microsoft/qlib · error · NotImplementedError

Please implement the `__rsub__` method

Error message

Please implement the `__rsub__` method

What it means

BaseSingleMetric.__rsub__ implements reflected subtraction (other - self). Unlike __radd__ (which the base class implements as self + other), __rsub__ is left abstract because operand order matters: the base class cannot compute scalar-minus-metric. Only SingleMetric (line 436) implements it, so 1.0 - base_metric on a bare BaseSingleMetric raises NotImplementedError.

Source

Thrown at qlib/backtest/high_performance_ds.py:240

                SH600079    1.0
                SH600266    NaN
                           ...
                SZ300692    NaN
                SZ300719    NaN,
        """
        raise NotImplementedError(f"Please implement the `__init__` method")

    def __add__(self, other: Union[BaseSingleMetric, int, float]) -> BaseSingleMetric:
        raise NotImplementedError(f"Please implement the `__add__` method")

    def __radd__(self, other: Union[BaseSingleMetric, int, float]) -> BaseSingleMetric:
        return self + other

    def __sub__(self, other: Union[BaseSingleMetric, int, float]) -> BaseSingleMetric:
        raise NotImplementedError(f"Please implement the `__sub__` method")

    def __rsub__(self, other: Union[BaseSingleMetric, int, float]) -> BaseSingleMetric:
        raise NotImplementedError(f"Please implement the `__rsub__` method")

    def __mul__(self, other: Union[BaseSingleMetric, int, float]) -> BaseSingleMetric:
        raise NotImplementedError(f"Please implement the `__mul__` method")

    def __truediv__(self, other: Union[BaseSingleMetric, int, float]) -> BaseSingleMetric:
        raise NotImplementedError(f"Please implement the `__truediv__` method")

    def __eq__(self, other: object) -> BaseSingleMetric:
        raise NotImplementedError(f"Please implement the `__eq__` method")

    def __gt__(self, other: Union[BaseSingleMetric, int, float]) -> BaseSingleMetric:
        raise NotImplementedError(f"Please implement the `__gt__` method")

    def __lt__(self, other: Union[BaseSingleMetric, int, float]) -> BaseSingleMetric:
        raise NotImplementedError(f"Please implement the `__lt__` method")

    def __len__(self) -> int:
        raise NotImplementedError(f"Please implement the `__len__` method")

View on GitHub (pinned to 79633dd950)

Solutions

  1. Use SingleMetric, which implements both __sub__ and __rsub__
  2. In custom subclasses, implement __rsub__ as e.g. type(self)(other - self.storage) mirroring SingleMetric
  3. Rewrite scalar - m as m.__rsub__(scalar) only after confirming a concrete implementation exists, or compute (m * -1) + other

Example fix

# before
inv = 1.0 - BaseSingleMetric(s)
# after
from qlib.backtest.high_performance_ds import SingleMetric
inv = 1.0 - SingleMetric(s)
Defensive patterns

Strategy: type-guard

Validate before calling

from qlib.backtest.high_performance_ds import BaseSingleMetric
assert type(m).__rsub__ is not BaseSingleMetric.__rsub__, "object does not implement __rsub__"

Type guard

def supports_rsub(m) -> bool:
    from qlib.backtest.high_performance_ds import BaseSingleMetric
    return isinstance(m, BaseSingleMetric) and type(m).__rsub__ is not BaseSingleMetric.__rsub__

Prevention

When it happens

Trigger: Expressions like 1.0 - m or np.float64(x) - m where m is a raw BaseSingleMetric; broadcasting code (numpy, pandas ops) that triggers the reflected operator because the left operand's __sub__ returns NotImplemented; custom subclasses that override __sub__ but forget __rsub__.

Common situations: Writing inverse/excess-return metrics (benchmark - portfolio); mixing metric objects with numpy scalars; subclassing BaseSingleMetric and implementing only the forward operators.

Related errors

AI-assisted analysis of microsoft/qlib@79633dd950 (2026-08-15). Data as JSON: /api/errors/77335ed3dafba0cc. Report an issue: GitHub.