j1dZddlZddlZddlZddlZddlZddlZddlmZddl m Z ddl m Z m Z ddlmZdagdZddZd ZGd d e ZGd d ZedkrddlmZmZdZeeeZdSdS)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/wekacXt||att}dtjvr&|dtjd|D]}tjtj|dr}tj|datt}|rtdtd|dntdtzttttddS) 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 \/lsinfo/ai/hellotax_ai/base_platform/venv/lib/python3.11/site-packages/nltk/classify/weka.py config_wekar"sMMM#!  # #   aJ!7 8 8 8 5 5Dw~~bgll4<<== 5"$',,tZ"@"@-o>>@P/PPWPPPQQQQ,>???#O444 4   cD tj|}n"#ttf$rt$rYdSwxYw |d|S#t$rY|dSwxYw#|wxYw)Nzweka/core/version.txt)zipfileZipFile SystemExitKeyboardInterrupt ExceptionreadcloseKeyError)jarzfs rrrCs _S ! ! ) * tt 77233          s/66A$$ B.B BB BcbeZdZdZdZdZdZdZdZddd d d d d Z e dgdfdZ dS)WekaClassifierc"||_||_dSN) _formatter_model)self formattermodel_filenames r__init__zWekaClassifier.__init__Ts#$ rc2||gdS)N)-p0z -distribution_classify_manyr- featuresetss rprob_classify_manyz!WekaClassifier.prob_classify_manyXs"";0L0L0LMMMrc2||ddgS)Nr2r3r4r6s r classify_manyzWekaClassifier.classify_many[s"";s <<fs rr5zWekaClassifier._classify_many^s #%%% GLL;??M O ! !- = = =4    C $)!!     VV Uf U3v==$" %%JV%STTT))&--*G*G*M*Md*S*STTZ)) 5 5 "',,x334444 HX    Z)) 5 5 "',,x334444 HX    s C&E''AGcdtjd|D}tt|j|}t |S)NcTg|]%}|t|&S)stripfloat).0vs r z:WekaClassifier.parse_weka_distribution..s+EEEa17799EqEEErz[*,]+)rerIdictzipr+labelsr)r-sprobss rparse_weka_distributionz&WekaClassifier.parse_weka_distributionsTEE28GQ#7#7EEES//11599::!%(((rct|D]8\}}|dr ||d}n9|dgdkrd|ddDS|dgdkrfd|ddDSt jd|dr d |DS|dd D]}t |td |dz) Ninst#r)rbactual predictederror predictioncg|]I}||dddJS):rUrIrWlines rrYz4WekaClassifier.parse_weka_output..sDXXXd4::<<XDJJLLO))#..q1XXXrrj)rbrcrdre distributioncg|]C}||dDS))rUr`rI)rWrmr-s rrYz4WekaClassifier.parse_weka_output..sR::<<,,TZZ\\"-=>>rz^0 \w+ [01]\.[0-9]* \?\s*$cjg|]0}||d1S)rjrkrls rrYz4WekaClassifier.parse_weka_output..s1FFFFDJJLLOFFFr zRUnhandled output format -- your version of weka may not be supported. Header: %s) enumeraterU startswithrIrZmatchrrE)r-linesirms` rrFz WekaClassifier.parse_weka_outputsf ''  GAtzz||&&w// abb   8>>  VVV V VXXeABBiXXX X 1X^^  " " "   !!""I X3U1X > > FFFFF Fcrc   d !&q* rr<zweka.classifiers.trees.J48z#weka.classifiers.functions.Logisticzweka.classifiers.functions.SMOzweka.classifiers.lazy.KStarzweka.classifiers.rules.JRip) naivebayesC4.5log_regressionsvmkstarripperrxTc tt|}tj} t j|d}|||||j vr|j |} n0||j vr|} ntd|z| d|d|g} | t|z } |r tj} nd} t| t | t#||t j|D]4} t jt j|| 5t j|S#t j|D]4} t jt j|| 5t j|wxYw)Nz train.arffzUnknown classifier %sz-dz-t)rr=)rARFF_Formatter from_trainr@rAr rrrB_CLASSIFIER_CLASSvaluesrElistrCrDrr r(rJrKrL) clsr/r7 classifierrMquietr.rNtrain_filename javaclassrPr=rQs rtrainzWekaClassifier.trains  #--k:: #%% W\\(LAAN OONK 8 8 8S2221*= s4;;====&  !8:!EFFFdND.IC 4== C #  ? ? ? ?")^<<Z)) 5 5 "',,x334444 HX    Z)) 5 5 "',,x334444 HX    s C E((AGN) __name__ __module__ __qualname__r0r8r:r5r`rFr classmethodrrTrrr(r(Ss%%%NNN===***X))) !!!b:,?/./    ***[***rr(cTeZdZdZdZdZdZdZedZ dZ d d Z d Z dS) rz 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"||_||_dS)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-r]featuress rr0zARFF_Formatter.__init__s !rcV|||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.)rr)r-s rr]zARFF_Formatter.labelssDL!!!rct|dst|d}||||dS)z.Writes ARFF data to a file for the given data.rBwN)hasattropenrBrr#)r-outfilers rrBzARFF_Formatter.writesRw(( )7C((G dkk&))*** rcHd|D}i}|D]\}}|D]\}}tt|trd}nltt|tt tfrd}n:tt|t rd}n|td|z||||krtd|z|||<Ȍt|}t||S)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. ch|]\}}|SrTrT)rWtoklabels r z,ARFF_Formatter.from_train..!s333LS%%333rz {True, False}NUMERICSTRINGNzUnsupported value type %rzInconsistent type for %s) items issubclasstypeboolintrVstrrEgetsortedr)rr]rrrfnamefvalftypes rrzARFF_Formatter.from_trains543F333  ( (JC"yy{{ ( ( td4jj$// J+EET S%,>??J%EET C00J$EE\$%@5%HIII<<u--66$%?%%GHHH"' ((..**++fh///rcddtjzz}|dz }|jD]\}}|d|dd|dz }|dddd d |jd z }|S) z#Returns an ARFF header as a string.z3% Weka ARFF file % Generated automatically by NLTK z%% %s z@RELATION rel z @ATTRIBUTE 30 r?z-label-z {,z} )timectimerrr)r-r^rrs rrzARFF_Formatter.header_section9s 4DJLL( )   !N : :LE5 AEEEE5559 9AA ))))SXXdl5K5K5K5K LLrNc .|$|o!t|dttf}|s d|D}d}|D][\}}|jD]3\}}|d|||zz }4|d||zz }\|S)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. Nrcg|]}|dfSr*rT)rWrs rrYz/ARFF_Formatter.data_section..]s444csDk444rz @DATA z%s,z%s ) isinstancetuplerr _fmt_arff_valr)r-rlabeledr^rrrrs rrzARFF_Formatter.data_sectionNs ?EF1It}!E!EG 544V444F   4 4JC $ @ @ uUT//???? $,,U333 3AArc|dSt|ttfrd|zSt|trd|zSd|zS)N?z%sz%r)rrrrV)r-rs rrzARFF_Formatter._fmt_arff_valhsR <3 tSk * * $;  e $ $ $; $; rr*) rrr__doc__r0rr]rB staticmethodrrrrrTrrrrs " " "AAA"""00\0>*4rr__main__)binary_names_demo_features names_democ:td|dS)Nz/tmp/name.modelry)r(r)r7s rmake_classifierrvs##$5{FKKKrr*)rr rZrCr@rrsysrnltk.classify.apirnltk.internalsrrnltk.probabilityrr r rrr(rrnltk.classify.utilrrrrrTrrrs  )))))),,,,,,,,//////     B    `````[```Fzzzzzzzzz zIIIIIIIILLLO-GHHJJJ r