File size: 30,059 Bytes
6c49103
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.

# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
# --------------------------------------------------------
# References:
# DeiT: https://github.com/facebookresearch/deit
# BEiT: https://github.com/microsoft/unilm/tree/master/beit
# --------------------------------------------------------

import builtins
import datetime
import os
import time
from collections import defaultdict, deque
from pathlib import Path

import torch
import torch.distributed as dist
from torch import inf
import numpy as np
from torchvision.transforms import functional as F

from typing import Optional, Tuple, Union, List

def randn_tensor(
    shape: Union[Tuple, List],
    generator: Optional[Union[List["torch.Generator"], "torch.Generator"]] = None,
    device: Optional["torch.device"] = None,
    dtype: Optional["torch.dtype"] = None,
    layout: Optional["torch.layout"] = None,
):
    """A helper function to create random tensors on the desired `device` with the desired `dtype`. When
    passing a list of generators, you can seed each batch size individually. If CPU generators are passed, the tensor
    is always created on the CPU.
    """
    # device on which tensor is created defaults to device
    rand_device = device
    batch_size = shape[0]

    layout = layout or torch.strided
    device = device or torch.device("cpu")

    if generator is not None:
        gen_device_type = generator.device.type if not isinstance(generator, list) else generator[0].device.type
        if gen_device_type != device.type and gen_device_type == "cpu":
            rand_device = "cpu"
            if device != "mps":
                print(
                    f"The passed generator was created on 'cpu' even though a tensor on {device} was expected."
                    f" Tensors will be created on 'cpu' and then moved to {device}. Note that one can probably"
                    f" slighly speed up this function by passing a generator that was created on the {device} device."
                )
        elif gen_device_type != device.type and gen_device_type == "cuda":
            raise ValueError(f"Cannot generate a {device} tensor from a generator of type {gen_device_type}.")

    # make sure generator list of length 1 is treated like a non-list
    if isinstance(generator, list) and len(generator) == 1:
        generator = generator[0]

    if isinstance(generator, list):
        shape = (1,) + shape[1:]
        latents = [
            torch.randn(shape, generator=generator[i], device=rand_device, dtype=dtype, layout=layout)
            for i in range(batch_size)
        ]
        latents = torch.cat(latents, dim=0).to(device)
    else:
        latents = torch.randn(shape, generator=generator, device=rand_device, dtype=dtype, layout=layout).to(device)

    return latents

class DiagonalGaussianDistribution(object):
    def __init__(self, parameters: torch.Tensor, deterministic: bool = False):
        self.parameters = parameters
        self.mean, self.logvar = torch.chunk(parameters, 2, dim=1)
        self.logvar = torch.clamp(self.logvar, -30.0, 20.0)
        self.deterministic = deterministic
        self.std = torch.exp(0.5 * self.logvar)
        self.var = torch.exp(self.logvar)
        if self.deterministic:
            self.var = self.std = torch.zeros_like(
                self.mean, device=self.parameters.device, dtype=self.parameters.dtype
            )

    def sample(self, generator: Optional[torch.Generator] = None) -> torch.Tensor:
        # make sure sample is on the same device as the parameters and has same dtype
        sample = randn_tensor(
            self.mean.shape,
            generator=generator,
            device=self.parameters.device,
            dtype=self.parameters.dtype,
        )
        x = self.mean + self.std * sample
        return x

    def kl(self, other: "DiagonalGaussianDistribution" = None) -> torch.Tensor:
        if self.deterministic:
            return torch.Tensor([0.0])
        else:
            sum_dim = self.mean.dim()
            if other is None:
                return 0.5 * torch.sum(
                    torch.pow(self.mean, 2) + self.var - 1.0 - self.logvar,
                    dim=list(range(1,sum_dim)),
                )
            else:
                return 0.5 * torch.sum(
                    torch.pow(self.mean - other.mean, 2) / other.var
                    + self.var / other.var
                    - 1.0
                    - self.logvar
                    + other.logvar,
                    dim=list(range(1,sum_dim)),
                )

    def nll(self, sample: torch.Tensor, dims: Tuple[int, ...] = [1, 2, 3]) -> torch.Tensor:
        if self.deterministic:
            return torch.Tensor([0.0])
        logtwopi = np.log(2.0 * np.pi)
        return 0.5 * torch.sum(
            logtwopi + self.logvar + torch.pow(sample - self.mean, 2) / self.var,
            dim=dims,
        )

    def mode(self) -> torch.Tensor:
        return self.mean
    
def set_for_tuning_decoder(args, model):
    args.mask_ratio = 0.0
    model.mask_token = None
    for name, param in model.named_parameters():
        if 'decoder' not in name and 'to_latent' not in name:
            param.requires_grad = False
    
    for name, param in model.named_parameters():
        if param.requires_grad == False:
            print(f"{name}: requires_grad = {param.requires_grad}")
    
def set_for_tuning_decoder_vae(args, model):
    for name, param in model.named_parameters():
        if 'post_quant_conv' in name or 'decoder' in name:
            param.requires_grad = True
        else:
            param.requires_grad = False
            
    for name, param in model.named_parameters():
        print(f"{name}: requires_grad = {param.requires_grad}")

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 v is None:
                continue
            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):
        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(len(iterable)))) + 'd'
        log_msg = [
            header,
            '[{0' + space_fmt + '}/{1}]',
            'eta: {eta}',
            '{meters}',
            'time: {time}',
            'data: {data}'
        ]
        if torch.cuda.is_available():
            log_msg.append('max mem: {memory:.0f}')
        log_msg = self.delimiter.join(log_msg)
        MB = 1024.0 * 1024.0
        for obj in iterable:
            data_time.update(time.time() - end)
            yield obj
            iter_time.update(time.time() - end)
            if i % print_freq == 0 or i == len(iterable) - 1:
                eta_seconds = iter_time.global_avg * (len(iterable) - i)
                eta_string = str(datetime.timedelta(seconds=int(eta_seconds)))
                if torch.cuda.is_available():
                    print(log_msg.format(
                        i, len(iterable), 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, len(iterable), 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: {} ({:.4f} s / it)'.format(
            header, total_time_str, total_time / len(iterable)))


def setup_for_distributed(is_master):
    """
    This function disables printing when not in master process
    """
    builtin_print = builtins.print

    def print(*args, **kwargs):
        force = kwargs.pop('force', False)
        force = force or (get_world_size() > 8)
        if is_master or force:
            now = datetime.datetime.now().time()
            builtin_print('[{}] '.format(now), end='')  # print with time stamp
            builtin_print(*args, **kwargs)

    builtins.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 args.dist_on_itp:
        args.rank = int(os.environ['OMPI_COMM_WORLD_RANK'])
        args.world_size = int(os.environ['OMPI_COMM_WORLD_SIZE'])
        args.gpu = int(os.environ['OMPI_COMM_WORLD_LOCAL_RANK'])
        args.dist_url = "tcp://%s:%s" % (os.environ['MASTER_ADDR'], os.environ['MASTER_PORT'])
        os.environ['LOCAL_RANK'] = str(args.gpu)
        os.environ['RANK'] = str(args.rank)
        os.environ['WORLD_SIZE'] = str(args.world_size)
        # ["RANK", "WORLD_SIZE", "MASTER_ADDR", "MASTER_PORT", "LOCAL_RANK"]
    elif 'RANK' in os.environ and 'WORLD_SIZE' in os.environ:
        # HERE
        args.rank = int(os.environ["RANK"])
        args.world_size = int(os.environ['WORLD_SIZE'])
        args.gpu = int(os.environ['LOCAL_RANK'])
    elif 'SLURM_PROCID' in os.environ:
        args.rank = int(os.environ['SLURM_PROCID'])
        args.gpu = args.rank % torch.cuda.device_count()
    else:
        print('Not using distributed mode')
        setup_for_distributed(is_master=True)  # hack
        args.distributed = False
        return

    args.distributed = True
        
    torch.cuda.set_device(args.gpu)
    args.dist_backend = 'nccl'
    print('| distributed init (rank {}): {}, gpu {}'.format(
        args.rank, args.dist_url, args.gpu), flush=True)
    from datetime import timedelta
    torch.distributed.init_process_group(backend=args.dist_backend, init_method=args.dist_url,
                                         world_size=args.world_size, rank=args.rank,
                                            timeout=timedelta(minutes=30) )
    torch.distributed.barrier()
    setup_for_distributed(args.rank == 0)
    


class NativeScalerWithGradNormCount:
    state_dict_key = "amp_scaler"

    def __init__(self):
        # self._scaler = torch.cuda.amp.GradScaler()
        self._scaler = torch.amp.GradScaler("cuda")

    def __call__(self, loss, optimizer, clip_grad=None, parameters=None, create_graph=False, update_grad=True):
        # loss.backward()
        # optimizer.step()
        self._scaler.scale(loss).backward(create_graph=create_graph)
        if update_grad:
            if clip_grad is not None:
                assert parameters is not None
                self._scaler.unscale_(optimizer)  # unscale the gradients of optimizer's assigned params in-place
                norm = torch.nn.utils.clip_grad_norm_(parameters, clip_grad)
            else:
                self._scaler.unscale_(optimizer)
                norm = get_grad_norm_(parameters)
            self._scaler.step(optimizer)
            self._scaler.update()
        else:
            norm = None
        return norm

    def state_dict(self):
        return self._scaler.state_dict()

    def load_state_dict(self, state_dict):
        self._scaler.load_state_dict(state_dict)


def get_grad_norm_(parameters, norm_type: float = 2.0) -> torch.Tensor:
    if isinstance(parameters, torch.Tensor):
        parameters = [parameters]
    parameters = [p for p in parameters if p.grad is not None]
    norm_type = float(norm_type)
    if len(parameters) == 0:
        return torch.tensor(0.)
    device = parameters[0].grad.device
    if norm_type == inf:
        total_norm = max(p.grad.detach().abs().max().to(device) for p in parameters)
    else:
        total_norm = torch.norm(torch.stack([torch.norm(p.grad.detach(), norm_type).to(device) for p in parameters]), norm_type)
    return total_norm


def save_model_vqvae(args, epoch, model, model_without_ddp, optimizer_ae, optimizer_disc):
    output_dir = Path(args.output_dir)
    epoch_name = str(epoch)
    checkpoint_paths = [output_dir / ('checkpoint-%s.pth' % epoch_name)]
    for checkpoint_path in checkpoint_paths:
        to_save = {
            'model': model_without_ddp.state_dict(),
            'optimizer_ae': optimizer_ae.state_dict(),
            'optimizer_disc': optimizer_disc.state_dict(),
            'epoch': epoch,
            'args': args,
        }

        save_on_master(to_save, checkpoint_path)

def save_model(args, epoch, model, model_without_ddp, optimizer, loss_scaler):
    output_dir = Path(args.output_dir)
    epoch_name = str(epoch)
    if loss_scaler is not None:
        checkpoint_paths = [output_dir / ('checkpoint-%s.pth' % epoch_name)]
        for checkpoint_path in checkpoint_paths:
            to_save = {
                'model': model_without_ddp.state_dict(),
                'optimizer': optimizer.state_dict(),
                'epoch': epoch,
                'scaler': loss_scaler.state_dict(),
                'args': args,
            }

            save_on_master(to_save, checkpoint_path)
    else:
        client_state = {'epoch': epoch}
        model.save_checkpoint(save_dir=args.output_dir, tag="checkpoint-%s" % epoch_name, client_state=client_state)

        
def resize_pos_embed(pos_embed, new_size):
    _, HW, D = pos_embed.shape
    H = int(HW ** 0.5)
    assert H * H == HW
    pos_embed_2d_resized = torch.nn.functional.interpolate(
        pos_embed.reshape(1,H,H,D).permute(0, 3, 1, 2),  # (batch, channels, height, width)
        size=(new_size, new_size),
        mode='bilinear',
        align_corners=False
    ).permute(0, 2, 3, 1).reshape(1,-1,D)  # (batch, height, width, channels)
    
    return pos_embed_2d_resized

def load_model(args, model_without_ddp, optimizer, loss_scaler):
    if args.resume:
        if args.resume.startswith('https'):
            checkpoint = torch.hub.load_state_dict_from_url(
                args.resume, map_location='cpu', check_hash=True)
        else:
            checkpoint = torch.load(args.resume, map_location='cpu')
            
        state_dict = checkpoint['model']
        
        if state_dict['pos_embed'].shape[1] != model_without_ddp.pos_embed.shape[1]:
            new_size = int(model_without_ddp.pos_embed.shape[1] ** 0.5)
            print(f'latent resolution is {new_size} x {new_size}, reshape pos embedding')
            print(f"prev pos embedding size: {state_dict['pos_embed'].shape}")
            state_dict['pos_embed'] = resize_pos_embed(state_dict['pos_embed'], new_size)
            print(f"new pos embedding size: {state_dict['pos_embed'].shape}")
            
            print(f"prev dec pos embedding size: {state_dict['decoder_pos_embed'].shape}")
            state_dict['decoder_pos_embed'] = resize_pos_embed(state_dict['decoder_pos_embed'], new_size)
            print(f"new dec pos embedding size: {state_dict['decoder_pos_embed'].shape}")
        
        msg = model_without_ddp.load_state_dict(state_dict, strict=False)
        print(msg)
        print("Resume checkpoint %s" % args.resume)
        if not args.tune_decoder:
            if 'optimizer' in checkpoint and 'epoch' in checkpoint and not (hasattr(args, 'eval') and args.eval):
                optimizer.load_state_dict(checkpoint['optimizer'])
                args.start_epoch = checkpoint['epoch'] + 1
                if 'scaler' in checkpoint:
                    loss_scaler.load_state_dict(checkpoint['scaler'])
                print("With optim & sched!")


def all_reduce_mean(x):
    world_size = get_world_size()
    if world_size > 1:
        x_reduce = torch.tensor(x).cuda()
        dist.all_reduce(x_reduce)
        x_reduce /= world_size
        return x_reduce.item()
    else:
        return x

def all_reduce_sum(x):
    world_size = get_world_size()
    if world_size > 1:
        x_reduce = torch.tensor(x).cuda()
        dist.all_reduce(x_reduce)
        return x_reduce.item()
    else:
        return x

def write_stat(t, num_rows, path, len_dataset):
    t_np = t.numpy().reshape(num_rows, -1)
    if os.path.isfile(path):
        stat = np.loadtxt(path, delimiter=',').astype(np.int64)
        stat = stat.reshape(num_rows, -1)
        stat = np.concatenate((stat, t_np), axis=1)
        save = np.savetxt(path, stat, delimiter=',', fmt='%d')
    else:
        save = np.savetxt(path, t_np, delimiter=',', fmt='%d')
    
    check = np.loadtxt(path, delimiter=',').astype(np.int64)/len_dataset
    print(f'count_convergence is activated.\n{check.round(3)*100}')


import math
class SequentialDistributedSampler(torch.utils.data.sampler.Sampler):
    """
    Distributed Sampler that subsamples indicies sequentially,
    making it easier to collate all results at the end.
    Even though we only use this sampler for eval and predict (no training),
    which means that the model params won't have to be synced (i.e. will not hang
    for synchronization even if varied number of forward passes), we still add extra
    samples to the sampler to make it evenly divisible (like in `DistributedSampler`)
    to make it easy to `gather` or `reduce` resulting tensors at the end of the loop.
    """
 
    def __init__(self, dataset, batch_size, rank=None, num_replicas=None):
        if num_replicas is None:
            if not torch.distributed.is_available():
                raise RuntimeError("Requires distributed package to be available")
            num_replicas = torch.distributed.get_world_size()
        if rank is None:
            if not torch.distributed.is_available():
                raise RuntimeError("Requires distributed package to be available")
            rank = torch.distributed.get_rank()
        self.dataset = dataset
        self.num_replicas = num_replicas
        self.rank = rank
        self.batch_size = batch_size
        self.num_samples = int(math.ceil(len(self.dataset) * 1.0 / self.batch_size / self.num_replicas)) * self.batch_size
        self.total_size = self.num_samples * self.num_replicas
 
    def __iter__(self):
        indices = list(range(len(self.dataset)))
        # add extra samples to make it evenly divisible
        indices += [indices[-1]] * (self.total_size - len(indices))
        # subsample
        indices = indices[self.rank * self.num_samples : (self.rank + 1) * self.num_samples]
        return iter(indices)
 
    def __len__(self):
        return self.num_samples

@torch.no_grad()
def update_mask(model, data_loader, device, dataset_train, target_attn, 
                mask_ratio = 0.75, ref_cluster = 'eigen', store_mask = False):
    print("Starts upadating informed mask...")
    len_ds = len(dataset_train)
    model.eval()

    metric_logger = MetricLogger(delimiter="  ")
    header = 'Upadating informed mask...'
    print_freq = 20
    masks_weights =[]
    mask_indices = []

    for data_iter_step, (samples, _, index, _, path_first, path_second, _) in enumerate(metric_logger.log_every(data_loader, print_freq, header)):
        samples = samples.to(device, non_blocking=True)
        path_first = path_first.to(device, non_blocking=True)
        path_second = path_second.to(device, non_blocking=True)
        index = index.to(device, non_blocking=True)
        with torch.cuda.amp.autocast():
            if ref_cluster == 'alternate':
                new_ids_shuffle_1, ref_cluster_size = model.forward_encoder_inference(samples, target_attn, 
                                                        mask_ratio = mask_ratio, ref_cluster = 'small',
                                                        return_score = True)
                new_ids_shuffle_2, ref_cluster_size = model.forward_encoder_inference(samples, target_attn, 
                                                        mask_ratio = mask_ratio, ref_cluster = 'small',
                                                        return_score = True, force_flip=True)
                new_ids_shuffle = torch.stack([new_ids_shuffle_1, new_ids_shuffle_2], dim=0) # 2 B N
                new_ids_shuffle = new_ids_shuffle.permute(1,0,2) # B 2 N
                # print(f'alternate: {new_ids_shuffle.shape}')
            else:
                new_ids_shuffle, ref_cluster_size = model.forward_encoder_inference(samples, target_attn, 
                                                        mask_ratio = mask_ratio, ref_cluster = ref_cluster,
                                                        return_score = True)
        # print(torch.cat([index, path_first, path_second]))
        # mask_info = torch.cat([new_ids_shuffle, ref_cluster_size.unsqueeze(-1), path_first.unsqueeze(-1), path_second.unsqueeze(-1)], dim=-1)
        mask_info = new_ids_shuffle
        # mask_index = torch.cat([index.unsqueeze(-1),path_first.unsqueeze(-1), path_second.unsqueeze(-1)], dim=-1)
        if data_iter_step % 200 == 0:
            print(f'ids_shuffle: {new_ids_shuffle.shape}, ref_cluster_size: {ref_cluster_size.shape}')
            print(f'mask_info: {mask_info.shape}')
        if data_iter_step ==0:
            print('Saving...')
            examples = mask_info.detach().cpu().numpy() 
            store_path = f'/data2/projects/jeongwoo/jeongwoo/mae/analysis/convergence/mask_samples_{ref_cluster}'
            save = np.save(store_path, examples)
        masks_weights.append(mask_info)
        # mask_indices.append(mask_index)
    

    masks_weights = torch.cat(masks_weights, dim=0)
    # mask_indices = torch.cat(mask_indices, dim=0)
    print(f'masks_weights: {masks_weights.shape}')
    dist.barrier()
    gather_masks = [torch.ones_like(masks_weights) for _ in range(dist.get_world_size())]
    # gather_mask_index = [torch.ones_like(mask_indices) for _ in range(dist.get_world_size())]
    dist.all_gather(gather_masks, masks_weights)
    # dist.all_gather(gather_mask_index, mask_indices)
    all_mask_weights = torch.cat(gather_masks)
    # all_mask_indices = torch.cat(gather_mask_index)
    
    all_mask_weights = all_mask_weights[:len_ds]
    # all_mask_indices = all_mask_indices[:len_ds]
    
    if store_mask:
        weights_to_store = all_mask_weights.cpu().numpy()
        store_path = f'/data2/projects/jeongwoo/jeongwoo/mae/analysis/convergence/stored_masks_{ref_cluster}'
        save = np.save(store_path, weights_to_store)
    
    dataset_train.mask = all_mask_weights.cpu()
    # dataset_train.mask_index = all_mask_indices.cpu()
    print("Informed masks have been updated")


import torchvision.transforms as transforms
class maskRandomResizedCrop(transforms.RandomResizedCrop):
    def __init__(self, size, **kwargs):
        super().__init__(size, **kwargs)
        self.mask_size = 14
    
    def forward(self, img, mask):
        mask = mask.reshape(14,14)
        i, j, h, w = self.get_params(img, self.scale, self.ratio)
        m_h_s = int(14 * (i/img.size[1]))
        m_h_e = int(14 * ((i+h)/img.size[1])) + 1
        m_w_s = int(14 * (j/img.size[0]))
        m_w_e = int(14 * ((j+w)/img.size[0])) + 1
        mask = mask[m_h_s:m_h_e, m_w_s:m_w_e]

        img = F.resized_crop(img, i, j, h, w, self.size, self.interpolation)
        mask = F.resize(mask.unsqueeze(0), (self.mask_size, self.mask_size))
        mask = mask.flatten()
        
        return img, mask
    
class maskRandomHorizontalFlip(transforms.RandomHorizontalFlip):
    def __init__(self):
        super().__init__()

    def forward(self, img, mask):
        mask = mask.reshape(14, 14)
        if torch.rand(1) < self.p:
            img, mask = F.hflip(img), F.hflip(mask)
        mask = mask.flatten()
        return img, mask

class trainCompose(transforms.Compose):
    def __call__(self, img, mask, hint_prob=False): 

        #mask needs to be processed individually for some operations
        for i in self.transforms[:2]:
            img, mask = i(img,mask)

        # if not hint_prob:
        #     mask = torch.argsort(
        #         mask, dim=0, descending=False
        #     ) 
        
        for t in self.transforms[2:]:
            img = t(img)

        return img, mask
    
# def schedule_hint(hint_ratio, hint_portion, do_schedule, cur_epoch, total_epoch):
#     if hint_ratio is None: return None
#     L = 196
#     if do_schedule:
#         alpha = 1 - (cur_epoch/total_epoch)**3 # 1 to 0
#         hint_ratio = hint_ratio * alpha
#         hint_portion = max(hint_portion, 0.2)
#         hint_portion = alpha * (hint_portion - 0.2) + 0.2
#     cluster_size = int(hint_portion*L)
#     hint_num = max(int(hint_ratio * cluster_size), 2)
#     print(f'{hint_num} tokens for hint in epoch {cur_epoch}')
#     return hint_num
def schedule_hint(hint_ratio, hint_portion, do_schedule, cur_epoch, total_epoch, min_portion, min_ratio, schedule_exp, full_schedule = False):
    if hint_ratio is None: return None
    L = 196
    if do_schedule:
        assert hint_portion >= min_portion, 'min_portion is bigger than hint_portion.'
        assert hint_ratio >= min_ratio, 'min_ratio is bigger than hint_ratio.'
        
        if full_schedule:
            total_epoch = 800
            alpha = 1 - ((cur_epoch-0)/(total_epoch-0))**schedule_exp # 1 to 0
        else:
            alpha = 1 - ((cur_epoch-100)/(total_epoch-100))**schedule_exp # 1 to 0
        
        
        hint_ratio = alpha * (hint_ratio - min_ratio) + min_ratio
        hint_portion = alpha * (hint_portion - min_portion) + min_portion
        
    hint_num = max(int(hint_ratio * L), 2)
    print(f'{hint_num} tokens for hint in epoch {cur_epoch}')
    print(f'Hint ratio & hint_portion: {hint_ratio, hint_portion} in epoch {cur_epoch}')
    return hint_ratio, hint_portion


import torchvision.datasets as datasets
import random
class NormalImgDataset(datasets.ImageFolder):
    def __init__(self, **kwargs):
        super().__init__(**kwargs)
        self.num_retries = 10
        
    def __getitem__(self, index: int):
        """
        Args:
            index (int): Index

        Returns:
            tuple: (sample, target) where target is class_index of the target class.
        """
        failed = []
        for _ in range(self.num_retries):
            path, target = self.samples[index]
            try:
                sample = self.loader(path)
            except:
                try:
                    sample = self.loader(path) # one more time
                except:
                    failed.append(path)
                    index = random.randint(0, len(self.samples) - 1)
                    continue
            if self.transform is not None:
                sample = self.transform(sample)
            if self.target_transform is not None:
                target = self.target_transform(target)
            
            return sample, target, torch.tensor(1)
        else:
            print('Failed to load {} after {} retries'.format(
                failed, self.num_retries
            ))