KNN_Classifier.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Welcome To Colaboratory",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/codingCapricorn/KNN_Classifier/blob/master/KNN_Classifier.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ax8zw-3D35Ht",
        "colab_type": "text"
      },
      "source": [
        "1.KNN Distance Algo -->>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CfH1xeFw1bJY",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "71a1f88e-c2d6-4dab-a797-df46644a7410"
      },
      "source": [
        "import numpy as np\n",
        "def distance(p1,p2):\n",
        "    return np.sqrt(np.sum(np.power(p2-p1,2))) #distance between two points\n",
        "p1 = np.array([1, 1])   #coordinate x=1,y=1\n",
        "p2 = np.array([4, 4])   #coordinate x=4,y=4\n",
        "distance(p1,p2)"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "4.242640687119285"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 1
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XqeKgUbg4Isk",
        "colab_type": "text"
      },
      "source": [
        "2.kNN_majority_vote+nearest_neighbour -->>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aD-yLDU71ey8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 303
        },
        "outputId": "a3b7841f-75fe-4565-eb26-9c142afb43b4"
      },
      "source": [
        "import numpy as np\n",
        "import random\n",
        "import scipy.stats as ss\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "points=np.array([[1,1],[1,2],[1,3],[2,1],[2,2],[2,3],[3,1],[3,2],[3,3]])\n",
        "p=np.array([2.5,2])\n",
        "\n",
        "def majority_vote(votes):\n",
        "    vote_counts = {}\n",
        "    for vote in votes:\n",
        "        if vote in vote_counts:\n",
        "           vote_counts[vote]+=1\n",
        "        else:\n",
        "            vote_counts[vote]=1\n",
        "    winners = []\n",
        "    max_count = max(vote_counts.values())\n",
        "    for vote, count in vote_counts.items():\n",
        "        if count == max_count:\n",
        "            winners.append(vote)\n",
        "    return random.choice(winners)    #returns winner randomly if there are more than 1 winner\n",
        "\n",
        "# >>>votes=[1,2,3,2,2,3,1,1,1,2,3,1,1,1,2,3,3,3,2,2,2,3,2,3,1,1,2] #sample vote counts\n",
        "# >>>vote_counts=majority_vote(votes)\n",
        "def majority_vote_short(votes):\n",
        "    mode,count = ss.mstats.mode(votes)\n",
        "    return mode\n",
        "\n",
        "def find_nearest_neighbours(p,points,k=5):  #algorithm to find the nearest neighbours\n",
        "    distances=np.zeros(points.shape[0])\n",
        "    for i in range(len(distances)):\n",
        "        distances[i]=distance(p,points[i])\n",
        "    ind = np.argsort(distances)            #returns index,according to sorted values in array\n",
        "    return ind[:k]\n",
        "\n",
        "ind=find_nearest_neighbours(p,points,2);print(points[ind]) #gives the nearest neighbour's for this sample case\n",
        "\n",
        "plt.plot(points[:,0],points[:,1],\"ro\")\n",
        "plt.plot(p[0],p[1], \"bo\")\n",
        "plt.axis([0.5,3.5,0.5,3.5])\n",
        "plt.show()"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[2 2]\n",
            " [3 2]]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UR9w1_Sg4SDC",
        "colab_type": "text"
      },
      "source": [
        "3.kKK_predict+synthetic_points -->>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Xgw2eWsk1pSW",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "outputId": "c3d4ed01-bd61-4318-8619-674d5713bb82"
      },
      "source": [
        "import numpy as np\n",
        "import random\n",
        "import scipy.stats as ss\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "''' add the functions and libraries from previous programmes '''\n",
        "\n",
        "def knn_predict(p,points,outcomes,k=5):\n",
        "    ind=find_nearest_neighbours(p,points,k)\n",
        "    return majority_vote(outcomes[ind])\n",
        "\n",
        "outcomes=np.array([0,0,0,0,1,1,1,1,1])\n",
        "knn_predict(np.array([2.5,2.7]),points,outcomes,k=2)\n",
        "\n",
        "def generate_synth_data(n=50):\n",
        "    points = np.concatenate((ss.norm(0,1).rvs((n,2)),ss.norm(1,1).rvs((n,2))),axis=0)\n",
        "    outcomes = np.concatenate((np.repeat(0,n),np.repeat(1,n)))\n",
        "    return (points,outcomes)\n",
        "\n",
        "n=20\n",
        "plt.figure()\n",
        "plt.plot(points[:n,0], points[:n,1], \"ro\")\n",
        "plt.plot(points[n:,0], points[n:,1], \"bo\")\n",
        "plt.show()"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZdhjyaMc4XEC",
        "colab_type": "text"
      },
      "source": [
        "4.kNN_prediction_grids -->>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QyPzSdzi1zIW",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 589
        },
        "outputId": "e94aad74-d72c-4ab8-d65a-aaadc3ec8185"
      },
      "source": [
        "import numpy as np\n",
        "import random\n",
        "import scipy.stats as ss\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "def make_prediction_grid(predictors,outcomes,limits,h,k):\n",
        "    (x_min,x_max, y_min,y_max) = limits\n",
        "    xs=np.arange(x_min,x_max,h)\n",
        "    ys = np.arange(y_min,y_max, h)\n",
        "    xx, yy = np.meshgrid(xs,ys)\n",
        "\n",
        "    prediction_grid = np.zeros(xx.shape, dtype = int)\n",
        "    for i,x in enumerate(xs):\n",
        "        for j,y in enumerate(ys):\n",
        "            p=np.array([x,y])\n",
        "            prediction_grid[j,i] = knn_predict(p,predictors,outcomes,k)\n",
        "    return (xx,yy,prediction_grid)\n",
        "\n",
        "def plot_prediction_grid (xx, yy, prediction_grid, filename):\n",
        "    \"\"\" Plot KNN predictions for every point on the grid.\"\"\"\n",
        "    from matplotlib.colors import ListedColormap\n",
        "    background_colormap = ListedColormap ([\"hotpink\",\"lightskyblue\", \"yellowgreen\"])\n",
        "    observation_colormap = ListedColormap ([\"red\",\"blue\",\"green\"])\n",
        "    plt.figure(figsize =(10,10))\n",
        "    plt.pcolormesh(xx, yy, prediction_grid, cmap = background_colormap, alpha = 0.5)\n",
        "    plt.scatter(predictors[:,0], predictors [:,1], c = outcomes, cmap = observation_colormap, s = 50)\n",
        "    plt.xlabel('Variable 1'); plt.ylabel('Variable 2')\n",
        "    plt.xticks(()); plt.yticks(())\n",
        "    plt.xlim (np.min(xx), np.max(xx))\n",
        "    plt.ylim (np.min(yy), np.max(yy))\n",
        "    plt.savefig(filename)\n",
        "\n",
        "(predictors, outcomes) = generate_synth_data()\n",
        "# >>>predictors.shape\n",
        "# >>>outcomes.shape\n",
        "k=5; filename=\"knn_synth_5.pdf\"; limits=(-3,4,-3,4); h=0.1\n",
        "(xx,yy,prediction_grid) = make_prediction_grid(predictors,outcomes,limits,h,k)\n",
        "plot_prediction_grid(xx,yy,prediction_grid,filename)\n",
        "plt.show()"
      ],
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "U2qYC-kW4cqz",
        "colab_type": "text"
      },
      "source": [
        "5.kNN_scikitlearn_vS_organic_on_IRIS -->>\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "--UeEi2x13x8",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 908
        },
        "outputId": "4d21571f-dfb0-4489-f844-b4c77fb931ed"
      },
      "source": [
        "from sklearn import datasets\n",
        "import numpy as np\n",
        "import random\n",
        "import matplotlib.pyplot as plt\n",
        "'''\n",
        "        We'll be applying both the SciKitLearn and our homemade classifier to a classic data set created by Ron Fisher in 1933.\n",
        "        It consists of 150 different iris flowers.50 from each of three different species.For each flower, we have the following\n",
        "        covariates: sepal length, sepal width,petal length, and petal width.\n",
        " '''\n",
        "iris=datasets.load_iris()\n",
        "# >>>iris[\"data\"]\n",
        "predictors = iris.data[:, 0:2]\n",
        "outcomes = iris.target\n",
        "\n",
        "plt.plot(predictors[outcomes==0][:,0], predictors[outcomes==0][:,1], \"ro\")\n",
        "plt.plot(predictors[outcomes==1][:,0], predictors[outcomes==1][:,1], \"go\")\n",
        "plt.plot(predictors[outcomes==2][:,0], predictors[outcomes==2][:,1], \"bo\")\n",
        "\n",
        "k=5; filename=\"iris_grid.pdf\"; limits=(4,8,1.5,4.5); h=0.1\n",
        "(xx,yy,prediction_grid) = make_prediction_grid(predictors,outcomes,limits,h,k)\n",
        "plot_prediction_grid(xx,yy,prediction_grid,filename)\n",
        "plt.show()\n",
        "\n",
        "from sklearn.neighbors import KNeighborsClassifier #predictions from skikit\n",
        "knn = KNeighborsClassifier(n_neighbors=5)\n",
        "knn.fit(predictors,outcomes)\n",
        "sk_predictions  = knn.predict(predictors)\n",
        "\n",
        "my_predictions = np.array([knn_predict(p,predictors,outcomes,5) for p in predictors])\n",
        "\n",
        "# >>>sk_predictions == my_predictions\n",
        "# >>>np.mean(sk_predictions == my_predictions)\n",
        "print(\" prediction by scikit learn : \")\n",
        "print(100*np.mean(sk_predictions == outcomes))\n",
        "print(\" prediction by own model : \")\n",
        "print(100*np.mean(my_predictions == outcomes))\n",
        "\n",
        "# our homemade predicter is actually somewhat better"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            " prediction by scikit learn : \n",
            "83.33333333333334\n",
            " prediction by own model : \n",
            "84.66666666666667\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iWjBbqf52L-j",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}