stacked_ensemble_learning.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "project5_final.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "gpuClass": "standard"
  },
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "!nvidia-smi"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yZkcpHKPqLj3",
        "outputId": "12ba4863-b82f-430e-8bfa-951a8db10509"
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mon Jul 11 12:13:17 2022       \n",
            "+-----------------------------------------------------------------------------+\n",
            "| NVIDIA-SMI 460.32.03    Driver Version: 460.32.03    CUDA Version: 11.2     |\n",
            "|-------------------------------+----------------------+----------------------+\n",
            "| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n",
            "| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n",
            "|                               |                      |               MIG M. |\n",
            "|===============================+======================+======================|\n",
            "|   0  Tesla T4            Off  | 00000000:00:04.0 Off |                    0 |\n",
            "| N/A   67C    P8    10W /  70W |      0MiB / 15109MiB |      0%      Default |\n",
            "|                               |                      |                  N/A |\n",
            "+-------------------------------+----------------------+----------------------+\n",
            "                                                                               \n",
            "+-----------------------------------------------------------------------------+\n",
            "| Processes:                                                                  |\n",
            "|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n",
            "|        ID   ID                                                   Usage      |\n",
            "|=============================================================================|\n",
            "|  No running processes found                                                 |\n",
            "+-----------------------------------------------------------------------------+\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qdksjgEdamYa",
        "outputId": "6ce51cbd-addf-44a4-ebc0-8b2b34dd409e"
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "CdXsRd2NpXWe"
      },
      "outputs": [],
      "source": [
        "#import necessary Libraries\n",
        "import keras\n",
        "import numpy as np \n",
        "import pandas as pd \n",
        "import matplotlib.pyplot as plt\n",
        "import cv2\n",
        "import scipy.ndimage\n",
        "import tensorflow as tf\n",
        "from PIL import Image\n",
        "from tensorflow.keras.preprocessing import image\n",
        "import os\n",
        "from tensorflow.keras.utils import to_categorical\n",
        "from keras.models import Sequential, load_model\n",
        "from keras.layers import Conv2D, MaxPooling2D, Dense, Flatten, Dropout,Activation,GlobalAveragePooling2D\n",
        "import random\n",
        "import pickle\n",
        "import skimage.morphology as morp\n",
        "from skimage.filters import rank\n",
        "from sklearn.utils import shuffle\n",
        "import csv\n",
        "from sklearn.metrics import confusion_matrix\n",
        "from keras.utils.vis_utils import plot_model"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "training_file = \"/content/drive/MyDrive/train.p\"\n",
        "validation_file= \"/content/drive/MyDrive/valid.p\"\n",
        "testing_file = \"/content/drive/MyDrive/test.p\"\n",
        "\n",
        "with open(training_file, mode='rb') as f:\n",
        "    train = pickle.load(f)\n",
        "with open(validation_file, mode='rb') as f:\n",
        "    valid = pickle.load(f)\n",
        "with open(testing_file, mode='rb') as f:\n",
        "    test = pickle.load(f)"
      ],
      "metadata": {
        "id": "gxc1Ds05rWP1"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# from PIL import Image\n",
        "# cur_path =r'/content/drive/MyDrive/'\n",
        "# print(cur_path)\n",
        "# path = os.path.join(cur_path,'Train',str(0))\n"
      ],
      "metadata": {
        "id": "EOrXqFCr5vgJ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# data = []\n",
        "# labels = []\n",
        "# classes = 43\n",
        " \n",
        "\n",
        "# #Retrieving the images and their labels \n",
        "# for i in range(classes):\n",
        "#     path = os.path.join(cur_path,'Train',str(i))\n",
        "#     images = os.listdir(path)\n",
        "\n",
        "#     for a in images:\n",
        "#         try:\n",
        "#             image = Image.open(path + '/'+ a)\n",
        "#             image = image.resize((30,30))\n",
        "#             image = np.array(image)\n",
        "    \n",
        "#             #data.append(image)\n",
        "#             #labels.append(i)\n",
        "#             data.append([image,i]) #appending all value together \n",
        "#         except:\n",
        "#             print(\"Error loading image\")"
      ],
      "metadata": {
        "id": "YnU3n1iRuq3F"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Split Train, Validation and Test samples and labels\n",
        "X_train, y_train = train['features'], train['labels']\n",
        "X_valid, y_valid = valid['features'], valid['labels']\n",
        "X_test, y_test = test['features'], test['labels']\n",
        "\n",
        "# Check to ensure features equals labels for each data set\n",
        "assert(len(X_train) == len(y_train))\n",
        "assert(len(X_valid) == len(y_valid))\n",
        "assert(len(X_test) == len(y_test))\n",
        "\n",
        "# Print shapes of training, validation and test data\n",
        "print(\"X_train shape:\", X_train[0].shape)\n",
        "print(\"y_train shape:\", y_train.shape)\n",
        "print(\"X_valid shape:\", X_valid[0].shape)\n",
        "print(\"y_valid shape:\", y_valid.shape)\n",
        "print(\"X_test shape:\", X_test[0].shape)\n",
        "print(\"y_test shape:\", y_test.shape)\n",
        "\n",
        "n_train = len(X_train)\n",
        "n_validation = len(X_valid)\n",
        "n_test = len(X_test)\n",
        "image_shape = X_train[0].shape\n",
        "n_classes = len(np.unique(y_train)) \n",
        "print(\"Number of classes =\", n_classes)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HvcjURvkNm-p",
        "outputId": "83536966-bfd9-4621-d317-21ccaa37c144"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "X_train shape: (32, 32, 3)\n",
            "y_train shape: (34799,)\n",
            "X_valid shape: (32, 32, 3)\n",
            "y_valid shape: (4410,)\n",
            "X_test shape: (32, 32, 3)\n",
            "y_test shape: (12630,)\n",
            "Number of classes = 43\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fig, axs = plt.subplots(2,10, figsize=(15, 6)) \n",
        "fig.subplots_adjust(hspace = .2, wspace=.1)\n",
        "axs = axs.ravel()\n",
        "for i in range(20):\n",
        "    index = random.randint(0, len(X_train))\n",
        "    image = X_train[index]\n",
        "    axs[i].axis('off')\n",
        "    axs[i].imshow(image)\n",
        "    axs[i].set_title(y_train[index])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 302
        },
        "id": "b-wahyCBN7M5",
        "outputId": "720b623d-cb19-418c-eb75-88808b5b930e"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1080x432 with 20 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Histogram of Image Class Distribution\n",
        "\n",
        "a = y_train\n",
        "b = y_valid\n",
        "c = y_test\n",
        "bins = n_classes\n",
        "plt.hist(a, bins, alpha = 1.0, label='Training data')\n",
        "plt.hist(b, bins, alpha = 1.0, label='Validation data')\n",
        "plt.hist(c, bins, alpha = 0.6, label='Test data')\n",
        "plt.legend(loc='upper center')\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "cZ0HQCZdOIhB",
        "outputId": "ece53256-8055-4764-ecc4-57f91b9d9531"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## Data augmentation-Rotation\n",
        "def rotate_image(imager,degree):\n",
        "    X_rotated=[]\n",
        "    for im in (imager):\n",
        "      rotated_image = scipy.ndimage.rotate(im, degree)\n",
        "      X_rotated.append(rotated_image)\n",
        "    return X_rotated\n",
        "\n",
        "def  crop_image(imager):\n",
        "    for i in range(len(imager)):\n",
        "      imager[i] = imager[i][2:34,2:34]   # box=(y:y+crop, x:x+crop)\n",
        "    return imager\n",
        "\n",
        "X_train_rotated_positive=rotate_image(X_train,10)\n",
        "X_train_rotated_negative=rotate_image(X_train,350)\n",
        "X_train_rotated_positive=crop_image(X_train_rotated_positive)\n",
        "X_train_rotated_negative=crop_image(X_train_rotated_negative)\n",
        "    \n",
        "# appending rotated images to training set\n",
        "# Convert the data into list type to use the method \"append\"\n",
        "X_train = list(X_train)\n",
        "\n",
        "# combine the Lists\n",
        "for i in range(len(X_train_rotated_positive)):\n",
        "    X_train.append(X_train_rotated_positive[i])\n",
        "    X_train.append(X_train_rotated_negative[i])\n",
        "    \n",
        "#Convert the data back to a np.array\n",
        "X_train = np.array(X_train)\n",
        "\n",
        "# New number of training examples (after data augmentation)\n",
        "new_n_train = len(X_train)\n",
        "print(\"New number of training samples after data augmentation =\", new_n_train)\n",
        "# do the same for the labels y_train\n",
        "# Convert the data into list type to use the method \"append\"\n",
        "y_train = list(y_train)\n",
        "\n",
        "# lengthen the list\n",
        "for i in range(len(y_train)):\n",
        "    y_train.append(y_train[i])\n",
        "    y_train.append(y_train[i]) \n",
        "#Convert the data back to a np.array\n",
        "y_train = np.array(y_train)\n",
        "# New length of ground truth labels\n",
        "new_n_train_y = len(y_train)\n",
        "print(\"New length of ground truth labels =\", new_n_train_y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UBnuS5LzOOyZ",
        "outputId": "4d8a486f-0408-497e-dd93-f4698d46ac9f"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "New number of training samples after data augmentation = 104397\n",
            "New length of ground truth labels = 104397\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Display an example for rotation\n",
        "\n",
        "fig, (ax1, ax2, ax3) = plt.subplots(1,3)\n",
        "ax1.imshow(X_train[2018])\n",
        "ax1.set_title('Original')    \n",
        "ax1.axis('ON')  # clear x- and y-axes\n",
        "ax2.imshow(X_train_rotated_positive[2018])\n",
        "ax2.set_title('Rotated +10°')    \n",
        "ax2.axis('ON')  # clear x- and y-axes\n",
        "ax3.imshow(X_train_rotated_negative[2018])\n",
        "ax3.set_title('Rotated -10°')   \n",
        "ax3.axis('ON')  # clear x- and y-axes\n",
        "plt.show()\n",
        "X_train, y_train = shuffle(X_train, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 162
        },
        "id": "-KC5JnUOOvKz",
        "outputId": "63cbcce5-7502-4483-9b19-6537c535c5b8"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def normalize(image):\n",
        "    normal=[]\n",
        "    for im in image:\n",
        "      norm=cv2.equalizeHist(norm2)\n",
        "      normal.append(np.reshape(norm, (im.shape[0],im.shape[1],im.shape[2])))          \n",
        "    return normal"
      ],
      "metadata": {
        "id": "fav6_LSfO_aL"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Normalize images\n",
        "import gc\n",
        "gc.collect()\n",
        "X_train_normalized = []\n",
        "X_valid_normalized = []\n",
        "X_test_normalized = []\n",
        "X_train_normalized=normalize(X_train)\n",
        "X_valid_normalized=normalize(X_valid)\n",
        "X_test_normalized=normalize(X_test)\n",
        "\n",
        "X_train_normalized=np.array(X_train_normalized)\n",
        "X_valid_normalized=np.array(X_valid_normalized)\n",
        "X_test_normalized=np.array(X_test_normalized)\n",
        "\n",
        "print(X_train_normalized.shape)\n",
        "print(X_valid_normalized.shape)\n",
        "print(X_test_normalized.shape)\n",
        "\n"
      ],
      "metadata": {
        "id": "PInfUQLRPNAz"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#Building the CNN model\n",
        "modelCNN=None\n",
        "modelCNN = Sequential()\n",
        "modelCNN.add(Conv2D(32, (3, 3), input_shape=(32, 32, 3), padding='same'))\n",
        "modelCNN.add(Activation('relu'))\n",
        "modelCNN.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "modelCNN.add(Conv2D(64, (3, 3), padding='same'))\n",
        "modelCNN.add(Activation('relu'))\n",
        "modelCNN.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "modelCNN.add(Conv2D(128, (3, 3), padding='same'))\n",
        "modelCNN.add(Activation('relu'))\n",
        "modelCNN.add(MaxPooling2D(pool_size=(2, 2)))\n",
        "\n",
        "modelCNN.add(Flatten())  # this converts our 3D feature maps to 1D feature vectors\n",
        "modelCNN.add(Dense(512))\n",
        "modelCNN.add(Activation('relu'))\n",
        "modelCNN.add(Dropout(0.5))\n",
        "modelCNN.add(Dense(43, activation='softmax'))\n",
        "\n",
        "\n",
        "modelCNN.compile(loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False),\n",
        "                                  optimizer=tf.keras.optimizers.Adam(learning_rate=0.001),\n",
        "                                  metrics=['accuracy'])\n",
        "\n",
        "plot_model(modelCNN, to_file='/content/drive/MyDrive/modelCNN.png', show_shapes=True, show_layer_names=True)\n",
        "modelCNN.summary()\n",
        "\n",
        "historyCNN=modelCNN.fit(X_train, y_train, batch_size=64, epochs=20, validation_data=(X_valid, y_valid))\n",
        "modelCNN.save('/content/drive/MyDrive/modelCNN.h5')"
      ],
      "metadata": {
        "id": "F5gvHGvSSDA2",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "59b71a15-fd44-4921-f3f7-3162bf7dade9"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_8\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d_24 (Conv2D)          (None, 32, 32, 32)        896       \n",
            "                                                                 \n",
            " activation_40 (Activation)  (None, 32, 32, 32)        0         \n",
            "                                                                 \n",
            " max_pooling2d_24 (MaxPoolin  (None, 16, 16, 32)       0         \n",
            " g2D)                                                            \n",
            "                                                                 \n",
            " conv2d_25 (Conv2D)          (None, 16, 16, 64)        18496     \n",
            "                                                                 \n",
            " activation_41 (Activation)  (None, 16, 16, 64)        0         \n",
            "                                                                 \n",
            " max_pooling2d_25 (MaxPoolin  (None, 8, 8, 64)         0         \n",
            " g2D)                                                            \n",
            "                                                                 \n",
            " conv2d_26 (Conv2D)          (None, 8, 8, 128)         73856     \n",
            "                                                                 \n",
            " activation_42 (Activation)  (None, 8, 8, 128)         0         \n",
            "                                                                 \n",
            " max_pooling2d_26 (MaxPoolin  (None, 4, 4, 128)        0         \n",
            " g2D)                                                            \n",
            "                                                                 \n",
            " flatten_8 (Flatten)         (None, 2048)              0         \n",
            "                                                                 \n",
            " dense_16 (Dense)            (None, 512)               1049088   \n",
            "                                                                 \n",
            " activation_43 (Activation)  (None, 512)               0         \n",
            "                                                                 \n",
            " dropout_8 (Dropout)         (None, 512)               0         \n",
            "                                                                 \n",
            " dense_17 (Dense)            (None, 43)                22059     \n",
            "                                                                 \n",
            " activation_44 (Activation)  (None, 43)                0         \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 1,164,395\n",
            "Trainable params: 1,164,395\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n",
            "Epoch 1/20\n",
            "1632/1632 [==============================] - 9s 5ms/step - loss: 0.6988 - accuracy: 0.8193 - val_loss: 0.3335 - val_accuracy: 0.9170\n",
            "Epoch 2/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.1167 - accuracy: 0.9678 - val_loss: 0.2106 - val_accuracy: 0.9535\n",
            "Epoch 3/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.1048 - accuracy: 0.9727 - val_loss: 0.2274 - val_accuracy: 0.9583\n",
            "Epoch 4/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.0927 - accuracy: 0.9776 - val_loss: 0.2846 - val_accuracy: 0.9556\n",
            "Epoch 5/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.0835 - accuracy: 0.9805 - val_loss: 0.5551 - val_accuracy: 0.9615\n",
            "Epoch 6/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.0944 - accuracy: 0.9804 - val_loss: 0.6083 - val_accuracy: 0.9331\n",
            "Epoch 7/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.0969 - accuracy: 0.9816 - val_loss: 0.4834 - val_accuracy: 0.9463\n",
            "Epoch 8/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.0975 - accuracy: 0.9823 - val_loss: 0.6509 - val_accuracy: 0.9474\n",
            "Epoch 9/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.1009 - accuracy: 0.9832 - val_loss: 0.6037 - val_accuracy: 0.9569\n",
            "Epoch 10/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.0957 - accuracy: 0.9844 - val_loss: 0.6517 - val_accuracy: 0.9576\n",
            "Epoch 11/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.0858 - accuracy: 0.9865 - val_loss: 0.3893 - val_accuracy: 0.9615\n",
            "Epoch 12/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.0934 - accuracy: 0.9859 - val_loss: 0.9291 - val_accuracy: 0.9376\n",
            "Epoch 13/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.1016 - accuracy: 0.9859 - val_loss: 0.6439 - val_accuracy: 0.9628\n",
            "Epoch 14/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.1057 - accuracy: 0.9866 - val_loss: 1.2979 - val_accuracy: 0.9646\n",
            "Epoch 15/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.1133 - accuracy: 0.9873 - val_loss: 1.4378 - val_accuracy: 0.9467\n",
            "Epoch 16/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.1119 - accuracy: 0.9871 - val_loss: 1.8406 - val_accuracy: 0.9615\n",
            "Epoch 17/20\n",
            "1632/1632 [==============================] - 9s 5ms/step - loss: 0.1041 - accuracy: 0.9883 - val_loss: 1.6361 - val_accuracy: 0.9519\n",
            "Epoch 18/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.1292 - accuracy: 0.9873 - val_loss: 2.1549 - val_accuracy: 0.9596\n",
            "Epoch 19/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.1199 - accuracy: 0.9894 - val_loss: 0.7994 - val_accuracy: 0.9696\n",
            "Epoch 20/20\n",
            "1632/1632 [==============================] - 8s 5ms/step - loss: 0.1022 - accuracy: 0.9903 - val_loss: 0.5274 - val_accuracy: 0.9748\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import accuracy_score,recall_score\n",
        "results = modelCNN.evaluate(X_train, y_train, batch_size=32)\n",
        "results = modelCNN.evaluate(X_valid, y_valid, batch_size=32)\n",
        "results = modelCNN.evaluate(X_test, y_test, batch_size=32)\n",
        "\n",
        "pred1 = np.argmax(modelCNN.predict(X_test), axis=-1)\n",
        "print(accuracy_score(y_test, pred1))\n",
        "print(recall_score(y_test, pred1, average='micro'))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Meba41XctDSZ",
        "outputId": "6d5f560f-2209-40a3-d0fd-d8007b2c99ce"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "138/138 [==============================] - 0s 3ms/step - loss: 0.5274 - accuracy: 0.9748\n",
            "395/395 [==============================] - 1s 3ms/step - loss: 1.2261 - accuracy: 0.9639\n",
            "0.963895486935867\n",
            "0.963895486935867\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(historyCNN.history['accuracy'])\n",
        "plt.plot(historyCNN.history['val_accuracy'])\n",
        "plt.title('model accuracy')\n",
        "plt.ylabel('accuracy')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'val'], loc='upper left')\n",
        "plt.show()\n",
        "\n",
        "plt.plot(historyCNN.history['loss'])\n",
        "plt.plot(historyCNN.history['val_loss'])\n",
        "plt.title('model loss')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'val'], loc='upper left')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 573
        },
        "id": "JJfiGT9MdSKq",
        "outputId": "2ac2c98b-f108-429c-eba8-52d36712b8f5"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Model Mobile Net\n",
        "\n",
        "base_modelMN = tf.keras.applications.MobileNet(classes=43,input_shape = (32, 32, 3), include_top = False, weights = 'imagenet')\n",
        "\n",
        "modelMN = Sequential()\n",
        "modelMN.add(base_modelMN)\n",
        "modelMN.add(Flatten())\n",
        "modelMN.add(Dense(512, activation='relu'))\n",
        "modelMN.add(Dropout(0.6))\n",
        "modelMN.add(Dense(43, activation='sigmoid'))\n",
        "\n",
        "base_learning_rate = 0.001\n",
        "modelMN.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=base_learning_rate),\n",
        "              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False),\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "plot_model(modelMN, to_file='/content/drive/MyDrive/modelMN.png', show_shapes=True, show_layer_names=True)\n",
        "modelMN.summary()\n",
        "\n",
        "historyMN = modelMN.fit(X_train,y_train,epochs = 20,batch_size=64 , validation_data = (X_valid, y_valid))\n",
        "modelMN.save('/content/drive/MyDrive/modelMN.h5')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "S3I6yCBDAHcN",
        "outputId": "c4df5e35-2e7c-46aa-d3d0-fd47a0cb8847"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "WARNING:tensorflow:`input_shape` is undefined or non-square, or `rows` is not in [128, 160, 192, 224]. Weights for input shape (224, 224) will be loaded as the default.\n",
            "Model: \"sequential_11\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " mobilenet_1.00_224 (Functio  (None, 1, 1, 1024)       3228864   \n",
            " nal)                                                            \n",
            "                                                                 \n",
            " flatten_9 (Flatten)         (None, 1024)              0         \n",
            "                                                                 \n",
            " dense_18 (Dense)            (None, 512)               524800    \n",
            "                                                                 \n",
            " dropout_9 (Dropout)         (None, 512)               0         \n",
            "                                                                 \n",
            " dense_19 (Dense)            (None, 43)                22059     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 3,775,723\n",
            "Trainable params: 3,753,835\n",
            "Non-trainable params: 21,888\n",
            "_________________________________________________________________\n",
            "Epoch 1/20\n",
            "1632/1632 [==============================] - 50s 16ms/step - loss: 0.7320 - accuracy: 0.8006 - val_loss: 0.4994 - val_accuracy: 0.8819\n",
            "Epoch 2/20\n",
            "1632/1632 [==============================] - 25s 16ms/step - loss: 0.1643 - accuracy: 0.9605 - val_loss: 0.3065 - val_accuracy: 0.9256\n",
            "Epoch 3/20\n",
            "1632/1632 [==============================] - 25s 15ms/step - loss: 0.1188 - accuracy: 0.9731 - val_loss: 0.3231 - val_accuracy: 0.9293\n",
            "Epoch 4/20\n",
            "1632/1632 [==============================] - 26s 16ms/step - loss: 0.1072 - accuracy: 0.9769 - val_loss: 0.2217 - val_accuracy: 0.9549\n",
            "Epoch 5/20\n",
            "1632/1632 [==============================] - 25s 15ms/step - loss: 0.0827 - accuracy: 0.9825 - val_loss: 0.2542 - val_accuracy: 0.9549\n",
            "Epoch 6/20\n",
            "1632/1632 [==============================] - 25s 16ms/step - loss: 0.0765 - accuracy: 0.9846 - val_loss: 0.2554 - val_accuracy: 0.9578\n",
            "Epoch 7/20\n",
            "1632/1632 [==============================] - 26s 16ms/step - loss: 0.0619 - accuracy: 0.9878 - val_loss: 0.3887 - val_accuracy: 0.9483\n",
            "Epoch 8/20\n",
            "1632/1632 [==============================] - 25s 15ms/step - loss: 0.0504 - accuracy: 0.9895 - val_loss: 0.2857 - val_accuracy: 0.9467\n",
            "Epoch 9/20\n",
            "1632/1632 [==============================] - 25s 15ms/step - loss: 0.0740 - accuracy: 0.9867 - val_loss: 0.1813 - val_accuracy: 0.9717\n",
            "Epoch 10/20\n",
            "1632/1632 [==============================] - 27s 16ms/step - loss: 0.0435 - accuracy: 0.9911 - val_loss: 0.2128 - val_accuracy: 0.9585\n",
            "Epoch 11/20\n",
            "1632/1632 [==============================] - 26s 16ms/step - loss: 0.0378 - accuracy: 0.9932 - val_loss: 0.1598 - val_accuracy: 0.9683\n",
            "Epoch 12/20\n",
            "1632/1632 [==============================] - 25s 15ms/step - loss: 0.0313 - accuracy: 0.9937 - val_loss: 0.1486 - val_accuracy: 0.9710\n",
            "Epoch 13/20\n",
            "1632/1632 [==============================] - 25s 15ms/step - loss: 0.0424 - accuracy: 0.9924 - val_loss: 0.2762 - val_accuracy: 0.9646\n",
            "Epoch 14/20\n",
            "1632/1632 [==============================] - 25s 16ms/step - loss: 0.0289 - accuracy: 0.9944 - val_loss: 0.1428 - val_accuracy: 0.9723\n",
            "Epoch 15/20\n",
            "1632/1632 [==============================] - 25s 15ms/step - loss: 0.0617 - accuracy: 0.9917 - val_loss: 0.1361 - val_accuracy: 0.9710\n",
            "Epoch 16/20\n",
            "1632/1632 [==============================] - 25s 16ms/step - loss: 0.0312 - accuracy: 0.9944 - val_loss: 0.1156 - val_accuracy: 0.9721\n",
            "Epoch 17/20\n",
            "1632/1632 [==============================] - 25s 15ms/step - loss: 0.0442 - accuracy: 0.9916 - val_loss: 0.1643 - val_accuracy: 0.9689\n",
            "Epoch 18/20\n",
            "1632/1632 [==============================] - 25s 16ms/step - loss: 0.0157 - accuracy: 0.9968 - val_loss: 0.1318 - val_accuracy: 0.9737\n",
            "Epoch 19/20\n",
            "1632/1632 [==============================] - 25s 16ms/step - loss: 0.0234 - accuracy: 0.9949 - val_loss: 0.1258 - val_accuracy: 0.9753\n",
            "Epoch 20/20\n",
            "1632/1632 [==============================] - 25s 15ms/step - loss: 0.0288 - accuracy: 0.9955 - val_loss: 0.2184 - val_accuracy: 0.9587\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import accuracy_score,recall_score\n",
        "results = modelMN.evaluate(X_train, y_train, batch_size=64)\n",
        "results = modelMN.evaluate(X_valid, y_valid, batch_size=64)\n",
        "results = modelMN.evaluate(X_test, y_test, batch_size=64)\n",
        "\n",
        "pred1 = np.argmax(modelMN.predict(X_test), axis=-1)\n",
        "print(accuracy_score(y_test, pred1))\n",
        "print(recall_score(y_test, pred1, average='micro'))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qLu45Ksx1r3J",
        "outputId": "76ebc6c6-6e88-4ec6-f79a-49fe7e9fde01"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1632/1632 [==============================] - 11s 7ms/step - loss: 0.0632 - accuracy: 0.9858\n",
            "69/69 [==============================] - 1s 8ms/step - loss: 0.2184 - accuracy: 0.9587\n",
            "198/198 [==============================] - 2s 8ms/step - loss: 0.2570 - accuracy: 0.9444\n",
            "0.9444180522565321\n",
            "0.9444180522565321\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(historyMN.history['accuracy'])\n",
        "plt.plot(historyMN.history['val_accuracy'])\n",
        "plt.title('model accuracy')\n",
        "plt.ylabel('accuracy')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'val'], loc='upper left')\n",
        "plt.show()\n",
        "\n",
        "plt.plot(historyMN.history['loss'])\n",
        "plt.plot(historyMN.history['val_loss'])\n",
        "plt.title('model loss')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'val'], loc='upper left')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 573
        },
        "id": "2PbHuK9G1tWr",
        "outputId": "1fc6f759-568e-4957-967b-d754e658cfdd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#model Densnet121\n",
        "\n",
        "base_modelDN = tf.keras.applications.DenseNet121(classes=43,input_shape = (32, 32, 3), include_top = False, weights = 'imagenet')\n",
        "modelDN = Sequential()\n",
        "modelDN.add(base_modelDN)\n",
        "modelDN.add(Flatten())\n",
        "modelDN.add(Dense(512, activation='relu'))\n",
        "modelDN.add(Dropout(0.5))\n",
        "modelDN.add(Dense(43, activation='sigmoid'))\n",
        "\n",
        "base_learning_rate = 0.001\n",
        "modelDN.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=base_learning_rate),\n",
        "              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False),\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "plot_model(modelDN, to_file='/content/drive/MyDrive/modelDN.png', show_shapes=True, show_layer_names=True)\n",
        "modelDN.summary()\n",
        "\n",
        "historyDN = modelDN.fit(X_train,y_train,epochs = 10,batch_size=64 , validation_data = (X_valid, y_valid))\n",
        "modelDN.save('/content/drive/MyDrive/modelDN.h5')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iBC9shgFa0xl",
        "outputId": "ef3122ef-711c-4c02-aa43-1fbbc80fd562"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_14\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " densenet121 (Functional)    (None, 1, 1, 1024)        7037504   \n",
            "                                                                 \n",
            " flatten_11 (Flatten)        (None, 1024)              0         \n",
            "                                                                 \n",
            " dense_22 (Dense)            (None, 512)               524800    \n",
            "                                                                 \n",
            " dropout_11 (Dropout)        (None, 512)               0         \n",
            "                                                                 \n",
            " dense_23 (Dense)            (None, 43)                22059     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 7,584,363\n",
            "Trainable params: 7,500,715\n",
            "Non-trainable params: 83,648\n",
            "_________________________________________________________________\n",
            "Epoch 1/10\n",
            "1632/1632 [==============================] - 119s 65ms/step - loss: 0.3498 - accuracy: 0.9034 - val_loss: 0.2808 - val_accuracy: 0.9435\n",
            "Epoch 2/10\n",
            "1632/1632 [==============================] - 101s 62ms/step - loss: 0.0809 - accuracy: 0.9797 - val_loss: 0.2569 - val_accuracy: 0.9460\n",
            "Epoch 3/10\n",
            "1632/1632 [==============================] - 102s 63ms/step - loss: 0.0700 - accuracy: 0.9827 - val_loss: 0.2872 - val_accuracy: 0.9508\n",
            "Epoch 4/10\n",
            "1632/1632 [==============================] - 102s 62ms/step - loss: 0.0603 - accuracy: 0.9855 - val_loss: 0.2141 - val_accuracy: 0.9587\n",
            "Epoch 5/10\n",
            "1632/1632 [==============================] - 101s 62ms/step - loss: 0.0627 - accuracy: 0.9863 - val_loss: 0.2142 - val_accuracy: 0.9712\n",
            "Epoch 6/10\n",
            "1632/1632 [==============================] - 101s 62ms/step - loss: 0.0402 - accuracy: 0.9902 - val_loss: 0.3504 - val_accuracy: 0.9317\n",
            "Epoch 7/10\n",
            "1632/1632 [==============================] - 99s 61ms/step - loss: 0.0748 - accuracy: 0.9843 - val_loss: 0.1669 - val_accuracy: 0.9719\n",
            "Epoch 8/10\n",
            "1632/1632 [==============================] - 99s 61ms/step - loss: 0.0653 - accuracy: 0.9864 - val_loss: 0.1664 - val_accuracy: 0.9603\n",
            "Epoch 9/10\n",
            "1632/1632 [==============================] - 99s 61ms/step - loss: 0.0130 - accuracy: 0.9963 - val_loss: 0.1989 - val_accuracy: 0.9692\n",
            "Epoch 10/10\n",
            "1632/1632 [==============================] - 100s 61ms/step - loss: 0.0189 - accuracy: 0.9951 - val_loss: 0.2856 - val_accuracy: 0.9571\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "results = modelDN.evaluate(X_train, y_train, batch_size=64)\n",
        "results = modelDN.evaluate(X_valid, y_valid, batch_size=64)\n",
        "results = modelDN.evaluate(X_test, y_test, batch_size=64)\n",
        "\n",
        "pred1 = np.argmax(modelDN.predict(X_test), axis=-1)\n",
        "print(accuracy_score(y_test, pred1))\n",
        "print(recall_score(y_test, pred1, average='micro'))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ESXJskQ0hg82",
        "outputId": "bf849f95-d4ba-495d-94d9-16769c875c35"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1632/1632 [==============================] - 28s 17ms/step - loss: 0.0192 - accuracy: 0.9941\n",
            "69/69 [==============================] - 1s 17ms/step - loss: 0.2856 - accuracy: 0.9571\n",
            "198/198 [==============================] - 4s 21ms/step - loss: 0.2466 - accuracy: 0.9503\n",
            "0.95027711797308\n",
            "0.95027711797308\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(historyDN.history['accuracy'])\n",
        "plt.plot(historyDN.history['val_accuracy'])\n",
        "plt.title('model accuracy')\n",
        "plt.ylabel('accuracy')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'val'], loc='upper left')\n",
        "plt.show()\n",
        "\n",
        "plt.plot(historyDN.history['loss'])\n",
        "plt.plot(historyDN.history['val_loss'])\n",
        "plt.title('model loss')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'val'], loc='upper left')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 573
        },
        "id": "fNGIyHP55Biu",
        "outputId": "e24ddc70-c462-423d-ec58-c42dc0417596"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deXiU5dX48e/JRggkARIIkLDvi6wRtWrFta6oaN032qrV2mrf2rd2s62t1bc/3/atra1bVdwX3C0uqOBSN9ZEENlkySQQlpBJQhLIcn5/3E9giAEmMJNnkjmf6+Ji8iwzJ6PMmXs7t6gqxhhjTHMJfgdgjDEmNlmCMMYY0yJLEMYYY1pkCcIYY0yLLEEYY4xpkSUIY4wxLbIEYQwgIo+IyB/CvHadiJwU7ZiM8ZslCGOMMS2yBGFMByIiSX7HYDoOSxCm3fC6dn4qIoUiskNE/iUiOSLyuohUisjbItI95PppIrJMRMpFZJ6IjAo5N1FEFnn3PQOkNnutM0VkiXfvRyIyLswYzxCRxSJSISJFIvLbZueP8Z6v3Dt/lXe8s4j8r4isF5GgiHzoHZsqIoEW3oeTvMe/FZFZIvK4iFQAV4nIFBH52HuNjSLydxFJCbl/jIjMEZEyESkVkV+ISG8RqRaRrJDrJonIFhFJDud3Nx2PJQjT3pwHnAwMB84CXgd+AfTE/f/8IwARGQ48BdzknZsNvCoiKd6H5UvAY0AP4DnvefHunQg8BFwLZAH3Aa+ISKcw4tsBXAF0A84ArhORc7znHeDF+zcvpgnAEu++u4DJwDe8mP4baAzzPTkbmOW95hNAA/BjIBs4CjgRuN6LIR14G3gD6AsMBd5R1U3APOCCkOe9HHhaVevCjMN0MJYgTHvzN1UtVdVi4APgU1VdrKq1wIvARO+6C4F/q+oc7wPuLqAz7gP4SCAZ+D9VrVPVWcD8kNe4BrhPVT9V1QZVnQns9O7bL1Wdp6qfq2qjqhbiktRx3ulLgLdV9Snvdbep6hIRSQC+A9yoqsXea36kqjvDfE8+VtWXvNesUdWFqvqJqtar6jpcgmuK4Uxgk6r+r6rWqmqlqn7qnZsJXAYgIonAxbgkauKUJQjT3pSGPK5p4eeu3uO+wPqmE6raCBQBud65Yt27UuX6kMcDgJ94XTTlIlIO9PPu2y8ROUJE5npdM0Hg+7hv8njPsaaF27JxXVwtnQtHUbMYhovIayKyyet2+mMYMQC8DIwWkUG4VlpQVT87yJhMB2AJwnRUJbgPegBERHAfjsXARiDXO9akf8jjIuB2Ve0W8idNVZ8K43WfBF4B+qlqJnAv0PQ6RcCQFu7ZCtTu49wOIC3k90jEdU+Fal6S+Z/Al8AwVc3AdcGFxjC4pcC9VtizuFbE5VjrIe5ZgjAd1bPAGSJyojfI+hNcN9FHwMdAPfAjEUkWkenAlJB7HwC+77UGRES6eIPP6WG8bjpQpqq1IjIF163U5AngJBG5QESSRCRLRCZ4rZuHgD+LSF8RSRSRo7wxj5VAqvf6ycCvgAONhaQDFUCViIwErgs59xrQR0RuEpFOIpIuIkeEnH8UuAqYhiWIuGcJwnRIqroC9034b7hv6GcBZ6nqLlXdBUzHfRCW4cYrXgi5dwFwNfB3YDuw2rs2HNcDt4lIJXArLlE1Pe8G4HRcsirDDVCP907fDHyOGwspA/4HSFDVoPecD+JaPzuAvWY1teBmXGKqxCW7Z0JiqMR1H50FbAJWAceHnP8PbnB8kaqGdruZOCS2YZAxJpSIvAs8qaoP+h2L8ZclCGPMbiJyODAHN4ZS6Xc8xl/WxWSMAUBEZuLWSNxkycGAtSCMMcbsQ9RaECLykIhsFpGl+zgvInK3iKwWVzphUsi5K0VklffnymjFaIwxZt+i1oIQkW8CVcCjqjq2hfOnAz/Ezeo4Avirqh4hIj2ABUA+bn73QmCyqm7f3+tlZ2frwIEDI/tLGGNMB7dw4cKtqtp8bQ0AUav8qKrvi8jA/VxyNi55KPCJiHQTkT7AVGCOqpYBiMgc4FRcyYJ9GjhwIAsWLIhE6MYYEzdEZJ/Tmf0cpM5l7xIBAe/Yvo5/jYhcIyILRGTBli1bohaoMcbEo3Y9i0lV71fVfFXN79mzxRaSMcaYg+RngijG1cZpkucd29dxY4wxbcjP3adeAW4Qkadxg9RBVd0oIm8Cf5Q9G7+cAvz8YF6grq6OQCBAbW1tZCKOYampqeTl5ZGcbHu7GGMiI2oJQkSewg04Z3s7Yv0GV4MfVb0Xt4HL6bg6N9XADO9cmYj8nj31+W9rGrBurUAgQHp6OgMHDmTvwp0di6qybds2AoEAgwYN8jscY0wHEc1ZTBcf4LwCP9jHuYdw1S0PSW1tbYdPDgAiQlZWFjZQb4yJpHY9SB2Ojp4cmsTL72mMaTt+jkEYY0y7VRgo5/2VW8jonEy3tBS6pyXTPS2Fbt7faSmJ7f6LmyWIKCsvL+fJJ5/k+uuvb9V9p59+Ok8++STdunWLUmTGmIMRrKnjrjdX8Pin69lfIYqUxITdyaLp7+5d9iQT93fo42QyOyeTlBg7HTuWIKKsvLycf/zjH19LEPX19SQl7fvtnz17drRDM8a0gqrySkEJv39tOWU7dnLVNwZy44nDqGtQyqt3sb26ju3Vu/Z+vKPpWB1rtlSxfX0d5dW7qG/cd2bJSE2ie5eUFlsloYmlW1oy3bu4Y52To9NasQQRZbfccgtr1qxhwoQJJCcnk5qaSvfu3fnyyy9ZuXIl55xzDkVFRdTW1nLjjTdyzTXXAHtKh1RVVXHaaadxzDHH8NFHH5Gbm8vLL79M586dff7NjIkfa7fu4NcvLeXD1VsZn5fJIzMOZ2xu5u7zPdMPtAvsHqpK1c56yr0ksr3aJY3tO0Iee+e2Ve1i9eYqyqvrqNpZv8/nnNCvGy/94OhD+h1bEjcJ4nevLuOLkoqIPufovhn85qwx+73mzjvvZOnSpSxZsoR58+ZxxhlnsHTp0t3TUR966CF69OhBTU0Nhx9+OOeddx5ZWVl7PceqVat46qmneOCBB7jgggt4/vnnueyyyyL6uxhjvq62roF731vDP+atoVNiAr8/ewyXHDGAxISD/7YuIqSnJpOemky/Hmlh37ervpHyGtcaaZ5MuqVFZ/1T3CSIWDFlypS91ircfffdvPjiiwAUFRWxatWqryWIQYMGMWHCBAAmT57MunXr2ixeY+LVh6u28uuXl7J26w6mje/Lr84YRa+MVN/iSUlKoFd6Kr3S2y6GuEkQB/qm31a6dOmy+/G8efN4++23+fjjj0lLS2Pq1Kktrvru1GlP8zUxMZGampo2idWYeLS5spbb/72cl5eUMDArjce+O4Vjh8Vnrbe4SRB+SU9Pp7Ky5d0bg8Eg3bt3Jy0tjS+//JJPPvmkjaMzxjRpaFSe/GwDf3rjS3bWNXLjicO4buoQUpMT/Q7NN5YgoiwrK4ujjz6asWPH0rlzZ3JycnafO/XUU7n33nsZNWoUI0aM4Mgjj/QxUmPi19LiIL98aSkFReUcPTSL3589lsE9u/odlu86zJ7U+fn52nzDoOXLlzNq1CifImp78fb7GnOoqnbW8+e3VvLIR2vp0SWFX585mmnj+7b7BW6tISILVTW/pXPWgjDGxB1V5Y2lm/jdq19QWlnLpUf056enjCQzSrOB2itLEMaYuFJUVs2tLy9l7ootjO6TwT8vm8TE/t0PfGMcsgRhjIkLu+obeeCDr/jbu6tIFOHXZ47myqMGxFRpi1hjCcKYGFa1s56F67ezeMN2hvTsysmjc+J6Vs3B+vSrbfzypaWs3lzFaWN7c+tZo+mTadUIDsQShDExJFhdx/x1ZXy6dhufrS1jaUkFDSF1e9I7JXHGuD5Mn5TH4QO7x9Vg6sEo27GLO2Yv57mFAfK6d+ahq/I5YWTOgW80gCUIY3y1tWon89eW8an358tNFai6SqAT+nXj+qlDmDKoBxP6daMwEOT5RQFeKSjh6flF9OvRmekT85g+KZcBWV0O/GJxpLFReW5hEXe8/iVVtfVcN3UIPzphGJ1TrPXVGpYgYkzXrl2pqqryOwwTJZuCtXy6dhufri3js7VlrN7s/lunJicweUB3fnzS8N0JoXlX0tFDszl6aDa/P7ueN5dt4oVFxdz97ir++s4q8gd0Z/qkPM4Y14fMzvE9E2fFpkp+9dLnzF+3nSkDe/CHc8cyPCfd77DaJUsQxkSJqhLYXuNaB19t47N1ZazfVg1A105J5A/sznmT8pgyqAeH5WaSkhTeYGmXTklMn5TH9El5bAzW8NLiEp5fFOAXL37Ob19dxsmjczhvUi7HDutJchwNwFbvqufud1bz4AdfkZ6axJ/OH8e3J+dZN9whsAQRZbfccgv9+vXjBz9w22//9re/JSkpiblz57J9+3bq6ur4wx/+wNlnn+1zpOZQqSpfbd3Bp1+V8Zk3hlASdLW1uqUlc/jAHlx+5ACOGJTF6L4Zh1QRtEmfzM5cN3UI3z9uMJ8XB3lhUTGvFJTw78KNZHdNYdr4XKZPymVM34wO/UH5zvJSbn15GcXlNVyQn8ctp42iR5cUv8Nq9+JnJfXrt8CmzyP7or0Pg9Pu3O8lixcv5qabbuK9994DYPTo0bz55ptkZmaSkZHB1q1bOfLII1m1ahUickhdTLaSum01NiorN1d6CcGNIWyt2glAdtdOHDGoB0cM7sGUQT0Y3iudhAgkhHDsqm/kvZVbeGFRgHeWb2ZXQyMjctI5b3IuZ0/IJcfHiqSRVlJew+9eXcaby0oZntOVP5xzGFMG9fA7rHbFVlL7aOLEiWzevJmSkhK2bNlC9+7d6d27Nz/+8Y95//33SUhIoLi4mNLSUnr37u13uGY/6hsaWb6xcvcYwvx1ZZRX1wHQJzOVY4ZmccTgLKYM6sHg7C6+fWNPSUrg5NE5nDw6h/LqXbxWuJHnFwX44+wvufP1LzlmWE/Om5TLKaN7t9tB2/qGRh75aB1/nrOSRlV+dupIvnvMoLC76Ux44idBHOCbfjR9+9vfZtasWWzatIkLL7yQJ554gi1btrBw4UKSk5MZOHBgi2W+TdtRVXbWN1K1s54dO+u9vxvYsbOe5Zsq+GxtGQvWbd+9q9eArDROHpXDEYOzOGJQD/K6d47JLpxuaSlcduQALjtyAF9tqeLFxcW8sKiYG59eQtdOSZw2tjfnTc5jysAebdbCaa2GRmVjsIYNZdUUlVWzoayad5Zv5stNlZwwshe/mzamVRvvmPDFT4Lw0YUXXsjVV1/N1q1bee+993j22Wfp1asXycnJzJ07l/Xr1/sdYrtU19C414d504f7nmP17NjV0OxDfx/X7mrYa71Bc0N7dWXahL6u22hQFr0z2183zeCeXfnJKSP48UnD+WxdGS8sCjD78008tzBAbrfOTJ+Uy7kTc32pYlpZW7dXAnB/aigqqyawvZq6hj3/bZIShME9u3DvZZP51picmEzMHYUliDYwZswYKisryc3NpU+fPlx66aWcddZZHHbYYeTn5zNy5Ei/Q4wJDY3Kum07WLmpki83VbIxWLPXh7n7IN/zAb+rvjGs501KELp0SqJrpyS6dEqkS6ck0lOT6J2R6h13x/Zcs/ex/j3SyO4a/p7DsS4hQThycBZHDs7id9PG8tYXbsrsPXNX87d3VzOxfzemT8rjrHF96JYWmYHelloBG8rczxu27WC711XXpFtaMv17pDG6bwanju1N/x5pu//0yUy18hhtJH4GqeNAe/l9VZVNFbWs2FTp/pS6v1dvrmKn96EvAr3SO9E15EO7+Yd815Q9x9KajnVKokvK3td1Skqwb5lhKK2o5eUlxTy/sJgVpZWkJCZw4qheTJ+Ux9QRB54yu69WwIZtOygur/laKyC3e2f690ijX480Bngf/v28P/G+lqMt2SC18U159S5WbKpkZalrFaz0kkFFbf3ua3IyOjE8J50rjhrA8Jx0RvbOYGivru12ALW9yslI5ZpvDuHqYwfzxcYKXlhUzMtLinl96SZ6dElh2vi+nDmuD7saGncngfXb9iSElloBA3qkMTY3k9MP67O7BdDPWgHthiUIExE1uxpYvbmKLzdVuCRQWsWKTRWUVuzcfU16ahIjctI5a3xfRvROZ0ROOsNz0ulu89Vjiogwpm8mY/pm8vPTRvLBqq08vyjAk59t4JGP1u2+LilByOvemX490r6WAKwV0DF0+AShqnHRvdBWXYX1DY2s21Yd0jVUwcrSKtZt20FTCClJCQzr1ZWjh2Qzonc6w71k0CczNS7+W3QkSYkJHD+yF8eP7EWwpo6PVm8l0xsf6J1hrYCOrkMniNTUVLZt20ZWVlaH/mBSVbZt20ZqauRm1qgqJcFaVoaMEazYVMnqLVW7B4cTBAZmdWFETjrTmloFvdMZ0CPNPjg6oMzOyZx2WB+/wzBtqEMniLy8PAKBAFu2bPE7lKhLTU0lLy/voO7dWd/AqtIqlhYHWVZSwRcbK1i5qZLKnXvGCXpnpDKidzrHDMtmRI5LBEN7dbW9CYzpwDp0gkhOTmbQoEF+hxFTduysZ/nGit3JYFlJBas2V+6eYdK1UxKj+qRz9sS+jOid4ZJBTrrt1WtMHOrQCSLebd+xi2UlFSwt8ZJBcZC1IWMFWV1SGN03g+NGDGZM3wzG9M1kQI+0mF1Ra4xpW5YgOoCmdQVLiytYFpIMmiqJAuR268yYvhmcPcFV9hybm0lORqcOPTZjjDk0UU0QInIq8FcgEXhQVe9sdn4A8BDQEygDLlPVgHfuT8AZQAIwB7hRO8qqvkPQ2KisL6sO6SJyf5ft2AW4BWaDsruQP7DH7lbBmL4ZNpXUGNNqUUsQIpII3AOcDASA+SLyiqp+EXLZXcCjqjpTRE4A7gAuF5FvAEcD47zrPgSOA+ZFK95YVNfQyKrSqj2tgpIgyzdW7i4Yl5woDM9J56RRvRjTN5OxuRmM7J1Bl07WMDTGHLpofpJMAVar6lcAIvI0cDYQmiBGA//lPZ4LvOQ9ViAVSAEESAZKoxir71SVZSUVLC4q54uSIEuLK1hRWrl7Smnn5ERG983YvfnLmL6ZDM9Jt/LGxpioiWaCyAWKQn4OAEc0u6YAmI7rhjoXSBeRLFX9WETmAhtxCeLvqrq8+QuIyDXANQD9+/eP/G/Qhj5bW8aF938CuBIFY/pmcNU3Bu5OBoOyu0RkBzJjjAmX330RNwN/F5GrgPeBYqBBRIYCo4Cmif1zRORYVf0g9GZVvR+4H1yxvjaLOgoWrN8OwDs/Oc7XzWaMMaZJNBNEMdAv5Oc879huqlqCa0EgIl2B81S1XESuBj5R1Srv3OvAUcBeCaIjWVJUzuCeXRjiQy1+Y4xpSTQ7sOcDw0RkkIikABcBr4ReICLZItIUw89xM5oANgDHiUiSiCTjBqi/1sXUkRQGyhmf183vMIwxZreoJQhVrQduAN7Efbg/q6rLROQ2EZnmXTYVWCEiK4Ec4Hbv+CxgDfA5bpyiQFVfjVasftsUrKW0Yifj8jL9DsUYY3aL6hiEqs4GZjc7dmvI41m4ZND8vgbg2mjGFksKAuUAjO9nLQhjTOywOZIxoDBQTlKCMLpPht+hGGPMbpYgYkBhIMiI3ulWGdUYE1MsQfhMVSkoKrfuJWNMzLEE4bN126qpqK1nvA1QG2NijCUInxV6A9TjbIqrMSbGWILw2ZKicjonJzKsly2QM8bEFksQPisMBBmbm2F7OBtjYo59KvmorqGRZSVB614yxsQkSxA+WllaSW1do81gMsbEJEsQPioMBAFsBpMxJiZZgvBRYaCcbmnJ9O+R5ncoxhxYXQ28ezuULvM7EtNGLEH4aElRkMNyM23vh1i2a4f7YDTw+XPw/p/ggRNg/oNgW8R3eJYgfFKzq4GVpZVMsPGH2LX6HfjLGHhuht+RxIaFMyFrKAw8Bv79E3jmMqgu8zsqE0WWIHzyxcYgDY1qM5hiUWMjvH8XPH4e1O+Ela9D2Vq/o/LXpqVQvADyvwOXPAen3A4r34R7j4F1//E7OhMlliB8sqTIBqhjUm3QfTN+9/dw2Plw7QcgCbBopt+R+WvRTEhMgfEXQ0ICfOMG+N4cSOoEM8+EuXdAQ73fUZoIswThk8JAOX0yU+mVkep3KKZJ6Rdw//Gw6k049X9g+gOQPRSGnwqLH4f6XX5H6I+6Gih8BkZNg7Qee473nQjXvg/jLoT37oSZZ0F5kX9xmoizBOGTwkDQdpCLJUufhwdPhF1VcOVrcOT3oWnyQP53YMcWWPFvf2P0yxcvu5bV5Ku+fq5TOpx7L5x7P2wqdF1OX7zy9es6muoyePOXrhty+zq/o4kaSxA+CFbXsXbrDht/iAUNdfDGL2DWd6D3OPeNeMBRe18z5ATI7A8LHmr5OTq6hY9AjyFucHpfxl/o3rseg+DZy+G1H3fM2V/1O+Gjv8PdE+CTf8D6j+G+b8KXHfPLgyUIHxQWuwquNoPJZ1Wb4dFz4JN7YMq1cOWrkN7769clJMLkK2Dt+7BtTdvH6actK2DDxzDpij0tqn3JGgLfeQu+8SOXTO8/3nXbdQSqsOxFuGcKvPVLyM2H738I1/0Hug+Cpy9xLYqGOr8jjShLED5oWkE9Nte6mHxTNN998yte6LpHTv8TJKXs+/qJl0NCEix8uO1ijAWLHoWEZJhwaXjXJ6XAKb+Hy16A6q3wwPEw/1/te81E0Wfwr1PguasguQtc9jxc/gLkjHEtpu++BYdfDR//HR4+HYIBvyOOGEsQPigoKmdwdhcyOyf7HUr8UYXPHoCHT3MzcL43x3WPHEh6bxhxGix50nUzxIP6ne73HXk6dO3ZunuHngjXfQQDjoZ//5frdmpvaya2r3NJ4V8nQ/l6mPY3+P4HMPSkva9L6gRn3AXnPwybl7txmJVv+RFxxFmC8EFBwLYY9UVdDbx0Hcy+2Y0rXDMPeh8W/v2TZ0D1Nlj+arQijC3LX4WaMph05cHd37UXXDoLTvkDrHgD7j0W1n8U2RijoWa76y76++Furcdxt8APF7lutoT97Bs/drr7fyojF578Nrz923Y/9dcSRBsrrailtGKnzWBqa9vXuW+CBU/D1F/AxU9D5+6te47Bx0P3gbAgTrqZFs2Ebv3d732wEhLgGz903TBJKfDIGTDvf6CxIXJxRkr9Lvjkn3D3RPj4Hhh3AfxwIRz/c+gU5oZe2UPhe2+7pPrhX+DRaVCxMbpxR5EliDZWUGRbjLa5VW/DfcdB+Qa45FmY+jP3wdVaCQnuH/76D2HLysjHGUu2rXGD8pOuOLj3qrncSW6W02Hfhnl/dGsmYqWvXtVNzf3HEfDGLdBnvOtKOvseyOjb+udL7gzT7oZz74OSxa7Lac3cyMfdBixBtLGCQDlJCcKYvhl+h9LxNTbCe3+CJ86HzH6u+T/8lEN7zomXeYPVj0QgwBi26FGQRJhwWeSes1M6TL/ffXBuLIB/Hg3LX4vc8x+MwEI3HvXs5W6l+KWz4PKXWtf1uC/jL4Kr50KXbHjsXLfaPBZbTvthCaKNFQaCjOidTmryfvoyzaGrKXdTD+fe7roKvvsW9Bh86M/btReMPBMKnoS62kN/vlhUvwuWPOFWkGf0ifzzj7/ItSa6D4RnLnWF/9p6zcT29W7ty4MnwLbVcOb/wff/A8NOPvB03tboNRKufnfPavPHznXTq9sJSxBtSFUpKCq37qVoK13mpleungOn/T/3jTUlgntu5M9wA5lfvBy554wlK193K8cnH+TgdDiyhsB357jxifkPuhLim5dH7/Wa1JTDW7+Gv+fDl7Phmz+FHy12/00Tk6Lzmild3GrzaX+Dok/dYP26D6PzWhFmCaINrdtWTUVtPRP62QB11BQ+Bw+cCLuq4ap/wxHXRPYbIcDAb7qVxR11ZfXCmW4mTvPpnJGWlOJmOF36vEtI9x/vJgBEY81EQx18ep8bgP7obzD2fDcAfcKvXNdXtIm48ZzvveMSxsyz4IP/dd2gMcwSRBsqDNgAddQ01MHrP4MXvreniFz/I6PzWgkJri5R0Sdt8623LW1fD2ve9RYGtlE36LCTXPdO/yPhtZvguStdCy0SVN04xz1HwOv/Db3HwrXvwbn/hMzcyLxGa/Qe68bCRp8D79wGT14AO7a1fRxhsgTRhgqKgqQmJzCsV5hT5kx4Kje5b2Sf3gtHXAdXvgLpOdF9zQmXukHNjjbldfFj7u+JERycDkd6jlt9ffJtrq7RvcfChk8O7TmLF7lptc9c6iYWXPIsXPGKm6Xkp9QMOP8hOP0uWPse3HcsbPjU35j2wRJEGyoIlHNYbiZJifa2R8yGT9wU1o0FcN6/4LQ7IbENVqh3yXLlrwuedt1ZHUFDvStrPvQk6Nav7V8/IQGOvtFNKEhIcrOL3vtT62f+lG+A569241BbVsAZf3aruod/K/LdjQdLBKZcved3feR01/UVYyVJ7JOqjdQ3NLKsJGjdS5Gi6vqUHznDDUB/7223wU9byp8BO4OuiFtHsOotqNzYclnvtpQ72XURjj3fzUKbOQ2CxQe+rzboVi//LR+WvwLH/sQNQB/+3egNQB+qpu7Q4afCW7+Cpy+NXPdaBFiCaCMrS6uorWu0FdSRsKsaXrzW9SkPPdnNNc8Z0/ZxDDgasod3nAJ+i2ZC1xz3TdtvqRluzcQ593qLzY7ed0nthjpXX+vuiW718phz4YYFcOKt7nliXeducOHj8K073GZVTUUkY0BUE4SInCoiK0RktYjc0sL5ASLyjogUisg8EckLOddfRN4SkeUi8oWIDIxmrNFW4A1Qj7cWxKEp+8qVzCh8Fo7/FVz0pPsH5gcR9207MN/t2dyeBYtdC2LCpW3TRRcOEZhwsfuG3a2/W9cy+6d71p+oworX4R9HufpaPUe5AeDp9/nTRXYoROCo62HGG25m07++BZ/e73uXU9QShIgkAvcApwGjgYtFZHSzy+4CHlXVccBtwB0h5x4F/p+qjgKmAO1ndUkLCgPlZHZOZkBWBOfjx5uVb8L9U12JhktnwXE/jUwZiEMx/mJI7NT+WxGLHwdtdFMxY032ULdm4qgb4LP73ZqJ5a+6iQlPXSuXIbkAAB0eSURBVOSuuegpuOo112XTnvU73JX5GHICvP5TV022tsK3cKL5r2sKsFpVv1LVXcDTwNnNrhkNvOs9ntt03kskSao6B0BVq1S1XY8EFhS5LUYlVgbJ2pPGRlem4MkL3DfJa99zUyNjQVoPGHOOa9Hs2uF3NAenscHNXho81e1vEIuSOsG3bndfDKpK4ZnLYPMXbibQ9R+7kuQd5d9WWg9XTPKk37lEeP9xsLHQl1CimSBygdAdzAPesVAFwHTv8blAuohkAcOBchF5QUQWi8j/81okexGRa0RkgYgs2LJlSxR+hcio2dXAitJK6146GDXb4SmvTMH4S9w3ye4D/Y5qb5NnwM4Kt691e7RmLgSLDr6sd1sadrLbxe3Mv7gB6ClXx06XWCQlJMAxN7lWUV0NPHiSq//Vxl1Ofg9S3wwcJyKLgeOAYqABSAKO9c4fDgwGrmp+s6rer6r5qprfs2crNzRpQ19sDNLQqLYHRGtt+tx1Ka2ZC2f8L5zzD1cpM9b0P9L1f7fXNRELH4a0bFdjqj1I7w3534HUOJjwMeAbcO0H7u9Xb4QXroGdVW328mElCO+b/Bki0pqEUgyEjhTlecd2U9USVZ2uqhOBX3rHynGtjSVe91Q98BIwqRWvHVMKitwWo+NtBlN46mpdXf4HT3a7ms2YDYd/L3a7EETclNeSRVCyxO9oWqeyFFa+4QaD97flqvFP155um9PjfwmfP+fWd7TRCv5wP/D/AVwCrBKRO0VkRBj3zAeGicggEUkBLgJeCb1ARLJDks7PgYdC7u0mIk3NghOAdrv7eUGgnN4ZqfTKSPU7lNhWV+tmbtw9wdXl73+km8HSb4rfkR3YuAshqXP7G6xe8gQ01reP7qV4lpAIx/03XPGS63a9/3i3HWy0Xzaci1T1bVW9FPctfh3wtoh8JCIzRKTFDkDvm/8NwJvAcuBZVV0mIreJyDTvsqnAChFZCeQAt3v3NuC6l94Rkc8BAR44yN/Rd4WBIOOtQN++hSaG138K3Qe5kgiXv+jKa7cHnbu5LSc/nwU7K/2OJjyNjW7tw4BjIHuY39GYcAyeCt//0C0mfOk6ePkHUV3JH/byQm/w+DLgcmAx8ARwDHAl7oP+a1R1NjC72bFbQx7PAmbt4945wLhw44tVweo61m7dwfmT8w58cbypq3Ub03z4Z7eCt/833OKogcfGbnfS/kye4b6Rf/6c6yOPdeved1uxHv9LvyMxrZHeG654GebdAR/cBcWL4YKZUUny4Y5BvAh8AKQBZ6nqNFV9RlV/CFjluf0oLLYFcl9TV+uVXg5pMVz5qhtrGPTN9pkcAPLyIWds9EpWR9rCmZDazdWUMu1LYhKc+GtXKr1yoyvREYXS4eG2IO5W1RY3VVXV/AjG0+EUBtwA9WE2QO21GGa6cgiVG12pivbcYmiuaWX17JvdgHXuZL8j2rcdW90c+8O/B8k2NtZuDTvJLazbsTUqi0bDfcbRIrL7K7CIdBeR6yMeTQdUUFTO4OwuZHbugHO1w7VXi+G/3dafV77qNvRpzy2Gloy7AJLTYn/Ka8FT0FgX3V3jTNvIzIO+E6Ly1OEmiKu96acAqOp24OqoRNTBFATK47dAX10NfHIv/HV8SGJ4rf13Je1PaiaMPc8tmqsN+h1Ny1Rd91K/I6DXKL+jMTEs3ASRKCE1IrxVzTZp+gBKK2oprdgZfwvkdieGCfDGz9z+w7sTw7F+Rxd9+TOgrtqV34hF6z+Cbatsaqs5oHDHIN4AnhGR+7yfr/WOmf0oKIqzLUbratw30w//AlWb3PTJ8x6Mj6QQqu8k6D3OdTPF4gK/RTOhU4arIWXMfoSbIH6GSwrXeT/PAR6MSkQdSEGgnKQEYUzfdlCT/lBYYtibiJvm+tpNrhR4LC30qy6DZS/BpMshpYvf0ZgYF1aCUNVG4J/eHxOmwkCQEb3TSU1uo83f21pdjSsg9uFfXIXNgcfGd2IIddj5boewBQ/HVoIofBYadlr3kglLWAlCRIbh9moYDeyeE6eqg6MUV7unqhQUlXPGuL5+hxJ5LSaGf1liCNUpHQ77tpstdOofoXN3vyNyg9OLZro9E/q0+zWopg2EO0j9MK71UA8cj9vM5/FoBdURrNtWTUVtfccq0FdX44ro/XW8q5WUPdwNPl/1miWHluTPgPpaKHjG70icwAK3h4Lfe06bdiPcMYjOqvqOiIiqrgd+KyILgVsPdGO8KmzaYrQjzGBqqcVw/kMw8Bi/I4ttfca7AeuFD8MR1/o/WL3wEUju4qbhGhOGcBPETq/q6ioRuQFXtttKbOxHQVGQ1OQEhvVqx29TXY3rQ//P/1liOFj5M+CVH8KGT2DAUf7FUVsBy15wYyOd0v2Lw7Qr4XYx3Yirw/QjYDKuaJ+Ncu1HQaCcsX0zSUr0e0+mg1BXAx//w3Ulvflz15V01b9dV5Ilh9YZe56bUup3GfDPn3NrM6x7ybTCAVsQ3qK4C1X1ZqAKmBH1qNq5+oZGlpUEuWTKAL9DCV9DPRR94jaPKXzWWgyRktLFld9Y9Biceqfbb9gPCx+BnMNcl5cxYTpgglDVBhGxT4hWWFlaRW1dY+zvAVGzHVa/45LCqjlQWw4JyTDkeDj6RksMkTJ5Bsx/0M1oOuoHbf/6JYthUyGcfpf/4yCmXQl3DGKxiLwCPAfsaDqoqi9EJap2riAQwyW+t652CWHlG67kgjZ4+xGfAcO/BUNOsD7qSOs9FvIOd+M5R17f9h/SC2e63e4O+3bbvq5p98JNEKnANtzWn00UsATRgsJAOZmdkxmQleZ3KK7raMPHe5LCttXueK8xrpUw4jRXljqhgy7mixWTZ8DL18P6/7Rty2xnlRt/GHOu2/XOmFYIdyW1jTu0QkFRkHF5mYhfzfmmrqMVr8PqOa6qaGKK+2Cacq1rKXRvR+MjHcGYc92A/4I2HtNZ9gLsqrKy3uaghLuS+mFci2EvqtoO9lVsWzW7GlhRWsl1I4e07QtvXeVaCCvecC2G3V1HZ8LwU924gnUd+SclDcZfDPP/5TZ36ZLdNq+7cCb0HOlKexvTSuF2Mb0W8jgVOBcoiXw47d8XG4M0NGr094BoqHNz61e+4VoKZWvc8V5j4JibYHhT11E7nGbbUU2eAZ/e6/atPvrG6L/epqVQvAC+dYcNTpuDEm4X0/OhP4vIU8CHUYmonSsocpvETIjGCuqa7bDqbVj5Oqx+O6Tr6Fg48jrXddStf+Rf10RGr5HQ/yg35fSoH0Y/eS+a6f7/GH9RdF/HdFjhtiCaGwb0imQgHUVhoJzeGan0yojAPr+qblB5xeuupbDhE9d11KUnjDwLRpwKg6da11F7MnkGvHgNrHvf/beLll3VUPgMjD7bv7UXpt0Ldwyikr3HIDbh9ogwzRQEgofWvdRQ58YQVrzhWgplX7njOWPhmB+78QTrOmq/Rp/tdtlb8HB0E8QXL7sWppX1Nocg3C4m+4oahmBNHWu37uD8yXnh3dBQB2VrYetK2LoCNhbCmrmw0+s6GvRNN2/euo46juRUGH8JfHYfVG2GrlFqiC+aCT2G2GJHc0jCbUGcC7yrqkHv527AVFV9KZrBtTefB9z4w9cWyNUG3SyjrSthy4o9j7evhcb6Pddl5MKopq6j46FTOy70Z/Zt8lXwyT2w+HE49r8i//xbVrhW6Mm32eC0OSThjkH8RlVfbPpBVctF5DeAJYgmjY2sWb2cYxMKmbxpHaxY47UMVrq6Rk0SkiFriBuwHH22K4SXPcz9sbGE+NBzuNuWddFMOPqmyHcXLpzp/j8bf0lkn9fEnXATREv/Bx/sAHf7VlfrppRuXbmnJeA9vrKumitTgHeA1EzIHgFDT3Yf/j1HuGTQbQAkxudbZ0Lkz4DnvwtfzYWhJ0bueet3uppPI0+Hrj0j97wmLoX7SbVARP4M3OP9/ANgYXRCihHVZSFdQiHJoHw9aOOe67r1dx/8A47mzvkNdMkdxQ8vONMthLLmvdmXUWdBWpZbWR3JBLH8Vagps7LeJiLCTRA/BH4NPIObzTQHlyTav7oaWPvBnoHipkRQvW3PNUmpkDXM7eU77kKvS2g4ZA11K2SB0opa7p33DrcOH23f3MyBJXWCCZfCx/dAxUbI6BOZ5134iGulDpoameczcS3cWUw7gFuiHIs/du2AJ70ql2nZ7oN/1Fne2IA3PpDZ74DF7AqKOtAWo6ZtTL4KPrrbDVYf99NDf75ta2DdB3DCr2watImIcGcxzQG+rarl3s/dgadV9VvRDK5NpGXBd+e41sAhLCgqDARJShDG9M2IYHCmQ8saAoOOc4PVx/7XoVfUXfQoSCJMuCwy8Zm4F+7XjOym5ACgqtvpKCupRaDflENebVoQKGd4TjqpyVY227RC/gwIFrnqu4eifper8TT81Mh1V5m4F26CaBSR3Su1RGQgLVR3jVeqSmEgaN1LpvVGnOFKpxzqntUrX4cdW2xw2kRUuIPUvwQ+FJH3AAGOBa6JWlTtzPpt1QRr6hgf7QqupuNJSoGJl8F//grBYsjMPbjnWTgTMvIiOyPKxL2wWhCq+gaQD6wAngJ+AtQc6D4ROVVEVojIahH52iC3iAwQkXdEpFBE5olIXrPzGSISEJG/h/Xb+KRpi9FxsbjFqIl9k650U6cXP3Zw929fD2vedYnGdgY0ERRWghCR7+GWf/0EuBl4DPjtAe5JxK2bOA0YDVwsIqObXXYX8KiqjgNuA+5odv73wPvhxOingqIgqckJDM+x0hjmIPQY5PYCX/So2yK2tRY/5sbSJtrgtImscMcgbgQOB9ar6vHARKB8/7cwBVitql+p6i7gaeDsZteMBt71Hs8NPS8ik4Ec4K0wY/RNYaCcsX0zSUq0qYXmIE2eARXFbovY1miod9Nkh54E3fpFJzYTt8L9RKtV1VoAEemkql8CIw5wTy5QFPJzwDsWqgCY7j0+F0gXkSwRSQD+F9da2ScRuUZEFojIgi1btoT5q0RWfUMjS0uC1r1kDs2I06BrjltZ3Rqr3oLKjVbW20RFuAki4FVwfQmYIyIvA+sj8Po3A8eJyGLgOKAYaACuB2aramB/N6vq/aqar6r5PXv6s3p5ZWkVtXWNjO9nA9TmECQmw6QrYNUcKN8Q/n2LZkLX3q4kvDERFu4g9bmqWq6qv8WV3PgXcM4BbisGQtu8ed6x0OctUdXpqjoRN1MKb73FUcANIrION05xhYjcGU6sba3QG6D+WolvY1pr0hXu70WPhnd9sNi1ICZe6hKMMRHW6k5zVX1PVV/xxhX2Zz4wTEQGiUgKcBHwSugFIpLtdScB/Bx4yHuNS1W1v6oOxLUyHlXVmCz1URAoJ7NzMgOy0vwOxbR33frDsJNh0WNuM6kDWfy4m/008fLox2biUtRGVVW1HrgBeBNYDjyrqstE5DYRmeZdNhVYISIrcQPSt0crnmgpKHJbjIpVbjWRMHkGVG1ye5DvT2ODm700+Hg3C8qYKIjqxgSqOhuY3ezYrSGPZwGzDvAcjwCPRCG8Q1Zb18CK0kquGznE71BMRzHsFEjv6/asHnXWvq9bM9eV6Djl920Xm4k7Ni/zECwrCdLQqIyzFdQmUhKT3FjEmndh+7p9X7fwYVd9eMQZbRaaiT+WIA5BQZG3B7XVYDKRNOkKt/Bt4cyWz1eWui6oCZe4Uh3GRIkliENQGCind0YqORmpfodiOpLMXBj2LTcI3dJg9ZInoLHe1j6YqLMEcQgKAkHrXjLRkT8DdmyGL/+99/HGRrf2YcAxkD3Un9hM3LAEcZCCNXWs3brDupdMdAw9ye1k2LwM+Lr33diElfU2bcASxEH6POCNP9gCORMNCYluLOKreW4r0SYLZ0Ln7vuf4WRMhFiCOEhNJb4Psy4mEy0TL3dbiC58xP28YyssfxXGXQTJNu5los8SxEEqKCpnUHYXMjtbiQMTJRl9XBG/JU9A/U4oeAoa62CyDU6btmEJ4iAVBoK2g5yJvvwZUL3NtRwWzoR+R0CvUX5HZeKEJYiDUFpRy6aKWivxbaJv8AnQbQC89SvYtsqmtpo2ZQniIBQUeRVcrcS3ibaEBNelVLkROmXCmHP9jsjEEUsQB6EwECQxQRjT1xKEaQMTL4fETjD+QkixqsGm7US1WF9HVRAoZ0ROOqnJtkG8aQNde8F1/4GM5hsyGhNd1oJoJVV1A9TWvWTaUvYwaz2YNmcJopXWb6smWFNnA9TGmA7PEkQrFdgWo8aYOGEJopUKioKkJicwPKer36EYY0xUWYJopcJAOWP6ZpKUaG+dMaZjs0+5VqhvaGRpSdC6l4wxccESRCusLK2itq7RZjAZY+KCJYhWKPQGqG0GkzEmHliCaIWCQJCM1CQGZtl8dGNMx2cJohUKisoZ368bIuJ3KMYYE3WWIMJUW9fAitJK24PaGBM3LEGEaVlJBQ2NajOYjDFxwxJEmPaU+LYEYYyJD5YgwlQYKCcnoxM5GbYXsDEmPliCCJPbYtRaD8aY+GEJIgzBmjq+2rrDupeMMXHFEkQYPg8EAWwGkzEmrliCCENTie9xudaCMMbED0sQYSgoKmdQdhcy05L9DsUYY9qMJYgwFAaC1r1kjIk7UU0QInKqiKwQkdUicksL5weIyDsiUigi80Qkzzs+QUQ+FpFl3rkLoxnn/pRW1LKpotZmMBlj4k7UEoSIJAL3AKcBo4GLRWR0s8vuAh5V1XHAbcAd3vFq4ApVHQOcCvyfiPjyCb1ngZy1IIwx8SWaLYgpwGpV/UpVdwFPA2c3u2Y08K73eG7TeVVdqaqrvMclwGagZxRj3afCQJDEBGF0H0sQxpj4Es0EkQsUhfwc8I6FKgCme4/PBdJFJCv0AhGZAqQAa6IU534VBMoZnpNO55REP17eGGN84/cg9c3AcSKyGDgOKAYamk6KSB/gMWCGqjY2v1lErhGRBSKyYMuWLREPTlUpDASZYN1Lxpg4FM0EUQz0C/k5zzu2m6qWqOp0VZ0I/NI7Vg4gIhnAv4FfquonLb2Aqt6vqvmqmt+zZ+R7oNZvqyZYU2c7yBlj4lI0E8R8YJiIDBKRFOAi4JXQC0QkW0SaYvg58JB3PAV4ETeAPSuKMe7X7gVyNsXVGBOHopYgVLUeuAF4E1gOPKuqy0TkNhGZ5l02FVghIiuBHOB27/gFwDeBq0RkifdnQrRi3ZeCoiCpyQkMz0lv65c2xhjfJUXzyVV1NjC72bFbQx7PAr7WQlDVx4HHoxlbOAoD5Yzpm0lyot9DNcYY0/bsk28f6hsaWVpiK6iNMfHLEsQ+rCytoraukQlW4tsYE6csQexD4e4BaksQxpj4ZAliHwoCQTJSkxiYleZ3KMYY4wtLEPtQUFTO+H7dEBG/QzHGGF9YgmhBbV0DK0orbYDaGBPXLEG0YFlJBQ2NauMPxpi4ZgmiBU0lvm0GkzEmnlmCaEFhoJycjE7kZKT6HYoxxvjGEkQL3Baj1nowxsQ3SxDNBGvq+GrrDuteMsbEPUsQzXweCAJWwdUYYyxBNLO7xHeutSCMMfHNEkQzBUXlDMruQmZast+hGGOMryxBNOMGqK17yRhjLEGE2FxRy6aKWpvBZIwxWILYS4E3QD3eWhDGGGMJIlRBUTmJCcKYvpYgjDHGEkSIgkA5w3PS6ZyS6HcoxhjjO0sQHlWlMBC07iVjjPFYgvCs31ZNsKaO8baC2hhjAEsQu+1eIGctCGOMASxB7FYYCNIpKYHhOel+h2KMMTHBEoSnoKicsbmZJCfaW2KMMWAJAoD6hkaWltgKamOMCWUJAli1uYraukbG2wpqY4zZzRIEe7YYtRlMxhizhyUIXImNjNQkBmal+R2KMcbEDEsQuD2ox+V1Q0T8DsUYY2JG3CeI2roGvtxUyfh+NkBtjDGh4j5BVNbWc+a4PnxjSLbfoRhjTExJ8jsAv/VM78RfL5rodxjGGBNz4r4FYYwxpmWWIIwxxrQoqglCRE4VkRUislpEbmnh/AAReUdECkVknojkhZy7UkRWeX+ujGacxhhjvi5qCUJEEoF7gNOA0cDFIjK62WV3AY+q6jjgNuAO794ewG+AI4ApwG9EpHu0YjXGGPN10WxBTAFWq+pXqroLeBo4u9k1o4F3vcdzQ85/C5ijqmWquh2YA5waxViNMcY0E80EkQsUhfwc8I6FKgCme4/PBdJFJCvMexGRa0RkgYgs2LJlS8QCN8YY4/8g9c3AcSKyGDgOKAYawr1ZVe9X1XxVze/Zs2e0YjTGmLgUzXUQxUC/kJ/zvGO7qWoJXgtCRLoC56lquYgUA1Ob3TsvirEaY4xpRlQ1Ok8skgSsBE7EJYb5wCWquizkmmygTFUbReR2oEFVb/UGqRcCk7xLFwGTVbVsP6+3BVh/CCFnA1sP4f6OxN6Lvdn7sTd7P/boCO/FAFVtsQsmai0IVa0XkRuAN4FE4CFVXSYitwELVPUVXCvhDhFR4H3gB969ZSLye1xSAbhtf8nBu+eQ+phEZIGq5h/Kc3QU9l7szd6Pvdn7sUdHfy+i1oJobzr6f+jWsPdib/Z+7M3ejz06+nvh9yC1McaYGGUJYo/7/Q4ghth7sTd7P/Zm78ceHfq9sC4mY4wxLbIWhDHGmBZZgjDGGNOiuE8QB6o4G09EpJ+IzBWRL0RkmYjc6HdMfhORRBFZLCKv+R2L30Skm4jMEpEvRWS5iBzld0x+EpEfe/9OlorIUyKS6ndMkRbXCSLMirPxpB74iaqOBo4EfhDn7wfAjcByv4OIEX8F3lDVkcB44vh9EZFc4EdAvqqOxa31usjfqCIvrhME4VWcjRuqulFVF3mPK3EfAF8rkhgvvP1JzgAe9DsWv4lIJvBN4F8AqrpLVcv9jcp3SUBnr2pEGlDiczwRF+8JIqyqsfFIRAYCE4FP/Y3EV/8H/DfQ6HcgMWAQsAV42Otye1BEuvgdlF9UtRi3n80GYCMQVNW3/I0q8uI9QZgWeIUTnwduUtUKv+Pxg4icCWxW1YV+xxIjknC10f6pqhOBHUDcjtl5G5idjUucfYEuInKZv1FFXrwniANWnI03IpKMSw5PqOoLfsfjo6OBaSKyDtf1eIKIPO5vSL4KAAFVbWpRzmJPMc14dBKwVlW3qGod8ALwDZ9jirh4TxDzgWEiMkhEUnCDTK/4HJNvRERwfczLVfXPfsfjJ1X9uarmqepA3P8X76pqh/uGGC5V3QQUicgI79CJwBc+huS3DcCRIpLm/bs5kQ44aB/N/SBi3r4qzvoclp+OBi4HPheRJd6xX6jqbB9jMrHjh8AT3pepr4AZPsfjG1X9VERm4bYiqAcW0wHLblipDWOMMS2K9y4mY4wx+2AJwhhjTIssQRhjjGmRJQhjjDEtsgRhjDGmRZYgjIkBIjLVKsaaWGMJwhhjTIssQRjTCiJymYh8JiJLROQ+b7+IKhH5i7c3wDsi0tO7doKIfCIihSLyole/BxEZKiJvi0iBiCwSkSHe03cN2W/hCW+FrjG+sQRhTJhEZBRwIXC0qk4AGoBLgS7AAlUdA7wH/Ma75VHgZ6o6Dvg85PgTwD2qOh5Xv2ejd3wicBNub5LBuJXtxvgmrkttGNNKJwKTgfnel/vOwGZcOfBnvGseB17w9k/opqrvecdnAs+JSDqQq6ovAqhqLYD3fJ+pasD7eQkwEPgw+r+WMS2zBGFM+ASYqao/3+ugyK+bXXew9Wt2hjxuwP59Gp9ZF5Mx4XsHOF9EegGISA8RGYD7d3S+d80lwIeqGgS2i8ix3vHLgfe8nfoCInKO9xydRCStTX8LY8Jk31CMCZOqfiEivwLeEpEEoA74AW7znCneuc24cQqAK4F7vQQQWv30cuA+EbnNe45vt+GvYUzYrJqrMYdIRKpUtavfcRgTadbFZIwxpkXWgjDGGNMia0EYY4xpkSUIY4wxLbIEYYwxpkWWIIwxxrTIEoQxxpgW/X9Mni9xGl9urwAAAABJRU5ErkJggg==\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deXhU9dXA8e/JTsgKSUDIsChL2CSRgLu1dUNBcCmCW93q0mrtYm21i/a1te1b+7bWVqtYaaVVEbEqVizuWqtI2GTfRZOwhQBZgOzn/ePegQEGSGAmd2ZyPs+TZ2buMnMykDlzz28TVcUYY4w5UJzXARhjjIlMliCMMcYEZQnCGGNMUJYgjDHGBGUJwhhjTFCWIIwxxgRlCcKYEBCRv4nIL1p57AYROfdYn8eYcLMEYYwxJihLEMYYY4KyBGE6DLe0c7eILBaRXSLylIh0E5HXRaRGRN4SkeyA48eJyDIR2Ski74nIoIB9RSKywD3veSDlgNcaKyKL3HM/EpETjzLmm0VkrYhsF5GZItLD3S4i8nsR2Soi1SKyRESGuvsuEpHlbmzlIvL9o3rDTIdnCcJ0NJcD5wEDgIuB14EfAbk4fw93AojIAOA54DvuvlnAqyKSJCJJwMvA34EuwAvu8+KeWwRMAW4FugJPADNFJLktgYrIV4BfAVcAxwGfA9Pc3ecDZ7m/R6Z7TKW77yngVlVNB4YC77TldY3xswRhOpo/quoWVS0H/gN8oqoLVbUOeAkoco+bCLymqm+qaiPwW6ATcBpwCpAIPKyqjao6AygJeI1bgCdU9RNVbVbVp4F697y2uBqYoqoLVLUeuBc4VUT6AI1AOlAAiKquUNVN7nmNwGARyVDVHaq6oI2vawxgCcJ0PFsC7u8J8jjNvd8D5xs7AKraApQCPd195br/TJefB9zvDdzllpd2ishOwOee1xYHxlCLc5XQU1XfAf4EPApsFZHJIpLhHno5cBHwuYi8LyKntvF1jQEsQRhzKBtxPugBp+aP8yFfDmwCerrb/HoF3C8FHlTVrICfVFV97hhj6IxTsioHUNVHVHUEMBin1HS3u71EVccDeTilsOltfF1jAEsQxhzKdGCMiJwjIonAXThloo+Aj4Em4E4RSRSRy4BRAec+CdwmIie7jcmdRWSMiKS3MYbngBtEpNBtv/glTklsg4iMdJ8/EdgF1AEtbhvJ1SKS6ZbGqoGWY3gfTAdmCcKYIFR1FXAN8EdgG06D9sWq2qCqDcBlwPXAdpz2in8GnDsPuBmnBLQDWOse29YY3gJ+CryIc9VyAjDJ3Z2Bk4h24JShKoGH3H3XAhtEpBq4Dactw5g2E1swyBhjTDB2BWGMMSYoSxDGGGOCsgRhjDEmKEsQxhhjgkrwOoBQycnJ0T59+ngdhjHGRJX58+dvU9XcYPtiJkH06dOHefPmeR2GMcZEFRH5/FD7rMRkjDEmKEsQxhhjgrIEYYwxJqiYaYMIprGxkbKyMurq6rwOJexSUlLIz88nMTHR61CMMTEiphNEWVkZ6enp9OnTh/0n3owtqkplZSVlZWX07dvX63CMMTEirCUmERktIqvcJRPvCbL/NnepxEUi8qGIDHa39xGRPe72RSLy+NG8fl1dHV27do3p5AAgInTt2rVDXCkZY9pP2K4gRCQeZzGT84AyoEREZqrq8oDDnlXVx93jxwG/A0a7+9apamEI4jjWp4gKHeX3NMa0n3BeQYwC1qrqend65GnA+MADVLU64GFnoN2nllVVGppaaGq2KfNNBFv6T9hZ6nUUpoMJZ4LoibOyll+Zu20/InK7iKwDfoO7YLyrr4gsdJdMPDPYC4jILSIyT0TmVVRUHFWQjc0trNxcTdWexqM6/0h27tzJY4891ubzLrroInbu3BmGiEzU2boCZtwA7/7S60hMB+N5N1dVfVRVTwB+CPzE3bwJ6KWqRcD3gGcD1tsNPHeyqharanFubtCR4keUGB9HQlwcuxuaj/I3OLxDJYimpqbDnjdr1iyysrLCEpOJMnMnO7er/w3Nh/9/Y0wohTNBlOOs4euX7247lGnAJQCqWq+qle79+cA6nDV3Q05ESE2KD1uCuOeee1i3bh2FhYWMHDmSM888k3HjxjF48GAALrnkEkaMGMGQIUOYPHny3vP69OnDtm3b2LBhA4MGDeLmm29myJAhnH/++ezZsycssZoItGcnfDoNMnvBnu3wxUdeR2Q6kHB2cy0B+otIX5zEMAm4KvAAEemvqmvch2OANe72XGC7qjaLyPFAf2D9sQTzP68uY/nG6qD7GptbaGhqoXNy296OwT0yuP/iIYc95te//jVLly5l0aJFvPfee4wZM4alS5fu7Y46ZcoUunTpwp49exg5ciSXX345Xbt23e851qxZw3PPPceTTz7JFVdcwYsvvsg111zTplhNlFr0LDTuhssmw98vgZWvQd+zvI7KdBBhu4JQ1SbgDmA2sAKYrqrLROQBt8cSwB0iskxEFuGUkq5zt58FLHa3zwBuU9Xt4Yo1zu0B1NwS/jbyUaNG7TdW4ZFHHmH48OGccsoplJaWsmbNmoPO6du3L4WFToeuESNGsGHDhrDHaSJASwuUPAm+k6H3qXDCOU6CsGWCTTsJ60A5VZ0FzDpg230B9799iPNexFmoPWQO902/qaWF5Rur6Z6RQl5GSihf9iCdO3fee/+9997jrbfe4uOPPyY1NZWzzz476FiG5OTkvffj4+OtxNRRrHsbtq+HL//YeVwwBla9Bps+hR7H3APcmCPyvJE6EiTExZGcEJ52iPT0dGpqaoLuq6qqIjs7m9TUVFauXMmcOXNC/vomis2dDGndYJB7wT3wQpA4WPkvb+MyHYYlCJe/oVpDfPnetWtXTj/9dIYOHcrdd9+9377Ro0fT1NTEoEGDuOeeezjllFNC+tomilWugzVvwogbICHJ2ZbaBXqf7pSZjGkHMT0XU1ukJsWzY3cDjc0tJCXEh/S5n3322aDbk5OTef3114Pu87cz5OTksHTp0r3bv//974c0NhOhSp6CuHgovmH/7QVj4d8/dBJI1xO8ic10GHYF4UpNcpJCuLq7GtNq9bWw8B8weDykd99/X8FFzq1dRZh2YAnClZwYT5yIJQjjvSXTob4KRt168L6sXnDccGuHMO3CEoQrToROieEbMGdMq6jCJ5Oh+4ngGxX8mIKxUDoXara0b2ymw7EEEaBTUjx1jc20WD9z45UNH0LFCjj5VjjUDL0FYwGFVbOC7zcmRCxBBEhNiqdFlbpGu4owHpn7BHTqAkMvP/QxeYMgu6+1Q5iwswQRwN9QvcfKTMYLO0udD/2TvgaJnQ59nAgMGgufvQ91waePMSYULEEECPfMrq2Rlpbm2Wsbj82b4tyOvOnIxxaMheYGWPtmeGMyHZoliADhntnVmENqrIMFT8PAi5yeSkeSPxI658EK681kwscGyh0gNSme6rpGmlpaSIg79vx5zz334PP5uP322wH42c9+RkJCAu+++y47duygsbGRX/ziF4wfP/4Iz2Ri2rKXYHcljLq5dcfHxTtTbyz9JzTVQ0Lykc8xpo06ToJ4/R7YvOSIh3VpaSG1sQUS4+BICaL7MLjw14c9ZOLEiXznO9/ZmyCmT5/O7NmzufPOO8nIyGDbtm2ccsopjBs3ztaV7qhUncbpnIHQ90utP2/Qxc5Vx2cfQP/zwhef6bA6ToJopfg450M6VDN/FxUVsXXrVjZu3EhFRQXZ2dl0796d7373u3zwwQfExcVRXl7Oli1b6N69+5Gf0MSe8vmwcSFc9NtDd20Npu9ZkJQOK161BGHCouMkiCN80/cToHxzDckJcfTJ6XzE41tjwoQJzJgxg82bNzNx4kSeeeYZKioqmD9/PomJifTp0yfoNN+mg/jkCeeDfviktp2XkOwkhlWzoOX3TtnJmBCyRuogQj2z68SJE5k2bRozZsxgwoQJVFVVkZeXR2JiIu+++y6ff/55SF7HRKHarU77Q9HVkJze9vMLxsCuCigrCX1spsOzBBFEalI8TS0tNDa3hOT5hgwZQk1NDT179uS4447j6quvZt68eQwbNoypU6dSUFAQktcxUWj+36ClEUa2snH6QP3Ph7hEm5vJhEXHKTG1QeDMrqGa+nvJkn0N5Dk5OXz88cdBj6utrQ3J65ko0NzojH044RzI6Xd0z5GSAcd/yenuet7P29aGYcwR2BVEECk2s6tpDytehZpNMOqWY3uegrGw4zPYuiI0cRnjsgQRhNjMrqY9zH0Ssvscew+kgRcBYmUmE3IxnyCOtqE5NSmePVE0s2uol0o1YbZ5CXzxEYz8+rH3Pkrv5kwNbgnChFhYE4SIjBaRVSKyVkTuCbL/NhFZIiKLRORDERkcsO9e97xVInLB0bx+SkoKlZWVR/Xh2SkpHo2SmV1VlcrKSlJSUrwOxbTW3MmQ0AmKrgnN8xWMhU2fws4vQvN8xhDGRmoRiQceBc4DyoASEZmpqssDDntWVR93jx8H/A4Y7SaKScAQoAfwlogMUNU2fVrn5+dTVlZGRUVFm+NvalG2VNVRvy2RtOTIb8tPSUkhPz/f6zBMa+zeDotfgBOvgE7ZoXnOgjHw5k9h5Sw45bbQPKfp8ML5yTcKWKuq6wFEZBowHtibIFQ1cK7izoD/q/54YJqq1gOficha9/mCd/05hMTERPr27XtUwasqN/7ybc7sl8PvJg47qucwJqiF/4CmPcfeOB2o6wmQN9gpM1mCMCESzhJTT6A04HGZu20/InK7iKwDfgPc2cZzbxGReSIy72iuEg5HRCj0ZbGodGdIn9d0cC3NUPIk9D4dug8N7XMXjIHP/+tcoRgTAp43Uqvqo6p6AvBD4CdtPHeyqharanFubm7IYyv0ZbF+2y6qdjeG/LlNB7XmDaedIJRXD34FY0FbYNXroX9u0yGFM0GUA76Ax/nutkOZBlxylOeGRZEvC4BFZXYVYUJk7mRI7+F82w+144ZDps+WIjUhE84EUQL0F5G+IpKE0+g8M/AAEekf8HAMsMa9PxOYJCLJItIX6A/MDWOsQQ3Lz0QEFn1hCcKEwLY1sO4dGHkjxCeG/vlFnMSz7m1o2BX65zcdTtgShKo2AXcAs4EVwHRVXSYiD7g9lgDuEJFlIrII+B5wnXvuMmA6ToP2v4Hb29qDKRTSUxLpn5fGotId7f3SJhbNfRLik+Ck68P3GgVjoKkO1r4dvtcwHUZY+2+q6ixg1gHb7gu4/+3DnPsg8GD4omudQl8Wby7fgqragj7m6NXXwKJnYchlkBb69rK9ep3mdJ1d+RoMHnfk4405DM8bqSNdoS+bHbsb+WL7bq9DaR/b1sKyl51VzkzofDoNGmrC0zgdKD4BBlwIq193JgM05hhYgjiCQn9Ddax3d60qg5nfgkdHwQvXOeUQExqqTuN0zxGQPyL8rzdoLNRVOV1eTexb/AIsnh6WL3WWII5gQLc0OiXGszBWG6prK+Df98IjRc633FG3OGsMzL4XPm/TuERzKOvfg22rw3/14Hf8l51pPFbY3Ewxr6ocXvsezH/aEoQXEuLjGNYzM/auIOqq4J1fwB+GwyePO9M+fGuBszTrZU9CVi/nSqJ6k9eRRr+5kyE1B4Zc2j6vl5QK/c5x2iGsVBi7VOHVO6GlCcb/CeJC/3FuCaKpAaZcCB8/BnuC91Yq7JXF8o3V1DdF/sR9R9SwGz58GB4+ET54CAacD7fPhfGPQpY79KRTFkx8xmlYfeE65z0yR2fH587AtRHXO2tIt5eCsVCzETYuaL/XNO1r4T9g7Vtw7v9Al6ObUuhILEHUbnYy8Ox74f8GwSt3wMaF+x1S6MuiobmFlZtqPAoyBJoanHaFRwrhrfud6aFv/QAm/A1y+h98fLfBTtIo/QRm/6jdw40ZJX8BiYPiG9v3dQdcABJvg+ZiVVW583fZ+wxnyvgwifxpSsMtqxd8/U3YtBjmPeU09iz8u9OgOPLrMOTS/Rqqh7v3o0ZLs/M7vfdLZ4qHXqfBhKeh96lHPnfoZc430I/+CD1PgsKrwh9vLGnYDQumOo3GmQdNJRZeqV2gz+lOO8Q59x35eBM92qG05GdXEH7HnQgX/wHuWgkX/sYpr7z8DfjdII775EGK0nZEVzuEKiyfCX8+DV6+DVKy4OoX4YZZrUsOfuf8DPqcCf/6LmxcFLZwY9LSGVC3s/0apw9UcDFsW+WM4Daxox1KS36WIA6Ukgkn3+rU5a97Ffqehcx5jJeabmfS6u869eSWCG6LUHWmc3jyyzD9WmfytglPwy3vQ/9z276ofXyCU4ZKzYHnr4VdlWEJO+b4u7bmDXFmbvVCwUXOra00FzvaqbTkZwniUESg71lwxVT47jLm9rmV3k2fwXOTnJ4/H/wWard6HeX+vvgEnr4Y/n4p7NoG4x+Db3wMQy45tsvQzjkwcarTXvPijZGdICPFF3OcZUVH3dz2pBwqmfnQo8jaIWJFO5aW/CxBtEbGcTSd+QPOqP8DS8/4E3Q5Ht75OfxuMMy4yRkv4GV3ws1L4NmJMOV8qFgFFz4E35oPRVc7VwCh0HMEjPk/p0//Oz8PzXPGsrmTnavRE6/wNo6CsVBWYt2VY8GiZ9qttORnCaKVTszPolkSeIuT4bqZcHuJc4m35k3462j48+lQ8pTTdtFeKtc5CerxM+CLj53GyG8vgpNvCU+XypO+5nTX/PD3sPyV0D9/rKjeBCtmQtG1kNTZ21gKxjq3q2Yd/jgT2arKnQGt7VRa8rME0UppyQkMyEvf11CdO8AZVHbXCrj4EYiLd0Y0/t8geO37sHVF+IKpKoOZd8KfRjp/+GfeBd/+1LkN9wfShb+BnsXw8jedqxVzsPl/dcpwI2/yOhLIHQhd+1k7RDTzoLTkZwmiDQp9WXxauhMNLCcldYYR1zljCr7+ttOlccFUeOwU+OsYWPpi6Aaa7doG//4RPHISfPqcU9/+9qfOlUOn7NC8xpEkJDvtMomdYNrVUFd95HM6kqYGmPdXZ7qSLsd7Hc2+NSI++wD2RFEvPLOPB6UlP0sQbVDYK4sduxv5vDLIzK4ikF8Mlz4O31sB5z0AVaUw40b4/RBnWouqsqN74boqeOdBd1qMP8OwCU4bw4X/C2l5x/ZLHY3Mnk7Ppu3rna7ALS3tH0OkWv4K7NrqlPkiRcHFzrfPNW96HYlpK49KS36WINqg1TO7du4Kp38b7lwEV89wBpl98Ft4eJjzrXvdO637UG3YDf/9g9tr6jfQ71z45idwyaPOAD8v9TkDzv+FU7r48HfexhJJ5k6GLifA8V/xOpJ9eo6AtO6w8lWvIzFtoQqvftstLf2xXUtLfjaSug0GdEsnNSmeRaU7uaSoFSNj4+Kg/3nOz47Pndr0gqnOh2qXE5wadeFVB5eHmhpg4VR4/yGna2m/8+ArP4EeheH5xY7WKd+A8vnO1VGPQieBdWQbF0LZXBj9v578MR9SXJwzJuLT56GxDhJTvI7ItMaiZ2Dtm067n0flygj6Xxz54uOEYT0zWXg0I6qze8O5P3PKT5c9CZ1znQEv/1cAL98O5Quchs1Pp8GfiuG1u5x64w2vwzUzIi85gFNWG/cIdBvi9KbascHriLw190lI7AyFV3odycEKxkDjLqebsol8+5WWbvYsDEsQbVTYK4sVxzKza0Ky0zf+ptlw24cw/EpY9pIz8vmhfvDSrU7/+atnOMmh92mh/QVCLakzTPw7oPD8NU5ZrCPatQ2WzHCSQ0qm19EcrM9ZkJxhvZmiQQSUlvwsQbRRkTuz6/KNIei9030YXPyw01X2woecZDDhb+60GOd5NwK3rbocD5c/BZuXOnM2dcQ1CBZMheZ6T7/tHVZCktOzKtKnijH7Skvn/szznnBhTRAiMlpEVonIWhG5J8j+74nIchFZLCJvi0jvgH3NIrLI/ZkZzjjbotDntBeEdOK+lEyn18ukZ5xFZSKpft1a/c+DL/8IFk/reMuVNjc5gyT7ngV5BV5Hc2iDxsLubc4U7iYyRUhpyS9sn0QiEg88ClwIDAauFJHBBxy2EChW1ROBGcBvAvbtUdVC92dcuOJsq+6ZKXTPSImumV3by5nfhwEXdrzlSle/DtVlMOpWryM5vH7nQnyyLUUaqSKotOQXzghGAWtVdb2qNgDTgPGBB6jqu6rqL1rPAfLDGE/IFPqyLEEEExcHlz0BWb071nKlnzwBmT4YMNrrSA4vOR2OP9tph+iIZcBIF0GlJb9wJoieQGnA4zJ326HcBLwe8DhFROaJyBwRuSTYCSJyi3vMvIqKimOPuJUKe2XxeeVutu+ypTgPkpLplMrqa2H612J/udKtK2DDf5wuy6GaGDGcCsbAzs9hy1KvIzGBqsqdWRJ6nx4RpSU/769hABG5BigGHgrY3FtVi4GrgIdF5IQDz1PVyaparKrFubm57RTtvgFzn9pVRHB5g5zBfGVznXJTLJs72SnbFH3N60haZ+BFgNgU4JFkb2mpsd3nWjqScEZSDvgCHue72/YjIucCPwbGqWq9f7uqlru364H3gKIwxtomw3pmEicc3XiIjmLIpXDanc6azAuf8Tqa8Niz0xm3MmyCM3o+GqTlQq9TrB0ikkRgackvnAmiBOgvIn1FJAmYBOzXG0lEioAncJLD1oDt2SKS7N7PAU4Hlocx1jbpnJzAgG7p1g5xJOfc7/TsidXlShc9C427nUkTo0nBWNiyxAY2RoIILS35hS1BqGoTcAcwG1gBTFfVZSLygIj4eyU9BKQBLxzQnXUQME9EPgXeBX6tqhGTIACKegWZ2dXsLz4BvvpXZ9R4rC1X2tICJU+C7+TIHOV+OAVjnFsrM3krgktLfmGNSFVnqeoAVT1BVR90t92nqjPd++eqarcDu7Oq6keqOkxVh7u3T4UzzqNR6Muiak8jn23b5XUoka1zjjPSunZLbC1Xuu4dZzbbURE0a2trdekL3YZamclrEVxa8ou8lBUlhrd2ZlfjzGY79nfOPEBvP+B1NKEx9wlI6waDImaITtsUjIXSOVDbfr3/TIAILy35WYI4Sv3z0unszuxqWqHoGii+Ef77cPQvV1q5zllbYcQNzhQW0ahgDGiLM8jPtK8oKC35RW5kES4+ThiWn2ldXdti9K8hf6SzXOnWlV5Hc/RKnnKWmC2+wetIjl73Yc6aItYO0f4WPRvxpSU/SxDHoNCXzfJN1dQ1xkhdPdwClyt9/mpnpbxo07ALFv4DBo+H9O5eR3P0RJwy07p3ob7G62g6juqN7lxLkV1a8rMEcQwKfVk0NivLN9m6zK2W0QMmPA3bP4OXonC50sXPQ31V5M+71BoFY50ZaNe+7XUkHYMqzLwzKkpLfpEfYQQr6uU2VH9hZaY26XM6XPAgrHotupYrVXVmqu1+IvhGeR3Nset1CqR2tTUi2ksUlZb8LEEcg24ZKRyXaTO7HpWTb4NhVzjLla59y+toWmfDh7B1OZx8a/Ss1XE4cfEw8EJY/Ubsz5nltSgrLflZgjhGNrPrURKBi//g9MefcZNTcop0cyc764cPvdzrSEKnYKxTMtvwH68jiV3+0lJzQ9SUlvyiJ9IIVejL4ovtu6msrT/ywWZ/SakBy5VeG9nLlVaVOT1+Tvqa08geK44/21lH23ozhU8Ulpb8LEEco70zu5bZVcRR6dLXWa50y1Knb3ikTl0ybwqgUHyT15GEVmIn6HeOkyCircNANAgsLUXhqHtLEMdoWH4m8XFiDdXHov958OUfw5LpThkn0jTWwfy/OavlZfc+4uFRZ9DFULsZNi7wOpLY4h8QF4WlJb/oizjCpCY5M7va1N/H6My7nLUKZv8IPv/I62j2t+wl2F3prBsei/qfB3EJsOJVryOJLYuehTVvRGVpyc8SRAgU+pyZXVtaIrQ8Eg3i4uDSxyG7D0yPoOVKVZ15l3IGQt8veR1NeHTKhj5nWjtEKPlLS71Oi8rSkp8liBAo8mVRXdfEZ5U2s+sxScmEic84o5UjZbnS8vmwcaGz5kMsdG09lIIxULkGKlZ5HUn0i4HSkl8ULKIb+QoDBsydkJvmcTRRLq8ALnkMXrgOpo535gvCvTLb24Dd1setPSbI/m1rICkdhk86yl8oShSMgVnfd8pMuQO9jia6+UtLo/8Xuh60UnJUsQQRAifkppGWnMCi0p1cPiLf63Ci35BLYMf/OD2Hqt1Vavd+e5c2Pj6ac2TfTWIKfOXHkJx+9L9PNMjoAT1HOGWms77vdTTRK0ZKS36WIEIgPk44MT/TBsyF0hnfcX5M+ykYC2//j7NWQWZPr6OJPjFUWvKL/t8gQhT6slhhM7uaaFYw1rldNcvbOKJVYK+lKC8t+VmCCJFCXxZNLcqyjVE4hbUxALkDIGeAdXc9GjFWWvKzBBEi/obqhTZgzkSzgjHOpIR7dngdSfSIwdKSX1h/ExEZLSKrRGStiNwTZP/3RGS5iCwWkbdFpHfAvutEZI37c1044wyFvPQUemZ1snYIE90KLgZthtWzvY4kesRgackvbAlCROKBR4ELgcHAlSIy+IDDFgLFqnoiMAP4jXtuF+B+4GRgFHC/iGSHK9ZQsZldTdTrUQTpPWyNiNaK0dKSXzivIEYBa1V1vao2ANOA8YEHqOq7quqfwnMO4O8jegHwpqpuV9UdwJvA6DDGGhKFvizKduxhm83saqJVXBwUXOSsMte4x+toIlsMl5b8wvkb9QRKAx6XudsO5Sbg9bacKyK3iMg8EZlXUVFxjOEeu0JbYc7EgoKx0LjbWa/aBNe4x+kSHKOlJb+ISHkicg1QDDzUlvNUdbKqFqtqcW5ubniCa4OhPdyZXa3MZKJZnzOcaU+szHSwlhZYPB3+WAwf/t5ZFTEGS0t+4RwoVw74Ah7nu9v2IyLnAj8GvqSq9QHnnn3Aue+FJcoQ6pQUT0H3dEsQJrrFJ8KA0bDqdWhugngbTwvAF3Oc2YbL58Nxw+GyJ5xkGsPCeQVRAvQXkb4ikgRMAmYGHiAiRcATwDhV3RqwazZwvohku43T57vbIt5wm9nVxIKCMbBnO3zxsdeReG/HBmeG4SkXOI3SlzwON78X88kBwpggVLUJuAPng30FMF1Vl5lhqS8AABv/SURBVInIAyIyzj3sISANeEFEFonITPfc7cDPcZJMCfCAuy3iFfqyqKlvYv02m9nVRLF+50JCSseeAryuCt68D/400mlrOPte+NZ8KLwyJhukg2nVtaOIfBv4K1AD/AUoAu5R1TcOd56qzgJmHbDtvoD75x7m3CnAlNbEF0mK3CVIF5XupF+ezexqolRSZzj+y047xOhfxfZU5wdqboIFT8O7v4Td22D4VXDOT50JDTuY1qbBG1W1GqfUkw1cC/w6bFFFsRNy00hPTmBRqY1ENVFu0FioKoXNi72OpP2seQsePx1e+54z7fkt78Glf+6QyQFa30jt//pwEfB3t1TUgb5StF5cnHCiz2Z2NTFgwGiQOFjxL6dRNpZtXQFv/ATWvgXZfWHiP5zuvh38Y661VxDzReQNnAQxW0TSgZbwhRXdCn1ZrNxUYzO7mujWOccZIRzL3V1rK+Bf34U/nwZlJXDBL+H2uTDo4g6fHKD1VxA3AYXAelXd7U6FcUP4wopuhb5smlqUpeVVFPfp4nU4xhy9gjEw+16oXBdbg8Ea6+CTx+E//+cscTvyZjj7Hki1v9dArb2COBVYpao73UFtPwFsXutDKAxoqDYmqhWMcW5jpTeTKix7CR4dBW/dD71Ph2/OgYt+Y8khiNYmiD8Du0VkOHAXsA6YGraoolxuejI9szqx0BKEiXbZvaH7sNhIEGXznbEML1zvLCF77ctw1TRnHQwTVGsTRJOqKs5ke39S1UeBGF+k99gU9sqyOZlMbCi4GEo/gdqtRz42Eu0shRdvhr98BbZ/BuP+CLd+ACd82evIIl5rE0SNiNyL0731NRGJAxLDF1b0K/JlUb5zDxU1NrOriXIFYwCNvqVI62vhnV/An4phxUw48/tw5wI46WsQF+91dFGhtQliIlCPMx5iM87cSG2aWK+jsXYIEzO6DYHsPk5312jQ0gwLpsIfT4IPHnJ6JN0xzxnslmyFj7ZoVYJwk8IzQKaIjAXqVNXaIA5jaM9MEuLEBsyZ6CfijAn47H2oq/Y6msNb/z488SWY+S0nqX39bbj8L5DlO+Kp5mCtShAicgUwF5gAXAF8IiJfDWdg0S4lMZ6C42xmVxMjCsY6C+OsfdPrSILbtgaenQRTx0F9FXz1r3DjbMgv9jqyqNbacRA/Bkb6Z1wVkVzgLZxlQs0hFPqyeGXhRlpalLg4G3RjophvFKTmOL2Zhl7udTT77N4O7/8vlPwFEjo5i/ec/A1ITPE6spjQ2jaIuAOm465sw7kdVqEvm5r6JtZV1HodijHHJi7eWYp09RvQFAEdL5oa4OPH4JEimDvZaXi+cyGc8V1LDiHU2iuIf4vIbOA59/FEDpil1RzM31C9sHQn/btZ45iJcgUXO42/v+jmzNEUF+/c7v2Jd9or/I8P2h/kZ+8xEvAcrdhfucZZp+GEr8D5D0K3wV6/OzGpVQlCVe8WkcuB091Nk1X1pfCFFRuOz+lMekoCi0p3ckWxNZKZKNfvHGeuoj07QFv2/bQ0OyOUA7dp8wGP1T0u2DF6wHO1HPzTcsDrZfrgot9C//O8fldiWqvXElTVF4EXwxhLzImLEwp9NmDOxIi4eDj1dq+jMO3osO0IIlIjItVBfmpEJML7u0WGQl8Wq7bUsKfBZnY1xkSXw15BqKoVzo9RoS+L5hZlSXkVo/raZGDGmOhhPZHCbN+IahswZ4yJLpYgwqxrWjK+Lp1swJwxJuqENUGIyGgRWSUia0XkniD7zxKRBSLSdODIbBFpFpFF7s/McMYZboW+bGuoNsZEnbAlCBGJBx4FLgQGA1eKyIGdlb8ArgeeDfIUe1S10P0ZF64420OhL4uNVXVsra7zOhRjjGm1cF5BjALWqup6VW0ApuGsJ7GXqm5Q1cXE+PrWgQPmjDEmWoQzQfQESgMel7nbWitFROaJyBwRuSS0obWvIT0y3JldLUEYY6JHqwfKeaC3qpaLyPHAOyKyRFXXBR4gIrcAtwD06tXLixhbJSUxnkHHZfCpJQhjTBQJ5xVEORA4v0S+u61VVLXcvV0PvAcUBTlmsqoWq2pxbm7usUUbZoW+LBaXVdHcol6HYowxrRLOBFEC9BeRviKSBEwCWtUbSUSyRSTZvZ+DMwfU8rBF2g4KfVnU2syuxpgoErYEoapNwB3AbGAFMF1Vl4nIAyIyDkBERopIGc5CRE+IyDL39EHAPBH5FHgX+LWqRneC6OUOmLPursaYKBHWNghVncUB04Kr6n0B90twSk8HnvcRMCycsbW3vl07k5GSwMLSnVwx0mZ2NcZEPhtJ3U7i4oThvizryWSMiRqWINpRkS+LVZur2d3Q5HUoxhhzRJYg2lFhryxaFJaUVXkdijHGHJEliHY0PN8/s6uVmYwxkc8SRDvqmpZMry6pliCMMVHBEkQ7K7SGamNMlLAE0c4KfVlsqqpji83saoyJcJYg2pl/wNxCGzBnjIlwliDa2eDjMkiMt5ldjTGRzxJEO0tJjGfwcRm2RrUxJuJZgvBAoS+LJTazqzEmwlmC8EBhryx2NTSzZmuN16EYY8whWYLwQKEvG7CZXY0xkc0ShAf6dE0lKzXRGqqNMRHNEoQHRITh+TZgzhgT2SxBeKTQl8XqLTXsqreZXY0xkckShEf8M7sutpldjTERyhKERwptZldjTISzBOGR7M5J9OmaagPmjDERyxKEh2xmV2NMJLME4aFCXxZbquvZVLXH61CMMeYgYU0QIjJaRFaJyFoRuSfI/rNEZIGINInIVw/Yd52IrHF/rgtnnF4Z7nPaIT61qwhjTAQKW4IQkXjgUeBCYDBwpYgMPuCwL4DrgWcPOLcLcD9wMjAKuF9EssMVq1cG98ggKT6OhZYgjDERKJxXEKOAtaq6XlUbgGnA+MADVHWDqi4GWg449wLgTVXdrqo7gDeB0WGM1RPJCfEM6pFhU24YYyJSOBNET6A04HGZuy1k54rILSIyT0TmVVRUHHWgXiryZbGk3GZ2NcZEnqhupFbVyaparKrFubm5XodzVAp9WexuaGb1FpvZ1RgTWcKZIMoBX8DjfHdbuM+NKoU+GzBnjIlM4UwQJUB/EekrIknAJGBmK8+dDZwvItlu4/T57raY07trKtmpidYOYYyJOGFLEKraBNyB88G+ApiuqstE5AERGQcgIiNFpAyYADwhIsvcc7cDP8dJMiXAA+62mCMiDLcBc8aYCJQQzidX1VnArAO23RdwvwSnfBTs3CnAlHDGFykKfVm8v3oNtfVNpCWH9Z/EGGNaLaobqWNFoS8LVVhcZlcRxpjIYQkiAlhDtTEmElmCiABZqUn0zelsDdXGmIhiCSJC+Gd2VbUBc8aYyGAJIkIU+rLYWlPP5A/W88n6SqrrGr0OyRjTwVmXmQjxpQG5dM9I4Vevr9y7rVeXVIb0yGDwcRkM7pHBkB6ZdMtIRkQ8jNQY01FYgogQfXI6M+dH57C1po5lG6tZ7v/ZVM3rSzfvPa5L5yQGH5fhJA43eRyfm0Z8nCUNY0xoWYKIMHnpKeQNTOHLA/P2bqutb2LlJidZLN9YzbKN1fz1vxtoaHYmwU1JjGNg94z9EkdB93RSk+yf14RWY3MLCXFiV7EdhMRKo2hxcbHOmzfP6zDaTWNzC+sqavdeaSxzrzaq9jhtF3ECfXM6M7hH5n6JIyct2ePITTSqrW/iD2+t5m8fbaBzcgID8tLp1y2N/nlpDOiWTv+8NHLTrfwZjURkvqoWB91nCSJ2qCobq+pYVl7F8k3Ve0tV5Tv3LWmal54cUJ7KZEiPDHp1SSXOSlQmCFVl5qcb+eWsFWyprufSop6kJMazZksNq7fUUF3XtPfYzE6J9M9Lo3+3NPrnpe+9tXazyGYJooOr2t3Isk1Ve9s0lm+sZs3W2r1rUHROimdQwFXGoOMy6J+XTqekeI8jN15avaWG+15Zypz12xnWM5MHxg+hqNe+hR1VlYraetZsqWXNlhrWbK1lzZZaVm+tYefufb3w0lMSnMThTxrd0hnQLY3uGSmWOCKAJQhzkLrGZtZsqWW5mziWbaxmxaZqdjU0AyAC+dmdGJCXTn+3hDCgWzr98tIsccS4mrpG/vDWmr3lpB+MHsikkb1a3RFCVanc1cDqLTWs9ScN937lroa9x6UlJ9AvL40B7pVGv27O/7EemZY42pMlCNMqLS3KF9t3s2JTNau31LJmaw1rttSyflstjc3O/xMR8GWnuqWE9H1/3JY4op6/nPTgayuoqK1n0kgfd19QQJfOSSF7jcraeudKY6t71bHFub+ttn7vMZ2T4unnfikJ/GLSM6uTlULDwBKEOSaNzS18XrnbrTtb4ohFqzY75aRPPtvOifmZPDB+6N45wtrDjl0NbuLwJw3ndmvNvsSRmhRPv7w0+uWlcWLPTK4+pTeJ8TbW91hZgjBh4SSOXW4Jwak9r7XEEVVq6hp52C0npackcPcFbSsnhdvO3Q1OmWpr7X4lq83VdZw9MJfHrj7JunMfI0sQpl21JXEM6JZGvzwncQzols4JuW1PHM0typ7GZnY3NFHX0MLuxib2NDQ7P43N7HZvAx/XucfvaWhhj3v8vu37H983pzMTR/oYP7wnmamJ4XjL2p2q8sqijTw4awXbwlROCqdpc7/gRy8t4cT8LKZcPzJq4o5EliBMRGhr4shNTw7yoX7wh31DU0ubY0lJjCM1KYFOifF0Sorf/zYxntSkeFKS4klJiGfO+kqWb6omOSGOC4d2Z+LIXpxyfJeobUhdtbmGn76ylLkelZNC5Y1lm/nWcwvpmd2JqTeOIj871euQopIlCBPR/Ilj9ZZ93STXbKlh+65GUpPcD2v3Qzvwg9z/IZ6amECnpDg6uR/4QY/zb3c/9Nva2Lm0vIrnS0p5eVE5NXVN9OmayoRiHxNG5JOXkRKmdya0Diwn/eCCAiaO9EVMOelozP1sO19/uoROSfFMvfFkBnZP9zqkqGMJwpgQ2dPQzOtLNzGtpJS5n20nPk748sA8Jo30cfbAXBIisNH04HJSL35wwUCyY6Qss3JzNddNmcuehmb+ct1IRvXt4nVIUcUShDFhsL6ilunzypgxv4xttfXkpSfz1RH5XFHso09OZ6/DA5wPz/teWcbcz7Yz3C0nDY/CctKRlO3YzdemzKV8xx7+eGUR5w/p7nVIUcOzBCEio4E/APHAX1T11wfsTwamAiOASmCiqm4QkT7ACmCVe+gcVb3tcK9lCcJ4pbG5hXdWbmV6SSnvrtpKi8Kpx3dl4kgfo4d2JyWx/XtrVdc18vCba3j6Y6ec9MPRBUws9sX0OILtuxq48W8lLC7byS8vHcakUb28DikqeJIgRCQeWA2cB5QBJcCVqro84JhvAieq6m0iMgm4VFUnugniX6o6tLWvZwnCRILNVXXMmF/K8/NKKd2+h4yUBC4t6snEkb0Y3CMj7K+vqry8qJwHX1tJ5a56rhzVi7vPj51y0pHsbmjim88s4L1VFdx13gDu+Eq/qO1M0F68ShCnAj9T1Qvcx/cCqOqvAo6Z7R7zsYgkAJuBXKA3liBMFGtpUeasr2RaSSn/XrqZhuYWTszPZOJIHxcP70FGSui7y67YVM39ryxj7gannPTzS4ZyYn7slZOOpLG5hR/OWMw/F5bztVN7c//FQ6K6IT7cDpcgwjnCpCdQGvC4DDj5UMeoapOIVAFd3X19RWQhUA38RFX/c+ALiMgtwC0AvXrZ5aSJHHFxwmn9cjitXw47djXw8qJyni8p5ccvLeXn/1rOmGE9mDTKR3Hv7GP+hltd18jv31zN1I8/JyMlgV9fNowrYrycdDiJ8XH8dsJwctKTmfzBeiprG/jdxOEkJ9jAzLaK1CGIm4BeqlopIiOAl0VkiKpWBx6kqpOByeBcQXgQpzFHlN05iRtO78v1p/VhcVkV00pKefXTjby4oIzjczszsdjHZSflk5vetrU6VJWXFpbzy1lOOemqUb34fgcqJx1OXJzwo4sGkZuWzIOzVrBjdwNPXDuC9DBcucWycCaIcsAX8Djf3RbsmDK3xJQJVKpT96oHUNX5IrIOGABYDclELRFhuC+L4b4sfjp2EK8t3sTzJaX86vWVPDR7FecO6sbEUT7O6p97xJLIik3V3PfKUko27GC4L4sp1xd3yHLSkdx81vHkpCdx9wuLmTR5Dn+9YSR56dExbiUShLMNIgGnkfocnERQAlylqssCjrkdGBbQSH2Zql4hIrnAdlVtFpHjgf+4x20/1OtZG4SJVmu31jJ9Xikvzi+jclcDx2WmMGFEPhOKffi67D86uLqukd+9sZq/z3HKST8cXdChy0mt9d6qrXzjHwvITU9m6o2jIqYbciTwspvrRcDDON1cp6jqgyLyADBPVWeKSArwd6AI2A5MUtX1InI58ADQCLQA96vqq4d7LUsQJto1NLXw9ootPD+vlPdXVwBwRr8crij2cd7gbry2eBO/en1fOenuCwaSlWrlpNZa+MUObvxbCfFxwt9uGMXQnplehxQRbKCcMVGmfOceZswrY/q8Usp37iEpIY6GphYKfVk8MH6IlZOO0rqKWr721Fx27m5g8teKOb1fjtchec4ShDFRqqVF+e+6bcxasokiXzZfHZFv5aRjtLmqjuumzGX9tlp+P7GQsSf28DokT1mCMMaYAFV7Grn56XmUfL6d+8cO5vrT+3odkmcOlyAib2YxY4wJs8xOiUy9aRTnDerGz15dzkOzVxIrX5ZDyRKEMaZDSkmM57GrT+LKUb149N113PPiEpqa2762SCyL1IFyxhgTdgnxcfzy0qHkpiXxyDtrqdzVwB+vLLLlcF12BWGM6dBEhO+dP5Cfjx/C2yu3cO1Tn1C1u9HrsCKCJQhjjAGuPbUPj151EovLqpjwxEdsqtrjdUieswRhjDGui4Ydx99uHMnGnXVc/thHrN1a43VInrIEYYwxAU47IYdpt5xCQ7Py1cc/ZsEXO7wOyTOWIIwx5gBDe2byz2+cRmanRK56cg7vrtzqdUiesARhjDFB9OqayozbTqNfXhpfnzqPF+eXeR1Su7MEYYwxh5Cbnsy0W07l1OO7ctcLn/LE++siZkDd7oYmPtu2iznrK/lkfWVYXsPGQRhjzGGkJScw5fqR3PXCp/zq9ZVsrannxxcNCtucWA1NLWytqWNLdT1bq+vYXL3v/hZ3+5aqOmrqm/aeMzw/k1fuOCPksViCMMaYI0hKiOMPEwvp2jmJpz78jG219Tz01eEkJbS+CNPcolTW1jsf8P4P+yr3Az8gIVTuajjo3MR4IS89hW4ZyfTPS+OMfjnkZSTTPSOFbhkp9MjqFMpfdy9LEMYY0wpxccL9Fw8mLyOZ3/x7Fdt3NfD4NSNITYpn5+7Gfd/uq90P/oAP/S3V9VTU1tPcsn95SgRy05LplpFCz6wUinpl0S09he6ZyeRlpNDNTQrZqUmezOJrCcIYY1pJRPjm2f3I6ZzMvS8t4dRfvU1dUwsNTQfP4ZSVmkj3jBTyMlIY0C2dbhkpdMtMoVu6kxC6ZaSQk5ZEQnzkNgVbgjDGmDa6YqSPbpkpzFy0kZy0JPIyUtxyj/Phn5ueTEpi9M/nZAnCGGOOwpcG5PKlAblehxFWkXttY4wxxlOWIIwxxgQV1gQhIqNFZJWIrBWRe4LsTxaR5939n4hIn4B997rbV4nIBeGM0xhjzMHCliBEJB54FLgQGAxcKSKDDzjsJmCHqvYDfg/8r3vuYGASMAQYDTzmPp8xxph2Es4riFHAWlVdr6oNwDRg/AHHjAeedu/PAM4REXG3T1PVelX9DFjrPp8xxph2Es4E0RMoDXhc5m4LeoyqNgFVQNdWnmuMMSaMorqRWkRuEZF5IjKvoqLC63CMMSamhDNBlAO+gMf57ragx4hIApAJVLbyXFR1sqoWq2pxbm5s90c2xpj2JuGautb9wF8NnIPz4V4CXKWqywKOuR0Ypqq3icgk4DJVvUJEhgDP4rQ79ADeBvqravNhXq8C+PwYQs4Bth3D+bHE3ov92fuxP3s/9omF96K3qgb9hh22kdSq2iQidwCzgXhgiqouE5EHgHmqOhN4Cvi7iKwFtuP0XMI9bjqwHGgCbj9ccnDPOaZLCBGZp6rFx/IcscLei/3Z+7E/ez/2ifX3ImxXENEm1v+h28Lei/3Z+7E/ez/2ifX3IqobqY0xxoSPJYh9JnsdQASx92J/9n7sz96PfWL6vbASkzHGmKDsCsIYY0xQliCMMcYE1eETxJFmnO1IRMQnIu+KyHIRWSYi3/Y6Jq+JSLyILBSRf3kdi9dEJEtEZojIShFZISKneh2Tl0Tku+7fyVIReU5EUryOKdQ6dIJo5YyzHUkTcJeqDgZOAW7v4O8HwLeBFV4HESH+APxbVQuA4XTg90VEegJ3AsWqOhRnrNckb6MKvQ6dIGjdjLMdhqpuUtUF7v0anA+ADjtJoojkA2OAv3gdi9dEJBM4C2dwK6raoKo7vY3KcwlAJ3fWiFRgo8fxhFxHTxA2a+whuIs3FQGfeBuJpx4GfgC0eB1IBOgLVAB/dUtufxGRzl4H5RVVLQd+C3wBbAKqVPUNb6MKvY6eIEwQIpIGvAh8R1WrvY7HCyIyFtiqqvO9jiVCJAAnAX9W1SJgF9Bh2+xEJBun2tAXZ764ziJyjbdRhV5HTxCtmjW2IxGRRJzk8Iyq/tPreDx0OjBORDbglB6/IiL/8DYkT5UBZarqv6KcgZMwOqpzgc9UtUJVG4F/Aqd5HFPIdfQEUQL0F5G+IpKE08g00+OYPOOu5vcUsEJVf+d1PF5S1XtVNV9V++D8v3hHVWPuG2JrqepmoFREBrqbzsGZTLOj+gI4RURS3b+bc4jBRvuwzeYaDQ4146zHYXnpdOBaYImILHK3/UhVZ3kYk4kc3wKecb9MrQdu8Dgez6jqJyIyA1iA0/tvITE47YZNtWGMMSaojl5iMsYYcwiWIIwxxgRlCcIYY0xQliCMMcYEZQnCGGNMUJYgjIkAInK2zRhrIo0lCGOMMUFZgjCmDUTkGhGZKyKLROQJd72IWhH5vbs2wNsikuseWygic0RksYi85M7fg4j0E5G3RORTEVkgIie4T58WsN7CM+4IXWM8YwnCmFYSkUHAROB0VS0EmoGrgc7APFUdArwP3O+eMhX4oaqeCCwJ2P4M8KiqDseZv2eTu70I+A7O2iTH44xsN8YzHXqqDWPa6BxgBFDifrnvBGzFmQ78efeYfwD/dNdPyFLV993tTwMviEg60FNVXwJQ1ToA9/nmqmqZ+3gR0Af4MPy/ljHBWYIwpvUEeFpV791vo8hPDzjuaOevqQ+434z9fRqPWYnJmNZ7G/iqiOQBiEgXEemN83f0VfeYq4APVbUK2CEiZ7rbrwXed1fqKxORS9znSBaR1Hb9LYxpJfuGYkwrqepyEfkJ8IaIxAGNwO04i+eMcvdtxWmnALgOeNxNAIGzn14LPCEiD7jPMaEdfw1jWs1mczXmGIlIraqmeR2HMaFmJSZjjDFB2RWEMcaYoOwKwhhjTFCWIIwxxgRlCcIYY0xQliCMMcYEZQnCGGNMUP8PQkCB7FrzgFUAAAAASUVORK5CYII=\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from keras.layers.normalization.batch_normalization import BatchNormalization\n",
        "\n",
        "#Resnet50\n",
        "from tensorflow.keras.applications import ResNet50V2\n",
        "\n",
        "base_modelRN = ResNet50V2(input_shape = (32, 32, 3), include_top = False, weights = 'imagenet')\n",
        "modelRN = Sequential()\n",
        "modelRN.add(base_modelRN)\n",
        "modelRN.add(GlobalAveragePooling2D())\n",
        "modelRN.add(Flatten())\n",
        "modelRN.add(Dense(43, activation='softmax'))\n",
        "                               \n",
        "base_learning_rate = 0.001\n",
        "modelRN.compile(optimizer=tf.keras.optimizers.RMSprop(lr=0.001),\n",
        "              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False),\n",
        "              metrics=['accuracy'])\n",
        "\n",
        "plot_model(modelRN, to_file='/content/drive/MyDrive/modelRN.png', show_shapes=True, show_layer_names=True)\n",
        "modelRN.summary()\n",
        "\n",
        "historyRN = modelRN.fit(X_train,y_train,epochs = 10 , validation_data = (X_valid, y_valid),steps_per_epoch = 100)\n",
        "modelRN.save('/content/drive/MyDrive/modelRN.h5')"
      ],
      "metadata": {
        "id": "UfjSadCCL-0T",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "db5e987e-6ee9-4062-9a63-bc4c3cab845f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/keras/optimizer_v2/rmsprop.py:130: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  super(RMSprop, self).__init__(name, **kwargs)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_5\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " resnet50v2 (Functional)     (None, 1, 1, 2048)        23564800  \n",
            "                                                                 \n",
            " global_average_pooling2d_1   (None, 2048)             0         \n",
            " (GlobalAveragePooling2D)                                        \n",
            "                                                                 \n",
            " flatten_5 (Flatten)         (None, 2048)              0         \n",
            "                                                                 \n",
            " dense_3 (Dense)             (None, 43)                88107     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 23,652,907\n",
            "Trainable params: 23,607,467\n",
            "Non-trainable params: 45,440\n",
            "_________________________________________________________________\n",
            "Epoch 1/10\n",
            "100/100 [==============================] - 48s 378ms/step - loss: 1.7083 - accuracy: 0.5458 - val_loss: 73.1668 - val_accuracy: 0.5810\n",
            "Epoch 2/10\n",
            "100/100 [==============================] - 36s 360ms/step - loss: 0.1458 - accuracy: 0.9616 - val_loss: 91.5643 - val_accuracy: 0.5773\n",
            "Epoch 3/10\n",
            "100/100 [==============================] - 36s 357ms/step - loss: 0.0763 - accuracy: 0.9845 - val_loss: 53.9360 - val_accuracy: 0.6474\n",
            "Epoch 4/10\n",
            "100/100 [==============================] - 37s 371ms/step - loss: 0.0401 - accuracy: 0.9903 - val_loss: 4.4466 - val_accuracy: 0.8036\n",
            "Epoch 5/10\n",
            "100/100 [==============================] - 36s 359ms/step - loss: 0.0356 - accuracy: 0.9935 - val_loss: 8.2385 - val_accuracy: 0.8005\n",
            "Epoch 6/10\n",
            "100/100 [==============================] - 36s 360ms/step - loss: 0.0299 - accuracy: 0.9933 - val_loss: 3.0604 - val_accuracy: 0.8222\n",
            "Epoch 7/10\n",
            "100/100 [==============================] - 36s 360ms/step - loss: 0.0227 - accuracy: 0.9952 - val_loss: 1.1891 - val_accuracy: 0.8816\n",
            "Epoch 8/10\n",
            "100/100 [==============================] - 36s 359ms/step - loss: 0.0230 - accuracy: 0.9955 - val_loss: 0.5140 - val_accuracy: 0.9209\n",
            "Epoch 9/10\n",
            "100/100 [==============================] - 36s 359ms/step - loss: 0.0159 - accuracy: 0.9967 - val_loss: 0.3130 - val_accuracy: 0.9458\n",
            "Epoch 10/10\n",
            "100/100 [==============================] - 36s 359ms/step - loss: 0.0174 - accuracy: 0.9969 - val_loss: 0.5364 - val_accuracy: 0.9206\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import accuracy_score,recall_score\n",
        "results = modelRN.evaluate(X_train, y_train, batch_size=64)\n",
        "results = modelRN.evaluate(X_valid, y_valid, batch_size=64)\n",
        "results = modelRN.evaluate(X_test, y_test, batch_size=64)\n",
        "\n",
        "pred1 = np.argmax(modelRN.predict(X_test), axis=-1)\n",
        "print(accuracy_score(y_test, pred1))\n",
        "print(recall_score(y_test, pred1, average='micro'))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FHa1uqho6k5m",
        "outputId": "3868a3fc-c6f4-46b3-d448-622ba03d1173"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1632/1632 [==============================] - 23s 14ms/step - loss: 0.0944 - accuracy: 0.9829\n",
            "69/69 [==============================] - 2s 24ms/step - loss: 0.5364 - accuracy: 0.9206\n",
            "198/198 [==============================] - 3s 16ms/step - loss: 0.6191 - accuracy: 0.9178\n",
            "0.9178147268408551\n",
            "0.9178147268408551\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(historyRN.history['accuracy'])\n",
        "plt.plot(historyRN.history['val_accuracy'])\n",
        "plt.title('model accuracy')\n",
        "plt.ylabel('accuracy')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'val'], loc='upper left')\n",
        "plt.show()\n",
        "\n",
        "plt.plot(historyRN.history['loss'])\n",
        "plt.plot(historyRN.history['val_loss'])\n",
        "plt.title('model loss')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'val'], loc='upper left')\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "uuqO1Lng6lRl",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 573
        },
        "outputId": "14710345-facf-457e-e009-e438ec4ee9f0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#Efficient Model\n",
        "base_modelEN=tf.keras.applications.efficientnet.EfficientNetB0(input_shape = (32, 32, 3), classes=43,include_top = False, weights = 'imagenet', pooling='max')\n",
        "  \n",
        "modelEN = Sequential()\n",
        "modelEN.add(base_modelEN)\n",
        "modelEN.add(Dense(43, activation='sigmoid'))\n",
        "\n",
        "modelEN.compile(optimizer='adam', loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False), metrics = ['accuracy'])\n",
        "\n",
        "plot_model(modelEN, to_file='/content/drive/MyDrive/modelEN.png', show_shapes=True, show_layer_names=True)\n",
        "modelEN.summary()\n",
        "\n",
        "historyEN = modelEN.fit(X_train,y_train, validation_data = (X_valid, y_valid), steps_per_epoch = 100, epochs = 20)\n",
        "modelEN.save('/content/drive/MyDrive/modelEN.h5')\n"
      ],
      "metadata": {
        "id": "CEN1aySgPky3",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "149495ec-213c-49ae-9717-b25ef0d62a28"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_13\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " efficientnetb0 (Functional)  (None, 1280)             4049571   \n",
            "                                                                 \n",
            " dense_13 (Dense)            (None, 43)                55083     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 4,104,654\n",
            "Trainable params: 4,062,631\n",
            "Non-trainable params: 42,023\n",
            "_________________________________________________________________\n",
            "Epoch 1/20\n",
            "100/100 [==============================] - 32s 228ms/step - loss: 0.6732 - accuracy: 0.8116 - val_loss: 1.0744 - val_accuracy: 0.7181\n",
            "Epoch 2/20\n",
            "100/100 [==============================] - 22s 216ms/step - loss: 0.0320 - accuracy: 0.9900 - val_loss: 0.6632 - val_accuracy: 0.8302\n",
            "Epoch 3/20\n",
            "100/100 [==============================] - 21s 214ms/step - loss: 0.0132 - accuracy: 0.9958 - val_loss: 0.6456 - val_accuracy: 0.8574\n",
            "Epoch 4/20\n",
            "100/100 [==============================] - 22s 221ms/step - loss: 0.0106 - accuracy: 0.9967 - val_loss: 0.6745 - val_accuracy: 0.8624\n",
            "Epoch 5/20\n",
            "100/100 [==============================] - 23s 228ms/step - loss: 0.0075 - accuracy: 0.9978 - val_loss: 0.7900 - val_accuracy: 0.8549\n",
            "Epoch 6/20\n",
            "100/100 [==============================] - 21s 214ms/step - loss: 0.0066 - accuracy: 0.9981 - val_loss: 0.7704 - val_accuracy: 0.8730\n",
            "Epoch 7/20\n",
            "100/100 [==============================] - 22s 225ms/step - loss: 0.0068 - accuracy: 0.9978 - val_loss: 0.7640 - val_accuracy: 0.8728\n",
            "Epoch 8/20\n",
            "100/100 [==============================] - 23s 231ms/step - loss: 0.0069 - accuracy: 0.9978 - val_loss: 0.6981 - val_accuracy: 0.8882\n",
            "Epoch 9/20\n",
            "100/100 [==============================] - 23s 227ms/step - loss: 0.0053 - accuracy: 0.9984 - val_loss: 0.6050 - val_accuracy: 0.9048\n",
            "Epoch 10/20\n",
            "100/100 [==============================] - 22s 220ms/step - loss: 0.0043 - accuracy: 0.9987 - val_loss: 0.7516 - val_accuracy: 0.8744\n",
            "Epoch 11/20\n",
            "100/100 [==============================] - 23s 227ms/step - loss: 0.0065 - accuracy: 0.9980 - val_loss: 0.7678 - val_accuracy: 0.8794\n",
            "Epoch 12/20\n",
            "100/100 [==============================] - 23s 227ms/step - loss: 0.0036 - accuracy: 0.9989 - val_loss: 0.7598 - val_accuracy: 0.8878\n",
            "Epoch 13/20\n",
            "100/100 [==============================] - 22s 215ms/step - loss: 0.0058 - accuracy: 0.9983 - val_loss: 0.7741 - val_accuracy: 0.8776\n",
            "Epoch 14/20\n",
            "100/100 [==============================] - 23s 230ms/step - loss: 0.0035 - accuracy: 0.9989 - val_loss: 0.6223 - val_accuracy: 0.8932\n",
            "Epoch 15/20\n",
            "100/100 [==============================] - 23s 228ms/step - loss: 0.0039 - accuracy: 0.9989 - val_loss: 0.5864 - val_accuracy: 0.9082\n",
            "Epoch 16/20\n",
            "100/100 [==============================] - 22s 219ms/step - loss: 0.0026 - accuracy: 0.9992 - val_loss: 0.6360 - val_accuracy: 0.9093\n",
            "Epoch 17/20\n",
            "100/100 [==============================] - 21s 213ms/step - loss: 0.0035 - accuracy: 0.9990 - val_loss: 0.5467 - val_accuracy: 0.9068\n",
            "Epoch 18/20\n",
            "100/100 [==============================] - 21s 214ms/step - loss: 0.0048 - accuracy: 0.9987 - val_loss: 0.7428 - val_accuracy: 0.8955\n",
            "Epoch 19/20\n",
            "100/100 [==============================] - 23s 228ms/step - loss: 0.0054 - accuracy: 0.9984 - val_loss: 0.6690 - val_accuracy: 0.9041\n",
            "Epoch 20/20\n",
            "100/100 [==============================] - 23s 226ms/step - loss: 0.0041 - accuracy: 0.9987 - val_loss: 0.5206 - val_accuracy: 0.9107\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "results = modelEN.evaluate(X_train, y_train, batch_size=64)\n",
        "results = modelEN.evaluate(X_valid, y_valid, batch_size=64)\n",
        "results = modelEN.evaluate(X_test, y_test, batch_size=64)\n",
        "\n",
        "pred1 = np.argmax(modelEN.predict(X_test), axis=-1)\n",
        "print(accuracy_score(y_test, pred1))\n",
        "print(recall_score(y_test, pred1, average='micro'))"
      ],
      "metadata": {
        "id": "63yVZM7fEtZV",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2183a4a0-6069-49aa-8327-d5890c7d0fc2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1632/1632 [==============================] - 22s 13ms/step - loss: 0.0077 - accuracy: 0.9976\n",
            "69/69 [==============================] - 1s 19ms/step - loss: 0.5206 - accuracy: 0.9107\n",
            "198/198 [==============================] - 3s 13ms/step - loss: 0.4565 - accuracy: 0.9196\n",
            "0.919477434679335\n",
            "0.919477434679335\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(historyEN.history['accuracy'])\n",
        "plt.plot(historyEN.history['val_accuracy'])\n",
        "plt.title('model accuracy')\n",
        "plt.ylabel('accuracy')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'val'], loc='upper left')\n",
        "plt.show()\n",
        "\n",
        "plt.plot(historyEN.history['loss'])\n",
        "plt.plot(historyEN.history['val_loss'])\n",
        "plt.title('model loss')\n",
        "plt.ylabel('loss')\n",
        "plt.xlabel('epoch')\n",
        "plt.legend(['train', 'val'], loc='upper left')\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "jJjCKUPHEyZC",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 573
        },
        "outputId": "d6ffa005-df5a-410b-8f41-ea653ff85748"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.datasets import make_blobs\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from keras.models import load_model\n",
        "from tensorflow.keras.utils import to_categorical\n",
        "from numpy import dstack\n",
        "\n",
        "# load models from file\n",
        "\n",
        "\n",
        "def load_all_models():\n",
        "  all_models = list()\n",
        "  model = load_model('/content/drive/MyDrive/modelCNN.h5')\n",
        "  all_models.append(model)\n",
        "  model = load_model('/content/drive/MyDrive/modelMN.h5')\n",
        "  all_models.append(model)\n",
        "  model = load_model('/content/drive/MyDrive/modelDN.h5')\n",
        "  all_models.append(model)\n",
        "  model = load_model('/content/drive/MyDrive/modelRN.h5')\n",
        "  all_models.append(model)\n",
        "  model = load_model('/content/drive/MyDrive/modelEN.h5')\n",
        "  all_models.append(model)\n",
        "  return all_models\n",
        "\n",
        "def stacked_dataset(members, inputX):\n",
        "\tstackX = None\n",
        "\tfor model in members:\n",
        "\t\t# make prediction\n",
        "\t\tyhat = model.predict(inputX, verbose=0)\n",
        "\t\t# stack predictions into [rows, members, probabilities]\n",
        "\t\tif stackX is None:\n",
        "\t\t\tstackX = yhat\n",
        "\t\telse:\n",
        "\t\t\tstackX = dstack((stackX, yhat))\n",
        "\t# flatten predictions to [rows, members x probabilities]\n",
        "\tstackX = stackX.reshape((stackX.shape[0], stackX.shape[1]*stackX.shape[2]))\n",
        "\treturn stackX\n",
        "\n",
        "  # fit a model based on the outputs from the ensemble members\n",
        "def fit_stacked_model(members, inputX, inputy):\n",
        "\t# create dataset using ensemble\n",
        "\tstackedX = stacked_dataset(members, inputX)\n",
        "\t# fit standalone model\n",
        "\tmodel = LogisticRegression()\n",
        "\tmodel.fit(stackedX, inputy)\n",
        "\treturn model\n",
        "def stacked_prediction(members, model, inputX):\n",
        "\t# create dataset using ensemble\n",
        "\tstackedX = stacked_dataset(members, inputX)\n",
        "\t# make a prediction\n",
        "\tyhat = model.predict(stackedX)\n",
        "\treturn yhat\n",
        "\n",
        "members = load_all_models()\n",
        "# evaluate standalone models on test dataset\n",
        "for model in members:\n",
        "\t_, acc = model.evaluate(X_test, y_test, verbose=0)\n",
        "\tprint('Model Accuracy: %.3f' % acc)\n",
        "\n",
        "\n",
        "# fit stacked model using the ensemble\n",
        "model = fit_stacked_model(members, X_test, y_test)\n",
        "# create stacked model input dataset as outputs from the ensemble\n",
        "# evaluate model on test set\n",
        "yhat = stacked_prediction(members, model, X_test)\n",
        "acc = accuracy_score(y_test, yhat)\n",
        "print('Stacked Test Accuracy: %.3f' % acc)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GUQnuK_MjPMj",
        "outputId": "4b6d14d3-9e0f-42a0-8c9b-db1eff828d88"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model Accuracy: 0.964\n",
            "Model Accuracy: 0.944\n",
            "Model Accuracy: 0.950\n",
            "Model Accuracy: 0.918\n",
            "Model Accuracy: 0.920\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/linear_model/_logistic.py:818: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
            "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
            "\n",
            "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
            "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
            "Please also refer to the documentation for alternative solver options:\n",
            "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
            "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG,\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Stacked Test Accuracy: 1.000\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "  def most_frequent(List):\n",
        "    counter = 0\n",
        "    num = List[0]\n",
        "     \n",
        "    for i in List:\n",
        "        curr_frequency = List.count(i)\n",
        "        if(curr_frequency> counter):\n",
        "            counter = curr_frequency\n",
        "            num = i\n",
        " \n",
        "    return num\n",
        "  \n",
        "  \n",
        "  img = image.load_img('/content/drive/MyDrive/00008.png', target_size = (32,32)) #load the image\n",
        "  x = image.img_to_array(img)\n",
        "  #x = np.expand_dims(x, axis=0)\n",
        "  #images = np.vstack([x])\n",
        "  image1=np.reshape(x,(1,32,32,3)) \n",
        "  valu=[]\n",
        "  valu.append(np.argmax(members[0].predict(image1),axis=-1)) #predict the label for the image\n",
        "  valu.append(np.argmax(members[1].predict(image1),axis=-1))\n",
        "  valu.append(np.argmax(members[2].predict(image1),axis=-1))\n",
        "  valu.append(np.argmax(members[3].predict(image1),axis=-1))\n",
        "  valu.append(np.argmax(members[4].predict(image1),axis=-1))\n",
        "  print(valu[0],valu[1],valu[2],valu[3],valu[4])\n",
        "  classes=most_frequent(valu)\n",
        "  if classes[0]==0:\n",
        "    print('Speed limit (20km/h)') #print the content\n",
        "\n",
        "  elif classes[0]==1:\n",
        "      print('Speed limit (30km/h)') #print the content\n",
        "\n",
        "  elif classes[0]==2:\n",
        "      print('Speed limit (50km/h)') #print the content\n",
        "\n",
        "  elif classes[0]==3:\n",
        "      print(' Speed limit (60km/h)') #print the content\n",
        "\n",
        "  elif classes[0]==4:\n",
        "      print('Speed limit (70km/h)') #print the content\n",
        "\n",
        "  elif classes[0]==5:\n",
        "      print('Speed limit (80km/h)') #print the content\n",
        "\n",
        "  elif classes[0]==6:\n",
        "      print('End of speed limit (80km/h)') #print the content\n",
        "\n",
        "  elif classes[0]==7:\n",
        "      print('Speed limit (100km/h)') #print the content\n",
        "\n",
        "  elif classes[0]==8:\n",
        "      print('Speed limit (120km/h)') #print the content\n",
        "\n",
        "  elif classes[0]==9:\n",
        "      print('No passing') #print the content\n",
        "        \n",
        "  elif classes[0]==10:\n",
        "      print('No passing veh over 3.5 tons') #print the content \n",
        "        \n",
        "  elif classes[0]==11:\n",
        "      print(', Right-of-way at intersection') #print the content\n",
        "        \n",
        "  elif classes[0]==12:\n",
        "      print('Priority road') #print the content\n",
        "        \n",
        "  elif classes[0]==13:\n",
        "      print('Yield') #print the content       \n",
        "        \n",
        "  elif classes[0]==14:\n",
        "      print('Stop') #print the content       \n",
        "        \n",
        "  elif classes[0]==15:\n",
        "      print('No vehicles') #print the content\n",
        "                \n",
        "  elif classes[0]==16:\n",
        "      print('Veh > 3.5 tons prohibited') #print the content       \n",
        "        \n",
        "  elif classes[0]==17:\n",
        "      print('No entry') #print the content        \n",
        "                       \n",
        "  elif classes[0]==18:\n",
        "      print('General caution') #print the content       \n",
        "                \n",
        "  elif classes[0]==19:\n",
        "      print('Dangerous curve left') #print the content        \n",
        "        \n",
        "  elif classes[0]==20:\n",
        "      print('Dangerous curve right') #print the content        \n",
        "        \n",
        "  elif classes[0]==21:\n",
        "      print('Double curve') #print the content       \n",
        "        \n",
        "  elif classes[0]==22:\n",
        "      print('Bumpy road') #print the content        \n",
        "                \n",
        "  elif classes[0]==23:\n",
        "      print('Slippery road') #print the content        \n",
        "        \n",
        "  elif classes[0]==24:\n",
        "      print('Road narrows on the right') #print the content               \n",
        "        \n",
        "  elif classes[0]==25:\n",
        "      print('Road work') #print the content        \n",
        "        \n",
        "  elif classes[0]==26:\n",
        "      print('Traffic signals') #print the content\n",
        "        \n",
        "  elif classes[0]==27:\n",
        "      print('Pedestrians') #print the content       \n",
        "                \n",
        "  elif classes[0]==28:\n",
        "      print('Children crossing') #print the content\n",
        "        \n",
        "  elif classes[0]==29:\n",
        "      print( 'Bicycles crossing') #print the content\n",
        "               \n",
        "  elif classes[0]==30:\n",
        "      print('Beware of ice/snow') #print the content        \n",
        "        \n",
        "  elif classes[0]==31:\n",
        "      print('Wild animals crossing') #print the content        \n",
        "               \n",
        "  elif classes[0]==32:\n",
        "      print('End speed + passing limits') #print the content        \n",
        "        \n",
        "  elif classes[0]==33:\n",
        "      print('Turn right ahead') #print the content\n",
        "        \n",
        "  elif classes[0]==34:\n",
        "      print('Turn left ahead') #print the content        \n",
        "        \n",
        "  elif classes[0]==35:\n",
        "      print('Ahead only') #print the content        \n",
        "        \n",
        "  elif classes[0]==36:\n",
        "      print('Go straight or right') #print the content\n",
        "        \n",
        "  elif classes[0]==37:\n",
        "      print('Go straight or left') #print the content        \n",
        "        \n",
        "  elif classes[0]==38:\n",
        "      print('Keep right') #print the content\n",
        "        \n",
        "  elif classes[0]==39:\n",
        "      print('Keep left') #print the content       \n",
        "        \n",
        "  elif classes[0]==40:\n",
        "      print('Roundabout mandatory') #print the content\n",
        "\n",
        "  elif classes[0]==41:\n",
        "      print('End of no passing') #print the content        \n",
        "        \n",
        "  else:\n",
        "      print('End no passing veh > 3.5 tons') #print the content"
      ],
      "metadata": {
        "id": "a0OwFfeo2EIX",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "632678ab-7200-4a1b-a98c-b58eefbec1d3"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[25] [25] [25] [25] [25]\n",
            "Road work\n"
          ]
        }
      ]
    }
  ]
}