tommy/libpostal
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[addresses] Adding digit spellout and the list form of field combinations to existing configs

Browse Source
This commit is contained in:
Al
2016-07-04 13:46:19 -04:00
parent 64f167f045
commit af11db1488
20 changed files with 353 additions and 124 deletions
Download Patch File Download Diff File Expand all files Collapse all files

59
resources/addresses/de.yaml
View File

@@ -26,7 +26,7 @@ components:
combinations:
# e.g. 2/34, more common way to specify a unit number in German
# if unit exists in the first place
house_number_unit:
-
components:
- house_number
- unit
@@ -117,6 +117,9 @@ levels:
direction: right
ordinal:
direction: right
digits:
ascii_probability: 0.8
spellout_probability: 0.2
numeric_probability: 0.3
numeric_affix_probability: 0.5
ordinal_probability: 0.2
@@ -129,6 +132,9 @@ levels:
direction: right
ordinal:
direction: right
digits:
ascii_probability: 0.8
spellout_probability: 0.2
numeric_probability: 0.4
ordinal_probability: 0.6
stock: &stock
@@ -140,6 +146,9 @@ levels:
direction: right
ordinal:
direction: right
digits:
ascii_probability: 0.8
spellout_probability: 0.2
numeric_probability: 0.1
ordinal_probability: 0.9
erdgeschoss: &erdgeschoss
@@ -613,29 +622,8 @@ countries:
# Combined apartment numbers are very common
combinations:
# e.g. Neubaugasse 55/5
house_number_unit:
probability: 0.7
separators:
- separator: /
probability: 0.98
- separator: "-"
probability: 0.02
# e.g. Neubaugasse 55/1/5
house_number_staircase_unit:
components:
- house_number
- staircase
- unit
label: house_number
separators:
- separator: /
probability: 0.98
- separator: "-"
probability: 0.02
probability: 0.8
# e.g. Neubaugasse 55/A/1/5
house_number_entrance_staircase_unit:
-
components:
- house_number
- entrance
@@ -648,6 +636,31 @@ countries:
- separator: "-"
probability: 0.02
probability: 0.9
# e.g. Neubaugasse 55/1/5
-
components:
- house_number
- staircase
- unit
label: house_number
separators:
- separator: /
probability: 0.98
- separator: "-"
probability: 0.02
probability: 0.8
# e.g. Neubaugasse 55/5
-
components:
- house_number
- unit
label: house_number
probability: 0.7
separators:
- separator: /
probability: 0.98
- separator: "-"
probability: 0.02
units:
top: &top