g 1dZddlZddlZddlZddlZddlZddlZddlmZddl m Z ddl m Z m Z ddlmZdagdZddZd ZGd d e ZGd d Zedk(rddlmZmZdZeeeZyy)z; Classifiers that make use of the external 'Weka' package. N)stdin) ClassifierI) config_javajava)DictionaryProbDist).z/usr/share/wekaz/usr/local/share/wekaz /usr/lib/wekaz/usr/local/lib/wekac t||att}dtjvr#|j dtjd|D]}tj jtj j|dsAtj j|datt}|rtdtd|dntdtzttt tdy) NWEKAHOMErzweka.jarz [Found Weka: z (version z)]z[Found Weka: %s]zUnable to find weka.jar! Use config_weka() or set the WEKAHOME environment variable. For more information about Weka, please see https://www.cs.waikato.ac.nz/ml/weka/) r_weka_classpath _weka_searchosenvironinsertpathexistsjoin_check_weka_versionprint LookupError) classpath searchpathrversions G/var/www/openai/venv/lib/python3.12/site-packages/nltk/classify/weka.py config_wekar"sM#!  #   aJ!7 8Dww~~bggll4<="$'',,tZ"@-o>M/):*WIRPQ,>?#O4 4  c tj|} |j d|j S#ttf$rYyxYw#t $rY|j ywxYw#|j wxYw)Nzweka/core/version.txt)zipfileZipFile SystemExitKeyboardInterruptreadcloseKeyError)jarzfs rrrCs| __S !  7723   ) *       s,:AA  A,A/+A,,A//BcZeZdZdZdZdZdZdZdZddd d d d d Z e dgdfdZ y)WekaClassifierc ||_||_yN) _formatter_model)self formattermodel_filenames r__init__zWekaClassifier.__init__Ts#$ rc*|j|gdS)N)-p0z -distribution_classify_manyr, featuresetss rprob_classify_manyz!WekaClassifier.prob_classify_manyXs"";0LMMrc*|j|ddgS)Nr1r2r3r5s r classify_manyzWekaClassifier.classify_many[s"";s <> V V>CABiXid4::% XX3U1X >05FDJJLOF Fcr d #!&q* +YGsE%/'E%89E*E/'E/r;zweka.classifiers.trees.J48z#weka.classifiers.functions.Logisticzweka.classifiers.functions.SMOzweka.classifiers.lazy.KStarzweka.classifiers.rules.JRip) naivebayesC4.5log_regressionsvmkstarripperrnTc ttj|}tj} t j j|d}|j||||jvr|j|} n-||jjvr|} ntd|z| d|d|g} | t|z } |rtj} nd} t| t | t#||t j$|D]5} t j&t j j|| 7t j(|S#t j$|D]5} t j&t j j|| 7t j(|wxYw)Nz train.arffzUnknown classifier %sz-dz-t)rr<)rARFF_Formatter from_trainr?r@r rrrA_CLASSIFIER_CLASSvaluesrDlistrBrCrr r'rIrJrK) clsr.r6 classifierrLquietr-rMtrain_filename javaclassrOr<rPs rtrainzWekaClassifier.trains^  #--k: ##% WW\\(LAN OONK 8S22211*= s44;;==&  !8:!EFFdND.IC 4= C#  ?")^<ZZ) "'',,x34* HHX ZZ) "'',,x34* HHX s CEA$GN) __name__ __module__ __qualname__r/r7r9r4r[rErw classmethodrrrr'r'Ss^%N=*X) !b:,?/./    **rr'cLeZdZdZdZdZdZdZedZ dZ d d Z d Z y) ruz Converts featuresets and labeled featuresets to ARFF-formatted strings, appropriate for input into Weka. Features and classes can be specified manually in the constructor, or may be determined from data using ``from_train``. c ||_||_y)a) :param labels: A list of all class labels that can be generated. :param features: A list of feature specifications, where each feature specification is a tuple (fname, ftype); and ftype is an ARFF type string such as NUMERIC or STRING. N)_labels _features)r,rWfeaturess rr/zARFF_Formatter.__init__s !rcF|j|j|zS)zBReturns a string representation of ARFF output for the given data.)header_section data_section)r,tokenss rformatzARFF_Formatter.format s!""$t'8'8'@@@rc,t|jS)zReturns the list of classes.)ryr)r,s rrWzARFF_Formatter.labelssDLL!!rct|ds t|d}|j|j||j y)z.Writes ARFF data to a file for the given data.rAwN)hasattropenrArr")r,outfilers rrAzARFF_Formatter.writes6w(7C(G dkk&)* rc|Dchc]\}}| }}}i}|D]\}}|jD]\}}tt|trd}nTtt|tt tfrd}n-tt|t rd}n|htdz|j|||k7rtd|z|||<t|j}t||Scc}}w)z Constructs an ARFF_Formatter instance with class labels and feature types determined from the given data. Handles boolean, numeric and string (note: not nominal) types. z {True, False}NUMERICSTRINGzUnsupported value type %rzInconsistent type for %s) items issubclasstypeboolintrTstrrDgetsortedru)rtoklabelrWrfnamefvalftypes rrvzARFF_Formatter.from_trains-33FLS%%F3 JC"yy{ td4j$/+ET S%,>?%ET C0$E\$%@5%HII<<u-6$%?%%GHH"' +! (..*+fh//-4s C6cddtjzz}|dz }|jD]\}}|d|dd|dz }|dddd d j|jd z }|S) z#Returns an ARFF header as a string.z3% Weka ARFF file % Generated automatically by NLTK z%% %s z@RELATION rel z @ATTRIBUTE 30 r>z-label-z {,z} )timectimerrr)r,rXrrs rrzARFF_Formatter.header_section9sr 4DJJL( )   !NNLE5 E59 9A+ )SXXdll5K LLrNc .||xrt|dttf}|s|Dcgc]}|df}}d}|D]V\}}|jD]+\}}|d|j |j |zz }-|d|j |zz }X|Scc}w)a Returns the ARFF data section for the given data. :param tokens: a list of featuresets (dicts) or labelled featuresets which are tuples (featureset, label). :param labeled: Indicates whether the given tokens are labeled or not. If None, then the tokens will be assumed to be labeled if the first token's value is a tuple or list. Nrz @DATA z%s,z%s ) isinstancetupleryr _fmt_arff_valr)r,rlabeledrrXrrrs rrzARFF_Formatter.data_sectionNs ?EF1It}!EG-34VcsDkVF4  JC $ uUT//???!/ $,,U33 3A! 5s Bcr|yt|ttfrd|zSt|trd|zSd|zS)N?z%sz%r)rrrrT)r,rs rrzARFF_Formatter._fmt_arff_valhs@ < tSk *$;  e $$; $; rr)) rrr__doc__r/rrWrA staticmethodrvrrrrrrrurus? "A"00>*4rru__main__)binary_names_demo_features names_democ0tjd|dS)Nz/tmp/name.modelro)r'r)r6s rmake_classifierrvs##$5{FKKrr))rr rRrBr?rrsysrnltk.classify.apirnltk.internalsrrnltk.probabilityrr r rrr'rurnltk.classify.utilrrrr{rrrrs  ),/  B  `[`Fzzz zILO-GHJ r