02-MNIST_testing_FC_architectures.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Testing Fully Connected Architectures\n",
    "\n",
    "This notebook tests several model architectures for the MNIST dataset. The main goal is both to explore the definition of the architecture, as well as analyzing its impact in the number of parameters.\n",
    "\n",
    "Although a performance test in terms of accuracy is also performed, note that to draw any meaningful conclusions, at the very least it is required to run each model multiple times and compare the averages. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torchvision\n",
    "import torchvision.transforms as transforms\n",
    "\n",
    "import torchinfo\n",
    "# importing a module with utilities for displaying stats and data\n",
    "import sys\n",
    "sys.path.insert(1, 'util')\n",
    "import vcpi_util\n",
    "\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import time"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Auxiliary functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def show_bad_preds(model, dataset, classes):\n",
    "    k = 0\n",
    "    iters = 0\n",
    "\n",
    "    preds = []\n",
    "    ground_truth = []\n",
    "    imgs = torch.Tensor(15, 1, 28,28)\n",
    "\n",
    "    iterator = iter(dataset)\n",
    "\n",
    "    max_iters = test_set.__len__() / batch_size\n",
    "    while k < 15 and iters < max_iters:\n",
    "\n",
    "        images, targets = next(iterator)\n",
    "        #print(images[0].shape)\n",
    "        logits = model(images.to(device))\n",
    "        \n",
    "        #print(predictions[0])\n",
    "        predictions = torch.nn.functional.softmax(logits, dim=1).cpu().detach().numpy()\n",
    "        for i in range(len(predictions)):\n",
    "\n",
    "            if np.argmax(predictions[i]) != targets[i] and k < 15:\n",
    "\n",
    "                preds.append(predictions[i])\n",
    "                ground_truth.append(targets[i])\n",
    "                imgs[k, :, :, :] = images[i]\n",
    "                k += 1\n",
    "\n",
    "        iters += 1\n",
    "\n",
    "    vcpi_util.plot_predictions(imgs, preds, ground_truth, classes, 5, 3)\n",
    "\n",
    "\n",
    "def evaluate(model, data_loader):\n",
    "\n",
    "    # sets the model in evaluation mode.\n",
    "    # although our model does not have layers which behave differently during training and evaluation\n",
    "    # this is a good practice as the models architecture may change in the future\n",
    "    model.eval()\n",
    "\n",
    "    correct = 0\n",
    "    \n",
    "    for i, (images, targets) in enumerate(data_loader):\n",
    "         \n",
    "        # forward pass, compute the output of the model for the current batch\n",
    "        outputs = model(images.to(device))\n",
    "\n",
    "        # \"max\" returns a namedtuple (values, indices) where values is the maximum \n",
    "        # value of each row of the input tensor in the given dimension dim; \n",
    "        # indices is the index location of each maximum value found (argmax).\n",
    "        # the argmax effectively provides the predicted class number        \n",
    "        _, preds = torch.max(outputs, dim=1)\n",
    "\n",
    "        correct += (preds.cpu() == targets).sum()\n",
    "\n",
    "    return (correct / len(data_loader.dataset)).item()\n",
    "\n",
    "\n",
    "def train(model, data_loader, epochs, loss_fn, optimizer):\n",
    "\n",
    "    # sets the model in training mode.\n",
    "    # although our model does not have layers which behave differently during training and evaluation\n",
    "    # this is a good practice as the models architecture may change in the future\n",
    "    model.train()\n",
    "\n",
    "    history = {}\n",
    "    history['accuracy'] = []\n",
    "    history['loss'] = []\n",
    "\n",
    "    for epoch in range(epochs):  # loop over the dataset multiple times\n",
    "\n",
    "        start_time = time.time() \n",
    "        correct = 0\n",
    "        running_loss = 0.0\n",
    "        for i, (inputs, targets) in enumerate(data_loader, 0):\n",
    "            \n",
    "            # send data to device, hopefully the GPU\n",
    "            inputs = inputs.to(device)\n",
    "            targets = targets.to(device)\n",
    "\n",
    "            # forward pass, compute the output of the model for the current batch\n",
    "            outputs = model(inputs)\n",
    "\n",
    "\n",
    "            _, predicted = torch.max(outputs, 1)\n",
    "\n",
    "            #compute the loss between the model's outputs and the labels, or targets.\n",
    "            loss = loss_fn(outputs, targets)\n",
    "            \n",
    "            # zero the parameter gradients\n",
    "            # this function must be called before calling \"backward\" on the losses, otherwise the gradients get accumulated with previous batches.\n",
    "            optimizer.zero_grad()\n",
    "\n",
    "            # perform backpropagation\n",
    "            loss.backward()\n",
    "            #update the model's parameters\n",
    "            optimizer.step()\n",
    "\n",
    "            running_loss += loss\n",
    "            correct += (predicted == targets).sum()\n",
    "\n",
    "        accuracy = 100 * correct / len(train_set)\n",
    "        stop_time = time.time()\n",
    "        print(f'Epoch: {epoch:03d}; Accuracy = {accuracy:0.4f}; Elapsed time: {(stop_time - start_time):0.4f}')\n",
    "        history['accuracy'].append(accuracy.cpu().numpy())\n",
    "        history['loss'].append(running_loss.cpu().detach().numpy())\n",
    "        \n",
    "    print('Finished Training')\n",
    "\n",
    "    return(history)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "def build_confusion_matrix(model, dataset):\n",
    "\n",
    "    preds = []\n",
    "    ground_truth = []\n",
    "\n",
    "    for images, targets in dataset:\n",
    "\n",
    "        predictions = model(images.to(device))\n",
    "        preds_sparse = [np.argmax(x) for x in predictions.cpu().detach().numpy()]\n",
    "        preds.extend(preds_sparse)\n",
    "        ground_truth.extend(targets.numpy())\n",
    "\n",
    "    vcpi_util.show_confusion_matrix(ground_truth, preds, len(dataset.dataset.classes))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cuda:0\n"
     ]
    }
   ],
   "source": [
    "device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')\n",
    "print(device)\n",
    "\n",
    "epochs = 50\n",
    "train_models = True\n",
    "\n",
    "eval_acc = []\n",
    "train_acc = []"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load MNIST dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "transform = transforms.Compose(\n",
    "    [transforms.ToTensor()]) \n",
    "\n",
    "batch_size = 128\n",
    "\n",
    "train_set = torchvision.datasets.MNIST(root='./data', train=True,\n",
    "                                        download=True, transform=transform)\n",
    "train_loader = torch.utils.data.DataLoader(train_set, batch_size=batch_size,\n",
    "                                          shuffle=True)\n",
    "\n",
    "test_set = torchvision.datasets.MNIST(root='./data', train=False,\n",
    "                                       download=True, transform=transform)\n",
    "test_loader = torch.utils.data.DataLoader(test_set, batch_size=batch_size,\n",
    "                                         shuffle=False)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Display a few images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x600 with 24 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "images, targets = next(iter(train_loader))\n",
    "\n",
    "# these are the transformed images\n",
    "vcpi_util.show_transformed_images(3,8, train_set, train_set.classes) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Create first model\n",
    "\n",
    "Single hidden layer with 64 units"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "class FullyConnected_I(torch.nn.Module):\n",
    "\n",
    "    def __init__(self, num_classes):\n",
    "        super().__init__()\n",
    "        self.fc1 = torch.nn.Linear(784, 64)\n",
    "        self.relu1 = torch.nn.ReLU()\n",
    "        self.fc2 = torch.nn.Linear(64,num_classes)\n",
    "        \n",
    "\n",
    "    def forward(self, x):    \n",
    "\n",
    "        x = torch.flatten(x,1)\n",
    "        x = self.fc1(x)\n",
    "        x = self.relu1(x)\n",
    "        x = self.fc2(x)\n",
    "        return(x)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Display model information"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "FullyConnected_I(\n",
      "  (fc1): Linear(in_features=784, out_features=64, bias=True)\n",
      "  (relu1): ReLU()\n",
      "  (fc2): Linear(in_features=64, out_features=10, bias=True)\n",
      ")\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "==========================================================================================\n",
       "Layer (type:depth-idx)                   Output Shape              Param #\n",
       "==========================================================================================\n",
       "FullyConnected_I                         [128, 10]                 --\n",
       "├─Linear: 1-1                            [128, 64]                 50,240\n",
       "├─ReLU: 1-2                              [128, 64]                 --\n",
       "├─Linear: 1-3                            [128, 10]                 650\n",
       "==========================================================================================\n",
       "Total params: 50,890\n",
       "Trainable params: 50,890\n",
       "Non-trainable params: 0\n",
       "Total mult-adds (Units.MEGABYTES): 6.51\n",
       "==========================================================================================\n",
       "Input size (MB): 0.40\n",
       "Forward/backward pass size (MB): 0.08\n",
       "Params size (MB): 0.20\n",
       "Estimated Total Size (MB): 0.68\n",
       "=========================================================================================="
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = FullyConnected_I(len(train_set.classes))\n",
    "model.to(device)\n",
    "print(model)\n",
    "\n",
    "torchinfo.summary(model, input_size=(batch_size, 1, 28, 28))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Specify the optimizer, loss and metrics\n",
    "\n",
    "optimizer: gradient descent optimization algorithm\n",
    "\n",
    "loss: the loss function upon which derivatives are computed\n",
    "\n",
    "metrics: metrics to be evaluated during tranining and evaluation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train and evaluate the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "optimizer = torch.optim.Adam(model.parameters())\n",
    "loss_fn = torch.nn.CrossEntropyLoss()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 000; Accuracy = 87.5917; Elapsed time: 6.3734\n",
      "Epoch: 001; Accuracy = 93.2717; Elapsed time: 6.2455\n",
      "Epoch: 002; Accuracy = 94.7267; Elapsed time: 6.2817\n",
      "Epoch: 003; Accuracy = 95.6367; Elapsed time: 6.3693\n",
      "Epoch: 004; Accuracy = 96.2583; Elapsed time: 6.3443\n",
      "Epoch: 005; Accuracy = 96.7517; Elapsed time: 6.2058\n",
      "Epoch: 006; Accuracy = 97.1033; Elapsed time: 6.2478\n",
      "Epoch: 007; Accuracy = 97.4183; Elapsed time: 6.1984\n",
      "Epoch: 008; Accuracy = 97.6850; Elapsed time: 6.1869\n",
      "Epoch: 009; Accuracy = 97.9100; Elapsed time: 6.1817\n",
      "Epoch: 010; Accuracy = 98.1483; Elapsed time: 6.3081\n",
      "Epoch: 011; Accuracy = 98.3233; Elapsed time: 6.3829\n",
      "Epoch: 012; Accuracy = 98.4300; Elapsed time: 6.3949\n",
      "Epoch: 013; Accuracy = 98.5817; Elapsed time: 6.3412\n",
      "Epoch: 014; Accuracy = 98.7283; Elapsed time: 6.3847\n",
      "Epoch: 015; Accuracy = 98.8383; Elapsed time: 6.4070\n",
      "Epoch: 016; Accuracy = 98.9617; Elapsed time: 6.4729\n",
      "Epoch: 017; Accuracy = 99.0250; Elapsed time: 6.4096\n",
      "Epoch: 018; Accuracy = 99.1717; Elapsed time: 6.4044\n",
      "Epoch: 019; Accuracy = 99.2367; Elapsed time: 6.3320\n",
      "Epoch: 020; Accuracy = 99.3017; Elapsed time: 6.2774\n",
      "Epoch: 021; Accuracy = 99.3650; Elapsed time: 6.3888\n",
      "Epoch: 022; Accuracy = 99.4250; Elapsed time: 6.4252\n",
      "Epoch: 023; Accuracy = 99.5083; Elapsed time: 6.3813\n",
      "Epoch: 024; Accuracy = 99.5567; Elapsed time: 6.4681\n",
      "Epoch: 025; Accuracy = 99.6033; Elapsed time: 6.3229\n",
      "Epoch: 026; Accuracy = 99.6450; Elapsed time: 6.2698\n",
      "Epoch: 027; Accuracy = 99.7200; Elapsed time: 6.2866\n",
      "Epoch: 028; Accuracy = 99.7283; Elapsed time: 6.3916\n",
      "Epoch: 029; Accuracy = 99.7933; Elapsed time: 6.4451\n",
      "Epoch: 030; Accuracy = 99.7617; Elapsed time: 6.3921\n",
      "Epoch: 031; Accuracy = 99.8317; Elapsed time: 6.3275\n",
      "Epoch: 032; Accuracy = 99.8700; Elapsed time: 6.3436\n",
      "Epoch: 033; Accuracy = 99.8483; Elapsed time: 6.3574\n",
      "Epoch: 034; Accuracy = 99.9000; Elapsed time: 6.3370\n",
      "Epoch: 035; Accuracy = 99.9050; Elapsed time: 6.3160\n",
      "Epoch: 036; Accuracy = 99.8900; Elapsed time: 6.3587\n",
      "Epoch: 037; Accuracy = 99.8900; Elapsed time: 6.2652\n",
      "Epoch: 038; Accuracy = 99.9000; Elapsed time: 6.2013\n",
      "Epoch: 039; Accuracy = 99.9683; Elapsed time: 6.1751\n",
      "Epoch: 040; Accuracy = 99.9850; Elapsed time: 6.1949\n",
      "Epoch: 041; Accuracy = 99.9717; Elapsed time: 6.2143\n",
      "Epoch: 042; Accuracy = 99.9350; Elapsed time: 6.1976\n",
      "Epoch: 043; Accuracy = 99.8717; Elapsed time: 6.3252\n",
      "Epoch: 044; Accuracy = 99.9600; Elapsed time: 6.2735\n",
      "Epoch: 045; Accuracy = 99.9883; Elapsed time: 6.4206\n",
      "Epoch: 046; Accuracy = 99.9833; Elapsed time: 6.2596\n",
      "Epoch: 047; Accuracy = 99.9850; Elapsed time: 6.2117\n",
      "Epoch: 048; Accuracy = 99.9650; Elapsed time: 6.2106\n",
      "Epoch: 049; Accuracy = 99.9683; Elapsed time: 6.1829\n",
      "Finished Training\n"
     ]
    }
   ],
   "source": [
    "if train_models:\n",
    "    history = train(model, train_loader, epochs, loss_fn, optimizer)\n",
    "\n",
    "    ###### Saving ######\n",
    "    torch.save({\n",
    "        'epoch': epochs,\n",
    "        'model':model.state_dict(),\n",
    "        'optimizer': optimizer.state_dict(),\n",
    "        'history' : history\n",
    "        },\n",
    "        f'model_{epochs}.pt')    \n",
    "else:\n",
    "    retrieved = torch.load(f'model_{epochs}.pt')\n",
    "    \n",
    "    model.load_state_dict(retrieved['model'])\n",
    "    model.to(device)\n",
    "\n",
    "    optimizer.load_state_dict(retrieved['optimizer'])\n",
    "    history = retrieved['history']\n",
    "\n",
    "train_acc.append(history['accuracy'][-1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## History graphs for accuracy and loss\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "vcpi_util.show_history(history)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluate the model on the test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "97.46000170707703\n"
     ]
    }
   ],
   "source": [
    "eval_acc.append(evaluate(model, test_loader)*100)\n",
    "\n",
    "print(eval_acc[-1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model 2\n",
    "\n",
    "wider model, doubling the number of units in the hidden layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "class FullyConnected_II(torch.nn.Module):\n",
    "\n",
    "    def __init__(self, num_classes):\n",
    "        super().__init__()\n",
    "        self.fc1 = torch.nn.Linear(784, 128)\n",
    "        self.relu1 = torch.nn.ReLU()\n",
    "        self.fc2 = torch.nn.Linear(128,num_classes)\n",
    "        \n",
    "\n",
    "    def forward(self, x):    \n",
    "\n",
    "        x = torch.flatten(x,1)\n",
    "        x = self.fc1(x)\n",
    "        x = self.relu1(x)\n",
    "        x = self.fc2(x)\n",
    "        return(x)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Display model information"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "FullyConnected_II(\n",
      "  (fc1): Linear(in_features=784, out_features=128, bias=True)\n",
      "  (relu1): ReLU()\n",
      "  (fc2): Linear(in_features=128, out_features=10, bias=True)\n",
      ")\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "==========================================================================================\n",
       "Layer (type:depth-idx)                   Output Shape              Param #\n",
       "==========================================================================================\n",
       "FullyConnected_II                        [128, 10]                 --\n",
       "├─Linear: 1-1                            [128, 128]                100,480\n",
       "├─ReLU: 1-2                              [128, 128]                --\n",
       "├─Linear: 1-3                            [128, 10]                 1,290\n",
       "==========================================================================================\n",
       "Total params: 101,770\n",
       "Trainable params: 101,770\n",
       "Non-trainable params: 0\n",
       "Total mult-adds (Units.MEGABYTES): 13.03\n",
       "==========================================================================================\n",
       "Input size (MB): 0.40\n",
       "Forward/backward pass size (MB): 0.14\n",
       "Params size (MB): 0.41\n",
       "Estimated Total Size (MB): 0.95\n",
       "=========================================================================================="
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_II = FullyConnected_II(len(train_set.classes))\n",
    "model_II.to(device)\n",
    "print(model_II)\n",
    "\n",
    "torchinfo.summary(model_II, input_size=(batch_size, 1, 28, 28))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train the model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 000; Accuracy = 99.9167; Elapsed time: 6.3570\n",
      "Epoch: 001; Accuracy = 99.9683; Elapsed time: 6.4604\n",
      "Epoch: 002; Accuracy = 99.9567; Elapsed time: 6.2764\n",
      "Epoch: 003; Accuracy = 99.9750; Elapsed time: 6.3001\n",
      "Epoch: 004; Accuracy = 100.0000; Elapsed time: 6.4035\n",
      "Epoch: 005; Accuracy = 100.0000; Elapsed time: 6.3833\n",
      "Epoch: 006; Accuracy = 100.0000; Elapsed time: 6.3231\n",
      "Epoch: 007; Accuracy = 100.0000; Elapsed time: 6.2234\n",
      "Epoch: 008; Accuracy = 100.0000; Elapsed time: 6.2063\n",
      "Epoch: 009; Accuracy = 100.0000; Elapsed time: 6.2803\n",
      "Epoch: 010; Accuracy = 100.0000; Elapsed time: 6.1810\n",
      "Epoch: 011; Accuracy = 100.0000; Elapsed time: 6.2046\n",
      "Epoch: 012; Accuracy = 100.0000; Elapsed time: 6.2395\n",
      "Epoch: 013; Accuracy = 100.0000; Elapsed time: 6.2006\n",
      "Epoch: 014; Accuracy = 100.0000; Elapsed time: 6.2418\n",
      "Epoch: 015; Accuracy = 99.6317; Elapsed time: 6.1500\n",
      "Epoch: 016; Accuracy = 99.9500; Elapsed time: 6.2121\n",
      "Epoch: 017; Accuracy = 99.9867; Elapsed time: 6.3085\n",
      "Epoch: 018; Accuracy = 100.0000; Elapsed time: 6.3037\n",
      "Epoch: 019; Accuracy = 100.0000; Elapsed time: 6.2087\n",
      "Epoch: 020; Accuracy = 100.0000; Elapsed time: 6.2019\n",
      "Epoch: 021; Accuracy = 100.0000; Elapsed time: 6.3891\n",
      "Epoch: 022; Accuracy = 100.0000; Elapsed time: 6.3075\n",
      "Epoch: 023; Accuracy = 100.0000; Elapsed time: 6.1925\n",
      "Epoch: 024; Accuracy = 100.0000; Elapsed time: 6.1733\n",
      "Epoch: 025; Accuracy = 100.0000; Elapsed time: 6.1523\n",
      "Epoch: 026; Accuracy = 100.0000; Elapsed time: 6.2278\n",
      "Epoch: 027; Accuracy = 100.0000; Elapsed time: 6.2846\n",
      "Epoch: 028; Accuracy = 100.0000; Elapsed time: 6.2652\n",
      "Epoch: 029; Accuracy = 100.0000; Elapsed time: 6.2763\n",
      "Epoch: 030; Accuracy = 100.0000; Elapsed time: 6.2973\n",
      "Epoch: 031; Accuracy = 100.0000; Elapsed time: 6.3053\n",
      "Epoch: 032; Accuracy = 100.0000; Elapsed time: 6.2842\n",
      "Epoch: 033; Accuracy = 99.9267; Elapsed time: 6.1766\n",
      "Epoch: 034; Accuracy = 99.7450; Elapsed time: 6.2618\n",
      "Epoch: 035; Accuracy = 99.9517; Elapsed time: 6.3159\n",
      "Epoch: 036; Accuracy = 99.9850; Elapsed time: 6.2866\n",
      "Epoch: 037; Accuracy = 100.0000; Elapsed time: 6.2945\n",
      "Epoch: 038; Accuracy = 100.0000; Elapsed time: 6.2965\n",
      "Epoch: 039; Accuracy = 100.0000; Elapsed time: 6.3119\n",
      "Epoch: 040; Accuracy = 100.0000; Elapsed time: 6.2711\n",
      "Epoch: 041; Accuracy = 100.0000; Elapsed time: 6.3240\n",
      "Epoch: 042; Accuracy = 100.0000; Elapsed time: 6.2666\n",
      "Epoch: 043; Accuracy = 100.0000; Elapsed time: 6.1762\n",
      "Epoch: 044; Accuracy = 100.0000; Elapsed time: 6.2201\n",
      "Epoch: 045; Accuracy = 100.0000; Elapsed time: 6.1913\n",
      "Epoch: 046; Accuracy = 100.0000; Elapsed time: 6.1748\n",
      "Epoch: 047; Accuracy = 100.0000; Elapsed time: 6.1869\n",
      "Epoch: 048; Accuracy = 100.0000; Elapsed time: 6.1616\n",
      "Epoch: 049; Accuracy = 100.0000; Elapsed time: 6.2836\n",
      "Finished Training\n"
     ]
    }
   ],
   "source": [
    "optimizer_II = torch.optim.Adam(model_II.parameters())\n",
    "\n",
    "if train_models:\n",
    "    history_II = train(model_II, train_loader, epochs, loss_fn, optimizer_II)\n",
    "\n",
    "    ###### Saving ######\n",
    "    torch.save({\n",
    "        'epoch': epochs,\n",
    "        'model':model_II.state_dict(),\n",
    "        'optimizer': optimizer_II.state_dict(),\n",
    "        'history' : history_II\n",
    "        },\n",
    "        f'model_II_{epochs}.pt')    \n",
    "else:\n",
    "    retrieved = torch.load(f'model_II_{epochs}.pt')\n",
    "    \n",
    "    model_II.load_state_dict(retrieved['model'])\n",
    "    model_II.to(device)\n",
    "\n",
    "    optimizer_II.load_state_dict(retrieved['optimizer'])\n",
    "    history_II = retrieved['history']\n",
    "\n",
    "\n",
    "\n",
    "train_acc.append(history_II['accuracy'][-1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## History graphs for accuracy and loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "vcpi_util.show_history(history_II)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Evaluate the model on the test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "97.97000288963318\n"
     ]
    }
   ],
   "source": [
    "eval_acc.append(evaluate(model_II, test_loader)*100)\n",
    "print(eval_acc[-1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model 3\n",
    "\n",
    "Adding a second hidden layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "class FullyConnected_III(torch.nn.Module):\n",
    "\n",
    "    def __init__(self, num_classes):\n",
    "        super().__init__()\n",
    "        self.fc1 = torch.nn.Linear(784, 128)\n",
    "        self.relu1 = torch.nn.ReLU()\n",
    "        self.fc2 = torch.nn.Linear(128,128)\n",
    "        self.relu2 = torch.nn.ReLU()\n",
    "        self.fc3 = torch.nn.Linear(128,num_classes)\n",
    "        \n",
    "\n",
    "    def forward(self, x):    \n",
    "\n",
    "        x = torch.flatten(x,1)\n",
    "        x = self.fc1(x)\n",
    "        x = self.relu1(x)\n",
    "        x = self.fc2(x)\n",
    "        x = self.relu2(x)\n",
    "        x = self.fc3(x)\n",
    "\n",
    "        return(x)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "FullyConnected_III(\n",
      "  (fc1): Linear(in_features=784, out_features=128, bias=True)\n",
      "  (relu1): ReLU()\n",
      "  (fc2): Linear(in_features=128, out_features=128, bias=True)\n",
      "  (relu2): ReLU()\n",
      "  (fc3): Linear(in_features=128, out_features=10, bias=True)\n",
      ")\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "==========================================================================================\n",
       "Layer (type:depth-idx)                   Output Shape              Param #\n",
       "==========================================================================================\n",
       "FullyConnected_III                       [128, 10]                 --\n",
       "├─Linear: 1-1                            [128, 128]                100,480\n",
       "├─ReLU: 1-2                              [128, 128]                --\n",
       "├─Linear: 1-3                            [128, 128]                16,512\n",
       "├─ReLU: 1-4                              [128, 128]                --\n",
       "├─Linear: 1-5                            [128, 10]                 1,290\n",
       "==========================================================================================\n",
       "Total params: 118,282\n",
       "Trainable params: 118,282\n",
       "Non-trainable params: 0\n",
       "Total mult-adds (Units.MEGABYTES): 15.14\n",
       "==========================================================================================\n",
       "Input size (MB): 0.40\n",
       "Forward/backward pass size (MB): 0.27\n",
       "Params size (MB): 0.47\n",
       "Estimated Total Size (MB): 1.15\n",
       "=========================================================================================="
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_III = FullyConnected_III(len(train_set.classes))\n",
    "model_III.to(device)\n",
    "print(model_III)\n",
    "\n",
    "torchinfo.summary(model_III, input_size=(batch_size, 1, 28, 28))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 000; Accuracy = 89.8233; Elapsed time: 6.4406\n",
      "Epoch: 001; Accuracy = 95.4767; Elapsed time: 6.4657\n",
      "Epoch: 002; Accuracy = 96.8417; Elapsed time: 6.4415\n",
      "Epoch: 003; Accuracy = 97.4867; Elapsed time: 6.5456\n",
      "Epoch: 004; Accuracy = 98.0033; Elapsed time: 6.4648\n",
      "Epoch: 005; Accuracy = 98.3783; Elapsed time: 6.6152\n",
      "Epoch: 006; Accuracy = 98.6217; Elapsed time: 6.5059\n",
      "Epoch: 007; Accuracy = 98.8933; Elapsed time: 6.4300\n",
      "Epoch: 008; Accuracy = 99.1383; Elapsed time: 6.5838\n",
      "Epoch: 009; Accuracy = 99.2633; Elapsed time: 6.6567\n",
      "Epoch: 010; Accuracy = 99.3383; Elapsed time: 6.5029\n",
      "Epoch: 011; Accuracy = 99.4950; Elapsed time: 6.4415\n",
      "Epoch: 012; Accuracy = 99.5783; Elapsed time: 6.3972\n",
      "Epoch: 013; Accuracy = 99.6567; Elapsed time: 6.5366\n",
      "Epoch: 014; Accuracy = 99.4883; Elapsed time: 6.5188\n",
      "Epoch: 015; Accuracy = 99.6233; Elapsed time: 6.5263\n",
      "Epoch: 016; Accuracy = 99.5983; Elapsed time: 6.4057\n",
      "Epoch: 017; Accuracy = 99.6933; Elapsed time: 6.4863\n",
      "Epoch: 018; Accuracy = 99.7983; Elapsed time: 6.4381\n",
      "Epoch: 019; Accuracy = 99.7133; Elapsed time: 6.3981\n",
      "Epoch: 020; Accuracy = 99.7917; Elapsed time: 6.3942\n",
      "Epoch: 021; Accuracy = 99.6133; Elapsed time: 6.3865\n",
      "Epoch: 022; Accuracy = 99.8767; Elapsed time: 6.3875\n",
      "Epoch: 023; Accuracy = 99.8433; Elapsed time: 6.7198\n",
      "Epoch: 024; Accuracy = 99.7067; Elapsed time: 6.4160\n",
      "Epoch: 025; Accuracy = 99.7267; Elapsed time: 6.3463\n",
      "Epoch: 026; Accuracy = 99.8300; Elapsed time: 6.3439\n",
      "Epoch: 027; Accuracy = 99.7767; Elapsed time: 6.3616\n",
      "Epoch: 028; Accuracy = 99.7067; Elapsed time: 6.2968\n",
      "Epoch: 029; Accuracy = 99.9433; Elapsed time: 6.4252\n",
      "Epoch: 030; Accuracy = 99.9733; Elapsed time: 6.5716\n",
      "Epoch: 031; Accuracy = 100.0000; Elapsed time: 6.6412\n",
      "Epoch: 032; Accuracy = 100.0000; Elapsed time: 6.4229\n",
      "Epoch: 033; Accuracy = 100.0000; Elapsed time: 6.3763\n",
      "Epoch: 034; Accuracy = 100.0000; Elapsed time: 6.3393\n",
      "Epoch: 035; Accuracy = 100.0000; Elapsed time: 6.4110\n",
      "Epoch: 036; Accuracy = 100.0000; Elapsed time: 6.4183\n",
      "Epoch: 037; Accuracy = 100.0000; Elapsed time: 6.6023\n",
      "Epoch: 038; Accuracy = 99.0217; Elapsed time: 6.3094\n",
      "Epoch: 039; Accuracy = 99.7667; Elapsed time: 6.3751\n",
      "Epoch: 040; Accuracy = 99.9233; Elapsed time: 6.5032\n",
      "Epoch: 041; Accuracy = 99.9300; Elapsed time: 6.5118\n",
      "Epoch: 042; Accuracy = 99.7850; Elapsed time: 6.5053\n",
      "Epoch: 043; Accuracy = 99.7317; Elapsed time: 6.4555\n",
      "Epoch: 044; Accuracy = 99.9250; Elapsed time: 6.4185\n",
      "Epoch: 045; Accuracy = 99.9217; Elapsed time: 6.5362\n",
      "Epoch: 046; Accuracy = 99.8217; Elapsed time: 6.4918\n",
      "Epoch: 047; Accuracy = 99.8633; Elapsed time: 6.3310\n",
      "Epoch: 048; Accuracy = 99.8367; Elapsed time: 6.4548\n",
      "Epoch: 049; Accuracy = 99.8367; Elapsed time: 6.5026\n",
      "Finished Training\n"
     ]
    }
   ],
   "source": [
    "optimizer_III = torch.optim.Adam(model_III.parameters())\n",
    "\n",
    "if train_models:\n",
    "    history_III = train(model_III, train_loader, epochs, loss_fn, optimizer_III)\n",
    "\n",
    "    ###### Saving ######\n",
    "    torch.save({\n",
    "        'epoch': epochs,\n",
    "        'model':model_III.state_dict(),\n",
    "        'optimizer': optimizer_III.state_dict(),\n",
    "        'history' : history_III\n",
    "        },\n",
    "        f'model_II_{epochs}.pt')    \n",
    "else:\n",
    "    retrieved = torch.load(f'model_III_{epochs}.pt')\n",
    "    \n",
    "    model_III.load_state_dict(retrieved['model'])\n",
    "    model_III.to(device)\n",
    "\n",
    "    optimizer_III.load_state_dict(retrieved['optimizer'])\n",
    "    history_III = retrieved['history']\n",
    "\n",
    "\n",
    "train_acc.append(history_III['accuracy'][-1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "###### Saving ######\n",
    "torch.save({\n",
    "    'epoch': epochs,\n",
    "    'model':model_III.state_dict(),\n",
    "    'optimizer': optimizer_III.state_dict(),\n",
    "    'history' : history_III\n",
    "    },\n",
    "    f'model_III_{epochs}.pt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "vcpi_util.show_history(history_III)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "97.83999919891357\n"
     ]
    }
   ],
   "source": [
    "eval_acc.append(evaluate(model_III, test_loader)*100)\n",
    "print(eval_acc[-1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model 4\n",
    "\n",
    "Adding a third hidden layer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "class FullyConnected_IV(torch.nn.Module):\n",
    "\n",
    "    def __init__(self, num_classes):\n",
    "        super().__init__()\n",
    "        self.fc1 = torch.nn.Linear(784, 256)\n",
    "        self.relu1 = torch.nn.ReLU()\n",
    "        self.fc2 = torch.nn.Linear(256,128)\n",
    "        self.relu2 = torch.nn.ReLU()\n",
    "        self.fc3 = torch.nn.Linear(128,64)\n",
    "        self.relu3 = torch.nn.ReLU()\n",
    "        self.fc4 = torch.nn.Linear(64,num_classes)\n",
    "        \n",
    "\n",
    "    def forward(self, x):    \n",
    "\n",
    "        x = torch.flatten(x,1)\n",
    "        x = self.fc1(x)\n",
    "        x = self.relu1(x)\n",
    "        x = self.fc2(x)\n",
    "        x = self.relu2(x)\n",
    "        x = self.fc3(x)\n",
    "        x = self.relu3(x)\n",
    "        x = self.fc4(x)\n",
    "\n",
    "        return(x)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "FullyConnected_IV(\n",
      "  (fc1): Linear(in_features=784, out_features=256, bias=True)\n",
      "  (relu1): ReLU()\n",
      "  (fc2): Linear(in_features=256, out_features=128, bias=True)\n",
      "  (relu2): ReLU()\n",
      "  (fc3): Linear(in_features=128, out_features=64, bias=True)\n",
      "  (relu3): ReLU()\n",
      "  (fc4): Linear(in_features=64, out_features=10, bias=True)\n",
      ")\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "==========================================================================================\n",
       "Layer (type:depth-idx)                   Output Shape              Param #\n",
       "==========================================================================================\n",
       "FullyConnected_IV                        [128, 10]                 --\n",
       "├─Linear: 1-1                            [128, 256]                200,960\n",
       "├─ReLU: 1-2                              [128, 256]                --\n",
       "├─Linear: 1-3                            [128, 128]                32,896\n",
       "├─ReLU: 1-4                              [128, 128]                --\n",
       "├─Linear: 1-5                            [128, 64]                 8,256\n",
       "├─ReLU: 1-6                              [128, 64]                 --\n",
       "├─Linear: 1-7                            [128, 10]                 650\n",
       "==========================================================================================\n",
       "Total params: 242,762\n",
       "Trainable params: 242,762\n",
       "Non-trainable params: 0\n",
       "Total mult-adds (Units.MEGABYTES): 31.07\n",
       "==========================================================================================\n",
       "Input size (MB): 0.40\n",
       "Forward/backward pass size (MB): 0.47\n",
       "Params size (MB): 0.97\n",
       "Estimated Total Size (MB): 1.84\n",
       "=========================================================================================="
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_IV = FullyConnected_IV(len(train_set.classes))\n",
    "model_IV.to(device)\n",
    "print(model_IV)\n",
    "\n",
    "torchinfo.summary(model_IV, input_size=(batch_size, 1, 28, 28))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "optimizer_IV = torch.optim.Adam(model_IV.parameters())\n",
    "\n",
    "if train_models:\n",
    "    history_IV = train(model_IV, train_loader, epochs, loss_fn, optimizer_IV)\n",
    "\n",
    "    ###### Saving ######\n",
    "    torch.save({\n",
    "        'epoch': epochs,\n",
    "        'model':model_IV.state_dict(),\n",
    "        'optimizer': optimizer_IV.state_dict(),\n",
    "        'history' : history_IV\n",
    "        },\n",
    "        f'model_IV_{epochs}.pt')    \n",
    "else:\n",
    "    retrieved = torch.load(f'model_IV_{epochs}.pt')\n",
    "    \n",
    "    model_IV.load_state_dict(retrieved['model'])\n",
    "    model_IV.to(device)\n",
    "\n",
    "    optimizer_IV.load_state_dict(retrieved['optimizer'])\n",
    "    history_IV = retrieved['history']\n",
    "\n",
    "train_acc.append(history_IV['accuracy'][-1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjsAAAHHCAYAAABZbpmkAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAABHQklEQVR4nO3deXhU5d3/8c9km+wJSSAkEEICkggCyiIGEbcoKioqCvizglhrFVBARUVFbdGCVH0qakWfqghYNyiujyLIVjTs+75DIIEA2ffJzPn9ETOaAhbCzJzJ5P26rrlIzpw5850DdT69v/e5j8UwDEMAAAA+ys/sAgAAANyJsAMAAHwaYQcAAPg0wg4AAPBphB0AAODTCDsAAMCnEXYAAIBPI+wAAACfRtgBAAA+jbADAAB8GmEHAAD4NMIOAADwaYQdAE1aWVmZ2SUAcDPCDgCXOnDggEaMGKG0tDSFhIQoNjZWd9xxh/bv33/SvoWFhRo7dqzatm0rq9Wq1q1ba+jQoTp+/Lhzn8rKSj3//PPq0KGDgoODlZCQoNtuu0179uyRJC1evFgWi0WLFy+ud+z9+/fLYrFo+vTpzm333HOPwsPDtWfPHt1www2KiIjQXXfdJUn697//rTvuuENt2rSR1WpVUlKSxo4dq4qKipPq3r59uwYNGqTmzZsrJCREaWlpevrppyVJixYtksVi0dy5c0963T//+U9ZLBZlZWWd7WkFcA4CzC4AgG9ZtWqVfvrpJw0ZMkStW7fW/v379dZbb+mKK67Q1q1bFRoaKkkqLS3VZZddpm3btunee+9Vt27ddPz4cX355Zc6dOiQ4uLiZLfbdeONN+qHH37QkCFDNHr0aJWUlGj+/PnavHmz2rVrd9b11dTUqF+/furTp49efvllZz2fffaZysvL9eCDDyo2NlYrV67U66+/rkOHDumzzz5zvn7jxo267LLLFBgYqPvvv19t27bVnj179NVXX+nFF1/UFVdcoaSkJH344Ye69dZb6733hx9+qHbt2ikjI+MczjCAs2YAgAuVl5eftC0rK8uQZMyYMcO57dlnnzUkGf/6179O2t/hcBiGYRjvvfeeIcl49dVXT7vPokWLDEnGokWL6j2/b98+Q5Lx/vvvO7cNGzbMkGQ8+eSTZ1T3pEmTDIvFYhw4cMC5rW/fvkZERES9bb+uxzAMY/z48YbVajUKCwud2/Ly8oyAgADjueeeO+l9ALgXbSwALhUSEuL82Waz6cSJE2rfvr2io6O1du1a53Nz5sxR165dTxr9kCSLxeLcJy4uTg899NBp92mIBx988DfrLisr0/Hjx9W7d28ZhqF169ZJko4dO6alS5fq3nvvVZs2bU5bz9ChQ1VVVaXZs2c7t33yySeqqanR7373uwbXDaBhCDsAXKqiokLPPvuskpKSZLVaFRcXp+bNm6uwsFBFRUXO/fbs2aMLLrjgN4+1Z88epaWlKSDAdR33gIAAtW7d+qTtBw8e1D333KOYmBiFh4erefPmuvzyyyXJWffevXsl6b/WnZ6erp49e+rDDz90bvvwww91ySWXqH379q76KADOEHN2ALjUQw89pPfff19jxoxRRkaGoqKiZLFYNGTIEDkcDpe/3+lGeOx2+ym3W61W+fn5nbTvNddco/z8fD3xxBNKT09XWFiYDh8+rHvuuadBdQ8dOlSjR4/WoUOHVFVVpeXLl+uNN9446+MAOHeEHQAuNXv2bA0bNkyvvPKKc1tlZaUKCwvr7deuXTtt3rz5N4/Vrl07rVixQjabTYGBgafcp1mzZpJ00vEPHDhwxjVv2rRJO3fu1AcffKChQ4c6t8+fP7/efqmpqZL0X+uWpCFDhuiRRx7RRx99pIqKCgUGBmrw4MFnXBMA16GNBcCl/P39ZRhGvW2vv/76SSMtAwcO1IYNG055iXbd6wcOHKjjx4+fckSkbp/k5GT5+/tr6dKl9Z7/+9//flY1//qYdT+/9tpr9fZr3ry5+vbtq/fee08HDx48ZT114uLidP3112vWrFn68MMPdd111ykuLu6MawLgOozsAHCpG2+8UTNnzlRUVJQ6duyorKwsLViwQLGxsfX2GzdunGbPnq077rhD9957r7p37678/Hx9+eWXmjZtmrp27aqhQ4dqxowZeuSRR7Ry5UpddtllKisr04IFCzRixAgNGDBAUVFRuuOOO/T666/LYrGoXbt2+vrrr5WXl3fGNaenp6tdu3Z67LHHdPjwYUVGRmrOnDkqKCg4ad+pU6eqT58+6tatm+6//36lpKRo//79+uabb7R+/fp6+w4dOlS33367JGnixIlnfzIBuIaJV4IB8EEFBQXG8OHDjbi4OCM8PNzo16+fsX37diM5OdkYNmxYvX1PnDhhjBo1ymjVqpURFBRktG7d2hg2bJhx/Phx5z7l5eXG008/baSkpBiBgYFGy5Ytjdtvv93Ys2ePc59jx44ZAwcONEJDQ41mzZoZf/zjH43Nmzef8tLzsLCwU9a9detWIzMz0wgPDzfi4uKMP/zhD8aGDRtOOoZhGMbmzZuNW2+91YiOjjaCg4ONtLQ0Y8KECScds6qqymjWrJkRFRVlVFRUnP3JBOASFsP4j7FXAIBL1NTUKDExUTfddJPeffdds8sBmizm7ACAm3z++ec6duxYvUnPADyPkR0AcLEVK1Zo48aNmjhxouLi4uotpgjA8xjZAQAXe+utt/Tggw+qRYsWmjFjhtnlAE0eIzsAAMCnMbIDAAB8GmEHAAD4NBYVlORwOJSTk6OIiIhzupMyAADwHMMwVFJSosTExJPuefdrhB1JOTk5SkpKMrsMAADQANnZ2WrduvVpnyfsSIqIiJBUe7IiIyNNrgYAAJyJ4uJiJSUlOb/HT4ewIzlbV5GRkYQdAAAamf82BYUJygAAwKcRdgAAgE8j7AAAAJ9G2AEAAD6NsAMAAHwaYQcAAPg0wg4AAPBphB0AAODTCDsAAMCnEXYAAIBPMzXsLF26VDfddJMSExNlsVj0+eef13veMAw9++yzSkhIUEhIiDIzM7Vr1656++Tn5+uuu+5SZGSkoqOj9fvf/16lpaUe/BQAAMCbmRp2ysrK1LVrV7355punfH7KlCmaOnWqpk2bphUrVigsLEz9+vVTZWWlc5+77rpLW7Zs0fz58/X1119r6dKluv/++z31EQAAgJezGIZhmF2EVHsTr7lz5+qWW26RVDuqk5iYqEcffVSPPfaYJKmoqEjx8fGaPn26hgwZom3btqljx45atWqVevToIUn67rvvdMMNN+jQoUNKTEw8o/cuLi5WVFSUioqKuBEoAJjIMAzVOAzZHYZsdofsjtrfa+yGahwOs8vDOYiPDFagv2vHWM70+9tr73q+b98+HTlyRJmZmc5tUVFR6tWrl7KysjRkyBBlZWUpOjraGXQkKTMzU35+flqxYoVuvfXWUx67qqpKVVVVzt+Li4vd90EAAKdUXGnTgq1H9X+bcpW154Qqa2rDDXzTwkcvV2rzcFPe22vDzpEjRyRJ8fHx9bbHx8c7nzty5IhatGhR7/mAgADFxMQ49zmVSZMm6U9/+pOLKwYA/De/DjhLdx5Xtf3MRmv8/Szy97PI4ub64D4Wi3l/e14bdtxp/PjxeuSRR5y/FxcXKykpycSKAMA8DoehwgqbjpdW6XhJlY6VVim/rFo928boglZR53z83wo47VuE64bOCbq2Y7yaR1jl72dRoJ+f/P0tCvCrffj7WUz9okTj57Vhp2XLlpKko0ePKiEhwbn96NGjuvDCC5375OXl1XtdTU2N8vPzna8/FavVKqvV6vqiATRqNrtD2fnl2n+iTAVlNl2V3kLNwoLc/r5FFTb5WaSI4EC3v1d2frneWbpXB/PLdaykSsd/DjY1p2gf+ftZ9JdbL9Dgnm0a9F41doemzNuh6T/uP2XAubFLgjrERzT4swBnymvDTkpKilq2bKkffvjBGW6Ki4u1YsUKPfjgg5KkjIwMFRYWas2aNerevbskaeHChXI4HOrVq5dZpQPwYoZhKKeoUvuOlWnf8VLtO17+859lyi6oqDdnJCzIX/f2SdF9fVIVFeqaIGIYhg7ml2v1/gKtPpCv1fsLtCuvVIH+Ft3UNVG/75OiTonnPprynxwOQzOy9mvKvB0qr7afcp+okEDFhQcpLtyqGoehNQcK9MScTTpcUKGx13Q4q9GVgrJqjfznWv2054QkqV3zMPXvkqj+nRPUIT6ckRp4lKlXY5WWlmr37t2SpIsuukivvvqqrrzySsXExKhNmzZ66aWXNHnyZH3wwQdKSUnRhAkTtHHjRm3dulXBwcGSpOuvv15Hjx7VtGnTZLPZNHz4cPXo0UP//Oc/z7gOrsYCmoZ1Bwv0/FdbtSG78LT7hAT6q21cmGx2h3bn1a7ZFREcoPv6pGp4n7aKPMvRF5vdoa05xVq1P19rDhRo9YECHSup+s3XXJIao9/3SdXV6S3k53fuoWB3XqmemLNRaw4USJIubhuj27u3VlxEkJqHBysuIkixYVYFBfxypYxhGHrl+516Y1Htf6Nv69ZKk2/rUm+f09lxpER/mLFaB/PLFRrkr1cHdVW/Ti0JOHC5M/3+NjXsLF68WFdeeeVJ24cNG6bp06fLMAw999xzeuedd1RYWKg+ffro73//uzp06ODcNz8/X6NGjdJXX30lPz8/DRw4UFOnTlV4+JnP+CbsoKk4Wlypvy3YqSqbQ5e0i1VGaqySYkLNLsvtjpVU6aXvtmv2mkOSpAA/i5JjQ5USF66UuNo/28aFKjUuXPGRVlksFjkchr7fekT/M3+XdhwtkVQ78nF/31Td07utwqynHhivrnFo0+FCLd+br+V7T2jNgYKTRlIC/S3q3CpKPdrGqEdyM3VPbqbsggq9u2yf/m9TrnN0qW1sqIZfmqLbu7c+7fv9FpvdoXeW7tVrC3ap2u5QWJC/nrzhfN11cZszDlH/XHFQE77YLLvDUJ/2cfr777r9ZuD7bvMRPfLpepVX25UUE6L/HdpD6S357yrco1GEHW9B2EFTsGLvCY385zodL60/qtC6WYgyUmOV0a72kRAVctpjlFTalFtUqZzCCuUWVarKZlf7FhFKaxmh5hFnPw+u0mbXnmOl2nusTBU2u2rshuwOx6/WVTFUY3c4111JiQs7q3k0NrtDH/y0X68t2KWSqhpJ0u3dW+vx69LUIiL4jI7hcBj6v825+tuCXc6RnpiwIP2xb6qGZrSVv59FGw8VavneE1q+t3b0psJWP9xEhQSqe3Iz9WjbTD2SY9SldZSCA/1P+X45hRX6IGu/PlpxUMWVtTVHBgfozovb6HeXJJ9xON18uEiPz96orbm1S2tc3qG5/nJbZ7WKPv3f7+ks2p6nkf9cq/Jqu9JbRuj94T1P+nficBh6feFu/c+CnZKk3u1i9eb/6+aROU9ougg7Z4GwA19mGIbeXbZPk77dLrvDUHrLCF2V3kLL957QxkNFJ01MbRsb6gw9uUUVyimsVG5RhXILK52B4VRiw4KU1rI2+KS3jFBay0h1iA9XaFCAyqpqtDuvVLvySrUrr0S7j5Zq97FSHcwv19n+F8jfz6KebZvp2o4tdU3H+NN++f971zH96autzoDSpXWUnr+5k7q1aXZ2b/gzu8PQVxty9NoPu7TveJkkKTo0UJU2uypt9S+fbhYaqF4psbokNUaXtItVhxYRZ92OKquq0Zy1h/Tesn3af6K83rHbtwhXu+bh9f5sFR0iPz+LKm12Tf1hl95euld2h6Ho0EA9e2NH3XpRq3NqI20+XKTh01fpWEmVWkYG6/3hPXV+QqSz1kc/3aDvttQu+XFP77Z6pv/5CnDxAnLAfyLsnAXCDnxVaVWNnpi9Ud9sypUk3XJhov5yW2eFBtW2RMqqarRqf76y9p7Q8j0ntOlwkf7bmm6RwQFKjA5RQlTtaqi78kq1/0TZKUOLxSLFhAbpRFn1aY8XHRqo9s3DFRkSKP+fLzUO8PdzXnIc6F/7pySt3l+g7UdK6r3+/IRIXdMxXtd2jFenxEgdKqjQC99s1bwtRyXVhrDHr0vTHd2TXDL/pcbu0OfrczT1h106mF8bQmLCgnRJaszPASdW57UId8l7SbUjJgu35+ndZfu0fN+J04ZDa4CfUpuHq6yqxllX/84Jev7mTg0adTuV7PxyDZ++SrvzShVhDdBbv+uu5NhQ/WHGam0/UqJAf4tevKWzBvVkKQ94BmHnLBB24It255XqgVlrtDuvVAF+Fj17U0fdfUnyb/6/++JKm1bty1fWnhMqrrQpISpEidHB9f481dyRimq7duWVaPuREu34+bH9SEm9lllcuFXntQjXefHhOq9FuNq3iNB58eGKDQs6qxGH7Pxyfb/1qL7fckSr9ufXC2etokN0vLRKVTUO+ftZdPclyRp7TQdFhbj+km6b3aGV+/LVPKL2c3li8m1FdW3bb8+xUu3JK9WeY2XanVd7JdmvL+1uHmHVxAEX6LoLTr8ER0MVldt0/8zVWrEvXwF+FoVZA1RUYVNcuFVv391N3ZNjXP6ewOkQds4CYQfuZBiGSqpqVFBWrRNl1Sooq1Z+WbUKyn/5vaSyRja7Q9V2Q7Yah2x2h/P3mp9/9rNY1Ll1lC5JrR09aBsbetov2G835eqxzzaorNqu+Eir/n6XOV9CJ0qrlFNYqdbNQtwydyO/rFoLt+fp+y1HtHTXMWc7KSM1Vs/f3ElpLZvGGi52h6Hs/HLtOVaqwnKbMs+Pd9ml8qdSVWPXY59t1FcbciTVtgjfvrv7b873AtyBsHMWCDtwpZJKm5bvzdePu4/rx93Hte942SkXbDtX8ZFWZ9vkktQYpcSFye4wNGXeDr2zdK8kqVdKjN74f91c1sbwZhXVdmXtPa7gQH9lpMZymbObORyG3v9pvwrKqjXqqvannXANuBNh5ywQdlCnqsauWcsP6pNVBxVmDVBqXLhSm4epXfMwpTYPV3JsqKwB9f+jXl3j0PrsQi3bfVzLdh3ThkNFp7yZYWiQv5qFBikm7JdH7e+BigoJVFCAnwL8/BQY4Kcgf4sC/f2cj6AAi8qr7Vq1v0DL957Q+oOFJ91TqHmEVc1CA7XzaO2E3Pv7purxfmlMEgXgsxr9Xc8BT7I7DH2x/rBe+X6nDhdWOLevO1hYbz8/i9S6WahSm4epbWyYDpwo04p9+Seto9I2NlSXto9Tn/Zx6pIUrdiwIJf8P9/LzmsuqfaS7XUH6y53PqF12YU6VlKlYyVVCgvy11/v6KobOif8l6MBQNPAyI4Y2WnKDMPQ4p3H9NK3251X+cRHWvXQVeepWWiQ9h4r1d7jZdp7rHYyaOlpLr2ODQtS7/Zx6tM+Vr3bxXl8ob5Km13rswu1K69Ufc+LU3JsmEffHwDMwMgOmhTDMPT91qPacrhIKc3DlBYfqXYtwk5qOf3auoMFmvztdq3Yly+p9pYAD17RTsN7pygk6OTXGYahYyVV2nOsTHuPl2r/8TK1iAjWpe3jlN7y7NdRcaXgQH/nxGUAQH2EHTRqdSMzr3y/Q5sPF9d7LsDPopS4sHqL3KW3jFBVjV0vz9vpXAAtKMBP9/RuqxFXtFN06OmvGLJYLGoRGawWkcHKaEeoAIDGgrCDRmvF3hN6+fsdWrW/9uaGYUH+uqZjvA4XVmj7kRKVVNb8vGJvqb7emHvS6/0s0sBurTX2mg5KbMAS+gCAxoGwg0ZnQ3ahXv5+h/6967ik2pVjh2Yk64HL2yk2vPYSa8MwlFtUqR1H6y9ytyevVNV2hzLPj9fj16WpQ3zTWIcFAJoywg4ajR1HSvTK9zv0/dba2wAE+Fk05OIkjbryPLWMqn9TR4vFosToECVGh+jKtBbO7Ta7Q+VVdrcuuAYA8C6EHXgFm92h/LJqHSup0vHSKh0v/fXPVTpSVKmV+/NlGLXtp1suaqUxV3dQm9izu+op0N9PUaGsOwMATQlhB6apqLbr41UHNf2n/Trwq7s6/5brL2ipR67poPNoPwEAzhBhBx5XVGHTzKz9eu/H/cr/1d2w/SxSbLhVceFWxYUHqXm4Vc0jfv49IkidEqOYYwMAOGuEHXjMsZIqvbtsn2YtP+BcnK9NTKj+eHmq+nVqqZjQIFPXqgEA+CbCDtwuO79c7yzdq09XZ6uqpvZ+TmnxERpxZTv175zAvZsAAG5F2IHb5JVUavK32/XF+hznjTEvahOtkVe011XpLRjFAQB4BGEHbrF4R54e/XSDTvw8J+ey8+I04or2uiQ1RhYLIQcA4DmEHbhUdY1Df523Xf/7732SpPSWEZo8sIsuTIo2tzAAQJNF2IHL7D9epoc+WqdNh4skSff0bqsnr09XcODpb8YJAIC7EXbgEnPXHdIzczerrNqu6NBA/fX2rrqmY7zZZQEAQNjBuSmtqtGzX2zWv9YeliRdnBKj14ZcqIQobqwJAPAOhB002KZDRXr443Xad7xMfhZp9NUdNOqq9vLnKisAgBch7KBB5q47pMdnb5TNbigxKlh/G3KRLk6JMbssAABOQtjBWZu/9age+2yj7A5D13aM15Tbuyg6NMjssgAAOCXCDs7Kir0nNOqfa2V3GLqje2tNub0L6+YAALwa6/TjjG3NKdZ9H6xWVY1DmefHa9JtnQk6AACvR9jBGTl4olxD31upkqoaXdw2Rm/8v4u4pxUAoFHg2wr/VV5JpX737godL63S+QmR+t9hPVgoEADQaBB28JuKK20a9t4qHcwvV5uYUH1wb09FhQSaXRYAAGeMsIPTqrTZdd8Hq7Utt1hx4VbN/P3FahERbHZZAACcFcIOTqnG7tBDH63Tyn35irAG6IN7eyo5NszssgAAOGuEHZzEMAw9NXeT5m89qqAAP/3vsB7qlBhldlkAADQI6+ygHofD0AvfbNOnqw/JzyK9fudFuiQ11uyyAABoMMIOnCptdj366QZ9sylXkjT5ti7q16mlyVUBAHBuCDuQJOWXVesPM1ZrzYECBfpbNPm2LhrYvbXZZQEAcM4IO9C+42Ua/v5K7T9RrsjgAL19dw9ltKN1BQDwDYSdJm7V/nz9YcZqFZbb1LpZiKYP76n2LSLMLgsAAJch7DRhX6w/rHGfbVS13aGuSdH6x9Aeah5hNbssAABcirDTBBmGob8v3qO/ztshSerXKV5/G3yRQoK4BQQAwPcQdpoYm92hZ+Zu1iersyVJv++ToqduOF/+fty9HADgmwg7TYjDYeiPM9do4fY8+Vmk52/upKEZbc0uCwAAtyLsNCFz1h7Swu15Cg700xt3dlNmx3izSwIAwO24XUQTUVxp00vf1c7ReeSaDgQdAECTQdhpIl7/YZeOl1YpNS5M9/ROMbscAAA8hrDTBOzOK9X7P+6XJD17U0cFBfDXDgBoOvjW83GGYejPX29VjcNQ5vktdEVaC7NLAgDAowg7Pu6HbXlauvOYgvz99Ez/jmaXAwCAxxF2fFilza4/f71VkvT7y1LUNi7M5IoAAPA8rw87JSUlGjNmjJKTkxUSEqLevXtr1apVzudLS0s1atQotW7dWiEhIerYsaOmTZtmYsXe491l+3Qwv1zxkVaNurK92eUAAGAKr19n57777tPmzZs1c+ZMJSYmatasWcrMzNTWrVvVqlUrPfLII1q4cKFmzZqltm3b6vvvv9eIESOUmJiom2++2ezyTZNbVKE3Fu6WJI2//nyFWb3+rxoAALfw6pGdiooKzZkzR1OmTFHfvn3Vvn17Pf/882rfvr3eeustSdJPP/2kYcOG6YorrlDbtm11//33q2vXrlq5cqXJ1Ztr8rfbVWGzq3tyMw24MNHscgAAMI1Xh52amhrZ7XYFBwfX2x4SEqJly5ZJknr37q0vv/xShw8flmEYWrRokXbu3Klrr732tMetqqpScXFxvYcvWbU/X1+sz5HFIv3p5k6yWLjvFQCg6fLqsBMREaGMjAxNnDhROTk5stvtmjVrlrKyspSbmytJev3119WxY0e1bt1aQUFBuu666/Tmm2+qb9++pz3upEmTFBUV5XwkJSV56iO5nd1h6Pkvt0iShvRM0gWtokyuCAAAc3l12JGkmTNnyjAMtWrVSlarVVOnTtWdd94pP7/a0l9//XUtX75cX375pdasWaNXXnlFI0eO1IIFC057zPHjx6uoqMj5yM7O9tTHcbtPVmVrS06xIoID9Ni1aWaXAwCA6SyGYRhmF3EmysrKVFxcrISEBA0ePFilpaWaPXu2oqKiNHfuXPXv39+573333adDhw7pu+++O6NjFxcXKyoqSkVFRYqMjHTXR3C7onKbrnh5kQrKbXrupo4afim3hQAA+K4z/f72+pGdOmFhYUpISFBBQYHmzZunAQMGyGazyWazOUd56vj7+8vhcJhUqXn+Z8FOFZTb1CE+XL+7JNnscgAA8Apefz3yvHnzZBiG0tLStHv3bo0bN07p6ekaPny4AgMDdfnll2vcuHEKCQlRcnKylixZohkzZujVV181u3SP2nOsVDOXH5AkPXdTJwX6N5ocCwCAW3l92CkqKtL48eN16NAhxcTEaODAgXrxxRcVGBgoSfr44481fvx43XXXXcrPz1dycrJefPFFPfDAAyZX7lnvLNkru8PQ1ektdGn7OLPLAQDAazSaOTvu1Njn7OQVV6rPS4tUbXdozoMZ6p4cY3ZJAAC4nc/N2cHpvf/TflXbHeqR3IygAwDAfyDsNHIllTbN+nmuzh8vb2dyNQAAeB/CTiP38cpslVTWqF3zMF2d3sLscgAA8DqEnUasusahd5ftkyT9sW87+flxWwgAAP4TYacR+3JDjo4UV6pFhFUDLuJmnwAAnAphp5FyOAy9s3SPJOnePimyBvibXBEAAN6JsNNILd6Zp51HSxVuDdD/69XG7HIAAPBahJ1G6u0leyVJ/69XG0UGB5pcDQAA3ouw0witO1igFfvyFehv0fBL25pdDgAAXo2w0wi9s7R2VGfAha2UEBVicjUAAHg3wk4js+94mb7bckSSdH/fVJOrAQDA+xF2Gpn//fdeGYZ0dXoLdYiPMLscAAC8HmGnETlWUqXZaw5J4tYQAACcKcJOIzIja7+qaxy6qE20erZtZnY5AAA0CoSdRqKsqkYzsn6+4WffVFks3BoCAIAzQdhpJD5Zla2iCptS4sJ0TceWZpcDAECjQdhpBGz2X274+YfLUuXPDT8BADhjhJ1G4PstR3W4sEJx4UG6rVsrs8sBAKBRIew0Ast2H5ck3XpRKwUHcsNPAADOBmGnEVh3sECS1D05xuRKAABofAg7Xq6k0qYdR0skSd2So80tBgCARoiw4+XWZxfKMKTWzULUIiLY7HIAAGh0CDtebu2BQklStzYsIggAQEMQdrzcWud8HcIOAAANQdjxYg6H4ZyczMgOAAANQ9jxYnuPl6q4skbBgX5KT+AO5wAANARhx4vVzdfp0jpagf78VQEA0BB8g3qxtbSwAAA4Z4QdL/ZL2Ik2txAAABoxwo6XKq60aVdeqSSpG1diAQDQYIQdL7X+YO1igm1iQhUXbjW7HAAAGi3CjpeihQUAgGsQdrzU2oOFkmhhAQBwrgg7XojFBAEAcB3Cjhfac6xUJZU1Cgn0V3pLFhMEAOBcEHa8UN18nS6toxTAYoIAAJwTvkm90JoDP7ewmK8DAMA5I+x4IefkZObrAABwzgg7Xqao3KbdPy8meBGXnQMAcM4IO15mXXZtCys5lsUEAQBwBcKOl6GFBQCAaxF2vMw6Vk4GAMClCDtexOEwtJ6VkwEAcCnCjhfZlVeqkqoahQb5Ky2exQQBAHAFwo4XqVtMsGvraBYTBADARfhG9SJrnYsJRptbCAAAPoSw40XWcvNPAABcjrDjJQrLq7XnWJkk6SLCDgAALkPY8RLrsgslSSlxYYoJCzK3GAAAfAhhx0us+3m+DreIAADAtQg7XoKVkwEAcA/CjhewO4xfrZxM2AEAwJW8PuyUlJRozJgxSk5OVkhIiHr37q1Vq1bV22fbtm26+eabFRUVpbCwMPXs2VMHDx40qeKzt/Noicqq7QoL8ldaSxYTBADAlbw+7Nx3332aP3++Zs6cqU2bNunaa69VZmamDh8+LEnas2eP+vTpo/T0dC1evFgbN27UhAkTFBwcbHLlZ865mGBStPz9LCZXAwCAb7EYhmGYXcTpVFRUKCIiQl988YX69+/v3N69e3ddf/31euGFFzRkyBAFBgZq5syZDX6f4uJiRUVFqaioSJGRka4o/aw8+ukGzVl7SKOubK/H+qV5/P0BAGiMzvT726tHdmpqamS3208apQkJCdGyZcvkcDj0zTffqEOHDurXr59atGihXr166fPPP//N41ZVVam4uLjew0zO+TqsnAwAgMt5ddiJiIhQRkaGJk6cqJycHNntds2aNUtZWVnKzc1VXl6eSktLNXnyZF133XX6/vvvdeutt+q2227TkiVLTnvcSZMmKSoqyvlISkry4Keqr6CsWnuP/7yYYBKTkwEAcDWvDjuSNHPmTBmGoVatWslqtWrq1Km688475efnJ4fDIUkaMGCAxo4dqwsvvFBPPvmkbrzxRk2bNu20xxw/fryKioqcj+zsbE99nJOsy64d1UmNC1MzFhMEAMDlvD7stGvXTkuWLFFpaamys7O1cuVK2Ww2paamKi4uTgEBAerYsWO915x//vm/eTWW1WpVZGRkvYdZ1h4olMQtIgAAcBevDzt1wsLClJCQoIKCAs2bN08DBgxQUFCQevbsqR07dtTbd+fOnUpOTjap0rOz/0RtC+v8BC45BwDAHQLMLuC/mTdvngzDUFpamnbv3q1x48YpPT1dw4cPlySNGzdOgwcPVt++fXXllVfqu+++01dffaXFixebW/gZKq+2S5Iigr3+rwIAgEbJ60d2ioqKNHLkSKWnp2vo0KHq06eP5s2bp8DAQEnSrbfeqmnTpmnKlCnq3Lmz/vGPf2jOnDnq06ePyZWfmfLqGklSaBBhBwAAd/DqdXY8xcx1dm5+Y5k2HirSu8N66Orz4z363gAANGY+sc5OU1DXxgoJ8je5EgAAfBNhx2QVP4edMNpYAAC4BWHHZGXOOTuM7AAA4A6EHZPRxgIAwL0IOyaqsTtUXVO7CjRtLAAA3IOwY6Jym935MyM7AAC4B2HHRHWTk/0skjWAvwoAANyhQd+wixYtcnUdTVL5r67EslgsJlcDAIBvalDYue6669SuXTu98MILpt4xvLErq6q9EosWFgAA7tOgsHP48GGNGjVKs2fPVmpqqvr166dPP/1U1dXVrq7Pp1X8PGeHy84BAHCfBoWduLg4jR07VuvXr9eKFSvUoUMHjRgxQomJiXr44Ye1YcMGV9fpk+pGdrgvFgAA7nPOs2K7deum8ePHa9SoUSotLdV7772n7t2767LLLtOWLVtcUaPPqpugzMgOAADu0+CwY7PZNHv2bN1www1KTk7WvHnz9MYbb+jo0aPavXu3kpOTdccdd7iyVp9TN0E51MrIDgAA7tKgb9mHHnpIH330kQzD0N13360pU6boggsucD4fFhaml19+WYmJiS4r1BeV190qIpCRHQAA3KVBYWfr1q16/fXXddttt8lqtZ5yn7i4OC5R/y/KaWMBAOB2DQo7P/zww38/cECALr/88oYcvsn4pY1F2AEAwF0aNGdn0qRJeu+9907a/t577+mll14656KaCmcbi6uxAABwmwaFnbffflvp6eknbe/UqZOmTZt2zkU1Fc47njNnBwAAt2lQ2Dly5IgSEhJO2t68eXPl5uaec1FNRd2l52G0sQAAcJsGhZ2kpCT9+OOPJ23/8ccfuQLrLJRV190ugjYWAADu0qBv2T/84Q8aM2aMbDabrrrqKkm1k5Yff/xxPfrooy4t0Jc5JyjTxgIAwG0aFHbGjRunEydOaMSIEc77YQUHB+uJJ57Q+PHjXVqgL6ONBQCA+zUo7FgsFr300kuaMGGCtm3bppCQEJ133nmnXXMHp1ZWN0GZNhYAAG5zTt+y4eHh6tmzp6tqaXIqnJeeM7IDAIC7NDjsrF69Wp9++qkOHjzobGXV+de//nXOhTUFrKAMAID7NehqrI8//li9e/fWtm3bNHfuXNlsNm3ZskULFy5UVFSUq2v0Wb+EHdpYAAC4S4PCzl/+8hf9z//8j7766isFBQXptdde0/bt2zVo0CC1adPG1TX6JMMwfrWCMiM7AAC4S4PCzp49e9S/f39JUlBQkMrKymSxWDR27Fi98847Li3QV1XVOOQwan8m7AAA4D4NCjvNmjVTSUmJJKlVq1bavHmzJKmwsFDl5eWuq86H1bWwJNpYAAC4U4O+Zfv27av58+erc+fOuuOOOzR69GgtXLhQ8+fP19VXX+3qGn1SXQsrKMBP/n4Wk6sBAMB3NSjsvPHGG6qsrJQkPf300woMDNRPP/2kgQMH6plnnnFpgb7KuaAgLSwAANzqrMNOTU2Nvv76a/Xr10+S5OfnpyeffNLlhfm6Mq7EAgDAI856zk5AQIAeeOAB58gOGqbceRNQRnYAAHCnBk1Qvvjii7V+/XoXl9K00MYCAMAzGtRDGTFihB555BFlZ2ere/fuCgsLq/d8ly5dXFKcL/vlvliEHQAA3KlBYWfIkCGSpIcffti5zWKxyDAMWSwW2e32070UP/vlvljM2QEAwJ0a9E27b98+V9fR5HBfLAAAPKNBYSc5OdnVdTQ5hB0AADyjQWFnxowZv/n80KFDG1RMU1JOGwsAAI9o0Dft6NGj6/1us9lUXl6uoKAghYaGEnbOACM7AAB4RoMuPS8oKKj3KC0t1Y4dO9SnTx999NFHrq7RJ5VXEXYAAPCEBoWdUznvvPM0efLkk0Z9cGrltrpLz2ljAQDgTi4LO1Lt6so5OTmuPKTPqrv0nEUFAQBwrwYNK3z55Zf1fjcMQ7m5uXrjjTd06aWXuqQwX1dWxaKCAAB4QoPCzi233FLvd4vFoubNm+uqq67SK6+84oq6fF5dGyuMNhYAAG7VoG9ah8Ph6jqanF9WUGZkBwAAd3LpnB2cOdpYAAB4RoPCzsCBA/XSSy+dtH3KlCm64447zrmopqCiro1lpY0FAIA7NSjsLF26VDfccMNJ26+//notXbr0nItqCupWUA4JZGQHAAB3alDYKS0tVVBQ0EnbAwMDVVxcfM5F+Tq7w1ClrXbeE3N2AABwrwaFnc6dO+uTTz45afvHH3+sjh07nnNRvq6uhSXRxgIAwN0aFHYmTJigiRMnatiwYfrggw/0wQcfaOjQoXrxxRc1YcIElxZYUlKiMWPGKDk5WSEhIerdu7dWrVp1yn0feOABWSwW/e1vf3NpDa5W18KyWCRrAHPEAQBwpwYNK9x00036/PPP9Ze//EWzZ89WSEiIunTpogULFujyyy93aYH33XefNm/erJkzZyoxMVGzZs1SZmamtm7dqlatWjn3mzt3rpYvX67ExESXvr87OO+LFegvi8VicjUAAPi2BvdQ+vfvr/79+7uylpNUVFRozpw5+uKLL9S3b19J0vPPP6+vvvpKb731ll544QVJ0uHDh/XQQw9p3rx5bq/JFZx3PKeFBQCA2zXo23bVqlVyOBzq1atXve0rVqyQv7+/evTo4ZLiampqZLfbFRwcXG97SEiIli1bJql2gcO7775b48aNU6dOnc7ouFVVVaqqqnL+7ulJ1RU2FhQEAMBTGjRhZOTIkcrOzj5p++HDhzVy5MhzLqpORESEMjIyNHHiROXk5Mhut2vWrFnKyspSbm6uJOmll15SQECAHn744TM+7qRJkxQVFeV8JCUluazmM+FcUJDLzgEAcLsGhZ2tW7eqW7duJ22/6KKLtHXr1nMu6tdmzpwpwzDUqlUrWa1WTZ06VXfeeaf8/Py0Zs0avfbaa5o+ffpZzX0ZP368ioqKnI9TBTd3qmtjcSUWAADu16CwY7VadfTo0ZO25+bmKiDAtV/g7dq105IlS1RaWqrs7GytXLlSNptNqamp+ve//628vDy1adNGAQEBCggI0IEDB/Too4+qbdu2v1l/ZGRkvYcn0cYCAMBzGhR2rr32WufoSJ3CwkI99dRTuuaaa1xW3K+FhYUpISFBBQUFmjdvngYMGKC7775bGzdu1Pr1652PxMREjRs3TvPmzXNLHa5AGwsAAM9p0DDMyy+/rL59+yo5OVkXXXSRJGn9+vWKj4/XzJkzXVrgvHnzZBiG0tLStHv3bo0bN07p6ekaPny4AgMDFRsbW2//wMBAtWzZUmlpaS6tw5UqaGMBAOAxDfq2bdWqlTZu3KgPP/xQGzZsUEhIiIYPH64777xTgYGBLi2wqKhI48eP16FDhxQTE6OBAwfqxRdfdPn7eFLdnB3ueA4AgPs1eGghLCxMffr0UZs2bVRdXS1J+vbbbyVJN998s2uqkzRo0CANGjTojPffv3+/y97bXepWUA6ljQUAgNs1KOzs3btXt956qzZt2iSLxSLDMOpdDWW323/j1WBRQQAAPKdBE5RHjx6tlJQU5eXlKTQ0VJs3b9aSJUvUo0cPLV682MUl+h5n2KGNBQCA2zVoaCErK0sLFy5UXFyc/Pz85O/vrz59+mjSpEl6+OGHtW7dOlfX6VOcbSzCDgAAbtegkR273a6IiAhJUlxcnHJyciRJycnJ2rFjh+uq81G/jOzQxgIAwN0a9G17wQUXaMOGDUpJSVGvXr00ZcoUBQUF6Z133lFqaqqra/Q5FbSxAADwmAaFnWeeeUZlZWWSpD//+c+68cYbddlllyk2NlaffPKJSwv0RWU/t7G49BwAAPdrUNjp16+f8+f27dtr+/btys/PV7Nmzc7qHlVNlXNRQdpYAAC4ncu+bWNiYlx1KJ/H1VgAAHhOgyYo49zQxgIAwHMIOyagjQUAgOcQdjysusahGochiZEdAAA8gbDjYXULCkrM2QEAwBMIOx5WNzk5yN9Pgf6cfgAA3I1vWw+rCzu0sAAA8AzCjodxXywAADyLsONhrLEDAIBnEXY8rIKbgAIA4FGEHQ8ro40FAIBHEXY8jDYWAACeRdjxsPKqupEd2lgAAHgCYcfDym2M7AAA4EmEHQ+roI0FAIBHEXY8rKyqblFB2lgAAHgCYcfDKmy1c3bCGNkBAMAjCDsexu0iAADwLMKOh9W1sbgaCwAAzyDseJizjWVlZAcAAE8g7HiYs40VSNgBAMATCDseVk4bCwAAjyLseFj5z22sUNpYAAB4BGHHw1hUEAAAzyLseJjzaqxA2lgAAHgCYceDHA5DFXX3xqKNBQCARxB2PKiyxu78mTYWAACeQdjxoLoWliQFBxB2AADwBMKOB/16crKfn8XkagAAaBoIOx7kvOycFhYAAB5D2PGgujYWNwEFAMBzCDseVNfGCmP1ZAAAPIaw40Hl1bVtLEZ2AADwHMKOB5WzejIAAB5H2PGgX8IObSwAADyFsONBdW0sRnYAAPAcwo4H0cYCAMDzCDseRBsLAADPI+x4UAVtLAAAPI6w40Fl1SwqCACApxF2PIhFBQEA8DzCjgexqCAAAJ5H2PGgMq7GAgDA4wg7HkQbCwAAzyPseBBtLAAAPM/rw05JSYnGjBmj5ORkhYSEqHfv3lq1apUkyWaz6YknnlDnzp0VFhamxMREDR06VDk5OSZXfWrljOwAAOBxXh927rvvPs2fP18zZ87Upk2bdO211yozM1OHDx9WeXm51q5dqwkTJmjt2rX617/+pR07dujmm282u+xTKufScwAAPM5iGIZhdhGnU1FRoYiICH3xxRfq37+/c3v37t11/fXX64UXXjjpNatWrdLFF1+sAwcOqE2bNmf0PsXFxYqKilJRUZEiIyNdVv9/6vD0t6q2O/TTk1cpMTrEbe8DAEBTcKbf317dT6mpqZHdbldwcHC97SEhIVq2bNkpX1NUVCSLxaLo6OjTHreqqkpVVVXO34uLi11S72+x2R2qtjsk0cYCAMCTvLqNFRERoYyMDE2cOFE5OTmy2+2aNWuWsrKylJube9L+lZWVeuKJJ3TnnXf+ZsKbNGmSoqKinI+kpCR3fgxJv7SwJNpYAAB4kleHHUmaOXOmDMNQq1atZLVaNXXqVN15553y86tfus1m06BBg2QYht56663fPOb48eNVVFTkfGRnZ7vzI0j65bLzAD+LggK8/rQDAOAzvL6f0q5dOy1ZskRlZWUqLi5WQkKCBg8erNTUVOc+dUHnwIEDWrhw4X+dd2O1WmW1Wt1dej1l3AQUAABTNJohhrCwMCUkJKigoEDz5s3TgAEDJP0SdHbt2qUFCxYoNjbW5EpPrcK5erLX50sAAHyK13/zzps3T4ZhKC0tTbt379a4ceOUnp6u4cOHy2az6fbbb9fatWv19ddfy26368iRI5KkmJgYBQUFmVz9L8q5VQQAAKbw+rBTVFSk8ePH69ChQ4qJidHAgQP14osvKjAwUPv379eXX34pSbrwwgvrvW7RokW64oorPF/waTjbWFbCDgAAnuT1YWfQoEEaNGjQKZ9r27atvHiZoHqcbaxArz/lAAD4lEYzZ6exY/VkAADMQdjxkLqbgIbRxgIAwKMIOx7iHNmhjQUAgEcRdjyEq7EAADAHYcdDyqu4GgsAADMQdjyk3MbVWAAAmIGw4yEVtLEAADAFYcdDymhjAQBgCsKOh1TYGNkBAMAMhB0P4dJzAADMQdjxkLo2FosKAgDgWYQdD6GNBQCAOQg7HkIbCwAAcxB2PKScNhYAAKYg7HiAYRjORQW56zkAAJ5F2PGAqhqHDKP259Ag2lgAAHgSYccD6q7EkqSQQEZ2AADwJMKOB9RNTg4O9JO/n8XkagAAaFoIOx7wy2XntLAAAPA0wo4HOO+LxeRkAAA8jrDjAdzxHAAA8xB2PKC8mjYWAABmIex4QFk1bSwAAMxC2PEA2lgAAJiHsOMBtLEAADAPYccDymljAQBgGsKOBzjveE7YAQDA4wg7HlAXdsJoYwEA4HGEHQ+oa2MxsgMAgOcRdjygnKuxAAAwDWHHA2hjAQBgHsKOB9DGAgDAPIQdD2BRQQAAzEPY8YAyFhUEAMA0hB0PYGQHAADzEHY8gBWUAQAwD2HHA5xtLCttLAAAPI2w42Z2h6HqGockKTSQkR0AADyNsONmdS0siUvPAQAwA2HHzeoWFPT3s8gawOkGAMDT+PZ1M+etIgL9ZbFYTK4GAICmh7DjZqyeDACAuQg7bua8LxZXYgEAYArCjpvVhZ0QrsQCAMAUhB03q2BBQQAATEXYcbOyKhYUBADATIQdNyu3/XI1FgAA8DzCjpvRxgIAwFyEHTf7pY1F2AEAwAyEHTerqGtjBTFnBwAAMxB23My5qCBzdgAAMIXXh52SkhKNGTNGycnJCgkJUe/evbVq1Srn84Zh6Nlnn1VCQoJCQkKUmZmpXbt2mVhxfeVVdYsKEnYAADCD14ed++67T/Pnz9fMmTO1adMmXXvttcrMzNThw4clSVOmTNHUqVM1bdo0rVixQmFhYerXr58qKytNrryWc1FB2lgAAJjCq8NORUWF5syZoylTpqhv375q3769nn/+ebVv315vvfWWDMPQ3/72Nz3zzDMaMGCAunTpohkzZignJ0eff/652eVL4tJzAADM5tVhp6amRna7XcHBwfW2h4SEaNmyZdq3b5+OHDmizMxM53NRUVHq1auXsrKyTnvcqqoqFRcX13u4S3lV7Zwd2lgAAJjDq8NORESEMjIyNHHiROXk5Mhut2vWrFnKyspSbm6ujhw5IkmKj4+v97r4+Hjnc6cyadIkRUVFOR9JSUlu+wy0sQAAMJdXhx1JmjlzpgzDUKtWrWS1WjV16lTdeeed8vNreOnjx49XUVGR85Gdne3Ciuuru/Q8jEUFAQAwhdeHnXbt2mnJkiUqLS1Vdna2Vq5cKZvNptTUVLVs2VKSdPTo0XqvOXr0qPO5U7FarYqMjKz3cJeyn9tYIYQdAABM4fVhp05YWJgSEhJUUFCgefPmacCAAUpJSVHLli31ww8/OPcrLi7WihUrlJGRYWK1v6ioZlFBAADM5PXfwPPmzZNhGEpLS9Pu3bs1btw4paena/jw4bJYLBozZoxeeOEFnXfeeUpJSdGECROUmJioW265xezSZRiG82os2lgAAJjD68NOUVGRxo8fr0OHDikmJkYDBw7Uiy++qMDAQEnS448/rrKyMt1///0qLCxUnz599N133510BZcZqmocsjsMSbSxAAAwi8UwDMPsIsxWXFysqKgoFRUVuXT+TkFZtS6aOF+StOcvN8jfz+KyYwMA0NSd6fd3o5mz0xjVtbCsAX4EHQAATELYcaO6BQVDaWEBAGAawo4blXMlFgAApiPsuNEvYYeRHQAAzELYcaPyatpYAACYjbDjRr/cF4uwAwCAWQg7blS3enIYc3YAADANYceNyqq5LxYAAGYj7LgRE5QBADAfYceNuAkoAADmI+y4URlXYwEAYDrCjhtV0MYCAMB0hB03YgVlAADMR9hxIxYVBADAfIQdN2JRQQAAzEfYcaNyFhUEAMB0hB03oo0FAID5CDtu5O/np0B/C20sAABMRH/Fjb4dfZkkyTAMkysBAKDpYmTHAywWi9klAADQZBF2AACATyPsAAAAn0bYAQAAPo2wAwAAfBphBwAA+DTCDgAA8GmEHQAA4NMIOwAAwKcRdgAAgE8j7AAAAJ9G2AEAAD6NsAMAAHwaYQcAAPi0ALML8AaGYUiSiouLTa4EAACcqbrv7brv8dMh7EgqKSmRJCUlJZlcCQAAOFslJSWKioo67fMW47/FoSbA4XAoJydHERERslgsLjtucXGxkpKSlJ2drcjISJcdF6fG+fYszrdncb49i/PtWQ0934ZhqKSkRImJifLzO/3MHEZ2JPn5+al169ZuO35kZCT/Y/Egzrdncb49i/PtWZxvz2rI+f6tEZ06TFAGAAA+jbADAAB8GmHHjaxWq5577jlZrVazS2kSON+exfn2LM63Z3G+Pcvd55sJygAAwKcxsgMAAHwaYQcAAPg0wg4AAPBphB0AAODTCDtu9Oabb6pt27YKDg5Wr169tHLlSrNL8glLly7VTTfdpMTERFksFn3++ef1njcMQ88++6wSEhIUEhKizMxM7dq1y5xiG7lJkyapZ8+eioiIUIsWLXTLLbdox44d9faprKzUyJEjFRsbq/DwcA0cOFBHjx41qeLG76233lKXLl2ci6tlZGTo22+/dT7P+XafyZMny2KxaMyYMc5tnG/Xev7552WxWOo90tPTnc+763wTdtzkk08+0SOPPKLnnntOa9euVdeuXdWvXz/l5eWZXVqjV1ZWpq5du+rNN9885fNTpkzR1KlTNW3aNK1YsUJhYWHq16+fKisrPVxp47dkyRKNHDlSy5cv1/z582Wz2XTttdeqrKzMuc/YsWP11Vdf6bPPPtOSJUuUk5Oj2267zcSqG7fWrVtr8uTJWrNmjVavXq2rrrpKAwYM0JYtWyRxvt1l1apVevvtt9WlS5d62znfrtepUyfl5uY6H8uWLXM+57bzbcAtLr74YmPkyJHO3+12u5GYmGhMmjTJxKp8jyRj7ty5zt8dDofRsmVL469//atzW2FhoWG1Wo2PPvrIhAp9S15eniHJWLJkiWEYtec2MDDQ+Oyzz5z7bNu2zZBkZGVlmVWmz2nWrJnxj3/8g/PtJiUlJcZ5551nzJ8/37j88suN0aNHG4bBv293eO6554yuXbue8jl3nm9Gdtygurpaa9asUWZmpnObn5+fMjMzlZWVZWJlvm/fvn06cuRIvXMfFRWlXr16ce5doKioSJIUExMjSVqzZo1sNlu9852enq42bdpwvl3Abrfr448/VllZmTIyMjjfbjJy5Ej179+/3nmV+PftLrt27VJiYqJSU1N111136eDBg5Lce765EagbHD9+XHa7XfHx8fW2x8fHa/v27SZV1TQcOXJEkk557uueQ8M4HA6NGTNGl156qS644AJJtec7KChI0dHR9fblfJ+bTZs2KSMjQ5WVlQoPD9fcuXPVsWNHrV+/nvPtYh9//LHWrl2rVatWnfQc/75dr1evXpo+fbrS0tKUm5urP/3pT7rsssu0efNmt55vwg6AMzJy5Eht3ry5Xn8d7pGWlqb169erqKhIs2fP1rBhw7RkyRKzy/I52dnZGj16tObPn6/g4GCzy2kSrr/+eufPXbp0Ua9evZScnKxPP/1UISEhbntf2lhuEBcXJ39//5NmkB89elQtW7Y0qaqmoe78cu5da9SoUfr666+1aNEitW7d2rm9ZcuWqq6uVmFhYb39Od/nJigoSO3bt1f37t01adIkde3aVa+99hrn28XWrFmjvLw8devWTQEBAQoICNCSJUs0depUBQQEKD4+nvPtZtHR0erQoYN2797t1n/fhB03CAoKUvfu3fXDDz84tzkcDv3www/KyMgwsTLfl5KSopYtW9Y798XFxVqxYgXnvgEMw9CoUaM0d+5cLVy4UCkpKfWe7969uwIDA+ud7x07dujgwYOcbxdyOByqqqrifLvY1VdfrU2bNmn9+vXOR48ePXTXXXc5f+Z8u1dpaan27NmjhIQE9/77PqfpzTitjz/+2LBarcb06dONrVu3Gvfff78RHR1tHDlyxOzSGr2SkhJj3bp1xrp16wxJxquvvmqsW7fOOHDggGEYhjF58mQjOjra+OKLL4yNGzcaAwYMMFJSUoyKigqTK298HnzwQSMqKspYvHixkZub63yUl5c793nggQeMNm3aGAsXLjRWr15tZGRkGBkZGSZW3bg9+eSTxpIlS4x9+/YZGzduNJ588knDYrEY33//vWEYnG93+/XVWIbB+Xa1Rx991Fi8eLGxb98+48cffzQyMzONuLg4Iy8vzzAM951vwo4bvf7660abNm2MoKAg4+KLLzaWL19udkk+YdGiRYakkx7Dhg0zDKP28vMJEyYY8fHxhtVqNa6++mpjx44d5hbdSJ3qPEsy3n//fec+FRUVxogRI4xmzZoZoaGhxq233mrk5uaaV3Qjd++99xrJyclGUFCQ0bx5c+Pqq692Bh3D4Hy723+GHc63aw0ePNhISEgwgoKCjFatWhmDBw82du/e7XzeXefbYhiGcW5jQwAAAN6LOTsAAMCnEXYAAIBPI+wAAACfRtgBAAA+jbADAAB8GmEHAAD4NMIOAADwaYQdAPgPixcvlsViOekePQAaJ8IOAADwaYQdAADg0wg7ALyOw+HQpEmTlJKSopCQEHXt2lWzZ8+W9EuL6ZtvvlGXLl0UHBysSy65RJs3b653jDlz5qhTp06yWq1q27atXnnllXrPV1VV6YknnlBSUpKsVqvat2+vd999t94+a9asUY8ePRQaGqrevXtrx44d7v3gANyCsAPA60yaNEkzZszQtGnTtGXLFo0dO1a/+93vtGTJEuc+48aN0yuvvKJVq1apefPmuummm2Sz2STVhpRBgwZpyJAh2rRpk55//nlNmDBB06dPd75+6NCh+uijjzR16lRt27ZNb7/9tsLDw+vV8fTTT+uVV17R6tWrFRAQoHvvvdcjnx+Aa3EjUABepaqqSjExMVqwYIEyMjKc2++77z6Vl5fr/vvv15VXXqmPP/5YgwcPliTl5+erdevWmj59ugYNGqS77rpLx44d0/fff+98/eOPP65vvvlGW7Zs0c6dO5WWlqb58+crMzPzpBoWL16sK6+8UgsWLNDVV18tSfq///s/9e/fXxUVFQoODnbzWQDgSozsAPAqu3fvVnl5ua655hqFh4c7HzNmzNCePXuc+/06CMXExCgtLU3btm2TJG3btk2XXnppveNeeuml2rVrl+x2u9avXy9/f39dfvnlv1lLly5dnD8nJCRIkvLy8s75MwLwrACzCwCAXystLZUkffPNN2rVqlW956xWa73A01AhISFntF9gYKDzZ4vFIql2PhGAxoWRHQBepWPHjrJarTp48KDat29f75GUlOTcb/ny5c6fCwoKtHPnTp1//vmSpPPPP18//vhjveP++OOP6tChg/z9/dW5c2c5HI56c4AA+C5GdgB4lYiICD322GMaO3asHA6H+vTpo6KiIv3444+KjIxUcnKyJOnPf/6zYmNjFR8fr6efflpxcXG65ZZbJEmPPvqoevbsqYkTJ2rw4MHKysrSG2+8ob///e+SpLZt22rYsGG69957NXXqVHXt2lUHDhxQXl6eBg0aZNZHB+AmhB0AXmfixIlq3ry5Jk2apL179yo6OlrdunXTU0895WwjTZ48WaNHj9auXbt04YUX6quvvlJQUJAkqVu3bvr000/17LPPauLEiUpISNCf//xn3XPPPc73eOutt/TUU09pxIgROnHihNq0aaOnnnrKjI8LwM24GgtAo1J3pVRBQYGio6PNLgdAI8CcHQAA4NMIOwAAwKfRxgIAAD6NkR0AAODTCDsAAMCnEXYAAIBPI+wAAACfRtgBAAA+jbADAAB8GmEHAAD4NMIOAADwaYQdAADg0/4/Rl1vtpPGry8AAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "vcpi_util.show_history(history_IV)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "98.55999946594238\n"
     ]
    }
   ],
   "source": [
    "eval_acc.append(evaluate(model_IV, test_loader)*100)\n",
    "print(eval_acc[-1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Show Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "vcpi_util.show_histories([history, history_II, history_III, history_IV], ['model 1', 'model 2', 'model 3', 'model4'])       "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAigAAAGiCAYAAADNzj2mAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAAkw0lEQVR4nO3deXRU5eH/8c9kYbKQDI1hSUISA0jAqmCBYiIq1HwxgLJIEQIa2bQuWMMSv8SqyA8w7v2W1ppqFRC0uKG4VAIVxcJBFhUERRZFSEIALTiTBJiQzP39oc5pZJEkM5lnwvt1Ts5hZu7cee48CXnnzp07NsuyLAEAABgkJNADAAAA+CkCBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABin3oHywQcf6JprrlFiYqJsNptef/31OrdblqX77rtPCQkJioyMVFZWlnbu3FlnmUOHDmnMmDGKjY1Vq1atNGHCBFVWVjZqQwAAQPNR70CpqqpSt27d9MQTT5z09ocfflhz585VUVGR1q1bp+joaF111VU6duyYd5kxY8bos88+04oVK/TWW2/pgw8+0M0339zwrQAAAM2KrTEfFmiz2fTaa69p6NChkr7fe5KYmKipU6dq2rRpkiSn06m2bdtq/vz5GjVqlLZt26bzzz9fGzZsUM+ePSVJy5Yt08CBA1VaWqrExMTGbxUAAAhqYb5c2e7du7V//35lZWV5r3M4HOrdu7fWrl2rUaNGae3atWrVqpU3TiQpKytLISEhWrdunYYNG3bCet1ut9xut/eyx+PRoUOHdM4558hms/lyEwAAgJ9YlqWKigolJiYqJOT0L+L4NFD2798vSWrbtm2d69u2beu9bf/+/WrTpk3dQYSFKS4uzrvMTxUWFmrmzJm+HCoAAAiQkpIStW/f/rTL+DRQ/KWgoEBTpkzxXnY6nUpJSVFJSYliY2MDODKgCTkcgR5BwzmdgR4B0PSC+WdW8svPrcvlUnJysmJiYn52WZ8GSrt27SRJBw4cUEJCgvf6AwcOqHv37t5lDh48WOd+NTU1OnTokPf+P2W322W320+4PjY2lkD5qWB/yavhh0TBZPycAsHHjz+3Z3J4hk/Pg5KWlqZ27drp3Xff9V7ncrm0bt06ZWRkSJIyMjL03Xff6aOPPvIus3LlSnk8HvXu3duXwwEAAEGq3ntQKisrtWvXLu/l3bt3a9OmTYqLi1NKSory8vI0e/ZsnXfeeUpLS9O9996rxMRE7zt9unbtquzsbN10000qKirS8ePHNWnSJI0aNYp38AAAAEkNCJSNGzeqX79+3ss/Hhty4403av78+brrrrtUVVWlm2++Wd9995369OmjZcuWKSIiwnuf559/XpMmTdKVV16pkJAQDR8+XHPnzvXB5gAAgOagUedBCRSXyyWHwyGn08kxKD/FMSjNVzDPLfOKs1Ew/8xKfvm5rc/vbz6LBwAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGCYrP4gGAZou3ogInxR4UAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYBwCBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYBwCBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYBwCBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYBwCBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYBwCBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADG8UugVFRUKC8vT6mpqYqMjFRmZqY2bNjgvf3AgQMaO3asEhMTFRUVpezsbO3cudMfQwEAAEHIL4EyceJErVixQgsXLtSWLVvUv39/ZWVlqaysTJZlaejQofrqq6+0dOlSffLJJ0pNTVVWVpaqqqr8MRwAABBkbJZlWb5c4dGjRxUTE6OlS5dq0KBB3ut79OihAQMGKDc3V+np6dq6dat++ctfSpI8Ho/atWunBx54QBMnTvzZx3C5XHI4HHI6nYqNjfXl8IOfzRboETSOb78dm5dgnlvm9dSCeV4l5vZ0mNsT1Of3t8/3oNTU1Ki2tlYRERF1ro+MjNTq1avldrslqc7tISEhstvtWr169UnX6Xa75XK56nwBAIDmy+eBEhMTo4yMDM2aNUv79u1TbW2tFi1apLVr16q8vFxdunRRSkqKCgoKdPjwYVVXV+uhhx5SaWmpysvLT7rOwsJCORwO71dycrKvhw0AAAzil2NQFi5cKMuylJSUJLvdrrlz5yonJ0chISEKDw/XkiVLtGPHDsXFxSkqKkrvvfeeBgwYoJCQkw+noKBATqfT+1VSUuKPYQMAAEOE+WOlHTt21KpVq1RVVSWXy6WEhASNHDlSHTp0kPT98SibNm2S0+lUdXW1Wrdurd69e6tnz54nXZ/dbpfdbvfHUAEAgIH8eh6U6OhoJSQk6PDhwyouLtaQIUPq3O5wONS6dWvt3LlTGzduPOF2AABwdvLLHpTi4mJZlqX09HTt2rVL+fn56tKli8aNGydJevnll9W6dWulpKRoy5YtuvPOOzV06FD179/fH8MBAABBxi+B4nQ6VVBQoNLSUsXFxWn48OGaM2eOwsPDJUnl5eWaMmWKDhw4oISEBOXm5uree+/1x1AAAEAQ8vl5UJoC50E5Dd5333wF89wyr6cWzPMqMbenw9yeIKDnQQEAAGgsAgUAABiHQAEAAMYhUAAAgHEIFAAAYBwCBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYBwCBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYJywQA8AAIBTsdkCPYKGswI9gCDHHhQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYBwCBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYBwCBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYBwCBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYBy/BEpFRYXy8vKUmpqqyMhIZWZmasOGDd7bKysrNWnSJLVv316RkZE6//zzVVRU5I+hAACAIBTmj5VOnDhRW7du1cKFC5WYmKhFixYpKytLn3/+uZKSkjRlyhStXLlSixYt0rnnnqvly5frtttuU2JiogYPHuyPIQEAgCBisyzL8uUKjx49qpiYGC1dulSDBg3yXt+jRw8NGDBAs2fP1gUXXKCRI0fq3nvvPentP+V2u+V2u72XXS6XkpOT5XQ6FRsb68vhBz+bLdAjaBzffjs2L8E8t8zrqQXzvEp+n9tgfnosBfHgJb/MrcvlksPhOKPf3z5/iaempka1tbWKiIioc31kZKRWr14tScrMzNQbb7yhsrIyWZal9957Tzt27FD//v1Pus7CwkI5HA7vV3Jysq+HDQAADOLzPSjS9wHSokULvfDCC2rbtq3+8Y9/6MYbb1SnTp20fft2ud1u3XzzzXruuecUFhamkJAQPf3008rNzT3p+tiDUg/B/OeGxF/apxPMc8u8nlowz6vEHpTTYA/KieqzB8Uvx6AsXLhQ48ePV1JSkkJDQ/WrX/1KOTk5+uijjyRJf/7zn/Xhhx/qjTfeUGpqqj744APdfvvtSkxMVFZW1gnrs9vtstvt/hgqAAAwkF/2oPyoqqpKLpdLCQkJGjlypCorK/XKK6/I4XDotddeq3OMysSJE1VaWqply5b97HrrU2BnnWD+c0Py619jQf/UBPNfY+xBObWg/8ZkD8qpBPXPrBTwPSh+PQ9KdHS0EhISdPjwYRUXF2vIkCE6fvy4jh8/rpCQug8dGhoqj8fjz+EAAIAg4ZeXeIqLi2VZltLT07Vr1y7l5+erS5cuGjdunMLDw3XFFVcoPz9fkZGRSk1N1apVq/Tcc8/p8ccf98dwAABAkPFLoDidThUUFKi0tFRxcXEaPny45syZo/DwcEnS4sWLVVBQoDFjxujQoUNKTU3VnDlzdMstt/hjOAAAIMj49RgUf+EYlNMI5hdsJY5BOY2gfj07+P6baTpB/43JMSinEtQ/s1LzPgYFAACgIQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcv5wHBQCaUnC/FRXAybAHBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYBwCBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYBwCBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYJyzQAzCRzRboETScFegBAADgA+xBAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYBwCBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYBwCBQAAGIdAAQAAxiFQAACAcQgUAABgHAIFAAAYh0ABAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMbxS6BUVFQoLy9PqampioyMVGZmpjZs2OC93WaznfTrkUce8cdwAABAkPFLoEycOFErVqzQwoULtWXLFvXv319ZWVkqKyuTJJWXl9f5evbZZ2Wz2TR8+HB/DAcAAAQZm2VZli9XePToUcXExGjp0qUaNGiQ9/oePXpowIABmj179gn3GTp0qCoqKvTuu++e0WO4XC45HA45nU7Fxsb6bOw/stl8vsomYymIBy9Jvv12rCOY51UK8rn147xKwT23QT2vEnN7Gsztierz+zvM1w9eU1Oj2tpaRURE1Lk+MjJSq1evPmH5AwcO6O2339aCBQtOuU632y232+297HK5fDdgAABgHJ+/xBMTE6OMjAzNmjVL+/btU21trRYtWqS1a9eqvLz8hOUXLFigmJgYXXvttadcZ2FhoRwOh/crOTnZ18MGAAAG8csxKAsXLpRlWUpKSpLdbtfcuXOVk5OjkJATH+7ZZ5/VmDFjTtjj8t8KCgrkdDq9XyUlJf4YNgAAMITPX+KRpI4dO2rVqlWqqqqSy+VSQkKCRo4cqQ4dOtRZ7t///re2b9+uF1988bTrs9vtstvt/hgqAAAwkF/PgxIdHa2EhAQdPnxYxcXFGjJkSJ3bn3nmGfXo0UPdunXz5zAAAECQ8cselOLiYlmWpfT0dO3atUv5+fnq0qWLxo0b513G5XLp5Zdf1mOPPeaPIQAAgCDmlz0oTqdTt99+u7p06aLc3Fz16dNHxcXFCg8P9y6zePFiWZalnJwcfwwBAAAEMZ+fB6UpcB6UU+N996cWzPMqBfnccq6MUwrqeZWY29Ngbk9Un9/ffBYPAAAwDoECAACMQ6AAAADjECgAAMA4BAoAADAOgQIAAIxDoAAAAOMQKAAAwDgECgAAMA6BAgAAjEOgAAAA4xAoAADAOAQKAAAwDoECAACMQ6AAAADjECgAAMA4BAoAADAOgQIAAIxDoAAAAOMQKAAAwDgECgAAMA6BAgAAjEOgAAAA4xAoAADAOAQKAAAwDoECAACMQ6AAAADjECgAAMA4BAoAADAOgQIAAIxDoAAAAOMQKAAAwDgECgAAMA6BAgAAjEOgAAAA4xAoAADAOAQKAAAwDoECAACMQ6AAAADjECgAAMA4BAoAADAOgQIAAIxDoAAAAOMQKAAAwDgECgAAMA6BAgAAjEOgAAAA4xAoAADAOAQKAAAwDoECAACMQ6AAAADjECgAAMA4BAoAADAOgQIAAIxDoAAAAOMQKAAAwDgECgAAMA6BAgAAjEOgAAAA4xAoAADAOAQKAAAwDoECAACMQ6AAAADjECgAAMA4YYEeAAAAJgkL8yghoVohjfwT/phSfTOgQDl2rN53CQ8PV2hoqE8enkABAOAHbdpU69FHdys+3iObrXHr2q0i3wwqUHbvbtDdWrVqpXbt2snWyCeQQAEAQJLNZumWW8qVlhaqiIhkNfYoiDRV+WZggZKWVq/FLcvSkSNHdPDgQUlSQkJCox6eQAEAQFKrVjXq2fOIIiISJUU1en0RjR9SYEXUfwsiIyMlSQcPHlSbNm0a9XIPB8kCACApJqZWYWGS1CLQQwlqUVHfx93x48cbtR4CBQAASTabfjjupJEHn5zlGnvsyY8IFAAAYBy/BEpFRYXy8vKUmpqqyMhIZWZmasOGDXWW2bZtmwYPHiyHw6Ho6Gj16tVLe/fu9cdwAABAkPHLQbITJ07U1q1btXDhQiUmJmrRokXKysrS559/rqSkJH355Zfq06ePJkyYoJkzZyo2NlafffaZIhpwQA4AAP7Uq1dD79mzQfeyNmxs6AP6zfz585WXl6fvvvuuyR7T54Fy9OhRvfrqq1q6dKkuv/xySdL999+vN998U08++aRmz56tP/zhDxo4cKAefvhh7/06dux4ynW63W653W7vZZfL5ethAwAAg/j8JZ6amhrV1taesDckMjJSq1evlsfj0dtvv63OnTvrqquuUps2bdS7d2+9/vrrp1xnYWGhHA6H9ys5OdnXwwYAAAbxeaDExMQoIyNDs2bN0r59+1RbW6tFixZp7dq1Ki8v18GDB1VZWakHH3xQ2dnZWr58uYYNG6Zrr71Wq1atOuk6CwoK5HQ6vV8lJSW+HjYAAEHL4/GocN48pQ0Zosg+fdRt9Gi98u678ng8aj9okJ585ZU6y3+yfbtCfv1r7SkvlyQ9/vzzunDUKEVfdpmSBw3SbQ8+qMrKykBsipdfDpJduHChLMtSUlKS7Ha75s6dq5ycHIWEhMjj8UiShgwZosmTJ6t79+6aPn26rr76ahUVnfy0wHa7XbGxsXW+AADA9wrnz9dz//yniqZP12eLF2tyTo6uv+8+/fuTT5TTv79eKC6us/zz77yjS7t1U+oPZ3sNsdk0d9o0ffbii1pw//1auXGj7rrrrkBsipdfAqVjx45atWqVKisrVVJSovXr1+v48ePq0KGD4uPjFRYWpvPPP7/Ofbp27cq7eAAAqCd3dbUemDdPz957r67KyFCH9u019pprdP2AAfrba69pTHa21mzerL3790v6fm/L4hUrNCY727uOvNGj1a9nT52bmKjf9Oql2bfeqpdeeilQmyTJz6e6j46OVnR0tA4fPqzi4mI9/PDDatGihXr16qXt27fXWXbHjh1KTQ3yT34EAKCJ7Sop0ZFjx/Q/kybVub76+HFdnJ6u7unp6nruuXph2TJNHztWqz7+WAcPHdKIK6/0LvuvdetUOH++vtizR66qKtXU1uqY260jR454zwzb1PwSKMXFxbIsS+np6dq1a5fy8/PVpUsXjRs3TpKUn5+vkSNH6vLLL1e/fv20bNkyvfnmm3r//ff9MRwAAJqtyqNHJUlv//GPSmrTps5t9vBwSdKYAQP0QnGxpo8dqxeWLVN2RobOadVKkvT1vn26esoU3Tp8uObcdpviYmO1evNmTZg1S9XV1c0rUJxOpwoKClRaWqq4uDgNHz5cc+bMUfgPT9SwYcNUVFSkwsJC/f73v1d6erpeffVV9enTxx/DAQCg2To/LU32Fi2098ABXdGjx0mXGX3VVbrnySf10bZtemXlShVNn+697aMvvpDH49FjeXkKCfn+yI+X/vWvJhn76fglUK677jpdd911p11m/PjxGj9+vD8eHgCAs0ZMdLSmXX+9Jj/+uDwej/p07y5nZaXWbN6s2Oho3Xj11To3MVGZF12kCbNmqdbj0eAfzlMmSZ3at9fxmhr9+cUXdc1ll2nNp5+qaMmSAG7R9/x6DAoAAMHuJ5/UcsZ6qunOCDvrllvUulUrFc6fr6/KytQqJka/Sk/X3T8cWiFJY7KzddtDDyl30CBF/te5yrp17qzHJ0/WQ889p4InntDlF1+swttvV+6MGU02/pOxWZZlBXQEDeByueRwOOR0Ov3ylmMffRBjQFjB/imcfvx2DOZ5lYJ8bv3830wwz21Qz6vUrOY2NfWYiop2Kz4+TVLjP3qlKQPFL3o27FT9x44d0+7du5WWlnbCSVvr8/ubTzMGAADGIVAAAIBxCBQAAGAcAgUAABiHQAEAAMYhUAAAgHEIFAAAYBwCBQAAGIdAAQAAxuFU9wAAnEbPXk18tt+GnlvfB84dPFh5o0Ypb/TogI3hRwQKAABBrO/vfqfunTvr/6ZObfS6NixYoOjISB+MqvEIFAAAmjHLslRbW6uwsJ//ld/6F79oghGdGY5BAQAgSI29/36t+vhj/WnxYtl69ZKtVy/Nf/NN2Xr10jtr1qjHDTfInpmp1Zs368vSUg2ZOlVtr7pKLS+/XL1yc/WvdevqrO/cwYP1fy+84L1ss9n097//XcOGDVNUVJTOO+88vfHGG02ybQQKAABB6k/Tpinjwgt109ChKn/nHZW/846S27aVJE1/4gk9OGmStr38si7q1EmVR45o4KWX6t0nntAnixYpOyND10ydqr3795/2MWbOnKnrrrtOn376qQYOHKgxY8bo0KFDft82AgUAgCDlaNlSLcLDFRURoXbx8WoXH6/Q0FBJ0v/73e/0P717q2P79opzONStc2f97tprdUGnTjovJUWzbr1VHZOS9MYHH5z2McaOHaucnBx16tRJDzzwgCorK7V+/Xq/bxvHoAAA0Az17Nq1zuXKI0d0/1NP6e01a1T+7beqqa3VUbf7Z/egXHTRRd5/R0dHKzY2VgcPHvTLmP8bgQIAQDP003fjTPvTn7Ri3To9eued6pScrEi7Xb/93/9V9fHjp11PeHh4ncs2m00ej8fn4/0pAgUAgCDWIjxctWcQDGs2b9bYq6/WsH79JH2/R+Xr8nJ/D6/BCBQAAILYuQkJWrd1q77et08to6JOuXfjvORkLXnvPV1z2WWy2Wy6t6hIHstq4tGeOQIFAIDT2LihYb/Ee2qjj0dyctOuv143zpyp86+7Tkfdbs27776TLvf45MkaP2uWMidMUHyrVvrf3Fy5qqqaZIwNYbMsg/PpFFwulxwOh5xOp2JjY32+flsTn9XYlywF8eAlyY/fjsE8r1KQz62f/5sJ5rkN6nmVmtXcpqYeU1HRbsXHp0mKaPT6mipQ/KZnzwbd7dixY9q9e7fS0tIUEVH3eazP72/eZgwAAIxDoAAAAOMQKAAAwDgECgAAMA6BAgCAvj/e9/tjfoPuvSNG8dVJ3HibMQAAkr75JlxOp01xcd8oNLS11Mh3WB3zzbAC51j9tsCyLFVXV+ubb75RSEiIWrRo0aiHJ1AAAJB09Gio5sxprz/8oVQOx9eNfovzbn3rm4EFyu7dDbpbVFSUUlJSFBLSuBdpCBQAAH6wdWtLjR9/nlq3Pt7oQPlCA3wzqED54ot63yU0NFRhYWGy+eAENgQKAAD/5ejRUO3dG9ro9URojw9GE0ARjT9ZXWNwkCwAADAOgQIAAIxDoAAAAOME5TEoP36+ocvlCvBIzBP0zwhzekpB/cwwr6cU9M8Mc3tKQf/M+GFuf/y9fSafUxyUn2ZcWlqq5OTkQA8DAAA0QElJidq3b3/aZYIyUDwej/bt26eYmBifvJWpuXC5XEpOTlZJScnPfow1ggtz2zwxr80Xc3tylmWpoqJCiYmJP3uelKB8iSckJORny+tsFhsbyw9EM8XcNk/Ma/PF3J7I4XCc0XIcJAsAAIxDoAAAAOMQKM2I3W7XjBkzZLfbAz0U+Bhz2zwxr80Xc9t4QXmQLAAAaN7YgwIAAIxDoAAAAOMQKAAAwDgECgAAMA6B0sz07dtXeXl5Z7z8/Pnz1apVK7+NB77D3DZPzGvzxvw2HIGCnzVnzhxlZmYqKiqKH5xm5Ouvv9aECROUlpamyMhIdezYUTNmzFB1dXWgh4ZGGjx4sFJSUhQREaGEhATdcMMN2rdvX6CHhUZas2aNwsLC1L1790APpUkQKPhZ1dXVGjFihG699dZADwU+9MUXX8jj8ehvf/ubPvvsM/3xj39UUVGR7r777kAPDY3Ur18/vfTSS9q+fbteffVVffnll/rtb38b6GGhEb777jvl5ubqyiuvDPRQmgyB0gT69u2rO+64Q3l5efrFL36htm3b6umnn1ZVVZXGjRunmJgYderUSe+8806d+61atUq//vWvZbfblZCQoOnTp6umpsZ7e1VVlXJzc9WyZUslJCToscceO+Gx3W63pk2bpqSkJEVHR6t37956//336zX+mTNnavLkybrwwgsbtP3NWTDPbXZ2tubNm6f+/furQ4cOGjx4sKZNm6YlS5Y0+PloLoJ5XiVp8uTJuuSSS5SamqrMzExNnz5dH374oY4fP96g56M5asgcB3J+b7nlFo0ePVoZGRk+2f5gQKA0kQULFig+Pl7r16/XHXfcoVtvvVUjRoxQZmamPv74Y/Xv31833HCDjhw5IkkqKyvTwIED1atXL23evFlPPvmknnnmGc2ePdu7zvz8fK1atUpLly7V8uXL9f777+vjjz+u87iTJk3S2rVrtXjxYn366acaMWKEsrOztXPnzibd/uasOc2t0+lUXFxcg+/fnDSXeT106JCef/55ZWZmKjw8vOFPSDNUnzkO5PzOmzdPX331lWbMmOG358JIFvzuiiuusPr06eO9XFNTY0VHR1s33HCD97ry8nJLkrV27VrLsizr7rvvttLT0y2Px+Nd5oknnrBatmxp1dbWWhUVFVaLFi2sl156yXv7f/7zHysyMtK68847LcuyrD179lihoaFWWVlZnfFceeWVVkFBgWVZljVv3jzL4XCc0XbUZ9mzRXOZW8uyrJ07d1qxsbHWU089dcb3aa6aw7zeddddVlRUlCXJuuSSS6xvv/223s9Dc1bfOQ7U/O7YscNq06aNtX37dsuyLGvGjBlWt27dfPU0GC0soHV0Frnooou8/w4NDdU555xT5yWTtm3bSpIOHjwoSdq2bZsyMjJks9m8y1x66aWqrKxUaWmpDh8+rOrqavXu3dt7e1xcnNLT072Xt2zZotraWnXu3LnOWNxut8455xzfbuBZrDnMbVlZmbKzszVixAjddNNN9b5/cxTs85qfn68JEyZoz549mjlzpnJzc/XWW2/VGd/Zrj5zHIj5ra2t1ejRozVz5swT7nM2IFCayE93rdpstjrX/fhN7/F4fPaYlZWVCg0N1UcffaTQ0NA6t7Vs2dJnj3O2C/a53bdvn/r166fMzEw99dRTPhtjsAv2eY2Pj1d8fLw6d+6srl27Kjk5WR9++OFZdQzDz2nqOa7v/FZUVGjjxo365JNPNGnSJO9YLMtSWFiYli9frt/85jc+GZuJCBRDde3aVa+++qosy/L+kKxZs0YxMTFq37694uLiFB4ernXr1iklJUWSdPjwYe3YsUNXXHGFJOniiy9WbW2tDh48qMsuuyxg24K6TJrbsrIy9evXTz169NC8efMUEsJhaQ1l0rz+1I+/YN1ut8/WebYJxPzGxsZqy5Ytda7761//qpUrV+qVV15RWlqaj7fSLPxvZKjbbrtNJSUluuOOO/TFF19o6dKlmjFjhqZMmaKQkBC1bNlSEyZMUH5+vlauXKmtW7dq7NixdX7BdO7cWWPGjFFubq6WLFmi3bt3a/369SosLNTbb799xmPZu3evNm3apL1796q2tlabNm3Spk2bVFlZ6Y9Nb/ZMmduysjL17dtXKSkpevTRR/XNN99o//792r9/v782vVkzZV7XrVunv/zlL9q0aZP27NmjlStXKicnRx07dmTvSSMEYn5DQkJ0wQUX1Plq06aNIiIidMEFFyg6Oropn4Imxx4UQyUlJemf//yn8vPz1a1bN8XFxWnChAm65557vMs88sgjqqys1DXXXKOYmBhNnTpVTqezznrmzZun2bNna+rUqSorK1N8fLwuueQSXX311Wc8lvvuu08LFizwXr744oslSe+995769u3buA09C5kytytWrNCuXbu0a9cutW/fvs5tlmU1fkPPMqbMa1RUlJYsWaIZM2aoqqpKCQkJys7O1j333CO73e7TbT6bmDK/ZxObxf9EAADAMLzEAwAAjEOgAAAA4xAoAADAOAQKAAAwDoECAACMQ6AAAADjECgAAMA4BAoAADAOgQIAAIxDoAAAAOMQKAAAwDj/HyDcysP6Iu+QAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "vcpi_util.show_accuracies(train_acc, eval_acc, ['model 1', 'model 2', 'model 3', 'model4'])  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x1000 with 30 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "images, targets = next(iter(test_loader))\n",
    "logits = model_IV(images.to(device))\n",
    "\n",
    "predictions = torch.nn.functional.softmax(logits, dim=1).cpu().detach().numpy()\n",
    "\n",
    "vcpi_util.plot_predictions(images, predictions, targets, train_set.classes, 5, 3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x1000 with 30 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_bad_preds(model_IV, test_loader, test_set.classes)    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "build_confusion_matrix(model_IV, test_loader)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.3"
  },
  "vscode": {
   "interpreter": {
    "hash": "8ca9a0304c78ff1aed536949a52a998bc579376efcff4fb61d932d04b9b39864"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}