<![CDATA[
<p><a href="https://www.notion.so/Logistic-Regression-35f3d4314e0080d3b438ee0125a4b5a1?source=copy_link" target="_blank" class="ke-link">https://www.notion.so/Logistic-Regression-35f3d4314e0080d3b438ee0125a4b5a1?source=copy_link</a></p><div class="figure-open" contenteditable="false" data-ke-type="opengraph" data-ke-align="alignCenter" data-og-type="website" data-og-title="로지스틱 회귀 (Logistic Regression) (수정중) | Notion" data-og-description="로지스틱 회귀 (Logistic Regression)" data-og-host="www.notion.so" data-og-source-url="https://www.notion.so/Logistic-Regression-35f3d4314e0080d3b438ee0125a4b5a1?source=copy_link" data-og-url="https://www.notion.so/Logistic-Regression-35f3d4314e0080d3b438ee0125a4b5a1" data-og-image="https://scrap.kakaocdn.net/dn/Rgxmk/dJMb9kmikYx/EZtEQ09uPmUccfkJISe1G0/img.png?width=1200&height=630&face=0_0_1200_630,https://scrap.kakaocdn.net/dn/hyFTa/dJMb9g5gLx4/4CLP134YRBPfnyduDz6Vxk/img.png?width=1200&height=630&face=0_0_1200_630"><a href="https://www.notion.so/Logistic-Regression-35f3d4314e0080d3b438ee0125a4b5a1?source=copy_link" target="_blank" data-source-url="https://www.notion.so/Logistic-Regression-35f3d4314e0080d3b438ee0125a4b5a1?source=copy_link"><div class="og-image"><img class="thumb_img" src="https://scrap.kakaocdn.net/dn/Rgxmk/dJMb9kmikYx/EZtEQ09uPmUccfkJISe1G0/img.png?width=1200&height=630&face=0_0_1200_630,https://scrap.kakaocdn.net/dn/hyFTa/dJMb9g5gLx4/4CLP134YRBPfnyduDz6Vxk/img.png?width=1200&height=630&face=0_0_1200_630" alt="" xxxxonerror="this.src="//img1.kakaocdn.net/thumb/C200x200/?fname=https%3A%2F%2Ft1.daumcdn.net%2Fcafe_image%2Fcafe_meta_image_190529.png""></div><div class="og-text"><p class="og-title">로지스틱 회귀 (Logistic Regression) (수정중) | Notion</p><p class="og-desc">로지스틱 회귀 (Logistic Regression)</p><p class="og-host">www.notion.so</p></div></a></div><p> </p>
<!-- -->
]]>
카페 게시글
과제게시판
[딥러닝] 로지스틱 회귀 (Logistic Regression)
조한빛
추천 0
조회 40
26.05.20 16:54
댓글 2
다음검색
첫댓글 인공지능 교과서 같은 보고서입니다. 다들 꼭 읽어보세요
1. 로지스틱회귀에서 손실함수로 바이너리 크로스엔트로피(BCE)함수 대신에 MSE를 사용하면 어떻게 될까요 학습이 잘 안되나요?
error = 레이블(정수, 0,1) - 출력(실수, 0~1) -> 0<=error<=1
error =-log출력(실수) -> 0<=error<=inf
2.BCE에서 정답이 1일때 오차만 계산하면 되지 않나요 왜 1일때 오차+0일때오차로 계산하나요