PaddleOCR/configs/rec/my_chinese_lite_large.yml

#Global:
#  use_gpu: True
#  epoch_num: 10
#  log_smooth_window: 20
#  print_batch_step: 10
#  save_model_dir: output/rec/my_rec_chinese_lite/
#  save_epoch_step: 5
#  # evaluation is run every 100000 iterations
#  eval_batch_step: [0, 100000]
#  # if pretrained_model is saved in static mode, load_static_weights must set to True
#  cal_metric_during_train: True
##  pretrained_model: output/rec/my_rec_chinese_lite/best_accuracy
#  pretrained_model:
#  checkpoints:
#  save_inference_dir:
#  use_visualdl: False
#  infer_img: doc/imgs_words_en/word_10.png
#  # for data or label process
#  character_dict_path: ppocr/utils/ppocr_keys_v1.txt
#  character_type: ch
#  max_text_length: 128
#  infer_mode: False
#  use_space_char: True
#
#
#Optimizer:
#  name: Adam
#  beta1: 0.9
#  beta2: 0.999
#  lr:
#    name: Cosine
#    learning_rate: 0.0003
#  regularizer:
#    name: 'L2'
#    factor: 0.00001
#
#Architecture:
#  model_type: rec
#  algorithm: CRNN
#  Transform:
#  Backbone:
#    name: MobileNetV3
#    scale: 0.5
#    model_name: large
#  Neck:
#    name: SequenceEncoder
#    encoder_type: rnn
#    hidden_size: 96
#  Head:
#    name: CTCHead
#    fc_decay: 0.00001
#
#Loss:
#  name: CTCLoss
#
#PostProcess:
#  name: CTCLabelDecode
#
#Metric:
#  name: RecMetric
#  main_indicator: acc
#
#Train:
#  dataset:
#    name: SimpleDataSet
#    data_dir: train_data/bidi_data/mix_data4/
#    label_file_list: ["./train_data/bidi_data/mix_data4/rec_gt_train.txt"]
#    transforms:
#      - DecodeImage: # load image
#          img_mode: BGR
#          channel_first: False
#      - CTCLabelEncode: # Class handling label
#      - RecResizeImg:
#          image_shape: [3, 32, 1000]
#      - KeepKeys:
#          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
#  loader:
#    shuffle: True
#    batch_size_per_card: 50
#    drop_last: True
#    num_workers: 0
#
#Eval:
#  dataset:
#    name: SimpleDataSet
#    data_dir: train_data/bidi_data/mix_data4/
#    label_file_list: ["./train_data/bidi_data/mix_data4/rec_gt_test.txt"]
#    transforms:
#      - DecodeImage: # load image
#          img_mode: BGR
#          channel_first: False
#      - CTCLabelEncode: # Class handling label
#      - RecResizeImg:
#          image_shape: [3, 32, 1000]
#      - KeepKeys:
#          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
#  loader:
#    shuffle: False
#    drop_last: False
#    batch_size_per_card: 50
#    num_workers: 0


Global:
  use_gpu: True
  epoch_num: 10
  log_smooth_window: 20
  print_batch_step: 20
  save_model_dir: output/rec/my_rec_chinese_lite/
  save_epoch_step: 5
  # evaluation is run every 100000 iterations
  eval_batch_step: [0, 100000]
  # if pretrained_model is saved in static mode, load_static_weights must set to True
  cal_metric_during_train: True
  # pretrained_model .pdmodel .pdiparams .pdiparams.info
  pretrained_model: output/rec/my_rec_chinese_lite/best_accuracy
#  pretrained_model:
  # checkpoints .pdparams .pdopt .states
#  checkpoints: output/rec/my_rec_chinese_lite/best_accuracy
  save_inference_dir:
  use_visualdl: False
  infer_img: doc/imgs_words_en/word_10.png
  # for data or label process
  character_dict_path: ppocr/utils/ppocr_keys_v1.txt
  character_type: ch
  max_text_length: 128
  infer_mode: False
  use_space_char: True

Optimizer:
  name: Adam
  beta1: 0.9
  beta2: 0.999
  lr:
    learning_rate: 0.0005
  regularizer:
    name: 'L2'
    factor: 0

Architecture:
  model_type: rec
  algorithm: CRNN
  Transform:
  Backbone:
    name: MobileNetV3
    scale: 0.5
    model_name: large
  Neck:
    name: SequenceEncoder
    encoder_type: rnn
    hidden_size: 96
  Head:
    name: CTCHead
    fc_decay: 0

Loss:
  name: CTCLoss

PostProcess:
  name: CTCLabelDecode

Metric:
  name: RecMetric
  main_indicator: acc

Train:
  dataset:
    name: SimpleDataSet
    data_dir: train_data/bidi_data/mix_data4/
    label_file_list: ["./train_data/bidi_data/mix_data4/rec_gt_train.txt"]
    transforms:
      - DecodeImage: # load image
          img_mode: BGR
          channel_first: False
      - CTCLabelEncode: # Class handling label
      - RecResizeImg:
          image_shape: [3, 32, 1000]
      - KeepKeys:
          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
  loader:
    shuffle: True
    batch_size_per_card: 80
    drop_last: True
    num_workers: 0

Eval:
  dataset:
    name: SimpleDataSet
    data_dir: train_data/bidi_data/mix_data4/
    label_file_list: ["./train_data/bidi_data/mix_data4/rec_gt_test.txt"]
    transforms:
      - DecodeImage: # load image
          img_mode: BGR
          channel_first: False
      - CTCLabelEncode: # Class handling label
      - RecResizeImg:
          image_shape: [3, 32, 1000]
      - KeepKeys:
          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
  loader:
    shuffle: False
    drop_last: False
    batch_size_per_card: 80
    num_workers: 0