Update app.py

app.py

@@ -120,15 +120,15 @@ def respond(
     message,
     history: list[tuple[str, str]],
     system_message,
-    max_tokens,
-    temperature,
-    top_p,
+    max_tokens = None,
+    temperature = None,
+    top_p = None,
 ):
     # Load model
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     
-    dataset_path = "/IndexedDataFiles/my_knowledge_dataset"
-    index_path = "/IndexedDataFiles/my_knowledge_dataset_hnsw_index.faiss"
+    dataset_path = "./IndexedDataFiles/my_knowledge_dataset"
+    index_path = "./IndexedDataFiles/my_knowledge_dataset_hnsw_index.faiss"
     
     tokenizer = AutoTokenizer.from_pretrained("facebook/rag-sequence-nq")
     retriever = RagRetriever.from_pretrained("facebook/rag-sequence-nq", index_name="custom",
@@ -138,21 +138,15 @@ def respond(
     rag_model = RagSequenceForGeneration.from_pretrained('facebook/rag-sequence-nq', retriever=retriever)
     rag_model.retriever.init_retrieval()
     rag_model.to(device)
-    messages = [{"role": "system", "content": system_message}]
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+    if message:  # If there's a user query
+        response = retrieved_info(rag_model, message)  # Get the answer from your local FAISS and Q&A model
+        return response
 
-    messages.append({"role": "user", "content": message})
-
-    #response = ""
+    # In case no message, return an empty string
+    return ""
+    
     
-    response = retrieved_info(rag_model, message)
-
-    yield response
 
 """
 For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface