nielsr HF staff commited on
Commit
d44ba07
1 Parent(s): 2521adb

Upload Base-RCNN-FPN.yml

Browse files
Files changed (1) hide show
  1. Base-RCNN-FPN.yml +69 -0
Base-RCNN-FPN.yml ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ MODEL:
2
+ MASK_ON: True
3
+ META_ARCHITECTURE: "GeneralizedRCNN"
4
+ PIXEL_MEAN: [123.675, 116.280, 103.530]
5
+ PIXEL_STD: [58.395, 57.120, 57.375]
6
+ BACKBONE:
7
+ NAME: "build_vit_fpn_backbone"
8
+ VIT:
9
+ OUT_FEATURES: ["layer3", "layer5", "layer7", "layer11"]
10
+ DROP_PATH: 0.1
11
+ IMG_SIZE: [224,224]
12
+ POS_TYPE: "abs"
13
+ FPN:
14
+ IN_FEATURES: ["layer3", "layer5", "layer7", "layer11"]
15
+ ANCHOR_GENERATOR:
16
+ SIZES: [[32], [64], [128], [256], [512]] # One size for each in feature map
17
+ ASPECT_RATIOS: [[0.5, 1.0, 2.0]] # Three aspect ratios (same for all in feature maps)
18
+ RPN:
19
+ IN_FEATURES: ["p2", "p3", "p4", "p5", "p6"]
20
+ PRE_NMS_TOPK_TRAIN: 2000 # Per FPN level
21
+ PRE_NMS_TOPK_TEST: 1000 # Per FPN level
22
+ # Detectron1 uses 2000 proposals per-batch,
23
+ # (See "modeling/rpn/rpn_outputs.py" for details of this legacy issue)
24
+ # which is approximately 1000 proposals per-image since the default batch size for FPN is 2.
25
+ POST_NMS_TOPK_TRAIN: 1000
26
+ POST_NMS_TOPK_TEST: 1000
27
+ ROI_HEADS:
28
+ NAME: "StandardROIHeads"
29
+ IN_FEATURES: ["p2", "p3", "p4", "p5"]
30
+ NUM_CLASSES: 5
31
+ ROI_BOX_HEAD:
32
+ NAME: "FastRCNNConvFCHead"
33
+ NUM_FC: 2
34
+ POOLER_RESOLUTION: 7
35
+ ROI_MASK_HEAD:
36
+ NAME: "MaskRCNNConvUpsampleHead"
37
+ NUM_CONV: 4
38
+ POOLER_RESOLUTION: 14
39
+ DATASETS:
40
+ TRAIN: ("publaynet_train",)
41
+ TEST: ("publaynet_val",)
42
+ SOLVER:
43
+ LR_SCHEDULER_NAME: "WarmupCosineLR"
44
+ AMP:
45
+ ENABLED: True
46
+ OPTIMIZER: "ADAMW"
47
+ BACKBONE_MULTIPLIER: 1.0
48
+ CLIP_GRADIENTS:
49
+ ENABLED: True
50
+ CLIP_TYPE: "full_model"
51
+ CLIP_VALUE: 1.0
52
+ NORM_TYPE: 2.0
53
+ WARMUP_FACTOR: 0.01
54
+ BASE_LR: 0.0004
55
+ WEIGHT_DECAY: 0.05
56
+ IMS_PER_BATCH: 32
57
+ INPUT:
58
+ CROP:
59
+ ENABLED: True
60
+ TYPE: "absolute_range"
61
+ SIZE: (384, 600)
62
+ MIN_SIZE_TRAIN: (480, 512, 544, 576, 608, 640, 672, 704, 736, 768, 800)
63
+ FORMAT: "RGB"
64
+ DATALOADER:
65
+ FILTER_EMPTY_ANNOTATIONS: False
66
+ VERSION: 2
67
+ AUG:
68
+ DETR: True
69
+ SEED: 42