![](http://i1.daumcdn.net/thumb/C148x148.fwebp.q85/?fname=https://blog.kakaocdn.net/dn/Eebq9/btsIJXsFQw6/IZc4BGzs6xEN2SXSZMnwLk/img.png)
Why does the probability is constrained under the 1/(n+1) factor, when n is the size of the calibration dataset?이 그림의 밑이 설명함. Calibration dataset에서 quantile 포인트와 그 직전 point (그게 뭘지는 모르고)의 사이에 있을 확률. calibration data point가 어떤 order로 있을 지 모르니까 각 point 사이에 새로운 test point의 score가 있을 확률이 1 / (n+1)라고 함. https://www.youtube.com/watch?v=TRx4a2u-j7M&t=12shttps://people.eecs.berkeley.edu/~angelopoulos/pu..
Research (연구 관련)
2024. 7. 23. 16:22
공지사항
최근에 올라온 글
최근에 달린 댓글
- Total
- Today
- Yesterday
링크
TAG
- nerf
- nohup
- 인터뷰
- densepose
- 문경식
- 2d pose
- Interview
- Transformation
- Generative model
- 컴퓨터비전
- 컴퓨터비젼
- 에디톨로지
- VAE
- pyrender
- 헬스
- demo
- focal length
- deep learning
- pytorch
- Virtual Camera
- camera coordinate
- spin
- 비전
- Docker
- 머신러닝
- Machine Learning
- world coordinate
- Pose2Mesh
- part segmentation
- 피트니스
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