PaddleOCR/configs/e2e/e2e_r50_vd_pg.yml

Global:
  use_gpu: True
  epoch_num: 600
  log_smooth_window: 20
  print_batch_step: 10
  save_model_dir: ./output/pg_r50_vd_tt/
  save_epoch_step: 10
  # evaluation is run every 0 iterationss after the 1000th iteration
  eval_batch_step: [ 0, 1000 ]
  # 1. If pretrained_model is saved in static mode, such as classification pretrained model
  #    from static branch, load_static_weights must be set as True.
  # 2. If you want to finetune the pretrained models we provide in the docs,
  #    you should set load_static_weights as False.
  load_static_weights: True
  cal_metric_during_train: False
  pretrained_model:
  checkpoints:
  save_inference_dir:
  use_visualdl: False
  infer_img:
  save_res_path: ./output/pg_r50_vd_tt/predicts_pg.txt

Architecture:
  model_type: e2e
  algorithm: PGNet
  Transform:
  Backbone:
    name: ResNet
    layers: 50
  Neck:
    name: PGFPN
    model_name: large
  Head:
    name: PGHead
    model_name: large

Loss:
  name: PGLoss

Optimizer:
  name: Adam
  beta1: 0.9
  beta2: 0.999
  lr:
    learning_rate: 0.001
  regularizer:
    name: 'L2'
    factor: 0


PostProcess:
  name: PGPostProcess
  score_thresh: 0.8
  cover_thresh: 0.1
  nms_thresh: 0.2

Metric:
  name: E2EMetric
  Lexicon_Table: [ '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z' ]
  main_indicator: f_score_e2e

Train:
  dataset:
    name: PGDateSet
    label_file_list: [./train_data/total_text/train/]
    ratio_list: [1.0]
    data_format: icdar
    transforms:
      - DecodeImage: # load image
          img_mode: BGR
          channel_first: False
      - PGProcessTrain:
          batch_size: 14
          min_crop_size: 24
          min_text_size: 4
          max_text_size: 512
          Lexicon_Table: [ '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z' ]
      - KeepKeys:
          keep_keys: [ 'images', 'tcl_maps', 'tcl_label_maps', 'border_maps','direction_maps', 'training_masks', 'label_list', 'pos_list', 'pos_mask' ] # dataloader will return list in this order
  loader:
    shuffle: True
    drop_last: True
    batch_size_per_card: 14
    num_workers: 16

Eval:
  dataset:
    name: PGDataSet
    data_dir: ./train_data/
    label_file_list: [./train_data/total_text/test/]
    transforms:
      - DecodeImage: # load image
          img_mode: BGR
          channel_first: False
      - E2ELabelEncode:
          Lexicon_Table: [ '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X', 'Y', 'Z' ]
          max_len: 50
      - E2EResizeForTest:
          valid_set: totaltext
          max_side_len: 768
      - NormalizeImage:
          scale: 1./255.
          mean: [ 0.485, 0.456, 0.406 ]
          std: [ 0.229, 0.224, 0.225 ]
          order: 'hwc'
      - ToCHWImage:
      - KeepKeys:
          keep_keys: [ 'image', 'shape', 'polys', 'strs', 'tags' ]
  loader:
    shuffle: False
    drop_last: False
    batch_size_per_card: 1 # must be 1
    num_workers: 2