open-mmlab/mmdetection · error · TypeError

pretrained must be a str or None

Error message

pretrained must be a str or None

What it means

MobileNetV2.__init__ throws TypeError when the deprecated `pretrained` argument is neither a str nor None. The constructor accepts pretrained only as a checkpoint path string (converted to init_cfg) or None; any other type (dict, bool, list) is rejected. New configs should use init_cfg instead.

Source

Thrown at mmdet/models/backbones/mobilenet_v2.py:77

        self.pretrained = pretrained
        assert not (init_cfg and pretrained), \
            'init_cfg and pretrained cannot be specified at the same time'
        if isinstance(pretrained, str):
            warnings.warn('DeprecationWarning: pretrained is deprecated, '
                          'please use "init_cfg" instead')
            self.init_cfg = dict(type='Pretrained', checkpoint=pretrained)
        elif pretrained is None:
            if init_cfg is None:
                self.init_cfg = [
                    dict(type='Kaiming', layer='Conv2d'),
                    dict(
                        type='Constant',
                        val=1,
                        layer=['_BatchNorm', 'GroupNorm'])
                ]
        else:
            raise TypeError('pretrained must be a str or None')

        self.widen_factor = widen_factor
        self.out_indices = out_indices
        if not set(out_indices).issubset(set(range(0, 8))):
            raise ValueError('out_indices must be a subset of range'
                             f'(0, 8). But received {out_indices}')

        if frozen_stages not in range(-1, 8):
            raise ValueError('frozen_stages must be in range(-1, 8). '
                             f'But received {frozen_stages}')
        self.out_indices = out_indices
        self.frozen_stages = frozen_stages
        self.conv_cfg = conv_cfg
        self.norm_cfg = norm_cfg
        self.act_cfg = act_cfg
        self.norm_eval = norm_eval
        self.with_cp = with_cp

View on GitHub (pinned to cfd5d3a985)

Solutions

1. Use init_cfg instead: MobileNetV2(init_cfg=dict(type='Pretrained', checkpoint='mobilenet_v2_8xb32_cifar10...'))
2. If using pretrained, pass a str URL/path or None only
3. Remove duplicate init_cfg when passing pretrained as str (they cannot be combined)

Example fix

// before
model = dict(backbone=dict(type='MobileNetV2', pretrained=dict(type='Pretrained', checkpoint='...')))
// after
model = dict(backbone=dict(type='MobileNetV2', init_cfg=dict(type='Pretrained', checkpoint='...')))

assert pretrained is None or isinstance(pretrained, str), 'pretrained must be str or None'

def is_valid_pretrained(p) -> bool:\n    return p is None or (isinstance(p, str) and len(p) > 0)

When it happens

Trigger: Calling MobileNetV2(pretrained=dict(checkpoint=...)) or passing a Path/bool/list as pretrained; mixing old-style pretrained configs with new init_cfg usage.

Common situations: Migrating old mmdet v1/v2 configs that used pretrained='open-mmlab://...' or pretrained=dict(...); copy-pasting configs across mmdet versions where pretrained semantics changed.

Related errors

• pretrained must be a str or None
• pretrained must be a str or None
• `init_cfg` must contain the key "type"
• out_indices must be a subset of range(0, 8). But received {o
• pretrained must be a str or None

AI-assisted analysis of open-mmlab/mmdetection@cfd5d3a985 (2026-08-27). Data as JSON: /api/errors/ae378de9e0c2c55f. Report an issue: GitHub.