handwriting  recognition (number).ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "46efa79e-62c4-416a-a2f8-5b9c4c6d25d1",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "import tensorflow.keras.utils as np_utils\n",
    "from keras.models import Sequential, load_model\n",
    "from keras.layers import Dense,Dropout,Flatten,Conv2D,MaxPooling2D\n",
    "from sklearn import metrics\n",
    "import seaborn as sns "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1efade7f",
   "metadata": {},
   "source": [
    "# Load the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "bcc40089-8d37-47a5-a6de-4ae3c8aa0103",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "x_train shape: (60000, 28, 28)\n",
      "y_train shape: (60000,)\n",
      "x_test shape: (10000, 28, 28)\n",
      "y_test shape: (10000,)\n"
     ]
    }
   ],
   "source": [
    "dataset = keras.datasets.mnist.load_data()\n",
    "(x_train, y_train), (x_test, y_test) = dataset\n",
    "print(\"x_train shape:\", x_train.shape)\n",
    "print(\"y_train shape:\", y_train.shape)\n",
    "print(\"x_test shape:\", x_test.shape)\n",
    "print(\"y_test shape:\", y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "87ebefa9",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_image(image):\n",
    "    fig = plt.gcf()\n",
    "    fig.set_size_inches(2, 2)\n",
    "    plt.imshow(image, cmap='binary')\n",
    "    plt.show()\n",
    "    \n",
    "def plot_images_labels_prediction(images,labels,\n",
    "                                  prediction,idx,num=10):\n",
    "    fig = plt.gcf()\n",
    "    fig.set_size_inches(12, 14)\n",
    "    if num>25: num=25 \n",
    "    for i in range(0, num):\n",
    "        ax=plt.subplot(5,5, 1+i)\n",
    "        ax.imshow(images[idx], cmap='binary')\n",
    "        title= \"label=\" +str(labels[idx])\n",
    "        if len(prediction)>0:\n",
    "            title+=\",predict=\"+str(prediction[idx]) \n",
    "            \n",
    "        ax.set_title(title,fontsize=10) \n",
    "        ax.set_xticks([]);ax.set_yticks([])        \n",
    "        idx+=1 \n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5e935f83",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 200x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "correct number: 2\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 200x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "correct number: 1\n"
     ]
    }
   ],
   "source": [
    "plot_image(x_train[5])\n",
    "print(\"correct number:\", y_train[5])\n",
    "\n",
    "plot_image(x_test[5])\n",
    "print(\"correct number:\", y_test[5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "7972d0bf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7oAAAGrCAYAAADuEet6AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/NK7nSAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA2QUlEQVR4nO3deXRV1d038B1ABpHBAUU04nKoiPMsKoJaQa0C1T7WaquAgtYq9HEesFKxVhzqrK1aEUdQaxW1Ty0OIBasqAVBnJc4lNcBUUBEoeS8f7yvPNqzr+YmN7ncnc9nLdaSb3b2+SXuE/LLSX6pyrIsCwAAAJCIZuUuAAAAAEpJowsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJEWjCwAAQFI0ugAAACRFowsAAEBSNLoAAAAkRaMb0bt37/DLX/6yVmsnTZoUqqqqwqefflqva2688cbhyiuvrNceUCruAZoy55+mzPmnqXMPpEOjm4C5c+eGqqqq3J9nnnmm3KVBo3nxxRdDz549Q+vWrUN1dXW45JJLyl0SNLo33ngjtGvXLnTs2LHcpUCj+OKLL8LAgQPDNttsE1q0aBEGDBhQ7pKg0d1zzz1h++23D6uvvnro2rVruPTSS8td0iqhRbkLoHQee+yxsNVWW638+9prr13GaqDxLFq0KPTp0yd8//vfD7///e/DrFmzwuDBg0PHjh3D0KFDy10eNIrly5eHn/zkJ6Fnz55h6tSp5S4HGsWKFStCmzZtwrBhw8Kf/vSncpcDje5//ud/wlFHHRWuueaa0KdPn/Dyyy+HIUOGhDZt2oSTTjqp3OWVlSe63+H2228PO++8c2jXrl3o3LlzOPLII8OHH36YW/f3v/89bLvttqF169Zh9913D7Nnz/7Gy59++unQs2fP0KZNm1BdXR2GDRsWlixZUtJa11577dC5c+eVf1ZbbbWS7k/TVAn3wJ133hmWLVsWbrnllrDVVluFI444IgwbNiz87ne/K8n+NF2VcP6/MmLEiNCtW7dw+OGHl3Rfmq5KOP9t27YNN9xwQxgyZEjo3LlzSfaEr1TCPXD77beHAQMGhBNOOCFssskm4Qc/+EE4++yzw+jRo0OWZSW5RqXS6H6H5cuXh1GjRoWZM2eGBx54IMydOzcMHDgwt+70008Pl19+eZg+fXro1KlTOOSQQ8Ly5ctDCCG8+eab4YADDgiHHXZYePHFF8P48ePD008//a1fZTnwwAPDGmusUfDP15/cfqVfv35h3XXXDXvttVeYMGFCyd4HNG2VcA9MmzYt7L333qFly5Yrs759+4ZXX301fPLJJ6V7Z9DkVML5DyGEJ554Itx7773huuuuK+nbT9NWKecfGkol3ANffvllaN269Tdev02bNuG9994Lb7/9dmneEZUqI6dXr17Z8OHDoy+bPn16FkLIFi9enGVZlj355JNZCCEbN27cyjUff/xx1qZNm2z8+PFZlmXZsccemw0dOvQb+0yZMiVr1qxZtnTp0izLsqxr167ZFVdcsfLl7733Xvb6668X/DN37tyVaz/66KPs8ssvz5555pns2Wefzc4888ysqqoqe/DBB0vx7qAJqrR7YP/998/t/9JLL2UhhGzOnDl1fj/QNFXa+Z8/f35WXV2dTZ48OcuyLBszZkzWoUOH+r4baKIq7fx/3THHHJP179+/jm85/D+Vdg/84Q9/yFZfffXssccey1asWJG9+uqrWbdu3bIQQjZ16tRSvEsqlp/R/Q7PP/98GDlyZJg5c2b45JNPQk1NTQghhHfeeSd079595boePXqs/O+11lorbLHFFuHll18OIYQwc+bM8OKLL4Y777xz5Zosy0JNTU146623wpZbbpm77gYbbFDrGtdZZ51wyimnrPz7LrvsEubNmxcuvfTS0K9fv9q/sRBRCfcANJRKOP9DhgwJRx55ZNh7772Lfvvg21TC+YeGVAn3wJAhQ8Kbb74ZDj744LB8+fLQvn37MHz48DBy5MjQrFnT/uZdje63WLJkSejbt2/o27dvuPPOO0OnTp3CO++8E/r27RuWLVtW630+++yzcPzxx4dhw4blXrbRRhtFX+fAAw8MU6ZMKbhn165dw0svvVTw5bvttluYOHFirWuEmEq5Bzp37hw++OCDb7z8q7/7mS3qqlLO/xNPPBEmTJgQLrvsshDC/34C1aJFi3DjjTeGwYMH17pW+EqlnH9oKJVyD1RVVYXRo0eHiy66KLz//vuhU6dO4fHHHw8hhLDJJpvUus4UaXS/xSuvvBI+/vjjcPHFF4fq6uoQQgjPPfdcdO0zzzyz8rB+8skn4bXXXlv5FZodd9wxzJkzJ2y22Wa1vvbNN98cli5dWvDl3zVoasaMGWH99dev9fUgplLugR49eoRzzz03LF++fGU+ceLEsMUWW4Q111yz1teEr6uU8z9t2rSwYsWKlX9/8MEHw+jRo8PUqVM9GaPOKuX8Q0OptHugefPmKz/m33333aFHjx6hU6dOtb5mijS632KjjTYKLVu2DNdcc0044YQTwuzZs8OoUaOiay+44IKw9tprh/XWWy+ce+65YZ111ln5u9zOPPPMsPvuu4eTTjopHHfccaFt27Zhzpw5YeLEieHaa6+N7lfMJydjx44NLVu2DDvssEMIIYT7778/3HLLLeHmm28u7g2G/1Ap98CRRx4Zfv3rX4djjz02nHnmmWH27NnhqquuCldccUXRbzN8pVLO/39+29tzzz0XmjVrFrbeeuta7wH/qVLOfwghzJkzJyxbtiwsWLAgLF68OMyYMSOEEML2229f1D7wdZVyD8yfPz/cd999oXfv3uGLL74IY8aMCffee2+YPHly0W9zapr2N25/h06dOoVbb7013HvvvaF79+7h4osvXvmtYf/p4osvDsOHDw877bRTeP/998NDDz20cgLstttuGyZPnhxee+210LNnz7DDDjuEX/3qV6FLly4lq3XUqFFhp512Crvttlt48MEHw/jx48OgQYNKtj9NU6XcAx06dAh/+9vfwltvvRV22mmncOqpp4Zf/epXfocu9VIp5x8aQiWd/4MOOijssMMO4aGHHgqTJk0KO+yww8ov/kNdVdI9MHbs2LDzzjuHPffcM7z00kth0qRJYddddy3Z/pWqKsua+C9YAgAAICme6AIAAJAUjS4AAABJ0egCAACQFI0uAAAASdHoAgAAkBSNLgAAAElpUddXrKmpCfPmzQvt2rULVVVVpawJvlOWZWHx4sWhS5cuoVmzxv96jfNPubkHaMqcf5qycp//ENwDlFdt74E6N7rz5s0L1dXVdX11KIl33303bLjhho1+XeefVYV7gKbM+acpK9f5D8E9wKrhu+6BOje67dq1W3mB9u3b13UbqJNFixaF6urqleewsTn/lJt7gKbM+acpK/f5D8E9QHnV9h6oc6P71bcptG/f3gGnbMr17TLOP6sK9wBNmfNPU1bObxl2D7Aq+K57wDAqAAAAkqLRBQAAICkaXQAAAJKi0QUAACApGl0AAACSotEFAAAgKRpdAAAAkqLRBQAAICkaXQAAAJKi0QUAACApGl0AAACSotEFAAAgKRpdAAAAkqLRBQAAICkaXQAAAJKi0QUAACApGl0AAACSotEFAAAgKRpdAAAAktKi3AUAfN3zzz8fza+99tpoPnbs2Fx2zDHHRNeefPLJ0XzHHXesZXUAAFQCT3QBAABIikYXAACApGh0AQAASIpGFwAAgKRodAEAAEiKqcsltmLFimi+cOHCeu9daOrs559/nsteffXV6Nrrrrsump922mnR/O67785lrVu3jq4966yzovn5558fzWnaZsyYEc2///3vR/NFixZF86qqqlx22223Rdc++OCD0XzBggXRHJqKxx9/PJofddRRuWzy5MnRtVtssUVJa4K6uvDCC6P5r371q2ieZVk0nzRpUi7r1atXnesCGpcnugAAACRFowsAAEBSNLoAAAAkRaMLAABAUprkMKp33nknmi9btiyXTZ06Nbr26aefjuaffvppNL/vvvtqV1yJVFdXR/OTTz45mv/5z3+O5u3atctl2223XXStAQ0U8uyzz+ayww47LLq20OC22NCpEEJo3759LmvZsmV07fz586P5tGnTovlOO+1U670praeeeiqaf/zxx7nshz/8YUOXk7zp06dH85133rmRK4Hi3Hrrrbns4osvjq5t3rx5NC80SLTQvztAZfBEFwAAgKRodAEAAEiKRhcAAICkaHQBAABIikYXAACApCQ9dfmf//xnNN93332jeaFpr6u62BTBCy+8MLq2bdu20fyoo46K5l26dMlla665ZnTtFltsUahEEvP5559H8xdeeCGa//SnP81l8+bNK0ktm2++eS4744wzomt//OMfR/M999wzmsfuo3POOaeI6qirSZMmRfPXX389l5m6XHs1NTXR/K233ormsd9SkGVZSWuC+nj77bdz2ZdfflmGSiCEf/zjH9H89ttvz2WFfrvA7Nmzi7rm5Zdfnstin7+HEMKUKVOi+c9+9rNovttuuxVVy6rGE10AAACSotEFAAAgKRpdAAAAkqLRBQAAICkaXQAAAJKS9NTlrl27RvN11lknmjf21OVCk8wKTTV+8skno3nLli1zWaHpaVAKxx9/fDS/6667GrmSEJ5//vlc9tlnn0XX9urVK5oXmvA7a9asOtdF/YwdOzaa77HHHo1cSVr+z//5P9H8xhtvjOaxf0u6detW0pqgNh577LFofvXVV9d6j0Jn9+GHH47m6623Xq33pmkZP358NB8+fHg0/+ijj3JZoQn2vXv3jubz58+P5qeddlo0jyl0zUJ7jxs3rtZ7r4o80QUAACApGl0AAACSotEFAAAgKRpdAAAAkqLRBQAAIClJT11ea621ovmll14azR966KFctsMOO0TXDhs2rKhatt9++1xWaIJg27Zto/ns2bOjeTETB6EYsYnGIRSeUFloml9MoamCBx98cDQvNFWwS5cuuazQfVvsRPNi3h5Kq6amptwlJOm4444rav3mm2/eQJVA3NNPPx3NBw4cGM0XLVpU671PP/30aF7ot3TQtPz73//OZdOnT4+uHTJkSDRfsmRJNI/91ofzzjsvunavvfaK5l9++WU0P/zww3PZo48+Gl1byM4771zU+krhiS4AAABJ0egCAACQFI0uAAAASdHoAgAAkJSkh1EVMmDAgGi+77775rJ27dpF17744ovR/Oabb47msUE6hYZOFbL11ltH8xtvvLGofSBmxowZuez73/9+dG2h4R9VVVXR/KCDDspld999d3TtpEmTovlvfvObaB4brtOpU6fo2u222y6aF6r7kUceyWUvvPBCdO2OO+4Yzfl2hT6WfvDBB41cSdPw6aefFrV+//33b5hCoICxY8dG83nz5tV6j0LDDo8++ui6lEQTcccdd+SyY489tqg9+vTpE83Hjx+fy9q3b1/U3rE9Qihu8FR1dXU0P+aYY4qqpVJ4ogsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJEWjCwAAQFKa5NTlQoqZftahQ4ei9o5NYz7iiCOia5s18/UHGs5rr70WzS+55JJctnDhwujaQlON119//Wgem+a3xhprRNcefPDBReUN6fPPP89ll112WXTtXXfd1dDlJOkvf/lLNF+6dGkjV5KWQlOr586dW9Q+G2ywQQmqgbz58+dH8z/+8Y/RvHnz5tG8Y8eOuWzEiBF1rov0FTofF110US4r9FsZfvGLX0TzCy+8MJoXO2E5ptBvnyjG1VdfHc0LfV5X6XRUAAAAJEWjCwAAQFI0ugAAACRFowsAAEBSNLoAAAAkxdTlOho5cmQ0f/7556P5pEmTctljjz0WXdunT5+6lgUrffnll9H8tNNOi+aPPPJILis0JfC2226L5jvvvHM0T2mC7rvvvlvuEpLy6quvFrV+q622aqBK0lLoPn///fej+RZbbBHN27VrV7KaaLpi074PPfTQkux98skn57J99923JHtT2S644IJoHpuuHEIIrVq1ymV9+/aNrh09enQ0b9OmTS2rC+GLL76I5n/729+i+dtvvx3NsyzLZeedd150bf/+/WtZXRo80QUAACApGl0AAACSotEFAAAgKRpdAAAAkqLRBQAAICmmLtdR27Zto/lNN90UzXfcccdcNmTIkOjaffbZJ5oXmmj7i1/8IpdVVVVF19J0vPDCC9E8Nl25kAcffDCa9+rVq041QX3tsssu5S6hwS1atCia//Wvf43md9xxRy4rNLWzkBEjRkTzjh07FrUPxMTO7qxZs4raY7/99ovmw4cPr1NNpOPTTz+N5tdff300L/Q5cmzC8gMPPFDXsr7hjTfeyGVHHXVUdO1zzz1X1N7/9V//lcvOOOOMovZIlSe6AAAAJEWjCwAAQFI0ugAAACRFowsAAEBSDKMqsU033TSa33rrrbls0KBB0bW33XZbUfmSJUty2dFHHx1du/7660dz0nPKKadE8yzLonnv3r1zWVMZOlXofVLftZTeggULGmzvmTNn5rKampro2scffzyav/fee9F82bJluezOO++Mri10zTZt2kTz3XbbLZe1atUqunb58uXRvNCwQyhGocE9Z511Vq336NmzZzQfO3ZsNO/QoUOt9yZNsY+vIYTw0UcfFbXP1Vdfncs+/PDD6NoxY8ZE80JDPF966aVctnjx4ujaQsOymjWLP5/86U9/mssKDc1tajzRBQAAICkaXQAAAJKi0QUAACApGl0AAACSotEFAAAgKaYuN5If/vCHuWyzzTaLrj311FOj+WOPPRbNzz777Fz29ttvR9eee+650XyDDTaI5qz6Hn744Wg+Y8aMaF5oml+/fv1KVVLFKfQ+ieXbb799A1fTtBSaJFzo/8nxxx+fyy666KKS1BKbulxoyvZqq60WzVdfffVovuWWW+aywYMHR9futNNO0Tw2GT2EENZbb71ctuGGG0bXLl26NJp369YtmkPM3Llzo/mhhx5a77032WSTaB475xBCCC1btozm6667bjQvNEl54403zmWF/i0qVuzz7Pbt20fXzps3L5qvs8460fyQQw6pe2GJ80QXAACApGh0AQAASIpGFwAAgKRodAEAAEiKRhcAAICkmLpcRttss000v+eee6L5Qw89FM0HDhyYy37/+99H177++uvRfOLEidGcVV+hKarLli2L5oWmEP74xz8uWU3l9uWXX0bzkSNHFrXPfvvtl8suvvjiupREAddff30079q1azSfOnVqg9Wy0UYb5bL+/ftH13bv3j2a77777iWtqTZuvPHGXFZoqmihibZQjNGjR0fz5s2b13vvs846q9570LR07Ngxmj/wwAPR/OCDD47mH3/8cS4r9BtSCv3bEPucPIQQ1lprrVx2xBFHRNcWmrpcaD2FeaILAABAUjS6AAAAJEWjCwAAQFI0ugAAACRFowsAAEBSTF1eBRWaHvezn/0smh933HG5bPny5dG1Tz31VDSfNGlSNO/du3c0p3K1bt06mq+//vqNXElpxCYsX3jhhdG1l1xySTSvrq6O5qeeemouW2ONNYqojro688wzy11CxXj88cdrvfZHP/pRA1ZCambMmBHNH3300Xrv3a9fv2i+xRZb1HtvCCGE3XbbLZp/9NFHjVxJ/PPvyZMnR9dWVVVFc1Pzi+eJLgAAAEnR6AIAAJAUjS4AAABJ0egCAACQFMOoyujFF1+M5vfdd180nz59ejQvNHgqpnv37tF87733rvUeVLZCA0BWdYWGosQGTI0fPz66tn///tH8/vvvr3NdUEkGDBhQ7hKoIH369Inmn3zySVH7xIYCjR07tk41QSVaunRpLis0dKpQfsQRR5S0pqbAE10AAACSotEFAAAgKRpdAAAAkqLRBQAAICkaXQAAAJJi6nKJvfrqq9H8mmuuyWWFJr2+//779a6jRYv4/9r1118/mjdr5mselSrLsqLyBx54IJpfddVVpSqpXn73u99F81GjRkXzhQsX5rKf/vSn0bW33XZb3QsDaGLmz58fzZs3b17UPr/4xS9y2RprrFGnmqAS9e3bt9wlNEm6GwAAAJKi0QUAACApGl0AAACSotEFAAAgKRpdAAAAkmLq8ncoNAH5rrvuiubXXnttNJ87d26pSsrZZZddctm5554bXduvX78Gq4PyqKqqKiovdKaHDRuWywYPHhxdu/baa0fzZ555JprffvvtuWzmzJnRte+++24079q1azQ/4IADctmJJ54YXQtN3euvvx7Ne/To0ciVsKoZNGhQLis0vX/FihVF7b3HHnvUqSZIxaOPPlruEpokT3QBAABIikYXAACApGh0AQAASIpGFwAAgKRodAEAAEhKk5y6/MEHH0Tzl156KZeddNJJ0bWvvPJKSWv6ut122y2an3HGGdG8f//+uaxZM1/DIO7f//53NL/uuuty2X333Rdd26FDh2j+2muv1b2w/6/QdM599903ml9wwQX1viY0FTU1NeUugTKbMWNGNJ84cWIuKzS9v1WrVtG80MT79dZbr3bFQaLefPPNcpfQJOmGAAAASIpGFwAAgKRodAEAAEiKRhcAAICkJDGMasGCBdH8+OOPj+aFBjE05A+K77nnnrns1FNPja7t27dvNG/Tpk1JayINPXr0iOa77rprNH/22Wdrvff7778fzQsNdCtknXXWyWVHHHFEdO1VV11V1N5A7U2bNi2aDxw4sHELoWw+/fTTaF7Mx/UuXbpE88svv7wuJUHyevbsmcuyLCtDJU2LJ7oAAAAkRaMLAABAUjS6AAAAJEWjCwAAQFI0ugAAACRllZ26/I9//COaX3LJJbls+vTp0bXvvfdeSWv6utVXXz2aDxs2LJqfe+65uaxt27YlrYmmacMNN4zm999/fzT/wx/+EM1HjRpV71qGDx8ezX/+85/nss0337ze1wMAWNVts802uazQ50GFfgtMobxTp051LyxxnugCAACQFI0uAAAASdHoAgAAkBSNLgAAAEnR6AIAAJCUVXbq8p///Oei8mJ07949mh9yyCG5rHnz5tG1p512WjTv2LFjneuCUlp//fWj+ciRI4vKgVXbgQcemMvuueeeMlRCJejWrVs032OPPXLZlClTGrocaLLOOeecaH7ssccWtf7aa6/NZYV6nabGE10AAACSotEFAAAgKRpdAAAAkqLRBQAAICkaXQAAAJJSlWVZVpdXXLRoUejQoUNYuHBhaN++fanrgm9V7vNX7utDuc9gua9P01bu81fu69O0rQrnb1WoodItWrQomh9++OHRfOLEidH8sMMOy2VjxoyJrm3btm0tq1u11fb8eaILAABAUjS6AAAAJEWjCwAAQFI0ugAAACSlRbkLAAAAaEoKDVG65557ovm5554bza+//vpcNnLkyOja7t271664RHiiCwAAQFI0ugAAACRFowsAAEBSNLoAAAAkRaMLAABAUkxdBgAAWAUUmsZ8zTXXFJXjiS4AAACJ0egCAACQFI0uAAAASdHoAgAAkJQ6D6PKsiyEEMKiRYtKVgzU1lfn7qtz2Nicf8rNPUBT5vzTlJX7/H/92u4ByqG290CdG93FixeHEEKorq6u6xZQb4sXLw4dOnQoy3VDcP4pP/cATZnzT1NWrvP/1bVDcA9QXt91D1RldfxyUE1NTZg3b15o165dqKqqqnOBUBdZloXFixeHLl26hGbNGv878J1/ys09QFPm/NOUlfv8h+AeoLxqew/UudEFAACAVZFhVAAAACRFowsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJEWjCwAAQFI0ugAAACRFowsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJEWjCwAAQFI0ugAAACRFowsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJEWjCwAAQFI0ugAAACRFowsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJEWjCwAAQFI0ugAAACRFowsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJEWjCwAAQFI0ugAAACRFowsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJEWjCwAAQFI0ugAAACRFowsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJEWjG9G7d+/wy1/+slZrJ02aFKqqqsKnn35ar2tuvPHG4corr6zXHlAq7gGaMuefpsz5p6lzD6RDo5uASZMmhf79+4f1118/tG3bNmy//fbhzjvvLHdZ0Gi++OKLMHDgwLDNNtuEFi1ahAEDBpS7JGg0r776athnn33CeuutF1q3bh022WSTMGLEiLB8+fJylwYNzsd/+F9vvPFGaNeuXejYsWO5S1kltCh3AdTf1KlTw7bbbhvOPPPMsN5664WHH344HH300aFDhw7h4IMPLnd50OBWrFgR2rRpE4YNGxb+9Kc/lbscaFSrrbZaOProo8OOO+4YOnbsGGbOnBmGDBkSampqwkUXXVTu8qBB+fgP/8/y5cvDT37yk9CzZ88wderUcpezSvBE9zvcfvvtYeeddw7t2rULnTt3DkceeWT48MMPc+v+/ve/h2233Ta0bt067L777mH27NnfePnTTz8devbsGdq0aROqq6vDsGHDwpIlS0pS4znnnBNGjRoV9thjj7DpppuG4cOHhwMOOCDcf//9Jdmfpq0S7oG2bduGG264IQwZMiR07ty5JHtCCJVx/jfZZJMwaNCgsN1224WuXbuGfv36haOOOipMmTKlJPvTdFXC+ffxn4ZUCffAV0aMGBG6desWDj/88JLuW8k0ut9h+fLlYdSoUWHmzJnhgQceCHPnzg0DBw7MrTv99NPD5ZdfHqZPnx46deoUDjnkkJXfNvbmm2+GAw44IBx22GHhxRdfDOPHjw9PP/10OOmkkwpe98ADDwxrrLFGwT9bbbXVt9a9cOHCsNZaa9XrbYcQKvcegFKoxPP/xhtvhL/+9a+hV69e9X77adoq8fxDKVXKPfDEE0+Ee++9N1x33XUlffsrXkZOr169suHDh0dfNn369CyEkC1evDjLsix78sknsxBCNm7cuJVrPv7446xNmzbZ+PHjsyzLsmOPPTYbOnToN/aZMmVK1qxZs2zp0qVZlmVZ165dsyuuuGLly997773s9ddfL/hn7ty5BesfP3581rJly2z27Nl1efOhou+BY445Juvfv38d33Ko3PPfo0ePrFWrVlkIIRs6dGi2YsWK+rwbaKIq9fxnmY//lEal3QPz58/Pqqurs8mTJ2dZlmVjxozJOnToUN93QxL8jO53eP7558PIkSPDzJkzwyeffBJqampCCCG88847oXv37ivX9ejRY+V/r7XWWmGLLbYIL7/8cgghhJkzZ4YXX3zxGwOisiwLNTU14a233gpbbrll7robbLBBnep98sknw6BBg8JNN93kK56URKXdA1BKlXT+x48fHxYvXhxmzpwZTj/99HDZZZeFM844o+h94CuVdP6hIVTCPTBkyJBw5JFHhr333rvoty91Gt1vsWTJktC3b9/Qt2/fcOedd4ZOnTqFd955J/Tt2zcsW7as1vt89tln4fjjjw/Dhg3LvWyjjTaKvs6BBx74rT9f1bVr1/DSSy99I5s8eXI45JBDwhVXXBGOPvroWtcHhVTaPQClVGnnv7q6OoQQQvfu3cOKFSvC0KFDw6mnnhqaN29e61rhK5V2/qHUKuUeeOKJJ8KECRPCZZddFkL43ya6RYsW4cYbbwyDBw+uda2p0eh+i1deeSV8/PHH4eKLL175CcRzzz0XXfvMM8+sPKyffPJJeO2111Z+hWbHHXcMc+bMCZtttlmtr33zzTeHpUuXFnz5aqut9o2/T5o0KRx88MFh9OjRYejQobW+DnybSroHoNQq+fzX1NSE5cuXh5qaGo0udVLJ5x9KoVLugWnTpoUVK1as/PuDDz4YRo8eHaZOndrkvztCo/stNtpoo9CyZctwzTXXhBNOOCHMnj07jBo1Krr2ggsuCGuvvXZYb731wrnnnhvWWWedlb/L7cwzzwy77757OOmkk8Jxxx0X2rZtG+bMmRMmTpwYrr322uh+xRzMJ598Mhx88MFh+PDh4bDDDgvvv/9+CCGEli1bGkhFvVTKPRBCCHPmzAnLli0LCxYsCIsXLw4zZswIIYSw/fbbF7UPfKVSzv+dd94ZVltttbDNNtuEVq1aheeeey6cffbZ4cc//rGGgDqrlPMfgo//NIxKuQf+81ufn3vuudCsWbOw9dZb13qPZJXzB4RXVV//IfS77ror23jjjbNWrVplPXr0yCZMmJCFELJ//vOfWZb97w+hP/TQQ9lWW22VtWzZMtt1112zmTNnfmPPZ599Ntt///2zNdZYI2vbtm227bbbZr/5zW9Wvvw/fwi9GMccc0wWQsj96dWrV532g0q7B756/dh9AMWqtPM/bty4bMcdd1y5d/fu3bOLLrpo5ZATKEalnf+vXt/Hf0qlEu+BrzOM6n9VZVmWNVpXDQAAAA3M79EFAAAgKRpdAAAAkqLRBQAAICkaXQAAAJKi0QUAACApGl0AAACS0qKur1hTUxPmzZsX2rVrF6qqqkpZE3ynLMvC4sWLQ5cuXUKzZo3/9Rrnn3JzD9CUOf80ZeU+/yG4Byiv2t4DdW50582bF6qrq+v66lAS7777bthwww0b/brOP6sK9wBNmfNPU1au8x+Ce4BVw3fdA3VudNu1a7fyAu3bt6/rNlAnixYtCtXV1SvPYWNz/ik39wBNmfNPU1bu8x+Ce4Dyqu09UOdG96tvU2jfvr0DTtmU69tlnH9WFe4BmjLnn6asnN8y7B5gVfBd94BhVAAAACRFowsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJEWjCwAAQFI0ugAAACRFowsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJEWjCwAAQFI0ugAAACRFowsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJEWjCwAAQFI0ugAAACSlRbkLANIxfPjwaH711Vfnsq233jq69uGHH47mXbt2rXthAADU27777lvU+ieeeKKBKvlunugCAACQFI0uAAAASdHoAgAAkBSNLgAAAEnR6AIAAJAUU5dXQYsXL47mn332WTR/5JFHctmHH34YXXvqqadG81atWtWyOghh7ty50fz222+P5lVVVblszpw50bWvvPJKNDd1mVXJa6+9Fs2XLVsWzadMmZLLTjzxxOja2P3S0AYMGBDNx40bl8tatmzZwNVQqZYvXx7Np06dmsvOPvvsWq8FyuO///u/c9m0adOia48++uiGLqdonugCAACQFI0uAAAASdHoAgAAkBSNLgAAAEkxjKqRvPXWW7nskksuia4t9EPes2bNqncd77//fjS/+uqr6703TUenTp2iea9evaL5gw8+2JDlQEnMnj07l40dOza69t57743mNTU10fxf//pXLis0dKocw6gK3aMnnHBCLrvyyiuja9u3b1/KkqhACxcujOa9e/fOZZ07d46uLfR5SqH1QP2dddZZ0fz3v/99LltttdWia/fbb7+S1lQKnugCAACQFI0uAAAASdHoAgAAkBSNLgAAAEnR6AIAAJAUU5fr6JVXXonmhaZR3nHHHbls6dKl0bVZlkXzjTbaKJq3a9cul82ZMye69p577onmJ554YjTv1q1bNKdpa9u2bTTv2rVrI1cCpXPOOefkskceeaQMlaw6YlOnBw8eHF271157NXQ5JKTQdGVTl6HxPfPMM9F82bJluazQx/rDDz+8pDWVgie6AAAAJEWjCwAAQFI0ugAAACRFowsAAEBSNLoAAAAkxdTlr1m4cGEuO/PMM6Nrx48fH80XLVpU7zq+973vRfNHH300mscmohWalvzRRx9F8/nz59eyOgjh008/jeYzZ85s3EKghPbff/9cVuzU5XXXXTeaH3vssbmspqYmurZZs+K+Bj116tRcNnny5KL2AKD+nnrqqWj+m9/8Jpfdfffd0bVrrbVWSWuqzTVnzZoVzTfbbLNcdtlll5W0pobkiS4AAABJ0egCAACQFI0uAAAASdHoAgAAkBSNLgAAAEkxdflr/vznP+eym266qcGuF5tkFkIIEydOjObV1dXR/PXXXy9ZTVAbn3/+eTR/++2367339OnTo3mhSeJdu3at9zUhhBB+/vOf57IBAwYUtcdqq60WzTt37lyXkmolNu1/6623jq7917/+VdTesbd/l112KWoPKMbSpUvLXQLU2dChQ6P5a6+9lsvmzJkTXbvXXnuVtKavi01/DiGEBQsWRPObb745l2233XYlrakheaILAABAUjS6AAAAJEWjCwAAQFI0ugAAACTFMKqvueeee+q9x8YbbxzNd91111w2evTo6NpCQ6cKeeWVV4paD/XVpUuXaD5o0KBofv7559d670JrO3bsGM1POumkWu8N36ZFi/w/icV+PC6HRx99NJd98sknJdk79va3atWqJHtDzPPPPx/Ne/To0ciVQPHatGkTzauqqnLZF1980WB1zJgxI5q/88470TxWXwgNW2Nj8EQXAACApGh0AQAASIpGFwAAgKRodAEAAEiKRhcAAICkmLr8NTfffHMuu/HGG6Nr+/TpE80322yzaL7uuuvWvbDv8MEHHzTY3lCM8847L5oXM3UZiBs3blw0j/079fnnn5fkmhdccEFJ9qFpiE0uDyE+Nf/TTz+Nrn3zzTdLWBE0jEKf78yePTuab7nllrlsu+22K0ktS5YsyWWFfrNLbG0IIey+++7R/Ec/+lHdC1sFeKILAABAUjS6AAAAJEWjCwAAQFI0ugAAACRFowsAAEBSTF3+mi5duuSykSNHNn4hRZo6dWq5S4BvlWVZuUuAVc4dd9wRzS+++OJoXmga7bJly+pdy/bbbx/NV1tttXrvTdMRm64cQgg9e/bMZQ899FADVwP19+6770bzm266KZoXmjx+3XXX5bJOnTrVvbCvOeWUU3LZPffcE127wQYbRPNUewlPdAEAAEiKRhcAAICkaHQBAABIikYXAACApGh0AQAASIqpy43k6quvzmVLliyJri00obaqqiqaz549u9Z17LnnntG8R48etd4DihU7u4XOM5TL3Llzc9ntt98eXfvYY4/V+3pTpkyJ5qW4N9q3bx/NR48eHc0POuigaN6mTZt61wJQCWbNmpXLDj300Ojajz76KJoPGzYsmvfq1avuhf1/l112WTS/9dZba73HueeeW+86KoknugAAACRFowsAAEBSNLoAAAAkRaMLAABAUgyj+g6ff/55NH/ppZei+QUXXBDNH3nkkVpfs9hhVDFdunSJ5mPGjInmzZs3r/XeAJUsNnAkhBD69euXy955552GLqdB7L333tF86NChjVwJFOfjjz8udwkk4t///nc0v+OOO6L54MGDc1mxn5NPmzYtml900UW57NRTT42uXbBgQTS/9957o3msxmOOOSa69vjjj4/mqfJEFwAAgKRodAEAAEiKRhcAAICkaHQBAABIikYXAACApDTJqcvLly+P5v/85z9z2WGHHRZdO2/evGi++uqrR/PYFOQ99tgjuvavf/1rNF+yZEk0j1mxYkU0v//++6P58OHDo3nLli1rfU2A1BSauLmq7/3QQw9F87/85S/R/KCDDmqwWqAYEyZMKHcJJGLcuHHR/Nhjj43mxfx2k8033zyaT58+vdZ5obP+r3/9K5oX6j3WXXfdXHbLLbdE1zY1nugCAACQFI0uAAAASdHoAgAAkBSNLgAAAEnR6AIAAJCUpKcuL1u2LJoXmmr8wx/+sNZ7jxw5Mprvs88+0XyvvfbKZQsWLIiu3XfffaP5rFmzaldcCOHDDz+M5meddVY032ijjaL5gAEDclmrVq1qXQeEUJrpsk899VQ0P+mkk+q9N03LNttsE80nTZqUy26//fbo2gMOOCCat27dus51fZc//vGP0fzqq69usGtCKcQ+Nyo0GRyKNX78+Gg+aNCgaF7oN4p07Ngxl911113RtWuuuWY0P+WUU6L55MmTc1mhCc2FPmcqNBV6/vz5uay6ujq6NvbvXAghbLrpptG80nmiCwAAQFI0ugAAACRFowsAAEBSNLoAAAAkRaMLAABAUpKYurx8+fJofv7550fzSy65pNZ7H3jggdH85JNPjuaxiW0hhPDRRx/lsoMOOii69sUXX4zmhaYdn3HGGbms0ITmBx98MJofeeSR0Xz//fev1fVCKDyBrpAddtihqPVUrtikwELTAwv505/+FM3nzJmTy7p3717U3hBCCF27ds1lI0aMKEMlcYWm/Zu6zKqu0G92iCn0GzPefvvtaB67b2la/vCHP0TzQpOHC31cHzx4cL1rufbaa6P50KFDc9m0adPqfb0QQqipqcllhX4LTKrTlQvxRBcAAICkaHQBAABIikYXAACApGh0AQAASErFDaNasWJFLjvvvPOiay+99NJovsYaa0Tz3/72t7nsJz/5SXRtoaFT06dPj+ax4VUvvPBCdO33vve9aH7DDTdE89gPnC9atCi6durUqdH8zjvvjOYTJkzIZbEBVd+m0BCKt956q6h9qFwnnHBCLis0PKJYN954Yy678sorS7I3rEoeffTRcpcAddKiRe0/3cyyLJp/+eWXpSqHxPTv3z+aH3roodG80JCqUpg/f340f+mll2q9x7hx46L51ltvXes9Ntxww1qvTZknugAAACRFowsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJKXipi7HJqwWmq7ctm3baF5o2mufPn1y2TPPPBNdO2bMmGj+l7/8JZovXbo0l51//vnRtYMGDYrmxUyJa9++fTQ/4IADisrvvvvuXFZoQnMhV1xxRVHrSc+WW25Z7hJI2PLly6N5oSnF++23XzRv06ZNyWqqj1tuuSWa//KXv2zcQqBEYlNxu3XrFl37yiuvRPNC0/Svv/76OtdFGoYPH97o11y4cGE0v+eee2q9frPNNouuPfzww+teGN/giS4AAABJ0egCAACQFI0uAAAASdHoAgAAkBSNLgAAAEmpyrIsq8srLlq0KHTo0CEsXLiw4ITfhrD++uvnsg8//DC6tlWrVtG80KS/zz//PJe9/vrrRVRX2K9//etcdvbZZ0fXNm/evCTXTFm5zt+qcv0UfO9734vmb7zxRlH7xD6EFdpj0003LWrvVVm5z2C5rj9lypRcdtFFF0XX/u1vf4vmc+fOjebFTLYv1oIFC3JZoSn9J598cjRftGhRra+3+uqrR/MJEyZE83322afWe68Kmur5T0mhKeKFfqvFBx98EM1bt25dqpIqxqpw/laFGsrpt7/9bTQfMWJENF933XVz2fTp06NrN9xww7oX1kTU9vx5ogsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJEWjCwAAQFJalLuAYnXu3DmXFZq6/OWXX0bzmTNn1vp6P/jBD6L53nvvHc0HDBgQzTfeeONcZroyTdlWW20Vzd98881GroRKEptIPGvWrKL2uOSSS6J5u3bt6lRTbUycODGXPf/889G1VVVVRe3du3fvXHbiiSdG11badGWankLnv2XLlo1cCYTw9ttvR/ObbropmjdrFn+GOHTo0FxmunLD80QXAACApGh0AQAASIpGFwAAgKRodAEAAEhKxQ2jeuqpp3LZAw88EF37wgsvRPN11103mg8ePDiXrbnmmtG1hiJA/cQGM4QQwoQJExq5Epqa66+/vtwlfKtC/0b169cvml911VW5rHXr1iWtCRrLwoULo3mhz/UOPfTQBqyGpm7//feP5oWGVP3sZz+L5r/+9a9LVhO154kuAAAASdHoAgAAkBSNLgAAAEnR6AIAAJAUjS4AAABJqbipy+3atctlhSacFcqB8uvevXtR+Zw5cxqyHCrEmDFjctk111wTXTt27NiGLidns802i+arr756LuvZs2d07ZAhQ6L5NttsU/fCYBUzfvz4aF5oYnihfxugIQ0cODCan3feedG80HR8ysMTXQAAAJKi0QUAACApGl0AAACSotEFAAAgKRpdAAAAklJxU5eBNHTt2jWaz5o1q5EroZLssMMOueyGG26Irt1tt92i+YgRI6L5ggULctmAAQOia/v06RPN+/fvH807d+4czaGp6tWrVzR/+eWXo3mbNm0ashyIOuecc4rKWbV4ogsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJEWjCwAAQFJMXQagorVq1SqaH3/88UXlQOMZN25cuUsAEueJLgAAAEnR6AIAAJAUjS4AAABJ0egCAACQFI0uAAAASdHoAgAAkBSNLgAAAEnR6AIAAJAUjS4AAABJ0egCAACQFI0uAAAASdHoAgAAkBSNLgAAAEnR6AIAAJAUjS4AAABJaVHXV8yyLIQQwqJFi0pWDNTWV+fuq3PY2Jx/ys09QFPm/NOUlfv8f/3a7gHKobb3QJ0b3cWLF4cQQqiurq7rFlBvixcvDh06dCjLdUNw/ik/9wBNmfNPU1au8//VtUNwD1Be33UPVGV1/HJQTU1NmDdvXmjXrl2oqqqqc4FQF1mWhcWLF4cuXbqEZs0a/zvwnX/KzT1AU+b805SV+/yH4B6gvGp7D9S50QUAAIBVkWFUAAAAJEWjCwAAQFI0ugAAACRFowsAAEBSNLoAAAAkRaMLAABAUjS6AAAAJEWjCwAAQFI0ugAAACRFowsAAEBSNLoAAAAkRaMLAABAUv4vZuMdkAY88B4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1200x1400 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x1400 with 10 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_images_labels_prediction(x_train, y_train, [], 0)\n",
    "plot_images_labels_prediction(x_test, y_test, [], 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "25c63681",
   "metadata": {},
   "source": [
    "# Preprocess dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "341d7ce6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 多加一個顏色的維度 \n",
    "x_train4D = x_train.reshape(x_train.shape[0],28,28,1).astype('float32')\n",
    "x_test4D = x_test.reshape(x_test.shape[0],28,28,1).astype('float32')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "d1a55e2b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 將數值縮小到0~1\n",
    "x_train4D_normalize = x_train4D / 255\n",
    "x_test4D_normalize = x_test4D / 255"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ffb00c8a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(60000, 10)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., ..., 0., 0., 0.],\n",
       "       [1., 0., 0., ..., 0., 0., 0.],\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., 1., 0.]], dtype=float32)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Performing Onehot encoding\n",
    "y_TrainOneHot = np_utils.to_categorical(y_train)\n",
    "y_TestOneHot = np_utils.to_categorical(y_test)\n",
    "print(y_TrainOneHot.shape)\n",
    "y_TrainOneHot"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "551295cb",
   "metadata": {},
   "source": [
    "# Build CNN Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "45005757",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv2d (Conv2D)             (None, 28, 28, 16)        416       \n",
      "                                                                 \n",
      " max_pooling2d (MaxPooling2D  (None, 14, 14, 16)       0         \n",
      " )                                                               \n",
      "                                                                 \n",
      " conv2d_1 (Conv2D)           (None, 14, 14, 36)        14436     \n",
      "                                                                 \n",
      " max_pooling2d_1 (MaxPooling  (None, 7, 7, 36)         0         \n",
      " 2D)                                                             \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 7, 7, 36)          0         \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 1764)              0         \n",
      "                                                                 \n",
      " dense (Dense)               (None, 128)               225920    \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, 128)               0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 10)                1290      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 242,062\n",
      "Trainable params: 242,062\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "model.add(Conv2D(filters=16,\n",
    "                 kernel_size=(5,5),\n",
    "                 padding='same',\n",
    "                 input_shape=(28,28,1), \n",
    "                 activation='relu'))\n",
    "\n",
    "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "model.add(Conv2D(filters=36,\n",
    "                 kernel_size=(5,5),\n",
    "                 padding='same',\n",
    "                 activation='relu'))\n",
    "model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "model.add(Dropout(0.25)) # avoid overfitting problem\n",
    "model.add(Flatten())\n",
    "model.add(Dense(128, activation='relu')) # fully connected layer\n",
    "model.add(Dropout(0.5))\n",
    "model.add(Dense(10, activation='softmax')) # There 10 outputs\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ed4ef49f",
   "metadata": {},
   "source": [
    "# Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "6368ad71",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/20\n",
      "180/180 - 19s - loss: 0.4691 - accuracy: 0.8551 - val_loss: 0.0848 - val_accuracy: 0.9743 - 19s/epoch - 105ms/step\n",
      "Epoch 2/20\n",
      "180/180 - 18s - loss: 0.1354 - accuracy: 0.9594 - val_loss: 0.0611 - val_accuracy: 0.9818 - 18s/epoch - 100ms/step\n",
      "Epoch 3/20\n",
      "180/180 - 19s - loss: 0.0971 - accuracy: 0.9707 - val_loss: 0.0441 - val_accuracy: 0.9873 - 19s/epoch - 105ms/step\n",
      "Epoch 4/20\n",
      "180/180 - 21s - loss: 0.0799 - accuracy: 0.9756 - val_loss: 0.0416 - val_accuracy: 0.9890 - 21s/epoch - 115ms/step\n",
      "Epoch 5/20\n",
      "180/180 - 18s - loss: 0.0695 - accuracy: 0.9792 - val_loss: 0.0393 - val_accuracy: 0.9907 - 18s/epoch - 101ms/step\n",
      "Epoch 6/20\n",
      "180/180 - 19s - loss: 0.0618 - accuracy: 0.9814 - val_loss: 0.0334 - val_accuracy: 0.9917 - 19s/epoch - 103ms/step\n",
      "Epoch 7/20\n",
      "180/180 - 16s - loss: 0.0522 - accuracy: 0.9842 - val_loss: 0.0344 - val_accuracy: 0.9897 - 16s/epoch - 92ms/step\n",
      "Epoch 8/20\n",
      "180/180 - 17s - loss: 0.0505 - accuracy: 0.9844 - val_loss: 0.0316 - val_accuracy: 0.9915 - 17s/epoch - 92ms/step\n",
      "Epoch 9/20\n",
      "180/180 - 17s - loss: 0.0453 - accuracy: 0.9860 - val_loss: 0.0300 - val_accuracy: 0.9915 - 17s/epoch - 93ms/step\n",
      "Epoch 10/20\n",
      "180/180 - 17s - loss: 0.0411 - accuracy: 0.9878 - val_loss: 0.0294 - val_accuracy: 0.9923 - 17s/epoch - 93ms/step\n",
      "Epoch 11/20\n",
      "180/180 - 17s - loss: 0.0369 - accuracy: 0.9883 - val_loss: 0.0287 - val_accuracy: 0.9925 - 17s/epoch - 93ms/step\n",
      "Epoch 12/20\n",
      "180/180 - 17s - loss: 0.0346 - accuracy: 0.9891 - val_loss: 0.0295 - val_accuracy: 0.9920 - 17s/epoch - 95ms/step\n",
      "Epoch 13/20\n",
      "180/180 - 17s - loss: 0.0340 - accuracy: 0.9892 - val_loss: 0.0307 - val_accuracy: 0.9930 - 17s/epoch - 93ms/step\n",
      "Epoch 14/20\n",
      "180/180 - 17s - loss: 0.0303 - accuracy: 0.9904 - val_loss: 0.0297 - val_accuracy: 0.9920 - 17s/epoch - 93ms/step\n",
      "Epoch 15/20\n",
      "180/180 - 17s - loss: 0.0291 - accuracy: 0.9906 - val_loss: 0.0267 - val_accuracy: 0.9932 - 17s/epoch - 94ms/step\n",
      "Epoch 16/20\n",
      "180/180 - 17s - loss: 0.0275 - accuracy: 0.9906 - val_loss: 0.0286 - val_accuracy: 0.9928 - 17s/epoch - 96ms/step\n",
      "Epoch 17/20\n",
      "180/180 - 17s - loss: 0.0267 - accuracy: 0.9913 - val_loss: 0.0308 - val_accuracy: 0.9930 - 17s/epoch - 96ms/step\n",
      "Epoch 18/20\n",
      "180/180 - 17s - loss: 0.0246 - accuracy: 0.9921 - val_loss: 0.0297 - val_accuracy: 0.9933 - 17s/epoch - 95ms/step\n",
      "Epoch 19/20\n",
      "180/180 - 17s - loss: 0.0241 - accuracy: 0.9924 - val_loss: 0.0263 - val_accuracy: 0.9942 - 17s/epoch - 95ms/step\n",
      "Epoch 20/20\n",
      "180/180 - 18s - loss: 0.0227 - accuracy: 0.9924 - val_loss: 0.0305 - val_accuracy: 0.9923 - 18s/epoch - 97ms/step\n"
     ]
    }
   ],
   "source": [
    "model.compile(loss='categorical_crossentropy', optimizer='adam',metrics=['accuracy']) \n",
    "\n",
    "train_history=model.fit(x=x_train4D_normalize, \n",
    "                        y=y_TrainOneHot,validation_split=0.1, \n",
    "                        epochs=20, batch_size=300,verbose=2)\n",
    "model.save('my_model.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "b4e77694",
   "metadata": {},
   "outputs": [],
   "source": [
    "def show_train_history(train_acc,test_acc):\n",
    "    plt.plot(train_history.history[train_acc])\n",
    "    plt.plot(train_history.history[test_acc])\n",
    "    plt.title('Train History')\n",
    "    plt.ylabel('Accuracy')\n",
    "    plt.xlabel('Epoch')\n",
    "    plt.legend(['train', 'test'], loc='upper left')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "e2467708",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "show_train_history('accuracy','val_accuracy')\n",
    "show_train_history('loss','val_loss')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "5cb19765",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "313/313 [==============================] - 2s 6ms/step - loss: 0.0210 - accuracy: 0.9928\n",
      "accuracy: 0.9927999973297119\n"
     ]
    }
   ],
   "source": [
    "model = load_model('my_model.h5')\n",
    "loss, accuracy = model.evaluate(x_test4D_normalize, y_TestOneHot)\n",
    "print(\"accuracy:\", accuracy)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "effdb2d2",
   "metadata": {},
   "source": [
    "# Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "b0b4a54d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.99      1.00      0.99       980\n",
      "           1       1.00      1.00      1.00      1135\n",
      "           2       0.99      1.00      1.00      1032\n",
      "           3       0.99      1.00      0.99      1010\n",
      "           4       1.00      0.99      0.99       982\n",
      "           5       0.99      0.99      0.99       892\n",
      "           6       1.00      0.99      0.99       958\n",
      "           7       1.00      0.99      0.99      1028\n",
      "           8       0.99      0.99      0.99       974\n",
      "           9       0.99      0.99      0.99      1009\n",
      "\n",
      "    accuracy                           0.99     10000\n",
      "   macro avg       0.99      0.99      0.99     10000\n",
      "weighted avg       0.99      0.99      0.99     10000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "prediction = model(x_test4D_normalize)\n",
    "prediction = np.argmax(prediction, 1)\n",
    "\n",
    "actual = np.argmax(y_TestOneHot, 1)\n",
    "\n",
    "cm = metrics.confusion_matrix(actual, prediction)\n",
    "cm_display = metrics.ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=[i for i in range(10)])\n",
    "cm_display.plot(cmap = 'Blues')\n",
    "plt.show()\n",
    "print(metrics.classification_report(actual, prediction))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8e30ff01",
   "metadata": {},
   "source": [
    "# My Own Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "ba220699",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction is wrong. The correct answer is 8. The prediction is 3\n"
     ]
    }
   ],
   "source": [
    "import cv2\n",
    "\n",
    "path = \"C:/Users/Microsoft/Desktop/handwriting recognition/8.jpg\"\n",
    "img = cv2.imread(path)\n",
    "cv2.imshow('image', img)\n",
    "cv2.waitKey(0)\n",
    "cv2.destroyAllWindows()\n",
    "#print(img.shape)\n",
    "image = cv2.resize(img, (28, 28), interpolation=cv2.INTER_AREA)\n",
    "image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)\n",
    "image = image.reshape(1,28,28,1).astype('float32')\n",
    "#print(image.shape)\n",
    "\n",
    "y_predict = model([image])\n",
    "if(np.argmax(y_predict, 1) == 8):\n",
    "    print(\"Prediction is right.\")\n",
    "else:\n",
    "    print(f\"Prediction is wrong. The correct answer is 8. The prediction is {np.argmax(y_predict, 1)[0]}\")"
   ]
  }
 ],
 "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": "3.7.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}