587 lines
16 KiB
YAML
587 lines
16 KiB
YAML
# hr.yaml
|
|
# -------
|
|
# Croatian language specification
|
|
|
|
components:
|
|
level:
|
|
null_probability: 0.9
|
|
alphanumeric_probability: 0.1
|
|
|
|
staircase:
|
|
null_probability: 0.99
|
|
alphanumeric_probability: 0.01
|
|
|
|
entrance:
|
|
null_probability: 0.999
|
|
alphanumeric_probability: 0.001
|
|
|
|
unit:
|
|
null_probability: 0.7
|
|
alphanumeric_probability: 0.3
|
|
|
|
combinations:
|
|
-
|
|
components:
|
|
- house_number
|
|
- staircase
|
|
- level
|
|
- unit
|
|
label: house_number
|
|
separators:
|
|
- separator: "/"
|
|
probability: 0.95
|
|
- separator: "-"
|
|
probability: 0.05
|
|
probability: 0.005
|
|
-
|
|
components:
|
|
- house_number
|
|
- level
|
|
- unit
|
|
label: house_number
|
|
separators:
|
|
- separator: "/"
|
|
probability: 0.95
|
|
- separator: "-"
|
|
probability: 0.05
|
|
probability: 0.005
|
|
-
|
|
components:
|
|
- house_number
|
|
- level
|
|
label: house_number
|
|
separators:
|
|
- separator: "/"
|
|
probability: 0.95
|
|
- separator: "-"
|
|
probability: 0.05
|
|
probability: 0.1
|
|
# For unit types like 2/34
|
|
-
|
|
components:
|
|
- house_number
|
|
- unit
|
|
label: house_number
|
|
separators:
|
|
- separator: "/"
|
|
probability: 0.95
|
|
- separator: "-"
|
|
probability: 0.05
|
|
probability: 0.005
|
|
|
|
|
|
numbers:
|
|
no_number:
|
|
default:
|
|
canonical: bez broja
|
|
abbreviated: bb
|
|
sample: true
|
|
canonical_probability: 0.3
|
|
abbreviated_probability: 0.4
|
|
sample_probability: 0.3
|
|
|
|
default: &broj
|
|
canonical: broj
|
|
abbreviated: br
|
|
sample: true
|
|
canonical_probability: 0.3
|
|
abbreviated_probability: 0.6
|
|
sample_probability: 0.1
|
|
numeric:
|
|
direction: left
|
|
numeric_affix:
|
|
affix: "br."
|
|
whitespace_probability: 0.6
|
|
direction: left
|
|
numeric_probability: 0.4
|
|
numeric_affix_probability: 0.6
|
|
|
|
alphanumeric_phrase_probability: 0.05
|
|
no_number_probability: 0.05
|
|
|
|
|
|
and:
|
|
default: &i
|
|
canonical: i
|
|
sample: true
|
|
canonical_probability: 0.8
|
|
sample_probability: 0.2
|
|
|
|
|
|
cross_streets:
|
|
i: *i
|
|
at: &na
|
|
canonical: na
|
|
sample: true
|
|
canonical_probability: 0.8
|
|
sample_probability: 0.2
|
|
corner: &ugao
|
|
canonical: ugao
|
|
sample: true
|
|
canonical_probability: 0.8
|
|
sample_probability: 0.2
|
|
corner_of: &uglu
|
|
canonical: uglu
|
|
sample: true
|
|
canonical_probability: 0.8
|
|
sample_probability: 0.2
|
|
na_uglu: &na_uglu
|
|
canonical: na uglu
|
|
sample: true
|
|
canonical_probability: 0.8
|
|
sample_probability: 0.2
|
|
intersection:
|
|
default: *i
|
|
probability: 0.65
|
|
alternatives:
|
|
- alternative: *na
|
|
probability: 0.1
|
|
- alternative: *uglu
|
|
probability: 0.1
|
|
- alternative: *na_uglu
|
|
probability: 0.1
|
|
- alternative: *ugao
|
|
probability: 0.05
|
|
|
|
izmedu: &izmedu
|
|
canonical: između
|
|
sample: true
|
|
canonical_probability: 0.8
|
|
sample_probability: 0.2
|
|
parentheses_probability: 0.5
|
|
between:
|
|
default: *izmedu
|
|
|
|
levels:
|
|
kat: &kat
|
|
canonical: kat
|
|
sample: true
|
|
canonical_probability: 0.8
|
|
sample_probability: 0.2
|
|
numeric:
|
|
direction: left
|
|
direction_probability: 0.9
|
|
digits:
|
|
ascii_probability: 0.7
|
|
roman_numeral_probability: 0.3
|
|
add_number_phrase: true
|
|
add_number_phrase_probability: 0.1
|
|
ordinal:
|
|
direction: right
|
|
digits:
|
|
ascii_probability: 0.3
|
|
roman_numeral_probability: 0.7
|
|
add_number_phrase: true
|
|
add_number_phrase_probability: 0.1
|
|
numeric_probability: 0.4
|
|
ordinal_probability: 0.6
|
|
etaza: &etaza
|
|
canonical: etaža
|
|
abbreviated: et
|
|
sample: true
|
|
canonical_probability: 0.4
|
|
abbreviated_probability: 0.4
|
|
sample_probability: 0.2
|
|
numeric:
|
|
direction: left
|
|
direction_probability: 0.9
|
|
digits:
|
|
ascii_probability: 0.7
|
|
roman_numeral_probability: 0.3
|
|
add_number_phrase: true
|
|
add_number_phrase_probability: 0.1
|
|
ordinal:
|
|
direction: right
|
|
digits:
|
|
ascii_probability: 0.3
|
|
roman_numeral_probability: 0.7
|
|
add_number_phrase: true
|
|
add_number_phrase_probability: 0.1
|
|
numeric_probability: 0.4
|
|
ordinal_probability: 0.6
|
|
prizemlje: &prizemlje
|
|
canonical: prizemlje
|
|
sample: true
|
|
canonical_probability: 0.9
|
|
sample_probability: 0.1
|
|
parter: &parter
|
|
canonical: parter
|
|
sample: true
|
|
canonical_probability: 0.9
|
|
sample_probability: 0.1
|
|
mezanino: &polukat
|
|
canonical: polukat
|
|
half_floors: true
|
|
canonical_probability: 0.8
|
|
sample_probability: 0.2
|
|
sample: true
|
|
# e.g. polukat 2
|
|
numeric:
|
|
direction: left
|
|
# e.g. 2. entresuelo
|
|
ordinal:
|
|
direction: right
|
|
numeric_probability: 0.1
|
|
ordinal_probability: 0.2
|
|
standalone_probability: 0.6
|
|
podrum: &podrum
|
|
canonical: podrum
|
|
sample: true
|
|
canonical_probability: 0.7
|
|
sample_probability: 0.3
|
|
# e.g. подрум 1
|
|
numeric:
|
|
direction: left
|
|
direction_probability: 0.8
|
|
# e.g. 1. подрум
|
|
ordinal:
|
|
direction: right
|
|
digits:
|
|
ascii_probability: 0.7
|
|
roman_numeral_probability: 0.3
|
|
standalone_probability: 0.99
|
|
number_abs_value: true
|
|
number_min_abs_value: 1
|
|
numeric_probability: 0.005
|
|
ordinal_probability: 0.005
|
|
|
|
aliases:
|
|
"<-1":
|
|
default: *podrum
|
|
"-1":
|
|
default: *podrum
|
|
# Special token for half-floors
|
|
half_floors:
|
|
default: *polukat
|
|
"0":
|
|
default: *prizemlje
|
|
probability: 0.5
|
|
alternatives:
|
|
- alternative: *parter
|
|
probability: 0.4
|
|
- alternative: *kat
|
|
probability: 0.1
|
|
|
|
numbering_starts_at: 0
|
|
|
|
alphanumeric:
|
|
default: *kat
|
|
probability: 0.95
|
|
alternatives:
|
|
- alternative: *etaza
|
|
probability: 0.05
|
|
numeric_probability: 0.69 # With this probability, pick an integer
|
|
roman_numeral_probability: 0.3 # Pick a Roman numeral for the actual value
|
|
alpha_probability: 0.0098 # With this probability, pick a letter e.g. A
|
|
numeric_plus_alpha_probability: 0.0001 # e.g. 2A
|
|
alpha_plus_numeric_probability: 0.0001 # e.g. A2
|
|
|
|
|
|
categories:
|
|
near:
|
|
default:
|
|
canonical: u blizini
|
|
nearby:
|
|
default:
|
|
canonical: u blizini
|
|
probability: 0.6
|
|
alternatives:
|
|
- alternative:
|
|
canonical: u blizini ovdje
|
|
probability: 0.3
|
|
- alternative:
|
|
canonical: oko ovdje
|
|
probability: 0.1
|
|
|
|
near_me:
|
|
default:
|
|
canonical: u blizini mene
|
|
|
|
# Don't worry about agreement
|
|
in:
|
|
default:
|
|
canonical: u
|
|
|
|
# Probabilities of each phrase
|
|
near_probability: 0.35
|
|
nearby_probability: 0.2
|
|
near_me_probability: 0.1
|
|
in_probability: 0.35
|
|
|
|
directions:
|
|
right: &desno
|
|
canonical: desno
|
|
sample: true
|
|
canonical_probability: 0.8
|
|
sample_probability: 0.2
|
|
numeric:
|
|
direction: right
|
|
left: &lijevo
|
|
canonical: lijevo
|
|
sample: true
|
|
canonical_probability: 0.8
|
|
sample_probability: 0.2
|
|
numeric:
|
|
direction: right
|
|
alternatives:
|
|
- alternative: *desno
|
|
probability: 0.5
|
|
- alternative: *lijevo
|
|
probability: 0.5
|
|
|
|
cardinal_directions:
|
|
east: &istok
|
|
canonical: istok
|
|
abbreviated: i
|
|
canonical_probability: 0.95
|
|
abbreviated_probability: 0.05
|
|
numeric:
|
|
direction: right
|
|
numeric_affix:
|
|
affix: i
|
|
direction: right
|
|
numeric_probability: 0.5
|
|
numeric_affix_probability: 0.5
|
|
|
|
west: &zapad
|
|
canonical: zapad
|
|
abbreviated: z
|
|
canonical_probability: 0.95
|
|
abbreviated_probability: 0.05
|
|
numeric:
|
|
direction: right
|
|
numeric_affix:
|
|
affix: z
|
|
direction: right
|
|
numeric_probability: 0.5
|
|
numeric_affix_probability: 0.5
|
|
|
|
north: &sjever
|
|
canonical: sjever
|
|
abbreviated: s
|
|
canonical_probability: 0.95
|
|
abbreviated_probability: 0.05
|
|
numeric:
|
|
direction: right
|
|
numeric_affix:
|
|
affix: s
|
|
direction: right
|
|
numeric_probability: 0.5
|
|
numeric_affix_probability: 0.5
|
|
|
|
south: &jug
|
|
canonical: jug
|
|
abbreviated: j
|
|
sample: true
|
|
canonical_probability: 0.75
|
|
abbreviated_probability: 0.1
|
|
sample_probability: 0.15
|
|
numeric:
|
|
direction: right
|
|
numeric_affix:
|
|
affix: j
|
|
direction: right
|
|
numeric_probability: 0.5
|
|
numeric_affix_probability: 0.5
|
|
|
|
alternatives:
|
|
- alternative: *sjever
|
|
probability: 0.25
|
|
- alternative: *istok
|
|
probability: 0.23
|
|
- alternative: *jug
|
|
probability: 0.23
|
|
- alternative: *zapad
|
|
probability: 0.23
|
|
|
|
entrances:
|
|
ulaz: &ulaz
|
|
canonical: ulaz
|
|
sample: true
|
|
canonical_probability: 0.8
|
|
sample_probability: 0.2
|
|
numeric:
|
|
direction: left
|
|
|
|
# Ulaz 1, Ulaz A, etc.
|
|
alphanumeric: &entrance_alphanumeric
|
|
default: *ulaz
|
|
numeric_probability: 0.1 # e.g. Ulaz 1
|
|
alpha_probability: 0.85 # e.g. Ulaz A
|
|
numeric_plus_alpha_probability: 0.025 # e.g. 1A
|
|
alpha_plus_numeric_probability: 0.025 # e.g. A1
|
|
|
|
alpha_plus_numeric:
|
|
whitespace_probability: 0.1
|
|
|
|
numeric_plus_alpha:
|
|
whitespace_probability: 0.1
|
|
|
|
|
|
staircases:
|
|
stubiste: &stubiste
|
|
canonical: stubište
|
|
sample: true
|
|
canonical_probability: 0.8
|
|
sample_probability: 0.2
|
|
numeric:
|
|
direction: left
|
|
|
|
|
|
alphanumeric: &staircase_alphanumeric
|
|
default: *stubiste
|
|
numeric_probability: 0.75
|
|
alpha_probability: 0.2
|
|
numeric_plus_alpha_probability: 0.025
|
|
alpha_plus_numeric_probability: 0.025
|
|
|
|
alpha_plus_numeric:
|
|
whitespace_probability: 0.1
|
|
|
|
numeric_plus_alpha:
|
|
whitespace_probability: 0.1
|
|
|
|
directional:
|
|
direction: right
|
|
direction_probability: 0.85
|
|
modifier:
|
|
alternatives:
|
|
- alternative: *desno
|
|
probability: 0.2
|
|
- alternative: *lijevo
|
|
probability: 0.2
|
|
- alternative: *sjever
|
|
probability: 0.15
|
|
- alternative: *jug
|
|
probability: 0.15
|
|
- alternative: *istok
|
|
probability: 0.15
|
|
- alternative: *zapad
|
|
probability: 0.15
|
|
|
|
po_boxes:
|
|
postanski_pretinac: &postanski_pretinac
|
|
canonical: poštanski pretinac
|
|
abbreviated: p.p
|
|
sample: true
|
|
canonical_probability: 0.2
|
|
abbreviated_probability: 0.4
|
|
sample_probability: 0.4
|
|
numeric:
|
|
direction: left
|
|
add_number_phrase: true
|
|
add_number_phrase_probability: 0.2
|
|
|
|
alphanumeric:
|
|
default: *postanski_pretinac
|
|
numeric_probability: 0.9 # pp 123
|
|
alpha_probability: 0.05 # p.p A
|
|
numeric_plus_alpha_probability: 0.04 # pp 123G
|
|
alpha_plus_numeric_probability: 0.01 # pp A123
|
|
alpha_plus_numeric:
|
|
whitespace_probability: 0.1
|
|
numeric_plus_alpha:
|
|
whitespace_probability: 0.1
|
|
|
|
digits:
|
|
- length: 1
|
|
probability: 0.05
|
|
- length: 2
|
|
probability: 0.1
|
|
- length: 3
|
|
probability: 0.2
|
|
- length: 4
|
|
probability: 0.5
|
|
- length: 5
|
|
probability: 0.1
|
|
- length: 6
|
|
probability: 0.05
|
|
|
|
units:
|
|
stan: &stan
|
|
canonical: stan
|
|
sample: true
|
|
canonical_probability: 0.8
|
|
sample_probability: 0.2
|
|
numeric:
|
|
direction: left
|
|
add_number_phrase: true
|
|
add_number_phrase_probability: 0.1
|
|
apartman: &apartman
|
|
canonical: apartman
|
|
abbreviated: ap
|
|
sample: true
|
|
canonical_probability: 0.4
|
|
abbreviated_probability: 0.2
|
|
sample_probability: 0.4
|
|
numeric:
|
|
direction: left
|
|
add_number_phrase: true
|
|
add_number_phrase_probability: 0.1
|
|
|
|
soba: &soba
|
|
canonical: soba
|
|
sample: true
|
|
canonical_probability: 0.8
|
|
sample_probability: 0.2
|
|
numeric:
|
|
direction: left
|
|
add_number_phrase: true
|
|
add_number_phrase_probability: 0.1
|
|
ured: &ured
|
|
canonical: ured
|
|
sample: true
|
|
canonical_probability: 0.6
|
|
sample_probability: 0.4
|
|
numeric:
|
|
direction: left
|
|
add_number_phrase: true
|
|
add_number_phrase_probability: 0.1
|
|
|
|
alphanumeric: &unit_alphanumeric
|
|
default: *stan
|
|
probability: 0.6
|
|
alternatives:
|
|
- alternative: *apartman
|
|
probability: 0.3
|
|
- alternative: *soba
|
|
probability: 0.1
|
|
numeric_probability: 0.9 # e.g. stan. 1
|
|
numeric_plus_alpha_probability: 0.03 # e.g. 1A
|
|
alpha_plus_numeric_probability: 0.03 # e.g. A1
|
|
alpha_probability: 0.04 # e.g. stan A
|
|
|
|
alpha_plus_numeric:
|
|
whitespace_probability: 0.1
|
|
numeric_plus_alpha:
|
|
whitespace_probability: 0.1
|
|
|
|
# If there are 10 floors, create unit numbers like #301 or #1032
|
|
use_floor_probability: 0.05
|
|
|
|
zones:
|
|
commercial: &commercial_unit_types
|
|
default: *soba
|
|
probability: 0.6
|
|
alternatives:
|
|
- alternative: *ured
|
|
probability: 0.4
|
|
numeric_probability: 0.95 # e.g. soba 1
|
|
numeric_plus_alpha_probability: 0.01 # e.g. soba 1A
|
|
alpha_plus_numeric_probability: 0.01 # e.g. soba A1
|
|
alpha_probability: 0.03 # e.g. soba A
|
|
alpha_plus_numeric:
|
|
whitespace_probability: 0.1
|
|
numeric_plus_alpha:
|
|
whitespace_probability: 0.1
|
|
university:
|
|
default: *soba
|
|
numeric_probability: 0.95 # e.g. soba 1
|
|
numeric_plus_alpha_probability: 0.01 # e.g. soba 1A
|
|
alpha_plus_numeric_probability: 0.01 # e.g. soba A1
|
|
alpha_probability: 0.03 # e.g. soba A
|
|
alpha_plus_numeric:
|
|
whitespace_probability: 0.1
|
|
numeric_plus_alpha:
|
|
whitespace_probability: 0.1
|