Training using CNN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "data=np.load('data.npy')\n",
    "target=np.load('target.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import cv2\n",
    "import os\n",
    "import keras\n",
    "from keras.layers import *\n",
    "from keras.models import * \n",
    "from keras.preprocessing import image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras import Input\n",
    "from keras.callbacks import ModelCheckpoint\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(392, 224, 224, 3)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(392,)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "target.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n",
      "1\n"
     ]
    }
   ],
   "source": [
    "for i in range(target.shape[0]):\n",
    "    print(target[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# CNN Based Model in Keras\n",
    "\n",
    "model = Sequential()\n",
    "model.add(Conv2D(32,kernel_size=(3,3),activation='relu',input_shape=(224,224,3)))\n",
    "model.add(Conv2D(64,(3,3),activation='relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "model.add(Dropout(0.25))\n",
    "\n",
    "model.add(Conv2D(64,(3,3),activation='relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "model.add(Dropout(0.25))\n",
    "\n",
    "model.add(Conv2D(128,(3,3),activation='relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "model.add(Dropout(0.25))\n",
    "\n",
    "model.add(Flatten())\n",
    "model.add(Dense(64,activation='relu'))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(1,activation='sigmoid'))\n",
    "\n",
    "model.compile(loss=keras.losses.binary_crossentropy,optimizer='adam',metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_1 (Conv2D)            (None, 222, 222, 32)      896       \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 220, 220, 64)      18496     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 110, 110, 64)      0         \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 110, 110, 64)      0         \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 108, 108, 64)      36928     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 54, 54, 64)        0         \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 54, 54, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_4 (Conv2D)            (None, 52, 52, 128)       73856     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_3 (MaxPooling2 (None, 26, 26, 128)       0         \n",
      "_________________________________________________________________\n",
      "dropout_3 (Dropout)          (None, 26, 26, 128)       0         \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 86528)             0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 64)                5537856   \n",
      "_________________________________________________________________\n",
      "dropout_4 (Dropout)          (None, 64)                0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 1)                 65        \n",
      "=================================================================\n",
      "Total params: 5,668,097\n",
      "Trainable params: 5,668,097\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "train_data,test_data,train_target,test_target=train_test_split(data,target,test_size=0.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(352, 224, 224, 3)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(40, 224, 224, 3)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(352,)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_target.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 316 samples, validate on 36 samples\n",
      "Epoch 1/10\n",
      "316/316 [==============================] - 72s 228ms/step - loss: 1.6736 - accuracy: 0.5095 - val_loss: 0.6989 - val_accuracy: 0.4444\n",
      "Epoch 2/10\n",
      "316/316 [==============================] - 62s 195ms/step - loss: 0.7008 - accuracy: 0.5253 - val_loss: 0.6832 - val_accuracy: 0.5556\n",
      "Epoch 3/10\n",
      "316/316 [==============================] - 69s 218ms/step - loss: 0.6633 - accuracy: 0.5854 - val_loss: 0.6482 - val_accuracy: 0.9167\n",
      "Epoch 4/10\n",
      "316/316 [==============================] - 69s 217ms/step - loss: 0.6083 - accuracy: 0.6741 - val_loss: 0.5774 - val_accuracy: 0.8333\n",
      "Epoch 5/10\n",
      "316/316 [==============================] - 70s 223ms/step - loss: 0.4485 - accuracy: 0.8165 - val_loss: 0.3262 - val_accuracy: 0.9444\n",
      "Epoch 6/10\n",
      "316/316 [==============================] - 68s 217ms/step - loss: 0.2817 - accuracy: 0.8892 - val_loss: 0.1917 - val_accuracy: 0.9444\n",
      "Epoch 7/10\n",
      "316/316 [==============================] - 67s 212ms/step - loss: 0.2157 - accuracy: 0.9177 - val_loss: 0.1522 - val_accuracy: 0.9722\n",
      "Epoch 8/10\n",
      "316/316 [==============================] - 68s 216ms/step - loss: 0.1984 - accuracy: 0.9304 - val_loss: 0.1299 - val_accuracy: 0.9444\n",
      "Epoch 9/10\n",
      "316/316 [==============================] - 57s 179ms/step - loss: 0.1474 - accuracy: 0.9430 - val_loss: 0.1293 - val_accuracy: 0.9444\n",
      "Epoch 10/10\n",
      "316/316 [==============================] - 56s 176ms/step - loss: 0.1691 - accuracy: 0.9367 - val_loss: 0.1067 - val_accuracy: 0.9444\n"
     ]
    }
   ],
   "source": [
    "hist = model.fit(train_data,train_target,epochs=10,validation_split=0.1,batch_size=64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(hist.history['loss'],'r',label='training loss')\n",
    "plt.plot(hist.history['val_loss'],label='validation loss')\n",
    "plt.xlabel('# epochs')\n",
    "plt.ylabel('loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEGCAYAAABo25JHAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAAznklEQVR4nO3dd3hU1db48e9KCIYAwUCQDgFRISAECEVFQQUEVLCgoKJXFBAU0Ndru1694rX8fBURvSqIvKgURaSKF6mCiNSEJr2XGIEAoYaQtn9/nElIwiRMIJMzM2d9nicPmZkzk5UB9pq91zlrizEGpZRSzhVkdwBKKaXspYlAKaUcThOBUko5nCYCpZRyOE0ESinlcKXsDqCoIiMjTVRUlN1hKKWUX4mPjz9ijKns7jG/SwRRUVHExcXZHYZSSvkVEdlX0GO6NKSUUg6niUAppRxOE4FSSjmcJgKllHI4TQRKKeVwmgiUUsrhNBEopZTD+d11BEop70lNz2T5rqPsSjrNjVdH0rBaeUTE7rCUl2kiUMrhks+k8cvWw8zffIglO5JIScvMeaxmRBk6NKxCp+gqtKxbkZBgXUQIRJoIlHKgfUfPMH/zIeZtPkTc3mNkGagSfgX3NqtBx+gqXFulPL/tSGL+5kN8t2o/Xy/bS3hoKW5tcBUdo6vQ7trKlA8NsfvXUMVE/G2HstjYWKMtJpQqmqwsw/qE4yzYcoj5mw+x/dBpABpULU/H6Cp0jK7C9TUquF0GSknL4LcdR1iw+RALtx7m2Jk0QoKFNvUq0Sm6Ch2iq1CtQpmS/pVUEYlIvDEm1u1jmgiUCkzZ6/3zNh9i4ZZDHD51juAgoVVUxZzBv1bFsCK9ZmaWYc3+ZOZvthLKniNnALi+RgU6RlehQ8MqWlfwUZoIlHIId+v9ZUsH0/66q+gQfRW3XncVV4aVLraft/PwaVdSOMjaA8cxRusKvkoTgVIBrKD1/g4NrU/9N1xdiStKBXs9jqRT5/hlqzVT+G3HEc5lZGldwYdoIlAqgGSv92cvz+w47Pl6f0lJSctg6Y4jzNe6gs/QRKCUn0tNz2TZriPM33y42Nb7S0p2XWGBK3HtdtUVGtcIp2PDqnSM1rpCSdBEoJQfKmy9v2N0FdpfV7lY1/tLSnZdYcGWQ6zZn4wxUOPKMjkJrZXWFbxCE4FSF5GSlsGp1Ay7w+BUajqLtyXlWe+vGh5Kh+ir6NCw5Nb7S0phdYUODavQok4EwUE6U8gWVjr4kussmgiUKsTGP0/w0OgVnDpnfyLI5kvr/SXFXV1B5TWg3dW80qXBJT23sESgVxYrR0s6dY5+4+IILxPCK10bINg74GYXVH11vd+bwkqXolOjqnRqVDWnrrDDdeFbscvKguPJcPgwHE6CpCRIOmzdTkqCtFxJKCgIgoIhI73g1wsKhrJloWwYhLn+LFv2/FdB95UpY72+hxpWK38Zv3TBNBEox0rLyOLpifEkp6QxZcCNNK5Rwe6QlEtwkNAyqiItoype+oukp8P+/bBz54Vfu3fnHexLl4arr4b69aFt/fPf168PtWtDSAicPQvHjkFysvVngV8JsDfX7VOnCo5RBK68EipWPP8VEZH3du6vUnWBiEt/TwqgiUA51tBZm1i9N5n/PNRMk4C/OncO9uxxP9jv3QuZ5xvoERZmDezR0XD33ecH+vr1oUYNCL5I7aVMGeu4GjWKFmN6Ohw/fpHkketr1y4r2SQnWzOX3F56Cf73f4v28z2giUA50oQV+/h25X6ebn81dzetbnc4qjApKdbgmHuQz769fz/krnOGh8M110BsLPTsmXewr1rV+gRe0kJCoHJl66sosrLg5Mm8SaJWLa+EqIlAOc7K3UcZ+uMmbmtwFX/vdJ3d4SiwBrz8g332V2Ji3mMjI11LOG3zDvT160OlSvYM9t4QFGQtG115JdSr59UfpYlAOUpCcgpPT1xD7UphjOgVo6cmlqRjx9wP9Dt3WgXa3KpVswb2Tp3yDvRXX20NjKpYaSJQjnE2LZP+4+JJy8ziy8diCde+N8XLGOusG3cDffa6d261almD+z335B3s69WDcuVs+RWcShOBcgRjDC9OWc+WgycZ+3hLrq6sA80lycqylmrcrdfv3Amnc53uGRQEUVHW4N6qVd7Bvm5dCA217ddQeWkiUI7w+eJd/LThL17p0oBbr7vK7nB8mzGwb1/Bn+xTU88fGxJifYKvXx/atct72mWdOtZpmcrnaSJwqLmbDrLv6Bn6tq1HUICvky/ccohh87bRrWl1nrrFu0U3v5WZCb//DtOnw4wZ1qmX2UJDzw/wnTvn/WRfq9bFT7tUPk8TgUMNm7uNHYdPs/WvU/xvjyYB2+Rr5+FTPDtpHY2qh/O/9zdxRKsGj6WmwoIF1uD/449w5Ij1Cb5jR+t89ehoa7CvVq1IV78q/6OJwIGOnD7HjsOnia4WzrS1f3IyNZ1PH25OaEhgfbI7cTadfuPiCQ0J4otHYylTOrB+v0ty4gTMnm0N/j//bK3ph4fDnXfCvfdan/jLe6eNgfJdmggcaOXuYwC8c29jNiae5F8zN/LY/61izOOBcyZNZpZhyHdrSUhO4dt+bahxpYM3QTl4EGbOtAb/X36xrnStWhUeecQ6Y+e223Qt3+G8mghEpDPwMRAMjDHGvJfv8QhgLHA1kAo8YYzZ6M2YFKzYfZSypYNpXKMCzWpHcGWZEJ6fvI5eX6zgmydaUbn8FXaHeNnen7uVX7cn8e69119evxp/tXOnNfBPnw4rVlgF4KuvhmeftT75t2mjyz0qh9cSgYgEA58BHYEEYLWI/GiM2ZzrsFeBdcaYe0Wkgev4270Vk7Ks2H2U2Kjzm3/c3bQ64WVCGDA+ngdGLWP8k639uvvlzHV/8sWvu+ndpjYPt65tdzglwxhYu/Z8sXej6/NUs2bw5pvW4N+oUeBcdauKlTc/ErQCdhpjdhtj0oBJQPd8x0QDCwGMMVuBKBGp4sWYHC+7PtCmXqU897e7tjIT+rYmOSWdHqOWse1gIR0TfdgfCSd4acoGWtWtyL/uamR3ON6VkQGLF8Nzz1nn5bdoAe++a7VZGDHCOvNnzRp4/XVo3FiTgCqQNxNBDeBArtsJrvtyWw/cByAirYA6QM38LyQi/UUkTkTikvJfiq6KJLs+0KbehcslLepEMPmpGzAGHvxiOfH7ki84xpclnTpH//FxRJa7gs8faU7pUgG49HH2LMyaBU88Ya3z33orjBoFTZrA2LFWPWDxYmsJqE4du6NVfsKb/1PcffzIvx3ae0CEiKwDBgNrgQu2iTLGjDbGxBpjYisXtYOfyiN3fcCd66qWZ+rAG4kIC6H3mJX8ut0/Em9aRhYDJ1h7C4x+rAWR5fy/zpHj+HGYOBF69LA6WHbrBtOmWWf4/PCDddrnjz9Cnz5F73CpFN4tFicAuXum1gTytBE0xpwE+gCIdYL3HteX8pL89QF3alUM44cBN/LY2FX0/WY1wx+M8elWzcYY3vhxI3H7kvn04WY0qh4AewskJp4/02fRImsZqFo1ePRRa72/fXs900cVG28mgtXANSJSF/gT6AU8nPsAEbkSSHHVEPoCS1zJQXlBdn3gvuYXrL5doHL5K/j+qTb0/TqOIZPWcvxsOo+28c2lhgkr9/PdqgM8c+vV3NXEdxPWRZ09C599BlOnWmf6gNVb//nnrcG/VSs900d5hdcSgTEmQ0QGAXOxTh8da4zZJCIDXI+PAhoC40QkE9gMPOmteFTh9QF3wkNDGPdkK56ZuIbXZ2zk+Jk0Bt1W36euzl2x+yhv/riJ2xtcxd87+vHeAvv3W4P9mjVW0fftt61z/KOjtcirvM6r1xEYY2YDs/PdNyrX98uBa7wZgzrvYvUBd0JDghn1aAtenrKBD+dvJzklndfubOgT/Ymy9xaoUymMj3rF+ERMl2TJEmv9PzXVWuu/+267I1IOo1cWO4gn9QF3QoKDGPZAU64MK83Y3/dwPCXN9v5EKWkZ9B8XT7o/7y1gDIwcaZ3hU6+eVRNo0MDuqJQDaSJwiKLUB9wJChJev6shFcuGMGzedk6cTeezR+zpT2SM4cUfNrDVtbdAPX/cW+DcORg0CMaMsfr8TJwIFQKgyK38klaeHKKo9QF3RIRBt13D2/c05pdth3ns/1Zx4mx6cYXosc8X7+K/f/zFy50b0N4f9xZITLTO+hkzBv75T2smoElA2UgTgUNcSn2gIL3b1OGTXs1YeyCZXqNXkHTqXDFE6JkFm629BbrHVKe/P+4tsGIFxMbCH39Y1wC8/bb281e200TgEJdaHyjI3U2rM+ZvLdl75AwPjFrGgWMpxfK6hdl5+BTPfe/HewuMHWvt4hUaCsuXWwVipXyAJgIHKKi/0OVqd21lJvaz+hPdP9K7/YlOpJzfW2D0o7H+tXdCerpVD3jySbjlFli9Gq6/3u6olMqhicABiqM+UJDmtSP4YcANiHivP1FmlmHwJGtvgZG9W1Ddn/YWOHwYOnSwLhT7+9+tzWAqFW9CVupyaSJwgOKsD7hzbZXyTBngvf5E78/ZypLtSfy7e2P/2lsgPt6qB6xaBRMmwLBhUEpP1FO+RxOBAxR3fcCd7P5EdSPL0veb1cxan3jxJ3lgxto/+WLJbh5tU4eHWvnR3gITJ0Lbttb3S5dau4Ep5aM0EQQ4b9UH3Klc/gomPdWGZrUjGDJpLeNX7Lus19uQcJyXp7r2Frg7upii9LKMDHjhBejd2+oNFBdntYxQyodpIghw3qwPuBMeGsK4J1pxe4OreH3GRj5ZuANj8ncfv7jDp1J5anw8keWuYOQjzW29itljx45B167w4YfwzDOwYAFc5YfXOSjH8YP/XepyeLs+4E5oSDAje7fgvmY1GD5/O//+aTNZWZ4ng3MZmQycsCZnb4FK/rC3wB9/QMuW8Ouv1oVin34KIX7Y9kI5klauAlxJ1AfcubA/UTrve9CfyBjDGzM3Ee9PewtMnQp/+xuEh1uJoE0buyNSqkh0RhDASrI+4E52f6IXOl3L9LV/MmB8PKnpmYU+Z8KKfUxa7Sd7C2RlwWuvWReGXX+9VQ/QJKD8kCaCAFbS9QF3itKfaPmuo7w5a7N/7C1w4oS1ZeQ771gXii1eDNV9PHEpVQBNBAHMjvpAQXq3qcN/Hjrfn+jwqdQ8jx84lsIz364hKrIsI3x9b4GtW60zgubOtS4U+/JLuMIP6hhKFUATQQCzqz5QkLua5O5PtDynP1FKWgb9x8eT4dpboLwv7y0wa5aVBJKTYeFCePpp3UFM+T3fGCFUsbO7PlCQ7P5Ex139ibb8dZIXf9jAtoMn+c/DzakbWdbuEN3LyrI6hXbvbu0jHB9v9Q1SKgBoIghQvlAfKEju/kTdPl3Kf//4i1e6NKDdtZXtDs29U6fggQfg9detK4SXLoVateyOSqlio4kgQPlSfcCd7P5E9SLL0atlLfrd7KN7C+zaBTfcADNmWBeKjRsHZfyo6Z1SHtDrCAKUr9UH3KlVMYw5z93su/sKzJsHPXtCUJBVGO7Qwe6IlPIK3x0l1CXz1fqAOz6ZBIyxOoV26WItAa1erUlABTRNBAHIl+sDPi8lxaoDvPgi3H+/tZNYPR9dtlKqmGgiCEC+Xh/wWfv2Wa2jJ02Cd9+F77+Hsj56FpNSxUhrBAHIH+oDPmfxYuvMoLQ061qBO++0OyKlSoyOFAHGn+oDPmPTJrjjDoiMtHYT0ySgHEZnBAFG6wNFlJUF/fpB+fKwZAlU9tFrGZTyIk0EAUbrA0U0apRVEB43TpOAcixdGgowWh8ogj//hFdegY4dra0llXIor44WItJZRLaJyE4RecXN4xVEZJaIrBeRTSLSx5vxBDqtDxTR4MHWHsOjRmnjOOVoXksEIhIMfAZ0AaKBh0Qk/w7kzwCbjTFNgfbAhyJS2lsxBTqtDxTB9OnW19Chep2AcjxvzghaATuNMbuNMWnAJKB7vmMMUF6sy0vLAceADC/GFNC0PuChEydg0CBo2hT+53/sjkYp23mzWFwDOJDrdgLQOt8xnwI/AolAeaCnMSYr/wuJSH+gP0Dt2rW9Emwg0PqAh159FQ4etBrJ6QbzSnl1RuBu0dXku30HsA6oDsQAn4pI+AVPMma0MSbWGBNbWc/scEvrAx5atgxGjoQhQ6BlS7ujUconeDMRJAC5m7bXxPrkn1sfYJqx7AT2AA28GFPA0vqAB9LSrGsGatWCt96yOxqlfIY3E8Fq4BoRqesqAPfCWgbKbT9wO4CIVAGuA3Z7MaaAlV0fuF7rAwV7/33YvNmaEZQrZ3c0SvkMr9UIjDEZIjIImAsEA2ONMZtEZIDr8VHAW8DXIvIH1lLSy8aYI96KKZCt2H2UlnUrUkrrA+5t22bNAnr2hK5d7Y5GKZ/i1SuLjTGzgdn57huV6/tEoJM3Y3CC7PrA/S1q2h2Kb8rKgv79ISwMRoywOxqlfI62mAgA5+sDWih266uvrD5CY8ZA1ap2R6OUz9F1hACQc/1A9QtOuFIHD8ILL0C7dvDEE3ZHo5RP0kQQALQ+UIjnnoOzZ+GLL7SNhFIF0JHDz+n1A4X473+tXcZeew2uu87uaJTyWZoI/JzWBwpw+jQ8/TRER8NLL9kdjVI+TYvFfk7rAwV4/XU4cACWLoXS2sdQqcLojMDPaX3AjdWr4ZNPYOBAuPFGu6NRyufp6OHHtD7gRnq61UaialV49127o1HKL+jSkB/T+oAbH30E69fDtGlQQdttKOUJnRH4Ma0P5LNrl7XRzD33wL332h2NUn5DE4Ef0/pALsbAgAFQqhR8+qnd0SjlV3QE8VNaH8hnwgRYsADeew9q1LA7GqX8iiYCP6X1gVyOHLG2nLzhBmtWoJQqEk0EfkrrA7k8/zycPAlffglB+k9aqaLS/zV+SusDLvPnw/jx8PLL0KiR3dEo5ZccPor4J60PuKSkWEtB114L//yn3dEo5bf0OgI/pPUBlzffhN27YfFiCA21Oxql/JbOCPzQ8t1HtD6wbh18+CE8+aS114BS6pJpIvBDK3Yfc3Z9IDPTaiNRqZK1Ib1S6rI4dCTxX0mnzrHT6fWB//wH4uKsxnIVK9odjVJ+TxOBn1m55yjg4PrAvn3WRjNdu8KDD9odjVIBQROBn3H09QPGwDPPWN9//rluPalUMfEoEYjIsyISLpb/E5E1ItLJ28GpCzm6PjB5srX95NtvQ506dkejVMDwdDR5whhzEugEVAb6AO95LSrllqPrA8nJMGQIxMbC4MF2R6NUQPH0OoLsOXhX4CtjzHoRnZeXNEfXB156CY4ehblzITjY7miUCiiezgjiRWQeViKYKyLlgSzvhaXccWx94NdfYcwY+PvfISbG7miUCjiezgieBGKA3caYFBGpiLU8pEqQI+sDqanQvz/UrQtvvGF3NEoFJE9HlBuAbcaY4yLSG3gNOOG9sFR+jq0PvPsubN8Oo0ZBWJjd0SgVkDxNBCOBFBFpCrwE7APGXexJItJZRLaJyE4RecXN4y+KyDrX10YRyXTNNlQ+jqwPbNpkbTTTuzd00pPUlPIWTxNBhjHGAN2Bj40xHwPlC3uCiAQDnwFdgGjgIRGJzn2MMeYDY0yMMSYG+AfwqzHmWBF/B0dwXH0gK8tqIxEeDsOH2x2NUgHN0xrBKRH5B/AocLNrkA+5yHNaATuNMbsBRGQSViLZXMDxDwHfeRiP4ziuPvDFF7B8OXzzDVSubHc0SgU0T0eVnsA5rOsJDgI1gA8u8pwawIFctxNc911ARMKAzsDUAh7vLyJxIhKXlJTkYciBw3H1gT//tDaa6dABHn3U7miUCngeJQLX4D8RqCAidwGpxpiL1QjcXWdgCjj2buD3gpaFjDGjjTGxxpjYyg78dOi4+sDgwZCebhWI9XIVpbzO0xYTDwKrgAeAB4GVItLjIk9LAGrlul0TSCzg2F7oslCBHFUfmD7d+ho6FK6+2u5olHIET2sE/wRaGmMOA4hIZWABMKWQ56wGrhGRusCfWIP9w/kPEpEKQDugdxHidhTH1AdOnIBBg6BpU2tDeqVUifA0EQRlJwGXo1xkNmGMyRCRQcBcIBgYa4zZJCIDXI+Pch16LzDPGHOmaKE7Q3Z9oEeLmnaH4n2vvgp//QUzZkDIxc5FUEoVF08TwRwRmcv55ZuewOyLPckYMzv/cbkSQPbtr4GvPYzDcRxTH1i2DEaOtBrLtWxpdzRKOYpHicAY86KI3A/chFUEHm2Mme7VyBTgkPpAWprVRqJmTXjrLbujUcpxPJ0RYIyZSgGndyrvcUR94P33rauIf/oJyhd6naJSygsKTQQicgr3p3wKYIwxAfwx1X6OqA9s22bNAh58EO680+5olHKkQhOBMUY/ntko4OsDKSnQp4/VTO7jj+2ORinH8nhpSJW8gK4PnD4Nd90FK1fCd99B1ap2R6SUY2ki8GEBWx84dQq6drXOFBo/3loWUkrZJsBGmMARsP2FTpywWkovXw6TJsHDF1xjqJQqYToj8FEBWR9IToY77oC1a2HyZLjvPrsjUkqhicBnBVx94OhRaybwxx8wdSp062Z3REopF00EPiqg6gNHjlgtpbdutdpHdO1qd0RKqVwCYJQJPAFVHzh8GG67zbpe4McfNQko5YN0RuCDAqY+cPAg3H477NljXTV8++12R6SUckMTgQ8KiPpAYqI1E0hIgJ9/hnbt7I5IKVUATQQ+yO/rAwcOWEng4EGYMwfatrU7IqVUIfx0pAlcfl8f2LfP+vR/+DDMm6dJQCk/oDMCH+PX9YHdu62ZwIkTMH8+tGpld0RKKQ9oIvAxflsf2LnTSgKnT8PChdC8ud0RKaU8pInAx/hlfWD7drj1Vjh3DhYtsvYcVkr5DT8abQKfX9YHtmyxagLp6ZoElPJTmgh8iN/VBzZuhPbtwRhYvBiuv97uiJRSl0ATgQ/xq/rA+vXWclCpUvDrrxAdbXdESqlLpInAh/hNfWDNGqswHBpqJYHrrrM7IqXUZfDxEcc5/KY+sHq11SqiXDkrCdSvb3dESqnLpInAR/hFfWDFCquLaESElQTq1bM7IqVUMdBE4CN8vj7w++/WfgKVK1tJICrK7oiUUsVEE4GP8On6wJIl1s5i1apZSaBWLbsjUkoVIx8cdZzHp+sDv/wCXbpA7drWKaI1atgdkVKqmGki8AE+Wx+YPx/uvNOqBSxaZM0IlFIBx6uJQEQ6i8g2EdkpIq8UcEx7EVknIptE5FdvxuOrfLI+8PPPcPfdcO211qygShW7I1JKeYnXeg2JSDDwGdARSABWi8iPxpjNuY65Evgc6GyM2S8iV3krHl/mc/WBWbOgRw9o1MiaFVTysZmKUqpYeXPkaQXsNMbsNsakAZOA7vmOeRiYZozZD2CMOezFeHySz9UHpk+H+++3egYtXKhJQCkH8GYiqAEcyHU7wXVfbtcCESKyWETiReQxL8bjk3yqPjBlCjz4ILRoYc0EIiLsjkgpVQK82YZa3Nxn3Pz8FsDtQBlguYisMMZsz/NCIv2B/gC1a9f2Qqj28Zn6wPffwyOPQJs2MHs2hPtQvUIp5VXenBEkALlPOK8JJLo5Zo4x5owx5giwBLigj7ExZrQxJtYYE1u5cmWvBWwHn6gPTJgADz9sbSs5Z44mAaUcxpujz2rgGhGpKyKlgV7Aj/mOmQncLCKlRCQMaA1s8WJMPsUn6gNffw2PPWa1k/7vf60eQkopR/Ha0pAxJkNEBgFzgWBgrDFmk4gMcD0+yhizRUTmABuALGCMMWajt2LyNbbXB8aMgf79oWNHmDEDypSxJw6llK28ulWlMWY2MDvffaPy3f4A+MCbcfgq2+oDWVnw4Yfw0kvWVcPTplktpZVSjqR7FtvIlvrAzp3wxBPw22/WaaITJ8IVV5Tcz1dK+RwfuYLJeUq8PpCVBZ98Ak2awIYNVm3ghx80CSildEZglxKtD+SeBXTtCqNHa/M4pVQOnRHYpETqA1lZ8PHHeWcBP/2kSUAplYfOCGzi9fpA7lnAnXfCF19oAlBKuaUzAht4tT7gbhYwa5YmAaVUgXRGYAOv1Qd27LBmAUuX6ixAKeUxnRHYoNjrA1lZMGKE1TF040b45hudBSilPKYzAhsUa30g/yxg9GioXv3yX1cp5Rg6IyhhxVYfyMyEjz6yagEbN8K4cdYsQJOAUqqIdEZQwlbsLob6wPbt1izg99/hrrusWoAmAKXUJdIZQQk6lZrOiAXbiSx3xaXVB7JnAU2bwqZN1izgxx81CSilLovOCEpIZpbhuUnr2Hc0hQl9Wxe9PrB9O/TpA8uWWZvKf/EFVKvmnWCVUo6iM4ISMnz+NhZuPcwbd0cXbVkoMxOGD7dmAVu2wPjxMHOmJgGlVLHRGUEJmLU+kc8W7eKhVrXo3aaO50/UWYBSqgTojMDLNiWe4MUp64mtE8Gb3Roj4m4r53x0FqCUKkE6I/Cio6fP0X9cPBFhpRnZuwWlS3mQd7dts84IWrYMunWDUaM0ASilvEpnBF6SnpnFwIlrOHL6HF882oLK5S/S9z8z09o1LCbGmgVMmGBtH6lJQCnlZToj8JJ/z9rMqj3HGNEzhiY1ryz84G3brFrA8uU6C1BKlTidEXjBtyv3M37FPp66pR73NCuk309mJgwbZs0Ctm7VWYBSyhY6Iyhmq/ce440fN9Lu2sq81LlBwQdu3WrNAlasgO7drVlA1aolF6hSSrnojKAYJR4/y8AJ8dSMCOOTXs0IDnJzhtC5c+dnAdu3W5vHT5+uSUApZRudERSTs2mZ9B8fR2p6FpP6t6BCaDDs3g1//JH3a/t2a0lIZwFKKR+hiaAYmMOHeeXbNWw6mMGXJ1dQv9sbVkfQM2fOH1S3LjRuDPfeCzffDHfcAZ5cU6CUUl6miaAoUlKsZm8bN+b5lD+6zk3MvPUJXvx1HB12/ALXX29dC3D99dZXo0ZQvrzd0SullFuaCNzJzLQ2f8+/rLNrFxhjHRMaCo0asbj747wX0ZY7rwri6WkfW0s9+klflZD09HQSEhJITU21OxTlI0JDQ6lZsyYhISEeP8fZicAY+OuvCwf8LVsg+z9WUBDUr2+1e+jd+/yn/Hr12H3sLIM/+50GEWF8MPAGpLSz305V8hISEihfvjxRUVGetS9RAc0Yw9GjR0lISKBu3boeP885I9eZM7B+/QXLOhw7dv6YatWsQf6226w/GzeG6GgoU+aClzuZmk6/cXGEBAcx+tEWhGkSUDZITU3VJKByiAiVKlUiKSmpSM9zzug1Y4b1iR6s9frGjaFHD+vP7E/5lTxrD51/b4FaFcO8F7dSF6FJQOV2Kf8evJoIRKQz8DEQDIwxxryX7/H2wExgj+uuacaYf3slmNtvt3bzuv56qFPnstbxh8/fxi9bD/NW90aXv/ewUkrZzGsXlIlIMPAZ0AWIBh4SkWg3h/5mjIlxfXknCYBVxL37boiKuqwkcMl7CygVgI4fP87nn39+Sc/t2rUrx48fL/SYf/3rXyxYsOCSXl95zptXFrcCdhpjdhtj0oBJQHcv/jyvu6S9BZQKYIUlgszMzEKfO3v2bK688spCj/n3v/9Nhw4dLjU8W2RkZNgdQpF5c2moBnAg1+0EoLWb424QkfVAIvCCMWZT/gNEpD/QH6B27dpeCPXiLmlvAaVK0nPPwbp1xfuaMTEwYkSBD7/yyivs2rWLmJgYOnbsyJ133smbb75JtWrVWLduHZs3b+aee+7hwIEDpKam8uyzz9K/f38AoqKiiIuL4/Tp03Tp0oW2bduybNkyatSowcyZMylTpgyPP/44d911Fz169CAqKoq//e1vzJo1i/T0dH744QcaNGhAUlISDz/8MEePHqVly5bMmTOH+Ph4IiMj88Q6cOBAVq9ezdmzZ+nRowdvvvkmAKtXr+bZZ5/lzJkzXHHFFSxcuJCwsDBefvll5s6di4jQr18/Bg8enBNzZGQkcXFxvPDCCyxevJihQ4eSmJjI3r17iYyM5N133+XRRx/ljOui0k8//ZQbb7wRgPfff5/x48cTFBREly5d6NevHw888ABr1qwBYMeOHfTq1Yv4+Pji/bsshDcTgbuPyybf7TVAHWPMaRHpCswArrngScaMBkYDxMbG5n8Nr8u9t8APA264+N4CSjnEe++9x8aNG1nnSkCLFy9m1apVbNy4Mef0xbFjx1KxYkXOnj1Ly5Ytuf/++6mU78SMHTt28N133/Hll1/y4IMPMnXqVHpnn9yRS2RkJGvWrOHzzz9n2LBhjBkzhjfffJPbbruNf/zjH8yZM4fRo0e7jfWdd96hYsWKZGZmcvvtt7NhwwYaNGhAz549+f7772nZsiUnT56kTJkyjB49mj179rB27VpKlSrFsdxnFxYgPj6epUuXUqZMGVJSUpg/fz6hoaHs2LGDhx56iLi4OH7++WdmzJjBypUrCQsL49ixY1SsWJEKFSqwbt06YmJi+Oqrr3j88ceL9hdxmbyZCBKAWrlu18T61J/DGHMy1/ezReRzEYk0xhzxYlxFVqS9BZSySyGf3EtSq1at8pzD/sknnzB9+nQADhw4wI4dOy5IBHXr1iUmJgaAFi1asHfvXrevfd999+UcM23aNACWLl2a8/qdO3cmIiLC7XMnT57M6NGjycjI4K+//mLz5s2ICNWqVaNly5YAhIeHA7BgwQIGDBhAqVLWEFmxYsWL/t7dunWjjOtU8/T0dAYNGsS6desIDg5m+/btOa/bp08fwsLC8rxu3759+eqrrxg+fDjff/89q1atuujPK07eTASrgWtEpC7wJ9ALeDj3ASJSFThkjDEi0gqrZnHUizEVmcd7CyilAChbtmzO94sXL2bBggUsX76csLAw2rdv7/Yq6CuuOD/LDg4O5uzZs25fO/u44ODgnLV4Yy6+SLBnzx6GDRvG6tWriYiI4PHHHyc1NRVjjNtaX0H3lypViqysLIALfo/cv/dHH31ElSpVWL9+PVlZWYSGhhb6uvfff3/OzKZFixYXJEpv89pCtzEmAxgEzAW2AJONMZtEZICIDHAd1gPY6KoRfAL0Mp78rZYQj/cWUMqhypcvz6lTpwp8/MSJE0RERBAWFsbWrVtZsWJFscfQtm1bJk+eDMC8efNITk6+4JiTJ09StmxZKlSowKFDh/j5558BaNCgAYmJiaxevRqAU6dOkZGRQadOnRg1alROssleGoqKispZu586dWqBMZ04cYJq1aoRFBTE+PHjcwrnnTp1YuzYsaSkpOR53dDQUO644w4GDhxInz59Lvs9KSqvVjyNMbONMdcaY642xrzjum+UMWaU6/tPjTGNjDFNjTFtjDHLvBlPUeTZW+ChAvYWUMrhKlWqxE033UTjxo158cUXL3i8c+fOZGRk0KRJE15//XXatGlT7DG88cYbzJs3j+bNm/Pzzz9TrVo1yudr8ti0aVOaNWtGo0aNeOKJJ7jpppsAKF26NN9//z2DBw+madOmdOzYkdTUVPr27Uvt2rVp0qQJTZs25dtvv835Wc8++yw333wzwcHBBcb09NNP880339CmTRu2b9+eM1vo3Lkz3bp1IzY2lpiYGIYNG5bznEceeQQRoVOnTsX9Fl2U+NAHcI/ExsaauLg4r/6Ms2mZPPDFMvYdSWH6MzdR/6pyXv15Sl2qLVu20LBhQ7vDsNW5c+cIDg6mVKlSLF++nIEDB+YUr/3JsGHDOHHiBG+99dZlv5a7fxciEm+MiXV3vHNaTHjIGMMr0zawKfEkYx6L1SSglI/bv38/Dz74IFlZWZQuXZovv/zS7pCK7N5772XXrl388ssvtvx8TQT5jF6ym5nrEnnxjuu4vWEVu8NRSl3ENddcw9q1a+0O47Jkn/VkF70qKpfF2w7z3pyt3NWkGk+3v9rucJRSqkRoInDZnXSawd+tpWHVcN7v0UTbRyilHEMTAdbeAn2z9xZ4TPcWUEo5i+MTQfbeAvuPpvD5I82pGaF7CyilnMXxieDDedbeAm90070FlCoJ5cpZZ+IlJibSo0cPt8e0b9+ei50mPmLEiJwLs8CzttbKPUcnglnrE/l88S4ealWb3q3t6WqqlFNVr16dKVOmXPLz8ycCT9pa+xJjTE67Crs5djF845+59xZopMVh5ffenLWJzYknL35gEURXD+eNuxsV+PjLL79MnTp1ePrppwEYOnQo5cuX56mnnqJ79+4kJyeTnp7O22+/Tffuebcj2bt3L3fddRcbN27k7Nmz9OnTh82bN9OwYcM8vYbctY/+5JNPSExM5NZbbyUyMpJFixblaRE9fPhwxo4dC1gN3Z577jn27t1bYLvr3GbNmsXbb79NWloalSpVYuLEiVSpUoXTp08zePBg4uLiEBHeeOMN7r//fubMmcOrr75KZmYmkZGRLFy4kKFDh1KuXDleeOEFABo3bsxPP/0EQJcuXbj11ltZvnw5M2bM4L333vO4PXbXrl35z3/+k9Og76abbmLkyJE0adLkMv6WHZoIjpw+x1PjdW8BpS5Xr169eO6553ISweTJk5kzZw6hoaFMnz6d8PBwjhw5Qps2bejWrVuBH7hGjhxJWFgYGzZsYMOGDTRv3jznMXfto4cMGcLw4cNZtGjRBfsOxMfH89VXX7Fy5UqMMbRu3Zp27doRERHhUbvrtm3bsmLFCkSEMWPG8P777/Phhx/y1ltvUaFCBf744w8AkpOTSUpKol+/fixZsoS6det61K5627ZtfPXVVzkb+hSlPXbfvn35+uuvGTFiBNu3b+fcuXOXnQTAgYkgLSOLpydYewtMGXCj7i2gAkZhn9y9pVmzZhw+fJjExESSkpKIiIigdu3apKen8+qrr7JkyRKCgoL4888/OXToEFWrVnX7OkuWLGHIkCEANGnSJM/g5q59dGGD39KlS7n33ntz+vvcd999/Pbbb3Tr1s2jdtcJCQn07NmTv/76i7S0tJyW2gsWLGDSpEk5x0VERDBr1ixuueWWnGM8aVddp06dPD2XitIe+4EHHuCtt97igw8+YOzYscW2b4HjEsG/f9rEqr3H+LhXDNfXrGB3OEr5vR49ejBlyhQOHjxIr169AJg4cSJJSUnEx8cTEhJCVFSU2/bTubmbLRTUProwhfVP86Td9eDBg3n++efp1q1bzu5j2a+bP0ZP2lVD3pbVudtVF7U9dlhYGB07dmTmzJlMnjz5ogV1TzlqTWTiyn1MWLGfp9rVo3uM7i2gVHHo1asXkyZNYsqUKTlnAZ04cYKrrrqKkJAQFi1axL59+wp9jVtuuYWJEycCsHHjRjZs2AAU3D4aCm6BfcsttzBjxgxSUlI4c+YM06dP5+abb/b49zlx4gQ1aljjwzfffJNzf6dOnfj0009zbicnJ3PDDTfw66+/smfPHiBvu+rsrSfXrFmT83h+RW2PDVbNY8iQIbRs2dKjGYgnHJMIVu05xhszN1l7C9yhewsoVVwaNWrEqVOnqFGjBtWqVQOslspxcXHExsYyceJEGjQo/P/cwIEDOX36NE2aNOH999+nVatWQMHtowH69++fU3jNrXnz5jz++OO0atWK1q1b07dvX5o1a+bx7zN06FAeeOABbr755jz1h9dee43k5GQaN25M06ZNWbRoEZUrV2b06NHcd999NG3alJ49ewLWRjPHjh0jJiaGkSNHcu2117r9WUVtjw3WklZ4eHix7lvgmDbUmxNP8v9+3sKnDzenQpkQL0SmVMnTNtTOk5iYSPv27dm6dStBQe4/yxe1DbVjZgTR1cMZ/2RrTQJKKb81btw4WrduzTvvvFNgErgUjisWK6WUv3rsscd47LHHiv11HTMjUCpQ+dvyrvKuS/n3oIlAKT8WGhrK0aNHNRkowEoCR48eJTQ0tEjP06UhpfxYzZo1SUhIICkpye5QlI8IDQ2lZs2aRXqOJgKl/FhISEjOVa1KXSpdGlJKKYfTRKCUUg6niUAppRzO764sFpEkoPDGJQWLBI4UYzj+Tt+PvPT9OE/fi7wC4f2oY4yp7O4Bv0sEl0NE4gq6xNqJ9P3IS9+P8/S9yCvQ3w9dGlJKKYfTRKCUUg7ntEQw2u4AfIy+H3np+3Gevhd5BfT74agagVJKqQs5bUaglFIqH00ESinlcI5JBCLSWUS2ichOEXnF7njsJCK1RGSRiGwRkU0i8qzdMdlNRIJFZK2I/GR3LHYTkStFZIqIbHX9G7nB7pjsIiL/4/o/slFEvhORorX19BOOSAQiEgx8BnQBooGHRCTa3qhslQH83RjTEGgDPOPw9wPgWWCL3UH4iI+BOcaYBkBTHPq+iEgNYAgQa4xpDAQDveyNyjsckQiAVsBOY8xuY0waMAnobnNMtjHG/GWMWeP6/hTWf/Qa9kZlHxGpCdwJjLE7FruJSDhwC/B/AMaYNGPMcVuDslcpoIyIlALCgESb4/EKpySCGsCBXLcTcPDAl5uIRAHNgJU2h2KnEcBLQJbNcfiCekAS8JVrqWyMiJS1Oyg7GGP+BIYB+4G/gBPGmHn2RuUdTkkE4uY+x583KyLlgKnAc8aYk3bHYwcRuQs4bIyJtzsWH1EKaA6MNMY0A84AjqypiUgE1spBXaA6UFZEetsblXc4JREkALVy3a5JgE7xPCUiIVhJYKIxZprd8djoJqCbiOzFWjK8TUQm2BuSrRKABGNM9gxxClZicKIOwB5jTJIxJh2YBtxoc0xe4ZREsBq4RkTqikhprILPjzbHZBsREaw14C3GmOF2x2MnY8w/jDE1jTFRWP8ufjHGBOSnPk8YYw4CB0TkOtddtwObbQzJTvuBNiIS5vo/czsBWjh3xFaVxpgMERkEzMWq/I81xmyyOSw73QQ8CvwhIutc971qjJltX0jKhwwGJro+NO0G+tgcjy2MMStFZAqwButMu7UEaKsJbTGhlFIO55SlIaWUUgXQRKCUUg6niUAppRxOE4FSSjmcJgKllHI4TQTKsUTk/4lIexG5x66OtCKyWEQCdlN05R80ESgna43VY6kd8JvNsShlG00EynFE5AMR2QC0BJYDfYGRIvIvN8dWFpGpIrLa9XWT6/6hIjJeRH4RkR0i0s91v7hef6OI/CEiPXO91kuu+9aLyHu5fswDIrJKRLaLyM2uYxu57lsnIhtE5BovviXK4RxxZbFSuRljXhSRH7Curn4eWGyMuamAwz8GPjLGLBWR2lhXpzd0PdYEaz+HssBaEfkvcAMQg9XHPxJYLSJLXPfdA7Q2xqSISMVcP6OUMaaViHQF3sDqcTMA+NgYk32Fb3Dx/PZKXUgTgXKqZsA6oAGF99LpAERbrWYACBeR8q7vZxpjzgJnRWQR1r4XbYHvjDGZwCER+RVr5tEO+MoYkwJgjDmW62dkN/2LB6Jc3y8H/unaK2GaMWbHpf6iSl2MJgLlKCISA3yN1YH2CNZmI+LquXSDa2DPLcjd/a7EkL8/i8F9y3Nc9xfUz+Wc689MXP8njTHfishKrA1z5opIX2PML4X9bkpdKq0RKEcxxqwzxsQA27G2Lf0FuMMYE+MmCQDMAwZl33AlkmzdRSRURCoB7bG63C4Berr2QK6MtdvXKtfrPCEiYa7Xyb00dAERqQfsNsZ8gtUpt8kl/LpKeUQTgXIc1wCdbIzJAhoYYwpbGhoCxLoKtpux1u6zrQL+C6wA3jLGJALTgQ3Aeqwk85Ix5qAxZg7WgB7nmn28cJEwewIbXcc2AMYV8ddUymPafVSpSyAiQ4HTxphhdsei1OXSGYFSSjmczgiUUsrhdEaglFIOp4lAKaUcThOBUko5nCYCpZRyOE0ESinlcP8fByRCRUDuh/8AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(hist.history['accuracy'],'r',label='training accuracy')\n",
    "plt.plot(hist.history['val_accuracy'],label='validation accuracy')\n",
    "plt.xlabel('# epochs')\n",
    "plt.ylabel('loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "40/40 [==============================] - 2s 41ms/step\n",
      "[0.16494626998901368, 0.925000011920929]\n"
     ]
    }
   ],
   "source": [
    "print(model.evaluate(test_data,test_target))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred=model.predict(test_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(40, 224, 224, 3)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred=y_pred.reshape(-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_pred=[]\n",
    "for i in range(y_pred.shape[0]):\n",
    "    if y_pred[i]<=0.5:\n",
    "        final_pred.append(0)\n",
    "    else:\n",
    "        final_pred.append(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_pred=np.array(final_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "import itertools"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "cm = confusion_matrix(test_target, final_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_confusion_matrix(cm, classes,\n",
    "                        normalize=False,\n",
    "                        title='Confusion matrix',\n",
    "                        cmap=plt.cm.Blues):\n",
    "    \"\"\"\n",
    "    This function prints and plots the confusion matrix.\n",
    "    Normalization can be applied by setting `normalize=True`.\n",
    "    \"\"\"\n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(classes))\n",
    "    plt.xticks(tick_marks, classes, rotation=45)\n",
    "    plt.yticks(tick_marks, classes)\n",
    "\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "        print(\"Normalized confusion matrix\")\n",
    "    else:\n",
    "        print('Confusion matrix, without normalization')\n",
    "\n",
    "    print(cm)\n",
    "\n",
    "    thresh = cm.max() / 2.\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(j, i, cm[i, j],\n",
    "            horizontalalignment=\"center\",\n",
    "            color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.ylabel('True label')\n",
    "    plt.xlabel('Predicted label')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "cm_plot_labels = ['covid_positive','covid_negative']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix, without normalization\n",
      "[[24  1]\n",
      " [ 2 13]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_confusion_matrix(cm=cm, classes=cm_plot_labels, title='Confusion Matrix')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "     class 0       0.92      0.96      0.94        25\n",
      "     class 1       0.93      0.87      0.90        15\n",
      "\n",
      "    accuracy                           0.93        40\n",
      "   macro avg       0.93      0.91      0.92        40\n",
      "weighted avg       0.93      0.93      0.92        40\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "target_names=['class 0','class 1']\n",
    "print(classification_report(test_target,final_pred,target_names=target_names))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_roc_cur(fper, tper):  \n",
    "    plt.plot(fper, tper, color='orange', label='ROC')\n",
    "    plt.plot([0, 1], [0, 1], color='darkblue', linestyle='--')\n",
    "    plt.xlabel('False Positive Rate')\n",
    "    plt.ylabel('True Positive Rate')\n",
    "    plt.title('Receiver Operating Characteristic (ROC) Curve')\n",
    "    plt.legend()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve\n",
    "fper, tper, threshold = roc_curve(test_target, y_pred)\n",
    "plot_roc_cur(fper, tper)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9786666666666667"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score\n",
    "roc_auc_score(test_target,y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.6.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}