Skip to content

armgit / mwmbl

Public Imported from GitHub
0

An open source, non-profit web search engine

Exclude zero predictions

Daoud Clarke27fb80000d9e
diff --git a/mwmbl/tinysearchengine/completer.py b/mwmbl/tinysearchengine/completer.py
index a9a9204..d18c35a 100644
--- a/mwmbl/tinysearchengine/completer.py
+++ b/mwmbl/tinysearchengine/completer.py
@@ -28,7 +28,6 @@ class Completer:
matching_terms = zip(self.counts[start_index:end_index], self.terms[start_index:end_index])
top_terms = sorted(matching_terms, reverse=True)[:self.num_matches]
- print("Top terms, counts", top_terms)
if not top_terms:
return []
diff --git a/mwmbl/tinysearchengine/ltr.py b/mwmbl/tinysearchengine/ltr.py
index 52f1453..18e3d68 100644
--- a/mwmbl/tinysearchengine/ltr.py
+++ b/mwmbl/tinysearchengine/ltr.py
@@ -37,7 +37,6 @@ class FeatureExtractor(BaseEstimator, TransformerMixin):
def transform(self, X: DataFrame, y=None):
features = X.apply(get_features_as_series, axis=1)
- print("Features", features.columns)
return features
diff --git a/mwmbl/tinysearchengine/ltr_rank.py b/mwmbl/tinysearchengine/ltr_rank.py
index 353f9f0..7b30123 100644
--- a/mwmbl/tinysearchengine/ltr_rank.py
+++ b/mwmbl/tinysearchengine/ltr_rank.py
@@ -35,7 +35,7 @@ class LTRRanker(HeuristicRanker):
dataframe = DataFrame(data)
predictions = self.model.predict(dataframe)
indexes = np.argsort(predictions)[::-1]
- return [top_pages[i] for i in indexes]
+ return [top_pages[i] for i in indexes if predictions[i] > 0.0]
def external_search(self, query: str) -> list[Document]:
if self.include_wiki: