app16

app.py

@@ -52,7 +52,7 @@ data=loader.load()
 text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=150)
 docs = text_splitter.split_documents(data)
 # define embedding
-embeddings = HuggingFaceEmbeddings(model_name='thenlper/gte-base')
+embeddings = HuggingFaceEmbeddings(model_name='thenlper/gte-small')
 # create vector database from data
 persist_directory = 'docs/chroma/'
 
@@ -105,9 +105,7 @@ qa_chain = ConversationalRetrievalChain.from_llm(
 )
 
 
-def chat_interface(inputs, chat_history):
-    question = inputs[0]  # Use integer index 0 for the "Question" input
-    chat_history = inputs[1]  # Use integer index 1 for the "Chat History" input
+def chat_interface(question, chat_history):
     # ConversationalRetrievalChain
     result = qa_chain.run({"question": question, "chat_history": chat_history})
     print("Debug: Result from qa_chain.run:", result)  # Add this line for debugging