Update doc_preprocessing.py

doc_preprocessing.py

@@ -6,6 +6,8 @@ import streamlit as st
 import numpy as np
 import os
 
+import builtins 
+
 emb_model = "intfloat/multilingual-e5-base"
 
 def extract_text(file):
@@ -69,7 +71,7 @@ def get_embeddings(texts)-> np.ndarray:
         st.error(f"Error generating embeddings: {e}")
         return np.array([])
 
-@st.cache_data
+@st.cache_ressource(hash_funcs={builtins.SwigPyObject: lambda _: None})
 def process_files(files):
     all_chunks = []
     all_embeddings = []