vit-l_struct-token-sse_ade20k_640x640.py

_base_ = [
    '../_base_/struct_token.py',
    '../_base_/datasets/ade20k.py', '../_base_/default_runtime.py',
    '../_base_/schedules/schedule_160k.py'
]

norm_cfg = dict(type='LN', requires_grad=True)
model = dict(pretrained='pretrain/vit_large_p16_384.pth',
             backbone=dict(img_size=(640, 640),
                           embed_dims=1024,
                           num_layers=24,
                           num_heads=16,),
             decode_head=dict(type='SelfSliceExtracStructTokenHead',
                              image_h=640,
                              image_w=640,
                              in_channels=1024,
                              channels=1024,),
             test_cfg=dict(mode='slide', crop_size=(640, 640), stride=(426, 426)))

img_norm_cfg = dict(
    mean=[127.5, 127.5, 127.5], std=[127.5, 127.5, 127.5], to_rgb=True)
crop_size = (640, 640)
train_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(type='LoadAnnotations', reduce_zero_label=True),
    dict(type='Resize', img_scale=(2560, 640), ratio_range=(0.5, 2.0)),
    dict(type='RandomCrop', crop_size=crop_size, cat_max_ratio=0.75),
    dict(type='RandomFlip', prob=0.5),
    dict(type='PhotoMetricDistortion'),
    dict(type='Normalize', **img_norm_cfg),
    dict(type='Pad', size=crop_size, pad_val=0, seg_pad_val=255),
    dict(type='DefaultFormatBundle'),
    dict(type='Collect', keys=['img', 'gt_semantic_seg'])
]
test_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(
        type='MultiScaleFlipAug',
        img_scale=(2560, 640),
        # img_ratios=[0.5, 0.75, 1.0, 1.25, 1.5, 1.75],
        flip=False,
        transforms=[
            dict(type='Resize', keep_ratio=True),
            dict(type='RandomFlip'),
            dict(type='Normalize', **img_norm_cfg),
            dict(type='ImageToTensor', keys=['img']),
            dict(type='Collect', keys=['img'])
        ])
]
data = dict(
    # By default, models are trained on 8 GPUs with 2 images per GPU
    samples_per_gpu=2,
    train=dict(pipeline=train_pipeline),
    val=dict(pipeline=test_pipeline),
    test=dict(pipeline=test_pipeline))

# AdamW optimizer, no weight decay for position embedding & layer norm
# in backbone
optimizer = dict(
    _delete_=True,
    type='AdamW',
    lr=0.00002,
    betas=(0.9, 0.999),
    weight_decay=0.01,
    paramwise_cfg=dict(
        custom_keys={
            'pos_embed': dict(decay_mult=0.),
            'cls_token': dict(decay_mult=0.),
            'norm': dict(decay_mult=0.),
            'decode_head': dict(lr_mult=6.),
        }))

lr_config = dict(
    _delete_=True,
    policy='poly',
    warmup='linear',
    warmup_iters=1500,
    warmup_ratio=1e-6,
    power=1.0,
    min_lr=0.0,
    by_epoch=False)

checkpoint_config = dict(by_epoch=False, interval=80000)
evaluation = dict(interval=8000, metric='mIoU', pre_eval=True)