recognition_config.yml

Global:
  debug: false
  use_gpu: true
  epoch_num: 50
  log_smooth_window: 20
  print_batch_step: 10
  save_model_dir: ./output/rec_ppocr_v4
  save_epoch_step: 10
  eval_batch_step:
  - 0
  - 2000
  cal_metric_during_train: true
  pretrained_model: null
  checkpoints: null
  save_inference_dir: null
  use_visualdl: false
  infer_img: doc/imgs_words/ch/word_1.jpg
  character_dict_path: /kaggle/working/Fine-Tuning-OCR-Model-with-PaddleOCR/PaddleOCR/ppocr/utils/en_dict.txt
  infer_mode: false
  use_space_char: true
  distributed: true
  save_res_path: ./output/rec/predicts_ppocrv3.txt
  max_text_length: 25
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: null
  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: 1.0e-05
    - NRTRHead:
        nrtr_dim: 384
        max_text_length: 25
Loss:
  name: MultiLoss
  loss_config_list:
  - CTCLoss: null
  - NRTRLoss: null
PostProcess:
  name: CTCLabelDecode
Metric:
  name: RecMetric
  main_indicator: acc
  ignore_space: false
  
Train:
  dataset:
    name: SimpleDataSet
    data_dir: "/kaggle/working/Fine-Tuning-OCR-Model-with-PaddleOCR/dataset"
    ext_op_transform_idx: 1
    label_file_list:
    - dataset/recognition_train.txt
    transforms:
    - DecodeImage:
        img_mode: BGR
        channel_first: false
    - RecConAug: # Concat Augmentation (jika digunakan)
        prob: 0.5
        ext_data_num: 2
        image_shape: [48, 320, 3]
        max_text_length: 25
    - RecAug: # <<< Andalkan operator ini untuk augmentasi blur, noise, light/color
              # Jangan tambahkan GaussianBlur, SaltAndPepperNoise, ColorJitter secara terpisah
    # --- HAPUS Augmentasi Tambahan yang Menyebabkan Error ---
    # - GaussianBlur:
    #     prob: 0.5
    # - SaltAndPepperNoise:
    #     prob: 0.5
    # - ColorJitter:
    #     prob: 0.5
    # --- Akhir Bagian yang Dihapus ---
    - MultiLabelEncode:
    - RecResizeImg:
        image_shape: [3, 48, 320]
    - KeepKeys:
        keep_keys:
        - image
        - label_ctc
        - label_sar
        - length
        - valid_ratio
  loader:
    shuffle: true
    batch_size_per_card: 128
    drop_last: true
    num_workers: 4

Eval:
  dataset:
    name: SimpleDataSet
    data_dir: "/kaggle/working/Fine-Tuning-OCR-Model-with-PaddleOCR/dataset"
    label_file_list:
    - dataset/recognition_test.txt
    transforms:
    - DecodeImage:
        img_mode: BGR
        channel_first: false
    - MultiLabelEncode:
    - RecResizeImg:
        image_shape: [3, 48, 320]
    - KeepKeys:
        keep_keys:
        - image
        - label_ctc
        - label_sar
        - length
        - valid_ratio
  loader:
    shuffle: false
    drop_last: false
    batch_size_per_card: 128
    num_workers: 4