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

[osm] Using the default probabilities for abbreviations in ways training data

Browse Source
This commit is contained in:
Al
2016-01-24 00:53:41 -05:00
parent bab7a0f961
commit 00ce71223f
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
scripts/geodata/osm/osm_address_training_data.py
View File

@@ -502,7 +502,7 @@ def build_ways_training_data(language_rtree, infile, out_dir):
for s in v.split(';'):
if lang in languages:
writer.writerow((lang, country, tsv_string(s)))
abbrev = osm_abbreviate(street_types_gazetteer, s, lang, abbreviate_prob=1.0, separate_prob=0.5)
abbrev = osm_abbreviate(street_types_gazetteer, s, lang)
if abbrev != s:
writer.writerow((lang, country, tsv_string(abbrev)))
if i % 1000 == 0 and i > 0: