_base_ = './yolox_s_8x8_300e_coco.py'

# model settings
model = dict(
    backbone=dict(deepen_factor=0.33, widen_factor=0.375),
    neck=dict(in_channels=[96, 192, 384], out_channels=96),
    bbox_head=dict(in_channels=96, feat_channels=96))

# dataset settings
img_norm_cfg = dict(
    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)

img_scale = (640, 640)

train_pipeline = [
    dict(type='Mosaic', img_scale=img_scale, pad_val=114.0),
    dict(
        type='RandomAffine',
        scaling_ratio_range=(0.5, 1.5),
        border=(-img_scale[0] // 2, -img_scale[1] // 2)),
    dict(
        type='PhotoMetricDistortion',
        brightness_delta=32,
        contrast_range=(0.5, 1.5),
        saturation_range=(0.5, 1.5),
        hue_delta=18),
    dict(type='RandomFlip', flip_ratio=0.5),
    dict(type='Resize', keep_ratio=True),
    dict(type='Pad', pad_to_square=True, pad_val=114.0),
    dict(type='Normalize', **img_norm_cfg),
    dict(type='DefaultFormatBundle'),
    dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels'])
]

test_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(
        type='MultiScaleFlipAug',
        img_scale=(416, 416),
        flip=False,
        transforms=[
            dict(type='Resize', keep_ratio=True),
            dict(type='RandomFlip'),
            dict(type='Pad', size=(416, 416), pad_val=114.0),
            dict(type='Normalize', **img_norm_cfg),
            dict(type='DefaultFormatBundle'),
            dict(type='Collect', keys=['img'])
        ])
]

train_dataset = dict(pipeline=train_pipeline)

data = dict(
    train=train_dataset,
    val=dict(pipeline=test_pipeline),
    test=dict(pipeline=test_pipeline))

resume_from = None
interval = 10

# Execute in the order of insertion when the priority is the same.
# The smaller the value, the higher the priority
custom_hooks = [
    dict(type='YOLOXModeSwitchHook', num_last_epochs=15, priority=48),
    dict(
        type='SyncRandomSizeHook',
        ratio_range=(10, 20),
        img_scale=img_scale,
        priority=48),
    dict(
        type='SyncNormHook',
        num_last_epochs=15,
        interval=interval,
        priority=48),
    dict(type='ExpMomentumEMAHook', resume_from=resume_from, priority=49)
]
checkpoint_config = dict(interval=interval)
evaluation = dict(interval=interval, metric='bbox')