ppyolov2训练官方指定roadsign数据集loss不下降 #3032
Description
根据官方教程跑了ppyolov2模型，利用roadsign数据集训练，Loss训练到最后一直卡在10几不下降，AP不到0.1
单卡训练，学习率也缩小了10倍
BASE: [
'../datasets/coco_detection.yml',
'../runtime.yml',
'./base/ppyolov2_r50vd_dcn.yml',
'./base/optimizer_365e.yml',
'./base/ppyolov2_reader.yml',
]
snapshot_epoch: 8
weights: output/ppyolov2_r50vd_dcn_365e_coco/model_final
architecture: YOLOv3
pretrain_weights: https://paddledet.bj.bcebos.com/models/pretrained/ResNet50_vd_ssld_pretrained.pdparams
norm_type: sync_bn
use_ema: true
ema_decay: 0.9998
YOLOv3:
backbone: ResNet
neck: PPYOLOPAN
yolo_head: YOLOv3Head
post_process: BBoxPostProcess
ResNet:
depth: 50
variant: d
return_idx: [1, 2, 3]
dcn_v2_stages: [3]
freeze_at: -1
freeze_norm: false
norm_decay: 0.
PPYOLOPAN:
drop_block: true
block_size: 3
keep_prob: 0.9
spp: true
YOLOv3Head:
anchors: [[10, 13], [16, 30], [33, 23],
[30, 61], [62, 45], [59, 119],
[116, 90], [156, 198], [373, 326]]
anchor_masks: [[6, 7, 8], [3, 4, 5], [0, 1, 2]]
loss: YOLOv3Loss
iou_aware: true
iou_aware_factor: 0.5
YOLOv3Loss:
ignore_thresh: 0.7
downsample: [32, 16, 8]
label_smooth: false
scale_x_y: 1.05
iou_loss: IouLoss
iou_aware_loss: IouAwareLoss
IouLoss:
loss_weight: 2.5
loss_square: true
IouAwareLoss:
loss_weight: 1.0
BBoxPostProcess:
decode:
name: YOLOBox
conf_thresh: 0.01
downsample_ratio: 32
clip_bbox: true
scale_x_y: 1.05
nms:
name: MatrixNMS
keep_top_k: 100
score_threshold: 0.01
post_threshold: 0.01
nms_top_k: -1
background_label: -1
worker_num: 8
TrainReader:
inputs_def:
num_max_boxes: 100
sample_transforms:
- Decode: {}
- Mixup: {alpha: 1.5, beta: 1.5}
- RandomDistort: {}
- RandomExpand: {fill_value: [123.675, 116.28, 103.53]}
- RandomCrop: {}
- RandomFlip: {}
batch_transforms:
- BatchRandomResize: {target_size: [320, 352, 384, 416, 448, 480, 512, 544, 576, 608, 640, 672, 704, 736, 768], random_size: True, random_interp: True, keep_ratio: False}
- NormalizeBox: {}
- PadBox: {num_max_boxes: 100}
- 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]}
batch_size: 1
shuffle: true
drop_last: true
mixup_epoch: 25000
use_shared_memory: true
EvalReader:
sample_transforms:
- Decode: {}
- Resize: {target_size: [640, 640], 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: 8
drop_empty: false
epoch: 365
LearningRate:
base_lr: 0.0005
schedulers:
- !PiecewiseDecay
gamma: 0.1
milestones:- 243
- !LinearWarmup
start_factor: 0.
steps: 4000
OptimizerBuilder:
clip_grad_by_norm: 35.
optimizer:
momentum: 0.9
type: Momentum
regularizer:
factor: 0.0005
type: L2
TestReader:
inputs_def:
image_shape: [3, 640, 640]
sample_transforms:
- Decode: {}
- Resize: {target_size: [640, 640], 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
metric: COCO
num_classes: 4
TrainDataset:
!COCODataSet
image_dir: images
anno_path: annotations/train.json
dataset_dir: E:/Download/Dataset/roadsign_coco
data_fields: ['image', 'gt_bbox', 'gt_class', 'is_crowd']
EvalDataset:
!COCODataSet
image_dir: images
anno_path: annotations/valid.json
dataset_dir: E:/Download/Dataset/roadsign_coco
TestDataset:
!ImageFolder
anno_path: E:/Download/Dataset/roadsign_coco/annotations/valid.json