tensorflow/models · error · ValueError

The linear layering config cannot be `None` when using the l

Error message

The linear layering config cannot be `None` when using the linear layering method.

What it means

Error "The linear layering config cannot be `None` when using the linear layering method." thrown in tensorflow/models.

Source

Thrown at official/recommendation/uplift/layers/heads/two_tower_logits_head.py:122

    Args:
      control_head: computes control logits from the control embedding. Its
        input and output is expected to be a dense tensor.
      treatment_head: computes treatment logits from the treatment embedding.
        Its input and output is expected to be a dense tensor.
      layering_config: configuration for the layering method. Defaults to no
        layering.
      **kwargs: base layer keyword arguments.
    """
    super().__init__(**kwargs)

    self._control_head = control_head
    self._treatment_head = treatment_head
    self._layering_config = layering_config

  def build(self, input_shapes: tuple[tf.TensorShape, tf.TensorShape]):
    if self._layering_config.layering_method == LayeringMethod.LINEAR_LAYERING:
      if self._layering_config.linear_layering_config is None:
        raise ValueError(
            "The linear layering config cannot be `None` when using the linear"
            " layering method."
        )
      # Build a learnable weight matrix that projects from the control embedding
      # space to the treatment embedding space.
      _, treatment_embedding_shape = input_shapes
      self._linear_layering = tf_keras.layers.Dense(
          units=treatment_embedding_shape[-1],
          activation=None,
          use_bias=True,
          kernel_initializer=(
              self._layering_config.linear_layering_config.kernel_initializer
          ),
          kernel_regularizer=(
              self._layering_config.linear_layering_config.kernel_regularizer
          ),
      )
    super().build(input_shapes)

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/recommendation/uplift/layers/heads/two_tower_logits_head.py:122 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/3f90756e94902ba9. Report an issue: GitHub.