CategoricalPrecision metric

@synalinks_export("synalinks.metrics.CategoricalPrecision")
class CategoricalPrecision(CategoricalFBetaScore):
    """Computes precision on categorical (list / label) structures.

    Mirrors `CategoricalF1Score`. Supports the optional `labels=` parameter:
    when provided, accumulation is per-label (sklearn-style) and
    `result()` returns a `{label: precision}` dict for `average=None`.


    Example:

    ```python
    program.compile(
        metrics=[
            synalinks.metrics.CategoricalPrecision(),
        ],
    )
    ```

    Args:
        average (str): One of `None`, `"micro"`, `"macro"`, `"weighted"`.
        labels (list): (Optional) Explicit list of label names to track.
        name (str): (Optional) string name of the metric instance.
        in_mask (list): (Optional) list of keys to keep.
        out_mask (list): (Optional) list of keys to remove.
        in_mask_pattern (str): (Optional) Regex pattern.
        out_mask_pattern (str): (Optional) Regex pattern.
    """

    def __init__(
        self,
        average=None,
        labels=None,
        name="categorical_precision",
        in_mask=None,
        out_mask=None,
        in_mask_pattern=None,
        out_mask_pattern=None,
    ):
        super().__init__(
            average=average,
            beta=1.0,
            labels=labels,
            name=name,
            in_mask=in_mask,
            out_mask=out_mask,
            in_mask_pattern=in_mask_pattern,
            out_mask_pattern=out_mask_pattern,
        )
        self._formula = "precision"

    def get_config(self):
        base_config = super().get_config()
        del base_config["beta"]
        return base_config