ó ‡ˆ\c@ sdZddlmZddlmZmZddlZddlZddlZ ddl j Z ddl mZddl m Z ddlmZdd lmZmZdd lmZmZmZdd lmZmZdd lmZdd lmZddlmZddl m!Z!m"Z"ddl#m$Z$ddl%m&Z&m'Z'ddlm(Z(ddl)m*Z*ddl+m,Z-dZ.dd„Z0e1e2de1e2d„Z3d„Z4dej5eeƒfd„ƒYZ6defd„ƒYZ7de8fd„ƒYZ9de6efd „ƒYZ:e2d!„Z;dS("s Generalized Linear models. iÿÿÿÿ(tdivision(tABCMetatabstractmethodN(tlinalg(tsparsei(tsix(tParalleltdelayed(t BaseEstimatortClassifierMixintRegressorMixin(t check_arrayt check_X_y(t FLOAT_DTYPES(tcheck_random_state(tsafe_sparse_dot(tmean_variance_axistinplace_column_scale(t sparse_lsqr(t ArrayDatasett CSRDataset(tcheck_is_fitted(tNotFittedError(t normalizeg{®Gáz„?cC s”t|ƒ}|jdtjtjƒjƒ}tj|ƒrlt|j |j |j ||d|ƒ}t }nt |||d|ƒ}d}||fS(sCreate ``Dataset`` abstraction for sparse and dense inputs. This also returns the ``intercept_decay`` which is different for sparse datasets. Parameters ---------- X : array_like, shape (n_samples, n_features) Training data y : array_like, shape (n_samples, ) Target values. sample_weight : numpy array of shape (n_samples,) The weight of each sample random_state : int, RandomState instance or None (default) Determines random number generation for dataset shuffling and noise. Pass an int for reproducible output across multiple function calls. See :term:`Glossary `. Returns ------- dataset The ``Dataset`` abstraction intercept_decay The intercept decay itseedgð?(Rtrandinttnptiinfotint32tmaxtsptissparseRtdatatindptrtindicestSPARSE_INTERCEPT_DECAYR(tXtyt sample_weightt random_statetrngRtdatasettintercept_decay((s8lib/python2.7/site-packages/sklearn/linear_model/base.pyt make_dataset0s !  c C st|tjƒrd}n|rHt|d|dddgdtƒ}n9|rtj|ƒrl|jƒ}q|jddƒ}nt j |d|j ƒ}|rítj|ƒrTt |dd ƒ\}} |sâ|j j d ƒ|(n|r2| |jd 9} t j| | ƒ} ~ d | | d k0 means this class would be predicted. RVs(This %(name)s instance is not fitted yettnameR-R.is*X has %d features per sample; expecting %dRXN(thasattrRVR9RR;R_R R<t ValueErrorRRYRBRZtravel(RTR$t n_featurestscores((s8lib/python2.7/site-packages/sklearn/linear_model/base.pytdecision_functionëscC s\|j|ƒ}t|jƒdkr?|dkjtjƒ}n|jddƒ}|j|S(s&Predict class labels for samples in X. Parameters ---------- X : array_like or sparse matrix, shape (n_samples, n_features) Samples. Returns ------- C : array, shape [n_samples] Predicted class label per sample. iiR3(RjtlenR<tastypeRtinttargmaxtclasses_(RTR$RiR"((s8lib/python2.7/site-packages/sklearn/linear_model/base.pyR\ s cC s |j|ƒ}|d9}tj||ƒ|d7}tj||ƒ|jdkrltjd||gƒjS||jddƒj|j ddfƒ}|SdS(sîProbability estimation for OvR logistic regression. Positive class probabilities are computed as 1. / (1. + np.exp(-self.decision_function(X))); multiclass is handled by normalizing that over all classes. iÿÿÿÿiR3iN( RjRtexpt reciprocalRDtvstackRYtsumtreshapeR<(RTR$tprob((s8lib/python2.7/site-packages/sklearn/linear_model/base.pyt_predict_proba_lr s  ,(R_R`RaRjR\Rv(((s8lib/python2.7/site-packages/sklearn/linear_model/base.pyRcås ! tSparseCoefMixincB s eZdZd„Zd„ZRS(slMixin for converting coef_ to and from CSR format. L1-regularizing estimators should inherit this. cC sDd}t|dd|ƒtj|jƒr@|jjƒ|_n|S(s…Convert coefficient matrix to dense array format. Converts the ``coef_`` member (back) to a numpy.ndarray. This is the default format of ``coef_`` and is required for fitting, so calling this method is only required on models that have previously been sparsified; otherwise, it is a no-op. Returns ------- self : estimator s6Estimator, %(name)s, must be fitted before densifying.RVtmsg(RRRRVttoarray(RTRx((s8lib/python2.7/site-packages/sklearn/linear_model/base.pytdensify:s cC s2d}t|dd|ƒtj|jƒ|_|S(svConvert coefficient matrix to sparse format. Converts the ``coef_`` member to a scipy.sparse matrix, which for L1-regularized models can be much more memory- and storage-efficient than the usual numpy.ndarray representation. The ``intercept_`` member is not converted. Notes ----- For non-sparse models, i.e. when there are not many zeros in ``coef_``, this may actually *increase* memory usage, so use this method with care. A rule of thumb is that the number of zero elements, which can be computed with ``(coef_ == 0).sum()``, must be more than 50% for this to provide significant benefits. After calling this method, further fitting with the partial_fit method (if any) will not work until you call densify. Returns ------- self : estimator s7Estimator, %(name)s, must be fitted before sparsifying.RVRx(RRt csr_matrixRV(RTRx((s8lib/python2.7/site-packages/sklearn/linear_model/base.pytsparsifyLs(R_R`RaRzR|(((s8lib/python2.7/site-packages/sklearn/linear_model/base.pyRw4s tLinearRegressioncB s/eZdZeeedd„Zdd„ZRS(s Ordinary least squares Linear Regression. Parameters ---------- fit_intercept : boolean, optional, default True whether to calculate the intercept for this model. If set to False, no intercept will be used in calculations (e.g. data is expected to be already centered). normalize : boolean, optional, default False This parameter is ignored when ``fit_intercept`` is set to False. If True, the regressors X will be normalized before regression by subtracting the mean and dividing by the l2-norm. If you wish to standardize, please use :class:`sklearn.preprocessing.StandardScaler` before calling ``fit`` on an estimator with ``normalize=False``. copy_X : boolean, optional, default True If True, X will be copied; else, it may be overwritten. n_jobs : int or None, optional (default=None) The number of jobs to use for the computation. This will only provide speedup for n_targets > 1 and sufficient large problems. ``None`` means 1 unless in a :obj:`joblib.parallel_backend` context. ``-1`` means using all processors. See :term:`Glossary ` for more details. Attributes ---------- coef_ : array, shape (n_features, ) or (n_targets, n_features) Estimated coefficients for the linear regression problem. If multiple targets are passed during the fit (y 2D), this is a 2D array of shape (n_targets, n_features), while if only one target is passed, this is a 1D array of length n_features. intercept_ : array Independent term in the linear model. Examples -------- >>> import numpy as np >>> from sklearn.linear_model import LinearRegression >>> X = np.array([[1, 1], [1, 2], [2, 2], [2, 3]]) >>> # y = 1 * x_0 + 2 * x_1 + 3 >>> y = np.dot(X, np.array([1, 2])) + 3 >>> reg = LinearRegression().fit(X, y) >>> reg.score(X, y) 1.0 >>> reg.coef_ array([1., 2.]) >>> reg.intercept_ # doctest: +ELLIPSIS 3.0000... >>> reg.predict(np.array([[3, 5]])) array([16.]) Notes ----- From the implementation point of view, this is just plain Ordinary Least Squares (scipy.linalg.lstsq) wrapped as a predictor object. cC s(||_||_||_||_dS(N(RERtcopy_Xtn_jobs(RTRERR~R((s8lib/python2.7/site-packages/sklearn/linear_model/base.pyt__init__ªs   c  s |j}tˆˆddddgdtdtƒ\‰‰|dk rltj|ƒjdkrltdƒ‚n|jˆˆd |j d |j d |j d |ƒ\‰‰}}}|dk rÕt ˆˆ|ƒ\‰‰nt jˆƒr©ˆjd krtˆˆƒ}|d|_|d|_qâtd|ƒ‡‡fd†tˆjdƒDƒƒ} tjg| D]}|d^qaƒ|_tjg| D]}|d^qŠƒ|_n9tjˆˆƒ\|_|_|_|_|jj|_ˆjdkr tj|jƒ|_n|j|||ƒ|S(sC Fit linear model. Parameters ---------- X : array-like or sparse matrix, shape (n_samples, n_features) Training data y : array_like, shape (n_samples, n_targets) Target values. Will be cast to X's dtype if necessary sample_weight : numpy array of shape [n_samples] Individual weights for each sample .. versionadded:: 0.17 parameter *sample_weight* support to LinearRegression. Returns ------- self : returns an instance of self. R-R.R/RWt y_numerict multi_outputis)Sample weights must be 1D array or scalarRERR,R&iiiRc3 s:|]0}ttƒˆˆdd…|fjƒƒVqdS(N(RRRg(t.0tj(R$R%(s8lib/python2.7/site-packages/sklearn/linear_model/base.pys ßsN(RR RBR9Rt atleast_1dRDRfRLRERR~RRRRRRVt _residuesRtrangeR<RrRtlstsqtrank_t singular_RYRgR^( RTR$R%R&tn_jobs_RHRKRJtouttouts((R$R%s8lib/python2.7/site-packages/sklearn/linear_model/base.pyRU±s4 $!   ),*N(R_R`RaRBRAR9R€RU(((s8lib/python2.7/site-packages/sklearn/linear_model/base.pyR}js>  c C sŠ|j\}} tj|ƒrct}t||d|d|dtdtd|ƒ\}}} } } n6t||d|d|d|d|ƒ\}}} } } t|dƒr |rÊtj| tj | ƒƒ sì|r tj| tj | ƒƒ r t j dt ƒd}d}nt|tjƒr8|dkr8|| k}n|tkr‡tjd | | fd |jd d ƒ}tj|j|d |ƒnt|dƒsŸd}nt|dƒrq|dkrqtj|j|jggƒ} |jdkr!tjd | d | d d ƒ}tj|j|d |ƒqq|jd}tjd | |fd | d dƒ}tj|j|d |jƒn||| | | ||fS(s6Aux function used at beginning of fit in linear modelsRERR,RFRGt __array__slGram matrix was provided but X was centered to fit intercept, or X was normalized : recomputing Gram matrix.tautoR<R0R1tCRŒitFN(R<Rt isspmatrixRARLRBReRtallcloseRCR>twarningstwarnt UserWarningR9R6Rt string_typestemptyR0R]RYtfind_common_typeRD(R$R%tXyt precomputeRRER,RGRPRhRHRKRJt common_dtypet n_targets((s8lib/python2.7/site-packages/sklearn/linear_model/base.pyt_pre_fitîsD'""       (<Rat __future__RtabcRRR7R”tnumpyRt scipy.sparseRRtscipyRt externalsRt utils._joblibRRtbaseRR R tutilsR R tutils.validationR Rt utils.extmathRtutils.sparsefuncsRRt utils.fixesRtutils.seq_datasetRRRt exceptionsRtpreprocessing.dataRR@R#R9R+RARBRLRRtwith_metaclassRSRctobjectRwR}Rž(((s8lib/python2.7/site-packages/sklearn/linear_model/base.pyts>     -O "+O6…