Skip to content

feat(hierarchical): support sample_weight #737

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
AhmedThahir wants to merge 4 commits into
base: main
Choose a base branch
from
Open

feat(hierarchical): support sample_weight #737

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
c578f53
feat(hierarchical): support sample_weight
Mar 25, 2025
36e201d
fix(hierarchical): convert narwhals to numpy
Mar 29, 2025
3988b94
ref(hierarchical): remove and self.has_sw_ for passing weights
Mar 29, 2025
d9de96b
fix(hierarchical): check sw supported to fit
Mar 29, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
51 changes: 41 additions & 10 deletions sklego/meta/hierarchical_predictor.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@
is_regressor,
)
from sklearn.utils.metaestimators import available_if
from sklearn.utils.validation import check_is_fitted
from sklearn.utils.validation import check_is_fitted, _check_sample_weight
from sklearn_compat.utils.validation import check_array

from sklego.common import as_list, expanding_list
Expand All @@ -26,6 +26,7 @@
min_n_obs_shrinkage,
relative_shrinkage,
)
import inspect


def _get_estimator(estimators, grp_values, grp_names, return_level, fallback_method):
Expand Down Expand Up @@ -198,6 +199,7 @@ class HierarchicalPredictor(ShrinkageMixin, MetaEstimatorMixin, BaseEstimator):

_GLOBAL_NAME = "__sklego_global_estimator__"
_TARGET_NAME = "__sklego_target_value__"
_SAMPLE_WEIGHT_NAME = "__sklego_sample_weight__"
_INDEX_NAME = "__sklego_index__"

_required_parameters = ["estimator", "groups"]
Expand All @@ -220,13 +222,13 @@ def __init__(
self.n_jobs = n_jobs
self.check_X = check_X
self.shrinkage_kwargs = shrinkage_kwargs

@property
def _estimator_type(self):
"""Computes `_estimator_type` dynamically from the wrapped model."""
return self.estimator._estimator_type

def fit(self, X, y=None):
def fit(self, X, y=None, sample_weight=None):
"""Fit one estimator for each hierarchical group of training data `X` and `y`.

Will also learn the groups that exist within the training dataset.
Expand All @@ -237,6 +239,8 @@ def fit(self, X, y=None):
Training data.
y : array-like of shape (n_samples,), default=None
Target values, if applicable.
sample_weight : array-like of shape (n_samples,), default=None
Sample weight values, if applicable.

Returns
-------
Expand Down Expand Up @@ -281,12 +285,29 @@ def fit(self, X, y=None):
if self.n_features_in_ < 1:
msg = "Found 0 features, while a minimum of 1 if required."
raise ValueError(msg)

try:
self.estimator_supports_sample_weight_ = "sample_weight" in inspect.signature(self.estimator.fit).parameters
except Exception:
self.estimator_supports_sample_weight_ = False

self.has_sw_ = sample_weight is not None

if self.has_sw_ and not self.estimator_supports_sample_weight_:
msg = f"Estimator does not support sample_weight."
raise ValueError(msg)
sample_weight = _check_sample_weight(sample_weight, X.to_native(), ensure_non_negative=True)

native_namespace = nw.get_native_namespace(X)
target_series = nw.new_series(name=self._TARGET_NAME, values=y, native_namespace=native_namespace)
global_series = nw.new_series(
name=self._GLOBAL_NAME, values=np.ones(n_samples), native_namespace=native_namespace
)
sample_weight_series = nw.new_series(
name=self._SAMPLE_WEIGHT_NAME,
values=sample_weight,
native_namespace=native_namespace,
)
if len(target_series) != n_samples:
msg = f"Found input variables with inconsistent numbers of samples: {[n_samples, len(target_series)]}"
raise ValueError(msg)
Expand All @@ -295,6 +316,7 @@ def fit(self, X, y=None):
**{
self._TARGET_NAME: target_series,
self._GLOBAL_NAME: global_series,
self._SAMPLE_WEIGHT_NAME: sample_weight_series,
}
).pipe(self.__validate_frame)

Expand Down Expand Up @@ -378,10 +400,15 @@ def _predict_estimators(self, X, method_name):

def _fit_single_estimator(self, grp_frame):
"""Shortcut to fit an estimator on a single group"""
_X = nw.to_native(grp_frame.drop([*self.groups_, self._TARGET_NAME]))
_X = nw.to_native(grp_frame.drop([*self.groups_, self._TARGET_NAME, self._SAMPLE_WEIGHT_NAME]))
_y = nw.to_native(grp_frame[self._TARGET_NAME])

return clone(self.estimator).fit(_X, _y)

args = [_X, _y]
if self.estimator_supports_sample_weight_:
_sample_weight = nw.to_native(grp_frame[self._SAMPLE_WEIGHT_NAME])
args.append(_sample_weight)

return clone(self.estimator).fit(*args)

def _fit_estimators(self, frame: nw.DataFrame):
"""Fits one estimator per level of the group column(s), and returns a dictionary of the fitted estimators.
Expand Down Expand Up @@ -506,7 +533,7 @@ class HierarchicalRegressor(RegressorMixin, HierarchicalPredictor):
```
"""

def fit(self, X, y):
def fit(self, X, y, sample_weight=None):
"""Fit one regressor for each hierarchical group of training data `X` and `y`.

Will also learn the groups that exist within the training dataset.
Expand All @@ -517,6 +544,8 @@ def fit(self, X, y):
Training data.
y : array-like of shape (n_samples,)
Target values.
sample_weight : array-like of shape (n_samples,), default=None
Sample weight values, if applicable.

Returns
-------
Expand All @@ -531,7 +560,7 @@ def fit(self, X, y):
if not is_regressor(self.estimator):
raise ValueError("The supplied estimator should be a regressor")

super().fit(X, y)
super().fit(X, y, sample_weight)
return self

def predict(self, X):
Expand Down Expand Up @@ -625,7 +654,7 @@ class HierarchicalClassifier(ClassifierMixin, HierarchicalPredictor):
```
"""

def fit(self, X, y):
def fit(self, X, y, sample_weight=None):
"""Fit one classifier for each hierarchical group of training data `X` and `y`.

Will also learn the groups that exist within the training dataset, the classes and the number of classes in the
Expand All @@ -637,6 +666,8 @@ def fit(self, X, y):
Training data.
y : array-like of shape (n_samples,)
Target values.
sample_weight : array-like of shape (n_samples,), default=None
Sample weight values, if applicable.

Returns
-------
Expand All @@ -657,7 +688,7 @@ def fit(self, X, y):
self.classes_ = np.unique(y)
self.n_classes_ = len(self.classes_)

super().fit(X, y)
super().fit(X, y, sample_weight)
return self

def predict(self, X):
Expand Down