animal_person_other_v2_final_version.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "animal_person_other_v2_final_version.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "machine_shape": "hm",
      "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/oleksandr-g-rock/How_to_create-image-classification-for-recognizing-persons-animals-others/blob/main/animal_person_other_v2_final_version.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vMC7Xf1v6Ydk"
      },
      "source": [
        "# Notebook for predict:\n",
        "# person, animals (dog, squirrel, horse, chicken, cats, raccoon, deer),other.\n",
        "# used MobileNetV2 model"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OeVxYAPJ7Ud7"
      },
      "source": [
        "connect to GD"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "a388Wkj27X2C"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BKbzm6lHUWgY"
      },
      "source": [
        "Need download Dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "q5lmfgsPnEcr"
      },
      "source": [
        "!cp drive/My\\ Drive/animals_and_person.zip /content/\n",
        "!wget https://raw.githubusercontent.com/oleksandr-g-rock/How_to_create-image-classification-for-recognizing-persons-animals-others/main/animal_person_other_sort.sh"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ss2-7DkXPNti"
      },
      "source": [
        "Need unzip folder"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XatbU-BJUesS"
      },
      "source": [
        "!jar xvf /content/animals_and_person.zip\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OQGDcVtW7uMR"
      },
      "source": [
        "sorted all files to train and test"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XGLNptb3otut"
      },
      "source": [
        "RENAME MAIN FOLDER"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "u6Y1Z13totuv"
      },
      "source": [
        "!mv /content/animals_and_person /content/animals_and_person"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kbKGu3Bc7001"
      },
      "source": [
        "!cat /content/animal_person_other_sort.sh\n",
        "!bash /content/animal_person_other_sort.sh"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mNJfXYZFNRho"
      },
      "source": [
        "Convert full data to jpg and after that remove all png "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "A_-ZGb7pF3VH"
      },
      "source": [
        "!apt-get install parallel \n",
        "!apt-get install imagemagick\n",
        "!apt -y update\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dkwTxuJZFfMJ"
      },
      "source": [
        "!chmod -R 777 /content/animals_and_person/\n",
        "\n",
        "!ls /content/animals_and_person/train/ | while read -r NAME; do ls /content/animals_and_person/train/$NAME/*.png | parallel convert '{}' '{.}.jpg' || true ; done\n",
        "\n",
        "!ls /content/animals_and_person/val/ | while read -r NAME; do ls /content/animals_and_person/val/$NAME/*.png | parallel convert '{}' '{.}.jpg' || true ; done\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nppyM5yKYy7D"
      },
      "source": [
        "!ls -la /content/animals_and_person/train/other"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IUmxU0MwJyPU"
      },
      "source": [
        "!find /content/animals_and_person -name '*.png' -exec rm -f {} +"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3VSa3Kd-4CZu"
      },
      "source": [
        "add testing folder and move 10 files from val from each class"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6nnisQ8b4C_Q"
      },
      "source": [
        "!mkdir /content/animals_and_person/testing\n",
        "!cp -r /content/animals_and_person/train/* /content/animals_and_person/testing/"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "s-0pz_pbaIuJ"
      },
      "source": [
        "!ls /content/animals_and_person/testing/ | while read -r NAME; do cd /content/animals_and_person/testing/\"${NAME}\" && rm -rf * ; done"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "d7Gl3ru_aG-_"
      },
      "source": [
        "!ls /content/animals_and_person/val/ | while read -r NAME; do cd /content/animals_and_person/val/\"${NAME}\" && ls | tail -100 | xargs -I{} sudo mv {} /content/animals_and_person/testing/\"${NAME}\"/ ; done"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "k-kJsVU6925E"
      },
      "source": [
        "**START PREPARE AND RUN TRAIN OUR MODEL**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1mIwTYFprNsR"
      },
      "source": [
        "Download modules"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3NFfovZxN9Ar"
      },
      "source": [
        "from tensorflow.keras.preprocessing import image\n",
        "from tensorflow.keras.models import Model, load_model\n",
        "from tensorflow.keras.callbacks import ModelCheckpoint, ReduceLROnPlateau\n",
        "from tensorflow.keras.layers import Dense, GlobalAveragePooling2D, Dropout\n",
        "from tensorflow.keras.layers import BatchNormalization, Activation\n",
        "from tensorflow.python.keras.optimizers import Adam, RMSprop\n",
        "from tensorflow.python.keras.preprocessing.image import ImageDataGenerator\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import tensorflow as tf\n",
        "from numpy import loadtxt\n",
        "import tensorflow.keras\n",
        "from PIL import Image, ImageOps\n",
        "%matplotlib inline"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Aa9dmY8OrRbK"
      },
      "source": [
        "Set image size & batch size"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EVeEdtpWraWm"
      },
      "source": [
        "image_size = 299\n",
        "batch_size = 64"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oBZs2T4kscQW"
      },
      "source": [
        "# Create image generator"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ua6InEjAs5FE"
      },
      "source": [
        "# folders with train dir & val dir\n",
        "train_dir = '/content/animals_and_person/train/'\n",
        "test_dir = '/content/animals_and_person/val/'\n",
        "testing_dir = '/content/animals_and_person/testing/'\n",
        "input_shape = (image_size, image_size, 3)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OcudjcroS-QY"
      },
      "source": [
        "Create image data generator for train data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "c2YgG5lHtK7Q"
      },
      "source": [
        "train_datagen = ImageDataGenerator(rescale=1. / 255,\n",
        "                              height_shift_range= 0.02, \n",
        "                              width_shift_range=0.02, \n",
        "                              rotation_range=0.02, \n",
        "                              shear_range = 0.01,\n",
        "                              fill_mode='nearest',\n",
        "                              zoom_range=0.01)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8QLN9jAPtQv_"
      },
      "source": [
        "Check generator"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SnOJaWgttOG8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 522
        },
        "outputId": "ec159612-d3ab-4c51-880f-9802f15d9621"
      },
      "source": [
        "img_path = '/content/animals_and_person/testing/person/012414.jpg'\n",
        "img = image.load_img(img_path, target_size=(image_size, image_size))\n",
        "x = image.img_to_array(img)\n",
        "x = x.reshape((1,) + x.shape)\n",
        "i = 0\n",
        "for batch in train_datagen.flow(x, batch_size=1):\n",
        "    plt.figure(i)\n",
        "    imgplot = plt.imshow(image.array_to_img(batch[0]))\n",
        "    i += 1\n",
        "    if i % 2 == 0:\n",
        "        break\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7T_Sg3IotdcQ"
      },
      "source": [
        "Setting train generator"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FcJO5RMotZbR",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "bdce1aa6-8857-48a2-9e35-e1ce5a6b8cc2"
      },
      "source": [
        "train_generator = train_datagen.flow_from_directory(\n",
        "    train_dir,\n",
        "    target_size=(image_size, image_size),\n",
        "    batch_size=batch_size,\n",
        "    class_mode='categorical')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 90486 images belonging to 3 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zJnTI2uGtl03"
      },
      "source": [
        "Setting test generator without data augmentation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-S81DZFhAZ_a",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "46eeb86d-554c-4d59-b0e4-d3c6b30e6726"
      },
      "source": [
        "test_datagen = ImageDataGenerator(rescale=1. / 255)\n",
        "\n",
        "test_generator = test_datagen.flow_from_directory(\n",
        "    test_dir,\n",
        "    target_size=(image_size, image_size),\n",
        "    batch_size=batch_size,\n",
        "    class_mode='categorical')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 22516 images belonging to 3 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Efx6MU_74P9y"
      },
      "source": [
        "Setting tesing generator without data augmentation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "i6AJvcb14QXj",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "605b702f-e756-44d9-f7f8-39fdda5b8cf4"
      },
      "source": [
        "testing_datagen = ImageDataGenerator(rescale=1. / 255)\n",
        "\n",
        "testing_generator = testing_datagen.flow_from_directory(\n",
        "    testing_dir,\n",
        "    target_size=(image_size, image_size),\n",
        "    batch_size=batch_size,\n",
        "    shuffle=False,\n",
        "    class_mode='categorical')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 300 images belonging to 3 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2OylaWP8tucp"
      },
      "source": [
        "Print count classes"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hrdpTxphtyfF",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d1f56e87-4d0c-418c-bb4c-758603ffcac7"
      },
      "source": [
        "train_generator.class_indices"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'animals': 0, 'other': 1, 'person': 2}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gtmFqHM4rM_4"
      },
      "source": [
        "Graph count images per classes \"train\"\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OqhWnpW4rWBu",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 592
        },
        "outputId": "7623b545-1530-40d7-c842-95f7f0a276c3"
      },
      "source": [
        "from sklearn.utils.class_weight import compute_class_weight\n",
        "cls_train = train_generator.classes\n",
        "from collections import OrderedDict\n",
        "classes = list(train_generator.class_indices.keys())\n",
        "num_values = []\n",
        "unique, counts = np.unique(cls_train, return_counts=True)\n",
        "valdict=OrderedDict(zip(unique, counts))\n",
        "for i in range(3):\n",
        "    num_values.append(valdict[i])\n",
        "plt.figure(figsize=(50,10))\n",
        "x = np.arange(len(num_values))\n",
        "xlabel = classes\n",
        "plt.bar(x, num_values)\n",
        "plt.xticks(x, xlabel)\n",
        "plt.show()   "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 3600x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ztB8QXO41GAa",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 280
        },
        "outputId": "21084472-6234-40d8-f488-e83762d169aa"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "goal_types = list(train_generator.class_indices.keys())\n",
        "unique, counts = np.unique(cls_train, return_counts=True)\n",
        "goals = [(counts)]\n",
        "colors = ['y','r', 'r']\n",
        "plt.pie(goals, labels = goal_types, colors=colors ,shadow = True, explode = (0.05, 0.05, 0.05), autopct = '%1.1f%%')\n",
        "plt.axis('equal')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:6: MatplotlibDeprecationWarning: Non-1D inputs to pie() are currently squeeze()d, but this behavior is deprecated since 3.1 and will be removed in 3.3; pass a 1D array instead.\n",
            "  \n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jDOGpfd4ESyU"
      },
      "source": [
        "Graph count images per classes \"test\""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "q6WxcWMdueeL",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 592
        },
        "outputId": "c718f7b7-75d5-4cab-ec0e-85dcca01cbb1"
      },
      "source": [
        "from sklearn.utils.class_weight import compute_class_weight\n",
        "cls_test = test_generator.classes\n",
        "from collections import OrderedDict\n",
        "\n",
        "\n",
        "classes = list(test_generator.class_indices.keys())\n",
        "num_values = []\n",
        "unique, counts = np.unique(cls_test, return_counts=True)\n",
        "valdict=OrderedDict(zip(unique, counts))\n",
        "for i in range(3):\n",
        "    num_values.append(valdict[i])\n",
        "plt.figure(figsize=(50,10))\n",
        "x = np.arange(len(num_values))\n",
        "xlabel = classes\n",
        "plt.bar(x, num_values)\n",
        "plt.xticks(x, xlabel)\n",
        "plt.show() \n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 3600x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cj4dHf491J59",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 280
        },
        "outputId": "8f8b2c91-f005-4ba9-e7aa-c299adc82210"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "goal_types = list(test_generator.class_indices.keys())\n",
        "unique, counts = np.unique(cls_test, return_counts=True)\n",
        "goals = [(counts)]\n",
        "colors = ['y','r', 'r']\n",
        "plt.pie(goals, labels = goal_types, colors=colors ,shadow = True, explode = (0.05, 0.05, 0.05), autopct = '%1.1f%%')\n",
        "plt.axis('equal')\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:7: MatplotlibDeprecationWarning: Non-1D inputs to pie() are currently squeeze()d, but this behavior is deprecated since 3.1 and will be removed in 3.3; pass a 1D array instead.\n",
            "  import sys\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mGrXRECZtzXM"
      },
      "source": [
        "# Download MobileNetV2 model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aroHtGSGuFxj"
      },
      "source": [
        "pretrained_model = tf.keras.applications.MobileNetV2(weights='imagenet', include_top=False)\n",
        "pretrained_model.summary()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Wn4LPhLYuNMN"
      },
      "source": [
        "\"Frozen\" all weights pretrained models "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "denack_UuJkC"
      },
      "source": [
        "pretrained_model.trainable = False"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hADU7F9xuTDB"
      },
      "source": [
        "# Created main layers for end MobileNetV2 model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dEIzXn4Qn9o2"
      },
      "source": [
        "x = pretrained_model.output\n",
        "x = GlobalAveragePooling2D()(x)\n",
        "x = Dense(512)(x)\n",
        "x = BatchNormalization()(x)\n",
        "x = Activation('relu')(x)\n",
        "x = Dropout(0.5)(x)\n",
        "predictions = Dense(3, activation='softmax')(x)\n",
        "model = Model(inputs=pretrained_model.input, outputs=predictions)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xJ245knNmOiy"
      },
      "source": [
        "View all model before training"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1UuMlE8wmTR-"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MqllByDruXdG"
      },
      "source": [
        "# Train a composite neural network based on MobileNetV2 model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eBheP5vMuaul",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "818b8d11-6420-4187-dffe-e4d341cf0c69"
      },
      "source": [
        "model.compile(loss='categorical_crossentropy',\n",
        "              optimizer='adam', \n",
        "              metrics=[\"accuracy\"])\n",
        "\n",
        "filepath=\"inceptionv3_best.h5\"\n",
        "checkpoint = ModelCheckpoint(filepath, monitor='val_loss', verbose=1, \n",
        "                             save_best_only=True, mode='min')\n",
        "learning_rate_reduction = ReduceLROnPlateau(monitor='val_loss', \n",
        "                                            patience=10, \n",
        "                                            verbose=1, \n",
        "                                            factor=0.5, \n",
        "                                            min_lr=0.00001)\n",
        "callbacks_list = [checkpoint, learning_rate_reduction]\n",
        "\n",
        "steps_per_epoch = 32\n",
        "history = model.fit_generator(\n",
        "    train_generator,\n",
        "    steps_per_epoch=steps_per_epoch,\n",
        "    validation_data=test_generator,\n",
        "    validation_steps=5,\n",
        "    callbacks=callbacks_list,\n",
        "    epochs=10,\n",
        "    verbose=1)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From <ipython-input-18-e24fb36bad32>:23: Model.fit_generator (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Please use Model.fit, which supports generators.\n",
            "Epoch 1/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.1780 - accuracy: 0.9390\n",
            "Epoch 00001: val_loss improved from inf to 0.07611, saving model to inceptionv3_best.h5\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.1780 - accuracy: 0.9390 - val_loss: 0.0761 - val_accuracy: 0.9781\n",
            "Epoch 2/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0306 - accuracy: 0.9907\n",
            "Epoch 00002: val_loss improved from 0.07611 to 0.00908, saving model to inceptionv3_best.h5\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.0306 - accuracy: 0.9907 - val_loss: 0.0091 - val_accuracy: 0.9969\n",
            "Epoch 3/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0361 - accuracy: 0.9893\n",
            "Epoch 00003: val_loss did not improve from 0.00908\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.0361 - accuracy: 0.9893 - val_loss: 0.0161 - val_accuracy: 0.9969\n",
            "Epoch 4/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0288 - accuracy: 0.9907\n",
            "Epoch 00004: val_loss did not improve from 0.00908\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.0288 - accuracy: 0.9907 - val_loss: 0.0132 - val_accuracy: 0.9937\n",
            "Epoch 5/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0201 - accuracy: 0.9927\n",
            "Epoch 00005: val_loss improved from 0.00908 to 0.00515, saving model to inceptionv3_best.h5\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.0201 - accuracy: 0.9927 - val_loss: 0.0051 - val_accuracy: 1.0000\n",
            "Epoch 6/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0196 - accuracy: 0.9941\n",
            "Epoch 00006: val_loss improved from 0.00515 to 0.00347, saving model to inceptionv3_best.h5\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.0196 - accuracy: 0.9941 - val_loss: 0.0035 - val_accuracy: 1.0000\n",
            "Epoch 7/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0178 - accuracy: 0.9937\n",
            "Epoch 00007: val_loss did not improve from 0.00347\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.0178 - accuracy: 0.9937 - val_loss: 0.0086 - val_accuracy: 0.9969\n",
            "Epoch 8/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0188 - accuracy: 0.9946\n",
            "Epoch 00008: val_loss did not improve from 0.00347\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.0188 - accuracy: 0.9946 - val_loss: 0.0421 - val_accuracy: 0.9844\n",
            "Epoch 9/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0207 - accuracy: 0.9937\n",
            "Epoch 00009: val_loss did not improve from 0.00347\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.0207 - accuracy: 0.9937 - val_loss: 0.0216 - val_accuracy: 0.9937\n",
            "Epoch 10/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0195 - accuracy: 0.9941\n",
            "Epoch 00010: val_loss did not improve from 0.00347\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.0195 - accuracy: 0.9941 - val_loss: 0.0048 - val_accuracy: 1.0000\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7sxYe7KB4bO5"
      },
      "source": [
        "Test Score"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Tfyai6_L4dMV",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "220187c9-f04a-496d-93db-d80518ad8ac3"
      },
      "source": [
        "test_score = model.evaluate_generator(testing_generator, batch_size)\n",
        "\n",
        "print(\"[INFO] accuracy: {:.2f}%\".format(test_score[1] * 100)) \n",
        "print(\"[INFO] Loss: \",test_score[0])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From <ipython-input-19-988243f5b692>:1: Model.evaluate_generator (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Please use Model.evaluate, which supports generators.\n",
            "WARNING:tensorflow:Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches (in this case, 64 batches). You may need to use the repeat() function when building your dataset.\n",
            "[INFO] accuracy: 99.67%\n",
            "[INFO] Loss:  0.00737465126439929\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SrbkH-WYWXon"
      },
      "source": [
        "Visualize the history of network learning for testing data in confusion matrix format"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "B7XsVmQ3Wcdq",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "8756d8d7-3c32-4a00-fd05-d2049a30354b"
      },
      "source": [
        "#Plot the confusion matrix. Set Normalize = True/False\n",
        "\n",
        "def plot_confusion_matrix(cm, classes, normalize=True, title='Confusion matrix', cmap=plt.cm.Blues):\n",
        "    \"\"\"\n",
        "    This function prints and plots the confusion matrix.\n",
        "    Normalization can be applied by setting `normalize=True`.\n",
        "    \"\"\"\n",
        "    plt.figure(figsize=(20,20))\n",
        "\n",
        "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
        "    plt.title(title)\n",
        "    plt.colorbar()\n",
        "\n",
        "    tick_marks = np.arange(len(classes))\n",
        "    plt.xticks(tick_marks, classes, rotation=45)\n",
        "    plt.yticks(tick_marks, classes)\n",
        "\n",
        "    if normalize:\n",
        "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
        "        cm = np.around(cm, decimals=2)\n",
        "        cm[np.isnan(cm)] = 0.0\n",
        "        print(\"Normalized confusion matrix\")\n",
        "    else:\n",
        "        print('Confusion matrix, without normalization')\n",
        "    thresh = cm.max() / 2.\n",
        "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
        "        plt.text(j, i, cm[i, j],\n",
        "                 horizontalalignment=\"center\",\n",
        "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
        "    plt.tight_layout()\n",
        "    plt.ylabel('True label')\n",
        "    plt.xlabel('Predicted label')\n",
        "\n",
        "#Print the Target names\n",
        "from sklearn.metrics import classification_report, confusion_matrix\n",
        "import itertools \n",
        "#shuffle=False\n",
        "\n",
        "target_names = []\n",
        "for key in train_generator.class_indices:\n",
        "    target_names.append(key)\n",
        "\n",
        "# print(target_names)\n",
        "\n",
        "#Confution Matrix \n",
        "\n",
        "Y_pred = model.predict_generator(testing_generator)\n",
        "y_pred = np.argmax(Y_pred, axis=1)\n",
        "print('Confusion Matrix')\n",
        "cm = confusion_matrix(testing_generator.classes, y_pred)\n",
        "plot_confusion_matrix(cm, target_names, title='Confusion Matrix')\n",
        "\n",
        "#Print Classification Report\n",
        "print('Classification Report')\n",
        "print(classification_report(testing_generator.classes, y_pred, target_names=target_names))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From <ipython-input-20-b4d754394145>:47: Model.predict_generator (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Please use Model.predict, which supports generators.\n",
            "Confusion Matrix\n",
            "Normalized confusion matrix\n",
            "Classification Report\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "     animals       0.99      1.00      1.00       100\n",
            "       other       1.00      1.00      1.00       100\n",
            "      person       1.00      0.99      0.99       100\n",
            "\n",
            "    accuracy                           1.00       300\n",
            "   macro avg       1.00      1.00      1.00       300\n",
            "weighted avg       1.00      1.00      1.00       300\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x1440 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "t7dbNTT04fBZ"
      },
      "source": [
        "Visualize the history of network learning: accuracy, loss in graphs.\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "g6Js7Xh14sI7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 822
        },
        "outputId": "d047b4ac-19f2-41bd-bb63-72d2ee6b652d"
      },
      "source": [
        "#Plot the Graph\n",
        "\n",
        "# Loss Curves\n",
        "plt.figure(figsize=[8,6])\n",
        "plt.plot(history.history['loss'],'r',linewidth=3.0)\n",
        "plt.plot(history.history['val_loss'],'b',linewidth=3.0)\n",
        "plt.legend(['Training loss', 'Validation Loss'],fontsize=18)\n",
        "plt.xlabel('Epochs ',fontsize=16)\n",
        "plt.ylabel('Loss',fontsize=16)\n",
        "plt.title('Loss Curves',fontsize=16)\n",
        "  \n",
        "# Accuracy Curves\n",
        "plt.figure(figsize=[8,6])\n",
        "plt.plot(history.history['accuracy'],'r',linewidth=3.0)\n",
        "plt.plot(history.history['val_accuracy'],'b',linewidth=3.0)\n",
        "plt.legend(['Training Accuracy', 'Validation Accuracy'],fontsize=18)\n",
        "plt.xlabel('Epochs ',fontsize=16)\n",
        "plt.ylabel('Accuracy',fontsize=16)\n",
        "plt.title('Accuracy Curves',fontsize=16)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Accuracy Curves')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAf0AAAGKCAYAAAAG65jxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3xUVfrH8c9DKKGE3jtKx74Uy7oiImDvYEFFV1nddcWuYOOHKLa1rGUVF1RcUbGjixXEXdcGdimRKlIFKYoYIMn5/XFmmMmkMElm5k6S7/v1mldum3ufSXvuKfccc84hIiIilV+1oAMQERGR1FDSFxERqSKU9EVERKoIJX0REZEqQklfRESkilDSFxERqSKU9EWSyMweMzNnZvcGHUtFYWZ1zWy0mX1uZr+YWY6ZZZvZg2bWOej4RCoy03P6IslhZrWBtUB94EegjXMuN9io0puZtQLeBVoDDwIfADuAnsD5QDXn3P7BRShSsVUPOgCRSuxEfMKfARwNDAFeDzSiGGaWgb/5T5ebkaeAVkBf59yiqO3vmdnDwAmJuIiZ1XLObU/EuUQqElXviyTPucAmYATwW2i9EDM7ycz+Z2ZbzexnM/vUzI6P2l/dzK41s/mhqu71ZvammXUP7R8RakLoGHPesWbmYrY5M7vVzK4zs2X4UvTeZpZpZvea2behONaa2Wvha8Sco5OZPRU6ZruZLTWz+0P7rgxtaxbzHgsd92xx3ywz6wMcAdwWk/ABcN4rMZ9lbMw5Ooa2j4ja9oSZrTSzg8zsQzP7DbjTzP5tZp8XEUcrM8s1s8tjPvPToe/9djP70sxOinlfVzN72cx+DP2cVpjZ82amwpWkDf0yiiSBmbUGBgKPOefWm9krwMlm1sg5tynquL8Cfwdewd8UbAUOADpGne5ZfK3Bffiq70zgD/gS8cIyhDcCWApcBfwKrAZqAVnAeGAN0Bj4M/CRmfVwzq0NxdsJ+BTYBtwELALaA4NC5348dI7zgDujrjkI6ISvoi/OkaGv08vwmXanAf77eDcwBn8T1gl4xsx6OufmRx17ZujrVAAzawd8gm+iuRxYDwwDXjSzE51z4Xj/jb/JuxjYALTB1/CocCXpwzmnl156JfgFXAM44KDQ+uDQ+kVRx9QHfgFeKuE8A0Lvu7SEY0aEjukYs32s/xMvsM3hk3zt3cSfAdQJxXd51PYp+BuT1iW89wlgMaE+Q6FtLwELdnPNf4TiqxXn99gBY2O2dQxtHxETjwNOiDm2NrAFmBCz/UtgRtT6JHyibxJz3DvAl6HlpqFrHB/0755eepX00h2oSHKcCyxyzn0UWn8Xn2yjq/gPBuoBE0s4zyB8MnksgbG96Zz7LXajmQ01s0/MbDOQi68FqAd0i4nndefc6hLO/zCwJ76qPtw57zhK/pzJtpOY/hSh78ELwFlmZgBmtjewL75vQdgQfL+MLaGmluqhKvu3gH3NrD7wE7725HYzu9DMuiT9E4mUgZK+SIKZWW98b/OXzKyhmTXEV52/BBxoZl1DhzYJfV1ZwumaABuLStLlsKaImI8DngMW4Ku3+wF98CXczJh4SooX59ynwGfARaFNF+BvIp7cTVw/hL522M1xZbHeOZdXxPangHZA/9D62fjajVeijmkOnIO/cYh+3RXa38Q55/DNE3OBCcB3oT4MFyf4c4iUi5K+SOKFS/PX4tt4w69LQtvPCX3dEPrapoRzbQAahx7/K05O6GvNmO1NYg8MKeo53dOBxc65Ec65GaHE/RW+bT82npLiDXsYOMHM2uCT/vPOuY27ec+7oa/HxXF+gO2U7zMDvA+sAIabWTX8Dc8LMTdZP+FrBPoU81oN4Jxb6pw7B2gG7A/MAh42s6Pi/DwiSaekL5JAZlYTOAPf8evwIl5fAmeHqpM/xLePjyzhlG8Dhk+cxfk+9HWvqDiqE+lcF486+NJ4tLPxbfux8RwbqrIvyTP4EvNUfEe/R3YXQOhGYxYwprhBeMws+pG974n6zCHH7O46Mdd0wL+AU/Gd7tpQsGof4E1gH2Cec25uEa/tsed0zn0JXBHaFBujSGDUe18ksY7BlzavdM7Njt1pZo/iO6z1d869Z2ajgQfM7EXgaXyi3A/Icc49EDrmReCeUC/yWUANfO/9f4euMQdYAtwVKq1ux/e8r1WKuN8ETjQ/cuDrQG/gr8DmmONuxifHD83sNnyHvTbAEOfc8PBBzrnfzOwJfG/3b5xzH8YZx3B8iX+OmT1AZHCe7vie/zWAV0PHPgvcYGbXAx8Dh+JvuErrKXyP/kfwpf7ZMftvwj+x8B8zexBYDjTCJ/M9nHPnm9k+wP34JpLF+JulEfgbqVlliEkkOYLuSaiXXpXphW8L/hmoU8z+BvjH3Z6I2nYqvmbgt9B7PwGOjdpfHbge+A6fANfjO5Z1izqmFz5ZbcUnrisovvf++CLiqoZ/1G51KL738VXUy6NjDR27J74kvwHftLAEuKeIcx4Uut5fSvk9rIdPwl/gOxNuB7LxSXWPqOMyQ9vW4G+WngP6UnTv/ZW7ueac0PtuK2Z/W+CfwKrQz2ANvvf+8ND+5vg+C9+Fvn8bQ9/DwUH/TuqlV/RLw/CKSFKY2a3AKPzjfT8HHY+IqHpfRBLMzPbHP+Y3CpiohC+SPlTSF5GEMrPlQAv8c+xnO+d+CTYiEQlT0hcREaki9MieiIhIFaGkLyIiUkVU6o58TZs2dR07dgw6DBERkZT57LPPNjjnmhW1r1In/Y4dOzJ37tygwxAREUkZM/u+uH2q3hcREakilPRFRESqCCV9ERGRKkJJX0REpIpQ0hcREakilPRFRESqCCV9ERGRKkJJX0REpIpQ0hcREakiUpr0zWyymf1oZt8Ws9/M7O9mttjMvjazA6L2nWtmi0Kvc1MXtYiISOWQ6mF4nwAeBKYUs/8ooEvo1Q/4B9DPzBoDNwO9AQd8ZmbTnXObkh6xSCXlHOTkBB1F6dWsCRkZQUchkhjTp8MBB0Dbtqm5XkqTvnPuP2bWsYRDTgCmOOcc8LGZNTSzVkB/4B3n3EYAM3sHGAI8k9yIRSqnd96BSy+FhQuDjqT0MjNhyBAYNgyOPRbq1Qs6IpHS27QJRo2Cp57yv88zZoBZ8q+bbm36bYAfotZXhrYVt70QMxtpZnPNbO769euTFqhIRfTLL3DRRTBoUMVM+OBrJ155Bc44A5o1g1NPheefh19/DToykfj8+9+w114+4QO8+SY8+WRqrl3pZtlzzk0EJgL07t3bBRyOSNp47z04/3xYvjyyLSMDatQILKRScw62b4+s5+TAiy/6V506vuQ/dCgcdZRfF0knW7bA5ZfD448X3D58OJxwQmpiSLekvwpoF7XeNrRtFb6KP3r77JRFJVKB/forjB4NDzxQcPtJJ8E//gEtWgQTV1llZ/uS/bRp8M03ke3btvlt06ZB3bpw3HG+CWDIEN8kIBKkt9+GP/4RVq6MbGveHB59FE48MXVxpFv1/nTgnFAv/gOBLc65NcBbwCAza2RmjYBBoW0iUoIPPoB99y2Y8Bs1gqef9qXjipbwAbp1gxtugK+/hnnzYOxY6NGj4DG//grPPutvbJo39yWp6dML1hKIpEK4SW3w4IIJf9gw//ubyoQPYL7PXIouZvYMvsTeFFiH75FfA8A594iZGb53/xBgG3Cec25u6L3nA2NCp7rVORdTQVJY79693dy5cxP9MUTS3m+/wfXXw333+SrxsGOPhYkToVWr4GJLlnnz4Lnn/Ou774o+pn59/0926FA48kj/JIBIshTVpNakia9hO+205F3XzD5zzvUucl8qk36qKelLVfTxx3DuuQUTX4MGcP/9cM45qekhHCTnfLX/tGn+BmDx4qKPa9jQ3wAMGwZHHFGx+jZIevv1V7juOnjwwYLbU9WkpqQvUgXk5Piq7rvugvz8yPbBg+Gf/0zdc8DpxDn46iuf/KdNg6VLiz6ucWP/D3noUDj8cN0ASNn9978wYkTB37VGjfwNwBlnpOamW0lfpJKbO9eX7ufPj2zLyoJ77vGdhyp76T4ezsHnn0c6+0VXuUZr0gROPtnXABx2GFRPt+7OkpbSqUlNSV+kktqxA265BSZMgLy8yPYBA2DyZOjQIbjY0plzMGdO5Abghx+KPq5ZMzjlFF8D8Ic/aCRAKVq6Nakp6YtUQl9+6f/RfP11ZFudOr56/6KLoFq6PZuTpvLz4dNPIzcAq1YVfVyLFn4goKFD4ZBDdAMgvknt5pvh7rvTq0lNSV+kEtm505fsb7kFcnMj2//wBz/oxx57BBdbRZefDx995JP/88/DmjVFH9eqlb8BGDYMDjpIN1hV0Zw5vu0+HZvUlPRFKolvv/Wl+88/j2yrXdvfBPz1r0o+iZSfD//7X+QGYN26oo9r08Y/fjV0KPTrp59BZbd9u7/hvv329G1SU9IXqeByc321/dixvh0/7KCD4IknoGvXoCKrGvLyfK/sadPghReguGk92rXzyX/oUOjTRx0oK5svvvA33dEjQdat6/82//Sn9LnhU9IXqcAWLPDViJ9+GtlWqxaMH+/H8Vbbcmrl5sL77/sbgBdfhJ9+Kvq4Dh188h82zE+dqhuAimvnTrjtNv83F92kdthhvnSfbk1qSvpV1MKF/u6zuHbJdNahg+81ffLJfhjVqigvD+691w85Gz18bJ8+fkau2KFnJfVyc/2oa9OmwUsvwcaNRR/XpYtPDr//fWrjk/L75htfuv/ii8i22rV99f4ll6RP6T6akn4V5JzvYfzRR0FHUj7VqvnBUoYN84OnNG0adESpsWiRL91/+GFkW40a8H//B1dfrWfH09HOnTBzpr8BePll2Ly54P5GjfwTF+3bBxOflE64Se3mm/3PNuzgg32H2XRuUlPSr4LefdePLV6ZZGT44VKHDvU3AI0bBx1R4uXn+8lxRo/2g32E7b+/L93vvXdwsUn8duzwf4PhGoBffvHbDz4YZs/WiH/pbsECX7qfMyeyrSI1qSnpVzHO+ce3PvjAr48Y4ZNIRbFzJ8ya5f9hhj9DrOrV/U3N0KF+HupGjVIbYzIsXQrnnQf/+U9kW/XqcOON/uenRFExffih/3sM9/S+9lpfNSzpJy/PP3J3440Vu0mtpKSPc67Svn73u9+5qmjmTOd86neuRg3nvv8+6IjK7ocfnLvvPucOPjjymWJfNWo4d8wxzk2Z4tzmzUFHXHp5ec49/LBzdesW/Fx77+3c558HHZ0kwu23F/zZvvFG0BFJrOxs5w46qODPqWZN5yZMcG7nzqCjKx1grismL6qkXwkddliktDhyJDz6aLDxJMqKFf5xqWnT4JNPij6mZk0YMsTXABx3nJ9KNZ19/70fyGPmzMi2jAxfsr/xRk39Wlnk58PRR8Nbb/n1pk39RECtWwcblxTfpHbAAb50v9dewcVWVqrer0Lefx/69/fL1av7DmEdOwYZUXIsX+4HTJk2zU82U5Ratfw/2qFD/aQX9eqlNMQSOQeTJsEVV0TaewF69vTP3ffpE1hokiQ//gj77Rd5muaww/zNXrq3D1dmS5b4+e4rW5NaSUk/DR82kPL4v/+LLI8YUTkTPvjPdfXVvqPNkiW+jfSAAwoes32770V9xhl+4pRTT/U3Cb/+GkjIu6xcCUcdBRdeGEn41ar5tt7PPlPCr6yaN4epUyOPeL3/vh/ZTVIvPx8efhj22adgwt9nH/8/5aabKm7C3x2V9CuR//7XdxgCX3r47rv0GzQi2RYvjkyc8tVXRR9Tp44v+Q8d6pNvnTqpic05mDIFRo2CLVsi27t29aX7gw5KTRwSrHHj/GNg4AfsefddP4SrpMby5b5JbdasyLbK1qSm6v0qYuDASNvweef5wUCqsuzsSBNA9LCZ0erW9W3/w4b5vgCZmcmJZc0aP1DSa69FtpnBZZfBrbf6wT6kasjL80+evPeeX2/Z0t+gVtVBqFLFOT/z3RVXwNatke09e/q2+95F93WvkJT0q4D//S8y2ldGhk94e+4ZbEzpZMECn/yfe84vFyUrC44/3tcADB7s+wSUl3PwzDN+5K5NmyLb99zTD/Bx6KHlv4ZUPGvWwL77RsbwHzQI3ngjPUd3qwxWroQLLoh0pAT/vb76aj+fRbJu9oOipF8FDB4Mb7/tl88911cXS9HmzYvcAGRnF31M/fr++f9hw3yprCxVfj/+CBdf7AdniXbJJb4PQt26pT+nVB5vv+3/bsMmTIDrrkvRxXfsgNWrYdWq4l9btvjqhzZtin81b57WPRGd86X4yy4r2KTWrZv/H3nggYGFllRK+pXcxx9H2oOrVfNj7nfpEmxMFYFzfqra557zr8WLiz6uYUM48URfAzBwYHwdfJ5/Hv78Z9iwIbKtY0ff5HL44QkJXyqBMWN8sgefO99/3w+fXWbO+QkAikri0Um+uGkCSysjw7dPlHRj0KZNII/OrF7tm9Refz2yzcyPqDd+fOVuUlPSr+SOPtpXDQIMHw5PPRVsPBWRc75dNVwDsHRp0cc1auQnARo61Cfv2BuADRt8Sf655wpu/9Of/DjeWVnJib/S+/VX3zM1O9tPa1evnn9lZflX7HIi2mZSIDfXP2L7v//59Xbt/MQuTZoUcfD27T6TlVRCX70acnJS+RHiU79+4RuB1q0LrrdokZBaA+f8UxJ//WuATWrOwbZt8PPP/hGd4r6Gl7du9c/wJmgqRiX9SuzTT6FfP79s5turu3ULNqaKzjn/j/e55/xNwPLlRR/XpEnkBqB/f1+i+NOffLV+WNu2/m950KBURF7BOeeT1sKF/pWdHfm6YkXpzlWjRuGbgaJuDuJdzsxM2ty4P6xw7LefY+Mm36B/3L7f8+qJT2CrYxJ6dLVReVWrVnwJPZyMGzWCdetKbgIobl7hsoin1qB16xLvnNet801qL79ccHtcTWrO+ZvLkpJ0bLIu7uvWrf65wNLYti1h1Q9K+pXYscfCv//tl884w9/hSuI45wf/CT8GWFzuadCgYJsh+EE/7rnH75Mov/3mR40KJ/VwYs/OLtitOp1kZJT9hiEry98wxJbQw+urV/NazkCOJ/Joxz1czuXcV7ZYs7KKL01Hl6oTMVVjTs7uax9WrfJ9CBIl+vNFvaav+h3nP/w7ftoSqX7r2ORnJp/0Ooc3n7f7RF6WRJ1Ia9f6n0sCKOlXUnPnRgZyMfMd1CrKhBAVkXN++N9p03yb/cqVRR/XqhU89hgcc0xq40srzvliV1Gl9uXL/f7SyMjw9bPdu/tv8LZtkRLX1q2R5fArPLtNBXIFf+NergCgBjv4gN/Tl6hp3koqnUe/0q0NyTlfI7C7fgblqMmYxmmcwTPkE2keuIh/cCfXkEVAN5K1a/tmjays+L6edlrC+j4o6VdSJ5wA06f75WHD4Nlng42nKsnP9x0owzcAq1f77cOHw9//Xjlm/YvL9u2+B2RsqX3hQl+CKq2GDX1iD7+6dfNf99gj/kconPNxxd4QlHY5ej2RJdWiZGWxo3VHDln9PHN/8e1znZps4fP7P6Bh1+aRnvKJKJ2nq3CfhZJqDFavLjj9HfACp3A6z5KH/9605Qcmcz5H8m7pYyhtoo7+Grsc4M9KSb8S+uKLgsPOfvst9OoVXDxVWX6+H7qzRo3CQwFXCs75UlhRpfalS0tfJVqtGnTqVDCph5ebNUta23m57NhRvhuIvDxfQ1FcdXuodL50Key/f+R+KTx0dDp+SwIRrjUI3Ry89FoNhj4ygDzn+0P0yFzKe/tfSYumecUn5JKSdiW5qVLSr4ROOgleecUvn3qqL22KlMvOnX4ig+ikHk700d2g41W/fuGk3r07dO5cYXrXB+H5533n0LCHH/ad06SgV17xNeK5uX69e3c/ymHLlsHGlQ6U9CuZr77ys3WFff017L13cPFIBZOf7zvSzZnjxycOJ/glSyL/QeNlBh06FEzq4a8tW6qIWkZ//jP84x9+uVYt35QU/Tdf1U2fDqecEvl17doVZs/2lSlSctKvHHUZVUz0zFwnn6yEL7uxZo1/tjP8mjOn8KMGu1O3buGk3r27HwWqMo9yEpB77oEPP/Q3+Nu3+z47c+emXx+9ILz+uq/dDCf8Ll18CV8JPz5K+hXMN9/Aiy9G1m+8MbhYJA39/LPPDtFJftWq+N/frl3RVfJt2qjUnkKZmb4t/4ADIuMSXXyxH3irKv8YZszwJfydO/16584+4bduHWxcFYmSfgUTXco/4QRV+VVp27f7tp05cyIJfuHC+B6Ha9oU+vaF3/3OP+fZvbuvI9WEAGmja1d49FH/RAjA00/DEUf4GTSrojff9H2Zwg9S7LGHT/ht2gQbV0WjNv0KZN48X5Uf/pF99lkl7S0uhYXb4aNL8F9+Gd+jZHXq+OTet2/k1aFD1S4yViB//GNkmuzatX1FTs+ewcaUam+95Qs54af1OnXybfjt2wcaVtpSm34lMX58JOEfd5wSfqW2enXBBD93bnzt8BkZ/s6wb18/clPfvj5DVJJHkaqiv//dd+SbP98PZjh0qP+VqFMn6MhS4513/IRX4YTfsaMv4Svhl43+E1QQCxYUnMTlppuCi0USbMsWX21Tlnb4PfYoWILff/+qkw2qiLp1fft+nz4+6c+bB5deCv/8Z9CRJd/MmXD88ZE5hNq39wm/Q4dg46rIlPQriOhS/tFHQ+8iK24k7YXb4aMT/MKF8b23WbOCCb53b982L5Ver17wwANwwQV+fdIkP8vjWWcFG1cyvfeer9EMJ/x27XyVfseOQUZV8alNvwLIzvY1tOGBzz75xP/PlzSXn++7XUc/Kqd2eCkj53ynvvCkWvXq+Qqirl2DjSsZ3n8fjjrK12yAn61y9mw//YLsntr0K7hbb40k/CFDlPDTVmw7/Jw58Y0/H90OH3716KF2eCnADB55xP9qLV7sR/kdNgw++sg/4ldZ/Oc/vjYznPDbtPGlfiX8xNB/lTS3aJF/VCesVG35X38NZ58Nmzf7R7LCj2aFl5s3V8mxtPLz/fR6Cxf6jhbhrwsWwI8/xneOPfeMJPc+fdQOL3HLyvLt+wce6CuMvvwSrroKHnww6MgS44MPfMLfts2vt2rlE37nzsHGVZko6ae56FL+kUfCQQeV4s3jxvnED34i+LffLri/YcPIjUD0DUGnTiplhmePi07s4eFqf/01/vPEtsP36QNNmiQvbqn09t/fj9h3ySV+/aGHfPv+KacEG1d5ffihr9IP/3m1bOkTfpcuwcZV2aS8Td/MhgD3AxnAP51zt8fs7wBMBpoBG4HhzrmVoX13AOFZym9xzj1HCSp6m/6SJX5AtPDU4B98AIccUooT7LEHLFtW+gvXrOn/0qJvBnr08MFUtsFbNm2KTCoTneDLMntcnTq+c110km/fXrUpknDO+aFoX3rJrzdo4Gfe7NQp2LjK6uOPYdAgPyEhQIsWvg2/e/dAw6qw0qZN38wygIeAI4GVwBwzm+6cmx912N3AFOfck2Y2AJgAnG1mxwAHAPsBtYDZZvaGc64Mk3ZXDLfdFkn4RxxRyoT/88+RhF+9Ojz7rO9UFk5sCxdG/sJi7djhnwuaN6/wvnbtiq4daNEifZNbuEo++rOHl9etK/35Gjcu/D3o0cN3tMvISHz8IjHMfA/+zz+H5cv9U5+nnw7//a+/Z69IPv0UBg+O/Dtq3tyX8JXwkyPVdbh9gcXOuaUAZvYscAIQnfR7AleElt8DXona/h/nXC6Qa2ZfA0OAaakIPNWWLYMpUyLrpX4u/9tvI8vduxeu+3POdzyLToDh5dWriz/vDz/4V1FNBUX1G0hlU8H27b4TROxnWrgw0kgYr/DscbGfp3t3X2UvErCGDf3YHYcc4ief+fRTGDMG7r476MjiN2eOL+GH+7s2awazZvk/NUmOVCf9NsAPUesrgX4xx3wFnIxvAjgJyDKzJqHtN5vZ34A6wOEUvFkAwMxGAiMB2lfgIZsmTIjMItW/P/zhD6U8QbgtH2CffQrvN/PdYtu08dUI0X7+uWDCDCfQxYuLn3p182ZfR/fxxwW3RzcVRCfObt38M0dlEa6Sj21vL0uVfK1akUllokvtXbqoc52kvb594Y474Mor/frf/ub/Xxx7bKBhxeWzz3zCDw802bSpH4ynV69g46rs0rG31lXAg2Y2AvgPsArIc869bWZ9gA+B9cBHQF7sm51zE4GJ4Nv0UxV0In3/PTz+eGT95pvLcJLdJf2S1K8faZOOtnOn72hQVDV5WZsKYpNtuKnAOV+jUFR7e7y95KM1aVJ0qV1V8lLBXX65rw5//XW/fu65fkretm2Djaskn3/uOyZv3uzXmzTxCV/ThCdfqpP+KqBd1Hrb0LZdnHOr8SV9zKwecIpzbnNo363AraF9U4HvUhBzykWX8g89FA47rAwnKU/SL06NGpGkGS26qSA2QcfTVPDOOwW3N2jgbzDKUiXfsWPRzQwauU4qKTN44gk/4+bKlbBxI5xxhr8RSMeHcL78EgYO9BV24LvIvPtu4v5NSclS2nvfzKrjE/UR+GQ/BzjTOTcv6pimwEbnXL6Z3Yov5d8U6gTY0Dn3k5ntA0wF9gu18RepIvbeX7HCP5Mani/63XcL177vlnO+wS/cULZyZXDzT/78s3/MLbYqvqSmgt3JzPTDkMXWEKhKXqqwDz7wVfvhzr9jxvhHftPJV1/BgAH+xgSgUSNfwt9//2DjqmzSpve+cy7XzC4B3sI/sjfZOTfPzMYBc51z04H+wAQzc/jq/b+E3l4D+K/5HuI/4x/lK2PWSF933BFJ+Icc4v9ASm3FikjCb9wYWrdOWHylVr++fza9T5+C28NNBUV1JAw3FTRtWrg6vkcP/xicquRFCvj97/3QHNdf79cnTPC1hIMGBRtX2Dff+AJMOOE3bOgr+ZTwU0tj76eRlSv9YG3hodnfftu3e5Xaa6/5qanA/9XPnp2oEJPPOVi71jclqEpepFTy8/1Q3eEWs+bNfXV6q1bBxvXtt34AoQ0b/HqDBr4WUxOHJUdJJf1qqQ5GinfHHZGEf+CBvt2rTJLRnp8qZv4/lBK+SKlVqwZPPeX7woLv8zp8eKTKPwjz5/say3DCr1/fF4RPsawAACAASURBVGiU8IOhpJ8mVq+Gxx6LrN98cznGuqnISV9EyqVFCz9fR/j/x6xZfqCvICxY4BP++vV+PSvLJ3xNGhYcJf00ceedfmwZ8H8QgweX42RK+iJV2hFHwA03RNbHjvXT1aZSdrZP+OFBL+vVg7fegn6xI7NISinpp4E1a+DRRyPr5Srl//abH24X/Ek00oVIlXTTTZFBvfLz4cwzIyXuZPvuO9+Gv3atX69bF958s5QThklSKOmngbvugpwcv9y7t59pqszmz4+MSte5c+WbIEdE4lK9OkydGukes3q1H7intINWltaiRT7hr1nj1+vWhTfeKOXcIZI0SvoBW7cOHnkksn7TTeWct0ZV+yIS0qZNwTk83njDD9WbLIsX+4QfHpOrTh2YMcMPMibpQUk/YHff7WvkwT+vWu4xs5X0RSTKUUfBNddE1seMgY8+Svx1li71CX9VaIzV2rXh3/8uw7whklRK+gH68Ud4+OHIerlL+aCkLyKFjB/vHwMGPxDm6adHhsFNhOXLfcJfudKvZ2b6uQD690/cNSQxlPQD9Le/RYaX33dfOOGEcp7QOSV9ESmkRg149lk/Ch74QTvPO8//yyiv77/3yX3FCr+emenHByvTaKKSdEr6AdmwAR56KLKekFL+unWRETDq1fOTz4iI4CeUjJ6989VX4YEHynfOFSt8wv/+e79eq5Y/b5kHFpOkU9IPyD33wK+/+uW994YTT0zASaNL+Xvt5YfnEhEJOfFEuPTSyPpVV0FZRyr/4Qdfpb98uV+vWRNeeSV9xvqXoikrBOCnnwreYd94Y4Lys6r2RWQ37rwTDjjAL+/cCcOGwZYtpTvHypU+4S9d6tdr1oSXX/bj/kt6U9IPwL33wtatfrlXLzjllASdWElfRHajVi147jk/JC74xD1yZPzt+6tW+fb6JUv8eo0a8OKLcPTRyYlXEktJP8U2bYK//z2ynrBSPijpi0hcOneGiRMj69OmFZz7ozhr1viEv2iRXw8n/HI/aiwpo6SfYvfdF5kuvkcPOPXUBJ14504/Gl/Y3nsn6MQiUhmdfrov4YeNGlWw3BBr7VpfpR8e5bt6dXj+eTjuuOTGKYmlpJ9CmzfD/fdH1m+4ATIyEnTy7Gyf+AHat488myMiUoz77ouUD3JyfPt+uOkx2rp1PuFnZ/v16tV97UC5HzOWlFPST6H77490mOnWzf+BJYyq9kWklGrX9sm7Th2/vnAhXHJJwWN+/NFX6S9c6NczMvwz/yedlNpYJTGU9FNkyxZ/Vx2W0FI+KOmLSJl07w7/+Edk/ckn/Qv8rHwDBkRaDjMy4JlnEtj5WFJOST9FHnjAV++D70Rz+ukJvoCSvoiU0Tnn+Bn4wv78Z/jgAzjiCJg3z2+rVg2efhpOOy2YGCUxlPRT4Oef/WA8YTfc4NvEEuqbbyLLSvoiUkoPPuhL/eCHBz/00Mi/lWrV4KmnEtwkKYFQ0k+Bhx6KTG6x555w1lkJvsDGjZGZLmrVgi5dEnwBEans6tXzz+9nZhbcbuar+888M5i4JLGU9JNs69aC81dff32SS/m9eiXhAiJSFeyzT8EnjMzgiSdg+PDAQpIEU3ZIsoce8sPuAnTqlKQ/nuj2fD2fLyLlcOGFsHo1/PvfcM01asOvbJT0k2jrVrj77sj6mDF+BKuEUyc+EUkQMxg71r+k8lH1fhI98khkptsOHXwP2aRQ0hcRkTgo6SfJtm1w112R9TFj/ExUCZeXB99+G1lX0hcRkWIo6SfJI4/4kawA2rWDESOSdKGlS/0dBkCLFtC8eZIuJCIiFZ2SfhJs2+bnrA4bPTpJpXxQ1b6IiMRNST8JHnvMT1AB0KYNnH9+Ei+mpC8iInFS0k+wnBy4447I+nXX+fFykkZJX0RE4qSkn2CPPQZr1vjl1q3hgguSfEENvysiInFS0k+gnBy4/fbI+rXXFh7SMqG2boUlS/xyRgb06JHEi4mISEWnpJ9Akyf7kawAWrb0I1slVfSjet27J7kdQUREKjol/QTZvh0mTIisX3MN1K6d5Itq+F0RESkFJf0EefzxyER3LVrAn/6UgouqE5+IiJSCkn4C7NhRsJR/9dVQp04KLqykLyIipaCknwBPPgkrVvjlZs3gootScFHnlPRFRKRUlPTLaedOuO22yPpVV0Hduim48A8/wJYtfrlhQ2jbNgUXFRGRikxJv5ymTIHly/1y06bw5z+n6MKxpXyzFF1YREQqqpQnfTMbYmbZZrbYzK4rYn8HM5tpZl+b2Wwzaxu1704zm2dmC8zs72bBZrqdO+HWWyPrV14J9eql6OKq2hcRkVJKadI3swzgIeAooCdwhpn1jDnsbmCKc24fYBwwIfTeg4FDgH2AvYA+wGEpCr1ITz8Ny5b55caN4S9/SeHFlfRFRKSUUl3S7wssds4tdc7tAJ4FTog5picwK7T8XtR+B2QCNYFaQA1gXdIjLkZuLowfH1m/4grIykphABp+V0RESinVSb8N8EPU+srQtmhfASeHlk8CssysiXPuI/xNwJrQ6y3n3IIkx1usZ56JjIDbqBH89a8pvHhODmRn+2Uz6NUrhRcXEZGKKh078l0FHGZmX+Cr71cBeWbWGegBtMXfKAwws0Nj32xmI81srpnNXb9+fVICzMsrWMq//HKoXz8plyraggU+CIA990xhRwIREanIUp30VwHtotbbhrbt4pxb7Zw72Tm3P3B9aNtmfKn/Y+fcVufcVuAN4KDYCzjnJjrnejvnejdr1iwpH+LZZ+G77/xygwYpLuWDht8VEZEySXXSnwN0MbNOZlYTOB2YHn2AmTU1s3Bco4HJoeUV+BqA6mZWA18LkPLq/bw8uOWWyPpll/nH5FNKnfhERKQMUpr0nXO5wCXAW/iEPc05N8/MxpnZ8aHD+gPZZvYd0AIIPxT3ArAE+Abf7v+Vc+61VMYP8Pzzkeb0+vVh1KhUR4CSvoiIlEn1VF/QOTcDmBGz7aao5RfwCT72fXlAKqaxKVZ+fsFS/qhRvhNfyinpi4hIGaRjR7609cILMH++X87K8lX7KbduHfz4o1+uUwf22COAIEREpCJS0o9Tfj6MGxdZ/+tf/YA8KRfbia+afoQiIhIfZYw4vfwyzJvnl+vW9YPxBEJV+yIiUkZK+nGILeVfcgk0aRJQMEr6IiJSRkr6cVi/HmrU8Mt16viJdQKj4XdFRKSMUt57vyJq0QLmzIEZM/w09kka82f3cnMjbQyggXlERKRUlPTjZAbHHBNwEN99Bzt2+OV27QJ6XlBERCoqVe9XJBp+V0REykFJvyJRJz4RESkHJf2KRElfRETKQUm/IlHSFxGRclDSryg2bfKPDgDUrAlduwYbj4iIVDhxJX0zm2pmhyY7GClB9PP5PXtGBg4QERGJU7wl/QOB2WY2z8wuNbNUzyAvqtoXEZFyiivpO+f2AI4GsoG7gVVm9riZHZjM4CSKkr6IiJRT3G36zrm3nHMnA+2B24HDgf+Z2RdmdpGZ1UtWkIKG3xURkXIrdUc+59xa59wtwMHAf4F9gYeB1WZ2l5nVTXCMkp+vpC8iIuVW6qRvZgPMbBqwDNgbuBd/A/AAcBEwJaERCixbBr/+6pebN/eTAYiIiJRSXGPvm1kT4DxgJLAn8Dk+wT/jnMsJHfaxmX0DTEpGoFWaht8VEZEEiHfCnVVAPvAccJZzbk4xxy0EfkxEYBJFnfhERCQB4k36Y4DHnXObSjrIOfcl0KncUUlBSvoiIpIAcSV959w9yQ5ESqCkLyIiCRDviHz3mtlTxex7yszuSmxYssvWrbBkiV+uVs2PxiciIlIG8fbePx54u5h9bwEnJiYcKWTePHDOL3frBpmZwcYjIiIVVrxJvw2woph9K0P7JRlUtS8iIgkSb9LfBHQuZl9nYGtiwpFClPRFRCRB4k367wI3mFmBUWFC62OAdxIdmIRoJD4REUmQeB/ZuxGYAywys9eJVOkfC+QANyQnvCrOOZX0RUQkYeJ9ZG+5mfUBxgFHAk2ADcDLwM3Oue+TF2IVtmoVbAoNjdCgAbRrF2w8IiJSocVb0sc5txw4J3mhSCGxw++aBReLiIhUeKWecEdSSFX7IiKSQHGX9M2sOXAG0A2IfVjcOef+mMjABCV9ERFJqHhn2esGfBQ6vi6+Pb8xkIF/nG9LsgKs0pT0RUQkgeKt3r8L33u/BWDAUUBt4AJgG3BSUqKryrZvh4ULI+t77RVcLCIiUinEW73fB7gI2B5ar+acywUmm1kz4D7g8CTEV3UtWAB5eX55jz0gKyvYeEREpMKLt6RfD9jonMvHV+U3jdo3B39TIImkqn0REUmweJP+cqBlaDkbOC1q37HA5gTGJKCR+EREJOHiTfrv4AflAbgHOM/Mss1sHjAKmJyM4Ko0lfRFRCTB4m3THw3UAnDOTTOz34BhQB3gfuCx5IRXhSnpi4hIgu026ZtZBtAdWB3e5px7DXgtiXFVbT/+CGvX+uU6dXxHPhERkXKKp3rfAXOB/RNxQTMbEmoaWGxm1xWxv4OZzTSzr81stpm1DW0/3My+jHrlmNmJiYgp7US35/fqBRkZwcUiIiKVxm6TfqjH/g/4QXnKJVRr8BD+Of+ewBlm1jPmsLuBKc65ffAT/EwIxfGec24/59x+wAD8+ABvlzemtKSqfRERSYJ4O/I9ClxmZjXLeb2+wGLn3FLn3A7gWeCEmGN6ArNCy+8VsR/gVOAN59y2csaTnpT0RUQkCeLtyJcF7AksNbM3gTX4av8w55y7OY7ztMHXGoStBPrFHPMVcDK+g+BJQJaZNXHO/RR1zOn4pwgKMbORwEiA9u3bxxFSGlLSFxGRJIg36Y+JWj6/iP0OiCfpx+Mq4EEzGwH8B1gF5IV3mlkrYG/graLe7JybCEwE6N27tyvqmLSWmwvz5kXW9947uFhERKRSiSvpO+cSNQXvKqBd1Hrb0Lboa63Gl/Qxs3rAKc656MF/hgIvO+d2Jiim9LJokR93H6BNG2jSJNh4RESk0khUMo/XHKCLmXUK9Q84HZgefYCZNTWzcFyjKTzwzxnAM0mPNCiq2hcRkSRJadIPTdJzCb5qfgEwzTk3z8zGmdnxocP6A9lm9h1+Vr9bw+83s474moL3Uxh2amn4XRERSZK4qvfNLJ+CHfcKcc7F9TC5c24GMCNm201Ryy8ALxTz3uX4zoCVl0r6IiKSJPF25BtH4aTfBBiEH573iQTGVLUp6YuISJLE25FvbFHbQ4PtvIafblfKa8sW+P57v1yjBnTrFmw8IiJSqZSrTd85lwc8DFyWmHCquOj2/J49feIXERFJkER05KsFNE7AeSS6al/P54uISILF25GvqKHtagJ7AbfjJ+SR8lJ7voiIJFG8HfmWU3TvfQOWAH9JVEBVmpK+iIgkUbxJ/3wKJ/0c4HtgTqhtX8ojP1/P6IuISFLF23v/iSTHIcuXw9atfrlpU2jZMtBwRESk8omrI5+ZdTWzw4rZ9wcz65LYsKqg2Kp9s+BiERGRSine3vv3AccVs+9Y4N7EhFOFqWpfRESSLN6k3xs/zW1R/gP0SUw4VZg68YmISJLFm/Sz8B33irITaJCYcKowJX0REUmyeJP+UuCIYvYNwD/SJ2W1bRssWuSXq1Xzo/GJiIgkWLxJfwpwuZn9xcxqAZhZLTP7C34I3ieTFWCVMG8euNATkV27Qu3awcYjIiKVUrzP6d+Nb7d/ALjfzDbih96tBrwI3JGc8KoIDb8rIiIpEO9z+nnAqWY2ADgSP63uBuBt59zs5IVXRag9X0REUiDekj4AzrlZwKwkxVJ1KemLiEgKxDs4z7Fmdkkx+/5iZkcnNqwqxDklfRERSYl4O/LdCNQtZl/t0H4pi9WrYeNGv5yVBR06BBuPiIhUWvEm/e7A58Xs+xLokZhwqiANvysiIikSb9KvBtQrZl8WUCMx4VRBGn5XRERSJN6k/xVwVjH7zgK+Lmaf7I7a80VEJEXi7b3/N+BFM3seeAxYCbQBRgInAaclJ7wqQElfRERSJN7n9F82s1HArcDJoc0GbAUudc69lKT4KrcdO2DBgsj6XnsFF4uIiFR68Vbv45x7AF+6PwY4GxgCtAa+NbPJyQmvklu4EHJz/XKnTlC/frDxiIhIpRZ30gdwzv3inHsT+BT4PfANfrCeoUmIrfLT8LsiIpJCcSd9M2tgZiPN7H9ANnA9sAn4M77EL6Wl9nwREUmhEpO+mVUzs6PN7DlgDfAI0AF4KHTIZc65R51zPyc5zspJSV9ERFKo2I58ZvY34EygOZADvIyfQvddoD5Q5LC8UgpK+iIikkIl9d6/HHDADGCEc+6n8A4zc8kOrNJbvx7WrPHLmZnQuXOw8YiISKVXUvX+JOAXfG/9bDN70Mz6piasKiB6JL699oKMjOBiERGRKqHYpO+cuxBoiR9xby7wJ+AjM1sAXIuvBZCy0vC7IiKSYiV25HPO5TjnnnHODQHaA6OBPOA6/OA8t5vZcDPLTH6olYza80VEJMVKMzjPGufcnc65vYC++B78XYAp+J79UhpK+iIikmKlGpwnzDk31zn3V/zz+acAsxMZVKWXlwfffhtZ18A8IiKSAvFOuFMk59xO/KN8LycmnCpi8WLIyfHLrVtD06bBxiMiIlVCmUr6Uk4afldERAKgpB8EteeLiEgAlPSDoKQvIiIBSHnSN7MhZpZtZovN7Loi9ncws5lm9rWZzTaztlH72pvZ22a2wMzmm1nHVMaeMEr6IiISgJQmfTPLwD/qdxTQEzjDzHrGHHY3MMU5tw8wDpgQtW8KcJdzrgf+scEfkx91gm3ZAsuX++Xq1aF790DDERGRqiPVJf2+wGLn3FLn3A7gWeCEmGN6ArNCy++F94duDqo7594BcM5tdc5tS03YCRT9qF6PHlCzZnCxiIhIlZLqpN8G+CFqfWVoW7SvgJNDyycBWWbWBOgKbDazl8zsCzO7K1RzULFo+F0REQlIOnbkuwo4zMy+AA4DVuGH/q0OHBra3wfYAxgR+2YzG2lmc81s7vr161MWdNzUni8iIgFJddJfBbSLWm8b2raLc261c+5k59z+wPWhbZvxtQJfhpoGcoFXgANiL+Ccm+ic6+2c692sWbNkfY6yU9IXEZGApDrpzwG6mFknM6sJnA5Mjz7AzJqaWTiu0cDkqPc2NLNwJh8AzE9BzInjnJK+iIgEJqVJP1RCvwR4C1gATHPOzTOzcWZ2fOiw/kC2mX0HtABuDb03D1+1P9PMvsHP8vdYKuMvt++/h19+8ctNmkCrVsHGIyIiVUq5xt4vC+fcDGBGzLabopZfAF4o5r3vABW3eBw7/K5ZcLGIiEiVk44d+SovVe2LiEiAlPRTSUlfREQCpKSfSkr6IiISICX9VNm2DRYt8stm0KtXsPGIiEiVo6SfKvPnQ36+X+7SBerUCTYeERGpcpT0U0XD74qISMCU9FNF7fkiIhIwJf1UUdIXEZGAKemngnPw1VeRdSV9EREJgJJ+KqxdCz/95JezsqBDh2DjERGRKklJPxWiq/b32guq6dsuIiKpp+yTCmrPFxGRNKCknwpK+iIikgaU9FNBSV9ERNKAkn6y7dgBCxZE1vfeO7hYRESkSlPST7bsbNi50y936AANGgQbj4iIVFlK+smm4XdFRCRNKOknm9rzRUQkTSjpJ5uSvoiIpAkl/WRT0hcRkTShpJ9MP/0Eq1b55cxM6Nw52HhERKRKU9JPpuhOfL16QfXqwcUiIiJVnpJ+MkVX7ev5fBERCZiSfjKpPV9ERNKIkn4yKemLiEgaUdJPlrw8+PbbyLqSvoiIBExJP1mWLIHffvPLLVtCs2bBxiMiIlWekn6yaPhdERFJM0r6yaL2fBERSTNK+smipC8iImlGST9ZlPRFRCTNKOknwy+/wNKlfrl6dejePdh4REREUNJPjuhH9bp3h1q1gotFREQkREk/GTT8roiIpCEl/WRQe76IiKQhJf1kUNIXEZE0pKSfaM5pYB4REUlLSvqJ9sMPsGWLX27UCNq0CTYeERGRECX9RIut2jcLLhYREZEoSvqJpvZ8ERFJUylP+mY2xMyyzWyxmV1XxP4OZjbTzL42s9lm1jZqX56ZfRl6TU9t5HFS0hcRkTRVPZUXM7MM4CHgSGAlMMfMpjvn5kcddjcwxTn3pJkNACYAZ4f2/eac2y+VMZeakr6IiKSpVJf0+wKLnXNLnXM7gGeBE2KO6QnMCi2/V8T+9JWTA9nZftkMevUKNh4REZEoKS3pA22AH6LWVwL9Yo75CjgZuB84CcgysybOuZ+ATDObC+QCtzvnXom9gJmNBEYCtG/fPvGfoCTz50N+vl/u3Bnq1k3t9UV2Y8uWLWzYsIEdO3YEHYqIxCkjI4OsrCwaN25MrXIO657qpB+Pq4AHzWwE8B9gFZAX2tfBObfKzPYAZpnZN865JdFvds5NBCYC9O7d26UubDT8rqS1nJwc1q1bR9u2balduzamJ0tE0p5zjp07d/Lzzz+zYsUK2rdvX67En+qkvwpoF7XeNrRtF+fcanxJHzOrB5zinNsc2rcq9HWpmc0G9gcKJP1AqT1f0tj69etp1qwZderUCToUEYmTmVGzZk2aNm0KwMaNG2nVqlWZz5fqNv05QBcz62RmNYHTgQK98M2sqZmF4xoNTA5tb2RmtcLHAIcA0R0Ag6ekL2ksJyeHevXqBR2GiJRR/fr1+eWXX8p1jpQmfedcLnAJ8BawAJjmnJtnZuPM7PjQYf2BbDP7DmgB3Bra3gOYa2Zf4Tv43R7T6z94SvqSxnJzc6lePR1b9EQkHjVq1CAvL2/3B5Yg5f8BnHMzgBkx226KWn4BeKGI930IpG9D+bp1sH69X65bFzp1CjYekSKoHV+k4krE369G5EuU2E581fStFRGR9KLMlCiq2hcRkTSnpJ8oSvoiVc7y5csxM8aOHVvmc4wYMULNLpIySvqJoqQvEjgzi/u1fPnyoMNNW/369cPM+OMf/xh0KJJg5lxqx69Jpd69e7u5c+cm/0I7d0K9ehAe5WzTJmjYMPnXFSmFBQsW0KNHj6DDSKp//etfBdb/+9//MnHiREaOHMmhhx5aYN9JJ51E3XKOmumcY/v27VSvXr3MT0bs3LmTvLw8MjMzyxVLonz77bfsvffe7Lnnnqxbt461a9eW+/skiRPP37GZfeac613UPj2/kwjffRdJ+O3bK+GLBGT48OEF1nNzc5k4cSIHHXRQoX2xfvnlF7Kyskp1PTMrd7KuUaMGNWrUKNc5EmnSpElkZWXxr3/9i4MOOohp06Zx3nnnBR3WbpXl51cVqXo/ETT8rkiF0rFjR/r3788XX3zB4MGDadCgAfuEmuV++eUXbrjhBvr160fTpk2pVasWnTt35rrrrmPbtm0FzlNUm370ttdff50+ffqQmZlJq1atuPrqq8nNzS1wjqLa9MPbtmzZwsUXX0zz5s3JzMzkkEMO4ZNPPin0eX766SfOP/98mjRpQr169RgwYABffPEF/fv3p2PHjnF/X3bs2MG//vUvTj31VA488ED2339/Jk2aVOzxL774Iv3796dhw4bUqVOHbt26cemllxaY28E5x2OPPUa/fv2oV68e9erVY++99+amm3Y9qc3YsWOLbXIJ/6yimRkjRoxg5syZ/P73v6devXocd9xxAKxevZorr7yS/fbbj0aNGpGZmUnPnj254447inzGfceOHdx5553st99+1KlThwYNGtC7d28efPBBAO69917MjHfeeafQe7dv306TJk0YMGBAid/XdKKSfiKoPV+kwlmxYgUDBgzgtNNO45RTTmHr1q0ArFq1in/+85+ccsopnHnmmVSvXp3333+fO++8ky+++IK33norrvPPmDGDhx9+mIsuuojzzz+fV199lbvvvptGjRoxZsyYuM4xePBgmjVrxk033cRPP/3EPffcwzHHHMOyZct2lWq3b9/OwIED+fLLLxkxYgR9+/bl66+/ZuDAgTRu3LhU35NXX32VDRs2cO655wL+5mPUqFFkZ2fTrVu3Asdef/313HbbbfTs2ZPLL7+cVq1asWTJEl588UXGjRtHzZo1ATj77LN5+umn6devH9dffz0NGzZk4cKFvPDCC4wbN65U8UWbO3cuL774IhdeeOGueAG+/vprXnrpJU466ST23HNPdu7cyZtvvsl1113H0qVLefTRR3cdu2PHDgYPHszs2bMZNGgQw4cPJzMzk2+++YaXXnqJSy65hHPOOYfRo0czefJkjjzyyAIxvPzyy2zcuJELLrigzJ8j5Zxzlfb1u9/9zqXE0Uc7B/71zDOpuaZIKc2fP7/oHeHf3XR8ldPjjz/uAPf4448X2N6hQwcHuMcee6zQe7Zv3+527NhRaPsNN9zgAPfJJ5/s2rZs2TIHuJtvvrnQtjp16rhly5bt2p6fn+969erlWrZsWeC85557riPms4a3XXzxxQW2T5s2zQHukUce2bXtoYcecoAbP358gWPD2zt06FDosxRnyJAhrmPHji4/P98559z69etdjRo13DXXXFPguE8++cQB7vDDD3e//fZbgX35+fm73v/cc885wA0fPtzl5eUVOC56/eabb3ZAge9XWIcOHdxhhx1WYBvgAPfOO+8UOn7btm27rh9t+PDhrlq1am716tW7tt1xxx0OcKNHjy50fHR8Z5xxhqtVq5b76aefChwzcOBA16hRo0Lfg2Qq9u84CjDXFZMXVb2fCCrpi1Q4jRs3LrKtumbNmrva2HNzc9m0aRMbNmxg4MCBAEVWrxflxBNPLFC1bmYcfvjhrF27dletwu5cfvnlBdbD1ciLFi3ate21114jIyODUaNGFTj2ggsuoEGDBnFdB+CHH37g7bff5pxzztnV3NC0aVOOOeYYpkyZUqBZ4umnnwZgwoQJhfo0hJ+OiD7u7rvvplrMEhB8vQAAHTZJREFUgGWx66W177777vqZRIueQXLHjh1s3LiRDRs2MHjwYPLz84nu3P3000/TqFGjAk0NRcU3cuRItm/fvuvzgG/GmTlzJmeddVbadMKMh5J+eW3aBCtX+uWaNaFr12DjEZG47LnnnmRkZBS57+GHH2afffahVq1aNG7cmGbNmu1qV960aVNc599jjz0KbWvSpAng2+DLco6i3r9s2TJat25daDKlmjVr0qkUw4E/8cQT5Ofnc8ghh7B48eJdrwEDBrB27VpmzIiMnr5o0SLMjH333bfEcy5atIhWrVrRokWLuOOIV9di/tfm5uYyfvx4unbtSmZmJk2aNKFZs2acffbZQMGf36JFi+jevftuk3b//v3p2rVrgf4Njz/+OM65ilW1j9r0y++bbyLLvXqBJjSRiqYSP7ZbkuKmGL7nnnu48sorGTRoEJdeeimtW7emZs2arFq1ihEjRpCfnx/X+Yu7oQDfrFqec8T7/ng553j88ccB34+gKJMnT+b444/ftR5doi+vks4T2/ExrLif3xVXXMEDDzzAsGHDuP7662nevDk1atTg888/59prr4375xfrwgsv5Oqrr+azzz5j//3354knnqB37967vfFJN8pQ5aWqfZFK5amnnqJjx4688cYbBap433zzzQCjKl7Hjh1599132bp1a4HS/s6dO1m2bBkN43iE+L333mPZsmVcdtllHHLIIYX2P/PMM0yfPp1169bRokULunbtyhtvvMFXX31F3759iz1v165defXVV3e9rzjhDocbN24s0CSSk5PDmjVr6Ny5824/Q9hTTz3FH/7wB5599tkC2xcvXlxkfAsXLmT79u3UqlWrxPOOGDGC66+/nkmTJnHCCSewYsUKRo8eHXdc6ULV++WlpC9SqWRkZGBmBUrTubm53H777QFGVbzjjjuOvLw87r///gLbH3vsMbZs2RLXOSZNmkRGRgZjxozh1FNPLfS69NJLyc3NZcqUKQCceeaZAIwZM6bA43lh4e/dWWedBcA111xTqIQd/f0NV9W/++67BY659957S10yz8jIKFQT8uuvv3LvvfcWOvass85i06ZNjB8/vtjPENa0aVNOPPFEpk6dyoMPPkidOnV2fR8qEpX0y0tJX6RSOfXUUxk9ejRHHXUUJ598Mj///DNTp05NqwF0ol1wwQU8+uij3HDDDSxevHjXI3vTpk2jc+fOxVaPh23evJmXXnqJQw89lGbNmhV5zKGHHkrz5s2ZPHkyV199NX379uXaa6/ljjvu4IADDmDYsGG0bNmSZcuW8cILL/Dpp5/SsGFDTjvtNIYNG8aUKVNYtGgRxx9/PI0aNeK7777jrbfe4ttvvwVg4MCBdOvWbdejiZ06deKDDz7g448/pmnTpqX6fpx66qk8+uijDBs2jIEDB7Ju3TomT568qz9EtFGjRvHaa68xfvx45syZw6BBg8jMzGTevHlkZ2cXugkZOXIk06ZN4/XXX+fcc8+lfv36pYotHSjpl0d+fsE2fSV9kQrv6quvxjnHpEmTGDVqFC1btmTYsGGcd9559OzZM+jwCqlVqxYzZ87k6quv5tVXX2XatGn069ePmTNncsEFFxQaUCjW008/TU5ODieffHKxx1SrVo0TTzyRiRMn8uGHH3LwwQdz++23s++++/Lggw9y5513kp+fT7t27Tj66KMLtLdPnTqVQw89lEmTJjFu3DgyMjLo1KkTp5122q5jMjIymD59OpdeeikPPPAANWvWZNCgQbz//vtFNjeU5J577iErK4tp06bx6quv0q5dO0aOHEmfPn0K9favWbMmb7/9Nn/729+YOnUqY8aMITMzky5duhT5ZMeAAQPo3LkzixcvrrDzEmjs/fJYvBi6dPHLLVrA2rXJu5ZIOVWFsfclIi8vj6ZNm9KvX7+07Y9QEfXq1Yu8vDwWLlwYyPXLO/a+2vTLQ8Pvikga+O233wpte+SRR9i8eXOhUeSk7GbNmsX8+fO58MILgw6lzFS9Xx5qzxeRNHDhhReSk5PDwQcfTK1atfjoo4+YOnUqnTt3ZuTIkUGHV+HNmjWLJUuWMGHCBJo1a6akX2Up6YtIGhg0aBAPPfQQt9xyC1u3bqVFixZccMEF3HLLLZp5LgHGjRvHBx98QM+ePXnyyScrZAe+MCX98lDSF5E0cM4553DOOecEHUalNXv27KBDSBi16ZfV1q2wdKlfzsgAdZASEZE0p6RfVvPmRYYv7dYNKtCECyIiUjUp6ZeVqvZFRKSCUdIvKyV9ERGpYJT0y0pJX0REKhgl/bJwTklfREQqHCX9sli5EjZv9ssNG0LbtsHGIyIiEgcl/bKIHX7XLLhYRERE4qSkXxaq2hepcpYvX46ZMXbs2ALbzYwRI0bEdY6xY8diZixfvjzh8T3x/+3de1RV9bbA8e8UhI2p+CKPoRSHJDVL8ZFWXu16CnsNT5nefEuZmJJ1K/XkY6QVQztej5UDNVN801WzTPOaNrpq3jEc2UM8x1QUh4+ObxCMSA8KzvvH3tCGvVFQYCN7fsbYg73W+q215v6tDZP1W7/fWkuWICI16kYypuJZ0r8elvSNqZb69euHiLB79+5Sy6gqkZGRNGjQwOuDaqqzbdu2MXXqVM4XXl6sxgoKCggPD0dEeOedd3wdjnGxpH89LOkbUy0VPuN88eLFpZbZunUrR48epX///oSEhNzwPi9evMiCBQtueDtlsW3bNt566y2vSX/IkCFcvHiR7t27V0ks1/Lll19y8uRJoqKiWLJkCTX5Me43E0v65ZWXBwcO/D7dtq3vYjHGFBMbG0uLFi1ISUnh0qVLXssU/kNQ+A/CjXI4HNSuXbtCtnUjAgICcDgc1KpVPf6sJycnExUVxaxZszh8+PBNc9nh119/9XUIlap6fDtuJvv3Q0GB831UFNSt69t4jDFFatWqRVxcHOfOnWP9+vUey3Nycvj0009p27YtnTt35tdff2Xy5Ml06dKFJk2aEBwczJ133skbb7zBhQsXyrRPb9f0r1y5wvTp04mMjMThcNC2bVtSUlK8rp+Wlsbo0aO5++67qVevHnXq1KFjx44sXLiwWLm4uDjeeustACIjIxGRYn0MSrumn5mZSUJCAi1atCAoKIgWLVqQkJDAuXPnipUrXH/Lli3MnDmTqKgogoODiY6OZunSpWWqi0Jnzpxhw4YNDB06lMcff5xbb72V5ORkr2VVlQULFtClSxfq1q1L3bp1ueeee3jzzTeLlbt06RIzZsygffv21KlTh9DQUDp16kRSUlKxOpJSOlaXPE7ufTRWrVpFx44dCQkJYcyYMUDZj0uhnJwcJk2aROvWrXE4HDRu3Jhu3bqxcuVKAF555RVEhPT0dI91T506RWBgIM8//3zplVpB7Cl75WVN+8ZUa8899xyJiYksXryYvn37Flu2cuVKLl68WHSWf+LECRYuXMgzzzzDwIEDCQwM5JtvvmHGjBmkpqayefPm64rhtdde44MPPqB79+68+uqrnD17loSEBP74xz96lN22bRvbt2/nySefJDIykt9++41PPvmEESNGkJGRwYQJEwAYOXIkOTk5rF27lvfee48mTZoAcO9V/g798ssvPPDAAxw6dIjnn3+eDh06kJqayrx589iyZQvfffedx6N3J06cyMWLFxk5ciTBwcHMmzePuLg47rzzTh588MEyff5ly5ZRUFDA0KFDCQwMZNCgQXz44Yf88ssvhIaGFis7ZMgQUlJS6NKlC5MmTaJBgwakpaWxZs0a3n77bcCZ8Hv16sW2bduIjY1l8ODBOBwO9uzZw2effcZLL71Upri8+fzzz5k9ezajRo3ixRdfLHpsblmPC8D58+fp1q0be/fupW/fvowaNYqCggJSU1PZsGED/fv3Z8SIEcyePZtFixYxffr0YjEsXbqUgoICXnjhhev+HGWmqjX21bFjR61wr7+u6rw9j+qUKRW/fWMqyb59+7zOL/w6V8fX9erZs6cGBAToyZMni83v2rWrBgUFaUZGhqqq5uXl6aVLlzzWnzx5sgK6c+fOonlHjhxRQKeU+L0HdNiwYUXTaWlpKiLas2dPzc/PL5r/448/qogooEeOHCman5ub67H/goIC7dGjh9avX79YfFOmTPFYv9DixYsV0K1btxbNmzhxogI6Z86cYmWTkpIU0MmTJ3us3759e83Lyyuaf/z4cQ0KCtL+/ft77LM0rVq10h49ehRN7969WwGdO3dusXKrVq1SQAcPHqwFBQUedVDor3/9qwI6YcIEj325lxs2bJhSyhen5HEqPJ6BgYFefzfKc1xGjRqlgM6fP/+q8d1///3arFmzYt8LVdWWLVtq69atvcZdUmm/x+6AH7SUvGjN++VlZ/rGVHvDhw+noKCAZcuWFc1LS0vj22+/pXfv3kVnyUFBQUXX4/Pz88nOziYzM5OHH34YgJ07d5Z73+vWrUNVee211wgICCia36FDBx555BGP8rfcckvR+3/961+cO3eOrKwsYmNjycnJIS0trdwxFFq7di1hYWHEx8cXmz9y5EjCwsJYu3atxzqjR48mKCioaDo8PJzo6GivzdLe7Nixg7S0NIYNG1Y0r127drRv355FixYVK1t4yWPmzJkefRHcp1NSUmjYsKFHk3/JctfjiSeeoLWXR6OX9bhcuXKFlStX0rp1a496LhlffHw8p06dYuPGjUXztm/fTnp6eoX1MbkWS/rlZUnfmGqvT58+NGjQoFgv/sKEU/K66dy5c7n33nsJDg6mUaNGhIWF8dBDDwGQnZ1d7n0fPnwYgFatWnksa9Omjce83Nxcxo4dS0REBCEhITRp0oSwsDAmTZp03TEUOnLkCHfddReBgcWv5AYGBhIdHV0UqztvlyAaN27s0QegNMnJydSuXZuYmBgOHTpU9OrVqxc//PAD/3D7G5qenk6zZs1o2rTpVbeZnp5Oq1atcFTCI8yjo6O9zi/rccnMzCQ7O5v27dtfc1/PPvssoaGhxfo3JCcnExQUxNChQyvg01ybXdMvjzNnnC+AOnXAyy+HMTebmjiSyuFwMHDgQObOncuOHTvo0qULy5cvp3nz5vTq1auo3KxZs3j99deJjY3l5Zdf5rbbbiMoKIgTJ04QFxfHlStXKj3WgQMHsmHDBuLj4+nevTuNGzcmICCAjRs38t5771VJDO7cWyfcaRm+KLm5uaxevZrLly8TExPjtcyiRYt4//33byjG0pTWiS8/P7/UderUqeN1fmUcl5CQEAYPHsz8+fM5c+YMISEhrFmzht69exMWFlbu7V0PS/rlsWfP7+/btoVqMjTGGONp+PDhzJ07l8WLF5OVlcXp06eZNGlSsebW5cuXc8cdd/Dll18Wm79p06br3m/hmXJaWhpRUVHFlu3bt6/Y9Pnz59mwYQNDhgzhww8/LLbs66+/9th2aUntarEcOHCA/Pz8Ymf7+fn5HDx40OtZ/Y1YvXo1ubm5TJs2jZYtW3osnz17NitWrGDGjBkEBQURHR3NunXrOHPmzFXP9qOjo0lLSyMvL4/g4OBSyzVq1AiArKysoveA1xaNqynPcWnSpAkNGzbk73//e5m2HR8fz5w5c1i6dCmhoaFcuHChypr2wQfN+yLyqIgcEJFDIvKGl+W3i8j/isg/RGSbiDQvsby+iBwXkaSS61Y6a9o35qbRoUMH2rdvz6pVq5gzZw4i4tG0HxAQgIgUO4vNz8/n3Xffve799u7dGxFh1qxZFBQO7wV27drlkTAKz6pLnkWfOnXK69Cwuq4hwllZWWWK5amnniIjI8NjWwsWLCAjI4Onn366TNspq+TkZBo1asS4cePo27evx2v48OGcO3eOdevWATBo0CAAxo8f73Hm7F4ngwYNIjs7m8TERI99upcrbKovWc9/+9vfyvU5ynNcatWqxYABA9i3b5/XYYklt3Hvvfdy3333sWjRIpKTk4mIiCA2NrZc8d2IKj3TF5EAYA7wCHAc+F5E1quq+7+/M4FlqrpURHoC04EhbsvfAbZXVczFWNI35qYyfPhwxowZw6ZNm3jooYc8zmz79u3LhAkTeOyxx+jTpw85OTl8/PHHN3SznVatWpGQkEBSUhI9e/bkmWee4ezZsyQlJdGuXTtSU1OLytarV4/Y2FhWrFhBSEgInTt35tixY8yfP5/IyEiP6+hdu3YF4C9/+QuDBg0qugdA21JuEjZ+/Hg++eQTEhIS2LVrFzExMaSmppKcnMxdd93F+PHjr/tzlpSWlsaOHTuIi4vz6ENQqHfv3tSuXZvk5GT69etHv379ePbZZ1m2bBnp6en07t2bhg0bcvDgQTZv3sxPP/0EOMe4f/HFFyQmJvL9998TGxuLw+Fg7969HDhwoCjJDxgwgIkTJxIfH09aWhqNGjVi06ZNZGZmluuzlPe4JCYmsmXLFl544QW++uorunXrhqqSmppKfn4+y5cvL1Y+Pj6+aHjelClTqvaGSqV166+MF3A/sNltegIwoUSZvUAL13sBctyWdQRWAnFA0rX2V+FD9mJifh9LtG1bxW7bmEpWlqE+NU1WVpY6HA4FdNmyZR7L8/Pzddq0aRoVFaVBQUEaERGh48aN03379nkMzyvrkD1V5zCtxMREjYiI0KCgIL377rt1xYoVXofcZWRk6PDhw7VZs2YaHBysbdu21Y8++sjrEDxV5/C1yMhIDQwMLBZPaeXPnj2ro0aN0vDwcA0MDNTw8HAdPXp00bDFQqWtr6rao0cPvf32273U8O/Gjh2rgK5fv/6q5WJjY7VWrVr6888/F9VVUlKSxsTEaEhIiNatW1fvuecenTp1arH1Ll68qImJidqmTRsNDg7W0NBQ7dSpk8dwxG+//VYfeOABDQ4O1saNG+uIESM0Ozu71CF7JY9nofIel+zsbB03bpxGRUVp7dq1tVGjRtqtWzddtWqVx7Zzc3O1fv36WqtWLT169OhV66ukGx2yJ1qFvXhEpC/wqKq+4JoeAnRR1ZfcynwM7FTVD0SkD/Ap0ATIBrYAg4GHgU7u67mtHw/EA0RERHQ8duxYxQSfn++8+15ennP63Dlwu2ZkTHW3f/9+r0OTjDFVKy8vj2bNmtG5c+dy3wCqLL/HIvKjqnbytqw69kQbC/QQkVSgB3ACKABGAxtV9fjVVlbVj1S1k6p2qtDekOnpvyf85s0t4RtjjLkuKSkpZGdnex3XX9mquvf+CaCF23Rz17wiqnoS6AMgInWBZ1T1vIjcD/ybiIwG6gJBIpKrqh6dASuFXc83xhhzA7744guOHTvG1KlTadOmDU899VSVx1DVSf97oKWIROJM9v2Bge4FRKQJkKWqV3Be818EoKqD3MrE4Wzer5qED5b0jTHG3JAxY8Zw8uTJogf3lHZPhMpUpUlfVfNF5CVgMxAALFLVvSLyNs6OB+uBh4DprntUbwcSqjLGUlnSN8YYcwOOHj3q6xCq/uY8qroR2Fhi3ptu79cAa66xjSXAkkoIr3SW9I0xxtzk7I58ZbVggTPx79kDpdyr2RhjjKnOLOmXVWys82XMTUxVy30rV2NM9VARQ+yr45A9Y0wlCAwMvOqDR4wx1dvly5dvuPOfJX1j/ITD4SA3N9fXYRhjrlNOTg716tW7oW1Y0jfGT4SFhZGRkcGFCxcqpJnQGFP5VJVLly6RmZlJdnZ2sacHXg+7pm+Mn3A4HDRt2pTTp0+TV3h3SWNMtRcQEEC9evWIiIi46qOFy8KSvjF+JDQ0lNDQUF+HYYzxEWveN8YYY/yEJX1jjDHGT1jSN8YYY/yEJX1jjDHGT1jSN8YYY/yEJX1jjDHGT1jSN8YYY/yE1OQ7c4lIBnCsgjfbBMis4G0aT1bPVcPquWpYPVcNq2en21U1zNuCGp30K4OI/KCqnXwdR01n9Vw1rJ6rhtVz1bB6vjZr3jfGGGP8hCV9Y4wxxk9Y0i+/j3wdgJ+weq4aVs9Vw+q5alg9X4Nd0zfGGGP8hJ3pG2OMMX7Ckn4ZicijInJARA6JyBu+jqcmEpEWIrJVRPaJyF4RecXXMdVkIhIgIqkissHXsdRkItJARNaISJqI7BeR+30dU00kIq+6/m78JCL/LSIOX8dUHVnSLwMRCQDmAI8BbYABItLGt1HVSPnA66raBugKJFg9V6pXgP2+DsIPfABsUtVWQDusziuciIQDLwOdVLUtEAD0921U1ZMl/bK5DzikqodV9RKwEvizj2OqcVT1lKrucr3/Fecfx3DfRlUziUhz4Algoa9jqclEJBToDiQDqOolVT3v26hqrEAgREQCgTrASR/HUy1Z0i+bcOCfbtPHsWRUqUTkDiAG2OnbSGqs94HxwBVfB1LDRQIZwGLXpZSFInKLr4OqaVT1BDAT+Bk4Bfyiql/5NqrqyZK+qXZEpC7wKfCfqprj63hqGhF5Ejirqj/6OhY/EAh0AOapagzwG2B9giqYiDTE2foaCdwG3CIig30bVfVkSb9sTgAt3Kabu+aZCiYitXEm/BRV/czX8dRQDwK9ReQozktVPUVkhW9DqrGOA8dVtbDFag3OfwJMxXoYOKKqGap6GfgMeMDHMVVLlvTL5nugpYhEikgQzg4i630cU40jIoLz2ud+VZ3l63hqKlWdoKrNVfUOnN/lLapqZ0WVQFVPA/8Ukbtcs/4E7PNhSDXVz0BXEanj+jvyJ6zDpFeBvg7gZqCq+SLyErAZZ6/QRaq618dh1UQPAkOAPSKy2zVvoqpu9GFMxtyoMUCK64ThMPCcj+OpcVR1p4isAXbhHAWUit2dzyu7I58xxhjjJ6x53xhjjPETlvSNMcYYP2FJ3xhjjPETlvSNMcYYP2FJ3xhjjPETlvSNqcFEJE5EtJSXT+8BLyJLROS4L2Mwxt/YOH1j/EM/nHeHc5fvi0CMMb5jSd8Y/7BbVQ/5OghjjG9Z874xxv0yQHcR+VxEckXknIjMEZGQEmWbicgyEckUkTwR+Ye3h5u4blu9XEROu8odFpEPvJSLEZH/E5ELIpIuIi+WWP4HEVkqIidd2zklIhtE5NaKrwljajY70zfGPwS4njPu7oqqlny07gpgNTAXuA94E7gFiANwPRb2G6AhMBHnI6cHA8tFpI6qfuQqFwl8B1xwbSMdiABiS+yvPvAxzkf9vo3zFrXzROSAqm51lVkO3A6Mc+2vKc57q9e5noowxp9Z0jfGP6R5mfc/wJMl5m1U1bGu91+JiAJvi8g0VT2IMym3BP5dVbe5yn0pIk2BRBFJVtUC4C0gBGinqifdtr+0xP7qAaMLE7yIbAd6AQOAwqR/P85nMKS4rfdJmT61MaYYS/rG+Ien8ezI5633/uoS0yuBRJxn/QeB7sAJt4RfaAWwGGgD7MF5Rr+hRML35oLbGT2qmiciB3G2ChT6HhjnenraFuAntYeGGHNdLOkb4x9+KmNHvjOlTIe7fjYCTnlZ77TbcoDGeP6T4U22l3l5gMNt+llgCjAe52WAUyLyIZDo5fKEMeYqrCOfMcZd01KmT7h+ZgF/8LLeH9yWA2Ty+z8KN0RVz6pqgqqGA62AJTgvH4ysiO0b408s6Rtj3P1Hien+wBVgp2v6G6C5iDxYotxA4CywzzX9FfCkiDSryOBU9YCqTsTZQtC2IrdtjD+w5n1j/EN7EWniZf4Pqup+k57HReS/cCbt+3A2qy9T1XTX8iXAK8BnIjIJZxP+IOARYKSrEx+u9R4HdojINOAQzjP/R1XVY3hfaUQkFPgaSMHZGfEy8Gecowe+Kut2jDFOlvSN8Q+l9XYPw9kUX2gw8DowCrgELAAKe/Ojqr+JSA9gBvAuzt73B4AhqrrCrdxREemKsxPgdKAuzksE68oZ97+AXcAInMP2rrj2N0hVy7stY/yeWCdYY4yIxOHsfd/S7txnTM1l1/SNMcYYP2FJ3xhjjPET1rxvjDHG+Ak70zfGGGP8hCV9Y4wxxk9Y0jfGGGP8hCV9Y4wxxk9Y0jfGGGP8hCV9Y4wxxk/8P4F8OaJuXFGPAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_V0JIBUiuk55"
      },
      "source": [
        "# Fine tuning convolutional layers MobileNetV2 model"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kKVlfogkupkQ"
      },
      "source": [
        "Download the best weights"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ht4fiHErunl3"
      },
      "source": [
        "model.load_weights(\"inceptionv3_best.h5\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DAI07hsdusB6"
      },
      "source": [
        "Defrost the last 2 blocks of MobileNetV2 model\n",
        "\n",
        "> Блок с отступами\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PSgcVDfxuu1B"
      },
      "source": [
        "pretrained_model.trainable = False\n",
        "for layer in model.layers[:290]:\n",
        "   layer.trainable = False\n",
        "for layer in model.layers[290:]:\n",
        "   layer.trainable = True"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "prGhlafxuxx-"
      },
      "source": [
        "\n",
        "Compile a network with defrosted layers. And start learning"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qB8egg2zu0WD",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3842b52c-ba73-4b8e-c6b2-f139378f5873"
      },
      "source": [
        "model.compile(loss='categorical_crossentropy',\n",
        "              optimizer='adam', \n",
        "              metrics=[\"accuracy\"])\n",
        "\n",
        "filepath=\"inceptionv3_fine_tuned.h5\"\n",
        "checkpoint = ModelCheckpoint(filepath, monitor='val_loss', verbose=1, save_best_only=True, mode='min')\n",
        "callbacks_list = [checkpoint, learning_rate_reduction]\n",
        "\n",
        "\n",
        "steps_per_epoch = 32\n",
        "history = model.fit_generator(\n",
        "    train_generator,\n",
        "    steps_per_epoch=steps_per_epoch,\n",
        "    validation_data=test_generator,\n",
        "    validation_steps=5,\n",
        "    callbacks=callbacks_list,\n",
        "    epochs=10,\n",
        "    verbose=1)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0153 - accuracy: 0.9956\n",
            "Epoch 00001: val_loss improved from inf to 0.01012, saving model to inceptionv3_fine_tuned.h5\n",
            "32/32 [==============================] - 40s 1s/step - loss: 0.0153 - accuracy: 0.9956 - val_loss: 0.0101 - val_accuracy: 0.9969\n",
            "Epoch 2/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0176 - accuracy: 0.9941\n",
            "Epoch 00002: val_loss improved from 0.01012 to 0.00894, saving model to inceptionv3_fine_tuned.h5\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.0176 - accuracy: 0.9941 - val_loss: 0.0089 - val_accuracy: 0.9969\n",
            "Epoch 3/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0178 - accuracy: 0.9956\n",
            "Epoch 00003: val_loss improved from 0.00894 to 0.00609, saving model to inceptionv3_fine_tuned.h5\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.0178 - accuracy: 0.9956 - val_loss: 0.0061 - val_accuracy: 0.9969\n",
            "Epoch 4/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0170 - accuracy: 0.9951\n",
            "Epoch 00004: val_loss did not improve from 0.00609\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.0170 - accuracy: 0.9951 - val_loss: 0.0120 - val_accuracy: 0.9969\n",
            "Epoch 5/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0145 - accuracy: 0.9966\n",
            "Epoch 00005: val_loss did not improve from 0.00609\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.0145 - accuracy: 0.9966 - val_loss: 0.0067 - val_accuracy: 1.0000\n",
            "Epoch 6/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0196 - accuracy: 0.9937\n",
            "Epoch 00006: val_loss did not improve from 0.00609\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.0196 - accuracy: 0.9937 - val_loss: 0.0093 - val_accuracy: 0.9969\n",
            "Epoch 7/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0243 - accuracy: 0.9917\n",
            "Epoch 00007: val_loss did not improve from 0.00609\n",
            "32/32 [==============================] - 38s 1s/step - loss: 0.0243 - accuracy: 0.9917 - val_loss: 0.0099 - val_accuracy: 0.9937\n",
            "Epoch 8/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0185 - accuracy: 0.9932\n",
            "Epoch 00008: val_loss did not improve from 0.00609\n",
            "32/32 [==============================] - 39s 1s/step - loss: 0.0185 - accuracy: 0.9932 - val_loss: 0.0130 - val_accuracy: 0.9969\n",
            "Epoch 9/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0140 - accuracy: 0.9956\n",
            "Epoch 00009: val_loss did not improve from 0.00609\n",
            "32/32 [==============================] - 38s 1s/step - loss: 0.0140 - accuracy: 0.9956 - val_loss: 0.0083 - val_accuracy: 0.9969\n",
            "Epoch 10/10\n",
            "32/32 [==============================] - ETA: 0s - loss: 0.0301 - accuracy: 0.9937\n",
            "Epoch 00010: val_loss did not improve from 0.00609\n",
            "32/32 [==============================] - 38s 1s/step - loss: 0.0301 - accuracy: 0.9937 - val_loss: 0.0142 - val_accuracy: 0.9937\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QE9W46fi5p3e"
      },
      "source": [
        "Test Score"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GK8mctuZ5p3f",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "15d35435-175c-4334-ac0f-1757c976a4dd"
      },
      "source": [
        "test_score = model.evaluate_generator(testing_generator, batch_size)\n",
        "\n",
        "print(\"[INFO] accuracy: {:.2f}%\".format(test_score[1] * 100)) \n",
        "print(\"[INFO] Loss: \",test_score[0])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches (in this case, 64 batches). You may need to use the repeat() function when building your dataset.\n",
            "[INFO] accuracy: 100.00%\n",
            "[INFO] Loss:  0.0037735644727945328\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lkxfY8jF5p3h"
      },
      "source": [
        "Visualize the history of network learning: accuracy, loss in graphs.\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dXjVyHfN5p3h",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 822
        },
        "outputId": "3f78f729-95e2-4f19-bd18-3b7c7489d04c"
      },
      "source": [
        "#Plot the Graph\n",
        "\n",
        "# Loss Curves\n",
        "plt.figure(figsize=[8,6])\n",
        "plt.plot(history.history['loss'],'r',linewidth=3.0)\n",
        "plt.plot(history.history['val_loss'],'b',linewidth=3.0)\n",
        "plt.legend(['Training loss', 'Validation Loss'],fontsize=18)\n",
        "plt.xlabel('Epochs ',fontsize=16)\n",
        "plt.ylabel('Loss',fontsize=16)\n",
        "plt.title('Loss Curves',fontsize=16)\n",
        "  \n",
        "# Accuracy Curves\n",
        "plt.figure(figsize=[8,6])\n",
        "plt.plot(history.history['accuracy'],'r',linewidth=3.0)\n",
        "plt.plot(history.history['val_accuracy'],'b',linewidth=3.0)\n",
        "plt.legend(['Training Accuracy', 'Validation Accuracy'],fontsize=18)\n",
        "plt.xlabel('Epochs ',fontsize=16)\n",
        "plt.ylabel('Accuracy',fontsize=16)\n",
        "plt.title('Accuracy Curves',fontsize=16)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Accuracy Curves')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gUz8hhZw5p3j"
      },
      "source": [
        "Visualize the history of network learning for testing data in confusion matrix format"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "g5pRUJQS5p3j",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "e44087a4-7e46-4113-9ee0-605804b0eac8"
      },
      "source": [
        "#Plot the confusion matrix. Set Normalize = True/False\n",
        "\n",
        "def plot_confusion_matrix(cm, classes, normalize=True, title='Confusion matrix', cmap=plt.cm.Blues):\n",
        "    \"\"\"\n",
        "    This function prints and plots the confusion matrix.\n",
        "    Normalization can be applied by setting `normalize=True`.\n",
        "    \"\"\"\n",
        "    plt.figure(figsize=(20,20))\n",
        "\n",
        "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
        "    plt.title(title)\n",
        "    plt.colorbar()\n",
        "\n",
        "    tick_marks = np.arange(len(classes))\n",
        "    plt.xticks(tick_marks, classes, rotation=45)\n",
        "    plt.yticks(tick_marks, classes)\n",
        "\n",
        "    if normalize:\n",
        "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
        "        cm = np.around(cm, decimals=2)\n",
        "        cm[np.isnan(cm)] = 0.0\n",
        "        print(\"Normalized confusion matrix\")\n",
        "    else:\n",
        "        print('Confusion matrix, without normalization')\n",
        "    thresh = cm.max() / 2.\n",
        "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
        "        plt.text(j, i, cm[i, j],\n",
        "                 horizontalalignment=\"center\",\n",
        "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
        "    plt.tight_layout()\n",
        "    plt.ylabel('True label')\n",
        "    plt.xlabel('Predicted label')\n",
        "\n",
        "#Print the Target names\n",
        "from sklearn.metrics import classification_report, confusion_matrix\n",
        "import itertools \n",
        "#shuffle=False\n",
        "\n",
        "target_names = []\n",
        "for key in train_generator.class_indices:\n",
        "    target_names.append(key)\n",
        "\n",
        "# print(target_names)\n",
        "\n",
        "#Confution Matrix \n",
        "\n",
        "Y_pred = model.predict_generator(testing_generator)\n",
        "y_pred = np.argmax(Y_pred, axis=1)\n",
        "print('Confusion Matrix')\n",
        "cm = confusion_matrix(testing_generator.classes, y_pred)\n",
        "plot_confusion_matrix(cm, target_names, title='Confusion Matrix')\n",
        "\n",
        "#Print Classification Report\n",
        "print('Classification Report')\n",
        "print(classification_report(testing_generator.classes, y_pred, target_names=target_names))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Confusion Matrix\n",
            "Normalized confusion matrix\n",
            "Classification Report\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "     animals       1.00      1.00      1.00       100\n",
            "       other       1.00      1.00      1.00       100\n",
            "      person       1.00      1.00      1.00       100\n",
            "\n",
            "    accuracy                           1.00       300\n",
            "   macro avg       1.00      1.00      1.00       300\n",
            "weighted avg       1.00      1.00      1.00       300\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x1440 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}