What causes "max_value must be greater than min_value" in apache/beam?

Swap or correct the values so max_value > min_value.

apache/beam · error · ValueError

max_value must be greater than min_value

Error message

max_value must be greater than min_value

What it means

ScaleMinMax validates its constructor arguments: the max_value used for scaling must be strictly greater than min_value, otherwise tft.scale_by_min_max would be mathematically invalid. __init__ raises a ValueError when max_value <= min_value.

Solutions

Swap or correct the values so max_value > min_value.
If bounds come from data, add a guard that skips or defaults scaling for constant columns where min == max.
Validate the config values before instantiating the transform.

Example fix

# before
tft.ScaleMinMax(columns=['x'], min_value=10, max_value=10)

# after
tft.ScaleMinMax(columns=['x'], min_value=0, max_value=10)

Defensive patterns

Strategy: validation

Validate before calling

def make_scale_min_max(columns, min_value, max_value):
    if not max_value > min_value:
        raise ValueError('max_value must be greater than min_value')
    return tft.ScaleMinMax(columns=columns, min_value=min_value, max_value=max_value)

Type guard

def valid_min_max(min_value, max_value) -> bool:
    return max_value > min_value

Try / catch

try:
    op = tft.ScaleMinMax(columns=['x'], min_value=mn, max_value=mx)
except ValueError as e:
    if 'max_value must be greater' in str(e):
        mn, mx = min(mn, mx), max(mn, mx)  # normalize order
        op = tft.ScaleMinMax(columns=['x'], min_value=mn, max_value=mx)
    else:
        raise

Prevention

Validate min < max in config-loading code before instantiating transforms.
Sort or normalize bounds when they are computed from data.
Reject constant columns (min == max) or fall back to ScaleToZScore.
Add unit tests for bound-order validation in transform configs.

When it happens

Trigger: Constructing tft.ScaleMinMax(columns=[...], min_value=a, max_value=b) with b <= a (equal values or inverted order), including default misuse where both are set to the same number.

Common situations: Loading min/max from a config where order was flipped, computing bounds from data and getting equal values for constant columns, or typos swapping the arguments.

Understand the failure class

Background: "Must be a positive integer", "Invalid value", "Unsupported": the invalid-argument-value error family, when a library rejects the value you pass — this error's family across 35 libraries.

Related errors

AI-assisted analysis of apache/beam@12126d8942 (2026-09-13). Data as JSON: /api/errors/c96402b6b5667fb3. Report an issue: GitHub.

Appendix: source

Thrown at sdks/python/apache_beam/ml/transforms/tft.py:552

      name: Optional[str] = None):
    """
    This function applies a scaling transformation on the given columns
    of incoming data. The transformation scales the input values to the
    range [min_value, max_value].

    Args:
      columns: A list of column names to apply the transformation on.
      min_value: The minimum value of the output range.
      max_value: The maximum value of the output range.
      name: A name for the operation (optional).
    """
    super().__init__(columns)
    self.min_value = min_value
    self.max_value = max_value
    self.name = name

    if self.max_value <= self.min_value:
      raise ValueError('max_value must be greater than min_value')

  def apply_transform(
      self, data: common_types.TensorType,
      output_column_name: str) -> common_types.TensorType:

    output = tft.scale_by_min_max(
        x=data, output_min=self.min_value, output_max=self.max_value)
    return {output_column_name: output}


@register_input_dtype(str)
class NGrams(TFTOperation):
  def __init__(
      self,
      columns: list[str],
      split_string_by_delimiter: Optional[str] = None,
      *,
      ngram_range: tuple[int, int] = (1, 1),

View on GitHub (pinned to 12126d8942)