tensorflow/models · error · ValueError
add_tpu_function_alias is only allowed for input_type of: im
Error message
add_tpu_function_alias is only allowed for input_type of: image_tensor, image_bytes, tf_example.
What it means
Error "add_tpu_function_alias is only allowed for input_type of: image_tensor, image_bytes, tf_example." thrown in tensorflow/models.
Source
Thrown at official/vision/serving/export_saved_model_lib.py:157
input_image_size=input_image_size,
input_type=input_type,
num_channels=num_channels,
input_name=input_name,
)
else:
raise ValueError(
'Export module not implemented for {} task.'.format(type(params.task))
)
if add_tpu_function_alias:
if input_type == 'image_tensor':
inference_func = export_module.inference_from_image_tensors
elif input_type == 'image_bytes':
inference_func = export_module.inference_from_image_bytes
elif input_type == 'tf_example':
inference_func = export_module.inference_from_tf_example
else:
raise ValueError(
'add_tpu_function_alias is only allowed for input_type of:'
' image_tensor, image_bytes, tf_example.'
)
save_options = tf.saved_model.SaveOptions(
function_aliases={
'tpu_candidate': inference_func,
}
)
export_base.export(
export_module,
function_keys=function_keys if function_keys else [input_type],
export_savedmodel_dir=output_saved_model_directory,
checkpoint=checkpoint,
checkpoint_path=checkpoint_path,
timestamped=False,
save_options=save_options,
)View on GitHub (pinned to e006f5f0d5)
Solutions
- Set input_type to 'image_tensor', 'image_bytes', or 'tf_example' when add_tpu_function_alias is enabled.
- Disable add_tpu_function_alias for other input types.
When it happens
Trigger: Thrown at official/vision/serving/export_saved_model_lib.py:157 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/baebfe780c902fef.
Report an issue: GitHub.