ai_project_v1/uv_module/unetformer_UAV_6000X4000.py

crop_size = (
    1920,
    1920,
)
data_preprocessor = dict(
    bgr_to_rgb=True,
    mean=[
        123.675,
        116.28,
        103.53,
    ],
    pad_val=0,
    seg_pad_val=255,
    size=(
        1920,
        1920,
    ),
    std=[
        58.395,
        57.12,
        57.375,
    ],
    type='SegDataPreProcessor')
data_root = '/media/data2/ZJB/MPNet/data_root/UAV'
dataset_type = 'UVA_dataset'
default_hooks = dict(
    checkpoint=dict(
        by_epoch=False,
        interval=2000,
        max_keep_ckpts=3,
        save_best='mIoU',
        type='CheckpointHook'),
    logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'),
    param_scheduler=dict(type='ParamSchedulerHook'),
    sampler_seed=dict(type='DistSamplerSeedHook'),
    timer=dict(type='IterTimerHook'),
    visualization=dict(type='SegVisualizationHook'))
default_scope = 'mmseg'
env_cfg = dict(
    cudnn_benchmark=True,
    dist_cfg=dict(backend='nccl'),
    mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0))
img_ratios = [
    0.5,
    0.75,
    1.0,
    1.25,
    1.5,
    1.75,
]
launcher = 'none'
load_from = None
log_level = 'INFO'
log_processor = dict(by_epoch=False)
model = dict(
    auxiliary_head=dict(
        align_corners=False,
        channels=48,
        concat_input=False,
        dropout_ratio=0.1,
        in_channels=8,
        in_index=0,
        loss_decode=[
            dict(loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False),
            dict(loss_weight=0.4, type='DiceLoss', use_sigmoid=True),
        ],
        norm_cfg=dict(requires_grad=True, type='SyncBN'),
        num_classes=7,
        num_convs=1,
        type='DirectHead'),
    backbone=dict(
        backbone_name='swsl_resnet18',
        decode_channels=64,
        dropout=0.1,
        num_classes=7,
        pretrained=False,
        type='UNetFormer',
        window_size=8),
    data_preprocessor=dict(
        bgr_to_rgb=True,
        mean=[
            123.675,
            116.28,
            103.53,
        ],
        pad_val=0,
        seg_pad_val=255,
        size=(
            1920,
            1920,
        ),
        std=[
            58.395,
            57.12,
            57.375,
        ],
        type='SegDataPreProcessor'),
    decode_head=dict(
        align_corners=False,
        channels=48,
        concat_input=False,
        dropout_ratio=0.1,
        in_channels=8,
        in_index=0,
        loss_decode=[
            dict(loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False),
            dict(loss_weight=1.0, type='DiceLoss', use_sigmoid=True),
        ],
        norm_cfg=dict(requires_grad=True, type='SyncBN'),
        num_classes=7,
        num_convs=1,
        type='DirectHead'),
    pretrained=None,
    test_cfg=dict(mode='whole'),
    train_cfg=dict(),
    type='EncoderDecoder_unetformer')
norm_cfg = dict(requires_grad=True, type='SyncBN')
optim_wrapper = dict(
    clip_grad=None,
    optimizer=dict(lr=0.04, momentum=0.9, type='SGD', weight_decay=0.0005),
    type='OptimWrapper')
optimizer = dict(lr=0.04, momentum=0.9, type='SGD', weight_decay=0.0005)
param_scheduler = [
    dict(
        begin=0,
        by_epoch=False,
        end=20000,
        eta_min=0.0004,
        power=0.9,
        type='PolyLR'),
]
resume = False
test_cfg = dict(type='TestLoop')
test_dataloader = dict(
    batch_size=1,
    dataset=dict(
        data_prefix=dict(img_path='img/test', seg_map_path='mask/test'),
        data_root='/media/data2/ZJB/MPNet/data_root/UAV',
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(keep_ratio=False, scale=(
                6016,
                3968,
            ), type='Resize'),
            dict(type='LoadAnnotations'),
            dict(type='PackSegInputs'),
        ],
        type='UVA_dataset'),
    num_workers=4,
    persistent_workers=True,
    sampler=dict(shuffle=False, type='DefaultSampler'))
test_evaluator = dict(
    iou_metrics=[
        'mIoU',
    ], type='IoUMetric')
test_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(keep_ratio=False, scale=(
        6016,
        3968,
    ), type='Resize'),
    dict(type='LoadAnnotations'),
    dict(type='PackSegInputs'),
]
train_cfg = dict(max_iters=20000, type='IterBasedTrainLoop', val_interval=2000)
train_dataloader = dict(
    batch_size=4,
    dataset=dict(
        data_prefix=dict(img_path='img/train', seg_map_path='mask/train'),
        data_root='/media/data2/ZJB/MPNet/data_root/UAV',
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(type='LoadAnnotations'),
            dict(
                keep_ratio=True,
                ratio_range=(
                    1.0,
                    1.0,
                ),
                scale=(
                    6000,
                    4000,
                ),
                type='RandomResize'),
            dict(
                cat_max_ratio=0.75,
                crop_size=(
                    1920,
                    1920,
                ),
                type='RandomCrop'),
            dict(prob=0.5, type='RandomFlip'),
            dict(type='PhotoMetricDistortion'),
            dict(type='PackSegInputs'),
        ],
        type='UVA_dataset'),
    num_workers=4,
    persistent_workers=True,
    sampler=dict(shuffle=True, type='InfiniteSampler'))
train_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(type='LoadAnnotations'),
    dict(
        keep_ratio=True,
        ratio_range=(
            1.0,
            1.0,
        ),
        scale=(
            6000,
            4000,
        ),
        type='RandomResize'),
    dict(cat_max_ratio=0.75, crop_size=(
        1920,
        1920,
    ), type='RandomCrop'),
    dict(prob=0.5, type='RandomFlip'),
    dict(type='PhotoMetricDistortion'),
    dict(type='PackSegInputs'),
]
val_cfg = dict(type='ValLoop')
val_dataloader = dict(
    batch_size=1,
    dataset=dict(
        data_prefix=dict(img_path='img/val', seg_map_path='mask/val'),
        data_root='/media/data2/ZJB/MPNet/data_root/UAV',
        pipeline=[
            dict(type='LoadImageFromFile'),
            dict(keep_ratio=True, scale=(
                6000,
                4000,
            ), type='Resize'),
            dict(type='LoadAnnotations'),
            dict(type='PackSegInputs'),
        ],
        type='UVA_dataset'),
    num_workers=4,
    persistent_workers=True,
    sampler=dict(shuffle=False, type='DefaultSampler'))
val_evaluator = dict(
    iou_metrics=[
        'mIoU',
    ], type='IoUMetric')
vis_backends = [
    dict(type='LocalVisBackend'),
]
visualizer = dict(
    name='visualizer',
    type='SegLocalVisualizer',
    vis_backends=[
        dict(type='LocalVisBackend'),
    ])
work_dir = './work_dirs/unetformer_UAV_6000X4000'