pandas-dev/pandas · error · ValueError

Cannot convert from to . Supported resolutions are 's'…

Error message

Cannot convert from {self.dtype} to {dtype}. Supported resolutions are 's', 'ms', 'us', 'ns'

What it means

TimedeltaArray.astype refuses conversion to a numpy timedelta64 dtype whose unit is outside the supported resolution set {s, ms, us, ns}. Other resolutions (m, h, D, etc.) cannot be represented without overflow ambiguity, so pandas raises rather than silently coerce.

Solutions

1. Cast to one of the supported units: s.astype('timedelta64[s]').
2. For coarser display, use .as_unit('s') or format the output instead.
3. Convert via int64 nanoseconds and reapply a unit only if necessary.

Example fix

// before
s.astype('timedelta64[D]')  # ValueError
// after
s.astype('timedelta64[s]')
# or for day-grained display
s.dt.to_pytimedelta()

SUPPORTED = {'s','ms','us','ns'}
def safe_td_astype(s, unit):
    if unit not in SUPPORTED:
        raise ValueError(f"Unsupported timedelta unit {unit!r}; use one of {SUPPORTED}")
    return s.astype(f'timedelta64[{unit}]')

def is_supported_td64_dtype_str(dtype_str) -> bool:
    import re
    m = re.match(r'timedelta64\[(.+)\]', dtype_str)
    return bool(m) and m.group(1) in {'s','ms','us','ns'}

try:
    s.astype(dtype_str)
except ValueError as e:
    if 'Supported resolutions' in str(e):
        s.as_unit('s')  # or pick an appropriate supported unit
    else:
        raise

When it happens

Trigger: s.astype('timedelta64[m]'); s.astype('timedelta64[D]'); arr.astype(np.dtype('timedelta64[h]').

Common situations: Downstream code casts timedelta Series to coarser units expecting numpy parity; receiving a dtype string from config that includes an unsupported unit.

Related errors

• 'unit' must be one of 's', 'ms', 'us', 'ns'
• Supported units are 's', 'ms', 'us', 'ns'
• 'unit' must be one of 's', 'ms', 'us', 'ns'
• Values resolution does not match dtype.
• Cannot add or subtract timedelta64[ns] dtype from

AI-assisted analysis of pandas-dev/pandas@3b7651241d (2026-08-11). Data as JSON: /api/errors/9a3460a541ab2b45. Report an issue: GitHub.

Appendix: source

Thrown at pandas/core/arrays/timedeltas.py:382

    def astype(self, dtype, copy: bool = True):
        # We handle
        #   --> timedelta64[ns]
        #   --> timedelta64
        # DatetimeLikeArrayMixin super call handles other cases
        dtype = pandas_dtype(dtype)

        if lib.is_np_dtype(dtype, "m"):
            if dtype == self.dtype:
                if copy:
                    return self.copy()
                return self

            if is_supported_dtype(dtype):
                # unit conversion e.g. timedelta64[s]
                res_values = astype_overflowsafe(self._ndarray, dtype, copy=False)
                return type(self)._simple_new(res_values, dtype=res_values.dtype)
            else:
                raise ValueError(
                    f"Cannot convert from {self.dtype} to {dtype}. "
                    "Supported resolutions are 's', 'ms', 'us', 'ns'"
                )

        return dtl.DatetimeLikeArrayMixin.astype(self, dtype, copy=copy)

    def _iter_convert_chunk(self, data: np.ndarray) -> np.ndarray:
        return ints_to_pytimedelta(data, box=True)

    # ----------------------------------------------------------------
    # Reductions

    def sum(
        self,
        *,
        axis: AxisInt | None = None,
        dtype: NpDtype | None = None,
        out=None,

View on GitHub (pinned to 3b7651241d)