week3_knn_강미경.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "과제 #1은 IRIS 데이터를 아래와 같이 불러와서    \n",
    "간단한 전처리(스케일링 등)와 EDA를 하고,     \n",
    "'#1_참고자료.html'을 참고하여 KNN에 GridSearchCV를 적용해보는 과제입니다!   \n",
    "그리고 BestParameter와 BestScore, TestsetScore를 구합니다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# KNN - IRIS DATA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.datasets import load_iris\n",
    "from sklearn.model_selection import train_test_split \n",
    "from sklearn.neighbors import KNeighborsClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 그래프 한글 깨짐 방지 폰트 설정\n",
    "plt.rc('font', family='Malgun Gothic')\n",
    "plt.rc('axes', unicode_minus=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## DATA LOAD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 아이리스 데이터 불러오기\n",
    "iris = load_iris() #iris 데이터를 dict 형식으로 불러옴."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#iris 데이터의 feature 부분과 label 부분을 분리\n",
    "X = iris.data\n",
    "y = iris.target"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- About data : https://www.kaggle.com/uciml/iris"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## SCAILING"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 거리 기반의 모델을 이용할 때에는 데이터의 스케일링이 필수!  \n",
    "- 가장 기본인 StandardScaler를 이용합니다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "scaler = StandardScaler()\n",
    "X = scaler.fit_transform(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "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>sepal length (cm)</th>\n",
       "      <th>sepal width (cm)</th>\n",
       "      <th>petal length (cm)</th>\n",
       "      <th>petal width (cm)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>-0.900681</td>\n",
       "      <td>1.019004</td>\n",
       "      <td>-1.340227</td>\n",
       "      <td>-1.315444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>-1.143017</td>\n",
       "      <td>-0.131979</td>\n",
       "      <td>-1.340227</td>\n",
       "      <td>-1.315444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>-1.385353</td>\n",
       "      <td>0.328414</td>\n",
       "      <td>-1.397064</td>\n",
       "      <td>-1.315444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>-1.506521</td>\n",
       "      <td>0.098217</td>\n",
       "      <td>-1.283389</td>\n",
       "      <td>-1.315444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>-1.021849</td>\n",
       "      <td>1.249201</td>\n",
       "      <td>-1.340227</td>\n",
       "      <td>-1.315444</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   sepal length (cm)  sepal width (cm)  petal length (cm)  petal width (cm)\n",
       "0          -0.900681          1.019004          -1.340227         -1.315444\n",
       "1          -1.143017         -0.131979          -1.340227         -1.315444\n",
       "2          -1.385353          0.328414          -1.397064         -1.315444\n",
       "3          -1.506521          0.098217          -1.283389         -1.315444\n",
       "4          -1.021849          1.249201          -1.340227         -1.315444"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#feature를 데이터 프레임 형태로 변환\n",
    "X_df = pd.DataFrame(X, columns = iris['feature_names'])\n",
    "X_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Preprocessing & EDA\n",
    "붖꽃 데이터 iris는 어떤 형태로 이루어져 있는지 봅시다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['data', 'target', 'target_names', 'DESCR', 'feature_names', 'filename'])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['sepal length (cm)',\n",
       " 'sepal width (cm)',\n",
       " 'petal length (cm)',\n",
       " 'petal width (cm)']"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris['data'] \n",
    "#각 꽃의 feature에 대한 값들\n",
    "iris['feature_names']\n",
    "#각 데이터들의 의미, sepal은 꽃받침, petal은 꽃잎이며 각각 이에 대한 길이와 넓이를 나타냄."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['setosa', 'versicolor', 'virginica'], dtype='<U10')"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris['target'] \n",
    "#target 변수, 즉 label에 대한 정보가 담겨져 있음.\n",
    "iris['target_names'] \n",
    "#target의 0은 setosa를, 1은 Versicolor를, 2는 Virginica를 의미함."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 꽃받침과 꽃잎의 너비, 길이로 구성되어 있는 feature가 있고 총 세 가지의 종으로 분류되어 있는 데이터 입니다.\n",
    "- feature들은 어떻게 구성되어 있는지 보겠습니다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.subplot(2,2,1)\n",
    "sns.boxplot(y, X_df['sepal length (cm)'])\n",
    "\n",
    "plt.subplot(2,2,2)\n",
    "sns.boxplot(y, X_df['sepal width (cm)'])\n",
    "\n",
    "plt.subplot(2,2,3)\n",
    "sns.boxplot(y, X_df['petal length (cm)'])\n",
    "\n",
    "plt.subplot(2,2,4)\n",
    "sns.boxplot(y, X_df['petal width (cm)'])\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "종 마다 feature 별로 특징이 두드러지나 sepal width의 경우에는 좀 애매합니다  \n",
    "어떤 식으로 분포하고 있는 지 알아보기 위해 산점도를 이용해 시각화 해보겠습니다  \n",
    "([이곳](http://theyoonicon.com/scikit-learn%EC%9D%98-iris-%EB%8D%B0%EC%9D%B4%ED%84%B0%EC%85%8B-%EB%B6%84%EB%A5%98%ED%95%98%EA%B8%B0/)을 참고하였습니다)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#종 별 sepal length / width에 따른 2차원 산점도 그리기\n",
    "for target in range(3):\n",
    "    if target == 0: #종이 setosa인 경우 red\n",
    "        c='r'\n",
    "        marker='>'\n",
    "    elif target == 1: #종이 versicolor인 경우 green\n",
    "        c='g'\n",
    "        marker='o'\n",
    "    elif target == 2: #종이 virginica인 경우 blue\n",
    "        c='b'\n",
    "        marker='x'\n",
    "\n",
    "    plt.scatter(X[y == target, 0], # sepal length\n",
    "                X[y == target, 1], # sepal width\n",
    "                marker = marker, c = c)\n",
    "    plt.xlabel(\"sepal length\")\n",
    "    plt.ylabel(\"sepal width\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#종 별 petal length / width에 따른 2차원 산점도 그리기\n",
    "for target in range(3):\n",
    "    if target == 0: #종이 setosa인 경우 red\n",
    "        c='r'\n",
    "        marker='>'\n",
    "    elif target == 1: #종이 versicolor인 경우 green\n",
    "        c='g'\n",
    "        marker='o'\n",
    "    elif target == 2: #종이 virginica인 경우 blue\n",
    "        c='b'\n",
    "        marker='x'\n",
    "\n",
    "    plt.scatter(X[y == target, 2], # petal length\n",
    "                X[y == target, 3], # petal width\n",
    "                marker = marker, c = c)\n",
    "    plt.xlabel(\"petal length\")\n",
    "    plt.ylabel(\"petal width\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "petal로는 육안으로 보아도 구분이 확실히 됩니다. 따라서 sepal, 즉 꽃받침의 너비가 요주의 변수겠네요"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "--------------------------------------------------------------------------------------\n",
    "# 2. Modeling\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2-1. KNN\n",
    "KNN모델을 생성하고 train set을 fit 시켜봅시다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Train, Test 셋 나누기 - 80대 20으로 나눔.\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state=77)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Documentation for \"KNeighborsClassifier\" : https://scikit-learn.org/stable/modules/generated/sklearn.neighbors.KNeighborsClassifier.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# build KNN model (k = 3) KNN 모델을 생성하고 하이퍼파라미터 k를 3으로 지정\n",
    "knn = KNeighborsClassifier(n_neighbors = 3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski',\n",
       "                     metric_params=None, n_jobs=None, n_neighbors=3, p=2,\n",
       "                     weights='uniform')"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# train the model 생성한 KNN모델에 train 데이터 셋을 fit 시켜 모델 학습\n",
    "knn.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get the predict value(y_pred) from X_test #학습한 데이터를 바탕으로 X_test에 대한 예측값 y_pred생성\n",
    "y_pred = knn.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy:  0.8333333333333334\n"
     ]
    }
   ],
   "source": [
    "# X_test를 투입시켰을 때의 y_pred값과 y_test의 값을 비교하여 mean accurcy반환\n",
    "print('accuracy: ', knn.score(X_test, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy:  0.8666666666666667\n"
     ]
    }
   ],
   "source": [
    "#k=5로 지정한 경우\n",
    "knn2 = KNeighborsClassifier(n_neighbors = 5)\n",
    "knn2.fit(X_train, y_train)\n",
    "y_pred = knn2.predict(X_test)\n",
    "print('accuracy: ', knn2.score(X_test, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "k=5일 때의 accuracy 값이 더 높게 나왔다. 그렇다면 더 좋은 결과를 낼 수는 없을까?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### [참고] KNeighborsClassifier's options\n",
    "1. n_neighbors = int (default = 5)\n",
    "2. weights = str or callable (default = 'uniform')\n",
    "3. metric = string or callable (default = 'minkowski')\n",
    "\n",
    "따라서 이 결과는 \n",
    "1. n_neighbors = 3 or 5\n",
    "2. weights = uniform \n",
    "3. metric = minkowski 일 때의 score 값이라고 할 수 있다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2-2 GridSearchCV\n",
    "다양한 하이퍼파라미터의 조합으로 가장 좋은 score 값을 추출하고 보다 신뢰성있는 모델 평가를 위해 교차검증을 실시해보자!\n",
    "\n",
    "1. 교차검증(k-fold-validation) : 모든 데이터 셋을 train에도 사용하고, validation set에도 사용함으로써 데이터의 수가 적은 경우나 데이터가 한쪽으로 쏠려 있는 경우에도 신뢰성 있는 모델 평가를 가능케 함.\n",
    "\n",
    "2. Validation set : 학습 과정에서 여러 가지 하이퍼 파라미터들로 생성된 모델 중 어떤 것이 성능이 좋은지 평가할 때 소모되는 데이터 셋 *Test set과 헷갈리지 말 것\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import GridSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "grid_params = {\n",
    "    'n_neighbors' : [3, 5, 11, 15], #인접한 이웃 몇 개를 볼 것인지를 결정하는 하이퍼 파라미터 k의 개수를 각각 지정\n",
    "    'weights' : ['uniform', 'distance'], #가중치 결정 방법\n",
    "    'metric' : ['euclidean', 'manhattan'] #거리계산 방식\n",
    "} #따라서 각 경우에 대해 최종 결과는 4 * 2 * 2 = 16가지의 형태로 나타날 것이다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "gs = GridSearchCV(\n",
    "    KNeighborsClassifier(), #KNN을 이용\n",
    "    grid_params, #위에서 설정한 파라미터를 인자로 전달\n",
    "    verbose = 1, #?\n",
    "    cv = 3, # 3-fold-cross-validation을 이용함.\n",
    "    n_jobs = -1 #?\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 16 candidates, totalling 48 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  48 out of  48 | elapsed:    0.0s finished\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py:814: DeprecationWarning: The default of the `iid` parameter will change from True to False in version 0.22 and will be removed in 0.24. This will change numeric results when test-set sizes are unequal.\n",
      "  DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "gs_results = gs.fit(X_train, y_train)\n",
    "#train 데이터 셋에 fit 시킨 결과를 gs_results에 저장함."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Parameter: {'metric': 'euclidean', 'n_neighbors': 11, 'weights': 'uniform'}\n"
     ]
    }
   ],
   "source": [
    "print(\"Best Parameter: {}\".format(gs.best_params_))\n",
    "#가장 좋은 결과를 얻는 parameter는 이웃 수 (n_neighbors)가 11일 때, \n",
    "#거리 계산 방식(metric)이 유클리디안일 때, \n",
    "#가중치(weights)를 uniform 방식으로 주었을 때이다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Cross-validity Score: 0.992\n"
     ]
    }
   ],
   "source": [
    "print(\"Best Cross-validity Score: {:.3f}\".format(gs.best_score_))\n",
    "#위의 파라미터를 이용해 계산한 최종 스코어는 0.992이다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test set Score: 0.867\n"
     ]
    }
   ],
   "source": [
    "print(\"Test set Score: {:.3f}\".format(gs.score(X_test, y_test)))\n",
    "#grid_search 알고리즘을 test_set에 적용시켜 나온 score값은 다음과 같다.\n",
    "#위의 이웃수를 5로 지정해 주었을 때의 값과 별반 다를 것이 없다.더 좋은 결과를 어떻게 얻을까?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[Score 값의 비교]\n",
    "- Gridsearch의 score값은 별 다른 옵션을 지정해주지 않는 이상 estimator의 scoring 방식을 사용!\n",
    "- 따라서 위에서 knn모델을 이용해 출력한 score 값과 gridsearch에서 출력한 score값은 그 산출 방식이 같다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2-3 GridSearchCV :: N_Neighbors Tuning\n",
    "위에서 교차검증으로 세 가지 하이퍼 파라미터를 튜닝해보았다면 이번에는 오로지 이웃의 개수인 n_neighbors만 세밀하게 조정해보자"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "#GridSearch의 범위를 1에서 10까지로 지정함\n",
    "param_grid = {'n_neighbors': [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "#KNN 알고리즘에 GridSearch를 적용함\n",
    "grid_search = GridSearchCV(KNeighborsClassifier(), param_grid, cv=5, return_train_score=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py:814: DeprecationWarning: The default of the `iid` parameter will change from True to False in version 0.22 and will be removed in 0.24. This will change numeric results when test-set sizes are unequal.\n",
      "  DeprecationWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score='raise-deprecating',\n",
       "             estimator=KNeighborsClassifier(algorithm='auto', leaf_size=30,\n",
       "                                            metric='minkowski',\n",
       "                                            metric_params=None, n_jobs=None,\n",
       "                                            n_neighbors=5, p=2,\n",
       "                                            weights='uniform'),\n",
       "             iid='warn', n_jobs=None,\n",
       "             param_grid={'n_neighbors': [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]},\n",
       "             pre_dispatch='2*n_jobs', refit=True, return_train_score=True,\n",
       "             scoring=None, verbose=0)"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Train 데이터에 GridSearch를 적용함\n",
    "grid_search.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "다섯 번의 교차검증을 실시하는데, n_neighbors이외의 다른 파라미터값을 지정하지 않았기 때문에 default 값인 weights(uniform), metric(minkowski) 방식으로 계산된다. 이를 바탕으로 계산한 결과는 아래와 같다. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Parameter: {'n_neighbors': 1}\n"
     ]
    }
   ],
   "source": [
    "#가장 좋은 결과를 얻는 그리드를 출력함\n",
    "print(\"Best Parameter: {}\".format(grid_search.best_params_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Cross-validity Score: 0.975\n"
     ]
    }
   ],
   "source": [
    "#grid_search의 최종 스코어를 출력함\n",
    "print(\"Best Cross-validity Score: {:.3f}\".format(grid_search.best_score_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test set Score: 0.833\n"
     ]
    }
   ],
   "source": [
    "#Test 데이터에 대해 grid_search 알고리즘을 적용함\n",
    "print(\"Test set Score: {:.3f}\".format(grid_search.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "result_grid = pd.DataFrame(grid_search.cv_results_)\n",
    "#결과를 데이터 프레임 형태로 저장"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "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>mean_fit_time</th>\n",
       "      <th>std_fit_time</th>\n",
       "      <th>mean_score_time</th>\n",
       "      <th>std_score_time</th>\n",
       "      <th>param_n_neighbors</th>\n",
       "      <th>params</th>\n",
       "      <th>split0_test_score</th>\n",
       "      <th>split1_test_score</th>\n",
       "      <th>split2_test_score</th>\n",
       "      <th>split3_test_score</th>\n",
       "      <th>...</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>std_test_score</th>\n",
       "      <th>rank_test_score</th>\n",
       "      <th>split0_train_score</th>\n",
       "      <th>split1_train_score</th>\n",
       "      <th>split2_train_score</th>\n",
       "      <th>split3_train_score</th>\n",
       "      <th>split4_train_score</th>\n",
       "      <th>mean_train_score</th>\n",
       "      <th>std_train_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.000591</td>\n",
       "      <td>0.000483</td>\n",
       "      <td>0.001596</td>\n",
       "      <td>7.981778e-04</td>\n",
       "      <td>1</td>\n",
       "      <td>{'n_neighbors': 1}</td>\n",
       "      <td>0.961538</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.956522</td>\n",
       "      <td>0.956522</td>\n",
       "      <td>...</td>\n",
       "      <td>0.975000</td>\n",
       "      <td>0.020498</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.000199</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>0.001795</td>\n",
       "      <td>3.983749e-04</td>\n",
       "      <td>2</td>\n",
       "      <td>{'n_neighbors': 2}</td>\n",
       "      <td>0.923077</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.956522</td>\n",
       "      <td>0.913043</td>\n",
       "      <td>...</td>\n",
       "      <td>0.958333</td>\n",
       "      <td>0.036845</td>\n",
       "      <td>6</td>\n",
       "      <td>0.989362</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.997872</td>\n",
       "      <td>0.004255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.000200</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>0.001197</td>\n",
       "      <td>3.989460e-04</td>\n",
       "      <td>3</td>\n",
       "      <td>{'n_neighbors': 3}</td>\n",
       "      <td>0.923077</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.956522</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.975000</td>\n",
       "      <td>0.031928</td>\n",
       "      <td>1</td>\n",
       "      <td>0.989362</td>\n",
       "      <td>0.989474</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.989691</td>\n",
       "      <td>0.989691</td>\n",
       "      <td>0.991643</td>\n",
       "      <td>0.004180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.000798</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>0.001397</td>\n",
       "      <td>4.887723e-04</td>\n",
       "      <td>4</td>\n",
       "      <td>{'n_neighbors': 4}</td>\n",
       "      <td>0.923077</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.913043</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.958333</td>\n",
       "      <td>0.036845</td>\n",
       "      <td>6</td>\n",
       "      <td>0.989362</td>\n",
       "      <td>0.989474</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.989691</td>\n",
       "      <td>0.979381</td>\n",
       "      <td>0.989582</td>\n",
       "      <td>0.006522</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.000798</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>0.001197</td>\n",
       "      <td>3.980877e-04</td>\n",
       "      <td>5</td>\n",
       "      <td>{'n_neighbors': 5}</td>\n",
       "      <td>0.884615</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.913043</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.950000</td>\n",
       "      <td>0.046864</td>\n",
       "      <td>9</td>\n",
       "      <td>0.989362</td>\n",
       "      <td>0.978947</td>\n",
       "      <td>0.989691</td>\n",
       "      <td>0.979381</td>\n",
       "      <td>0.989691</td>\n",
       "      <td>0.985414</td>\n",
       "      <td>0.005106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.000199</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>0.001200</td>\n",
       "      <td>4.042626e-04</td>\n",
       "      <td>6</td>\n",
       "      <td>{'n_neighbors': 6}</td>\n",
       "      <td>0.923077</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.956522</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.966667</td>\n",
       "      <td>0.029926</td>\n",
       "      <td>3</td>\n",
       "      <td>0.989362</td>\n",
       "      <td>0.989474</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.979381</td>\n",
       "      <td>0.991643</td>\n",
       "      <td>0.007745</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>0.000489</td>\n",
       "      <td>0.000798</td>\n",
       "      <td>3.989935e-04</td>\n",
       "      <td>7</td>\n",
       "      <td>{'n_neighbors': 7}</td>\n",
       "      <td>0.884615</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.966667</td>\n",
       "      <td>0.046215</td>\n",
       "      <td>3</td>\n",
       "      <td>0.989362</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.989691</td>\n",
       "      <td>0.989691</td>\n",
       "      <td>0.993749</td>\n",
       "      <td>0.005106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>0.000489</td>\n",
       "      <td>0.000998</td>\n",
       "      <td>1.507891e-07</td>\n",
       "      <td>8</td>\n",
       "      <td>{'n_neighbors': 8}</td>\n",
       "      <td>0.884615</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.966667</td>\n",
       "      <td>0.046215</td>\n",
       "      <td>3</td>\n",
       "      <td>0.989362</td>\n",
       "      <td>0.989474</td>\n",
       "      <td>0.989691</td>\n",
       "      <td>0.969072</td>\n",
       "      <td>0.979381</td>\n",
       "      <td>0.983396</td>\n",
       "      <td>0.008166</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.000798</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>4.886360e-04</td>\n",
       "      <td>9</td>\n",
       "      <td>{'n_neighbors': 9}</td>\n",
       "      <td>0.923077</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.913043</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.958333</td>\n",
       "      <td>0.036845</td>\n",
       "      <td>6</td>\n",
       "      <td>0.989362</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.989691</td>\n",
       "      <td>0.989691</td>\n",
       "      <td>0.993749</td>\n",
       "      <td>0.005106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.000798</td>\n",
       "      <td>0.000399</td>\n",
       "      <td>0.000997</td>\n",
       "      <td>2.780415e-07</td>\n",
       "      <td>10</td>\n",
       "      <td>{'n_neighbors': 10}</td>\n",
       "      <td>0.923077</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.956522</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.950000</td>\n",
       "      <td>0.049043</td>\n",
       "      <td>9</td>\n",
       "      <td>0.978723</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.989691</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.979381</td>\n",
       "      <td>0.989559</td>\n",
       "      <td>0.009371</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   mean_fit_time  std_fit_time  mean_score_time  std_score_time  \\\n",
       "0       0.000591      0.000483         0.001596    7.981778e-04   \n",
       "1       0.000199      0.000399         0.001795    3.983749e-04   \n",
       "2       0.000200      0.000399         0.001197    3.989460e-04   \n",
       "3       0.000798      0.000399         0.001397    4.887723e-04   \n",
       "4       0.000798      0.000399         0.001197    3.980877e-04   \n",
       "5       0.000199      0.000399         0.001200    4.042626e-04   \n",
       "6       0.000399      0.000489         0.000798    3.989935e-04   \n",
       "7       0.000399      0.000489         0.000998    1.507891e-07   \n",
       "8       0.000798      0.000399         0.000399    4.886360e-04   \n",
       "9       0.000798      0.000399         0.000997    2.780415e-07   \n",
       "\n",
       "  param_n_neighbors               params  split0_test_score  \\\n",
       "0                 1   {'n_neighbors': 1}           0.961538   \n",
       "1                 2   {'n_neighbors': 2}           0.923077   \n",
       "2                 3   {'n_neighbors': 3}           0.923077   \n",
       "3                 4   {'n_neighbors': 4}           0.923077   \n",
       "4                 5   {'n_neighbors': 5}           0.884615   \n",
       "5                 6   {'n_neighbors': 6}           0.923077   \n",
       "6                 7   {'n_neighbors': 7}           0.884615   \n",
       "7                 8   {'n_neighbors': 8}           0.884615   \n",
       "8                 9   {'n_neighbors': 9}           0.923077   \n",
       "9                10  {'n_neighbors': 10}           0.923077   \n",
       "\n",
       "   split1_test_score  split2_test_score  split3_test_score  ...  \\\n",
       "0                1.0           0.956522           0.956522  ...   \n",
       "1                1.0           0.956522           0.913043  ...   \n",
       "2                1.0           0.956522           1.000000  ...   \n",
       "3                1.0           0.913043           1.000000  ...   \n",
       "4                1.0           0.913043           1.000000  ...   \n",
       "5                1.0           0.956522           1.000000  ...   \n",
       "6                1.0           1.000000           1.000000  ...   \n",
       "7                1.0           1.000000           1.000000  ...   \n",
       "8                1.0           0.913043           1.000000  ...   \n",
       "9                1.0           0.956522           1.000000  ...   \n",
       "\n",
       "   mean_test_score  std_test_score  rank_test_score  split0_train_score  \\\n",
       "0         0.975000        0.020498                1            1.000000   \n",
       "1         0.958333        0.036845                6            0.989362   \n",
       "2         0.975000        0.031928                1            0.989362   \n",
       "3         0.958333        0.036845                6            0.989362   \n",
       "4         0.950000        0.046864                9            0.989362   \n",
       "5         0.966667        0.029926                3            0.989362   \n",
       "6         0.966667        0.046215                3            0.989362   \n",
       "7         0.966667        0.046215                3            0.989362   \n",
       "8         0.958333        0.036845                6            0.989362   \n",
       "9         0.950000        0.049043                9            0.978723   \n",
       "\n",
       "   split1_train_score  split2_train_score  split3_train_score  \\\n",
       "0            1.000000            1.000000            1.000000   \n",
       "1            1.000000            1.000000            1.000000   \n",
       "2            0.989474            1.000000            0.989691   \n",
       "3            0.989474            1.000000            0.989691   \n",
       "4            0.978947            0.989691            0.979381   \n",
       "5            0.989474            1.000000            1.000000   \n",
       "6            1.000000            1.000000            0.989691   \n",
       "7            0.989474            0.989691            0.969072   \n",
       "8            1.000000            1.000000            0.989691   \n",
       "9            1.000000            0.989691            1.000000   \n",
       "\n",
       "   split4_train_score  mean_train_score  std_train_score  \n",
       "0            1.000000          1.000000         0.000000  \n",
       "1            1.000000          0.997872         0.004255  \n",
       "2            0.989691          0.991643         0.004180  \n",
       "3            0.979381          0.989582         0.006522  \n",
       "4            0.989691          0.985414         0.005106  \n",
       "5            0.979381          0.991643         0.007745  \n",
       "6            0.989691          0.993749         0.005106  \n",
       "7            0.979381          0.983396         0.008166  \n",
       "8            0.989691          0.993749         0.005106  \n",
       "9            0.979381          0.989559         0.009371  \n",
       "\n",
       "[10 rows x 21 columns]"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_grid\n",
    "#각 n_neighbors에 따른 다양한 score 값을 보여줌."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1e4f94c5b08>"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(result_grid['param_n_neighbors'], result_grid['mean_train_score'], label=\"Train\")\n",
    "plt.plot(result_grid['param_n_neighbors'], result_grid['mean_test_score'], label=\"Test\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 이웃 수에 따른 train / test set의 mean_score 그래프\n",
    "- 이 그래프를 그리는 이유는 train / test 셋의 score를 같이 확인해봄으로써 모델이 over/under fitting 되진 않았는지 확인해보기 위해서다.\n",
    "- Train 스코어가 항상 더 높은 score을 가지기 때문에 과적합 된 것은 아닌지 생각해 볼 필요가 있다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3. Evaluation\n",
    "지금까지 진행 한 분석의 결과를 정리하고, 더 좋은 Score 값을 찾는 파라미터의 조합을 연구해보자"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[현재까지의 상황]\n",
    "1. n_neighbors = 5 / weights = uniform / metric = minkowski  \n",
    "{test set score : 0.86666667}\n",
    "\n",
    "2. n_neighbors = 11 / weights = uniform / metric = euclidean  \n",
    "{model score : 0.992  / test set score : 0.867}\n",
    "\n",
    "3. n_nieghbors = 1 / weights = uniform / metric = minkowski  \n",
    "{model score : 0.975  / test set score : 0.833}\n",
    "\n",
    "==> 파라미터를 더 다양하게 지정해보자\n",
    "metric의 경우 ([이곳](https://scikit-learn.org/stable/modules/generated/sklearn.neighbors.DistanceMetric.html)을 참고하였음)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [],
   "source": [
    "grid_params = {\n",
    "    'n_neighbors' : [1, 2, 3, 4, 5, 6, 7, 10, 11, 13, 15, 17, 20], #이웃 수\n",
    "    'weights' : ['uniform', 'distance'], #가중치 결정 방법\n",
    "    'metric' : ['euclidean', 'manhattan', 'minkowski', 'chebyshev'] #거리계산 방식\n",
    "} # 13 * 2 * 4 = 104개의 조합이 생성된다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [],
   "source": [
    "grid_tn = GridSearchCV(\n",
    "    KNeighborsClassifier(),\n",
    "    grid_params,\n",
    "    verbose = 1,\n",
    "    cv = 3,\n",
    "    n_jobs = -1\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 104 candidates, totalling 312 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done 312 out of 312 | elapsed:    0.1s finished\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_search.py:814: DeprecationWarning: The default of the `iid` parameter will change from True to False in version 0.22 and will be removed in 0.24. This will change numeric results when test-set sizes are unequal.\n",
      "  DeprecationWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=3, error_score='raise-deprecating',\n",
       "             estimator=KNeighborsClassifier(algorithm='auto', leaf_size=30,\n",
       "                                            metric='minkowski',\n",
       "                                            metric_params=None, n_jobs=None,\n",
       "                                            n_neighbors=5, p=2,\n",
       "                                            weights='uniform'),\n",
       "             iid='warn', n_jobs=-1,\n",
       "             param_grid={'metric': ['euclidean', 'manhattan', 'minkowski',\n",
       "                                    'chebyshev'],\n",
       "                         'n_neighbors': [1, 2, 3, 4, 5, 6, 7, 10, 11, 13, 15,\n",
       "                                         17, 20],\n",
       "                         'weights': ['uniform', 'distance']},\n",
       "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n",
       "             scoring=None, verbose=1)"
      ]
     },
     "execution_count": 131,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid_tn.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Parameter: {'metric': 'euclidean', 'n_neighbors': 10, 'weights': 'uniform'}\n",
      "Best Cross-validity Score: 1.000\n",
      "Test set Score: 0.900\n"
     ]
    }
   ],
   "source": [
    "print(\"Best Parameter: {}\".format(grid_tn.best_params_))\n",
    "print(\"Best Cross-validity Score: {:.3f}\".format(grid_tn.best_score_))\n",
    "print(\"Test set Score: {:.3f}\".format(grid_tn.score(X_test, y_test)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 결론\n",
    "여러 번의 시도 끝에 찾은 best score는 0.900이었으며 이때의 model score는 1이었다.  \n",
    "train set에서의 모델 스코어가 1이기 때문에 과적합을 우려할 수 있으나 다행히 test set score도 더 높게 나왔다.\n",
    "\n",
    "처음 cv값이 5였을 때에는 아무리해도 위에서 진행했던 score 값보다 낮게 나왔으나\n",
    "cv를 3으로 지정한 후에 겨우 더 높은 값을 찾을 수 있었다.\n",
    "하이퍼 파라미터를 튜닝할 때 이웃 수, 거리계산방식, 가중치계산방식만 고려했었는데 교차검증에 이용되는 fold 값 또한 영향을 줄 수 있다는 사실을 알게 되었다.\n",
    "\n",
    "그래도 마지막에는 더 좋은 값을 찾아서 다행이다.  \n",
    "이후로도 여러 번 조정해가면서 해보았지만 저 스코어 이상의 값을 찾을 수 없었다  \n",
    "만약 있다면.. 어떤 조합인지 알고싶다...\n",
    "\n"
   ]
  }
 ],
 "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}