pandas-dev/pandas · error · ValueError

No accumulation for implemented on BaseMaskedArray

Error message

No accumulation for {func} implemented on BaseMaskedArray

What it means

ValueError raised in _cum_func (datetimelike_accumulations) when the requested numpy accumulation function is not one of the supported ones (np.maximum.accumulate, np.cumsum, np.minimum.accumulate). The lookup table of fill values has no entry, indicating an unsupported accumulation on a datetimelike masked array.

Solutions

Use a supported accumulation on datetimelike data: cumsum, cummin (minimum.accumulate), cummax (maximum.accumulate).
Convert the values to numeric (e.g. int64 nanoseconds) before applying a different accumulation.

Defensive patterns

Strategy: validation

Validate before calling

SUPPORTED = {'cumsum', 'cummin', 'cummax'}
def safe_cum(s, method):
    if method not in SUPPORTED:
        raise ValueError(f'{method} is unsupported on datetimelike arrays; use one of {SUPPORTED}')
    return getattr(s, method)()

Try / catch

try:
    s.cumsum()
except ValueError as e:
    if 'No accumulation' in str(e):
        s.astype('int64').cumsum()  # fall back to numeric
    else:
        raise

Prevention

Restrict datetimelike accumulations to cumsum/cummin/cummax.
Convert to int64 nanoseconds for custom accumulations.

When it happens

Trigger: Internal dispatch only: cumprod or another unsupported accumulation routed onto a datetimelike (datetime64/timedelta64) masked array.

Common situations: Rare; surfaces when custom array code or monkeypatching invokes an unsupported cum-method on datetime/timedelta data.

Related errors

AI-assisted analysis of pandas-dev/pandas@3b7651241d (2026-08-11). Data as JSON: /api/errors/c436b2beb515c682. Report an issue: GitHub.

Appendix: source

Thrown at pandas/core/array_algos/datetimelike_accumulations.py:78

    Accumulations for 1D datetimelike arrays.

    Parameters
    ----------
    func : np.cumsum, np.maximum.accumulate, np.minimum.accumulate
    values : np.ndarray
        Numpy array with the values (can be of any dtype that support the
        operation). Values is changed is modified inplace.
    skipna : bool, default True
        Whether to skip NA.
    """
    try:
        fill_value = {
            np.maximum.accumulate: np.iinfo(np.int64).min,
            np.cumsum: 0,
            np.minimum.accumulate: np.iinfo(np.int64).max,
        }[func]
    except KeyError as err:
        raise ValueError(
            f"No accumulation for {func} implemented on BaseMaskedArray"
        ) from err

    mask = isna(values)
    y = values.view("i8")
    y[mask] = fill_value

    if not skipna:
        mask = np.maximum.accumulate(mask)

    # GH 57956
    result = func(y, axis=0)
    if func is np.cumsum:
        # GH#66551: cummin/cummax cannot leave the range, cumsum can
        _check_cumsum_overflow(y, result, mask)
    result[mask] = iNaT

    if values.dtype.kind in "mM":

View on GitHub (pinned to 3b7651241d)