keras-team/keras · error · ValueError

`mask` should be a list. Received mask=

Error message

`mask` should be a list. Received mask={mask}

What it means

Raised by the `Concatenate` layer when the `mask` argument is not a list. When masking is used with `Concatenate`, masks must be provided as a list with one mask per input tensor. Wrap the mask(s) in a list before passing them to the layer.

Solutions

  1. Pass mask as a list, one per input: Concatenate()([x1, x2], mask=[m1, m2])
  2. Use None entries for unmasked inputs
  3. Let Keras propagate masks automatically instead of hand-passing them

Example fix

# before
out = layers.Concatenate()([x1, x2], mask=m)

# after
out = layers.Concatenate()([x1, x2], mask=[m, None])
Defensive patterns

Strategy: validation

Validate before calling

assert mask is None or isinstance(mask, (list, tuple)), 'mask must be a list parallel to inputs'

Type guard

def is_mask_list(mask) -> bool:
    return mask is None or isinstance(mask, (list, tuple))

Prevention

When it happens

Trigger: Concatenate()([x1, x2], mask=mask_tensor); custom layers passing a bare mask when delegating to a Concatenate layer.

Common situations: Sequence models with mask_zero embeddings feeding a concat; manually constructing mask arguments in custom training loops.

Related errors

AI-assisted analysis of keras-team/keras@7a34a03db6 (2026-08-25). Data as JSON: /api/errors/3d35a3d0b2afdd9b. Report an issue: GitHub.

Appendix: source

Thrown at keras/src/layers/merging/concatenate.py:126

            raise ValueError(
                "A `Concatenate` layer should be called on a list of inputs. "
                f"Received: input_shape={input_shape}"
            )
        input_shapes = input_shape
        output_shape = list(input_shapes[0])

        for shape in input_shapes[1:]:
            if output_shape[self.axis] is None or shape[self.axis] is None:
                output_shape[self.axis] = None
                break
            output_shape[self.axis] += shape[self.axis]
        return tuple(output_shape)

    def compute_mask(self, inputs, mask=None):
        if mask is None:
            return None
        if not isinstance(mask, (tuple, list)):
            raise ValueError(f"`mask` should be a list. Received mask={mask}")
        if not isinstance(inputs, (tuple, list)):
            raise ValueError(
                f"`inputs` should be a list. Received: inputs={inputs}"
            )
        if len(mask) != len(inputs):
            raise ValueError(
                "The lists `inputs` and `mask` should have the same length. "
                f"Received: inputs={inputs} of length {len(inputs)}, and "
                f"mask={mask} of length {len(mask)}"
            )
        if all(m is None for m in mask):
            return None
        # Make a list of masks while making sure
        # the dimensionality of each mask
        # is the same as the corresponding input.
        masks = []
        for input_i, mask_i in zip(inputs, mask):
            if mask_i is None:

View on GitHub (pinned to 7a34a03db6)