Difference between BN and IN
ProgrammingIt is well known that IN is simply 1-batch of BN. It is true in training phase, but not in inference.
In inference phase, BN uses popular mini-batch statistic for mean and standard variation, while IN uses them unchanged from training time, as noted in AdaIN paper below.
Reference: arxiv.org/abs/1703.06868
'Programming' 카테고리의 다른 글
| TCMalloc 사용하여 TF 메모리 누수 방지 (0) | 2022.03.10 |
|---|---|
| ML 프레임워크 GPU 연산 지원 확인하기 (0) | 2020.09.16 |
| mkvirtualenv 실행시 ailed to find interpreter for Builtin discover of python_spec 에러 (0) | 2020.08.06 |
| 공용 .gitignore 를 건드리지 않고 .gitignore 설정하기 (0) | 2020.08.03 |
| Python 업그레이드 후 virtualenv 문제 (0) | 2020.01.23 |