PaddleDetection——YOLO系列模型参数配置教程

内容参考自：README_cn.md · PaddlePaddle/PaddleDetection - 码云 - 开源中国 (gitee.com)

说明：用于帮助自己理解参数，后续会更新，可能有错误的地方，请不吝赐教。

YOLO系列模型参数配置教程

标签： 模型参数配置

++++++++++++++++++++++++++ppyolo_r18vd.yml++++++++++++++++++++++++++++++++architecture: YOLOv3 #模型的名称use_gpu: true #是否使用GPU

max_iters: 15500     #最大的迭代次数
log_iter: 10　　　　　#输出指定区间的平均结果，如10次的平均结果，也即打印log的间隔
save_dir: output
snapshot_iter: 1550
metric: COCO
pretrain_weights: https://paddle-imagenet-models-name.bj.bcebos.com/ResNet18_vd_pretrained.tar
weights: output/ppyolo_r18/model_final
num_classes: 9
use_fine_grained_loss: true
use_ema: true
ema_decay: 0.9998

YOLOv3:
  backbone: ResNet
  yolo_head: YOLOv3Head
  use_fine_grained_loss: true

ResNet:
  norm_type: sync_bn
  freeze_at: 0
  freeze_norm: false
  norm_decay: 0.
  depth: 18
  feature_maps: [4, 5]
  variant: d

YOLOv3Head:
  anchor_masks: [[3, 4, 5], [0, 1, 2]]
  anchors: [[10, 14], [23, 27], [37, 58],
            [81, 82], [135, 169], [344, 319]]
  norm_decay: 0.
  conv_block_num: 0
  scale_x_y: 1.05
  yolo_loss: YOLOv3Loss
  nms: MatrixNMS
  drop_block: true

YOLOv3Loss:
  ignore_thresh: 0.7
  scale_x_y: 1.05
  label_smooth: false
  use_fine_grained_loss: true
  iou_loss: IouLoss

IouLoss:
  loss_weight: 2.5
  max_height: 640
  max_width: 640

MatrixNMS:
    background_label: -1
    keep_top_k: 100
    normalized: false
    score_threshold: 0.01
    post_threshold: 0.01

LearningRate:
  base_lr: 0.004 #学习率决定了权值更新的速度，学习率大，更新的就快，但太快容易越过最优值，而学习率太小又更新的慢，效率低，一般学习率随着训练的进行不断更改，先高一点，然后慢慢降低
  schedulers:
  - !PiecewiseDecay
    gamma: 0.1
    milestones:#学习率变动因子：如迭代到10000次时，学习率衰减十倍，15000次迭代时，学习率又会在前一个学习率的基础上衰减十倍
    - 10000
    - 15000
  - !LinearWarmup
    start_factor: 0. 
    steps: 4000  #学习率变动步长

OptimizerBuilder:
  optimizer:
    momentum: 0.9 #动量，影响梯度下降到最优的速度，一般默认0.9
    type: Momentum
  regularizer:
    factor: 0.0005  #权重衰减正则系数，防止过拟合
    type: L2

_READER_: 'ppyolo_reader.yml'
TrainReader:
  inputs_def:
    fields: ['image', 'gt_bbox', 'gt_class', 'gt_score']
    num_max_boxes: 50
  dataset:
    !COCODataSet
      image_dir: images
      anno_path: annotations/val.json
      dataset_dir: /home/aistudio/data/data101204/NGdet_v4-concat
      with_background: false
  sample_transforms:
    - !DecodeImage
      to_rgb: True
      with_mixup: True
    - !MixupImage
      alpha: 1.5
      beta: 1.5
    - !ColorDistort {
    
    }
    - !RandomExpand
      fill_value: [123.675, 116.28, 103.53]
    - !RandomCrop {
    
    }
    - !RandomFlipImage
      is_normalized: false
    - !NormalizeBox {
    
    }
    - !PadBox
      num_max_boxes: 50
    - !BboxXYXY2XYWH {
    
    }
  batch_transforms:
  - !RandomShape
    sizes: [ 352, 384, 416, 448, 480, 512, 544, 576, 608,640]
  - !NormalizeImage
    mean: [0.485, 0.456, 0.406]
    std: [0.229, 0.224, 0.225]
    is_scale: True
    is_channel_first: false
  - !Permute
    to_bgr: false
    channel_first: True
  # Gt2YoloTarget is only used when use_fine_grained_loss set as true,
  # this operator will be deleted automatically if use_fine_grained_loss
  # is set as false
  - !Gt2YoloTarget
    anchor_masks: [[3, 4, 5], [0, 1, 2]]　　#anchors是可以事先通过cmd指令计算出来的，是和图片数量，width,height以及cluster(就是下面的num的值，即想要使用的anchors的数量)相关的预选框，可以手工挑选，也可以通过k-means算法从训练样本中学出
    anchors: [[10, 14], [23, 27], [37, 58],
              [81, 82], [135, 169], [344, 319]]
    downsample_ratios: [32, 16]
  batch_size: 8
  shuffle: true
  mixup_epoch: 100 #大于最大epoch，表示训练过程一直使用mixup数据增广
  drop_last: true
  worker_num: 8
  bufsize: 8
  use_process: true

标签： 模型参数配置
以ppyolo_r50vd_dcn_1x_coco.yml为例，这个模型由五个子配置文件组成：

• 数据配置文件coco_detection.yml

# 数据评估类型
metric: COCO
# 数据集的类别数
num_classes: 80

# TrainDataset
TrainDataset:
  !COCODataSet
    # 图像数据路径，相对 dataset_dir 路径，os.path.join(dataset_dir, image_dir)
    image_dir: train2017
    # 标注文件路径，相对 dataset_dir 路径，os.path.join(dataset_dir, anno_path)
    anno_path: annotations/instances_train2017.json
    # 数据文件夹
    dataset_dir: dataset/coco
    # data_fields
    data_fields: ['image', 'gt_bbox', 'gt_class', 'is_crowd']

EvalDataset:
  !COCODataSet
    # 图像数据路径，相对 dataset_dir 路径，os.path.join(dataset_dir, image_dir)
    image_dir: val2017
    # 标注文件路径，相对 dataset_dir 路径，os.path.join(dataset_dir, anno_path)
    anno_path: annotations/instances_val2017.json
    # 数据文件夹，os.path.join(dataset_dir, anno_path)
    dataset_dir: dataset/coco

TestDataset:
  !ImageFolder
    # 标注文件路径，相对 dataset_dir 路径
    anno_path: annotations/instances_val2017.json

• 优化器配置文件optimizer_1x.yml

# 总训练轮数
epoch: 405

# 学习率设置
LearningRate:
  # 默认为8卡训学习率
  base_lr: 0.01
  # 学习率调整策略
  schedulers:
  - !PiecewiseDecay
    gamma: 0.1
    # 学习率变化位置(轮数)
    milestones:
    - 243
    - 324
  # Warmup
  - !LinearWarmup
    start_factor: 0.
    steps: 4000

# 优化器
OptimizerBuilder:
  # 优化器
  optimizer:
    momentum: 0.9
    type: Momentum
  # 正则化
  regularizer:
    factor: 0.0005
    type: L2

• 数据读取配置文件ppyolo_reader.yml

# 每张GPU reader进程个数
worker_num: 2
# 训练数据
TrainReader:
  inputs_def:
    num_max_boxes: 50
  # 训练数据transforms
  sample_transforms:
    - Decode: {
    
    }
    - Mixup: {
    
    alpha: 1.5, beta: 1.5}
    - RandomDistort: {
    
    }
    - RandomExpand: {
    
    fill_value: [123.675, 116.28, 103.53]}
    - RandomCrop: {
    
    }
    - RandomFlip: {
    
    }
  # batch_transforms
  batch_transforms:
    - BatchRandomResize: {
    
    target_size: [320, 352, 384, 416, 448, 480, 512, 544, 576, 608], random_size: True, random_interp: True, keep_ratio: False}
    - NormalizeBox: {
    
    }
    - PadBox: {
    
    num_max_boxes: 50}
    - BboxXYXY2XYWH: {
    
    }
    - NormalizeImage: {
    
    mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
    - Permute: {
    
    }
    - Gt2YoloTarget: {
    
    anchor_masks: [[6, 7, 8], [3, 4, 5], [0, 1, 2]], anchors: [[10, 13], [16, 30], [33, 23], [30, 61], [62, 45], [59, 119], [116, 90], [156, 198], [373, 326]], downsample_ratios: [32, 16, 8]}
  # 训练时batch_size
  batch_size: 24
  # 读取数据是是否乱序
  shuffle: true
  # 是否丢弃最后不能完整组成batch的数据
  drop_last: true
  # mixup_epoch，大于最大epoch，表示训练过程一直使用mixup数据增广
  mixup_epoch: 25000
  # 是否通过共享内存进行数据读取加速，需要保证共享内存大小(如/dev/shm)满足大于1G
  use_shared_memory: true

# 评估数据
EvalReader:
  # 评估数据transforms
  sample_transforms:
    - Decode: {
    
    }
    - Resize: {
    
    target_size: [608, 608], keep_ratio: False, interp: 2}
    - NormalizeImage: {
    
    mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
    - Permute: {
    
    }
  # 评估时batch_size
  batch_size: 8
  # 是否丢弃没有标注的数据
  drop_empty: false

# 测试数据
TestReader:
  inputs_def:
    image_shape: [3, 608, 608]
  # 测试数据transforms
  sample_transforms:
    - Decode: {
    
    }
    - Resize: {
    
    target_size: [608, 608], keep_ratio: False, interp: 2}
    - NormalizeImage: {
    
    mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
    - Permute: {
    
    }
  # 测试时batch_size
  batch_size: 1

• 模型配置文件ppyolo_r50vd_dcn.yml

# 模型结构类型
architecture: YOLOv3
# 预训练模型地址
pretrain_weights: https://paddledet.bj.bcebos.com/models/pretrained/ResNet50_vd_ssld_pretrained.pdparams
# norm_type
norm_type: sync_bn
# 是否使用ema
use_ema: true
# ema_decay
ema_decay: 0.9998

# YOLOv3
YOLOv3:
  # backbone
  backbone: ResNet
  # neck
  neck: PPYOLOFPN
  # yolo_head
  yolo_head: YOLOv3Head
  # post_process
  post_process: BBoxPostProcess


# backbone
ResNet:
  # depth
  depth: 50
  # variant
  variant: d
  # return_idx, 0 represent res2
  return_idx: [1, 2, 3]
  # dcn_v2_stages
  dcn_v2_stages: [3]
  # freeze_at
  freeze_at: -1
  # freeze_norm
  freeze_norm: false
  # norm_decay
  norm_decay: 0.

# PPYOLOFPN
PPYOLOFPN:
  # 是否coord_conv
  coord_conv: true
  # 是否drop_block
  drop_block: true
  # block_size
  block_size: 3
  # keep_prob
  keep_prob: 0.9
  # 是否spp
  spp: true

# YOLOv3Head
YOLOv3Head:
  # anchors
  anchors: [[10, 13], [16, 30], [33, 23],
            [30, 61], [62, 45], [59, 119],
            [116, 90], [156, 198], [373, 326]]
  # anchor_masks
  anchor_masks: [[6, 7, 8], [3, 4, 5], [0, 1, 2]]
  # loss
  loss: YOLOv3Loss
  # 是否使用iou_aware
  iou_aware: true
  # iou_aware_factor
  iou_aware_factor: 0.4

# YOLOv3Loss
YOLOv3Loss:
  # ignore_thresh
  ignore_thresh: 0.7
  # downsample
  downsample: [32, 16, 8]
  # 是否label_smooth
  label_smooth: false
  # scale_x_y
  scale_x_y: 1.05
  # iou_loss
  iou_loss: IouLoss
  # iou_aware_loss
  iou_aware_loss: IouAwareLoss

# IouLoss
IouLoss:
  loss_weight: 2.5
  loss_square: true

# IouAwareLoss
IouAwareLoss:
  loss_weight: 1.0

# BBoxPostProcess
BBoxPostProcess:
  decode:
    name: YOLOBox
    conf_thresh: 0.01
    downsample_ratio: 32
    clip_bbox: true
    scale_x_y: 1.05
  # nms 配置
  nms:
    name: MatrixNMS
    keep_top_k: 100
    score_threshold: 0.01
    post_threshold: 0.01
    nms_top_k: -1
    background_label: -1

• 运行时置文件runtime.yml

# 是否使用gpu
use_gpu: true
# 日志打印间隔
log_iter: 20
# save_dir
save_dir: output
# 模型保存间隔时间
snapshot_epoch: 1