U 2d+@sddlZddlmZddlZddlmZddlmZddl m Z m Z ddl m Z dd lmZdd lmZdd lmZdd lmZmZdd lmZmZmZGddde e e ZdS)N)Real)sparse)linprog) BaseEstimatorRegressorMixin) LinearModel)ConvergenceWarning)_safe_indexing)_check_sample_weight) sp_version parse_version)HiddenInterval StrOptionsc @seZdZUdZeeddddgeeddddgdged d d d d heedhgedgdZ ee d<ddddddddZ dddZ dS)QuantileRegressoraI Linear regression model that predicts conditional quantiles. The linear :class:`QuantileRegressor` optimizes the pinball loss for a desired `quantile` and is robust to outliers. This model uses an L1 regularization like :class:`~sklearn.linear_model.Lasso`. Read more in the :ref:`User Guide `. .. versionadded:: 1.0 Parameters ---------- quantile : float, default=0.5 The quantile that the model tries to predict. It must be strictly between 0 and 1. If 0.5 (default), the model predicts the 50% quantile, i.e. the median. alpha : float, default=1.0 Regularization constant that multiplies the L1 penalty term. fit_intercept : bool, default=True Whether or not to fit the intercept. solver : {'highs-ds', 'highs-ipm', 'highs', 'interior-point', 'revised simplex'}, default='interior-point' Method used by :func:`scipy.optimize.linprog` to solve the linear programming formulation. From `scipy>=1.6.0`, it is recommended to use the highs methods because they are the fastest ones. Solvers "highs-ds", "highs-ipm" and "highs" support sparse input data and, in fact, always convert to sparse csc. From `scipy>=1.11.0`, "interior-point" is not available anymore. .. versionchanged:: 1.4 The default of `solver` will change to `"highs"` in version 1.4. solver_options : dict, default=None Additional parameters passed to :func:`scipy.optimize.linprog` as options. If `None` and if `solver='interior-point'`, then `{"lstsq": True}` is passed to :func:`scipy.optimize.linprog` for the sake of stability. Attributes ---------- coef_ : array of shape (n_features,) Estimated coefficients for the features. intercept_ : float The intercept of the model, aka bias term. n_features_in_ : int Number of features seen during :term:`fit`. .. versionadded:: 0.24 feature_names_in_ : ndarray of shape (`n_features_in_`,) Names of features seen during :term:`fit`. Defined only when `X` has feature names that are all strings. .. versionadded:: 1.0 n_iter_ : int The actual number of iterations performed by the solver. See Also -------- Lasso : The Lasso is a linear model that estimates sparse coefficients with l1 regularization. HuberRegressor : Linear regression model that is robust to outliers. Examples -------- >>> from sklearn.linear_model import QuantileRegressor >>> import numpy as np >>> n_samples, n_features = 10, 2 >>> rng = np.random.RandomState(0) >>> y = rng.randn(n_samples) >>> X = rng.randn(n_samples, n_features) >>> # the two following lines are optional in practice >>> from sklearn.utils.fixes import sp_version, parse_version >>> solver = "highs" if sp_version >= parse_version("1.6.0") else "interior-point" >>> reg = QuantileRegressor(quantile=0.8, solver=solver).fit(X, y) >>> np.mean(y <= reg.predict(X)) 0.8 rrZneither)closedNleftbooleanhighs-ds highs-ipmhighsinterior-pointzrevised simplexwarnquantilealpha fit_interceptsolversolver_options_parameter_constraintsg?g?TcCs"||_||_||_||_||_dS)Nr)selfrrrrr r#B/tmp/pip-unpacked-wheel-zrfo1fqw/sklearn/linear_model/_quantile.py__init__s zQuantileRegressor.__init__cCs2||j||dddgddd\}}t||}|jd}|}|jrL|d7}t||j}|jdkrxt d t d }n4|jd krt t d krtd |jdt n|j}|d krt t dkrtd |dt|r|dkrtd |jd|jdkr|d krddi}n|j}t|d} t| } | t|krZ|| }t|| }t|| }ttjd||d||j|d|jg} |jrd| d<d| |<|dkrtj| |jdd} |jrttj| df|jd} tj| || | | | gdd}ntj|| | | gdd}n\t| } |jr^t| df} tj| || | | | gdd}ntj|| | | gdd}|}t| ||||d}|j}|jsdddd d!}t d"|j d#|!|j d$d%d&|j"t#|d|||d|}|j$|_%|jr"|dd|_&|d|_'n ||_&d'|_'|S)(aFit the model according to the given training data. Parameters ---------- X : {array-like, sparse matrix} of shape (n_samples, n_features) Training data. y : array-like of shape (n_samples,) Target values. sample_weight : array-like of shape (n_samples,), default=None Sample weights. Returns ------- self : object Returns self. ZcscZcsrZcooTF)Z accept_sparseZ y_numericZ multi_outputrrzThe default solver will change from 'interior-point' to 'highs' in version 1.4. Set `solver='highs'` or to the desired solver to silence this warning.r)rrrz1.6.0zSolver z* is only available with scipy>=1.6.0, got z1.11.0z- is not anymore available in SciPy >= 1.11.0.)rrrz; does not support sparse X. Use solver 'highs' for example.NZlstsqrr)Z fill_value)dtypeformat)shaper&)r')Zaxis)cA_eqb_eqmethodoptionszIteration limit reached.z!Problem appears to be infeasible.z Problem appears to be unbounded.z#Numerical difficulties encountered.)rrzDLinear programming for QuantileRegressor did not succeed. Status is z: zunknown reason zResult message of linprog: g)(Z_validate_paramsZ_validate_datar r(rnpsumrrwarningsr FutureWarningr r ValueErrorrissparser Znonzerolenr Z concatenatefullreyer&Z csc_matrixonesZhstackrxsuccessstatus setdefaultmessager ZnitZn_iter_Zcoef_Z intercept_)r"XyZ sample_weightZ n_featuresZn_paramsrrr indicesZ n_indicesr)r9r:r*r+resultZsolutionZfailureparamsr#r#r$fits             " "    zQuantileRegressor.fit)N) __name__ __module__ __qualname____doc__rrrrdictr!__annotations__r%rEr#r#r#r$rs. Z   r)r3ZnumbersrZnumpyr1ZscipyrZscipy.optimizerbaserr_baser exceptionsr utilsr Zutils.validationr Z utils.fixesr rZutils._param_validationrrrrr#r#r#r$s