-
Notifications
You must be signed in to change notification settings - Fork 307
/
rec_r34_vd_tps_bilstm_att.yml
100 lines (91 loc) · 2.16 KB
/
rec_r34_vd_tps_bilstm_att.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
Global:
device: gpu
epoch_num: 72
log_smooth_window: 20
print_batch_step: 10
output_dir: ./output/rec/r34_vd_tps_bilstm_att
eval_epoch_step: [0, 1]
cal_metric_during_train: true
pretrained_model:
checkpoints:
use_tensorboard: false
infer_mode: false
infer_img: doc/imgs_words/en/word_1.png
character_dict_path: &character_dict_path
max_text_length: &max_text_length 25
use_space_char: &use_space_char False
Export:
export_dir:
export_shape: [ 1, 3, 32, 100 ]
dynamic_axes: [ 0, 2, 3 ]
Optimizer:
name: Adam
lr: 0.0005
weight_decay: 0.0
LRScheduler:
name: ConstLR
warmup_epoch: 0
Architecture:
model_type: rec
algorithm: RARE
Transform:
name: TPS
num_fiducial: 20
loc_lr: 0.1
model_name: large
Backbone:
name: ResNet
layers: 34
Neck:
name: SequenceEncoder
encoder_type: rnn
hidden_size: 256 #96
Head:
name: AttentionHead # AttentionHead
hidden_size: 256 #
l2_decay: 0.00001
Loss:
name: AttentionLoss
PostProcess:
name: AttnLabelDecode
character_dict_path: *character_dict_path
use_space_char: *use_space_char
Metric:
name: RecMetric
main_indicator: acc
Train:
dataset:
name: LMDBDataSet
data_dir: ./train_data/data_lmdb_release/training/
transforms:
- DecodeImage: # load image
img_mode: BGR
channel_first: False
- AttnLabelEncode: # Class handling label
- RecResizeImg:
image_shape: [3, 32, 100]
- KeepKeys:
keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
loader:
shuffle: True
batch_size_per_card: 256
drop_last: True
num_workers: 8
Eval:
dataset:
name: LMDBDataSet
data_dir: ./train_data/data_lmdb_release/validation/
transforms:
- DecodeImage: # load image
img_mode: BGR
channel_first: False
- AttnLabelEncode: # Class handling label
- RecResizeImg:
image_shape: [3, 32, 100]
- KeepKeys:
keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
loader:
shuffle: False
drop_last: False
batch_size_per_card: 256
num_workers: 8