11_KNN(Classfication_iris).ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "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": "3.7.6"
    },
    "colab": {
      "provenance": [],
      "include_colab_link": true
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Sadcato/ai-way-of-mine/blob/main/11_KNN(Classfication_iris).ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0ox2wDT5ygSx",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 71
        },
        "outputId": "ce67ff0b-4113-4a7e-d68b-0cd04e991b45"
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "import io\n",
        "import requests"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
            "  import pandas.util.testing as tm\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "d_Vr-gJnygS4"
      },
      "source": [
        "## 1) 載入資料集"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PIq7xZR5ygS6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "outputId": "8d332b51-7783-4596-e87e-67cbc0ed3b7d"
      },
      "source": [
        "url = 'https://github.com/1010code/iris-dnn-tensorflow/raw/master/data/Iris.csv'\n",
        "s=requests.get(url).content\n",
        "df_data=pd.read_csv(io.StringIO(s.decode('utf-8')))\n",
        "df_data = df_data.drop(labels=['Id'],axis=1) # 移除Id\n",
        "df_data"
      ],
      "execution_count": null,
      "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>SepalLengthCm</th>\n",
              "      <th>SepalWidthCm</th>\n",
              "      <th>PetalLengthCm</th>\n",
              "      <th>PetalWidthCm</th>\n",
              "      <th>Species</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>5.1</td>\n",
              "      <td>3.5</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>4.9</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>4.7</td>\n",
              "      <td>3.2</td>\n",
              "      <td>1.3</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4.6</td>\n",
              "      <td>3.1</td>\n",
              "      <td>1.5</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5.0</td>\n",
              "      <td>3.6</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>145</th>\n",
              "      <td>6.7</td>\n",
              "      <td>3.0</td>\n",
              "      <td>5.2</td>\n",
              "      <td>2.3</td>\n",
              "      <td>Iris-virginica</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>146</th>\n",
              "      <td>6.3</td>\n",
              "      <td>2.5</td>\n",
              "      <td>5.0</td>\n",
              "      <td>1.9</td>\n",
              "      <td>Iris-virginica</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>147</th>\n",
              "      <td>6.5</td>\n",
              "      <td>3.0</td>\n",
              "      <td>5.2</td>\n",
              "      <td>2.0</td>\n",
              "      <td>Iris-virginica</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>148</th>\n",
              "      <td>6.2</td>\n",
              "      <td>3.4</td>\n",
              "      <td>5.4</td>\n",
              "      <td>2.3</td>\n",
              "      <td>Iris-virginica</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>149</th>\n",
              "      <td>5.9</td>\n",
              "      <td>3.0</td>\n",
              "      <td>5.1</td>\n",
              "      <td>1.8</td>\n",
              "      <td>Iris-virginica</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>150 rows × 5 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     SepalLengthCm  SepalWidthCm  PetalLengthCm  PetalWidthCm         Species\n",
              "0              5.1           3.5            1.4           0.2     Iris-setosa\n",
              "1              4.9           3.0            1.4           0.2     Iris-setosa\n",
              "2              4.7           3.2            1.3           0.2     Iris-setosa\n",
              "3              4.6           3.1            1.5           0.2     Iris-setosa\n",
              "4              5.0           3.6            1.4           0.2     Iris-setosa\n",
              "..             ...           ...            ...           ...             ...\n",
              "145            6.7           3.0            5.2           2.3  Iris-virginica\n",
              "146            6.3           2.5            5.0           1.9  Iris-virginica\n",
              "147            6.5           3.0            5.2           2.0  Iris-virginica\n",
              "148            6.2           3.4            5.4           2.3  Iris-virginica\n",
              "149            5.9           3.0            5.1           1.8  Iris-virginica\n",
              "\n",
              "[150 rows x 5 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wRNkTMeQygTA"
      },
      "source": [
        "## 2) 手動編碼\n",
        "處理名目資料 (Nominal variables) - 資料前處理\n",
        "依據特徵資料的特性，可以選擇手動編碼或自動編碼。\n",
        "\n",
        "### 使用編碼時機?\n",
        "進行深度學習時，神經網路只能處理數值資料。因此我們需要將所有非數字型態的特徵進行轉換。\n",
        "\n",
        "ex:\n",
        "\n",
        "|  Iris-setosa |  Iris-versicolor | Iris-virginica  |\n",
        "|:---:|:---:|:---:|\n",
        "|  1 | 2  | 3  |"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iADNqWQNygTA"
      },
      "source": [
        "label_map = {'Iris-setosa': 0, 'Iris-versicolor': 1, 'Iris-virginica': 2}\n",
        "\n",
        "#將編碼後的label map存至df_data['Species']中。\n",
        "df_data['Class'] = df_data['Species'].map(label_map)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "J2ZamFnHygTE",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "outputId": "0a5e85b1-7127-4b22-d9e9-4f0152c5ccbc"
      },
      "source": [
        "df_data"
      ],
      "execution_count": null,
      "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>SepalLengthCm</th>\n",
              "      <th>SepalWidthCm</th>\n",
              "      <th>PetalLengthCm</th>\n",
              "      <th>PetalWidthCm</th>\n",
              "      <th>Species</th>\n",
              "      <th>Class</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>5.1</td>\n",
              "      <td>3.5</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>4.9</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>4.7</td>\n",
              "      <td>3.2</td>\n",
              "      <td>1.3</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4.6</td>\n",
              "      <td>3.1</td>\n",
              "      <td>1.5</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5.0</td>\n",
              "      <td>3.6</td>\n",
              "      <td>1.4</td>\n",
              "      <td>0.2</td>\n",
              "      <td>Iris-setosa</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>145</th>\n",
              "      <td>6.7</td>\n",
              "      <td>3.0</td>\n",
              "      <td>5.2</td>\n",
              "      <td>2.3</td>\n",
              "      <td>Iris-virginica</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>146</th>\n",
              "      <td>6.3</td>\n",
              "      <td>2.5</td>\n",
              "      <td>5.0</td>\n",
              "      <td>1.9</td>\n",
              "      <td>Iris-virginica</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>147</th>\n",
              "      <td>6.5</td>\n",
              "      <td>3.0</td>\n",
              "      <td>5.2</td>\n",
              "      <td>2.0</td>\n",
              "      <td>Iris-virginica</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>148</th>\n",
              "      <td>6.2</td>\n",
              "      <td>3.4</td>\n",
              "      <td>5.4</td>\n",
              "      <td>2.3</td>\n",
              "      <td>Iris-virginica</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>149</th>\n",
              "      <td>5.9</td>\n",
              "      <td>3.0</td>\n",
              "      <td>5.1</td>\n",
              "      <td>1.8</td>\n",
              "      <td>Iris-virginica</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>150 rows × 6 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "     SepalLengthCm  SepalWidthCm  ...         Species  Class\n",
              "0              5.1           3.5  ...     Iris-setosa      0\n",
              "1              4.9           3.0  ...     Iris-setosa      0\n",
              "2              4.7           3.2  ...     Iris-setosa      0\n",
              "3              4.6           3.1  ...     Iris-setosa      0\n",
              "4              5.0           3.6  ...     Iris-setosa      0\n",
              "..             ...           ...  ...             ...    ...\n",
              "145            6.7           3.0  ...  Iris-virginica      2\n",
              "146            6.3           2.5  ...  Iris-virginica      2\n",
              "147            6.5           3.0  ...  Iris-virginica      2\n",
              "148            6.2           3.4  ...  Iris-virginica      2\n",
              "149            5.9           3.0  ...  Iris-virginica      2\n",
              "\n",
              "[150 rows x 6 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pYUfEb84ygTI"
      },
      "source": [
        "## 3) 檢查缺失值\n",
        "使用 numpy 所提供的函式來檢查是否有 NA 缺失值，假設有缺失值使用dropna()來移除。使用的時機在於當只有少量的缺失值適用，若遇到有大量缺失值的情況，或是本身的資料量就很少的情況下建議可以透過機器學習的方法補值來預測缺失值。\n",
        "\n",
        "```python\n",
        "# 移除缺失值\n",
        "train=train.dropna()\n",
        "```"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kwnSwhlGygTI",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "3a8cdafb-eade-4749-c9af-4bb1d3a30a00"
      },
      "source": [
        "X = df_data.drop(labels=['Species','Class'],axis=1).values # 移除Species (因為字母不參與訓練)\n",
        "# checked missing data\n",
        "print(\"checked missing data(NAN mount):\",len(np.where(np.isnan(X))[0]))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "checked missing data(NAN mount): 0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CWUr-s6sygTL"
      },
      "source": [
        "## 4) 切割訓練集與測試集"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7aVcFk4DygTL"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "X=df_data.drop(labels=['Class','Species'],axis=1)\n",
        "y=df_data['Class'].values\n",
        "X_train , X_test , y_train , y_test = train_test_split(X,y , test_size=.3 , random_state=42)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GoDSIZ8GygTN",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "f3902fad-a6b7-4f8a-9131-13949d76f222"
      },
      "source": [
        "print('Training data shape:',X_train.shape)\n",
        "print('Testing data shape:',X_test.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Training data shape: (105, 4)\n",
            "Testing data shape: (45, 4)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yEUuYs_sygTQ"
      },
      "source": [
        "## 建立 k-nearest neighbors(KNN) 模型\n",
        "Parameters:\n",
        "- n_neighbors: 設定鄰居的數量(k)，選取最近的k個點，預設為5。\n",
        "- algorithm: 搜尋數演算法{'auto'，'ball_tree'，'kd_tree'，'brute'}，可選。\n",
        "- metric: 計算距離的方式，預設為歐幾里得距離。\n",
        "\n",
        "Attributes:\n",
        "- classes_: 取得類別陣列。\n",
        "- effective_metric_: 取得計算距離的公式。\n",
        "\n",
        "Methods:\n",
        "- fit: 放入X、y進行模型擬合。\n",
        "- predict: 預測並回傳預測類別。\n",
        "- score: 預測成功的比例。"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "StPXF57mygTQ"
      },
      "source": [
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "\n",
        "# 建立KNN模型\n",
        "knnModel = KNeighborsClassifier(n_neighbors=3)\n",
        "# 使用訓練資料訓練模型\n",
        "knnModel.fit(X_train,y_train)\n",
        "# 使用訓練資料預測分類\n",
        "predicted= knnModel.predict(X_train)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RaDwWANPygTS"
      },
      "source": [
        "## 使用Score評估模型"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wKX_cGvlygTS",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "74e094b9-a5fe-42a7-956f-2659230c6f4e"
      },
      "source": [
        "# 預測成功的比例\n",
        "print('訓練集: ',knnModel.score(X_train,y_train))\n",
        "print('測試集: ',knnModel.score(X_test,y_test))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "訓練集:  0.9428571428571428\n",
            "測試集:  1.0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kVycw-l1ygTU"
      },
      "source": [
        "## 真實分類"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FjESXkz3ygTU"
      },
      "source": [
        "# 建立訓練集的 DataFrme\n",
        "df_train=pd.DataFrame(X_train)\n",
        "df_train['Class']=y_train\n",
        "# 建立測試集的 DataFrme\n",
        "df_test=pd.DataFrame(X_test)\n",
        "df_test['Class']=y_test"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-2sPt9f6ygTW",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 386
        },
        "outputId": "230b0a6d-d323-41d7-9a00-4b0e2a227ecc"
      },
      "source": [
        "sns.lmplot(\"PetalLengthCm\", \"PetalWidthCm\", hue='Class', data=df_train, fit_reg=False)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<seaborn.axisgrid.FacetGrid at 0x7f2fc0bf17f0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 402.375x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1M-AImwdygTX"
      },
      "source": [
        "## KNN (訓練集)預測結果"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LmJ2RxsrygTY",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 369
        },
        "outputId": "ae9e453a-107d-4e94-f236-0292aab8fa93"
      },
      "source": [
        "df_train['Predict']=predicted\n",
        "sns.lmplot(\"PetalLengthCm\", \"PetalWidthCm\", data=df_train, hue=\"Predict\", fit_reg=False)\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 402.5x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "meFAB7PSygTZ"
      },
      "source": [
        "## 查看不同的K分類結果\n",
        "為了方便視覺化我們將原有的測試集特徵使用PCA降成2維。接著觀察在不同 K 的狀況下，分類的情形為何。"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "izFuEiBhygTa"
      },
      "source": [
        "from matplotlib.colors import ListedColormap\n",
        "\n",
        "def plot_decision_regions(X, y, classifier, test_idx = None, resolution=0.02):\n",
        "    # setup marker generator and color map\n",
        "    markers = ('s','x','o','^','v')\n",
        "    colors = ('red','blue','lightgreen','gray','cyan')\n",
        "    cmap = ListedColormap(colors[:len(np.unique(y))])\n",
        "\n",
        "    # plot the decision surface\n",
        "    x1_min, x1_max = X[:,0].min() - 1, X[:,0].max() + 1\n",
        "    x2_min, x2_max = X[:,1].min() - 1, X[:,1].max() + 1\n",
        "\n",
        "    xx1, xx2 = np.meshgrid(np.arange(x1_min,x1_max,resolution),\n",
        "                            np.arange(x2_min,x2_max,resolution))\n",
        "\n",
        "    Z = classifier.predict(np.array([xx1.ravel(),xx2.ravel()]).T)\n",
        "\n",
        "    Z = Z.reshape(xx1.shape)\n",
        "\n",
        "    plt.contourf(xx1, xx2, Z, alpha=0.4, cmap=cmap)\n",
        "    plt.xlim(xx1.min(),xx1.max())\n",
        "    plt.ylim(xx2.min(),xx2.max())\n",
        "\n",
        "    for idx, cl in enumerate(np.unique(y)):\n",
        "        plt.scatter(x=X[y==cl,0], y=X[y==cl,1],\n",
        "            alpha=0.8, c=[cmap(idx)], marker=markers[idx],label=cl)\n",
        "    if test_idx:\n",
        "        X_test, y_test = X[test_idx,:], y[test_idx]\n",
        "        plt.scatter(X_test[:, 0], X_test[:,1], c='',\n",
        "            alpha=1.0, linewidth=1, marker='o',\n",
        "            s=55, label='test set')\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nrrOQ6vaygTb"
      },
      "source": [
        "def knn_model(plot_dict, X, y, k):\n",
        "    #create model\n",
        "    model = KNeighborsClassifier(n_neighbors=k)\n",
        "\n",
        "    #training\n",
        "    model.fit(X, y)\n",
        "\n",
        "    # Plot the decision boundary. For that, we will assign a color to each\n",
        "    if k in plot_dict:\n",
        "        plt.subplot(plot_dict[k])\n",
        "        plt.tight_layout()\n",
        "        plot_decision_regions(X,y,model)\n",
        "        plt.title('Plot for K: %d'%k )"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "H69Lonx9ygTd"
      },
      "source": [
        "from sklearn.decomposition import PCA\n",
        "pca = PCA(n_components=2, iterated_power=1)\n",
        "train_reduced = pca.fit_transform(X_train)\n",
        "test_reduced = pca.fit_transform(X_test)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RGi5wH5NygTe"
      },
      "source": [
        "### KNN 訓練集 PCA 2 features"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "l2_cDBlFygTf",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 441
        },
        "outputId": "750532c4-d293-4079-ff65-5829de356442"
      },
      "source": [
        "plt.figure(figsize=(8.5, 6))\n",
        "\n",
        "# 調整 K\n",
        "plot_dict = {1:231,2:232,3:233,6:234,10:235,15:236}\n",
        "for i in plot_dict:\n",
        "    knn_model(plot_dict, train_reduced, y_train, i)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 612x432 with 6 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8MHNNYOuygTg"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "H8hwK21dygTi"
      },
      "source": [],
      "execution_count": null,
      "outputs": []
    }
  ]
}