KNN算法实现.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import operator\n",
    "\n",
    "# 已知分类的数据\n",
    "x1 = np.array([3, 2, 1])\n",
    "y1 = np.array([104, 100, 81])\n",
    "x2 = np.array([101, 99, 98])\n",
    "y2 = np.array([10, 5, 2])\n",
    "scatter1 = plt.scatter(x1, y1, c='r')\n",
    "scatter2 = plt.scatter(x2, y2, c='b')\n",
    "\n",
    "# 未知数据\n",
    "x = np.array([18])\n",
    "y = np.array([90])\n",
    "scatter3 = plt.scatter(x, y, c='k')\n",
    "\n",
    "\n",
    "def gaussian(dist, sigma=10.0):\n",
    "    \"\"\" Input a distance and return it`s weight\"\"\"\n",
    "    weight = np.exp(-dist ** 2 / (2 * sigma ** 2))\n",
    "    return weight\n",
    "\n",
    "\n",
    "#画图例\n",
    "plt.legend(handles=[scatter1, scatter2, scatter3], labels=['labelA', 'labelB', 'X'], loc='best')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 已知分类的数据\n",
    "x_data = np.array([[3, 104],\n",
    "                   [2, 100],\n",
    "                   [1, 81],\n",
    "                   [101, 10],\n",
    "                   [99, 5],\n",
    "                   [81, 2]])\n",
    "y_data = np.array(['A', 'A', 'A', 'B', 'B', 'B'])\n",
    "x_test = np.array([18, 90])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 计算样本数量\n",
    "x_data_size = x_data.shape[0]\n",
    "x_data_size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[18, 90],\n",
       "       [18, 90],\n",
       "       [18, 90],\n",
       "       [18, 90],\n",
       "       [18, 90],\n",
       "       [18, 90]])"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 复制x_test\n",
    "np.tile(x_test, (x_data_size, 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 15, -14],\n",
       "       [ 16, -10],\n",
       "       [ 17,   9],\n",
       "       [-83,  80],\n",
       "       [-81,  85],\n",
       "       [-63,  88]])"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 计算x_test与每一个样本的差值\n",
    "diffMat = np.tile(x_test, (x_data_size, 1)) - x_data\n",
    "diffMat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 225,  196],\n",
       "       [ 256,  100],\n",
       "       [ 289,   81],\n",
       "       [6889, 6400],\n",
       "       [6561, 7225],\n",
       "       [3969, 7744]])"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 计算差值的平方\n",
    "sqDiffMat = diffMat ** 2\n",
    "sqDiffMat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([  421,   356,   370, 13289, 13786, 11713])"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 求和\n",
    "sqDistances = sqDiffMat.sum(axis=1)\n",
    "sqDistances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 20.51828453,  18.86796226,  19.23538406, 115.27792503,\n",
       "       117.41379817, 108.2266141 ])"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 开方\n",
    "distances = sqDistances ** 0.5\n",
    "distances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 2, 0, 5, 3, 4], dtype=int64)"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 从小到大排序\n",
    "sortedDistances = distances.argsort()\n",
    "sortedDistances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.99501248, 0.98019867, 1.        , 0.8824969 , 0.95599748,\n",
       "       0.92311635])"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 求权\n",
    "distancesByWeight = gaussian(sortedDistances)\n",
    "distancesByWeight"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "classCount = {}\n",
    "# 设置k\n",
    "k = 5\n",
    "for i in range(k):\n",
    "    # 获取标签\n",
    "    votelabel = y_data[sortedDistances[i]]\n",
    "    # 统计标签数量\n",
    "    classCount[votelabel] = classCount.get(votelabel, 0) + distancesByWeight[i]*1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'A': 2.975211152499438, 'B': 1.8384943844176953}"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "classCount"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('A', 2.975211152499438), ('B', 1.8384943844176953)]"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 根据operator.itemgetter(1)-第1个值对classCount排序，然后再取倒序\n",
    "sortedClassCount = sorted(classCount.items(), key=operator.itemgetter(1), reverse=True)\n",
    "sortedClassCount"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'A'"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 获取数量最多的标签\n",
    "knnclass = sortedClassCount[0][0]\n",
    "knnclass"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "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": "3.10.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}