Lightning-AI/pytorch-lightning · error · MisconfigurationException

Accumulation factor should be an int greater than 0. Got {li

Error message

Accumulation factor should be an int greater than 0. Got {list(scheduling.values())}.

What it means

All values of the `scheduling` dict must be ints >= 1 (accumulation factors). Zero, negative, float, or string values raise MisconfigurationException listing the offending values. Validated in `__init__` immediately after the key check.

Source

Thrown at src/lightning/pytorch/callbacks/gradient_accumulation_scheduler.py:79

        # because epoch (key) should be zero-indexed.
        >>> accumulator = GradientAccumulationScheduler(scheduling={4: 2})
        >>> trainer = Trainer(callbacks=[accumulator])

    """

    def __init__(self, scheduling: dict[int, int]):
        super().__init__()

        if not scheduling:  # empty dict error
            raise TypeError("Empty dict cannot be interpreted correct")

        if any(not isinstance(key, int) or key < 0 for key in scheduling):
            raise MisconfigurationException(
                f"Epoch should be an int greater than or equal to 0. Got {list(scheduling.keys())}."
            )

        if any(not isinstance(value, int) or value < 1 for value in scheduling.values()):
            raise MisconfigurationException(
                f"Accumulation factor should be an int greater than 0. Got {list(scheduling.values())}."
            )

        minimal_epoch = min(scheduling.keys())
        if minimal_epoch < 0:
            raise IndexError(f"Epochs indexing from 1, epoch {minimal_epoch} cannot be interpreted correct")
        if minimal_epoch != 0:  # if user didn't define first epoch accumulation factor
            scheduling.update({0: 1})

        self.scheduling = scheduling
        self.epochs = sorted(scheduling.keys())

    def going_to_accumulate_grad_batches(self) -> bool:
        return any(v > 1 for v in self.scheduling.values())

    def get_accumulate_grad_batches(self, epoch: int) -> int:
        accumulate_grad_batches = 1
        for iter_epoch in reversed(self.epochs):

View on GitHub (pinned to 9fed5c27d2)

Solutions

1. Use integers >= 1; to disable accumulation for an epoch use 1, never 0
2. Cast values: `{k: int(v) for k, v in schedule.items()}`
3. Inspect the value list in the error message to locate bad entries

Example fix

# before
GradientAccumulationScheduler({0: 0, 3: 4})
# after
GradientAccumulationScheduler({0: 1, 3: 4})

if any(not isinstance(v, int) or v < 1 for v in scheduling.values()):
    scheduling = {k: int(v) for k, v in scheduling.items()}
assert all(isinstance(v, int) and v >= 1 for v in scheduling.values())

def valid_factors(s: dict) -> bool:
    return all(type(v) is int and v >= 1 for v in s.values())

When it happens

Trigger: `GradientAccumulationScheduler({0: 0})`, `{0: 2.5}`, or `{'0': '8'}` — e.g. values loaded from YAML/JSON as strings or computed with float math.

Common situations: Config parsed from YAML without type coercion; wanting to 'disable' accumulation for an epoch by setting 0 (use 1 instead); fractional accumulation factors.

Understand the failure class

Background: "Invalid value" and "allowed values are" config errors: what your library rejected and how to fix it — this error's family across 41 libraries.

Related errors

• Epoch should be an int greater than or equal to 0. Got {list
• Filter should be a dictionary, given {filter!r}
• Expected `fabric.save(filter=...)` for key {k!r} to be a cal
• Empty dict cannot be interpreted correct
• Epochs indexing from 1, epoch {minimal_epoch} cannot be inte

AI-assisted analysis of Lightning-AI/pytorch-lightning@9fed5c27d2 (2026-08-28). Data as JSON: /api/errors/a505e83fc7a495b4. Report an issue: GitHub.