Implementing_CNN.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Implementing CNN.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyOjBDUSRgjMgzW0OyepGlFT",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/daaaanish17/Implementing-CNN/blob/main/Implementing_CNN.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "edq2PV2cw3yn"
      },
      "outputs": [],
      "source": [
        "import tensorflow as tf\n",
        "\n",
        "from tensorflow.keras import datasets, layers, models\n",
        "import matplotlib.pyplot as plt"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "(train_images, train_labels), (test_images, test_labels) = datasets.cifar10.load_data()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pqtMnN0y183V",
        "outputId": "de8c3fe7-7794-452a-dc43-3deb6e678fa7"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz\n",
            "170500096/170498071 [==============================] - 3s 0us/step\n",
            "170508288/170498071 [==============================] - 3s 0us/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "np.unique(train_labels)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_ydg6Vmt2paA",
        "outputId": "6714b229-ddc4-44ab-bad5-3c9aeae5b2e0"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], dtype=uint8)"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_images.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Q3a6UVVQ6IkU",
        "outputId": "bcd4b27a-d475-40e9-9f06-fe0c5f463f1c"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(50000, 32, 32, 3)"
            ]
          },
          "metadata": {},
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Normalize pixel values to be between 0 and 1"
      ],
      "metadata": {
        "id": "TDAA3tZf2mST"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_images, test_images = train_images/255.0, test_images/255.0 "
      ],
      "metadata": {
        "id": "KVxONLJu18z5"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Verify the data\n",
        "\n",
        "\n",
        "To verify that the dataset looks correct, let's plot the first 5 images from the training set and display the class name below each image:"
      ],
      "metadata": {
        "id": "pQcdr00r33pU"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "class_names = ['airplane', 'automobile', 'bird', 'cat', 'deer',\n",
        "               'dog', 'frog', 'horse', 'ship', 'truck']"
      ],
      "metadata": {
        "id": "tJXpXWxM18xX"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "for i in range(5):\n",
        "  plt.imshow(train_images[i])\n",
        "  plt.xlabel(class_names[train_labels[i][0]])\n",
        "  plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "kGE4wRSb18ur",
        "outputId": "264ab740-a3b0-48dc-d94d-b17cbf0852f0"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAEHCAYAAABoVTBwAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO2da5CcZ5Xf/6cvc7/PaEYjaaSRZEnIyEY2QmuDA86yXNcpQ+2Ggg+ED9R6K4HaUNn94GKrAkmlKmwSoKjUhi0TXGsSlksWvLgMm8UYLwaMbeSbLFm2LOsuzYykkUZz63uffOh2aux9/s+MNZoewfv/VanU85x+3vf02+953+7n3+ccc3cIIX77Sa22A0KIxqBgFyIhKNiFSAgKdiESgoJdiISgYBciIWSWM9nM3g/gKwDSAP6nu38h9vzOrm7vHxwK2or5eTqvXMwHx92Nzsk2tVBbUzO3pbNN1JZKhfeXz83SOcVCjtq8UqE2A39tqXSaz0uFr9/tHZ10TnPkeHilTG25HH/PgLCkW/UqnZHP8WNVifgRk4+ZqVzmflSrse3xeZkMD6dMhr9njvB5EFPFq8SN3HwOhUIxePJccbCbWRrAXwJ4D4DTAH5tZg+4+wtsTv/gEP78S/8jaDv94lN0X+ePHQqOVyrc/aGNb6K2jVt3Ulvv2o3U1tIa3t/hg4/ROSeO7Ke20gy/SKQjr62rt5vaMi1twfG973gnnXPddn6s8pcvUtvBA89QW7VaDI4XS+ELNwC8cPB5apueukBthWKB2krFcJBdnOQXqtl57mO5wve1Zk0ftfX2dVBbxWfC+yrRKcjnwleCf3zkcTpnOR/j9wI44u5H3b0I4NsA7lzG9oQQK8hygn09gFML/j5dHxNCXIOs+AKdmd1lZvvMbN/M9OWV3p0QgrCcYD8DYGTB3xvqY6/B3e9x9z3uvqezi3/XFEKsLMsJ9l8D2GZmm82sCcBHATxwddwSQlxtrng13t3LZvZpAP+AmvR2r7sfjM2pVCqYvhRe3e3v4SuZviYs13mmi84Z3riF+1Hly5ypKl+lrc6H5Z/8pUk6x3N8ZXf9wCC1bRy5jtpGrttEbevWbwiODxLJEwCy2WZqK/eEV/cBYGTDWj6vHF6Nz+e5vDZ1iasTFy5wVSATkVlh4dX43n7+mlvauY+Xpy9RW3MLD6eqc+kwmwn7Mn15is4pFsKr8c40OSxTZ3f3HwH40XK2IYRoDPoFnRAJQcEuREJQsAuREBTsQiQEBbsQCWFZq/FvGHegFJa9igUuh83Ph2Wc0e3817mzc3PUFkvG6BuIJJlkw9fGbdu20zlvv2UPta0fCstkANDdvYbaShmeLdfWEpZxMpEMKitHMtvmuBxWIO8lALS1hiW73h4uN27dcj21HTr0ErXBuB+FQlhK7e7qpXMiiY+4PD1BbY7weQrEM+kuXQqfq7l5nnTDMuJiGYC6swuREBTsQiQEBbsQCUHBLkRCULALkRAauhrv1SrKJBHCynyFubmpNTh++QIvVdS/lq90b3wzTzIZHFlHbVm2TBupH1Qq85X/F8d4As380fN8mym+6vvS888Fx9+2k690v3Pv26gttro7HalPcPLE2eB4UzZSG7CJJzYNrOHKy8lTL/NtkjJdszmu1kxP8/Mqk+W1Abu6eNJQrF4fK68Xq5PX3Bw+F427pzu7EElBwS5EQlCwC5EQFOxCJAQFuxAJQcEuREJouPRWmA9LHh2tXJLp6gsnhdz8lt10zsiWbdQ2E0n8eOnoKWqbng/LJ7NTvFbY5BSX18bGeT2zrkgiDFI8QeLB73wvOJ79CL+uv+vW26gtm+Wy4tq1XKaEh+WrqUvh7icA8PQzvHtOJlInr72TS3blSlg6LM7y9ywduQXGur5UKlwSnbzI5bwUwpJdrJ1UT084YSsdaTOlO7sQCUHBLkRCULALkRAU7EIkBAW7EAlBwS5EQliW9GZmxwHMAKgAKLs7L7gGwFKG5uZs0FZKd9J5udZwI/tj07xNz7O/eJLaLk7yumpnzvIaY9l0OKUom+LZSQXSBgkA8nluG17D35pz4yeorYtkQ81MTdM5h48d434MD1BbNst9HB4Jt4ZaR8YB4OQ4lz1fep7bBoe5THn8JJG8Svw9qxa5rRKp/9fSxOXB5kz4vAeAXD68za4uLilmSMsoi9y/r4bO/s/diagqhLhm0Md4IRLCcoPdAfzYzJ4ys7uuhkNCiJVhuR/jb3P3M2Y2COAhM3vR3R9d+IT6ReAuAOjp5T81FEKsLMu6s7v7mfr/5wDcD2Bv4Dn3uPsed9/T3hFeaBNCrDxXHOxm1m5mna8+BvBeAAeulmNCiKvLcj7GDwG432oV7jIA/sbd/29sQiqVQVvbUNB2bopnoh05FZZdXjjIry2piCxUibSays3wQoRpIrHlClzWmprhtplIa6Xjpw9RW3srlyl3bN0RNkQkwF/+/B+pbdPmzdS2fQdve9XfH87Kam7h70t3F5euUmVe3HKuwO9ZrIVSbopn31UqvEhoSyuX0Gan+Ta7Ipl5zS3hTLViMdYSLZyBWa1y2fCKg93djwJ4y5XOF0I0FklvQiQEBbsQCUHBLkRCULALkRAU7EIkhIYWnEynM+jpC2dRHTl1mM4bOx7OymrL8sKLl+d4McfZ6XPUZhHpYmomLJVN5bhUkyFZfgAwMDRIba2dYekKANaPchFkhMg4x577FZ2TNi7LlSo8y+v8BV5M84YbdgbHr9u2hc4ZiWSvddxyE7Xtf/EktRXy4UKmhWwk6w1cJqs6l4jHx8P97QCgqZnLit297DzgMnAuF874rDp/XbqzC5EQFOxCJAQFuxAJQcEuREJQsAuREBq6Gl8ozOGVV8K14V585Qidd3bsleB4JZK00tndTm07to1S266du6ht7Hx4BfTEee7HmrXhxB8A2LSVJ5l09vOV+olLfH9+IaxcnDzBV6zPR1pU7byemvCe7eEVdwCYmyWrxXxxH17kqsDBx7masG0HbwM2tL4nOP74k48GxwFgfIInL5VKfDU+n+P+X4q0vWrtCPsYW1mfI23UYokwurMLkRAU7EIkBAW7EAlBwS5EQlCwC5EQFOxCJISGSm9zs9N4/NGHwo4MkdppALbuvCE43hpp07Pz+m3UtmP7Bmqr5MOJJADgqbCcNAfeECeTDSdiAEA6HZZcAKBU5okTczMXqa27GJaGyhWnc06e40lDLR1n+L66eqlty9bR4LhH7i+5qXBdNQB48Ylnqc1z/DzY9b73B8dvuJEn5OT2centlSPHqa2tjVdP7u7pp7Za97R/yvQ0f18KhfCxcklvQggFuxAJQcEuREJQsAuREBTsQiQEBbsQCWFR6c3M7gVwB4Bz7r6rPtYH4DsARgEcB/ARd+c6QZ1SsYxzp8Iy1U1v+X06r7k5XJusj6tkGF7H64hdjLT+OXWEy1rFalgOSxlP5UpnuBRScV5DD+VY+6qwBAgAXgnvr6M7XPsPACZneRZdqolnD1ady3m1bt6hSXxGRwt/z0bXjVBbS5r7kUK4buANu3jGYU8Pl0QfyP2Y2sbHeAisH1xHbRUL1zDMRlqYTU+H5cFD2XCrNGBpd/a/BvB6sfJuAA+7+zYAD9f/FkJcwywa7PV+66+/3d0J4L764/sAfOgq+yWEuMpc6Xf2IXcfqz8eR62jqxDiGmbZP5d1dzcz+qXJzO4CcBcAZLO8hroQYmW50jv7hJkNA0D9f9p1wd3vcfc97r4nk2noT/GFEAu40mB/AMAn6o8/AeAHV8cdIcRKsRTp7VsAbgcwYGanAXwOwBcAfNfMPgngBICPLGVnqVQGbR19QVs2ouJMTYU/ODT3cYlkvsw1njzv1oTW3k5qa64a2SCX3jxyhPMlnuXV0sonpiLtmqqp8LyOfi79NDmXG9OtPLPNm7j2WbXwa7MKl/JSaf6as+1N1NbawW3lQlhmnTwzQef0t/M2VHd+8H3Utu+549Q2GylGmS+cD44XSIsnAOjpDJ/7mTR/TxYNdnf/GDG9e7G5QohrB/2CToiEoGAXIiEo2IVICAp2IRKCgl2IhNDQX7k0NTVjeGM428hS/LqTz4czfCamuftNPTzLq1TmUo1FfuWXmw1nUJWc+57J8MKR5TS3tXXxDLDB/ilq84thuaYY6VFmVe5/a2srtaUiWYdVD++vUuEyZSobKfaZ5j7OzvEsRiMFGJsj59v0eS7LtbaFpWMAeOetN1LbS6+coLYDL4wHx2eneTZiEylkWq3GMgCFEIlAwS5EQlCwC5EQFOxCJAQFuxAJQcEuREJoqPTmBriF5ZVSRBqanwlLK80RWWhmOlI4Ms8LPc5PcxknS5LeOtu5hLaml0s1XX08A2xND39tlUw3teWaw8fx4iae9VaojFEbIpl5lXIk+45kCFZSPBvRItJbTx/PvqtWIj6S86q7mx/fJl6LBVMzEdmzFJZmAWD3zrXU1tMZPn8efJAXtzw/ES7cWo7Eke7sQiQEBbsQCUHBLkRCULALkRAU7EIkhMaWe3UHyApupspXdrvDv/nHSDdZHgfwpi28Pl1HC1+JTRu//s1Nh1di8/OX6ZzW9hK17djGV+pHNm2gtlR2E7XNToV9HBke5n4co8WB0dVHDj6Avl6erJPJhJONInka8EhiTUt7G7WV85EVaLK/bCzxClyt6R/ooLbZea4KzE2Fk10AYP2acM27D/2L99I5f/fDnwTHMxl+EHVnFyIhKNiFSAgKdiESgoJdiISgYBciISjYhUgIS2n/dC+AOwCcc/dd9bHPA/gjAK/2rfmsu/9osW11trfhXbe+NWjbcv1b6LyzZ84Ex9ev49LV9m1bqW3tmkFqSzuX82ZIEkQhkixiKb69jnaeCNPRwSWvdBOXDrNEwszNhVsMAcDNu7iUN7p9lNpKVS4rOrmPlKtcJvM0P1bpLD9VS3mu51VJYkgqw+9z1sL9QGReocSPRybNaxtWiuHzak1E5rvtn70tOP6rJ5+nc5ZyZ/9rAO8PjH/Z3XfX/y0a6EKI1WXRYHf3RwHwfFEhxG8Ey/nO/mkz229m95oZTzYWQlwTXGmwfxXAVgC7AYwB+CJ7opndZWb7zGzf7BxP7hdCrCxXFOzuPuHuFXevAvgagL2R597j7nvcfU9HO19wEEKsLFcU7Ga2MKviwwAOXB13hBArxVKkt28BuB3AgJmdBvA5ALeb2W4ADuA4gD9eys7a2lrx1hvfFLS9+SYuveV2hWW09m6edcUrnQFuXFpJRSSSvvZwHbFI96fo1bRKWhMB8VpiiEg8hUK4/dPW6zbSOa1NXALMzfGMPk9FTh8L2zxS363q3FaJvGexlkfFXPh4VKr8NacykfMj8o7OTHIJ9sSxU9T2jttuCo7Pl3g9xDYiD0aU3sWD3d0/Fhj++mLzhBDXFvoFnRAJQcEuREJQsAuREBTsQiQEBbsQCaGhBSdTqRRaSaZXRwtvodTeRtyMFNeLFTa0mPQWk3g8LJVVS1xCi8lJFil6WI6IhzF5xUnBzI4eniFYrvB9VaqRKpCkxRMAOCrB8VTM+Qq3VTJcEnVE3mxS4NSqYf8AoDnymrMV/p615/k8nwhLgABw/uhEcHzDDl509EIq/GvU2OHVnV2IhKBgFyIhKNiFSAgKdiESgoJdiISgYBciITRUekun0+jsDktAHsk2my+E5RMv8J5cBTIHAOZm56itWOLzCoVwtlm5zKWrUiRDrRTZ13ykb9j8HM+GKpNMus6+bjqns5v3xevpHKC2lqZwPzcAqLDefRbpywZu6+zkBTgnz/HjmM+FJapqlRdXMvDXVa3wc66rk8vHmzYOUVtuPnw+eqQ4Z3dnWMJOR+Rc3dmFSAgKdiESgoJdiISgYBciISjYhUgIDV2Nn5qaxt898PdBWyX7czrv0qVwosDs5Qt0TiqSGxFbqZ+YCO8LACoku6Yv0k6qd6Cf2prT/PDPXQy3BAKAwy8forbp2fDq88hm3uIpneVKSFcn93/zZl7XbsNIuF7f5i3r6Zy+Zp7F0dnCfaxGahEiHU5OKVX4Snc60uIpHfFxaDSiXHTxlfqSh5Ny0lwUQF9f+DVnIslhurMLkRAU7EIkBAW7EAlBwS5EQlCwC5EQFOxCJISltH8aAfANAEOotXu6x92/YmZ9AL4DYBS1FlAfcfdLsW1Nz8zioUceC9p6Nuyg87wSlpOeeewROmfTBl6/a6Cfy0lnTo9TW5nULWvr44kkxRRPkpk4zVsCvXvvrdS2+8Y3U9t8IR8cT2X5W33s5AlqO/zyK9T2/IFnqK2nO9zE8w/+8MN0zjvevJ3amiI9tjYMj1BbkUhvFinWFqsbWCK19QAglYnUtevhiTytJHmlmuYSMRMiIyUUl3RnLwP4U3e/HsAtAD5lZtcDuBvAw+6+DcDD9b+FENcoiwa7u4+5+9P1xzMADgFYD+BOAPfVn3YfgA+tlJNCiOXzhr6zm9kogJsAPAFgyN3H6qZx1D7mCyGuUZYc7GbWAeB7AD7j7tMLbe7uQLh4t5ndZWb7zGxfscgT/4UQK8uSgt3MsqgF+jfd/fv14QkzG67bhwGcC81193vcfY+772lq4r8PFkKsLIsGu9Xap3wdwCF3/9IC0wMAPlF//AkAP7j67gkhrhZLyXp7B4CPA3jezJ6tj30WwBcAfNfMPgngBICPLLah3r5+/MuP/augrXlwG503PxOWw15+/jk6Z3gtl2NSkTpdrS08g6pYDbfw2b6L+947zDPi5gd4HbQ7PvB71NbW2Uptc0R6i3RqQpm0tQKAfDm8PQA4d+4itZ04djY43tbGj+/46UlqO37wZWpL5bmPR8eDHzix97176JxNo+uoLZYtl2qJpKlluSxnrNac8TlNFn7PYtLbosHu7r8AwDbx7sXmCyGuDfQLOiESgoJdiISgYBciISjYhUgICnYhEkJDC06aAc1N4evL4RcP0HnTl8PSm8eyk4o8Y2g20v7JItpFS3M416g0z9sxXT7PfZw4ybPe/v4fwoU5AeDSTGR/s5eD451dXPLq7g235AKA9kihxNOnw/IaAAwOhAtLtnRxKfLnP+Sv+eLL+6mtUuQtto6MhwuIno600Nq2k0up3V1t3NbLW2y1tvGst+728HmVbeHFI9vawu+LOz9/dWcXIiEo2IVICAp2IRKCgl2IhKBgFyIhKNiFSAgNld6q5RJmJsMy2k9/8EM679T46eB4qhTOQgOA/funqS2WGlQu86wmkEyjhx78KZ3SlOXS1e6bbqa2YlMntU0X5qnt6MlwltfkJO8PV8zzrLez48ep7dhxvs09N701OP4nn/p3dM6Tj/+K2sqXeUbcdIEXRcmFa6rg6D4ue/78qTFqa89wmS/bxKWydDM/DzqJ9LZh0yidc+cffDQ4Xizz+7fu7EIkBAW7EAlBwS5EQlCwC5EQFOxCJISGrsZns00YHhoO2raNbqbzHOHV4kyktVI6suKeSvNrnFd54kpTS3vYkOVJDuvWhRNCAOD2972P2jrbIgkXLbx23QsHwnX5Dh/hbZzWrh+ltnyk7VK6lft44PCLwfEXDh+mc9pGd1Lb2bP8Nff2cNtgU7guXFsHr+N3cZy3w5o8c4Tazl8IJ90AQL4SSdoiBQLHpnh4vv3d4TllXrZOd3YhkoKCXYiEoGAXIiEo2IVICAp2IRKCgl2IhLCo9GZmIwC+gVpLZgdwj7t/xcw+D+CPAJyvP/Wz7v6j2LbK5TIung+3DLrld95O5739Xe8Kjjc388SDTERei7V/qkZaIaUR3l+pyPWOXJEnrUyePkZtF/M84eLiBd526SiR2M6eCycgAUDHIG93hGYuK1oTl96K5XByykM/+wWds2nrDdQ20sclzJYUP43bSCJSIc9r0B2dPkhtHZ28ll/FeRLV+KVZahsYGA2Oz5f4ufjTnz0ZHJ+Z4fUVl6KzlwH8qbs/bWadAJ4ys4fqti+7+39bwjaEEKvMUnq9jQEYqz+eMbNDAPhlVghxTfKGvrOb2SiAmwA8UR/6tJntN7N7zYz/jEkIseosOdjNrAPA9wB8xt2nAXwVwFYAu1G783+RzLvLzPaZ2b6ZWf49SQixsiwp2M0si1qgf9Pdvw8A7j7h7hV3rwL4GoC9obnufo+773H3PZ0dvPqKEGJlWTTYrdYi5esADrn7lxaML8xo+TAA3tJFCLHqLGU1/h0APg7geTN7tj72WQAfM7PdqMlxxwH88WIbSqUM7aRtzeR0ns57Zv9TwfHBQb5MMDQ4QG2lEpe1Ll2aojbkwz5mqnx76zdzWWukl3/SOXOY10Gbm+U11waH1gbH2/p76Jx0C5eT5nP8fRke3kht42fDdQMvTIbbUwHA8LpIW65Iq6/ZAj/+yITPt1KVy6XNrSS7EUBzJJuyOHme2pAK15kDgCGSdVgs8BZm7HDwo7S01fhfAAi9wqimLoS4ttAv6IRICAp2IRKCgl2IhKBgFyIhKNiFSAgNLTiZMqA5G87kKeS55PXYYw8Hx73EZaGuNl5QsFTi2Un5HG8plSHXxk2jI3TOrluup7atG7ksN3UqLF0BwPilC9TW1BqWmrb2hyU5ADh/nmdk3bBjF7W9+YYd1Pbt//2N4HgG4QKQAFCa4+9nschtHquy2BJ+r2PtmEY3b6G2c6de4vtK8SzM1na+v507twfH8/P8fRkZHgyO/6yJS3y6swuREBTsQiQEBbsQCUHBLkRCULALkRAU7EIkhIZKb9VqFfM5UoAxUgTyfR+4I7y9Is+SSkfktWqFF/LzNJdP0pmwbNTSzgsvjk9xKW9mivc9u5jj/lsLLwL50rNHg+OTv+IZWVs2cwntbddto7ZiJCOutSksNXkk4zCWYZdK81OVtEoDAOSqpE9ghR/fTRu49JafnaS267t4ttyTTz1DbWdPhOW83Bw/v33+UnC8WOAZkbqzC5EQFOxCJAQFuxAJQcEuREJQsAuREBTsQiSExma9pQztHWH5qjtSKa9zTTgrqBCRGVoi17Em45lX3sqz5ZrbwvOqeZ6dNDMzTW3pNl7ocXArLxC5tY1nvb18LNzrDcYlxSwpAgoAZ8ZOUlv/AC/4yWzFHJeTCgVejHIukhFXiGSHlQphqTfTwuXSoXVrqO3E2AS1TZwkxx5Afpa/tlcOPhsc7+/nfnhvX3g8UphTd3YhEoKCXYiEoGAXIiEo2IVICAp2IRLCoqvxZtYC4FEAzfXn/627f87MNgP4NoB+AE8B+Li78341AKrVPOZnSPJHlV93stYRHJ+Y4CucL79wnNpaMnzFvambr4IPkHZT6wa66ZxMJMGnv7uf2iK5OsjnwkkQADA4GF7hX78uvHoLAGPj49R2+PAhahstbqY2ppTMzPD3bH6er3RPX+aqRmw1vlIMJyKlm3nSysEDvHVYrCXT4OAQta2/kdfyG1wTnjewhtcNbCH+P/zLR+icpdzZCwB+193fglp75veb2S0A/gLAl939OgCXAHxyCdsSQqwSiwa713j10pmt/3MAvwvgb+vj9wH40Ip4KIS4Kiy1P3u63sH1HICHALwCYMrdX00KPg1g/cq4KIS4Giwp2N294u67AWwAsBfAm5a6AzO7y8z2mdm+mRlSuEIIseK8odV4d58C8AiAWwH0mNmrC3wbAJwhc+5x9z3uvqezk/9EUQixsiwa7Ga2xsx66o9bAbwHwCHUgv4P60/7BIAfrJSTQojls5REmGEA95lZGrWLw3fd/UEzewHAt83sPwF4BsDXF91S1VElbXxSketOphRO4ugiraQA4KnHf0Zt4xM8kcSyPClk7963Bsdvu3UPnXP5Mpea9j/9BLXN5Xnix+GTp6jt6PHjwfHcPP8K5c6LuLV08WSM6ekZapshLarmprlsGCklh0yaW7sjnxjXbQ7Lg739w3TO4Douea276QZq64vUoGuK1TZktkjyEjwcL6lIC6pFg93d9wO4KTB+FLXv70KI3wD0CzohEoKCXYiEoGAXIiEo2IVICAp2IRKCxWpWXfWdmZ0HcKL+5wAAroE1DvnxWuTHa/lN82OTuwf10oYG+2t2bLbP3blALT/kh/y4qn7oY7wQCUHBLkRCWM1gv2cV970Q+fFa5Mdr+a3xY9W+swshGos+xguREFYl2M3s/Wb2kpkdMbO7V8OHuh/Hzex5M3vWzPY1cL/3mtk5MzuwYKzPzB4ys5fr//PeSivrx+fN7Ez9mDxrZh9sgB8jZvaImb1gZgfN7N/Wxxt6TCJ+NPSYmFmLmT1pZs/V/fgP9fHNZvZEPW6+YxbpYxbC3Rv6D0AatbJWWwA0AXgOwPWN9qPuy3EAA6uw33cCuBnAgQVj/wXA3fXHdwP4i1Xy4/MA/qzBx2MYwM31x50ADgO4vtHHJOJHQ48Jatm+HfXHWQBPALgFwHcBfLQ+/lcA/vUb2e5q3Nn3Ajji7ke9Vnr62wDuXAU/Vg13fxTAxdcN34la4U6gQQU8iR8Nx93H3P3p+uMZ1IqjrEeDj0nEj4biNa56kdfVCPb1ABZWX1jNYpUO4Mdm9pSZ3bVKPrzKkLuP1R+PA+BFyFeeT5vZ/vrH/BX/OrEQMxtFrX7CE1jFY/I6P4AGH5OVKPKa9AW629z9ZgAfAPApM3vnajsE1K7sqF2IVoOvAtiKWo+AMQBfbNSOzawDwPcAfMbdX9MVopHHJOBHw4+JL6PIK2M1gv0MgJEFf9NilSuNu5+p/38OwP1Y3co7E2Y2DAD1/8+thhPuPlE/0aoAvoYGHRMzy6IWYN909+/Xhxt+TEJ+rNYxqe/7DRd5ZaxGsP8awLb6ymITgI8CeKDRTphZu5l1vvoYwHsBHIjPWlEeQK1wJ7CKBTxfDa46H0YDjomZGWo1DA+5+5cWmBp6TJgfjT4mK1bktVErjK9bbfwgaiudrwD481XyYQtqSsBzAA420g8A30Lt42AJte9en0StZ97DAF4G8BMAfavkx/8C8DyA/agF23AD/LgNtY/o+wE8W//3wUYfk4gfDT0mAG5ErYjrftQuLP9+wTn7JIAjAP4PgOY3sl39gk6IhJD0BTohEoOCXYiEoGAXIiEo2IVICAp2IRKCgj1BmFmPmf2bq7St283swauxLdEYFOzJogfAPwn2BfIymXwAAAFISURBVL/KEr/FKNiTxRcAbK3nZP/azH5uZg8AeMHMRl+X1/5nZvb5+uPrzOwn9fzqp81s68KNmtnbzOyZ14+Lawtd0ZPF3QB2uftuM7sdwA/rfx+rZ3kxvgngC+5+v5m1oHaTGAEAM3s7gP8O4E53P7mSzovloWBPNk+6+7HYE+r5A+vd/X4AcPd8fRwAdqJWCPG97n52hX0Vy0Qf45PN3ILHZbz2fGhZwvwxAHnU8r7FNY6CPVnMoFZuKcQEgEEz6zezZgB3AP+/YstpM/sQAJhZs5m11edMAfh9AP+5/rVAXMMo2BOEu08C+GV9Ie6/vs5WAvAfUcuqegjAiwvMHwfwJ2a2H8BjANYumDeB2oXhL83sd1b2FYjloKw3IRKC7uxCJAQFuxAJQcEuREJQsAuREBTsQiQEBbsQCUHBLkRCULALkRD+H9Ey3yQKgA+yAAAAAElFTkSuQmCC\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAEHCAYAAABoVTBwAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAf60lEQVR4nO2dW4xk13We/1Wn7l19mZ7u6em58W5ZhBGNlAEtW4qh2LHByE4oAbYgIRD4QHjswEoiwH4gZCBigDzIQSRBL1EwigjThqKLLQkiAtmWTAihlQdaQ5kiKVIiR+RQw+HM9Fz6XveqlYeqAZrE/nc3p7urR9r/BzS6eu/e56za56xzqvZ/1lrm7hBC/PyT22sDhBCjQc4uRCLI2YVIBDm7EIkgZxciEeTsQiRCfjuDzexeAJ8BkAH4X+7+idj/53Lm+Xz4+pIzi+0o3By3LtJ3Y3Jjt9cLtueMXzNjV9N+TPbMcftjc5XLhfeYZfxQ93pd2tfv39hcORsXO8yR7VnkPWcZ7yvkw++70+nQMb3IcYnNY+xw9vvhcwcAioXwMYu9Z9a3Xm+j1e4GO+1GdXYzywC8AOA3AbwK4HsAPuTuz7ExxWLmczPlYF+lUontK9iez2V0DDvpAaAbmXh2YQGApeWVYHs5V6RjxnL85FhtNWhfrlqifZVSZH9jY8H2yckpOmZx8Rrta6+3aF/szOm0iTNFPDrL8+PJHAIAJsfC5xQAzM/uC7afv3SJjllv8/NjYiK8PQDodviMrK8v074jhyeC7YUCP3fy5CL29//wAq4t1YOzvJ2P8fcAOOPuL7l7G8CXANy3je0JIXaR7Tj7YQDnNvz96rBNCHETsq3v7FvBzE4COAnEv1sJIXaX7dzZzwM4uuHvI8O21+Hup9z9hLufyEUWnYQQu8t2nP17AO4ys9vMrAjggwAe3RmzhBA7zQ1/jHf3rpl9BMDfYSC9PezuP4yNMQCFLLzi2utyKaTf64e3V+Sr0q0ul5Niq76x1fip8WqwfYKsgANAe3Wd9vUbbdpXLXB1YrLK+6qV8Mp0rVigY640+Ip733lfucwVg9nZmWD74uIi3x6xHQAOzR+gfVlEFzhwYDrYXojs6+Vzr9G+YiFyfkzx86DGu7B/cjLYbhHpYr1OzquIRLKt7+zu/k0A39zONoQQo0FP0AmRCHJ2IRJBzi5EIsjZhUgEObsQibDrT9BtxMxQJFFvFokc2zezP9i+3qjTMYUel9e6EVnOIoFB8wfD8s/B2bB9APDymZ/Qvpl8WHIBgIOHDtK+XDcSZUekw4mI1LR/cpz2eRaRAIlkBADVsbBMmeX43M/OheU6AChHpMPVFR5k0vWwpDs5xW0/3I1EvUU8Jl/g40oZlyn7JPBmYjwcIAMA3gnL0dGISNojhPi5Qs4uRCLI2YVIBDm7EIkgZxciEUa6Gp9lOUxOhFd+Y0EQBw6EV8EXrl6lY8olvvq5vLhE++ZmZmlfqRRe4a9U+Erx4aN8VZ2lkAKATpuvWhfBA4BKxfD7rjd4Cqyjh3iQiRfCq74AUIykx2q3w0E+M/v5Kng+x/fVavGAovGJ8Mo/ADRI6q/VZR6Q02rxtFT7Z7hyURmLpJEyvs18OzyPzXV+zLqtsMoQSzOnO7sQiSBnFyIR5OxCJIKcXYhEkLMLkQhydiESYaTSWz6fxwwJaun3uezSbjaD7XMkMAUAqmUewFEiefAAYH6WS2+dTjjw5uqVBTpmnEiNAJCPVDnpt/l8FPKx8k9h6aVRD1ezARCt0pIr87lqtbk01GqHc9eVIpLo2soq7RurcXmtR8pyAcDVa2GJrVTgsmesElmbvC8AWF1bo325yCS3V8L2t1lVHQA1ItvSslvQnV2IZJCzC5EIcnYhEkHOLkQiyNmFSAQ5uxCJsC3pzczOAlgF0APQdfcT0f8HkENYUmq3wvIaAPSI3NGNRUk1eX66fMavcStL12ifISyReET6OX/hAu2brHFZrprnEWUrLZ5zjUU9Fcv8UHcipbc6EanJchHpsBuek37G56oUyTMXK2tUj5SvKpbCkl2xwCXAapnLZKVIpN/yEo+mXF7ix6xWJuWfIhJxdSI8JhcZsxM6+7909ys7sB0hxC6ij/FCJMJ2nd0BfMvMnjSzkzthkBBid9jux/h3u/t5MzsA4Ntm9iN3f3zjPwwvAicBoFKKfCcTQuwq27qzu/v54e8FAF8HcE/gf065+wl3P1EsjvRRfCHEBm7Y2c1szMzGr78G8FsAnt0pw4QQO8t2brVzAL5ugxChPID/7e5/Gx/iMKKhxO76TE7q9rhk1GryiKx9FR7xVMhx2SWfC38Naba53FEs8USa7VY4KSMAtFd4gsVijUf0FYthacgK3MZel0tXlUj0YCcSlTU+MRVsL5f5fFgkKWMsoqxDyicBgBGJLWYHOpHzqs7nqtfm985ivkb7JqaniRk86ejKelha7kWiR2/Y2d39JQBvu9HxQojRIulNiESQswuRCHJ2IRJBzi5EIsjZhUiEET/lYsiRSKlYorzKWFj+aVqkDlmkjlpvncsnMD4lB+fmgu3dq5GQrC6X18ZIXTYAaK1yqWnyYFiqAYB6nUf7MWbmeJLN1hq3PzP+RGSBSV4lLuU1G/w9l4p8XK7IZa1lcqw7HS7XZT0ueTWbXJZDn8ublYjUlydyabPD5/7ylcvB9k6X2647uxCJIGcXIhHk7EIkgpxdiESQswuRCCNdje90ezh/OZyLiwW7AMBYK7zqXpvkK+7NSHBELeMro4fn99G+UjUcJJOFKwwBAPZVec6yqSq3Y/zgDO1rkRJPAPDCxdfC+5qa4Ntb52+gWeeru4XIPHZWwuOaLa6E9I2vZmeRQJ61NV42qkviodo9PoezU7zU1PQEPz9eXH2J9u3fx8extz1BVCgA6HfC+Qvz2VU6Rnd2IRJBzi5EIsjZhUgEObsQiSBnFyIR5OxCJMJIpTd3R6sbltGuXeNll6r1cGmo6UigQCHy1sq1iGRXX6F9a0yG4mnrkEUCE1qrXIaaHefBHT9+8WXaVyuHZaNahcs4rVYkX988D7qxHg+E6ZJcbZEqVFhtRkpDRXL5XbwUlhsBAP3w+65NhnPkAUCzwYOJupH8dJUylwfHx7gEe40EPTUjJdHGa+HzI1b+SXd2IRJBzi5EIsjZhUgEObsQiSBnFyIR5OxCJMKm0puZPQzgdwAsuPsvDdumAXwZwK0AzgL4gLtHYr+GO8tnODAdjtbpNnn+sfFaOJ+ZR/K7ZXl+HatUuAwSCb5DvRHeX7vL91WKaE1vfcudtO/ixUu0r9XiRs7MhvPJxUpl9cEltGpEpmzXeQ7ArEIiBHNcXlu/Fo6IBIDlOu+bnOARfWv18Fz1+nw+SgU+H7Ecb4ePHaV9/Yg+u7gSPvf7kVJOU9Ph48xyPAJbu7P/OYB739D2IIDH3P0uAI8N/xZC3MRs6uzDeutvfOLlPgCPDF8/AuB9O2yXEGKHudHv7HPufmH4+iIGFV2FEDcx216g80GKGfol0sxOmtlpMzsdy9UthNhdbtTZL5nZPAAMfy+wf3T3U+5+wt1PFCKphYQQu8uNOvujAO4fvr4fwDd2xhwhxG6xFentiwDeA2DGzF4F8HEAnwDwFTN7AMArAD6wlZ3lzFArhe/ub73jGB1XqYYjuXIZN//iuQu0r9vl0WZjtQO0b2ktHIWUGZfyLCK5rC7zRImXF67QvkjgFUBktLU1Lm32nW+wXl+nfWsrPCprohqWWNvg+3LjslYWkZQmxsP7AoBKNXyO5PORCLVxHmGX5fi4mFT28k/P0T7Lh8+fYiSCbZVEgvYiZdQ2dXZ3/xDp+o3Nxgohbh70BJ0QiSBnFyIR5OxCJIKcXYhEkLMLkQgjTTiZGVArhuWEsSqPrioUw3LS5BRPhkiCrgAAi1d5PawfPv8C7ev2w9fGUpEnh5we4zW+Xjt/nvZdvcKlt2aXS0MrTM4zfl13rhhhaYkHM0byfaLdCndWq1xOmt4/SfssYn+ry5/MdCJFNZo8yaaDS7PdWALRSB27Xp/bWImc+4x8ISzXmfETX3d2IRJBzi5EIsjZhUgEObsQiSBnFyIR5OxCJMJIpbdioYAjB8NRZTFpYt9UWL7KjMs4hRkueR2c3U/7HvvO/6V9/X54f1PjXO64eIFHhs3t4xLa1CSX85YWuGx0ZeFieHv7eFLGsUgdssnIuPExLn2OT4ZltLFapD5cg7+vl868QvsyEjUGAHUiAbbbXDdst/i5mGX8/mjgGmalHE6aCgA9C89JJxLe2CF14DwSeac7uxCJIGcXIhHk7EIkgpxdiESQswuRCCNdjXc4nERdlEiwC8BXQDvrPD9aKeMr5F7gfT0S7AIAuVzYxugVM1Jm6JZbbqN9rIwTABy5wPPJlUphGycmebBFFpmrhQUerPOrv3wP7Tt46FCwvetcnVi5epn2LV7hATlXl/h5kM/CgTCzMzzoph/J49bv8ZX6yRpXUBYj+QY9F57/doPPVa8TDshh/gXozi5EMsjZhUgEObsQiSBnFyIR5OxCJIKcXYhE2Er5p4cB/A6ABXf/pWHbQwB+H8B1reRj7v7NzbbVbnfw03OvBvtqY1waWl0NSytTJR4AESsz1Mtzma8aKSXUboTljgOzPOimlOPBHXfcfpiPi7y3XKFC+4pEeqtU+HvOEekHALzBJaPWCpcAO5Ph971/nkteuS6fq1uOHqF9pfIK7VtZXwq2F4v81M8b7+tGglOySEmpHgnIAYCsHD73PVKmrEaCkEoFHjC0lTv7nwO4N9D+aXc/PvzZ1NGFEHvLps7u7o8DuDYCW4QQu8h2vrN/xMyeNrOHzYx/jhVC3BTcqLN/FsAdAI4DuADgk+wfzeykmZ02s9Mt8oifEGL3uSFnd/dL7t7zwYO4nwNAH5J291PufsLdT5QKI30UXwixgRtydjOb3/Dn+wE8uzPmCCF2i61Ib18E8B4AM2b2KoCPA3iPmR0H4ADOAviDreys3++j3gjLCX1w+adNyvtMz/IcaP0+/8rQbHL55OjRo7TvuWd/HGwv5Lnt8wd59NpsRLLLjEcvFbiKhmIpfEirVZ7vLhb1hsZB3rXCJa9rlxeC7Z7jkVyVMrcjZv/EOI9SW6mH15a9x8+BSplLmxbJd9eJ1MOaqFRpX4+cPxNVvq8CUfki1Z82d3Z3/1Cg+fObjRNC3FzoCTohEkHOLkQiyNmFSAQ5uxCJIGcXIhFG+pSLmSGXhXWjVpPLFiUid7TaPCqoVI4kjuxwWavX5pFXq4vhCKr6Gpegbjt2B+2rlLhOUqvy6LvJfVwa6nTDklKvF4m6ipQ0mpnhdixEylBduByWvJ589mk65s47j/F9XeZz/NoFnqiyi/A5MjXB31chUsapVOISYDcS9dZqcsmxT06D6vQUHbOyFo44jChvurMLkQpydiESQc4uRCLI2YVIBDm7EIkgZxciEUYqvRXyBRycCUdRlQr8ulMlyRcrVS40dCNSUyFSy2uizKPl7jg8F2yfqnIp7NABLp/USlyqmRjjEk8zF0k42Q/P1coyf1/lMb69QpWH2F28zBNOnrtWD7b/+Mwlvr2FSB245Uhyyw7vu/ut88H2Wpm/r16dS7ro82Pmzs+rcqSWYY9EdVoWSXzZI7XewG3QnV2IRJCzC5EIcnYhEkHOLkQiyNmFSISRrsa7AZ4LX1/KkRxdhXx4TKHEr1XNVb6i2umEVz8BYHJ8gvYdPz4TbK8U+ApoocDziOUj+cx6fR6MgUgetxIpa1Sr8dXgYiQgx/v8FCmQYwkAz/0onK9vvc5zv6EXLvMFAK0WH1ckwVUAkMuVgu0eSdbWz/HzY6URCZSq8+OSzyKlytrhlfVui2+v3Qqf3x45b3RnFyIR5OxCJIKcXYhEkLMLkQhydiESQc4uRCJspfzTUQB/AWAOg3JPp9z9M2Y2DeDLAG7FoATUB9x9MbYt7wNtUsl1dT0cOAEAufGwLNdYWqVjWC42AKhWeP6xLMclkqWry8H2VkR6W17jUk2nx8s/eYsHrsTKTRVy4UCNei8S3MGVJrRJuS4AqJJSUwBw8eKFYHvLeYBPK4vIaxGZMivz4JR6Pfzmuu1IzsMi39dykx/Pi1f56e/gNsLDx9OMH5gKm/uIpLiVO3sXwB+7+90A3gngj8zsbgAPAnjM3e8C8NjwbyHETcqmzu7uF9z9+8PXqwCeB3AYwH0AHhn+2yMA3rdbRgohts+b+s5uZrcCeDuAJwDMufv1z2oXMfiYL4S4Sdmys5tZDcBXAXzU3V+XxNsHUfvBL65mdtLMTpvZ6WY78qikEGJX2ZKzm1kBA0f/grt/bdh8yczmh/3zAIIFud39lLufcPcTsWwdQojdZVNnNzPDoB778+7+qQ1djwK4f/j6fgDf2HnzhBA7xVai3t4F4MMAnjGzp4ZtHwPwCQBfMbMHALwC4AObbajb6+IKKaF06MB+Oo7Jct0+jwqa3j/Nt7fCZb5ul/e1iFwTSWmHH515mfbljEcoFSMlmY7deohvsxaO8mqucxmnF5GhupFyWKWIjUuLYZnyhfOv0DG3zYbzxQHA9Pgk7ctP80jF9fXwV8fFbtg+AMiTyEEAWG3wc24x0td3PldG3LBgXH5dJ3nyuiSfHbAFZ3f374KXkPqNzcYLIW4O9ASdEIkgZxciEeTsQiSCnF2IRJCzC5EII0042e50cO6114J9hQKPCmLyz9Gj4VJSAJcmAGBlLSa9cR0tYxFlXS5dPX/mJdqXJ9sDgNfOhaPGAGBmmkfLTU6Gy029+OIZOiZWMujf/vav0L6Sc8lr31Q4srCywp+ivLoUlmUBoN/mMmXs3FlZC0dMrrd4cst6RG7MFcPSJgA0O9zGWCmnPkkSubjG5cGZcV6yi6E7uxCJIGcXIhHk7EIkgpxdiESQswuRCHJ2IRJhtLXeAHQ9LPNcXeYyw0Q1nKQwJqFl+YjUEUn+t96IJL4kl0bvc6lmvML3tXCN7+upZ3h02FjlMu1rNZm0FYmwiyRsfP5FbsdcNVz7DgDGx8K5Cw4e5GOuvnKR9lkkyebCZT4fR46Eoyl7fb69VkR+ra/zJKfdyDZ7sXNkohZsb0fCKdeJFNmLRGDqzi5EIsjZhUgEObsQiSBnFyIR5OxCJMJIV+PzWR779odXYycmxui4ciFs5rUVvjJaqYQDIACg0+Z5utqxHF6F8LWxWOLlgto9HvixcI3b3+zy6/D0eDjYBQCO3B6e3w4puwUAK6s8AOXsq3yluzjLswXnPLy/WpXPlR3gAT4TFR50s7a0QvvOvnI22H7HLxyjY9qkHBMAtHs8z1xE8Iiu4h8jOfQqZT5XrQYLvtpe+SchxM8BcnYhEkHOLkQiyNmFSAQ5uxCJIGcXIhE2ld7M7CiAv8CgJLMDOOXunzGzhwD8PoDr2szH3P2bsW31+n2s1sPBH/0+l6gOzR0Ithcj8lq9xfPCjVW5jGN5Lr1ZFo4yKBQjucciElq9wfdVrISDfwCgtj8cOAEAnVxY8urmufRWnuLz2M9zeW01Eoh01+23hO24uEbHdNd5sMjy2jW+rzvvon2vnnsx2N6JSKysHBMArEVKh/Uj985alc8xkyPXSdkzAMiq4Rx/iOQ13IrO3gXwx+7+fTMbB/CkmX172Pdpd//vW9iGEGKP2UqttwsALgxfr5rZ8wAO77ZhQoid5U19ZzezWwG8HcATw6aPmNnTZvawmfHHn4QQe86Wnd3MagC+CuCj7r4C4LMA7gBwHIM7/yfJuJNmdtrMTnd7kecJhRC7ypac3cwKGDj6F9z9awDg7pfcvefufQCfA3BPaKy7n3L3E+5+Ih+p5y2E2F029T4zMwCfB/C8u39qQ/v8hn97P4Bnd948IcROsZXV+HcB+DCAZ8zsqWHbxwB8yMyOYyDHnQXwB5ttKJflUB0LSxC9SAmlVicsy+UjZX8KBR4xlGV8XOz6lyMqVL5wY19PWhG50fLcxuokf2+rq+HoqkqFlwu6fJnLWvk8kXgA7KvwuapOheXNWpnLa3Ozk7Tvii/yfVW5PHjgQDgH3eoKj5SLBEUix4PKMEFKbwHA+ASf/5XlcNThlStX6BjPheXXbpdLrFtZjf8uwnFzUU1dCHFzoS/RQiSCnF2IRJCzC5EIcnYhEkHOLkQijDThZM4M5UpYNsoZl5Ma7VawvdTn8lQlkgTSwOWJYkTOQxbWXSYmp+mQ5gova9XOc7kxX+JyXqPNkx5mWfh9d8JTOLCjwWsGXWhy+Wf6MA+R6FxYCLZXjO+rPM7nfnYyHPkIAFeu/pT2TU+SCEemowJY6/LJesv8IdrXd25/vc5l1vp6uG86IuWx/KFZRBvUnV2IRJCzC5EIcnYhEkHOLkQiyNmFSAQ5uxCJMFLpzcxQJDHt1UhCvl4vHIaUgYcnZUQmG2yPyyDdSPSdE9tXV7nk0ohEV8XsL5f5oWlH6rZ1GuG++jKXk4p5HpE1Ps3lHxRL3I56OLotK3LpLVYzz0m9PyAeUVYi0YNT07N8Xys8CtBy/Jg1V9dpX6MeOdbk3B9ElxM8PI9ZJGeE7uxCJIKcXYhEkLMLkQhydiESQc4uRCLI2YVIhJFHvY0RuSYfTHM3HEfay2VeD21tjdcUiyWcLJa4nFQhyTKjYyKX0wZJNAgAcweO0b5mRLKbGgvPSWE2ImtF8mV2wCW7bo9LgJXaWNgOUtcMQDjT4XU7IjLUzCyvfVfsh0/xLFLDrlTi55U7n49qldtRib1vcj42Gjw5J+tzIskBurMLkQxydiESQc4uRCLI2YVIBDm7EImw6Wq8mZUBPA6gNPz/v3b3j5vZbQC+BGA/gCcBfNjdeRQJBoutBbJamIus7BazsJkWW8HP8etYv8+Xn4sFvkrLSuv0+9z2csSOyXG+ehsrM1Qu8qChPqldVK3xMZ0WP2zNRp32tbpcFagWw8esEAmeWa/zfZXHSS45AI02n/8GeW8F58c5y3G1Jpfxlfpe5NZZb/BzbmkpXNoqVsqpWGSr+9vLQdcC8Ovu/jYMyjPfa2bvBPBnAD7t7ncCWATwwBa2JYTYIzZ1dh9wXbQuDH8cwK8D+Oth+yMA3rcrFgohdoSt1mfPhhVcFwB8G8BPACy5+/XPGa8C4HmFhRB7zpac3d177n4cwBEA9wD4xa3uwMxOmtlpMzvdiny3EkLsLm9qNd7dlwB8B8CvAJgys+urMEcAnCdjTrn7CXc/USKLNkKI3WdTZzezWTObGr6uAPhNAM9j4PS/O/y3+wF8Y7eMFEJsn63caucBPGJmGQYXh6+4+/8xs+cAfMnM/iuAfwLw+c02lDNDpRiWPFieOQDwPslBl3H5ZGKCSzUx6S2W94tJJB6R3iYrPD9aLfJJxyOlrRotPlfWD0ub/Q4v4zQ+xiXASFxFJBwHWCcluwodfswajUjQTY4HhVxZXqV9a1fDOQCnpmbomKvr4eMMAOVIZJM7P56L17isuEokx0rk3GF9sXN7U2d396cBvD3Q/hIG39+FED8D6Ak6IRJBzi5EIsjZhUgEObsQiSBnFyIRLJazasd3ZnYZwCvDP2cAcD1odMiO1yM7Xs/Pmh23uHuwttVInf11OzY77e4n9mTnskN2JGiHPsYLkQhydiESYS+d/dQe7nsjsuP1yI7X83Njx559ZxdCjBZ9jBciEfbE2c3sXjP7sZmdMbMH98KGoR1nzewZM3vKzE6PcL8Pm9mCmT27oW3azL5tZi8Of+/bIzseMrPzwzl5yszeOwI7jprZd8zsOTP7oZn9p2H7SOckYsdI58TMymb2j2b2g6Ed/2XYfpuZPTH0my+bWaSmVAB3H+kPgAyDtFa3AygC+AGAu0dtx9CWswBm9mC/vwbgHQCe3dD23wA8OHz9IIA/2yM7HgLwJyOej3kA7xi+HgfwAoC7Rz0nETtGOicYpIitDV8XADwB4J0AvgLgg8P2/wng37+Z7e7Fnf0eAGfc/SUfpJ7+EoD79sCOPcPdHwdw7Q3N92GQuBMYUQJPYsfIcfcL7v794etVDJKjHMaI5yRix0jxATue5HUvnP0wgHMb/t7LZJUO4Ftm9qSZndwjG64z5+4Xhq8vApjbQ1s+YmZPDz/m7/rXiY2Y2a0Y5E94Ans4J2+wAxjxnOxGktfUF+je7e7vAPCvAfyRmf3aXhsEDK7sGFyI9oLPArgDgxoBFwB8clQ7NrMagK8C+Ki7vy7FzCjnJGDHyOfEt5HklbEXzn4ewNENf9NklbuNu58f/l4A8HXsbeadS2Y2DwDD3wt7YYS7XxqeaH0An8OI5sTMChg42Bfc/WvD5pHPSciOvZqT4b7fdJJXxl44+/cA3DVcWSwC+CCAR0dthJmNmdn49dcAfgvAs/FRu8qjGCTuBPYwged15xryfoxgTmyQ+O/zAJ53909t6BrpnDA7Rj0nu5bkdVQrjG9YbXwvBiudPwHwp3tkw+0YKAE/APDDUdoB4IsYfBzsYPDd6wEMauY9BuBFAH8PYHqP7PhLAM8AeBoDZ5sfgR3vxuAj+tMAnhr+vHfUcxKxY6RzAuCfYZDE9WkMLiz/ecM5+48AzgD4KwClN7NdPUEnRCKkvkAnRDLI2YVIBDm7EIkgZxciEeTsQiSCnF1cj+r6k722Q+wucnaxI2x4skvcpMjZE8XM/tTMXjCz7wJ4y7DtDjP722Fg0D+Y2S8O22fN7Ktm9r3hz7uG7Q+Z2V+a2f/D4METcROjq3GCmNk/x+Ax5eMYnAPfB/AkBnnO/tDdXzSzXwbwPzAIq/wMgE+7+3fN7BiAvwPw1uHm7sYgoIjXVBY3BXL2NPkXAL7u7nUAMLNHAZQB/CqAv9pQo740/P2vANy9oX1iGBkGAI/K0X82kLOL6+QwiJc+Tvre6e7NjY1D518fgW1iB9B39jR5HMD7zKwyjPz7NwDqAF42s98DBhFgZva24f9/C8B/uD7YzEIXBHGTI2dPEB+kXvoyBhF/f4NB2DEA/DsAD5jZ9UjA6+nC/iOAE8NMLc8B+MMRmyx2AEW9CZEIurMLkQhydiESQc4uRCLI2YVIBDm7EIkgZxciEeTsQiSCnF2IRPj/zKA2uVH21IkAAAAASUVORK5CYII=\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Creating CNN"
      ],
      "metadata": {
        "id": "HIov6CuJ4x0F"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model = models.Sequential()\n",
        "model.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=(32, 32, 3)))\n",
        "model.add(layers.MaxPooling2D((2, 2)))\n",
        "model.add(layers.Conv2D(64, (3, 3), activation='relu'))\n",
        "model.add(layers.MaxPooling2D(2, 2))\n",
        "model.add(layers.Conv2D(64, (3,3), activation='relu'))"
      ],
      "metadata": {
        "id": "3lbr7WbB18sI"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Display the architecture of model"
      ],
      "metadata": {
        "id": "eXVcNlTb7QTo"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "K8QFoZAp18pY",
        "outputId": "25695232-22da-4e11-bbbe-058589f56574"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d (Conv2D)             (None, 30, 30, 32)        896       \n",
            "                                                                 \n",
            " max_pooling2d (MaxPooling2D  (None, 15, 15, 32)       0         \n",
            " )                                                               \n",
            "                                                                 \n",
            " conv2d_1 (Conv2D)           (None, 13, 13, 64)        18496     \n",
            "                                                                 \n",
            " max_pooling2d_1 (MaxPooling  (None, 6, 6, 64)         0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " conv2d_2 (Conv2D)           (None, 4, 4, 64)          36928     \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 56,320\n",
            "Trainable params: 56,320\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Adding Dense layers"
      ],
      "metadata": {
        "id": "LQwYZ3Wq7bt9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model.add(layers.Flatten())\n",
        "model.add(layers.Dense(64, activation='relu'))\n",
        "model.add(layers.Dense(10))"
      ],
      "metadata": {
        "id": "UNe4z0uj7V_z"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kFrDmZ677V8P",
        "outputId": "daf94040-ef3b-4e2e-9f35-f1b2e3f39436"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d (Conv2D)             (None, 30, 30, 32)        896       \n",
            "                                                                 \n",
            " max_pooling2d (MaxPooling2D  (None, 15, 15, 32)       0         \n",
            " )                                                               \n",
            "                                                                 \n",
            " conv2d_1 (Conv2D)           (None, 13, 13, 64)        18496     \n",
            "                                                                 \n",
            " max_pooling2d_1 (MaxPooling  (None, 6, 6, 64)         0         \n",
            " 2D)                                                             \n",
            "                                                                 \n",
            " conv2d_2 (Conv2D)           (None, 4, 4, 64)          36928     \n",
            "                                                                 \n",
            " flatten (Flatten)           (None, 1024)              0         \n",
            "                                                                 \n",
            " flatten_1 (Flatten)         (None, 1024)              0         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 64)                65600     \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 10)                650       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 122,570\n",
            "Trainable params: 122,570\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Compile and train the model"
      ],
      "metadata": {
        "id": "KnO2wnva76c7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model.compile(optimizer='adam', loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True), metrics=['accuracy'])\n",
        "history = model.fit(train_images, train_labels, epochs=3, validation_data=(test_images, test_labels))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "atC6sb-K7V6C",
        "outputId": "f8a7946a-bed4-4e5f-a82c-b49d39d2c080"
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/3\n",
            "1563/1563 [==============================] - 90s 57ms/step - loss: 0.7816 - accuracy: 0.7263 - val_loss: 0.8894 - val_accuracy: 0.6921\n",
            "Epoch 2/3\n",
            "1563/1563 [==============================] - 81s 52ms/step - loss: 0.7326 - accuracy: 0.7407 - val_loss: 0.8487 - val_accuracy: 0.7064\n",
            "Epoch 3/3\n",
            "1563/1563 [==============================] - 85s 54ms/step - loss: 0.6760 - accuracy: 0.7617 - val_loss: 0.8385 - val_accuracy: 0.7118\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Evaluate the model"
      ],
      "metadata": {
        "id": "OcqlBzff9wi9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(history.history['accuracy'], label='accuracy')\n",
        "plt.plot(history.history['val_accuracy'], label='val_accuracy')\n",
        "plt.xlabel('Epochs')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.ylim([0.5, 1])\n",
        "plt.legend(loc='lower right')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "id": "KbVsZqt87V3X",
        "outputId": "74ab5640-341f-46b7-a43a-f677acd0a90d"
      },
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_loss, test_acc = model.evaluate(test_images,  test_labels, verbose=2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yTPJpVb37V08",
        "outputId": "d8c70923-b5b2-444c-dec1-2ca3edc47853"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "313/313 - 5s - loss: 0.8385 - accuracy: 0.7118 - 5s/epoch - 15ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(test_acc)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IgEAZ7A418kJ",
        "outputId": "931b6224-cdd9-4247-d8c0-9d2edab65889"
      },
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.7117999792098999\n"
          ]
        }
      ]
    }
  ]
}