
과제 Assignment #2 - 1. word2vec의 핵심적인 통찰은 '단어가 보유하고있는 그룹에 의해 알려진다'는 것입니다. 구체적으로 '중심'단어 C와 C를 둘러싼 상황 별 창이 있다고 가정합니다. 이 상황 별 창에있는 단어를 '외부 단어'라고합니다. 예를 들어, 그림 1에서 중심 단어 c가 'banking'임을 알 수 있습니다. 컨텍스트 창 크기가 2이므로 외부 단어는`turning ',`into',`crises '및`as'입니다. context window란 근방의 단어를 몇 길이나 볼것인지를 뜻하는 크기임 -- skip-gram word2vec 알고리즘의 목표는 확률 분포 P (O|C)를 정확하게 얻는 것입니다. 특정 단어 o와 특정 단어 c가 주어지면, 우리는 단어 o가 c에 대한 '외부'..
AI/자연어 처리
2021. 7. 1. 05:31
공지사항
최근에 올라온 글
최근에 달린 댓글
- Total
- Today
- Yesterday
링크
TAG
- /var/log/fontconfig.log
- /var/log/tallylog
- 첫논문
- Deformable Convolutional Networks
- 샴네트워크
- textual guidance
- Deformable Part Model
- torchmodel
- 도메인 어댑테이션
- alternatives.log
- Group Acitivty Recognition
- tensorflow convert
- yolov3 #tensorflow_convert
- max-margin_loss
- 딥페이크탐지
- neural network applications
- /var/log/alternatives.log
- 1-stage detection
- TransferLearning
- StyleCLIP
- Arbitrary image stylization
- /var/log/faillog
- tf version porting
- batch란
- Chatgpt4
- Source-free unsupervised domain adaptation
- ctfloader
- DetectoRS
- tf2to1
- 전이학습
일 | 월 | 화 | 수 | 목 | 금 | 토 |
---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | ||
6 | 7 | 8 | 9 | 10 | 11 | 12 |
13 | 14 | 15 | 16 | 17 | 18 | 19 |
20 | 21 | 22 | 23 | 24 | 25 | 26 |
27 | 28 | 29 | 30 | 31 |
글 보관함