Update app.py

app.py

@@ -74,11 +74,11 @@ else:
 embedding_model = SentenceTransformerEmbeddings(model_name='thenlper/gte-large')
 # Define the embedding model and the vectorstore
 
-client = PersistentClient(path="./report_10kdb")
-print("Available collections:", client.list_collections())
+# client = PersistentClient(path="./report_10kdb")
+# print("Available collections:", client.list_collections())
 
 
-collection_name = 'report-10k-2024'
+# collection_name = 'report-10k-2024'
 collection_name ='report_10k_collection'
 vectorstore_persisted = Chroma(
     collection_name=collection_name,
@@ -87,13 +87,6 @@ vectorstore_persisted = Chroma(
 )
 
 
-results = vectorstore_persisted.similarity_search("Amazon", k=5)
-print(f"Retrieved {len(results)} results")
-for doc in results:
-    print("Doc:", doc.page_content[:200])
-    print("Metadata:", doc.metadata)
-
-
 # Load the persisted vectorDB
 
 retriever = vectorstore_persisted.as_retriever(
@@ -252,7 +245,7 @@ def get_predict(question, company):
     else:
         return "Invalid company selected"
 
-    output = predict(question, selectedCompany)    
+    output = predict(question, company)    
     return output
 
 # Set-up the Gradio UI