forked from lanl/OpenFWI
-
Notifications
You must be signed in to change notification settings - Fork 0
/
utils.py
474 lines (398 loc) · 16.6 KB
/
utils.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
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
# © 2022. Triad National Security, LLC. All rights reserved.
# This program was produced under U.S. Government contract 89233218CNA000001 for Los Alamos
# National Laboratory (LANL), which is operated by Triad National Security, LLC for the U.S.
# Department of Energy/National Nuclear Security Administration. All rights in the program are
# reserved by Triad National Security, LLC, and the U.S. Department of Energy/National Nuclear
# Security Administration. The Government is granted for itself and others acting on its behalf a
# nonexclusive, paid-up, irrevocable worldwide license in this material to reproduce, prepare
# derivative works, distribute copies to the public, perform publicly and display publicly, and to permit
# others to do so.
from collections import defaultdict, deque
import datetime
import time
import torch
import torch.distributed as dist
import torch.autograd as autograd
from torch.autograd import Variable
import torch.nn as nn
import errno
import os
import itertools
from torchvision.models import vgg16
import numpy as np
class SmoothedValue(object):
"""Track a series of values and provide access to smoothed values over a
window or the global series average.
"""
def __init__(self, window_size=20, fmt=None):
if fmt is None:
fmt = "{median:.4f} ({global_avg:.4f})"
self.deque = deque(maxlen=window_size)
self.total = 0.0
self.count = 0
self.fmt = fmt
def update(self, value, n=1):
self.deque.append(value)
self.count += n
self.total += value * n
def synchronize_between_processes(self):
"""
Warning: does not synchronize the deque!
"""
if not is_dist_avail_and_initialized():
return
t = torch.tensor([self.count, self.total], dtype=torch.float64, device='cuda')
dist.barrier()
dist.all_reduce(t)
t = t.tolist()
self.count = int(t[0])
self.total = t[1]
@property
def median(self):
d = torch.tensor(list(self.deque))
return d.median().item()
@property
def avg(self):
d = torch.tensor(list(self.deque), dtype=torch.float32)
return d.mean().item()
@property
def global_avg(self):
return self.total / self.count
@property
def max(self):
return max(self.deque)
@property
def value(self):
return self.deque[-1]
def __str__(self):
return self.fmt.format(
median=self.median,
avg=self.avg,
global_avg=self.global_avg,
max=self.max,
value=self.value)
class MetricLogger(object):
def __init__(self, delimiter="\t"):
self.meters = defaultdict(SmoothedValue)
self.delimiter = delimiter
def update(self, **kwargs):
for k, v in kwargs.items():
if isinstance(v, torch.Tensor):
v = v.item()
assert isinstance(v, (float, int))
self.meters[k].update(v)
def __getattr__(self, attr):
if attr in self.meters:
return self.meters[attr]
if attr in self.__dict__:
return self.__dict__[attr]
raise AttributeError("'{}' object has no attribute '{}'".format(
type(self).__name__, attr))
def __str__(self):
loss_str = []
for name, meter in self.meters.items():
loss_str.append(
"{}: {}".format(name, str(meter))
)
return self.delimiter.join(loss_str)
def synchronize_between_processes(self):
for meter in self.meters.values():
meter.synchronize_between_processes()
def add_meter(self, name, meter):
self.meters[name] = meter
def log_every(self, iterable, print_freq, header=None):
if isinstance(iterable, list):
length = max(len(x) for x in iterable)
iterable = [x if len(x) == length else itertools.cycle(x) for x in iterable]
iterable = zip(*iterable)
else:
length = len(iterable)
i = 0
if not header:
header = ''
start_time = time.time()
end = time.time()
iter_time = SmoothedValue(fmt='{avg:.4f}')
data_time = SmoothedValue(fmt='{avg:.4f}')
space_fmt = ':' + str(len(str(length))) + 'd'
if torch.cuda.is_available():
log_msg = self.delimiter.join([
header,
'[{0' + space_fmt + '}/{1}]',
'eta: {eta}',
'{meters}',
'time: {time}',
'data: {data}',
'max mem: {memory:.0f}'
])
else:
log_msg = self.delimiter.join([
header,
'[{0' + space_fmt + '}/{1}]',
'eta: {eta}',
'{meters}',
'time: {time}',
'data: {data}'
])
MB = 1024.0 * 1024.0
for obj in iterable:
data_time.update(time.time() - end)
yield obj # <-- yield the batch in for loop
iter_time.update(time.time() - end)
if i % print_freq == 0:
eta_seconds = iter_time.global_avg * (length - i)
eta_string = str(datetime.timedelta(seconds=int(eta_seconds)))
if torch.cuda.is_available():
print(log_msg.format(
i, length, eta=eta_string,
meters=str(self),
time=str(iter_time), data=str(data_time),
memory=torch.cuda.max_memory_allocated() / MB))
else:
print(log_msg.format(
i, length, eta=eta_string,
meters=str(self),
time=str(iter_time), data=str(data_time)))
i += 1
end = time.time()
total_time = time.time() - start_time
total_time_str = str(datetime.timedelta(seconds=int(total_time)))
print('{} Total time: {}'.format(header, total_time_str))
# Legacy code
class ContentLoss(nn.Module):
def __init__(self, args):
super(ContentLoss, self).__init__()
names = ['l1', 'l2']
self.loss_names = ['loss_' + n for n in names]
for key in ['lambda_' + n for n in names]:
setattr(self, key, getattr(args, key))
self.l1loss = nn.L1Loss()
self.l2loss = nn.MSELoss()
def forward(self, model, input, target):
pred = model(input)
loss_l1 = self.l1loss(target, pred)
loss_l2 = self.l2loss(target, pred)
loss = loss_l1 * self.lambda_l1 + loss_l2 * self.lambda_l2
scope = locals()
return loss, { k: eval(k, scope) for k in self.loss_names }
# Legacy code
class IdenticalLoss(nn.Module):
def __init__(self, args):
super(IdenticalLoss, self).__init__()
names = ['id1s', 'id2s']
self.loss_names = ['loss_' + n for n in names]
for key in ['lambda_' + n for n in names]:
setattr(self, key, getattr(args, key))
self.l1loss = nn.L1Loss()
self.l2loss = nn.MSELoss()
def forward(self, model_s2v, model_v2s, input):
mid = model_s2v(input)
pred = model_v2s(mid)
cal_loss = lambda x, y: (self.l1loss(x, y), self.l2loss(x, y))
loss_id1s, loss_id2s = cal_loss(input, pred)
loss = loss_id1s * self.lambda_id1s + loss_id2s * self.lambda_id2s
scope = locals()
return loss, { k: eval(k, scope) for k in self.loss_names }
# Implemented according to H-PGNN, not useful
class NMSELoss(nn.Module):
def __init__(self):
super(NMSELoss, self).__init__()
def forward(self, pred, gt):
return torch.mean(((pred - gt) / (torch.amax(gt, (-2, -1), keepdim=True) + 1e-5)) ** 2)
class CycleLoss(nn.Module):
def __init__(self, args):
super(CycleLoss, self).__init__()
names = ['g1v', 'g2v', 'g1s', 'g2s', 'c1v', 'c2v', 'c1s', 'c2s']
self.loss_names = ['loss_' + n for n in names]
for key in ['lambda_' + n for n in names]:
setattr(self, key, getattr(args, key))
self.l1loss = nn.L1Loss()
self.l2loss = nn.MSELoss()
def forward(self, data, label, pred_s=None, pred_v=None, recon_s=None, recon_v=None):
cal_loss = lambda x, y: (self.l1loss(x, y), self.l2loss(x, y))
loss_g1v, loss_g2v, loss_g1s, loss_g2s = [0] * 4
if pred_v is not None:
loss_g1v, loss_g2v = cal_loss(pred_v, label)
if pred_s is not None:
loss_g1s, loss_g2s = cal_loss(pred_s, data)
loss_c1v, loss_c2v, loss_c1s , loss_c2s = [0] * 4
if recon_v is not None:
loss_c1v, loss_c2v = cal_loss(recon_v, label)
if recon_s is not None:
loss_c1s, loss_c2s = cal_loss(recon_s, data)
loss = loss_g1v * self.lambda_g1v + loss_g2v * self.lambda_g2v + \
loss_g1s * self.lambda_g1s + loss_g2s * self.lambda_g2s + \
loss_c1v * self.lambda_c1v + loss_c2v * self.lambda_c2v + \
loss_c1s * self.lambda_c1s + loss_c2s * self.lambda_c2s
scope = locals()
return loss, { k: eval(k, scope) for k in self.loss_names }
# Legacy code
class _CycleLoss(nn.Module):
def __init__(self, args):
super(_CycleLoss, self).__init__()
names = ['g1v', 'g2v', 'g1s', 'g2s', 'c1v', 'c2v', 'c1s', 'c2s']
self.loss_names = ['loss_' + n for n in names]
for key in ['lambda_' + n for n in names]:
setattr(self, key, getattr(args, key))
self.l1loss = nn.L1Loss()
self.l2loss = nn.MSELoss()
def forward(self, data, label, pred_s=None, pred_v=None, recon_s=None, recon_v=None):
cal_loss = lambda x, y: (self.l1loss(x, y), self.l2loss(x, y))
loss_g1v, loss_g2v, loss_g1s, loss_g2s = [0] * 4
if pred_v is not None and (self.lambda_g1v != 0 or self.lambda_g2v != 0):
loss_g1v, loss_g2v = cal_loss(pred_v, label)
if pred_s is not None and (self.lambda_g1s != 0 or self.lambda_g2s != 0):
loss_g1s, loss_g2s = cal_loss(pred_s, data)
loss_c1v, loss_c2v, loss_c1s , loss_c2s = [0] * 4
if recon_v is not None and (self.lambda_c1v != 0 or self.lambda_c2v != 0):
loss_c1v, loss_c2v = cal_loss(recon_v, label)
if recon_s is not None and (self.lambda_c1s != 0 or self.lambda_c2s != 0):
loss_c1s, loss_c2s = cal_loss(recon_s, data)
loss = loss_g1v * self.lambda_g1v + loss_g2v * self.lambda_g2v + \
loss_g1s * self.lambda_g1s + loss_g2s * self.lambda_g2s + \
loss_c1v * self.lambda_c1v + loss_c2v * self.lambda_c2v + \
loss_c1s * self.lambda_c1s + loss_c2s * self.lambda_c2s
scope = locals()
return loss, { k: eval(k, scope) for k in self.loss_names }
def accuracy(output, target, topk=(1,)):
"""Computes the accuracy over the k top predictions for the specified values of k"""
with torch.no_grad():
maxk = max(topk)
batch_size = target.size(0)
_, pred = output.topk(maxk, 1, True, True)
pred = pred.t()
correct = pred.eq(target[None])
res = []
for k in topk:
correct_k = correct[:k].flatten().sum(dtype=torch.float32)
res.append(correct_k * (100.0 / batch_size))
return res
def mkdir(path):
try:
os.makedirs(path)
except OSError as e:
if e.errno != errno.EEXIST:
raise
def setup_for_distributed(is_master):
"""
This function disables printing when not in master process
"""
import builtins as __builtin__
builtin_print = __builtin__.print
def print(*args, **kwargs):
force = kwargs.pop('force', False)
if is_master or force:
builtin_print(*args, **kwargs)
__builtin__.print = print
def is_dist_avail_and_initialized():
if not dist.is_available():
return False
if not dist.is_initialized():
return False
return True
def get_world_size():
if not is_dist_avail_and_initialized():
return 1
return dist.get_world_size()
def get_rank():
if not is_dist_avail_and_initialized():
return 0
return dist.get_rank()
def is_main_process():
return get_rank() == 0
def save_on_master(*args, **kwargs):
if is_main_process():
torch.save(*args, **kwargs)
def init_distributed_mode(args):
if 'RANK' in os.environ and 'WORLD_SIZE' in os.environ:
args.rank = int(os.environ["RANK"])
args.world_size = int(os.environ['WORLD_SIZE'])
args.local_rank = int(os.environ['LOCAL_RANK'])
elif 'SLURM_PROCID' in os.environ and args.world_size > 1:
args.rank = int(os.environ['SLURM_PROCID'])
args.local_rank = args.rank % torch.cuda.device_count()
elif hasattr(args, "rank"):
pass
else:
print('Not using distributed mode')
args.distributed = False
return
args.distributed = True
torch.cuda.set_device(args.local_rank)
args.dist_backend = 'nccl'
print('| distributed init (rank {}): {}'.format(
args.rank, args.dist_url), flush=True)
torch.distributed.init_process_group(backend=args.dist_backend, init_method=args.dist_url,
world_size=args.world_size, rank=args.rank)
setup_for_distributed(args.rank == 0)
class Wasserstein_GP(nn.Module):
def __init__(self, device, lambda_gp):
super(Wasserstein_GP, self).__init__()
self.device = device
self.lambda_gp = lambda_gp
def forward(self, real, fake, model):
gradient_penalty = self.compute_gradient_penalty(model, real, fake)
loss_real = torch.mean(model(real))
loss_fake = torch.mean(model(fake))
loss = -loss_real + loss_fake + gradient_penalty * self.lambda_gp
return loss, loss_real-loss_fake, gradient_penalty
def compute_gradient_penalty(self, model, real_samples, fake_samples):
alpha = torch.rand(real_samples.size(0), 1, 1, 1, device=self.device)
interpolates = (alpha * real_samples + ((1 - alpha) * fake_samples)).requires_grad_(True)
d_interpolates = model(interpolates)
gradients = autograd.grad(
outputs=d_interpolates,
inputs=interpolates,
grad_outputs=torch.ones(real_samples.size(0), d_interpolates.size(1)).to(self.device),
create_graph=True,
retain_graph=True,
only_inputs=True,
)[0]
gradients = gradients.view(gradients.size(0), -1)
gradient_penalty = ((gradients.norm(2, dim=1) - 1) ** 2).mean()
return gradient_penalty
# Modified from https://gist.github.com/alper111/8233cdb0414b4cb5853f2f730ab95a49
class VGGPerceptualLoss(nn.Module):
def __init__(self, resize=True):
super(VGGPerceptualLoss, self).__init__()
blocks = []
blocks.append(vgg16(pretrained=True).features[:4].eval()) # relu1_2
blocks.append(vgg16(pretrained=True).features[4:9].eval()) # relu2_2
blocks.append(vgg16(pretrained=True).features[9:16].eval()) # relu3_3
blocks.append(vgg16(pretrained=True).features[16:23].eval()) # relu4_3
for bl in blocks:
for p in bl:
p.requires_grad = False
self.blocks = nn.ModuleList(blocks)
self.transform = nn.functional.interpolate
self.resize = resize
self.register_buffer("mean", torch.tensor([0.485, 0.456, 0.406]).view(1, 3, 1, 1))
self.register_buffer("std", torch.tensor([0.229, 0.224, 0.225]).view(1, 3, 1, 1))
self.l1loss = nn.L1Loss()
self.l2loss = nn.MSELoss()
def forward(self, input, target, rescale=True, feature_layers=[1]):
input = input.view(-1, 1, input.shape[-2], input.shape[-1]).repeat(1, 3, 1, 1)
target = target.view(-1, 1, target.shape[-2], target.shape[-1]).repeat(1, 3, 1, 1)
if rescale: # from [-1, 1] to [0, 1]
input = input / 2 + 0.5
target = target / 2 + 0.5
input = (input-self.mean) / self.std
target = (target-self.mean) / self.std
if self.resize:
input = self.transform(input, mode='bilinear', size=(224, 224), align_corners=False)
target = self.transform(target, mode='bilinear', size=(224, 224), align_corners=False)
loss_l1, loss_l2 = 0.0, 0.0
x = input
y = target
for i, block in enumerate(self.blocks):
x = block(x)
y = block(y)
if i in feature_layers:
loss_l1 += self.l1loss(x, y)
loss_l2 += self.l2loss(x, y)
return loss_l1, loss_l2
def cal_psnr(gt, data, max_value):
mse = np.mean((gt - data) ** 2)
if (mse == 0):
return 100
return 20 * np.log10(max_value / np.sqrt(mse))