keras-team/keras · error · ValueError

Must specify exactly two of top_cropping, bottom_cropping, t

Error message

Must specify exactly two of top_cropping, bottom_cropping, target_height. Received: top_cropping={top_cropping}, bottom_cropping={bottom_cropping}, target_height={target_height}

What it means

crop_images (and its layer) requires exactly two of top_cropping, bottom_cropping, target_height; the third is derived. Zero, one, or three specified values make the vertical crop ambiguous and the validator raises.

Source

Thrown at keras/src/ops/image.py:1581

    _validate_non_negative(top_padding, "top_padding")
    _validate_non_negative(bottom_padding, "bottom_padding")
    _validate_non_negative(target_height, "target_height")
    _validate_non_negative(left_padding, "left_padding")
    _validate_non_negative(right_padding, "right_padding")
    _validate_non_negative(target_width, "target_width")


def _validate_crop_images_args(
    top_cropping,
    left_cropping,
    bottom_cropping,
    right_cropping,
    target_height,
    target_width,
):
    if [top_cropping, bottom_cropping, target_height].count(None) != 1:
        raise ValueError(
            "Must specify exactly two of "
            "top_cropping, bottom_cropping, target_height. "
            f"Received: top_cropping={top_cropping}, "
            f"bottom_cropping={bottom_cropping}, "
            f"target_height={target_height}"
        )
    if [left_cropping, right_cropping, target_width].count(None) != 1:
        raise ValueError(
            "Must specify exactly two of "
            "left_cropping, right_cropping, target_width. "
            f"Received: left_cropping={left_cropping}, "
            f"right_cropping={right_cropping}, "
            f"target_width={target_width}"
        )

    _validate_non_negative(top_cropping, "top_cropping")
    _validate_non_negative(bottom_cropping, "bottom_cropping")
    _validate_non_negative(target_height, "target_height")

View on GitHub (pinned to 7a34a03db6)

Solutions

1. Specify exactly two, e.g. crop_images(img, target_height=224, top_cropping=0)
2. Center-crop style: top_cropping=(H-target)//2 style values on both sides
3. Remove the redundant third argument

Example fix

# before
out = crop_images(img, target_height=224)

# after
out = crop_images(img, target_height=224, top_cropping=0)

assert [top_cropping, bottom_cropping, target_height].count(None) == 1

def crop_height_triple_ok(t, b, th) -> bool:
    return [t, b, th].count(None) == 1

When it happens

Trigger: keras.ops.image.crop_images(images) with all cropping args None, a single value, or all three given.

Common situations: Default-constructing the op expecting a no-op; mixing pad-style kwargs (padding) into crop; config templates where one bound is always set plus target, accidentally also setting the third.

Related errors

• Must specify exactly two of left_cropping, right_cropping, t
• Expected `size` to be a tuple of 2 integers. Received: size=
• `size` must have positive height and width. Received: size={
• Only one of `pad_to_aspect_ratio` & `crop_to_aspect_ratio` c
• {name} must be >= 0. Received: {name}={value}

AI-assisted analysis of keras-team/keras@7a34a03db6 (2026-08-25). Data as JSON: /api/errors/a4d224dbc9f5c155. Report an issue: GitHub.