sklearn GridSearchCV not using sample_weight in score function

Question

I have data with differing weights for each sample. In my application, it is important that these weights are accounted for in estimating the model and comparing alternative models. I'm using `sklearn` to estimate models and to compare alternative hyperparameter choices. But this unit test shows that `GridSearchCV` does not apply `sample_weights` to estimate scores. Is there a way to have `sklearn` use `sample_weight` to score the models? Unit test: ``` from __future__ import division import numpy as np from sklearn.datasets import load_iris from sklearn.ensemble import RandomForestClassifier from sklearn.metrics import log_loss from sklearn.model_selection import GridSearchCV, RepeatedKFold def grid_cv(X_in, y_in, w_in, cv, max_features_grid, use_weighting): out_results = dict() for k in max_features_grid: clf = RandomForestClassifier(n_estimators=256, criterion="entropy", warm_start=False, n_jobs=-1, random_state=RANDOM_STATE, max_features=k) for train_ndx, test_ndx in cv.split(X=X_…

Accepted Answer

sklearn version >= 1.4 and 1.4 nightly releases

From sklearn 1.4 (release date somewhere around October 2023), and nightly releases already available from September 2023 which you can install following guides from here, you can use the new metadata routing mechanism.

You can simply decide which objects should or should not receive metadata such as `sample_weight`, like the following script:

```
import numpy as np
from sklearn import set_config
from sklearn.datasets import load_iris
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import log_loss
from sklearn.model_selection import GridSearchCV, RepeatedKFold
from sklearn.metrics import make_scorer

set_config(enable_metadata_routing=True)

if __name__ == "__main__":
    RANDOM_STATE = 1337
    X, y = load_iris(return_X_y=True)
    sample_weight = np.array([1 + 100 * (i % 25) for i in range(len(X))])
    search_params = {"max_features": [1, 2, 3, 4]}
    cv = RepeatedKFold(n_splits=3, n_repeats=1, random_state=RANDOM_STATE)

rfc_weighted = RandomForestClassifier(
        n_estimators=256,
        criterion="entropy",
        warm_start=False,
        n_jobs=1,
        random_state=RANDOM_STATE,
    ).set_fit_request(sample_weight=True)

rfc_unweighted = RandomForestClassifier(
        n_estimators=256,
        criterion="entropy",
        warm_start=False,
        n_jobs=1,
        random_state=RANDOM_STATE,
    ).set_fit_request(sample_weight=False)

scorer_weighted = make_scorer(
        log_loss, greater_is_better=False, needs_proba=True, needs_threshold=False
    ).set_score_request(sample_weight=True)

scorer_unweighted = make_scorer(
        log_loss, greater_is_better=False, needs_proba=True, needs_threshold=False
    ).set_score_request(sample_weight=False)

for rfc, rfc_is_weighted in zip([rfc_weighted, rfc_unweighted], [True, False]):
        for scorer, scorer_is_weighted in zip(
            [scorer_weighted, scorer_unweighted], [True, False]
        ):
            grid_clf = GridSearchCV(
                estimator=rfc,
                scoring=scorer,
                cv=cv,
                param_grid=search_params,
                refit=True,
                return_train_score=False,
            )
            if rfc_is_weighted or scorer_is_weighted:
                grid_clf.fit(X, y, sample_weight=sample_weight)
            else:
                grid_clf.fit(X, y)
            print(
                "This is the best out-of-sample score using GridSearchCV with "
                f"(is scorer weighted: {scorer_is_weighted}), (is rfc weighted: "
                f"{rfc_is_weighted}): {-grid_clf.best_score_}"
            )
```

which will output:

```
This is the best out-of-sample score using GridSearchCV with (is scorer weighted: True), (is rfc weighted: True): 0.09180030568650309
This is the best out-of-sample score using GridSearchCV with (is scorer weighted: False), (is rfc weighted: True): 0.1225297422810374
This is the best out-of-sample score using GridSearchCV with (is scorer weighted: True), (is rfc weighted: False): 0.09064253271691491
This is the best out-of-sample score using GridSearchCV with (is scorer weighted: False), (is rfc weighted: False): 0.12187958644498716
```

sklearn version < 1.4

The `GridSearchCV` takes a `scoring` as input, which can be callable. You can see the details of how to change the scoring function, and also how to pass your own scoring function here. Here's the relevant piece of code from that page for the sake of completeness:

EDIT: The fit_params is passed only to the fit functions, and not the score functions. If there are parameters which are supposed to be passed to the `scorer`, they should be passed to the `make_scorer`. But that still doesn't solve the issue here, since that would mean that the whole `sample_weight` parameter would be passed to `log_loss`, whereas only the part which corresponds to `y_test` at the time of calculating the loss should be passed.

`sklearn` does NOT support such a thing, but you can hack your way through, using a `padas.DataFrame`. The good news is, `sklearn` understands a `DataFrame`, and keeps it that way. Which means you can exploit the `index` of a `DataFrame` as you see in the code here:

```
  # more code

X, y = load_iris(return_X_y=True)
  index = ['r%d' % x for x in range(len(y))]
  y_frame = pd.DataFrame(y, index=index)
  sample_weight = np.array([1 + 100 * (i % 25) for i in range(len(X))])
  sample_weight_frame = pd.DataFrame(sample_weight, index=index)

# more code

def score_f(y_true, y_pred, sample_weight):
      return log_loss(y_true.values, y_pred,
                      sample_weight=sample_weight.loc[y_true.index.values].values.reshape(-1),
                      normalize=True)

score_params = {"sample_weight": sample_weight_frame}
  my_scorer = make_scorer(score_f,
                          greater_is_better=False, 
                          needs_proba=True, 
                          needs_thresho…

sklearn GridSearchCV not using sample_weight in score function

Solution

Problem