tensorflow/models · error · ValueError

Inconsistent decoder type {decoder_type}. Need to be `aspp`.

Error message

Inconsistent decoder type {decoder_type}. Need to be `aspp`.

What it means

Error "Inconsistent decoder type {decoder_type}. Need to be `aspp`." thrown in tensorflow/models.

Source

Thrown at official/vision/modeling/decoders/aspp.py:184

  Args:
    input_specs: A `dict` of input specifications. A dictionary consists of
      {level: TensorShape} from a backbone. Note this is for consistent
        interface, and is not used by ASPP decoder.
    model_config: A OneOfConfig. Model config.
    l2_regularizer: A `tf_keras.regularizers.Regularizer` instance. Default to
      None.

  Returns:
    A `tf_keras.Model` instance of the ASPP decoder.

  Raises:
    ValueError: If the model_config.decoder.type is not `aspp`.
  """
  del input_specs  # input_specs is not used by ASPP decoder.
  decoder_type = model_config.decoder.type
  decoder_cfg = model_config.decoder.get()
  if decoder_type != 'aspp':
    raise ValueError(f'Inconsistent decoder type {decoder_type}. '
                     'Need to be `aspp`.')

  norm_activation_config = model_config.norm_activation
  return ASPP(  # pyrefly: ignore[bad-return]
      level=decoder_cfg.level,
      dilation_rates=decoder_cfg.dilation_rates,
      num_filters=decoder_cfg.num_filters,
      use_depthwise_convolution=decoder_cfg.use_depthwise_convolution,
      pool_kernel_size=decoder_cfg.pool_kernel_size,
      dropout_rate=decoder_cfg.dropout_rate,
      use_sync_bn=norm_activation_config.use_sync_bn,
      norm_momentum=norm_activation_config.norm_momentum,
      norm_epsilon=norm_activation_config.norm_epsilon,
      activation=norm_activation_config.activation,
      kernel_regularizer=l2_regularizer,
      spp_layer_version=decoder_cfg.spp_layer_version,
      output_tensor=decoder_cfg.output_tensor)

View on GitHub (pinned to e006f5f0d5)

Solutions

1. Set the decoder type to 'aspp' when building an ASPP decoder.
2. Use the matching factory/config class so the decoder type is consistent.

When it happens

Trigger: Thrown at official/vision/modeling/decoders/aspp.py:184 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/eb563877b77f6028. Report an issue: GitHub.