CIFAR-10 image classification.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow.keras import layers, models, datasets\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz\n",
      "170500096/170498071 [==============================] - 203s 1us/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(50000, 32, 32, 3)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(X_train, y_train), (X_test, y_test)= datasets.cifar10.load_data()\n",
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 32, 32, 3)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[ 59,  62,  63],\n",
       "        [ 43,  46,  45],\n",
       "        [ 50,  48,  43],\n",
       "        ...,\n",
       "        [158, 132, 108],\n",
       "        [152, 125, 102],\n",
       "        [148, 124, 103]],\n",
       "\n",
       "       [[ 16,  20,  20],\n",
       "        [  0,   0,   0],\n",
       "        [ 18,   8,   0],\n",
       "        ...,\n",
       "        [123,  88,  55],\n",
       "        [119,  83,  50],\n",
       "        [122,  87,  57]],\n",
       "\n",
       "       [[ 25,  24,  21],\n",
       "        [ 16,   7,   0],\n",
       "        [ 49,  27,   8],\n",
       "        ...,\n",
       "        [118,  84,  50],\n",
       "        [120,  84,  50],\n",
       "        [109,  73,  42]],\n",
       "\n",
       "       ...,\n",
       "\n",
       "       [[208, 170,  96],\n",
       "        [201, 153,  34],\n",
       "        [198, 161,  26],\n",
       "        ...,\n",
       "        [160, 133,  70],\n",
       "        [ 56,  31,   7],\n",
       "        [ 53,  34,  20]],\n",
       "\n",
       "       [[180, 139,  96],\n",
       "        [173, 123,  42],\n",
       "        [186, 144,  30],\n",
       "        ...,\n",
       "        [184, 148,  94],\n",
       "        [ 97,  62,  34],\n",
       "        [ 83,  53,  34]],\n",
       "\n",
       "       [[177, 144, 116],\n",
       "        [168, 129,  94],\n",
       "        [179, 142,  87],\n",
       "        ...,\n",
       "        [216, 184, 140],\n",
       "        [151, 118,  84],\n",
       "        [123,  92,  72]]], dtype=uint8)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x272fab5e970>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,2))\n",
    "plt.imshow(X_train[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x272fa8f6190>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,2))\n",
    "plt.imshow(X_train[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50000, 1)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[6],\n",
       "       [9],\n",
       "       [9],\n",
       "       [4],\n",
       "       [1]], dtype=uint8)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([6, 9, 9, ..., 9, 1, 1], dtype=uint8)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train= y_train.reshape(-1,)\n",
    "y_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "classes=['airplane','automobile','bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'bird'"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classes[2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_sample(X, y, index):\n",
    "    plt.figure(figsize=(15,2))\n",
    "    plt.imshow(X[index])\n",
    "    plt.xlabel(classes[y[index]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_sample(X_train, y_train, 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_sample(X_train, y_train, 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAI4AAACcCAYAAACp45OYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAYSklEQVR4nO1daWxc13X+zrxZONxJkZIoStYWVbGz2HVcx3ZiJ45twC3axikcIAaapkAK/2nRBF3QNM2fboCDAImBBEnrJq6FJmiQ1nHjLG3jpnWcxUksW5GsxYssiVopiRIpcvZ5753+mMd3z7kckqMnaUSK9wMInZl73313ns67Z73nEjPDweFikbraE3BYnnCM45AIjnEcEsExjkMiOMZxSATHOA6JcEmMQ0T3E9GrRHSQiD5xuSblsPRBSf04ROQBeA3AfQCOA3gBwEPMvP/yTc9hqSJ9CdfeCuAgMx8CACL6OoD3A5iXcXr7+nl49Uj0STMskVn8UilSbSwWRnkVQfcjlm02ZCM1+3bu+HNeKnGdaJrz6s29ueh88S/qxVyhh5//Sl6olxhk7I39E8w8bHe5FMYZBXBMfD4O4J0LXTC8egSPPPo4ACAMQ9WWz+ViOtvRodpCz7T5bJgoDU/18wJDZ/Tw6mFw2oxRJ/3Y5KdUYD1Szph51E1bkAp0vwUYR67wc1Z7cV0YivFhv0jNx2tcZ354EFjzmmcM3xqD2Yzx0fe/dazZ9Zei4zR7PHNfPqKHiWgnEe2cvjB5CbdzWEq4lBXnOIAN4vN6ACftTsz8GIDHAGDrtus5jNgtncuofrXQvB3FCzOqLdNleNTL5MXgmndDwcu+tZIElXpMVy6UYzrbkdP9YN62Qrmg2lJk+nZ39ZlpQC9voXjTiVpfLeSU5YoT2iuO6hdabWKlWmAeoZhJuMCqNR8uZcV5AcA2ItpMRFkAHwLw9CWM57CMkHjFYWafiP4IwH8D8AA8zsz7LtvMHJY0LkVUgZm/B+B7l2kuDssIl8Q4F4sgDDBdbOgN9XpdtU2cPRfTx0+cUW1eR1dMd/cMxHQupfUTqfLUfD1+WPdjujRjdJd8Ro+BlJHvMzWta9Vq5gZbNm+L6Tdt3aj65YVVaOsL6rNlXrD4IqT57f0FLbN5YOs4KXkvLK7T2HAhB4dEcIzjkAhtFVWFYhE//dnzEW2ZujDmebmql99KYMRYJmtoL9R8H4jVuMK+1WbG7MoaUZIn/Qg6csapGKRqqq1YNOJv555dMX1mQnshtmzeHNNDQ0OqLd/ZGdMcWi4DYT6HwglH1u9M5H22zXbpPW+zOe6wguEYxyERHOM4JEJ7zfEgxFSh4e5nK1xAwuZMZ3U4olPoIV7K0FlkVb8KjI7gW+/ETKkY0+WioXOkA6XdbMxzz3o6mZwJd1QKlZh+49gJ1W/s1HhM9/f2qbYN69fH9PDQKtXWP2BcDemUmZfH84cVbMi4bIj59RgZyJwbclhch3IrjkMiOMZxSIS2iqqQGeVaY4nMZOxbi2U10F5fhvlMIunGCoCjVjfio24N39PZHdMz06WYnq6VVb+qMEWzWS0Ke7Lmhp5n2op+VfWTboLqxAXVNjVl3BBd3XnVNjKyLqa3bt4S091Z7d3OiXnZHvi6kGos8pXCBcSdLfnsNKRmcCuOQyI4xnFIhPaLqmpDnFTrmmdlEK7DSh1VyU8yvdKSVfJz0fJMd+TNhbmM8A7XtXVXqRrR5ZPtbTXjZ4XVM/f1ExZiWlttcoyZkp7jhdcPxPTEuYmY7unQltn6UWOZDQhLDACyOSn+RCDT1550X/w02wINeP6U01m4FcchERzjOCSCYxyHRGirjsPMqEVmIQXzJziFqQX2l+SE2e5pvg9TRo6nrV9WF2Z3Nm10qO68NrlLNWPS+9B6gQzaV33zIZfSN/OEGczWu1kPzZg+tC6RSpm+4+dNMtvJ6jnV7+DY0ZgeHtbR93XrzP6B7u6emO7IWXqj0NHqbOk4C2yriee6aA8HhyZwjOOQCO0VVQB8bp4kFIglvFLQub5pIXdkslbaSrSSpnomo8VdWv5UlferTfpuEWD1rddK5lPVxRh+oOeREtuZ2de/NxDiKfBsl60hpTeXSAd9feEenj6pNzmOnToS0zmRsNYpEsgA7fLIWR7yTEbfrxnciuOQCI5xHBLBMY5DIrTdHK/WG/rAnL3M4fx7hXwRBihXTWQ7YyV8eUK3yKV1G4vwAbGIGttJ3KFMGNfzLwVGD6uJvUiplA4r1MRvy1gJayz2bdWtKhfyZ6c8MSZVVD9htc+p8hAKRawm9r5PFy0TW+plVR36sP9vmmHRFYeIHieiM0S0V3w3SETPENHr0b8DC43hcO2hFVH1BID7re8+AeAHzLwNwA+izw4rCIuKKmZ+jog2WV+/H8B7I3oHgGcB/MViY4VhiFKlseymU7atO4+5DKBcPB3TWZFMNbhmveqXF6txKtBeX094iDllkp8uTGqvbLkwHdMbN29XbTN1sxV5ctIkaOVy2tSt140YIMs7rPJ79RRVm0ymykIna6U84X22ovuB9BlIt0C1qPqFU6Ym1rkTh/REePH1JKlyvIaZTwFA9O/qhOM4LFNcceWYiB4G8DAApEXxAIfljaSMc5qIRpj5FBGNADgzX0dZkSvXN8jBbEKRZQ4MiASk3i699Jc7xTTJiIFMQecLdwhX7+rVehGs5I2nVFayyHfoe3mdZh6dvb2qrb9rJKbXDpk8Y9syqwiRU7Laxs8asVsvTqm2DJt5pX1jSXmh9kzX68aznvb0/EOIShky+FrW3vjpk0diujp5WrUVCjqHuhmSiqqnAXwkoj8C4FsJx3FYpmjFHP9XAM8D2E5Ex4noowAeAXAfEb2ORp3jR67sNB2WGlqxqh6ap+meyzwXh2WEtnqOwQz4DXnd19mjmvqFHnPi1FHVVhb7iqrCzKZxXYJ38yqj16zeMKraXjlpSpHEpU8BdBa1ntTXZXSEl4/tVm3da41J2y2qph5+TdcED7qMP7R/29v1GOveFNPFsQOqzROugF423txSQetCpRmjUmYz3aptumI8zvl+U9d6VV6b7QVp4luOYlKukubZDC5W5ZAIjnEcEqHtoioVbe9d262X2NOTZvmt91hJWD1GrKVEdQm/rpOYNt78lpietJbY2oAxWz1R/SLVq3Nxp6aN2TpT0WIsLBmRUa0YkdlnjXGsYMRM8az2TG/s74/pddu1GJvab0zw4gkjhidPa5E8XTRjBla22YWyeXb5ASOqejbo4xj8khGLlbI2v+2gbTO4FcchERzjOCSCYxyHRGirjpP2PAz2NvSVoW5tjk+dN27vwQ6dhJUTiee+KHS9equOXm8ZMXuK9h3VEd/+nImO+yJ6vXptv+qXGjK6VzGt36tUjxlj8qypurVxtY7Sl7Jm/MlAR6XPT541441cp9rW33BbTJ84/kpMV8ol1S/jyZIwOnbjhcbMrk4ZvfEsdMjBL5kxU569rwqLwq04DongGMchEdoqqrIZDxvXDgIAfufX36faxg5tiumZis6BrVbM0u9XjajatE4v9bLgNA+tVW0XhHgqivIi64d0FF3u+yoUda4vi7Otutl4h71Qr+1r+kyEvXjmrGornDAiol7VLoMukZi27i13xnRY11W9zpx8I6ZL1h40iLn0dhmzOg3tWmDxP18vWbnPC54N2YBbcRwSwTGOQyK0VVR5xOj1Gsv/7TdrMXPrW0xQcqakPZmymkJdVInwS3r5LVfMdZtrOshZqprluCACm3YRy8lp41Ht2Ky3xparZnzuN1UiToyfUv1eP2yCtDcMaFF49Ox58yHUHtqgw1ia3Rtvjuk7t25S/c4fM6Lq1ZdeVG1nxl+N6S4SnnUr57gSmHuTlWyWFhXLKr5OIpuFW3EcEsExjkMiOMZxSIT2Vh31fRTON+Tu8cN7Vdv6UXPG0+jIGtWWFklfoYhsT09MqH5TU0amrxrU5yQUy8ajWioL07ygZf9MwVT43L51i2orijMgKmWjJw3ndQHrTNXc6x3vvEO1nS+ZtiPj2syupUyUPSgLV8CAjmyve7t5VsNvv0+1+SLx/PyBn8f04b0vqH4Tb7wW06msfgaptNB5qk7HcbiMcIzjkAjtNcdTHvrzjU15M+fGVdspYRIOrdWeyz5x/k9XjwhK9ulAqUdGDPToYxLQJ4KqnGoe8ASAA/tNcHF4WIuIzk7jQigJEXfjJm36v+cWY0qXfR2ELIltv9s2aI/t6XNG/J0cN2b7+OFjqt9REdisWLnb+X7jfe5/q9nyf9P221W/0cN7YnrPT/UJ4GfHD4tP02gGt+I4JIJjHIdEcIzjkAht1XEynoeRwYa5SzVduuP8aZN0tHvPQdW2a69xo68ZNclad77nLtVvdNiY0pVJnfzkpYXSI3SctFVJ+7p1JuqdtxPKsuY9682KPds9OjRRD8wYM2X9O8uibOqB14+otsmqiaTfvMXoV4XVeo6HxdGNB8ZeUW27D5lnN5Mz+uBQr95jfsMao5fdcpc26Xc9/0xMT081LwvQyhbgDUT0f0R0gIj2EdHHou9dVa4VjFZElQ/gT5n5egC3AfhDIroBrirXikYre8dPAZgtojRDRAcAjCJBVa5yqYg9uxoeTD6n9wr1rTJL84v79PL7iljS33W32bL+1a/9i+r3W/e8O6YHOrQZ3JE3Zms6Y5btckWLtGGxjTjM6Xo+k9Xm5T/Iytmti/eRMnrP1cGx4zH9uc9+TrVNnDEm+DtvM7/lNz/4YdVv9VrzrLp8nSGwzjeicN+UPB9Dl/86c9Q8/23XaU/9lu03xPRrL/8czXBRynFU0u1XAfwcrirXikbLjENE3QCeBPBxZm7uFWp+3cNEtJOIdlatg0cdli9aYhxqHCbwJICvMfM3o69PR9W4sFBVLmZ+jJlvYeZbci2cEeCwPLCojkONaslfAXCAmT8rmmarcj2CFqty1YMQZ6caOsUrGZ3E7Z0x+6GPntIZdXfd896Y/uSn/iqmP/+FL6p+3/320zH95lEdHc9kTVZbV48p0WafzTTYNxjTw4NWlF6Y7vJo6RTpx1iQhbStvVlf+od/jun9r7ys2nIZM+ZTT/9bTK/f/jbV723bfiWm89Y5VL1s7r1ObM/3rXkUhVuAa1p32ziqszOboRU/zrsAfBjAy0T0y+i7T6LBMN+IKnQdBfDBFsZyuEbQilX1Y8wpvRPDVeVaoWjvvqpcDqObGhWpAmtLar1uEpeyXboEyoioriWPX96wTm+9/Z9vPRnTM+PaH9kpkq1y+eZHLAP6DIjuTj2Pzrwx47NCrHRkdShe7r86a1X73HfAVO+691793t14040x/U9fNiLt+ef+U/XbIrYtZzt1wvvEuPEq737dJGtluvQc1/SaMYKyFtf57OKqr4tVOSSCYxyHRGjz0Yocn3wbWGf6ZMV5CF26LjWmC8a7e1psqZ04rytyHR83lhn72mfUIQpw1+tmabaP7cmJfVZdOetYo7QozCiOJuywimyHoprE0bO6+LQ8//GBD3xANd1xh8lPPnbMeJifevrbqt+u3RtjOqjoRLTJ0yaPuXbuREynA53wVfLNNuhDkzpRrDOng7bN4FYch0RwjOOQCI5xHBKhrTqO7weYmGroIXVflxCR51exr83DXXvMHqy33fgO8b32vMqodC2tzc9a3egnp06Z/ViVqp5HVniHM1bxTWm4y2Md7f3ngSyVYlUuHRwy3uihVdq7PSP2ra8dMWVaZBUvAPj+901yecXaF3bunKh4Ks6rSlt7vzyhaw2s0Un5q9foEjHN4FYch0RwjOOQCO01x4kRRKfxkqdNvoIoZlgu6Ipc46LI9KOf/0JMjx3UyWCFmhFxB0/o5V1W65KBzXqgS3xQYAJ+nvVekRBWJLytTL7VT95YG/z5LjP+uXO6eHZOBE6nLxixVa3q8Y8cMaY6WWK9Ln4OCzeB7XaQnu+unPaQl+wTg5vArTgOieAYxyERHOM4JEJ7C2Sn0xhcNZsopW3dsjArq1Z0PCXMyqlJcxDHqmGd5tw3aMxK3wpphCwql9aNnhGfERpBhiPCuh5D6kZVUf4jtPQYCHM8Zb2bU8Lk/slPf6La7r777pjet9+cZWUXrK6J3+ZZzzEUz0rqb0HVStutmTGOjemQg5fT4YlmcCuOQyI4xnFIhLZHx4PoHCn7yOW0iMjmcjraLHN9BwZMtU9YpmgolvCUp5dwv2bM/TAwYiawzHE5L1sCyXMkCkXjMqha+63qonRKYM1R9v3Od7+r2vbuN0leO198KaYppaP0gTD4fWuS0mvNvthXFWiRLD/Z51N18OK7UdyK45AIjnEcEqGtoopAoOhoxEzG8sqK5CcEOg84I/djiZWZycoXluLJasuKX0owSVhS/ABAIEWoJQak+Fs1ZLbR1K0xWIiLuaLQiK5iUW8/Hj9tkr42bTIFImeKWnSUyjJwqucoRZcSW5ZqIH9LKqX/L1Ip8+xK09q7Hfdp+q2DwyJwjOOQCI5xHBKhzeY4gbkhWznUOoiKPFvb/6SJrPSdtDYjSVyYsgcRfT0h0zOWh7kuCiPY24Nl2FtG2z3S5rIvTF/LK4CMuHe+Rx/rOHqdcUlI10K5ZkXAhU5luzVkyRUW+o7dzxMTs3+ndBmcGDuMZmilIlcHEf2CiHZHFbn+OvreVeRawWhFVFUBvI+ZbwRwE4D7ieg2uIpcKxqt7B1nALNu0kz0x0hQkYtDRq3SWBbJEiWyqFXGMg/lMusJLzJZooqFaRpaZiqJ4F9KiJZMXosZ9oyoynkLvVei2oNltvsicFqv6X1PoTCRfSvAWqpJM96Ij4q1R0w9O08/RxbXSRNcVtcA5hbNlOjs7Jy3bRat1sfxokoVZwA8w8yuItcKR0uMw8wBM98EYD2AW4nora3eQFbkst8+h+WLizLHmXkKDZF0PxJU5MpkF99a6rA80EpFrmEAdWaeIqI8gHsBfBoJKnIBAMf7ebRsVlFk0uZhLmf2BGlzWct+uddpTvQdpi0Q5qxv52Dx/HqSdMVLPYMsnSyTE6Z/Rr8s8jrbDJZzrgu9JhVqXSgU1/nWGHK/VCh0KFsPsz9L2CGIZmjFjzMCYAc1gkwpAN9g5u8Q0fNwFblWLFqxqvagUaLW/v4cXEWuFQtaaMm67DcjOgtgDMAQgIlFuq8kLOXnsZGZh+0v28o48U2JdjLzLW2/8RLFcnweLsjpkAiOcRwS4WoxzmNX6b5LFcvueVwVHcdh+cOJKodEaCvjENH9RPQqER0kohWXhnEtnTbYNlEVeZ5fA3AfgOMAXgDwEDPvX/DCawhRTG+EmV8ioh4ALwJ4AMDvAzjPzI9EL9QAMy+YonK10c4V51YAB5n5EDPXAHwdjZyeFQNmPsXML0X0DAB52uCOqNsONJhpSaOdjDMKQJZFOB59tyKx3E8bbCfjNDuBZkWadElPG1xKaCfjHAewQXxeD+BkG++/JHAppw0uJbSTcV4AsI2INhNRFsCH0MjpWTFo4bRB4CJym64m2h0d/w0Aj6JRjutxZv77tt18CYCI3g3gRwBeBjCbtfVJNPScbwC4DlFuEzOfbzrIEoHzHDskgvMcOySCYxyHRHCM45AIjnEcEsExjkMiOMaJQEQPENENV/gem4ho7zxtX569PxEdIaKhZv2WChzjGDwA4IoyzkJg5j9YTpkC1zTjENF/ENGLUe7Lw9F3BdH+IBE9QUR3APhtAJ8hol8S0VYiuomIfkZEe4joqdkcGSJ6log+R0TPRXk1v0ZE34xyaf5OjP0nRLQ3+vu4mFaaiHZE4/47EXWKcefsdCCi343qE/2SiP6RZqtvXm0w8zX7B2Aw+jcPYC+AVQAKov1BAE9E9BMAHhRtewC8J6L/BsCjEf0sgE9H9MfQiLeNAMihEY9bBeAdaHiHuwB0A9iHRiR8ExqB3XdF1z8O4M/EuLdE9BE09lpdD+DbADLR918E8HtX+7ky87W94gD4YyLaDeBnaARYt7VyERH1Aehn5h9GX+0AcJfoMhtjexnAvijPpgrgUHSfdwN4ipmLzFwA8E0Ad0bXHGPm2dM/vhr1nQ/3oMGEL0RlZu4BsKWV33Cl0dYagO0EEb0XjQIJtzNziYieBdABncrR0eTSVjBbJC8U9OznNJqnkMzCjvEsFPMhADuY+S8veoZXGNfyitMHYDJimjcDuC36/jQRXU+NEl0fEP1nAPQAADNfADBJRLOrxIcB/BCt4zkADxBRJxF1Rff5UdR2HRHdHtEPAfjxAuP8AMCDRLQaiHOTN17EPK4YrmXG+S80FNE9AP4WDXEFNGoVfgfA/wI4Jfp/HcCfE9EuItqKRnrDZ6Lrb0JDz2kJUXroEwB+gUbk+8vMvCtqPgDgI9G4gwC+tMA4+wF8CsD3o/7PoKFPXXW46LhDIlzLK47DFYRjHIdEcIzjkAiOcRwSwTGOQyI4xnFIBMc4DongGMchEf4ff7fwHdiQTdoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1080x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_sample(X_train, y_train, 4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_sample(X_train, y_train, 12)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[[0.00090734, 0.00095348, 0.00096886],\n",
       "         [0.00066128, 0.00070742, 0.00069204],\n",
       "         [0.00076894, 0.00073818, 0.00066128],\n",
       "         ...,\n",
       "         [0.00242983, 0.00202999, 0.0016609 ],\n",
       "         [0.00233756, 0.00192234, 0.00156863],\n",
       "         [0.00227605, 0.00190696, 0.00158401]],\n",
       "\n",
       "        [[0.00024606, 0.00030757, 0.00030757],\n",
       "         [0.        , 0.        , 0.        ],\n",
       "         [0.00027682, 0.00012303, 0.        ],\n",
       "         ...,\n",
       "         [0.00189158, 0.00135333, 0.00084583],\n",
       "         [0.00183007, 0.00127643, 0.00076894],\n",
       "         [0.0018762 , 0.00133795, 0.00087659]],\n",
       "\n",
       "        [[0.00038447, 0.00036909, 0.00032295],\n",
       "         [0.00024606, 0.00010765, 0.        ],\n",
       "         [0.00075356, 0.00041522, 0.00012303],\n",
       "         ...,\n",
       "         [0.00181469, 0.00129181, 0.00076894],\n",
       "         [0.00184544, 0.00129181, 0.00076894],\n",
       "         [0.00167628, 0.00112265, 0.00064591]],\n",
       "\n",
       "        ...,\n",
       "\n",
       "        [[0.00319877, 0.00261438, 0.00147636],\n",
       "         [0.00309112, 0.00235294, 0.00052288],\n",
       "         [0.00304498, 0.00247597, 0.00039985],\n",
       "         ...,\n",
       "         [0.00246059, 0.00204537, 0.00107651],\n",
       "         [0.00086121, 0.00047674, 0.00010765],\n",
       "         [0.00081507, 0.00052288, 0.00030757]],\n",
       "\n",
       "        [[0.00276817, 0.00213764, 0.00147636],\n",
       "         [0.00266052, 0.00189158, 0.00064591],\n",
       "         [0.00286044, 0.00221453, 0.00046136],\n",
       "         ...,\n",
       "         [0.00282968, 0.00227605, 0.0014456 ],\n",
       "         [0.00149173, 0.00095348, 0.00052288],\n",
       "         [0.00127643, 0.00081507, 0.00052288]],\n",
       "\n",
       "        [[0.00272203, 0.00221453, 0.00178393],\n",
       "         [0.00258362, 0.00198385, 0.0014456 ],\n",
       "         [0.00275279, 0.00218378, 0.00133795],\n",
       "         ...,\n",
       "         [0.0033218 , 0.00282968, 0.00215302],\n",
       "         [0.00232218, 0.00181469, 0.00129181],\n",
       "         [0.00189158, 0.00141484, 0.00110727]]],\n",
       "\n",
       "\n",
       "       [[[0.00236832, 0.00272203, 0.00287582],\n",
       "         [0.00193772, 0.00210688, 0.0020915 ],\n",
       "         [0.00161476, 0.00159938, 0.00146098],\n",
       "         ...,\n",
       "         [0.00139946, 0.00146098, 0.00109189],\n",
       "         [0.00133795, 0.00138408, 0.00109189],\n",
       "         [0.00121492, 0.00124567, 0.00107651]],\n",
       "\n",
       "        [[0.00215302, 0.00246059, 0.002599  ],\n",
       "         [0.00222991, 0.00235294, 0.00236832],\n",
       "         [0.00192234, 0.00192234, 0.00181469],\n",
       "         ...,\n",
       "         [0.00147636, 0.00152249, 0.00119954],\n",
       "         [0.00118416, 0.0012303 , 0.00095348],\n",
       "         [0.00109189, 0.00112265, 0.0009381 ]],\n",
       "\n",
       "        [[0.00215302, 0.0023837 , 0.00252211],\n",
       "         [0.00213764, 0.00224529, 0.00229143],\n",
       "         [0.00176855, 0.00176855, 0.00172241],\n",
       "         ...,\n",
       "         [0.00121492, 0.00126105, 0.00098424],\n",
       "         [0.00104575, 0.00107651, 0.00084583],\n",
       "         [0.00103037, 0.00106113, 0.00084583]],\n",
       "\n",
       "        ...,\n",
       "\n",
       "        [[0.00269127, 0.00256824, 0.00255286],\n",
       "         [0.00239908, 0.00236832, 0.00246059],\n",
       "         [0.00236832, 0.00246059, 0.00261438],\n",
       "         ...,\n",
       "         [0.00064591, 0.00052288, 0.00055363],\n",
       "         [0.0009381 , 0.00081507, 0.00087659],\n",
       "         [0.00143022, 0.00127643, 0.00139946]],\n",
       "\n",
       "        [[0.00253749, 0.00236832, 0.00196847],\n",
       "         [0.00239908, 0.00233756, 0.00199923],\n",
       "         [0.00244521, 0.00247597, 0.00218378],\n",
       "         ...,\n",
       "         [0.00158401, 0.00143022, 0.00147636],\n",
       "         [0.00189158, 0.00175317, 0.00184544],\n",
       "         [0.00201461, 0.00186082, 0.00201461]],\n",
       "\n",
       "        [[0.00250673, 0.00227605, 0.00184544],\n",
       "         [0.00242983, 0.00227605, 0.0018762 ],\n",
       "         [0.00250673, 0.00239908, 0.00204537],\n",
       "         ...,\n",
       "         [0.00219915, 0.00204537, 0.00213764],\n",
       "         [0.00219915, 0.00206075, 0.00218378],\n",
       "         [0.00219915, 0.00204537, 0.00221453]]],\n",
       "\n",
       "\n",
       "       [[[0.00392157, 0.00392157, 0.00392157],\n",
       "         [0.00389081, 0.00389081, 0.00389081],\n",
       "         [0.00389081, 0.00389081, 0.00389081],\n",
       "         ...,\n",
       "         [0.00389081, 0.00389081, 0.00389081],\n",
       "         [0.00389081, 0.00389081, 0.00389081],\n",
       "         [0.00389081, 0.00389081, 0.00389081]],\n",
       "\n",
       "        [[0.00392157, 0.00392157, 0.00392157],\n",
       "         [0.00392157, 0.00392157, 0.00392157],\n",
       "         [0.00392157, 0.00392157, 0.00392157],\n",
       "         ...,\n",
       "         [0.00392157, 0.00392157, 0.00392157],\n",
       "         [0.00392157, 0.00392157, 0.00392157],\n",
       "         [0.00392157, 0.00392157, 0.00392157]],\n",
       "\n",
       "        [[0.00392157, 0.00392157, 0.00392157],\n",
       "         [0.00390619, 0.00390619, 0.00390619],\n",
       "         [0.00390619, 0.00390619, 0.00390619],\n",
       "         ...,\n",
       "         [0.00390619, 0.00390619, 0.00390619],\n",
       "         [0.00390619, 0.00390619, 0.00390619],\n",
       "         [0.00390619, 0.00390619, 0.00390619]],\n",
       "\n",
       "        ...,\n",
       "\n",
       "        [[0.00173779, 0.00184544, 0.00172241],\n",
       "         [0.00170704, 0.00181469, 0.00170704],\n",
       "         [0.00161476, 0.00172241, 0.00163014],\n",
       "         ...,\n",
       "         [0.00110727, 0.00124567, 0.0012303 ],\n",
       "         [0.00110727, 0.0012303 , 0.00121492],\n",
       "         [0.00110727, 0.0012303 , 0.00121492]],\n",
       "\n",
       "        [[0.00170704, 0.00181469, 0.00169166],\n",
       "         [0.00159938, 0.00170704, 0.00159938],\n",
       "         [0.00152249, 0.00163014, 0.00150711],\n",
       "         ...,\n",
       "         [0.00104575, 0.0011534 , 0.00112265],\n",
       "         [0.00107651, 0.00116878, 0.0011534 ],\n",
       "         [0.00119954, 0.00129181, 0.00126105]],\n",
       "\n",
       "        [[0.00163014, 0.00173779, 0.00161476],\n",
       "         [0.00152249, 0.00163014, 0.00150711],\n",
       "         [0.00146098, 0.00156863, 0.0014456 ],\n",
       "         ...,\n",
       "         [0.00119954, 0.00130719, 0.00127643],\n",
       "         [0.00121492, 0.00130719, 0.00127643],\n",
       "         [0.0012303 , 0.00132257, 0.00129181]]],\n",
       "\n",
       "\n",
       "       ...,\n",
       "\n",
       "\n",
       "       [[[0.00053825, 0.00273741, 0.00361399],\n",
       "         [0.00061515, 0.00270665, 0.00367551],\n",
       "         [0.00064591, 0.00270665, 0.00370627],\n",
       "         ...,\n",
       "         [0.00152249, 0.00272203, 0.00336794],\n",
       "         [0.00121492, 0.00226067, 0.0030296 ],\n",
       "         [0.0013687 , 0.00227605, 0.00290657]],\n",
       "\n",
       "        [[0.00087659, 0.00279892, 0.00359862],\n",
       "         [0.00067666, 0.00282968, 0.00384468],\n",
       "         [0.00076894, 0.0028143 , 0.00369089],\n",
       "         ...,\n",
       "         [0.00239908, 0.00279892, 0.00307574],\n",
       "         [0.0021684 , 0.00272203, 0.00316801],\n",
       "         [0.00178393, 0.00229143, 0.00269127]],\n",
       "\n",
       "        [[0.00150711, 0.0030296 , 0.00364475],\n",
       "         [0.00098424, 0.00290657, 0.00387543],\n",
       "         [0.00106113, 0.00295271, 0.00376778],\n",
       "         ...,\n",
       "         [0.0028912 , 0.00299885, 0.00316801],\n",
       "         [0.00183007, 0.00207612, 0.00226067],\n",
       "         [0.0009381 , 0.00121492, 0.00138408]],\n",
       "\n",
       "        ...,\n",
       "\n",
       "        [[0.00112265, 0.00121492, 0.00118416],\n",
       "         [0.00081507, 0.00096886, 0.00104575],\n",
       "         [0.00083045, 0.00104575, 0.0012303 ],\n",
       "         ...,\n",
       "         [0.00026144, 0.00061515, 0.00098424],\n",
       "         [0.00032295, 0.00055363, 0.00078431],\n",
       "         [0.0005075 , 0.00073818, 0.00075356]],\n",
       "\n",
       "        [[0.0009381 , 0.00104575, 0.0011534 ],\n",
       "         [0.00084583, 0.00107651, 0.00132257],\n",
       "         [0.00087659, 0.00121492, 0.00158401],\n",
       "         ...,\n",
       "         [0.00036909, 0.00073818, 0.00110727],\n",
       "         [0.00026144, 0.00053825, 0.00081507],\n",
       "         [0.00010765, 0.00035371, 0.00049212]],\n",
       "\n",
       "        [[0.00067666, 0.00086121, 0.00112265],\n",
       "         [0.00070742, 0.00101499, 0.00135333],\n",
       "         [0.00075356, 0.00118416, 0.00161476],\n",
       "         ...,\n",
       "         [0.00041522, 0.00079969, 0.00118416],\n",
       "         [0.00032295, 0.00066128, 0.00101499],\n",
       "         [0.00018454, 0.00047674, 0.00076894]]],\n",
       "\n",
       "\n",
       "       [[[0.00290657, 0.00324491, 0.00369089],\n",
       "         [0.00286044, 0.00319877, 0.00362937],\n",
       "         [0.00284506, 0.00318339, 0.00361399],\n",
       "         ...,\n",
       "         [0.00269127, 0.00299885, 0.00344483],\n",
       "         [0.00264514, 0.00298347, 0.00341407],\n",
       "         [0.002599  , 0.00298347, 0.00338331]],\n",
       "\n",
       "        [[0.00298347, 0.00322953, 0.00367551],\n",
       "         [0.00293733, 0.00318339, 0.00362937],\n",
       "         [0.00292195, 0.00316801, 0.00361399],\n",
       "         ...,\n",
       "         [0.00266052, 0.00295271, 0.00338331],\n",
       "         [0.00262976, 0.00293733, 0.00335256],\n",
       "         [0.00256824, 0.00292195, 0.0033218 ]],\n",
       "\n",
       "        [[0.00319877, 0.00336794, 0.0037524 ],\n",
       "         [0.00315263, 0.0033218 , 0.00369089],\n",
       "         [0.00313725, 0.00330642, 0.00367551],\n",
       "         ...,\n",
       "         [0.00269127, 0.00293733, 0.00333718],\n",
       "         [0.00264514, 0.00292195, 0.0033218 ],\n",
       "         [0.002599  , 0.00293733, 0.00330642]],\n",
       "\n",
       "        ...,\n",
       "\n",
       "        [[0.00318339, 0.00306036, 0.00278354],\n",
       "         [0.00312188, 0.00299885, 0.00269127],\n",
       "         [0.00312188, 0.00301423, 0.00266052],\n",
       "         ...,\n",
       "         [0.00207612, 0.00202999, 0.00195309],\n",
       "         [0.00249135, 0.00242983, 0.00230681],\n",
       "         [0.00258362, 0.00250673, 0.00232218]],\n",
       "\n",
       "        [[0.00304498, 0.00292195, 0.00261438],\n",
       "         [0.00290657, 0.00278354, 0.00244521],\n",
       "         [0.00276817, 0.00264514, 0.00226067],\n",
       "         ...,\n",
       "         [0.00273741, 0.00262976, 0.00246059],\n",
       "         [0.00269127, 0.002599  , 0.00239908],\n",
       "         [0.00269127, 0.002599  , 0.00236832]],\n",
       "\n",
       "        [[0.00304498, 0.00290657, 0.00266052],\n",
       "         [0.00290657, 0.00278354, 0.00249135],\n",
       "         [0.00273741, 0.00261438, 0.00229143],\n",
       "         ...,\n",
       "         [0.00299885, 0.00282968, 0.002599  ],\n",
       "         [0.00301423, 0.00290657, 0.00262976],\n",
       "         [0.00299885, 0.00292195, 0.00262976]]],\n",
       "\n",
       "\n",
       "       [[[0.00352172, 0.00352172, 0.00367551],\n",
       "         [0.00362937, 0.00364475, 0.00379854],\n",
       "         [0.00359862, 0.00362937, 0.00379854],\n",
       "         ...,\n",
       "         [0.00333718, 0.00336794, 0.00358324],\n",
       "         [0.00339869, 0.00342945, 0.00359862],\n",
       "         [0.00341407, 0.00342945, 0.00358324]],\n",
       "\n",
       "        [[0.00341407, 0.00339869, 0.00352172],\n",
       "         [0.00367551, 0.00367551, 0.0038293 ],\n",
       "         [0.00358324, 0.00359862, 0.00378316],\n",
       "         ...,\n",
       "         [0.00342945, 0.00342945, 0.00362937],\n",
       "         [0.00349097, 0.00350634, 0.00366013],\n",
       "         [0.00322953, 0.00324491, 0.00338331]],\n",
       "\n",
       "        [[0.00327566, 0.00316801, 0.00324491],\n",
       "         [0.00359862, 0.00356786, 0.00367551],\n",
       "         [0.00355248, 0.00358324, 0.0037524 ],\n",
       "         ...,\n",
       "         [0.00338331, 0.00338331, 0.00356786],\n",
       "         [0.00338331, 0.00336794, 0.00356786],\n",
       "         [0.0031065 , 0.00312188, 0.00330642]],\n",
       "\n",
       "        ...,\n",
       "\n",
       "        [[0.00230681, 0.00219915, 0.00207612],\n",
       "         [0.00215302, 0.00207612, 0.00195309],\n",
       "         [0.00202999, 0.00195309, 0.00184544],\n",
       "         ...,\n",
       "         [0.00344483, 0.00341407, 0.00335256],\n",
       "         [0.0035371 , 0.00350634, 0.00346021],\n",
       "         [0.00370627, 0.00370627, 0.00366013]],\n",
       "\n",
       "        [[0.00210688, 0.00202999, 0.00193772],\n",
       "         [0.00199923, 0.00195309, 0.00184544],\n",
       "         [0.00192234, 0.00186082, 0.00176855],\n",
       "         ...,\n",
       "         [0.00278354, 0.00276817, 0.00273741],\n",
       "         [0.0031065 , 0.00309112, 0.00304498],\n",
       "         [0.00326028, 0.00324491, 0.00318339]],\n",
       "\n",
       "        [[0.0018762 , 0.00183007, 0.00175317],\n",
       "         [0.00181469, 0.00178393, 0.00169166],\n",
       "         [0.00184544, 0.00178393, 0.00170704],\n",
       "         ...,\n",
       "         [0.00275279, 0.00272203, 0.00266052],\n",
       "         [0.00252211, 0.00252211, 0.00249135],\n",
       "         [0.00250673, 0.00250673, 0.00247597]]]])"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train=X_train/255\n",
    "X_test=X_test/255\n",
    "X_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "1563/1563 [==============================] - 116s 74ms/step - loss: 2.3019 - accuracy: 0.1238\n",
      "Epoch 2/5\n",
      "1563/1563 [==============================] - 109s 70ms/step - loss: 2.3015 - accuracy: 0.1506\n",
      "Epoch 3/5\n",
      "1563/1563 [==============================] - 108s 69ms/step - loss: 2.3013 - accuracy: 0.1476\n",
      "Epoch 4/5\n",
      "1563/1563 [==============================] - 117s 75ms/step - loss: 2.3011 - accuracy: 0.1344\n",
      "Epoch 5/5\n",
      "1563/1563 [==============================] - 104s 66ms/step - loss: 2.3009 - accuracy: 0.1527\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x272f8227df0>"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ann=models.Sequential([\n",
    "    layers.Flatten(input_shape=(32,32,3)),\n",
    "    layers.Dense(3000, activation='relu'),\n",
    "    layers.Dense(1000, activation='relu'),\n",
    "    layers.Dense(10, activation='sigmoid')\n",
    "])\n",
    "\n",
    "ann.compile(optimizer='SGD', loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n",
    "ann.fit(X_train, y_train, epochs=5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 7s 23ms/step - loss: 2.3007 - accuracy: 0.1537\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[2.3007285594940186, 0.15369999408721924]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ann.evaluate(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification Report:\n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       0.15      0.87      0.25      1000\n",
      "           1       0.00      0.00      0.00      1000\n",
      "           2       0.00      0.00      0.00      1000\n",
      "           3       0.33      0.00      0.00      1000\n",
      "           4       0.40      0.01      0.02      1000\n",
      "           5       0.16      0.42      0.23      1000\n",
      "           6       0.00      0.00      0.00      1000\n",
      "           7       0.16      0.24      0.19      1000\n",
      "           8       0.00      0.00      0.00      1000\n",
      "           9       0.00      0.00      0.00      1000\n",
      "\n",
      "    accuracy                           0.15     10000\n",
      "   macro avg       0.12      0.15      0.07     10000\n",
      "weighted avg       0.12      0.15      0.07     10000\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1221: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix, classification_report\n",
    "import numpy as np\n",
    "y_pred=ann.predict(X_test)\n",
    "y_pred_classes=[np.argmax(element) for element in y_pred]\n",
    "\n",
    "print('Classification Report:\\n', classification_report(y_test, y_pred_classes))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "1563/1563 [==============================] - 59s 37ms/step - loss: 2.0639 - accuracy: 0.2422\n",
      "Epoch 2/10\n",
      "1563/1563 [==============================] - 64s 41ms/step - loss: 1.8138 - accuracy: 0.3490\n",
      "Epoch 3/10\n",
      "1563/1563 [==============================] - 61s 39ms/step - loss: 1.6894 - accuracy: 0.3898\n",
      "Epoch 4/10\n",
      "1563/1563 [==============================] - 57s 37ms/step - loss: 1.6209 - accuracy: 0.4151\n",
      "Epoch 5/10\n",
      "1563/1563 [==============================] - 51s 33ms/step - loss: 1.5731 - accuracy: 0.4339\n",
      "Epoch 6/10\n",
      "1563/1563 [==============================] - 61s 39ms/step - loss: 1.5380 - accuracy: 0.4488\n",
      "Epoch 7/10\n",
      "1563/1563 [==============================] - 55s 35ms/step - loss: 1.5071 - accuracy: 0.4585\n",
      "Epoch 8/10\n",
      "1563/1563 [==============================] - 52s 33ms/step - loss: 1.4692 - accuracy: 0.4739\n",
      "Epoch 9/10\n",
      "1563/1563 [==============================] - 56s 36ms/step - loss: 1.4392 - accuracy: 0.48450s - loss: 1.4388 - accu - ETA: 0s - loss: 1.4391 - accuracy: 0.\n",
      "Epoch 10/10\n",
      "1563/1563 [==============================] - 45s 29ms/step - loss: 1.4178 - accuracy: 0.4917\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x272f7ec4c70>"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cnn=models.Sequential([\n",
    "    layers.Conv2D(filters=32, kernel_size=(3,3), activation='relu', input_shape=(32,32,3)),\n",
    "    layers.MaxPooling2D((2,2)),\n",
    "    \n",
    "    layers.Conv2D(filters=64, kernel_size=(3,3), activation='relu', input_shape=(32,32,3)),\n",
    "    layers.MaxPooling2D((2,2)),\n",
    "    \n",
    "    layers.Flatten(),\n",
    "    layers.Dense(64, activation='relu'),\n",
    "    layers.Dense(10, activation='softmax')\n",
    "    \n",
    "])\n",
    "\n",
    "cnn.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n",
    "cnn.fit(X_train, y_train, epochs=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 3s 8ms/step - loss: 1.4355 - accuracy: 0.4857\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[1.4355486631393433, 0.48570001125335693]"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cnn.evaluate(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10000, 1)"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[3],\n",
       "       [8],\n",
       "       [8],\n",
       "       [0],\n",
       "       [6]], dtype=uint8)"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([3, 8, 8, 0, 6], dtype=uint8)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test= y_test.reshape(-1,)\n",
    "y_test[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_sample(X_test, y_test, 4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[3.14806332e-03, 6.12902222e-03, 6.77718446e-02, 4.78925377e-01,\n",
       "        3.92541848e-02, 2.58859426e-01, 1.24940708e-01, 9.15090367e-03,\n",
       "        8.13172478e-03, 3.68879386e-03],\n",
       "       [1.76762924e-01, 2.83114016e-01, 1.69828138e-03, 1.28077099e-03,\n",
       "        2.41927663e-03, 4.03770886e-04, 2.02607698e-04, 9.67743341e-04,\n",
       "        3.56667787e-01, 1.76482841e-01],\n",
       "       [7.88015723e-02, 1.61857113e-01, 5.61125763e-03, 4.68267966e-03,\n",
       "        5.33223478e-03, 1.10699260e-03, 1.50540296e-03, 1.20735599e-03,\n",
       "        6.15934491e-01, 1.23960905e-01],\n",
       "       [2.14763060e-01, 9.82125178e-02, 1.30921323e-02, 7.15627568e-03,\n",
       "        1.64500885e-02, 1.61271216e-03, 1.19002154e-02, 4.27767774e-03,\n",
       "        6.06036484e-01, 2.64987834e-02],\n",
       "       [4.11486346e-03, 6.16441830e-04, 1.28421068e-01, 1.42833054e-01,\n",
       "        2.89495975e-01, 9.52922180e-02, 2.90092915e-01, 4.62212898e-02,\n",
       "        2.40618200e-03, 5.05979988e-04]], dtype=float32)"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred=cnn.predict(X_test)\n",
    "y_pred[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_classes= [np.argmax(element) for element in y_pred]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[3, 8, 8, 8, 6, 6, 3, 6, 5, 1]"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_classes[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([3, 8, 8, 0, 6, 6, 1, 6, 3, 1], dtype=uint8)"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_sample(X_test, y_test, 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'ship'"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classes[y_classes[2]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAI4AAACcCAYAAACp45OYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAGiklEQVR4nO3dSYhcVRjF8f+xjaioOEuTyYEsFEHFEFy4EoXWjYIICSJxgCAoRuJCcaWC4EpcuJAGgxFECSgas9EQFBWnjqJxiNHg2CQkxOAQETX6uXhXbdpK6uWr6pre+cGj3rvd1e+mc+q+oW5/pYjA7HAd0e8O2HBycCzFwbEUB8dSHBxLcXAspaPgSJqQtF3SDkn3dKtTNviUvY8jaQz4HLgCmAamgBUR8Wn3umeD6sgOnrsM2BERXwJIega4GjhocCT5buPw2RsRp81u7ORQNR/4bsb2dGmz0fJNq8ZORhy1aPvfiCJpFbCqg/3YAOokONPAwhnbC4Cds78pIiaBSfChapR0cqiaApZIOkvSUcByYEN3umWDLj3iRMQBSbcDLwFjwNqI+KRrPbOBlr4cT+3Mh6ph9F5ELJ3d6DvHluLgWIqDYykOjqU4OJbi4FiKg2MpDo6lODiW4uBYioNjKQ6OpTg4luLgWIqDYykOjqU4OJbi4FiKg2MpDo6lODiW4uBYioNjKQ6OpTg4ltI2OJLWStoj6eMZbSdL2iTpi/J40tx20wZNnRHnCWBiVts9wOaIWAJsLtvWIG2DExGvAftmNV8NrCvr64BrutwvG3DZc5wzImIXQHk8vXtdsmHQSWGlWlyRazRlR5zdksYByuOeg31jRExGxNJWpTJseGWDswFYWdZXAi90pzs2NCLikAvwNLAL+IOq7t8twClUV1NflMeT2/2c8rPCy9AtW1r9X7oil7XjilzWPQ6OpTg4luLgWIqDYykOjqU4OJbi4FiKg2MpDo6lODiW4uBYioNjKQ6OpTg4luLgWIqDYykOjqU4OJbi4FiKg2MpDo6lODiW4uBYioNjKXUqci2U9IqkbZI+kbS6tLsqV4PVGXEOAHdFxLnAJcBtks7DVbkarU5Frl0R8X5Z/xnYBszHVbka7bDOcSSdCVwEvIOrcjVa7Ypcko4DngXujIifJNV9nityjaKadW3mAS8Ba2a0bQfGy/o4sN31cUZyaVkfp85VlYDHgW0R8fCML7kqV5PVGCUupUreVuCDslxFoioX/X/1eOnSiOOKXNaOK3JZ9zg4luLgWIqDYykOjqU4OJbi4FiKg2MpDo6lODiW4uBYioNjKQ6OpTg4luLgWIqDYykOjqU4OJbi4FiKg2MpDo6lODiW4uBYioNjKQ6OpdT52/GjJb0r6cNSkev+0u6KXA1WZ8T5DbgsIi4ALgQmJF2CK3I1Wp2KXBER+8vmvLIErsjVaLXOcSSNSfoA2ANsighX5Gq4WsGJiD8j4kJgAbBM0vl1dyBplaQtkrZkO2mD57CuqiLiB+BVYALYLWkcoDzuOchzJiNiaatSGTa86lxVnSbpxLJ+DHA58BmuyNVodYpHjgPrJI1RBW19RGyU9BawXtItwLfAdXPYTxswrshl7bSsyFW7XG2X7AW+AU4t61YZ5N/H4laNPR1x/t2ptMUny/8Zxt+H36uyFAfHUvoVnMk+7XdQDd3voy/nODb8fKiylJ4GR9KEpO2Sdkhq3DSMUfq0wZ4dqsqd58+BK4BpYApYERGf9qQDA6C8pzceEe9LOh54j2o6yo3Avoh4qLygToqIu/vY1bZ6OeIsA3ZExJcR8TvwDNWcnsYYpU8b7GVw5gPfzdieLm2NNOyfNtjL4LT6SL1GXtLN/rTBfvcno5fBmQYWztheAOzs4f4HgqR5VKF5KiKeK8215jYNkl4GZwpYIuksSUcBy6nm9DTGKH3aYK+nVVwFPAKMAWsj4sGe7XwASLoUeB34CPirNN9LdZ6zHlhEmdsUEfv60smafOfYUnzn2FIcHEtxcCzFwbEUB8dSHJwaJN1R3tF+qt99GRS+HK9B0mfAlRHx1Yy2IyPiQB+71VcecdqQ9BhwNrBB0o+SJiW9DDwpabGkzZK2lsdF5TnnSHpb0pSkByTtP+ROhlFEeGmzAF9T/e3TfVRzaI4p7S8CK8v6zcDzZX0j1VwjgFuB/f3+N3R78aGqBklfA0uB26lKBv1TlWwv1cSsP8qbl7si4lRJ31NNlTgg6QRgZ0Qc16/+zwUfqg7fL4f4WmNehQ5OZ96kepcf4HrgjbL+NnBtWV8++0mjwMHpzB3ATZK2AjcAq0v7ncAaSe9SVfv4sU/9mzM+x5kDko4Ffo2IkLSc6kR5pOZX97paRVNcDDxaJm79QHXFNVI84liKz3EsxcGxFAfHUhwcS3FwLMXBsZS/AQQmgB06iBDWAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1080x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_sample(X_test, y_test, 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'frog'"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classes[y_classes[5]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x144 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_sample(X_test, y_test, 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'ship'"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classes[y_classes[10]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification Report:\n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       0.61      0.52      0.56      1000\n",
      "           1       0.70      0.47      0.56      1000\n",
      "           2       0.37      0.39      0.38      1000\n",
      "           3       0.33      0.27      0.30      1000\n",
      "           4       0.44      0.36      0.39      1000\n",
      "           5       0.43      0.42      0.43      1000\n",
      "           6       0.44      0.68      0.53      1000\n",
      "           7       0.51      0.56      0.54      1000\n",
      "           8       0.53      0.70      0.61      1000\n",
      "           9       0.56      0.48      0.52      1000\n",
      "\n",
      "    accuracy                           0.49     10000\n",
      "   macro avg       0.49      0.49      0.48     10000\n",
      "weighted avg       0.49      0.49      0.48     10000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print('Classification Report:\\n', classification_report(y_test, y_classes))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}