Rooba
/
spaCy
mirror of https://github.com/explosion/spaCy.git
1
1
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'develop' of https://github.com/explosion/spaCy into develop

This commit is contained in:
Matthew Honnibal 2017-11-06 19:20:41 +01:00
parent 8fea512ac8 64d0f97c67
commit ffb9101f3f
18 changed files with 99 additions and 96 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
spacy/deprecated.py
View File

@ -1,4 +0,0 @@
# coding: utf8
from __future__ import unicode_literals
PRON_LEMMA = "-PRON-"

3
spacy/lang/bn/morph_rules.py
View File

@ -1,8 +1,7 @@
# coding: utf8
from __future__ import unicode_literals
from ...deprecated import PRON_LEMMA
from ...symbols import LEMMA
from ...symbols import LEMMA, PRON_LEMMA
MORPH_RULES = {

4
spacy/lang/da/morph_rules.py
View File

@ -1,8 +1,8 @@
# coding: utf8
from __future__ import unicode_literals
from ...symbols import LEMMA
from ...deprecated import PRON_LEMMA
from ...symbols import LEMMA, PRON_LEMMA
MORPH_RULES = {
"PRON": {

3
spacy/lang/de/tokenizer_exceptions.py
View File

@ -1,8 +1,7 @@
# coding: utf8
from __future__ import unicode_literals
from ...symbols import ORTH, LEMMA, TAG, NORM
from ...deprecated import PRON_LEMMA
from ...symbols import ORTH, LEMMA, TAG, NORM, PRON_LEMMA
_exc = {

3
spacy/lang/en/morph_rules.py
View File

@ -1,8 +1,7 @@
# coding: utf8
from __future__ import unicode_literals
from ...symbols import LEMMA
from ...deprecated import PRON_LEMMA
from ...symbols import LEMMA, PRON_LEMMA
MORPH_RULES = {

3
spacy/lang/en/tokenizer_exceptions.py
View File

@ -1,8 +1,7 @@
# coding: utf8
from __future__ import unicode_literals
from ...symbols import ORTH, LEMMA, TAG, NORM
from ...deprecated import PRON_LEMMA
from ...symbols import ORTH, LEMMA, TAG, NORM, PRON_LEMMA
_exc = {}

3
spacy/lang/es/tokenizer_exceptions.py
View File

@ -1,8 +1,7 @@
# coding: utf8
from __future__ import unicode_literals
from ...symbols import ORTH, LEMMA, TAG, NORM, ADP, DET
from ...deprecated import PRON_LEMMA
from ...symbols import ORTH, LEMMA, TAG, NORM, ADP, DET, PRON_LEMMA
_exc = {

3
spacy/lang/fr/tokenizer_exceptions.py
View File

@ -7,8 +7,7 @@ from ._tokenizer_exceptions_list import FR_BASE_EXCEPTIONS
from .punctuation import ELISION, HYPHENS
from ..tokenizer_exceptions import URL_PATTERN
from ..char_classes import ALPHA_LOWER
from ...symbols import ORTH, LEMMA, TAG, NORM
from ...deprecated import PRON_LEMMA
from ...symbols import ORTH, LEMMA, TAG, NORM, PRON_LEMMA
def upper_first_letter(text):

3
spacy/lang/nb/morph_rules.py
View File

@ -1,8 +1,7 @@
# encoding: utf8
from __future__ import unicode_literals
from ...symbols import LEMMA
from ...deprecated import PRON_LEMMA
from ...symbols import LEMMA, PRON_LEMMA
# Used the table of pronouns at https://no.wiktionary.org/wiki/Tillegg:Pronomen_i_norsk

3
spacy/lang/pt/tokenizer_exceptions.py
View File

@ -1,8 +1,7 @@
# coding: utf8
from __future__ import unicode_literals
from ...symbols import ORTH, LEMMA, NORM
from ...deprecated import PRON_LEMMA
from ...symbols import ORTH, LEMMA, NORM, PRON_LEMMA
_exc = {

3
spacy/lang/sv/morph_rules.py
View File

@ -1,8 +1,7 @@
# coding: utf8
from __future__ import unicode_literals
from ...symbols import LEMMA
from ...deprecated import PRON_LEMMA
from ...symbols import LEMMA, PRON_LEMMA
# Used the table of pronouns at https://sv.wiktionary.org/wiki/deras

3
spacy/lang/sv/tokenizer_exceptions.py
View File

@ -1,8 +1,7 @@
# coding: utf8
from __future__ import unicode_literals
from ...symbols import ORTH, LEMMA, TAG, NORM
from ...deprecated import PRON_LEMMA
from ...symbols import ORTH, LEMMA, TAG, NORM, PRON_LEMMA
_exc = {}

4
website/_includes/_mixins.jade
View File

@ -452,8 +452,8 @@ mixin head-cell()
//- Table cell (only used within +row in +table)
mixin cell()
td.c-table__cell.u-text&attributes(attributes)
mixin cell(align)
td.c-table__cell.u-text(class=align ? "u-text-" + align : null)&attributes(attributes)
block

2
website/_includes/_page_models.jade
View File

@ -71,7 +71,7 @@ for id in CURRENT_MODELS
+label=label
if MODEL_META[field]
| #[+help(MODEL_META[field]).u-color-subtle]
+cell.u-text-right(data-tpl=id data-tpl-key=field)
+cell("right")(data-tpl=id data-tpl-key=field)
| n/a
p.u-text-small.u-color-dark(data-tpl=id data-tpl-key="notes")

2
website/models/_data.json
View File

@ -43,7 +43,7 @@
"en": ["en_core_web_sm", "en_core_web_lg", "en_vectors_web_lg"],
"de": ["de_core_news_sm"],
"es": ["es_core_news_sm", "es_core_news_md"],
"pt": [],
"pt": ["pt_core_news_sm"],
"fr": ["fr_core_news_sm"],
"it": ["it_core_news_sm"],
"nl": ["nl_core_news_sm"],

24
website/usage/_facts-figures/_benchmarks-choi-2015.jade
View File

@ -5,41 +5,41 @@
+cell #[strong spaCy v2.x]
+cell 2017
+cell Python / Cython
+cell.u-text-right #[strong 92.6]
+cell.u-text-right #[em n/a]
+cell("right") #[strong 92.6]
+cell("right") #[em n/a]
| #[+help("This table shows speed as benchmarked by Choi et al. We therefore can't provide comparable figures, as we'd be running the benchmark on different hardware.").u-color-dark]
+row
+cell #[strong spaCy v1.x]
+cell 2015
+cell Python / Cython
+cell.u-text-right 91.8
+cell.u-text-right 13,963
+cell("right") 91.8
+cell("right") 13,963
+row
+cell ClearNLP
+cell 2015
+cell Java
+cell.u-text-right 91.7
+cell.u-text-right 10,271
+cell("right") 91.7
+cell("right") 10,271
+row
+cell CoreNLP
+cell 2015
+cell Java
+cell.u-text-right 89.6
+cell.u-text-right 8,602
+cell("right") 89.6
+cell("right") 8,602
+row
+cell MATE
+cell 2015
+cell Java
+cell.u-text-right 92.5
+cell.u-text-right 550
+cell("right") 92.5
+cell("right") 550
+row
+cell Turbo
+cell 2015
+cell C++
+cell.u-text-right 92.4
+cell.u-text-right 349
+cell("right") 92.4
+cell("right") 349

81
website/usage/_facts-figures/_benchmarks-models.jade
View File

@ -4,7 +4,9 @@ p
| In this section, we provide benchmark accuracies for the pre-trained
| model pipelines we distribute with spaCy. Evaluations are conducted
| end-to-end from raw text, with no "gold standard" pre-processing, over
| text from a mix of genres where possible.
| text from a mix of genres where possible. For are more detailed
| comparison of the available models, see the new
| #[+a("/models/comparison") model comparison tool].
+aside("Methodology")
| The evaluation was conducted on raw text with no gold standard
@ -16,55 +18,70 @@ p
+table(["Model", "spaCy", "Type", "UAS", "NER F", "POS", "WPS", "Size"])
+row
+cell #[+a("/models/en#en_core_web_sm") #[code en_core_web_sm]] 2.0.0a5
+cell #[+a("/models/en#en_core_web_sm") #[code en_core_web_sm]] 2.0.0a8
each data in ["2.x", "neural"]
+cell.u-text-right=data
+cell.u-text-right 91.4
+cell.u-text-right 85.5
+cell.u-text-right 97.0
+cell.u-text-right 8.2k
+cell.u-text-right #[strong 36 MB]
+cell("right")=data
+cell("right") 91.7
+cell("right") 85.3
+cell("right") 97.0
+cell("right") 10.1k
+cell("right") #[strong 35 MB]
+row
+cell #[+a("/models/en#en_core_web_lg") #[code en_core_web_lg]] 2.0.0a0
+cell #[+a("/models/en#en_core_web_lg") #[code en_core_web_lg]] 2.0.0a3
each data in ["2.x", "neural"]
+cell.u-text-right=data
+cell.u-text-right #[strong 91.9]
+cell.u-text-right #[strong 86.4]
+cell.u-text-right #[strong 97.2]
+cell.u-text-right #[em n/a]
+cell.u-text-right 667 MB
+cell("right")=data
+cell("right") #[strong 91.9]
+cell("right") #[strong 85.9]
+cell("right") #[strong 97.2]
+cell("right") 5.0k
+cell("right") 812 MB
+row("divider")
+cell #[code en_core_web_sm] 1.2.0
each data in ["1.x", "linear", 86.6, 78.5, 96.6]
+cell.u-text-right=data
+cell.u-text-right #[strong 25.7k]
+cell.u-text-right 50 MB
+cell("right")=data
+cell("right") #[strong 25.7k]
+cell("right") 50 MB
+row
+cell #[code en_core_web_md] 1.2.1
each data in ["1.x", "linear", 90.6, 81.4, 96.7, "18.8k", "1 GB"]
+cell.u-text-right=data
+cell("right")=data
+h(4, "benchmarks-models-spanish") Spanish
+aside("Evaluation note")
| The NER accuracy refers to the "silver standard" annotations in the
| WikiNER corpus. Accuracy on these annotations tends to be higher than
| correct human annotations.
+table(["Model", "spaCy", "Type", "UAS", "NER F", "POS", "WPS", "Size"])
+row
+cell #[+a("/models/es#es_core_web_sm") #[code es_core_web_sm]] 2.0.0a0
+cell.u-text-right 2.x
+cell.u-text-right neural
+cell.u-text-right #[strong 90.1]
+cell.u-text-right 89.0
+cell.u-text-right #[strong 96.7]
+cell.u-text-right #[em n/a]
+cell.u-text-right #[strong 36 MB]
+cell #[+a("/models/es#es_core_news_sm") #[code es_core_news_sm]] 2.0.0a0
+cell("right") 2.x
+cell("right") neural
+cell("right") 89.8
+cell("right") 88.7
+cell("right") #[strong 96.9]
+cell("right") #[em n/a]
+cell("right") #[strong 35 MB]
+row
+cell #[+a("/models/es#es_core_news_md") #[code es_core_news_md]] 2.0.0a0
+cell("right") 2.x
+cell("right") neural
+cell("right") #[strong 90.2]
+cell("right") 89.0
+cell("right") 97.8
+cell("right") #[em n/a]
+cell("right") 93 MB
+row("divider")
+cell #[code es_core_web_md] 1.1.0
each data in ["1.x", "linear", 87.5]
+cell.u-text-right=data
+cell #[strong 94.2]
+cell #[strong 96.7]
+cell.u-text-right #[em n/a]
+cell.u-text-right 377 MB
+cell("right")=data
+cell("right") #[strong 94.2]
+cell("right") 96.7
+cell("right") #[em n/a]
+cell("right") 377 MB

44
website/usage/_facts-figures/_benchmarks.jade
View File

@ -50,55 +50,55 @@ p
+cell spaCy v2.0.0
+cell 2017
+cell neural
+cell.u-text-right 94.48
+cell("right") 94.48
+row
+cell spaCy v1.1.0
+cell 2016
+cell linear
+cell.u-text-right 92.80
+cell("right") 92.80
+row("divider")
+cell
+a("https://arxiv.org/pdf/1611.01734.pdf") Dozat and Manning
+cell 2017
+cell neural
+cell.u-text-right #[strong 95.75]
+cell("right") #[strong 95.75]
+row
+cell
+a("http://arxiv.org/abs/1603.06042") Andor et al.
+cell 2016
+cell neural
+cell.u-text-right 94.44
+cell("right") 94.44
+row
+cell
+a("https://github.com/tensorflow/models/tree/master/research/syntaxnet") SyntaxNet Parsey McParseface
+cell 2016
+cell neural
+cell.u-text-right 94.15
+cell("right") 94.15
+row
+cell
+a("http://static.googleusercontent.com/media/research.google.com/en//pubs/archive/43800.pdf") Weiss et al.
+cell 2015
+cell neural
+cell.u-text-right 93.91
+cell("right") 93.91
+row
+cell
+a("http://research.google.com/pubs/archive/38148.pdf") Zhang and McDonald
+cell 2014
+cell linear
+cell.u-text-right 93.32
+cell("right") 93.32
+row
+cell
+a("http://www.cs.cmu.edu/~ark/TurboParser/") Martins et al.
+cell 2013
+cell linear
+cell.u-text-right 93.10
+cell("right") 93.10
+h(4, "ner-accuracy-ontonotes5") NER accuracy (OntoNotes 5, no pre-process)
@ -110,38 +110,38 @@ p
+table(["System", "Year", "Type", "Accuracy"])
+row
+cell spaCy #[+a("/models/en#en_core_web_lg") #[code en_core_web_lg]] v2.0.0
+cell spaCy #[+a("/models/en#en_core_web_lg") #[code en_core_web_lg]] v2.0.0a3
+cell 2017
+cell neural
+cell.u-text-right 86.45
+cell("right") 85.85
+row("divider")
+cell
+a("https://arxiv.org/pdf/1702.02098.pdf") Strubell et al.
+cell 2017
+cell neural
+cell.u-text-right #[strong 86.81]
+cell("right") #[strong 86.81]
+row
+cell
+a("https://www.semanticscholar.org/paper/Named-Entity-Recognition-with-Bidirectional-LSTM-C-Chiu-Nichols/10a4db59e81d26b2e0e896d3186ef81b4458b93f") Chiu and Nichols
+cell 2016
+cell neural
+cell.u-text-right 86.19
+cell("right") 86.19
+row
+cell
+a("https://www.semanticscholar.org/paper/A-Joint-Model-for-Entity-Analysis-Coreference-Typi-Durrett-Klein/28eb033eee5f51c5e5389cbb6b777779203a6778") Durrett and Klein
+cell 2014
+cell neural
+cell.u-text-right 84.04
+cell("right") 84.04
+row
+cell
+a("http://www.aclweb.org/anthology/W09-1119") Ratinov and Roth
+cell 2009
+cell linear
+cell.u-text-right 83.45
+cell("right") 83.45
+h(3, "spacy-models") Model comparison
@ -183,24 +183,24 @@ p
+row
+cell #[strong spaCy]
each data in [ "0.2ms", "1ms", "19ms"]
+cell.u-text-right #[strong=data]
+cell("right") #[strong=data]
each data in ["1x", "1x", "1x"]
+cell.u-text-right=data
+cell("right")=data
+row
+cell CoreNLP
each data in ["2ms", "10ms", "49ms", "10x", "10x", "2.6x"]
+cell.u-text-right=data
+cell("right")=data
+row
+cell ZPar
each data in ["1ms", "8ms", "850ms", "5x", "8x", "44.7x"]
+cell.u-text-right=data
+cell("right")=data
+row
+cell NLTK
each data in ["4ms", "443ms"]
+cell.u-text-right=data
+cell.u-text-right #[em n/a]
+cell("right")=data
+cell("right") #[em n/a]
each data in ["20x", "443x"]
+cell.u-text-right=data
+cell.u-text-right #[em n/a]
+cell("right")=data
+cell("right") #[em n/a]