{"id":690,"date":"2020-12-16T21:44:33","date_gmt":"2020-12-16T12:44:33","guid":{"rendered":"http:\/\/cedartrees.co.kr\/?p=690"},"modified":"2021-04-03T19:10:29","modified_gmt":"2021-04-03T10:10:29","slug":"logistic","status":"publish","type":"post","link":"http:\/\/blog.cedartrees.co.kr\/index.php\/2020\/12\/16\/logistic\/","title":{"rendered":"\ub85c\uc9c0\uc2a4\ud2f1 \ud68c\uadc0(Logistic Regression)"},"content":{"rendered":"\n
import numpy as np\nimport pandas as pd\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.svm import LinearSVC\nimport matplotlib.pyplot as plt<\/pre>\n\n\n\n
\uc544\ub798\uc640 \uac19\uc740 \ub370\uc774\ud130\uac00 \uc788\ub2e4\uace0 \uc0dd\uac01\ud574\ubcf4\uaca0\uc2b5\ub2c8\ub2e4. \ub3c5\ub9bd\ubcc0\uc218 x\ub294 0-9\uae4c\uc9c0\uc758 \uac12\uc774 \uc785\ub825\ub418\uc5b4 \uc788\uace0 \uc885\uc18d\ubcc0\uc218 y\ub294 \uc5f0\uc18d\ub41c \uc218\uce58 \ud615\ud0dc\uc758 \uac12\uc774 \uc544\ub2cc 0,1\uc758 \uc774\ud56d\ubcc0\uc218\ub85c \ub418\uc5b4 \uc788\uc2b5\ub2c8\ub2e4. \ub370\uc774\ud130\ub97c \ud1b5\ud574 \ubcf4\ub2c8 0-4\uae4c\uc9c0\ub294 0, 5-9\uae4c\uc9c0\ub294 1\uc744 \ub9ac\ud134\ud558\ub294 \ud615\ud0dc\uc785\ub2c8\ub2e4. \uc608\ub97c \ub4e4\uc5b4\uc11c \uccb4\ub825 \ud14c\uc2a4\ud2b8\ub97c \ud558\ub294\ub370 \ud131\uac78\uc774 0-4\uae4c\uc9c0\ub294 0(\ubd88\ud569\uaca9), 5-9\uae4c\uc9c0\ub294 \ud569\uaca9(1) \uc774\ub77c\uace0 \uac00\uc815\ud574\ubcf4\uaca0\uc2b5\ub2c8\ub2e4. <\/p>\n\n\n\n
\uc774\ub7ec\ud55c \ubb38\uc81c\ub97c \uc120\ud615\ud68c\uadc0\uc758 \ud615\ud0dc\ub85c \ubd84\uc11d\ud574\ubcf8\ub2e4\uba74 \uc5b4\ub5bb\uac8c\ub420\uae4c\uc694? \ubd84\uc11d\uc774 \uac00\ub2a5\ud560\uae4c\uc694? 
\uac00\ub2a5\ud574\ubcf4\uae30\ub3c4 \ud569\ub2c8\ub2e4. \uba3c\uc800 \ub370\uc774\ud130\ub97c \uc2dc\uac01\ud654 \ud574\ubcf4\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n
x = [[1],[2],[3],[4],[0],[5],[6],[7],[8],[9]]\ny = [0,0,0,0,0,1,1,1,1,1]<\/pre>\n\n\n\n
line = np.linspace(0,1,num=10)\nyline = np.linspace(0.5,0.5,num=10)\n\nplt.scatter(x,y)\nplt.plot(line)\nplt.plot(yline)<\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\ub370\uc774\ud130\ub97c \uc2dc\uac01\ud654 \ud574\ubcf4\uba74 \uc704\uc640 \uac19\uc740 \ud615\ud0dc\uc758 \uadf8\ub798\ud504\uac00 \ud45c\uc2dc\ub429\ub2c8\ub2e4. \uadf8\ub9ac\uace0 \uc774\ub7ec\ud55c \ub370\uc774\ud130\ub77c\uba74 \uc120\ud615\ud68c\uadc0\ub85c\ub3c4 \uac00\ub2a5\ud560\ub4ef \ubcf4\uc785\ub2c8\ub2e4. 0-4, 5-9\uc758 \ub370\uc774\ud130\ub97c \uc911\uac04\uc5d0 \uc8fc\ud669\uc0c9 \uc120\uc744 0.5\ub97c \uae30\uc900\uc73c\ub85c \uc785\ub825\ub418\ub294 \ub370\uc774\ud130\uac00 0.5 \ubcf4\ub2e4 \ud06c\uba74 1\uc744 \ucd9c\ub825\ud558\uace0 \uc791\uc73c\uba74 0\uc744 \ucd9c\ub825\ud558\uac8c \ud558\uba74 \ub420\uac83\uac19\uc2b5\ub2c8\ub2e4.
\uc774\ub7ec\ud55c \uae30\ubcf8 \ucee8\uc149\uc744 \uac00\uc9c0\uace0 \uc120\ud615\ud68c\uadc0\ub97c \ud1b5\ud574\uc11c \ud574\ub2f9 \ub0b4\uc6a9\uc744 \uad6c\ud604\ud574\ubcf4\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n\n\n\n
from sklearn.linear_model import LinearRegression\nlinear = LinearRegression().fit(x,y)\npredict = linear.predict([[50]])\nprint(predict)\n# 7.39393939<\/pre>\n\n\n\n
\uc704\uc640 \uac19\uc740 \ubaa8\ub378\uc744 \uad6c\ud604\ud558\uace0 \ud14c\uc2a4\ud2b8 \ud574\ubcf4\ub2c8 0-9\uae4c\uc9c0\uc758 \ud131\uac78\uc774 \uac2f\uc218\ub85c\ub294 \uc608\uce21\uc774 \ube44\uad50\uc801 \uc798\ub418\ub294\ub4ef \ubcf4\uc785\ub2c8\ub2e4. \ud558\uc9c0\ub9cc \uc5b4\ub5a4 \uc0ac\ub78c\uc774 \uc640\uc11c \ud131\uac78\uc774\ub97c 50\ud68c \ud639\uc740 \uadf8 \uc774\uc0c1\uc744 \ud558\uac8c\ub418\uba74 \uc608\uce21\uce58\uac00 \ud569\uaca9(1)\/\ubd88\ud569\uaca9(0)\uc744 \ubc97\uc5b4\uc11c\ub098 7\uc774\ub77c\ub294 \uc22b\uc790\uac00 \ub098\uc624\uac8c\ub429\ub2c8\ub2e4. \ub354 \ud070 \uc22b\uc790\ub97c \uc785\ub825\ud558\uba74 \uc608\uce21\uce58\uac00 \ub354 \ub192\uac8c \ub098\uc624\uaca0\uc8e0. <\/p>\n\n\n\n
\uc774\ub807\uac8c\ub418\uba74 \uc774 \ubaa8\ub378\uc740 \uc0ac\uc6a9\ud560 \uc218\uac00 \uc5c6\uc2b5\ub2c8\ub2e4. 
\ub9cc\uc57d \uc774\ub7f4 \ub54c\uc5d0 \ud131\uac78\uc774\ub97c \uc544\ubb34\ub9ac \ub9ce\uc774 \ud574\ub3c4 \ud569\uaca9\uacfc \ubd88\ud569\uaca9\uc73c\ub85c\ub9cc \uacb0\uacfc\uce58\ub97c \ud45c\uc2dc\ud558\uace0 \uc2f6\ub2e4\uba74 \uc5b4\ub5bb\uac8c \ud558\ub294\uac8c \uc88b\uc744\uae4c\uc694?<\/p>\n\n\n\n
\uc774\ub7f4 \ub54c\uc5d0 \uc0ac\uc6a9\ud558\ub294 \uac83\uc774 \ub85c\uc9c0\uc2a4\ud2f1 \ud68c\uadc0\uc785\ub2c8\ub2e4. \uac19\uc740 \ub370\uc774\ud130\ub85c sklearn\uc758 \ub85c\uc9c0\uc2a4\ud2f1\ud68c\uadc0(LogisticRegression)  \ud328\ud0a4\uc9c0\ub97c \uc0ac\uc6a9\ud574\uc11c \ubaa8\ub378\uc744 \ub9cc\ub4e4\uc5b4\ubcf4\uba74 50\uc774\ub77c\ub294 \ud070 \uc22b\uc790\ub97c \uc785\ub825\ud574\ub3c4 \ud569\uaca9\uc744 \uc758\ubbf8\ud558\ub294 1\uc774\ub77c\ub294 \uac12\uc744 \ubcf4\uc5ec\uc8fc\uac8c\ub429\ub2c8\ub2e4.<\/p>\n\n\n\n
from sklearn.linear_model import LogisticRegression\nlogistic = LogisticRegression().fit(x,y)\npredict = logistic.predict([[50]])\nprint(predict)\n# 1<\/pre>\n\n\n\n
\uc774\ub7ec\ud55c \uac83\uc744 \uac00\ub2a5\ud558\uac8c \ud574\uc8fc\ub294 \uac83\uc740 Hypothesis\uac00 \uae30\uc874\uc758 \uc120\ud615\ud68c\uadc0 \ubaa8\ub378\uacfc \ub2e4\ub978 \ud2b9\ubcc4\ud55c \ud615\ud0dc\ub97c \uac00\uc9c0\uae30 \ub54c\ubb38\uc785\ub2c8\ub2e4. \uc774\uac83\uc744 Sigmoid\ub77c\uace0 \ubd80\ub985\ub2c8\ub2e4.<\/p>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\uc774 \ud568\uc218\uc5d0\uc11c \ubcf4\uba74 z\uc758 \uac12\uc740 \uc120\ud615\ud68c\uadc0\uc758 Wx+b\uc758 \uac12, \uc989 \uc120\ud615\ud68c\uadc0\uc5d0\uc11c\uc758 \ucd9c\ub825\uac12\uc744 \uc0ac\uc6a9\ud569\ub2c8\ub2e4. \uc218\uc2dd\uc5d0 \ub530\ub77c\uc11c \ucd9c\ub825\uc740 0-1 \uc0ac\uc774\uc758 \uc2e4\uc218 \uac12\uc744 \ub9ac\ud134\ud558\uac8c \ub418\uace0 0.5\ub97c \uae30\uc900\uc73c\ub85c \uc791\uc73c\uba74 0\uc744 \ud06c\uba74 1\uc744 \ucd9c\ub825\ud574\uc904 \uc218 \uc788\uac8c\ub429\ub2c8\ub2e4.<\/p>\n\n\n\n
\uadf8\ub807\ub2e4\uba74 \ucd5c\uc801\ud654\ub294 \uc5b4\ub5bb\uac8c \ud558\uac8c\ub420\uae4c? \uc120\ud615\ud68c\uadc0\uc5d0\uc11c\ub294 MSE(Mean Square Error)\ub97c \uc0ac\uc6a9\ud558\uc9c0\ub9cc \ub85c\uc9c0\uc2a4\ud2f1\uc5d0\uc11c\ub294 \uc544\ub798\uc640 \uac19\uc740 Cross-Entropy\ub97c \uc0ac\uc6a9\ud569\ub2c8\ub2e4.<\/p>\n\n\n\n
\"\"\/
https:\/\/towardsdatascience.com\/cross-entropy-for-classification-d98e7f974451<\/figcaption><\/figure>\n\n\n\n
\uc704\uc758 \uc2dd\uc740 \ub450\uac1c\uc758 \uc2dd\uc744 \ud558\ub098\ub85c \ud569\uce5c \uac83\uc785\ub2c8\ub2e4.<\/p>\n","protected":false},"excerpt":{"rendered":"
\uc544\ub798\uc640 \uac19\uc740 \ub370\uc774\ud130\uac00 \uc788\ub2e4\uace0 \uc0dd\uac01\ud574\ubcf4\uaca0\uc2b5\ub2c8\ub2e4. \ub3c5\ub9bd\ubcc0\uc218 x\ub294 0-9\uae4c\uc9c0\uc758 \uac12\uc774 \uc785\ub825\ub418\uc5b4 \uc788\uace0 \uc885\uc18d\ubcc0\uc218 y\ub294 \uc5f0\uc18d\ub41c \uc218\uce58 \ud615\ud0dc\uc758 \uac12\uc774 \uc544\ub2cc 0,1\uc758 \uc774\ud56d\ubcc0\uc218\ub85c \ub418\uc5b4 \uc788\uc2b5\ub2c8\ub2e4. \ub370\uc774\ud130\ub97c \ud1b5\ud574 \ubcf4\ub2c8 0-4\uae4c\uc9c0\ub294 0, 5-9\uae4c\uc9c0\ub294 1\uc744 \ub9ac\ud134\ud558\ub294 \ud615\ud0dc\uc785\ub2c8\ub2e4. \uc608\ub97c \ub4e4\uc5b4\uc11c \uccb4\ub825 \ud14c\uc2a4\ud2b8\ub97c \ud558\ub294\ub370 \ud131\uac78\uc774 0-4\uae4c\uc9c0\ub294 0(\ubd88\ud569\uaca9), 5-9\uae4c\uc9c0\ub294 \ud569\uaca9(1) \uc774\ub77c\uace0 \uac00\uc815\ud574\ubcf4\uaca0\uc2b5\ub2c8\ub2e4. \uc774\ub7ec\ud55c \ubb38\uc81c\ub97c \uc120\ud615\ud68c\uadc0\uc758 \ud615\ud0dc\ub85c \ubd84\uc11d\ud574\ubcf8\ub2e4\uba74 \uc5b4\ub5bb\uac8c\ub420\uae4c\uc694? \ubd84\uc11d\uc774 \uac00\ub2a5\ud560\uae4c\uc694? \uac00\ub2a5\ud574\ubcf4\uae30\ub3c4 \ud569\ub2c8\ub2e4. \uba3c\uc800 \ub370\uc774\ud130\ub97c … <\/p>\n
\ub354 \ubcf4\uae30 “\ub85c\uc9c0\uc2a4\ud2f1 \ud68c\uadc0(Logistic Regression)”<\/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":[105,54,48,45,55],"_links":{"self":[{"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/posts\/690"}],"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=690"}],"version-history":[{"count":4,"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/posts\/690\/revisions"}],"predecessor-version":[{"id":714,"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/posts\/690\/revisions\/714"}],"wp:attachment":[{"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/media?parent=690"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/categories?post=690"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/blog.cedartrees.co.kr\/index.php\/wp-json\/wp\/v2\/tags?post=690"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}