Various_CNN_Architectures.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 34
    },
    "colab_type": "code",
    "id": "4AS5GyQLVzam",
    "outputId": "75c888fb-b90e-4870-9176-d3234e9faa4c"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "# Importing the Required Libraries\n",
    "\n",
    "import keras\n",
    "from keras.datasets import mnist\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout, Flatten\n",
    "from keras.layers import Conv2D, MaxPooling2D\n",
    "from keras import backend as K\n",
    "from keras.utils import to_categorical\n",
    "from keras.initializers import he_normal\n",
    "from keras.layers.normalization import BatchNormalization\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "K.set_image_data_format('channels_last') # set the data format for our model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 51
    },
    "colab_type": "code",
    "id": "VFMaDic7WlLS",
    "outputId": "38fe3853-97db-45cc-9a21-bcfaa2441574"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://s3.amazonaws.com/img-datasets/mnist.npz\n",
      "11493376/11490434 [==============================] - 4s 0us/step\n"
     ]
    }
   ],
   "source": [
    "# Loading MNIST Dataset\n",
    "\n",
    "(x_train, y_train), (x_test, y_test) = mnist.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 68
    },
    "colab_type": "code",
    "id": "HZyO0_NEWtq1",
    "outputId": "7674a91a-9198-44d9-a4dc-406af8567495"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_train shape: (60000, 28, 28, 1)\n",
      "60000 train samples\n",
      "10000 test samples\n"
     ]
    }
   ],
   "source": [
    "# Data Preprocessing Steps\n",
    "\n",
    "# Given dimensions of image is 28X28X1 a grayscale image\n",
    "\n",
    "x_train = x_train.reshape(x_train.shape[0], 28, 28, 1)\n",
    "x_test = x_test.reshape(x_test.shape[0], 28, 28, 1)\n",
    "input_shape = (28, 28, 1)\n",
    "\n",
    "x_train = x_train.astype('float32')\n",
    "x_test = x_test.astype('float32')\n",
    "x_train /= 255\n",
    "x_test /= 255\n",
    "print('x_train shape:', x_train.shape)\n",
    "print(x_train.shape[0], 'train samples')\n",
    "print(x_test.shape[0], 'test samples')\n",
    "\n",
    "# convert class vectors to binary class matrices\n",
    "y_train = to_categorical(y_train, 10)\n",
    "y_test = to_categorical(y_test, 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "K0dSXyC8Ydtl"
   },
   "source": [
    "#MODEL-1 CNN with 3 Layers\n",
    "\n",
    "---\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 527
    },
    "colab_type": "code",
    "id": "MV8nJjdnX7Mz",
    "outputId": "d8764041-87fb-4416-9e24-afc9c64a82ed"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 60000 samples, validate on 10000 samples\n",
      "Epoch 1/50\n",
      "60000/60000 [==============================] - 14s 235us/step - loss: 0.0347 - acc: 0.9881 - val_loss: 0.0072 - val_acc: 0.9975\n",
      "Epoch 2/50\n",
      "60000/60000 [==============================] - 11s 185us/step - loss: 0.0118 - acc: 0.9962 - val_loss: 0.0051 - val_acc: 0.9982\n",
      "Epoch 3/50\n",
      "60000/60000 [==============================] - 11s 185us/step - loss: 0.0081 - acc: 0.9973 - val_loss: 0.0047 - val_acc: 0.9983\n",
      "Epoch 4/50\n",
      "60000/60000 [==============================] - 11s 185us/step - loss: 0.0066 - acc: 0.9978 - val_loss: 0.0045 - val_acc: 0.9984\n",
      "Epoch 5/50\n",
      "60000/60000 [==============================] - 11s 184us/step - loss: 0.0056 - acc: 0.9982 - val_loss: 0.0039 - val_acc: 0.9987\n",
      "Epoch 6/50\n",
      "60000/60000 [==============================] - 11s 185us/step - loss: 0.0046 - acc: 0.9984 - val_loss: 0.0053 - val_acc: 0.9984\n",
      "Epoch 7/50\n",
      "60000/60000 [==============================] - 11s 185us/step - loss: 0.0041 - acc: 0.9986 - val_loss: 0.0037 - val_acc: 0.9987\n",
      "Epoch 8/50\n",
      "60000/60000 [==============================] - 11s 186us/step - loss: 0.0034 - acc: 0.9989 - val_loss: 0.0041 - val_acc: 0.9987\n",
      "Epoch 9/50\n",
      "60000/60000 [==============================] - 11s 187us/step - loss: 0.0039 - acc: 0.9987 - val_loss: 0.0032 - val_acc: 0.9989\n",
      "Epoch 10/50\n",
      "60000/60000 [==============================] - 11s 186us/step - loss: 0.0029 - acc: 0.9990 - val_loss: 0.0039 - val_acc: 0.9986\n",
      "Epoch 11/50\n",
      "60000/60000 [==============================] - 11s 185us/step - loss: 0.0028 - acc: 0.9991 - val_loss: 0.0039 - val_acc: 0.9987\n",
      "Epoch 12/50\n",
      "60000/60000 [==============================] - 11s 185us/step - loss: 0.0027 - acc: 0.9991 - val_loss: 0.0035 - val_acc: 0.9989\n",
      "Epoch 13/50\n",
      "60000/60000 [==============================] - 11s 185us/step - loss: 0.0023 - acc: 0.9992 - val_loss: 0.0039 - val_acc: 0.9988\n",
      "Epoch 14/50\n",
      "60000/60000 [==============================] - 11s 186us/step - loss: 0.0022 - acc: 0.9992 - val_loss: 0.0036 - val_acc: 0.9990\n",
      "Epoch 00014: early stopping\n"
     ]
    }
   ],
   "source": [
    "from keras.callbacks import EarlyStopping\n",
    "\n",
    "early_stoppping = EarlyStopping(monitor='val_loss',patience = 5,verbose=1)\n",
    "\n",
    "model_1 = Sequential()\n",
    "\n",
    "model_1.add(Conv2D(32, kernel_size=(3, 3),activation='relu',input_shape=(28,28,1)))\n",
    "\n",
    "model_1.add(Conv2D(64, (3, 3), activation='relu'))\n",
    "\n",
    "model_1.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "\n",
    "model_1.add(Dropout(0.25))\n",
    "\n",
    "model_1.add(Conv2D(128, (3, 3), activation='relu'))\n",
    "\n",
    "model_1.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "\n",
    "model_1.add(Flatten())\n",
    "\n",
    "model_1.add(Dense(256, activation='relu',kernel_initializer=he_normal(seed=None)))\n",
    "\n",
    "model_1.add(Dropout(0.5))\n",
    "\n",
    "model_1.add(Dense(10, activation='softmax'))\n",
    "\n",
    "model_1.compile(loss='binary_crossentropy', optimizer=\"adam\", metrics=['accuracy'])\n",
    "\n",
    "my_model_1 = model_1.fit(x_train, y_train,batch_size=128,epochs=50,verbose=1,validation_data=(x_test, y_test),callbacks=[early_stoppping])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 299
    },
    "colab_type": "code",
    "id": "ZklJMnC0wVQB",
    "outputId": "ee715ef8-f33e-42c5-eeb2-02a2f7b14dcf"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test score: 0.003630750371064505\n",
      "Test accuracy: 0.9989699977874756\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD4CAYAAAAAczaOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEA19pZSKIf8PlkcgzxZRf9wAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "score_1 = model_1.evaluate(x_test, y_test, verbose=0) \n",
    "print('Test score:', score_1[0]) \n",
    "print('Test accuracy:', score_1[1])\n",
    "\n",
    "plt.plot(my_model_1.history['acc'], 'r')\n",
    "plt.plot(my_model_1.history['val_acc'], 'b')\n",
    "plt.legend({'training accuracy': 'r', 'validation accuracy':'b'})\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 265
    },
    "colab_type": "code",
    "id": "qhbCk7pIx-4H",
    "outputId": "e1cfd76d-8a8b-4fd5-f372-a6240d77a35b"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD4CAYAAAAAczaOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(my_model_1.history['loss'], 'r')\n",
    "plt.plot(my_model_1.history['val_loss'], 'b')\n",
    "plt.legend({'training loss': 'r', 'validation loss':'b'})\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "xMsArRUx6cyv"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "o_miUJJa7VHK"
   },
   "source": [
    "# MODEL-2 CNN with 5 Layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 442
    },
    "colab_type": "code",
    "id": "TiCA6Q8r7Xvf",
    "outputId": "89b36f9c-851b-4c6f-99d9-90d52bb17952"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 60000 samples, validate on 10000 samples\n",
      "Epoch 1/12\n",
      "60000/60000 [==============================] - 9s 158us/step - loss: 0.4075 - acc: 0.8695 - val_loss: 0.0538 - val_acc: 0.9829\n",
      "Epoch 2/12\n",
      "60000/60000 [==============================] - 8s 140us/step - loss: 0.0980 - acc: 0.9725 - val_loss: 0.0440 - val_acc: 0.9867\n",
      "Epoch 3/12\n",
      "60000/60000 [==============================] - 8s 141us/step - loss: 0.0728 - acc: 0.9794 - val_loss: 0.0255 - val_acc: 0.9923\n",
      "Epoch 4/12\n",
      "60000/60000 [==============================] - 8s 141us/step - loss: 0.0602 - acc: 0.9825 - val_loss: 0.0237 - val_acc: 0.9927\n",
      "Epoch 5/12\n",
      "60000/60000 [==============================] - 8s 141us/step - loss: 0.0548 - acc: 0.9843 - val_loss: 0.0217 - val_acc: 0.9934\n",
      "Epoch 6/12\n",
      "60000/60000 [==============================] - 8s 142us/step - loss: 0.0460 - acc: 0.9871 - val_loss: 0.0291 - val_acc: 0.9916\n",
      "Epoch 7/12\n",
      "60000/60000 [==============================] - 9s 143us/step - loss: 0.0431 - acc: 0.9878 - val_loss: 0.0271 - val_acc: 0.9909\n",
      "Epoch 8/12\n",
      "60000/60000 [==============================] - 8s 141us/step - loss: 0.0403 - acc: 0.9885 - val_loss: 0.0270 - val_acc: 0.9929\n",
      "Epoch 9/12\n",
      "60000/60000 [==============================] - 8s 141us/step - loss: 0.0389 - acc: 0.9892 - val_loss: 0.0219 - val_acc: 0.9944\n",
      "Epoch 10/12\n",
      "60000/60000 [==============================] - 9s 142us/step - loss: 0.0345 - acc: 0.9905 - val_loss: 0.0252 - val_acc: 0.9923\n",
      "Epoch 11/12\n",
      "60000/60000 [==============================] - 9s 142us/step - loss: 0.0316 - acc: 0.9913 - val_loss: 0.0243 - val_acc: 0.9930\n",
      "Epoch 12/12\n",
      "60000/60000 [==============================] - 9s 142us/step - loss: 0.0303 - acc: 0.9912 - val_loss: 0.0174 - val_acc: 0.9942\n"
     ]
    }
   ],
   "source": [
    "model_2 = Sequential()\n",
    "\n",
    "\n",
    "model_2.add(Conv2D(8, kernel_size=(5, 5),padding='same',activation='relu',input_shape=(28,28,1)))\n",
    "\n",
    "\n",
    "model_2.add(Conv2D(16, (5, 5), activation='relu'))\n",
    "\n",
    "\n",
    "model_2.add(MaxPooling2D(pool_size=(2, 2),padding='same'))\n",
    "\n",
    "\n",
    "model_2.add(Dropout(0.25))\n",
    "\n",
    "\n",
    "model_2.add(Conv2D(32, (5, 5),padding='same', activation='relu'))\n",
    "\n",
    "\n",
    "model_2.add(MaxPooling2D(pool_size=(2, 2),padding='same'))\n",
    "\n",
    "\n",
    "model_2.add(Dropout(0.25))\n",
    "\n",
    "\n",
    "model_2.add(Conv2D(64, (5, 5),padding='same',activation='relu'))\n",
    "\n",
    "\n",
    "model_2.add(Conv2D(64, (5, 5), activation='relu'))\n",
    "\n",
    "\n",
    "model_2.add(MaxPooling2D(pool_size=(2, 2),padding='same'))\n",
    "\n",
    "\n",
    "model_2.add(Dropout(0.25))\n",
    "\n",
    "\n",
    "model_2.add(Flatten())\n",
    "\n",
    "\n",
    "model_2.add(Dense(256, activation='relu',kernel_initializer=he_normal(seed=None)))\n",
    "\n",
    "\n",
    "model_2.add(BatchNormalization())\n",
    "\n",
    "\n",
    "model_2.add(Dropout(0.5))\n",
    "\n",
    "\n",
    "model_2.add(Dense(10, activation='softmax'))\n",
    "\n",
    "\n",
    "model_2.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "\n",
    "my_model_2 = model_2.fit(x_train, y_train,batch_size=128,epochs=12,verbose=1,validation_data=(x_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 302
    },
    "colab_type": "code",
    "id": "yjxTCF_28YZk",
    "outputId": "c9afa289-f1d4-4a77-9b61-3d3e6e1d9402"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test score: 0.017415812282267688\n",
      "Test accuracy: 0.9942\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD7CAYAAACRxdTpAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzt3Xl8VPW9//HXLNkXCGEghEVEwodF\nFAEValEUWxdEW8XaXuveeu3Vlu7XVltxpfWKGxQKdWv9VW1vq177EAXBigvtrYLXKiRfimxCWAYI\nEJKQZGbO748zM5nsQzJhMmc+z8cjj5mzzXy/Wd7zzfd8z/e4LMtCKaWUs7iTXQCllFKJp+GulFIO\npOGulFIOpOGulFIOpOGulFIOpOGulFIO5I1nJxE5Gfgf4BFjzMIW284HHgCCwDJjzL3h9Y8AUwAL\nmGOMeT+RBVdKKdW+TsNdRPKABcCqdnZ5HLgA2AmsFpE/Az6gzBgzVUTGAE8BUxNTZKWUUp2Jp+Ve\nD1wM/GfLDSIyAjhgjPksvLwMmIEd7i8DGGPKRaRIRAqNMYfbexO/v7rLV1MVFeVSVVXb1cN7NSfX\nDZxdP61b6kql+vl8Ba621nca7saYABAQkbY2lwD+mOW9wElAf2BtzHp/eN92w72oKBev19NZcdrl\n8xV0+djezsl1A2fXT+uWulK9fnH1uR+DNj9BOlgf1Z1PSZ+vAL+/usvH92ZOrhs4u35at9SVSvVr\n70Oou+Feid0ijxgcXtfQYn0psKub76WUUipO3RoKaYzZChSKyHAR8QKXACvCX7MBRGQiUGmMSY2P\nQaWUcoB4RstMAuYDw4FGEZkNvAJsMca8BHwLeD68+x+MMRuBjSKyVkTWACHg1p4ovFJKqbbFc0J1\nLTC9g+1v08YwR2PM7d0qmVJKqS7TK1SVUsqBNNyVUsqBEj0UUinlAJYFNTWwb58r5svN/v3280AA\n+vTJZOhQi8GDQwwdGmLwYIvs7GSXvHeLfF+rqlwcPGh/ZWTAmWcGcXU6YPzYaLirpAiFoL7e/mVX\nx0djI+zf78Lvd7UIbTu4W66rq+ssbbJarfH5QgwZYjFkiP0YCf0hQ+wPgD59SHiIJUNDA9FwPngw\n9rmLqioXhw61fGzaJxBo/Q14660axo4NJbSMGu6qlVAI6uqgrs5FXR3U1rqaLdfUNF+uq3NRW9t8\n/9jlltvtR/sXPDMTSkryGDjQYtCgECUlVrPnJSX2Y35+kr8pvVRjI+za5aKy0o3f315w2+F98GDn\nqZqZaeHzWZSVhejf34r5spd9Pnt5yJA8NmyoZccOF5995mbnTnf0+fr1bj78sO2rzfPzmwd+0wdA\niKFD7Z+9u4c6i4NB2vx9jl2urbWXg0HYuTOLgwdpM6xra+P/hPJ6Lfr2tSgqshg+3H7s06fpcdiw\nEKNGJTbYQcO9V/P7Xfztb7B3r5eGBvsPuaHBFX6EQMDV5vrGRmhsjF1uvl8gQDvH2ft13mKLn8tl\nkZsLOTn244ABFjk59ldWFtTUeNmxA9atcxMMtj/9RH6+HfSDBtkBEHlufxjYzwcMsF/TKUIh+3dg\n504XO3e6o4+VlU3Le/e6sKz2f14ul0Vxsf39Ovlkq0Vgtw7u/Pz4WtY+HxQXBzss92efudixww59\n+7HpeXl522+SkWFRWtoU/JEWf0EBrRoQLRsOTY9th3d9/bH+Xmc2WyostEN65MhQs3AuKrLX9+1L\n+LH5V15ecv5b0XDvZXbudLFsmZdXX/Xy9797CIUAchL2+l6vRWYmZGTYf0iZmeD1Qm6uRUaGHY65\nuRY5OU2BbIexvT52ue3tTdtyciArq+NfbPsy7xqCQbt/d88eF7t2udi9283u3a7wV9PzTZs6nn+o\nuDjS4m9q9UeeDxhg/6HFljs7mx5rKXbEsuDQIVqF9o4dTeG9a5eLxsb2Q3DQIIspU4IMHmz3ew8Y\n0Dq4+/Wz8HR9yqYucbth4ED7Q3jy5LZbpIcPE27x2639HTtin7t4772uR5PH0/z3ctCgUIe/t5Fl\n+3fXXjdsWA5QEw3uwkL77ySVuKxe0unZnVkhU2keiLZs2uTi1VczWLbMG/131uWyOP30IJde6sXr\nPRoO5KYwzsy0yMggGtSR5djQbr0+OUHWkWP92dXXw9699gfAnj126Ec+DGI/GI4cib+pZP9xN/9Q\nausDruVjZ9tzc/P5+OPaVqEdCfP2/rV3uez/QoYMsSgtDUVbsqWldogPHmy3spP5s+zpv7n6eqLf\ntx07XNTUuOJudGR4LVyNDbjqanHV1kYfqa3DVVuDq671IzH7uWpryc70cLS+MaZELX5WLVss3VgO\nFfWj5vY7ITe3S9+rLs8KqRLPsuCTT9y8+qrdQjfGDnSv1+KccwLMnBngoosCDBxohf+IGjt5xfSR\nlQVDh1oMHWphX/zctiNHiLb6I4Hv93d+LqCqyu4Gqaujw+6O+LX+gy0qsjjxxFC0xT14sB3ikTAv\nKbE/nNNOIIDrwAHcVQfIP7Af2b+fMQf246o6gPvQIVx1tc1DOBLO4dCmrq4pzINtdxkdi+M18Mfy\neDh6zfUEy0Yl9HU13I+TUAjef9/Dq696WbbMy/btdrMrO9viwgsbmTkzwBe/GKCoKMkFdYj8fBg5\n0mLkyK79kVuW3XpsGf4tTya3dTIuslxQkEG/fvUxrW47vPPyElzZ3igQwFVVhbvqAO4D+3Ht328/\nVh3AHXl+IPJ4APeBA7gPHTzmt7EyM7Fyc7FycrEKCggNGAi5uVg5OVi5efZjXl50mZycpv1bPJIX\nXs7JoXhAH/bvPxJ+kxadCgletvLzsfoVH3PdO6Ph3oMaG+G99+xAf+01L3v32oGen29x+eV2oJ97\nbkBHgvRCLhdkZxMetx35Yzy2nkOfLwO/vyHRRetZoRAcPWq3iuvCreOjdXaXRuyyO0jOtkrcB2KC\nen84vA/sx30wvqC2MjII9SsmVDqYwPhTsIr62cvF/bD6FRMq6keouBirsE9TWOfmQW6OHcg91RHu\nKyCUkbpdvaDhnnB1dfDWW3Z3y4oV3ujws+LiEFdf3cDMmQGmTQs6alSHOk4CgXA3RK3dDVFTG33u\nqq3FdbQuGsrRLoqjR6OhTDiYXTFBTcwxdnAfjbs4LdskltdrB/OgUgLjxrcZ1FZxsb2uXzFWv35Y\n+QXOGPjeC2m4J0B1Nbzxhh3oq1Z5oyfKBg0KMXu23UI/88xgyp1tV8coPJDadfSo3brd78a7Y29T\nGNfW4qqpiZ60iw3q2HW0t70hcf8FWF5vtAuCnBxChSV2qzi6LhcrOzvcbRFuJefkYGXnYOXkUFBS\nzCFvTvOgLijUoO5FNG66aN8+F8uX24H+9tseGhrsX+oTTwxxySV2C33ChFCvG52Stmpr8ez4LNyq\ntcM3GsJHjza1ciPLdS22H63DVddiObxfdLmx9Ynvrp5CsXLz7L7g3DxCg0qb9w+H11u5uRDdL7wu\nHMiRbgsrJpCj67Nz7KFT3VDgK6AhhUeopQMN9zg1NMD69W7+8Q8Pr7/u5W9/8xAK2YE+blyQmTPt\nUS6jR4e08ZJErurDeDYaPP/aiNdU4NlYgdcY3J9tw5WAYb+W2w3ZOVg52VjZOYQKC2HAwHCo2iFK\ndg5WdjbZRYXUujKw8joI4zYeycnRFrDqNg33NlgW7NjhYu1aT/Tr44/dza5wmzw5yMyZjVx8cYAT\nT+wd1wqkE9eB/Xg2bsS7MRzgG40d6pU7W+0b6u+j8XOfJ3jiCKy8/HAAZ8e0aGOWw4/ktL1MRkbc\nwZvtK6BGW7cqSTTcscdEf/SRHeIffOBm3TpPdGQL2Fe8jR0bYtKkIBMnBpk+PUhJiQZ6j7MsXHv3\nhgPcND2aCtz7/K12D5YOpuGccwnIaIKjRhMYNZrgqFE9MsxMqd4u7cI9FIJ//cvN2rXuaKu8osId\n7WIBKCkJcckljUycGGTy5BCnnBLs6sVjKh6WhbtyJx5TERPkBs/GilZD6iyXi9DQE6j/wkQ7wGU0\nwbJRBEeJfUJPKQWkQbjv2+di3Tq7Nf7BBx4+/NBDdXVTkOfk2Jf5T5pkt8wnTQpSWqqt8oSqr8e9\nZzfu3btx79mFZ/cu+/muSti2meIN5bhrjjQ7xPJ4CJ44gsbPTSMgQrBMCMpoAieVdfkybaXSiaPC\nPXLS0+5e8bBunYetW5sPVznppBAXXRQIt8qDjBkT6u7AgfTV0IB77x7ckbDeY3/Z4b0rHOi7cFdV\ntf8amZmETiqjIRzgkS6V4Ikj0IsBlOq6lA/3jz9285e/wLvv5rY66dm3r8V55zUF+WmnBfXy/ngE\nArj9e5tCe/cu3HuaAtwTboG79+3r8GVCffoSKikhcPKphEpKCJUMIlRSQrBkEKGBgwgNHEjxhDFU\nVdUdp4oplT5SPtzvvjuLt98Gj8fNuHEhJk60u1YmTw4yYoSlI8o64DpYhbd8A54N6/GWb8C74RPc\n27fh9u/tcNhgKL/ADu3RYwkNLLG/ouE9iGB4XVzdJ3pll1I9IuX/shYsOEp1dT5DhhzRrtj21Nfb\n4743fGKHePl6POUb8OyqbLab5fEQGjqMximfs8N64KAWwW2HtpVfkKSKKKXiFVe4i8gjwBTsmZPm\nGGPej9l2GXAnUA+8YIxZKCL5wO+wL9DLAu42xixPdOEBBg2yOOUU8LceGZd+QiHcn22PCXC7Re7Z\n9K9WU6AGB5XScN75BMaMIzB2HIEx4+wpR7WfWylH6DTcReQcoMwYM1VExgBPAVPD29zAQmAisB94\nTUReBr4EGGPMT0SkFHgTGN1DdUhLrqoDdnCXr8e7oak13nLUSSi/gMDEyXaIjxlLcOw4AqPHYBX1\nS1LJlVLHQzwt9xnAywDGmHIRKRKRQmPMYaA/cNAY4wcQkVXA+cA+4JTw8UXhZdUV9fV4Pv4n3vL1\neDesb+pS2b2r2W6W10twZBkNY8YSHDMu2iIPDRmql7IrlYbiCfcSYG3Msj+87nD4eYGIlAFbgXOB\nt4wxvxSR60VkE3a4z+zsTYqKcvF6u36zR5/PQf3AwSAsXw5LlsCyZfQLBJpvHzIELroITjkFxo+H\n8eNxieDNykrJkyiO+tm1oHVLXalev65kQbQZaIyxROQ67K6aQ8AWwCUiXwe2G2MuFJFTgSeByR29\naFVVbReKYkv1e6hGuCt3kv3cs2T//nd4du6wV556KnUTJjfvUunbxnjOww1Ait0YAuf87NqidUtd\nqVS/9j6E4gn3SuyWekQpEO0TMMasBqYBiMg87Bb8OcDy8PaPRKRURDzGmO7f2NBpgkEy33yD7N89\nTeYby3GFQoTyC6i77iaOXnMdRTOmcSRFfsmUUr1HPOG+ArgbWCIiE4FKY0w0bUTkNeA6oAaYBcwH\nBgNnAn8WkROAIxrszbkrd5L9+9+R/dyz0VZ642kTOXrNDRz90hXovfeUUt3RabgbY9aIyFoRWYN9\nu/lbReR64JAx5iXgN9gfABYwzxizT0SWAE+JyOrwe9zSYzVIJcEgmatWkP3sM61b6ddeT2D8qcku\noVLKIVxWAm5gkAh+f3WXC9Lb+8fcO3c0tdLD8403Tpxkt9Ivu7zDVnpvr1t3Obl+WrfUlUr18/kK\n2hwOl4qDK1JDIEDmqjfIfvZpMleuaGqlX38TddfcQHD8KZ2/hlJKdZGGe4K120q/9ka7lZ6Xl+QS\nKqXSgYZ7IgQCZK5cYbfSV71ht9ILCqm74RvUff16baUrpY47DfducO/4rKmVHp6Eq3HSZOquvZH6\nS7+srXSlVNJouB+rSCv9d0/ZrXTLslvpN37TbqWfPD7ZJVRKKQ33Y+E6fIii86bh2b4VgMZJp1N3\n7Q3aSldK9Toa7sfAu24tnu1baZh+Hkd+fq+20pVSvZa7811UhNeUA3D06ms12JVSvZqG+zHwVNjh\nHhg9NsklUUqpjmm4HwNvRTlWRgbBEScluyhKKdUhDfd4WRYeU0FwZBlkZCS7NEop1SEN9zi5d+7A\nfaSagOjdApVSvZ+Ge5y8FRsACMqYJJdEKaU6p+EeJ09FBaAnU5VSqUHDPU7Rlvto7ZZRSvV+Gu5x\n8pgKrKwsgsNHJLsoSinVKQ33eIRCeDdWECgT8HiSXRqllOqUhnsc3Nu24qqrIzhaT6YqpVKDhnsc\nvNErUzXclVKpQcM9DpE5ZbTlrpRKFRrucfCER8oEdIy7UipFaLjHwVtRgZWbR2josGQXRSml4hLX\nfO4i8ggwBbCAOcaY92O2XQbcCdQDLxhjFobXXw38GAgAPzfGvJrgsh8fgQCeTRsJjDsZ3PpZqJRK\nDZ2mlYicA5QZY6YCNwGPx2xzAwuBi4GzgVkiMkREioG7gM8DlwCX9UDZjwvPls24Ghr0ylSlVEqJ\np+U+A3gZwBhTLiJFIlJojDkM9AcOGmP8ACKyCjgfqANWGmOqgWrg5h4p/XEQmcNd55RRSqWSeMK9\nBFgbs+wPrzscfl4gImXAVuBc4K3wfrki8gpQBMw1xqxKUJmPq8i0AwGddkAplUK6cg9VV+SJMcYS\nkeuAp4BDwJaY7cXAl4ETgL+KyAnGGKu9Fy0qysXr7frVnz5fQZeP7dDWTQD0PesM6Kn36ESP1a2X\ncHL9tG6pK9XrF0+4V2K31CNKgV2RBWPMamAagIjMw27B5wBrjDEB4FMRqQZ8wN723qSqqvZYyx7l\n8xXg91d3+fiOFH30T9wFhezPLIQeeo+O9GTdegMn10/rlrpSqX7tfQjFM/xjBTAbQEQmApXhvnTC\n614TkQEikgfMAlaGjzlPRNzhk6v5wL7uVSEJ6uvxbP7UvnjJ5ep8f6WU6iU6bbkbY9aIyFoRWQOE\ngFtF5HrgkDHmJeA32GFuAfOMMfsARORPwN/DL/NtY0yoJyrQkzyfbsIVCOi0A0qplBNXn7sx5vYW\nqz6K2fYi8GIbxywBlnSrdEmm0w4opVKVXpXTAZ12QCmVqjTcO+DVW+sppVKUhnsHPBUbCPXrh+Xz\nJbsoSil1TDTc21NXh2frFrvVriNllFIpRsO9Hd5/GVyWRVD0ylSlVOrRcG+HJ3r3Je1vV0qlHg33\ndkRurafDIJVSqUjDvR2e8Bj3gHbLKKVSkIZ7O7ymguCAgVj9ipNdFKWUOmYa7m05cgTP9m06h7tS\nKmVpuLfBuzF88dIYDXelVGrScG+DV+++pJRKcRrubWgaBqnhrpRKTRrubYjcWk8vYFJKpSoN9zZ4\nTAXBwUOwCvskuyhKKdUlGu4tuA4dxLOrUlvtSqmUpuHegken+VVKOYCGewuR/nY9maqUSmUa7i14\n9NZ6SikH0HBvITLGPVAmSS6JUkp1nYZ7C96KcoInDIe8vGQXRSmlukzDPYZr3z7c+/za366USnka\n7jG8RqcdUEo5gzeenUTkEWAKYAFzjDHvx2y7DLgTqAdeMMYsjNmWA3wC3GuMeSaB5e4RHh0po5Ry\niE5b7iJyDlBmjJkK3AQ8HrPNDSwELgbOBmaJyJCYw+8EDiS0xD3IGxnjri13pVSKi6dbZgbwMoAx\nphwoEpHC8Lb+wEFjjN8YEwJWAecDiMhoYCzwasJL3UM8phzL7SZYNirZRVFKqW6Jp1umBFgbs+wP\nrzscfl4gImXAVuBc4K3wfvOB24Dr4ilIUVEuXq8nrkK3xecr6PKxAFgWmHIYORLfUF/3XivBul23\nXs7J9dO6pa5Ur19cfe4tuCJPjDGWiFwHPAUcArYALhG5FvibMWaLSHzjxauqartQFJvPV4DfX93l\n4wHce3ZTXFVF/eemcbibr5VIiahbb+bk+mndUlcq1a+9D6F4wr0Su6UeUQrsiiwYY1YD0wBEZB52\nC/7LwAgRuQQYAtSLyA5jzMquFP548JSHT6bqhGFKKQeIJ9xXAHcDS0RkIlBpjIl+pInIa9hdLzXA\nLGC+MeaFmO1zga29OdghZhjkGJ0wTCmV+joNd2PMGhFZKyJrgBBwq4hcDxwyxrwE/Ab7A8AC5hlj\n9vVkgXtK9O5LOlJGKeUAcfW5G2Nub7Hqo5htLwIvdnDs3C6V7DjzVpRjZWQQHHFSsouilFLdpleo\nAliWffelk0ZCZmayS6OUUt2m4Q64d+7AfaRar0xVSjmGhjuxN8TWcFdKOYOGOzG31tNwV0o5hIY7\nMS33MRruSiln0HAHPKYCKyuL4PARyS6KUkolhIZ7KIR3Y4V9Wz1P1+e2UUqp3iTtw929bSuuujqC\nOu2AUspB0j7cozfE1mkHlFIOouGut9ZTSjlQ2od7dE4ZvYBJKeUgaR/u3opyrNxcQkOHJbsoSimV\nMOkd7oEAnk0b7Tnc3en9rVBKOUtaJ5pny2ZcDQ3a366Ucpz0Dvdof7uOlFFKOUtah3tk2oHAaB3j\nrpRylrQOd4+xJwwLastdKeUwaR3u3ooNhAoKCQ0qTXZRlFIqodI33Ovr8Wz+1J52wOVKdmmUUiqh\n0jbcPZ9uwhUI6LQDSilHSttwb5p2QE+mKqWcJ23D3RMdKaMtd6WU86RtuHv11npKKQfzxrOTiDwC\nTAEsYI4x5v2YbZcBdwL1wAvGmIXh9Q8C08LvMc8Y82KCy94tnooNhPr1wxowINlFUUqphOu05S4i\n5wBlxpipwE3A4zHb3MBC4GLgbGCWiAwRkXOBk8PHXAg82hOF77K6Ojxbt9itdh0po5RyoHi6ZWYA\nLwMYY8qBIhEpDG/rDxw0xviNMSFgFXA+8DZwZXifg0CeiPSae9h5N23EZVkEdZpfpZRDxdMtUwKs\njVn2h9cdDj8vEJEyYCtwLvCWMSYI1IT3vwlYFl7XrqKiXLzerue/z1cQ/847twCQM/k0co7luCQ5\nprqlICfXT+uWulK9fnH1ubcQ7ccwxlgich3wFHAI2BK7PdwffxPwxc5etKqqtgtFsfl8Bfj91XHv\nn/f+h+QCB4eMoPEYjkuGY61bqnFy/bRuqSuV6tfeh1A84V6J3VKPKAV2RRaMMauxT5wiIvOwW/CI\nyAXAHcCFxphDXSl0T4kOg9Qx7koph4qnz30FMBtARCYClcaY6EeaiLwmIgNEJA+YBawUkT7AfwGX\nGGMO9EC5u8VrKggOGIjVrzjZRVFKqR7RacvdGLNGRNaKyBogBNwqItcDh4wxLwG/wf4AsLCHPO4T\nkZuxT7b+UUQiL3WtMWZ7T1TimBw5gmf7NhqmTU92SZRSqsfE1edujLm9xaqPYra9CLzYYv+lwNJu\nl64HeDeGL17SOdyVUg6WdleoesN3X9I53JVSTpZ24R69tZ5OO6CUcrC0C/fIrfWC2i2jlHKwtAt3\nj6kgWDoYq7BPsouilFI9Jq3C3XXoIJ5dlTrtgFLK8dIq3D06za9SKk2kVbhH+tv11npKKadLq3D3\n6K31lFJpIq3CPTLGPTBKw10p5WxpF+7BYcMhLy/ZRVFKqR6VNuHu2rcP9z4/gTF6MlUp5XxpE+7e\naH+7hrtSyvnSJtybph3Q/nallPOlTbhHT6bqhGFKqTSQNuHuMeVYbjfBslHJLopSSvW49Ah3y8Jb\nsYHgiSMgOzvZpVFKqR6XFuHu3rsH98GDejJVKZU20iLcPeXhaQd0wjClVJpIi3CPDoPUcFdKpYm0\nCHePjpRRSqWZtAh3b0U5ltdLcMRJyS6KUkodF84Pd8uy7740sgwyM5NdGqWUOi688ewkIo8AUwAL\nmGOMeT9m22XAnUA98IIxZmFnxxxP7p07cB+ppkFHyiil0kinLXcROQcoM8ZMBW4CHo/Z5gYWAhcD\nZwOzRGRIR8ccb003xNZwV0qlj3i6ZWYALwMYY8qBIhEpDG/rDxw0xviNMSFgFXB+J8ccV3prPaVU\nOoon3EsAf8yyP7wu8rxARMpEJAM4FxjYyTHHVbTlrlP9KqXSSFx97i24Ik+MMZaIXAc8BRwCtsRu\nb+uY9hQV5eL1erpQHJvPV9D2hk83QlYW/U4/FTxdf/1karduDuHk+mndUleq1y+ecK+keau7FNgV\nWTDGrAamAYjIPGArkN3RMW2pqqqNq8Bt8fkK8PurW28Ihei/YQPBkaOoOtD110+mduvmEE6un9Yt\ndaVS/dr7EIqnW2YFMBtARCYClcaYaK1F5DURGSAiecAsYGVnxxwv7m1bcdXV6bQDSqm002nL3Riz\nRkTWisgaIATcKiLXA4eMMS8Bv8EOcwuYZ4zZB+xreUyP1aADXhM+marhrpRKM3H1uRtjbm+x6qOY\nbS8CL8ZxzHHXNAxSpx1QSqUXR1+hqrfWU0qlK0eHu7eiHCs3l9CwE5JdFKWUOq6cG+6BAJ5NGwmM\nEnA7t5pKKdUWx6aeZ8tmXA0N2t+ulEpLzg33aH+7jpRRSqUfx4Z7ZKRMQKcdUEqlIceGuyc8xl1v\niq2USkeODXdvxQZCBYWESgcnuyhKKXXcOTPc6+vxbP6UoIwGV6dzlimllOM4Mtw9n27CFQjotANK\nqbTlyHD3GnukjN59SSmVrhwZ7h6jwyCVUunNkeHuLQ+Hu17ApJRKU44Md48pJ1RUhDVgQLKLopRS\nSeG8cK+rw7Nls91q15EySqk05bhw927aiMuy7GGQSimVphwX7p7y8LQD2t+ulEpjjgv3yK31dBik\nUiqdOS7cPZEJw3QYpFIqjTku3L2mgpBvAFZxcbKLopRSSeOscD9yBM/2bdrfrlSSvPXWqrj3feyx\n+VRW7mx3++23fz8RRUpbjgp370a7vz0wWkfKKHW87dpVycqVy+Pef86cH1Dawaytv/jFw4koVtry\nJrsAieStiMwpoy13ld7y5t5J1l9eTuhr1s/6EjVz72t3+8MP/5Ly8vU8/fRvCIVCVFbuZNeuSh59\ndBHz5t2D37+Xuro6brzxZs46axq33XYz3//+j/nrX1dRU3OE7du3sXPnDr7znR8wdepZzJw5g1df\nXcVtt93M6aefybp1H3Dw4EF++ctH6N+/P/fc8zN2797F+PGn8OabK3nppWXNyvP88/+Pt95aRSgU\nYurUs7jxxpuprq7mnnvupKamhvz8fObOfYBgMNhq3YIFz5CRkcMVV1zF5s2bePjhB1m4cClf/eqX\nGTVqNGeccSYDBw7iiSd+TUZGBgUFBdxzzy/IyMjg0UcfYsOGT/B4PPzoRz/hmWee5NJLv8zkyWfQ\n0NDA179+Jc8992e83p6N37gF0IwyAAAPt0lEQVReXUQeAaYAFjDHGPN+zLZbga8DQeADY8x3RaQU\neArIAjzA94wxaxNd+Jb01npKJc/XvnYNL774R2644Zs8+eQSAoFGFi16gqqqA5xxxhQuuugSdu7c\nwc9+djtnnTWt2bF79+7hoYce5+9/X8P//M+fmTr1rGbb8/LyeOyxxSxevIC3336T0tIhNDTUs3Tp\nM7z33jv88Y/Pt1mmRYuewO1285WvXMZVV/0bzz//LGecMZUrr/wqf/jD7/ngg39QUbGh1br2VFbu\n5IEHHmLEiJN4882V3HXXfZSWDubee3/O//7v38jKymLv3j0sXfoM//d/61i16g0uuOBiVq16g8mT\nz2Dt2n8wZcrnejzYIY5wF5FzgDJjzFQRGYMd2lPD2wqBHwEjjTEBEVkhIlOA2cBLxpglIvI54H7g\nwh6rRVjk1npBkZ5+K6V6tZq593XYyj4exowZB0BBQSHl5et55ZUXcbncHD58qNW+p5wyAYABAwZw\n5MiRVttPPfW06PZDhw6xbdsWxo8/FYCpU8/C4/G0OiY7O5vbbrsZj8fDwYMHOXz4MBs3VvCNb3wL\ngKuuuhqAV155sdW6F154ps06ZWfnMGLESQD07duXX/7yPoLBIJWVO5k06XSqqg5EyzVhwkQmTJhI\nIBBg8eLHCQQCvPPOai6+eFYc373ui6fPfQbwMoAxphwoCoc6QEP4K19EvEAucADYB0SGqxSFl3uc\nx1QQLB2M1afv8Xg7pVQHMjIyAHjjjdc5fPgwv/rVEzzwwENt7hsbzpZldbrdsixcLju+XC4XrhZT\njezevYs//OH3zJ+/gIULl1JSUgKA2+3BskLN9m1rXezrBQKBmDo1tYfnzbuX733vxyxcuJTPf/7s\ndl/L6/Vy+ulT+OCDf7Bly2ZOPvmUNr8HiRbP/wYlQGyXij+87rAx5qiI3A1sBuqAF4wxG8PdOP8Q\nkWuBQuDznb1JUVEuXm/rT994+TKCsKsSLrgAn6+gy6/TGzmtPi05uX7pVLd+/fLxeFz4fAXk5WWR\nn5+Nz1dAIFDHyJEnMnBgH95663WCwQA+XwGZmV6KivKa7VtVlUdmphefrwCXy9VsP5+vgPz8bBob\nsygrK2P58uX4fAW88847BIPBZuXZs2cbPl9/TjhhIOvXr2fPnt0UFGQyadIEKir+ydlnT+GFF14g\nKyurzXX5+flUVVXh8xXw+usVrcoEUFdXw7hxIwkEAvzznx8yYcJ4xo6dzNKlS/H5bmXDhg3893//\nN3fddRdf/eps5s6dy1lnnXXcfie60vET/UgLt+B/CowCDgNvisipwCzgj8aY+0XkEuAh4PKOXrSq\nqrYLRbH5fAVUvfs+RUDtiFHU+Ku7/Fq9jc9XgN9B9WnJyfVLt7r16TOQjz/+hJ/9bC55eflkZBzF\n769m8uSzuP327/P++2uZOfNS+vf38eCDD9PQEKCqqoaamvrovlVVNTQ0BPD7q7EsC7+/Orqf31/N\nkSNHqamp5+STJ/P8839g9uyvcNppkygs7NOsPP37DyEjI4vZs69k/PgJXHrp5dxxx8+5//4Hue++\nn7Nq1V/Jzc1j7tz7mDDBarXO6w1y003f4IMP1jFhwsRWZQL40pdmc+WVVzF06DCuuurrLF78axYv\nfopBg4Zy5ZVXAfCDH9yO31/NwIEncOBAFWeddV7Cfyfa+7BwtfUvUCwRmQvsMsYsCS9vBk41xlSL\nyJnAncaYWeFt84B/AVeG168VkSzgX8aYYR29j99f3XFBOuDzFVA9/3EKfjiHw48tov5rX+/qS/U6\nTg4IcHb9tG495/DhQ6xb9wHTp8/A79/LnDnf4rnn/pyw1090/bZv38b8+b/ksccWJew1I3y+gjan\nv42n5b4CuBtYIiITgUpjTKTWW4ExIpJjjKkDJgPLgE3AmdjdOadjB36Pikw7oHPKKOV8ubl5vPnm\nSp577lksK8S3v917L3h6+eU/8corL3HHHXcf1/fttOUOICK/AM4GQsCtwGnAIWPMSyLy78ANQABY\nY4z5sYgMAp7EPsEK8B1jzD87eo/uttwbzp5O5jur8W+uhPz8rr5Ur5PsFlJPc3L9tG6pK5Xq117L\nPa5wPx66G+4h3wCs3DwOfNDhZ0jKSaVfsq5wcv20bqkrlerXXrg7Y/oBvx/3Pr9OO6CUUmHOCPf1\n6wGddkAppSIcFe4BvbWeUkoBTgn3Tz4B9NZ6SqWK2bNnUVtby7PPPsMnnzQ/T1ZbW8vs2R1foh+Z\nWnjZsr+wevVfe6ycqcwZs0KuX4/ldhMsG5XskiiljsE111x/zMdEphaePn3GcZunJRWlfrhbFnzy\nCcHhJ0J2drJLo1SvMHduFn/5S2L/vGfNCjB3bn2722+88WoeeGA+JSUl7N69i5/+9EcsWPBr7r77\nTurq6jh69Cjf+96PGDv25Ogx998/l+nTZzBhwmncccePaWhoiE4iBrBixWv86U9/wONxM3z4Sfzn\nf97Ramrhvn37csUVV7Fo0WN8/PFHBAJBrrjiK1x44cw2pwuOzDMD9myU9977c8CeQ+bOO+9m8OAh\nvPzyyzz99G9xuVx89atXM2PGF3n99Vf505/+0GxdZFpigDvv/DGXX/4VPvxwbafTHW/cWMH8+b/E\n7XZx8smnMnPmpTz44P0sWvQEAL/97ZPk5uZx5ZVf7fLPK+W7Zdx790BVlZ5MVSrJzj77XN57720A\n3nlnNdOnn8f+/fu55JIvsWDBEm655TZ+//vftnns8uWvMWLESSxa9ARlMf+B19XVMX/+AhYvfort\n27fy6aeb+NrXrmHChInccMM3o/v93/+tY/PmT1m8+Ckef/zXPPXUUmpra4Cm6YKnTPkcb7/9ZrP3\n3b9/Hzfc8E0WLFjCzJmX8uKL/01tbQ2LFi3iV79aysMPL+SNN16ntraGZ555otm6jkSmO66pOcIZ\nZ0xh4cKl3HPPPJ58cgkAjz76ED/60U9ZvPgpDhzYT3Z2No2NDezduweANWveZcaMLxzjT6C5lG+5\ne8rDN8TWYZBKRc2dW99hK7snnH32uSxc+ChXXPEV3n13NT/4we3061fMb3/7BM8//yyNjY1kt/Pf\n9datm5kwYRIAp502Kbq+sLCQn/zkBwBs27aFQ4cOtnl8RcUGJkyYCEBOTg7Dh4/gs88+A1pPFxyr\nX79iHn30IZ58cgnV1YcRGcPWrVsYMWIEWVnZZGVl84tfPMyGDZ8wbNjwZus60tl0x9u3b2PkyDIA\nfvazewD44hcv5s033+D88y8gLy+ffv26dx/olG+5e43efUmp3mDEiJPYv9/Pnj27qa6uZtiwE/jj\nH5+jf/8BLF78JD/84e3tHmtZ4Hbb1+KEQvb1jI2NjTz88IPcffcDLFy4tFl3Tksul4vY6zEDgcbo\n63U0nfCTTy7hzDOn8Ktf/Sb6n4Db7SEU6nxa4JaaTw3c8XTHbnfr6D3//At4553VvPvu23zhCxd0\n+F7xSPlw17svKdV7TJ36eZYuXcS0aecAcOjQQQYPHgLA6tV/bRaAsYYNO4GK8N/yunUfAFBbW4PH\n46G4uD979uymoqKcQCCA2+0mGAw2O3706HF8+OHa8HG17Ny5gyFDOpyrEICDB+3yWZbFu++uprGx\nkRNOGM6WLVuora2lvr6e7373PzjhhOFs376t2Tp7TnkXR48e5ejRo2zcaNp8/UGDSnG73axe/SaN\njY0ADB9+IuvX26P85s27h61bt1BUVERhYSHLly/jnHPO7bTsnUn5bhnvvzaC10vwpJHJLopSae+c\nc87llltu5Jln7NveXXjhTO677y7++teVXHHFV1i5cgWvvvpKq+MuvHAmP/3pD5kz51uccsoEXC4X\nffr05fTTz+Qb37iWkSPL+Ld/u4bHH3+YBQuWYEwFjz8+n7w8ex6pU0+dgMhobr31mwQCAW655TZy\ncnI6Le9ll13OI4/8FyUlpcyefRUPPng/H3/8Ed/5znf47nf/A4Crrvo3cnJyuOmmW5qtc7lcfOlL\ns7n55usYPnwE0kYDc/r087j99u+zYcMnzJx5KQMGDODpp3/DnDk/5KGH5gEwbtx4hg8/Mbz/DN57\n7x1yc/O68N1vLuXnlsn59ULyG+vwf/tHiS5Sr5BKc1x0hZPrp3VLXcmq33333cXFF89i4sTJcR/T\nnSl/e7W6W24j31cADv5FU0o5W319Pd/+9r8zZszYYwr2jqR8uCulVKrLyspi6dJnEvqaKX9CVSml\nVGsa7kop5UAa7kop5UAa7kop5UAa7kop5UAa7kop5UAa7kop5UC95gpVpZRSiaMtd6WUciANd6WU\nciANd6WUciANd6WUciANd6WUciANd6WUciANd6WUcqCUn89dRB4BpgAWMMcY836Si5QwIvIgMA37\n5zTPGPNikouUUCKSA3wC3GuMeSbJxUkoEbka+DEQAH5ujHk1yUVKCBHJB34HFAFZwN3GmOXJLVX3\nicjJwP8AjxhjForIUOBZwAPsAq4xxtQns4zHKqVb7iJyDlBmjJkK3AQ8nuQiJYyInAucHK7bhcCj\nSS5ST7gTOJDsQiSaiBQDdwGfBy4BLktuiRLqesAYY84FZgOPJbc43SciecACYFXM6nuAXxljpgGb\ngBuTUbbuSOlwB2YALwMYY8qBIhEpTG6REuZt4Mrw84NAnoh4kliehBKR0cBYwBEt2hbOB1YaY6qN\nMbuMMTcnu0AJtA8oDj8vCi+nunrgYqAyZt10IHIn779g/0xTSqqHewngj1n2h9elPGNM0BhTE168\nCVhmjAkms0wJNh/4frIL0UOGA7ki8oqIvCMiM5JdoEQxxrwADBORTdgNkB8muUjdZowJGGPqWqzO\ni+mG2QsMOs7F6rZUD/eW2rwLeCoTkcuww/22ZJclUUTkWuBvxpgtyS5LD3Fht24vx+7GeFpEHPG7\nKSJfB7YbY0YC5wELk1yk4yElf3apHu6VNG+pl2Kf/HAEEbkAuAO4yBhzKNnlSaCZwGUi8nfgG8DP\nRCTl/u3twB5gTbhF+ClQDfiSXKZEOQtYDmCM+QgodVJ3YYwj4RP+AINp3mWTElI93Fdgn9RBRCYC\nlcaY6uQWKTFEpA/wX8AlxhhHnXQ0xlxljDndGDMFeAJ7tMzKZJcrgVYA54mIO3xyNR9n9E2DfXLx\nTAAROQE44rDuwoiVwBXh51cAryexLF2S0kMhjTFrRGStiKwBQsCtyS5TAl0F9Af+KCKRddcaY7Yn\nr0gqHsaYnSLyJ+Dv4VXfNsaEklmmBFoCPCUiq7Hz45Ykl6fbRGQS9jmg4UCjiMwGrgaeEZF/B7YB\nv01eCbtG53NXSikHSvVuGaWUUm3QcFdKKQfScFdKKQfScFdKKQfScFdKKQfScFdKKQfScFdKKQf6\n/4AL1FbqXMkpAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "score_2 = model_2.evaluate(x_test, y_test, verbose=0) \n",
    "print('Test score:', score_2[0]) \n",
    "print('Test accuracy:', score_2[1])\n",
    "\n",
    "plt.plot(my_model_2.history['acc'], 'r')\n",
    "plt.plot(my_model_2.history['val_acc'], 'b')\n",
    "plt.legend({'training accuracy': 'r', 'validation accuracy':'b'})\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 265
    },
    "colab_type": "code",
    "id": "FzJA3qOd8j41",
    "outputId": "ba481082-3d34-4dfb-9dbd-437f8e60f3c5"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD4CAYAAAAXUaZHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(my_model_2.history['loss'], 'r')\n",
    "plt.plot(my_model_2.history['val_loss'], 'b')\n",
    "plt.legend({'training loss': 'r', 'validation loss':'b'})\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "vWIBcdwZ9Adp"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "98FQG6SB9E9v"
   },
   "source": [
    "# MODEL-3 CNN with 7 Layers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 374
    },
    "colab_type": "code",
    "id": "rR28Yuyx9Huv",
    "outputId": "2ec9266b-21c1-4c75-ccd2-5d307d91b2c8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 60000 samples, validate on 10000 samples\n",
      "Epoch 1/10\n",
      "60000/60000 [==============================] - 19s 312us/step - loss: 0.4120 - acc: 0.8620 - val_loss: 0.0495 - val_acc: 0.9837\n",
      "Epoch 2/10\n",
      "60000/60000 [==============================] - 17s 286us/step - loss: 0.0929 - acc: 0.9726 - val_loss: 0.0439 - val_acc: 0.9855\n",
      "Epoch 3/10\n",
      "60000/60000 [==============================] - 17s 285us/step - loss: 0.0666 - acc: 0.9801 - val_loss: 0.0303 - val_acc: 0.9910\n",
      "Epoch 4/10\n",
      "60000/60000 [==============================] - 17s 283us/step - loss: 0.0561 - acc: 0.9836 - val_loss: 0.0287 - val_acc: 0.9918\n",
      "Epoch 5/10\n",
      "60000/60000 [==============================] - 17s 283us/step - loss: 0.0502 - acc: 0.9855 - val_loss: 0.0212 - val_acc: 0.9930\n",
      "Epoch 6/10\n",
      "60000/60000 [==============================] - 17s 284us/step - loss: 0.0432 - acc: 0.9871 - val_loss: 0.0237 - val_acc: 0.9918\n",
      "Epoch 7/10\n",
      "60000/60000 [==============================] - 17s 284us/step - loss: 0.0382 - acc: 0.9886 - val_loss: 0.0252 - val_acc: 0.9926\n",
      "Epoch 8/10\n",
      "60000/60000 [==============================] - 17s 284us/step - loss: 0.0351 - acc: 0.9893 - val_loss: 0.0234 - val_acc: 0.9923\n",
      "Epoch 9/10\n",
      "60000/60000 [==============================] - 17s 283us/step - loss: 0.0332 - acc: 0.9903 - val_loss: 0.0282 - val_acc: 0.9918\n",
      "Epoch 10/10\n",
      "60000/60000 [==============================] - 17s 283us/step - loss: 0.0305 - acc: 0.9907 - val_loss: 0.0205 - val_acc: 0.9943\n"
     ]
    }
   ],
   "source": [
    "model_3 = Sequential()\n",
    "\n",
    "\n",
    "model_3.add(Conv2D(32, kernel_size=(2, 2),padding='same',activation='relu',input_shape=(28,28,1)))\n",
    "\n",
    "\n",
    "model_3.add(Conv2D(32, (2, 2), activation='relu'))\n",
    "\n",
    "\n",
    "model_3.add(MaxPooling2D(pool_size=(3, 3), strides=(1,1)))\n",
    "\n",
    "\n",
    "model_3.add(Dropout(0.3))\n",
    "\n",
    "\n",
    "model_3.add(Conv2D(64, (2, 2), activation='relu'))\n",
    "\n",
    "\n",
    "model_3.add(MaxPooling2D(pool_size=(2, 2),padding='same'))\n",
    "\n",
    "\n",
    "model_3.add(Conv2D(64, (2, 2),padding='same',activation='relu'))\n",
    "\n",
    "\n",
    "model_3.add(Conv2D(128, (2, 2), activation='relu'))\n",
    "\n",
    "\n",
    "model_3.add(MaxPooling2D(pool_size=(3, 3),padding='same'))\n",
    "\n",
    "\n",
    "model_3.add(Dropout(0.25))\n",
    "\n",
    "\n",
    "model_3.add(Conv2D(128, (2, 2),padding='same',activation='relu'))\n",
    "\n",
    "\n",
    "model_3.add(Conv2D(256, (2, 2), activation='relu'))\n",
    "\n",
    "\n",
    "model_3.add(MaxPooling2D(pool_size=(2, 2), strides=(1,1)))\n",
    "\n",
    "\n",
    "model_3.add(Dropout(0.25))\n",
    "\n",
    "\n",
    "model_3.add(Flatten())\n",
    "\n",
    "\n",
    "model_3.add(Dense(256, activation='relu',kernel_initializer=he_normal(seed=None)))\n",
    "\n",
    "\n",
    "model_3.add(BatchNormalization())\n",
    "\n",
    "\n",
    "model_3.add(Dropout(0.5))\n",
    "\n",
    "\n",
    "model_3.add(Dense(128, activation='relu',kernel_initializer=he_normal(seed=None)))\n",
    "\n",
    "\n",
    "model_3.add(Dropout(0.25))\n",
    "\n",
    "\n",
    "model_3.add(Dense(10, activation='softmax'))\n",
    "\n",
    "\n",
    "model_3.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "\n",
    "my_model_3 = model_3.fit(x_train, y_train,batch_size=128,epochs=10,verbose=1,validation_data=(x_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 302
    },
    "colab_type": "code",
    "id": "Z3DdzNDf959y",
    "outputId": "09dea816-a7a5-4083-d99c-77931414d808"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test score: 0.020454944854525092\n",
      "Test accuracy: 0.9943\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD7CAYAAACRxdTpAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzt3Xt8VPWd//HXmQsJuUkggxhYQSR+\nhCIg3kitXIrWG4hV1G5bq61df7W6i9tuu9bS1jvVilrK6mrV0u222q5VqlWrBZVqo1UhiaLwRbmI\nEi6JJCTkPjPn98eZSWZyIUNukznzeT4e85iZc5n5zBd4z5fvnPM9lm3bKKWUchdPsgtQSinV/zTc\nlVLKhTTclVLKhTTclVLKhTTclVLKhTTclVLKhXyJbCQiU4E/AfcaY1Z2WHcmcAcQAp4zxtwaWX4v\nMAuwgSXGmLf6s3CllFLd6zHcRSQb+AWwtptNVgBnA7uAdSLyRyAAFBljikVkMvAoUNw/JSullOpJ\nIj33ZuA84D87rhCRicB+Y8zHkefPAfNxwn01gDFmk4jki0ieMaa2uzeprKzr9dlU+flZVFc39HZ3\n19H2aKdtEU/bI54b2iMQyLW6Wt7jmLsxJmiMaexm9RigMub5PuCoLpZXRpYNCJ/PO1AvnZK0Pdpp\nW8TT9ojn5vZIaMz9MHT5DXKI5W3y87P61NCBQG6v93UjbY922hbxtD3iubU9+hruFcT3yMdGlrV0\nWF4I7D7UC/Xlv0aBQC6VlXW93t9ttD3aaVvE0/aI54b26O7LqU+HQhpjdgB5IjJBRHzAAuDFyG0x\ngIjMBCqMMandgkoplUISOVrmJGA5MAFoFZHFwNPAdmPMU8A1wGORzX9vjNkCbBGR9SJSAoSBawei\neKWUUl2zhsqUv305WsYN/7XqT9oe7bQt4ml7xHNDe/T6aBmllFKpR8NdKaVcqL8PhVRKKdWFcBgq\nKy0qKiwqKjzs3m2xa5eHUAh+8INmhg/v3/fTcFdKqT4KhZzg3rWrPbgrKjyRILfYvdtZFgx2Hh73\n+20uv7yVoqJwv9ak4a5SRjAIBw9CXZ3FwYMWdXVw8KAVuTnLY9fV11vYNvj9meTk2OTm2uTkEPfY\nuY9fl50NVo+n3aWOUAgaG6GpySIchupqC5/Pxusl7ubxuOtz95dgEPbta+9xd+x5795tsWePRSjU\ndeN5PDZjxtjMmBGmsDDMUUfZFBaGGTvW5qijwkyYYBMI9P+BLRruakCFQlBf3zmQ6+qstuXOuvZt\nOoe1s09jY2+Tx39YW1tWx+CH7OzOXwhdfTnErsvJscnK6hyYra3Q1AQNDRZNTU7oNjVBY2P0eVfL\nun4eXRYN7+i62OetrR3bLafbz+712vh87WHvPO78ReD10vYF0b5d9GZ32M5ZFrudzwfDhtn4/bTd\nhg2zI8ud3mz7cue9nOXt65zl8a8T/xrx23q7OAG+tRU++aS9d71rl3MfG+R791qEw13/3fN6bY46\nymbmzDBjx7YHd2Fh+30g4NQ02DTc01A0XLoPjs4B0X3YdL1NY6MTyA0Nve8KZmU5oZmXB2PHhsnN\ntSMh27n3HV0eG8g5OTaFhTl89NHBti+U6BdF5y+Urr+APv3UYvt2q4uATIzH49Ti99ttbdddD6+v\nsrJsMjNtMjMhNxcCgTDDh9O2LDPTZvhwPw0NrYTDEAxahELE3YJBp3fvPCbmcfu2LS3R7a3ILXb/\ngft8fWVZsV8Qzp9NdTWEw11/2fn9TnCfckqIwkLncWyAjx1rU1Bgd/mlMRRouLtAMAjvvefh9de9\nvP22l4MHobY2q9te4ED847MsOy5IcnJgzJhwF0Mh8eGcne0Ecseeb3Z21z2tw1VQAM65HH37b29z\nMz0MBXX3ReE8b22FMWPa26hz6MY/z8yE4cOjz3veJyMjsSGVQMBPZWVTn9oiEc6XB91+ebS2Rm8W\nLS3Oti0tVsxy57mz3NmufbnzOs7y9nWxr5PIfsEgTJkCgUBrTG/bbhs6CQSc/230yLadD9faihUK\nRooKtj8OBrGCwch95I1j1tt5eQRnzOz3MTEN9xTU0gJlZR7eeMNHSYmXN9/0cvBg/F8Mr9fTFhDD\nh0N+vtPzSCQoBjJcUlVGBmRk2IwaBX39okgHHo8zFNJZx7brQ1vaNjQ3YzXUYzU0RG71bfdEl9Uf\nxKpv6LxdYyMZVpiW6kaoDMH61kgIt0Iw1BbEVmtre3h3tT4Y7P1niNj/yuuEpnymz68TS8M9BTQ0\nwIYNXkpKvLzxhpf1671x48+TJoUoLnZup50W4oQTcqipOZjEilXas20n+JqboKkZq7kp/nFjoxO6\nDQ2REK7vOoDr6+OW0XF9KNTnUmO/g2yvF/x+bK8P/D7w+bB9fmdwPzOTsN8fs8wLPj+2zweR57bf\nD16fc3+I9fijy32ECwKEjpM+f46ONNyHoLo6ePNNL6+/7uX1132UlXnaxnwty2bKlHBcmI8eHd/7\n8R/e74fK7UIhrAM1WDU1UGHh2/0pVnMzNDVhNTdHQjf2cTNWS3PMNk1OD7m5CaupGZoj2zZ1WN4S\nDfHIfuH+O7TP9nqxs7Kxs7KcW0Gg/XFWNnZ2dvz6rGzsbOee2GVZWdjZOc798CwKCkdSWdPkhLfP\n56r/fmq4DwGffmrxj384Yf7GG17efdfT9uu812szfXqYWbNCFBcHOfXUEPn5SS5YDb7WVqwDB/Ac\nqMaqqcE6UIOnxglsTyS4rQM1eA4caAvy6HJPXfwF0Prrr4/t92NnZEJmBnZGphOY+SOxI88ZloGd\nmQEZmdgZGdiR8Tt7WEaHsHXuiV3WFtiRgB42bGCCNy8Xmt0T6LE03JNg714r0it3bps3t/9yOGyY\nzamnOr3yWbNCnHJKiJzuj1xTqaSlJSZ0q+NDOTaoIwHtiayzamrw1B/eMFs4Jxd7xAjCR48nOGIE\n9hEjCI8YwfCCfBrCnviwzciMe9wW1hkZkJnpPB42LPI4sk1GRv/84q0GjIb7ALNt+Pjj2DD3sX17\n+0/wWVk2s2cHKS4O8dnPhjjxxBCZmUksWPWsqQlPTTVWdbVzv39/+/Pq/U5wV1djVe937msi2zUc\n3gVpwnlHYI8YQeiYiXEBbR8xwgnuDvfO43zsI46guwOrhwdyqU/xWRBVYjTc+5ltw9atFiUlvrZh\nll272sM8L8/mrLOCFBc7gT5tWljHyJOludkJ455CuqYaz/79vQrpcG4edv5IgpOOaw/lRII67wjt\nGas+0XA/TLYNtbXOOPmnn1rs3+/cV1V5KC93jjWvqmoP84KCMAsWtLYNs0yZEtZ/swOhtRXP/k+x\nKivxVO7DU1WJp6oKGmvJ2bWn/0K6SJwQHjkSe0Q+4fz8yP1I7Px8wiNGYo8cSXhEpAet39wqSdI+\n3JuaaAvo2LCOfbx/v3OrqrKoru568p+oo44Kc9FFrW1HsxQVhd30A/zgsW2sg3WRsK6MhHXkVrkP\nq6oq7rmnurrbl4qdbM8Z6sjvOqTzI6GsIa1cwFXhHp0UqWNAR4M5+jg2uOvrE0veESNsRo60mTAh\nzKhRYUaOtBk1yo67P+64MOPH2xrm3Yn2rvftiwnrqraettUW1s691dx8yJezLcsJ4NFHEpwylXAg\nQLgggF3g3IcDozli0tHstzI1pFXaSflwf+ABP48/Dvv2ZVNd3f0EP7EyM51AnjixPaRjgzr2efSW\njIl/Uko4jOeTj/F++AG+D7fg3b7N6XVXxQyTHKJ3HWVnZhIOjCY45TOEA6Njwrqg7Xn0Zo8a1e0P\nh20CuYT0B0SVhlI+sj75xENlJYwaZVNU1H1Yx/awu5qpTyXGOliH98MP4m6+Dz/Au+1D56SWDhLp\nXYcLCpxlgQB2do7+4SjVD1I+3G+/vZmHHhpGZeXhHWamDqFjL/zDD/Bu/RDvB1vw7tndaXM7K5vg\ncccTmjSJ0LFFhIqOIzTxWEJHHpVY71op1e8S+lcnIvcCs3Bm+VlijHkrZt0iYCnQDDxujFkpIjnA\n/+CcDJcB3GyMeaG/i1d9c7i98NC4f6JlzjyCRcc5IT7JCfLwUYXa21ZqiOkx3EVkDlBkjCkWkcnA\no0BxZJ0HWAnMBD4FnheR1cCFgDHG/EBECoGXgOMH6DOoQ+mHXnjw2CJCE4+F7OwkfAClVG8k0nOf\nD6wGMMZsEpF8EckzxtQCBUCNMaYSQETWAmcCVcC0yP75kedqgFm1Bxj28lrYuZXc8o3aC1cqjSUS\n7mOA9THPKyPLaiOPc0WkCNgBzANeMcbcKSJXisiHOOF+fr9WrdpY1fsZ9sLzZDyzmmGvvOTMPQ1k\nor1wpdJZb37pauvWGWNsEbkCZ6jmALAdsETkq8BOY8w5IjIdeAQ4+VAvmp+fhc/X+1M3A4HcXu+b\ncqqqYPVqeOIJWLvWubILwPTpcPHFUFwMxx+PNXYsfss6zCuIuk9a/d1IgLZHPLe2RyLhXoHTU48q\nBNoGa40x64AzAERkGU4Pfg7wQmR9uYgUiojXGNPtzPrV1b0/2iUQyKXS5ccyW/v2kfHcM2Q88yf8\nJa+2XaSgdfqJNC9cRMuCCwhNnATEtEeVXrAjHf5uHA5tj3huaI/uvpwSCfcXgZuBB0VkJlBhjGlr\nDRF5HrgCqAcWAsuBscBpwB9FZDxw8FDBrrrm2bObYc8+7QT663/Hsp2LcrSedDLNCy6kecEFhMdP\nSG6RSqkhqcdwN8aUiMh6ESkBwsC1InIlcMAY8xTwS5wvABtYZoypEpEHgUdFZF3kPb41YJ/AZTy7\nPiEjEui+N99oD/RTZ9G8cBHN519AeNw/JblKpdRQZ9n20LjYb2VlXa8LSfX/Wnl2fkTGn58m45nV\n+Nc7pxDYlkVr8enOkMt5C52jWBKU6u3Rn7Qt4ml7xHNDewQCuV0e3qanDiaJZ9tWJ9D/vBp/WSkA\ntsdDyxlznR76uQuwjzwyyVUqpVKVhvsg8n74gXPI4jN/wr/xHcC58G/L3M/TfMEXaT7nfOyCgiRX\nqZRyAw33AebdvImMZ1aT8ec/4dv0PuBcWLj5zC/QvPBCWs4+F3vkqCRXqZRyGw33/mbbeN9/rz3Q\ntxhncUYGzeecR/OCRU6gHzEiyYUqpdxMw72feLcYMv/vcYY9sxrftq2AMzd58/kXOD+KnnU2dm5e\nkqtUSqULDff+0NRE/hfmYjXUY2dl0bToIifQP38W5OQkuzqlVBrScO8Hvo3vYDXU03TRYuruWQlZ\nWckuSSmV5jzJLsANfOXOoYwt887UYFdKDQka7v0gepx68MSTklyJUko5NNz7ga9sA+HsHELHTkp2\nKUopBWi4993Bg3i3GILTpoO391MWK6VUf9Jw7yP/xnewbJvgjJnJLkUppdpouPeRr3QDAMEZJya5\nEqWUaqfh3ke+cifcW6druCulhg4N9z7ylZUSPmIE4WMmJrsUpZRqo+HeB9aBGnzbthKcfiJYXU6p\nrJRSSaHh3ge+8jJAx9uVUkOPhnsf+Moi4+16pIxSaojRcO+DtjNTteeulBpiNNz7wFdeSrggQHjs\nuGSXopRScTTce8mqqsL78U5aZ+iPqUqpoSehKX9F5F5gFmADS4wxb8WsWwQsBZqBx40xKyPLvwJ8\nHwgCPzbGPNvPtSeVP3J8e1CPb1dKDUE99txFZA5QZIwpBq4CVsSs8wArgfOA2cBCERknIqOAnwCf\nAxYAiwag9qTytc0EqT+mKqWGnkR67vOB1QDGmE0iki8iecaYWqAAqDHGVAKIyFrgTKARWGOMqQPq\ngKsHpPokih4poz13pdRQlEi4jwHWxzyvjCyrjTzOFZEiYAcwD3glsl2WiDwN5AM3GWPWHupN8vOz\n8Pl6P6tiIJDb63175Z0yGDuWUVOLBvd9EzTo7TGEaVvE0/aI59b26M1l9tp+PTTG2CJyBfAocADY\nHrN+FPBFYDzwsoiMN8bY3b1odXVDL0pxBAK5VFbW9Xr/w+XZXcGo3btpPncBtYP4voka7PYYyrQt\n4ml7xHNDe3T35ZRIuFfg9NSjCoHd0SfGmHXAGQAisgynBz8cKDHGBIGtIlIHBIB9vah9yPHp8e1K\nqSEukXB/EbgZeFBEZgIVkbF0AETkeeAKoB5YCCwHMoBVInInzrBMDlDVz7UnTdtMkHpmqlJqiOox\n3I0xJSKyXkRKgDBwrYhcCRwwxjwF/BLnC8AGlhljqgBE5AngjcjL/KsxJjwQHyAZ/NE53KfPSHIl\nSinVNcu2ux0GH1SVlXW9LmRQx81sm1GTj8HOyWP/2+8MznseJjeMI/YXbYt42h7x3NAegUBul2dR\n6hmqh8nz8U48+/fTqse3K6WGMA33w6THtyulUoGG+2HSmSCVUqlAw/0w+coj4a4/piqlhjAN98MR\nDuMrKyU4qQg7Ny/Z1SilVLc03A+Dd/tWPHW1Ot6ulBryNNwPgy96fLseKaOUGuI03A9DdLy9dbqG\nu1JqaNNwPwz+slJsj4fgCdOSXYpSSh2ShnuigkF875YTksmQlZXsapRS6pA03BPk/WALVkODc81U\npZQa4jTcE9R2fLvOBKmUSgEa7gnylzoXo9IzU5VSqUDDPUG+8lJsv5/glKnJLkUppXqk4Z6IlhZ8\nG991gj0jI9nVKKVUjzTcE+Db/D5WS4uemaqUShka7glou2aqnpmqlEoRGu4JiM7h3qo9d6VUitBw\nT4CvrBQ7M5OQHJ/sUpRSKiEa7j1pbMS3+X2CU6eB35/sapRSKiEa7j3wvfcuVjCoZ6YqpVKKL5GN\nROReYBZgA0uMMW/FrFsELAWagceNMStj1g0HNgK3GmNW9WPdg6b9yksa7kqp1NFjz11E5gBFxphi\n4CpgRcw6D7ASOA+YDSwUkXExuy8F9vdrxYOs7ZqpJ56U5EqUUipxiQzLzAdWAxhjNgH5IhK9xlwB\nUGOMqTTGhIG1wJkAInI8MAV4tt+rHkS+sg2Es3MIHTsp2aUopVTCEhmWGQOsj3leGVlWG3mcKyJF\nwA5gHvBKZLvlwHXAFYkUkp+fhc/nTajorgQCub3et1sHD8IWA7NnExgzov9ffwANSHukKG2LeNoe\n8dzaHgmNuXdgRR8YY2wRuQJ4FDgAbAcsEfka8LoxZruIJPSi1dUNvSjFEQjkUllZ1+v9u+N//e+M\nsG0apkyjfgBef6AMVHukIm2LeNoe8dzQHt19OSUS7hU4PfWoQmB39IkxZh1wBoCILMPpwX8RmCgi\nC4BxQLOIfGKMWdOb4pOl7cxUPVJGKZViEgn3F4GbgQdFZCZQYYxp+6oTkedxhl7qgYXAcmPM4zHr\nbwJ2pFqwA/jKI2em6hzuSqkU02O4G2NKRGS9iJQAYeBaEbkSOGCMeQr4Jc4XgA0sM8ZUDWTBg8lX\nuoHwESMITzgm2aUopdRhSWjM3RhzQ4dF5THrngSePMS+N/WqsiSzaqrxbd9Gy+x5YFk976CUUkOI\nnqHaDd87zveXzgSplEpFGu7d0JkglVKpTMO9G349UkYplcI03LvhK9tAuCBAeOy4njdWSqkhRsO9\nC1ZlJd5PPnZmgtQfU5VSKUjDvQv+d6JDMvpjqlIqNWm4d8FX6vyYquPtSqlUpeHehegc7q3Tteeu\nlEpNGu5d8JWVEioci33kkckuRSmlekXDvQPP7gq8e/folZeUUilNw70DnQlSKeUGGu4d6EyQSik3\n0HDvwB89Umb6jCRXopRSvafhHsu28ZWXEho/AXvkqGRXo5RSvabhHsOz8yM8+/frkIxSKuVpuMeI\nHt+uR8oopVKdhnuMtpkgdQ53pVSK03CPEZ3DPThtepIrUUqpvtFwjwqH8ZWXEZxUhJ2bl+xqlFKq\nTzTcI7zbt+Kpq9WZIJVSrqDhHqEzQSql3MSXyEYici8wC7CBJcaYt2LWLQKWAs3A48aYlZHldwFn\nRN5jmTHmyX6uvV+1zQQ546QkV6KUUn3XY89dROYARcaYYuAqYEXMOg+wEjgPmA0sFJFxIjIPmBrZ\n5xzgvoEovj/5SzdgezwEp56Q7FKUUqrPEhmWmQ+sBjDGbALyRST6i2MBUGOMqTTGhIG1wJnA34BL\nItvUANki4u3XyvtTMIhv4zuEZDJkZSW7GqWU6rNEhmXGAOtjnldGltVGHueKSBGwA5gHvGKMCQH1\nke2vAp6LLOtWfn4WPl/v8z8QyO31vmzcCA0N+IpP69vrDCFu+Rz9QdsinrZHPLe2R0Jj7h20XTHa\nGGOLyBXAo8ABYHvs+sh4/FXAF3p60erqhl6U4ggEcqmsrOv1/hkvvUoeUCdTaerD6wwVfW0PN9G2\niKftEc8N7dHdl1Mi4V6B01OPKgR2R58YY9bh/HCKiCzD6cEjImcDPwTOMcYc6E3Rg8VfpkfKKKXc\nJZEx9xeBxQAiMhOoMMa0fdWJyPMiMlpEsoGFwBoROQL4GbDAGLN/AOruV77yUmy/n+CUqckuRSml\n+kWPPXdjTImIrBeREiAMXCsiVwIHjDFPAb/E+QKwcQ55rBKRq3F+bP2DiERf6mvGmJ0D8SH6pKUF\n38Z3nWDPyEh2NUop1S8SGnM3xtzQYVF5zLongSc7bP8Q8FCfqxsEvs3vY7W06EyQSilXSfszVNvO\nTNWZIJVSLqLhHj0zVXvuSikX0XAvK8XOzCR0/ORkl6KUUv0mvcO9sRHfpvcITp0Gvt4c8q+UUkNT\nWoe77713sUIhWvX4dqWUy6R3uEevmapzuCulXCatw93fNoe7hrtSyl3SOtx95aWEs3MIHTsp2aUo\npVS/St9wP3gQ7xZDcPoM8A7d2YiVUqo30jbc/e+WY9m2npmqlHKltA13X1nkx1Q9M1Up5UJpHO7O\n9Uf0zFSllBulcbiXEh4xgvCEY5JdilJK9bu0DHerphrf9m3OeLtl9byDUkqlmLQMd195GaDHtyul\n3CtNw11nglRKuVtahrtfj5RRSrlcWoa7r2wD4YIA4cKxyS5FKaUGRNqFu1VZifeTj52ZIPXHVKWU\nS6VduPvLdbIwpZT7pV24t52ZqnO4K6VcLKHLD4nIvcAswAaWGGPeilm3CFgKNAOPG2NW9rRPMrUf\nKaM9d6WUe/XYcxeROUCRMaYYuApYEbPOA6wEzgNmAwtFZNyh9kkq28ZXuoFQ4VjsI49MdjVKKTVg\nEhmWmQ+sBjDGbALyRSQvsq4AqDHGVBpjwsBa4Mwe9kkaz57dePft1ZkglVKul8iwzBhgfczzysiy\n2sjjXBEpAnYA84BXetinS/n5Wfh8vZ9XPRDI7Xmjks0AZHyuOLHtU5jbP9/h0LaIp+0Rz63tkdCY\newdtxw8aY2wRuQJ4FDgAbI9d39U+3amubuhFKY5AIJfKyroet8ta9xrZQM2kKbQmsH2qSrQ90oG2\nRTxtj3huaI/uvpwSCfcKnF53VCGwO/rEGLMOOANARJbh9OAzD7VPsrSdmTp9RpIrUUqpgZXImPuL\nwGIAEZkJVBhj2r7qROR5ERktItnAQmBNT/skhW3jKy8lNH4C9shRSS1FKaUGWo89d2NMiYisF5ES\nIAxcKyJXAgeMMU8Bv8QJcxtYZoypAqo67jNgnyBBnp0f4dm/n6Yz5ia7FKWUGnAJjbkbY27osKg8\nZt2TwJMJ7JNU0ePb9cxUpVQ6SJszVP2l0WkH9DBIpZT7pU24+8pLsS2L4LTpyS5FKaUGXHqEeziM\nr7yM0KQi7Nykn0ullFIDLi3C3bttK566Wj0zVSmVNtIi3H1lOt6ulEov6RHu0ZkgZ5yU5EqUUmpw\npEW4+0s3YHu9BKeekOxSlFJqULg/3INBfBvfISSTISsr2dUopdSgcH24e7cYrIYG55qpSimVJlwf\n7m1npuqRMkqpNOL6cPdHj5Q5UacdUEqlD9eHu69sA7bfT3DyZ5JdilJKDRp3h3tLC773NhKcMhUy\nMpJdjVJKDRpXh7tv8/tYLS06E6RSKu24O9x1JkilVJpyd7hHz0zVI2WUUmnG3eFeVoqdmUno+MnJ\nLkUppQaVe8O9sRHfpvcITp0GvoQuOKWUUq7h2nD3vfcuVihEqx7frpRKQ+4N9+jJSzrerpRKQwmN\nV4jIvcAswAaWGGPeill3LfBVIAS8bYy5XkQKgUeBDMAL/LsxZn1/F38o/jK9ILZSKn312HMXkTlA\nkTGmGLgKWBGzLg/4HnCGMeZzwBQRmQV8B3jKGDMPuAG4fSCKPxRfeSnh7BxCk4oG+62VUirpEhmW\nmQ+sBjDGbALyI6EO0BK55YiID8gC9gNVwKjINvmR54PGOliHd4shOH0GeFw78qSUUt1KJPnGAJUx\nzysjyzDGNAE3A9uAj4B/GGO2APcCl4nIZuCXwI/7s+ie+N59B8u2dbxdKZW2enOMoBV9EOnB3wgc\nB9QCL4nIdGAh8AdjzO0isgC4G7joUC+an5+Fz+ftRTmOQCC3/cmH7wOQNed0smKXp5FAmn7urmhb\nxNP2iOfW9kgk3CuI9NQjCoHdkceTgW3GmCoAEXkVOAk4HVga2eavwP09vUl1dUOCJXcWCORSWVnX\n9jz3tRIygU+POZ5wzPJ00bE90pm2RTxtj3huaI/uvpwSGZZ5EVgMICIzgQpjTLQ1dgCTRWR45PnJ\nwAfAh8BpkWWnRJYNGl9ZKeERIwhPOGYw31YppYaMHsPdGFMCrBeREpwjZa4VkStF5IvGmL3Az4CX\nReQ1oNQY8ypwB7BARF4BbgP+fcA+QQdWTTW+7duc8XbL6nkHpVS/eeWVtQlv+/OfL6eiYle362+4\n4Tv9UVLaSmjM3RhzQ4dF5THrHgQe7LD9buC8PlfXC77yMkCPb1dqsO3eXcGaNS8wd+78hLZfsuS7\nh1z/05/e0x9lpS3XTbrSNhOkhrtSg+qee+5k06b3+NWvfkk4HKaiYhe7d1dw3333s2zZLVRW7qOx\nsZFvfONqTj/9DK677mq+853v8/LLa6mvP8jOnR+xa9cn/Nu/fZfi4tM5//z5PPvsWq677mpOOeU0\nNmx4m5qaGu68814KCgq45ZYfsWfPbk44YRovvbSGp556Lq6exx77X155ZS3hcJji4tP5xjeupq6u\njltuWUp9fT05OTmsXLmi07KbbrqDxx77DSNGjODiiy9j27YPueeeu1i58iG+9KUvctxxx3Pqqadx\n5JFH8fDD/43f7yc3N5dbbvmzhwj4AAAOxklEQVQpfr+f++67m/ff34jX6+V73/sBq1Y9wgUXfJGT\nTz6VlpYWvvrVS/jd7/6Ib4DnvHJduPt1DnelyL5pKRnPrO68wmMxMmz36jWbF15I/U23dbv+n//5\ncp588g98/ev/wiOPPEgw2Mr99z9MdfV+Tj11Fueeu4Bduz7hRz+6gdNPPyNu33379nL33St4440S\n/vSnP1JcfHr858nO5uc/f4AHHvgFf/vbSxQWjqOlpZmHHlrF3//+Kn/4w2Nd1nT//Q/j8Xi49NJF\nXHbZl3nssd9w6qnFXHLJl/j973/L66+/zj/+sT5u2dtvv9ntZ6yo2MUdd9zNxInH8tJLa/jJT26j\nsHAst976Y/7xj9fJyMhg3769PPTQKsrKNrB27V85++zzWLv2r5x88qmsX/8ms2Z9dsCDHVwY7r7y\nUsIFAcKFY5NdilJpbXLkusW5uXls2vQeTz/9JJblobb2QKdtp02bAcDo0aM5ePBgp/XTI+esjB49\nmgMHDvDRR9s54YTpABQXn47X2/kw6szMTK677mq8Xi81NTXU1tayZctmvvnNawC47LKvEAjk8pvf\n/DZuGcAHH5guP1Nm5nAmTjwWgBEjRnDnnbcRCoWoqNjFSSedQnX1/ra6ZsyYyYwZMwkGgzzwwAqC\nwSCvvrqO885bmGAL9o2rwt2qrMT7ycc0n3W2/piq0lr9Tbd12csOBHLZP0iH/vn9fgD++te/UFtb\ny3/918PU1tbyzW9e3mnb2HC27c7/s+i43rZtPB5nmWVZWB3+ve/Zs5vf//63PProb8nKyuLyyy8F\nwOPxYtvhuG27Whb7esFgMOYztUfmsmW38rOf3ceECcdwzz13dvtaPp+PU06Zxdtvv8n27duYOnVa\np883EFx1br6/XGeCVCpZPB4PoVCo0/KamhqOOqoQj8fDunUv0dra2uf3Gjt2HMY4Jyu++eYbnd63\npqaG/Px8srKyMGYze/bsobW1lcmTp7B+vTPv4erVf+Spp57qtOz55/9MdnY2VVXOrCnvvFPWZQ31\n9Qc58sgx1NXVsWHD+rbX37DhbQC2bNnM8uVO6J999nk88sh/c+KJJ/X5syfKVeHui84EqXO4KzXo\nxo8/BmM2s2LF8rjlc+d+npKSV1my5BqGDx/O6NGj+dWvftmn9/rsZ8+gvr6ea665ivLyUvLyjohb\nX1R0HMOHZ3HNNd9g7doXWbToIpYvv5NLLvlnNm58h+uuu5qSktc466yzOi2bM2cec+Z8ntdeW8f1\n13+7y2EigIsuuoRrrrmKu+66na985Wv87/+uYty4oxk//hi+/e1vct99d3PhhRcDcPzxk6mtreWs\ns87p0+c+HFZX/wVKhsrKul4XEj3LLO/yy8h44Xmq3v0A+8gj+7O8lOKGs+76i7ZFPLe0R23tATZs\neJu5c+dTWbmPJUuu4Xe/++Nhv85gtcfOnR+xfPmd/PznPZ6sf9gCgdwux6DdM+Zu2/hKNxAqHJvW\nwa5UOsjKyuall9bwu9/9BtsO86//OnRPeFq9+gmefvopfvjDmwf1fV0T7p49u/Hu20vzIP0SrZRK\nHp/Pxy23LEt2GQm58MLFXHjh4kF/X9eMufv0+HallGrjnnCPHCnTqkfKKKWUe8K9/ZqpGu5KKeWO\ncLdtfGUbCI2fgJ0/MtnVKKVU0rkj3HfswFNdrZOFKZUiFi9eSENDA7/5zSo2bnwnbl1DQwOLFx/6\nwIjo1MLPPfcM69a9PGB1pjJ3HC3ztnNGmE7zq1RqufzyKw97n9iphQdrnpZU5I5wf8s5dVjH25VK\nnm984yvcccdyxowZw549u7nxxu/xi1/8NzffvJTGxkaampr493//HlOmTG3b5/bbb2Lu3PnMmHEi\nP/zh92lpaWmbRAzgxRef54knfo/X62HChGP5z//8YaephaNT895//895991ygsEQF198Keecc36X\n0wWPGdN+1dA9e/Zw/fXOMfLBYJClS29m7Nhx/OUvz/LEE7/Hsiy+9KWvMH/+F7pcFp2WGGDp0u9z\n0UWXUlq6vsfpjqNTE3g8FlOnTuf88y/grrtu5/77Hwbg179+hKysbC655Eu9/vNwR7i//Ta2ZRGc\nNj3ZlSg1JNx0UwbPPNP5n7fHA+Fwdq9ec+HCIDfd1Nzt+tmz5/H3v/+Niy++lFdfXcfcuZ/n008/\nZcGCC5k9ey7r17/Fb3/7a26//Wed9n3hheeZOPFY/u3fvsvatS+yZs0LADQ2NrJ8+S/Izc3l2mv/\nha1bP+w0tTBAWdkGtm3bygMPPEpjYyNXXPElZs+eC3SeLvjSS7/c9r779u3j61//F2bOPJk///lP\nPPnk/3HVVVezatXD/PrXj9HS0srtt/+E4uLTOy2bP/8L3bZFT9Md33ff3XzvezcyaVIRt976YzIz\nM2ltbWHfvr2MHn0kJSWvsWzZ3b35Y2qT+uEeDsP69YQmFWHn5iW7GqXS1uzZ81i58j4uvvhSXntt\nHd/97g2MHDmKX//6YR577De0traSmZnZ5b47dmxjxgxnUq3YybXy8vL4wQ+cKzZ99NF2Dhyo6XL/\nzZvfZ0ZkWHb48OFMmDCRjz/+GOg8XXCsQCDAihX/xSOPPEhdXS0ik9mxYztHHz2BjIxMMjIy+elP\n7+H99zd2WnYoPU13vHPnR0yaVATAj350CwBf+MJ5vPTSXznzzLPJzs5h5MhRh3yPnqR8uHu3bYXa\nWoJfODfZpSg1ZNx0U3OXvWxnLpX6AXnPiROP5dNPK9m7dw91dXUcffR4Hn30IQoKRvOjH93K5s3v\ns3LlfV3ua9vg8ThTpIQjFxNpbW3lnnvuYtWq3zFqVAHf//713b63ZVnETpMVDLa2vd6hphNesWIF\np502iwsvXMzLL6+hpOS1hKcF7ih+auBDT3fs8XQ+luXMM89m6dLvk5k5nLPOOvuQ75WIlD9axlcW\nOTNVZ4JUKumKiz/HQw/dzxlnzAHgwIEaxo4dB8C6dS/HBWCso48ez+bNmwDapsxtaKjH6/UyalQB\ne/fuYfPmTQSDwS6nFj7++M9QWro+sl8Du3Z9wrhxR/dYb3V1NWPHjsO2bV57bR2tra2MHz+BnTs/\noqGhgebmZq6//ttdLrNtG8uyaGpqoqmpiS1bOl/go7vpjidMOIb33tsIwLJlt7Bjx3by8/PJy8vj\nhReeY86ceT3W3pOU77lHw711uoa7Usk2Z848vvWtb7BqlXPZu3POOZ/bbvsJL7+8hosvvpQ1a17k\n2Wef7rTfOeecz403/gdLllzDtGkzsCyLI44YwSmnnMY3v/k1Jk0q4stfvpwVK+7hF794sG1q4ezs\nHACmT5+ByPFce+2/EAwG+da3rmP48OE91nvZZZdxxx3LGDOmkMWLL+Ouu27n3XfLueqqb3H99d+O\nbPNlhg8f3mmZZVlceOFirr76CiZMmIjI5E6vP3fu57nhhu/w/vsbOf/8C9qmO16y5D+4+25nbpzP\nfOYEJkw4JrL9fP7+91fJyurd7yKxEpryV0TuBWYBNrDEGPNWzLprga8CIeBtY8z1keX/EVneCnw7\ndp+u9HbK3yMWncuwN9+gcusuyMrqzUu4jlumde0P2hbxtD3iDbX2uO22n3DeeQuZOfPkhPfpbsrf\nHodlRGQOUGSMKQauAlbErMsDvgecYYz5HDBFRGaJyGeALwEnA/8PWJBwpYep5Zzz4cYbNdiVUimr\nubmZq6++kuzs7MMK9kNJZFhmPrAawBizSUTyRSTPGFMLtERuOSJyEMgC9gNfBP5gjAkCGyK3AdF4\nzXXkBHJhCH37KqXU4cjIyOChh1b162sm8oPqGKAy5nllZBnGmCbgZmAb8BHwD2PMFmACcLSI/EVE\n1oqIHoCulFKDqDc/qLaN70SGZW4EjgNqgZciQW4BXuBc4HTgYeCUQ71ofn4WPp/3UJscUiCQ2+t9\n3Ujbo522RTxtj3hubY9Ewr2CSE89ohDYHXk8GdhmjKkCEJFXgZOAvcBmY4wNvCYiE3p6k+rqhsMo\nO95Q+1Ek2bQ92mlbxNP2iOeG9ujuyymRYZkXgcUAIjITqDDGRFtjBzBZRKLHHJ0MfAA8D5wd2ed4\n4OPeFq6UUurw9dhzN8aUiMh6ESkBwsC1InIlcMAY85SI/Ax4WUSCQIkx5lUAETlXRF6PvMy1A1S/\nUkqpLiR0nPtg6O1x7uCO/1r1J22PdtoW8bQ94rmhPXp9nLtSSqnUo+GulFIuNGSGZZRSSvUf7bkr\npZQLabgrpZQLabgrpZQLabgrpZQLabgrpZQLabgrpZQLpfxl9g51lah0IyJ3AWfg/LkuM8Y8meSS\nki4y79FG4FZjzKokl5NUIvIV4PtAEPixMebZJJeUFCKSA/wPkA9kADcbY15IblX9L6V77oe6SlS6\nEZF5wNRIW5wDdH2Z+fSzFOcCMmlNREYBPwE+h3NltEXJrSiprgSMMWYezqSIP09uOQMjpcOdDleJ\nAvIjc8yno78Bl0Qe1wDZItL7CfJdIDIj6RQgLXuoHZwJrDHG1Bljdhtjrk52QUlUBYyKPM6PPHed\nVA/3bq8SlW6MMSFjTH3k6VXAc8aYUDJrGgKWA99JdhFDxAQgS0SeFpFXRWR+sgtKFmPM4zhXivsQ\np1P0H0kuaUCkerh31OXsaOlERBbhhPt1ya4lmUTka8Drxpjtya5liLBweqsX4QxL/EpE0vLfi4h8\nFdhpjJkEfB5YmeSSBkSqh/uhrhKVdkTkbOCHwLnGmAPJrifJzgcWicgbwDeBH4nImUmuKZn24lxv\nIWiM2QrUAYEk15QspwMvABhjyoFCNw5hpvrRMi/iXKD7wS6uEpVWROQI4GfAmcaYtP8B0RhzWfSx\niNwE7DDGrEleRUn3IrBKRO7EGWfOwaVjzQn4EDgN+KOIjAcOunEIM6XDvaurRCW7piS6DCgA/iAi\n0WVfM8bsTF5JaqgwxuwSkSeANyKL/tUYE05mTUn0IPCoiKzDycBvJbmeAaFT/iqllAul+pi7Ukqp\nLmi4K6WUC2m4K6WUC2m4K6WUC2m4K6WUC2m4K6WUC2m4K6WUC2m4K6WUC/1/dYW3ZfWmPX8AAAAA\nSUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "score_3 = model_3.evaluate(x_test, y_test, verbose=0) \n",
    "print('Test score:', score_3[0]) \n",
    "print('Test accuracy:', score_3[1])\n",
    "\n",
    "plt.plot(my_model_3.history['acc'], 'r')\n",
    "plt.plot(my_model_3.history['val_acc'], 'b')\n",
    "plt.legend({'training accuracy': 'r', 'validation accuracy':'b'})\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 265
    },
    "colab_type": "code",
    "id": "LtJPL05q-CgA",
    "outputId": "2f03bb8a-3a94-4ae7-851b-48229d5ee1fa"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD4CAYAAAAXUaZHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/5X42BY9QQ9OAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(my_model_3.history['loss'], 'r')\n",
    "plt.plot(my_model_3.history['val_loss'], 'b')\n",
    "plt.legend({'training loss': 'r', 'validation loss':'b'})\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "JojyQpON-qhq"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "7G6owLFyupGF"
   },
   "source": [
    "# Performance Comparison of the Models Using Pretty Table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 136
    },
    "colab_type": "code",
    "id": "prO8c6xnuuOm",
    "outputId": "006195f5-c3ef-4ede-9865-c17e1f3a34a8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+-------+-------------------------------------+--------------------+--------------------+\n",
      "| S.NO. |                MODEL                | Training Accuracy  |   Test Accuracy    |\n",
      "+-------+-------------------------------------+--------------------+--------------------+\n",
      "|   1   | Model 1 -- CNN with 3 Hidden Layers | 0.9992333401997884 | 0.9989699977874756 |\n",
      "|   2   | Model 2 -- CNN with 5 Hidden Layers | 0.9912500000317891 |       0.9942       |\n",
      "|   3   | Model 3 -- CNN with 7 Hidden Layers | 0.9907166666984558 |       0.9943       |\n",
      "+-------+-------------------------------------+--------------------+--------------------+\n"
     ]
    }
   ],
   "source": [
    "from prettytable import PrettyTable\n",
    "\n",
    "# Names of models\n",
    "names = ['Model 1 -- CNN with 3 Hidden Layers','Model 2 -- CNN with 5 Hidden Layers','Model 3 -- CNN with 7 Hidden Layers']\n",
    "         \n",
    "\n",
    "# Training accuracies\n",
    "train_acc = [max(my_model_1.history['acc']),max(my_model_2.history['acc']),max(my_model_3.history['acc'])]\n",
    "\n",
    "# Test accuracies\n",
    "test_acc = [score_1[1],score_2[1],score_3[1]]\n",
    "\n",
    "numbering = [1,2,3]\n",
    "\n",
    "# Initializing prettytable\n",
    "ptable = PrettyTable()\n",
    "\n",
    "# Adding columns\n",
    "ptable.add_column(\"S.NO.\",numbering)\n",
    "ptable.add_column(\"MODEL\",names)\n",
    "ptable.add_column(\"Training Accuracy\",train_acc)\n",
    "ptable.add_column(\"Test Accuracy\",test_acc)\n",
    "\n",
    "# Printing the Table\n",
    "print(ptable)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "cI91kT47v_QB"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "Various_CNN_Architectures.ipynb",
   "provenance": [],
   "version": "0.3.2"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}