legacy_materials/1_1_linear_regression_and_classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "881ecc53",
   "metadata": {},
   "source": [
    "# Neural Network 실습 전 개념 잡기  \n",
    "- scikit-learn으로 선형회귀 및 선형분류 모델 구축하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6aaad987",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 필요 패키지 및 모듈 불러오기\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "np.random.seed(42)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4c452180",
   "metadata": {},
   "source": [
    "## Linear Regression   \n",
    "- Goal : 데이터가 주어질 때 데이터를 가장 잘 표현할 수 있는 선형 방정식을 찾아내는 것"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "bcee8d15",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 선형 데이터 생성\n",
    "# np.random.randn : 평균 0, 편차 1인 정규 분포표 난수 생성\n",
    "# np.random.rand : 0~1 사이의 균일 분포 난수 생성\n",
    "\n",
    "# train 데이터\n",
    "X_train = 2*np.random.rand(100,1)\n",
    "y_train = 4 + 3*X_train + np.random.randn(100,1)\n",
    "\n",
    "# test 데이터\n",
    "X_test = 2*np.random.rand(20,1)\n",
    "y_test = 4 + 3*X_test + np.random.randn(20,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "224f7f92",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# train 데이터 scatter 시각화\n",
    "plt.figure(figsize=(15,8))\n",
    "plt.scatter(X_train, y_train)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "3df310ee",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# test 데이터 scatter 시각화\n",
    "plt.figure(figsize=(15,8))\n",
    "plt.scatter(X_test, y_test)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "d6448dc5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression()"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 모델 생성\n",
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "model = LinearRegression()\n",
    "model.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "21559ffe",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "기울기:  [2.77011339]\n",
      "y 절편:  [4.21509616]\n"
     ]
    }
   ],
   "source": [
    "# weight(기울기) 및 bias(y 절편) 구하기 \n",
    "bias, weight = model.intercept_, model.coef_[0]\n",
    "print('기울기: ', weight)\n",
    "print('y 절편: ', bias)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "147c4c1a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 모델이 생성한 선형 회귀 방정식이 train data에 대해 잘 나타내었는지 확인\n",
    "sample = np.linspace(0,2)\n",
    "\n",
    "plt.figure(figsize=(15,8))\n",
    "plt.scatter(X_train, y_train)\n",
    "plt.plot(sample, weight*sample+bias, color='red')\n",
    "# plt.plot(sample, bias*sample+weight, color='red')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "de094445",
   "metadata": {},
   "source": [
    "- 따라서 모델이 생성한 선형 회귀 방정식은 다음과 같이 나타낼 수 있음\n",
    "$$Linear\\, Regression\\, Equation : y = wx + b$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ee2c54a2",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# test 데이터에 대해 predict 진행\n",
    "output = model.predict(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "54b2bbae",
   "metadata": {},
   "source": [
    "### loss function은 MAE로 진행\n",
    "$$loss\\, function\\, : mean\\, absolute\\, error\\, = \\frac{1}{n}\\sum_{i=1}^n|y_{pred}-y_{label}|$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c754674c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mae :  0.6860341986110177\n"
     ]
    }
   ],
   "source": [
    "# loss function 만들기 (mean absolute error)\n",
    "def mae(output, target):\n",
    "    loss = np.abs(output-target)\n",
    "    loss = loss.sum()\n",
    "    loss = loss/len(output)\n",
    "    return loss\n",
    "\n",
    "# 아니면 다음과 같이 정의할 수 있음\n",
    "mae2 = lambda output, target: np.abs(output-target).sum()/len(output)\n",
    "\n",
    "# loss 구하기\n",
    "print('mae : ', mae(output, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "6b9a37c7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "rmse : 0.8049770741149787\n"
     ]
    }
   ],
   "source": [
    "# RMSE 구현\n",
    "def rmse(output, target):\n",
    "    loss = (output-target)**2  \n",
    "    loss = loss.sum()/len(output)\n",
    "    loss = np.sqrt(loss)\n",
    "    return loss\n",
    "\n",
    "print('rmse :', rmse(output, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "a112282c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# predicted와 label 비교\n",
    "plt.figure(figsize=(15,8))\n",
    "plt.scatter(X_test, y_test, label='Label', color='blue')\n",
    "plt.scatter(X_test, output, label='Predicted', color='red')\n",
    "plt.legend(loc='best')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a5dc0fb7",
   "metadata": {},
   "source": [
    "## Linear Classification  \n",
    "- Goal : 여러 클래스의 데이터가 주어지면 이 클래스들을 구별할 수 있는 분류 경계를 그리는 것"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "c4eed43d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0.2  0.4 ]\n",
      " [0.4  0.2 ]\n",
      " [0.65 0.3 ]\n",
      " [0.8  0.5 ]\n",
      " [0.5  0.8 ]\n",
      " [0.25 0.8 ]]\n",
      "[[0]\n",
      " [0]\n",
      " [0]\n",
      " [1]\n",
      " [1]\n",
      " [1]]\n"
     ]
    }
   ],
   "source": [
    "# 데이터 생성\n",
    "X_train = np.array([[0.2,0.4], [0.4,0.2], [0.65,0.3],  # class 1\n",
    "                    [0.8,0.5], [0.5,0.8], [0.25,0.8]]) # class 2\n",
    "y_train = np.array([[0], [0], [0],   # class 1\n",
    "                    [1], [1], [1]])  # class 2\n",
    "\n",
    "print(X_train)\n",
    "print(y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "6a8bd909",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2oAAAHSCAYAAACdLTg6AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8/fFQqAAAACXBIWXMAAAsTAAALEwEAmpwYAAAhN0lEQVR4nO3dcYzd5X3n+893jRHOkoVN7N2CTYrpUsjGeJzsBJpLA81yuaZ3SyAJzRKiILqroOxVAt0obmDhEgQhSkSljfaSyoIkqiKhJU5LqbNN6l0FbbsityrDQm0MODF0Ew/+o46paRKMsN1n/5gBBnuMz5gZn8ee10uy7PM7j3/zjZ+cwW+f3/ymWmsBAACgH/9g2AMAAADwWkINAACgM0INAACgM0INAACgM0INAACgM0INAACgM8cN6wMvXry4nX766cP68AAAAEP1yCOP/KS1tmS654YWaqeffnrGxsaG9eEBAACGqqp+dLDnXPoIAADQGaEGAADQGaEGAADQmaF9jRoAAHDs2bNnT8bHx/Piiy8Oe5RunHDCCVm2bFkWLlw48O8RagAAwKwZHx/Pm9/85px++umpqmGPM3SttezcuTPj4+NZvnz5wL/PpY8AAMCsefHFF/PWt75VpE2qqrz1rW+d8TuMQg0AAJhVIu21DufPQ6gBAADHvFtvvTW/+7u/Oyfnvummm3LaaaflxBNPnLVzCjUAAIA34NJLL81f/uVfzuo5hRoAADA0Dzz6bM7/4oNZfsOf5PwvPpgHHn32DZ/zG9/4RlauXJmRkZF87GMfO+D5e+65J+9+97szMjKSD33oQ3nhhReSJN/61reyYsWKjIyM5IILLkiSbN68Oeeee25WrVqVlStX5oc//OEB5/uVX/mVnHLKKW947qnc9REAABiKBx59Njfevym79+xLkjy7a3duvH9TkuTydy49rHNu3rw5d9xxRx566KEsXrw4zz333AFrPvjBD+bjH/94kuTmm2/O1772tXzqU5/Kbbfdlg0bNmTp0qXZtWtXkmTt2rW5/vrr89GPfjQvvfRS9u3bd1hzzZR31AAAgKG4c8OWVyLtZbv37MudG7Yc9jkffPDBXHHFFVm8eHGS5C1vecsBax5//PG8973vzTnnnJN77703mzdvTpKcf/75ueaaa3LPPfe8EmTvec978oUvfCFf+tKX8qMf/SiLFi067NlmQqgBAABDsX3X7hkdH0Rr7ZB3Wbzmmmty1113ZdOmTfnc5z73yq3z165dm89//vPZtm1bVq1alZ07d+aqq67K+vXrs2jRoqxevToPPvjgYc82E0INAAAYilNPnv7dqYMdH8RFF12UdevWZefOnUky7aWPP/3pT3PKKadkz549uffee185/vTTT+e8887LbbfdlsWLF2fbtm155plncsYZZ+S6667L+9///mzcuPGwZ5uJgUKtqi6pqi1VtbWqbpjm+ZOq6ttV9VdVtbmqfmv2RwUAAI4la1aflUULF7zm2KKFC7Jm9VmHfc53vOMduemmm3LhhRdmZGQkn/70pw9Yc/vtt+e8887LxRdfnLPPPvvVedasyTnnnJMVK1bkggsuyMjISL75zW9mxYoVWbVqVZ566qlcffXVB5zvd37nd7Js2bK88MILWbZsWW699dbDnv9l1Vp7/QVVC5L8IMnFScaTPJzkI621J6as+Q9JTmqtfbaqliTZkuQXWmsvHey8o6OjbWxs7A3/D2CWbFyXfO+25Pnx5KRlyUW3JCs/POypAOaGz3kAc+bJJ5/M29/+9oHXP/Dos7lzw5Zs37U7p568KGtWn3XYNxLp2XR/LlX1SGttdLr1g9z18dwkW1trz0ye7L4klyV5YsqaluTNNXEx6IlJnkuyd+bjMxQb1yXfvi7ZM3kt8PPbJh4n/uICHHt8zgPoyuXvXHpMhtkbNcilj0uTbJvyeHzy2FR3JXl7ku1JNiW5vrX297MyIXPve7e9+heWl+3ZPXEc4Fjjcx4AR4FBQm26W6bsf73k6iSPJTk1yaokd1XVPzrgRFXXVtVYVY3t2LFjhqMyZ54fn9lxgKOZz3kAHAUGCbXxJKdNebwsE++cTfVbSe5vE7Ym+eskZ++3Jq21u1tro6210SVLlhzuzMy2k5bN7DjA0cznPACOAoOE2sNJzqyq5VV1fJIrk6zfb82Pk1yUJFX1T5OcleSZ2RyUOXTRLcnC/W6BunDRxHGAY43PeQAcBQ55M5HW2t6q+mSSDUkWJPl6a21zVX1i8vm1SW5P8vtVtSkTl0p+trX2kzmcm9n08hfPuwMaMB/4nAfAUWCQuz6mtfadJN/Z79jaKb/enuT/mt3ROKJWfthfUoD5w+c8gHnn1ltvzYknnpjPfOYzs3reF154Ib/5m7+Zp59+OgsWLMill16aL37xi2/4vAN9w2sAAACm95nPfCZPPfVUHn300Tz00EP57ne/+4bPKdQAAIDh2bgu+Y8rkltPnvh547o3fMpvfOMbWblyZUZGRvKxj33sgOfvueeevPvd787IyEg+9KEP5YUXXkiSfOtb38qKFSsyMjKSCy64IEmyefPmnHvuuVm1alVWrlyZH/7wh68515ve9Ka8733vS5Icf/zxede73pXx8Td+J+GBLn0EAACYdRvXJd++7tXvb/n8tonHyWFfor558+bccccdeeihh7J48eI899xzB6z54Ac/mI9//ONJkptvvjlf+9rX8qlPfSq33XZbNmzYkKVLl2bXrl1JkrVr1+b666/PRz/60bz00kvZt2/fQT/2rl278u1vfzvXX3/9Yc0+lXfUAACA4fjeba9G2sv27J44fpgefPDBXHHFFVm8eHGS5C1vecsBax5//PG8973vzTnnnJN77703mzdvTpKcf/75ueaaa3LPPfe8EmTvec978oUvfCFf+tKX8qMf/SiLFi064HxJsnfv3nzkIx/JddddlzPOOOOw53+ZUAMAAIbj+YNcIniw4wNoraWqXnfNNddck7vuuiubNm3K5z73ubz44otJJt49+/znP59t27Zl1apV2blzZ6666qqsX78+ixYtyurVq/Pggw9Oe85rr702Z555Zn77t3/7sGefSqgBAADDcdKymR0fwEUXXZR169Zl586dSTLtpY8//elPc8opp2TPnj259957Xzn+9NNP57zzzsttt92WxYsXZ9u2bXnmmWdyxhln5Lrrrsv73//+bNy48YDz3XzzzXn++efz5S9/+bDn3p9QAwAAhuOiW5KF+11KuHDRxPHD9I53vCM33XRTLrzwwoyMjOTTn/70AWtuv/32nHfeebn44otz9tlnv3J8zZo1Oeecc7JixYpccMEFGRkZyTe/+c2sWLEiq1atylNPPZWrr776NecaHx/PHXfckSeeeCLvete7smrVqnz1q1897PlfVq21N3ySwzE6OtrGxsaG8rEBAIC58eSTT+btb3/74L9h47qJr0l7fnzinbSLbjkmv9fldH8uVfVIa210uvXu+ggAAAzPyg8fk2H2Rrn0EQAAoDNCDQAAoDNCDQAAmFXDug9Grw7nz0OoAQAAs+aEE07Izp07xdqk1lp27tyZE044YUa/z81EAACAWbNs2bKMj49nx44dwx6lGyeccEKWLZvZ94YTagAAwKxZuHBhli9fPuwxjnoufQQAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOjMQKFWVZdU1Zaq2lpVN0zz/Jqqemzyx+NVta+q3jL74wIAABz7DhlqVbUgyVeS/HqSf57kI1X1z6euaa3d2Vpb1VpbleTGJH/WWntuDuYFAAA45g3yjtq5Sba21p5prb2U5L4kl73O+o8k+c+zMRwAAMB8NEioLU2ybcrj8cljB6iqNyW5JMkfvvHRAAAA5qdBQq2mOdYOsvbSJA8d7LLHqrq2qsaqamzHjh2DzggAADCvDBJq40lOm/J4WZLtB1l7ZV7nssfW2t2ttdHW2uiSJUsGnxIAAGAeGSTUHk5yZlUtr6rjMxFj6/dfVFUnJbkwyR/P7ogAAADzy3GHWtBa21tVn0yyIcmCJF9vrW2uqk9MPr92cukHkvzX1trP52xaAACAeaBaO9iXm82t0dHRNjY2NpSPDQAAMGxV9UhrbXS65wb6htcAAAAcOUINAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgM0INAACgMwOFWlVdUlVbqmprVd1wkDW/VlWPVdXmqvqz2R0TAABg/jjuUAuqakGSryS5OMl4koeran1r7Ykpa05O8ntJLmmt/biq/skczQsAAHDMG+QdtXOTbG2tPdNaeynJfUku22/NVUnub639OElaa38zu2MCAADMH4OE2tIk26Y8Hp88NtUvJ/nHVfXfq+qRqrp6uhNV1bVVNVZVYzt27Di8iQEAAI5xg4RaTXOs7ff4uCT/Ism/SrI6yf9bVb98wG9q7e7W2mhrbXTJkiUzHhYAAGA+OOTXqGXiHbTTpjxelmT7NGt+0lr7eZKfV9WfJxlJ8oNZmRIAAGAeGeQdtYeTnFlVy6vq+CRXJlm/35o/TvLeqjquqt6U5LwkT87uqAAAAPPDId9Ra63trapPJtmQZEGSr7fWNlfVJyafX9tae7Kq/jTJxiR/n+SrrbXH53JwAACAY1W1tv+Xmx0Zo6OjbWxsbCgfGwAAYNiq6pHW2uh0zw30Da8BAAA4coQaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZwYKtaq6pKq2VNXWqrphmud/raqer6rHJn/cMvujAgAAzMDGdcl/XJHcevLEzxvXDXuigR13qAVVtSDJV5JcnGQ8ycNVtb619sR+S/9Ha+035mBGAACAmdm4Lvn2dcme3ROPn9828ThJVn54eHMNaJB31M5NsrW19kxr7aUk9yW5bG7HAgAAeAO+d9urkfayPbsnjh8FBgm1pUm2TXk8Pnlsf++pqr+qqu9W1TumO1FVXVtVY1U1tmPHjsMYFwAAYADPj8/seGcGCbWa5ljb7/H/TPKLrbWRJP9fkgemO1Fr7e7W2mhrbXTJkiUzGhQAAGBgJy2b2fHODBJq40lOm/J4WZLtUxe01v6utfazyV9/J8nCqlo8a1MCAADMxEW3JAsXvfbYwkUTx48Cg4Taw0nOrKrlVXV8kiuTrJ+6oKp+oapq8tfnTp5352wPCwAAMJCVH04u/U/JSaclqYmfL/1PR8WNRJIB7vrYWttbVZ9MsiHJgiRfb61trqpPTD6/NskVSf5dVe1NsjvJla21/S+PBAAAOHJWfvioCbP91bB6anR0tI2NjQ3lYwMAAAxbVT3SWhud7rmBvuE1AAAAR45QAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6IxQAwAA6MxAoVZVl1TVlqraWlU3vM66d1fVvqq6YvZGPDIeePTZnP/FB7P8hj/J+V98MA88+uywRwIAAOap4w61oKoWJPlKkouTjCd5uKrWt9aemGbdl5JsmItB59IDjz6bG+/flN179iVJnt21OzfevylJcvk7lw5zNAAAYB4a5B21c5Nsba0901p7Kcl9SS6bZt2nkvxhkr+ZxfmOiDs3bHkl0l62e8++3Llhy5AmAgAA5rNBQm1pkm1THo9PHntFVS1N8oEka1/vRFV1bVWNVdXYjh07ZjrrnNm+a/eMjgMAAMylQUKtpjnW9nv85SSfba3tm2btq7+ptbtba6OttdElS5YMOOLcO/XkRTM6DgAAMJcGCbXxJKdNebwsyfb91owmua+q/leSK5L8XlVdPhsDHglrVp+VRQsXvObYooULsmb1WUOaCAAAmM8OeTORJA8nObOqlid5NsmVSa6auqC1tvzlX1fV7yf5L621B2ZvzLn18g1D7tywJdt37c6pJy/KmtVnuZEIAAAwFIcMtdba3qr6ZCbu5rggyddba5ur6hOTz7/u16UdLS5/51JhBgAAdGGQd9TSWvtOku/sd2zaQGutXfPGxwIAAJi/BvqG1wAAABw5Qg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzQg0AAKAzA4VaVV1SVVuqamtV3TDN85dV1caqeqyqxqrqV2d/VAAAgPnhuEMtqKoFSb6S5OIk40kerqr1rbUnpiz7XpL1rbVWVSuTrEty9lwMDADAzD3w6LO5c8OWbN+1O6eevChrVp+Vy9+5dNhjAQdxyFBLcm6Sra21Z5Kkqu5LclmSV0KttfazKev/YZI2m0MCAHD4Hnj02dx4/6bs3rMvSfLsrt258f5NSSLWoFODXPq4NMm2KY/HJ4+9RlV9oKqeSvInSf7N7IwHAMAbdeeGLa9E2st279mXOzdsGdJEwKEMEmo1zbED3jFrrf1Ra+3sJJcnuX3aE1VdO/k1bGM7duyY0aAAABye7bt2z+g4MHyDhNp4ktOmPF6WZPvBFrfW/jzJL1XV4mmeu7u1NtpaG12yZMmMhwUAYOZOPXnRjI4DwzdIqD2c5MyqWl5Vxye5Msn6qQuq6p9VVU3++l1Jjk+yc7aHBQBg5tasPiuLFi54zbFFCxdkzeqzhjQRcCiHvJlIa21vVX0yyYYkC5J8vbW2uao+Mfn82iQfSnJ1Ve1JsjvJv26tuaEIAEAHXr5hiLs+wtGjhtVTo6OjbWxsbCgfGwAAYNiq6pHW2uh0zw30Da8BAAA4coQaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZ4QaAABAZwYKtaq6pKq2VNXWqrphmuc/WlUbJ398v6pGZn9UAACA+eG4Qy2oqgVJvpLk4iTjSR6uqvWttSemLPvrJBe21v62qn49yd1JzpuLgQFmywOPPps7N2zJ9l27c+rJi7Jm9Vm5/J1Lhz0WAMChQy3JuUm2ttaeSZKqui/JZUleCbXW2venrP+LJMtmc0iA2fbAo8/mxvs3ZfeefUmSZ3ftzo33b0oSsQYADN0glz4uTbJtyuPxyWMH82+TfPeNDAUw1+7csOWVSHvZ7j37cueGLUOaCADgVYO8o1bTHGvTLqx6XyZC7VcP8vy1Sa5Nkre97W0Djggw+7bv2j2j4wAAR9Ig76iNJzltyuNlSbbvv6iqVib5apLLWms7pztRa+3u1tpoa210yZIlhzMvwKw49eRFMzoOAHAkDRJqDyc5s6qWV9XxSa5Msn7qgqp6W5L7k3ystfaD2R8TYHatWX1WFi1c8JpjixYuyJrVZw1pIgCAVx3y0sfW2t6q+mSSDUkWJPl6a21zVX1i8vm1SW5J8tYkv1dVSbK3tTY6d2MDvDEv3zDEXR8BgB5Va9N+udmcGx0dbWNjY0P52AAAAMNWVY8c7A2ugb7hNQAAAEeOUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOiMUAMAAOhMtdaG84GrdiT50VA++OtbnOQnwx6CobD385e9n7/s/fxl7+cn+z5/9br3v9haWzLdE0MLtV5V1VhrbXTYc3Dk2fv5y97PX/Z+/rL385N9n7+Oxr136SMAAEBnhBoAAEBnhNqB7h72AAyNvZ+/7P38Ze/nL3s/P9n3+euo23tfowYAANAZ76gBAAB0Zt6GWlVdUlVbqmprVd0wzfMfraqNkz++X1Ujw5iT2TfA3l82ue+PVdVYVf3qMOZk9h1q76ese3dV7auqK47kfMyNAV7zv1ZVz0++5h+rqluGMSezb5DX/OT+P1ZVm6vqz470jMyNAV73a6a85h+f/Jz/lmHMyuwaYO9PqqpvV9VfTb7uf2sYcw5iXl76WFULkvwgycVJxpM8nOQjrbUnpqz5P5I82Vr726r69SS3ttbOG8rAzJoB9/7EJD9vrbWqWplkXWvt7KEMzKwZZO+nrPtvSV5M8vXW2h8c6VmZPQO+5n8tyWdaa78xjBmZGwPu/clJvp/kktbaj6vqn7TW/mYY8zJ7Bv18P2X9pUn+fWvtXx65KZkLA77u/0OSk1prn62qJUm2JPmF1tpLw5j59czXd9TOTbK1tfbM5Kbcl+SyqQtaa99vrf3t5MO/SLLsCM/I3Bhk73/WXv0XjH+YZP79a8ax6ZB7P+lTSf4wib+sHRsG3XeOPYPs/VVJ7m+t/ThJRNoxY6av+48k+c9HZDLm2iB735K8uaoqyYlJnkuy98iOOZj5GmpLk2yb8nh88tjB/Nsk353TiThSBtr7qvpAVT2V5E+S/JsjNBtz65B7X1VLk3wgydojOBdza9DP9++ZvAzmu1X1jiMzGnNskL3/5ST/uKr+e1U9UlVXH7HpmEsD/z2vqt6U5JJM/AMdR79B9v6uJG9Psj3JpiTXt9b+/siMNzPHDXuAIalpjk37rklVvS8ToebrlI4NA+19a+2PkvxRVV2Q5PYk/+dcD8acG2Tvv5zks621fRP/0MYxYJB9/59JfrG19rOq+r+TPJDkzLkejDk3yN4fl+RfJLkoyaIk/39V/UVr7QdzPRxzauC/5yW5NMlDrbXn5nAejpxB9n51kseS/Mskv5Tkv1XV/2it/d0czzZj8/UdtfEkp015vCwTVf0ak1+f9NUkl7XWdh6h2ZhbA+39y1prf57kl6pq8VwPxpwbZO9Hk9xXVf8ryRVJfq+qLj8i0zFXDrnvrbW/a639bPLX30my0Gv+mDDIa348yZ+21n7eWvtJkj9P4uZhR7+Z/Lf+yrjs8VgyyN7/ViYueW6tta1J/jpJl/cimK+h9nCSM6tqeVUdn4kX6fqpC6rqbUnuT/Ix/7J2TBlk7//Z5HXLqap3JTk+iVA/+h1y71try1trp7fWTk/yB0n+n9baA0d8UmbTIK/5X5jymj83E/9t9Jo/+h1y75P8cZL3VtVxk5fAnZfkySM8J7NvkL1PVZ2U5MJM/P+AY8Mge//jTLyLnqr6p0nOSvLMEZ1yQPPy0sfW2t6q+mSSDUkWZOLObpur6hOTz69NckuSt2biX9STZG9rbXRYMzM7Btz7DyW5uqr2JNmd5F9PubkIR6kB955jzID7fkWSf1dVezPxmr/Sa/7oN8jet9aerKo/TbIxyd8n+Wpr7fHhTc1smMHn+w8k+a+ttZ8PaVRm2YB7f3uS36+qTZm4VPKzk++od2de3p4fAACgZ/P10kcAAIBuCTUAAIDOCDUAAIDOCDUAAIDOCDUAAIDOCDUAAIDOCDUAAIDOCDUAAIDO/G9WNjq8QFNWSgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 데이터 분포 시각화\n",
    "plt.figure(figsize=(15,8))\n",
    "plt.scatter(X_train[:3,0], X_train[:3,1], label='class 1')\n",
    "plt.scatter(X_train[3:,0], X_train[3:,1], label='class 2')\n",
    "plt.legend(loc='best')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "828f15f1",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\user\\anaconda3\\envs\\tf2\\lib\\site-packages\\sklearn\\utils\\validation.py:985: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n",
      "C:\\Users\\user\\anaconda3\\envs\\tf2\\lib\\site-packages\\sklearn\\linear_model\\_stochastic_gradient.py:696: ConvergenceWarning: Maximum number of iteration reached before convergence. Consider increasing max_iter to improve the fit.\n",
      "  warnings.warn(\n",
      "C:\\Users\\user\\anaconda3\\envs\\tf2\\lib\\site-packages\\sklearn\\utils\\validation.py:985: DataConversionWarning: A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().\n",
      "  y = column_or_1d(y, warn=True)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "SGDClassifier(max_iter=100)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 모델 학습\n",
    "from sklearn.linear_model import SGDClassifier\n",
    "\n",
    "model1 = SGDClassifier(max_iter=10)\n",
    "model1.fit(X_train, y_train)\n",
    "\n",
    "model2 = SGDClassifier(max_iter=100)\n",
    "model2.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "12fc6b9f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accuracy of training 1 : 100.0 %\n",
      "accuracy of training 2 : 100.0 %\n"
     ]
    }
   ],
   "source": [
    "# train data에 대해 모델의 정확도 체크\n",
    "train_acc1 = model1.score(X_train, y_train)\n",
    "print('accuracy of training 1 :', train_acc1*100, '%')\n",
    "\n",
    "train_acc2 = model2.score(X_train, y_train)\n",
    "print('accuracy of training 2 :', train_acc2*100, '%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "efd61712",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "first weight 1: 17.94145420207743\n",
      "second weight 1: 20.774315391879117\n",
      "bias 1: [-19.46186535]\n"
     ]
    }
   ],
   "source": [
    "# weight 및 bias 구하기\n",
    "weight11, weight21 = model1.coef_[0]\n",
    "bias1 = model1.intercept_\n",
    "print('first weight 1:', weight11)\n",
    "print('second weight 1:', weight21)\n",
    "print('bias 1:', bias1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "3f08e876",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "first weight 2: 5.223171889838555\n",
      "second weight 2: 14.245014245014238\n",
      "bias 2: [-9.94075081]\n"
     ]
    }
   ],
   "source": [
    "# weight 및 bias 구하기\n",
    "weight12, weight22 = model2.coef_[0]\n",
    "bias2 = model2.intercept_\n",
    "print('first weight 2:', weight12)\n",
    "print('second weight 2:', weight22)\n",
    "print('bias 2:', bias2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9fc7057d",
   "metadata": {},
   "source": [
    "### weight들과 bias를 가지고 decision boundary 그리기   \n",
    "- $n$차원 공간에 있는 데이터 사이의 decision boundary는 $n-1$ 차원임  \n",
    "- 즉, 2차원 공간에 있는 데이터 사이의 decision boundary는 1차원 즉, line 임  \n",
    "- 이 예제에서 사용한 데이터는 2차원임    \n",
    "- 따라서 이에 맞춰 식을 변경할 필요가 있음  \n",
    "- 위의 데이터 정의에 의해 $x$축은 $w_1$에 해당, $y$축은 $w_2$에 해당  \n",
    "$$decision\\, boundary\\, :\\, w_1x_1+w_2x_2+b=0$$  \n",
    "- $x_2$에 대한 식으로 표현하면   \n",
    "$$\\Rightarrow x_2=-\\frac{w1}{w2}x-\\frac{b}{w2}$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "426789cd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 결정 경계선\n",
    "x1 = np.linspace(0,1,100)\n",
    "y1 = (-weight11/weight21) * x - (bias1/weight21)\n",
    "\n",
    "plt.figure(figsize=(15,8))\n",
    "plt.plot(x1,y1)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "c5341fbf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 결정 경계선\n",
    "x2 = np.linspace(0,1,100)\n",
    "y2 = (-weight12/weight22) * x - (bias2/weight22)\n",
    "\n",
    "plt.figure(figsize=(15,8))\n",
    "plt.plot(x2,y2)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "dc6e1927",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,8))\n",
    "plt.scatter(X_train[:3,0], X_train[:3,1], label='class 0')\n",
    "plt.scatter(X_train[3:,0], X_train[3:,1], label='class 1')\n",
    "plt.plot(x1, y1, label='decision boundary max iter 10')\n",
    "plt.plot(x2, y2, label='decision boundary max iter 100')\n",
    "plt.legend(loc='best')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f5e3a5f7",
   "metadata": {},
   "source": [
    "# Homework  \n",
    "- 실습한 내용들 가지고 복습  \n",
    "- 데이터를 더 많이 늘려서 regression, classification 모델 학습 시켜보기  \n",
    "- linear regression model의 loss function을 RMSE로 변경해서 실습  \n",
    "- 텐서플로, Neural Network에 대해서 간단하게 살펴보기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "eb249cf0",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}