switched back to langchain

app.py

@@ -1,9 +1,9 @@
 import os
+import spaces
 from threading import Thread
 from typing import Iterator
 from backend2 import load_documents, prepare_documents, get_context_sources
 import gradio as gr
-import spaces
 import torch
 from transformers import AutoModelForCausalLM, GemmaTokenizerFast, TextIteratorStreamer
 from huggingface_hub import login

backend.py

@@ -1,4 +1,5 @@
 import torch
+import spaces
 import os
 from transformers import AutoModelForCausalLM, GemmaTokenizerFast, TextIteratorStreamer, AutoTokenizer
 from interface import GemmaLLMInterface
@@ -7,7 +8,6 @@ from llama_index.embeddings.instructor import InstructorEmbedding
 import gradio as gr
 from llama_index.core import Settings, VectorStoreIndex, SimpleDirectoryReader, ChatPromptTemplate, PromptTemplate, load_index_from_storage, StorageContext
 from llama_index.core.node_parser import SentenceSplitter
-import spaces
 from huggingface_hub import login
 from llama_index.core.memory import ChatMemoryBuffer
 from typing import Iterator, List, Any