DocumentAIKit/configs/smrt/ppyoloe/ppyoloe_crn_m_300e_renche_1024.yml

weights: output/ppyoloe_crn_m_300e_renche_1024/model_final
pretrain_weights: https://paddledet.bj.bcebos.com/models/ppyoloe_crn_m_300e_coco.pdparams
depth_mult: 0.67
width_mult: 0.75

worker_num: 4
eval_height: &eval_height 1024
eval_width: &eval_width 1024
eval_size: &eval_size [*eval_height, *eval_width]

metric: COCO
num_classes: 22

TrainDataset:
  !COCODataSet
    image_dir: train_images
    anno_path: train.json
    dataset_dir: /paddle/dataset/renche
    data_fields: ['image', 'gt_bbox', 'gt_class', 'is_crowd']

EvalDataset:
  !COCODataSet
    image_dir: train_images
    anno_path: test.json
    dataset_dir: /paddle/dataset/renche

TestDataset:
  !ImageFolder
    anno_path: test.json
    dataset_dir: /paddle/dataset/renche

epoch: 30
LearningRate:
  base_lr: 0.0005
  schedulers:
    - !CosineDecay
      max_epochs: 36
    - !LinearWarmup
      start_factor: 0.
      epochs: 3

TrainReader:
  sample_transforms:
    - Decode: {}
    - RandomFlip: {}
  batch_transforms:
    - BatchRandomResize: {target_size: [960, 992, 1024, 1056, 1088], random_size: True, random_interp: True, keep_ratio: False}
    - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
    - Permute: {}
    - PadGT: {}
  batch_size: 4
  shuffle: true
  drop_last: true
  use_shared_memory: true
  collate_batch: true

EvalReader:
  sample_transforms:
    - Decode: {}
    - Resize: {target_size: *eval_size, keep_ratio: False, interp: 2}
    - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
    - Permute: {}
  batch_size: 2

TestReader:
  inputs_def:
    image_shape: [3, *eval_height, *eval_width]
  sample_transforms:
    - Decode: {}
    - Resize: {target_size: *eval_size, keep_ratio: False, interp: 2}
    - NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
    - Permute: {}
  batch_size: 1

use_gpu: true
use_xpu: false
log_iter: 100
save_dir: output
snapshot_epoch: 5
print_flops: false

# Exporting the model
export:
  post_process: True  # Whether post-processing is included in the network when export model.
  nms: True           # Whether NMS is included in the network when export model.
  benchmark: False    # It is used to testing model performance, if set `True`, post-process and NMS will not be exported.

OptimizerBuilder:
  optimizer:
    momentum: 0.9
    type: Momentum
  regularizer:
    factor: 0.0005
    type: L2

architecture: YOLOv3
norm_type: sync_bn
use_ema: true
ema_decay: 0.9998

YOLOv3:
  backbone: CSPResNet
  neck: CustomCSPPAN
  yolo_head: PPYOLOEHead
  post_process: ~

CSPResNet:
  layers: [3, 6, 6, 3]
  channels: [64, 128, 256, 512, 1024]
  return_idx: [1, 2, 3]
  use_large_stem: True

CustomCSPPAN:
  out_channels: [768, 384, 192]
  stage_num: 1
  block_num: 3
  act: 'swish'
  spp: true

PPYOLOEHead:
  fpn_strides: [32, 16, 8]
  grid_cell_scale: 5.0
  grid_cell_offset: 0.5
  static_assigner_epoch: 100
  use_varifocal_loss: True
  loss_weight: {class: 1.0, iou: 2.5, dfl: 0.5}
  static_assigner:
    name: ATSSAssigner
    topk: 9
  assigner:
    name: TaskAlignedAssigner
    topk: 13
    alpha: 1.0
    beta: 6.0
  nms:
    name: MultiClassNMS
    nms_top_k: 1000
    keep_top_k: 100
    score_threshold: 0.01
    nms_threshold: 0.6