Qgud1%)US0FO=Xs{@+0!XK)xSzacYKUq5vV?=h4<jE2UX7{R_H0Ug;YrPoU|5v
zddwoHZ{Ok9FN3A@q6{q}UnMy&J(<)Lud8>;P$OcR&74n_|Kg{H_J`bBnOd0FshO;B
z<AK3*=6MFJ`L~WICl;D+U9+11)Y6$D8%|1|cvJD8;p-J4=alVig(;smzBs19d0e6S
z<+ZI(3#yx*?%e<3g71o(uf)UU650#;OpEVNs!qN<asJx;*c(Tdu>HEWz5UI;lMjD~
zXf4?BS(;hCdQM%$s;GwhmTy?ALl&R*Rm&GmNoG*2U48jy*x}75)42~-D6fAU`CaU^
zh~68A1Gd@AHf)i<S@!D3-KS0N3y=ARly5L!+cUdtXYn@2iI-<x-{fQ$W@)kPOm{x_
ziGt9OjgPm+Y;)}SU~z(Jt>T>fy-Ox>n`ntFmJz6*TDd#oR{Wj`_qVKx?Q$zTcs}{%
zbqV>Xw|0v0CwEy}9DDSRfw4UO{*CM0wad<@b#J`q<o{8IRe56o`+Cvm$<lkTZ8MmB
zW%u7%Yyb80uJOn{cKwEfN(n=q`=^y98Lazft$xa>@^f-g``2^)aeE%yo=x`7RiB{u
z?ox%1?!OJ{zmBQTGcWF&^L@pdr<1%}x-?@Gg(}+xRx%{B7d~c|4}KdH_uZ)M%G|GJ
zTh03{jq<`+Kb_EUu=)1-_T-<%p=;*dIV{qxI8TLV=YzZ?i`(=3EnmKkjG7#~b=66=
z8J|2S_Z!vZ1|NH8rN2LF(RJZzg54c8$KrQ9E?U1L$Wqq&`G+-6mTTIY7#B@WcQ^gI
zb@yuTVj*9pXRa}u7yW0Lr?r1-e!*GYt#_1*-sY4tBtMmDF#c<~rgH6-6Mfz)lk}Ju
zNO#wlTl{CRT&Zi)8RAhsQ?$)kFL2i<_vA0HgDv)M7vFPpWyXX%C-!H}*upWJsZ-Re
z=Qm&7{Iy|w+Y5I@M~bBv?i8J$cZuKmdghDTe_NfT?WfL5@}Fi9vQql^ipP>it=|Y*
z&t%!Gc#AFK^OqAcbFLg+w&sHNoA{oeWt%#<HGEI1SeG--Tk>5YL1xzWs%hI!eTX#N
zAvnP?)#9tP@y}9Qt%=*G30=_sDIpo@`Gom=T*V*bm!XmGo`)O67)#Uzo$|Dl{dnd1
z^8XC$wgf%xSK{8XmHYNh{dH+!LQ3qsTATkInA|h<m2u<SOKaTKr{@3Oz25AO*!fd3
zDf1T|O^V1YI$4)?dz*PduR>nRlfu3mU*3l4<~*Gqpqcx7g>q8wv%V$i+pHg7o8<qf
z?@gh*b3WIdRp(7Fd0d#Foavs!w|C>4?5ODIB}~=qfg6i1zPq*j!+!?*sHb9&JmxH&
zcXi?~d5y<^AKzR5cjc1STh+5OoAd-z6{hDePvF0{Yo*_*Ygex6FYTUgo-g!v@7?@D
z10{<mGIL5TWf;H8`&<hNJSkEnvG#;p5eK_gyX4ywEHclRD_`&CU3c#eOT5xO)dM_d
zckVgA=&gl);~%ZHZ@*mIeeH@7+ws{lI^yho{xy$p?~S=q^~y4H$CL0ri9FpKqUxC*
z^G|ae_&Cqwb$E?&cIE0<r*oG#U01fZeBickjbGi&btQ2p4lLdHsatRd+wbSw+`aC8
zn76-pY5bifK{Mu+NZ<JylPpvAPw~fJ?O7rZo=n+hxU0(en#p;-7gkX!zgF};d%5)1
zrFTm9#w!`p=aj2oT9;*~xaHE9UWd#LuEI{U7r8jBdf0xdZs&<VwyzKW34iR}z4Gj=
z<KL<)Qp#KtuilvCf93hlYr(HKdmiz5y+a{^FJOh`VV<Ij^Uvd&jyTPm(AyoVw8edm
z-PUEkOJZ)^yT|Slq?mE8+=bzmpY?;D`7ytwK3Cef{F1$VGr~__N6b;_><(?wya@Tj
zUm3oy3tj*0@}CbsD%O2EP*855vXY_6lSTggy`A0NGmh-|Y*%|^P2ik4^<LA1f;YQZ
zO{$iCxAL*tH5pf(#F`Bcy!WtA5VA<T{KKs(eH&}|ma9EQe=d0~=Mi`^!9n8XkNy+a
z<(p+!%v{WVUcy0Ky7IaDcIEzKi+1dta8`ift*fN^*RzR1LMz@ETUy$ld$~1XveUVR
z3HkRF5~sbh{8Q}j8<;!K&3M(CnV%lVKl-k|fcfL|&+C#TTaQ%FIVPcZHSD>}ue#E#
z3!OL4T+2#p-CwrmCi|7g_Dho<rmobqj=8sOxA)0g5zA9&tb3kf$^MVYx@z9NuUVfS
znYQ`g>&U(8Y{?}VK7S8`<tqo7$M-hAH;vyIz2#&&vyg)>*V9M3*PmK4NTtqw9lPYS
z(m^TD8BZgHvjl1#U(SCSo~?3fqfXz%jdSj)PgZ(&*zivEbB4$EA9VPGqojE6&cA(4
z_PUz%RKa#@ubwANh41_3ZJn~MS9r#T)?^l*hf^AF6rQgyj-B3OV{}hr@6K(fol^yZ
ze){MnOGvz(WL;<Cq?N0&ATaku=8yE#vAT<V51*L%DB-wjgv`WW*BAV#N}Z)BJ>N%a
znk?JZwJl)}wj})dY#_a?h9P1Cn?LicN!F~Fm)-FC`uJts7KI0stRKwk@_8Y4*Jhn%
zSkM|<*P8E-_L}C}_!i5}n|M@s(vg!!jgu^HPoAGsRJp!%>y?G~b$2FAKK@6i?$f*;
zTi1&9N31XWQ=Rtg+V$f4)l;vt2>1WnTl_dA-9NNg#>vmz^7Y9ZcWqK7mrh^_lw(wu
zFRgu*y+Z7;e=-MCp5>?09G|Mh{(bi<Js7>S=5$NN)04MOEw((y!DDIr!){J!uhHK%
z=^5YFp7C=!F}s1?rsVa+zU50RUK#qGPo8{s(#wzA?i)$$In@5_ZUdXu7yGEy`ODWn
zJ^gdFVpz?(qk57X52$}$+Z&{vYZR{;c-rD#n7Wq0H<l-#k1KzD9qQP+uyXCYnTvZ^
z51jvA<oR5Azr}IcS9eXTOQZX4%{14vn{1`&x!7RB!Mpwb>wmtC3bSf`w$`w9S^bPX
zO-@VZJhoK7{PD}$-4=4+48K(dSu8%Z+2i&VJ*&3|CFP%s{Qj}HW&gLDjX6Q(0#8p1
zetlAF9b6*1zLq;GEB)}YySwTaZ@9(B`I#%_*MmU+ZJytRu0NIRd$j44a^n0XtLlqR
z%h}ei$nc6Qad@p@xLLrFp(pwN)7M`gt$+D_L$~CaZQuSSI61qBnlteG?l}IW_O%xG
z*VIp94_j~iUg!5j`AYJf!t+b5uARG^D!BIfxp3v4V?W={_p_3}w906uq}oiGGZ71g
zZlAfdW}~NcO~>1xGLM(pPh6QWOZ#PtIma`TP^afVc9$Km(!Dz=&ATkpF@}|8(!P6-
zSG=6qwKS(urJPesS@6)D4;3erm)rJfo4z}#`dResG0(7573-e1E}S5*Wg62s&r>GF
z+;Zt=Lucve+cQ2Y9zT0feS4g(;a_dN=$Xp9T&F&XxFg_b^Y!gzn|hm&&XViw&jPib
zx1XNsCQ*D|LV5d@uzeTJ^H%>#)$H2N7$O-N%YW*COyhY4zQb!XbhdG(1uDhQewHYe
zr_TJYKj!nT%||YtEmNB<xOkF-{GprAd;V6)O_lQH5YoAM;|SwCPg#$cGL`=fRsmDC
z`#b$z94&M^OYD-9;?}JZiye3R-FYziLap8YQ|q{&%<)kXR654^X{NjO>)Y$?rzZc@
zHobc2q_o?|w<TQnwI)^kF|ZH~i%1EdTo$oW(dg!>^!?knroT;>j`7+uZP}f&YoE5y
z)_Jd_v+7jWnN>Sq?3BCrRcvX>`W3S~pT(Y8yFs~di*W1h3EJ}C?}hH^`xo_bZ}9}_
zOpDhy5}wc7-x|@D@g$|vnIkr0<H@AL^U8OckAGQPdFQ%LyVkns+1I|aJ!1IIbKhvf
zr4)g~b7Xiv-~P{V)zc(XcUN><?OfR>4BMAENF02=Hu2`Z9|4_*9k_G1IVqpddEWf;
z4%e(t8M?~B9KUU58wqya;W+tR=2z9P%R#=yBCEG-No&z9T;_M6b<&X?w>Hh0ze1Uz
z*78qQ`1Wb<Pfv`Ey6bnP%YMql>+{cid3`i>-Acx(Ax)pQRQJy)XDSywR`s#(uh!YQ
zcM83AbR2~AJ}^FMzTCI-SJ=cUsf(SS`xxs^T~jzUwBX^=vY(wQk0t&Vep%Ol_=4~1
z*ApU7u}@mysc0j_Uc<*;_~R<C{vmUz)lqlu-CdPeD!8Ol>v=w7U%{T_D-$eU$1Pu&
z5mnzF(y_Gi-8#FVw++$E77S$v)^$Z>dM>-yaY=Be+Iod1*_+4sL>*qgjqyGEW&SNb
zSuqut&k}|zpK9&bmPmcQvu%2|>6w7P)7(ByoK?Ft;_HcH>^`M`c7Li<yIr_(ap!_P
z4ql6`nqJ>2sFbz$dN+CJe#3>&mTh^NopQWzs`G|7JD*%nUSsUI{Di`RnGJnrU(&1{
z|7Lf7(D5+YxmDXBDkMiG;%DC_bN%aovKR0F$lLSHAja+fi)ShK&Nyhcu}{2G@;>;?
zns;B5cImvDu;B99l@tEFe)>*#+QG-_*XQsbn_XZ0%_GX~+yORi>0`ofze1*$O}Dq*
z%;1tD+@`GXi}~Om#x=X&7Czaqvz_&Bh<43^rYa%+<&uAV%>&gFy)vz3p6##u{L0s`
zRQvlC;oiN?b^DhlxU=3px8nNi^4*mhFM>Hdl%GkyanJv@V&@g7E%{I1&h=<DGM6cG
zwD@kna%t<+Exk%zL3{_^tyVbiBT)ImmVIgErInR22g5R_FVZakH&L?8Q-pcPG$+pA
zORI!$FJI~{I>*_^Bh*E4*}XM;iko{E+itpBWUqC0`_A8DlXKpj-j|ue(X~_JZu5@>
zsqHKLlKZYnJkc!KuTZeNaGw3N%-Aia!QMvE-PxO0%zS6cT+qnm$*gz5;+1ugz13P3
zft^f-TZ^~G+59>1?a5>>(Rmi#G1ETB>zuPZ)KmKQ$FJb0dnVnwI=O|5<FsJwH};qF
zPi-pPy<z9&<diL~DFPhj3+3`pc^v$D);&LOTE5z)i$~4~=yAq}onOW%KQ(Hz(Vw-7
zydoQWl=aV5B};s<Kh^zwQp~};Rmbx8ggl+x_@Ck0xr(Ezj?1gN=jMgaRI14H-F#f-
zcJ#A^8-I2@m-w@)^+0x7bY^_Xq}8Gd0z1AOjhJPwW-(*=%}Y*k0R<i`ld7vjijU^%
zKJD0Jv}<La<FS99>eu`1rc6~&*dG$mX1!v@p0uZl6Fnx~Kl4-8f1P_?t$ozyWRtQ_
z3#KWbyUMzRM|kp%s)Xm3Uzzv%O%L02?)v0o*Ir88>YB@_zRd9Now}`V)wj#{d%X4O
ztE}@>NS<c%$<N~W{tH*D&8Dv75BhhlIrxm{F45Q%OE=E>`(}RNuO}gs+;z{-+L)fS
z+5S{<?_o((sY7kT6$g*~aO#c}Ss=r+;eFp+!>_9*_)RpKq^S69j`Di013l?~KKFgq
zo89-<bHDbL&FZ(F&WKun;=|#|`*rp<9^YPnXwl5vMW4ldKb~n}3Mn|{@mTWw`sbe9
z`CZy4y2aw^(vLmtndfJ5JTCK^*-^>Rxr`H1+sadzxHeeYl-X|dHNVJbS{ZP=Q#W^M
zzTWe*DJ`4bzk97Xc;fiAl?%FliA6m6c`7evg7Nz~^;WfK-|jMg&m^?ww)aC1zrtUI
z=c>3S-THS_v9RnCvyvz;*OT01tgchPRxz*GCGo3yzrkTHn@aV+9(Nqy&)YwJX?n9s
zu9<nKY>3r63!DDamM`DO@|@?`7AfVV*0ZTCanf;y{=@qphc0xR^;0&I*~Inc&5c|8
z{eK8B_vbIwUVY<Km}b1I%(rdtZFChA9&CJibyLOh0{g=s&iL9zhkssP;?Z1l=H|)Q
z-`y<V{;OE~Z2PQd*Ey9d_dGBU`}+1u;m^>$dScE7A@{d@k~HV9N&nAq^=S6FMw8U|
z$R(3ackR}@X1MO+_crz-tLJuCI}0D&-+b3!<RJTZr=D$v6Y786`+Co3+Oyb?VvSP1
zM;3jxc_R9Jz1_}LlVaANR55n_wxDsYgn+X1Mb8&?AO7qP4w_OYblEeb$f~{lX0BqB
z`b-JlrBgkXCwVa28b!}vRuwGnyJOP*H`^!d*=WF|$*J>GQe{r@_5P3VZ*-qnl94n=
zV@F!XhXwB*Y*$J6_4w<`+*_*K^SAkg7MX2e+;m5Vf!{~)*EO##moH>_&N*Vmd^n$b
ziQfag2W#ryxfLGYe{I?2HK|*ZwKnKD+!y|M@Rj2+!=F+wq8=npyC=z**fY=TAG76u
zhS#-O9uL|cKIsjeqxH<kK<BXKdG@czv!+_^H{7<(xa7CU@0f3wnG)s}^eQl3`1072
zXXo1&RYB|9b0%F2X>r?h!Xc^1N&b8B@#S%Us=~fxeS0`bgYD9f!<My2N-bWnn+Cml
zbg!~S$W8Z2*}X~&_P>kgS=U;utz4RGB5t}(!+G%?SEa)&6OA@awRoPX-6QjDz3=4|
zDYNNvvzNay+8~#**Jq8x&kot$jPtHO-?aMWwIc6o`*$~1<+e`c5}sG&?!P>0yX?^`
z*G?DOOxHimd)zv<N~q%CasSU}16RJ8`*z3UK=tA)Cc0S_3H!6VcP#I_&wR3!`{~>T
ze;?i`vSeQ#b!y)==d5tw>Vj$J)&eZLbGs)`TJrVvm-Tj=_XylMD--lt&EVj_e*w>L
ze1H8qy3Td&(>3Lj?=|WO@$0J8Bs{jRdOSyN-_)x-F3Y9GR=-S&bv_~{=OXr<LwS~X
zX;bbFozu7HIohOuT9+o>zdcK6bKX6VM8h859mkb#U-?i!)w8-j-<W^HrnX3P5gmyq
zjq|7PkV%o#vdy#K5_3>xyL5N%1djd_5&s#4CKR|o)1I_u{YPU-zh_f|P2wuVPh8BG
zJ`x}n#*)iCr}+3z%lBuKVs6}MYSVr`>33oB)K5LNg%j>fzED)VF^bu}Kd6n{Z12QS
zqc>tc$u?h@E#IEA-RN;Gc+HAK-jZu&If~Vyv*tDLRQVUHxhwP7`$Cg>Pi-%Gbxks#
ztukqX1Ebf*=W~8PXTE=J&5x{zh2jBgHm=N>sKvXc;@pM1(=BB=R=m|SsjlAnfZL{f
zdv^3x<qDZU>kbGQEnfdT<mi)tNRLS!fsbdisai*{J=YO`&v<<OTIc5OXFILdo-RDU
z?OEoYOB0L@A|5H^o|kysx8twY;hX2B+%9;fY@Z+?*|oyS;z4us?Y+*G5})JWJeA_@
zR9+x$;Pv_PO7_po4zJ)$a+KYDq(%Ob!imZ!&A$rlqn<zARXAb&ow~IBNm2JAZ6-fy
zckXF?{r;?IwOfm{;&W%?lRKtqS}?0m{_)v0^UZVJ{2#3ASF2olo3F3?E-idVui(T;
z?*`A}2L3aA#a|D+p0(OyrqDZ<-MvpkDkLS%PD+2V$l<TM_jS>>f*W?PJykMqr!#)2
zJo#MW@s3q*^$xw&F}ti~oWLN?QaI<=lgFP|l`6}Z<XyA;ZMMIIZI`FH=#jjNq#OTq
z3hEZ`%~(5asfn)cBO85n*$-}aK5;W%DSLhY+V<S{QxB^ysa8wgXu$b%<Bh_fR{~?>
zbj&w+nP>Scr}wn|=~uqL)a&WAH(^FkyMB0WVvBrx)ojW{pLB%>b3Sk!-=CyqJ#~Ba
zj>OFqQ!_atE)=m<J?XoDY1x)7)y=sjDxPMORPG!x*Rf>2Kd*hq+FIkoyK28)?DAZ^
zxy|(MGo7qY38gw#MOm>HS1zwmEl=su-YF->#W9iN4&!;nc6N37rAFM7eg`t^tp3#d
zW(_CvMp?(B*+07d`~7EGn@_4-km=YuPg(A%ayd)$vC8wdRae5rPN$l5Kl6Hb_DREW
zEv~OQ^NNFB>`mGtGC4|gZQQGO4^%gD?=12boVR2n`-SwMSBg$b#d`8CKK^{=X%EAW
z%kzpSPkWHSe{4(1ZABULaG{BeGovhyJ)AW2^aja4_QEd@=dWEW@zbpJlEC4Ti9DM(
zzA$`wXWp_cJyxgdUnkynEahFh>`=~vkW|K~%nsJi*SRWeJ{ezIb^F|jRhxgNiYuRU
zGN?G_AzS?R{5~zqwNJc79r8|V3r)OZEtYt_`AUi1RHtW;z8w>tdR6&>o5%|FjR&7g
z{z}_2`D*ln*$VrnxGIJUO<^&RZh9BPc<>41(Tv-#%*+F}7_45ivG<hhh6Cx#Z3O2B
zY3u5_lovX#l2q=!B5?8GGreaouQZ-K=V03}YIFI(<y+C70s=?M#oUz{9vCDzJm;$Z
zc+NK4;_$S$U%5TAqNllrW~S`uSoQSko;AsrRe1g>E3LYDca4$KJKvp(@`@V_Z~S7q
zYd2S8PfzrU?u<#E!iEL~8&7aZXum!A*ml+CgDJ}K3d}5rHSIWW_!V);9z5|cCirRL
zm8j*#b=M>3a{1^jsZg3=6|H$oCC<<AvVw8KS4q36m!m2kz0xcfH|EjVcG*&b`JJ7S
zQPscF!;uRw&(aJ~Hd%T=|HL^z!|OfCU)OTvr@MsiXz92k#2IhQ^7oGLJpZa&qUx;K
zvz&@LFK7Kz?T@?cBUssYony_4<rCHGmlah0XOQycKm7T}tWUT7udNreuDIeU-F_$4
zY5CL_MutuP>;1O&FMd^K<+SzcfxG2u&jc6#oar=wN<uoj{g3y19oJ-umkT`i>5Arb
z-hFGc;bA84TnpKLpI@IFvNNte|GxaYl-BgC^G>fkn*Q)SzpLu>dCFanPkB7H-f5Jq
z=r1WVhbJW^ocWQi)1k!b(!2$S-iSR@O<2?NoNw_uPRXyb2hPj%Sp8cdbyx4n#68o)
zTc>W8pW%64R^;2!q<dNS^0t<%Y&)nIrTX-s<{pJpQ)50J7JS{cYU|E_f~U5A@wn9;
zIRCfSr|bGpdDm^~v<rLQ^Tk4DV*kIbp0(djua;hKsy0zQ^wvkSKfI@yetp=<ZujT=
zI@e{>&-Hn(mps8b{}jsxW!e7XhU{gXOLlGFvM;=HVgk#l#fkqJV*C}V6jlB+?0z3}
zadwr&XS0;QYmS!6Y~1aw^ykgex*zQ8?9az%?fE|O&?ntf@A#H;RV1{(`_JI`^H*5t
z-s^Kdo_n#(;_l3o0g}__?y&vj?)*0NePZ?MJ8LFsZ<?vnGiQVD4gQ2G8L1iZWh(E=
zX34D9JKfn?@%G^<Z3YodhXci5C2fmpS<_D2ERT#i&RM1tIU&XUep&VB?e>4GRxXJ1
zxESNae3vy*?tz8I35El2ID0FeueDrMU9CNHeeM%A;qDnzH>#|jxII(yz_~k5%4Dux
z%G|Oj_P6*~?k(FtUDBxO4%23sIkO{1?Co*MCylRzbtbup=C9qhs7)zSMY>Yftswf9
z@vWV+-$p(1je1sHGk=qZkc3O5K;yy3^Zx#=x+=YU>C9=mPRYhk9zXke;7guC<qO6o
zcU#XVFSj~a@uT3f{<$^M)^1fj$?|^-y{=#P`?z(JPtV0^6OPZ>R{8Sn#DA{It~}ym
zXTELw%qhU8IGxe&{dCFKjm_u3*IKOlCN+Pe$)abTi4E&+GWSo5+goq*pz8TruV?!f
zho?KLE&AvvK3%8yWV7O@bJ6M+K@R8qvZB=|eYz?XR;bA`=bd$U|DGEUzRRu_x+b!_
zqGm@qyK13u8}mLRi$BHTDt*}}*WG;NyY2C`boPaZ)?b$WTR5kDUfFlvay9Q2fqNG7
z^t|~O<KgC*#9{fk@2ly(r6NW`f_62Fo|-Mm;;^urx4*+=mrpsb(VBPNO1l^JK6q|<
zyGQun#XnhLpN?Gm+30(Esty<9Gzpdl%VR|ahF{-%M`r}32)lpNzF4fZ!-#R3g>75G
zgOsl?kJqFx4Vw1LVx7#AbDKQ{{0$kEm;258&oH$ls-$nNqMn#pUceD1nZI{5e-#ya
z1Zk>trL{3j`aaj<lfEqTpCNAXwf3Ml)6Qr|XSZ!^InH&m?XKD(zI!ttzYt%Z6>TB@
z_S$;2XFIt3C%vB@cwRy#VSZf23+<|kGZSWth6(ej1T;%u?^(g{@%?qXt5tWE%~q=z
zuFm7!Is3iv#4W<tf9#Jc=zHVwZ2gxTXHFK@GRMT+d3%5A-R*O}z2XnPH7QL#Ju^2c
z)Z_k1K>=y)EpL@=`6a%}&&|9O<*fVZ@4BaJ&%NFFAk~Y3BkfOlsQBhjw=<^bPgQ$-
zGUr#atlj*mpzH+_+s=HqOS*mRT)@xX9e<x4<IhPDu=xIMrAe8ma*59V%pPV1g_=Dd
z*k#*{e}-+b$P`^-_Q_La!Q`hrll(ZIPd<4(Z%@dk8#iy?z9)NhQgo}ELh&4#m9H;+
ze0kJ$a=1rtxu$dW4#}IxPn8{5IkA(^K_y*f0^=XsR~F8~iZgcSNt?}CXZ6tEufVD>
zTygGNH8pW@HRkru;XENlp2zF%w|a~Ho)+*<GPcj~*qzlWzmxkf^-71`DR(Qmn7d7%
z#rU)$*CC}MY4(*ikN00&xbu?hogjayE#ZO>%lo)LRsCnE=#Trl>R0R9jk6S)E3FR*
zF*jAE@btCxU%7l}%jGxb2FKPgZ=Z34`;~j%j_<3ctXZ>N-fWseu9c;ZsLh900!0z;
zeEQehKYtl;d-`v?YZ^BU*F{h2?BQSj@O8Lq<ATdi`;6Ah-;camHmAwRs>tH~+mI(2
zmn4<DJY$vZwM9?u7T`So%<`|c)YQ_A=Gz|663h!to@G9Hl9c_=%!ywQ{IUKQTA8rN
z-Mb^s-!w$<M2?2N;eUpGGv+Ti8nSt-RoA5R>N8n#7A%lV`(k_G;f8D{#(LJ(4`k1^
z2+T}ke6dF+dBySl55tyUJL)<m>bcC4iq7QUEtBUxnOFV&Kf~+GInSmpnXp)=<A4Ek
z(_N#TGJQr_M~}|v7Tn@1{Aa0fVc`VJyqQ1y{Op-81nJE$(_5#f7?_`BUon3dSFrMO
zTR(p1m&YttU%M{1W{tDXB$YRxj(nWV>|ric%m3S1;_LF8cRVYTvlO;Z?<`=9-*X^+
zetvz<w<WV)vaY%8bYM$Pa>b43nJiNiUw9n<&#-FU;qv0LZAAx#t&+1gFoxT;-(POO
zUexBScTm}drKSzuk5+a}uU^ry-Fm0OoZrv)e_1Q*bvAKpHg~e|W{zdIs@|VwGXC*z
z+2-D@k6x8e`&4snlfjG=-l@{ry=!VKA8!|VvHrcG?)2Hee_T=B(^34qO06ng*7lQK
z|LP5s)vlT>-@0{2*SBk`TX(1#nmGEOJt!et{3B#-zKT@s+<m>WyP}SZ&)xdrN%Ki%
zWo8Roh9^>It~}in-LUS&+wEnlX{9Wm=V&~b#2%lue|fxVOVjDo>75t-4t^F@NzPDT
zZhP?0s)uHOCj8tbctfY?R!rrFN{>6%E(XW!A01AX`Ogq_VeeI)u&at2`E!*RP6d8=
zI9-yr&(HCM<-t|rV(ZRk7Z)VY(l==qu{`ylW{+Nht^4I`i#IiSZ{NwraC%E??hY;f
z>+?U=76!c9-=_WNKf`*3XIf{U#>Cw@_50AN+Yu)mH4ZctHs8P9$C~~s%0{!2Icrke
zj@dero2S?wWDxQ^aWv;_z8uG`a{iff82>0d@7wcd|JsO2zn-OXv)P<zU~_6Ne^6BY
z>v7*ztu^_T27B)<Tp7VxbK_0iUEL>+TCN4LlJ12Q_FDB6UY>Qn-|xyByJ8cgXwBcJ
zce&mwoO(h=u<uFf+Q_`9=ib|zO4Yt5DSwZdE_v+DudnJavLY63_n7ioCCFvriQe|v
z=U3Rsc}jd)Rn+@#>V^Bd@6tBgK2zP^B`GYylXS;_Upe26MN8%#3akGy-D>SV0}<g*
Tzvb$tgxhbaoheq^|NkZcNIR{#

literal 0
HcmV?d00001

diff --git a/images_test/toilet_paper.png b/images_test/toilet_paper.png
new file mode 100644
index 0000000000000000000000000000000000000000..d9c4ddb3cb843df5e1a2d72014d0284c476a80eb
GIT binary patch
literal 10660
zcmWIYbaR`d$-ofq>J$(bV4*NUlYwCaFXLQBtuKsU811II&Gk`TymE!|#9tQ;ggP_x
z@;=wJ-klw@?#;8`KjoiQ6_iZ8|ISZ1=8xc?_Y0p)e^UJF{<HtTKCb^d|9<`Z|IT-w
z|GV?}ul;{HoBy|8^#A)W{J-)4{29Uj_sITUXKerP<Ej65-z@*XZ>4?H|5tYDamQ}Y
znOiNLS5*2Ycjw9L8@D}I$0Fowyr%Tln)}kB(gzitR{Wf*C;DaWR|AD}n}vCpB_G{2
z@VEE5wvpNFxq9qXmo2{<H>~N5D|6hZwf)H>o0r+T#mlBv&yKqM&p5AO#j1{X=e^Q5
zxO6mEXvR<4vEco#hxQw%mtUT-l5yS}k@^}fzm*BRNo={(%$G^ui8!nBR(*qk5Od-C
zSNr!(QIHE{)4XDAcq{vhb;-;FtEy+|+?0yEd}8DN#UB~=u6i(;dslLm{I&{5hohN4
z`Zj1;PqW^1Dd_Uw!?*5!Po7$mG<)XZ37(%K`*!)<Uuyntr-=mDrw~5l^8V+={p_}E
zd;D&8wXNu=<niAUl4LcxO<4Vs`_FZ4%l%Vl8aq}W`F1M*XuksctEGF~xO1Y?RJ~1?
z_(X7VgnO8+T3B6>o#Ypwusy>|nNRLh|JCT_^K;jSy0BO9-~KJLdL3KAt)nX#4|uM-
zy3RJ_0{5gv|4O;17Co5|X&|U9SS+jlHSTGY#LQx+tmQ2Wlr=Ux->gwEXt^fEvN-w9
z3W>9`I2iV<V{q(UxXEe8<tHii_aE$0yHKU|*IeRz5{H#g!RD6ApzqGB^B(thRV;aE
zU$sv}TYhPDsLk<h)d3Rb8Oq!BK1D2QF81O7Rdsh!)RZ%AtPPza3(|#V@v~fte9~zg
z=p&WV?OuNJ)=c()PcN*!;a9xkRY`5iiC4$6O?TT)c$HXFaW`?>4UV#aiw^Gm@1}Fg
z9R1fAy!z?<oZBhu_C8~g@j7#PX_?N^*0}~%Cc@vJw0_<gT3B;5?!SXY!^Nxm?~8T6
z{mtyT$|&N~`QJ)~`Mq%L_B#!q<qMykw0L#XI`x$Dy=U{CcWym$eMPDE(OoI$IQQtZ
zU*}p>$y$9yT!~Hbn27WJN5R(J`@c#~(6LI4HJBow#Bt+#?ZK~UDG3o5>=v&udHOmg
zddu1=YcH=2j0|Nsly)t<`+`Eq(Hf@ow{71uvUg2TcC`$Om-_d9XWZg5ioCqZqCr~A
zmPP3Pe!p?srwcA~elVEJ|G)IL>6gs<rc?#F<DSi;X01{x2R5(XbW*-rVG_63V}FK@
zzI=wpPp>#;2JB3&{t_pVIEz<Rrtf?Aao&*Dt($7aK8LB8ZAxkUx9I)hm#k&06pT{-
zTyrV<&7Z@*v_jkU<oPtcw2#Lm6Rt<epV6LhFY=6;fx)H|8b7tJEO=V?^yTy_iz?II
zTj#OPDHYI{?9FIZIC4MsF8l4iu)`()Bj&2;@O2wpmas1@kiGueAVYa=h0VJD0)=J2
zxb7;7oX?rPZdSqk#d=RP{ybxtH>-F-+oQOv@{*}8*?gVrm-RC+9Eq5(@3S}2fBz%R
zC1(^A)6VCZzkFYOBrWs8B2f)a9{r3nJR$!*iZ}QjH8EfN`)R`cq7@BKZRQ?ZBNBf>
zI=MagY*{qtvBUDqdDBeJoRU-xx$-?p&hgB9tNyxUax7w#YZ*(<F5PPKbN=<HOOq%3
z72x{zdUIE6(~fRNjuu|O1=}M^y(TNEvwyo@Z~R!7TaWR%+R{f{2Yk=%SkZ4L)cGRs
zhrwl`kheK)!n}XYttX}YdNTXYm4hueuPFC+B^}zZL2?ntwSol?7=v%kJ}PO|xX>-8
zGGyZ-K|N)Q(-%&(_wT;&$xxA#e^CP)>&&pLFE(e3J!!wlTTxt*uU5$PdR0`e_g2UG
z=Gmex99=I9Z*;9cdbzxIUW#(oj}@yvr{z4*x+Q-0?bi9nR5z$*c$qI*-11~9W8SRF
z&Zp<vMXa@U{8!!_yIoY@@bpQamn@31Z$Ezt|2LQY2=i>64dutJ-|L<}%yGc#N_bCY
z%lFF3K4lvpX&qd1j-{-I-{-uo=0x>J%PsOMTzzvkuG;#oOLK|FZuU2~UOc)dbYtp;
zbBqs~y8Alv+>h`4TmA3}Pk)iXkLi&-hmZb$;iUZe_>8_8=G=UIbFYf$3WazcvB{av
zlPaCzGK2HcqSGnQui8I4ey8ToDxav&w!c@YDYwpue4g;N<Bihg)fu9)%f2mteQ@ns
z*-g>W{jC4`9WMTK=$Uz$JLkYO)g>1;OZXeTT5D=*(6nR5fti}jw~aGH?H}d+y}+nq
z(!lO!@tb>fl(^fsGaHwD{2%>Y!12ENa%*emcwhcb6UNl7af~vyC+0l=J@fPy1(q%U
zS3fU4sUG|(*_!vruEGcF;tKZ!GbJomKQRAGQip#0h2^VnuJ&4>u;IWY;eXX@U-;|!
zODaFQXPWYB!k_58D~|hALge3y?e2Lu`Pb%!G43ZdKe>LL+jAn(|6RqtgzxQY(&t{s
z=AK?Cr`zQ`@m0jkW5J1G!3y<T{u*d7O`YLzUf|b~J?RHzUsYHXecx0lQ2Fij>5aeB
zMIG)e)_xNA!giqvyMD^%yXG~K*@bV;z5Tu6lfBdi0oN0o|FD;MeGvDG|J!1_lyCVi
zv)ji%RxR1+m#&~|dw6Z>wOwiJ9&h-WShkh>o*9eAZX3g<XM6e!54{t-Usmu*%j{R2
zk#fBbYv!r1%RV?h4(U1fd#m}u3-%{t-%d^~kkp**9_zQ<?}K-4neyYkyx*E)zU<?F
zb#L|Wb-mdF#(Bj?3=y$Qif5_vaqiY@*7Wzh#L_fRr{s`@W9jlJ{qKr90*gK|<UCtn
zEdC<tO=FbR?@K#XZ=I}QzNXTYQ@sDbY>Z&yf648ZR^QC-Jo$e*WztlkolEQ*9QQX?
zmmc-E7Fcrga<0jlj|yS?Z`B{mJ?FQ#7pnI4joZbO*C(S2Enr~y_wU671}Lz*l-;~%
z>dTLhx{N0Uaw;8}$jDcgR(+~RVZP#Z<*)JQCmo%uv8{^TX@|j@CvhthF8Yc_c`K*P
zdaiD#{<TMV_lo}^Cyph3Xt&xMHND~E#vW(=*&hX~PVw4xe=*fD3~8>s_l|*K?o2ID
zpF`I58!z%OD<&2OPmxqQq<_`fIw1T-^L3M>vkr*;t(eTPZQbh~3=GG2?PST775gO4
zoW#p?RN~%-6CTeUO12ay-!f#_|BP?arQf%;@+RIhpM3m?((LG#&u;yEj30XypNVwr
zJ|>~S**>GW_U*q*`|qc={Il%+Vj*%vQtK)2*5EmzlhS@V+{))-U<e4<=>Kk?>mKhd
zuBFF+z4~!`ReycDLa3W;Y4l+)!+i-0gl!)ub{w6_WZZM<!D0dR55i}4`$dzg(zYM^
zeB5MJ=7)LH3ukieyrmvAE52v@X-@w0$Gh6yO<qJ4+^Ac)>EWy(MK;ept(Ml+Cr@Ah
zZSVFnKF4L9xTE2rFqQZBT;%inOS2fG?Shn9i`+9m^W6K(VEIbg<YOLJayU<fjKHCF
zE~nN1b(~))-t<@F>aXLr4{ZhZY^{!5G%2aeY|+Qcb&FEm*2i`C2pz6f{O-`o@hIw-
z>W&ky(w7AKUA&i<UfnhI?CdT+(U*@8w>b*ct6g|<bI%v)*E6PU?b_VmvFSq)>z}ag
zv))Zuqp)4Lr8eje=K|%jV~?9|FMPJfLgVYY-G^tE2=E@M%vZZ;WZlJJdE9JH+oXM)
zbpotJZoD^VDm9iWSMoaC_iV$yQ^#$WKXh{rm-tfp)<{QeS$B1KS!@L>_tB|Rt7Ok8
zrWSFB*<4k=Qp7UtOX<9b<@#Ydi+kJWl}^mO<JPcHYpLK`rSI!XBX=+KWj*1g*{Of+
zC*xhuk4H}&{i~bQcQNFYP2nEt30j^`$1b;NRX<<6p>L6qVoX@L=$nFEw?D-<&zyG*
zW@V6Hl99%A_>P3mz3HLb&hp)jxMaV3_C=2Zz17Z_TCZkYx~9x?(N;<5U0?FT^l+nn
zp=Q!HR%dhg`v0#G>ymiA_I6*&u08B~esP@nWtuR<I-=F>=S?^1<iuwuKFF{xzrCUI
zRn&=E{m%JE8M7+XuL{g*TDB$6)VsvT@WYce+93*uTBZd&?vnKBn|bcxWEK5wvJ;|Q
z*^Wr)m`Ir!mj7LvW|n2d%d7M~>HepUja`LXf(_flE#+;E&0k5+K3e#wy-Y*QCx8Dn
znWI5pLhQCmyfoT1smZ?Zf#8F(WnpoT;$L6ac(T`yP3hmBlWad{nt7&st({R+YVj-U
z))B^nzdxil-g4oqKT;L>Rp`?X5n+Wzm(*lV)mc2^674Crz4YLhWBH2gId>8!O<ljX
zSZUjK;rH{FFlD^{A$#}Nrd~D{=H~`~KUc0--@R}{mdPtlE4Jx&!o?d-nkKT&J!n5C
zj`#V0|B2naadljEYgI0~FKzgroW3KxXui=$pFcm=>Iwf1*J*nu&R-fmIaw?2bZzAA
z2UW)pzn^XPP-O3&M^Z(GC**f~8*H$jbmf@AF_ZAeCz?e!A3OewDPhrukC$Zse6T(F
zy>vJ4QwH}n{b$a`M}J(H6@G*{hhJf>`pxdVQ*XUbTneB2y?<5w<Oi&;E-7DHmFxL)
z-I4=ycK-QiCv;n>Au~Al+tJqns?YoxGMBWii+On3tTI<!kLjM+-t<79P$B-G`TK0I
z&40F@<;XRKYA->Xvr~6pKk&dsM*B@yXkF#oo0rz^eSbv7R^;Ns6{jzsZSLOPtTXNJ
zUibO(ySd-a(mpO&maz9nkbK^gFfrjazcacAwx}Qd7bn2FST;WT)wjUlgVk=5yWX5p
z;xz6(^``iqw|8Vz8;ivXH$IuWH!e9Z$nU;aUiebhF)m2p5|gV*fbA=xOMHc|+V~h)
zGJ{&?%y7B2mA@t8>dEc=FP}<Frl^ZtR+D0wvc9#g@QvQ(Z6R`;s-4V26J%~iamh$*
zk?)GWcSM0#;gR?HuDa}nGVJRdBb5%t#OF<u)0xZBEYuq1rr`6wbl%@(ve9uj@8&JL
zTyV#!X{%1M@!CCq{=LgPv(%-Sxy<t6H%_x7fzMZPzg(KDo@BC{^{e~C<_#TZb{Q0h
zPv04RAb94Bolc$6dje0%Z(3wFC#A?Zm(QGQv(qi-&+)BI%Z>N%pQztmmLaF$dcS`Y
z$L;@-H?pptFIVGmWtCZXu!4D#+po-*wuKz!3tGd<j_vyNrX%m(uElceC-Ct(zBX%I
z@R!}RIHP>`%wu~q*1K;h-TO1KUUB!_^N+g%M4!&vcx;x~l4+3-e$9VxyJ*JE`%=7x
zLb|<8+wY`B=51-HHokrG=c5N76w}tqU#?wf?R|iUvF^_9H?hxtpV6~WTF>0NlX2c6
z@h9GIWHmRi_<xT*d;e|vr<{gmVKYm9^|&khzu|hUYO(*QS_xOeOHr8^kBb}oFI`jZ
zy<5KL--2VZ2Qq6F`krUZ==<>bsb-t8oZ`jGpOV>g-3~53p7(0E%fzOfjSJ5<Yd1RC
zTnyQ8O5ZhGN-H{*<4kYvv`_12i?RJU(0bBp%aPhPEm!>l!7~q68c*D~-g(LE3g4Fh
z2R>_8Mc3S{tC--!_U(|u(a_~bO0RJo&gpXH7GrzBEqJ&lgW=@YtHKAGH|Fr%YzXc?
zG2L}iaAo(E8`iC#Cm6i03%YXaba00t<F$_;j_tU0>`mEPi9ZqfQ>@xf-`y8ulkHhF
zv9N{X9NWP>(ciU`Hft6fO>o#<nsS)&Moy13o6fJP>yNw6m=oVnb>!OxtvN;8j*GIr
znWA{pEu}-s#=_G%!eDVW>!$C#+naeFH^iq+_*m(ua_J(M1Ap`V`ow$gg@1*P-IcpB
zx&LgfedC6P%4XHW5eDVlp_=aPd6Pc8tl4V1U3f|Jrkk$oZhV>Ue__^gwT(ah^XG7=
z&1$z5E-pJz9XdsnF{Ebxybp5?!+6AwK4FdCxU2K;_S;-m$)TTI|F#NG**|OjdbQo3
zUngk)2~x<D`Lr;{;n4PC$1_*hFJ+uid7r~sRTpV_=iSAXd%sn>zI^=sg^t4fyp5Am
zCT(BveBVvp=tn%9Y?lpZGOu2l?j4s>Q`e`rb4|PU^<NtfhGor>visHjNz}A8L-oRK
zZH>jU#~U@*1YK#Ksc<bZVaEiEjeBgB?XA9Qe_g-(fS-z1;W6#6%U`O;DbIdnKbb`;
zySn>`+ulnhK82Zj0uNlo&t9~e6uxuEy><nj?t5GJzLlO8Et2(iCcpRQ$Cl5c<dS!5
zJUV>uxrkut8+%rxX<L}Bt6NUXa$0|KZPZ=5EA;Qi1-v1Jn=8|-En=!wuC2MzxaGgG
zmVfRHrhB@I&is*YOe9Xo?S63G@}76IJEz>%r2J{_OFv!|_+83*^NNPjyKASz;se%}
zg~V(Xst}La@wDiVcKNiYGk%{?=v#Iu>A`~vf!*^L@ufWx*tv5?>o@JAr5C?nZ@#(f
zci}It_t{>t{@0l^XZENx@x>Q6E>Au2vFlum*6|z9mc1?yDf-xXqNiukqSH@rFmir(
z+WxR_pWuYg@oQBJEBM$#nW~DCYIPqmtgL>Yq+_;Sqj#TmUdkqqr(fo8Q)=lt93Hm#
zJyXi!O()X0d~2N?qc5G}JR|OVJuLLMtCioxyz>{GJO5sPfB)p;n6Jfg#Vhi(H=91F
z+t|0>#@gQYzoE$<u?hR4vMw{Y3U7LU;`QvyhaM|GpLjIz=cH$;_2%l<I}f&>-{7pj
zXX2Z*MWVGv{s*LDb_cdvtoE+b;0tHz-#6{tj^rW_>v^s-mp}QVXy7=xx!U=ckLk<O
zxN|%!vK#K^HANH#m~3NNb#2~^&x)nOyZR1D&)*xu`o-s;<h;2~wX@YX{$=}lvgBu(
z=LRM>SB6c29LL%1+Vd4Occs-yOf>g;x~NX@T5=<6(AhnPp^6hO`|*`buzBzA;-Q(v
z`F`EU#j%XD%YMY1_|Da~@z~2B8b#4b&FiGMd|70{^<@8*%-S7HX9}h?ob=r^e@@mi
z-n)mkTKqoCp;EZu?vidE+loiQ8>-H)+OcG2!LqPf!CBlp+lvk{zjr*6IrG)(%UbU3
z0)G@9Y$)7){+QtUAoZVZWeQv7y_MM~dcKggC1ln6hCMfKy{XYoaI=i;da=1=J>#6e
zM~<2<<*B-*mE*=47SsD>gP^6-<wBkavlZ7L9p1>rnk+Qq%2Ua>yKWmM78q){U*_6m
zVH$VZC{_8m$>o2!EY4b=lmF<<3vS{3@v`e_P3c7y?+%C6)9%|}cFne*yG!f9dgD5-
zl|sK>H!sY!k@fd8O<ekY%dRafk6Ku&+gq6ALs#GYcDaD*d((#h3nS!Bgn87a&XU}?
z{wLoYCtfY-Px6z_aP?kVW?A^w@EGg;uDoB8lXtz1FP@gvk@NPkX!E^m`Sbqy$!*tM
zU3WvmB<*iwzWPz8@Yoew$~M*4N~H7L%38RTO(DebTtRGTMBTLt=8|iUcUtd?^_hvS
z-*fg}SZK4+z8A|34dh+#eXpp~J!YSN*y8Idi&**T&PBTaOE>QNDA<3ZAUDnLSHgz!
z+>p6Gmt{^A{=c|=`m*0DX|vM59v05xndbL4eb1`&n_S;-u&p|NNBv|(qg8HnZSwDY
zHmL{YTVJ11>C*i;<8_JTu28#j_qNkT6SXu>l+9fBEaB!QqsfZuYriJMd6^hJ5a*J}
zXbJ39uIc-|J@)o;sWY4@=}I@-KQ7aFKcRl6Tx8sbJsK;|X#6q=E8clBLhH3zo~jK8
z(_OXx8qbE=|9VpQPi<${$X9Lam#yyM37%_E!jzyTv!ed%k9m0=d(5|(xaAmLSzyQN
z7~6d|VD-T%>D_gwj{TnK;^-JEC#WOXezd1pYu?k(Q;vNSS^97G%<WIhd}LeN{5QY0
znUejw!_d7niTRNIxwXE>9fcFVlmrSYHtc)6cwV5^<jwh+{oA%}G0L8LsBnpbh3{(-
z*Ug`}{+)7OAYA6f%~hz>D*N%$#q_2BE97jh?dga$yMOoHMu!=H8jGiIT*oeWZk5IE
ztk9U+{Q(#6H<aj9N}bJBpOC?`wC2i`Ik!*xzwdh$;afe6{~h~V9^sIL|2hhX3w)bJ
ze80V~l$F|+@bB}?!>?r2@29UlUia8--t5&yBC~cIMjdXvR__<Uo)d9Gw(x5D-hC59
z&piKjEnv^AevO~1d3DEJf9z?Sdhe-T+>X_6_cL-@PF^xMNXp<P|0%f_9g$tNiJgD@
zC9}U?S@vS(Lru=PuQdL#t#XPsF1hrWjo&IyyUFe1@AJ7i<wXw3=Kaiz`f@bObmRBc
zXfc_XyuQA@yj1hr*~M&0O5bk&)cF3g%>Qf3VZTL5{H#7Jf}d=zVN|aaRMKW%oAFtt
zG0P~s;f&_-sJ9<_OZ#Ip%xC?!xPO20`{hkPN`-d>r^vpq@xH_Ju;q046Lal<hrgN%
zwx3j(uYG31@-MD$o;BT$I9&LmT!OLA*K=b2hd7xcV~OjFKSr}GN)3<lZ*17Rajs~T
zT(ENIM6byym$h!Yto<CU%glfIA4`1js%fpWnEbEXDx7TIrSYxtr1Fn9$9^Zw3ObhE
zc2Tgrmfx#(b<p04Tjp(AQLQerz;;FGtQP5iszrgm2WD2U4&Jq^yng$utWB<uJg#T_
z6uC7ug!SQaej5R^qq9mjXmuASe_u8s=8-}HgU>3C^)-oY^A2ZqdBqq?Jxe||uRMp(
zRr~6{c}buC_}+ftz`09c;SRkAz7dkPoRJ+3hN;({oGQ$@D6_J8z1O^1Usy6sp53$C
zUbXu|e6rKaC`EZm*=crP)!ymY?+QLSL+Nzb+`@`gS$1z`yio9r{&Z#E^+F5xsHzCb
zWi7U?D>N<5ORAVc6uqXsYTnkv{C$D(rX|J~otpM-6nb^ihvRo&ZK#mcHrIbw47=TW
zpC^c)tNY&EU0USw?CX;yyKBy9N-z7$q<*il<oRMY_E3*ak8hjJJmhzxoMp${MGcpP
z(~m5DVw8Nwq58mPtLf9ki+g+Dnb*klCPnlmys;}XWxB1L&-2Sw@nhk6*6h&ZjNg}S
z5SV*FO6sBRSKWq2iC1+RT6G0hvtJYi%|H4$K+$7sg~WWxIm$a{NIAF{uWOFt-Bw!m
z=;F2I)@!azIAP$%ry6W|_1p6BOEO7O;cvRXq{*G&3wRi|Nyax<PisX>_KwOjceb(^
zu_EEpLtWp_GUeZGjo+!Hw=ZE{@(qpb$~~JtYR_X{Gdcew^D}l0O$NF6{%Wy?Wyd;<
zx89$ZbpA`+rKoi0jXU?Xi=Essb0<@U;%%n&_r5I%nVZscb>^of1s0W67RS95cHQ)!
zn0jmWx0oBZ-_MrkcqCqzu}5N^&SuV%cgc%d^Z%>-(P=)WZcx?tb-r+T#|e|@l6<x;
zTbXX=`oGZrIce7o8*{O&gO_atww_#QI5FsTYm3WD3+BZhvrmQ<8zt;y;oSBo$Wh2>
z>V13eyS6_s-8K02#n-UbtM_oO+`F|57Vqb5ex!8jXydI1opuwSri3vkF{gcux};Sq
z9(T%Oqn>`0Dc|Iy$vaIxPF$S-F81jEdp?_ZFa4dhB>uPIV-4#G0&(1*9Pd`l_|De!
zk>i-p`m>tXHiVzoza7c_QoBR8T9iNY`lWf%?B1(HD`%gHJpILax84=gkHQfl5oTLc
zImM@+dYW-qdGdzMZ?;`_c{x2Puf@9M<NO2SldSdp#cf3{>)r3^+8gqz^hAem%_NOg
zyK^@uIBfQ5FPY{1#lfb-g5$uV-ru%UR!@s;-26hQOe1vCjfIDkSGdjGGjYH1Z?%KL
zjp0StdOHrBFFs^gTKRNGqt^}opRY=NTK>lci)LK@dgA^nuLn<3{r<;4tdA?Unl@MJ
z?uo7+#l4TJ8z(kyRM~CEy?WBKe6QCF-{tO6S^Vo&WPNbLV&)}A!8Z+F?O;6hdKJ%q
zE$1ycAAfN3r+jZ(xWLcihstxWT_tk_lB50z9^G*-e7#fB6!zNam;X#G3@vMaZ;!fS
z<h5q*b0<~7mGaBjcC^`T&d6HzTVUb3??GHT9~U`)n*8r}>ow&%eLlaO*DG~>trT?k
zp2*C9^ew)1rsp+b_Mb{B{Z_T@d4G@nKi1VQBbB`P=k%O21tt><SFhrk(|+R2yU4d!
z9KXaAeCSrW_UVGd`U{*+miK~WdegrcYuN|0IK61RBz3M&_0g_xkEX`;?n~;v^<g2e
zxZ2tI@;@Ke#71QAzHpaWcAo1DHui&))*dos``qf3Ygf{c@%AgD>4w|0n#9k!2&(c+
ztha4F6L#Tz{L}_XgX?=j*Jd7>#4mf`%SrpWRpsn5PZdH+pDp|9Y*u(6zVmXAcd)pi
zLX|$(`_SGarDjWfH+>3M6?`ah`qmUDBkzbF!(cT_-CLP+U(P=+drGox^5JDmwD~MA
zUQoBOohYbLmGW!8mhHo?eas@7np<L3%Nt}1OcvbB`1s`4Jxi;eAiEzW_xaxZaWY)J
zao+aN(aynq+xw1+v9sjlrlv3{&YHQ#{fxz$f;T*uGPys__c;7x^7?74+;_Koa^!vq
zN%zpVf18}Z8S-R{(&?%7zbE|sq{4puic=oPzkm9!={l=c?q9UE>u}ko)_-M^rQ&r?
zE)r^SFMoz7`fSOYP$L?6pyr3&l@ca~6#H{kR;H^;89#G5Jlmq%_+lGxyTJL2D&lqz
z&GQ?D^KS436vlGyUHENM__pAbQ!!q%%h$c_|Cn*-v89LexdRgW8l3mqSuSP2Rz3N#
z$N3Y68oFZh)p>LlWPf<HvwY?>AKtQtG<~!G*UUO+99}Wi<j$G^UZd^h`@`oI`1Va}
zSs!G3dgCsEBjOW1CFPk~Zk#wZ^<}wE&SBk-%_ft$^tN`mxJ|h6`SDX1>*a4x*$E!_
zaeK#N#Wh*S4lP_4G;^WVn>U>c!>0T8*#20@#klLX<E{s$mwR4qVM^g&?Y*h@Zb}_%
zdhfgT?@jmm%{HEQ?~XO$`<{`xm_@0hyKVRS=^0nrLQZ9B><QqCYI9zD(D;1InMKE1
zj`=;G^mN;`Km$pOfc$?7;(>qTqgOoE)jP{5Wmu~gQNANOcKxEb^9{VgLE&vSM(vT$
zs|6pZ+uTm#QMkFhO3&E#Li{X^0$;v%{UbKpy2EB_r@j6zy2+#P`<d3IXU;obmUUL<
z2vK{Mo&9CQc75)XLR%I*XgIC5k)131a*a*kRg)8~{3{|}G5Y#;l(d{+`=D(b$nc3_
zN_Kw(pMUX-GtJw=)s4Ai8!qi#CiGEMc0Kpq>HUkioafx_@aCAQKHImg+VU4;qsF~y
zC$w~)`)W3>oLCZM`~Lr$rDkFU{rp*?|Bi*Oy*~RC-+fN5o_cxN+NBvw^rPpUX6_Qp
z`xMBg^Xqr07c=*jv`>X*Q&qnBKXu{J|E?t=w#aV(%4)_m``tfccBSiv**0HG*tT&&
zMZViBp7bMA-=-C1ewEAMe#60WHG@G(d$o7qg0RWYO1A9T?>SL|N0;5~(B9%3pKPVR
zPc#uv%lDbKKwe?$wpHy@QtVG!-xbgE+`_b4h5zXDyv7Yjzwkc3*|k18g57huTIrNU
z%Ns&IJ=_-8(OJZO&S*lBk>~0s(rhuxv$6vArEPssFq{4M`YR00>y}q?R`T4uQqXty
z+o_y&F5Wf<GST|)RqQTFO<JRBbklRgWFe3Bt4rQYo29{7AC-FYQ17%Svi=`(H_w>P
zdDVV#TvpBO@|_6^i8f5LIYa+1ni;C<egD7eCDF?~I?Eq4Z(6wZ#LJ!P*EmoA-*tAu
zrczDs^>4*^OpnZcnRdosrT0c)!CU#>w`(3xRC&E+snmbfZ@rJIxaRG$ntkQXoYv%j
zUwaO}x^|*>PjJNs!{_3RC5t~*_P=zpR4zPUAGaVxaid!}pZWQ+W+na1#pxH5^<Rg)
zkJ#4JAry9=E%e7mt~p{XH?J;u{#SI`!F?GE`5t9?=-5B_E_S`s_X+PVW!1)0VV|Xv
z4>+?PzI?GjOm^J~r|BtI?r&rk4)yuwc*Jn+d#~U?o1SC`OT|F8>-*X_?M~RU{POcA
zp3YTjYgv}B<%(uK+m-$G*E0|2XT7_k*sXS5S{A;Q+gZlmijm*ktTk*M-<-ziwvvGB
z{EH@Q=1i-Oxxb!GG~w3y)|Rc39=jJB+Z<{%ZJew1SBFV-Wtw3?{jb8A3>%#zjz{_C
zRXhlKBHen?{hNBXuh*=T)2pM){(pbbd`P^R*Y&5P=e_I6{B!v?Uc4n3<^SS(*uU%9
z?)`bryDolUHOu+>qn@cIn6dRt^b7kXQF9o#Uf;oU`QfX74VF*4HmuCrp}RCh=;f6e
zZ;zU;dQftrN!P=gb@>-l&rX9C1*hvJ)!gP>UA9Sh*}M}$CU)kw+Z5KEu!~pa32I5M
zkv{+I;D6(U-A*r8Pc_M@(q&fs<ejx~*8jP0>P3#(3iX<=_F8oLNn^vY_Ipx_(H<|8
zR`ooK*tgmu__jP-M9B}!dA_!*6CM_xkYHG)Dac#ym)dpuhm>w{b%E8nrCy%v>bCeD
zt-3K?B<hw?n)?mIyDulooxZ!gGIveWQ_;xNS57FUob_F2s_1o(A#Ka);0u#yq@EHz
zxo_rzq^+}*8s2iP)M2?|vu>Kv6N!SdD+_O$hPW=vo@8fsEXt^<d#>WHKCQw`Db8Q%
z2S3OLu3Y&mGEey$yNR(={`I1Zy*UE2ZYm@=S#{}aN*(*>>wQ<^!TgS!uQtB)ov_3)
z_ur-cQKBNZPiw}v+H1(J`WczJ$NHW_s`QM-M;FUw3P#_KPV4gvepPO~bE?ITb#CJE
zAEs1he31NnJLSiY%8h3~{0oj<yd`4#s-{~%?$24KE~qQZ%h1`G;_>_b4v!oK>nY}Q
z8O}a9^XbjOsp)6e9N3=V&a-T{VeF-|6Bks>pEgndV!f5GmrS2ZTVu`oSsr#T+x5)0
z=PRmBofG*)(TBOt`D}Q#z^Y!`>`fcnJ{GDZ-jlhk^=R@sCU&P^v-m3Qt_ePR6z;~d
z@6##%|I3(|t_ksPQhS(v>&Kh3C*3vMXW6ei!)_d8_b){L?z9srYbPB0wo$AxDl=1I
z;tA_hRVQ`wFU)S<;4*t_3(NXt$EI0voSm7zRxzkpO7ZKBQr>u{q<N;BzZO^q?Up(g
zxj9+kTdqdc_qx+ZeWjMVH*Sf3l(ST{u|Zk!H0OW5u8SSMb51zy`zWNM8z#ak@PEO4
zE}nHamWOZGeXvj7n_W6tdhezAJJqK;NWYb>fBgCgS9*WS8(IGITt$!ez1X;H>7;p*
z(N*_V)>L1+Vp@?}P~`P_cB!kvjjS6<FMi$fI?(FCH_L%hpnIo+uj0b>wWXJAHzXLW
zmpT6QiRi{dHx_ZPQnh5`b~(I4v0Pn{+hCKW-L97Q`9WP8HR=7KOzUsn`}6pqSZLs`
zyhsr}9rFvfwlB<27I?L3)7Pg<EVqiK?0B_i(^m;5ugPbU;%0KX+ST4#I$P=k$DHy<
zt#{<I+*<OPI9#6Fa_fnAKDp?q$;Xs?c3-u7<*S=r>!t^P>QD=w!Zo9GamLcpqR^1#
z{30(ui2Lk%bKfV-j9qc^##~9+u4BLdO_)4;$8*)~%#~UaOnMVqTx}SO?XO9&&SP^m
z?zNjXRpS2A9RjQ4pJu<#RSsG6QhxuIxy<$6H(s7UoD*oXXT#C(0^Rvmt_eN=qxUz*
z@H`Rz`!1N3u_MAMZmHi{&Xf#R=^44<$G@<%O<1Vdz?(ji?d1%Y^S<AN@9B8dP3d^J
zR>4#E^2EUV3vc8r`YT>?c4*aRxS4ogN4%6nVmD_8|M%a|Zn`9R@>uM0w-FYv{a$yP
zr%<KD^}cOd!S*SfvMCb<cW?33RW{sjvi-^-=gk*88eZl1`YpQIwr>*CISo^R&5R!V
z&czyR;aI+O>I=;!`-7KV+duoSbMWhoK+zDkDD7XbAFkZ6P0sp1zwo{nu@?^ANQ^06
zkSEYJ*TkT};`HUWKPqfD6g_V<5U*QbQB<=b<h8P!;{s(xgKsUBz4cuO_6y`PGyQXV
z8?c=x%V=)S;+FK#PraJnDW@}5Jc_n)Z+7P2Z{@eh@<!2KX8*4|(<c3q6`H%#xa0ZM
z=h6z^j&6%R)*t<*W%rQ(=OW{{#Dt!1?$GJ)S8Q9;ka?n)`<GCcgGfV(cj4Dwm8EZ!
zx&KcMY@73<_`aHZ`dSB`+`^BW9(t~rJE45R%^eI3Ob3MQa?JgEJH%3rcq9+j$=|k#
zYw*)__?th~?{ih^yMhIr$7jz<^lCIN2~xNB_S-kFPQXZ|p0D}xmDJ^ntUaSw@aZXh
z(pVe2sUY~#oKhwGqb4j}_k8ZjuvdN%e)0Z8^SzsjR)zr&);e7ZG_0SZd!BvS1K%6T
zYO1R=<yTxd)xxK1?)W0mBf7#@^}+YZNpEs@HCjJymdSoxq|p;_Y?4TX-->LHEr*v)
z<e7avT{*}ux!L&!``5>2W!KIx<Z{ZqEPO(K-ere3Dsf>Ke9tGVz1zY4nO#Tq&cylq
zs<-Zy^^<-+;r`JNw`Z|Q+~kQ%zpbpZ*exhA=3Cy>zmp=<c<R>1MJXJN+4?c<px3{f
zCm0wQA~wfyirCeC;yMyjeT&mU#`6e^$ka@Z=RVge%Ga2?xBX~n*})JaFEs1mLc_`%
J&}9%13ILtgiZ}oO

literal 0
HcmV?d00001

-- 
GitLab