Update app.py

app.py

@@ -150,43 +150,6 @@ def get_field(x):
         return np.nan
 
 
-#def query_records(search_term):
-
-
-#     # Fetch records based on the search term in the abstract!
-#     query = Works().search([search_term])
-#     query_length = Works().search([search_term]).count()
-    
-#     records = []
-#     #total_pages = (query_length + 199) // 200  # Calculate total number of pages
-#     progress=gr.Progress()
-    
-#     for i, record in progress.tqdm(enumerate(chain(*query.paginate(per_page=200)))):
-#         records.append(record)
-        
-#         # Calculate progress from 0 to 0.1
-#         #achieved_progress = min(0.1, (i + 1) / query_length * 0.1)
-        
-#         # Update progress bar
-#         #progress(achieved_progress, desc="Getting queried data...")
-
-
-    
-#     records_df = pd.DataFrame(records)
-#     records_df['abstract'] = [invert_abstract(t) for t in records_df['abstract_inverted_index']]
-
-#     records_df['parsed_publication'] = [get_pub(x) for x in records_df['primary_location']]
-
-    
-#     records_df['parsed_publication'] = records_df['parsed_publication'].fillna(' ')
-#     records_df['abstract'] = records_df['abstract'].fillna(' ')
-#     records_df['title'] = records_df['title'].fillna(' ')
-
-    
-#     return records_df
-
-
-################# Setting up the model for specter2 embeddings ###################
 
 
 
@@ -206,7 +169,7 @@ model = SentenceTransformer("m7n/discipline-tuned_specter_2_024")
 @spaces.GPU(duration=60)
 def create_embeddings(texts_to_embedd):
  
-    embeddings = model.encode(texts_to_embedd,show_progress_bar=True,batch_size=32)
+    embeddings = model.encode(texts_to_embedd,show_progress_bar=True,batch_size=192)
 
     return embeddings