forked from microsoft/scene_graph_benchmark
-
Notifications
You must be signed in to change notification settings - Fork 1
/
rel_danfeiX_FPN50_reldn.yaml
90 lines (90 loc) · 2.68 KB
/
rel_danfeiX_FPN50_reldn.yaml
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
MODEL:
META_ARCHITECTURE: "SceneParser"
WEIGHT: "models/vgvrd/vgnm_usefpTrue_objctx0_edgectx2/model_final.pth"
USE_FREQ_PRIOR: False
FREQ_PRIOR: "visualgenome/label_danfeiX_clipped.freq_prior.npy"
RESNETS:
TRANS_FUNC: "BottleneckWithFixedBatchNorm"
BACKBONE_OUT_CHANNELS: 256
BACKBONE:
CONV_BODY: "R-50-FPN"
ATTRIBUTE_ON: False
RELATION_ON: True
RPN:
USE_FPN: True
ANCHOR_STRIDE: (4, 8, 16, 32, 64)
PRE_NMS_TOP_N_TRAIN: 2000
PRE_NMS_TOP_N_TEST: 1000
POST_NMS_TOP_N_TEST: 1000
FPN_POST_NMS_TOP_N_TEST: 1000
ROI_HEADS:
BATCH_SIZE_PER_IMAGE: 512
POSITIVE_FRACTION: 0.5
USE_FPN: True
SCORE_THRESH: 0.05 # 0.0001
DETECTIONS_PER_IMG: 100 # 600
MIN_DETECTIONS_PER_IMG: 1
ROI_BOX_HEAD:
NUM_CLASSES: 151
POOLER_RESOLUTION: 7
POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125)
POOLER_SAMPLING_RATIO: 2
FEATURE_EXTRACTOR: "FPN2MLPFeatureExtractor"
PREDICTOR: "FPNPredictor"
MLP_HEAD_DIM: 1024
ATTRIBUTE_ON: False
RELATION_ON: True
ROI_RELATION_HEAD:
DETECTOR_PRE_CALCULATED: False
FORCE_RELATIONS: False
ALGORITHM: "sg_reldn"
MODE: 'sgdet'
USE_BIAS: False
FILTER_NON_OVERLAP: True
UPDATE_BOX_REG: False
SHARE_CONV_BACKBONE: False
SHARE_BOX_FEATURE_EXTRACTOR: False
SEPERATE_SO_FEATURE_EXTRACTOR: True
NUM_CLASSES: 51 # 10
POOLER_RESOLUTION: 7
POOLER_SCALES: (0.25, 0.125, 0.0625, 0.03125)
POOLER_SAMPLING_RATIO: 2
FEATURE_EXTRACTOR: "FPN2MLPRelationFeatureExtractor"
PREDICTOR: "FPNRelationPredictor"
CONTRASTIVE_LOSS:
USE_FLAG: True
TRIPLETS_PER_IMG: 100
POSTPROCESS_METHOD: 'constrained' # previously it's unconstrained
INPUT:
MIN_SIZE_TRAIN: (600,)
MAX_SIZE_TRAIN: 1000
MIN_SIZE_TEST: 600
MAX_SIZE_TEST: 1000
PIXEL_MEAN: [103.530, 116.280, 123.675]
DATASETS:
FACTORY_TRAIN: ("VGTSVDataset",)
FACTORY_TEST: ("VGTSVDataset",)
TRAIN: ("visualgenome/train_danfeiX_relation_nm.yaml",)
TEST: ("visualgenome/test_danfeiX_relation.yaml",)
DATALOADER:
SIZE_DIVISIBILITY: 32
NUM_WORKERS: 4
SOLVER:
BASE_LR: 0.015
WEIGHT_DECAY: 0.0001
MAX_ITER: 40000
STEPS: (20000,30000)
IMS_PER_BATCH: 16
CHECKPOINT_PERIOD: 5000
TEST:
IMS_PER_BATCH: 8
SAVE_PREDICTIONS: False
SAVE_RESULTS_TO_TSV: True
TSV_SAVE_SUBSET: ['rect', 'class', 'conf', 'relations', 'relation_scores', 'relation_scores_all']
GATHER_ON_CPU: True
SKIP_PERFORMANCE_EVAL: False
OUTPUT_RELATION_FEATURE: False
OUTPUT_FEATURE: True # need this to output scores_all and box_all
OUTPUT_DIR: "./models/relation_danfeiX_FPN50/"
DATA_DIR: "./datasets"
DISTRIBUTED_BACKEND: 'gloo'