Chapter8_Convlutional_Neural_Networks.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Chapter8_Convlutional_Neural_Networks",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyOEO3JlxKPCiQOgDTd60/iO",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/SokichiFujita/PyTorch-for-Deep-Learning-and-Computer-Vision/blob/master/Chapter8_Convlutional_Neural_Networks.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3RqovMu_PnJQ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import torch\n",
        "import matplotlib.pyplot as plt \n",
        "import numpy as np\n",
        "from torch import nn\n",
        "import torch.nn.functional as F\n",
        "from torchvision import datasets, transforms"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "snpdiujd7lMT",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "9a231f54-9edf-4cea-d401-f4bfaed17373"
      },
      "source": [
        "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
        "print(device)"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "cuda:0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8a__D4ENSAXW",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 255
        },
        "outputId": "1c3f47e4-8c52-4a51-da4f-8e3541864ad2"
      },
      "source": [
        "transform = transforms.Compose([transforms.Resize((28,28)),\n",
        "                                transforms.ToTensor(), \n",
        "                                transforms.Normalize((0.5,),(0.5,))\n",
        "                                ])\n",
        "training_dataset = datasets.MNIST(root='./data', train=True, download=True, transform=transform)\n",
        "validation_dataset = datasets.MNIST(root='./data', train=False, download=True, transform=transform)\n",
        "\n",
        "training_loader = torch.utils.data.DataLoader(training_dataset, batch_size=100, shuffle=True)\n",
        "validation_loader = torch.utils.data.DataLoader(validation_dataset, batch_size = 100, shuffle=False)"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "  0%|          | 16384/9912422 [00:00<01:30, 109893.84it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz to ./data/MNIST/raw/train-images-idx3-ubyte.gz\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "9920512it [00:00, 24030717.05it/s]                           \n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Extracting ./data/MNIST/raw/train-images-idx3-ubyte.gz to ./data/MNIST/raw\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "32768it [00:00, 323317.31it/s]                           \n",
            "0it [00:00, ?it/s]"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Downloading http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz to ./data/MNIST/raw/train-labels-idx1-ubyte.gz\n",
            "Extracting ./data/MNIST/raw/train-labels-idx1-ubyte.gz to ./data/MNIST/raw\n",
            "Downloading http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz to ./data/MNIST/raw/t10k-images-idx3-ubyte.gz\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "1654784it [00:00, 5268606.41it/s]                           \n",
            "8192it [00:00, 132908.37it/s]\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Extracting ./data/MNIST/raw/t10k-images-idx3-ubyte.gz to ./data/MNIST/raw\n",
            "Downloading http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz to ./data/MNIST/raw/t10k-labels-idx1-ubyte.gz\n",
            "Extracting ./data/MNIST/raw/t10k-labels-idx1-ubyte.gz to ./data/MNIST/raw\n",
            "Processing...\n",
            "Done!\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yuJZYOCYSPLi",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def image_convert(tensor):\n",
        "  image = tensor.cpu().clone().detach().numpy() #create numpy array from tensor\n",
        "  image = image.transpose(1, 2, 0)\n",
        "  print(image.shape)\n",
        "  image = image * np.array((0,5,0.5,0.5)) + np.array((0,5,0.5,0.5))\n",
        "  image = image.clip(0, 1)\n",
        "  return image\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cR36txpgTid9",
        "colab_type": "code",
        "outputId": "ca12539b-8a1b-484e-95f3-593d174d6aff",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 621
        }
      },
      "source": [
        "dataiter = iter(training_loader)\n",
        "images, labels = dataiter.next()\n",
        "fig = plt.figure(figsize=(25,4))\n",
        "\n",
        "for i in np.arange(20):\n",
        "  ax = fig.add_subplot(2, 10, i+1, xticks=[], yticks=[])\n",
        "  fig.add_subplot(2, 10, i+1)\n",
        "  plt.imshow(image_convert(images[i]))\n",
        "  ax.set_title([labels[i].item()])\n"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABYsAAAEICAYAAAAeHat/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAACS/g/gEJ+Jk04jKAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 1800x288 with 40 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ub4bIYHUT3gE",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "class LeNet(nn.Module):\n",
        "    def __init__(self):\n",
        "        super().__init__()\n",
        "        self.conv1 = nn.Conv2d(1, 20, 5, 1)\n",
        "        self.conv2 = nn.Conv2d(20, 50, 5, 1)\n",
        "        self.fc1 = nn.Linear(4 * 4 * 50, 500)\n",
        "        self.dropout1 = nn.Dropout(0.5)\n",
        "        self.fc2 = nn.Linear(500, 10)\n",
        "    def forward(self, x):\n",
        "        x = F.relu(self.conv1(x))\n",
        "        x = F.max_pool2d(x, 2, 2)\n",
        "        x = F.relu(self.conv2(x))\n",
        "        x = F.max_pool2d(x, 2, 2)\n",
        "        x = x.view(-1, 4 * 4 * 50)\n",
        "        x = F.relu(self.fc1(x))\n",
        "        x = self.dropout1(x)\n",
        "        x = self.fc2(x)\n",
        "        return x"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "h7iRzAmmgbnw",
        "colab_type": "code",
        "outputId": "82073ac2-0d11-418b-bc57-c33394048b48",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        }
      },
      "source": [
        "model = LeNet().to(device)\n",
        "print(model)"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "LeNet(\n",
            "  (conv1): Conv2d(1, 20, kernel_size=(5, 5), stride=(1, 1))\n",
            "  (conv2): Conv2d(20, 50, kernel_size=(5, 5), stride=(1, 1))\n",
            "  (fc1): Linear(in_features=800, out_features=500, bias=True)\n",
            "  (dropout1): Dropout(p=0.5, inplace=False)\n",
            "  (fc2): Linear(in_features=500, out_features=10, bias=True)\n",
            ")\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DEfjEbERgsN_",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "criterion = nn.CrossEntropyLoss()\n",
        "optimizer = torch.optim.Adam(model.parameters(), lr = 0.0001)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jmTc9a6Dg2cy",
        "colab_type": "code",
        "outputId": "84151653-5857-4c04-8cf9-4ef64f5d8ed8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 357
        }
      },
      "source": [
        "epochs = 10\n",
        "running_loss_history = []\n",
        "running_corrects_history = []\n",
        "val_running_loss_history = []\n",
        "val_running_corrects_history = []\n",
        "\n",
        "for e in range(epochs):\n",
        "  \n",
        "  running_loss = 0.0\n",
        "  running_corrects = 0.0\n",
        "  val_running_loss = 0.0\n",
        "  val_running_corrects = 0.0\n",
        "  \n",
        "  for inputs, labels in training_loader:\n",
        "    inputs = inputs.to(device)\n",
        "    labels = labels.to(device)\n",
        "    outputs = model(inputs)\n",
        "    loss = criterion(outputs, labels)\n",
        "    \n",
        "    optimizer.zero_grad()\n",
        "    loss.backward()\n",
        "    optimizer.step()\n",
        "    \n",
        "    _, preds = torch.max(outputs, 1)\n",
        "    running_loss += loss.item()\n",
        "    running_corrects += torch.sum(preds == labels.data)\n",
        "\n",
        "  else:\n",
        "    with torch.no_grad():\n",
        "      for val_inputs, val_labels in validation_loader:\n",
        "        val_inputs = val_inputs.to(device)\n",
        "        val_labels = val_labels.to(device)\n",
        "        val_outputs = model(val_inputs)\n",
        "        val_loss = criterion(val_outputs, val_labels)\n",
        "        \n",
        "        _, val_preds = torch.max(val_outputs, 1)\n",
        "        val_running_loss += val_loss.item()\n",
        "        val_running_corrects += torch.sum(val_preds == val_labels.data)\n",
        "      \n",
        "    epoch_loss = running_loss/len(training_loader)\n",
        "    epoch_acc = running_corrects.float()/ len(training_loader)\n",
        "    running_loss_history.append(epoch_loss)\n",
        "    running_corrects_history.append(epoch_acc)\n",
        "    \n",
        "    val_epoch_loss = val_running_loss/len(validation_loader)\n",
        "    val_epoch_acc = val_running_corrects.float()/ len(validation_loader)\n",
        "    val_running_loss_history.append(val_epoch_loss)\n",
        "    val_running_corrects_history.append(val_epoch_acc)\n",
        "\n",
        "    print('epoch: {} training loss: {:.7f}, accuracy: {:.7f} '.format(e+1, epoch_loss, epoch_acc.item()))\n",
        "    print('epoch: {} validate loss: {:.7f}, accuracy: {:.7f} '.format(e+1, val_epoch_loss, val_epoch_acc.item()))\n"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "epoch: 1 training loss: 0.5861845, accuracy: 84.4316711 \n",
            "epoch: 1 validate loss: 0.1848690, accuracy: 94.7099991 \n",
            "epoch: 2 training loss: 0.1565206, accuracy: 95.3166656 \n",
            "epoch: 2 validate loss: 0.1112543, accuracy: 96.6599960 \n",
            "epoch: 3 training loss: 0.1060580, accuracy: 96.8316727 \n",
            "epoch: 3 validate loss: 0.0824014, accuracy: 97.4700012 \n",
            "epoch: 4 training loss: 0.0811711, accuracy: 97.5083389 \n",
            "epoch: 4 validate loss: 0.0613420, accuracy: 98.1800003 \n",
            "epoch: 5 training loss: 0.0664433, accuracy: 98.0116653 \n",
            "epoch: 5 validate loss: 0.0536786, accuracy: 98.3699951 \n",
            "epoch: 6 training loss: 0.0579428, accuracy: 98.2550049 \n",
            "epoch: 6 validate loss: 0.0459960, accuracy: 98.6199951 \n",
            "epoch: 7 training loss: 0.0512607, accuracy: 98.4616699 \n",
            "epoch: 7 validate loss: 0.0425509, accuracy: 98.6599960 \n",
            "epoch: 8 training loss: 0.0438316, accuracy: 98.6866684 \n",
            "epoch: 8 validate loss: 0.0376707, accuracy: 98.7099991 \n",
            "epoch: 9 training loss: 0.0410614, accuracy: 98.7500000 \n",
            "epoch: 9 validate loss: 0.0391879, accuracy: 98.7500000 \n",
            "epoch: 10 training loss: 0.0363976, accuracy: 98.8833389 \n",
            "epoch: 10 validate loss: 0.0371443, accuracy: 98.6999969 \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NtEG_BWsi80_",
        "colab_type": "code",
        "outputId": "e12ef686-7278-475f-e2de-04f2a3b4738f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        }
      },
      "source": [
        "plt.plot(running_loss_history, label='training loss')\n",
        "plt.plot(val_running_loss_history, label='validation loss')\n",
        "plt.legend()"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f0173d89be0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD4CAYAAAD8Zh1EAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjIsIGh0\ndHA6Ly9tYXRwbG90bGliLm9yZy8li6FKAAAgAElEQVR4nO3deXxcdb3/8ddn9uwzTQJdkpCWFrqm\nnRJKsWKBgrel1yIo24WreEWu/OSiF+VSvArKffD4ofYiogUuoMgPF6xFvFUKFbAVkK0rpSvdmzS0\nTdImzT7b9/fHmSwNWSbNpJOZ+Twfj3nMzJkz53w6yvt88/2e8z1ijEEppVTysyW6AKWUUvGhga6U\nUilCA10ppVKEBrpSSqUIDXSllEoRjkTtuKCgwJSWliZq90oplZTWr19fY4wp7OmzhAV6aWkp69at\nS9TulVIqKYnIgd4+i6nLRUTmi8hOEdktIot7WedaEdkmIltF5DenWqxSSqlT028LXUTswFLgcqAS\nWCsiK4wx27qsMwG4B5hjjDkuImcMVcFKKaV6FksLfRaw2xiz1xgTAJ4Druy2zleApcaY4wDGmKPx\nLVMppVR/YulDHwNUdHlfCVzQbZ1zAETk74Ad+J4x5uXuGxKRW4FbAUpKSk6lXqXUIASDQSorK2lt\nbU10KaofHo+HoqIinE5nzN+J16CoA5gAXAwUAa+LyDRjTF3XlYwxTwBPAJSXl+skMkqdZpWVleTk\n5FBaWoqIJLoc1QtjDLW1tVRWVjJ27NiYvxdLl8shoLjL+6Losq4qgRXGmKAxZh/wIVbAK6WGkdbW\nVvLz8zXMhzkRIT8/f8B/ScUS6GuBCSIyVkRcwPXAim7r/BGrdY6IFGB1wewdUCVKqdNCwzw5nMr/\nTv0GujEmBNwOrAK2A8uMMVtF5H4RWRRdbRVQKyLbgNXAXcaY2gFXE4P1B47xg5d3oNP+KqXUyWI6\nD90Ys9IYc44x5mxjzAPRZfcaY1ZEXxtjzJ3GmMnGmGnGmOeGquAth07w2Jo9VNXroI5Syaauro5H\nH330lL57xRVXUFdX1+c69957L6+++uopbb+70tJSampq4rKt0yXp5nLxl3gB2HjweIIrUUoNVF+B\nHgqF+vzuypUr8Xq9fa5z//33c9lll51yfcku6QJ94shc3A4bGw/2faRWSg0/ixcvZs+ePcyYMYO7\n7rqLNWvWcNFFF7Fo0SImT54MwGc/+1nOO+88pkyZwhNPPNHx3fYW8/79+5k0aRJf+cpXmDJlCp/+\n9KdpaWkB4Oabb2b58uUd6993333MnDmTadOmsWPHDgCqq6u5/PLLmTJlCrfccgtnnXVWvy3xhx56\niKlTpzJ16lQefvhhAJqamli4cCHTp09n6tSp/O53v+v4N06ePJmysjK+9a1vxfcH7EfC5nI5VS6H\njWlj8rSFrtQgff9PW9lWdSKu25w8Opf7PjOl188ffPBBtmzZwqZNmwBYs2YNGzZsYMuWLR2n5/3i\nF79gxIgRtLS0cP755/O5z32O/Pz8k7aza9cufvvb3/Lkk09y7bXX8vzzz3PTTTd9bH8FBQVs2LCB\nRx99lCVLlvDUU0/x/e9/n0svvZR77rmHl19+mZ///Od9/pvWr1/P008/zbvvvosxhgsuuIC5c+ey\nd+9eRo8ezYsvvghAfX09tbW1vPDCC+zYsQMR6beLKN6SroUOVrfLlqoTtIXCiS5FKTVIs2bNOulc\n60ceeYTp06cze/ZsKioq2LVr18e+M3bsWGbMmAHAeeedx/79+3vc9tVXX/2xdd58802uv/56AObP\nn4/P5+uzvjfffJOrrrqKrKwssrOzufrqq3njjTeYNm0ar7zyCnfffTdvvPEGeXl55OXl4fF4+PKX\nv8wf/vAHMjMzB/pzDErStdAB/CU+nnxjH9s/amBGcd99akqpnvXVkj6dsrKyOl6vWbOGV199lbff\nfpvMzEwuvvjiHs/FdrvdHa/tdntHl0tv69nt9n776AfqnHPOYcOGDaxcuZLvfOc7zJs3j3vvvZf3\n3nuP1157jeXLl/Ozn/2Mv/71r3Hdb1+StoUOOjCqVLLJycmhoaGh18/r6+vx+XxkZmayY8cO3nnn\nnbjXMGfOHJYtWwbAX/7yF44f7ztHLrroIv74xz/S3NxMU1MTL7zwAhdddBFVVVVkZmZy0003cddd\nd7FhwwYaGxupr6/niiuu4Mc//jHvv/9+3OvvS1K20EflZTAy18PGg3V8aU6iq1FKxSo/P585c+Yw\ndepUFixYwMKFC0/6fP78+Tz++ONMmjSJc889l9mzZ8e9hvvuu48bbriBZ599lgsvvJCRI0eSk5PT\n6/ozZ87k5ptvZtasWQDccsst+P1+Vq1axV133YXNZsPpdPLYY4/R0NDAlVdeSWtrK8YYHnroobjX\n3xdJ1AU65eXlZjA3uLjtV+vZUlXPG/9xaRyrUiq1bd++nUmTJiW6jIRqa2vDbrfjcDh4++23ue22\n2zoGaYebnv73EpH1xpjyntZPyhY6WN0uL205THVDG4U57v6/oJRSwMGDB7n22muJRCK4XC6efPLJ\nRJcUN0kc6NbI9KaKOi6ffGaCq1FKJYsJEyawcePGRJcxJJJyUBRg6ug8HDbRgVGllIpK2kDPcNmZ\nNCpXrxhVSqmopA10sPrR36+sIxzRmReVUiqpA31miY/mQJgPj/R+XqtSSqWLpA70zguMtNtFqVSV\nnZ0NQFVVFZ///Od7XOfiiy+mv9OgH374YZqbmzvexzIdbyy+973vsWTJkkFvJx6SOtBLRmQyIsul\nA6NKpYHRo0d3zKR4KroHeizT8SabpA50EcFf7GWDBrpSSWHx4sUsXbq0431767axsZF58+Z1THX7\nv//7vx/77v79+5k6dSoALS0tXH/99UyaNImrrrrqpLlcbrvtNsrLy5kyZQr33XcfYE34VVVVxSWX\nXMIll1wCnHwDi56mx+1rmt7ebNq0idmzZ1NWVsZVV13VMa3AI4880jGlbvvEYH/729+YMWMGM2bM\nwO/39zklQqyS9jz0dv4SL6/tOEp9c5C8TGeiy1Eqeby0GA5/EN9tjpwGCx7s9ePrrruOb3zjG3zt\na18DYNmyZaxatQqPx8MLL7xAbm4uNTU1zJ49m0WLFvV6X83HHnuMzMxMtm/fzubNm5k5c2bHZw88\n8AAjRowgHA4zb948Nm/ezB133MFDDz3E6tWrKSgoOGlbvU2P6/P5Yp6mt90XvvAFfvrTnzJ37lzu\nvfdevv/97/Pwww/z4IMPsm/fPtxud0c3z5IlS1i6dClz5syhsbERj8cT88/cm6RuoUOXC4wqtR9d\nqeHO7/dz9OhRqqqqeP/99/H5fBQXF2OM4dvf/jZlZWVcdtllHDp0iCNHjvS6nddff70jWMvKyigr\nK+v4bNmyZcycORO/38/WrVvZtm1bnzX1Nj0uxD5NL1gTi9XV1TF37lwAvvjFL/L666931HjjjTfy\nq1/9CofDakfPmTOHO++8k0ceeYS6urqO5YOR9C30sqI8RKyZF+eeU5jocpRKHn20pIfSNddcw/Ll\nyzl8+DDXXXcdAL/+9a+prq5m/fr1OJ1OSktLe5w2tz/79u1jyZIlrF27Fp/Px80333xK22kX6zS9\n/XnxxRd5/fXX+dOf/sQDDzzABx98wOLFi1m4cCErV65kzpw5rFq1iokTJ55yrZACLfQcj5NzzsjR\nM12UShLXXXcdzz33HMuXL+eaa64BrNbtGWecgdPpZPXq1Rw4cKDPbXzqU5/iN7/5DQBbtmxh8+bN\nAJw4cYKsrCzy8vI4cuQIL730Usd3epu6t7fpcQcqLy8Pn8/X0bp/9tlnmTt3LpFIhIqKCi655BJ+\n8IMfUF9fT2NjI3v27GHatGncfffdnH/++R23yBuMpG+hQ+dEXZGIwWbruc9NKTU8TJkyhYaGBsaM\nGcOoUaMAuPHGG/nMZz7DtGnTKC8v77eletttt/GlL32JSZMmMWnSJM477zwApk+fjt/vZ+LEiRQX\nFzNnTuf82rfeeivz589n9OjRrF69umN5b9Pj9tW90ptnnnmGr371qzQ3NzNu3DiefvppwuEwN910\nE/X19RhjuOOOO/B6vXz3u99l9erV2Gw2pkyZwoIFCwa8v+6Sdvrcrn639iB3P/8Br31zLmcXZsdl\nm0qlIp0+N7kMdPrcpO9ygc6BUe12UUqls5QI9PGF2eS4HXqBkVIqraVEoNtswvRir7bQlYpBorpZ\n1cCcyv9OKRHoYA2M7jh8guZAfO/srVQq8Xg81NbWaqgPc8YYamtrB3yxUUxnuYjIfOAngB14yhjz\nYLfPbwZ+BByKLvqZMeapAVUySP4SLxEDmyvrmT0u/3TuWqmkUVRURGVlJdXV1YkuRfXD4/FQVFQ0\noO/0G+giYgeWApcDlcBaEVlhjOl++dXvjDG3D2jvcTSjuHNgVANdqZ45nU7Gjh2b6DLUEImly2UW\nsNsYs9cYEwCeA64c2rIGbkSWi9L8TB0YVUqlrVgCfQxQ0eV9ZXRZd58Tkc0islxEinvakIjcKiLr\nRGTdUPzJ5y/xsbGiTvsHlVJpKV6Don8CSo0xZcArwDM9rWSMecIYU26MKS8sjP+8K/4SL9UNbRyq\nO7X5FpRSKpnFEuiHgK4t7iI6Bz8BMMbUGmPaom+fAs6LT3kD4y/WC4yUUukrlkBfC0wQkbEi4gKu\nB1Z0XUFERnV5uwjYHr8SYzdxVA5uh00DXSmVlvo9y8UYExKR24FVWKct/sIYs1VE7gfWGWNWAHeI\nyCIgBBwDbh7CmnvltNsoK8pjY4UOjCql0k9M56EbY1YCK7stu7fL63uAe+Jb2qmZWeLj6b/vpy0U\nxu2wJ7ocpZQ6bVLmStF2/hIvgXCEbVUnEl2KUkqdVikY6DowqpRKTykX6Gfmehid52FjhQa6Uiq9\npFygQ/QCI71iVCmVZlI00L1UHm/haMOp3xxWKaWSTcoGOsAm7UdXSqWRlAz0KaPzcNpF+9GVUmkl\nJQPd47QzeVQuGw5oP7pSKn2kZKCDNTC6ubKeUDiS6FKUUuq0SOFA99ISDLPzSEOiS1FKqdMidQNd\nZ15USqWZlA304hEZ5Ge5NNCVUmkjZQNdRPCXeHXmRaVU2kjZQAdrYHRvdRN1zYFEl6KUUkMutQO9\nOHqBkZ6PrpRKAykd6GXFXkR0YFQplR5SOtCz3Q7OPTNHrxhVSqWFlA50sM5H33TwOJGISXQpSik1\npFI/0It9nGgNsbemKdGlKKXUkEr9QI/OvKjzoyulUl3KB/rZhdnkuB3aj66USnkpH+g2mzCjxKtn\nuiilUl7KBzpYFxjtPHyCprZQoktRSqkhkyaB7iViYHNlfaJLUUqpIZMWgT6jKDowqvO6KKVSWFoE\nui/LxbiCLO1HV0qltLQIdKBjYNQYvcBIKZWaYgp0EZkvIjtFZLeILO5jvc+JiBGR8viVGB/+Eh81\njW1UHm9JdClKKTUk+g10EbEDS4EFwGTgBhGZ3MN6OcDXgXfjXWQ8tM+8qOejK6VSVSwt9FnAbmPM\nXmNMAHgOuLKH9f4L+AHQGsf64mbiyBw8TpteMaqUSlmxBPoYoKLL+8rosg4iMhMoNsa82NeGRORW\nEVknIuuqq6sHXOxgOOw2yor0AiOlVOoa9KCoiNiAh4Bv9reuMeYJY0y5Maa8sLBwsLseMH+Jl21V\nJ2gLhU/7vpVSaqjFEuiHgOIu74uiy9rlAFOBNSKyH5gNrBiWA6PFPgLhCFurTiS6FKWUirtYAn0t\nMEFExoqIC7geWNH+oTGm3hhTYIwpNcaUAu8Ai4wx64ak4kHonHlRu12UUqmn30A3xoSA24FVwHZg\nmTFmq4jcLyKLhrrAeDoz18MYbwYbdGBUKZWCHLGsZIxZCazstuzeXta9ePBlDZ0ZJV42aQtdKZWC\n0uZK0Xb+Yi+H6lo4cmJYnl2plFKnLP0CvcQHaD+6Uir1pF2gTxmdi9MuOvOiUirlpF2ge5x2Jo/O\n0xa6UirlpF2gg9WPvrmyjlA4kuhSlFIqbtIz0Eu8tAYj7DjckOhSlFIqbtIy0Ge2D4zqzItKqRSS\nloFe5MugINulMy8qpVJKWga6iOAv8ekFRkqplJKWgQ5WP/remiaONwUSXYpSSsVF+gZ6sdWPvqlS\nW+lKqdSQtoFeVpSHTfSKUaVU6kjbQM9yOzh3ZK4OjCqlUkbaBjpY/eibKuqIREyiS1FKqUFL70Av\n9tLQGmJvTWOiS1FKqUFL70CPXmC0QfvRlVIpIK0DfVxBFrkehw6MKqVSQloHus0mzCjx6cCoUiol\npHWgg9WP/uGRBhrbQokuRSmlBkUDvcRLxMBmvcBIKZXk0j7QZxR7Ab3ASCmV/NI+0L2ZLsYVZmmg\nK6WSXtoHOljzumyqOI4xeoGRUip5aaBj9aPXNAaoPN6S6FKUUuqUaaBjBTrABj19USmVxDTQgXPP\nzCHDadd+dKVUUtNABxx2G2VFeXqBkVIqqcUU6CIyX0R2ishuEVncw+dfFZEPRGSTiLwpIpPjX+rQ\n8pf42Fp1gtZgONGlKKXUKek30EXEDiwFFgCTgRt6COzfGGOmGWNmAD8EHop7pUPMX+IlFDFsrapP\ndClKKXVKYmmhzwJ2G2P2GmMCwHPAlV1XMMac6PI2C0i68//aB0a1H10plawcMawzBqjo8r4SuKD7\nSiLyNeBOwAVc2tOGRORW4FaAkpKSgdY6pM7I8VDky9BAV0olrbgNihpjlhpjzgbuBr7TyzpPGGPK\njTHlhYWF8dp13Ph15kWlVBKLJdAPAcVd3hdFl/XmOeCzgykqUfzFXqrqWzlc35roUpRSasBiCfS1\nwAQRGSsiLuB6YEXXFURkQpe3C4Fd8Svx9GnvR99Uoa10pVTy6TfQjTEh4HZgFbAdWGaM2Soi94vI\nouhqt4vIVhHZhNWP/sUhq3gITR6di8tu0350pVRSimVQFGPMSmBlt2X3dnn99TjXlRBuh50pY3I1\n0JVSSUmvFO3GX+xj86E6guFIoktRSqkB0UDvxl/ipTUYYefhhkSXopRSA6KB3k3nBUY6MKqUSi4a\n6N2M8WZQmOPWfnSlVNLRQO9GRPAXe9lYoYGulEouGug98Jf42FfTxPGmQKJLUUqpmGmg96DzAiNt\npSulkocGeg/KivKwiQ6MKqWSiwZ6DzJdDiaOzNV+dKVUUtFA74W/xMumg3VEIkk3tbtSKk1poPfC\nX+KjoS3EnurGRJeilFIx0UDvhd7BSCmVbDTQezE2P4u8DCcbdSpdpVSS0EDvhc0mzCj2agtdKZU0\nNND74C/xsvNIA41toUSXopRS/dJA74O/xIcx8L6evqiUSgIa6H2YUawzLyqlkocGeh/yMpyMPyNb\n+9GVUklBA70f7TMvGqMXGCmlhjcN9H74S3wcawpw8FhzoktRSqk+aaD3Qy8wUkolCw30fpxzZg6Z\nLrsOjCqlhj0N9H7YbcL0Ir2DkVJq+NNAj4G/xMu2qhO0BsOJLkUppXqlgR4Df4mPUMSw5VB9oktR\nSqleaaDHoPMCI+12UUoNXxroMSjMcVM8IkNnXlRKDWsxBbqIzBeRnSKyW0QW9/D5nSKyTUQ2i8hr\nInJW/EtNLH+xT1voSqlhrd9AFxE7sBRYAEwGbhCRyd1W2wiUG2PKgOXAD+NdaKL5S7x8VN/KR/Ut\niS5FKaV6FEsLfRaw2xiz1xgTAJ4Druy6gjFmtTGm/VLKd4Ci+JaZeP4SHwCbtJWulBqmYgn0MUBF\nl/eV0WW9+TLwUk8fiMitIrJORNZVV1fHXuUwMHlULi6HTc9HV0oNW3EdFBWRm4By4Ec9fW6MecIY\nU26MKS8sLIznroecy2Fj6uhcvWJUKTVsxRLoh4DiLu+LostOIiKXAf8JLDLGtMWnvOHFX+Jjc2U9\nwXAk0aUopdTHxBLoa4EJIjJWRFzA9cCKriuIiB/4H6wwPxr/MocHf4mXtlCEHR81JLoUpZT6mH4D\n3RgTAm4HVgHbgWXGmK0icr+ILIqu9iMgG/i9iGwSkRW9bG7wAs2w/+9Dtvm+tA+M6vnoSqnhyBHL\nSsaYlcDKbsvu7fL6sjjX1bs3/hvefAgu+U/45J1gO33XRo3O83BGjpuNB+v4woWnbbdKKRWT5LtS\n9KI7YcpV8Nf/gt/dBK2nb34VEcFf4tWBUaXUsJR8ge7Kgs/9HOb/AHatgicuhiPbTtvu/SU+9tc2\nc6wpcNr2qZRSsUi+QAcQgdlfhS/+GQJN8NQ8+GD5adm1PzpR1ybtR1dKDTPJGejtzroQ/vV1GDUd\nnv8yvLQYwsEh3WVZkRe7TXReF6XUsJPcgQ6QMxK++Ce44DZ49zF45jPQcHjIdpfhsjNpVI4GulJq\n2En+QAewO2HBg1bf+kfvw/98Cg68PWS78xf72FRRRzhihmwfSik1UKkR6O2mfR5uedUaOH3mH+Gd\nx8HEP3T9JV4a20LsPtoY920rpdSpSq1ABzhzCty6BiZ8Gl6+G/7wFWvgNI46LjDS0xeVUsNI6gU6\ngCcPrvs1XPpd6+yXpy6D2j1x23xpfibeTKf2oyulhpXUDHSwriD91LfgpuetQdInLoYdL8Zl0yLC\nzBIfL289zLJ1FdqXrpQaFlI30NuNnwf/+jcYMQ6e+yd47X6IhAe92cULJlJakMV/LN/M/Idf5y9b\nD2OGoL9eKaVilfqBDuAtgX9ZBf5/tuaC+dXnoKl2UJs858wc/vh/PsHjN80kHDHc+ux6Pv/426zd\nfyxORSul1MBIolqV5eXlZt26dad/x+ufgZXfguwz4dr/B2NmDnqToXCE36+v5MevfMjRhjYum3QG\nd/3DRM4dmROHgpVSqpOIrDfGlPf4WdoFOsCh9bDsi9B4FBYugZlfiMtmWwJhnn5rH4+t2UNjW4ir\n/UX8++UTKPJlxmX7Simlgd6Tplp4/l9g7xor0Bf8CJyeuGy6rjnAo2v28Mu39oOBL1x4Fl+7ZDy+\nLFdctq+USl8a6L2JhGH1A1a/+mi/1QXjLYnb5qvqWnj41Q9Zvr6SLJeDf507jn/55FgyXTFNQ6+U\nUh+jgd6fHS/CC18FmwM+/3M4+9K4bn7XkQZ+uGonr2w7QmGOm6/Pm8B15xfjtKfHmLRSKn76CnRN\nFICJC+Erq62B0mevhteXQCR+N4KecGYOT36hnOdvu5DS/Ey+88ctfPrHr/Pi5o/0VEelVNxoC72r\nQBOs+DfY8jycuxCuesy66jSOjDH8dcdRfvjyTnYeaaCsKI+7509kzviCuO5HKZWatMtlIIyBdx+H\nv3zH6k+/7tdw5uS47yYcMbyw8RA/fuVDDtW1cNGEAu6eP5GpY+J7AFFKpRYN9FNx4C34/c3Q1gCL\nfmrN5DgEWoNhfvXOAX62ejd1zUE+M3003/r0OZyVnzUk+1NKJTcN9FPVcNg6X73iHesGGp/+L2vu\n9SFwojXIE3/by1Nv7iUUNvzTBSX826UTKMxxD8n+lFLJSQN9MMJBq/vl3ceh5EK45pfWXZKGyNET\nrfzktV08t7YCt8PGLReN4ysXjSXHMzQHEqVUctFAj4fNv4c/3QHuHLjmGet+pkNob3Uj//2XD3nx\ng48YkeXi9kvGc+PsEtwO+5DuVyk1vOlpi/FQds1puRtSu3GF2Sy9cSYrbp/DxJE53P/nbcz777/x\nwsZKIjpdr1KqB9pCH6iWOusipA9fgvGXwbRrYcLlkDliyHZpjOGNXTX84OUdbK06wcSROdy9YCIX\nn1OIiAzZfpVSw492ucRbJAJvPQLvPAqNR0DscNYn4NwFcO4VMGLsEO3W8OcPPmLJqp0cPNbMBWNH\nsHjBxI5b4imlUt+gA11E5gM/AezAU8aYB7t9/ingYaAMuN4Ys7y/bSZ1oLeLRKBqI+x8EXa+BEe3\nWcvPmBwN94XWHDG2+PZsBUIRnlt7kEde20VNY4DzS318cnwhn5yQT1mRV6cUUCqFDSrQRcQOfAhc\nDlQCa4EbjDHbuqxTCuQC3wJWpE2gd3dsL+x8GXautM5jN2HIHgnnzrda7mPnxm1GR4CmthC/fGs/\nq7Ye5oND9RgDWS47s8flM2d8AZ+cUMCEM7K1W0apFDLYQL8Q+J4x5h+i7+8BMMb83x7W/SXw57QN\n9K6aj8GuV6zW++7XINAIziw4+xJr7pgJ/wBZ+XHbXV1zgLf31PLm7hre2lPLvpomAApz3Mw52wr4\nOeMLGO3NiNs+lVKnX1+BHss8rmOAii7vK4ELTrGQW4FbAUpK4jdN7bCUOQKmX2c9Qm2w7w2r5b7z\nJdjxZxAbFM+GiVdYrff8swe1O2+miwXTRrFg2igAKo8389ZuK+Df3F3DHzdVATCuIKsj3C8cl09e\npp7frlSqiKWF/nlgvjHmluj7fwYuMMbc3sO6v0Rb6H0zBj7aBDui4X7kA2t5wblWv/vEhTCmPK79\n7sYYdh5p4M1dNfx9dw3v7jtGcyCMTWDamDyre2Z8ATPP8uFx6nnuSg1n2uUynB0/AB++bM3JfuDv\nEAlBViGcE+13H3cxuOJ7C7tAKML7lXUdAb+poo5QxOB22Di/dERHwE8enYvdpv3vSg0ngw10B9ag\n6DzgENag6D8ZY7b2sO4v0UA/dS11sPtVq2tm1yvQdgIcGdYNN85dYIV8dmHcd9vYFuK9fbW8uauW\nv++uYeeRBgDyMpx8okv/e2l+pg6wKpVg8Tht8Qqs0xLtwC+MMQ+IyP3AOmPMChE5H3gB8AGtwGFj\nzJS+tqmB3o9QwGqxt/e711cAAsWzrJb7xIVQMGFIdn20odUaYI224KvqWwEY481gzngr4D9xdoFO\nHKZUAuiFRcnOGDj8QTTcV8JH71vL88db4X72pTDmPPDkDsGuDftrm3lzdw1/31XDW3tqONEaAmDi\nyJyO7plZY0eQ5dZ7pSo11DTQU019pdVq37nSOnsmEgQEzpwCRedbrfiiWdaZM3HuIglHDFur6q2A\n313D2v3HCYQiOGzCuMIsxp+RzdmFnY9xhVka9ErFkQZ6Kms9AZVrrUfFe1C5Dtrqrc8y862ALzof\nii+AMTOtycXiuftgmPUHjvPWnhp2Hm5kb3UjB441E+4ygdioPE804LM4u0vgn5nr1j55pQZIAz2d\nRCJQs9MK94r3oPI9qPnQ+h5csYQAAAqhSURBVEzsViu+eJYV8EXng6807q34tlCYg7XN7KluZE91\nE3uONna8bmwLdayX7XZYIV+YHQ166/VZ+Vm4HDp9gVI90UBPd83HrJZ7ZTTkD623rlwF6xTJ9nAv\nnmXNPeMcmqtJjTEcbWhjz9FGdlc3RoO+iT3VjXwUHXgFsNuEkhGZnWFfmM3ZZ1ivvZmuIalNqWSh\nga5OFglbE4lVvBftqnnXmocGwOaAkWXRfvhoV01eUdxb8d01toXYFw33jsfRJvbVNBEIRzrWK8h2\nMa6jj97qwhlfmM1ob4aeM6/Sgga66l9TTWe4V6yFqg0QbLY+yxnVGe7Fs2DUdHCcnlMWwxFDxbHm\nk0J+T7XVwq9rDnas53bYGFuQFe2b91CQ46Ig201htpuCbDcFOS7ys9zalaOSnga6GrhwEI5sscK9\n8j0r6OsOWp/ZXTBqRpdW/CzIHX3aSzzWFIiGvBX2u482sq+miaMNbTQHwj1+x5vptAI+2xV9dlOY\nEw3+nM5l+dkuvd2fGpY00FV8NBzpDPeKtdZc8OE267PcImuANSsfMgsgqyD63PV9vvWwD/2EYM2B\nEDUNAaob26iJPqoboq8bAh3LahoDJw3UdpXrcVDQEfbtrf3O0C/IsQ4GBRr+6jTSQFdDIxSAw5uj\nA63r4ESV1XXTXAMtx3v/nievn9DvtmyIBmnbtQTCVuA3tlHTYIV8TZcDQfsBoLqhjYZewj/H4zip\ne6cg240304U3w4k303rkZbjwZTrxZrrI9Thw6I1I1CkY7PS5SvXM4YKicuvRXThkhXpzTWfIN9VA\nc631aF92fL91MGiutSYm64kzqzPgM/M7g7/jgND1wJAP7twBDeJmuOwUj8ikeET/k6C1BsMdLfua\nhraTWvrVDdZBYcfhBmobaznRGuzzPuI5HocV9hmuaOh3HgDyMpz4Ml0nHQysdZ16IFC90kBXQ8Pu\nsCYSi3UyMWOgtf7ksO94jh4EmmugqRqqd1ifhVp62ber55Z+19DvuizDF/N0xR6nnSJfJkW+/sM/\nHDE0tAapaw5yvDlAXUuQ+uYgddHXdd1eVx5voa45QH1LkEhfBwK3g7xo0HceDDpf52VYBwdfdPmI\nLDd5GU49CygNaKCr4UEEMrzWI9abfQSauhwAejsQ1FhTFDcf67yC9mP7tkHGiD5Cv/tfA7GNA9ht\nYrW6M12UEvsVupGIoaE1RF1LwAr9lmjwN0cPAi0B68DQYh0oqupaOtbp7UAgAt4MJ74sFyMyXYzI\nsh69vs92keWy65W8SUYDXSUvV5b18MZ496tQoEtLv4+DwNHt1vuW40AvCenJ6zv0M0ZYff/OTOs+\nsg5P52tnpvVXRC9habMJeZlO8jKdnDWAuxRGIobGQIi6pmDHweB4c4DjTQGONQet5+jj4LFmNlXU\ncbw5QDDc87/RZbfhy7K6fvKzXfiiwe/rcgDo+t6X5dTB4QTTQFfpw+GC3FHWIxaRsBXqvbX8O8YB\nDlhX3/Y1DvAxEg38DGvOe6eny+v25V0OAh3Lo8scno993+bMJNfhIdeZSUmOB3wZ4BphrdvLwcMY\nQ0NbqCPsjzcHqG20no81RQ8CzdZn26pOcCz6l0Jvst0OfFnOjlZ/e4s/x+PE5bDhdti6Pdtxd3vf\n23pOu+hfDP3QQFeqNza71drOKoht/a7jAM3HrAuzQq0QbLEeoZYur1u7vW6GYGvnd1qO97BeC5hI\n/3V0J3ZwZ1uDxa5s67UrG9w5iDuHXFc2ue4czuqynDNyOtd153W+dmYRMljdPV0OAseaghxrarMO\nAtEDQG1TgA+PNHKsKUBLsOfrAgaq9wNB/wcEt8NGhtNOlttBtttBprvLa5edbLeDLLeDLJcDj9OW\nlAcPDXSl4uVUxgEGwhjrgq+OA0X0INBxoGg9+aARbIFgE7Q1WnP3tDVCoAHaoo+Gj6xlbQ3W8pgO\nFoLDlU2BO5uC9vB3Z4Mr+uzOgRHZMKp9mbU8Ik5CoQChYJBg9DkcChEKBQl3eURCQSLhEJFwMPoI\nEQmHMOEQJhzEREIQDlnP7Y9gCNrCSCSEmDBiQtiir20djxA2EyaMjRB2gjgI4CBk7DThoC66LNTx\n7MDYnYjdCTYnNocLm8OFOFzYHS7sThcOpwuH04PD5cLlcuN0eXC7XbhcHtxuN25PBh63iwxPBh6P\nB7G7rLEXm9MaiI/zrSVBA12p5CFidRs5hmCCMmOsA0BbQzT8uzx3HAgae/+8+YB1y8T2g0c4cNLm\nbYAr+oiZ2Ky5hToe9uizs/O9o6fPM6LBaT/5+2IjYgzhYJv1CAUxoQCRcBDCLZhwEAkHIBJCwkFs\nJogtEsIeCmILhnAQn78yADZPv5eyq74Zt+2100BXSlkHC1dmtNV45uC3F2o7+UAQCXYJYsfHw7b9\nvT26jthjPpV0IGzRxyldqxyJWH8RhAPWvyccJBhopaWtjdaWFlpaW2lpbaWtrY22Nus5EGgl2NZG\nINBGKNhGMBAgFGxjYvEn4vsPi9JAV0rFn8NtPbIGcJrOcGezge3kv5Cc0Uf8b/54avSSM6WUShEa\n6EoplSI00JVSKkVooCulVIrQQFdKqRShga6UUilCA10ppVKEBrpSSqWIhN2CTkSqgQOn+PUCoCaO\n5SQ7/T1Opr9HJ/0tTpYKv8dZxpge7xyTsEAfDBFZ19s99dKR/h4n09+jk/4WJ0v130O7XJRSKkVo\noCulVIpI1kB/ItEFDDP6e5xMf49O+lucLKV/j6TsQ1dKKfVxydpCV0op1Y0GulJKpYikC3QRmS8i\nO0Vkt4gsTnQ9iSIixSKyWkS2ichWEfl6omsaDkTELiIbReTPia4l0UTEKyLLRWSHiGwXkQsTXVOi\niMi/R/872SIivxURT6JrGgpJFegiYgeWAguAycANIjI5sVUlTAj4pjFmMjAb+Foa/xZdfR3Ynugi\nhomfAC8bYyYC00nT30VExgB3AOXGmKmAHbg+sVUNjaQKdGAWsNsYs9cYEwCeA65McE0JYYz5yBiz\nIfq6Aes/1jGJrSqxRKQIWAg8lehaEk1E8oBPAT8HMMYEjDF1ia0qoRxAhog4gEygKsH1DIlkC/Qx\nQEWX95WkeYgBiEgp4AfeTWwlCfcw8B9AJNGFDANjgWrg6WgX1FMikpXoohLBGHMIWAIcBD4C6o0x\nf0lsVUMj2QJddSMi2cDzwDeMMScSXU+iiMg/AkeNMesTXcsw4QBmAo8ZY/xAE5CWY04i4sP6S34s\nMBrIEpGbElvV0Ei2QD8EFHd5XxRdlpZExIkV5r82xvwh0fUk2BxgkYjsx+qKu1REfpXYkhKqEqg0\nxrT/1bYcK+DT0WXAPmNMtTEmCPwB+ESCaxoSyRboa4EJIjJWRFxYAxsrElxTQoiIYPWPbjfGPJTo\nehLNGHOPMabIGFOK9f+LvxpjUrIVFgtjzGGgQkTOjS6aB2xLYEmJdBCYLSKZ0f9u5pGiA8SORBcw\nEMaYkIjcDqzCGqn+hTFma4LLSpQ5wD8DH4jIpuiybxtjViawJjW8/Bvw62jjZy/wpQTXkxDGmHdF\nZDmwAevssI2k6BQAeum/UkqliGTrclFKKdULDXSllEoRGuhKKZUiNNCVUipFaKArpVSK0EBXSqkU\noYGulFIp4v8DrmDiyc2lgHIAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "m5Wingd2i9jk",
        "colab_type": "code",
        "outputId": "edaacc31-e86a-4b33-c7ab-0f5167fb3728",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        }
      },
      "source": [
        "plt.plot(running_corrects_history, label='training accuracy')\n",
        "plt.plot(val_running_corrects_history, label='validation accuracy')\n",
        "plt.legend()"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f0173a47780>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXAAAAD4CAYAAAD1jb0+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nK6WUTWmAK6WUTWmAK6WUTf1/M3vmyz8+vdkAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8bBHP-3Tujg5",
        "colab_type": "code",
        "outputId": "19edaa2e-e0e0-459a-bf97-f65bb5031876",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "!pip install pillow"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: pillow in /usr/local/lib/python3.6/dist-packages (6.2.2)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "k-SUReA8tEVz",
        "colab_type": "code",
        "outputId": "4c5ad5b4-68f3-4749-a92a-002125a43061",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 303
        }
      },
      "source": [
        "import requests\n",
        "from PIL import Image\n",
        "\n",
        "url = 'https://images.homedepot-static.com/productImages/007164ea-d47e-4f66-8d8c-fd9f621984a2/svn/architectural-mailboxes-house-letters-numbers-3585b-5-64_1000.jpg'\n",
        "response = requests.get(url, stream=True)\n",
        "print(response)\n",
        "img = Image.open(response.raw)\n",
        "plt.imshow(img)"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<Response [200]>\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f0173e7cba8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQEAAAD8CAYAAAB3lxGOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nPgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RHabVrOotzQW",
        "colab_type": "code",
        "outputId": "a372d6a7-02cd-4f93-ebe9-92966f4db50d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 299
        }
      },
      "source": [
        "import PIL.ImageOps\n",
        "\n",
        "img = PIL.ImageOps.invert(img)\n",
        "img = img.convert('1')\n",
        "img = transform(img) \n",
        "plt.imshow(image_convert(img))"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(28, 28, 1)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.image.AxesImage at 0x7f01c7de81d0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD4CAYAAAAq5pAIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSfwPt/iuapolhWUAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GjXZaPeet10V",
        "colab_type": "code",
        "outputId": "13d79d76-d456-47c3-b3c7-3f0342007793",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "images = img.to(device)\n",
        "image = images[0].unsqueeze(0).unsqueeze(0)\n",
        "output = model(image)\n",
        "_, pred = torch.max(output, 1)\n",
        "print(pred.item())"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "5\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uF3b2GPwu3uJ",
        "colab_type": "code",
        "outputId": "12d6a219-d24c-4808-f29e-0312a4e6bf3c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 608
        }
      },
      "source": [
        "dataiter = iter(validation_loader)\n",
        "images, labels = dataiter.next()\n",
        "images = images.to(device)\n",
        "labels = labels.to(device)\n",
        "output = model(images)\n",
        "_, preds = torch.max(output, 1)\n",
        "\n",
        "fig = plt.figure(figsize=(25, 4))\n",
        "\n",
        "for idx in np.arange(20):\n",
        "  ax = fig.add_subplot(2, 10, idx+1, xticks=[], yticks=[])\n",
        "  plt.imshow(image_convert(images[idx]))\n",
        "  ax.set_title(\"{} ({})\".format(str(preds[idx].item()), str(labels[idx].item())), color=(\"red\" if preds[idx]==labels[idx] else \"red\"))"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n",
            "(28, 28, 1)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABXEAAAD7CAYAAAAsAtcsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 1800x288 with 20 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    }
  ]
}