paddlers/tutorials/train/object_detection/ppyolo.py

import sys

sys.path.append("/ssd2/pengjuncai/PaddleRS")

import paddlers as pdrs
from paddlers import transforms as T

train_transforms = T.Compose([
    T.MixupImage(mixup_epoch=-1), T.RandomDistort(),
    T.RandomExpand(im_padding_value=[123.675, 116.28, 103.53]), T.RandomCrop(),
    T.RandomHorizontalFlip(), T.BatchRandomResize(
        target_sizes=[320, 352, 384, 416, 448, 480, 512, 544, 576, 608],
        interp='RANDOM'), T.Normalize(
            mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])

eval_transforms = T.Compose([
    T.Resize(
        target_size=608, interp='CUBIC'), T.Normalize(
            mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])


train_dataset = pdrs.datasets.VOCDetection(
    data_dir='insect_det',
    file_list='insect_det/train_list.txt',
    label_list='insect_det/labels.txt',
    transforms=train_transforms,
    shuffle=True)

eval_dataset = pdrs.datasets.VOCDetection(
    data_dir='insect_det',
    file_list='insect_det/val_list.txt',
    label_list='insect_det/labels.txt',
    transforms=eval_transforms,
    shuffle=False)


num_classes = len(train_dataset.labels)
model = pdrs.tasks.det.PPYOLO(num_classes=num_classes, backbone='ResNet50_vd_dcn')

model.train(
    num_epochs=200,
    train_dataset=train_dataset,
    train_batch_size=8,
    eval_dataset=eval_dataset,
    pretrain_weights='COCO',
    learning_rate=0.005 / 12,
    warmup_steps=500,
    warmup_start_lr=0.0,
    save_interval_epochs=5,
    lr_decay_epochs=[85, 135],
    save_dir='output/ppyolo_r50vd_dcn',
    use_vdl=True)