From 5781813cbd48ee32ae21336ee572f7a786dc989a Mon Sep 17 00:00:00 2001 From: Al Date: Wed, 25 Nov 2015 00:39:52 -0500 Subject: [PATCH] [fix] For countries like Denmark, removing country with a smaller probability --- scripts/geodata/osm/osm_address_training_data.py | 3 +++ 1 file changed, 3 insertions(+) diff --git a/scripts/geodata/osm/osm_address_training_data.py b/scripts/geodata/osm/osm_address_training_data.py index bee89c27..933af991 100644 --- a/scripts/geodata/osm/osm_address_training_data.py +++ b/scripts/geodata/osm/osm_address_training_data.py @@ -707,6 +707,9 @@ def build_address_format_training_data_limited(language_rtree, infile, out_dir): have_country = True _ = value.pop(k, None) + if have_country and random.random() < 0.4: + have_country = False + if not value: continue