CIFAR10-CNN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From C:\\Users\\Arnav\\Anaconda3\\envs\\gpu\\lib\\site-packages\\tensorflow_core\\python\\compat\\v2_compat.py:88: disable_resource_variables (from tensorflow.python.ops.variable_scope) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "non-resource variables are not supported in the long term\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import tensorflow.compat.v1 as tf\n",
    "tf.disable_v2_behavior()\n",
    "import numpy as np\n",
    "import matplotlib\n",
    "from matplotlib import pyplot as plt\n",
    "from keras.models import Sequential\n",
    "from keras.optimizers import Adam, SGD\n",
    "from keras.callbacks import ModelCheckpoint\n",
    "from keras.constraints import maxnorm\n",
    "from keras.models import load_model\n",
    "from keras.layers import GlobalAveragePooling2D, Lambda, Conv2D, MaxPooling2D, Dropout, Dense, Flatten, Activation\n",
    "from keras.preprocessing.image import ImageDataGenerator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "tf.compat.v1.enable_eager_execution(\n",
    "    config=None, device_policy=None, execution_mode=None\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from helper import get_class_names, get_train_data, get_test_data, plot_images\n",
    "from helper import plot_model, predict_classes, visualize_errors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "matplotlib.style.use('ggplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Decoding file: data/batches.meta\n",
      "['airplane', 'automobile', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck']\n"
     ]
    }
   ],
   "source": [
    "class_names = get_class_names()\n",
    "print(class_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10\n"
     ]
    }
   ],
   "source": [
    "num_classes = len(class_names)\n",
    "print(num_classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Hight and width of the images\n",
    "IMAGE_SIZE = 32\n",
    "# 3 channels, Red, Green and Blue\n",
    "CHANNELS = 3\n",
    "# Number of epochs\n",
    "NUM_EPOCH = 110\n",
    "# learning rate\n",
    "LEARN_RATE = 1.0e-4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Decoding file: data/data_batch_1\n",
      "Decoding file: data/data_batch_2\n",
      "Decoding file: data/data_batch_3\n",
      "Decoding file: data/data_batch_4\n",
      "Decoding file: data/data_batch_5\n"
     ]
    }
   ],
   "source": [
    "images_train, labels_train, class_train = get_train_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Decoding file: data/test_batch\n"
     ]
    }
   ],
   "source": [
    "images_test, labels_test, class_test = get_test_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set size:\t 50000\n",
      "Testing set size:\t 10000\n"
     ]
    }
   ],
   "source": [
    "print(\"Training set size:\\t\",len(images_train))\n",
    "print(\"Testing set size:\\t\",len(images_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def pure_cnn_model():\n",
    "    \n",
    "    model = Sequential()\n",
    "    \n",
    "    model.add(Conv2D(96, (3, 3), activation='relu', padding = 'same', input_shape=(IMAGE_SIZE,IMAGE_SIZE,CHANNELS)))    \n",
    "    model.add(Dropout(0.2))\n",
    "    \n",
    "    model.add(Conv2D(96, (3, 3), activation='relu', padding = 'same'))  \n",
    "    model.add(Conv2D(96, (3, 3), activation='relu', padding = 'same', strides = 2))    \n",
    "    model.add(Dropout(0.5))\n",
    "    \n",
    "    model.add(Conv2D(192, (3, 3), activation='relu', padding = 'same'))    \n",
    "    model.add(Conv2D(192, (3, 3), activation='relu', padding = 'same'))\n",
    "    model.add(Conv2D(192, (3, 3), activation='relu', padding = 'same', strides = 2))    \n",
    "    model.add(Dropout(0.5))    \n",
    "    \n",
    "    model.add(Conv2D(192, (3, 3), padding = 'same'))\n",
    "    model.add(Activation('relu'))\n",
    "    model.add(Conv2D(192, (1, 1),padding='valid'))\n",
    "    model.add(Activation('relu'))\n",
    "    model.add(Conv2D(10, (1, 1), padding='valid'))\n",
    "\n",
    "    model.add(GlobalAveragePooling2D())\n",
    "    \n",
    "    model.add(Activation('softmax'))\n",
    "\n",
    "    model.summary()\n",
    "    \n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_1\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_1 (Conv2D)            (None, 32, 32, 96)        2688      \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 32, 32, 96)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_2 (Conv2D)            (None, 32, 32, 96)        83040     \n",
      "_________________________________________________________________\n",
      "conv2d_3 (Conv2D)            (None, 16, 16, 96)        83040     \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 16, 16, 96)        0         \n",
      "_________________________________________________________________\n",
      "conv2d_4 (Conv2D)            (None, 16, 16, 192)       166080    \n",
      "_________________________________________________________________\n",
      "conv2d_5 (Conv2D)            (None, 16, 16, 192)       331968    \n",
      "_________________________________________________________________\n",
      "conv2d_6 (Conv2D)            (None, 8, 8, 192)         331968    \n",
      "_________________________________________________________________\n",
      "dropout_3 (Dropout)          (None, 8, 8, 192)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_7 (Conv2D)            (None, 8, 8, 192)         331968    \n",
      "_________________________________________________________________\n",
      "activation_1 (Activation)    (None, 8, 8, 192)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_8 (Conv2D)            (None, 8, 8, 192)         37056     \n",
      "_________________________________________________________________\n",
      "activation_2 (Activation)    (None, 8, 8, 192)         0         \n",
      "_________________________________________________________________\n",
      "conv2d_9 (Conv2D)            (None, 8, 8, 10)          1930      \n",
      "_________________________________________________________________\n",
      "global_average_pooling2d_1 ( (None, 10)                0         \n",
      "_________________________________________________________________\n",
      "activation_3 (Activation)    (None, 10)                0         \n",
      "=================================================================\n",
      "Total params: 1,369,738\n",
      "Trainable params: 1,369,738\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model = pure_cnn_model()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "checkpoint = ModelCheckpoint('best_model_improved.h5', \n",
    "                             monitor='val_loss', \n",
    "                             verbose=0, \n",
    "                             save_best_only= True,\n",
    "                             mode='auto')  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(loss='categorical_crossentropy', \n",
    "              optimizer=Adam(lr=LEARN_RATE),\n",
    "              metrics = ['accuracy']) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 50000 samples, validate on 10000 samples\n",
      "Epoch 1/110\n",
      "50000/50000 [==============================] - 61s 1ms/step - loss: 1.9836 - accuracy: 0.2394 - val_loss: 1.7870 - val_accuracy: 0.3341\n",
      "Epoch 2/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 1.6825 - accuracy: 0.3686 - val_loss: 1.6742 - val_accuracy: 0.3853\n",
      "Epoch 3/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 1.5722 - accuracy: 0.4170 - val_loss: 1.4860 - val_accuracy: 0.4593\n",
      "Epoch 4/110\n",
      "50000/50000 [==============================] - 61s 1ms/step - loss: 1.4976 - accuracy: 0.4486 - val_loss: 1.4560 - val_accuracy: 0.4625\n",
      "Epoch 5/110\n",
      "50000/50000 [==============================] - 60s 1ms/step - loss: 1.4422 - accuracy: 0.4705 - val_loss: 1.3735 - val_accuracy: 0.5011\n",
      "Epoch 6/110\n",
      "50000/50000 [==============================] - 1079s 22ms/step - loss: 1.3849 - accuracy: 0.4938 - val_loss: 1.3202 - val_accuracy: 0.5248\n",
      "Epoch 7/110\n",
      "50000/50000 [==============================] - 56s 1ms/step - loss: 1.3415 - accuracy: 0.5131 - val_loss: 1.2716 - val_accuracy: 0.5373\n",
      "Epoch 8/110\n",
      "50000/50000 [==============================] - 56s 1ms/step - loss: 1.2961 - accuracy: 0.5305 - val_loss: 1.2457 - val_accuracy: 0.5509\n",
      "Epoch 9/110\n",
      "50000/50000 [==============================] - 60s 1ms/step - loss: 1.2537 - accuracy: 0.5493 - val_loss: 1.1997 - val_accuracy: 0.5705\n",
      "Epoch 10/110\n",
      "50000/50000 [==============================] - 79s 2ms/step - loss: 1.2115 - accuracy: 0.5648 - val_loss: 1.1582 - val_accuracy: 0.5919\n",
      "Epoch 11/110\n",
      "50000/50000 [==============================] - 62s 1ms/step - loss: 1.1794 - accuracy: 0.5781 - val_loss: 1.1367 - val_accuracy: 0.5927\n",
      "Epoch 12/110\n",
      "50000/50000 [==============================] - 60s 1ms/step - loss: 1.1457 - accuracy: 0.5897 - val_loss: 1.1350 - val_accuracy: 0.5930\n",
      "Epoch 13/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 1.1198 - accuracy: 0.6001 - val_loss: 1.0886 - val_accuracy: 0.6144\n",
      "Epoch 14/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 1.1086 - accuracy: 0.6039 - val_loss: 1.0776 - val_accuracy: 0.6115\n",
      "Epoch 15/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 1.0863 - accuracy: 0.6152 - val_loss: 1.0669 - val_accuracy: 0.6216\n",
      "Epoch 16/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 1.0592 - accuracy: 0.6213 - val_loss: 1.0359 - val_accuracy: 0.6401\n",
      "Epoch 17/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 1.0383 - accuracy: 0.6320 - val_loss: 1.0009 - val_accuracy: 0.6456\n",
      "Epoch 18/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 1.0205 - accuracy: 0.6363 - val_loss: 0.9919 - val_accuracy: 0.6502\n",
      "Epoch 19/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 0.9998 - accuracy: 0.6451 - val_loss: 0.9942 - val_accuracy: 0.6504\n",
      "Epoch 20/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 0.9863 - accuracy: 0.6496 - val_loss: 0.9537 - val_accuracy: 0.6652\n",
      "Epoch 21/110\n",
      "50000/50000 [==============================] - 61s 1ms/step - loss: 0.9682 - accuracy: 0.6573 - val_loss: 0.9428 - val_accuracy: 0.6674\n",
      "Epoch 22/110\n",
      "50000/50000 [==============================] - 60s 1ms/step - loss: 0.9501 - accuracy: 0.6658 - val_loss: 0.9397 - val_accuracy: 0.6682\n",
      "Epoch 23/110\n",
      "50000/50000 [==============================] - 60s 1ms/step - loss: 0.9297 - accuracy: 0.6714 - val_loss: 0.9180 - val_accuracy: 0.6791\n",
      "Epoch 24/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.9239 - accuracy: 0.6753 - val_loss: 0.9038 - val_accuracy: 0.6833\n",
      "Epoch 25/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.9098 - accuracy: 0.6803 - val_loss: 0.8944 - val_accuracy: 0.6903\n",
      "Epoch 26/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.8900 - accuracy: 0.6860 - val_loss: 0.8743 - val_accuracy: 0.6957\n",
      "Epoch 27/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.8733 - accuracy: 0.6919 - val_loss: 0.8633 - val_accuracy: 0.6975\n",
      "Epoch 28/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 0.8674 - accuracy: 0.6953 - val_loss: 0.8640 - val_accuracy: 0.6981\n",
      "Epoch 29/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.8520 - accuracy: 0.6993 - val_loss: 0.8901 - val_accuracy: 0.6871\n",
      "Epoch 30/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 0.8432 - accuracy: 0.7054 - val_loss: 0.8261 - val_accuracy: 0.7111\n",
      "Epoch 31/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.8266 - accuracy: 0.7120 - val_loss: 0.8389 - val_accuracy: 0.7044\n",
      "Epoch 32/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.8208 - accuracy: 0.7099 - val_loss: 0.8464 - val_accuracy: 0.7042\n",
      "Epoch 33/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.8024 - accuracy: 0.7167 - val_loss: 0.8580 - val_accuracy: 0.7035\n",
      "Epoch 34/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.7970 - accuracy: 0.7216 - val_loss: 0.7928 - val_accuracy: 0.7227\n",
      "Epoch 35/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.7879 - accuracy: 0.7247 - val_loss: 0.7920 - val_accuracy: 0.7226\n",
      "Epoch 36/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.7755 - accuracy: 0.7286 - val_loss: 0.7828 - val_accuracy: 0.7223\n",
      "Epoch 37/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.7661 - accuracy: 0.7306 - val_loss: 0.8214 - val_accuracy: 0.7120\n",
      "Epoch 38/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 0.7572 - accuracy: 0.7338 - val_loss: 0.7792 - val_accuracy: 0.7281\n",
      "Epoch 39/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 0.7451 - accuracy: 0.7391 - val_loss: 0.7809 - val_accuracy: 0.7248\n",
      "Epoch 40/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 0.7402 - accuracy: 0.7418 - val_loss: 0.7565 - val_accuracy: 0.7380\n",
      "Epoch 41/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 0.7317 - accuracy: 0.7435 - val_loss: 0.7702 - val_accuracy: 0.7300\n",
      "Epoch 42/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.7222 - accuracy: 0.7467 - val_loss: 0.7415 - val_accuracy: 0.7397\n",
      "Epoch 43/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 0.7115 - accuracy: 0.7493 - val_loss: 0.7287 - val_accuracy: 0.7439\n",
      "Epoch 44/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.6994 - accuracy: 0.7547 - val_loss: 0.7254 - val_accuracy: 0.7459\n",
      "Epoch 45/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.6997 - accuracy: 0.7552 - val_loss: 0.7189 - val_accuracy: 0.7457\n",
      "Epoch 46/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.6846 - accuracy: 0.7588 - val_loss: 0.7542 - val_accuracy: 0.7383\n",
      "Epoch 47/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.6799 - accuracy: 0.7639 - val_loss: 0.7243 - val_accuracy: 0.7458\n",
      "Epoch 48/110\n",
      "50000/50000 [==============================] - 59s 1ms/step - loss: 0.6714 - accuracy: 0.7645 - val_loss: 0.7040 - val_accuracy: 0.7518\n",
      "Epoch 49/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.6668 - accuracy: 0.7652 - val_loss: 0.7208 - val_accuracy: 0.7512\n",
      "Epoch 50/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.6616 - accuracy: 0.7688 - val_loss: 0.7191 - val_accuracy: 0.7468\n",
      "Epoch 51/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.6511 - accuracy: 0.7711 - val_loss: 0.6814 - val_accuracy: 0.7605\n",
      "Epoch 52/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.6436 - accuracy: 0.7758 - val_loss: 0.7017 - val_accuracy: 0.7544\n",
      "Epoch 53/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.6321 - accuracy: 0.7793 - val_loss: 0.6742 - val_accuracy: 0.7685\n",
      "Epoch 54/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.6283 - accuracy: 0.7794 - val_loss: 0.7025 - val_accuracy: 0.7479\n",
      "Epoch 55/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.6228 - accuracy: 0.7812 - val_loss: 0.6795 - val_accuracy: 0.7613\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 56/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.6135 - accuracy: 0.7840 - val_loss: 0.6815 - val_accuracy: 0.7637\n",
      "Epoch 57/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.6073 - accuracy: 0.7883 - val_loss: 0.6548 - val_accuracy: 0.7708\n",
      "Epoch 58/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.6048 - accuracy: 0.7893 - val_loss: 0.6564 - val_accuracy: 0.7712\n",
      "Epoch 59/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5977 - accuracy: 0.7902 - val_loss: 0.6543 - val_accuracy: 0.7716\n",
      "Epoch 60/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5910 - accuracy: 0.7923 - val_loss: 0.6780 - val_accuracy: 0.7637\n",
      "Epoch 61/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5822 - accuracy: 0.7960 - val_loss: 0.6529 - val_accuracy: 0.7714\n",
      "Epoch 62/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5724 - accuracy: 0.7998 - val_loss: 0.6418 - val_accuracy: 0.7770\n",
      "Epoch 63/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5708 - accuracy: 0.8012 - val_loss: 0.6317 - val_accuracy: 0.7790\n",
      "Epoch 64/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5636 - accuracy: 0.8014 - val_loss: 0.6480 - val_accuracy: 0.7723\n",
      "Epoch 65/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5588 - accuracy: 0.8048 - val_loss: 0.6554 - val_accuracy: 0.7710\n",
      "Epoch 66/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5543 - accuracy: 0.8046 - val_loss: 0.6220 - val_accuracy: 0.7799\n",
      "Epoch 67/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5512 - accuracy: 0.8065 - val_loss: 0.6373 - val_accuracy: 0.7803\n",
      "Epoch 68/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5434 - accuracy: 0.8091 - val_loss: 0.6237 - val_accuracy: 0.7837\n",
      "Epoch 69/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5318 - accuracy: 0.8116 - val_loss: 0.6264 - val_accuracy: 0.7793\n",
      "Epoch 70/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5286 - accuracy: 0.8155 - val_loss: 0.6077 - val_accuracy: 0.7863\n",
      "Epoch 71/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5210 - accuracy: 0.8177 - val_loss: 0.6161 - val_accuracy: 0.7865\n",
      "Epoch 72/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5159 - accuracy: 0.8196 - val_loss: 0.6141 - val_accuracy: 0.7886\n",
      "Epoch 73/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5146 - accuracy: 0.8188 - val_loss: 0.6039 - val_accuracy: 0.7890\n",
      "Epoch 74/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5097 - accuracy: 0.8206 - val_loss: 0.5957 - val_accuracy: 0.7943\n",
      "Epoch 75/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.5010 - accuracy: 0.8244 - val_loss: 0.6188 - val_accuracy: 0.7862\n",
      "Epoch 76/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.4948 - accuracy: 0.8266 - val_loss: 0.6194 - val_accuracy: 0.7859\n",
      "Epoch 77/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.4945 - accuracy: 0.8260 - val_loss: 0.6030 - val_accuracy: 0.7930\n",
      "Epoch 78/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.4857 - accuracy: 0.8298 - val_loss: 0.6241 - val_accuracy: 0.7881\n",
      "Epoch 79/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.4860 - accuracy: 0.8299 - val_loss: 0.6116 - val_accuracy: 0.7862\n",
      "Epoch 80/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.4803 - accuracy: 0.8318 - val_loss: 0.5977 - val_accuracy: 0.7953\n",
      "Epoch 81/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.4721 - accuracy: 0.8342 - val_loss: 0.6144 - val_accuracy: 0.7905\n",
      "Epoch 82/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.4640 - accuracy: 0.8357 - val_loss: 0.6185 - val_accuracy: 0.7899\n",
      "Epoch 83/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.4627 - accuracy: 0.8367 - val_loss: 0.6052 - val_accuracy: 0.7976\n",
      "Epoch 84/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.4581 - accuracy: 0.8384 - val_loss: 0.5966 - val_accuracy: 0.7934\n",
      "Epoch 85/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.4575 - accuracy: 0.8390 - val_loss: 0.5696 - val_accuracy: 0.8006\n",
      "Epoch 86/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.4486 - accuracy: 0.8428 - val_loss: 0.5766 - val_accuracy: 0.8047\n",
      "Epoch 87/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.4453 - accuracy: 0.8435 - val_loss: 0.5833 - val_accuracy: 0.8035\n",
      "Epoch 88/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.4396 - accuracy: 0.8441 - val_loss: 0.5715 - val_accuracy: 0.8073\n",
      "Epoch 89/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.4340 - accuracy: 0.8482 - val_loss: 0.5980 - val_accuracy: 0.8009\n",
      "Epoch 90/110\n",
      "50000/50000 [==============================] - 60s 1ms/step - loss: 0.4322 - accuracy: 0.8470 - val_loss: 0.5827 - val_accuracy: 0.8046\n",
      "Epoch 91/110\n",
      "50000/50000 [==============================] - 60s 1ms/step - loss: 0.4279 - accuracy: 0.8488 - val_loss: 0.5881 - val_accuracy: 0.8068\n",
      "Epoch 92/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.4279 - accuracy: 0.8491 - val_loss: 0.5900 - val_accuracy: 0.7980\n",
      "Epoch 93/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.4201 - accuracy: 0.8515 - val_loss: 0.5774 - val_accuracy: 0.8031\n",
      "Epoch 94/110\n",
      "50000/50000 [==============================] - 57s 1ms/step - loss: 0.4136 - accuracy: 0.8547 - val_loss: 0.5839 - val_accuracy: 0.8066\n",
      "Epoch 95/110\n",
      "50000/50000 [==============================] - 58s 1ms/step - loss: 0.4133 - accuracy: 0.8543 - val_loss: 0.5643 - val_accuracy: 0.8073\n",
      "Epoch 96/110\n",
      "50000/50000 [==============================] - 60s 1ms/step - loss: 0.4082 - accuracy: 0.8558 - val_loss: 0.5637 - val_accuracy: 0.8085\n",
      "Epoch 97/110\n",
      "50000/50000 [==============================] - 63s 1ms/step - loss: 0.4041 - accuracy: 0.8566 - val_loss: 0.5789 - val_accuracy: 0.8083\n",
      "Epoch 98/110\n",
      "50000/50000 [==============================] - 64s 1ms/step - loss: 0.3990 - accuracy: 0.8588 - val_loss: 0.5717 - val_accuracy: 0.8046\n",
      "Epoch 99/110\n",
      "50000/50000 [==============================] - 63s 1ms/step - loss: 0.3950 - accuracy: 0.8608 - val_loss: 0.5757 - val_accuracy: 0.8055\n",
      "Epoch 100/110\n",
      "50000/50000 [==============================] - 64s 1ms/step - loss: 0.3895 - accuracy: 0.8635 - val_loss: 0.5721 - val_accuracy: 0.8090\n",
      "Epoch 101/110\n",
      "50000/50000 [==============================] - 64s 1ms/step - loss: 0.3920 - accuracy: 0.8610 - val_loss: 0.5775 - val_accuracy: 0.8105\n",
      "Epoch 102/110\n",
      "50000/50000 [==============================] - 64s 1ms/step - loss: 0.3837 - accuracy: 0.8639 - val_loss: 0.5503 - val_accuracy: 0.8139\n",
      "Epoch 103/110\n",
      "50000/50000 [==============================] - 66s 1ms/step - loss: 0.3804 - accuracy: 0.8652 - val_loss: 0.5749 - val_accuracy: 0.8106\n",
      "Epoch 104/110\n",
      "50000/50000 [==============================] - 65s 1ms/step - loss: 0.3744 - accuracy: 0.8675 - val_loss: 0.5596 - val_accuracy: 0.8145\n",
      "Epoch 105/110\n",
      "50000/50000 [==============================] - 65s 1ms/step - loss: 0.3716 - accuracy: 0.8697 - val_loss: 0.5662 - val_accuracy: 0.8126\n",
      "Epoch 106/110\n",
      "50000/50000 [==============================] - 66s 1ms/step - loss: 0.3690 - accuracy: 0.8681 - val_loss: 0.5725 - val_accuracy: 0.8069\n",
      "Epoch 107/110\n",
      "50000/50000 [==============================] - 65s 1ms/step - loss: 0.3638 - accuracy: 0.8699 - val_loss: 0.5666 - val_accuracy: 0.8147\n",
      "Epoch 108/110\n",
      "50000/50000 [==============================] - 65s 1ms/step - loss: 0.3629 - accuracy: 0.8729 - val_loss: 0.5819 - val_accuracy: 0.8060\n",
      "Epoch 109/110\n",
      "50000/50000 [==============================] - 65s 1ms/step - loss: 0.3590 - accuracy: 0.8716 - val_loss: 0.5733 - val_accuracy: 0.8099\n",
      "Epoch 110/110\n",
      "50000/50000 [==============================] - 65s 1ms/step - loss: 0.3551 - accuracy: 0.8742 - val_loss: 0.5618 - val_accuracy: 0.8198\n"
     ]
    }
   ],
   "source": [
    "model_details = model.fit(images_train, class_train,\n",
    "                    batch_size = 128,\n",
    "                    epochs = NUM_EPOCH, # number of iterations\n",
    "                    validation_data= (images_test, class_test),\n",
    "                    callbacks=[checkpoint],\n",
    "                    verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 81.98%\n"
     ]
    }
   ],
   "source": [
    "scores = model.evaluate(images_test, class_test, verbose=0)\n",
    "print(\"Accuracy: %.2f%%\" % (scores[1]*100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "correct, labels_pred = predict_classes(model, images_test, labels_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 81.98%\n"
     ]
    }
   ],
   "source": [
    "num_images = len(correct)\n",
    "print(\"Accuracy: %.2f%%\" % ((sum(correct)*100)/num_images))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "ans = []\n",
    "for i in range(10000):\n",
    "    ans.append(class_names[labels_pred[i]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['dog',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'airplane',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'automobile',\n",
       " 'frog',\n",
       " 'cat',\n",
       " 'automobile',\n",
       " 'airplane',\n",
       " 'truck',\n",
       " 'dog',\n",
       " 'horse',\n",
       " 'truck',\n",
       " 'ship',\n",
       " 'dog',\n",
       " 'horse',\n",
       " 'ship',\n",
       " 'frog',\n",
       " 'automobile',\n",
       " 'airplane',\n",
       " 'deer',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'bird',\n",
       " 'deer',\n",
       " 'airplane',\n",
       " 'truck',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'dog',\n",
       " 'deer',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'automobile',\n",
       " 'deer',\n",
       " 'truck',\n",
       " 'truck',\n",
       " 'dog',\n",
       " 'deer',\n",
       " 'frog',\n",
       " 'dog',\n",
       " 'frog',\n",
       " 'airplane',\n",
       " 'truck',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'horse',\n",
       " 'frog',\n",
       " 'truck',\n",
       " 'ship',\n",
       " 'horse',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'horse',\n",
       " 'cat',\n",
       " 'cat',\n",
       " 'dog',\n",
       " 'horse',\n",
       " 'dog',\n",
       " 'frog',\n",
       " 'cat',\n",
       " 'frog',\n",
       " 'bird',\n",
       " 'automobile',\n",
       " 'bird',\n",
       " 'cat',\n",
       " 'horse',\n",
       " 'bird',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'airplane',\n",
       " 'bird',\n",
       " 'truck',\n",
       " 'cat',\n",
       " 'dog',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'automobile',\n",
       " 'automobile',\n",
       " 'horse',\n",
       " 'bird',\n",
       " 'dog',\n",
       " 'airplane',\n",
       " 'airplane',\n",
       " 'ship',\n",
       " 'truck',\n",
       " 'airplane',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'frog',\n",
       " 'deer',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'airplane',\n",
       " 'airplane',\n",
       " 'horse',\n",
       " 'deer',\n",
       " 'dog',\n",
       " 'frog',\n",
       " 'cat',\n",
       " 'automobile',\n",
       " 'automobile',\n",
       " 'bird',\n",
       " 'frog',\n",
       " 'ship',\n",
       " 'horse',\n",
       " 'deer',\n",
       " 'airplane',\n",
       " 'frog',\n",
       " 'bird',\n",
       " 'automobile',\n",
       " 'cat',\n",
       " 'airplane',\n",
       " 'deer',\n",
       " 'cat',\n",
       " 'horse',\n",
       " 'ship',\n",
       " 'cat',\n",
       " 'automobile',\n",
       " 'bird',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'cat',\n",
       " 'frog',\n",
       " 'bird',\n",
       " 'deer',\n",
       " 'automobile',\n",
       " 'ship',\n",
       " 'truck',\n",
       " 'automobile',\n",
       " 'bird',\n",
       " 'truck',\n",
       " 'horse',\n",
       " 'bird',\n",
       " 'automobile',\n",
       " 'frog',\n",
       " 'dog',\n",
       " 'frog',\n",
       " 'dog',\n",
       " 'ship',\n",
       " 'bird',\n",
       " 'frog',\n",
       " 'dog',\n",
       " 'dog',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'truck',\n",
       " 'cat',\n",
       " 'airplane',\n",
       " 'airplane',\n",
       " 'dog',\n",
       " 'bird',\n",
       " 'truck',\n",
       " 'horse',\n",
       " 'deer',\n",
       " 'bird',\n",
       " 'automobile',\n",
       " 'cat',\n",
       " 'frog',\n",
       " 'airplane',\n",
       " 'deer',\n",
       " 'ship',\n",
       " 'frog',\n",
       " 'dog',\n",
       " 'airplane',\n",
       " 'truck',\n",
       " 'airplane',\n",
       " 'truck',\n",
       " 'ship',\n",
       " 'truck',\n",
       " 'truck',\n",
       " 'frog',\n",
       " 'horse',\n",
       " 'cat',\n",
       " 'airplane',\n",
       " 'airplane',\n",
       " 'dog',\n",
       " 'bird',\n",
       " 'frog',\n",
       " 'dog',\n",
       " 'ship',\n",
       " 'frog',\n",
       " 'cat',\n",
       " 'cat',\n",
       " 'airplane',\n",
       " 'dog',\n",
       " 'ship',\n",
       " 'airplane',\n",
       " 'automobile',\n",
       " 'horse',\n",
       " 'deer',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'horse',\n",
       " 'ship',\n",
       " 'cat',\n",
       " 'automobile',\n",
       " 'ship',\n",
       " 'horse',\n",
       " 'automobile',\n",
       " 'cat',\n",
       " 'airplane',\n",
       " 'dog',\n",
       " 'horse',\n",
       " 'truck',\n",
       " 'cat',\n",
       " 'deer',\n",
       " 'dog',\n",
       " 'truck',\n",
       " 'ship',\n",
       " 'bird',\n",
       " 'horse',\n",
       " 'truck',\n",
       " 'ship',\n",
       " 'bird',\n",
       " 'horse',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'deer',\n",
       " 'truck',\n",
       " 'frog',\n",
       " 'deer',\n",
       " 'cat',\n",
       " 'deer',\n",
       " 'dog',\n",
       " 'automobile',\n",
       " 'dog',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'airplane',\n",
       " 'deer',\n",
       " 'airplane',\n",
       " 'dog',\n",
       " 'dog',\n",
       " 'automobile',\n",
       " 'automobile',\n",
       " 'automobile',\n",
       " 'truck',\n",
       " 'airplane',\n",
       " 'bird',\n",
       " 'automobile',\n",
       " 'ship',\n",
       " 'bird',\n",
       " 'airplane',\n",
       " 'dog',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'airplane',\n",
       " 'cat',\n",
       " 'airplane',\n",
       " 'airplane',\n",
       " 'truck',\n",
       " 'ship',\n",
       " 'automobile',\n",
       " 'dog',\n",
       " 'horse',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'bird',\n",
       " 'deer',\n",
       " 'horse',\n",
       " 'cat',\n",
       " 'bird',\n",
       " 'cat',\n",
       " 'frog',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'dog',\n",
       " 'airplane',\n",
       " 'bird',\n",
       " 'deer',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'automobile',\n",
       " 'frog',\n",
       " 'automobile',\n",
       " 'truck',\n",
       " 'truck',\n",
       " 'automobile',\n",
       " 'ship',\n",
       " 'horse',\n",
       " 'truck',\n",
       " 'automobile',\n",
       " 'bird',\n",
       " 'frog',\n",
       " 'automobile',\n",
       " 'cat',\n",
       " 'deer',\n",
       " 'frog',\n",
       " 'airplane',\n",
       " 'airplane',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'cat',\n",
       " 'bird',\n",
       " 'frog',\n",
       " 'automobile',\n",
       " 'ship',\n",
       " 'bird',\n",
       " 'truck',\n",
       " 'bird',\n",
       " 'automobile',\n",
       " 'frog',\n",
       " 'airplane',\n",
       " 'airplane',\n",
       " 'deer',\n",
       " 'airplane',\n",
       " 'horse',\n",
       " 'horse',\n",
       " 'dog',\n",
       " 'dog',\n",
       " 'cat',\n",
       " 'dog',\n",
       " 'bird',\n",
       " 'cat',\n",
       " 'horse',\n",
       " 'automobile',\n",
       " 'horse',\n",
       " 'dog',\n",
       " 'deer',\n",
       " 'frog',\n",
       " 'automobile',\n",
       " 'truck',\n",
       " 'cat',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'truck',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'airplane',\n",
       " 'automobile',\n",
       " 'deer',\n",
       " 'frog',\n",
       " 'bird',\n",
       " 'dog',\n",
       " 'ship',\n",
       " 'dog',\n",
       " 'deer',\n",
       " 'frog',\n",
       " 'ship',\n",
       " 'truck',\n",
       " 'truck',\n",
       " 'automobile',\n",
       " 'ship',\n",
       " 'bird',\n",
       " 'frog',\n",
       " 'deer',\n",
       " 'cat',\n",
       " 'bird',\n",
       " 'ship',\n",
       " 'airplane',\n",
       " 'truck',\n",
       " 'dog',\n",
       " 'ship',\n",
       " 'automobile',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'automobile',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'automobile',\n",
       " 'bird',\n",
       " 'horse',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'bird',\n",
       " 'horse',\n",
       " 'truck',\n",
       " 'horse',\n",
       " 'ship',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'truck',\n",
       " 'airplane',\n",
       " 'truck',\n",
       " 'dog',\n",
       " 'ship',\n",
       " 'horse',\n",
       " 'bird',\n",
       " 'dog',\n",
       " 'dog',\n",
       " 'automobile',\n",
       " 'bird',\n",
       " 'frog',\n",
       " 'bird',\n",
       " 'truck',\n",
       " 'frog',\n",
       " 'bird',\n",
       " 'cat',\n",
       " 'airplane',\n",
       " 'ship',\n",
       " 'truck',\n",
       " 'automobile',\n",
       " 'horse',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'deer',\n",
       " 'airplane',\n",
       " 'automobile',\n",
       " 'ship',\n",
       " 'bird',\n",
       " 'horse',\n",
       " 'airplane',\n",
       " 'cat',\n",
       " 'frog',\n",
       " 'automobile',\n",
       " 'truck',\n",
       " 'cat',\n",
       " 'horse',\n",
       " 'cat',\n",
       " 'horse',\n",
       " 'deer',\n",
       " 'dog',\n",
       " 'airplane',\n",
       " 'horse',\n",
       " 'bird',\n",
       " 'truck',\n",
       " 'cat',\n",
       " 'horse',\n",
       " 'airplane',\n",
       " 'frog',\n",
       " 'cat',\n",
       " 'dog',\n",
       " 'cat',\n",
       " 'deer',\n",
       " 'horse',\n",
       " 'horse',\n",
       " 'bird',\n",
       " 'dog',\n",
       " 'cat',\n",
       " 'automobile',\n",
       " 'automobile',\n",
       " 'deer',\n",
       " 'truck',\n",
       " 'truck',\n",
       " 'dog',\n",
       " 'horse',\n",
       " 'dog',\n",
       " 'airplane',\n",
       " 'bird',\n",
       " 'airplane',\n",
       " 'truck',\n",
       " 'truck',\n",
       " 'horse',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'deer',\n",
       " 'dog',\n",
       " 'deer',\n",
       " 'dog',\n",
       " 'dog',\n",
       " 'frog',\n",
       " 'automobile',\n",
       " 'deer',\n",
       " 'cat',\n",
       " 'deer',\n",
       " 'deer',\n",
       " 'cat',\n",
       " 'horse',\n",
       " 'ship',\n",
       " 'cat',\n",
       " 'horse',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'dog',\n",
       " 'horse',\n",
       " 'frog',\n",
       " 'airplane',\n",
       " 'dog',\n",
       " 'deer',\n",
       " 'truck',\n",
       " 'frog',\n",
       " 'ship',\n",
       " 'dog',\n",
       " 'dog',\n",
       " 'truck',\n",
       " 'truck',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'airplane',\n",
       " 'automobile',\n",
       " 'airplane',\n",
       " 'ship',\n",
       " 'automobile',\n",
       " 'automobile',\n",
       " 'ship',\n",
       " 'airplane',\n",
       " 'bird',\n",
       " 'bird',\n",
       " 'airplane',\n",
       " 'deer',\n",
       " 'frog',\n",
       " 'cat',\n",
       " 'deer',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'horse',\n",
       " 'truck',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'dog',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'automobile',\n",
       " 'dog',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'truck',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'dog',\n",
       " 'horse',\n",
       " 'airplane',\n",
       " 'horse',\n",
       " 'airplane',\n",
       " 'cat',\n",
       " 'deer',\n",
       " 'airplane',\n",
       " 'deer',\n",
       " 'frog',\n",
       " 'truck',\n",
       " 'ship',\n",
       " 'truck',\n",
       " 'dog',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'truck',\n",
       " 'airplane',\n",
       " 'automobile',\n",
       " 'horse',\n",
       " 'frog',\n",
       " 'horse',\n",
       " 'dog',\n",
       " 'truck',\n",
       " 'automobile',\n",
       " 'frog',\n",
       " 'bird',\n",
       " 'dog',\n",
       " 'dog',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'cat',\n",
       " 'deer',\n",
       " 'cat',\n",
       " 'bird',\n",
       " 'airplane',\n",
       " 'horse',\n",
       " 'frog',\n",
       " 'deer',\n",
       " 'bird',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'bird',\n",
       " 'bird',\n",
       " 'horse',\n",
       " 'automobile',\n",
       " 'cat',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'horse',\n",
       " 'dog',\n",
       " 'deer',\n",
       " 'airplane',\n",
       " 'ship',\n",
       " 'deer',\n",
       " 'airplane',\n",
       " 'truck',\n",
       " 'cat',\n",
       " 'deer',\n",
       " 'ship',\n",
       " 'truck',\n",
       " 'frog',\n",
       " 'truck',\n",
       " 'bird',\n",
       " 'frog',\n",
       " 'automobile',\n",
       " 'deer',\n",
       " 'horse',\n",
       " 'cat',\n",
       " 'cat',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'dog',\n",
       " 'airplane',\n",
       " 'bird',\n",
       " 'automobile',\n",
       " 'frog',\n",
       " 'deer',\n",
       " 'cat',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'frog',\n",
       " 'truck',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'dog',\n",
       " 'ship',\n",
       " 'frog',\n",
       " 'dog',\n",
       " 'bird',\n",
       " 'automobile',\n",
       " 'horse',\n",
       " 'horse',\n",
       " 'automobile',\n",
       " 'deer',\n",
       " 'horse',\n",
       " 'truck',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'deer',\n",
       " 'automobile',\n",
       " 'airplane',\n",
       " 'bird',\n",
       " 'frog',\n",
       " 'ship',\n",
       " 'horse',\n",
       " 'frog',\n",
       " 'ship',\n",
       " 'bird',\n",
       " 'airplane',\n",
       " 'dog',\n",
       " 'dog',\n",
       " 'cat',\n",
       " 'airplane',\n",
       " 'horse',\n",
       " 'truck',\n",
       " 'automobile',\n",
       " 'cat',\n",
       " 'deer',\n",
       " 'deer',\n",
       " 'dog',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'dog',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'automobile',\n",
       " 'automobile',\n",
       " 'deer',\n",
       " 'automobile',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'horse',\n",
       " 'frog',\n",
       " 'cat',\n",
       " 'airplane',\n",
       " 'truck',\n",
       " 'airplane',\n",
       " 'automobile',\n",
       " 'bird',\n",
       " 'frog',\n",
       " 'horse',\n",
       " 'frog',\n",
       " 'cat',\n",
       " 'bird',\n",
       " 'airplane',\n",
       " 'deer',\n",
       " 'automobile',\n",
       " 'airplane',\n",
       " 'dog',\n",
       " 'truck',\n",
       " 'frog',\n",
       " 'airplane',\n",
       " 'automobile',\n",
       " 'truck',\n",
       " 'bird',\n",
       " 'truck',\n",
       " 'bird',\n",
       " 'cat',\n",
       " 'frog',\n",
       " 'cat',\n",
       " 'bird',\n",
       " 'airplane',\n",
       " 'horse',\n",
       " 'ship',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'cat',\n",
       " 'horse',\n",
       " 'dog',\n",
       " 'horse',\n",
       " 'airplane',\n",
       " 'deer',\n",
       " 'ship',\n",
       " 'horse',\n",
       " 'deer',\n",
       " 'bird',\n",
       " 'truck',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'bird',\n",
       " 'airplane',\n",
       " 'ship',\n",
       " 'horse',\n",
       " 'deer',\n",
       " 'cat',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'automobile',\n",
       " 'ship',\n",
       " 'automobile',\n",
       " 'automobile',\n",
       " 'dog',\n",
       " 'frog',\n",
       " 'dog',\n",
       " 'deer',\n",
       " 'bird',\n",
       " 'horse',\n",
       " 'truck',\n",
       " 'automobile',\n",
       " 'bird',\n",
       " 'automobile',\n",
       " 'deer',\n",
       " 'automobile',\n",
       " 'dog',\n",
       " 'dog',\n",
       " 'horse',\n",
       " 'airplane',\n",
       " 'horse',\n",
       " 'truck',\n",
       " 'horse',\n",
       " 'deer',\n",
       " 'frog',\n",
       " 'bird',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'airplane',\n",
       " 'truck',\n",
       " 'automobile',\n",
       " 'bird',\n",
       " 'bird',\n",
       " 'frog',\n",
       " 'ship',\n",
       " 'bird',\n",
       " 'automobile',\n",
       " 'cat',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'airplane',\n",
       " 'automobile',\n",
       " 'bird',\n",
       " 'horse',\n",
       " 'truck',\n",
       " 'dog',\n",
       " 'deer',\n",
       " 'airplane',\n",
       " 'automobile',\n",
       " 'frog',\n",
       " 'deer',\n",
       " 'airplane',\n",
       " 'cat',\n",
       " 'bird',\n",
       " 'frog',\n",
       " 'airplane',\n",
       " 'frog',\n",
       " 'truck',\n",
       " 'automobile',\n",
       " 'horse',\n",
       " 'frog',\n",
       " 'horse',\n",
       " 'airplane',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'frog',\n",
       " 'ship',\n",
       " 'airplane',\n",
       " 'airplane',\n",
       " 'frog',\n",
       " 'deer',\n",
       " 'horse',\n",
       " 'horse',\n",
       " 'automobile',\n",
       " 'deer',\n",
       " 'bird',\n",
       " 'deer',\n",
       " 'horse',\n",
       " 'automobile',\n",
       " 'deer',\n",
       " 'horse',\n",
       " 'deer',\n",
       " 'deer',\n",
       " 'ship',\n",
       " 'deer',\n",
       " 'horse',\n",
       " 'horse',\n",
       " 'dog',\n",
       " 'dog',\n",
       " 'horse',\n",
       " 'bird',\n",
       " 'airplane',\n",
       " 'ship',\n",
       " 'truck',\n",
       " 'cat',\n",
       " 'automobile',\n",
       " 'cat',\n",
       " 'frog',\n",
       " 'bird',\n",
       " 'airplane',\n",
       " 'ship',\n",
       " 'horse',\n",
       " 'cat',\n",
       " 'horse',\n",
       " 'frog',\n",
       " 'dog',\n",
       " 'cat',\n",
       " 'automobile',\n",
       " 'cat',\n",
       " 'bird',\n",
       " 'bird',\n",
       " 'dog',\n",
       " 'cat',\n",
       " 'automobile',\n",
       " 'bird',\n",
       " 'truck',\n",
       " 'bird',\n",
       " 'horse',\n",
       " 'airplane',\n",
       " 'horse',\n",
       " 'bird',\n",
       " 'automobile',\n",
       " 'airplane',\n",
       " 'bird',\n",
       " 'airplane',\n",
       " 'bird',\n",
       " 'deer',\n",
       " 'horse',\n",
       " 'truck',\n",
       " 'ship',\n",
       " 'automobile',\n",
       " 'airplane',\n",
       " 'cat',\n",
       " 'horse',\n",
       " 'airplane',\n",
       " 'horse',\n",
       " 'ship',\n",
       " 'deer',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'cat',\n",
       " 'airplane',\n",
       " 'automobile',\n",
       " 'cat',\n",
       " 'cat',\n",
       " 'airplane',\n",
       " 'automobile',\n",
       " 'cat',\n",
       " 'automobile',\n",
       " 'deer',\n",
       " 'bird',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'deer',\n",
       " 'bird',\n",
       " 'cat',\n",
       " 'horse',\n",
       " 'ship',\n",
       " 'deer',\n",
       " 'airplane',\n",
       " 'automobile',\n",
       " 'truck',\n",
       " 'airplane',\n",
       " 'airplane',\n",
       " 'automobile',\n",
       " 'airplane',\n",
       " 'frog',\n",
       " 'deer',\n",
       " 'frog',\n",
       " 'horse',\n",
       " 'cat',\n",
       " 'automobile',\n",
       " 'automobile',\n",
       " 'cat',\n",
       " 'horse',\n",
       " 'cat',\n",
       " 'dog',\n",
       " 'cat',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'dog',\n",
       " 'ship',\n",
       " 'horse',\n",
       " 'automobile',\n",
       " 'frog',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'dog',\n",
       " 'cat',\n",
       " 'cat',\n",
       " 'deer',\n",
       " 'ship',\n",
       " 'automobile',\n",
       " 'deer',\n",
       " 'ship',\n",
       " 'ship',\n",
       " 'airplane',\n",
       " 'frog',\n",
       " 'truck',\n",
       " 'truck',\n",
       " 'cat',\n",
       " 'dog',\n",
       " 'cat',\n",
       " 'bird',\n",
       " 'frog',\n",
       " 'airplane',\n",
       " 'airplane',\n",
       " 'deer',\n",
       " 'bird',\n",
       " 'cat',\n",
       " 'cat',\n",
       " 'horse',\n",
       " 'cat',\n",
       " 'bird',\n",
       " 'dog',\n",
       " 'automobile',\n",
       " 'ship',\n",
       " 'truck',\n",
       " 'automobile',\n",
       " 'horse',\n",
       " 'deer',\n",
       " 'airplane',\n",
       " 'cat',\n",
       " 'airplane',\n",
       " 'automobile',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'frog',\n",
       " 'automobile',\n",
       " 'deer',\n",
       " 'horse',\n",
       " 'bird',\n",
       " 'cat',\n",
       " 'horse',\n",
       " 'ship',\n",
       " 'truck',\n",
       " 'automobile',\n",
       " 'automobile',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'frog',\n",
       " 'cat',\n",
       " 'truck',\n",
       " 'automobile',\n",
       " 'truck',\n",
       " 'truck',\n",
       " 'deer',\n",
       " 'cat',\n",
       " 'automobile',\n",
       " 'bird',\n",
       " 'airplane',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'automobile',\n",
       " 'truck',\n",
       " 'frog',\n",
       " 'truck',\n",
       " 'horse',\n",
       " 'deer',\n",
       " 'horse',\n",
       " 'ship',\n",
       " 'dog',\n",
       " 'automobile',\n",
       " 'horse',\n",
       " 'airplane',\n",
       " 'automobile',\n",
       " 'dog',\n",
       " 'ship',\n",
       " 'horse',\n",
       " 'cat',\n",
       " 'cat',\n",
       " 'ship',\n",
       " 'automobile',\n",
       " 'cat',\n",
       " 'ship',\n",
       " ...]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.savetxt(\"cifar10CNN.csv\",ans, fmt = '%s')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "gpu",
   "language": "python",
   "name": "gpu"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}