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

[osm] Adding abbreviated street names in addition to the originals

Browse Source
This commit is contained in:
Al
2016-01-23 23:23:58 -05:00
parent 67130383ce
commit 7646adfc0f
1 changed files with 3 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
scripts/geodata/osm/osm_address_training_data.py
View File

@@ -500,8 +500,10 @@ def build_ways_training_data(language_rtree, infile, out_dir):
for lang, val in name_language.iteritems(): for lang, val in name_language.iteritems():
for s in val: for s in val:
if lang in languages: if lang in languages:
s = osm_abbreviate(street_types_gazetteer, s, lang)
writer.writerow((lang, country, tsv_string(s))) writer.writerow((lang, country, tsv_string(s)))
abbrev = osm_abbreviate(street_types_gazetteer, s, lang, abbreviate_prob=1.0, separate_prob=0.5)
if abbrev != s:
writer.writerow((lang, country, tsv_string(abbrev)))
if i % 1000 == 0 and i > 0: if i % 1000 == 0 and i > 0:
print('did {} ways'.format(i)) print('did {} ways'.format(i))
i += 1 i += 1