forked from yikang-li/FactorizableNet
-
Notifications
You must be signed in to change notification settings - Fork 1
/
eval_39588769.out
111 lines (110 loc) · 3.04 KB
/
eval_39588769.out
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
test
## args
{'MPS_iter': None,
'clip_gradient': True,
'dataset_option': None,
'dir_logs': None,
'dropout': None,
'epochs': None,
'eval_epochs': 1,
'evaluate': True,
'evaluate_object': False,
'infinite': False,
'iter_size': 1,
'learning_rate': None,
'loss_weight': True,
'model_name': None,
'nms': -1.0,
'optimize_MPS': False,
'optimizer': None,
'path_opt': 'options/models/VRD.yaml',
'pretrained_model': 'output/trained_models/Model-VRD.h5',
'print_freq': 1000,
'resume': None,
'rpn': None,
'save_all_from': None,
'seed': 1,
'start_epoch': 0,
'step_size': None,
'triplet_nms': 0.4,
'use_gt_boxes': False,
'use_normal_anchors': False,
'warm_iters': -1,
'workers': 1}
## options
{'data': {'batch_size': 1L,
'dataset': 'VRD',
'dataset_version': None,
'opts': 'options/data_VRD.yaml',
'use_region': False},
'logs': {'dir_logs': 'output/', 'model_name': 'FN_VRD'},
'model': {'MPS_iter': 1,
'activation': 'relu',
'arch': 'FN_v4s',
'bottleneck': True,
'cls_obj': 1.0,
'cls_pred': 2.0,
'dim_ho': 512,
'dim_hp': 512,
'dim_hr': 512,
'dim_mm': 256,
'dropout': 0.0,
'fusion': 'PI_v5',
'op': 'Sum',
'pool_size': 5,
'reg_obj': 0.5,
'rpn_opts': 'options/RPN/RPN_FN_VRD.yaml',
'use_bias': True,
'use_loss_weight': True},
'optim': {'clip_gradient': True,
'epochs': 15,
'lr': 0.01,
'lr_decay': 0.1,
'lr_decay_epoch': 3,
'momentum': 0.9,
'nesterov': True,
'optimizer': 0,
'weight_decay': 1e-05},
'opts': {'dir': 'data/VRD',
'test': {'MAX_SIZE': 1000, 'SCALES': [600]},
'train': {'MAX_SIZE': 1000,
'SCALES': [450, 500, 550, 600, 600, 650, 700, 750]}}}
Checkpoints are saved to: output/FN_VRD_1_iters_SGD
Loading training set and testing set... #################### MY DEBUG SECTION start ####################
len of test_set: 954
traverse test_set start
traverse test_set done
options['data']['dataset'] VRD
data_opts
{'dir': 'data/VRD',
'test': {'MAX_SIZE': 1000, 'SCALES': [600]},
'train': {'MAX_SIZE': 1000,
'SCALES': [450, 500, 550, 600, 600, 650, 700, 750]}}
dataset_option = None
use_region = False
#################### MY DEBUG SECTION end ####################
Done
hello from RPN.py init()
using k-means anchors: data/VRD/kmeans_anchors.json
1 MPS modules are used.
Bottleneck enabled.
Done.
vgg feature length: 26
rpn feature length: 6
HDN feature length: 37
MPS feature length: 20
Logs directory already exists in output/FN_VRD_1_iters_SGD. Erase?
Loading pretrained model: output/trained_models/Model-VRD.h5
Model has 32954137 parameters
========== Testing =======
eval done
0
hello
(<class 'torch.Tensor'>, <type 'list'>)
hello from fn_v4s forward_eval()
hello from RPN.py forward
done till #108 in RPN.py
done till #114 in RPN.py
done till #117 in RPN.py
parameters no problem
hello from proposal_layer() in RPN.py