Update app.py

app.py

@@ -2,22 +2,24 @@ import gradio as gr
 import os
 import openai
 import pandas as pd
-from langchain.embeddings import HuggingFaceBgeEmbeddings
 from langchain.vectorstores import FAISS
 from langchain.prompts import ChatPromptTemplate, SystemMessagePromptTemplate, HumanMessagePromptTemplate
 from langchain.chains import LLMChain
 from langchain_core.output_parsers.string import StrOutputParser
 from langchain.chat_models import ChatOpenAI
+from langchain.embeddings.openai import OpenAIEmbeddings
 
 from sentence_transformers import SentenceTransformer
 
-embedding_function = SentenceTransformer("BAAI/bge-large-en-v1.5")
+embeddings = OpenAIEmbeddings()
+
+#embedding_function = SentenceTransformer("BAAI/bge-large-en-v1.5")
 
 # Set the OpenAI API key
 openai.api_key = os.getenv("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")
 
 # Load the FAISS index using LangChain's FAISS implementation
-db = FAISS.load_local("Faiss", embedding_function, allow_dangerous_deserialization=True)
+db = FAISS.load_local("Faiss", embeddings, allow_dangerous_deserialization=True)
 parser = StrOutputParser()
 
 # Load your data (e.g., a DataFrame)
@@ -25,7 +27,7 @@ df = pd.read_pickle('df_news (1).pkl')
 
 # Search function to retrieve relevant documents
 def search(query):
-    query_embedding = embedding_function.embed_query(query).reshape(1, -1).astype('float32')
+    query_embedding = embeddings.embed_query(query).reshape(1, -1).astype('float32')
     D, I = db.similarity_search_with_score(query_embedding, k=10)
     results = []
     for idx in I[0]: