[formatting] conditional probabilities of template swaps in formatter config

This commit is contained in:
Al
2016-05-17 21:20:36 -04:00
parent a948e97fe6
commit aeffac1047

View File

@@ -1,22 +1,9 @@
global: global:
# Add these components to templates that don't have them # Add these components to templates that don't have them
admin_components: admin_components:
subdivision:
after:
- road
before:
- suburb
- city_district
- city
- island
- state_district
- state
- postcode
- country
suburb: suburb:
after: after:
- road - road
- subdivision
before: before:
- city_district - city_district
- city - city
@@ -36,7 +23,7 @@ global:
- state - state
- postcode - postcode
- country - country
# This is added to all the templates but only makes it in # This is added to all the templates but only makes it into island nations, Hawaii, etc.
island: island:
after: after:
- road - road
@@ -134,10 +121,87 @@ global:
first: true first: true
probability: 0.1 probability: 0.1
care_of:
care_of_after_attention:
after: attention
probability: 0.9
care_of_after_house:
after: house
probability: 0.1
subdivision:
subdivision_before_suburb:
before: suburb
probability: 1.0
# Overrides for languages (better for e.g. covering all French-speaking countries) # Overrides for languages (better for e.g. covering all French-speaking countries)
languages: languages:
en: en:
insertions: insertions:
building:
building_after_house:
after: house
probability: 0.6
building_after_road:
after: road
probability: 0.3
building_before_suburb:
before: suburb
probability: 0.1
conditional:
- component: subdivision
probabilities:
building_after_house:
after: house
probability: 0.6
building_after_road:
after: road
probability: 0.2
building_before_subdivision:
before: subdivision
probability: 0.2
entrance:
# e.g. 123 East 45th St, 6th Floor, NYC
entrance_after_road:
after: road
probability: 0.75
entrance_before_house:
before: house
probability: 0.1
# e.g. Da Vinci House, 1st Floor, 44 Saffron Hill, London
entrance_before_house_number:
before: house_number
probability: 0.15
conditional:
- component: building
probabilities:
# default: 0.8
entrance_after_building:
after: building
probability: 0.2
staircase:
# e.g. 123 East 45th St, Staircase C, NYC
staircase_after_road:
after: road
probability: 0.5
# e.g. Floor 1, Da Vinci House, 44 Saffron Hill, London
staircase_before_house:
before: house
probability: 0.1
# e.g. Da Vinci House, 1st Floor, 44 Saffron Hill, London
staircase_before_house_number:
before: house_number
probability: 0.4
conditional:
- component: entrance
probabilities:
# default: 0.1
staircase_after_entrance:
after: entrance
probability: 0.9
level: level:
# e.g. 123 East 45th St, 6th Floor, NYC # e.g. 123 East 45th St, 6th Floor, NYC
level_after_road: level_after_road:
@@ -148,10 +212,24 @@ languages:
before: house before: house
probability: 0.25 probability: 0.25
# e.g. Da Vinci House, 1st Floor, 44 Saffron Hill, London # e.g. Da Vinci House, 1st Floor, 44 Saffron Hill, London
level_before_road: level_before_house_number:
before: road before: house_number
probability: 0.25 probability: 0.25
conditional:
- component: staircase
probabilities:
# default: 0.4
level_after_staircase:
after: staircase
probability: 0.6
- component: entrance
probabilities:
# default: 0.4
level_after_entrance:
after: entrance
probability: 0.6
unit: unit:
# e.g. Flat 18, Da Vinci House, 44 Saffron Hill, London # e.g. Flat 18, Da Vinci House, 44 Saffron Hill, London
unit_before_house: unit_before_house:
@@ -166,20 +244,87 @@ languages:
# e.g. Da Vinci House, 44 Saffron Hill, Flat 18, London (not as common in UK) # e.g. Da Vinci House, 44 Saffron Hill, Flat 18, London (not as common in UK)
unit_after_road: unit_after_road:
after: road after: road
probability: 0.1 probability: 0.2
# e.g. Floor 5, Apt 6 conditional:
unit_after_level: - component: level
after: level probabilities:
probability: 0.09 unit_before_house:
before: house
# e.g. Apt. 6, 5/F (less common) probability: 0.1
unit_before_level: # e.g. Da Vinci House, Flat 18, 44 Saffron Hill, London
before: level unit_before_house_number:
probability: 0.01 before: house_number
probability: 0.1
# e.g. Floor 5, Apt 6
unit_after_level:
after: level
probability: 0.79
# e.g. Apt. 6, 5/F (less common)
unit_before_level:
before: level
probability: 0.01
es: es:
insertions: insertions:
building:
building_after_house_number:
after: house_number
probability: 0.8
building_after_house:
after: house
probability: 0.05
building_before_suburb:
before: suburb
probability: 0.15
conditional:
- component: subdivision
probabilities:
building_after_house_number:
after: house_number
probability: 0.8
building_before_subdivision:
before: subdivision
probability: 0.2
entrance:
# e.g. 123 East 45th St, 6th Floor, NYC
entrance_after_house_number:
after: house_number
probability: 0.8
entrance_before_house:
before: house
probability: 0.2
conditional:
- component: building
probabilities:
entrance_after_building:
after: building
probability: 0.9
staircase:
# e.g. 123 East 45th St, Staircase C, NYC
staircase_after_house_number:
after: house_number
probability: 0.8
# e.g. Floor 1, Da Vinci House, 44 Saffron Hill, London
staircase_before_house:
before: house
probability: 0.2
conditional:
- component: entrance
probabilities:
# default: 0.1
staircase_after_entrance:
after: entrance
probability: 0.9
- component: building
probabilities:
# default: 0.1
staircase_after_building:
after: building
probability: 0.9
level: level:
# e.g. Calle Ruiz de Alarcón 23 piso 3 # e.g. Calle Ruiz de Alarcón 23 piso 3
level_after_house_number: level_after_house_number:
@@ -194,21 +339,67 @@ languages:
before: road before: road
probability: 0.1 probability: 0.1
conditional:
- component: staircase
probabilities:
# default: 0.1
level_after_staircase:
after: staircase
probability: 0.9
- component: entrance
probabilities:
# default: 0.1
level_after_entrance:
after: entrance
probability: 0.9
- component: building
probabilities:
# default: 0.1
level_after_building:
after: building
probability: 0.9
unit: unit:
unit_before_house: unit_before_house:
before: house before: house
probability: 0.05 probability: 0.2
unit_before_house_number: unit_after_house_number:
before: house_number after: house_number
probability: 0.05 probability: 0.7
# e.g. Piso 3 Dpto 12 (most common) unit_before_road:
unit_after_level: before: road
after: level
probability: 0.8
# e.g. Apto 6, 2o piso (less common)
unit_before_level:
before: level
probability: 0.1 probability: 0.1
conditional:
- component: level
probabilities:
# default: 0.1
# e.g. Piso 3 Dpto 12 (most common)
unit_after_level:
after: level
probability: 0.85
# e.g. Apto 6, 2o piso (less common)
unit_before_level:
before: level
probability: 0.05
- component: staircase
probabilities:
# default: 0.1
unit_after_staircase:
after: staircase
probability: 0.9
- component: entrance
probabilities:
# default: 0.1
unit_after_entrance:
after: entrance
probability: 0.9
- component: building
probabilities:
# default: 0.1
unit_after_building:
after: building
probability: 0.9
fr: fr:
# libpostal issue #27 # libpostal issue #27
@@ -247,15 +438,15 @@ countries:
# e.g. 123 East 45th St, 6th Floor, NYC # e.g. 123 East 45th St, 6th Floor, NYC
level_after_road: level_after_road:
after: road after: road
probability: 0.75 probability: 0.875
# e.g. Floor 1, Da Vinci House, 44 Saffron Hill, London # e.g. Floor 1, Da Vinci House, 44 Saffron Hill, London
level_before_house: level_before_house:
before: house before: house
probability: 0.125 probability: 0.005
# e.g. Da Vinci House, 1st Floor, 44 Saffron Hill, London # e.g. Da Vinci House, 1st Floor, 44 Saffron Hill, London
level_before_road: level_before_house_number:
before: road before: house_number
probability: 0.125 probability: 0.12
unit: unit:
# e.g. Flat 18, Da Vinci House, 44 Saffron Hill, London # e.g. Flat 18, Da Vinci House, 44 Saffron Hill, London