changed to FAISS vs

app.py

@@ -4,11 +4,7 @@ import time
 import boto3
 from botocore import UNSIGNED
 from botocore.client import Config
-
-from langchain.document_loaders import WebBaseLoader
-
-from langchain.text_splitter import RecursiveCharacterTextSplitter
-text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
+import zipfile
 
 from langchain.llms import HuggingFaceHub
 model_id = HuggingFaceHub(repo_id="HuggingFaceH4/zephyr-7b-beta", model_kwargs={"temperature":0.1, "max_new_tokens":1024})
@@ -16,16 +12,19 @@ model_id = HuggingFaceHub(repo_id="HuggingFaceH4/zephyr-7b-beta", model_kwargs={
 from langchain.embeddings import HuggingFaceHubEmbeddings
 embeddings = HuggingFaceHubEmbeddings()
 
-from langchain.vectorstores import Chroma
+from langchain.vectorstores import FAISS
 
 from langchain.chains import RetrievalQA
 
 s3 = boto3.client('s3', config=Config(signature_version=UNSIGNED))
-s3.download_file('rad-rag-demos', 'vectorstores/chroma.sqlite3', './chroma_db/chroma.sqlite3')
-     
-db = Chroma(persist_directory="./chroma_db", embedding_function=embeddings)
-db.get()
-retriever = db.as_retriever()
+s3.download_file('rad-rag-demos', 'vectorstores/faiss_db_ray.zip', './chroma_db/faiss_db_ray.zip')
+with zipfile.ZipFile('./chroma_db/faiss_db_ray.zip', 'r') as zip_ref:
+    zip_ref.extractall('./chroma_db/')
+
+FAISS_INDEX_PATH='./chroma_db/faiss_db_ray'
+embeddings = HuggingFaceEmbeddings("multi-qa-mpnet-base-dot-v1")
+db = FAISS.load_local(FAISS_INDEX_PATH, embeddings)
+retriever = db.as_retriever(search_type = "mmr")
 
 global qa 
 qa = RetrievalQA.from_chain_type(llm=model_id, chain_type="stuff", retriever=retriever)