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

[fix] state full name probabilities for limited vs. full formatted OSM training sets

Browse Source
This commit is contained in:
Al
2016-01-22 03:54:20 -05:00
parent d1cf253092
commit bca3dae004
1 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
scripts/geodata/osm/osm_address_training_data.py
View File

@@ -710,7 +710,7 @@ class OSMAddressFormatter(object):
venue_names.append(venue_name) venue_names.append(venue_name)
return venue_names return venue_names
def state_name(self, address_components, country, language, non_local_language=None, state_full_name_prob=0.3): def state_name(self, address_components, country, language, non_local_language=None, state_full_name_prob=0.4):
''' '''
States States
------ ------
@@ -1038,7 +1038,7 @@ class OSMAddressFormatter(object):
if address_country: if address_country:
address_components[AddressFormatter.COUNTRY] = address_country address_components[AddressFormatter.COUNTRY] = address_country
address_state = self.state_name(address_components, country, language, non_local_language=non_local_language, state_full_name_prob=1.0) address_state = self.state_name(address_components, country, language, non_local_language=non_local_language)
if address_state: if address_state:
address_components[AddressFormatter.STATE] = address_state address_components[AddressFormatter.STATE] = address_state
@@ -1100,7 +1100,7 @@ class OSMAddressFormatter(object):
if address_country: if address_country:
address_components[AddressFormatter.COUNTRY] = address_country address_components[AddressFormatter.COUNTRY] = address_country
address_state = self.state_name(address_components, non_local_language) address_state = self.state_name(address_components, country, language, non_local_language=non_local_language, state_full_name_prob=1.0)
if address_state: if address_state:
address_components[AddressFormatter.STATE] = address_state address_components[AddressFormatter.STATE] = address_state