ó ‡ˆ\c@ s!dZddlmZddlZddlZddlZddlmZm Z ddl m Z ddl m Z mZdd l mZmZdd lmZmZdd lmZdd lmZdd lmZmZddlmZmZddlmZm Z m!Z!m"Z"ddlm#Z#m$Z$ddl%m&Z&ddl'm(Z(ddl)m*Z*ddl+m,Z,m-Z-ddgZ.ej/ej0ƒj1Z2d„Z3d„Z4d„Z5d„Z6d„Z7d„Z8d„Z9deee ƒfd„ƒYZ:de:efd „ƒYZ;de:efd!„ƒYZ<dS("sBagging meta-estimator.iÿÿÿÿ(tdivisionN(tABCMetatabstractmethod(twarni(t BaseEnsemblet_partition_estimatorsi(tClassifierMixintRegressorMixin(tParalleltdelayed(twith_metaclass(tzip(tr2_scoretaccuracy_score(tDecisionTreeClassifiertDecisionTreeRegressor(tcheck_random_statet check_X_yt check_arrayt column_or_1d(tindices_to_masktcheck_consistent_length(tif_delegate_has_method(tcheck_classification_targets(tsample_without_replacement(thas_fit_parametertcheck_is_fittedtBaggingClassifiertBaggingRegressorcC s7|r|jd||ƒ}nt||d|ƒ}|S(sDraw randomly sampled indices.it random_state(trandintR(Rt bootstrapt n_populationt n_samplestindices((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyt_generate_indices#s   c C s@t|ƒ}t||||ƒ}t||||ƒ}||fS(s)Randomly draw feature and sample indices.(RR#( Rtbootstrap_featurestbootstrap_samplest n_featuresR!t max_featurest max_samplestfeature_indicestsample_indices((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyt_generate_bagging_indices/s      c C së|j\}} |j} |j} |j} |j} t|jdƒ}| rg|d k rgtdƒ‚ng}g}xkt |ƒD]]}|dkr«d|d||fGHnt j j ||ƒ}|j dtd|ƒ}t|| | | || | ƒ\}}|r›|d kr$t j|fƒ}n |jƒ}| rXt j|d|ƒ}||9}nt||ƒ}d||<|j|d d …|f|d|ƒn(|j||d d …|f||ƒ|j|ƒ|j|ƒq€W||fS( sBPrivate function used to build a batch of estimators within a job.t sample_weights0The base estimator doesn't support sample weightis?Building estimator %d of %d for this parallel run (total %d)...tappendRt minlengthiN(tshapet _max_featurest _max_samplesRR$Rtbase_estimator_tNonet ValueErrortrangetnptrandomt RandomStatet_make_estimatortFalseR+tonestcopytbincountRtfitR-(t n_estimatorstensembletXtyR,tseedsttotal_n_estimatorstverboseR!R&R'R(RR$tsupport_sample_weightt estimatorstestimators_featurestiRt estimatortfeaturesR"tcurr_sample_weightt sample_countstnot_indices_mask((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyt_parallel_build_estimators?sJ            )( c C s'|jd}tj||fƒ}xþt||ƒD]í\}}t|dƒrÏ|j|dd…|fƒ}|t|jƒkrŽ||7}q|dd…|jfc|dd…tt|jƒƒf7¯s(tsumR (RGRHRA((RAs7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyt_parallel_decision_function­sc s#t‡fd†t||ƒDƒƒS(s:Private function used to compute predictions within a job.c3 s4|]*\}}|jˆdd…|fƒVqdS(N(RU(RjRJRK(RA(s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pys ¶s(RkR (RGRHRA((RAs7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyt_parallel_predict_regression´st BaseBaggingc B sŒeZdZed dddeeeed d dd„ ƒZd d„Zd d d d„Z ed„ƒZ d„Z d „Z e d „ƒZRS( s‚Base class for Bagging meta-estimator. Warning: This class should not be used directly. Use derived classes instead. i gð?ic C sttt|ƒjd|d|ƒ||_||_||_||_||_||_| |_ | |_ | |_ dS(Ntbase_estimatorR?( tsuperRnt__init__R(R'RR$t oob_scoret warm_starttn_jobsRRE( tselfRoR?R(R'RR$RrRsRtRRE((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyRqÂs         cC s|j|||jd|ƒS(s Build a Bagging ensemble of estimators from the training set (X, y). Parameters ---------- X : {array-like, sparse matrix} of shape = [n_samples, n_features] The training input samples. Sparse matrices are accepted only if they are supported by the base estimator. y : array-like, shape = [n_samples] The target values (class labels in classification, real numbers in regression). sample_weight : array-like, shape = [n_samples] or None Sample weights. If None, then samples are equally weighted. Note that this is supported only if the base estimator supports sample weighting. Returns ------- self : object R,(t_fitR((RuRARBR,((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyR>Ýsc  s#tˆjƒ}tˆˆddgdddtdtƒ\‰‰ˆdk rmtˆdtƒ‰tˆˆƒnˆj\}ˆ_ |ˆ_ ˆj ˆƒ‰ˆj ƒ|dk r¼|ˆj _n|dkrÔˆj}n5t|tjtjfƒs t|ˆjdƒ}nd|ko'ˆjdkns;tdƒ‚n|ˆ_tˆjtjtjfƒrnˆj}n4tˆjtjƒr–ˆjˆj }n td ƒ‚d|ko¼ˆj knsÐtd ƒ‚ntd t|ƒƒ}|ˆ_ˆj rˆjrtd ƒ‚nˆjr1ˆjr1td ƒ‚ntˆdƒrRˆjrRˆ` nˆj sltˆdƒ rgˆ_!gˆ_"nˆj#t$ˆj!ƒ} | dkrÈtdˆj#t$ˆj!ƒfƒ‚n| dkrât%dƒˆSt&| ˆj'ƒ\} ‰‰t(ˆƒ‰ˆjrFt$ˆj!ƒdkrF|j)t*dt$ˆj!ƒƒn|j)t*d| ƒ‰ˆˆ_+t,d| dˆj-ƒ‡‡‡‡‡‡‡‡fd†t.| ƒDƒƒ} ˆj!t/t0j1j2d„| Dƒƒƒ7_!ˆj"t/t0j1j2d„| Dƒƒƒ7_"ˆjrˆj3ˆˆƒnˆS(sBBuild a Bagging ensemble of estimators from the training set (X, y). Parameters ---------- X : {array-like, sparse matrix} of shape = [n_samples, n_features] The training input samples. Sparse matrices are accepted only if they are supported by the base estimator. y : array-like, shape = [n_samples] The target values (class labels in classification, real numbers in regression). max_samples : int or float, optional (default=None) Argument to use instead of self.max_samples. max_depth : int, optional (default=None) Override value used when constructing base estimator. Only supported if the base estimator has a max_depth parameter. sample_weight : array-like, shape = [n_samples] or None Sample weights. If None, then samples are equally weighted. Note that this is supported only if the base estimator supports sample weighting. Returns ------- self : object tcsrtcscR[tforce_all_finitet multi_outputt ensure_2dis%max_samples must be in (0, n_samples]s!max_features must be int or floats'max_features must be in (0, n_features]is6Out of bag estimation only available if bootstrap=Trues6Out of bag estimate only available if warm_start=Falset oob_score_t estimators_sTn_estimators=%d must be larger or equal to len(estimators_)=%d when warm_start==TruesJWarm-start fitting without increasing n_estimators does not fit new trees.tsizeRtREc 3 sS|]I}ttƒˆ|ˆˆˆˆˆˆ|ˆ|d!ˆdˆjƒVqdS(iREN(R RORE(RjRI(RAR?R,RCRutstartsRDRB(s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pys qs cs s|]}|dVqdS(iN((Rjtt((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pys ~scs s|]}|dVqdS(iN((RjR€((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pys €sN(4RRRR3R:tTrueRRR/t n_features_t _n_samplest _validate_yt_validate_estimatorR2t max_depthR(t isinstancetnumberstIntegralR6tintegerR`R4R1R'tfloattmaxR0RRrRsRRR|R}testimators_features_R?RSRRRtRkRtMAX_INTt_seedsRRER5tlistt itertoolstchaint from_iterablet_set_oob_score( RuRARBR(R†R,RR!R'tn_more_estimatorsRtt all_results((RAR?R,RCRuRRDRBs7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyRvös|      #            !  cC sdS(s+Calculate out of bag predictions and score.N((RuRARB((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyR”‡scC s@t|jƒdks(|jddkr8t|dtƒS|SdS(NiR(RSR/RR(RuRB((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyR„‹s(c c skxd|jD]Y}tjj|ƒ}t||j|j|j|j|j |j ƒ\}}||fVq WdS(N( RR6R7R8R+R$RR‚RƒR0R1(RutseedRR)R*((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyt_get_estimators_indices‘scC s#g|jƒD]\}}|^q S(sÂThe subset of drawn samples for each base estimator. Returns a dynamically generated list of indices identifying the samples used for fitting each member of the ensemble, i.e., the in-bag samples. Note: the list is re-created at each call to the property in order to reduce the object memory footprint by not storing the sampling data. Thus fetching the property may be slower than expected. (R˜(Rut_R*((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pytestimators_samples_žs N(t__name__t __module__t__doc__RR3RR:RqR>RvR”R„R˜tpropertyRš(((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyRn»s& ‘  c B s†eZdZddddeeeedddd„ Zd„Zd„Zd„Z d„Z d „Z d „Z e d d ƒd „ƒZRS(s A Bagging classifier. A Bagging classifier is an ensemble meta-estimator that fits base classifiers each on random subsets of the original dataset and then aggregate their individual predictions (either by voting or by averaging) to form a final prediction. Such a meta-estimator can typically be used as a way to reduce the variance of a black-box estimator (e.g., a decision tree), by introducing randomization into its construction procedure and then making an ensemble out of it. This algorithm encompasses several works from the literature. When random subsets of the dataset are drawn as random subsets of the samples, then this algorithm is known as Pasting [1]_. If samples are drawn with replacement, then the method is known as Bagging [2]_. When random subsets of the dataset are drawn as random subsets of the features, then the method is known as Random Subspaces [3]_. Finally, when base estimators are built on subsets of both samples and features, then the method is known as Random Patches [4]_. Read more in the :ref:`User Guide `. Parameters ---------- base_estimator : object or None, optional (default=None) The base estimator to fit on random subsets of the dataset. If None, then the base estimator is a decision tree. n_estimators : int, optional (default=10) The number of base estimators in the ensemble. max_samples : int or float, optional (default=1.0) The number of samples to draw from X to train each base estimator. - If int, then draw `max_samples` samples. - If float, then draw `max_samples * X.shape[0]` samples. max_features : int or float, optional (default=1.0) The number of features to draw from X to train each base estimator. - If int, then draw `max_features` features. - If float, then draw `max_features * X.shape[1]` features. bootstrap : boolean, optional (default=True) Whether samples are drawn with replacement. If False, sampling without replacement is performed. bootstrap_features : boolean, optional (default=False) Whether features are drawn with replacement. oob_score : bool, optional (default=False) Whether to use out-of-bag samples to estimate the generalization error. warm_start : bool, optional (default=False) When set to True, reuse the solution of the previous call to fit and add more estimators to the ensemble, otherwise, just fit a whole new ensemble. See :term:`the Glossary `. .. versionadded:: 0.17 *warm_start* constructor parameter. n_jobs : int or None, optional (default=None) The number of jobs to run in parallel for both `fit` and `predict`. ``None`` means 1 unless in a :obj:`joblib.parallel_backend` context. ``-1`` means using all processors. See :term:`Glossary ` for more details. random_state : int, RandomState instance or None, optional (default=None) If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. verbose : int, optional (default=0) Controls the verbosity when fitting and predicting. Attributes ---------- base_estimator_ : estimator The base estimator from which the ensemble is grown. estimators_ : list of estimators The collection of fitted base estimators. estimators_samples_ : list of arrays The subset of drawn samples (i.e., the in-bag samples) for each base estimator. Each subset is defined by an array of the indices selected. estimators_features_ : list of arrays The subset of drawn features for each base estimator. classes_ : array of shape = [n_classes] The classes labels. n_classes_ : int or list The number of classes. oob_score_ : float Score of the training dataset obtained using an out-of-bag estimate. oob_decision_function_ : array of shape = [n_samples, n_classes] Decision function computed with out-of-bag estimate on the training set. If n_estimators is small it might be possible that a data point was never left out during the bootstrap. In this case, `oob_decision_function_` might contain NaN. References ---------- .. [1] L. Breiman, "Pasting small votes for classification in large databases and on-line", Machine Learning, 36(1), 85-103, 1999. .. [2] L. Breiman, "Bagging predictors", Machine Learning, 24(2), 123-140, 1996. .. [3] T. Ho, "The random subspace method for constructing decision forests", Pattern Analysis and Machine Intelligence, 20(8), 832-844, 1998. .. [4] G. Louppe and P. Geurts, "Ensembles on Random Patches", Machine Learning and Knowledge Discovery in Databases, 346-361, 2012. i gð?ic C sVtt|ƒj|d|d|d|d|d|d|d|d| d | d | ƒ dS( NR?R(R'RR$RrRsRtRRE(RpRRq( RuRoR?R(R'RR$RrRsRtRRE((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyRq)s cC s tt|ƒjdtƒƒdS(s:Check the estimator and set the base_estimator_ attribute.tdefaultN(RpRR…R(Ru((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyR…Csc C sÆ|jd}|j}tj||fƒ}xt|j|j|jƒD]ó\}}}t||ƒ} t |dƒr½|| dd…fc|j || dd…fdd…|fƒ7³sN(RRR3R:R‚R/R4tformatRR?RtRRER5Rk(RuRARtR?t all_probaRW((RARuRs7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyRP‹s    c sJtˆdƒtˆjdƒr0tˆdddgdddtƒ‰ˆjˆjdkr~td j ˆjˆjdƒƒ‚nt ˆj ˆj ƒ\}}‰t d |d ˆjƒ‡‡‡fd †t|ƒDƒƒ}|d }x3tdt|ƒƒD]}tj|||ƒ}qöW|tjˆj ƒ8}|StjˆjˆƒƒSdS(sÈPredict class log-probabilities for X. The predicted class log-probabilities of an input sample is computed as the log of the mean predicted class probabilities of the base estimators in the ensemble. Parameters ---------- X : {array-like, sparse matrix} of shape = [n_samples, n_features] The training input samples. Sparse matrices are accepted only if they are supported by the base estimator. Returns ------- p : array of shape = [n_samples, n_classes] The class log-probabilities of the input samples. The order of the classes corresponds to that in the attribute `classes_`. RTRaR¯RwRxR[RyisjNumber of features of the model must match the input. Model n_features is {0} and input n_features is {1} RtREc3 sY|]O}ttƒˆjˆ|ˆ|d!ˆjˆ|ˆ|d!ˆˆjƒVqdS(iN(R RhR}RR¡(RjRI(RARuR(s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pys åsiN(RRRR2RR3R:R‚R/R4R°RR?RtRRER5RSR6RbtlogRP(RuRARtR?t all_log_probaRdR©((RARuRs7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyRa¿s&    tdelegateRoc sÛtˆdƒtˆdddgdd dtƒ‰ˆjˆjdkrltdjˆjˆjdƒƒ‚ntˆj ˆj ƒ\}}‰t d |d ˆj ƒ‡‡‡fd †t |ƒDƒƒ}t|ƒˆj }|S( s­Average of the decision functions of the base classifiers. Parameters ---------- X : {array-like, sparse matrix} of shape = [n_samples, n_features] The training input samples. Sparse matrices are accepted only if they are supported by the base estimator. Returns ------- score : array, shape = [n_samples, k] The decision function of the input samples. The columns correspond to the classes in sorted order, as they appear in the attribute ``classes_``. Regression and binary classification are special cases with ``k == 1``, otherwise ``k==n_classes``. RTR¯RwRxR[RyisjNumber of features of the model must match the input. Model n_features is {0} and input n_features is {1} RtREc3 sS|]I}ttƒˆjˆ|ˆ|d!ˆjˆ|ˆ|d!ˆƒVqdS(iN(R RlR}R(RjRI(RARuR(s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pys sN(RRR3R:R‚R/R4R°RR?RtRRER5Rk(RuRARtR?t all_decisionst decisions((RARuRs7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyRiùs    N(R›RœRR3RR:RqR…R”R„RURPRaRRi(((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyR®s&z   %   4 :c B sSeZdZddddeeeedddd„ Zd„Zd„Zd„Z RS( sšA Bagging regressor. A Bagging regressor is an ensemble meta-estimator that fits base regressors each on random subsets of the original dataset and then aggregate their individual predictions (either by voting or by averaging) to form a final prediction. Such a meta-estimator can typically be used as a way to reduce the variance of a black-box estimator (e.g., a decision tree), by introducing randomization into its construction procedure and then making an ensemble out of it. This algorithm encompasses several works from the literature. When random subsets of the dataset are drawn as random subsets of the samples, then this algorithm is known as Pasting [1]_. If samples are drawn with replacement, then the method is known as Bagging [2]_. When random subsets of the dataset are drawn as random subsets of the features, then the method is known as Random Subspaces [3]_. Finally, when base estimators are built on subsets of both samples and features, then the method is known as Random Patches [4]_. Read more in the :ref:`User Guide `. Parameters ---------- base_estimator : object or None, optional (default=None) The base estimator to fit on random subsets of the dataset. If None, then the base estimator is a decision tree. n_estimators : int, optional (default=10) The number of base estimators in the ensemble. max_samples : int or float, optional (default=1.0) The number of samples to draw from X to train each base estimator. - If int, then draw `max_samples` samples. - If float, then draw `max_samples * X.shape[0]` samples. max_features : int or float, optional (default=1.0) The number of features to draw from X to train each base estimator. - If int, then draw `max_features` features. - If float, then draw `max_features * X.shape[1]` features. bootstrap : boolean, optional (default=True) Whether samples are drawn with replacement. If False, sampling without replacement is performed. bootstrap_features : boolean, optional (default=False) Whether features are drawn with replacement. oob_score : bool Whether to use out-of-bag samples to estimate the generalization error. warm_start : bool, optional (default=False) When set to True, reuse the solution of the previous call to fit and add more estimators to the ensemble, otherwise, just fit a whole new ensemble. See :term:`the Glossary `. n_jobs : int or None, optional (default=None) The number of jobs to run in parallel for both `fit` and `predict`. ``None`` means 1 unless in a :obj:`joblib.parallel_backend` context. ``-1`` means using all processors. See :term:`Glossary ` for more details. random_state : int, RandomState instance or None, optional (default=None) If int, random_state is the seed used by the random number generator; If RandomState instance, random_state is the random number generator; If None, the random number generator is the RandomState instance used by `np.random`. verbose : int, optional (default=0) Controls the verbosity when fitting and predicting. Attributes ---------- estimators_ : list of estimators The collection of fitted sub-estimators. estimators_samples_ : list of arrays The subset of drawn samples (i.e., the in-bag samples) for each base estimator. Each subset is defined by an array of the indices selected. estimators_features_ : list of arrays The subset of drawn features for each base estimator. oob_score_ : float Score of the training dataset obtained using an out-of-bag estimate. oob_prediction_ : array of shape = [n_samples] Prediction computed with out-of-bag estimate on the training set. If n_estimators is small it might be possible that a data point was never left out during the bootstrap. In this case, `oob_prediction_` might contain NaN. References ---------- .. [1] L. Breiman, "Pasting small votes for classification in large databases and on-line", Machine Learning, 36(1), 85-103, 1999. .. [2] L. Breiman, "Bagging predictors", Machine Learning, 24(2), 123-140, 1996. .. [3] T. Ho, "The random subspace method for constructing decision forests", Pattern Analysis and Machine Intelligence, 20(8), 832-844, 1998. .. [4] G. Louppe and P. Geurts, "Ensembles on Random Patches", Machine Learning and Knowledge Discovery in Databases, 346-361, 2012. i gð?ic C sVtt|ƒj|d|d|d|d|d|d|d|d| d | d | ƒ dS( NR?R(R'RR$RrRsRtRRE(RpRRq( RuRoR?R(R'RR$RrRsRtRRE((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyRq›s c s tˆdƒtˆdddgdd dtƒ‰tˆjˆjƒ\}}‰td|dˆjƒ‡‡‡fd †t |ƒDƒƒ}t |ƒˆj}|S( s&Predict regression target for X. The predicted regression target of an input sample is computed as the mean predicted regression targets of the estimators in the ensemble. Parameters ---------- X : {array-like, sparse matrix} of shape = [n_samples, n_features] The training input samples. Sparse matrices are accepted only if they are supported by the base estimator. Returns ------- y : array of shape = [n_samples] The predicted values. RR¯RwRxR[RyRtREc3 sS|]I}ttƒˆjˆ|ˆ|d!ˆjˆ|ˆ|d!ˆƒVqdS(iN(R RmR}R(RjRI(RARuR(s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pys ÑsN( RRR3R:RR?RtRRER5Rk(RuRARtR?t all_y_hatty_hat((RARuRs7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyRU´s   cC s tt|ƒjdtƒƒdS(s:Check the estimator and set the base_estimator_ attribute.RŸN(RpRR…R(Ru((s7lib/python2.7/site-packages/sklearn/ensemble/bagging.pyR…Üsc C s|jd}tj|fƒ}tj|fƒ}x…t|j|j|jƒD]h\}}}t||ƒ} || c|j|| dd…fdd…|fƒ7<|| cd7sB   "   ;    óÿ~