tensorflow/models · error · ValueError

`layer` must be a `tf_keras.layer.Layer`. Observed `{}`

Error message

`layer` must be a `tf_keras.layer.Layer`. Observed `{}`

What it means

Error "`layer` must be a `tf_keras.layer.Layer`. Observed `{}`" thrown in tensorflow/models.

Source

Thrown at official/nlp/modeling/layers/spectral_normalization.py:75

        value estimate.
      aggregation: (tf.VariableAggregation) Indicates how a distributed variable
        will be aggregated. Accepted values are constants defined in the class
        tf.VariableAggregation.
      inhere_layer_name: (bool) Whether to inhere the name of the input layer.
      **kwargs: (dict) Other keyword arguments for the layers.Wrapper class.
    """
    self.iteration = iteration
    self.do_power_iteration = training
    self.aggregation = aggregation
    self.norm_multiplier = norm_multiplier

    # Set layer name.
    wrapper_name = kwargs.pop('name', None)
    if inhere_layer_name:
      wrapper_name = layer.name

    if not isinstance(layer, tf_keras.layers.Layer):
      raise ValueError('`layer` must be a `tf_keras.layer.Layer`. '
                       'Observed `{}`'.format(layer))
    super().__init__(
        layer, name=wrapper_name, **kwargs)

  def build(self, input_shape):  # pytype: disable=signature-mismatch  # overriding-parameter-count-checks
    super().build(input_shape)
    self.layer.kernel._aggregation = self.aggregation  # pylint: disable=protected-access
    self._dtype = self.layer.kernel.dtype

    self.w = self.layer.kernel
    self.w_shape = self.w.shape.as_list()

    self.v = self.add_weight(
        shape=(1, np.prod(self.w_shape[:-1])),
        initializer=tf.initializers.random_normal(),
        trainable=False,
        name='v',
        dtype=self.dtype,

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/nlp/modeling/layers/spectral_normalization.py:75 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of tensorflow/models@e006f5f0d5 (2026-08-24). Data as JSON: /api/errors/a5c95a31bbbe09d3. Report an issue: GitHub.