pandas-dev/pandas · error · NotImplementedError

does not implement interpolate

Error message

{type(self).__name__} does not implement interpolate

What it means

The base ExtensionArray.interpolate raises NotImplementedError naming the subclass: interpolation is dtype-specific (linear needs numeric ordering, time-based needs a DatetimeIndex, etc.) and the base class cannot provide a correct default. Subclasses opt in by overriding interpolate. The source notes the method must return type(self) even when copy=False.

Solutions

1. Interpolate only the numeric/temporal columns: df.select_dtypes(include='number').interpolate(), or exclude the unsupported column with df.drop(columns=[...]).interpolate().
2. If you own the EA, implement interpolate returning type(self) with the desired method handling.
3. Use fillna (ffill/bfill) as a fallback where appropriate, since _pad_or_backfill has its own (also overridable) default.

Example fix

// before
df.interpolate()  # NotImplementedError on custom EA column

// after
df.select_dtypes(include='number').interpolate()
# or fill the unsupported column separately
df['custom_col'] = df['custom_col'].fillna(method='ffill')

from pandas.api.extensions import ExtensionArray
if getattr(type(arr), 'interpolate', None) is ExtensionArray.interpolate:
    # base will raise; use fillna instead
    out = arr.fillna(method='ffill')
else:
    out = arr.interpolate(method='linear', axis=0, index=..., limit=None, limit_direction='both', limit_area=None, copy=True)

def supports_interpolate(cls) -> bool:
    return getattr(cls, 'interpolate', None) is not ExtensionArray.interpolate

try:
    out = series.interpolate()
except NotImplementedError:
    out = series.ffill()

When it happens

Trigger: Calling Series.interpolate / DataFrame.interpolate on a column backed by an ExtensionArray whose class did not override interpolate (e.g. a custom categorical-like or string dtype). Internally pandas calls arr.interpolate(...) which hits the base NotImplemented.

Common situations: Applying df.interpolate() to a mixed-dtype frame where one non-numeric EA column lacks an interpolate override. Using interpolate(method='time') on a custom EA that only supports numeric data. Third-party dtype without interpolation support.

Related errors

• cannot perform with type
• Default 'empty' implementation is invalid for dtype=
• {dtype}
• function is not implemented for this dtype
• can only convert an array of size 1 to a Python scalar

AI-assisted analysis of pandas-dev/pandas@3b7651241d (2026-08-11). Data as JSON: /api/errors/bf43a601da7c8462. Report an issue: GitHub.

Appendix: source

Thrown at pandas/core/arrays/base.py:1298

        Interpolating values in a FloatingArray:

        >>> arr = pd.array([1.0, pd.NA, 3.0, 4.0, pd.NA, 6.0], dtype="Float64")
        >>> arr.interpolate(
        ...     method="linear",
        ...     axis=0,
        ...     index=pd.Index(range(len(arr))),
        ...     limit=None,
        ...     limit_direction="both",
        ...     limit_area=None,
        ...     copy=True,
        ... )
        <FloatingArray>
        [1.0, 2.0, 3.0, 4.0, 5.0, 6.0]
        Length: 6, dtype: Float64
        """
        # NB: we return type(self) even if copy=False
        raise NotImplementedError(
            f"{type(self).__name__} does not implement interpolate"
        )

    def _pad_or_backfill(
        self,
        *,
        method: FillnaOptions,
        limit: int | None = None,
        limit_area: Literal["inside", "outside"] | None = None,
        copy: bool = True,
    ) -> Self:
        """
        Pad or backfill values, used by Series/DataFrame ffill and bfill.

        This method propagates the last valid observation forward (pad/ffill)
        or the next valid observation backward (backfill/bfill) to fill NaN
        values.

View on GitHub (pinned to 3b7651241d)