VGG11-random-rotation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "r6XcGTCxbZ-D"
   },
   "outputs": [],
   "source": [
    "import tensorflow\n",
    "from keras.optimizers import SGD\n",
    "from keras.preprocessing.image import ImageDataGenerator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "kdq9JxH0kYqY"
   },
   "outputs": [],
   "source": [
    "from __future__ import print_function\n",
    "import keras\n",
    "import numpy\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 PIL import Image\n",
    "import cv2\n",
    "batch_size = 128\n",
    "num_classes = 10\n",
    "epochs = 12\n",
    "\n",
    "# input image dimensions\n",
    "img_rows, img_cols = 32, 32\n",
    "\n",
    "# the data, split between train and test sets\n",
    "(x_train, y_train), (x_test, y_test) = mnist.load_data()\n",
    "X_train = []\n",
    "X_test = []\n",
    "for i in range(len(x_train)):\n",
    "  X_train.append(cv2.resize(x_train[i], (32,32)))\n",
    "\n",
    "for i in range(len(x_test)):\n",
    "  X_test.append(cv2.resize(x_test[i], (32,32)))\n",
    "\n",
    "X_train = numpy.asarray(X_train) \n",
    "X_test = numpy.asarray(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "H_6rbCJ8k9po"
   },
   "outputs": [],
   "source": [
    "if K.image_data_format() == 'channels_first':\n",
    "    x_train = X_train.reshape(X_train.shape[0], 1, img_rows, img_cols)\n",
    "    x_test = X_test.reshape(X_test.shape[0], 1, img_rows, img_cols)\n",
    "    input_shape = (1, img_rows, img_cols)\n",
    "else:\n",
    "    x_train = X_train.reshape(X_train.shape[0], img_rows, img_cols, 1)\n",
    "    x_test = X_test.reshape(X_test.shape[0], img_rows, img_cols, 1)\n",
    "    input_shape = (img_rows, img_cols, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 67
    },
    "colab_type": "code",
    "id": "01NUaGoQkzYf",
    "outputId": "dd69579a-d4b6-4c64-d50a-0fac76f59672"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_train shape: (60000, 32, 32, 1)\n",
      "60000 train samples\n",
      "10000 test samples\n"
     ]
    }
   ],
   "source": [
    "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')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "PElbQTzclZWT"
   },
   "outputs": [],
   "source": [
    "y_train = keras.utils.to_categorical(y_train, num_classes)\n",
    "y_test = keras.utils.to_categorical(y_test, num_classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 753
    },
    "colab_type": "code",
    "id": "Df3WgLBZleg-",
    "outputId": "db75b2c7-e111-4341-f4a1-b1a9281099c7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_1 (Conv2D)            (None, 32, 32, 64)        640       \n",
      "_________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2 (None, 16, 16, 64)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 16, 16, 128)       73856     \n",
      "_________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2 (None, 8, 8, 128)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 8, 8, 256)         295168    \n",
      "_________________________________________________________________\n",
      "conv2d_4 (Conv2D)            (None, 8, 8, 256)         590080    \n",
      "_________________________________________________________________\n",
      "max_pooling2d_3 (MaxPooling2 (None, 4, 4, 256)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_5 (Conv2D)            (None, 4, 4, 512)         1180160   \n",
      "_________________________________________________________________\n",
      "conv2d_6 (Conv2D)            (None, 4, 4, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "max_pooling2d_4 (MaxPooling2 (None, 2, 2, 512)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_7 (Conv2D)            (None, 2, 2, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "conv2d_8 (Conv2D)            (None, 2, 2, 512)         2359808   \n",
      "_________________________________________________________________\n",
      "max_pooling2d_5 (MaxPooling2 (None, 1, 1, 512)         0         \n",
      "_________________________________________________________________\n",
      "flatten_1 (Flatten)          (None, 512)               0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 4096)              2101248   \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 4096)              16781312  \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 512)               2097664   \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 10)                5130      \n",
      "=================================================================\n",
      "Total params: 30,204,682\n",
      "Trainable params: 30,204,682\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "model.add(Conv2D(64, (3, 3), activation='relu', input_shape=input_shape, padding='same'))\n",
    "model.add(MaxPooling2D(pool_size=(2, 2), strides = (2, 2)))\n",
    "model.add(Conv2D(128, (3, 3), activation='relu', padding='same'))\n",
    "model.add(MaxPooling2D(pool_size=(2, 2), strides = (2, 2)))\n",
    "model.add(Conv2D(256, (3, 3), activation='relu', padding='same'))\n",
    "model.add(Conv2D(256, (3, 3), activation='relu', padding='same'))\n",
    "model.add(MaxPooling2D(pool_size=(2, 2), strides = (2, 2)))\n",
    "model.add(Conv2D(512, (3, 3), activation='relu', padding='same'))\n",
    "model.add(Conv2D(512, (3, 3), activation='relu', padding='same'))\n",
    "model.add(MaxPooling2D(pool_size=(2, 2), strides = (2, 2)))\n",
    "model.add(Conv2D(512, (3, 3), activation='relu', padding='same'))\n",
    "model.add(Conv2D(512, (3, 3), activation='relu', padding='same'))\n",
    "model.add(MaxPooling2D(pool_size=(2, 2), strides = (2, 2))) \n",
    "model.add(Flatten()) \n",
    "model.add(Dense(4096, activation='relu'))\n",
    "model.add(Dense(4096, activation='relu'))\n",
    "model.add(Dense(512, activation='relu'))\n",
    "model.add(Dense(10, activation='softmax'))\n",
    "print(model.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "jo7DZFzzmytr"
   },
   "outputs": [],
   "source": [
    "sgd = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)\n",
    "model.compile(loss='categorical_crossentropy', optimizer=sgd, metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ePyraNbWm3_3"
   },
   "outputs": [],
   "source": [
    "datagen = ImageDataGenerator(rotation_range = 40) #applying random rotations to the data set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "el-Xi8dynLNj"
   },
   "outputs": [],
   "source": [
    "validation_generator = ImageDataGenerator()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "ZpcdM1FfnV05"
   },
   "outputs": [],
   "source": [
    "validation_generator.fit(x_test)\n",
    "\n",
    "datagen.fit(x_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 184
    },
    "colab_type": "code",
    "id": "chjc_wgVnXS-",
    "outputId": "7766d621-463b-4021-9536-1f08af0343f8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "1875/1875 [==============================] - 129s 69ms/step - loss: 0.8379 - acc: 0.7005 - val_loss: 0.1132 - val_acc: 0.9651\n",
      "Epoch 2/5\n",
      "1875/1875 [==============================] - 126s 67ms/step - loss: 0.1005 - acc: 0.9700 - val_loss: 0.0748 - val_acc: 0.9763\n",
      "Epoch 3/5\n",
      "1875/1875 [==============================] - 126s 67ms/step - loss: 0.0720 - acc: 0.9785 - val_loss: 0.0472 - val_acc: 0.9858\n",
      "Epoch 4/5\n",
      "1875/1875 [==============================] - 126s 67ms/step - loss: 0.0559 - acc: 0.9830 - val_loss: 0.0398 - val_acc: 0.9882\n",
      "Epoch 5/5\n",
      "1875/1875 [==============================] - 127s 68ms/step - loss: 0.0477 - acc: 0.9861 - val_loss: 0.0321 - val_acc: 0.9906\n"
     ]
    }
   ],
   "source": [
    "history = model.fit_generator(datagen.flow(x_train, y_train, batch_size=32),steps_per_epoch = len(x_train) / 32, epochs = 5, validation_data = validation_generator.flow(x_test, y_test, batch_size = 32),validation_steps = 1000)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 376
    },
    "colab_type": "code",
    "id": "b7ZcNXbonljq",
    "outputId": "b8863835-3541-480f-da3b-a017e8b96cff"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfUAAAFnCAYAAAC/5tBZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nERFpIRTqIiIiLYRCXUREpIX4/zHdtYDPUl7xAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2fb4e8aa90>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "# summarize history for accuracy\n",
    "plt.plot(history.history['acc'])\n",
    "plt.title('Train_Accuracy vs Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 376
    },
    "colab_type": "code",
    "id": "53tubbZOs7TF",
    "outputId": "b9964a19-f05d-4882-f1de-faa9d9c0a5fa"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfsAAAFnCAYAAAChL+DqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n2IuIiDg5hb2IiIiTU9iLiIg4uf8PxlFmZXwReHUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2fb4671c50>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['val_acc'])\n",
    "plt.title('Test(Validation)_Accuracy vs Epochs')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('Epochs')\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 376
    },
    "colab_type": "code",
    "id": "fdpZ5HDus9NJ",
    "outputId": "b6597868-a4e6-4322-c6ed-560d0a4ffecb"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2fb45c5198>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['loss'])\n",
    "plt.title('Train_loss vs Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 376
    },
    "colab_type": "code",
    "id": "aD3Io7khs-5W",
    "outputId": "6297ac9e-5943-4439-b9ff-41436a392b57"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfUAAAFnCAYAAAC/5tBZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzs3XlYVPXiP/D3mWFnYGCAQRRBxAUF\nzO1WRokLiKK5UCmWXr3Zqv7qpnVL7jW7N7Xyki1m2eLNvllJmms3MzO1VG5KGAiICCqyMyPrsA9z\nfn+go6QiKsNhZt6v5/F5mvkwM+8zp4c3Z/scQRRFEURERGT2ZFIHICIioo7BUiciIrIQLHUiIiIL\nwVInIiKyECx1IiIiC8FSJyIishAsdbJoy5Ytw/jx4zF+/HgEBwdj9OjRxsc6ne6W3vPrr79u9biw\nsBATJkzA2rVrERsbe9XPV1ZWYtCgQcjLy7vue+bm5iI0NBQA8Nlnn2HNmjXX/LkxY8YgKSmpzXz1\n9fXYsWOHMdvkyZPb/Pm2fPHFF1iyZEm7fvbKZTAHubm5CAoKMv7/cOU/rVbboZ+l1+vRv39/aDSa\nDn1foj+ykToAkSn985//NP73mDFjsGrVKgwfPvyW36+pqQnx8fGYPn268bm///3veOaZZxAWFoaP\nPvoIeXl56Nmzp3F89+7duOOOO1o915Y5c+bccj4ASEtLw86dOzFlyhR0794dO3fuvOX3evjhh/Hw\nww/jwIEDGDVq1G3l6opsbW3x/fffSx2DqMNwS52sWmFhIZ544glERUUhKioKv/zyC4CWLaslS5Zg\n/PjxiIiIwDPPPIOamhrMnTsXVVVVGD9+PAoLC5GcnIzCwkJERUXB1dUVY8aMMW4lX7Jz505MmzYN\nAJCTk4PY2FhMmDAB48aNw3fffXdVprfeegsvv/wyACA1NRXR0dGIiorCG2+80ernNm3ahPHjxyMy\nMhKzZ89GUVERSkpK8Ne//hXJycmYPXt2q63n5uZmvPnmm8at0bi4ONTV1QEAZs6ciQ0bNiA2Nhb3\n3Xcfnn/+eQCAIAh4/PHHsXbt2tv6nsvLy/HMM88gKioK0dHRWL9+vXHszTffNH7/c+fORWlpaZvP\nX7Jv3z5MnTq11XOTJk3CkSNHkJiYiKlTpyI6OhoTJkzADz/8cNOZN2/ejKeeegqLFy/G2LFjcf/9\n9+P8+fM3XJ6DBw8a19lTTz2FyspK49j+/fsxbdo03HvvvdiwYQMAQKfT4emnn8aECRMQERGBl19+\nGXq9/qbzEgEARCIrMXr0aPHYsWOtnnvkkUfENWvWiKIoimfOnBHvvPNOsaKiQty7d6/46KOPigaD\nQTQYDOKbb74pHj58WDx37pwYEhJifP3KlSvF119/3fj44MGD4rhx44yP8/LyxCFDhog6nU4URVGc\nN2+e+Mknn4iiKIpHjhwRBw8eLOr1+lbvu3r1anHp0qWiKIri1KlTxc2bN4uiKIq7du0Sg4KCxGPH\njonFxcViaGioWFxcLIqiKL7wwgvG13z99dfio48+Koqi2Op9t2/fLsbExIi1tbWiXq8Xn3zySfHD\nDz8URVEUY2NjxTlz5oj19fVidXW1eOedd4q///67KIqi2NDQIN5xxx1iQUFBm9/vH7+bK8XFxYmv\nvPKKKIqiWFZWJo4cOVI8fvy4ePLkSXHChAliU1OTKIqi+Omnn4o7duy47vNXamhoEIcOHSoWFhaK\noiiKZ8+eFe+++25Rr9eLU6ZMEZOSkkRRFMWcnBxx8eLFN5X30vcYHBwspqamiqIoiv/+97/F//f/\n/l+by1NdXS3+6U9/ErOzs0VRFMV//vOf4quvvio2NTWJ/fr1E9966y1RFEXx+PHj4h133CHq9Xpx\nw4YN4t///ndRFEWxsbFRXLp0qZiZmdnmd010PdxSJ6tVXV2NpKQkzJ07FwAQEBCAwYMH4+eff4ZK\npcKpU6ewb98+1NXVYdGiRbjnnnuueo/U1NRWx5HDwsJQW1uLlJQUAC1b6ePGjYOzszMA4KOPPjJ+\n3vDhw1FbW3vd47c1NTXIyMhAdHQ0AGDixImws7MDAHh7eyMpKQne3t7G98rPz29zeQ8ePIiYmBg4\nOjpCLpdj2rRpOHTokHF8/PjxsLe3h0KhQK9evVBYWAgAsLOzw4ABA3D8+PE23/9Gn/3www8DANzd\n3REREYFDhw7B1dUVGo0G3377LaqqqjB37lxMnjz5us9fyc7ODqNGjcJPP/0EANi7dy8iIyMhl8vh\n4eGBbdu24cyZM+jduzfi4+Ovmaupqemq4+mLFy82jvfr18+4fseNG2f8Dq63PL/99ht8fX0RGBgI\nAHjppZfwt7/9zfh+U6ZMAQAMHDgQdXV1qKyshIeHB5KTk3H48GGIooh//etf6N+//y1/12TdWOpk\ntaqrqyGKIh588EHjL/STJ0+iqqoKQ4cORVxcHDZs2ICwsDA8//zzqK6uvuo9ysrKoFKpjI/lcjmm\nTJliPI69a9cu4653oKUMHnnkEURFReH+++8HAIjXuf1CZWUl5HI5nJycALTsCndxcQHQcnjg7bff\nxoQJExAVFYV3330XBoOhzeUtKyuDq6ur8bGrqyvKysqMjy+9NwDIZLJW7+fh4YELFy60+f638tnd\nu3fHu+++i2+//Rbh4eF46qmnUFJSct3n/ygqKspY6vv27TP+AfT666/Dzs4Oc+bMQVRUFPbu3XvN\nXJeOqV/578033zSOK5XKVv9dVVXV5vKUl5e3et7Ozs74hxgAKBQKAC3fL9BySGTSpEmYNWsW3nrr\nLYwYMQLLly9HY2NjO79ZotZY6mS1PD09IZPJsH37duMv9J9//hmPPPIIACA6OhobN27ETz/9BJ1O\nh08//fSq97hWIU+bNg27d+9GWloaGhsbceeddwIAGhsb8eyzz2LBggXYs2cPtm/f3mY+V1dXNDc3\no7a2FkBLAVwqlW+//Ra//PILvvzyS+zZswfz58+/4fJ6eHigoqLC+LiiogIeHh43fF1HaOuzR4wY\ngU8++QSHDh2Cl5cXVq9e3ebzVxo5ciROnDiBc+fOIT8/3/hde3l54eWXX8Yvv/yCuLg4/O1vfzOe\nP3AzysvLjf9dWVlpLPnrLY+7u3ur19TV1V3zj5E/evjhh7Flyxb897//xe+//45du3bddFYigKVO\nVszOzg733XcfNm3aBACora3FkiVLUFJSgs2bN+PDDz8E0LJ7NSAgAIIgwMbGplXRqlSqVlu7ABAY\nGAhfX1/Ex8dj6tSpEAQBQMsJUY2NjQgJCYEoivi///s/2Nraoqam5pr5FAoF+vbta9zK3LVrl/EE\nqgsXLqBHjx5wd3dHWVkZvv/+e2MmGxsb6HS6q/7gGD16NHbs2IH6+nro9Xps2bKl3We0/3GPxM0a\nPXq08VLACxcuYO/evQgPD8fBgwexfPlyGAwGODs7o3///hAE4brP/5GDgwNGjBiBf//734iMjIRM\nJkNjYyNmz55tvHwsNDQUcrkccrn8pnNnZ2cjMzMTAPD9998br5y43vIMHz4cRUVFSE9PBwC8++67\n+OCDD9r8jDVr1hj/wOvWrRu6d+9+zWUlag9e0kZW7dVXX8XSpUuNxT5t2jR4e3sjIiICS5Yswbhx\n4yCXyxEQEIDXX38dzs7OGDRoEMLDw7F+/XoMGjQIJ06cMO72vWTatGn45z//iVdffdX4nEqlwl/+\n8hdMmTIFHh4emD9/PsaMGYPHH38c69atu2a+V155Bf/4xz/w/vvvY/To0ejVqxcA4P7778d3332H\nyMhI+Pn5YdGiRXj66aexatUqzJw5E2+++Sbuu+8+fPHFF8b3io6OxunTp41njN9zzz3GvRJtaWpq\nwsmTJzF48OB2/ez48eNbPfef//wHixYtwssvv4zx48dDLpdjwYIFCAkJQX19Pb777jtERUXB1tYW\nnp6eeO211+Dh4XHN568lKioKzz33HD7//HMALX+sxcTEYM6cORBFEXK5HMuWLWu1G7ytvACMx8GH\nDRuGTz75BMnJyXB2djYW9PWWB2gp8kWLFkEURfTu3fu6uS+ZOnUqlixZgg8++ACCIGDIkCGYNGnS\nDb5pomsTxOsd0COiG0pKSsI//vEP7N6922K3rg4ePIh33nkHW7dulTpKp9q8eTO+//77VperEXV1\n3P1OdBuGDx8OLy8v/Pjjj1JHMZmPP/64XcfsiUh63P1OdJtee+01PPnkkxg2bNhtHXfuir788kv0\n6NEDERERAICYmBjjsfs/2rZtGxwdHTszHhH9AXe/ExERWQjuficiIrIQLHUiIiILYfbH1DWaq2f5\nIiIislReXi7XHeOWOhERkYVgqRMREVkIljoREZGFYKkTERFZCJY6ERGRhWCpExERWQiWOhERkYVg\nqRMREVkIljoREZGFYKkTERFZCJY6ERGRhWCpX6FAo0PamQtSxyAiIrolLPUr7Dx8Dqu/TsHxLI3U\nUYiIiG4aS/0KE0f4w85Gho+/zUDRhRqp4xAREd0UlvoV/LxdMGdCEOobm7F2WxrqGvRSRyIiImo3\nlvofjAjuhojhvijU1uDT705CFEWpIxEREbULS/0apo/ug36+SiSd0uD7X89LHYeIiKhdWOrXYCOX\n4empIXBT2GHLwRyknyuTOhIREdENsdSvQ6mwx/xpoZAJAj7ckQ5tZZ3UkYiIiNpk0lJfuXIlZsyY\ngdjYWKSmprYaa2howIsvvoiYmJhWz2dlZSEiIgIbN240ZbR26dNDiYcj+0FX14S1W9PQ2NQsdSQi\nIqLrMlmpHz16FLm5uUhISMCKFSuwYsWKVuOrVq3CgAEDWj1XW1uLV199FSNGjDBVrJs2anB33Bvq\ng9ySanz+wymeOEdERF2WyUo9MTERERERAIDAwEBUVlZCp9MZx5977jnj+CV2dnb4+OOPoVarTRXr\npgmCgFnj+sG/mwsOnyjGgd8LpY5ERER0TSYrda1WC3d3d+NjlUoFjebyTG0KheKq19jY2MDBwcFU\nkW6Zna0cC6aFQOFoiy/3ZiG7oFLqSERERFfptBPlzH23tafSEU9OCYZBFPH+thOo1DVIHYmIiKgV\nk5W6Wq2GVqs1Pi4tLYWXl5epPq5TBPdS4cFRgajQNeKD7WnQNxukjkRERGRkslIPCwvDnj17AADp\n6elQq9XX3OVubsbf6Yfh/b2QlV+Jr/dnSx2HiIjISBBNuF88Pj4eSUlJEAQBy5YtQ0ZGBlxcXBAZ\nGYlnnnkGxcXFOH36NEJCQjB9+nQEBATgjTfeQEFBAWxsbODt7Y01a9bAzc3tup+h0VSbKv511TXo\nseLz31CorcHj9w/EiOBunZ6BiIisk5eXy3XHTFrqnUGKUgeAogs1WP5/SWhuFhE3exj8vK//JRMR\nEXWUtkqdM8rdIh8PZzw2cSAa9Qa8t/UEdHVNUkciIiIrx1K/DUP6eWHSPb2grazHR7vSYTCY9U4P\nIiIycyz12zT13gCE9FYh7UwZdhw6K3UcIiKyYiz12ySTCXji/mB4Kh2w68g5HD+tufGLiIiITICl\n3gEUjrZYGBMKOxsZPvk2A8VltVJHIiIiK8RS7yB+3i6YMz4IdQ3NeG/rCdQ36qWOREREVoal3oFG\nhHRDxDBfFGpr8J/vMs1+alwiIjIvLPUONn1MH/TzVSIpsxTfHz0vdRwiIrIiLPUOZiOX4empIVAq\n7LDlQA4yzpVJHYmIiKwES90ElAp7LJgWCpkgYN2OdGgr66SOREREVoClbiJ9eijxcERf6OqasHZb\nGpr0zVJHIiIiC8dSN6FRQ3ogLLQbcour8fmeLJ44R0REJsVSNyFBEDB7XH/4e7vg0IkiHPy9UOpI\nRERkwVjqJmZnK8eCmBAoHG3xxd4s5BRUSh2JiIgsFEu9E3gqHfHk5GAYRBFrt51AZU2j1JGIiMgC\nsdQ7SXCACg+EB6JC14gPtqdB32yQOhIREVkYlnonmnCXH4b190JWXgU278+ROg4REVkYlnonEgQB\nj0YPgI+HE/Ym5eF/6cVSRyIiIgvCUu9kjvY2WBgTCgc7OTbszkReqU7qSEREZCFY6hLw8XDGY5MG\nolFvwHtbU1FT3yR1JCIisgAsdYkM7eeFSff4Q1NRj492ZsDAiWmIiOg2sdQlNPXe3ggJUOHEmQvY\neeis1HGIiMjMsdQlJJMJeGJyMDyVDth5+Bx+P62VOhIREZkxlrrEFI62WBgTClsbGT7+Nh0lZbVS\nRyIiIjPFUu8C/LxdMHd8EOoamvHe1hOob9RLHYmIiMyQSUt95cqVmDFjBmJjY5GamtpqrKGhAS++\n+CJiYmLa/RpLNiKkG8YO80WBtgaffpfJO7oREdFNM1mpHz16FLm5uUhISMCKFSuwYsWKVuOrVq3C\ngAEDbuo1lm7GmD7o66vEscxS7DmaJ3UcIiIyMyYr9cTERERERAAAAgMDUVlZCZ3u8kQrzz33nHG8\nva+xdDZyGZ6eGgKlwg6bD2Tj5LkyqSMREZEZMVmpa7VauLu7Gx+rVCpoNBrjY4VCcdOvsQZuCnss\nmBoKmSDggx3puFBZL3UkIiIyE512otytHCO21uPKfXyVmBnRF7q6JqzddgJN+mapIxERkRkwWamr\n1WpotZevuy4tLYWXl1eHv8ZSjR7SA2Eh3XCuuBqf/5BltX/gEBFR+5ms1MPCwrBnzx4AQHp6OtRq\n9TV3ud/uayyVIAiYHdUfft4KHEotwsGUQqkjERFRFyeIJtwEjI+PR1JSEgRBwLJly5CRkQEXFxdE\nRkbimWeeQXFxMU6fPo2QkBBMnz4d999//1WvCQoKavMzNJpqU8XvErQVdfjnhmOob2zGS48MRWAP\npdSRiIhIQl5eLtcdM2mpdwZLL3UASD9bhtVf/w43hT1envsnKJ3tpI5EREQSaavUOaOcGQgOUCFm\nZG+UVzdg3fY06JsNUkciIqIuiKVuJqLv9sfQfl44lVeBLQdypI5DRERdEEvdTAiCgHkTB8DHwwk/\nHMvD/zKKpY5ERERdDEvdjDja22BhTCgc7OTY8F0m8kqtZ7Y9IiK6MZa6mfHxcMa8iQPRqDdg7dYT\nqKlvkjoSERF1ESx1MzSsvxcmjvBHaUUdPt6VAYN5X8BAREQdhKVupqbd1xvBASqk5lzAzkNnpY5D\nRERdAEvdTMlkAp6cHAxPpQN2Hj6H37O1N34RERFZNJa6GVM42mLBtFDY2sjw8a4MlJTVSh2JiIgk\nxFI3c/7dXDBnfH/UNejx3rYTqG/USx2JiIgkwlK3APeE+GDsUF8UaGqwYXcm7+hGRGSlWOoWYsbY\nPujjq8TRk6X44Vie1HGIiEgCLHULYSOXYf7UECid7bB5fw5O5pZLHYmIiDoZS92CuCnsMX9aCAQB\nWLcjDWVV9VJHIiKiTsRStzB9fd0QO7YvqmubsHbbCTTpm6WOREREnYSlboHGDO2Be0K64WxRNb7Y\nmyV1HCIi6iQsdQskCAL+HNUffmoFfk4pwsHfC6SOREREnYClbqHsbOVYEBMKZwcbfLE3CzmFlVJH\nIiIiE2OpWzAvN0c8OSUYzc0i3t+WhsqaRqkjERGRCbHULVxIgAdiwnujvLoB67anodlgkDoSERGZ\nCEvdCkTf7Y+h/bxwKq8Cm/fnSB2HiIhMhKVuBQRBwLyJA9BN5YQfjuXh14wSqSMREZEJsNSthKO9\nDRbGhMLeTo5Pd59EfqlO6khERNTBWOpWpLunMx6bOBCNTQa8t/UEauubpI5EREQdyKSlvnLlSsyY\nMQOxsbFITU1tNXbkyBE8+OCDmDFjBtauXQsAMBgMWLp0KWJjYzF79mzk5PD4b0cb1t8LE0f4o7Si\nDh/tyoCBd3QjIrIYJiv1o0ePIjc3FwkJCVixYgVWrFjRanz58uVYs2YNvvrqKxw+fBjZ2dnYt28f\nqqursWnTJqxYsQKrVq0yVTyrNu2+3gju5Y7UnAvYdfic1HGIiKiDmKzUExMTERERAQAIDAxEZWUl\ndLqW47h5eXlQKpXw8fGBTCZDeHg4EhMTce7cOQwaNAgA4Ofnh8LCQjQ3c+7yjiaTCXhySgg8XB2w\n49BZ/J6tlToSERF1AJOVularhbu7u/GxSqWCRqMBAGg0GqhUqqvG+vXrh0OHDqG5uRlnzpxBXl4e\nyst5C1FTUDjaYmFMKGxtZPh4VwZKymuljkRERLep006UE9tx7DY8PByhoaF45JFH8Nlnn6F3797t\neh3dGv9uLvhzVH/UNejx3tYTaGjkXhEiInNmY6o3VqvV0Gov79YtLS2Fl5fXNcdKSkqgVqsBAM89\n95zx+YiICHh4eJgqIgEIC/XB2aIq/JRcgE93n8STk4MhCILUsYiI6BaYbEs9LCwMe/bsAQCkp6dD\nrVZDoVAAAHx9faHT6ZCfnw+9Xo/9+/cjLCwMmZmZWLJkCQDg559/xsCBAyGT8ao7U4sd2xd9eihx\n9GQp9h7LkzoOERHdIpNtqQ8dOhTBwcGIjY2FIAhYtmwZtm7dChcXF0RGRuKVV17B4sWLAQDR0dEI\nCAiAwWCAKIp48MEHYW9vj/j4eFPFoyvYyGV4emoI/rXhGL7enwM/bxcE+bvf+IVERNSlCKKZH7TW\naKqljmAxsvIq8O+vjsPJwQbL5v4JKlcHqSMREdEfeHm5XHeM+7bJqF9PN8SO7Yvq2ias3ZaGJj3v\n6EZEZE5Y6tTKmKE9MCK4G84WVeGLvVlSxyEiopvAUqdWBEHAn8f3h59agZ9TCnHw9wKpIxERUTux\n1Okq9rZyLIgJhbODDb7Ym4UzhVVSRyIionZgqdM1ebk54snJwWhuFrF22wlU1TRKHYmIiG6ApU7X\nFdLbA9NG9kZ5dQPW7UhDs4EnzhERdWUsdWpT9Ah/DOnriczzFdhygLfCJSLqyljq1CaZIOCxSQPh\nrXLCnqN5OHqyROpIRER0HSx1uiFHexssjAmFvZ0c//nuJPI1OqkjERHRNbDUqV16eDpjXvQANDYZ\n8N7WE6itb5I6EhER/QFLndpteJAaE+72Q2l5HT7elQGDec8wTERkcVjqdFMeGBmIgb3ckZJzAd8e\nPid1HCIiugJLnW6KTCbgycnB8HB1wI5DZ5GSrZU6EhERXcRSp5vm4mSHhTGhsLGR4eNdGSgpr5U6\nEhERgaVOt8i/mwv+HNUftQ16rN16Ag2NzVJHIiKyeix1umVhoT4YPbQH8jU12PB9JkSeOEdEJCmW\nOt2WmWP7IrCHK37NKMHepHyp4xARWTWWOt0WG7kM86eGwtXZDl//lI1T58uljkREZLVY6nTb3F3s\nMX9qCAQB+GB7Gsqq6qWORERklVjq1CH69XTDjDF9UFXbhPe3p6FJzzu6ERF1NpY6dZixw3wxItgb\nZwqr8NWPWVLHISKyOix16jCCIODP44PQU63Agd8L8XNKodSRiIisCkudOpS9rRwLYkLh7GCDjT+c\nwpnCKqkjERFZDZY6dTi1myOemByM5mYRa7edQFVNo9SRiIisAkudTCK0twemjuyN8uoGrNuRhmYD\nT5wjIjI1k5b6ypUrMWPGDMTGxiI1NbXV2JEjR/Dggw9ixowZWLt2LQCgpqYGCxcuxOzZsxEbG4tf\nfvnFlPHIxCaO8MeQvp7IPF+Bbw6ckToOEZHFM1mpHz16FLm5uUhISMCKFSuwYsWKVuPLly/HmjVr\n8NVXX+Hw4cPIzs7Gtm3bEBAQgM8//xzvvPPOVa8h8yITBMybOBDeKid8f/Q8jp4skToSEZFFM1mp\nJyYmIiIiAgAQGBiIyspK6HQ6AEBeXh6USiV8fHwgk8kQHh6OxMREuLu7o6KiAgBQVVUFd3d3U8Wj\nTuLkYIOFMaGwt5Xj0+8yka/RSR2JiMhimazUtVptq1JWqVTQaDQAAI1GA5VKddXYxIkTUVhYiMjI\nSMyaNQsvvviiqeJRJ+rh6Yx5EwegoakZa7eeQG19k9SRiIgsUqedKNeeO3jt2LED3bt3x969e/HZ\nZ5/hX//6Vycko84wPEiNCXf5oaS8Dp98exIG3tGNiKjDmazU1Wo1tFqt8XFpaSm8vLyuOVZSUgK1\nWo3k5GTce++9AICgoCCUlpaiuZn36bYUMeG9McDfHb9na/HtkXNSxyEisjgmK/WwsDDs2bMHAJCe\nng61Wg2FQgEA8PX1hU6nQ35+PvR6Pfbv34+wsDD4+/sjJSUFAFBQUABnZ2fI5XJTRaROJpfJ8NSU\nYHi42mPHL2eRmqO98YuIiKjdBLE9+8VvUXx8PJKSkiAIApYtW4aMjAy4uLggMjISx44dQ3x8PABg\n3LhxmDdvHmpqahAXF4cLFy5Ar9fj2WefxYgRI9r8DI2m2lTxyUTOFVdh5efJsLOR4eW5w6F2d5I6\nEhGR2fDycrnumElLvTOw1M3TodQi/Oe7k/D1csbfZw+HvR33yBARtUdbpc4Z5UgS9w7yweghPZCv\nqcFn32e260RKIiJqG0udJDMzoi8Cu7vifxkl+DEpX+o4RERmj6VOkrGRyzB/Wihcne2Q8FM2Tp0v\nlzoSEZFZY6mTpNxd7DF/aggEAfhgexrKqxukjkREZLZY6iS5fj3dMH1MH1TVNuH9bSfQpOcd3YiI\nbgVLnbqEiGG+uDvYGzmFVfhq32mp4xARmSWWOnUJgiBgzvgg9FQrcOB4AX5JKZQ6EhGR2WGpU5dh\nbyvHgphQONnb4PMfsnC2qErqSEREZoWlTl2K2s0RT0wORnOzAWu3nUBVbaPUkYiIzAZLnbqcQYEe\nmHpfAMqqGvDhjnQ0G3jiHBFRe7Sr1NPS0rB//34AwFtvvYU5c+YgKSnJpMHIuk28pxcG9/HEydxy\nfHPwjNRxiIjMQrtKffny5QgICEBSUhJOnDiBpUuX4t133zV1NrJiMkHAY5MGwtvdEd//eh4Hfy+Q\nOhIRUZfXrlK3t7dHr169sG/fPkyfPh19+vSBTMY992RaTg42WPjAICgcbfHZ96fw38RznCOeiKgN\n7Wrmuro67N69Gz/++CPuvfdeVFRUoKqKZyaT6fXwdMaSWUOhcrXHNwfPIOGnbBhY7ERE19SuUl+0\naBF27dqF5557DgqFAp9//jnmzp1r4mhELXw8nBE3axh8PJzww7E8rP82A/pmnjxHRPRH7b6fuk6n\ng0KhgFarxblz5zB06NAusQtkOuOTAAAgAElEQVSe91O3Hrq6JryzOQU5hVUI7e2B+VNDeB92IrI6\nbd1PXf7KK6+8cqM3ePXVV1FRUYEePXpg+vTpKC4uRmJiIkaPHt2ROW9JLa9jthp2tnLcNcAbuSXV\nSDtThszz5Rjazwt2tix2IrIezs721x1r16Z2RkYGHnroIezevRvTpk3D22+/jdzc3A4LSNRe9nZy\nPPPAoJZ54guq8PoXySirqpc6FhFRl9CuUr+0h/7AgQMYM2YMAKCxkVvIJA0buQyPTRqIiOG+KNTW\n4LWNv6HoQo3UsYiIJNeuUg8ICEB0dDRqamowYMAAbN++HUql0tTZiK5LJgiYObYvHgjvjQtVDXht\nYzLniiciq9euE+Wam5uRlZWFwMBA2NnZIS0tDX5+fnB1de2MjG3iiXL0c0ohPvs+E3Y2ciyMCUVw\ngErqSEREJtPWiXLtKvWamhps2LABJ06cgCAIGDx4MObMmQMHB4cODXorWOoEAL+d0uDDnekQRRGP\n3z8Qdw7wljoSEZFJ3HapL1q0CN7e3rjrrrsgiiKOHDmC8vJyxMfHd2jQW8FSp0syc8vx7jepaGhs\nxiPj+mHMUF+pIxERdbi2St2mPW+g1WqxevVq4+PRo0dj9uzZt5+MqAMF+bvjxYeH4q2vf8fGH7JQ\nVdOIKfcGQBAEqaMREXWKdpV6XV0d6urq4OjoCACora1FQ0PDDV+3cuVKpKSkQBAExMXFYdCgQcax\nI0eOYPXq1ZDL5Rg5ciQWLFiAzZs3Y+fOncafSUtLw/Hjx292mciK+XdzwZLZw/Dmpt+x8/A5VNc1\n4ZGIfpDJWOxEZPnaVeozZszAhAkTEBISAgBIT0/Hs88+2+Zrjh49itzcXCQkJCAnJwdxcXFISEgw\nji9fvhzr16+Ht7c3Zs2ahaioKDz00EN46KGHjK/fvXv3rS4XWTFvdyfEzR6G1Qkp2J9cAF1tEx6b\nNBC2NtLPgEhEZErt+i334IMP4quvvsLUqVMxbdo0bNq0CdnZ2W2+JjExEREREQCAwMBAVFZWQqfT\nAQDy8vKgVCrh4+MDmUyG8PBwJCYmtnr92rVrMX/+/FtZJiK4Kezx0iND0M9XiWOZpXhnSwrqGvRS\nxyIiMql2b7r4+PggIiICY8eOhbe3N1JTU9v8ea1WC3d3d+NjlUoFjUYDANBoNFCpVNccA4DU1FT4\n+PjAy8ur3QtC9EdODrZYNGMwBvfxRMa5cvz7q+Oo4rTCRGTBbnl/5M3e1/pmfn7Lli2YNm3azUYi\nuoqdrRwLYkJwb6gPzhVX47WNydBW1kkdi4jIJG651G90RrFarYZWqzU+Li0tNW55/3GspKQEarXa\n+PjXX3/FkCFDbjUaUStymQx/iQ7ChLv9UFJWi5Wf/4YCjU7qWEREHa7NE+XCw8OvWd6iKKK8vLzN\nNw4LC8OaNWsQGxuL9PR0qNVqKBQKAICvry90Oh3y8/PRrVs37N+/33jNe0lJCZydnWFnZ3ery0R0\nFUEQ8NCoPnBxtMPX+7Px+hfJePbBO9DHl9MdE5HlaLPUv/zyy1t+46FDhyI4OBixsbEQBAHLli3D\n1q1b4eLigsjISLzyyitYvHgxACA6OhoBAQEArj7eTtSRxt/lBxcnW3z6XSbiNx3H/GkhGBToKXUs\nIqIO0a4Z5boyzihHtyIlW4sPtqdB3yzi0YlBuCfER+pIRETt0taMcrxwl6zSHX08sTh2MBzs5Pjk\n25P44eh5qSMREd02ljpZrb6+bnhp1lC4Keyw6adsbDmQc9NXdRARdSUsdbJqvl4KxM0aBm93R3z3\nv1x8ujsTzQaD1LGIiG4JS52snqebI5bMGgb/bi44lFqE97elobGpWepYREQ3jaVOBMDV2Q5/mzkE\nA/zdcfy0Fqu/TkFtPaeVJSLzwlInusjR3gZ/fegODO/vhay8CrzxZTIqdTe+GyERUVfBUie6gq2N\nDE9NCcGoIT2QV6rDyo2/obS8VupYRETtwlIn+gOZTMDscf0wOawXNBX1WLkxGedLOB8CEXV9LHWi\naxAEAVPv641HIvuhuqYRb3yZjFPn254amYhIaix1ojaMHeaLJyYHo7HJgDcTUpCcpbnxi4iIJMJS\nJ7qBuwZ649mHBkEuE7B22wn8nFIodSQiomtiqRO1Q0iAB16YOQTODrbYsDsT/008x9nniKjLYakT\ntVPv7q5YMmsoVK72+ObgGST8lA0Di52IuhCWOtFN8PFwRtysYfDxcMIPx/Kw/tsM6Js5rSwRdQ0s\ndaKbpHJ1wJJZwxDY3RWJ6SVY880JNDRyWlkikh5LnegWKBxt8XzsEIT0VuHEmQuITzgOXV2T1LGI\nyMqx1Ilukb2dHM88MAh3B3sjp6AKr3+RjLKqeqljEZEVY6kT3QYbuQyPTRqIiOG+KNTW4LWNv6Ho\nQo3UsYjISrHUiW6TTBAwc2xfPBDeGxeqGvDaxmScLaqSOhYRWSGWOlEHEAQBE0f0wtwJQaipb8Kq\nL48j/WyZ1LGIyMqw1Ik60Mg7umP+1FA0G0S8vTkFR0+WSB2JiKwIS52ogw3r74VF0++ArY0MH+5I\nx0/J+VJHIiIrwVInMoEgf3e8+PBQuDjZYuMPWdj+yxlOK0tEJsdSJzIR/24uWDJ7GDyVDth5+Bw2\n7s2CwcBiJyLTYakTmZC3uxPiZg+Dr5cC+5ML8OHOdDTpOa0sEZmGSUt95cqVmDFjBmJjY5Gamtpq\n7MiRI3jwwQcxY8YMrF271vj8zp07MXnyZMTExODAgQOmjEfUKdwU9njpkSHo56vEscxSvLMlBXUN\neqljEZEFMlmpHz16FLm5uUhISMCKFSuwYsWKVuPLly/HmjVr8NVXX+Hw4cPIzs5GeXk51q5diy+/\n/BLr1q3Dvn37TBWPqFM5Odhi0YzBGNzHExnnyvHvr46jqrZR6lhEZGFMVuqJiYmIiIgAAAQGBqKy\nshI6nQ4AkJeXB6VSCR8fH8hkMoSHhyMxMRGJiYkYMWIEFAoF1Go1Xn31VVPFI+p0drZyLIgJwb2h\nPjhXXI3XNiZDW1kndSwisiAmK3WtVgt3d3fjY5VKBY1GAwDQaDRQqVRXjeXn56O+vh5PPfUUHn74\nYSQmJpoqHpEk5DIZ/hIdhAl3+6GkrBYrP/8NBRqd1LGIyEJ02oly7b2cp6KiAu+99x5ef/11LFmy\nhJcBkcURBAEPjeqD6aP7oELXiNe/SEZ2fqXUsYjIApis1NVqNbRarfFxaWkpvLy8rjlWUlICtVoN\nDw8PDBkyBDY2NvDz84OzszPKyjjVJlmm8Xf5Yd7EAahraEb8puNIzdHe+EVERG0wWamHhYVhz549\nAID09HSo1WooFAoAgK+vL3Q6HfLz86HX67F//36EhYXh3nvvxf/+9z8YDAaUl5ejtra21S58IksT\nFuqD//dAKADg3S0ncCStSOJERGTOBNGE+7fj4+ORlJQEQRCwbNkyZGRkwMXFBZGRkTh27Bji4+MB\nAOPGjcO8efMAAJs2bcKWLVsAAE8//TTGjh3b5mdoNNWmik/UaU7nV+CdzamobdAjdkwfjLvTT+pI\nRNRFeXm5XHfMpKXeGVjqZCnyNTqsTvgdFbpGRN/tjwfCe0MQBKljEVEX01apc0Y5oi7C10uBuFnD\n4O3uiO/+l4tPd2ei2cDZ54io/VjqRF2Ip5sjlswaBv9uLjiUWoT3t6WhsalZ6lhEZCZY6kRdjKuz\nHf42cwgG+Lvj+GktVn+dgtp6TitLRDfGUifqghztbfDXh+7A8P5eyMqrwBtfJqNS1yB1LCLq4ljq\nRF2UrY0MT00JwaghPZBXqsPKjb+htLxW6lhE1IWx1Im6MJlMwOxx/TA5rBc0FfVYuTEZ50t4xQcR\nXRtLnaiLEwQBU+/rjUci+6G6phFvfJmMU+fLpY5FRF0QS53ITIwd5osnJgejscmANxNSkJylkToS\nEXUxLHUiM3LXQG88+9AgyGUC1m47gZ9TCqWORERdCEudyMyEBHjghZlD4Oxgiw27M/HfxHO8myER\nAWCpE5ml3t1dsWTWUKhc7fHNwTNI+CkbBhY7kdVjqROZKR8PZ8TNGgYfDyf8cCwP67/NgL6Z08oS\nWTOWOpEZU7k6YMmsYQjs7orE9BKs+eYEGho5rSyRtWKpE5k5haMtno8dgpDeKpw4cwHxCcehq2uS\nOhYRSYClTmQB7O3keOaBQbg72Bs5BVV4/YtklFXVSx2LiDoZS53IQtjIZXhs0kBEDPdFobYGr238\nDUUXaqSORUSdiKVOZEFkgoCZY/vigfDeuFDVgNc2JuNsUZXUsYiok7DUiSyMIAiYOKIX5k4IQk19\nE1Z9eRzpZ8ukjkVEnYClTmShRt7RHfOnhqLZIOLtzSk4erJE6khEZGIsdSILNqy/FxZNvwO2NjJ8\nuCMdPyXnSx2JiEyIpU5k4YL83fHiw0Ph4mSLjT9kYfsvZzitLJGFYqkTWQH/bi5YMnsYPJUO2Hn4\nHDbuzYLBwGInsjQsdSIr4e3uhLjZw+DrpcD+5AJ8uDMdTXpOK0tkSVjqRFbETWGPlx4Zgn6+ShzL\nLMU7W1JQ16CXOhYRdRCWOpGVcXKwxaIZgzG4jycyzpXj318dR1Vto9SxiKgDCKIJz5hZuXIlUlJS\nIAgC4uLiMGjQIOPYkSNHsHr1asjlcowcORILFizAr7/+imeffRZ9+/YFAPTr1w9Lly5t8zM0mmpT\nxSeyaM0GAz7bfQqHThTBW+WExTPugKfSUepYRHQDXl4u1x2zMdWHHj16FLm5uUhISEBOTg7i4uKQ\nkJBgHF++fDnWr18Pb29vzJo1C1FRUQCAO++8E++++66pYhHRRXKZDH+JDoKLsy12/+88Vn7+GxbP\nGIweXgqpoxHRLTLZ7vfExEREREQAAAIDA1FZWQmdTgcAyMvLg1KphI+PD2QyGcLDw5GYmGiqKER0\nHYIg4KFRfTB9dB9U6Brx+hfJyM6vlDoWEd0ik5W6VquFu7u78bFKpYJGowEAaDQaqFSqa45lZ2fj\nqaeewsyZM3H48GFTxSOiK4y/yw/zJg5AXUMz4jcdR2qOVupIRHQLTLb7/Y/ac+i+V69eWLhwISZM\nmIC8vDz8+c9/xg8//AA7O7tOSEhk3cJCfaBwtMUH29Pw7pYTeHRiEO4J8ZE6FhHdBJNtqavVami1\nl//aLy0thZeX1zXHSkpKoFar4e3tjejoaAiCAD8/P3h6eqKkhPNVE3WWO/p4YnHsYDjYyfHJtyfx\nw9HzUkcioptgslIPCwvDnj17AADp6elQq9VQKFpOwPH19YVOp0N+fj70ej3279+PsLAw7Ny5E+vX\nrwfQsov+woUL8Pb2NlVEIrqGvr5ueGnWULgp7LDpp2xsOZDDaWWJzIRJL2mLj49HUlISBEHAsmXL\nkJGRARcXF0RGRuLYsWOIj48HAIwbNw7z5s2DTqfD888/j6qqKjQ1NWHhwoUIDw9v8zN4SRuRaWgr\n6vBmwu8oKa/DvYN8MGd8f8hlnNqCSGptXdJm0lLvDCx1ItOpqmnEW5tTkFtcjSF9PfHk5GDY2cql\njkVk1doqdf7ZTUTX5epsh7/NHIIB/u44flqL1V+noLae08oSdVUsdSJqk6O9Df760B0Y3t8LWXkV\neOPLZFTqGqSORUTXwFInohuytZHhqSkhGDWkB/JKdVi58TcUl9VKHYuI/oDH1Imo3URRxI5DZ7Hz\n8DkAQHdPZ/Tr6YZ+PZXo39Md7i720gYksgI8UY6IOtT/0otxOK0Y2fmVaGhqNj6vdnO8WPJu6O/n\nBk+lAwRBkDApkeVhqRORSeibDThfokNWXoXxX+0V92d3d7FH/ytKvpvKiSVPdJtY6kTUKQwGEfma\n1iVfVdtkHHdxsr28Jd/TDb5eCshkLHmim8FSJyJJiKKI4rJanLpY8KfOV6C8+vKZ8072Nujrq0Q/\nv5ai9/d2gY2c5+8StYWlTkRdgiiK0FbWtxT8xaIvLa8zjtvbytGnh6txa753d1fY2nCyG6IrsdSJ\nqMsqr25otbu+QFtjHLORy9C7u6txd32fHkrY27Hkybqx1InIbFTVNuJ0XqWx5M+XVuPSbym5TIB/\nNxfjlnw/XyWcHGylDUzUyVjqRGS2auv1yC64vLv+XFE1mg0tv7YEAD3Vissl39MNrs520gYmMjGW\nOhFZjIbGZpwprDSWfE5hFZr0BuO4j4eT8TK6fj3doHJ1kDAtUcdjqRORxWrSG3CuuAqnzreU/OmC\nSjQ0Xp4Qx8vNodVldF5ujrxWnswaS52IrEazoWVCHGPJ51egpr71hDhX7q7v7sEJcci8sNSJyGoZ\nRBGFmhqcuuIyuqqaRuO4wtG21e76nmpOiENdG0udiOgiURRRUl6HU+fLjdfLl1VdnhDH8eKEOJeK\n3r8bJ8ShroWlTkTUBm1lnXHGu6y8CpRcMSGOna0MfXoojcfkOSEOSY2lTkR0Eyp0Da1mvSvQXDkh\njoDePq7GqW379FDCwc5GwrRkbVjqRES3QVfXhNMXS/5UXgXOl1yeEEcmtEyIc2l3fd+eSjhzQhwy\nIZY6EVEHqmvQI7ug0ri7/mxRVasJcXwvTojTv6cb+vZ0g5IT4lAHYqkTEZlQQ1MzzhRWXTwuX37V\nhDjdVE7o73f5WnlOiEO3g6VORNSJ9M0GnCuqxqm8cpzKq0B2fiXqr5gQx1PpcPkyOj83qDkhDt0E\nljoRkYSaDQbkleqQdf7yyXdXTojjprAzbsX36+kGH09nyFjydB2SlfrKlSuRkpICQRAQFxeHQYMG\nGceOHDmC1atXQy6XY+TIkViwYIFxrL6+HpMmTcL8+fMRExPT5mew1InI3BhEEYXamlaX0VX+YUKc\nK6e25YQ4dKW2St1k12EcPXoUubm5SEhIQE5ODuLi4pCQkGAcX758OdavXw9vb2/MmjULUVFR6NOn\nDwDggw8+gFKpNFU0IiJJyQQBvl4K+HopMGaoL0RRRGl5XcvZ9RdLPjlLg+QsDQDA0V6Ovr6XZ73r\nxQlx6DpMVuqJiYmIiIgAAAQGBqKyshI6nQ4KhQJ5eXlQKpXw8fEBAISHhyMxMRF9+vRBTk4OsrOz\nMWrUKFNFIyLqUgRBgLfKCd4qJ4y8ozuAlglxTudVGi+jS825gNScCwAAOxsZAntcnvWud3dX2Nly\nQhwyYalrtVoEBwcbH6tUKmg0GigUCmg0GqhUqlZjeXl5AIA33ngDS5cuxfbt200VjYioy/NUOsJT\n6YgRId0AAJW6BmTlV148Ll+Ok7kt/wBALhMQ0N0V/Xu6IcjPHX18lbBnyVulTpsGqT2H7rdv347B\ngwejZ8+enZCIiMh8KBX2+FOQGn8KUgO4OCFOfoXxuHxOQSWy8yvx38RcyGUCend3RZCfOwb4uyOw\nB6e2tRYmK3W1Wg2tVmt8XFpaCi8vr2uOlZSUQK1W48CBA8jLy8OBAwdQXFwMOzs7dOvWDffcc4+p\nYhIRmSWFoy2G9PXCkL4tv1cvTYiTmVuOzPPlyC6oxOn8Suw6cg42chn69HBFkH9LyQf4uPKYvIUy\nWamHhYVhzZo1iI2NRXp6OtRqNRQKBQDA19cXOp0O+fn56NatG/bv34/4+HjMmjXL+Po1a9agR48e\nLHQionZwtLdBaG8PhPb2AADU1uuRlVeBzPPlyMwtx6nzFcg8X4Htv5yFna0MfX3dEOTnhiB/d/Tq\n5gK5jCVvCUxW6kOHDkVwcDBiY2MhCAKWLVuGrVu3wsXFBZGRkXjllVewePFiAEB0dDQCAgJMFYWI\nyOo4OdhgcF9PDO7rCaBld/2p8+XIzG0p+vSzZUg/WwYAcLCTo9/F4/ED/N15CZ0Z4+QzRERWqKqm\n0bgVf/J8BUrKao1jTvY26H9xK36Anzu6e3EynK6EM8oREVGbyqsbLpd8bjm0lfXGMYWjrXFX/QB/\nd3RTOXFaWwmx1ImI6KZoK+uMu+ozz5ejrKrBOKZ0tkOQv7ux6Dl3fediqRMR0S0TRRGlFXUXz6yv\nQGZueatpbd1d7DHA3x1Bfu4I8neDp9JRwrSWj6VOREQdRhRFFJfV4mRuubHodXVNxnFPpYNxV32Q\nnzvcXewlTGt5WOpERGQyBlFEoaYGJ6+4fK624fJd6LxVThhwcVd9fz93KJ3tJExr/ljqRETUaQwG\nEXmlupYt+fPlyMqraHU/+R6ezsZd9f393KFwtJUwrflhqRMRkWSaDQacK6427qo/nV+BxiYDAEAA\n0FOtuHjinTv69XSDk0OnzWBulljqRETUZeibDThTWGW8hC67oAr65oslLwD+3i7GY/J9fZVwsGPJ\nX4mlTkREXVaTvhk5BS0lfzK3HGcKq9BsaKkmuUxALx+Xi7vr3dGnB+9Ax1InIiKz0dDY3HJzmotb\n8meLqmG4WFU2cgG9uysR5OeGAf7u6N1dCVsb65q3nqVORERmq65Bj9P5FcjMrcDJ8+U4X1yNS8Vl\nayNDnx5K45S2vXxcLP4OdCx1IiKyGDX1Tcg6X3HxEroK5Gt0xjF7Wzn6+iqNx+T9vBUWdwc6ljoR\nEVms6tpGnDKWfDmKLly+OY2jvRz9fN2MZ9f39FaY/c1pWOpERGQ1KnUNLdPZXiz5kvI645izgw36\n+12et76Hp7PZzVvPUiciIqtVVlV/seBbiv7KO9C5ONkaz6wP8nMzizvQsdSJiIgu0hhvTtMyGU55\n9RV3oFPYYcClkvd3h5fSocuVPEudiIjoGkRRRGl5nXFK28zcclTVXr45jYer/RVb8u7wUDpImLYF\nS52IiKgdRFFE4YXay1vyueWoqb98cxq1myOC/N2MRe+m6Pw70LHUiYiIboFBFJFfqjPeR/5UXgXq\nrrgDXTeVU8stZv3d0d/PDa5Opr8DHUudiIioAxgMInJLqo0n3mXlV6DhyjvQebXcgW7AxZJ3duj4\nO9Cx1ImIiExA32xAbnG18Zh8dn4lGvVX3IHOW4Hou/1x5wDvDvtMljoREVEnaNIbcLaoCpm5LTen\nySmsxPAgNZ64P7jDPoOlTkREJAF9swFymdChl8W1Veq8SS0REZGJdPbNZSxrlnsiIiIrZtIt9ZUr\nVyIlJQWCICAuLg6DBg0yjh05cgSrV6+GXC7HyJEjsWDBAtTV1eGll17ChQsX0NDQgPnz52P06NGm\njEhERGQxTFbqR48eRW5uLhISEpCTk4O4uDgkJCQYx5cvX47169fD29sbs2bNQlRUFLKyshASEoLH\nH38cBQUFePTRR1nqRERE7WSyUk9MTERERAQAIDAwEJWVldDpdFAoFMjLy4NSqYSPjw8AIDw8HImJ\niZg9e7bx9UVFRfD27rhLAIiIiCydyUpdq9UiOPjyKfwqlQoajQYKhQIajQYqlarVWF5envFxbGws\niouLsW7dOlPFIyIisjiddqLczVw5t2nTJnzwwQd44YUXbup1RERE1sxkpa5Wq6HVao2PS0tL4eXl\ndc2xkpISqNVqpKWloaioCAAwYMAANDc3o6yszFQRiYiILIrJSj0sLAx79uwBAKSnp0OtVkOhUAAA\nfH19odPpkJ+fD71ej/379yMsLAxJSUn4z3/+A6Bl931tbS3c3d1NFZGIiMiimHRGufj4eCQlJUEQ\nBCxbtgwZGRlwcXFBZGQkjh07hvj4eADAuHHjMG/ePNTX1+Pvf/87ioqKUF9fj4ULF2LMmDFtfgZn\nlCMiImvCaWKJiIgsRFulzhnliIiILITZb6kTERFRC26pExERWQiWOhERkYVgqRMREVkIljoREZGF\nYKkTERFZCJY6ERGRhTDZXdq6upUrVyIlJQWCICAuLg6DBg0yjh05cgSrV6+GXC7HyJEjsWDBAgmT\n3lhbyzJmzBh069YNcrkcQMssf135lrZZWVmYP38+5s6di1mzZrUaM7f10taymNt6WbVqFX777Tfo\n9Xo8+eSTGDdunHHMnNZLW8thTuukrq4OL730Ei5cuICGhgbMnz8fo0ePNo6b0zq50bKY03oBgPr6\nekyaNAnz589HTEyM8flOWyeiFfr111/FJ554QhRFUczOzhanT5/eanzChAliYWGh2NzcLM6cOVM8\nffq0FDHb5UbLMnr0aFGn00kR7abV1NSIs2bNEv/xj3+In3/++VXj5rRebrQs5rReEhMTxccee0wU\nRVEsKysTw8PDW42by3q50XKY0zr573//K3700UeiKIpifn6+OG7cuFbj5rJORPHGy2JO60UURXH1\n6tViTEyM+M0337R6vrPWiVXufk9MTERERAQAIDAwEJWVldDpdACAvLw8KJVK+Pj4QCaTITw8HImJ\niVLGbVNby2Ju7Ozs8PHHH0OtVl81Zm7rpa1lMTd/+tOf8M477wAAXF1dUVdXh+bmZgDmtV7aWg5z\nEx0djccffxwAUFRU1GrL1ZzWCdD2spibnJwcZGdnY9SoUa2e78x1YpW737VaLYKDg42PVSoVNBoN\nFAoFNBoNVCpVq7G8vDwpYrZLW8tyybJly1BQUIBhw4Zh8eLFEARBiqg3ZGNjAxuba/8vaW7rpa1l\nucRc1otcLoeTkxMAYMuWLRg5cqRxV6g5rZe2luMSc1knl8TGxqK4uBjr1q0zPmdO6+RK11qWS8xl\nvbzxxhtYunQptm/f3ur5zlwnVlnqfyRa0Ey5f1yWZ555Bvfddx+USiUWLFiAPXv2YPz48RKlo0vM\ncb38+OOP2LJli/H2yObqesthjutk06ZNOHnyJF544QXs3Lmzy5Zde1xvWcxlvWzfvh2DBw9Gz549\nJc1hlbvf1Wo1tFqt8XFpaSm8vLyuOVZSUtKld6G2tSwAMHXqVHh4eMDGxgYjR45EVlaWFDFvm7mt\nlxsxt/Xyyy+/YN26dfj444/h4nL5DlHmtl6utxyAea2TtLQ0FBUVAQAGDBiA5uZmlJWVATC/ddLW\nsgDms14OHDiAffv2Yfr06di8eTPef/99HDlyBEDnrhOrLPWwsDDs2bMHAJCeng61Wm3cXe3r6wud\nTof8/Hzo9Xrs378fYSIwtAYAAAQbSURBVGFhUsZtU1vLUl1djXnz5qGxsREAcOzYMfTt21eyrLfD\n3NZLW8xtvVRXV2PVqlX48MMP4ebm1mrMnNZLW8thbuskKSnJuKdBq9WitrYW7u7uAMxrnQBtL4s5\nrZe3334b33zzDb7++ms89NBDmD9/Pu655x4AnbtOrPYubfHx8UhKSoIgCFi2bBkyMjLg4uKCyMhI\nHDt2DPHx8QCAcePGYd68eRKnbVtby/LZZ59h+/btsLe3x8CBA7F06dIuu4suLS0Nb7zxBgoKCmBj\nYwNvb2+MGTMGvr7/v727B2ktCcM4/o85qARs/ABBm4gasBA/IDaBoJVgGSwsLEQbRREUgyLxo9Gk\nEUkqQatgojZio5UgGNQUIhHUVkSwimhASBHDFrKXjdddl/XelZw8v3KKk7wMnOe8MzBTm3fz8lkt\n+TQv29vbhEIh7Hb7j7GOjg4cDkdezctndeTTnKTTaWZnZ3l4eCCdTjM6OsrT01NevsM+qyWf5uVP\noVCImpoagP99Tgo21EVERMymIJffRUREzEihLiIiYhIKdREREZNQqIuIiJiEQl1ERMQkdKKcSAG6\nv7+nu7ub1tbWnHG3283Q0NCXnx+Px1ldXSUajX75WSLy7ynURQpUeXk54XD4u/+GiPxCCnURydHU\n1MTIyAjxeJyXlxf8fj+NjY0kEgn8fj+GYWCxWJibm6O+vp7b21t8Ph/ZbJaSkhKWl5cByGazzM/P\nc3NzQ3FxMWtrawBMTk6SSqXIZDJ0dnYyPDz8neWKmIr21EUkx+vrKw0NDYTDYfr6+ggGgwB4vV5m\nZmYIh8MMDAywuLgIvN2gNTg4yObmJh6Ph4ODA+DtGsqxsTF2dnYwDINYLMbJyQmZTIZIJMLW1hY2\nm41sNvtttYqYjTp1kQL1+PhIf39/ztjU1BQALpcLgLa2NjY2NkilUiSTSZqbmwFwOp1MTEwAcHl5\nidPpBKCnpwd421Ovq6ujsrISgOrqalKpFF1dXQSDQcbHx3G73fT29lJUpN5C5FdRqIsUqH/aU//r\n6dEWi+Wns7bfny79Ubf9/q5ygIqKCvb29ri4uODw8BCPx8Pu7i6lpaX/pQQReUefyCLyk7OzMwDO\nz89xOByUlZVRVVVFIpEA4PT0lJaWFuCtmz8+PgZgf3+flZWVv31uLBbj6OiI9vZ2vF4vNpuNZDL5\nm6sRKRzq1EUK1EfL77W1tQBcX18TjUZ5fn4mEAgAEAgE8Pv9WK1WioqKWFhYAMDn8+Hz+YhEIhiG\nwdLSEnd3dx/+pt1uZ3p6mvX1daxWKy6X68dtViLydbqlTURyOBwOrq6uMAx984vkGy2/i4iImIQ6\ndREREZNQpy4iImISCnURERGTUKiLiIiYhEJdRETEJBTqIiIiJqFQFxERMYk/AD/ihNKaxK9YAAAA\nAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2fb456d278>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('Test(Validation)_Loss vs Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 652
    },
    "colab_type": "code",
    "id": "aG3IXsj9tA2Z",
    "outputId": "4e25b6da-3fba-4ad7-e695-1dd97a46a353"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10000/10000 [==============================] - 4s 395us/step\n",
      "Rotation degree: -45 Score:[0.6708947430133841, 0.9578]\n",
      "10000/10000 [==============================] - 4s 381us/step\n",
      "Rotation degree: -40 Score:[0.42005940365791994, 0.9737]\n",
      "10000/10000 [==============================] - 4s 380us/step\n",
      "Rotation degree: -35 Score:[0.28725108520687465, 0.9817]\n",
      "10000/10000 [==============================] - 4s 384us/step\n",
      "Rotation degree: -30 Score:[0.23969086404441478, 0.9849]\n",
      "10000/10000 [==============================] - 4s 380us/step\n",
      "Rotation degree: -25 Score:[0.208757014851282, 0.987]\n",
      "10000/10000 [==============================] - 4s 382us/step\n",
      "Rotation degree: -20 Score:[0.19374479918574064, 0.9878]\n",
      "10000/10000 [==============================] - 4s 385us/step\n",
      "Rotation degree: -15 Score:[0.2094399516776467, 0.9866]\n",
      "10000/10000 [==============================] - 4s 384us/step\n",
      "Rotation degree: -10 Score:[0.21994463266357114, 0.9861]\n",
      "10000/10000 [==============================] - 4s 381us/step\n",
      "Rotation degree: -5 Score:[0.22984542411128422, 0.9856]\n",
      "10000/10000 [==============================] - 4s 380us/step\n",
      "Rotation degree: 0 Score:[0.1743536612887455, 0.989]\n",
      "10000/10000 [==============================] - 4s 382us/step\n",
      "Rotation degree: 5 Score:[0.2615081715133654, 0.9832]\n",
      "10000/10000 [==============================] - 4s 383us/step\n",
      "Rotation degree: 10 Score:[0.21820959478021715, 0.9858]\n",
      "10000/10000 [==============================] - 4s 381us/step\n",
      "Rotation degree: 15 Score:[0.1969586051166155, 0.9876]\n",
      "10000/10000 [==============================] - 4s 383us/step\n",
      "Rotation degree: 20 Score:[0.22820646737330572, 0.9854]\n",
      "10000/10000 [==============================] - 4s 385us/step\n",
      "Rotation degree: 25 Score:[0.23869694012984702, 0.9847]\n",
      "10000/10000 [==============================] - 4s 381us/step\n",
      "Rotation degree: 30 Score:[0.2911111881519574, 0.9817]\n",
      "10000/10000 [==============================] - 4s 383us/step\n",
      "Rotation degree: 35 Score:[0.3645175949275337, 0.9772]\n",
      "10000/10000 [==============================] - 4s 385us/step\n",
      "Rotation degree: 40 Score:[0.5063861561462402, 0.9681]\n",
      "10000/10000 [==============================] - 4s 385us/step\n",
      "Rotation degree: 45 Score:[0.7404538876952609, 0.953]\n"
     ]
    }
   ],
   "source": [
    "s_loss = []\n",
    "s_acc = []\n",
    "for j in range(-45,50,5):\n",
    "  rotate_test_x = []\n",
    "  for i in range(len(X_test)):\n",
    "    X_test_rotate = Image.fromarray(X_test[i])\n",
    "    new_image = X_test_rotate.rotate(j)\n",
    "    img_array = numpy.array(new_image)\n",
    "    rotate_test_x.append(img_array)\n",
    "  X_test_rotate = numpy.asarray(rotate_test_x)\n",
    "  X_test_rotate = X_test_rotate.reshape(X_test_rotate.shape[0], img_rows, img_cols, 1)\n",
    "  score = model.evaluate(X_test_rotate, y_test, batch_size=32)\n",
    "  print(\"Rotation degree: \" +str(j) + \" Score:\" + str(score))\n",
    "  s_loss.append(score[0])\n",
    "  s_acc.append(score[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 334
    },
    "colab_type": "code",
    "id": "AQev_eKNtC-e",
    "outputId": "89585b85-85e5-4390-e890-ac4b6de766d2"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.6708947430133841,\n",
       " 0.42005940365791994,\n",
       " 0.28725108520687465,\n",
       " 0.23969086404441478,\n",
       " 0.208757014851282,\n",
       " 0.19374479918574064,\n",
       " 0.2094399516776467,\n",
       " 0.21994463266357114,\n",
       " 0.22984542411128422,\n",
       " 0.1743536612887455,\n",
       " 0.2615081715133654,\n",
       " 0.21820959478021715,\n",
       " 0.1969586051166155,\n",
       " 0.22820646737330572,\n",
       " 0.23869694012984702,\n",
       " 0.2911111881519574,\n",
       " 0.3645175949275337,\n",
       " 0.5063861561462402,\n",
       " 0.7404538876952609]"
      ]
     },
     "execution_count": 29,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 334
    },
    "colab_type": "code",
    "id": "DZI1FTYftfhd",
    "outputId": "93cabb29-8053-4048-9495-a0206fdb02f5"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.9578,\n",
       " 0.9737,\n",
       " 0.9817,\n",
       " 0.9849,\n",
       " 0.987,\n",
       " 0.9878,\n",
       " 0.9866,\n",
       " 0.9861,\n",
       " 0.9856,\n",
       " 0.989,\n",
       " 0.9832,\n",
       " 0.9858,\n",
       " 0.9876,\n",
       " 0.9854,\n",
       " 0.9847,\n",
       " 0.9817,\n",
       " 0.9772,\n",
       " 0.9681,\n",
       " 0.953]"
      ]
     },
     "execution_count": 30,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 376
    },
    "colab_type": "code",
    "id": "UqrhBfnNthH3",
    "outputId": "ee77bedf-9a71-45a9-e218-2006ebb29bc5"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDMObiIhIYhjeREREEsPwJiIikpj/B88TEw9vURF1AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2fb44cb208>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot( range(-45,50,5), s_loss)\n",
    "plt.title('Test_Loss vs degree')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('degree')\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 376
    },
    "colab_type": "code",
    "id": "H5MilqU-tiwz",
    "outputId": "084c7db1-f68d-47c5-bec2-bdda40c43adb"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfsAAAFnCAYAAAChL+DqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nKexFRERsTmEvIiJic/8Lf+8+9/uD0noAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2fb4382860>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot( range(-45,50,5), s_acc)\n",
    "plt.title('Test_Accuracy vs degree')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('degree')\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 251
    },
    "colab_type": "code",
    "id": "zC1cEzKOtklq",
    "outputId": "5214b2be-9066-47db-d832-674b65ec2ef3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10000/10000 [==============================] - 4s 406us/step\n",
      "Filter radius: 0 Score:[0.1743536612887455, 0.989]\n",
      "10000/10000 [==============================] - 4s 382us/step\n",
      "Filter radius: 1 Score:[0.17037417346491332, 0.989]\n",
      "10000/10000 [==============================] - 4s 379us/step\n",
      "Filter radius: 2 Score:[0.665875063097405, 0.9559]\n",
      "10000/10000 [==============================] - 4s 380us/step\n",
      "Filter radius: 3 Score:[3.5227216413497926, 0.7512]\n",
      "10000/10000 [==============================] - 4s 380us/step\n",
      "Filter radius: 4 Score:[6.652305493927002, 0.4918]\n",
      "10000/10000 [==============================] - 4s 384us/step\n",
      "Filter radius: 5 Score:[7.920265014648438, 0.3377]\n",
      "10000/10000 [==============================] - 4s 384us/step\n",
      "Filter radius: 6 Score:[8.107508961486817, 0.2369]\n"
     ]
    }
   ],
   "source": [
    "from PIL import ImageFilter\n",
    "s_loss = []\n",
    "s_acc = []\n",
    "for j in range(0,7):\n",
    "  filter_test_x = []\n",
    "  for i in range(len(X_test)):\n",
    "    X_test_filter = Image.fromarray(X_test[i])\n",
    "    new_image = X_test_filter.filter(ImageFilter.GaussianBlur(radius=j))\n",
    "    img_array = numpy.array(new_image)\n",
    "    filter_test_x.append(img_array)\n",
    "  X_test_filter = numpy.asarray(filter_test_x)\n",
    "  X_test_filter = X_test_filter.reshape(X_test_filter.shape[0], img_rows, img_cols, 1)\n",
    "  score = model.evaluate(X_test_filter, y_test, batch_size=32)\n",
    "  print(\"Filter radius: \" +str(j) + \" Score:\" + str(score))\n",
    "  s_loss.append(score[0])\n",
    "  s_acc.append(score[1])\n",
    "  #We can see accuracy doesnt improve in this case as the data augmentation was done only for rotations "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 376
    },
    "colab_type": "code",
    "id": "jIE7kCoUtmj4",
    "outputId": "15180d2d-6bff-4818-b2c9-00d8b5dd6693"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAecAAAFnCAYAAACcvYGMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n5UxERCQzLGciIiKZ+f9ZjQw4lpLIjgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2fb43a24a8>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot( range(0,7), s_loss)\n",
    "plt.title('Test_Loss vs Filter_radius')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Filter_radius')\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 376
    },
    "colab_type": "code",
    "id": "8bvKAl5ytvFW",
    "outputId": "5b9e1a8e-f4ce-4117-b68f-c4751f08fd31"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n5U1ERGRkWN5ERERG5v8D6iFt3rO7ObQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f2fb43a2b70>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot( range(0,7), s_acc)\n",
    "plt.title('Test_Accuracy vs Filter_radius')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('Filter_radius')\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 0,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "wDcK46uJtyYD"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "Q1.4-b.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
}