File size: 2,778 Bytes
708dec4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
MODEL:
  META_ARCHITECTURE: "GeneralizedVLRCNN"
  WEIGHT: "swin_large_patch4_window12_384_22k.pth"
  RPN_ONLY: True
  RPN_ARCHITECTURE: "VLDYHEAD"

  BACKBONE:
    CONV_BODY: "SWINT-FPN-RETINANET"
    OUT_CHANNELS: 256

  SWINT:
    EMBED_DIM: 192
    DEPTHS: (2, 2, 18, 2)
    NUM_HEADS: (6, 12, 24, 48)
    WINDOW_SIZE: 12
    OUT_CHANNELS: (192, 384, 768, 1536)
    DROP_PATH_RATE: 0.4

  LANGUAGE_BACKBONE:
    FREEZE: False
    MODEL_TYPE: "bert-base-uncased" # "roberta-base", "clip"
    MASK_SPECIAL: False

  RPN:
    USE_FPN: True
    ANCHOR_SIZES: (64, 128, 256, 512, 1024)
    ANCHOR_STRIDE: (8, 16, 32, 64, 128)
    ASPECT_RATIOS: (1.0,)
    SCALES_PER_OCTAVE: 1

  DYHEAD:
    CHANNELS: 256
    NUM_CONVS: 8
    USE_GN: True
    USE_DYRELU: True
    USE_DFCONV: True
    USE_DYFUSE: True
    TOPK: 9 # topk for selecting candidate positive samples from each level
    SCORE_AGG: "MEAN"
    LOG_SCALE: 0.0

    USE_CHECKPOINT: True
    FUSE_CONFIG:
      USE_FUSED_FEATURES_DOT_PRODUCT: True
      EARLY_FUSE_ON: True
      TYPE: "MHA-B"
      USE_CLASSIFICATION_LOSS: False
      USE_TOKEN_LOSS: False
      USE_CONTRASTIVE_ALIGN_LOSS: False
      CONTRASTIVE_HIDDEN_DIM: 64
      USE_DOT_PRODUCT_TOKEN_LOSS: True
      USE_LAYER_SCALE: True
      CLAMP_MIN_FOR_UNDERFLOW: True
      CLAMP_MAX_FOR_OVERFLOW: True
      CLAMP_BERTATTN_MIN_FOR_UNDERFLOW: True
      CLAMP_BERTATTN_MAX_FOR_OVERFLOW: True
      CLAMP_DOT_PRODUCT: True

DATASETS:

  TRAIN: ("mixed_train_no_coco",) # Place holder dataset for now. To be updated in the next version
  TEST: ("coco_2017_val", )

  ONE_HOT: False
  FLICKR_COPY: 8 # 0.15 * 8 = ~1.2M
  MIXED_COPY: 4 # 0.6 * 4 = ~2.4M
  OBJECT365_COPY: 2 # 1.4 * 2 = ~2.8M
  VG_COPY: 3 # 0.4 * 3 = ~1.2M
  IN_COPY: 2 # 0.67 * 2 = ~1.33M
  OI_COPY: 1 # 2M * 1 = 2M

  DISABLE_SHUFFLE: False
  ADD_DET_PROMPT: False
  RANDOM_SAMPLE_NEG: 85
  CONTROL_PROB: (0.0, 0.0, 0.5, 0.0)
  FURTHER_SCREEN: True
  CAPTION_CONF: 0.5
  CAPTION_NMS: -1.0
  CAPTION_MIN_BOX: 1

  SEPARATION_TOKENS: ". "

  PACK_RANDOM_CAPTION_NUMBER: 20
  NO_RANDOM_PACK_PROBABILITY: 0.4
  RANDOM_PACK_PROB: 0.5
  CAPTION_FORMAT_VERSION: "v2"

INPUT:
  PIXEL_MEAN: [ 103.530, 116.280, 123.675 ]
  PIXEL_STD: [ 57.375, 57.120, 58.395 ]
  MIN_SIZE_TRAIN: 800
  MAX_SIZE_TRAIN: 1333
  MIN_SIZE_TEST: 800
  MAX_SIZE_TEST: 1333

AUGMENT:
  MULT_MIN_SIZE_TRAIN: (480,560,640,720,800)

DATALOADER:
  SIZE_DIVISIBILITY: 32

SOLVER:
  OPTIMIZER: ADAMW
  BASE_LR: 0.0001
  LANG_LR: 0.00001
  WEIGHT_DECAY: 0.01
  WEIGHT_DECAY_SCHEDULE: True
  STEPS: (0.67, 0.89)
  MAX_ITER: 1000000
  IMS_PER_BATCH: 64
  WARMUP_ITERS: 2000
  WARMUP_FACTOR: 0.001

  FIND_UNUSED_PARAMETERS: False

  CLIP_GRADIENTS:
    ENABLED: True
    CLIP_TYPE: "full_model"
    CLIP_VALUE: 1.0
    NORM_TYPE: 2.0