keras-team/keras · error · ValueError

Unsupported reduction: {reduction}

Error message

Unsupported reduction: {reduction}

What it means

Error "Unsupported reduction: {reduction}" thrown in keras-team/keras.

Source

Thrown at keras/src/backend/openvino/core.py:1348

def scatter_update(inputs, indices, updates, reduction=None):
    inputs = get_ov_output(inputs)
    indices = get_ov_output(indices)
    updates = get_ov_output(updates)

    inputs, updates = align_operand_types(inputs, updates, "scatter_update")

    # Map Keras reduction to OpenVINO ScatterNDUpdate reduction.
    # OpenVINO Opset 15 supports: "none", "sum", "sub", "prod", "min", "max".
    if reduction is None:
        ov_reduction = "none"
    elif reduction == "add":
        ov_reduction = "sum"
    elif reduction == "mul":
        ov_reduction = "prod"
    elif reduction in ("max", "min"):
        ov_reduction = reduction
    else:
        raise ValueError(f"Unsupported reduction: {reduction}")

    result = ov_opset.scatter_nd_update(
        inputs, indices, updates, reduction=ov_reduction
    ).output(0)
    return OpenVINOKerasTensor(result)


def slice(inputs, start_indices, shape):
    inputs = get_ov_output(inputs)
    if isinstance(start_indices, (list, np.ndarray)):
        start_indices = tuple(start_indices)
    if isinstance(shape, (list, np.ndarray)):
        shape = tuple(shape)
    if not isinstance(start_indices, tuple):
        raise ValueError(
            "`slice` operation requires tuple for `start_indices with the "
            f"openvino backend. Received: start_indices={start_indices}"
        )

View on GitHub (pinned to 7a34a03db6)

When it happens

Trigger: Thrown at keras/src/backend/openvino/core.py:1348 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of keras-team/keras@7a34a03db6 (2026-08-25). Data as JSON: /api/errors/e5e27ed2cf73f7d7. Report an issue: GitHub.