文本检测微调模型 转推理模型后识别效果不一致

[zhongmy20230403]

问题描述 / Problem Description

训练文本检测模型，训练后的checkpoints模型与转换后的inference模型识别效果不一致

运行环境 / Runtime Environment

• OS: Windows10
• Paddle: 2.4.2
• PaddleOCR: 2.7.0

复现代码 / Reproduction Code

训练：tools/train.py -c ./configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_student.yml
评估：tools/eval.py -c ./configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_student.yml -o Global.checkpoints=./output/1-fapiao_putong_ch_PP-OCR_V3_det/best_accuracy.pdparams
预测：tools/infer_det.py -c configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_student.yml -o Global.infer_img=./train_data/wenbenjiance_test Global.checkpoints=./output/1-fapiao_putong_ch_PP-OCR_V3_det/best_accuracy.pdparams
预测结果正常：

转推理模型：tools/export_model.py -c configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_student.yml -o Global.checkpoints=./output/1-fapiao_putong_ch_PP-OCR_V3_det/best_accuracy Global.save_inference_dir=./output/fapiao_putong_ch_PP-OCR_v3_det_inferer

推理模型预测：tools/infer/predict_det.py --image_dir=./train_data/wenbenjiance_test --det_model_dir=./output/fapiao_putong_ch_PP-OCR_v3_det_inferer

预测结果不正常：

配置文件ch_PP-OCRv3_det_student.yml
Global:
debug: false
use_gpu: true
epoch_num: 500
log_smooth_window: 20
print_batch_step: 10
save_model_dir: ./output/fapiao_putong_ch_PP-OCR_V3_det/
save_epoch_step: 100
eval_batch_step:

0
500
cal_metric_during_train: false
pretrained_model: ./pretrained_model/ch_PP-OCRv3_det_distill_train/student.pdparams
checkpoints: null
save_inference_dir: null
use_visualdl: false
infer_img: doc/imgs_en/img_10.jpg
save_res_path: ./output/fapiao_putong_checkpoints/det_db/predicts_db.txt
distributed: true
Architecture:
model_type: det
algorithm: DB
Transform:
Backbone:
name: MobileNetV3
scale: 0.5
model_name: large
disable_se: True
Neck:
name: RSEFPN
out_channels: 96
shortcut: True
Head:
name: DBHead
k: 50

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: 0.00005
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: ./train_data/wenbenjiance/imgs
label_file_list:

• ./train_data/wenbenjiance/train.txt
  ratio_list: [1.0]
  transforms:
• DecodeImage:
  img_mode: BGR
  channel_first: false
• DetLabelEncode: 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:
• 960
• 960
  max_tries: 50
  keep_ratio: true
• MakeBorderMap:
  shrink_ratio: 0.4
  thresh_min: 0.3
  thresh_max: 0.7
• MakeShrinkMap:
  shrink_ratio: 0.4
  min_text_size: 8
• 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: 4
  num_workers: 0
  Eval:
  dataset:
  name: SimpleDataSet
  data_dir: ./train_data/wenbenjiance/imgs
  label_file_list:
• ./train_data/wenbenjiance/val.txt
  transforms:
• DecodeImage:
  img_mode: BGR
  channel_first: false
• DetLabelEncode: null
• DetResizeForTest: null
• 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: 0

完整报错 / Complete Error Message

可能解决方案 / Possible solutions

附件 / Appendix

[UserWangZz]

转inference模型后，会有一套默认配置，需要确保这个配置和动态图推理时一致。