More hits for crossencoder

app.py

@@ -170,7 +170,9 @@ def search_func(query, bi_encoder_type, top_k=top_k):
     else:
          question_embedding = bi_encoder.encode(query, convert_to_tensor=True)
     question_embedding = question_embedding.cpu()
-    hits = util.semantic_search(question_embedding, corpus_embeddings, top_k=top_k,score_function=util.dot_score)
+
+    HITS_NUM=20 # Number of hits for reranker
+    hits = util.semantic_search(question_embedding, corpus_embeddings, top_k=HITS_NUM,score_function=util.dot_score)
     hits = hits[0]  # Get the hits for the first query
 
     ##### Re-Ranking #####
@@ -229,7 +231,7 @@ if search:
         with st.spinner(
             text=f"Loading {bi_encoder_type} bi-encoder and embedding document into vector space. This might take a few seconds depending on the length of your document..."
         ):
-            bi_encoder, corpus_embeddings = bi_encode(bi_encoder_type,passages)
+            bi_encoder, corpus_embeddings = bi_encode(bi_encoder_type,passages, dataset_name=dataset_type)
             cross_encoder = cross_encode(cross_encoder_type)
             bm25 = bm25_api(passages)
             

mcontriever-scifact-pl-corpus

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:291dd709b7bbe015f1176c9fe37fe8e7a64d55673cac2729a304c7b26a40addd
+size 15923056