explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2024-12-25 01:16:28 +03:00
Code Issues Packages Projects Releases Wiki Activity

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

Browse Source
This commit is contained in:
Explosion Bot 2017-10-30 16:15:05 +01:00
commit 2d2cc294b4
9 changed files with 285 additions and 257 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
spacy/glossary.py
View File

@ -300,5 +300,15 @@ GLOSSARY = {
'MONEY': 'Monetary values, including unit',
'QUANTITY': 'Measurements, as of weight or distance',
'ORDINAL': '"first", "second", etc.',
'CARDINAL': 'Numerals that do not fall under another type'
'CARDINAL': 'Numerals that do not fall under another type',
# Named Entity Recognition
# Wikipedia
# http://www.sciencedirect.com/science/article/pii/S0004370212000276
# https://pdfs.semanticscholar.org/5744/578cc243d92287f47448870bb426c66cc941.pdf
'PER': 'Named person or family.',
'MISC': ('Miscellaneous entities, e.g. events, nationalities, '
'products or works of art'),
}

244
website/_includes/_mixins-base.jade
View File

@ -1,244 +0,0 @@
//- 💫 MIXINS > BASE
//- Section
id - [string] anchor assigned to section (used for breadcrumb navigation)
mixin section(id)
section.o-section(id="section-" + id data-section=id)
block
//- Aside wrapper
label - [string] aside label
mixin aside-wrapper(label)
aside.c-aside
.c-aside__content(role="complementary")&attributes(attributes)
if label
h4.u-text-label.u-text-label--dark=label
block
//- SVG from map (uses embedded SVG sprite)
name - [string] SVG symbol id
width - [integer] width in px
height - [integer] height in px (default: same as width)
mixin svg(name, width, height)
svg(aria-hidden="true" viewBox="0 0 #{width} #{height || width}" width=width height=(height || width))&attributes(attributes)
use(xlink:href="#svg_#{name}")
//- Icon
name - [string] icon name (will be used as symbol id: #svg_{name})
width - [integer] icon width (default: 20)
height - [integer] icon height (defaults to width)
mixin icon(name, width, height)
- var width = width || 20
- var height = height || width
+svg(name, width, height).o-icon(style="min-width: #{width}px")&attributes(attributes)
//- Pro/Con/Neutral icon
icon - [string] "pro", "con" or "neutral" (default: "neutral")
size - [integer] icon size (optional)
mixin procon(icon, label, show_label, size)
- var colors = { yes: "green", no: "red", neutral: "subtle" }
span.u-nowrap
+icon(icon, size || 20)(class="u-color-#{colors[icon] || 'subtle'}").o-icon--inline&attributes(attributes)
span.u-text-small(class=show_label ? null : "u-hidden")=(label || icon)
//- Headlines Helper Mixin
level - [integer] 1, 2, 3, 4, or 5
mixin headline(level)
if level == 1
h1.u-heading-1&attributes(attributes)
block
else if level == 2
h2.u-heading-2&attributes(attributes)
block
else if level == 3
h3.u-heading-3&attributes(attributes)
block
else if level == 4
h4.u-heading-4&attributes(attributes)
block
else if level == 5
h5.u-heading-5&attributes(attributes)
block
//- Permalink rendering
id - [string] permalink ID used for link anchor
mixin permalink(id)
if id
a.u-permalink(href="##{id}")
block
else
block
//- Quickstart widget
quickstart.js with manual markup, inspired by PyTorch's "Getting started"
groups - [object] option groups, uses global variable QUICKSTART
headline - [string] optional text to be rendered as widget headline
mixin quickstart(groups, headline, description, hide_results)
.c-quickstart.o-block-small#qs
.c-quickstart__content
if headline
+h(2)=headline
if description
p=description
for group in groups
.c-quickstart__group.u-text-small(data-qs-group=group.id)
if group.title
.c-quickstart__legend=group.title
if group.help
| #[+help(group.help)]
.c-quickstart__fields
for option in group.options
input.c-quickstart__input(class="c-quickstart__input--" + (group.input_style ? group.input_style : group.multiple ? "check" : "radio") type=group.multiple ? "checkbox" : "radio" name=group.id id="qs-#{option.id}" value=option.id checked=option.checked)
label.c-quickstart__label.u-text-tiny(for="qs-#{option.id}")!=option.title
if option.meta
| #[span.c-quickstart__label__meta (#{option.meta})]
if option.help
| #[+help(option.help)]
if hide_results
block
else
pre.c-code-block
code.c-code-block__content.c-quickstart__code(data-qs-results="")
block
//- Quickstart code item
data - [object] Rendering conditions (keyed by option group ID, value: option)
style - [string] modifier ID for line style
mixin qs(data, style)
- args = {}
for value, setting in data
- args['data-qs-' + setting] = value
span.c-quickstart__line(class="c-quickstart__line--#{style || 'bash'}")&attributes(args)
block
//- Terminal-style code window
label - [string] title displayed in top bar of terminal window
mixin terminal(label)
.x-terminal
.x-terminal__icons: span
.u-padding-small.u-text-label.u-text-center=label
+code.x-terminal__code
block
//- Chart.js
id - [string] chart ID, will be assigned as #chart_{id}
mixin chart(id, height)
figure.o-block&attributes(attributes)
canvas(id="chart_#{id}" width="800" height=(height || "400") style="max-width: 100%")
//- Gitter chat button and widget
button - [string] text shown on button
label - [string] title of chat window (default: same as button)
mixin gitter(button, label)
aside.js-gitter.c-chat.is-collapsed(data-title=(label || button))
button.js-gitter-button.c-chat__button.u-text-tag
+icon("chat", 16).o-icon--inline
!=button
//- Badge
image - [string] path to badge image
url - [string] badge link
mixin badge(image, url)
+a(url).u-padding-small.u-hide-link&attributes(attributes)
img.o-badge(src=image alt=url height="20")
//- spaCy logo
mixin logo()
+svg("spacy", 675, 215).o-logo&attributes(attributes)
//- Landing
mixin landing-header()
header.c-landing
.c-landing__wrapper
.c-landing__content
block
mixin landing-banner(headline, label)
.c-landing__banner.u-padding.o-block.u-color-light
+grid.c-landing__banner__content.o-no-block
+grid-col("third")
h3.u-heading.u-heading-1
if label
div
span.u-text-label.u-text-label--light=label
!=headline
+grid-col("two-thirds").c-landing__banner__text
block
mixin landing-logos(title, logos)
.o-content.u-text-center&attributes(attributes)
h3.u-heading.u-text-label.u-color-dark=title
each row, i in logos
- var is_last = i == logos.length - 1
+grid("center").o-inline-list.o-no-block(class=is_last ? "o-no-block" : null)
each details, name in row
+a(details[0]).u-padding-medium
+icon(name, details[1], details[2])
if is_last
block
//- Under construction (temporary)
Marks sections that still need to be completed for the v2.0 release.
mixin under-construction()
+infobox("Under construction", "🚧")
| This section is still being written and will be updated for the v2.0
| release. Is there anything that you think should definitely mentioned or
| explained here? Any examples you'd like to see? #[strong Let us know]
| on the #[+a(gh("spacy") + "/issues/1105") v2.0 alpha thread] on GitHub!
//- Alpha infobox (temporary)
Added in the templates to notify user that they're visiting the alpha site.
mixin alpha-info()
+infobox("You are viewing the spaCy v2.0.0 alpha docs", "⚠️")
strong This page is part of the alpha documentation for spaCy v2.0.
| It does not reflect the state of the latest stable release.
| Because v2.0 is still under development, the implementation
| may differ from the intended state described here. See the
| #[+a(gh("spaCy") + "/releases/tag/v2.0.0-alpha") release notes]
| for details on how to install and test the new version. To
| read the official docs for spaCy v1.x,
| #[+a("https://spacy.io/docs") go here].

235
website/_includes/_mixins.jade
View File

@ -1,7 +1,39 @@
//- 💫 INCLUDES > MIXINS
include _functions
include _mixins-base
//- Section
id - [string] anchor assigned to section (used for breadcrumb navigation)
mixin section(id)
section.o-section(id="section-" + id data-section=id)
block
//- Headlines Helper Mixin
level - [integer] 1, 2, 3, 4, or 5
mixin headline(level)
if level == 1
h1.u-heading-1&attributes(attributes)
block
else if level == 2
h2.u-heading-2&attributes(attributes)
block
else if level == 3
h3.u-heading-3&attributes(attributes)
block
else if level == 4
h4.u-heading-4&attributes(attributes)
block
else if level == 5
h5.u-heading-5&attributes(attributes)
block
//- Headlines
@ -18,6 +50,18 @@ mixin h(level, id, source)
span Source #[+icon("code", 14).o-icon--inline]
//- Permalink rendering
id - [string] permalink ID used for link anchor
mixin permalink(id)
if id
a.u-permalink(href="##{id}")
block
else
block
//- External links
url - [string] link href
trusted - [boolean] if not set / false, rel="noopener nofollow" is added
@ -63,6 +107,18 @@ mixin help(tooltip, icon_size)
+icon("help_o", icon_size || 16).o-icon--inline
//- Aside wrapper
label - [string] aside label
mixin aside-wrapper(label)
aside.c-aside
.c-aside__content(role="complementary")&attributes(attributes)
if label
h4.u-text-label.u-text-label--dark=label
block
//- Aside for text
label - [string] aside title (optional)
@ -112,6 +168,37 @@ mixin infobox-logos(...logos)
| #[+icon(logo[0], logo[1], logo[2]).u-color-dark]
//- SVG from map (uses embedded SVG sprite)
name - [string] SVG symbol id
width - [integer] width in px
height - [integer] height in px (default: same as width)
mixin svg(name, width, height)
svg(aria-hidden="true" viewBox="0 0 #{width} #{height || width}" width=width height=(height || width))&attributes(attributes)
use(xlink:href="#svg_#{name}")
//- Icon
name - [string] icon name (will be used as symbol id: #svg_{name})
width - [integer] icon width (default: 20)
height - [integer] icon height (defaults to width)
mixin icon(name, width, height)
- var width = width || 20
- var height = height || width
+svg(name, width, height).o-icon(style="min-width: #{width}px")&attributes(attributes)
//- Pro/Con/Neutral icon
icon - [string] "pro", "con" or "neutral" (default: "neutral")
size - [integer] icon size (optional)
mixin procon(icon, label, show_label, size)
- var colors = { yes: "green", no: "red", neutral: "subtle" }
span.u-nowrap
+icon(icon, size || 20)(class="u-color-#{colors[icon] || 'subtle'}").o-icon--inline&attributes(attributes)
span.u-text-small(class=show_label ? null : "u-hidden")=(label || icon)
//- Link button
url - [string] link href
@ -238,6 +325,14 @@ mixin graphic(original)
+button(original, false, "secondary", "small") View large graphic
//- Chart.js
id - [string] chart ID, will be assigned as #chart_{id}
mixin chart(id, height)
figure.o-block&attributes(attributes)
canvas(id="chart_#{id}" width="800" height=(height || "400") style="max-width: 100%")
//- Labels
mixin label()
@ -353,8 +448,8 @@ mixin grid(...style)
width - [string] "quarter", "third", "half", "two-thirds", "three-quarters"
see $grid in assets/css/_variables.sass
mixin grid-col(width)
.o-grid__col(class="o-grid__col--#{width}")&attributes(attributes)
mixin grid-col(...style)
.o-grid__col(class=prefixArgs(style, "o-grid__col"))&attributes(attributes)
block
@ -445,3 +540,137 @@ mixin annotation-row(annots, style)
else
+cell=cell
block
//- spaCy logo
mixin logo()
+svg("spacy", 675, 215).o-logo&attributes(attributes)
//- Gitter chat button and widget
button - [string] text shown on button
label - [string] title of chat window (default: same as button)
mixin gitter(button, label)
aside.js-gitter.c-chat.is-collapsed(data-title=(label || button))
button.js-gitter-button.c-chat__button.u-text-tag
+icon("chat", 16).o-icon--inline
!=button
//- Badge
image - [string] path to badge image
url - [string] badge link
mixin badge(image, url)
+a(url).u-padding-small.u-hide-link&attributes(attributes)
img.o-badge(src=image alt=url height="20")
//- Quickstart widget
quickstart.js with manual markup, inspired by PyTorch's "Getting started"
groups - [object] option groups, uses global variable QUICKSTART
headline - [string] optional text to be rendered as widget headline
mixin quickstart(groups, headline, description, hide_results)
.c-quickstart.o-block-small#qs
.c-quickstart__content
if headline
+h(2)=headline
if description
p=description
for group in groups
.c-quickstart__group.u-text-small(data-qs-group=group.id)
if group.title
.c-quickstart__legend=group.title
if group.help
| #[+help(group.help)]
.c-quickstart__fields
for option in group.options
input.c-quickstart__input(class="c-quickstart__input--" + (group.input_style ? group.input_style : group.multiple ? "check" : "radio") type=group.multiple ? "checkbox" : "radio" name=group.id id="qs-#{option.id}" value=option.id checked=option.checked)
label.c-quickstart__label.u-text-tiny(for="qs-#{option.id}")!=option.title
if option.meta
| #[span.c-quickstart__label__meta (#{option.meta})]
if option.help
| #[+help(option.help)]
if hide_results
block
else
pre.c-code-block
code.c-code-block__content.c-quickstart__code(data-qs-results="")
block
//- Quickstart code item
data - [object] Rendering conditions (keyed by option group ID, value: option)
style - [string] modifier ID for line style
mixin qs(data, style)
- args = {}
for value, setting in data
- args['data-qs-' + setting] = value
span.c-quickstart__line(class="c-quickstart__line--#{style || 'bash'}")&attributes(args)
block
//- Terminal-style code window
label - [string] title displayed in top bar of terminal window
mixin terminal(label)
.x-terminal
.x-terminal__icons: span
.u-padding-small.u-text-label.u-text-center=label
+code.x-terminal__code
block
//- Landing
mixin landing-header()
header.c-landing
.c-landing__wrapper
.c-landing__content
block
mixin landing-banner(headline, label)
.c-landing__banner.u-padding.o-block.u-color-light
+grid.c-landing__banner__content.o-no-block
+grid-col("third")
h3.u-heading.u-heading-1
if label
div
span.u-text-label.u-text-label--light=label
!=headline
+grid-col("two-thirds").c-landing__banner__text
block
mixin landing-logos(title, logos)
.o-content.u-text-center&attributes(attributes)
h3.u-heading.u-text-label.u-color-dark=title
each row, i in logos
- var is_last = i == logos.length - 1
+grid("center").o-inline-list.o-no-block(class=is_last ? "o-no-block" : null)
each details, name in row
+a(details[0]).u-padding-medium
+icon(name, details[1], details[2])
if is_last
block
//- Under construction (temporary)
Marks sections that still need to be completed for the v2.0 release.
mixin under-construction()
+infobox("Under construction", "🚧")
| This section is still being written and will be updated for the v2.0
| release. Is there anything that you think should definitely mentioned or
| explained here? Any examples you'd like to see? #[strong Let us know]
| on the #[+a(gh("spacy") + "/issues/1105") v2.0 alpha thread] on GitHub!

3
website/_includes/_page-docs.jade
View File

@ -25,9 +25,6 @@ main.o-main.o-main--sidebar.o-main--aside
+button(gh("spacy", source), false, "secondary", "small").u-nowrap
| Source #[+icon("code", 14)]
//-if ALPHA
//- +alpha-info
if IS_MODELS
include _page_models
else

38
website/api/_annotation/_named-entities.jade
View File

@ -1,5 +1,10 @@
//- 💫 DOCS > API > ANNOTATION > NAMED ENTITIES
p
| Models trained on the
| #[+a("https://catalog.ldc.upenn.edu/ldc2013t19") OntoNotes 5] corpus
| support the following entity types:
+table(["Type", "Description"])
+row
+cell #[code PERSON]
@ -45,9 +50,6 @@
+cell #[code LANGUAGE]
+cell Any named language.
p The following values are also annotated in a style similar to names:
+table([ "Type", "Description" ])
+row
+cell #[code DATE]
+cell Absolute or relative dates or periods.
@ -75,3 +77,33 @@ p The following values are also annotated in a style similar to names:
+row
+cell #[code CARDINAL]
+cell Numerals that do not fall under another type.
+h(4, "ner-wikipedia-scheme") Wikipedia scheme
p
| Models trained on Wikipedia corpus
| (#[+a("http://www.sciencedirect.com/science/article/pii/S0004370212000276") Nothman et al., 2013])
| use a less fine-grained NER annotation scheme and recognise the
| following entities:
+table(["Type", "Description"])
+row
+cell #[code PER]
+cell Named person or family.
+row
+cell #[code LOC]
+cell
| Name of politically or geographically defined location (cities,
| provinces, countries, international regions, bodies of water,
| mountains).
+row
+cell #[code ORG]
+cell Named corporate, governmental, or other organizational entity.
+row
+cell #[code MISC]
+cell
| Miscellaneous entities, e.g. events, nationalities, products or
| works of art.

3
website/assets/css/_base/_grid.sass
View File

@ -48,6 +48,9 @@
flex: 0 0 100%
flex-flow: column wrap
&.o-grid__col--no-gutter
margin-top: 0
// Fix overflow issue in old browsers
& > *

2
website/assets/css/_components/_navigation.sass
View File

@ -8,7 +8,7 @@
align-items: center
display: flex
justify-content: space-between
flex-flow: row wrap
flex-flow: row nowrap
padding: 0 2rem 0 1rem
z-index: 30
width: 100%

1
website/assets/css/_components/_tables.sass
View File

@ -51,6 +51,7 @@
@include scroll-shadow-base($color-front)
display: inline-block
overflow-x: auto
overflow-y: hidden
width: auto
-webkit-overflow-scrolling: touch

2
website/usage/_install/_changelog.jade
View File

@ -3,7 +3,7 @@
+h(2, "changelog") Changelog
+button(gh("spacy") + "/releases", false, "secondary", "small").u-float-right.u-nowrap View releases
div(data-tpl="changelog" data-tpl-key="error")
div(data-tpl="changelog" data-tpl-key="error" style="display: none")
+infobox
| Unable to load changelog from GitHub. Please see the
| #[+a(gh("spacy") + "/releases") releases page] instead.