File size: 7,913 Bytes
626eca0 |
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 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 |
import math
import torch
from torch.optim import Optimizer
class RAdamW(Optimizer):
r"""Implements RAdamW algorithm.
RAdam from `On the Variance of the Adaptive Learning Rate and Beyond
<https://arxiv.org/abs/1908.03265v1>`_
* `Adam: A Method for Stochastic Optimization
<https://arxiv.org/abs/1412.6980>`_
* `Decoupled Weight Decay Regularization
<https://arxiv.org/abs/1711.05101>`_
* `On the Convergence of Adam and Beyond
<https://openreview.net/forum?id=ryQu7f-RZ>`_
* `On the Variance of the Adaptive Learning Rate and Beyond
<https://arxiv.org/abs/1908.03265v1>`_
Arguments:
params (iterable): iterable of parameters to optimize or dicts defining
parameter groups
lr (float, optional): learning rate (default: 1e-3)
betas (Tuple[float, float], optional): coefficients used for computing
running averages of gradient and its square (default: (0.9, 0.999))
eps (float, optional): term added to the denominator to improve
numerical stability (default: 1e-8)
weight_decay (float, optional): weight decay coefficient (default: 1e-2)
"""
def __init__(
self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8, weight_decay=1e-2
):
if not 0.0 <= lr:
raise ValueError("Invalid learning rate: {}".format(lr))
if not 0.0 <= eps:
raise ValueError("Invalid epsilon value: {}".format(eps))
if not 0.0 <= betas[0] < 1.0:
raise ValueError("Invalid beta parameter at index 0: {}".format(betas[0]))
if not 0.0 <= betas[1] < 1.0:
raise ValueError("Invalid beta parameter at index 1: {}".format(betas[1]))
defaults = dict(lr=lr, betas=betas, eps=eps, weight_decay=weight_decay)
super(RAdamW, self).__init__(params, defaults)
def step(self, closure=None):
"""Performs a single optimization step.
Arguments:
closure (callable, optional): A closure that reevaluates the model
and returns the loss.
"""
loss = None
if closure is not None:
loss = closure()
for group in self.param_groups:
for p in group["params"]:
if p.grad is None:
continue
# Perform optimization step
grad = p.grad.data
if grad.is_sparse:
raise RuntimeError(
"Adam does not support sparse gradients, please consider SparseAdam instead"
)
state = self.state[p]
# State initialization
if len(state) == 0:
state["step"] = 0
# Exponential moving average of gradient values
state["exp_avg"] = torch.zeros_like(p.data)
# Exponential moving average of squared gradient values
state["exp_avg_sq"] = torch.zeros_like(p.data)
exp_avg, exp_avg_sq = state["exp_avg"], state["exp_avg_sq"]
beta1, beta2 = group["betas"]
eps = group["eps"]
lr = group["lr"]
if "rho_inf" not in group:
group["rho_inf"] = 2 / (1 - beta2) - 1
rho_inf = group["rho_inf"]
state["step"] += 1
t = state["step"]
# Decay the first and second moment running average coefficient
exp_avg.mul_(beta1).add_(grad, alpha=1 - beta1)
exp_avg_sq.mul_(beta2).addcmul_(grad, grad, value=1 - beta2)
rho_t = rho_inf - ((2 * t * (beta2**t)) / (1 - beta2**t))
# Perform stepweight decay
p.data.mul_(1 - lr * group["weight_decay"])
if rho_t >= 5:
var = exp_avg_sq.sqrt().add_(eps)
r = math.sqrt(
(1 - beta2**t)
* ((rho_t - 4) * (rho_t - 2) * rho_inf)
/ ((rho_inf - 4) * (rho_inf - 2) * rho_t)
)
p.data.addcdiv_(exp_avg, var, value=-lr * r / (1 - beta1**t))
else:
p.data.add_(exp_avg, alpha=-lr / (1 - beta1**t))
return loss
# from typing import List
# import collections
# import torch
# import transformers
# from classy.optim.factories import Factory
# from transformers import AdamW
# class ElectraOptimizer(Factory):
# def __init__(
# self,
# lr: float,
# warmup_steps: int,
# total_steps: int,
# weight_decay: float,
# lr_decay: float,
# no_decay_params: List[str],
# ):
# self.lr = lr
# self.warmup_steps = warmup_steps
# self.total_steps = total_steps
# self.weight_decay = weight_decay
# self.lr_decay = lr_decay
# self.no_decay_params = no_decay_params
# def group_layers(self, module) -> dict:
# grouped_layers = collections.defaultdict(list)
# module_named_parameters = list(module.named_parameters())
# for ln, lp in module_named_parameters:
# if "embeddings" in ln:
# grouped_layers["embeddings"].append((ln, lp))
# elif "encoder.layer" in ln:
# layer_num = ln.replace("transformer_model.encoder.layer.", "")
# layer_num = layer_num[0 : layer_num.index(".")]
# grouped_layers[layer_num].append((ln, lp))
# else:
# grouped_layers["head"].append((ln, lp))
# depth = len(grouped_layers) - 1
# final_dict = dict()
# for key, value in grouped_layers.items():
# if key == "head":
# final_dict[0] = value
# elif key == "embeddings":
# final_dict[depth] = value
# else:
# # -1 because layer number starts from zero
# final_dict[depth - int(key) - 1] = value
# assert len(module_named_parameters) == sum(
# len(v) for _, v in final_dict.items()
# )
# return final_dict
# def group_params(self, module) -> list:
# optimizer_grouped_params = []
# for inverse_depth, layer in self.group_layers(module).items():
# layer_lr = self.lr * (self.lr_decay**inverse_depth)
# layer_wd_params = {
# "params": [
# lp
# for ln, lp in layer
# if not any(nd in ln for nd in self.no_decay_params)
# ],
# "weight_decay": self.weight_decay,
# "lr": layer_lr,
# }
# layer_no_wd_params = {
# "params": [
# lp
# for ln, lp in layer
# if any(nd in ln for nd in self.no_decay_params)
# ],
# "weight_decay": 0,
# "lr": layer_lr,
# }
# if len(layer_wd_params) != 0:
# optimizer_grouped_params.append(layer_wd_params)
# if len(layer_no_wd_params) != 0:
# optimizer_grouped_params.append(layer_no_wd_params)
# return optimizer_grouped_params
# def __call__(self, module: torch.nn.Module):
# optimizer_grouped_parameters = self.group_params(module)
# optimizer = AdamW(optimizer_grouped_parameters, lr=self.lr)
# scheduler = transformers.get_linear_schedule_with_warmup(
# optimizer, self.warmup_steps, self.total_steps
# )
# return {
# "optimizer": optimizer,
# "lr_scheduler": {
# "scheduler": scheduler,
# "interval": "step",
# "frequency": 1,
# },
# }
|