model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[PhysicalDevice(name='/physical_device:CPU:0', device_type='CPU')]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.config.experimental.list_physical_devices()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.test.is_built_with_cuda()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "(X_train, y_train), (X_test, y_test) = tf.keras.datasets.cifar10.load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50000, 32, 32, 3)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "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": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50000, 1)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[6],\n",
       "       [9],\n",
       "       [9],\n",
       "       [4],\n",
       "       [1]], dtype=uint8)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train[0:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_sample(index):\n",
    "    plt.figure(figsize=(10,1))\n",
    "    plt.imshow(X_train[index])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAHwAAAB9CAYAAABgQgcbAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAzDklEQVR4nO29W6ylaVX3+xvP4X3naa1VVV3d1ScaUPeOEty6g9ASEyBKJPHKyIWJN2oMRNMQtROjeCHRGy410dYrbS62bIyJhEQTblrBrYEQ2xADhN4fbj9poA8UVbUO8/C+z2Hsi/HMuaq+FqzGLnpZVSOZtapmzTXnO9/xPOMZh//4D1FV5a7cMeJe7Qu4K99duavwO0zuKvwOk7sKv8PkrsLvMLmr8DtM7ir8DpO7Cr/D5K7C7zC5q/A7TG6Zwp944gle97rXMZlMePTRR/nsZz97qz7qrrwMuSUK/4u/+Asef/xxPvjBD/LP//zP/NAP/RDvete7ePHFF2/Fx92VlyFyK4onjz76KG9+85v5oz/6IwBqrbzmNa/h/e9/P7/1W7/1bX+31srXv/519vb2EJFX+tJuS1FVjo+PefDBB3Hu2+/h8Ep/+DiOPP3003zgAx/YPeec453vfCef/vSnX/L6YRgYhmH376997Wu84Q1veKUv646QZ599locffvjbvuYVV/jly5cppXDp0qUbnr906RJf+tKXXvL6D33oQ/zu7/7uS57/P9/0I+SSGYYNDiU4xQvMvNI5WPSBg3lH9I7FvCNGz7TvmPYBJw4vDhGH+IhzHhUB56ilslqvSLlQq1Ir+BCYTSc4cYgWRCu1VnLJ9jMVqlZC8IQQUJRKRVFSGik5E7rIdNI3q1QBpZRKKQVBcO1WOxFEhOADXd/hnafrI955nAs4H6hVyblQUUBQBERAHKpQ2nWpChVhPYx88I/+b/b29v5T/bziCn+58oEPfIDHH3989++joyNe85rXcOXqFcZhYNhsCA4mXglO8BOHBKHUCL7DdYFO53R4Yi247JB2kxRBXKCKoyqUqtRaGYbBlFGVWhXnHToEnAheFIe9rpSKqqIKqKIhUENAtZJKQlXJJVG14GsHfmoKF0UALQXNBSdCkIAgVFVUlSLCODjECbXr8d7jxCPOo0BVALGF2n6vVDPfJRdUK+BAHDklgJs6Al9xhV+8eBHvPS+88MINz7/wwgvcf//9L3l93/f0ff+S569evcqwXjNsNkQH8wjRQzePSOcZNZJ9jy8RZoK4SFUhVQFVqKDbryiOnAtj29VjSjtlVzXlrJ0gAtEp3tF2v72DE4+IkL3Hh0CthTENVK3UWlAq2vcEEs45pB2jtWRqMYWLiwhiO7dWtFkREaFrCkccIh4QcL4tHnvkUkmpUFWppaCqiHic9wyp3LR+XnEvves63vSmN/HUU0/tnqu18tRTT/HWt771pt/HC3gH0UHnYRod0+iZdYFpH5h2kb7r6LqIDwHnbXfYDfNUFWoVSq2kXCjVdqoC4hzOOZx39jvOgW0kfPCEGIgxtJ+R2NkjxIj3Hu8DIURC6PAh4n1AxKMqZg0QRFwz0RHnIz5EXIg4H0yZziPtmlWkHQKgaNO3w3kPIs3K6M46bL3sqpWcMinlm76vt8SkP/744/z8z/88P/IjP8Jb3vIW/uAP/oDlcskv/uIv3vyFOUG8EIIp+/w80AfHhb2eeR9ZTHv2FjNi9PR9Twx2c0RAtVJqpqqSS7UbRdvxIrgQcCoUVbOdqqAZcRD7yKT31Aq1KILgfcSJt5cpVFVcCM2kZ0opeO+o1Y4RH4J5y85usDgh+GhmXhI1J0yvFQDFUaGd9bYgY4yIc2aNcqVqpdS2s9tvlVxIKb+sHX5LFP6zP/uzfOMb3+B3fud3eP755/nhH/5hPvGJT7zEkft24kTN/jiIXuiDow+ePni64Aje4Z3gnNiNBETVHJmtud6Z7bozjfbetqNr3T6lzT0C7x3ee0QUUXvWNXOv222IKWf3Xk4RkXYN2/+nLUDZOWp2HNvuVxRUrn/xdd9eTx963eN02baFrZRqR8PNyi2Jw/8rcnR0xMHBAQ/fdw5XBiQP7E0jD16Y0QfP/sR2ehcDkz7inBCjmWUFUKEUJY123iGKCoToiV00BfgAIoxjYkyZ7R5zTji/P2U+7ezMH7Pd6yrNO67kYjd/e4u99zjnUK2oFjuT+7B73jm3UzxAzmYRaq3UYjvTOWcRAu1kESGEsDu7azUHM5dTk66qrDaJ5WpkSIX/6//5/zg8PGR/f//b3t9X3Uv/VuJRggcvjknnmU8iffBMO0fnBe/AaUGqULIZRFXbZabwiiqItwdqvyNO8N6UkN2pP++2PkNwdjyoUtouyrWiVam5Uktlt0VEcN7jnUOrUqqZW1GzDlvlmQlStqeva9bAPG0QHLI1HbZqqSXt/q1VEQUnNA/e3qtWJeVKyje/w8+swjvv6MXjtDKNnj4G+uCZNIU7Z6ay3VVozrniAKWGZgDFdrm468wr7RyshVqy/d0p4BiGhEPa+Wg7MaXSdlghlYKIw7uA4MhV0Wq+gKpeZ5ntM523s76oLRSHOWpOHeLrLuSrzZfYGtztEZFKIRdbTGPW5pfY69dDYT1kxttB4dMgTJwnUJl1gUlnO3zWOybRNa/VvqgdrZZYsdjbzmsFspppFyc4522hYLuQWtCcUGlm21U265GSiu3oYs7SZkyUUkhayLXgXKCPgnMKxVIwO0uhLVUi5nxJCGit1GxRQpRAEE8l4SgtyZKp1c7jUraOnB0bmyEx5kLKyrotvGGsbRFAKkK+HRQevSN6JeAJYeug0XYqIILH7czr1sqqCFSoTq/3cXamcvvTflR2b9AWSNGKq6feeG3Jmt3f7VCnVjOrSiXXglBxWAzvsxK0Uqim1FIZh2yWIHhwDvBAh5OK+ZC1OXyV2j5bVSlFybmSSmVsCt8MmZQrpQq5OvLLcNrOrMIX08gseqKzuLuLjhgcLT+Bd87OTmgxtjZDLRSxPV8rFNQir1qpuaBCswxKzQmt2Q5HAIRUEgVBK1QVtCpDO58t22WftakFQRlrotSKiOKl4hxMB/P2VSrVJWoppLWFYvvTwKzzdF1gOp0iXsFvUDLDMKDFFtg4FkqtrIfEMGaGXFhuMmOqXDvasNlkyyDiLLy8STmzCo/B0UUhOiEGj/On4dH2sasMad2ZdWV3pFtoxS6nstvNqhW0Nqfp+jjKdrhWqFWo2pyxlvSwc5ad9w/KmCupWBhpD0GoeG9HTRU7GvKQERVGD504vHcIwY4BZwtRXN5dSNWW7WvndymWUyjFFsMwFltQ4lrO/ebkzCp8MYsseuh8pQueSR/wIjhX2dYSTOzGC5ZyLLVYHrpWBKELnk48MXj6LqIoecxUreaN01OB3G7wMCRSyWYuSzsySoVqRZvO2UJyLpuPsIF10hsW28o7nAPvLQvYh457Du4h+ogvE1wJSBY0gfMQe8uqaXCU4LZVACwHYJas7xzOd5SqdN2EMVXGqgxZyVX59+OrN3Vfz6zC57PAohc6Vwne00WPE3BsEyJwwwGtSq2ZkvMuiSI4fOhwPhJjoOssD16TVbu8c0gXbAflgtbKepNZDwOpCGMWC+e2OaDomAaxI8VnqkIeM5uhUlRo/j4eOyX6rqPvO+Ks4/z+eSZxwnCkjOuKFCDbtUcn+ODILaGk1RI5gkUV3jmC8/R9RIHZXKgKq7FwMmRSLsB/c4VH73clUSfawjDw4s3LRgDX8syCojjnCe0b6XU7xJIe2pIjarsGTy1CKZgHvM7kUllvCuuxkotjKKboCRY7z+cdl/ameC/0nbdj5NoGvxrZZOVkKJzu9RYOWvxHyYkijlKqxfNems+ooAW02jdysju+7O8Or9rOKfsZQmil34wKjPnmSyJnVuGzPtK7hK8VL47oHM4LsSU6aqmUXFuMW1AVYoiE0N/guVd1VDWFl5bM8DHiCYyrwpAq603lm1cHxpQ52WQ2qZJVGYsniOKjI3i4f3HA//Ga+5j0gXP7HRX4/Nev8JUrS66cbHj2ypJcC1VbBk3Ai4V/42YJKTGuM3mwWL5MrDxqroTiXSVGD0CIDhUliiKufZdacd6ztzeh63tWm8R0Pbz6ufRXQoL3eCnbRd0cMdmVDBXZVkFvKIxIy5O/JF67TmRXYy4W8uRKSpUxKakoudqxXWqLoJqP1nnHohVXFs28zruOeZdYxULwttNy3SZUaKdNJeWMVEjZLIkvjlI9snUOW8QoIi1nYDvdq9ulUsGygd4LwQsxCH30p9/1Zu7rd6SN74LMp1M6FepYrIzY6sQpK1AoZZtSPHVwsoKWbbrS7kOulaJKkLDLTxe1kG05rLl8bckwFk7WhVwqWZ1tTdmaVeiDMgkw95mF29Crp1sXFDiQyL2TfTRHjmbCmAurYUUqCadCTplVVp4fLuPEUceK5sqsBkrf0xWHTmb0LlhJVxzioe8jofpd7F+reeki4MmQlUkQ+kVkSLeBSe+7Hp8zpdW4d7uyVouvq3mnoHgsXCvbvLPILgwrTeEeRbyhSaiWch1TZbkeGVNlSNWccSdN4a1cKeC9Er3SuUIvmU4rPpnvMJXIIkRWEeZdxksi5bGFfUrNSiEzbjKoItsUaoj0qVIkMCk9rvr2zc1ChehxVXabt9SKy9ujwjJD0QdC8LjbISzrJz1RhRq9gQmCmdCUEqUWyIVibm7z2nUXcFuatZUvPYhaDbzSUWrhZLVmTJnlZmgQJTPdIgLevPDmihMFpqHSeWXIlctHa4I4gnoqwrUyZVMjlUrfecQpuURGZ6a8Uhqkycx8zS3BJ6flW63t/NhVS5sDd12NANeK69vULdxQRb1ZObMKn84WTHyFPG+1ZI+qst6sSTmBS2QZ267ZpkhbRcq2pXm0voERCFQCY0lcObzKarPmeLlmyCOoa6gVCM1DDiEQukgUWLhKL8pqzHztmyegQi0OxZF6KNESNrOJpysC2jNmT63ZQkWtjDlZdUst5lesoFJqge0DmmVRrMBnK1jFQr3gfbMa2YAQzaLdFgo3GJKAKw3r5Uy5W4dmW2tWO8O3odAWeNC2LM4HnPPUaoiUWiDnQk4ZUSV6qznWlp0L3uLeLlrcGxxMqERRnPcUaencVqIs2C5GFO9MQV0XG7atQ0QpNbMeV5RSWEsGiq1HFBHFzpjCdoubA7q9Eadgi+3TKttt//LlzCocMYgQfnt+A2qYM6RSBUp72iHNm2/IFKCIWhkz9ngXGYbMOGSGoTIuN6TViolT5rNoiNZsygox4ENgbz7lwvk9vEAo2cq03lO9p5TCIBvbZX40xAuePgQQz2x2DvAs9mbs7c8Z04prh19nGDdcuXrC8fEaH4QYMsFVhAErmji02nWI881xdIa5U1q9wCL9yvb7OtjBMf5zObMK192BJjc8b1GZmAXwHrQpHNn9n8EharthAfEBRJuJbajPXPC9YxodtW7NrKFnfAjM+sjetMOLQHZItcwc0kCH3rJd3ikipWXEQMQRYodzHfv7e1y4sM8wRmq9ymZU1usNw+BwznLvZpkqW5itqrvuLLc/BTnd8dt6OTfel5uVM6vwnCuJilSDIG0zZWMaKCVbPbgoqKBiHu4W31W0MOaMOGE2F7oOSh7ROuBILHqlV9ifCYuZKTwlAXHM9s/RTxdMOs9iGkEbXKooMXaE2OFzRoMBH2azGV3XkSuMFbwPLA7O0XUT5ntTFntTxgQuXGAYZoxDIQ+ZosUSQcoO3SrOI+pPqwMN63YKkqjtCDP5TlR+ZhVeciVLaQq3HaC1krYKr1CKpVdrA4KPKbfERmY9bgzhEiLiPaUmtI44EvNeUSec33OcXzhKFcYE4iIH9x4w278HRyFoptbC0iVKUUIfCZMpJWVqC/T39vaZT6eMObMeR0KMXLy0z3Q2p59E+mkkJUeMFxiGDcfXlqyOVqSUWGWroIkEnIsoAVFL2ZZWydtmGbSVeHULaNyeci/zLD+zClfZ5r5bLu26TJRzjtAcOUXIxXbBmAqbYaTUQi4GKBzHhPcb8jCShg2UkegU1wn7i44L53tyUVabChKYzTsms44ghpatpUDw5JKJkwmxn1JLZdJ1pvDZnEnfE1ICDy54QlCcz1SUlDMpjwxjYhxzCwE9zltRyHmrDNR6fd0PtsHXNgOnu2ftD/nOfLazq3DzfathibkuteiDVblcQFwkF+VoaQiQ49XA4dFJw7AZCtWHY3LeUFYb0smS6Ar3TCvTLvDwA3u8/pFzbMbMN4/WFHVMzi3oZnv0Xc9sOqfWyvHJMSmN9JOOftIhuvUYpRV5PJthzcnJETilm2ZcWJNqYb3OjEPh2tGaccikpHjfAQ6ZNOxaFVLCoFYtf7BNESsN56bsnNft+X26OG5ezqzCRa7Lm5/CRAHzTHf/j7bGv0LOxXLWoriGVM0p4wVKTha/+try0I7JJDKbT5CQ6IdEVkfsvEGaJz393BQ+1oJLkUkf6HvrERO1nelbCbNooOsDKtpQORVKQdUaERULp5z3hBgbmGNb2291d9lBNRoEq/Wo7Wp/so3LrkN1vLyT/MwqfDLt6J1HU8ODZwMWmjer5FQsPEqFa9eWbIbEyWrFMAyWlApCFVjmkY1AVKWv0MXAYjFlfx7ZO7jAdP8CdTPgN0otINGa2Lq9PQ7ufw2qSjg6JKWRThLRJbYOJKh1hdRKFxyum+8SPyrgakVqxfWgQclZif2CvZOBnAbGzZJaC8PQsOriqdsO0Vra7tWW2hdLDjXwpvWWCXiHL/7b3cob5MwqPMZAdIJSESnmkYtlSLRCzQYfHofEarUy7NdmIOe8S5MqWDOfKuIdk+Dx4pn0U6bTnn66oJstGPH4bknJigQP3hEmU2bnLth7OE8eB0JdEXSFHTjJooaxoLnivSPE3hbjFuyI+VeuCho8tYCTjskkkYY1w9KTc0J1yTDCttWmctpsYM6hpVmdc6bwbbKhhafiboM4PKWR4ApkA/Zt2+3EHHNLprVccymZnBNOlC76U0gQoGNCS6GLkem0YzrrmZ27yGx/RvYLri47lsvE4ZFVow4mgV46qJ5xtPpzN90jTOYEneF1g2qhVPvpWjdozYU6tpRna1aI3jPxnqJCzFYNm8wqZYQ8rhmWR6RxoPAcnCwZU+tFh5ZhczuzryJsq95lm3hRkFYculk5swof10tCqLiSUCy5YufWti4slNZXVtJIGjeEEJn2RhIw7SKCklUompj2HfO9GXv7Cw4uPcL+uQNGKs8fKsuTkRe/aWHf4iDSuxlSI5tNJXSB2cE9hBgREkJCa6LkE2stoiKqpM3IsFyhpbUQVSVMJoTphKKOVQ1UHI4pQqSkDWl1xGZ1wpAMuVJO1pTN2jav961w1horZJsc2gapYt0tzX+5WTmzCrfSUkW3Z9m28e4GyEPLZzf4sFRFquXFrX9Q2tnn6LqO2WLOdL6gmy6IkwXr9Yr1esWwKWiL6aWFuFoLaRwtxYvHha71iXeoJvBqCq8GT3LF4UJBXQUxfBzeYmucx7seJ57gZng/IY8BJ9Zb7mPAeQEHSjWfTNx1DqstcmtX113D6/Y+nTYQ/+dyZhVu3ZoZTcb/og2mtDXtWu1vhcy6KqsCroLL0DlFcyY4YRY8ceK5596LPPK/vZ75Yo97H/peJrMF1/7H/8vX/v0bSB2J1VtfmSouJ9JyyaCX6WcL9u65hA8zQt/h+w6lQF2jmsmbNTUNqBupzNFaKWk0aHKtbAZFfEecnseFjunsHF0/p+QlaZiyOp7Q7fXIMTBUakiWbNnCfNQWvS33bbBe7B6ooMXxMjb42VU4YDv8Jd+mtujT6DBU7QzLzadz7WduWDAXDd8+mU7ZP3eO2WKPyXyfbjKnFMfyeE10hUnniA0giRZqSaTNBuc7VFtOPkxw3RQooL69TqjVIcHjouXppTpzNMfEmBMeoXMd3vXEbmqfnUFkTUprXHRW//TtC8Du6NLmwimK7HD02yzctuXqdtjhLYWouzpxAzRsmwuqQlKDNLX7EKOni4HewSJCFxwX71mwt+i59PB9XHz4QWI/Rb1jzCNaBnwd6IPjYDGh7wMxehQlhECczeimExxCzRXGgkoBTmk78gAlOWrxFI2UKhyvl4zDwNHxIYdHh0ymMy4pTCYzumlHP+2pdcM4rhjTGlzBR+hmnmmOALv6gBbL2au2n7WSh9GcxBKoNZJuizMcdl8QYNtjbSetmnXLCrna2a3WrTLpIxOvzLvKpPNcvHfB+Qt73PfQvVx8+CEQz+o4MQ4jWkZTuOvYn/dMJh1hixoNnslsSpxMEDWFq7T8n1oHSFWlJqEWQYun0pEVjleJ1WrN5ctXePHF51gs5kynHaUuWOR9YM8UnpaktEal4CJ0U8dMDNmjroWVuVqjY5Vdu3LWkaq5pZDr7dFbtq3/lmp1bXFtxe/KpYpIxYm1Ek+LMomeLhhbxGQamXSB+f4+i3PnrAo23zMw4NUVm80aKJY9m3T08zndpCfM9vDTGWE6x3d9g1Zpg1WNzZcsDOOKWgt5WFHSiNZMzYmcE3kc0JIRKj6AuMKYloRBGYcVOa3JabAybS2WPq7akC9WHdTmoFG3mPVqXTdaCVG2BWG02mlws3JmFV5LbRBiPeVIcULdshuVipNCFzznFzNmXSJ2gRCE2aTjnvNzprMJD7z2ddz3wCXOPfgI5+5/mPVyxfLL/5PDy98AHTh/fsZif4+DBx6gm84Ii3twkwUu9oR+jvhI0UpNG/IwtMrawMnRITmNrE8OSRtz4KyJX5vHrQSfWSw83meOl88xjBYpdJ2FZWVYU8bBcgUpwTgg4wrVgpLZdqeCRRzBG9YpOKjVkwbHsBZkvPn7emYVDrQUowEFioJToeqWI4WWhbP+MdVqLcUCITi6yYTJdMZkvmCy2KefLgjdBDckSsmkcURQO/e7SJxMidMpfjLD9TNwoTmFSsoJqQ2zXpRxHFgvj5vCjxk3TUklIU7ophHf+su66HHOUrG15kblZWVXVHe+ilS13HvOoAXVtAvRQBvTlGeLzxQH1YOPgq83n08/swr34hhLZb0ZwVfIpugYA955a9iXSvUwnQRiFDZjYrMZme8dcP7S97J3cMD5B3+Acw88wGTvANwcdKSOlTqMBITYT+jn+8wv3k8326N2e9QwZbVccnT5SusNN994TCPjmEjDwPG1K9Y+NG7QPDYYneKDZ1EXdH1kPg2cP7jHlOWNZXE6OYeXqVXyfKG6kVAUlxK6WpOuHVr/m9qCqNJSEC1vLiL4PiDeGZvjfk8cboMzXESskT4XtEAtxjrYqeDDaRswTomdI1QYxmScZRKZnbvE3oV7WJx/iPn5Bwhdh8gE1WDOT8oIgo8dcTKl3ztPP99jcDNUOtJyzeHJCTml3Xk7DBvGcSANG46vXqHmhGjBaW3kQo4QA9OJUF1Pv7fPuf09EE9xPeICXZwjEs3/cCPRBcv1l4IOI3m5MqBDsQpbbWlzFajOId7RqeK7gI9CP42ou0XEfB/60Id485vfzN7eHvfddx8//dM/zTPPPHPDazabDY899hj33HMPi8WCd7/73S9hZbwZqS3Z4N1pW00MYg2GKJ6Kp+CoRA8xOs7fc46HX/swDz3yEJcefoCLD1xitr9H7CcgMK5OSOslWq2EWrU0cMKGYXnE5uSQ5bVvcnz1Ra6+8HVeePZ/8uJX/50rz32Nay9+ncNvPMfh5edYXrsMeYnTDVEyMVSEkc1myXp9wsnyhJPlkvUmkRJUjUym55jNLhC7Gc5FUE9OlTRWcsaaGov1m49FSVWsgzXBMMJmFFYDLDfK8apydFJYbYz6I98qk/6pT32Kxx57jDe/+c3knPnt3/5tfvInf5IvfvGLzOdzAH7913+dv/mbv+Ev//IvOTg44H3vex8/8zM/wz/+4z++LIWXZAwLXXAGOY6ukekYxda2qV8EpLNK0rlL97N/8QHuuf8hXv+G/53Z3OBK3WTO5uSQ1bUrrI+uoWVEpJCLoWDC6oSTq5eJ6yVHg7LO8PWvfIV//cIXqDm1ggykPJLLSAyevUVPCO38j4HVesPx8THiPMUp3XpKnM6ZH8As9Cz27qfrpggdaCDryLApbDaFcVTGLGwSrNO2L631qSVjpcoKA0ZBxlDAVy74QreAl0HE+PIU/olPfOKGf3/4wx/mvvvu4+mnn+Ztb3sbh4eH/Omf/ikf+chH+PEf/3EAnnzySX7gB36Az3zmM/zoj/7ozX+Y7qh6OAVtaktAaSPhs4zbFvbUTyYs9hbM53O6rifE0HBwI2kYGIcNadwYP6oqtZZdpS2NGxBIg7UPp2FDHjfUko0Y1wmUjNSCaxRgXgwdq7VQS2mOmOHenTdKzO0jp4JzFa0WwqVhMPxdLi136Kk4slr2zLWmaG3VQcCQNjvKL3Ngt2XUm5X/0hl+eHgIwIULFwB4+umnSSnxzne+c/ea7//+7+eRRx7h05/+9MtSeMXiTyt+O4tTxfhZToE9Vuf23uNj5Pw99/Ca176W2f4FJn2HEzi8coXNmBmWh6yufZNhdcRmvTIlDwNptcJ5x9XLLxC7nnXxjOrI62N8Sbha6MQRjNLB/IgusDeb4pxjM6zZrNfG9JQL4pQ0DKCwPDrhan+F9TIBc2I3IY2mZNGMlIFhWDFqRwl7DHLMsk4BJYRoNfDOE8TjxdGJZQHHkihacK6FjPpdcNpqrfzar/0aP/ZjP8Yb3/hGAJ5//nm6ruPcuXM3vPbSpUs8//zz/+H7/K8E+UdHR9tPOK0CbUMXbbll3TbaWwFBxDDqfd8zXyyYNO5zqrLZrFkuNwyrE9arE9J6teMbL3m7u0eG9YqaM4lIIVDzaNzpVLyo0Wo7a+WNwShEnHOs17VxphultVZzNkuxattmvQENrE6WhJgZh4GUR2twcJDGRFGHuo4ikaS+4esD6lwj5DV2QXGNq31s1kasdPpdqZY99thjfP7zn+cf/uEfvtO3AL41Qb6l2lqlyFlnCUKrEeqOjVC8Af9j7Ll29ZB/+x9fppsu2PvGMc5HcjFEaBpWpPWaPNri8s4TQkRjTwwRL+4UDYtj0kXmsym1ZgRFS0bF4XBoKWzWBoNOY24020LnAyoOSqGMI6ujQ2qqxK5nOF7jfWDMI6VkC+O8IXJOTq4xpAH1lbgXQRTXGS5OQjE0bFt4AKEIWgMHF6ZcuLjPZrjFufT3ve99/PVf/zV///d/f8PIhfvvv59xHLl27doNu/xbcaXDtybIR63ebMWT1jvW0q2ocbRUKk49oeuIXce1q1e5eu2E2M+Y73+TEDsWe+foJ1NKHinjhpoHBOs5C6FDYm/m0zkjsvceJ5aTn8+nlJRIw4paMs55Y2MoyrBuExFy3rEuRx9QxCxIKSzHQ1ZHx3gXOOkuI+LINVG3VGHOWbjVzmr1StyLVlPv7XiQUBCvOBWCGlFRJOBxXLhnxr33Htw6hasq73//+/nYxz7GJz/5SV7/+tff8P9vetObiDHy1FNP8e53vxuAZ555hq985Svfkiv9WxHkb1sEt1AmZzjdXe/YFr3qG84LVUpK5NZO5F0kxI4+BLxUas7UbA6TF4cLgTCZgKs2gsJHRLwBJlUp2wrULuKR3YIz1q7Tlh/X8PHbPP+2V6D9go3UKMkyhDW3hWyuiWHmWnLFJXyvqKvULoNT1AvFASpIMQvTOYcXIXSOrovkeosIAR577DE+8pGP8PGPf5y9vb3duXxwcMB0OuXg4IBf+qVf4vHHH+fChQvs7+/z/ve/n7e+9a0vz0Ongf9ECI3bxejQr+u4EI+KP0Vy5sKwXrIeRpzzbI6uEkLEpxWy2LMb296z7zq86+j25nRbGjBnvdrDyZLVMLLZbNiR34qzuB3QWq3f3PDFeBcI3sAY7bRB6rb4QYswCpSB7RqQxkszlIRKJfuMukKdJCbTRHGZ1G2orpC1Wgq5CGEUgnjm/QW6MGW6F1nsLXAx3fR9fVkK/5M/+RMA3vGOd9zw/JNPPskv/MIvAPD7v//7OOd497vfzTAMvOtd7+KP//iPX87HNNnu8O0uPw3RLKFsmTfXOiu3tfOaE7hCVkVKpoxrSooNtG+dmILHOU+Mjj5avry03PUWVmQKO+0CYccERYsW7Hq2LFFuFzjqjvV4+2+AWo1ObEfFXYuREUiliileJOOigSByqOBa/r2xRpuFU/BWoLGpCQ7vbhFM+WbivclkwhNPPMETTzzxct76JSIEvA8Q4q7hADF+ci+CCx0hTqAVUiiZ6GHW+92wGu8LkteUtaDizCo4TwqKamDSz/GTqSm4GCXHfpwx08q1b36T1WpjyiFRtVqXqtCIdy3b0UXBB29ZwdYtItSm1NpGcFRO1gOlKGPLrGXNJB3AK/05xfdK74XpxBksuikx5cwwVlwRfPLgBO97usnMOm+SktLtgHgRhxNr9T0dYKHNhAZC7Oj6qcWlje80CPjOI2qTiZyrUAZqMrYGFQ8hUJK39xPBx952a4M3zWbWfDhskoEQc8Wq0Q1xg+3anC15433Aa+vl3tbq3RZ6ZJskl8LJemBMhfW6Mo7G+5IZcBH25kLXCdEFfGfEQw7XyHwgZ0WKAT58kEY0OEEknHbR3qScWYW7tqtLawrEN6yLeFRsRkipZceH7gTrrW670IvDid8xMYrzuMbiVGtGUyXnRozrPHE6B8QcNiD0U/bP30MeB4a+p+SRnAZyHhqLhHXAlFzIYqS/rg3L2Tb65ZRYbzZshszqZMOYC2MScgG84qPHRzG2ia4ld2oLQbNRevlNR9xAFE/vOia+ZxbmTOMUsrA8Glitb9EZ/t0U58QQoDm3yUOdecMSUPEUQGuyJIWzRIz3bZSEeLzvLK7uZlYRi5FJHym1crw8sSF48wVjLnSTjvn+ecR5jpYrxmGkm5/j0sORkhPDyTXKOHJ49TKH165QymDk+aXaEDyFGDGGRKSBFirDsOHo2iHrIXP1ypoxGXQZF4gTR99FYu+YzSZMZ4EQCpIzFEGWHVIq3crhh45JHzmYz5jEnguT8+xP9zk5SVy+fMxqffMIiDOr8OBBnbUZ0ThNrh/nJLXuGurcNteup/NFZMsLA7t26h2dV5tikFOjrBZPnzLilTSOjMOA5tTQoFuIVaPeYJvcbsDhLZ/6dcNmLFdfb2jgD95ZVjBY83+INrclRuNR9WIsEySxztTskAy+ekSFjo5pnDCJPV3oiD6CZvKYSbc68fLdkMUcio6wWVIJFp5UT0pQSmDXMdB2togQO0OYQJtroo6SMlqtjBqcM7RpFjQ7rl2+ytWr15jMZlxYHuG85/DwmNVqg6EkTYGe0sx3xbmA95UYe6qzmH8cEjllxnFkewxpmynWtdlns+keiMN3PT5E8CARxGvrsFHqupJXUIvAJiBFmbgO74Rz0wUP3Xsvfd9zfv8808mM4ysjq6M1q9VtsMP7DlKoeN9mfNXSAITZWsZF0UZeq+pbh0lH9baTW/tGgxMr1ftW0VLTY4XNuGaTR8Y00k9s2N3J4THr9ZqtL+AahMo5aTn8bZnWKC9ryc0bp7X8GJfM1hoF74z0N07tZz8hdNF41KUhYWVsFTcxnt0iyCBQIXQ2tmsSJuxNF/STnmk3oQvWp56GRBpvgzN8b89CqInzDEk4XtuNHVNjJEZOOzSiFTZyzogbLHVJ2c00EzGOl2Ez2nCYdSLnylgGxjJQy0DfKd57aioEtX6tMZni1i1jMo6JlBKi2vL30TJnte7ialRxwbDl2xloThwhmA+Ct1Pemv8tpMx6anlKcrgKUu3aF/2E2azjYG/GYj4lxGAl13HFajWwXmWGzW1g0g/2PX4SGPrIat3oK5MwjpCzp6inVN8SH94qlzmDVIoTlMF2aEuJ1NKaBQpsBms/znXTQBABz4YQPH0/JfqOkgaGkxWlVoYh2fSiljLtYmBvMbOBdo36ehxH1hvb1c6bdZj0xpfunBBajF62YVTjklMEslBqU/bom0WzMG9vMuXgYMbB/oL9vRnihKvXbATGybGZ8/XmVR5F+UqID4IvRintvRKiFRlCZ2GTFEWzUXsopQ2FMWCAmX5DtGrjIa9GyNBIanOzEgXEzupainVo5gwqNkpat86YNozZ9r0tDStSraNbtpMLrF695Umvqm1CsRjgVGU3FhNg2whoIyhbhOHcjnPOO0fsOyaT3qyJGrHgajVwcjJwstywWic2t4PCJ1OPUCnrkdjDdCHE4qCHPsOwqayX2jJaNrJCXG/VLDWqTrYdKhWqlqb0wiatyaUQgjOstxRyStRajKFZvLFIbXu/SxtQh4A6qlSGTcJ7R995XPD4AJNpW1A52byzMRupkHN0fWwjMFtyZtvpT5te7D1hy+0u5tV77zk4v8/F+87jo6cUYRgSX332Ci9+44jnXjzh2eePbg++dN/A/OKqFU8i4JWoinrbrT7U1rRhTprumuykhWKnefFtOFe1mvI1Ax4RuwXb3q0tH8vW2Tt9tAtTc/q0tME6ailQg1m1RdpCw1KMdlO9xxffosaWE2/vteV12bEqOtdID4w834dA7CI42U0iXK5Gjo5WnCwH1ptk8f1NyplVuIsO6T1h4umqMI1CUUFGpSuZ0IGPSikwboxH1TvFtTHOMcwM6lxpYyJzK1Hae/s2rtk5j3cB5zp7Pdv5ZdJmhF6v7O1PtfFXCJpBxW64VdAsrlYRSk6MQ0JcJteKa1ODfbBZqa6915Zms4gj4dFSyWPCO+Fkk1kNp9OFT042/PuzV/j3r3yDw2XmynEi1dsgtSre4YMjdJbJ6jtTuMaKL+C8Ik6t7dY1rtSWcvfO48MEo8xQ1ClaRwQbPeGCg2q1dFsgW072linX7VGx7dxk2w1gF6etS6T1Z293rnOWAwjOt5lnkFMGEXLV3cTE7VhMtBHzNFNfcGRcG0ZXEIH1YGM6cikM48jh8YbnXzziq1+/xirBSYJ6Oyi8FIeoBwmIKEFahyjgihqCU8UUjlCKUHPrz5aMMqC4RscJuaQd7swTLFSSQHAWNmlVc+Bkm89p+Xlhx8oE5ng5L/i4HVMB2zkdO/I8Z9FBCI4YvZH3akVLCx0FfGuStHVk/sEwZk7WI2lMHB8uUYV7L5ywt5i2vENpJEB2dHnv6HE28fgm5cwqfBgcQgTpca7QRWuu80afQt+JDXYvsJk2IP9QjEA/K+Og1CqM2eiytkNxDIrU21D2NjAHVbS0iUQNqOhbmw/uevK7Bon2Qte51i7kjBVET2eh+mDncJwEJtqRciGvB8vKUanJYFmuoWBzNajy0XLgG1dXrFcDLzx3jZoLs77Di9JPPPP9wJisVRhRm9QYbZYZLG/qvp5ZhecEWRylbM1sK082tKp3oN642Do1jjXTzJbmqxUhnJn+UzjKaQatuUtWcWvOWXt74EZ00xaQsc3Vb+ei3HC8t6T9Dimjxocuch0NV1UKBWlTBhHIKFWFccwMQ2IYEptNombLAaQxW9sxseHubOS18bp5y8HfpJxZhR9ehQ2eMgRcJ8SZgfqcr0Y97SD2AMJsZjd+HGwEdMrCZjWSi7BeOVIS0qAMG0Ad3jgdoOG8TREGSpSyxaVdx1q8Y58wXjScDXlli8sojVGpYeZrSohWchqpybJ7VYTi/O5YCW3EpCKsx0wqypVra755ZcWwySyPN1CVcVPIY8VPA4vJHp1L3HtwjuFCIVfHWByp3AYK32xsAHwZvZ13wZmytxgDv0M5WT+2CN5XQqqEBFDIGWoNOxhUbvNGpKE/d5WvXRdLQ/UIuyrXVuvCdryUnbtmUGTHTS+q+KZ0agI1uJWWbM4d3n5Pt0kcw5VXhc3GBu2slmvWy7Ut3DGfMk8UEHX0ocOpZz6ZsDebMmbwCcLtoHB1UyrnKQ4z0YyoVlwdQLLlptoQ2jix4kZo87dzgbiBkqGbBsbBMwzQLY06Q3M0sIQ2fvLr0KiGhjbv3JCrjQGRRsrvDHyRDDXdqDet381pq8qpx1EQGVFJFho2z118wdVikUSMOIWOHgmFeY4clI5SlL19m6B43wOXOH/xIufOLdg/d5GcCxfvO6JoYDMUluvKkG+DTFv1MyqR6mcUVyiam8JPQEZLZ0bw0TGZB3y0URAq1WLzsVKKsF5FxtEzboT10lGLkEbX2Jc8Wi2U2rrh1YZ3W+dIah61D40M1xy1oiCpTRseM5pah4pmnCi9JCupykiVwcLJYlm0IAa/Cs4TfLAF5jKhVrKM1DAiInRdT4yBhx55gIv3X+Rgf8H5ey5SSuGBB9fEbsrJMnF0PLL571wt25rW1SaRSJQx43MlY8140RerRxtHAD4p1QkhVgurgFyVMVlSZrOxgsu4ETZjNYUPlqbcjoA0adzkrXxaig3CAQihoVOddasWhU2Lf/OYrE6uxa6RikrBU9hkI//dVBiqJ2FHgcMwDurtM1M+HbyXiiVofK24NiR+PSTiMLJcj5YaHjNDyow5N5/lRorxbyeiL5dS/xbLV7/6Ves8uSsvW5599tkbOoH+IzlzCq+18swzz/CGN7yBZ599lv39/Vf7kr4rsm2x+k6+s6pyfHzMgw8+2Ga0fWs5cybdOcdDDz0EwP7+/h2j8K18p9/54ODgpl73sig/7sp/f7mr8DtMzqTC+77ngx/84H/YVXq7ynfrO585p+2u3Fo5kzv8rtw6uavwO0zuKvwOk7sKv8PkTCr8iSee4HWvex2TyYRHH32Uz372s6/2Jb0icjPUpe94xzvYDufZPn75l3/5lbsIPWPy0Y9+VLuu0z/7sz/TL3zhC/qe97xHz507py+88MKrfWn/ZXnXu96lTz75pH7+85/Xz33uc/pTP/VT+sgjj+jJycnuNW9/+9v1Pe95jz733HO7x+Hh4St2DWdO4W95y1v0scce2/27lKIPPvigfuhDH3oVr+rWyIsvvqiAfupTn9o99/a3v11/9Vd/9ZZ95pky6eM48vTTT99A3emc453vfCef/vSnX8UruzXyv1KXbuXP//zPuXjxIm984xv5wAc+wGq1esU+80wVTy5fvkwphUuXLt3w/KVLl/jSl770Kl3VrZH/iLoU4Od+7ud47Wtfy4MPPsi//Mu/8Ju/+Zs888wz/NVf/dUr8rlnSuF3knwr6tL3vve9u7//4A/+IA888AA/8RM/wb/+67/yvd/7vf/lzz1TJv3ixYt4719CqP/tqDv/O8qWuvTv/u7v/lPAwqOPPgrAl7/85Vfks8+Uwruu401vehNPPfXU7rlaK0899dS3pO787ySqyvve9z4+9rGP8bd/+7cvoS79j+Rzn/scAA888MArdhFnSj760Y9q3/f64Q9/WL/4xS/qe9/7Xj137pw+//zzr/al/ZflV37lV/Tg4EA/+clP3hB2rVYrVVX98pe/rL/3e7+n//RP/6T/9m//ph//+Mf1e77ne/Rtb3vbK3YNZ07hqqp/+Id/qI888oh2Xadvectb9DOf+cyrfUmviLBrd7zx8eSTT6qq6le+8hV929vephcuXNC+7/X7vu/79Dd+4zde0Tj8bnn0DpMzdYbflVsvdxV+h8ldhd9hclfhd5jcVfgdJncVfofJXYXfYXJX4XeY3FX4HSZ3FX6HyV2F32FyV+F3mPz/QK9lzgUp7qIAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 1000x100 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_sample(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'frog'"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classes = [\"airplane\", \"automobile\", \"bird\", \"cat\", \"deer\", \"dog\", \"frog\", \"horse\", \"ship\", \"truck\"]\n",
    "classes[y_train[0][0]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x100 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_sample(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'truck'"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classes[y_train[1][0]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50000, 32, 32, 3)"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_scaled = X_train / 255\n",
    "X_test_scaled = X_test / 255\n",
    "\n",
    "X_train_scaled.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., 0., 0., 0., 1., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 1.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 0., 1.],\n",
       "       [0., 0., 0., 0., 1., 0., 0., 0., 0., 0.],\n",
       "       [0., 1., 0., 0., 0., 0., 0., 0., 0., 0.]], dtype=float32)"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train_categorical = keras.utils.to_categorical(\n",
    "    y_train, num_classes=10, dtype='float32'\n",
    ")\n",
    "y_train_categorical[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., 1., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 1., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 0., 0., 1., 0.],\n",
       "       [1., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 0., 0., 1., 0., 0., 0.]], dtype=float32)"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test_categorical = keras.utils.to_categorical(\n",
    "    y_test, num_classes=10, dtype='float32'\n",
    ")\n",
    "y_test_categorical[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/50\n",
      "1563/1563 [==============================] - 82s 52ms/step - loss: 1.8130 - accuracy: 0.3546\n",
      "Epoch 2/50\n",
      "1563/1563 [==============================] - 63s 40ms/step - loss: 1.6217 - accuracy: 0.4267\n",
      "Epoch 3/50\n",
      "1563/1563 [==============================] - 57s 37ms/step - loss: 1.5395 - accuracy: 0.4586\n",
      "Epoch 4/50\n",
      "1563/1563 [==============================] - 55s 35ms/step - loss: 1.4822 - accuracy: 0.4799\n",
      "Epoch 5/50\n",
      "1563/1563 [==============================] - 58s 37ms/step - loss: 1.4319 - accuracy: 0.4974\n",
      "Epoch 6/50\n",
      "1563/1563 [==============================] - 56s 36ms/step - loss: 1.3896 - accuracy: 0.5112\n",
      "Epoch 7/50\n",
      "1563/1563 [==============================] - 56s 36ms/step - loss: 1.3502 - accuracy: 0.5273\n",
      "Epoch 8/50\n",
      "1563/1563 [==============================] - 57s 37ms/step - loss: 1.3160 - accuracy: 0.5367\n",
      "Epoch 9/50\n",
      "1563/1563 [==============================] - 57s 36ms/step - loss: 1.2822 - accuracy: 0.5520\n",
      "Epoch 10/50\n",
      "1563/1563 [==============================] - 56s 36ms/step - loss: 1.2510 - accuracy: 0.5617\n",
      "Epoch 11/50\n",
      "1563/1563 [==============================] - 56s 36ms/step - loss: 1.2229 - accuracy: 0.5727\n",
      "Epoch 12/50\n",
      "1563/1563 [==============================] - 56s 36ms/step - loss: 1.1940 - accuracy: 0.5823\n",
      "Epoch 13/50\n",
      "1563/1563 [==============================] - 57s 37ms/step - loss: 1.1655 - accuracy: 0.5924\n",
      "Epoch 14/50\n",
      "1563/1563 [==============================] - 62s 40ms/step - loss: 1.1386 - accuracy: 0.6032\n",
      "Epoch 15/50\n",
      "1563/1563 [==============================] - 64s 41ms/step - loss: 1.1121 - accuracy: 0.6128\n",
      "Epoch 16/50\n",
      "1563/1563 [==============================] - 56s 36ms/step - loss: 1.0866 - accuracy: 0.6218\n",
      "Epoch 17/50\n",
      "1563/1563 [==============================] - 55s 35ms/step - loss: 1.0616 - accuracy: 0.6325\n",
      "Epoch 18/50\n",
      "1563/1563 [==============================] - 56s 36ms/step - loss: 1.0339 - accuracy: 0.6400\n",
      "Epoch 19/50\n",
      "1563/1563 [==============================] - 65s 41ms/step - loss: 1.0089 - accuracy: 0.6504\n",
      "Epoch 20/50\n",
      "1563/1563 [==============================] - 70s 45ms/step - loss: 0.9830 - accuracy: 0.6602\n",
      "Epoch 21/50\n",
      "1563/1563 [==============================] - 69s 44ms/step - loss: 0.9586 - accuracy: 0.6684\n",
      "Epoch 22/50\n",
      "1563/1563 [==============================] - 69s 44ms/step - loss: 0.9351 - accuracy: 0.6753\n",
      "Epoch 23/50\n",
      "1563/1563 [==============================] - 69s 44ms/step - loss: 0.9075 - accuracy: 0.6875\n",
      "Epoch 24/50\n",
      "1563/1563 [==============================] - 69s 44ms/step - loss: 0.8809 - accuracy: 0.6938\n",
      "Epoch 25/50\n",
      "1563/1563 [==============================] - 69s 44ms/step - loss: 0.8568 - accuracy: 0.7044\n",
      "Epoch 26/50\n",
      "1563/1563 [==============================] - 70s 45ms/step - loss: 0.8305 - accuracy: 0.7135\n",
      "Epoch 27/50\n",
      "1563/1563 [==============================] - 69s 44ms/step - loss: 0.8039 - accuracy: 0.7238\n",
      "Epoch 28/50\n",
      "1563/1563 [==============================] - 69s 44ms/step - loss: 0.7794 - accuracy: 0.7335\n",
      "Epoch 29/50\n",
      "1563/1563 [==============================] - 68s 44ms/step - loss: 0.7567 - accuracy: 0.7403\n",
      "Epoch 30/50\n",
      "1563/1563 [==============================] - 68s 44ms/step - loss: 0.7312 - accuracy: 0.7519\n",
      "Epoch 31/50\n",
      "1563/1563 [==============================] - 72s 46ms/step - loss: 0.7057 - accuracy: 0.7600\n",
      "Epoch 32/50\n",
      "1563/1563 [==============================] - 69s 44ms/step - loss: 0.6841 - accuracy: 0.7666\n",
      "Epoch 33/50\n",
      "1563/1563 [==============================] - 58s 37ms/step - loss: 0.6571 - accuracy: 0.7781\n",
      "Epoch 34/50\n",
      "1563/1563 [==============================] - 57s 37ms/step - loss: 0.6350 - accuracy: 0.7848\n",
      "Epoch 35/50\n",
      "1563/1563 [==============================] - 57s 36ms/step - loss: 0.6125 - accuracy: 0.7949\n",
      "Epoch 36/50\n",
      "1563/1563 [==============================] - 56s 36ms/step - loss: 0.5834 - accuracy: 0.8059\n",
      "Epoch 37/50\n",
      "1563/1563 [==============================] - 55s 35ms/step - loss: 0.5600 - accuracy: 0.8151\n",
      "Epoch 38/50\n",
      "1563/1563 [==============================] - 55s 36ms/step - loss: 0.5382 - accuracy: 0.8223\n",
      "Epoch 39/50\n",
      "1563/1563 [==============================] - 57s 36ms/step - loss: 0.5163 - accuracy: 0.8301\n",
      "Epoch 40/50\n",
      "1563/1563 [==============================] - 54s 34ms/step - loss: 0.4927 - accuracy: 0.8397\n",
      "Epoch 41/50\n",
      "1563/1563 [==============================] - 55s 35ms/step - loss: 0.4745 - accuracy: 0.8447\n",
      "Epoch 42/50\n",
      "1563/1563 [==============================] - 53s 34ms/step - loss: 0.4480 - accuracy: 0.8540\n",
      "Epoch 43/50\n",
      "1563/1563 [==============================] - 54s 34ms/step - loss: 0.4267 - accuracy: 0.8642\n",
      "Epoch 44/50\n",
      "1563/1563 [==============================] - 61s 39ms/step - loss: 0.4085 - accuracy: 0.8687\n",
      "Epoch 45/50\n",
      "1563/1563 [==============================] - 65s 41ms/step - loss: 0.3882 - accuracy: 0.8785\n",
      "Epoch 46/50\n",
      "1563/1563 [==============================] - 57s 37ms/step - loss: 0.3703 - accuracy: 0.8837\n",
      "Epoch 47/50\n",
      "1563/1563 [==============================] - 59s 38ms/step - loss: 0.3532 - accuracy: 0.8901\n",
      "Epoch 48/50\n",
      "1563/1563 [==============================] - 58s 37ms/step - loss: 0.3331 - accuracy: 0.8973\n",
      "Epoch 49/50\n",
      "1563/1563 [==============================] - 59s 38ms/step - loss: 0.3138 - accuracy: 0.9041\n",
      "Epoch 50/50\n",
      "1563/1563 [==============================] - 65s 41ms/step - loss: 0.2954 - accuracy: 0.9121\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.src.callbacks.History at 0x220654e7ed0>"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = keras.Sequential([\n",
    "    keras.layers.Flatten(input_shape=(32,32,3)),\n",
    "    keras.layers.Dense(3000, activation='relu'),\n",
    "    keras.layers.Dense(1000, activation='relu'),\n",
    "    keras.layers.Dense(10, activation='sigmoid')\n",
    "])\n",
    "\n",
    "model.compile(optimizer='SGD',\n",
    "              loss='categorical_crossentropy', #because of one hot encoding, else it would be sparse_categorical_crossentropy\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "model.fit(X_train_scaled, y_train_categorical, epochs=50)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 2s 7ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'frog'"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classes[np.argmax(model.predict(X_test_scaled)[7])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'frog'"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classes[y_test[7][0]]"
   ]
  }
 ],
 "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": "3.11.2"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}