Artificial Intelligence Major Project - CNN for Sign Language Digit Recognition.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "cb13f12e",
   "metadata": {},
   "source": [
    "## The sign-language-digits dataset is taken from the following link:\n",
    "## https://www.kaggle.com/datasets/ardamavi/sign-language-digits-dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 194,
   "id": "d2a621f0",
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow\n",
    "from tensorflow import keras\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from keras.layers import Conv2D, Dense, Dropout, Flatten, MaxPooling2D\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e63a7224",
   "metadata": {},
   "outputs": [],
   "source": [
    "? np.load"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "dc2d9a5f",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = np.load('E:\\Datasets\\X.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "44c3024e",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[0.46666667, 0.4745098 , 0.47843137, ..., 0.5176471 ,\n",
       "         0.5137255 , 0.5019608 ],\n",
       "        [0.49803922, 0.4862745 , 0.4862745 , ..., 0.5254902 ,\n",
       "         0.5176471 , 0.50980395],\n",
       "        [0.54509807, 0.49411765, 0.49019608, ..., 0.5294118 ,\n",
       "         0.5254902 , 0.5137255 ],\n",
       "        ...,\n",
       "        [0.5019608 , 0.5137255 , 0.5176471 , ..., 0.5529412 ,\n",
       "         0.54509807, 0.53333336],\n",
       "        [0.49803922, 0.5058824 , 0.5137255 , ..., 0.54509807,\n",
       "         0.53333336, 0.52156866],\n",
       "        [0.49019608, 0.49803922, 0.5019608 , ..., 0.5294118 ,\n",
       "         0.52156866, 0.50980395]],\n",
       "\n",
       "       [[0.59607846, 0.60784316, 0.61960787, ..., 0.4627451 ,\n",
       "         0.4509804 , 0.44313726],\n",
       "        [0.60784316, 0.61960787, 0.6313726 , ..., 0.4745098 ,\n",
       "         0.4627451 , 0.4509804 ],\n",
       "        [0.6156863 , 0.627451  , 0.6431373 , ..., 0.4862745 ,\n",
       "         0.4745098 , 0.4627451 ],\n",
       "        ...,\n",
       "        [0.50980395, 0.5176471 , 0.5254902 , ..., 0.40392157,\n",
       "         0.39607844, 0.3882353 ],\n",
       "        [0.5019608 , 0.50980395, 0.5176471 , ..., 0.4       ,\n",
       "         0.39215687, 0.38431373],\n",
       "        [0.49411765, 0.5058824 , 0.50980395, ..., 0.3882353 ,\n",
       "         0.38039216, 0.37254903]],\n",
       "\n",
       "       [[0.5882353 , 0.6039216 , 0.61960787, ..., 0.53333336,\n",
       "         0.5137255 , 0.5019608 ],\n",
       "        [0.5921569 , 0.6117647 , 0.6156863 , ..., 0.5411765 ,\n",
       "         0.52156866, 0.50980395],\n",
       "        [0.5882353 , 0.61960787, 0.627451  , ..., 0.54901963,\n",
       "         0.5294118 , 0.5137255 ],\n",
       "        ...,\n",
       "        [0.4745098 , 0.4862745 , 0.49411765, ..., 0.42745098,\n",
       "         0.41960785, 0.40784314],\n",
       "        [0.47058824, 0.47843137, 0.4862745 , ..., 0.41568628,\n",
       "         0.4117647 , 0.4       ],\n",
       "        [0.45882353, 0.47058824, 0.47843137, ..., 0.40392157,\n",
       "         0.4       , 0.3882353 ]],\n",
       "\n",
       "       ...,\n",
       "\n",
       "       [[0.5254902 , 0.54509807, 0.5568628 , ..., 0.49803922,\n",
       "         0.49019608, 0.4862745 ],\n",
       "        [0.5372549 , 0.5529412 , 0.5686275 , ..., 0.5019608 ,\n",
       "         0.49803922, 0.49019608],\n",
       "        [0.54509807, 0.5568628 , 0.57254905, ..., 0.5058824 ,\n",
       "         0.49803922, 0.49019608],\n",
       "        ...,\n",
       "        [0.5058824 , 0.50980395, 0.5137255 , ..., 0.4862745 ,\n",
       "         0.47843137, 0.47058824],\n",
       "        [0.49411765, 0.5058824 , 0.50980395, ..., 0.47843137,\n",
       "         0.47058824, 0.4627451 ],\n",
       "        [0.48235294, 0.49411765, 0.5058824 , ..., 0.47058824,\n",
       "         0.4627451 , 0.45490196]],\n",
       "\n",
       "       [[0.5137255 , 0.5254902 , 0.5372549 , ..., 0.45490196,\n",
       "         0.44705883, 0.43529412],\n",
       "        [0.5176471 , 0.53333336, 0.54901963, ..., 0.46666667,\n",
       "         0.45882353, 0.44705883],\n",
       "        [0.5254902 , 0.5411765 , 0.56078434, ..., 0.4745098 ,\n",
       "         0.4627451 , 0.4509804 ],\n",
       "        ...,\n",
       "        [0.53333336, 0.5411765 , 0.5529412 , ..., 0.54509807,\n",
       "         0.5372549 , 0.5294118 ],\n",
       "        [0.52156866, 0.53333336, 0.54509807, ..., 0.5372549 ,\n",
       "         0.5294118 , 0.52156866],\n",
       "        [0.5137255 , 0.5254902 , 0.5372549 , ..., 0.5294118 ,\n",
       "         0.5254902 , 0.5137255 ]],\n",
       "\n",
       "       [[0.92941177, 0.9372549 , 0.9372549 , ..., 0.9647059 ,\n",
       "         0.972549  , 0.972549  ],\n",
       "        [0.92941177, 0.93333334, 0.93333334, ..., 0.95686275,\n",
       "         0.9607843 , 0.9647059 ],\n",
       "        [0.93333334, 0.93333334, 0.93333334, ..., 0.9490196 ,\n",
       "         0.9490196 , 0.9529412 ],\n",
       "        ...,\n",
       "        [0.76862746, 0.76862746, 0.77254903, ..., 0.73333335,\n",
       "         0.74509805, 0.68235296],\n",
       "        [0.7647059 , 0.7647059 , 0.76862746, ..., 0.7254902 ,\n",
       "         0.7372549 , 0.6784314 ],\n",
       "        [0.75686276, 0.75686276, 0.7607843 , ..., 0.72156864,\n",
       "         0.73333335, 0.67058825]]], dtype=float32)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "3e53aae8",
   "metadata": {},
   "outputs": [],
   "source": [
    "Y = np.load('E:\\Datasets\\Y.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "a08d86e4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1., 0., 0., ..., 0., 0., 0.],\n",
       "       [1., 0., 0., ..., 0., 0., 0.],\n",
       "       [1., 0., 0., ..., 0., 0., 0.],\n",
       "       ...,\n",
       "       [0., 0., 0., ..., 0., 0., 1.],\n",
       "       [0., 0., 0., ..., 0., 0., 1.],\n",
       "       [0., 0., 0., ..., 0., 0., 1.]])"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "87435dae",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2062, 64, 64)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.shape      # A stack of 2062 images each of size 64-by-64 - expressed as a 3 dimensional tensor."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "ad94cb01",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2062, 10)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "8bcad07a",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x2709a58a460>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X[260])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "0b1ea4c5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x2709ef63f40>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X[900])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "fec00042",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x2709e67bfd0>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X[2000])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "744d422d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x2709e899a30>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X[2061])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "6b005c21",
   "metadata": {},
   "outputs": [],
   "source": [
    "import sklearn\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "05b41bce",
   "metadata": {},
   "outputs": [],
   "source": [
    "? train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "50317992",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.46666667, 0.4745098 , 0.47843137, ..., 0.5176471 , 0.5137255 ,\n",
       "        0.5019608 ],\n",
       "       [0.49803922, 0.4862745 , 0.4862745 , ..., 0.5254902 , 0.5176471 ,\n",
       "        0.50980395],\n",
       "       [0.54509807, 0.49411765, 0.49019608, ..., 0.5294118 , 0.5254902 ,\n",
       "        0.5137255 ],\n",
       "       ...,\n",
       "       [0.5019608 , 0.5137255 , 0.5176471 , ..., 0.5529412 , 0.54509807,\n",
       "        0.53333336],\n",
       "       [0.49803922, 0.5058824 , 0.5137255 , ..., 0.54509807, 0.53333336,\n",
       "        0.52156866],\n",
       "       [0.49019608, 0.49803922, 0.5019608 , ..., 0.5294118 , 0.52156866,\n",
       "        0.50980395]], dtype=float32)"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X[0]    # a 64-by-64 pixel grid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "59d92911",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x270a301a130>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "bb93d0b3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1., 0., 0., 0., 0., 0., 0., 0., 0., 0.])"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Y[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "469d9146",
   "metadata": {},
   "outputs": [],
   "source": [
    "? plt.imshow    # in Grayscale"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "d54c7de4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x270a6a3fbb0>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X[0], 'gray')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "8e43f88e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "412.4"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# I am using one-fifth of the dataset for validation i.e 2062/5=413 for testing the trained neural network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "bdc64a57",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.20, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "9ac57750",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[0.56078434, 0.5647059 , 0.5647059 , ..., 0.5411765 ,\n",
       "         0.54901963, 0.54509807],\n",
       "        [0.56078434, 0.57254905, 0.5764706 , ..., 0.56078434,\n",
       "         0.56078434, 0.5529412 ],\n",
       "        [0.5647059 , 0.5764706 , 0.5803922 , ..., 0.57254905,\n",
       "         0.5686275 , 0.5568628 ],\n",
       "        ...,\n",
       "        [0.5137255 , 0.5176471 , 0.5176471 , ..., 0.49411765,\n",
       "         0.49411765, 0.4862745 ],\n",
       "        [0.50980395, 0.50980395, 0.50980395, ..., 0.49411765,\n",
       "         0.49019608, 0.47843137],\n",
       "        [0.5019608 , 0.5019608 , 0.5058824 , ..., 0.4862745 ,\n",
       "         0.48235294, 0.47058824]],\n",
       "\n",
       "       [[0.49411765, 0.5058824 , 0.52156866, ..., 0.46666667,\n",
       "         0.4509804 , 0.4392157 ],\n",
       "        [0.50980395, 0.5176471 , 0.5372549 , ..., 0.47843137,\n",
       "         0.45882353, 0.44705883],\n",
       "        [0.52156866, 0.5411765 , 0.56078434, ..., 0.48235294,\n",
       "         0.46666667, 0.4509804 ],\n",
       "        ...,\n",
       "        [0.4862745 , 0.5019608 , 0.5058824 , ..., 0.46666667,\n",
       "         0.4509804 , 0.4392157 ],\n",
       "        [0.48235294, 0.49411765, 0.49803922, ..., 0.45490196,\n",
       "         0.44705883, 0.43529412],\n",
       "        [0.47843137, 0.4862745 , 0.49411765, ..., 0.44313726,\n",
       "         0.43529412, 0.42352942]],\n",
       "\n",
       "       [[0.47843137, 0.49803922, 0.5058824 , ..., 0.5137255 ,\n",
       "         0.5019608 , 0.49411765],\n",
       "        [0.49019608, 0.5058824 , 0.5137255 , ..., 0.5254902 ,\n",
       "         0.5137255 , 0.5019608 ],\n",
       "        [0.49411765, 0.50980395, 0.52156866, ..., 0.5294118 ,\n",
       "         0.5176471 , 0.5058824 ],\n",
       "        ...,\n",
       "        [0.44705883, 0.45882353, 0.46666667, ..., 0.49411765,\n",
       "         0.48235294, 0.4745098 ],\n",
       "        [0.44705883, 0.45882353, 0.46666667, ..., 0.4862745 ,\n",
       "         0.4745098 , 0.46666667],\n",
       "        [0.4392157 , 0.4509804 , 0.45882353, ..., 0.4745098 ,\n",
       "         0.46666667, 0.45882353]],\n",
       "\n",
       "       ...,\n",
       "\n",
       "       [[0.5294118 , 0.5411765 , 0.54901963, ..., 0.5137255 ,\n",
       "         0.5058824 , 0.49411765],\n",
       "        [0.5372549 , 0.5529412 , 0.56078434, ..., 0.5254902 ,\n",
       "         0.5176471 , 0.5058824 ],\n",
       "        [0.54901963, 0.5647059 , 0.5764706 , ..., 0.53333336,\n",
       "         0.5254902 , 0.5137255 ],\n",
       "        ...,\n",
       "        [0.49019608, 0.5019608 , 0.5058824 , ..., 0.4627451 ,\n",
       "         0.45882353, 0.44705883],\n",
       "        [0.4862745 , 0.49411765, 0.5019608 , ..., 0.4509804 ,\n",
       "         0.44313726, 0.43529412],\n",
       "        [0.4745098 , 0.4862745 , 0.49019608, ..., 0.44313726,\n",
       "         0.43137255, 0.42352942]],\n",
       "\n",
       "       [[0.5137255 , 0.52156866, 0.53333336, ..., 0.49803922,\n",
       "         0.49019608, 0.47843137],\n",
       "        [0.52156866, 0.5294118 , 0.54509807, ..., 0.50980395,\n",
       "         0.49411765, 0.48235294],\n",
       "        [0.5294118 , 0.5411765 , 0.5568628 , ..., 0.5137255 ,\n",
       "         0.49803922, 0.4862745 ],\n",
       "        ...,\n",
       "        [0.4745098 , 0.4862745 , 0.49411765, ..., 0.4862745 ,\n",
       "         0.4745098 , 0.4627451 ],\n",
       "        [0.46666667, 0.47843137, 0.48235294, ..., 0.47843137,\n",
       "         0.47058824, 0.4627451 ],\n",
       "        [0.4627451 , 0.4745098 , 0.4745098 , ..., 0.46666667,\n",
       "         0.4627451 , 0.45490196]],\n",
       "\n",
       "       [[0.5372549 , 0.5411765 , 0.56078434, ..., 0.5019608 ,\n",
       "         0.49019608, 0.47843137],\n",
       "        [0.54509807, 0.5568628 , 0.5686275 , ..., 0.50980395,\n",
       "         0.49411765, 0.48235294],\n",
       "        [0.5529412 , 0.57254905, 0.5882353 , ..., 0.5137255 ,\n",
       "         0.5019608 , 0.4862745 ],\n",
       "        ...,\n",
       "        [0.52156866, 0.53333336, 0.54509807, ..., 0.43529412,\n",
       "         0.42745098, 0.41568628],\n",
       "        [0.5254902 , 0.53333336, 0.5411765 , ..., 0.43137255,\n",
       "         0.41960785, 0.40784314],\n",
       "        [0.52156866, 0.52156866, 0.5254902 , ..., 0.41960785,\n",
       "         0.4117647 , 0.4       ]]], dtype=float32)"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "0ddf938c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1649, 64, 64)"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "c5490d49",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.56078434, 0.5647059 , 0.5647059 , ..., 0.5411765 , 0.54901963,\n",
       "        0.54509807],\n",
       "       [0.56078434, 0.57254905, 0.5764706 , ..., 0.56078434, 0.56078434,\n",
       "        0.5529412 ],\n",
       "       [0.5647059 , 0.5764706 , 0.5803922 , ..., 0.57254905, 0.5686275 ,\n",
       "        0.5568628 ],\n",
       "       ...,\n",
       "       [0.5137255 , 0.5176471 , 0.5176471 , ..., 0.49411765, 0.49411765,\n",
       "        0.4862745 ],\n",
       "       [0.50980395, 0.50980395, 0.50980395, ..., 0.49411765, 0.49019608,\n",
       "        0.47843137],\n",
       "       [0.5019608 , 0.5019608 , 0.5058824 , ..., 0.4862745 , 0.48235294,\n",
       "        0.47058824]], dtype=float32)"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "370dc72c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(413, 64, 64)"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "id": "b1022303",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[0.78039217, 0.7921569 , 0.80784315, ..., 0.8509804 ,\n",
       "         0.84313726, 0.83137256],\n",
       "        [0.78431374, 0.80784315, 0.8156863 , ..., 0.85490197,\n",
       "         0.84705883, 0.8392157 ],\n",
       "        [0.79607844, 0.8117647 , 0.8156863 , ..., 0.85490197,\n",
       "         0.84705883, 0.8392157 ],\n",
       "        ...,\n",
       "        [0.6431373 , 0.64705884, 0.654902  , ..., 0.69411767,\n",
       "         0.6901961 , 0.68235296],\n",
       "        [0.63529414, 0.6392157 , 0.64705884, ..., 0.6862745 ,\n",
       "         0.68235296, 0.6745098 ],\n",
       "        [0.62352943, 0.627451  , 0.63529414, ..., 0.6745098 ,\n",
       "         0.67058825, 0.6627451 ]],\n",
       "\n",
       "       [[0.5529412 , 0.5647059 , 0.57254905, ..., 0.49411765,\n",
       "         0.47843137, 0.46666667],\n",
       "        [0.5647059 , 0.5764706 , 0.58431375, ..., 0.5137255 ,\n",
       "         0.49803922, 0.4862745 ],\n",
       "        [0.5764706 , 0.5882353 , 0.6       , ..., 0.5176471 ,\n",
       "         0.5137255 , 0.5019608 ],\n",
       "        ...,\n",
       "        [0.52156866, 0.52156866, 0.52156866, ..., 0.48235294,\n",
       "         0.4745098 , 0.46666667],\n",
       "        [0.5176471 , 0.5254902 , 0.5254902 , ..., 0.4745098 ,\n",
       "         0.47058824, 0.45882353],\n",
       "        [0.5137255 , 0.5176471 , 0.5254902 , ..., 0.4627451 ,\n",
       "         0.45882353, 0.45490196]],\n",
       "\n",
       "       [[0.6       , 0.6156863 , 0.627451  , ..., 0.53333336,\n",
       "         0.5294118 , 0.52156866],\n",
       "        [0.6117647 , 0.627451  , 0.6392157 , ..., 0.54509807,\n",
       "         0.5411765 , 0.5294118 ],\n",
       "        [0.61960787, 0.63529414, 0.6431373 , ..., 0.5529412 ,\n",
       "         0.54509807, 0.53333336],\n",
       "        ...,\n",
       "        [0.5372549 , 0.54509807, 0.54901963, ..., 0.47058824,\n",
       "         0.45490196, 0.44313726],\n",
       "        [0.53333336, 0.5411765 , 0.54509807, ..., 0.4627451 ,\n",
       "         0.4509804 , 0.4392157 ],\n",
       "        [0.5254902 , 0.5294118 , 0.53333336, ..., 0.45490196,\n",
       "         0.44313726, 0.43137255]],\n",
       "\n",
       "       ...,\n",
       "\n",
       "       [[0.5294118 , 0.5411765 , 0.5568628 , ..., 0.47058824,\n",
       "         0.46666667, 0.45882353],\n",
       "        [0.53333336, 0.54509807, 0.56078434, ..., 0.48235294,\n",
       "         0.4745098 , 0.4627451 ],\n",
       "        [0.5372549 , 0.5529412 , 0.5686275 , ..., 0.49411765,\n",
       "         0.48235294, 0.47058824],\n",
       "        ...,\n",
       "        [0.5254902 , 0.54509807, 0.56078434, ..., 0.4745098 ,\n",
       "         0.46666667, 0.45490196],\n",
       "        [0.5372549 , 0.54901963, 0.56078434, ..., 0.4627451 ,\n",
       "         0.45490196, 0.44313726],\n",
       "        [0.54509807, 0.5568628 , 0.5647059 , ..., 0.45490196,\n",
       "         0.44313726, 0.43529412]],\n",
       "\n",
       "       [[0.4509804 , 0.4627451 , 0.47058824, ..., 0.4392157 ,\n",
       "         0.43529412, 0.43137255],\n",
       "        [0.45882353, 0.47058824, 0.47843137, ..., 0.4509804 ,\n",
       "         0.44313726, 0.43529412],\n",
       "        [0.4627451 , 0.4745098 , 0.49019608, ..., 0.45882353,\n",
       "         0.44705883, 0.43529412],\n",
       "        ...,\n",
       "        [0.43137255, 0.4392157 , 0.44705883, ..., 0.40784314,\n",
       "         0.40392157, 0.39215687],\n",
       "        [0.41960785, 0.43137255, 0.4392157 , ..., 0.4       ,\n",
       "         0.39215687, 0.38039216],\n",
       "        [0.40784314, 0.41960785, 0.43137255, ..., 0.3882353 ,\n",
       "         0.38431373, 0.3764706 ]],\n",
       "\n",
       "       [[0.48235294, 0.5019608 , 0.50980395, ..., 0.45882353,\n",
       "         0.4509804 , 0.4392157 ],\n",
       "        [0.49411765, 0.5137255 , 0.52156866, ..., 0.4627451 ,\n",
       "         0.45490196, 0.44705883],\n",
       "        [0.5019608 , 0.5254902 , 0.53333336, ..., 0.46666667,\n",
       "         0.45882353, 0.44705883],\n",
       "        ...,\n",
       "        [0.5019608 , 0.5058824 , 0.50980395, ..., 0.48235294,\n",
       "         0.46666667, 0.45490196],\n",
       "        [0.49411765, 0.49803922, 0.5019608 , ..., 0.46666667,\n",
       "         0.4509804 , 0.4392157 ],\n",
       "        [0.47843137, 0.4862745 , 0.49019608, ..., 0.4509804 ,\n",
       "         0.4392157 , 0.42745098]]], dtype=float32)"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "6de6f17d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 1.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       ...,\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.]])"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "id": "d7128de5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1649, 10)"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "6bdd0c11",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 1.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       ...,\n",
       "       [0., 1., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 1., 0., ..., 0., 0., 0.]])"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "c09dca7f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(413, 10)"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "9d966161",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "a1737513",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "efc07a18",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.003921569"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "2fb3e7e9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.02745098"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "b94b00f8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x270acef0940>"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_train[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "0d6d504c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 0., 0., 0., 0., 1., 0., 0., 0., 0.])"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "4044824c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x270acf600d0>"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPsAAAD7CAYAAACscuKmAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAA500lEQVR4nO19abAc13Xed3qWt2NfCZAEKFEUNxEkYS6mZFFcZIaSRTm2UlZZMuNiFatkJZHLUizKqaTKSZwo5US2Eyt2sWzFdEzJprVR0U6D2iVTAkVSAsEFFEUSIEAABAjgPbxlZnpufrx5fb9zZrrfAHiYB2vOV/Xq3e57+/bt5U6fc88535EQAhwOx88+ksUegMPh6A18sjscfQKf7A5Hn8Anu8PRJ/DJ7nD0CXyyOxx9glOa7CJyi4g8JSLPiMhdCzUoh8Ox8JCTtbOLSAnA0wBuBrAHwA8AvDOEsHPhhudwOBYK5VM49ioAz4QQngUAEflbALcByJ3s5eGRUFm6AgAQxFRKTtm2LWgHhIK6LvsI+XUinX8Y8/afaNvQdlM6H8OtbG9Fl31yDfNPdjKfiaLbfbK98Pcq7x7adu2VdJxtp7bty9lFGUDhK8KvbUEfue2orjZ+GI3p4x1vwqlM9g0AdtP2HgBXFx1QWboCm3/zdwAAzZKuC5VYTgf0lTTL3C50LANAKNF2ualPwJefxHZSKniDzS0rVxtZuUTHlUrmXNyFeSrVcuzDvpj11NyUFiqlVG2XaPxpU3LrilCmPosmCCNtJma7u+P4HiTmkGaXs53HaMfboHE1GvoeNmmMzTS2s30061RXN9ptGttKai8gFpMatavrdkmDy7pOqE7oUSd1fSrebqtrzN7Ipz71R8jDqejsnZ5026MTkTtFZLuIbG9MHj+F0zkcjlPBqXzZ9wA4m7Y3AthrG4UQ7gZwNwAMnnV2mPtKB/Mz06SvdLMacuv45ynYrzJtS9mIvvZLP9ed+RLyL37bV7kaf3b5a2vbFX1di76o3A/X2f5KSbyWsr0HhAq142MAICmQK5s5X/q2r2GXEkER0oI+uh1HQl9s+yxSqkuprlkglTRNH/ylD/b7yO8jf/XN68YqRNN8E0t0XFDvd34fIa//gkdyKl/2HwA4X0Q2i0gVwK8B+Nwp9OdwOE4jTvrLHkJoiMi/AvAVACUAHwshPL5gI3M4HAuKUxHjEUL4IoAvLtBYHA7HacQpTfYThgChdca0TS/nstXFY7FoxZ319KRi6nJ01FI57bgfAMqm/6FqXALtdjXbnreS5K/cs17N+qrVr7kP2z+PZaAUl3ntij6jLPljapASWaSj27q6NbfktOvcahZ83c2C1fhSkq+NlqiPBq192FV7gO6P6P747rS9RQ3S5+17q/qg52nraBbyqn2RCdqueQWZf/3E3WUdjj6BT3aHo0/QczF+TnxnsR0AApvKrIhSJUFK8s1rCZnGrKMLi7uVSmzHTi6AdvooG9F3uGI8GVqwommRWYtRJD7zeMtG9K+SrGfPlSdq2z64XdH4q+TxYdvpPvRDs+frBkX9F6FEKkPJ3I86O9yQ09JM2zPiqWDGXsmtMU5zSecKg2bhJzbffMcWWOsNeLpNbw6H458QfLI7HH0Cn+wOR5+gpzp7AJkMrJsqmdSC0cXZxCZ0XGLalcmMZs1mCemQA+V8nZ0xaOrYlNWtXl5N7JoA6cOJXhNgMxfr83xeC+tuyjprkSuqOq8xk/EYKzRGu8bA480ztQFa964VnOtE1gQYvD5gx5ik8RXv9pklxpRX56Wash0juzhTuegE5jrZLVa9+yZgRrnStpne5j+vf9kdjj6BT3aHo0/Qew+6lsjSFtHD3kfWM45Fd/KMKxvvNxbdE2P6YdGdTWrWo63I62y4XMvKRWahIlG9SCxOckxx1pzEOFkxnsVi2z+rHhUa01CpptrV6SHWmvpVSpXoHutORCVp5njvWVWATVRNEzA/KFEGl7S7133GbDfJVhZsAD5zI9C7aaPjgrbR6f55m6LxxAyXX9U2raYLjc2/7A5Hn8Anu8PRJ+itGI+46t624s7ikPWMo232jLNecix2l00dr7pXqd1QWXvFsdhdNSLnYKmzB53FAPUxYFbjy1RXavPH6oy2QBjJD2rJQ93oTUWBNjzmUoGXHyM1cuUMie58rtR8X9gS0CyQRXm1v9y01xL7tCI+U1bxddprLlKVNCVWbjOkFBRjrU2gQBX77iuqqyJyloJgl+wWuAedw+Hwye5w9Al8sjscfYLem94qOTp7mXV2Y5IiHZsJIq3Oznr6kIlQY2II1tOtHs7eWCNlbWqy3nB5YHPVgOH8HSiIWFPjLdCVE7LdWP2ddfNudXurzzO67sPQUPD6CeviVre3prK8cU2hmpWtiZJNe7aumXSns6vzmrC0SpN19iJeevL4MyY6ZbJLrQcded6xN529N0XkFV18tv3L7nD0CXyyOxx9gh6L8SF6yrXxx3UOdgG06F6p5Hu4sehuTWosxrMnnPVwYw+voZI1y3UnxrNJzfbBYvGgTetByPOmm+0/XwQdROzTqhCMosASFp9Tkh27PW9Rf9OGtaRZYE5KyM6lPBYt80nCBBvGtJfTh/WAbJC4P1juTlRvr2NeQlunBqwr83jjbW4F5qqzr4d70Dkcjjn4ZHc4+gQ+2R2OPkHP3WUzGFdAKciKytFsrKdXTdQb6+VWn2cTG+vpNgqL9XKrow8n2hQ3hyJ3VquXc501a7GeXuRKyy6ntl23BA2cLpRJJYF2vbqbcVjkrgkUfF7aziud1z7a3HtJj55qVlUd3556AYknb7PuDQA1IVdd825ypB6TXth1J94OluufXcWL+N/VcSeeZ2/eL7uIfExEDojIDtq3QkQeEJFdrf/LT/jMDoejp+hGjP8rALeYfXcB2BZCOB/Atta2w+E4gzGvGB9C+KaIbDK7bwNwfat8D4CvA/jgvGcTZOK7lVa0ec0QPrAHHR1XRDxhPeMGmZSCxPgi81q7mN3Zc822Y+83W8fmsCIR3IrWeUhMHxy9xf1bkZvFz6rkqzx8XBEZxnTQ4nOTzq3MiJaSveA+5pnsrJrHXnlWxGfPu6JUVtUknx+/WpA6ix3jmOSiLSW0MvsZMZ4etYp0Mx50SjM6iWzZJ7tAtzaEsA8AWv/XnGQ/DoejRzjtq/EicqeIbBeR7en48dN9OofDkYOTXY3fLyLrQwj7RGQ9gAN5DUMIdwO4GwAGNm0Mc0H97cEu+fxxvALK/HGlghVVK86xKMncbwnyV9KLgkC6rWsT8UmOtSv1ViTPjikKmClYtVdWAemOeMNCBa4UiY5mGNMkPiOQ92JiGd4iKkZ1YRWCryU1A6mTa1lRAFEROYZqZywJKa2yWw+6lN7NesrvsOmDn6F9nqruJORznF7yis8BuL1Vvh3A/SfZj8Ph6BG6Mb19AsD3AFwgIntE5A4AHwZws4jsAnBza9vhcJzB6GY1/p05VTcu8FgcDsdpRM/JKzKd3XoYFXg3Kc84TuNU4P1mTW95JJBF5BJW/2P9/mTNa6yzWl2c9erx5mBWfmxio2o3kQ5k5UtG9qq6c6sHc8fPSBVvvG5XD/G1YD26aQRBXmMYMTJiKXQ2AVrUCogzmjmegtaMOMPPJRTo7F2mw7Jgk13D6PP8bvJ7m5p3WJjcww6DySvIlldIONlD05vD4fgnBp/sDkefoOdi/FzAS1GwS8mI+OzBpLjhC0xvVkROcnjhLK97tya1pECM1yYvw2NX0P+hdDQr/8lfvz0rn/WtSdWOL+2RV29RdW/4Nw9l5V9c+uOsbEX1oiAWHj8fNx10oMqDRy/Kyp/+/lY9xnqUM7dseTYr//q6h9AtanSdbKasBZsXKRZT47nGqpd91gz2DrSegkVpv3h7mvI1WQuasFm4ZMx3vMk8c/lOeG11Tl7hcDgy+GR3OPoEPtkdjj5Bj8krQmZyE8k3vVk3WHZ9Lee4vc5XV8rhDK/YlMo5+qqtKzKvsZ7eFlFWQAL57WOvycrrvzedlcsTWu9vjEZX1NXf2qfqvnTONVn559/9TFZeWZpQ7UrKpVfrsmx6Y9fU8XRItfuH/xvPddFn9yAPExs2ZOUvf+RSVXfjsp00jvz1DNbfbUQg682Dto7cVjmKrp1wJD7DWlsK67RjGdCutEymUk+1SVEk38TYlhcuO6hg2/LGu87ucDjm4JPd4egT9JyDbk6MbzO9kejerYhfNiKVTu9jUjYXkFLkocgzrlvzWlFkm1UTLiRvuB+uvTwrj6T6ftSWRhNYY2SVqtv4YDTTffWXLs7Kv7HqO+gWeRFyu43pbeBIHFe6fEzVhQpFrO05lJW//rXXqXZv+ZXHuh5XhHnu9E5MtpFjRPE5j0MQ0KoAp5sGgJTsaPadKyWUbivHmw4A6iSqp23vdyyrmhMxvXUB/7I7HH0Cn+wOR5+g92L83P82DrrOwS5A/mpoEUGF9aArKU40EvcNeUVRgAtv84qwPZfytGvjiMsP1Di7EsXdwxfHGzR00GQ+rRKfWVWv8g4cjoE22751WVb+zV/+tmqnaavzV6a5zl4LMz+nIyat00AcV+mVWLfkWdUMx9IY8LOspD0F9fOkIBNLLkHfrDaa6YT59OJ9q5gAHM4may00jQKOu7z30aqirLamJotrUzp/c4NdpT9JYotsfKd0tMPh+CcDn+wOR5/AJ7vD0Sfoqc4uEoklrU6j+OBL1rzR2dzWlm5Z1TVMXWfTmzWNsXmmzXwnnfuwHnTskTZSQLBoo8i4beXSo1m5+d0Rfdwy4nIfUFVI6tHLbS0FmD33Vm2iu2IwerwNdpkyal35qNqujcWH1hjV18IqcWPNkqw8uk8/s6en12flm8YeV3WsY/O9Ssw3qmgdhL3mFJGkcWhjMsohY6KrN2NjS4rCJjsmU6lX9BjZo86SUXLqs25540PZcM+30kwXmeT8y+5w9Al8sjscfYLem95aImMb8QSTUljut5wgFitmF2VBZXPKYI4nnO3DmoLYs8wGjzCKRHfuw5q8uO7nN/w0K/9oxWWqXWOITG/mCTYGKSjk5djfFw/rAJQ3btidla0/YV7IxpJkWm2zCtEYMubBSmd5cuCQvjfPHo/qxdhS3T/z8FV4lLZruo323h9vxkEWPbM61dUTfQf43bEcdEmOd6c1H/P7nu/HZ2CvU2WCNSK+B8I4HI45+GR3OPoEPtkdjj5Bj01vIXMbHKho/WmwTO6nbe6ysa5c5BJL25ZckN1i88gnAWCC3Df/6JGb9Pj3xLp0NPbxukueU+1+a8ODWdnq/aynj1kTD+mDb14WzVDf2ny5ajf2AkVQGdMbo3Is6q/f++l5qm747KiXTjYNwQa5b6YF402HaBxVrTTydlIn/vdJfb8PTUez4rDVqWmNQKVvNiZLjkpjHR3IXz8Zp+cMABXOJWBcafn9YzPcbF137rIqctNEfDYU4WQ+b3xgU1ybPp+zv0OTXIjI2SLyNRF5QkQeF5H3tfavEJEHRGRX6//y+fpyOByLh27E+AaA94cQLgRwDYD3ishFAO4CsC2EcD6Aba1th8NxhqKbXG/7AOxrlcdF5AkAGwDcBuD6VrN7AHwdwAfn6y/J4aDLM68B+amYrZccez4VccsV4c+/86as/No/17xttRUkjs7E/o6OnaPaffY/XZmV71j1TTOO/Ggz5oXbVHk5K0+v02Nf8jx5lq3Uv9eNIfIYq0Qe+spO3W7y9Uy+Yc1JndkUxmw0GFPSWZ70pr62OCgtwh48HsX4QWNKbbYZBefGp9tNkuhuiTfqZEjk67Tmtclmvj6Uly/AwpqMGUmRaYzfdxbdraWQH4v1ACzn3G8ew7wt+FwimwBcDuAhAGtbPwRzPwhrTqQvh8PRW3Q92UVkFMCnAPx2COHYCRx3p4hsF5Ht6bHJ+Q9wOBynBV1NdhGpYHai3xtC+HRr934RWd+qXw/gQKdjQwh3hxC2hhC2lpYML8SYHQ7HSWBenV1EBMBfAngihPARqvocgNsBfLj1//6FGpTVxVkXYl28PZ9bdww0VldmDD8fb0ltpeZJHz878rVzF8uf0Lr9l/8xurfe8VajsxekL2aMkUlweIPuP63GKDJrelPus6SjLtulz/uDmah13TB0WPefc3/sl6E5HPtsDOra2hKKiCMT3dABbfI6pi9Ngdc3UqWw6nPlEWTOto3FOvVnWXfYBDvdNMSaST4rUZ4Ob9edzgR0Y2e/DsC7AfxYRB5t7fs9zE7y+0TkDgAvAHjHaRmhw+FYEHSzGv9t5Jvqb1zY4TgcjtOFRYt6OxEwYUVZiWKWXDCKW5aUIi91kzXJTa2nbTPUlKT4lKLLJjZpcon134wHfuNNr1V1bx2LaZQrbapGLLOgumXdi6rdThLjrQmGSQ2a07FDe9s/tvcNWfmGV2kNrES/7c0ClSdQmm0b9abUCwrJYs57AAgHCtIiEZT6Yz49ikjS2Ku4LgndEZLad2eG0mG1q4eUmozeU2v65ajOJNHTTjnQ5XHIQxNQ2qi3yOSKXLhvvMPRJ/DJ7nD0CXouxp9OMB+Y5RZnbrmi1E3rX3Mw9jG4Mv9cJI1OrdCy07Jnoyh5987rVN1br45ifMmIXDxiFm4vHdNi/OOli+IxxsuKx8VlS27w2BPnZuX6eZZDj3nS44F1s/AsFOBiySqUekFVqVm1H9wft6fbVtnjCetdfpeKLC26XT5BShFs+qdys7N354kgYU555qczgTD8wrQFyXTBKe9fdoejT+CT3eHoE/hkdzj6BD0nr5jjhG8j5JN83aeUQzJpiScU+V9BBFK1gHjwzeufzMpfWPNGPX4OSCLnOkvcMLk63tbhr2lT0+Gt0YNsRUkTLA7n6HwXDO5T28rBy6YDa5JeR8NiIkoAGHs6jvHZX9SvwasrnaPNxoNuVx4nkouqbsvnVjnhjG4/9kJ8TntTnfb57HIMwWCTV1pgX7IebaXQ+Z2w5jVLiKH6pHO36/rdeUSeFEyuN9bTJc/0VtTdQozJ4XCc+fDJ7nD0CRbN9NYmqicnLg6diKkjyQmCqBoPOhaZ732N7mPsuVhm/jXLi8FBICP79XU9OBHNZr+x7GFVx9Yq7nJZokODWSwu4uRgs5wNmFnyfDzwK+OaU/6CFY/ROOJ1jje1rF45lu+hp0B11uNv4Fi8P7vr2tTJYrzyeAvWBhiL0+YEzBXPaZrTRfzOWS/SkHfvbMrmU4R/2R2OPoFPdoejT+CT3eHoE/RcZw9dJKUq0sVZd2uavkrKfGdMMOgc8WRNMKzPr7vyJVU39cy62K5O+mqBy2plQivV9+7ampXvvOqHyMMgKaKrS8d1//TUyobpi9VqvgX2lo68OJWVP/ncFlX3/pU74kaI47c69cARisKyjIqceZhusXXv5XbjqSYLsVGBGcypmrlKrwaTTzYLvnM2EpLTOdt3sxlO/XvJVjRhPb1pXJBP8Tz+ZXc4+gQ+2R2OPkFvPegAlBbQnHCyUUZFnGWsCvzShh+pur9esT6em7IKzawy6XxGWPwyrmXb4/b+K/Vv7VqOflJpl3T/9SWkhhzS3U+NUuTfcGzHaZ4BYOBoJP889pQm38CV1E7yPcvqdJ1GAgc7miVM+mFuRzlqExgrTaEbWPE+VR5uuq6Ana5rWMIKVbcAHnSVHI/FetqFW9wJwL/sDkefwCe7w9En6KkYHwCkrRXGovRPRWCRqtuUTkX9W0phxgUDOgDl+CVRdh9+MrqkpcO6j+ZAFL8mze/psqdi+Q9evFXV/c9zPp+VmUBizPKevTryL88c0cEjtRVxLM1BCh4xgTAT62P/S57V/R9O43WuL8cUUofSUdWOb2l9tMCCQlx4tVE9Ds7wap8Fx8xwTdrG6UDvhOljmtpWyC+xbl59PrcVzdliM9PMnzLWOqT6pzEOV7VyUWuQZ18a74e0EVTMn+G1aBb5l93h6BP4ZHc4+gQ+2R2OPkGPPegk86CznnSs71QtqR/pUEz4Z/XwAXJlsyYY7p+JCirG/a1OBA3DMqPqrnjV81n5qR0xJC6Z1r+ZzeE4xqYhtphZGts+/DXNKb/3XV/KymcRz3jFpFS+eF1cS9hZ1jp7c4C92sjb0FjQ6pF6HgMv6LqnqHI9ccMf4IOgPQebFcuiEYs6+k7fDzbF2QjEPEb57ldq9HugPCxN5Bzr/Zagogh5HnT23RytRGLTmVRPu7TZuY82DknW080h2TBOhTdeRAZF5Psi8piIPC4iv9/av0JEHhCRXa3/y+fry+FwLB66EeNnANwQQrgMwBYAt4jINQDuArAthHA+gG2tbYfDcYaim1xvAcCcrafS+gsAbgNwfWv/PQC+DuCD3Z64yExxspikKJC6ITEYJpe3IjGNxfq6ESTftDLazXYORjFeGvnXYsXbmZVxm81wAPC/9sfUef99wz9k5UHRj2nj8JGs/IS5FE7/hEqsDEYVaAzni/gPTb4qK18/tCsrTzQMAwaj4LOh+OsL3rgTMaV2Czaj5fHRAcXvRFLAj8h1rH42SybAip9LTWeynaZ0UEkRiYvklNFuiuuEbvOzl1oZXA8AeCCE8BCAtSGEfQDQ+r+moAuHw7HI6GqyhxDSEMIWABsBXCUil3R7AhG5U0S2i8j2xtHj8x/gcDhOC07I9BZCOIJZcf0WAPtFZD0AtP4fyDnm7hDC1hDC1vLSkU5NHA5HDzCvzi4iqwHUQwhHRGQIwE0A/huAzwG4HcCHW//vz+9lFiEAjZaZoZZqfZh1eKvPN8i8cZz0xi//5EI91p3RDJVoqxkmXx1NHx+49itZ2ZJXjBhzG+P8aiSzmNpA6aH3GVPKGI3f2I9ULjZz97/36cuy8t73PJCVzypr8+DPjf40K381uUrVsRkwDEYdMhhTUzrAv/P6fn9p38VZ+QMr4sLC4Xr+j7VaKzBoqlPnr28sMyQdFZW/mNYYTBf1BQikZAJKS0ZZpwdVtNbEZuGVZZ1D8Gg96unWVZxNb4Up25hz00SPShfE8d3Y2dcDuEdESpiVBO4LIXxeRL4H4D4RuQPACwDe0UVfDodjkdDNavyPAFzeYf8hADe2H+FwOM5E9JyDbk6EsV5D9WaUdxvGK4kNPg/tOycrr7tHmzCalSjuNgZ0H+seimLP/6jcnJU/cu19uWO1XngjSUzXdN75UaTf/9xG1S6ZiufmiC8ASGpxuzGsqrBqR1Qp3v/TX83Kf/3qv1ftLh2IKZxnVpo0WhMUYTaWL9qxJFkymsvzz6/Oys2LY8O9x5eqdtVjLFeaE6j+4zjKkzalUSyPJTodVp1EdzbK2ag3fYzxZgwsnp+cuZfNdJOGO58971ZXYzTihCXqL+qfzG1KxLeRmqdorXbfeIejT+CT3eHoE5wxVNK8ytkwIn5KdeP74or7qroWYTl76sxSfZ76SBS/zr03ist/vO4m1e73zvtCVh40S/osBr59fUyR9L8HtBg/cCiO33qnsafTzHI9xlcqsfEr92/Kyl95zzmq3S8Ox8iV8gbNJV3dHgkm6kujjGy9/FidsBLn4O44jn1p5IXbtXutareCRHCp5/fPaaLKU5aQIZZHTFDSyazGV8SSV3R+xdMToICepoc4nOhV9qVEoscehjVjauH3286BOhFWNO3FLSD8y+5w9Al8sjscfQKf7A5Hn2DRUjYXpWguSqmz7KyYxndmqTYF8WFWVZtZRjrk8VjZuHe9avfIBzZl5auHf6LqOCLu0sHdWXn6HK3HLfthXB+YOMd4S43G67aeXzMr4xiX74zlf//QbardrW/60ziOs/aqup9Mxmi8EpkA00FDikkWpPqIvt/Ln45tf+vZ6CtV3qvNTsdicByCie7TGZkofZK5aE4bZdMtM81kUTwc88hPmzo2nxbp6RwlOWMWWjgab7ii13Em0mj+VV54Ri+fTmOftaa+Ttbhm7xeVeCtJ0a374av1b/sDkefwCe7w9EnWDTTG5sbAGCwnC+yNEjsuXRNFFu3v2aFaje6h8wz5spYghs/J26seViLZX/+nTdl5a03P2s6IX46MsvddOkTqtlDT8SAFuaEmx0kmZdsRlASzY5tjiLzsu9q29i3rl2VlTeP6PxPL0xwFlCqMLc3Ia824xSGqRXx/vzkwc1ZOYyY8VbjtjXtSU7qomZF7y/VYh+HUh1oc1bplXgc7bcedPUC1zI2lzIZyWRT31M2r1lPu6WUKtd60CnPTypbVXSqEfufmNHnnqlToE2R6Y3rjBacd78Z/mV3OPoEPtkdjj6BT3aHo0+wCDr7/G1s1BuDo4zClnFVV9kVdT5rmuAIswYRLRy6WOtPmz4TzWhfu+YiVXfT2ONZmXWyt698WLX72lmXxjGNm3GsicrW4JA22dXrUeerrSQSij3aFPR7P/7lrPy2zT9WdRVyR+V7UKTTNUbytxuDHIWl2ymX2yGjRM50Pl/S0C9AZTIe99TMWarutdVX0AlWR2eTnTXfcR4ALluCCt5eXdbvFYNNbbN9xvN988Xz4v5HNbP6ip3xOpe8rAlTltKkmFoVn/WhS/S1pOvoODNFmgXkITmHOByOn1X4ZHc4+gS9TdkcJPMQKkr/VMRBx0L3peu199jOzRdk5epRLdbUKUsSawlWhE2HYuV9X3q9qrvwn8fzbaoczMqrS1rse+2VMU3U85/frOpmanTyIX1uJi6Qoei1Nb3KeHR9O3oOfqWiU0il68iLi1JJW15xZXkyErfkmOysF14RjzmnfGKav2bJRsfFcT0xqb0Z3zr6ODrBeqcVgc1oyrxm+lhWiua1ERPt+FI93u+Hj+oIxMe+ET0Wz/1ijIArHz6o2skU9Tmj1bfQoFRfk3EcS/6f9hs8+O5IGHV4i3kWC5H+yeFw/GzAJ7vD0SdYtNV4S1BRp20rxufR9y6vauKG5Nq4elv+7DJVV54kT6qxKDrWlmrx9uWL4y1Z+wMtRv3nV9+alf/0io9nZctVd+eGb2Tl3z5Hi33VF6MH1qQRrYXogQMF69SXapFtZB+tgn95laqb3EBehAPEbWasEyzi29989obj9FU22AW0AixTuo+E+mAxPjHub7w6/729m1Td9Gri8uMsq8bzkGNrrOeazthLAS1GVB+UOMjxptav/u7FrVn5+N9oi8GmpzonPknH9Kq9jES9KRnX4TrCYn0j6j/p0WOq3ZrPPpOVj21+tT7fnNWkYFHev+wOR5/AJ7vD0Sfwye5w9AkWgTe+8/48IkogX2e3utvV6yMR4zdes0zVje4mL6V1pIdak1Q5/v4d3aw9mFZ9PLrhfXTtDVn5/Ru/otptqhzOym+8aqeq2/7J6F3XGNYRVE3WoznAaVCPcfxcIi80aYAaS0hPJ/NdG7tBndZIBkwkGnnbKRpzQyoJbmfWBHSqaiaO1F0kNeJk36G9zvZfEskz15YiJ3uzyyg3QBNWjCXRNGbNa4fSeK5/nHiVqjvy2Q1ZeckhTYo5cU58J5iYo3Jcr/eUpmjbrFuUpmksScG1HYzmvOqx81Xd1GCBsj7X9bwtWmilbX5ERD7f2l4hIg+IyK7W/+Xz9eFwOBYPJyLGvw8AB27fBWBbCOF8ANta2w6H4wxFV2K8iGwE8BYAfwDgd1q7bwNwfat8D2ZTOX9wYYfXDiYLqCZapBopR3Fo89W7Vd3ew9EEViZrSX2ZEePJS2xmhf4tnCE++EN/GD3jPnqXTnn3/rOiWP+etQ+qul/bFINrhvdoNWHyLFIvOLCkpkW7OonqNktsGI73pFzNZ25jDzKrymj6dhLV7aeh0bHZ7Cadmh9T05CU1MfiK3jWN3WAyH+9Jpo6P/KqmAIrMbrAcQpwOW5IKZZQSilOLzXe1KaxwyTGf+q7OjPu2lfi/Z44y/DH0T0pT7PJUl9npcIqjxHjX6HjJuLLKRU9PY+9/eey8vRqbY5NGvnif9Zm3haz+GMAvwvNj7E2hLAPAFr/13TZl8PhWATMO9lF5K0ADoQQHp6vbc7xd4rIdhHZnh7r7IDgcDhOP7oR468D8DYRuRXAIIAlIvI3APaLyPoQwj4RWQ/gQKeDQwh3A7gbAAZftWH+JUOHw3Fa0E1+9g8B+BAAiMj1AD4QQniXiPwhgNsBfLj1//75+hIJSFp88dZsxhFfdcOrzS6QOjpOCyZM+HfhspdU3dMXxzxlSx6Nel1jJF/XKU3quqmYyRgDlL/siXsvVO2++95IVPnzw7tU3euvjGucP3r6ElU3vDeOf2otXaclhijTtjFLJpVYVx2IOvDMjEk6xyiIWGswYWZRRJVZV+BxMflnfdSY6Krxmkf36ndizz/EdZaXNkWdenVJS4hMIFEVvU7BbY8r8go9jgdfidGDy3+k36taPDVSY6ZkinnmtbA5/li5r75iuPOPHM3KyWgMw9xzu45onDiPrs0GvXWRI+5UnGo+DOBmEdkF4ObWtsPhOENxQk41IYSvY3bVHSGEQwBuLGrvcDjOHPScvCJt8cU3rOdXGkWxtKRFMSav4NQ5hncCdWpXNgmDbrroyaz83acjr/vAQRN9RxFxJvuvEtMOXUzjeFFfy5994i1x451fUHW3roiccduXaDGe+5li20ZZD6Q0lG9SK5U71wUj5gXyoGvjfKe2peh01hZ9x2iTIml7ijI9l4y4X56I29Vxrb6t+WGMBvu3j/9qVv7IxX+v2nFaLva0s2Bx30a2fetHkfhklbbooraEzWa6jinvuGyyT6N6jAgqdmuu/7B6ZVb+ye3xZtXWaZILmaYU3J7+yeFw5MEnu8PRJ1i09E/tHHRcNl5WJOKXSY6aSvWS51Aprj7XTP6n1dXIE5dcdSQrD3xBZ4KtkRdUbYkeO3uusddZbYn+zVzzcGz3Z3/7FlX3u7/+yTj+zVpMG3s+Xk+g1EowHm55ojoANCmtVkqyXbOuxyjTcduupDMtccoOaebToDzvjBjPz1Mo+slSSbOn2fRym0KKVtnvi6EXn3zfz6l271n9deRhnMgr2Lvu+xPnqXZDu6ndBlWlRPLSjKmjR1EmGu/qhJb3h5/cn5VDTT/3Xe89Oys3NxKxxYR+v0uT9ADsanxrHEXivH/ZHY4+gU92h6NP4JPd4egT9Nj0FnXKYPTOJhFO1lITykVgT7tykq+7JsZGcqwRTS03nvN0Vr7/6stVu9XfibeEPeYA48lG40irWlE6fFEc/4ZvaCXvvwz9SlYevfCoqptavSwrByZzTPIVMZtOK62RCbNBv+Wp1Yc56k33UVtF95U88mD1ftWHGQjr7NSFjQbjdrWldu2AvevimL75iStVuyXvjvbBNy/ZoerYU27HdNSNP/Go1vtH2Gtw2KyR5JBnAkCFUmQPvxzHOPLky3oca5Zl5Wd/eVTVhbOjnt6kvALJtL3fncsARb25zu5wOHyyOxx9gh6b3iQTLeuJlh3zuOkAICURv67Kuo+l1SjOWfPdQCnKaUOIstibL9Ni37YjW7Ly6Au6j2PDJLYO5MtLbK46cLkmU1jzcJS/Jl/QZj+T/ShDqaJVEmVea+jfayXWTxWoQ8Qn1zRpnUpj8f40SYS1Xnhsvgsjug/mxkuJsMKK8aWE+jBvYzpIx1HAzODL+t5/9aPXZeVPXH21qlu3MfIB7n8mcuxz0BEA8KtUOWbGSJrY4Cv6OtnENvIsqWWJfi5P3x656qprtJdfWqMLp+dpRXVNCGLVslbBxXiHw+GT3eHoE/hkdzj6BL3V2QOy8KimMa/V6x3ad0CzQCdhPX2orDscJFfamVL+Za+9NLo1Tj27VtWxnnd8E9uWDBkB/YTOrNR1r1Ty9Whl8qFIN3vJ1vWVocx0HNlmON9L5LGZjpnx591j46KpcsmZdQWOqlMXYC6fc84ZOnWUplh/JbfaVeZaKDJv2SPaxXRyR3yGw0wuoSn7tflRe7OiOs588HqQ1aNxLUiORl38+Xedq9pVVse6RsOQVrJZVJkpTV48cmsOVp/H/PAvu8PRJ/DJ7nD0CXorxguAlpgpRvQtk0fdHE9dp+1KKZ9AgSPpphpanJtJiVucXLpqzXyx+sjF+lxrvkepkqv5KZXTsXgtwaQyZpE2NSl7OP1TMpjvHWh551QVic8Jie6VY3oc6RCdu1JwT8kjT6yZjzjq20R/Fk3pWQfjbahUIKOeBDJ5sYrTxu9G5BIlnQ1ZqStFsi5HtiVGjC/VyJvRXGjlUEwbPnlJTOc8dcmUapekrNYYIpE8/jir8vAzs2m0FpA33uFw/BOHT3aHo0/QYw+6gKS1ylyuaDG1SmL8YNWspJejjMWecIMlTfRVVXW6jwEKmqkUBdCQfLRhVAeqPDL5mqy8/PHYbmKjSRPFSUvNimp9OYnM5XyRlkW7XDEPZiUXgHAgBa3eWm8sDlxhbjNbx+J4MFx4WlQ3A7Nt52BUgSLRmlf7i75LQsv4oWSDaagdi+p25Z9UBiahmO2fjquZFfKJqDccuHxFVh4e0XTXrGKmqb4WkXj/QzVes7W6sDdjYqjB567NySscDodPdoejX+CT3eHoE/RWZ28K0qnWKa35gXSNRlP/BjUqrNOQDmlTHzFLgjFbcN0Q2VYGTNpn3l5Z1dFJg2+I6wCPvRg535c8p/XTcRr/1EajLJNOVjLmtWbaWYG1ejlIL2/7uaYuWQ9tjOYrczJj9FxOccS6t03tXKBvKwtVznXNNizQ+5mTndY+7HO36xF5dfx6WP533rYpnngNY+CwYZxsEgHqa6P+bnMasJ7etBz+nCqLowztfVPrJ/Z5zm966zY/+3MAxjH7KjVCCFtFZAWAvwOwCcBzAP5FCOGVbvpzOBy9x4mI8W8KIWwJIWxtbd8FYFsI4XwA21rbDofjDMWpiPG3Abi+Vb4HszngPjjvUS0xxZoVGtId71wpiUO2BBVF0JlgY3mkrMUyThFUMfLh2oHIPV97fSwv+fiwasc8ZVZEDsRlYb0IKwPxfI06qS6GhELygkwAlCeI3IMCXKwJMOEgE2uHovtaGiEiCxuAU3D/uU/Vuwl4kiLzXR7nmhXBqY82wgdqq/jjbOZaMqklNnXTeOyk/Nx+VXfs2k1ZeXgsqn31uiHHING92cw3qakAImOmFA6EMY9iZtXsGJsFM7rbL3sA8FUReVhE7mztWxtC2AcArf9rco92OByLjm6/7NeFEPaKyBoAD4jIk/Me0ULrx+FOACitWHbiI3Q4HAuCrr7sIYS9rf8HAHwGwFUA9ovIegBo/T+Qc+zdIYStIYStpTG7RulwOHqFeb/sIjICIAkhjLfKbwbwHwF8DsDtAD7c+n//vGcj8oowY/QWKhuVSZl4Jml/WjbplkkXahilRuvsktuuSgpbyejU7Ep7zdnPZeWHLr5UtRvdQ2SLg9aVNt7yYFIvsw4fmHjQ6G6sb5cntfLJkXRMJGlT/LK/aJvqzQQKdO5QpLNbvZ/10gLee5suWg2R1juUHm29jAtMb6o/yjNnj2Ez5dAhXTny0+g2HVK9YPDStfE6OaAvtQQVbF4zJrU8Pd0SjiQ0Z9p447vI9daNGL8WwGdayfnKAD4eQviyiPwAwH0icgeAFwC8o4u+HA7HImHeyR5CeBbAZR32HwJw4+kYlMPhWHj0nDd+zkxiRUf2FmoasZUlOCayML5Mip8uMfJMnug+2LRcdVH8simkWKxncT+5QkfHDT0W1ybSqhbnGsQ9X6/p6wxEnM4eaDalcpl5EQr475KZ/CUZKfTGov5miLzC9KeOK5If1Ynzq5JpI7bSg1fmtQL+i0IwX8ekMXvS9sArhr3ipYNZsfa6TaqqeVb0mqtRJFpbSnIS1ds843JE9zYxXt0PmDpP/+RwOFrwye5w9Al8sjscfYIeR72Ry5/V2cm+FsxvUJNMN9YNUfVR4L7J+eK4XcOY7xpEQGlTQvM6AOeqe93avardo5ddmJVX7dCGxNqSeMvTIc020hym85E+rFJFA6iTYm71eXYJbdJNtimVlc5n3gJl6mO1vG4fGhcLvhs55JMW1k1VLZnwcfY557nEwuZHy2/H/O+l40Znr8e6F984qKqSJBqD2SXW6uWBTYzWDTZHTy96ttZk6bneHA5HBp/sDkefoOfpn+a4zNs5FEnMNmFNoUQpinm/iR5KiFPeivR18nyqE5HAjElDNZNDbgkAZTL7Ndj0ZkTTwasOxXHsXKbqRvdSuqMB4wG4lMyPZNJJhvQ4UsoJbcXFPPOM5RVX4r5JhYRaZ3XIio5KNbA85tOdVYG2VNe8aU+bp5YVEE9Y5Inu5RljVp2mvAWvjKu6+qXnZeWZzZqYnj0T2eux0LxmRXAW3bncFsFH4r655m486PzL7nD0CXyyOxx9gp6K8YIofiRmZdckfFJb7BWmuM1s5lAinrDpiJKEAmFoBd4SCdQogKFa1renQmJ8ico2Y+y6sSgGPn/BSlW3/h+jGNgYHFB1ExXKVMoZWKvmOjmwxASZNMudxcBkJkckRnvKJA6mCQUL6Yp4wnw2+PmyOmGDQIpE8KIF+Lw+SsatkjngK5P0/KbMGzcT5f3my4dV3cu3boztyvpm5XLGWVGdtq0Yn3ev2og4CjzofDXe4XBk8MnucPQJfLI7HH2CHke9Rf3K6mBKd7OeWmyWY2IFY7+TMpveDOEf6bZ8XCPRv3elEpvodB9lpbPH/mqm3WglemA1r9BmHPlWbDt4ROuNM8tjXf3laA/jVM4ACskg1M+3ZQHhcbCea6LNmLRQ8bUbLzzWNUObjtq5nTQLPPkK+OC5xnq/lYnRxOZpK0931tnLx/XNSY7GTmRsVNWNnxvLlnRTEU8o4st881qu9xusqbCgnbkH2XGuszscDp/sDkefoOcedHMiTGLkjabKJZRPyCAs0hv7mtq0dSz6UlrfphErG2R6Y6KM2e3Ytkwqw1RNB7TUqvG2LhudVHX7r1qbldd/V6eXml7GpBdEcmHJPEZIXbHqEAeuNDvvB4o51MtTcQdTrhnNSIvgNhNzjkmtbbxcLiKhCDllc662tE6sChAHXWnG8P9NRZtdc502lzaWUVvrGcfn5jp7/QU8H/qZUR9F7ey96oLAw7/sDkefwCe7w9En8MnucPQJeh/11jIRNI2iyDq8zeGWcGQRm3/M6EPI1+c59bDiRCj4uWta8x3p98wLztF2gE7PWy5r3fD4ZdHdsr5Dh5uNvBTtLulAvDixJily8WXXVgsevkzpOqVvGzW0pNqS/m4i1nLzqAFoqmUM4mtv0zUL9Fzun0xS1r03qZMuXtMn4O0SRbpJXT+XMBHXTw7deI4+QYUuzvLc54y/zSWW3WXzXF0Lym3bOfp8gVexf9kdjn6BT3aHo0/Q26i3ED2r7K8Mi/VtZjnVB5nN8h3tOkRJ5ZBjFJgs2qtIFCN1wnrrpSSrpsZUUx2KIuGBK8ZU3dlfPpKV6yOxTlJ9t9jEU9ddtJnHckHDSg15hTJlMe/ZCbwtKlqO+24Tb7lO96HIJki1KM0YLznymqse17pA5ViUfcvHY4eWoCJQvoCj5+txKNG9zcuvs2dckfdbm3mwncllXhRFC+ahqy+7iCwTkU+KyJMi8oSIXCsiK0TkARHZ1fq//MRP73A4eoVuxfg/AfDlEMJrMZsK6gkAdwHYFkI4H8C21rbD4ThD0U0W1yUAfgHAvwSAEEINQE1EbgNwfavZPQC+DuCD8/Y3FwhjYztIhLNSTZ5znUj34o/Nx5rtt0ElRUEmbBWglXk7DA7ISZv5ATm1C/Wy8uSOGIAx9kxcHU4268AMXo234mJKTMd8apOsVgW7WLD4rNSCKXOhBWoTexvySnRiyCVU4Ectv64yQavqdf2MquPxflePaLNAaSJ2WiLRvbFb039Pve3KrFxbZbzr2IOxbYU8R3Qv4MkrXI3n+1FArd2mCixQIMx5AA4C+D8i8oiI/EUrdfPaEMI+AGj9X9NFXw6HY5HQzWQvA7gCwJ+FEC4HcBwnILKLyJ0isl1Etjcmj5/kMB0Ox6mim8m+B8CeEMJDre1PYnby7xeR9QDQ+n+g08EhhLtDCFtDCFvLwyOdmjgcjh6gm/zsL4nIbhG5IITwFGZzsu9s/d0O4MOt//fPe7YQdQ2rc7CqYfVQxTteZl3QjpXMch3OHdvR/jaFm8kcbScEJhq0p+LUxjYajEkyzfrA3jdEBfn8v4i2ppHdhh+/PpSVp2f0mkCd1PvGEJF0DKlmoOzQbSYvda4CksNQsCbQ9gyz/XqbTWqWeILbDozHk5emdbvKeGxYGtfrIMlE3G4ejHz+5bWrVbt911Jq6tS8PUqn7tIzro2cM78P5XmXE7U4e1zn8uw4QsfzMrq1nP5rAPeKSBXAswB+E7Ov8X0icgeAFwC8o8u+HA7HIqCryR5CeBTA1g5VNy7oaBwOx2lDzzno5sSMtoAIdmqzkjWLQCyql00nSgI3vGp55AdG/GTx1ppZctEt6QJMyirD6dZYEWWzg2+IJBdrtu1R7QYqxCnflruJL0DZKVUr9porJLbI6xpQ96fNTMQipwposd5vscwmNEB7ypUnKT1TPZ/zXWb0QMLRY7E8E+1+h264VLVLRzhIpiBLbJEIXhAIUxTgkuSY3tracZ0J9Cok/pg7fv4mDofjZwE+2R2OPoFPdoejT9B73vg53cLqw2xBKtCVldXM2tdYTbL6DnLMckUet9YCmOdKeyJBS7wgYfU/wuHXxXON7Vmr6oZ27qOtVaquMdg5p501jRWBdXbWDVNzobzVRhpBFrAiN88y6eUVE7HG3O6laVJYG0ZnnyQf3P0HVV165GhWDtdtycovb7G6d77SqyLb2tIoU7lLgoq2VMw5LrLtfZDZOacPT9nscDh8sjsc/QJp42o7nScTOQjgeczKni/37MT58HFo+Dg0zoRxnOgYzg0hrO5U0dPJnp1UZHsIoZOTjo/Dx+HjOE1jcDHe4egT+GR3OPoEizXZ716k81r4ODR8HBpnwjgWbAyLorM7HI7ew8V4h6NP0NPJLiK3iMhTIvKMiPSMjVZEPiYiB0RkB+3rORW2iJwtIl9r0XE/LiLvW4yxiMigiHxfRB5rjeP3F2McNJ5Si9/w84s1DhF5TkR+LCKPisj2RRzHaaNt79lkF5ESgI8C+GcALgLwThG5qEen/ysAt5h9i0GF3QDw/hDChQCuAfDe1j3o9VhmANwQQrgMwBYAt4jINYswjjm8D7P05HNYrHG8KYSwhUxdizGO00fbHkLoyR+AawF8hbY/BOBDPTz/JgA7aPspAOtb5fUAnurVWGgM9wO4eTHHAmAYwA8BXL0Y4wCwsfUC3wDg84v1bAA8B2CV2dfTcQBYAuCnaK2lLfQ4einGbwCwm7b3tPYtFhaVCltENgG4HMBDizGWluj8KGaJQh8Is4Sii3FP/hjA70LHJy3GOAKAr4rIwyJy5yKN47TStvdysncK8epLU4CIjAL4FIDfDiEcm6/96UAIIQ0hbMHsl/UqEbmk12MQkbcCOBBCeLjX5+6A60IIV2BWzXyviPzCIozhlGjb50MvJ/seAGfT9kYAe3Pa9gJdUWEvNESkgtmJfm8I4dOLORYACCEcwWw2n1sWYRzXAXibiDwH4G8B3CAif7MI40AIYW/r/wEAnwFw1SKM45Ro2+dDLyf7DwCcLyKbWyy1vwbgcz08v8XnMEuBDXRLhX2KkNl8VX8J4IkQwkcWaywislpElrXKQwBuAvBkr8cRQvhQCGFjCGETZt+HB0MI7+r1OERkRETG5soA3gxgR6/HEUJ4CcBuEbmgtWuOtn1hxnG6Fz7MQsOtAJ4G8BMA/66H5/0EgH0A6pj99bwDwErMLgztav1f0YNxvB6zqsuPADza+ru112MB8DoAj7TGsQPAf2jt7/k9oTFdj7hA1+v7cR6Ax1p/j8+9m4v0jmwBsL31bD4LYPlCjcM96ByOPoF70DkcfQKf7A5Hn8Anu8PRJ/DJ7nD0CXyyOxx9Ap/sDkefwCe7w9En8MnucPQJ/j/FqEg6XjduuQAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_train[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "f396c441",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 0., 0., 0., 0., 1., 0., 0., 0., 0.])"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "a2e8e0d1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x270aae368b0>"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_train[100])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "2131ef33",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 0., 0., 0., 1., 0., 0., 0., 0., 0.])"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train[100]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "4ac061b6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1649, 10)"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "5e120281",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1649, 64, 64)\n",
      "(1649, 10)\n",
      "(413, 64, 64)\n",
      "(413, 10)\n"
     ]
    }
   ],
   "source": [
    "print(X_train.shape)\n",
    "print(y_train.shape)\n",
    "\n",
    "print(X_test.shape)\n",
    "print(y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "91ed7f8c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 1.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       ...,\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.]])"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "8804a57f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x270a8c4d520>"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_train[2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "06b2d385",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x270a8cb1670>"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(X_train[3])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9abe2ae9",
   "metadata": {},
   "source": [
    "#### Splitting the Training and Test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "id": "ecb5f336",
   "metadata": {},
   "outputs": [],
   "source": [
    "? train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "08111434",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_reshaped = X_train.reshape(-1, 64, 64, 1)\n",
    "X_test_reshaped = X_test.reshape(-1, 64, 64, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "id": "92a522b5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(413, 64, 64, 1)"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test_reshaped.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "id": "ec9fcc1a",
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.optimizers import SGD, Adam\n",
    "from keras.layers import BatchNormalization"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c20406d3",
   "metadata": {},
   "source": [
    "#### Building the layers of the neural network..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "id": "9f638a05",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(Conv2D(filters = 64, kernel_size=(3,3), activation='relu', input_shape = (64, 64, 1)))\n",
    "model.add(MaxPooling2D(pool_size = (2,2)))\n",
    "model.add(BatchNormalization())\n",
    "\n",
    "model.add(Conv2D(filters= 32, kernel_size=(3,3), activation='relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "model.add(BatchNormalization())\n",
    "\n",
    "model.add(Conv2D(filters=16, kernel_size=(3,3), activation='relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2,2)))\n",
    "model.add(BatchNormalization())\n",
    "\n",
    "model.add(Flatten())\n",
    "\n",
    "model.add(Dense(512, activation='relu'))\n",
    "model.add(Dropout(0.22))\n",
    "model.add(Dense(1024, activation='relu'))\n",
    "model.add(Dropout(0.4))\n",
    "\n",
    "model.add(Dense(10, activation='softmax'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "id": "8d38ff83",
   "metadata": {},
   "outputs": [],
   "source": [
    "# model.compile(optimizer=SGD(learning_rate=0.1), loss=\"categorical_crossentropy\", metrics=['accuracy'])\n",
    "model.compile(optimizer=Adam(learning_rate=0.001, beta_1=0.9, beta_2=0.999), loss=\"categorical_crossentropy\", metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "id": "54fd4853",
   "metadata": {},
   "outputs": [],
   "source": [
    "? ImageDataGenerator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "id": "e05fb8a3",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_gen = ImageDataGenerator(featurewise_center = False,                      # set input mean to 0 over the dataset\n",
    "        samplewise_center = False,                                              # set each sample mean to 0\n",
    "        featurewise_std_normalization = False,                                  # divide inputs by std of the dataset\n",
    "        samplewise_std_normalization = False,                                   # divide each input by its std\n",
    "        zca_whitening = False,                                                  # reducing the dimensions\n",
    "        rotation_range = 10,                                                    # randomly rotate images in the range 10 degrees\n",
    "        zoom_range = 0.1,                                                       # Randomly zoom image 1%\n",
    "        width_shift_range = 0.1,                                                # randomly shift images horizontally 1%\n",
    "        height_shift_range = 0.1,                                               # randomly shift images vertically 1%\n",
    "        horizontal_flip = True,                                                 # randomly flip images\n",
    "        vertical_flip = True)                                                   # randomly flip images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "id": "0e7f334d",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_gen.fit(X_train_reshaped)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "id": "7562c019",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Sai Ganesh\\AppData\\Local\\Temp\\ipykernel_16952\\1503049174.py:1: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n",
      "  History = model.fit_generator(train_gen.flow(X_train_reshaped, y_train, batch_size=32),\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "52/52 [==============================] - 10s 168ms/step - loss: 0.2977 - accuracy: 0.8993 - val_loss: 0.0877 - val_accuracy: 0.9709\n",
      "Epoch 2/20\n",
      "52/52 [==============================] - 9s 170ms/step - loss: 0.3380 - accuracy: 0.8939 - val_loss: 0.1058 - val_accuracy: 0.9588\n",
      "Epoch 3/20\n",
      "52/52 [==============================] - 10s 186ms/step - loss: 0.2879 - accuracy: 0.9048 - val_loss: 0.1333 - val_accuracy: 0.9564\n",
      "Epoch 4/20\n",
      "52/52 [==============================] - 9s 176ms/step - loss: 0.3079 - accuracy: 0.9018 - val_loss: 0.2638 - val_accuracy: 0.9322\n",
      "Epoch 5/20\n",
      "52/52 [==============================] - 9s 169ms/step - loss: 0.2618 - accuracy: 0.9121 - val_loss: 0.0615 - val_accuracy: 0.9758\n",
      "Epoch 6/20\n",
      "52/52 [==============================] - 9s 169ms/step - loss: 0.3129 - accuracy: 0.8975 - val_loss: 0.1877 - val_accuracy: 0.9370\n",
      "Epoch 7/20\n",
      "52/52 [==============================] - 10s 185ms/step - loss: 0.2787 - accuracy: 0.9181 - val_loss: 0.0920 - val_accuracy: 0.9685\n",
      "Epoch 8/20\n",
      "52/52 [==============================] - 9s 168ms/step - loss: 0.3009 - accuracy: 0.9084 - val_loss: 0.0941 - val_accuracy: 0.9709\n",
      "Epoch 9/20\n",
      "52/52 [==============================] - 9s 164ms/step - loss: 0.2824 - accuracy: 0.9066 - val_loss: 0.1319 - val_accuracy: 0.9467\n",
      "Epoch 10/20\n",
      "52/52 [==============================] - 9s 173ms/step - loss: 0.2386 - accuracy: 0.9339 - val_loss: 0.0948 - val_accuracy: 0.9685\n",
      "Epoch 11/20\n",
      "52/52 [==============================] - 9s 177ms/step - loss: 0.2591 - accuracy: 0.9242 - val_loss: 0.6867 - val_accuracy: 0.8402\n",
      "Epoch 12/20\n",
      "52/52 [==============================] - 10s 198ms/step - loss: 0.2729 - accuracy: 0.9242 - val_loss: 0.1533 - val_accuracy: 0.9370\n",
      "Epoch 13/20\n",
      "52/52 [==============================] - 10s 183ms/step - loss: 0.2498 - accuracy: 0.9230 - val_loss: 0.3635 - val_accuracy: 0.8886\n",
      "Epoch 14/20\n",
      "52/52 [==============================] - 10s 198ms/step - loss: 0.2582 - accuracy: 0.9090 - val_loss: 0.1092 - val_accuracy: 0.9637\n",
      "Epoch 15/20\n",
      "52/52 [==============================] - 8s 161ms/step - loss: 0.2296 - accuracy: 0.9224 - val_loss: 0.0691 - val_accuracy: 0.9806\n",
      "Epoch 16/20\n",
      "52/52 [==============================] - 8s 163ms/step - loss: 0.2162 - accuracy: 0.9321 - val_loss: 0.5469 - val_accuracy: 0.8692\n",
      "Epoch 17/20\n",
      "52/52 [==============================] - 8s 161ms/step - loss: 0.2283 - accuracy: 0.9284 - val_loss: 0.0648 - val_accuracy: 0.9806\n",
      "Epoch 18/20\n",
      "52/52 [==============================] - 9s 165ms/step - loss: 0.2123 - accuracy: 0.9242 - val_loss: 0.0727 - val_accuracy: 0.9758\n",
      "Epoch 19/20\n",
      "52/52 [==============================] - 8s 160ms/step - loss: 0.2178 - accuracy: 0.9290 - val_loss: 0.0550 - val_accuracy: 0.9758\n",
      "Epoch 20/20\n",
      "52/52 [==============================] - 8s 161ms/step - loss: 0.2198 - accuracy: 0.9272 - val_loss: 0.0854 - val_accuracy: 0.9637\n"
     ]
    }
   ],
   "source": [
    "History = model.fit_generator(train_gen.flow(X_train_reshaped, y_train, batch_size=32), \n",
    "                             epochs=20, validation_data=(X_test_reshaped, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "id": "7495ce19",
   "metadata": {},
   "outputs": [],
   "source": [
    "epochs = [i for i in  range(1, 21)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "id": "c38300bf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20]"
      ]
     },
     "execution_count": 158,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epochs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "id": "a8840f2e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['loss', 'accuracy', 'val_loss', 'val_accuracy'])"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "History.history.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "id": "791e22fa",
   "metadata": {},
   "outputs": [],
   "source": [
    "training_accuracy = History.history['accuracy']\n",
    "validation_accuracy = History.history['val_accuracy']\n",
    "\n",
    "training_loss = History.history['loss']\n",
    "validation_loss = History.history['val_loss']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "id": "3345dcef",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "13/13 [==============================] - 0s 32ms/step - loss: 0.0854 - accuracy: 0.9637\n"
     ]
    }
   ],
   "source": [
    "score = model.evaluate(X_test_reshaped, y_test, verbose='auto')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "id": "3a25e6d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.08537979423999786, 0.9636803865432739]"
      ]
     },
     "execution_count": 160,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "id": "98447f73",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The final accuracy is:  96.36803865432739\n",
      "The final loss is:  0.08537979423999786\n"
     ]
    }
   ],
   "source": [
    "print(\"The final accuracy is: \", score[1]*100)\n",
    "print(\"The final loss is: \", score[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cd150a35",
   "metadata": {},
   "source": [
    "### Analyzing the Model's Evaluated Performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "id": "3e8afacf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(epochs, training_accuracy, 'g', label = 'train_acc')\n",
    "plt.plot(epochs, validation_accuracy, 'b', label='val_acc')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracy Metric')\n",
    "plt.title('Training vs Validation Accuracy')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "id": "ab089b51",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(epochs, training_loss, 'g', label = 'train_loss')\n",
    "plt.plot(epochs, validation_loss, 'b', label='val_loss')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss Metric')\n",
    "plt.title('Training vs Validation Loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "id": "57d589c0",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "13/13 [==============================] - 10s 36ms/step\n"
     ]
    }
   ],
   "source": [
    "predictions = model.predict(X_test_reshaped)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "id": "2ae35445",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1.2740564e-02, 8.0347592e-05, 3.2773800e-04, ..., 6.0994469e-04,\n",
       "        1.9817080e-06, 4.8995456e-03],\n",
       "       [1.0723685e-08, 6.1754976e-18, 2.5933968e-15, ..., 3.8926029e-10,\n",
       "        3.8547730e-22, 1.0000000e+00],\n",
       "       [3.9930347e-07, 5.7989677e-08, 3.0542567e-04, ..., 8.0896752e-06,\n",
       "        2.8174938e-04, 4.5509627e-10],\n",
       "       ...,\n",
       "       [3.2432212e-15, 1.0000000e+00, 1.3007486e-14, ..., 3.8910821e-17,\n",
       "        4.3171660e-18, 4.7478093e-20],\n",
       "       [9.1356860e-04, 1.2233512e-03, 2.5385753e-03, ..., 1.0494658e-03,\n",
       "        1.3189601e-03, 3.3620174e-05],\n",
       "       [8.4932136e-14, 1.0000000e+00, 3.1417220e-16, ..., 2.1421779e-20,\n",
       "        1.0624998e-20, 7.4686364e-21]], dtype=float32)"
      ]
     },
     "execution_count": 167,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "id": "0ec006d2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(413, 10)"
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bfce0d1e",
   "metadata": {},
   "source": [
    "#### Some Diagnostic Plots..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "id": "673ed7b3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1.2740564e-02, 8.0347592e-05, 3.2773800e-04, 4.0914990e-02,\n",
       "       1.4785788e-07, 1.8065232e-04, 9.4024414e-01, 6.0994469e-04,\n",
       "       1.9817080e-06, 4.8995456e-03], dtype=float32)"
      ]
     },
     "execution_count": 169,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "id": "8250e67c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 0., 0., 0., 0., 0., 1., 0., 0., 0.])"
      ]
     },
     "execution_count": 170,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "id": "06abf650",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1.07236851e-08, 6.17549760e-18, 2.59339677e-15, 3.71554228e-19,\n",
       "       1.02850177e-22, 3.57575124e-12, 1.15002488e-08, 3.89260291e-10,\n",
       "       3.85477298e-22, 1.00000000e+00], dtype=float32)"
      ]
     },
     "execution_count": 171,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "id": "c15ce360",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 0., 0., 0., 0., 0., 0., 0., 0., 1.])"
      ]
     },
     "execution_count": 172,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "id": "6db3136f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x270b541a0d0>]"
      ]
     },
     "execution_count": 173,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(predictions[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "id": "54505596",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x270b577f790>]"
      ]
     },
     "execution_count": 177,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(predictions[4])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "id": "a9f5d766",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([9.73002553e-01, 2.12124769e-06, 2.37244694e-06, 1.02953805e-07,\n",
       "       2.30091203e-07, 2.69846953e-02, 6.53835968e-06, 1.25299562e-06,\n",
       "       1.52522844e-07, 1.44777959e-08], dtype=float32)"
      ]
     },
     "execution_count": 178,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions[4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "id": "479e30c4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1., 0., 0., 0., 0., 0., 0., 0., 0., 0.])"
      ]
     },
     "execution_count": 179,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test[4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "id": "efc7b920",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x270b50bd3d0>]"
      ]
     },
     "execution_count": 180,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(predictions[100])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "id": "96b100ea",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([6.1060797e-04, 6.1783830e-05, 4.0888813e-01, 1.0609956e-04,\n",
       "       9.4344718e-03, 5.7732773e-01, 1.5156460e-04, 8.2405307e-04,\n",
       "       2.5934272e-03, 2.2306190e-06], dtype=float32)"
      ]
     },
     "execution_count": 181,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions[100] # As can be seen, there's a significant flaw in the prediction here. It's supposed to be at index-2 but there is an even bigger spike at index-5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "id": "9ed75311",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0., 0., 1., 0., 0., 0., 0., 0., 0., 0.])"
      ]
     },
     "execution_count": 182,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test[100]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb033206",
   "metadata": {},
   "source": [
    "### Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "id": "7632b139",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 1.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       ...,\n",
       "       [0., 1., 0., ..., 0., 0., 0.],\n",
       "       [0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 1., 0., ..., 0., 0., 0.]])"
      ]
     },
     "execution_count": 189,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "id": "cff62fb6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(413, 10)"
      ]
     },
     "execution_count": 192,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "id": "f1bc3674",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([6, 9, 3, 9, 0, 5, 8, 2, 5, 9, 4, 9, 7, 1, 3, 3, 0, 5, 0, 7, 0, 8,\n",
       "       3, 6, 9, 2, 7, 3, 5, 9, 8, 5, 4, 6, 4, 6, 3, 1, 9, 2, 7, 7, 3, 1,\n",
       "       1, 2, 0, 7, 8, 9, 1, 9, 6, 2, 1, 0, 6, 8, 2, 8, 8, 7, 2, 7, 5, 9,\n",
       "       2, 3, 6, 4, 1, 1, 5, 7, 4, 9, 9, 4, 3, 8, 8, 9, 2, 0, 9, 0, 0, 4,\n",
       "       1, 5, 5, 4, 7, 4, 7, 4, 2, 2, 8, 7, 2, 0, 9, 0, 2, 1, 7, 8, 8, 7,\n",
       "       2, 8, 3, 3, 2, 2, 6, 1, 5, 5, 5, 0, 1, 5, 8, 2, 6, 5, 1, 0, 3, 1,\n",
       "       9, 9, 8, 3, 8, 9, 2, 2, 2, 6, 2, 6, 6, 1, 6, 2, 5, 4, 9, 2, 1, 2,\n",
       "       6, 2, 6, 6, 1, 1, 7, 5, 9, 8, 6, 2, 4, 7, 6, 9, 8, 7, 2, 9, 1, 6,\n",
       "       7, 6, 0, 6, 1, 7, 4, 8, 4, 3, 2, 2, 4, 2, 8, 6, 8, 3, 2, 0, 8, 8,\n",
       "       8, 5, 4, 7, 0, 8, 2, 4, 2, 5, 5, 8, 9, 0, 2, 8, 5, 3, 1, 3, 5, 0,\n",
       "       6, 4, 6, 1, 7, 5, 4, 1, 9, 5, 8, 4, 7, 6, 3, 1, 5, 4, 6, 4, 9, 6,\n",
       "       1, 0, 9, 0, 1, 2, 3, 9, 2, 4, 2, 0, 0, 8, 2, 5, 9, 2, 4, 7, 9, 8,\n",
       "       3, 7, 3, 7, 7, 1, 1, 1, 8, 9, 5, 7, 7, 3, 1, 7, 5, 3, 6, 9, 1, 9,\n",
       "       2, 9, 8, 3, 0, 7, 0, 8, 0, 6, 5, 2, 1, 0, 0, 3, 3, 3, 8, 3, 2, 9,\n",
       "       6, 9, 6, 6, 1, 9, 9, 4, 6, 3, 0, 5, 5, 8, 0, 0, 1, 2, 8, 9, 3, 1,\n",
       "       0, 4, 1, 6, 2, 0, 9, 7, 9, 4, 8, 3, 6, 0, 5, 9, 2, 5, 5, 7, 4, 7,\n",
       "       7, 8, 8, 1, 0, 7, 0, 1, 1, 9, 1, 9, 9, 7, 5, 4, 6, 3, 5, 3, 1, 1,\n",
       "       1, 5, 4, 0, 6, 4, 7, 8, 1, 4, 6, 3, 8, 9, 9, 7, 6, 0, 2, 8, 6, 8,\n",
       "       0, 8, 1, 2, 2, 0, 5, 2, 1, 2, 0, 5, 3, 8, 1, 6, 1], dtype=int64)"
      ]
     },
     "execution_count": 190,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_True = np.argmax(y_test, axis=1)\n",
    "y_True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "id": "2382f659",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(413,)"
      ]
     },
     "execution_count": 191,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_True.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 188,
   "id": "a458d04c",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "Confusion_Matrix = confusion_matrix(y_true= y_True, y_pred= np.argmax(predictions, axis=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "id": "eabe05dc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[40,  0,  0,  0,  0,  0,  0,  0,  0,  0],\n",
       "       [ 0, 48,  0,  0,  0,  0,  0,  0,  0,  0],\n",
       "       [ 0,  0, 49,  0,  0,  1,  0,  0,  0,  0],\n",
       "       [ 0,  0,  1, 31,  0,  0,  1,  0,  2,  0],\n",
       "       [ 0,  1,  0,  0, 31,  0,  0,  0,  0,  0],\n",
       "       [ 0,  1,  1,  0,  1, 35,  0,  0,  0,  0],\n",
       "       [ 0,  0,  0,  2,  0,  0, 38,  0,  0,  1],\n",
       "       [ 0,  0,  0,  0,  0,  0,  0, 38,  0,  0],\n",
       "       [ 0,  0,  0,  0,  3,  0,  0,  0, 42,  0],\n",
       "       [ 0,  0,  0,  0,  0,  0,  0,  0,  0, 46]], dtype=int64)"
      ]
     },
     "execution_count": 193,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Confusion_Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 195,
   "id": "092f8b3d",
   "metadata": {},
   "outputs": [],
   "source": [
    "? sns.heatmap"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "id": "62be446f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(10, 10))\n",
    "sns.heatmap(Confusion_Matrix, annot=True, cmap='Blues', linecolor='gray', fmt='.1f', ax = ax, linewidths=0.01)\n",
    "plt.xlabel('Predicted Label')\n",
    "plt.ylabel('True Label')\n",
    "plt.title('Confusion Matrix')\n",
    "# plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "24a4348d",
   "metadata": {},
   "source": [
    "### Classification Report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 198,
   "id": "266fbde2",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import classification_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 199,
   "id": "53f1c365",
   "metadata": {},
   "outputs": [],
   "source": [
    "Classification_Report = classification_report(y_true= y_True, y_pred= np.argmax(y_test, axis=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 200,
   "id": "0c2f8a80",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'              precision    recall  f1-score   support\\n\\n           0       1.00      1.00      1.00        40\\n           1       1.00      1.00      1.00        48\\n           2       1.00      1.00      1.00        50\\n           3       1.00      1.00      1.00        35\\n           4       1.00      1.00      1.00        32\\n           5       1.00      1.00      1.00        38\\n           6       1.00      1.00      1.00        41\\n           7       1.00      1.00      1.00        38\\n           8       1.00      1.00      1.00        45\\n           9       1.00      1.00      1.00        46\\n\\n    accuracy                           1.00       413\\n   macro avg       1.00      1.00      1.00       413\\nweighted avg       1.00      1.00      1.00       413\\n'"
      ]
     },
     "execution_count": 200,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Classification_Report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "25ebbcf2",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}