2023年12月4日
[pytorch] 余弦退火+warmup实现调研
摘要： tl;dr: pytorch的 torch.optim.lr_scheduler.OneCycleLR 就很不错，能兼顾warmup和余弦学习率，也不用下载额外的包 import torch from torch.optim.lr_scheduler import CosineAnnealingLR 阅读全文
posted @ 2023-12-04 15:57 NoNoe 阅读(1532) 评论(0) 推荐(0)
[论文速览] Randomized Quantization@ A Generic Augmentation for Data Agnostic Self-supervised Learning
摘要： Pre title: Randomized Quantization: A Generic Augmentation for Data Agnostic Self-supervised Learning accepted: ICCV 2023 paper: https://arxiv.org/abs 阅读全文
posted @ 2023-12-04 15:55 NoNoe 阅读(484) 评论(0) 推荐(0)