mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-07 12:21:27 +01:00
63219f1f9f
Summary: Fixes : https://github.com/pytorch/pytorch/issues/24892 In the paper : https://arxiv.org/pdf/1908.03265.pdf Liyuan Liu et al. suggested a new optimization algorithm with an essence of similar to Adam Algorithm. It has been discussed in the paper that, without warmup heuristic, in the early stage of adaptive optimization / learning algorithms sometimes we can get undesirable large variance which can slow overall convergence process. Authors proposed the idea of rectification of variance of adaptive learning rate when it is expected to be high. Differing from the paper, we selected variance tractability cut-off as 5 instead of 4. This adjustment is common practice, and could be found in the code-repository and also tensorflow swift optim library as well :2f03dd1970/radam/radam.py (L156)f51ee4618d/Sources/TensorFlow/Optimizers/MomentumBased.swift (L638)Pull Request resolved: https://github.com/pytorch/pytorch/pull/58968 Reviewed By: vincentqb Differential Revision: D29310601 Pulled By: iramazanli fbshipit-source-id: b7bd487f72f1074f266687fd9c0c6be264a748a9
6 lines
237 B
Python
6 lines
237 B
Python
from typing import Tuple
|
|
from .optimizer import _params_t, Optimizer
|
|
|
|
class RAdam(Optimizer):
|
|
def __init__(self, params: _params_t, lr: float=..., betas: Tuple[float, float]=..., eps: float=..., weight_decay: float=...) -> None: ...
|