pastk/organicmaps
1
Fork 0
forked from organicmaps/organicmaps
Code Pull requests Activity

[search][search_quality] Drop categorial requests cause we use different ranking model for them.

Browse source
This commit is contained in:
tatiana-yan 2019-06-19 13:54:07 +03:00 • committed by mpimenov
parent 1da67cc18f
commit 2392ef7711
1 changed files with 8 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
search/search_quality/scoring_model.py
View file

@ -20,8 +20,8 @@ MAX_POPULARITY = 255
RELEVANCES = {'Harmful': -3, 'Irrelevant': 0, 'Relevant': 1, 'Vital': 3}
NAME_SCORES = ['Zero', 'Substring', 'Prefix', 'Full Match']
SEARCH_TYPES = ['POI', 'Building', 'Street', 'Unclassified', 'Village', 'City', 'State', 'Country']
FEATURES = ['DistanceToPivot', 'Rank', 'Popularity', 'Rating', 'FalseCats', 'ErrorsMade', 'AllTokensUsed',
'CategorialRequest', 'HasName'] + NAME_SCORES + SEARCH_TYPES
FEATURES = ['DistanceToPivot', 'Rank', 'Popularity', 'Rating', 'FalseCats', 'ErrorsMade',
'AllTokensUsed'] + NAME_SCORES + SEARCH_TYPES
BOOTSTRAP_ITERATIONS = 10000
@ -272,6 +272,12 @@ def get_normalized_coefs(clf):
def main(args):
data = pd.read_csv(sys.stdin)
# Drop categorial requests cause we use different ranking model for them.
data.drop(data[data['IsCategorialRequest'] == 1].index, inplace=True)
data.reset_index(inplace=True, drop=True)
data.drop(columns=['IsCategorialRequest', 'HasName'], inplace=True)
normalize_data(data)
ndcgs = compute_ndcgs_without_ws(data);