<a id="skfolio-metrics-make-scorer"></a>

# skfolio.metrics.make_scorer

<a id="skfolio.metrics.make_scorer"></a>

### skfolio.metrics.make_scorer(score_func, greater_is_better=None, response_method='predict', \*\*kwargs)

Make a scorer from a [measure](https://skfolio.org/api.html.md#measures-ref), a portfolio score
function, or a non-predictor estimator score function.

This function wraps scoring functions for use in model selection:

* `response_method="predict"` (default): for portfolio optimization
  estimators (e.g.
  [`MeanRisk`](https://skfolio.org/generated/skfolio.optimization.MeanRisk.html.md#skfolio.optimization.MeanRisk)). Compatible with
  `GridSearchCV` and
  `cross_val_score`.
* `response_method=None`: for non-predictor estimators
  (covariance, expected returns, prior) that implement `fit` but not
  `predict`. Compatible with both sklearn cross-validation utilities
  and skfolio online utilities
  ([`OnlineGridSearch`](https://skfolio.org/generated/skfolio.model_selection.OnlineGridSearch.html.md#skfolio.model_selection.OnlineGridSearch),
  [`online_score`](https://skfolio.org/generated/skfolio.model_selection.online_score.html.md#skfolio.model_selection.online_score)).

#### NOTE
For online evaluation of portfolio optimization estimators, pass a
[measure](https://skfolio.org/api.html.md#measures-ref) directly to the `scoring` parameter instead of using
`make_scorer`. Online evaluation scores the full
aggregated [`MultiPeriodPortfolio`](https://skfolio.org/generated/skfolio.portfolio.MultiPeriodPortfolio.html.md#skfolio.portfolio.MultiPeriodPortfolio) rather than averaging
per-fold scores.

* **Parameters:**
  **score_func** *Measure | callable*
  : If `score_func` is a [measure](https://skfolio.org/api.html.md#measures-ref), we return the
    measure of the predicted [`Portfolio`](https://skfolio.org/generated/skfolio.portfolio.Portfolio.html.md#skfolio.portfolio.Portfolio) times
    `1` or `-1` depending on `greater_is_better`.
    `response_method` must be `"predict"` in this case.
    <br/>
    If `response_method="predict"`, `score_func` must be a score
    function (or loss function) with signature
    `score_func(pred, **kwargs)` where `pred` is the predicted
    [`Portfolio`](https://skfolio.org/generated/skfolio.portfolio.Portfolio.html.md#skfolio.portfolio.Portfolio).
    <br/>
    If `response_method=None`, `score_func` must be a score function
    (or loss function) with signature
    `score_func(estimator, X_test, **kwargs)` where `estimator` is
    the fitted non-predictor estimator and `X_test` the realized
    returns.

  **greater_is_better** *bool, optional*
  : Whether `score_func` is a score function (high is good) or a loss
    function (low is good).  In the latter case the scorer sign-flips
    the outcome so that higher values always indicate a better model.
    The default (`None`) is:
    * If `score_func` is a [measure](https://skfolio.org/api.html.md#measures-ref):
      > * `True` for [`PerfMeasure`](https://skfolio.org/generated/skfolio.measures.PerfMeasure.html.md#skfolio.measures.PerfMeasure) and
      >   [`RatioMeasure`](https://skfolio.org/generated/skfolio.measures.RatioMeasure.html.md#skfolio.measures.RatioMeasure).
      > * `False` for [`RiskMeasure`](https://skfolio.org/generated/skfolio.measures.RiskMeasure.html.md#skfolio.measures.RiskMeasure) and
      >   [`ExtraRiskMeasure`](https://skfolio.org/generated/skfolio.measures.ExtraRiskMeasure.html.md#skfolio.measures.ExtraRiskMeasure).
    * Otherwise, `True`.

  **response_method** *str or None, default=”predict”*
  : Determines how the scorer obtains predictions. Only `"predict"` and
    `None` are supported:
    * `"predict"`: call `estimator.predict(X)` and pass the resulting
      [`Portfolio`](https://skfolio.org/generated/skfolio.portfolio.Portfolio.html.md#skfolio.portfolio.Portfolio) to `score_func`. Use for
      portfolio optimization estimators (e.g.
      [`MeanRisk`](https://skfolio.org/generated/skfolio.optimization.MeanRisk.html.md#skfolio.optimization.MeanRisk)).
    * `None`: pass `(estimator, X_test)` directly to `score_func`
      without calling any response method. Use for non-predictor
      estimators (covariance, expected returns, prior).

  **\*\*kwargs** *additional arguments*
  : Additional parameters to be passed to `score_func`.
* **Returns:**
  **scorer** *callable*
  : Callable object with signature `scorer(estimator, X, y=None)`
    that returns a scalar score (higher is better).

### Examples

Portfolio scorer from a measure:

```pycon
>>> from skfolio.measures import RatioMeasure
>>> scorer = make_scorer(RatioMeasure.SHARPE_RATIO)
```

Portfolio scorer from a custom function:

```pycon
>>> def custom(pred):
...     return pred.mean - 2 * pred.variance
>>> scorer = make_scorer(custom)
```

Non-predictor estimator scorer for covariance evaluation:

```pycon
>>> from skfolio.metrics import portfolio_variance_qlike_loss
>>> import numpy as np
>>> scorer = make_scorer(
...     portfolio_variance_qlike_loss,
...     greater_is_better=False,
...     response_method=None,
...     portfolio_weights=np.ones(20) / 20,
... )
```