keras-team/keras · error · ValueError

Must specify exactly two of left_padding, right_padding, tar

Error message

Must specify exactly two of left_padding, right_padding, target_width. Received: left_padding={left_padding}, right_padding={right_padding}, target_width={target_width}

What it means

Horizontal counterpart of the pad_images triple rule: exactly two of left_padding, right_padding, target_width must be specified; otherwise the width is over- or under-determined and the validator raises.

Source

Thrown at keras/src/ops/image.py:1556

def _validate_pad_images_args(
    top_padding,
    left_padding,
    bottom_padding,
    right_padding,
    target_height,
    target_width,
):
    if [top_padding, bottom_padding, target_height].count(None) != 1:
        raise ValueError(
            "Must specify exactly two of "
            "top_padding, bottom_padding, target_height. "
            f"Received: top_padding={top_padding}, "
            f"bottom_padding={bottom_padding}, "
            f"target_height={target_height}"
        )
    if [left_padding, right_padding, target_width].count(None) != 1:
        raise ValueError(
            "Must specify exactly two of "
            "left_padding, right_padding, target_width. "
            f"Received: left_padding={left_padding}, "
            f"right_padding={right_padding}, "
            f"target_width={target_width}"
        )

    _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,

View on GitHub (pinned to 7a34a03db6)

Solutions

1. Give exactly two of the three: e.g. target_width=224, right_padding=4
2. Symmetric case: left_padding=p, right_padding=p
3. Check both axes independently — height passing does not imply width is valid

Example fix

# before
out = pad_images(img, target_width=224)

# after
out = pad_images(img, target_width=224, right_padding=0)

assert [left_padding, right_padding, target_width].count(None) == 1

def width_triple_ok(l, r, tw) -> bool:
    return [l, r, tw].count(None) == 1

When it happens

Trigger: pad_images where the height triple is fine but width has 0, 1, or 3 specified values, e.g. only target_width=224 given.

Common situations: Asymmetric configs copied from a symmetric example; forgetting the second width argument when pad-to-target width; YAML keys typos (e.g. width_padding) silently leaving values None.

Related errors

• {name} must be >= 0. Received: {name}={value}
• Must specify exactly two of top_padding, bottom_padding, tar
• The `weights` argument should be either `None` (random initi
• `padding` should have two elements. Received: padding={paddi
• Invalid padding: {padding}

AI-assisted analysis of keras-team/keras@7a34a03db6 (2026-08-25). Data as JSON: /api/errors/9652eccb10dbc100. Report an issue: GitHub.