configs/rec/lpv/svtr_base_lpv_wo_glrm.yml

Global:
  device: gpu
  epoch_num: 20
  log_smooth_window: 20
  print_batch_step: 10
  output_dir: ./output/rec/u14m_filter/svtr_base_lpv_wo_glrm/
  save_epoch_step: 1
  # evaluation is run every 2000 iterations
  eval_batch_step: [0, 500]
  eval_epoch_step: [0, 1]
  cal_metric_during_train: True
  pretrained_model:
  checkpoints:
  use_tensorboard: false
  infer_img:
  # for data or label process
  character_dict_path: &character_dict_path ./tools/utils/EN_symbol_dict.txt # 96en
  # ./tools/utils/ppocr_keys_v1.txt  # ch
  max_text_length: &max_text_length 25
  use_space_char: &use_space_char False
  save_res_path: ./output/rec/u14m_filter/predicts_svtr_base_lpv_wo_glrm.txt
  use_amp: True
  grad_clip_val: 20

Optimizer:
  name: Adam
  lr: 0.0001 # for 4gpus bs128/gpu
  weight_decay: 0.0
  filter_bias_and_bn: False
  betas: [0.9, 0.99]

LRScheduler:
  name: MultiStepLR
  milestones: [12]
  gamma: 0.1

Architecture:
  model_type: rec
  algorithm: LPV
  in_channels: 3
  Transform:
  Encoder:
    name: SVTRNet
    img_size: [32, 128]
    out_char_num: 25
    out_channels: 256
    patch_merging: 'Conv'
    embed_dim: [128, 256, 384]
    depth: [6, 6, 6]
    num_heads: [4, 8, 12]
    mixer: ['Conv','Conv','Conv','Conv','Conv','Conv', 'Conv','Conv', 'Global','Global','Global','Global','Global','Global','Global','Global','Global','Global']
    local_mixer: [[5, 5], [5, 5], [5, 5]]
    sub_k: [[1, 1], [1, 1]]
    feature2d: True
    last_stage: False
    prenorm: True
  Decoder:
    name: LPVDecoder
    num_layer: 3
    max_len: *max_text_length
    use_mask: False
    dim_feedforward: 1536
    nhead: 12
    dropout: 0.1
    trans_layer: 3

Loss:
  name: LPVLoss

PostProcess:
  name: ARLabelDecode
  character_dict_path: *character_dict_path
  use_space_char: *use_space_char

Metric:
  name: RecMetric
  main_indicator: acc
  is_filter: True

Train:
  dataset:
    name: LMDBDataSet
    data_dir: ../Union14M-L-LMDB-Filtered
    transforms:
      - DecodeImagePIL: # load image
          img_mode: RGB
      - PARSeqAugPIL:
      - ARLabelEncode: # Class handling label
          character_dict_path: *character_dict_path
          use_space_char: *use_space_char
          max_text_length: *max_text_length
      - RecTVResize:
          image_shape: [32, 128]
          padding: False
      - KeepKeys:
          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
  loader:
    shuffle: True
    batch_size_per_card: 128
    drop_last: True
    num_workers: 4

Eval:
  dataset:
    name: LMDBDataSet
    data_dir: ../evaluation/
    transforms:
      - DecodeImagePIL: # load image
          img_mode: RGB
      - ARLabelEncode: # Class handling label
          character_dict_path: *character_dict_path
          use_space_char: *use_space_char
          max_text_length: *max_text_length
      - RecTVResize:
          image_shape: [32, 128]
          padding: False
      - KeepKeys:
          keep_keys: ['image', 'label', 'length'] # dataloader will return list in this order
  loader:
    shuffle: False
    drop_last: False
    batch_size_per_card: 128
    num_workers: 4