Update app.py

app.py

@@ -14,9 +14,6 @@ nltk.download('punkt')
 
 from nltk import sent_tokenize
 
-
-
-
 def extract_text_from_url(url: str):
     '''Extract text from url'''
 
@@ -109,7 +106,6 @@ def preprocess_plain_text(text, window_size=3):
     return passages
 
 
-#@st.experimental_memo(suppress_st_warning=True)
 def bi_encode(bi_enc, passages):
     global bi_encoder
     # We use the Bi-Encoder to encode all passages, so that we can use it with sematic search
@@ -122,12 +118,11 @@ def bi_encode(bi_enc, passages):
     with st.spinner('Encoding passages into a vector space...'):
         corpus_embeddings = bi_encoder.encode(passages, convert_to_tensor=True, show_progress_bar=True)
 
-    st.success(f"Embeddings computed. Shape: {corpus_embeddings.shape}")
+    st.success(f"Embeddings computed.")
 
     return bi_encoder, corpus_embeddings
 
 
-#@st.experimental_singleton(suppress_st_warning=True)
 def cross_encode():
     global cross_encoder
     # We use a cross-encoder, to re-rank the results list to improve the quality
@@ -144,7 +139,7 @@ def display_as_table(model, top_k=2, score='score'):
     return df
 
 
-# Streamlit App
+
 
 st.title("Search with Retrieve & Rerank")
 window_size = 3
@@ -178,7 +173,7 @@ def search_func(query, top_k=2):
     for idx in range(len(cross_scores)):
         hits[idx]['cross-score'] = cross_scores[idx]
 
-    # Output of top-3 hits from bi-encoder
+    # Output of top hits from bi-encoder
     st.markdown("\n-------------------------\n")
     st.subheader(f"Top-{top_k} Bi-Encoder Retrieval hits")
     hits = sorted(hits, key=lambda x: x['score'], reverse=True)
@@ -186,7 +181,7 @@ def search_func(query, top_k=2):
     cross_df = display_as_table(hits, top_k)
     st.write(cross_df.to_html(index=False), unsafe_allow_html=True)
 
-    # Output of top-3 hits from re-ranker
+    # Output of top hits from cross encoder
     st.markdown("\n-------------------------\n")
     st.subheader(f"Top-{top_k} Cross-Encoder Re-ranker hits")
     hits = sorted(hits, key=lambda x: x['cross-score'], reverse=True)
@@ -204,9 +199,7 @@ def clear_search_text():
     st.session_state["text_input"] = ""
 
 
-url_text = st.text_input("Please Enter a url here",
-                         value="https://en.wikipedia.org/wiki/Virat_Kohli",
-                         key='text_url', on_change=clear_search_text)
+url_text = st.text_input("Please Enter a url here",value="https://en.wikipedia.org/wiki/Virat_Kohli",key='text_url', on_change=clear_search_text)
 
 st.markdown(
     "<h3 style='text-align: center; color: red;'>OR</h3>",
@@ -234,8 +227,7 @@ with col1:
     search = st.button("Search", key='search_but', help='Click to Search!!')
 
 with col2:
-    clear = st.button("Clear Text Input", on_click=clear_text, key='clear',
-                      help='Click to clear the URL input and search query')
+    clear = st.button("Clear Text Input", on_click=clear_text, key='clear',help='Click to clear the URL and query')
 
 if search:
     if bi_encoder_type: