keras-team/keras · error · ValueError

`dtype` was passed both positionally and as a keyword argume

Error message

`dtype` was passed both positionally and as a keyword argument.

What it means

Layer.add_weight() in Keras 3 rejects calls where dtype is supplied both positionally (the 3rd positional argument) and via the dtype= keyword. Positional args map to (shape, initializer, dtype), so any overlap with the corresponding keyword raises this ValueError.

Source

Thrown at keras/src/layers/layer.py:605

                    f"Received: add_weight('{shape_arg}', ...). "
                    f"Use: add_weight(shape=..., name='{shape_arg}')."
                )
            if shape is not None:
                raise ValueError(
                    "`shape` was passed both positionally and as "
                    "a keyword argument."
                )
            shape = shape_arg
            if len(args) > 1:
                if initializer is not None:
                    raise ValueError(
                        "`initializer` was passed both positionally and "
                        "as a keyword argument."
                    )
                initializer = args[1]
            if len(args) > 2:
                if dtype is not None:
                    raise ValueError(
                        "`dtype` was passed both positionally and as a "
                        "keyword argument."
                    )
                dtype = args[2]
        if shape is None:
            shape = ()
        if dtype is not None:
            dtype = backend.standardize_dtype(dtype)
        else:
            dtype = self.variable_dtype
        if initializer is None:
            if "float" in dtype:
                initializer = "glorot_uniform"
            else:
                initializer = "zeros"
        initializer = initializers.get(initializer)
        with backend.name_scope(self.name, caller=self):
            variable = backend.Variable(

View on GitHub (pinned to 7a34a03db6)

Solutions

1. Drop one of the two dtype specifications and pass all arguments as keywords
2. Audit the layer for other positional/keyword overlaps (shape, initializer) since they raise sibling errors

Example fix

# before
self.add_weight((64, 64), 'zeros', 'float32', dtype='float32')
# after
self.add_weight(shape=(64, 64), initializer='zeros', dtype='float32')

# keep add_weight calls keyword-only: layer.add_weight(shape=s, initializer=i, dtype=d)

When it happens

Trigger: add_weight(shape, init, 'float32', dtype='float32') — the 3rd positional collides with the dtype keyword. Typical when Keras 2 code that passed dtype positionally is edited to add dtype= for mixed precision.

Common situations: Migrating mixed-precision Keras 2 layers; editing generated add_weight calls; custom layers whose signature changed across Keras versions.

Related errors

• `initializer` was passed both positionally and as a keyword
• Only input tensors may be passed as positional arguments. Th
• `add_loss()` can only be called from inside `build()` or `ca
• Quantization mode='{mode}' doesn't work well with compute_dt
• Feature '{name}' has `output_mode='one_hot'`. Thus its prepr

AI-assisted analysis of keras-team/keras@7a34a03db6 (2026-08-25). Data as JSON: /api/errors/23fbfd372aae6a35. Report an issue: GitHub.