DACON
search Subscribe
noti 0

    MY XP

    XP Info

    LEVEL 1

    Until Next Level 150 XP

    0/150 XP

    Today's XP

    0 / 800

    Total XP

    0

    Earned XP

    0

    Spent XP

    0

    Sign in with your SNS accounts

    google
    Sign in with Google
    Don't have an account? Sign Up
    Verify your email

    To sign up, you must verify your email. Do you want to resend the code?
    [Dacon] sign up verification

    Your email must be verified to complete the sign up process. Please verify your email below to complete.

    Terms of Use

    Privacy policy

    Consent to receive marketing information

    1. Promotional Information Usage


    a. DACON provides promotional information such as user-tailored services and product recommendations, various prize events, promotions,

    and competition announcements to users through email, postal mail, text messages (SMS or KakaoTalk Alert), push notifications, or phone calls


    b. Users may refuse marketing communications and can withdraw consent at any time.

    Refusing consent will not restrict access to DACON's core services.

    However, marketing information services such as discounts, event notifications, and personalized recommendations will be limited.


    2. Disadvantages of Non-Consent

    a. Under Article 22(5) of the Personal Information Protection Act, refusal of optional information consent does not affect service availability.

    b. However, marketing information services including discounts, events, and personalized recommendations will be limited


    3. Withdrawing Service Communication Consent

    a. To opt out of DACON's marketing communications, go to 'Home > Account Management Page > Marketing (Competitions, Education, etc.) Information Reception Consent (Optional)' at the bottom of the page

    b. Consent can be reinstated anytime through the same path ('Home > Account Management Page > Marketing (Competitions, Education, etc.) Information Reception Consent (Optional)’) for future marketing benefits.


    2021.05.25

    2022 AI 대학원 챌린지

    LG | 채용 | B-cell | Epitope | Classification | macro-f1

    • moneyIcon Prize : Total 2,500만원
    • 2022.07.04 ~ 2022.07.29 16:59 + Google Calendar
    • 324 Users Completed
    Practice
    Overview Data Code (share) Talk Leaderboard

    Setup D-0

     

    0.9에 가까운 macro-f1 score가 나올 수 있었던 이유

    안녕해요
    공동작성자
    2022.07.26 15:51 2,334 Views language
    comments 6

    현재 토크 게시판에 새로운 규정이 추가되었는데 관련해서 궁금해 하실 분들이 많을 것 같아 내용을 정리해봤습니다. 정리된 내용처럼 epitope과 antigen에 대한 정보를 사용하지 않고도 좋은 성능이 나오게 됩니다. 관련해 validation 결과와 리더보드 성능도 비슷한 결과를 보여주었습니다.

    Code
    6 Comments
    Login Required
    0 / 1000
    Yun_Geon_Il
    2022.07.26 16:39

    Deleted Comment

    Yun_Geon_Il
    2022.07.26 16:40

    공유 감사드립니다 !

    ZINZINBIN
    2022.07.26 23:53

    우와..여태껏 epitope이랑 antigen을 가지고 bert랑 transformer로 이것저것 하고 있었는데.... 공유 감사드립니다 ㅜㅜ

    안녕해요
    2022.07.27 00:08

    저희도 처음에 이런저런 시도들을 하다가 알게 되었습니다. 기존방식대로 bert, transformer활용하는게 좋을 것 같아요!

    콩뮤
    2022.07.28 10:03

    number of test, number of response 사용 불가능한거 아닌가용..?

    안녕해요
    2022.07.28 16:07

    네 맞습니다. 사용이 불가능하고, number of test, number of response와 관련된 데이터를 사용하면 어떤 결과가 나오는지 왜 사용하면 안되는 것인지 궁금증을 가지고 있을 분들이 있다고 생각되어서 코드 공유를 진행했습니다.

    📣 The comment input field has been moved to the top of the comment list!
    to_list List
    Previous
    Amino acid feature로 계산된 Protein 단위의 feature를 추출하는 방법
    Competition - 2022 AI 대학원 챌린지
    Likes 12
    Views 2,838
    Comments 3
    3yr ago
    Current
    0.9에 가까운 macro-f1 score가 나올 수 있었던 이유
    Competition - 2022 AI 대학원 챌린지
    Likes 13
    Views 2,334
    Comments 5
    3yr ago
    Next
    [Baseline] Embedding Layer + LSTM
    Competition - 2022 AI 대학원 챌린지
    Likes 8
    Views 3,393
    Comments 2
    3yr ago