Update space

app.py

@@ -1,8 +1,8 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
 
-from langchain.embeddings import HuggingFaceEmbeddings
-from langchain.vectorstores import Chroma
+from langchain_huggingface import HuggingFaceEmbeddings
+from langchain_ollama import Ollama
 
 from langchain_community.llms import Ollama
 
@@ -21,9 +21,6 @@ embedding_model = HuggingFaceEmbeddings(model_name="all-MiniLM-L6-v2")
 # Load the existing ChromaDB database
 vector_store = Chroma(persist_directory="./chroma_db", embedding_function=embedding_model)
 
-# Now you can use `vector_store` as your retriever
-retriever = vector_store.as_retriever()
-
 # Function to call a RAG LLM query
 def rag_query(query, history):
     # Invoke the chain
@@ -36,6 +33,7 @@ def rag_query(query, history):
     output = f"Answer: {answer}\n\nSources:\n" + "\n".join(unique_sources)
     return output
 
+
 # See full prompt at https://smith.langchain.com/hub/rlm/rag-prompt
 # Basically a solid prompt for RAG 
 prompt = hub.pull("rlm/rag-prompt")

requirements.txt

@@ -4,4 +4,7 @@ langchain
 langchain-community
 langchain-core
 chromadb
-ollama
+ollama
+sentence-transformers
+langchain-huggingface
+langchain-ollama