Files
INTUIA/Programa final/spacy/tests/pipeline/__pycache__/test_tok2vec.cpython-312.pyc
T

315 lines
30 KiB
Plaintext
Raw Normal View History

2026-03-15 13:27:50 +00:00
Ë
?û g¢Wãó<ddlZddlmZddlmZmZddlmZddlm Z ddl
m Z m Z m
Z
ddlmZmZddlmZdd lmZdd
lmZdd lmZd d
lmZmZmZdZej:j=dgd¢gd¢gd¢g«d«Zej:j@ej:j=ddg«ej:j=ddddgddgddœfdddgddgddœfd dd!ddd"œfd dd#d dd"œfg«ej:j=d$d%d&d'd(d d)œfd*d+d'd(d d)œfd%d,d'd-d.œfd*d/d'd-d.œfg«d0„««««Z!d1„Z"d2Z#d3gd4¢d5d6d7œd8œfd9gd:¢d6d5d7œd8œfgZ$ej:j=d;d<«d=„«Z%d>„Z&d?„Z'd@„Z(dA„Z)dB„Z*dCZ+dD„Z,dEZ-dF„Z.dG„Z/dH„Z0y)IéN)Úassert_array_equal)ÚConfigÚget_current_ops)Úutil)ÚEnglish)ÚMaxoutWindowEncoderÚMultiHashEmbedÚbuild_Tok2Vec_model)ÚTok2VecÚTok2VecListener)ÚDoc)ÚExample)Úregistry)ÚVocabé)Úadd_vecs_to_vocabÚ get_batchÚ make_tempdirc ó(d}d}t«}t|g¬«}tt|||||gdgd¢¬«t |ddd ¬
««}|j «|j
|g«\}}t|«dk(sJ|d jd |fk(sJy) Né€éÐ)ÚwordsF©ÚNORMÚPREFIXÚSUFFIXÚSHAPE©ÚwidthÚrowsÚinclude_static_vectorsÚattrsééé©rÚdepthÚ window_sizeÚ
maxout_piecesr) rr
r
r rÚ
initializeÚ begin_updateÚlenÚshape)rÚ
embed_sizeÚvocabÚdocÚtok2vecÚvectorsÚbackprops úbC:\Users\garci\AppData\Roaming\Python\Python312\site-packages\spacy/tests/pipeline/test_tok2vec.pyÚtest_empty_docr5Ø €EØ€JÜ ‹G€EÜ
ˆe˜
€CÜØØ˜j¨*°jÐAØ#(Ú
ô  %¨q¸aÈqÔ€Gð ×ÑÔØ×,¨c¨UÓ€GˆXÜ ˆw‹<˜1Ò ÐÐ Ø 1‰:× Ñ   5˜zÒ  zbatch_size,width,embed_size)r$rr)rrr)r%éé?c óLt|«}tt||gdzdgd¢¬«t|ddd¬««}|j «|j |«\}}t
|«t
|«k(sJt||«D]!\}}|jt
|«|fk(rŒ!Jy)Nr#Frrr$r%r&) rr
r rr*r+r,Úzipr-) Ú
batch_sizerr.Úbatchr1r2r3Údoc_vecr0s r4Útest_tok2vec_batch_sizesr>)ô
 !€EÜØØ Ñ!Ø#(Ú
ô  %¨q¸aÈqÔ€Gð ×ÑÔØ×,¨UÓ€GˆXÜ ˆw<œ3˜u  ˜G UÖ+‰ ˆØ}‰}¤ S£¨5Ð 1Ó,r6rr7zembed_arch,embed_configzspacy.MultiHashEmbed.v1édrÚLOWERF)r r"r!éÚORTHrzspacy.CharacterEmbed.v1é@)r ÚnMÚnCr!éz&tok2vec_arch,encode_arch,encode_configzspacy.Tok2Vec.v1zspacy.MaxoutWindowEncoder.v1r$r%)r(r)r'úspacy.Tok2Vec.v2zspacy.MaxoutWindowEncoder.v2zspacy.MishWindowEncoder.v1é)r(r'zspacy.MishWindowEncoder.v2có¬tjd|«}tjd|«}tjd|«}||d<||d<td«} ||di|¤Ž|di|¤Ž«}
|
j| «|
j | «\} } t | «t | «k(sJ| dj t | d«|fk(sJ| | «y)
architecturesrr%r©)rÚgetrr*r+r,r-)
rÚ tok2vec_archÚ
embed_archÚ embed_configÚ encode_archÚ
encode_configÚembedÚencodeÚ
tok2vec_modelÚdocsr1r2r3s
r4Útest_tok2vec_configsrV>ô6
L‰L˜¨*Ó 5€EÜ
\‰\˜/¨;Ó
7€FÜ—L‘L °,Ó?€Mà!€LÑØ"€MÜ Q<€DÙ™EÑ1 1±6Ñ3J¸MÑ3JÓK€GØ ×ÑØ×,¨TÓ€GˆXÜ ˆw<œ3˜t  1‰:× Ñ ¤ D¨¡G£ ¨eÐ   ˆr6có¶t«}|jd«}|jgk(sJ|j«|jj d«sJy)Nr1ÚnO)rÚadd_pipeÚ listenersr*ÚmodelÚget_dim)Únlpr1s r4Útest_init_tok2vecr^hsMä
‹)€CØl‰l˜9Ó%€GØ × Ñ  Ò  ‡NØ =‰=× Ñ  Ô  &r6a„
[nlp]
lang = "en"
pipeline = ["tok2vec","tagger"]
[components]
[components.tagger]
factory = "tagger"
[components.tagger.model]
@architectures = "spacy.Tagger.v2"
nO = null
[components.tagger.model.tok2vec]
@architectures = "spacy.Tok2VecListener.v1"
width = ${components.tok2vec.model.encode.width}
[components.tok2vec]
factory = "tok2vec"
[components.tok2vec.model]
@architectures = "spacy.Tok2Vec.v2"
[components.tok2vec.model.embed]
@architectures = "spacy.MultiHashEmbed.v1"
width = ${components.tok2vec.model.encode.width}
rows = [2000, 1000, 1000, 1000]
attrs = ["NORM", "PREFIX", "SUFFIX", "SHAPE"]
include_static_vectors = false
[components.tok2vec.model.encode]
@architectures = "spacy.MaxoutWindowEncoder.v2"
width = 96
depth = 4
window_size = 1
maxout_pieces = 3
úI like green eggs©ÚJraçð?ç)Ú
preferenceÚ
imperative)ÚtagsÚcatsú Eat blue ham©rbrcraÚ with_vectors)FTc ó
t«jt«}||ddddd<tj|dd¬«}|r†t «}d|j
gd ¢«fd
|j
gd ¢«fd |j
gd
¢«fd|j
gd¢«fd|j
gd¢«fg}t|j|«|jddgk(sJ|jd«}|jd«}|jjd«}t|t«sJt|t«sJgŠt D]Z}j#t%j&|j)|d«|d««|ddD]} |j+| «ŒŒ\|j-ˆfd«}
|j.|gk(sJt1d«D]} i} |j3|
| ¬«Œ|d«}
|j5|
g«d}t «}t7|j9|
j:«|j9|««|d«}
|j=d¬«|jdgk(sJ|d«y)
componentsr1r[rRr!Ú auto_fillÚvalidateÚapple)r$rr%Úorange)éÿÿÿÿéþÿÿÿéýÿÿÿÚand)rtrtrtÚjuice)éryé
Úpie)ég333333@gÍÌÌÌÌÌ!@Útaggerrr$rhcóS©NrK©Útrain_exampless€r4ú<lambda>z'test_tok2vec_listener.<locals>.<lambda>Ä󸀡~r6ry©ÚsgdÚlossesz Running the pipeline as a whole.Ú)Údisablez9Running the pipeline with the Tok2Vec component disabled.)rÚfrom_strÚ
cfg_stringrÚload_model_from_configrÚasarrayrr/Ú
pipe_namesÚget_piper[Úget_refÚ
isinstancer r Ú
TRAIN_DATAÚappendrÚ from_dictÚmake_docÚ add_labelr*rZÚrangeÚupdateÚpredictrÚto_numpyÚtensorÚ select_pipes)rlÚ orig_configr]Úopsr2r}r1Útagger_tok2vecÚtagÚ optimizerÚir†r0Ú
doc_tensorrs @r4Útest_tok2vec_listenerr¤¤s^ø€ä“(×#¤JÓ/€Kð ð јiÑÑ1°'Ñ ñô ×
% k¸TÈDÔ
Q€CáÜÓˆà
c—kk¢)Ó
s—{{¢<Ó
C—K Ó
c—kk¢*Ó
C—K
Ó 
ˆô ˜#Ÿ)™)  >‰>˜Ð 
\‰\˜
#€FØl‰l˜9Ó%€GØ—\‘\×)¨)Ó4€NÜ gœwÔ  n¤oÔ  €Nß
ˆØ×Ñœg×· ± ¸Q¸q¹TÓ0BÀAÀaÁDÓQ‘4˜”<ˆCØ × Ñ ˜SÕ  ðð Ó6€IØ × Ñ  Ð 0Ò 
1ŽXˆØˆØ
> y¸ˆ
Õñ Ð
1€CØ×¨Ó.¨qÑ1€JÜ
Ó
€CÜs—|‘| C§J¡JÓ·±¸jÓ1IÔ ˆb‹'€Cð×Ñ˜Ô >‰>˜h˜  ÐDr6c óöt«jt«}tj|dd¬«}|j
ddgk(sJ|j
d«}|j
d«}|jd«g}|jj|¬«|Dcgc]}dDcgc]}dŒc}Œ}}}|jjj|d ¬
«g}|jj||¬ «|jd «g}|j|D cgc]} tj| i«Œc} «|jj|«\}
} | |
«Jycc}wcc}}wcc} w)
NTror1r}zA random sentence)ÚX)rbÚZrdÚfloat32)Údtype)ÚYz Another entirely random sentence)rr‰rrrr”r[r*rr—rr“r+) r]r}r1rUÚwordr Ú
gold_arrayÚ label_sampleÚxrªÚget_dXs r4Útest_tok2vec_listener_callbackr°ÙsMÜ“(×#¤JÓ/€KÜ
×
% k¸TÈDÔ
Q€CØ >‰>˜Ð 
\‰\˜
#€FØl‰l˜%€GØ L‰LÐ .€DØ ‡MM×јÔ$Ù:>Ô?¹$°$¡*Ó-¡*˜33 -¸$€JÑ—L‘L×,¨Z¸yÐJ€LØ
‡Lјd lÐÔ L‰LÐ =€DØ ‡NN±dÓ;±d°”G×% Õ,°dÑ ×)¨$Ó/I€A€vá !‹9Ð Ð Ñ ùò.ùÓ?ùò<s E0Â' E+ Â0E0ÄE6Å+E0c ó| t«jt«}tj|dd¬«}gŠ t
D]<} j
tj|j|d«|d««Œ>|jˆ fd¬«}td«D]}i}|j ||dg¬ «Œd
d ksJd }||«}|djd
k(sJ|djdk(sJ|djdk(sJ|djd
k(sJt«5}|j|«tj |«} | |«}
|
djd
k(sJ|
djdk(sJ|
djdk(sJ|
djd
k(sJ ddd«y#1swYyxYw)ziTest that a pipeline with a listener properly overfits, even if 'tok2vec' is in the annotating componentsTrorr$cóSrrKr€s€r4rz3test_tok2vec_listener_overfitting.<locals>.<lambda>òóø€±Nr6©Ú get_examplesé2r1)r…r†Ú annotatesr}gñh㈵øä>úI like blue eggsrarbrrcr%rr‰rrrrrr“r”r*rr—Útag_rÚto_diskÚload_model_from_path© r]r†Ú test_textr0Útmp_dirÚnlp2Údoc2rs @r4Ú!test_tok2vec_listener_overfittingrÂëø€ä“(×#¤JÓ/€KÜ
×
% k¸TÈDÔ
Q€CØ€Nß
ˆØ×Ñœg×· ± ¸Q¸q¹TÓ0BÀAÀaÁDÓàÓ,BÓC€Iä
2ŽYˆØˆØ
> y¸ÈIÈ;ˆ
Õð  ˜gÒ  #€IÙ
ˆi‹.€CØ ˆq‰6;‰;˜#Ò ÐÐ Ø ˆq‰6;‰;˜#Ò ÐÐ Ø ˆq‰6;‰;˜#Ò ÐÐ Ø ˆq‰6;‰;˜#Ò ÐÐ ô
Œ˜7Ø Ü×ÓÙI‹ˆØA‰w|‰|˜A‰w|‰|˜A‰w|‰|˜A‰w|‰|˜
ús
Ä)A?F2Æ2F;c óØt«jt«}tj|dd¬«}gŠt
D]<}j
tj|j|d«|d««Œ>|jˆfd¬«}td«D]>}i}tjtd¬ «5|j||d
g¬ «d d d «Œ@y #1swYŒKxYw)
z]Test that a pipeline with a frozen tok2vec raises an error when the tok2vec is not annotatingTrorr$cóSrrKr€s€r4rz4test_tok2vec_frozen_not_annotating.<locals>.<lambda>r6r´rz*the tok2vec embedding layer is not updated)Úmatchr1)r…r†ÚexcludeN)rr‰rrrrrr“r”r*rÚpytestÚraisesÚ
ValueErrorr—)r]r†rs @r4Ú"test_tok2vec_frozen_not_annotatingrÊ ø€ä“(×#¤JÓ/€KÜ
×
% k¸TÈDÔ
Q€CØ€Nß
ˆØ×Ñœg×· ± ¸Q¸q¹TÓ0BÀAÀaÁDÓàÓ,BÓC€Iä
1ŽXˆØˆÜ
]‰]Ü Ð
ð
J‰JØ I°fÀyÀkð
ô
÷
ð
ñ÷
ð
ús Â>C à C) c ó€ t«jt«}tj|dd¬«}gŠ t
D]<} j
tj|j|d«|d««Œ>|jˆ fd¬«}td«D]}i}|j ||dgdg¬ «Œd
d ksJd }||«}|djd
k(sJ|djdk(sJ|djdk(sJ|djd
k(sJt«5}|j|«tj |«} | |«}
|
djd
k(sJ|
djdk(sJ|
djdk(sJ|
djd
k(sJ ddd«y#1swYyxYw)zITest that a pipeline with a frozen & annotating tok2vec can still overfitTrorr$cóSrrKr€s€r4rz1test_tok2vec_frozen_overfitting.<locals>.<lambda>&r6r´r?r1)r…r†r}g-Cëâ6?r¸rarbrrcr%Nr¹s @r4Útest_tok2vec_frozen_overfittingrÍø€ä“(×#¤JÓ/€KÜ
×
% k¸TÈDÔ
Q€CØ€Nß
ˆØ×Ñœg×· ± ¸Q¸q¹TÓ0BÀAÀaÁDÓàÓ,BÓC€Iä
3ŽZˆØˆØ
Ø ØØØ õ
ðð  ˜fÒ  #€IÙ
ˆi‹.€CØ ˆq‰6;‰;˜#Ò ÐÐ Ø ˆq‰6;‰;˜#Ò ÐÐ Ø ˆq‰6;‰;˜#Ò ÐÐ Ø ˆq‰6;‰;˜#Ò ÐÐ ô
Œ˜7Ø Ü×ÓÙI‹ˆØA‰w|‰|˜A‰w|‰|˜A‰w|‰|˜A‰w|‰|˜
ús
Ä+A?F4Æ4F=cóøt«jt«}tj|dd¬«}t j |jd«dddgi«gŠ|jˆfd«|jd«}|jd «}t|jjd
t«sJ|jd d
|jjd
k(sJ|jd dd d
dk(sJ|jd d d dd
dk(sJ|j!dd dg«t|jjd
t«rJ|jd dd }|d
dk(sJ|jd d d d|k(sJt#j$t&«5|j!dd dg«ddd«t#j$t&«5|j!dddg«ddd«t#j$t&«5|j!dd dg«ddd«t#j$t&«5|j!dd ddg«ddd«|jˆfd«}t)d«D],}i}|j+||¬«|ddk(sJ|d dkDrŒ,Jy#1swYŒûxYw#1swYŒÑxYw#1swYŒ§xYw#1swYŒ|xYw)NTroúx yrhrbcóSrrK©Úexampless€r4rz(test_replace_listeners.<locals>.<lambda>Jóø€™8r6r1r}rrnr[ú@architecturesrGúspacy.Tok2VecListener.v1ú
model.tok2vecÚinvalidÚparserz
model.yolocóSrrKs€r4rz(test_replace_listeners.<locals>.<lambda>esø€¡xr6rr„re)rr‰rrrr“r”r*rr[Úlayersr Ú listener_mapÚconfigÚreplace_listenersrÇrr—) rœr]r1r}Út2v_cfgr¡r†s @r4Útest_replace_listenersrßFø€Ü“(×#¤JÓ/€KÜ
×
% k¸TÈDÔ
Q€CÜ×! #§,¡,¨uÓ"5¸ÀÀcÀ
Ð7KÓM€H؇Nl‰l˜9Ó%€GØ
\‰\˜
#€FÜ f—ll×)¨!Ñ,¬oÔ  × Ñ  Ñ )¨!Ñ · ± ×0CÑ0CÀAÑ0FÒ  
  Ñ+¨GÑ4Ð5EÑ ò ðð ð
  Ñ*¨7Ñ3°IÑ>Ð?OÑ ð×ј) Ð/@Ô˜&Ÿ,™,×-¨aÑ0´ j‰j˜Ñ& 1°'Ñ:€GØ Ð $Ð(:Ò  :‰: #  -¨gÑ 6°yÑ AÀWÒ  ”zÕ "Ø ×ј°OÐ3DÔ
”zÕ "Ø ×ј°OÐ3DÔ
”zÕ "Ø ×ј°L°>Ô
”zÕ "Ø ×ј°OÀ\Ð3RÔ
Ó0€IÜ
1ŽXˆØˆØ
8 °6ˆ
Ô   ÷
#Ð "úç "Ð "úç "Ð "úç "Ð "ús0Æ=K Ç3KÈ)K$ÉK0Ë KËK!Ë$K-Ë0K9a
[nlp]
lang = "en"
pipeline = ["tok2vec","tagger", "ner"]
[components]
[components.tagger]
factory = "tagger"
[components.tagger.model]
@architectures = "spacy.Tagger.v2"
nO = null
[components.tagger.model.tok2vec]
@architectures = "spacy.Tok2VecListener.v1"
width = ${components.tok2vec.model.encode.width}
[components.ner]
factory = "ner"
[components.ner.model]
@architectures = "spacy.TransitionBasedParser.v2"
[components.ner.model.tok2vec]
@architectures = "spacy.Tok2VecListener.v1"
width = ${components.tok2vec.model.encode.width}
[components.tok2vec]
factory = "tok2vec"
[components.tok2vec.model]
@architectures = "spacy.Tok2Vec.v2"
[components.tok2vec.model.embed]
@architectures = "spacy.MultiHashEmbed.v1"
width = ${components.tok2vec.model.encode.width}
rows = [2000, 1000, 1000, 1000]
attrs = ["NORM", "PREFIX", "SUFFIX", "SHAPE"]
include_static_vectors = false
[components.tok2vec.model.encode]
@architectures = "spacy.MaxoutWindowEncoder.v2"
width = 96
depth = 4
window_size = 1
maxout_pieces = 3
cóª t«jt«}tj|d¬«}ddgddgdœ}t j |jd«|«gŠ |jˆ fd „«|jd
«}|jd «}|jd «}|jd d gk(sJtd
|jj«D««sJtd|jj«D««sJt«5}|j|«t!|«}dgd¢dœd|i|d dgdœ|d dœ|d dœdœdœ}tj|d¬«} ddd« jˆ fd«| jd
«}| jd«}| jd«}d | j"vsJd | j"vsJ|jddgk(sJtd|jj«D««sJtd|jj«D««rJ| j$dd
d}
|
dd k(sJ| j$dddd
|
k(sJ| j$dddd
dd!k(sJ| j$dddd
dd!k(sJy#1swYŒYxYw)"NT)rprb)rr$ÚA)r$rÚB)rhÚentitiesrÏcóSrrKs€r4rz4test_replace_listeners_from_config.<locals>.<lambda>¤r6r1r}Únerc3ó<K|]}t|t«Œy­wr©rr ©Ú.0Únodes r4ú <genexpr>z5test_replace_listeners_from_config.<locals>.<genexpr>©óèø€ÐNÑ=M°TŒz˜$¤×0Ñ=Mùóc3ó<K|]}t|t«Œy­wrs r4z5test_replace_listeners_from_config.<locals>.<genexpr>ªsèø€ÐQÑ=P°TŒz˜$¤×0Ñ=PùríÚen)r1Útagger2Úner3Útagger4)ÚlangÚpipelineÚsourcerÖ)Ú componentrÝ))r]rncóSrrKs€r4rz4test_replace_listeners_from_config.<locals>.<lambda>Åsø€™xr6c3ó<K|]}t|t«Œy­wrs r4z5test_replace_listeners_from_config.<locals>.<genexpr>Ìc3ó<K|]}t|t«Œy­wrs r4z5test_replace_listeners_from_config.<locals>.<genexpr>Ísèø€ÐUÑAT¸”:˜d¤O×4ÑATùrírnr[rG)rr‰Úcfg_string_multirrrr“r”r*Úlistening_componentsÚanyr[ÚwalkrÚstrr) r]Úannotsr1r}Údir_pathÚ
base_modelÚ
new_configÚnew_nlprÞs @r4Ú"test_replace_listeners_from_configrŸø€Ü“(×#Ô$4Ó5€KÜ
×
% k¸
B€CؘCj¨{¸KÐ.HÑ
I€FÜ×! #§,¡,¨uÓ"5°vÓ?€H؇Nl‰l˜9Ó%€GØ
\‰\˜
#€FØ
,‰,
€CØ × '¨H°eÐ+<Ò  ÑN¸S¿Y¹Y¿^¹^Ô=MÓ  ÑQ¸V¿\¹\×=NÑ=NÔ=PÓ  Œ˜8Ø Ü˜“]ˆ
ðÚð
% (Ø!)Ø*9Ð):ñð )Ø!&ñð
)Ø!)ñññ 
ˆ
ô,×-¨jÀDÔ÷3
ð4 ×ÑÓ×ј)€GØ
×
Ñ
˜
(€FØ
×
Ñ