Mipa SSKrSSKJr SSKJr SSKJrJrJrJ r J r J r J r J r Jr SSKrSSKJrJrJrJrJrJr SSKJrJrJrJr SSKJrJr SS KJ r SS K!J"r" SS K#J$r$ SS K%J&r&J'r'J(r( SS K)J*r*J+r+ SSK,J-r- SSK.J/r/ Sr0Sr1Sr2\"5Rg\05Sr4\"5Rg\15Sr5\"SS\55r6SS.S\ \&S\ \7S\ \S\4Sjjr8SS.S\ \&S\9S\ \S\4Sjjr:S\ \7S\64Sjr;S \7S!\7S\64S"jrS&r?\"S'S(55r@"S)S*\/5rAS+rBg),N) dataclass)partial) AnyCallableDictIterableListOptionalTupleUnioncast)ConfigModelOps Optimizerget_current_opsset_dropout_rate)Floats2dInts1dInts2dRagged)Protocolruntime_checkable)Errors)Language)Scorer)DocSpan SpanGroup)Examplevalidate_examples)Vocab) TrainablePipea4 [model] @architectures = "spacy.SpanCategorizer.v1" scorer = {"@layers": "spacy.LinearLogistic.v1"} [model.reducer] @layers = spacy.mean_max_reducer.v1 hidden_size = 128 [model.tok2vec] @architectures = "spacy.Tok2Vec.v2" [model.tok2vec.embed] @architectures = "spacy.MultiHashEmbed.v2" width = 96 rows = [5000, 1000, 2500, 1000] attrs = ["NORM", "PREFIX", "SUFFIX", "SHAPE"] include_static_vectors = false [model.tok2vec.encode] @architectures = "spacy.MaxoutWindowEncoder.v2" width = ${model.tok2vec.embed.width} window_size = 1 maxout_pieces = 3 depth = 4 a& [model] @architectures = "spacy.SpanCategorizer.v1" scorer = {"@layers": "Softmax.v2"} [model.reducer] @layers = spacy.mean_max_reducer.v1 hidden_size = 128 [model.tok2vec] @architectures = "spacy.Tok2Vec.v2" [model.tok2vec.embed] @architectures = "spacy.MultiHashEmbed.v1" width = 96 rows = [5000, 1000, 2500, 1000] attrs = ["NORM", "PREFIX", "SUFFIX", "SHAPE"] include_static_vectors = false [model.tok2vec.encode] @architectures = "spacy.MaxoutWindowEncoder.v2" width = ${model.tok2vec.embed.width} window_size = 1 maxout_pieces = 3 depth = 4 scmodelc>\rSrSrSS.S\\S\\S\4Sjjr Sr g) SuggesterONopsdocsr,returncgN)selfr-r,s p/var/www/eduai.edurigo.com/storigo/production/storigo_env/lib/python3.13/site-packages/spacy/pipeline/spancat.py__call__Suggester.__call__QsUXr1) __name__ __module__ __qualname____firstlineno__rrr rrr4__static_attributes__r1r6r3r)r)Os"DHXXc]XHSMXVXXr6r)r+r-sizesr,r.c Uc [5n/n/nUHnURR[U5SS9nUR S5nSnUHnU[U5::aVUS[U5US- - n UR URR XU-455 XsSRS- nU(dMqUSRS:XaMUSR5e UR U5 M URU5n [U5S:a&[URRU5U 5n O$[URRSSS9U 5n U RRS:XdeU $) Nidtype)r$rr$rArrr)rxparangelenreshapeappendhstackshapendim asarray1irvstackzerosdataXd) r-r<r,spanslengthsdocstartslengthsize starts_size lengths_arrayoutputs r3ngram_suggesterrXTs\ { EGs3xs3(Ds3x$%r?rB) rrOrGstartendr rasarrayrErrCrM) r-rYr,rOrPrQrSspanrVrWs r3preset_spans_suggesterr_qs { EG 99Y  ), djj$((34! - vWC!@AM 5zA~ E 5}E M V3 7G Mr6c[[US9$)zSuggest all spans of the given lengths. Spans are returned as a ragged array of integers. The array has two columns, indicating the start and end position.r<)rrXras r3build_ngram_suggesterrbs ?% 00r6min_sizemax_sizecF[[XS-55n[U5$)zSuggest all spans of the given lengths between a given min and max value - both inclusive. Spans are returned as a ragged array of integers. The array has two columns, indicating the start and end position.r$)listrangerb)rcrdr<s r3build_ngram_range_suggesterrhs" xA. /E  ''r6c[[US9$)zSuggest all spans that are already stored in doc.spans[spans_key]. This is useful when an upstream component is used to set the spans on the Doc such as a SpanRuler or SpanFinder.rY)rr_rjs r3build_preset_spans_suggesterrks )Y ??r6examplesc ^^[U5nSmUSmURSTT35 URSS5 URSU4Sj5 URSU4S j5 [R"U40UD6$) Nspans_rYattr allow_overlapTgettercT>URRU[T5S/5$r0)rOgetrE)rQkey attr_prefixs r3spancat_score..s!399==S5E5G1H"#Mr6has_annotationc">TUR;$r0)rO)rQrts r3rvrwsC3994Dr6)dict setdefaultr score_spans)rlkwargsrurts @@r3 spancat_scorer~s &\FK  C f cU34 ot, M &(DE   h 1& 11r6c[$r0)r~r1r6r3make_spancat_scorerrs r6c*\rSrSrSrSrSrSrSrg) _Intervalsz2 Helper class to avoid storing overlapping spans. c"[5Ulgr0)setrangesr2s r3__init___Intervals.__init__s e r6c^[X5HnURRU5 M gr0)rgradd)r2r>jes r3r_Intervals.adds qA KKOOA r6cRUup#[X#5HnX@R;dM g g)NTF)rgr)r2rangr>rrs r3 __contains___Intervals.__contains__s)qAKKr6)rN) r7r8r9r:__doc__rrrr;r1r6r3rrsr6rc\rSrSrSrS>SSSSSS\S .S \S \\\ \ \ 4\ 4S \ S \S\S\S\\S\\S\\S\\S\\SS4Sjjjr\S\4Sj5rS?SjrS\S\4Sjr\S\\4Sj5r\S\ \4Sj5r\S\\\44Sj5r\S\4Sj5r\S\\S44Sj5rS \ \ 4S!jr!S"S#.S \ \ S$\SS4S%jjr"S \ \ SS4S&jr#S'SSS(.S)\ \$S*\S+\\%S,\\\\4S\\\44 S-jjr&S)\ \$S.\\ \ 4S\\\44S/jr'SSS0.S1\/\ \$4S2\\(S3\\ \SS4S4jjr)S)\ \$4S5jr*S6\$4S7jr+S8\ S9\,S:\ S\-4S;jr.S@S8\ S9\,S:\ S\S\-4 S<jjr/S=r0g)ASpanCategorizerzWPipeline component to label spans of text. DOCS: https://spacy.io/api/spancategorizer FrO?TNg?)add_negative_labelrYnegative_weightrp max_positive thresholdscorervocabr' suggesternamerrYrrprrrr.c/UU U UUS.UlXlX0lX lX@lXlXPlU(d1U b-U S:a&[[RRU S95eggg)a}Initialize the multi-label or multi-class span categorizer. vocab (Vocab): The shared vocabulary. model (thinc.api.Model): The Thinc Model powering the pipeline component. For multi-class classification (single label per span) we recommend using a Softmax classifier as a the final layer, while for multi-label classification (multiple possible labels per span) we recommend Logistic. suggester (Callable[[Iterable[Doc], Optional[Ops]], Ragged]): A function that suggests spans. Spans are returned as a ragged array with two integer columns, for the start and end positions. name (str): The component instance name, used to add entries to the losses during training. spans_key (str): Key of the Doc.spans dict to save the spans under. During initialization and training, the component will look for spans on the reference document under the same key. Defaults to `"spans"`. add_negative_label (bool): Learn to predict a special 'negative_label' when a Span is not annotated. threshold (Optional[float]): Minimum probability to consider a prediction positive. Defaults to 0.5. Spans with a positive prediction will be saved on the Doc. max_positive (Optional[int]): Maximum number of labels to consider positive per span. Defaults to None, indicating no limit. negative_weight (float): Multiplier for the loss terms. Can be used to downweight the negative samples if there are too many when add_negative_label is True. Otherwise its unused. allow_overlap (bool): If True the data is assumed to contain overlapping spans. Otherwise it produces non-overlapping spans greedily prioritizing higher assigned label scores. Only used when max_positive is 1. scorer (Optional[Callable]): The scoring method. Defaults to Scorer.score_spans for the Doc.spans[spans_key] with overlapping spans allowed. DOCS: https://spacy.io/api/spancategorizer#init )labelsrYrrrrpNr$)r) cfgrrr'rrr ValueErrorrE1051format) r2rr'rrrrYrrprrrs r3rSpanCategorizer.__init__sxf""(.*   "   "4!9lQ>NV\\00l0KL L?O!9}r6c2[URS5$)zKey of the doc.spans dict to save the spans under. During initialization and training, the component will look for spans on the reference document under the same key. rY)strrrs r3rtSpanCategorizer.keys 488K())r6cPSnURRS5(aURRS5nOyURRS5(aYURR S5RS5(a*URR S5RS5nUblXR :Xa\UR (dJ[[RRURURRS5S95eggg)zbNN2$$ KK&&DII$**:L:LT:R&S%3>r6labelc&[U[5(d[[R5eXR ;agUR 5 URSRU5 URRRU5 g)zAdd a new label to the pipe. label (str): The label to add. RETURNS (int): 0 if label is already present, otherwise 1. DOCS: https://spacy.io/api/spancategorizer#add_label rrr$) isinstancerrrE187rrrrGrstringsr)r2rs r3 add_labelSpanCategorizer.add_label%sk%%%V[[) ) KK  ! !!%( u%r6c2[URS5$)zvRETURNS (Tuple[str]): The labels currently added to the component. DOCS: https://spacy.io/api/spancategorizer#labels r)tuplerrs r3rSpanCategorizer.labels6s TXXh'((r6c,[UR5$)zvRETURNS (List[str]): Information about the component's labels. DOCS: https://spacy.io/api/spancategorizer#label_data )rfrrs r3 label_dataSpanCategorizer.label_data>s DKK  r6cb[UR5VVs0sHupX!_M snn$s snnf)z(RETURNS (Dict[str, int]): The label map.) enumerater)r2r>rs r3 _label_mapSpanCategorizer._label_mapFs+*34;;)?@)?XQ)?@@@s+c~UR(a[UR5S-$[UR5$)z RETURNS (int): Number of labels.r$)rrErrs r3rSpanCategorizer._n_labelsKs0  " "t{{#a' 't{{# #r6cPUR(a[UR5$g)z8RETURNS (Union[int, None]): Index of the negative label.N)rrErrs r3_negative_label_i!SpanCategorizer._negative_label_iSs  " "t' 'r6r-cURXRRS9nURR 5S:Xa)URRR SS5nX#4$URR X45nX#4$)zApply the pipeline's model to a batch of docs, without modifying them. docs (Iterable[Doc]): The documents to predict. RETURNS: The models prediction for each document. DOCS: https://spacy.io/api/spancategorizer#predict r+r)rr'r,rPsumalloc2fpredict)r2r-indicesscoress r3rSpanCategorizer.predict[su..::>>.: ??   A %ZZ^^++Aq1FZZ''8Fr6 candidates)candidates_keyrcURXRRS9n[X15HMupE/URU'UR H)nURUR XVSUS5 M+ MO g)aGUse the spancat suggester to add a list of span candidates to a list of docs. This method is intended to be used for debugging purposes. docs (Iterable[Doc]): The documents to modify. candidates_key (str): Key of the Doc.spans dict to save the candidate spans under. DOCS: https://spacy.io/api/spancategorizer#set_candidates r+rr$N)rr'r,ziprOrNrG)r2r-rsuggester_outputrrQindexs r3set_candidatesSpanCategorizer.set_candidatesjsq >>$JJNN>C"#3:OJ(*CIIn %#** .)001Xa1IJ+ ;r6c Uup4Sn[U5HupgX6Rn[[URS5n URSS:Xa=UR UUXEXSR U-U 5URUR'O;URUUXEXSR U-5URUR'XSR U- nM g)zModify a batch of Doc objects, using pre-computed scores. docs (Iterable[Doc]): The documents to modify. scores: The scores to set, produced by SpanCategorizer.predict. DOCS: https://spacy.io/api/spancategorizer#set_annotations rrprr$N) rrNr boolr_make_span_group_singlelabelrPrOrt_make_span_group_multilabel) r2r-indices_scoresrroffsetr>rQ indices_irps r3set_annotationsSpanCategorizer.set_annotations|s)oFA ))I txx'@AMxx'1,&*&G&GF__Q-?$?@! ' $((#'+&F&FF__Q-?$?@' $((# ooa( (F!&r6)dropsgdlossesrlrrrcpUc0nURURS5 [US5 URU5 [ SU55(dU$UVs/sHoUR PM nnUR X`RRS9nURR5S:XaU$[URU5 URRXg45upURXU45upU "U 5 UbURU5 X@R==U - ss'U$s snf)aLearn from a batch of documents and gold-standard information, updating the pipe's model. Delegates to predict and get_loss. examples (Iterable[Example]): A batch of Example objects. drop (float): The dropout rate. sgd (thinc.api.Optimizer): The optimizer. losses (Dict[str, float]): Optional record of the loss during training. Updated using the component name as the key. RETURNS (Dict[str, float]): The updated losses dictionary. DOCS: https://spacy.io/api/spancategorizer#update rzSpanCategorizer.updatec3p# UH,oR(a[UR5OSv M. g7f)rN) predictedrE).0egs r3 )SpanCategorizer.update..s"Oh 3r||$!;hs46r+r)r{rr"_validate_categoriesanyrrr'r,rPrr begin_updateget_loss finish_update) r2rlrrrrr-rOrbackprop_scoreslossd_scoress r3updateSpanCategorizer.updates ( >F$))S)($<= !!(+OhOOOM'/0x x0t8 ==   ! #MT*"&**"9"94-"HxA! ?   s #yyT! 1sD3 spans_scorescUup4[URRRUR5URRRUR 55n[ R"URURS9nUR(a#[ R"URS5nSnURn[U5Hup0n X9Rn [UR U 5H)n [!XS45n[!XS45nX}-XU4'M+ UR#U 5HVnUR$UR&4nUU ;dM#U UnUUR(nSUUU4'UR(dMQSWU'MX XsR U - nM URRR+USS9nUR(a*[ R,"W5SnSUUUR.4'XE- nUR(a0[1[2UR4S5nUS:wa UW==U-ss'[3US-R755nUU4$) a;Find the loss and gradient of loss for the batch of documents and their predicted scores. examples (Iterable[Examples]): The batch of examples. spans_scores: Scores representing the model's predictions. RETURNS (Tuple[float, float]): The loss and the gradient. DOCS: https://spacy.io/api/spancategorizer#get_loss r?rr$rrfrr)rr'r,to_numpydatarPnumpyrMrIr@ronesrrrNrgint_get_aligned_spansr[r\label_r]nonzerorr floatrr)r2rlrrOrtargetnegative_spansr label_mapr>r spans_indexspans_irr[r\ gold_spanrtrowknegative_samplesr neg_weightrs r3rSpanCategorizer.get_losss-%  JJNN # #EJJ /1H1H1W V\\>  " ""ZZa:NOO x(EAKhooG5==+,GqDM*'Q$-(,2J CL)-"44R8   6+%%c*C!)"2"23A%(F36N....1s+9 mmA& &F+),''c':  " "$}}^r  attrs_scoresr[r\rkeeps r3r+SpanCategorizer._make_span_group_multilabel.s#HH- ;;! L((0**..))'2HH[) xx/ #  #lC00 00&&"'**VAt7M7M4M-N"O5:YYJq$0001 2+..04Cq$0001 2+..0 LM!12K#K0 %f 1 w}}Q'(AqDMEQ$-C$UX.44D222 T#cQ%PQ$++Fa4L9 /)!& L 9 H r6c URS:Xa[XRS9$URRR U5nURRR U5nUR SS9n[R"U[R"US5SS9n[R"UR[S9nUR(a#[R"XuUR:g5nUR SnUb'[R"XvU:R#55nU(d1UR#5S-R%5n XinXYnX)nXyn['5n [XRS9n /n [)URS5H{n X}(dMX]nX-S4nX-S4nU(dUU4U ;aM/U R+UU5 U R-Xm5 U R-[/XUUR0US95 M} [R2"U 5U R4S 'U $) z$Find the argmax label for each span.rr"r$)axisr?rrAr#r)rTr rtr'r,rargmaxrtake_along_axis expand_dimsrrIrr logical_andrrsqueezer&rrgrrGrrr'r()r2rQrrrpr argmax_scoresr)rsort_idxseenrOr-r>rr[r\s r3r,SpanCategorizer._make_span_group_singlelabelWs ;;! Sxx0 0((0**..))'2MMqM) -- E%%i3!  9??$7  " "%%e$:P:P-PQEHH[)  %%ey.H-Q-Q-STE%--/"4==?H)3M!+I'GOE|#HH- w}}Q'(A8LEqDMEQ$-C 3<4'HHUC(    0 1 LLc#T[[5GH I) !& L 9 H r6)rrr'rrrr)spancat)r.N)T)1r7r8r9r:rr~r#rr r rrrr)rrr rrrrpropertyrtrrrrrrrr rrrrrr!rrrrrrrrr rrr;r1r6r3rrs AM$) +.(,&*%(%2AMAMU49f,-x78AM AM  AM!AMAM"%AM ~AMsmAME?AM"AM AMF*S** ss")c ))!DI!!ADcNAA$3$$5d#3 HSM  =IKSMK69K K$)HSM)d)@#'-1 '7#' ' i ' c5j)* ' c5j 'R< )<9>vx?O9P< ue| <D#'&* $$r8G#445$$h  $$ c# $$  $$L Xg->  W ' '' '  '\# 5 55 5  5  55r6rcUS:Xa"[R"S5nUR$US:Xa"[R"S5nUR$[ S[ SU35e)N make_spancatzspacy.pipeline.factoriesmake_spancat_singlelabelzmodule z has no attribute ) importlib import_moduler>r?AttributeErrorr7)rmodules r3 __getattr__rDsf ~(()CD""" + +(()CD... 78*,>tfE FFr6)Cr@ dataclassesr functoolsrtypingrrrrr r r r r r thinc.apirrrrrr thinc.typesrrrrcompatrrerrorsrlanguagerrrtokensrrr trainingr!r"rr#trainable_piper%spancat_default_config"spancat_singlelabel_default_configDEFAULT_SPANS_KEYfrom_strDEFAULT_SPANCAT_MODEL!DEFAULT_SPANCAT_SINGLELABEL_MODELr)rrXrr_rbrhrkr~rrrrDr1r6r3rVs!TTT VV880))1)6&"4))*@A'J$*H$5$5&% % ! YYY DH 3- $S 3;C= <BF 3-$'19# .1c1y1(#(((@C@I@ 2HW- 2DcN 2  (GmGVGr6