ProfLogitClassifier#

class empulse.models.ProfLogitClassifier(loss, C=1.0, fit_intercept=True, soft_threshold=False, l1_ratio=1.0, optimize_fn=None, optimizer_params=None, n_jobs=None)[source]#

Profit-driven logistic regression classifier.

Maximizing empirical (Expected) maximum profit score by optimizing the regression coefficients of the logistic model through a Real-coded Genetic Algorithm (RGA).

Read more in the User Guide.

Parameters:
lossCallable or empulse.metrics.Metric

Loss function to optimize.

  • If Callable it should have a signature loss(y_true, y_score).

  • If :class`~empulse.metrics.Metric`, metric parameters are passed as loss_params to the fit method.

By default, loss function is maximized, customize behaviour in optimize_fn. If the loss function in an instance of Metric then the optimization direction is automatically determined.

Cfloat, default=1.0

Inverse of regularization strength; must be a positive float. Like in support vector machines, smaller values specify stronger regularization.

fit_interceptbool, default=True

Specifies if a constant (a.k.a. bias or intercept) should be added to the decision function.

soft_thresholdbool, default=False

If True, apply soft-thresholding to the regression coefficients.

l1_ratiofloat, default=1.0

The ElasticNet mixing parameter, with 0 <= l1_ratio <= 1. For l1_ratio = 0 the penalty is a L2 penalty. For l1_ratio = 1 it is a L1 penalty. For 0 < l1_ratio < 1, the penalty is a combination of L1 and L2.

optimize_fnCallable, optional

Optimization algorithm. Should be a Callable with signature optimize(objective, X). See Profit-Driven Logistic Regression (ProfLogit) for more information.

optimizer_paramsdict[str, Any], optional

Additional keyword arguments passed to optimize_fn.

By default, the optimizer is a Real-coded Genetic Algorithm (RGA) with the following parameters:

  • max_iterint, default=1000

    Maximum number of iterations.

  • patienceint, default=250

    Number of iterations with no improvement to wait before stopping the optimization.

  • tolerancefloat, default=1e-4

    Relative tolerance to declare convergence.

  • boundstuple[float, float], default=(-5, 5)

    Lower and upper bounds for the regression coefficients.

  • all other parameters are passed to the Generation initializer.

n_jobsint, optional

Number of parallel jobs to run. None means 1 unless in a joblib.parallel_backend context. -1 means using all processors.

Attributes:
classes_numpy.ndarray

Unique classes in the target found during fit.

result_scipy.optimize.OptimizeResult

Optimization result.

coef_numpy.ndarray

Coefficients of the logit model.

intercept_float

Intercept of the logit model. Only available when fit_intercept=True.

References

[1]

Stripling, E., vanden Broucke, S., Antonio, K., Baesens, B. and Snoeck, M. (2017). Profit Maximizing Logistic Model for Customer Churn Prediction Using Genetic Algorithms. Swarm and Evolutionary Computation.

[2]

Stripling, E., vanden Broucke, S., Antonio, K., Baesens, B. and Snoeck, M. (2015). Profit Maximizing Logistic Regression Modeling for Customer Churn Prediction. IEEE International Conference on Data Science and Advanced Analytics (DSAA) (pp. 1–10). Paris, France.

Examples

from empulse.metrics import mpc_score
from empulse.models import ProfLogitClassifier
from sklearn.datasets import make_classification

X, y = make_classification()

model = ProfLogitClassifier(loss=mpc_score, C=0.1, l1_ratio=0.5)
model.fit(X, y, clv=200, incentive_cost=10)
fit(X, y, **loss_params)[source]#

Fit ProfLogit model.

Parameters:
X2D array-like, shape=(n_samples, n_features)

Training data.

y1D array-like, shape=(n_samples,)

Target values.

loss_paramsdict

Additional keyword arguments passed to loss.

Returns:
selfProfLogitClassifier

Fitted ProfLogit model.

get_metadata_routing()#

Get metadata routing of this object.

Please check User Guide on how the routing mechanism works.

Returns:
routingMetadataRequest

A MetadataRequest encapsulating routing information.

get_params(deep=True)#

Get parameters for this estimator.

Parameters:
deepbool, default=True

If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns:
paramsdict

Parameter names mapped to their values.

predict(X)#

Compute predicted labels.

Parameters:
X2D array-like, shape=(n_samples, n_dim)

Features.

Returns:
y_pred1D numpy.ndarray, shape=(n_samples,)

Predicted labels.

predict_proba(X)#

Compute predicted probabilities.

Parameters:
X2D array-like, shape=(n_samples, n_features)

Features.

Returns:
y_pred2D numpy.ndarray, shape=(n_samples, 2)

Predicted probabilities.

score(X, y, sample_weight=None)#

Return the mean accuracy on the given test data and labels.

In multi-label classification, this is the subset accuracy which is a harsh metric since you require for each sample that each label set be correctly predicted.

Parameters:
Xarray-like of shape (n_samples, n_features)

Test samples.

yarray-like of shape (n_samples,) or (n_samples, n_outputs)

True labels for X.

sample_weightarray-like of shape (n_samples,), default=None

Sample weights.

Returns:
scorefloat

Mean accuracy of self.predict(X) w.r.t. y.

set_params(**params)#

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters:
**paramsdict

Estimator parameters.

Returns:
selfestimator instance

Estimator instance.

set_score_request(*, sample_weight='$UNCHANGED$')#

Request metadata passed to the score method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

  • True: metadata is requested, and passed to score if provided. The request is ignored if metadata is not provided.

  • False: metadata is not requested and the meta-estimator will not pass it to score.

  • None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

  • str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:
sample_weightstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for sample_weight parameter in score.

Returns:
selfobject

The updated object.