11장_비지도학습_5절_가우시안_혼합_모델.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 개별 코드 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 데이터 생성\n",
    "from sklearn.datasets import make_blobs\n",
    "X,y = make_blobs(n_samples=200, \n",
    "                 n_features=2, \n",
    "                 centers=2, \n",
    "                 random_state=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(200, 2)\n",
      "(200,)\n"
     ]
    }
   ],
   "source": [
    "# 생성된 데이터 차원 확인\n",
    "print(X.shape)\n",
    "print(y.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "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": [
    "# 라벨링 되지 않은 데이터 플랏\n",
    "import matplotlib.pyplot as plt\n",
    "plt.scatter(X[:,0], X[:,1], \n",
    "            c='gray', \n",
    "            edgecolor='black', \n",
    "            marker='o')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1 0 1 1 0 1 0 0 1 0 1 0 0 1 0 1 1 1 1 1 1 0 1 1 1 1 0 1 1 0 0 0 1 0 0 1 0\n",
      " 1 1 0 1 0 1 1 0 0 0 0 0 1 0 1 0 0 1 1 1 0 1 0 1 0 1 0 0 1 0 0 1 1 1 0 0 1\n",
      " 0 1 1 1 1 0 1 0 0 0 0 1 1 0 1 1 0 0 0 0 1 0 1 1 1 0 0 1 0 0 1 0 0 0 1 0 0\n",
      " 1 1 1 1 0 1 1 1 1 1 1 0 0 0 1 0 0 1 1 1 0 0 1 0 0 0 0 1 1 0 1 1 0 0 0 0 1\n",
      " 1 0 1 0 1 1 0 0 1 1 0 0 1 1 1 0 0 1 1 0 0 1 1 0 0 0 0 0 1 1 0 1 0 1 0 1 1\n",
      " 0 0 1 0 0 1 1 0 1 1 0 0 0 1 0]\n"
     ]
    }
   ],
   "source": [
    "# 가우시안 혼합 모형\n",
    "from sklearn.mixture import GaussianMixture\n",
    "gmm = GaussianMixture(n_components=2, \n",
    "                      random_state=0)\n",
    "gmm.fit(X)\n",
    "label_gmm = gmm.predict(X)\n",
    "print(label_gmm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>gmm_comp1</th>\n",
       "      <th>gmm_comp2</th>\n",
       "      <th>target</th>\n",
       "      <th>label_gmm</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.381155</td>\n",
       "      <td>-1.644815</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.583904</td>\n",
       "      <td>-8.784059</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.830865</td>\n",
       "      <td>-1.425235</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.813135</td>\n",
       "      <td>-1.924671</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-1.655771</td>\n",
       "      <td>-9.555946</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   gmm_comp1  gmm_comp2  target  label_gmm\n",
       "0   0.381155  -1.644815       1          1\n",
       "1  -0.583904  -8.784059       0          0\n",
       "2   1.830865  -1.425235       1          1\n",
       "3   1.813135  -1.924671       1          1\n",
       "4  -1.655771  -9.555946       0          0"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 시각화를 위한 데이터프레임 생성\n",
    "import pandas as pd\n",
    "gmm_columns = ['gmm_comp1', 'gmm_comp2']\n",
    "X_gmm_df = pd.DataFrame(X, columns=gmm_columns)\n",
    "X_gmm_df['target'] = y\n",
    "X_gmm_df['label_gmm'] = label_gmm\n",
    "X_gmm_df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{0, 1}\n",
      "{0, 1}\n"
     ]
    }
   ],
   "source": [
    "# 타겟 클래스 종류 확인\n",
    "print(set(X_gmm_df['target']))\n",
    "print(set(X_gmm_df['label_gmm']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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": [
    "# 가우시안 혼합 모형을 이용한 데이터 플랏\n",
    "df = X_gmm_df\n",
    "markers=['o','x']\n",
    "\n",
    "for i, mark in enumerate(markers):\n",
    "    df_i = df[df['label_gmm']==i]\n",
    "    target_i = i\n",
    "    X1 = df_i['gmm_comp1']\n",
    "    X2 = df_i['gmm_comp2']\n",
    "    plt.scatter(X1, X2, \n",
    "                marker=mark, \n",
    "                label=target_i)\n",
    "\n",
    "plt.xlabel('gmm_component1')\n",
    "plt.ylabel('gmm_component2')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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": [
    "# 실제 타겟 플랏\n",
    "df = X_gmm_df\n",
    "markers=['o','x']\n",
    "\n",
    "for i, mark in enumerate(markers):\n",
    "    df_i = df[df['target']==i]\n",
    "    target_i = i\n",
    "    X1 = df_i['gmm_comp1']\n",
    "    X2 = df_i['gmm_comp2']\n",
    "    plt.scatter(X1, X2, \n",
    "                marker=mark, \n",
    "                label=target_i)\n",
    "\n",
    "plt.xlabel('gmm_component1')\n",
    "plt.ylabel('gmm_component2')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7842908753561848\n"
     ]
    }
   ],
   "source": [
    "# 모형 평가\n",
    "from sklearn.metrics import silhouette_score\n",
    "sil_score = silhouette_score(X, label_gmm)\n",
    "print(sil_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 통합 코드"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(200, 2)\n",
      "(200,)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1 0 1 1 0 1 0 0 1 0 1 0 0 1 0 1 1 1 1 1 1 0 1 1 1 1 0 1 1 0 0 0 1 0 0 1 0\n",
      " 1 1 0 1 0 1 1 0 0 0 0 0 1 0 1 0 0 1 1 1 0 1 0 1 0 1 0 0 1 0 0 1 1 1 0 0 1\n",
      " 0 1 1 1 1 0 1 0 0 0 0 1 1 0 1 1 0 0 0 0 1 0 1 1 1 0 0 1 0 0 1 0 0 0 1 0 0\n",
      " 1 1 1 1 0 1 1 1 1 1 1 0 0 0 1 0 0 1 1 1 0 0 1 0 0 0 0 1 1 0 1 1 0 0 0 0 1\n",
      " 1 0 1 0 1 1 0 0 1 1 0 0 1 1 1 0 0 1 1 0 0 1 1 0 0 0 0 0 1 1 0 1 0 1 0 1 1\n",
      " 0 0 1 0 0 1 1 0 1 1 0 0 0 1 0]\n",
      "{0, 1}\n",
      "{0, 1}\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7842908753561848\n"
     ]
    }
   ],
   "source": [
    "from sklearn.datasets import make_blobs\n",
    "\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.mixture import GaussianMixture\n",
    "from sklearn.metrics import silhouette_score\n",
    "\n",
    "# 데이터 생성\n",
    "X,y = make_blobs(n_samples=200, \n",
    "                 n_features=2, \n",
    "                 centers=2, \n",
    "                 random_state=2)\n",
    "\n",
    "# 생성된 데이터 차원 확인\n",
    "print(X.shape)\n",
    "print(y.shape)\n",
    "\n",
    "# 라벨링 되지 않은 데이터 플랏\n",
    "plt.scatter(X[:,0], X[:,1], \n",
    "            c='gray', \n",
    "            edgecolor='black', \n",
    "            marker='o')\n",
    "plt.show()\n",
    "\n",
    "# 가우시안 혼합 모형\n",
    "gmm = GaussianMixture(n_components=2, \n",
    "                      random_state=0)\n",
    "gmm.fit(X)\n",
    "label_gmm = gmm.predict(X)\n",
    "print(label_gmm)\n",
    "\n",
    "# 시각화를 위한 데이터프레임 생성\n",
    "gmm_columns = ['gmm_comp1', 'gmm_comp2']\n",
    "X_gmm_df = pd.DataFrame(X, columns=gmm_columns)\n",
    "X_gmm_df['target'] = y\n",
    "X_gmm_df['label_gmm'] = label_gmm\n",
    "X_gmm_df.head(5)\n",
    "\n",
    "# 타겟 클래스 종류 확인\n",
    "print(set(X_gmm_df['target']))\n",
    "print(set(X_gmm_df['label_gmm']))\n",
    "\n",
    "# 가우시안 혼합 모형을 이용한 데이터 플랏\n",
    "df = X_gmm_df\n",
    "markers=['o','x']\n",
    "\n",
    "for i, mark in enumerate(markers):\n",
    "    df_i = df[df['label_gmm']==i]\n",
    "    target_i = i\n",
    "    X1 = df_i['gmm_comp1']\n",
    "    X2 = df_i['gmm_comp2']\n",
    "    plt.scatter(X1, X2, \n",
    "                marker=mark, \n",
    "                label=target_i)\n",
    "\n",
    "plt.xlabel('gmm_component1')\n",
    "plt.ylabel('gmm_component2')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "# 실제 타겟 플랏\n",
    "df = X_gmm_df\n",
    "markers=['o','x']\n",
    "\n",
    "for i, mark in enumerate(markers):\n",
    "    df_i = df[df['target']==i]\n",
    "    target_i = i\n",
    "    X1 = df_i['gmm_comp1']\n",
    "    X2 = df_i['gmm_comp2']\n",
    "    plt.scatter(X1, X2, \n",
    "                marker=mark, \n",
    "                label=target_i)\n",
    "\n",
    "plt.xlabel('gmm_component1')\n",
    "plt.ylabel('gmm_component2')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "# 모형 평가\n",
    "sil_score = silhouette_score(X, label_gmm)\n",
    "print(sil_score)"
   ]
  },
  {
   "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
}