차원축소_LDA_Quiz(0303).ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyO4UDLJahVQG3Cf93hNeQ4x",
      "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/cnddlf/2022_1_Study_AI/blob/main/%EC%B0%A8%EC%9B%90%EC%B6%95%EC%86%8C_LDA_Quiz(0303).ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "> 보스턴 주택 가격 데이터 LDA를 이용하여 차원축소하고 KNN을 이용하여 정확도를 구하시오(다음 순서를 참고하여 작성하시오)\n",
        "> 1. 데이터 얻기\n",
        "> 2. 학습 및 데이터 나누기\n",
        "> 3. 데이터 표준화\n",
        "> 4. 주성분데이터 찾기\n",
        "> 5. KNN 모델 적용"
      ],
      "metadata": {
        "id": "nE75VCPCY1Sd"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#1. 데이터 얻기"
      ],
      "metadata": {
        "id": "Riysf2cga9FT"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Xng__7b6Xt84",
        "outputId": "e2fae2cc-abb3-4691-95d5-e01ab79de52a"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/sklearn/utils/deprecation.py:87: FutureWarning: Function load_boston is deprecated; `load_boston` is deprecated in 1.0 and will be removed in 1.2.\n",
            "\n",
            "    The Boston housing prices dataset has an ethical problem. You can refer to\n",
            "    the documentation of this function for further details.\n",
            "\n",
            "    The scikit-learn maintainers therefore strongly discourage the use of this\n",
            "    dataset unless the purpose of the code is to study and educate about\n",
            "    ethical issues in data science and machine learning.\n",
            "\n",
            "    In this special case, you can fetch the dataset from the original\n",
            "    source::\n",
            "\n",
            "        import pandas as pd\n",
            "        import numpy as np\n",
            "\n",
            "\n",
            "        data_url = \"http://lib.stat.cmu.edu/datasets/boston\"\n",
            "        raw_df = pd.read_csv(data_url, sep=\"\\s+\", skiprows=22, header=None)\n",
            "        data = np.hstack([raw_df.values[::2, :], raw_df.values[1::2, :2]])\n",
            "        target = raw_df.values[1::2, 2]\n",
            "\n",
            "    Alternative datasets include the California housing dataset (i.e.\n",
            "    :func:`~sklearn.datasets.fetch_california_housing`) and the Ames housing\n",
            "    dataset. You can load the datasets as follows::\n",
            "\n",
            "        from sklearn.datasets import fetch_california_housing\n",
            "        housing = fetch_california_housing()\n",
            "\n",
            "    for the California housing dataset and::\n",
            "\n",
            "        from sklearn.datasets import fetch_openml\n",
            "        housing = fetch_openml(name=\"house_prices\", as_frame=True)\n",
            "\n",
            "    for the Ames housing dataset.\n",
            "    \n",
            "  warnings.warn(msg, category=FutureWarning)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[6.3200e-03, 1.8000e+01, 2.3100e+00, 0.0000e+00, 5.3800e-01,\n",
              "        6.5750e+00, 6.5200e+01, 4.0900e+00, 1.0000e+00, 2.9600e+02,\n",
              "        1.5300e+01, 3.9690e+02, 4.9800e+00],\n",
              "       [2.7310e-02, 0.0000e+00, 7.0700e+00, 0.0000e+00, 4.6900e-01,\n",
              "        6.4210e+00, 7.8900e+01, 4.9671e+00, 2.0000e+00, 2.4200e+02,\n",
              "        1.7800e+01, 3.9690e+02, 9.1400e+00],\n",
              "       [2.7290e-02, 0.0000e+00, 7.0700e+00, 0.0000e+00, 4.6900e-01,\n",
              "        7.1850e+00, 6.1100e+01, 4.9671e+00, 2.0000e+00, 2.4200e+02,\n",
              "        1.7800e+01, 3.9283e+02, 4.0300e+00],\n",
              "       [3.2370e-02, 0.0000e+00, 2.1800e+00, 0.0000e+00, 4.5800e-01,\n",
              "        6.9980e+00, 4.5800e+01, 6.0622e+00, 3.0000e+00, 2.2200e+02,\n",
              "        1.8700e+01, 3.9463e+02, 2.9400e+00],\n",
              "       [6.9050e-02, 0.0000e+00, 2.1800e+00, 0.0000e+00, 4.5800e-01,\n",
              "        7.1470e+00, 5.4200e+01, 6.0622e+00, 3.0000e+00, 2.2200e+02,\n",
              "        1.8700e+01, 3.9690e+02, 5.3300e+00]])"
            ]
          },
          "metadata": {},
          "execution_count": 1
        }
      ],
      "source": [
        "from sklearn import datasets\n",
        "\n",
        "boston = datasets.load_boston()\n",
        "\n",
        "data = boston['data']\n",
        "data[:5]"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#2. 표준화"
      ],
      "metadata": {
        "id": "TnMGXdCJu_aF"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.preprocessing import StandardScaler\n",
        "\n",
        "scaler = StandardScaler()\n",
        "data_std = scaler.fit_transform(data)"
      ],
      "metadata": {
        "id": "5mjZT6fCu-2o"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#3. 훈련 및 검증 데이터"
      ],
      "metadata": {
        "id": "wfO4xvhpa7d2"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "X_train, X_test, y_train, y_test = train_test_split(data_std, boston['target'], random_state=62)"
      ],
      "metadata": {
        "id": "22AF3uHZvWcM"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#4. 주성분 데이터 찾기"
      ],
      "metadata": {
        "id": "APyjME5kbxEC"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# LDA 사용시 label 정보는 정수이어야 함.\n",
        "y_train = y_train.astype('int')\n",
        "y_train"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DHrWnyb7v7Uc",
        "outputId": "0ae3231a-0814-47e4-c32f-61f3bd2c991c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([15, 48, 19, 26, 23, 23, 18, 26, 23, 16, 32, 23, 24, 23, 21, 12, 20,\n",
              "       13, 37, 22, 19, 14, 33, 17, 18, 22, 28, 24, 24, 11, 10, 37, 24, 20,\n",
              "       24, 21, 20, 24, 22, 23, 16, 22, 20, 17, 17, 35, 19, 21, 14, 18, 23,\n",
              "       21, 31,  7, 30, 24, 32, 20, 14, 22, 25, 18, 19, 20, 43, 23, 16, 20,\n",
              "       23, 26, 26, 15, 10, 21, 21, 50, 13, 43, 23, 25, 20, 23, 24, 19, 22,\n",
              "        8, 18, 32, 13, 17, 30, 10, 22, 29, 24, 29, 18, 23, 20, 17, 22, 11,\n",
              "       19, 28, 14, 27, 19, 31, 16, 18, 13, 10, 21, 20, 29, 17, 23, 50, 14,\n",
              "       18, 14, 34, 50, 13, 22, 22, 16, 13, 45, 50, 20, 26, 10, 14, 22, 14,\n",
              "       22, 13, 18, 23, 32, 16, 28, 16, 41,  5, 34, 13, 27, 25, 28, 24, 26,\n",
              "       23, 15, 24, 21, 14, 43, 11, 28,  6, 23, 15, 19, 31, 19, 24, 22, 15,\n",
              "       21, 16, 24, 21, 22, 12, 30, 17, 27, 10, 10, 21, 34, 18, 14, 22, 25,\n",
              "       11, 18, 20, 14, 23, 22, 16, 25, 21, 17, 17, 14, 17,  8, 33, 19, 21,\n",
              "       21, 36, 15, 19, 21, 29, 23, 24, 37,  7, 23, 29, 30, 23, 15, 13, 13,\n",
              "       19, 22, 20, 21, 17, 24, 27, 19, 20, 36, 50, 21, 13, 27, 15, 18, 19,\n",
              "       24, 15, 29, 17, 12, 28, 17,  5, 21, 19, 20, 27, 23,  8, 24, 14, 29,\n",
              "       22, 19, 50, 17, 20, 21, 21, 20, 19, 31,  5, 20, 50, 12, 33, 24, 21,\n",
              "       28, 18, 32, 22, 24,  8, 50, 28, 29, 42, 12, 19, 20, 22, 21, 24,  7,\n",
              "       21, 20, 20, 18, 37, 42, 20, 23, 13, 22, 12, 19, 17, 19, 22, 25, 50,\n",
              "       15, 22, 18, 23, 19, 17, 27, 18, 30, 28, 16, 23,  7, 32, 11, 33, 23,\n",
              "       48, 20, 19, 34, 18,  8, 18, 50, 13, 24,  8, 13, 20, 11, 50,  9, 27,\n",
              "       18, 14, 12, 31, 29, 20, 23, 21, 14, 14, 18, 20,  9, 19, 19, 20, 19,\n",
              "       12, 44, 31, 18, 22, 15, 23, 25, 38, 14,  9,  7, 17, 17, 36, 23, 50,\n",
              "       18, 13, 23, 15, 20])"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
        "\n",
        "lda = LinearDiscriminantAnalysis()\n",
        "lda.fit(X_train, y_train)\n",
        "lda.explained_variance_ratio_"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gJdSufTqcIDA",
        "outputId": "51176574-15d3-40c6-b39f-18001b356748"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0.6226411 , 0.13428482, 0.0665114 , 0.04085967, 0.03246797,\n",
              "       0.02569701, 0.02043952, 0.01881548, 0.01160513, 0.01102314,\n",
              "       0.00739516, 0.00451698, 0.00374263])"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "\n",
        "x_length = len(X_train[0])\n",
        "# 막대그래프의 위치를 지정할 정보\n",
        "x = np.arange(x_length)\n",
        "x"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zqov5BHyiXQw",
        "outputId": "792da528-4a99-4282-d143-3e367a1953f6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12])"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "plt.bar(x, lda.explained_variance_ratio_)\n",
        "# plt.xticks(x, columns)\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "ozyhAIuMwSXv",
        "outputId": "fd9ec526-9411-423f-841b-d50d45a34ced"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#5. K 최근접 이웃모델 적용"
      ],
      "metadata": {
        "id": "N_b8-ZOFwpe4"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "##5.1 차원 축소"
      ],
      "metadata": {
        "id": "gTE0pUSbzF1v"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# 주성분 분석(PCA) 특성 추출 기법을 적용\n",
        "n_component_size = 7\n",
        "lda = LinearDiscriminantAnalysis(n_components=n_component_size)\n",
        "X_train_lda = lda.fit_transform(X_train, y_train)\n",
        "X_test_lda = lda.transform(X_test)"
      ],
      "metadata": {
        "id": "s6ch_DVvzFsd"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "##5.2 KNN 모델 생성 및 학습"
      ],
      "metadata": {
        "id": "Sy4h18V-zs4o"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.neighbors import KNeighborsRegressor\n",
        "\n",
        "knn = KNeighborsRegressor(n_neighbors = n_component_size)\n",
        "knn.fit(X_train_lda, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8JTAXhiJzFpN",
        "outputId": "5abd855f-fd83-4ed0-93f0-5b748887ebcb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "KNeighborsRegressor(n_neighbors=7)"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##5.3 예측"
      ],
      "metadata": {
        "id": "lnZ4FyVZz9qq"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "knn.score(X_test_lda, y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RqJBlUmWz9TC",
        "outputId": "9cf53c3e-2c6d-4a03-a430-c228df2fa0af"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.8017046115408937"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##5.4 차트 확인"
      ],
      "metadata": {
        "id": "Qlysl5WC1ZYv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "y_predicted = knn.predict(X_test_lda)"
      ],
      "metadata": {
        "id": "xGPr40KD1b3C"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "number_of_sample = 40\n",
        "plt.plot(range(number_of_sample), y_test[:number_of_sample], label='real target')\n",
        "plt.plot(range(number_of_sample), y_predicted[:number_of_sample], label='predicted target')\n",
        "plt.ylabel('price')\n",
        "plt.legend()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "AqWzv4BE1gpS",
        "outputId": "1418bcd6-1763-498b-cdc7-17174a80caec"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#6. 하이퍼 파라미터"
      ],
      "metadata": {
        "id": "o33BbRjH1o0T"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import GridSearchCV\n",
        "\n",
        "def get_best_hyper_paramerter(dt_clf, n_neighbors_list, X_train, y_train):\n",
        "  params =  {\n",
        "    'n_neighbors' : n_neighbors_list,\n",
        "    'weights' : [\"uniform\", \"distance\"],\n",
        "    'metric' : ['euclidean', 'manhattan', 'minkowski']\n",
        "  }\n",
        "\n",
        "  grid_cv = GridSearchCV(dt_clf, param_grid=params, scoring='explained_variance', cv=5, verbose=1)\n",
        "  grid_cv.fit(X_train, y_train)\n",
        "  print('Best Score: ', grid_cv.best_score_)\n",
        "  print('Best Hyper Parameter: ', grid_cv.best_params_)"
      ],
      "metadata": {
        "id": "xLMectYL1rhF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "> 위의 하이퍼 파라미터 함수를 이용하여 최적의 파라미터를 구하시오."
      ],
      "metadata": {
        "id": "wPvejG9X1y7b"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "y_train = boston['target'].astype('int')\n",
        "\n",
        "n_component_size = 7\n",
        "lda = LinearDiscriminantAnalysis(n_components=n_component_size)\n",
        "X_train_lda = lda.fit_transform(data_std, y_train)"
      ],
      "metadata": {
        "id": "icoehFlf5DOB"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# 10단위 리스트\n",
        "\n",
        "n_neighbors_list = range(1, 91, 10)\n",
        "knn = KNeighborsRegressor()"
      ],
      "metadata": {
        "id": "TcAAxv5O5Std"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "get_best_hyper_paramerter(knn, n_neighbors_list, X_train_lda, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "o65yfnkRBinw",
        "outputId": "942f0010-0948-4286-a287-23fdfe9dd3d6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Fitting 5 folds for each of 54 candidates, totalling 270 fits\n",
            "Best Score:  0.6445111179972716\n",
            "Best Hyper Parameter:  {'metric': 'manhattan', 'n_neighbors': 11, 'weights': 'distance'}\n"
          ]
        }
      ]
    }
  ]
}