huggingface/pytorch-image-models · error · TypeError

ScheduledBatchSampler requires a sampler with a length.

Error message

ScheduledBatchSampler requires a sampler with a length.

What it means

ScheduledBatchSampler must iterate a sized sampler to compute batch counts and budgets; the passed sampler lacks __len__ (e.g. an infinite or generator-backed sampler), so a TypeError is raised at construction.

Source

Thrown at timm/data/scheduled_sampler.py:61

            choices mixed into the progressive choice probabilities.
    """

    def __init__(
            self,
            sampler: Sampler,
            batch_sizes: Sequence[int],
            choice_weights: Optional[Sequence[float]] = None,
            seed: int = 0,
            drop_last: bool = True,
            shuffle_schedule: bool = True,
            num_batches: Optional[int] = None,
            choice_schedule: str = 'constant',
            schedule_epochs: Optional[int] = None,
            schedule_spread: float = 0.65,
            schedule_random_mix: float = 0.1,
    ) -> None:
        if not hasattr(sampler, '__len__'):
            raise TypeError('ScheduledBatchSampler requires a sampler with a length.')
        if len(sampler) <= 0:
            raise ValueError('ScheduledBatchSampler requires a non-empty sampler.')
        if not batch_sizes:
            raise ValueError('batch_sizes must contain at least one value.')
        if any(int(batch_size) != batch_size or batch_size <= 0 for batch_size in batch_sizes):
            raise ValueError('All scheduled batch sizes must be positive integers.')
        if num_batches is not None and (int(num_batches) != num_batches or num_batches <= 0):
            raise ValueError('num_batches must be a positive integer when specified.')
        if choice_schedule not in ('constant', 'progressive'):
            raise ValueError("choice_schedule must be 'constant' or 'progressive'.")
        if choice_schedule == 'progressive':
            if len(batch_sizes) < 2:
                raise ValueError('A progressive schedule requires at least two choices.')
            if schedule_epochs is None or int(schedule_epochs) != schedule_epochs or schedule_epochs <= 0:
                raise ValueError('schedule_epochs must be a positive integer for a progressive schedule.')
            if schedule_spread < 0:
                raise ValueError('schedule_spread must be non-negative.')
            if not 0 <= schedule_random_mix <= 1:

View on GitHub (pinned to 9a5261e31b)

Solutions

  1. Implement __len__ on the custom sampler (usually returning the underlying dataset length).
  2. Use a sized sampler such as RandomSampler(dataset) or the dataset itself.
  3. For infinite streams, pre-materialize an index list and wrap it in a sized Sampler.

Example fix

# before
class MySampler(Sampler):
    def __iter__(self):
        while True:
            yield random.randrange(n)

# after
class MySampler(Sampler):
    def __init__(self, n): self.n = n
    def __iter__(self):
        for _ in range(self.n):
            yield random.randrange(self.n)
    def __len__(self):
        return self.n
Defensive patterns

Strategy: type-guard

Validate before calling

assert hasattr(sampler, '__len__') and callable(getattr(sampler, '__len__')), 'sampler must be sized'

Type guard

def is_sized_sampler(s) -> bool:
    return hasattr(s, '__iter__') and hasattr(s, '__len__') and callable(s.__len__)

Try / catch

try:
    sched = ScheduledBatchSampler(sampler, batch_sizes=[128])
except TypeError as e:
    if 'sampler with a length' in str(e):
        sampler = RandomSampler(dataset)  # sized fallback
        sched = ScheduledBatchSampler(sampler, batch_sizes=[128])
    else:
        raise

Prevention

When it happens

Trigger: Passing sampler=iter(dataset), a custom Sampler without __len__, or a DistributedSamplerWrapper built over an unsized iterable to ScheduledBatchSampler.__init__.

Common situations: Wrapping streaming/infinite samplers; custom sampler subclasses that forgot to implement __len__; adapting example code that used itertools.cycle.

Related errors

AI-assisted analysis of huggingface/pytorch-image-models@9a5261e31b (2026-08-27). Data as JSON: /api/errors/84c35b7fdd10dc3f. Report an issue: GitHub.