keras-team/keras · error · ValueError

Invalid `output_shape`: {output_shape}. When specified, the

Error message

Invalid `output_shape`: {output_shape}. When specified, the `output_shape` should be of type tuple, list, or int.

What it means

Error "Invalid `output_shape`: {output_shape}. When specified, the `output_shape` should be of type tuple, list, or int." thrown in keras-team/keras.

Source

Thrown at keras/src/layers/attention/multi_head_attention.py:169

                "Received an invalid value for argument `value_dim`, "
                "expected a positive integer or `None`. Received: "
                f"value_dim={value_dim}"
            )
        super().__init__(**kwargs)
        self.supports_masking = True
        self._num_heads = num_heads
        self._key_dim = key_dim
        self._value_dim = value_dim if value_dim else key_dim
        self._dropout = dropout
        self._use_bias = use_bias
        self._use_gate = use_gate
        if output_shape:
            if isinstance(output_shape, int):
                output_shape = (output_shape,)
            try:
                output_shape = tuple(output_shape)
            except:
                raise ValueError(
                    f"Invalid `output_shape`: {output_shape}. When "
                    "specified, the `output_shape` should be of type tuple, "
                    "list, or int."
                )
        self._output_shape = output_shape
        self._flash_attention = flash_attention or is_flash_attention_enabled()
        self._kernel_initializer = initializers.get(kernel_initializer)
        self._bias_initializer = initializers.get(bias_initializer)
        self._kernel_regularizer = regularizers.get(kernel_regularizer)
        self._bias_regularizer = regularizers.get(bias_regularizer)
        self._activity_regularizer = regularizers.get(activity_regularizer)
        self._kernel_constraint = constraints.get(kernel_constraint)
        self._bias_constraint = constraints.get(bias_constraint)
        if isinstance(attention_axes, int):
            attention_axes = (attention_axes,)
        elif attention_axes and not isinstance(attention_axes, (list, tuple)):
            raise ValueError(
                "`attention_axes` must be an int, list, or tuple."

View on GitHub (pinned to 7a34a03db6)

When it happens

Trigger: Thrown at keras/src/layers/attention/multi_head_attention.py:169 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/acee4da0dd423763. Report an issue: GitHub.