210 lines
20 KiB
Plaintext
210 lines
20 KiB
Plaintext
|
Ë
|
|||
|
|
>û g< ã ó — d dl mZ d dlmZmZmZmZmZmZm Z d dl
|
|||
|
|
Z
|
|||
|
|
d dl m Z m
|
|||
|
|
Z
|
|||
|
|
mZmZmZ d dlmZ ddlmZ ddlmZ dd lmZ dd
|
|||
|
|
lmZ dd lm Z m Z m Z dd l m Z dd
|
|||
|
|
l!m"Z" ddl#m$Z$ dZ% e « jM e%« d Z'dZ(dZ) ejT ddgde'ddidœdddddddddddœ
|
|||
|
|
¬ « d ed e+de
|
|||
|
|
ee ee f d e,d ee d!d"f d#„« Z-d$ee d!ee+ef fd%„Z. e j^ d« d&„ « Z0 G d'„ d"e$« Z1y)(é )Úislice)ÚAnyÚCallableÚDictÚIterableÚListÚOptionalÚTupleN)ÚConfigÚModelÚ OptimizerÚget_array_moduleÚset_dropout_rate)ÚFloats2dé )ÚErrors)ÚLanguage)ÚScorer)ÚDoc)ÚExampleÚvalidate_examplesÚvalidate_get_examples)Úregistry)ÚVocabé )Ú
|
|||
|
|
TrainablePipeaW
|
|||
|
|
[model]
|
|||
|
|
@architectures = "spacy.TextCatEnsemble.v2"
|
|||
|
|
|
|||
|
|
[model.tok2vec]
|
|||
|
|
@architectures = "spacy.Tok2Vec.v2"
|
|||
|
|
|
|||
|
|
[model.tok2vec.embed]
|
|||
|
|
@architectures = "spacy.MultiHashEmbed.v2"
|
|||
|
|
width = 64
|
|||
|
|
rows = [2000, 2000, 500, 1000, 500]
|
|||
|
|
attrs = ["NORM", "LOWER", "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 = 2
|
|||
|
|
|
|||
|
|
[model.linear_model]
|
|||
|
|
@architectures = "spacy.TextCatBOW.v3"
|
|||
|
|
exclusive_classes = true
|
|||
|
|
length = 262144
|
|||
|
|
ngram_size = 1
|
|||
|
|
no_output_layer = false
|
|||
|
|
Úmodelz€
|
|||
|
|
[model]
|
|||
|
|
@architectures = "spacy.TextCatBOW.v3"
|
|||
|
|
exclusive_classes = true
|
|||
|
|
length = 262144
|
|||
|
|
ngram_size = 1
|
|||
|
|
no_output_layer = false
|
|||
|
|
a`
|
|||
|
|
[model]
|
|||
|
|
@architectures = "spacy.TextCatReduce.v1"
|
|||
|
|
exclusive_classes = true
|
|||
|
|
use_reduce_first = false
|
|||
|
|
use_reduce_last = false
|
|||
|
|
use_reduce_max = false
|
|||
|
|
use_reduce_mean = true
|
|||
|
|
|
|||
|
|
[model.tok2vec]
|
|||
|
|
@architectures = "spacy.HashEmbedCNN.v2"
|
|||
|
|
pretrained_vectors = null
|
|||
|
|
width = 96
|
|||
|
|
depth = 4
|
|||
|
|
embed_size = 2000
|
|||
|
|
window_size = 1
|
|||
|
|
maxout_pieces = 3
|
|||
|
|
subword_features = true
|
|||
|
|
Útextcatzdoc.catsç z@scorerszspacy.textcat_scorer.v2)Ú thresholdr Úscorerç ð?)
|
|||
|
|
Ú
|
|||
|
|
cats_scoreÚcats_score_descÚ cats_micro_pÚ cats_micro_rÚ cats_micro_fÚ cats_macro_pÚ cats_macro_rÚ cats_macro_fÚcats_macro_aucÚcats_f_per_type)ÚassignsÚdefault_configÚdefault_score_weightsÚnlpÚnamer r! ÚreturnÚTextCategorizerc ó6 — t | j ||||¬« S )aÏ Create a TextCategorizer component. The text categorizer predicts categories
|
|||
|
|
over a whole document. It can learn one or more labels, and the labels are considered
|
|||
|
|
to be mutually exclusive (i.e. one true label per doc).
|
|||
|
|
|
|||
|
|
model (Model[List[Doc], List[Floats2d]]): A model instance that predicts
|
|||
|
|
scores for each category.
|
|||
|
|
threshold (float): Cutoff to consider a prediction "positive".
|
|||
|
|
scorer (Optional[Callable]): The scoring method.
|
|||
|
|
)r r! )r3 Úvocab)r0 r1 r r r! s úWC:\Users\garci\AppData\Roaming\Python\Python312\site-packages\spacy/pipeline/textcat.pyÚ make_textcatr7 M s € ôJ ˜3Ÿ9™9 e¨T¸YÈvÔ VÐVó Úexamplesc ó4 — t j | dfddi|¤ŽS )NÚcatsÚ multi_labelF)r Ú
|
|||
|
|
score_cats)r9 Úkwargss r6 Ú
|
|||
|
|
textcat_scorer? u s/ € Ü × Ñ ØØñ ð ð ð ñ ð r8 c ó — t S ©N)r? © r8 r6 Úmake_textcat_scorerrC ~ s € ä Ðr8 c ó — e Zd ZdZ d$edœdedededede e
|
|||
|
|
dd f d
|
|||
|
|
„Z e d „ « Z
|
|||
|
|
e dee fd „« Ze dee fd
|
|||
|
|
„« Zdee fd„Zdee dd fd„Zdd d dœdee dede e de eeef deeef f
|
|||
|
|
d„Zdd d dœdee dede e de eeef deeef f
|
|||
|
|
d„Zdee deej8 ej8 f fd„Z dee deeef fd„Z dede fd „Z d d d d œd e
|
|||
|
|
g ee f d e e! d e ee d!e e dd f
|
|||
|
|
d"„Z"dee fd#„Z#y )%r3 zmPipeline component for single-label text classification.
|
|||
|
|
|
|||
|
|
DOCS: https://spacy.io/api/textcategorizer
|
|||
|
|
)r! r5 r r1 r r! r2 Nc óv — || _ || _ || _ d| _ g |ddœ}t |« | _ || _ y)aa Initialize a text categorizer for single-label classification.
|
|||
|
|
|
|||
|
|
vocab (Vocab): The shared vocabulary.
|
|||
|
|
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): Unused, not needed for single-label (exclusive
|
|||
|
|
classes) classification.
|
|||
|
|
scorer (Optional[Callable]): The scoring method. Defaults to
|
|||
|
|
Scorer.score_cats for the attribute "cats".
|
|||
|
|
|
|||
|
|
DOCS: https://spacy.io/api/textcategorizer#init
|
|||
|
|
N)Úlabelsr Úpositive_label)r5 r r1 Ú_rehearsal_modelÚdictÚcfgr! )Úselfr5 r r1 r r! rJ s r6 Ú__init__zTextCategorizer.__init__‰ sE € ð, ˆŒ
|
|||
|
|
؈Œ
|
|||
|
|
؈Œ Ø $ˆÔ àØ"Ø "ñ
|
|||
|
|
ˆô
|
|||
|
|
˜“9ˆŒØ ˆ� r8 c ó — y)NFrB ©rK s r6 Úsupport_missing_valuesz&TextCategorizer.support_missing_values« s € ð
|
|||
|
|
r8 c ó2 — t | j d « S )z†RETURNS (Tuple[str]): The labels currently added to the component.
|
|||
|
|
|
|||
|
|
DOCS: https://spacy.io/api/textcategorizer#labels
|
|||
|
|
rF )ÚtuplerJ rN s r6 rF zTextCategorizer.labels² s € ô �T—X‘X˜hÑ'Ó(Ð(r8 c ó — | j S )z†RETURNS (List[str]): Information about the component's labels.
|
|||
|
|
|
|||
|
|
DOCS: https://spacy.io/api/textcategorizer#label_data
|
|||
|
|
)rF rN s r6 Ú
|
|||
|
|
label_datazTextCategorizer.label_dataº s € ð �{‰{Ðr8 Údocsc óš — t d„ |D « « ss|D �cg c] }|j ‘Œ }}| j j j }|j t
|
|||
|
|
t |« « t
|
|||
|
|
| j « f« }|S | j j |« }| j j j |« }|S c c}w )zþApply the pipeline's model to a batch of docs, without modifying them.
|
|||
|
|
|
|||
|
|
docs (Iterable[Doc]): The documents to predict.
|
|||
|
|
RETURNS: The models prediction for each document.
|
|||
|
|
|
|||
|
|
DOCS: https://spacy.io/api/textcategorizer#predict
|
|||
|
|
c 3 ó2 K — | ] }t |« –— Œ y wrA ©Úlen©Ú.0Údocs r6 ú <genexpr>z*TextCategorizer.predict.<locals>.<genexpr>Ê ó è ø€ Ð,¡t ”3�s—8¡tùó ‚) ÚanyÚtensorr ÚopsÚxpÚzerosrX ÚlistrF ÚpredictÚasarray)rK rT r[ Útensorsrb Úscoress r6 re zTextCategorizer.predict sœ € ô Ñ,¡tÓ,Ô,á-1Ó2©T c�s—z“z¨TˆGÐ2Ø—‘— ‘ ×"Ñ"ˆBØ—X‘Xœs¤4¨£:›´°D·K±KÓ0@Ð AÓBˆF؈MØ—‘×#Ñ# DÓ)ˆØ—‘— ‘ ×'Ñ'¨Ó/ˆØˆ
|
|||
|
|
ùò
|
|||
|
|
3s —Cc óž — t |« D ]? \ }}t | j « D ]" \ }}t |||f « |j |< Œ$ ŒA y)a Modify a batch of Doc objects, using pre-computed scores.
|
|||
|
|
|
|||
|
|
docs (Iterable[Doc]): The documents to modify.
|
|||
|
|
scores: The scores to set, produced by TextCategorizer.predict.
|
|||
|
|
|
|||
|
|
DOCS: https://spacy.io/api/textcategorizer#set_annotations
|
|||
|
|
N)Ú enumeraterF Úfloatr; )rK rT rh Úir[ ÚjÚlabels r6 Úset_annotationsz TextCategorizer.set_annotationsÔ sG € ô –o‰FˆAˆsÜ % d§k¡kÖ 2‘��5Ü"'¨¨q°!¨t© Ó"5�—‘˜’ñ 3ñ &r8 r )ÚdropÚsgdÚlossesr9 rp rq rr c óØ — |€i }|j | j d« t |d« | j |« t d„ |D « « s|S t | j |« | j j |D �cg c] }|j ‘Œ c}« \ }}| j ||« \ }} || « |�| j |« || j xx |z
|
|||
|
|
cc< |S c c}w )a1 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 (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/textcategorizer#update
|
|||
|
|
r zTextCategorizer.updatec 3 ób K — | ]' }|j rt |j « nd –— Œ) yw)r N)Ú predictedrX )rZ Úegs r6 r\ z)TextCategorizer.update.<locals>.<genexpr>ù s% è ø€ ÐOÁhÀ¨¯ ª ”3�r—|‘|Ô$¸!Ó;Áhùs ‚-/) Ú
|
|||
|
|
setdefaultr1 r Ú_validate_categoriesr_ r r Ú begin_updateru Úget_lossÚ
|
|||
|
|
finish_update)
|
|||
|
|
rK r9 rp rq rr rv rh Ú bp_scoresÚlossÚd_scoress
|
|||
|
|
r6 ÚupdatezTextCategorizer.updateà sÖ € ð( ˆ>؈FØ×ј$Ÿ)™) SÔ)ܘ(Ð$<Ô=Ø ×!Ñ! (Ô+ÜÑOÁhÓOÔOàˆMܘŸ™ TÔ*Ø ŸJ™J× 3Ñ 3ÉHÓ4UÉHÀb°R·\³\ÈHÑ4UÓ Vш� Ø Ÿ ™ x°Ó8‰ˆˆhÙ�(ÔØ ˆ?Ø × Ñ ˜sÔ #؈t�y‰yÓ˜TÑ!Ó؈
|
|||
|
|
ùò
|
|||
|
|
5Vs ÂC'c ó4 — |€i }|j | j d« | j €|S t |d« | j |« |D �cg c] }|j
|
|||
|
|
‘Œ }}t
|
|||
|
|
d„ |D « « s|S t | j |« | j j |« \ }}| j j |« \ } }
|
|||
|
|
|| z
|
|||
|
|
} || « |�| j |« || j xx | dz j « z
|
|||
|
|
cc< |S c c}w )a« Perform a "rehearsal" update from a batch of data. Rehearsal updates
|
|||
|
|
teach the current model to make predictions similar to an initial model,
|
|||
|
|
to try to address the "catastrophic forgetting" problem. This feature is
|
|||
|
|
experimental.
|
|||
|
|
|
|||
|
|
examples (Iterable[Example]): A batch of Example objects.
|
|||
|
|
drop (float): The dropout rate.
|
|||
|
|
sgd (thinc.api.Optimizer): The optimizer.
|
|||
|
|
losses (Dict[str, float]): Optional record of the loss during training.
|
|||
|
|
Updated using the component name as the key.
|
|||
|
|
RETURNS (Dict[str, float]): The updated losses dictionary.
|
|||
|
|
|
|||
|
|
DOCS: https://spacy.io/api/textcategorizer#rehearse
|
|||
|
|
r zTextCategorizer.rehearsec 3 ó2 K — | ] }t |« –— Œ y wrA rW rY s r6 r\ z+TextCategorizer.rehearse.<locals>.<genexpr># r] r^ r ) rw r1 rH r rx ru r_ r r ry r{ Úsum) rK r9 rp rq rr rv rT rh r| ÚtargetÚ_Úgradients r6 ÚrehearsezTextCategorizer.rehearse s € ð, ˆ>؈FØ×ј$Ÿ)™) SÔ)Ø × Ñ Ð (؈Mܘ(Ð$>Ô?Ø ×!Ñ! (Ô+Ù'/Ó0¡x �— “ xˆÐ0ÜÑ,¡tÓ,Ô,àˆMܘŸ™ TÔ*Ø ŸJ™J× 3Ñ 3°DÓ 9ш� Ø×)Ñ)×6Ñ6°tÓ<‰ ˆ�ؘF‘?ˆÙ�(ÔØ ˆ?Ø × Ñ ˜sÔ #؈t�y‰yÓ˜h¨™k× .Ñ .Ó 0Ñ0Ó؈
|
|||
|
|
ùò 1s ÁDc ó" — t t |« « }t j |t | j « fd¬« }t j
|
|||
|
|
|t | j « fd¬« }t
|
|||
|
|
|« D ]m \ }}t
|
|||
|
|
| j « D ]P \ }}||j j v r |j j | |||f< Œ=| j sŒJd|||f< ŒR Œo | j j j |« }||fS )NÚf)Údtyper )
|
|||
|
|
rX rd Únumpyrc rF Úonesrj Ú referencer; rO r ra rf ) rK r9 Ú nr_examplesÚtruthsÚ not_missingrl rv rm rn s r6 Ú_examples_to_truthz"TextCategorizer._examples_to_truth0 sâ € ô œ$˜x›.Ó)ˆ Ü—‘˜k¬3¨t¯{©{Ó+;Ð <ÀCÔHˆÜ—j‘j +¬s°4·;±;Ó/?Ð!@ÈÔLˆ Ü ˜xÖ(‰EˆAˆrÜ % d§k¡kÖ 2‘��5ؘBŸL™L× -Ñ -Ñ-Ø#%§<¡<×#4Ñ#4°UÑ#;�F˜1˜a˜4’LØ×0Ó0Ø(+�K 1 Ò%ñ 3ð )ð —‘— ‘ ×'Ñ'¨Ó/ˆØ�{Ð"Ð"r8 c ó — t |d« | j |« | j |« \ }}| j j j |« }||z
|
|||
|
|
}||z }|dz j
|
|||
|
|
« }t |« |fS )ae Find 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, float]): The loss and the gradient.
|
|||
|
|
|
|||
|
|
DOCS: https://spacy.io/api/textcategorizer#get_loss
|
|||
|
|
zTextCategorizer.get_lossr )r rx r� r ra rf Úmeanrk )rK r9 rh rŽ r� r~ Úmean_square_errors r6 rz zTextCategorizer.get_loss? s� € ô ˜(Ð$>Ô?Ø ×!Ñ! (Ô+Ø "× 5Ñ 5°hÓ ?ш� Ø—j‘j—n‘n×,Ñ,¨[Ó9ˆ ؘF‘?ˆØ�KÑ ˆØ % q™[× .Ñ .Ó 0ÐÜÐ&Ó'¨Ð1Ð1r8 rn c óæ — t |t « st t j « ‚|| j
|
|||
|
|
v ry| j
|
|||
|
|
« | j d j |« | j rZd| j j v rB | j j d | j t | j
|
|||
|
|
« « | _ | j j j |« y)zÎAdd 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/textcategorizer#add_label
|
|||
|
|
r rF Ú
|
|||
|
|
resize_outputr )Ú
|
|||
|
|
isinstanceÚstrÚ
|
|||
|
|
ValueErrorr ÚE187rF Ú_allow_extra_labelrJ Úappendr ÚattrsrX r5 ÚstringsÚadd)rK rn s r6 Ú add_labelzTextCategorizer.add_labelR s® € ô ˜%¤Ô%Ü œVŸ[™[Ó)Ð )Ø �D—K‘KÑ ØØ × Ñ Ô!Ø �‰�Ñ×!Ñ! %Ô(Ø �:Š:˜/¨T¯Z©Z×-=Ñ-=Ñ=Ø:˜Ÿ™×)Ñ)¨/Ñ:¸4¿:¹:ÄsÈ4Ï;É;ÓGWÓXˆDŒJØ �
|
|||
|
|
‰
|
|||
|
|
×Ñ× Ñ ˜uÔ%Ør8 )r0 rF rG Ú get_examplesr0 rF rG c ó| — t |d« | j |« « |€9 |« D ]. }|j j D ] }| j |« Œ Œ0 n|D ] }| j |« Œ t | j « dk rt t j « ‚|�’|| j vr6t j j || j ¬« }t |« ‚t | j « dk7 r6t j j || j ¬« }t |« ‚|| j d< t t |« d« « } | D �
|
|||
|
|
cg c] }
|
|||
|
|
|
|
|||
|
|
j ‘Œ } }
|
|||
|
|
| j# | « \ } }
|
|||
|
|
| j% « t | « dkD s/J t j& j | j( ¬« « ‚t | « dkD s/J t j& j | j( ¬« « ‚| j* j- | | ¬ « yc c}
|
|||
|
|
w )
|
|||
|
|
a Initialize 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 (Language): The current nlp object the component is part of.
|
|||
|
|
labels (Optional[Iterable[str]]): The labels to add to the component, typically generated by the
|
|||
|
|
`init labels` command. If no labels are provided, the get_examples
|
|||
|
|
callback is used to extract the labels from the data.
|
|||
|
|
positive_label (Optional[str]): The positive label for a binary task with exclusive classes,
|
|||
|
|
`None` otherwise and by default.
|
|||
|
|
|
|||
|
|
DOCS: https://spacy.io/api/textcategorizer#initialize
|