23 - Sklearn Feature Selection.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Untitled5.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyPR1U8M7FFMUXY3LrCtpzUF",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/MachineLearnia/Python-Machine-Learning/blob/master/23%20-%20Sklearn%20Feature%20Selection.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TXrkkQh-w7ug",
        "colab_type": "text"
      },
      "source": [
        "# 23/30 Sklearn : Feature Selection"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "89idlZn1xJr9",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qo4eBVStxB1N",
        "colab_type": "text"
      },
      "source": [
        "## 1. Variance Threshold"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mZtVwaMYvwgs",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.feature_selection import VarianceThreshold"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cUnJ1Ce7xNWG",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "outputId": "ab740c95-6be0-455c-dfe5-cb548de3081a"
      },
      "source": [
        "from sklearn.datasets import load_iris\n",
        "\n",
        "iris = load_iris()\n",
        "X = iris.data\n",
        "y = iris.target\n",
        "\n",
        "plt.plot(X)\n",
        "plt.legend(iris.feature_names)"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f58916182e8>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 3
        },
        {
          "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": {
        "id": "N16z31mtxa70",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "f044a7a5-d62d-4d17-e419-bd276d81b998"
      },
      "source": [
        "X.var(axis=0)"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0.68112222, 0.18871289, 3.09550267, 0.57713289])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "v22pDx0UxQsd",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "dd038702-1f5c-4153-f06b-116e52c4558e"
      },
      "source": [
        "selector = VarianceThreshold(threshold=0.2)\n",
        "selector.fit(X)"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "VarianceThreshold(threshold=0.2)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mq6ci-8IxSW9",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "45e67e2f-f6e4-46a1-d64d-195bd801149e"
      },
      "source": [
        "selector.get_support()"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ True, False,  True,  True])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sJ3cxjQExU2h",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "c86d48b2-b3e0-4082-b599-d4b53e2d4f77"
      },
      "source": [
        "np.array(iris.feature_names)[selector.get_support()]"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['sepal length (cm)', 'petal length (cm)', 'petal width (cm)'],\n",
              "      dtype='<U17')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Mm-aEqZKxWOF",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "285fc652-b53c-40c7-d480-e0a81d8706d3"
      },
      "source": [
        "selector.variances_"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0.68112222, 0.18871289, 3.09550267, 0.57713289])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7znIa48Sxfu5",
        "colab_type": "text"
      },
      "source": [
        "## 2. SelectKBest"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "G4g8vA2oxcV8",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.feature_selection import SelectKBest, chi2, f_classif"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tBCK6Bb6xxUK",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "f029bed8-de65-474e-adbf-4c2aa0eb738b"
      },
      "source": [
        "chi2(X, y)"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([ 10.81782088,   3.7107283 , 116.31261309,  67.0483602 ]),\n",
              " array([4.47651499e-03, 1.56395980e-01, 5.53397228e-26, 2.75824965e-15]))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VypJt69yxyQx",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "38cf9486-d8f3-46ba-8f7d-849c3ac059d1"
      },
      "source": [
        "selector = SelectKBest(f_classif, k=2)\n",
        "selector.fit(X, y)\n",
        "selector.scores_"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ 119.26450218,   49.16004009, 1180.16118225,  960.0071468 ])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yk22XF0cx01W",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "fc338097-ecfd-4b69-84f3-636346a2d051"
      },
      "source": [
        "np.array(iris.feature_names)[selector.get_support()]"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['petal length (cm)', 'petal width (cm)'], dtype='<U17')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "06PFdmouyCDi",
        "colab_type": "text"
      },
      "source": [
        "## 3. Recursive feature Elimination"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ci09C91vx8Ld",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.feature_selection import RFECV\n",
        "from sklearn.linear_model import SGDClassifier"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_WqGRXy2yA-I",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "e993a244-2443-4742-b875-bac2d4a68016"
      },
      "source": [
        "selector = RFECV(SGDClassifier(random_state=0), step=1, min_features_to_select=2, cv=5)\n",
        "selector.fit(X, y)\n",
        "print(selector.ranking_)\n",
        "print(selector.grid_scores_)"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[2 1 1 1]\n",
            "[0.8        0.84666667 0.77333333]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MCR4s_LbyHq3",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "6afd2a99-5ee3-47ba-8aab-6364e61bacbf"
      },
      "source": [
        "np.array(iris.feature_names)[selector.get_support()]"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['sepal width (cm)', 'petal length (cm)', 'petal width (cm)'],\n",
              "      dtype='<U17')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OTkfkAQAya9x",
        "colab_type": "text"
      },
      "source": [
        "## 4. SelectFromModel"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TcWx2_H0yWhN",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.feature_selection import SelectFromModel"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KutnODU6ydjo",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "60cd8b83-6d42-41c3-a8a5-ab26aa8b99bc"
      },
      "source": [
        "X = iris.data\n",
        "y = iris.target\n",
        "selector = SelectFromModel(SGDClassifier(random_state=0), threshold='mean')\n",
        "selector.fit(X, y)\n",
        "selector.estimator_.coef_"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[  8.64029104,  27.2851296 , -40.01819009, -17.73533424],\n",
              "       [ -5.48888269, -58.79616709,  22.88584985, -54.14457159],\n",
              "       [-81.28026953, -75.17372078, 130.76437145, 131.39608339]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "34sgbtJAyelk",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "8909096b-afab-4c25-8c51-029f92b4ffe1"
      },
      "source": [
        "np.array(iris.feature_names)[selector.get_support()]"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['petal length (cm)', 'petal width (cm)'], dtype='<U17')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 29
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CljC3bCjygMI",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}