NN from scratch - mnist digits.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "2ca60b6f",
   "metadata": {},
   "outputs": [],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np \n",
    "from NN import Layer_Dense, Train_Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "df472a6d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1, 0, 0, ..., 0, 0, 0],\n",
       "       [0, 0, 0, ..., 0, 0, 0],\n",
       "       [1, 0, 0, ..., 0, 0, 0],\n",
       "       ...,\n",
       "       [7, 0, 0, ..., 0, 0, 0],\n",
       "       [6, 0, 0, ..., 0, 0, 0],\n",
       "       [9, 0, 0, ..., 0, 0, 0]])"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# eash row is an image\n",
    "# 42000 images each having 784 pixels (featres)\n",
    "# first column is the number vale, the rest 784 is the image pixels\n",
    "# (42000, 785)\n",
    "data = pd.read_csv('train.csv')\n",
    "data = np.array(data)\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "69f9d18b",
   "metadata": {},
   "outputs": [],
   "source": [
    "np.random.shuffle(data) # shuffle before splitting into test and training sets\n",
    "\n",
    "\n",
    "data_test = data[0:1000,:].T #1000 image for testing, each col is an image \n",
    "m, n = data.shape\n",
    "\n",
    "Y_test = data_test[0,:] \n",
    "X_test = data_test[1:n,:]\n",
    "X_test = X_test / 255."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b3b51e80",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0., ..., 0., 0., 0.],\n",
       "       [0., 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., 0., 0.]])"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f2fa31bb",
   "metadata": {},
   "outputs": [],
   "source": [
    "data_train = data[1000:m].T\n",
    "Y_train = data_train[0,:]\n",
    "X_train = data_train[1:n,:]\n",
    "X_train = X_train / 255."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "67c368cd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch:  0\n",
      "[1 1 1 ... 1 1 9] [3 2 7 ... 1 7 7]\n",
      "acc 0.1075609756097561\n",
      "----------------------------------------------\n",
      "epoch:  10\n",
      "[1 1 4 ... 1 1 9] [3 2 7 ... 1 7 7]\n",
      "acc 0.1536829268292683\n",
      "----------------------------------------------\n",
      "epoch:  20\n",
      "[1 1 4 ... 1 1 9] [3 2 7 ... 1 7 7]\n",
      "acc 0.193\n",
      "----------------------------------------------\n",
      "epoch:  30\n",
      "[1 1 4 ... 1 1 9] [3 2 7 ... 1 7 7]\n",
      "acc 0.24582926829268292\n",
      "----------------------------------------------\n",
      "epoch:  40\n",
      "[1 1 4 ... 1 1 4] [3 2 7 ... 1 7 7]\n",
      "acc 0.2955121951219512\n",
      "----------------------------------------------\n",
      "epoch:  50\n",
      "[1 1 7 ... 1 1 4] [3 2 7 ... 1 7 7]\n",
      "acc 0.36665853658536585\n",
      "----------------------------------------------\n",
      "epoch:  60\n",
      "[1 2 7 ... 1 1 4] [3 2 7 ... 1 7 7]\n",
      "acc 0.4316341463414634\n",
      "----------------------------------------------\n",
      "epoch:  70\n",
      "[1 2 7 ... 1 1 4] [3 2 7 ... 1 7 7]\n",
      "acc 0.5439756097560976\n",
      "----------------------------------------------\n",
      "epoch:  80\n",
      "[1 2 7 ... 1 1 4] [3 2 7 ... 1 7 7]\n",
      "acc 0.6274634146341463\n",
      "----------------------------------------------\n",
      "epoch:  90\n",
      "[3 2 7 ... 1 1 4] [3 2 7 ... 1 7 7]\n",
      "acc 0.6748048780487805\n",
      "----------------------------------------------\n",
      "epoch:  100\n",
      "[3 2 7 ... 1 1 4] [3 2 7 ... 1 7 7]\n",
      "acc 0.7055121951219512\n",
      "----------------------------------------------\n",
      "epoch:  110\n",
      "[3 2 7 ... 1 1 4] [3 2 7 ... 1 7 7]\n",
      "acc 0.7288780487804878\n",
      "----------------------------------------------\n",
      "epoch:  120\n",
      "[3 2 7 ... 1 1 4] [3 2 7 ... 1 7 7]\n",
      "acc 0.7466829268292683\n",
      "----------------------------------------------\n",
      "epoch:  130\n",
      "[3 2 7 ... 1 1 9] [3 2 7 ... 1 7 7]\n",
      "acc 0.7626829268292683\n",
      "----------------------------------------------\n",
      "epoch:  140\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.774390243902439\n",
      "----------------------------------------------\n",
      "epoch:  150\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.7849512195121952\n",
      "----------------------------------------------\n",
      "epoch:  160\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.7946585365853659\n",
      "----------------------------------------------\n",
      "epoch:  170\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.8030731707317074\n",
      "----------------------------------------------\n",
      "epoch:  180\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.8110487804878049\n",
      "----------------------------------------------\n",
      "epoch:  190\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.8189756097560975\n",
      "----------------------------------------------\n",
      "epoch:  200\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.8250243902439024\n",
      "----------------------------------------------\n",
      "epoch:  210\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.8305853658536585\n",
      "----------------------------------------------\n",
      "epoch:  220\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.8355365853658536\n",
      "----------------------------------------------\n",
      "epoch:  230\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.8396585365853658\n",
      "----------------------------------------------\n",
      "epoch:  240\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.8441951219512195\n",
      "----------------------------------------------\n",
      "epoch:  250\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.8477073170731707\n",
      "----------------------------------------------\n",
      "epoch:  260\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.8508536585365853\n",
      "----------------------------------------------\n",
      "epoch:  270\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.8536829268292683\n",
      "----------------------------------------------\n",
      "epoch:  280\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.8566829268292683\n",
      "----------------------------------------------\n",
      "epoch:  290\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.859\n",
      "----------------------------------------------\n",
      "epoch:  300\n",
      "[3 2 7 ... 1 1 7] [3 2 7 ... 1 7 7]\n",
      "acc 0.8609756097560975\n",
      "----------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "model = [Layer_Dense(10,784,'ReLU'), Layer_Dense(30,10,'ReLU'), Layer_Dense(10, 30,'Softmax')]\n",
    "\n",
    "train = Train_Model(model,X_train, Y_train, epochs=301, learning_rate=0.1)\n",
    "train.Train()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "7b554e70",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(784,)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test[0:,1].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "3d84efd9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [[6.01521986e-07]\n",
      " [1.57073175e-08]\n",
      " [1.01054224e-05]\n",
      " [1.92360484e-07]\n",
      " [9.19446416e-01]\n",
      " [7.23689916e-06]\n",
      " [1.29778233e-04]\n",
      " [4.10101189e-04]\n",
      " [1.26376516e-04]\n",
      " [7.98691762e-02]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "example = X_test[0:,2].reshape(784,1)\n",
    "train.predict_probability(example)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "00ee24bc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Prediction:  [4]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaEAAAGdCAYAAAC7EMwUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8li6FKAAAZXklEQVR4nO3dcWyU9R3H8c9Z4KjkeqSB9u5GbTqEaShhEbDYIBQSGrqEDDsdamLKEpnGlgQrIUOy2OwPuhAhZulE8I8CGSjJIowNInaBFhxjq0wiYQZrrKOOXioN9toOrqM8+4Nw2dFSeI67fnvt+5Vc4t09P+7r4xPePr27px7HcRwBAGDgAesBAABjFxECAJghQgAAM0QIAGCGCAEAzBAhAIAZIgQAMEOEAABmxlkPcLsbN27o0qVL8vl88ng81uMAAFxyHEfd3d0KhUJ64IGhz3VGXIQuXbqkvLw86zEAAPepra1N06ZNG3KbEffjOJ/PZz0CACAJ7uXv85RF6O2331ZBQYEmTpyouXPn6uTJk/e0jh/BAcDocC9/n6ckQvv379e6deu0adMmffrpp3ryySdVVlamixcvpuLlAABpypOKq2gXFRXpscce0/bt22OPPfroo1q5cqVqa2uHXBuJROT3+5M9EgBgmHV1dSkrK2vIbZJ+JtTX16czZ86otLQ07vHS0lKdOnVqwPbRaFSRSCTuBgAYG5IeocuXL6u/v1+5ublxj+fm5iocDg/Yvra2Vn6/P3bjk3EAMHak7IMJt78h5TjOoG9Sbdy4UV1dXbFbW1tbqkYCAIwwSf+e0JQpU5SRkTHgrKejo2PA2ZEkeb1eeb3eZI8BAEgDST8TmjBhgubOnauGhoa4xxsaGlRcXJzslwMApLGUXDGhurpaL7zwgubNm6cnnnhCO3fu1MWLF/Xyyy+n4uUAAGkqJRFatWqVOjs79atf/Urt7e0qLCzUkSNHlJ+fn4qXAwCkqZR8T+h+8D0hABgdTL4nBADAvSJCAAAzRAgAYIYIAQDMECEAgBkiBAAwQ4QAAGaIEADADBECAJghQgAAM0QIAGCGCAEAzBAhAIAZIgQAMEOEAABmiBAAwAwRAgCYIUIAADNECABghggBAMwQIQCAGSIEADBDhAAAZogQAMAMEQIAmCFCAAAzRAgAYIYIAQDMECEAgBkiBAAwQ4QAAGaIEADADBECAJghQgAAM0QIAGCGCAEAzBAhAIAZIgQAMEOEAABmiBAAwAwRAgCYIUIAADNECABghggBAMwQIQCAGSIEADBDhAAAZogQAMAMEQIAmCFCAAAzRAgAYIYIAQDMECEAgBkiBAAwQ4QAAGaSHqGamhp5PJ64WyAQSPbLAABGgXGp+ENnzZqlP//5z7H7GRkZqXgZAECaS0mExo0bx9kPAOCuUvKeUEtLi0KhkAoKCvTss8/qq6++uuO20WhUkUgk7gYAGBuSHqGioiLt2bNHR48e1bvvvqtwOKzi4mJ1dnYOun1tba38fn/slpeXl+yRAAAjlMdxHCeVL9Db26vp06drw4YNqq6uHvB8NBpVNBqN3Y9EIoQIAEaBrq4uZWVlDblNSt4T+n+TJk3S7Nmz1dLSMujzXq9XXq831WMAAEaglH9PKBqN6vPPP1cwGEz1SwEA0kzSI7R+/Xo1NTWptbVVf/vb3/T0008rEomooqIi2S8FAEhzSf9x3DfffKPnnntOly9f1tSpU7VgwQKdPn1a+fn5yX4pAECaS/kHE9yKRCLy+/3WYyBF3nvvPddrXn31VddrwuGw6zW4Px6Px/WaRN4Pvnbtmus1sHEvH0zg2nEAADNECABghggBAMwQIQCAGSIEADBDhAAAZogQAMAMEQIAmCFCAAAzRAgAYIYIAQDMECEAgJmU/1I74P89+uijrtc0NDS4XrN161bXayRp165dCa0bbSZPnux6zU9/+lPXa1544QXXa1577TXXa/7+97+7XoPhwZkQAMAMEQIAmCFCAAAzRAgAYIYIAQDMECEAgBkiBAAwQ4QAAGaIEADADBECAJghQgAAM0QIAGCGCAEAzHAVbYx4s2bNcr1mx44dCb0WV9G+6Wc/+5nrNYleudyt6dOnu17DVbRHLs6EAABmiBAAwAwRAgCYIUIAADNECABghggBAMwQIQCAGSIEADBDhAAAZogQAMAMEQIAmCFCAAAzXMAUwADPPffcsLzOtWvXXK9pb29PwSSwwpkQAMAMEQIAmCFCAAAzRAgAYIYIAQDMECEAgBkiBAAwQ4QAAGaIEADADBECAJghQgAAM0QIAGCGC5gCo9j3v//9hNbNnDkzyZMM7ttvv3W9prGxMfmDwAxnQgAAM0QIAGDGdYROnDihFStWKBQKyePx6ODBg3HPO46jmpoahUIhZWZmqqSkROfPn0/awACA0cN1hHp7ezVnzhzV1dUN+vyWLVu0bds21dXVqbm5WYFAQMuWLVN3d/d9DwsAGF1cfzChrKxMZWVlgz7nOI7eeustbdq0SeXl5ZKk3bt3Kzc3V/v27dNLL710f9MCAEaVpL4n1NraqnA4rNLS0thjXq9Xixcv1qlTpwZdE41GFYlE4m4AgLEhqREKh8OSpNzc3LjHc3NzY8/drra2Vn6/P3bLy8tL5kgAgBEsJZ+O83g8cfcdxxnw2C0bN25UV1dX7NbW1paKkQAAI1BSv6waCAQk3TwjCgaDscc7OjoGnB3d4vV65fV6kzkGACBNJPVMqKCgQIFAQA0NDbHH+vr61NTUpOLi4mS+FABgFHB9JtTT06Mvv/wydr+1tVVnz55Vdna2HnroIa1bt06bN2/WjBkzNGPGDG3evFkPPvignn/++aQODgBIf64j9Mknn2jJkiWx+9XV1ZKkiooK7dq1Sxs2bNDVq1f1yiuv6MqVKyoqKtJHH30kn8+XvKkBAKOC6wiVlJTIcZw7Pu/xeFRTU6Oampr7mQtAEixatCihdVlZWUmeBBgc144DAJghQgAAM0QIAGCGCAEAzBAhAIAZIgQAMEOEAABmiBAAwAwRAgCYIUIAADNECABghggBAMwQIQCAmaT+ZlWMLcuWLXO9Zvbs2SmYZKC//OUvw/I6I92rr75qPcKQErla9x//+EfXa95//33XayRp7969Ca3DveNMCABghggBAMwQIQCAGSIEADBDhAAAZogQAMAMEQIAmCFCAAAzRAgAYIYIAQDMECEAgBkiBAAwwwVMkbCMjAzXazweTwomGejkyZPD8joj3fjx461HGJLf73e95oc//KHrNVVVVa7XYHhwJgQAMEOEAABmiBAAwAwRAgCYIUIAADNECABghggBAMwQIQCAGSIEADBDhAAAZogQAMAMEQIAmOECpkjYj370I+sR0lZxcbHrNT//+c9drwmFQq7XjHS7du1yveZf//pX8gdBUnAmBAAwQ4QAAGaIEADADBECAJghQgAAM0QIAGCGCAEAzBAhAIAZIgQAMEOEAABmiBAAwAwRAgCY4QKmSFhhYaH1CHf0zDPPJLTuxo0brtfMnTvX9ZrHH3/c9ZqcnBzXazwej+s1kuQ4TkLr3Dp79qzrNe+8804KJoEVzoQAAGaIEADAjOsInThxQitWrFAoFJLH49HBgwfjnl+9erU8Hk/cbcGCBUkbGAAweriOUG9vr+bMmaO6uro7brN8+XK1t7fHbkeOHLmvIQEAo5PrDyaUlZWprKxsyG28Xq8CgUDCQwEAxoaUvCfU2NionJwczZw5U2vWrFFHR8cdt41Go4pEInE3AMDYkPQIlZWVae/evTp27Ji2bt2q5uZmLV26VNFodNDta2tr5ff7Y7e8vLxkjwQAGKGS/j2hVatWxf65sLBQ8+bNU35+vg4fPqzy8vIB22/cuFHV1dWx+5FIhBABwBiR8i+rBoNB5efnq6WlZdDnvV6vvF5vqscAAIxAKf+eUGdnp9ra2hQMBlP9UgCANOP6TKinp0dffvll7H5ra6vOnj2r7OxsZWdnq6amRj/5yU8UDAb19ddf6/XXX9eUKVP01FNPJXVwAED6cx2hTz75REuWLIndv/V+TkVFhbZv365z585pz549+u677xQMBrVkyRLt379fPp8veVMDAEYFjzNcVyq8R5FIRH6/33oM3IOdO3e6XvPiiy+mYBKMBFevXnW95uGHH3a9pr293fUa2Ojq6lJWVtaQ23DtOACAGSIEADBDhAAAZogQAMAMEQIAmCFCAAAzRAgAYIYIAQDMECEAgBkiBAAwQ4QAAGaIEADADBECAJhJ+W9Wxej1m9/8xvWa8ePHu17z9NNPu14zadIk12uGk8fjcb0mkQveJ/I6ib7WjRs3XK/hitjgTAgAYIYIAQDMECEAgBkiBAAwQ4QAAGaIEADADBECAJghQgAAM0QIAGCGCAEAzBAhAIAZIgQAMONxErlSYQpFIhH5/X7rMTCCTJ061fWahQsXJvRaia5z65tvvnG9Zt++fa7XnDx50vUaSXr44Yddr+nt7XW9xufzuV6D9NHV1aWsrKwht+FMCABghggBAMwQIQCAGSIEADBDhAAAZogQAMAMEQIAmCFCAAAzRAgAYIYIAQDMECEAgBkiBAAwM856AOBuvv32W9drDhw4kNBrJbpupLp+/fqwvdabb745bK+F0YMzIQCAGSIEADBDhAAAZogQAMAMEQIAmCFCAAAzRAgAYIYIAQDMECEAgBkiBAAwQ4QAAGaIEADADBcwBZAU/f391iMgDXEmBAAwQ4QAAGZcRai2tlbz58+Xz+dTTk6OVq5cqQsXLsRt4ziOampqFAqFlJmZqZKSEp0/fz6pQwMARgdXEWpqalJlZaVOnz6thoYGXb9+XaWlpert7Y1ts2XLFm3btk11dXVqbm5WIBDQsmXL1N3dnfThAQDpzdUHEz788MO4+/X19crJydGZM2e0aNEiOY6jt956S5s2bVJ5ebkkaffu3crNzdW+ffv00ksvJW9yAEDau6/3hLq6uiRJ2dnZkqTW1laFw2GVlpbGtvF6vVq8eLFOnTo16J8RjUYViUTibgCAsSHhCDmOo+rqai1cuFCFhYWSpHA4LEnKzc2N2zY3Nzf23O1qa2vl9/tjt7y8vERHAgCkmYQjVFVVpc8++0zvvffegOc8Hk/cfcdxBjx2y8aNG9XV1RW7tbW1JToSACDNJPRl1bVr1+rQoUM6ceKEpk2bFns8EAhIunlGFAwGY493dHQMODu6xev1yuv1JjIGACDNuToTchxHVVVV+uCDD3Ts2DEVFBTEPV9QUKBAIKCGhobYY319fWpqalJxcXFyJgYAjBquzoQqKyu1b98+/eEPf5DP54u9z+P3+5WZmSmPx6N169Zp8+bNmjFjhmbMmKHNmzfrwQcf1PPPP5+SfwEAQPpyFaHt27dLkkpKSuIer6+v1+rVqyVJGzZs0NWrV/XKK6/oypUrKioq0kcffSSfz5eUgQEAo4erCDmOc9dtPB6PampqVFNTk+hMAJKko6MjoXWPPPJIkicBBse14wAAZogQAMAMEQIAmCFCAAAzRAgAYIYIAQDMECEAgBkiBAAwQ4QAAGaIEADADBECAJghQgAAM0QIAGAmod+sCiA97Nq1K6F1ixYtSu4gwB1wJgQAMEOEAABmiBAAwAwRAgCYIUIAADNECABghggBAMwQIQCAGSIEADBDhAAAZogQAMAMEQIAmOECpsAo9qc//SmhdVeuXEnyJMDgOBMCAJghQgAAM0QIAGCGCAEAzBAhAIAZIgQAMEOEAABmiBAAwAwRAgCYIUIAADNECABghggBAMx4HMdxrIf4f5FIRH6/33oMYEx75plnXK958803Xa/Jz893vQbpo6urS1lZWUNuw5kQAMAMEQIAmCFCAAAzRAgAYIYIAQDMECEAgBkiBAAwQ4QAAGaIEADADBECAJghQgAAM0QIAGCGC5gCSIru7m7XaxYvXux6zT/+8Q/Xa2CDC5gCAEY0IgQAMOMqQrW1tZo/f758Pp9ycnK0cuVKXbhwIW6b1atXy+PxxN0WLFiQ1KEBAKODqwg1NTWpsrJSp0+fVkNDg65fv67S0lL19vbGbbd8+XK1t7fHbkeOHEnq0ACA0WGcm40//PDDuPv19fXKycnRmTNntGjRotjjXq9XgUAgORMCAEat+3pPqKurS5KUnZ0d93hjY6NycnI0c+ZMrVmzRh0dHXf8M6LRqCKRSNwNADA2JBwhx3FUXV2thQsXqrCwMPZ4WVmZ9u7dq2PHjmnr1q1qbm7W0qVLFY1GB/1zamtr5ff7Y7e8vLxERwIApJmEvydUWVmpw4cP6+OPP9a0adPuuF17e7vy8/P1/vvvq7y8fMDz0Wg0LlCRSIQQAWmI7wnhdvfyPSFX7wndsnbtWh06dEgnTpwYMkCSFAwGlZ+fr5aWlkGf93q98nq9iYwBAEhzriLkOI7Wrl2rAwcOqLGxUQUFBXdd09nZqba2NgWDwYSHBACMTq7eE6qsrNTvfvc77du3Tz6fT+FwWOFwWFevXpUk9fT0aP369frrX/+qr7/+Wo2NjVqxYoWmTJmip556KiX/AgCA9OXqTGj79u2SpJKSkrjH6+vrtXr1amVkZOjcuXPas2ePvvvuOwWDQS1ZskT79++Xz+dL2tAAgNHB9Y/jhpKZmamjR4/e10AAgLEjoQ8mAMDt+GkHEsEFTAEAZogQAMAMEQIAmCFCAAAzRAgAYIYIAQDMECEAgBkiBAAwQ4QAAGaIEADADBECAJghQgAAM0QIAGCGCAEAzBAhAIAZIgQAMEOEAABmiBAAwAwRAgCYIUIAADNECABghggBAMwQIQCAGSIEADAz4iLkOI71CACAJLiXv89HXIS6u7utRwAAJMG9/H3ucUbYqceNGzd06dIl+Xw+eTyeuOcikYjy8vLU1tamrKwsowntsR9uYj/cxH64if1w00jYD47jqLu7W6FQSA88MPS5zrhhmumePfDAA5o2bdqQ22RlZY3pg+wW9sNN7Ieb2A83sR9ust4Pfr//nrYbcT+OAwCMHUQIAGAmo6ampsZ6CDcyMjJUUlKiceNG3E8ShxX74Sb2w03sh5vYDzel034YcR9MAACMHfw4DgBghggBAMwQIQCAGSIEADCTVhF6++23VVBQoIkTJ2ru3Lk6efKk9UjDqqamRh6PJ+4WCASsx0q5EydOaMWKFQqFQvJ4PDp48GDc847jqKamRqFQSJmZmSopKdH58+eNpk2du+2H1atXDzg+FixYYDRtatTW1mr+/Pny+XzKycnRypUrdeHChbhtxsLxcC/7IV2Oh7SJ0P79+7Vu3Tpt2rRJn376qZ588kmVlZXp4sWL1qMNq1mzZqm9vT12O3funPVIKdfb26s5c+aorq5u0Oe3bNmibdu2qa6uTs3NzQoEAlq2bNmouw7h3faDJC1fvjzu+Dhy5MgwTph6TU1Nqqys1OnTp9XQ0KDr16+rtLRUvb29sW3GwvFwL/tBSpPjwUkTjz/+uPPyyy/HPfbII484v/jFL4wmGn5vvPGGM2fOHOsxTElyDhw4ELt/48YNJxAIOL/+9a9jj127ds3x+/3OO++8YzHisLh9PziO41RUVDg//vGPjSay0dHR4UhympqaHMcZu8fD7fvBcdLneEiLM6G+vj6dOXNGpaWlcY+Xlpbq1KlTRlPZaGlpUSgUUkFBgZ599ll99dVX1iOZam1tVTgcjjs2vF6vFi9ePOaODUlqbGxUTk6OZs6cqTVr1qijo8N6pJTq6uqSJGVnZ0sau8fD7fvhlnQ4HtIiQpcvX1Z/f79yc3PjHs/NzVU4HDaaavgVFRVpz549Onr0qN59912Fw2EVFxers7PTejQzt/77j/VjQ5LKysq0d+9eHTt2TFu3blVzc7OWLl2qaDRqPVpKOI6j6upqLVy4UIWFhZLG5vEw2H6Q0ud4GPnXdPg/t/9qB8dxBjw2mpWVlcX+efbs2XriiSc0ffp07d69W9XV1YaT2Rvrx4YkrVq1KvbPhYWFmjdvnvLz83X48GGVl5cbTpYaVVVV+uyzz/Txxx8PeG4sHQ932g/pcjykxZnQlClTlJGRMeD/ZDo6Ogb8H89YMmnSJM2ePVstLS3Wo5i59elAjo2BgsGg8vPzR+XxsXbtWh06dEjHjx+P+9UvY+14uNN+GMxIPR7SIkITJkzQ3Llz1dDQEPd4Q0ODiouLjaayF41G9fnnnysYDFqPYqagoECBQCDu2Ojr61NTU9OYPjYkqbOzU21tbaPq+HAcR1VVVfrggw907NgxFRQUxD0/Vo6Hu+2HwYzU4yFtrqKdlZWlX/7yl/re976niRMnavPmzTp+/Ljq6+s1efJk6/GGxfr16+X1euU4jr744gtVVVXpiy++0I4dO0b1Pujp6dE///lPhcNh7dixQ0VFRcrMzFRfX58mT56s/v5+1dbW6gc/+IH6+/v12muv6d///rd27twpr9drPX7SDLUfMjIy9Prrr8vn86m/v19nz57Viy++qP/+97+qq6sbNfuhsrJSe/fu1e9//3uFQiH19PSop6dHGRkZGj9+vDwez5g4Hu62H3p6etLneLD7YJ57v/3tb538/HxnwoQJzmOPPRb3ccSxYNWqVU4wGHTGjx/vhEIhp7y83Dl//rz1WCl3/PhxR9KAW0VFheM4Nz+W+8YbbziBQMDxer3OokWLnHPnztkOnQJD7Yf//Oc/TmlpqTN16lRn/PjxzkMPPeRUVFQ4Fy9etB47qQb795fk1NfXx7YZC8fD3fZDOh0P/CoHAICZtHhPCAAwOhEhAIAZIgQAMEOEAABmiBAAwAwRAgCYIUIAADNECABghggBAMwQIQCAGSIEADBDhAAAZv4HtgH5tJufBdgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train.predict_label(example)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "85381cf0",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_predictions = train.make_predictions(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "c790c904",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[5 6 4 6 7 8 0 0 7 5 3 6 3 1 1 6 9 3 9 6 8 2 3 5 0 2 0 5 1 6 2 3 6 1 0 2 1\n",
      " 7 8 9 9 2 9 1 7 0 7 6 9 8 3 1 6 7 6 1 0 7 9 8 1 1 5 4 6 7 4 2 6 6 6 2 3 3\n",
      " 0 8 4 4 5 1 1 3 5 0 9 8 5 9 7 4 7 5 2 5 0 8 9 4 7 3 2 6 9 3 7 8 3 7 6 3 1\n",
      " 0 1 2 0 1 1 2 2 3 8 5 3 4 6 6 2 2 5 7 1 1 9 6 5 8 7 0 3 4 3 0 4 0 3 8 7 9\n",
      " 1 8 2 1 5 5 6 7 6 0 9 1 2 8 3 3 8 7 3 1 2 6 2 9 5 7 7 4 0 3 5 8 7 2 4 5 6\n",
      " 2 6 3 1 9 4 6 2 2 5 4 2 4 9 4 6 7 6 7 9 4 4 9 9 7 4 7 6 8 3 2 4 5 6 9 2 4\n",
      " 0 1 0 3 7 1 3 6 3 8 7 2 7 4 4 4 6 5 8 4 5 0 9 5 7 0 3 0 1 5 8 3 8 3 7 4 6\n",
      " 3 3 0 3 6 0 2 3 5 3 7 3 5 0 8 9 0 8 5 1 8 1 7 8 0 4 9 7 6 8 3 8 7 0 4 5 2\n",
      " 4 3 1 7 2 9 0 2 5 0 8 6 3 3 3 3 1 1 1 2 6 0 5 3 1 7 8 3 4 4 2 8 2 9 9 1 3\n",
      " 7 6 7 3 8 0 6 0 8 3 8 7 5 4 1 5 8 0 3 0 3 6 1 6 8 9 4 8 1 0 4 1 2 0 9 0 3\n",
      " 7 4 4 1 1 2 3 1 1 0 0 7 3 1 8 0 9 9 4 6 0 0 6 9 2 5 9 4 4 5 0 6 3 3 2 3 8\n",
      " 1 8 0 7 6 4 8 5 6 9 6 9 9 6 4 7 0 2 1 2 9 6 6 7 5 4 5 7 3 2 7 1 9 4 7 2 9\n",
      " 2 9 4 1 5 5 7 6 9 6 5 7 1 7 7 6 4 7 1 7 1 7 8 8 0 3 4 4 0 6 7 4 8 8 4 1 1\n",
      " 1 6 1 2 6 9 6 4 2 8 5 8 6 6 9 9 3 7 6 4 5 8 1 7 1 4 5 9 2 2 0 6 4 2 6 1 6\n",
      " 5 1 4 4 2 5 7 3 9 1 0 8 1 2 5 9 4 7 2 7 6 2 2 0 6 6 0 5 7 5 0 2 2 1 0 2 4\n",
      " 4 1 8 3 5 2 6 7 7 9 6 6 5 8 2 2 4 6 4 6 1 4 5 9 1 2 1 9 6 7 1 4 4 3 1 7 9\n",
      " 0 9 1 6 4 0 3 9 8 2 0 7 2 4 3 2 2 1 9 6 5 9 5 4 4 1 5 5 7 8 1 9 6 4 0 4 4\n",
      " 2 5 7 5 9 6 7 9 9 8 8 2 7 9 7 0 2 1 9 4 2 1 6 2 8 0 3 8 2 7 2 5 9 1 5 8 4\n",
      " 9 9 0 6 8 4 7 6 6 6 8 8 3 4 6 5 2 5 2 9 3 1 5 9 0 2 2 1 7 8 7 6 5 3 1 9 6\n",
      " 4 6 6 6 3 6 0 7 5 9 7 7 0 5 3 5 9 0 6 1 8 0 7 3 7 1 8 6 6 1 6 2 9 2 7 8 2\n",
      " 3 5 1 8 2 5 4 3 8 8 0 2 8 1 3 3 4 7 9 8 6 6 5 7 3 5 8 0 5 7 3 8 3 4 0 3 2\n",
      " 4 4 0 8 2 0 5 7 1 5 0 9 5 1 3 3 5 6 3 1 9 1 0 1 7 0 0 3 6 9 8 5 6 7 7 4 4\n",
      " 5 6 4 7 7 0 3 8 3 0 4 3 2 5 5 3 2 8 7 8 5 1 9 5 6 8 0 1 8 1 8 2 9 3 7 6 6\n",
      " 8 0 0 9 8 6 9 5 4 3 2 1 1 9 8 9 9 2 5 6 0 1 4 2 3 2 5 2 9 1 7 7 2 5 0 5 0\n",
      " 3 7 6 3 2 6 0 4 8 4 5 9 5 0 1 3 5 0 6 1 8 2 7 3 1 9 3 0 1 1 9 8 8 3 5 4 4\n",
      " 8 1 9 9 3 7 4 1 4 6 7 0 0 3 5 5 8 6 1 0 7 1 8 4 5 7 5 3 9 7 1 1 7 6 8 8 6\n",
      " 3 8 9 0 2 2 9 2 2 1 7 3 7 9 7 9 9 6 0 4 7 3 9 9 7 7 3 0 2 8 4 7 9 6 1 1 3\n",
      " 9] [3 6 4 8 7 3 0 0 7 5 3 6 9 1 1 6 9 8 9 6 8 2 3 5 0 2 0 5 1 6 2 3 6 1 0 2 1\n",
      " 7 8 9 9 2 9 1 7 0 7 6 9 8 8 1 6 7 6 1 0 7 9 8 1 1 5 4 6 2 4 2 6 6 6 2 3 3\n",
      " 0 8 4 2 5 7 1 3 5 0 7 8 5 9 7 4 7 5 8 5 0 8 9 4 7 3 2 6 7 3 7 8 3 7 5 3 1\n",
      " 0 1 2 0 1 1 2 2 3 2 5 3 4 6 6 2 2 5 7 1 1 9 6 5 8 7 0 2 4 3 0 9 0 3 8 2 9\n",
      " 1 8 2 1 5 5 6 7 3 0 9 1 2 8 3 3 8 7 3 1 2 6 2 9 5 7 9 4 0 3 5 8 7 2 4 5 6\n",
      " 2 6 3 1 9 4 2 2 2 5 4 2 4 8 9 6 7 6 7 9 4 4 9 9 7 4 7 2 8 3 2 4 3 6 9 2 4\n",
      " 0 1 0 3 7 1 3 6 3 8 7 2 7 4 4 4 6 5 8 4 5 0 9 5 7 0 3 0 1 5 8 3 8 3 9 4 6\n",
      " 3 3 0 3 6 0 3 3 5 3 7 2 5 0 8 1 0 8 5 8 1 7 7 8 0 4 9 7 6 1 3 8 7 0 4 5 2\n",
      " 4 3 1 7 2 7 0 2 5 0 3 6 3 5 3 3 6 1 1 2 2 0 5 1 1 7 8 8 4 9 2 8 2 9 9 1 3\n",
      " 7 6 7 3 8 0 5 0 8 3 2 7 5 4 1 3 8 0 3 0 3 6 1 6 8 9 4 8 1 0 4 1 2 0 9 0 3\n",
      " 7 4 9 1 1 2 3 1 1 0 0 7 3 1 8 0 9 9 4 6 0 0 6 9 2 5 7 4 4 5 0 6 3 3 5 3 8\n",
      " 9 8 8 7 6 9 8 5 6 9 8 7 9 6 4 7 0 2 1 2 9 6 6 7 5 4 5 7 3 2 7 1 9 4 7 2 9\n",
      " 2 9 4 1 5 3 7 6 9 2 5 7 1 7 7 6 4 7 1 3 1 7 8 8 9 3 4 9 0 6 7 4 8 8 6 1 1\n",
      " 8 6 7 2 6 8 6 4 0 8 5 5 6 6 9 9 8 7 6 4 5 8 1 7 1 4 8 4 8 2 0 6 4 2 6 1 6\n",
      " 5 1 4 4 2 5 3 8 9 1 0 8 1 2 5 9 4 7 2 7 6 2 2 0 6 6 0 5 9 5 0 2 2 1 0 2 4\n",
      " 4 1 3 3 5 2 6 7 7 9 6 6 5 8 2 2 4 6 4 6 1 4 8 9 2 2 1 8 2 7 1 9 4 3 1 7 9\n",
      " 0 2 1 6 4 0 3 9 8 2 0 7 7 4 3 2 2 1 7 6 5 9 5 4 4 1 3 5 7 2 1 9 6 4 2 9 9\n",
      " 2 3 7 5 9 6 7 7 9 8 7 2 7 7 7 0 2 1 9 5 3 1 6 2 8 0 5 8 2 3 3 3 9 1 3 8 4\n",
      " 7 2 0 6 8 4 7 6 6 6 6 8 3 4 6 5 2 5 2 4 5 1 5 9 0 2 2 1 3 8 7 2 5 3 1 9 6\n",
      " 4 6 6 6 8 6 0 7 3 7 7 9 0 5 5 5 9 0 8 1 8 0 7 3 7 1 8 6 6 1 4 2 9 1 7 8 2\n",
      " 3 8 1 8 2 3 4 3 8 8 0 2 8 1 3 8 4 7 9 8 6 6 5 7 3 5 8 0 5 7 3 8 8 4 0 3 2\n",
      " 4 4 0 8 2 0 5 7 1 5 0 8 5 1 3 3 5 6 5 1 4 1 0 1 9 0 5 5 6 9 8 5 6 7 7 4 4\n",
      " 5 6 4 7 9 0 3 8 8 0 4 3 2 5 9 3 2 8 7 0 8 1 9 5 6 8 0 1 8 1 8 2 9 3 7 6 6\n",
      " 8 0 0 9 9 6 7 5 4 3 2 1 1 9 8 9 9 2 5 6 0 1 4 2 3 2 5 2 9 1 7 7 2 5 0 5 0\n",
      " 5 7 6 5 3 8 0 4 8 4 5 9 0 0 1 3 5 0 6 1 8 2 7 3 1 9 3 0 1 1 7 4 8 5 5 4 4\n",
      " 8 1 9 9 3 7 4 1 4 4 7 0 0 3 5 5 8 6 1 0 7 1 8 4 5 7 5 3 9 7 1 1 7 6 8 8 6\n",
      " 3 8 9 0 2 2 9 2 2 1 7 3 7 9 9 9 8 6 0 3 7 9 9 8 7 7 3 0 2 8 4 7 9 5 1 1 3\n",
      " 9]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.859"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.get_accuracy(test_predictions, Y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9610f762",
   "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}