clusterMetrics.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "clusterMetrics.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyO3tH0fbEL36ZtblD5Pxpvw",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/burakugurr/AI-ML-DL/blob/master/clusterMetrics.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FZKGbw23UN2f"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "from sklearn.datasets import make_blobs\n",
        "from sklearn.cluster import KMeans\n",
        "import pandas as pd \n",
        "from sklearn import metrics\n",
        "from sklearn.metrics import silhouette_score\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "import numpy as np\n"
      ],
      "execution_count": 118,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "thea2RSxUijV"
      },
      "source": [
        "features, true_labels = make_blobs(\n",
        "    n_samples=2000,\n",
        "    centers=3,\n",
        "    cluster_std=1.55,\n",
        "    random_state=42\n",
        ")\n",
        "datas = {\n",
        "    'X1':features[:,0],\n",
        "    'X2':features[:,1],\n",
        "    'label':true_labels\n",
        "}"
      ],
      "execution_count": 85,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 343
        },
        "id": "Q1The66UU43B",
        "outputId": "226f9ee8-1c40-4bce-f547-a4950647cced"
      },
      "source": [
        "data = pd.DataFrame(data=datas,columns=['X1','X2','label'])\n",
        "data.head(10)"
      ],
      "execution_count": 122,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>X1</th>\n",
              "      <th>X2</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>4.440646</td>\n",
              "      <td>-0.943696</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5.479146</td>\n",
              "      <td>3.149860</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2.545714</td>\n",
              "      <td>-0.076640</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>-6.963180</td>\n",
              "      <td>-8.402680</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>-3.792720</td>\n",
              "      <td>-8.169805</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>-5.119416</td>\n",
              "      <td>-8.594865</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>-9.000516</td>\n",
              "      <td>-3.799832</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>3.905428</td>\n",
              "      <td>1.950769</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>-3.794140</td>\n",
              "      <td>8.146006</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>-5.051053</td>\n",
              "      <td>-6.579243</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         X1        X2  label\n",
              "0  4.440646 -0.943696      1\n",
              "1  5.479146  3.149860      1\n",
              "2  2.545714 -0.076640      1\n",
              "3 -6.963180 -8.402680      2\n",
              "4 -3.792720 -8.169805      2\n",
              "5 -5.119416 -8.594865      2\n",
              "6 -9.000516 -3.799832      2\n",
              "7  3.905428  1.950769      1\n",
              "8 -3.794140  8.146006      0\n",
              "9 -5.051053 -6.579243      2"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 122
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "gETeONR1cPWu",
        "outputId": "494e2493-0f6d-4448-9fad-7d4c75877701"
      },
      "source": [
        "plt.scatter(data.iloc[:, 0], data.iloc[:, 1],c=['#9467bd'],marker='X');"
      ],
      "execution_count": 87,
      "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": "TfV-cZpKit7a"
      },
      "source": [
        "* Verileri kendimiz yarattığımız için harici kaynaklar sınıfında bulunan değerlendirme metriklerine bakmamız gerekmektedir."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aMWvz7GpU6QP"
      },
      "source": [
        "scaler = StandardScaler()\n",
        "scaled_features = scaler.fit_transform(data.iloc[:,:2])"
      ],
      "execution_count": 88,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ItRdrYlLjorG"
      },
      "source": [
        "## 1-KMeans"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "twxx1yENU7qq"
      },
      "source": [
        "kmeans = KMeans(\n",
        "    init=\"k-means++\",\n",
        "    n_clusters=3,\n",
        "    n_init=10,\n",
        "    max_iter=5000,\n",
        "    random_state=42\n",
        ").fit(scaled_features)\n",
        "y_kmeans = kmeans.predict(scaled_features)"
      ],
      "execution_count": 95,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "UM7vKWuMW2ie",
        "outputId": "427be717-cecb-4930-e7b4-006703dfcd2a"
      },
      "source": [
        "plt.scatter(data.iloc[:, 0], data.iloc[:, 1], c=y_kmeans, cmap='viridis')\n",
        "\n",
        "centers = kmeans.cluster_centers_\n",
        "plt.scatter(centers[:, 0], centers[:, 1], c='black', s=200, alpha=0.5);"
      ],
      "execution_count": 96,
      "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": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cnxHc0vuevmL",
        "outputId": "dc213139-c678-4564-8f80-e6754acfe1d2"
      },
      "source": [
        "kmeans_score = kmeans.score(scaled_features,y_kmeans)\n",
        "kmeans_score"
      ],
      "execution_count": 97,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "-295.6866707456031"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 97
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IIrv-JTInyrN"
      },
      "source": [
        "### Mutual Information based Score"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5_ItXCrhn0sn",
        "outputId": "81b228a6-f009-49f5-f2c1-fb8712a93f9f"
      },
      "source": [
        "metrics.adjusted_mutual_info_score(true_labels, y_kmeans)   #AMI"
      ],
      "execution_count": 100,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9965822161626248"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 100
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9nwgpyPZn-f-",
        "outputId": "33e08361-0c19-4374-aa21-b569b3f5d5cb"
      },
      "source": [
        "metrics.normalized_mutual_info_score(true_labels, y_kmeans)   #NMI"
      ],
      "execution_count": 102,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9965853313238792"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 102
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VEmushocrdKe"
      },
      "source": [
        "### Rand Index"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iE6S9-sFoGSy",
        "outputId": "11edc11b-ca5b-4779-bf56-a97b086b4114"
      },
      "source": [
        "metrics.cluster.adjusted_rand_score(true_labels, y_kmeans)   "
      ],
      "execution_count": 103,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.998498874063205"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 103
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "r0e0O6pYsSu3"
      },
      "source": [
        "### Fowlkes-Mallows Score"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KdS8o1mKrkbT",
        "outputId": "23ef348b-fdb9-4ac0-9d47-b06eac4cd65d"
      },
      "source": [
        "metrics.cluster.fowlkes_mallows_score(true_labels, y_kmeans)   "
      ],
      "execution_count": 105,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9989987493759085"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 105
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eMqZ21Yvsdox"
      },
      "source": [
        "### Precision"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jvNif1-MsXhv",
        "outputId": "37d77123-60dd-4254-dad1-207db681be7f"
      },
      "source": [
        "metrics.precision_score(true_labels, y_kmeans,pos_label='positive',average='micro')"
      ],
      "execution_count": 114,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/metrics/_classification.py:1321: UserWarning: Note that pos_label (set to 'positive') is ignored when average != 'binary' (got 'micro'). You may use labels=[pos_label] to specify a single positive class.\n",
            "  % (pos_label, average), UserWarning)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.0005"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 114
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vSzZYiaAuucV"
      },
      "source": [
        "### Recall"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "x2FlfP6kuFKu",
        "outputId": "95c240a3-495e-4ad3-d26a-bc707504a191"
      },
      "source": [
        "metrics.recall_score(y_kmeans,true_labels ,pos_label='positive',average='micro')"
      ],
      "execution_count": 116,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/metrics/_classification.py:1321: UserWarning: Note that pos_label (set to 'positive') is ignored when average != 'binary' (got 'micro'). You may use labels=[pos_label] to specify a single positive class.\n",
            "  % (pos_label, average), UserWarning)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.0005"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 116
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZEQE7AFSwJ56"
      },
      "source": [
        "### Purity"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "k4z_D--zuxsX",
        "outputId": "91502859-a246-4d4a-880a-821b587c3797"
      },
      "source": [
        "def purity_score(y_true, y_pred):\n",
        "    contingency_matrix = metrics.cluster.contingency_matrix(y_true, y_pred)\n",
        "    return np.sum(np.amax(contingency_matrix, axis=0)) / np.sum(contingency_matrix) \n",
        "\n",
        "purity_score(true_labels,y_kmeans)  "
      ],
      "execution_count": 119,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9995"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 119
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vmaBwtFx2UJK"
      },
      "source": [
        "### Jaccard similarity coefficient score"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "j3d2JLd31Ydu",
        "outputId": "a385eba7-6f4f-47fa-9eb2-695585e491b2"
      },
      "source": [
        "metrics.jaccard_score(true_labels,y_kmeans,pos_label='positive',average='micro')"
      ],
      "execution_count": 124,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/metrics/_classification.py:1321: UserWarning: Note that pos_label (set to 'positive') is ignored when average != 'binary' (got 'micro'). You may use labels=[pos_label] to specify a single positive class.\n",
            "  % (pos_label, average), UserWarning)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.00025006251562890725"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 124
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Fx1MWruF2Zza"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}