t]eY6dZddlZddlmZddlmZddlmZmZddl m Z m Z ddl Z ddlmZddlmZdd lmZmZdd lmZdd lmZd d lmZdZdZGddZdZd$dZdZ d%dZ!dZ"dZ#dZ$dZ%dZ&dZ'dZ(dZ)dZ*d Z+e&e#e%e$e"e'd!Z,d"Z-d#Z.dS)&zUtilities for input validationN) OrderedDict)wraps) Parameter signature)IntegralReal)clone)NearestNeighbors) check_array column_or_1d)type_of_target) _num_samples) _is_pandas_df) over-samplingunder-samplingclean-samplingensemblebypass)binary multiclassmultilabel-indicatorc*eZdZdZdZdZdZdZdS)ArraysTransformerzAA class to convert sampler output arrays to their original types.cn|||_|||_dSN) _gets_propsx_propsy_propsselfXys :lib/python3.11/site-packages/imblearn/utils/_validation.py__init__zArraysTransformer.__init__"s0''** ''** c|||j}|||j}|jddkr-|jddvr |j|_||fS)Ntype dataframe>seriesr))_transfrom_onerrlowerindexr s r$ transformzArraysTransformer.transform&s   4< 0 0   4< 0 0 <  % % ' '; 6 64< < %'',<-<- gAG!t r&ci}|jj|d<t|dd|d<t|dd|d<t|dd|d<|S)Nr(columnsnamedtypes) __class____name__getattr)r!arraypropss r$rzArraysTransformer._gets_props1sZ0f "5)T::ivt44f !%488h r&cX|d}|dkr|}nt|dkr=ddl}|||d}||d}n1|dkr)ddl}|||d|d  }n|}|S) Nr(listr)rr0)r0r2r*r1)dtyper1)r,tolistpandas DataFrameastypeSeries)r!r6r7type_retpds r$r+z ArraysTransformer._transfrom_one9sf ##%% F??,,..CC k ! !    ,,ueI.>,??C**U8_--CC h      ))ExuV})MMCCC r&N)r4 __module__ __qualname____doc__r%r.rr+r&r$rrsVKK+++   r&rcBddg}tfd|DS)aCheck that the estimator exposes a KNeighborsMixin-like API. A KNeighborsMixin-like API exposes the following methods: (i) `kneighbors`, (ii) `kneighbors_graph`. Parameters ---------- estimator : object A scikit-learn compatible estimator. Returns ------- is_neighbors_object : bool True if the estimator exposes a KNeighborsMixin-like API. kneighborskneighbors_graphc38K|]}t|VdSr)hasattr).0attr estimators r$ z'_is_neighbors_object..\s-IIDwy$''IIIIIIr&)all)rNneighbors_attributess` r$_is_neighbors_objectrRKs5 )*<= IIII4HIII I IIr&cpt|trt||zSt|S)aCheck the objects is consistent to be a k nearest neighbors. Several methods in `imblearn` relies on k nearest neighbors. These objects can be passed at initialisation as an integer or as an object that has KNeighborsMixin-like attributes. This utility will create or clone said object, ensuring it is KNeighbors-like. Parameters ---------- nn_name : str The name associated to the object to raise an error if needed. nn_object : int or KNeighborsMixin The object to be checked. additional_neighbor : int, default=0 Sometimes, some algorithm need an additional neighbors. Returns ------- nn_object : KNeighborsMixin The k-NN object. ) n_neighbors) isinstancerr r )nn_name nn_objectadditional_neighbors r$check_neighbors_objectrY_s<0)X&&MI8K,KLLLL   r&cntj|d\}}tt||S)NT) return_counts)npuniquedictzip)r#r]countss r$_count_class_samplera}s2Yq555NFF FF## $ $$r&Fct|}|dkrRtj|ddkrt d|d}nt |}|r||dkfn|S)aCheck the target types to be conform to the current samplers. The current samplers should be compatible with ``'binary'``, ``'multilabel-indicator'`` and ``'multiclass'`` targets only. Parameters ---------- y : ndarray The array containing the target. indicate_one_vs_all : bool, default=False Either to indicate if the targets are encoded in a one-vs-all fashion. Returns ------- y : ndarray The returned target. is_one_vs_all : bool, optional Indicate if the target was originally encoded in a one-vs-all fashion. Only returned if ``indicate_multilabel=True``. rr)axiszImbalanced-learn currently supports binary, multiclass and binarized encoded multiclasss targets. Multilabel and multioutput targets are not supported.)r r\anysum ValueErrorargmaxr )r#indicate_one_vs_alltype_ys r$check_target_typerjs.A  F ''' 6!%%Q%--!# $ $ 9  HH!H   OO4G NAv// 0 0QNr&cbt|}|dkrBt|fd|D}nU|dks|dkrBt |fd|D}nt |S)z1Returns sampling target by targeting all classes.rc"i|] \}}||z  SrFrF)rLkeyvaluen_sample_majoritys r$ z*_sampling_strategy_all..s2   /;UC"U*   r&rrci|]}|SrFrF)rLrmn_sample_minoritys r$rpz*_sampling_strategy_all..sSSSS"3SSSr&)ramaxvaluesitemsminkeysNotImplementedError)r# sampling_type target_statssampling_strategyrorrs @@r$_sampling_strategy_allr|s&q))L'' 3 3 5 566    ?K?Q?Q?S?S    * * *m?O.O.O 3 3 5 566SSSS|?P?P?R?RSSS!! r&c*|dkrtd|dks|dkrht|}t||jt |fd|D}nt|S)z=Returns sampling target by targeting the majority class only.rz@'sampling_strategy'='majority' cannot be used with over-sampler.rrrmc"i|] }|k| SrFrFrLrmclass_majorityrrs r$rpz/_sampling_strategy_majority..1   n$$ "$$$r&)rfrarsgetrvrtrwrx)r#ryrzr{rrrs @@r$_sampling_strategy_majorityrs'' N    * * *m?O.O.O*1-- \|/?@@@ 3 3 5 566     #((**    "! r&ct|}|dkrYt|t||jfd|D}nl|dks|dkrYt |t||jfd|D}nt|S)zJReturns sampling target by targeting all classes but not the majority.rr~c.i|]\}}|k ||z SrFrF)rLrmrnrros r$rpz3_sampling_strategy_not_majority..:   en$$ "U*$$$r&rrc"i|] }|k| SrFrFrs r$rpz3_sampling_strategy_not_majority..rr&)rarsrtrrurvrwrx)r#ryrzr{rrorrs @@@r$_sampling_strategy_not_majorityr'q))L'' 3 3 5 566\|/?@@@     , 2 2 4 4    * * *m?O.O.O 3 3 5 566\|/?@@@     #((**    "! r&ct|}|dkrYt|t||jfd|D}nl|dks|dkrYt|t||jfd|D}nt|S)zJReturns sampling target by targeting all classes but not the minority.rr~c.i|]\}}|k ||z SrFrFrLrmrnclass_minorityros r$rpz3_sampling_strategy_not_minority..rr&rrc"i|] }|k| SrFrF)rLrmrrrs r$rpz3_sampling_strategy_not_minority..rr&)rarsrtrvrrurwrx)r#ryrzr{rrorrs @@@r$_sampling_strategy_not_minorityrrr&c*t|}|dkrYt|t||jfd|D}n"|dks|dkrt dt|S)z=Returns sampling target by targeting the minority class only.rr~c.i|]\}}|k ||z SrFrFrs r$rpz/_sampling_strategy_minority..rr&rrzS'sampling_strategy'='minority' cannot be used with under-sampler and clean-sampler.)rarsrtrvrrurfrx)r#ryrzr{rros @@r$_sampling_strategy_minorityrs&q))L'' 3 3 5 566\|/?@@@     , 2 2 4 4    * * *m?O.O.O 0   "! r&cj|dkrt||S|dks|dkrt||SdS)zWReturns sampling target auto for over-sampling and not-minority for under-sampling.rrrN)rr)r#rys r$_sampling_strategy_autorsO''.q-@@@ * * *m?O.O.O.q-@@@/P.Or&cdt|}t|t|z }t|dkrt d|dt d|Drt d|i}|dkrt|t||j| D];\}}|||krt d||d |d |||z ||<./s& E EY9q= E E E E E Er&zfThe number of samples in a class cannot be negative.'sampling_strategy' contains some negative value: rr~zWith over-sampling methods, the number of samples in a class should be greater or equal to the original number of samples. Originally, there is z samples and z samples are asked.rzWith under-sampling methods, the number of samples in a class should be less or equal to the original number of samples. Originally, there is rz'sampling_strategy' as a dict for cleaning methods is not supported. Please give a list of the classes to be targeted by the sampling.) rasetrwlenrfrdrtrsrrurx)r{r#ryrz!set_diff_sampling_strategy_targetsampling_strategy_ class_samplers r$_sampling_strategy_dictr!s'q))L(+,=,B,B,D,D(E(EII)% ,--11 $4 $ $ $    E E*;*B*B*D*D E E EEE  UAR U U   '' L   ! !""" Ll.////'8'>'>'@'@ V V #L)< 555 B.:,-GBB$- BBB09< ;U/U | , , V * * *'8'>'>'@'@ 9 9 #L)< 555 B.:,-GBB$- BBB09 | , , 9 * * * (   "! r&c|dkrtdt|t|tz }t |dkrtd|dfd|DS)z[With cleaning methods, sampling_strategy can be a list to target the class of interest.rzQ'sampling_strategy' cannot be a list for samplers which are not cleaning methods.rrrcTi|]$}|t%SrF)rvrt)rLrrzs r$rpz+_sampling_strategy_list..ms<   5A c,--//00   r&)rfrarrwr)r{r#ryrrzs @r$_sampling_strategy_listrYs((( .   'q))L(+,=(>(>BB)% ,--11 $4 $ $ $       EV   r&c t|}|dkrtdt| |dkrt t jfd D}td|Drtdn|dkrt t jfd  D}t fd |Drtd ntd |S) zrTake a proportion of the majority (over-sampling) or minority (under-sampling) class in binary classification.rzg"sampling_strategy" can be a float only when the type of target is binary. For multi-class, use a dict.rr~cNi|]!\}}|k |tz|z "SrFint)rLrmrnrror{s r$rpz,_sampling_strategy_float..sH   en$$ &)::UBCC$$$r&cg|]}|dk SrrFrs r$ z,_sampling_strategy_float..sLLL9 QLLLr&zThe specified ratio required to remove samples from the minority class while trying to generate new samples. Please increase the ratio.rcHi|]\}}|k |tz SrFr)rLrmrnrrrr{s r$rpz,_sampling_strategy_float..sC   en$$ &)::;;$$$r&c.g|]\}}||kSrFrF)rLtargetrrzs r$rz,_sampling_strategy_float..s7   %FIL00   r&zThe specified ratio required to generate new sample in the majority class while trying to remove samples. Please increase the ratio.zD'clean-sampling' methods do let the user specify the sampling ratio.) r rfrarsrtrrurdrv) r{r#ryrirrrrorrrzs ` @@@@@r$_sampling_strategy_floatrrs A  F  @   'q))L'' 3 3 5 566\|/?@@@      , 2 2 4 4    LL0B0I0I0K0KLLL M M    * * * 3 3 5 566\|/?@@@      , 2 2 4 4        );)A)A)C)C      =   R    r&c |tvrtdtd|dtj|jdkr*tdtj|jd|dvr|St |t rx|tvrtdtd|d ttt||| St |tr=ttt||| St |tr=ttt||| St |t r\|d ks|dkrtd |dttt#||| St%|rF||fi|}ttt||| Sd S) aJSampling target validation for samplers. Checks that ``sampling_strategy`` is of consistent type and return a dictionary containing each targeted class with its corresponding number of sample. It is used in :class:`~imblearn.base.BaseSampler`. Parameters ---------- sampling_strategy : float, str, dict, list or callable, Sampling information to sample the data set. - When ``float``: For **under-sampling methods**, it corresponds to the ratio :math:`\alpha_{us}` defined by :math:`N_{rM} = \alpha_{us} \times N_{m}` where :math:`N_{rM}` and :math:`N_{m}` are the number of samples in the majority class after resampling and the number of samples in the minority class, respectively; For **over-sampling methods**, it correspond to the ratio :math:`\alpha_{os}` defined by :math:`N_{rm} = \alpha_{os} \times N_{m}` where :math:`N_{rm}` and :math:`N_{M}` are the number of samples in the minority class after resampling and the number of samples in the majority class, respectively. .. warning:: ``float`` is only available for **binary** classification. An error is raised for multi-class classification and with cleaning samplers. - When ``str``, specify the class targeted by the resampling. For **under- and over-sampling methods**, the number of samples in the different classes will be equalized. For **cleaning methods**, the number of samples will not be equal. Possible choices are: ``'minority'``: resample only the minority class; ``'majority'``: resample only the majority class; ``'not minority'``: resample all classes but the minority class; ``'not majority'``: resample all classes but the majority class; ``'all'``: resample all classes; ``'auto'``: for under-sampling methods, equivalent to ``'not minority'`` and for over-sampling methods, equivalent to ``'not majority'``. - When ``dict``, the keys correspond to the targeted classes. The values correspond to the desired number of samples for each targeted class. .. warning:: ``dict`` is available for both **under- and over-sampling methods**. An error is raised with **cleaning methods**. Use a ``list`` instead. - When ``list``, the list contains the targeted classes. It used only for **cleaning methods**. .. warning:: ``list`` is available for **cleaning methods**. An error is raised with **under- and over-sampling methods**. - When callable, function taking ``y`` and returns a ``dict``. The keys correspond to the targeted classes. The values correspond to the desired number of samples for each class. y : ndarray of shape (n_samples,) The target array. sampling_type : {{'over-sampling', 'under-sampling', 'clean-sampling'}} The type of sampling. Can be either ``'over-sampling'``, ``'under-sampling'``, or ``'clean-sampling'``. **kwargs : dict Dictionary of additional keyword arguments to pass to ``sampling_strategy`` when this is a callable. Returns ------- sampling_strategy_converted : dict The converted and validated sampling target. Returns a dictionary with the key being the class target and the value being the desired number of samples. z!'sampling_type' should be one of z. Got 'z instead.rz4The target 'y' needs to have more than 1 class. Got z class instead)rrzO   '(9:1mLLRRTT U U    %t , ,  *+.inner_f..Vs6D##r&zPass z, z` as keyword args. From version 0.9 passing these as positional arguments will result in an errorci|]\}}|| SrFrF)rLkrs r$rpz?_deprecate_positional_args..inner_f..`sFFF&!Sq#FFFr&rF)rr_warningswarnjoin FutureWarningupdate parameters)argsr extra_argsargs_msgall_argsf kwonly_argssigs r$inner_fz+_deprecate_positional_args..inner_fQsYYX. >>!$[*%=tZKLL?Q!R!RH M& (++&&&      FFC,E,EFFFGGGq{{6{{r&) rrrukindrPOSITIONAL_OR_KEYWORDappend KEYWORD_ONLYr)rr1paramrrrrs` @@@r$_deprecate_positional_argsr<s A,,CKH~++--%% e :8 8 8 OOD ! ! ! ! Z91 1 1   t $ $ $ 1XXX" Nr&ct|}|dkrtd|dt|r|St|dddgdS) z+Check X and do not check it if a dataframe.rzFound array with z, sample(s) while a minimum of 1 is required.NcsrcscF)r: accept_sparseforce_all_finite)rrfrr )r"rs r$_check_XrfsyQI1}}        Q  eU^e   r&r)F)/rEr collectionsr functoolsrinspectrrnumbersrrnumpyr\ sklearn.baser sklearn.neighborsr sklearn.utilsr r sklearn.utils.multiclassr sklearn.utils.validationrfixesrr TARGET_KINDrrRrYrarjr|rrrrrrrrrrrrrFr&r$rsV$$ ######((((((((""""""""......33333333333333111111  ? ))))))))XJJJ(<%%% #O#O#O#OL"*44,AAA555p2000fJ J J \,+33 ! # '''T     r&