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
|
import torch
from .optimizer import Optimizer
class Rprop(Optimizer):
"""Implements the resilient backpropagation algorithm.
Arguments:
params (iterable): iterable of parameters to optimize or dicts defining
parameter groups
lr (float, optional): learning rate (default: 1e-2)
etas (Tuple[float, float], optional): pair of (etaminus, etaplis), that
are multiplicative increase and decrease factors
(default: (0.5, 1.2))
step_sizes (Tuple[float, float], optional): a pair of minimal and
maximal allowed step sizes (default: (1e-6, 50))
"""
def __init__(self, params, lr=1e-2, etas=(0.5, 1.2), step_sizes=(1e-6, 50)):
if not 0.0 <= lr:
raise ValueError("Invalid learning rate: {}".format(lr))
if not 0.0 < etas[0] < 1.0 < etas[1]:
raise ValueError("Invalid eta values: {}, {}".format(etas[0], etas[1]))
defaults = dict(lr=lr, etas=etas, step_sizes=step_sizes)
super(Rprop, self).__init__(params, defaults)
@torch.no_grad()
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:
with torch.enable_grad():
loss = closure()
for group in self.param_groups:
for p in group['params']:
if p.grad is None:
continue
grad = p.grad
if grad.is_sparse:
raise RuntimeError('Rprop does not support sparse gradients')
state = self.state[p]
# State initialization
if len(state) == 0:
state['step'] = 0
state['prev'] = torch.zeros_like(p, memory_format=torch.preserve_format)
state['step_size'] = grad.new().resize_as_(grad).fill_(group['lr'])
etaminus, etaplus = group['etas']
step_size_min, step_size_max = group['step_sizes']
step_size = state['step_size']
state['step'] += 1
sign = grad.mul(state['prev']).sign()
sign[sign.gt(0)] = etaplus
sign[sign.lt(0)] = etaminus
sign[sign.eq(0)] = 1
# update stepsizes with step size updates
step_size.mul_(sign).clamp_(step_size_min, step_size_max)
# for dir<0, dfdx=0
# for dir>=0 dfdx=dfdx
grad = grad.clone(memory_format=torch.preserve_format)
grad[sign.eq(etaminus)] = 0
# update parameters
p.addcmul_(grad.sign(), step_size, value=-1)
state['prev'].copy_(grad)
return loss
|
