TheAlgorithms/Python · error · ValueError

rbf kernel requires gamma

Error message

rbf kernel requires gamma

What it means

Raised by the SupportVectorMachine constructor when kernel='rbf' is requested but gamma is left at its default 0.0. The RBF kernel needs a positive width parameter; unlike sklearn this class has no automatic default (the source comments note a future default may be added), so the caller must supply gamma explicitly.

Source

Thrown at machine_learning/support_vector_machines.py:67

    Traceback (most recent call last):
        ...
    ValueError: gamma must be > 0
    """

    def __init__(
        self,
        *,
        regularization: float = np.inf,
        kernel: str = "linear",
        gamma: float = 0.0,
    ) -> None:
        self.regularization = regularization
        self.gamma = gamma
        if kernel == "linear":
            self.kernel = self.__linear
        elif kernel == "rbf":
            if self.gamma == 0:
                raise ValueError("rbf kernel requires gamma")
            if not isinstance(self.gamma, (float, int)):
                raise ValueError("gamma must be float or int")
            if not self.gamma > 0:
                raise ValueError("gamma must be > 0")
            self.kernel = self.__rbf
            # in the future, there could be a default value like in sklearn
            # sklear: def_gamma = 1/(n_features * X.var()) (wiki)
            # previously it was 1/(n_features)
        else:
            msg = f"Unknown kernel: {kernel}"
            raise ValueError(msg)

    # kernels
    def __linear(self, vector1: ndarray, vector2: ndarray) -> float:
        """Linear kernel (as if no kernel used at all)"""
        return np.dot(vector1, vector2)

    def __rbf(self, vector1: ndarray, vector2: ndarray) -> float:

View on GitHub (pinned to f5988cc097)

Solutions

1. Pass an explicit positive gamma, e.g. SupportVectorMachine(kernel='rbf', gamma=0.5) or gamma=1/n_features.
2. If you do not need nonlinearity, keep kernel='linear' which needs no gamma.
3. Compute a sklearn-style default yourself: gamma = 1 / (n_features * X.var()).

Example fix

# before
svm = SupportVectorMachine(kernel='rbf')

# after
svm = SupportVectorMachine(kernel='rbf', gamma=1.0 / X.shape[1])

gamma = gamma if gamma else 1.0 / n_features  # avoid default-0 trap
svm = SupportVectorMachine(kernel='rbf', gamma=gamma)

def has_positive_gamma(kernel: str, gamma: float) -> bool:
    return kernel != 'rbf' or (isinstance(gamma, (int, float)) and gamma > 0)

When it happens

Trigger: Constructing SupportVectorMachine(kernel='rbf') without passing gamma, or explicitly passing gamma=0 with the rbf kernel.

Common situations: Porting code from sklearn.svm.SVC where gamma='scale' works out of the box, then hitting this class's stricter contract; or toggling the kernel string from 'linear' to 'rbf' in a config without adding the gamma key.

Related errors

• gamma value must be non-negative
• gamma must be > 0
• Length of predicted and actual array must be same.
• y_true can have values -1 or 1 only.
• Test samples' feature length does not equal to that of train

AI-assisted analysis of TheAlgorithms/Python@f5988cc097 (2026-08-14). Data as JSON: /api/errors/d63c5a4e5178fdd8. Report an issue: GitHub.