03. K Nearest Neighbour.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# K Nearest Neighbour\n",
    "\n",
    "It assumes that similar things exist in close proximity.   \n",
    "Used for both classification and regression.  \n",
    "\n",
    "Algorithm:  \n",
    "Step 1: Choose the no. K of neighbours  \n",
    "Step 2: Take the K nearest neighbours of the new data points by Euclidean distance  \n",
    "Step 3: Among K Neighbours, count the no. of data points in each category  \n",
    "Step 4: Assign new data point to the category where you counted most neighbour  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Case 1: K = 3, Solid Circle  - Assigned to Red Triangle Group  \n",
    "Case 2: K = 5, Dashed Circle  - Assigned to Blue Square Group  \n",
    "<img src=\"Image/KNN.JPG\" width=\"200\">\n",
    "<font size=1><center>Image Source:Wikipedia</center></font>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[64  4]\n",
      " [ 3 29]]\n"
     ]
    }
   ],
   "source": [
    "# Importing the libraries\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "\n",
    "# Importing the dataset\n",
    "dataset = pd.read_csv('Data/Classification/Social_Network_Ads.csv')\n",
    "X = dataset.iloc[:, [2, 3]].values\n",
    "y = dataset.iloc[:, 4].values\n",
    "\n",
    "X = X.astype(float)\n",
    "y = y.astype(float)\n",
    "# Splitting the dataset into the Training set and Test set\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.25, random_state = 0)\n",
    "\n",
    "# Feature Scaling(Standardization)\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "sc = StandardScaler()\n",
    "X_train = sc.fit_transform(X_train)\n",
    "X_test = sc.transform(X_test)\n",
    "\n",
    "# Fitting K-NN to the Training set\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "classifier = KNeighborsClassifier(n_neighbors = 5)\n",
    "classifier.fit(X_train, y_train)\n",
    "\n",
    "# Predicting the results on Test set\n",
    "y_pred = classifier.predict(X_test)\n",
    "\n",
    "# Creating the Confusion Matrix\n",
    "from sklearn.metrics import confusion_matrix\n",
    "cm = confusion_matrix(y_test, y_pred)\n",
    "\n",
    "print(cm)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAEWCAYAAABv+EDhAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAABD80lEQVR4nO2de5gcZZXwf2cmHTJJIEPIJEAmk+DCzgqGywqjy8qSj0tg1aCg7mecBdm4ZvWRRdH9ViWIIA64ruINXQ0fKsiIeAEhQSWBT4KIOoKGYLisupLJhJBMICMhmZDOzPn+qOpMX6p6qqaququ7z+955pnuurz1VnX3Oe8573nPEVXFMAzDaDyaqt0BwzAMozqYAjAMw2hQTAEYhmE0KKYADMMwGhRTAIZhGA2KKQDDMIwGxRSAYaQEEekQkZdEpDnAsQtEREVkks/+q0Tk1vh7adQTpgCM2BGRZ0TkrLz3bxeRnSJyusexi1xB9uWi7Q+JyMUV6G5g3H4eXWb/xe4x/6do+4CILBqvfVXtV9XpqjoSvbeGMT6mAIxEEZF3Al8G3qCq63wO2w1cJCILKtAfzxFzjLwAfFhEDkn4OqmgAs/TSBBTAEZiiMhy4LPAOar6cJlDh4BvAh8v09YyEXnStSTuFZH5efu+ICKbReRFEXlURE7L23eViHxfRG4VkReBi0VkhojcJCJbRWSLiHwy53YRkaNFZJ2I/FlEdojI7e72B90mH3PdNP/bp6tPAr8ALvO5jyYR+YiI/FFEnheR74rITHdfgVtHRI4SkQdFZJeI3CciX/Zw63SLSL/b1xVF+6aIyO3u+b8RkRPy+vFKEXlARIZEZKOInJe37wER+ee89xeLyEN571VE3icivwd+7/McjBrAFICRFO8FrgHOVNVHAhzfA7xFRDqLd4jIm4HLgQuANuBnwG15h/waOBGYCXwb+J6ITMnb/ybg+0Ar0AvcDOwHjgZOAhYDOYF3DbAGOBRoB74EoKp/5+4/wXXT3F7mXj4GXJYT7EVcCrwZOB04EtiJYyF58W2gDzgMuAq40OOY1wGdwJnAlSLyyrx9bwK+x9hz+aGIZEQkA6xy73M28K9Ar9ezL8ObgdcAx4Y4x0gZpgCMpDgb+CXweJCDVfU54KvAJzx2/wtwnao+qar7gWuBE3NWgKreqqrPq+p+Vf0scBCOUMzxC1X9oaqOAocAfw98QFV3q+p24HPA291js8B84EhV3auqDxESVV2PI1w/7HMvK1R1QFVfxhHsby12pYhIB3AKcKWq7nP7cbdHe1er6rCqPgY8BpyQt+9RVf2+qmaB64EpwGvdv+nAp9y2/x+wGlga4javU9UXVHU4xDlGyjAFYCTFe4C/BP6viEjAc/4DOCffVeEyH/iC664YwvGzCzAXQEQ+5LqH/uzunwHMyjt/c1FbGWBrXntfwxkJA/y723af6xpZFrDvxVwJvFdEDve4lzvzrv0kMALMKTruSOAFVd3jcx85nst7vQdHsJcc7yq/AbfdI4HN7rYcm3CfZ0C8+mLUGDaBYyTFdhy3xDrgKzguobKo6vMi8nkcN0w+m4EeVe0tPsf193/YvdZGVR0VkZ04QvxA00VtvQzMcq2J4j48B7zbbft1wH0i8qCq/mG8/he185SI3IHjuiq+l2Wq+nOPe1mQ93YrMFNEpuYpgXlh+pB/vIg04bi0ns3tE5GmPCXQAfy3+3o3MDWvnWIlBoXP1KhRzAIwEkNVnwXOAM4Vkc8FPO164FQg35f9VeCjInIcgDuJ+zZ338E4/vxBYJKIXInj5vHr01Yc98xnReQQd1L2L8QNURWRt4lIu3v4ThxBlwvL3Aa8IuB9AFwN/BPO3EP+vfTk3Fci0iYib/Lo5ybgEeAqEZksIn8DLAlxbYBXi8gFrnvpAziK75fAr3CE/L+7cwKL3La/4563HrhARKaKE/b6rpDXNWoEUwBGoqjqZhwl8FYRuS7A8S8Cn8aZuMxtuxPHPfQdN5Lndzh+fIB7gR/jjF43AXsZ3z1xETAZeAJHyH8fOMLddwrwKxF5Ccfn/n5V/ZO77yrgZtd98w8B7uVPwLeAaXmbv+C2u0ZEduEI5Nf4NNEN/A3wPPBJ4HYcIR6Uu4D/jXOPFwIXqGpWVfcB5+E8wx04FtpFqvqUe97ngH04Cu9mnIlzow4RKwhjGLWBG5L6lKr6hssaRhjMAjCMlCIip7juqSYRORcnrPOHVe6WUUfYJLBhpJfDgTtw1gEMAO9V1d9Wt0tGPWEuIMMwjAbFXECGYRgNSk25gGZNnqwLpk4d/0DDMAzjAI/++c87VLWteHtNKYAFU6fyyGmnjX+gYRiGcQBZvXqT13ZzARmGYTQopgAMwzAaFFMAhmEYDUpNzQEYhmHETXbaNAbOP5+9c+ZA4MS1KUSVKdu20X7nnWR27w50iikAwzAamoHzz+fgY49lwUEHETxzefpQVZ4/7DAGgKNuLS4c5425gAzDaGj2zpnDYTUu/AFEhMMOOsixZAJiCsAwjMZGpOaFfw4RCeXGqpoCEJEpItInIo+5lZeurlZfDMMwGpFqWgAvA2eo6gk4Bb3PFZHXVrE/hmEYVeEnDz5I5znncPRZZ/Gpr32tZL+qcuk113D0WWdx/JIl/GbjxliuWzUFoA4vuW8z7p9lpjMMo6EYGRnhfVdfzY9vvJEnfvQjblu9mif+UFiB9Mfr1vH7Z57h92vXsvKaa3jvx+MpCVHVOQARaRaR9Tj1Y9eq6q88jlkuIo+IyCOD+/ZVvI+GYRj59N49nQWLFtDUeTQLFi2g9+7pkdrr27CBo+fP5xUdHUyePJm3v+EN3HXffQXH3HX//Vx0/vmICK898USGdu1i6/btka4LVVYAqjqiqifiFKvuEpFXeRyzUlVPVtWT2yZPrngfDcMwcvTePZ3lV8xh07MZVIVNz2ZYfsWcSEpgy7ZtzDv88APv2w8/nC3btpU/Zs6ckmMmQiqigFR1CHgAOLe6PTEMw/BnxfWz2LO3UGzu2dvEiutnTbhNr5osxVFJQY6ZCNWMAmoTkVb3dQtwFvBU2ZMMwzCqSP9W77WzftuD0H744Wx+7rkD7weee44jZ88uf8y2bSXHTIRqWgBHAD8VkQ3Ar3HmAFZXsT+GYRhl6Thif6jtQThl4UJ+/8wz/GnzZvbt28d37rmH8848s+CY8844g1vuvBNV5Zfr1zNj+nSOiEEBVC0VhKpuAE6q1vUNwzDC0vPBHSy/Yk6BG2jqlFF6Prhjwm1OmjSJG668knPe9S5GRkZY9ta3ctwxx/DV224D4D1Ll/L6RYv40bp1HH3WWUxtaeEb110X+V7AcgEZhmEEpvs8J3J9xfWz6N86iY4j9tPzwR0Htk+U1y9axOsXLSrY9p6lSw+8FhG+fNVVka7hhSkAwzCMEHSf91JkgZ8WUhEFZBiGYVQeUwCGYRgNiikAwzCMBsUUgGEYRoNiCsAwDKNBMQVgGIZRZZZ99KPMfu1redUb3uC5v+7SQRuGYRgOF19wAT+56Sbf/XWZDtowDKPmuPtuWLQIOjud/3ffHbnJvzvlFGbOmOG7vy7TQRuGYdQUd98NV1wBzz4Lqs7/K66IRQmUo67TQRuGYdQE118Pe/cWbtu719meIHWXDtowDKPm2Lo13PaYqMd00IZhGLXFEUeE2x4TdZcO2jAMo+b44Acdn3++G2jKFGd7BJZedhkP9PWxY+dO2k87jasvvZTsfqfGgKWDNgzDSAPnnef8v/56x+1zxBGO8M9tnyC3fe5zZfdbOmjDMGqTgQF4+mkYHoaWFid8sr292r2aOOedF1ngpwVTAEbtUw8Cph7uwYuBAXj8cRgZcd4PDzvvoT7ur8YxBWBUlrgFXT0ImHq4Bz+efnrsvnKMjDjb03JvqqhqLGGV1UZVnfUJATEFYFSOJARdWAGTxpF2LQjJiTI8HG57FZiybRvPH3YYhx10UE0rAVXl+ZdfZkqIBWKmAIzKkYSgCyNg0jrSrgEhGQgv5drS4n0fLS2V758P7XfeyQAwOGcO1LACQJUp27bRfuedgU8xBWBUjiQEXRgBk9aRdg0IyXHxU65z58KWLYXPvbnZUQ4pIbN7N0fdemu1u1EVbCGYUTn8BFoUQdfZ6QiUfPwETFgFNDAA998Pq1c7/wcGJt7PcoS5h7Tip1wHB2HhwrHPuKXFeV/rrq06wSwAo3J0dhaOEiG6oMsJkiB+/TAj7Uq6i8LcQ1opp1zb20vvJQ1zMWnoQ5UxBWBUjqQEnZeA8SKMAorDXRRGwAS9h7SSVuXqRxr6kAJMARiVpZqCLowCijpfUYMCpnfgdFY8fRH9w7PoaNlBT+ctdLevC3ZypZVrVNLQhxRQWwpg927o6yvd3tVV+b4YtUlQBRR1YrbGBEzvwOksf/wS9oxMAWDT8GyWP34JQDAlUEnlGgdp6EMKqC0F4MXQEKxZA4sXV7snRj0Rdb4iSQGTgO96xdMXHRD+OfaMTGHF0xcFtwIqpVzjIA19SAFViwISkXki8lMReVJENorI+8c7Z2QEdg4V/dHq7Fyzxts6MIyJ0N4eLXoliYgnGHMt5YRXzrUUMUKpf3hWqO2RSEPUUxr6kAKqaQHsBz6kqr8RkYOBR0Vkrao+4XfCY5zA4TxSsn0gO4cZDDF5cNCsASM+osxXJBHxBIm5ljpadnDq8FquZQUd9NNPB5fTw8MtZ0frL3hbLAsXVjcCpx4ir2KgagpAVbcCW93Xu0TkSWAu4KsADjkETjutcFtfH7QPbeOO7BK66KONIUcJtLZ6N2LzBfFTr+F0Ue4rKQGTkGvp1rYPcFL/XUxjDwAL2MSNLOe3bW+K1K7vZPjChXDmmdHajkqtR17FQCrmAERkAXAS8CuPfcuB5QAtLR0l5+bk+bK+VQwN5VkDQ0PeF8spB1ME8VCD0S6BiOO+khAwCfmuXzf4Q6Cw3WnscbdHENQ1NhneaFRdAYjIdOAHwAdU9cXi/aq6ElgJ0Np6sm+au66uQmsg43HMSdk+cxXFTaXj5StlbaRVcCXlWkpq0tqibVJNVRWAiGRwhH+vqt4Rtb18a8CLoSG8XUVmDUycSsbLV9LaCHtfGzbA5s1OKl4RmDcPjj8+3j5Bcq6lGCwLz3UELfdbtE2KqZoCECfv6k3Ak6p6fZxtl5PnJa6inDXghSmH8alkvHwlR+Vh7mvDBujvH3uvOvY+ohLwXZwV9/22tRXeQ/72gP30Wkcwf+5veN2W78ZvsdTrvFOFKRsGKiLniMi7XB99/vZlMVz7b4ELgTNEZL379/oY2i1LV5cj19sz21jLYnZm2thJa8nfvixOIisLLS1P1HC6MCPtSroTwtzX5s3ebfhtD0hOqG4ano3SdECo9g6cHqldTwYHw20vwm8dwT8Ofj7+ZHAJhcI2Ir4WgIhcC7wO+A1wuYh8XlW/5O6+BPh6lAur6kNAVZJvF08cewUMfWtoCafQx6E5C8GsAW+iuiTCjLQruXgnzH35VWAKUZnJi9CLs6KMiiMq17LrCOKeDE/r/EwNUs4FtAQ4SVX3i8hVwLdF5BWqehlVEtxxk5s49uLCVtdVxBxmZF1XUV+fKQEvKhUvn9QEqB9B70vEW9hHLC4SanFW1PmRTAayWe/tAeho2cGm4dme22PHJpZjo5wCmKSq+wFUdUhElgArReR7wOSK9K4ClJPn+VFFp2b6ODSXdsILsxAmRpiRdloX78yb5+0/nzcvUrOhFmdVeVTc03lLwRwAwNTmvfR03hL/xSyNQ2yUUwB/FJHTVXUdgKqOAO8SkU8Cb6lI76pMvqvo64NO9NAknwGRuYoiEMaCSOPindxEb8xRQKEWZ0UdFXuN/sttLyLnkppwNtEwVNoSrGNEffyUItICoKol3yARmauqWxLuWwmtrSfraaeVpoKoBH19ThipF7nQ0hkMMZmsEzlhSsCIyv1lQiiLV9GGOTbqtdKARQGFQlavflRVTy7e7msBeAn+vH0VF/7VJkho6QFXUb41ENdFjGRJo0AJM6qPOiqutVF1Gi3BGqTqK4HrAS9X0YxsmXQUXmSzNslcLQYG4LHHxiZyh4ed91A7xeKjzo+kdX4F0qmc6wRTADHS1VVoDQSLn3A4pdhyMEVQOTZuLI3iUXW211Lah6ij4jSOqus111RKCKQAROR1wDGq+g0RaQOmq+qfku1abTJeOgo/CiwHCzmtLBEnQBMjzaPySmEx/4kyrgIQkY8DJwOdwDeADHArzkpew4ewstt3HsGLWrAQzGyPhzSOyiuJxfwnShAL4HycVM2/AVDVZ90CLkaMeIacAoe2ehycdldRrZntERdBAabwksJi/hMliALYp6oqIgogItMS7lNDUzKPMFR6zEn5k8xpdBXVmtl+3HGFk8DgxPIfd1yw82tN4dUStRadVGMEUQDfFZGvAa0i8m5gGXBjst1qbIKktV7FEhZl14xlM02TNVBjBdEj+9prTeHVEjYPkijjKgBV/YyInA28iDMPcKWqrk28Z0ZZeX6hV8hpWqyBpMz2JEfaUXzt9eKnTqsbq9HnQRIkUBSQK/BN6KeIfFdRpawB39z0xcRhtnsJoyRH2lGEXz34qc2N1ZCUSwe9C/DKEyGAquohifXKCEROxi9Zs4o7SNYa8Cv4AZQqgahmu58wKhb+OaKOtKMKvzT7qYMqNnNjNSTlUkFYpE+NsHhx8tZA6Nz0Icz2Ysvid/s7me4ljPwIE63jRVThl1Y/dRjFVi9uLCMUQReC/TVOcRgFHlLV3ybaKyM0SVsDoXLTh8DLspjKnyO1GZo4hF8a/dRhFFs9uLGM0JQtCQkgIlcCNwOHAbOAb4rIFUl3zJgYixfDsrZVtGe28UBm8VhpyzVrIpW39CvsEbXgh5dl0U9HuEairtj1E3K1LvzCJpOLUtqzDL0Dp7Pg/ptoWn0XC+6/KZmSlsaEGFcBAEuBU1T146r6ceC1QHey3TKi0NXlKIIlrGItixmkzVEEOWtgAvR03sLU5r0F2+Io+OFlQVxOD7uZWrixudnX1fNSpjWagAkr/AYGnPTJq1c7/9Nai9ZPgWUypf1vb4+/di8VrmtshCaIC+gZYAqQ+/UfBPwxqQ4Z8VF2bsCLMvMFSRX88ColeBvdzMq8yBcnfajQpw4lk637JcO/7v8cm7JOGwWT0/T6++VzirCrK5wPf2CA/Y9tZJK6VsfwsPMe0ucC8pqcFoH9+8espuJ5gZjvIfTckVFRykUBfQnH5/8ysFFE1rrvzwYeqkz3jKh4zQ14ppgYGhpzFS1e7NlWd/u62H+0fqUEX3Pcbmj3KUSSJ6g/uP+zfDN7ccHuPSNT+NXGaXSPekyA/vGP8PLLjgDMZAonygMIv5c2PsN0LXQ5TdKssz1tCsBLseUL/xwJRvskNXdkxEM5CyBXeutR4M687Q8k1hsjMYqtAYZKjzmQYqKCK4tDWxZFo9QbVv+L52EfzH4K8JgA3bWLfTKZP9MGWZwqbuMovnymZr0nqP22V53iUf3q1d7HJRTtU9Fi8UZoyoWB3lzJjhjJk5PnF4ZJMRFAKEYlimXhK2DY7Hm8Ag9kFrME5xmsYgmn0MehDAUKm+1nHgsoLQDvbK8BKhztU9Fi8UZogqSDPga4DjgWZy4AAFV9RYL9MhKk3MA+31XUlhOKflRAOYyHn4DZ0zSD6dmhkuP3NbVwYesqWt33F7KKVe79BuH6zEe4LvtvBwq1A+xmKtdnPsIX+VGEO6kQFV60VtFi8UZogkwCfwP4OPA54H8B/4SzGtioQ4pdRaf4CMagI+ak8RMw01nguXp4debN7NgBo6PQ1ATTpgEhBr+vOW43lzz2ZT6uV9FBP/10cLVcxVnHRXRpVCoPTxUWrYWy8Kqdj6ja168wosWl8IoPEHlUVV8tIo+r6kJ3289U9bSK9DCP1taT9bTTHhn/QCMWyg3+D7iKcCdTo1gDSf3o8tttauIh/RvO1vvYS2FUytUHf4YrW/5zbMM4Ci1wTqSg91W8YhecUXkMYZg1RbWfQ7WvnyCyevWjqnpyyfYACuDnwGnA94H/B2wBPqWqFU90YgogPaxZ49Qr6KKPtlzRgtZW74PLCdRK/ej6+liwvY9NzC/Z1c5mNk8+2nkzUYumWNi3tcGWLcHu6/77/f3yZ/pEQtUj1X4O1b5+gvgpgCALwT4ATAUuBV4NXAi8M9beGTVH/orjwWzr2EIzr79yq5DLpSuIk64u+pnnuWsLc50XUYT/44+PCY/hYejvD35flofHodrPodrXrwJB6gH82n35Eo7/PzZE5OvAG4HtqvqqONs2kicnK9vXbHOql1FqBBwz1Me0bJlQywr+6Hwjhpq2RHNheSkxP/xGmI2Wh8fLPRbmOSThNmzAz6HcQrDPq+oHRGQVHmmhVfW8GK7/TeAGwGLCapj8iWOv9QUlUUX5I+0K/ug8I4bYQ8/xt0drOIyy8rqvNKeTTgK/LKUzZng/y7a2YOdDNCXQaJ8D5S2Ab7n/P5PUxVX1QRFZkFT7RuUo5znJKYeB7JzChVetrTDJ5ys4aVKh2yiGSKPEQhL9lFgxfsIkremkk8LP7bdzp/fxg4PBzo+6mrm9HV54ATZvdupDi8DcufX7OVB+IdijItIMvFtV/7GCfSpARJYDywFaWkJmiTRSQVeXI8vbh7aNTRwz5MwPgCMYi0ddL7/s/OWIKeQ0iXQWviPHuXMd4RVEqKcxnXRS+ClLv4CU4uPDug3DRGNt2TLWD1Xn/cyZdfvZlJ0DUNUREWkTkcmquq9SnSrqw0pgJThRQNXogxGd/EL3B9JR5Gh2/3w4Kds3ZjmsW+fks0nTSLnRRvBR8bOYRLyVQLHbLOxcQVB3UQNWRQuaDfTnInI3sDu3UVWvT6pTRv2SswYuxDsdhRc5hXHqyDqm79o1tiNNdWsbaQQflXIWk1fobLHbLIyvPoxQtyggT551/5oAKxNpRCaMF6evz1lvcBJ9TNNdpQfU+QitLilnMc2cOb4lFcbiCiPULQqoFFW9OqmLi8htwCJglogMAB9X1ZuSup5RO/T1OSP/O7JLODXTx6HZQf+Dw4zQGmypf2rxs5iCWlJBjwsj1C0KqBQRaQP+HTiOwmRwZ0S9uKoujdqGUX/khP+BqKFs1gkF3LUr2ggtqfDBXNspVCyB01ZAau8hEmGEegPO5QRxAfUCt+Ms2HoPzirgMsMxw5g4OeG/iiV5KSbaHL+RX9qIoCO0pCb5klQsEciVY8yteyiollasBFJ6D5EJG9rZYHM5QVJBHOa6ZbKquk5Vl+HUBTaM2OnqcqI9l7BqLMXE4KAbRxqxbm1Sk3yVSmcRknLlGEtI6T1Exi+0M611nCtMEAsgVz9uq4i8AWdCuHFUpFFxitcNnJrp49D8xWMTTcyV1CRfSqNHQpVjTOk9RKYBQzvDUC4VREZVs8AnRWQG8CHgS8AhwGUV6p/RoBSvGxhgjlOuMmcNTGRBWNhJvqA+8ZRGj4Qqx5jSe4hMvSq2mCjnAtoiIjcCe4AXVfV3qvq/VPXVqnp3hfpnNDg5l1B7ZhtrWczOTNtY8Xq/DKN+hHEheWX4fPxxb9dBZ6ejSPJJQfRIT+ctTG3eW7DNtxxjSu8hMn4KrNYVW0yUUwCvxCkM/zFgs4h8XkReU5luGcYYXV1j6acfznYxSFvh3EAShPGJR52bSIju9nWsXHgD81u2I4wyv2U7Kxfe4B0FlNJ7iEy9KraYGLcgDICIHAm8DXg7MBv4jqquSLhvJVhBmMoRJiKwktGDnusDMplgeYLCFJ9Zvdq/nTe+caLdL9+3Bgo/rCj2bH0LwgSZBEZVnxWRm4CdwAeBfwYqrgCMyhAmItDv2BdeCJ4Hzev3Cd6/2fy5ga8POonlAs8NhJkQrKRPPM0hmPUgPBsstDMMZRWAiEwBlgBLgb8FfgJ8FChTLdaodcLISb9j+/vH3odVII895rzOGade53d1jU0Ql0QKedHaOv6EYM68yN2EF/v2FV4jiOWR365fm8VMNFIl/1qtrXDkkRMX4AMDzoeR/0HkPhyofcVglI0C+jZwFvAg8G3gHaq61+94o34IEzgRNJgijALx8kp6ne9lDfh9oQ8dHOQlpjF9LJ/hAV7KtDJ9zRrIZtlHht2ZVmiGDMO0jOymiVFGaWK4eRrZ5paSdsumqi5uF8iMFLbbBIhXp4eHw0U8ucJ/nxu4Pfn552H79sL2wlgWGzeWfhiqsGGDs6AqjRaLEYpyFsC9wL+oemXgMuIijRZ2JgPZbOn25uaxutnjVfHzIooCKXdsvjUApWUpvzXkJJO7hsu5jsuZxp4D+3YzlcuzV/HFyf/OIG38NtPFha3BMpXm2p2RHWLy0FChsM4biee3e/5wL9fvWk4zowA0M8oogpQW3UMBKW7XjzxF80DGKW959r7VpYoljGXh9SUAGB0t3Wax9TVJuYIwN1eyI41Iml2/XoyMlEZFemXw9cPLfR5GgZRzv5eTjxe6yiG7b5QdzOJaVtBBP/10cDk9fIelfCx7Le2ZbaFqzlyYV9vgFIpcUHnCeElm1YF2P3H/CqbmKSCAJhQtUgK7mcofm/+Sv8o+UVhBzQt31P9n2uijiwvcVNt7ve2K5GLgLba+5giSCsJIiLSuvvcb+BUzMuJM9BZHD86c6X18cWlX8I7SE3H+8okSuZcLI5WmJm6jm6N4hmZGOYpnuI1uZrEjtPDPb/fC1qLwVJxR/1oWs4RVLF481u6s4X7PthRle8t8RhG2t8znv05cySnNv2UtiwvaLcEV/g9kFtOe2cayNud6ixfDQNN873OaAv7sM5lgx+Ww2PqaI0g20INU9eXxthnhSesixbBuneIgi3vv9T5261bvdO8LFwaPAorC8cfD+vWl2/cePJvFp0+83XwXVK7SWW7Uv7hIoexo6WD28KaSNna0zOddZz5TsG1xe2G7pwyVrnnYn209MOpfvLhw3+3H93DJ48uZMjJmcSggIsHcSscdVzgJDI5mnjcvWOEWI/UECQP9BfDXAbYZIQkbabhhQ2FSw3nzHKEWlKDzDV4ZE/woHqmDvwWRzXq7vBYu9E7vE7cbrDjbL8DBB8PpEYR/jpwszbmFioVxjls6S4Xy3uap3NLZU7bdJWt85iXcJRDFigZgXXs3z78A7928gnZ1XF6/lFO5YPQHY24lv45C9MIt5Ujj5FcDUi4K6HBgLtAiIicxFqhwCDC1An2re8KkptmwoTC0UnXsfRAlEGa+wet3H7aOtx/VzsuVdEj4eIPqde3dAFz09ApmDfezo6WDWzp7Dmz3o5yc9mNgAH68pZtPaV7bCh/iM3xY/tOZvC4XwQTRC7f4dayWJr/qmHIWwDnAxTiZP/Pr/+4CLk+wTw1DmPoTmzd7t7F5czAFEDUpYlOTd/CHl5vYL4rIj+Hh0uiiNMgBv0Fq1MFrL91cSTfDQAvQCZDAgNjrMwf4XNO/8aVJ/8YqlrAouyaYNRAnlqEzNYwXBXSziLxFVX9QwT41FEEHUn4j7aAj8DDzDV4DND+8lIKf63jSJH/FUBxdBNWVBeVWOOe7v8P2d6IL3yaC3+c2OurI+iVrVnEHzvqJNobKr2eIk7ROfjUgQeYAVovIO4AF+cer6ieS6pRRioi3sPfywXsRZr7Bb+TohddxXkWY5s1z3MZB5haSHAwGHb37DVJz91S8fePGibcbdOFb2HvLZOCt2d6SsNcftjguocWLCyeZK2YN1Gvq6RokSDzYXcCbgP3A7rw/o4LMmxduezFhkiJGHYj5FWGC0pBRP5IYDIbJ8Dw8DEvp5U8sYIQm/sQCltLra3Fls6XtbtjguLZWr3b+DwzEs/At6L29LdvLjSxnAZtoQlnAJm5kOe9v6z1wXi6UdQmrDoScAhNLtx0Uy9CZGoJYAO2qem7iPTHKkvPzTzQKKMx8g5+14YXXHEA5F++ZZxZeM+f7LyaJwWAY1/M7M718Obv8wKrhnPAU4NuUn7DNteuXDykoYZ6B1731sKJg1TPANPZw2eAK3lV0DxW1Bhqw+HpaCaIAHhaRhaoa8utrxM3xx4cL+ywm6nxDMSKOv7+YMC5ev0iotrb4J4bD9OtaH+H5H80ruJ3uwC6yfMKcE3ZA7HUPHXgvOvNbjJYfcpr43EAaMnRaKGogF9DrgEdF5GkR2SAij4vIhqQ7ZlQPv5FnJlPovjnhBP8KiX7tDgwUukWg1C2USy8RxFUThjD9OjzrLSSPHOkv6W8m4+0uCkNz89h8jojzDMLIIq9766fD89gdLd7bc+SK77RntjGYbR0rvuOXabUWCeMPrGOCWAB/n3gvjEBUasDiNyo/7rhg1ys3qg+yEOz++5OJEmxrK3TL5GhpKe1XPx0soHTF7nOZjpLB66s29HJFf6m7COC2IldLU1NhIk1w3o+Ols6ZzJwZ/H69nvkV0sNKljNVxyyZYfFfdHb6QG/h+oRjezji8W3MHNnODmbRsW8zPT/+GN2H3Zt8pNBEiTrL32ChqONaAKq6CZgHnOG+3hPkPCNeKjlgiVod0O/8wcFguY+SihIcHPTevnNnab8up4fdResddzOVD4+UCs/LBr3dRdd61EzKFSDLfzaTJsHbtdCC+IeR3lA5obye+c/mdbOclTyDk2PoGebzblbS6zGHcfpAL5c8vpzZw86E8ezhTbzvseW8fbSXQWajNLGJ+Swf+S96t5+VTmsg7Cy/Fw0WijpuSUgR+ThwMtCpqn/plof8nqr+bSU6mE8jl4T0myzNZBwBUgk3ZlQLJGiVxXvu8Q95fcMbJt7fsL/tpZSGUN5Gd0lFyLtWN9Hkkc55FDmQ9jmfadNgd14c3VKcaJ3iNNXvZiW73jj+hLMf5SbYi1Nv3HT/As8cRc8wn6N4pmBbO5vZPPlo581E5waSMGfD3HCYY+sAv5KQQUby5wPn4YZ+quqzwMHxds8YDz/h5RV+mIRVEIcFUs4Hn0/URW/g3d+weGUOBSe88557HIV2zz2w2cfX/jwzPecFdhcFUftNOH9KolVdDTPI9ZsY9ppIHmAeO2l13pSrdOZHUuZs2OgDC0UNpAD2qWMmKICITIvr4iJyrju5/AcR+Uhc7dY6xROSAwPBQwKTSicdR+rqoL85vyzEYbITh1nMlutHMX4Tu/39hf76j3q4i15mMgfzYkkMvtfksF+0Trt6bw9KUIUL/hPDXhPJTU3w+9Yu//oE41Huy+T15Q9KmBuO6uesE4IogO+KyNeAVhF5N3AfcGPUC4tIM/BlnEnmY4GlInJs1HZrHb/BUVubt5DyIgk3Zhwu06R+c14yI+wzKO5Xzi0TRIDfRjfvLvK1v8jBTKEw74XfvMBEo3XGI8wg95bOHvY2FyqxYZnKFVI65zEt6hCw3JcpimUQdlTf3u64e974xtIFKg3CuFFAqvoZETkbeBEnb9WVqro2hmt3AX9Q1f8BEJHv4Kw4fiKGtmsWv8FRrvBKvtt0/37v3DpJLKKKa/V+kPDvcumki/HL2RM2IV1xv65d7T+xWxzZA44SyN8+4jO28hrtX05PyRxAuRTRQQmz3sovS+lGumnJO3/SpBi+X35fpuLwKAgXmWMLzEITJAwUV+DHIfTzmQvk57gcAF4T8zVqjnKDo2IhVZwiOodX5a2ohEldHRU/4R1m1XFTk9O/IG4grwJZfm4Zv+3F+IWReo32c4rj8y3hUkQHIcx6q3Xt3SXXbKfw/FiyQ/h9mfw+rLBmpgn8wJSrB7ALPEIbXFT1kIjX9kpjVnI9EVkOTlB1S0STuBYIM9L2C2v02x4l8CKtg6tyk+MnnljYX791AF6rq5/LdHBktlSAb2nqwCOwp4TL6eH/sryg/u+wTOVr83rw0iFPntjNu2IQ+DWB35cpv1JPPpYkLjHKpYM+GEBEPgE8B3wLR2h3E08U0ADO+oIc7cCzHv1YCawEJww0huummjAj7agpnsOmHK7U4CqMC6icwvTqb9BCVt8+rof3PbaclqJFVLcf30OHR6ZTKNz283ndfGVmqUvld+3dnBixmFZd4PdlqpSZaQDBXEDnqGq+a+a/RORXwKcjXvvXwDEichSwBXg78I6IbdY8YUbaUVM8p3XhY5jU10m5pspV7pqJY2UND8OUKWMrdostid6B0sIvOZdK8TOPWu4zKYqtxljmAPxIq5lZxwRRACMi0g18B8dFsxSYQCqsQlR1v4hcAtwLNANfV9WNUdutB4KOtJOyFqpNmHUAYWRGWCvIyycetI0w14pa7jMpyhYGSlIJmMCvGEEUwDuAL7h/CvycmEbqqvoj4EdxtNWIJGUtVJuwfQ0qM8ZbyxBnCpkw1/JTwkHLfSaF31qK3btJTgEYFSVIGOgzOOGZRgpJwlqoNMVuhra2wrKLEE9fxws/DzJaD2pJhbmWH2FWPidBuZKSRn0wrgIQkTbg3ZSWhFyWXLeMuEmLe3U8YT887LyfO3fMzx5XX+MIP/drI5MprF/gF8rqdS0/gpb7DErYEM6mJm9hP5vtY2+yWafhtGYHNcoSxAV0F/AznBXAkX3/RvWotnvVy6fsFZaZW/gWNCdX0PBWvzBQv5G2nxvKa3s2Oybw4xrVBy33OR59fWMpe8Jkb5g2DXbtKtx2EHt538G3uPK+y2k8VyugEgXlc1gxl1gIogCmquqHE++JUVd4RbV4pYP2I+jkdJjJ1q1bg7WZw2vh2c6d4dooxi/Cqbl5rCZAnFFA69Y5QlwYZS5b+NjzPbyt+c7A53+v+XyuGVnBFuYyly186uAeuk/fcmB/H10cRR8zskNOCclKWANxxDQbQDAFsFpEXu9O2BrGuJSLaglKsUsljroeYVJD+BHVL+93/ty58U/4/vjHMGlkLzAFpYkB5nHZyGeZxh66m28P1Mby5m+wvPkbYxteysKaDLS2smaoi2wWTsts447sEk7N9HFoJayBWoppTjlBFMD7gctFZB+wD2cxmMawEtioE6Lm3S9GpDDPURyTshMhmy1VQknht3p7IuRcPkeMbGaAQj/SHqaxYvJn6D5zgqZMrvGhIU7K9tHHmJDfnyVcytaJUksxzSknSBSQ5f43fCkbK+5BccqX5ubSCV+vJHdhJ2XjCm8tTkyZFHHJrjVr4I6sU9B9Dt4+r/7hWRO/QG5Uv2YNbQxyNmsYyM5hBkNMJgtknKLCSVJLMc0pJ0gUUC79w1Gqeo2IzAOOUNU40kIZCRF0jizqXFqYvPsipRlNva7nVznMr65HHOGtQRLHjYz4+/CjElV25Qbmq1jCItYwmSwdbGYT80uO7WjZEe1i4Aj5vj4mDw3RxpCzrbWtMpPAaY5prjGCuIC+gpP+6gzgGuAlnDz+pyTYLyMCSaxW9SPMyHXevGCRSGEGeGHCW/3CGpuaHCskf9I6jqpkfnhZQS0tY6Uww04C54T/QHYObZkhyACtbfQc+T2WP34Je0amHDh2avNeejpviX4TMCbsKx0GmpaY5jogiAJ4jar+tYj8FkBVd4rI5IT7ZUQgjtWqQX9L5eLdYWICLewAL2h4ay7SxostWwqrfPnhpxyKazOXU4zFVlBLC7zwwtj+iaSCaG2FSUOMxXl2ddHNOgBWPH0R/cOz6GjZQU/nLXS3rwvWaFCqsQag2jHNdUIQBZB1q3flSkK2ESghrlEt4lit6oWXu8iPSZPgnHPG76sXSQ3w/KKAgq5sLWcZZLOF9/vww4VCPcfMmaXbvI4DRwlEjQzqbl8Xv8A36oYgCuCLwJ3AHBHpAd4KXJFor4xIBHWhhHG1+LmL/PzmUUMuwwzwgs5jxOG/97N4ip/ZqaeWKoGZM6GjI9ykeSLYIirDJUgUUK+IPArk1mW+WVWfTLZbRhSCulDCuFr83EV+QrVSARlh5jHiiOH3sxa8qrCdemrptvvvD1esPnZsEZWRR5Ci8ABTcVI2N2F5AFNP0OLrYYq0+41UVcPV4Y6b8bJu5hOHUvIT3kHj+MOM+L1KVUYmzAMz6p4gYaBXAm8DfoCzCOwbIvI9Vf1k0p0zCgljuQd1oQQ9rpy7KL+aX6U9CmHmMfxyAflFB8XRj2LCLJRLJBW0LaIy8ggyB7AUOElV9wKIyKeA3wCmACpItS33cu6iagZkxFFDOajwL7dWIGjmTr/nmET2U8MYjyAK4BlgCrDXfX8Q8MekOmR4U+30J3FF5sQ9/xhHVbRy5BRMrq/r13sfF3R+Ie4Ip/yVv4dmhmCIcCk/GwGb9PYliAJ4GdgoImtxQkHPBh4SkS8CqOqlCfbPcEmD5R51pJ+EFRNHVbRyE9nFKalz1/E6Nkyfo8ofr5W/kBk3CZvi+HGDbq95wn7pGkxZBFEAd7p/OR5IpitGOeoh/UlSVkzUqmhz5wavQBZHFoKoMiZ/1D+DISa7K3858kin4dWrfRt+gUM5jNJEcM72OiTMl67aftYqECQM9GYRaQE6VNVCBWImqDCotfQnXvdVbSumvd2Jy89P+ZBLwzxzZrDPIaoLJ4qMyR/1n5Lp49ADOXhaHeEfoOFL+SI3sYwpjC1m2EuGS/kivQRLEQ3QO3B68iuM4yDMl67aftYqECQKaAnwGWAycJSInAh8QlXPS7hvdU8YYZCG9CdhEsx53VfQRVRx9cuv/GR+yoctWxzhH8YtE8WFE0XGdLkFuLKDsB8nN/vknCAP2PDPWxazbPgbXMsKOuinnw4up4eHW86GgAqgd+D0ghxDm4Zns/zxSwDSpwTCmM7VHqFUgSAuoKuALlzXj6quF5GjEuxTwxBWGFQz2iaMsvK7L78ImigKwK9fL7xQWmvYr/xkHAM8rwpoXmGcUWVMVxcs61vF0BCFRVgC1rXs6byF5Y9fwm0j3Qe2TW3ey8rOG4J1ACe3UH6COYA9I1NY8fRF6VMAZUznYivmd5lOpmeHStuoJT9rSIIsNdmvqn8u2pZAQtzGo5YGHGHWD4Xtf5Qyi3792rw5/vKTfuQqoOVbFv39zvZi/GRJGBnT1eVkY17WtoqHs10M0ub/gyxquLt9HSsX3sD8lu0Io8xv2c7KhTc4gntgwFmqvHq1839gwLNJv3oCkeoMJIXPasdeuln++CVsGp6N0sSm4dn0Zv/B+zl6LfOuE4JYAL8TkXcAzSJyDHAp8HCy3WoMamliN4yyClsVLEqKhnIrlIMS9Xlv3uy/vdgKiHMuJ98a+PXISRw/sr7wAJ+GPRPEhTDxOlp2sGl4dkm7sdQZSAIP03nF/aVWzDnc6x0JFWe5tpQRRAH8K7ACJxz0NuBenLoARkRqaWI3jLLyu6+oi6j8zo2iQESiP+8wtQPimMvJn9toYztf4Er+Sp5wHnIuYVHYhv1MqY0bSzqbcyMlVmegmARCM72slQ58Clen0SSPiSBRQHtwFMCK5LvTWKRhYjcoYZSV33298IK3H37evNJtQSkn/IuVjpeyiKPAi58S8lNsUeZyBgYKF6MNMpsP8VkObtpD92H3Tjw3v5+Qy2ZLijN3L+yFhSHqDEQR4AmFZnpZMf10sIBNpQen0SSPCV8FICKrKOPrtyigeKiVuhZxKKucOyTIZGlQwuQoevllb0G9cWO0z2DevPgVmx9e8woHCr13RZhMCeq3cyd+us8MWGcgqgBPKDTTy4q5Wq7iRpYzSfNC1dJqksdEOQvgM+7/C4DDgVvd90tx0kMYDUZQZVXuN3/88fEmOQuTo8iv1nDU2gVJKLZicmsAdHQUr9iNyBOwXg/SjzAukagCPKFIiZzyyrdizurcwSSOqw2TPCZ8FYCqrgMQkWtU9e/ydq0SkQejXFRE3oYTXvpKoEtVH4nSXqNQK6vUK7meJi1utLgVWz59ffD1QWfl78n00Z9EoXevB7l/v//CjaBfxqgCPMFICe9qaTViksdEkEngNhF5har+D4C7BiBqXNTvcCyLr0Vsp2GopVXqlQ5vDWqZ+C1Ey2Ti71MclKz8zQ5ybfPHWD7yVfYw9cBxiU3AHnGEd46MtrbgX8aoAryWIiVqkCAK4DLgARH5H/f9AuBfolw0V1FMooR/NBi1tEo9reGtxx0Hjz1WOA8g4mxPGznhP5CdM5bvp63N8fMPfCVaGga/4s7FQn3LFu881WG+jFEFeKVNvFoxs2MiSBTQT9z4/79yNz2lqi8n260xRGQ5sBygpaWjUpdNHbW0aCytg7ZyssRPJlZTFrS2wqQhmJxL7+xG+EQq9O5nSjY1eQv1wcHSlKh+ObG9voxxCPAkCkT7nVsrZnZMBLEAAF6NM/KfBJwgIqhqWZtTRO7DmTwuZoWq3hW0g6q6ElgJ0Np6csOuQE5qVB3m9xL0WL+ka2n4DXnJEq/f/WOPOa9z1kLNyoLiD23//nB5OsKs9PP7MlYq1C2lEUdpJkgyuG8BfwGsB3JPR4GyCkBVz4raOWOMJEbVYX4v5Y6FcEnX4iaq1e71u/cKF605WeD1oYXFy02bVhMvpRFHaSaIBXAycKxqHEtmjImShCs0zO+l3ELR0dHKJV0rJg6rPczvOw5ZMJ7Cyp/8BZw3E6ny5fWhhSWp5cxJkOKIo7QSKBcQjitna1wXFZHzgS/hRBPdIyLrVfWcuNqvV+K2pMP8XsotFI16vSjEYbWHyV0Uh8utnMLKD/mcwZCT7rmtbWIrfON44H4hUmlcwRhWgPvlC0+bZZMgQRTALOAJEenDyQcERFsJrKrFVcaMKhDm9xI2wZvf9eIOsojDavfyaOQ8H/kD4DhkgZ/C2rABnngiL8VzdtARvq0TFP4Qz4c2MuJkBk3TSN+PMK4pL03sF/WU1vuNgaD1AIwaI4ig7ez0DosMUwqxqSmYFRA2fDwoYWP7yz2XSkQB+WYvHR1lIHuEM+rPRhj159PW5u2PC8Po6Fin0z4THsY15aeJvaKe6pggYaApq/BgjEcli6+Df53dKOHjSTDecylXAjIu/Ablc9lCGxFH/V4RP3GT9AdWqTj8Bpzw9aJcMrhdeCeDE0BV9ZDEetXgJBHV4vW7ffpp7+yYE6lIFqS/YcLHg+JnfXhtr7YCAm9Laiq7+RQfiTbqjxLxI+L8jY4GOz4pIRl15BLm/Aac8PWiXC6ggyvZEcMhyaiW4u1xDYKCzgcm8ZurtZKvzz7r/G9jOzuYRQf99DRfSfdh90Vz+USN+GlvH1u4MR5JCcmoGrqSK5TrhKALwYwKMV7pxSAj7aBCsdKDoCR+c2HaDJu3P25y4Z1bR+cwQ4ojfCL6+6NoMVXHX3fCCdGiYqKarlE1dLnz/Say0xbKWmFMAaSMct/hoJZBUKHoN0fY1paMKzaJ31yYNsNU7koKv/QOkfHR5goFZQ6L3x9geNjblJs5M9jDjcN0jToiKRf15DeR3WACvxhTACnD7zssEty6DSoU/Uqdbt1aOPCLM/gjid9c0DarbQEkioc29xL2vrcatXxZHBMsUU3EoDUNam5Jd3KYAkgZfqGZfqNUvwFPkN9tmMVd9fCbSYMFkBge2jyUXov6EMqNvMMkkYKJm4he56dh4ifFmAKoEfzi3ZOYQPWj1n8zdR34EccqvTAUC/VyCzLCuIaimojF5+d8/8XUxYcendLackZV8QvNBMcazieOCdRiy1/EfxFVrf9mOjvjf4apwefDGS2yA7I0exf6bgtR4ynn78/3q+/fX/plyj3sclENSVPXH3p0zAJIGeXcMieeGH/qZi/8CkFV8jfj1/8o91XXgR8e/u/9kuEm/SfO4V466KefDqbzErN4vvR8vwkhL/zSp2YyMGlS4cNNYvFHGOr6Q4+OKYCUUW6iMqh1HDQgw8/aGByEhQu9V/16RdNFjRgaLydXrv8vvBB9crpuAz88BN2kzk6mcwyLnr7mQPWwPw3P8T4/Dl9gNgvnnFO4LdefYippTtbthx4dUwApI+xEpZfwDRqQUW5+rPg3MzBQODk9POy8jyqUvZSVXzppr3VK9TA5HRsegq6bouph98cwERJmMsUWXKUamwNIGX6/Q6/tXq7Y/PfFFG8Pc62NG72thf7+aC7eMAtYw0ZCGR7E4RMP00Z7u2NO5r5ULS3Oe9PYqcAsgJQRZsDkN9L3cyMVC/Yw1wqT9x+iL970Iuh9GWWIwycetg1zwaQWUwApI8xvyze1sDqCfDzBnuT8WFChXG6NQz65DKPVnpyuC+IQyCbU6wJTACkkjuRqubmAIGtvov6OgygbP8oJ/9z95fc/aGYCwzDGxxRADVPOhVPJAZpXxFAc6V+86nLYwNMw4sMUQA1TyRDnjg7v6JyOjmhC2YJEDKN6mAKocSo1Ij7+eOd/LhRTBObNG9s+UWydjmFUD1MARmCOPz66wPfC3DqGUR1sHYBhGEaDYgrAMAyjQTEFYBiG0aCYAjAMw2hQTAEYhmE0KKYADMMwGhRTAIZhGA1KVRSAiPyniDwlIhtE5E4Raa1GPwzDMBqZalkAa4FXqerxwH8DH61SPwzDMBqWqigAVV2jqvvdt78EbB2oUfcMDcG3hpZUuxuGcYA0pIJYBtzut1NElgPLAVpaOirVJ8OIjb4+R/ivYgmLsmuYTBaGMrB4cbW7ZjQ4iSkAEbkPONxj1wpVvcs9ZgWwH+j1a0dVVwIrAVpbTw5QOsQw0sOaNXBHdgld9NGWGYIM0NoGXV3V7pphJKcAVPWscvtF5J3AG4EzVYPUhDKM2qFg1I876sdG/Ua6qIoLSETOBT4MnK6qe6rRB8NIinzhfwp9TG5rdXbYqN9IGdWaA7gBOAhYKyIAv1TV91SpL4YRO62t0AocWuV+GEY5qqIAVPXoalzXMAzDGMNWAhuGYTQopgAMwzAaFFMAhmEYDYopAMMwjAbFFIBhGEaDYgrAMAyjQTEFYBiG0aCYAjAMw2hQTAEYhmE0KKYADMMwGhRTAIZhGA2KKQDDMIwGxRSAYSTA0BAMDsLOIffN0JCTJ9owUoQpAMOIma4up+7LsrZVPJztYjDbyr4sjkYwJWCkCFMAhpEQXV2OEmjPbGMti9mZaXOUwJo1pgiMVGAKwDASpMQaoM3ZYS4hIwVILZXjFZFBYFO1+5EAs4Ad1e5EQti91SZ2b7WJ373NV9W24o01pQDqFRF5RFVPrnY/ksDurTaxe6tNwt6buYAMwzAaFFMAhmEYDYopgHSwstodSBC7t9rE7q02CXVvNgdgGIbRoJgFYBiG0aCYAjAMw2hQTAGkBBH5TxF5SkQ2iMidItJa7T7FhYi8TUQ2isioiNRF+J2InCsiT4vIH0TkI9XuT1yIyNdFZLuI/K7afYkbEZknIj8VkSfd7+P7q92nOBCRKSLSJyKPufd1ddBzTQGkh7XAq1T1eOC/gY9WuT9x8jvgAuDBanckDkSkGfgy8PfAscBSETm2ur2KjW8C51a7EwmxH/iQqr4SeC3wvjr53F4GzlDVE4ATgXNF5LVBTjQFkBJUdY2q7nff/hJor2Z/4kRVn1TVp6vdjxjpAv6gqv+jqvuA7wBvqnKfYkFVHwReqHY/kkBVt6rqb9zXu4AngbnV7VV01OEl923G/QsU3WMKIJ0sA35c7U4YvswFNue9H6AOBEkjISILgJOAX1W5K7EgIs0ish7YDqxV1UD3NSnRXhkFiMh9wOEeu1ao6l3uMStwTNXeSvYtKkHurY4Qj20WT10jiMh04AfAB1T1xWr3Jw5UdQQ40Z07vFNEXqWq487jmAKoIKp6Vrn9IvJO4I3AmVpjCzTGu7c6YwCYl/e+HXi2Sn0xQiAiGRzh36uqd1S7P3GjqkMi8gDOPM64CsBcQClBRM4FPgycp6p7qt0foyy/Bo4RkaNEZDLwduDuKvfJGAcREeAm4ElVvb7a/YkLEWnLRQ2KSAtwFvBUkHNNAaSHG4CDgbUisl5EvlrtDsWFiJwvIgPA3wD3iMi91e5TFNzJ+kuAe3EmEr+rqhur26t4EJHbgF8AnSIyICLvqnafYuRvgQuBM9zf2HoReX21OxUDRwA/FZENOIOTtaq6OsiJlgrCMAyjQTELwDAMo0ExBWAYhtGgmAIwDMNoUEwBGIZhNCimAAzDMBoUUwBG4ojIS+MfVXD8IhEJFMaWBGH7W3TuxSJypM/224q2zRKRQRE5KGDbJ4vIFwNc/waffWE/hx4R2RzleRjpxhSAYcTLxUCJAgDuAM4Wkal5294K3K2qL4/XqIhMUtVHVPXSeLoZiFU4ie+MOsUUgFEx3JH9AyLyfbf2Qa+7OjOXX/8pEXkIJ3V07pxpbo76X4vIb0XkTe72i0XkLhH5iZuX/+N55/yjmx99vYh8zU3fjIi85I5qHxORX4rIHHf7USLyC/ca1xT1+f+42zfk8qyLyAI3p/yNbv71NSLSIiJvBU4Get1rt+TacXPOPAgsyWv+7cBtIrJERH7l3t99ef26SkRWisga4JZ8y0hEukTkYfech0WkM6/deV7PZbz7KkZVf6mqW8t8pEato6r2Z3+J/gEvuf8XAX/GyZ3ThLPi9HXAFJzsmsfgJFr7LrDaPeda4B/d1604tRKm4Yy0twKHAS04eU9OBl6JM3LNuOd8BbjIfa3AEvf1p4Er3Nd35x3zvrz+LsYpsi1uf1cDfwcswEnYd6J73Hfz+vgAcLLPc3gbcKf7+kic/EHNwKGMLcr8Z+Cz7uurgEeBlrznl3suhwCT3NdnAT9wX3s+l6LPwfO+xvv87K/+/iwZnFFp+lR1AMBNX7sAeAn4k6r+3t1+K7DcPX4xcJ6I/Jv7fgrQ4b5eq6rPu+fcgaNM9gOvBn7tGhctOClyAfbhCDtwBOvZ7uu/Bd7ivv4W8B95114M/NZ9Px1HSfW7/V2f19aCAPe+GviKiBwC/APwfVUdEZF24HYROQKYDPwp75y7VXXYo60ZwM0icgyOYsvk7fN6Lo/k7fe7r7oo2GMExxSAUWny/d0jjH0H/XKSCPAWLSooIyKv8ThH3eNvVlWvimpZVc2dk39tv+sLcJ2qfq3o2gs87qOFcVDVYRH5CXA+jvvnMnfXl4DrVfVuEVmEM/LPsdunuWuAn6rq+W5/HihzL8XvPe/LaDxsDsBIA08BR4nIX7jvl+btuxf417y5gpPy9p0tIjNdX/ubgZ8D9wNvFZHZ7vEzRWT+ONf/OY5ABuguuvYycfLHIyJzc+2WYRdOUj8/bgM+CMzBqfwGzmh+i/v6neO0nyP/nIuL9nk9l3wmcl9GHWIKwKg6qroXx+VzjzsJvClv9zU47o0N4hQqz5+kfQjHZbMexwf+iKo+AVwBrBEnO+JanGyJ5Xg/Tn3YX+MI1ly/1gDfBn4hIo8D36e8cAenpu5XiyeB81iD4/+/Pc8auQr4noj8DNgxTvs5Pg1cJyI/x5lHyKfkueTvDHpfIvJpcbK4ThUnM+hVAftm1AiWDdSoSUTkYpzJzUuq3RfDqFXMAjAMw2hQzAIwDMNoUMwCMAzDaFBMARiGYTQopgAMwzAaFFMAhmEYDYopAMMwjAbl/wNPfGXRla6mMwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualising the Training set results\n",
    "from matplotlib.colors import ListedColormap\n",
    "#Define Variables\n",
    "clf = classifier\n",
    "h = 0.1\n",
    "X_plot, z_plot = X_train, y_train \n",
    "\n",
    "#Template for all Regression\n",
    "x_min, x_max = X_plot[:, 0].min() - 1, X_plot[:, 0].max() + 1\n",
    "y_min, y_max = X_plot[:, 1].min() - 1, X_plot[:, 1].max() + 1\n",
    "xx, yy = np.meshgrid(np.arange(x_min, x_max, h),\n",
    "                     np.arange(y_min, y_max, h))\n",
    "\n",
    "\n",
    "# Plot the decision boundary. For that, we will assign a color to each\n",
    "# point in the mesh\n",
    "Z = clf.predict(np.array([xx.ravel(), yy.ravel()]).T)\n",
    "Z = Z.reshape(xx.shape)\n",
    "plt.contourf(xx, yy, Z,\n",
    "             alpha = 0.7, cmap = ListedColormap(('blue', 'red')))\n",
    "\n",
    "\n",
    "for i, j in enumerate(np.unique(z_plot)):\n",
    "    plt.scatter(X_plot[z_plot == j, 0], X_plot[z_plot == j, 1],\n",
    "                c = ['blue', 'red'][i], cmap = ListedColormap(('blue', 'red')), label = j)\n",
    "   #X[:, 0], X[:, 1] \n",
    "plt.xlim(xx.min(), xx.max())\n",
    "plt.ylim(yy.min(), yy.max())\n",
    "plt.title('K Nearest Neighbour')\n",
    "plt.xlabel('Independent Variable 1')\n",
    "plt.ylabel('Independent Variable 2')\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Minimum error:- 0.07 at K = 2\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Determining optimum value of K\n",
    "#Choose correct value of K\n",
    "from matplotlib.colors import ListedColormap\n",
    "import numpy as np\n",
    "error_value = []\n",
    "for i in range(1,40):\n",
    " classifier = KNeighborsClassifier(n_neighbors=i)\n",
    " classifier.fit(X_train,y_train)\n",
    " i_pred = classifier.predict(X_test)\n",
    " error_value.append(np.mean(i_pred != y_test))\n",
    "\n",
    "plt.figure(figsize=(10,6))\n",
    "plt.plot(range(1,40),error_value,color='green', linestyle='dashed', \n",
    "         marker='o',markerfacecolor='orange', markersize=10)\n",
    "plt.title('Error Value vs. K Value')\n",
    "plt.xlabel('K')\n",
    "plt.ylabel('Error Value')\n",
    "print(\"Minimum error:-\",min(error_value),\"at K =\",error_value.index(min(error_value)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1\n"
     ]
    }
   ],
   "source": [
    "#K-NN - Implement from scratch\n",
    "import pandas as pd\n",
    "from sklearn.datasets import load_iris\n",
    "iris = load_iris()\n",
    "iris.feature_names\n",
    "iris.target_names\n",
    "\n",
    "df = pd.DataFrame(iris.data, columns=iris.feature_names)\n",
    "df['type'] = iris.target\n",
    "\n",
    "import numpy as np\n",
    "class MyKNN:\n",
    "    def __init__(self,k=5):\n",
    "        self.k = k\n",
    "    \n",
    "    #convert and assign feature and target data to variable   \n",
    "    def my_fit(self, feature_data, target_data):\n",
    "        self.feature_data = np.array(feature_data)\n",
    "        self.target_data = np.array(target_data)\n",
    "     \n",
    "    #Calculate Euclidean Distance, axis-> 1: Column, 0:row    \n",
    "    #[3.5, 2.5, 3.4, 2.3] - [1,2,3,4] - like this for all rows\n",
    "    def calculate_distance_vector_matrix(self, one_data):\n",
    "        distances = np.sqrt(np.sum(np.square(self.feature_data - one_data), axis = 1))\n",
    "        return distances\n",
    "    \n",
    "    #Sort the distance and take top k=5 which are closest\n",
    "    #argsort() will give index of the above values\n",
    "    def find_k_neighbours(self,one_data_feature):\n",
    "        res = self.calculate_distance_vector_matrix(one_data_feature)\n",
    "        return res.argsort()[:self.k]\n",
    "    \n",
    "    #find the index of the closest    \n",
    "    def find_k_neighbours_class(self, one_data_feature):\n",
    "        indexs_of_neighbours = self.find_k_neighbours(one_data_feature)\n",
    "        return self.target_data[indexs_of_neighbours]\n",
    "    \n",
    "    def my_predict(self, one_data_feature):\n",
    "        classes = self.find_k_neighbours_class(one_data_feature)\n",
    "        return np.bincount(classes).argmax()\n",
    "    \n",
    "model = MyKNN(k=5)\n",
    "df.columns\n",
    "feature_data = df.drop(columns=['type'],axis=1)\n",
    "target_data = df.type\n",
    "model.my_fit(feature_data, target_data)\n",
    "#print(feature_data)\n",
    "\n",
    "one_data = [1,2,3,4]\n",
    "model.find_k_neighbours_class(one_data)\n",
    "print(model.my_predict([1,2,3,4]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}