P02_SimpleRegression.ipynb

{
  "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/P02_SimpleRegression.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "id": "2a4339b1",
      "metadata": {
        "id": "2a4339b1"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "from sklearn.linear_model import LinearRegression\n",
        "import scipy as sp\n",
        "import matplotlib.pyplot as plt\n",
        "from scipy.optimize import fsolve\n",
        "from sklearn.metrics import confusion_matrix\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.utils import shuffle"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "id": "64cb5e01",
      "metadata": {
        "id": "64cb5e01"
      },
      "outputs": [],
      "source": [
        "x1=np.array([5, 15, 25, 35, 45, 55])\n",
        "y1=np.array([5, 20, 14, 32, 22, 38])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "id": "0be7f277",
      "metadata": {
        "id": "0be7f277"
      },
      "outputs": [],
      "source": [
        "x=x1.reshape((-1, 1)) # этот массив должен быть двумерным: иметь одну колонку и необходимое количество рядов\n",
        "y=y1"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "id": "3cca38a4",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3cca38a4",
        "outputId": "b6c813f0-5db6-4dca-8300-60e56d220379"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[ 5]\n",
            " [15]\n",
            " [25]\n",
            " [35]\n",
            " [45]\n",
            " [55]]\n",
            "[ 5 20 14 32 22 38]\n"
          ]
        }
      ],
      "source": [
        "print(x)\n",
        "print(y)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "id": "afd47752",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 313
        },
        "id": "afd47752",
        "outputId": "ff796ac3-28ec-4a86-b8dc-c3f36c37ce28"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.collections.PathCollection at 0x7f8c0f264640>"
            ]
          },
          "metadata": {},
          "execution_count": 5
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "axes = plt.axes()\n",
        "axes.set_xlim([0, 60])         \n",
        "axes.set_ylim([0, 40])         \n",
        "axes.set_xlabel(\"X\")             \n",
        "axes.set_ylabel(\"Y\")             \n",
        "legend = []\n",
        "plt.title('Исходные данные')\n",
        "plt.scatter(x1, y1, color='green', marker=\"o\")\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "id": "e84502f3",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "e84502f3",
        "outputId": "261e343c-f187-4936-901a-70593b11fbdf"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "coefficient of determination: 0.7158756137479542\n",
            "intercept: 5.633333333333329\n",
            "slope: [0.54]\n"
          ]
        }
      ],
      "source": [
        "model = LinearRegression()\n",
        "model.fit(x, y) # вычисляются оптимальные значение весов, используя существующие вход и выход в качестве аргументов\n",
        "print('coefficient of determination:', model.score(x, y))\n",
        "print('intercept:', model.intercept_) # b₀\n",
        "print('slope:', model.coef_) # b₁"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "id": "54999b4c",
      "metadata": {
        "scrolled": true,
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "54999b4c",
        "outputId": "14981485-3881-4982-b143-b301e239b718"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "actual response:\n",
            "[ 5 20 14 32 22 38]\n",
            "predicted response:\n",
            "[ 8.33333333 13.73333333 19.13333333 24.53333333 29.93333333 35.33333333]\n",
            "predicted response2:\n",
            "[[ 8.33333333]\n",
            " [13.73333333]\n",
            " [19.13333333]\n",
            " [24.53333333]\n",
            " [29.93333333]\n",
            " [35.33333333]]\n"
          ]
        }
      ],
      "source": [
        "y_pred = model.predict(x)\n",
        "print('actual response:', y, sep='\\n') \n",
        "print('predicted response:', y_pred, sep='\\n')\n",
        "\n",
        "z_pred = model.intercept_ + model.coef_ * x\n",
        "\n",
        "print('predicted response2:', z_pred, sep='\\n')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "id": "348b875b",
      "metadata": {
        "id": "348b875b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 313
        },
        "outputId": "134b65a0-1903-4936-b4c9-c4a220e698f2"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x7f8c0ed53e80>"
            ]
          },
          "metadata": {},
          "execution_count": 8
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "axes = plt.axes()\n",
        "axes.set_xlim([0, 60])         \n",
        "axes.set_ylim([0, 40])         \n",
        "axes.set_xlabel(\"X\")             \n",
        "axes.set_ylabel(\"Y\")             \n",
        "legend = []\n",
        "plt.title('Линейная регрессия')\n",
        "plt.scatter(x1, y1, color='green', marker=\"o\")\n",
        "legend.append(\"исходные данные\")\n",
        "\n",
        "plt.scatter(x1, y_pred, color='red', marker=\"s\")\n",
        "legend.append(\"предсказанные данные\")\n",
        "\n",
        "plt.plot(x1, y_pred, color='black', linewidth=1)\n",
        "legend.append(\"регрессия\")\n",
        "\n",
        "plt.vlines(x1, y1, y_pred, color='black', linewidth=2, linestyles='dashed')\n",
        "legend.append(\"остатки\")\n",
        "\n",
        "plt.legend(legend, loc=\"upper left\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "id": "9b24622f",
      "metadata": {
        "id": "9b24622f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 422
        },
        "outputId": "01fa256d-088d-4499-cf1a-801a4f4b24a1"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " \n",
            "0.54 x + 5.633\n",
            "           2\n",
            "-0.001964 x + 0.6579 x + 4.438\n",
            "           3          2\n",
            "0.0007315 x - 0.0678 x + 2.263 x - 4.23\n",
            "           4            3           2\n",
            "1.875e-05 x - 0.001519 x + 0.02073 x + 1.002 x + 0.4562\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "# получаем параметры модели для полинома степени d\n",
        "axes = plt.axes()\n",
        "axes.set_xlim([0, 65])         \n",
        "axes.set_ylim([0, 55])         \n",
        "axes.set_xlabel(\"X\")             \n",
        "axes.set_ylabel(\"Y\")\n",
        "fx = np.linspace(x1[0], x1[-1] + 5, 1000)\n",
        "legend = []\n",
        "plt.scatter(x1, y1, color='green', marker=\"o\")\n",
        "legend.append(\"исходные данные\")\n",
        "\n",
        "for d in range(1, 5):\n",
        "    fp = np.polyfit(x1, y1, d)\n",
        "    # функция-полином, если её напечатать, то увидите математическое выражение\n",
        "    f = sp.poly1d(fp)\n",
        "    print(f)\n",
        "    # рисуем график модельной функции\n",
        "    \n",
        "    plt.plot(fx, f(fx), linewidth=2)\n",
        "    legend.append(\"d=%i\" % f.order)\n",
        "    \n",
        "plt.legend(legend, loc=\"upper left\")\n",
        "plt.grid()\n",
        "plt.show()"
      ]
    }
  ],
  "metadata": {
    "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": "3.9.13"
    },
    "colab": {
      "provenance": [],
      "include_colab_link": true
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}