upernet_vheat_160k_ade20k_512x512_small.py

_base_ = [
    './configs/swin/swin-small-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py'
]

crop_size = (512, 512)
data_preprocessor = dict(size=crop_size)

model = dict(
    data_preprocessor=data_preprocessor,
    backbone=dict(
        type='MMSEG_VHEAT',
        img_size=512,
        post_norm=True,
        layer_scale=1.e-5, 
        drop_path_rate=0.4,
        depths=(2, 2, 18, 2),
        dims=96,
        out_indices=(0, 1, 2, 3),
        pretrained="path/to/pth",
    ),
    decode_head=dict(num_classes=150, in_channels=[96, 192, 384, 768]),
    test_cfg=dict(mode='slide', crop_size=crop_size, stride=(341, 341)),
)

train_dataloader = dict(batch_size=2)
val_dataloader = dict(batch_size=1)
test_dataloader = val_dataloader