key, val = kv.split('=', maxsplit=1) ValueError: not enough values to unpack (expected 2, got 1)

[LZYmixiu]

I defined a new config that attempt to train on my custom dataset (VOC format).
The command I used is : python tools/train.py configs/swin/upernet_swin_tiny_patch4_window7_512x512_160k_mfdataset.py --options model.pretrained = checkpoints/swin_tiny_patch4_window7_224.pth --gpus=1 --gpu-ids=1

The config I defined is:

base_ = [
    '../_base_/models/upernet_swin.py', '../_base_/datasets/pascal_voc12.py',
    '../_base_/default_runtime.py', '../_base_/schedules/schedule_160k.py'
]
data_root ='/storage/zdxy01/dataset/dataset'
model = dict(
    backbone=dict(
        embed_dim=96,
        depths=[2, 2, 6, 2],
        num_heads=[3, 6, 12, 24],
        window_size=7,
        ape=False,
        drop_path_rate=0.3,
        patch_norm=True,
        use_checkpoint=False
    ),
    decode_head=dict(
        in_channels=[96, 192, 384, 768],
        num_classes=150
    ),
    auxiliary_head=dict(
        in_channels=384,
        num_classes=150
    ))

# AdamW optimizer, no weight decay for position embedding & layer norm in backbone
optimizer = dict(_delete_=True, type='AdamW', lr=0.00006, betas=(0.9, 0.999), weight_decay=0.01,
                 paramwise_cfg=dict(custom_keys={'absolute_pos_embed': dict(decay_mult=0.),
                                                 'relative_position_bias_table': dict(decay_mult=0.),
                                                 'norm': dict(decay_mult=0.)}))

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)

# By default, models are trained on 8 GPUs with 2 images per GPU
data=dict(samples_per_gpu=2,
    train=dict(

            data_root=data_root,
            img_dir='images',
            ann_dir='labels',
            split='train.txt'),
    val=dict(
        
        data_root=data_root,
        img_dir='images',
        ann_dir='labels',
        split='val.txt',
        ),
    test=dict(
        
        data_root=data_root,
        img_dir='images',
        ann_dir='labels',
        split='val.txt',
        ))

Is there have any problem?
Or there is have another way to train a VOC format dataset with swin-upernet？
Thank you for your concern!