sparsereg.util package

Submodules

sparsereg.util.net module

sparsereg.util.net.complexity(estimator)

sparsereg.util.net.net(estimator, x, y, attr='alpha', max_coarsity=2, filter=True, r_max=1000.0, **kw)

sparsereg.util.pipeline module

class sparsereg.util.pipeline.ColumnSelector(index=slice(None, None, None))

Bases: sklearn.base.TransformerMixin, sklearn.base.BaseEstimator

fit(x, y=None)

transform(x, y=None)

get_feature_names(input_features=None)

Module contents

sparsereg.util.dominates(a, b)

sparsereg.util.pareto_front(models, *attrs, all=False)

Simple cull. Can recursively determine all fronts.

sparsereg.util.crowding_distance(models, *attrs)

Assumes models in lexicographical sorted.

sparsereg.util.sort_non_dominated(models, *attrs, index=False)

NSGA2 based sorting

sparsereg.util.normalize(x, order=2)

sparsereg.util.cardinality(x, null=1e-09)

sparsereg.util.rmse(x)

sparsereg.util.nrmse(x, y)

class sparsereg.util.ReducedLinearModel(mask, lm)

Bases: sklearn.linear_model.base.LinearModel

fit(x, y)

Fit model.

predict(x)

Predict using the linear model

Parameters:X (array_like or sparse matrix, shape (n_samples, n_features)) – Samples. Returns:C – Returns predicted values. Return type:array, shape (n_samples,)

scores(x, y)

sparsereg.util.aic(residuals, k, correct=False)

Akaike information criterion