apache/beam · error · AssertionError

Unsupported serialization type.

Error message

Unsupported serialization type.

What it means

Raised in sklearn_inference._load_model as an AssertionError when the provided file_type is neither PICKLE nor JOBLIB. It is the terminal fall-through guard after the ModelFileType branches, meaning an unsupported serialization type reached the loader.

Source

Thrown at sdks/python/apache_beam/ml/inference/sklearn_inference.py:69

class ModelFileType(enum.Enum):
  """Defines how a model file is serialized. Options are pickle or joblib."""
  PICKLE = 1
  JOBLIB = 2


def _load_model(model_uri, file_type):
  file = FileSystems.open(model_uri, 'rb')
  if file_type == ModelFileType.PICKLE:
    return pickle.load(file)
  elif file_type == ModelFileType.JOBLIB:
    if not joblib:
      raise ImportError(
          'Could not import joblib in this execution environment. '
          'For help with managing dependencies on Python workers.'
          'see https://beam.apache.org/documentation/sdks/python-pipeline-dependencies/'  # pylint: disable=line-too-long
      )
    return joblib.load(file)
  raise AssertionError('Unsupported serialization type.')


def _default_numpy_inference_fn(
    model: BaseEstimator,
    batch: Sequence[numpy.ndarray],
    inference_args: Optional[dict[str, Any]] = None) -> Any:
  inference_args = {} if not inference_args else inference_args
  # vectorize data for better performance
  vectorized_batch = numpy.stack(batch, axis=0)
  return model.predict(vectorized_batch, **inference_args)


class SklearnModelHandlerNumpy(ModelHandler[numpy.ndarray,
                                            PredictionResult,
                                            BaseEstimator]):
  def __init__(
      self,
      model_uri: str,

View on GitHub (pinned to 12126d8942)

Solutions

  1. Pass model_file_type as ModelFileType.PICKLE or ModelFileType.JOBLIB explicitly
  2. Verify the ModelFileType enum members available in your installed Beam version
  3. Upgrade apache-beam on workers if the serialization type was added in a newer release

Example fix

// before
handler = SklearnModelHandler(model_uri=uri, model_file_type='joblib')
// after
handler = SklearnModelHandler(model_uri=uri, model_file_type=ModelFileType.JOBLIB)
Defensive patterns

Strategy: validation

Validate before calling

from apache_beam.ml.inference.sklearn_inference import ModelFileType
assert model_file_type in (ModelFileType.PICKLE, ModelFileType.JOBLIB), f'Unsupported type: {model_file_type}'

Type guard

def is_supported_model_file_type(t) -> bool:
    return t in (ModelFileType.PICKLE, ModelFileType.JOBLIB)

Try / catch

try:
    model = handler.load_model()
except AssertionError as e:
    if 'Unsupported serialization type' in str(e):
        handler.model_file_type = ModelFileType.PICKLE
        model = handler.load_model()
    else:
        raise

Prevention

When it happens

Trigger: Calling SklearnModelHandler (or _load_model) with a model_file_type value outside the ModelFileType enum's PICKLE/JOBLIB members, or constructing the enum with an arbitrary value.

Common situations: Passing a raw string instead of a ModelFileType enum member; an enum version mismatch where a new member exists in a newer Beam version than the worker runs; typo-ed enum construction via ModelFileType('pickle') with wrong case.

Related errors

AI-assisted analysis of apache/beam@12126d8942 (2026-09-13). Data as JSON: /api/errors/5f541a9cba445e90. Report an issue: GitHub.