lllyasviel/ControlNet · error · ValueError

inverted_residual_setting should be non-empty or a 4-element

Error message

inverted_residual_setting should be non-empty or a 4-element list, got {}

What it means

Identical validation in the MLSd tiny variant: the constructor requires inverted_residual_setting to be a non-empty list of 4-element (t, c, n, s) tuples. An empty list or malformed first row triggers the error.

Source

Thrown at annotator/mlsd/models/mbv2_mlsd_tiny.py:185

        input_channel = 32
        last_channel = 1280
        width_mult = 1.0
        round_nearest = 8

        inverted_residual_setting = [
            # t, c, n, s
            [1, 16, 1, 1],
            [6, 24, 2, 2],
            [6, 32, 3, 2],
            [6, 64, 4, 2],
            #[6, 96, 3, 1],
            #[6, 160, 3, 2],
            #[6, 320, 1, 1],
        ]

        # only check the first element, assuming user knows t,c,n,s are required
        if len(inverted_residual_setting) == 0 or len(inverted_residual_setting[0]) != 4:
            raise ValueError("inverted_residual_setting should be non-empty "
                             "or a 4-element list, got {}".format(inverted_residual_setting))

        # building first layer
        input_channel = _make_divisible(input_channel * width_mult, round_nearest)
        self.last_channel = _make_divisible(last_channel * max(1.0, width_mult), round_nearest)
        features = [ConvBNReLU(4, input_channel, stride=2)]
        # building inverted residual blocks
        for t, c, n, s in inverted_residual_setting:
            output_channel = _make_divisible(c * width_mult, round_nearest)
            for i in range(n):
                stride = s if i == 0 else 1
                features.append(block(input_channel, output_channel, stride, expand_ratio=t))
                input_channel = output_channel
        self.features = nn.Sequential(*features)

        self.fpn_selected = [3, 6, 10]
        # weight initialization
        for m in self.modules():

View on GitHub (pinned to ed85cd1e25)

Solutions

1. Supply a valid non-empty list of [t, c, n, s] rows
2. Diff your config against the default in mbv2_mlsd_tiny.py
3. Guard with an assertion before model construction

Example fix

# before
model = MobileNetV2([[6, 24, 2]])  # 3-element row
# after
model = MobileNetV2([[6, 24, 2, 2]])  # [t,c,n,s]

assert len(inverted_residual_setting) > 0 and all(len(r) == 4 for r in inverted_residual_setting), 'need non-empty list of [t,c,n,s] rows'

def is_valid_setting(s) -> bool:
    return isinstance(s, (list, tuple)) and len(s) > 0 and all(len(r) == 4 for r in s)

When it happens

Trigger: Building mbv2_mlsd_tiny with a custom inverted_residual_setting that is empty or has a first element whose length != 4.

Common situations: Porting layer configs between tiny/large variants; YAML overrides that drop the list; experimental pruning scripts emptying the setting.

Related errors

• inverted_residual_setting should be non-empty or a 4-element
• resize_method {self.__resize_method} not implemented
• provide num_res_blocks either as an int (globally constant)
• 'order' must be '1' or '2' or '3'.
• 'solver_type' must be either 'dpm_solver' or 'taylor', got {

AI-assisted analysis of lllyasviel/ControlNet@ed85cd1e25 (2026-08-27). Data as JSON: /api/errors/613bbb420e839430. Report an issue: GitHub.