What causes "mode[{model_name}_model] is not implemented!" in PaddlePaddle/PaddleOCR?

Set model_name: large or model_name: small (lowercase) in the det_mobilenet_v3 backbone config

PaddlePaddle/PaddleOCR · error · NotImplementedError

mode[{model_name}_model] is not implemented!

Error message

mode[{model_name}_model] is not implemented!

What it means

The MobileNetV3 detection backbone selects its inverted-residual block table by model_name and only implements 'large' and 'small' (the SOURCE shows the tail of the 'small' table, cls_ch_squeeze=576). Any other value raises NotImplementedError with the message 'mode[<name>_model] is not implemented!'. Note the message is slightly misleading — it is model_name, not 'mode', that is wrong.

Source

Thrown at ppocr/modeling/backbones/det_mobilenet_v3.py:87

            cls_ch_squeeze = 960
        elif model_name == "small":
            cfg = [
                # k, exp, c,  se,     nl,  s,
                [3, 16, 16, True, "relu", 2],
                [3, 72, 24, False, "relu", 2],
                [3, 88, 24, False, "relu", 1],
                [5, 96, 40, True, "hardswish", 2],
                [5, 240, 40, True, "hardswish", 1],
                [5, 240, 40, True, "hardswish", 1],
                [5, 120, 48, True, "hardswish", 1],
                [5, 144, 48, True, "hardswish", 1],
                [5, 288, 96, True, "hardswish", 2],
                [5, 576, 96, True, "hardswish", 1],
                [5, 576, 96, True, "hardswish", 1],
            ]
            cls_ch_squeeze = 576
        else:
            raise NotImplementedError(
                "mode[" + model_name + "_model] is not implemented!"
            )

        supported_scale = [0.35, 0.5, 0.75, 1.0, 1.25]
        assert (
            scale in supported_scale
        ), "supported scale are {} but input scale is {}".format(supported_scale, scale)
        inplanes = 16
        # conv1
        self.conv = ConvBNLayer(
            in_channels=in_channels,
            out_channels=make_divisible(inplanes * scale),
            kernel_size=3,
            stride=2,
            padding=1,
            groups=1,
            if_act=True,
            act="hardswish",

View on GitHub (pinned to 2661c7c0ef)

Solutions

Set model_name: large or model_name: small (lowercase) in the det_mobilenet_v3 backbone config
Check for a capitalized variant like 'Large' — the comparison is case-sensitive
Confirm scale is also one of 0.35/0.5/0.75/1.0/1.25, since the next line asserts it

Example fix

# before
Backbone:
  name: MobileNetV3
  model_name: Medium

# after
Backbone:
  name: MobileNetV3
  model_name: small

Defensive patterns

Strategy: validation

Validate before calling

model_name = model_name.lower()
assert model_name in ('large', 'small'), f"model_name must be 'large' or 'small', got {model_name!r}"
backbone = MobileNetV3(model_name=model_name, scale=scale)

Type guard

def is_mobilenet_v3_name(n: str) -> bool:
    return isinstance(n, str) and n.lower() in ('large', 'small')

Prevention

Lowercase model_name defensively before passing it into the backbone
When copying configs between det and rec models, keep the model_name/scale keys consistent
Also pre-check scale in {0.35, 0.5, 0.75, 1.0, 1.25}

When it happens

Trigger: Backbone config Backbone.name=MobileNetV3 with model_name='medium', 'Large' (capitalized), or None; the exception fires during network construction.

Common situations: Copying a recognition MobileNetV3 config into a detection config (or vice versa) with an extra/renamed key; case-sensitive typos; assuming intermediate widths exist.

Related errors

AI-assisted analysis of PaddlePaddle/PaddleOCR@2661c7c0ef (2026-08-14). Data as JSON: /api/errors/88d79b0d10450bc2. Report an issue: GitHub.