eva_alk_p1-v2.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "eva-alk-p1",
      "provenance": [],
      "mount_file_id": "1WtcXGZobVoQKAOrrx5amxD0mD1fRfnNa",
      "authorship_tag": "ABX9TyPWGwjwzvUVts9pihSqbhF9",
      "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/pjablonski123/alk_eye-tech/blob/master/eva_alk_p1-v2.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Clh5cxfJ3RHq",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import keras\n",
        "import numpy as np\n",
        "from keras.preprocessing.image import ImageDataGenerator \n",
        "from keras.applications.inception_v3 import InceptionV3 \n",
        "from keras.layers import Dense, GlobalAveragePooling2D \n",
        "from keras.preprocessing.image import ImageDataGenerator, load_img, img_to_array\n",
        "from keras.models import Model \n",
        "from keras.models import Sequential\n",
        "from keras import optimizers \n",
        "from keras import layers\n",
        "from keras import models\n",
        "import matplotlib.pyplot as plt \n",
        "%matplotlib inline\n",
        "from keras.models import load_model"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "c_4Q70Ly3Wlf",
        "colab_type": "code",
        "outputId": "381427ac-e7ff-4c23-d6df-5aecb7f2993c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "cd '/content/drive/My Drive/TEST1'"
      ],
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/content/drive/My Drive/TEST1\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Uq0jL2uEe3nq",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# pre-trained Inception V3 model \n",
        "base_model = InceptionV3(weights='imagenet', include_top=False) \n",
        "base_model.trainable = False\n",
        "\n",
        "#x = base_model.output\n",
        "#x = GlobalAveragePooling2D()(x) \n",
        "#x = Dense(64, activation='relu')(x) \n",
        "#predictions = Dense(2, activation='softmax')(x) \n",
        "\n",
        "#model = Model(inputs=base_model.input, outputs=predictions) \n",
        "\n",
        "#model.trainable = False\n",
        "\n",
        "model = models.Sequential()\n",
        "model.add(base_model)\n",
        "#model.add(layers.Flatten()) \n",
        "model.add(GlobalAveragePooling2D())\n",
        "model.add(layers.Dense(64, activation='relu')) \n",
        "model.add(layers.Dense(2, activation='softmax'))\n",
        "#model.add(layers.Dense(1, activation='sigmoid'))\n",
        "\n",
        "adam = optimizers.adam(lr=0.01)\n",
        "model.compile(optimizer=adam, loss='categorical_crossentropy', \n",
        "              metrics=['accuracy'])"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1GaFfpZI1Mus",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 289
        },
        "outputId": "3d79869b-755c-435f-9743-9c94ff939658"
      },
      "source": [
        "model.summary()"
      ],
      "execution_count": 61,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Model: \"sequential_5\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "inception_v3 (Model)         (None, None, None, 2048)  21802784  \n",
            "_________________________________________________________________\n",
            "global_average_pooling2d_6 ( (None, 2048)              0         \n",
            "_________________________________________________________________\n",
            "dense_15 (Dense)             (None, 64)                131136    \n",
            "_________________________________________________________________\n",
            "dense_16 (Dense)             (None, 2)                 130       \n",
            "=================================================================\n",
            "Total params: 21,934,050\n",
            "Trainable params: 131,266\n",
            "Non-trainable params: 21,802,784\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RVig2h_-9E36",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "train_datagen = ImageDataGenerator(rescale=1./255) \n",
        "test_datagen = ImageDataGenerator(rescale=1./255) \n",
        "train_dir = '/content/drive/My Drive/TEST1/' \n",
        "test_dir = '/content/drive/My Drive/TEST2/'"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jYWEYv5n9alC",
        "colab_type": "code",
        "outputId": "9ec80f84-00f0-44cb-9d43-46f16b0c15be",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        }
      },
      "source": [
        "train_generator = train_datagen.flow_from_directory( train_dir,\n",
        "  target_size=(299, 299), batch_size=16, class_mode='categorical')\n",
        "\n",
        "test_generator = test_datagen.flow_from_directory( test_dir,\n",
        "  target_size=(299, 299), batch_size=16, class_mode='categorical')"
      ],
      "execution_count": 70,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 500 images belonging to 2 classes.\n",
            "Found 31 images belonging to 2 classes.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f3VZpDeDIujA",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "# Setup a callback to save the best model \n",
        "callbacks = [keras.callbacks.ModelCheckpoint(\n",
        "  'model.{epoch:02d}-{val_accuracy:.2f}.hdf5', \n",
        "  monitor='val_accuracy', verbose=1, save_best_only=True, \n",
        "  mode='max', period=1)]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PAyz__fzLQET",
        "colab_type": "code",
        "outputId": "3bfd4d7f-7d9a-4318-91fc-ab7cf5d076bc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "source": [
        "history = model.fit_generator(train_generator, verbose=1, \n",
        "  steps_per_epoch=len(train_generator), epochs=100,\n",
        "  validation_data=test_generator,\n",
        "  validation_steps=len(test_generator), callbacks=callbacks)"
      ],
      "execution_count": 72,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "32/32 [==============================] - 10s 302ms/step - loss: 0.8200 - accuracy: 0.6240 - val_loss: 0.4325 - val_accuracy: 0.6774\n",
            "\n",
            "Epoch 00001: val_accuracy improved from -inf to 0.67742, saving model to model.01-0.68.hdf5\n",
            "Epoch 2/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.5174 - accuracy: 0.7580 - val_loss: 1.7193 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00002: val_accuracy did not improve from 0.67742\n",
            "Epoch 3/100\n",
            "32/32 [==============================] - 4s 111ms/step - loss: 0.3582 - accuracy: 0.8460 - val_loss: 0.9542 - val_accuracy: 0.7419\n",
            "\n",
            "Epoch 00003: val_accuracy improved from 0.67742 to 0.74194, saving model to model.03-0.74.hdf5\n",
            "Epoch 4/100\n",
            "32/32 [==============================] - 3s 100ms/step - loss: 0.3889 - accuracy: 0.8340 - val_loss: 0.5291 - val_accuracy: 0.8387\n",
            "\n",
            "Epoch 00004: val_accuracy improved from 0.74194 to 0.83871, saving model to model.04-0.84.hdf5\n",
            "Epoch 5/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.2935 - accuracy: 0.8740 - val_loss: 0.5972 - val_accuracy: 0.6129\n",
            "\n",
            "Epoch 00005: val_accuracy did not improve from 0.83871\n",
            "Epoch 6/100\n",
            "32/32 [==============================] - 3s 107ms/step - loss: 0.2765 - accuracy: 0.8780 - val_loss: 2.7264 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00006: val_accuracy did not improve from 0.83871\n",
            "Epoch 7/100\n",
            "32/32 [==============================] - 4s 110ms/step - loss: 0.2381 - accuracy: 0.9020 - val_loss: 1.4756 - val_accuracy: 0.5161\n",
            "\n",
            "Epoch 00007: val_accuracy did not improve from 0.83871\n",
            "Epoch 8/100\n",
            "32/32 [==============================] - 3s 102ms/step - loss: 0.1943 - accuracy: 0.9160 - val_loss: 0.9209 - val_accuracy: 0.5161\n",
            "\n",
            "Epoch 00008: val_accuracy did not improve from 0.83871\n",
            "Epoch 9/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.1527 - accuracy: 0.9500 - val_loss: 3.2492 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00009: val_accuracy did not improve from 0.83871\n",
            "Epoch 10/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.2478 - accuracy: 0.8940 - val_loss: 3.5191 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00010: val_accuracy did not improve from 0.83871\n",
            "Epoch 11/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.1790 - accuracy: 0.9280 - val_loss: 3.6226 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00011: val_accuracy did not improve from 0.83871\n",
            "Epoch 12/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.1538 - accuracy: 0.9400 - val_loss: 1.4336 - val_accuracy: 0.5806\n",
            "\n",
            "Epoch 00012: val_accuracy did not improve from 0.83871\n",
            "Epoch 13/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.3320 - accuracy: 0.8680 - val_loss: 2.3750 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00013: val_accuracy did not improve from 0.83871\n",
            "Epoch 14/100\n",
            "32/32 [==============================] - 3s 101ms/step - loss: 0.2175 - accuracy: 0.9000 - val_loss: 3.0822 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00014: val_accuracy did not improve from 0.83871\n",
            "Epoch 15/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.1722 - accuracy: 0.9340 - val_loss: 2.3301 - val_accuracy: 0.4839\n",
            "\n",
            "Epoch 00015: val_accuracy did not improve from 0.83871\n",
            "Epoch 16/100\n",
            "32/32 [==============================] - 3s 100ms/step - loss: 0.2165 - accuracy: 0.9100 - val_loss: 1.9051 - val_accuracy: 0.4839\n",
            "\n",
            "Epoch 00016: val_accuracy did not improve from 0.83871\n",
            "Epoch 17/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.2525 - accuracy: 0.8920 - val_loss: 2.5001 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00017: val_accuracy did not improve from 0.83871\n",
            "Epoch 18/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.2068 - accuracy: 0.9180 - val_loss: 1.9986 - val_accuracy: 0.6129\n",
            "\n",
            "Epoch 00018: val_accuracy did not improve from 0.83871\n",
            "Epoch 19/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.2099 - accuracy: 0.9380 - val_loss: 2.3621 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00019: val_accuracy did not improve from 0.83871\n",
            "Epoch 20/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.1938 - accuracy: 0.9200 - val_loss: 1.4442 - val_accuracy: 0.5484\n",
            "\n",
            "Epoch 00020: val_accuracy did not improve from 0.83871\n",
            "Epoch 21/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.2154 - accuracy: 0.9080 - val_loss: 0.9859 - val_accuracy: 0.7742\n",
            "\n",
            "Epoch 00021: val_accuracy did not improve from 0.83871\n",
            "Epoch 22/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.3526 - accuracy: 0.8640 - val_loss: 1.0670 - val_accuracy: 0.6452\n",
            "\n",
            "Epoch 00022: val_accuracy did not improve from 0.83871\n",
            "Epoch 23/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.1665 - accuracy: 0.9340 - val_loss: 1.7322 - val_accuracy: 0.6129\n",
            "\n",
            "Epoch 00023: val_accuracy did not improve from 0.83871\n",
            "Epoch 24/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.1215 - accuracy: 0.9540 - val_loss: 1.8681 - val_accuracy: 0.7097\n",
            "\n",
            "Epoch 00024: val_accuracy did not improve from 0.83871\n",
            "Epoch 25/100\n",
            "32/32 [==============================] - 3s 100ms/step - loss: 0.1536 - accuracy: 0.9360 - val_loss: 0.4590 - val_accuracy: 0.5484\n",
            "\n",
            "Epoch 00025: val_accuracy did not improve from 0.83871\n",
            "Epoch 26/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.2222 - accuracy: 0.9180 - val_loss: 1.0123 - val_accuracy: 0.7419\n",
            "\n",
            "Epoch 00026: val_accuracy did not improve from 0.83871\n",
            "Epoch 27/100\n",
            "32/32 [==============================] - 3s 100ms/step - loss: 0.2583 - accuracy: 0.9040 - val_loss: 1.4548 - val_accuracy: 0.5806\n",
            "\n",
            "Epoch 00027: val_accuracy did not improve from 0.83871\n",
            "Epoch 28/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.1984 - accuracy: 0.9140 - val_loss: 2.1844 - val_accuracy: 0.6129\n",
            "\n",
            "Epoch 00028: val_accuracy did not improve from 0.83871\n",
            "Epoch 29/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.2520 - accuracy: 0.9160 - val_loss: 0.4676 - val_accuracy: 0.7097\n",
            "\n",
            "Epoch 00029: val_accuracy did not improve from 0.83871\n",
            "Epoch 30/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.1701 - accuracy: 0.9300 - val_loss: 3.6327 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00030: val_accuracy did not improve from 0.83871\n",
            "Epoch 31/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.2114 - accuracy: 0.9120 - val_loss: 1.8631 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00031: val_accuracy did not improve from 0.83871\n",
            "Epoch 32/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.1141 - accuracy: 0.9600 - val_loss: 2.9762 - val_accuracy: 0.4839\n",
            "\n",
            "Epoch 00032: val_accuracy did not improve from 0.83871\n",
            "Epoch 33/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.1548 - accuracy: 0.9320 - val_loss: 2.5105 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00033: val_accuracy did not improve from 0.83871\n",
            "Epoch 34/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.3627 - accuracy: 0.8620 - val_loss: 0.4973 - val_accuracy: 0.6129\n",
            "\n",
            "Epoch 00034: val_accuracy did not improve from 0.83871\n",
            "Epoch 35/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.4066 - accuracy: 0.9260 - val_loss: 0.6411 - val_accuracy: 0.6129\n",
            "\n",
            "Epoch 00035: val_accuracy did not improve from 0.83871\n",
            "Epoch 36/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.2561 - accuracy: 0.9460 - val_loss: 0.7285 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00036: val_accuracy did not improve from 0.83871\n",
            "Epoch 37/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.1184 - accuracy: 0.9540 - val_loss: 2.5601 - val_accuracy: 0.5806\n",
            "\n",
            "Epoch 00037: val_accuracy did not improve from 0.83871\n",
            "Epoch 38/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.1643 - accuracy: 0.9360 - val_loss: 2.2258 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00038: val_accuracy did not improve from 0.83871\n",
            "Epoch 39/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.2287 - accuracy: 0.9320 - val_loss: 2.0724 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00039: val_accuracy did not improve from 0.83871\n",
            "Epoch 40/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.1856 - accuracy: 0.9240 - val_loss: 2.7023 - val_accuracy: 0.5806\n",
            "\n",
            "Epoch 00040: val_accuracy did not improve from 0.83871\n",
            "Epoch 41/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.1732 - accuracy: 0.9240 - val_loss: 0.5419 - val_accuracy: 0.6774\n",
            "\n",
            "Epoch 00041: val_accuracy did not improve from 0.83871\n",
            "Epoch 42/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.1515 - accuracy: 0.9360 - val_loss: 1.9685 - val_accuracy: 0.5806\n",
            "\n",
            "Epoch 00042: val_accuracy did not improve from 0.83871\n",
            "Epoch 43/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.1895 - accuracy: 0.9220 - val_loss: 2.9146 - val_accuracy: 0.4839\n",
            "\n",
            "Epoch 00043: val_accuracy did not improve from 0.83871\n",
            "Epoch 44/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.1639 - accuracy: 0.9340 - val_loss: 0.4877 - val_accuracy: 0.7419\n",
            "\n",
            "Epoch 00044: val_accuracy did not improve from 0.83871\n",
            "Epoch 45/100\n",
            "32/32 [==============================] - 3s 100ms/step - loss: 0.1414 - accuracy: 0.9500 - val_loss: 1.4541 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00045: val_accuracy did not improve from 0.83871\n",
            "Epoch 46/100\n",
            "32/32 [==============================] - 3s 100ms/step - loss: 0.1756 - accuracy: 0.9360 - val_loss: 1.3275 - val_accuracy: 0.6774\n",
            "\n",
            "Epoch 00046: val_accuracy did not improve from 0.83871\n",
            "Epoch 47/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.1285 - accuracy: 0.9500 - val_loss: 2.9242 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00047: val_accuracy did not improve from 0.83871\n",
            "Epoch 48/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.1179 - accuracy: 0.9500 - val_loss: 1.7550 - val_accuracy: 0.6452\n",
            "\n",
            "Epoch 00048: val_accuracy did not improve from 0.83871\n",
            "Epoch 49/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.0814 - accuracy: 0.9700 - val_loss: 1.6525 - val_accuracy: 0.6452\n",
            "\n",
            "Epoch 00049: val_accuracy did not improve from 0.83871\n",
            "Epoch 50/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.1058 - accuracy: 0.9600 - val_loss: 2.0271 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00050: val_accuracy did not improve from 0.83871\n",
            "Epoch 51/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.2955 - accuracy: 0.8600 - val_loss: 2.2418 - val_accuracy: 0.6452\n",
            "\n",
            "Epoch 00051: val_accuracy did not improve from 0.83871\n",
            "Epoch 52/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.2219 - accuracy: 0.9360 - val_loss: 1.2948 - val_accuracy: 0.7419\n",
            "\n",
            "Epoch 00052: val_accuracy did not improve from 0.83871\n",
            "Epoch 53/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.1142 - accuracy: 0.9560 - val_loss: 3.7598 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00053: val_accuracy did not improve from 0.83871\n",
            "Epoch 54/100\n",
            "32/32 [==============================] - 3s 101ms/step - loss: 0.0983 - accuracy: 0.9640 - val_loss: 0.6297 - val_accuracy: 0.6452\n",
            "\n",
            "Epoch 00054: val_accuracy did not improve from 0.83871\n",
            "Epoch 55/100\n",
            "32/32 [==============================] - 3s 100ms/step - loss: 0.0654 - accuracy: 0.9800 - val_loss: 1.9179 - val_accuracy: 0.5806\n",
            "\n",
            "Epoch 00055: val_accuracy did not improve from 0.83871\n",
            "Epoch 56/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.1055 - accuracy: 0.9540 - val_loss: 1.2153 - val_accuracy: 0.6129\n",
            "\n",
            "Epoch 00056: val_accuracy did not improve from 0.83871\n",
            "Epoch 57/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.0687 - accuracy: 0.9760 - val_loss: 2.3014 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00057: val_accuracy did not improve from 0.83871\n",
            "Epoch 58/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.1073 - accuracy: 0.9600 - val_loss: 2.4411 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00058: val_accuracy did not improve from 0.83871\n",
            "Epoch 59/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.1317 - accuracy: 0.9540 - val_loss: 2.6982 - val_accuracy: 0.5806\n",
            "\n",
            "Epoch 00059: val_accuracy did not improve from 0.83871\n",
            "Epoch 60/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.2367 - accuracy: 0.9140 - val_loss: 2.1251 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00060: val_accuracy did not improve from 0.83871\n",
            "Epoch 61/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.0828 - accuracy: 0.9680 - val_loss: 4.1384 - val_accuracy: 0.5484\n",
            "\n",
            "Epoch 00061: val_accuracy did not improve from 0.83871\n",
            "Epoch 62/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.0751 - accuracy: 0.9660 - val_loss: 2.6787 - val_accuracy: 0.6774\n",
            "\n",
            "Epoch 00062: val_accuracy did not improve from 0.83871\n",
            "Epoch 63/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.0505 - accuracy: 0.9740 - val_loss: 3.2004 - val_accuracy: 0.5806\n",
            "\n",
            "Epoch 00063: val_accuracy did not improve from 0.83871\n",
            "Epoch 64/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.1332 - accuracy: 0.9460 - val_loss: 4.9895 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00064: val_accuracy did not improve from 0.83871\n",
            "Epoch 65/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.1917 - accuracy: 0.9300 - val_loss: 2.3154 - val_accuracy: 0.6774\n",
            "\n",
            "Epoch 00065: val_accuracy did not improve from 0.83871\n",
            "Epoch 66/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.0628 - accuracy: 0.9800 - val_loss: 2.0988 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00066: val_accuracy did not improve from 0.83871\n",
            "Epoch 67/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.0843 - accuracy: 0.9600 - val_loss: 2.6636 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00067: val_accuracy did not improve from 0.83871\n",
            "Epoch 68/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.1184 - accuracy: 0.9560 - val_loss: 1.9697 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00068: val_accuracy did not improve from 0.83871\n",
            "Epoch 69/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.0773 - accuracy: 0.9680 - val_loss: 3.5366 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00069: val_accuracy did not improve from 0.83871\n",
            "Epoch 70/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.1504 - accuracy: 0.9500 - val_loss: 1.7486 - val_accuracy: 0.5484\n",
            "\n",
            "Epoch 00070: val_accuracy did not improve from 0.83871\n",
            "Epoch 71/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.1057 - accuracy: 0.9700 - val_loss: 1.5442 - val_accuracy: 0.6129\n",
            "\n",
            "Epoch 00071: val_accuracy did not improve from 0.83871\n",
            "Epoch 72/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.0691 - accuracy: 0.9780 - val_loss: 1.5936 - val_accuracy: 0.4839\n",
            "\n",
            "Epoch 00072: val_accuracy did not improve from 0.83871\n",
            "Epoch 73/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.1164 - accuracy: 0.9600 - val_loss: 1.6644 - val_accuracy: 0.6452\n",
            "\n",
            "Epoch 00073: val_accuracy did not improve from 0.83871\n",
            "Epoch 74/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.1701 - accuracy: 0.9400 - val_loss: 0.3912 - val_accuracy: 0.8065\n",
            "\n",
            "Epoch 00074: val_accuracy did not improve from 0.83871\n",
            "Epoch 75/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.2737 - accuracy: 0.8920 - val_loss: 1.2895 - val_accuracy: 0.7097\n",
            "\n",
            "Epoch 00075: val_accuracy did not improve from 0.83871\n",
            "Epoch 76/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.1543 - accuracy: 0.9560 - val_loss: 1.1823 - val_accuracy: 0.7419\n",
            "\n",
            "Epoch 00076: val_accuracy did not improve from 0.83871\n",
            "Epoch 77/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.2691 - accuracy: 0.9080 - val_loss: 0.8519 - val_accuracy: 0.5806\n",
            "\n",
            "Epoch 00077: val_accuracy did not improve from 0.83871\n",
            "Epoch 78/100\n",
            "32/32 [==============================] - 3s 98ms/step - loss: 0.1639 - accuracy: 0.9680 - val_loss: 0.5629 - val_accuracy: 0.6452\n",
            "\n",
            "Epoch 00078: val_accuracy did not improve from 0.83871\n",
            "Epoch 79/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.1318 - accuracy: 0.9720 - val_loss: 0.9319 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00079: val_accuracy did not improve from 0.83871\n",
            "Epoch 80/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.1094 - accuracy: 0.9780 - val_loss: 0.7382 - val_accuracy: 0.5484\n",
            "\n",
            "Epoch 00080: val_accuracy did not improve from 0.83871\n",
            "Epoch 81/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.1027 - accuracy: 0.9640 - val_loss: 3.5693 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00081: val_accuracy did not improve from 0.83871\n",
            "Epoch 82/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.1930 - accuracy: 0.9220 - val_loss: 1.4823 - val_accuracy: 0.6774\n",
            "\n",
            "Epoch 00082: val_accuracy did not improve from 0.83871\n",
            "Epoch 83/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.1381 - accuracy: 0.9560 - val_loss: 3.1480 - val_accuracy: 0.6774\n",
            "\n",
            "Epoch 00083: val_accuracy did not improve from 0.83871\n",
            "Epoch 84/100\n",
            "32/32 [==============================] - 3s 99ms/step - loss: 0.0669 - accuracy: 0.9660 - val_loss: 2.3622 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00084: val_accuracy did not improve from 0.83871\n",
            "Epoch 85/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.0574 - accuracy: 0.9800 - val_loss: 3.1606 - val_accuracy: 0.4839\n",
            "\n",
            "Epoch 00085: val_accuracy did not improve from 0.83871\n",
            "Epoch 86/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.1077 - accuracy: 0.9620 - val_loss: 2.8606 - val_accuracy: 0.6774\n",
            "\n",
            "Epoch 00086: val_accuracy did not improve from 0.83871\n",
            "Epoch 87/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.0867 - accuracy: 0.9740 - val_loss: 1.9991 - val_accuracy: 0.5161\n",
            "\n",
            "Epoch 00087: val_accuracy did not improve from 0.83871\n",
            "Epoch 88/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.0879 - accuracy: 0.9620 - val_loss: 2.9131 - val_accuracy: 0.4839\n",
            "\n",
            "Epoch 00088: val_accuracy did not improve from 0.83871\n",
            "Epoch 89/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.0878 - accuracy: 0.9740 - val_loss: 2.7994 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00089: val_accuracy did not improve from 0.83871\n",
            "Epoch 90/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.1937 - accuracy: 0.9280 - val_loss: 3.6337 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00090: val_accuracy did not improve from 0.83871\n",
            "Epoch 91/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.2312 - accuracy: 0.9080 - val_loss: 2.0470 - val_accuracy: 0.4839\n",
            "\n",
            "Epoch 00091: val_accuracy did not improve from 0.83871\n",
            "Epoch 92/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.0956 - accuracy: 0.9660 - val_loss: 4.3960 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00092: val_accuracy did not improve from 0.83871\n",
            "Epoch 93/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.1389 - accuracy: 0.9500 - val_loss: 2.1292 - val_accuracy: 0.5161\n",
            "\n",
            "Epoch 00093: val_accuracy did not improve from 0.83871\n",
            "Epoch 94/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.0896 - accuracy: 0.9740 - val_loss: 2.7775 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00094: val_accuracy did not improve from 0.83871\n",
            "Epoch 95/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.0409 - accuracy: 0.9840 - val_loss: 3.1929 - val_accuracy: 0.4839\n",
            "\n",
            "Epoch 00095: val_accuracy did not improve from 0.83871\n",
            "Epoch 96/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.1601 - accuracy: 0.9440 - val_loss: 2.5233 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00096: val_accuracy did not improve from 0.83871\n",
            "Epoch 97/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.1110 - accuracy: 0.9520 - val_loss: 0.6741 - val_accuracy: 0.7419\n",
            "\n",
            "Epoch 00097: val_accuracy did not improve from 0.83871\n",
            "Epoch 98/100\n",
            "32/32 [==============================] - 3s 97ms/step - loss: 0.1901 - accuracy: 0.9340 - val_loss: 2.2150 - val_accuracy: 0.4839\n",
            "\n",
            "Epoch 00098: val_accuracy did not improve from 0.83871\n",
            "Epoch 99/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.1355 - accuracy: 0.9580 - val_loss: 1.7726 - val_accuracy: 0.5161\n",
            "\n",
            "Epoch 00099: val_accuracy did not improve from 0.83871\n",
            "Epoch 100/100\n",
            "32/32 [==============================] - 3s 96ms/step - loss: 0.0995 - accuracy: 0.9620 - val_loss: 3.1933 - val_accuracy: 0.4516\n",
            "\n",
            "Epoch 00100: val_accuracy did not improve from 0.83871\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "K59QtXezLV7Q",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 545
        },
        "outputId": "d0fd5a2e-24b2-4474-90eb-c7cb614d43f2"
      },
      "source": [
        "def plot_history(history): \n",
        "    acc = history.history['accuracy'] \n",
        "    val_acc = history.history['val_accuracy'] \n",
        "    loss = history.history['loss'] \n",
        "    val_loss = history.history['val_loss'] \n",
        "    epochs = range(1, len(acc) + 1) \n",
        "    \n",
        "    plt.figure() \n",
        "    plt.title('Training and validation accuracy') \n",
        "    plt.plot(epochs, acc, 'bo', label='Training acc')\n",
        "    plt.plot(epochs, val_acc, 'b', color='red', \\\n",
        "      label='Validation acc') \n",
        "    plt.legend() \n",
        "    plt.show() \n",
        "   \n",
        "    plt.figure() \n",
        "    plt.title('Training and validation loss')\n",
        "    plt.plot(epochs, loss, 'bo', label='Training loss')\n",
        "    plt.plot(epochs, val_loss, 'b', color='red', \\\n",
        "      label='Validation loss') \n",
        "    plt.legend() \n",
        "    plt.show() \n",
        "    return acc, val_acc, loss, val_loss \n",
        "\n",
        "acc, val_acc, loss, val_loss = plot_history(history)"
      ],
      "execution_count": 73,
      "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": "iVBORw0KGgoAAAANSUhEUgAAAWoAAAEICAYAAAB25L6yAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nO29eZgU5bn+fz/MDLPDsIsMCoqCiOy4oUaNOXE3cYkLUTkuqPHEaHI0RuNycsJPfwknUa9EE9RoVAwaTTxuxETF/biAIouCCzuCwCgzA8zAMPN+/3j6od6uqaquqq7qru55P9c1Vy/TXfV2LXfddb8bKaVgMBgMhuTSLd8FMBgMBoM3RqgNBoMh4RihNhgMhoRjhNpgMBgSjhFqg8FgSDhGqA0GgyHhGKHuYhDRHCK6MOrP5hMiWklEx8WwXEVEw1LP/0BEN/n5bIj1TCGif4Ytp8dyjyaitVEv15B7SvNdAENmiGir9rIKwA4A7anXlymlZvldllLqhDg+W+wopS6PYjlENATACgBlSqldqWXPAuB7Hxq6HkaoCwClVI08J6KVAC5RSr1o/xwRlcrJbzAYigcTfRQwcmtLRD8log0AHiCiXkT0LBFtIqKvU8/rte+8QkSXpJ5PJaI3iGhG6rMriOiEkJ8dSkSvEVEzEb1IRL8nokdcyu2njP9NRG+mlvdPIuqr/f98IlpFRA1EdKPH9jmEiDYQUYn23neJaGHq+cFE9H9EtIWI1hPR74iou8uyHiSiX2qvr0195wsiusj22ZOI6AMiaiKiNUR0q/bv11KPW4hoKxEdJttW+/7hRPQeETWmHg/3u228IKIDUt/fQkRLiOhU7X8nEtFHqWWuI6L/TL3fN7V/thDRV0T0OhEZ3cgxZoMXPnsA6A1gbwDTwPv0gdTrvQC0APidx/cPAbAMQF8AvwJwPxFRiM8+CuBdAH0A3ArgfI91+injeQD+HUB/AN0BiHCMBHBPavl7ptZXDweUUu8A2AbgWNtyH009bwdwTer3HAbgmwB+4FFupMpwfKo83wKwHwB7Pr4NwAUA6gCcBOAKIvpO6n9HpR7rlFI1Sqn/sy27N4DnANyV+m2/AfAcEfWx/YZO2yZDmcsAPAPgn6nv/RDALCIanvrI/eAYrRbAKAAvp97/CYC1APoBGADgBgBm3IkcY4S68OkAcItSaodSqkUp1aCUelIptV0p1QxgOoBveHx/lVLqXqVUO4A/AxgIPiF9f5aI9gIwCcDNSqmdSqk3ADzttkKfZXxAKfWJUqoFwOMAxqbePxPAs0qp15RSOwDclNoGbvwFwLkAQES1AE5MvQel1Hyl1NtKqV1KqZUA/uhQDie+lyrfYqXUNvCFSf99ryilFimlOpRSC1Pr87NcgIX9U6XUw6ly/QXAUgCnaJ9x2zZeHAqgBsDtqX30MoBnkdo2ANoAjCSiHkqpr5VS72vvDwSwt1KqTSn1ujIDBOUcI9SFzyalVKu8IKIqIvpjKhpoAt9q1+m3/zY2yBOl1PbU05qAn90TwFfaewCwxq3APsu4QXu+XSvTnvqyU0LZ4LYusHs+nYjKAZwO4H2l1KpUOfZP3dZvSJXj/wO760yklQHAKtvvO4SI5qainUYAl/tcrix7le29VQAGaa/dtk3GMiul9IuavtwzwBexVUT0KhEdlnr/1wA+A/BPIlpORNf7+xmGKDFCXfjY3c1PAAwHcIhSqgesW223OCMK1gPoTURV2nuDPT6fTRnX68tOrbOP24eVUh+BBekEpMceAEcoSwHslyrHDWHKAI5vdB4F31EMVkr1BPAHbbmZ3OgX4EhIZy8A63yUK9NyB9vy5d3LVUq9p5Q6DRyLPAV26lBKNSulfqKU2gfAqQB+TETfzLIshoAYoS4+asGZ75ZU3nlL3CtMOdR5AG4lou4pN3aKx1eyKeMTAE4moiNSFX+/QObj+FEAPwJfEP5qK0cTgK1ENALAFT7L8DiAqUQ0MnWhsJe/FnyH0UpEB4MvEMImcFSzj8uynwewPxGdR0SlRHQ2gJHgmCIb3gG77+uIqIyIjgbvo9mpfTaFiHoqpdrA26QDAIjoZCIalqqLaATn+l5RkyEGjFAXH3cAqASwGcDbAP6Ro/VOAVfINQD4JYDHwO29nQhdRqXUEgBXgsV3PYCvwZVdXkhG/LJSarP2/n+CRbQZwL2pMvspw5zUb3gZHAu8bPvIDwD8goiaAdyMlDtNfXc7OJN/M9WS4lDbshsAnAy+62gAcB2Ak23lDoxSaidYmE8Ab/e7AVyglFqa+sj5AFamIqDLwfsT4MrSFwFsBfB/AO5WSs3NpiyG4JCpFzDEARE9BmCpUip2R28wFDvGURsigYgmEdG+RNQt1XztNHDWaTAYssT0TDRExR4A/gau2FsL4Aql1Af5LZLBUByY6MNgMBgSjok+DAaDIeHEEn307dtXDRkyJI5FGwwGQ1Eyf/78zUqpfk7/i0WohwwZgnnz5sWxaIPBYChKiMjeI3U3JvowGAyGhGOE2mAwGBKOEWqDwWBIODlrR93W1oa1a9eitbU184cNeaWiogL19fUoKyvLd1EMBgNyKNRr165FbW0thgwZAvdx6Q35RimFhoYGrF27FkOHDs13cQwGA3wKNfE8fc3gkbN2KaUmBl1Ra2urEekCgIjQp08fbNq0Kd9FMRgMKYI46mOyHcHLiHRhYPaTwZAsTGWiwZAk2tqAP/0J6DBDPhss/Aq1Ak/FM5+Ipjl9gIimEdE8IpqXtNvmhoYGjB07FmPHjsUee+yBQYMG7X69c+dOz+/OmzcPV111VcZ1HH744Rk/44dXXnkFJ598ciTLMhQgc+cCF18MvP12vktiSBB+hfoIpdR48KDjVxLRUfYPKKVmKqUmKqUm9uvn2AsyELNmAUOGAN268eOsWeGX1adPHyxYsAALFizA5ZdfjmuuuWb36+7du2PXrl2u3504cSLuuuuujOt46623whfQYBCamvhx+3bvzxm6FL6EWikl86ptBPB3AAfHWahZs4Bp04BVqwCl+HHatOzE2s7UqVNx+eWX45BDDsF1112Hd999F4cddhjGjRuHww8/HMuWLQOQ7nBvvfVWXHTRRTj66KOxzz77pAl4TU3N7s8fffTROPPMMzFixAhMmTIFMkLh888/jxEjRmDChAm46qqrMjrnr776Ct/5zncwevRoHHrooVi4cCEA4NVXX919RzBu3Dg0Nzdj/fr1OOqoozB27FiMGjUKr7/+enQby5A7tm3jR9OM1aCRsTKRiKoBdFNKNaee/xt4nrrYuPHGzoZi+3Z+f8oU5++EYe3atXjrrbdQUlKCpqYmvP766ygtLcWLL76IG264AU8++WSn7yxduhRz585Fc3Mzhg8fjiuuuKJTe+MPPvgAS5YswZ577onJkyfjzTffxMSJE3HZZZfhtddew9ChQ3HuuedmLN8tt9yCcePG4amnnsLLL7+MCy64AAsWLMCMGTPw+9//HpMnT8bWrVtRUVGBmTNn4tvf/jZuvPFGtLe3Y7txZIWJCPUOt1nMDF0RP60+BgD4e6olQCmAR5VSsc7Dt3p1sPfDctZZZ6GkpAQA0NjYiAsvvBCffvopiAhtbW2O3znppJNQXl6O8vJy9O/fH19++SXq6+vTPnPwwQfvfm/s2LFYuXIlampqsM8+++xum3zuuedi5syZnuV74403dl8sjj32WDQ0NKCpqQmTJ0/Gj3/8Y0yZMgWnn3466uvrMWnSJFx00UVoa2vDd77zHYwdOzarbWPIE0aoDQ5kjD6UUsuVUmNSfwcqpabHXai99gr2fliqq6t3P7/ppptwzDHHYPHixXjmmWdce1CWl5fvfl5SUuKYb/v5TDZcf/31uO+++9DS0oLJkydj6dKlOOqoo/Daa69h0KBBmDp1Kh566KFI12nIESb6MDiQyOZ506cDVVXp71VV8ftx0djYiEGDBgEAHnzwwciXP3z4cCxfvhwrV64EADz2WOYJr4888kjMSgXzr7zyCvr27YsePXrg888/x0EHHYSf/vSnmDRpEpYuXYpVq1ZhwIABuPTSS3HJJZfg/fffj/w3GHKAcdQGBxIp1FOmADNnAnvvDRDx48yZ0ebTdq677jr87Gc/w7hx4yJ3wABQWVmJu+++G8cffzwmTJiA2tpa9OzZ0/M7t956K+bPn4/Ro0fj+uuvx5///GcAwB133IFRo0Zh9OjRKCsrwwknnIBXXnkFY8aMwbhx4/DYY4/hRz/6UeS/wZADjFAbHIhlzsSJEycq+8QBH3/8MQ444IDI11VIbN26FTU1NVBK4corr8R+++2Ha665Jt/FcsTsrzxx4YXAQw8Bt90GXH99vktjyCFENN9teI5EOupi5d5778XYsWNx4IEHorGxEZdddlm+i2RIGsZRGxzI2eh5BuCaa65JrIM2JARTmWhwwDhqgyFJGEdtcMAItcGQJIxQGxwwQm0wJAkTfRgcMEJtMCQJ46gNDnQZoT7mmGPwwgsvpL13xx134IorrnD9ztFHHw1pZnjiiSdiy5YtnT5z6623YsaMGZ7rfuqpp/DRRx/tfn3zzTfjxRdfDFJ8R8yQqEWIEWqDA11GqM8991zMnj077b3Zs2f7GhwJ4JHv6urqQq3bLtS/+MUvcNxxx4ValqHIMdGHwYEuI9Rnnnkmnnvuud0TBaxcuRJffPEFjjzySFxxxRWYOHEiDjzwQNxyyy2O3x8yZAg2b+aZyKZPn479998fRxxxxO7hUAFuJz1p0iSMGTMGZ5xxBrZv34633noLTz/9NK699lqMHTsWn3/+OaZOnYonnngCAPDSSy9h3LhxOOigg3DRRRdhR8pJDRkyBLfccgvGjx+Pgw46CEuXLvX8fWZI1CJg1y5AJrIwjtqgkZ921FdfDSxYEO0yx44F7rjD9d+9e/fGwQcfjDlz5uC0007D7Nmz8b3vfQ9EhOnTp6N3795ob2/HN7/5TSxcuBCjR492XM78+fMxe/ZsLFiwALt27cL48eMxYcIEAMDpp5+OSy+9FADw85//HPfffz9++MMf4tRTT8XJJ5+MM888M21Zra2tmDp1Kl566SXsv//+uOCCC3DPPffg6quvBgD07dsX77//Pu6++27MmDED9913n+vvM0OiFgHipgEj1IY0uoyjBtLjDz32ePzxxzF+/HiMGzcOS5YsSYsp7Lz++uv47ne/i6qqKvTo0QOnnnrq7v8tXrwYRx55JA466CDMmjULS5Ys8SzPsmXLMHToUOy///4AgAsvvBCvvfba7v+ffvrpAIAJEybsHszJjTfeeAPnn38+AOchUe+66y5s2bIFpaWlmDRpEh544AHceuutWLRoEWpraz2XbcgRW7daz030YdDIj6P2cL5xctppp+Gaa67B+++/j+3bt2PChAlYsWIFZsyYgffeew+9evXC1KlTXYc4zcTUqVPx1FNPYcyYMXjwwQfxyiuvZFVeGS41m6FSr7/+epx00kl4/vnnMXnyZLzwwgu7h0R97rnnMHXqVPz4xz/GBRdckFVZDRFgHLXBhS7lqGtqanDMMcfgoosu2u2mm5qaUF1djZ49e+LLL7/EnDlzPJdx1FFH4amnnkJLSwuam5vxzDPP7P5fc3MzBg4ciLa2tt3DkwJAbW0tmpubOy1r+PDhWLlyJT777DMAwMMPP4xvfOMboX6bGRK1CBChLiszjtqQRpcb6+Pcc8/Fd7/73d0RiAwNOmLECAwePBiTJ0/2/P748eNx9tlnY8yYMejfvz8mTZq0+3///d//jUMOOQT9+vXDIYccsluczznnHFx66aW46667dlciAkBFRQUeeOABnHXWWdi1axcmTZqEyy+/PNTvkvkcR48ejaqqqrQhUefOnYtu3brhwAMPxAknnIDZs2fj17/+NcrKylBTU2MmGUgKItR9+hhHbUjDDHNqcMTsrzzwwgvA8ccDI0cCX38NfPFFvktUPGzdCvz7v3PsmpogJGmYYU4NhkJAd9Qm+oiWjz4CnngCeOutfJckFEaoDYakIELdu7eJPqJGLnwtLfktR0hyKtRxxCyG6DH7KU8YoY4PEeoCvVPJmVBXVFSgoaHBiEDCUUqhoaEBFRUV+S5K10OPPtrbuaeiIRrESReoo85Zq4/6+nqsXbsWmzZtytUqDSGpqKhAfX19vovR9RCh7tWLH3fsAEq7XMOseCjw6CNnR0FZWRmGDh2aq9UZDIXHtm1A9+5ATQ2/3rEDqK7Ob5mKBRN9GAyGSNi2jYU51SO1UEUlkRR49GGE2mBICnahNhWK0VHg0YcRaoMhKYhQS0WucdTRIQJdoNvUCLXBkCu++AJ46in3/xtHHR/GURsMBl/cdx9wxhlAW5vz/+2O2gh1dBihNhgMvmhuBjo6+NEJU5kYHyb6MBgMvpCZdJqanP9voo/46CqOmohKiOgDIno2zgIZDEWLCHUmR22ij+jpKkIN4EcAPo6rIAZD0RPUURfobXoi6QrRBxHVAzgJgPvsqgaDwRsRC78ZtXHU0dFFHPUdAK4D0OH2ASKaRkTziGieGc/DYHDAy1G3t7OYmHbU8VDsPROJ6GQAG5VS870+p5SaqZSaqJSa2K9fv8gKaDAUDV4ZtfzPOOp46AJjfUwGcCoRrQQwG8CxRPRIrKUyGIoRL0ctI+cZoY6HYo8+lFI/U0rVK6WGADgHwMtKqe/HXjKDodjwyqhFqGtqTPQRB8UefRgMhojw66i7d+fnxlFHh1z02tq4PqDACCTUSqlXlFInx1UYg6Go8cqodaEm4vjDCHV06E66AO9UjKM2GHKFX0cNsFAXoKAkltZWvgACBRl/GKE2GHJBR4clvJkcNWAcddS0tgI9e/JzI9QGQwHz5pvxncS6O/bjqCsqjFBHhVK8X3v35tcFeKdihNpgAICGBuDII4HZs+NZvsQegH9HXYCCkkh27uRHmTTYOGqDoUBpamLn1dgYz/J1ofabURtHHQ1ywTNCbTAUOHIyi/uKGhHqujp/jrqiwjjqqBBhFqEuwO1qhNpgAHIn1AMGuDvq0lKrDbVx1NFhHLWh4Fi3DjCDZnVGTua4xFHEYcAAXof9giAj5wlGqMPR3g68/Xb6e3ZHbYS6QFm0CJg0Cfj883yXJH7OOw/44Q/zXYrkkStHvcce/GiPP+xCbaKPcDz9NHDYYennst1RF+B2NUK9dStw1lnAvHnA4sX5Lk38NDQAX36Z71Ikj1xGH0BmoTaOOhwrV/Kjftco+1aa5xlHXYBceSWwbBk/L8AdGJgdO6yKK4NF3NGH3VHbc2onR10MQq0UcPXVwDvv5GZ9Gzbw49at1nsm+ihwHnwQeOghYOpUfl2AOzAwO3caoXYiiY46zC36G29YrjIJ7NwJ3Hkn8GyOplqVu0V9+8p2rKtLf11AdF2hXrGC3fTRRwPTp/N7Rqi7LnELtV6ZCGR21GGjj7PPBm6/PVwZ40B+Q67OLS9HbaKPAuSxx9jlPPggjwEMFOQODMzOnekHsYHJVfTh11GHjT6++grYsiVcGeNAfoPe4SdOnIRa9m1tLVBSUpDneWm+C5A35swBxo4F9t6bx6gFCnIHBmbnTmDXrnyXInnkKvro358f/TjqoLfou3bxd5J0x5RrR+0VfVRUFGxrmq7pqBsbeQCeE07g12VlBXulDcyOHXygFuDg6bGSC6GurLRGcPPb6kMp/+sQF5lEoc6Fo25vBzZu5OdO0UdFBe+DAjzPu6ZQv/gi71QRaqBgd2AglLLuHnJ1K1oo5CKjrqzk22/AX6sPfX/5IclCnYtza/NmHk4WcHbUlZUFe553TaGeM4edzWGHWe8V6A4MhH7SJ+lkTgK5yKirqvjOraoqXUiU4v/bHXXQ8sgyk7Rv5cKXC2Og9w9wyqgl+ijA87zrCbVSwD/+AXzrWzy2gtAVhFp3i0k6mZNALqKPqip+3qNHuqNuaeHj0u6ogWBC3dUdtVQkAp2jj9JS/qusNBl1QbBoEY93occeQNcQav2kT9LJnARyKdS1temO2j5yHmA56iCikkRHncuMWhx1jx6dow+58BXoed71hHrOHH48/vj09wt0BwZCFyHTRC+dXEUfQGdH7SXUhR595MNRDxvW2VEboS4w5swBxowB9twz/f0C3YGBMNGHO7mqTAT8Oepsoo/t261KtXyTa6GuquJu+nZHLdveNM8rAJqauFme3U0DRqi7OvnMqKOOPoDkHMu5jj722IMvhPbKROOoC4hXXuFOAUaojVDbSWpGHcZR68vMN7l21AMGcE9j+0XLCHUBIY3h99238/8KdAcGwlQmuiPbJikZdZjow0n8843uqIN03gnDhg3sqGtqOjtqE30UEHoPJTtdQaiNo3YnVz0Tgc6OWp5nG30k2VErFd+2FezRh1wYgjjqG27gEQjD8JvfAN/9brjvZqBrCbXe8N2OEequTS4qE3VHrU/HJcOSDh5sfT6bVh9AcvavXn6/OfXHHwMffBBsPW1t3DNRog/pRASkO2qv87yhAbjtNuDhh4OtW1i0CHj//XDfzUDXEmrZQbLTdIxQd21EqNvaom8xIaKhZ9SAJayffAIMHGi9D1hmImxlYlL2ry7U9vPr4ot5rGo7V18NnHlmsPXIjC7iqAHrDkOvTJTowymG+fBDflyzJti6haYmvgjHQNcS6tZW7sJb6jBoYFcQav2kMe2o09EFMcj4GkGWrTtqwMqpP/0U2G+/9O+ErUwk4udJFGq7o37hBe4lbGftWmD58mATIEgbanHUgHXhskcfbjGMCPXq1f7Xq2OEOiL0WyA7lZV8gnqNKrdsWXJOgDAYR+2OLtRRxx8iUF6Oev/9078TtjKxTx9+npT96+Wot24F1q/v/B15b+5c/+sRoXZz1Hr04VQWAFiwgB+No84z+pXVjtcOBLhZ34QJwG9/G0/ZcoERandaW4FuqdMh6pYfItRyjOmO+uuv+bbdLtRhKxNlTsak7F83oVbKWah37OBtAgAvv+x/PdJ9XFp9AO7Rh7xnRxx1UxMPhRyUfAo1EVUQ0btE9CERLSGi/4qlJLkgk6MG3IV6wwY++PVp6AsNEeoePZJzIieF1lbrJIvaUcsx5eSoP/2Un7sJdVBHLTPIJGX/ukUfO3bw3eumTekTWYgzLitjR+23SV+m6COTo965E/joI2DIEH4dxlXn2VHvAHCsUmoMgLEAjieiQ2MpTdxk46jXruXHdeuiL1euEAHq3Ts5J3ISUIqFIy6htkcfuqP+5BN+HkX0UUiOWtyuUunDk4rDPv54PtfkQpaJDRt4u+pjfjs5arfz/OOPOfo85RR+HVao9QrhCMko1IqRmqey1F/MLddjIhtHLUL9xRfRlytXyEnjJtRxd0hIKrJdZPaVXGbUn37Kkcs++6R/p7SUKwaDtvro14+/l0Sh1h21Xj49/hBnfN55/Og3/vjyS+tuQnfU7e0swHahtm9XiT1OPpkfg1YodnTwhSGfGTURlRDRAgAbAfxLKfWOw2emEdE8Ipq3SZrKJI0oHHUhC7UIUK9enU/kn/4UOPbY3JcpCchJKydZXBm1m6Pee28r6hCIgs1ELplvbS13nEmSUEtLFCdHDaQLtTw/6iigvt5/haL0SgTSHbU+uwtgnf/283zBAv7MN77BF86gjlp+Tz6FWinVrpQaC6AewMFENMrhMzOVUhOVUhP79esXdTmjIQpH/fXXhduMTxdqe/O8RYu4R1bUTdMKAbtQx5VRyzGmOz6nFh9CkO7O0kW7piZ5Ql1Xx891R60ff7r5Wb+ehb1/f+CYY1io/bRrl16JQPr2tXdyczvPP/wQGDWKL46DBgV31NLUMgmtPpRSWwDMBeAwqlEBEIWjBpybFBUCXhn1V19xpc5nn+W+XDpPPBGsC++f/wwceWR265STOVfRR0kJi6k4ajehDuKopeIsiY66Vy9+7sdRb9jA8U1pKd/hbdoELFnSebnLlwN/+IMV18mATABvt7IyXod92Ain81wpdtRjxvDrwYODO+p8CzUR9SOiutTzSgDfArA0ltLETbaOuqSEnxdqhaLchvbs6SzUgPNJkUuuu47HTPDLvHnZ3wnkOvoAWFA//ZTFxN7ZRQgi1CJ8heio7dHHwIH8/Jhj+NEp/vjTn4ArrgAefJD335YtlqMGrIGZ3KIP/U5l3To+/seO5deFKNQABgKYS0QLAbwHzqifjaU0cZOtox49mp8Xak69cyfQvTufyK2t6Z17pO3qRx/lp2zCxo3huk1v2RJ+nXFHH05C3aMHX2SAaKKPpDrqnTut7erkqCsr3YV67725ktWpQlFGwvzRj6ztKI4asIY69RN9SEWiOOq99mKhDjKUQL6FWim1UCk1Tik1Wik1Sin1i1hKkgvCOuqODr7qHnwwvy4GoQYsAenosBx1PoW6pYUFJsyIcXKhCUOuow+ABVUEKoroI8mOury88xANUr5hwzpHH7oznjwZmD+/83I3bmRB7+gAvv99fk//noygZ68fcDrPpUeiGLHBg/kYCNIoQoQ6X83zioqwjloa5Y8axd8vFqGWk6W52XIP+RTqzZv5MV9CnavKRH1d3buzg3OioqI4Murycr5IOUUf++1nCXVHB1cKiqMGeMq8jRs7Nx3dtAkYORL4n/8BVq3i9+zRh5Ojdoo+PvyQnbvsE9kfQSoU8+2oi4qwjloqEgcP5gOn2IRa3HT//jyeid5TLJeIgwkTfUQp1HFk1OXlVhd1wHJe++5r1X3YKS8v/OjDzVGLUA8bxuLc0cEX6l270oV6wAA+bu3R1saNXOk4bRrwb//G7+nfsztqr+hDr0gErOFmg+TUsv2NUEeA3kPJjh+hrq9noS7kysTy8s5jIYjIHXEEnxT56iYfRqgLwVHrQ5wKsi632AMorujDyVFXVvI5tWsXi7Q+sJIgubPeexFgoe7fnyvHZ80CHn2Um9UJmSoT5TxvaeGWThJ7ANk5ahN9ZEAp7/BfRsZzE2q3hvCAJdSDBhW3o548mR/zFX9kI9RRVCbGmVHbhVpOaC+hLqbow8lR19RYLnj9eisC0Z1x//78qAv1jh0sjNJfo29f4Nxz09frFn0QpVfSrl7N2jFsmPXdPn34M0EcdVMT72OnIZQjoHiE+l7v+qsAACAASURBVPbbgYkT3f9vv7LakR3oJtSlpXzQDBrEQl2I3a0zCfXhh/NjXE30Wlq8t5sIdZAORYUSfdiPO7+O2u9Fy8lRJ+EY9XLU1dWZhdrJUctxIiLuhFv0Ic/lfcm39XoCouBN9GIckAkoJqH+5BNvJ+g1X6LgNnnA2rUs0N26saPeti19No1CwU2oReT22otHD4vDUW/bxttw9mz3zxRLZeIvfwmcfbb1Wp+GS/DjqIN2eJGOHjU1fHcZ10S9QfBq9WF31F7RhzTHAyyh9uoBbXfU+oVSL4vEG3vvnf79vfYKHn0YofZBSwsfFG4neSZHLf9zE+r6en6+5578WIjxh5w0bo66Vy+uSY9DqJcvZzFdscL9M0Gjj/Z2a38lpXmeUtxj7plnLEfrFH0MHMh3aSNGuC/LrR21UsDvfpcuJBIlAJ33bz7xctRO0YdEN0KfPmyQdEctop3JUbe2Whdy3aBVVqZHH2LAdMI46pjyaaCYhFoOArcBv7N11MUg1F7RR0UF//6RI4GlS71nugmD3GJ6xRq6UPu5bdd7t0Uh1HKiZeNEP/iAK5tbWixxcRLq738fWLjQW2zcHPX69cAPfwg88ID1XnOzVf6kCHVHB9cNeWXUlZV8gRSh1mMPgFvE9OvnHH1kctT6Z3WDZo8+9tyT70R09tqLy+P3om0ctU9kw7tVKoV11Eo5C3Uhtvzwij569+bnI0eyOCxfHu26Rai9ZqLWOxj4OUGiFuqqKhaGbBz1s1qnXWk94yTU3bsDBxzgvSw3oZbl6uPPJNFRy3b0EmqAxVmiDz32EAYMCOeoAStO00cntEcf9tgDYEetlH9DZoTaJ5mEOqyj/vprPpGL3VGLUB94ID9GHX8EcdSAv/hDryfIVqjLy62hRbMR6meescRGLnb6DCNBcIs+RKj1W/MkOmq5yHhFH4Al1E6OGugs1Js2sQOWqMoJ3VHLvhX06GPVKucOR0Gb6DU3G6H2hQiAW/QR1lHrTfMAPgB69ChMoZa8sKzMGl0MYKGWEc7E5QUR6l27gMce824eKTNKZxJq6fzhR6il/HV12TfPkwt49+7hhfqLL3jcicsuY2EQoXZy1H6Qi4Y9BnJy1EkXaidHLeXMJNT9+3d21DJBghu6UNvPeYk+2tt5G7o5asB/Tm0ctU/kap2No66qchdqcdRA4balFkcN8IHs5Khra/kgDdJEb+5c4JxzgBdecP9MJke9axe7YrkgBhHqwYOzd9S6UIfNqJ9/nh/POIOPF6/oww9SJvuFo1CiDydHLRcdafUBsDivXs3veUUf8t1Nm7xjDyA9+rCf83LR2LCBM3QnRx1EqJUyQu2buDLqYhVqvVOEnlEDHH8EcdSyzb3Gkc4k1A0N/CjbOUj0UV/Pd1JhK0CjctTPPMPubNQoHjsiCkct5dMRoW5stLZBIThqqVzs6Ogs1LLv3KKPlhbrwiyO2gvdUbsJtcQaTkJdU8N3an6ijx07+HcZofZBXBn12rXcfEe/0nt1I58yBbjkEn9lzjVuQq07aoDHn5Cowg+ynDffdP6/3gLCTagln5aTxk+nFzlx5TtusVcmdKEOm1G3tAD/+hdPjkrE2/Dzz9ltZSvUdof/+edWPitGohAcNcDbQu5+daEW3IQasCoRgzjqLVuco4/WVvc21EKvXlbXcC9i7j4OFKNQx5FR77FHevMdcdROTchef93/PG+5xkmod+zgE0cyaoBPIK/WGXZEEN5913kAf92VZBJqueUMGn0A4eOPKKKPl1/m3yaTo+6zD99ab9nCx0nYykQgvTyNjXz3cdRR/FqEuhAcNZDujJ2E2i36AKyLfRBHDbg7aqdeiTp+u+LHPHIeUExCHUVG7SbUeuwBsFC3tVm368LOnfz5lSuD9a7LFVKZCFgHoYib7qgrK/mzfqME2fYtLdyO2I6cELW10Qq13PZHLdRhHPXTT7M4HH00v5ZZxSXrjyr6kNjjG9/gxzVruLxtbYXjqO1CrXc28XLUX35pCb1fRw10vkjq0UddnbvAGqGOGBlwCYgno7YLtVR42XNqGeCloyP/cw864eSopVeiLtRyUvm92OgHs1P8IUI9YoS7UEt712wcddiWH9lGH2vWAA89xJWIIq777suPixbxY1TRhwi17qj1AZkA6xhPklA7OWq91QfAd636cSjoAzP56eyiLxtwdtStre5N8wR7k0I3Yh7iFCgWodZP/kxC7cdR65GGjPOh49aWWu8evWyZd5lzjVKdhXrr1vTu44LufvywbRtvuyFDnIV65Upudrfvvu7LlBMwaKuPigrrpM2Xo/75z3n7/td/We+Jo168mB+zafXh5KhHjGABW7PGEgpxqN268fqSJNTy+2UWH8Aqb20t/3/AgPQxuwVdqP10dgH4eLMPbSpUVLC5W7HCW6iNo44YXai9upB369a5q6iO7Fg5wJqb+S8KoV64kFsF5AuZDMDuqJ2ijzBCXV3Nw6S++Wbn7H7VKr4r6dHDO/ro1ctyhX6jD6mdB/KTUS9YADz8MHDVVemVUn368O/NRqjdHHW/flYzyrVrLYeq3+4nYahTJ0ftFH0Qsat2ij0Ay2kHcdRA5zsMQV5/9pl7RSLA29DPOWCE2if6xvRy1BUV3o3k7ZMHyNXbXsEhB5S95ceKFTzQzsCBPF6Gzo03Apde6r7uuNFPGsBf9OFXqKVVw+TJXIFmH3hp1So+IdzGUgH4BOzb19lFuiEtHeRuINeOWing2mt5/TfckP4/InbVItRRVSYuX27FKvX1ztEHkFyhdqpMBLi1zIknui9L2lL7ddT68p2iDylfpujDOOoIkZO/utq7MtEr9gA6C7Xb1bu8nN+zt7FcvpwFaeTIdEetFPDOO7y8IDMbR4mIj73Di1P0obsfP4ijlvGs7fHHqlUci2QS6n79ggu13DaXlUUj1EEy6hdeAF58EbjpJsvV6+yzj7V9o6xMFKGWEd6chC8JQq2P9eFVmQgAv/0tcOut7ssaMIBF2s9Y1IKbo9Z1IMrowzTPy4Cc/AMHejvqTK7Gr1AD3ClEKoqEFSuAoUM5P1y2zIoAVq2yRFpO3FxjF+rqat4mmzez+9PHTdDzRD+IUI8axa5CF+q2Nr7zEEetV/zqhBFqiT6I+EKT6+jjV7/i/f2DHzj/XwQVCCfUQ4bwb3vnHX69cycLs+6oGxutAfcL2VFnQnfU+nRyXmRy1EB00YeeicdA8Qn1tm3Ok7Nm46j79u382dGjWah1hyxCPXw4n0DS7vPdd63PBJmCPkqchBrgW+devdIrccJm1CUlwKGHpgv12rW8jUSoAecLwObN6ULtt8OLiFOUQu3XUa9Zw79XtqkdqVAEwgn1gAHAscfynIBKcaVsR0e6UANWzJY0Ry1C3b27s6PWW2ZkQoRaOrt4RZiCH6HOFH3s3Jl5smfpPu6nTCEpDqEWQZEs2alCMWpHPXo0r1e6CW/dymKzzz4s1IAVfyRJqPWMGmCxsTeJCptRA5xTL1li3dlI0zyJPgDnoWRFqN26TTuh98bzOzDTpk3ABRekt4EPG300NDg3JxN0oQ7rtqZM4bjj3XetFh969AEAH3/Mj4XiqLdtY1ELsk3692dBXL3aX0UikDn6KCtzr8AE/LdHj3mcD6BYhFpOfGmN4XTChnXUFRXOV36ZXv7DD/lRKtDEUQPpQi3Rgj6lUC7R3Q2QLtR6Pg2Ed9QAz2SuFPDPf/Jr6Yru5ai3bGHXIiOiuQ3vaUeiD8C/o/7Tn7iVxltv8WuleNsEjT7a27ncffq4fybb6AMATj+dhW7WrM5CLY5axmVJqqN2yqglsvKLdHpZvNhfPg1kdtT19c7NAQU5pu3nwd//Dvz1r9brmIc4BYpNqOXq6CTUYR2123CKBx7IO3nhQn6tC/XgwbyspUtZgObPB779bWuZueDVV9ObD3pFH3ZXGLYyEeCOGMOGAbfdxiIojlq2CdBZqO13Ln6FOmj0oRSLtL5OEZOg0UdjIy/Py1HvtZclBGGFumdPbhHx2GN84Zf2xoDVbPTzz3k9+vGdNKHWIy39Tsgv+ngfQR21m1B7xR6Atc/s23HGDODmm63XxlH7xC7UTtFHWEftdlBUVvLEpE6Ouls3/t+yZRwDbN9uNT3KhaPu6ABOOIFr0gU3oW5tdY8+glYmAtw88aabuH3x//4vC/XAgZ3b0urY6wKCCHUQR71wodWl2z4/Y9Dow6lZo52yMksMsqlomjKFj5tHH+U4RYxDeTm7y/Z2FiXdUCRFqEtK+E+ijmyFGgjuqN3aUXtVJALu0UdzM/Dpp9axY4TaJ/aMOmpH7cbo0emOurraEpvhw1moJZ+ePJnFJBeOevNm/g36dnATaiDa6AMAzjsP2G8/7qm3ciXn0/py7RcA6T4exFHv3Ml/ulBv2eLd/PGRR/hC0r27u1D7ddR+hBrgmKJ7d15vWE44gTP4r75Kj1MAK/6wC19ShNo+BZZEH0EqEoF0oc7WUcvrTI7aLfrYupUvjlKJG/PEtkCxCHWcGXUmoV6xgneUtPgQVzNiBL/3+ut8Mu+7LzuBXDhq6Yijn6hOHV4Eu9jIdvIj1B0dLHb6rb3uql991XIuUUYf9t54vXpxWfTpuXTa24G//IVFb489vIW6vT3zgFR+hXr//b2njPJDeTlw5pn83C7UUqFoF4rqam4K6TSaYa6wC7VMzJFvR92nDx+jMu2cG27Rhxxj0jzXOGqfxJ1RuyEViosWWUItDB/OwvH3vwMHH8wC3q9fbhy1CLU++atThxfBLjZE/gekkc/YHdK557JISdM8IB6h1lt9AO4tP159lbfL97+fvh+chBrI7Kr9CvVNN0UzdMD3v8+P++2X/r6Xowby66rdHLU+aYBfKiosMfTrqN0qE/v2BT75BPje97y/7xV9AEaoA9PSwlfIXr1YZNya5wVx1DLAeSZHDXBOvXx5Z6EGWFAOPpif59NRe0UfTmLj1YtQR9ZhF2px1YAVfXgJdXV1+iA6mYTaPhBRpm7kjzzCrvOUU3ifStzilFEDmYVamvdlEuqBA4FDDvH+jB+OOoorFKdMSX/fy1EDyRPqsI4asJy0X0ftFn0AVl2SF07RR1ubdXe6eDHfeW3bln+hJqLBRDSXiD4ioiVE9KNYSxSG7dv5IOjWjTeYW/QRxFHbc1MnBg9mJ/fyy7yz9Haz++9vPRehTpKj9sqoAf+O2k2oAXbVf/wjcPbZ/NpLqPXtHDb6AJyFuqUFePJJHoa0stKfo87URM+p632cELEDtAtyITlqOabCCrXEH34dtYhn0DxccIo+9Ght0SLrOMy3UAPYBeAnSqmRAA4FcCURjYy1VEFpabE2qlvHBz+OuqSEa+pbWvyN0kXErvof/+DXuqOurbWaT02aZC1r8+bwc/v5xctR+8mogeDRh1Pzs5ISYNo0S8z8CrUfN2+PPryE+sUX+fZU3GhU0UfPntlVEkaBCHVXcNQi1H4d9THHAL/7HXDYYcHXBTg7ajnu9tuPm7bKeD/5Fmql1Hql1Pup580APgYwyPtbOUZ3y3V1naOP9na+ZfHTREoOJr/DKY4ebZ0MulADwAEH8G2/fsumVPzjfTg5anuHl7Iya8jXbITay1HbcRPqzZvTu+lHHX28/TYL6uTJ/LpvXy53S0v46MM+z2S+kOij0Bx1GJc7aBBfkPx+t6wMuPLK8BdTp20ox52Iv3ScyrdQ6xDREADjALzj8L9pRDSPiOZtynU3aV2oe/bs7Kj9TBogSIWHX6GWCkWgs1DfcQcwe7b1WpYVd07tJ6MGrAMxrujDTr6ij3nzeMAoWb+sa9Om7Bx1EoRa7toKxVE3NaU3qwzCT38KPPtsdOXLRHk53zU7CbV9pMiYm+f5vtQQUQ2AJwFcrZTqNDWvUmomgJkAMHHiRIdZX2NEMmqAHbX0hhP8zJcohHHUALs0+8Fnb/4jzjruC5mfjBqwhoV1Emq5YGUiiFDLgW9f7pYt6aIXptVHTQ3XUdiFWikW6jPOsN7zI9R+MuokCHV5OXD33dbUXEIShbqqyqr7CSPUgwZ1nsQjTog6j6AnQn3AAeyik+SoiagMLNKzlFJ/i7VEYciUUfuZL1HQhbq0NHMb2FGjeIfa3bQTuXDULS0sVuXl/LslD7dn1AAfhFVVzhewKDJqO3rvNEHaPusHepjoo1s3532/fDmL6sSJ1nuyHzZv9ueof/xjYM6c9OVmGpApl1xxRWdTkEShrqy09k8Yoc4H9o5DeuXhqFHW+Cv5FmoiIgD3A/hYKfWbWEsTlkwZdVhH3bdv5oFjqqqAcePSIxA3cuGoZXwPaW8rB5k9owbSZ0exE0f0AXQW6m3b2PXqt45BHLW99YrdUc+bx49SoQt4O2p7Rq0UV0g9/nj6cpPiqN1IolDrF/NCEWr7LC/6bDoHHWS9n2+hBjAZwPkAjiWiBak/jzlz8oA9o25sTO9KHNZR+20G9NJLwJ13Zv6cjLQW1FGvWAGcdJI1vrUXEntIO245yER49IqV6mp3scmVUDtNY+RXqKur09vCOgn1e++xWIwaZb0nFZd+oo/WVq6IXrvW+n5HB6/Ha+S8fJNEodbPv0IRarfoo6Ym/ZiKWagzZtRKqTcAxDcidhTYM2qleINKbBHUUUuLAL9C7TQNkxOlpXxyB3XUP/kJ8PzzwPvvcxdoL0SopR23OE+ZgVy/QzjlFHcxjiOjluXqQi0Hvl2oZSaYkhLn5ehDnApujnrs2PRJjevqeF9s2mSt1y36kFt1XaibmlisjaP2xstRh23bnGvs0Yebo05KZWKisWfUAJ9gItRBHfXmzSxwEyZEX9Z+/YI56lde4W7ogOU+vfBy1PaZSK691n05Mi5DJrZvt8aQ9oMfR61PPuqWfetDnAq9eqVXJLe38xCzU6emf46IXfWmTdY2EUGxRx8So61ZwwaAyH/38XxSUsK/JUlCXYiO2in6kCFlxVHL7EYxUjxdyHVHDaTn1GEyaplxJGr69/fvqNvbuSJLfpNfoa6utsY90YVaP2kyIUKdaTLebdv4s34HgfcbfQDeFwqnThN1demOetky/pxekShIp5fWVqs1CtA5+pDjaNs263khCDWQ/xH07MdcIQq1PfoQg0DEd8cDB8YeewDFKNTiovXa/6COuqmJvx+HUAdx1H/+M/DBB8Dtt/Nrv0I9aJB1Ikj0sWOH+9x+ToiTzZQV24c4zYSbUNsrEzOt2y362LLFmlTYqSJRkF6i9h6rbtEHYMUfRqj9UQyViU7Rh172gw7yH31mQeELtVKdM2rAWaj9OuoNG/h5Ph311q3AjTfy5KmXXMLvhRFqr+jDC79jUkcl1E6O2kuonaKPQYM425ZpwN57j8smMZCORB92oXaLPgBLqP0OyJRvamutdsv5oFiiD3tlon7c/c//8Hg2MZNcof7hD4GHHsr8OblFdcqoBREGv45aHFlcjrqhIfPMxs8/zxeM227j/Ku2NphQi3jaKxP94jVjuM727cGE2n7gZyPU9pP94ou5PfH553MzxXnzuJ7BKT/Uow8vR+0k1IXiqI8+msc58TPpb9R0dPAxXoyVibpQjxoFHHlk7MVIrlA/+qhVieaFXYSdMuqgjlqIy1ED6bNgOyGTwo4fz489emQW6o4OFig3Rx00owb8Oeog8wG6tfqIIvqoquJJR7dtA845hycucIo9AN63X3/Ny3ESajEAhRx9TJ3K29DeBjwX2CeqAKxzq6wsmGnIJ06ViTG38HAimUKtFAvt+vWZP2sXanFm2ThqIS5HDWTOqdes4bxdfo8fod68mW/9nRx12Iw6F9FHRUV62cJGHwB37f3DH3hmndZW54pEwNoP69ZldtTduvGsMLpQ19amN/lLIhMn8h3GAw+4f+bJJ4HfxNCPzamDlRxThRJ7AHxs79xp3QG7HXcxk0yh3raNWzzos2i7IUKiX61lDAshiY46U069enX6nG5+hFqa5tXXdx5LN8kZtf3AzyTU0k7e7YQ//3yOQYg443dC9u2aNZkz6p49eZQ6XaiT7qYB/v1Tp/LogTK/n86dd/IUX7/8ZfTr9nLUhSbUgHUeeB13MZJMoZbYYv36zM3D5KTXb7/tYz7IZ/zc+svBRBTPyejXUWcj1IMGsQusqgqfUfsV6qAZtZNQ25s3ZRLq1lY+LrxOmHvu4Q5CMruMHemduGFDulCLS9ab5/XsyRe/QhNqgKfwKikBHnzQek8p4Oc/B66+mrdhU5NVLxMVTkJdiI7abnhM9KEhQr1rV+Zaa6dYwz7eh1QY+WnrK8vp0yeeRux+HfWaNdZYw0BwoQbSK0KCZtR+KxPDZNQ7d1qDRTkJtazbTajtQ5w6UVbGPRLdkAtmR0e6UBPxd/XmeXahTtKATJnYYw/uzfrww7zNGxpYvKdP57uOG26wppOKkmJ21EaoU+huOFNO7STU9jGp/UzDJdjHLI6a3r1ZDLyEevt2PqHCOOpu3ayZMGpq4nfUYaIPwBJh+8h5QOYOL/aR88JgH/9ap7y8c/RRX8/Pm5sLy1EDHH988QUPRXDAAVy5+ItfAPfea41X4jTPaDYUi6PWu+LLfIlGqFPoB02mnNqeUQOdow8/03AJshx9xpEoKSnhk8Mr+lizhh/DCPUee1gDL+mOOkmViYAlwmGiD/tY1GHQB1SyHxvdu6cLdV2dNeXVunWFJ9SnnMK/9847eUb4efN44mGizLO3h8XLURdK0zwgPfrwcycXE8kc6yOIULtl1MuWpX8mqFDH5aiBzJ1eZB42e/TR3My36m6zJ0sbaiFuR71rFy83aDtqfblhKhOjOGFKS1lsv/rKWaj15nmjR1tCvWYNfyfJI+fZ6d4duP9+vju95JL0ERSlJ28uHLVs50J01Nu3R3MnF5LiEepMGXVSog9ZtpejFqG2O2ql+MruJlDr1gHDhlmv7Rl11EIdZNIAIQpHHdUJ07evs1C7RR8A8PHHnOkWkqMGgNNOc34/l466Wzfe1oUo1Nu2Obf5zxHJjD7koKmqCpdRy+A8UmFVaI56zRq+LdXdsYiZPl29nUyOOkhlop+BkYIOcQqEF+qGBuuiHUX0AVj72C366Ojg8tXVAXvuyf9buJAfC02o3cilowZ4+F2nLv1JRTcseRTq5DrqkhJg333DZdR7780ivW4du9IkOmqvSQBWr+ZRufQOFSJmTU2WaOg0N/PFSZwfkF1GLe7Hy1FnK9Q7d3K57EJdWsr7Xxfq//gPHvJ16dLoskIvod6xg9fT0cFiVlHBn//wQ/5MsQl1Lhw1wNsv6qaAcaI76jxm1Ml01HK7ueee4TJquf2X+cyCVCYOGsQVcnGMRS0MHconhls38jVr0mMPIF2onViyhB/1ufOyyaiBzLO8yP/CCrXTpAGCfZaXVau4zfPNN0cXfWRy1OIyRczq64HFi/l5sQi105ALUeAm1ID/IXGTgFP0YTq8pNiyxbrd9CvU+sm27778+Nln1mf8Ouq6Oo5bJk8OVuYgiJiKuNpZvTq9IhHILNSLFvGjPj1QNhk1kFmos3XUTgMyCXah3ryZT/Df/Y67hwPxCbVk1CJeImb19VaZikWoKyr4zi1XjrrQSEj0kUyh1h31hg3evRNbWvhg0FtC1NezKIVx1LnAS6iV6twrEfAn1DU1HPsINTXW5LFBM2og8ywvItRhKxOdxqLWP6cL9aZNwHnncQXg449bs2xkgzTBdIs+RLx0Ry0Ui1BLE71cOupCQjrKmcpEB0SoBw7krDlT5xD7CVtSwvFCGEedCwYP5p3tJNQNDSxQQR314sV8AdAvWNXV1rgYSgV31JnmTYzbUctFoq2NRXO//YBf/5rfq6nJ/hY6TPQhFItQA507iEVBsQg1kXVnaoTahh59AN7xhz5fos6wYcl11ETAyJHOQu3UNA/wFmql2FHrk20CVjQgw3IWakYt5e/blwdcOuooq/dlNviNPuxCXV1d+AKkYxy1N3IebN0azZ1cCJIp1Hr0AXg30XNzy/vuy45aqeQ5aoDdr5NQO/VKBKyruJNQb9zIGa6eTwOWgMYl1GEctZ75+c2o5Y6qXz++yD37LPCvf/lfpxsy7oqboxaXqWfUQHG5acDZUS9aBFxzjSW4QSkmodYddRR3ciEoDKHO5KidRHjYML4CbtyYPEcNsFBv2tQ51nHqlQhwhY/M52hHKhLdHLVM+Bomo/Yj1Nlm1JmEWgbmkky5tjY9iw/LmDEcpZx4Yvr7klG7OepiE2onR/23vwF33MGj7IVBOgwVygQBXuhCnYfYA0iiUHd08Aapq7Nub72E2imjBqyWH59+yiddEh010NlVr17NgurUjtttvA9pMha1o7YPSWonjKOWGb8zVSa6Oeoo6dYN+M//7Hyh0DPqsjLrIi+diYpNqJ0ctfScnTGDp/MKyo4dvO3chjsoJMSwGKHWkLFxe/bkE6Zfv/AZNWAJYRIdNdBZqGV4U6fbKzehXrSIb+PlVl6IO/rYvp0rboMsl8iqKGxqsipr7Hg56rjRM+q6OmtfVFfzTOfFJtROjnrjRh7L+4ADgAsu8Dchs459YttCxjhqB+y3m3vuGS6jHjKEr+biNpPmqAcNYuH96KP0952a5gluE9wuXtzZTQPxVybKyHlBMztx6nLgO7kuJ6HO1UBIevM8OQ6F667jCs1iomdP3pf6hMsbN/Jx+Je/cEukiy8O1qOw2IRaKhONUKewdzLI1OnFTai7d+cDTYQ6aY6ayLlC0alXouDkqDs6eBn2fBqwnKpk1HEIdZB8WhChdhrnQ7BHH3KHlQv06MMu1Ndf7z7AUaHi1Dtx40a+QxszhseufuYZqy7ED8Uk1DLBbZ6m4QKSKNT2TgYDB4bLqAHOqZPqqIHOQt3Wxr/VXpEoOAn1ihV8EPlx1GErE92cVNCxqAU/Qq13eNm8OXexB9A5+ih2nAZmEqEGgO99jx/feMP/MotJqE304YBT09PFLAAAE7RJREFU9LFhgzUSnh23jBrgnFpum5PmqAEW6s2brYqbL75ghxzEUbtVJALRVCYC7k20gs6XqC9XhNrtwNc7vGzeHO8gWXb05nl2R12M2AdmamvjY0aEesgQPg/ffNP/MotNqE1log2n6KOjw70yw6uNtLT8AJIr1IDlqqUNdRBHLbej+mBMQhSViYB7/BHWUUvX9CDRRy4ddffufBfR0NA1hNoefchgYSLURDz2TVd11Hr0kVShJqI/EdFGIlqciwJ1ij4ytaX2Emp9EP2kRh+AJdQyhGYmR61HEYsXs+NxOoDKylh0ssmogeiFWrqmO82XKIhQK5Wf6AOwsvFix+6o5Q5Pb0V0xBFc0S1mIhNBh9VNMtXV/Ht27Eh0Rv0ggONjLoeFPfoYOJAfnYS6vd27jXTSHfXAgdzca8kSYO5cbtM7cSKPaeFEjx7WBJuCU9dxnZqa7DJqwFuo46xMlMGkNm3KffQB8J1cV8io7Y7aTagB//FHMTlq3Ywk1VErpV4D8FUOysI0NvJJKjvZy1HLrbGbWOhCnURHLS0//vEPnoB0n32AOXPS57TTsY/3sWMH8Mknzvm0UF0dn6OOIqP2EmqAb8N37Mh99CEYR82MHs372m/8UUxCretLUoU659ibRA0YwILm1JbaaRounepqngQASKajBlioV67kcr74orcg2YV6+XJu+zpypPt3ZKhTIHxlolvvxGyjDz9CvXYtP+bDUQNdQ6hlH3g56tJS4LDDuqZQF4Kj9gsRTSOieUQ0b1PQXkw69pr2sjI+YJwcdSahBqycOomOGgBOOolnk3npJSvmccMu1DKMq57F29EPsiRl1A0NHG14tfoArEw0Hxk10DWij9JSvqDrjrq0tPNvP+IIjtr8jLRnhDpSIhNqpdRMpdREpdTEftm4H6e2q4MG8VRMdpzmS7Qj8UdSHfUppwDz5vkbZCiMUOuVH0nKqL0GZJLPAJajNtFHvOjdyKUNtb3H6eTJnNu//Xbn7yvFnWKk/qSYhFo/xhNcmZhbnHqDjRrl3CvKab5EO1IxF8b5JQ27UH/+OW8rr67VcTnqtjaOXcI6asFEH8lAH5hJ7+yic8ghPLaLU/wxfz5w6qnAb3/Lr4tJqAvBURPRXwD8H4DhRLSWiC6OoyCzZnErs4/f3oLn3uyJWbO0f44ezdGHdF4R/EQfl13GCy+GW1gnR73vvt5jbegOIKxQO2XUYUbOE4IItYk+coOTo7ZTW8tdyp1afrzzDj/eey+7biPUkeKn1ce5SqmBSqkypVS9Uur+qAsxaxYwbRqnGz3RiC+212HaNFhiPXo0P9pdtR+h7tuX59orBpwctVfsAWTnqGW7OjnqbIRavwPyI9Slpbl1tl3VUWcSaoBz6rff5jsqnXff5cfly7mpaTEJtWn1wdx4o6UFPdGIRvTE9u38PgC+igNWhxDBT0ZdTOhC3dbGrUX0JohO6EJaVhZsfV7RR5hJAwR9f/mpTOzbN7ezanRFoa6ryxx9ACzULS0cdei89x5w3HE8BOzMmcUl1Po51JUzapnUpBRtqMZ2NKJn2vvo35+b6S1cmP5FPxl1MVFRwe6yqYk3zq5dmR21HFjdu4cbjhSI3lEHiT6++CK3sQdgCXVVVfCLW6EijnrbNv5zE+qjj+bHl16y3mtqApYuBY48kseu/vvfuY9DsQl1t25505pECLX0mO4JvvXagrq09wFw/OEm1F3FURNZ3chl4l6/0UeY7rwlJXyyOQl1mIlthSBC3d6e24pEwBKYruKmAasyUdpQu23zfv2AsWPTZ32ZP59bfRx8MHDppXy3p1TxCLWIc57mSwQSItTTp/O2EKFuRE9UVfH7uxkzhse10Ac372pCDVhCLU3zMkUfuqMOgwygZCcqR50p+gDy56i7klDX1fG5Jc1g3Rw1AHzrW1yhKMfAe+/x48SJ3Plq8mR+XSxCXVHBAp2nfBpIiFBPmcKx1og9WKjL+/XEzJn8/m5Gj+bc69NPrfe6WkYNpDvqysrMnWSycdSA1YvQThQZdXm5+8lshDq3yG/95BN+9BLq445j1/z66/z6vfeAoUOt/TRtGj8Wi1AT8XHe1YUaYFF+bhZXZtz7eM90kQaslh96/NHVMmqAhbq52Wqal2nyUHHUYU8ap1lelAL++ld+HiaWEKH2OvD1i2++oo+u0jQPsH6rGCEvoT7iCL6YSfzx7rscewhnnQWceSZn1sVCdXXeKhKBBAk1gM5jUeuMGMEVaXrLj5YWvtoVy3CKftCjj0yxB5C9o3YS6htv5Ln0pk+3ZuYOgoiwWz4NGEeda+yO2uviWFXF8caLL3KmvXo1MGmS9f/KSr6Qjx8fX3lzTXW1cdS7sQ9xqlNezjMi2x11ZWXeAv680KMHV/osX565IhGIJqPWhfoPfwBuu41vb3/2s3DL9CPU+h1Arh11VxRq3VFXV2euezjuODZNzz3Hr3VHXYz06ZP741AjWUJtnzTAjr3lh9d8icVKjx48T2Jra24ctQxJCgAvvwxceSUPJPX734e/QEpU5SXURJZY59pRd8XoQ865zz/3jj2Eb32LH3/1K47fisk9O/HII8CMGXlbfbKEWhy12wk8ejR3gJCB8L3mSyxWevSw5o8M4qizzajb24Grr+ZKo8cecx8z2w9+HDVgxR/5EOr+/d0ncChG5KK0c6c/oR4/nr+zdCkP1VsMY+l4MXy4+xR5OSB5Ql1T4y4C0kNx0SIeDH/+/LzmRnlBFzc/Qh1VRv3oo7zdp0/P/qQMKtS5vuUsKeG7lotjGdYmmeh3sX6EuqQEOPZYfq7n04ZYSJZQZ5r1WVp+PPss1zwvXQr84he5KVtSEHErLfV3hY8io96yBbj5ZmDcOK7RzxY/rT6A/DlqgH93phY1xURVlWWQ/Ag1wDk1YIQ6ByTrSEyNRS0j6XXrxo+7B2faYw8+aWfMANatA154ATjjjDwWOA+IUA8d6i9+iMJRr1/P44rcfns04iUC7Eeoa2uLpz1ukiGyTJJfoT79dOCEE3hMdUOsZBE0xkBjIzbt7Ilp06yGBqtWWe3np0wh4Bvf4CEV58zxniuwWBGh9hN7ACxyJSXZVSYCfJsrFUjZQsTjFn/zm96fq6jIj5vuqtTV8cw7foV6wADg+efjLZMBQAKFevGaftjemv62jKQ3ZQqAhx9mJ9lVBsuxI0Ltp8UHwKJYXR3elcr6br892maQV1+d+TOVlV2rjXy+CeqoDTkjWUK9ZQvWtzrXtO8eSa+rNcezE9RRA5xThxW8yy8HDj00Pznk2Wd3rZw43xihTizJOgsaG9FR41yZuNdecM+uuxLDhgH7728NN+mHX/8a+MEPwq1vwADg29/Oz7a/+mrgqqtysCIDAKuJnhHqxJEsR93YiHH/1hNVL6d3hquqAk48ER7Zde6Lmjf69AGWLQv2nZAz3MyaxZHTqlWceijF73fZbV/sGEedWJLlqLduxYF/vRUzZ/Kk3ESsS5WVwD33dB5yIm0WmBjIpYtM2t2CPj0aYIm0EPe2N+QBcdSmAjd5KKUi/5swYYKKgkceUaqqSimWCec/okhW5WvdVVX8fiGvyy977+293ePc9oY8MX++Ur/6Vb5L0WUBME+5aGqyHLUNfS5FN5SKx4E6rTsuF5nLdflld+WtB2kz8BQISbtzSRTjxwPXXpvvUhgcSLRQ+xELgG/Pzz+foxI/J5+fk9Vt3X7LFIQo1hW1AGUSYSLe7oUkdnqco5SVtRdK+XOF27FkLnJ5xM1qZ/MXVfTh5/bb7ZYc4O/b4wO/MYPbuvfe2728jzzC/ydyXnfQ3+m1Lr+/KWyZnJYp21UekxLT+CXb7RyUsNs+n7gdS1dckbx4zotC3PbwiD4SLdRuB00Q0bYfTH5PVi+h0ne8HBBOAuZ1wcj0XT8ngf59p78+fbITcKfP5FrsosS+je3lj/JkTmK9gx/c9m9JSeHs90Ld9gUr1EoFEwuvP/mu18lqF61MQurkMryWLeLZp084Ybdvl6AXLT8Cngm37ZepYjHshSFKMh03UZ7MUVzQ8uEKvc6PMPtdJ+7fk0kbknhR0SlooXYirEj5PQjtghk2ggnzp6/T6YCOszyZ1q1UZgHSv+/nohT0wpXNye7nuInqZA57QfMqa5yuMNNx5eaoS0r87Yu4f0/ULcTcjrM4LzZFJ9RKebvdKP/CutYo16nHFXGXJ9PJlCkPj6t8bvl4kLsQ+3GT7ckcZvl+6ziyjRqCCEqm/ea27d0+F+YCny1+zEu2dT5x5/RFKdQ6cYu220kTx5+Xc8lVGTId4E6u2UtccvUXxIFnEtMg0Yzf48/rpPZ7kbNfSJzK5Ne9+rk7c7swZtrXTndMUW1vJzKd9/rvD3vXGOY3BKHohVonSDQgO8rPZ4Nm0WEuGNm60aqqzCdFmD83l5kLh59NmZ32g1/3bxf9IC7L6U+/oGUjDvY4zk1IM30/m30nIhfl/vJz52YXf3u85lXeK67wX+kf9rcFvbuz06WEWvCbRwb5XKaTKYzLiioPzyQo2Qi42y1jLrP7KP/sebrfEzCbv0yi42cZ2ZiAKH9PEIMT9C/I9vCzzYNelKIyO2EikS4p1Er5b/4WpJlcmEoRt6jA6cob1Ol43dL6uR3O5oAL00Igk1BEJUaZypHN7wj6l00kJN+Nu4xO+91NsPwanHz++Y25vI6RKMoQhKyFGsDxAJYB+AzA9Zk+nxSh1vGbdyWhGZmfSiX7wRh0uX4cfKble7W59boo+blw+bm9D3sy2U+gOO8MshEzuUjm+s7Fb9Ydtm4o7rqMqC7EUdx5BCEroQZQAuBzAPsA6A7gQwAjvb6TRKEuVPLRrMnv8nPVhCzbSrxM5YvaHUYRZ+m/M1du2u/dmdc+yrQvgvY/CPMX5YXYrc+Bn9+QU0cN4DAAL2ivfwbgZ17fMUIdLbl08FG49Hzgx4FHITRBBFZfdhBhCio0XhWnfkUxyn0X9I7Jz1/UF+IgTQ7DGIWcZ9QAzgRwn/b6fAC/8/qOEWpDPsn24pHpBAzTnjabJnxeQuOnKWLUghIlmS5iTkMeOLX6CBtp+on//P6ObA1LToQawDQA8wDM22uvvYKX0mBIIJlO8GzvQrIVmqh+Tz4Jsz3iKEO+xwfxEmri/7tDRIcBuFUp9e3U65+lRt27ze07EydOVPPmzfNcrsFgMCQJmXpu9Woe5nf69NxONUdE85VSE53+52fOxPcA7EdEQwGsA3AOgHCT8BkMBkNCmTIluXOAZhRqpdQuIvoPAC+AW4D8SSm1JPaSGQwGgwGAz1nIlVLPA3g+5rIYDAaDwYFET8VlMBgMBiPUBoPBkHiMUBsMBkPCydg8L9RCiTYBWBXgK30BbI68IMmmK/5moGv+7q74m4Gu+buz+c17K6X6Of0jFqEOChHNc2s/WKx0xd8MdM3f3RV/M9A1f3dcv9lEHwaDwZBwjFAbDAZDwkmKUM/MdwHyQFf8zUDX/N1d8TcDXfN3x/KbE5FRGwwGg8GdpDhqg8FgMLhghNpgMBgSTl6FmoiOJ6JlRPQZEV2fz7LECRENJqK5RPQRES0hoh+l3u9NRP8iok9Tj73yXdaoIaISIvqAiJ5NvR5KRO+k9vljRNQ932WMGiKqI6IniGgpEX1MRIcV+74momtSx/ZiIvoLEVUU474moj8R0UYiWqy957hvibkr9fsXEtH4sOvNm1ATUQmA3wM4AcBIAOcS0ch8lSdmdgH4iVJqJIBDAVyZ+q3XA3hJKbUfgJdSr4uNHwH4WHv9/wP4rVJqGICvAVycl1LFy50A/qGUGgFgDPj3F+2+JqJBAK4CMFEpNQo8yuY5KM59/SB4sm8dt317AoD9Un/TANwTeq1uMwrE/YcQczEWyx+A/wXwLfDM7gNT7w0EsCzfZYv4d9anDtxjATwLgMC9tkqdjoFi+APQE8AKpCrqtfeLdl8DGARgDYDe4BE5nwXw7WLd1wCGAFicad8C+COAc50+F/Qvn9GH7Fxhbeq9ooaIhgAYB+AdAAOUUutT/9oAYECeihUXdwC4DkBH6nUfAFuUUrtSr4txnw8FsAnAA6nI5z4iqkYR72ul1DoAMwCsBrAeQCOA+Sj+fS247dvINM5UJuYQIqoB8CSAq5VSTfr/FF9yi6atJBGdDGCjUmp+vsuSY0oBjAdwj1JqHIBtsMUcRbivewE4DXyR2hNANTrHA12CuPZtPoV6HYDB2uv61HtFCRGVgUV6llLqb6m3vySigan/DwSwMV/li4HJAE4lopUAZoPjjzsB1BGRTFhRjPt8LYC1Sql3Uq+fAAt3Me/r4wCsUEptUkq1AfgbeP8X+74W3PZtZBqXT6HePRdjqjb4HABP57E8sUFEBOB+AB8rpX6j/etpABemnl8Izq6LAqXUz5RS9UqpIeB9+7JSagqAueCZ7YEi+80AoJTaAGANEQ1PvfVNAB+hiPc1OPI4lIiqUse6/Oai3tcabvv2aQAXpFp/HAqgUYtIgpHnUP5EAJ8A+BzAjfmuJIjxdx4Bvh1aCGBB6u9EcGb7EoBPAbwIoHe+yxrT7z8awLOp5/sAeBfAZwD+CqA83+WL4feOBTAvtb+fAtCr2Pc1gP8CsBTAYgAPAygvxn0N4C/gHL4NfPd0sdu+BVee/z6lb4vArWJCrdd0ITcYDIaEYyoTDQaDIeEYoTYYDIaEY4TaYDAYEo4RaoPBYEg4RqgNBoMh4RihNhgMhoRjhNpgMBgSzv8Dbb1wNaBV/AoAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gs6ckcFQLl6_",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "2e58e5d4-75ee-4a9e-f72e-091cad66b12e"
      },
      "source": [
        "cd '/content/drive/My Drive/TEST1'"
      ],
      "execution_count": 74,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/content/drive/My Drive/TEST1\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MWfDktW-nOal",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "model = load_model('model.31-0.94.hdf5')"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QTvRL0NlnzxH",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 71
        },
        "outputId": "7a1f78fe-3183-434e-e2f7-8002ece20108"
      },
      "source": [
        "model = load_model('model.04-0.84.hdf5')"
      ],
      "execution_count": 75,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/keras/engine/saving.py:384: UserWarning: Error in loading the saved optimizer state. As a result, your model is starting with a freshly initialized optimizer.\n",
            "  warnings.warn('Error in loading the saved optimizer '\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IBLZs1kMnaC6",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "classes = ['CNV',  'NORMAL']\n",
        "\n",
        "def preprocess_input(x):\n",
        "    x = img_to_array(x) / 255.\n",
        "    return np.expand_dims(x, axis=0) \n",
        "\n",
        "# Prediction for an image path in the local directory\n",
        "def predict_from_image_path(image_path):\n",
        "    return predict_image(load_img(image_path, target_size=(299, 299)))\n",
        "\n",
        "# Prediction for an image URL path\n",
        "def predict_from_image_url(image_url):\n",
        "    res = requests.get(image_url)\n",
        "    im = Image.open(BytesIO(res.content))\n",
        "    return predict_from_image_path(im.fp)\n",
        "    \n",
        "# Predict an image\n",
        "def predict_image(im):\n",
        "    x = preprocess_input(im)\n",
        "    pred = np.argmax(model.predict(x))\n",
        "    return pred, classes[pred]"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lHQXkyXhnPtU",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "d187e0bf-7e5c-4063-ab63-7efdc7ad7357"
      },
      "source": [
        "#Picture with CNV\n",
        "print(predict_from_image_path('/content/drive/My Drive/TEST2/CNV/test2c.jpeg'))"
      ],
      "execution_count": 77,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(0, 'CNV')\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jmk73ohjoG1Q",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "6c940290-691f-4705-8f95-75c9de502ad7"
      },
      "source": [
        "#Picture without an eye disease\n",
        "print(predict_from_image_path('/content/drive/My Drive/TEST2/CNV/test1n.jpeg'))"
      ],
      "execution_count": 78,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(1, 'NORMAL')\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "syZs5i1iqTN8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "1273978f-9b36-4948-d776-99b866810cb9"
      },
      "source": [
        "#CNV again\n",
        "print(predict_from_image_path('/content/drive/My Drive/TEST2/CNV/no-disease-cnv-28682-8.jpeg'))"
      ],
      "execution_count": 79,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(0, 'CNV')\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9sR10KOaqlf6",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}