File size: 36,487 Bytes
3e06e1c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
# Copyright (c) OpenMMLab. All rights reserved.
from typing import List, Optional, Tuple

import torch
import torch.nn as nn
import torch.nn.functional as F
from mmcv.cnn import ConvModule, Scale
from mmcv.ops import deform_conv2d
from mmengine import MessageHub
from mmengine.config import ConfigDict
from mmengine.model import bias_init_with_prob, normal_init
from mmengine.structures import InstanceData
from torch import Tensor

from mmdet.registry import MODELS, TASK_UTILS
from mmdet.structures.bbox import distance2bbox
from mmdet.utils import (ConfigType, InstanceList, OptConfigType,
                         OptInstanceList, reduce_mean)
from ..task_modules.prior_generators import anchor_inside_flags
from ..utils import (filter_scores_and_topk, images_to_levels, multi_apply,
                     sigmoid_geometric_mean, unmap)
from .atss_head import ATSSHead


class TaskDecomposition(nn.Module):
    """Task decomposition module in task-aligned predictor of TOOD.

    Args:
        feat_channels (int): Number of feature channels in TOOD head.
        stacked_convs (int): Number of conv layers in TOOD head.
        la_down_rate (int): Downsample rate of layer attention.
            Defaults to 8.
        conv_cfg (:obj:`ConfigDict` or dict, optional): Config dict for
            convolution layer. Defaults to None.
        norm_cfg (:obj:`ConfigDict` or dict, optional):  Config dict for
        normalization layer. Defaults to None.
    """

    def __init__(self,
                 feat_channels: int,
                 stacked_convs: int,
                 la_down_rate: int = 8,
                 conv_cfg: OptConfigType = None,
                 norm_cfg: OptConfigType = None) -> None:
        super().__init__()
        self.feat_channels = feat_channels
        self.stacked_convs = stacked_convs
        self.in_channels = self.feat_channels * self.stacked_convs
        self.norm_cfg = norm_cfg
        self.layer_attention = nn.Sequential(
            nn.Conv2d(self.in_channels, self.in_channels // la_down_rate, 1),
            nn.ReLU(inplace=True),
            nn.Conv2d(
                self.in_channels // la_down_rate,
                self.stacked_convs,
                1,
                padding=0), nn.Sigmoid())

        self.reduction_conv = ConvModule(
            self.in_channels,
            self.feat_channels,
            1,
            stride=1,
            padding=0,
            conv_cfg=conv_cfg,
            norm_cfg=norm_cfg,
            bias=norm_cfg is None)

    def init_weights(self) -> None:
        """Initialize the parameters."""
        for m in self.layer_attention.modules():
            if isinstance(m, nn.Conv2d):
                normal_init(m, std=0.001)
        normal_init(self.reduction_conv.conv, std=0.01)

    def forward(self,
                feat: Tensor,
                avg_feat: Optional[Tensor] = None) -> Tensor:
        """Forward function of task decomposition module."""
        b, c, h, w = feat.shape
        if avg_feat is None:
            avg_feat = F.adaptive_avg_pool2d(feat, (1, 1))
        weight = self.layer_attention(avg_feat)

        # here we first compute the product between layer attention weight and
        # conv weight, and then compute the convolution between new conv weight
        # and feature map, in order to save memory and FLOPs.
        conv_weight = weight.reshape(
            b, 1, self.stacked_convs,
            1) * self.reduction_conv.conv.weight.reshape(
                1, self.feat_channels, self.stacked_convs, self.feat_channels)
        conv_weight = conv_weight.reshape(b, self.feat_channels,
                                          self.in_channels)
        feat = feat.reshape(b, self.in_channels, h * w)
        feat = torch.bmm(conv_weight, feat).reshape(b, self.feat_channels, h,
                                                    w)
        if self.norm_cfg is not None:
            feat = self.reduction_conv.norm(feat)
        feat = self.reduction_conv.activate(feat)

        return feat


@MODELS.register_module()
class TOODHead(ATSSHead):
    """TOODHead used in `TOOD: Task-aligned One-stage Object Detection.

    <https://arxiv.org/abs/2108.07755>`_.

    TOOD uses Task-aligned head (T-head) and is optimized by Task Alignment
    Learning (TAL).

    Args:
        num_classes (int): Number of categories excluding the background
            category.
        in_channels (int): Number of channels in the input feature map.
        num_dcn (int): Number of deformable convolution in the head.
            Defaults to 0.
        anchor_type (str): If set to ``anchor_free``, the head will use centers
            to regress bboxes. If set to ``anchor_based``, the head will
            regress bboxes based on anchors. Defaults to ``anchor_free``.
        initial_loss_cls (:obj:`ConfigDict` or dict): Config of initial loss.

    Example:
        >>> self = TOODHead(11, 7)
        >>> feats = [torch.rand(1, 7, s, s) for s in [4, 8, 16, 32, 64]]
        >>> cls_score, bbox_pred = self.forward(feats)
        >>> assert len(cls_score) == len(self.scales)
    """

    def __init__(self,
                 num_classes: int,
                 in_channels: int,
                 num_dcn: int = 0,
                 anchor_type: str = 'anchor_free',
                 initial_loss_cls: ConfigType = dict(
                     type='FocalLoss',
                     use_sigmoid=True,
                     activated=True,
                     gamma=2.0,
                     alpha=0.25,
                     loss_weight=1.0),
                 **kwargs) -> None:
        assert anchor_type in ['anchor_free', 'anchor_based']
        self.num_dcn = num_dcn
        self.anchor_type = anchor_type
        super().__init__(
            num_classes=num_classes, in_channels=in_channels, **kwargs)

        if self.train_cfg:
            self.initial_epoch = self.train_cfg['initial_epoch']
            self.initial_assigner = TASK_UTILS.build(
                self.train_cfg['initial_assigner'])
            self.initial_loss_cls = MODELS.build(initial_loss_cls)
            self.assigner = self.initial_assigner
            self.alignment_assigner = TASK_UTILS.build(
                self.train_cfg['assigner'])
            self.alpha = self.train_cfg['alpha']
            self.beta = self.train_cfg['beta']

    def _init_layers(self) -> None:
        """Initialize layers of the head."""
        self.relu = nn.ReLU(inplace=True)
        self.inter_convs = nn.ModuleList()
        for i in range(self.stacked_convs):
            if i < self.num_dcn:
                conv_cfg = dict(type='DCNv2', deform_groups=4)
            else:
                conv_cfg = self.conv_cfg
            chn = self.in_channels if i == 0 else self.feat_channels
            self.inter_convs.append(
                ConvModule(
                    chn,
                    self.feat_channels,
                    3,
                    stride=1,
                    padding=1,
                    conv_cfg=conv_cfg,
                    norm_cfg=self.norm_cfg))

        self.cls_decomp = TaskDecomposition(self.feat_channels,
                                            self.stacked_convs,
                                            self.stacked_convs * 8,
                                            self.conv_cfg, self.norm_cfg)
        self.reg_decomp = TaskDecomposition(self.feat_channels,
                                            self.stacked_convs,
                                            self.stacked_convs * 8,
                                            self.conv_cfg, self.norm_cfg)

        self.tood_cls = nn.Conv2d(
            self.feat_channels,
            self.num_base_priors * self.cls_out_channels,
            3,
            padding=1)
        self.tood_reg = nn.Conv2d(
            self.feat_channels, self.num_base_priors * 4, 3, padding=1)

        self.cls_prob_module = nn.Sequential(
            nn.Conv2d(self.feat_channels * self.stacked_convs,
                      self.feat_channels // 4, 1), nn.ReLU(inplace=True),
            nn.Conv2d(self.feat_channels // 4, 1, 3, padding=1))
        self.reg_offset_module = nn.Sequential(
            nn.Conv2d(self.feat_channels * self.stacked_convs,
                      self.feat_channels // 4, 1), nn.ReLU(inplace=True),
            nn.Conv2d(self.feat_channels // 4, 4 * 2, 3, padding=1))

        self.scales = nn.ModuleList(
            [Scale(1.0) for _ in self.prior_generator.strides])

    def init_weights(self) -> None:
        """Initialize weights of the head."""
        bias_cls = bias_init_with_prob(0.01)
        for m in self.inter_convs:
            normal_init(m.conv, std=0.01)
        for m in self.cls_prob_module:
            if isinstance(m, nn.Conv2d):
                normal_init(m, std=0.01)
        for m in self.reg_offset_module:
            if isinstance(m, nn.Conv2d):
                normal_init(m, std=0.001)
        normal_init(self.cls_prob_module[-1], std=0.01, bias=bias_cls)

        self.cls_decomp.init_weights()
        self.reg_decomp.init_weights()

        normal_init(self.tood_cls, std=0.01, bias=bias_cls)
        normal_init(self.tood_reg, std=0.01)

    def forward(self, feats: Tuple[Tensor]) -> Tuple[List[Tensor]]:
        """Forward features from the upstream network.

        Args:
            feats (tuple[Tensor]): Features from the upstream network, each is
                a 4D-tensor.

        Returns:
            tuple: Usually a tuple of classification scores and bbox prediction
                cls_scores (list[Tensor]): Classification scores for all scale
                    levels, each is a 4D-tensor, the channels number is
                    num_anchors * num_classes.
                bbox_preds (list[Tensor]): Decoded box for all scale levels,
                    each is a 4D-tensor, the channels number is
                    num_anchors * 4. In [tl_x, tl_y, br_x, br_y] format.
        """
        cls_scores = []
        bbox_preds = []
        for idx, (x, scale, stride) in enumerate(
                zip(feats, self.scales, self.prior_generator.strides)):
            b, c, h, w = x.shape
            anchor = self.prior_generator.single_level_grid_priors(
                (h, w), idx, device=x.device)
            anchor = torch.cat([anchor for _ in range(b)])
            # extract task interactive features
            inter_feats = []
            for inter_conv in self.inter_convs:
                x = inter_conv(x)
                inter_feats.append(x)
            feat = torch.cat(inter_feats, 1)

            # task decomposition
            avg_feat = F.adaptive_avg_pool2d(feat, (1, 1))
            cls_feat = self.cls_decomp(feat, avg_feat)
            reg_feat = self.reg_decomp(feat, avg_feat)

            # cls prediction and alignment
            cls_logits = self.tood_cls(cls_feat)
            cls_prob = self.cls_prob_module(feat)
            cls_score = sigmoid_geometric_mean(cls_logits, cls_prob)

            # reg prediction and alignment
            if self.anchor_type == 'anchor_free':
                reg_dist = scale(self.tood_reg(reg_feat).exp()).float()
                reg_dist = reg_dist.permute(0, 2, 3, 1).reshape(-1, 4)
                reg_bbox = distance2bbox(
                    self.anchor_center(anchor) / stride[0],
                    reg_dist).reshape(b, h, w, 4).permute(0, 3, 1,
                                                          2)  # (b, c, h, w)
            elif self.anchor_type == 'anchor_based':
                reg_dist = scale(self.tood_reg(reg_feat)).float()
                reg_dist = reg_dist.permute(0, 2, 3, 1).reshape(-1, 4)
                reg_bbox = self.bbox_coder.decode(anchor, reg_dist).reshape(
                    b, h, w, 4).permute(0, 3, 1, 2) / stride[0]
            else:
                raise NotImplementedError(
                    f'Unknown anchor type: {self.anchor_type}.'
                    f'Please use `anchor_free` or `anchor_based`.')
            reg_offset = self.reg_offset_module(feat)
            bbox_pred = self.deform_sampling(reg_bbox.contiguous(),
                                             reg_offset.contiguous())

            # After deform_sampling, some boxes will become invalid (The
            # left-top point is at the right or bottom of the right-bottom
            # point), which will make the GIoULoss negative.
            invalid_bbox_idx = (bbox_pred[:, [0]] > bbox_pred[:, [2]]) | \
                               (bbox_pred[:, [1]] > bbox_pred[:, [3]])
            invalid_bbox_idx = invalid_bbox_idx.expand_as(bbox_pred)
            bbox_pred = torch.where(invalid_bbox_idx, reg_bbox, bbox_pred)

            cls_scores.append(cls_score)
            bbox_preds.append(bbox_pred)
        return tuple(cls_scores), tuple(bbox_preds)

    def deform_sampling(self, feat: Tensor, offset: Tensor) -> Tensor:
        """Sampling the feature x according to offset.

        Args:
            feat (Tensor): Feature
            offset (Tensor): Spatial offset for feature sampling
        """
        # it is an equivalent implementation of bilinear interpolation
        b, c, h, w = feat.shape
        weight = feat.new_ones(c, 1, 1, 1)
        y = deform_conv2d(feat, offset, weight, 1, 0, 1, c, c)
        return y

    def anchor_center(self, anchors: Tensor) -> Tensor:
        """Get anchor centers from anchors.

        Args:
            anchors (Tensor): Anchor list with shape (N, 4), "xyxy" format.

        Returns:
            Tensor: Anchor centers with shape (N, 2), "xy" format.
        """
        anchors_cx = (anchors[:, 2] + anchors[:, 0]) / 2
        anchors_cy = (anchors[:, 3] + anchors[:, 1]) / 2
        return torch.stack([anchors_cx, anchors_cy], dim=-1)

    def loss_by_feat_single(self, anchors: Tensor, cls_score: Tensor,
                            bbox_pred: Tensor, labels: Tensor,
                            label_weights: Tensor, bbox_targets: Tensor,
                            alignment_metrics: Tensor,
                            stride: Tuple[int, int]) -> dict:
        """Calculate the loss of a single scale level based on the features
        extracted by the detection head.

        Args:
            anchors (Tensor): Box reference for each scale level with shape
                (N, num_total_anchors, 4).
            cls_score (Tensor): Box scores for each scale level
                Has shape (N, num_anchors * num_classes, H, W).
            bbox_pred (Tensor): Decoded bboxes for each scale
                level with shape (N, num_anchors * 4, H, W).
            labels (Tensor): Labels of each anchors with shape
                (N, num_total_anchors).
            label_weights (Tensor): Label weights of each anchor with shape
                (N, num_total_anchors).
            bbox_targets (Tensor): BBox regression targets of each anchor with
                shape (N, num_total_anchors, 4).
            alignment_metrics (Tensor): Alignment metrics with shape
                (N, num_total_anchors).
            stride (Tuple[int, int]): Downsample stride of the feature map.

        Returns:
            dict[str, Tensor]: A dictionary of loss components.
        """
        assert stride[0] == stride[1], 'h stride is not equal to w stride!'
        anchors = anchors.reshape(-1, 4)
        cls_score = cls_score.permute(0, 2, 3, 1).reshape(
            -1, self.cls_out_channels).contiguous()
        bbox_pred = bbox_pred.permute(0, 2, 3, 1).reshape(-1, 4)
        bbox_targets = bbox_targets.reshape(-1, 4)
        labels = labels.reshape(-1)
        alignment_metrics = alignment_metrics.reshape(-1)
        label_weights = label_weights.reshape(-1)
        targets = labels if self.epoch < self.initial_epoch else (
            labels, alignment_metrics)
        cls_loss_func = self.initial_loss_cls \
            if self.epoch < self.initial_epoch else self.loss_cls

        loss_cls = cls_loss_func(
            cls_score, targets, label_weights, avg_factor=1.0)

        # FG cat_id: [0, num_classes -1], BG cat_id: num_classes
        bg_class_ind = self.num_classes
        pos_inds = ((labels >= 0)
                    & (labels < bg_class_ind)).nonzero().squeeze(1)

        if len(pos_inds) > 0:
            pos_bbox_targets = bbox_targets[pos_inds]
            pos_bbox_pred = bbox_pred[pos_inds]
            pos_anchors = anchors[pos_inds]

            pos_decode_bbox_pred = pos_bbox_pred
            pos_decode_bbox_targets = pos_bbox_targets / stride[0]

            # regression loss
            pos_bbox_weight = self.centerness_target(
                pos_anchors, pos_bbox_targets
            ) if self.epoch < self.initial_epoch else alignment_metrics[
                pos_inds]

            loss_bbox = self.loss_bbox(
                pos_decode_bbox_pred,
                pos_decode_bbox_targets,
                weight=pos_bbox_weight,
                avg_factor=1.0)
        else:
            loss_bbox = bbox_pred.sum() * 0
            pos_bbox_weight = bbox_targets.new_tensor(0.)

        return loss_cls, loss_bbox, alignment_metrics.sum(
        ), pos_bbox_weight.sum()

    def loss_by_feat(
            self,
            cls_scores: List[Tensor],
            bbox_preds: List[Tensor],
            batch_gt_instances: InstanceList,
            batch_img_metas: List[dict],
            batch_gt_instances_ignore: OptInstanceList = None) -> dict:
        """Calculate the loss based on the features extracted by the detection
        head.

        Args:
            cls_scores (list[Tensor]): Box scores for each scale level
                Has shape (N, num_anchors * num_classes, H, W)
            bbox_preds (list[Tensor]): Decoded box for each scale
                level with shape (N, num_anchors * 4, H, W) in
                [tl_x, tl_y, br_x, br_y] format.
            batch_gt_instances (list[:obj:`InstanceData`]): Batch of
                gt_instance.  It usually includes ``bboxes`` and ``labels``
                attributes.
            batch_img_metas (list[dict]): Meta information of each image, e.g.,
                image size, scaling factor, etc.
            batch_gt_instances_ignore (list[:obj:`InstanceData`], Optional):
                Batch of gt_instances_ignore. It includes ``bboxes`` attribute
                data that is ignored during training and testing.
                Defaults to None.

        Returns:
            dict[str, Tensor]: A dictionary of loss components.
        """
        num_imgs = len(batch_img_metas)
        featmap_sizes = [featmap.size()[-2:] for featmap in cls_scores]
        assert len(featmap_sizes) == self.prior_generator.num_levels

        device = cls_scores[0].device
        anchor_list, valid_flag_list = self.get_anchors(
            featmap_sizes, batch_img_metas, device=device)

        flatten_cls_scores = torch.cat([
            cls_score.permute(0, 2, 3, 1).reshape(num_imgs, -1,
                                                  self.cls_out_channels)
            for cls_score in cls_scores
        ], 1)
        flatten_bbox_preds = torch.cat([
            bbox_pred.permute(0, 2, 3, 1).reshape(num_imgs, -1, 4) * stride[0]
            for bbox_pred, stride in zip(bbox_preds,
                                         self.prior_generator.strides)
        ], 1)

        cls_reg_targets = self.get_targets(
            flatten_cls_scores,
            flatten_bbox_preds,
            anchor_list,
            valid_flag_list,
            batch_gt_instances,
            batch_img_metas,
            batch_gt_instances_ignore=batch_gt_instances_ignore)
        (anchor_list, labels_list, label_weights_list, bbox_targets_list,
         alignment_metrics_list) = cls_reg_targets

        losses_cls, losses_bbox, \
            cls_avg_factors, bbox_avg_factors = multi_apply(
                self.loss_by_feat_single,
                anchor_list,
                cls_scores,
                bbox_preds,
                labels_list,
                label_weights_list,
                bbox_targets_list,
                alignment_metrics_list,
                self.prior_generator.strides)

        cls_avg_factor = reduce_mean(sum(cls_avg_factors)).clamp_(min=1).item()
        losses_cls = list(map(lambda x: x / cls_avg_factor, losses_cls))

        bbox_avg_factor = reduce_mean(
            sum(bbox_avg_factors)).clamp_(min=1).item()
        losses_bbox = list(map(lambda x: x / bbox_avg_factor, losses_bbox))
        return dict(loss_cls=losses_cls, loss_bbox=losses_bbox)

    def _predict_by_feat_single(self,
                                cls_score_list: List[Tensor],
                                bbox_pred_list: List[Tensor],
                                score_factor_list: List[Tensor],
                                mlvl_priors: List[Tensor],
                                img_meta: dict,
                                cfg: Optional[ConfigDict] = None,
                                rescale: bool = False,
                                with_nms: bool = True) -> InstanceData:
        """Transform a single image's features extracted from the head into
        bbox results.

        Args:
            cls_score_list (list[Tensor]): Box scores from all scale
                levels of a single image, each item has shape
                (num_priors * num_classes, H, W).
            bbox_pred_list (list[Tensor]): Box energies / deltas from
                all scale levels of a single image, each item has shape
                (num_priors * 4, H, W).
            score_factor_list (list[Tensor]): Score factor from all scale
                levels of a single image, each item has shape
                (num_priors * 1, H, W).
            mlvl_priors (list[Tensor]): Each element in the list is
                the priors of a single level in feature pyramid. In all
                anchor-based methods, it has shape (num_priors, 4). In
                all anchor-free methods, it has shape (num_priors, 2)
                when `with_stride=True`, otherwise it still has shape
                (num_priors, 4).
            img_meta (dict): Image meta info.
            cfg (:obj:`ConfigDict`, optional): Test / postprocessing
                configuration, if None, test_cfg would be used.
            rescale (bool): If True, return boxes in original image space.
                Defaults to False.
            with_nms (bool): If True, do nms before return boxes.
                Defaults to True.

        Returns:
            tuple[Tensor]: Results of detected bboxes and labels. If with_nms
                is False and mlvl_score_factor is None, return mlvl_bboxes and
                mlvl_scores, else return mlvl_bboxes, mlvl_scores and
                mlvl_score_factor. Usually with_nms is False is used for aug
                test. If with_nms is True, then return the following format

                - det_bboxes (Tensor): Predicted bboxes with shape \
                    [num_bboxes, 5], where the first 4 columns are bounding \
                    box positions (tl_x, tl_y, br_x, br_y) and the 5-th \
                    column are scores between 0 and 1.
                - det_labels (Tensor): Predicted labels of the corresponding \
                    box with shape [num_bboxes].
        """

        cfg = self.test_cfg if cfg is None else cfg
        nms_pre = cfg.get('nms_pre', -1)

        mlvl_bboxes = []
        mlvl_scores = []
        mlvl_labels = []
        for cls_score, bbox_pred, priors, stride in zip(
                cls_score_list, bbox_pred_list, mlvl_priors,
                self.prior_generator.strides):
            assert cls_score.size()[-2:] == bbox_pred.size()[-2:]

            bbox_pred = bbox_pred.permute(1, 2, 0).reshape(-1, 4) * stride[0]
            scores = cls_score.permute(1, 2,
                                       0).reshape(-1, self.cls_out_channels)

            # After https://github.com/open-mmlab/mmdetection/pull/6268/,
            # this operation keeps fewer bboxes under the same `nms_pre`.
            # There is no difference in performance for most models. If you
            # find a slight drop in performance, you can set a larger
            # `nms_pre` than before.
            results = filter_scores_and_topk(
                scores, cfg.score_thr, nms_pre,
                dict(bbox_pred=bbox_pred, priors=priors))
            scores, labels, keep_idxs, filtered_results = results

            bboxes = filtered_results['bbox_pred']

            mlvl_bboxes.append(bboxes)
            mlvl_scores.append(scores)
            mlvl_labels.append(labels)

        results = InstanceData()
        results.bboxes = torch.cat(mlvl_bboxes)
        results.scores = torch.cat(mlvl_scores)
        results.labels = torch.cat(mlvl_labels)

        return self._bbox_post_process(
            results=results,
            cfg=cfg,
            rescale=rescale,
            with_nms=with_nms,
            img_meta=img_meta)

    def get_targets(self,
                    cls_scores: List[List[Tensor]],
                    bbox_preds: List[List[Tensor]],
                    anchor_list: List[List[Tensor]],
                    valid_flag_list: List[List[Tensor]],
                    batch_gt_instances: InstanceList,
                    batch_img_metas: List[dict],
                    batch_gt_instances_ignore: OptInstanceList = None,
                    unmap_outputs: bool = True) -> tuple:
        """Compute regression and classification targets for anchors in
        multiple images.

        Args:
            cls_scores (list[list[Tensor]]): Classification predictions of
                images, a 3D-Tensor with shape [num_imgs, num_priors,
                num_classes].
            bbox_preds (list[list[Tensor]]): Decoded bboxes predictions of one
                image, a 3D-Tensor with shape [num_imgs, num_priors, 4] in
                [tl_x, tl_y, br_x, br_y] format.
            anchor_list (list[list[Tensor]]): Multi level anchors of each
                image. The outer list indicates images, and the inner list
                corresponds to feature levels of the image. Each element of
                the inner list is a tensor of shape (num_anchors, 4).
            valid_flag_list (list[list[Tensor]]): Multi level valid flags of
                each image. The outer list indicates images, and the inner list
                corresponds to feature levels of the image. Each element of
                the inner list is a tensor of shape (num_anchors, )
            batch_gt_instances (list[:obj:`InstanceData`]): Batch of
                gt_instance.  It usually includes ``bboxes`` and ``labels``
                attributes.
            batch_img_metas (list[dict]): Meta information of each image, e.g.,
                image size, scaling factor, etc.
            batch_gt_instances_ignore (list[:obj:`InstanceData`], Optional):
                Batch of gt_instances_ignore. It includes ``bboxes`` attribute
                data that is ignored during training and testing.
                Defaults to None.
            unmap_outputs (bool): Whether to map outputs back to the original
                set of anchors.

        Returns:
            tuple: a tuple containing learning targets.

                - anchors_list (list[list[Tensor]]): Anchors of each level.
                - labels_list (list[Tensor]): Labels of each level.
                - label_weights_list (list[Tensor]): Label weights of each
                  level.
                - bbox_targets_list (list[Tensor]): BBox targets of each level.
                - norm_alignment_metrics_list (list[Tensor]): Normalized
                  alignment metrics of each level.
        """
        num_imgs = len(batch_img_metas)
        assert len(anchor_list) == len(valid_flag_list) == num_imgs

        # anchor number of multi levels
        num_level_anchors = [anchors.size(0) for anchors in anchor_list[0]]
        num_level_anchors_list = [num_level_anchors] * num_imgs

        # concat all level anchors and flags to a single tensor
        for i in range(num_imgs):
            assert len(anchor_list[i]) == len(valid_flag_list[i])
            anchor_list[i] = torch.cat(anchor_list[i])
            valid_flag_list[i] = torch.cat(valid_flag_list[i])

        # compute targets for each image
        if batch_gt_instances_ignore is None:
            batch_gt_instances_ignore = [None] * num_imgs
        # anchor_list: list(b * [-1, 4])

        # get epoch information from message hub
        message_hub = MessageHub.get_current_instance()
        self.epoch = message_hub.get_info('epoch')

        if self.epoch < self.initial_epoch:
            (all_anchors, all_labels, all_label_weights, all_bbox_targets,
             all_bbox_weights, pos_inds_list, neg_inds_list,
             sampling_result) = multi_apply(
                 super()._get_targets_single,
                 anchor_list,
                 valid_flag_list,
                 num_level_anchors_list,
                 batch_gt_instances,
                 batch_img_metas,
                 batch_gt_instances_ignore,
                 unmap_outputs=unmap_outputs)
            all_assign_metrics = [
                weight[..., 0] for weight in all_bbox_weights
            ]
        else:
            (all_anchors, all_labels, all_label_weights, all_bbox_targets,
             all_assign_metrics) = multi_apply(
                 self._get_targets_single,
                 cls_scores,
                 bbox_preds,
                 anchor_list,
                 valid_flag_list,
                 batch_gt_instances,
                 batch_img_metas,
                 batch_gt_instances_ignore,
                 unmap_outputs=unmap_outputs)

        # split targets to a list w.r.t. multiple levels
        anchors_list = images_to_levels(all_anchors, num_level_anchors)
        labels_list = images_to_levels(all_labels, num_level_anchors)
        label_weights_list = images_to_levels(all_label_weights,
                                              num_level_anchors)
        bbox_targets_list = images_to_levels(all_bbox_targets,
                                             num_level_anchors)
        norm_alignment_metrics_list = images_to_levels(all_assign_metrics,
                                                       num_level_anchors)

        return (anchors_list, labels_list, label_weights_list,
                bbox_targets_list, norm_alignment_metrics_list)

    def _get_targets_single(self,
                            cls_scores: Tensor,
                            bbox_preds: Tensor,
                            flat_anchors: Tensor,
                            valid_flags: Tensor,
                            gt_instances: InstanceData,
                            img_meta: dict,
                            gt_instances_ignore: Optional[InstanceData] = None,
                            unmap_outputs: bool = True) -> tuple:
        """Compute regression, classification targets for anchors in a single
        image.

        Args:
            cls_scores (Tensor): Box scores for each image.
            bbox_preds (Tensor): Box energies / deltas for each image.
            flat_anchors (Tensor): Multi-level anchors of the image, which are
                concatenated into a single tensor of shape (num_anchors ,4)
            valid_flags (Tensor): Multi level valid flags of the image,
                which are concatenated into a single tensor of
                    shape (num_anchors,).
            gt_instances (:obj:`InstanceData`): Ground truth of instance
                annotations. It usually includes ``bboxes`` and ``labels``
                attributes.
            img_meta (dict): Meta information for current image.
            gt_instances_ignore (:obj:`InstanceData`, optional): Instances
                to be ignored during training. It includes ``bboxes`` attribute
                data that is ignored during training and testing.
                Defaults to None.
            unmap_outputs (bool): Whether to map outputs back to the original
                set of anchors.

        Returns:
            tuple: N is the number of total anchors in the image.
                anchors (Tensor): All anchors in the image with shape (N, 4).
                labels (Tensor): Labels of all anchors in the image with shape
                    (N,).
                label_weights (Tensor): Label weights of all anchor in the
                    image with shape (N,).
                bbox_targets (Tensor): BBox targets of all anchors in the
                    image with shape (N, 4).
                norm_alignment_metrics (Tensor): Normalized alignment metrics
                    of all priors in the image with shape (N,).
        """
        inside_flags = anchor_inside_flags(flat_anchors, valid_flags,
                                           img_meta['img_shape'][:2],
                                           self.train_cfg['allowed_border'])
        if not inside_flags.any():
            raise ValueError(
                'There is no valid anchor inside the image boundary. Please '
                'check the image size and anchor sizes, or set '
                '``allowed_border`` to -1 to skip the condition.')
        # assign gt and sample anchors
        anchors = flat_anchors[inside_flags, :]
        pred_instances = InstanceData(
            priors=anchors,
            scores=cls_scores[inside_flags, :],
            bboxes=bbox_preds[inside_flags, :])
        assign_result = self.alignment_assigner.assign(pred_instances,
                                                       gt_instances,
                                                       gt_instances_ignore,
                                                       self.alpha, self.beta)
        assign_ious = assign_result.max_overlaps
        assign_metrics = assign_result.assign_metrics

        sampling_result = self.sampler.sample(assign_result, pred_instances,
                                              gt_instances)

        num_valid_anchors = anchors.shape[0]
        bbox_targets = torch.zeros_like(anchors)
        labels = anchors.new_full((num_valid_anchors, ),
                                  self.num_classes,
                                  dtype=torch.long)
        label_weights = anchors.new_zeros(num_valid_anchors, dtype=torch.float)
        norm_alignment_metrics = anchors.new_zeros(
            num_valid_anchors, dtype=torch.float)

        pos_inds = sampling_result.pos_inds
        neg_inds = sampling_result.neg_inds
        if len(pos_inds) > 0:
            # point-based
            pos_bbox_targets = sampling_result.pos_gt_bboxes
            bbox_targets[pos_inds, :] = pos_bbox_targets

            labels[pos_inds] = sampling_result.pos_gt_labels
            if self.train_cfg['pos_weight'] <= 0:
                label_weights[pos_inds] = 1.0
            else:
                label_weights[pos_inds] = self.train_cfg['pos_weight']
        if len(neg_inds) > 0:
            label_weights[neg_inds] = 1.0

        class_assigned_gt_inds = torch.unique(
            sampling_result.pos_assigned_gt_inds)
        for gt_inds in class_assigned_gt_inds:
            gt_class_inds = pos_inds[sampling_result.pos_assigned_gt_inds ==
                                     gt_inds]
            pos_alignment_metrics = assign_metrics[gt_class_inds]
            pos_ious = assign_ious[gt_class_inds]
            pos_norm_alignment_metrics = pos_alignment_metrics / (
                pos_alignment_metrics.max() + 10e-8) * pos_ious.max()
            norm_alignment_metrics[gt_class_inds] = pos_norm_alignment_metrics

        # map up to original set of anchors
        if unmap_outputs:
            num_total_anchors = flat_anchors.size(0)
            anchors = unmap(anchors, num_total_anchors, inside_flags)
            labels = unmap(
                labels, num_total_anchors, inside_flags, fill=self.num_classes)
            label_weights = unmap(label_weights, num_total_anchors,
                                  inside_flags)
            bbox_targets = unmap(bbox_targets, num_total_anchors, inside_flags)
            norm_alignment_metrics = unmap(norm_alignment_metrics,
                                           num_total_anchors, inside_flags)
        return (anchors, labels, label_weights, bbox_targets,
                norm_alignment_metrics)