-
-
Notifications
You must be signed in to change notification settings - Fork 78
/
rsgd.py
200 lines (185 loc) · 7.35 KB
/
rsgd.py
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
import torch.optim.optimizer
from ..manifolds import Euclidean
from ..tensor import ManifoldParameter, ManifoldTensor
from .mixin import OptimMixin
from .tracing import create_traced_update
__all__ = ["RiemannianSGD"]
class RiemannianSGD(OptimMixin, torch.optim.Optimizer):
r"""Riemannian Stochastic Gradient Descent with the same API as :class:`torch.optim.SGD`
Parameters
----------
params : iterable
iterable of parameters to optimize or dicts defining
parameter groups
lr : float
learning rate
momentum : float (optional)
momentum factor (default: 0)
weight_decay : float (optional)
weight decay (L2 penalty) (default: 0)
dampening : float (optional)
dampening for momentum (default: 0)
nesterov : bool (optional)
enables Nesterov momentum (default: False)
Other Parameters
----------------
stabilize : int
Stabilize parameters if they are off-manifold due to numerical
reasons every ``stabilize`` steps (default: ``None`` -- no stabilize)
"""
def __init__(
self,
params,
lr,
momentum=0,
dampening=0,
weight_decay=0,
nesterov=False,
use_momentum=None,
stabilize=None,
):
if lr < 0.0:
raise ValueError("Invalid learning rate: {}".format(lr))
if momentum < 0.0:
raise ValueError("Invalid momentum value: {}".format(momentum))
if weight_decay < 0.0:
raise ValueError("Invalid weight_decay value: {}".format(weight_decay))
defaults = dict(
lr=lr,
momentum=momentum,
dampening=dampening,
weight_decay=weight_decay,
nesterov=nesterov,
use_momentum=use_momentum or bool(momentum),
)
if nesterov and (momentum <= 0 or dampening != 0):
raise ValueError("Nesterov momentum requires a momentum and zero dampening")
super().__init__(params, defaults, stabilize=stabilize)
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()
with torch.no_grad():
for group in self.param_groups:
if "step" not in group:
group["step"] = 0
weight_decay = self.group_param_tensor(group, "weight_decay")
momentum = self.group_param_tensor(group, "momentum")
dampening = self.group_param_tensor(group, "dampening")
nesterov = group["nesterov"]
learning_rate = self.group_param_tensor(group, "lr")
for p in group["params"]:
if p.grad is None:
continue
state = self.state[p]
# State initialization
if len(state) == 0:
if momentum != 0:
state["momentum_buffer"] = p.grad.clone()
if "traced_step" not in state:
if isinstance(p, (ManifoldParameter, ManifoldTensor)):
manifold = p.manifold
else:
manifold = Euclidean()
if group["use_momentum"]:
state["traced_step"] = create_traced_update(
self.perform_step,
manifold,
p,
weight_decay.type_as(p),
momentum.type_as(p),
state["momentum_buffer"],
dampening=dampening,
nesterov=nesterov,
use_momentum=group["use_momentum"],
)
else:
state["traced_step"] = create_traced_update(
self.perform_step,
manifold,
p,
weight_decay.type_as(p),
momentum=None,
momentum_buffer=None,
dampening=dampening,
nesterov=nesterov,
use_momentum=group["use_momentum"],
)
if group["use_momentum"]:
state["traced_step"](
p,
p.grad,
learning_rate.type_as(p),
weight_decay.type_as(p),
momentum.type_as(p),
state["momentum_buffer"],
)
else:
state["traced_step"](
p, p.grad, learning_rate.type_as(p), weight_decay.type_as(p)
)
group["step"] += 1
if self._stabilize is not None and group["step"] % self._stabilize == 0:
self.stabilize_group(group)
return loss
@staticmethod
def perform_step(
manifold,
point,
grad,
lr,
weight_decay,
momentum,
momentum_buffer,
dampening,
nesterov,
use_momentum,
):
grad.add_(weight_decay, point)
grad = manifold.egrad2rgrad(point, grad)
if use_momentum:
momentum_buffer.mul_(momentum).add_(1 - dampening, grad)
if nesterov:
grad = grad.add_(momentum, momentum_buffer)
else:
grad = momentum_buffer
# we have all the things projected
new_point, new_momentum_buffer = manifold.retr_transp(
point, momentum_buffer, u=grad, t=-lr
)
momentum_buffer.set_(new_momentum_buffer)
point.set_(new_point)
else:
new_point = manifold.retr(point, grad, -lr)
point.set_(new_point)
def stabilize_group(self, group):
with torch.no_grad():
for p in group["params"]:
if not isinstance(p, (ManifoldParameter, ManifoldTensor)):
continue
manifold = p.manifold
momentum = group["momentum"]
p.set_(manifold.projx(p))
if momentum > 0:
param_state = self.state[p]
if not param_state: # due to None grads
continue
if "momentum_buffer" in param_state:
buf = param_state["momentum_buffer"]
buf.set_(manifold.proju(p, buf))
def _sanitize_group(self, group):
group = group.copy()
if isinstance(group["weight_decay"], torch.Tensor):
group["weight_decay"] = group["weight_decay"].item()
if isinstance(group["dampening"], torch.Tensor):
group["dampening"] = group["dampening"].item()
if isinstance(group["momentum"], torch.Tensor):
group["momentum"] = group["momentum"].item()
return group