Image_processing_Assignment_3.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4",
      "authorship_tag": "ABX9TyN2AX3Q4Dfn9aMTvIs5LaBw",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/SasikaA073/image-processing-assignments-m/blob/main/Image_processing_Assignment_3.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# EN3160 Assignment 3 on Neural Networks\n",
        "\n",
        "- 210035A\n",
        "- Sasika Amarasinghe\n",
        "\n",
        "- [https://github.com/SasikaA073/image-processing-assignments-m](https://github.com/SasikaA073/image-processing-assignments-m)"
      ],
      "metadata": {
        "id": "YwBVkWD7nEv9"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "---\n",
        "## Q1\n",
        "---"
      ],
      "metadata": {
        "id": "aPqnz-CjikG4"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 796
        },
        "id": "qtS42BfMifNv",
        "outputId": "007e649d-a63f-49f4-9ba1-4654d6fa4f3c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Using device: cuda\n",
            "Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 170M/170M [00:04<00:00, 41.9MB/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting ./data/cifar-10-python.tar.gz to ./data\n",
            "Files already downloaded and verified\n",
            "Epoch 1/10, Loss: 1.7614432820081711\n",
            "Epoch 2/10, Loss: 1.6302554644346237\n",
            "Epoch 3/10, Loss: 1.5702226233482361\n",
            "Epoch 4/10, Loss: 1.5232444244623184\n",
            "Epoch 5/10, Loss: 1.480954245209694\n",
            "Epoch 6/10, Loss: 1.4469501773715019\n",
            "Epoch 7/10, Loss: 1.4172854322195052\n",
            "Epoch 8/10, Loss: 1.3824967207312584\n",
            "Epoch 9/10, Loss: 1.3553754616379738\n",
            "Epoch 10/10, Loss: 1.326730316758156\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x500 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training accuracy: 56.34%\n",
            "Test accuracy: 46.18%\n"
          ]
        }
      ],
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torchvision\n",
        "import torchvision.transforms as transforms\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Check if CUDA is available\n",
        "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
        "print(f\"Using device: {device}\")\n",
        "\n",
        "# 1. Data loading and transformation\n",
        "transform = transforms.Compose([\n",
        "    transforms.ToTensor(),\n",
        "    transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))\n",
        "])\n",
        "\n",
        "batch_size = 50\n",
        "trainset = torchvision.datasets.CIFAR10(root='./data', train=True, download=True, transform=transform)\n",
        "trainloader = torch.utils.data.DataLoader(trainset, batch_size=batch_size, shuffle=True, num_workers=2)\n",
        "\n",
        "testset = torchvision.datasets.CIFAR10(root='./data', train=False, download=True, transform=transform)\n",
        "testloader = torch.utils.data.DataLoader(testset, batch_size=batch_size, shuffle=False, num_workers=2)\n",
        "\n",
        "# Class names\n",
        "classes = ('plane', 'car', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck')\n",
        "\n",
        "# 2. Define the neural network as a proper PyTorch module\n",
        "class SimpleNet(nn.Module):\n",
        "    def __init__(self, input_size, hidden_size, num_classes):\n",
        "        super(SimpleNet, self).__init__()\n",
        "        self.flatten = nn.Flatten()\n",
        "        self.fc1 = nn.Linear(input_size, hidden_size)\n",
        "        self.sigmoid = nn.Sigmoid()\n",
        "        self.fc2 = nn.Linear(hidden_size, num_classes)\n",
        "\n",
        "    def forward(self, x):\n",
        "        x = self.flatten(x)\n",
        "        x = self.sigmoid(self.fc1(x))\n",
        "        x = self.fc2(x)\n",
        "        return x\n",
        "\n",
        "# Initialize network and move to device\n",
        "Din = 3 * 32 * 32  # Input size (flattened CIFAR-10 image size)\n",
        "H = 100            # Hidden layer size\n",
        "K = 10             # Output size (number of classes)\n",
        "model = SimpleNet(Din, H, K).to(device)\n",
        "\n",
        "# Define loss function and optimizer\n",
        "criterion = nn.CrossEntropyLoss()\n",
        "optimizer = torch.optim.Adam(model.parameters(), lr=0.001)\n",
        "\n",
        "# Training settings\n",
        "epochs = 10\n",
        "loss_history = []\n",
        "\n",
        "# 3. Training Loop\n",
        "for epoch in range(epochs):\n",
        "    running_loss = 0.0\n",
        "    for i, data in enumerate(trainloader, 0):\n",
        "        inputs, labels = data\n",
        "        inputs, labels = inputs.to(device), labels.to(device)\n",
        "\n",
        "        # Zero the gradients\n",
        "        optimizer.zero_grad()\n",
        "\n",
        "        # Forward pass\n",
        "        outputs = model(inputs)\n",
        "\n",
        "        # Calculate loss\n",
        "        loss = criterion(outputs, labels)\n",
        "        loss_history.append(loss.item())\n",
        "        running_loss += loss.item()\n",
        "\n",
        "        # Backward pass and optimize\n",
        "        loss.backward()\n",
        "        optimizer.step()\n",
        "\n",
        "    print(f\"Epoch {epoch + 1}/{epochs}, Loss: {running_loss / len(trainloader)}\")\n",
        "\n",
        "# 4. Plotting the Loss History\n",
        "plt.figure(figsize=(10, 5))\n",
        "plt.plot(loss_history)\n",
        "plt.title(\"Loss History\")\n",
        "plt.xlabel(\"Iteration\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.show()\n",
        "\n",
        "# 5. Calculate Accuracy on Training Set\n",
        "model.eval()  # Set the model to evaluation mode\n",
        "correct_train = 0\n",
        "total_train = 0\n",
        "with torch.no_grad():\n",
        "    for data in trainloader:\n",
        "        inputs, labels = data\n",
        "        inputs, labels = inputs.to(device), labels.to(device)\n",
        "        outputs = model(inputs)\n",
        "        _, predicted = torch.max(outputs.data, 1)\n",
        "        total_train += labels.size(0)\n",
        "        correct_train += (predicted == labels).sum().item()\n",
        "\n",
        "train_acc = 100 * correct_train / total_train\n",
        "print(f\"Training accuracy: {train_acc:.2f}%\")\n",
        "\n",
        "# 6. Calculate Accuracy on Test Set\n",
        "correct_test = 0\n",
        "total_test = 0\n",
        "with torch.no_grad():\n",
        "    for data in testloader:\n",
        "        inputs, labels = data\n",
        "        inputs, labels = inputs.to(device), labels.to(device)\n",
        "        outputs = model(inputs)\n",
        "        _, predicted = torch.max(outputs.data, 1)\n",
        "        total_test += labels.size(0)\n",
        "        correct_test += (predicted == labels).sum().item()\n",
        "\n",
        "test_acc = 100 * correct_test / total_test\n",
        "print(f\"Test accuracy: {test_acc:.2f}%\")"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The above graph, corresponds to the loss in each batch iteration."
      ],
      "metadata": {
        "id": "VNQN-RDGqDTJ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "---\n",
        "## Q2\n",
        "---"
      ],
      "metadata": {
        "id": "oWtNuz0kjHOs"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.optim as optim\n",
        "import torchvision\n",
        "import torchvision.transforms as transforms\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Check if CUDA is available\n",
        "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
        "print(f\"Using device: {device}\")\n",
        "\n",
        "# 1. Data loading and transformation for MNIST\n",
        "transform = transforms.Compose([\n",
        "    transforms.Resize((32, 32)),  # LeNet-5 uses 32x32 input\n",
        "    transforms.ToTensor(),\n",
        "    transforms.Normalize((0.5,), (0.5,))\n",
        "])\n",
        "\n",
        "batch_size = 64\n",
        "trainset = torchvision.datasets.MNIST(root='./data', train=True, download=True, transform=transform)\n",
        "trainloader = torch.utils.data.DataLoader(trainset, batch_size=batch_size, shuffle=True, num_workers=2)\n",
        "\n",
        "testset = torchvision.datasets.MNIST(root='./data', train=False, download=True, transform=transform)\n",
        "testloader = torch.utils.data.DataLoader(testset, batch_size=batch_size, shuffle=False, num_workers=2)\n",
        "\n",
        "# 2. Define LeNet-5 Model\n",
        "class LeNet5(nn.Module):\n",
        "    def __init__(self):\n",
        "        super(LeNet5, self).__init__()\n",
        "        # First convolutional layer: 1 input channel, 6 output channels, 5x5 kernel\n",
        "        self.conv1 = nn.Conv2d(1, 6, kernel_size=5)  # Remove padding=2\n",
        "        # Second convolutional layer: 6 input channels, 16 output channels, 5x5 kernel\n",
        "        self.conv2 = nn.Conv2d(6, 16, kernel_size=5)\n",
        "        # Fully connected layers\n",
        "        self.fc1 = nn.Linear(16 * 5 * 5, 120)  # The 5x5 is the size of the feature maps after conv2\n",
        "        self.fc2 = nn.Linear(120, 84)\n",
        "        self.fc3 = nn.Linear(84, 10)\n",
        "\n",
        "    def forward(self, x):\n",
        "        # First conv + activation + pooling\n",
        "        x = torch.tanh(self.conv1(x))        # Output: 6 x 28 x 28\n",
        "        x = torch.max_pool2d(x, 2)           # Output: 6 x 14 x 14\n",
        "\n",
        "        # Second conv + activation + pooling\n",
        "        x = torch.tanh(self.conv2(x))        # Output: 16 x 10 x 10\n",
        "        x = torch.max_pool2d(x, 2)           # Output: 16 x 5 x 5\n",
        "\n",
        "        # Flatten the feature maps\n",
        "        x = x.view(-1, 16 * 5 * 5)           # Output: batch_size x 400\n",
        "\n",
        "        # Fully connected layers\n",
        "        x = torch.tanh(self.fc1(x))\n",
        "        x = torch.tanh(self.fc2(x))\n",
        "        x = self.fc3(x)\n",
        "        return x\n",
        "\n",
        "# Move the model to the selected device\n",
        "model = LeNet5().to(device)\n",
        "\n",
        "# Define the loss function and optimizer\n",
        "criterion = nn.CrossEntropyLoss()\n",
        "optimizer = optim.Adam(model.parameters(), lr=0.001)\n",
        "\n",
        "# 3. Training the LeNet-5 Model\n",
        "epochs = 10\n",
        "train_losses = []\n",
        "\n",
        "for epoch in range(epochs):\n",
        "    running_loss = 0.0\n",
        "    for i, data in enumerate(trainloader, 0):\n",
        "        inputs, labels = data\n",
        "        inputs, labels = inputs.to(device), labels.to(device)\n",
        "\n",
        "        # Zero the parameter gradients\n",
        "        optimizer.zero_grad()\n",
        "\n",
        "        # Forward pass\n",
        "        outputs = model(inputs)\n",
        "        loss = criterion(outputs, labels)\n",
        "\n",
        "        # Backward pass and optimize\n",
        "        loss.backward()\n",
        "        optimizer.step()\n",
        "\n",
        "        running_loss += loss.item()\n",
        "\n",
        "    # Log the loss for this epoch\n",
        "    epoch_loss = running_loss / len(trainloader)\n",
        "    train_losses.append(epoch_loss)\n",
        "    print(f\"Epoch {epoch + 1}/{epochs}, Loss: {epoch_loss:.4f}\")\n",
        "\n",
        "# 4. Plotting the Training Loss History\n",
        "plt.figure(figsize=(10, 5))\n",
        "plt.plot(train_losses, label='Training Loss')\n",
        "plt.title(\"Training Loss History\")\n",
        "plt.xlabel(\"Epoch\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.legend()\n",
        "plt.show()\n",
        "\n",
        "# 5. Calculate Training Accuracy\n",
        "model.eval()  # Set model to evaluation mode\n",
        "correct_train = 0\n",
        "total_train = 0\n",
        "with torch.no_grad():\n",
        "    for data in trainloader:\n",
        "        images, labels = data\n",
        "        images, labels = images.to(device), labels.to(device)\n",
        "        outputs = model(images)\n",
        "        _, predicted = torch.max(outputs.data, 1)\n",
        "        total_train += labels.size(0)\n",
        "        correct_train += (predicted == labels).sum().item()\n",
        "\n",
        "train_accuracy = 100 * correct_train / total_train\n",
        "print(f\"Training accuracy: {train_accuracy:.2f}%\")\n",
        "\n",
        "# 6. Calculate Test Accuracy\n",
        "correct_test = 0\n",
        "total_test = 0\n",
        "with torch.no_grad():\n",
        "    for data in testloader:\n",
        "        images, labels = data\n",
        "        images, labels = images.to(device), labels.to(device)\n",
        "        outputs = model(images)\n",
        "        _, predicted = torch.max(outputs.data, 1)\n",
        "        total_test += labels.size(0)\n",
        "        correct_test += (predicted == labels).sum().item()\n",
        "\n",
        "test_accuracy = 100 * correct_test / total_test\n",
        "print(f\"Test accuracy: {test_accuracy:.2f}%\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "74r4i2WljGaa",
        "outputId": "8e289d93-3610-4bf9-b009-7a9bd023cbc3"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Using device: cuda\n",
            "Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz\n",
            "Failed to download (trying next):\n",
            "HTTP Error 403: Forbidden\n",
            "\n",
            "Downloading https://ossci-datasets.s3.amazonaws.com/mnist/train-images-idx3-ubyte.gz\n",
            "Downloading https://ossci-datasets.s3.amazonaws.com/mnist/train-images-idx3-ubyte.gz to ./data/MNIST/raw/train-images-idx3-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 9.91M/9.91M [00:00<00:00, 18.1MB/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting ./data/MNIST/raw/train-images-idx3-ubyte.gz to ./data/MNIST/raw\n",
            "\n",
            "Downloading http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz\n",
            "Failed to download (trying next):\n",
            "HTTP Error 403: Forbidden\n",
            "\n",
            "Downloading https://ossci-datasets.s3.amazonaws.com/mnist/train-labels-idx1-ubyte.gz\n",
            "Downloading https://ossci-datasets.s3.amazonaws.com/mnist/train-labels-idx1-ubyte.gz to ./data/MNIST/raw/train-labels-idx1-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 28.9k/28.9k [00:00<00:00, 496kB/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting ./data/MNIST/raw/train-labels-idx1-ubyte.gz to ./data/MNIST/raw\n",
            "\n",
            "Downloading http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz\n",
            "Failed to download (trying next):\n",
            "HTTP Error 403: Forbidden\n",
            "\n",
            "Downloading https://ossci-datasets.s3.amazonaws.com/mnist/t10k-images-idx3-ubyte.gz\n",
            "Downloading https://ossci-datasets.s3.amazonaws.com/mnist/t10k-images-idx3-ubyte.gz to ./data/MNIST/raw/t10k-images-idx3-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 1.65M/1.65M [00:00<00:00, 4.50MB/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting ./data/MNIST/raw/t10k-images-idx3-ubyte.gz to ./data/MNIST/raw\n",
            "\n",
            "Downloading http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz\n",
            "Failed to download (trying next):\n",
            "HTTP Error 403: Forbidden\n",
            "\n",
            "Downloading https://ossci-datasets.s3.amazonaws.com/mnist/t10k-labels-idx1-ubyte.gz\n",
            "Downloading https://ossci-datasets.s3.amazonaws.com/mnist/t10k-labels-idx1-ubyte.gz to ./data/MNIST/raw/t10k-labels-idx1-ubyte.gz\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 4.54k/4.54k [00:00<00:00, 4.23MB/s]"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Extracting ./data/MNIST/raw/t10k-labels-idx1-ubyte.gz to ./data/MNIST/raw\n",
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10, Loss: 0.2155\n",
            "Epoch 2/10, Loss: 0.0614\n",
            "Epoch 3/10, Loss: 0.0432\n",
            "Epoch 4/10, Loss: 0.0346\n",
            "Epoch 5/10, Loss: 0.0257\n",
            "Epoch 6/10, Loss: 0.0216\n",
            "Epoch 7/10, Loss: 0.0177\n",
            "Epoch 8/10, Loss: 0.0150\n",
            "Epoch 9/10, Loss: 0.0127\n",
            "Epoch 10/10, Loss: 0.0120\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x500 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA18AAAHWCAYAAACIZjNQAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAABn3klEQVR4nO3deXxU5d3///fMJDPZ9x2CYZN9UZaICyCkAlorigqIN4tWWwUUUr8VvBVQWoNKLbdiwVq3KhS0v0qVKi2ExS0KglhFiCyykw3IQkIyycz8/kgyMCZACElOMnk9H495JHPNdc58JqY1b6/rfI7J5XK5BAAAAABoVGajCwAAAACA1oDwBQAAAABNgPAFAAAAAE2A8AUAAAAATYDwBQAAAABNgPAFAAAAAE2A8AUAAAAATYDwBQAAAABNgPAFAAAAAE2A8AUAaFCTJ09WUlJSvY6dN2+eTCZTwxbUym3cuFEmk0kbN240uhQAaPUIXwDQSphMpjo9Wusf6ZMnT1ZQUJDRZVzQ0KFD1bNnz1pf279/v0wmkxYuXHjJ7/P0009r1apVl3weAMAZPkYXAABoGm+99ZbH87/+9a9au3ZtjfFu3bpd0vu88sorcjqd9Tr28ccf16xZsy7p/eFp8ODBOn36tKxW60Ud9/TTT+v222/X6NGjG6cwAGiFCF8A0ErcfffdHs+/+OILrV27tsb4T5WUlCggIKDO7+Pr61uv+iTJx8dHPj78q6khmc1m+fn5GV2GJKm4uFiBgYFGlwEAhmHbIQDArXpL29atWzV48GAFBATosccekyT985//1E033aSEhATZbDZ17NhR8+fPl8Ph8DjHT6/5Onsr3J///Gd17NhRNptNAwYM0JYtWzyOre2aL5PJpGnTpmnVqlXq2bOnbDabevTooTVr1tSof+PGjerfv7/8/PzUsWNHvfzyyw1+Hdm7776rfv36yd/fX1FRUbr77rt15MgRjzlZWVmaMmWK2rZtK5vNpvj4eN1yyy3av3+/e85XX32lESNGKCoqSv7+/mrfvr3uueeeBquzWm3XfO3evVtjxoxRXFyc/Pz81LZtW40bN04FBQWSKn/mxcXFevPNN93bUSdPnuw+/uuvv9aoUaMUEhKioKAgDR8+XF988YXH+77xxhsymUzatGmTHnzwQcXExKht27basGGDTCaT3nvvvRq1Ll++XCaTSRkZGQ3+cwCA5oD/vAgA8HD8+HGNGjVK48aN0913363Y2FhJlX9MBwUFKTU1VUFBQVq/fr3mzJmjwsJCPffccxc87/Lly1VUVKRf/epXMplMevbZZ3Xbbbdp3759F1wt+/TTT/WPf/xDDz74oIKDg/XCCy9ozJgxOnjwoCIjIyVVBoKRI0cqPj5eTz75pBwOh5566ilFR0df+g+lyhtvvKEpU6ZowIABSktLU3Z2tv7v//5Pn332mb7++muFhYVJksaMGaMdO3Zo+vTpSkpKUk5OjtauXauDBw+6n99www2Kjo7WrFmzFBYWpv379+sf//hHnepwOBzKy8urMX7y5MkLHmu32zVixAiVlZVp+vTpiouL05EjR7R69Wrl5+crNDRUb731ln75y19q4MCBuv/++yVJHTt2lCTt2LFD1113nUJCQvTb3/5Wvr6+evnllzV06FBt2rRJycnJHu/34IMPKjo6WnPmzFFxcbGGDh2qxMRELVu2TLfeeqvH3GXLlqljx44aNGhQnX4OANDiuAAArdLUqVNdP/3XwJAhQ1ySXEuXLq0xv6SkpMbYr371K1dAQICrtLTUPTZp0iTXZZdd5n7+448/uiS5IiMjXSdOnHCP//Of/3RJcn3wwQfusblz59aoSZLLarW69uzZ4x775ptvXJJcL774onvs5ptvdgUEBLiOHDniHtu9e7fLx8enxjlrM2nSJFdgYOA5X7fb7a6YmBhXz549XadPn3aPr1692iXJNWfOHJfL5XKdPHnSJcn13HPPnfNc7733nkuSa8uWLRes66eq/xmd73H2e2/YsMElybVhwwaXy+Vyff311y5Jrnffffe87xMYGOiaNGlSjfHRo0e7rFara+/eve6xo0ePuoKDg12DBw92j73++usuSa5rr73WVVFR4XGO2bNnu2w2mys/P989lpOT4/Lx8XHNnTv3In4aANCysO0QAODBZrNpypQpNcb9/f3d3xcVFSkvL0/XXXedSkpKtGvXrgued+zYsQoPD3c/v+666yRJ+/btu+CxKSkp7pUXSerdu7dCQkLcxzocDq1bt06jR49WQkKCe16nTp00atSoC56/Lr766ivl5OTowQcf9LiG6qabblLXrl31r3/9S1Llz8lqtWrjxo3nXImqXiFbvXq1ysvLL7qWpKQkrV27tsbj7bffvuCxoaGhkqR///vfKikpuaj3dTgc+s9//qPRo0erQ4cO7vH4+Hjddddd+vTTT1VYWOhxzH333SeLxeIxNnHiRJWVlenvf/+7e2zlypWqqKi44DWIANCSEb4AAB7atGlTa2e8HTt26NZbb1VoaKhCQkIUHR3t/kO5+lqh82nXrp3H8+ogVpetcj89tvr46mNzcnJ0+vRpderUqca82sbq48CBA5KkLl261Hita9eu7tdtNpueeeYZffTRR4qNjdXgwYP17LPPKisryz1/yJAhGjNmjJ588klFRUXplltu0euvv66ysrI61RIYGKiUlJQaj2uuueaCx7Zv316pqan6y1/+oqioKI0YMUIvvfRSnf4Z5ubmqqSkpNafQbdu3eR0OnXo0KEa7/dTXbt21YABA7Rs2TL32LJly3TVVVc12D8vAGiOCF8AAA9nr3BVy8/P15AhQ/TNN9/oqaee0gcffKC1a9fqmWeekaQ6tZb/6epHNZfL1ajHGmHGjBn64YcflJaWJj8/Pz3xxBPq1q2bvv76a0mVDS3+/ve/KyMjQ9OmTdORI0d0zz33qF+/fjp16lSj1/eHP/xB//3vf/XYY4/p9OnTeuihh9SjRw8dPny4wd+rtt8nqXL1a9OmTTp8+LD27t2rL774glUvAF6P8AUAuKCNGzfq+PHjeuONN/Twww/r5z//uVJSUjy2ERopJiZGfn5+2rNnT43Xahurj8suu0ySlJmZWeO1zMxM9+vVOnbsqN/85jf6z3/+o++++052u11/+MMfPOZcddVV+v3vf6+vvvpKy5Yt044dO7RixYoGqfdCevXqpccff1wff/yxPvnkEx05ckRLly51v15bh8jo6GgFBATU+jPYtWuXzGazEhMT6/T+48aNk8Vi0d/+9jctW7ZMvr6+Gjt2bP0/EAC0AIQvAMAFVa88nb3SZLfb9ac//cmokjxYLBalpKRo1apVOnr0qHt8z549+uijjxrkPfr376+YmBgtXbrUY3vgRx99pJ07d+qmm26SVHlftNLSUo9jO3bsqODgYPdxJ0+erLFq17dvX0mq89bD+iosLFRFRYXHWK9evWQ2mz3eOzAwUPn5+R7zLBaLbrjhBv3zn//0aJufnZ2t5cuX69prr1VISEid6oiKitKoUaP09ttva9myZRo5cqSioqLq/bkAoCWg1TwA4IKuvvpqhYeHa9KkSXrooYdkMpn01ltvNattf/PmzdN//vMfXXPNNXrggQfkcDi0ePFi9ezZU9u3b6/TOcrLy/W73/2uxnhERIQefPBBPfPMM5oyZYqGDBmi8ePHu1vNJyUlaebMmZKkH374QcOHD9edd96p7t27y8fHR++9956ys7M1btw4SdKbb76pP/3pT7r11lvVsWNHFRUV6ZVXXlFISIhuvPHGBvuZ1Gb9+vWaNm2a7rjjDl1++eWqqKjQW2+9JYvFojFjxrjn9evXT+vWrdPzzz+vhIQEtW/fXsnJyfrd736ntWvX6tprr9WDDz4oHx8fvfzyyyorK9Ozzz57UbVMnDhRt99+uyRp/vz5Dfo5AaA5InwBAC4oMjJSq1ev1m9+8xs9/vjjCg8P1913363hw4drxIgRRpcnqTIsfPTRR3rkkUf0xBNPKDExUU899ZR27txZp26MUuVq3hNPPFFjvGPHjnrwwQc1efJkBQQEaMGCBXr00UcVGBioW2+9Vc8884y7g2FiYqLGjx+v9PR0vfXWW/Lx8VHXrl31zjvvuMPNkCFDtHnzZq1YsULZ2dkKDQ3VwIEDtWzZslobVDSkPn36aMSIEfrggw905MgRBQQEqE+fPvroo4901VVXuec9//zzuv/++/X444/r9OnTmjRpkpKTk9WjRw998sknmj17ttLS0uR0OpWcnKy33367xj2+LuTmm29WeHi4nE6nfvGLXzT0RwWAZsfkak7/2RIAgAY2evRo7dixQ7t37za6FPxERUWFEhISdPPNN+vVV181uhwAaHRc8wUA8BqnT5/2eL579259+OGHGjp0qDEF4bxWrVql3NxcTZw40ehSAKBJsPIFAPAa8fHxmjx5sjp06KADBw5oyZIlKisr09dff63OnTsbXR6qfPnll/rvf/+r+fPnKyoqStu2bTO6JABoElzzBQDwGiNHjtTf/vY3ZWVlyWazadCgQXr66acJXs3MkiVL9Pbbb6tv37564403jC4HAJoMK18AAAAA0ASaxTVfL730kpKSkuTn56fk5GRt3rz5nHNfeeUVXXfddQoPD1d4eLhSUlI85peXl+vRRx9Vr169FBgYqISEBE2cONHjvi+SlJSUJJPJ5PFYsGBBo31GAAAAAK2b4eFr5cqVSk1N1dy5c7Vt2zZ3C9ycnJxa52/cuFHjx4/Xhg0blJGRocTERN1www06cuSIpMqbW27btk1PPPGEtm3bpn/84x/KzMystYXtU089pWPHjrkf06dPb9TPCgAAAKD1MnzbYXJysgYMGKDFixdLkpxOpxITEzV9+nTNmjXrgsc7HA6Fh4dr8eLF5+yWtGXLFg0cOFAHDhxQu3btJFWufM2YMUMzZsyoV91Op1NHjx5VcHCwTCZTvc4BAAAAoOVzuVwqKipSQkKCzOZzr28Z2nDDbrdr69atmj17tnvMbDYrJSVFGRkZdTpHSUmJysvLFRERcc45BQUFMplM7htgVluwYIHmz5+vdu3a6a677tLMmTPl41P7j6SsrExlZWXu50eOHFH37t3rVCMAAAAA73fo0CG1bdv2nK8bGr7y8vLkcDgUGxvrMR4bG6tdu3bV6RyPPvqoEhISlJKSUuvrpaWlevTRRzV+/HiFhIS4xx966CFdeeWVioiI0Oeff67Zs2fr2LFjev7552s9T1pamp588ska44cOHfI4LwAAAIDWpbCwUImJiQoODj7vvBbdan7BggVasWKFNm7cKD8/vxqvl5eX684775TL5dKSJUs8XktNTXV/37t3b1mtVv3qV79SWlqabDZbjXPNnj3b45jqH3BISAjhCwAAAMAFL0cyNHxFRUXJYrEoOzvbYzw7O1txcXHnPXbhwoVasGCB1q1bp969e9d4vTp4HThwQOvXr79gQEpOTlZFRYX279+vLl261HjdZrPVGsoAAAAAoC4M7XZotVrVr18/paenu8ecTqfS09M1aNCgcx737LPPav78+VqzZo369+9f4/Xq4LV7926tW7dOkZGRF6xl+/btMpvNiomJqd+HAQAAAIDzMHzbYWpqqiZNmqT+/ftr4MCBWrRokYqLizVlyhRJ0sSJE9WmTRulpaVJkp555hnNmTNHy5cvV1JSkrKysiRJQUFBCgoKUnl5uW6//XZt27ZNq1evlsPhcM+JiIiQ1WpVRkaGvvzyS11//fUKDg5WRkaGZs6cqbvvvlvh4eHG/CAAAAAAeDXDw9fYsWOVm5urOXPmKCsrS3379tWaNWvcTTgOHjzo0a5xyZIlstvtuv322z3OM3fuXM2bN09HjhzR+++/L0nq27evx5wNGzZo6NChstlsWrFihebNm6eysjK1b99eM2fO9LimCwAAAK2Dy+VSRUWFHA6H0aWgmbJYLPLx8bnkW0wZfp+vlqqwsFChoaEqKCig4QYAAEALZbfbdezYMZWUlBhdCpq5gIAAxcfHy2q11nitrtnA8JUvAAAAwAhOp1M//vijLBaLEhISZLVaL3llA97H5XLJbrcrNzdXP/74ozp37nzeGymfD+ELAAAArZLdbpfT6VRiYqICAgKMLgfNmL+/v3x9fXXgwAHZ7fZab3NVF4Z2OwQAAACMVt9VDLQuDfF7wm8aAAAAADQBwhcAAAAANAHCFwAAAAAlJSVp0aJFdZ6/ceNGmUwm5efnN1pN3obwBQAAALQgJpPpvI958+bV67xbtmzR/fffX+f5V199tY4dO6bQ0NB6vV9deVPIo9uhl3A4XbKYaY0KAADg7Y4dO+b+fuXKlZozZ44yMzPdY0FBQe7vXS6XHA6HfHwu/Gd/dHT0RdVhtVoVFxd3Uce0dqx8tXCHT5Zo0mub9bPnN4n7ZQMAAFwal8ulEnuFIY+6/i0XFxfnfoSGhspkMrmf79q1S8HBwfroo4/Ur18/2Ww2ffrpp9q7d69uueUWxcbGKigoSAMGDNC6des8zvvTbYcmk0l/+ctfdOuttyogIECdO3fW+++/7379pytSb7zxhsLCwvTvf/9b3bp1U1BQkEaOHOkRFisqKvTQQw8pLCxMkZGRevTRRzVp0iSNHj263v/MTp48qYkTJyo8PFwBAQEaNWqUdu/e7X79wIEDuvnmmxUeHq7AwED16NFDH374ofvYCRMmKDo6Wv7+/urcubNef/31etdyIax8tXCRgTZ9se+4yiqc+iH7lLrEBRtdEgAAQIt1utyh7nP+bch7f//UCAVYG+bP81mzZmnhwoXq0KGDwsPDdejQId144436/e9/L5vNpr/+9a+6+eablZmZqXbt2p3zPE8++aSeffZZPffcc3rxxRc1YcIEHThwQBEREbXOLykp0cKFC/XWW2/JbDbr7rvv1iOPPKJly5ZJkp555hktW7ZMr7/+urp166b/+7//06pVq3T99dfX+7NOnjxZu3fv1vvvv6+QkBA9+uijuvHGG/X999/L19dXU6dOld1u18cff6zAwEB9//337tXBJ554Qt9//70++ugjRUVFac+ePTp9+nS9a7kQwlcL52+16OqOkdqQmav0XdmELwAAAOipp57Sz372M/fziIgI9enTx/18/vz5eu+99/T+++9r2rRp5zzP5MmTNX78eEnS008/rRdeeEGbN2/WyJEja51fXl6upUuXqmPHjpKkadOm6amnnnK//uKLL2r27Nm69dZbJUmLFy92r0LVR3Xo+uyzz3T11VdLkpYtW6bExEStWrVKd9xxhw4ePKgxY8aoV69ekqQOHTq4jz948KCuuOIK9e/fX1Ll6l9jInx5gWHdYrUhM1frd+bowaGdjC4HAACgxfL3tej7p0YY9t4NpTpMVDt16pTmzZunf/3rXzp27JgqKip0+vRpHTx48Lzn6d27t/v7wMBAhYSEKCcn55zzAwIC3MFLkuLj493zCwoKlJ2drYEDB7pft1gs6tevn5xO50V9vmo7d+6Uj4+PkpOT3WORkZHq0qWLdu7cKUl66KGH9MADD+g///mPUlJSNGbMGPfneuCBBzRmzBht27ZNN9xwg0aPHu0OcY2Ba768wLCuMZKkbQdP6mSx3eBqAAAAWi6TyaQAq48hD5Op4ZqnBQYGejx/5JFH9N577+npp5/WJ598ou3bt6tXr16y28//t6Ovr2+Nn8/5glJt843uS/DLX/5S+/bt0//8z//o22+/Vf/+/fXiiy9KkkaNGqUDBw5o5syZOnr0qIYPH65HHnmk0WohfHmBNmH+6hoXLKdL2vRDrtHlAAAAoJn57LPPNHnyZN16663q1auX4uLitH///iatITQ0VLGxsdqyZYt7zOFwaNu2bfU+Z7du3VRRUaEvv/zSPXb8+HFlZmaqe/fu7rHExET9+te/1j/+8Q/95je/0SuvvOJ+LTo6WpMmTdLbb7+tRYsW6c9//nO967kQth16ieu7xmhXVpHW78rR6CvaGF0OAAAAmpHOnTvrH//4h26++WaZTCY98cQT9d7qdymmT5+utLQ0derUSV27dtWLL76okydP1mnV79tvv1Vw8Jn+BiaTSX369NEtt9yi++67Ty+//LKCg4M1a9YstWnTRrfccoskacaMGRo1apQuv/xynTx5Uhs2bFC3bt0kSXPmzFG/fv3Uo0cPlZWVafXq1e7XGgPhy0sM7xqjJRv3amNmjiocTvlYWNQEAABApeeff1733HOPrr76akVFRenRRx9VYWFhk9fx6KOPKisrSxMnTpTFYtH999+vESNGyGK58PVugwcP9nhusVhUUVGh119/XQ8//LB+/vOfy263a/Dgwfrwww/dWyAdDoemTp2qw4cPKyQkRCNHjtQf//hHSZX3Kps9e7b2798vf39/XXfddVqxYkXDf/AqJpfRmzBbqMLCQoWGhqqgoEAhISFGlyOH06V+v1ur/JJyrbz/KiV3iDS6JAAAgGattLRUP/74o9q3by8/Pz+jy2mVnE6nunXrpjvvvFPz5883upzzOt/vS12zAcsjXsJiNmno5ZV3JV+fee4ONAAAAIBRDhw4oFdeeUU//PCDvv32Wz3wwAP68ccfdddddxldWpMgfHmRYd1iJUnrdxK+AAAA0PyYzWa98cYbGjBggK655hp9++23WrduXaNeZ9WccM2XFxnSOVoWs0m7c07p0IkSJUYEGF0SAAAA4JaYmKjPPvvM6DIMw8qXFwkN8FW/y8IlSet3sfoFAAAANCeELy8zvOqGy+mELwAAgDqh/xzqoiF+TwhfXmZYVfj6Yu9xFZdVGFwNAABA81XdirykpMTgStASVP+eVP/e1AfXfHmZTjFBSozw16ETp/XZnjzd0CPO6JIAAACaJYvForCwMOXkVO4YCggIqNPNftG6uFwulZSUKCcnR2FhYXW6J9m5EL68jMlk0rAuMXoz44A2ZOYQvgAAAM4jLq7yb6XqAAacS1hYmPv3pb4IX15oWLdYvZlxQOk7c+RyufgvOAAAAOdgMpkUHx+vmJgYlZeXG10OmilfX99LWvGqRvjyQsntIxRgtSinqEw7jhaqZ5tQo0sCAABo1iwWS4P8cQ2cDw03vJCfr0XXdIqSRMt5AAAAoLkgfHkpWs4DAAAAzQvhy0tdXxW+/ns4X7lFZQZXAwAAAIDw5aViQ/zUs02IXC5pYyarXwAAAIDRCF9ebFjXWEnSBsIXAAAAYLhmEb5eeuklJSUlyc/PT8nJydq8efM5577yyiu67rrrFB4ervDwcKWkpNSY73K5NGfOHMXHx8vf318pKSnavXu3x5wTJ05owoQJCgkJUVhYmO69916dOnWqUT6fUYZVbT38+Ic82SucBlcDAAAAtG6Gh6+VK1cqNTVVc+fO1bZt29SnTx+NGDHinDe627hxo8aPH68NGzYoIyNDiYmJuuGGG3TkyBH3nGeffVYvvPCCli5dqi+//FKBgYEaMWKESktL3XMmTJigHTt2aO3atVq9erU+/vhj3X///Y3+eZtS7zahigqy6lRZhbbsP2F0OQAAAECrZnK5XC4jC0hOTtaAAQO0ePFiSZLT6VRiYqKmT5+uWbNmXfB4h8Oh8PBwLV68WBMnTpTL5VJCQoJ+85vf6JFHHpEkFRQUKDY2Vm+88YbGjRunnTt3qnv37tqyZYv69+8vSVqzZo1uvPFGHT58WAkJCRd838LCQoWGhqqgoEAhISGX8BNoXI+8+43+vvWw7r22vZ74eXejywEAAAC8Tl2zgaErX3a7XVu3blVKSop7zGw2KyUlRRkZGXU6R0lJicrLyxURESFJ+vHHH5WVleVxztDQUCUnJ7vPmZGRobCwMHfwkqSUlBSZzWZ9+eWXtb5PWVmZCgsLPR4tQXXLee73BQAAABjL0PCVl5cnh8Oh2NhYj/HY2FhlZWXV6RyPPvqoEhIS3GGr+rjznTMrK0sxMTEer/v4+CgiIuKc75uWlqbQ0FD3IzExsU71Ge3azlHytZj0Y16x9uV61zVtAAAAQEti+DVfl2LBggVasWKF3nvvPfn5+TXqe82ePVsFBQXux6FDhxr1/RpKsJ+vBravXBVk9QsAAAAwjqHhKyoqShaLRdnZ2R7j2dnZiouLO++xCxcu1IIFC/Sf//xHvXv3do9XH3e+c8bFxdVo6FFRUaETJ06c831tNptCQkI8Hi1Fdct5whcAAABgHEPDl9VqVb9+/ZSenu4eczqdSk9P16BBg8553LPPPqv58+drzZo1HtdtSVL79u0VFxfncc7CwkJ9+eWX7nMOGjRI+fn52rp1q3vO+vXr5XQ6lZyc3FAfr9mobjm/+ccTKiotN7gaAAAAoHUyfNthamqqXnnlFb355pvauXOnHnjgARUXF2vKlCmSpIkTJ2r27Nnu+c8884yeeOIJvfbaa0pKSlJWVpaysrLc9+gymUyaMWOGfve73+n999/Xt99+q4kTJyohIUGjR4+WJHXr1k0jR47Ufffdp82bN+uzzz7TtGnTNG7cuDp1Omxp2kcFqkNUoCqcLn2yO8/ocgAAAIBWycfoAsaOHavc3FzNmTNHWVlZ6tu3r9asWeNumHHw4EGZzWcy4pIlS2S323X77bd7nGfu3LmaN2+eJOm3v/2tiouLdf/99ys/P1/XXnut1qxZ43Fd2LJlyzRt2jQNHz5cZrNZY8aM0QsvvND4H9gg13eN0b5Pf9T6XTm6sVe80eUAAAAArY7h9/lqqVrKfb6qfb4nT3f95UtFBlq15X9TZDabjC4JAAAA8Aot4j5faDr9kyIUbPPR8WK7vjmcb3Q5AAAAQKtD+GolrD5mXXd5lCRpA10PAQAAgCZH+GpFqlvOpxO+AAAAgCZH+GpFhnaJlskk7ThaqKyCUqPLAQAAAFoVwlcrEhVkU5+2YZKkDZmsfgEAAABNifDVygyvuuFy+k7CFwAAANCUCF+tzPVV4euzPXkqLXcYXA0AAADQehC+WpkeCSGKDbHpdLlDX+w7bnQ5AAAAQKtB+GplTCaThlWtftFyHgAAAGg6hK9W6OyW8y6Xy+BqAAAAgNaB8NUKXdMpUlYfsw6fPK3dOaeMLgcAAABoFQhfrVCA1UeDOkRKktaz9RAAAABoEoSvVmp4t8rrvtbTch4AAABoEoSvVur6LpXha+vBk8ovsRtcDQAAAOD9CF+tVGJEgC6PDZLD6dKmH3KNLgcAAADweoSvVqy66yHXfQEAAACNj/DVilXf72vTD7mqcDgNrgYAAADwboSvVuzKdmEK9fdVfkm5vj6Ub3Q5AAAAgFcjfLViPhazhlweLYmthwAAAEBjI3y1crScBwAAAJoG4auVG3J5tMwmKTO7SIdPlhhdDgAAAOC1CF+tXFiAVf0uC5ckbWDrIQAAANBoCF9wt5xPJ3wBAAAAjYbwBXfL+c/3HleJvcLgagAAAADvRPiCLo8NUpswf9krnPp8z3GjywEAAAC8EuELMplM7q6HbD0EAAAAGgfhC5Kk66u2Hm7YlSOXy2VwNQAAAID3IXxBkjSoQ6T8fS3KKizV98cKjS4HAAAA8DqEL0iS/HwtuqZTpCRazgMAAACNgfAFN1rOAwAAAI2H8AW36pbz2w/l6/ipMoOrAQAAALwL4QtucaF+6h4fIpdL2piZa3Q5AAAAgFcxPHy99NJLSkpKkp+fn5KTk7V58+Zzzt2xY4fGjBmjpKQkmUwmLVq0qMac6td++pg6dap7ztChQ2u8/utf/7oxPl6LU91yfj1bDwEAAIAGZWj4WrlypVJTUzV37lxt27ZNffr00YgRI5STU/sf/iUlJerQoYMWLFiguLi4Wuds2bJFx44dcz/Wrl0rSbrjjjs85t13330e85599tmG/XAtVHXL+Y9/yFW5w2lwNQAAAID3MDR8Pf/887rvvvs0ZcoUde/eXUuXLlVAQIBee+21WucPGDBAzz33nMaNGyebzVbrnOjoaMXFxbkfq1evVseOHTVkyBCPeQEBAR7zQkJCGvzztUR92oYpMtCqorIKbdl/wuhyAAAAAK9hWPiy2+3aunWrUlJSzhRjNislJUUZGRkN9h5vv/227rnnHplMJo/Xli1bpqioKPXs2VOzZ89WSUnJec9VVlamwsJCj4c3sphNGtIlWhIt5wEAAICGZFj4ysvLk8PhUGxsrMd4bGyssrKyGuQ9Vq1apfz8fE2ePNlj/K677tLbb7+tDRs2aPbs2Xrrrbd09913n/dcaWlpCg0NdT8SExMbpMbmaDgt5wEAAIAG52N0AY3p1Vdf1ahRo5SQkOAxfv/997u/79Wrl+Lj4zV8+HDt3btXHTt2rPVcs2fPVmpqqvt5YWGh1waw6y6Pko/ZpH25xdqfV6ykqECjSwIAAABaPMNWvqKiomSxWJSdne0xnp2dfc5mGhfjwIEDWrdunX75y19ecG5ycrIkac+ePeecY7PZFBIS4vHwViF+vhqQFCGJrocAAABAQzEsfFmtVvXr10/p6enuMafTqfT0dA0aNOiSz//6668rJiZGN9100wXnbt++XZIUHx9/ye/rLWg5DwAAADQsQ7sdpqam6pVXXtGbb76pnTt36oEHHlBxcbGmTJkiSZo4caJmz57tnm+327V9+3Zt375ddrtdR44c0fbt22usWDmdTr3++uuaNGmSfHw8d1bu3btX8+fP19atW7V//369//77mjhxogYPHqzevXs3/oduIapbzn/543GdKqswuBoAAACg5TP0mq+xY8cqNzdXc+bMUVZWlvr27as1a9a4m3AcPHhQZvOZfHj06FFdccUV7ucLFy7UwoULNWTIEG3cuNE9vm7dOh08eFD33HNPjfe0Wq1at26dFi1apOLiYiUmJmrMmDF6/PHHG++DtkAdogKVFBmg/cdL9OnuXI3syaogAAAAcClMLpfLZXQRLVFhYaFCQ0NVUFDgtdd/PfXB93rtsx91R7+2eu6OPkaXAwAAADRLdc0Ghm47RPM2rGrr4YbMXDmdZHQAAADgUhC+cE4D20co0GpR3qkyfXukwOhyAAAAgBaN8IVzsvqYdV3naEl0PQQAAAAuFeEL5zWMlvMAAABAgyB84byu71IZvr49UqCcwlKDqwEAAABaLsIXzis62KY+bUMlSRsyWf0CAAAA6ovwhQsa1rXyvmvpOwlfAAAAQH0RvnBB1S3nP92Tp7IKh8HVAAAAAC0T4QsX1CMhRDHBNpXYHfpy3wmjywEAAABaJMIXLshsNrlXv+h6CAAAANQP4Qt1cn1V+ErflS2Xy2VwNQAAAEDLQ/hCnVzbKUpWi1mHTpzW3txTRpcDAAAAtDiEL9RJoM1HyR0iJLH1EAAAAKgPwhfqbHj11kNazgMAAAAXjfCFOqu+39dXB06qoKTc4GoAAACAloXwhTprFxmgTjFBcjhd+nh3rtHlAAAAAC0K4QsXZTgt5wEAAIB6IXzholS3nN+YmSOHk5bzAAAAQF0RvnBR+l0WrhA/H50sKdf2QyeNLgcAAABoMQhfuCi+FrOGdKHrIQAAAHCxCF+4aMO6Rkviui8AAADgYhC+cNGGXB4js0nalVWkI/mnjS4HAAAAaBEIX7hoEYFWXdEuXJK0gdUvAAAAoE4IX6iXYbScBwAAAC4K4Qv1MrxbZfj6bE+eTtsdBlcDAAAANH+EL9RLl9hgJYT6qazCqYx9eUaXAwAAADR7hC/Ui8lk0rButJwHAAAA6orwhXqrvu5rw64cuVwug6sBAAAAmjfCF+rt6o5R8vM162hBqXZlFRldDgAAANCsEb5Qb36+Fl3TMUoSXQ8BAACACyF84ZJcT8t5AAAAoE4IX7gk1dd9bTt4UieK7QZXAwAAADRfhoevl156SUlJSfLz81NycrI2b958zrk7duzQmDFjlJSUJJPJpEWLFtWYM2/ePJlMJo9H165dPeaUlpZq6tSpioyMVFBQkMaMGaPs7OyG/mitQkKYv7rGBcvlkjb9wOoXAAAAcC6Ghq+VK1cqNTVVc+fO1bZt29SnTx+NGDFCOTm1/xFfUlKiDh06aMGCBYqLizvneXv06KFjx465H59++qnH6zNnztQHH3ygd999V5s2bdLRo0d12223Nehna02G03IeAAAAuCBDw9fzzz+v++67T1OmTFH37t21dOlSBQQE6LXXXqt1/oABA/Tcc89p3Lhxstls5zyvj4+P4uLi3I+oqCj3awUFBXr11Vf1/PPPa9iwYerXr59ef/11ff755/riiy8a/DO2BtVbDz/+IVflDqfB1QAAAADNk2Hhy263a+vWrUpJSTlTjNmslJQUZWRkXNK5d+/erYSEBHXo0EETJkzQwYMH3a9t3bpV5eXlHu/btWtXtWvX7rzvW1ZWpsLCQo8HKvVNDFd4gK8KSyu09cBJo8sBAAAAmiXDwldeXp4cDodiY2M9xmNjY5WVlVXv8yYnJ+uNN97QmjVrtGTJEv3444+67rrrVFRUeR+qrKwsWa1WhYWFXdT7pqWlKTQ01P1ITEysd43exmI26foudD0EAAAAzsfwhhsNbdSoUbrjjjvUu3dvjRgxQh9++KHy8/P1zjvvXNJ5Z8+erYKCAvfj0KFDDVSxd6DlPAAAAHB+Pka9cVRUlCwWS40ug9nZ2edtpnGxwsLCdPnll2vPnj2SpLi4ONntduXn53usfl3ofW0223mvM2vtBl8eLYvZpD05p3TweInaRQYYXRIAAADQrBi28mW1WtWvXz+lp6e7x5xOp9LT0zVo0KAGe59Tp05p7969io+PlyT169dPvr6+Hu+bmZmpgwcPNuj7tjah/r7qf1m4JGn9Ltr2AwAAAD9l6LbD1NRUvfLKK3rzzTe1c+dOPfDAAyouLtaUKVMkSRMnTtTs2bPd8+12u7Zv367t27fLbrfryJEj2r59u3tVS5IeeeQRbdq0Sfv379fnn3+uW2+9VRaLRePHj5ckhYaG6t5771Vqaqo2bNigrVu3asqUKRo0aJCuuuqqpv0BeBl3y3m2HgIAAAA1GLbtUJLGjh2r3NxczZkzR1lZWerbt6/WrFnjbsJx8OBBmc1n8uHRo0d1xRVXuJ8vXLhQCxcu1JAhQ7Rx40ZJ0uHDhzV+/HgdP35c0dHRuvbaa/XFF18oOjrafdwf//hHmc1mjRkzRmVlZRoxYoT+9Kc/Nc2H9mLDusbq6Q936ct9J1RcVqFAm6G/XgAAAECzYnK5XC6ji2iJCgsLFRoaqoKCAoWEhBhdTrPgcrk05LmNOniiRC//Tz+N6NFw1+4BAAAAzVVds4HXdTuEcUwmk/uGy+t3svUQAAAAOBvhCw2qOnxtyMyR08miKgAAAFCN8IUGldwhQgFWi3KKyrTjaKHR5QAAAADNBuELDcrmY9F1naMkSem0nAcAAADcCF9ocO6th7ScBwAAANwIX2hw13epDF/fHC5QTlGpwdUAAAAAzQPhCw0uJsRPvdqESpI2ZuYaXA0AAADQPBC+0ChoOQ8AAAB4InyhUQzvVhm+Ptmdq7IKh8HVAAAAAMYjfKFR9EwIVVSQTcV2h7b8eNLocgAAAADDEb7QKMxmk4Z1jZZEy3kAAABAInyhEbmv+9qVI5fLZXA1AAAAgLEIX2g013aOlq/FpAPHS7Qvr9jocgAAAABDEb7QaIJsPrqqQ6Qkuh4CAAAAhC80quobLq/fRfgCAABA60b4QqOqbjm/Zf8JFZaWG1wNAAAAYBzCFxrVZZGB6hAdqAqnS5/8kGd0OQAAAIBhCF9odMOruh7Sch4AAACtGeELjW5Y11hJ0sbMXDmctJwHAABA60T4QqPrnxSuYD8fnSi265vD+UaXAwAAABiC8IVG52sxa/Dl0ZJoOQ8AAIDWi/CFJjGMlvMAAABo5QhfaBJDu0TLZJK+P1aoYwWnjS4HAAAAaHKELzSJyCCbrkgMkyRt2JVrbDEAAACAAQhfaDLDulZvPaTlPAAAAFofwheaTHXL+U/35Km03GFwNQAAAEDTInyhyXSLD1Z8qJ9Ky53K2Hfc6HIAAACAJkX4QpMxmUy6vnrrIS3nAQAA0MoQvtCkhnc903Le5XIZXA0AAADQdAhfaFJXd4ySzcesI/mn9UP2KaPLAQAAAJoM4QtNyt9q0dUdIyVJ6XQ9BAAAQCtC+EKTq245v2EX130BAACg9TA8fL300ktKSkqSn5+fkpOTtXnz5nPO3bFjh8aMGaOkpCSZTCYtWrSoxpy0tDQNGDBAwcHBiomJ0ejRo5WZmekxZ+jQoTKZTB6PX//61w390XAO1U03th44qZPFdoOrAQAAAJqGoeFr5cqVSk1N1dy5c7Vt2zb16dNHI0aMUE5O7SsiJSUl6tChgxYsWKC4uLha52zatElTp07VF198obVr16q8vFw33HCDiouLPebdd999OnbsmPvx7LPPNvjnQ+3ahgeoS2ywnC7p4925RpcDAAAANAkfI9/8+eef13333acpU6ZIkpYuXap//etfeu211zRr1qwa8wcMGKABAwZIUq2vS9KaNWs8nr/xxhuKiYnR1q1bNXjwYPd4QEDAOQMcGt+wbjHKzC5S+s4c3dK3jdHlAAAAAI3OsJUvu92urVu3KiUl5UwxZrNSUlKUkZHRYO9TUFAgSYqIiPAYX7ZsmaKiotSzZ0/Nnj1bJSUl5z1PWVmZCgsLPR6ov+qW8xszc1ThcBpcDQAAAND4DFv5ysvLk8PhUGxsrMd4bGysdu3a1SDv4XQ6NWPGDF1zzTXq2bOne/yuu+7SZZddpoSEBP33v//Vo48+qszMTP3jH/8457nS0tL05JNPNkhdkK5oF66wAF/ll5Rr28F8DWwfceGDAAAAgBbM0G2HjW3q1Kn67rvv9Omnn3qM33///e7ve/Xqpfj4eA0fPlx79+5Vx44daz3X7NmzlZqa6n5eWFioxMTExim8FbCYTRp6ebRWbT+q9F3ZhC8AAAB4PcO2HUZFRclisSg72/NeT9nZ2Q1yLda0adO0evVqbdiwQW3btj3v3OTkZEnSnj17zjnHZrMpJCTE44FLcz0t5wEAANCKGBa+rFar+vXrp/T0dPeY0+lUenq6Bg0aVO/zulwuTZs2Te+9957Wr1+v9u3bX/CY7du3S5Li4+Pr/b64eEMuj5bFbNIP2ad06MT5r7kDAAAAWjpDtx2mpqZq0qRJ6t+/vwYOHKhFixapuLjY3f1w4sSJatOmjdLS0iRVNun4/vvv3d8fOXJE27dvV1BQkDp16iSpcqvh8uXL9c9//lPBwcHKysqSJIWGhsrf31979+7V8uXLdeONNyoyMlL//e9/NXPmTA0ePFi9e/c24KfQeoUFWNXvsnBt/vGE1u/K0aSrk4wuCQAAAGg0hoavsWPHKjc3V3PmzFFWVpb69u2rNWvWuJtwHDx4UGbzmcW5o0eP6oorrnA/X7hwoRYuXKghQ4Zo48aNkqQlS5ZIqryR8tlef/11TZ48WVarVevWrXMHvcTERI0ZM0aPP/54435Y1GpY1xjCFwAAAFoFk8vlchldREtUWFio0NBQFRQUcP3XJdidXaSf/fFjWX3M2j7nZwqwenUPGAAAAHihumYDw675AiSpU0yQ2ob7y17h1Gd7jhtdDgAAANBo6hW+Dh06pMOHD7ufb968WTNmzNCf//znBisMrYPJZHLfcHn9ruwLzAYAAABarnqFr7vuuksbNmyQJGVlZelnP/uZNm/erP/93//VU0891aAFwvsN61Z5jd/6XTliFywAAAC8Vb3C13fffaeBAwdKkt555x317NlTn3/+uZYtW6Y33nijIetDK5DcPkL+vhZlF5Zpx9FCo8sBAAAAGkW9wld5eblsNpskad26dfrFL34hSeratauOHTvWcNWhVfDztejazlGSKle/AAAAAG9Ur/DVo0cPLV26VJ988onWrl2rkSNHSqpsBR8ZGdmgBaJ1GOa+7ovwBQAAAO9Ur/D1zDPP6OWXX9bQoUM1fvx49enTR5L0/vvvu7cjAhfj+i6V4eubw/nKLSozuBoAAACg4dXrpkpDhw5VXl6eCgsLFR4e7h6///77FRAQ0GDFofWIC/VTzzYh+u5IoTZm5uiO/olGlwQAAAA0qHqtfJ0+fVplZWXu4HXgwAEtWrRImZmZiomJadAC0XoMq1r92pDJ1kMAAAB4n3qFr1tuuUV//etfJUn5+flKTk7WH/7wB40ePVpLlixp0ALRelS3nP/4hzzZK5wGVwMAAAA0rHqFr23btum6666TJP39739XbGysDhw4oL/+9a964YUXGrRAtB6924QqKsiqU2UV+mr/CaPLAQAAABpUvcJXSUmJgoODJUn/+c9/dNttt8lsNuuqq67SgQMHGrRAtB5ms0lDq7YeptP1EAAAAF6mXuGrU6dOWrVqlQ4dOqR///vfuuGGGyRJOTk5CgkJadAC0boMp+U8AAAAvFS9wtecOXP0yCOPKCkpSQMHDtSgQYMkVa6CXXHFFQ1aIFqXaztHyddi0o95xdqXe8rocgAAAIAGU6/wdfvtt+vgwYP66quv9O9//9s9Pnz4cP3xj39ssOLQ+gT7+Wpg+whJrH4BAADAu9QrfElSXFycrrjiCh09elSHDx+WJA0cOFBdu3ZtsOLQOl1Py3kAAAB4oXqFL6fTqaeeekqhoaG67LLLdNlllyksLEzz58+X00mLcFya4VUt57/cd0JFpeUGVwMAAAA0DJ/6HPS///u/evXVV7VgwQJdc801kqRPP/1U8+bNU2lpqX7/+983aJFoXdpHBapDVKD25RXr0915GtUr3uiSAAAAgEtWr/D15ptv6i9/+Yt+8YtfuMd69+6tNm3a6MEHHyR84ZJd3zVG+z79Uem7cghfAAAA8Ar12nZ44sSJWq/t6tq1q06c4Oa4uHTVLec3ZubI6XQZXA0AAABw6eoVvvr06aPFixfXGF+8eLF69+59yUUB/ZMiFGTzUd4pu/57pMDocgAAAIBLVq9th88++6xuuukmrVu3zn2Pr4yMDB06dEgffvhhgxaI1snqY9bgy6P04bdZWr8zW30Tw4wuCQAAALgk9Vr5GjJkiH744Qfdeuutys/PV35+vm677Tbt2LFDb731VkPXiFZqWNfKrofraTkPAAAAL2ByuVwNdkHNN998oyuvvFIOh6OhTtlsFRYWKjQ0VAUFBQoJCTG6HK+Ud6pMA36/Ti6X9MXs4YoL9TO6JAAAAKCGumaDet9kGWhsUUE29WkbJokbLgMAAKDlI3yhWRtW1fVw/S7CFwAAAFo2wheaterw9enuPJWWe/92VgAAAHivi+p2eNttt5339fz8/EupBaihR0KIYkNsyi4s05c/ntCQy6ONLgkAAACol4sKX6GhoRd8feLEiZdUEHA2k8mkYV1j9LfNh7R+ZzbhCwAAAC3WRYWv119/vbHqAM5pWNdY/W3zIaXvytG8X7hkMpmMLgkAAAC4aFzzhWbvmk6RsvqYdfjkae3JOWV0OQAAAEC9EL7Q7AVYfTSoQ6QkKZ2uhwAAAGihDA9fL730kpKSkuTn56fk5GRt3rz5nHN37NihMWPGKCkpSSaTSYsWLarXOUtLSzV16lRFRkYqKChIY8aMUXZ2dkN+LDQwWs4DAACgpTM0fK1cuVKpqamaO3eutm3bpj59+mjEiBHKyan9D+ySkhJ16NBBCxYsUFxcXL3POXPmTH3wwQd69913tWnTJh09evSCnRxhrOrwtfXASeWX2A2uBgAAALh4JpfL5TLqzZOTkzVgwAAtXrxYkuR0OpWYmKjp06dr1qxZ5z02KSlJM2bM0IwZMy7qnAUFBYqOjtby5ct1++23S5J27dqlbt26KSMjQ1dddVWdai8sLFRoaKgKCgoUEhJykZ8c9XHDHzfph+xT+r9xfXVL3zZGlwMAAABIqns2MGzly263a+vWrUpJSTlTjNmslJQUZWRkNNo5t27dqvLyco85Xbt2Vbt27c77vmVlZSosLPR4oGldX7X6tYGthwAAAGiBDAtfeXl5cjgcio2N9RiPjY1VVlZWo50zKytLVqtVYWFhF/W+aWlpCg0NdT8SExPrVSPqb3jXyn+uG3/IlcNp2IItAAAAUC+GN9xoKWbPnq2CggL349ChQ0aX1Opc2S5Mof6+yi8p19cHTxpdDgAAAHBRDAtfUVFRslgsNboMZmdnn7OZRkOcMy4uTna7Xfn5+Rf1vjabTSEhIR4PNC0fi1lDLo+WRMt5AAAAtDyGhS+r1ap+/fopPT3dPeZ0OpWenq5BgwY12jn79esnX19fjzmZmZk6ePBgvd8XTWd4t6qW8zsJXwAAAGhZfIx889TUVE2aNEn9+/fXwIEDtWjRIhUXF2vKlCmSpIkTJ6pNmzZKS0uTVNlQ4/vvv3d/f+TIEW3fvl1BQUHq1KlTnc4ZGhqqe++9V6mpqYqIiFBISIimT5+uQYMG1bnTIYwz5PJomU1SZnaRDp8sUdvwAKNLAgAAAOrE0PA1duxY5ebmas6cOcrKylLfvn21Zs0ad8OMgwcPymw+szh39OhRXXHFFe7nCxcu1MKFCzVkyBBt3LixTueUpD/+8Y8ym80aM2aMysrKNGLECP3pT39qmg+NSxIWYFW/y8K1Zf9JbdiVo/8ZlGR0SQAAAECdGHqfr5aM+3wZ508b9+jZNZm6vku0Xp8y0OhyAAAA0Mo1+/t8AfVV3XL+s73HVWKvMLgaAAAAoG4IX2hxLo8NUpswf9krnPp8z3GjywEAAADqhPCFFsdkMmlY16quh5l0PQQAAEDLQPhCizSsquX8hl054rJFAAAAtASEL7RIgzpEys/XrGMFpdp5rMjocgAAAIALInyhRfLztejaTlGSpPW7sg2uBgAAALgwwhdarGFVXQ/Td3HdFwAAAJo/whdarOu7RkuSth/K1/FTZQZXAwAAAJwf4QstVnyov7rHh8jlkjZm5hpdDgAAAHBehC+0aLScBwAAQEtB+EKLVt1y/uPMXJU7nAZXAwAAAJwb4QstWp+2YYoMtKqorEJf7T9pdDkAAADAORG+0KJZzCYN6VLZeIOW8wAAAGjOCF9o8YbTch4AAAAtAOELLd51l0fJx2zSvtxi7c8rNrocAAAAoFaEL7R4IX6+GpAUIUlaz+oXAAAAminCF7zC8KquhxtoOQ8AAIBmivAFr3B91f2+vth3XKfKKgyuBgAAAKiJ8AWv0CEqUEmRASp3uPTp7lyjywEAAABqIHzBK5hMJvfqF9d9AQAAoDkifMFrVLecX78rV06ny+BqAAAAAE+EL3iNge0jFGi1KO9Umb47WmB0OQAAAIAHwhe8htXHrOs6R0uS0ney9RAAAADNC+ELXmVYN677AgAAQPNE+IJXGdqlcuXr2yMFyiksNbgaAAAA4AzCF7xKTLCf+rQNlcQNlwEAANC8EL7gdYa5ux4SvgAAANB8EL7gdYZV3e/rk915KqtwGFwNAAAAUInwBa/TIyFEMcE2ldgd+nLfCaPLAQAAACQRvuCFzGaTru9C10MAAAA0L4QveKWzW867XC6DqwEAAAAIX/BS13aKktVi1sETJdqbW2x0OQAAAEDzCF8vvfSSkpKS5Ofnp+TkZG3evPm8899991117dpVfn5+6tWrlz788EOP100mU62P5557zj0nKSmpxusLFixolM+Hphdo81FyhwhJ0vpd2QZXAwAAADSD8LVy5UqlpqZq7ty52rZtm/r06aMRI0YoJ6f2a3U+//xzjR8/Xvfee6++/vprjR49WqNHj9Z3333nnnPs2DGPx2uvvSaTyaQxY8Z4nOupp57ymDd9+vRG/axoWsOruh6m7+S6LwAAABjP5DL4gpjk5GQNGDBAixcvliQ5nU4lJiZq+vTpmjVrVo35Y8eOVXFxsVavXu0eu+qqq9S3b18tXbq01vcYPXq0ioqKlJ6e7h5LSkrSjBkzNGPGjHrVXVhYqNDQUBUUFCgkJKRe50DjOni8RIOf2yCL2aRtT/xMof6+RpcEAAAAL1TXbGDoypfdbtfWrVuVkpLiHjObzUpJSVFGRkatx2RkZHjMl6QRI0acc352drb+9a9/6d57763x2oIFCxQZGakrrrhCzz33nCoqKs5Za1lZmQoLCz0eaN7aRQaoU0yQHE6XPv4h1+hyAAAA0MoZGr7y8vLkcDgUGxvrMR4bG6usrKxaj8nKyrqo+W+++aaCg4N12223eYw/9NBDWrFihTZs2KBf/epXevrpp/Xb3/72nLWmpaUpNDTU/UhMTKzLR4TBqm+4vIGW8wAAADCYj9EFNLbXXntNEyZMkJ+fn8d4amqq+/vevXvLarXqV7/6ldLS0mSz2WqcZ/bs2R7HFBYWEsBagGFdY/Tnj/dpQ2aOHE6XLGaT0SUBAACglTJ05SsqKkoWi0XZ2Z7d6LKzsxUXF1frMXFxcXWe/8knnygzM1O//OUvL1hLcnKyKioqtH///lpft9lsCgkJ8Xig+et3WbhC/Hx0sqRc2w+dNLocAAAAtGKGhi+r1ap+/fp5NMJwOp1KT0/XoEGDaj1m0KBBHvMlae3atbXOf/XVV9WvXz/16dPngrVs375dZrNZMTExF/kp0Jz5WswafHm0pMobLgMAAABGMbzVfGpqql555RW9+eab2rlzpx544AEVFxdrypQpkqSJEydq9uzZ7vkPP/yw1qxZoz/84Q/atWuX5s2bp6+++krTpk3zOG9hYaHefffdWle9MjIytGjRIn3zzTfat2+fli1bppkzZ+ruu+9WeHh4435gNLnh3Wg5DwAAAOMZfs3X2LFjlZubqzlz5igrK0t9+/bVmjVr3E01Dh48KLP5TEa8+uqrtXz5cj3++ON67LHH1LlzZ61atUo9e/b0OO+KFSvkcrk0fvz4Gu9ps9m0YsUKzZs3T2VlZWrfvr1mzpzpcU0XvMeQy2NkMkm7sop0NP+0EsL8jS4JAAAArZDh9/lqqbjPV8syZsnn2nrgpH43uqfuvuoyo8sBAACAF2kR9/kCmkp1y3mu+wIAAIBRCF9oFarD12d78nTa7jC4GgAAALRGhC+0Cl3jgpUQ6qeyCqcy9uUZXQ4AAABaIcIXWgWTyaTr2XoIAAAAAxG+0GpUt5xfvzNH9JkBAABAUyN8odW4umOU/HzNOlpQql1ZRUaXAwAAgFaG8IVWw8/Xoqs7Rkli6yEAAACaHuELrQot5wEAAGAUwhdaleqmG18fPKkTxXaDqwEAAEBrQvhCq9ImzF9d44LldEmbfmD1CwAAAE2H8IVWx931cFeuwZUAAACgNSF8odWpvu5rU2aOyh1Og6sBAABAa0H4QqvTNzFc4QG+Kiyt0NYDJ40uBwAAAK0E4QutjsVs0tAulatfG+h6CAAAgCZC+EKrVL31MJ3wBQAAgCZC+EKrNPjyaFnMJu3JOaV/bj/CtV8AAABodIQvtEqh/r4acnm0JOnhFdt1zYL1WvjvTB06UWJwZQAAAPBWJpfL5TK6iJaosLBQoaGhKigoUEhIiNHloB7yS+xaummf/r71kPJOVd5w2WSSruscrbsGttPwbjHytfDfJwAAAHB+dc0GhK96Inx5D3uFU2u/z9bfNh/Up3vy3OPRwTbd2b+txg1op8SIAAMrBAAAQHNG+GpkhC/vdOB4sVZsOaR3v6ptNSxRw7vFshoGAAAAD4SvRkb48m72CqfW7czW8i9ZDQMAAMD5Eb4aGeGr9TjXati1naI0Ibkdq2EAAACtHOGrkRG+Wh97hVPpO7O1fPNBfbL7zGpYVFDlatj4gayGAQAAtEaEr0ZG+GrdDh4v0d+2HPRYDZOk6zpH6a6B7ZTSndUwAACA1oLw1cgIX5AuvBo2bkA7tYtkNQwAAMCbEb4aGeELP3XweIlWbDmod746rLxTZe5xVsMAAAC8G+GrkRG+cC7lDqfWfV/7atgd/dtq3IBEXRYZaGCFAAAAaEiEr0ZG+EJdHDxeopVfVa6G5RZ5roaNH9hOP2M1DAAAoMUjfDUywhcuRrmj8tqwZV+yGgYAAOBtCF+NjPCF+jp04sy1YWevhl3bKUp3JbdTSrdYWX1YDQMAAGgpCF+NjPCFS1W5GpZTdW1Yrqr/lxgVZNXt/RI1fiCrYQAAAC0B4auREb7QkM63GlZ9bRirYQAAAM1TXbNBs/hr7qWXXlJSUpL8/PyUnJyszZs3n3f+u+++q65du8rPz0+9evXShx9+6PH65MmTZTKZPB4jR470mHPixAlNmDBBISEhCgsL07333qtTp041+GcD6iIxIkD/b0RXfT5rmJbe3U9DLo+WySR9uidPU5dv09UL0rXgo13an1dsdKkAAACoJ8PD18qVK5Wamqq5c+dq27Zt6tOnj0aMGKGcnJxa53/++ecaP3687r33Xn399dcaPXq0Ro8ere+++85j3siRI3Xs2DH3429/+5vH6xMmTNCOHTu0du1arV69Wh9//LHuv//+RvucQF34Wswa2TNOb94zUB//v+s17fpOig62Ke+UXUs37dXQhRs14S9f6F//PSZ7hdPocgEAAHARDN92mJycrAEDBmjx4sWSJKfTqcTERE2fPl2zZs2qMX/s2LEqLi7W6tWr3WNXXXWV+vbtq6VLl0qqXPnKz8/XqlWran3PnTt3qnv37tqyZYv69+8vSVqzZo1uvPFGHT58WAkJCResm22HaCrV14b9bfNBfXzWtWGRgVbd3r+txg9op6Qorg0DAAAwSovYdmi327V161alpKS4x8xms1JSUpSRkVHrMRkZGR7zJWnEiBE15m/cuFExMTHq0qWLHnjgAR0/ftzjHGFhYe7gJUkpKSkym8368ssva33fsrIyFRYWejyApvDT1bDpwzopJtim48V2vbxpn3s1bPV/j7IaBgAA0Iz5GPnmeXl5cjgcio2N9RiPjY3Vrl27aj0mKyur1vlZWVnu5yNHjtRtt92m9u3ba+/evXrsscc0atQoZWRkyGKxKCsrSzExMR7n8PHxUUREhMd5zpaWlqYnn3yyPh8TaDCJEQH6zQ1d9NDwzlq/q3I1bNMPufpsz3F9tuc4q2EAAADNmKHhq7GMGzfO/X2vXr3Uu3dvdezYURs3btTw4cPrdc7Zs2crNTXV/bywsFCJiYmXXCtQH74Ws0b0iNOIHnE6dKJE73x1SCu3HFJOUZle3rRPL2/ap6s7Ruqu5Ha6oXscnRIBAACaAUPDV1RUlCwWi7Kzsz3Gs7OzFRcXV+sxcXFxFzVfkjp06KCoqCjt2bNHw4cPV1xcXI2GHhUVFTpx4sQ5z2Oz2WSz2erysYAmVb0a9nDVatjyqtWwz/ce1+d7q1bD+rXVuIHt1J7VMAAAAMMY+p/DrVar+vXrp/T0dPeY0+lUenq6Bg0aVOsxgwYN8pgvSWvXrj3nfEk6fPiwjh8/rvj4ePc58vPztXXrVvec9evXy+l0Kjk5+VI+EmAYH4tZN/SI0xtTBuqT316vh4Z1UmxI1bVhH+/T9Qs36q5XvtAH3xxVWYXD6HIBAABaHcO7Ha5cuVKTJk3Syy+/rIEDB2rRokV65513tGvXLsXGxmrixIlq06aN0tLSJFW2mh8yZIgWLFigm266SStWrNDTTz+tbdu2qWfPnjp16pSefPJJjRkzRnFxcdq7d69++9vfqqioSN9++6179WrUqFHKzs7W0qVLVV5erilTpqh///5avnx5neqm2yFaggqH031t2MYfznRKjAi06g5WwwAAABpEXbOB4eFLkhYvXqznnntOWVlZ6tu3r1544QX3CtTQoUOVlJSkN954wz3/3Xff1eOPP679+/erc+fOevbZZ3XjjTdKkk6fPq3Ro0fr66+/Vn5+vhISEnTDDTdo/vz5Ho06Tpw4oWnTpumDDz6Q2WzWmDFj9MILLygoKKhONRO+0NIcPlmid7Yc0sqvDim7sMw9PqhD1bVhPWJl87EYWCEAAEDL1KLCV0tE+EJLdb7VsNv7tdV4VsMAAAAuCuGrkRG+4A0OnyzRO18d1sotB2usho1PbqcRrIYBAABcEOGrkRG+4E0qHE5tyMzV3zYf1IbMnBqrYeMGJKpDdN225AIAALQ2hK9GRviCtzqSf1ortxzSO1sOKauw1D1+VYcI3ZV8GathAAAAP0H4amSEL3i7CodTGzNztXzzQW3MzJGz6v8pwgN8NebKtrq2c5S6J4QoJtjP2EIBAAAMRvhqZIQvtCZH8k9Xdkr8yWqYJEUH29Q9PkTdE0LUIyFE3eNDlBQZKLPZZFC1AAAATYvw1cgIX2iNqlfD3v/mqL47WqAf84pV2/+DBFgt6hoXrB4JoepeFci6xAXLz5ftigAAwPsQvhoZ4QuQSuwV2pVVpB1HC/X90UJ9f6xQu44VqqzCWWOuxWxSx+hA9ypZ9/jKYBYRaDWgcgAAgIZD+GpkhC+gdhUOp37MK9b3x84Esh1HC3Wi2F7r/PhQP3WPr9qyWBXKEiP8ZTKxbREAALQMhK9GRvgC6s7lcim7sEzfHyvQ90crw9j3xwp14HhJrfODbT7qVrVdsXrb4uWxwbL6mJu4cgAAgAsjfDUywhdw6YpKy7XzWJG+P1pQuVJ2rFA/ZJ2S3VFz26KvxaROMcEeq2Td4kMU6u9rQOUAAABnEL4aGeELaBz2Cqf25p46a8ti5WpZYWlFrfPbhvtXdVmsau6REKKEUD+2LQIAgCZD+GpkhC+g6bhcLh3JP+3R2OP7o4U6kn+61vlhAb6VWxbdLfBD1SE6UL4Wti0CAICGR/hqZIQvwHj5JXaPxh7fHy3U7pxTcjhr/t+a1cesLrHBZzX2CFHX+BAF2XwMqBwAAHgTwlcjI3wBzVNpuUN7ck5VNfaovJZs57EinSqruW3RZJKSIgM9Gnv0SAhRdLCNbYsAAKDOCF+NjPAFtBxOp0uHTpbU2LaYVVha6/yoIKu6nbVlsXt8iNpHBcpiJpABAICaCF+NjPAFtHx5p8q0syqIVbe/35d7SrXsWpS/r0Vd44M9Vsm6xoXI32pp+sIBAECzQvhqZIQvwDudtjuUmV3k7rL4/bFC7TpWpNPljhpzzSapQ3TQWY09KkNZZJDNgMoBAIBRCF+NjPAFtB4Op0s/5hW7tytWB7PjxfZa58eG2NzbFatXydpFBMjMtkUAALwS4auREb6A1s3lcim3qEw7qrstVq2S/ZhXXOv8IJuPusUHq0dCqHokhKhX21B1ig6SD+3vAQBo8QhfjYzwBaA2p8oqtOtYoUcL/F1ZRbJXOGvMtfmY1S0+RD3bhKhnQqh6tgnV5bHBsvoQyAAAaEkIX42M8AWgrsodTu3LLdaOowXacbRQ3x6p3LZYW/t7X4tJXeKC1atNqHokhKpXm1B1iQuWny+NPQAAaK4IX42M8AXgUjidLh04UaJvjxRox5ECfXukQN8dKVBhac1A5mM2qXNssHomhKhnm8oVsu7xdFoEAKC5IHw1MsIXgIbmcrl0+ORpdxD77mihvjtSoBO1NPYwm6SO0UGVK2RtKlfIuieEKMjmY0DlAAC0boSvRkb4AtAUXC6XjhWUulfIvqvatphbVFZjrskktY8MrFodq1wl65EQqlB/XwMqBwCg9SB8NTLCFwAj5RSWVq2QFeq7o5UrZccKSmud2y4ioGqF7Exjj4hAaxNXDACA9yJ8NTLCF4DmJu9Umb47UtnU47uq68gOnzxd69w2Yf4eXRZ7tglVdDA3hwYAoD4IX42M8AWgJcgvsbs7LH5X9dh/vKTWubEhNo8w1qtNqGJDbDKZuDk0AADnQ/hqZIQvAC1VYWm5vq9aHateIduXV6za/m0QFWR1t7yvvo6sTZg/gQwAgLMQvhoZ4QuANykuq9DOY4Xu68h2HC3Q7pxTcjhr/isiLMD3rBWyyq2Ll0UGEMgAAK0W4auREb4AeLvTdod2ZVWvkFU29sjMKlJFLYEs2M9HPRJCqlbIKh/tIwNlNhPIAADej/DVyAhfAFqjsgqHfsg6VblCdrSy/f3OrCLZK5w15gZaLepefWPoqpWyjtGB8rGYDagcAIDGQ/hqZIQvAKhU7nBqd/Ypd8v7744U6PtjhSotrxnI/HzN6hZftUKWUNn+/vLYYPkSyAAALViLCl8vvfSSnnvuOWVlZalPnz568cUXNXDgwHPOf/fdd/XEE09o//796ty5s5555hndeOONkqTy8nI9/vjj+vDDD7Vv3z6FhoYqJSVFCxYsUEJCgvscSUlJOnDggMd509LSNGvWrDrVTPgCgHOrcDi1L6/Y3dBjR9V1ZMV2R425VotZXeODPRp7dIkLls3HYkDlAABcvBYTvlauXKmJEydq6dKlSk5O1qJFi/Tuu+8qMzNTMTExNeZ//vnnGjx4sNLS0vTzn/9cy5cv1zPPPKNt27apZ8+eKigo0O2336777rtPffr00cmTJ/Xwww/L4XDoq6++cp8nKSlJ9957r+677z73WHBwsAIDA+tUN+ELAC6O0+nSj8eL3atj1deRFZVW1JjrYzbp8thg9WxTuUrWOTZY0cE2RQXZFOLnQ3MPAECz0mLCV3JysgYMGKDFixdLkpxOpxITEzV9+vRaV6HGjh2r4uJirV692j121VVXqW/fvlq6dGmt77FlyxYNHDhQBw4cULt27SRVhq8ZM2ZoxowZ9aqb8AUAl87lcungiRJ9d6Sy0+KOo5UrZfkl5ec8xupjVnSQTVFBVkUFVQayqOAz31eHtOggm0L8CWoAgMZX12zg04Q11WC327V161bNnj3bPWY2m5WSkqKMjIxaj8nIyFBqaqrH2IgRI7Rq1apzvk9BQYFMJpPCwsI8xhcsWKD58+erXbt2uuuuuzRz5kz5+NT+IykrK1NZWZn7eWFh4QU+HQDgQkwmky6LDNRlkYG6qXe8pMpAdiT/tLvl/bdHCrQ/r1h5p+w6VVYhe4VTR/JP60j+6Que32oxK9Id0qzuYFYZ2KrGqp6HBfgS1AAAjcrQ8JWXlyeHw6HY2FiP8djYWO3atavWY7Kysmqdn5WVVev80tJSPfrooxo/frxHCn3ooYd05ZVXKiIiQp9//rlmz56tY8eO6fnnn6/1PGlpaXryyScv5uMBAOrBZDKpbXiA2oYHaGTPOI/XSssdyi0qU96pMuWdsivvVNlZz8uUV1Q1dqpMRaUVsjucOlZQqmMFpRd8X1+LSZGBnqtoZ4e2aHdgsynM35c2+gCAi2Zo+Gps5eXluvPOO+VyubRkyRKP185ePevdu7esVqt+9atfKS0tTTabrca5Zs+e7XFMYWGhEhMTG694AEANfr4WJUYEKDEi4IJzS8sdZ0La2QHtlF25Z4e2ojIVllao3OFSVmGpsgovHNR8zCZFBFo9tjlGBZ9ZRTt7K2REgJWgBgCQZHD4ioqKksViUXZ2tsd4dna24uLiaj0mLi6uTvOrg9eBAwe0fv36C16XlZycrIqKCu3fv19dunSp8brNZqs1lAEAmic/X4t7Be1CyiocOl5jJc3uuapW9Xp+SbkqnC7lFJUpp6hMOnb+c1vOCmrubY7BtWyDDLIpItAqC0ENALyWoeHLarWqX79+Sk9P1+jRoyVVNtxIT0/XtGnTaj1m0KBBSk9P92iUsXbtWg0aNMj9vDp47d69Wxs2bFBkZOQFa9m+fbvMZnOtHRYBAN7N5mNRQpi/EsL8LzjXXuHU8WLPLY5nQpvnKtvJknI5nC7lFlW+fiFmk2quqHk0FjkT2iICrNywGgBaGMO3HaampmrSpEnq37+/Bg4cqEWLFqm4uFhTpkyRJE2cOFFt2rRRWlqaJOnhhx/WkCFD9Ic//EE33XSTVqxYoa+++kp//vOfJVUGr9tvv13btm3T6tWr5XA43NeDRUREyGq1KiMjQ19++aWuv/56BQcHKyMjQzNnztTdd9+t8PBwY34QAIAWwepjVnyov+JDLxzUyh1OnSiuXEHLrdriWL2ClveT0HayxC6nS1Wv27Urq+i85zaZpIgAa41uj2eCm9VjRc3qQ1ADAKMZHr7Gjh2r3NxczZkzR1lZWerbt6/WrFnjbqpx8OBBmc1n/oVx9dVXa/ny5Xr88cf12GOPqXPnzlq1apV69uwpSTpy5Ijef/99SVLfvn093mvDhg0aOnSobDabVqxYoXnz5qmsrEzt27fXzJkza3RRBADgUvhazIoN8VNsiN8F51ZUB7WzrlM7E9g8Q9vxYrtcLul4sV3Hi+3KzL7g6RXs56PIQKsiAq2KDLIpMtCqyCCrIgIrg1pEoFWRgbaqMat8WVUDgAZn+H2+Wiru8wUAMEqFw6kTJXb31keP69KqQtvZK2oO58X/qz7Ez8e9alYjpHmENytbIAG0ei3iPl8AAODi+VjMign2U0zwhVfUnE6XCkvLlXfKrhPFdh2vWjk7fsquE8Vlyiu268Qpu44Xl+lEceUcp0sqLK1QYWmF9uUV16mmsADfyuvVAs8EtsiqoHbmeeXKWngAjUUAtE6ELwAAvJjZbFJYgFVhAdY6zXc4XSo4Xe4OaTUDW+Vq24mqLY8nSyq3QOaXlCu/pFz7ci8c1kwmKTygeqvjmWBW2WykcpXt7PDGfdUAeAvCFwAAcKtujR8RaFXnOsx3OF06WfKTUHbKXhXcytzfH6967WRJuVwuuVfZ9tThPcxVYS3yp9seA22KCLIq6ifXsoUS1gA0U4QvAABQbxazyd1V8fLY4AvOr3A4dbKkvHKb4yl71bbHMnfzkOM/CXAFp8vlPKu5SF1rCg+w1rxGLdCqiCDPxiJRgTaF+PvIZCKsAWh8hC8AANBkfCxmRQdXtsOvi3KHUyfdwazy2rTq7Y/Hz1pZq155KyqtkMPpcjchqVNNZ632VTcZCfX3VYi/j4L9fBXs56OQqq/Bfr4KrRoP8fOVn6+Z4AagzghfAACg2fK1mBUT4qeYOrTrlypvgl0dzM5eQateUcurajRyvKrRSFFZhSqcLuUUlSmnqEzS+e+v9lM+ZlNlOPOvCme22kNbiN+Z8TNBrvI592ADWg/CFwAA8BpWH7PiQv0UF1q3sFZa7tDJkpohrfB0eVXHx3IVlVao8HTl16KychWerlBRaeV2yAqnSydLynWypLzeNfv5mqtC2Vmhzf/Mc/dX/8pw5xH2/HwVbPPhGjeghSB8AQCAVsvP16L4UH/Fh/pf1HEul0vFdoeKfhLOCksrQ1tt40VV49XhrdjukCSVljtVWl55b7b6MJmkIKtPjVDmEdrOs30y2M9H/r4Wtk8CTYDwBQAAcJFMJpOCbD4KsvkoPrR+56hwOHWqrOJMaKsKZbWFt7NX3M6eb3c45XJJRWUVKiqr0NGC0nrVUr198uwVtgttn/xpmGP7JHBhhC8AAAAD+FjMF3UPttqUljvOrKjVFtpKz2yf/Gl4q369IbdPBtl8FWSzKMjPR4HWynAaWPUIrhoLtFnc40F+VXPccyuPtflY6l0H0JwRvgAAAFooP1+L/Hwtde4e+VN12T55rtB2ru2Teacu/XP5WkyVoa0qlAX5VYU1m6UqwFWFOZvnuDvM2c58DfC1cE0cmg3CFwAAQCvVkNsnC09X6FRZ5aP4J1/PfO9QcdX3RVVfzx4/XV4Z5ModLuWXlCv/ElbiznxGuVfcqkOZZ0CrHA+2nR3mfH4y98xqna+F7ZWoP8IXAAAA6q0htk9Wq3A4VWx3nBXKaga32gJecZmj1nGnS3K55H5Nql9Tk7PZfMwe2ymDzt5GefY2S/f3llrCXOVX7hPX+hC+AAAA0Cz4WMwK9Tcr1N/3ks/lcrlUWu5UUVm5is8KbucMdaUVOmWvuRpXPdde4ZQklVU4VVZReWuCS2UxmxRotSjA6iOL2eT5MHk+9zGbZK76eq45Z4/7WEwymzyPcx9vMsliNstiludXk2SxmCuPr+W46vPV+p4/qafGcZbz12sxm1pFECV8AQAAwOuYTCb5Wy3yt1qk4Es/X7nDef7VuNKqMfuZ1biiqrFie81VOklyOF1VDVEqLr1AL2A2ST5ms8zmqq+mykB+ztBnMumm3vF6aHhno0uvM8IXAAAAcAG+Dbi90ul0qaT8THA7bXfI4XSpwumS0+VShaPqq9MlZ9W4o/rhcsnhdMrh1E++nnX8Wce5j3e55HBUH1/L4xzHVddzvuN+OnbmOGdVN02nnM6qr67z/Fxckt3hlByS5KzTz3JAUfgl//NoSoQvAAAAoAmZzWcancQaXUwTc7nOHdrcIfCC4fPMIzakfp0+jUL4AgAAANAkTFXXf7XWEEKvTAAAAABoAoQvAAAAAGgChC8AAAAAaAKELwAAAABoAoQvAAAAAGgChC8AAAAAaAKELwAAAABoAoQvAAAAAGgChC8AAAAAaAKELwAAAABoAoQvAAAAAGgChC8AAAAAaAKELwAAAABoAoQvAAAAAGgCPkYX0FK5XC5JUmFhocGVAAAAADBSdSaozgjnQviqp6KiIklSYmKiwZUAAAAAaA6KiooUGhp6ztdNrgvFM9TK6XTq6NGjCg4OlslkMrSWwsJCJSYm6tChQwoJCTG0FrQO/M6hKfH7hqbG7xyaEr9v3sHlcqmoqEgJCQkym899ZRcrX/VkNpvVtm1bo8vwEBISwv9o0aT4nUNT4vcNTY3fOTQlft9avvOteFWj4QYAAAAANAHCFwAAAAA0AcKXF7DZbJo7d65sNpvRpaCV4HcOTYnfNzQ1fufQlPh9a11ouAEAAAAATYCVLwAAAABoAoQvAAAAAGgChC8AAAAAaAKELwAAAABoAoQvL/DSSy8pKSlJfn5+Sk5O1ubNm40uCV4oLS1NAwYMUHBwsGJiYjR69GhlZmYaXRZaiQULFshkMmnGjBlGlwIvduTIEd19992KjIyUv7+/evXqpa+++srosuClHA6HnnjiCbVv317+/v7q2LGj5s+fL3rheTfCVwu3cuVKpaamau7cudq2bZv69OmjESNGKCcnx+jS4GU2bdqkqVOn6osvvtDatWtVXl6uG264QcXFxUaXBi+3ZcsWvfzyy+rdu7fRpcCLnTx5Utdcc418fX310Ucf6fvvv9cf/vAHhYeHG10avNQzzzyjJUuWaPHixdq5c6eeeeYZPfvss3rxxReNLg2NiFbzLVxycrIGDBigxYsXS5KcTqcSExM1ffp0zZo1y+Dq4M1yc3MVExOjTZs2afDgwUaXAy916tQpXXnllfrTn/6k3/3ud+rbt68WLVpkdFnwQrNmzdJnn32mTz75xOhS0Er8/Oc/V2xsrF599VX32JgxY+Tv76+3337bwMrQmFj5asHsdru2bt2qlJQU95jZbFZKSooyMjIMrAytQUFBgSQpIiLC4ErgzaZOnaqbbrrJ4//ngMbw/vvvq3///rrjjjsUExOjK664Qq+88orRZcGLXX311UpPT9cPP/wgSfrmm2/06aefatSoUQZXhsbkY3QBqL+8vDw5HA7FxsZ6jMfGxmrXrl0GVYXWwOl0asaMGbrmmmvUs2dPo8uBl1qxYoW2bdumLVu2GF0KWoF9+/ZpyZIlSk1N1WOPPaYtW7booYcektVq1aRJk4wuD15o1qxZKiwsVNeuXWWxWORwOPT73/9eEyZMMLo0NCLCF4CLNnXqVH333Xf69NNPjS4FXurQoUN6+OGHtXbtWvn5+RldDloBp9Op/v376+mnn5YkXXHFFfruu++0dOlSwhcaxTvvvKNly5Zp+fLl6tGjh7Zv364ZM2YoISGB3zkvRvhqwaKiomSxWJSdne0xnp2drbi4OIOqgrebNm2aVq9erY8//lht27Y1uhx4qa1btyonJ0dXXnmle8zhcOjjjz/W4sWLVVZWJovFYmCF8Dbx8fHq3r27x1i3bt30//1//59BFcHb/b//9/80a9YsjRs3TpLUq1cvHThwQGlpaYQvL8Y1Xy2Y1WpVv379lJ6e7h5zOp1KT0/XoEGDDKwM3sjlcmnatGl67733tH79erVv397okuDFhg8frm+//Vbbt293P/r3768JEyZo+/btBC80uGuuuabG7TN++OEHXXbZZQZVBG9XUlIis9nzT3GLxSKn02lQRWgKrHy1cKmpqZo0aZL69++vgQMHatGiRSouLtaUKVOMLg1eZurUqVq+fLn++c9/Kjg4WFlZWZKk0NBQ+fv7G1wdvE1wcHCN6wkDAwMVGRnJdYZoFDNnztTVV1+tp59+Wnfeeac2b96sP//5z/rzn/9sdGnwUjfffLN+//vfq127durRo4e+/vprPf/887rnnnuMLg2NiFbzXmDx4sV67rnnlJWVpb59++qFF15QcnKy0WXBy5hMplrHX3/9dU2ePLlpi0GrNHToUFrNo1GtXr1as2fP1u7du9W+fXulpqbqvvvuM7oseKmioiI98cQTeu+995STk6OEhASNHz9ec+bMkdVqNbo8NBLCFwAAAAA0Aa75AgAAAIAmQPgCAAAAgCZA+AIAAACAJkD4AgAAAIAmQPgCAAAAgCZA+AIAAACAJkD4AgAAAIAmQPgCAAAAgCZA+AIAoImZTCatWrXK6DIAAE2M8AUAaFUmT54sk8lU4zFy5EijSwMAeDkfowsAAKCpjRw5Uq+//rrHmM1mM6gaAEBrwcoXAKDVsdlsiouL83iEh4dLqtwSuGTJEo0aNUr+/v7q0KGD/v73v3sc/+2332rYsGHy9/dXZGSk7r//fp06dcpjzmuvvaYePXrIZrMpPj5e06ZN83g9Ly9Pt956qwICAtS5c2e9//77jfuhAQCGI3wBAPATTzzxhMaMGaNvvvlGEyZM0Lhx47Rz505JUnFxsUaMGKHw8HBt2bJF7777rtatW+cRrpYsWaKpU6fq/vvv17fffqv3339fnTp18niPJ598Unfeeaf++9//6sYbb9SECRN04sSJJv2cAICmZXK5XC6jiwAAoKlMnjxZb7/9tvz8/DzGH3vsMT322GMymUz69a9/rSVLlrhfu+qqq3TllVfqT3/6k1555RU9+uijOnTokAIDAyVJH374oW6++WYdPXpUsbGxatOmjaZMmaLf/e53tdZgMpn0+OOPa/78+ZIqA11QUJA++ugjrj0DAC/GNV8AgFbn+uuv9whXkhQREeH+ftCgQR6vDRo0SNu3b5ck7dy5U3369HEHL0m65ppr5HQ6lZmZKZPJpKNHj2r48OHnraF3797u7wMDAxUSEqKcnJz6fiQAQAtA+AIAtDqBgYE1tgE2FH9//zrN8/X19XhuMpnkdDoboyQAQDPBNV8AAPzEF198UeN5t27dJEndunXTN998o+LiYvfrn332mcxms7p06aLg4GAlJSUpPT29SWsGADR/rHwBAFqdsrIyZWVleYz5+PgoKipKkvTuu++qf//+uvbaa7Vs2TJt3rxZr776qiRpwoQJmjt3riZNmqR58+YpNzdX06dP1//8z/8oNjZWkjRv3jz9+te/VkxMjEaNGqWioiJ99tlnmj59etN+UABAs0L4AgC0OmvWrFF8fLzHWJcuXbRr1y5JlZ0IV6xYoQcffFDx8fH629/+pu7du0uSAgIC9O9//1sPP/ywBgwYoICAAI0ZM0bPP/+8+1yTJk1SaWmp/vjHP+qRRx5RVFSUbr/99qb7gACAZoluhwAAnMVkMum9997T6NGjjS4FAOBluOYLAAAAAJoA4QsAAAAAmgDXfAEAcBZ24wMAGgsrXwAAAADQBAhfAAAAANAECF8AAAAA0AQIXwAAAADQBAhfAAAAANAECF8AAAAA0AQIXwAAAADQBAhfAAAAANAE/n9Lyj796QMETAAAAABJRU5ErkJggg==\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Training accuracy: 99.69%\n",
            "Test accuracy: 98.79%\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "---\n",
        "## Q3\n",
        "---"
      ],
      "metadata": {
        "id": "YQ-iRhITj-Km"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# getting the dataset in\n",
        "!wget https://download.pytorch.org/tutorial/hymenoptera_data.zip"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NgcG2inYlH3j",
        "outputId": "6a2fdfbd-4522-4159-9fe3-65c3eae28607"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2024-11-14 15:58:16--  https://download.pytorch.org/tutorial/hymenoptera_data.zip\n",
            "Resolving download.pytorch.org (download.pytorch.org)... 18.65.25.112, 18.65.25.85, 18.65.25.45, ...\n",
            "Connecting to download.pytorch.org (download.pytorch.org)|18.65.25.112|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 47286322 (45M) [application/zip]\n",
            "Saving to: ‘hymenoptera_data.zip’\n",
            "\n",
            "hymenoptera_data.zi 100%[===================>]  45.10M  98.6MB/s    in 0.5s    \n",
            "\n",
            "2024-11-14 15:58:16 (98.6 MB/s) - ‘hymenoptera_data.zip’ saved [47286322/47286322]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# unzipping the dataset\n",
        "!unzip hymenoptera_data.zip -d /content/data"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IM0wKamplJuH",
        "outputId": "3c75dcc8-c9c1-43b6-d6bf-fa24c93b3274"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Archive:  hymenoptera_data.zip\n",
            "   creating: /content/data/hymenoptera_data/\n",
            "   creating: /content/data/hymenoptera_data/train/\n",
            "   creating: /content/data/hymenoptera_data/train/ants/\n",
            "  inflating: /content/data/hymenoptera_data/train/ants/0013035.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1030023514_aad5c608f9.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1095476100_3906d8afde.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1099452230_d1949d3250.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/116570827_e9c126745d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1225872729_6f0856588f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1262877379_64fcada201.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1269756697_0bce92cdab.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1286984635_5119e80de1.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/132478121_2a430adea2.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1360291657_dc248c5eea.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1368913450_e146e2fb6d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1473187633_63ccaacea6.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/148715752_302c84f5a4.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1489674356_09d48dde0a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/149244013_c529578289.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/150801003_3390b73135.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/150801171_cd86f17ed8.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/154124431_65460430f2.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/162603798_40b51f1654.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1660097129_384bf54490.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/167890289_dd5ba923f3.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1693954099_46d4c20605.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/175998972.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/178538489_bec7649292.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1804095607_0341701e1c.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1808777855_2a895621d7.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/188552436_605cc9b36b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1917341202_d00a7f9af5.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/1924473702_daa9aacdbe.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/196057951_63bf063b92.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/196757565_326437f5fe.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/201558278_fe4caecc76.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/201790779_527f4c0168.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/2019439677_2db655d361.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/207947948_3ab29d7207.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/20935278_9190345f6b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/224655713_3956f7d39a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/2265824718_2c96f485da.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/2265825502_fff99cfd2d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/226951206_d6bf946504.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/2278278459_6b99605e50.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/2288450226_a6e96e8fdf.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/2288481644_83ff7e4572.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/2292213964_ca51ce4bef.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/24335309_c5ea483bb8.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/245647475_9523dfd13e.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/255434217_1b2b3fe0a4.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/258217966_d9d90d18d3.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/275429470_b2d7d9290b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/28847243_e79fe052cd.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/318052216_84dff3f98a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/334167043_cbd1adaeb9.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/339670531_94b75ae47a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/342438950_a3da61deab.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/36439863_0bec9f554f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/374435068_7eee412ec4.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/382971067_0bfd33afe0.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/384191229_5779cf591b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/386190770_672743c9a7.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/392382602_1b7bed32fa.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/403746349_71384f5b58.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/408393566_b5b694119b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/424119020_6d57481dab.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/424873399_47658a91fb.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/450057712_771b3bfc91.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/45472593_bfd624f8dc.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/459694881_ac657d3187.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/460372577_f2f6a8c9fc.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/460874319_0a45ab4d05.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/466430434_4000737de9.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/470127037_513711fd21.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/474806473_ca6caab245.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/475961153_b8c13fd405.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/484293231_e53cfc0c89.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/49375974_e28ba6f17e.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/506249802_207cd979b4.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/506249836_717b73f540.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/512164029_c0a66b8498.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/512863248_43c8ce579b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/518773929_734dbc5ff4.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/522163566_fec115ca66.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/522415432_2218f34bf8.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/531979952_bde12b3bc0.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/533848102_70a85ad6dd.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/535522953_308353a07c.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/540889389_48bb588b21.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/541630764_dbd285d63c.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/543417860_b14237f569.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/560966032_988f4d7bc4.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/5650366_e22b7e1065.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/6240329_72c01e663e.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/6240338_93729615ec.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/649026570_e58656104b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/662541407_ff8db781e7.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/67270775_e9fdf77e9d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/6743948_2b8c096dda.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/684133190_35b62c0c1d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/69639610_95e0de17aa.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/707895295_009cf23188.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/7759525_1363d24e88.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/795000156_a9900a4a71.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/822537660_caf4ba5514.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/82852639_52b7f7f5e3.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/841049277_b28e58ad05.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/886401651_f878e888cd.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/892108839_f1aad4ca46.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/938946700_ca1c669085.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/957233405_25c1d1187b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/9715481_b3cb4114ff.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/998118368_6ac1d91f81.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/ant photos.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/Ant_1.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/army-ants-red-picture.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/formica.jpeg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/hormiga_co_por.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/imageNotFound.gif  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/kurokusa.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/MehdiabadiAnt2_600.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/Nepenthes_rafflesiana_ant.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/swiss-army-ant.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/termite-vs-ant.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/trap-jaw-ant-insect-bg.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/ants/VietnameseAntMimicSpider.jpg  \n",
            "   creating: /content/data/hymenoptera_data/train/bees/\n",
            "  inflating: /content/data/hymenoptera_data/train/bees/1092977343_cb42b38d62.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/1093831624_fb5fbe2308.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/1097045929_1753d1c765.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/1232245714_f862fbe385.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/129236073_0985e91c7d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/1295655112_7813f37d21.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/132511197_0b86ad0fff.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/132826773_dbbcb117b9.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/150013791_969d9a968b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/1508176360_2972117c9d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/154600396_53e1252e52.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/16838648_415acd9e3f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/1691282715_0addfdf5e8.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/17209602_fe5a5a746f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/174142798_e5ad6d76e0.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/1799726602_8580867f71.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/1807583459_4fe92b3133.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/196430254_46bd129ae7.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/196658222_3fffd79c67.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/198508668_97d818b6c4.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2031225713_50ed499635.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2037437624_2d7bce461f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2053200300_8911ef438a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/205835650_e6f2614bee.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/208702903_42fb4d9748.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/21399619_3e61e5bb6f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2227611847_ec72d40403.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2321139806_d73d899e66.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2330918208_8074770c20.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2345177635_caf07159b3.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2358061370_9daabbd9ac.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2364597044_3c3e3fc391.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2384149906_2cd8b0b699.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2397446847_04ef3cd3e1.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2405441001_b06c36fa72.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2445215254_51698ff797.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2452236943_255bfd9e58.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2467959963_a7831e9ff0.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2470492904_837e97800d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2477324698_3d4b1b1cab.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2477349551_e75c97cf4d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2486729079_62df0920be.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2486746709_c43cec0e42.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2493379287_4100e1dacc.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2495722465_879acf9d85.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2528444139_fa728b0f5b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2538361678_9da84b77e3.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2551813042_8a070aeb2b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2580598377_a4caecdb54.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2601176055_8464e6aa71.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2610833167_79bf0bcae5.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2610838525_fe8e3cae47.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2617161745_fa3ebe85b4.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2625499656_e3415e374d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2634617358_f32fd16bea.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2638074627_6b3ae746a0.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2645107662_b73a8595cc.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2651621464_a2fa8722eb.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2652877533_a564830cbf.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/266644509_d30bb16a1b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2683605182_9d2a0c66cf.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2704348794_eb5d5178c2.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2707440199_cd170bd512.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2710368626_cb42882dc8.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2722592222_258d473e17.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2728759455_ce9bb8cd7a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2756397428_1d82a08807.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2765347790_da6cf6cb40.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2781170484_5d61835d63.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/279113587_b4843db199.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2792000093_e8ae0718cf.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2801728106_833798c909.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2822388965_f6dca2a275.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2861002136_52c7c6f708.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2908916142_a7ac8b57a8.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/29494643_e3410f0d37.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2959730355_416a18c63c.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/2962405283_22718d9617.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/3006264892_30e9cced70.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/3030189811_01d095b793.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/3030772428_8578335616.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/3044402684_3853071a87.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/3074585407_9854eb3153.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/3079610310_ac2d0ae7bc.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/3090975720_71f12e6de4.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/3100226504_c0d4f1e3f1.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/342758693_c56b89b6b6.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/354167719_22dca13752.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/359928878_b3b418c728.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/365759866_b15700c59b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/36900412_92b81831ad.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/39672681_1302d204d1.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/39747887_42df2855ee.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/421515404_e87569fd8b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/444532809_9e931e2279.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/446296270_d9e8b93ecf.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/452462677_7be43af8ff.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/452462695_40a4e5b559.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/457457145_5f86eb7e9c.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/465133211_80e0c27f60.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/469333327_358ba8fe8a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/472288710_2abee16fa0.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/473618094_8ffdcab215.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/476347960_52edd72b06.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/478701318_bbd5e557b8.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/507288830_f46e8d4cb2.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/509247772_2db2d01374.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/513545352_fd3e7c7c5d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/522104315_5d3cb2758e.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/537309131_532bfa59ea.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/586041248_3032e277a9.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/760526046_547e8b381f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/760568592_45a52c847f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/774440991_63a4aa0cbe.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/85112639_6e860b0469.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/873076652_eb098dab2d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/90179376_abc234e5f4.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/92663402_37f379e57a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/95238259_98470c5b10.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/969455125_58c797ef17.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/train/bees/98391118_bdb1e80cce.jpg  \n",
            "   creating: /content/data/hymenoptera_data/val/\n",
            "   creating: /content/data/hymenoptera_data/val/ants/\n",
            "  inflating: /content/data/hymenoptera_data/val/ants/10308379_1b6c72e180.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/1053149811_f62a3410d3.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/1073564163_225a64f170.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/1119630822_cd325ea21a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/1124525276_816a07c17f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/11381045_b352a47d8c.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/119785936_dd428e40c3.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/1247887232_edcb61246c.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/1262751255_c56c042b7b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/1337725712_2eb53cd742.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/1358854066_5ad8015f7f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/1440002809_b268d9a66a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/147542264_79506478c2.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/152286280_411648ec27.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/153320619_2aeb5fa0ee.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/153783656_85f9c3ac70.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/157401988_d0564a9d02.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/159515240_d5981e20d1.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/161076144_124db762d6.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/161292361_c16e0bf57a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/170652283_ecdaff5d1a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/17081114_79b9a27724.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/172772109_d0a8e15fb0.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/1743840368_b5ccda82b7.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/181942028_961261ef48.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/183260961_64ab754c97.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/2039585088_c6f47c592e.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/205398178_c395c5e460.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/208072188_f293096296.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/209615353_eeb38ba204.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/2104709400_8831b4fc6f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/212100470_b485e7b7b9.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/2127908701_d49dc83c97.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/2191997003_379df31291.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/2211974567_ee4606b493.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/2219621907_47bc7cc6b0.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/2238242353_52c82441df.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/2255445811_dabcdf7258.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/239161491_86ac23b0a3.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/263615709_cfb28f6b8e.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/308196310_1db5ffa01b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/319494379_648fb5a1c6.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/35558229_1fa4608a7a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/412436937_4c2378efc2.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/436944325_d4925a38c7.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/445356866_6cb3289067.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/459442412_412fecf3fe.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/470127071_8b8ee2bd74.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/477437164_bc3e6e594a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/488272201_c5aa281348.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/502717153_3e4865621a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/518746016_bcc28f8b5b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/540543309_ddbb193ee5.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/562589509_7e55469b97.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/57264437_a19006872f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/573151833_ebbc274b77.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/649407494_9b6bc4949f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/751649788_78dd7d16ce.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/768870506_8f115d3d37.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/800px-Meat_eater_ant_qeen_excavating_hole.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/8124241_36b290d372.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/8398478_50ef10c47a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/854534770_31f6156383.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/892676922_4ab37dce07.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/94999827_36895faade.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/Ant-1818.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/ants-devouring-remains-of-large-dead-insect-on-red-tile-in-Stellenbosch-South-Africa-closeup-1-DHD.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/desert_ant.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/F.pergan.28(f).jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/ants/Hormiga.jpg  \n",
            "   creating: /content/data/hymenoptera_data/val/bees/\n",
            "  inflating: /content/data/hymenoptera_data/val/bees/1032546534_06907fe3b3.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/10870992_eebeeb3a12.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/1181173278_23c36fac71.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/1297972485_33266a18d9.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/1328423762_f7a88a8451.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/1355974687_1341c1face.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/144098310_a4176fd54d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/1486120850_490388f84b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/149973093_da3c446268.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/151594775_ee7dc17b60.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/151603988_2c6f7d14c7.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/1519368889_4270261ee3.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/152789693_220b003452.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/177677657_a38c97e572.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/1799729694_0c40101071.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/181171681_c5a1a82ded.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/187130242_4593a4c610.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/203868383_0fcbb48278.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2060668999_e11edb10d0.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2086294791_6f3789d8a6.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2103637821_8d26ee6b90.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2104135106_a65eede1de.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/215512424_687e1e0821.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2173503984_9c6aaaa7e2.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/220376539_20567395d8.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/224841383_d050f5f510.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2321144482_f3785ba7b2.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/238161922_55fa9a76ae.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2407809945_fb525ef54d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2415414155_1916f03b42.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2438480600_40a1249879.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2444778727_4b781ac424.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2457841282_7867f16639.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2470492902_3572c90f75.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2478216347_535c8fe6d7.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2501530886_e20952b97d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2506114833_90a41c5267.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2509402554_31821cb0b6.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2525379273_dcb26a516d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/26589803_5ba7000313.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2668391343_45e272cd07.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2670536155_c170f49cd0.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2685605303_9eed79d59d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2702408468_d9ed795f4f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2709775832_85b4b50a57.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2717418782_bd83307d9f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/272986700_d4d4bf8c4b.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2741763055_9a7bb00802.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2745389517_250a397f31.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2751836205_6f7b5eff30.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2782079948_8d4e94a826.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2809496124_5f25b5946a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2815838190_0a9889d995.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2841437312_789699c740.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/2883093452_7e3a1eb53f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/290082189_f66cb80bfc.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/296565463_d07a7bed96.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/3077452620_548c79fda0.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/348291597_ee836fbb1a.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/350436573_41f4ecb6c8.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/353266603_d3eac7e9a0.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/372228424_16da1f8884.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/400262091_701c00031c.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/416144384_961c326481.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/44105569_16720a960c.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/456097971_860949c4fc.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/464594019_1b24a28bb1.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/485743562_d8cc6b8f73.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/540976476_844950623f.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/54736755_c057723f64.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/57459255_752774f1b2.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/576452297_897023f002.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/586474709_ae436da045.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/590318879_68cf112861.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/59798110_2b6a3c8031.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/603709866_a97c7cfc72.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/603711658_4c8cd2201e.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/65038344_52a45d090d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/6a00d8341c630a53ef00e553d0beb18834-800wi.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/72100438_73de9f17af.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/759745145_e8bc776ec8.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/936182217_c4caa5222d.jpg  \n",
            "  inflating: /content/data/hymenoptera_data/val/bees/abeja.jpg  \n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import torch.optim as optim\n",
        "import torchvision\n",
        "import torchvision.transforms as transforms\n",
        "import torchvision.datasets as datasets\n",
        "from torchvision import models\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Check if CUDA is available\n",
        "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
        "print(f\"Using device: {device}\")\n",
        "\n",
        "# 1. Data loading and transformation for Hymenoptera dataset\n",
        "data_dir = './data/hymenoptera_data'\n",
        "input_size = 224\n",
        "batch_size = 8\n",
        "\n",
        "data_transforms = {\n",
        "    'train': transforms.Compose([\n",
        "        transforms.RandomResizedCrop(input_size),\n",
        "        transforms.RandomHorizontalFlip(),\n",
        "        transforms.ToTensor(),\n",
        "        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])\n",
        "    ]),\n",
        "    'val': transforms.Compose([\n",
        "        transforms.Resize(input_size),\n",
        "        transforms.CenterCrop(input_size),\n",
        "        transforms.ToTensor(),\n",
        "        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])\n",
        "    ]),\n",
        "}\n",
        "\n",
        "image_datasets = {x: datasets.ImageFolder(root=f\"{data_dir}/{x}\", transform=data_transforms[x]) for x in ['train', 'val']}\n",
        "dataloaders = {x: torch.utils.data.DataLoader(image_datasets[x], batch_size=batch_size, shuffle=True, num_workers=4) for x in ['train', 'val']}\n",
        "dataset_sizes = {x: len(image_datasets[x]) for x in ['train', 'val']}\n",
        "class_names = image_datasets['train'].classes\n",
        "\n",
        "# 2. Load the pre-trained ResNet-18 model\n",
        "model_ft = models.resnet18(pretrained=True)\n",
        "\n",
        "# Fine-tuning: Set requires_grad = True for all parameters to train the whole model\n",
        "for param in model_ft.parameters():\n",
        "    param.requires_grad = True\n",
        "\n",
        "# Modify the final layer to match the number of classes in the Hymenoptera dataset (2 classes)\n",
        "num_ftrs = model_ft.fc.in_features\n",
        "model_ft.fc = nn.Linear(num_ftrs, len(class_names))\n",
        "\n",
        "# Move the model to the selected device\n",
        "model_ft = model_ft.to(device)\n",
        "\n",
        "# Define loss function and optimizer for fine-tuning\n",
        "criterion = nn.CrossEntropyLoss()\n",
        "optimizer_ft = optim.SGD(model_ft.parameters(), lr=0.001, momentum=0.9)\n",
        "exp_lr_scheduler = optim.lr_scheduler.StepLR(optimizer_ft, step_size=7, gamma=0.1)\n",
        "\n",
        "# 3. Training Function\n",
        "def train_model(model, criterion, optimizer, scheduler, num_epochs=10):\n",
        "    train_losses = []\n",
        "    val_accuracies = []\n",
        "\n",
        "    for epoch in range(num_epochs):\n",
        "        print(f\"Epoch {epoch + 1}/{num_epochs}\")\n",
        "        print(\"-\" * 10)\n",
        "\n",
        "        # Each epoch has a training and validation phase\n",
        "        for phase in ['train', 'val']:\n",
        "            if phase == 'train':\n",
        "                model.train()  # Set model to training mode\n",
        "            else:\n",
        "                model.eval()   # Set model to evaluate mode\n",
        "\n",
        "            running_loss = 0.0\n",
        "            running_corrects = 0\n",
        "\n",
        "            # Iterate over data\n",
        "            for inputs, labels in dataloaders[phase]:\n",
        "                inputs, labels = inputs.to(device), labels.to(device)\n",
        "\n",
        "                # Zero the parameter gradients\n",
        "                optimizer.zero_grad()\n",
        "\n",
        "                # Forward pass\n",
        "                with torch.set_grad_enabled(phase == 'train'):\n",
        "                    outputs = model(inputs)\n",
        "                    _, preds = torch.max(outputs, 1)\n",
        "                    loss = criterion(outputs, labels)\n",
        "\n",
        "                    # Backward pass and optimize in training phase\n",
        "                    if phase == 'train':\n",
        "                        loss.backward()\n",
        "                        optimizer.step()\n",
        "\n",
        "                # Statistics\n",
        "                running_loss += loss.item() * inputs.size(0)\n",
        "                running_corrects += torch.sum(preds == labels.data)\n",
        "\n",
        "            epoch_loss = running_loss / dataset_sizes[phase]\n",
        "            epoch_acc = running_corrects.double() / dataset_sizes[phase]\n",
        "\n",
        "            if phase == 'train':\n",
        "                scheduler.step()\n",
        "                train_losses.append(epoch_loss)\n",
        "            else:\n",
        "                val_accuracies.append(epoch_acc.item())\n",
        "\n",
        "            print(f\"{phase} Loss: {epoch_loss:.4f} Acc: {epoch_acc:.4f}\")\n",
        "\n",
        "    return model, train_losses, val_accuracies\n",
        "\n",
        "# Train and evaluate the model\n",
        "model_ft, train_losses, val_accuracies = train_model(model_ft, criterion, optimizer_ft, exp_lr_scheduler, num_epochs=10)\n",
        "\n",
        "# Plotting Training Loss and Validation Accuracy\n",
        "plt.figure(figsize=(10, 5))\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.plot(train_losses, label=\"Training Loss\")\n",
        "plt.xlabel(\"Epoch\")\n",
        "plt.ylabel(\"Loss\")\n",
        "plt.title(\"Training Loss\")\n",
        "plt.legend()\n",
        "\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.plot(val_accuracies, label=\"Validation Accuracy\")\n",
        "plt.xlabel(\"Epoch\")\n",
        "plt.ylabel(\"Accuracy\")\n",
        "plt.title(\"Validation Accuracy\")\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "VFseKXfHkAUC",
        "outputId": "3af77540-665a-40a2-abaa-99a01b070fa2"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Using device: cuda\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/torch/utils/data/dataloader.py:617: UserWarning: This DataLoader will create 4 worker processes in total. Our suggested max number of worker in current system is 2, which is smaller than what this DataLoader is going to create. Please be aware that excessive worker creation might get DataLoader running slow or even freeze, lower the worker number to avoid potential slowness/freeze if necessary.\n",
            "  warnings.warn(\n",
            "/usr/local/lib/python3.10/dist-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",
            "/usr/local/lib/python3.10/dist-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=ResNet18_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet18_Weights.DEFAULT` to get the most up-to-date weights.\n",
            "  warnings.warn(msg)\n",
            "Downloading: \"https://download.pytorch.org/models/resnet18-f37072fd.pth\" to /root/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth\n",
            "100%|██████████| 44.7M/44.7M [00:00<00:00, 126MB/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "----------\n",
            "train Loss: 0.5309 Acc: 0.7008\n",
            "val Loss: 0.2555 Acc: 0.9281\n",
            "Epoch 2/10\n",
            "----------\n",
            "train Loss: 0.4023 Acc: 0.8197\n",
            "val Loss: 0.1897 Acc: 0.9412\n",
            "Epoch 3/10\n",
            "----------\n",
            "train Loss: 0.3137 Acc: 0.8607\n",
            "val Loss: 0.3525 Acc: 0.8693\n",
            "Epoch 4/10\n",
            "----------\n",
            "train Loss: 0.3146 Acc: 0.8811\n",
            "val Loss: 0.1886 Acc: 0.9346\n",
            "Epoch 5/10\n",
            "----------\n",
            "train Loss: 0.2931 Acc: 0.8811\n",
            "val Loss: 0.2036 Acc: 0.9346\n",
            "Epoch 6/10\n",
            "----------\n",
            "train Loss: 0.2414 Acc: 0.9016\n",
            "val Loss: 0.2011 Acc: 0.9412\n",
            "Epoch 7/10\n",
            "----------\n",
            "train Loss: 0.3699 Acc: 0.8361\n",
            "val Loss: 0.2844 Acc: 0.8889\n",
            "Epoch 8/10\n",
            "----------\n",
            "train Loss: 0.1932 Acc: 0.9221\n",
            "val Loss: 0.1920 Acc: 0.9346\n",
            "Epoch 9/10\n",
            "----------\n",
            "train Loss: 0.1513 Acc: 0.9467\n",
            "val Loss: 0.1795 Acc: 0.9412\n",
            "Epoch 10/10\n",
            "----------\n",
            "train Loss: 0.1416 Acc: 0.9426\n",
            "val Loss: 0.1741 Acc: 0.9412\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x500 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "In order to produce two classes that correspond to the Hymenoptera dataset, I initialize the pre-trained ResNet18 model and alter its last fully connected layer. With this configuration, the model may change to precisely identify ants and bees while maintaining the traits it has learnt from ImageNet.\n",
        "\n",
        "I updated all parameters using an SGD optimizer with a learning rate of 0.001 and momentum of 0.9, using cross-entropy loss as the criteria. To improve the model's fine-tuning over time, a learning rate scheduler decays the learning rate by 0.1 every seven epochs.\n",
        "\n",
        "Finally got a validation accuracy around 94.12%."
      ],
      "metadata": {
        "id": "In95u3QBXv2e"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "class_names"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WHeG7uc8Z2Zv",
        "outputId": "ea0e7c1d-b725-4abc-c280-7b4b18b68527"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['ants', 'bees']"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "# 4. Feature Extraction: Freeze all layers except the final fully connected layer\n",
        "model_ft = models.resnet18(pretrained=True)\n",
        "for param in model_ft.parameters():\n",
        "    param.requires_grad = False  # Freeze parameters\n",
        "\n",
        "# Replace the final layer\n",
        "model_ft.fc = nn.Linear(num_ftrs, len(class_names))\n",
        "model_ft = model_ft.to(device)\n",
        "\n",
        "# Define optimizer for feature extraction (only updates final layer)\n",
        "optimizer_fe = optim.SGD(model_ft.fc.parameters(), lr=0.001, momentum=0.9)\n",
        "\n",
        "# Train the model again using feature extraction\n",
        "model_fe, _, val_accuracies_fe = train_model(model_ft, criterion, optimizer_fe, exp_lr_scheduler, num_epochs=10)\n",
        "\n",
        "# Plot Validation Accuracy for Feature Extraction\n",
        "plt.plot(val_accuracies_fe, label=\"Validation Accuracy (Feature Extraction)\")\n",
        "plt.xlabel(\"Epoch\")\n",
        "plt.ylabel(\"Accuracy\")\n",
        "plt.title(\"Validation Accuracy (Feature Extraction)\")\n",
        "plt.legend()\n",
        "plt.show()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "McebJh9kXt74",
        "outputId": "42841ca0-d3c5-4acc-9729-ab879d7f1627"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "----------\n",
            "train Loss: 0.5517 Acc: 0.7090\n",
            "val Loss: 0.3528 Acc: 0.9020\n",
            "Epoch 2/10\n",
            "----------\n",
            "train Loss: 0.3489 Acc: 0.8443\n",
            "val Loss: 0.2225 Acc: 0.9281\n",
            "Epoch 3/10\n",
            "----------\n",
            "train Loss: 0.3302 Acc: 0.8402\n",
            "val Loss: 0.2350 Acc: 0.9150\n",
            "Epoch 4/10\n",
            "----------\n",
            "train Loss: 0.2884 Acc: 0.8730\n",
            "val Loss: 0.2838 Acc: 0.8824\n",
            "Epoch 5/10\n",
            "----------\n",
            "train Loss: 0.3804 Acc: 0.8156\n",
            "val Loss: 0.1875 Acc: 0.9477\n",
            "Epoch 6/10\n",
            "----------\n",
            "train Loss: 0.3535 Acc: 0.8484\n",
            "val Loss: 0.1911 Acc: 0.9542\n",
            "Epoch 7/10\n",
            "----------\n",
            "train Loss: 0.3284 Acc: 0.8320\n",
            "val Loss: 0.2367 Acc: 0.9281\n",
            "Epoch 8/10\n",
            "----------\n",
            "train Loss: 0.2331 Acc: 0.9016\n",
            "val Loss: 0.2241 Acc: 0.9281\n",
            "Epoch 9/10\n",
            "----------\n",
            "train Loss: 0.3966 Acc: 0.8402\n",
            "val Loss: 0.2775 Acc: 0.8824\n",
            "Epoch 10/10\n",
            "----------\n",
            "train Loss: 0.3130 Acc: 0.8770\n",
            "val Loss: 0.2017 Acc: 0.9412\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "To use the pre-trained ResNet18 model solely as a feature extractor, I load it and freeze all layers by setting requires_grad = False for each parameter. This means only the last layer—which I modified to output two classes—will be trained. With this setup, the model adapts the final layer to classify the Hymenoptera dataset while leveraging the general features learned from the ImageNet dataset.\n",
        "\n",
        "I achieved a validation accuracy of approximately 95%.\n",
        "\n",
        "Since the pre-trained layers capture generic visual features from a large dataset (ImageNet), using ResNet18 as a feature extractor improved accuracy. Freezing these layers helps prevent overfitting to the smaller Hymenoptera dataset by preserving the robust, transferable patterns learned from ImageNet. This allows the final layer to focus on distinguishing between ants and bees, resulting in more precise and efficient classification."
      ],
      "metadata": {
        "id": "RPz8L9r1ZxlM"
      }
    }
  ]
}