Notice
Recent Posts
Recent Comments
Link
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 |
Tags
- irregularly sampled time series
- nccl 설치
- 패혈증 관련 급성 호흡곤란 증후군
- machine learning
- netflix thumbnail
- causal machine learning
- GaN
- doubleml
- 의료
- multi gpu
- 인과추론 의료
- causal reasoning
- 리뷰
- 의료정보
- 분산 학습
- nccl 업데이트
- 불규칙적 샘플링
- 의료 의사결정
- 토픽모델링
- ERD
- 딥러닝
- first pg on this rank that detected no heartbeat of its watchdog.
- Transformer
- pytorch
- gru-d
- causal ml
- nccl 업그레이드
- NTMs
- causal inference
- Time Series
Archives
- Today
- Total
목록경사하강법 (1)
데알못정을

파이토치의 구성 torch numpy와 같은 구성 torch.autograd 자동 미분을 위한 함수 포함 torch.nn 신경망을 구축하기 위한 다양한 데이터 구조나 레이어등이 정의돼 있다.(활성함수, MSEloss 등) torch.optim SGD(확률적 경사하강법) 을 중심으로 한 파라미터 최적화 알고리즘이 구현돼 있다 torch.utils.data SGD의 반복 연산을 실핼할 때 사용하는 미니 배치용 유틸리티 함수가 포함돼 있다. 1. torch.nn , optim 모듈을 사용하지 않고 수동으로 SGD를 구현하기 import torch y = 1 + 2x_1 + 3x_2 #우리가 만들 함수 w_true = torch.Tensor([1,2,3]) # 정답 파라미터 X = torch.cat([torch..
Coding
2022. 9. 30. 17:15