{"id":697,"date":"2020-12-17T10:45:13","date_gmt":"2020-12-17T01:45:13","guid":{"rendered":"http:\/\/cedartrees.co.kr\/?p=697"},"modified":"2021-04-03T19:10:05","modified_gmt":"2021-04-03T10:10:05","slug":"svm","status":"publish","type":"post","link":"http:\/\/blog.cedartrees.co.kr\/index.php\/2020\/12\/17\/svm\/","title":{"rendered":"\uc11c\ud3ec\ud2b8\ubca1\ud130\uba38\uc2e0(Support Vector Machine)"},"content":{"rendered":"\n

\uc11c\ud3ec\ud2b8\ubca1\ud130\uba38\uc2e0(SVM,Support Vector Machine, \uc774\ud558 SVM)\uc740 \uac15\ub825\ud55c \uba38\uc2e0\ub7ec\ub2dd \ubaa8\ub378\uc774\uba70 \ub610 \ub2e4\uc591\ud55c \uc815\ud615 \ub370\uc774\ud130\uc14b\uc5d0\uc11c \uc88b\uc740 \uc131\ub2a5\uc744 \ubcf4\uc5ec\uc90d\ub2c8\ub2e4. SVM\uc740 \ub370\uc774\ud130\uc758 \ud2b9\uc9d5\uc774 \uba87\uac1c \ub418\uc9c0 \uc54a\ub354\ub77c\ub3c4 \ubcf5\uc7a1\ud55c \uacb0\uc815\uacbd\uacc4(Decision Boundary)\ub97c \ub9cc\ub4e4\uc5b4\uc90d\ub2c8\ub2e4.<\/p>\n\n\n\n

SVM\uc740 \uc800\ucc28\uc6d0, \uace0\ucc28\uc6d0\uc758 \ubaa8\ub378\uc5d0\uc11c \uc88b\uc740 \uc131\ub2a5\uc744 \ubc1c\ud718\ud558\uc9c0\ub9cc \uc0d8\ud50c\uc774 \ub9ce\uc744 \uacbd\uc6b0 \uc608\ub97c \ub4e4\uc5b4 100,000\uac1c \uc774\uc0c1\uc758 \ub370\uc774\ud130\ub97c \ucc98\ub9ac\ud558\ub294\ub370\ub294 \uc131\ub2a5\uc774 \uc88b\uc9c0 \uc54a\uc2b5\ub2c8\ub2e4. \ub610 \ub370\uc774\ud130 \uc804\ucc98\ub9ac(MinMaxScaler\uc640 \uac19\uc740 \uc791\uc5c5\uc774 \ud544\uc694)\uc640 \ub9e4\uac1c\ubcc0\uc218 \uc124\uc815\uc5d0 \uc880 \ub354 \uace0\ubbfc\ud574\uc57c \ud560 \ubd80\ubd84\ub4e4\uc774 \ub9ce\uc774 \uc788\uace0 \uc608\uce21\uc774 \uc5b4\ub5a4 \uacb0\uacfc\ub97c \ud1b5\ud574\uc11c \uc774\ub904\uc84c\ub294\uc9c0\uc5d0 \ub300\ud55c \ubaa8\ub378 \uc124\uba85\uc774 \uc27d\uc9c0 \uc54a\uc740 \ubd80\ubd84\uc774 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n

\uadf8\ub7fc\uc5d0\ub3c4 SVM\uc740 \ud2b9\uc9d5\uc774 \ube44\uc2b7\ud55c \ub2e8\uc704\uc774\uace0 \uc0d8\ud50c \uc218\uac00 \ub9ce\uc9c0 \uc54a\ub2e4\uba74 \uc88b\uc740 \uacb0\uacfc\ub97c \uc5bb\uc744 \uc218 \uc788\ub294 \ubaa8\ub378\uc785\ub2c8\ub2e4.<\/p>\n\n\n\n

\uc544\ub798\uc640 \uac19\uc740 \uc608\uc81c\ub97c \ud1b5\ud574\uc11c SVM \ubaa8\ub378\uc744 \ud14c\uc2a4\ud2b8\ud574\ubcf4\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n

from sklearn.datasets import make_classification\nfrom sklearn.datasets import make_blobs\nfrom sklearn import svm\n\nimport numpy as np\n\nimport matplotlib.pyplot as plt\nfrom sklearn.model_selection import train_test_split, GridSearchCV<\/pre>\n\n\n\n
\ud14c\uc2a4\ud2b8 \ub370\uc774\ud130\ub294 sklearn\uc758 make_classification() \ud568\uc218\ub97c \uc0ac\uc6a9\ud574\uc11c \ub9cc\ub4e4\uc5c8\uc2b5\ub2c8\ub2e4. \uadf8\ub807\uae30 \ub54c\ubb38\uc5d0 \uc774 \ucf54\ub4dc\ub97c \ud1b5\ud574\uc11c \ud14c\uc2a4\ud2b8\ud558\uc2dc\ub294 \ubd84\ub4e4\uc740 \ub2e4\ub978 \ud615\ud0dc\uc758 \uadf8\ub798\ud504\uac00 \uadf8\ub824\uc9d1\ub2c8\ub2e4. \ub370\uc774\ud130\ub97c \uacc4\uc18d \uc0c8\ub85c \ub9cc\ub4e4\uba74\uc11c \uacb0\uc815\uacbd\uacc4\uac00 \uc5b4\ub5bb\uac8c \ud45c\uc2dc\ub418\ub294\uc9c0 \ud14c\uc2a4\ud2b8\ud574\ubcf4\uc2dc\ub294 \uac83\ub3c4 \uc88b\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n
\uc800\ub294 2\uac1c\uc758 feature\uac00 \uc788\ub294 1,000\uac1c\uc758 \uc0d8\ud50c \ub370\uc774\ud130\ub97c \uc0dd\uc131\ud588\uc2b5\ub2c8\ub2e4. \uc0dd\uc131\ud55c \ub370\uc774\ud130\ub97c \ud1b5\ud574\uc11c \uc2dc\uac01\ud654 \ud574\ubcf4\uba74 \uc544\ub798\uc640 \uac19\uc740 \uadf8\ub798\ud504\uac00 \ud45c\uc2dc\ub429\ub2c8\ub2e4. \uc774 \uadf8\ub798\ud504\ub294 \uc120\ud615\uc73c\ub85c \ubd84\ub958\uac00 \ubd88\uac00\ub2a5\ud569\ub2c8\ub2e4. \ub450\uac1c\uc758 \ud074\ub798\uc2a4\uac00 \uacb9\uccd0\uc788\uae30 \ub54c\ubb38\uc785\ub2c8\ub2e4. \uadf8\ub807\uae30 \ub54c\ubb38\uc5d0 \ucc28\uc6d0\uc744 \ub192\uc5ec\uc11c \ube44\uc120\ud615(Nonlinear) \ubc29\uc2dd\uc758 \ubd84\ub958\uac00 \ud544\uc694\ud569\ub2c8\ub2e4.<\/p>\n\n\n\n
\uc608\uc81c\ub97c \uc218\ud589\ud558\uae30\uc5d0 \uc55e\uc11c \uac04\ub2e8\ud788 \uc124\uba85\ud558\uba74 \uc11c\ud3ec\ud2b8\ubca1\ud130\uba38\uc2e0(Support Vector Machine)\uc740 \uc120\ud615 \ub370\uc774\ud130 \ubd84\ub958\uc640 \ube44\uc120\ud615 \ub370\uc774\ud130 \ubd84\ub958 \ubaa8\ub450 \uac00\ub2a5\ud569\ub2c8\ub2e4. \uadf8 \ubc29\ubc95\uc740 \ub370\uc774\ud130\ub97c \ubd84\ub958\ud560 \uc218 \uc788\ub294 \uc9c1\uc120\uc744 \uae0b\uace0 \uc9c1\uc120\uc744 \uae30\uc900\uc73c\ub85c \ub370\uc774\ud130\ub97c \ubd84\ub958\ud558\ub294 \uac83\uc785\ub2c8\ub2e4. \ud14c\uc2a4\ud2b8 \uc608\uc81c\uc778 \uadf8\ub9bc 1\uc758 \uacbd\uc6b0\ub294 \uc9c1\uc120\uc73c\ub85c \ubd84\ub958\ud560 \uc218 \uc5c6\uae30 \ub54c\ubb38\uc5d0 \ucee4\ub110(kernel)\uc774\ub77c\ub294 \ud615\ud0dc\uc758 \ubc29\ubc95\uc774 \ud544\uc694\ud558\uc9c0\ub9cc \uadf8\ub9bc2\uc758 \uacbd\uc6b0\ub294 \ub370\uc774\ud130\uac00 \uad49\uc7a5\ud788 \uba85\ud655\ud558\uac8c \uad6c\ubd84\ub418\uc5b4 \uc788\uae30 \ub54c\ubb38\uc5d0(\uba87\uac1c\uc758 \uc544\uc6c3\ub77c\uc774\uc5b4\uac00 \uc788\uc9c0\ub9cc…) \ud558\ub098\uc758 \uc9c1\uc120\uc73c\ub85c \ub450\uac1c\uc758 \ub370\uc774\ud130 \uadf8\ub8f9\uc744 \ubd84\ub9ac \ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.  \uc815\ub9ac\ud558\uba74 \uadf8\ub9bc1\uc740 \ube44\uc120\ud615, \uadf8\ub9bc 2\ub294 \uc120\ud615 \uc774\ub77c\uace0 \ud558\uace0 SVM\uc740 \uc774 \ub450\uac00\uc9c0 \ubaa8\ub450 \uac00\ub2a5\ud569\ub2c8\ub2e4.<\/p>\n\n\n\n
\uadf8\ub9bc2\uc5d0\uc11c \uc880 \ub354 \uc790\uc138\ud788 \uc124\uba85\ud558\uba74 \ub450\uac1c\uc758 \ub370\uc774\ud130 \uadf8\ub8f9\uc774 \uc788\uace0 \ud55c \uadf8\ub8f9\ub9c8\ub2e4 \ub370\uc774\ud130\uac00 \ub9ce\uc774 \uc874\uc7ac\ud558\uace0 \uc788\uc9c0\ub9cc \uc2e4\uc81c\ub85c \ubaa8\ub4e0 \ub370\uc774\ud130\uac00 \uacb0\uc815\uacbd\uacc4(Decision Boundary)\ub97c \ub9cc\ub4dc\ub294 \uac83\uc740 \uc544\ub2c8\uace0 \uc77c\ubd80\uc758 \ub370\uc774\ud130\ub9cc \uacb0\uc815\uacbd\uacc4\ub97c \ub9cc\ub4dc\ub294\ub370 \uc0ac\uc6a9\ub429\ub2c8\ub2e4. \uc774\uac83\uc740 \ube44\uad50\uc801 \uc801\uc740 \uc5f0\uc0b0 \uacfc\uc815\uc73c\ub85c\ub3c4 \uacb0\uacfc\ub97c \uc5bb\uc744 \uc218 \uc788\ub2e4\ub294 \uc758\ubbf8\uc774\uba70 \ub610 \uc18d\ub3c4\uac00 \ube60\ub974\ub2e4\ub294 \uc758\ubbf8\ub3c4\ub429\ub2c8\ub2e4.
\uc774\ub7ec\ud55c \ub370\uc774\ud130\ub4e4\uc744 \uc11c\ud3ec\ud2b8 \ubca1\ud130(Support Vector)\ub77c\uace0 \ud45c\ud604\ud569\ub2c8\ub2e4. \ud480\uc5b4\uc11c \ub9d0\ud558\uba74 \uacb0\uc815\uacbd\uacc4\ub97c \uc874\uc7ac\ud558\uac8c \ud558\ub294 \uc9c0\uc9c0 \ubca1\ud130\ub77c\uace0 \ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uadf8\ub7f0\ub370 \uc65c \uc810\uc774\ub77c\uace0 \ud558\uc9c0 \uc54a\uace0 \ubca1\ud130\ub77c\uace0 \ud588\uc744\uae4c\ub97c \uc0dd\uac01\ud574\ubcf4\uba74 \uc5b4\ub5a4 \ud3ec\uc778\ud2b8\uc758 \uc5f0\uc18d\ub41c \ub370\uc774\ud130 \ud615\ud0dc\ub97c \ud45c\ud604\ud558\uae30 \uc704\ud574\uc11c \ub77c\uace0 \uc774\ud574\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n
\uc0c9\uc744 \ud45c\uc2dc\ud558\uae30 \uc704\ud574\uc11c \uc544\ub798\uc640 \uac19\uc740 \uceec\ub7ec \ucf54\ub4dc\ub97c \uc0ac\uc6a9\ud588\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n
red_RGB = (1,0,0)\nblue_RGB = (0,0,1)\ndata_colors = [red_RGB, blue_RGB]\n\ndef get_colors(y):\n    return [data_colors[label] for label in y]<\/pre>\n\n\n\n
X,y = make_classification(n_samples=1000, n_features=2, n_informative=1, n_redundant=0, n_clusters_per_class=1)\nplt.scatter(X[:,0], X[:,1], c=get_colors(y), s=10)<\/pre>\n\n\n\n
\"\"
\uadf8\ub9bc 1<\/figcaption><\/figure><\/div>\n\n\n\n
\"\"
\uadf8\ub9bc 2<\/figcaption><\/figure><\/div>\n\n\n\n
\uc774\uc81c \uadf8\ub9bc1\uc758 \ub370\uc774\ud130\ub85c \ub2e4\uc2dc \ub3cc\uc544\uc640\uc11c train \ub370\uc774\ud130\uc640  test \ub370\uc774\ud130\ub85c \ub098\ub220\uc90d\ub2c8\ub2e4. \ubd84\ub9ac\ub294 8:2\uc758 \ud615\ud0dc\ub85c \ub098\ub204\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state=0) # 8:2<\/pre>\n\n\n\n
\ub370\uc774\ud130\ub97c \ud45c\uc2dc\ud558\uae30 \uc704\ud574\uc11c \uc2dc\uac01\ud654 \ud568\uc218\ub97c \uc815\uc758\ud569\ub2c8\ub2e4.<\/p>\n\n\n\n
def plot_decision_function_helper(X, y, clf, show_only_decision_function = False):\n\n    colors = get_colors(y)\n    plt.axis('equal')\n    plt.tight_layout()\n\n    plt.scatter(X[:, 0], X[:, 1], c=colors, s=10, edgecolors=colors)\n    ax = plt.gca()\n    xlim = ax.get_xlim()\n    ylim = ax.get_ylim()\n\n    # Create grid to evaluate model\n    xx = np.linspace(xlim[0], xlim[1], 30)\n    yy = np.linspace(ylim[0], ylim[1], 30)\n    YY, XX = np.meshgrid(yy, xx)\n    xy = np.vstack([XX.ravel(), YY.ravel()]).T # xy.shape = (900, 2)\n    Z = clf.decision_function(xy).reshape(XX.shape)\n\n    if  show_only_decision_function:\n        ax.contour(XX, YY, Z, colors='k', levels=[0], alpha=0.5,\n                 linestyles=['-'])\n    else :\n        ax.contour(XX, YY, Z, colors='k', levels=[-1, 0, 1], alpha=0.5,\n                 linestyles=['--', '-', '--'])\n\ndef plot_decision_function(X_train, y_train, clf):\n    colors = get_colors(y_train)\n    \n    plt.figure(figsize=(8,4), dpi=150)\n    plt.title('Support Vector Machine Classifier')\n    plot_decision_function_helper(X_train, y_train, clf, False)<\/pre>\n\n\n\n
SVM\uc740 \ube44\uc120\ud615 \ub370\uc774\ud130\ub97c \uace0\ucc28\uc6d0\uc73c\ub85c \ud655\uc7a5\ud558\uc5ec \ubd84\ub9ac\ud560 \uc218 \uc788\ub3c4\ub85d rbf, linear, poly \ucee4\ub110\uc744\u001b \uc81c\uacf5\ud569\ub2c8\ub2e4. \uae30\ubcf8\uac12\uc740 rbf \ucee4\ub110\uc785\ub2c8\ub2e4.<\/p>\n\n\n\n
clf = svm.SVC() #default kernel ='rbf'    \nprint(clf)\nclf.fit(X_train, y_train)\n\nplot_decision_function(X, y, clf)\nprint(\"Accuracy: {}%\".format(clf.score(X_test, y_test) * 100 ))\n#Accuracy: 92.5%<\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
clf2 = svm.SVC(kernel='linear', C=1)    \nclf2.fit(X, y)\nplot_decision_function(X, y, clf2)\nprint(\"Accuracy: {}%\".format(clf2.score(X_test, y_test) * 100 ))\n#Accuracy: 91.5%<\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
clf4 = svm.SVC(kernel='poly', C=1)    \nclf4.fit(X, y)\nplot_decision_function(X, y, clf4)\nprint(\"Accuracy: {}%\".format(clf4.score(X_test, y_test) * 100 ))\n#Accuracy: 90.5%<\/pre>\n\n\n\n
\ucc38\uace0\ub85c \uacb0\uc815\uacbd\uacc4\ub97c \ub9cc\ub4e4\ub54c \uace0\ub824\ud574\uc57c \ud560 \ub0b4\uc6a9\uc740 \ub9c8\uc9c4\uc744 \uc5b4\ub5bb\uac8c \uacb0\uc815\ud558\ub294\uac00? \uc5d0 \ub300\ud55c \ub0b4\uc6a9\uc785\ub2c8\ub2e4. \uc774\uac83\uc740 “C”\ub97c \uc5b4\ub5bb\uac8c \uc124\uc815\ud560 \uac83\uc778\uac00\uc5d0 \ub530\ub77c \ub2ec\ub77c\uc9d1\ub2c8\ub2e4. C\ub294 \uae30\ubcf8\uac12\uc740 1\uc774\uc9c0\ub9cc \ub370\uc774\ud130\uc758 \ud2b9\uc9d5\uc5d0 \ub530\ub77c \ub354 \ub9ce\uc774 \uc124\uc815\ud560 \uc218\ub3c4 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n
\ub9cc\uc57d \uc544\uc6c3\ub77c\uc774\uc5b4\ub97c \ucd5c\ub300\ud55c \ud5c8\uc6a9\ud558\uc9c0 \uc54a\uace0 \ubaa8\ub4e0 \ub370\uc774\ud130\ub97c \ub300\uc0c1\uc73c\ub85c \uae4c\ub2e4\ub86d\uac8c(?) \ub9cc\ub4dc\ub294 \uacbd\uc6b0 \uc774\uac83\uc744 \ud558\ub4dc \ub9c8\uc9c4(hard margin)\uc774\ub77c\uace0 \ubd80\ub985\ub2c8\ub2e4. \uc774\ub807\uac8c \ub418\uba74 \uc11c\ud3ec\ud2b8 \ubca1\ud130\uc640 \uacb0\uc815 \uacbd\uacc4 \uc0ac\uc774\uc758 \uac70\ub9ac\uac00 \ub9e4\uc6b0 \uc881\ub2e4. \uc989, \ub9c8\uc9c4\uc774 \ub9e4\uc6b0 \uc791\uc544\uc9c4\ub2e4. \uc774\ub807\uac8c \uac1c\ubcc4\uc801\uc778 \ud559\uc2b5 \ub370\uc774\ud130\ub4e4\uc744 \ub2e4 \ub193\uce58\uc9c0 \uc54a\uc73c\ub824\uace0 \uc544\uc6c3\ub77c\uc774\uc5b4\ub97c \ud5c8\uc6a9\ud558\uc9c0 \uc54a\ub294 \uae30\uc900\uc73c\ub85c \uacb0\uc815 \uacbd\uacc4\ub97c \uc815\ud574\ubc84\ub9ac\uba74 \uc624\ubc84\ud53c\ud305(overfitting) \ubb38\uc81c\uac00 \ubc1c\uc0dd\ud560 \uc218 \uc788\ub2e4. <\/p>\n\n\n\n
\ubc18\ub300\ub85c \ub9c8\uc9c4(margin)\uc744 \uc5b4\ub290 \uc815\ub3c4 \uc624\ub958\uac12\ub4e4\uc774 \ud3ec\ud568\ub418\ub3c4\ub85d \ub108\uadf8\ub7fd\uac8c \uae30\uc900\uc744 \uc138\uc6b4\ub2e4\uba74 \uc774\uac83\uc744 \uc18c\ud504\ud2b8 \ub9c8\uc9c4(soft margin)\uc774\ub77c\uace0 \ubd80\ub985\ub2c8\ub2e4. \uc774\ub807\uac8c \ub108\uadf8\ub7fd\uac8c(?) \uc7a1\uc544 \ub193\uc73c\uba74 \uc11c\ud3ec\ud2b8 \ubca1\ud130\uc640 \uacb0\uc815 \uacbd\uacc4 \uc0ac\uc774\uc758 \uac70\ub9ac\uac00 \uba40\uc5b4\uc9c0\ub294 \uc989, \ub9c8\uc9c4\uc774 \ud604\uc0c1\uc774 \ub098\ud0c0\ub0a9\ub2c8\ub2e4. \uadf8\ub807\uac8c\ub418\uba74 \uc5b8\ub354\ud53c\ud305(underfitting) \ubb38\uc81c\uac00 \ubc1c\uc0dd\ud560 \uc218 \uc788\ub2e4.  \uadf8\ub807\uae30 \ub54c\ubb38\uc5d0 \ud6c8\ub828 \ub370\uc774\ud130\uc758 \uc624\ub958\ub97c \uc5b4\ub290 \uc815\ub3c4\ub294 \ud5c8\uc6a9\ud560 \uac83\uc778\uac00\uc5d0 \ub300\ud55c \uae30\uc900\uc744 \uc7a1\uace0 \ubaa8\ub378\uc744 \uc124\uacc4\ud558\ub294 \uac83\uc774 \ud544\uc694\ud569\ub2c8\ub2e4.<\/p>\n\n\n\n
\ub610 \ud558\ub098 \uace0\ub824\ud574\uc57c \ud560 \ubd80\ubd84\uc740 gamma \uac12\uc785\ub2c8\ub2e4. gamma\ub294 \ud6c8\ub828 \uc0d8\ud50c\uc774 \ubbf8\uce58\ub294 \uc601\ud5a5\uc758 \ubc94\uc704\ub97c \uc758\ubbf8\ud558\ub294 \uac83\uc73c\ub85c \uc791\uc740 \uac12\uc740 \ub113\uc740 \uc601\uc5ed\uc744 \ud070 \uac12\uc774\ub77c\uba74 \ubbf8\uce58\ub294 \ubc94\uc704\uac00 \uc81c\ud55c\uc801\uc778 \ud2b9\uc9d5\uc774 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n
Reference<\/h2>\n\n\n\n
https:\/\/www.learnopencv.com\/svm-using-scikit-learn-in-python\/<\/pre>\n\n\n\n
<\/p>\n","protected":false},"excerpt":{"rendered":"
\uc11c\ud3ec\ud2b8\ubca1\ud130\uba38\uc2e0(SVM,Support Vector Machine, \uc774\ud558 SVM)\uc740 \uac15\ub825\ud55c \uba38\uc2e0\ub7ec\ub2dd \ubaa8\ub378\uc774\uba70 \ub610 \ub2e4\uc591\ud55c \uc815\ud615 \ub370\uc774\ud130\uc14b\uc5d0\uc11c \uc88b\uc740 \uc131\ub2a5\uc744 \ubcf4\uc5ec\uc90d\ub2c8\ub2e4. SVM\uc740 \ub370\uc774\ud130\uc758 \ud2b9\uc9d5\uc774 \uba87\uac1c \ub418\uc9c0 \uc54a\ub354\ub77c\ub3c4 \ubcf5\uc7a1\ud55c \uacb0\uc815\uacbd\uacc4(Decision Boundary)\ub97c \ub9cc\ub4e4\uc5b4\uc90d\ub2c8\ub2e4. SVM\uc740 \uc800\ucc28\uc6d0, \uace0\ucc28\uc6d0\uc758 \ubaa8\ub378\uc5d0\uc11c \uc88b\uc740 \uc131\ub2a5\uc744 \ubc1c\ud718\ud558\uc9c0\ub9cc \uc0d8\ud50c\uc774 \ub9ce\uc744 \uacbd\uc6b0 \uc608\ub97c \ub4e4\uc5b4 100,000\uac1c \uc774\uc0c1\uc758 \ub370\uc774\ud130\ub97c \ucc98\ub9ac\ud558\ub294\ub370\ub294 \uc131\ub2a5\uc774 \uc88b\uc9c0 \uc54a\uc2b5\ub2c8\ub2e4. \ub610 \ub370\uc774\ud130 \uc804\ucc98\ub9ac(MinMaxScaler\uc640 \uac19\uc740 \uc791\uc5c5\uc774 \ud544\uc694)\uc640 \ub9e4\uac1c\ubcc0\uc218 \uc124\uc815\uc5d0 \uc880 \ub354 \uace0\ubbfc\ud574\uc57c … <\/p>\n
\ub354 \ubcf4\uae30 “\uc11c\ud3ec\ud2b8\ubca1\ud130\uba38\uc2e0(Support Vector Machine)”<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[17,70],"tags":[54,48,103,45,104,55],"_links":{"self":[{"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/posts\/697"}],"collection":[{"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/comments?post=697"}],"version-history":[{"count":5,"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/posts\/697\/revisions"}],"predecessor-version":[{"id":709,"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/posts\/697\/revisions\/709"}],"wp:attachment":[{"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/media?parent=697"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/categories?post=697"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/tags?post=697"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}