tensorflow/models · error · ValueError

Unrecognized `input_type`

Error message

Unrecognized `input_type`

What it means

Sentinel fallback in ExportModule.get_inference_signatures: the signature builder branches on the configured serving input key ('image_tensor', 'tf_example', 'tflite', etc.) and this raise is the else-branch reached when `input_type` matches none of them — i.e. the export config names an input format the module does not know how to expose as a concrete function.

Solutions

1. Set input_type to 'image_tensor', 'image_bytes', or 'tf_example'.
2. Check the serving input_type config for typos.

When it happens

Trigger: Thrown at official/vision/serving/export_base.py:221 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/5827afefedc8ce1e. Report an issue: GitHub.

Appendix: source

Thrown at official/vision/serving/export_base.py:221

        )
        signatures[def_name] = (
            self.inference_from_tf_example.get_concrete_function(
                input_signature
            )
        )
      elif key == 'tflite':
        input_signature = tf.TensorSpec(
            shape=[self._batch_size]
            + self._input_image_size
            + [self._num_channels],
            dtype=tf.float32,
            name=self._input_name,
        )
        signatures[def_name] = self.inference_for_tflite.get_concrete_function(
            input_signature
        )
      else:
        raise ValueError('Unrecognized `input_type`')
    return signatures

View on GitHub (pinned to e006f5f0d5)