9iJSSKJrJr SSKrSSKJr SSKJr "SS\5rg))AnyDictN)ModeKeys)OfaBasePreprocessorc^\rSrSrSr\R 4U4SjjrS\\ \ 4S\\ \ 44Sjr Sr S\\ \ 4S\\ \ 44Sjr S\\ \ 4S\\ \ 44S jrS rU=r$) !OfaTextClassificationPreprocessor z1 OFA preprocessor for text classification tasks. c:>[[U] "XU/UQ70UD6 g)zpreprocess the data Args: cfg(modelscope.utils.config.ConfigDict) : model config model_dir (str): model path, mode: preprocessor mode (model mode) N)superr __init__)selfcfg model_dirmodeargskwargs __class__s p/var/www/html/land-doc-ocr/venv/lib/python3.13/site-packages/modelscope/preprocessors/ofa/text_classification.pyr *OfaTextClassificationPreprocessor.__init__s8 /   T D48 D>NN5)E&&5'{E&B   v %"J    .OAb$99#EFJ% %))T]]JsO$DE'+}} #S^O$% "1    ( rcURU5nURS:Xa/nURnO!URS:Xa USSnUSSnO[eUUUS.nSU;aURUSUS'U$)a Building inference samples. step 1. Building instruction for text classification using `_build_instruction`. step 2. Whether or not to add `prefix_token`. step 3. Whether or not to add `label` data. Args: data (`Dict[str, Any]`): Input data, should contains the key of `text` and `text2`, both of them refer to a text input, and the target of this job is to find whether or not `text` imply `text2`. Return: A dict object, contains source text input, prefix tokens and label data. r3r4Nr5)r6 prefix_tokendecoder_promptr1)r.r:r>r=r9)rrr-rHrIrEs rr5OfaTextClassificationPreprocessor._build_infer_samplefs11$7   v %L!]]N    .*3B/L,Sb1N% %%(,  d?"nnT'];F7O r)__name__ __module__ __qualname____firstlineno____doc__r INFERENCEr rstrrrr.rr__static_attributes__ __classcell__)rs@rr r s((D 2T#s(^2S#X2 ,)S#X)4S>)VS#X4S>rr ) typingrrr;modelscope.utils.constantrbaserr rKrrrXs  .%{(;{r