PaddleX/paddlex/repo_apis/PaddleDetection_api/configs/YOLOX-M.yaml

use_gpu: true
use_xpu: false
use_mlu: false
use_npu: false
log_iter: 100
save_dir: output
snapshot_epoch: 10
print_flops: false
print_params: false


metric: COCO
num_classes: 80

TrainDataset:
  name: COCODataSet
  image_dir: train2017
  anno_path: annotations/instances_train2017.json
  dataset_dir: dataset/coco
  data_fields: ['image', 'gt_bbox', 'gt_class', 'is_crowd']

EvalDataset:
  name: COCODataSet
  image_dir: val2017
  anno_path: annotations/instances_val2017.json
  dataset_dir: dataset/coco
  allow_empty: true

TestDataset:
  name: ImageFolder
  anno_path: annotations/instances_val2017.json # also support txt (like VOC's label_list.txt)
  dataset_dir: dataset/coco # if set, anno_path will be 'dataset_dir/anno_path'


worker_num: 4
TrainReader:
  sample_transforms:
    - Decode: {}
    - Mosaic:
        prob: 1.0
        input_dim: [640, 640]
        degrees: [-10, 10]
        scale: [0.1, 2.0]
        shear: [-2, 2]
        translate: [-0.1, 0.1]
        enable_mixup: True
        mixup_prob: 1.0
        mixup_scale: [0.5, 1.5]
    - AugmentHSV: {is_bgr: False, hgain: 5, sgain: 30, vgain: 30}
    - PadResize: {target_size: 640}
    - RandomFlip: {}
  batch_transforms:
    - Permute: {}
  batch_size: 8
  shuffle: True
  drop_last: True
  collate_batch: False
  mosaic_epoch: 285

EvalReader:
  sample_transforms:
    - Decode: {}
    - Resize: {target_size: [640, 640], keep_ratio: True, interp: 1}
    - Pad: {size: [640, 640], fill_value: [114., 114., 114.]}
    - Permute: {}
  batch_size: 4

TestReader:
  inputs_def:
    image_shape: [3, 640, 640]
  sample_transforms:
    - Decode: {}
    - Resize: {target_size: [640, 640], keep_ratio: True, interp: 1}
    - Pad: {size: [640, 640], fill_value: [114., 114., 114.]}
    - Permute: {}
  batch_size: 1


architecture: YOLOX
norm_type: sync_bn
use_ema: True
ema_decay: 0.9999
ema_decay_type: "exponential"
act: silu
find_unused_parameters: True

depth_mult: 0.67
width_mult: 0.75

YOLOX:
  backbone: CSPDarkNet
  neck: YOLOCSPPAN
  head: YOLOXHead
  size_stride: 32
  size_range: [15, 25] # multi-scale range [480*480 ~ 800*800]

CSPDarkNet:
  arch: "X"
  return_idx: [2, 3, 4]
  depthwise: False

YOLOCSPPAN:
  depthwise: False

YOLOXHead:
  l1_epoch: 285
  depthwise: False
  loss_weight: {cls: 1.0, obj: 1.0, iou: 5.0, l1: 1.0}
  assigner:
    name: SimOTAAssigner
    candidate_topk: 10
    use_vfl: False
  nms:
    name: MultiClassNMS
    nms_top_k: 10000
    keep_top_k: 1000
    score_threshold: 0.001
    nms_threshold: 0.65
    # For speed while keep high mAP, you can modify 'nms_top_k' to 1000 and 'keep_top_k' to 100, the mAP will drop about 0.1%.
    # For high speed demo, you can modify 'score_threshold' to 0.25 and 'nms_threshold' to 0.45, but the mAP will drop a lot.


epoch: 300

LearningRate:
  base_lr: 0.01
  schedulers:
  - !CosineDecay
    max_epochs: 300
    min_lr_ratio: 0.05
    last_plateau_epochs: 15
  - !ExpWarmup
    epochs: 5

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


# 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.
  fuse_conv_bn: False