PaddlePaddle/PaddleOCR · error · ValueError

translation values should be between 0 and 1

Error message

translation values should be between 0 and 1

What it means

CVRandomAffine's constructor validates the translate argument: it must be a 2-element tuple/list where every element t satisfies 0.0 <= t <= 1.0 (fractions of image size, mirroring torchvision semantics). Any value outside [0,1] raises ValueError during transform construction.

Source

Thrown at ppocr/data/imaug/abinet_aug.py:98

        flags = get_interpolation()
        return cv2.warpAffine(
            img, M, (dst_w, dst_h), flags=flags, borderMode=cv2.BORDER_REPLICATE
        )


class CVRandomAffine(object):
    def __init__(self, degrees, translate=None, scale=None, shear=None):
        assert isinstance(degrees, numbers.Number), "degree should be a single number."
        assert degrees >= 0, "degree must be positive."
        self.degrees = degrees

        if translate is not None:
            assert (
                isinstance(translate, (tuple, list)) and len(translate) == 2
            ), "translate should be a list or tuple and it must be of length 2."
            for t in translate:
                if not (0.0 <= t <= 1.0):
                    raise ValueError("translation values should be between 0 and 1")
        self.translate = translate

        if scale is not None:
            assert (
                isinstance(scale, (tuple, list)) and len(scale) == 2
            ), "scale should be a list or tuple and it must be of length 2."
            for s in scale:
                if s <= 0:
                    raise ValueError("scale values should be positive")
        self.scale = scale

        if shear is not None:
            if isinstance(shear, numbers.Number):
                if shear < 0:
                    raise ValueError(
                        "If shear is a single number, it must be positive."
                    )
                self.shear = [shear]

View on GitHub (pinned to 2661c7c0ef)

Solutions

1. Express translation as fractions of image dimensions within [0,1], e.g. translate=(0.1, 0.2).
2. If you need pixel units, divide by image size first.
3. Validate the config values before building transforms.

Example fix

# before
CVRandomAffine(degrees=10, translate=(20, 30))  # ValueError
# after
CVRandomAffine(degrees=10, translate=(20/img_w, 30/img_h))  # fractions in [0,1]

def valid_translate(t) -> bool:
    return (t is None
            or (isinstance(t, (tuple, list)) and len(t) == 2
                and all(isinstance(x, (int, float)) and 0.0 <= x <= 1.0 for x in t)))

def is_valid_translate(v) -> bool:
    return v is None or (isinstance(v, (tuple, list)) and len(v) == 2 and all(0.0 <= x <= 1.0 for x in v))

When it happens

Trigger: CVRandomAffine(degrees=10, translate=(0.1, 1.5)) or translate=(-0.1, 0.2) — an element outside [0.0, 1.0]; values supplied as pixel counts (e.g. translate=(20, 30)) also trigger it.

Common situations: Porting torchvision affine configs but entering pixel offsets instead of fractions; typos/negatives in augmentation YAML; assuming larger-than-1 means 'more pixels'.

Related errors

• scale values should be positive
• If shear is a single number, it must be positive.
• Shear should be a single value or a tuple/list containing tw
• Unsupported format: .{ext}\nSupported formats: {supported}
• Invalid mode {mode}, must be one of ['union', 'small', 'larg

AI-assisted analysis of PaddlePaddle/PaddleOCR@2661c7c0ef (2026-08-14). Data as JSON: /api/errors/c1d9c5c546dd66d7. Report an issue: GitHub.