Skip to content

Helpers API

lefts.helpers.tabular_model(estimator_class, features, target)

Wraps a sklearn-compatible estimator constructor into a Lefts model factory.

The returned factory can be passed directly to leaf(model_constructor=...).

Parameters:

Name Type Description Default
estimator_class

A callable that constructs a sklearn-compatible estimator (must implement .fit(X, y) and .predict(X)). Use functools.partial to pre-bind constructor arguments.

 required
features list[str]

Column names to use as model inputs.

 required
target str

Column name of the target variable.

 required
Source code in src/lefts/helpers.py 
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
def tabular_model(
    estimator_class,
    features: list[str],
    target: str,
) -> Callable[..., Any]:
    """
    Wraps a sklearn-compatible estimator constructor into a Lefts model factory.

    The returned factory can be passed directly to `leaf(model_constructor=...)`.

    Parameters
    ----------
    estimator_class:
        A callable that constructs a sklearn-compatible estimator (must implement .fit(X, y) and .predict(X)).
        Use ``functools.partial`` to pre-bind constructor arguments.
    features:
        Column names to use as model inputs.
    target:
        Column name of the target variable.
    """

    def factory(**hyperparameters):
        est = estimator_class(**hyperparameters)

        class _Model:
            def fit(self, training_set: pl.DataFrame):
                X = training_set.select(features).to_numpy()
                y = training_set[target].to_numpy()
                est.fit(X, y)

            def predict(self, df: pl.DataFrame):
                return est.predict(df.select(features).to_numpy())

        return _Model()

    return factory