diff --git "a/Week10_\354\225\210\354\204\234\354\227\260_preview.pdf" "b/Week10_\354\225\210\354\204\234\354\227\260_preview.pdf"
new file mode 100644
index 0000000..7939539
Binary files /dev/null and "b/Week10_\354\225\210\354\204\234\354\227\260_preview.pdf" differ
diff --git "a/Week16_\354\225\210\354\204\234\354\227\260_preview.pdf" "b/Week16_\354\225\210\354\204\234\354\227\260_preview.pdf"
new file mode 100644
index 0000000..8959fa8
Binary files /dev/null and "b/Week16_\354\225\210\354\204\234\354\227\260_preview.pdf" differ
diff --git a/Week17_preview.md b/Week17_preview.md
new file mode 100644
index 0000000..08435b0
--- /dev/null
+++ b/Week17_preview.md
@@ -0,0 +1,20 @@
+# GAN, Image-to-Image Translation with Conditional Adversarial Networks 
+
+
+- image to image translation에서 general-purpose solution 제시
+- Conditional GAN을 이용한 한 유형의 이미즐 다른 유형의 이미지로 변환하는 framework 제시하여 image to image translation 작업에
+  처음 적용해서 좋은 결과를 얻음 
+
+
+
+# Bringing Old Photos Back to Life 
+
+- 이미지  복원에서 화질을 복원하는 연구이고,Triplet domain translation을 제시하여 기존 이미지 복원 연구에서 가지고 있던 한계점인
+  Generalization issue와 mixed degradation issue 에 대한 해결방안을 제시함.
+- scratch가 이미지 전제적으로 일관적으로 복원될수 있게됨   
+
+# Denoising Diffusion Probabilistic Models (DDPM) 
+
+- nonequilibrium thermodynamics로부터 고안된 잠재 변수 모델 중 하나인 diffusion probabilistic models를 제세.
+- forward diffusen process: noise를 점점 증가 시켜가면서 학습 데이터를 특정한(Guassian) noise distribution으로 변환
+- reverse generative proccess: noise distribution으로부터 학습 데이터를 복원(denoising)하는 학습 단위과정을 Markov chain으로 표현
diff --git a/Week18_preview.md b/Week18_preview.md
new file mode 100644
index 0000000..375650f
--- /dev/null
+++ b/Week18_preview.md
@@ -0,0 +1,19 @@
+# Learning Deep Features for Discriminative Localization
+
+- CNN의 결과 설명
+- Class Activation Maps(CAM): Convolution-Global average pooling -Softmax 구조
+- Object loaclization도 함 
+
+# MediaPipe Hands: On-device Real-time Hand Tracking
+
+- 하드웨어 필요x, 2개 이상 손 탐지 및 일부 가려져도 탐지가능, 모바일 실시간 가능
+- 1) Bounding Box를 찾는 palm detector
+- 2) Bounding Box 당 21개 keypoints 감지
+- 성능이 좋음
+
+# Mask R-CNN 
+
+- 이미지 내 instance segmentation mask 
+- Mask R-CNN=Faster R-cNN + mask branch
+- BackBone: ResNet & REsNeXt , Head: Faster R-CNN head 
+- COCO 2016 challenge 1st
diff --git a/Week8_preview_Training_a_Classifier.ipynb b/Week8_preview_Training_a_Classifier.ipynb
new file mode 100644
index 0000000..3e49a03
--- /dev/null
+++ b/Week8_preview_Training_a_Classifier.ipynb
@@ -0,0 +1,1001 @@
+{
+  "nbformat": 4,
+  "nbformat_minor": 0,
+  "metadata": {
+    "colab": {
+      "name": "Training a Classifier.ipynb",
+      "provenance": [],
+      "collapsed_sections": []
+    },
+    "kernelspec": {
+      "name": "python3",
+      "display_name": "Python 3"
+    },
+    "language_info": {
+      "name": "python"
+    },
+    "widgets": {
+      "application/vnd.jupyter.widget-state+json": {
+        "ce0f512204bc4540a56fa1363e85e257": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HBoxModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HBoxModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HBoxView",
+            "box_style": "",
+            "children": [
+              "IPY_MODEL_e7ff134b9a404fd284a0f614d5c5903b",
+              "IPY_MODEL_d75f11c2a2cc47d7af05c500dae96355",
+              "IPY_MODEL_869d9806a41f4bac938f2053a66683d7"
+            ],
+            "layout": "IPY_MODEL_12bdb2122e2c4158884947e52c9e5b9d"
+          }
+        },
+        "e7ff134b9a404fd284a0f614d5c5903b": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_ca04de8c81944319ab6f92b7074f08b1",
+            "placeholder": "​",
+            "style": "IPY_MODEL_af05e518ab24430d9095be0d5af78e17",
+            "value": ""
+          }
+        },
+        "d75f11c2a2cc47d7af05c500dae96355": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "FloatProgressModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "FloatProgressModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "ProgressView",
+            "bar_style": "success",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_a664d31fd7bb4fd2b9b030fb4295eb62",
+            "max": 170498071,
+            "min": 0,
+            "orientation": "horizontal",
+            "style": "IPY_MODEL_25e0eeb20dd4459bb7941ec4f13f8ae7",
+            "value": 170498071
+          }
+        },
+        "869d9806a41f4bac938f2053a66683d7": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_7038855e026244b9b894b498a076069f",
+            "placeholder": "​",
+            "style": "IPY_MODEL_2fa799f1c9484f098944c1c1fc9925f5",
+            "value": " 170499072/? [00:08<00:00, 30735702.61it/s]"
+          }
+        },
+        "12bdb2122e2c4158884947e52c9e5b9d": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": null,
+            "bottom": null,
+            "display": null,
+            "flex": null,
+            "flex_flow": null,
+            "grid_area": null,
+            "grid_auto_columns": null,
+            "grid_auto_flow": null,
+            "grid_auto_rows": null,
+            "grid_column": null,
+            "grid_gap": null,
+            "grid_row": null,
+            "grid_template_areas": null,
+            "grid_template_columns": null,
+            "grid_template_rows": null,
+            "height": null,
+            "justify_content": null,
+            "justify_items": null,
+            "left": null,
+            "margin": null,
+            "max_height": null,
+            "max_width": null,
+            "min_height": null,
+            "min_width": null,
+            "object_fit": null,
+            "object_position": null,
+            "order": null,
+            "overflow": null,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "ca04de8c81944319ab6f92b7074f08b1": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": null,
+            "bottom": null,
+            "display": null,
+            "flex": null,
+            "flex_flow": null,
+            "grid_area": null,
+            "grid_auto_columns": null,
+            "grid_auto_flow": null,
+            "grid_auto_rows": null,
+            "grid_column": null,
+            "grid_gap": null,
+            "grid_row": null,
+            "grid_template_areas": null,
+            "grid_template_columns": null,
+            "grid_template_rows": null,
+            "height": null,
+            "justify_content": null,
+            "justify_items": null,
+            "left": null,
+            "margin": null,
+            "max_height": null,
+            "max_width": null,
+            "min_height": null,
+            "min_width": null,
+            "object_fit": null,
+            "object_position": null,
+            "order": null,
+            "overflow": null,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "af05e518ab24430d9095be0d5af78e17": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "a664d31fd7bb4fd2b9b030fb4295eb62": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": null,
+            "bottom": null,
+            "display": null,
+            "flex": null,
+            "flex_flow": null,
+            "grid_area": null,
+            "grid_auto_columns": null,
+            "grid_auto_flow": null,
+            "grid_auto_rows": null,
+            "grid_column": null,
+            "grid_gap": null,
+            "grid_row": null,
+            "grid_template_areas": null,
+            "grid_template_columns": null,
+            "grid_template_rows": null,
+            "height": null,
+            "justify_content": null,
+            "justify_items": null,
+            "left": null,
+            "margin": null,
+            "max_height": null,
+            "max_width": null,
+            "min_height": null,
+            "min_width": null,
+            "object_fit": null,
+            "object_position": null,
+            "order": null,
+            "overflow": null,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "25e0eeb20dd4459bb7941ec4f13f8ae7": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "ProgressStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "ProgressStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "bar_color": null,
+            "description_width": ""
+          }
+        },
+        "7038855e026244b9b894b498a076069f": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": null,
+            "bottom": null,
+            "display": null,
+            "flex": null,
+            "flex_flow": null,
+            "grid_area": null,
+            "grid_auto_columns": null,
+            "grid_auto_flow": null,
+            "grid_auto_rows": null,
+            "grid_column": null,
+            "grid_gap": null,
+            "grid_row": null,
+            "grid_template_areas": null,
+            "grid_template_columns": null,
+            "grid_template_rows": null,
+            "height": null,
+            "justify_content": null,
+            "justify_items": null,
+            "left": null,
+            "margin": null,
+            "max_height": null,
+            "max_width": null,
+            "min_height": null,
+            "min_width": null,
+            "object_fit": null,
+            "object_position": null,
+            "order": null,
+            "overflow": null,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "2fa799f1c9484f098944c1c1fc9925f5": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        }
+      }
+    }
+  },
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "source": [
+        "### Image Classifier 학습시키기\n",
+        "\n",
+        "**학습단계**\n",
+        "1. torchvision으로 CIFAR10의 train & test 데이터셋을 load하고 normalize하기\n",
+        "2. Convoultional Neural Network 정의하기\n",
+        "3. Loss function 정의하기\n",
+        "4. 신경망 training data로 학습시키기\n",
+        "5. test data로 신경망 검증하기 \n"
+      ],
+      "metadata": {
+        "id": "qygJPRr1TjBu"
+      }
+    },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "## 1. torchvision으로 CIFAR10의 train & test 데이터셋을 load하고 normalize하기"
+      ],
+      "metadata": {
+        "id": "lYhWhGNSUZrt"
+      }
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "import torch\n",
+        "import torchvision\n",
+        "import torchvision.transforms as transforms"
+      ],
+      "metadata": {
+        "id": "i4-ItRwkTdDU"
+      },
+      "execution_count": 17,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "transform = transforms.Compose(\n",
+        "    [transforms.ToTensor(),\n",
+        "     transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])\n",
+        "\n",
+        "batch_size = 4\n",
+        "\n",
+        "trainset = torchvision.datasets.CIFAR10(root='./data', train=True,\n",
+        "                                        download=True, transform=transform)\n",
+        "trainloader = torch.utils.data.DataLoader(trainset, batch_size=batch_size,\n",
+        "                                          shuffle=True, num_workers=2)\n",
+        "\n",
+        "testset = torchvision.datasets.CIFAR10(root='./data', train=False,\n",
+        "                                       download=True, transform=transform)\n",
+        "testloader = torch.utils.data.DataLoader(testset, batch_size=batch_size,\n",
+        "                                         shuffle=False, num_workers=2)\n",
+        "\n",
+        "classes = ('plane', 'car', 'bird', 'cat',\n",
+        "           'deer', 'dog', 'frog', 'horse', 'ship', 'truck')"
+      ],
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 104,
+          "referenced_widgets": [
+            "ce0f512204bc4540a56fa1363e85e257",
+            "e7ff134b9a404fd284a0f614d5c5903b",
+            "d75f11c2a2cc47d7af05c500dae96355",
+            "869d9806a41f4bac938f2053a66683d7",
+            "12bdb2122e2c4158884947e52c9e5b9d",
+            "ca04de8c81944319ab6f92b7074f08b1",
+            "af05e518ab24430d9095be0d5af78e17",
+            "a664d31fd7bb4fd2b9b030fb4295eb62",
+            "25e0eeb20dd4459bb7941ec4f13f8ae7",
+            "7038855e026244b9b894b498a076069f",
+            "2fa799f1c9484f098944c1c1fc9925f5"
+          ]
+        },
+        "id": "y8OhsIfDCXb2",
+        "outputId": "d084b3ec-cb75-4ba9-f09a-c9a9d4895ee5"
+      },
+      "execution_count": 2,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to ./data/cifar-10-python.tar.gz\n"
+          ]
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "  0%|          | 0/170498071 [00:00<?, ?it/s]"
+            ],
+            "application/vnd.jupyter.widget-view+json": {
+              "version_major": 2,
+              "version_minor": 0,
+              "model_id": "ce0f512204bc4540a56fa1363e85e257"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Extracting ./data/cifar-10-python.tar.gz to ./data\n",
+            "Files already downloaded and verified\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "import matplotlib.pyplot as plt\n",
+        "import numpy as np\n",
+        "\n",
+        "# functions to show an image\n",
+        "\n",
+        "\n",
+        "def imshow(img):\n",
+        "    img = img / 2 + 0.5     # unnormalize\n",
+        "    npimg = img.numpy()\n",
+        "    plt.imshow(np.transpose(npimg, (1, 2, 0)))\n",
+        "    plt.show()\n",
+        "\n",
+        "\n",
+        "# get some random training images\n",
+        "dataiter = iter(trainloader)\n",
+        "images, labels = dataiter.next()\n",
+        "\n",
+        "# show images\n",
+        "imshow(torchvision.utils.make_grid(images))\n",
+        "# print labels\n",
+        "print(' '.join(f'{classes[labels[j]]:5s}' for j in range(batch_size)))"
+      ],
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 156
+        },
+        "id": "XhjtVgj4CfN7",
+        "outputId": "8bc0c399-c982-413b-a6a6-d644822f2910"
+      },
+      "execution_count": 3,
+      "outputs": [
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ],
+            "image/png": "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\n"
+          },
+          "metadata": {
+            "needs_background": "light"
+          }
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "deer  dog   deer  dog  \n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "### 2. Convoultional Neural Network 정의하기\n",
+        "- 3-channel image 학습가능하게 하기"
+      ],
+      "metadata": {
+        "id": "f3lfjD9UUhLM"
+      }
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "import torch.nn as nn\n",
+        "import torch.nn.functional as F\n",
+        "\n",
+        "\n",
+        "class Net(nn.Module):\n",
+        "    def __init__(self):\n",
+        "      \n",
+        "        super().__init__()\n",
+        "        self.conv1 = nn.Conv2d(3, 6, 5)\n",
+        "        self.pool = nn.MaxPool2d(2, 2)\n",
+        "        self.conv2 = nn.Conv2d(6, 16, 5)\n",
+        "        self.fc1 = nn.Linear(16 * 5 * 5, 120)\n",
+        "        self.fc2 = nn.Linear(120, 84)\n",
+        "        self.fc3 = nn.Linear(84, 10)\n",
+        "\n",
+        "    def forward(self, x):\n",
+        "        x = self.pool(F.relu(self.conv1(x)))\n",
+        "        x = self.pool(F.relu(self.conv2(x)))\n",
+        "        x = torch.flatten(x, 1) # flatten all dimensions except batch\n",
+        "        x = F.relu(self.fc1(x))\n",
+        "        x = F.relu(self.fc2(x))\n",
+        "        x = self.fc3(x)\n",
+        "        return x\n",
+        "\n",
+        "\n",
+        "net = Net()"
+      ],
+      "metadata": {
+        "id": "jIY4esZkCjRN"
+      },
+      "execution_count": 4,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "### 3.Loss function 정의하기\n",
+        "- Cross-Entropy loss와 SGD momentum"
+      ],
+      "metadata": {
+        "id": "rzlfQaY5Ulgp"
+      }
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "import torch.optim as optim\n",
+        "\n",
+        "criterion = nn.CrossEntropyLoss()\n",
+        "optimizer = optim.SGD(net.parameters(), lr=0.001, momentum=0.9)"
+      ],
+      "metadata": {
+        "id": "MGoDiJmBCjT1"
+      },
+      "execution_count": 5,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "### 4.신경망 training data로 학습시키기"
+      ],
+      "metadata": {
+        "id": "RlOUhBLUUnlJ"
+      }
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "for epoch in range(2):  # loop over the dataset multiple times\n",
+        "\n",
+        "    running_loss = 0.0\n",
+        "    for i, data in enumerate(trainloader, 0):\n",
+        "        # get the inputs; data is a list of [inputs, labels]\n",
+        "        inputs, labels = data\n",
+        "\n",
+        "        # zero the parameter gradients\n",
+        "        optimizer.zero_grad()\n",
+        "\n",
+        "        # forward + backward + optimize\n",
+        "        outputs = net(inputs)\n",
+        "        loss = criterion(outputs, labels)\n",
+        "        loss.backward()\n",
+        "        optimizer.step()\n",
+        "\n",
+        "        # print statistics\n",
+        "        running_loss += loss.item()\n",
+        "        if i % 2000 == 1999:    # print every 2000 mini-batches\n",
+        "            print(f'[{epoch + 1}, {i + 1:5d}] loss: {running_loss / 2000:.3f}')\n",
+        "            running_loss = 0.0\n",
+        "\n",
+        "print('Finished Training')"
+      ],
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "bg-f3Al5CjWL",
+        "outputId": "87d772e1-ab8d-444c-d2a6-9c4550a81343"
+      },
+      "execution_count": 6,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "[1,  2000] loss: 2.179\n",
+            "[1,  4000] loss: 1.840\n",
+            "[1,  6000] loss: 1.637\n",
+            "[1,  8000] loss: 1.550\n",
+            "[1, 10000] loss: 1.494\n",
+            "[1, 12000] loss: 1.463\n",
+            "[2,  2000] loss: 1.375\n",
+            "[2,  4000] loss: 1.371\n",
+            "[2,  6000] loss: 1.329\n",
+            "[2,  8000] loss: 1.346\n",
+            "[2, 10000] loss: 1.301\n",
+            "[2, 12000] loss: 1.291\n",
+            "Finished Training\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "PATH = './cifar_net.pth'\n",
+        "torch.save(net.state_dict(), PATH)"
+      ],
+      "metadata": {
+        "id": "vtCSImZFCjYj"
+      },
+      "execution_count": 7,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "### 5.test data로 신경망 검증하기"
+      ],
+      "metadata": {
+        "id": "GpiMR6jjUyB-"
+      }
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "dataiter = iter(testloader)\n",
+        "images, labels = dataiter.next()\n",
+        "\n",
+        "# print images\n",
+        "imshow(torchvision.utils.make_grid(images))\n",
+        "print('GroundTruth: ', ' '.join(f'{classes[labels[j]]:5s}' for j in range(4)))"
+      ],
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 156
+        },
+        "id": "euz6kki4Cja4",
+        "outputId": "2db8dcc0-5b4d-4159-c3bf-7d0e02bbf896"
+      },
+      "execution_count": 8,
+      "outputs": [
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ],
+            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXAAAAB5CAYAAAAgYXpDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOy9u49sSbbe91srIvbOrKrz7J6eOzOkeAnp6soQIIeQXAGCABkC6BGiHBkCriVf9AR6/ANk0SAkOXp4kkFAhgBBJklBjkRJFMX74JBDTvf0eVRVZu4dj0VjRezMOl1n5nT3vWxesGImu+pkZe5H7IgvvvWtR4iZ8dSe2lN7ak/tT1/TH/oCntpTe2pP7al9t/YE4E/tqT21p/antD0B+FN7ak/tqf0pbU8A/tSe2lN7an9K2xOAP7Wn9tSe2p/S9gTgT+2pPbWn9qe0fS8AF5H/QET+XxH5+yLyV/64LuqpPbWn9tSe2m9u8l3jwEUkAH8P+PeBnwN/G/jLZvZ3//gu76k9taf21J7ax1r8Ht/9t4G/b2b/AEBE/jvgLwIfBfCrqyt7+fLl9zjlU3tqT+2p/cvXfvGLX3xlZj/68P3vA+A/A/7hxb9/Dvw7v+4LL1++5Pd+7/e+xymf2lN7ak/tX772V//qX/3Dx97/E3diisjvicjfEZG/czgc/qRP99Se2lN7av/StO8D4P8I+LMX//4z/b0Hzcz+upn9BTP7C1dXV9/jdE/tqT21p/bULtv3kVD+NvA7IvLnceD+j4D/+NscQG0l1TuEur0ngIg8+Jxh+P8NhtNVZPvc445YGx/r/9R+9Mu/2fn38beLU4/jPjy+PPh9/Ol8pIeXMN4b12Fm5+/YOJ5dHNXO/zVo/XM2vgtIukGnm+1aRIQXL15wc3Pz4Ho/7McfrD3yeOyRS9veerS7H3vG3+/+cs68efOGdV239zROpKsXiMbtnK01Wi392fkrxkhKEyKCqiIiiJyvaPS9v3f5+/nKL8ezqn9H+h8fjLIxzlujttY/rxef386wnQcejKRH2zYbzGjNLs4riEq/bns4D8xABBW5mFxwe3vLu7dvHsyVGAMhBEQVDRMiitD7qfcbQOuHNau0fn/yyNWPe9V+DJAzA/01Q8Hn0jf7wT7oG9k64LI3P2iPvvnrA0GkY5X35+UI6PchYK3Q6kprjcNxIZf68QNetO8M4GZWROQ/A/5nIAB/w8z+r29zjFRveXn6fwh2QswhUEUIwcFW+ohvtVJb9YFmDczQEAiqGEbrCGcD9TAH5z7GRMBahJYAEIdFNFRUqw8q9PzsZEzUhhm0ZhiGmZxBV/waW+sDEKE8guIfDpLWGrW2/l0dH+oD1raf1YxmUAxyg2ZGrv5eevWvo6//jW0ChRD4nd/5HX73d3+3T8a2ne+PG8Q/+Xj2zd8fvLX1MwPNGEP84awdnWr+HP3N8cffNHceve4BMu/eveNv/a2/xZdffrn9Le6f8+LP/VvE3TPoYy2vC4fDnY/DUmitcf3sGS9fviKEwH43E0JARVD1SRlU/WcHOxUhdKBV8QXbAaugAtMUCCqodDBH+k8c+ETIOXM6HQGYpokYIyJCCAEARTdw9G40mtWt7x70x9YnQimFkosfoy8MA3ybGbX63LNaaa0SNGznFlEM+L//7v/J//G//21KKVt/7/czV9c7QtozXX2GhkSSQBAlxsg8TZjAWozSjJIX8npkzN+xyI0houpzNGokiBJEvJ+3IWHfBFgb/fBwEFp/ttbPQT8WHWhH3zw23J34AdLHZMeKx8ac92V/ThJQCZ1MBj9+UFCo6x35+BWn04nf//kveXf7aXLz92HgmNnfBP7md/2+UgntQGwHBy4zVJUo2tmAggnNCtoBvDYfkKEPBDM6gNvFItt65/aBKwItYS05HkhDMIIUVD4AcAwxaDSsNQyozQeAIZiNVdQfdmtCs8EiHODPPP/MXjZG1hrU6sdp3Sqw9hDAzc8n5tfS1xusND9HXR70o0+WPS9evPgTBfBvdaxPAvCLBW4D8EtQtg7gAO1RABfkIxbYx697fL7WuoHfaKqBuLshXT2H1sAapolQ/Lk1yUirhOmatH9GSpHdfk8IA4AdhENwMI2qDsoixA4OG4C3QmsZVZinQAhKELZFQGX8dFBd15WYnITsdjtiDKiGDXTDBcNFBLNGtQJm9Cn1AcOUbWHIawacDAzQiTHSWqMUtz5aLVhrBA2klB4A+H5/xYdop+pAHVNimvdomEgdfFOM7HazX0VpxGasGvoDbn3ungHW+82vN2ns/aok6aArY+KZj62LIWGtnQG8D6th/Yz/QbekdAC4Phw7D6yiQQ7H1fV5bucu2Fi3SO8r7ZgVAAWL9BUeUaHExmozWNssk09p3wvAv28zoG6TtgIVRcECghDcWALTDnANVe+hvlj67wY2TFihs+LW2V0frAZifhRnKuZmnFMhlNqJXmcstM3ssg7ODaWhmAjWH0QVoZlz+tL8+DI0D6yDM4Q+KcV8qTABURnjFbPBvmtnpB21L4bNYFePswIfgH9S9d2/9UJw8fFzf/QfF5bkuGsExHScrN/6eVKeRYJxf+eDiHwcxB+77vH5McEe/E2FKQWmFPsYaAiRskZqAVqmmhHVSAFiEFIUYnSgDsMq6uQhiG3AnsIAcL+nVg1aQxGiCqG/zgDu96n9+y0IMThgqowesP5iO88YI62NRb9tXSmi2xwaR7j8jqq6taC6Acn4tPYzBXUAlgsZJajwYU8bihFpTDTZg05UTaCRECJNZgdfdQLTMHJZMatgxZ97n+fjHsFo2ih97hbt1x7cKLbWrfSLZ19qpdZ6wazPAOuksF+ttm2U0WVd0d5H6v3mWNDGDcIH41H69aSYSFNCNRDitPXrYPY2GH5UBzNbsGWmxdqt+09rPziAN4SK4DSzYWaoOfsOo1MGyHf9iwszcWhIF8SiSyad29mZbfTlYAN6GZNALgG39qtyAHVc0G2a+LtCk4ANAO8gXqwDj10sAOaeYhXZ7kb7wLHOQE2G/jNA25mn2Zkf0Aef8nFQugSkP24g/65Mfky6y39vRszle/35XorjQ0Ib1smHC9rDxe1xEP/YdT8G3uP9GAIpDvIgWA0sMSAYZfXFVxVCEGKAGJQU+7PZXh3A1cE2qIP95a1LAxPbGHpQ7QB+OTYHoHcg1/O4Fxl9a9vfN6LQz1PGiO1dpxhhmyw+pgfTd0IzwKoDuA25x7Y5Nq5VLpiiPtrPillwEJcJZKbphEqkScJ09mPqmGsLuQasmU8FvJ+HlDTGkQk0cSmqjcUOQZovWsNPMK6v5EbpmvJmrYSAKtQqXdLkQrbZTrUBcgh0gtQl1W/IpeMZjb5LTLJHNKBht1lkm8wzXjGAClYSLSRM4zcsmV/XflAAP/eUg7eDpz4AUwfjhnXpRPQDMOBSG+3vi5ud3sddT7cODnIGN19NHSJlAO/GFPU807pe1YhUItWENbt0U6pQG7RqZLdWEWsuf4gR8Akca2dNIhdaZQdqFMN1cRmzxKxPRGcard+XXY6ux7rU7E+EhQ/G+q2/xzdB3NsHKM5YdM2X2vFM+0L3zU9/8/q+zXV/lLEzpDfz59h/V7EOtr6Uq9iwxwgDhGGz6LBh0XXWy1kSGdIMGx3ApZrmlmRf2c9SgIQOJOcxT5cANvLSrb1m6sfHqKWwLgutNVIKZ7lInc6MIX4GfHuw+OiYMv33Ns7JGMedgNhHRuRmVRqXYoVPxYZRnYEHIwSIc2S+mrEaoQBWiTQCDYwuDfYFLSgSIhpTl3w6kW2VZp36DQAPSilxu24ROoBrl4baePgX185GFC+B2QxabWyBFd18HPN2+EKmKZFSRNVlsXP/2vn5baM6bM/h27YfFsD7oPNXBYozHtNuTivg4G21goKGwcAv1NDO0rYoDVyDNIRmwYdlC5g5o+jw7drYJUOBbmoKw0kpophGzJRqkWyBXI33p0apsFYH8VKNdWlnAMdIQZiCA3Dsq/uUAlP0hxqjizIm9YJ4dj6qipoROhkRg2qDjT7Sk39CwP3hOT6lyXmlgf5cvnHdA5X9G1iD0rz/NJw9Emcrxh5OsH7sT73ux5yYH2vaxTIRX4iDNGJfbIM2mrT+XiOqv5KYs3IR141z7otwREUJBFJnz41Cs4pY7dKdgBWnlQ715zEqQghnH40Ma82G/epzRExpNNSMWjK1FnLO3N/dYa1xfbUjzBNYJw/qCyUdiLXPHL2QFkNHlbBZmT5nXKd1UK/1/N4jPd/JjMs47kdqtP7fJsXZfHJW3DTB9AxqQU6KlMpUVlLJ7vy3vhQExaYEKWHzHlFhF4WgPvewtlkUALW4hGIXzz7G2Be0C9K2jYuzTPmY5NJq2x7B5oT3riJqQERJKZJS6n3a5zTnRbttT7gvmxsAfTsU/4EBnIteuLBJbJiF/fbGvfWvjPCph/cqm5lnclZMu98f0YCIh4adweAiqsRPDJ1XnCeSYgSMQG1KrkJuUKqSq1Gakk2pDXJ3pErDGdV2FVA7ixk3E9RAfdJvmkJn3IDrxhcy0TZ5P9L+hQkZ5MLchs2aehCNY9LZZXcM062Z4o7qAFgIdHvVP6HnkLHvc6+fsghtLNzG9LLze3Ael/LB3xnP6sw8heajsDN5b8Np/QEDN+XsXOudN5xj2zEvGfeZEQ9HWjMjl0JeV3JeWZcFM2M/Jx44+i/wYptXl/fPmTU+1n8Xy2G/ll/Xm2e+6Qx8zL8z4UJcohigqiEg1oi5EWpGa8PW4qeJCVrEmmJWURO0if8cfXtJdkaoGOf7j+azGutzbXTJFoUgD/plXHLrx90WAzMne12a0c7sh1TzQWdtfTB+bn+2h+f81PaDAriYoVZQKx31utpstZumlYCcZRMdoYVd71W5GDgOeGetzodfiBEkME/P2M0vsFbJ65HWCusKay4IPYSoM27pVkAz52KrJaop74+ZN/cr1ZTMRCNQuqzSVChTwMwo60ItGdYKdUXBmbiCWkXaSorw/FpJEaYopBD7wtIdKdJDHa0LOAbRmjtB9fHHfOmc+xdls+ohA3wQBOBxxzhwtwZrXnnz7j21Vvb7PfNudumsVlSE/exhc0HFrTDOj/67yjuPNY9OMqI0d6bREKsIzpZVmjsmpRGoBFPUsjN1PNJAtSHq/pwo3ZEpELpzyqRiVMwqzdzqbMXHbyCCRDDbtFyLEQibRr2B3kWnWgduM+PrN2948/XX1FpYT6dNs57SRIxnpX4D6b5YbRZxU4bz3V+tM9uKdTZtrbl22//9GIK7A3NGmNEuhpgIps5Ai3XQG0RLlEmjz5cUPELszTv0q39KyZVyWECE3WevSc9uPFY/ThgK6vM84n1uZuRaaNYIZmiXR3Wccz+hU4RmSCcJYwS1GGgjNLN18DYQE5o01lgxDF1XpBRkvkavX2Ix0tJEi6lTFn9+wzYJCGJxUwtEBJo/W7WANkHa6PVPaz+4hCLW0DEAlM56gK4vusTgXm6ks+tLB9RGni9Cyi7NHnX2Pc8z1zc3tFo4qpuZawk068fz7gUiRnDnagfxigP4KVduD40mAjFi6s6Zpokm6j+bkUUoKLUu1MXZ1xyFIGC1YSUzJSHEwNR0Yx+Im+Dn0LquRWqXETYd/+MP+EPw/k669eYj+B7AeGHUDCPHzrfl5rQNcmSsa+H+/uAgJJwZenEzO/W4ZHskA+jbXONv+uwlwx7sUhgT3O0x17utu7Y7wNtm62FdL4ehmbfzMcQlmjre60A59G9rXeYzo9W6kZLz9Y/f7Nypdl4Ma2vcHw58/fatx63nlaDKZ69ebc63MUXOAE2/z348OUd8nFffIdF1kB8Pth/wMcLgszei3YIF7VKh0cTtkOGb8ugsRdUXwrkv1Lae4PYNbS3InQN42kV2CUQCSAJxQmUEogqTeBRaXVes1c52DUVJfSGUVqFEpBlSqz/b3rc1RWqKYBBqx6QmSIUaDJkL0AinE7quft3755jBqhEJ0Re7NgDcn5VusbNyZuIjwbD3gYxAiE9sP7yEMoaQJEZg/OAH1iM9toEk9NV/eJM7o+lm5jbAtuaJQaLC9fPn/OgnP6OsK19/9SXrcuKYj9TlBAjV3FGJJbDAkhunpVJa5X5ZydW4W41TDkicmNMNEhIWJppONNS96wZoQvNKK5k6naBVrGaKVSQERCdaMI6tkLM76Zq5Q9qiuoOsg4SZuXPv0lP0ax7wh1r4YyD8qcD8x8JqZcTU9meED9BWGq1BLpVSGsc1czwt5FJI80RMyWOnu1NoPOtLmenDJKlvex+Pvi8jakQ2fHK9d4Byg1Y7Q3WtfPhwBgiZNVrrbFXUF50AInEjJ5seal0N3RyYLh1Zq5ScMYwpRVoMDkY2xoRryX4LSjNYe7jckgtL9pjtjTmaj7HW3OLxb3VTfgPoPr/MHYaDabfm9yMMR6wD/ZARvjnvLjpza2PBGguPbCy+XQCXa/mBKqnDXvI5qQHmCcSDCUpuUDK23GEGVdwi9hwZD0rMrfRwYr+2YEK0CAi2JEih52X4Ito6xtQUKSl6Pkg1pIFWN4prhHrlY8HuD8iaCVmJ6TlMBvMVhC5X9ezWti1a2kF8OHTlLLXJ5euThjXwLwSAg8+a6A/p4uLdedd1MhkB891ROOJON8fKALkLxW04loLy8rPX/Nl/9V/jdDxSGtzf3XF7OlIPR8yE2rTLJhNmgbvTwtt3C2su/OrdgdNS0OkKna9JsmM/vSBNOyzMEBJNAiVE13CnhVIyVjMtr9SaOdy9J6+rxwyrUMnclVuEQjFjrTBHEJSgkILHqpuV7vVmrGH0Xx+0jzkxPxYq98+rjfDPTejsIFLM08JPubAsmcNh4fZwpORMnBMhRaYYmebZkzZ0GNpnzjimwuMnPn/m210vPR7b/2UGrf/bqst7WOmyygClgjRDzXMXGj2OufXEq47LOqIuLqOurPh5WpcAm8fCWS2s6wkzY54SKQZ35JvLHSPJTAS0Z0yuObPWynHJHE4ZxUjqBKY2j5aSBrFH+4Q2FqWHIO4gX2lNqa1Qq2efet/oFgbrlkNPePuoZNefgBkjYU3oi0qrCIo2X+CbjZwQKEyoKMpEsMlX1f0EolSDvFTa/T3tzVusNqq4lNIiWHTQXLTRLpy+yZRUo1tJUbEhwdbGWTM36hTJkwO4Zl/5tLqc0ialPFcH8NsDdlxJr4D0Etk37NqBfCu70IUUo3vXRDb4ZkimIj3SCT6ijn60/cAALoxgLCxgFnpaOgwn0fj90gQHzmYd4ycPTDnpq54DfqWUzLqcWBdneblWcjNKH9i5eAJAKYXaGodT4ZgbazaW4nJ2NCWJLzSi0V8hQIgg2lP76fqZ63xWnTVpSIRkW80LWiM3H0BRus6IUpL3SwzjJrozZBtgH+E6/xxA+WMZjbb9p19y12mxc5aa0Rmged2NnCulNnKurLlSanXwgDOI2DbMxxWwAcL24zfct/zGT3zzKww/il38uzMjs03z3WSWC+JwzhQYHGscsz2UJh7oxn0cDyrOWUJx524m50itI5qiszsMDaMcg2cpl1JY88rpdHKwnRJh68ezw3gQI//quZetNQcZq91J2C50bnnQ3+faMO2iDx5p2wCx7d9j4dgs5yEDAVtehgGiWHB9W8IORCE1LBoWo0ekiBNAk+DgHYdsUc/nNcOsR5QhtBSwoJ603YuxDDRpKSBzgmaYtGEoudw3CUydaE7dXzAlN5+DnjNePyAacmE59mHE5suTTxjHH2k/KIA3UYrM7hioiVLO5qOqsU/NM96kkHoqtQwgs4bVMQjcQbENpM72zIxSTpSa+fIXP2ddlZwrX7+9Z1kzb24XbhfhtFTe363k0nh/Xzkt7UIDh6UlmkY0XaG7G8JuR5h2hGlG44TEPih67HmURKuBdYFaMkEi18+e9XsrtFpZV+Xu/paS4aCVJJXrWRGZmKOnMk+qPTN1gIQzuiaX5b/OzTPF7IMwvj+eJo8Boch2mg0OLsKtanOmeFxXXzBrY82V1ow1V2oz1lxY10IphSYBiUJpsGRnpnPqZq2NaPzNvvq06/4N9/TY56MYSZqb33jIYBDnUtYKNReslh4i5yzahtNPpDs7q7OqvmCrto2B04q/LkC89e83azQqpRXWdaGUSquFw/0B5BJ0XQe5kmvm/UxplePhwOG08NWXX/Hzf/SP2O9mvvjscxChYpjSnYcu70TTLlC6Q9XsoliWVIK0vmisWDMnLBI2rR2MUjOtFq9V9EFfbmUdWnMZsNnZhzFwFeuhvDDi7WnNyxXUiqQJnj1D4ky6eoFpAFloUrDdDpJDmO1uIE3U4fWvFTkc0e5H8JeSQwJV7GoPs5fWUNc4KObx3WFKhCl5VueSsWa+FLgJgux63PZphVwJV9eE56/8WncBQo+uMnfIJvUb7kn0WM9z8GvysEMPWx6vT28/PAOX4NEcLbK2iFmlVUOloUFo4hNguBjDYDV2ZjjCFtgDG5Px1pqLV4fDHcZX5Ap3h0IuldNaWatwKsbdyZngm9uFw6kgISLRU32b9bonGpGYkJCQEDxkKCgaQgeVzjxVaUANI5RIidPkqb9ZKZKhBNYq5AqtGJlGECGXnkBh2h9u60lJHpHShhTxYU9evDVAjt5PH+39wZR/01O6OPg50xM+LmB0ccO8KNdaKkvOvc+LA3hpGxNfS6XV2tOLfdGstVGDbYy9mb/0kvXDxpI/ePPBj37hOAeVX7sCXIYGqlgH5rNvxoGodgA9x2GfL6DHjw/zmFH7ZIQR+mc2sjE6U0ZIpS8aZo1aC7VW1tUopbhsuNnYg6XPDIZZevjg8Xjk/v4ea43ysl7U8umA2dq2yF6y6KHdg9Ga0pr6vfY6RFg4n7cN52vb9PJvtAt9fFThfGBRdT18aO8q0LpUKtbQZpgqxIRME7q/woL7xZqt/v28+rO9cQC3AFUNKdW161zOdTc0UFOCEODZFezmzXHolokvJZoiMk9OuU8rVhtNXQqToEjqaDRnP8+8Q65md14GuRiTI0t0LOyXoaBjURkIdmFdfov2gwJ4bXAsAlU51sTaJu7v73jz9g0ijVfPld0svLpSPr9WAhBFtzSH0UTYgNL6ZPDWTSCEZV1YyztKFW4XPPGmiqfzhtgfSnPTyQqaEmFyAK9DI0txYzJpjkxzIiUPbyu1sqxeDjLnxauylUrq9RVCdGdcK9kr2pmhaUZR8npgzW513B4yOSk3c2COYXOMbSDxEYdRrY3cK8oN77ezZjmb53J+7wHGyeWk2rCOi09sC+Vln9eefHMpXQWJBHG5pLZGqY3jkjmuK8taOCwrrVkHFSjFP+NWScZaI5fC/UGZUuR4OJJi5NXzG/a7HfvdREyxA+pvWH6MD4J2PpRfHjFdDbRVtBVaK5gVpBWPWmiFllfyutLyCrW4Flo8hpsgaFRaqx5ZZc1T03FWLi3746sZq7UnY9qDa+o2Pc0KuazkNXeAc7kkpniWqACJEUmJUip5WWm5IAZB3V45rSuo8ubdWzR4qv0UlRCUF8+eMc+zJ/70bOdasrNzhar4wlpX99MqiEaqGNbr/lhdoWR37H7Y5Dy2zouEv3m59G/gbj2yp2TC+/eE00r98p9iv/oFtr/22m8pYb2PLBfoiXqa+/G10aQiuRDvFzQXREOvaVBdVdHqjsYqbrl3/0PsMkhohZgrVhssK9IaOgVsClBA788gLIDUI2VZPOJtdjyx3Y62v0JQWq+nsuURjIFpPUxRLoyxb2k4/7AAbsKxCFaUQ5tY2o5fvn/HP/ijrxGp/NYXEzdXAdjz4voK9BzKNZjDMFvBa1Oc64fUjWiZwLIcOa2NUpX7HCmmWJwh7iE2dJqdKe1OiBR0SqT93o+RPROMmHzwqpDmxG4/MU+JKSWWZaEu2U3K5UBZVzyu1UMZU3JH3Lp4CKM1Q+IOlYllzSzZEy52srIm5eXVxNXksaNh5DxfpGd/2EqtrDl368Tf0wuH0/ZeGJUe5VKm2/DDLvW4i/Oc1c9LDbRRumNtAxTzuKFmXiI018phWbk7LhyXlfvjcq7sCB3MHSjW09pN7gpWiUHZTZGUIkbjWc/Gvbreb7Vlzvrq+Uo/bI8R7kfBG59cYhVpBVp2qaNm/3et1DVTTgt17cBF694tQaIiVVFrBPN+CfgCHKz6cQDbAFy4LAzjZKx1SaNQ8sK6rqxrJudCTJFpNzt77uytu1Gozcinhbp6BmhUZ4mndaWY8dWbN6xl9TC76JmCZo2rq6stZa21SskeepfUC3a12rCybgDrxbcaLYpHlJTF+6GWb/Tlw2dgW+josN62YdbfpzXPEC2Z8P4d4e6A/fIX1F/+ETx7jl1FbN6BuM/J6855VIlm/34lU8noWtC7k/+Mw1fl3zFVRCJkzmU6VInXe19sSyXU4iG/y+KRKtcTphPkCu9Xl17mGUkRqwu1HEBApx0SAvbqM2yasQCNwEa4uyUd7Kwk6KUL4VuC+A8roYhCmMCgWaQSqBaoBMQg18BaldKUaurOGLlYquySLo5VTrg010oPoVqrsJReZ1u8Hq/ECU0zwRohFgKNGBupCSEmQojel83cLO61Ezy8yk1LlUQMSlWPHmn9eqwPRghgzjCHuXk2Ka2bn67PGY1cPJEkV2MtxqRGDDKQlSGdXTY3/5zVblmq21rvXxrx8ZfEejvOB/KLzy355vsXPw0vx5rzCpwLLg2H5XZv3dTeijXFgG7riVs3Ir7cjjoS1hnJqNyGDaBvZylALh9/Z3QyAPHcSXLZYY9NjG9g+CAJbM41B/Xxs/UENLzUARC67KrNkNqZdw/9VBXUzP/Weg2NvkjJKBkhzsik/81q8ZDB3mejnjSqm0ada3P/zFq5Oyy00ri/O5Fz5t27Ww7HE5oL2XzzCWuVw+HgSUoKKUXymrm5vmY3TVztemEpc6d7G5r1kEiaYdUwCoEEpXdczb7Q2XAYPuzcS9/JGFvbGNlGqI9LNSPUipYK6wrLgtSKDnDP2eeJGkhAmmB1jJHsSTXa0NATj9SwYLQR6QGepm+KrNnDBC8AnCBQouvnZQD4yZ+XVkQrrA1bsxvEnbFbKdTFSzxbw6WUdfWxQA+BFRi17YI3CQEAACAASURBVF0fPwdeDAt2RKx8m/YbAVxE/gbwHwK/NLN/s7/3Gvjvgd8G/gD4S2b25lue27Xk+TUtrtRjZK1K1j0l3CBUTjUgq3DME2uJSGi0UFwnM2fZYoLayKJ0JwuGh00BhwLZ4LgKh1Pz6mDzzmsTXz0n7W7QpbLaiZgbCyd0zUiIhGlyR2iYfCDXQu3F7/OyUFQI+x1XU0JbZQ0BaRW1Sqsr0gIazOt85OzHWtfu1Kme7lzcUtAweWjh4qFg744FDcr1DCn1ALrzqH/YzMi1ccilV9JLPf49bI7N1jM8rY8lvcC5b6pvF4lSjw8KBFiWE+/fvyOEwG7nNbFj6AkbrVLLSikVVWGKvjNLSL5gj1sppZJLpdVGDl5/2jdNKFsVPw1KMWMphdxar155jg23rZokZ7Du9/Wguy5vyc5Y//DOfVKkbu5WIIxkj1oI1og0khmzCZPBDiGixOJZtoIhPW46mhGCy366usIteUVyQQh9vI7FRrDcqBUohTkpQRM72SEaOC0r7++PrLnw9vae07Jyf3/k3e0dJRcOd8cuP524P558Ye9M10vf4sy/LKQY+MkXX/Ds5oaf/eS3+PO//a8wTYmb/Y6UAik3SmjU4uPUqssrrVRkN2N27Z2VTw6c5fSNvvRFO/g47EKSO/J9HFrXwAVfBGNpzHkl3J+wt7fU21vimgkhOjm5vaPFhRp3XjcFdQZuztppjbAPhOtOmqbmKysVk4aaENfsV3Ja8HqnPeNWhHYXsRCQWp1pt4odj9RWCTcz4WrGqlEXPL4/gAQoxwPHN28xM+LkEkqKO9KzBYmxyzaKZ1d7iY3So/AVz64uzS3WemHNfkr7FAb+XwH/JfDfXLz3V4D/xcz+moj8lf7v//xbnNebBDTOfVWVXiYygiasB/Q3hGaeHbkFYwAj5Ac4R14Mb27/3RrkZqwGuRq5eZXCSd0RGeJETDOxFUKsPjljIVTzSmfqqfEebSQ++IdTrYO5dOeL7w7Sd11xCorHHzlrr7WxZdf5RW9sHlxmseYPUoozrLU25tYDDDeHx+OPt3WHoZgRu4ntr3Olw0v2bRf/foBj9iGYf7y53p+782vqKf5nW3A4sBT6Rh3j3GzXM/hZE69Caa1RRai1O4DECEE262Fj3wKXWZkPwVg26ezRe/kIeJ+/PQZaTzzZCq51dt61WrWe5GH+orYtvG84KL2KoEHtoX9mUKuzOvEiVYOBO/NwWdwZuEcWheghq2up1OqlUU8nl6Pe3t7xq1+9IXcAL8Ud9GvxEMS1bwWnVERcDinLkRQC0uD+5t6jVX70ObXO7FL0IlXRd46q1UHcaqGsnpxW1GglubO9rJvF8OHQHP6Wi6T9Pi44T2TYSt2KGVobUqpvelIrIGiINFWkevFppNKReetrK8UX2clh0TBa6HNgG7Dev9afm7Qej9+zvT1TR/28xV9tXdyntIqjZQOKeDSMVUw8/LiWAs2oEtAKrfRnXL2swrhR69dsfaJttVU26ffbtd8I4Gb2v4nIb3/w9l8E/t3++38N/K98BwCf5h0vnv+Wm3e/vOdYF+K0Y5qvCEH44rde8/xmx+urQpozgZXWVjcjN7Okx3aKIswgCeLENO+xapze3nG/ZEwimhJp2nHz/EfEaSakPSHtqG0lhEKopedXeLiYb6V2DpVzYFBqLnz91dfcpUAUj+NurRFCIE0TaUqkPFFK4Xg8PoykaH4cVWWXvLZHxsjWsCosuVHEeH9YgYbKxPMbX+QkFPeaf7hjh4jLQbPXH7YQaJ2BO/C1vsh1YO1a3LYADkj9DaPnkuRaB9tWq0cKjV2FLpx38xSJVSkVhMIItURcohIRcq7kXLoe3gG6Sw2j5raq9C3EArv9jImXLIL2DYAe0s+4py3bkE9bmGop3L5/T8mFWvPmXD0eDuScKTnTWmU5nXh/e0sKgWXy7f1aB7KgypwSqso0BWIMrhae/KqWNbt01PMJ2J4VXSaq7j8oXs27UhEVjqeVd3cHTkvmzfsj98eF+0NhLYFqoNOeFBqkhlQfb6lnekbt0tzpyLGncp8Wj+T48stfcbXbsd/NHF+/Yj+7pLLfTbRaWE8HWq3k5UjNCzc3110+hLIcqaVwd3/4BvyoKDFECNHvT0dGrp1X4T6wxJyorEGQOVF+9Bny/Ia5vCa1hYqwqHoIbQd0LyLlIX1kZ+A6R3SfelGv/CA6JphLLmJQzG2CKkbug2SUs4jNKNUjo5bTCWuNuEvEOW2yrKkSnj1DdzuYErs4+wIfPUyRFy9YU3ee9hLfvtB3388DodOL3VXYJMJPbd9VA/+xmf2i//5PgB9/l4OkNPPi9UvA+OXdL5FDI6SZNO+ZUuKzz3/KZ6+e81xvmeQt2o7Y+p5muQfAK6bBA/0JGDNYIqRr0vySmitLKdwdjTTvmHZXxHnP1c1rpu4MMQnkIgQ9oerx5NRe8nJoWKHvX2eCilJL5d2bN6jA9TyznycH7zS5wzJOpJgpuXI6LQ+SUmSYlOI7v4QgHs1QlYwDuNK4Pfrg282BKgFR6ybfpXh9bhITOu0/cHD14dFT2UcquDD2CDyH5T0cNd/k5hvYXwiaZn2z3415n5lqUJhTooZG9k09uxnp1dqmeYeGQM6FNRdfoNQdUtLr3njUxdgw2N8L6vfX42w2Hj+u6SwH+TU1HpGIPm7IUEvl/u7OgbrWbmUUltPJy5LmjLXKui7c390RQ2CNvjnAuq7kdSGlyLPra2IIzDmRUvAIBB2x031B1YAG325rjLFc8wbcReihlV5y9rhmbu9PnE4r7++O3B9XlrWy1m6Ox9hZonlySr9NES+mFlU4hVvKsmDNM2DLmvn6V18TVbnaz7RSuNrN26vVwrocupP5npIXluXEFCOqwnI8UEvmcDxxwXS8m1UJYWRYO+CORKfNnBoM3TxENqvAnOCzV+40VmVWfVDbvKnXUiGo72jT5yzN0BjRFDGMpXm019C/1QRrHvgZxB38RYUcfXSMxb6YEbrMuZ4WWqtMKWIxUlUoQTAV4jwTYiTudszTjVMK9YJd9XpHTqEjtw/MtN26bFr3kDBHQNJletentO/txDQzk19T51REfg/4PYAXL148/JuKF2QHpt2e/b6yX1aurp8xpcj+6pr9/ppZComTOwWyF8axrWyk7/ohGtntX5DSDWl+xu76M9bc+KLM7G6Ovu2ZwTTtmdKOFGb6NoceDdDZ9VbAnREN7Owao+vfGbFGsNa3RBtOiPOYDDEwTROlVlJK1NZc6+5et1FkKwR/wDF5Cr7VSIgJWiWXxvFUOZwih1MhBWEXXFO0D4L9zYx3797zj3/xT3rYmpuv2gdpr3bpTDb53ou7FNmn6HGvF0PmUk5+cI4P1gwzenRNIsb4zRKadAmkO+GcdJ0BPAbf0cVC6OHAsrHRIZOInBNG2qgHM4pCmfVolWGHu+YaoxcrmlL0PSPHDOkQ/qDU7SOtlMLt+1uO6dTP6RLCuq7UWjmcTuTVC0Rp8B10UggeVlkytWRSjORaiSF4Yf/YNzjupvTYJFs1OuiKov3e15J7QS+hhuCLozbQyPHkoZhrqQOvQHRzto9p2Eq7gIi+Xtl5r9QQAhKEm/3MnAL7q32Xv7yWeW3NI5qsUUrmeH/rlQ2XE6VkmgTS7hYVYT0dKSVzf1o+wG8ffyGoO/r61kTSB88o9oXYCAf380vfl3LyRcjwvR1Kg5V1KwvW8K+P1G0Zi0EzpLgFtzb3OYyNVEKfpIr0pBqvw1L74N5AtbkFY63vK4FgJp2IQe7nbs0IzZ/BtJths44FYsSr8w3C9QGNuCBhNibUEFHssRn4ePuuAP5PReQnZvYLEfkJ8MuPfdDM/jrw1wF++tOfPrgy1cC0u0JVef7SqHpFmHY0C6QU+PxHP+blixuuWuK6CW1NnNavaFYYFQbMAtYmgs786Is/z6vXP+Xq5jNevPoZtTY++9mvOB6OfPX1G7786leEmLjevSTEyN3hRFkWaja0BdQaUQMpBM+8sh4Olz29fl1OLKcTQWCfBInhgcnT+kicdztC8NClUj3U7nA8kUv1B6y+QEyaMPP6EiVNxBAoeaWWzP3xlttyQjCu5sRuCnz2IrGbIvbBY6u18ft/8Af8wZdfb3WLwWOBVZzFTj0c77NXL9jtZr54/YovPnvl2v0mr9vGEPqo6myeM5WDbb/RkKKHoYVATG6FDHnHsw8DAdjvlCkltt2RVF3XFaUED2nzVcZBrPR9DXMtLMvJCzT1Egglr+TBxHLeYt4Rj7a4utoRY+T1qxfc3Nz4XpNBH8go2608YskcTyf+8B/+nMskeAc/jyJa1oVaC29u3/OPfzVKvI7p6Ut+jJHr/Y4QfG/N2DcrHjuzqDrr1BA8xE0E0QQi5O7UlRiQefakk16+4e5w4t39iZwrS2lU80Uv7aYxAvszy9SeEyDDEqnZnXUG8zyTYuDP/PQLXjy78b+V3GUgjywqy4mDVU6nE7/61Ve+8XFtlGbs3x748v0JEXFWXAsV91ldthCVaZogJlpya9fnbkfckd3ax5uIQEwoQkx7FCi5UEpltYV7lGpnm8p9UX1R6ky+1UZbvdxA7bpyUCctHr7nzykF3fYctRY6Vrn12IrvZuRhkwKmhGJo80Sf6k4dkjYCxk2cubq+3gQRMZAoI4OHzcq4wHAZc+vCAP5YPaNf174rgP9PwH8C/LX+83/8TkcZkRLBWUQIkZQmdvs9U/JCRilNxJYINXkpMDnvTwnOx2rDOzntma6eM189Z3fzklaNZ6dKijOHYybFO0KIxBBRdQCxQffoHFzG1kmjRvK5Y92B5KmS1hMqRhF6uaiOGIJLAe5BD5uT9SwF6PYAzax/hp7V2SMxGtTiGYun1UPK3C9yya1GM46nE6XKGAm4hekAk1JgmhPTlNj1FOH1We4WwfaVB44lbJjvgyENmWRYGrYB8UP2rQ9K/hpdW9yembOSwYxVz7lp5y1FnQXWWslrprTKaVn67ukr63Ki1VEu9JwFORJ8UorkctOfm5xvqD/qDbYfoeK1Vg7HpYdkah8jIwnFyNmZuGhBij/vbfPhXpAo9pK4QZWUnRD4rkw+xsLY4zIENIbOwH0z21y8RozEgErwuuJiIJXTkt1nUOq5NCwPI4Yunfzn2+5st49Dj1SKzNPMbjeTF1hL7oTWrZ2xk44vnitrzuTq8oLJiiYH8LyunrgUEjE9BPCxT+fYvsfL5V76lEYI4YXc0yVC6XHsXljLS1t4Edc+U6X7JLeDOXyOcNoRd+7XYVwQfgQvzTBW81E1cARJ1H6MbbBv1pvjee3MXz2fyKXKXg9pc17L5R39OpuPC/b97dunhBH+t7jD8nMR+TnwX+DA/T+IyH8K/CHwl77DuRFRNHipyOOSeX97j4bAj3/yU6YUuX72nGk/EfMRzROtJMzEYy3dLqJkeH9YiSlS2BP2r9HdK2R6gdRGmhasBqZ4R9IJDZEYJl8swkoKkaLVH6zAfr8jTon704nj3Z0PZhWCBJo1lmXBpkSab9jvZl68fM3nX/x400rNjDTZlm2ot3f+lMWjDWKaSPNEa401e+JKHSaeRmKaEQmUZaVWWHLg63cru7kyRSEXI02V+ebcjx7Y0MhWNscijMpxyroKy8kzG3cxUtbMq2c3tDpix5uThTYSoxTVcGbjshl3fr4xeFJid3WNiBJD2KShIYOMUjW+W8l5coJ03fwMOKVk7k53lFq5P504rZ7xeDjcO4isqxca6zp0a4269vttldYqaUq8eHHDPE9cX11xc32NRMNMH41IeWzCrLnwyzfvWHMhhJ4LYD1C1xq5rDSrxJhIsyfVhOCOtJEUErRwyPVid3chqjKFOGDGx78qEvqCpqmTE69BrzES1ogEpZpnY56WhdvbQy9a1Rxk5Jxt23o9kyVn1rX0qCdnlVdTZIrBr3sfuuzTWNeTJ7kd75liIO8jQSdePrvh+c0V796/5+37tzSMVbyq4bEKp/ceNtjM5cWruXEdH5Yp9kXKE5zi5PM8l0ypsm3MsTmdpdtnW7Zi7dp4waiYVES8JOCQRATr7h3DWo+2GYuGgPQEuLFZtNCjxAbj7yULRn5cG6V9ab0irDCig8aG0V5qxcdxRPsmDF4h8YHMaEYcZKqDul4MOtt+tTNz3/776e1TolD+8kf+9O99qzM91kSR4KbjmguH04mbZ8949fozB5uriRSVYBPSzjHeThz7VKjG4VAJqdBkQqcbdLqBeI1II8Z7LBoxzARNDpIa0RAJGrfkkQ5VzNNEElhr6cDQi/ioOnPNxeOdY2Kad1zf3PDixUtyzhwOB2p1gcPAHTsXbFTEQ8KmaabWSqleBGkMOteLfd9CDRMSqid+HQpraTy7cmZ//awx8RCQWm3UVi72Y+wAri4frAIlRQ77A7TmKdrNI1JaH8iDnTiTHKV9z5E4F8YKhod3TcPxSDeB+wQeJqE/Z7mQLIbY8BBAS63cH+5Z1pX39/fcH4+sy8L9/b0vdr1MgWcGerheyV5q12uGZKYpAY3dbvbM2Fp9IxDYrkk+OO+HIJ5r5e3dPcdlZUozMU6MuidmjdIyzSrTbOyCWx+RoYG7U1eAmMvZeSZC1MCcPFmmNTfx0UFEfBdzEQVNiCQkQmwFUaX2qoTLsnA4nHis7ojRn785kcjF+6Zkv559vCJoYIqR6zn0fnGLIq/umDSLrnGnwG438fr1SwwjTcmLi5lrxqU2ltPSH23f5T0a1x9O725haRBCchmomdL6rlLDChvDZiRQua3WQ5JG6d1RzVF6Uh1ebrnHtVCtdkv3bA2G4c/qGckqct4VqddF96S8EcLXScqIRgZPcWfMT//dJ0EPKei1VNowZbfrp+vj1heLy465MIz44A/fsv3AmZjSK5z5LdfqE9QLyJs/JA0OtsG9wCEED+HpjrmofcNZq3z1yy8h/gEvXi382G5QYD0dqMuJ2goj8mgUTp/nCLrHaMTbc6lNj/0WdvPkpnzT7itxbVZE2O/3XN/ccHNzw83NM06nk0ecmCelVPNQpFIruZ4LCrUuDZiZO9l6QftmxT3aaUIkEKfZnSTBM8sQYVkLQRpTKR90o/Dq5Uvis1e0Wimru15i8PC2IWPHGHj53Bnq1f7KmVsvobvFOZtLI0NCoJvAW+w1bH1h2+4hY+PWPqE/KGo8GNOGoGNRYHjfXQmt1jcp6LHPm7LbGvM8u2Vh4mZrbeQldxP/xLqemHcT11fX7HYuvYnoxk4Hz9ku4SPNcJmgGH0HpAFSvY6cghBI8479/voMGMBq9A1vveoiwBRj30E9eZo1Qi0rtdVefM6jqTTNoAHVGdHJw0E1YaJU8Gp/4lEro0Qy1rZna65H9TC1ziODa7QiwvNnV7y4uWY/RZ5fzQQVYnRQOkyhV0vEa5C3TC4vEPUxs9/tqM041Ext1eum9DFAD0/9hnQrZwlFuxVC90eYBdelu8Q3Clh5PkEPsBu1wTWgSYkizD0rdBtPdM3Y+qJo5o7TLtkF0c2RHno9+TFObZOg3FHuY9G2hXXECVi3MLoP1oljB3AN/pxCDH4PMqjJGcQvbddhnAwb40FfMRayR0zFX9N+2D0xRdHoZmhrdOZQKaVuGVwhTiRmJtmhdWZKE6FMBPUyn1WNORi5Lvz+//f3+P9//5f87M/9LpVnpBjR01soC6UshCg9oqmh2ri+3nEdrggJ3r4P2FrIqyctxCDcXF/12uCZUnsIWK8K9+LVSz57/YrPPv+Cz3/0Be/fv+f27s71s86A1pxdCsjFF6fmG/dqLqgK8zT1p7Z0qSKya3FLnlCNBGmoFEwq98d71qUyv14fIFFQ5c/87Gf85Ld/xxn4mgHcgaah97Wbl3OKBFX2u7ilEudlOVedaw2V4E5YUa/MFkKvuez6u+81aHgEUN8tfTjmLvRw1RF9MUC9D2o5J1fU8TLf5KGY67+TTCgzz25cK2rlfG1RIq02jvcnSi7c3b3j9u49+6sdn73+rBe92qNylnM82uYhA39MQmlmrA1O1byOhfl9TMlZdpCIiHF185wXLz7zsVtrTztXSm6Umjn1sDq5ToQ0IWlPvHrhC/HxQCnrtrmvhkDc3aBxIoQdIcyu+Tb3kay27TlAnLoeX1fMCqNqIviO6AaE6H4Pa40SXcb58Y9e8ePPX/Psas/nL58RVNyaaJXb929583VkXU68/dWXlLzyW1+8JgRhmiPPXzwjpMjb4ztyPlGNrjM78wc2EBzYI/QkpKBoEGLymG0j9HEdScH9SL5DvZHX1ufZuQppjO7cj9aI896XYnv4BOX86wbSbvnoNj/CcK5/+OyNbXOMAeAm0Po+vOdkNNss1IB26cfHual4GduL3XTUPET38n/+tw7efd5sY7JbK1uxuU9sP3g52eHUMxt1CdpFyFhPl95Mb2WE+vXEXHcOqQ+odTlxOt1yONyzLCdajYS8ImWllEytZXva7ndQwpRIKRJTINaAioOfag9Jq0qMnrZ/CUwpTX2TWJdXhvYLdJZ9tiQe+Ab7fY77Gg91FHb30rTOQEONBNoW5dC6I7V9SHcEUkrsdjtn/12/nqbk2jTnFX7qoXXxYsuwzVFUKrUW37i3M/DYTd2GR+3V1jguJ49lRvE9rNjAWnWw/h790Z1mo+9k9GGPvjjLLJ0pBSURN7M3jAzN4CAZJBA1OdOtRgmBXHaseWE3z+ymmXmeN0cy1kPQ6V7/3zg9Rv0YN/G8HEH3CTg1BFpn3l1/FvUaPSIXTvHhwA3+LGMipAQIIaUuCvj/fDcq37QA7a8uFbb+qu3Cqc4jPomuQ4h4woiZ30eKzTc0nhK7yYuv7eaZEIRSxSWXZWa/m4de0KUq3zwCs83pKXgmsZ+/y2rCZk19oyfl4QtxOan169zYM5WGZzPmde3+oNR/Th3U1Ovud0A9axADxP2pXu7b9BiAf9jcka+bTWhA24B4hPWNUeF23MWM2n6cZ7g8+HGmDLJ9nC4Bj+e1SawMPPj09gMzcEGkB773cpjrslLWTOlRHSFE2qosq1GzYT3VXlgJGFOAZzthyvD+3S33hwN3bz/n3duvSDEg6zuoC+/fv+fdu7fMux27ZzMpRPbXM1c3N6CNF7fPmE6J8qZRrXm97zR57PjOJ9DxcOTN/I7d/opnz57z7PkLRJXj6eROtuoM4nBaOB5PHHts7DbBVD1DLBdqE5DmEko7Ox2nSXshoT0pBlQKkQK2wqn/7ZG+HMWGaHaxwS5b0kLSiKhsCRhGIWffXCJXl3Xubu84HA49ftfDAp+9eMa823FcfMuzw/HEH/zRP+T97X3PiPTnFKP/TME3Hw5B/fpVmSdf6GKMTJNHw3z++efsdvP273mKvH7+7LwTu/n4CGPLgV4vRUU3h/J6vafWxovjnuPhGfNu4rPPXnsmbIrQnFXltfYYcbc+ton0CJS7njsRqrLf37DfeVZwir47zWm5o5a+8826IChjz9JSCrm6w3q+ukFVefbqFTc3N6RpZn/1DESYrq6prZK7heYLuG/lV/sm17V5JmFrjbwulLJ2n4xXGzTzmIzhXPPoFgc9sUagohGm3cwUAs+vr7i52nG1m3xbv6Ds5ytUhV0IXE0T93e3HN++4b428mnlzVdfU5txNXnZY0rhePve63tMvjtOZZTAeGR+Y92CPG9mEQREAzU3TuuR1hqnfE9pmdPhwOH+DhVlN+0JIfLq5Stunr/w++qLch2c1nqkyYjttwsAH8DI47LeeeJwljg62GpfPB9aFGcpchtAFxNR+xe8dML58y6b++LvhHPoKH3B6eTPpaQRqfTp7YffE7Mvz0MbHq+hhWln57l4YL27LcL2kKLAHAWP31xYT8bpeMfpeEeJEdY7qAuH4y33x1uvF0xF1EhzZHc1s8s7dlc7mhjxLqBL17iIWPc2NxPmeSbGSEqJebdj3u1AxFOse43vsTfhaVk83Ms4Mzq6Dl6r753YQ6q25BL1SBRtRkqpb2zszhJaoyznxKFvNKPX2rgYdJ1iexRE6D8dwEv1KIY6iug047is3N0fOiC7Jj9d7dA0cVxW3t/e8/72jj/8o3/MP2PuTX4s27b1rt+sVrGriMjqFLd4NkZug0QDiSZ/AD33EEhIbiPRwOIvcAuJriUaINEACSToIiQadGiAQAgZZPv5+fn5nnuKzMiIXaxqFjTGmGvvyMz7fE7rsK7iZpzMKPZea84xx/jGN77v/YePZGN18OIarIMX3rN3jiYEnLP0nXCz26ah71u6rqPfbBTrthgT8N7Sm1Zed82htNICtGGZdDhJqH1LELipaSyd0iR3u60ODlU1RGTDm4JzejBwzRw/2y6mUlvBh5am7WRYxxttpl6IWkXFGKWS0kSwTtwaa/E64NT2G7rNTiq2rpMDIjdrUzIVrT7VqqRQDYjLug+qsUP1p2QFhbSJpw06q+4uxgqDw1vogsA/InvsCd6p56ela4Ictsg+shTa0DA7T4riAmStZN+lGEiJOI2Y0uJCC6aQTdUb+dJhWG4+pOaoBtWpJOIUiSkyThfmODKcT5yPz2JN2M0E35A3Pa7sBdZTbDuiz7SUm94Na4Ct6+c2u5XX8/lrXDHvWggWjUl1T/Lye1/8jFvLvWuSvVYDplbWGtRvoHX92utrq01W+wuz8F8/gK8BR0rEYC1929K3LSHIAjvNC88fnmE5Y4aIiYWd14aIqtW5nHE2Y02i5JlxlDHnslwgzxhn2R32dF2PD+Ko43zANy2h7ej6LRlL1w0si0y6LQndUPJIvPf0XacluhOxK1gDYcqVv6wfOa8c8trtrprIOYu4kOBeTjWvLaBlYnZEHRRwJMgO3zpIHu8/mcTMhfcfHlnsX0l2lkTa0ztp9nrraX2j2XDAWodvFJc0GWOiBHrf0G52OOdpWxlGavotvu1gmJhjYZwiT88XHj+eSBhi7e47gTzkd8qkZasBCYKQvgAAIABJREFU4rDb0TYt+/2WTdfhjVDqWh9orCOoFZ2zpVJiZT8pF7eUQl5E2fB2ba8mEg6axmOdFc/TeWHWQ9XqIWGtpe+z9EWsxTuzjrTfXtZYmtCSyDovoBVAksZh/fIcM9M0A8o+yYV5XoQ6imDnKcMwLRg34H1cxdSMk2pF9Kx0iCTXcr5OGJcVovFenHBskfssm0YDuEJcViUcrLXC4c6Jxlm2nRiDtI3IHhuKyAGkiC0J55woZeaMw7DZbMVCLMHHj0dCCGy3uyt0lzNpWZgvF3HLUYeblMKf3OJrIDM3XP9c1sRnHAfmZWQaL8zzmZILw+kZg2Eejjw//kRoGrabA857QtdLX8EK3bLCgy+A7Zo/lwLUGYXrWH39UotZobX6vTWpuvUKpR4IGF66In0SyMxt+F4XqkJaL6tnOSu0k1BjhzaIf+716wbwimMJ/o+h0HjPftPT9x2dunGfLzN/9ccPmDTQp0goBdcbWjUn8E6CrHMSxFMauZw/Cp4cz5Ajfb/h7vBA07T4tsUGj29amran7ROb7QHjGs6nkRjFbLdMy4pBFqAJge12y6bf4H1QtUKzTlvWwB1TYqmc5XzFwnNlpqjEas4SYPrNlrZtkWPZQS44AilK88QbB8mRS4CYaILjNpLlnPnujz/yz38cqNrjlKsCXbCeRiGRtuvw3nG433F42BOCY7dpcM6KTkzoaJqGzXYv+i5twHlHOV6Y5sx5WPjpw4k//vBRRpx1n+ij0KpBcNO2lef35v4Vu00vbkr3D3jr6UPDtmnFeV7xyarxXRu+lZVQcibOEpilypHBphAkaHsvza6UMsMwkFLmchkYx0mgnEbeQ4yZthNRrBA8k1ZNt5e1jrbtKa7IQW8s6EBLWTeXIcbEcBkFAhjrcI8MdBWuuhvnYWJaMsEHuimq9G6H90Fkpq1XyCyt+jRGxcecsTK+WYLKAouxBWhBZyrSLhlc0zRCHS2ip9N6y/22pQmWTdvQOIMtmWUaMQbyrHRCLf+tsRz2dzSh5XQ68sOPH+j7Hh86qQ5U63yZZy7jJGywfoMLgbhpX2IO8iqvvR2t1Gq/oFJD50W4/tN8YRyeGS9PxGXh9Ch//tj2tG3Hpt/y5s3XtG3H/et3MiPSdDTb3TUrNqzQyrqgjKG2sA1XCGN9mfYavK92gTfsltX67vo11Uj6RTC+WUefTVNqnCvl6lC//oMG/Aq/fimp+OuuX5lGKP8nneOrJKtXHLWelpUZYFLBRmEtLBnB37TUXk8+I2PP8zLhshddAwrGOZq2E3qZk+m3mDLjLEMPSyqkJA0N5zw5g3MiOFE0oADXZqV+XJuA1w9QNoYTHZCkNEJ5XhaVQBEtGHvL3LDaE8gi2mSlIWXUZ9E6CfD2k4ZMAaZpYsiyGYvqh5gixaa3nsbNeO/IBpocmOZZLeCsakVbrSqkYWe9eH5i7Qo1iO+EXWEkU3G/mwAenASFJng2fUcInsN+x7bfsNts2HSdNNFqc1K5yrW5W5k6UeGDrLS14TJoAM+kLOp8NYALXVDMDqZpIafM5TIyTRPOO0JOeO9o2hbjqu0eK53z0zVprMVaeY5FYagq9ylfohu4vPy4KuwZKvXQGKnUzPp8rx/WSn/CWAhIf8RakTGGawZoLNhUjXDrOqyltsAIUjlo87eIJn3jBdq6jo3LCy3KsS5IpbC+x1zWpmvOhWleMNap0qBh1gGqalhNUd3zbG8mYr+80dctY0W/X8TfJGtv2xZjM6YsmDKxTJYpeDk4c2QaBwxwOj0zLxO+aSlA1y3yrJzDO39V6axQxoqBG43TV9LAi91zG9SvyTXFCCpQ1l1m1udy82XXP78Ibd5k3eVlBn67iNas/+Zn/pzr121igo4VS7ndt4G+a+g7aWo5MawjFstYRNryPMr0U+MtbSOqwN6JKL6olCXGZeDD0wd8aNg0nuAt7WbLq7dfS9ZsAsUYPjxd+OmjNBvffzgJjp0dXbvB+4T3C3OMnMcj4zSLyNDKMLA6xl80405M88K8RHwI9NuteOMZS4yJ4/kMS8TrdKQxKnNhDG0rWW9t2spQzSBhIkWIA5ZE13u8MXStf7kEc+HD4xM/PH8Q1oepZb+ozjlnNCNu+fY337BzWxjORBPx1nB+lknBVw8P7PcHnJVBJeus0PtiJJWC856263j39TtC266BtjJOnBWfxd12w3az4e2bVzQh8LDf0qu63X4nKn1dCJQlchnEIzTGxGUaZaBnGBinWUfpZw3gIucqZaYMyYhMq2quOCdsDR3yWZZITAkfPKFrCU1gTondfkcTPG3XcLoMa9P0uiaNBlylyi2RykCQxqrX4K6mv0UakNbVDW7BeJwTpbq+39G2nYyudwJxNI1UQ5XlckvTuLKuak+jkHTS1HvpK1SNcYFq86od7tU8mbRAWgjOsGkswQrriBy1qpHhnmTrGr7Cfi4EgjFMqfDh+QzPZ7778T0pZ96/fyTGRXomMQv2bS05JvJ+owfcFwDc9VC0oINz222gC3tyzrzJB1KZWaYz83RkHC58v/sD4/nM44f3PH18ZBiOPH18j7WO3f4Vfb9jt7/j1ZuvaNqO+4dXdF1HCIGmbXRNGs38b+h55tPQ+/klZtxZ4cj8AvgogFHxuzWbLvCF0Pz5bVgD/BW2qZOnqRRS4bOK8F92/fqDPBVDVdqeNMDko05lFQwJR8QSowxyTEl4slTivq0YVCHmKKJDpdB6iysW5xu6fquZt4x5j8PMOC2M08IwyJBCMEYoaIqZZcWvJbi/bGygp2bNvGvTya5ZataBkoizjmiSMjT8NWs13Kj5iXZ0KRCtHkg5C9/XZII3BOdFgvbmKhTGaeL5OCr+LJSvFMcVpnHe0OfInCKxZOa4YCexBMsWvLUcdvsVT6xSBaXicqDaJ47tdkOMSdkYGjx0aOj1wwN3hz2H/Y6vv/6KtgncbXu6JtB4TxtkyZWcdOhoYRon5hg5Xi4sMXI8n7iMozA9NJAPw8CiWtypsnZ8nQKUQS9pJNZnItm8C54mRcISOEwjoW1kRTkr04qfZeDadFJZi1y0tWi4ZtfW3WTbArtQteMxmnV7rA1439A0ncI2ohHudCDN6Fqg6uloL0Aqu0KV3Es6HBS8p+sa7Ztory0n6lCPQ/4syVCi0Gsbb3QIRSuzKvmL3CuDDB9FNYCoSUTKhXEW6m2dah3GUQ8M/TBicpJ1nfxLtjooVo8RaYrGtbLHjCMTiUvDMgXGtmM4nvDWcjx+JCXRT1+mCBiWJXNp5UD3oaXreoF6lP1UQZE64bxqsmDWuF1u/r9m1tfddO3LiUHA7RoxminXaeLPse0v9iDLJ8dAtcYqdZ19Us39zOtXb2IahHPdtZ1M0TVSXhuQRRITwQe2uzumwfJ4fE9KkbFYJmNIJoFZSDbh2kTbG0LX0LYyIeWCOO90mx1396+xzmNcIBf453/1Haf3onux6pLkCVOkO75EZZfkJPrW3rDpLF3IlPlMHA1LKCwuk9NEEyRwTtPEOE0M48jlcpFGzTSyLJFQYRTdtM4axY2zYJcpUkrCxDMmnmjKjAsJZwq9F1uszyhRpbCMI8PpWTLtpkWkYy2uaek3LbvDRrDtTYP3MI0nTk8j3sDWCf2v92K+G7qWaZ50glAyxLxMdMFith2/+/Ydb1/dU70SrfKMnXPcHw7sdlsNNi3WGKbpwjTma7mYy9UYYZqZx5klJU7DQEyJ03BhmCZVxpOR/5jiKoeKBlQbrWLhajunh5c1hqBTftL0aoUV0rRY7ymo6t8XIBRrHW3XiORcAZOVxePk95bUyvNRUVJrRP7gmklLSd+qxVzX9zRNi3OiCCmHqcJgRu6vfJtby/vVWFqhjRiFheLcjRCWrpti5H6SCzlG0fm+nJjHM87A7IW6NwdD46R523rl2DuPNVYnhuswi0q6Oo8NLUuGcT6TUpTDO/Q0ueCiBKRkHcVY3BepE1c82qzVhUIQN3CGMVYGtHyrwz8tX32TWcaJvu95uL9nuFz46af3xEWC+TCe9X4LNJpLZrvdsd/vyff3eB/Ybjea/ctaeJEpKzx1y+pawZIbdstn8apCMXqvpHd5w4Nfm5hXLPzFT5EyboUl9Umv0MlfL3b8+fWrQyiVPtO1Hbvtjq7tRGy9IBZVpRB8w+5wD9Ywm5YxRYYSGAl4kzB2opCwbaLdGJq+XQO49wHrG/rNnrtXbwlNQ9ttyQW+++MHjseLOrqr9skyUNIsAXyZtSucsBbaxkBy9L5Q5iNxSCy+sFjR52gaEe6PcWa4nLkMI+fTWTvtkzi4IxvHGaH2ValRUwN4FvdyG0+Y5UhwmU2TVMJWGrbRiU5xvQowjwPn5ycaHzAbcTDf9jv6vuX+1YF3X7+RCicI5np8OvLxxx/wxjD7QOMdrQWbFnzbcB5EWKzpepwXvemucXSNY7/5RtehLD7nPX3X471ju9nQdR3LsjAMF+Ky8PTxzDgMLPPMOEo2Nw6TZtjSg4gpcZkmYpZMb1SlwVSFubQn4pxUAYCKFhm6LJvJO09oJLttlMVkncMrDz00Lc7LKPqs1mOfNTGdo+s3uOLFxDeLqXQbnPZjZsWcHdZ4vAvs93eE0CgkIrBIbYAKFKW8+GDX4GuqCmPFX+u0lv6dxAkpA5ZZBtGkwJTDqw5YrVg2mZwWcpwZz0dOz2I4MjrZY2OwNF6og2x6aVp7ua+1eQZG9NkBXMCGlrJEhmkmpcjdfsO2b4m50KpB+BgzsUgF92noubI5WKvta7+qUuiKYteF0HQY00GX2W32lJx4eHXP+fiOj4+PGAPn84XHx2cuw0UaoONECA0xLmw2O968kXXetuIo5Ksfqb64KpVRM2KRJagiVrfBvf73dbpSHk3FzNUzwNx8S32P6/sv19/5WeTTCnf9WiFi/DIA5f8PGbipwu8yrWadX0+nemoH7+nblmVuMUYm1rJpiKbFmETEUEzC+onQiOjNPJ3xKdI1e5zzxJi5nC+4eWaapHQexws5L+QUiYtqTMeZkpZVwySrNjUIq8ObhDeRYBPBREyeSYuK/2uJKk7ijqCQUClO+a/aBMlZZGFzkakvVIOEJAEiR2yOBDKNLbQqpOONjvR+ehOLZLTzcKGEgDOFkgLh1YFt33HYbXl1L9oW8zKJSath9fCUzE4hq1rDCUUCZ6TBaepkoqQ08ux0EQqzotXAKbKiMS3EtLDEhXmZmZaZZZGeQk6ZWNSSwzpwomERMLicJQP0geqIDtw0WK1CJ2gzUJpglaPf9xt5HU3A+6BUM6V8ajZbm6PLEtXD8mY9Ilm4LW5tXIogk3hFes29rHEYpHnmnBdpWOdWJsoqReCuDepVIcncDBOtDIrbAF5fTR3u1rlNVTs0ALkG8Awl6RoWQ4llmZnnWZ5xECW+7ITLvQ6/ZBlYoZbv3OC5BREqa0Q183A4kHPi9f2e3bZjjonLFIkpY4eJOSY1s/5y9rjS7/RN17dcJQAwqARxleaVQ7Bof6jkHcsS2e0P0lQ9jRgjcE5cJiiZcThjKJyOnq4LtG2Hd8LMabuOpmkFwjE1eNY3fJMF1wy9XO/FFcTV18mfmOT90l9Kk0I/vQnsn3zxCszW1/YLrl99EtM5jw+BttvSb2ahLBVHzk71Eiz77VZwWge+2VDGyOLuGd2BaCLZTBgizdbj2guny5kPP/4lodmw272jaTecTmf+6T/9CwpFFNdy5v37n0jxxDyNnI7PV1ZCKcQMU6pPUnjSvkwEc2FnC3d+YBNkwGSKJxKOREvJhq7zWLvB6/TevEQ1/5XAuCwzJXtmF0iu4MsiCH+eiOmILZGmDHgzs3GOfSv6G0XF8DNFXEH0KqUwHJ/4+P13eO8Z+pbtdsPf/lu/4XffvOXb337Nv/q3/xYxLvzFX/45z8/PdN6y7zu8NeyDTOv1jadRG79gxBZt04g+u/ee0EqJK7rbVpUMUWkBtQbTDGsYI6fzM+M08Xh85jIMpFSIUWEQ68B6XGNFW8IY9k7CY22O6psDWGGHNYOtWLURTZmmaUV4qe+0qSl4q7ghidToPEcmxdRjXJgHqY5erElrCbbFEMhRxNWCt5LZW0vbOrxT/Dvr1wdJPHwQ3RMw62F3ZaAYSqnsoZsg8snHtZgvq0RqyjMpL6RlYRoGxbs1+UG0UHKKTMORGGeOz088ffxI8I7SNwTnaL0FE8jFMCfV6bCoK42t/lZrQG+7jvv7B6Dwzddv8dby9bsHHg47zsPI+8dnxmnmux9+4nQZ2Pbdi0xVN/h6cJk6ZITBOElYlI1HNlXut2a8MrBOsYTDA4f9ge3mANlyPp8Yh8RZdXCG4SjVTBppQuD48Xt++uGf03Ydb99+Rdd1PLx+w939AyE0bLdbfSayniSRcvq+83poS1b+Uqt+pRled97tm9X//kJwvrkxpg4N1dBSK7Ab1tLnN/JPX796ADf6kO1tBl4qydVgsIrrNjRNIzffODKBSAM4vDFYE7G+xdqIuZxZ5gGQEtQYyzwvHI9H5Z8O2hg7k/NCSjPLMmi5LpstZqRJWtCsN2NMEfVDm3Am4lgES82WbBrJcrhtyOpYecqrwE5tsCUjvHEAUxIQKWnBxAlKxLlIsIVgxc8QxFbqk4RxvVJcWKZBbNlsIQZpGO62PYfdlvu7PfM80figk4U63GHFgDd4cexxqmXirOigCA1Np08bYaaEEDDOriqCQi2UMW7haQNG2BMpRWIWjDVnKbuFQueVmVA3k1sPgRzTDd6NZvk1k2U9JKxmbY2ujVoJVPU7Y9DmHOLpmPKqiRPjsuq3v1iTVCs6d6VxVmql0kKDdypHKoqZa9OxNulqlVLhAnO7iRX//HyvAxWHrYNfSTNm0flOORLjDOoMY5DD1hnEFSfGldETY8RSyDmoMFMdpDF1Pkp7pNfJz2sIBWc9oWlw1tA2Uk0+3N/z6n5HOA2iSx8mno4nYkw6bPTJ/tZn93Iysd6BG+EnUFaGfs+KkVc99UCKid1ujzGWRgfppDewQIJ5ghxncl6IcaKdOtomMM8bGq3QSk60bYNzRaabK65SbkYxqZn45wH5iubf/M21iPj8eX4S328Qsj/x+P9Edv/XXL+6GmFoAk3T0PUyXu29pzhHspZkHFFmmvEUQtPKCPs0sSyJp+cT267h7mGPd4W0ZEryhC4RmhPWNYzjiH1+4nQ8Yq1oaIiuhIjzp7TIQMEoMIhvOqyTEe1pkSDSeIuzjnazY+87vC28/3jiw8cjH44jz5eZbnvg4e1vFWMVD71pHPn4+IF5XjidLuq+rl1458lJBjs2jcGL3j3eCD7eNZ42FEK1vUe0TnIpMgH3yZNOKREX0cQ+HA7cP9zxm9/+lr/5r/xNXr9+xbbf0oaWb776lv3uQFpm4jzjrGHrG7y17LYb+q4TM4G2VTx4SwitTLvqAJFVb7C0zMJm0Wy6SuMWwDcdd6/esFkiTX/HNIubzDQv+tzl50t2I5mZu2k21XSw6J95HaxIKy5+u0PmmChL5DxcFAGSoJdjWvnjcRGmUY2naR7XQ7ReuRSWlIhELNoId+pZCUzTwjwJfr8sCxSzZpe7/YHtbod1Wq0YNS24SVSgBghkvkCHdlJJiKZ30teeyEmDdRIj5bwI1AeFsAY4ry5LFmd6UgqrmqZ3jl3f4p1lu+no2mYNqtlYYrGiZZ11BuLmflpn6TuRnW28xTnDPI08PSWmaYGS8dZwv9/TtR3lht3yYo/Lr1sP26q+Zw2rAVtRLLgGb4ypnhdycCXhp+92B5wL3N8/8Pz0zPH5mcvpqCYmhmwS8xxJcWKazsQ44X3D09MHvv/+QNf13D+8IoSG3W5P2/ai69+IcqVVJpXVJmOd8SgoUqXvZeWb6+t92aRkraRe/MttxF5v9fXQMrwU4vq5189x5Pkd8F8izvMF+AellP/MGPMK+K+BvwH8BfB3SimPv+B3Y63RzE90ItpeNncVSIrGYbAUF/DG4JuWtmnFMDgmjvMFby2h29IGyzIvpMUQ2hnfbDHGMU0zpZwEf50XzWgk86pJUowLswrUu9AJxl5gjkJbqnhm07Ucdo44j3z88APTOPKHHx/58fGZh9fvCP0dbbvB+YCxjmmaePr4JA29cdZOvywO5yw5R7y3NLaFEATntk7cW3pL1xicLWAzpWrXlbJi0PUS5xUJ4NYYdvsd9/f3fPPN1/z+979ns+nZ9BtSSrx7+xV3h3sqj9gZQ+cCzhjatlGecXVKsjgXlGHBakRbqMFGhj2cF79Aa64IvQst+7tX5Jzpd+IgM00zl2HCWEvX9+vASHXuMQoxVMfwqwzBlYmxRDlwK1OglCI+mXEhRjHViJphpyj6KWkWCKwOoXjvCI2j6EH+6b2MKbFgRXrXhTX7pxSmeSQnCeLjNKqQlVFIyeCbQAiF0DZX/rGt21IOvjqgk7IE7lwycxQt+RgjMS2UHElplMiRE6ZkkUiIwuF23qn2taNpPJSCU9qpoWj/xbFpZcq2b0XgK2dR/QOIWExRKYh0y4EoqwiZBHBhvCzLTFoGsTDM0kM57Hdsc2EYBy6Xy4uK5oXKHjVAlZcB3FwDuPxqswbx2gCVvWrZbHd4HzjsDxwOB+Iyr7x2oeFZljkzq2/m6fyMMZYPjz/Rtj19v+HNm7e0bcfr12/Z7/d03Yb9/g7nPF3XYQnUMdeCYUE0mNaga1gPYnQ3rHD6F1gnV5ee8iKIf4Y2XVfIF/71T18/JwOPwH9USvnfjTF74H8zxvyPwL8P/E+llL9vjPl7wN8D/uOf/ZsB1gaNUKS8coRzEfV06a7XDq9grU3b0HUtcZg1oyrEhHCjTcC6Dhe2dP1eBjtSYRonpnliHCestcKl9VXKNMuIuWpzd11HaDfMZYIhrhzNnAvFWKwP2JxxvsU1hW6zY5cMbdcTY8aYiTRM5AKX82WlaC2K61or04PSEJvxxdIaQ+8hmEzvCt4UWp+RhDdjlN8r0v4QX575GGNU7W7H/u6Bd199w+vXr2m6DRnLEguXQWh7S4SURaO5SntG40XiM1ty1CxDScLWRqoBraZIZHUnHMbIMEacg8SMtVXZUD6E2VOY56R+oollll+QkjzPeYnMsWaysjGCFyin3ATd6gq/LDPjNEnGpyqAwhMWoadxGtdmX84qdFTFtOuq04ytpM+pYoaKLYvO95IhxsI8yVpZloGsbk0pibmH+IIK1i0Qksoiq8SsrY3CEjUg5Wu2rQE8pmUNRuLxVTTACT1SNDhURwWBTSxVl6NoZaQeqD6Qm1agMIXMxDil2t7VRl0dRjE3db226fTgTDkzz3Wyd4Eiw2wp1wPpim1/aX+/gEuqN2XlVptMtSwT55ry2fOQHyOvyerUpVNlS/tiCOr6iOukLbq/pcE7MxnD8fmJMQwYCuNwput6hstJG+A77WcEvJdpz5zlHYg/qdNZDX/TlKw88GsoN0j7We7Zta+xRvoCL+gr5vbVf/r5X3/9HEu174Dv9POjMeYfAr8B/h3EKxPgvwD+Z35hAC+lUPICxRMaR79tSamwRHkDER020DrMe8/d4YCzhsf4yOU8Mi2JcckU42j8Dud7+l3DffEs88Lz45F5HDidz5xOZzabnt/97lu6rpOpshQJJogBgLNsDq9p+i3pwzM/PcsJH5NSgYzHNRuwgXa34JaZN80du1eitTCOM5fLyNPzkWGYmOZF6IMxMY6iTth1lqZpCTbRcSaQOfieV223SuM6UxRn1w2e43V5GEMkXU99JODsDve8+fq3/P7P/ox/7d/4N3l4eGD/8I45O+YhcjwfqSPQGdHUdi5ggIjBFMNlrm2btJbDFQ4yq5RaoRApiD/oNE0Ys+B8pGA4XS4MCkctUTS885IoScybvfPyE5IErufjiefjUVgICqH0fU/TtaKBEgVSGEdpWk2TcOtLZf1cj3lqdQCssrbeWjoXJFutDkipsORMSV/AwI3QO0uxLNOsrugz0ziscq6l5HXiz/vApu/FfDt4Uk7YZElJnlkpmWwtMad1FL0OJNVna0rRwFlW7N7asropeee1tIeglUBaRF7WOdGytsZIDwH0c79auVkjTCKrE5ygw2Ily+CTvHHNlisGX0hJKph5lvce51EwZ2OxRpg3u92B0NSs9eUl21aarOL2KVWTKZK0YYRWe5sc3OLKugCvz0VVJptGhndCE1gdbD5t/JW6LpDXvcxMlxOn50eMMfxRtfybVhr+PgT2+3vapqPf7Nhu74QO6juscXTdlrbdEJoG2zYiBVGpn6VSCypNUd5LpqwHIfXtreeVPLdVsE3Xc9Vg+bnXL8LAjTF/A/jXgf8V+EqDO8AfEYjlS9/zd4G/C3B3d/fJv5bVCcaagnNGq6ebhoqeUAYw1qw6HNU5vuJ3KaOqgaIXLmwWizFnXYyJeZ5Ff8HIEE1S/0JrLD64lTrV3ighSgaebway7Np0taXQEMCjEMHCopzvy2UQyVZt3skHin87nC14Wwgm01j5aJ1ZWSACJAmuWIOSrPHP+bYYQ9N2bHZ7tvs9+8M92/0B6wJLEnXCvFQhfi1ZbSHXPhto9lOHbVgXUZW8NVIDg7kG8GWKzHMCk1EfDM7nifMwqtyqNOHKItmy6IHLgl4Ujz6fL5xOZwpmpeClXGiT6qAo7W8YRuKyME4jw/mih79mhgZeSpcaSvCELBo4GQcWnL7h24GNLyV8NRO8NjsX5nlSuEU+vMJkVv1RZU26uqoFyjBGm43ys6q+d1RrMuoUJTeaSkWgRUsd2rma8jpjlJZaKEkYTrfZZ4UsrHOiAwQrA+a2iWpqsCyGahlWX3fRNZBWeeeoKotJpmLjjDFOhOOcJ+W8esB+cf/f7PVri7QG6rJ+wVVz5PO1Xf+hDv+sb6XUJFArtVoB5bI2SyWoJqxYZ5AXHV1fFq0AJ1KeCT5QCrRNJ/MBKjTmw6yCY2gsSDRtQ/ZeqtjK519v4hX47fRBAAAgAElEQVSKqknWi2Tr+tJf/P3NbfpF188O4MaYHfDfAv9hKeX5E8J6McZ88VeXUv4B8A8Avv322xdfk1NkHk7YEiFDcHIiRaMBqyi3tAiu54yh7zrImafQgLGkDOdxYUkwuKzUJIt1O3ybuH/tSQfRb/jw+KTqgBLw50UaMm3bsNltaZqG+7sHtvsDSzQ8Po8yov70yDjPnDvLeSPTlhiP85bGexyOYRh5Ok1Mc+Q0zJwuo3b3jcRFG7ABul5gjtbM3Fto7cJD3/LQBryDzupwbmbV9y7ZrYdVuQmu9bLW8u1vf8vm/je8efeOw8Nrmr7j+TRzGj6KS0+UAJ50oxgnjccVc7tJgtaAznU0/QrQabmL4O51rL0U2Tyn84lhvFB1rFcMqhSa0NA1HZXKmXLmfB44n2Y07cQYwzhmnJ8k60viPL/Ms7AslkUMmbluaOfMOkLtlO3jisNkwTITecXvXRE1OXEL4gajlCvlzLgMzMlxOp7EGk0DjnOW/f4gvPOuo99stfQWmuVtUJqmsa5/eb8lM6dFKqn6ur3qsxupGFY5XlcDOGuwXcO03s+auBijQmcVazb1wNXvKNwwg2oz1enPkkpL4CCBnKZxlEGrcWAaR2ISMSn5u5F5mXHKGAqhwTYbsvEy1fr57lf4o6iPJStUUqEVNFst5tOAVoO1NOzreqrm4cfjM8fjkePxmRgj0yQzF/KdhapNX7XwcU7v6U0CVIqYZSRZfx+PR03OAs7WqkKgp81mT9dv2Ww2vHv3lqZp2O83dL1UYW0r7CcfWoHTnMNbz4vJjaKHisYFebGGirfLfO8vu35WADfGBCR4/1ellP9O//p7Y8w3pZTvjDHfAD/8wt8tjbdpJJqCKV6cVmzCkqg5uMFUATasgbYJ5NSueiWpFKZZ9LuNEdPZ4GWy07rCzgdMSXz48Cgi+blm6yLhKOJTDSG0tDoNut/tOV1mdpsNBvioi3caG6ZxXgOGtY5gxe1+WgpLzExzEm2VaUGEjfQQsh6DIbQ9/WZHZ2f2jLRWeN7b4HBGMnIQHnpSdx2rwVGadp/DhNZYXr954P4rcQnq9zK8dD5PzPOkmLRkwzJAU/MhgUdWjQ+q6OY1m1gbSbdby1wDC1TzAcnezucT41ThhnmN9wZowkLf6fvTzHocZqZBudhV6XFMojGdEzFJwzLPi2LXSZp5FfOtE4la0q7C+MWo6JBMKZaK/Wo55/x1mOZFxa6HxRQNw3BhuFx0MKhZG2n7/Z6269lstjKt2khFKI3TRSUAZh0YkvcZS2YmYozY33nltTdBXYvULcg7K1ONxlxZGiVpLyArQ0V7MqXI4WwyxV5d2KUqsfr8dK9Rs796v2oVklVCQhqoVXPmMlwYhkGniMdVC2WeZ+0ZtTRNZH83izrgJ30GqGd+XWmfZp+32fjL7yn6YNbkwjiMNupTTKK8OQxchotUunEhRjESWd3nnTRonXNYb3GmSAVfmUAIsyktiVlliucYpZpPmRjl3lWsf7M50PVb9vsd5+EjXdfx9u09h/1W4sZOdPR7s8OEBkujzCUD6HRrbXKWIhU960Za78/tffo5189hoRjgPwf+YSnlP735p/8B+PeAv69//ve/4PcCQn27XC6ipdxscN6SDFiTtcehpZCtRqVysjonok/VpqvCKl3f0gTPsowMw0mm1tIAKTIuCRsacJ5hXih2Eiy46Wn7HZvdvTQwm046+6Fhu+kB6LsNWaGQ4/GksUJW2FIcS3FchpGPp4F5nhljYSlW3GBK0lO9wTrPUizDnMAVltbjHMzOMzuHRauPkplRLXE9mQuFpQineakLQq9CkQ2XEtMs4+jWOqYpERdhrpRKhzJ16+hkIHXA5OoGskKKpoZ13VAG+bobaty6FTOUYrCKr8u07G14FDgjadZVwQi05Be4TL5c/C6zMDQUp84xSTlQa1lVnzR6EEPBatVigazqiMUaXHDa7HMKmclBT4FKSFzXuxWKY3GW7S7LcI4T/XDnZBzfqJFHTEnxXIVtFNeuE6S3eubOGjovzJTQhGuwVl2SmjFLkpGgyM8XBs6iHpVVsycxzyMpRbq2ZbPZiMSAkQzzdnGsQbNI9SXvWCCHyzSwKINnmidSTC+C9jhNpBhXXZpV0M2IuFsBHh8fGcaRLlj6xn0SfIquzuvxLwmD1eRBQ3k18qgHMQi8U9cNMmn8/PzMcLkwXs7EeYKcCM5BkcCelrjGhlIMycv3W5MlmNuC9ZUPf4VlbZEDwjuDtfL+5NBUx/oisyNZ5zXevxfSQ1xOPPWt0qDFI2C7O9AE9RloNxjrCEGGy7zzeE08ZaLcaNJSYcDrnvy518/JwP8t4N8F/i9jzP+hf/efIIH7vzHG/AfAPwP+zs/+rXot88zjh0fatuHulaNvGorNRKOud0kCufWiOyHSsaLy1zTCRum6TiiIbcvbt2/Y73d8/8fv+P7771mmiWV8JsWZcZwJ/Q6c4+k8cp4S+92eze6O7d09r979hq5t6VvJjjb9hlf3D3Rtx+V0xjtHmo58/8NPyiKQDOg4ZS5zZo6Zy6Rl6JJJ2a+8Z5HAbAi+ZUyWx8vC3GR2fQPBM3hP450EA+UDD7mwZOFIYz0ZiAhdypVATSZBssaPHx/58WmWhfLP/kJLaQ/lqj8twdJTJ77MTYYAdcLRruV9bRzVEXZhWpi1C2+cF59CDdwUcDR0zmGojSqRC8gqYhR1U6daSTiHaxoJ2DranlJcm5wpCfXTpNpYlZNERJwkRAgEIBWKLZrlOhlEaoInmAaj6pZizSV9Bgpk8yLe4ZyjD1sCgX67owovOdU5aZoG6+V5TPMMFCbqXlSXKC33BZ+VgB5CQ7/drMNAlYm0jpNXPriqP0p1OpFTZhwuMqswT1yGs7JxRFPm/v6ONzWrr/DAFVpeIeTq9FL7CjFFno5PjNOwiq4JS0kgq3kW6QOZdhTvSqvZu/SUMkwTp9OJUgpfv33N779995IjTUEalTrhmAUOJEsVaBQ+yXUtayO7Ytu3WPJ4Gfn+uz9wPp14fnzPdDlS4kzfeJzJPB3PzPMiWkdtK1PbLpOdHKzJplXxUaQRjAZS8VgFUEFLlmXBGpXTiNJwPp8nUs6cjp7z6b1UIa0MOFnnCF6kG+7v39B1Pdvtnt3uQBMa9od7QtOy2xzY9lucD4Sml56JN1dCvPHXwP4zr5/DQvlf+NNHwr/9s3/TF67KIy4YdqVcTx/DtRlQCtdJKT3RS1kDKKDCPI6+79ntdpy2W/p+gzWGuFzEPso3NJ0wIVxoZfS57Wj7DU23ITQ9PjRiEGCF+x1CQ4iJ0LSERqY1YzHkzGo4ME6Jy5SJubCkol190Q3BGfEzdR7jG4xvKMaTsEQMkcACjMXRZKvvV39uycxZysdSWQPZkTO0ONoX7SFkc58vL7JmYxoMddE6bXBpALf+RYmH4qdVaMk6+TP4Kw+6bgCnQd24IB8YLAIRkdPaszBGsr1kC8UIva6OKydtcKYYyaqPkqIEcGEHySBOzIKjm5wE96r4jRETWEmlClcmSl7HxCmWlyReqBZWxpQVF//0Et721cl8DeBK6au3PpesJX9RT0Z7LSNuH079uXKXqVDwNdLeCCwpE0HYH8JlH0cxp5jnSR2HEjHO2pgXfr/Tpn5lMdTXJNOWArXUw3FR9tW0zEyL2M/Niw47KbS1xCiuUinqcFEVnVI9lRt6pzBr9p/dybpHiwbktWlcrlPJaxKij/YFPFgq1FG1a6ShDKyTu5tNj5s94xS18aj9Iv0wOUsFW/sCRqqsWlkVY0j2Go9qr8mYqD0Jp08sYZJU3lnX+EQkLvI7Z40XfT+qOmcrB6W1pBhF1iEtMoRmLD4nPcQqO75Clb/s+lUnMcc58t2HM01Y2LyGnW+FwcYsD1II2MJJLpJpTrPwgIfhwjCcaNtA3wX2+w1/489+w2++/Zbf/uYdv//9t5xPJ/7RP/5HfPz4uHannXP0/QbvPIfDHbvdTvSE7x9k9J2IJRN6x2bvMM3E/hIx7Y6565najnEaef7pJ6Z55nmE8wwZRzJCIfFdK56PmuVKWd5rAHQUa5mD5b1xhAIfR0OIWtoDkJlm2WS5GGIRnegklTVftwe+5aY8zpmffvyef/Lnf5CR6mWhFNSlXDSnvW80C69OR2KX/ELW0tUgdc3Agxo7CAXQ6eKvfoQd1rcY4/C2lYNihWLKNbv0iKrBzQGcomSmMvqtmbdqlgg3upbe8uytBsvbAF68A2sIzqy6195LI7Nx0ASjGuoFawveJBxFXl/Ww+aTqyBa8cmwHlzW2FXHGnPFky3SfKuN01WnG4Px4qAcSaQIJSWWccZaQ1wHfHSqlZvKWSGXaRj48OOPzNPE89MTl/NZnezr8IoccN45Dvsduc1sum5lV1ULuKSwxzxNRJVNHseBlBPDNLCotMDyguIo+2yepjWgVZjIWiuDaReZcTifZEjufr9ZE6p6J+v4v836eshi5VmTFYXTiqI+JRfkCyowBzlGUlyYpplljqSY2W33OGMo5R2//92fscwLf/j+R47nM8/Pzzw+faQk0TO3ph7imU3fszvIWP1XX3/N/f2dNM6dI+XM8XRkmmY+fHjPOIqD1evdftXwd94LjDVNa8N3iQvGCMvNmELX9Rz2dxzu7ni4f3XljVslbUwDLooxtbUOaxqMccQ0rzMjv8TU4VcN4DFlLmMkNLAkAzaAuREXKjUDR7MT2fhihrpooyjig6NtGx4e7vj66zccDlsOhz3PxyMfnp9J6+4wq2Jdneja6nRX02+kpM2qBhggdIVkPN1mp2PHCXIkFsucLcNSGCOMSaEOJx3rEHq8SpeGpl872xirp6whOcvZtDgMp1gwUTr11a94Vt3jVGDOFcOU03qfG25V0UopnM8n3v/0A3FZWEah2XnXqLBTI9itucm2iwTx2sCU8tVII9AarOcawFVdURai6l87h/U9NvRY6wmuX53DLRUe0B8dDNwwKtAAXqcC188rd7qipvoGxf9IN7VCQcYaSOpqE1QQy0ngFOodauisU3+2rIeK4cbY4AubZc2hTZWHFXU88/JfWVEsw8osAf3cqQ5P0hy9iF53nX8BCd5rP8HVylNe07zMnE4nkWP4+JHz6aSZaOWuS6Abx5FlWdSRqGaQlQKojkYpMQ6axS8z54seBlHogZIZC16/6PfNy8IUtfpZ36e81hrg47JwPp8F3pmmz+5l/bm1J2BM1lhqFNqRZ6wtDf3+vBbhUOGxuHLSS8mqNCja703TiIE1lvb5SMyZD09POhyVsUYPu5wJTYNx4oV7d//A23cC+djgxb3JBy7DwPl8karLObab7apo2LYt0zRxfH6WamBeKHlZK42cRfq66zo2/YbNZqP3S9dLFhG1Ugo2OqxLuGykQZvj6p/7S9LwXzeAZ8Pz4vA4zskzliA4r20oJWkjTQSmKDKTiBGxITEn6Lk77Pjq7Wvu7+/Y9g3OFtrGsd9tsNbw7W9/R7s78PR05PHjEzkb4hixNhMZuGgJ5NxFwISyYEtimCPHQXwj//h44XwZiONMHDPTYklhD6XFBUOTtJlnBZ5wTQs+gAuU0IKxZOPXAJ6L2FmlEjBZPDKtNka9okWz8SJ4hbjziF6wZiXmE+EgAyE4+i6w2ExeKkdaNDRs4wit4ts60JxjJsdFGSSKUa47tfKp0Qz06vFZDyNx59lgQ48xHud7MA5LVZ+7ZqRutT67NtlSkqw7J2WWUEchFP4wNeOTjNbZK7tAHIfE1Np5x6YLbLoGZ6FrrEoSgDMFVydZb5q03jma4MhWnY9u12SMHC/PLMWz2+/ouk6+t3Z29RmsdYsGnkJhXtT4OCcxVyhX+uMKT1ErEWjahn4jOjNNK/BdWbu5rNl/8J6mUdd30+grlWcUmrBm5sfjkcF7llkakznllQ0zTbOM6ccokr45a5KQ1oAvzVHRwF/mmXlZ5PlpsjAsiRmYxpFxGMSVSbH1PxVzbhlMN20bPSBvDkBQk6OrdKvcA5l07Tcb3n719SqZm5Ng2sF7lhhx3ZbT+cLh4RXbw53ouNf7rQyuTd/z9u0buq7j3Ve/4dXr1xgnkGDKidBumMYJ3/T4psday267FbVUFcSKMcp7T0kO2EoXxdCEhq+++R2H/Z7tbs92d2BlAwDeBJzxAtF6rwnTy7H8L4N6f/r6VQP4nA2Pk8PkwHNsuJRW3oQVM16RKYsidlTsypYw1tJ1LYf9ltevHviz333D/f0dh70ITbku0HUtm+2WyTa8HUb+8T/5C358HmWcexKXl+NYCH6iGumWUvA6QDPFzHmRRf7D+2exk0ojxETOjqW9hwDBeCxeFh/I6zNimozzFN9o003w5lSEApaNJSLdco8Tx3ZrCBYohcksJCLZiKhXEQwJQyaaawamv5G2Cex2HdMlEYdCQhgRlIKznrYXHNxpBj4PC3OcKEU2bU55HeuWrEk53J9mqEZweYwEcBe2YDw41ZDByli+ZsnGGFVAlC58EwJwxWTX3W+Qho6lzomrwbVdJwudtTQh4K24zOyCowme/abnsOtxFpogLzHFiZwi1mQsEVsqxVBopn0XiIth+qRhFJeFj4/vGZPwsdtWxMkqglJvyRV40o2nMMU0DVJmj+OqT5Ni5VvLM8tZmuCHw46Hci+caj3k0CzaFFQn36kud6K6+awHm4GuayXD1sajgdXtXbBsWesp12B7zRZnDepJnY9EqVP591EycUkqpDk6ThNZcWhpeipmfkNXfHHd9AOuw1PXIG4wVAaG/EWFaq4/wnqrFeCB7WanP1f/TXs7KSfefPPEOI/89P4nvv3+e0rOK6e69h2a0LDb7WiawJs3ooVSM/BSMg/DyBIjD2++5vXbbzGInrh1VsgSjVjAFfUIuJzPTPOkzWGBZ+8O97RtK9OibS97oEohYFTvJyujpZCc9HKKMmPKl8dp/uT1qwbwVAxTkqz6PCWOlxmbIz4LFOa4Oo5XJoV3Hu8TbduKSNOmo+9bOnXfqCVuzpByYpxnLsPMZRT3jhilOVRyYVoyzsU1gFMKrghpb84Cj8wxMs6JKWZhYOlkYsKTLRQTyCZwe24WIzlLMZL/GcT8FiNYdqkPrJj1z1wMJtewbBAalVXwQN6XlM6Wz8lvFWaoWXOtRlXF76Zx573AHCYHTE64KJofySodL+sCrWYC2nTUdyavzQgKbEzA2gXjrgbNkdqcKqv0rUgRiP7Hyj/WrL+iKsYKrENRDjdXr1QZypADoGsCfdfSeBFqaoOj845gjcAjJYs5hh52snYUedHmrFNMH52WvL0KRUt2q1x2CWJFtWFqA/161+shVISOp9S7SQN4TtKgvTYY9Z6C4tTXxmOVA6Go0FiVI1aqrFGWC7BWSDmLgXOlf9bXIewJkSyovqY5Z+0FSTN5WVSCNl0HeXJSD9Ta0CtVbOzacL6aYuQvB+6be1khnUo/RGFuo8ykVSzEXDPQ61pbt/MnaWmtZK4fopESZCJ5s10PQd1NGIxWMp1oLhlHrubEsQ7XWMDhXEvbbgHwQVlYrsFarYIs2FIIbQHr5fGrtaB1DeDJxbKkIu/VKgxVtG+i+wqjpE6HzmroPMUvyMF/3Qw8wU+jKH79v3945Gkp7FvHm21DsLA1nlBxVWdxjZWyNnbs+ob4+p5vvv6Kd+9es9tuaFsPyHj3eVx4Og38n//3P+YPPz7yL/7wPX/5V98pLqg4Yqld9evGWafTjCda8c6cYiaVyhcNgmGjWDaBbLzirDW7qAHYiLciEqgFAbqa1xp0MWt2nVcYAUwuOHVdqdKWKUuGVgdUrpe87pJmHImuceRUmGMi5oKxiVwmvG25u9/Qtx15XkiTeH6ezuLbOYwD4ywBaBllg8YlrVzYleFQFsDgCTS2petaHt494EPLeYmMMbKkyKCqgSlHUonEbJmW+YqpFlTvWasPJ1olbfCEYMULVZ1+dvsdbduw6zse9lsaazm0nsZaSpGR9JQT8yh0uDr2bxV+sN7RdqJ53qnhxTLPPDn78k5mcWVKEeI8sIwtxTtMFhikbi4Rl9K1ow3Fp6ePfHx8lKbbMKzBv3J8rao6Viu6qNh/SdKoN9VooYA3jk3X03hZW00Iggers3xUeltKSct44TIY1MenGhcjh86SBGOVqUXhdU+XgbTEmzUr3qMVD68uU7PCQWmeKTpBW1kofyLCynpNUhVkM2PHEeuKHM7FYMlETUSyUfBMKwP5XIkL9eCryxx0pkF2kDyDwhwXYs4YHzg8vNYvNuvBKFOqVrVhLOdLYhjPAmfq/q+N85wDTfdqfc4YWCJiqG5YqZLWBVpXqw/5XUsszAuUSySXuL5uMKKyqQmFc1qKNBa8IU4Ly2WRRm16eS//uutXzsBhiIa5FB5PEy6ciNuWTeNonaVx1bxVcFVrHcE3OrlmoDTs91s2fSeWXlqCphyZ5pnLMPDTh0f+8P1P/PjhIx+fT6qbnVZtkmuVZ/RzDeA2UHwnAvjqkqHkMnntWP03T8ZJ1W9gHZsF6qTjLdPDKoYPqIiR/v4qarOuUl10ZfUp0S/5NHjr71IOuSHrQWKxOcnvWEs28e3s+ga8pQRHXMREOUaLOP4koimURZQKc0JLf/k56+vDYEg4I6qJu40ML5hpxiwWuxiWImqRJdaNKQcVVGlRVqxcmJdXA+HGVQ9HT/CeXS/j63fbnof9lmAte28I1jAvE/MssFvWwGa0QWgpys9WDRptxobgKV/IwKVETqriGsVnkkJexUrka7B2lb+tzTWh+V0kAx8EYnA6UGRUGfCWkphr5lAbIzd/moKaJBRCCIpTG9G0qdZoGkirqdCKya+pp1mNElLOxJxYdBgoRdU2WeLa45D3ogNI+vOLUgrlYIuiq3MDxXwqBnZ7VWzdxERcIjbLOLstRsWRperNtd4rrFn9n/odcotqU9RQ8ZZYqtSyITTqilT3UQ3k+r9SDPMi7zNlEc8rsP48qXzCi99d7/WtI9RqTr3GjyLUzyTyyTFWmYnroeyME4axoqymOEiGNGfinFgWhRZ/5vWrBvCCYTGWMRt+fD5znmc+9p7z6UgfPL+937HrApvWs3WWGCPPp2fSMnO369hvWradWEY5a8nFkrLhw9PIn//VD3x4OvMvvn/ih/cnns8LS3LkIgG52LI+/BXXkMJWNpB1VAGHmncJdKIyt3VxIWPMOcvE5BqQ10t/hu6uyugwyMTpzVfpn/LZ+vPLVTfB+kYCnpu5hVEKsEwTw+lEjgtxETGpJUsGXsZRqHHzzLnrKdNMKOCLEZw4ztgkpnAe4UfL4pKgXYP3Le4uMNVCjAPObfnqzR27/R2+6XChYZwmnk7PxCR6M3WqsHbk63i9tXZd1I0aPG+aQOsdjfdslRrXdKIe52zBTaKseEwz5EQqmVQEnjFWRNFc8FjvCG0gtAHnHW3X0rRB8My2pZSyBq7b5yDCUQWyHggUcqoj6Ob6+vVzq01m7z1t25CcxSAbMQQZ8TbWYas2h/p7dl27Uveen56xRrTp4yKV0WW4KCNk1slHgT6qfkkdMqpTwZXlU31cc85MldetPq9SValFoPN449YspijkUrPvCrUUDeRXvERXq/Y6agD7dHc/PX1knk7gemiOYMIaxMgJowyO6uy0wjKgXOvKjS4vgnmdJagmGQWItWJY15hZkx2rEKyrCeDqCOV0pqO+BX22lTa67i5uDpHrYWedW5PLOgOwEgLybTUhP8Yao6+lrD0MGoPxlhJH8ngm5+kzm7+/7vpVA3g2hsU4xmL448czlMSuMfzUOXZdg+NrXu+3JGMIrWdeIs9PT8R54H77jsPujm3fErzDWceUDDEbfnoa+X/+/I+8fzrxl9898uPjUXSns2Pls2lArQGz5Nr9FnDkCipyLeXgGsj1v0UeUzDsiLslKtwwMa4NPe8tITjp1WnZu/byzCq6uWpdr6bKxuBUhdG4DEzXG1kKyzhwOT4LA0KnFxdExzCmyDhNLKHh6AP/X3vnFnNbltX135hzXfbe37l1Vxf0jUgTCAhEgRAC0RjTagQl4AMPKFGMJLyYiJfEQHhQE1+IRtQEIQSU1hBabUE6JBgROyE+gIKYFoGG5iL0rbqqzu277L3XWnMOH8aYc639ne9UnWq6zzlf3KNqn29f15rXMcf1P/J2yzq2rL1wsIwjIScazTSe4BIOEl0K866qAQApD4wTxJh464t3ePEtn8VnvfhW3nTnTVxst9y7d5dxHDnfbs2plrLVuqzmKqpGEsTrcAqcNI3ZtWPDuu8secJ14e3ugtOLM6Zp4OHZQysx1pgRPrYNq5snxCbS9g2xa63AQtfRtJHVekW/6unXlr1rdtlDxiNicd1NwA6HaTBmHC0tn1L3U23uKhPD0u37viM3kcbLdXWtFVIozjIJDoAWo8cyK+M4sD01p+t+t2PvUQ7bYVeZaNZcMz0RHKlwzhBdBDuQ1YtSjCOnF5a5OXhyVCkiHhDWTUsr0TFGJiaS2cBTqtARJQzwEeTGutDlUd6N7Zf79+/xyf05SVu2+nFUG7rQGshTTlZAXOekm6wFHE3rAVWuZaa4dMjA3fymCuNU2ktl4LVASGhce2/ZrA2zpO9XNE1HrtjmdUk6Hy/MfP4rCwZvf5vK7A2JUqp5rGjciy06o+aC+QPAGTiQ9sh4hjAhMiJXjOlV9MxrYsamIarUsKspw36ciEE42+7pYqCNsO4C0zg5nrVlFLZt57UYbUD3w8R2TDw8O+fug1MePLxgP05U3H7feLUWng94DaGrlukSiVzs47ONr0wy4qqZ/4IqnS3jgUP1QJfTuqSlV6AupPgY7X5iKpkG4+ohQPBsyqbrPOvU0fsWon6I5uhLKZNT0Q78rx9GJToiKkicCM1UnVlWcmwwB55Xsil4HnMG3SXVriDZTSP77Tm7izNk2LFKyRJvQmBqIuu2YRR1R0bR+RIAACAASURBVE04MBnVzES0wit3ZNqsBE1WzFfwyjXKbtgx7M49yWkik4nBAYuaYFELTUAaK34QolelX1SOL1mp85wdrMpDxaxOeOmzH9w5V/ME9XtapfRSwceGyUxPeXK1P1nlqSgw+udpsLT5cdgzToPXEJ0TmiqDNtHf59bbUVth95kmgzWeyjwmy2K1nztTdgZcHGrFXJIPHjUi3/ohoA4iogUAzc01j5qi5uVp4zVL0jk7mlwuY+edU8e897/WRHfWkt0f5Fm0VQI3jTgGP1DV4BGAysAPkrHcjFfCPdVNLbNWPe9n8T5bQMKsHZccDJEJkUCuDJzZPMpibZUl5ImJC9uONTYKkneEdI5Iom3T9WDgMTbmMZ4C250wjntGHXm4HdiPid9/6VXu3W85f/MN0nQL8siUQGjoVifcvPUm1ic3CU1H1sDL9+7xyv1TfvN3PsqvfOi3Od+OPNw5sh+GKVKYdiXBIy9S5U8mWasjCJd9qLMdGpNiShxnBaEJJuEXcCKzl1l4WN/3C4edMfCgi2gSFaiw91JtoW3XsO4tVtikhsgmJ0Qfzl0QYb3quHXb4lgfjgMwRxTYaWM2uQfpLuchmAQe3avuEs6giSEnppzZJ4vOSW7/LCMzyxWCIe0lLs7u87Hf+zC7+6/wdpQbEmhUWZEhCPuuJbXR0QTdxpsL7nTyyjQwaHJ0Sau1OUwjD7YXTDlxMe0ZU2IkM5AhCmHVIW2kWTf0mzWhaWg3nTFpN6HEvqVbr8zu3fc0jmVixSMOzSfzeJqai2KHYSySuqWrG9PM5NGTOIpDbxppQ7BMTWeImhJDmtA8kbIXCHFccBcX7aa+/mRuBEVbFPcFIRi4JUpKgGbGrH5gW61X82eYCWaZ8m6xTG47TwlEyDqRgs3x6L8Zx/m31WxQTDNYLc2yF0SKBkWVQJcUQqSNHaItSToUC2MN6oJNU/DT3Z+kbv+V+bZzDHkJw9RL9zFmaIWfywaeBTQzi5dotjkXIojlBwTUShaqSf3kgrEuB3cgzw7hXIqRF3gFkcX17b5LiZ3yu3EkT6PtNwdgmzyMMMiehnNiFG7evEnXdTwJPXMJvGl8UmNAUiBPYuE3mjjf7iFNnK1bLvYDQZM5d7yggkmkLVb+QNjtBk7Ptzw8P+fB6Tnb/cSgnYcHmfrrN66Sd6FZxp5fw8y87fm8eJZofLMzpWQJhnlSF0BQJoXP0opSpAYPJ8SvgYBL4IaAZ7gNXd9Zuv8UTXRbUGwM1tSSRhaNLxtClSxiUQFAiIkQvZiAf33SYk/WWfo7cCItTQ7q/xvj2J6dciFCOj8j7i7oEW6GQFBlYC5JoVIcRn5PDCt7InOhFvGSpoFx3JP3e/YXZxbRMu4Z8kQKkBpBckO7NtU4NFZwOTiwUHDJW2KRwmfJu5Q+OzCRPYbMljkX5Ahhwax1DrUrUSg5JVfRtErfJskmSrX0EndfsNKXtmWB+YAPVo29SKLVj8JSknU42ak4M0eP4bb0+DJNUqRHXwflYdqVZVamnA6k76uckwXmVV2IKfMolxge3pvyX4F4zUXb9E+D/069lqoQDGHyUHi1Hqu6z0jr/UpGqqIEj68ucnJpozWyZGLjPnTzI6gDx81gW5M7KqmSfW1HMaO6nRvFzaLUvQ5Xm1wKpXFPGoeZgasyBsNiibInhwsajahuXnthLugZM3AObEhW3TyiEplUOdtNjMOE8JA0jKzawFtudjR9Q+xv0K5uIm3P4OBSn3j1Lr/7kY/z8t17DHnyklHJMZDMu1xonozEzKTcduZx12XNFyU1uP3TQsgKo3VJTiLEtpqFlmqbMfjgtkkP8dI51Ctli8gJsTHnW4isVi0xNnRtZNXbdTUlR+lLl8ZRWJ2suf3CbWIX2e63DMPIeLE3D83caVJhoEqFXg1iiRqjOwOnAjoFzGBQUJj2ZflnGgcevPoSev6Q+7HjwcNzsgRueDx8iTxRrXEsTI7ouM0TO00Mmrg/jQyaOZ0GLqaJQRMXyUwo2jfQReKqpz9Zm4PyZENoLTU6dC5Vew5/bOyA77qetu9sTryKjsHBXkap9v7kzLDfst8nNI3s25amadieWUJWNS1lZwAu0ZoEXApQ6IIJlrWVgNHXWqpmAptAFuGGs5xhDkshjRmCmzrUYsdLuFnJZgWL4inSsiydMdltysW5mYrUb4Woi+RdGPmhYPOoWFN8Otb0q0axrBXPSlUYxx1Jozkx3c8zm2aMgZc4ddwePkvW2VuyBCYrh5j5VIqZb3F2VghlKcKbapVnguP+2Ipcwvbmgz5XDcBNiMVmjy7MWhRGPzPv+dWin+MIvj40ZVRgbBumJtB1ynqj9HTcJNE/Mp5X0zNl4CxO58LANVgJrKSJ893EDoMZ3V1subVZcfvGZ0G7IXQnNP0J0jSMWdgNiZfvPuAjH3+JVx88sAgMlKSZrGJ2UvceL0ODUrrEDP3f4oxYVgqZPc6uy9bdJlbhpjGnRkE6K9IU4NVr1EOMCmiTtcWqk0PTCuvWoQL6NX3f0cZA3xpE6bjfGyaEpMMdI7Da9Ny4cwMV5eHpQwiw2w14h+tXC2MecakSocEcY8lj0ZObjw6TVgrpfFP/N40Dp/deITcNDyfh9N4pUSJJWlPbvQ2WbWYq4yBW2/6CiTOd2ObMy9PAXjP388R5TqQoDF2AJrLqbpsZZNOxvnOL2EaatTFyyryEUCOHomPEt+7ENAZuTL7gqVxlaMyaGIc9w35kGvf1EG5c3Z/Giblo8uRCQPk7Z1uWNTZLaEqMJt2FUEwiC1NbdMFgoeGYhAyZZHC7yepzGpDSaNnD7l8OQehWhhxptuRZdFTFoBNqhImHt6Vkfgpn4NXkcnnO1aX+OvOuZSoVD/0qMk3FoH7HcW9YPtJUPPWygLIfAXOyUSalRcX5bOBPiMETq3e64ucoBlOr4k5/qoaCa81Foj6cF7CSdn4A11qlrmktvl8YuA1HMSnOfS+js5S6y7MCgRFSctOa44wDw7ondS3rTUNe9aYBH0R7vTY9YwbOQqUDmFUqAJVMVphUGCZlN2bOtiMhDmyHxJBAgxLGxG6a2E8Tu3Gy5JUYCaoG15jn0L9yUJbNUp2Yi4k94I7lu1AZxcKjNNu/CbZw0Gp3nlVNrb6LrHOI4JQ8ssUPhRAbmrbzakPFuz45TrYlUpATqR2hPWiioZ61FnXRdR05KSFsq8pah9v/ZrFssgykYu9VY6qZeTxeY+Lqza1/2aMZRtIwkKUUURDHOVdj3AiTKKeSGVFOSZyS2JF5qIlRlSFCbiM0kWZtNutus6Zdr2j73rQUr8ZT+VSZNZkjfurfpamrakRSJvewVykz7rfst8OBL6NkQFrGoqvupV+aK8MoELPlwvY7eeSsOIxtLkxJyoACQsYd2j4rlrgzOmOywyA2ccbXcJVe8yIML82SaonsWAowLOzky9jregDUP0WMsaELrsbpAgzr8vKYpj373Tn7Cc63kaSBMXgsNLVODQ4+PINxacGBL+GSk02VlGxYy2lYCmDZnePFalik8royHrXwFKMKuASecjmc53DGAwb+WGHmsqZySUv13+ZsCVtZIHviSFivCOueZt3QrNqK9Pmk9CQVeVbAzwG9f/99qvr3RORdwHuBF4BfAv6yqg5PfGfvWfbU7bJolYiGxtR9n9RdTqR9ZtRE98opD84HPu/hntNBaXJiq3t2+z33Lnbcv9ixnbIVCQhKlmgJEJ6FJiJegFasgrgGWwhTkTwspM1shx5B4kxbFxJ3KYcUQ+OVZhzjRJWUR4TJk0Zat8nZ1VPKTMmkbstwg7Y39MJ2tWZ145ZjECtTVobtBduz+4btkBNBlRdu7tBmiRlhyQvrzQkpKTdu3iQ2LacPzq8edoxZj2KJ/iXxK7tZY5kRetVv7Y7LN82hl1Jm3G4ZOGUKLTRWBECmiZATk8AZmR3Kx5g4R7kfMvfEnDnbVqySzqYnrFra1Yr1rZvEtuXkxk26fkXsWtpVb+YMB6KaAzyhCQV4Sy7Zkx99bbjfhzt7mkbO7t/l4dnFLHHppQNN8LT8UF8fnPmOYWLgW61HvczS+AEDhcpgValFsI15lKPAe1kwavDDPgRWqzWb9QmKSbk5JYZpYnK4iFxMhJ62X5qrWHm9wrgLI6wMCz0cG1kkX2HQx6qZMQ21WtBlhrbdnvLg3sfYDsrLp4kpQRcaR6xUL9o9w0cUH5Dq7JRUt1VLsNqn1kZn4Itx8+hDlol0oZh5lmfRwpdA1Y7CPB/qjvVUDmc9PAOkmFjL7+xi9XtLk8sBaYWT0CaS+wZpW268+Cb627fpW1ivhE4aq8f5hPQkEvgeeLeqnonVxvxvIvLTwN8GvldV3ysiPwB8G/D9T3znRcdm73F5LzgLCSDG5JMa49sNiRis5uTFfqTNgSzB4rzHxOAIaW5YZOmsmkOPgHBJQhMubeZH1WtZWPyq46ZIabpY/NliqCWHmr1VN60WduNhimAMJdpBUCR69QiPcRzZ73aQMw0WJ102/rKpxrTinG0Yi7Nu0ZvFolpKDIVZLy2Ml77+yKiY5kAF8coLRjClRNJACskKV3hRgD3KlswW5YyJc1HOgPNgUtgYGjQKoY3IqiWsWppVb5pF39F0rSfDBA80OJyj2VyxMEcsNh6Lzxd61QGpZtMixoESGlikQrej2QqIgaCxMvNyCxArCSigYS5XZ9mTZQ1Sn+uCiS7NMFpLijlrk8LAbW0Hx54JXpygtFslu807H9riizSq8xyr25yX8dZXTvZitETwBK8yNnNS0WUqDtVxTIxDYkwKsaERS3QKi3R/Pxaq5lmA2NQ1HRszM8jVZKbiq1kw8MWEzwdgDXVcmDiWa6Vo/NWOPvs3Zjs6VldV5+sf2Ff9t49n4Pb1LEJBXZOuIa462lVPjNkykSVwxbJ8LD1JRR4Fzvxl6w8F3g38JX//PcDf51Nh4HU12ROhxNA6s1Qlh0AmMqI83I7sxsRv/N+PkUW5dWPD2972AilPvPLgjLsPz9lPhiZYwk3VoO0MeD9YXLBIoMHwsrNmYmpsI02WxFA2rq0Jm0KJVKlrOekpWcDhQdgQvoDHWcqa51eJIbDqWwiR9eYG3foEEWHnIPrj7sLiq88ecnH/FYIIt26cENv2ytURQkPbWFx80zSmWnuFc1uIHCyMwoCpx8jM1K/UFi9N2UwmH46q7FEe5MQn08AEbLSlAe6nC7Zp4IEmXskT+wB328BehLzaoJs1oWvZ3NoQuobVzQ3dpid2Lf1mUyEULDt2LnktfkAHdyTXaJNLh1epzDK3uGzeR/smKKIjomPhsov43boUjNGk6Bv4MKoli4AmsgiaJ6axmN2Y18JjTFSzYDBrgG5M8abYAdC4BtA2kaZtKhxssWcbPvd8cJcjq8Z8q6eka4kF10facdkZZ5C+lm6fCtrhsHNArEthUQAEREv6PEQ3ada4p2KFkyI5GzcUByXT8q6bw4L7akxYEjOPuuRbfS3LSdV5Pc8+3cLMpcw2BWRu1kCD47ZrRQecdRdn5My4OFduljB/185wIa5WtH1Du16xfssdX983aPsVur9gfPgQJKKbE2ifzI35pFXpI2Ym+Xzg+4DfAu6rFrQWPgK84zG//Xbg2wFu3759+OGCcVemJIU5ah1181IbROrFfmQ/THz0pbvkPPGWF+7QnxgM7en5jrOLHVOGUYvUE+dDORYV2hI5ohhDzjlbYd2sjHnvnmjq6QsWGxok+F6VxQLwE7swlNolj3KhSFl1PHxBWmKOhIZ+vWa1Xnv6tBWr3Z6dMw5b9mcP2T54QNtETvrOsg6vZOBhZt4uiVtZNFxMXkhadY2r92Ee+/ltufI+9ZeXDqsJY+JnOXE/TQQJ3NVMQPlE3nM67bmnIy/lgTEI503HJJF1J6xOVnSrnv7Nt2n7lnVh4I1FmIiEmuy0bG+RQEvBhSXzLnNh8+F9V6r9ezYGXO6YFXwQLXC6Nr9yYOMVT+TxmOGCPjmLzCTXAHKearZgWe6PaFDg1Zo84avAj7ojPTCH34ViaBCH2o3B6nC607lmVTpIVSxOd+a1WdARH2cqu4p5u9Jq2aOoVX4qpd0m0xYvX8mKhsQKXmX4Yn6YaZjXXZVqFytKXbDwtoeSro5pALoIGbRflAN7eaGFk7Gyk2J6nDXoIsJkpOKpEDzmXzjYByWUcn5j1lnriLlwoL7W1OtvhnVHe7JidfOEO299i2mUNAiBcbhge3FBkIbcpwMf12vREzFwNaPTl4nIHeAngC96ssuDqv4g8IMAb3/72w/mOASzR3chMk4jMU0eBeFxltVUURacbdhM5my745W7ypQT/boDUR6cXTBOltywn5Ss4uFexf7pm+MA6wAbcM/SCjES8zzp9nGJ7zYzBwsGnsEkAFlIsAvVuzh3xCe2iVbLL8SGtl97RqlhXKs7otDZZlsiKxQHJEqperUPxtlVS/vOZKWsIrSOTFg827NWIXXZVbvhwt5XpbcrqUTolA1gcd6jwKkmXskjYxaatCMAr0rmvIHz0LCPkRwD7cmapo1sbp5wcnNN13fcWJu5pOs62qYFZ2rgarYs7i+zTTvEgjNiB5dUc1LjB1mYGeOCOT2W3Dk5H3r2mJ17aqaRapooY1FlSGocsWdk2oE3ZwbOPcG1CHdkL0x6ZUVpNmjckhiEWEhpDsHtz1aVvRRsKOXUqlZwhanmcoTR5SiKsgxKG00gtXHI02S44V6+r1TLuXzgm2Tv5gg36aibQtTNILCQvgUzFTG3rzLoRcZpcRrPj5KZujSJXFrANQB97mc1ZRUNJOfFnC7DPAtgFYU/1x4udQffFDZGwbR+iZF2vUKayOr2CasbG9q+hZTJ+5Fhf0EaE+nijHG7RWK7SJx7fXpDUSiqel9EPgB8DXBHRBqXwt8JfPSNXAsMDGa9WkFqmMY9eYIpw+SxnTG0XtOuRJAI2WN4X75/yqt3X2Xzcs9Lr95FgnDvwQO2Q2I/WkWdrEqeTDJqm9YZpxUILnUeoyd1hBDMtZ4b20RFAnKpSDxWtWCVFFu2CQsLu7ZqBf8pcbWCFSQIYuh66/WKpu1Yn9wmNlZCwlLoA10TmVDaJqLZoi00RMONyYZpcSnyscyNJcGkiWGydOzYCP26tc09cLBxFUMbhGJmogosVdq9UgBfaBmhqJ/KqOZd/6ROjGnHhomXxSTxs6AMjaCrHk4siuTGnRu0Xcvtkw23Nye0bcuNzYlV21n1hLYlCQzejCTWTsv6MwkpeuJO07aG9Be84LJEDyHsaNrW63oarnYsAESPX+UHMd5lbFWLk8/7nxcDtGBchSFXCTZ4gYsQPIBJanHo2V5PTf5iOezlECkRJCmRxxERIUUL+5taS3qaJqvVODgeuSEthmqtSGkiL9HxSmPlMdNcO1RcdlAKVqdhz/b81Jx902ARUleYUNTDZg1bZbQsXHGNehnOqGYyrQFdhfEzp8ujWrUgLc5At1HbvWxfBjFk0IXSYYeTml/C3pf5VFK1mGzwpKp5Puu8+DiUN+Vgg1iKf/ABzX6gEIXcNoRVR//im4mrnpt3bnFy66ZB8l5cMI17HrzyChenZ4Q0EMcd9Gvymz+NYFYi8iIwOvNeA38G+B7gA8A3YZEo3wr85BPftVy7PGTxQP2kX4zx4jflvEtZSWNiF0fOLraEIIzT5HZvraD0yQ/6ECLBpZIpTV6ZZqYaI74wJ1QG7mnz6hmfS+VS/H4ZRarJZVYBxVW2on7GECyeuWnoWi+UqhYJEouDrArEerC5y4YojrElHUhVahvFKri4BBg9687NPQfpF6oLpsGCiV8tqVY5VAJacF1cZJuiMJjwwc435NhEUhBC1xL7nqZtaPuOrmvpupa2i7ReNLkwN6SM8xxjUpymlUHKfMjWefK5Okhxrs9ZSLdX08FB7GPDPKtX2s0LlXUxS67FUOOfLDS3yrjL+4v7L+eUIhkW7A7HYMk5VSd5hX5dSKV15Ipwkef3rSMl0mXxVjlAfN1paVcxZdasUn+kyQ+7JVLho6N51UOLffmKnz1Wdjj4BnVMVakRRTEEF8qgLMqC/WJ7yc1SwSAvTJv12HkJtTZnsaRIcSrqvAoOlkBhUHXSfX4bA1MLXWcO+JWZBEtAQxos3HbaD6T9ADoRU6oZmk9KTyKBvw14j9vBA/DvVPWnRORXgfeKyD8Efhn44Se+q1OtSp5NPWvcDBE0gQpRsjlNRM32vBggKyDcM6ny4HQLYkw7q1VvL7AGhhNhEoiImKQyDLP91P8WG2TpZIzRoVMEJLvZhRrdUWK1TcpWK4iw3zsTt0QDEa3hbOvO7NMn6xU3NmvaruPmrQ0xtgxTYkhKRNl5EdZp2LPfblFV1usNqBWSHfeZ4XYPl3K1VDNJJzIJiUpooN80hhMzZabRQzNLXDBSJe9Ubb06gx4daHFlI/u9ypMQnYEbww6CHUhNJDUtU7+miZEb/ZqmaWlXHf1mRWwC65OOpg2su4Z1H920ZCaRJJ5kBSRHpNGwgCmIBX9moeQ7Q68JVPVvqNpWDSN8DS5s8fmJ0cuJlXsUs1M9NIp5C6mp+aa1OeMukKPlM7er1t8W/R0ODguLeLGDuCSiyDQhaULTRNrvETG3Zmwzoe1q6nwuphNnwKhWW3iVLmtH50Nx2feaD+EMrky7CAzDnmm/Yxi2pHFPKb93SW9YLpsSWGImQTW4IImm7FY/RT2ZZ4Zcxn22Vy9VQwuzLUJWkEDX9qa9iVSHZs3g9GgV2+s2PwXLW31PLLsw+6xmgWwcJ8f3NmkepAoTpimqze9qhcTI6tZNNrdvE9uOze1bxKZl3O04e+lVpu2O7d27pGFk2u0MDVQyjYhVI3vs6nyUniQK5YPAl1/x/m8DX/UG7vXotcFVQ0HUQuQSeEmsIidaunde7NV50zSkNDHu94DWqATLc1jayCxGuwb9l4gCXyQxesV1EfpoOOAWQ1oiTvzhJblCDHSdoSCWuF1B2Q9+ANXFZhSC0DYON9o19C55rnvDuJZBkCkzjRYfi2ZympjGgYDSdC2aEuN+53bOiF5m4EXt1OzFEaBpLfMtx0wMBchIPZloTioK1d7LgQS+lCTLOFTm7QwJl3iDmwdCGy3OtWnJ6x6NDf36hFW7outbNpsVIQqrVSQ2QtcKbVPKnNm1SiKRKahzAnVRfZdSd1WHfT5rDcIwf7dU/Dlw5nFoA14uyoIHgjOxeu96j5KRWxhyXEjXM1M5bKOpYfV1HdDiYC0S8czEU2HGaUKmCZ0MKRJA2hGCVMZdCjHoYm2rV9SpMcpLyW6h5V1FUtpVTRiQpolhv2OchpqFurzWoxeZHxKqQEypg1kcv0sH9bzaIiznR4QS8CqFn/v8xBBZdb3tpYUGNCcuzdmlZZ02lYFDvnSrMi4lfNTi6BWV7ENYHOkFw0iZwPhG15imeWPD+s4tmrZjszkhhMh0vmU4u2A4v+Di3il5HAlqNXiDKDHiVeofM55X0LMt6KAFYtTKJJSq7E1Z3x4LWmdYyqbz4VvYc+164htwfv6IWrl4DjaZVszVcpCS2IHRlFJcIdB1VtYrtl7pWkwtM0dVwQ0xR1Oo2oLUjVigUkvVHkd+QNNoKeuT2SfztGca9kzDQJ4m60jwECk3B9TY9svkmyQ2ga5rCCETxOBbNanhaVQJ3M0+RUNH68J0weYRqNyi6oPUECkJ4tIwdVysgk5k1XbcXq1pQsNJv6ZrWrq2ZdU3fgBask1sBBpBQ2CKxnQnYFItJUhRStECw5eJByBhxWyykHQX41McdyAek5/9N1ytpwsW0RLmaJZF2cZZei73hAXLqPyp3H2hXStcFoIXB2epk1kBpRxbRVWRcYBkle7zsDfNLwRiMgl8HKxKe57cxJJytXdTmPdVpPMprYs2l37EIIjmmgk8DntGhxwuh/3BYfrI9QXJ9pglBpdeK8Y8oJGCxnl5LLWoCqWu69LAkt2hq+o2eLeZl18W5sscTFBqPWQMXx2V4kud140uZG+PMzdwiUtYLGU7xEjbRqSN9LdvEFcd/aZHAqQ0cHbP/AQX9x+we3BqGdVpRLz+bunxY4WK16Bny8DdhJKSVhMKYmXizEExUcJ2TAi4zMTLM0+SmJE5ayDBcjwuJyyURbesyk0eISerndhZWvt6s7HoCLQyLKV1ocbqMaLJMstQxKuqJytBb3gjnkkWxVDVRBM67sk5kMeJPCam/ZZhe2FlmcYBzQkwnOvs0qXmAqh1SEVDaNrAat1Z9mkrrlbrXLi1mFB0rlyi9RolVGuWY+qzak9mjr+OjlcuVtosBqHrO7q2pW8abvVrk46a1qq/hEjv+O1Na6nwOZgElCUwRGPCo1oWqorjZMiM0hdDnEuU1YiTIgnLzMQXc25+A8tyFCm4iOFQIq1jIDNuyoH93K5ZzSpSEN0Xvz1Y3M4pDgSQRXq2apWQTUOciw+XAsPjOFhW5jTYIb9wYk6qFmLZtHTrve8lL8gwWVRK6c/jVXJ1IaeypIP11LhwMw5WoHm/37LbbbGTaN5DCzfn4VgqhCxmjksKk6JYvVIHcfAVpj4jAXHfVFlzujTTlAiSImWUjFUJTONgfoGDA9x7pQt+4c9DmoHpStsP4IUXl7D9kqzthbELXuBDkK6lv9ETuo4bb7lDt1n7mlGm/cjpJ+8y7faMZxekiy3kjCQLsZ0DJnQxD09OzxYPXJR1MDVCxGoaJs2MpSBBcQqLH9ClirUA1RwlqAP4lAFITWSMBd/BAd9LJXLF3lP1MD0hJ/HvC5oUtFSwNmbRd0LTQNtA1xgoUS8TIpEUJssSjYnY+qb0jZEjTEGIATZNpo2JlUx0DERNxCQEDTSehLFiYBMn2maCzowkMdh9NQv7HNAUWDWPbpYmNvRth+Q1aX3DVOt2pNR31HSZgTPbXgEo4VcL15vUTxbmACrGSQhKiPgBZWanrm1p2oY+Nqa5hEATMdG6OgAACFRJREFUW3cuRSQaY6SgwZW9HwJIRB0Jsty7QI4KESkSmqW0LQ5sM7agQpKEBrVaosmKKJDN7KEp08SmAqfttlsLf1uuyRi5eWJwngcOx6UJpZhVLkmL9XllDEtlYGbk1caqhYEv8EhSqqBOk1eW12n0VPhEHh0OYnNCbDvWN064se7RbCJFduycampZzOOihTZuB0JQZU0AdO5o15QY+4acEm1QhlVPxYGB6lM42Zw8IoXHpqXtN2SZ2IwNU8rgWGJ213KYRl91VvS3DJxQJOByWBSTjTtZsTUsIsTOkvIOD/CZcRepulSDLyauMkbFfFMncqmhKcRmpKmlzowHqWcES98SeoucWjcdbWjrNYNk+tAQY6Zpe1LPbCJWrWGjUZRGlN6FnielZ8rAbzYDX7i5Z0kP1eEwQ0QWlf4gGKJMogJE0IDqYYdntXl+vdyElyXwRyI41CI4CgZKjEVNhxCKA3R0AWs20eRNvtQOULWFEuOeICMh7oj5IaJCzHb9iqESMm+9bZs23c7kIm2LDULOK1SVm3042JRBhNvr22y6EwM8uvNOl/LyrIIWTboOny6e+9AuF/CVzGn+XMrLA0aL25qLbbJgyUhlauUQyCKHsd0uOQlCg5mbdHlbDSbJ5YJJA9NUnMWpblyR3YHUXJ2GzM7D0paUEtuLi4M5u33zBl/xJX/YSr8tOnrAmqT+8wg9XtpdfLiQ/OfIkNkGPjOb2T6+tJUDNWGpAKApaky+SPZLM+Fj2lH50xUiX/UZlL3kZoqcHkUfFLEiBEu4ZpHAye230G9ukrPy4pSrUjLffbnylpxzYc1YLtp67UX7q52+7Am5dI/Dp0Vkucrk81q6Sl4cJEvSovlF1wrbxg8k/7xRphdu1hh+LcUgqghQDhE7oEKIrNY3HtuOy/RMGXgrmTvt/jW+8XrKxKz+PD0qTPqKYOwrs6dK+9L8m7J2L8frC9BdfmNJjzuZha7p6ZonRRF+9vRaMyvymFldSESHKv8bsxu+FnVty4svvOnTdr3/n6nt17T9+lk349lSAFafuX35WhkNRzrSkY50pOeYjgz8SEc60pGuKR0Z+JGOdKQjXVOSxwXyf0ZuJvIycA688tRu+pmht3C9+3Dd2w/Xvw/Xvf1w/ftwndr/h1T1xctvPlUGDiAiv6iqX/lUb/pppuveh+vefrj+fbju7Yfr34fr3n44mlCOdKQjHena0pGBH+lIRzrSNaVnwcB/8Bnc89NN170P1739cP37cN3bD9e/D9e9/U/fBn6kIx3pSEf69NDRhHKkIx3pSNeUnioDF5GvFZEPiciHReQ7n+a9PxUSkc8RkQ+IyK+KyP8Rke/w998sIj8jIr/pf5/r3GsRiSLyyyLyU/76XSLyCz4P/1ZEute7xrMkEbkjIu8TkV8XkV8Tka+5hnPwt3wN/YqI/JiIrJ7neRCRfykinxSRX1m8d+WYi9E/9358UES+4tm1fKbH9OEf+Tr6oIj8hFid3/LZd3kfPiQif/bZtPqN0VNj4GIVfb4P+Drgi4G/KCJf/LTu/ynSBPwdVf1i4KuBv+5t/k7gZ1X1C4Cf9dfPM30H8GuL198DfK+qfj5wD/i2Z9KqJ6d/BvwnVf0i4I9ifbk2cyAi7wD+BvCVqvqlGKjNN/N8z8OPAF976b3HjfnXAV/gj28Hvv8ptfH16Ed4tA8/A3ypqv4R4DeA7wLwff3NwJf4b/6F86znmp6mBP5VwIdV9bdVdcBqaX7jU7z/GyZV/biq/k9/fooxjndg7X6Pf+09wF94Ni18fRKRdwJ/Hvghfy3Au4H3+Vee9/bfBv4EXrJPVQdVvc81mgOnBliLSANsgI/zHM+Dqv4ccPfS248b828E/rUa/TxW8PxtT6elj6er+qCq/1mtEDvAz2MF2cH68F5V3avq7wAf5g9Ycexp0NNk4O8Afn/x+iP+3rUgEflcrLTcLwCfraof948+AXz2M2rWk9A/Bf4uM/bhC8D9xSJ+3ufhXcDLwL9yM9APicgJ12gOVPWjwD8Gfg9j3A+AX+J6zQM8fsyv697+a8BP+/Nr2YejE/MJSERuAP8B+Juq+nD5mZYSH88hicjXA59U1V961m35A1ADfAXw/ar65RgUw4G55HmeAwC3FX8jdhi9HTjhUdX+WtHzPuavRyLy3ZiJ9EefdVv+IPQ0GfhHgc9ZvH6nv/dck4i0GPP+UVX9cX/7paIi+t9PPqv2vQ79MeAbROR3MZPVuzF78h1X5eH5n4ePAB9R1V/w1+/DGPp1mQOAPw38jqq+rKoj8OPY3FyneYDHj/m12tsi8leBrwe+Rec46mvVh0JPk4H/D+AL3PPeYQ6D9z/F+79hcnvxDwO/pqr/ZPHR+4Fv9effCvzk027bk5CqfpeqvlNVPxcb7/+qqt8CfAD4Jv/ac9t+AFX9BPD7IvKF/tafAn6VazIHTr8HfLWIbHxNlT5cm3lwetyYvx/4Kx6N8tXAg4Wp5bkiEflazKT4Daq6LMf0fuCbRaQXkXdhDtn//iza+IZoWRLsM/0A/hzm+f0t4Luf5r0/xfb+cUxN/CDwv/zx5zA78s8Cvwn8F+DNz7qtT9CXPwn8lD//PGxxfhj490D/rNv3Om3/MuAXfR7+I/Cm6zYHwD8Afh34FeDfAP3zPA/Aj2H2+hHTgr7tcWOOFVD6Pt/X/xuLtnle+/BhzNZd9vMPLL7/3d6HDwFf96zb/ySPYybmkY50pCNdUzo6MY90pCMd6ZrSkYEf6UhHOtI1pSMDP9KRjnSka0pHBn6kIx3pSNeUjgz8SEc60pGuKR0Z+JGOdKQjXVM6MvAjHelIR7qmdGTgRzrSkY50Ten/AWyWt5m08snnAAAAAElFTkSuQmCC\n"
+          },
+          "metadata": {
+            "needs_background": "light"
+          }
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "GroundTruth:  cat   ship  ship  plane\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "net = Net()\n",
+        "net.load_state_dict(torch.load(PATH))"
+      ],
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "T4FAA5xVCjdJ",
+        "outputId": "1e36594f-c9f8-4852-f4e7-a8f23e7bd78c"
+      },
+      "execution_count": 9,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/plain": [
+              "<All keys matched successfully>"
+            ]
+          },
+          "metadata": {},
+          "execution_count": 9
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "outputs = net(images)"
+      ],
+      "metadata": {
+        "id": "_XV7L3nQCv8o"
+      },
+      "execution_count": 10,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "_, predicted = torch.max(outputs, 1)\n",
+        "\n",
+        "print('Predicted: ', ' '.join(f'{classes[predicted[j]]:5s}'\n",
+        "                              for j in range(4)))"
+      ],
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "QL_gLuJ4Cv_O",
+        "outputId": "58eb3752-720b-49c4-f8ab-e7b766b8b4f1"
+      },
+      "execution_count": 11,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Predicted:  cat   car   car   plane\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "correct = 0\n",
+        "total = 0\n",
+        "# since we're not training, we don't need to calculate the gradients for our outputs\n",
+        "with torch.no_grad():\n",
+        "    for data in testloader:\n",
+        "        images, labels = data\n",
+        "        # calculate outputs by running images through the network\n",
+        "        outputs = net(images)\n",
+        "        # the class with the highest energy is what we choose as prediction\n",
+        "        _, predicted = torch.max(outputs.data, 1)\n",
+        "        total += labels.size(0)\n",
+        "        correct += (predicted == labels).sum().item()\n",
+        "\n",
+        "print(f'Accuracy of the network on the 10000 test images: {100 * correct // total} %')"
+      ],
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "oN8AQJ9MCwBi",
+        "outputId": "89210846-4c63-4183-b13e-e9eb8da3d70c"
+      },
+      "execution_count": 12,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Accuracy of the network on the 10000 test images: 55 %\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "# prepare to count predictions for each class\n",
+        "correct_pred = {classname: 0 for classname in classes}\n",
+        "total_pred = {classname: 0 for classname in classes}\n",
+        "\n",
+        "# again no gradients needed\n",
+        "with torch.no_grad():\n",
+        "    for data in testloader:\n",
+        "        images, labels = data\n",
+        "        outputs = net(images)\n",
+        "        _, predictions = torch.max(outputs, 1)\n",
+        "        # collect the correct predictions for each class\n",
+        "        for label, prediction in zip(labels, predictions):\n",
+        "            if label == prediction:\n",
+        "                correct_pred[classes[label]] += 1\n",
+        "            total_pred[classes[label]] += 1\n",
+        "\n",
+        "\n",
+        "# print accuracy for each class\n",
+        "for classname, correct_count in correct_pred.items():\n",
+        "    accuracy = 100 * float(correct_count) / total_pred[classname]\n",
+        "    print(f'Accuracy for class: {classname:5s} is {accuracy:.1f} %')"
+      ],
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "-r-2ZeavCwGN",
+        "outputId": "791e75b6-b7d2-4e5a-d69f-eb4772add792"
+      },
+      "execution_count": 13,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Accuracy for class: plane is 60.6 %\n",
+            "Accuracy for class: car   is 68.7 %\n",
+            "Accuracy for class: bird  is 39.0 %\n",
+            "Accuracy for class: cat   is 43.3 %\n",
+            "Accuracy for class: deer  is 43.7 %\n",
+            "Accuracy for class: dog   is 51.9 %\n",
+            "Accuracy for class: frog  is 63.4 %\n",
+            "Accuracy for class: horse is 63.2 %\n",
+            "Accuracy for class: ship  is 52.8 %\n",
+            "Accuracy for class: truck is 71.6 %\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')\n",
+        "\n",
+        "# Assuming that we are on a CUDA machine, this should print a CUDA device:\n",
+        "\n",
+        "print(device)"
+      ],
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "krOTFV38CwJK",
+        "outputId": "13bbd8da-71aa-447b-de08-2734244841fd"
+      },
+      "execution_count": 14,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "cpu\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "net.to(device)"
+      ],
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "GCamA75kCwMB",
+        "outputId": "a680b7d1-e0bc-4ce0-e8c6-eb251a55feb6"
+      },
+      "execution_count": 15,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/plain": [
+              "Net(\n",
+              "  (conv1): Conv2d(3, 6, kernel_size=(5, 5), stride=(1, 1))\n",
+              "  (pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)\n",
+              "  (conv2): Conv2d(6, 16, kernel_size=(5, 5), stride=(1, 1))\n",
+              "  (fc1): Linear(in_features=400, out_features=120, bias=True)\n",
+              "  (fc2): Linear(in_features=120, out_features=84, bias=True)\n",
+              "  (fc3): Linear(in_features=84, out_features=10, bias=True)\n",
+              ")"
+            ]
+          },
+          "metadata": {},
+          "execution_count": 15
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "inputs, labels = data[0].to(device), data[1].to(device)"
+      ],
+      "metadata": {
+        "id": "rgRb2ncwC7JN"
+      },
+      "execution_count": 16,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        ""
+      ],
+      "metadata": {
+        "id": "lwmGSBsNC7LX"
+      },
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        ""
+      ],
+      "metadata": {
+        "id": "JcpBntSWC7Nh"
+      },
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        ""
+      ],
+      "metadata": {
+        "id": "44EpJRyvC7QS"
+      },
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        ""
+      ],
+      "metadata": {
+        "id": "JZySeHD8C7Th"
+      },
+      "execution_count": null,
+      "outputs": []
+    }
+  ]
+}
diff --git a/Week8_preview_Transfer_Learning.ipynb b/Week8_preview_Transfer_Learning.ipynb
new file mode 100644
index 0000000..243b338
--- /dev/null
+++ b/Week8_preview_Transfer_Learning.ipynb
@@ -0,0 +1,1378 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "gQaz-4aNQWB4"
+      },
+      "source": [
+        "### **전이학습(Transfer Learning)으로 학습하기** \n",
+        ": 충분한 크기의 데이터셋을 확보하기 어려울때 매우 큰 데이터셋에서 합성곱 신경망을 미리 학습하고 관심있는 작업을 위한 초기 설정/고정된 특징 추출기로 이를 사용함 \n",
+        "\n",
+        "### **전이학습의 과정**\n",
+        "1.ImageNet 데이터셋과 같은 대규모 데이터셋으로 신경망 학습시켜초기화하기\n",
+        "\n",
+        "2.마지막 와전히 연결된 계층 제외 고정된 특징 추출기로써 합성곱 신경망 사용하기"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "bGaPE2EuRcei"
+      },
+      "outputs": [],
+      "source": [
+        "# License: BSD\n",
+        "# Author: Sasank Chilamkurthy\n",
+        "\n",
+        "from __future__ import print_function, division\n",
+        "\n",
+        "import torch\n",
+        "import torch.nn as nn\n",
+        "import torch.optim as optim\n",
+        "from torch.optim import lr_scheduler\n",
+        "import torch.backends.cudnn as cudnn\n",
+        "import numpy as np\n",
+        "import torchvision\n",
+        "from torchvision import datasets, models, transforms\n",
+        "import matplotlib.pyplot as plt\n",
+        "import time\n",
+        "import os\n",
+        "import copy\n",
+        "\n",
+        "cudnn.benchmark = True\n",
+        "plt.ion()   # 대화형 모드"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "### 1. 데이터 불러오기\n",
+        "-  ImageNet에서 개미/벌에 대한 일부 학습데이터셋"
+      ],
+      "metadata": {
+        "id": "N9PKgerOSZMx"
+      }
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "u-N6utS4G4O1",
+        "outputId": "976f8cff-8888-497b-d37c-b7f1b712134e"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Mounted at /content/drive\n"
+          ]
+        }
+      ],
+      "source": [
+        "from google.colab import drive\n",
+        "drive.mount('/content/drive', force_remount=True)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "6y2lMVDtG7sX",
+        "outputId": "40a9a609-b725-402f-9129-fa6f193da987"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "/content/drive/My Drive/hymenoptera_data\n"
+          ]
+        }
+      ],
+      "source": [
+        "%cd /content/drive/My\\ Drive/hymenoptera_data"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "0qEanSjxDbgE",
+        "outputId": "30726a7d-ef94-4ed4-cf42-f0389523d269"
+      },
+      "outputs": [
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py:490: 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",
+            "  cpuset_checked))\n"
+          ]
+        }
+      ],
+      "source": [
+        "# 학습을 위해 데이터 증가(augmentation) 및 일반화(normalization)\n",
+        "# 검증을 위한 일반화\n",
+        "data_transforms = {\n",
+        "    'train': transforms.Compose([\n",
+        "        transforms.RandomResizedCrop(224),\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(256),\n",
+        "        transforms.CenterCrop(224),\n",
+        "        transforms.ToTensor(),\n",
+        "        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])\n",
+        "    ]),\n",
+        "}\n",
+        "\n",
+        "data_dir = '/content/drive/MyDrive/hymenoptera_data'\n",
+        "image_datasets = {x: datasets.ImageFolder(os.path.join(data_dir, x),\n",
+        "                                          data_transforms[x])\n",
+        "                  for x in ['train', 'val']}\n",
+        "dataloaders = {x: torch.utils.data.DataLoader(image_datasets[x], batch_size=4,\n",
+        "                                             shuffle=True, num_workers=4)\n",
+        "              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",
+        "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "### 2.  일부 학습용 이미지 시각화하기"
+      ],
+      "metadata": {
+        "id": "GI14VU2zSkv_"
+      }
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 204
+        },
+        "id": "5YEOiCkeDbiO",
+        "outputId": "3fd8254d-5c64-4ab9-a8e7-caf552c6e9c7"
+      },
+      "outputs": [
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py:490: 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",
+            "  cpuset_checked))\n"
+          ]
+        },
+        {
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "def imshow(inp, title=None):\n",
+        "    \"\"\"Imshow for Tensor.\"\"\"\n",
+        "    inp = inp.numpy().transpose((1, 2, 0))\n",
+        "    mean = np.array([0.485, 0.456, 0.406])\n",
+        "    std = np.array([0.229, 0.224, 0.225])\n",
+        "    inp = std * inp + mean\n",
+        "    inp = np.clip(inp, 0, 1)\n",
+        "    plt.imshow(inp)\n",
+        "    if title is not None:\n",
+        "        plt.title(title)\n",
+        "    plt.pause(0.001)  # 갱신이 될 때까지 잠시 기다립니다.\n",
+        "\n",
+        "\n",
+        "# 학습 데이터의 배치를 얻습니다.\n",
+        "inputs, classes = next(iter(dataloaders['train']))\n",
+        "\n",
+        "# 배치로부터 격자 형태의 이미지를 만듭니다.\n",
+        "out = torchvision.utils.make_grid(inputs)\n",
+        "\n",
+        "imshow(out, title=[class_names[x] for x in classes])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "### 3. 모델 학습하기 \n",
+        "- 하급률(learning rate)관리\n",
+        "- 최적의 모델 구하기 "
+      ],
+      "metadata": {
+        "id": "H-fTw3BMSqlD"
+      }
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "mi5dVSfzDbkf"
+      },
+      "outputs": [],
+      "source": [
+        "def train_model(model, criterion, optimizer, scheduler, num_epochs=25):\n",
+        "    since = time.time()\n",
+        "\n",
+        "    best_model_wts = copy.deepcopy(model.state_dict())\n",
+        "    best_acc = 0.0\n",
+        "\n",
+        "    for epoch in range(num_epochs):\n",
+        "        print(f'Epoch {epoch}/{num_epochs - 1}')\n",
+        "        print('-' * 10)\n",
+        "\n",
+        "        # 각 에폭(epoch)은 학습 단계와 검증 단계를 갖습니다.\n",
+        "        for phase in ['train', 'val']:\n",
+        "            if phase == 'train':\n",
+        "                model.train()  # 모델을 학습 모드로 설정\n",
+        "            else:\n",
+        "                model.eval()   # 모델을 평가 모드로 설정\n",
+        "\n",
+        "            running_loss = 0.0\n",
+        "            running_corrects = 0\n",
+        "\n",
+        "            # 데이터를 반복\n",
+        "            for inputs, labels in dataloaders[phase]:\n",
+        "                inputs = inputs.to(device)\n",
+        "                labels = labels.to(device)\n",
+        "\n",
+        "                # 매개변수 경사도를 0으로 설정\n",
+        "                optimizer.zero_grad()\n",
+        "\n",
+        "                # 순전파\n",
+        "                # 학습 시에만 연산 기록을 추적\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",
+        "                    # 학습 단계인 경우 역전파 + 최적화\n",
+        "                    if phase == 'train':\n",
+        "                        loss.backward()\n",
+        "                        optimizer.step()\n",
+        "\n",
+        "                # 통계\n",
+        "                running_loss += loss.item() * inputs.size(0)\n",
+        "                running_corrects += torch.sum(preds == labels.data)\n",
+        "            if phase == 'train':\n",
+        "                scheduler.step()\n",
+        "\n",
+        "            epoch_loss = running_loss / dataset_sizes[phase]\n",
+        "            epoch_acc = running_corrects.double() / dataset_sizes[phase]\n",
+        "\n",
+        "            print(f'{phase} Loss: {epoch_loss:.4f} Acc: {epoch_acc:.4f}')\n",
+        "\n",
+        "            # 모델을 깊은 복사(deep copy)함\n",
+        "            if phase == 'val' and epoch_acc > best_acc:\n",
+        "                best_acc = epoch_acc\n",
+        "                best_model_wts = copy.deepcopy(model.state_dict())\n",
+        "\n",
+        "        print()\n",
+        "\n",
+        "    time_elapsed = time.time() - since\n",
+        "    print(f'Training complete in {time_elapsed // 60:.0f}m {time_elapsed % 60:.0f}s')\n",
+        "    print(f'Best val Acc: {best_acc:4f}')\n",
+        "\n",
+        "    # 가장 나은 모델 가중치를 불러옴\n",
+        "    model.load_state_dict(best_model_wts)\n",
+        "    return model"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "### 4. 모델 예측값 시각화하기 "
+      ],
+      "metadata": {
+        "id": "TaXWj0tNSztx"
+      }
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "nDATmF47Dbmf"
+      },
+      "outputs": [],
+      "source": [
+        "def visualize_model(model, num_images=6):\n",
+        "    was_training = model.training\n",
+        "    model.eval()\n",
+        "    images_so_far = 0\n",
+        "    fig = plt.figure()\n",
+        "\n",
+        "    with torch.no_grad():\n",
+        "        for i, (inputs, labels) in enumerate(dataloaders['val']):\n",
+        "            inputs = inputs.to(device)\n",
+        "            labels = labels.to(device)\n",
+        "\n",
+        "            outputs = model(inputs)\n",
+        "            _, preds = torch.max(outputs, 1)\n",
+        "\n",
+        "            for j in range(inputs.size()[0]):\n",
+        "                images_so_far += 1\n",
+        "                ax = plt.subplot(num_images//2, 2, images_so_far)\n",
+        "                ax.axis('off')\n",
+        "                ax.set_title(f'predicted: {class_names[preds[j]]}')\n",
+        "                imshow(inputs.cpu().data[j])\n",
+        "\n",
+        "                if images_so_far == num_images:\n",
+        "                    model.train(mode=was_training)\n",
+        "                    return\n",
+        "        model.train(mode=was_training)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "### 5. 합성곱 신경망 미세조정(finetuning)\n",
+        "- 미리 학습한 모델에서 마지막 완전히 연결된 계층 초기화하기"
+      ],
+      "metadata": {
+        "id": "XPM-MNE1S3fC"
+      }
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 67,
+          "referenced_widgets": [
+            "c07c0605a77241628f499fbd3b39e9c5",
+            "ae060ba608934a5db38fddd462e60376",
+            "cc59758590f64205afbc980318fced25",
+            "05ab70175635434db813e1ca278816d6",
+            "ade88339e5c64e7091363b4c6c02d22c",
+            "f92f8a50f06d42eb818f1ea94136cbab",
+            "29db66ae692f4d71b436f4fc3a1678ec",
+            "17d1def0ff0344ea9c2809a213dae0b3",
+            "7b02c667a1da4fbc99e580fb6110ed7d",
+            "9b5d02cfd948481da3fbaee3e46b3698",
+            "f4f63484744a44fd90ded2ba9703a835"
+          ]
+        },
+        "id": "iThnR64rDbo2",
+        "outputId": "2ef1c4d4-1530-489e-bca8-fb26eaa93819"
+      },
+      "outputs": [
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "Downloading: \"https://download.pytorch.org/models/resnet18-f37072fd.pth\" to /root/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth\n"
+          ]
+        },
+        {
+          "data": {
+            "application/vnd.jupyter.widget-view+json": {
+              "model_id": "c07c0605a77241628f499fbd3b39e9c5",
+              "version_major": 2,
+              "version_minor": 0
+            },
+            "text/plain": [
+              "  0%|          | 0.00/44.7M [00:00<?, ?B/s]"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "model_ft = models.resnet18(pretrained=True)\n",
+        "num_ftrs = model_ft.fc.in_features\n",
+        "# 여기서 각 출력 샘플의 크기는 2로 설정합니다.\n",
+        "# 또는, nn.Linear(num_ftrs, len (class_names))로 일반화할 수 있습니다.\n",
+        "model_ft.fc = nn.Linear(num_ftrs, 2)\n",
+        "\n",
+        "model_ft = model_ft.to(device)\n",
+        "\n",
+        "criterion = nn.CrossEntropyLoss()\n",
+        "\n",
+        "# 모든 매개변수들이 최적화되었는지 관찰\n",
+        "optimizer_ft = optim.SGD(model_ft.parameters(), lr=0.001, momentum=0.9)\n",
+        "\n",
+        "# 7 에폭마다 0.1씩 학습률 감소\n",
+        "exp_lr_scheduler = lr_scheduler.StepLR(optimizer_ft, step_size=7, gamma=0.1)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "### 6. 학습 및 평가하기"
+      ],
+      "metadata": {
+        "id": "MZycy0ueTAmw"
+      }
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "background_save": true,
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "0D-usIQADbrR",
+        "outputId": "f52448e1-a6b6-4d34-a3d0-6bd243dadc6b"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Epoch 0/24\n",
+            "----------\n"
+          ]
+        },
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py:490: 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",
+            "  cpuset_checked))\n"
+          ]
+        },
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "train Loss: 0.6582 Acc: 0.7418\n",
+            "val Loss: 0.2540 Acc: 0.8889\n",
+            "\n",
+            "Epoch 1/24\n",
+            "----------\n",
+            "train Loss: 0.5127 Acc: 0.7746\n",
+            "val Loss: 0.4802 Acc: 0.8366\n",
+            "\n",
+            "Epoch 2/24\n",
+            "----------\n",
+            "train Loss: 0.5363 Acc: 0.8033\n",
+            "val Loss: 0.3488 Acc: 0.8693\n",
+            "\n",
+            "Epoch 3/24\n",
+            "----------\n",
+            "train Loss: 0.3828 Acc: 0.8320\n",
+            "val Loss: 0.3105 Acc: 0.8758\n",
+            "\n",
+            "Epoch 4/24\n",
+            "----------\n",
+            "train Loss: 0.6727 Acc: 0.7828\n",
+            "val Loss: 0.4062 Acc: 0.8954\n",
+            "\n",
+            "Epoch 5/24\n",
+            "----------\n",
+            "train Loss: 0.6912 Acc: 0.7582\n",
+            "val Loss: 0.3483 Acc: 0.9150\n",
+            "\n",
+            "Epoch 6/24\n",
+            "----------\n",
+            "train Loss: 0.4294 Acc: 0.8320\n",
+            "val Loss: 0.2568 Acc: 0.9150\n",
+            "\n",
+            "Epoch 7/24\n",
+            "----------\n",
+            "train Loss: 0.4225 Acc: 0.8197\n",
+            "val Loss: 0.2391 Acc: 0.9150\n",
+            "\n",
+            "Epoch 8/24\n",
+            "----------\n",
+            "train Loss: 0.2979 Acc: 0.8770\n",
+            "val Loss: 0.2949 Acc: 0.8954\n",
+            "\n",
+            "Epoch 9/24\n",
+            "----------\n",
+            "train Loss: 0.3632 Acc: 0.8238\n",
+            "val Loss: 0.2423 Acc: 0.9150\n",
+            "\n",
+            "Epoch 10/24\n",
+            "----------\n",
+            "train Loss: 0.3479 Acc: 0.8402\n",
+            "val Loss: 0.2775 Acc: 0.9281\n",
+            "\n",
+            "Epoch 11/24\n",
+            "----------\n",
+            "train Loss: 0.2997 Acc: 0.8770\n",
+            "val Loss: 0.2165 Acc: 0.9150\n",
+            "\n",
+            "Epoch 12/24\n",
+            "----------\n",
+            "train Loss: 0.3597 Acc: 0.8443\n",
+            "val Loss: 0.1950 Acc: 0.9346\n",
+            "\n",
+            "Epoch 13/24\n",
+            "----------\n",
+            "train Loss: 0.3496 Acc: 0.8484\n",
+            "val Loss: 0.2347 Acc: 0.9216\n",
+            "\n",
+            "Epoch 14/24\n",
+            "----------\n",
+            "train Loss: 0.3476 Acc: 0.8484\n",
+            "val Loss: 0.1903 Acc: 0.9281\n",
+            "\n",
+            "Epoch 15/24\n",
+            "----------\n",
+            "train Loss: 0.2872 Acc: 0.8852\n",
+            "val Loss: 0.2562 Acc: 0.9150\n",
+            "\n",
+            "Epoch 16/24\n",
+            "----------\n",
+            "train Loss: 0.2511 Acc: 0.8934\n",
+            "val Loss: 0.1979 Acc: 0.9216\n",
+            "\n",
+            "Epoch 17/24\n",
+            "----------\n",
+            "train Loss: 0.2962 Acc: 0.8689\n",
+            "val Loss: 0.2066 Acc: 0.9150\n",
+            "\n",
+            "Epoch 18/24\n",
+            "----------\n",
+            "train Loss: 0.2971 Acc: 0.8566\n",
+            "val Loss: 0.2147 Acc: 0.9281\n",
+            "\n",
+            "Epoch 19/24\n",
+            "----------\n",
+            "train Loss: 0.3063 Acc: 0.8689\n",
+            "val Loss: 0.2415 Acc: 0.9150\n",
+            "\n",
+            "Epoch 20/24\n",
+            "----------\n",
+            "train Loss: 0.3107 Acc: 0.8648\n",
+            "val Loss: 0.2346 Acc: 0.9150\n",
+            "\n",
+            "Epoch 21/24\n",
+            "----------\n",
+            "train Loss: 0.3151 Acc: 0.8811\n",
+            "val Loss: 0.2536 Acc: 0.9216\n",
+            "\n",
+            "Epoch 22/24\n",
+            "----------\n",
+            "train Loss: 0.2944 Acc: 0.8770\n",
+            "val Loss: 0.2023 Acc: 0.9216\n",
+            "\n",
+            "Epoch 23/24\n",
+            "----------\n",
+            "train Loss: 0.2723 Acc: 0.8975\n",
+            "val Loss: 0.2828 Acc: 0.9020\n",
+            "\n",
+            "Epoch 24/24\n",
+            "----------\n",
+            "train Loss: 0.2792 Acc: 0.8689\n",
+            "val Loss: 0.2162 Acc: 0.9150\n",
+            "\n",
+            "Training complete in 32m 32s\n",
+            "Best val Acc: 0.934641\n"
+          ]
+        }
+      ],
+      "source": [
+        "model_ft = train_model(model_ft, criterion, optimizer_ft, exp_lr_scheduler,\n",
+        "                       num_epochs=25)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 637
+        },
+        "id": "YfVNYL7_Dbth",
+        "outputId": "bbec4a56-cc80-4244-9936-73eddb3a193a"
+      },
+      "outputs": [
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py:490: 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",
+            "  cpuset_checked))\n"
+          ]
+        },
+        {
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "visualize_model(model_ft)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "### 7. 고정된 특징 추출기로써 합성곱 신경망\n",
+        "- 마지막 계층 제외 신경망 모든 부분 고정하기 "
+      ],
+      "metadata": {
+        "id": "E72o6hWtTIb-"
+      }
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "ZUPql_1QDbv6"
+      },
+      "outputs": [],
+      "source": [
+        "model_conv = torchvision.models.resnet18(pretrained=True)\n",
+        "for param in model_conv.parameters():\n",
+        "    param.requires_grad = False\n",
+        "\n",
+        "# 새로 생성된 모듈의 매개변수는 기본값이 requires_grad=True 임\n",
+        "num_ftrs = model_conv.fc.in_features\n",
+        "model_conv.fc = nn.Linear(num_ftrs, 2)\n",
+        "\n",
+        "model_conv = model_conv.to(device)\n",
+        "\n",
+        "criterion = nn.CrossEntropyLoss()\n",
+        "\n",
+        "# 이전과는 다르게 마지막 계층의 매개변수들만 최적화되는지 관찰\n",
+        "optimizer_conv = optim.SGD(model_conv.fc.parameters(), lr=0.001, momentum=0.9)\n",
+        "\n",
+        "# 7 에폭마다 0.1씩 학습률 감소\n",
+        "exp_lr_scheduler = lr_scheduler.StepLR(optimizer_conv, step_size=7, gamma=0.1)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "### 8. 학습 및 평가하기"
+      ],
+      "metadata": {
+        "id": "6cS6nHXwTUHl"
+      }
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "d76xPVR_DbyA",
+        "outputId": "98e9b65a-1fc4-4321-91b5-cc08da5f8784"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "Epoch 0/24\n",
+            "----------\n"
+          ]
+        },
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py:490: 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",
+            "  cpuset_checked))\n"
+          ]
+        },
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "train Loss: 0.7327 Acc: 0.6148\n",
+            "val Loss: 0.2635 Acc: 0.8889\n",
+            "\n",
+            "Epoch 1/24\n",
+            "----------\n",
+            "train Loss: 0.4836 Acc: 0.7582\n",
+            "val Loss: 0.1645 Acc: 0.9608\n",
+            "\n",
+            "Epoch 2/24\n",
+            "----------\n",
+            "train Loss: 0.5213 Acc: 0.7869\n",
+            "val Loss: 0.3503 Acc: 0.8431\n",
+            "\n",
+            "Epoch 3/24\n",
+            "----------\n",
+            "train Loss: 0.4484 Acc: 0.8197\n",
+            "val Loss: 0.2268 Acc: 0.9085\n",
+            "\n",
+            "Epoch 4/24\n",
+            "----------\n",
+            "train Loss: 0.3925 Acc: 0.8115\n",
+            "val Loss: 0.3317 Acc: 0.8693\n",
+            "\n",
+            "Epoch 5/24\n",
+            "----------\n",
+            "train Loss: 0.6248 Acc: 0.7500\n",
+            "val Loss: 0.3675 Acc: 0.8497\n",
+            "\n",
+            "Epoch 6/24\n",
+            "----------\n",
+            "train Loss: 0.5915 Acc: 0.7500\n",
+            "val Loss: 0.2445 Acc: 0.9216\n",
+            "\n",
+            "Epoch 7/24\n",
+            "----------\n",
+            "train Loss: 0.3916 Acc: 0.8033\n",
+            "val Loss: 0.2041 Acc: 0.9412\n",
+            "\n",
+            "Epoch 8/24\n",
+            "----------\n",
+            "train Loss: 0.3921 Acc: 0.8402\n",
+            "val Loss: 0.1711 Acc: 0.9608\n",
+            "\n",
+            "Epoch 9/24\n",
+            "----------\n",
+            "train Loss: 0.3482 Acc: 0.8484\n",
+            "val Loss: 0.1620 Acc: 0.9608\n",
+            "\n",
+            "Epoch 10/24\n",
+            "----------\n",
+            "train Loss: 0.3570 Acc: 0.8566\n",
+            "val Loss: 0.2108 Acc: 0.9216\n",
+            "\n",
+            "Epoch 11/24\n",
+            "----------\n",
+            "train Loss: 0.3315 Acc: 0.8484\n",
+            "val Loss: 0.1864 Acc: 0.9412\n",
+            "\n",
+            "Epoch 12/24\n",
+            "----------\n",
+            "train Loss: 0.3213 Acc: 0.8811\n",
+            "val Loss: 0.1714 Acc: 0.9608\n",
+            "\n",
+            "Epoch 13/24\n",
+            "----------\n",
+            "train Loss: 0.3009 Acc: 0.8730\n",
+            "val Loss: 0.1862 Acc: 0.9346\n",
+            "\n",
+            "Epoch 14/24\n",
+            "----------\n",
+            "train Loss: 0.3299 Acc: 0.8484\n",
+            "val Loss: 0.1811 Acc: 0.9477\n",
+            "\n",
+            "Epoch 15/24\n",
+            "----------\n",
+            "train Loss: 0.3236 Acc: 0.8484\n",
+            "val Loss: 0.1798 Acc: 0.9412\n",
+            "\n",
+            "Epoch 16/24\n",
+            "----------\n",
+            "train Loss: 0.2597 Acc: 0.9016\n",
+            "val Loss: 0.1704 Acc: 0.9477\n",
+            "\n",
+            "Epoch 17/24\n",
+            "----------\n",
+            "train Loss: 0.2945 Acc: 0.8770\n",
+            "val Loss: 0.1759 Acc: 0.9542\n",
+            "\n",
+            "Epoch 18/24\n",
+            "----------\n",
+            "train Loss: 0.3595 Acc: 0.8525\n",
+            "val Loss: 0.1690 Acc: 0.9542\n",
+            "\n",
+            "Epoch 19/24\n",
+            "----------\n",
+            "train Loss: 0.4043 Acc: 0.8238\n",
+            "val Loss: 0.1780 Acc: 0.9477\n",
+            "\n",
+            "Epoch 20/24\n",
+            "----------\n",
+            "train Loss: 0.3929 Acc: 0.8156\n",
+            "val Loss: 0.1908 Acc: 0.9477\n",
+            "\n",
+            "Epoch 21/24\n",
+            "----------\n",
+            "train Loss: 0.3386 Acc: 0.8730\n",
+            "val Loss: 0.1643 Acc: 0.9542\n",
+            "\n",
+            "Epoch 22/24\n",
+            "----------\n",
+            "train Loss: 0.2937 Acc: 0.8566\n",
+            "val Loss: 0.2269 Acc: 0.9150\n",
+            "\n",
+            "Epoch 23/24\n",
+            "----------\n",
+            "train Loss: 0.3227 Acc: 0.8689\n",
+            "val Loss: 0.1712 Acc: 0.9608\n",
+            "\n",
+            "Epoch 24/24\n",
+            "----------\n",
+            "train Loss: 0.3753 Acc: 0.8279\n",
+            "val Loss: 0.1713 Acc: 0.9542\n",
+            "\n",
+            "Training complete in 1m 53s\n",
+            "Best val Acc: 0.960784\n"
+          ]
+        }
+      ],
+      "source": [
+        "model_conv = train_model(model_conv, criterion, optimizer_conv,\n",
+        "                         exp_lr_scheduler, num_epochs=25)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 637
+        },
+        "id": "eFT9SuokDb0W",
+        "outputId": "21e37ebb-b5cf-4a9b-ef8c-cedfda09a864"
+      },
+      "outputs": [
+        {
+          "name": "stderr",
+          "output_type": "stream",
+          "text": [
+            "/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py:490: 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",
+            "  cpuset_checked))\n"
+          ]
+        },
+        {
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAGwAAABeCAYAAADG1PFrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO29eazlyXXf9zlV9dvuffft7/W+TM/0rCSHFClSiiiGliVYDhIIgRPHgSM4sQMlCAIF2RDLMZL8kQgynATIHzFsILHlOI7twIJE04ookRRFDmc0wyFnhrPv09vr1/36rXf9LbXkj/rd149NjjhE5lrqoE+ju3+/W/WrX906dc75nlOn6koIgXt095D64+7APfrR6B7D7jK6x7C7jO4x7C6jewy7y+gew+4y+mNnmIhcEpGfba//moj8b/8C3vl5Ebn2I9QPIvLALPv0Qcn8cXfgKIUQfvWD1BORXweuhRD++mx79CePPlQJE5E/URPg/5cUQvgj/wKXgF8BXgX2gL8H5G3Z54FrwH8F3AD+AXES/FXgHWAH+L+B5SPt/SJwuS37r9v2f7Yt+++A//NI3c8CTwH7wFXg3wV+CWiAGhgCX2zrngR+A7gFvAf88pF2CuDX2/6/CvyXRAn9od+/fT4Avwy8C2wDfxNQR8r/MvBa2/7vAueOlD0MfBnYBd4A/vyRsn+l7c8A2AD+ix/alw/IsJeBM8Ay8CTw3x9hmAX+BpC1A/OfAE8Dp9vP/g7wj9r6j7aD/Lm27H9un/8+hgHn2i/ybwMJsAJ8vC379Wkf2nsFfAf4b4AUuNAO7p9py38NeKLt/5n2+1w78vzfAv7WD2HY19rnzwJvAv9+W/YLwNvAI0QT89eBp9qybjvR/r227BMtwx9tyzeBn26vl4Af+7AY9h/eMSveOcKwmlbi2s9eA/70kfsTRIkw7YD+4yNl3fb5H8SwXwF+8336dCfDPgNcuaPOrwB/r71+F/j5I2W/xI8uYUef/4+Ar7bXvwP8lTsmz7idcP8W8MQdbf0d4L9tr68A/wEw/0H78kFt2NUj15eJ6mdKt0II5ZH7c8Bvisi+iOy3DHTAsfa5w7ZCCCOiavxBdIaoVj8InQNOTt/Zvvevte/kzve23+FHpfcbg3PA/3LkvbuAAKfass/c0a+/CBxvn/1zRAG4LCJfF5Gf/GGd+KAg4cyR67PA9SP3d4b7rwJ/OYTw5J2NiMgmUXVM7ztEVfeD6Crw6fcp+0HvfC+EcPF96m8Sv8Mr7f3Z96n3R9Gdz0/H4CrwP4QQ/uGdD4jIOeDrIYSf+0ENhhCeBX5BRBLgPyba+zM/qO7Rhz6ISnyJaJOWgW8Cv3pEJV67o/5/CvwBreEF1oBfaK8fI9qwzxJtzf/I+9uws0Qb9ueJE+uoDfs14P868k4NPEcEP0V7/xHgx9vyvwF8nWgnTgMv3tnvD6ASv9o+fwZ4HfiltuxfJ9rEx9r7BeDfbK97RGn8RaIdToAfJ07alChtC23dvwJc/rBs2BQl7gN/H+j8EQxTwH9GREQDolr71SPlf4mouz8ISvxp4BmgT5zJf6n9/CLwQtuf32o/Own8IyJa3SMCn2m7HeD/aOt/H0oE/jbwt38Iw6YocQf4nwB9pPwXiZN62s+/e6TsIeC3ieh1B/h94OMtw77U9rUPPAt89ofxQ9pG35dE5BIREX3lj6x4j/6F0B97aOoe/Wh0j2F3Gf1QlXiP/mTRPQm7y+gew+4ymll03e0Mw5d/75/yxde/jNMBEUGUAiXIYa32WsCohH/jp/4CP/WTPwMi+KqBsUMKjeQJiMAR9R1CbBPANxZcQOVJLGscftyg53MQsE3N3/8n/ysvbjwXm6H1vCW2F5D4r0z7FOL7pO2hKJCAYvpwOOwDSITcPn6mtCZPhV/75V+//TU/RJrdcohRZFmO0oqgiQwTCNIyqf3S00Ffzpd46P5HbzNm4hAXkPT7uxhC4JDtPhAqh+6m0A58qBwqM0yr3Ny6ztXt924z63AohSnLaJ9Gpoxra0isJajD9oJIy+tDXwsRQSkhSwUl9f/n4Xs/mhnDglFkeYFSmiCtNAjtgAggiI9DpUTx4KmHWFpeA8BXFkoPXQNqOkpHwFHgUJn7yqJSE0eWQHCeYD2qm8aq3vPdl5+h3wwJbZ2jzUT5ihIVZMqTKcfa/6fPhWmpEAgcra20kCUAE8bV8EMZwx9EM2OYiJBmOdpoUO2sFRCZKqD4/b2HTGc8fvET6MTEgolFGo8UP7h70jIxOB/bTDWH0lVaVJEcjvXuzk1efPtZAh51RKKmbBO5rZbhiAbgiEqMBS1DBd8q0amG1FqTmoD3E8Zln3E5+vAG8g6anUr0AW0MSWIQ8VGyCIiazk8hePA+cLx3jDOnLwBRIpg4KAwYBd8rFN/7itKi8ulXCAR7h3QFzwsvP83uZA/ktvo6bFRC25N2AtwhWOqItUWihAUC0po4EDKj0ToQfMmo7DMc9bF2dq7S7BjWrt0YFfD4djZGSZtqR5SgneKhkw+SF5342MQiNsBK0jbUVr7DXwzWR+nQt4GuL5vWlkXa3trk2VefwuFvP0fsx231R3sRGXf4KpFDcKKC4EKI2iFEqdRakxiFkgbbjBiODxiWI2zjCWF24HuGEgbaaJx3OOxtRCagBZRoCJpUZ1w4cT/K6KjiBg1kulVzfB+jDptvLDpPbt+XNg63iYPlrOUPn/19dsrd2wqwBRARQtyGG1FYbqtLQZAQxUhaFHi7spAkmtQogq8pJ32G4wMm9ZgQFMZkOPdhDuT30gwlLGIr6xw22KhuptAYIYigRLFYzHPi5BnQKkrX2MJ6wVFtBFFViorM8NZFyVKtuHpPKBt0Lz+sf+W9N3jh3W9H6b4DaIi0Go7bKHUqVVPygISjLkhARJElCqMtja0YjweMR0Nq2xDQKKUR0Wh9N0qYIw5kcHHQAngCwakoCSrakJOLp+kuLMYRO6hAgXSS72vuUBJCC+OndUKI0pUlUXSByWDA15/+HYbN6HueBzkEfFMbFNp7JZFJvq2nDmVRWhSoyYygpKGuhgzGQ8aTMd55Qoi+mvOB4BuUzmYypDBj0BF9E41WgrQqJw5KQIUoYRdOXESnKaHxMGpgMT8c+MOmvL/tJFcWpdRtaXCe0Hh0Lw5SCJ7vvvw0b2+/gxfP0VjpIX5RUxV3G6AT7oT07SskoJUhSQRNzaTqMxr3KSc1zsZ5FqY2MAhBAiHctpkfNs02j1CEJElRejqlBRc8wTtCCOSq4NTymWi7hk10lLu3peswmhFCVIchEKxDdW7PYDdqIlKcOsmbV3nihS9TUR+xXdEmhSkcD633JREJyiEChMNIB1F9a6VJEpBQMi4PGIwOKCsH3hB8wHtACUortFKHE2tWNDs/zAec9xiTIDogSsdZ7B1BFN471ubWWF5eA+thUEOq8EbAudbRlmi7DqXLobQ+dKZDaSNqawFKXZU88fSX2Cl3udNuTaUgKjnV+oN3+FihBSWtv6W0IkkERRWZNdymqhyQg0R3Raloq6OaNSQ6wc1wAWR2kQ4b8IDSJmZYRBxPIoIjDsippdMkaR7V57iB5YKAR4LgrUOnCc66Vro8VA7ptT6W84RRg1poVaH3vPjCk7x89QUc/ntct5a9HLoI0+upGpx+dkQ4tFakRqFoKMsB/dEude3QKseHBJxqoy3SzgYFohGVYOQuBB0SAs7WiIqzeTprY/BXk6A4vXoW0RrGMXiLMUiEZ+AiU6R2oDzegTLqUOrCsEYSDVoBga3NK3zj+S9Thpqj3raItLHH1v86ZFRoeXYbiEQYH9BKSBNBqKnq6GPZxqNUAZIgIUp0aO2wUklUs8Hj/OzsF8zYcba2jJEN8S1snkLsQKozji2egCYggxqMJuh2mIcWyQy+dIglRknqAEsJ3llC45HKI6sZCJSDIV974gvsTG4RpmGIKUz/nujGbQs1DTtFzRtRnhaNUQqUJVBRVgNG432qqkYwgIqIEEEpgxaFVgalo91tbIWzDUbrmQ3rTGF94y1KCT5ACI7g21keYCHvsdhbiYzoV9DNwQWkjgCEVCNVyzDvwShC8ATrkL5F5lKCBHxjee75r/Pq5su4w4USz21F+L3hqKmTLAQSlbLSWeXMyhnOrpxlvbeKaRreuv4aX7/8VUbjPSaTCcEnUTsIgEcpjdIaJQlKKbTWBCUkEl3yTH+/W/Jh0UwlzPk2JOXvTBsT1rqrFKaAsUVqCwvtrOw38XHnkSqAD9BYWM4IpUNKh3ioh3toV3D12jt848Xfow5NK1Dfy6jIoGlsPcKMwnR59OzH+LGHPsPp5VMUSYFKdJTGZkLRm+P33/gSk0mJdxqC4BGUENGgvo0Gvfco5REUSimKvEOi7kIbFjzY4A8XLiX4Q+dXAhyfP452CsYTcD7aoipE6eqlMPZQhQjHjSIoQUYOGTWEtYyDb34La+ArV77NfnWAF4m+eBvCmKLr21BdSFXKI6c+yk997POcPXUBlZo2ehKlJ0wqJMnp+oKfWP40vz3YpcQBOragWnslCSI6+mA+4LxFtX+0UgS5G1Wi9VhbRec5hgpQrYRpFKudNfAg44oIAQRKBzb6RKryUPtok7oplB4ZNIRM8N4zODjg3a23uVFeI+jQxgYFI4ZMp9S+ocEiEtAknF4+z5/6+M/x4NlHMUkKiQKtwTqCdZFnWhN2dynEc/z8A3AlRSkHKJRMnXUVHW6JoaiAx7sGgmDaOKa/K2OJztM0VWvgdXSe2/BURsai6UWwMSphbi7arrKCTkqo44oz3sfkZhHUoCGUDbI6B96xubfN5nCPtHbkhaI2nlQlLCaLGJVxMNxjqANF3uWnH/1TfPrxf5lCijjmiYpIvGygaiLaVILbuAl1jT99km998x9glUMR44PTSA1BoopXIYILbfBeYSQ62QGwYXYcm6FK9FSuaaGvbgMdggIKVTCXzkXbVFWwtASTimAtJAXSr6FxhGCRLCeMLeGghLmE4AL7eze4tHeNvquw3qFKj0khEU9VDyAPLOiCk911fu5P/wVOLp9BHJBASBVifVTD1kVmeY975zqSJZiL53nr3dd479Zb0G4ojW5VaCP3tDkcjuAdShuyNEMHwUmIYbRZDSozZJhvanxwaKXwnts5HQE6piAjgbomBOIK8qAPCx0oLYyr6AJ4T9AaBhVYB3mHajTka898kaGaoHSB2HFUs43gdWDiSlIMJ7qrnMvXOGYWkbGDTgKOeO0sYgSKFIYT3OWbqMUe6vw6ZTnha09+gTI0GH07PhmIUiPEeKhzFoBMgSbaM+8DPnA7gjIDml3WVF0iYsl1wtg3uODRPoZau7ogaQIcDBCTRElrajA9OBhD3TJMp0hpCftDWOoRCDz/8pO8euN11rKMTOd0mxQah4ihl8/RoHmgWOV01qPp9+lfv8zK2YdhWNEucUOREJIEtvv4zV30qVXk+CJB4K23X+PVzdcISkeoIQrvA841eG9bpBulTKk2KuOhsY7G24gg1V0IOmxdAY6uMgTlGLkG7x1ZWVHkBjUqCcMRLC0TxiPEWphUyGAYYbgLyGIP+mNIBBZSGsYk3ZQLaw/DZEhhBDs8AAV1XTPXMTxw8kFOmwX6OzeovWe4eZmVxdOI8wQlsDgXwcbVbfyoRJ87TljpEkKgrmueevFrVCEGjpVSiGiMURid4pzFOYdzEcwYZVBKg1KHC2xHdq3MhGbGsL1LrxOaPvOX38MWc9TLHdxkQu+Vdyg+9xEYTgh1g4iw++YbLDxwH6ZqYDiGzIBOQHlCHuDcCjYX3nrhFb7zrWe4unkdI3DfiRVs2WB0ga0syaThGBm4gHUNdTVg1N8jjMaQprCyBM4SrtzCO4u+7zhhqRPzSLTw9LNf5bVrL9A0VYyDKodSkJgMrROMSXHeEVwT18hEIdrEKImZ2rrZ0sxe8dxTv0XZ36N5/lXC1Rt0Jg1p2VC9s4kqLfXeLj54fFNx/fUXacoR+5vvEYYjwngCWhMSB+eXYLHD1Stv8c9/54t8+5XX2RuP0LlQ1jVrSxmdtEQksN8fcmPnJl4CKI1zjsngFl4sstCD/ohwdZvgPfr+k4ReHtWpEl767tN85dl/RuUqrLPYpqZpKqytW3UYl4SUCGmaUqQZxsQEV6UEYzTaJChtDrO6ZkEzk7DXXn+JudOPM964SWZyVjJhZ32Vnf6Yyf4ub7z0Gzz683+R4fYmG+++jv3KiOLUMXrzF0ELWge81KhOQl2N+NZT3+CdKxvML2T87M88xiMXlnjxmcus5xneJ2zmEzZv7lF3G26ON7HlCEtgb7iLtRV6aw83msBCF33fMbxRSOMIifDad5/hC9/8h/SbQZtso1FthMbjsNJEKRSFVhqdGJzExBxCVLXxjyeIj7ZyRjQzhtVbfca7Ay4PS87vj+i+8goH8z/Bjg3s33gP/dZ3eODz/yrvvvsGV66/wcH1F/nkz/87VGaEyTP2b1xirz9hIR+xtb/Nm2+8Tl5o/uzPfYTP/vSjJOK4eWmH8fYBJjh8U5JIQ5ZWFCnsDg44KIeEWjO5uYeTCfnxddSp9Rg81gofLC8/8xT/7Ll/ym61T2NjNEZpTZDooEMEHM7WiIBOsgjfReMDqDa3QKm42kwribOimTHsUal58+kXOWg8u7t73FIHjJ98CZ12uHXlEuXNS7z05G/w7RtbXN27zEpacP/ldzCjivX5T/Hsa9/huZ23KL61hPWWm5ubfOTj5/nkx+4j8ZZqv4+tS6reGa5ceouqydnY28K/eYVjvR5znQSfeHYOxnzzu0/x+IOf4HR6EgZjpJNT9/s8/fxX+Z1X/jkDO8I6T9MySEQhStAS1ap1NYInM3mUo1byaFMVfIhoMYrl0fy5D59mxrAyGdITjVlZYtdDVlnu39xgdPERLr3xNmfE8uIrX+WKWaYfHJIq3nvvm5jVf43y4Apvb77Ki995mnHWQffmITSsrRZsXLrJNmPeeXeTZ1+6yokLCop1tnYuszOBl595i0fuO8snP/4wdAM3rm6xcfVVOt1FVo+fIk8Ttrev8pXXvsy3Np5l4sc01uKsxfs2uKtiJB6lEaUwAbRJyEyGiI7RMxXQR3gz5ZNSt9fiZkEzY9gb2rEcKk6tneXWzi1uVYH7zB433niRW7WjM58heSCfNxzUBd1ji+i1dV7Kh1x9/Ztc7W/TH0+YjGoS0WAUw0rzv//dLxJsyeWtEQfDhk/7BR752McYjkd4hEanvLlxi7WTZ1DO47I5Nm/ucmnrKud2r3F181t848pTbNXb2OCwzuO9a0FFTKFDty5Am0WX5QVFmqOUpqo91nl0Cwk9cUFdQdyYwfdky33oNDOGvVcZstE+Zw5GDE+coNTzXLL7pCGlO9/h2rjPeHfC8kLC4//SJ1mQEvudl3Enz1OLw4th/vz9zGlD3ZSEYHnyyT/k5tY+eZowbixKhIcfHrC1cQUfBBfiMqMtK25tb5OnBusC1gdevfkWu2/ss9FsMqHB0a5cS8w3iXn9Aeci83yoUDjytKCbFSRJhvUQlEUr2hzEuLziRSFttq/3Dq3uQtAx112mn4xZZw83KHHS4c1hn89Qcm19jZ3hiHP9Pq/7y1zf3aNT3eJBU/Hu177MyKSErMOqSXBVydhaymrMrVu77I0tad3EEJCA84G6tgSEvb0he9u7GK3Z2LhJby6PUfzg2A0DynGDM44QND64mD8igtIJSlIEFW2WrbCuxiCkOkWJwQWwCKI1RlTrVEOwDd656JNJABwzzCOdHcMW5wyjtMO7W3tQ7lJnC6wfW+NYvcXm/g7zCxmZBDoLHdTJE+y+t88ro4rBpM9YK4aTDR5b1AwmMKwhzHUwSUIIDbWFPE+AQPCeojvPlavX2bi5xc72LkYUVd2wvtRjrluggCzvYMURMPgQ8C5u0EDFKHu0Wwaloh9lnCZLEoxJsAS89fg2SihaoVoor5Tg/DTeGBDxd2fmb3Nwg7KT0LOKH0tqnnGOLO/wlk4YCdSUvKeEW8N95i8Jc7uWUgp0kpItFtS+YjIZ05ceYxyrJ1c4JQeUFgbjkm6iccFj64rEJDSNY9AfUDtLFYR654Cb23usz3d46OJZpEvMprIW6xyeJvpNmDYz2RMkppQnWpPnOUZpXIDGxYRUaRN+QhAIkWnaaMRHKQ7BYUwMac2KZtbyR3eGTG4O2Mw09aqmCiXFjVsMtizX+yXDawNODTyJKDrlkI/KiOW8y1yWM76xhZpMWB15JjolWeiRVCVLRcHa0jwnjy2jjSYgjOuGsnGMJhVlWdPYgAuBcd2wO5xwffeAg8mQYOIGQmsbrGtiirWPMUvvHd7VNHaCcxVaa/Ikw+gkBnYbG0NS3rWpvm0yrPdtToeK2czKk2p9aM9mQTNruTmueOhkyqkHThFuBUQ7QlZzekFgXqNWEjproFKPX8+42RPeGQ/YchPy5Q73rxpcConydDs5Q5twq/RI0qWx/nBtqihybmzd4sbWLayffh5iYkxQjKqGwXiCEo3ztAPvY1iMmPvhvcV7i3UV4j0JGkHhg2B9iCrUO7w/kuZw5Lv64BBxMVKPoZ7djtnZqcSnTIpyNYnd4sFFz5ncsNvtcbC9R9rJmVtKKS5PML5kf9xwUAWGJtCZlKRi2BdPnXfprC7jO4vcuNUnWV5neHDApCzpJIZECfXoAFsts7O7B0pjJJAaQ29xkY16i+A8TWlx1uOUxwWHDzGlIPgQo0ji8cGhRMdAL0LjAjYQo/U6JXiPCQqNibmIojA6I9MJqdHkaUonK1AY6rtxQ99qqpjfcDyXVzwDiK+Z3xbe7DuaYsyF1HPiIY27Hrhv3OAnnmGWkQQ4X1fc8Dn7krKWZZQE8t48k7IilEMWOzlJYphUcH3zJt2FVZqmjjkW3pMazfxcl600ZTyqqFyAoIi5GQZRMbkHH+JaV/AoFGmWkopB+4ScLr3uAqvdFZaKReaLeebSgiLL0UlOkqTkWU6e5mRpjklTtIk7aGaoEWfHsH4SeHQZ3gqBt63AxPHn1IQmDbzgHWMCT5eKhYuncFdvsEtgIo65Ts6J/X2G6QnC2nGsSfn4Yxf4ylMvkTcTlo4tsDOYsLO7T+MsdV3y3eeexnhHb3GJwXBC5RrGk3GMAXpPOZqAFXQ3Q7xuGeXxzqODRwKsdlY4v3qOi+v3c3rlHMeWj9PrLpKmBZIkMQKv20zj9mQEjLRp6HJ7x42aZZxjhgx7u1aMveL4Lcf93vPkkuHZ2uMLw5L1FG+UPFUYjoU9PrpVsVEpFo4FbL3H64miWVtl4AQ7sTz9h8/TDEYcWywgz0grj9bC+nKXB445Vudz3r42YWN7h7nEcDBquL6xSVXGLNwsS2OePnF5RIBUNPN5l/O9kzy6/iAPnXqQlaV10qyLJBkkBpQiJOr2XmvVCmrSMmiasye0R1uodlPFXagSF0MgTOBaR/Ep6zmpNNcnwiQID3U8D87DS7lhLtQk6wFddtnbGvGYEvTZC/Tnl+lUlmp7g+39IcfPnKJqPKP9IaOyxoXAseWcTz6qObuquHhCcX3Hsj9RvL3heedaQyKBIjUszyUE8WirWc2Wubh6nsfWL3Ju4SSL3SWSJEeyFEwCaQ6pAdPucTAqqrlpMm+bxni4qWz6rw2EGUY4pjQzhp308NHgeaKX8Af7wvkm8OOm4bcLw7YTXikUpmfY2Cn5ijP0LixzfkHxsUlD5+OP8vaB49q77/KtjeuYYo61tTVuHoxJVEIaxvQ3Ntm4EXhrLiEXzdqy4fh6TtV4TiwqjvcMwSvy1HDy3DGKhfM8tvIQjy1fYKWzSJIXSJpClkKaEDeBqZjzMaV2T1rcPzSN9MZ9ScGAJMLheSsiBK1iPPFuXA97Tyz9FJYGDjNf0C88Jtes1IGt/cC7Y80oE5YaYV+nmLJmMesS7j/DzfIA9cYr7L9+i61ase7GvPzs83TnEkLlCV5xZmmOpU7B6tpJnnnpLZbmA4tdWOpBN6m5eMLTzRTLK8dYn/80Z3uf5sTaSYpOFzEGTHtoS0x3AueiNCmJ97bNVp5+pmgDwgHRgmiJo2f04Y6aqdqc5e8tzKzliz3NpURzcuJ4yNU8aRS/ve/JOzk/0TMcCyNe7hU8vF2xu36GylqW+0N+9+BtRnXJ6XrMmx5Oz3VJguPG3gFZP1A2sL62zInjx1Dieeem5buXU8ajfTrGc9+64uJZRbfIOb50hvW5n2LZP8RSb5k0y/Deoex0Q6hpd2NqxMSjj4IxiGkTTQ2QCMG0kfukXfNKFGL07R2iErOrgnNtmvhdmCJwa2D5CbpsM+L50lI0GadKzUbhuC6Wal5Y3djjrVSxbw8o6iFrozE3mpRF17BVaRZPZtgDx7JRHC8y3q0CXge6Rcyd8ErD4v10VoT90rPb7zOoIOkUdNM51rsfQZlzuCRgywqbViRpCp0UktZmZUnc0G4UIdFghJC0Rkq3q8jE8zkOD3tPp/YqLqeEKUKsPeICwdewPjeTcZ0Zw+arwM3OiMZ7zpaKfN9RNY4m5FyyESH/TNrwu6FL2Buy34z55rxhPcv5vLNc3YGtJcNGOeKjxjMcaN71Cb1E0+zu8OaNXVbWVskWj+ObhjxLMTKHOM/VG10+c+YB5qoVmoMStVDQzXKyLEOKPNqs6dlSEgh1Ba6BZroxokWBWsX4oVaQ6AjtlcRUch1iFqmJmyQCMSgsomYYP5plisDE8VyRsJMklHOW4wPLdhM4uVHRnUt4Y07xZCdh18Hn6pK3Q+AFK6QMeSsXNnNDeTBmUTneHAo3QsJcR9EbjzjewI1S2Lx5i1uDbxG0oejkdIoFlucXeWD+BJ8+9gCL+TyFTlhcXkbnKZaAlCVSVpEZSQI2hSyLCa0miVI23dmZyKFdinlvRPVoiNDwcDOzQ5K4wwZC3NAxoy1iM5sLf5AIUik+4RQ7RYabTzmTK3aSwCCDkHa5ORGysubNLFCe6vC5oFnxmm8PoF4puHhS8WcWFKMqoVMUEDwLCVwOit7aMucuXMBkKSKOVAKpCMtJwkO9VUwdsJMaI3H9qh5N8KMJqrYordFFgSoyJI9/IwhRiIkSchjo9YF2/aTdzRnawzzidWjz9MPE4s0WCbMAAAleSURBVEcl9cGAyc7erIZ1hqGpJKOTJTQKlo3iTWVJc01fJeQEklByEQ9jx7NdQ+YCF4qaAkXqMvb7Q66UlqtD4YoEzuaBxTrjsupw6uELHNOK8dYOwQg7paOwnm5d4yY3ubVXUZx9mN7xc3QXu4RUKI3D5gGVB0KYEGyFd6BVBmIIDVhrsa4hpAneCI2NuYm1a3DetUDFY72jrCdYHXDOY61lNBrTH+xzMDxg0kz4q//535zJuM6MYbkHPRxztZdz//WSsFawOijZmQucH3g2reJl58m6hsdroU4033QKKTRntLCwZXltaLh6YLlvTnhgPGZv+QTOJGy88w4nywHnS8ut9eMsn1pmsDdg3D+gs5ywdX/K4NQtXtZ9wv6L+IOY6NntzJMVBUGgbBpqV2PylCRLUca06doNznliXlTAhzbPUEGWKJSOp/tYP92AK/gmUFUWWzU0zQw3hzFDhm3ZhtSD4NkLnqRyDF1gUsGl2jLOoTuX8Ph+jR7Bu0s5aZpjfSDzFQ/mmuFCD71a8aAOnNixbEz6XD+oCLZCp57rRU690GVvNKA757hwdpliTjE8lTHODiAYqqbBOktHdZj4iqROcCgq7+PKckjQVmNCihaDV/Hwl5iUE5dhtAhZohBjcb6h8R4XVHvAWdzz5oPDBo8PYXoO2kxoZgw7s665PhaCeJq1BI3Di1A6GKXR71kcluSl4/kk4XrZ8ONzQveG4/ec48ZcxsX5NdKDDa40gUEvRZqGE2NPnc1zOUAx16HZ7eOH+ywaR3ELknPH8N0apS1JVuAKjWhDnudkqYkmKdiYFpAmaKXbTXqAgG63+LoQF0KVCGmiSIwQcDjvsYcHv0QkcniMbJB2P/RdGEvcsqB7gg+G4GBHCQ/O15QYHqkD71U1l73wBZfwcKo5VQfe267IgpB2e2wOx5xzlzk3sXSOzXH55piVsuZiKTwRBJ9oOBgw38nZbRzvlrBhNL1L+yxuT5hb66IuJCRdTS/tsVQss1Qs003nyU2XuWKOIuuipc3yhXZvl0dUwAahsjWVr2h8ycQOGFZ9+s0QH8oI40N7NF+A9vyz9zst8EOjmTFsQSvGBlySEaqSBaO5vuvxiaVxniwJfNzC8x2DKxvSxHCZggEVj/Uq1q3n5R3P2xqWt/ooZdjFcMtoNApvawosj9sBe2nKm0lOZ2WO1XPLZMcK8iznZPcU53vnOLtwjmO9dRa6S2RZB52miGrDUyaBPItxw2lUXsvtcJQWgndY31A3FeNyyMFwl/3hDnvlPnujPbYHu+zs7bBnBzgp787c+hUjGPHsTUoCmqZpyFNFEwI3rIdEuOk8aRq4kkbH5nzmed3BinJ8bj7whbHhRhL4VHDcUgUv5R3KfsNSVzO+vouSQH9unrrT41Sec+zCWY7PLXImPcbp7BRrc8eYK+bpJF2M6aAli1lRoXWOk6SNeGhINWF6dnDMtYlHKikBY0hISTpzdBdXWAvnoiiZmB7gqprRqM/27jbXt65xfWdjVsM6w6OLRlCmHpMLAQuSsliOGacwmtN8KlF8qfF8ZGxxwXAt0Ty4VXFrLueJSyXbxvOQEkYD+MORo3Ql+UpGmitS7whlw6Wg6SfCUtfQXL2BvrTNsc4SvYc/xeInHiRVMU1NiWqPhFAxIq91lK72KKUwPQvLxXOJg8QIBqqNIyqixGlFsCFmSXnaXSuKpJuw0OmysH6CBx78CKGysxrW2THsVWmYl4Kssux0NOfrhus+0EHIfeDJiePEUsbrNypIAyMC/48E1gTuLzQ6UcxVgTpPeHU3cKKTsGI0ajLkvs6ERzrC10pNxzmW05yz932Us93jnFk7xfrqKRJvUI0j9WCsR9kQz61yFUHXrXRlMcSUJNCv48Kj1nHvc24gVYRDKZMYM0ymP5YQo/qi2qCv89C0qdr5XXgSTjHfBaPZqwRcYG/U0JvPGY0aQqrJUscnRxVfMpqHrWXsE55WimOu5vjI853FhOd7HdLCcIoDlsZj9AhsCa+VmtpDncKCLeltXKYzX1Jm8+yNhuSdEVl3HtDUTUNQgmlMNE1ZCjpBlI4nCDQSV5dTTQgO0nYLrAPsFKLH7yCmPTck0ZF50h6POY05mvYXIu7G3Pp0HyTzzHUzltSYAwz5fsnVTPPIkrDb9/zBvkcvphwcWLIlQzeD5wc1c0nG0AZOd2BtAFIoPqMsN3c91+cXObATGj8hdx5wjANcKq+jRsKS2aHfjGgEjmenyDo9kqKLydugrzHxIDIlYOu4Gb6qItOSBEghbRGgEJdaCPE8++mZ+9aBxPP4CUdO9lbEkw9mNajMcnllMuG+PGFvr2RsPQNg0cKaDhQ3SrI6sFoGbN+yPVfw2QL2+/BWInSMYzJydENDsevZCIrfnAS6XY0RR20bVjuO9aFlL6RcX8hRqqGzfYPm5k3k5iVGe6/x8uMPkYY1TtVneMCd5TjHKdLlmJmrVWSQUvH4onbQg7MwtvFXJJRAqeP6V25idD5RUeqcj9uRTDwxQKbHAorcnethpzOH27DsZ8IqijmEd1Rg6DwHDvI8ZdBYMMLooOTKvqM0hsx5/uyK5tvbnhfTgpVVxWkjUHvWdyo2XIopCsKwYlGlqN4CfWlYWp5ncXGRQnosra3Q7c7TUydIwhxp0OyW+3gXWLQVnWKetNuNu1BIoi/lfSsxAAJ5XCsji+dRBR+QiSOULjJFA6knJHFDO0YfhtKDdVGaZ0CzQ4lVxuLJgHUBxhASeNALAyWsVZ6qrHkphwsaXs1TvtFvWOoWDJuS37rZUHc0mW84VuSMas3cyXn29ndjsmiaspv0eE4rVtMex2vH8tWK89uO0w8+RDecomi6LLtj9MwCnaxL2u2gspSQpaAV3npcOYxLLmmCmuuisjwuZk6zbCY1YVTG0wESDXmKFAnk0f4F2qMerAcbU79xbTS/M5txnRnDNhwsbtbcQjgdhE2g8Z7dVLipA91cs+QU9++UXF3WzCWOcjJkySh+shCuS8LOxNG7OqQqM554e0zQkNVD1L6l0DCPYcWvcv7sI9x34j5OLB+n11skSQsSY9Amwaj4AwahPf1faQWJQaVpRITtuRq+qnGDYTwwWgvSLdALPaTTgTQmn4a6gZ1hhP5aInPnMigS5PBkuJivPyu695OKdxnd+4W+u4zuMewuo3sMu8voHsPuMrrHsLuM7jHsLqP/F9tCFbWn5iwSAAAAAElFTkSuQmCC\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "visualize_model(model_conv)\n",
+        "\n",
+        "plt.ioff()\n",
+        "plt.show()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "lzqaConhDb28"
+      },
+      "outputs": [],
+      "source": [
+        ""
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "gM-_2m7PDb4x"
+      },
+      "outputs": [],
+      "source": [
+        ""
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "1ifpA2tGDb7l"
+      },
+      "outputs": [],
+      "source": [
+        ""
+      ]
+    }
+  ],
+  "metadata": {
+    "accelerator": "GPU",
+    "colab": {
+      "name": "Transfer Learning.ipynb",
+      "provenance": [],
+      "collapsed_sections": []
+    },
+    "kernelspec": {
+      "display_name": "Python 3",
+      "name": "python3"
+    },
+    "language_info": {
+      "name": "python"
+    },
+    "widgets": {
+      "application/vnd.jupyter.widget-state+json": {
+        "05ab70175635434db813e1ca278816d6": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_module_version": "1.5.0",
+          "model_name": "HTMLModel",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_9b5d02cfd948481da3fbaee3e46b3698",
+            "placeholder": "​",
+            "style": "IPY_MODEL_f4f63484744a44fd90ded2ba9703a835",
+            "value": " 44.7M/44.7M [00:00&lt;00:00, 78.1MB/s]"
+          }
+        },
+        "17d1def0ff0344ea9c2809a213dae0b3": {
+          "model_module": "@jupyter-widgets/base",
+          "model_module_version": "1.2.0",
+          "model_name": "LayoutModel",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": null,
+            "bottom": null,
+            "display": null,
+            "flex": null,
+            "flex_flow": null,
+            "grid_area": null,
+            "grid_auto_columns": null,
+            "grid_auto_flow": null,
+            "grid_auto_rows": null,
+            "grid_column": null,
+            "grid_gap": null,
+            "grid_row": null,
+            "grid_template_areas": null,
+            "grid_template_columns": null,
+            "grid_template_rows": null,
+            "height": null,
+            "justify_content": null,
+            "justify_items": null,
+            "left": null,
+            "margin": null,
+            "max_height": null,
+            "max_width": null,
+            "min_height": null,
+            "min_width": null,
+            "object_fit": null,
+            "object_position": null,
+            "order": null,
+            "overflow": null,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "29db66ae692f4d71b436f4fc3a1678ec": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_module_version": "1.5.0",
+          "model_name": "DescriptionStyleModel",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "7b02c667a1da4fbc99e580fb6110ed7d": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_module_version": "1.5.0",
+          "model_name": "ProgressStyleModel",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "ProgressStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "bar_color": null,
+            "description_width": ""
+          }
+        },
+        "9b5d02cfd948481da3fbaee3e46b3698": {
+          "model_module": "@jupyter-widgets/base",
+          "model_module_version": "1.2.0",
+          "model_name": "LayoutModel",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": null,
+            "bottom": null,
+            "display": null,
+            "flex": null,
+            "flex_flow": null,
+            "grid_area": null,
+            "grid_auto_columns": null,
+            "grid_auto_flow": null,
+            "grid_auto_rows": null,
+            "grid_column": null,
+            "grid_gap": null,
+            "grid_row": null,
+            "grid_template_areas": null,
+            "grid_template_columns": null,
+            "grid_template_rows": null,
+            "height": null,
+            "justify_content": null,
+            "justify_items": null,
+            "left": null,
+            "margin": null,
+            "max_height": null,
+            "max_width": null,
+            "min_height": null,
+            "min_width": null,
+            "object_fit": null,
+            "object_position": null,
+            "order": null,
+            "overflow": null,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "ade88339e5c64e7091363b4c6c02d22c": {
+          "model_module": "@jupyter-widgets/base",
+          "model_module_version": "1.2.0",
+          "model_name": "LayoutModel",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": null,
+            "bottom": null,
+            "display": null,
+            "flex": null,
+            "flex_flow": null,
+            "grid_area": null,
+            "grid_auto_columns": null,
+            "grid_auto_flow": null,
+            "grid_auto_rows": null,
+            "grid_column": null,
+            "grid_gap": null,
+            "grid_row": null,
+            "grid_template_areas": null,
+            "grid_template_columns": null,
+            "grid_template_rows": null,
+            "height": null,
+            "justify_content": null,
+            "justify_items": null,
+            "left": null,
+            "margin": null,
+            "max_height": null,
+            "max_width": null,
+            "min_height": null,
+            "min_width": null,
+            "object_fit": null,
+            "object_position": null,
+            "order": null,
+            "overflow": null,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "ae060ba608934a5db38fddd462e60376": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_module_version": "1.5.0",
+          "model_name": "HTMLModel",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_f92f8a50f06d42eb818f1ea94136cbab",
+            "placeholder": "​",
+            "style": "IPY_MODEL_29db66ae692f4d71b436f4fc3a1678ec",
+            "value": "100%"
+          }
+        },
+        "c07c0605a77241628f499fbd3b39e9c5": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_module_version": "1.5.0",
+          "model_name": "HBoxModel",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HBoxModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HBoxView",
+            "box_style": "",
+            "children": [
+              "IPY_MODEL_ae060ba608934a5db38fddd462e60376",
+              "IPY_MODEL_cc59758590f64205afbc980318fced25",
+              "IPY_MODEL_05ab70175635434db813e1ca278816d6"
+            ],
+            "layout": "IPY_MODEL_ade88339e5c64e7091363b4c6c02d22c"
+          }
+        },
+        "cc59758590f64205afbc980318fced25": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_module_version": "1.5.0",
+          "model_name": "FloatProgressModel",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "FloatProgressModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "ProgressView",
+            "bar_style": "success",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_17d1def0ff0344ea9c2809a213dae0b3",
+            "max": 46830571,
+            "min": 0,
+            "orientation": "horizontal",
+            "style": "IPY_MODEL_7b02c667a1da4fbc99e580fb6110ed7d",
+            "value": 46830571
+          }
+        },
+        "f4f63484744a44fd90ded2ba9703a835": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_module_version": "1.5.0",
+          "model_name": "DescriptionStyleModel",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "f92f8a50f06d42eb818f1ea94136cbab": {
+          "model_module": "@jupyter-widgets/base",
+          "model_module_version": "1.2.0",
+          "model_name": "LayoutModel",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": null,
+            "bottom": null,
+            "display": null,
+            "flex": null,
+            "flex_flow": null,
+            "grid_area": null,
+            "grid_auto_columns": null,
+            "grid_auto_flow": null,
+            "grid_auto_rows": null,
+            "grid_column": null,
+            "grid_gap": null,
+            "grid_row": null,
+            "grid_template_areas": null,
+            "grid_template_columns": null,
+            "grid_template_rows": null,
+            "height": null,
+            "justify_content": null,
+            "justify_items": null,
+            "left": null,
+            "margin": null,
+            "max_height": null,
+            "max_width": null,
+            "min_height": null,
+            "min_width": null,
+            "object_fit": null,
+            "object_position": null,
+            "order": null,
+            "overflow": null,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        }
+      }
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}
diff --git a/images/1.PNG b/images/1.PNG
new file mode 100644
index 0000000..6ec8281
Binary files /dev/null and b/images/1.PNG differ
diff --git a/images/2.PNG b/images/2.PNG
new file mode 100644
index 0000000..07c4ea7
Binary files /dev/null and b/images/2.PNG differ
diff --git a/images/3.PNG b/images/3.PNG
new file mode 100644
index 0000000..76c3c0f
Binary files /dev/null and b/images/3.PNG differ
diff --git a/images/4.PNG b/images/4.PNG
new file mode 100644
index 0000000..c511372
Binary files /dev/null and b/images/4.PNG differ
diff --git a/images/5.PNG b/images/5.PNG
new file mode 100644
index 0000000..1cc1482
Binary files /dev/null and b/images/5.PNG differ
diff --git a/images/6.PNG b/images/6.PNG
new file mode 100644
index 0000000..aee3508
Binary files /dev/null and b/images/6.PNG differ
diff --git a/images/7.PNG b/images/7.PNG
new file mode 100644
index 0000000..9d317c6
Binary files /dev/null and b/images/7.PNG differ
diff --git a/images/8.PNG b/images/8.PNG
new file mode 100644
index 0000000..42410e7
Binary files /dev/null and b/images/8.PNG differ
diff --git "a/week12_\354\225\210\354\204\234\354\227\260_preview.pdf" "b/week12_\354\225\210\354\204\234\354\227\260_preview.pdf"
new file mode 100644
index 0000000..4494783
Binary files /dev/null and "b/week12_\354\225\210\354\204\234\354\227\260_preview.pdf" differ
diff --git "a/week15_\354\225\210\354\204\234\354\227\260_preview.pdf" "b/week15_\354\225\210\354\204\234\354\227\260_preview.pdf"
new file mode 100644
index 0000000..1fc7475
Binary files /dev/null and "b/week15_\354\225\210\354\204\234\354\227\260_preview.pdf" differ
diff --git "a/week1_\354\225\210\354\204\234\354\227\260_preview.md" "b/week1_\354\225\210\354\204\234\354\227\260_preview.md"
new file mode 100644
index 0000000..c5519f9
--- /dev/null
+++ "b/week1_\354\225\210\354\204\234\354\227\260_preview.md"
@@ -0,0 +1,68 @@
+# Lecture 1: Introduction
+
+## 1.1 Computer vison 의 중요성
+
++ 스마트폰에 내장되어 있는 카메라와 같이 무수한 센서의 사용으로 데이터, 특히 시각 데이터가 많아짐
++ CISCO에서 2015-2017 조사한 통계 결과 인터넷 트래픽 중 80%가 video data 이고, 이런 시각 데이터를 잘 이해하고 분석하는것이 필요
++ 시각 데이터 처리가 쉽지 않아서 "**__dark matter(암흑물질)__**" 라고도 불림(존재하는지는 알 수 있지만 직접적으로 관찰하고 의미를 도출하는것이 어려우므로)
+
++ conmputer vison은 interdisciplinary field의 성격을 가져 다른 science, engineering, technology 와 연관되어 해당 분야 연구에 있어서 physics, mathematics 과 같이 다른 영역의 지식에 대한 이해도 필요함 
+
+<img src="./images/1.PNG" width="40%" height="30%"></img>
+
+
+## 1.2 Computer vison 의 역사 
+
+    Vison(시각)의 시작과 연관지어 Computer Vision의 시작을 생각할수 있다.
+
++ **Biological Vision** 의 시작
+
+  543백만년 전 동물의 종은 단순했지만, 화석을 연구하는 동물학자에 따르면 짧은 기간 동안(10만년) 생물의 종이 급증했고(evolution's Big Bang), 가장 신빙성있는 이론은 동물이 처음으로 eyes를 발달   시킨것이 원인이라는 것이다. 동물에게 있어서 Vision 이 생긴것이 능동적이고,진화적 군비경쟁(evolutionary arms race)를 촉진시켜 살아남기 위해 종의 진화를 활발하게 하였다. 이후 지능을 가진 동     물, 인간에게도 Vison은 가장 큰 감각 체계로 중요한 역할을 담당(일하고, 의사소통 등)
+
++ **Mechanical Vision** - computer vison의 시작
+  + 르네상스 시대의 **camera obscura** : pinhole camera 이론을 바탕으로 만들어진것으로, 초창기 동물에게 있는 눈의 구조와 비슷하다. (빛을 모음->뒤쪽의 카메라에서 정보를 모으고 시상을 project함)
+  + computer vision에 영향을 준 시각의 메커니즘 과정에 대한 이론은 **Hubel과 Wiesel의 전기 생리학을 이용한 연구**
+  :영장류, 포유류의 "visual processing mechanism"을 찾는것으로, 인간과 비슷한 고양이를 대상으로 연구를 진행, visual processing이 단순한 구조로 시작  
+  하는것을 발견( edge → pathway → brain ) <br>
+    
+  <img align="left" src="./images/2.PNG" width="40%" height="30%"></img>   <br><br><br><br><br><br><br><br><br><br><br><br>
+  
+  
+  + **BLock world** : 1963년 Larry Roberts의 시각적 세계를 단순한 기하학적 형태로 단순화 시켜 형태를 인식하고 이것이 무엇인지 재구조화 하는 연구 
+  + **The summer vision project, MIT, 1966** : 시각 시스템 구조 설계 
+  + **David Marr,1970s** : Hubel 과 Wisel 의 연구와 비슷하게 Biological vision처럼 처음에는 단순한 edge에서 시각 처리과정이 시작함. (이미지 입력 → 기본적인 스케치 → 2-1/2D 스케치(깊이,불연속성 등 고려) → 3-D 모델)  1970년대에는 데이터 적고, computer 속도가 느렸지만, Block 단위가 아닌 실제 세계의 물체를 인지 및 재표현 하고자 하는 노력이 이어짐
+  + **Generalized Cylinder, Pictorial Structure**: 모든 물체는 단순한 primitive로 표현 가능하다는 생각을 바탕으로 복잡한 구조의 물체를 단순한 primitve의 조합으로 바꾸고 이를 기하학적으로   구성함 <br>
+  <img align="left" src="./images/3.PNG" width="40%" height="30%"></img>  <br><br><br><br><br><br><br><br><br><br>
+  
+  +  **David Lowe,1987** : 면도기를 선과 엣지들의 조합으로 인식함 
+  
+  Object recognition이 여전히 어려움 → 그래프 이론을 이용한 **Object Segmentation** (이미지 내의 픽셀들을 의미있는 영역으로서 그룹화 시킴) 을 시도함  <br>
+  
+  
+  <img align="left" src="./images/4.PNG" width="40%" height="30%"></img> <br><br><br><br><br><br><br><br><br><br><br><br>
+
+  + **Face Detection**:1999-2000년 통계학적 기계학습(ex) support vector machines,boosting, graphical models)이 등장하고 2001년 Viola & Jones가 AdaBoost alorithm을 이용하여 거의 실시간     얼굴 인식 구현 - 2006년 후지필름이 얼굴 탐지 카메라 출시 
+  
+  + **feature-based(1990 후반 ~ 2010년)**:  객체 탐지, 불변하는 객체의 특징을 다른 객체에 매칭(ex.SIFT, Spatial pyramid matching) <br>   
+  
+     <img align="left" src="./images/5.PNG" width="40%" height="30%"></img> <br><br><br><br> *"stop"부분의 글자는 변하지 않는 부분으로 이부분만을 매칭*   <br><br><br><br><br>
+  
+  + **Pascal Visual Object challenge** : 20 object 카테고리의 데이터셋 형성
+  + **ImageNet** : 시각 데이터는 기계학습 알고리즘에 train 시킬때 high demension input, parameter 수가 증가하는 상태에서 data 적으면 overfiting 문제발생함. 이를 극복하려 15만 장의 이미지, 22만가지의 카테고리 데이터셋 형성
+  + **ImageNet Large scale Visual recognition challenge**  :  1,431,167 이미지를 1000개 카테고리로 분류하는 챌린지 개최<br>
+    
+     <img align="left" src="./images/6.PNG" width="40%" height="30%"></img> 
+      
+     <img align="left" src="./images/7.PNG" width="40%" height="30%"></img>    <br> <br><br><br><br><br><br><br><br><br><br>
+     
+      2011-2012 사이에 CNN 등장으로 급격한 error rate 감소, 이후에도 이를 tuning 한 매우 많은 layer의 NN가 계속 수상    <br>
+      
+      <img src="./images/8.PNG" width="40%" height="30%"></img>   <br>
+      
+      1998년에 이미 digit을 인식하기 위하여 연구가 있었으나 무어의 법칙으로 computation이 빠른 병렬 처리가 강력한 gpu 등장과, dataset의 많은 확보로 CNN과 같이
+      더 높은 capacity mdoel을 학습시키는것이 가능해짐 <br><br>
+
+## 1.3 Challenges
+    행동 인식, 상황 맥락을 파악하는것과 AR,VR과 같은 새로운 분야에서의 객체 탐지는 아직 도전과제이다. 어떤 그림을 보고 어떤 상황인지 알려면 배경지식이 필요하다.
+    이런 연구가 지속되면서 computer vision은 의료 진단, 자율주행,로보틱스 등에 효과적으로 적용될수 있고 인간의 지능을 이해하는데 도움이 될것이다.
+    
diff --git "a/week3_\354\225\210\354\204\234\354\227\260_preview.pdf" "b/week3_\354\225\210\354\204\234\354\227\260_preview.pdf"
new file mode 100644
index 0000000..f317434
Binary files /dev/null and "b/week3_\354\225\210\354\204\234\354\227\260_preview.pdf" differ
diff --git "a/week4_\354\225\210\354\204\234\354\227\260_preview.pdf" "b/week4_\354\225\210\354\204\234\354\227\260_preview.pdf"
new file mode 100644
index 0000000..bf62486
Binary files /dev/null and "b/week4_\354\225\210\354\204\234\354\227\260_preview.pdf" differ
diff --git "a/week6_\354\225\210\354\204\234\354\227\260_preview.pdf" "b/week6_\354\225\210\354\204\234\354\227\260_preview.pdf"
new file mode 100644
index 0000000..a8fe127
Binary files /dev/null and "b/week6_\354\225\210\354\204\234\354\227\260_preview.pdf" differ
diff --git "a/week7_\354\225\210\354\204\234\354\227\260_preview.pdf" "b/week7_\354\225\210\354\204\234\354\227\260_preview.pdf"
new file mode 100644
index 0000000..7fab33c
Binary files /dev/null and "b/week7_\354\225\210\354\204\234\354\227\260_preview.pdf" differ