-
Notifications
You must be signed in to change notification settings - Fork 2.9k
/
dark_hrnet_w32_256x192.yml
141 lines (124 loc) · 3.01 KB
/
dark_hrnet_w32_256x192.yml
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
use_gpu: true
log_iter: 5
save_dir: output
snapshot_epoch: 10
weights: output/hrnet_w32_256x192/model_final
epoch: 210
num_joints: &num_joints 17
pixel_std: &pixel_std 200
metric: KeyPointTopDownCOCOEval
num_classes: 1
train_height: &train_height 256
train_width: &train_width 192
trainsize: &trainsize [*train_width, *train_height]
hmsize: &hmsize [48, 64]
flip_perm: &flip_perm [[1, 2], [3, 4], [5, 6], [7, 8], [9, 10], [11, 12], [13, 14], [15, 16]]
#####model
architecture: TopDownHRNet
pretrain_weights: https://paddledet.bj.bcebos.com/models/pretrained/Trunc_HRNet_W32_C_pretrained.pdparams
TopDownHRNet:
backbone: HRNet
post_process: HRNetPostProcess
flip_perm: *flip_perm
num_joints: *num_joints
width: &width 32
loss: KeyPointMSELoss
HRNet:
width: *width
freeze_at: -1
freeze_norm: false
return_idx: [0]
KeyPointMSELoss:
use_target_weight: true
#####optimizer
LearningRate:
base_lr: 0.0005
schedulers:
- !PiecewiseDecay
milestones: [170, 200]
gamma: 0.1
- !LinearWarmup
start_factor: 0.001
steps: 1000
OptimizerBuilder:
optimizer:
type: Adam
regularizer:
factor: 0.0
type: L2
#####data
TrainDataset:
!KeypointTopDownCocoDataset
image_dir: train2017
anno_path: annotations/person_keypoints_train2017.json
dataset_dir: dataset/coco
num_joints: *num_joints
trainsize: *trainsize
pixel_std: *pixel_std
use_gt_bbox: True
EvalDataset:
!KeypointTopDownCocoDataset
image_dir: val2017
anno_path: annotations/person_keypoints_val2017.json
dataset_dir: dataset/coco
bbox_file: bbox.json
num_joints: *num_joints
trainsize: *trainsize
pixel_std: *pixel_std
use_gt_bbox: True
image_thre: 0.0
TestDataset:
!ImageFolder
anno_path: dataset/coco/keypoint_imagelist.txt
worker_num: 2
global_mean: &global_mean [0.485, 0.456, 0.406]
global_std: &global_std [0.229, 0.224, 0.225]
TrainReader:
sample_transforms:
- RandomFlipHalfBodyTransform:
scale: 0.5
rot: 40
num_joints_half_body: 8
prob_half_body: 0.3
pixel_std: *pixel_std
trainsize: *trainsize
upper_body_ids: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
flip_pairs: *flip_perm
- TopDownAffine:
trainsize: *trainsize
- ToHeatmapsTopDown_DARK:
hmsize: *hmsize
sigma: 2
batch_transforms:
- NormalizeImage:
mean: *global_mean
std: *global_std
is_scale: true
- Permute: {}
batch_size: 64
shuffle: true
drop_last: false
EvalReader:
sample_transforms:
- TopDownAffine:
trainsize: *trainsize
batch_transforms:
- NormalizeImage:
mean: *global_mean
std: *global_std
is_scale: true
- Permute: {}
batch_size: 16
TestReader:
inputs_def:
image_shape: [3, *train_height, *train_width]
sample_transforms:
- Decode: {}
- TopDownEvalAffine:
trainsize: *trainsize
- NormalizeImage:
mean: *global_mean
std: *global_std
is_scale: true
- Permute: {}
batch_size: 1