try_log_regr.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyOzsS6zz6ZjaMgatEDr2nVS",
      "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/Sinrez/pyCoursera/blob/master/try_log_regr.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn import datasets\n",
        "\n",
        "\n",
        "iris = datasets.load_iris()\n",
        "X = iris.data[:, [0,3]] \n",
        "#выбрали sepal_length petal_width\n",
        "Y = iris.target\n",
        "# Y = Y0[Y0 == 1]\n",
        "Y"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IOKTDYceqTGA",
        "outputId": "2414a6c7-c639-426a-f779-f31c183f35f6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
              "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
              "       0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
              "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,\n",
              "       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,\n",
              "       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2])"
            ]
          },
          "metadata": {},
          "execution_count": 41
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Создать экземпляр классификатора логистической регрессии и подогнать под данные.\n",
        "logreg = LogisticRegression(C=1e5)\n",
        "logreg.fit(X, Y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zliTkgA7q5iJ",
        "outputId": "ba6d2937-b200-4617-c042-491fa5bcfd8e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LogisticRegression(C=100000.0)"
            ]
          },
          "metadata": {},
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# точка в сетке [x_min, x_max] x [y_min, y_max].\n",
        "x_min, x_max = X[:, 0].min() - .5, X[:, 0].max() + .5\n",
        "y_min, y_max = X[:, 1].min() - .5, X[:, 1].max() + .5\n",
        "h = .02  # размер шага в сетке\n",
        "xx, yy = np.meshgrid(np.arange(x_min, x_max, h), np.arange(y_min, y_max, h))\n",
        "Z = logreg.predict(np.c_[xx.ravel(), yy.ravel()])"
      ],
      "metadata": {
        "id": "3QWO64Gzq6x1"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "# результат в цветной график\n",
        "Z = Z.reshape(xx.shape)\n",
        "plt.figure(1, figsize=(4, 3))\n",
        "plt.pcolormesh(xx, yy, Z, cmap=plt.cm.Paired)\n",
        "\n",
        "#тренировочные точки\n",
        "plt.scatter(X[:, 0], X[:, 1], c=Y, edgecolors='k', cmap=plt.cm.Paired)\n",
        "plt.xlabel('Sepal length')\n",
        "plt.ylabel('Petal_width')\n",
        "\n",
        "plt.xlim(xx.min(), xx.max())\n",
        "plt.ylim(yy.min(), yy.max())\n",
        "plt.xticks(())\n",
        "plt.yticks(())\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 208
        },
        "id": "FL2owvU3q1gQ",
        "outputId": "8156170b-86da-4bdc-9a12-972863a11db5"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 288x216 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
}