i,)>ddlZddlmZmZmZmZmZmZmZddl m Z m Z m Z m Z ddlmZddlmZddlmZddlmZdd lmZmZdd lmZd d lmZd d lmZdZe jAedZ!dZ"deedee#effdZ$dedee%e%ffdZ&GddeZ'dZ(y)N)AnyCallableDictIterableListOptionalTuple)ConfigModel Optimizerset_dropout_rate)Floats2d)Errors)Language)Scorer)DocSpan)Example)DEFAULT_SPANS_KEY) TrainablePipea [model] @architectures = "spacy.SpanFinder.v1" [model.scorer] @layers = "spacy.LinearLogistic.v1" nO = 2 [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 modelctSN)span_finder_scorek/var/www/vps2.regionflexible.com/Desarrollo/venv/lib/python3.12/site-packages/spacy/pipeline/span_finder.pymake_span_finder_scorerr ,s rexamplesreturnc Rt|}d|d|jd|jdfd|jdfd|jdd |jd d tj|fi|}|j |dd d|S) Nspans_ spans_keyattrgettercT|jj|tdgSr)spansgetlen)dockey attr_prefixs rz#span_finder_score..6s#399==S5E5G1H"#Mrhas_annotationc |jvSr)r))r,r-s rr/z#span_finder_score..8sC3994Dr allow_overlapTlabeledF _per_type)dict setdefaultr score_spanspop)r!kwargsscoresr.r-s @@rrr0s &\FK  C f cU34 M &(DE ot, i'    3F 3F JJ&.!+T2 Mrspancd|dj}|djt|dz}||fS)Nr)idxr+)r;startends r _char_indicesrA@s3 GKKE r(,,T"X &C #:rcDeZdZdZ dedddeddedeee e fde de d e d e ed e ed e ed dfdZdee fdZdee de d dfdZdddddeede de ede ee e fd ee e ff dZd ee e ffdZd ee e ffdZdddegeefde ed dfdZy) SpanFinderzUPipeline that learns span boundaries. DOCS: https://spacy.io/api/spanfinder g?N)r% threshold max_length min_lengthscorernlprnamer%rDrErFrGr"c|j|_||dks|/|dkr*ttjj ||||_||_||_||||d|_y)aInitialize the span finder. model (thinc.api.Model): The Thinc Model powering the pipeline component. name (str): The component instance name, used to add entries to the losses during training. threshold (float): Minimum probability to consider a prediction positive. scorer (Optional[Callable]): The scoring method. 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. max_length (Optional[int]): Maximum length of the produced spans, defaults to None meaning unlimited length. min_length (Optional[int]): Minimum length of the produced spans, defaults to None meaning shortest span length is 1. DOCS: https://spacy.io/api/spanfinder#init Nr)rFrE)rFrErDr%) vocab ValueErrorrE1053formatrrIrGcfg) selfrHrrIr%rDrErFrGs r__init__zSpanFinder.__init__Lsw<YY  "zA~  "zA~ ##zj#Q    $$"" $ rdocsc<|jj|}|S)aApply 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/spanfinder#predict )rpredict)rPrRr:s rrTzSpanFinder.predict{s##D) rr:c d}t|D]K\}}g|j|jd<g}g}|||t|z}t ||D]f\} } | d|jdk\r|j | j | d|jdk\sL|j | j h|D]} |D]} | dz| z } | dkr|jd|jd| ks3|jd| |jdksU|j|jdj || | dz|t|z }Ny)aModify a batch of Doc objects, using pre-computed scores. docs (Iterable[Doc]): The documents to modify. scores: The scores to set, produced by SpanFinder predict method. DOCS: https://spacy.io/api/spanfinder#set_annotations rr%rDrrFNrE) enumerater)rOr+zipappendi)rPrRr:offsetrYr,startsends doc_scorestoken token_scorer?r@ span_lengths rset_annotationszSpanFinder.set_annotationssro FAs/1CIIdhh{+ ,FD#c():;J&)#z&: )"{q>TXXk%::MM%''*q>TXXk%::KK(  )  V VC"%'E/K"Q .688L1[@.6&$((<*@@ $((;"78??ECRSG@TU V V c#h F3 r)dropsgdlossesr!rcrdrecx|i}|j|jd|Dcgc]}|j}}t|j||jj |\}}|j ||\} } || ||j|||jxx| z cc<|Scc}w)a?Learn from a batch of documents and gold-standard information, updating the pipe's model. Delegates to predict and get_loss. examples (Iterable[Example]): A batch of Example objects. drop (float): The dropout rate. sgd (Optional[thinc.api.Optimizer]): The optimizer. losses (Optional[Dict[str, float]]): Optional record of the loss during training. Updated using the component name as the key. RETURNS (Dict[str, float]): The updated losses dictionary. DOCS: https://spacy.io/api/spanfinder#update rb)r6rI predictedr r begin_updateget_loss finish_update) rPr!rcrdreegrgr:backprop_scoreslossd_scoress rupdatezSpanFinder.updates& >F$))S),45bR\\5 5T*"&**"9"9)"Dx8h! ?   s #tyyT! 6sB7c|j||jj\}}||jjj|z }||z}t |dzj }||fS)abFind the loss and gradient of loss for the batch of documents and their predicted scores. examples (Iterable[Examples]): The batch of examples. scores: Scores representing the model's predictions. RETURNS (Tuple[float, Floats2d]): The loss and the gradient. DOCS: https://spacy.io/api/spanfinder#get_loss r)_get_aligned_truth_scoresrops asarray2ffloatsum)rPr!r:truthsmasksrnrms rrizSpanFinder.get_losssj66xP DJJNN44V<<Ehk&&()X~rcg}g}|D]}|jj|jjk7r)ttj j dt|j}|jj|dfd}|jj|dfd}|jd|jjvr|jj|jdD]} t| \} } |jj!| | d} t| \} }| | k(}|| k(}|rd || d j"d f<nd || d j"d f<|rd || d j"d f<d || d j"d f<|j%||j%||jj'|d }|jj'|d }||fS) z\Align scores of the predictions to the references for calculating the loss. span_finder) componentrfloat32)dtyper%expand)alignment_moderrr=)axis)xtextyrLrE1054rNr+rgxpzerosonesrO referencer)rA char_spanrYrX concatenate)rPr!rrrvrwrkn_tokenstruthmaskr;ref_start_char ref_end_char pred_spanpred_start_char pred_end_char start_match end_matchs rrqz$SpanFinder._get_aligned_truth_scoress BttyyBDDII% !4!4}!4!MNN2<<(HFFLL(AiL@E66;;!}I;>Dxx $ (:(::LL..txx /DE5D3@3F0NL " 6 6& X!7!I6C95M2O]"1^"CK - =I"34ilnna/023Yq\^^Q./ 45imooq0134Yr]__a/05 MM% LL 1 2##F#3""5q"1u}r)rH get_examplescbg}|D]"}t|dks|j|$|r`|Dcgc]}|j}}|j||jj \}}|jj ||y|jj ycc}w)aInitialize the pipe for training, using a representative set of data examples. get_examples (Callable[[], Iterable[Example]]): Function that returns a representative sample of gold-standard Example objects. nlp (Optional[Language]): The current nlp object the component is part of. DOCS: https://spacy.io/api/spanfinder#initialize )XYN)r+rXrrqrrr initialize)rPrrHsubbatchrkrRr_s rrzSpanFinder.initializes#%. $B8}r!# $ +34RBLL4D411(DJJNNKDAq JJ ! !DA ! . JJ ! ! # 5sB,)ry)__name__ __module__ __qualname____doc__rrrr rrrstrrtrintrrQrTrarr rror rirqrrrrrCrCFs" - +$($(%6- - Xc]H,--  -  - - SM- SM- "-  - ^ HSM !HSM!8!!N#'-1 7#  i  c5j)*  c5j @ E%/,B !%(@R:S!N#' $r8G#445$h  $  $rrCcx|dk(r!tjd}|jStdtd|)Nmake_span_finderzspacy.pipeline.factorieszmodule z has no attribute ) importlib import_modulerAttributeErrorr)rImodules r __getattr__rsA !!(()CD&&& 78*,>tfE FFr))rtypingrrrrrrr thinc.apir r r r thinc.typesrerrorsrlanguagerrGrtokensrrtrainingrspancatrtrainable_piperspan_finder_default_configfrom_strDEFAULT_SPAN_FINDER_MODELr rrrrArCrrrrrsGGG@@ &)4#H--.HI'R  1 S#X  sCx P$P$hGr