08장_지도학습_4절_선형회귀분석.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 개별 코드 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 데이터 불러오기\n",
    "from sklearn import datasets\n",
    "raw_boston = datasets.load_boston()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 피쳐, 타겟 데이터 지정\n",
    "X = raw_boston.data\n",
    "y = raw_boston.target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 트레이닝/테스트 데이터 분할\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_tn, X_te, y_tn, y_te=train_test_split(X,y,random_state=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#데이터 표준화\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "std_scale = StandardScaler()\n",
    "std_scale.fit(X_tn)\n",
    "X_tn_std = std_scale.transform(X_tn)\n",
    "X_te_std  = std_scale.transform(X_te)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None, normalize=False)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 선형 회귀분석 학습\n",
    "from sklearn.linear_model import LinearRegression \n",
    "clf_lr =  LinearRegression()\n",
    "clf_lr.fit(X_tn_std, y_tn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-1.07145146  1.34036243  0.26298069  0.66554537 -2.49842551  1.97524314\n",
      "  0.19516605 -3.14274974  2.66736136 -1.80685572 -2.13034748  0.56172933\n",
      " -4.03223518]\n",
      "22.344591029023768\n"
     ]
    }
   ],
   "source": [
    "# 선형 회귀분석 모형 추정 계수 확인\n",
    "print(clf_lr.coef_)\n",
    "print(clf_lr.intercept_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ridge(alpha=1, copy_X=True, fit_intercept=True, max_iter=None, normalize=False,\n",
       "      random_state=None, solver='auto', tol=0.001)"
      ]
     },
     "execution_count": 178,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 릿지 회귀분석(L2 제약식 적용)\n",
    "from sklearn.linear_model import Ridge\n",
    "clf_ridge = Ridge(alpha=1)\n",
    "clf_ridge.fit(X_tn_std, y_tn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-1.05933451  1.31050717  0.23022789  0.66955241 -2.45607567  1.99086611\n",
      "  0.18119169 -3.09919804  2.56480813 -1.71116799 -2.12002592  0.56264409\n",
      " -4.00942448]\n",
      "22.344591029023768\n"
     ]
    }
   ],
   "source": [
    "# 릿지 회귀분석 모형 추정 계수 확인\n",
    "print(clf_ridge.coef_)\n",
    "print(clf_ridge.intercept_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Lasso(alpha=0.01, copy_X=True, fit_intercept=True, max_iter=1000,\n",
       "      normalize=False, positive=False, precompute=False, random_state=None,\n",
       "      selection='cyclic', tol=0.0001, warm_start=False)"
      ]
     },
     "execution_count": 193,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 라쏘 회귀분석(L1 제약식 적용)\n",
    "from sklearn.linear_model import Lasso\n",
    "clf_lasso = Lasso(alpha=0.01)\n",
    "clf_lasso.fit(X_tn_std, y_tn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 194,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-1.04326518  1.27752711  0.1674367   0.66758228 -2.41559964  1.99244179\n",
      "  0.14733958 -3.09473711  2.46431135 -1.60552274 -2.11046422  0.55200229\n",
      " -4.00809905]\n",
      "22.344591029023768\n"
     ]
    }
   ],
   "source": [
    "# 라쏘 회귀분석 모형 추정 계수 확인\n",
    "print(clf_lasso.coef_)\n",
    "print(clf_lasso.intercept_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 211,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ElasticNet(alpha=0.01, copy_X=True, fit_intercept=True, l1_ratio=0.01,\n",
       "           max_iter=1000, normalize=False, positive=False, precompute=False,\n",
       "           random_state=None, selection='cyclic', tol=0.0001, warm_start=False)"
      ]
     },
     "execution_count": 211,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 엘라스틱넷\n",
    "from sklearn.linear_model import ElasticNet\n",
    "clf_elastic = ElasticNet(alpha=0.01, l1_ratio=0.01)\n",
    "clf_elastic.fit(X_tn_std, y_tn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 212,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-1.02916603  1.23681955  0.15236504  0.67859622 -2.34646781  2.02965524\n",
      "  0.14575132 -2.98592423  2.32013379 -1.48829485 -2.09271972  0.56506801\n",
      " -3.9495281 ]\n",
      "22.344591029023768\n"
     ]
    }
   ],
   "source": [
    "# 엘라스틱넷 모형 추정 계수 확인\n",
    "print(clf_elastic.coef_)\n",
    "print(clf_elastic.intercept_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 213,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 예측\n",
    "pred_lr = clf_lr.predict(X_te_std)\n",
    "pred_ridge = clf_ridge.predict(X_te_std)\n",
    "pred_lasso = clf_lasso.predict(X_te_std)\n",
    "pred_elastic = clf_elastic.predict(X_te_std)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 214,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7789410172622859\n",
      "0.7789704562726603\n",
      "0.7787621490259895\n",
      "0.7787876079239252\n"
     ]
    }
   ],
   "source": [
    "# 모형 평가-R제곱값\n",
    "from sklearn.metrics import r2_score\n",
    "print(r2_score(y_te, pred_lr))\n",
    "print(r2_score(y_te, pred_ridge))\n",
    "print(r2_score(y_te, pred_lasso))\n",
    "print(r2_score(y_te, pred_elastic))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 215,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "21.897765396049483\n",
      "21.894849212618773\n",
      "21.915483810504824\n",
      "21.91296189093687\n"
     ]
    }
   ],
   "source": [
    "# 모형 평가-MSE\n",
    "from sklearn.metrics import mean_squared_error\n",
    "print(mean_squared_error(y_te, pred_lr))\n",
    "print(mean_squared_error(y_te, pred_ridge))\n",
    "print(mean_squared_error(y_te, pred_lasso))\n",
    "print(mean_squared_error(y_te, pred_elastic))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 통합 코드"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 216,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-1.07145146  1.34036243  0.26298069  0.66554537 -2.49842551  1.97524314\n",
      "  0.19516605 -3.14274974  2.66736136 -1.80685572 -2.13034748  0.56172933\n",
      " -4.03223518]\n",
      "22.344591029023768\n",
      "[-1.05933451  1.31050717  0.23022789  0.66955241 -2.45607567  1.99086611\n",
      "  0.18119169 -3.09919804  2.56480813 -1.71116799 -2.12002592  0.56264409\n",
      " -4.00942448]\n",
      "22.344591029023768\n",
      "[-1.04326518  1.27752711  0.1674367   0.66758228 -2.41559964  1.99244179\n",
      "  0.14733958 -3.09473711  2.46431135 -1.60552274 -2.11046422  0.55200229\n",
      " -4.00809905]\n",
      "22.344591029023768\n",
      "[-1.02916603  1.23681955  0.15236504  0.67859622 -2.34646781  2.02965524\n",
      "  0.14575132 -2.98592423  2.32013379 -1.48829485 -2.09271972  0.56506801\n",
      " -3.9495281 ]\n",
      "22.344591029023768\n",
      "0.7789410172622859\n",
      "0.7789704562726603\n",
      "0.7787621490259895\n",
      "0.7787876079239252\n",
      "21.897765396049483\n",
      "21.894849212618773\n",
      "21.915483810504824\n",
      "21.91296189093687\n"
     ]
    }
   ],
   "source": [
    "from sklearn import datasets\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "from sklearn.linear_model import Ridge\n",
    "from sklearn.linear_model import Lasso\n",
    "from sklearn.linear_model import ElasticNet\n",
    "from sklearn.linear_model import LinearRegression \n",
    "\n",
    "from sklearn.metrics import r2_score\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "\n",
    "# 데이터 불러오기\n",
    "raw_boston = datasets.load_boston()\n",
    "\n",
    "# 피쳐, 타겟 데이터 지정\n",
    "X = raw_boston.data\n",
    "y = raw_boston.target\n",
    "\n",
    "# 트레이닝/테스트 데이터 분할\n",
    "X_tn, X_te, y_tn, y_te = train_test_split(X,y,random_state=1)\n",
    "\n",
    "\n",
    "# 데이터 표준화\n",
    "std_scale = StandardScaler()\n",
    "std_scale.fit(X_tn)\n",
    "X_tn_std = std_scale.transform(X_tn)\n",
    "X_te_std  = std_scale.transform(X_te)\n",
    "\n",
    "# 선형 회귀분석 학습\n",
    "clf_lr =  LinearRegression()\n",
    "clf_lr.fit(X_tn_std, y_tn)\n",
    "\n",
    "# 선형 회귀분석 모형 추정 계수 확인\n",
    "print(clf_lr.coef_)\n",
    "print(clf_lr.intercept_)\n",
    "\n",
    "# 릿지 회귀분석(L2 제약식 적용)\n",
    "clf_ridge = Ridge(alpha=1)\n",
    "clf_ridge.fit(X_tn_std, y_tn)\n",
    "\n",
    "# 릿지 회귀분석 모형 추정 계수 확인\n",
    "print(clf_ridge.coef_)\n",
    "print(clf_ridge.intercept_)\n",
    "\n",
    "# 라쏘 회귀분석(L1 제약식 적용)\n",
    "clf_lasso = Lasso(alpha=0.01)\n",
    "clf_lasso.fit(X_tn_std, y_tn)\n",
    "\n",
    "# 라쏘 회귀분석 모형 추정 계수 확인\n",
    "print(clf_lasso.coef_)\n",
    "print(clf_lasso.intercept_)\n",
    "\n",
    "# 엘라스틱넷\n",
    "clf_elastic = ElasticNet(alpha=0.01, l1_ratio=0.01)\n",
    "clf_elastic.fit(X_tn_std, y_tn)\n",
    "\n",
    "# 엘라스틱넷 모형 추정 계수 확인\n",
    "print(clf_elastic.coef_)\n",
    "print(clf_elastic.intercept_)\n",
    "\n",
    "# 예측\n",
    "pred_lr = clf_lr.predict(X_te_std)\n",
    "pred_ridge = clf_ridge.predict(X_te_std)\n",
    "pred_lasso = clf_lasso.predict(X_te_std)\n",
    "pred_elastic = clf_elastic.predict(X_te_std)\n",
    "\n",
    "# 모형 평가-R제곱값\n",
    "print(r2_score(y_te, pred_lr))\n",
    "print(r2_score(y_te, pred_ridge))\n",
    "print(r2_score(y_te, pred_lasso))\n",
    "print(r2_score(y_te, pred_elastic))\n",
    "\n",
    "# 모형 평가-MSE\n",
    "print(mean_squared_error(y_te, pred_lr))\n",
    "print(mean_squared_error(y_te, pred_ridge))\n",
    "print(mean_squared_error(y_te, pred_lasso))\n",
    "print(mean_squared_error(y_te, pred_elastic))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}