fce_icdar15.yaml

system:
  mode: 0 # 0 for graph mode, 1 for pynative mode in MindSpore
  distribute: False
  amp_level: 'O0'
  seed: 42
  # log_interval: 10
  val_while_train: True
  drop_overflow_update: False

model:
  resume: False
  type: det
  transform: null
  backbone:
    name: det_resnet50
    pretrained: True
  neck:
    name: FCEFPN
    out_channels: 256
  head:
    name: FCEHead
    fourier_degree: 5

postprocess:
  name: FCEPostprocess
  box_type: 'quad'
  alpha: 1.2
  beta: 1.0
  scales: [ 8, 16, 32 ]

metric:
  name: DetMetric
  main_indicator: f-score

loss:
  name: FCELoss
  fourier_degree: 5
  num_sample: 50

scheduler:
  scheduler: polynomial_decay
  lr: 0.002
  min_lr: 0.0000002
  num_epochs: 1500
  warmup_epochs: 0
optimizer:
  opt: momentum
  weight_decay: 5.0e-4
  momentum: 0.9
  filter_bias_and_bn: False



train:
  ckpt_save_dir: './tmp_det'
  dataset_sink_mode: True
  ema: True
  dataset:
    type: DetDataset
    dataset_root: /data/ocr_datasets
    data_dir: ic15/det/train/ch4_training_images
    label_file: ic15/det/train/det_gt.txt
    sample_ratio: 1.0
    transform_pipeline:
      - DecodeImage:
          img_mode: RGB
          to_float32: False
      - DetLabelEncode:
      - DetResize:
          target_size: [ 800, 800 ]
          keep_ratio: True
          padding: True
      - RandomScale:
          scale_range: [ 0.75,2.5 ]
          p: 1.0
      - RandomHorizontalFlip:
          p: 0.5
      - RandomRotate:
          degrees: [ -10, 10 ]
          expand_canvas: False
          p: 1.0
      - RandomCropWithBBox:
          max_tries: 10
          min_crop_ratio: 0.1
          crop_size: [ 640, 640 ]
          p: 1.0
      - ValidatePolygons:
      - RandomColorAdjust:
          brightness: 0.1255  # 32.0 / 255
          saturation: 0.5
      - NormalizeImage:
          bgr_to_rgb: False
          is_hwc: True
          mean: imagenet
          std: imagenet
      - FCENetTargets:
          fourier_degree: 5
          level_proportion_range: [ [ 0, 0.25 ], [ 0.2, 0.65 ], [ 0.55, 1.0 ] ]
      - ToCHWImage:
    #  the order of the dataloader list, matching the network input and the input labels for the loss function, and optional data for debug/visualize
    output_columns: [ 'image', 'p3_maps','p4_maps','p5_maps' ]
    #    output_columns: ['image'] # for debug op performance
    net_input_column_index: [ 0 ] # input indices for network forward func in output_columns
    label_column_index: [ 1, 2, 3 ] # input indices marked as label

  loader:
    shuffle: True
    batch_size: 8
    drop_remainder: True
    num_workers: 8

eval:
  ckpt_load_path: './best.ckpt'
  dataset_sink_mode: False
  dataset:
    type: DetDataset
    dataset_root: /data/ocr_datasets
    data_dir: ic15/det/test/ch4_test_images
    label_file: ic15/det/test/det_gt.txt
    sample_ratio: 1.0
    shuffle: False
    transform_pipeline:
      - DecodeImage:
          img_mode: RGB
          to_float32: False
      - DetLabelEncode:
      - DetResize:
          target_size: [ 800, 800 ]
          keep_ratio: True
          padding: True
      - NormalizeImage:
          bgr_to_rgb: False
          is_hwc: True
          mean: imagenet
          std: imagenet
      - ToCHWImage:
    #  the order of the dataloader list, matching the network input and the labels for evaluation
    output_columns: [ 'image', 'polys', 'ignore_tags','img_path','texts', 'shape_list' ]
    net_input_column_index: [ 0 ] # input indices for network forward func in output_columns
    label_column_index: [ 1, 2,3,4 ] # input indices marked as label
  #    num_keys_of_labels: 2 # num labels

  loader:
    shuffle: False
    batch_size: 1 # TODO: due to dynamic shape of polygons (num of boxes varies), BS has to be 1
    drop_remainder: False
    num_workers: 1