Skip to content

Estimator recipes

myogestic.recipes.estimators ships constructor recipes for third-party estimators - thin wrappers that return a fitted-or-fittable object (.fit(X, y) + .predict(X)) with sane defaults. The library never owns the model lifecycle; that stays in your @pipeline.train. Optional dependencies are imported lazily, and each constructor raises a clear ImportError naming the extra to install.

To persist a trained model, use myogestic.ml.save_pickle / load_pickle (see the ML API).

CatBoost

catboost_classifier 

catboost_classifier(**kwargs: Any) -> Any

CatBoostClassifier with quiet defaults.

Source code in myogestic/recipes/estimators.py 
49
50
51
52
53
54
def catboost_classifier(**kwargs: Any) -> Any:
    """CatBoostClassifier with quiet defaults."""
    cb = _require("catboost", "examples")
    kwargs.setdefault("verbose", 0)
    kwargs.setdefault("allow_writing_files", False)
    return cb.CatBoostClassifier(**kwargs)

catboost_regressor 

catboost_regressor(**kwargs: Any) -> Any

CatBoostRegressor with quiet defaults.

Source code in myogestic/recipes/estimators.py 
57
58
59
60
61
62
def catboost_regressor(**kwargs: Any) -> Any:
    """CatBoostRegressor with quiet defaults."""
    cb = _require("catboost", "examples")
    kwargs.setdefault("verbose", 0)
    kwargs.setdefault("allow_writing_files", False)
    return cb.CatBoostRegressor(**kwargs)

scikit-learn

sklearn_classifier 

sklearn_classifier(**kwargs: Any) -> Any

RandomForestClassifier (sklearn).

Source code in myogestic/recipes/estimators.py 
68
69
70
71
def sklearn_classifier(**kwargs: Any) -> Any:
    """RandomForestClassifier (sklearn)."""
    skl = _require("sklearn.ensemble", "examples", pip_name="scikit-learn")
    return skl.RandomForestClassifier(**kwargs)

sklearn_regressor 

sklearn_regressor(**kwargs: Any) -> Any

RandomForestRegressor (sklearn).

Source code in myogestic/recipes/estimators.py 
74
75
76
77
def sklearn_regressor(**kwargs: Any) -> Any:
    """RandomForestRegressor (sklearn)."""
    skl = _require("sklearn.ensemble", "examples", pip_name="scikit-learn")
    return skl.RandomForestRegressor(**kwargs)

sklearn_extra_trees_classifier 

sklearn_extra_trees_classifier(**kwargs: Any) -> Any

ExtraTreesClassifier (sklearn). Defaults n_estimators=300, n_jobs=-1.

Source code in myogestic/recipes/estimators.py 
80
81
82
83
84
85
86
def sklearn_extra_trees_classifier(**kwargs: Any) -> Any:
    """ExtraTreesClassifier (sklearn). Defaults n_estimators=300, n_jobs=-1."""
    skl = _require("sklearn.ensemble", "examples", pip_name="scikit-learn")
    kwargs.setdefault("n_estimators", 300)
    kwargs.setdefault("random_state", 0)
    kwargs.setdefault("n_jobs", -1)
    return skl.ExtraTreesClassifier(**kwargs)

sklearn_extra_trees_regressor 

sklearn_extra_trees_regressor(**kwargs: Any) -> Any

ExtraTreesRegressor (sklearn). Same defaults as the classifier.

Source code in myogestic/recipes/estimators.py 
89
90
91
92
93
94
95
def sklearn_extra_trees_regressor(**kwargs: Any) -> Any:
    """ExtraTreesRegressor (sklearn). Same defaults as the classifier."""
    skl = _require("sklearn.ensemble", "examples", pip_name="scikit-learn")
    kwargs.setdefault("n_estimators", 300)
    kwargs.setdefault("random_state", 0)
    kwargs.setdefault("n_jobs", -1)
    return skl.ExtraTreesRegressor(**kwargs)

sklearn_logistic_classifier 

sklearn_logistic_classifier(**kwargs: Any) -> Any

Multinomial LogisticRegression (sklearn). max_iter=1000 default.

Source code in myogestic/recipes/estimators.py 
 98
 99
100
101
102
def sklearn_logistic_classifier(**kwargs: Any) -> Any:
    """Multinomial LogisticRegression (sklearn). max_iter=1000 default."""
    skl = _require("sklearn.linear_model", "examples", pip_name="scikit-learn")
    kwargs.setdefault("max_iter", 1000)
    return skl.LogisticRegression(**kwargs)

Dummy estimators (zero deps)

constant_classifier 

constant_classifier(class_index: int = 0) -> _ConstantClassifier

Estimator that always predicts class_index. No deps.

Source code in myogestic/recipes/estimators.py 
146
147
148
def constant_classifier(class_index: int = 0) -> _ConstantClassifier:
    """Estimator that always predicts ``class_index``. No deps."""
    return _ConstantClassifier(class_index=class_index)

mean_regressor 

mean_regressor() -> _MeanRegressor

Estimator that predicts the mean of training targets. No deps.

Source code in myogestic/recipes/estimators.py 
151
152
153
def mean_regressor() -> _MeanRegressor:
    """Estimator that predicts the mean of training targets. No deps."""
    return _MeanRegressor()