13 KiB
13 KiB
识别结果
参数基本采用默认参数，都训练 10轮 。
PP-OCRv5_server (极好)
识别效果：极好 ，基本都是别出来了，并且正确。
PP-OCRv5_server_det.yml
Global:
model_name: PP-OCRv5_server_det # To use static model for inference.
debug: false
use_gpu: true
epoch_num: &epoch_num 10
log_smooth_window: 20
print_batch_step: 10
save_model_dir: ./output-v5server/PP-OCRv5_server_det
save_epoch_step: 200
eval_batch_step:
- 0
- 200
cal_metric_during_train: false
checkpoints:
pretrained_model: ./pretrain_models/PP-OCRv5_server_det_pretrained.pdparams
save_inference_dir: null
use_visualdl: false
infer_img: doc/imgs_en/img_10.jpg
save_res_path: ./checkpoints/det_db/predicts_db.txt
distributed: false
Architecture:
model_type: det
algorithm: DB
Transform: null
Backbone:
name: PPHGNetV2_B4
det: True
Neck:
name: LKPAN
out_channels: 256
intracl: true
Head:
name: PFHeadLocal
k: 50
mode: "large"
Loss:
name: DBLoss
balance_loss: true
main_loss_type: DiceLoss
alpha: 5
beta: 10
ohem_ratio: 3
Optimizer:
name: Adam
beta1: 0.9
beta2: 0.999
lr:
name: Cosine
learning_rate: 0.001 #(8*8c)
warmup_epoch: 2
regularizer:
name: L2
factor: 1e-6
PostProcess:
name: DBPostProcess
thresh: 0.3
box_thresh: 0.6
max_candidates: 1000
unclip_ratio: 1.5
Metric:
name: DetMetric
main_indicator: hmean
Train:
dataset:
name: SimpleDataSet
data_dir: ./data/input0/images #图片文件夹路径
label_file_list:
- ./det_train_label.txt #标签路径
ratio_list: [1.0]
transforms:
- DecodeImage:
img_mode: BGR
channel_first: false
- DetLabelEncode: null
- CopyPaste: null
- IaaAugment:
augmenter_args:
- type: Fliplr
args:
p: 0.5
- type: Affine
args:
rotate:
- -10
- 10
- type: Resize
args:
size:
- 0.5
- 3
- EastRandomCropData:
size:
- 640
- 640
max_tries: 50
keep_ratio: true
- MakeBorderMap:
shrink_ratio: 0.4
thresh_min: 0.3
thresh_max: 0.7
total_epoch: *epoch_num
- MakeShrinkMap:
shrink_ratio: 0.4
min_text_size: 8
total_epoch: *epoch_num
- NormalizeImage:
scale: 1./255.
mean:
- 0.485
- 0.456
- 0.406
std:
- 0.229
- 0.224
- 0.225
order: hwc
- ToCHWImage: null
- KeepKeys:
keep_keys:
- image
- threshold_map
- threshold_mask
- shrink_map
- shrink_mask
loader:
shuffle: true
drop_last: false
batch_size_per_card: 10
num_workers: 1
Eval:
dataset:
name: SimpleDataSet
data_dir: ./data/input0/images
label_file_list:
- ./det_eval_label.txt
transforms:
- DecodeImage:
img_mode: BGR
channel_first: false
- DetLabelEncode: null
- DetResizeForTest:
- NormalizeImage:
scale: 1./255.
mean:
- 0.485
- 0.456
- 0.406
std:
- 0.229
- 0.224
- 0.225
order: hwc
- ToCHWImage: null
- KeepKeys:
keep_keys:
- image
- shape
- polys
- ignore_tags
loader:
shuffle: false
drop_last: false
batch_size_per_card: 1
num_workers: 1
profiler_options: null
PP-OCRv5_server_rec.yml
Global:
model_name: PP-OCRv5_server_rec # To use static model for inference.
debug: false
use_gpu: true
epoch_num: 10
log_smooth_window: 20
print_batch_step: 10
save_model_dir: ./output-v5server/PP-OCRv5_server_rec
save_epoch_step: 200
eval_batch_step: [0, 200]
cal_metric_during_train: true
calc_epoch_interval: 1
pretrained_model: ./pretrain_models/PP-OCRv5_server_rec_pretrained.pdparams
checkpoints:
save_inference_dir:
use_visualdl: false
infer_img: doc/imgs_words/ch/word_1.jpg
character_dict_path: ./ppocr/utils/dict/ppocrv5_dict.txt
max_text_length: &max_text_length 25
infer_mode: false
use_space_char: true
distributed: true
save_res_path: ./output-v5server/rec/predicts_ppocrv5.txt
d2s_train_image_shape: [3, 48, 320]
Optimizer:
name: Adam
beta1: 0.9
beta2: 0.999
lr:
name: Cosine
learning_rate: 0.0005
warmup_epoch: 1
regularizer:
name: L2
factor: 3.0e-05
Architecture:
model_type: rec
algorithm: SVTR_HGNet
Transform:
Backbone:
name: PPHGNetV2_B4
text_rec: True
Head:
name: MultiHead
head_list:
- CTCHead:
Neck:
name: svtr
dims: 120
depth: 2
hidden_dims: 120
kernel_size: [1, 3]
use_guide: True
Head:
fc_decay: 0.00001
- NRTRHead:
nrtr_dim: 384
max_text_length: *max_text_length
Loss:
name: MultiLoss
loss_config_list:
- CTCLoss:
- NRTRLoss:
PostProcess:
name: CTCLabelDecode
Metric:
name: RecMetric
main_indicator: acc
Train:
dataset:
name: MultiScaleDataSet
ds_width: false
data_dir: ./RecTrainData/
ext_op_transform_idx: 1
label_file_list:
- ./rec_train_label.txt
transforms:
- DecodeImage:
img_mode: BGR
channel_first: false
- RecAug:
- MultiLabelEncode:
gtc_encode: NRTRLabelEncode
- KeepKeys:
keep_keys:
- image
- label_ctc
- label_gtc
- length
- valid_ratio
sampler:
name: MultiScaleSampler
scales: [[320, 32], [320, 48], [320, 64]]
first_bs: &bs 12
fix_bs: false
divided_factor: [8, 16] # w, h
is_training: True
loader:
shuffle: true
batch_size_per_card: 10
drop_last: true
num_workers: 1
Eval:
dataset:
name: SimpleDataSet
data_dir: ./RecEvalData/
label_file_list:
- ./rec_eval_label.txt
transforms:
- DecodeImage:
img_mode: BGR
channel_first: false
- MultiLabelEncode:
gtc_encode: NRTRLabelEncode
- RecResizeImg:
image_shape: [3, 48, 320]
- KeepKeys:
keep_keys:
- image
- label_ctc
- label_gtc
- length
- valid_ratio
loader:
shuffle: false
drop_last: false
batch_size_per_card: 1
num_workers: 1
PP-OCRv5_mobile
PP-OCRv5_mobile_det.yml
Global:
model_name: PP-OCRv5_mobile_det # To use static model for inference.
debug: false
use_gpu: true
epoch_num: &epoch_num 100
log_smooth_window: 20
print_batch_step: 100
save_model_dir: ./output-v5mobile/PP-OCRv5_mobile_det
save_epoch_step: 200
eval_batch_step:
- 0
- 200
cal_metric_during_train: false
checkpoints:
pretrained_model: https://paddleocr.bj.bcebos.com/pretrained/PPLCNetV3_x0_75_ocr_det.pdparams
save_inference_dir: null
use_visualdl: false
infer_img: doc/imgs_en/img_10.jpg
save_res_path: ./checkpoints/det_db/predicts_db.txt
d2s_train_image_shape: [3, 640, 640]
distributed: true
Architecture:
model_type: det
algorithm: DB
Transform: null
Backbone:
name: PPLCNetV3
scale: 0.75
det: True
Neck:
name: RSEFPN
out_channels: 96
shortcut: True
Head:
name: DBHead
k: 50
fix_nan: True
Loss:
name: DBLoss
balance_loss: true
main_loss_type: DiceLoss
alpha: 5
beta: 10
ohem_ratio: 3
Optimizer:
name: Adam
beta1: 0.9
beta2: 0.999
lr:
name: Cosine
learning_rate: 0.001 #(8*8c)
warmup_epoch: 2
regularizer:
name: L2
factor: 5.0e-05
PostProcess:
name: DBPostProcess
thresh: 0.3
box_thresh: 0.6
max_candidates: 1000
unclip_ratio: 1.5
Metric:
name: DetMetric
main_indicator: hmean
Train:
dataset:
name: SimpleDataSet
data_dir: ./data/input0/images #图片文件夹路径
label_file_list:
- ./det_train_label.txt #标签路径
ratio_list: [1.0]
transforms:
- DecodeImage:
img_mode: BGR
channel_first: false
- DetLabelEncode: null
- CopyPaste: null
- IaaAugment:
augmenter_args:
- type: Fliplr
args:
p: 0.5
- type: Affine
args:
rotate:
- -10
- 10
- type: Resize
args:
size:
- 0.5
- 3
- EastRandomCropData:
size:
- 640
- 640
max_tries: 50
keep_ratio: true
- MakeBorderMap:
shrink_ratio: 0.4
thresh_min: 0.3
thresh_max: 0.7
total_epoch: *epoch_num
- MakeShrinkMap:
shrink_ratio: 0.4
min_text_size: 8
total_epoch: *epoch_num
- NormalizeImage:
scale: 1./255.
mean:
- 0.485
- 0.456
- 0.406
std:
- 0.229
- 0.224
- 0.225
order: hwc
- ToCHWImage: null
- KeepKeys:
keep_keys:
- image
- threshold_map
- threshold_mask
- shrink_map
- shrink_mask
loader:
shuffle: true
drop_last: false
batch_size_per_card: 14
num_workers: 1
Eval:
dataset:
name: SimpleDataSet
data_dir: ./data/input0/images
label_file_list:
- ./det_eval_label.txt
transforms:
- DecodeImage:
img_mode: BGR
channel_first: false
- DetLabelEncode: null
- DetResizeForTest:
- NormalizeImage:
scale: 1./255.
mean:
- 0.485
- 0.456
- 0.406
std:
- 0.229
- 0.224
- 0.225
order: hwc
- ToCHWImage: null
- KeepKeys:
keep_keys:
- image
- shape
- polys
- ignore_tags
loader:
shuffle: false
drop_last: false
batch_size_per_card: 1
num_workers: 1
profiler_options: null
PP-OCRv5_mobile_rec.yml
Global:
model_name: PP-OCRv5_mobile_rec # To use static model for inference.
debug: false
use_gpu: true
epoch_num: 75
log_smooth_window: 20
print_batch_step: 10
save_model_dir: ./output/PP-OCRv5_mobile_rec
save_epoch_step: 10
eval_batch_step: [0, 2000]
cal_metric_during_train: true
pretrained_model: https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-OCRv5_mobile_rec_pretrained.pdparams
checkpoints:
save_inference_dir:
use_visualdl: false
infer_img: doc/imgs_words/ch/word_1.jpg
character_dict_path: ./ppocr/utils/dict/ppocrv5_dict.txt
max_text_length: &max_text_length 25
infer_mode: false
use_space_char: true
distributed: true
save_res_path: ./output/rec/predicts_ppocrv5.txt
d2s_train_image_shape: [3, 48, 320]
Optimizer:
name: Adam
beta1: 0.9
beta2: 0.999
lr:
name: Cosine
learning_rate: 0.0005
warmup_epoch: 5
regularizer:
name: L2
factor: 3.0e-05
Architecture:
model_type: rec
algorithm: SVTR_LCNet
Transform:
Backbone:
name: PPLCNetV3
scale: 0.95
Head:
name: MultiHead
head_list:
- CTCHead:
Neck:
name: svtr
dims: 120
depth: 2
hidden_dims: 120
kernel_size: [1, 3]
use_guide: True
Head:
fc_decay: 0.00001
- NRTRHead:
nrtr_dim: 384
max_text_length: *max_text_length
Loss:
name: MultiLoss
loss_config_list:
- CTCLoss:
- NRTRLoss:
PostProcess:
name: CTCLabelDecode
Metric:
name: RecMetric
main_indicator: acc
Train:
dataset:
name: MultiScaleDataSet
ds_width: false
data_dir: ./RecTrainData/
ext_op_transform_idx: 1
label_file_list:
- ./rec_train_label.txt
transforms:
- DecodeImage:
img_mode: BGR
channel_first: false
- RecConAug:
prob: 0.5
ext_data_num: 2
image_shape: [48, 320, 3]
max_text_length: *max_text_length
- RecAug:
- MultiLabelEncode:
gtc_encode: NRTRLabelEncode
- KeepKeys:
keep_keys:
- image
- label_ctc
- label_gtc
- length
- valid_ratio
sampler:
name: MultiScaleSampler
scales: [[320, 32], [320, 48], [320, 64]]
first_bs: &bs 128
fix_bs: false
divided_factor: [8, 16] # w, h
is_training: True
loader:
shuffle: true
batch_size_per_card: 14
drop_last: true
num_workers: 1
Eval:
dataset:
name: SimpleDataSet
data_dir: ./RecEvalData/
label_file_list:
- ./rec_eval_label.txt
transforms:
- DecodeImage:
img_mode: BGR
channel_first: false
- MultiLabelEncode:
gtc_encode: NRTRLabelEncode
- RecResizeImg:
image_shape: [3, 48, 320]
- KeepKeys:
keep_keys:
- image
- label_ctc
- label_gtc
- length
- valid_ratio
loader:
shuffle: false
drop_last: false
batch_size_per_card: 14
num_workers: 1