pandas-dev/pandas · error · NotImplementedError

the 'numba' engine doesn't support using a numpy ufunc as…

Error message

the 'numba' engine doesn't support using a numpy ufunc as the callable function

What it means

Raised in `FrameApply.apply` ufunc branch when `self.func` is a `numpy.ufunc` and `engine='numba'`. Although ufuncs are callable, pandas routes them through its internal manager path (`_mgr.apply('apply', ...)`) rather than the numba JIT path, so they are explicitly rejected with NotImplementedError.

Solutions

Drop `engine='numba'`; the python engine handles ufuncs efficiently via the manager path.
Wrap the ufunc in a Python callable for numba: `df.apply(lambda x: np.log(x), engine='numba')` (may or may not compile depending on numba version).
For simple elementwise math, call the ufunc directly: `np.log(df)`.

Example fix

// before
df.apply(np.log, engine='numba')
// after
np.log(df)  # or df.apply(np.log)

Defensive patterns

Strategy: validation

Validate before calling

import numpy as np
def frame_apply_ufunc_engine(df, func, engine='python'):
    if engine == 'numba' and isinstance(func, np.ufunc):
        # apply directly — ufuncs are vectorized already
        return func(df)
    return df.apply(func, engine=engine)

Type guard

def numba_engine_supports_func_type(func) -> bool:
    import numpy as np
    return callable(func) and not isinstance(func, (str, np.ufunc))

Try / catch

try:
    out = df.apply(func, engine='numba')
except NotImplementedError as e:
    if 'numba' in str(e).lower() and 'ufunc' in str(e).lower():
        out = func(df)  # apply ufunc directly
    else:
        raise

Prevention

Apply numpy ufuncs directly to the frame (np.log(df)) instead of via apply.
Do not pair ufuncs with engine='numba'.
Document that numba needs pure-Python callables.

When it happens

Trigger: `df.apply(np.log, engine='numba')`, `df.apply(np.sqrt, engine='numba')`, or any `np.ufunc` instance with the numba engine.

Common situations: Users assume ufuncs (being numeric) are ideal for numba and pass them with `engine='numba'`; copy-paste of the engine flag.

Related errors

AI-assisted analysis of pandas-dev/pandas@3b7651241d (2026-08-11). Data as JSON: /api/errors/2c0ac5dc71d6f0c1. Report an issue: GitHub.

Appendix: source

Thrown at pandas/core/apply.py:1036

            return self.apply_list_or_dict_like()

        # all empty
        if len(self.columns) == 0 and len(self.index) == 0:
            return self.apply_empty_result()

        # string dispatch
        if isinstance(self.func, str):
            if self.engine == "numba":
                raise NotImplementedError(
                    "the 'numba' engine doesn't support using "
                    "a string as the callable function"
                )
            return self.apply_str()

        # ufunc
        elif isinstance(self.func, np.ufunc):
            if self.engine == "numba":
                raise NotImplementedError(
                    "the 'numba' engine doesn't support "
                    "using a numpy ufunc as the callable function"
                )
            with np.errstate(all="ignore"):
                results = self.obj._mgr.apply("apply", func=self.func)
            # _constructor will retain self.index and self.columns
            return self.obj._constructor_from_mgr(results, axes=results.axes)

        # broadcasting
        if self.result_type == "broadcast":
            if self.engine == "numba":
                raise NotImplementedError(
                    "the 'numba' engine doesn't support result_type='broadcast'"
                )
            return self.apply_broadcast(self.obj)

        # one axis empty
        elif not all(self.obj.shape):

View on GitHub (pinned to 3b7651241d)